THE UNIVERSITY OF ILLINOIS LIBRARY Return this book on or before the Latest Date stamped below. *' University of Illinois Library L161 H41 MIXED METALS OR METALLIC ALLOYS BY THE SAME AUTHOll. Practical Metallurgy and Assaying. A TEXT-BOOK FOR THE USE OF TEACHERS, STUDENTS, AND ASSAYERS. Globe 8vo. Price 6s. With Illustrations. Second edition. Completely revised. Elementary Metallurgy for Use of Students. Globe Svo. Price Ss. With Illustrations. Second edition. Completely revised. Examination Questions. Embracing the questions set by the Science and Art Department for twenty -Jive years. Price Is. Second edition. Steel and Iron. FOR ADVANCED STUDENTS. Second edition. Completely revised. Globe Svo. Price 10s. 6d. Iron and Steel Manufacture. A TEXT-BOOK FOR BEGINNERS. Globe Svo. Price 3s. 6d. With Illustrations. Fourth edition. Completely revised. Principles of Metallurgy. Globe Svo. Price 6s. Second edition. With Illustrations. Metal Colouring and Bronzing. Second edition. Revised. Globe Svo. Price 5s. With Illustrations. Metallography. AN INTRODUCTION TO THE STUDY OF THE STRUCTURE OF METALS, CHIEFLY BY THE AID OF THE MICROSCOPE. Globe 8v<>. Price 6s. Pp. 158. MIXED METALS OR METALLIC ALLOYS BY AETHUE H: HIOENS HEAD OF METALLURGICAL DEPARTMENT, BIRMINGHAM MUNICIPAL TECHNICAL SCHOOL THIRD EDITION COMPLETELY REVISED AND ENLARGED MACMILLAN AND CO., LIMITED ST. MARTIN'S STREET, LONDON 1912 COPYRIGHT First Edition, 1890 Second Edition, 1901 7Yiird Edition, 1912 PREFACE TO THE THIRD EDITION DURING the last decade much progress has been made in the study of metallic alloys, and a clearer conception has been formed of the causes of many properties which before weie inexplicable. Many new alloys have been introduced, and many others have been modified to suit modern requirements and new inventions. The author has endeavoured to incor- porate the substance of these investigations and productions in the present edition, so far as the aim and scope of the work permits. BIRMINGHAM MUNICIPAL TECHNICAL SCHOOL, February 1912. 428793 PREFACE TO THE SECOND EDITION ADVANTAGE has been taken of the issue of another edition to thoroughly revise and amplify the work, so as to include some of the results of recent research. Great advances have been made in the study of alloys during the last seven or eight years, notably by Roberts-Austen, Charpy, Osmond, Le Chatelier, Wright, Moissan, Stead, Heycock and Neville, etc., and references have been made to the published works of these and other eminent authorities. M. H. Le Chatelier, in the preface to the French translation of the present work, points out that the scientific study of alloys has not yet been sufficiently matured to permit of placing the subject on a logical scientific basis, and judging by the experience of different branches of applied chemistry, such as dissolution, fractional distillation, dissociation, etc., it will take about twenty-five years to enable the new theories to be properly sifted, tested, and co-ordinated, as well as to suitably connect such a scientific basis with practical industry. During this elaboration, the present work, it is hoped, will be a useful guide to manufacturers, workers, and students who are interested in the practical applications of metallic alloys. BIRMINGHAM MUNICIPAL TECHNICAL SCHOOL, August 1901. vii PEEFACE TO THE FIRST EDITION As a teacher of metallurgy, living and working amongst Birmingham people, it is not surprising that the subject of metallic alloys has claimed a considerable amount of the author's attention, seeing that Birmingham is the very centre of what may be termed the mixed-metals trades. A work adequate to the importance of the subject has yet to be written, for our knowledge of the phenomena which attend the union of metals is still very imperfect, and rests upon a comparatively slender experimental basis. But it is hoped that the present volume may at least supply a want which is becoming increasingly felt by practical men, as well as by a large number of students, who desire a more intimate acquaintance with the nature and properties of metals in the alloyed state, as well as with metals in the free state. The author has aimed at taking up the subject where ordinary metallurgical treatises leave off, dealing with the metals in a state of admixture with each other, showing how such mixtures are usefully employed. In cases, which have come under the author's notice, where the properties of certain alloys have been investigated by eminent men, who may be taken as authorities on the subject, an abstract of their researches is given in the succeeding pages, and, as far as possible, the source of the information is duly acknow- ix MIXED METALS ledged. In some instances where knowledge appeared to be lacking, such as in German silver, for example, the author has endeavoured to supply the deficiency by experiments of his own, as well as to test the accuracy of some published state- ments which appeared to him doubtful. It has been thought advisable to give a brief account of the main properties of the separate metals, and of the effect of certain elements upon them, seeing that commercial metals are not chemically pure substances, and that the presence of the common impurities often produces a characteristic result, which may be a useful guide to the manufacturer in special cases, and assist him to determine the cause of those anomalies which are constantly occurring in practice. As many persons engaged in the melting and mixing of metals are not very familiar with chemical processes, a short account of chemical terms and principles, and of a few of the non-metallic elements, is given for their guidance. The author has experienced considerable difficulty in dealing with the various alloys in a systematic manner without repetition, but hopes that, with the aid of the copious index, any given alloy may be easily referred to. The first portion deals with the principal chemical elements, and their classification into suitable groups ; the refractory materials used in making crucibles and in furnace construction ; as well as the properties and uses of various fluxes. The alloys themselves are arranged in the follow- ing order : I. Copper alloys, including brass and bronze. II. Nickel alloys, including German silver. III. Tin alloys, including soft solder, fusible metal, and Britannia metal. IV. Lead alloys, including pewter and shot metal. PREFACE TO THE FIRST EDITION xi V. Mercury alloys or amalgams. VI. Alloys of precious metals, including gold, silver, and platinum. VII. Iron alloys. VIII. Miscellaneous alloys. Under one or other of the above headings will be found alloys containing the metals platinum, silver, gold, mercury, copper, lead, cadmium, bismuth, tin, antimony, arsenic, iron, aluminium, chromium, nickel, cobalt, manganese, tungsten, titanium, sodium, potassium, and magnesium. BIRMINGHAM AND MIDLAND INSTITUTE, October 1890. CONTENTS SECT. PAGE 1. INTRODUCTION ...... 1 2. NATURE OF METALS ..... 4 3. NATURE OF NON-METALS .... 7 PROPERTIES OF METALS .... 13 4. NOBLE METALS ...... 13 5. COPPER GROUP ...... 16 6. TIN GROUP ...... 20 7. IRON GROUP ...... 22 8. ZINC GROUP ...... 31 9. ALUMINIUM ...... 33 10. ALKALINE EARTHS . . . . . .34 11. ALKALI METALS ..... 35 TABLES OF ELEMENTS, ATOMIC WEIGHTS, ETC. . 36 12. NATURE OF ALLOYS ..... 38 13. ELECTROLYSIS AND DIFFUSION . . . . 57, 58 METALS AND OXYGEN . . . . .59 STRENGTH AT HIGH TEMPERATURES . . .60 14. MICROSCOPIC EXAMINATION OF METALS AND ALLOYS 61 15. SLAGS ....... 67 16. FLUXES ....... 68 17. REFRACTORY MATERIALS . . . .75 18. CRUCIBLES ...... 76 19. FIRE-BRICKS ...... 79 20. PREPARATION AND PROPERTIES OF ALLOYS . . 80 METALS MELTED IN VACUO . . . .82 xiii xiv MIXED METALS PAGE QUENCHING AND TEMPERING . . . .89 ALUMINO-THERMICS 98 II 21. COPPER ALLOYS ...... 99 22. EFFECTS OF DIFFERENT ELEMENTS ON COPPER . 102 23. BRASS, ITS PROPERTIES . . . . .105 SHEPHERD'S CURVE ..... 109 24. PERCY'S TABLE OF COPPER-ZINC ALLOYS . .124 25. THURSTON'S TABLE OF COPPER-ZINC ALLOYS . .127 26. TABLE OF VARIOUS COPPER-ZINC ALLOYS . . 138 27. TABLE OF MODERN BRASS ALLOYS . . .139 FREEZING- POINTS OF COPPER-ZINC ALLOYS . .140 28. SHEET BRASS AND WIRE BRASS . . . 140 28A. CORROSION OF BRASS BY ACID AND SALINE LIQUIDS 144 29. CAST-BRASS . . . . . .147 DIFFERENT VARIETIES OF BRASS . . . 150 30. OREIDE . . . . . .150 31. TALMI GOLD 151 32. TOURNAY'S ALLOY . . . . .151 33. MANNHEIM GOLD, SIMILOR . . . .151 34. TOMBAC ....... 151 35. GILDING METAL ..... 152 36. HAMILTON'S METAL, CHRYSORIN, MOSAIC GOLD . 153 37. PRINCE'S METAL . . . . .153 38. BOBIERRE'S METAL . . . . .153 39. MACHT'S YELLOW METAL .... 154 39A. COMPLEX BRASSES ..... 154 40. BRASS CONTAINING IRON .... 158 41. STERRO-METAL ...... 158 42. AICH'S METAL ...... 159 43. DELTA-METAL . . . . . .160 DICK'S PROCESS ...... 165 44. WHITE BRASS . . . . . .166 45. BIRMINGHAM PLATINUM .... 166 46. SOREL'S ALLOYS . . . . .166 47. FONTAINEMOREAU'S BRONZES . .167 CONTENTS xv SECT. PAGE 48. BRASS SOLDERS ...... 167 49. MANUFACTURE OF BRASS .... 171 50. CALAMINE BRASS . . . . .171 51. DIRECT PREPARATION OF BRASS . . . 174 52. CONDENSATION OF ZINC FUME . . .183 53. CASTING OF BRASS ..... 184 54. PLATE OR STRIP CASTING .... 184 55. INGOT CASTING ...... 190 56. AIR-DRYING STOVE ..... 190 57. MOULDING AND CASTING .... 192 58. ODD-SIDE CASTING ..... 19"5 59. FINE CASTING . . . . . .197 60. CORES ....... 198 61. FIGURE-CASTING ..... 199 62. CLEANING, DIPPING, AND PICKLING . . . 201 62A. COPPER AND ITS ALLOYS . . . . 202 62B. COPPER, BRASS, BRONZE, GERMAN SILVER, ETC. . 203 62c. DIPPING IN AQUAFORTIS, COMMON SALT, AND SOOT 204 620. WHITENING BATH ..... 205 62E. DEAD DIPPING ..... 205 62r. OLD DIPPING LIQUID ..... 208 62G. ZINC ....... 208 62n. SILVER ....... 210 621. IRON AND STEEL ..... 211 62K. LEAD, TIN, AND THEIR ALLOYS . . . 212 62L. ALUMINIUM ...... 212 III 63. BRONZE . . . . . . . 213 64. PHOSPHOR-COPPER ..... 217 65. PHOSPHOR-TIN ...... 218 66. PHYSICAL PROPERTIES OF BRONZE . . . 219 HEYCOCK AND NEVILLE'S CURVE . . . 223 67. THURSTON'S TABLE OF COPPER-TIN ALLOYS . . 224 68. GUN-METAL . .... 237 69. BELL-METAL ...... 241 70. SPECULUM-METAL . ... . . 246 70A. VARIOUS BRONZES . . . . . .248 xvi MIXED METALS IV SECT. PAGE 71. MACHINE BRONZES ..... 249 72. ALLOYS SUITABLE FOR BEARINGS, ETC. . . 250 FRENCH RAILWAY BRONZES .... 252 SPECIAL BRONZES ..... 253 73. PHOSPHOR-BRONZE ..... 255 74. SILICON-BRONZE ..... 261 75. MANGANESE-BRONZE ..... 264 76. CUPRO-MANGANESE ..... 265 77. CUPRO-FERRO-MANGANESE .... 267 78. ALUMINIUM-BRONZE ..... 274 79. COWLES'S ELECTRICAL METHOD . . . 280 80. ALUMINIUM-BRASS ..... 284 81. TESTS OF ALUMINIUM BRONZE AND BRASS . . 285 82. BRAZING ,, ,, . . 287 83. SOLDERING ,, 287 84. CHINESE AND JAPANESE BRONZES . . . 289 84A. THE ACTION OF SEA-WATER ON ALLOYS . . 294 85. WHITE BEARING METALS .... 296 86. BABBIT'S ANTI-FRICTION METAL . . . 298 87. MAGNOLIA METAL . . . . . 298 88. TABLE OF BEARING ALLOYS . . . . 300 89. MELTING AND CASTING OF BRONZE 301 90. GERMAN SILVER . . ... . 303 90A. PLATINOID . . . . . . 307 91. EXPERIMENTS ON GERMAN SILVER . . . 307 92. ANALYSES OF GERMAN SILVER . . . 310 93. MANUFACTURE OF GERMAN SILVER . . . 312 94. GERMAN SILVER SOLDERS .... 315 94A. ALLOYS FOR COINAGE ..... 317 94B. CUPRO-NlCKEL ..... 317 CONTENTS xvh VI SECT. PAGE TIN ALLOYS .... . 320 95. TIN AND ZINC . 320 96. TIN AND LEAD . 324 97. PEWTER .... . 325 98. SOFT SOLDERS . 328 99. TIN WITH LEAD AND ZINC . . 329 100. TIN AND ANTIMONY . . 331 101. BRITANNIA METAL 332 FUSIBLE ALLOYS 337 102. TIN AND BISMUTH . 337 103. TIN WITH BISMUTH AND LEAD . 338 104. ALLOYS CONTAINING CADMIUM . 342 105. ,, ,, MERCURY . 344 VII 106. LEAD ALLOYS .... . 346 107. TYPE-METAL .... . 346 108. LEAD AND ARSENIC (SHOT-METAL) . . 348 109. ,, IRON . 350 110. ,, ,, COPPER . 350 111. ,, ,, MANGANESE . 351 112. ,, ,, BISMUTH . .351 VIII 113. AMALGAMS (MERCURY ALLOYS) . 352 114. LEAD-AMALGAM . 353 115. ZINC ,, ... . 353 116. TIN . 353 117. BISMUTH ,, ... . 354 118. CADMIUM ., . 355 119. COPPER ,, ... . 355 120. GOLD ,, ... . 356 xviii MIXED METALS SECT. PAGE 121. SILVER-AMALGAM ..... 358 122. MAGNESIUM ,, . . . . . 359 123. SODIUM ,,..... 359 124. POTASSIUM ,, . . . . . 360 124A. CHROMIUM 360 IX 125. GOLD ALLOYS . . . . . .361 126. GOLD AND COPPER . . . . .363 127. ,, ,, SILVER . . . . .364 128. ,, WITH SILVER AND COPPER . . . 365 129. ,, ,, SILVER, COPPER, AND ZINC . . 368 130. ,, AND TIN . . . . .369 131. ,, ,, LEAD ..... 369 132. ,, ,, BISMUTH ..... 369 133. ,, ANTIMONY .... 370 134. ,, ,, ARSENIC ..... 370 135. ,, ,, IRON . . . . .370 136. ,, ,, PLATINUM 371 137. ,, ,, PALLADIUM .... 372 138. ,, ,, ALUMINIUM .... 372 139. COLOURED GOLDS ..... 373 140. STANDARD GOLD ..... 374 141. TABLE OF EQUIVALENTS .... 377 142. PREPARATION OF GOLD ALLOYS . . 377 143. ,, PURE GOLD . . .382 144. REFINING GOLD ..... 382 145. GOLD PLATE . . . . .383 146. ,, SOLDERS ...... 384 147. COLOURING OF GOLD . 386 148. SILVER ALLOYS . , . . .391 149. SILVER AND ARSENIC ..... 394 150. ,, ANTIMONY 394 CONTENTS xix SECT. PAGE 151. SILVER AND BISMUTH . 395 152. TIN .... . 395 153. ,, ZINC .... . 396 154. IRON .... . 397 155. ,, NICKEL .... . 397 156. ,, ,, LEAD .... . 397 157. ALUMINIUM . 398 158. ,, COPPER .... . 399 159. STANDARD SILVER .... . 402 160. SILVER ALLOYS FOR MANUFACTURING PURPOSES . 402 161. SILVER SOLDERS .... . 408 162. IMITATION SILVER ALLOYS . 413 163. MIXING AND MELTING SILVER ALLOYS 414 164. LEMEL ...... . 415 165 HALL-MARKING, ETC. . 416 166. POLISHING AND FINISHING . 416 167. SILVER, HOW IMPORTED . 418 XI 168. PLATINUM ALLOYS .... . 420 169. PLATINUM AND SILVER . 421 170. ,, ,, COPPER . 422 171. ,, WITH COPPER AND OTHER METALS . 423 172. ,, AND IRIDIUM . 423 173. ,, WITH EASILY FUSIBLE METALS . . 425 174. AND NICKEL . 425 175. PLATINOR ..... . 425 176. PLATINUM-BRONZE .... . 426 176A. RHODIUM ALLOYS .... . 426 XII 177. IRON AND STEEL ALLOYS . 427 178. ,, ,, MANGANESE . 427 179. ,, ,, NICKEL .... . 430 180. ,, ,, COBALT .... . 434 MIXED METALS SECT. PAGE 181. IRON AND MOLYBDENUM .... 434 182. ,, ,, CHROMIUM .... 435 183. ,, TITANIUM .... 438 184. ,, TUNGSTEN ..... 438 184A . ,, ,, MOLYBDENUM, CHROMIUM, AND TUNGSTEN 439 185. ,, ,, COPPER ..... 441 186. TIN 442 187. ,, ZINC ..... 443 188. ,, ,, ALUMINIUM .... 443 188A . STEEL AND VANADIUM 446 XIII MISCELLANEOUS ALLOYS .... 448 189. ALLOYS FOR CALICO-PRINTING ROLLERS . . 448 190. NON-OXIDISABLE ALLOYS .... 451 191. AMALGAMS FOR SILVERING GLOBES . . . 451 192. GERSNEIN'S ALLOY ..... 452 193. SlDERAPHITE ...... 452 194. ALLOY FOR HOROLOGY .... 453 195. ALUMINIUM ALLOYS .... 453 196. ALLOYS FOR ELECTRIC RESISTANCE . 457 INDEX ... 459 CHAPTER I INTRODUCTION 1. That metals are capable of uniting with each other to form a series of bodies, having more or less the properties of their constituents, has long been known, and probably this knowledge has been usefully applied from remote antiquity. The ancients were acquainted with seven metals, viz. gold, silver, mercury, copper, iron, tin, and lead. They knew and employed various compounds of antimony, arsenic, and zinc ; although we have no evidence that these metals were known to them in the metallic state. Gold and silver, which occur in nature in the metallic state, were probably the first metals with which man became acquainted, and as other metals were discovered, especially copper, attempts would doubtless be made to alloy the base metals with gold, in consequence of the comparative rarity of the latter. Meteoric iron is a native alloy of iron, nickel, and small quantities of other elements, said to have been used by the Esquimaux and sundry tribes for making knives and similar weapons. Copper is also occasionally found "native," and to persons accustomed to melt the precious metals, there would be no difficulty in melting copper and alloying it with gold and silver. The Latin word aes in ancient writings sometimes signifies copper, and sometimes brass, so that the two metals were occasionally confounded together. Pliny says that " an B MIXED METALS CHAP. ore of aes, termed chalcitis, occurs in Cyprus, where aes was first discovered." Here aes obviously means copper. In another place he says that aes is obtained from a mineral called cadmia. Now cadmia is the ore known at the present time as calamine, which is chiefly a carbonate of zinc. This substance, when mixed with charcoal, and strongly heated for some time in a closed crucible, in admixture with metallic copper, or with an oxide of copper, forms brass. The proper name for brass was aurichalcum, or golden copper, and it may be inferred that ores of copper and zinc were sometimes smelted together, forming brass, as the metal zinc in the free state was probably unknown till the sixteenth century. Pliny describes four different varieties of what was known as Corinthian copper. 1. White. It resembles silver in lustre and contains an excess of that metal. 2. Red. In this kind there is an excess of gold. 3. In this kind, gold, silver, and copper are mixed in equal proportions. 4. This variety was termed hepatizon, from its having a liver colour, which gives it its value. Copper was used by the ancients for many of the purposes to which it is put by the moderns. The alloys of copper with tin were used in various proportions, thus : Bronze for statues was composed of 100 parts copper and 12| parts tin. Another mixture was 100 parts copper, 10 parts lead, and 5 parts tin. Culinary pots were made of an alloy of 100 parts copper and 3 to 4 parts tin. The arms of the ancients were often made of bronze, which was rendered hard by heating and allowing to cool slowly. Tin was in common use in the time of Moses. It was doubtless the Phrenicians who supplied the Egyptians with this metal, which the former obtained from the Scilly Islands and Cornwall. Cassiteros, or tin, is mentioned by Homer. In the time of Pliny, tin was used for coating the interior of copper and brass vessels. Mercury was used by the Eomans for alloying with gold INTRODUCTION and silver to form amalgams, which were used for gilding and plating, as at the present time, by laying the amalgam on the base metals, and subsequently volatilising the mercury by heat, leaving a thin coating of the precious metal on the article. Lead was well known to the ancient Egyptians and Arabians, as well as to the Romans, by whom it was termed plumbum niyrum. Large quantities of lead were obtained from Spain and Britain. Sheet lead and lead piping were used for similar purposes to those for which they are em- ployed at the present time. A mixture of lead and tin was also used as a solder. Iron was known in very early times, but in comparatively small quantities, being obtained from meteoric stones or easily reducible oxides. Moses speaks of iron being used for swords, knives, axes, etc., which seems to imply that steel was known at that early period. Homer represents warriors as armed with bronze swords, and never as using iron weapons. Achilles proposes a ball of iron as a valuable prize to be contended for in the games, which shows its scarcity at that period. That the Romans were acquainted with steel, as well as with the method of hardening and tempering it, we have abundant evidence. The steel was probably manu- factured direct from iron ore. Steel may be considered as an alloy of iron and carbon, into which manganese frequently enters in small proportion. Thus the ancients knew the six malleable metals and their alloys above referred to, but they have left us scant informa- tion respecting the methods of extracting them from the ores. It is probable that only those ores of a simple character, or those readily acted on by reducing agents, were employed, unless the appliances at their disposal and their general chemical knowledge were superior to what known facts warrant us in believing. In the eighteenth century the investigation of the nature of alloys began to receive systematic scientific attention. Thus Reaumur, an indefatigable French chemist, concluded that steel was iron impregnated with sulphurous MIXED METALS CHAP. and saline matters. The word " sulphurous " as used at that time is nearly synonymous with the present term " combus- tible." At the end of that century, Berthollet concluded from his own experiments and those of Reaumur that steel was a compound of iron and carbon. Reaumur also explained the principle of the method of tinning iron plates. Gellert in his Metallurgic Chemistry endeavours to point out the analogy between alloys and solutions, and gives a table showing the relative solubilities of metals in each other, such as copper in silver. He further clearly showed that with regard to the solution of metals in a triple alloy, he understood the possi- bility of the division of a metal between two metals acting as solvents. Musschenbroek in the early part of the eighteenth century made some experiments on the tensile strength of metals and alloys. He writes of the "absolute cohesion by which a body resists fracture when acted on by force- drawing according to its length." Duhamel in 1792 writes of the necessity for making exact experiments upon alloys, with metals which possess a high degree of purity, the union being effected in closed vessels. In the early part of the nineteenth century the researches on alloys became more numerous, and the interest excited in them has continued to grow to the present day. NATURE OP METALS 2. The term " metal " indicates a certain number of the chemical elements which have well-defined characters in common, such as metallic lustre, conductivity, and high specific gravity. When separated from their compounds by electrolytic action they appear at the negative electrode, and hence belong to the class of electro-positive bodies. These properties are by no means equally developed in all cases ; in some instances one or more of these characteristics may be absent. On the other hand, there are substances not metallic in character in which some of these properties are strongly displayed. Thus, graphite has a metallic lustre, gas-carbon INTRODUCTION and silicon are conductors of heat and electricity, and the alkali metals are lighter than water. In fact, while there are elements having strongly marked characteristics, which may unhesitatingly be classed as metallic, there are others, such as arsenic, which have properties either metallic or non-metallic according to the standpoint from which they are viewed. These bodies form a connecting link between metals and non-metals, making it impossible to draw a strict line of demarcation. The influence of heat upon metals is very varied : some fuse at a low temperature, others require a red heat, a strong red, or a white heat respectively, to melt them. The following table by Pouillet will explain the temperatures corresponding to different colours : Incipient red heat corresponds to 525 C. 797 F. Dull red 700 1292 Incipient cherry red 800 1472 Cherry red 900 1652 Clear cherry red 1000 1832 Deep orange .. 1100 2012 Clear orange 1200 2192 White 1300 2372 Bright white 1400 2552 Dazzling white 1500 2732 Metals expand when heated and contract on cooling, and, within certain limits, the expansion is proportional to the degree of heat. Certain anomalies, however, exist, thus : Molten cast-iron expands at the moment of becoming solid, and solidified bismuth occupies a larger space than bismuth in the liquid state. One of the most distinctive features of a metal is its internal mobility, in virtue of which its shape may be altered by pressure without disruption of the mass. This property is possessed by metals in various degrees, so that the " malleability " or capability of being extended by pressure without cracking, and " ductility," or 6 MIXED METALS CHAP. the capability of being permanently elongated by a tensile stress combined with lateral pressure, are by no means equal in extent ; nor is the order of their malleability the same as for ductility, for the former depends on the softness and tenacity, while the latter is much more dependent on tenacity. By " tenacity " or tensile strength is understood the resist- ance a body offers to an attempt to pull its particles asunder when a stretching force is applied. The tenacity is generally diminished by a rise in temperature, while the reverse is often the case with regard to malleability and ductility. Some metals have a feeble tenacity, and are then generally brittle. When a body resists rupture by a bending or twisting force, after the limit of elasticity has been reached, it is said to be "tough." Elasticity is a property of matter in virtue of which a body requires force to change its shape, and requires a continued application of the force to maintain the change, and recovers its original form when the force is removed. There is a limit in every solid body, beyond which it will not return to its original form on the withdrawal of the force : this is termed the limit of perfect elasticity. " Hardness," which is measured by resistance to a compressive force, like all the other physical properties of a metal, is modified by the presence of impurities, so that, in many cases, softness is a test of purity. Most malleable metals become hardened by pressure, and often require annealing during the process of manufacture. The fractured surface of metals is often characteristic, being spoken of as fibrous, crystalline, granular, columnar, and conchoidal, thus : Wrought-iron is fibrous, zinc is crystalline, steel is granular, tin is columnar, and hard steel is conchoidal. Crystalline structure is often accompanied by brittleness, and fibrous structure by high tenacity. Most metals are much heavier than water, and the ratio which expresses the number of times a body is heavier than an equal volume of water is termed its " specific gravity." The " specific heat " or the capacity of a body for heat extends over a wide range with regard to metals ; iron for example, being ^ that of water, while lead is less than INTRODUCTION 3^. The oxides of metals are usually basic in character, but this property is only relative, as an oxide which is basic in one compound may become acid when allied with a stronger base. NATURE OF NON-METALS 3. All the chemical elements devoid of that combination of physical and chemical properties which constitute the metals are termed non-metals. Only a few of them are found in nature in the free state ; namely, oxygen, nitrogen, sulphur, and sometimes carbon and selenium. As a general rule they are opposite to the metals in their electrical relations, being electro-negative, and, consequently, appearing at the positive electrode when their compounds are de- composed by the electric current. The positive or negative character, however, like that of the metals, is not absolute, except in the elements oxygen and fluorine, but varies according to the kind of elements existing in combination. Thus, in the compound known as hydrochloric acid, chlorine is negative ; but when chlorine is in combination with oxygen it plays the part of a positive element. The combination of a metal with a non-metal, such as oxygen, is characterised by a much more decisive change of properties than in the case of union between two metals. The affinity of the different metals for oxygen, sulphur, etc., is, however, very varied, being more intense the wider the difference between the electrical relations of the constituents. In the case of gold the affinities are so feeble that this metal does not unite directly with either of these non-metals. The compounds of most metals with oxygen form a class of bodies termed bases ; while the combination of non-metals with oxygen generally forms a class of bodies of opposite properties, termed acids. Those metals, however, which ex- hibit the metallic character in a feeble degree, more often have acid than basic properties when combined with oxygen. It is upon these chemical relations, taken in conjunction 8 MIXED METALS CHAP. with the physical properties, that a true classification of the elements into metals and non-metals is based. The following is a short description of the chief non- metals which are subsequently alluded to. Oxygen (0). This is the most abundant element, forming probably one-half the solid crust of the earth, -| of all water, and about 21 per cent by volume of the air. It is necessary for life and all ordinary processes of com- bustion. In the air it is a gas, but its compounds are chiefly solid or liquid. It is the chief supporter of com- bustion that is, it forms the active medium in which bodies burn. When bodies combine with oxygen they are said to be oxidised, and the substance which imparts the oxygen is termed an oxidising agent. Conversely, substances which remove oxygen from a body are termed reducing or de- oxidising agents. The chief oxidising bodies employed in metallurgy are Oxygen, air, higher oxides of metals, basic slags, carbon dioxide, and water. Oxides, as the compounds of oxygen with other elements are termed, may be roughly divided into two groups. 1. Those which have an acid character, chiefly oxides of the non-metals, and often termed acids, such as carbonic acid C0 2 , and silica Si0 2 . 2. Those of a basic character, chiefly oxides of the metals, which are termed bases. These two classes are opposite in character, and when united in equivalent proportions, generally neutralise each other, forming what are termed " neutral " bodies, which do not possess the characteristic properties of either kind. Thus silica Si0 2 will neutralise oxide of iron FeO, forming a silicate, which is neither acid nor basic. If any compound contain an excess of acid or base, it is classified either as an acid or as a basic substance, according to the kind which predominates. Thus, 3FeO . SiO 2 is a basic silicate, and FeO . Si0 2 an acid silicate, because in the former there is more FeO than is required to neutralise the acid Si0 2 , and in the latter less than is necessary for this purpose. Hydrogen (H) is chiefly found in nature in combination INTRODUCTION 9 with oxygen, forming water H 2 0, which contains ^ its weight of hydrogen. It differs from other non-metals in not generally uniting with metals to form compounds, but metals such as palladium and iron absorb it in large quantities, when it is said to be occluded. It burns in air or in pure oxygen, forming water, and evolving great 2H + = H 2 0. It is a constituent of wood, peat, coal, and coal-gas, part of it probably existing in these bodies as water ; and in combination with carbon it forms what are termed hydro- carbons, such as marsh-gas CH 4 , and olefiant-gas C 2 H 4 . When the latter are burnt the hydrogen forms water, thus CH 4 + 40 - C0 2 + 2H 2 O Marsh-gas. Oxygen. Carbonic acid. Water. C 2 H 4 + 60 2C0 2 + 2H 2 Olefiant-gas. Oxygen. Carbonic acid. Water. In some furnaces and gas-producers steam is introduced along with air to increase the volume of combustible gases, but only a very limited amount of steam can be used for this purpose., Nitrogen (N) forms about 79 per cent by volume of the air, its chief function being to modify the active properties of oxygen. It neither burns nor supports combustion, so that the nitrogen which enters a furnace, for the most part, comes out unchanged, thus robbing it of a large amount of heat, without contributing any itself. Air is simply a mixture of oxygen and nitrogen along with small quantities of water and carbonic acid. Omit- ting the latter, its composition may be taken as By Volume. By Weight. N 79 N 77 21 O 23 100 100 10 MIXED METALS CHAP. A ton of air thus contains about 515 Ibs. of oxygen. Air resembles oxygen in its properties, but is less active on account of the inactive nitrogen. Silicon (Si). This non-metal is a greyish-black sub- stance. It is generally present in iron and a few other metals, and is supposed to exist, like carbon, in the " free " and in the " combined " state. It is of little importance as an element, but in combination it forms about J of the earth's crust. It burns in oxygen, forming silica, thus Si + 20 = Si0 2 Silicon. Oxygen. Silica. Silica (Si0 2 ) plays a prominent part in the reduction of metals from their ores, being the chief slag-forming sub- stance. It exists largely as sand, and in combination with " bases " it forms silicates ; it is therefore a useful flux. The various slags are chiefly combinations of Si0 2 with alumina A1 2 3 , lime CaO, and other metallic oxides which fuse at high temperatures. Uncombined silica is practically in- fusible. Carbon (C). This non-metal is an essential constituent of all living matter, and of all ordinary fuels, such as coal. It exists in the free state as the diamond, and as graphite or black-lead. In the latter form it is used in the manufacture of crucibles, etc., because of its infusibility, and its non- tendency to form fusible slags with acid or basic substances. It will burn away in contact with air, but will not melt or vaporise. It exists in pig-iron and steel in the free and in the combined state. Part of the free carbon of pig-iron sometimes rises to the surface of the molten mass when allowed to stand, and is known as "kish." Charcoal and coke are almost entirely composed of carbon, with a little earthy matter, which is left as ash when the carbon is burnt. Either form of car- bon will burn in oxygen, forming oxides. When carbon is strongly heated in the presence of steam the latter is de- composed and the carbon oxidised thus INTRODUCTION 11 C 4 H 2 Q = H 2 + CO Carbon. Water. Hydrogen. Carbonic oxide. 3C 4 2H 9 = CH 4 + SCO Carbon. Water. Marsh-gas. Carbonic oxide. Carbon dioxide or Carbonic acid (C0 2 ) is a gas about l times the weight of air, and is formed when carbon is burned in oxygen or in a free supply of air, thus C + 20 = C0 2 Carbon. Oxygen. Carbonic acid. Also when carbonic oxide is burned in air or oxygen, thus CO + C0 2 Carbonic oxide. Oxygen. Carbonic acid. If carbon dioxide is brought in contact with red-hot carbon, it takes up some of the latter, forming twice its volume of carbonic oxide (CO), thus C0 2 + C = SCO Carbonic acid. Carbon. Carbonic oxide. In this case carbonic acid is oxidising. C0 2 is not poisonous, but it will not support life or ordinary combustion. Carbon monoxide or Carbonic oxide (CO) is a colour- less gas, about the same weight as air, extremely poisonous, and burns in air or oxygen with a blue flame producing carbonic acid, and evolving considerable heat. The com- bustible gas formed in gas-producers is chiefly carbonic oxide. It is a powerful reducing agent, probably the chief agent in reducing oxide of iron in the blast-furnace, and zinc oxide in zinc muffles. At high temperatures CO is decomposed, especially in the presence of other bodies, such as iron, which combine with carbon. 2CO = C + C0 2 Carbonic oxide. Carbon. Carbonic acid. This is probably the case in the blast-furnace, and in the cementation process for steel. Phosphorus (P). This non-metal is generally a waxy- 12 MIXED METALS CHAP. looking crystalline solid, which readily melts and vaporises. It is highly inflammable in air, forming a white cloud of phosphorus pentoxide P 2 5 , also called phosphoric acid. It combines with oxygen in two proportions, forming oxides of phosphorus. One of these oxides unites with bases to form compounds termed phosphates. It probably exists in metals as a phosphide, but in slags as a phosphate. Thus, in refin- ing pig-iron, phosphate of iron (3FeO . P 2 5 ) ' s f un d i n tap- cinder, and phosphate of lime (4CaO . P 2 5 ) in basic slag. Phosphates are decomposed by silica at high tempera- tures, because under these conditions Si0 2 is non-volatile, thus 3FeO.P 2 O 5 + Si0 2 = 3FeO.Si0 2 + P0 5 Iron phosphate. Silica. Iron silicate. Phosphoric acid. Phosphoric acid is reduced by carbon, or even by iron, the phosphorus uniting with the iron, thus P 2 5 + 50 = 5CO + 2P Phosphoric acid. Carbon. Carbonic oxide. Phosphorus. Sulphur (S) is a non-metal, and solid at ordinary temperatures. It readily melts and vaporises, and unites with metals forming sulphides, such as lead sulphide PbS. With oxygen it forms oxides, viz. sulphur dioxide S0 2 , and sulphur trioxide S0 3 . Chlorine. This element exists in nature mainly in combination with sodium, calcium, potassium, magnesium, etc. At ordinary temperatures and pressures chlorine is a greenish-yellow gas, having a pungent and irritating smell, but by great pressure may be liquefied to a dark greenish- yellow liquid, and at low temperatures may be solidified. It is readily soluble in water. It is an active chemical agent, and combines with most metals, forming a class of bodies termed chlorides. Indirectly it acts as a powerful oxidising agent, and is thus used in bleaching and as a dis- infectant. In combination with hydrogen it forms hydro- chloric acid HC1. I INTRODUCTION 13 PROPERTIES OF THE METALS NOBLE METALS. GOLD, PLATINUM, SILVER 4. Gold is usually found in the metallic state in nature (generally associated with silver, and sometimes with copper, iron, and platinum). It is often found in ores of lead, zinc, iron, and copper. Gold is a yellow metal, with a brilliant lustre ; it exceeds all others with regard to malleability and ductility; its specific gravity is 19-32; its melting point 1061 C. ; and it is only volatile at very high tempera- tures. It is almost as soft as lead, and can be welded by pressure in the cold ; it is one of the best conductors of heat and electricity. It has a tensile strength of 7 tons per square inch, and elongates 30 per cent, when stretched, before breaking. Pure gold is too soft for general use, so that it is usually alloyed with silver and copper, which harden it, without seriously impairing its malleability and ductility. Antimony, tin, and lead are most in- jurious substances in gold, even when present in minute quantities. Gold does not oxidise in air, nor is it acted on by any single acid except selenic, but is dissolved by substances like aqua regia, which yield chlorine. Chlorine gas unites directly with gold to form chloride of gold AuCl 3 . It is readily reduced and precipitated from its solution by oxalic acid, sulphur dioxide, most metals, and by salts, such as ferrous sulphate. With cyanogen it forms gold cyanide, which, with potassium cyanide, constitutes the ordinary gilding solution. To silicates, such as glass, it imparts a ruby colour. It is unacted upon by sulphur or its compounds, so that gold exposed to sulphurous fumes does not become tarnished like silver under similar circumstances. Platinum is a white metal, with a brilliant lustre ; highly malleable and ductile ; as soft as silver, and can readily be welded ; it is very tenacious, being only exceeded in this respect 14 MIXED METALS CHAP. by iron and copper among the elementary metals ; it only melts at the highest temperatures, such as those of the oxy- hydrogen flame and the electric arc. Its melting point is 1750 C. It does not oxidise at any temperature, and resists the action of all single acids, except boiling sulphuric acid, its best solvent being aqua regia. It also dissolves in a concentrated solution of potassium cyanide. It is one of the heaviest metals, having a specific gravity of 21'5. Like silver it absorbs oxygen when melted, giving it out again on cooling, causing the mass to spit. It absorbs considerable quantities of hydrogen and other gases when strongly heated with them, especially the spongy variety called platinum black ; if this substance be introduced into a mixture of oxygen and hydrogen it causes them to combine, with the development of great heat. Platinum occurs in nature, like gold, in the metallic state, in the form of grains or nuggets, and is often associated with iron, copper, gold, silver, and certain rare metals. Silver is remarkable for its whiteness and brilliant lustre, although when precipitated from its solutions it often forms a grey powder ; it is harder than gold, but softer than copper, the relative hardness being as 4 : 5 : 7 '2. Silver is extremely malleable and ductile, with a tenacity of about 7 tons per square inch, and it elongates 30 per cent before rupture ; its specific gravity is 10'5, which may be slightly increased by the operations of coining, rolling, hammering, etc. ; it melts at 960 C. ; is the best conductor of heat and electricity ; is volatile at high temperatures, and at the temperature of the electric arc it may be boiled and distilled. When heated in a current of hydrogen it volatilises at 1330 C. It does not oxidise when heated in air, but molten silver mechanically absorbs oxygen and emits it on solidifying ; this is termed " spitting." Silver in a finely divided state is oxidised when heated with certain metallic oxides, such as cupric oxide, manganese dioxide, red lead, etc., these bodies being reduced to lower oxides. Silver is soluble in nitric and sulphuric acids. Silver unites readily with sulphur INTRODUCTION 15 when heated, forming silver sulphide Ag 2 S, which is a dark grey crystalline body, with feeble lustre, somewhat soft and malleable. When heated in air it does not form oxide or sulphate, like most other metallic sulphides, and at a red heat is decomposed into metallic silver and sulphurous acid. Dilute hydrochloric acid has no action on silver sulphide, but the strong acid attacks it. Lead, copper, or iron de- composes it when the two bodies are fused together. When silver sulphide is heated with common salt, in the presence of moist air, silver chloride is formed. Silver and all its salts dissolve in sodium thiosulphate, forming a soluble double thiosulphate (Na 2 S 2 O 3 + Ag^Og), when the sodium salt is in excess. Silver combines directly with chlorine to form silver chloride AgCl. The same substance is formed by adding hydrochloric acid, or a solu- tion of common salt to a solution containing silver, when AgCl is precipitated as a white powder ; if, however, a large excess of strong salt solution be used the AgCl is dissolved, a double salt being formed thus AgN0 3 + NaCl = AgCl + NaN0 3 AgCl + NaCl = (AgCl, NaCl). Silver chloride fuses at a low red heat to a yellow liquid, and slightly volatilises at a strong red heat It is insoluble in acids, but soluble in ammonia, sodium chloride, sodium thiosulphate, and potassium cyanide. It may be reduced by hydrogen, carbonate of soda, zinc, iron, and several other metals, and partially by sulphur. It unites with oxide of lead in all proportions, and partially so with sulphide of lead and some other sulphides. Silver occurs in nature in the metallic state ; as sulphide in silver glance, which is often associated with the sulphides of lead, antimony, and iron ; as bromide and iodide ; as chloride in horn silver ; in many lead, zinc, and copper ores ; and sometimes in iron pyrites. Silver is too soft to be worked by itself for most purposes, pure silver being only used in special cases, where the 16 MIXED METAI CHAP. presence of another nietal would exert an injurious effect In most cases, silver is alloyed with copper, and occasionally with other metals, as in silver solders. COPPER GROUP. COPPER, MERCURY, LEAD, AND BISMUTH 5. Copper has a red colour ; is highly malleable, ductile, and tough ; it melts at about 1084 C., is not sensibly volatile, except at very high temperatures ; its specific gravity is 8-82 ; it has a tenacity of 8*5 tons per square inch, which may be slightly increased by hammering and rolling. Copper is one of the best conductors of heat and electricity, but this property is considerably interfered with by the presence of small traces of impurities. When a bar of pure copper is freshly broken it exhibits a fibrous silky fracture, of a light salmon colour. It readily unites with oxygen at a red heat, forming one or both of the two oxides, known respectively as black and red oxide, in virtue of their colour. The red or cuprous oxide is highly basic, and unites with acid sub- stances, such as silica, forming copper salts. Cuprous oxide is soluble in molten copper, making it dry in appearance and brittle in character. This may be remedied by remelting the copper with a little charcoal, and stirring with a pole of green wood. Commercial or tough-pitch copper is never pure, but the impurities are neutralised by the presence of a little oxygen. If the poling referred to above be continued too far, the neutralising oxygen is removed and the other impurities present act on the copper prejudicially, making it brittle. The copper is then said to be over-poled. Copper does not, however, unite with carbon. Copper unites directly with sulphur when the two bodies are fused together, forming cuprous sulphide, which is of a dark bluish-grey colour, shows a finely-granular fractured surface when broken, and has a metallic lustre. Phosphorus is highly injurious to copper when allowed to remain in it, but a small quantity may, under certain circum- INTRODUCTION 17 stances, exert a refining influence, provided the whole of it is afterwards removed. The element silicon, when reduced from sand by the action of carbon, unites with copper, making it much harder, and causing it somewhat to resemble gun-metal in colour, but diminishes its toughness and malleability. Lead, bismuth, and antimony have an injurious action on copper, tending to make it hard, brittle, and cold-short. The common impurities in copper are iron, bismuth, arsenic, antimony, and cuprous oxide ; sometimes tin, sul- phur, lead, nickel, and cobalt are present. The varieties of commercial copper are Rosette or Japan copper, the surface of which presents a peculiar red colour, due to a coating of oxide, formed by throwing water on the surface of the metal while in a heated state. Bean -shot and feathered -shot copper, which are obtained in the form of globules and flakes respectively, by running the metal into hot or cold water. Tough-cake is a variety cast into rectangular slabs, convenient for rolling, etc. Best-selected is the name applied to the purest variety of commercial copper, special care being taken to free it from sulphur, arsenic, antimony, and iron. Russian-copper, which generally contains traces of iron, but is otherwise very pure. Chili-bars. This variety, as imported into this country, is prepared in bars weighing about 200 Ibs. each ; the copper being in a raw state, requires to be refined. Electrolytic or deposited copper. The chief ores of copper are : Pyrites Cu 2 S, Fe 2 S 3 ; malachite CuC0 3 , CuH 2 2 ; azurite (2CuC0 3 , CuH 2 2 ) ; copper glance Cu S ; and cuprite Cu 2 0. Mercury. This is the only one of the useful metals which is liquid at ordinary temperatures ; it is also called quicksilver, and has been known from the most remote times. It has a silver-white colour with a brilliant lustre ; is devoid of taste or odour when pure ; at a temperature of 360 it boils, and at - 39 '4 C. it solidifies, forming a soft, white, ductile, and malleable mass, exhibiting a granular structure on the freshly fractured surface. It has a high and fairly C 18 MIXED METALS CHAP. regular coefficient of expansion for heat, which renders it suitable for thermometers and similar instruments ; its specific heat is -0332, and its density at 4 C. is 13*59. Liquid mercury does not oxidise in air, except when near its boiling point, which forms a ready means of detecting the presence of base metals, such as lead and antimony, added as adulterations, or present as impurities. Impure mercury, when exposed to air or oxygen, becomes coated with a grey film, due to the oxidation of the impurities. At its boiling point mercury is slowly oxidised to mercuric oxide HgO. It combines directly with sulphur, forming an important com- pound, mercuric sulphide or vermilion HgS. Mercury unites with most metals forming "amalgams," some of which are liquid, others semi-liquid, and some solid. The solid amalgams are regarded as chemical compounds, while the liquid amalgams may be solutions of compounds in excess of mercury, but the affinity is feeble as the mercury is partially expelled by pressure, and completely so, in most cases, by heat. Amalgams are formed (1) by rubbing the metal in a finely divided state with mercury, an increase of temperature facilitat- ing the amalgamation ; (2) by dipping a metal into the solution of a mercury salt ; (3) by voltaic action, as when a metal is placed in contact with mercury and an acid ; (4) by mixing a metal, such as gold, with an amalgam of a highly positive metal, such as sodium. Mercury sometimes occurs in the metallic state, sometimes as an amalgam with silver, and occasionally as chloride, bromide, and iodide of mercury. The chief source of the metal is the sulphide HgS, known as cinnabar. Lead. This metal has a bluish-grey colour, and possesses considerable lustre ; it is malleable, ductile, and tough, but has a feeble tenacity. The lustre of a freshly cut surface soon becomes dim when exposed to the air, owing to the formation of a film of suboxide of lead. Pure lead emits a dull sound when struck, but the presence of impurities renders it more sonorous ; also when the pure metal is cast INTRODUCTION 19 in the form of a hollow sphere it becomes somewhat sonorous. Its specific gravity is 11-37, and all base metals, when alloyed with it, lower its density. Its melting point is 326 C., and it is not well adapted for castings, since it contracts considerably on solidifying. It is so soft that it can easily be cut with a knife, and two clean surfaces of lead can easily be welded together by pressure in the cold, and also, when in a finely divided state, the metal can be pressed into a compact mass. Lead exhibits in a remarkable degree the property of flowing when in a viscous state. Lead pipes, rods, etc., are made by squirting the viscous metal through suitable cylinders by means of hydraulic pressure. Lead piping, termed composition, contains tin or antimony, whkh hardens it. Its specific heat between and 100 C. is 0314, and its coefficient of expansion is -00003 for each degree between and 100 C. If lead is boiled with water containing oxygen it is partially dissolved, and the liquid affords an alkaline reaction. The metal is oxidised when exposed to moist air ; it is somewhat volatile when heated in air, forms lead oxide PbO, and this oxide acts as an oxidising agent on many metals, such as copper, zinc, iron, etc. Two important oxides, viz. litharge PbO, and red lead PbgO^ are manufactured ex- tensively, having important industrial applications. Lead and sulphur unite when heated together, forming lead sul- phide PbS, which is a bluish-grey, brittle, and crystalline body. Commercial lead is often nearly pure, but it generally contains some silver, copper, antimony, tin, and sulphur ; and occasionally iron, arsenic, zinc, and manganese. Lead occurs in nature as galena PbS, cerussite PbC0 3 , and pyromorphite 3Pb 3 P 2 O 8 , PbCl 2 . Bismuth is a hard, greyish-white metal with a reddish tint and bright metallic lustre. Its specific gravity is 9*8, which may be reduced by pressure ; it melts at 268 C. and volatilises at a high temperature, burning with a blue flame, forming flowers of bismuth Bi<,0 3 ; it expands in the act of solidifying. When the metal is melted, and allowed to cool 20 MIXED METALS CHAP. until its surface begins to solidify, the crust broken, and the metal poured out, fine large crystals are obtained. They oxidise in air, and frequently become covered with an iridescent film of oxide. Bismuth unites with sulphur, forming a dark-grey, metallic-looking sulphide Bi 2 S 3 . Bismuth serves for the preparation of many pharmaceu- tical products and for cosmetics. The chief use of the metal is in the preparation of fusible alloys, the melting points of which can be altered according to the proportions of their constituents. It occurs in nature in the metallic state, as bismuth -glance Bi 2 S 3 , as bismuth-ochre Bi 2 3 , and often in company with silver, lead, tin, copper, and cobalt ores. TIN GROUP. TIN, ANTIMONY, AND ARSENIC 6. Tin is a white metal with a brilliant lustre ; very malleable, as seen by the thinness to which tin-foil can be reduced. It has a very low tenacity about two tons per square inch. A bar of tin when bent produces a crackling sound, known as the " cry of tin," supposed to be due to the grinding action of its crystals over each other. Its specific gravity is 7 '3, it melts at 232 C., and may be somewhat strongly heated without volatilising. When raised to a temperature near its melting point, and allowed to fall from a considerable height, the metal breaks up into the form of long grains, known as grain-tin. When tin is melted, and poured into a mould at a temperature little removed from the point at which it solidifies, the surface remains bright, if pure, but the presence of a little lead, iron, or other base metals imparts a more or less dull and frosted appearance, so that the brilliancy of the surface is a test of purity. Tin is easily crystallised superficially by treating its surface with a mixture of dilute sulphuric and nitric acids ; the orna- mental appearance, known as Moire'e Metallique, is obtained in this way. Tin is an inferior conductor of heat and electricity ; it takes a high polish, and the radiation of heat from its surface is small. It forms a valuable metal for i INTRODUCTION 21 coating culinary vessels. 4t is little affected by air at ordinary temperatures, and is therefore used for coating iron to protect it from rust. It unites readily with sulphur on the application of heat, forming stannous sulphide SnS. Commercial tin often contains small portions of lead, iron, copper, arsenic, antimony, bismuth, tungsten, and sometimes manganese and zinc. The tin of commerce is quoted as common, refined, and grain-tin. The refined-tin is made from the best ores, and is morfe perfectly refined than common tin. Grain-tin is obtained from the best pigs, which are heated and dropped from a height, as referred to on the previous page. The only important ore of tin is tin-stone, which contains tin dioxide SnOg. Antimony. Ordinary commercial antimony is often very impure, containing iron, lead, arsenic, and sulphur, and is called "regulus of antimony." Antimony is a brilliant bluish- white metal, highly crystalline, with fernlike markings on the surface, and very brittle, so that it may be easily powdered ; its specific gravity is 6*7 1 ; it melts at 632 C., and volatilises at a higher temperature. It does not oxidise at ordinary temperatures, but when heated in air, antimonious oxide Sb. 2 O 3 is formed ; and at a red heat antimony burns with a bluish- white flame producing dense white fumes of Sb 2 3 . Antimony and sulphur readily unite when heated together, forming Sb. 2 S 3 ; the same compound is also formed by heating the oxide with sulphur, thus 2Sb 2 3 x 9S = 3S0 2 + 2Sb 2 S 3 . Antimony unites with other metals to form valuable alloys, in consequence of its hardening properties, but it impairs the malleability and ductility of the malleable metals. The effect of even small quantities of antimony on the malleable metals, such as copper, gold, iron, etc., is most injurious, making them hard and brittle. Antimony occurs native, and in combination with other ores, but the chief ore is "stibnite" 22 MIXED METALS CHAP. Arsenic. This metal has brilliant, dark steel -grey colour and metallic lustre ; it is crystalline, exceedingly brittle, and may be readily reduced to powder. It is a comparatively poor conductor of heat and electricity. When heated to 180 C. in a closed vessel it begins to volatilise without fusing, and crystallises as it condenses in rhombo- hedrons, similar to those of antimony. At 500 C. it liquefies if heated under pressure. At 200 C. its vapour is phos- phorescent. Its vapour is colouTless, and possesses a peculiar garlic-like odour. The metal may be exposed to dry air without undergoing change. When the vapour of arsenic is condensed at a temperature a little below its volatilising point, it scarcely oxidises in air, even when heated to 80 C. When it is condensed on a cold surface and not completely surrounded by its own vapour, it solidifies as a dark grey crystalline powder, which is readily oxidisable. If heated in air it absorbs oxygen, and burns with a bluish flame, forming arsenious acid As 2 3 , which is condensed as a white powder when in contact with a cool body. The specific gravity of arsenic is 5-67. Arsenic occurs in nature as realgar As 2 S 2 , orpiment As 2 S 3 , mispickel FeAs + FeS, nickel pyrites NiAs, and kupfer-nickel or copper-nickel NiAs 2 . The metal is obtained by heating nickel pyrites, mispickel, etc., in closed retorts, when the arsenic is expelled, and sublimes into condensing chambers. Arsenic enters into the composition of some alloys, such as shot metal, its general effect being to harden and render the alloys brittle and more fusible. Its compounds are used in medicine, in bronzing, and in glass making. IRON GROUP. IRON, CHROMIUM, MANGANESE, MOLYBDENUM, TUNGSTEN, URANIUM, NICKEL, AND COBALT 7. Iron. Malleable-iron is of a greyish-white colour, having a granular, crystalline, or fibrous fracture, according INTRODUCTION 23 to the mode of treatment. When rolled or hammered hot, the iron becomes fibrous, but continued cold hammering induces a crystalline or granular structure, making it hard and brittle. The nature of the fractured surface varies also with the manner in which the iron has been broken, for specimens broken by progressively increasing stresses invariably show a fibrous structure, whilst the same specimen broken by a sudden blow may be crystalline. The presence of impurities generally tends to impart a granular or crystalline fracture, and makes the iron less malleable. When impurities, such as sulphur and arsenic, render the metal unworkable at a red heat, it is said to be hot- or red-short. On the other hand, some substances, such as phosphorus, cause iron to crack when hammered cold ; it is then termed cold-short. The specific gravity of pure iron is about 7 '8 6, and its fusing point is 1505 C. Before melting wrought iron assumes a pasty form, at which point two pieces may be joined together by welding. To ensure a good weld the surfaces must be clean, and the metal at a white heat. In order to dissolve any scale the smith adds a little sand, which unites with the oxide (produced by union of iron with the oxygen of the air), and forms a fusible silicate. The presence of any foreign bodies, such as carbon, silicon, sulphur, phosphorus, copper, oxygen, etc., increases the difficulty of welding. Iron possesses considerable malleability, ductility, and tenacity. Its tensile strength ranges from 17 to 25 tons per square inch, but this, like all the other physical properties, is modified by the presence of impurities, which tend to make it harder, more fusible, and brittle. When iron is heated to dazzling whiteness in air it burns, forming the black oxide Fe 3 O 4 ; the iron also becomes crystalline, which renders it friable and brittle, and is then termed " burnt iron." Iron may be magnetised by bringing it in contact with, or near to, a magnet, but soft iron loses its magnetism when the exciting magnet is withdrawn. Its specific heat is -110; its con- ductivity about 120, silver being taken at 1000. Its electric resistance is 5 '8 times that of pure copper. When iron is 24 MIXED METALS CHAP. exposed to moist air it readily rusts or oxidises, so that it is often coated with some substance to prevent this action, such as tinning, galvanising, and painting. Professor Barff preserves iron from rusting by exposing it at a red heat to superheated steam, which imparts to it a coating of the black oxide Fe 3 4 . Pure iron is not a commercial article, but it may be obtained in several ways. 1. By reducing pure ferric oxide in a porcelain tube by means of a current of hydrogen gas at 700 C. ; the iron is obtained in the form of a dark powder, which, when somewhat heated, fires spontaneously in contact with air, forming ferric oxide Fe 2 3 . When the reduction is effected at a much higher temperature, a spongy mass of a silvery grey colour is obtained. 2. By strongly heating the purest variety of iron wire with a little pure oxide of iron, covering the mixture with powdered glass free from lead, and exposing the whole to a high temperature in a covered clay crucible. The small portion of carbon present in the wire is absorbed in reducing the oxide, while the other impurities pass into the slag. 3. By electrolytic decomposition of a solution of pure ferrous chloride or sulphate a mass of silvery white, soft, malleable iron is obtained. Iron may be exposed to dry air for an indefinite period without alteration, but in the presence of moisture a layer of rust (Fe 2 3 , 3H 2 0) is formed. The oxidation is accelerated by the presence of carbonic acid, which is always present in the air, a carbonate of iron being formed. This rapidly absorbs a further portion of water and oxygen from the air, and in this way the rusting is slowly conveyed to the centre of the mass of iron. The layer of oxide and carbonate is electro-negative with regard to iron, so that a galvanic action is set up, causing decomposition of the water. This electrical condition still further augments the liability of iron to rust. Iron is readily attacked by dilute hydrochloric or sulphuric acid, but with concentrated sulphuric acid the iron is oxidised. Dilute nitric acid dissolves iron, but the fuming a.cid renders it passive. When iron is strongly heated in contact with air i INTRODUCTION 25 or oxygen its surface is rapidly coated with a scale of black oxide Fe 3 O 4 , which peels off when struck with a hammer. Iron and Sulphur. Compounds of iron and sulphur occur in nature as pyrites. These elements readily unite when heated together, forming ferrous sulphide FeS. Sulphur, even in small quantities, has a very injurious effect on wrought-iron, making it red-short, although the metal may be readily worked in the cold. With cast-iron a small quantity of sulphur is sometimes an advantage, making it stronger, more fusible, and more liquid when poured. Sul- phur in pig-iron tends to the production of the white variety. The surface and fractured portions often exhibit black patches, which are characteristic of sulphur in iron. Ferrous sulphide, heated with carbon, is but little affected, but it is decomposed at a high temperature by oxidising sub- stances. When this sulphide is heated in air it first forms iron sulphate, which is decomposed at a higher temperature, forming ferric oxide, the sulphur being removed as sulphur dioxide. Iron and Phosphorus. These bodies readily unite when phosphorus is dropped into red-hot iron, forming a phosphide of iron Fe 12 P. When oxide of iron is reduced in the presence of an earthy phosphate, phosphorus is separated, and unites with the iron. *3 per cent of phos- phorus in wrought-iron makes it harder and somewhat diminishes its tenacity. '5 per cent makes it cold-short but not red-short. 1 per cent makes it very brittle. The effect of phosphorus on iron is to impart a coarsely crystalline structure, diminish its strength, increase its fusibility, and make it cold-short. The presence of phos- phorus in cast-iron diminishes its strength, but on account of its imparting fluidity to the metal, its presence is beneficial in making fine castings. Phosphorus is generally injurious to steel, even in very minute quantities, especially when the carbon is high. Iron and Arsenic. The effect of arsenic on iron is much less injurious than that of sulphur, and when less than "25 26 MIXED METALS CHAP. per cent appears to exert but little influence upon it. Arsenic is also much less injurious to steel than is ordinarily supposed. Mr. Stead finds that -15 per cent has no material effect on its mechanical properties. With 1 per cent the tenacity is in- creased, and the elongation somewhat reduced. With hi'gher amounts, up to 4 per cent, the elongation diminishes but the tenacity increases. Several compounds of iron and arsenic are known, varying from grey to white in colour. Iron and Silicon. A compound of silicon and iron, highly crystalline, and of a silver-white colour, known as silicon-iron, is now an article of commerce, and is used for producing soundness in steel castings. It is obtained by reducing silica with carbon in the presence of iron. If iron be heated alone with silica no action takes place. The effect of silicon on cast-iron is to set the combined carbon free, so that, as a rule, the greyer the pig, the higher the amount of silicon present. Much silicon makes iron hard, more easily fusible and brittle. Iron and Carbon. Carbon unites with iron in various proportions up to about 4j per cent, forming the different varieties of steel and cast-iron. When manganese is present a larger proportion of carbon may be taken up. The difference between malleable-iron, steel, and cast-iron is chiefly dependent on the relative amounts of carbon in combination with the iron. The more the carbon, the harder and more fusible the metal becomes, and this effect is considerably increased by the presence of other bodies, such as phosphorus, sulphur, etc. Combination takes place when iron is heated in contact with gaseous fuel, such as carbonic oxide, cyanogen, and hydrocarbons, or by a pro- longed exposure to a high temperature in contact with solid carbon, such an operation being termed cementation. When the carbon present in iron reaches -25 per cent, the iron is sensibly harder ; this may be considered the greatest amount of carbon which can be present in malleable iron without materially diminishing its softness and malleability. Steel may be considered as iron containing from '25 to T5 per INTRODUCTION 27 cent carbon. When the proportion of carbon is low the metal is termed " mild steel," and in like manner those with the higher proportions of carbon are termed "hard steels." Arnold states that the constituents of steel may be 1. Crystals of pure iron, which remain bright on etching. 2. Crystals of impure iron, which become pale brown on etching. 3. Normal carbide of iron Fe 3 C, which exists in three modifications, viz. Finely divided carbide as in tempered steels ; normal carbide in ill-defined striae and granules ; crystallised carbide in annealed steels. 4. He also considers that there is a sub-carbide Fe 24 C, of great hardness and present in hardened and tempered steels. It is de- composed by most acids, and splits up into Fe 3 C and free iron at 400 C. It has a remarkable capacity for permanent magnetism. 5. Graphite or temper carbon. The sub-carbide theory accounts for the cause of weakness in steel with 1*3 per cent carbon, which would be useless for a cutting edge, and yet be able to withstand a shock. There are lines of weakness along the junction of the sub-carbide granules with the normal carbide membranes. This sub- carbide is the eutectic, and stated by others to be composed of alternate layers of Fe 3 C and pure iron, and contains *89 per cent of carbon. Carbon exists in pig-iron in two states free and combined. When the carbon is chiefly in the com- bined form, the iron is "white." On the other hand, when the carbon is free, being diffused through the iron in crystal- line scales, the iron is called " grey " ; but neither variety is entirely free from graphite or combined carbon respectively. In some varieties the carbon is partly combined and partly free, which gives to the fractured surface of the metal a speckled appearance, consisting of grey spots, enclosed by reticulating lines of white ; it is then termed " mottled iron." Chromium is a comparatively rare metal, which only occurs in nature in combination with other elements, the chief ore being chrome-iron-stone FeO,Cr 2 3 . Chromium, or its oxides, forms the colouring matter of several minerals ; the 28 MIXED METALS CHAP. green colour of the emerald, for example, is due to chromium oxide. The metal is obtained by the reduction of its oxide or chloride, or by the electrolysis of its chlorides, when chromium separates out in brittle glistening scales. It is also reduced by aluminium. It is tin-white in colour, having a specific gravity of 6'8. The fused metal is said to be as hard as corundum ; it melts at a much higher temperature than does platinum, and is only slowly oxidised when heated in air. It may be polished, and acquires a fine lustre. It is not oxidisable at ordinary temperatures, but when heated in air the green oxide Cr 3 4 is formed. It is soluble in hydrochloric and sulphuric acids, but not in nitric. It is used in the form of an alloy with iron and carbon, forming a hard, white, and brilliant steel, much esteemed for special purposes. Molybdenum. A somewhat rare metal, found in com- bination with lead as a molybdate, also as a sulphide. It can be reduced from its oxide by carbon and by hydrogen at a white heat, and when fused in the oxy-hydrogen or electric furnace it forms a silvery white mass, of specific gravity 8*6, and extremely hard. It is unalterable in air but easily oxidises when heated. It forms alloys with iron, lead, tin, copper, silver, platinum, and aluminium, making those metals harder, less malleable, and less fusible. Its symbol is Mo, and atomic weight 96. Moissan has obtained pure molybdenum in the electric furnace, which he states has a density of 9 '01, is as malleable as iron, can be filed cold and forged when hot. It unites with carbon to form a crystalline carbide, Mo 2 C. Tungsten. Is found as a tungstate of iron and manganese in Wolfram, a Cornish mineral ; as tungstate of calcium and also of lead in Bohemia. The metal may be obtained in powder by reducing the oxide by hydrogen, or by subjecting the oxide, mixed with oil, to an intense heat in a carbon crucible. It is melted in the electric or oxy-hydrogen furnace. Its specific gravity is 19, and its melting point about 2000 C. In the fused form it does not oxidise when heated in air. INTRODUCTION 29 Its forms alloy with steel of intense hardness and high melting points. Its symbol is W, and atomic weight 184. Uranium has the highest atomic weight of all elements, being 240, and a specific gravity of 18*7. The principal source of the metal is pitchblende, an impure oxide. It is a hard, greyish- white metal, not oxidisable except when heated, but then it burns brilliantly, forming the oxide, probably U 3 4 . Its oxide or nitrate may be reduced by sodium, potassium, or by aluminum. The compounds of uranium are used in enamel painting, in staining glass, and in photo- graphy. It has a very high melting point. Manganese. The pure metal, obtained by the reduction of its oxide, is a grey or reddish- white .body, hard, and brittle ; its specific gravity is 8, and its melting point about 1250 C. ; it oxidises more readily than iron, and must there- fore be excluded from air by keeping it under rock-oil, or in sealed vessels. Its chief use is in the formation of alloys with iron, steel, and copper. It is not used in fche unalloyed state. Compounds of manganese are very widely distributed in nature ; one of the most common is pyrolusite or black oxide of manganese Mn0 9 . Nickel. This is a brilliant- white, malleable, ductile, weld- able, and very tenacious metal, with a melting point nearly the same as that of iron, viz. 1450 C. ; but the presence of carbon, silicon, and other impurities considerably lowers its fusing point. It takes up carbon like iron by cementation, and the carbon may exist both in the combined and in the graphitic form by fusing the cemented metal. It does not possess the property of hardening and tempering like iron. Its specific gravity is 8-8 ; it is magnetic like iron, but in a less degree. It does not readily oxidise in air at ordinary temperatures, but when heated the monoxide NiO is formed. It readily unites with sulphur, forming nickel sulphide NiS, which is brass-yellow in colour ; and with arsenic, forming nickel arsenide NiAs. Nickel is readily soluble in nitric acid, and slowly soluble in dilute sulphuric and hydrochloric acids. Organic acids and caustic alkalies have very little action on it. 30 MIXED METALS CHAP. When pure dry carbon monoxide is passed through finely divided nickel at 30 C., a peculiar volatile liquid is obtained, termed nickel carbonyl, and having the formula Ni(CO) 4 . Nickel is found in commerce in the form of dull-grey cakes or cubes, and by melting these at a high temperature a com- pact, silver- white metal is obtained. The malleability of nickel allows of its being fashioned into various articles, which pos- sess great lustre, hardness, and durability. These properties render it valuable for coating base metals by the process of electro-plating, especially as it is little liable to oxidation. Commercial nickel was formerly very impure, due to the presence of iron, copper, silicon, sulphur, arsenic, and carbon, which makes it hard and brittle. Dr. Fleitmann and other metallurgists have devised simple and effective means of refining and toughening nickel, which are now largely practised. Fleitmann adds to the melted metal minute quantities of magnesium in several charges, and well stirs each time a dose is added. One ounce of magnesium is sufficient for refining 60 Ibs. of impure nickel. The mag- nesium is supposed to reduce the occluded carbonic oxide CO forming magnesia, and to cause the carbon to separate out as graphite. Aluminium is now generally used instead of magnesium in refining nickel. Manganese is also used. Nickel unites readily with most metals forming alloys, some of which are of great commercial utility. The most im- portant of these is German silver. Nickel occurs in nature as kupfer-nickel or copper-nickel NiAs, which is a copper-red coloured mineral, with a metallic lustre. As nickel pyrites NiS, which is brass-yellow in colour. As nickel-glance, which is a variable compound of nickel, arsenic, and sulphur. As garnierite, which is a hydrated silicate of nickel, iron, and magnesium. Cobalt. This metal resembles nickel in appearance and properties, and is generally associated with it in nature. Cobalt is a white metal, with a reddish cast, highly malleable, ductile, and tenacious ; its specific gravity is 8'5 ; it is magnetic like nickel, and has a melting point of 1500 C. ; INTRODUCTION 31 almost unalterable in air at ordinary temperatures, but oxidises when heated, and at a high temperature burns with a red flame. It is seldom used in the metallic state, but its compounds are largely employed in pigments. It unites with arsenic to form iron-grey, fusible, and brittle compounds. The principal ores are srnaltine CoAs, cobalt glance Co 2 AsS, and cobalt bloom (Co 3 As0 4 ,4H 2 0). ZINC GROUP. ZINC, CADMIUM, AND MAGNESIUM 8. Zinc, commonly known by the name of " spelter " when in the cast state, is a white metal, with a bluish shade and bright metallic lustre. Ordinary zinc is hard and brittle, and when fractured exhibits a highly crystalline structure. When pure it is feebly crystalline, much more malleable than common zinc, and malleable at the ordinary temperature ; common zinc, however, becomes malleable and ductile if heated to a temperature of 100 to 150 C., but beyond 200 C. it again becomes brittle. Its specific gravity in the cast state is 6'9, which may be increased to 7'15 by rolling or forging ; it contracts but slightly on cooling from the liquid state, and is thus well adapted for castings. The castings made at a high temperature are brittle and crystalline ; but when cast near the solidifying point are more malleable. Zinc melts at 419 C. and boils at 1040 C. At a red heat in air it rapidly oxidises, and burns with a greenish- white flame, forming zinc oxide ZnO ; if raised to a bright-red heat in a closed vessel, it may be readily distilled. When rolled zinc is exposed to air and moisture a grey film of suboxide is formed, which preserves the metal from further oxidation. Ordinary zinc readily dissolves in dilute hydrochloric and sulphuric acids, while the pure metal is unaffected if quite smooth ; both kinds dissolve in nitric acid and in alkalies. Zinc displaces silver, gold, platinum, bismuth, antimony, tin, mercury, and lead from their solutions. The chief impurities of the com- mercial metal are iron, lead, and arsenic. Zinc and sulphur do not readily unite, but when a mixture 32 MIXED METALS CHAP. of finely divided zinc and sulphur is projected into a red-hot crucible, some zinc sulphide ZnS is formed. It is also formed by heating zinc with cinnabar HgS. Zinc forms compounds with phosphorus and arsenic, when these bodies are heated with it, having a metallic lustre and somewhat vitreous fracture. The chief ores of zinc are The oxide ZnO, called zincite or red oxide of zinc, which is white when pure, but generally red from the presence of oxide of manganese. The sulphide ZnS, known as " blende " and " black jack," is the principal source of the metal, and generally black or yellowish-black in colour, but sometimes it has a reddish tint from the presence of galena ; when pure it is white, and contains 67'03 per cent of zinc. The carbonate ZnCO 3 , called calamine. And the silicate (2ZnO,Si0 2 ,OH 2 ), called electric calamine. Zinc forms with other metals a most important class of alloys, such as brass, German silver, etc. It is used in the form of sheets, worked into a variety of shapes ; it protects iron from rusting, as in galvanised-iron ; it forms the electro- positive element in many batteries; and in the form of fine dust it is obtained in large quantities mixed with zinc oxide, and forms a valuable reducing agent. Cadmium. In the process of zinc extraction it was ob- served that a volatile vapour, in some cases, was distilled off with the first portions of zinc ; this was found to be the metal cadmium. It possesses a tin-white colour, has a fibrous structure, and takes a high polish. It is harder than tin, malleable, ductile, and readily volatile. It has a density of 8-6 ; it melts at about 320 C., and boils at 860 C. Its vapour is of a dark-yellow colour, with a disagreeable odour. Like tin it emits a crackling sound when bent. It is used in alloys to produce a fusible metal, which melts below 100 C. ; and an amalgam of cadmium is employed as a stopping for teeth, such amalgam being soft when first prepared, but soon becomes hard. Magnesium. This metal possesses a brilliant -white colour, but soon tarnishes when exposed to moist air, due to INTRODUCTION 33 the formation of magnesium oxide. It is stated to possess great tensile strength, being nearly equal to that of aluminium bronze. Its specific gravity is 1'74, and its melting point 633 C. At a temperature of 450 C. it can be rolled and worked into a variety of forms. Screws and threads made of this metal are sharper and more exact than those made of aluminium. When ignited in a flame it burns with a dazzling white light, which is said to have been seen at sea from a distance of twenty-eight miles. This light is used for purposes of photography. Magnesium occurs abundantly in nature in combination with other elements forming com- pounds, such as magnesite MgC0 3 , dolomite MgCa(C0 3 ) 2 , etc. ALUMINIUM 9. Aluminium. With the exception of oxygen and silicon, this is the most widely distributed of the elements, and contained in the largest quantity in the solid crust of the earth. It occurs in a variety of forms as oxide, but more generally in combination with other metals, such as zinc, iron, magnesium, etc., forming aluminates ; as silicate in all clays, and as fluoride in cryolite (6NaF,Al 2 F 6 ). Aluminium is a white metal which takes a fine polish. It has no taste or odour. It is soft, very malleable, and ductile, with an elasticity and tenacity about equal to that of silver. Aluminium, like zinc, is most malleable at 150 C., but when rolled cold must be frequently annealed. If worked at temperatures above 200 C. it becomes hot-short. Its specific gravity is 2-56, which is increased by hammering ; it melts at a temperature of 657 C., and is not volatile when strongly heated out of contact with air. Its con- ductivity for heat and electricity when pure is said to be 60 per cent that of silver. It does not oxidise in air at the ordinary temperature, or combine with sulphur. At elevated temperatures it combines rapidly with oxygen, and therefore acts as a deoxidiser, removing oxygen from iron, steel, copper, etc. It has partly replaced manganese in steel- D 34 MIXED METALS CHAP. making, and wholly replaced magnesium in nickel purification. It forms a powerful reducing agent for difficultly reducible oxides, such as those of manganese, chromium, tungsten, uranium, etc. It is insoluble in cold nitric acid ; dilute sulphuric acid has no action on it ; but hydrochloric acid and alkalies dissolve it readily. Aluminium is valuable for making articles where light- ness is an important feature ; this, combined with its lustre, unalterability in air and sulphuretted hydrogen, non-poisonous properties, and ease of working, gives it a widespread interest. As it is highly sonorous as well as light it is useful for bells, wind instruments, drums, etc. It is, however, in its alloys that its greatest value appears. In some cases it imparts strength, in others it modifies the colour, while in others it promotes soundness in castings. ALKALINE-EARTHY METALS 10. The term " earth " was formerly used to denote those bodies which are insoluble or but slightly soluble in water, and unaltered by exposure to a high temperature. Some of these were found to have an alkaline reaction, and to easily neutralise acids ; hence the term " alkaline earth." These oxides viz. baryta, strontia, lime, and magnesia were found to be composed of metals in combination with oxygen. Barium is a pale yellow metal, malleable, and fusible at a white heat. It rapidly tarnishes in air, and burns brilliantly at a red heat forming barium oxide. Its melting point is 1200 C. It decomposes water rapidly at the ordinary temperature. Its specific gravity is 3'75, and its atomic weight 137. Strontium is similar to barium in colour ; it is malleable, fusible at a white heat, quickly oxidises on exposure to air, burns brilliantly in air when heated, and violently decom- poses water. Its specific gravity is 2 '5 4, and its atomic weight 87-2. Calcium is a yellowish -white metal, tenacious and malle- INTRODUCTION 35 able ; it melts at a white heat, oxidises in air, and burns when heated; it decomposes water rapidly. Its specific gravity is 1'57, and its atomic weight 40. The pure metal is obtained from the electrolysis of its chloride. It forms an amalgam with mercury. The alkaline-earthy metals, although their compounds are widely distributed, do not occur in nature in the metallic state, and the isolated metals have little application in the arts, on account of their easy oxidation. They may be useful in removing oxygen from other metals and their alloys. ALKALI METALS. SODIUM, POTASSIUM, LITHIUM, ETC. 11. The word "alkali" was originally used as the name of a soluble salt obtained from the ashes of sea-plants, and is now applied to a well-defined class of bodies having the following properties : They turn red litmus blue, completely neutralise acids, are soluble in water, and their solutions exert a caustic action upon animal matter. The alkalies proper are the oxides of sodium, potassium, lithium, rubidium, and caesium. To these is added the hypothetical metal ammonium NH 4 , which is called the volatile alkali in contradistinction to potash and soda. The metals of the alkalies are soft, readily fusible, volatile bodies, easily oxidised on exposure to air, and they rapidly decompose water at ordinary temperatures. Sodium. This metal melts at 95 C. and volatilises, form- ing a dark blue vapour. It rapidly oxidises in air, and when strongly heated burns with a yellow light. It decomposes water rapidly at ordinary temperatures. It is a silver-white metal, with a specific gravity of '97. Sodium may be used for the preparation of aluminium, magnesium, boron, and silicon. As an amalgam it is used in the extraction of gold, and in the laboratory as a reducing agent. It occurs very abundantly in nature in a state of combination in the forms of chloride, nitrate, borate, carbonate, and silicate. Potassium. This element is very similar to sodium in 36 MIXED METALS CHAP. appearance and properties. It is a silver-white lustrous metal having a specific gravity of '87 ; it is brittle at C., but at 15C. it becomes soft, malleable, and weldable ; it melts at 62 C., forming a liquid like mercury in appearance ; at a red heat it boils, emitting a green-coloured vapour. It has a strong affinity for oxygen, and decomposes water, with evolution of great heat. It is used for similar purposes to those of sodium, and occurs abundantly in nature in analogous forms. Lithium. This is a widely diffused element, being found in many micas and felspars, in the ashes of plants, and in sea-water. It has the colour and lustre of silver, is soft and weldable, melts at 180 C., is volatile at a high temperature, burning with a white flame, and rapidly oxidises in contact with air at ordinary temperatures ; its specific gravity is 59, and it is therefore the lightest of all solid and liquid bodies. METALS Names. Symbols. Atomic Weights. Specific Gravity. Melting Point. d Aluminium Al 27 2-56 657 Antimony Sb 120 6-71 632 Arsenic As 75 5-67 450 Barium Ba 137 3-75 1200 Bismuth . Bi 207-5 9'8 268 Cadmium . Cd 112 8-6 320 Caesium . Cs 133 26 Calcium Ca 40 1-57 800 Cerium Ce 141 6-68 623 Chromium Cr 52-4 6'8 1450 Cobalt Co 58-9 8-5 1500 Copper Didymium Cu Di 63-2 145 8-82 6'54 1084 Erbium E 112-6 ... Glucinum . Gl 9 2-1 Gold . Au 197-2 19-3 1061 INTRODUCTION 37 MET A LS Continued Names. Symbols Atomic Weights. Specific Gravity. Melting Point. Indium In 113-4 7-42 165 Iridium Ir 192-5 12-2 2000 Iron . Fe 56 7-86 1505 Lanthanium La 138-5 6-2 Lead Pb 207 11-37 326 Lithium . Li 7 -59 180 Magnesium Mg 24 1-74 633 Manganese Mn 55 8 1250 Mercury . Hg 200 13-59 -39 Molybdenum Nickel . Mo Ni 96 58-6 9 8-8 1450 Niobium . Nb 94 6-27 ... Osmium . . i Os 191 22-48 2000 ? Palladium . Pd 106-2 12 1500 Platinum . Pt 195-3 21-5 1750 Potassium . K 39 87 62 Rhodium . Rh 104 12-1 2000 Rubidium . Rb 85 1-52 38 Ruthenium Ru 104 12-26 1800 Silver Ag 107-9 10-53 960 Sodium Na 23 97 95 Strontium . Sr 87-2 2-54 ... Tantalum . Ta 182 10-8 Thallium . Tl 2037 11-9 288 Thorium . . Th 232 11-1 Tin . Sn 119 7-3 232 Titanium . Ti 48 ... Tungsten . \V 184 19 2000 Uranium . U 240 18-7 Vanadium . V 51 5-5 Yttrium . Y 617 ... Zinc . Zn 65 7-15 419 Zirconium . Zr 90-4 - 4-15 ... 38 MIXED METALS CHAP. NON-METALS Names. Symbols. Atomic Weights. Specific Gravity. Boron .... B 11 2-68 Bromine .... Br 80 2-96 (Graphite Carbon^ (Diamond C 12 2-2 "3-5 Chlorine . . . ci 35-5 Fluorine F 19 Hydrogen ? H 1 Iodine I 127 '4-95 Nitrogen N 14 Oxygen 16 Phosphorus P 31 1-8-2-1 Selenium Se 79'5 4-28-4-8 Silicon Si 28-1 2-49 Sulphur S 32 1-97-2-07 Tellurium . Te 126-3 6-18 NATURE OF ALLOYS 1 2. When two or more metals are caused permanently to unite the resulting mixture is termed an alloy. The term is also used for similar mixtures of metals and non-metals ; such as iron and carbon. Several suggestions have been made to account for the origin of the word alloy ; such as " adligo " to bind to : " linderen " to lessen, because a precious metal is lessened in value by the addition of a base one : also " aloi," i.e. standard of coin (a la loi). When mercury is a chief constituent the body is generally termed an amalgam. Although several alloys occur in the native state, they are seldom capable of application to useful purposes without preparation and purification. Artificial alloys on the other hand are of the greatest possible importance, since the power of uniting the elementary metals in various proportions gives an endless variety of bodies suitable for all the requirements I NATURE OF ALLOYS 39 of mankind. To give an idea of the countless number of combinations thus possible, Roberts -Austen states, that if only one proportion of each of the thirty common metals were considered, the number of binary alloys would be 435, of ternary alloys 4060, and of quaternary alloys 27,405. If four multiples of each of the thirty metals be taken, the binary compounds are 5655, ternary 247,660, and quaternary 1,013,98s. 1 If we take into account those metals which are employed in the arts in the unalloyed state, such as iron, copper, zinc, lead, etc., and consider that they are not sent into the market in a state of purity, but contain varying amounts of other metals and metalloids, and if we extend the meaning of the word alloy so as to embrace such impure bodies, then it is not too much to say that all metals are used only in the alloyed state. Most of the metals are only practically used in the arts in the form of alloys, using the term in the more limited sense, such as gold, silver, antimony, arsenic, nickel, cobalt, cadmium, manganese, chromium, ete. The general method of effecting the union of metals is by the agency of heat, but with certain soft metals true alloys may be formed by subjecting the constituents to considerable pressure, even at the ordinary temperature. Alloys, such as those briefly referred to in the historical sketch, were doubtless first discovered by the metallurgical treatment of mixed ores, from the simultaneous reduction of which, alloys would be formed ; or in some cases, as in ores of gold and silver, naturally formed alloys would be obtained by a simple melting process. The direct preparation of alloys by the simple melting together of the constituent metals has been enormously developed in modern times, and the attention which mixed metals are now receiving from chemists is far greater than in any period of history. Comparatively few of the metals possess properties such as render them suitable to be employed alone by the manufac- turer ; but most of them have important applications in the form of alloys. Even among the metals which can be 1 Introduction to Metallurgy, Roberts-Austen, p. 74. 40 MIXED METALS CHAP. used independently, it is often found expedient to add portions of other metals, to improve or otherwise modify their physical properties. Thus gold is hardened, and made to resist wear and tear, as well as to lower its cost, by the addition of copper ; silver is likewise hardened by alloying it with copper ; and the bronze coinage is formed of an alloy of copper, zinc, and tin for similar reasons. The purposes for which metals are alloyed are as various as the uses of the metals themselves, but, as a rule, the combination is employed to harden, render more fusible, alter the colour, or to reduce the cost of production. Thus the class of alloys known as solders, which are used for joining the several parts of a body or bodies together, are formed so as to possess melting points below that of the articles to be soldered. The well-known class of alloys termed "brass" furnishes a good illustration of the effect of alloying in pro- ducing different shades of colour. These bodies are composed of the metals copper and zinc in varying proportions, the colour depending to a great extent on the quantity of copper present. When the copper predominates the colour is yellow, or reddish ; when the two metals exist in equal proportions the colour is still yellow ; beyond this, when the zinc is in excess, the colour gets white, or bluish-white, resembling impure zinc. Nickel is added to brass to whiten it, forming German silver. Again, some metals, such as copper, can only with difficulty be made to produce sound castings ; and the metal is too tough to be conveniently wrought in the lathe or with the file, but when alloyed with zinc or tin, good castings can be readily obtained, and rolled, turned, or filed with con- siderable facility. In some cases the tensile strength of a metal is enormously increased by the addition of another metal, sometimes in very small proportions ; the various bronzes may be cited as examples. The addition of a second metal is often a source of weakness, as in the case of adding antimony to lead. One might be led to consider that the alloying of two malleable metals would produce a malleable i OYS 41 alloy, and while in many cases this undoubtedly is so, there are others in which the opposite is the fact. Thus lead added to gold in very small quantity makes the gold ex- ceedingly brittle and weak. The specific gravity of an alloy nearly always differs from the mean specific gravities of the constituents, sometimes being greater and sometimes less. When the density is in- creased it shows that contraction has occurred, and chemical combination has probably taken place between the com- ponents. This is the case with bronze rich in copper, while with similar alloys rich in tin expansion occurs, the specific gravity being less than the mean of the specific gravities of the two metals. One of the greatest difficulties connected with the subject of alloying is the tendency of the constituents to separate on cooling according to their specific gravities. As a rule, it is more difficult to alloy three or four metals than two metals, especially when the components differ widely in fusibility, unless the combination forms a true chemical compound. The mixture is promoted by constant agitation when the body is in the liquid condition, and by pouring the metal into the mould at the lowest possible temperature con- sistent with the proper degree of liquidity. The colour of alloys is a matter of considerable interest. Metals do not generally appear as highly-coloured bodies, though their brilliancy is often intense. The angle of in- cidence of the light has much to do with this. A polished plate of gold observed from a very obtuse angle appears nearly white, but by altering the angle only a few degrees the yellow colour appears. It may be still further enriched by repeated reflections at small angles of incidence ; hence chased and granulated gold appears to have a far richer colour than plain polished gold. Copper also, in this way, may be made to yield a nearly pure red light. By alloying metals together the colour of a pure metal is considerably modified. Gold with 8 per cent silver has a greenish-yellow colour, the greenness of which may be rendered still more decided by a small further addition of silver. Copper 42 MIXED METALS CHAP. reddens gold. A large proportion of silver, varying from 20 to 50 per cent, added to gold, renders it nearly white. The British gold coinage has a yellowish-orange tint. An alloy of 85 parts of copper and 15 parts of tin, or a mixture of tin and zinc, has a rich yellow colour. Five parts of aluminium added to 95 parts of copper produce a similar tint. If copper is mixed with 30 per cent tin the colour is greyish-white. The alloy of 51 per cent of copper with 49 per cent of antimony has a beautiful violet colour. Roberts- Austen l found that gold when alloyed with 1 per cent of aluminium is white, but on the addition of more aluminium the colour gradually deepens, and with 22 per cent alumi- nium an intensely ruby-coloured alloy, AuAl 2 , is formed. Copper-zinc alloys show great variations in colour, according to the relative proportions of the two metals. See 23. Ledebur arranges the colouring power of metals in the follow- ing order : Tin, nickel, aluminium, manganese, iron, copper, zinc, lead, platinum, silver, gold. 2 Each of the above metals has a greater decolourising action than the metals following it. Most metals are capable to some extent of existing in a state of chemical combination with each other, but, as a general rule, they are united by feeble affinities, for it is necessary, in order to produce energetic union, that the constituents should exhibit great dissimilarity in properties. It is probable that the metals do unite in definite proportions, but it is difficult to obtain these compounds in a separate con- dition, since they dissolve in all proportions in the melted metals, and do not generally differ so widely in their melting points from the metals they may be mixed with, as to be separated by crystallisation in a definite condition. For these reasons it has been questioned whether alloys are true chemical compounds. Definite compounds do, however, exist in definite proportions by weight in the native as well as in the artificial state. Such is the case with mercury and silver, which are found crystallised together in the proportion of one atom of silver to two or three atoms of mercury. 1 Proc. Roy. Soc. 1891. 2 Die Legierungen, by Ledebur. i NATURE OF ALLOYS 43 A good illustration of chemical combination between two metals is seen in the alloy of copper and tin, which may be represented by the formula SnCu 2 , containing 38'4 parts of tin and 6T6 parts of copper. It is distinguished by its peculiar colour, its homogeneity after repeated fusions, its brittleness, and by having a greater density than any other alloy of these metals. As a general rule, it may be stated that all metals which unite with oxygen to form only bases have a strong tendency to unite with metals some of whose oxides possess an acid character ; and those metals which are allied in regard to basicity or acidity, have little affinity for each other. Thus sodium and potassium, although miscible in various pro- portions, exhibit little or no tendency to unite in definite quantities ; and the same is true of antimony and tin. On the other hand copper and (tin form very stable alloys. In some instances, as in the case of lead and zinc, only very unequal portions can be made to unite, the main bulk of the metals separating on cooling; the lead retaining 1*6 per cent of zinc, and the zinc retaining 1-2 per cent of lead. In most cases of mixed metals there is not that total and complete alteration in properties which is the distinctive feature of chemical action between a metal and a non-metal. Alloys still possess the metallic character, but there is frequently a considerable evolution of heat evinced ; the melting points, as in fusible metal, are considerably lowered ; the mean density is increased, and the colour and other physical properties are considerably modified. Most often, however, as already indicated, alloys seem to be mixtures of definite compounds with an excess of one or other metal, and the separation of their components from each other is generally easily effected by simple means. Thus an alloy of lead and copper may be largely- separated by exposing the mixture to a temperature above that of the fusing point of lead, but below that of copper, when the lead liquates out, leaving behind a porous mass of copper, containing a little lead. Also in re-melting brass, a considerable quantity of 44 MIXED METALS CHAP. zinc is lost by volatilisation. When silver is amalgamated with mercury the amalgam is dissolved in an excess of mercury, which excess is removed by simple pressure ; and the remaining portion of mercury is completely separated by the agency of heat. Many metals combine together when melted, and only remain in union within certain ranges of temperature, as shown by the wide differences between their melting and solidifying points. Chemical combination is well illustrated in the formation of sodium-amalgam. If a little clean sodium is thrown into mercury, especially when heated, so much heat is developed that some of the mercury is vaporised. By dissolving the amalgam in water and acids, and deducting the heat of the solution of sodium, Berthelot found that for each atom of the sodium in amalgams, containing a large excess of mercury, the amount of heat evolved increases with the sodium, and also that when a composition is reached which approaches the compound NaHg 5 , the amount of heat decreases. The compound of sodium with hydrogen Na 2 H has all the appearance of a metal and is a most instructive example of the characteristics of alloys, which exhibit a close analogy to the class of indefinite compounds, termed solutions. In the solution of substances in water, for example, there proceeds a peculiar kind of indefinite combination ; there is formed in fact a new homogeneous substance from the two substances taken, but the bond connecting the substances is very unstable. Solutions are regarded by Mendeleef l as fluid, unstable, definite chemical compounds in a state of dissociation. Of such a kind are most metallic alloys. They may be con- sidered as solidified solutions of metals which contain definite compounds in an excess of one of the constituent metals. 2 According to some experiments by Laurie, the copper-zinc alloys, in respect to the electro-motive force in galvanic 1 Mendeleef s Chemistry, vol. ii. 123. 2 Jour. Cliem. Soc. 1888, p. 88, 1889, p. 667, 1894, p. 1030. i NATURE OF ALLOYS 45 batteries, behave just like zinc if the proportion of copper in the alloy does not exceed a certain percentage that is, until a definite compound is attained for then there are yet particles of free zinc ; but if a copper surface be taken and it be covered with only one-thousandth part of its area of zinc, then only the zinc will act in the battery. Similarly, in the case of copper-tin alloys, a sudden rise of electro- motive force is observed when the proportion of the tin in the alloy exceeds that of the compound corresponding to SnCu 3 . If an alloy containing a larger amount of tin than that in SnCu 3 , in a fine state of division, be placed in a copper cup and used in the place of the zinc in a copper chloride cell, the excess of tin is gradually eaten out, leaving approximately the alloy SiiCn 3 . Solutions may be cooled several degrees below their freezing points before solidifica- tion occurs, but mechanical agitation at once causes freezing. It may be incidentally remarked that a minute variation in composition sometimes alters the melting point of an alloy. Raoult has" proved that one molecular proportion of any .substance dissolved in 100 molecular proportions of any solvent lowers the freezing point of that solvent by a definite amount. Heycock and Neville 1 have shown that the temperature of solidification of molten tin is lowered by the presence of a small quantity of other metals in proportion to the concentration of the solution. The following metals in atomic proportion were dissolved in 11,800 parts of tin, and the reductions of temperature were Zinc, 2'53 ; copper, 2-47 ; silver, 2-67; cadmium, 2-16 ; lead, 2-22; mercury, 2-3 ; antimony, 2 ; aluminium, 1'34. The last is therefore exceptional. Dr. Alder Wright 2 investigated the subject of solution of metals in metals by a series of experiments ex- tending over several years. Of the 36 pairs which it is possible to form with the nine metals, lead, bismuth, alumi- nium, zinc, tin, silver, antimony, cadmium, and copper, the 1 Jour. Chem. Soc. p. 666, 1889. See Chemical News, Dec. 1892. - Proceedings Roy. Soc. 1889-93. See also Roberts- Austen's In- troduction to Metallurgy, p. 89. 46 MIXED METALS CHAP. great majority have been found to possess the property of completely blending with one another so as to form a homo- geneous fluid, stable for many hours when heated to such a temperature that the whole mass remains liquid. On cool- ing segregation occurs in many cases during solidification. Five pairs are, however, exceptional when the components lie within the undermentioned limits. When well mixed and allowed to stand several hours at an equable temperature, two alloys separate from each other, thus : At about 650" C. . between {Pb = <;76 and {% = Sn0 o r /Pb = 98-70 /Pb= 1-57 U ' ' " \Zn = 1-30 " \Zn = 98-43 flKAop /Bi =85-72 /Bi = 2-32 " \Zn=H-28 " \Zn=97'68 o nn o r (Bi = 84-17 /Bi = 2'52 " \Zn = 15-83 " \Zn = 97'48 /Pb = 99-93 /Pb = 1-91 JA1 = -07 " \A1 = 98-09 fBi =99-72 fBi= 2'02 \A1= -28 \A1=97'98 7Fl0 o p /Cd = 99'78 /Cd= 3'39 " Ul= -22 " | Al =96-61 Matthiessen regards it as probable that the condition of an alloy of two metals in the liquid state may be either that of (1) a solution of one metal in another ; (2) chemical combination ; (3) mechanical mixture ; (4) a solution or mixture of two or all of the above ; and that similar differ- ences may obtain as to its condition in the solid state. He also classifies the solid alloys composed of two metals accord- ing to their chemical nature. 1. Solidified solutions of one metal in another : lead-tin, cadmium-tin, zinc- tin, zinc- cadmium, and lead-cadmium alloys. 2. Solidified solutions of one metal in the allotropic modi- fication of another : lead-bismuth, tin-bismuth, tin-copper, zinc-copper, lead-silver, and tin-silver alloys. 3. Solidified solutions of allotropic modifications of the I NATURE OF ALLOYS 47 metals in each other : bismuth-gold, bismuth-silver, palladium- silver, platinum-silver, gold-copper, and gold-silver alloys. 4. Chemical compounds of the alloys corresponding to Sn 5 Au, Sn 2 Au, and Au 2 Sn. 5. Solidified solutions of chemical compounds in each other : the alloys intermediate between those corresponding to the above formulae. 6. Mechanical mixtures of solidified solutions of one metal in another : alloys of lead and zinc containing more than 1-2 per cent of lead, or 1 -6 per cent of zinc. 7. Mechanical mixtures of solidified solutions of one metal in the allotropic modification of another : alloys of zinc and bismuth containing more than 14 per cent zinc, or more than 24 per cent bismuth. 8. Mechanical mixtures of solidified solutions of allotropic modifications of the two metals in each other : most of the silver-copper alloys. 1 As before stated, many mixtures, the components of which may be united in the liquid condition, separate to some extent on cooling, forming two or more alloys of different densities and of different fusibilities. Or, many solid alloys, when heated, release certain of their constituents that have different melting points from the remainder. This is termed "liquation." The term is also used to express the change that occurs when certain components of alloys or of ores are actually separated by heat and allowed to flow away. Ex- amples of this are seen in the case of the purification of tin, and in the extraction of silver from argentiferous copper by the agency of lead. A good illustration of liquation is the almost complete separation of lead from zinc, when the two metals are melted together and then allowed to cool. In fact, a cooling mass of mixed metals often behaves like water containing suspended matter does in freezing, when the ice first formed rejects the foreign matter ; and so the portion of the alloy which first solidifies rejects certain other portions of the constituent metals. Thus a mixture of lead, antimony, 1 Matthiessen, Brit. Assoc. 1863. 48 MIXED METALS CHAP. and copper poured into a cylindrical mould, allowed to cool, and then broken, will show that while the copper and anti- mony have united the lead has been rejected, and driven to the centre of the mass. Silver and copper alloys behave in a similar manner, but in any mixture of fused silver and copper one alloy is formed, which is driven inwards or outwards according to whether copper or silver is in excess in the bath. In all cases the separation is never complete, a small quantity at least of the other metal being found in each portion of the separated constituents. The solid mass in the above cases is a mixture of solidified solutions of the metals in each other. 1 A mixture of lead and zinc largely separates accord- ing to gravity. Levol has shewn that, in cooling, alloys of lead with gold and silver, if the latter are present in small quantities they are driven towards the centre of the solidifying mass. Gold of high standard does not, like silver, shew any tendency of the constituents to undergo liquation. It is well known that in gold-platinum alloys certain portions of the con- stituents separate, and become concentrated, either in the centre, or in the external portions of the solidifying mass. 2 Iron and steel exhibit a marked tendency to undergo liqua- tion when slowly cooled. Lencanchez 3 heated fragments of pig-iron to a temperature of 940 C. for 100 hours and found that a number of spherical grains had extruded, which on analysis were found to contain 4 to 6 per cent phosphorus, whereas the original pig-iron contained only T9 per cent. By applying severe pressure to partly solidified cast-iron Mr. Stead caused the molten interior, into which much phos- phorus had penetrated, to burst the tender walls and gush out. This ejected matter contained 6'84 per cent of phos- phorus, while the cast-iron treated contained only 1-53 per cent. The causes of the separation of certain constituents of steel is attributed by Howe to the struggle between crystalline force and surface tension, aided by gravity ; on the one hand 1 Roberts- Ansten, Journal of Society of Arts, 1888, p. 1115. 2 Proc. Roy. Soc. Feb. 1890.' 3 Mem, Soc. Ing. Civils, 1889, i NATURE OF ALLOYS 49 tending towards differentiation, and on the other towards diffusion. As the temperature sinks towards the freezing point, the surface tension increases, the different constituents tend less powerfully to diffuse among each other, and more to draw apart in drops. Further, as the complex molten mass cools past the freezing point of a certain potentially present compound, this compound tends to form, to solidify, to crystallise, to expel the more fusible residue, in the same way as a salt in crystallising expels the mother liquor, which is gradually driven towards the last freezing region. As time is required to effect any considerable separation, we may infer that slow cooling favours separation. 1 Dr. Guthrie gave considerable attention to this subject some years ago, and came to the conclusion that certain alloys in cooling behave as a cooling mass of granite would ; clear molten granite in cooling would throw off " atomically definite " bodies, leaving behind a fluid mass, which is not definite in composition, as the quartz and felspar solidify before the mica. The same thing takes place in cooling fused mixed metals ; for when a mixture of lead and bismuth or bismuth and tin cools, a certain alloy of the metals falls out, and the most fusible alloy of the series is left, which Guthrie calls the eutectic alloy. Although it is the most fusible alloy of the series, the proportions between the constituent metals are not atomic proportions ; and Guthrie says " that the preconceived notion that the alloy of minimum temperature of fusion must have its constituents in simple atomic propor- tions, and that it must be a chemical compound, seems to have misled previous investigators ; but that certain metals may, and do unite with one another in small multiples of their combining weights may be conceded ; the constitution of eutectic alloys is not in the ratio of a simple multiple of the chemical equivalents of the constituents, but their com- position is not on that account less fixed, nor are their pro- perties the less definite." 2 1 Metallurgy of Steel, p. 207. 2 Phil. Mag. June 1884, p. 462. E 50 MIXED METALS CHAP. Freezing Points of Alloys. 1 The study of the freezing points of alloys is destined to furnish very precise informa- tion on the constitution of alloys in consequence of the very definite knowledge which we possess upon the freezing points of similar mixtures, such as solutions of salt and water, mixtures of salts between themselves, and mixtures of organic compounds. If we can extend to fusion the well- known laws of solution, and if we apply these results to alloys, we may, from the curve of fusibility, determine the constitution of the alloys, by recognising, whether, after solidi- fication, they are constituted by the simple juxtaposition of 350 Lead 300 1250 200 150 .00 in E tect 2O 3O 4O SO 6O Lead % too 90 eo TO eo so 40 FIG. 1. 70 8O OO \OO%Tin 30 20 10 o crystals of the two metals, or whether they are formed of definite combinations, or whether they are isomorphous mixtures. Thus : 1. The two metals give neither definite combinations nor isomorphous mixtures. The curve of fusi- bility is then composed of two branches only, as in the case of mixtures of sodium-chloride and sodium-carbonate. Similar curves have been obtained for a certain number of alloys, such as lead-tin, tin-zinc, tin-bismuth, etc. (Fig. 1). 2. The two metals yield one or several definite combinations. The curve of fusibility is then formed of a certain number of dis- tinct branches which meet at an acute angle, as in the case of solutions of sodium-sulphate. These curves present maxima 1 Boudouard, French translation of present work. NATURE OF ALLOYS 51 which may be considered as certain proofs of definite com- binations, but the opposite would not be true. Similar curves have been obtained with the alloys copper-tin, cop- per-antimony, and copper-aluminium (Fig. 2). 3. The two metals form isornorphous mixtures. The curve of fusibility is then continuous, and approximates more or less to the straight line which joins the points of fusion of the two separate bodies, such as gold-silver alloys. The general pro- Antimony Copper 10 20 30 40 50 60 7O 8O 9O >OO%Cu FIG. 2. perties of the preceding curves of fusibility give an explana- tion of the phenomenon of liquation. At the moment when solidification commences there is deposited either a pure metal or a definite compound ; generally this separation of a definite substance tends to change the composition of the remaining liquid portion, and the separation can only con- tinue if the temperature is being lowered. Theoretically, this would manifest itself by a total difference of composition in different parts of the ingot after solidification. One or other of the metals being at the one extremity and the alloy of lowest freezing point at the other : that is, the eutectic alloy. Practically these differences of composition in dif- 52 MIXED METALS CHAP. ferent parts of the ingot, as determined by chemical analysis are generally very small, in consequence of crystallisation of films or very fine needles which form a network in the still liquid portion, and so entangle each other as to render their separation according to their densities impossible. In the case of alloys yielding definite compounds, there will be generally two eutectic alloys, each intermediate between the definite compound and one of the pure metal?, as in the case of copper-antimony alloys. If there are two definite compounds there will be three eutectic alloys, as in the case of copper-aluminium alloys. Isomorphous alloys solidify either with a composition identical with that of the molten alloy, and thus at constant temperature, or they solidify with a different composition, and then the molten portion approximates to the composition of a eutectic alloy. Kesearches have been made by M. H. Gau- tier on antimony-aluminium, nickel-tin, nickel-copper, sil- ver-aluminium, silver-antimony, silver-zinc, silver-cadmium, and silver- tin alloys. Other researches have been made by M. Roland-Gosselin, including tin-antimony, tin-aluminium, tin-lead, tin-zinc, tin-bismuth, bismuth-copper, bismuth- zinc, zinc-aluminium, zinc-cadmium, and copper-lead alloys. The above experimenters infer from their results that alloys do not at all resemble glass, as some have stated. They are crystalline bodies, either formed by the juxtaposition of the crystals of the constituents, as in the case of tin-zinc, tin- bismuth, tin -lead, lead -antimony, cadmium -zinc, zinc- aluminium, and antimony-silver alloys ; or by the juxtaposi- tion of crystals of one of the metals with a definite compound. This is the case with the alloys tin-copper, antimony-copper, alu- minium-copper, lead-copper, bismuth-copper, tin-aluminium, tin-nickel, copper-nickel, zinc-antimony, aluminium-silver, and antimony-aluminium alloys. Others again, much more complex, would be composed of isomorphous mixtures, either the metals being really isomorphous, such as bismuth-anti- mony and silver-gold ; or they form isomorphous compounds with one of them. Such appears to be the case with zinc- i NATURE OF ALLOYS 53 copper, zinc -silver, tin -silver, cadmium -silver, and tin- antimony alloys. Very often the addition of a small quantity of a metal to another more fusible one lowers the point of fusion of the latter and nearly always the melting point of alloys is below that of the less fusible of the two constituents ; the exceptions being certain alloys of gold and aluminium, of antimony and aluminium, and certain sodium alloys. The solidification of a molten alloy always com- mences at the same temperature for a definite composition. Omitting the case of isomorphous mixtures, there is deposition during solidification either of a pure metal or of a definite compound, which causes a variation of the composition of the still liquid part, and the solidification cannot continue unless the temperature is further lowered. It therefore follows that the temperature does not remain constant during the complete solidification. Moreover, if the body which is deposited has a density different from that of the liquid, it will either pass to the top or the bottom of the vessel. Soli- dification, however, occurs at a constant temperature in two cases 1, where the alloy has a composition corresponding to a definite compound, in which case the deposition of this comr pound does not modify that of the remaining liquid ; 2, where its composition corresponds to an angular point of the curve of fusibility, that is to say, an eutectic alloy. In this case the two metals corresponding to the branches which cut one another are deposited simultaneously, and the composi- tion of the liquid bath still remains constant As to iso- morphous mixtures, they solidify either at a constant temperature or at a progressively decreasing temperature. Lastly, if the crystalline structure of alloys appears most often on the fractured surface, or, if it is easy to develop by polishing and etching, it is not a case of eutectic alloys or isomorphous mixtures, as the crystals of these alloys are of extreme fineness and give to the fractured surface an appear- ance similar to that of glass. " Matthiessen considers that the conductivity for heat and electricity is among the characters best calculated to indicate 54 MIXED METALS CHAP. the chemical nature of alloys. With respect to electric con- ductivity, he divides metals into two classes : " A. Metals which, when alloyed with each other, conduct electricity in the ratios of their relative volumes lead, tin, zinc, and cadmium. " B. Metals which, when alloyed with each other, or with a metal of class A, do not conduct electricity in the ratios of their relative volumes, but always in a lower degree than that calculated from the mean of their volumes bismuth, antimony, platinum, palladium, iron, aluminium, gold, copper, silver, etc. tot 90 40 30 20 _ J ~~! -"-"-" ~--~' ___ - 10 10 20 30 40 50 60 70 80 90 WO Lead Volumes per cent FIG. 3. " The curves representing the conductivity of different series of alloys have the relation shown in the accompanying diagrams. "Group I. Those belonging to the alloys of metals in class A are almost straight lines. That of lead-tin alloys is given as a type, Fig. 3. " Group II. The curves of alloys of metals in class B show a rapid decrement on both sides of the curve, the turning points being connected together by nearly straight lines. That of gold-silver alloys is given as the type, Fig. 4. NATURE OF ALLOYS 55 " Group III. The curves of alloys of metals in class A with those in class B show a rapid decrement on the side Sit Cop WO 10 20 30 40 50 60 70 80 90 700 Gold 22 Carat Gold with Silver 22 Carat Gold with Copper Volumes per cent FIG. 4; beginning with the metal belonging to class B, then turning and going in a straight line to the other side, beginning with metal belonging to class A. That of tin-copper alloys is given as the type, Fig. 5. Co^pei Brass i Metal 20 10 20 30 40 50 60 70 80 90 100 Zinc Tin Volumes per cent TIG. 5. r tu, o. "In regard to alloys of the first group, if they were mechanical mixtures, the metals composing them, unless 56 MIXED METALS CHAP. their specific gravities were the same, would separate into two layers when melted and slowly cooled, as in the case of lead-zinc alloys. But the alloys of lead and tin, for example, do not separate in the same way as lead and zinc. More- over, homogeneous wires could not be obtained by pressing, if these alloys were mechanical mixtures ; but wires of the same alloy have been proved to have the same conducting power, whether taken from the press at the beginning or end of the operation. " On the other hand, the agreement between the theo- retic and actual conductivity of these alloys, as well as between the calculated and actual percentage decrement in conductivity between and 100 C., indicates that, in the solid state, they are not chemical compounds. In regard to these particulars, the following law has been found to obtain for all alloys of the first and second groups, as well as for some of those belonging to the third group : " The actual percentage decrement in conductivity between and 100 C. is to the calculated decrement, as the actual is to the calculated conductivity. "Among the alloys of the second group, some may be regarded as mechanical mixtures. Silver and copper fused and well stirred together separate when slowly cooled, so that the mass contains different amounts of the metals at different parts. But these alloys are exceptional, and most alloys of this group may be regarded as solidified solutions of allotropic modifications of the metals in each other. " In the third group of alloys, the rapid decrement in the conductivity of those alloys of the several series, which contain but very small amounts of a metal belonging to class A, cannot be ascribed to the existence of chemical compounds of the metals. For, in the first place, the amount of one of the metals in the alloys corresponding to the turning-points of the curves representing the conductivity of the series is too small, as will be seen by the following instances : NATURE OF ALLOYS 57 Alloy. Percentage. Bismuth-tin Tin .... '6 Bismuth-lead Lead . . . 2'0 Silver-tin Tin . . . . 2'6 " Again, the great similarity of the curves representing the conductivity of the series of alloys belonging to this group is opposed to the existence of chemical compounds in the solid alloys. " The influence exercised upon the conductivity of metals by the presence of small quantities of other metals, does not appear to be in any way determined by the alteration of crystalline form, or tendency to crystallise, which are known to be influenced by that circumstance. " If it be assumed that the metals belonging to class B undergo a molecular change when alloyed with one another, or with metals belonging to class A, and that in each case an allotropic condition is induced by a small amount of other inetals, varying with the different metals, then many of the phenomena characteristic of alloys may be explained. Thus the curve representing the conductivity of zinc-copper alloys has the same form as those of other alloys, belonging to the same group, and the percentage decrement in their con- ductivity between and 100 C. is exactly what is indicated by the law above stated. Hence it may be inferred that solid alloys of zinc and copper are only solidified solutions of zinc, and of allotropic copper in each other. The different action of reagents upon alloys, and upon the metals con- stituting them, when in an isolated state, may also be referred to the existence of such allotropic modifications when they are alloyed, as well as to the existence of chemical compounds of the metals." l 13. Electrolysis of Alloys. Several attempts have been made to separate the constituents of alloys in a molten state by means of the passage of an electric current, in a 1 Watt's Diction, of vhem. vol. iii. pp. 943, 944 ; or Phil. Trans. Roy. Sac. 1860, p. 161. 58 MIXED METALS CHAP. similar way to that of saline solutions, but all efforts appear to have been unavailing. The electrical resistance of alloys increases as the temperature rises, except for metals which undergo a molecular change before fusion. In the case of molecular change in alloys, the effect is clearly indicated by a change in the resistance. Professors Dewar and Fleming have determined the conductivity of alloys at very low temperatures, and found that many of those alloys that are poor conductors and therefore offer much resistance to the passage of an electric current at ordinary temperatures, have the same property but little diminished at - 1 00 to - 200 C. This is the reverse to that of pure metals when unalloyed, as these decrease in resistance, with a lowering of temperature, to such an extent as to suggest that if the abso- lute zero of temperature could be reached the resistance would totally vanish. Diffusion of Metals. The subject of diffusion of metals in metals has been closely studied by Roberts- Austen. 1 The diffusion of molten metals is proved by placing metals, such as gold or platinum, at the bottom of a tube and filling the remainder with molten lead. These were placed in a bath or oven and heated to about 500 C. In twenty-four hours a quantity of solid gold or platinum had diffused into the molten lead. After allowing the metal to solidify, it was cut into sections and analysed. Other metals such as tin and bismuth were .also tested, and the diffusibility in cubic centimetres per day was found to be as follows : Gold in lead Gold in bismuth Gold in tin Silver in tin Lead in tin Rhodium in lead Platinum in lead Gold in lead Gold in mercury 3-19 at 550 C. 4-52 4-65 4-14 3-18 3-04 1-69 at 490 C. 3-03 0-72 at 11C. 1 Nature, May 21, 1896. I NATURE OF ALLOYS 59 It will be seen from the above that gold diffuses more rapidly into molten tin and bismuth than into lead. The heat was applied so that the top of the tube was hotter than the lower portion. Even after the lapse of an hour the lead in the upper part was found to contain gold. The rela- tively slow rate at which platinum diffuses into lead, as com- pared with gold, points to its having a more complex molecule than the latter. Some physicists hold what may be termed the gaseous theory of metallic solutions. It supposes that when the amount of added element is very small, its atoms are so widely separated by dilution that they act with no more mutual constraint than would the atoms of a gas. Experiments have been made on the diffusion of solid metals by placing gold at the bottom of a solid cylinder of lead and keeping the temperature uniform for thirty-one days. The following results in square centimetres per day were obtained : Diffusibility of gold in fluid lead at 550 C. . 3 '19 ,, solid lead at 251 C. . 0'03 200 C. . 0-007 ,, ,, 165 C. . 0-004 100 C. . 0-00002 If clean surfaces of lead and gold are held together in vacuo, at a temperature of 40 C. for four days, they will unite firmly, and can only be separated by a load equal to one- third the breaking stress of the lead itself. The act of carbon in uniting with iron in the process of cementation may be taken as an evidence of solids diffusing with each other. Metals and alloys absorb gases, the most common of which is oxygen, and as this gas is a constituent of the atmosphere and readily unites with metals, while the nitrogen does not, its influence is much more marked. R. H. Greaves has shown that the presence of oxygen in a metal tends to harden it, and up to a certain limit to increase the tenacity but to lower the ductility. In the case of copper, the addition 60 MIXED METALS CHAP. of small amounts of antimony or arsenic improves these properties, increasing the ductility and diminishing the hardness. In all cases, whether in the presence of arsenic or in pure copper, the effect of adding oxygen is to diminish the Conductivity, while in the presence of antimony the addition of a little oxygen raises the Conductivity. STRENGTH OF ALLOYS AT HIGH TEMPERATURES The influence of temperature on alloys is of great industrial importance, especially since the extended application of super- heated steam. As a rule an increase of temperature lowers the tenacity, and that in proportion to the duration of the exposure. Some alloys undergo a molecular change in their crystalline character which is accentuated as the temperature is raised. This change may simply be in the size of the crystals, or it may be a change of crystalline form. This alteration requires time and explains the influence of duration of exposure, so a mechanical test made at high temperatures is not sufficient to indicate the behaviour of the metal in ordinary practice. Gun metal sustains a considerable loss in tenacity above 200 C., while phosphor bronze only slowly deteriorates up to 250. A. Le Chatelier showed that a bronze containing 10 per cent tin and 3 per cent zinc had an increased tenacity and elongation at 140, and rapidly diminished above 200. Brass (70-30) remains nearly constant up to 160, and above that gradually diminishes in tenacity and elongation. Aluminium bronze has a sharp loss of mechanical properties above 80 ; Le Chatelier found that a 9 per cent aluminium bronze has a tenacity of 25 tons at 15 ; 20 tons at 117 ; and 11 tons at 180. With brass (60-40) the tenacity slowly diminishes and the elongation increases up to 200, but above that both the tenacity and elongation rapidly diminish. One important application of alloys employed for high temperatures is for firebox plates and firebox stays. Such a i NATURE OF ALLOYS 61 metal must be capable of being easily worked and riveted. It is subjected to a temperature of 200 and upwards. The best alloy for this purpose appears to be Cupro-Manganese (Cu 84-86, Mn 6-4) as long as it is not brought in contact with flame. It is very good for bolts. The best all-round material for plates seems to be Copper with 0-5 per cent of arsenic, and one of the worst is aluminium bronze, which undergoes a crystalline growth on heating. MICROSCOPIC EXAMINATION OF METALS AND ALLOYS 14. Micro-Structure. The prominence given to the microscopical examination of metals is mainly owing to the facts communicated by Dr. Sorby some forty years ago, and since that time this method of investigation has been taken up by numerous other workers, not only in relation to the metallurgy of iron and steel, but also with regard to alloys, and impurities in metals. Among other workers in micro- scopy as applied to metals may be mentioned the names of Osmond, Martens, Wedding, Charpy, Arnold, and Stead. As metals cannot be obtained in sufficiently transparent sections, they have to be viewed by reflected light For this purpose it is essential that they should be perfectly smooth, which is effected by grinding and polishing. Dr. Sorby's method of preparing sections consists in cutting slices about J^ of an inch thick, and fixing them on to a glass plate by means of Canada balsam. The upper surface is first ground by rubbing on emery paper, and finishing on the very finest quality of emery paper. The surface is next rubbed on Water-of-Ayr stone until all scratches are removed. It is then polished on wet cloth stretched on a piece of wood, using, in the first place, the finest grained crocus, and, kstly, the finest rouge, so as to give a high degree of polish. Some of Dr. Sorby's specimens were polished dry on parchment covered with rouge. When the polishing is complete the surface is etched with very dilute nitric acid, in which it is placed for a few seconds, then swilled in hot water and examined under the 62 MIXED METALS CHAP. microscope. If the etching has not been sufficient, the acid treatment is repeated. Instead of nitric acid, a mixture of equal parts of sulphuric acid and potassium bichromate in ten times their weight of wafcer, may be used. 1 Polishing with hard metals can be satisfactorily done and a fairly smooth surface obtained, even by a beginner, but with soft alloys much greater skill is required. Polishing alone often suffices to show the structure of coloured alloys, but etching is generally necessary. If a metal is being subjected to gradual mechanical treatment and specimens examined at various stages of the process, the etching liquid must be the same in each case, but in determining the constituents, of an alloy it is useful to employ various reagents. The etching liquids usually employed are acids (hydrochloric, sulphuric, nitric), potash or soda/ ammonia, alkaline sulphides, soluble chlorides, etc. Acids must be made very dilute with water, otherwise the action will be too rapid and violent. In some cases simple heating in air will suffice, so as to impart a film of oxide, which varies with the different constituents. Sulphur- etted hydrogen is sometimes very useful. Lastly, the electro- lytic cell may be used, making the metal the positive electrode in a saline solution. One constituent may be attacked and rendered dull, while another may remain bright. Of course only a feeble current must be employed in most cases. If it is desired to develop crystals of the same kind a stronger current is advisable. Ammonium' nitrate, sodium chloride, and sodium thiosulphate solutions yield good results, according to the alloy to be treated. Any microscope fitted with powers varying from 30 to 200 diameters is sufficient, higher powers only being requisite in extreme cases. Ordinary oblique illumination may be used, but in most cases by far the best results are obtained in viewing the objects in direct light, and then a vertical illuminator becomes necessary. Beck of London and most other makers supply such an appliance at about 15s. It consists of a circular disc of glass fitted in a socket which can 1 See Journal of Iron and Steel Institute, No. 1, p. 294, 1894. i NATURE OF ALLOYS 63 be screwed on to the bottom of the tube, above the objective. Instead of the above reflector, a right angled prism may be advantageously used and placed immediately below the eye-piece. A great deal of valuable information may be gained by placing the section at an angle of 45 on the stage of the microscope, and dispensing with the above- mentioned illuminator. Only a small part will be in focus, but it will often suffice to give a good idea of the general structure. If low powers only are used then excellent results may be obtained by fixing an ordinary .cover glass, or a thin disc of mica, between the objective and the section to be examined. This answers as well as an internal reflector. Mr. Stead has studied the effect of bending or otherwise deforming a polished section. On examination under the microscope the cracks thus produced enable one to recognise the constituents which are a cause of weakness and their mode of intervention. He has applied this method to lead- antimony, tin-antimony, and tin-arsenic alloys. The author has also found it very instructive with copper containing a little bismuth. From the microscopic study of iron it appears to be con- stituted, like crystalline minerals, of different bodies which have different crystallising temperatures. Take the case of the meteoric iron. 1 It appears to contain three separate components. The dominant constituent seems to have crystallised first in strongly marked thin meshes ; between these meshes the second constituent has crystallised. The third constituent, which is a phosphide of iron and nickel, occupies a position between the other two. Howe has proposed names for the crystallised bodies of which iron and steel appear to be made up when viewed through the microscope. Thus, in Fig. 6 the iron, independ- ent of the black patches of slag, is of nearly uniform character, to which he has given the name of ferrite. It consists of crystallised grains. Fig. 7 is a specimen of high J See Journal of Iron and Steel Institute, No. 1, 1887. 64 MIXED METALS CHAP. Fio. 7. NATURE OF ALLOYS 65 carbon steeL It consists of crystals of pearlite and an intensely hard compound of iron and carbon, which has been termed cementite. Pearlite, Fig. 8, is assumed to be made up of ferrite and cementite, and contains 0*9 per cent of carbon. Moderately hard steels consist largely of pearlite. Hardenite is another constituent of steel containing iron and up to 3 per cent of hardening carbon. The following examples will show the nature of some cop- Fio. 8. per alloys. Fig. 9 is brass containing 70 per cent copper, and is characterised by dendritic masses, formed by long needle-shaped crystals arranged in groups at right angles to each other, but too imperfectly formed to measure individual angles. The size of these crystalline grains varies with the rate of cooling, being smaller as the cooling is quicker. Fig. 10 contains 60 per cent copper, and is distinguished by curved crystals interwoven together, but not in groups at right angles to each other. Fig. 11 is same as Fig. 9, but has been burnt in annealing, and a highly crystalline structure Fio. 9. Fio. 10. FIG. 11. Fio. 12. Fio. 13. Fio. CHAP. T SLAGS 67 therefore induced. Fig. 12 is a sample of gun metal, with 90 per cent copper. Fig. 13 is bell metal, with 80 per cent copper. Fig. 14 contains about 62 per cent copper, and is almost entirely made up of the compound SLAGS, FLUXES, AND REFRACTORY MATERIALS 15. Slags. The slags formed in metallurgical operations are, with few exceptions, produced by the union of silica with metallic oxides, and termed silicates. Silicates may be divided into two classes, viz. "hydrated," which contain water; and "anhydrous," which are free from water. The silicates produced by heat in metallurgical operations belong to the latter class. The bases most often found in slags from smelting ores are Lime, magnesia, oxide of iron, and alumina. Both the silica and bases are derived from the earthy matter of the ore, the ashes of the fuel, the flux, and the material of which the furnaces or vessels are constructed. In melting and mixing base metals to form alloys, some of these elements oxidise and pass into the slag, and often form the chief bases when united with silica or other acid substances. The silicates most frequently produced are those with an iron base, of which the most fusible is the protosilicate 2FeO . Si0 2 . The protosilicate of manganese appears to have about the same fusing point as the protosilicate of iron ; it has an olive-green colour, is opaque, and but slightly crystalline. It does not absorb oxygen in contact with air like the silicates of iron, and cannot therefore be employed as an oxidising agent. A double or multiple silicate con- taining two or more bases is often more fusible than a single silicate ; hence a compound silicate of lime, magnesia, alumina, etc., is more fusible than either of these bases combined alone with silica, and melts readily when strongly heated. Likewise, oxide of manganese augments the fusibility of earthy silicates. Oxide of zinc forms single silicates, which are practically infusible, and, as a general rule, this oxide diminishes the fusibility of silicates of iron, lime, etc. Protoxide of tin 68 MIXED METALS CHAP. .augments the fusibility of multiple silicates. The proto- and bi-silicates of copper melt easily : they augment the fusibility of earthy silicates. Oxide of copper is easily separated, especially in the presence of sulphur or arsenic, with which copper forms regulus or speise respectively. The oxides of lead and bismuth are, next to the alkalies, the most fluxing bases. Silicates of lead melt at a red heat. Oxide of lead increases the fusibility of multiple silicates, but it is easily reduced by the presence of metallic iron. Slags are either vitreous or stony, and not infrequently they are more or less crystalline. Rapid cooling tends to produce the glassy variety, and by slow cooling the crystalline structure is induced. The stony condition is prevalent in slags in which earthy bases predominate. The object in forming slags in melting metals is to form readily fusible compounds, which shall contain the impurities it is desir- able to remove, and at the same time be as free as possible from the metal or metalliferous substance undergoing pre- paration. When the slag is free from useful metals it is said to be clean ; but when a slag contains useful ingredients it is often advisable to remelt it with another charge, so as to extract valuable components, and render it sufficiently clean to be thrown away. It frequently happens that the slags produced in dealing with a given kind of metalliferous matter are useful as a flux or other agent in extracting or melting other metals. When the charge is improperly prepared, and. either silica or lime predominate, or when the amount of alumina or magnesia is very large, or the temperature insufficient, the slags are imperfectly melted, and present a kind of granular fracture. 16. Fluxes. A flux is a substance added to metalli- ferous matter, in order to unite with the foreign ingredients and form a fusible slag. The flux employed in any given case varies with the nature of the bodies to be removed. Thus, if the impurity be of an acid nature, such as silica, the i FLUXES 69 flux should be basic or neutral. If, on the other hand, a basic substance has to be separated, an acid flux will be generally required. In many cases of simple melting of metals it is desirable to remove foreign metals, which are present as impurities ; it is then advisable to add an oxidising flux in order first to oxidise the impurities, in which state they readily combine with an acid flux to form a liquid slag, or volatilise, or simply rise to the surface in virtue of their lower specific gravities. It frequently happens that a slag is formed at one stage of an operation, and decomposed again by the action of some reducing agent present. With regard to silicates, the most difficult to reduce are those with earthy bases, such as lime, magnesia, alumina, etc. When a slag contains several oxides, the most feeble bases will be reduced first, and the others in the order of their basicity. Suppose a compound silicate to contain the oxides of manganese, iron, tin, and lead ; the oxide of lead will be the first to be reduced, then the oxide of tin, then the oxide of iron, and lastly the oxide of man- ganese. The composition of silicates has a marked influence upon their consistence when melted. The bisilicates, or silicates with an excess of silica, pass gradually from the solid to the liquid state, or conversely from the liquid to the solid state ; they preserve for a long time a " viscous " or " plastic " consistency. The protosilicates or basic silicates, on the contrary, pass more or less quickly from the liquid to the solid state, and on cooling they set rapidly. They are very liquid when melted, and crack at the moment of solidifying. The most fusible silicates are formed by union of silica with fusible bases, such as soda, potash, oxide of lead, oxide of bismuth, etc. ; and the silicate formed is fusible in propor- tion to the amount of base present. With infusible bases, on the contrary, the fusibility is greatest when neither acid nor base is in excess. Baryta BaO is more fluxing than lime CaO, and magnesia MgO is less fluxing than lime. Alumina A1 2 O 3 is, among the common bases, the least fluxing. Fluor-spar CaF 2 has a wide and well-deserved reputation 70 MIXED METALS CHAP. as a flux. At a high temperature it melts into a transparent liquid. It is partly decomposed by silica, with the liberation of the gas silicon fluoride, but the greater part is found in the slag, in which it favours fusion and fluidity. When the bodies to be fluxed off are sulphates of lime, baryta, strontia, and bone-ash (phosphate of lime), fluor-spar is specially use- ful. A tenth part of fluor-spar is sufficient to liquefy sulphate of lead at a bright red heat. Sulphur is often present in metalliferous substances in the form of sulphides, such as silver sulphide, for example. In that case metallic lead may be employed, which decom- poses the sulphide of silver, forming sulphide of lead, and isolating the silver, which then unites with the excess of lead. In the same way iron decomposes lead sulphide, forming sulphide of iron and metallic lead. The metals of the alkalies and alkaline-earths are very effectual in removing sulphur from metallic sulphides. The metals are obtained for this purpose by reducing their oxides by carbonaceous matter in admixture with the metallic sulphide. In this case an alkaline sulphide is generally formed, together with an oxide of the metal. In some instances the decomposition of the metallic sulphide is incomplete. Certain sulphides, such as zinc sulphide, are rendered fusible by the addition of sulphide of iron, or iron pyrites. Many sulphides are fusible alone ; and in most cases it is easy to transform them into oxides by roast- ing in atmospheric air, or by fusion with an oxidising flux. When litharge PbO is melted with sulphides and arsenides, sulphurous and arsenious acids are volatilised, and an alloy of the metals with lead is formed. Other metallic oxides act in a similar manner, but, as a general rule, the sulphide is more or less converted into oxide, depending upon the amount of oxide used, and the relatively chlorous and basylous characters of the metals, and of sulphur and oxygen. Lead carbonate acts on sulphides in the same way as lead oxide. Lead silicate is also oxidising, but in a less degree, the slags produced containing both metals. Lead sulphate is more oxidising than litharge. When it is melted in the FLUXES 71 right proportion with sulphide of lead, the whole of the lead is reduced. Copper and iron sulphates are generally oxidising agents for sulphides. Nitrates, such as saltpetre KN0 3 , are the most powerful oxidising agents in melting operations. Basic iron slags are very active agents in promoting oxidation. For example, when iron containing carbon, silicon, sulphur, and phosphorus is strongly heated with basic silicate of iron these impurities are largely oxidised, forming carbonic oxide, silica, sulphurous acid, and phosphoric acid. Following is a list of the most common fluxes, with a brief account of their properties and uses : 1. Ammonium chloride (AmCl), called sal - ammoniac. This substance is decomposed by several metals forming metallic chlorides and liberating ammonia, which property is taken advantage of in purifying gold. A similar reaction occurs with several metallic salts. 2. Sodium chloride (NaCl), or common salt, is employed for preserving the substance beneath from the action of the atmosphere, and to moderate the action of bodies which cause violent ebullition. It melts and volatilises at a red heat in an open crucible, but requires a white heat to vaporise it in a closed vessel. When heated to redness with silica it forms a readily fusible silicate. It forms fusible compounds with antimony and arsenic, thus removing them from other metals during the process of refining. As the crystals decrepitate when heated, common salt should be powdered before using as a flux. 3. Borax (B 4 O 7 Na,). In the crystalline form it may contain 5 or 10 molecules of water, which are given off on heating, causing an enormous increase in volume, so that the vitrified form is much more suitable for assaying. It forms fusible compounds with silica, and nearly all bases, being especially useful in uniting with metallic oxides, sulphides, and arsenides. The commercial salt is adulterated with common salt and alum. 4. Sodium carbonate (Na C0 3 ) has the property of oxidiskig many metals, such as tin, iron, zinc, etc., by the 72 MIXED METALS CHAP. action of its carbonic acid, and as a consequence of this action it acts as a desulphuriser. It forms fusible compounds with silica and many metallic oxides ; it also melts at a moderate temperature, absorbing many infusible substances, such as lime, alumina, charcoal, etc. In some cases it acts as a reducing agent, as in the case of chloride of silver. When mixed with carbonate of potash a double salt is formed, which fuses at a lower temperature than either taken alone, a property very useful in the fusion of silicates, etc. 5. Potassium nitrate (KN0 3 ), also called nitre and saltpetre, is largely used as an oxidising agent. It fuses below redness, and at a high temperature is decomposed, yielding a large volume of oxygen, whereby the sulphur of metallic sulphides is converted into sulphurous acid, and the metals into oxides. Sodium nitrate acts in the same way. 6. Potassium bitartrate (THoKo), known also as cream of tartar or tartar. When pure this substance is white, but the variety chiefly used on the large scale is coloured, and sold as red " argol " ; this is cheaper and contains other carbonaceous matters, which give it greater reducing power than pure cream of tartar. This reagent is very valuable in operations requiring much carbonaceous matter. 7. Potassium chlorate (KC10 3 ). This substance is sometimes used with nitre as an oxidising agent, especially in assaying. 8. Potassium cyanide (KCN). This flux is valuable on account of the facility with which it fuses, and the readi- ness with which it reduces many metallic compounds when mixed with carbonate of soda. Common cyanide is preferable as a reducing agent because it contains carbonate of potash. 9. Calcium oxide (CaO), or lime, is used in the caustic state, or combined with carbonic acid in the form of carbonate. It is a useful flux for silica and silicates, and is also used to remove sulphur and phosphorus from metals and their com- pounds. 10. Calcium fluoride (CaF 2 ) or fluor-spar. This sub- stance acts as a flux in two ways (1) by combining with FLUXES 73 silicates, forming fusible compounds ; (2) by reacting with silicates and evolving the gas silicon fluoride SiF 4 . It forms fusible compounds with sulphates, such as plaster of Paris, and with phosphate of lime (bone-ash). It should be free from pyrites, blende, and galena, with which it is likely to be contaminated. 11. Lead oxide. There are two oxides of lead of im- portance in treating metals, viz. litharge PbO, and red-lead Pb 3 4 . Both oxides are reduced by carbon or hydrogen, producing metallic lead. Lead oxides, when melted, oxidise nearly all metals, except mercury, gold, silver, and platinum. With other oxides they form easily fusible compounds. When heated with sulphur, lead oxides are reduced and sulphurous acid is liberated. When oxide of lead in sufficient quantity is melted with an infusible silicate, a fusible double silicate is formed. 12. Manganese dioxide (Mn0 2 ). This substance is black in colour, opaque, and a good conductor of electricity. When heated alone it is infusible, but gives off oxygen, forming Mn 9 3 or Mn 3 4> according to the degree of heat employed ; heated with charcoal it is reduced to MnO. The facility with which it gives up oxygen makes it a valuable oxidising agent. With hydrochloric acid it is extensively used for generating chlorine. When strongly heated in a crucible lined with a paste of carbon it is reduced to the metallic state. 13. Silica (Si0 3 ). This body occurs in crystalline and amorphous forms ; it is white, infusible, except at the very highest temperatures, non-volatile, insoluble in water and acids, except hydrofluoric ; after ignition it is decomposed by carbon in the presence of iron, copper, or silver at a white heat, forming silicides of those metals. The amorphous and gelatinous varieties are slightly soluble in alkaline carbonates, but readily soluble in caustic alkalies. It combines with all the bases, forming silicates, and is therefore frequently employed to effect the fusion and separation of gangues in ores, the best forms to use being pure white sand and quartz. 74 MIXED METALS CHAP. 14. China clay is essentially a hydrated silicate of alumina, and when pure may be represented by the formula (2A1 2 O 3 , 3Si0 2 ) + 30H 2 : but clay generally contains sand and other silicates. It is white, and infusible in an ordinary furnace when heated alone, but readily unites with earthy and metallic gangues to form a fusible slag. 1 5. Glass is a mixture of silicates of sodium and potassium with some insoluble silicate, such as silicate of barium, mag- nesium, aluminium, iron, or lead. Being a compound silicate it fuses easily at a high temperature, and readily combines with lime and other bases containing little or no silica, so that it is often preferred to pure silica, and serves to economise borax. It is also employed as a covering in melting metals, so as to exclude the air. Plate or window glass, or green bottle glass, is the most useful, but flint glass, which contains much oxide of lead, would be detrimental in many cases. 16. Ferrous sulphide (FeS) is chiefly used as a source of sulphuretted hydrogen. Roasted with easily decomposable sulphides, such as that of silver, it converts them into sul- phates. Heated with oxides of copper, nickel, etc., it forms regulus. Heated in air it is oxidised to sulphate, and at a high temperature to oxide. 1 7. Iron pyrites (FeS ). This body loses half its sulphur at a white heat, forming ferrous sulphide, and is used for similar purposes to that compound. It is chiefly employed in the metallurgy of copper, nickel, and cobalt. 18. Ferric oxide (Fe 2 3 ). This oxide is very stable, non-volatile, and of a red colour. At a white heat it gives up oxygen, forming Fe 3 4 . By heating with carbon or carbonic oxide, it is reduced to the metallic state, but if much carbonic acid is present ferrous oxide may be formed, which combines with any silica present, forming a fusible silicate. For this reason it is sometimes used as a flux. In refining iron it acts as an oxidising agent. In presence of sulphur it oxidises that element to sulphur dioxide. 19. Zinc oxide (ZnO) is a powerful base ; it forms com- binations with alkaline earths and several bases, and has a i REFRACTORY MATERIALS 75 strong affinity for alumina. It is reduced by carbon, carbonic oxide, and hydrogen. Zinc oxide and carbon in small quantity is added to molten copper for producing sound castings. REFRACTORY MATERIALS 17. For melting metals a furnace is required, built of, or at least lined inside with, a material capable of with- standing high temperatures without fusing, or softening, or decomposing by the heat to which it is subjected. As a rule the exterior is constructed of ordinary masonry, but the in- terior is lined with refractory material, the nature of which depends to a great extent upon the character of the operation to be performed in it. Refractory materials are either used in the natural state, such as silica, alumina, oxide of iron, magnesia, and fire-clay, or they undergo a preliminary pre- paration before use. In some cases the materials are moulded to the internal shape of the furnace. If they are not of a plastic nature like fire-clay, then clay, tar, or other binding material is intimately incorporated with them, in order to impart the necessary plasticity. Quartz or Silica. This substance neither softens nor melts at the highest furnace temperatures, and is therefore a valuable material for internal construction, either when mixed with refractory clay to form silica bricks, or when used alone as a lining for the beds of reverberatory furnaces. Sand is not composed of pure silica, but the small quantities of lime, oxide of iron, and clay usually present are not objec- tionable. Dinas clay is a highly refractory substance, occurring in the Vale of Neath, and contains about 97 per cent of silica, the remainder being lime, oxide of iron, alumina, alkali, and water. From this material, when mixed with 1 to 3 per cent of lime, bricks are made, which form valuable linings for the roofs of many reverberatory furnaces. Alumina is quite as infusible as silica, and has the advan- tage of not generally combining with bases, and when it does so combine the aluniinates formed are less fusible than silicates. 76 MIXED METALS CHAP. But pure alumina is rarely found in nature, the nearest approach to it in large quantity is bauxite, which Berthier found to consist of 52 per cent alumina, 2 7 '6 oxide of iron, and 20-4 per cent water ; but the composition varies in different specimens. The ordinary mineral contains 3 to 5 per cent silica, 24 to 25 per cent oxide of iron, 50 to 60 per cent alumina, and 10 to 15 per cent water. It is generally dark red in colour like an iron ore, but some varieties exist with but little iron, and are then white in colour and very refractory. Lime and Magnesia. These are infusible bodies, strongly basic in character, but they form fusible compounds with silica and other acid bodies. This property is utilised in some steel furnaces, the interiors of which are lined with these oxides, which abstract phosphoric acid from the charge, form ing stable phosphates. Lime and magnesia occur together in dolomite, from which material, after calcination, basic bricks are prepared. Fire-Clay. The refractory bodies already referred to may combine together in certain proportions without ceasing to be refractory. Fire-clay is a hydrated silicate of alumina with varying amounts of lime, magnesia, oxide of iron, alkali, etc., and some mechanically mixed silica. The plastic property which clays possess is due to the chemically combined water. In all cases the plasticity disappears when the clay has been baked, and it remains granular and powdery. The clays of the coal-measures, such as those of Stourbridge, are admirably adapted for making fire-bricks, although not pliant enough for pottery. In fact an excess of plasticity is a disadvantage for some metallurgical purposes, such as bricks, for example, which would crack at the time of baking. 18. Crucibles, etc. Earthen crucibles are made of fire-clay mixed with sand, burnt clay, or other infusible matter, so as to counteract the tendency which raw clay possesses of shrinking when heated. The bodies thus mixed with the clay expand, or do not contract on heating, having CRUCIBLES, ETC. been already shrunk when burnt, and therefore act in an opposite manner to the clay. Such a composition must be able to resist a high temperature without softening, must not be friable when hot, and be capable of withstanding sudden changes of temperature without cracking, as, for instance, when a white hot crucible is brought out of a furnace into cold air. Some crucibles are required to resist the corrosive action of metallic oxides in the material operated upon, and in the ashes of the fuel, so that a crucible should be selected which is best adapted to the special purpose to which it is to be applied. The component parts of a crucible are first crushed to a fine powder, and passed through a sieve, the fineness of whose meshes will vary with the desired fineness of the grain in the pot (the plasticity being closely connected with the fineness of the particles ; at any rate for small crucibles this closeness of grain appears to be indispensable), then the fine powder is mixed with water and kneaded to the right consistence for use. The best results are obtained by using a mixture of different fire-clays, the most infusible being those containing the largest amount of silica and the minimum of oxides of iron and lime. The presence of potash and soda in small quantity sensibly increases the fusibility, but they act advantageously in soldering the particles together. Iron pyrites, which is frequently disseminated through clays, especially those from the coal measures, is perhaps the most injurious constituent. A crucible made from such clay will become indented with small cavities, and even holes, when exposed to a prolonged high temperature. It follows then that the most refractory crucibles are those made from pure clays, the nearest approach to which is presented by some French clays. The fitness of a clay for making crucibles may be de- termined by moulding a portion into the shape of a prism, or any form containing sharp edges, carefully drying, baking, and exposing to a high temperature in a covered crucible for 78 MIXED METALS CHAP. some time. If very refractory, the test will show no signs of fusion. If the edges are rounded it is a proof of incipient fusion, and if melted, the clay is useless. Clay vessels of all kinds may be tested to ascertain their power of resisting corrosion by melting in them a mixture of litharge, red oxide of copper, and borax, and noticing the time this mixture will take to permeate them. Those which resist this destructive action the longest will of course be the best. Most crucibles are by this means eaten away irregularly, showing the necessity of uniformity of grain to resist perforation. All crucibles should be cautiously annealed before use by placing them in an inverted position over the furnace, other- wise they are liable to split when plunged into a red-hot fire. I have noticed this tendency with the best plumbago crucibles. Plumbago or black-lead pots are made from varying proportions of fire-clay mixed with powdered graphite or coke dust. Good graphite is neither altered nor fused by exposure to the highest temperatures (air being absent), so that it is an admirable substance for crucibles. The graphite is powdered, sifted, and mixed with sufficient clay to render it plastic. Good plumbago crucibles, after a careful pre- liminary annealing, withstand the greatest changes of temperature without cracking, and may be heated many times in succession. When an ordinary crucible requires to be protected from the corrosive action of metallic oxides, or when small amounts of metallic compounds have to be reduced, the inside is coated with a lining of charcoal. This is done by first mixing the charcoal with sufficient starch, paste, or treacle to make it adhere when pressed. The crucible is then loosely filled with the brasque, and a cavity of the desired size made by boring with a triangular-shaped piece of wood, and then made smooth with a round elongated wooden tool, whose size and shape are apportioned to the capacity of the cavity desired, or the brasque may be plastered on the inside of the crucible by the hand. FIRE-BRICKS 79 19. Fire-bricks. A fire-brick, employed to withstand high temperatures, must only contain small quantities of the alkalies, which should not exceed 1 per cent. Glenboig, Stour- bridge, and Wortley (Leeds) are the leading brands extensively used ; these are safe, though of course there are many other fairly serviceable bricks made of brands less in repute. Ganister bricks are exceedingly valuable for withstanding the very highest temperatures for the crowns of reverberatory furnaces. They do not crack on cooling so much as bricks composed almost entirely of silica. The Lowood bricks, made near Sheffield, have a very high reputation in this class. Ganister bricks should be set in thin ganister cement. Crowns are best put on dry, and just " slurried " over the top when finished. No fire-brick has a fair chance if set in a clay inferior to itself ; but however excellent the clay, a good furnace builder will use as little as possible. Dinas bricks are practically infusible, and composed almost wholly of silica. The fractured surface presents a coarse, irregular structure of a light-brown colour. The lime which is added exerts a fluxing action on the particles of quartz, and so causes them to agglutinate (see also p. 75). Mr. James Dunnachie, in a paper read before the British Association, said : " The great variety of purposes for which fire-bricks are used, and the various qualities required in them, make it impossible that one brick can answer for all descrip- tions of furnaces. They require to stand strong heat and changes of temperature ; in some cases heavy pressure and hard knocks ; in others they require to resist the fluxing and chemical actions of various kinds. " A brick high in silica, containing a fair proportion of alumina, and nearly free from alkalies and other impurities, is the one which combines, in the highest degree, infusibility and freedom from cracking. To get a really good furnace we must first procure the best materials for its construction ; but after that much depends upon how it is built. If we were as careful about the curves of our furnaces as we are about the lines of our ships, and as particular about the 80 MIXED METALS CHAP. quality of the materials and workmanship employed as the importance of the subject demands, we might, in many cases, double the duration of our furnaces, without waiting for the possible discoveries of the future." The following analysis will serve to show the composition of British materials used in furnace construction : Silica. Alu- mina. Ferric oxide. Liine. Mag- nesisv. Pofawh and Soda. Titanic acid. Fire-brick, Glenboig . 04-41 30-55 1-70 69 ill 55 1-33 Stour bridge 73-05 22-40 2-43 39 54 1W 77-63 19-48 1-29 18 31 91 Newcastle . 73-30 20-56 1-69 1-55 -2 2'50 Pensher 58-00 65-10 36-50 30-03 1-67 2-00 58 45 82 51 2-42 2-04 Flintshire . 88-10 4-50 6-10 1-2 .. Leeds . 77-30 19-17 1-43 45 63 1-20 72-65 23-75 1-75 30 36 90 07-40 25-56 2-00 1-12 72 2-47 Ell'and '. 62-33 35-59 1-25 73 10 :: Ganister bricks, Lowood 96-32 99 71 1-28 25 '26 ,, Witton 94-80 89 1-10 2-85 32 30 " Dinas bricks, Wales . 95-16 42 33 2-96 28 29 Maiirrn anga- nous oxide. Burnt Dolomite, Raisby Hill 5-60 2-35 1-70 M'56 34-18 1-2 Ganister, Lowood 89-55 4-85 1-28 80 3-52 Weardale . 87-93 9-10 64 '73 40 1-05 PREPARATION AND PROPERTIES OF ALLOYS. 20. An alloy may be prepared by fusion in a crucible or furnace by melting the constituent metals together ; by strongly compressing the powders of the constituent metals ; by electro-deposition ; by chemical reaction and by Gold- smith's method of alumino-thermics. An alloy may be produced by reducing one of the metals in the presence of the other. A metal A may be in the form of an oxide, sulphide, etc. If such a metal is reduced by a reducing agent in the presence of the other metal B, an alloy will be formed of A + B. Such was the case with the ancient i PREPARATION OF ALLOYS 81 method of producing Calarnine brass and the present method of obtaining alloys of silicon and some of the rarer metals in the electric furnace. Alloys are also formed by the simultaneous reduction of both metals from their oxides. Such is the case with nickel-vanadium. The recent method of alumino-thermics is an application of the same principle. The oxides of the two metals are mixed with aluminium in grains which reduces the oxides and the metals set free alloy together. Lastly, alloys may be prepared by cementation, as in the case of carbon and iron to produce steel ; copper and zinc to produce brass ; and copper and tin to produce bronze. The mode of procedure in the production of any alloy will be largely influenced by the nature of the metals to be operated upon. Some metals are volatile, and readily pass off as vapour when heated a few degrees above their melting points. Others have little tendency to vaporise, and may be raised to high temperatures without sensible volatilisation. When a volatile metal has to be alloyed with a non-volatile metal, and the fusing points of both are approximately the same, combination can be most readily effected by mixing the constituents and melting them to- gether in the same crucible or furnace. This is, however, seldom the case, and, as a general rule, the components of an alloy, one or all of which are volatile, have widely divergent melting points, and then it is requisite for the most refractors- constituent to be melted first, and for the others to be added in the solid state. Again, an alloy may contain one or more fixed metals and a volatile one, in which case the more volatile metal is added to the crucible, after the fixed metal or metals have been fused, and raised to a temperature necessary to melt the volatile constituent immediately it is introduced, so that combination may be effected before any serious loss, due to vaporisation, has occurred. Union between the components of an alloy is more perfectly secured by agitation of the contents with a stirring-rod, the most effective in many cases being a wooden or carbon rod, which promotes G 82 MIXED METALS CHAP. admixture without the introduction of any substance likely to contaminate the mixture and modify its properties. Professor Turner l has shown that certain metals, such as lead, zinc, and arsenic may be readily vaporised when melted "in vacuo" and thus removed from other metals with which they are alloyed. He melted common brass in vacuo and succeeded in completely removing the zinc and lead, leaving practically pure copper. He suggested this method for refining impure copper and hard zinc. A thing to be guarded against in the melting of all base metals, or alloys containing base metals as essential con- stituents, is oxidation. Various plans are adopted to avoid loss of metal and injury to the alloy from this cause. The most common one is to cover the metals with carbon, which not only excludes the air admitted to the furnace, but tends to absorb any oxygen liberated from the metals during fusion The gas thus formed by union of carbon with oxygen is termed carbonic oxide CO, and this gas being a reducing agent, is capable of taking up another atom of oxygen, forming carbonic acid C0 2 . Thus, as long as the mixture is covered with carbon, the carbonic oxide formed effectually shields it from oxidation. In the method already referred to of stirring, metals with a carbon rod to promote mixture, the same gas, carbonic oxide, is formed, and thus the rod not only promotes union by mechanical agitation, but generates a gas which protects the metals in a great measure from oxidation. In some cases this is not admissible, as com- mercial metals are impure, and it may be advisable to admit sufficient oxygen, either from the air or by means of a special oxidising agent, added along with the flux, to convert the impurities into oxides, which do not alloy with the metals, but either enter into combination with the flux to form a slag, or rise to the surface as dross or scum. In most cases it is advisable that the covering body should not exert any influence on the metals beneath. Some manufacturers are in the habit of throwing fat and 1 Institute of Metals, January 1912. i PREPARATION OF ALLOYS 83 resin on the lieated metals before fusion. These are de- composed by the heat, liberating gases, and when well stirred with the molten metal promote combination by the mechanical agitation imparted by their escape. They also act chemically in removing oxygen, by the union of that element with the carbon and hydrogen set free. When the evolution of gas has ceased a quantity of carbon remains in a finely divided state, which covers the metals and protects them from oxidation. Borax is sometimes used to exclude the air, but it is much more costly than carbon, and when it is not required as a flux its employment is accompanied with some evils. Now borax is composed of the base soda in combination with boric acid, which is only partly saturated with the soda, and the excess of acid unites with any metallic oxide present, forming double borates of a glassy nature. Commercial borax is often very impure, and is adulterated with common salt and alum ; these impurities are injurious to many metals. Sodium chloride, or common salt, is also employed for preserving molten metals from oxidation, and also to moderate the action of bodies which cause violent ebullition. Glass is frequently used for a similar purpose, and next to carbon is the least injurious to metals. It is a mixture of silicates, which easily fuses at high temperatures, forming compounds with lime and other bases, so that it acts almost as beneficially as borax when such a flux is required. Window glass or green bottle glass is the most useful, but flint glass, which contains much oxide of lead, would be detrimental in many cases. The nature of metallic alloys has already been discussed, from which we may assume that certain proportions of the constituents enter into chemical combination, and other portions are simply in a state of mixture or solution, and therefore, on gradually cooling, tend to separate in distinct layers according to their respective densities. This is especially the case when the constituents have widely divergent densities, so that the higher the temperature of 84 MIXED METALS CHAP. the alloy when removed from the furnace the longer will the period of cooling last, and the greater will be the facilities offered for separation. To obviate this defect the metal should be constantly agitated by stirring or otherwise, and poured into the moulds at the lowest temperature consistent with the requisite fluidity ; and cooled as rapidly as the nature of the alloy and the purposes for which it is designed will admit. With regard to the melting point of an alloy, it should be borne in mind that it fuses at a lower temperature than that at which the most refractory constituent melts, and sometimes below that of either, which knowledge should guide the operator in so regulating the temperature as not to make the charge unnecessarily hot. It is a well-known fact that the character of many alloys is altered by repeated remelting, and that the scrap obtained in working cannot be used again without the addition of a certain quantity of new metal. A given mixture may be employed for the formation of an alloy, which is highly malleable, ductile, and tenacious, and the scrap from the same alloy when remelted may be brittle and unworkable ; but when a suitable quantity of new metal is added, the combination may form an alloy even superior to the original one with regard to its good working properties. It is to the advantage of the manufacturer as regards economy, to use as much scrap as possible in alloying, and the quantity thus employed varies from one-third to two-thirds of the weight of the charge. Of course, in using old metal many more im- purities are liable to be introduced than with new metal, and although the same impurities may exist in the new metal, the quantities may be insufficient to produce a deteriorating effect, but when augmented from old metal may then rise to such proportions as to entirely alter the physical properties of the alloy. The presence of notable quantities of foreign matter is generally exhibited by increased hardness and a modification of the structure, as seen on a freshly fractured surface. The difficulty of maintaining uniformity in an alloy after i PREPARATION OF ALLOYS 85 repeated remelting is least when only two metals are mixed together, and increases when the combination requires the presence of three or more metals. Thus German silver requires much greater care in this respect than brass ; and soft solder, containing only lead and tin, requires less care than fusible alloy, containing bismuth or cadmium in addition to lead and tin. Those alloys which contain as an essential constituent a volatile metal, such as zinc or antimony, are generally altered most by remelting, and it is requisite to know, at any rate approximately, what the furnace loss is, so that the defection may be counterbalanced by the addition of the quantity of fresh metal requisite to maintain the right composition. Many errors arise from this cause, as well as from overdoing what is required. Where possible, a chemical analysis is the best means of solving the problem, but as this is out of the question in most cases, a few simple trials with weighed quantities, and careful observations of the results obtained, by testing its malleability, colour, and fracture, will generally afford sufficient evidence of the required amount to be added. In making experimental tests, a small melting furnace, such as that used in a metallurgical laboratory, a strong pair of hand rolls, and an anvil, would be very useful adjuncts to every casting shop. The quantity of metal operated upon need not exceed one pound in weight, and as this could be cast in a long strip, its suitability for stamping or rolling could be readily tested. Such test-pieces, if carefully labelled and preserved, would be most valuable for future reference, and there can be no doubt that both employers and employed would thus gain a vast amount of information which would prove of great benefit both as a standard of workmanship and of economy of production. It is a great annoyance to find, after a quantity of metal has been mixed and the castings made, that the alloy is unsuitable for the work required of it, either from unsuitable constituents, improper mixing, or impure materials which annoyance could be avoided by a few preliminary trials on the small scale. The casting of 86 MIXED METALS CHAP. such trial tests could be made in an iron or sand mould, and the time of cooling made to approximate to that of a large mass by judicious treatment. Another advantage of such an experimental plant would be that new combinations could be readily tried, and the effect of certain impurities on well- known alloys ascertained by purposely adding these bodies in definite amounts to a weighed quantity of the alloy. Casting Alloys. In the preparation of alloys by fusion there is generally a certain amount of loss which modifies the composition. This loss is due to volatilisation and oxidation. The temperature of casting has a considerable influence on the properties of the alloy. Thus, antifriction alloys when cast too ho't or too cold tend to become hot in wear and last much less time than when cast at the proper temperature. Bronze with ten per cent of tin is also very sensible to the temperature of casting ; the size of the grain increases with the increase of temperature. The same remarks apply to brass. There are certain casting temperatures at which these alloys give the best results as regards tenacity and elongation. Guillet l found that on causing the casting temperature of manganese brass to vary from 1100 to 950 C. the elongations varied from 4 to 20 per cent. Alloys may be cast in metallic or in sand moulds. When metals are required to be rolled or drawn the casting is done in ingot moulds. For articles to be simply cast for dressing, polishing, etc., sand moulds are largely used. The quick- ness of cooling after solidification has a marked influence on the properties of the metal. The slower the cooling the more the grain of the metal is developed and the more apparent is the crystallisation ; the quicker the cooling the finer and closer is the grain. In many cases a second fusion is made, either for employing metal previously cast into ingots, or to use up scrap, or, as in the case of Britannia metal, to get a sounder and more uniform structure. In a second fusion some of the metal is lost by volatilisation and oxidation, as in the case 1 Alliages Mttalliqiies, p. 80. i PREPARATION OF ALLOYS 87 of zinc, antimony, lead, manganese, and aluminium alloys. It is specially objectionable when some of the oxide formed is finally retained in the metal Castings are liable to certain defects which influence their appearances or properties. These are summarised by Guillet : pipes, risers, blowholes, cold drops, cracks, blisters, spilly places and liquation. Piping is due to contraction during solidification, the upper part serving to feed the metal below. Risers are due to excessive disengagement of gases. Blowholes are due to enclosed gases, by casting at too low or too high a temperature ; to impurities and lack of material. Cold drops are due to too rapid cooling of one part of the metal. Cracks are due to bad moulds, too little feeding, and too great contraction. Blisters are due to the inability of gases to escape at the surface ; such gases are often formed by the reaction of carbon on any oxide present, producing carbonic oxide. Spilly places are spongy or spotted parts probably due to oxides and other impurities which react on each other to produce gases and to imperfect mixing. Liquation is due to the difference of density of the constituents of an alloy. Such is the case with lead in brass. Treatment of Alloys. Alloys are submitted to certain operations in order to bring them into the desired condition for commerce, such as rolling, drawing, hammering, cleaning, heat treatment and chemical treatment. The principal heat treatments are : annealing, quenching, and tempering. In annealing the metal is raised to a certain temperature, depending on the nature of the metal or alloy. Annealing has for its object the complete destruction of the influence of mechanical or physical treatment which the metal has undergone. The result is that the tenacity is diminished, the elongation increased, and the tensions produced by compression are removed. A Le Chatelier has shown that each metal possesses a limit of compression, which is practically attained for a relatively small deformation which it is impossible to exceed. Impurities tend to raise the compressing limit. 88 MIXED METALS CHAP. It has been observed that cold working of metals often produces an augmentation of strength. Le Chatelier l finds that there is a limit to the increase of strength obtained by the cold working of pure metals or of those containing less than 1 to 2 per cent of impurities. For all metals examined excepting silver, the maximum strength after cold working is double that of the perfectly annealed specimens. In the case of alloys some follow the same law as pure metals ; others, such as bronze, copper-silver alloys, and aluminium bronze, become more and more brittle after each successive pass without annealing, and the strength increases regularly, but at last the metal becomes too brittle to be further worked, and gives way. In regard to annealing, five laws are formulated as the result of experiments : (1) Annealing is never instan- taneous ; its effects, rapid at first, become more and more slow, and the softening tends towards a limit for each temperature ; (2) This limit is lower and is attained more rapidly as the annealing temperature is raised ; (3) Above a certain temperature annealing is complete, and a further increase of temperature tends to diminish the strength, a crystallisation due to annealing occurs, and increases with the time of annealing, ultimately reducing the tensile strength and elongation to those of the cast metal ; (4) The presence of impurities retards the action of annealing, and demands a higher temperature -for its completion. (5) The crystallisation from annealing is chiefly due to the presence of impurities which have lower fusing points than the metal itself, or which form compounds which have those properties. Cold -worked metals tend to recover their malleability even at ordinary temperatures by a process which Le Chatelier terms spontaneous annealing. The maximum limit of strength attainable by cold-working is reached at the moment when the increase produced by continued working is just balanced by the diminution due to spon- 1 Butt, de la Soc. d' Encouragement, 1896, Apl. 564-570. i PREPARATION OF ALLOYS 89 taneous annealing. Similarly, in wire drawing, if the thickness of the metal be reduced too rapidly by successive passes without annealing, it will break, owing to the failure of the spontaneous annealing to keep pace with the distorting force ; but the metal may be fractured even in course of a very gradual reduction, unless it be allowed to remain at rest for five or ten minutes between the passes ; with this precaution, however, it may be drawn down indefinitely even without annealing. Spontaneous annealing affects the mechanical properties of metals under test, causing the breaking load at any given temperature to be greater in proportion to the rapidity with which the stress is applied, whilst the deformation produced is not instantaneous, but increases more and more slowly up to a certain limit. Quenching is a treatment based on the quick cooling of a metal which has previously been raised to a certain temperature. This method is particularly applicable to steels, but it is also valuable for other alloys, such as tin bronzes and aluminium bronzes. The effect of quenching is to preserve the state in which the metal exists at the temperature at which it is quenched. This effect on the properties of an alloy depends on its composition, the temperature of quenching, and the nature and volume of the quenching liquid. Of these, the composition has a preponderating influence. A metal that undergoes no transformation during heating such as copper will under- go no change either by quenching or annealing. The rapidity of cooling depends on the conductivity of the quenching liquid, on its volume up to a certain point, on its low viscosity and on its low vaporisation. The liquids most used are : water, lime water, salt water, dilute acids, mineral oils, colza oil, molten lead, etc. Tempering is an annealing at a low temperature and is chiefly used for steels. Its effect is to destroy internal tensions and diminish the brittleness which always accom- panies rapid quenching. The purposes for which alloys are required are endless. 90 MIXED METALS CHAP. Some are required to possess great malleability, for others hardness is the chief requisite ; others, again, must possess a high degree of elasticity, while some are useful on account of their low melting point, etc. These different demands can only be satisfied by uniting suitable metals in different proportions. The metals most often used for alloying at the present time are those which have been known the longest, such as copper, zinc, lead, tin, gold, and silver ; and although combinations of these metals have been known and employed for many centuries, it is only during the latter half of the present century that their intimate properties have been closely studied. Indeed, at the present day our information concerning the nature and properties of alloys is perhaps less than in any other branch of chemical science, and although chemical investigation may do much to enlighten our know- ledge, such information will be destitute of great commercial value unless accompanied with practical knowledge of the working, from observation of the physical properties, when alloys are worked in large quantities, by the manufacturers themselves. The number of simple metals is very limited, but they may be united in various proportions, forming an endless variety of modifications ; and since every alloy may be looked upon as a new metal, from the fact of its properties differing from those of its constituents, we have at command the necessary material for producing metals suitable for every requirement for which metallic matter is desirable. The action of metals upon each other is widely divergent ; sometimes one metal may be added to another in quantity without seriously altering its working properties ; in other cases a minute quantity of a second metal will altogether change the character of the first metal ; so that in alloying it by no means follows, because one metal may be freely added, that another, even of a similar nature, may be as liberally introduced. The man who aspires to the formation of new alloys, or who wishes to produce metals suitable for different requirements as circumstances arise, must be well i PREPARATION OF ALLOYS 91 acquainted with the nature and properties of the simple metals in order to successfully accomplish his object ; and although a knowledge of the components is not sufficient of itself, it is of immense advantage in assisting the operator who combines practical experience in mixing metals with this theoretical knowledge. It is for these reasons that a brief account of the elementary metals is included in this work. In chemical combinations it is a well-known fact that elements always combine with other elements in definite proportions by weight, termed atomic weight, producing compounds of fixed and decided properties, so that the same compounds can be always relied upon to contain the same elements, united in the same proportions. This same law applies to the union of two metals, when such metals are chemically combined, and the same alloy will always have pro- perties identically the same, however it may be tested. Several experimenters have directed their attention to the mixing of metals according to their atomic weights, so as to obtain alloys of determined characteristic properties, but up to the present time the number of such combinations of a useful character is very limited. They are by no means the ones most suited to the wants and requirements of industry. There is always one indispensable item from the manufacturer's point of view which the chemist is not concerned with that is, the cost of production and however nicely atomic propor- tions would suit the requirements of a given alloy, such an alloy would in most cases be useless unless the cost was consistent with the market value. The question then of cost must have consideration, and the proportions must, if possible, be made to fit in with commercial necessities. With regard to copper alloys, such as brass and bronze, the combinations which best exhibit the characters of chemical compounds are hard and brittle, and as copper alloys are much more widely used than any other, there is little inducement to encourage metallurgists to endeavour to alloy copper and zinc, or copper and tin, in atomic proportions, since malleability and tenacity are the properties most desired 92 MIXED METALS CHAP. in these alloys. Again, colour is the chief desideratum in many alloys, and this cannot be always obtained by mix- ing in atomic proportions, especially as it often happens that a very small addition of one of the constituents will alter the shade of colour so as to produce what is required. When it is desirable to add a non-metallic element to a metal or alloy, for the purpose of bringing about a certain result, very much greater care is generally required in apportioning the quantity to be added than with a metal, as non-metals combine much more actively with metals than the metals do with each other, and a very small quantity of a non-metal will suffice to alter the properties of a metal or alloy. It is very surprising to note how, in some instances, a mere trace of another element will alter the properties of a metal. For example, -^^ of carbon added to iron will convert it into mild steel ; y^j-Q" ^ phosphorus makes copper hot-short ; 2 ^ part of tellurium in bismuth maltes it minutely crystal- line ; 10 1 00 part of bismuth in copper renders it exceedingly bad in quality for certain purposes. Lothar Meyer has shown that a remarkable relation exists between the "atomic volumes of the elements." The rela- tive atomic volumes of the elements are found by dividing their atomic weights by their specific gravities. The atomic weight of lead is 207, and its specific gravity 11-45; = 18, the atomic volume of lead. It would appear 11-45 that the power of an element to produce weakness in a metal, when added in small quantity, is dependent on the atomic volume of the impurity. 1 Roberts-Austin tried the effect of various elements on pure gold, and found that when the body added had an atomic volume equal to, or less than that of gold, the strength was little affected, and in some cases, as copper for example, was increased ; but when the element added had an atomic volume much greater than that of gold the strength, with two exceptions, was greatly diminished. His results are embodied in the following table : 1 Jmirnal of Soc. of Arts, 19th October 1888. PREPARATION OF ALLOYS 93 Tensile strength. Elongation per cent on 3 inches. Impurity per cent. At vol. of im- purity. K. /Less \than -5 Not perceptible Less \ than '2 J 45-1 Bi 5 j j 21 20-9 Te 3-88 > j 186 25'5 Pb 4-17 4-9 240 18 Thallium 6-21 8-6 193 17-2 Sn 6-21 12-3 196 16-2 Sb 6-0 ? 203 17-9 Cd 6-88 44-0 202 12-9 Ag 7-1 33'3 2 10-1 Pa 7-1 32-6 205 9-4 Zi 7-54 28-4 205 9-1 Rh 776 25-0 21 8'4 Mn 7-99 297 207 6-8 In 7-99 26-5 29 15-3 Cu 8-22 43-5 193 7-0 Li 8-87 21-0 201 11-8 Al 8-87 25-5 186 10-6 The intervention of a third metallic solvent to the two constituents of an alloy, which do not naturally mix, often causes their assimilation by destroying the eutectic. Until recently the initial point of solidification was the only one considered, but it is necessary to know what subsidiary eutectic points may be present, as the action of impurities is very marked in connection with them. In some cases the foreign bodies remain in the mass, partly 'or wholly, as solidified solutions, the impurities being probably dissociated into their atoms in both the solid and the liquid mass. The relation between melting points of alloys and the atomic volumes of the constituent metals has been studied by Pictet, 1 who showed that there exists an intimate relation between the melting points of metals and the length of their molecular oscillation, viz. that the length of oscillation at ordinary temperature diminishes as the melting point rises ; and it is 1 Comp. Rend., vol. Ixxxviii. 1879. 94 MIXED METALS CHAP. known that generally the metals with the highest melting point are the most tenacious. The absolute temperature of the melting point of a metal must be closely connected with its atomic volume, because the former is inversely pro- portioned to the rate at which the amplitude of the oscilla- tions of the molecules increases with the temperature ; and the rate of increase of amplitude at any given temperature is obtained by multiplying the ordinary thermal coefficient of linear expansion by the cube root of the atomic volume. The relation between melting point and tenacity appears to be much more definite in worked alloys than in pure metals, probably because an alloyed metal is in a more definite condition than one nominally pure. An alloy is less radically changed by traces of an impurity than is a single metal. Pure metals are consistently weaker than alloys. Certain exceptions are found, but such alloys are composed of metals whose atomic volumes are practically identical and possess lower tenacities, as compared with their melting points, than alloys of metals having different atomic volumes. Examples of metals with same atomic volumes are found in copper- nickel ; gold-silver ; platinum-iridium, etc. 1 Alloys generally have properties differing from their constituents, and some have these differences very strongly marked, thus : If a very small quantity of arsenic be added to tin, the resulting alloy will have a crystalline fracture closely resembling zinc. Sometimes metals combine with evolution of heat and sometimes with an absorption of heat. The following metals, according to Roberts-Austin, evolve heat when they are united : Aluminium and copper, platinum and tin, arsenic and antimony, bismuth and lead, gold and just melted tin. On the other hand, lead and tin when they unite absorb heat. When lead, tin, and bismuth in equivalent proportions, and very finely divided, are mixed with eight equivalents of mercury, at the ordinary temperature, the temperature will fall from 17 C. to 10 C. If the vessel containing the 1 Fourth Report, Alloys Res. Com. lust. Mechan. Eng. PREPARATION OF ALLOYS 95 mixture stand on a wet board the water underneath will be frozen. This combination then will form a freezing mixture. The method of producing alloys by strongly compressing the powders of the constituent metals was shown by Professor Spring of Liege in 1878, and he has since devoted much study to the subject. 1 His experiments were made by the aid of a press, the form of which is shown in Fig. 1 5. The I I FIG. 15. metallic powder is placed under a short cylinder of steel, in the cavity of a steel block divided vertically, held together by a collar, and placed in a chamber of gun-metal, which may be rendered vacuous. The pressure is applied to a cylindrical rod passing through a stuffing-box. Under a pressure of 2000 atmospheres, or 13 tons to the square inch, lead, in the form of filings, becomes compressed to a solid block ; and under a pressure of 5000 atmospheres the 1 Bui. de VAcad. de Belgique (2), torn. xlv. No. 6, 1878 ; (2) torn. xlix. No. 5, 1880. 96 MIXED METALS CHAP. metal flows through all the cracks of the apparatus like a liquid. Spring obtained some important results with crystalline metals, such as bismuth and antimony. At a pressure of 6000 atmospheres finely powdered bismuth unites into a solid mass, having a crystalline fracture. Tin when similarly compressed in powder, is made to flow through a hole in the base forming a wire. The following table shows the amount of compression required to unite the powders of the respective metals : Lead unites at 13 tons per square incli. Tin ,, 19 Zinc 38 Antimony , . 3S Aluminium . , 38 Bismuth . , 38 Copper . 33 Lead flows at 33 Tin 47 It occurred to Professor Spring that the particles of different metals might be united by pressure, so as to form alloys, and he considered that the formation of such alloys by pressure would afford conclusive proof that there is true union between the particles of metals in the cold, when they are brought into intimate contact. He compressed a mixture of 15 parts bismuth, 8 parts lead, 4 parts tin, and 3 parts cadmium, and produced an alloy which fuses at 100 C. It is necessary to crush up the product of the first compression, and again submit the powder to compression in order to get a perfect alloy. The objection has been urged that the mixture may have been fused by the heat of compression. Professor Spring has experimentally proved that this was not the case. The compression was effected with extreme slow- ness, and he calculates that if all the work done in compressing the powders were translated into heat, it would only serve to heat a cylinder of iron 10 mm. in height and 8 mm. in diameter 40'64 C. Spring took the organic compound phorone, which melts at 28 C., and compressed it in precisely the same way as the metallic powders. Only imperfect i PREPARATION OF ALLOYS 97 union of the particles resulted, and the 28 of heat necessary to melt the phorone was not generated. 1 Electro-deposition is another means of producing alloys, and although it may not be practicable to produce a very thick coating, the application of the method for thick deposits is somewhat extensively practised. The difficulty lies in getting a mixed solution of two or more metals that are equally decomposable, or of equal conducting power. Take, for example, a mixture of the cyanides of gold, silver, and copper in a solution of potassium cyanide. The silver is the best conductor, and much of it will be deposited first, if a weak current is employed. If the solution is heated and the current so maintained that no gas is allowed to escape from the articles, the gold may be deposited next and the copper left in solution ; but if the gas is allowed to flow from the cathode the copper will probably be deposited more abundantly than the gold. Copper and zinc can be deposited together in a potassium cyanide solution to form brass. The solution should be worked at a temperature of 55 C., using a good brass anode and a current equal to that produced by three quart Bunsen's cells. Too much free cyanide will prevent the deposition. It is more difficult to deposit copper and zinc simultaneously than separately, and a good brass colour can only be obtained by properly adjusting the strength of the current. If too strong a pale brass results ; if too weak a red brass may be obtained. Solutions of alloys generally, such as German silver for example, may be made by dissolving one ounce of the alloy in dilute nitric acid, in such quantity as to leave a small portion of the metal undissolved, and thus avoid excess of acid. The solution is then diluted to one gallon with water, ammonia added until the precipitate first formed re-dissolves, then about one-sixth the amount of potassium cyanide solution. A strip of metal of the alloy is necessary for the anode. The solution should be worked hot, using a current strong enough to evolve hydrogen from the cathode. 1 Bui. Soc. Chim., Paris, 1884, torn. xli. p. 488. H 98 MIXED METALS CHAP, i Alumina - Thermics. Goldschniidt introduced a method of producing metals and alloys from their oxides by taking advantage of the heat of combination of aluminium and oxygen. The oxygen is taken, not from the air, but from a solid oxide, which is intimately mixed with finely granulated aluminium and termed " Thermit." To start the reaction a fuse of aluminium and barium peroxide is used which is ignited by a burning piece of magnesium. When the thermit is once ignited the combustion continues throughout the whole mass in a very short time. It is advisable to have the mass quite dry. and free from grease. The process is applied to welding iron and steel, in which case the thermit consists of iron oxide and aluminium, and the reduced iron runs into the junction of the two pieces and firmly unites them. The reaction is as follows : Fe 2 3 + 2 Al = A1 2 3 + 2Fe. This method of reduction is very useful in producing iron, chromium, manganese, etc., from their oxides, free from carbon. In the same way may be made alloys of manganese with copper, tin, zinc, etc., and various iron, copper, and other alloys by using their oxides mixed with aluminium. CHAPTER II COPPER ALLOYS 21. Copper forms with other metals a series of alloys far more numerous and important than that of any other metal, which may be accounted for by its red colour, high malle- ability, ductility, toughness, softness, and tenacity ; which properties it imparts in a great measure to many of its alloys, even when united with metals opposite to it in character, such as zinc. Next to iron, it may be considered the most useful of all the metals both from its valuable properties when used alone, and its intrinsic value as a constituent of alloys. The very properties which make copper so useful are sometimes a disadvantage for certain purposes ; for instance, the toughness and closeness of grain make it more difficult to turn in a lathe than brass ; and its softness makes it unfit to be used alone for objects subjected to great wear and tear. In the vast majority of cases in which copper is used it has to be melted and cast into moulds of various kinds, in order to prepare it for further treatment, and the difficulties in producing sound castings are so great, that it can only be successfully manipulated in the hands of a very skilful and experienced workman ; even then a common practice is to add some other substance, so that it is probably not too much to say that pure copper has never been cast in considerable quantity, so as to produce a good ductile casting free from blow-holes. Great attention has been directed to 99 100 MIXED METALS CHAP. this subject of late years in consequence of the demand for solid-drawn copper tubes, rollers, etc. ; arid different physics have been added to the copper in melting, with a view of overcoming the inherent defects. It must be borne in mind that pure copper is not a com- mercial article, and although copper is now manufactured on the large scale in a purer state than previously, the metal retains ingredients which modify the valuable properties the metal possesses in its chemically pure state. The common impurities are iron, arsenic, antimony, bismuth, and some- times sulphur, tin, lead, nickel, cobalt, and gold. These elements even in very small quantities seriously affect copper, but they may be neutralised to some extent by uniting them with oxygen. When such metal is melted in contact with air, some of the copper is oxidised, forming cuprous oxide Cu 2 0, which makes the whole mass brittle and unworkable ; some air or carbonic oxide is also retained in the gaseous state, and on the metal solidifying at the surface some of this gas will be enclosed in the interior of the mass, and produce a honeycombed structure. To prevent access of air, the metals may be covered with charcoal, and this when used judiciously may be effective, but the great difficulty is to know when the carbon has done its work. Carbon in contact with air will produce carbonic oxide, and this gas probably penetrates the copper to some extent and removes the oxygen ; but if any further excess of carbonic oxide is admitted, it may reduce the oxides present as impurities, as well as oxide of copper, and the elements being liberated, will alloy with the copper, making it brittle and analogous in properties to what is technically known as " overpoled " copper. Seeing then that commercial copper is impure, but that the impurities may be neutralised to a great extent by oxygen, the problem is to discover the point when this end has been attained, without introducing an excess of air. On the other hand, it may be as efficacious, and much more easy, to admit an excess of oxygen, and subsequently remove it, by adding some body which has a stronger affinity for ii COPPER ALLOYS 101 oxygen than copper has. To this end Mr. Walton, of the Ansoiiia copper works of the United States of America, has patented a process of preparing copper for casting, which, in the words of the patentee, is as follows : " My present invention is for treating copper in crucibles, so as to exclude the action of the atmosphere, and subject the copper to the action of carbon sufficiently to remove the oxygen, and render such copper solid when cast, and increase its malleability and ductility. I take eight pounds of zinc, either in the form of an oxide or carbonate, and mix it with one bushel of ground charcoal, wet it, and make it into a stiff paste, portion it out into, say twenty-four parts, make it into rough balls, and dry at a moderate heat. The copper is placed in a crucible in a furnace, and when on the point of melting, one of the balls is ^dropped upon the copper, and gradually falls in pieces and covers the copper as it melts down, thereby entirely excluding the atmosphere from the surface of the copper ; at the same time the zinc in the mass is evolved, and dispels any oxygen which may remain in the crucible. It may be supposed that the oxide of zinc will impregnate the copper, but such is not the case ; the charcoal coming in contact with the copper, and oxide of zinc being volatile under the action of heat, 1 no combina- tion with the copper takes place, and the zinc volatilised and carried off through the flue, while the charcoal remains on the surface of the copper, and combines with any excess of oxygen and burns. Copper treated in this way becomes per- fectly malleable, and is thoroughly toughened, and is, in fact, improved by this treatment. My improvement is especially available when the copper is melted in a crucible, but it may be used when melted on a hearth or otherwise. The im- purities of common copper are thrown to the surface in a slag, and the copper made so that it will work better either hot or cold, and stand a greater test for either tension or ductility. l . Zinc oxide is practically non-volatile. See Percy's Metallurgy, p. 532. 102 MIXED METALS CHAP. " The above method of treatment can be applied to copper that is to be used in the manufacture of any article. Copper so treated will remain in a very liquid state much longer than that treated in the ordinary manner, and the copper can be brought to a very great heat without losing its toughness. When casting such articles as tubes, or small or thin castings, a little phosphorus, added just before pouring, assists very materially in keeping the metal in a liquid state ; and also prevents the absorption of oxygen from the atmo- sphere while cooling in the moulds." 22. In addition to the statements already made con- cerning impurities in copper, the following summary of the effects of different elements will doubtless be of use to those interested in the manufacture of its alloys. Phosphorus. A small quantity does not sensibly alter the colour of copper, but a large quantity renders it grey. One half per cent makes it very hot-short, and only capable of being rolled in the cold, without cracking. A little phos- phorus, say O'l per cent, added to molten copper in a crucible promotes soundness in the subsequent casting. Phosphorus increases the fusibility and hardness of copper, and when present in quantity, renders it brittle at the ordinary temperature. Copper containing 11 per cent of phosphorus is extremely hard, of a steel-grey colour, is susceptible of a fine polish, but readily tarnishes. In making phosphorised copper, by adding phosphorus direct, the metal should not be stirred with an iron rod, as phosphorised iron will also be formed, and alloy with the copper. Silicon. Copper is contaminated with silicon when strongly heated in contact with sand and carbon. Copper containing 2 per cent of silicon resembles gun metal in colour ; is tough, harder than copper, red -short, but may be rolled in the cold. Mr. Anderson of Woolwich found copper containing 1'82 per cent of silicon tougher than gun-metal. If the temperature employed in heating the copper be too low, or not sufficiently prolonged, only a little silicon will be COPPER ALLOYS 103 reduced, and the metal will resemble slightly under-poled copper. Arsenic. Copper and arsenic readily combine when metallic arsenic is dropped into molten copper. When a very small quantity is thus added, the metal may be cast into a sound ingot, which contracts during solidification like phosphorised copper, and may be rolled cold, and afterwards drawn into fine wire. 0*1 to - 5 per cent increases the tensile strength of commercial copper. Quantities up to 0*5 per cent greatly increase the durability of copper which has to be considerably heated, as in the case of the plates of fire-boxes of locomotives. Much arsenic is highly injurious, making the metal hard and brittle. Arsenic readily combines with copper when one of its compounds is heated with charcoal in contact with copper. Iron. The malleability of copper is seriously impaired by the presence of iron, which renders the copper harder, paler in colour, and more infusible. It increases the tensile strength, but diminishes the ductility. It may, however, be removed by the use of an oxidising flux. Lead. One half per cent of lead in copper makes it both hot- and cold-short, and can only be removed by causing some of the copper to pass into the slag. In quantities less than this, lead is sometimes added to copper intended for rolling. Even *05 per cent in pure copper cannot be worked hot, but 0'2 per cent may be present in ordinary copper without impairing its tenacity. Antimony renders copper harder. Up to 0'5 per cent, it increases the strength and durability of copper, but when simultaneously present with other metals, except arsenic, this advantage probably ceases. One per cent and upwards imparts to the fractured surface a dull yellowish -grey colour. It is more injurious than arsenic. Bismuth. This metal exerts a specially injurious in-, fluence on copper, a very small quantity makes it hot-short and cold-short. O'l per cent makes copper so weak that at gradually rising temperatures the fall in tenacity is very 104 MIXED METALS CHAP. rapid. Even 0-002 per cent considerably reduces the elonga- tion. The explanation given for this deleterious action by Roberts-Austen is that, however poor or rich in bismuth the copper may be, a portion of bismuth, with perhaps a little copper, remains fluid until the temperature of the mass has fallen to 268 C., which is the melting point of bismuth. Zinc in very small quantity does not much alter the character of copper ; it tends to impart a yellow colour and fibrous fracture. 0'6 per cent of zinc in copper makes the latter hot-short but not cold-short. Nickel and Cobalt occasionally occur in copper ores, and are reduced along with the copper. These metals make copper less malleable, especially in the presence of a little antimony. The metal is then harder and paler in colour. Tin in very small quantity does not appear to affect the working properties of copper except to make it a little harder. The foregoing remarks concerning the effect of small quantities of foreign metals on the properties of copper do not apply in the same degree when metals are added to copper in considerable quantities to form what is generally under- stood by the term "alloys." For example, one per cent of zinc in copper will make it hard and red-short, but 20 per cent of zinc alloyed with 80 per cent copper will produce an exceedingly malleable alloy. Alloys of copper will be con- sidered as those alloys in which copper is the principal constituent, and not those in which copper plays a sub- ordinate part. Thus standard silver and standard gold contain copper, and these are not termed copper alloys, but silver and gold alloys respectively. The chief alloys of copper are Brass, bronze, and German silver, and these are the alloys most extensively used in the various industries. Alloys of copper and glucinum have been made in the electric furnace by Lebeau, by heating together the oxides of copper and glucinum. The alloy has a rose-red fracture, but is not homogeneous, and on heating yields alloys containing ii BRASS 105 5 to 10 per cent of glucinum. The 5 per cent alloy is said not to tarnish in air, and only slightly in sulphuretted hydrogen. It is malleable both hot and cold. The 1-3 per cent alloy is sonorous, and has a deep golden colour. Oxygen. If copper contains more than a small amount of oxygen, its mechanical properties and those of its alloys are injured. To remedy this it is usual to add a deoxidise r, such as phosphorus, manganese, aluminium, etc. Oxygen unites with copper to form the compound Cu 2 0. Heyn first showed that this compound forms with copper a eutectic mixture containing 3*5 per cent of Cu 2 O. With less than 3'5 per cent we have crystals of copper and eutectic. With more than 3'5 per cent we have crystals of Cu 2 and eutectic. The hardness and brittleness increase with increase of oxide. BRASS 23. Tiie term brass will be used in this work to signify all alloys of which copper and zinc are the essential and chief constituents ; but it is generally limited in the industrial arts to those alloys which are decidedly yellow, or have the yellowish tint characteristic of ordinary brass. Alloys of zinc and copper are known in commerce by a variety of names, and indeed, great confusion has been introduced by the multiplication of empirical names to represent one and the same substance. This is doubtless owing to the ignor- ance that formerly prevailed, when every mixture was jealously guarded as a great secret, and fanciful names given to hide the real composition. Moreover, some alloys have been handed down to us from very early times, and their names corrupted so as to have different appellations in different localities. Dr. Percy mentions "that the terms tombac, prince's metal, similar, and Mannheim gold are used by some authors to designate alloys consisting of about 85 per cent copper and 1 5 per cent zinc ; whereas, according to others, prince's metal and Mannheim gold are synonymous, and are 106 MIXED METALS CHAP. composed of 75 per cent copper and 25 per cent zinc ; according to another author, similar consists of aboiit 71 J per cent copper and 28| per cent zinc ; and Mannheim gold of 80 per cent copper and 20 per cent zinc ; and again, according to another author, similar and Mannheim gold are synonymous, and are applied to alloys of copper containing from 10 to 12 per cent zinc and from 6 to 8 per cent tin." l That brass was known to the ancients is beyond dispute, but its direct preparation from copper and zinc is an inven- tion of more modern times. " With the exception of iron, there is no produce of man's industry that is earlier spoken of than brass, but the word thus used in all probability referred to bronze, an alloy of copper and tin. The brazen serpent in the wilderness, the brass vessels of Solomon's temple, and the so-called brass armour and weapons of ancient Greece and Rome, were all doubtless of bronze. In England the earliest specimens we have of brass are monumental brasses or latten plates, of which some in Westminster Abbey are supposed to be 600 years old. St. Mary's Church, Warwick, has the celebrated Beauchamp Monument of the date of 1460, which is ornamented with brass. If the work was done in England the metal doubtless came from Flanders or Germany. Brass- working was practised in Bavaria in 1351, and Rudolph of Nuremberg was the first to draw wire after the present method. The art of making and working brass was definitely introduced into this country by Christopher Shutz, a native of Saxony, who, in partnership with William Humphrey, ' lay ' master of the Mint, obtained the exclusive right of getting calamine stone and working brass. Shutz is described as ' a man of greate cunnynge in ye mixed metalle called latten or brasse, alsoe in apting, manuring, and working ye same in alle sortes of batterye wares, caste worke, and wyre.' From this partnership arose a company or guild, having charters and privileges granted to it by successive monarchs. An old document states that Humphrey and Shutz devised tools or engines to draw brass wire, before which it was made 1 Percy's Metallurgy, p. 606. BRASS 107 and drawn by man's strength. Previous to that, wire, as well as latten plates and all kinds of brass vessels, was made by beating out or battery. An ancestor of Lord Byron estab- lished brass battery works at Nottingham. In 1700 the Bristol battery works were established, and this ancient style is continued in the battery company of Messrs. Gibbins of Birmingham. Bristol was the chief seat of the brass trade for the next 100 years, and the famous Cheadle Works were established in 1730. The place was chosen for its convenience of position, as Lancashire supplied the copper and Derbyshire the calamine. There was coal on the spot, and abundance of water power. For the old method of making brass, see 50. Stimulated by the success of Cheadle, other brass works were established in different parts of the country, "Wales taking the lead. The raw material was sent to various towns to be made up, of which a large part came to Birmingham. Here, in a population of metal workers, possessing hereditary aptitudes, which descended from father to son for many generations, was the destined home of the new industry. Mr. Aitken says, * In all probability there was queer fitting, abundance of material, while their finish was imperfect Of form there was little, of ornamentation less.' A very great leap in prosperity arose from the opening of the canal system in 1769, which caused a reduction in price of coal to one- third, and a cheapening of raw material as well as of finished goods. " At what time sheet-metal commenced to be made by rolling instead of the old battery process does not seem to be known ; but at the time of which we are speaking, metal- rolling mills, worked by water power, had been established. Of these,' one mill at Hockley passed into the hands of Matthew Boulton. The want of water power led some manufacturers to set up steam engines, on the old plan of Newcomen, but Boulton being aware of its many imperfections, was one of the first to see the advantage of the new engine of James Watt, which was beginning to be made known. From this the celebrated partnership followed, with such 108 MIXED METALS CHAP. marvellous results that these two names have become historic. Soho Works also became famous for the production of the copper coinage of 1797. The Boulton pennies, as they were termed, when first issued were sold at Ijd. and 2d. each, so great was the demand for them, as they were considered superior to anything of the kind ever produced before. The improved methods of rolling sheet brass led to its increased production, and by far the most important use of it was for the manu- facture of brass tubes for gasfittings, bedsteads, and boiler tubes. Solid drawn tubes have been forced into existence by the requirements of the railway system. The first recorded success was achieved by Mr. Charles Green in 1838, but since then many improved processes have been devised, and the manufacture of them brought to a high pitch of perfection. Another enormous development of the brass trade has resulted from the process of stamping by means of stamps and dies. The cheapness of the articles and their superior finish secure for them a large sale. This branch owes its existence to the large and perfect production of sheet metal. "The revival of ecclesiastical and mediaeval brasswork, chiefly by the great artist-architect Pugin, has tended to raise the character of brasswork to a higher level, and in the case of the best class of work to give to form and ornament their proper expression. Pugin himself says ' In metal work, such was the difficulty of procuring operatives, that we were compelled to employ for the first altar lamp an old German workman who made jelly moulds, as the only one who understood beating up copper to the old forms.' " 1 Mines containing the ores from which yellow metal was produced were highly esteemed, and much regret was ex- pressed when the ores were exhausted. In process of time it was observed that a certain ore (probably calamine ZnC0 3 ), when smelted in contact with copper, produced a metal of a yellow colour, and this process long continued to be used for making brass, without it becoming known what metal the ore contained, and what imparted the change of properties to 1 Short History of Brass Trade, by W. J. Davis, Birmingham. II BRASS 109 the copper. The preparation of brass, by direct mixing of the two metals, copper and zinc, was probably practised soon after the metal zinc was first isolated. In 1700, as already stated, zinc works were established at Bristol, and in 1758 a patent was granted to Mr. Champion of Bristol for making 1000 900 800 700 600 500 400 Liquid r c 90 80 70 60 50 Copper per Cent FIG. 16. " spelter " and brass, but the brass was still made by the indirect process from calamine ZnC0 3 , or calcined blende ZnO. Hutton states that the brass industry was introduced into Birmingham about 1740 A.D. by the family of Turner, and that the chief supply of the metal was derived from the Macclesfield, Cheadle, and Bristol companies. Fig. 16 is a portion of Shepherd's curve, who thoroughly 110 MIXED METALS CHAP. investigated the nature of the whole series of copper-zinc alloys. The liquidus curve, AB, BC shows the points of first solidification, and AD, DB, BE, the solidus curves, or points of complete solidification. These were obtained by means of microscopic study. The liquidus varies from 100 to 63 per cent copper, and the solidus from 100 to 71 per cent copper. Between 100 and 63 per cent copper the alloys consist of solutions, termed a, which begin to crystallise along AB, and are completely solid along AD. In the region marked a + /3 the mass consists of two solid solutions. In the region marked a there is only one solid solution. Along the curve BC we get the solid solution /3 beginning to crystallise, and in the region marked ft we have the solid solution /3 only. Alloys consisting of a are malleable and the same applies to those consisting of /3 with an excess of a. Alloys consisting of /?, or with only a small proportion of a, are brittle. All alloys with upwards of 71 per cent copper consist of a whether cooled quickly or slowly. The same applies to alloys between 71 and 63 per cent copper if quenched above 890 C. If quenched below that temperature they may consist of a, or of a + J3 according to the percentage. For example, if an alloy with 65 per cent copper is allowed to cool slowly below 750 C., it will consist entirely of a. If quenched above that temperature and below 890 C., it will be composed of a + /?. Thus some ft is transformed to a by slow cooling of alloys containing more than 60 per cent of copper. The presence of J3 along with a imparts to the alloy the property of being rolled hot or cold. If, however, an alloy with 58 to 60 per cent copper .is quenched above 780 C., it will consist entirely of /5 and will then be hard and brittle. Of all the constituents of brass the a is the most malleable and ductile, but with prolonged annealing at too high a temperature the crystals grow in size to such an extent that the metal becomes quite brittle. Prof. Carpenter l has shown that, in alloys containing the beta constituent, a thermal change occurs in brass at about 1 Institute of Metals, January 1911 and January 1912. BRASS 111 470 C. on cooling, which is due to a change of beta into alpha + gamma, the reverse change occurring on heating. Thus above 470 the stable phase is beta, below it is alpha + gamma, and consequently more brittle. It is sug- gested that in alloys containing only the alpha constituent below 470 are mixtures of copper + beta. The brittleness can be removed by heating to above 500 and allowing to cool This change by lapse of time explains the brittleness of some kinds of brass after being kept a long time in stock The line e e in Fig. 16 shows the critical point found by Carpenter. Commercial brass never consists entirely of copper and zinc, since whatever impurities exist in the separate metals will also be found in the alloy, though probably in smaller quantities, the most common of these being lead, tin, iron, and arsenic. It often happens that some of these are purposely added, to produce a given effect in the alloys. The colour of brass shows great variations, according to the proportions of the constituents, ranging from the red of copper at one end, to the bluish-white of zinc at the other. But the change from red to white is not so uniform as a casual observer might suppose. Thus alloys containing 94 to 99 per cent copper are red, with only a faint yellow tint ; with 87 to 93 per cent copper, the colour is reddish -yellow ; from 79 to 86 per cent copper, a yellowish-red tint prevails ; below this, down to 74 per cent copper, the alloys are yellow ; with a content of 67 J per cent copper, a reddish-yellow tint is obtained ; with 60 to 66 per cent copper the colour is a full yellow ; with 59 per cent copper, a reddish colour is ob- tained ; with 52 per cent copper the colour is nearly golden- yellow ; with a less quantity of copper than the above, the colour of the zinc begins to overpower the red colour of the copper, the alloys becoming more lead-like in appearance as the proportion of zinc increases. Brass may be made pliable and soft, hard or brittle, strong or weak, elastic or non- elastic, having a dead surface or one highly polished, by varying the composition, and by adding to the above list the 112 MIXED METALS CHAP. diverse changes of colour mentioned above ; it is perhaps not too much to say that it is more comprehensive in its pro- perties than that of any other kind of metallic matter. Copper and zinc may be united in all proportions, forming homogeneous alloys ; and the combination is usually attended with evolution of heat. Certain varieties of brass are ex- ceedingly malleable and ductile, and these properties, com- bined with the variety of shades of colour obtained by suitable mixing, and the moderate cost, render copper-zinc alloys most valuable for ornamental purposes. Brass possesses all the necessary advantages as a constructive material for works of art, and with the aid of transparent varnishes, termed lacquers, which have been brought to great perfection, it resists the action of the atmosphere remarkably well. The malleability of brass varies with the composition, with the temperature, and with the presence of foreign metals, which are sometimes in minute quantities. Some varieties are only malleable when rolled hot, others can be rolled at any temperature. Alloys containing up to 35 per cent zinc can be drawn into wire, but those containing 15 to 30 per cent of zinc are the most ductile. The alloy known as Dutch metal, which is an alloy of copper and zinc, containing more copper than ordinary brass, is an example of the great malle- ability of certain kinds of brass. The thickness of the leaves of Dutch metal is said not to exceed sa ^ 00 of an inch. Brass is harder than copper, and therefore better adapted to resist wear and tear. It acts well under the influence of a percussive force, as in the process of stamping, provided it is suitably annealed at proper intervals, in order to counteract the effects of local hardening, due to the com- pression of the particles into what may be termed unnatural positions. During the ordinary process of annealing the metal becomes coated with a scale of oxide, by union with the oxygen of the air, which oxide requires to be removed at each stage. This is done by dipping the metal in aqua- fortis, or dilute sulphuric acid, then scouring with sand if necessary, and finally well rinsing in water. A piece of BRASS 113 brass submitted to permanent deformation by mechanical treatment, such as rolling, is more or less hardened, and its limit of elasticity is raised. Between soft and hard brass there are many shades of difference. With the same rolled brass the author has obtained tensile strengths varying from fifteen to twenty-five tons per square inch before and after annealing. As will be afterwards pointed out, the temper- ature employed for annealing is of the greatest importance, for if the temperature is too low, the re-heating is inoperative and if too high, the metal is spoilt. With metal of large dimensions, such as sheets, bars, etc., it is necessary to raise the whole to the annealing temperature, otherwise some parts will not be sufficiently softened ; in other words, temperature and not time is all-important in annealing, and the subsequent cooling, whether quickly or slowly, is almost immaterial The melting point of brass is lower than that of copper, and this lower melting point is of the utmost importance in the case of re-melting, for casting or other purposes. The melting point of zinc is very much below that of copper, and when an alloy of these metals is strongly heated, the zinc volatilises, while the copper remains for the most part fixed, so that the loss of zinc will be considerable if the temperature is raised too high, and the operation of melting unduly prolonged ; moreover, the affinity of zinc for oxygen is far greater than that of copper, and if air is freely admitted a considerable portion of the zinc will be oxidised, so that the metal should be excluded from the air as much as possible, by covering it with a layer of charcoal, or other substance, which has no chemical action upon the metal. The moderate fusibility of brass, and its fluidity when melted, render it valuable for casting, as it is capable of receiving very fine impressions from the mould. Cast brass is generally more or less crystal- line, which is very pronounced in the brittle varieties. The mechanical properties of brass have been extensively studied by Charpy. 1 He took hard brass, produced by 1 Recherches sur les alliagts de cuivre et de zinc. Soc. d'encour. pour V Industrie nationale, 1886. 114 MIXED METALS CHAP. hammering and rolling cold, and tested its tensile strength before and after annealing at different temperatures. The test pieces were about *2 inches square, and 3^ inches long between the shoulders. For compression tests he used cylinders -52 inches high and '32 inches diameter. For fragility he employed a weight of 22 J Ibs. falling through a distance of 40, 80, and 120 inches, etc., till the bar was broken or bent to an angle of 90. The bars tested were 54 in. x -35 in. x 2-4 in. long. For all the alloys studied it was found that as a hard alloy was heated at gradually increasing temperatures the tenacity diminished, elongation increased, and the section of the ruptured part became less, except at temperatures approaching the fusing point. From the ordinary temperature up to a certain point the effect of re-heating is nil. The temperature at which annealing begins to be effective depends on the amount of compression the metal has undergone. With regard to brass with upwards of 40 per cent of copper, the temperature for annealing must be lower, as the quantity of zinc is increased, for the same treatment increases the hardness, as the zinc is greater. When the limit at which annealing begins to be effective is passed, then the softening effect is increased with each degree of temperature raised, as shown by the greater amount of elongation, till finally a point is reached at which the annealing effect is at a maximum. The metal then has the maximum malleability and ductility. It is important to know this zone of complete re-heating, since the lower the temperature required the quicker will the work be done, and the less will be the amount of fuel used. Take copper for example ; if annealed at 420 C. it will be as soft and malleable as if heated at 900 C. For brass with 30 per cent of zinc, the annealing temperature must be above 600 C. Beyond the zone of temperature within which annealing is most effective, the metal alters in structure and rapidly deteriorates as the temperature approaches that of the melting point. The metal is then said to be burnt. As a rule this burning zone will be higher the purer the metal. BRASS 115 Brass containing lead and tin will be burnt at temperatures at which pure brass will be unaffected. A sample of brass with 70 per cent copper, and containing '15 per cent of tin and *2 per cent lead, was burnt when heated above 800 C., while a similar alloy free from tin and lead was not sensibly burnt at 900 C. The tenacity of brass gradually increases as the zinc increases from to 35 per cent, then more rapidly up to about 45 per cent zinc, when it is at a maximum, then it rapidly decreases with more zinc. In a similar manner the elongation increases with the percentage of zinc, and reaches a maximum at about 30 per cent. Charpy summarises the mechanical properties of rolled brass perfectly annealed as follows : " The properties of copper-zinc alloys with 57 per cent of copper and upwards, vary continuously with the increase of zinc." With the exception of colour, there is little advantage in employing alloys with more than 70 per cent of copper, which become more costly, offer less resistance, and are less malleable. In varying the proportion of zinc between 30 and 43 per cent we get a variety of alloys of different shades, the most malleable having 60 per cent elongation and a tenacity of about 20 tons per square inch, while the most tenacious may reach 25 tons tenacity with 40 per cent elongation. The mechanical properties of the brasses have also been made the subject of investigation by Roberts- Austen. 1 He has shown that it is possible to trace the relation between the freezing point curve and the curves which represent the mechanical properties of tenacity and extensibility. Many of the brasses have more than one freezing point, and the subsidiary freezing points represent eutectic alloys. These alloys were separated by squeezing away the fluid portions in a press. The brass under examination was placed in a steel cylinder, fitted with loose plungers, and the whole placed between the jaws of a hydraulic press. The cylinder was 1 Fourth Report, Alloys Research Committee, Inst. Mechan. Eng., 1897. 116 MIXED METALS CHAP. then gradually heated, and the temperature at any period measured by means of a thermo-couple. At a certain definite temperature part of the alloy will liquefy, and can be squeezed out between the plungers and the wall of the cylinder. The temperature was carefully noted at which any extrusions took place, and the various samples were then analysed. The maximum strength in a series of cast brasses occurs in the alloy containing 60 per cent copper, and this has practically but one freezing point, although a feeble eutectic is revealed at 450 C. It is a well-known fact that the above composition, with a little iron added, has an increased tensile strength, but the cause was obscure. An alloy was made, consisting of 61 copper, 39 zinc, with ij per cent of iron. This, upon being tested and a cooling curve obtained, showed that the low eutectic point was absent, and therefore the source of weakness had been removed. More- over, the mean solidifying point was higher than that of the brass without the iron, which in itself is an indication of augmentation of strength. TENACITY OF BRASS AND OF AICH'S METAL Tenacity, per square inch. Temperature. Brass. Aich's Metal. 20 C. 207 Tons. 25-6 Tons. 100 C. 13-6 22-2 200 C. 12-6 17-9 300 C. 7'3 12-4 350 C. 6-8- 11-2 450 C. 3-6 5-1 500 C. 2-8 4-1 II BRASS 117 AVERAGE RESULTS OF VARIOUS COPPER-ZINC ALLOYS Composition. Tenacity in tons Elongation Crushing Strength in per sq. in. per cent. tons per sq. in. Cu. Zn. 90 10 15 27 15 80 20 16 32 16 70 30 22 29 21 66 33 21 27 21 60 40 28 20 37 Where strength is required it would appear to be advisable, whenever the presence of a low eutectic in an alloy is revealed, to add some third metal which will diminish the fusibility of the eutectic. The presence of an eutectic in an alloy also naturally diminishes the extensibility of any mass which contains more than a small amount of it. The formation of copper-zinc alloys is generally attended with contraction, which attains its maximum in the alloy CuZn 2 , containing 32'6 per cent of copper. This alloy is brittle, and exhibits none of the characteristic properties of the constituent metals. The density of brass is increased by mechanical treatment, but this effect is annulled by sudden, and still more by slow cooling after annealing. 1 In mixing brass for casting, old copper from worn-out articles, scrap brass, etc., are frequently used, and as these often contain injurious ingredients, which modify the pro- perties of the alloy, great care should be exercised in the selection. For some kinds of casting this is not important, but when the metal is required for rolling into sheet, or drawing into wire, or for making the best kinds of brass tubing, the use of impure metal is often fatal, entailing a considerable amount of expense on account of waste, and much annoyance to those concerned. The most common impurities, as already stated, are lead, tin, iron, and arsenic, 1 Riche, Ann. Ghim. Phys. (4), xxx. 118 MIXED METALS CHAP. all of which harden the metal and tend to make it brittle. For brass intended for filing and turning, 1 to 2 per cent of lead is added, in order to prevent the unpleasant fouling of the tools in working. Brass containing lead should be very thoroughly mixed before pouring, and the cast metal should be cooled as quickly as is expedient, otherwise the lead separates out in the lower portion of the casting, producing unsightly spots. A little tin is often an advantage in brass ; it renders the metal more easily fusible, less brittle, somewhat sounder, and enables it to take a better polish. A little iron considerably increases the hardness of brass, lightens the colour, and such metal is more easily tarnished by the atmosphere than brass free from iron. When an ingot of ordinary brass is broken while hot, its fracture is coarsely fibrous, but when broken while cold, it should be finely granular. When the fracture of a cast ingot of certain metals is fibrous, the directions of the fibres will be at right angles to the cooling surface. In the case of a sphere, the fibres will have the direction of radii ; and in the case of a square, two diagonals will be plainly visible on the transverse fracture, formed by the points of junction of the internal extremities of the fibres. 1 Mr. F. H. Storer 2 states " that the tendency to shoot into fibres extends from alloys containing 57 or 58 per cent of copper down to those con- taining 43 to 44 per cent, where it gradually disappears. The inclination to form fibres is strongest in those alloys which contain nearly equal atomic proportions of copper and zinc, being less clearly marked as one recedes in either direc- tion from this point, until a stringy texture, analogous to that of copper, is reached on one hand, and the peculiar pastiness of zinc on the other. In preparing crystals, this pastiness manifests itself decidedly in the alloys immediately below those which are fibrous, becoming more strongly marked as the alloys are richer in zinc. The fracture of these white alloys is for the most part vitreous or glassy." Brass is occasionally obtained in well defined crystals. 1 Percy's Metallurg., p. 608. 2 Mem. of Amer. Acad. 1860 (8), p. 35. ii BRASS 119 Storer prepared the most perfect individual crystals from brasiers' solder, which consists of equal parts by weight, of copper and zinc, and occurs in the state of coarse powder, produced by heating the alloy to a sufficient degree and pounding it in a mortar while hot The alloy containing 5 to 6 per cent of zinc was found to crystallise remarkably well It will be seen from the above remarks that the crystalline condition of copper- zinc alloys does not depend on an excess of zinc, as might be presupposed from the highly crystalline character of zinc. As before stated, zinc becomes malleable when worked at a temperature of from 100 to 1 50 C., but at higher temperatures it again becomes brittle. It is assumed that the brittleness is intimately connected with the crystalline condition- Kalischer examined different varieties of sheet brass having the following composition : I II III IV Copper ... 66 62 '5 60 56 -8 Zinc ... 34 37'5 40 43 '2 100 100-0 100 100-0 Nos. I and II were crystalline, No. Ill showed traces of crystallisation, and No. IV did not become crystalline even by heating. I III IV Copper. . . 73-64 80'38 90'09 Zinc . . . 25-96 19 '29 9 '91 Tin -40 "33 100-00 100-00 100-00 The above three specimens were all crystalline. i ii Copper .... 90 88 '23 Zinc .... 5 8 '82 Tin .... 5 2-95 100 100-00 120 MIXED METALS CHAP. In the above specimens no crystallisation could be detected. Charpy l states that brasses containing not more than 35 per cent of zinc are entirely made up of a network of needle- shaped crystals. Above 35 per cent of zinc the crystals do not fill the whole mass, but are embedded in glassy magma. The crystals become fewer as the zinc is increased, until when the zinc reaches 67 per cent, a homogeneous structure, with conchoidal fracture is obtained. These facts explain certain known mechanical properties of such alloys. Thus in com- mercial brasses, containing less than 35 per cent of zinc, small quantities of lead and tin form, as it were, a solder between the crystals, and weaken the alloys when they are raised above 200 C. When there is more than 40 per cent of zinc the lead and tin are embedded in the glassy magma, and do not weaken the structure ; therefore, such alloys can be rolled hot. It is probable that alloys from 1 to 34*5 per cent zinc are isomorphous mixtures of copper and Cu 2 Zn ; alloys with 35 to 67 per cent zinc are mixtures of Cu 2 Zn, a malleable substance, and of CuZn 2 , a hard and brittle substance ; alloys with more than 67 per cent zinc are mixtures of zinc and CuZn 2 . 2 The heats of formation of a number of alloys of copper and zinc, containing those metals in very diverse proportions, have been ascertained. The method consists in finding the difference between the heats of dissolution, in suitable solvents, of an alloy and of an equal weight of a mere mixture containing the metals in the same proportion. The first series of experiments was made with an aqueous solution of chlorine as solvent. Its application was limited to those alloys containing less than 40 per cent of copper, as it was impossible to obtain those richer in copper in a sufficiently fine state of division to enable them to dissolve. 1 Comp. Rend. 122, 670. 2 Baker, Phil. Trans. Roy. Soc., vol. 196, p. 529. ii BRASS 121 The results, though not altogether satisfactory, showed that the heat of dissolution of an alloy was sensibly less than that of the merely mixed metals. Incidentally it was found that the equation Cl 2 .Aq = 2600 (Thomson's Thermochemische Untersuchungeri) is erroneous, and, on inquiry, Professor Thomsen gave a corrected value, 4870. The author finds Cl 2 .Aq = 4970. The most suitable solvents of the alloys are (a) Mixture of ammonium chloride and ferric chloride solutions. (6) Mixture of ammonium chloride and cupric chloride solutions. The chemical actions involved are simple reductions, and no gases are involved. Two series of experiments made on twenty-one alloys yielded very concordant results. They show that heat is evolved in the formation of every alloy of copper and zinc yet tested. A sharply denned maximum heat of formation is found in the alloy containing 32 per cent of copper, i.e. correspond- ing to the formula CuZn . It amounts to 52*5 calories per gramme of alloy, or 10,143 calories per gramme-molecule. There is some evidence of a sub-maximum in the alloy nearly corresponding to CuZn. From these points there is a steady decrease in the heat of formation, both in the case of alloys containing less than 3 per cent of copper as the amount of copper decreases, and also in the case of those containing more than 50 per cent of copper as the quantity of copper increases. The results, in general, confirm the existence of inter- metallic compounds, and the values obtained are in accord- ance with those demanded by Lord Kelvin's calculation of the molecular dimensions of copper and zinc. The microscopic examination of alloys of copper and zinc reveals the existence of special characteristics which admit of classifying them into a number of groups, each group showing a similarity of structure under normal condi- 122 MIXED METALS CHAP. tions. 1 Alloys with 1 to 35 per cent of zinc are distinguished by dendritic crystals, formed of long rectilinear needles and united in groups, but too badly formed to admit of measuring the angles, although the network of crystals presents a large number of right angles. The size of these crystals varies with the rate of cooling, and generally the size is increased when the metal has been poured at a high temperature and allowed to cool slowly. Pouring at a lower temperature and quick cooling have the contrary effect. As the rate of cooling also modifies the mechanical properties, we may infer the properties from the relative size of the grain. For a given alloy the tenacity is greater as the grain is closer. When a metal is maintained at a high temperature, varying with the composition, the crystalline grains become more clearly defined and apparently larger. When a certain temperature is exceeded the dendritic structure is modified, the crystals are much larger, somewhat octagonal in shape, and with numerous well-defined angles. The temperature necessary to produce this change is, for alloys with 70 per cent copper, from 750 to 800 C., and for alloys with 80 to 90 per cent copper, 850 to 900 C. Alloys containing more than 35 per cent of zinc and less than 45 per cent, after being polished and etched, exhibit curved crystalline grains completely interlaced, which are larger or smaller according to the rate of cooling, and the temperature at which the metal has been cast or re-heated. The etching liquid seems to dissolve an amorphous magma, leaving imperfectly formed crystals limited by curved lines. These crystals seem to be malleable, and are broken with difficulty by mechanical treatment, which makes the rolled or hammered metal to differ but slightly from the cast metal, unless the pressure is carried very far. (See p. 110.) Alloys with more than 45 per cent of zinc show none of the crystalline needles so characteristic of the two former series. The surface is like mosaic, formed of fragments of 1 Charpy, " Recherches sur les alliages," Soc. d'encour. pour V Industrie nationale, 1886. BRASS 123 metal, limited by hexagons, and assumes different colours when attacked by the etching liquid. Solidification appears to have commenced round a great number of points simultane- ously, and around these points has been developed a solid core, sensibly homogeneous. Where these cores are in contact, they are bounded by surfaces of which the plane sections are hexagons. Alloys of this class are brittle and cannot be rolled. When cast in large masses, solidification is less regular, and [elongated patches are formed perpendicular to the surface of cooling. As the proportion of zinc increases smaller crystals are formed in the larger ones. The alloy with about 33 per cent of copper, corresponding to the definite compound CuZn , has a vitreous fracture, and etching does not reveal any variation in structure. When the copper is less than 33 per cent, the excess of zinc can be readily dis- solved by hot potash, and the surface then exhibits fine striae formed of parallelograms. As the proportion of zinc increases we find more numerous and larger crystals, which probably consist of zinc dissolved in the compound CuZn^ The microscope permits of determining whether brass has been burnt. When the alloy has been raised to a temperature at which the mechanical properties begin to alter, cavities appear like gas bubbles, which become more numerous as the temperature is raised, and form fissures around the crystals, thus producing a continuous network. The metal then has completely deteriorated. It would appear that a fusible alloy is formed, which melts and flows round the crystals, especially if lead and tin are present. These indications explain why commercial brass is more easily spoiled by annealing than pure brass. 24. The following table of the composition and pro- perties of copper-zinc alloys is taken from Mr. Mallet's table and Karsten's observations : 1 1 Percy's Metallurgy, p. 611. 124 MIXED METALS CHAP. Tensile No. Atomic constitution. Composition per cent. Specific gravity. Colour. Order of ntensity. Fracture. strength in tons per square inch. 1 Cu 100 8-667 24-6 2 10 Cu+Zn 90-7 + '3 8-605 Reddish-yellow . 1 c'.c. 12'1 3 9 Cu+Zn 89-8 +10-2 8-607 2 F.C. 11-5 4 8 Cu+Zn 88-6 +11-4 8-633 H 3 F.C. 12-8 5 7 Cu+Zn 87-3 +127 8-587 n 4 F.C. 13-2 6 86-3 +13-7 8-705 7 6 Cu+Zn 85-4 +14-6 8-591 Yellowish-red . 3 F.F. 14 : 1 8 84-0 +16-0 8-639 9 5 Cu+Zn 83-02+16-98 8-415 Yellowish-red . 2 F.C. 13 : 7 10 80-6 +19-4 8-679 11 4 Cu+Zn 79-65+20-35 8-448 j| 1 F.C. 14 : 7 12 75-9 +24-1 8-609 Brass-ye'llow 13 3 Cu+Zn 74-58+25-42 8-397 Pale yellow F.C. 13 : 1 14 73-8 +26-2 8-582 Brass-yellow 15 67-5 +32-3 8-499 Reddish brass- yellow . 16 2 Cu+Zn 66-18+33-82 8-299 Full yellow 1 F.C. 12-5 17 60-0 +40-0 Yellow 18 59-0 +41-0 8 : 375 Reddish-yellow . 19 52-0 +48-0 8-229 Nearly gold- .. yellow . 20 Cu+Zn 49-47+50-53 8-230 Full yellow 2 C.C. 9-2 21 46'5 +53-5 Reddish-white . 22 44-0 +56-0 White, but viewed obliquely deep yellow 23 35-5 +64-5 Bluish-white" . 24 Cu+2 Zn 3-2-85 + 67-15 8-283 Deep yellow? . c'-c. 19 : 3? 25 8Cu+17Zn 31-52+68-48 7-721 Silver white? . V C. 2-1 26 8 Cu+18Zn 30-30+69-7 7-836 1 V.C. 2-2 27 8Cu+19Zn 29-17+70-83 8-019 Silver-grey . 3 C. 0-7 28 8 Cu+20 Zn 28-12+71-88 7-603 Ash-grey . 3 V. 3-2 29 8 Cu+21 Zn 27-10+72-90 8-058 Silver-grey . 2 C. 0-9 30 8Cu+22Zn 26-24+73-76 7-882 1 C. 0-8 31 8Cu+23Zn 25-39+74-61 7-443 Ash-grey . 4 F.C. 5-9 32 24-8 +75'2 Bright lead-grey. 33 Cu+8 Zn 24-50+75-50 7-449 Ash-grey . 1 F'.C. 3 : 1 34 21-0 +79-0 Bright lead-grey. 35 Cu+4 Zn 19-65+80-35 7-371 Ash-grey . 2 pic. i'-9 36 37 Cu+5 Zn 16-36+83-64 14-7 +85-3 6-605 Very dark-grey . Bright lead-grey. F.C. 1-8 38 11-25+88-75 M 39 9-5 +90-5 Dark lead-grey . 40 Zn 100 6-895 15 : 2 "Remarks on the preceding Table. The figures in column 6 represent intensity of tint of the same colour. In column 7 the letters are abbreviations for the characters of the fractures according to Mallet's nomenclature, thus F.F. ii BRASS 125 fine-fibrous, C. conchoidal, V.C. vitreous-conchoidal, F.C. fine-crystalline, C.C. coarse-crystalline. The tensile strength was determined on prisms ^ inch square, without having being hammered or compressed after being cast ; and the weights given are those which each prism just sustained for a few seconds before disruption. " 6. Approaching tombac in colour, is not quite so good for rolling, hammering, or wire drawing as common red brass, which contains less copper in proportion to the zinc. " 8. Comes very near ' red brass,' which it quite equals with respect to its working qualities. " 1 0. The reddish colour is now beginning to pass into brass-yellow \ it behaves itself irreproachably under the rolls and hammer, and in wire drawing. " 13. Working qualities quite the same as those of com- mon brass. Karsten remarks that it was this alloy which was puffed for some time in England under the name of Mosaic gold. " 14. In working qualities not different from common brass. " 16. It contains 4 per cent more zinc and 4 per cent less copper than common brass. The colour is no longer pure brass-yellow, but has a reddish tint, works excellently under the rolls and hammer, as well as in wire drawing. " 1 9. Much to be recommended, on account of its golden colour, and though it possesses greater tenacity than the preceding, and less hardness, is brittle. Well adapted for cast ware. Very flexible after casting while still hot. When cold it is hard to cut, and is rather fractured than cut. The ingots after being cut, annealed, and cooled, can no longer be rolled. If, after heating and cooling to the temperature of boiling water, they are rolled, they produce only very brittle, fragile sheets, much cracked at the edges, notwithstanding that frequent annealing during rolling may have been resorted to ; it is quite unsuitable for rolling, hammering, or wire drawing. " 20. Very flexible while strongly heated. Rolls tolerably 126 MIXED METALS CHAP, n well cold, though it frequently cracks at the edges. During rolling into thin sheets, it requires more frequent annealing than common brass. It cannot be soldered, as it melts at the same temperature as the solder. It is not adapted for wire drawing. "21. Fracture reticulated. Possesses a tolerable degree of tenacity, but by hammering becomes so brittle that at no temperature can it be sufficiently drawn out for practical purposes. "22. The lustre is so imperfectly metallic that it would hardly be recognised as a metallic alloy. Its great brittleness and frangibility, combined with its perfectly conchoidal and smooth fracture, communicate to the alloy more the appear- ance of a sulphur compound than that of a mixture of two metals. " 23. More brittle than the last ; fracture smooth, even, small-conchoidal ; lustre strongly metallic. " 25, 26, 27. Very brittle. Too hard to file or turn ; lustre nearly equal to that of speculum metal. " 28, 29. Brittle. Too hard to file or turn ; lustre nearly equal to that of speculum metal. " 30. Very brittle. Too hard to file or turn ; lustre nearly equal to that of speculum metal. "31. Barely malleable. "32. Brittle ; fracture granular, uneven ; has a tolerable degree of tenacity, but cannot be worked under the rolls or hammer. "33. Brittle. " 34. In external character and working qualities closely resembles No. 32. "36. Brittle. " 38. Fracture finely granular. "39. Has sufficient tenacity to admit of being somewhat extended under the rolls and hammer, if, after annealing, it is not allowed to become cold." ]* . **!* g gig I : tl 00000000 00 position nalysis. Composition of original mixture. < Oi-ii-l es a UH^ 5 A& i ^ | ca g S -a - ggs Q - ag = =5= g^s g ^ *, 5 r-, OC 00 S .:::: : : : : : . : . : : :* . . 00 ... . . . -t~. . : : 2 : : : : :- I : : [TT^j ::::::::: : : - : : : : ::::; :::;-: ::::::::: o :| : : : : g :S : : : :| : : : g : : S r " 1 B .|| I .1 -I -1 - - 3 - - S 3 '^fsf o" *cT ' rn" cT oo" ooo eowtM oo eo(M : : ft : 1 : :?Sc ^ ^ & S> ^ ^ lt| ' ' ' i if!' '^ ' '. 8 " " 1 fl S3 - i| -if : Ig* 1 1 al .1.1.1 |= 2 - | -| a| 'f " ' ' t " ' I >>, >>>> >> >> 1^> 30 OO 00 00 0000 00 00 00 00 00 00 00 ? . irt OO t ^ o2JS* SS S K 1 s |5*N HI ?-l^ $ : o It ill 00 00 00 00 s g OOW-^D ^< p p b-,0 ,p C^cOOOOO COCO '' "*i* ^ ^|CO ^ ^ ^. QO COCCCOCO COCO CO CO CO CO CO CO CO a an tt^N DO oo 1 I ^*. J :=- s| I PIP . ' S s Sa IS : I" S rf 03 QQ PQ OQ QQ PQ 2 ? $ . 1 ' s" : fltitl . /; - ir P a? *^ .-3 iilll :: i! >>>> 3, 00 00 00 1~( I -^1 CO ; 55 : : i-H 1-H C* 9 ? 9 5 g O CM : g I p t- O rH (M t- t-t- S ? X N X S ^ ^ 00 -s~~$ i 1 s S3 S 5 Sg 5 ill . ,!..!! 1! ... 1|1 M '1 " ' JI! ' ' ' o - * ^ d s g o s g 5 .p^ "/ .OQcd-2 QQ3QGQ (::::::- . . . .... ... : : S : : O s S S :::*::: ; ; r e ; ; : : '.'.'. , a ' |= : - |= : : : : : 1 o . . o 2 : : : S : : : : : : : ti : : i :: I f : : : : : :| : r | ? I : o : :o : : : o : : : o : : : : : : : 1 : 1 1 : ! 1 1 : S : : 1 .3:811 "*" o" -J >J>J>J 5 c 2 1 C5 O o o y i s" -s >> g >:=>> ijC .2 ; tC .5 g .2 g - - - ^f - '5 c r S n ' 'g z. "z 2 " = 's "2 eg - - - - 5 1 CO O t O> O OO E || i 1 ?ils 1 S 1 H .,...??. Ol 1-* QD t* t*- : ' s : : s s s 1 1 S : :; 1 I : o o i o o CM tp ftp tf) OO 00 o p Op O IS 2 22 2 2 1 " 1 S 2 *"* *" 6 DC '="='= ' "s s ' ' s ... o oo o o o 2 g - ?jr, r oJ SS co cow w w co l m ill ililll P! lt Composition by analysis. omposition of original mixture. ii tn C t, C ^ 4S542S ' 11 111 000 1 1 S sss - i CHAP. IT BRASS 135 Professor R. H. Thurston, who conducted the investigations for the United States Board, makes the following remarks on the preceding table : "Alloys having the name of Bolley appended give com- positions and commercial names, and mention valuable properties, such as are given in the column of remarks, but do not give results in figures as recorded by other authorities. The same properties and the same name are accorded by Bolley to alloys of different compositions, such as those which in the column of remarks are said to be suitable for forging. It might be supposed that such properties belonged to those mixtures, and not to other mixtures of similar composition. It seems probable, however, that when two alloys of different mixtures of copper and zinc are found to have the same strength, colour, fracture, and malleability, it will also be found that all alloys between these compositions will possess the same pro- portions ; and hence, that instead of the particular alloys mentioned only being suitable for forging, all the alloys between the extreme compositions mentioned also possess that property. "In the figures given from Mallet under the heads of order of ductility, order of malleability, hardness, and order of fusibility, the maximum of each of these properties is represented by 1. " The figures given by Mallet for tenacity are confirmed by experiments of the author, with a few very marked exceptions. These exceptions are chiefly the figures for copper, for zinc, and for CuZn 2 (32-85 copper, 67 '15 of zinc). The figures for CuZn.,, as given by Mallet, can, in the opinion of the author, only be explained on the supposi- tion that the alloy tested was not CuZn , but another containing a percentage of copper, probably as high as 55. The figure for the specific gravity (8'283) given by Mallet in- dicates this to be the case, as does the colour. The figure for ductility would indicate even a higher percentage of copper. The name watchmakers' brass in the column of remarks 136 MIXED METALS CHAP. must be an error, as that alloy is brittle, silver-white, and extremely weak. " The figures of Calvert and Johnson and Riche, as well as those of the author, give a more regular curve than can be constructed from the figures of Mallet. "The specific gravities in Eiche's experiments were ob- tained both from the ingot and from powder. In some cases one, and in some cases the other, gave highest results. In the table under the head of specific gravity Riche's highest average figures are given, whether these are from the ingot or from the fine powder, as probably the most nearly correct. The figures by the other method, in each case, are given in the column of remarks. The figures of Riche and Calvert and Johnson are scarcely sufficient in number to show definitely the law regulating specific gravity to com- position, and the curves from their figures vary considerably. The figures of the author being much more numerous than those of earlier experimenters, a much more regular curve is obtained, especially in that part of the series which includes the yellow or useful metals. The irregularity in that part of the curve which includes the bluish -grey metals is, no doubt, due to blowholes, as the specific gravities were in all cases determined from pieces of considerable size. If they were determined from powder, it is probable that a more regular set of observations could be obtained, and that these would show a higher figure than 7-143, obtained from cast-zinc. Matthiessen's figure for pure zinc (7 '148) agrees very closely with that obtained by the author for the cast-zinc, which contained about 1 per cent of lead. " The figures for hardness given by Calvert and Johnson were obtained by means of an indenting tool. The figures are on a scale in which the figure for cast-iron is taken as 1000. The alloys opposite which the word ' broke ' appears were much harder than cast-iron ; and the indenting tool broke them instead of making an indentation. The figures of alloys containing 17'05, 20'44, 25*52, and 33'94 per BRASS 137 cent of zinc have nearly the same figures for hardness, varying only from 427-08 to 472-92. This corresponds with what has been stated in regard to the similarity in strength, colour, and other properties of alloys between these compositions." [TABLE 138 MIXED METALS CHAP. 26. TABLE OF VARIOUS COPPER-ZINC ALLOYS Name. Authority. Copper. Zinc. Tin. Lead. Iron. 1. Brass, English . Lavater 70-29 29-26 17 28 2. , , Heegermahl > 70-16 27-45 79 2 3. ,, Augsburg . 70-89 27-63 85 ... 4. ,, Neustadt . Kadernatsch 71-36 28-15 ... 5. ,, Romilly . Chaudet . 70-1 29-9 6. ,, Unknown . Karsten 71-5 28-5 7- Regnault . 71-0 27-6 trace 1-3 8- Chaudet . 61-59 35-33 25 2-86 9. Stolberg . 65-8 31-8 25 2-15 10. Watch wheels . Faisst 60-66 36-88 1-35 ... 74 11- ,, f 66-06 31-46 1-43 88 12. Ship nails, bad . Percy 53-73 41-18 ... 4-72 13. good. 63-62 24-64 2-64 8-69 14. Tombac, English Faisst 86-38 13-61 ... trace 15. ,, German Karsten 84-0 15-5 16. Coin of Titus Giraldin . 81-4 18-6 Claudius 17. Coin of Titus, 79 Phillips . 83-04 15-84 5 A.D. 18. Coin of Hadrian, )5 85-47 10-83 1-14 1-73 74 120 A.D. 19. Coin of Faustina, >5 * 79-15 6-67 4-97 9-18 jun., 165 A.D. 20. Antique bracelet, Goebel 83-08 15-38 1-54 ... Naumberg 21. Statue of Louis D'Arcet . 91-40 5-53 1-7 1-37 XIV. 22. Statue of Napoleon . 75 20 3 2 23. Brass for gilding . 82 15-5 2-5 ... 24. . . . , 64-5 32-5 2-5 25. ,, . , 82 15 3 26. . 78 20 2 27. Brass, colour pale Kbnig 83-33 16-60 ... yellow 28. , , deep yellow . 84-5 15-3 29. red yellow . 90 9-6 ... 30. ,, orange . 98-93 73 ... 31. ,, copper-red. 32. ,, violet 99-9 99-22 5 trace 08 trace 33. ,, green 84-32 15-02 " 3 BRASS 139 27. SOME VARIETIES OF MODERN BRASS Name. Colour. Copper. Zinc. Tin. Lead. Iron. Gold. Jewellers' gilding Red 94 6 alloy Pinchbeck . Reddish-yellow 90-5 88-8 7-9 11-2 1-6 ... Red 93-6 6'4 Oreide (French gold) Reddish-yellow 90 10 Talmi gold . Tissier's metal with Gold Red 9070 97 8-33 2 ... ... 97 one per cent of arsenic Tournay's alloy Yellow . 82'54 17-46 Rich sheet-brass 84 16 Bath metal, similor, etc. Dutch alloy . Bristol sheet-brass Brass wire . Prince's metal Bright yellow Yellow . 80 76 72-8 70 75 20 24 27 30 25 ... : 2 ... ... Sheet and wire brass Full yellow . 67 33 Mosaic gold ordin- 66 '6 33-3 ary brass Bobierre's metal 66 34 Muntz's metal 62 38 5 J } J Gedge's metal Common brass 60 60 64 40 38-5 36 ... 1-5 Aich's metal . French brass (Potiii jaune) Hamilton's metal, Chrysorin French brass for fine castings Sterro metal . Hard solder for copper or iron Hard solder for brass Grey-yellow . Full yellow . 60 71-9 64-5 71 55-5 57 50 38-2 24-9 32-5 24 42 43 50 i-2 3 2 2 : 6 2-7 3 1-8 2-5 ... Dipping brass White brass 53 34 47 66 ... Lap alloy 12-5 87-5 140 MIXED METALS CHAP. FREEZING-POINTS OF COPPER-ZINC ALLOYS Percentage of Temperature in degrees. Copper. Zinc. Centigrade. Fahrenheit. 100 1082 1980 96 4 1075 1967 86 14 1032 1890 80 20 1008 1846 76 24 980 1796 72 28 958 1756 71 29 952 1746 66'4 33-6 918 1684 63 37 908 1666 60 40 890 1634 50 50 880 1616 48 52 870 1598 41 59 840 1544 35 65 816 1501 33 67 S03 1477 29 71 786 1467 24 76 740 1364 20 80 705 1301 15 85 657 1212 11 89 588 1090 8 92 547 1017 3-5 96-5 465 869 2 98 425 797 100 419 786 SHEET BRASS AND WIRE BRASS 28. For rolling into sheets, and drawing into wire and tubes, it is in the highest degree essential that the metals employed in alloying should be pure and consist only of copper and zinc. Two of the most injurious substances to brass are bismuth and antimony, which are occasionally found in common copper. It is very important that all copper used in making brass for stamping and drawing should con- tain the smallest possible quantity of cuprous oxide that is, SHEET BRASS 141 it should be cast at the highest tough pitch ; because if it contained any considerable quantity of that oxide, the latter acts on the zinc, producing a highly infusible zinc oxide, which diffuses itself through the brass and impairs its strength. Mr. E. S. Sperry states that cracks produced during the rolling of brass are often due to antimony. In an alloy of 60 parts Cu and 40 parts zinc he found that -006 per cent of antimony was immaterial, but with '02 per cent the metal showed incipient cracks, and he considers that some kinds of electrolytic copper fail from this cause when used in making Brass laminates well in the rolling mill in the cold, as long as it is kept sufficiently soft by occasional annealing, which, on account of the expansion of the mass by heat, releases the strain put upon its particles by the pressure of the rolls. The malleability and ductility of brass depend upon the composition ; and for alloys of a full yellow colour, the most suitable proportions are found between 66 to 73 per cent of copper and 34 to 27 per cent of zinc. The following table gives the composition of a few English and foreign varieties : Copper. Zinc. Tin. Lead. 927 4-6 27 91-6 8-4 ... 90 10 85-5 14-5 ... 83 17 79'5 20 5 76 24 ... 75 25 ... 73-5 26-2 3 70 30 68 32 67 32 5 5 66 34 65 35 ... 142 MIXED METALS CHAP. For hot-rolling, the alloys known as Muntz's metal, or more commonly yellow metal, are employed. They vary in composition from 56 to 63 per cent of copper and 42 to 37 per cent of zinc. 1 Mr. George Frederick Muntz of Birming- ham took out his first patent for yellow metal as a sheathing for bottoms of ships and other vessels in 1832. The pro- portions specially recommended in the specification are 60 per cent of copper and 40 per cent of zinc ; but these pro- portions may be varied from 56 up to 63 per cent of copper. Best selected copper and foreign zinc are directed to be used. The metal is cast into ingots, and rolled while hot into sheets, which, when finished, are pickled in dilute sulphuric acid to remove the adhering scale, and afterwards swilled in water. In the same year Mr. Muntz obtained a second patent for an improved manufacture of bolts, and other ship fastenings. Precisely the same proportions of copper and zinc are claimed as in the first patent. In. 1846 a third patent was granted to Mr. Muntz for the use of an alloy consisting of 56 per cent of copper, 40j of zinc, and 3j of lead. In the specifi- cation it is directed that only the purest metals should be used, and that the alloy is to be cast into ingots, which are to be rolled at a red-heat, and treated in other respects in the same manner as stated in the specification of the first patent. Dr. Percy states that he has succeeded in rolling brass well, which, on subsequent analysis, was found to contain not less than 8 per cent of lead. The theory assigned by Mr. Muntz for the application of his alloy is, that by exposure to sea-water the zinc is slowly and uniformly corroded over the entire surface, whereby the attachment of barnacles, etc., is prevented. Yellow-metal sheathing soon entirely superseded copper sheathing in the merchant service. Its special advantages are stated to be, that it keeps the bottoms of ships cleaner, and costs consider- ably less than copper sheathing. An experienced yellow-metal manufacturer states that the proportion of zinc in the alloy should not exceed 38 per cent ; and that if it exceeds this 1 Percy's Metallurgy, p. 619. SHEET BRASS 143 proportion, the sheathing is apt to become friable ; and that if it is sensibly below this proportion, it wears away too rapidly. In 1874 an alloy composed of 62 parts copper, 37 parts zinc, and 1 part tin, was proposed by Mr. Farquharson for naval brass. To ensure the best results he recommends that Australian or English B.S. copper be used. Unannealed, in rods or sheets of moderate thickness, the metal has a tensile strength of 67,000 to 72,000 Ibs. per square inch, according to the amount of rolling it has received. The composition of brass for ordinary wire drawing varies from 67 to 72 per cent copper, and 33 to 28 per cent of zinc. It has been stated that a little lead is sometimes added to brass intended for rolling, but this is not admissible in brass intended for wire drawing, as less than -5 per cent diminishes the tenacity ; and the ductility of a metal largely depends upon its tensile strength. Wire-drawers state that if brass wire is annealed immediately it has been taken off the drum, on which it has been wound during the process of wire drawing, it will fly to pieces. This effect is prevented by subjecting the coil of wire, after its removal from the drum, to strong concussion, by seizing one part of the coil with the hands, raising it, and heavily striking a bench with the other part, repeating the process several times. Some kinds of brass wire undergo a molecular change in process of time, especially if subjected to vibration, and become extremely brittle. Brass chains, used to support objects, such as chandeliers, etc., have been known to lose their tenacity, become brittle, and break without any apparent cause. Ship bolts of Muntz's metal are sometimes found to undergo a singular kind of exfoliation, the metal on the exterior becoming separated, more or less completely, into concentric laminae, from a solid cylindrical nucleus within. 1 Sir William Anderson states that brass powder-cases for quick-firing guns are made from brass with 70 per cent of copper and 30 per cent of zinc. These cases are pressed and drawn from discs J-inch thick into cases 16 inches 1 Percy's Metallurgy, p. 621. 144 MIXED METALS CHAP. long, and tapering from 7 inches at the breach to 6^ inches at the muzzle end, being annealed after each draw. They are then headed, followed by turning, tapering, and finishing. These cases, made out of first-class material, sometimes crack spontaneously through the base while lying in the Arsenal Stores. The effort of the material to recover itself and to return to its normal internal state is sufficient to cause its rupture. Hence the necessity of doing the work more gradually, with more drawings and more annealings, so that the internal stresses arising from so extraordinary a flow of metal may be obliterated at a greater number of stages. CORROSION OF BRASS BY ACID AND SALINE LIQUIDS 28A. Dr. Tilden l has made an extensive investigation of the corrosion of copper and brass by saline waters and acids. With respect to sheet brass, he found that when the metal was bent or distorted and placed in cold hydrochloric acid, a very rapid action took place at the bend, resulting in the disintegration of the metal at that part. After a few weeks' immersion the zinc was largely removed and a brittle spongy mass left behind, while the portion not bent was much less affected, and with brass containing 70 per cent copper scarcely any effect was produced. The action of weak hydrochloric acid and of solutions of chlorides is quite differ- ent from that of strong hydrochloric acid. Particulars are shown in Table A. 1 Birm. Phil. Soc. [TABLE A II CORROSION OF BRASS 145 TABLE A. ACTION OF DILUTE HYDROCHLORIC ACID (7 PER CENT). TEMPERATURE 13 TO 15 Percentage Loss of Weight per Square Centimetre. Ratio of Cupper (taken = 100) to Zinc, etc. by Analysis. In 14 days. In 21 days. In Original Metal. In Dissolved Portion. Copper. Zinc. Copper. Zinc. Copper 187 grm. "239 100 100 Brass C 58-64 004 100 70 100 800(!) D 61-35 009 022 100 63 100 75 >. E 61-98 020 020 100 61 100 :64 F 70-22 036 042 100 42 100 35 Bronze 93-34 172 229 100 : 7 (zinc 100 35 (Coinage metal) and tin* Table B shows the action of dilute nitric acid on the above alloys. Here the attack is greater as the proportion of zinc is greater, and the action of the acid on copper and on the best brass is about the same. TABLE B. ACTION OF DILUTE NITRIC ACID. 1 VOL. (SP. GR. 1 -42) WITH 15 VOLS. WATER. TEMPERATURE 13 TO 15 Loss of Weight per Square Centimetre. 23 days. Ratio of Copper to Zinc in dissolved portions. Copper .... 069 grm. Cu 100 : Zn Brass C . 230 100 : 124 D ... 110 100: 90 E ... 130 100 : 171 F ... 057 100: 65 Bronze (coinage metal) . 034 100: 12Zn + Sn Table C shows the eifect of sulphuric acid. The poor brass and the bron/e were the most affected. 146 MIXED METALS CHAP. .TABLE C. ACTION OF H 2 S0 4 .2H 2 0. TEMPERATURE 185 TO 190 ; TIME, 1 HOURS Loss of Weight per Square Centimetre. Copper 062 Brass C 102 D . . 040 E . . 068 F . 063 Bronze 115 If copper is immersed in sea-water it is slowly dissolved. If any portion is exposed to the air it is soon covered with a green crust. Sea-water is a solution of many salts together with dissolved oxygen and carbonic acid, and of these potassium and sodium chlorides are the most corrosive. The action of a chloride on copper in air depends not only on the affinity of copper for chlorine, but also on the affinity of the metal of the chloride for oxygen, which is strong in the case of sodium and potassium. Brass is not affected by salt solution if the air is excluded ; hence the corrosion of ship -sheathing is the greatest near the water-line. The pitting of brass and copper tubes employed in connection with salt water is closely connected with the above. Dr. Tilden draws the following conclusions from his experiments, in which the various alloys were immersed in sea-water for seven hours a day during a period of eleven months : (1) Copper and brasses with more than 60 per cent of copper are attacked by sea- water and chlorides generally more rapidly than brasses containing about 60 per cent of copper. (2) One cause of unequal corrosion leading to pitting is the voltaic action set up by the presence of impurities or metallic oxides ; oxide of copper being one of the most active. (3) The most destructive agent of all is the green ii CAST-BRASS 147 crust produced when the metal is alternatively wetted with salt water and dried in air. Patches of this substance, oxy- chloride, act electro -chemically when under water and chemically by absorption of oxygen when dried up. The presence of any free foreign matter, as iron, will corrode brass after oxidation by electrical action. (4) Of course corrosion is intensified by imperfections in the metal, as mentioned above in the case of bent and distorted specimens, when acted upon by hydrochloric acid. Holes for rivets should be punched or bored out quite clean, without altering the form of the plate or tube. Contact with other metals should be avoided. Copper should not be placed in contact with gun-metal, bronze, or brass. CAST-BRASS 29. The composition of brass used for castings varies considerably, and depends to a large extent upon the uses to which the cast articles are to be put, as also with the particular shade of colour it is desired to obtain. Thus various alloys for jewellery, having a reddish-yellow colour, are composed of from 82 to 90 per cent of copper, while those containing 60 to 70 per cent copper are of a full yellow colour. The composition most generally used consists of 66*6 per cent copper and 33*4 per cent zinc, which is termed English standard brass. It works excellently under the rolls and hammer, and may be used also for wire drawing. Cast-brass generally contains more zinc than that which is to be worked into sheet or drawn into wire and tubing ; it is therefore more fusible. Moreover, the materials are not selected with such care as is the case with those for rolling and wire drawing, so that cast-brass is often more impure than rolled brass, a large quantity of scrap being generally melted with new metal. The most common impurities are lead, tin, and iron. These metals are not always a disadvantage, for lead is generally added to the extent of from 1 to 2 per cent in brass required for turning and filing, 148 MIXED METALS CHAP. as it gives sharpness to the metal and prevents fouling of the tools. The presence of a little tin is an advantage when hardness is requisite, as in the case of bushes, for example. A little iron hardens brass, and tends to increase its tenacity and elasticity. But it is not advisable to introduce metals other than copper and zinc, unless for specific purposes, as mentioned above. It should be borne in mind that however beneficial the addition of lead, tin, or iron may be in special cases, the simultaneous presence of two or all of these metals is generally injurious, whatever the action of one of them may be when present alone in the brass. In fact, it may be stated as a general rule, not without exceptions, that an alloy of two metals is more stable than that of three or more ; and the greater the number of metals alloyed together, the greater will be the difficulty in obtaining uniform and sound castings. Lead especially has a tendency to separate out from the mixture in cooling, producing unsightly black spots, 'especially when present to the extent of 2 or more per cent. When the metal is cast in a large mass, or the castings are of considerable size, there may be, as the metal solidifies, a greater proportion of copper in the under portion than in the upper part of the casting, and the alloy is thus liable to be deeper in colour below than above. The con- stituents of an alloy of two or more metals tend to separate according to their respective densities, and the more numer- ous the constituents, and the greater the difference of their specific gravities, the more marked will this tendency be; but when the metals are in chemical combination with each other, no such separation will take place. [TABLE II QUALITIES OF CAST-BRASS 149 TABLE OF DIFFERENT QUALITIES OF ORDINARY CAST-BRASS AS DETERMINED BY ANALYSIS OF COMMERCIAL SAMPLES Copper. Zinc. Tin. Iron. Lead. French . 71-9 25-1 I'O 2-0 )> 66-3 33-2 5 French navy . English clock 65-8 67-0 31-15 27-27 25 ' 1-0 : 04 2-8 4-69 metal English brass . 67-0 32-0 ... ... 1-0 64-6 33-7 2 ... 1-5 > 64-5 32-5 3 5 2-2 (very 61-6 34-8 5 6 2-5 common) 60-2 37-04 47 61 1-68 Brass employed for the production of fine castings re- quires to possess other properties than those of being able to be filed and turned in the lathe. It must be thinly liquid when melted, and not in any degree pasty, so that it will readily flow into the minutest and most intricate crevices of the mould, and the texture must be fine-grained and uniform throughout. Moreover, as such castings speci- ally require to be sound and free from pinholes, it must be capable of remaining thinly liquid until near the point at which it solidifies, so that the metal may be poured at the lowest possible temperature, thus enabling air and other gases, absorbed during the melting, to escape as much as possible before pouring. When the castings have to be subsequently gilded, it has been found that when the metal is similar in colour to gold, it requires less gold to produce the desired effect than when the brass is of a pale yellow colour. For this reason brass of reddish-yellow colour is preferred. The French have brought the casting of fine articles, as well as larger articles, to a high degree of per- fection ; clock-cases, statuettes, and other artistic ornamental 150 MIXED METALS CHAP. work, are manufactured by them on an extensive scale. The alloys used for these purposes are often called bronze, but they are generally brass, with the addition of a little lead and tin, which impart to the metal a bronze-like colour. The following are the proportions used for a few of the French mixtures : . I II III IV Copper 637 64'45 70-90 72-43 Zinc 33-55 32-44 24-05 22-75 Lead 2-5 25 2-00 1-87 Tin . 0-25 2-86 3-05 2-95 The above alloys can be readily cast, worked with files and turning tools, and readily gilt. DIFFERENT VARIETIES OF BRASS 30. French Oreid< Copper . Zinc Tin Iron 90 10 85-5 14-5 82-75 16-40 55 30 These alloys are distinguished by a beautiful gold colour, which very closely resembles that of gold alloys. They are very ductile and tenacious, can readily be stamped and rolled, and admit of a fine polish. According to an old recipe, oreide is prepared in the following manner : " Melt 100 parts of copper and add, with constant stirring, 6 parts of magnesia, 3 '6 parts of sal-ammoniac, 1'8 parts of lime, and 9 parts of crude tartar. Stir again thoroughly, and then add 17 parts of granulated zinc, and after mixing it with the copper by vigorous stirring, keep the alloy liquid for one hour. Then remove the cover of dross, and pour out the alloy." ii VARIETIES OF BRASS 151 31. Talmi or Talmi Gold. Also termed Abyssinian gold Copper . 93-5 90 88 87 85 86 '4 Zinc . 6-5 9 11-5 13 15 12'2 Gold . ... 1 -5 Tin . ... 1-4 Taliiii gold is the name given by the French, who use the foregoing alloys for cheap jewellery, which is distin- guished by good workmanship, great durability, and a beautiful golden colour. The colour is retained for a con- siderable time, as it possesses a thin coating of gold, firmly welded to the alloy by rolling, and when the gold is of suffi- cient thickness the characteristic appearance may be retained for years without alteration. But many cheap imitations are manufactured and sold under this name, which consist of common brass, gilded with a thin coating of gold. 32. Tournay's Alloy is also used in the manufacture of cheap jewellery, as well as for buttons and so-called bronze ornaments. It is composed of 8 2 '5 parts of copper and 17 '5 parts of zinc. 33. Mannheim Gold, Similor, Prince's Metal. The composition of this alloy varies considerably, as will be seen from the following analysis of three samples : Copper. . . . 83-1 88 '9 75 Zinc .... 10-0 10-3 25 Tin .... 6-9 -8 The first has a yellowish-red tint, and the second one a deeper red. Similor has been much used for buttons and other stamped work requiring a reddish cast of colour. 34. Tombac. As stated on a previous page, tombac is a name applied to alloys which by some manufacturers are termed prince's metal, similor, and Mannheim gold. The name is used for alloys so widely different in composition and properties as to lose any significance it may have 152 MIXED METALS CHAP. formerly possessed, and strikingly illustrates the great need of adopting Dr. Percy's proposition, that the different alloys of copper and zinc should be designated by their percentage composition. The following table will show the proportions of different mixtures. Copper. Zinc. Lead. Tombac for buttons . 99-15 85 Red tombac of Vienna 97'8 2'2 ,, ,, Paris 92 8 Tombac (Bolley) 91 9 Tombac of Oker . 84'6 15-4 .. Tombac for buttons 82-3 177 Tombac (Bolley) 70-1 29'9 French tombac . 80 17 3 )> 82 17-5 0-5 35. Gilding Metal. Alloys of copper and zinc contain- ing upwards of 80 per cent of copper have a reddish-yellow or red tint, and are used as a base for gilding. The more nearly the colour approaches to that of standard gold, the more highly is it appreciated for articles which are sub- sequently to be gilded. When such metal is required for rolling into sheet, it will be seen by reference to the table collated for the U.S. Board that the most malleable alloy is represented by the composition, 83 copper and 17 zinc. These alloys are also known by the name of red-brass, and as the reddish cast of colour is more agreeable to the eye than that of yellow-brass, they are frequently used for cast articles not requiring special hardness and strength. It has the reputation, however, of tarnishing quicker than ordinary brass. The properties of red-brass may be modified to suit special cases by the addition of tin, lead, and iron, the action of which has been already explained. The following varieties of brass are distinguished from the preceding alloys by a characteristic yellow colour as contrasted with red -brass. HAMILTON'S METAL 153 36. Hamilton's Metal, Chrysorin, and Mosaic Gold. Copper . . 72 66 '6 65 '3 50 Zinc . . 28 33-4 347 50 Hamilton and Parker in 1826 claimed for the 50 per cent alloy that, after casting in the mould and cooling, it exhibits the colour of gold, and does not tarnish by exposure to air, even in the neighbourhood of the sea. They prepared it by fusing together equal parts of copper and zinc in a crucible at the lowest possible temperature, stirring con- stantly, and then adding a further quantity of zinc in small portions, till the right colour was obtained. This alloy is very flexible when strongly heated, but not adapted for cold rolling or wire drawing. It melts at a comparatively low temperature, and is used as a hard solder. The higher alloys given under the above names are malleable and ductile, and, as before mentioned, well adapted for cast ware. The method of preparation described by Hamilton and Parker would be quite unsuitable for adoption on a work- ing scale, for independently of the great waste of zinc, the composition of the alloy would be of an uncertain character. The usual plan is to melt the copper first and then add the zinc. The following plan has been recommended, but is open to serious objection for the same reasons as mentioned above. Bring into the crucible one-half of the zinc to be used ; place upon this the copper, and fuse the mixture, under a cover of borax, at as low a temperature as possible. When the contents of the crucible are liquid, heat the other half of the zinc (cut into small pieces) until almost melted, and throw it into the crucible in portions ; stir constantly to effect as intimate a mixture of the metals as possible. 37. Prince's Metal. A name given to various yellow alloys varying from 60 to 75 per cent of copper and 40 to 25 per cent zinc. 38. Bobierre's Metal This is ordinary brass, con- 154 MIXED METALS CHAP. sisting of 66 parts copper and 34 parts zinc. Bobierre intro- duced this alloy as especially suitable for ships' sheathing. 39. Macht's Yellow Metal is composed of 57 parts copper and 43 parts zinc. It has a reddish-yellow colour, malleable when rolled hot, but not in the cold. It is said to be suitable for fine castings, as it possesses great strength. COMPLEX BRASSES 39A. By this term is meant copper-zinc alloys to which other constituents have been purposely added. Brass with Lead. Ordinary brass for filing and turning usually contains lead, which improves its working qualities. Lead exists in brass in the free state and tends to segregate in patches according to the amount present and the rate of cooling. Quick cooling lessens segregation. 6 to 8 per cent of lead may be present in brass without segregating in patches if cooled quickly, but the metal has a greyish yellow fracture. If worked hot the metal " sweats," and when the pressure is put on in rolling liquid lead flows out, but it facilitates the working. In sand casting, which allows of slow cooling, the lead tends to collect at the bottom of the casting. The tests of mechanical properties show that lead, exceeding 0'5 per cent, diminishes the tensile strength and elastic limit of rolled bars ; it also lowers the elongation, especially in Muntz's metal. It promotes brittleness. The following tests are given by Quillet : l 1. BARS ROLLED AND ANNEALED Copper. Zinc. Lead. T. 8. E. L. E. per cent. 59-5 39-7 0-85 19 7-6 45-5 58-45 39-35 1-90 17 7-2 43-4 57-85 38-85 3-05 14 7-0 41-1 56-15 40-05 4-02 13 6-0 31-9 54-85 39-25 5-15 12-5 6-0 30-2 1 Alliages Metalliques, p. 663. COMPLEX BRASSES 155 2. CAST BARS Copper. Zinc. Lead. T. S. E. L. E. per cent. 60 40 20 5 4-7 59 39'8 1-2 17 5'1 14-9 60 37-9 2-1 19 5 12-5 70-4 29-6 8-6 3'6 68 69-1 30-2 07 10 3-6 42 67-9 30-8 1-3 11-8 3 51 68-3 29-1 2-6 13 4 54 Brass with Tin. The effect of tin on copper is much the same as zinc up to a certain limit, but the tin is much more active. Brass with 60 per cent copper and containing more than 4 per cent tin is very brittle. The effect of more than 2 per cent tin on Muntz metal is to raise the tenacity but to considerably lower the elongation and increase the brittleness and hardness. For hot working, 2 per cent should not be exceeded. Naval brass contains from 60-62 copper, 1-1 tjr per cent tin, and the rest zinc. The Admiralty specify 62 copper, 37 zinc, and 1 per cent tin. The presence of tin in brass confers the property of greatly resisting corrosion by sea -water. Brass with Manganese. The action of manganese on brass is to strengthen and harden it. The addition of manganese produces the same effect as though the copper were reduced and the zinc increased. For example, an alloy of 54 copper, 40 zinc, and 6 manganese has practically the same structure as brass with 57 copper and 43 zinc. Again, an alloy of 55 copper, 10 manganese, and 35 zinc has the same structure as 60-40 brass. With regard to mechanical properties, manganese increases the tensile strength, limit of elasticity and hardness in proportion to the amount present, while at the same time the elongation steadily diminishes. These remarks equally apply whether the manganese replaces 156 MIXED METALS CHAP. the zinc or the copper, as shown in the following table by Guillet : Copper. Manganese. Zinc. T. S. E. per cent. Hardness. 57-5 2-6 39-9 22 42 571!,! 55'6 4-5 39-9 30 30 64 1 ^^ 54'8 5-6 39'6 31 29 64 f ,2 g 50-8 9-6 39'6 34 21 83 J ( 69-6 2-0 28-3 13 47 38] . 70-5 4-2 25-3 15 44 43 r 3 70-2 9-2 20-5 19 32 54j The types of brass with manganese most employed in industry are : 60 copper, 40 zinc, and a trace of manganese ; and 58 copper, 38 zinc, and 2 manganese. These alloys can be worked hot, and present facilities for forging, stamping, bending, etc. They offer a good resistance to sea-water and are non- magnetic. They are used for screws of torpedo tubes, rudders, hydraulic cylinders and valves ; forgings for shafts of screws, piston rods and slides, coils and bolts ; rolled bars for rivets and bolts; sheets, profiles, etc., for marine work, and plates subjected to moderately raised temperatures. It may be stated that manganese is generally added in the form of ferro- manganese, and therefore the brass contains a little iron and silicon. Brass with Aluminium. Aluminium is used in brass as a deoxidiser, when it is added in very small quantity. The addition of 2 or 3 per cent of aluminium to brass of the 2-1 type has the effect of enabling it to be rolled hot, and confers a structure similar to brass of the 60-40 type. It is not advisable to add more than 4 per cent as the metal becomes too brittle. With brass of the 70-30 type, 2 per cent aluminium does not alter the structure, as the aluminium passes into the a solution. With more than 2 per cent the /3 constituent appears, and the alloy is harder in proportion. The general effect of adding aluminium to BRASS WITH ALUMINIUM 157 brass is the same as though more zinc were added. Probably 1 part of aluminium has the same effect as adding 3j per cent of zinc. Brass containing aluminium in notable quantity has an extremely close grain, and has a tenacity higher than ordinary brass. Brasses with aluminium may be arranged under three heads : I. II. III. Copper . . . 68-70 64-66 58-61 Zinc .... 31-27 33-30 40-5-37 '5 Aluminium . . 1-3 1-4 G-3-1'5 Guillet, in his Alliages Metalliques, gives the following results of his mechanical tests : CAST BARS Copper. Zinc. Aluminium. T. S. E. per cent. Hardness. 60 40 20 47 50 59-9 39-3 0-8 19-3 45 52 59-6 37-5 2-9 29 14 101 60-4 34-9 4-7 28 2 148 70 30 8-6 50 32 70 29-1 0-9 14-2 67 34 70-5 26-4 3-1 21-2 50 65 70 24-8 5-2 31-8 11 107 ROLLED AND ANNEALED BARS Copper. Zinc. Aluminium. T. S. E. per cent. Hardness. 61-4 37-9 0-7 22 45 52 61 37-6 1-4 23 45-3 65 60 38 2'0 25 17 123 60 36-1 3-9 30 13 148 1 i Brass with aluminium may be used for similar purposes to those of brass with manganese. 158 MIXED METALS CHAP. BRASS CONTAINING IRON 40. Many samples of brass and bronze made by the ancients have been found on analysis to contain iron, and probably they knew that the addition of iron to these alloys would increase their hardness and strength, and introduced it with that view. In more modern times the combination of iron with brass has engaged the attention of metallurgists, and several alloys, containing iron as an essential constituent, have been introduced from time to time. In 1779 James Keir proposed an alloy of 10 parts iron, with 100 parts copper and 75 parts zinc. Similar alloys to this, but containing less iron and different proportions of copper and zinc, were introduced under the names of " sterro-metal " or " Gedge's alloy," and "Aich's metal." Sir John Anderson, late superintendent of the royal gun factories, carried out a series of experiments with brass containing iron, and obtained some good results. The increased strength and hardness of such alloys were acquired at the expense of ductility and toughness. The great difficulty the above experimenter had to contend with was the uncertainty in the properties of the alloys containing iron. See also p. 116. 41. "Sterro-metal" consists of 60 parts copper, 38 to 38-5 zinc, and 2 to 1-5 iron. It was recommended as an alloy for sheathing for ships and other objects which are subjected to the continued action of sea-water. The presence of iron in this alloy imparts to it a strength equal to that of mild steel, and superior to that of wrought-iron. Brannt mentions a case in which a wrought-iron pipe broke with a pressure of 267 atmospheres, while a pipe of sterro- metal stood the enormous pressure of 763 atmospheres without cracking. This alloy also possesses great elasticity, and is therefore specially adapted for hydraulic cylinders. Such cylinders, when subjected to very high pressures, begin to sweat, the water from the inside permeating the pores of the metaL With sterro-metal the pressure can be raised IT AICH'S METAL 159 considerably higher than with iron or steel, without moisture appearing on the outside of the cylinder. Sterro-metal can be made very hard and dense by suitable mechanical treatment, which has as great an influence in modifying its properties as has the chemical composition. In rolling or hammering this alloy when hot, special care is requisite in regulating the temperature to which it is raised, as by too much heat it becomes brittle, and cracks under the hammer, or between the rolls. Baron Rosthorn tested a sterro-metal containing copper 55'04, zinc 42*36, tin 0'83, and iron 1*77 per cent, which gave the following results: Condition. ; Tenacity in IDS. per Square Inch. Cast 60,480 Forged . . . 76,160 Cold drawu . . 85,120 Gun-metal Case . . . 40,320 The tenacity of ordinary gun-metal is given for comparison. The specific gravity of the alloy was from 8 -3 7 to 8 -40 when forged or drawn into wire. Another alloy from the same source contained copper 55, zinc 41'34, and iron 3'66 per cent. 42. Aich's Metal. This alloy is analogous to sterro- metal, and shows similar variations in composition from various analyses that have been made. Its chief properties are hardness and tenacity, the same remarks applying to this as to sterro-metal, with which it is practically identical. Alloys under this name contain from 0*4 to 3 - per cent of iron. It has a golden-yellow colour, and is recommended for articles exposed to sea-water. The following analyses will give an idea of the composition : Copper. . 60-66 60 60 -2 58 '26 Zinc . . 36-58 38'2 38*2 41'00 Tin . . 1-02 Iron . . 174 1-8 1'6 074 160 MIXED METALS CHAP. 43. Delta-metal. This alloy was brought out by Mr. Alexander Dick in 1883, and since that time has established a useful place for itself among modern brasses. The name " delta " was given to it by Mr. Dick, simply for the purpose of connecting it with his own name, delta being the Greek for the letter D, the initial of the inventor's surname. As already mentioned, the great difficulty former experi- mentalists had to contend with was the uncertainty in the properties of the alloys containing iron, and Mr. Dick set himself the task of ascertaining the cause of failure. He prepared various quantities of the alloy, apparently in exactly the same way, by dissolving wrought-iron in molten copper. The results showed many discrepancies, because the amount of iron dissolved in each sample was far from uniform. He then tried to find a method by which he might be enabled to introduce a known and definite quantity of iron, and succeeded by dissolving iron in molten zinc to saturation, and adding the same, with or without pure zinc, to the molten copper. But when the metals were re-melted, oxidation took place, and the castings again varied in character, owing to the oxides thus formed dissolving in the alloy, and diminishing its strength and toughness. This second difficulty was over- come by adding a small percentage of phosphorus in combina- tion with copper. In some cases Mr. Dick also introduces tin, manganese, or lead into the alloy, to impart special properties to it. The various alloys thus produced are now manufactured and sold under the name of " delta "-metal The inventor claims that by his process the iron is chemically combined in the brass and bronze, as proved by the alloys not rusting when exposed to moist air, and by their indifference to the magnetic needle. In a lecture by Mr. Macintyre before the Balloon Society on 15th November 1889 he states that "the properties which are combined in delta -metal great strength and toughness, durability, resistance to corrosion, and a compara- tively low price render it of the greatest value for purposes of construction generally ; and more especially for ship- II DELTA-METAL 161 building, marine engineering, and sanitary work. It can be equally well cast as forged, stamped and rolled hot, and drawn cold." The power of delta-metal to resist corrosion by the acid liquors of mines has been proved by the Bonifacius Coal Mining Company of Westphalia. The Company made a series of experiments with a view to finding the relative corrosion of metals of suitable strength. Brass and gun-metal were not strong enough, and trials were made with steel, iron, and delta-metal. Rolled bars of each of these were immersed during a period of six and a half months in the water issuing from the pits at Kray, and then carefully re-weighed and photographed. The bars were 7^ inches long, and had a sectional area of *62 square inch. The following were the weights of the three kinds of bars before and after the trial : , Wrought-iron. Steel. Delta-metal. Weight of bars when put in ,, ,, after 6 months Ibs. 1-1805 6393 Ibs. 1-2125 6614 Ibs. 1-2787 1-2633 In the Schweizerisches Gewerbeblatt of 8th June 1889 the following tests of delta-metal were recorded : " In the first locomotive engines of the Pilatus mountain railway, a material was required which could be cast, and possess at the same time great tensile strength and elasticity. Delta-metal was found to answer these requirements ; the worm-wheels of the brake gear for the engines were of this material and worked very satisfactorily. The castings were tested by Professor Tetmajer of Zurich, the results showing a tensile strength of 2l| to 23 J tons per square inch, with an elongation of 30 to 40 per cent on a length of 7 J inches." Captain Locher also had the following tests made : One of the delta-metal pinions having been in use for a long time, the teeth had worn about -% of an inch, so that their thickness at the root was | inch, and at the top ^ inch, by a breadth of 4 inches. 162 MIXED METALS CHAP. It was tested to show what power would be required to break such a tooth. The test, made by Professor Tetraajer, gave the following result : P ... 5 9 10 12 14 15 16 17 18 AL ... 0-015 0-019 0-059 0'098 0'133 0'169 0'208 0'244 P ... 19 20 21 214 AL 0-295 0-354 0'472 | broke. P indicating the stress in tons, and AL the shortening in decimals of an inch of the distance L, originally measuring 2^ inches. Both as regards P and AL the results were unexpectedly favourable. [TABLE II DELTA-METAL 163 1 : i Broke in ex- tension mark. * \t li e m M P P S f O OQ < IslS S -& 0) CO g CO (M T 1 r- 1 1-2 -S i| |l| CO p o CO CO CO CO CO 1 - s ' Not taken. CN CO (N p CO 11 || 00 1 (M CO CO lO 1 1 * * | il s Oi CO 00 1 o 00 oo CO oo 0* * * * " Size and specimen. IO 00 X 495 dr. turned i oo (M X I 1 Description. Square bar (annealed) o -M S 1 1 Hexagonal bar t & IM i i 6 CO 6 fc i to 164 MIXED METALS CHAP. 1 o o ri OS S ^1 ill 60 fl p^ 1 wJ'g d 1 P 5 J^ " * " 1 ll ID 1 1 o b a ^^ ^ ^1 2 CO a i J Sj ^ -u ^ S)'" o 03 60 a 8 1 Oi ? ^'g> H 1-1 I 1 * ^ rt *^ Per cent of con- traction. CO CO 1 a eo ? I i 1 |* CO CO i 1 60 .1 ^ g 3* d 'i S |1 CO ' 2 CO 1! I ,2 ^ 15 1 ^^ H l ! B "l g I 1 6 ^g ii DICK PKOCESS 165 In addition to iron and phosphorus, delta-metal often contains manganese, aluminium, tin or lead. The average composition is 55-57 copper, 42 zinc, and 1-2 per cent iron, and small quantities of the above as impurities. Brass containing iron is manufactured in Germany under the name of " Durana Metal" Dick Process. This consists of forcing the metal to be worked across a matrix situated at the end of a cylinder after raising it to a sufficient temperature to make it plastic. This principle of compression has long been used in making lead piping, etc. The compression chamber consists of an inner steel tube surrounded successively by other steel tubes, separated by insulating material The inner tube, which sustains the action of the hot plastic metal, is supported by the concentric tubes. The matrices employed are made of special hard steel, and contain one or several holes according to the section of bars to be produced, so that several bars may be produced at one time. The matrix is placed in a conical cavity in a support which is kept in position by a pair of jaws actuated by a hydraulic press. The matrix and support are previously heated to prevent expansion during the working. The receptacles vary in size so as to contain from 110 to 440 Ibs. of metal For conducting an operation the end of the cylinder is closed by a fastening plate, put in the vertical position, the metal run in from a crucible, and allowed to solidify until in a plastic state. A bevelled steel plate is then put on to pre- vent back flow, a block of steel interposed, having a cavity in which the end of the piston rests. Under the combined action of heat and pressure the disc expands and closes the valve. It is then placed in the horizontal position. The pump is then worked, and in four minutes the bars of metal are formed. 50 or more charges per day may be made, or a daily output of 5 tons, assuming each charge to be about 220 Ibs. This process is limited to forgeable brasses and bronzes, 166 MIXED METALS CHAP. such as Muntz metal, aluminium bronze, delta -metal, etc. The metal thus treated has a high tenacity and ductility, is smooth on the surface and requires no subsequent finishing work. WHITE BRASS 44. Alloys of copper and zinc containing less than 45 per cent of copper cease to have a yellow colour. The alloys containing from 40 down to 30 per cent of copper are silver- white, and with less than 30 per cent of copper the colour passes from grey to bluish-grey, having a greater resemblance to metallic zinc as the proportion of that metal is increased. The silver-white alloys break with a con- choidal fracture, and the more zinciferous alloys with a fracture more or less crystalline. In consequence of the brittle nature of white alloys they cannot be used for rolling and wire drawing, but some of them are used for pressed work, when too strong a pressure is not required. Some of them are known by special names, thus : 45. Birmingham Platinum and Platinum Lead are used for certain castings, but the composition is variable, according to the taste of the manufacturer. The following will illustrate this point : Copper Zinc 46-5 53-5 43 57 20 80 The above alloys are used for buttons by casting them in moulds giving sharp impressions, the letter or crest being subsequently brought out by careful pressing. Other alloys for buttons consist of Copper Zinc . Tin 54 50 43 45 3 5 60 60 ;3i 30 6i 10 46. Sorel's Alloys. These alloys are distinguished by great hardness and considerable tenacity. They cast well and can readily be detached from the mould. They are BRASS SOLDERS 167 largely used for statuettes and other artistic work, which, after suitable bronzing, are brought into commerce as cast- bronze. The following mixtures are recommended : Copper .... 1 10 Zinc .... 98 80 Iron .... 1 10 Iron is used in the form of turnings, and melted with the copper and zinc under a layer of charcoal But as zinc so readily volatilises it is advisable to employ zinc already containing iron, by which a more uniform alloy is obtained, with the minimum loss of zinc. 47. Fontainemoreau's Bronzes. These so-called bronzes are said to answer well for chill-casting, the metal being poured into iron moulds. By this means the alloys are rendered more homogeneous, because the rapid cooling prevents the separation of the constituents in accordance with their respective densities. The highly crystalline nature of the zinc is changed by the addition of copper, iron, or lead. The following are the proportions used : Zinc . 90 91 92 92 97 97 99 99 Copper . 8 8 8 7 2i 3 1 ... Cast-iron 1 1 4 Lead . 1 1 " BRASS SOLDERS 48. These are copper-zinc alloys employed for joining the various parts of articles together by fusion. The solder must, therefore, have a lower melting point than the body to be soldered, but the fusing point of the solder should approach, as nearly as it conveniently can, to that of the article, as a more perfect and more tenacious junction may thus be effected. Brass solder belongs to the class known as hard solders, or brazing solders. It may be stated as a general rule that the melting point of copper -zinc alloys is higher in proportion to the amount of copper present, and therefore any quality of brass may be made into a suitable solder by adding zinc or copper, as the case 168 MIXED METALS CHAP. may be. The alloy commonly used in soldering brass contains equal weights of copper and zinc. An easily fusible solder may be made with 34 copper and 66 zinc. In this case, however, it must be borne in mind that the joint would be much weaker than when the more difficultly fusible solder is employed, so that excess of zinc is to be avoided wherever possible. A readily fusible solder may be obtained by using 44 parts copper, 50 parts zinc, 4 parts tin, and 2 parts lead. Alloys containing much lead are not to be recommended, since the lead tends to separate out and produce unsightly black spots, besides decreasing the strength of the joint. A good hard solder for the richer alloys of copper and zinc may be produced from 53 parts copper and 47 parts zinc. Brass solder is sometimes used for soldering iron and copper, and as these metals have a much higher melting point than brass, a much better quality of solder can be employed, and is indeed advisable in many cases, being much stronger. In these cases tin is often added as one of the ingredients, but it should be only sparingly used, as it increases the brittleness of the solder and thus becomes a source of weakness. The addition of tin to brass causes the yellow colour to pass into grey, or white, according to the content of tin employed, and mix- tures may be obtained of a yellow, yellowish- white, and greyish- white colour. The following table indicates the different varieties : Copper. Zinc. Tin. Lead. Colour. Very strong . 58 42 Reddish yellow Strong . 53 47 J} Medium 50 50 > > *)4A 43i ji * Easily fusible 34 66 ... White 44 50 4 2 Grey White solder 57 28 15 ... White ii BRASS SOLDERS 169 In making solder it is very important that the constituent metals should be of good quality, as impurities seriously interfere with the colour, malleability, and strength of the solder ; great care should also be taken to ensure a thorough mixture, so that the alloys may be uniform in composition. Solder is often made by melting brass with the requisite addition of zinc, thereby ensuring a more perfect union and less loss of zinc than is the case when zinc is added to molten copper. Solder is most commonly used in the granulated form, which is effected by pouring the molten alloy into water, or by pounding it in an iron mortar when strongly heated. The most suitable mode of preparing brass solder is to melt the brass rapidly in a crucible, the metal being covered with a layer of the best powdered charcoal, and when thoroughly fused, to add the zinc, which has previously been heated to near its melting point. Stir vigorously for a few minutes to ensure a thorough incorporation of the contents, skim the dross from the surface, and then pour, taking care that no dross or charcoal is carried over with the metal into the mould, or into the water, as the case may be. One method of granulating is to take the ingot of metal from the mould immediately it has solidified and pound vigorously in a large iron mortar. Or to first raise the ingot of solder to the requisite temperature over a charcoal fire, and then crush to powder in an iron mortar. Some manufacturers pour the molten metal into a ladle, and empty the contents of the latter from a considerable height into cold water, the metal in its descent passing through a wet broom, or similar contrivance, so as to divide it into fragments. The granulated metal is afterwards sifted through sieves of different sized meshes, so as to obtain the grains of uniform size. Another plan is to pour the molten metal on to the surface of a large iron ball, placed in a shallow pan contain- ing cold water, so that the top of the ball projects above the 170 MIXED METALS CHAP. surface of the water. The metal is thus broken up into small fragments of fairly uniform size. According to Krupp, the finest and most uniform product is obtained in the following manner : " At some distance above the surface of the water, serving for 1 the collection of the grains, a horizontal pipe is arranged which is connected either with a powerful forcing pump or a water reservoir. Before pouring the metal the cock on the pipe is opened, so that the jet of water issuing from the pipe is thrown in a horizontal direction over the vessel containing the water. Upon this jet of water the molten metal is poured. The greater the force with which the water is forced from the pipe, the greater also the force with which the stream of melted metal is divided, and by this means it is possible, within certain limits, to obtain grains of a determined size." As will be seen from the above description, the scattering of the stream of molten metal is based on the same principle as that employed in diffusing fragrant liquids in the air. When solder is granulated by pouring into water, it is necessary to remove the grains from the vessel as soon as the operation is completed, and dry them quickly, so as to avoid unnecessary oxidation. ii CALAMINE BRASS 171 MANUFACTURE OF BRASS 49. Allusion has previously been made to the two distinct modes of making brass, known respectively as the calamine and direct methods ; the former being almost exclusively used until within the last fifty or sixty years, but is now practically obsolete, at any rate in this country. In the ancient or calamine method, metallic copper is mixed with oxide of zinc and charcoal, and the mixture strongly heated for twelve hours or more, when the zinc, reduced by the carbon and carbonic oxide present, alloys with the copper, forming brass. Such brass has been claimed to be of superior quality to that made by the direct process ; but as great improvement has taken place in the qualities of brass of late years, it is very doubtful, if the manufacture of calamine brass were to be resumed, whether the value of the metal for most purposes would be superior to what the market can at present command. The reputed high quality of brass made by the old method could only be maintained by exercising great care in the quality of the ore and copper employed, and selecting ores of uniform composition to produce a product of uniform and determined properties. The cost of production of calamine brass is less than that of brass made by the direct process, but the former method is much more tedious and troublesome, requiring a much longer time than is consistent with modern requirements, and for small founders is quite unsuitable. CALAMINE BRASS 50. For calamine brass the ores were submitted to a preliminary treatment in order to remove as far as possible other compounds, such as those of lead, antimony, and arsenic, which would injure the quality of the brass. Native calamine was calcined to remove carbonic acid, sulphur, or other volatile matter, and form zinc oxide. The calcined ore was then ground in a mill, any galena removed by 172 MIXED METALS CHAP. Fig. 17. Fig. 18. Scale of Feet ii CALAMINE BRASS 173 washing, and the dried oxide mixed with about one-third its weight of coal or charcoal, and introduced into crucibles with alternate layers of granulated copper. The crucibles employed were made of fire-clay, 12 J inches deep, 8j inches wide at the top, and 6| inches wide at the middle, inside measure. The king-pot, as the middle pot was termed, was 13^ inches deep, and would hold 120 Ibs. of metal, while the smaller ones would only hold 84 Ibs. The charge consisted of 100 Ibs. calcined ore and 40 Ibs. coal, to every 66 Ibs. of bean-shot copper. 1 The furnace used, Figs. 17 and 18, consisted of a circular chamber w, lined with fire-brick ; it was contracted above to a circular opening, in which was fixed a cast-iron collar ee, it was closed at the bottom by a cast-iron bed-plate oa, in which were twelve holes symmetrically arranged round one larger hole in the centre &, through which the ashes and clinkers could be withdrawn from time to time. Below this plate was the ash-pit n, communicating in front by means of an air-way c, with a vault i, through which air was admitted, and access gained by the workmen to the ash-pit. In the small holes in the bed-plate were placed cast-iron twyers //, tapering upwards. The space between the twyers was filled up with fire-bricks to form a solid bed. The air for the combustion of the fuel entered through these twyers. Several furnaces were constructed in a row, and the whole covered with a large cone like that of a glass- house. The crucibles were placed in the furnace so that the large or king-crucible occupied the central position. Each pot was loosely covered with a piece of coal, and smaller pieces of fuel were packed between the pots. When the operation was completed the king-pot was first removed, and the contents well stirred with an iron rod. Each of the side pots was then removed in succession, well stirred, and when the brass in each case had subsided to the bottom the contents were poured into the king-pot. The dross was 1 Percy's Metallurgy, p. 613 174 MIXED METALS CHAP, n then skimmed from the surface, and the metal poured into suitable moulds. In this process oxide of zinc is reduced at a temperature below the melting point of copper, which, being exposed to the action of the vapour of zinc, becomes permeated with this metal and converted into brass. If the temperature be raised too high at an early stage of the process the copper will melt, sink to the bottom of the crucible, and much of the zinc escape without alloying with the copper. When the metal obtained by the above process was not of the desired quality, it was necessary to undergo a fresh fusion with calamine and charcoal, or with copper, according as the copper or zinc was in excess. DIRECT PREPARATION OF BRASS 51. This method consists of melting copper and zinc together in the desired proportions, either in a crucible, or in a reverberatory furnace, the latter being used chiefly for Muntz's or yellow metal, where large ingots are required. Many attempts have been made to do away with crucibles for the manufacture of brass, and to substitute special furnaces of the reverberatory type ; but the loss of zinc is so great, and the composition of the brass so liable to be uncertain, that even for yellow metal many manufacturers have discarded the reverberatory, and gone back to the old crucible method. A crucible furnace is generally a rectangular chamber 12 to 16 inches square and 3 to 4 feet deep, lined inside with fire-brick, and connected near the top with a chimney by means of a flue, which is generally horizontal at the part adjoining the furnace, then inclining upwards into the chimney. This is especially the case when several furnaces open into one chimney. The proper construction of the furnace and disposition of the flues is a matter of the first importance, as a slight difference in the arrangement of the flues will considerably affect the draught and prevent the Scale % of an inch to a foot Vertical Section Fio. 19. Scale % of an inch to a foot Plan Fio. 20. 176 MIXED METALS CHAP. II attainment of that high temperature necessary in melting copper, brass, bronze, and similar metals. The section of the flue has a great influence on the working of a furnace, J - T o'boooooooood'o Scale ^ of an inch to a foot Front Elevation Fio. 2j. ' for if too narrow the friction will be great and the draught too sluggish. For an active and strong draught the flue must be wide and the chimney large and high. The section Scale | of an inch to a foot) Section through A.B. Fig. 24 Scale $ of an inch to a foot Section through C.D. Fig. 24 Fio. 23. Scale Y& of an inch to a foot 02468 FIG. 24. CHAP, ii BRASS 179 of the flue should be from one-sixth to one-fourth that of the fireplace. Defective draught very often arises from the bad construction of the flues, for when the draught holes from several furnaces open into a common conduit the currents, being continued beyond their orifices, modify each other. Figs. 18 to 24 show in plan and elevation a row of six large sized furnaces with their flues opening into a common stack. Furnaces for melting metals, heated with gas instead of solid fuel, have been adopted in some works. Fig. 25 shows an arrangement formerly used in the Berlin Porcelain Manufactory, and in the mint. A is a chamber of fire-brick having an ordinary flat grate, composed of iron bars, and another grate in which the bars are placed in steps 6. On a grate of this kind small free-burning coal may be used, which would in a great measure drop through an ordinary grate. The surface of the grate, by this step-like arrange- ment of bars, is considerably increased, and ample space allowed for admission of air. The two openings d, e are closed by iron doors. A blast of air is introduced under the grates by the pipe /. The fuel ia charged through the opening g, which is closed with an iron cover, and by a sliding damper h. The gases generated, chiefly carbonic oxide, pass through the pipe &, provided with a regulating valve I, and escape at m. Immediately in front of the opening m, and extending across the furnace, is an iron pipe n, from which air issues. The carbonic oxide here burns, and the heat is communicated to the furnace in which the crucible is placed. In order to represent both the gas- producer and the melting furnaces in the same vertical section, it has been necessary to deviate from the actual position of these furnaces at the mint, where the melting furnace is at the side of the gas furnace. The air is forced by a small fan through the pipes / and n, which are provided with regulating valves. 1 Great improvements in the construction of melting 1 Percy's Metallurgy, p. 201. 180 MIXED METALS CHAP. furnaces heated by gaseous fuel have been made during the last thirty years, notably through the labours of Messrs. Siemens. By the use of gas-producers fuel may be used which is too inferior for ordinary furnaces, and a higher temperature attained by the combustion of the gas formed than by using solid fuel directly. A gas-producer is generally a somewhat rectangular chamber, in which fuel is burnt for the formation of carbonic oxide as already explained. Siemens's original producer, Fig. 26, was lined with fire-brick ; the side A was formed of iron plates lined with fire-bricks, and had a step-grate B, with wrought-iron bars C. The fuel was charged Fio. 26. FIG. 27. through the hopper D. The gas passed up the pipe E, which was cased with iron, and issued into a horizontal wrought-iron pipe, which conveyed it to the regenerator, where it was highly heated before entering the furnace in which it \vas burnt, thus producing a much higher temperature. The waste heat of the furnace is not lost, but utilised in heating the regenerators. The arrangement of a Siemens's crucible furnace is shown in Fig. 27, where the pots are heated by the combustion of hot gas and hot air. The regenerators are chambers of open refractory brickwork, built in pairs, two pairs being required for each furnace ; each pair being used alternately for absorbing the heat of the gaseous products from the furnace, and heating the gas and air required for com- bustion. By means of a reversing valve, the waste gases BRASS 181 pass to the right or left pair at wilL When the waste gases are passing down through the right pair, the cold air and producer gas are passing up through the left pair, the direction being reversed when sufficient heat has been ab- sorbed. It will be observed that the combustible gas and air enter the furnace alternately at the right and left, according to which pair of regenerators is being used for heating them. By using gas for melting metals a neutral or non- oxidising flame can be obtained, which prevents loss of metal by oxidation, a manifest advantage in cases where it is Fio. 28. necessary to preserve the composition of an alloy intact, and should be of great advantage in the manufacture of such alloys as brass. Reverberatory furnaces are only used for the manufacture of brass where large quantities require to be cast, as in the case of ingots of yellow metal used for ships' sheathing, etc. The furnace commonly employed, Figs. 28, 29, is capable of holding about a ton of metal. It is somewhat rectangular in shape, with flat sloping bed, which inclines from the bridges and back towards the front, or working door. The bed is formed of fire-brick carefully set edgeways, or a well- rammed sand bottom is used. It is advisable that the copper should be melted first in 182 MIXED METALS CHAP. an atmosphere of flame, so as to exclude the air as much as possible. Then the scrap and zinc, previously heated, are introduced, and the mixture thoroughly stirred as rapidly as possible, to prevent loss of zinc. Another plan, which finds favour with some manufacturers, is to melt the copper in Fig. 29. Scale of Feet 3 4 $ 5 7 the furnace, tap it into a large iron ladle, and add the heated scrap and zinc to the ladle, so that the metal may be poured into the moulds immediately after the necessary stirring required to mix the contents is effected. The alloys of copper and zinc are easily formed, but when the zinc is added to the copper in the furnace the damper of the chimney should be nearly closed, and the fire should not be too brisk, since by having too high a tempera- ture at this stage much zinc would be wasted. Moreover, when the metals are thoroughly mixed the surface should be covered with charcoal or sand, especially when it is necessary to raise the temperature before tapping. When the metal is ready for tapping the tap-hole is opened with an iron bar, and the metal run into a ladle. The surface of the metal is covered with charcoal, which keeps in the heat, and preserves the metal from oxidation by the air. The temperature of molten brass and bronze becomes rapidly II BRASS 183 lowered, and iii order to produce sound castings, no time should be lost in pouring the metal into the moulds. All currents of air should be guarded against, and all openings tending to produce them should be closed during the time of casting. 52. Condensation of Zinc Fume. In the melting or Feet Fio. 30. manufacture of brass, the zinc, being of a volatile character, is to some extent volatilised, as before mentioned, especially when a reverberatory furnace is used, so that it becomes 184 MIXED METALS CHAP. advisable to adopt some means for its recovery. The following plan is employed in some works. A syphon-flue, Fig 30, is placed between the furnaces and the chimney, and at the bottom of the flue is placed a tank about 3 feet wide by 4 feet long, as shown at A. This tank is filled with water, and when the heated gases from the furnaces pass over the water, a vapour arises, which causes the zinc fume to condense and settle down into the tank. Zinc fume consists largely of zinc oxide, which is periodically removed and sold to zinc smelters. A second tank is sometimes placed near the top of the syphon-flue, as shown at B, and from this tank a spray of water is allowed to continually Irop, thus assisting in the condensation of zinc fume. The tank A is oval in plan, and the end projects outside the flue, and is fitted with an air-tight iron lid for cleaning purposes. CASTING OF BRASS 53. Great care and skill are required in casting brass after the alloying has taken place, as the success of the operation depends upon the discrimination displayed at this juncture. However perfectly the metal may be made in the furnace, the whole will be vitiated by an unsound or spilly casting, if the brass is required for sheet or wire. Two different modes of casting may be distinguished, viz. ingot casting, and plate or strip casting. In the former method the metal is poured into moulds producing brick-shaped ingots, which are to be re-melted for ordinary castings or for further mixing. In the latter method the metal is poured into flat closed moulds, producing a strip or plate of metal to be rolled into sheets or otherwise. 54. Plate or Strip Casting. The moulds for strip casting are made of iron, and consist of two halves fastened together by a ring and wedge, so arranged as to be easily detachable from each other. The plan and section, Figs. 31, 32, will show the construction without further explanation. ii BRASS CASTING 185 The sizes of the mould vary considerably, depending upon the dimensions of the sheet or wire strip desired. The following are the running sizes of the strips for sheet: 3j, 4, 4j, 5, 6, 7, 8, 9, 10, and 12 inches wide, 18 to 28 inches long, and f to | inch thick. The ordi- nary sizes of the strips for wire are 3^ inches wide, inches thick, and 7 feet to 7 feet 6 inches long. In order to obtain per- fect castings the metal must be poured at the proper temperature ; for if the metal is too hot a porous casting will result, and if too cold the mould will be imperfectly filled and the metal non-coherent in parts. Two grave defects are liable to occur in castings, arising from different causes, viz. blowholes and spillyness. When metals are melted in ordinary crucibles and furnaces a certain amount of air penetrates the metal, and certain gases, such as hydro- gen and carbonic oxide, which are generated during the process. As the metal cools these gases are gradually liberated, and this evolution is greatly facilitated by mechanical agitation, which is usually effected by vigorously stirring with an iron rod. If the metal be poured at too high a temperature into a closed mould, some gases will be retained in the metal, forming cavities, and producing a honeycombed appearance, termed by the workmen " spuey metal." This is especially the case in a large mass, when the surface is rapidly solidified, while the interior remains in the molten state ; the means of escape for gases from the interior being thus effectually 186 MIXED METALS CHAP. cut off. For this reason it is a disadvantage to cool a casting too rapidly. The mould is therefore heated previous to running in the metal. The best plan of heat- ing a mould is to have a plate of iron, which loosely fills the cavity of the mould, made redhot in a furnace, then placed inside the mould until the necessary temperature is attained. The iron plate is then withdrawn, the two halves of the mould detached, the interior surfaces oiled, and then powdered charcoal dusted on to prevent the metal sticking to the mould. Some coating material is absolutely necessary, but it is a frequent cause of defective casting, producing both unsoundness and spillyness. Now air is unavoidably carried into the mould with the molten metal, and the oxygen unites with the carbon present to form the gas carbonic oxide ; also hydrogen gas is generated by chemical change j if, therefore, such gases are not liberated before the metal solidifies, they will produce cavities in the metal. It is advisable then to use charcoal dust and oil very sparingly. The author is informed that a much better plan of blacking moulds is to use a mixture of resin and lard-oil. Three parts of resin are melted with one part of the best lard-oil, forming a viscous mass of about the consistence of treacle ; this is applied to the surface of the hot mould by means of an ordinary paint-brush. By the use of this composition much better and more uniformly sound castings are obtained. A spilly casting is produced by the admixture of im- . purities in the metal, causing imperfect cohesion. This may be caused by loose particles of metal in the unalloyed state, but a more common cause is the presence of charcoal or dross, run into the mould along with the metal ; or charcoal and dirt detached from the sides of the mould, when oil and charcoal are used as a preventative -for sticking of the metal, as stated above. Great care should always be given to properly skimming the metal, and seeing that no dirt is carried into the mould. The metal should be poured in a clear and uninterrupted stream, otherwise serious flaws may occur, and the casting be rendered useless. It was formerly ii BRASS CASTING 187 considered impossible to use iron moulds for plate-casting, as, through inexperience and ignorance of the conditions necessary for success, the castings very frequently turned out to be failures. Loam moulds were sometimes used, but, as they readily break, more trouble was experienced. For small work moulds of sand, [thoroughly dried, were employed in some foundries ; but sand is liable to crack and injure the plate of metal. For the most part granite moulds were used, and were considered to yield the best results. The preparation of granite moulds requires great care ; it is requisite to line them with a thin coating of clay, which must be kept in such a condition as to ensure the greatest uniformity of surface of the plates. The clay coating was covered with a thin layer of cow-dung to prevent it cracking. Brannt gives the following description of the method : "The prepared granite moulds 1 are arranged in the following manner. The upper plate is suspended over the lower one, the space or mould between the two being limited by iron bars laid on the lower stone, which is a little longer than the upper one, and projects to the front, so as to form a lip or mouthpiece for receiving the metal. The plates are bound together with iron, and raised on one side so that they stand at an angle of 45 while the metal is run in. As soon as the casting is finished, and the metal is supposed .to be solidified, the sheet of brass is carefully taken from the mould. With sufficient precautions such granite moulds can be used for a long time without the coating of clay being damaged, and the sheets turn out very uniform after the mould has once been heated by several castings. One and the same mould is frequently used continuously in order to keep it warm, and if it has to stand empty for some time, it is enveloped by a bad conductor, such as a coarse carpet, to prevent its cooling. If the mould is damaged it must be carefully mended, and the mended places sharply dried to prevent cracking." Plate brass after casting is carefully inspected, and 1 Such moulds are not now used. 188 MIXED METALS CHAP. subjected to a mechanical cleansing previous to rolling. During the rolling process the brass becomes hardened, and requires occasional annealing. As annealing blackens the metal, due to the formation of oxide, it is advisable to cleanse it in a bath of dilute sulphuric acid and scour with sand Fig. 34.. Scale of Feet if necessary, and lastly to well swill in water. Sheet brass is annealed in a reverberatory furnace represented in section and plan, Figs. 33, 34. The furnace is so con- structed as to prevent oxidation of the metal as much as possible. The fire-bridge a is high, and the charging door 6 is at the front. The waste gases and heat are drawn off ii BRASS CASTING 189 by the flue c. To facilitate the introduction and withdrawal of the sheets a roller d is arranged at the front of the charging door, and on the bed itself are movable cast-iron bars e e, which favour the sliding in of the sheets. They also assist in the heating by isolating the sheets from the brickwork of the bed. Annealing furnaces, heated by gas from Wilson's gas- producers, have been worked for some years in Lancashire with economical and advantageous results. Experience has now been gained, and it is stated that the sheets come out admirably in colour and condition of surface. After passing through the rolls, sheet brass may require to be left soft and flexible, or hard and elastic. For soft brass the sheets are finally annealed. For hard brass the metal is passed through the rolls two or three times after the last annealing. Plate brass intended for wire is first rolled to a certain gauge to obtain the requisite thinness. This preliminary rolling is not only advantageous in obtaining metal of a certain thickness, but the mechanical treatment imparts to the brass greater strength and ductility than if the metal were cast of the desired size for the slitting-rolls, without first passing through the flat rolls. The sheet-metal is first cut into strips, and then the strips are cut into rods by means of slitting-rolls. These consist of spindles carrying steel discs, fixed at suitable intervals. They are so arranged that the discs on the upper spindle project into the spaces of the lower series, and when revolving form a rotatory shearing-machine. On inserting one end of the brass strip between the guides it is drawn forward by the shearing discs, and cut into rods, which, if necessary, are afterwards cut to length. The exceedingly thin sheet -metal or leaf, known as Dutch-metal, is not produced entirely by rolling, but by a combination of rolling and hammering. The metal is cast in thin plates and reduced to a certain thickness by rolling until a thin ribbon is obtained, frequent annealing being 190 MIXED METALS CHAP. requisite. The ribbon is then cut into portions about one inch square, and a large number of these are piled on the top of each other, each piece being separated by a sheet of specially prepared tough paper, the whole forming a packet enclosed in parchment. The packet is then hammered on a block for some time with a heavy hammer, until each piece of metal is extended to sixteen times its former dimensions. These sheets are each cut into four, and the pieces so obtained made into a packet as before, except that a layer of gold- beater's skin is placed between each sheet, and the beating repeated as before until the requisite degree of thinness is obtained. Careful annealing at certain stages of the process is requisite to prevent the sheets of metal cracking. 55. Ingot Casting. In casting ingots of brass, which have to be subsequently re-melted, less care is required than with plate-casting, but when the metal is employed for casting various articles in sand moulds, equal care is ab- solutely necessary. Crucibles are invariably used for mixing and melting brass for the above uses, and the metal poured directly from the crucibles into the moulds. Brassfounders' melting-pots are made of plumbago mixed with fire-clay and coke-dust in varying proportions, a description of which has already been given. These crucibles, although more expensive than clay ones, are much more durable and stand a greater number of meltings, so that they are now most generally used. 56. Air-drying Stove. The flue from the casting furnaces is generally used for heating the stove, and so constructed as to pass underneath the drying stove, Fig. 35. Where such an arrangement is not convenient an ordinary fire-grate is employed. The flue in both instances passes under the stove, and the arch is constructed of specially shaped fire-bricks, as shown in Fig. 36. The size of drying stoves varies according to the requirements. Those used for drying cores for tube-casting are generally 6 feet II BRASS CASTING 191 long, 4 to 5 feet wide, and not less than 6 feet high. The bottoms are made of cast-iron plate, 1 inch thick, and Scale of Feet FIG. 35. perforated with J-inch holes to allow the heat to pass up- wards from the flue. DO LJ FIG. 36. The stoves for drying ordinary brassfounders' cores are . similar in construction, with this difference, that shelves are 192 MIXED METALS CHAP. also arranged in them, upon which the cores to be dried are placed. MOULDING AND CASTING 57. The valuable property of moderate fusibility, which many metals possess, enables articles of various kinds to be produced from a pattern, by impressing upon sand or other mobile material a copy of the pattern, and subsequently pouring molten metal into the cavity thus obtained. The solidified metal will be an exact counterpart of the impression in the sand. The term " moulding " is used to signify the various opera- tions concerned in preparing the impression to receive the metal, and the term " casting " is applied to the whole opera- tion of producing the object, and includes both the moulding and running in the metal. The principal materials used in moulding are Sand of various kinds, loam, plaster of Paris, blackening, pea-flour, etc. Sand is by far the most common, and certainly the most perfect and convenient moulding material. The properties which make it so valuable are its porosity, adhesiveness, mobility, practical infusibility, and unalterability. By means of its porous nature the gases generated during pouring of the metal can freely escape ; by its adhesive property a perfect impression is produced from any given pattern ; the metal is firmly retained when run into the mould ; by its mobility it gives way sufficiently, and the metal flows into the finest markings of the pattern when pressure is applied ; and in virtue of its non-fusibility and chemical unalterability, the heat of the molten metal does not fuse it, or change its chemical composition. The best kinds of moulding sand employed for casting brass have been found to have an almost uniform chemical composition, varying only in size of grain or aggregate form. It contains Silica . Clay . Iron oxide 93 to 95 parts 6 to 3 1 to 2 ,, ii MOULDING AND CASTING 193 Sand containing other metallic oxides, such as lime and magnesia, is too weak or too close that is to say, it will not retain its form, or it will cause the metal to boil by its closeness, the gases not having a free means of escape. When the oxide of iron is greater than the above proportions the sand is liable to fuse, and unite with and blister the surface of the casting, generate gases, and cause blowholes in the metal. Different kinds of castings require different kinds of sand. One class necessitates the use of a porous and yet adhesive sand ; in another the sand must be very fine, free from grit, and very adhesive, so as to conform to the finest parts of the pattern. The best moulding sand is often found along the banks of large rivers, in the vicinity of granite or slate mountains ; or in coal districts, where the river flats are largely composed of sand. In such localities it sometimes contains too much oxide of iron, and is liable to melt ; but this may be modified by mixing it with coke dust. Kampmann states that a good sand for moulds may be artificially made from the following mixture : Fine quartzose sand .... 93 Red English ochre 2 Aluminous earth, the least possible calcareous 5 A valuable casting sand is obtained from the new red sandstone at Birmingham. The value of the quarry of this sand at the Old Cemetery was estimated at 20,000 sterling. Gore-sand. This sand should be coarse, porous, and very adhesive, such as rock-sand, the fine material from abraded rocks ; free-sand from river-banks, or from the sea-shore, and pounded blast-furnace cinder, etc., are often mixed with fine, strong sand and a little clay to make it adhesive. In each case fresh sand must be used for a core, as old sand, burnt sand, or sand mixed with coal is not advisable. One part of clay mixed with nine parts of free-sand is sufficiently strong for small and simple cores, but for large and com- plicated ones a stronger sand is required. o 194 MIXED METALS CHAP. Parting-sand. A substance which does not retain moisture is required for this purpose. Red brick-dust is preferred, but free-sand, sea and river sand, and blast-furnace cinder are also used. Facing -sand. If molten metal is allowed to come in immediate contact with some kinds of fresh sand, a surface fusion of the sand takes place, with consequent roughness of the casting. If the raw sand is too coarse the metal will penetrate the sand to some extent, and also produce a rough casting. To avoid this defect the sand is coated with carbon or carbonaceous matter in the form of fine dust. Carbon and coal dust do not adhere well to old sand, so that in such a case the mould is first dusted with pea-flour and then with charcoal. Charcoal powder is also mixed with one-tenth of its volume of fine sand, and used as a facing for small castings. The success in casting is not only dependent on the manipulative skill of the founder, but also on that of the pattern-maker. If a wood-pattern is required, pine, mahogany, oak, and other kinds of wood are sometimes used. In the brass trade boxwood is most commonly employed. The pattern is made larger than the size of the required casting, about J inch per foot being allowed for shrinkage and finishing. Patterns with their edges at right angles do not leave the sand without disturbing the impression, so that they are made to taper in the parts which enter the sand. Sharp angles in a pattern should be avoided as much as possible, as they leave an edge of sand which is liable to break off on removing the pattern, and produce a defect in the casting. Such a detect in the mould is remedied by mending the broken places with sand before pouring. Wood patterns should either be varnished or brushed with black- lead to prevent absorption of moisture, and enable them to part from the sand more freely. Beeswax or plaster of Paris is used for stopping up holes and cracks in the wood. It often happens, when a permanent pattern is required, that a metal pattern is cast from a wooden one, and the ii MOULDING AND CASTING 195 former then used as a pattern from which castings are obtained. In such a case the wood pattern should be made much larger than the finished articles to be sub- sequently produced. This is to allow for the double shrinkage, and for the dressing of the cast work. Some- times patterns must have pegs of wire attached to enable the caster to lift them easily out of the sand. Besides wood and metal, casting patterns are often made of clay, plaster of Paris, or wax. Moulding-sand is kept in position by means of shallow iron frames, open at top and bottom, and called flasks. These are of various sizes, each side of a frame having a depth of about 3 inches. A flask consists of two FIG. -37. parts, as represented in Fig. 37, where A is the upper and B the lower frame ; or they may be termed the peg-side and the eye-side respectively. In addition to the two parts shown in the figure, a third frame is often used, termed the odd-side, which is subsequently described. 58. To make the odd-side. The sand is first tempered with water, and passed through a rough sieve with about five meshes to the linear inch. It should be noted here that black or old sand does not hold together so well as raw sand, the black sand containing burnt flour, charcoal, and brick-dust ; and probably the sharp edges of the particles of sand are partly worn off by frequent use, and cannot be wedged so firmly together. If black sand only be used the castings are liable to be blown, the material not being sufficiently porous to allow the air and other gases to escape. The moulding is commenced by placing one of the frames, often termed the she-side, on the top of a flat board, then 196 MIXED METALS CHAP. dusting the inside with parting sand, then adding some raw sand, and filling up with black sand. The sand is rammed down tightly with the palms of the hands, then with the knuckles, and finished with a mallet. The surface is next scraped level with a straight-edged piece of wood, a board placed on the top, and the whole frame with its contents inverted. The patterns are now carefully laid on the mould and the dust bag shaken over them ; this will leave a clear outline of the patterns upon the sand, which is then cut away in order to let them in half-way. The " odd-side," which is always the "peg-side," is now fixed to the lower frame, parting sand added, and then a mixture of raw sand and black sand, and finally black sand, to complete the mould as before. The facing-sand, which is sometimes mixed with black sand, half and half, must be pounded in a mortar and passed through the meshes of a fine sieve. The odd-side must be rammed down as tightly as possible, as upon the compact- ness of the sand the future success of the operations depends. A board is then placed on the top and the "flask" turned over. The board which now forms the top is hammered to loosen the patterns, and the "she-side" taken off and broken up. The "odd-side" contains the patterns, and is now ready to mould from. A frame is fixed to the " odd-side," sand pressed in as before, then covered with a board, and the whole inverted. The patterns are then loosened by hammer- ing the top board, and the odd-side, which now contains perfect impressions of half of each of the patterns, is removed, leaving the patterns in the .she-side. A frame containing pegs is now fixed to the she-side, sand added and pressed as before. This peg-side is then removed and placed aside, leaving the patterns in the "she-side." The "odd-side" is then placed on the "she-side," the flask again inverted, the patterns loosened with the hammer, and the she-side removed, leaving the patterns in the " odd-side " ready to start again. It will thus be seen that the "odd-side" is ii MOULDING AND CASTING 197 used over and over again to make other moulds from, a group of five to six moulds constituting a heat. About six heats of common brass work would be considered a fair day's work. The number of moulds constituting a heat is regulated, however, by the size of the crucibles and weight of the castings. Before the molten metal is poured in, the impressions from the patterns in the sand must be connected by the principal ingates or channels by which the metal is run into the mould, the sand being scooped out by a special tool, termed a "drawer." This operation, although apparently simple, really requires much judgment and experience, especially in large work, in order to obtain the requisite amount of metal to feed the impressions. The runners are generally made on the "she-side," but sometimes they arc employed on both the "she-side" and "peg-side." The "cores" are generally laid in the peg-side. For common work the metal is poured into the moulds while in the damp state, the impressions having been previously dusted with flour or charcoal. 59. Fine Work. For fine castings the moulds, after being made, are dried before a fire, previous to dusting, and then dusted with powdered charcoal. For single-face only one side of the mould is dried. For very delicate work the face of the mould is smoked with a torch composed of pitch. In this case, after drying and torching, the moulds must be again brought in contact with the patterns before being screwed up, and the soot being in a very finely-divided state, the impression is brought out clear and sharp. The caster uses a clayey sand, termed " loam," to mix with his ordinary sand for fine work, and for that reason the moulds must be well dried. By the use of loam the chased and sharp corners of the impressions hold together better. When loam is used facing-sand is necessary, and this being of a more porous nature. than loam-sand, permits the gases more freely to escape when pouring, and thus prevents blowholes in the work. 198 MIXED METALS CHAP. When moulding thin scrolls, or when the pattern has a thin part, the moulds must not be rammed too hard, for, as the metal contracts on cooling, if the sand does not give way the metal will crack and the castings become what is termed seared. The caster overcomes this difficulty by damping the thin part of the impression with charcoal and water, which renders that part of the mould soft and yielding when the metal cools. 60. Cores. When the objects to be cast are required to be hollow, they are then " cored," as it is termed. It is advisable in all cases, wherever possible, to allow the pattern to deliver its own core. This can be done by making the pattern to "leave" half-way, or by turning it taper all FIG. 38. through. For example, in Fig. 38, A to C, the light shaded portion marked d represents the core. In many cases this method cannot be adopted. Sometimes the pattern is moulded as though for a solid casting, and the sand core pushed out of the pattern, and fixed on the face of the mould with a splint of wood or a nail, before pouring in the metal. For many articles'; core-prints have to be made on the pattern, and these imprint in the mould the places where the ends of the core will subsequently be supported. The core-bearing must always be of the same diameter as the prints in the pattern. Core-stocks. These are core-boxes, generally made of plaster of Paris or wood, the interior of which contains a cavity, of the desired shape of the interior of the object to be cast. Core-stocks are made as follows : The core-bearing is placed half-way in the sand, and pieces of me'tal or wood CASTING 199 placed round to form an enclosure. Plaster of Paris is then poured over the bearing in this enclosure, and left until it has completely set. The plaster cast along with the bearing is then removed from the sand, trimmed up, and hollows cut out of the sides, and left to dry, when the bearing becomes loose. This forms one-half of the core-box. The plaster cast is then well oiled, placed in sand, an enclosure made as before, and plaster of Paris poured in. When the whole is set and dry the two parts are separated, and the bearing removed. The cork-stocks are then ready for use. Should any difficulty be experienced in getting the parts asunder, the stocks are placed in an oven and heated, when the parts are readily separated. These stocks are then generally used to cast from in brass or iron, so as to form permanent core-boxes. In casting from a pattern which has a hole at one end only, the core requires to be balanced that is, it must be heavier at one end than at the other, the light end pro- jecting into the part which is to be hollow. Very often the core is balanced in the middle, one core being made to do duty for two castings. This principle is taken advantage of in patterns which are undercut, as in the flange of a pulley. This process is called false coring. 61. Figure-Casting, etc. This most complex part of the founders' art is done by the false-coring process, the patterns being generally solid. The figure to be cast is laid as far into the sand as will enable it to leave properly, and when the caster comes to an undercut part, he dusts it with parting- sand, and fits in a piece of sand. When he comes to another undercut part he uses more parting- sand, and rams more sand in as before. In this way piece after piece is successively laid on the pattern, until he comes to a part which will leave the sand without breaking any portion of it. This part is then moulded, removed, and each separate piece of core taken out with two needles, in order to remove the pattern. The moulds and pieces of core have to be dried 200 MIXED METALS CHAP. and carefully replaced, and mended if required. The im- pression after dusting, etc., is then ready to receive the metal. The pieces of sand are removed with two needles stuck in a piece of wood. The relative positions of the cores are readily recognised by their irregular forms. Birds, insects, and parts of plants may be cast by fixing the object to be cast in the centre of a box by means of pieces of cotton. China clay mud, plaster of Paris, or any substance which is not combustible, is thrown into the box, and the object covered. Plaster of Paris crumbles to powder when heated, but if it be mixed with a little potassium sulphate, alum, or borax, the plaster can be heated to redness without crumbling. . When the object is covered with the non-combustible matter, sand is added until the box is full. It is then dried, and the box and its contents heated sufficiently to burn the object to ashes, which are carefully blown out. A place is left as usual for the metal to be poured in. Dr. Bransoris Method. This method was devised for taking copies of ferns, seaweed, etc. A sheet of gutta-percha is softened in boiling water, put upon a warm metal plate, and dusted over with bronze powder ; this dries the surface, makes it smoother, and prevents the specimens sticking to the plaster. A fern is then laid on the top of the gutta- percha, then a smooth plate, and pressure applied. The plate is removed when cold, and from the beautiful impression thus prepared a cast is taken in plaster of Paris, and a casting may be obtained in type-metal, brass, etc. Some articles, such as a pattern for a paraffin-lamp-stand, are moulded as follows : The pattern is moulded in wax or plaster, laid upon a board, and an impression of the outside taken. The mould is then inverted and the pattern re- moved. Parting-sand is dusted in, a frame fixed on the top, and an impression taken of the sunken sand-mould. The thickness of the casting is then determined by placing a layer of clay between the two moulds. Each part of the casting is, in this case, of the same thickness throughout. ii CLEANING, DIPPING, AND PICKLING 201 CLEANING, DIPPING, AND PICKLING. 62. Articles may be cleansed from dirt by washing with water and brushing with white sand, pumice, whiting, etc. Grease and fatty matter as well as lacquer on old work may be best removed by boiling in a hot solution of caustic potash or soda, contained in a cast-iron pot. After boiling for some time they should be removed, and if not perfectly clean it may be necessary to scour with fine sand, swill in water, and again suspend in the solution. Small articles may be freed from grease by dipping in ether, benzine, or paraffin. The best plan is to have three vessels containing the cleansing liquid. The first is used for dipping the articles so as to remove the greater portion of the grease by frequent dipping in and taking out to examine. The second dip is more pure and is used for removing the remainder of the dirt The third or clean dip is used for removing the last traces that may not have been completely removed in the second bath. The article after removal from the third bath is left in the air to dry. The first and second dipping liquid may very conveniently be paraffin, and the third benzine. It should be observed that these liquids are very inflammable, and therefore must be removed from the vicinity of a naked light, especially in the case of benzine. Large and bulky articles, such as copper, brass, iron, and bronze goods, are best cleansed from grease in boiling potash or soda solution. One Ib. of potash to 1 gallon of water is a convenient quantity to make a strong solution. It should be borne in mind that in boiling this solution the water only evaporates, leaving the remainder in a more concentrated state, so that the solution should be frequently made up with water, thus keeping the solution at a constant amount Zinc and tin articles may also be cleansed in potash solution, but care is required, as these metals are some- what soluble in this liquid, and the fine lines of the 202 MIXED METALS CHAP. pattern are liable to be removed. The best plan to adopt in the case of zinc is to allow it to remain only a short time in the solution, then to take it out and scrub it with a brush which has been dipped in fine wet sand. Care must be taken to thoroughly swill the cleaned articles in two or three lots of wash-water so as to remove every trace of potash, especially with zinc and tin articles. More- over, the articles must be kept immersed in perfectly clean water until they are ready for the acid dip and bronzing bath, as they will tarnish if exposed to the air. Caustic potash or soda solution lasts a long time, and when it is partially exhausted, may be renewed by adding fresh portions of the solid salt. Another liquid in constant use for cleansing purposes is a solution of potassium cyanide. Potash and soda solution themselves act as bronzing liquids in some cases, and most metals are coloured more or less when immersed in them, especially when the solutions have been in use for some time. This is not always a defect, but occasionally it is important that the articles should leave the dipping liquid perfectly uncoloured, and then a solution of potassium cyanide is valuable. About 1 Ib. of good potassium cyanide dissolved in 1 gallon of water makes a solution of convenient strength. Besides dirt and grease, which may be removed by the methods just enumerated, articles are often coated with a film of some firmly adherent chemical compound which can only be removed by means of an acid, and therefore is not soluble in water or the liquids already described. 6 2 A. Copper and its Alloys. Copper, brass, bronze, etc., become oxidised in ordinary moist air, and, in con- sequence of the simultaneous presence of carbonic acid, may become gradually converted into carbonates. In fact, the brownish - black to bluish - green deposit, often seen on copper, brass, and bronze goods, is a mixture of oxide and carbonate of copper, mixed with oxygen compounds of zinc or tin respectively when the copper is present as ii CLEANING, DIPPING, AND PICKLING 203 an alloy of these metals. Sulphur compounds are often formed on the surface, when the above metals are exposed for some time to the atmosphere of large towns or rooms where coal-gas is burned. These films may be removed by immersion in suitable acid dips. For this purpose a series of liquids is used : pickle or spent aquafortis is very generally employed for a preliminary dip. The articles are allowed to remain in it until the scale of oxide has disappeared, leaving, after rinsing, a uniform metallic lustre. Dipping in old aqua- fortis is recommended for two reasons : it economises the cost of new acid, and, as its action is slow, it prevents the too rapid corrosion of the cleansed copper during the time of the solution of the protoxide. A dipping liquid may consist of a mixture of 64 parts commercial sulphuric acid, 32 parts of aquafortis, 1 part hydrochloric acid, and 64 parts water. 62e. Copper, brass, bronze, German silver, etc., are often cleaned by heating them to dull redness, and then plung- ing into dilute sulphuric acid. (Those having solder upon them are not heated thus ; neither are articles of cast- bronze, because they would be liable to crack.) They are then soaked in old aquafortis, until, after rinsing, they look uniformly metallic ; they may then be dipped in strong aquafortis for a few seconds and swilled. The straw- coloured aquafortis acts the best ; the white variety acts too feebly, and the red too strongly. It is best to use the dips cold, and to have a considerable bulk of liquid to prevent them becoming too hot by the immersion of the heated metal. In diluting strong sulphuric acid with water a consider- able amount of heat is generated by the chemical action which takes place between them, and if the mixture is made too rapidly the vessel which contains it is liable to be cracked and the liquid to be projected on to the operator. The acid should always be poured into the water and not vice versa. 204 MIXED METALS CHAP To dip gilding metal bright : Immerse it in weak aqua- fortis until there is a black scale formed, then dip it in strong pickle for a few minutes (N.B. Strong pickle is partially exhausted aquafortis; weak pickle is the same diluted with the washings), then dip it quickly into aqua- fortis, then into several wash-waters in succession. There are various mixtures which may be employed for imparting a bright lustre to brass, German silver, etc., by dipping ; the following is one of them : 1 measure of nearly exhausted aquafortis, 2 of water, and 6 of hydrochloric acid ; the articles should be immersed in it a few minutes, or until, after washing off the black mud which entirely covers them, they look bright ; they are then cleaned and dipped again. It is convenient for remov- ing the sand, etc., which adheres to castings. Large articles may remain in this bath for twenty to thirty minutes. 62c. Dipping in Aquafortis, Common Salt, and Soot. Brass and similar articles, after cleaning in pickle, are rinsed in water, well shaken and drained, then dipped in a bath consisting of 100 parts nitric acid, 1 part of common salt, and 1 part of calcined soot. This mixture attacks the metal with great energy, and therefore it should only remain in it a few seconds. The volume of acid should be twenty times that of the articles immersed in it to prevent undue heating and too rapid weakening of the acid. When removed, the articles should be quickly rinsed in water to prevent the production of nitrous fumes. They then present a fine lustre varying from red to golden-yellow and greenish-yellow, according to the composition of the alloy. If the metal is not swilled in water after removing from the acid, there rises on its surface a green froth, and nitrous vapours are given off which indicate the decomposition of the acid with which the metal is covered. When the vapours have disappeared the metal remains dull black, even after swilling. This last mode of operating, called blacking by aquafortis, is used by some colourers to give ii CLEANING, DIPPING, AND PICKLING 205 a dull dark colour to brass and bronze work. Aquafortis is spent when its action on copper alloys is too slow, and when the objects removed from the bath are covered with a bluish-white film. Such acid is termed "pickle," and is used for the preliminary cleaning, or for forming what is termed a whitening bath. Very good aquafortis may appear too weak and cleanse imperfectly by dipping when the temperature is too high, or too low, as in the case of frosty weather. 62D. Whitening Bath. This consists of old aquafortis, sulphuric acid, common salt, and raw soot. Pour into a stoneware vessel a certain quantity of old aquafortis, and add twice the volume of commercial sulphuric acid. Allow the mixture to stand till the next day. The copper nitrate of the old aquafortis is converted into copper sulphate, which crystallises against the sides of the vessel Decant the clear liquid into another vessel and add 2 to 3 per cent of common salt, and an equal quantity of calcined soot. This mixture is less active than the acids used for a bright lustre. The bath may be strengthened when necessary by the addition of aquafortis and sulphuric acid. 62E. Another dipping liquid may be made with equal parts of aquafortis and sulphuric acid mixed with forty times their bulk of water and allowed to cool ; then adding a quantity of common salt equal to about one-fifth that of the strong acid present. Or the following may be used : 1 Ibs. nitric acid, 2 Ibs. sulphuric acid, 10 grains common salt. To the above ingredients add a mixture of the following if a dead surface is desired. 1 Ib. nitric acid, % Ib. strong sulphuric acid, 5 grains common salt,' 20 grains zinc sulphate. 206 MIXED METALS CHAP. The longer the articles remain in this dip the deader will be the surface. They are then thoroughly swilled and dried as quickly as possible. Or previous to swilling with water they may be momentarily dipped in the bright dipping liquid. Mr. Aitkin says that " dead dipping " was discovered in the following way : In the year 1832 a dipper in the employ of Mr. David Malins of Birmingham left throughout the night a quantity of articles in the pickle, and when he attempted to produce the bright appearance in the bright dip they presented a dull, frosted yellow surface. Charmed with the effect, certain portions were burnished and the whole lacquered. Acting on the accidental hint, dead dipping was originated, and has now become the recognised mode of finish for brass work generally. Another liquid for dead dipping may be made of 1 volume of a concentrated solution of potassium bichromate, 2 volumes of concentrated hydrochloric acid. The articles should be left in this solution for some hours, then well swilled in several wash -waters. If, however, they are left exposed to the air for some time without lacquering or further treatment, they become coated with a film of oxide. Dead-dipped articles, while waiting to be bronzed or lacquered, may be kept from oxidising by immersing in clean water, to which half its volume of alcohol has been added. In the case of copper alloys, such as brass, the surface colour will depend not only on the original composition of the alloy, but also on the length of time it has been exposed to the action of the acid. The zinc is oxidised more rapidly than the copper, so that the effect of dipping in aquafortis or other oxidising liquid is to increase the relative quantity of copper on the surface, and to give to the alloy a richer appearance and a deeper colour. When it is desired to clean very small articles and not to appreciably alter the composition, they may be dipped in ii CLEANING, DIPPING, AND PICKLING 207 a solution of 5 parts potassium cyanide dissolved in 95 parts of water. If the coloured brass articles show a granular appearance on the surface after dipping they should be immersed for twelve hours in a mixture of 1 volume of nitric acid, 1 volume of sulphuric acid, and 8 volumes of water. The greyish-black deposit is washed off with water, leaving an agreeable moire appearance. The articles are next immersed in one of the bright dips above described, then passed through a weak solution of caustic soda, or milk of lime, well wasHed in water, and dried out in sawdust. If an article remains too long in the bright dip, after being made dull in the dead dip, the dead lustre disappears. If the bath for giving a bright lustre is not available, the objects, after rinsing, may be rapidly passed through the dead dip to remove the dulness of the lustre caused by too long immersion. After long use the compound acids used for bright dipping will give a dead appearance to brass work. For large embossed work a hot bath for dead lustre is used, composed of Old aquafortis, 4 to 5 parts ; sulphuric acid, 1 part ; zinc sulphate, 8 to 10 per cent. More zinc sulphate is added when required for increasing the dulness of the lustre. The lustre, however, after rinsing the article, and passing it through the same bath for one or two seconds, and well swilling, becomes clearer. For the production of a granular appearance on brass, etc., a mixture of one part of a saturated solution of potassium bichromate in water, and two parts of concentrated hydro- chloric acid, may be employed. The metal is left for some hours until the desired granular effect is produced. It is then removed and well swilled with water. The operation may be considerably hastened by the aid of an electric current, attaching the metal article to the positive pole of the battery and using a brass plate as the cathode. The liquid for this method may be a very dilute solution of sulphuric and nitric acids, or of potassium bichromate and hydrochloric acid. 208 MIXED METALS CHAP. 62F. When a dipping liquid becomes nearly exhausted, or when an article is immersed for too long a time, the surface assumes a dark blackish-grey appearance, or becomes patchy in parts, as the metal is not acted upon. The metal may be restored to its right colour by dipping it in a solution of zinc chloride, taking out, heating till it is dry, and washing in water. This method will also serve for the bright dipping of brass in case it has only a very thin film of oxide to be removed ; the metal must in this case be dipped, boiled, and well washed. Old aquafortis may be revived, to a certain extent, by the addition of sulphuric acid and common salt ; the sulphuric acid decomposes the copper nitrate in it, and also the common salt, and sets free nitric and hydrochloric acids. Crystals of copper sulphate also form at the bottom of the liquid. All the nitric acid may be utilised in this way. For dipping small articles they may be either strung on wire of the same or similar metal, or put into a perforated stoneware basket and then dipped. It is best for the suspending wires to be of the same material as the articles, because they are then less liable to cause a stain. It is less economical, but sometimes necessary, to use baskets of brass or copper-wire cloth. Those who have frequently to cleanse very small articles will find it advantageous to employ a basket of perforated platinum foil, which, though expensive in the first cost, will be found the most economical in the end, as it is not acted on by single acids. 62a. Zinc. When clean zinc is exposed to the air, even at ordinary temperatures, a thin grey film of sub- oxide of zinc soon forms on the surface, which protects the metal beneath from further oxidation. If the metal has been exposed for a long time to the atmosphere of a large town or to the action of impure water, the film becomes more firmly adherent to the metal, and is composed of other bodies than the sub-oxide. When the film is composed only of oxide and is very thin, it is very readily ii CLEANING, DIPPING, AND PICKLING 209 dissolved in a dilute solution of sulphuric acid (15 to 20 parts water to 1 part of acid). For the thicker and more complicated film, mentioned above, a cold mixture of equal parts of sulphuric and nitric acids is best. As great heat is produced by this mixture, the whole must be cooled before using. The zinc article to be cleaned should be suspended for a second or two in this dip by means of a wooden support, then swilled several times with water so as to remove every trace of acid. It is a good plan to let water finally run on the article straight from the tap to ensure a perfect cleansing. The zinc should then be bright and clean. Acid dips for zinc become gradually weaker with use, since the oxide of zinc is dissolved, and combines with the acid to form zinc sulphate, or a mixture of zinc sulphate and nitrate, according to the composition of the solution. Such a liquid will still possess cleansing properties, but will act much more slowly, and the zinc immersed in it, instead of coming out bright, will be dull or crystalline. The dip may in such a case be renewed by adding a little concentrated sulphuric acid. If the zinc article has been coppered and it is required to clean it, it must be dipped in aquafortis till it becomes black, then dipped in one of the former solutions. If it is desired to produce a dull appearance on the surface of the zinc, it should be first dipped in dilute sulphuric acid or a mixture of sulphuric and nitric acids which has become nearly exhausted by use with zinc articles, then put into a bath consisting of zinc nitrate dissolved in very dilute nitric acid. The zinc nitrate may easily be prepared by dissolving zinc in nitric acid until the acid is saturated. Other zinc salts may also be used when acidified with nitric acid. A dipping-bath which contains copper in solution from previous operations will not suit for articles which consist of zinc, tin, lead, antimony, bismuth, or their alloys, as these metals cause a deposition of copper upon their surface. p 210 MIXED METALS CHAP. For small and medium sized articles the above solutions answer well, but with larger articles there are difficulties to contend with in consequence of the great bulk of solution required for immersion, and the size of vessels, which must be made of a material capable of resisting the action of acid liquids. For large and bulky articles a solution of potassium-ammonium tartrate may be used. This liquid has a weaker action on zinc than the ones mentioned before, but dissolves its sub-oxide. The solution is prepared by dissolving Ib. of cream of tartar in a pint and a half of water, adding ammonium carbonate till all effervescence ceases, then adding another 700 grains for excess. The solution is put on the articles with a brush, left on some time to act, then well rubbed in with a sponge, brush, or rag, which has been dipped in a mixture of whiting and water. Lastly, well wash with water. If the article is exposed to the air it soon oxidises, so that it should be plated or coloured as soon as possible after cleaning. Small articles may be kept from oxidising by immersion in a solution of potassium-ammonium tartrate, but large articles, if they cannot be so immersed, should be rubbed with a clean cloth till dry. 62n. Silver. Silver is not oxidised like base metals by contact with moist air, but in the presence of sulphuretted hydrogen is readily coated with a film of silver sulphide varying from yellow to black according to the thickness of the film. Very often the tarnish assumes iridescent colours. This film of sulphide may be removed in several ways. Adhering dirt and grease is displaced as described for copper and its alloys. Tarnish is removed by immersing the silver articles in dilute sulphuric acid (1 : 5), or by boiling in a solution of 1 part cream of tartar and 2 parts common salt. Silver sulphide is readily soluble in potassium cyanide, and in sodium thiosulphate, so that when a tarnished silver article is rubbed with a cloth which has been dipped in a solution ii CLEANING, DIPPING, AND PICKLING 211 of either of these salts it is rendered perfectly clean. A 5 per cent solution of potassium cyanide or a 30 per cent solution of the thiosulphate is a convenient strength for the purpose. Small articles may be immersed in a saturated solution of borax, in contact with a piece of zinc, when the surface after a time becomes very clean. The dull surface may be made bright with the scratch-brush. 62i. Iron and Steel. Articles of iron and steel after the removal of adherent dirt and grease, in the same manner as that described for copper, etc., may be cleaned by immersion in a mixture of lamp-black and concentrated nitric acid, washing with water, dipping in a soda solution, then well swilling with water, and drying out in sawdust. For cleaning iron articles generally a cold mixture of about twenty measures of water, and one of sulphuric acid, is frequently used ; but a better liquid is composed of one gallon of water, one pound of sulphuric acid, with one or two ounces of zinc dissolved in it ; to this is added half a pound of nitric acid. This mixture leaves the iron quite bright, whereas dilute sulphuric acid alone leaves it black, or of a different appearance at the edges. It should be scoured with sharp sand and brushed with a steel scratch- brush. For glassy patches upon cast-iron (which usually con- sist of iron silicate) hydrofluoric acid is used ; it is kept in a bottle of gutta-percha closed by a bung of india- rubber ; it must not be allowed to come in contact with glass vessels, nor must the mouth of the bottle be left open. The fumes from it are extremely dangerous to inhale. If a drop of it falls on the hand it should be thoroughly washed off at once, as it produces ulcers, and causes great pain after a few hours. Articles of iron and steel which have been cleaned in acids, and the adhering acids washed away with water, may be protected from rusting by continued immersion in lime- 212 MIXED METALS CHAP, n water, a solution of caustic soda, or water containing any caustic alkali, until required. Articles of polished steel are cleaned in a moderately strong solution of potassium bisulphate. The article is immersed in the solution in contact with a piece of clean zinc. The zinc decomposes the solution with the libera- tion of hydrogen gas, and the steel is allowed to remain in the bath until the oxide of iron or rust is removed. Steel may also be cleaned in a 20 per cent solution of hydrochloric acid. 62K. Lead, Tin, and their Alloys. These metals are cleaned to remove dirt and grease, as with other metals, by means of caustic alkali solution, and brushing with sand, etc. 62L. Aluminium. Articles of aluminium are cleaned in very dilute solution of potash, when the surface assumes a bright appearance ; wash well with warm water and dry with a warm cloth. Aluminium alloys are treated like copper alloys. CHAPTER III BRONZE 63. The term "bronze" will be applied in this work-to all alloys consisting chiefly of copper and tin. These metals have been known from very remote times, and the importance of the mixture of copper and tin appears to have been among the first discoveries of the metallurgist. Instruments of various kinds were fabricated from these alloys, and weapons were made with a keen cutting edge, harder than iron, and almost rivalling that of steel. The bronzes of the ancients varied considerably in the proportions of the ingredients, for in the main copper and tin only were used, according to the purposes for which they were intended. Sometimes other ingredients were added, either purposely to produce a given effect, or it may be, in some cases, that bodies other than copper and tin were present as accidental impurities. This would arise from the use of impure metals, derived largely from ores of copper or tin associated with other ores, which is often the case. Of late years very great attention has been devoted to the study of copper-tin alloys, and those proportions of the constituents which have been found by experience to give the greatest strength and the keenest cutting-edge are the same as those used by the Greeks and Bomans for their weapons of war and of the chase. The effect was produced by causing the bronze to undergo a process of hammering, as well as a method of hardening, by heating and slow cooling. 213 214 MIXED METALS CHAP. Many ancient coins were made of bronze, containing in some cases lead, zinc, or iron. The following table will show the composition of some ancient bronzes : COMPOSITION OF ANCIENT BRONZES Authority. Specific gravity. Copper. Tin. Zinc. Lead. Iron. Remarks. Old Attic coin . Mitscherlich 88-46 10-04 1-05 Athenian ,, 76-41 7-05 16-54 & Wa'gener 83-62 10-85 5-53 Macedonian ,, O. Monse 87-95 11-44 Coin of Alex, the Great E. Schmid 95-96 3-28 0-76 ?j 86-72 13-14 335B.C. Philippus V. ' . 89 11-0 200 Athens 89-41 9-95 Ptolemy IX. 84-21 15-59 70B.C. Pompey .. 74-11 8-56 16-15 0-28 53 The Atilia Family 69-72 4-77 25-43 45 Augustus & Agrippa 78-58 12-91 7-66 30 Roman sword blade . 91-39 8-38 > 89-3 10-7 Ancient bronze nails . 95-1 4-9 ,, soft bronze . 90 10 medium bronze 88-9 11-1 ,, hard bronze . 87-5 12-5 Roman As . Philiips 8-59 69-69 7-02 21-82 0-47 500B.C. Coin of Julius & Augustus M 8-64 79-13 8-0 12-80 42 Broken sword blade . tt 89-54 10-02 t> : 44 .. Sulphur. . , spear head Celt (Ireland) . 9971 90-68 7-43 1-28 0-28 Sulphur. 00-33 9-19 0-33 0-24 Nickel. Celtic weapon Fresenius 92 6-7 0-69 0-29 0-81 Silver. Coin of Claudius Goth- 8-81 81-6 7'41 8-11 1-86 inus Celtic vessels 88 12 Gallic bell . Girardin 85-9 14-1 Bells of twelfth century Drinking horn . Bronze ring . M Donovan Salvetat 76-1 7!) '34 75-55 20-7 10-87 23-52 1-6 9-11 47 I'-fl It will be observed from an examination of the foregoing table that the principal constituent of bronze is, in all cases, copper, the other components being added to harden or otherwise modify its properties, according to the purpose for which the alloy is intended. Tin has the property of in BRONZE 215 hardening copper, as already stated ; the alloys are capable of taking a high polish ; they present a beautiful metallic lustre, and with their moderate melting points, and fluidity when melted, form excellent alloys for casting. In certain proportions copper-tin alloys emit a beautifully clear sound when struck, the quality of which may be modified by slightly altering the composition of the mixture. Certain varieties of bronze containing, in addition to copper and tin, zinc, lead, manganese, iron, silicon, or phosphorus, are now largely manufactured for machine and engineering purposes. The great feature of modern bronzes is the substitution of triple and quadruple alloys for the old dual alloys. French bronzes nearly always contain the four metals, copper, tin, lead, and zinc, and in some cases small quantities of nickel, arsenic, antimony, and sulphur. Each of these elements exerts an influence on bronze in proportion to the amount present, and if such influence is prejudicial for certain uses, care must be taken in the selection of the metals employed for admixture. Impure copper is by no means a rarity in commerce, and may contain ingredients fatal to the properties of certain varieties of bronze. The difficulty of preparing alloys of definite composition is increased when scrap is re-melted with new metal, unless great care is taken to keep scrap of a given quality separate from other varieties ; such old metal is also liable to contain iron and other foreign metals mechanically mixed with it. Zinc in small quantity added to copper and tin has often a beneficial influence, as in casting, for instance, the metal runs thinner, fills up the mould, and is freer from pin-holes. The zinc probably acts favourably in uniting with any oxygen which may be present, forming oxide of zinc. If the addition of zinc much exceeds what is required for this purpose, the alloy will be weaker, although harder, and the colour will more or less resemble that of brass. For this reason the amount of zinc should not exceed 2 per cent when high tenacity and elasticity are desired as important properties of the alloy. 216 MIXED METALS CHAP. Lead alloys very imperfectly with bronze, showing a great tendency to liquate out on cooling, the greater portion being found in the lower part of the casting. A small quantity of lead is said to make the alloy more malleable and denser. The peculiar patina of a velvety black colour found on old Chinese bronzes is probably due to the presence of lead. Iron, in certain amounts, affects the properties of bronze very beneficially. It hardens the alloy and increases its resistance to wear in cases where the bronze is subjected to considerable friction, as in machinery bearings. Such alloys are paler in colour and more difficult to melt than with copper and tin alone. In small quantities iron increases the tenacity of bronze. In 1858 Parker noticed that the addition of phosphorus during the melting together of copper and tin improved the physical properties of bronze in some respects, and this addition was eventually introduced into bronze manufacture with very successful results. The action of phosphorus in phosphor-bronze is to exert a refining influence on the mixed metals, rather than to form a definite alloy of copper, tin, and phosphorus, since many samples of phosphor-bronze of excellent quality contain but the merest traces of phosphorus. During the melting of copper and tin a certain amount of oxides is formed, which, being soluble in the molten metals, exerts a weakening influence on the alloy by preventing that intimate union of the constituent metals which is necessary to give the strength, toughness, and durability for which some varieties of bronze are noted. Phosphorus has a strong affinity for oxygen, and when brought in contact with metallic oxides, such as those of tin and copper, reduces them, forming oxide of phosphorus. Now this oxide has an acid character, and readily unites with metallic oxides, which are generally basic, to form a fusible slag. This slag, being lighter than the metal and very fusible, floats on the surface and may be readily removed. If the requisite amount of phosphorus be added for the above purpose, the oxygen will be completely removed ; if any in BRONZE 217 excess of the required quantity be added, such phosphorus will unite with the alloy, and may become a source of weak- ness instead of strength. Some metallurgists have thought that the beneficial action of phosphorus is due to its combina- tion with the copper and tin, but such is not probably the case, since, if more than a small quantity be added, the metal is hardened at the expense of toughness ; but the alloy still possesses considerable tenacity, and, for special purposes, may be useful. Also, as mentioned above, chemical analysis proves that the strongest bronzes contain only minute quantities of phosphorus. Montefiori-Levi and Kiinzel, who introduced phosphor-bronze as a material to be used in construction in 1871, state that, besides the deoxidising influence of phosphorus on metals, it performs another very important function ; it has the power of imparting to tin a more crystalline nature, which enables it to form with copper a more intimate union, and thus produce a more homogeneous alloy. The question of producing various qualities of phosphor- bronze depends not so much upon the quantity of phosphorus as upon the correct proportioning of the various ingredients. The alloys are generally prepared by adding a specially prepared phosphor-copper or phosphor- tin (both these metals being sometimes used at the same time) to the bulk of the copper to be treated (see also 73). 64. Phosphor-copper may be prepared in a variety of ways. (1) By dropping phosphorus upon molten copper in a crucible an alloy rich in phosphorus is obtained, forming an extremely hard, steel-grey, fusible compound. (2) By reducing phosphate of copper with charcoal, or charcoal and carbonate of soda. (3) By heating a mixture of 4 parts bone-ash, 1 part charcoal, and 2 parts granulated copper at a moderate temperature. The melted phosphide of copper separates on the bottom of the crucible, and is stated to con- tain 14 per cent of phosphorus. (4) By adding phosphorus to copper-sulphate solution and boiling. The precipitate is 218 MIXED METALS CHAP. dried, melted, and cast into ingots. When of good quality and in proper condition it is quite black. (5) Copper- phosphide is easily prepared by adding to a crucible 14 parts sand, 18 parts bone-ash, 4 parts powdered coal, 4 parts sodium carbonate, and 4 parts powdered glass ; the whole being intimately mixed with 9 parts granulated copper. A lid is then luted on and the crucible exposed to a strong heat The sand acts on the bone-ash, forming silicate of lime. The liberated phosphoric acid is reduced by the coal, and the phosphorus thus set free unites with the copper. (6) Monte fiori-Levi and Kiinzel prepare phosphor-copper by putting sticks of phosphorus into crucibles containing molten copper. To avoid a too ready combustion the sticks of phosphorus are previously coated with a firm layer of copper, by placing them in a solution of copper sulphate. (7) By strongly heating in a crucible an intimate mixture of bone- ash, copper oxide, and charcoal, phosphor-copper is produced. 65. Phosphor-tin. (1) When finely divided tin is heated in the vapour of phosphorus, a silvery-white, very brittle phosphide is obtained, containing about 21 per cent of phosphorus. (2) When phosphorus is dropped into molten tin combination takes place with the formation of a white phosphide, containing about 1 5 per cent of phosphorus. (3) By placing a bar of zinc in an aqueous solution of chloride of tin, a spongy mass of metallic tin is obtained ; by placing this moist tin on the top of sticks of phosphorus in a crucible, pressing down tightly, and then exposing to a gentle heat until the flame of burning phosphorus ceases, a crystalline mass of phosphor-tin is obtained. The following plan has been adopted in some works for the manufacture of phosphor-copper and phosphor-tin. In a cast-iron crucible A, Fig. 39, is placed the requisite quantity of phosphorus, then the top crucible B is tightly joined to A by means of screw clamps d d. The molten metal is poured into B and runs through the opening c on to the phosphorus. The vaporised phosphorus can only Ill BRONZE 219 escape by passing through the molten metal, and is thus almost completely absorbed. Another method is to place the metal to be phosphorised in a crucible provided with a tightly fitting lid, having a hole through the top through which a pipe passes, leading to a similarly fitted crucible in which the phosphorus is placed. The crucible containing the metal is heated in a separate furnace until well melted, when the crucible containing the phosphorus is heated in another fire. The vapours of phosphorus pass through the connecting pipe into the molten metal, into which they are absorbed and combined. Scale of i/. 66. Very small quantities of Fia 39> sulphur, arsenic, and antimony ren- der bronze brittle, J_ per cent being sufficient to modify its properties. The physical properties of bronze depend upon the composition, mode of manufacture, mechanical treatment, and rate of cooling after heating. Riche l has examined a series of copper-tin alloys with regard to fusibility, liquation, and changes of density resulting from certain operations. The alloys having the chemical formulae SnCu 3 and SnCu 4 are the only ones which melt and solidify without decom- position, and their melting points lie between 600 and 700 C. ; all other alloys of tin and copper undergo liquation at the moment of solidification. The several alloys, in quantities of 500 to 700 grammes, were fused for ten hours in tubular moulds, and the top and bottom portions of the castings were analysed. Another portion of each of the melted alloys was stirred during 1 Ann. Chim. Phys. (4) vol. xxx. p. 351. 220 MIXED METALS CHAP. solidification, and the portion which last remained fluid was poured off and likewise analysed. The following table gives the results : RICHE'S TABLE OF COPPER-TIN ALLOYS Composition. Percentage of 1 tin. Percentage of tin in the portion last solidi- fied. Physical properties. Cu. Sn. top. xttom. 1. CllSllg 973 90-27 87-87 92-9 98-50 Tin-grey, soft as tin, non-crystalline. 2. CuSn 3 15-21 84-79 83-15 78-90 96-99 Tin-grey, crystallis- ing by slow cooling. 3. CuSn 2 21-21 78-79 74-97 77-4 64-40 Tin - grey, crystal- lised, moderately hard. 4. CuSn 34-99 65-01 55 80 82-83 Whitish-grey, crys- talline, and brittle. 5. Cu 2 Sn 51-84 48-16 ... 50-42 Bluish -grey like zinc, crystalline, very brittle. 6. Cu 3 Sn 61-79 38-21 37-29 37-66 37-37 Bluish, fine-grained, pulverisable in a mortar. 7. Cu 4 Sn 68-28 31-72 30-44 30-83 30-91 White, laminar, brittle as glass. 8. Cu 5 Sn 72-91 27-09 27-15 26-78 2776 White, with yellow- reflex, crystalline, very hard. 9. CucSii 76-31 23-69 23-37 23-69 25-17 Yellowish, very hard, fine-grained, malle- able at dul 1 red heat. 10. Cu 7 Sn 79-02 20-98 21-0 21-32 24-85 Like No. 9. 11. Cu 8 Sn 81-15 18-85 18-88 18-56 24-6 Like No. 9. 12. Cu 10 Sn 84-33 15-67 15-18 15-18 /20-06 1 24-50 Distinctly yellow. 13. Cu 15 Sn 89-00 11-00 ... 13-1 Gun-metal. m BRONZE 221 The specific gravity of these alloys is best determined by filing off portions from the upper and lower ends of the casting, and taking the mean of the two densities. In alloys rich in tin expansion takes place (that is to say, the specific gravity of the alloy is less than the mean specific gravities of the two metals) up to the proportion CuSn 2 ; alloys richer in copper exhibit contraction, which is small in the alloy SnCu 2 , then suddenly becomes very great, attains its maximum in SnCu 3 and then gradually diminishes ; the greatest density, 8'91, is found in the alloy SnCu 3 , even the more cupri- ferous alloys exhibiting lower densities, e.g. gun-metal, 8'84. The hardness of the alloys, reckoning from pure tin, increases with the proportion of copper down to CuSn. This and all the more cupriferous alloys down to Cu-Sn are ex- tremely brittle, and from this alloy the hardness diminishes as the proportion of copper increases. The hardness of the alloy consisting of 66*66 parts tin and 33-33 parts copper is said to be the same as that of pure copper. The alloy SnCu 3 is distinguished from all the rest by several characters ; it presents the same homogeneous com- position after repeated fusion, is peculiar in colour, has the highest density, exhibits the greatest degree of contraction, and is so brittle that it may be pounded in a mortar. Bronzes containing from 18 to 22 per cent of tin, such as are used for making wind-instruments, have their density in- creased by heating and suddenly plunging into cold water ; but on again raising them to a red heat and allowing them to cool slowly the density is lowered, but not to the value it had before the sudden cooling. By mechanical treatment, such as simple compression or the blow of a coining-press, followed by sudden or slow cooling, the density of these alloys is in- creased, more also (from 8'775 to 8*952) by pressure and sudden cooling than by pressure and slow cooling (from 8 '782 to 8-854). These bronzes, therefore, are affected by sudden cooling and by annealing in the opposite manner to steel. They cannot be worked at ordinary temperatures, because they break too easily ; they are likewise brittle at a red heat, 222 MIXED METALS CHAP. and between 100 and 200 C. But at temperatures a little below dull redness they may be forged like iron, easily hammered out into thin plates, and reduced from ^ inch to -~^ inch thickness by rolling. This property renders them avail- able for the fabrication of gongs, which in external appearance and sonorous qualities, as well as in chemical composition, are identical with the famous Chinese instruments. By the same treatment in the warm state these bronzes are, more- over, rendered denser, and more easily brought to any given density, than by similar treatment when cold. Alloys containing 94 to 88 per cent copper and 6 to 12 per cent tin can be rolled and hammered at ordinary temperatures, and are not increased in density by slow or sudden cooling ; if they are at the same time subjected to mechanical treatment their specific gravities are slightly increased. A bronze containing 6 per cent tin had its density increased from 8-924 to 8'93 2, by 72 blows alternating with 24 annealings ; and by similar treatment, substituting quick for slow cooling, the density was increased from 8 '928 to 8-935. According to the amount of tin present in bronze the colour varies between red and white, and with a large excess of tin it becomes steel-grey. Generally speaking, tin whitens copper more than zinc, 73 parts copper and 27 parts tin forming a white alloy. Alloys containing 89 per cent of copper and upwards are red or reddish-yellow, 88 per cent copper and 12 per cent tin is orange-yellow, and the alloy containing 85 per cent of copper is pure yellow ; from 85 to 74 per cent copper the yellow colour becomes fainter, and disappears with 72 per cent of copper. Alloys with a few per cents of tin are malleable, ductile, and tough, but less so than pure copper. With a greater content of tin the metal becomes less malleable ; the greatest strength, as previously remarked, being found in bronze used as gun-metal containing about 90 per cent copper and 10 per cent tin. From practical experience it has been found that the greatest strength is obtained by so working as to produce the crystals of the alloy as small as possible, even the kind of mould in which Ill BRONZE 223 the casting is effected exerting an influence upon the grain, and through this upon the strength. Articles must be cast at a higher temperature in iron moulds than in sand moulds. Heycock and Neville investigated the nature of copper- tin alloys (Trans. Roy. Soc. January 1904) by studying the cooling curves and micro-sections, and found clear indications v- ^ Qf)O \ a + "" ^ Liquid <^ Liquid 80O \ \ xJ3 p N +/> 700 a + ]8 \ P . a \ 2 o 200 a + 5 10O 5 1O 15 20 25 3C Tin per Cent Flo. 40. of the commencements and end of solidification, the disen- gagements of heat below the point of solidification being sometimes considerable. The diagram, Fig. 40, shows the portion of their curves, embracing the most important of copper-tin alloys. AB, BO give the points of commencing solidification, and all above these curves is a homogeneous liquid. Immediately below is a mixture of liquid and solid. The region a consists of isomorphous solid solutions of copper and tin, from to 9 per cent of tin. The region /3 consists of solid solutions of 22-5 to about 30 per cent of tin 224 MIXED METALS CHAP, in in copper. The region a + 8 consists of two solid solutions, of which 8 is hard and brittle and imparts brittleness to the whole mass. The region a + /3 is composed of two solid solu- tions, between 800 and 486 C., of which /3 is hard and brittle, but much less so than 8, hence by quenching an alloy of more than 9 and not more than 22-5 per cent tin between these temperatures, the a -f j3 condition is retained, and the alloy is malleable within these temperatures and may be forged. If, however, the alloy is allowed to cool below 486 C. without quenching, the /3 crystals break up into a mixture of a + 8, and is then brittle. Guillet tested bronzes con- taining 9 to 22 per cent tin, and found the tenacity and elongation was considerably increased by quenching between 550 and 700 C. With 79 copper and 21 tin the tenacity was 24 tons per square inch and 2*9 per cent elongation. With 87 copper and 13 tin the tenacity was 18 tons per square inch and 13 per cent elongation. With 91 copper and 9 tin the tenacity was 16 tons per square inch and 23 per cent elongation, while without quenching the elongation was only 16 '5 per cent. By annealing for several days at 540 C. an alloy of 81 copper and 19 tin, Shepherd found the elongation was considerably increased. The phenomenon of local contraction of area is hardly perceptible in these bronzes, consequently the ratio of length to diameter has no great effect on the percentage of elongation. 67. The following table of copper-tin alloys was prepared for the United States Board by the Committee on Alloys (Report, vol. i. 1879, p. 390). illl ili& ii.i P*f3 ? 32* !? s I ill . a 2 ::::::! ::: : r: I O ^** C<1 I ::::::: :| : : | : 1 _ I | I I ::::::: :| : : | ; |l 1-1 O> : S : S 6060 ; oo oo oo oo oo oo oo e g >> g ** | tH-!? I! ^ : : S -^P : : :::*-: 00 CO 00 00 00 00 It J fj : : : : :::::: : : : : : I* 3 : : : : :::::: : : : : : llg A fj C<> t~ t- CO 0000000000; |ii < oot- o ^< o -. ? ^- S- *-

>pO51^rH oo oo oo oooo oo ooooi) cs OO CO OO CO 00 CO CO .? fffc a O O C- fl oo oo 3 * i! = !l = = iH If f | 02 02 02 . .*> fii iuf : 8 g 8 ii rHO t3 S a? H S : I:: I : ! ."_ I ill" - __ ^ :r = ~ ~ 21 "tf^*^^ ^*j^^^Jj^,ZS^^;-Z.^^lS *^a;-2 x^*^3^ x^5'-^^3:^j'^jj as 2 : s3^3b2"&3&3&2^3 &3KS : : :">2">>32 &S&3 ^S^S SMit? -r^g-sC^C^u^S^S^ ^2;^ ^C^^ u^Cjfi-S-r^ -SbtcScicto-c tctc Sba; B ft ft ** oo oo oo t- oo oo oo . . 2 . . S . . a s 'is O4 C 1 ! 9 5 -> 9 T ? ?> 2 S S 2 8 S oo - fe 9 * ss ggi;^?^^ ? 5 3 55 S3S 'a S f 1 f 1 oo : : 7 -7 : *o 2 2 g S S S S S S S S 5 ||*|l fe* il* < ' : : g fe a 11 *.! it . co o o ^ "^ ii I! ' fc * a II! I! fc g 5 x : : r. i - OO OO' o o o> -oo o o t^ o?? OOO ) ^*OC<35' MI 2 I, g. s - * W ^ OOOOOOO tS S S & 33 : CO i> 91-8 8-20 ... ... 8-pounder guns . Prussian ordnance 91-66 90-91 8-33 9-09 ... ... French ,, 9073 9-27 90-09 9-9 American compressed 90 10 ... .. ... ordnance 238 MIXED METALS CHAP. Copper. Tin. Iron. Zinc. Lead. American compressed ordnance Russian ordnance (1819) Swiss ,, Chinese , , 90-27 88-61 88-93 77-18 93-19 973 107 10-38 3-42 5-43 69 11 1-16 1-38 '42 5-02 06 13-22 The proportions best suited for guns vary from 8 to 11 parts copper to 1 of tin ; the best of these proportions, according to Karsten, is 9 parts copper to 1 of tin. The alloy of 11 parts copper to 1 of tin appears uniform after sudden cooling to the unassisted sight, but when examined with a lens it appears to be composed of striated faces of a reddish alloy, mixed with a white one. If it be still more rapidly solidified by pouring into thick iron moulds, an alloy is obtained which appears perfectly uniform, even under the lens. When cooled in water, after continued strong ignition, it remains uniform ; but if suffered to cool slowly after con- tinued ignition, it becomes variable in composition, like that which has been slowly cooled after fusion. Hence the alloy which is uniform at the melting point, and likewise at a strong red heat, separates into two different alloys when slowly cooled. The large mass of a cannon cannot be cooled, even by moulds which conduct heat well, suddenly enough to prevent the separation of two distinct alloys, the one that is richer in copper solidifying first, while that which is richer in tin containing 82*3 per cent copper and 17 "7 per cent tin, partly rises to the top and partly sinks into the mould. Gun-metal must possess a considerable degree of hardness and elasticity, as in firing the gun the cartridge strikes several times against the sides, and if the metal yields permanently to the pressure thus exerted upon it, the bore gradually loses its cylindrical form after a time, and its accuracy for shooting at a given object is largely destroyed. Moreover, in firing a cannon a large quantity of gas is generated, which has a more or less corrosive action on metal, so that it is advisable to use that alloy which is least in GUN-METAL 239 affected by such gases, provided it has the requisite strength and endurance. In short, the properties desired for good gun-metal are that it should be very tenacious, sufficiently hard and elastic to resist distortion, indifferent to the ordinary chemical influences, thinly liquid when melted, and capable of settling down solid in the moulds when cast. Modern practice in producing gun-metal has resulted in the exclusive use of copper and tin, as combining the above advantages in the highest degree, and although the addition of a third metal may be advantageous in strengthening one particular property, it is injurious to the required properties taken as a whole. The addition of a little phosphorus is useful in special cases, but the quantity added must be very small. In fact the phosphorus is used only as a purifier, and the amount remaining in the bronze is almost infini- tesimal, and cannot be reckoned as a factor in the final alloy. In old guns many foreign ingredients are found, such as nickel, cobalt, lead, iron, bismuth, arsenic, etc. ; but in many cases the amounts are so small that they may be looked upon as adventitious impurities, and not as purposely introduced into the mixture for alloying. With regard to lead, zinc, and iron, these metals have doubtless been intentionally added in certain cases, with a view to producing a given effect. The other metals mentioned are occasional impurities found in commercial copper, which in former times was much more impure than the varieties obtainable at the present time. The term " gun-metal, " as understood at the present time by many engineers and brassfounders, is not confined exclusively to alloys of copper and tin, as zinc and other metals are very frequently added ; and, in fact, the term seems to be applied to any alloy in which copper largely predominates, and which possesses considerable strength and toughness. Several of the triple alloys of copper, tin, and zinc possess many excellent properties, which fit them for parts of machinery and for anti-friction metals. These will be referred to under the head of "Machine Brasses." The following table will show the composition of a few of these alloys : 240 MIXED METALS CHAP. No. Copper. Tin. Zinc. Colour. I 92 2 6 Pale red. II 90 8 2 Reddish-yellow. III 84 5 11 Yellow. IV 83 5 12 V VI 80 80 15 5 5 15 Pale yellowish-pink. Yellow. VII 75 5 20 Greenish-yellow. No. I is tough, malleable, and tenacious. No. II is hard, somewhat unyielding, and easily broken. Nos. Ill and IV work well under the file and chisel. No. V is hard, but somewhat malleable. No. VI is hard and resisting, tough, and works fairly well with the file and chisel. No. VII is hard and easily broken, but may be filed. The alloys are hard and brittle when the copper is less than 66 per cent of the mixture ; and when the copper is reduced to 50 per cent the alloys are extremely hard and brittle. The addition of a little lead improves the above alloys for turning and filing. A sample of so-called " gun- metal," stated by the user to be very strong and durable, and used for crown-wheel escapements, gave on analysis Copper Zinc Tin Lead Iron 87-85 5-07 4-96 1-84 28 100-00 An alloy prepared by Mr. Stirling, and tried in the Arsenal of Woolwich, has a resistance to flexion much greater than that of ordinary bronze ; it contains Copper .... 87 Tin .... 87 Zinc 4-3 100-0 Ill BELL-METAL 241 The above alloy is difficult to obtain in a sound and homogeneous state. BELL-METAL 69. The various alloys used in the manufacture of bells consist essentially of copper and tin, but in some cases other metals are added in small quantity either for cheapness, or to produce a desired quality of sound. The additional metals chiefly used are zinc, lead, iron, and sometimes bismuth, silver, antimony, and manganese. The following table will show a few of the proportions employed : Copper. Tin. Zinc. Lead. Iron. Silver. Bis- math. Anti- mony. Musical bells Sleigh bells . Gongs . House bells . i >i Large bells . Swiss clock bells Old bell at Rouen Clock bells . Alarm bell at Rouen Tam-tam Japanese kara BUM Japanese kara kane Japanese kara kane White table bells White table bells Small bells 84 84-5 82 80 78 76 74-5 71 72 75-1 79-0 64 70 61 17 40 16 15-4 18 20 22 24 25 26 26-56 22-3 20-3 24 19 18 80 87-5 60 1 1-8 i : b 9 3 6 5 1-2 52 8 12 3 1-44 1-6 18 ... 3 ... 3 12-5 242 MIXED METALS CHAP. In addition to the above alloys, small bells are made of ordinary brass, especially when a wrought pattern is required on the outside for the purpose of ornamentation ; but the sound from such bells is of inferior quality. Mr. P. M. Parson states that manganese-bronze is peculiarly adapted for large bells. The advantages claimed are, that bells cast from it possess the same or greater sonorousness, with a mellower tone, and are at the same time so tough that they cannot be cracked like bells made of ordinary bell-metal, which is made brittle in order to acquire the requisite sonorousness. Good bell - metal should give a pure full sound, the quality of which varies with the purpose for which it is designed. The sound may be modified by chemical composition, but it is also dependent upon the mode of manufacture. The alloy should be hard, homo- geneous, fine-grained, and strong. The colour of bell-metal of normal composition, containing from 76 to 80 per cent copper and 24 to 20 per cent tin, is yellowish -grey, and when very slowly cooled from a red heat is very hard, difficult to file, sonorous, brittle, and exhibits a fine-grained fracture when broken. When heated to redness and suddenly cooled by plunging into cold water it becomes moderately soft, and capable of being easily filed, turned, and otherwise worked. It may be hardened by heating to redness and allowing to cool slowly. At a temperature not far below redness it is malleable. An alloy of 80 per cent copper and 20 per cent tin, when slowly cooled after fusion, exhibits a dingy-grey striated appearance and is very brittle. If suddenly chilled in cold water from a low red heat it becomes yellowish-grey and extensible. During the ignition, if the temperature be raised too high, white globules of an alloy, rich in tin, separate out, so that bell-metal appears to be resolved into two alloys at a temperature below its melting point, which on slow cooling reunite and form a brittle alloy, but remain separated if the cooling be too sudden. 1 1 Watt's Die. vol. ii. p. 44. in BELL-METAL 243 The tone of a bell is influenced materially by its size and shape, by the thickness of the metal, and by the ratio of the height to the diameter. The skill of the bellfounder is not only exercised in finding the best composition to be employed, but also in determining the exact shape and size for a required note and tone, which is of special importance in chimes. Some dinner-bells are made of two halves, having a clapper on each side, the two having different sounds, of a major or minor third interval, and thus forming an agreeable combination when struck. Table-bells are sometimes preferred to be white, which is effected by casting in white metal, as given in the preceding table, or the ordinary alloys are whitened by the addition of tin or nickel. For boiling white, the bells are first highly polished and then placed in a hot bath of water with cream of tartar added, and a layer of granulated tin placed between each row of bells in the bath. The bath is kept at a tem- perature of 212 F. for two or three hours, then the bells are removed, washed in clean water, and finally polished with chamois leather. Many bells are now nickel plated. The melting and casting of bell-metal is similar to that of bronze. The copper is generally melted first, the tin subsequently added, and the whole vigorously stirred to promote intimate mixture. When scrap is used along with new metal, the copper and scrap are melted together, and the new tin added as before. In casting small bells no odd- side is required as in ordinary brass casting. The patterns, with the convex side upwards, are placed on a board, and a casting frame placed over them ; parting material (generally brick-dust) is then dusted over them to prevent the casting sand sticking to the patterns. A layer of raw sand, pounded very fine in a mortar, is then laid on the patterns as a first coat, and the frame filled up with ordinary sand. The sand is well rammed down, first with the hands, then with a mallet, and sometimes with the feet. A board is then placed on the top, and the frame inverted ; one side of the mould is now completed. The peg frame of the mould is 244 MIXED METALS CHAP. now fixed on the top of the other, and the inside of the bell pattern dusted. Sand is then added and rammed in as before, a board placed on the top, and the double frame in- verted. It is then beaten with a mallet to loosen the pattern, and the board taken off, thus exposing the patterns, which are now carefully taken off the cores, as they are termed, which are then dusted with a mixture of charcoal and flour. The frame is then screwed together, leaned against the spilling hearth, and the metal poured in. The caster usually makes about five moulds before pouring, this being termed a heat. Small bells are now also cast in iron moulds. It has often been observed that bells cast from metal which has been repeatedly re-melted acquire a disagreeable tone, and this has been attributed to the formation of metallic oxides and the solution of these oxides by the molten metal. If such metal be treated with a substance capable of exerting a deoxidising influence, such as phosphorus, silicon, manganese, magnesium, etc., a complete reduction of the metallic oxides takes place, the liberated oxygen uniting with the reducing agent added and passing into the slag. Deoxidising agents must be used very sparingly, otherwise the excess will enter into combination with the alloy and may be a greater evil than the one it is employed to remedy. Chinese tam-tams and gongs are characterised by a strong penetrating sound, which is conferred by the peculiar mechanical treatment they are made to undergo. As soon as the plates are well solidified they are taken from the mould, raised to a cherry-red heat in a furnace, then inserted between iron discs to prevent warping, and the whole plunged into cold water and allowed to cool. After this treatment they are found sufficiently malleable to be worked under the hammer. For large bells the metal is melted in a reverberatory furnace, and the molten alloy ought to be exposed to the heat for several hours, which produces a more homogeneous texture and less crystallisation. If any zinc is to be added to such an alloy, it is advisable to add it in the form of brass, calculating of course the quantity of copper it contains. The Ill BELL-METAL 245 relative quantity of the inetals forming the alloy can be calculated and mixed to this arrangement ; but the melting operation has an influence upon the strength of the metal. The more volatile constituents are volatilised to a greater extent than the copper, so that the founder takes proofs before casting and adds the constituent which is deficient. The trial sample is taken in a small iron ladle, the metal broken when cold, and the quality determined by the character of the fractured surface, tenacity, etc. The following description of the casting of a large bell is taken from Overman's Founder^ Guide : " In Fig. 41 a mould is represented as it is sunk in the pit for casting. The core is built in brick upon an iron plat- form, which is to have ' snugs ' in case the mould is made above ground. This brick core is covered with three - fourths of an inch or one inch thick of hair- loam, and the last surface - washing is given by a finely- ground composition consisting of clay and brick-dust. This latter is mixed with an extract of horse -dung, to which is added a little sal- ammoniac. Upon the core the ' thickness ' is laid in loam sand, but the 'thickness' is again washed with fine clay to give it a smooth surface. Ornaments which have been previously moulded, either in wax, wood, or metal, are now pasted on by means of wax, glue, or any other cement. If the ornaments are of such a nature as to prevent the if ting of the cope without them for the cope cannot FIG. 41. 246 MIXED METALS CHAP. be divided the ornaments are fastened to the 'thick- ness' by tallow, or a mixture of tallow and wax. A little heat given to the mould will melt the tallow, after which the ornaments adhere to the cope, from which they may be removed when the cope is lifted off the core. The ' thickness ' is to be well polished, and as no coal can be used for parting, the whole is slightly dusted over with wood-ashes. The parting between the core and the * thick- ness ' is also made with ashes. The cope is laid on at first by means of a paint-brush, the paint consisting of clay and ground bricks made thin by horse-dung water. This coating is to be thin and fine ; upon it hair-loam, and finally straw- loam is laid. The crown of the bell is moulded over a wood pattern after the spindle is removed, The iron staple for the hammer is set in the core, into the hollow left by the spindle. It projects into the thickness so as to be cast into the metal. The facing of the mould ought to be finished when the cope is lifted off. Small defects may occur, and are, if not very large, left as they are ; the excess of metal in those places is chiselled off after the bell is cast. All that can be done in polishing the facing of the mould is to give it a uniform dusting of ashes. When the mould is perfectly dry it is put together for casting. The core may be filled with sand if preferred, but there is no harm done if it is left open, for bell-metal does not generate much gas, and there is no danger of an explosion. The cope is in some measure secured by iron, but its chief security is in the strong, well-rammed sand of the pit. The cast-gate is on the top of the bell, either on the crown or, if the latter is ornamented, on one side of it. Flow-gates are of no use here, the metal is to be cleaned before it enters the mould ; there is no danger of sullage." SPECULUM-METAL 70. This is a perfectly white alloy which admits of a beautiful polish, used formerly for mirrors, but now only used Ill SPECULUM-METAL 247 for such purposes as the construction of mirrors for optical instruments, and even here they are being gradually displaced by glass mirrors. Its typical composition is represented by the formula Cu 4 Sn, containing 66 '6 per cent of copper and 33*4 per cent of tin. The speculum metal of Lord Ross's large telescope is composed of 68*21 copper and 31 '7 9 tin. This alloy is of a brilliant white lustre and has a specific gravity of 8'811 ; it is nearly as hard as steel, and brittle. The speculum is cast 6 feet in diameter and 5j inches thick, and weighs upwards of three tons. The casting of this mirror was only effected after repeated failures. A mould was made whose bottom consisted of a wrought-iron ring, packed full of hoop-iron laid edgeways, so close that air but no metal could escape through the crevices ; this bottom was turned convex on a lathe, true to the concavity of the speculum ; it was then placed upon a level floor and enclosed by a sand dam left open from above. The metal was melted in cast-iron crucibles, because wrought-iron or clay would have injured the alloy. The cast was carried while hot into the annealing oven, which was previously heated to a red heat, and left there sixteen weeks to cooL Good speculum-metal should be pure white, of a fine- grained structure, perfectly sound and homogeneous when cast, and sufficiently tenacious to stand grinding and polishing without rupture. It should contain 65 to 68 per cent of copper to comply with all these requisites. The following table exhibits different varieties : Copper. Tin. Zinc. Arsenic. Lead. Other metals. English alloy 66-6 33-4 Ross's alloy . 68-21 31-79 ... ... Ancient mirror . 62 32 6 Richardson's alloy 65-3 30 7 2 ! ... 2 silver. Sallit's alloy 64-6 31-3 ... 4-1 nickel. Chinese alloy 80-83 10-67 i ... 8 '5 antimony. 248 MIXED METALS CHAP. Ill 70A. Structural Bronzes are employed for machine purposes, such as gearing, or for bearings. Bronzes for gearing generally contain phosphorus, which must, however, be very small, just sufficient to act as a purifier, as stated on p. 239. Suitable proportions are: copper 88 to 91 and tin 12 to 9 per cent. Bearing Metals of Bronze often contain other metals besides copper and tin, as shown in the following table : Copper. Tin. Lead. Phos- phorus. Relative Wear. Phosphor- bronze Ordinary ,, 80 87-5 10 12-5 9'2 0-8 1 1-5 Lead 77 10-5 12-5 0-92 >> > 77 8'0 15-0 0-86 While the wear is least with lead bronze, the friction is much higher. An alloy containing nickel is now largely used, as it allows of a much greater proportion of lead with the consequent diminished wear. Copper 64, tin 5, lead 30, and nickel 1 per cent. It is not advisable to put zinc into bearing bronzes, as it increases the rate of wear and tends to segregate the lead. Bronze for Statues is generally composed of copper, tin, and zinc, the copper varying with the colour, varying from 87 for reddish yellow to 65 for pale yellow. The addition of zinc imparts fluidity and a greater ease in working. Sometimes lead is also added. The following are examples : Copper. 87 85 83 75 67 Zinc. 3-5 11 12 21 30 Tin. 6'5 4 5 4 3 Lead. 3 Bronze for Coinage and Medals. Bronze for coinage must be very malleable. For medals with high relief the copper should be 97, tin 2, and zinc 1. English and French coinage is composed of 95 copper, 4 tin, and 1 zinc per cent. CHAPTER IV MACHINE BRONZES OR BRASSES 71. Under this general term will be included various alloys employed for different parts of machinery, such as bearings, and parts subjected to great friction. These alloys very frequently contain ingredients other than copper and tin. They must be sufficiently hard to resist wear, capable of being easily cast into various shapes, worked with the file and turning tools, and otherwise pre- pared by mechanical treatment for the uses for which they are designed. For the bearings of large axles and shafting, especially those which revolve with great rapidity, alloys containing 80 to 90 per cent of copper are used. These alloys are capable of being forged at a red heat, a property often required for their manufacture into different shapes. Some alloys are required to possess great strength, so as to resist sudden shocks without yielding ; others are required to offer little frictional resistance under a heavy load when in contact with other metals. 249 250 MIXED METALS CHAP. 72. ALLOYS SUITABLE FOR BEARINGS Copper. Tin. Zinc. Ordinary bearings . 84-6 13-3 2'2 ? > M 83-6 12-6 3'8 Heavy ,, 84 12 4 >J 77 9 14 Main , , 75 4 21 Locomotive axles 86 14 82 10 8 Moderately hard axles 70 22 8 Hard axles 82 16 2 Very hard axles 89 11 Dudley has made a comparative study of the different metals employed for bearings in the United States. He considers the best alloys to contain 74 per cent copper, 8 of tin, and 15 of lead. For white alloys for bearings see p. 296. Different manufacturers hold different views as to the best mixtures to employ for a given purpose, as will be seen by the very varied proportions given in the previous table for bearings. Moreover, an alloy used for a special purpose will not possess the same properties under all circumstances, and the mixture must be modified to suit local requirements. It has been shown in the preceding pages how greatly even small quantities of impurities will affect metals, so that an alloy containing foreign ingredients will have its molecular structure disturbed by their presence, and this has doubt- less led some persons to alter their mixtures, when they should have looked to the purity of the constituent metals employed. Even chemical analysis, however, will not always reveal the reason why a certain alloy possesses certain pro- perties, as the qualities may have been conferred by special mechanical treatment. The mode of melting, mixing, and casting will also influence the final result. IV MACHINE BRASSES 251 The following table is intended to show the composition of various alloys, and may serve as a general guide : Copper. Tin. Zinc. Lead. Other metals. Eccentric rings 90 7-7 2-3 J > 9 J * 66 15-5 18-5 ... Pumps 84 7 9 ... ,, 34 50 16 Kingston valve 84-2 10-5 5'3 ... ... Cocks and glands 81 3 13 3 Paddle wheel pins . 76-8 17-4 5-8 ... Sluice cock-way 81 19 Propeller-blades and 57 u" 29 ... boxes Hydraulic pumps 81 19 Propeller-shaft liner . 80 5'4 14-6 White-metal bush for 5 26 69 propeller Cog-wheels 91 9 Steam whistles 80 17 3 Stuffing boxes . 86 11 3 Mechanical instru- 82 13 5 ments Piston rings 84 2-9 8-3 4-8 Stevenson's socket 19 31 19 31 alloy Iron Sterro-metal for pumps Valve balls . 55 87 6 12 22-5 . . 16'5 Antimony 1 The following tables represent the mixtures employed by large engineering firm, using scrap and new metal : Bearing brasses. Eccentric pumps. Pumps. Cocks and glands. Sluice cock-way. Copper . 38 38 38 38 38 Zinc 1 1 4 6 9 Lead H Tin. 7 4 3 H Old metal 54 57 55 53 53 252 MIXED METALS CHAP. Bearing brasses. Eccentric pumps. Kingston valve. Paddle wheel pins. Propeller blades and boxes. Copper . 56 28 112 56 16 Tin . H 6* 14 12* 4 Zinc 2* 7* 3} 8 Old metal 45 70 . 7 40 84 Heavy bearings. Heavy bearings. Main bearings. Propeller shaft liner. Ingot-copper 16 16 16 56 Block-tin . 2* 3 2-3 6 Zinc | Old brass . 13 32 50 Hydraulic Pumps. Ingot-copper Zinc Yellow brass Or spelter 14 Ibs. 1*,. H,, if,, White Metal Bushf&r Propeller Shaft. Ingot-copper Tin . Zinc Ibs. 32 84 Guillet 1 states that the following bronzes are used in the French State Railways : For Circular Friction. Bearings of grease-boxes, motor connecting-rods, couplings, and eccentric collars 82 copper, 16 tin, 2 zinc. For Alternating Friction. Slide-valves, rings of packing presses, safety-valves and their seats, spindles, screws, etc. 84 copper, 14 tin, and 2 zinc. Parts not subject to Continuous Friction. Cocks, whistles, oil -boxes, injectors, screws, etc. copper 90, tin 8, zinc 2. Bearings for common wagons copper 82, tin 18. Piston -rings, bearings for locomotives, tenders, portable engines, cranes, eccentric collars, valves and clacks, packing 1 Alliages metalliques, p. 538. iv MACHINE BRASSES 253 presses, all screwed friction pieces, stuffiDg-boxes, oil-boxes, rods of chain brakes, seats of balls, guide-boxes ; all parts connected with injectors except the body and steam parts, steam valves, sheaths of press linings, rings of packing presses, press linings and stuffing-boxes ; pulleys, wedges of slides for oil-boxes copper 85, tin 13, zinc 2. Wheels for escapement, steam parts, etc., of which the middle is screwed, water-levels, caps of safety-valves, plugs, stoppers, screw-jacks, seats of taps and cocks, screws of slides, gearing, screws and washers, air-valves for ejectors, shells of bearings for carriages and wagons copper 90, tin 8, zinc 2. Slides for distribution and regulators, bearing rings of connecting-rods, bearings of small heads of motor connecting- rods, main screws for reversing movements, shells of wagon bearings copper 78, tin 11, zinc 7'5, phosphor copper 3'5. The following are recomTne'nded by the French Admiralty : Bearings of large connecting-rods without anti-friction balls for bearings copper 84, tin 16, zinc up to 2 per cent. Guide -blocks, bearings of axles without anti- friction friction plates, keys for slide-valves and cranks, slide-valves, etc. copper 86, tin 14, zinc up to 2 per cent. Stuffing-boxes, frames for compensators, eccentric collars and carriers, cylinders of air-pumps, bearings carrying anti- friction sockets of packing presses, gearing, covers for pistons and plungers, feeding pumps, cock- valves and seats of metallic valves, detent cylinders and pistons copper 88, tin 2, and not more than 2 per cent of zinc. Springs, twyer boxes, bolts, heads of plummer blocks, covers of air-pumps, seats and catches for caoutchouc valves, bodies of feeding pumps, screws, fly-wheels, linings of water gauges, feeding-boxes, piston-rods, etc. copper 90, tin 10, and not more than 2 per cent of zinc. SPECIAL BRONZES When the alloy contains elements other than tin purposely alloyed with copper, it may be termed a " Special Bronze " ; thus we may have zinc, phosphorus, manganese, silicon, 254 MIXED METALS CHAP. aluminium, etc., alloyed with copper, with or without tin. The addition of zinc has already been considered, but it may be added that zinc promotes closer union of copper and tin, gives sounder castings, produces greater fluidity, and lowers the melting point. Guillet found that zinc up to about 1 per cent raises the tenacity and elastic limit ; beyond that amount the tenacity diminishes with increase of zinc, but the elongation remains unaltered until the zinc reaches 9 or 10 per cent. The hardness diminishes slowly with diminution of tin. The zinc passes into the a solution. The following table gives Guillet's l results : Cu. Sn. Zu. T.S. E%. Shock. Hardness. 90-34 9-62 0-04 17 24-5 6 58 90-45 8-93 0-62 17-8 23-5 6 53 90-24 6-90 2-86 14-4 27-5 6 54 91-36 4-52 4-12 10-3 16-5 6 49 90-47 2-90 6-63 11-4 23 4 49 90-35 9-65 6-4 13 12 39 T.S. = tensile stress in tons per square inch. E = elongation per cent. Bronzes with Lead. Lead does not alloy with bronze, but remains in the free state. If cooled quickly, the lead is distributed through the mass in small grains or patches. If cooled slowly, the lead liquates towards the bottom and edges. The general etfect of lead is to lower the tenacity, but it has little influence on the elongation when it does not exceed 3 per cent. Beyond that amount the tenacity and elongation are considerably diminished. The same remarks apply to bronze containing zinc. Lead slowly diminishes the hardness. Lead bronzes are chiefly employed for bearing metals in consequence of their greater plasticity than ordinary bronzes. A little nickel (about 1 per cent) prevents liquation of lead 1 Alliages metalliques, p. 546. iv PHOSPHOR-BRONZE 255 in bronze and enables a much larger proportion of lead to be present. The usual composition of such a bearing metal is copper 64, lead 30, tin 5, and nickel 1 per cent. The lead and nickel are uniformly distributed. It is not practicable to exceed 30 per cent lead with 5 per cent tin. PHOSPHOR-BRONZE 73. This is an alloy of copper and tin containing a small quantity of phosphorus and other ingredients in definite proportions, the general properties of which have been already discussed, so that the mode of manufacture and the suitability of the metal for machinery and engineering purposes need only be here stated. It is prepared by melting and mixing copper and tin in the usual way, and adding a certain quantity of phosphor-copper, or phosphor- tin, or both, and well incorporating the ingredients by vigorous stirring. A new plumbago pot is used so as to avoid contamination from other metals, and charcoal or coke added as a covering to prevent unnecessary oxidation. For large castings the moulds are thoroughly dried and dressed with a mixture of black-lead and water. Small work is cast in ordinary green sand. In cooling from the molten state it passes directly from the liquid to the solid condition, without passing through an intermediate pasty state If re-melted it does not sensibly alter in composition, except that when phosphorus is present in notable quantity, that element is slightly decreased by volatilisation. If the alloy be poured into the moulds at too high a temperature, a certain amount of separation of the constituents occurs as in ordinary bronze, so that it is advisable to pour phosphor-bronze only just before the setting takes place. This is effected by addiug ingots or runners to the molten metal, and when the metal no longer melts these, but adheres to them, it is a sign that pouring should take place. For rolling ; drawing into wire, rods, and tubes ; making bolts, springs, screws, etc., the tin should not exceed 4 to 5 per cent, and the phosphorus less than ^ per cent, the 256 MIXED METALS CHAP. remainder being copper. It can be forged into firearms of various descriptions. For pinions, valves, steam and boiler fittings, pumps and general ornamental castings, the tin is about 7 per cent, and the phosphorus varies between -15 and *25 per cent. This is a strong, tough metal, and much harder than the preceding alloy. For axle-bearings, slide-valves, bushes, cog-wheels, and all parts of machinery exposed to much friction, a metal of great hardness and strength is required, and should contain from 90 to 91 per cent of copper, 8 to 9 per cent of tin, and from | to 1 per cent of phosphorus. When the amount of phosphorus is much greater than the above, the alloys are harder, less malleable and tough, and in cases where hardness is the chief requisite, can be made to rival steel in this respect. With upwards of 4 per cent phosphorus the bronze is useless. Phosphor-bronze possesses the advantage of not becoming crystalline under the action of repeated shocks and bends, and is therefore well adapted for wire rope. It resists the action of sea-water better than copper, and also the corrosive action of water in mines better than iron or steel The following table published by the Phosphor-Bronze Company shows the results obtained with various axle- bearings ; but as the compositions of their phosphor-bronzes are omitted, the table is robbed of its chief value as a standard of comparison. [TABLE > 3 11 1 if i < .0 < II 1 SjteLf I oob i-l US i-H Tj< CO TfOO-r}* CO rH O OO o ooco OTH o O CO r^ CN O * O CO O CO 10040 "^ co o t i O CO i-l t^- CO CS O? ofco" oT r-T cT tCi-T i-H rH rH r- 1 O iKJlii so oooo Oi i IVOCO (N rH COCN Composition 100 parts alloy. - CO > ^ I . i a "| SI s g C5 J? J- . Ho efl ?' IB d(|- PQ ^^ Pi " g PM I S S S fc 00 H. s i o H =8 s 5^ 12 cj 1 PQ N- "^ 2 '" a -s f^ ^ *t io| ee * < pq ** -R *- I 0. 5 CO n|w OS HIM ee $* * < !< <3 1 ? (N H* efl c*o ee H^H S- C5|00 !< |aotw 's ^ 8 i -H R < H* ee <- * 10* cT c co* 1 s -* o* r4* r-T co* oo* i ;gg > o t- I C$ OiO 000000 tn eo o co eo o ^-eo^~ co5>p*-p5 esuig jo 'azuoag-mniuitnniv * g a 11 CO O CO O r-l 5 d 6 01 f rH t< J- IT ALUMINIUM-BRASS 287 Aluminium alloys containing varying proportions of nickel are made, and are said to be very ductile and to possess a tenacity of from 33 to 44 tons per square inch. Tests made by Kircaldy on similar alloys, manufactured by the " Crown Metal Company," gave results ranging from 39 to 42 tons per square inch. According to Mr. Webster, two alloys were employed by the Company for preparing the bronze, desig- nated as aluminium alloy A and nickel alloy B. The A alloy consists of 15 parts aluminium and 85 parts tin. The B alloy consists of 17 parts nickel, 17 parts copper, and 66 parts tin. The metals were melted in the usual way with the use of a flux, under a cover of common salt and potassium chloride. The two alloys were then, melted together with copper. It has been found that the bronze is the harder and better the more it contains of the two alloys, and vice versa. The following was recommended as the best proportion : Copper, 88 parts, and 8 parts of each of A and B. The copper is first melted and the alloys added ; the mixture is then stirred with a wooden or clay rod (an iron rod must not on any account be used) until the mass is homogeneous, as shown by a test -ingot. A second quality alloy, which is cheaper than the preceding, was composed of 92 parts copper, and 4 parts each of the alloys A and B. One great drawback to the use of aluminium - bronzes for manufactured articles is the difficulty experienced in soldering and brazing them. The Cowles Electric Smelting Company issued the following directions : 82. Brazing. Aluminium-bronze will braze as well as any other metal by using brass solder (copper 50 per cent, zinc 50 per cent) and J borax. 83. Soldering. To solder aluminium-bronze with soft solder : Cleanse well from dirt, and grease the parts to be joined. Then place the parts to be soldered in a strong solution of sulphate of copper, and place in the bath a rod 288 MIXED METALS CHAP. of soft, iron, touching the parts to be joined. After a while a copper-like surface will be seen on the metal. Remove from bath, rinse quite clean, and brighten the surfaces. These surfaces can then be tinned by using a fluid, consisting of zinc dissolved in hydrochloric acid, in the ordinary way, with soft solder. Mierzinski says that Hulot uses an alloy of the usual half-and-half lead and tin solder, with 12J, 25, and 50 per cent of zinc amalgam. Different workers have sought to improve aluminium bronzes by the addition of certain elements, either to alloys rich in aluminium or to those rich in copper. 2 to 3 per cent of cadmium is said to considerably improve the mechanical properties of copper - aluminium alloys, rich in aluminium. Aluminium bronze is improved by the addition of a little silicon and iron, as proved by the experiments of Tefmajer. A selection of his results are given below. (T. S. in tons per square inch.) : Composition. Mechanical Properties. Cu. Al. Fe. Si. S. T. S. E. L. E. % 93-35 4-62 0-89 0-98 0-04 24 6-1 46-6 1 ""' 91-17 5-92 0-78 2-12 0-04 31 10 28-1 89-67 7-08 072 2-72 0-12 34 14 7-4 fa 86-07 10-05 0-98 2-48 0-14 39-9 ? 0-15, f] 90-99 7-96 1-36 31-2 13-5 18-4 89-77 7-43 0-54 2-58 36-3 16 16-5 88-83 7-19 2-27 1-32 0-05 33 12-4 35-4 89-99 7-98 0-89 1-23 32 13-6 12-5 i? 86-71 9-80 0-78 2-38 43-5 19-6 0-4 1 88-16 11-01 0-34 0-80 32 19-5 0-2 We see from the above table that silicon and iron increase the tenacity and diminish the elongation. When the IV CHINESE AND JAPANESE BRONZES 289 aluminium plus silicon exceed 10 per cent the alloys are brittle. An alloy made by Helouis, containing 91 copper, 8 aluminium, and 1 per cent vanadium, gave a tenacity of 33 tons per square inch with an elongation of 32 per cent. CHINESE AND JAPANESE BRONZES 84. The Chinese and Japanese have attained great perfection in making bronzes for art metal-work, of which the following may be taken as typical The first is called shaku-do, and contains I II Copper Silver . 94-61 1-55 95-77 08 Gold . 373 4-15 Lead . 11 ... The above have been used for very large works of art, such as colossal statues. Professor Roberts -Austen states that the quantity of gold is very variable, some specimens which he analysed containing only 1-5 per cent gold. The other alloy is termed shibu-ichi, of which there are many varieties. The precious metals are employed to produce definite results. I II Copper 67-31 51-05 Silver . 32-17 48'83 Gold . traces '12 Iron . 52 290 MIXED METALS CHAP. The gold enables the metal to receive a beautiful rich purple " patina " or coating when treated with certain pick- ling solutions, while shibu-ichi possesses a silvery-grey tint which becomes very beautiful under ordinary atmospheric influences. There are three pickling solutions generally in use. They are made up respectively in the following pro- portions, and are used boiling : I II III Verdigris 438 grains 87 grains 220 grains Sulphate of 292 437 540 ,, copper Nitre 87 Common salt 146 Sulphur 233 Water 1 gallon 1 gallon Vinegar 1 gallon 5 fluid drachms 1 " That most widely employed is No. I. When boiled in No. Ill solution pure copper will turn a brownish red, and shaku-do, which contains gold, becomes purple. Copper containing a small quantity of antimony gives a very different shade to that resulting from the pickling of pure copper. But the copper produced in Japan is often the result of smelting complex ores, and the methods of purification are not so perfectly understood as in the West. The result is that the so-called ' antimony ' of the Japanese art metal-workers, which is present in the variety of copper called kuromi, is a complex mixture containing tin, cobalt, and other metals, so that the operator has a varied series of materials at command with which to secure any particular shade. Each particular tint is the result of very small quantities of metallic impurity. "Another art material termed mokume, which signifies 1 Roberts-Austen, Jour. Soc. of Arts, 26th October 1888. IV CHINESE AND JAPANESE BRONZES 291 wood grain, is very rare. It may be imitated by soldering thin sheets of different metals or alloys together, layer upon layer, 1 as shown in Fig. 43. Then drill conical holes of FIG. 43. varying depth, A, in the mass, or devices in trench-like cuts of V section, B, and hammer the mass until the holes dis- appear ; the holes will thus be replaced by banded circles, and the trenches by banded lines. Prominences as at C may be produced by bumping up the soldered layers from the back with blunt tools. These prominences are filed down until the sheet is flat ; the banded alloys then appear on the surface in complicated sections, and a remarkable effect is produced, especially when the colours of the alloys are developed by suitable pickles. " Oriental art metal-workers often blend metals and alloys of different colours, by pouring them together at a temperature near the solidifying point of the more infusible of the metals and alloys to be associated. In this way, by pouring the comparatively fusible, grey silver-copper alloy on to fused copper, which is just on the point of 'setting,' the metals unite, but do not thoroughly mix, and a mottled alloy is produced." 1 The solder used by Professor Roberts-Austen contained Silver 55'5 Zinc 26'0 Copper 18-5 100-0 MIXED METALS CHAP. Some Chinese and Japanese bronzes have an unusually deep bronze colour, and in some cases possess a very beautiful dead-black " patina." The following analyses by H. Morin show the composition : I II III IV V Tin 4-36 2-64 3-27 3'23 5-22 SST: : 82-72 9-90 82-90 10-46 81-30 11-15 84-29 11-50 72-09 20-31 Iron 55 64 67 22 1-73 Zinc 1-86 2-74 3-71 50 65 Arsenic . 25 25 The peculiar black colour was proved to belong to the substance of the bronze and not to a superficial coating of sulphide. It increases in intensity with the proportion of lead present. The presence of zinc rather impairs the colour. 1 " In imitation of the above bronzes the following alloys were made by Morin : ' II Tin 5-5 5 SET : 72-5 20-0 83 10 Iron 1-5 Zinc 5 2 No. I gave an alloy exceedingly difficult to work, and without giving any superior results as regards colour, furnished castings which were extremely brittle. No. II, on the contrary, gave an alloy exactly resembling Chinese bronze. Its fracture and polish were identical, and when heated in a muffle it quickly assumed the peculiar dead- 1 Comp. rend. torn. Ixxviii. p. 811. IV CHINESE AND JAPANESE BRONZES 293 black appearance so greatly admired in Chinese specimens. Hitherto it has been found difficult, if not impossible, to obtain this depth of colour with modern art bronzes, since the surface scales off when heated under similar conditions." Christophle and Bouillet confirm the results of Morin's analyses, but point out that the presence of lead is by no means essential to the production of a fine black * patina." By peculiar oxidation processes they profess to have succeeded in producing brown, orange-yellow, red, and black patina on pure copper. They are said to consist of the production of cuprous-oxide in two molecular states and of copper sulphide. Two Japanese bronzes, analysed by Kalischer, and having the colour of brass (Ding. pol. J. torn. ccxv. p. 93), con- tained Tin 4-48 4-36 Copper . 76-64 76-53 Lead 11-88 12-20 Zinc 6-53 6-58 Iron 47 33 Japanese bronzes analysed by Maumene (Gompt. rend. torn. Ixxx. p. 1009) gave the following results : I II III IV Copper . 86-46 80-91 88-70 92-07 Tin . 1-94 7-55 2-58 1-04 Antimony 1-61 44 10 ... Lead 5-68 5-33 3-54 ... Zinc 3-26 3-08 371 2-65 Iron 69 1-43 1-07 3-64 Silica 10 16 09 04 Sulphur . 31 ... ... Waste . 26 79 "-21 : 56 294 MIXED METALS CHAP. These alloys showed a hard, granular texture, with small cavities on the interior, and sound on the exterior. In the presence of much antimony their colour becomes sensibly violet, and red with the presence of much iron. The specimens were from 5 to 12 millimetres thick. The alloys were probably not made by melting the metals together, but prepared directly from the ores. 8 4 A. The Action of Sea -Water on Alloys. The use of copper and its alloys as a material for the sheathing of vessels, and in other locations where it is exposed to contact with sea -water, renders it most desirable that a knowledge of the behaviour of materials of a known com- position should be determined so far as corrosion is concerned. For this reason the German Government has caused a number of tests to be made, and from the report of Chief Engineer Diegel, very fully abstracted in Stahl und Eisen, the principal results of these investigations are taken. Since the experi- ments extended over a period of two years, and were conducted with great care and thoroughness, there is every reason to believe that the information thus acquired is the most com- plete contribution to knowledge in this direction which has yet been made. The tests were made by suspending the specimens in the sea-water from a bridge in the harbour of Kiel, so that they were exposed to the action of the water in a manner similar to that which occurs in actual practice. Twelve strips were cut from a sheet of the metal to be tested, nine being suspended in the water, and three were retained for comparison. At intervals of eight months three strips were taken out of the water and broken in the testing- machines, as was also one of the reserved pieces, and thus a comparison of the exact deterioration due to differences in location could be determined. Tests were thus made with immersions of eight, sixteen, and twenty-four months, and, in addition, some trials were made after immersions of sixteen and thirty - two months. As the atmosphere is known to have an injurious effect upon alloys containing iv ACTION OF SEA-WATER 295 much zinc, specimen alloys were also exposed to free contact with the air, in order that the loss of strength from this cause might be determined. The trials were limited to such alloys as are actually used in shipbuilding, or which are likely to be considered for such work, these being the follow- ing five groups : Copper alloys rich in zinc, bronzes containing little zinc, pure tin-bronzes, pure aluminium- bronze and iron-aluminium bronzes. The results of these tests are given at length in the original report, with diagrams showing the relations of the various alloys, but for this the original paper must be consulted. We can give here, how- ever, the general conclusions which were drawn from the behaviour of the different materials, and some comments on their application. Considering first the behaviour of the specimens in air, it was found that the iron-bronze alloys resisted atmospheric influences remarkably well, there being practically no deterioration in strength during two years ; those alloys containing much zinc did not behave so well, but no tests were made as to the resistance of the iron-zinc bronzes to the action of the air. When immersed in sea-water in contact with iron, the iron, tin, and aluminium-bronzes all retained their strength very welL After exposure to the water for two to two and a half years there was no marked difference in appearance, nor was there any reduction in weight, while tension tests showed no reduction in strength. Iron-bronzes in contact with tin-bronze showed a material loss after exposure to sea-water, the zinc dissolving out, but this action was less rapid with the aluminium -bronzes. A specimen of iron-bronze in contact with tin-bronze after two years' immersion in sea- water showed a loss of two- thirds of its original strength and four-fifths of its extensibility, and the structure was partially destroyed by the eating away of the zinc. Cast and wrought bronzes appeared to be equally affected. A wrought plate of iron-bronze in contact with a cast plate of the same material was rapidly affected, and in two years its strength was diminished by about 60 per cent An iron-bronze and a phosphor-bronze, attached to a piece 296 MIXED METALS CHAP. of oak, showed a slow corrosion of the iron-bronze, 20 per cent of its strength being lost in twenty-three months. The important conclusion which was drawn from the experiments was the influence of the metal to which the sheathing is attached. Thus, for instance, tin- and iron-bronzes both give good results in contact with iron, but are injuriously affected in contact with each other, while the iron itself is least corroded when sheathed with an iron-bronze. The principle which governs the corrosive action between different metals is that of their relative position in the electrical scale, and the same considerations which govern the choice of metals for galvanic action enter into the question of injurious corrosion. Pure aluminium-bronze, for example, resists the action of sea-water remarkably well when in contact with metals which are electro-negative towards it, while contact with electro-positive metals causes its rapid destruction. It is, therefore, most desirable to place those metals in contact which are close to each other in the electrical scale, and any statement as to the resistance of an alloy or sheathing metal to corrosion is misleading, unless it includes a description of the material with which it is said to be used. WHITE METALS FOR BEARINGS Anti-friction Metals 85. The metals entering into the composition of the various alloys employed for the above purpose are : Copper, tin, antimony, lead, and zinc ; but seldom more than three of these constituents are used in any one alloy. In machinery running at high speed, or with great pressure, the bearing surfaces are subject to considerable friction, and in many cases the object of the engineer is rather to reduce this friction to the lowest degree, than to provide a bearing which will withstand great pressure without wearing away. A common practice at the present time is to make the foundation of brass or bronze, and line the bearing surface iv WHITE METAL 297 with a renewable lining of comparatively soft white metal. One great advantage of white metal is its low melting point, so that a worn-out bearing can be readily melted out and replaced by a new one. The white metal is generally melted in an ordinary ladle, and when the journal or mandril is wiped dry and chalked, the molten metal is poured in. Care should be exercised not to raise the metal to too high a temperature, as it not only causes the constituent metals to oxidise unequally, but volatile metals escape, and thus the composition is considerably altered. No dross should be allowed to pass into the bearing along with the metal. The journal should also be warm, so as not to chill the metal too suddenly at the wearing surface. White metal bearings are indispensable for certain purposes ; as, for instance, where the shaft resting in the bearing does not run smoothly. If the bearing be made of hard metal considerable friction is set up, and a battle will take place between the axle and the bearing, the softer of the two being worn away, causing the axle to swerve considerably. By the use of a soft metal u liner" the axle is not worn, and adapts itself to the condition of the bearing, and runs with much less friction than in the previous case, and, as stated before, the lining is easily replaced when worn away too much. This tendency of white metal to reduce friction has given rise to the term " anti-friction " metal as a name for such alloys. The points of great importance in a bearing are : (1) that it should not cut the journal, (2) should be durable, (3) not become heated by friction, (4) sufficiently soft to adapt itself to the bearing surface, and (5) the metal should be capable of being readily melted and cast. Red metal bearings are distinguished by great hardness and power of resistance, and used where the speed is high and the pressure great. Professor Goodman l has investigated the subject of anti- friction metals, and found that some samples of presumably the same composition gave different frictional results, in one instance amounting to nearly 100 per cent The amount of 1 Inst. Mechan. Eng., April 1895. 298 MIXED METALS CHAP. lead, antimony, and tin were the same, but on further investigation minute quantities of impurities were found. By noting the atomic volumes of these foreign bodies he found that when the atomic volume of the impurity was less than that of the main body of the alloy, which was about 1 7, the friction was greater. With aluminium, for example, having an atomic volume of 10*6, the addition of - 1 per cent increased the friction 20 to 30 per cent. If the im- purity, such as bismuth, for example, was greater than that of the main alloy the friction was reduced provided the amount did not exceed -25 per cent. 86. Babbit's Anti-friction Metal. This was once largely used, but is now replaced in many cases by other alloys. Mr. Babbit recommended to melt together 4 Ibs. copper, 8 Ibs. antimony, and 24 Ibs. tin. This he called hardening. For every pound of the above he added 2 Ibs. more tin. Since its introduction many different mixtures have been sold under the same name, some of the tin being replaced by zinc and lead. One receipt gives 2 Ibs. anti- mony, 2 Ibs. tin, and 20 Ibs. lead. The bearing to be lined with Babbit's metal is recommended to be washed with alcohol and powdered over with sal-ammoniac ; and those surfaces which are not to be coated are to be covered with a clay- wash. It is then to be warmed sufficiently to volatilise a portion of the sal-ammoniac, and then tinned. Charpy recommends 83 tin, 11 J copper, and 5 antimony as giving the greatest compressive strength. 87. A white metal alloy for bearings has been intro- duced, under the fanciful title of " magnolia metal," and appears to have found considerable patronage for marine and railway work. Professor R. H. Smith, at Mason's College, Birmingham, submitted this metal to a variety of tests, and states " that it is much superior to either Babbit's metal or gun -metal. It produces less friction, keeps the bearing tem- perature lower, requires less lubrication, and possesses greater iv BABBITS METAL 299 durability. This characteristic of durability is a most im- portant one. Within the wide limits covered by my tests, it would be true to say that the longer the magnolia metal is used, and the more severe the duty imposed upon it, the better becomes its condition. The elevation of bearing temperature above that of the surrounding air is extremely low. The general conclusion I have arrived at from my experiments is, that magnolia metal is a very excellent metal for bearings ; that its specially good qualities appear more particularly when it is subjecte'd to intense pressures, such as could not be borne by other metals without firing or melting, and that under very trying circumstances magnolia metal may be trusted to remain cool that is, at a temperature that does not interfere with good working." An analysis of this metal, made in the author's laboratory, gave approximately Lead .... 78 Antimony . . 16 Tin .... 6 100 88. The following table shows the composition of various white alloys for bearings : [TABLE 300 MIXED METALS CHAP. Tin. Copper. Antimony. Lead. Zinc. Iron. Kingston's metal with 6 per ) cent of mercury added / 88 6 Fenton's metal for axle-] boxes of locomotives and > 14-5 5-5 ... 80 waggons J Stephenson's alloy 31 19 ... 19 31 For propeller boxes . 14 57 29 Dewrance's metal for loco-\ motives / 33-3 22-2 44-4 ... Hoyle's alloy for pivot \ bearings J 46 12 42 ... ... Jacoby's alloy . 85 5 10 For propeller bush 26 5 69 Very hard bearing 12 4 82 2 ... Anti-friction metal 14 6 80 For general bearings 81 5 14 ... j 81 5 14 5 > > 10 10 80 ... ... 12 88 ... Bearings for light work 85 5 10 ... ... > 73 9 18 ... } 76 7 17 ... ... ... heavy work . 90 2 8 > 87 6 7 common work 2 8 2 88 Soft alloy for pillow-blocks 15 85 Vaucher's alloy for lining^ journals ] 18 2-5 4-5 75 ... Charpy's alloy . For piston-rods . 83 12 11-5 5-5 8 80 ... For locomotives 10 ' 25 65 ... Clamer's alloy, for general) 5 64 30 Kiel ell purposes / C harpy states that anti-friction alloys are all formed of hard grains embedded in a plastic alloy. These two condi- tions are necessary, so that the pressure may be supported by the hard parts and the plasticity of the cement may enable the bearing to accommodate itself to the shaft. In this way local depressions are avoided. The following ternary groups iv BABBITS METAL 301 all fulfil the above conditions and are preferable to more complicated mixtures of 4 or 5 metals : Copper Tin - Antimony Lead Tin - Antimony Copper Lead Antimony Zinc Tin Antimony Copper - Tin Lead Bronzes with much zinc may be cheap, but zinc increases the rate of wear and tends to segregate lead. MELTING AND CASTING OF BRONZE 89. Bronze is prepared in crucibles, and in reverberatory furnaces, according to the quantity required for casting. The general rule is to melt the copper first, then add the scrap-metal, and when these have been thoroughly melted together, to add the tin, or tin and zinc, as the case may be, previously heated as much as possible without melting. The addition of the tin cools the copper, and as it is advis- able to get the whole charge melted and mixed as quickly as possible, in order to prevent loss by oxidation and volatilisation, the fire should be kept very brisk while the alloying is taking place. Bronze contracts on solidifying, as with other copper alloys, but the amount of contraction depends on the composition and on the temperature at which it is cast, varying from y^^ to T Y of its bulk. The difficulties attending the casting of bronze are much the same as those already discussed with regard to brass. The tin shows a greater tendency than copper to unite with oxygen, so that in re-melting bronze the alloy becomes a little richer in copper, and therefore a slight excess of tin should be added in the first place to supply the subsequent loss. The metals should be excluded as much as possible from the atmosphere during melting, by covering the surface with charcoal or anthracite powder, or oxides will be formed, 302 MIXED METALS CHAP, iv which will be dissolved by the molten alloy, and impair its strength and toughness. Moreover, bronze has the property of absorbing air and other gases when in the molten state, and emitting them as the surface solidifies. If the castings are thick, or the cast metal is rapidly chilled, the absorbed gases cannot escape from the interior, and therefore produce numerous small hollow spaces or pin-holes. Gases are removed more perfectly from molten metal by mechanical agitation, such as vigorous stirring with a suitable rod, and if the rod be of wood, or carbonaceous material, some of the dissolved oxides are decomposed, the oxygen combining with the carbon and escaping as a gas. The other impurities also more readily rise to the surface through the agitation, and form dross. It is probably for a similar reason that phos- phorus, manganese, and other reducing agents, when added in small quantity, are so efficacious in producing sound castings. The loss of metal in melting, as well as the quantity of the product, is influenced by the construction and arrangement of the furnace. In a reverberatory-furnace a neutral or reducing atmosphere should be maintained, so as to avoid unnecessary oxidation, but the chief point is to get the metals melted and mixed as rapidly as possible ; for, if bronze is kept fused for a considerable time, a white alloy rich in tin separates from the main mass, as previously mentioned. The crucible furnace employed for making bronze is similar to the brass-making furnace, Figs. 19-24. CHAPTER V GERMAN SILVER 90. The alloys, largely manufactured under this name, consist essentially of nickel, copper, and zinc. Different names are used to signify the same substance, such as : Nickel - Silver, Argentan, Packfong, White - Copper, Weiss- kupfer, Neu- Silver, and Mailkchort. Besides these, dif- ferent manufacturers employ fanciful names to denote alloys containing different proportions of the constituent metals, which they consider best suited to produce a given result, a good white colour being a great desideratum. Thus : Nevada Silver, Virginia Silver, Potosi Silver, Silveroid, Silverite, Electrum, Afenide, Agiroide, etc., are simply German silver ; but in some cases a little cobalt is present as well as nickel, and some makers add a small quantity of iron or manganese. 1 to 3 per cent of lead is sometimes added for cast work. The properties which make German silver so valuable are its white colour, lustre, hardness, tenacity, toughness, malleability, ductility, and power of resisting certain chemical influences. When carefully prepared it works well under the stamp and between the rolls, but it is advisable that the metals used in alloying should be as pure as possible, since small quantities of certain impurities, such as arsenic, seriously injure its working qualities. Cobalt frequently accompanies nickel in its ores, and becomes reduced at the same time as that metal, and alloys readily with it ; but as the chemical 303 304 MIXED METALS CHAP. and physical properties of both metals are closely allied to each other, cobalt does not often exert an injurious influence. Iron has a beneficial effect on German silver for some purposes. It makes the colour whiter, increases the tenacity and elasticity, but makes it harder. In some experiments made by the author, 1 to 2 per cent of iron was found to have no deteriorating effect, except with regard to hardness, and the colour of an alloy containing 12 per cent of nickel was equal to one containing 1 6 per cent in which no iron was present, the same quantity of zinc being used in each case. The metal rolled remarkably well. In the better qualities of German silver used for rolling and spinning, iron was found to be injurious, and for these purposes the purer the alloy the better it works. All commercial varieties of German silver contain iron, especially when scrap-metal is used in connection with new metal for the melting charge, the iron being probably obtained from files and other tools used in the fabrication of the various articles. Packfong, the original nickel alloy introduced from China, contains as much as 3 per cent of iron. Gersdorff states that iron is difficult ' to alloy with the other constituents, and when that metal is added to fused German silver it does not combine with it, and forms upon the surface of the fused mass a layer consisting of copper, nickel, and iron. He states that the iron must be previously fused with a portion of the copper, under a layer of charcoal-powder in a blast-furnace, and the alloy formed may then be used to alloy with the nickel, zinc, and the remainder of the copper required. This is quite unnecessary, as proved by the author's experiments, in which copper, nickel, and the best iron wire were strongly heated together in a covered crucible, and zinc added to the molten contents, then vigorously stirred. The metals were perfectly alloyed together, and no separation of iron could be detected when the ingot was rolled into a thin sheet and highly polished. The effect of tin when alloyed with German silver to the extent of from 2 to 4 per cent was found much more v GERMAN SILVER 305 injurious than that of iron, the strips being brittle when rolled, and the metal of a decidedly yellow cast when polished. From his experiments the author concludes that there is no advantage in adding tin to German silver, as it impairs the colour, hardens the metal, and makes it brittle. Moreover, tin is an expensive metal, and better effects can be obtained by the use of iron and zinc. An alloy, patented more than fifty years ago, had the following composition : Copper ... 55 Zinc .... 17 Nickel ... 23 Iron .... 3 Tin .... 2 100 Silver in certain proportions does not impair the malle- ability of German silver, but no particular advantage is gained by its use. M. Ruolz many years ago manufactured a series of alloys for jewellery having the following composition : Silver . . . 20 to 30 Nickel . . . 25 to 30 Copper . . . 35 to 50 By varying the proportions the alloys may be made to resemble silver very closely ; but the nickel and copper must be of good quality, or the alloys will be hard, brittle, and difficult to work. It is stated that French jewellers still use these and similar alloys. An old alloy, termed American nickel silver, contains Iron . . 1 part. Manganese . . 4 parts. Cobalt . 1 Nickel . . 24 ,, Silver . 2 parts. Zinc . . . 36 ,, Tin . . 2 ,, Copper . . 96 ,, Another alloy, having the same advantage of being com- plex, which is its chief merit, was introduced by Mr. Toucas, and consists of 306 MIXED METALS CHAP. Copper 35 '8 parts Lead . 7-1 parts. Nickel 28-7 Zinc . 7-1 Antimony 7-1 Iron . 7-1 Tin . 7-1 According to the inventor, it has nearly the colour of silver, may be worked like it, and laminated by the ordinary processes; it is resisting, susceptible of a fine polish, with the lustre of platinum, and may be silvered perfectly well. He recommends it for objects which are to be spun, hammered, or chased ; but for cast and adjusted pieces he prefers to increase the proportion of zinc in order to increase the fluidity of the metal. Modern proportions for German silver alloys vary consider- ably, each manufacturer having a preference for his own particular mixture for any given quality. For metal which is to be rolled, pressed, or 'stamped, the alloy must be tough and malleable ; and as whiteness in colour is an important consideration, it follows that the metals nickel and zinc must be present in considerable quantity in order to overcome the red colour of the copper. For cast work, which only requires to be filed and turned, malleability is not of first importance, it being sufficient if the metal has the requisite tenacity and toughness, and also sufficient liquidity when melted and adaptability to ordinary moulding purposes. Another item, which cannot be ignored, is that of cost, and as nickel is an expensive metal, it is to the advantage of the manufacturer to employ as much zinc as is consistent with the properties enumerated above. Founders whose specialty is the manufacture of German silver have agreed that the best alloy for beauty, lustre, and working properties consists of the following proportions : Copper ... 46 Nickel ... 34 Zinc ... 20 It is also the most costly among similar alloys on account of the large proportion of nickel present. GERMAN SILVER 307 90A. Platinoid. An alloy of 60 parts copper, 14 parts nickel, and 24 parts zinc, to which 1 to 2 per cent of tungsten is added, is largely used in electrical work on account of its high resistance. 91. The alloys most largely employed for spoons, forks, and other table and ornamental goods contain from 6 to 22 per cent nickel ; the lower qualities being little better than brass, and only used for the very commonest goods ; the higher qualities approximate to pure nickel in whiteness. In order to determine the best proportions to employ for alloys containing 8, 10, 12, 16, and 20 per cent respectively of nickel, the following experiments were made by the author. In all cases the copper and nickel were melted together under a layer of charcoal, and a little borax added as a flux. The zinc was added when the former were well fused ; and the whole was well stirred with a wooden rod before pouring the metal into the iron mould. Ordinary commercial metals were employed. ALLOYS CONTAINING 8 PER CENT NICKEL I II III IV Copper . 58 60 62 64 Nickel . 8 8 8 8 Zinc 34 32 30 28 No. I was the whitest, but the weakest, and rolled the worst of the series. No. II was whiter than Nos. Ill or IV, and was the toughest of the four ; an experimental ingot about \ inch thick was bent double in a vice without cracking. It rolled well, but was not so malleable as Nos. Ill and IV. Nos. Ill and IV rolled equally well, but No. Ill was a shade the whiter. 308 MIXED METALS CHAP- ALLOYS CONTAINING ABOUT 12 PER CENT NICKEL i II III IV V VI VII VIII IX X XI XII SSBT 57-5 12-5 57-5 12-5 57 12 55 12 58 12-5 57 12-5 56 12-5 55 12-5 .'56 12-5 55 12-5 54 12-5 53 12-5 Zinc 30 28 31 31 29-5 30-5 31-5 32-5 29-5 30-5 31-5 32-5 Iron . 2 2 Tin . 2 2 2 2 With regard to malleability, as proved by rolling, the following order was observed, the numbers given being in the order of their capability of being rolled : Nos. IV, III, VI, II, I, V, VII, VIII, IX, X, XI, XII. No. IV not only rolled the best, without cracking on the edges, but was also the whitest in colour. All the ingots were numbered, and passed in succession through the rolls, the ingots being of the same thickness at the commencement. The condi- tion of each was noticed after each squeeze of the rolls, and the thickness accurately measured. They were all likewise annealed together under the same conditions. The first six in order that is, Nos. IV, III, VI, II, I, V rolled well, Nos. I and V being equal in this respect. Nos. VII and VIII rolled moderately well ; and Nos. IX to XII rolled badly, the strips splitting down the middle long before they were reduced to the requisite thinness. For the above rolling tests the author is indebted to Mr. "Wilcox of Messrs. Kemp and Co., Birmingham. The above results may be summarised as follows : That 30 to 31 per cent of zinc, with less than double the amount of copper, gives the best results with respect to malleability and whiteness ; 32 per cent of zinc makes the alloy more brittle and requires more frequent annealing during the rolling process ; 1 to 2 per cent of iron may be present without seriously impairing the malle- ability, the alloys being whiter and harder ; tin is very injurious, and imparts a yellowish tint to the alloy, even 2 per cent making the alloys unworkable. GERMAN SILVER 309 ALLOYS CONTAINING 16 PER CENT NICKEL I II III IV V Copper . 62 60 58 56 54 Nickel . 16 16 16 16 16 Zinc . 22 24 26 28 30 No. V was the whitest and behaved itself irreproachably during rolling. Nos. Ill and IV were about equal in malle- ability, rolled better than No. II, and were whiter in colour. No. I was the worst of the series, both as regards colour and malleability. Any higher proportion of zinc, with 16 per cent of nickel, would probably be injurious. It is here noticeable that the best results were obtained with 30 per cent zinc, as in the 12 per cent nickel alloys, but with a less percentage of copper. ALLOYS CONTAINING 20 PER CENT NICKEL I II III IV V Copper . 60 58 56 54 50 Nickel . 20 20 20 20 20 Zinc 20 22 24 26 30 No. I rolled badly, but the casting was not sound, through being poured too hot, which somewhat impaired its malle- ability. No. Ill rolled well, although slightly cracked on the edges. No. IV rolled remarkably well with a clear feather edge. No. V was about equal to No. Ill, but whiter in colour. From these experiments it appears that 30 per cent zinc with 20 per cent nickel is rather too much ; prob- ably 28 per cent would combine the advantages of whiteness and malleability in the highest degree. 310 MIXED METALS CHAP. 92. The following analyses by the author of different articles made of German silver, by different makers, may be instructive : I II III Copper . 66-0 59-2 55'0 Nickel . 8-2 16-0 207 Zinc . 25'3 23-8 23-3 Iron 5 1-0 1-0 The above were three different qualities made by the same firm. No. I had been boiled white in a tinning solu- tion, and the tin was polished off before analysis. I II III IV V VI Copper 61-8 47-10 55-40 53-1 5575 f.3'3 Ziiic . 29-6 41-95 31-38 29-98 26-15 24-4 Nickel 8'3 10-95 11-64 16-25 18-10 21-0 Iron . 3 1-58 67 trace 1-3 The above were various qualities manufactured by another firm. No. VI was found to work very badly, probably due to iron, which is more injurious as the percentage of nickel is greater. The following table gives the composition of commercial varieties of German silver by the best makers, as used in Birmingham, exclusive of small proportions of lead and [TABLE GERMAN SILVER 311 XT A *i Percentage Composition. riame. Nickel. Copper. Zinc. Extra White Metal . 30 50 20 White Metal .... 24 54 22 Arguzoid 20 48^ 31 Best Best 21 50 29 Firsts or Best .... 16 56 28 Special Firsts .... 17 56 27 Seconds 14 62 24 Thirds 12 56 32 Special Thirds .... 11 56 32 Fourths 10 55 35 Fifths, for Plated Goods . 7 57 36 The following table gives the composition by analyses of German silver from various sources : Authority. Percentage Composition. Copper. NickeL CobaltJ Zinc. Iron. Lead. 1. Fyle . 40'4 31-6 25-4 2-6 2. Fricke 50 31-2 18-8 ... 3. Guettier . 53-3 26-6 ... 20-1 4. . 51-6 26 ... 22-4 5. Krupp 51-6 25-8 22-6 ... ... 6. ,, . 48'5 24-3 ... 24-3 2-9 7. Hiorns 56-98 24-3 1872 ... 8. Guettier 59 22-2 18-5 0-3 9. . 55-2 21-4 ... 23-4 ... ... 10. Hiorns 59-1 20-2 20-4 0-3 11. 56-5 20-3 23-2 ... 12. Henry 67 19-3 ... 13-6 13. Louyet 63-34 19-13 ... 17-41 ... 14. D'Arcet 50 18-75 31-25 ... 15. Hiorns 58 18-5 ... 23-5 ... ... 16. Smith 60 18-8 3-4 17-8 ... 17. Krupp 58-3 19-4 ... 19-4 2-9 312 MIXED METALS CHAP- ANALYSES OF GERMAN SILVER Continued Percentage Composition. Authority. Copper. Nickel. Cobalt. Zinc. Iron. Lead. 18. Hiorns 53-1 16-25 29-98 67 19. s 55-62 15-72 28-66 ... 20. 56-84 15-62 27-24 30 21. Louyet 62-4 15-05 22-15 trace ... 22. Krupp 57-8 14-3 27-1 '8 J23. Hiorns 58-72 13-85 26-43 1-00 ... 24. 57-0 13-4 27-60 2-00 25. Eisner 57'4 13'0 26-6 3-00 26. Hiorns 55-4 11-64 31-38 1-58 27. Louyet 28. Rochet 62-63 59-1 10-85 9-7 26-52 31-2 ... 29. Hiorns 66 8-2 25-3 '50 30. Krupp 63 6 31 The alloys in the preceding table containing lead are used for casting only, that metal making the alloy more fusible. MANUFACTURE OF GERMAN SILVER 93. The manufacture of German silver was formerly conducted in two distinct ways, known respectively as the German and English methods. German Process. In the German method the zinc and nickel to be used for a certain quantity of copper are divided into three equal portions. On the bottom of a graphite crucible, capable of holding 22 Ibs. of the alloy, is placed a layer of copper, and upon this a layer of zinc and nickel ; upon this another layer of copper is placed, and so on until all the copper is in the crucible. One-third each of zinc and nickel is retained for future addition. The contents of the crucible are then covered with charcoal powder, and the metals melted in an ordinary casting-furnace. When the contents are supposed to be liquefied, an iron rod is inserted, and if the whole is thoroughly fused, it is then v GERMAN SILVER 313 vigorously stirred. The remaining zinc and nickel are then added in portions at a time, and the whole well stirred after each addition, a brisk fire being maintained to prevent chilling of the alloy by the freshly added metals. After the introduction of the last portion, an additional piece of zinc is thrown into the crucible to compensate for loss of zinc by volatilisation. If the alloy is intended for rolling, it is recommended to keep the finished alloy liquid for some time longer before casting, keeping the surface well covered with charcoal in the meantime. English Process. The English differs from the German method chiefly in the manner in which the metals are melted together, no portion of the zinc and nickel being retained, but the entire quantity of metals, consisting of 8^ Ibs. copper, ^ Ib. zinc, and 2 to 3 Ibs. nickel, is melted at one time. The mixed metals are placed in a red-hot crucible and quickly covered with a thick layer of coal-dust. The furnace is urged to its highest pitch, so as to fuse the metals as quickly as possible. After ascertaining by an iron rod that the mass is liquefied, a previously prepared alloy, of 1 part by weight of zinc and ^ part by weight of copper, is added, the quantity for the above charge varying between 1^ and 2 Ibs. When this alloy is melted, and the entire contents of the crucible form a homogeneous whole, 2 Ibs. of zinc are finally added. The mass, being kept covered with coal-dust, is then heated as strongly as possible, and when thinly fluid, a sample is taken to test its qualities. The metals during melting always absorb oxygen, which is liable to produce blowholes when the alloy is cast. The nature of the alloy may be ascertained by taking a small quantity as a test, and if this exhibits unsoundness, some makers push a fire-clay pipe to the bottom of the crucible, and drop some pitch through it to the bottom of the metal, and then remove the tube. The pitch is decomposed, giving rise to reducing gases, which combine with the oxygen in the metal, and by vigorous stirring, before pouring, are removed. 314 MIXED METALS CHAP. It need hardly be stated that the distinctive methods of working, just described as the German and English methods, are not generally carried out at the present time a modifica- tion being found more rapid, economical, and attended with equally good results. The separate metals entering into the composition of German silver are not generally used in the free state, but are previously made into binary alloys. Thus, nickel is alloyed with copper by some manufacturers in the proportion of 2 parts by weight of copper to 1 part of nickel, while others use equal weights of these metals. Zinc is used in the form of brass common mixtures consisting of equal weights of copper and zinc, and 2 parts zinc to 1 of copper. The crystalline nature of German silver when cast in iron moulds renders the metal somewhat difficult of mechanical manipulation, especially when the amount of nickel present is high ; but this may be overcome by careful annealing at a moderate temperature, and by judicious hammering. Of course there are certain proportions which give better results than others, as shown by the author's experiments on the subject already enumerated. The crystalline structure gradu- ally disappears as the plates are reduced by rolling, with occasional annealing as the process proceeds ; and when the strips have been rolled sufficiently thin for ordinary stamp- ing purposes, the cause of brittleness has been practically eliminated, so that the alloy can be worked into any desired form. It has been found in practice that certain varieties of German silver become more homogeneous by re-melting, and can then be worked with greater facility ; but this is un- necessary if suitable proportions are selected at the commence- ment, and copper-nickel alloys with copper-zinc alloys used for melting together to form the desired alloy, instead of using the separate metals. It must also be remembered that each time German silver is remelted in a crucible in the ordinary way, a certain amount of oxide is formed, and a greater portion of the zinc than the copper or nickel is volatilised, so v GERMAN SILVER 315 that the relative amounts of the constituents are altered. In the case of re-melting it is necessary to add a portion of metallic zinc to compensate for the loss, and it is advisable to add this zinc after the fusion of the alloy has been effected. There is reason to believe that the zinc thus added assists in purifying the metal, by uniting with the absorbed oxygen. Whether this is so or not, the author has proved that such an addition is beneficial in many cases, if not in all. Great care is required in casting the alloy to avoid chill- ing of the metal, and as a high temperature is requisite to keep the metal sufficiently liquid for pouring into the moulds, the casting shop is kept closed and all spaces which ordinarily admit cold air are stopped up. All the pre- cautions with regard to blacking and heating of the moulds, etc., as described when speaking of brass strip-casting and moulding, apply also to German silver. But as German silver is more easily chilled than brass, it may be necessary, after one mould is filled, to re-heat the remaining portion of metal in the crucible before pouring it into a second mould. The moulds used for German silver strip-casting are of the same shape as those described for brass-strips, but differ in size. The running sizes of plate-ingots are 16 to 18 inches long, 4, 4^, and 5 inches wide, and 1 to 1 J inches thick. The wire -ingots are about 4 feet 6 inches to 5 feet long, 3j inches wide, and l inches thick. GERMAN SILVER SOLDERS. 94. Hard solders, employed for joining the parts of German silver articles together, are generally made of the same metals as those which compose the alloy to be soldered, but in such proportions as to have a lower melting point. In general the soldering is more perfect the nearer the fusing point of the solder approaches that of the metal of which the article is composed, but the greater is the care required to avoid melting it. In some cases silver solder is used for and German silver solder is also used 316 MIXED METALS CHAP. for soldering articles of iron and steel, on account of its high melting point and great tenacity. German silver solder is known by different names, such as "argentan solder," "arguzoid," etc. It is rendered moderately fusible by the addition of a large proportion of zinc to the copper and nickel. The mode of manufacture is similar to that already described for brass solder, and the proportions of the ingredients will depend upon the com- position of the alloy which it is required to solder. For the higher alloys, i.e. those rich in nickel, a more refractory solder is advisable than with the cheaper and more fusible alloys. In making argentan solder the copper and nickel should be melted first, and the zinc added in the free state, or in the form of brass, containing much zinc. The mixture is then cast in thin plates, which are broken into pieces while hot and crushed to powder in an iron mortar. The facility with which it can be pulverised will depend upon the proportion of zinc. If it is too brittle it indicates too much zinc, and even if somewhat malleable, too much zinc may be present ; in either case the defect may be remedied by adding the fresh metal required and re-melting. Excess of zinc may be removed by re-melting, when some of that metal will burn off, but this is of course wasteful and expensive. The right composition can always be determined by taking a small quantity and testing it before pouring the main portion. The colour is greyish-white, with a strong lustre. In order to test the best proportions for soldering articles containing 16 to 22 per cent of nickel, the author obtained a sample of the solder used by a large manufacturer for the above alloys, which, on analysis, gave the following pro- portions : Copper . . . . 47-10 Nickel . . . . 10-95 Zinc . . . . 41-95 100 GERMAN SILVER SOLDERS 317 The following trial samples were then prepared : I II III Copper 45 38 35 Nickel . 10 12 8 Zinc . . . | 45 50 57 100 100 100 The three samples were pulverised whilst hot, and tested by a workman accustomed to use the solder of which the analysis is given above. He considered No. II the best for soldering, but preferred the one he had been accustomed to use. No. Ill was pronounced porous, from which it was inferred that 57 per cent zinc was too much. 94A. Alloys for Coinage. Pure nickel is used as minor coinage in several countries. Austria - Hungary use an alloy of nickel 97 -4, cobalt 1*3, copper '32, iron '80 and some carbon. Too much iron and cobalt makes the metal too brittle. Excess of cobalt darkens the colour. In Servia they have used an alloy of 25 per cent nickel and 75 per cent copper, but it does not give so good or so clear an im- pression as that of pure nickel. It does not work any easier than nickel, and soon wears smooth. 94B. Oupro-Nickel. In the manufacture of cupro-nickel it is important to use nickel as pure as possible, since the im- purities present, such as silicon, carbon, iron and sulphur greatly diminish the malleability and ductility. There are no chemical compounds of copper and nickel, neither is there any eutectic mixture. The two metals form solid solutions or mixed crystals in all proportions. The freezing point is practically a continuous line. All the alloys with excess of nickel are magnetic, and those with excess of copper are non-magnetic. The uses of different alloys are given in the following table : 318 MIXED METALS CHAP. Composition. Copper 98, Nickel 2 95, ,, 90, 85, - 80, 75, 70, 60, 52, Copper 28, Nickel 70,) and Iron 2 per cent. / Uses. Tubes ; locomotive boilers. Projectile bands. Electric resistances. Rifle and pistol ammunition ; solid drawn tubes ; plumbers' fittings to resist corrosion. Coinage. 'Electric resistance in the form of tape or wire. In fact, the whole series from 95 copper to 52 copper are used for electric resistances, f A new alloy produced in America, called " Monel metal." It is hard and white in colour. It is a natural product of V. certain nickel- copper ores. The mechanical properties of copper-nickel alloys as used in this country are given in the following table : Copper. Nickel. Material. Tenacity in Tons per square inch. Elongation per cent. 98 98 2 2 Tubes, soft hard 17-5 28-3 44 5-5 95 95 5 5 Rolled, soft hard 17-5 30 50 4 80 80 20 20 Rolled, soft , , hard 21 40 35 4 80 80 20 + Iron 20 + Carbon Rolled, soft ,, hard 25 24 32 16 75 25 Rolled, soft 27 31 The magnetic alloys of cup ro- nickel cease to be magnetic at certain transition temperatures. Nickel itself has a change point at 320 C., above which it ceases to possess magnetic permeability. The presence of copper lowers this point, as shown in the following table by Guertler and Tamman : CUPKO-NICKEL 319 Nickel per cent. Loss of Magnetism on Heating. Return of Magnetism on Cooling. Average. 89-1 78-6 68-2 57-9 315 C. 215 125 Not appreciable 280 C. 190 105 . 295 C. 205 115 Alloys with 100 to 25 per cent of nickel have the colour of nickel, while alloys with 20 per cent and less of nickel are coloured by copper. The alloys up to 50 per cent nickel show magnetic properties, and those with less than 50 per cent do not exhibit magnetic properties. The magnetic alloy may be termed alpha and the non-magnetic beta. Cupro- nickel has a great tendency to develop blowholes on casting, due to occluded gases, so that the addition of a deoxidiser, such as aluminium or magnesium, is necessary. CHAPTER VI ALLOYS OF TIN 95. Tin and Zinc. Alloys of these inetals can be readily produced by fusion, forming combinations that are generally harder and less malleable than tin, softer than zinc, and more or less crystalline in structure. The colour of the fractured surface depends upon the nature of the mould and the temperature of the alloy at the time of cast- ing. The same observations also apply to the shrinkage upon solidification. Tin-zinc alloys are chiefly employed for casting ornamental objects and patterns. The following investigation into the character of tin-zinc alloys was made by Guettier l : " 1. Tin 30, Zinc 70. Texture of a dull white colour ; an average- shrinkage ; breaks easily ; on fracture shows larger and brighter facets than zinc ; the metal is denser at the bottom of the mould ; dry to the file ; breaks under the chipping chisel ; slightly sonorous ; and shows an appearance of crystallisation at the surface, with a slight bluish-yellow colour. " 2. Tin 25, Zinc 75. Texture of a white colour, inclin- ing to blue ; slight shrinkage of the bar ; bright fracture with large bluish facets like those of zinc ; the tin seems to be in larger proportion at the bottom of the button, the same as in No. 1 ; the surface is covered with a kind of skin, 1 Guettier, Guide Pratique des Attiages, 1865. 320 CHAP, vi TIN-ZINC ALLOYS 321 rather wrinkled than crystalline, and with variegated colours, light blue, violet, and golden yellow. " 3. Tin 50, Zinc 50. Texture, pallid white ; surface of the ingot is very smooth, granular, and lamellar, without the appearance of shrinkage ; the edges are somewhat round, and do not show plainly the iridescent colours ; the fracture is bright and finely granular, upon a tin- white ground ; and clogs the file a little ; the alloy is well mixed, tough, and malleable, without being soft. " 4. Tin 70, Zinc 30. Texture white ; and somewhat shining ; exhibits no settling ; feebly sonorous ; surface granular, and dead white with slightly yellow spots ; difficult to break ; bears hammering well ; easily worked with the chisel, which takes off long chips ; it clogs the file ; the fracture, like that of tin, is without brightness and crystallisation ; when polished it is less bright than tin ; the alloy is more complete and uniform in texture than the preceding ones. "5. Tin 75, Zinc 25. Texture tin-white, but without brightness ; exhibits no settling ; surface granular and studded with bright particles ; the upper surface has a tint changing from yellow to reddish-blue ; clogs the file more than No. 4 ; very malleable, although resisting the hammer and chisel more than No. 4 ; bends without the crackling sound of tin. " 6. Tin 10, Zinc 90. The bar shows at the fracture the characteristics of a zinc bar ; clogs the file more than zinc ; the fracture is not of so dull a grey ; the bottom of the button is soft and easily receives the impression of a punch ; like No. 2, tin appears to settle, and the metal at the bottom is softer than pure tin. "7. Tin 90, Zinc 10. The bar presents the jagged fracture of tin, and the runner could only be separated by cutting it ; the alloy clogs the file less than pure tin ; the button settles sensibly in the middle, although the edges are sharp ; the alloy is very malleable, although not very soft under the hammer. "8, Tin 1, Zinc 99. The fracture is like that of zinc, Y 322 MIXED METALS CHAP. but the facets are less ; lustre is slightly brighter after filing ; middle of the bar had settled ; the button also settles in the middle, and the lower part was soft like No. 6, although not so thick, on account of the small proportion of tin in the alloy ; the soft portions are bluish, like lead, and are easily streaked by the nail. " 9. Tin 99, Zinc 1. The fracture is slightly granular, and not so dull and jagged as that of tin ; when polished is not so bright as tin ; there is more shrinkage on the ingot than on the button, and the surface of the latter presents dark iridescent colours. " General Observations. The alloys, where the pro- portion of zinc is the greatest, present in their fracture a crystallisation whose facets shine like graphite. Very small proportions of tin added to zinc cause this crystallisation. In similar circumstances the exterior of the castings is covered with a yellowish-white moire. " In thick castings, where zinc predominates, there is a tendency for the metals to separate at the bottom of the mould ; and, what is remarkable, this tendency grows greater as the proportion of tin becomes smaller, which is exemplified by the separation being more sensible in No. 8 than in No. 6. We may add as a singular anomaly that the tin which has passed through the zinc, and has become precipitated, loses its distinctive qualities and acquires the softness and bluish dull colour of lead. " The colour of the alloy of zinc and tin, whether simply cast, or after being filed, becomes brighter in a direct ratio with the proportion of tin contained in it. " The alloys already rich in tin become granular when the proportion of zinc is increased. The alloy No. 3 tin 50, zinc 50 has the fracture of iron, but its colour is duller. " The alloy No. 9 tin 99, zinc 1 has a fracture present- ing no longer the jagged appearance of tin, and is dull-grey and finely granular in structure. " The specific gravity of the alloys of tin and zinc is in proportion to the mean specific gravity of the two metals ; vi TIN-ZINC ALLOYS 323 therefore the alloys where tin predominates are more dense. u The waste is greater where zinc is in excess ; the tin having been put into the crucible after the fusion of the zinc, we infer that most of the waste comes from the zinc. " The addition of 1 per cent of tin to zinc is sufficient to impart to the latter metal a greater resistance without diminishing its hardness. 1 per cent of zinc added to tin impairs the flexibility of the latter, and, what is remarkable, prevents its peculiar crackling noise. These two alloys, ' when the combination is intimate, present no other sensible changes. " The alloy of tin 50, zinc 50, is the best as regards stiff- ness and economy. More zinc would produce an alloy not so well mixed, more crystalline, and brittle ; more tin would give a metal too soft and clogging the file. However, for thin and resisting castings an alloy of tin 70, zinc 30, is well adapted. The alloys kept between these figures and the proportion of half and half are very resisting and tenacious. Their malleability increases with the proportion of tin. " The alloy of zinc 1, tin 99, without impairing the malle- ability of the latter metal, increases its hardness and tenacity for castings. " The alloys where the maximum of zinc is employed arc useful in foundries only for thick pieces ; they are then very economical. Up to the portions of tin 30, zinc 70, they re- main nearly as brittle as zinc itself. The proportion of tin 25, zinc 75, produces an alloy not so flexible as and less brittle than zinc, which could be adopted for foundry patterns. " The alloys Nos. 6 and 8 appear to us more brittle than zinc. In those experiments tin passing through the molten mass into the mould had become precipitated to the bottom. We may infer from this that a quantity of tin, sensibly less than 1 per cent, is sufficient to change the nature of zinc. " In the proportions of tin 40, zinc 60, the alloy possesses but little malleability." 324 MIXED METALS CHAP. An alloy of 11 parts tin to 1 part zinc, beaten out into leaf, forms spurious silver-leaf. According to Rudberg, ZnSn 6 (1 part zinc to lOf parts tin) solidifies completely at 204 C., but all the other alloys separate on cooling from a state of fusion into two portions, the one consisting of ZnSn 6 , not solidifying till cooled to 204 C., while the remainder, consisting of an alloy containing a larger proportion of one or the other metal, solidifies at a higher temperature : thus : Sn l2 Zn Sn 6 Zn Sn 4 Zn Sn 3 Zn Sn 2 Zn SnZn Variable point Fixed point . 210 204 204 230 204 250 204 280 204 320 204 A metal, now largely employed for buttons, is prepared by rolling a thin sheet of tin on each side of a thick sheet of zinc. The metals become firmly welded together, and the pressed work has the advantage of a coating of tin. 96. Tin and Lead. These metals are easily melted, and unite together in all proportions, forming a series of valuable alloys. Lead leaves a dark mark when drawn across paper, and when only a limited quantity of tin is alloyed with it, this property is still retained ; but if a certain limit be exceeded, the alloy no longer has the property of producing the dark mark ; hence it is possible to roughly estimate the quantity of the lead present by this test, taken in conjunction with the behaviour of the metal under the hammer, file, and chisel. Tin 90 parts, and lead 10 parts, does not streak paper. Tin 75, lead 25, gives a very faint mark. Between these limits no mark can be observed when the alloys are drawn across the paper. All lead-tin alloys containing less than 75 per cent of tin have the power of marking paper. Alloys of lead and tin shrink, or settle less on cooling than either of the metals taken singly ; they are not so fluid when melted, and the castings have not the same sharpness. VI TIN-LEAD ALLOYS 325 The effect of lead on tin is to increase its malleability and ductility, but to dimmish its tenacity and toughness. In the alloy, tin 90, lead 10, tin preserves the crackling noise, but in a less degree than in pure tin. On the contrary, 1 per cent of zinc in tin is sufficient to destroy the crackling noise when the metal is bent The following table contains the results of Kuffer's experiments with respect to the specific gravities and melting points of lead-tin alloys. The author has added the per- centage compositions : Composition of the alloys. Specific gravity. Per cent. Difference. Melting point. Formula. Calcu- lated. Found. Pb. Sn. Pb; 100 11-3803 335 C. Sn . 100 7-2911 ... 230 SnPb* 63-7 36-3 9-4366 9-4263 0-0103 241 ,, SnPbs 77-82 22-18 10-0936 10-0782 0-0154 SnPb 3 84-04 15-96 10-4122 10-3868 0-0254 239',, SnPb 4 87-42 12-58 10-6002 10-5551 0-0431 ... Sn 2 Pb 4673 53-27 8-7518 87454 0-0064 196 ,, Sn 3 Pb 36-90 63-10 8-3983 8-2914 0-0069 il86,, Sn 4 Pb 30-49 S 69-51 8-1516 8-1730 0-0096 i 189 ,, SnsPb 25-85 74-15 8-0372 8-0279 0-0093 194 Sn 6 Pb 17-04 82-96 7-9526 7-9210 0-0116 * Sn = 118. Pb = 207. " Alloys of lead and tin are distinguished by the facility with which they ignite and burn. The alloy of 4 or 5 parts lead and 1 part tin burns like charcoal at a red heat, the combustion continuing like that of an inferior peat, with the formation of cauliflower excrescences. The action appears to be due to the affinity which exists between the two oxides." l 97. Pewter is essentially an alloy of lead and tin, to which small quantities of other metals are sometimes added. 1 Watt's Diet, of Chem. p. 534. 326 MIXED METALS CHAP. Common pewter consists of tin 80 parts and lead 20 parts. Holtzapfel states that some pewters are made nearly as com- mon as that of equal parts of the metals ; when cast they are black, shining, and soft ; when turned they are dark and bluish. Other pewters only contain ^ or of lead ; these when cast are white, without gloss, and hard. Such are pronounced very good metals, and but little darker than tin. The French legislature sanctioned the employment of 1 8 per cent lead with 82 per cent tin, as quite harmless in vessels for wine and vinegar. The finest pewter, called tin and temper, consists chiefly of tin, with only a little lead and copper, which make it hard and somewhat sonorous, but the metal becomes brown in colour when the copper exceeds a certain quantity. The copper with twice its weight of tin is melted, and from ^ Ib. to 7 Ibs. of this alloy termed temper are added to a block of tin, weighing from 360 to 390 Ibs. Zinc in small quantity is added to the molten alloy, and the mixture well stirred. The operator considers that the zinc removes impurities, bringing them to the surface as dross, and also that the burning of the zinc during casting lessens the oxidation of the pewter. Unalloyed tin is now being largely used in place of pewter, and it is not only whiter in colour, but for domestic purposes it is certainly safer, on account of the poisonous nature of lead compounds, {although, ^as before stated, lead when thoroughly alloyed with tin may be present to the extent of 1 8 per cent without injury. An alloy of 3 parts lead and 5 parts tin is used for tinning certain articles of copper. Alloys of lead and tin have a bright lustrous appearance, and are used for the manufacture of stage jewellery. The so-called Fahlum brilliants are made of an alloy of 3 9 '6 parts lead and 60*4 parts tin. The molten alloy is poured into moulds faceted in the same manner as diamonds. Toys, such as tin-soldiers, and many other articles, are made of alloys of lead and tin. For tempering various articles of steel, where it is im- portant to have a definite and uniform temperature, Messrs. Parkes and Martin have proposed the following alloys : VI TIN-LEAD ALLOYS 327 Alloy. Temperature of melting. Articles. Lead. Tin. C. F. 14 8 213-4 415-4 Lancets. 15 8 221 429-8 Other surgical instruments. 16 8 228 442-4 Razors. 17 8 240 464 Penknives, cold chisels. 28 8 257 494-6 Shears, gardeners' tools. 36 8 262 503-6 Axes, planes, small scissors. 60 8 275 527 Table knives, large scissors. 96 8 284 543-2 Swords, watch-springs. 100 8 289 552-2 Strong springs, augers, saws, etc. The same authorities have determined the melting points of lead-tin alloys ; their results are embodied in the following table : l Composition. Melting point. Composition. Melting point. Tin. Lead. F. C. Tin. Lead. F. C. 50 50 372 189 12-5 87-5 527 275 60 40 336 169 11-8 88'2 530 277 66-7 33-3 340 170 11-1 88-9 532 278 71-4 28-6 348 175-5 10-5 89-5 535 279-5 75 25 354-2 179 10 90 538 281 77-8 22-2 362 183 9-5 90-5 540 282 80 20 367 186 9 91 542 283 81-8 18-2 372 189 87 91-3 544 284-5 83-3 16-7 378 192 8-3 91-7 546 285-5 84-6 15-4 380 193 8 92 548 286-5 857 14-3 382 194-4 7-7 92-3 550 287-7 40 60 412 211 7-4 92-6 551 288-3 33-3 667 442 228 7-1 92-9 552 289 28-6 71-4 470 243 6-7 93-3 554 290 25 75 482 250 6-4 93-6 555 290-5 22-9 77-1 490 254-4 6-3 93-7 556 291 20 80 498 259 6-2 93-8 557 2917 18-2 81-8 505 263 6 94 557 291-7 16-7 83-3 512 267 5-9 94-1 557 291-7 15-4 84-6 517 269-5 57 94-3 557 291-7 14-3 85-7 519 270-5 5-5 94-5 557 291-7 13-3 86-7 523 273 5'4 94-6 558 292 1 The melting points in these tables are much too high, according to more recent determinations. See next page. 328 MIXED METALS CHAP. The most fusible alloy of lead and tin consists of very nearly 2 parts tin to 1 part lead, or 3 -3 atoms of tin to 1 atom of lead. The melting points of this and other alloys are graphically represented by the curve, Fig. 1, p. 50. This alloy of minimum fusion is the eutectic alloy, and has only one melting point. All other proportions have more than one freezing point. For generations tinmen have used 2 parts tin and 1 part lead for making the most fusible solder. They do not trouble to weigh the metals, but add tin to the lead in a ladle and pour a little of the alloy on to a slab, and repeat until the metal sets with a perfectly bright surface. If a white spot appears, then more tin or lead is required. If too much of either metal is added the white spot increases. In cooling, one part of the alloy crystallises after another. The eutectic behaves like either of its con- stituents, and having but one freezing point, the surfaces of the small cakes remain bright. The moment the proportion of either constituent is increased, it begins to form a compound which crystallises before the eutectic alloy. SOFT SOLDERS 98. Soft solders usually consist of lead and tin in various proportions, and to a certain extent the fusibility of the alloy increases with the content of tin ; but? this does not apply when the tin exceeds 67 per cent. When a more fusible solder is required, the metal bismuth is added in addition, and sometimes cadmium. Metallic tin is some- times used alone, as in soldering fine utensils of tin plate. Lead is also soldered to lead by simply melting the edges by means of a blowpipe flame, as in the case of lead sheets for sulphuric acid chambers. This is termed autogenous soldering. Soft solders are termed common, medium, or best, accord- ing to the amount of tin, those containing most lead being the cheapest Fine or best solder is largely used for Britannia metal, best tin-plate, brass, and other metal SOFT SOLDERS 329 articles. The commoner varieties are used by plumbers. An alloy of 1 part tin to 2 parts lead is termed " plumbers' sealed solder," and stamped by the " Plumbers' Company." This alloy undergoes a prolonged pasty stage on cooling, due to the fact that it has two widely separated points of solidification, the mass consisting of granules of solid lead, in a fluid mother liquor, which is the eutectic mixture. It is on this property that the plumber depends in wiping a joint. To do this with the eutectic alloy referred to in the previous section would be manifestly impossible. The nearer the composition approximates to this alloy of minimum fusion, the less pasty is the mass when approaching the freezing point. The following table gives the proportions employed for different kinds of soft solder : Tin.' Lead. Tin. Lead. 1 10 H 1 1 5 2 1 1 3 3 1 1 2 4 1 1 1 5 1 The quality of the solder is roughly judged by the appearance of the surface when cast into a mould and allowed to cool. With excess of lead, the surface shows a uniformly greyish - white colour. With excess of tin, the surface is bright with dull greyish-white spots ; in fact, the appearance approximates to that of lead or tin according to the amount of lead or tin present. 99. Alloys of Tin, Lead, and Zinc. The presence of lead imparts more body and resistance to the alloys of tin and zinc than when tin and zinc alone are used, but the triple alloys clog the file as much as the latter. The frac- 330 MIXED METALS CHAP. tures are more decided than tin -zinc alloys. When the three metals are present in about equal proportions, the alloys are malleable, although not very ductile, and may be economically employed in some cases. 1 Various alloys are given in the following table : Tin. Zinc. Lead. 1 76 12 12 2 12 76 12 3 12 12 76 4 34 33 33 5 10 45 45 6 45 45 10 7 45 10 45 No. 2 is as hard and brittle as zinc, although more resisting. No. 3 resembles hard lead, leaves a mark on paper, appears to be uniform in composition, and has a leaden colour. Both these alloys are better adapted for castings than either of the three metals taken separately. Nos. 1, 3, and 7 appear to withstand friction very well. Nos. 2, 4, and 5 will do for work requiring more resistance than pure] zinc. No. 6 will answer for small castings requiring a certain malleability. It is serviceable for orna- ments, and will bear engraving and chasing. For these uses Nos. 2, 4, and 5 would be too brittle ; and Nos. 1, 2, and 7 too soft and yielding. All these alloys of tin, zinc, and lead have little lustre when polished, nd become readily tarnished by exposure to air. Some of them may be used as type metaL Most of them may be easily rolled. It is preferable to melt the zinc at the lowest possible temperature, to add the tin, and then the lead. The metals should be covered with charcoal, or with resin, and a little borax added, in order to prevent oxidation. For white 1 Guettier, Guide Pratique des Alliages, 1865. vi TIN-ANTIMONY ALLOYS 331 alloys, the best proportions are within the following limits : Tin 67 73 Zinc 16| 13 Lead 16f 13f The proportion of zinc is increased if toughness and hardness are desired. More tin increases the malleability, whiteness, and lustre. But the proportion of lead should not much exceed the amount indicated above. 100. Alloys of Tin and Antimony. These metals when united together form the base of what is termed Britannia metal, many varieties of which consist of tin hardened by antimony. Such alloys are as white as tin, but harder and less malleable. The brittleness increases as the proportion of antimony is greater. Guettier states that the specific gravity of tin-antimony alloys is below that which would be calculated from the specific gravity of each -metal taken singly. This indicates that expansion takes place by the union of these metals. An alloy of 80 parts tin and 20 of antimony is sufficiently malleable to be hammered and rolled in the cold. It is by keeping near these proportions that the best alloys of tin and antimony, for making pots and engravers' plates, are obtained. A white metal dessert- spoon analysed by the author was found to contain Tin 79 Antimony 20 Copper "6 Iron . . '4 100-0 The following table will show the proportions of different tin-antimony alloys : [TABLE 332 MIXED METALS CHAP. Tin. Antimony. Britannia metal 907 9'2 94 6 Algiers metal 90 10 75 25 For seats of stopcocks . For plugs of stopcocks . 86 80 14 20 Very hard alloy, and\ forms the extreme 1 limit of ordinary anti- f 67 33 (imony-tin alloys J Metal for bearings 79 21 According to Chaudet, 10 parts tin to 1 part antimony form a perfectly ductile alloy. Alloys of tin and antimony are made by fusing the two metals together. Also they may be made by reducing sulphide of antimony in contact with tin. In working with antimony alloys it is important to cast at a temperature as low as possible, and to pour into cool moulds, so as to aid rapid solidification. Then instead of getting the antimony separated from the eutectic alloy, it is fairly distributed through the mass. 101. Britannia Metal. Reference has just been made to some alloys called by this name, consisting of tin and antimony, but more generally other metals in small quantities are added to the mixture, such as copper, zinc, lead, bismuth, etc. Britannia metal has a white colour, with a bluish tint ; it takes a high polish, is hard, malleable, and ductile in proportion to the amount of tin and copper present. The latter metal, however, must always be limited in quantity, as it tends to impart a yellowish tint, and diminish the fusibility, for which reasons the quantity of copper used is always very small compared with the tin and antimony. Good alloys show a fine-grained jagged fracture. If the alloy exhibits a crystalline fracture, it either contains too vi BRITANNIA METAL 333 much antimony, or requires to be re-melted in order to promote more intimate union of the constituents. Iron and zinc appear to be very objectionable, as they considerably increase the hardness and brittleness. If much zinc is used to make a cheap alloy, the antimony must be in much smaller amount than is usual with better alloys. Lead is advantageous in cast work, making the alloy more fusible, but it impairs the colour and lustre, and the alloy tarnishes more readily in air than in alloys in which lead is absent. Arsenic, even in small quantity, induces brittleness, and should be avoided as much as possible. Alloys containing metals other than tin and antimony are less brilliant in lustre than when these two metals alone are employed. The following table shows the composition of a few varieties of Britannia metal : [TABLE 334 MIXED METALS CHAP. M 01 "3 81 ::;:::::::::: 1 S :;:: | "A 1 CO . . . 00 V s 5 V 05 J ,H '-' -Hw ^^ CO 5 OO 1 PQ : w : : : : : : ^ : ^ CQ : : : : CO 1 OO OS T*CO i COS i I (N O ' U5 CO I 1 I (H O00p ;)OS 10 8 3 8 167 > > 1 ... 2 3 203 > 1 ' ... 3 5 203 ' > 1 ... 1 2 203 Wood's alloy 1 2 2 4 1 2 4 5 150 160 Fusible alloy 2 2 4 187 Type-metal 22 50 36 ... Lipowitz's alloy is prepared by melting the constituents together, and well stirring with a stick of hard wood before pouring. This alloy has a silvery- white colour and lustre, and can be bent short, hammered, and turned. It is well adapted for castings of delicate objects, and for soldering Britannia metal and other articles which cannot be strongly heated, and are of a white colour. It begins to soften at 140 F. According to Hauer, the melting points of fusible alloys are proportionate to the atomic composition, thus : Melting point. 155-1 F. 153-5 150-0 , 190-4' 193-0 : 203-0 The above formulae correspond to the following per- centage compositions : Formula. 1. Cd Sn Pb Bi 2. Cd 3 Sn 4 Pb 4 Bi 4 3. Cd 4 Sn 5 Pb 5 Bi 5 4. Cd Pb 6 Ei 7 . 5. Cd Pb 3 Bij . 6. 344 MIXED METALS CHAP. I II III IV V VI Cadmium . 17-31 13-6 14-3 4 97 8-9 Tin . . 18-24 19 19 Lead . 32-00 33-4 33-1 44 54 57-7 Bismuth 32-45 34 33-6 52 36-3 33-4 105. Fusible Alloys containing Mercury. An alloy containing Lead 20 Bismuth 20 Mercury 60 is liquid at the ordinary temperature, and may be squeezed through chamois leather like mercury. A fusible alloy for casts is made by adding ^ its weight of mercury to the alloy already mentioned, as in Rose's alloy. The new compound is fusible at the tempera- ture of the human body. This alloy may be used for taking casts of certain portions of the human body after death, such as the ear. The animal substances are destroyed by a concentrated solution of caustic potash, and the metal remains. An alloy for fusible tea-spoons, etc., is composed of Bismuth Tin . Lead. Mercury 8 3 5 1 to 2 The following alloys are used for filling teeth : I II Cadmium . Mercury 25-99 74-01 2174 78-26 Any of the fusible alloys previously described may be vi FUSIBLE ALLOYS 345 used with the addition of a little mercury, and thus rendered still more fusible. Attoy for Anatomical Preparations. The following alloy melts at 169 F. and remains liquid at 140 F. : Bismuth . . . . 53-5 Lead 17*0 Tin 19-0 Mercury . . . . 10 '5 100-0 CHAPTER VII LEAD ALLOYS 106. Lead forms alloys with most metals, some of which have been already considered. It is a soft metal, readily fusible, and generally tends to impart these properties to those metals with which it is combined. Some of its alloys are considered to be definite chemical compounds, which are capable of dissolving in an excess of lead. An addition of any other metal to lead hardens it and impairs its malleability. Zinc and iron only alloy with lead in limited proportions, the constituents separating according to their specific gravities. The most important alloys of lead are : Pewter, soft solder, type-metal, and shot-metal. 107. Type-metal. An alloy for type-metal should be readily fusible, not show a great tendency to crystallise near the surface of the mould, sufficiently hard to prevent the crushing of the letters when printing, and capable of expanding on cooling so as to fill the moulds sharply. To fulfil these conditions lead has generally been adopted as the base for type-metal, other metals being added to harden it, and impart the properties enumerated above. Zinc has been tried, but it does not alloy well with lead. Antimony answers the purpose fairly well, but if present beyond a certain amount, the alloys become very crystalline, hard, and brittle. Lead-antimony alloys not exceeding 15 per cent of antimony have the important property of expand- 346 CHAP, vii TYPE-METAL 347 ing on cooling, which makes them very suitable for the manufacture of type. The alloy with 15 per cent of anti- mony is the most satisfactory as regards fluidity and expan- sion on cooling. It is more fusible than either of the con- stituent metals. However, this alloy of lead and antimony, notwithstanding the proper degree of hardness, has a vitreous structure, and imperfectly resists the action of the press, and of the scouring caustics. It was then tried to increase the resistance without losing the other qualities of the alloy. This result was obtained by the employment of tin or bismuth. The best proportion of tin appears to be from 6 to 8 per cent A greater amount causes waste by oxidation ; the alloy also becomes too brittle, with a great tendency for the tin and antimony to crystallise. Other metals have been added to lead and antimony for special purposes. Copper and iron in small quantity have been added to produce a hard resisting alloy for newspaper work. The following table gives the composition of several type alloys : [TABLE 348 MIXED METALS CHAP. ALLOYS USED FOR TYPE-METAL Lead. Anti- mony. Tin. Bis- muth. Copper. Zinc. Other metals. Printing types . 4 1 > * 5> Small types and stereotypes Small types and 7-5 9 64 9 16 2'5 1 8 2 4 12" 5 af 5 16 Arsenic 5 stereotypes Small types and 3 1 stereotypes Small types and 5 1 stereotypes Small types and stereotypes Plates for engrav- 10 2 5 5 i ing music, etc. Plates for engrav- ing music, etc. Plates for engrav- ing music, etc. Plates for engrav- ing music, etc. Linotype metal 64 60 83 2-5 8 2-5 12 7-5 12 37'5 5 ... 16 The manufacture of type from the alloy by stamping or pressing is only adopted in certain cases, the types being generally cast. The alloys, being well adapted for castings, are employed for certain kinds of ornamental work. An alloy for keys of flutes and similar parts of musical instruments consists of lead 2 parts, and antimony 1 part. 108. Lead and Arsenic. Arsenic unites readily with lead, forming alloys, which are hard and brittle if much arsenic be present. At the same time the fusibility of the lead is increased. A small quantity of arsenic in lead is vii SHOT-METAL 349 used to form shot-metal Shots are formed by letting drops of lead fall from a considerable height into water. It is found that the presence of arsenic enables the drops of metal to assume a spherical shape in thus falling. The lead is melted in a cast-iron pot with oxide of arsenic and charcoal. The latter reduces the oxide to the metallic state, and the two metals unite to form an alloy. In England 10 parts of arsenic are allowed for 1100 parts lead. In France 8 parts of arsenic per 1000 parts of lead are considered sufficient. The purer the lead employed the greater is the quantity of arsenic required. If the amount of arsenic is too small the shots will not be spherical. Some manufacturers add other metals in small quantities, such as antimony and copper. The molten alloy is poured into a perforated iron basin at the top of a tower and received in a vessel of water below. The perforations in the basins are regulated according to the size of the shot required, and each hole is three times its diameter distant from the next, otherwise two drops of lead in falling might coalesce. The water receiving the shot should be frequently changed, or it will get too hot. Some manufacturers prefer to have a layer of oil on the surface of the water, the shot retaining thereby its spherical shape better than when dropped directly into the water. To pre- vent the metal oxidising when removed from the water, a small quantity of (about '25 per cent) sodium sulphide is dissolved in the water which receives the shot, and the drops of metal are thus coated with a film of lead sulphide, of a dark-grey colour, which is not affected by the atmosphere. Instead of the method of dropping lead from the top of a tower, a plan for the manufacture of shot by centrifugal force has been introduced. The metal is poured in a thiu stream upon a rapidly revolving disc, surrounded by a screen against which the shot is thrown. The centrifugal action divides the metal into drops, the size of which is regulated by the rapidity with which the disc revolves. The drops are stopped by the surrounding screen. Mr. D. Smith of New York has devised a method of 350 MIXED METALS CHAP. making shot, in which the molten metal falls through an ascending current of air, travelling with considerable velocity, so that the descending metal conies in contact with as many particles of air in a short tower as it would by the usual method in falling through a high tower. 109. Lead and Iron. These metals only alloy together in small proportions, and do not form combinations of much use in the Arts. In some experiments made by Guettier at Angers in 1848, lead to the extent of 2 to 3 per cent was thoroughly mixed with molten cast-iron, but the lead was almost entirely oxidised or deposited at the bottom of the mould. The cast-iron thus treated was harder, and its grains were flatter and without lustre. As soon as lead is introduced into molten cast-iron a certain agitation appears at the surface, and even through the whole bath, and the cast-iron seems more fluid. Guettier states that when thin or large pieces are to be cast, some founders throw a quantity of lead into the molten iron to prevent the metal solidifying too soon against the sides of the casting-ladle. The want of affinity of iron for lead, and conversely, is made use of in separating lead from other metals having a greater affinity for iron. On the other hand, lead may be employed in separating iron from other metals, such as silver, for example. Thus, if lead is added in sufficient quantity to a fused alloy of iron and silver, it will combine with the silver, and the iron will swim on the surface of the bath. 1 | 110. Lead and Copper. These metals do not unite in all proportions, and combinations are more easily produced when the copper is in excess. When the lead is in excess, the metal largely separates from the copper, or becomes much oxidised if the temperature be too high. In preparing copper-lead alloys, the copper should be melted first under a layer of charcoal, and then the lead introduced. The mix- ture should then be vigorously stirred with an iron rod for some time before pouring into the moulds. If the alloy be 1 Guettier, Guide Pratique des Alliagcs, 1865. vii LEAD ALLOYS 351 re-melted, the union of the constituents becomes more inti- mate, the colour more uniform, the fracture more homo- geneous, and the alloy stronger. When copper containing a little silver and an excess of lead is slowly raised to the melting point of lead, the latter liquates out, carrying with it the silver and a little copper. The residual copper also retains a portion of lead. Lead is frequently added to certain copper alloys, such as brass, required for turning, as it renders them more fusible, and more easily worked with cutting or abrading tools. For those alloys, made on the large scale, into which it is desir- able to introduce lead, it is best to previously prepare an alloy of equal parts of copper and lead, and add a portion of this mixture to the desired alloy, instead of using free lead for admixture. An alloy of 4 parts lead to 1 part copper has been em- ployed for casting large movable types. 111. Lead and Manganese. "When a mixture of 892 parts of peroxide of manganese and 2789 parts of oxide of lead are heated with charcoal in a carbon lined crucible, a homogeneous, compact, and malleable alloy is obtained, which can be rolled into sheets having great lustre when polished." l 112. Lead and Bismuth. These metals unite in various proportions by simple fusion. The alloys are malle- able and ductile as long as the proportion of bismuth does not exceed that of lead ; they are also more tenacious than lead. According to Quettier, "the alloy of bismuth 2, and lead 3 parts is ten times harder than pure lead. An alloy of equal parts of bismuth and lead is harder and more malle- able than lead. The ductility and malleability diminish with an increase of bismuth." An alloy of bismuth 1, and lead 2 parts is very ductile and malleable : Berthier gives its fusing point as 166 C. 1 Watt's Diet, of Chem. vol. iii. p. 534. CHAPTEE VIII ALLOYS CONTAINING MERCURY AMALGAMS 113. Mercury or quicksilver readily unites with other metals, forming what are termed amalgams, some of which are liquid, while others are solid or semi-solid. As pre- viously stated, the solid amalgams appear for the most part to consist of metals united in atomic proportions ; and the liquid amalgams to contain a compound dissolved in excess of mercury. In all cases the bond of affinity between the constituents is somewhat feeble, for Joule has shown that many of those which contain equal numbers of atoms of the component metals may be partly decomposed by subjecting them to very great pressure, part of the mercury being forced out, and an amalgam containing a larger proportion of the other metals remaining behind. In many cases liquid amalgams after a time acquire a crystalline form, and the mercury in excess is separated. The crystallised body is the true amalgam, and is very probably a definite chemical com- pound. Mercury in small quantity is useful in alloys required to have low melting points, as in the case of fusible alloys used for stopping teeth. Amalgams of gold and silver were formerly much used for gilding and silvering, known as fire-gilding and silvering, because the mercury was volatilised by heat, leaving a deposit of gold or silver on the article. The process is still practised to a limited extent, when a very durable coating of gold is required. Mercury unites readily with lead, zinc, tin, bismuth, 352 CHAP, vm AMALGAMS 353 cadmium, copper, gold, silver, magnesium, potassium, and sodium. The following metals unite with mercury with difficulty : Iron, nickel, cobalt, manganese, and platinum in the compact state. 114. Lead -Amalgam. This substance may be formed by pouring molten lead into mercury, or by rubbing lead filings with mercury in a mortar by means of a pestle. The amalgam possesses a brilliant white colour, and remains liquid with as much as 33 per cent of lead, but it soils the fingers. The amalgam of equal parts can be crystallised, and a piece of lead plunged into this amalgam is found to be covered with crystals when it is withdrawn. The amalgam has a higher specific gravity than either of the metals, owing to the contraction they undergo in combining. The presence of *02 to "025 per cent of lead improves the mercury for use in barometers and thermometers, as the latter metal has then not so great a tendency to form globules on the surface of the glass. 115. Zinc- Amalgam. Mercury and zinc unite to form white brittle alloys, which become pasty when the mercury predominates. Zinc combines readily with mercury when the latter is near its boiling point, and still more readily when mercury is mixed with molten zinc ; 8 parts zinc to 1 part mercury is very brittle. 1 part zinc to 4 or 5 parts mercury form a brittle, pulverulent amalgam, sometimes used for rubbers of electric machines ; but the addition of tin is preferable. Singer recommends 2 parts zinc, 1 part tin, and 4 to 6 parts mercury for this purpose. Zinc plates, used in galvanic batteries, are generally coated with mercury by first cleaning the zinc plate in dilute sulphuric acid, and then rubbing in the mercury with a brush or rag. 116. Tin- Amalgam. Mercury and tin combine at the ordinary temperature, and in all proportions when heated, especially by adding mercury to molten tin. An amalgam of 10 parts mercury to 1 part tin is liquid like mercury, but 2 A 354 MIXED METALS CHAP. does not run so well. An alloy of equal parts is solid. Tin-amalgam has a tin-white colour, and, if the mercury is not in too great excess, is brittle and granular. The old method of silvering mirrors by placing the glass on a sheet of tin-foil covered with mercury is an instance of the applica- tion of tin-amalgam. Dentists' Amalgam. Tin-amalgam is used for filling teeth. 1 part of finely divided tin is rubbed together in a mortar with 4 parts of mercury. The excess of mercury is then removed by squeezing through a bag of chamois leather. A flexible mass is left which hardens in a few days. Another alloy is formed of 2 parts tin, 1 part cadmium and excess of mercury. The tin and cadmium are melted to- gether and mercury added. The whole is then poured into an iron mortar, and well stirred with a wooden pestle till it acquires a soft buttery consistence ; the excess of mercury is then squeezed off. The amalgam is soft and plastic when kneaded by the hand. See also fusible alloys, p. 337. An amalgam of tin, silver, gold, and mercury is used as a cement for teeth. 1 part gold and 3 parts silver are melted, and 1 part tin is added. The resulting alloy is pulverised when hot, and then well kneaded with an equal weight of mercury. Amalgam for Tinning. Small iron articles can be tinned by first pickling bright in sulphuric acid, and dipping in melted tin-amalgam, then blanching in dilute acid, drying, and polishing. 117. Bismuth- Amalgam. Mercury is capable of dis- solving a considerable quantity of bismuth without losing its liquid form, but a drop would be pear-shaped, and not globular like pure mercury. An amalgam of 4 parts mercury and 1 part bismuth may be used as a substitute for tin in tinning. An amalgam of 3 parts mercury, 1 part lead, and 1 part bismuth, is fluid, and has been used for adulterating mercury. Itjpasses through chamois leather like pure mercury, but the viii AMALGAMS 355 drops are pear-shaped, which serves to detect the adulteration. Lucas states that mercury containing only ^^ part of bis- muth becomes coated with a black film when agitated in air. 118. Cadmium- Amalgam. Cadimum readily combines with mercury in the cold, but much more perfectly when mercury is added to molten cadmium. When mercury is completely saturated with cadmium, the alloy consists of 78*26 parts mercury and 21 - 74 parts cadmium, agreeing with the formula Hg 2 Cd. Cadmium-amalgam is a tin-white, crystalline, brittle mass, which softens when moderately heated, and can be kneaded like wax. It is used by dentists for stopping teeth, but a better alloy containing tin has been previously described. Evans's Metallic Cement is made by dissolving cadmium- amalgam (26 cadmium and 74 mercury) in excess of mercury, and pressing out the excess of mercury. This amalgam becomes very plastic by kneading. 119. Copper -Amalgam. Amalgams of copper and mercury are somewhat extensively used in the Arts. An amalgam containing 3 parts copper and 7 parts mercury becomes quite hard by keeping for a day or two. It has the property of softening and acquiring the consistence and elasticity of clay by pounding or kneading, and recovering its hard crystalline character when left to stand for a few hours. It was formerly much used for stopping teeth. Its density is the same in the hard as in the soft state, so that it does not expand or contract on hardening, and therefore fills cavities air-tight. It may be used for sealing bottles, glass tubes, etc., and as a cement for metals. Copper-amalgam crystallises with great ease. It can be hammered or rolled, and admits of a good polish. It retains its lustre well in air, but is blackened by sulphur compounds, such as sulphuretted hydrogen. It becomes soft and flexible when placed in boiling water. Amalgamated copper plates are largely used for extracting gold and silver from certain ores of those metals. 356 MIXED METALS CHAP. Copper-amalgam may be formed (1) by immersing copper foil in a solution of mercuric nitrate. (2) By rubbing copper, which has been precipitated from its solutions by zinc, first with mercuric nitrate solution, then with mercury in a mortar. (3) Mercury placed in contact with the negative pole of a battery, and covered with a solution of copper sul- phate, into which the positive wire dips, becomes saturated with copper. The chemical formula given to this compound by Joule is CuHg, or 63 parts copper to 200 parts mercury. Ironier's Bronze consists of copper and tin with 1 per cent of mercury. An alloy of 86*4 parts copper and 13 '6 parts mercury is said to have a golden yellow colour, and to take a fine polish. 120. Gold -Amalgam. Mercury has a great solvent action on gold, and is capable of dissolving a considerable amount without losing its liquidity. The point of saturation, according to Guettier, is 2 parts of gold for 1 part of mercury, when it acquires a waxy consistence. Gold-amalgam may be produced at a very low temperature by exposing gold to the fumes of mercury. A piece of gold rubbed with mercury is immediately penetrated by it and becomes exceedingly brittle. Gold - amalgam dissolves in mercury, forming a liquid mass. When this solution is strained through chamois leather, mercury passes through, together with a small quantity of gold, and there remains a white amalgam of pasty consistence. The amalgam, saturated with gold, when produced hot and allowed slowly to cool, becomes crystalline. By dissolving 1 part of gold in 1000 parts mercury, pressing through leather, and treating the residue with dilute nitric acid at a moderate heat, a solid amalgam Au 8 Hg is obtained, which crystallises in four-sided prisms, retains its lustre in air, and does not melt, even when heated till the mercury volatilises. 1 The liquid amalgams obtained by squeezing through 1 T. H. Heury, Phil. Mag. (4) vol. ix. p. 468. viii AMALGAMS 357 chamois leather contain, at the ordinary temperature, *126 per cent of gold ; at C. the percentage is '110 ; at 100 C. -650 per cent. These amalgams therefore behave like aqueous solutions. The residues left after the action of nitric acid on solid or liquid gold amalgams are not homogeneous, which proves that there probably exist definite compounds of gold and mercury dissolved in excess of mercury. A mixture of gold and mercury heated to a temperature a little above the boiling point of mercury, till its weight became constant, left an amalgam containing 10*02 to 10*5 per cent of mercury, which corresponds to the formula Au Q Hg. The pasty amalgam of gold and mercury is the base of the various processes for gilding base metals by the old method, known "as fire-gilding. For gilding copper, brass, etc., 8 to 9 parts of mercury to 1 part gold are used. The amalgam is prepared by heating the gold, cut up small, in a plumbago crucible to a red heat, then adding one-eighth or one-ninth of its weight of the above-mentioned amalgam, previously heated to boiling. The contents of the crucible are then well stirred with an iron rod, and poured into a large vessel cooled on the outside by water. If the amalgam is allowed to cool in the crucible, or if kept for some time after being poured into the cold vessel, it crystallises. The crystalline amalgam may, however, be restored to working condition by heating in a crucible with excess of mercury. The articles to receive the deposit of gold are first cleansed by heating and dipping in dilute sulphuric acid, and some- times in nitric acid ; they are then dipped in a solution of nitrate of mercury, by which means they become coated with mercury they are afterwards pressed upon the pasty amalgam, a portion of which adheres to it. The mercury is then expelled by heat and the gold surface polished. A mixture of bees -wax, red -ochre, copper acetate, and alum is then applied for the removal of the last traces of mercury, and to impart a redder shade to the surface. The articles are then burnished, washed with dilute nitric acid, then with water, and finally dried. 358 MIXED METALS CHAP. A native amalgam of gold in small yellowish crystals is found in California having the formula Au 2 Hg g (gold 39'02 to 41-63 per cent, mercury 58*37 to 60'98 per cent). An amalgam of gold and silver is found in small white grains in New Granada containing 38'59 gold, 5 silver, 56'4 mercury per cent. 121. Silver -Amalgam. Silver and mercury unite slowly at ordinary temperatures, but much more quickly when at a red heat. The affinity of these metals for each other is nearly the same as that of mercury for gold, but with a greater .tendency towards crystallisation. The more finely divided the silver, the more rapidly does amalgamation take place. Silver precipitated from its solutions by metallic zinc, then washed and dried, is in a very favourable condi- tion for amalgamation. When such silver is thrown into 1 a hot crucible containing mercury, combination rapidly takes place. By pressing the product in a chamois leather bag, the free mercury runs through and a soft white amalgam is left. Silver-amalgam can be prepared by adding mercury to a solution of silver nitrate ; the amalgam is precipitated in a crystalline form called a silver tree, or Arbor Diance. t The silver solution should be somewhat acid, and not too concen- trated. Joule gives the formula Ag 2 Hg as the average composition of this amalgam. When the action is assisted by the aid of a battery, amalgams richer in silver are obtained. Silver-amalgam varies in character according to its com- position and mode of formation, being soft, granular, or crystalline. The soft amalgam obtained by uniting the two metals was formerly used for coating articles with silver in a manner similar to that described for gilding. A native compound, termed amalgam, and having the composition Ag 2 Hg 2 or Ag 2 Hg 3 , is found in a crystallised state, and sometimes massive. Another amalgam, termed arquerite, is found in Chili, having the formula Ag 2 Hg 6 , or 13*5 silver and 86-5 mercury per cent. vni AMALGAMS 359 122. Magnesium- Amalgam. l This amalgam is slowly formed by contact of mercury with pure magnesium in the cold, but quickly at the boiling point of mercury. In this amalgam the affinities of magnesium are exalted. An amalgam containing *5 per cent magnesium swells up instantly in' contact with air, and loses its lustre ; it decom- poses water readily. Magnesium -amalgam may also be prepared by covering sodium - amalgam with a solution of magnesium sulphate. 123. Sodium -Amalgam. Sodium combines rapidly with mercury at ordinary temperatures, the combination being attended with a hissing noise and vivid combustion. A piece of sodium forcibly thrown upon mercury is thrown out of the vessel with explosion, in consequence of the great heat produced. It can be prepared by melting sodium under petroleum, and introducing the mercury 'through a narrow glass tube. A silver-white solid amalgam is formed which must be kept under petroleum to prevent the oxidation of the sodium. Sodium-amalgam is also prepared by triturating the two metals together in a dry mortar, fitted with a cover, until the combustion ceases. With 30 parts mercury to 1 of sodium, it is tolerably hard under the file, and exhibits a crystalline laminar fracture. With 40 parts mercury to 1 of sodium the amalgam is still solid, but softer than the former. With 60 parts mercury to 1 of sodium it forms a stiff paste at 21 C. ; 100 parts mercury to 1 of sodium produces a viscous mass; and 128 parts mercury to 1 of sodium is liquid. Sodium-amalgam is used in the preparation of other amalgams. Metallic chlorides, such as those of silver and gold, for example, are decomposed by sodium-amalgam, and the reduced metal then unites with the mercury. Metals which do not readily unite directly with mercury may be amalgamated by the action of sodium -amalgam on certain solutions of their salts. Thus : iron-amalgam is obtained by 1 Chem. Soc. J. (2) vol. iv. p. 141. 360 MIXED METALS CHAP, vm immersing sodium-amalgam, containing 1 per cent of sodium, in a clear saturated solution of ferrous sulphate. Sodium-amalgam is used for the extraction of gold and silver from their ores instead of mercury. It is said to facilitate the amalgamation, and to prevent flouring of the mercury ; i.e. it prevents the formation of oxide, sulphide, arsenide, etc., which would form a coat on the mercury and prevent contact with the gold or silver. 124. Potassium -Amalgam. Potassium unites with mercury with great violence at ordinary temperatures, and forms amalgams similar in properties to those of sodium. 124A. Chromium- Amalgam. Fe"ree has obtained this amalgam by electrolysing a solution of chromium chloride. By distillation in vacuo at 300 an allotropic form of chromium is obtained, which fires spontaneously in air. CHAPTER IX GOLD ALLOYS 125. The properties which make gold so valuable in the Arts have been already mentioned in the introductory part of this work, and it has been shown that this metal is improved for manufacturing purposes by alloying with certain other metals, such as silver and copper, in small quantity. The great value of gold is the reason, in most cases, why the proportion of added metal should be small. Many metals not only reduce the value of gold, but impart qualities which destroy its high malleability and ductility, while others render the gold unfit for any useful purpose. With a few exceptions, it may be stated that the higher the value of a metal, the greater should be the proportion of this metal in the alloy. Taking into consideration its high intrinsic value, gold may be considered the most perfect of all the metals, but it is necessary to impart to it a greater degree of hardness than it alone possesses, for the manufacture of coins, medals, jewellery, etc. Silver and copper are the metals invariably used for this purpose, and, when added in small quantity, do not materially alter the malleability and other working properties of gold, but increase its fusibility. The above remarks are only true when the silver and copper employed for alloying are practically pure, as a small quantity of lead, arsenic, antimony, etc., in the copper would considerably reduce the malleability and ductility of the alloy. 361 362 MIXED METALS CHAP. When gold is pure, or as nearly pure as it can be obtained for commercial purposes, it is commonly expressed as fine gold or 24 carat fine, the pound or 1000 parts being divided into 24 equal parts. Thus, 22-carat gold signifies that in 24 parts there are 22 parts gold and 2 parts of other metals. 9-carat gold likewise contains -^ gold and ^ other metals. The metal or metals added to gold are technically termed the alloy. The following table 1 gives the relative value of the different carats, and the amount of alloy to be added, taking 24 as the unit of fine gold and the mint price of purchase as 85 shillings per ounce troy : Quality. s. d. Alloy to be added. 24 carats 450 23 4 1 6| 1 part 22 3 17 11 2 21 3 14 4 3 20 3 10 10 4 19 3 7 3i 5 , 18 ,, 339 6 , 17 3 2| 7 , 16 2 16 8 8 , 15 2 13 li 9 , 14 297 10 , 13 ,, 2 6 Oi 11 , 12 ,, 226 12 , 11 ,, 1 18 Hi 13 , 10 1 15 5 14 , 9 1 11 10* 15 , 8 184" 16 , 7 ,, 1 4 9)5 17 , 6 1 1 3 18 , 5 ,, 17 8i 19 , 4 14 2 20 , 3 ,, 10 7i 21 , 2 7 1 22 , 1 3 6 23 1 Streeter, Gold, p. 138. ix GOLD ALLOYS 363 The price of fine gold from refiners is a little higher than the mint price, by about Is. per ounce for a single ounce, and proportionately lower for larger quantities. From this it will be seen that the cost of material to the manufacturer cannot be calculated merely on the amount of gold the alloy contains, and allowance must also be made for the quantity of alloy added, which is a consideration when much silver is used. Take 18-carat gold, for example. The price given in table is 3 : 3 : 9, the refiner's price would be 3:4:3, and assuming that the six parts of alloy consist of half silver and half copper, its cost will be 9d. ; so that the 1 8-carat alloy costs 3 : 5s. per ounce before it is manufac- tured into articles. 126. Gold and Copper. These metals alloy well together in all proportions, and when the copper does not exceed 10 to 12 per cent the malleability is little altered that is to say, 21 -carat gold and upwards are practically as malleable as pure gold. Many of the alloys have a density less than the mean of that of the separate metals. The gold coinage of this country is made from 22-carat gold, or 916-66 parts gold to 8 3 '33 parts copper, the alloying metal being copper. This is termed standard gold. Standard gold is very largely used for the manufacture of wedding-rings. The makers formerly had to pay a duty of 17s. per ounce, one-sixth being remitted for loss in finishing. Wedding-rings of this quality must be Hall-marked, and this is required to be done when the rings are in an unfinished state, so that for every 6 ounces the maker paid for 5 ounces. Mourning-rings, watch-cases, etc., are also occasion- ally made of the above quality. The alloy of 20 parts gold and 4 parts copper is termed broion-gold, and is sometimes used by jewellers in decorative designs. The gold-copper alloys used for coinage are generally accepted as being homogeneous, but doubts were raised upon the truth of this some years since by Mr. Booth of the United States Mint. Much evidence was afforded on this 364 MIXED METALS CHAP. subject during the preparation of the Standard Gold Trial Plate, made in the Royal Mint in 1873. A mass of standard gold weighing 72 ounces was cast and rolled into a plate 37 inches long and 6'5 inches wide ; portions were cut from different parts of it and assayed. The greatest variation between any two assays was 1( )j) o> there being no evidence of concentration of the precious metal anywhere. Roberts- Austen l has conducted the following experiments to obtain evidence on the liquation of gold alloys. An ingot consisting of 984-7 parts gold and 15'3 parts silver was melted and cast into a spherical mould. From this sphere of gold a disc ^ of an inch thick was cut, which weighed 3 1 ounces. It was then rolled in two directions at right angles, and portions cut off for assaying. The result of 82 assays afforded no clear evidence of systematic rearrangement, for although there appeared to be an enrichment towards the upper part to the extent of TT y 8 oo0 , such small differences as existed in the assays made on metal taken from the same horizontal planes could not be regarded as being due to any definite redistribution of the metal. It may be taken as proved that gold of high standard does not, like silver, show any marked tendency to reject on solidification the baser metal with which it is associated. M. Peligot 2 has published the results of an investigation which point to the same conclusion. 127. Gold and Silver. These metals unite in all proportions, but do not generally appear to form true chemical combinations. Levol regards the union of these metals in equivalent proportions as being incapable of separating by gradual cooling. The alloys of gold and silver are harder, tougher, and more fusible than gold, and more sonorous and elastic than either metal taken singly. One-twentieth of silver is sufficient to modify the colour of gold, and is employed by jewellers to impart different shades of colour to 1 Annual Report of the Deputy-Master of the Mint, 1888. 2 Bull, de la Soc. d' Encouragement, torn. iv. 1889, p. 171. IX GOLD ALLOYS 365 gold. 27 to 30 per cent silver and 73 to 70 per cent gold form a green alloy. When the amount of silver exceeds 50 per cent the alloys are nearly white. The greenish-yellow cast of the sovereigns manufactured by the Sydney Mint, Australia, is due to the fact that the alloying metal used is silver, and not copper, as in the English sovereign. The Australian gold coins, however, are of the same standard in fineness, weight, and value as the English coins. Gold-silver alloys do not oxidise on exposure to air. 128. Gold, Silver, and Copper. These three metals are largely used by jewellers to form alloys more tough, malleable, and ductile than by using copper alone as the alloying metal. The alloy added to gold for manufacturing the old English guineas consisted of equal parts of copper and silver, which accounts for their yellow appearance. The guinea is of the same fineness as the sovereign, but differs in weight. It weighs 5 dwts. 9^ grains; and a sovereign 5 dwts. and a little over 3^ grains ; of which 4 dwts. 22^ grains, and 4 dwts. 17 grains respectively are fine gold. Guineas were not coined for circulation during the reign of Queen Victoria. Gee 1 gives in the following table the proportions of silver, copper, and gold used in jewellers' alloys : Carat Copper. Silver. Gold. 23 i * 23 22 1 1 22 20 2 2 20 18 3 3 18 15 6 3 15 13 8 3 13 12 84 3* 12 10 10 4 10 9 8 10* 10* 4* H 9 8 7 9 8 7 1 Goldsmiths' Handbook, pp. 41, 52. 366 MIXED METALS CHAP. Wigley l gives the following proportions for various gold alloys used in jewellery : Number. Carat. Gold. Silver. Copper. Character of Alloy. I 15 0-625 0-025 0-350 Red gold. II 15 0-625 0-375 ' III 15 0-625 0-375 ... Green gold. IV 18 0-750 0-108 0-142 For dry colouring. V 18 0750 0-129 0-121 > j VI ... 1-000 0-250 0-387 Zinc -063 \ For gold pens. VII 8 1-000 1-000 1-000 ,} T VIII 18 0-900 11 Oil 0-300 Blue gold. IX 12 0-600 0-600 White gold. X 20 1-000 0-075 0-125 For enamelling. XI 1-000 0-450 0-175 XII 1-000 0709 0-300 For transparent enamelling. With regard to 18 -carat gold, Gee states that, "if properly cast, it is malleable and tenacious. It is also exceedingly ductile. A hardness is imparted to this quality of gold which admirably adapts it to the manufacture of jewellery of the highest order. There is perhaps a difficulty in preparing 18 -carat gold not experienced in some other alloys." The general opinion is that the occasional want of cohesion is due to the copper employed, as by using a purer variety of copper the difficulty in working is diminished. The best alloy appears to be that given in the preceding table, for by increasing the amount of silver the colour of the alloy would suffer. 15 -carat gold is also largely used for articles required to be made of coloured gold, as it is technically termed. The colouring is effected by dissolving out the copper from the exterior by suitable solvents, and leaving the surface with a colour like that of pure gold. It can thus be made to look 1 Art of the Goldsmith and Jeweller, p. 44. ix GOLD ALLOYS 367 equal to fine gold. It is easy to work, and the 9 parts of alloy give to the articles the requisite strength and hardness necessary to resist wear, and to retain their shape when subjected to various uses. This alloy can be Hall-marked as a guarantee of its proper quality. Gee l cautions pur- chasers of 15-carat gold against an inferior quality of gold introduced into the trade, and called 15 carat, bearing a stamp similar to the Hall-mark ; however, this is not the Hall-mark, but the private mark of the manufacturer. 12^ to 13-carat gold is very extensively manufactured into all kinds of jewellery. This is the lowest quality that can be properly subjected to the colouring process, and retain a rich and uniform appearance without showing irregularities on the surface. The articles of the so-called 15-carat coloured gold, referred to above, were formerly made of this quality. 12-carat gold is known in the jewellery trade as the best of the bright golds that is, qualities which cannot be properly coloured, and therefore show the true colours of the alloys of which they are composed, and not a surface of superior metal, as is the case in coloured gold articles, which have a much richer appearance than bright gold, and con- sequently are in much greater demand. The 12-carat alloy, using the proportions given in the preceding table, is malleable and ductile, and tolerably soft, so that it possesses good working qualities. It may be Hall-marked as a guarantee of its purity. 10-carat gold is similar in physical properties to the 12- carat alloy, but has a different shade of colour, owing to the different proportions of the constituent metals. This quality is not Hall-marked. 9 -carat gold is used for manufacturing articles of almost every description of jewellery, and when up to standard fineness may be Hall-marked. The quality most extensively employed is somewhat below the standard, this being the extreme limit that will stand the test of nitric acid without 1 Goldsmiths' Handbook, p. 46. 368 MIXED METALS CHAP. exhibiting signs of corrosion. Gee l states that " 9-carat gold of the mixture given in the preceding table, p. 365, will stand more than ordinary treatment from the hands of the workman, and may be touched and removed from the annealing -pan while still red-hot, without injury to any subsequent manipulation of it ; it may also be quenched at any degree of heat in pickle and water, if any advantage is likely to accrue from it; but we strongly object to the continuous quenching of gold alloys at every subsequent process of annealing partly because, every time the metal is quenched in sulphuric acid pickle, a portion of the base metal in these low qualities is dissolved." 9-carat alloys are sometimes alloyed with zinc, or spelter, as it is generally termed in the trade, in small quantity ; but it must be very sparingly used, or the alloys will be hard, brittle, and difficult to work, and, moreover, more readily acted upon by acids. 8-carat gold and qualities inferior to this are harder, and require more careful working than the higher alloys. They are more liable to become brittle and to break, unless carefully annealed at the proper stages. 8-carat gold may be made to withstand the acid-test by using more silver than that given in the table (p. 365), but the alloy is paler to the eye. This quality works up well if proper judgment is exercised in the manipulation. 129. Gold, Silver, Copper, and Zinc. Mention has already been made of the fact that zinc is sometimes employed in gold alloys used by jewellers. It is generally used in the form of brass, termed composition, which varies in the pro- portions of its constituents with different makers, and may be typically represented as containing 2 parts copper to 1 part zinc. The effect of zinc on gold is to harden it and make it brittle. An alloy of 11 parts gold to 1 part zinc resembles pale yellow brass in colour, and does not tarnish in air. Gee gives 17 per cent of zinc in gold as thtj 1 Goldsmith' Handbook, p. 48. ix GOLD ALLOYS 369 maximum amount that can be safely worked. When silver, in ordinary jewellers' alloys, is partly replaced by composition, the alloy appears of a deeper colour, and may be made to resemble one containing a higher standard of gold, but it is more difficult to manipulate, and more liable to change colour, depending of course on the amount of composition, used. (See also gold solders, p. 384.) 130. Gold and Tin. These metals appear to mix in all proportions forming, for the most part, brittle alloys. Guettier states that when the tin does not exceed 8 per cent, the alloys have a certain amount of ductility. The colour is yellow, pale, or white, according to the quantity of tin present. Like the alloys of gold and zinc, the union of the two metals produces contraction that is, their specific gravities are in excess of the mean of their constituents. 131. Gold and Lead. These metals unite readily in all proportions, producing very brittle alloys, which are harder and more fusible than gold, and without any utility in the Arts. TrgVff part of lead melted with standard gold, and the alloy cast into a bar, can be broken with a slight tap with a hammer ; the colour is also altered to orange-brown, and experiments have shown that the tenacity of the metal is reduced from 18 to 5 tons per square inch. All the alloys of gold and lead expand on alloying, and this is greatest when copper is present and the quantity of lead is small. The greatest expansion, according to Guettier, takes place when the lead is only -001 of the alloy. An alloy of 1 1 parts gold and 1 part lead has the colour of gold and the fragility of glass. (See also p. 92.) 132. Gold and Bismuth. These metals alloy well together in various proportions forming bodies having the appearance of brass, when the gold is in excess. Bismuth is highly injurious to gold, making it hard and brittle. Bismuth gives a eutectic alloy of very low melting point, the pasty stage being maintained down to 350 C. A small quantity 2 B 370 MIXED METALS CHAP. will impart a lead-grey or almost purple colour to the frac- tured surface, due to the very distinct liquation that can be observed, the grains of nearly pure gold being surrounded by a brittle impure mass. 133. Gold and Antimony. Antimony has a strong affinity for gold, and dissolves it rapidly. Melted gold dissolves the vapour of antimony, and when this metal is present in gold in very minute quantity it renders the gold brittle. An alloy of 9 parts gold to 1 part antimony is white, brittle, and has a granular fracture. 2irb7F ^ ant ^" niony in gold hardens it, and considerably impairs its malleability. Antimony may be largely removed from gold by heat. 134. Gold and Arsenic. Arsenic, like antimony, readily unites with gold, and is somewhat injurious when present in minute quantities. The alloys are greyish-white when much arsenic is present, hard, more fusible than gold, and very brittle. 135. Gold and Iron. Iron in small quantity is some- times added to gold alloys for ornamental purposes, in order to impart a characteristic tint. Gold and iron combine in all proportions, the former increasing the fusibility of the latter. Gold in small quantity does not seem to impair the qualities of iron. Guettier states that an alloy of equal parts gold and iron is greyish-white, brittle, and slightly magnetic. The alloy containing ^ iron is pale yellow, and becomes greyish-yellow when the iron is increased to ^. This is known to some jewellers as grey gold. Gee gives 18 parts gold to 6 parts iron as the proportions of blue gold. With regard to this alloy, he directs that the gold should be melted first, and then iron wire in small pieces introduced successively into the molten metal. When cast it must be hammered on the edge and annealed, in order to give a closer grain, and pre- vent cracking during the rolling. The process may be wisely repeated upon the surface, and the ingot again annealed. The alloy may then be safely wrought into wire or sheets. ix GOLD ALLOYS 371 136. Gold and Platinum. These metals unite to form ductile and elastic alloys, but require a high temperature to effect their combination in consequence of the high melting point of platinum. This circumstance, combined with the effect the platinum possesses of making the colour of gold paler, considerably limits the application of these alloys for jewellery. Roberts- Austen states that '6 per cent of platinum will saturate gold. It also imparts a characteristic crystalline surface. Platinum, however, like gold, is not acted upon by nitric acid, or by the atmosphere. An alloy of 7 parts platinum and 3 parts gold is infusible in the strongest blast- furnace, but with a greater proportion of gold fusion takes place. 2 parts platinum and 1 part gold form a brittle alloy. 1 part platinum and 1 part gold form a malleable alloy of a pale gold colour. Clarke states that an alloy of 9*6 parts gold and 1 part platinum has the colour of gold and the density of platinum. It is well known that in gold-platinum alloys certain portions of the constituents separate and become concentrated either in the centre or in the external portions of the solidified mass. Mr. Edward Matthey 1 has investigated this subject ; he cast gold containing platinum into a spherical iron mould, 3 inches in diameter, and cut the metal so obtained into halves. The shrinkage was so great that the spheres had to be cast several times in order to get them solid. Portions were then taken from different parts of the spheres and assayed. A. Composed of about 880 gold and 050 platinum. B. Composed of about 700 gold and 120 platinum. In the sphere A the maximum difference between the gold percentage is a variation of 032, viz. 887 on the outside against 883'8 at the centre of the alloy ; and in the platinum 047-5 on the outside against 052-5 at the centre, showing an extreme variation of 005. In the sphere B the maximum difference between the gold percentage is a variation of 041, viz. 732-4 on the 1 Proc. Roy. Soc. 13th February 1890. 372 MIXK1) MKTALS ,-i.vr against 694*1 at the centre of the alloy ; and in the platinum 122 on the outside against 166 at the centre, giving an extreme variation of 044. These results show indisputably that the platinum in cooling liquates from the gold and becomes concentrated toward* the centre of the alloy. The above experiments were made on gold -platinum alloys containing silver and copper. In order to prove whether a similar liquation takes place \\ith alloys con- taining k'ohl anil plat ilium alone, 1)00 parts of lim- gold were repeatedly melted with I OO parts of pure platinum, and then cost as before. The result showed that the exterior con- tained 900 gold and 098 platinum, against 846 gold and 146 platinum at the centre of the sphere. 137. Gold and Palladium. Several alloys of these metals have been formed, the combination taking place with- out incandescence. 1 part palladium and 1 part gold form a grey alloy, having the colour of wrought-iron, less ductile than either of the component metals, ami of a coarse-brained fracture. 1 part palladium ami 4 parts gold yield a while, hard, ductile alloy. 1 part palladium and f> parts gold is almost white. Alloys of gold, silver, copper, and palladium have been used for hearings of the arbors in ^ood watches ; the colour is brownish-red, they are as hard as iron, do not rust, and cause the minimum of friction. The following is a typical alloy for watches : (Jold ;?7'f>, copper -J7'l, silver -J'J'i), palladium 12'5. Ber/e.lius analysed a native alloy from Porpez, containing 85'98 gold, 9 -85 palladium, and 4'17 silver per cent. 138. Gold and Aluminium. An alloy of gold, copper, and aluminium, known as Nurnberg gold, has been used in the manufacture of cheap gold-ware. Its colour resembles that of gold, and it 18 said to remain unchanged "I air. Aluminium combines with gold most readily and when cold shows a marked granular structure. It lowers the initial IX GOLD ALLOYS 373 livr/ing point, and the alloy only partly solidifies during a long range of temperature. This long pasty stage makes it ditlicult to determine the true freezing point. An alloy of gold and aluminium, having the formula AuAl 2 , has an intense purple colour and a melting point much higher than that of gold, which points to chemical union. *2 per cent aluminium enables gold to be cast with remarkable sound- ness. 139. Coloured Golds. Jewellers and goldsmiths use a variety of gold alloys for purposes of ornamentation, so as to produce a number of different shades of colour in the same article. For example, red and white are employed lor flowers, green for leaves, and yellow for stems, sprays, etc. The following table gives the composition per cent of alloys most in use : Colour. Gold. Silver. Copper. Iron. Platinum. Cadmium. White . 100 100 Grey . 857 8-6 5-7 88 -a ... 167 . t 72-6 27-5 ... _ Green . 75 2.". ... 75 16-6 84 74-6 11-4 97 i ;t , 75 12-5 12 5 Pale yellow 91-67 8-33 ... . ) 9 91'67 s-:;:; Very pale Yellow. 50 100 50 ... ... Deep yellow 90 ... 10 . j > 58 25 22 .. . Red . 75 ... 26 t Dark red 50 50 ii 25 75 Blue . 75 25 M 667 ... ... :;:;:; Japanese blue ItolO 99 to 90 ... gold 374 MIXED METALS CHAP. The necessary precautions for the preparation of gold alloys containing iron have been given (see p. 370). The alloys of gold, silver, and copper call for no special re- marks here. The alloys containing cadmium given in the above table are malleable and ductile. It should be borne in mind that cadmium is a volatile metal like zinc, and should therefore be added after the other metals are melted under a layer of charcoal. Much depends upon the rapidity with which it is added, and the perfect incorporation of this metal in the alloy will also depend on the vigour with which the metals are quickly stirred, preferably with a charcoal stick. In any case, some of the cadmium volatilises, and a little more than the stated amount should be added, to com- pensate for this loss. The term coloured golds in the above heading does not refer to the chemical process of colouring, which will be referred to hereafter. 140. Standard Gold. In most countries there are gold alloys of a certain degree of fineness, fixed by law, and used as the national standard of value for coinage. They nearly all consist of gold and copper, the latter metal being necessary to enable the gold to resist the wear to which it is subjected in commerce. Notwithstanding the hardening effect of the base metal, coins wear considerably when frequently used, causing them to become light in weight. An English sovereign weighs 123*27,447 grains, and remains a legal tender till it is reduced below 122 '5 grains, the difference between these two weights being the remedy allowed by English law for abrasion or loss by wear. Besides the standard fineness for coins, there is also a legal weight, fixed according to the regulations of the Royal Mint. 1 Ib. troy of standard gold is worth .46 : 14 : 11, and if a person were to take 1 Ib. troy of standard gold to the Mint to be coined, he would receive 46 sovereigns, 1 half- sovereign, and nearly 5 shillings in return without beingf charged anything for the coinage of the gold. Hence, standard gold contains nominally and intrinsically its full ix STANDARD GOLD 375 value of gold. The coinage is conducted with great exactness by the officers of the Mint with respect to weight, and the extreme accuracy with which they are compelled to work is seen by the following table : Coin. Weight in grains. Remedy in grains. Sovereign 123 "27,447 "20, 000 Half-sovereign 61 '63,723 0'10,000. The remedy, of ^ and T ^ of a grain respectively, is the difference allowed to the Deputy-Master of the Mint, between the standard and real weight of the manufactured coins ; and he invariably confines himself well within these limits. The British standard for gold coins is 22 carat or 916-666 parts of gold and 83-333 parts of copper per thousand, the remedy being '002 or g-J-^ of a grain. The old guinea is of the same standard as the sovereign, but contains a little silver and less copper. The standards of different countries are given in the following table : [TABLE 376 MIXED METALS CHAP. Name. Gold per 1000 parts. Carat. Egyptian, Mexican, Spanish, \ Philippine Islands / 875 21 Old German coins (pistoles) . 895 21-5 German coins Austrian crowns French coins Belgian Italian Swiss Spanish 900 21-6 Greek United States coins Japanese ,, Chinese ,, Russian (new) j Prussian Freidrich'sc 'or 902 21-65 British coins "| Prussian (old) coins I Portuguese ,, V . 916-6 22 Turkish Brazilian , , J Dutch ducats 982 23-57 Austrian ,, 986 23-66 Hungarian ,, 989 2374 Previous to the reign of Charles II. all the coin of this realm was made by. hand by forging pieces of gold to the proper thickness required for the coins, then cutting them into squares a little larger than required for the different sizes. The corners were afterwards removed from the squares, and the pieces rounded to the proper size, when they were adjusted to the desired weight for circulation. These round blanks were then placed consecutively between steel dies, containing the pattern of the intended coin ; the upper die was then struck with a hammer to produce the impression. The different alloys used by jewellers in this country have been already discussed. In France 18, 20, and 22 carat gold are assay-marked. In Germany 18, 14, 8, and a 6-carat alloy, termed joujou gold, used for electro - gilding, are IX STANDARD GOLD 377 recognised. Austria has three legal standards : No. I 326, No. II 545, No. Ill 767 degrees of fineness per thousand. Pforzheim Gold-ware (German) Ordinary ware (joujou) Finer quality Finest quality 130 to 250 parts gold per 1000 563 583 to 750 , 141. Since the introduction of the decimal system, the method of expressing the fineness of gold alloys in thousandths has been gradually gaming ground. Its simplicity, over the old system of carats and grains, is its great recommenda- tion. The carat consists of 4 carat-grains. The following table shows the equivalents of carat-grains and carats in thousandths : 11 carats = 458 -630 1 grain = 10 "414 2 20-828 3 31*242 4 41-66 lea at =41 -667 2 83-334 3 125-001 4 166-667 5 208-333 6 250-000 7 291-666 8 333-333 9 374-999 10 416-667 12 500-000 13 541-667 14 583-333 15 624-555 16 666*667 17 707-333 18 750-000 19 791-666 20 833-333 21 874-999 22 916-666 23 958-333 24 1000-000 PREPARATION OF GOLD ALLOYS 142. The gold and silver, as purchased from the refiner, is sufficiently pure, in most cases, for the purposes of the manufacturer. Various kinds of copper are used, but in all cases it is required to be of excellent quality. The impurities most likely to be present are : Iron, arsenic, antimony, and sulphur, the two latter being specially hurt- ful. Persons using copper for alloying should specify that 378 MIXED METALS CHAP. these elements be absent, or only present in extremely minute quantities. Arsenic and sulphur are much more commonly present than antimony. Dr. A. S. Taylor l states that he found arsenic in not fewer than forty samples of copper, employed in the form of wire, sheet, gauze, etc. He further states that he found arsenic in two out of five specimens of electrotype copper. He does not appear to have examined any specimens of best-selected copper. It should be stated that the better qualities of commercial copper are now superior in purity to those manufactured forty years ago. Gold and its alloys should be melted in good plumbago crucibles, which, if carefully annealed in an inverted position before use, will last a great number of times. The crucible should be covered with a lid. A new pot should be rubbed inside with charcoal powder, to prevent any particles of metal adhering to it. Some jewellers lay stress on the manner in which the metals are placed in the crucible, and it is a good plan to have the gold on the top, as it does not oxidise, and to some extent protects the copper, especially when covered with a layer of charcoal. In making alloys of which zinc is a constituent, this metal must be added after the other metals are melted, on account of its volatile nature. When the mixture is melted the whole must be vigorously stirred with an iron rod (previously made red-hot), in order to promote more intimate union between the constituents of the alloy. The metal is then ready for pouring into the mould if the proper temperature has been attained. Gold alloys are cast either into iron or sand moulds according to the purpose for which the metal is designed. Jewellers rarely use moulds for shaping the articles, except such as have considerable thickness, such as seal-rings, and articles with high relief. For small gold work the cuttle- fish bone is used as a mould, the pattern being pressed in, as in the case of sand. Sometimes the process is simply performed by rubbing two pieces of bone quite flat on a smooth stone, and then cutting in one of the pieces the shape 1 Research on Arsenic, Antimony, etc., p. 28. ix GOLD ALLOYS 379 required, leaving a hole through to the edge by which to pour in the metal. Bathbrick also forms a convenient mould, as the pattern may be cut in with a tool, and used for 2 or 3 different castings. For the most part gold is cast into ingot-moulds, and subsequently rolled into sheet, or rolled and drawn into wire. The various precautions, referred to in a previous part of this work, when treating of the casting of metals, apply in a great measure to gold alloys. The mould must be dry, suitably warmed, and blackened or greased to prevent the metal sticking. If the mould be too cold or too hot, the metal will spit, and thus incur loss of gold. Care must be taken not to let the charcoal or dross run with the metal into the mould; carelessness in this respect is often the cause of faulty castings. The dross, etc., may be prevented from passing into the mould with the gold by using a thin piece of flat wood, held in the left hand ; poplar wood is preferred, because it burns away very slowly. Many defects arise from the use of bad charcoal-powder, which is some- times contaminated with coal - dust. The latter may be detected by washing a portion, when the coal will impart a colour to the water, while water filtered from a good charcoal will be perfectly colourless. Gold sometimes cracks during rolling, due to its brittle- ness, from the presence of impurities. Many substances are recommended and used for the toughening and brighten- ing of jewellers' gold alloys, consisting, for the most part, of substances containing chlorine, such as common salt, bi- chloride of mercury, and sal-ammoniac. Common salt is not advisable, as it produces a very liquid slag, which is liable to run into the ingot-mould along with the metal, making the ingot irregular and full of small holes. The same remarks apply to borax. Bichloride of mercury is very useful when the brittleness of the gold arises from the presence of lead or tin ; the fractured surface of the bar then presents a close grain, of a pale yellow colour. In most cases sal-ammoniac is the best agent for producing tough 380 MIXED METALS CHAP. gold. The sal-ammoniac, like the mercury chloride, is a volatile substance, and partly vaporises, while another por- tion is decomposed into ammonia and hydrochloric acid. The latter probably acts on the base metals, yielding up to them its contained chlorine and forming volatile chlorides. Thus, no slag is left to run into the moulds and interfere with the ingot of purified metal. In most cases a very small amount of this flux is required, and, indeed, a great excess will be injurious. In all cases, where gold alloys containing copper are melted for toughening, a little charcoal should be added along with the chloride employed. In making gold alloys, the ingredients of which do not contain sufficient impurities to interfere with the toughness of the gold, it is not advisable to use anything else except a little powdered charcoal, which forms a protective coating and prevents the oxidation of the copper. In melting precious metals the quality of the coke employed is not such a matter of indifference as some manufacturers suppose. The following observations by Mr. C. Tookey, formerly assistant assayer at the Japanese Mint, bear on this subject. 1 In the Imperial Mint at Osaka, Japan, bars of gold from San Francisco were frequently converted into standard metal by melting them with the requisite proportion of Japanese copper to produce an alloy containing y 9 ^^- of gold. In melting one particular consignment there was an unusual loss. No good reason could be assigned in the Melting Department, but on investigation Mr. Tookey found that the plumbago muffles and stirrers used in the operation were coated with minute particles of the gold alloy. It appeared as though the particles had been projected from the surface of the molten metal by the escape of a gas. Mr. Tookey had often watched a similar phenomenon when pots containing a silver - copper alloy were removed from the furnaces previous to pouring. While one of these pots, holding 2500 ounces, was cooling on the floor of the 1 Percy's Metallurgy, Gold and Silver, Part I. p. 489. ix GOLD ALLOYS 381 melting room to the proper temperature, a vigorous effer- vescence took place at the surface of the metal, projecting it in most minute particles, which were deposited on the floor of the room. During the effervescence there was a powerful smell of sulphurous acid, which had no doubt been absorbed or occluded by the metal, while the pot containing it had been exposed, at a high temperature, to the products of combustion of coke containing much sulphur. The size of the cast bars varies according to the branch of the jewellery trade for which they are required. Locket -makers cast the gold in broad and thin plates. Chain-makers cast their metal in long and tolerably thick strips, which, when rolled to about Nos. 10, 11, or 12 of the Birmingham wire gauge, are annealed and cut into strips in the slitting mill, when they are drawn into wire. An ordinary jeweller's melting furnace is a wind furnace about 9 inches square and 18 inches deep, lined with fire-brick. The draught-hole is 6 by 3 inches, and the ash-pit should have a capacity at least equal to that of the furnace. The bars are 10 inches long, 1 J inches wide at the top, and gradually taper towards the bottom. The furnace must be connected with a chimney at least 40 feet high, in order to obtain an active and strong draught. The furnace mouth is closed by two fire-bricks, each of which is clamped by a piece of flat bar-iron, well wedged on. The draught is regulated by a suitable damper fixed in the flue. Coke is the fuel generally employed. This should be of good quality and practically free from sulphur. The furnaces used for melting gold for coinage are of larger dimensions than the one just described, and correspond in capacity and arrangement to those used for ordinary brass melting (see Figs. 20-24) 5 described under the head of " brass." The gold is sent from the Bank of England to the Mint in ingots of 400 ounces each. The crucibles are made of plum- bago, and are capable of melting 1200 ounces of standard gold. The pots are heated previously to the introduction of the charge, to prevent them cracking or flying when con- 382 MIXED METALS CHAP. taining the precious metal. To ascertain whether a crucible is sound, a cold bar of iron is put to the bottom, when, if any cracks exist they will become visible. When the mixture of gold and copper is melted it is thoroughly stirred with an iron rod, and then poured into ingot-moulds. The plates of gold are then assayed, two assays being made by two different men for each plate, and the correct standard thus determined. If found correct, the plates are then weighed and rolled to the desired size for cutting out the blanks. 143. Preparation of Pure Gold. The following method has been adopted for the manufacture of the pure gold Trial- Plate now in the custody of the Warden of the Standards. Fine gold is dissolved in aqua-regia, the excess of acid driven off, and alcohol and potassium chloride added to precipitate traces of platinum. The chloride of gold is then diluted with distilled water in the proportion of half an ounce to the gallon, when the solution is allowed to stand three weeks. The solution is then carefully syphoned off, and oxalic acid in crystals added from time to time until the solution is colourless, the precipitation of the gold towards the end being aided by a gentle heat. The spongy gold so obtained is washed repeatedly with hydrochloric acid, distilled water, ammonia water, and lastly with distilled water. It is then melted in a Picardy crucible with a little pure bisulphate of potash and borax, and poured into a stone mould. The Trial-Plate prepared in this way weighed 70 ounces, and was of the average purity of 999 '96 parts of gold per 1000. 144. Refining of Gold. The gold employed for coinage, when it does not contain more than 10 per cent of silver, is usually purified by Miller's process with chlorine gas. This process consists of melting the impure gold in a clay crucible, which has been glazed inside with borax, and passing chlorine through the molten metal by means of a clay pipe. The chlorine combines with the silver to form silver chloride, ix REFINING GOLD 383 which rises to the surface of the molten metal, whilst the chlorides of base metals which may be present, such as zinc, bismuth, antimony, arsenic, etc., are volatilised. A layer of borax is placed on the top to prevent the silver chloride from being volatilised. The gold thus refined varies in purity from 991 to 997 in 1000 parts, which is purer than ordinary fine gold. Dr. J. C. Booth, of the United States Mint, has dis- covered a general method of toughening gold and silver, which he described to the American Chemical Society a few years back. Some time ago Mr. Booth found that a quantity of brittle gold accidentally melted with some tough gold in a crucible had rendered the whole mass very brittle, crystalline in fracture, and therefore useless for coining. The whole was toughened by him in one and a half days, at a trifling cost, by the new process. The 75,000 ounces of gold were divided into 14 "melts" of 5400 ounces each, and each melt separately toughened. The ingots, which could be broken into pieces by striking them on the edge of a wooden box, were put into the crucible with soda ash and anhydrous fused borax, in the ratio of one or two ounces to a melt, until the crucible was nearly full. After melting it appeared as a quiet mass of metal covered with a viscid slag, disposed to swell and puff. A few crystals of saltpetre, say one or two ounces, were then dropped into the centre of the metallic surface, and as they melted, their spreading out over the whole surface was aided by the concentric motion of the bottom of a small crucible. The moment the visible oxidising action began to slacken, the melter skimmed off, by a small blacklead dripping crucible, the fluxed matter as rapidly as was con- sistent with the care necessary to avoid taking up metal. The remainder in the melting pot was the toughened metal. 145. Gold Plating. This consists of joining a bar of gold and gilding or other metal together by sweating or soldering. A bar of gold of a desired quality and a bar of 384 MIXED METALS CHAP. base metal are first made perfectly flat under a stamp or press; then the surfaces to be joined are filed or scraped clean ; borax is next prepared and well rubbed over the surfaces. The two bars of metal are firmly secured together by iron wire, placed in a muffle, and the temperature raised nearly to the point of fusion, when the metals unite into one compact bar. This is termed joining by " sweating." Another method, which is more generally adopted, is to join the two bars by soldering. The two bars are prepared as in the former case. The metal-bar, being larger than the gold-bar, supports the pieces of solder, which are placed along one side, and half way along each end. The whole is then strongly heated in a muffle until the solder melts, and joins them together. These compound bars may be rolled, stamped, spun, or otherwise manipulated, as though they had been melted to form one homogeneous mass. The gold follows the reduction of the base metal during the pro- cesses of rolling, etc., and retains the relative proportion of thickness between them. HARD SOLDERS FOR GOLD WORK 146. With respect to hard solders, it may be taken as a general guide that the more nearly the solder approxi- mates in composition and properties to that of the metal to be soldered, the more perfect will the union of the two parts be, and the stronger the point of juncture. On the other hand, the greater the difference between the melting points of the solder and the metal to be soldered the easier will the operation be. In all cases it is necessary that the solder should have a lower melting point than the metal to be soldered. Solder is said to be "hard" when it has a high melting point, i.e. at or above a red heat, and is literally hard with regard to its power of resisting the pressure of a cutting or abrading tool. As the melting point of the solder alloy is lowered by the addition of an easily fusible metal, or by the IX GOLD SOLDERS 385 addition of a greater amount of the more fusible constituents of the mixture, the solder is said to be " easy " or softer. Gold solders are made in a great variety of degrees of softness and hardness by the addition of different proportions of a more fusible metal, such as silver, to the gold. Thus, in 18-carat gold, for example, the solder may be made from the 18-carat alloy by adding a certain proportion of silver and copper, so as not to materially alter the colour. If the solder is made too poor in quality, the articles will not colour properly. Coloured gold solders contain 1 part of silver to 4, 5, or 6 parts of alloyed gold, according to the degree of fusibility desired. l Gee gives the following tables for coloured gold solders : Description. Fine gold. Fine silver. Copper. Best solder Medium solder Common solder 12i parts 10 81 41 parts 6 ,, i 3 parts 4 ,, 5 ,, These solders may be rolled thin, and cut with the shears and pressed into suitable pieces termed " pallions," or filed into dust, according to the needs of the workman. TABLE OF COLOURED-GOLD SOLDERS SUITABLE FOR THE FOLLOWING PROCESSES Description. Fine gold. Fine silver. Copper or compo. Total. Value per oz. oz. dwt. gr. oz. dwt gr. oz. dwt. gr. 07. dwt. gr. s. d. Dry colouring 1 060 040 1 10 2 18 French , , 100 080 070 1 15 2 10 London ,, 1 11 090 200 250 Birmingham ,, 1 12 12 10 2 2 12 226 German ,, 1 13 12 250 200 Best bright 1 1 1 100 310 1 10 1 1 Goldsmiths Handbook, pp. 136, 217. 1881. 2 C 386 MIXED METALS CHAP. The following table gives the composition of various solders for alloys of different standards : Gold. Silver. Copper. Zinc. Hard solder for gold 10 5 1-00 Hard solder for 16-carat gold 9 2 i" Easier ,, ,, ,, 12 7 3 Solder for about 14-carat gold 3 2 1 99 99 99 99 * 2 0-5 0-5 ,, for less than 14-carat gold . 1 2 1 j> 1 2 1 2 ... Very easy solder .... 11-54 5474 28-17 5-55 147. Colouring of Gold. This operation consists of imparting a colour to gold articles after every other process has been completed. It is stated that its object is to give to alloyed gold all the appearance of fine gold itself by dissolv- ing out the base metal from the surface of the articles and leaving a facing of gold of a deep rich colour. This is only partially true, as some of the gold is also dissolved. More- over, pure gold is of a pale yellow colour, and the tint of coloured gold is generally a reddish - yellow, showing that the surface cannot be pure gold. The special attraction of coloured gold is the rich dead appearance, due to the uniform matting of the surface by the solvent action of the acid. Two distinct modes of colouring are adopted by jewellers, termed respectively dry colouring and wet colouring. The latter is most frequently practised, as the former cannot well be applied to gold inferior to 18-carat. Wet Colouring. The ingredients of the mixture employed in this process have a powerfully solvent action on the base metal with which the gold is alloyed, and a weaker action on the gold itself, so that the article loses weight in direct ratio to the length of time it is submitted to the colouring process, and this loss is greater as the gold is lower in ix COLOUKING OF GOLD 387 quality. l Gee states that the colouring is hastened and the loss in weight reduced to a minimum by using old colouring liquid, and he assumes that the dissolved gold is, to some extent, deposited again on the article, because the loss in weight of some common qualities of gold was found to be very little, and the amount of gold recovered from the spent colouring liquid very small indeed. This statement is in accord with the well-known fact that in any liquid in which a metal say copper is electro-positive to the metal in solu- tion say gold the latter is deposited on the former. But this deposit is somewhat loosely adherent, and most of it is removed by the subsequent process of scratch-brushing. Many different mixtures are used for colouring gold, some of which will be afterwards given in tabular form. The following has been supplied to the author by an experienced Birmingham jeweller, which he has found to be effective : Potassium nitrate . 12 ounces Common salt . . . k 6 ,, Hydrochloric acid . . . 3 ,, The nitrate and salt are pounded to a fine powder and placed in a previously warmed plumbago crucible about 8 inches by 7 inches, then stirred with a wooden spoon for a minute or two. The acid is then added, with about 1 ounce of boiling water, and the mass constantly stirred until it boils up to the top of the pot. The work, which has been previously cleansed in hot potash or soda solution, is then suspended in the colouring liquid by means of a silver or platinum wire for about one minute, then well swilled in boiling water. A little more water is added to the colour- pot, and when the liquid boils up the work is again im- mersed for another minute, and swilled in boiling water as before. This operation of dipping and swilling is repeated several times, the colouring liquid being weakened by adding 1 Goldsmiths' Handbook, p. 161. 388 MIXED METALS CHAP. water before each immersion, until the desired appearance is attained. The work is finally well washed in hot water and dried in boxwood sawdust. The whole process takes five to seven minutes. The coloured work is next scratch-brushed, on a lathe, with a revolving brush made of very fine brass wire and having stale beer dropping on it. If the colouring has been properly conducted, a beautiful rich and dead colour will be produced. Dry Colouring. This term is applied to the colouring process when no liquids are used as constituents of the mixture. The ingredients used are Potassium nitrate . . .8 ounces Common salt . . . . 4 Alum 4 ,, These substances are ground to a fine powder, well mixed and placed in a previously -heated blacklead " colour "-pot, of the same dimensions as that described for use in wet colouring ; but the same pot must not be employed for dry colouring as has been used for the wet process. It is well to get the pot nearly red-hot before placing the "colour " in it. The mixture must then be constantly stirred with an iron rod. It will first boil up as a greenish liquid, then solidify, and afterwards boil up a second time and become thoroughly fused, having a brownish-yellow colour. At this stage the work, which has been previously annealed and dipped in dilute aquafortis, is dipped in the " colour," being suspended on a silver or platinum wire, the latter being preferred, and kept in motion for about a minute and a half, then immersed in boiling water containing a little aquafortis. The immer- sion and swilling are again repeated, when the articles possess a beautiful colour. They are then washed in hot water containing a little potash, and finally dried in warm boxwood sawdust. In dry-colouring the work should be as highly polished as possible previous to the colouring, for the brighter it is the IX COLOURING OF GOLD 389 better will be the final colour. The time given above is only intended as a general guide, as some work will colour much quicker than others, and the time can only be arrived at by experience. The following mixtures have been recom- mended for colouring : Dry process. Wet process. Potassium nitrate 8 oz. Common salt . 4 Alum . . . 4 ,, Potassium nitrate . Common salt Alum . Hydrochloric acid Water in each case 8 4 4 14 7 7 2 15 7 7 1 14 7 "5 Sal-ammoniac . 4 oz. Potassium nitrate 4 ,, Borax . . . 4 ,, The chemical reactions in the colour-pot may be briefly summarised as follows : Nitric and hydrochloric acids by their mutual reaction liberate chlorine, which attacks the gold and other metals rapidly when heated, forming chlorides, of which silver chloride is insoluble, and by coating the gold retards the action, thus preventing the inspection of the sur- face during the process. For this reason this mixture of acids is not suitable as a " colouring " liquid. If we consider the mixture of potassium nitrate, common salt, hydrochloric acid, and water, given on p. 387, the same formation of chlorine gas occurs, and the same reactions on the metals takes place, but the silver chloride is soluble in the salt solution if it is kept concentrated ; thus the work is clean and can be pro- perly inspected during the operation. By increasing the water the action is slower, but the proper proportion of salt must be maintained in order to prevent the silver chloride being deposited. 1 The following is a useful mixture for removing tarnish from coloured-gold articles which have been kept in stock for some time : See J. H. Stansbie, Art of Goldsmith and Jeweller, p. 158. 390 MIXED METALS CHAP, ix Bicarbonate of soda ... 2 ounces Chloride of lime .... 1 ounce Common salt 1 ,, Water ...... 16 ounces Well mix the above ingredients and apply with a soft brush. CHAPTER X SILVER ALLOYS 148. Articles are very seldom manufactured from fine silver free from alloy, as pure silver is far too soft to resist the wear to which most bodies are subjected. It is therefore alloyed with some other metal, chiefly copper, to impart the requisite degree of hardness. Fine silver, in consequence of its high ductility, is used in the manu- facture of silver lace and fine filigree-work, the latter being principally made in India, Sweden, Norway, and some parts of Germany, where labour is cheap. The purest commercial silver contains minute quantities of other elements, which, in the best varieties, do not materially affect its working properties. It may be necessary in some cases to obtain chemically pure silver, and this may be done in the following ways. Ordinary silver is dissolved in pure nitric acid, when any gold is left undissolved, and is removed by filtration. The silver solution is next evaporated to dryness, and the residue fused to decompose any platinum nitrate that may be present. The residue is then dissolved in dilute ammonia and filtered ; and the filtered blue liquid diluted with enough water to bring the strength down to 2 per cent of silver. A sufficient quantity of normal ammonium sulphate is now added to render the solution colourless on warming, and the liquid is allowed to stand in closed stoppered vessels for twenty-four hours, when a third of the silver separates out in the crystalline form. 391 392 MIXED METALS CHAP. The liquid, which is still blue when cold, is poured off and heated from 60 to 70 C., when the remainder of the silver is deposited. In order to remove every trace of copper, the metallic precipitate is washed with water, and allowed to stand several days in contact with strong ammonia ; it is then again washed, dried, and fused in an unglazed porcelain crucible with 5 per cent of pure borax and 5 per cent of pure sodium nitrate. Lastly, it is cast in moulds lined with a mixture of burnt and unburnt porcelain clay. The bars of silver must then be cleaned with sand, and heated with potash solution to remove every trace of adherent silicate, and finally washed with water. To prepare pure silver, Stas dissolved fine silver in dilute nitric acid, evaporated the solution to dryness, and ignited the residue until all the red fumes were evolved. The mass was dissolved in water, filtered, and diluted with rain-water (30 of water to 1 of silver), and the silver precipitated as chloride with pure hydrochloric acid. The precipitate was washed with dilute hydrochloric acid, then with pure water, and then dried, and the powder well rubbed in a clean porcelain mortar. This was then repeatedly digested with aqua-regia, and afterwards well washed. The silver chloride was then reduced to metal by boiling with dilute pure caustic potash and some milk-sugar. The reduced silver was washed with dilute sulphuric acid, then with water, then dried and fused. Metallic silver has the power of absorbing certain gases when melted in contact with them, but the gases are, for the most part, expelled during the solidification of the metal, raising blisters on the surface, or covering the same with a number of small excrescences, giving it a frosted appearance. This action is termed spitting or vegetating. When molten silver is allowed to cool slowly out of contact with air, the gases gradually escape, and little evidence of spitting is exhibited. The gas which produces this phenomenon is principally oxygen. When silver is melted under a suffi- ciently thick layer of non-oxidising material, such as common SILVER ALLOYS 393 salt, or potash, the metal solidifies with a bright surface, showing that oxygen had not been absorbed and afterwards emitted. When powdered charcoal is thrown on the surface of the metal, the carbon withdraws the absorbed oxygen and prevents the silver from spitting. Silver may be alloyed with gold, even to the extent of one-third of its weight, without losing its power of absorbing oxygen when melted, and of spitting during solidification. 1 Chevillot states that silver alloyed with copper, of the re- spective standards 990 and 995 parts of silver per 1000, exhibits the phenomenon of spitting ; that silver of the standard 952 does not evolve gas in sensible quantity; and he supposes that silver of the standard 980 is the limit at which spitting occurs. Silver is capable of absorbing oxygen and other gases, and retaining them when cold, by heating the metal to red- ness in contact with oxygen or other gas. Such gas is said to be occluded. Graham found that pure silver occluded 545 of its volume of oxygen ; and fine silver wire '002 inch diameter, yielded *289 of its volume of a gas, consisting chiefly of carbonic acid. When standard silver is heated to low-redness it becomes almost black on the surface, from the oxidation of the copper. Silver wire thus blackened was found to have occluded several times its volume of oxygen. Fine silver wire heated to redness in hydrogen, and cooled in that gas, occludes *211 of its volume of hydrogen. Fine silver is somewhat extensively used in the manu- facture of fine wire and filigree-work on the Continent and in India. The Indian workman accomplishes work of a very beautiful kind, representing flowers, animals, etc., with true artistic taste. The articles are " hand-made " with the aid of a few simple tools. The work is commenced by hammer- ing out the metal on an anvil, and when it has assumed a certain degree of thinness, it is cut into strips, and drawn 1 Memoirs of the Phil. Soc. of Manchester, second series, 1819, p. 271. 394 MIXED METALS CHAP. into very thin wire through perforated steel plates, a pair of strong pliers being used for the purpose. The wire is then used for fashioning various ornamental articles. Indian filigree-work is said to be the finest and cheapest in the world. It is of importance, in this class of work, that the various forms required in filigree-work should steadily retain their place when pressed into shape, and not rebound like metals of a highly elastic nature ; hence the need of using tine silver in preference to standard silver. 149. Silver and Arsenic. These metals are capable of uniting in several proportions, forming hard, grey, brittle, and readily fusible alloys. Gehlen produced an alloy con- taining 16 per cent arsenic, which is compact, brittle, steel- grey, and fine-grained. Berthier describes an alloy of 14*8 per cent arsenic as dull-grey, brittle, and crystalline ; by burnishing it acquires the lustre and colour of silver ; it is very fusible, and not decomposed on heating. Guettier describes an alloy containing 1,4 per cent arsenic, formerly used for table ware. Mr. R. Smith prepared a hard and brittle though somewhat tough alloy, which became white and lustrous on burnishing. It contained 18*54 per cent arsenic, and corresponded to the formula Ag 3 As. Silver-arsenic alloys may be prepared by direct fusion of the constituent metals, or by melting a mixture of silver, arsenious acid, and black flux. 150. Silver and Antimony. Alloys of these metals may be obtained in all proportions by direct fusion. They are hard, brittle, and grey or white in colour. The white- ness decreases with the proportion of antimony. The alloys are very fusible, and wholly decomposed by cupellation or by fusion with nitre, pure silver remaining. 1 Mr. R. Smith has prepared the following alloys : i II in Silver . . 72 '65 77 '98 84-16 Antimony . 27 '35 22'02 15 '84 1 Percy's Gold and Silver, vol. i. p. 143. SILVER ALLOYS 395 ^responding to the formulae 3Ag + Sb, 4Ag-f Sb, and 6Ag + Sb respectively. The silver was melted first under a layer of charcoal, and the antimony then added. No. I was hard, crystalline, and bluish- white. No. II was similar to No. I, but greyish-white. No. Ill was hard, granular, and greyish-white. The specific gravities of 48 silver-antimony alloys containing 50 per cent of silver, and upwards, has been determined by Cooke, of Harvard College, U.S., who found that the densities were above the mean densities of the constituents, the maximum being reached in the alloy containing 26*6 per cent of antimony. Cooke also found that the crystallisation of the alloys became marked in pro- portion as the same composition is approached. 151. Silver and Bismuth. Alloys of these metals are hard, easily fusible, brittle, and lamellar in structure. The colour of the 50 per cent silver alloy is the same as that of bismuth. An alloy containing 33*33 per cent silver is said to be steel-grey and to expand on solidification. Schneider states that when impure bismuth, containing sulphur, arsenic, iron, nickel, and silver, is fused and poured upon a cold plate, the globules of metal which are thrown up during solidification of the mass contain at least 99 '5 per cent bismuth, and of the heavy metals only silver is found in the bismuth. Even a very small quantity of bismuth renders silver ingots very brittle, and if dropped on the floor will break into several pieces. The metal is then very crystalline, each crystal itself being ductile, while the mass of the ingot is very brittle. This brittleness is probably due to the presence of a fusible eutectic, which surrounds each of the crystals. 152. Silver and Tin. A very small quantity of tin renders silver brittle. Alloys of tin and silver, according to Guettier, are harsh, very hard, and brittle. An alloy of 80 per cent tin is nearly as hard as bronze. An alloy of 52 per cent tin is somewhat malleable. These alloys are very easily 396 MIXED METALS CHAP. oxidised. They have a specific gravity less than the mean of the constituents. Tin may be removed from silver by fusion with bichloride of mercury (corrosive sublimate), leaving the silver pure. Dentists use an alloy of 60 parts silver and 40 parts tin, in admixture with mercury, for stopping teeth. 153. Silver and Zinc. These metals combine very readily, forming bluish-grey and, for the most part, brittle alloys. Those with excess of zinc are granular, but with an excess of silver the fracture becomes columnar. The metals combine much more readily at a high than at a moderate temperature. 1 Berthier prepared an alloy containing 80 per cent silver, which he states was rolled into very thin leaf ; it was rigid, elastic, very tenacious, and tough. 2 Mr. G. H. Godfrey prepared the following alloys by pouring molten zinc into molten silver : I II III IV Silver . 8 '16 22 -47 4972 67 '58 Zinc . 91-84 77'53 50-28 32 "42 " I. The surface was bluish-grey. The metal was hard, easily frangible, and easily scratched with a knife. Its fracture was bluish-grey, finely granular, and feebly lustrous. " II. The surface was bluish - grey. The metal was harder than No. I, easily frangible, but less easily scratched. Its fracture was bluish-grey, bright, and fibro-columnar. " III. The surface was copper - red after solidification. The metal was hard, brittle, and easily pulverised. The broken surface, when fractured cold, was white and very bright, and somewhat columnar. " IV. The surface had a faint reddish-yellow tint. The metal was hard and easily frangible ; its fracture white and very bright, but it soon tarnished ; it was columnar in structure." An alloy of 2 parts by weight of zinc and 1 part silver 1 Traite des Ess. (2) p. 798. 2 Percy's Gold and Silver, vol. i. p. 169. x SILVER ALLOYS 397 is said to be ductile, finely granular, and nearly as white as silver. 1 Silver-zinc alloys have been proposed for coinage purposes. Peligot prepared alloys containing 5, 10, and 20 per cent of zinc respectively. They were white, with a tinge of yellow. The coins were elastic and sonorous. These alloys are not so readily blackened by sulphuretted hydrogen as silver- copper alloys. 154. Silver and Iron. These metals do not alloy well together. Messrs. Stoddard and Faraday made some experiments with silver in steel, and concluded that ^J^ of silver corresponds to the best mixture. These alloys do not appear to present any practical interest. 155. Silver and Nickel. Berthier described an alloy of these metals containing 13*5 per cent nickel which was white, and capable of a high polish ; it rolled well, and was very tough. There appears to be very little known concerning alloys of these two metals alone. 156. Silver and Lead. Alloys of these metals are of little interest from a commercial point of view. The metals readily unite in all proportions. A very small amount of lead is sufficient to diminish the malleability and ductility of silver. Molten lead dissolves silver just as mercury does, and homogeneous mixtures are obtained only while the metals are liquid, a certain amount of liquation taking place as the metals cool. 2 Levol has investigated this subject, and his results are tabulated below. 1 Handworterbuch der Chemie (7), p. 958. 2 Mtmoire sur Us AUiages, Ann. de Chim. 1853 3, ser. 39, p. 173. 398 MIXED METALS CHAP. Atomic formula. Silver per 1C Calculated. K)0 of alloy. Found. Variations in the ingot. I A g20 Pb 912-5 914 7-5 II Ag 12 Pb 862-0 863 14-5 III Ag 10 Pb 839-1 840-5 23-5 IV Ag 4 Pb 675-9 676-5 49-5 V Ag 2 Pb 510-5 516-6 66-5 VI Ag Pb 342-8 347-5 11-0 VII Ag 2 Pb 3 258-0 262-0 13-0 VIII Ag Pb 2 206-8 206-0 6-5 IX Ag Pb 5 94-4 ... 19-5 X Ag 2 Pb 15 65-0 67-25 7-5 XI Ag Pb 10 49-4 46-00 2-5 XII Ag Pbso 10-3 9-75 25 I. Greyish - white, but little malleable, and contracts during solidification. II. Greyish - white, resembles platinum in colour and grain, contracts during solidification, and changes rapidly in moist air. III. Greyish-white ; contracts strongly during solidifica- tion ; heated in air it assumes a beautiful violet-blue tint. IV. Alloy tolerably malleable, but has only feeble tenacity, and melts near cherry-red heat ; it is bluish-grey in colour, and quickly oxidises in moist air. V. l Is much more like lead than silver, soft, and tolerably malleable and ductile. The others require no special comments. 157. Silver and Aluminium. Alloys of these metals were made some years ago, and it was thought that valuable metals of a white colour, and unaffected by the atmosphere, would be obtained, which would make them superior to ordinary silver-copper alloys ; but these great expectations have not as yet been realised. Aluminium hardens silver, and the alloys admit of a high polish. 1 Guettier, Guide Pratique des Alliages. 1865, p. 150. x SILVER ALLOYS 399 1 MM. C. and A. Tissier state that an alloy of 4'75 per cent silver and 95 '25 aluminium is more elastic and harder than aluminium, and as malleable as the latter metal. Aluminium alloyed with 10 per cent silver is no longer malleable. An alloy of equal parts of silver and aluminium is said to be as hard as bronze. 2 Lange uses an alloy of 100 parts aluminium and 5 parts silver for watch-springs. Such springs are said to be very elastic, hard, light, not so brittle as steel, and not to rust. Tiers-argent (one-third silver). This alloy is said to be manufactured at Paris into various utensils, and consists of 33*33 parts silver and 66*66 parts aluminium. MM. Tissier state that this same alloy may be used as a solder, but it runs with difficulty and produces a brittle joint. 158. Silver and Copper. These metals unite in all proportions, forming a series of most valuable alloys, having a great variety of applications in the Arts. Combination takes place with expansion, so that the specific gravities are less than the mean of their constituents. Most of the alloys are as ductile as silver, and possess more hardness, elasticity, and sonorousness. The colour of these alloys is white until the copper reaches nearly 50 per cent, and beyond that the colour is yellowish, up to about 70 per cent copper, when a red tint prevails. The hardest alloy is that containing 5 parts by weight of silver to 1 or 1 1 parts copper. 3 Professor Roberts- Austen has determined the melting points of certain silver-copper alloys, and states that the alloy containing 630'29 of silver per 1000 of alloy, and represented by the formula AgCu, has a lower melting point than silver, or than any other alloy of silver and copper. His results are given in the following table. He has since communicated to the author the fact that the results are too 1 L' Aluminium et les Mitaux alkalius, 1858, p. 173. 2 Jahresber. 1874, p. 1077. 3 Pro. Roy. Soc. vol. xxiiL pp. 349, 481. 400 MIXED METALS CHAP. high, as more recent determinations of the melting point of silver make it 967 C. instead of 1040 C. According to J. Violle its melting point is 954C. 1 No. Pure silver per 1000. Approximate formula. Melting point. 1 1000 Ag 1040 C. 2 925 Ag 7 Cu 931-1 C. 3 8207 Ag 3 Cu 886-2 4 798 Ag 5 Cu 2 887 5 773-6 Ag 2 Cu 858-3 6 750-3 Ag 7 Cu 4 850-4 7 718-93 Ag 3 Cu 2 870-5 8 630-29 Ag Cu 846-8 9 600 Ag 7 Cu 8 857 10 569-9 Ag 7 Cu 9 899-9 11 561-1 Ag 3 Cu 4 917-6 12 540-8 AgsoCu^ 919-8 13 500 Ag 3 Cu 5 940-8 14 497 AgasCutf 962-6 15 459-4 Ag Cu 2 960-8 16 250-5 Ag Cu 5 1114-1 17 Pure copper Cu 1330 The alloys numbered 7 and 8 are of especial interest. The first Ag 3 Cu 2 is Level's homogeneous alloy, which remains uniform in composition, while many others undergo liquation on slow cooling. The alloy AgCu has the simplest atomic relation and the lowest melting point. In studying the phenomena of liquation, the alloys were cast in moulds of fire-brick, in which the metal (about 50 ounces) could be quickly or slowly and uniformly cooled. The results showed that the homogeneity of Levol's alloy is slightly disturbed by this method of casting ; and, on the other hand, that alloys which contain more than 71-89 per cent of silver hardly show signs of rearrangement when the solidification takes place gradually. Two alloys containing 1 Compt. rend, torn, Ixxxv. p. 543. SILVER ALLOYS 401 63 and 33-3 per cent of silver respectively, were found to be far from homogeneous, and in the former the arrangement was influenced by gravity, the base of the casting being richer in silver. The following table contains a summary of Roberts- Austen's results : Maximum varia- No. Designation. Parts of silver per 1000. Rate of cooling. tion in the quan- tity of silver in different parts of the alloy. I.V Ib) British standard coins 925 ( r fP id V slow 12-8 per 1000 1-4 II a\ lib/ Old French stan- dard coin 900 f rapid \ slow 10-1 1-3 III Level's homogen- 718-93 slow 1-2 eous alloy IV AgCu 630-3 slow 21-1 V AgCu 4 333-3 slow 12-8 In I a and II a the centre of the cubes was the richest and the corners the poorest. In I b and II b the slight variations followed the same law. In III the corners were generally richer than the centre. In IV it was supposed that gravity had influenced the alloy, the lower parts being richer than the upper. In V the variations do not follow any known law. With regard to standard silver cast under ordinary con- ditions, the tendency of the copper and silver to separate appears to depend upon the inequality of the rate of cooling in the different parts of the ingot. The act of cooling causes a partial separation of the copper at the parts first cooled, and those parts which solidify last are generally richer in silver. Silver-copper alloys are subject to change when strongly heated in air, the copper being oxidised. If the alloy con- tains much copper, the silver will also, to some extent, be 2 D 402 MIXED METALS CHAP. oxidised, but in a less degree than the copper. When such alloys are just heated to redness in an ordinary muffle with accession of air the discoloration of the surface is proportionate to the amount of copper. The following table is due to Chaudet : : Silver in 1000 parts of alloy. Characters of the surfaces after heating. 1000, i.e. pure silver Dull but white. 950 . . . Uniform grey-white. 900 . . . Dull grey- white, black fillet at edges. 880 . . . Grey, almost black. 860 . . . ,, ,, 840 . . . Quite black. 820 . 800 . 159. Standard or Sterling Silver. British silver coin and plate contain definite proportions of silver and copper, regulated by law. Silver coin contains 11 oz. 2 dwts. of silver per Ib. troy, or 925 parts of silver per 1000, the remaining 75 parts being copper. It should be mentioned that the copper, added to the silver, is termed the alloy, whereas, strictly speaking, the whole body, containing silver and copper, is the alloy. Articles of silver to be Hall-marked must be made of standard silver, as no other quality is allowed by law to be assay-marked. In preparing standard silver for Hall-marking it is necessary to add a little more than the above stated amount, because commercial fine silver is never perfectly pure ; so that instead of using 18 dwts. 12 grs. per ounce, use 18 dwts. 14 grs. The quality commonly used by silversmiths is generally below the standard, when not Hall-marked, and contains 18 dwts. of silver per ounce, or 900 parts per 1000. 160. The various alloys of silver and copper employed for manufacturing purposes are represented in the following table 2 : 1 VArtde V Essay eur. Par M. Chaudet, pp. 77, 78. 2 Gee, Silversmiths' Handbook, pp. 64-67. SILVER ALLOYS 403 Silver Copper . I II III IV oz. dwt. gr. 18 020 oz. dwt. gr. 16 040 oz. dwt. gr. 15 050 oz. dwt. gr. 14 060 Silver Copper . V VI VII VIII oz. dwt gr. 13 12 6 12 oz. dwt. gr. 13 070 oz. dwt. gr. 12 12 7 12 oz. dwt. gr. 12 080 Xo. VIII is about the commonest alloy it is possible to make without a perceptible yellow cast being imparted to the colour. A commoner variety may, however, be made by adding a third metal such as nickel, which will be sub- sequently referred to. The alloys for coin and plate until recent years legal in France were as follows * : - M*-r BBSS. Errors toler- ated. Copper in 1000 parts. Silver coin . Coin named " billon " Silver plate . ,, medals ,, jewellery . solder 110 -26 to 10 '3 100 ; 9 -37 to 9-5 800 50 50 200 120 to 330 97 to 103 797 to 807 50 to 55 47 to 53 The 10 centimes pieces, termed "billon," have not been made since 1810. The standard for all French silver coins under the 5-franc piece has been lowered to 835 of silver per 1000 of alloy. 1 L'Art de I'Essayeur. Par M. Chaudet, p. 335. 404 MIXED METALS CHAP. In Germany there are four silver standards, which are as follows : Silver ware . 11 oz. 8 dwts. or 950 parts silver per 1000 Coinage . . 10 ,, 16 ,, or 900 ,, ,, ,, Silverware . 9 ,, 12 ,, or 800 ,, ,, ,, Silverware . 9 ,, or 750 ,, ,, No other metal than copper is allowed to be alloyed with the silver. Elliot and Storer x have detected the presence of lead in silver coins of the United States and other countries. Their results are given in the following table : Kind of coin. Silver per 1000. Lead per cent. 1 American half-dollar, 1824 . 900 3100 20 ,, 5-cent pieces, 1853 . 900 2090 10 ,, 1854 . 900 2282 2 ,, 25-cent ,, 1858 . 900 2305 Fine silver from U.S. Assay Office in New\ York, 1860 / ... 1611 1 Spanish dollar, 1793, Carolus IV. . 0558 1 Mexican ,, 1829 .... 0434 2 English shillings, 1816 .... 1 French 5-franc piece, 1852, Napoleon III. 925 900 4847 4282 The lead in American coins is accounted for by the zinc, used in the reduction of the chloride of silver, containing some lead. The silver at the French Mint is precipitated by copper from an acid solution of sulphate of silver, and the acid probably contains sulphate of lead. The lead in English coin may be due to imperfect cupellation. The following table gives the composition of coins of various countries : 1 Amer. Acad. of Arts and Sc. 1860, p. 52. SILVER ALLOYS 405 Country. Coins. Fineness. Great Britain Shillings, etc. 925 Australasia . Cape of Good Hope j> Canada 50 cents Newfoundland M j > East Indies, Burmah, Ceylon, \ and Mauritius / Rupee 916-667 1 France, Belgium, and Swit- \ zerland / 5 Francs 900 Italy .... 5 Liras it Greece .... 5 Drachms Servia .... 5 Dinars a Roumania . 5 Leys t ) Bulgaria 21 Lew 835 Austria, Hungary Florins 900 Spain .... 5 Pesedas, 2 escudos Argentine Republic Bolivia 1 Peso 1 Boliviano 11 Chili .... 1 Peso 5> Peru .... ISol Germany 1 Mark Russia .... 1 Rouble 868-056 Denmark, Sweden, and Norway 2 Crowns 800 Netherlands . 1 Florin 945 Portugal 5 Testoons 916-667 Ottoman Empire . 20 Piastres 830 Egypt .... 10 Piastres 900 T-fc Persia .... 10 Scahis ii United States 1 Dollar j Mexico. 1 Peso 902-778 Brazil .... 2 Milreis 916-667 Japan .... 1 Yen 900 Cochin China Trade piastre M Silver, Copper, Zinc, and Nickel. Some foreign coins have been made of alloys of these metals ; and certain French manufacturers have employed alloys of copper, nickel, and silver as a substitute for standard silver. A series of alloys known as Argent-Ruolz were patented 1 See p. 403. 406 MIXED METALS CHAP. by De Euolz and De Fontenay in 1853, which they state may be used for all purposes for which silver is usually employed. The alloys contain 20 parts silver, 30 to 75 parts nickel, and 70 to 75 parts copper. It is recommended to melt the copper and nickel first, and then to add the silver. Char- coal and borax are to be used as the flux, and the ingots obtained are to be rendered malleable by heating them for some time in powdered charcoal. A second patent was obtained in 1854 for the introduction of a little phosphorus into the same alloys. The advantages claimed are that the alloys are more fusible, the molten metal very liquid, and the castings free from porosity, closer in grain, and whiter in colour. The phosphorus, however, greatly lessens the malleability and ductility of the alloys. C. D. Abel of London also patented certain silver alloys containing nickel, having the following composition : I II III Silver 33 40 20 Nickel . 25 to 30 20 to 30 25 to 35 Copper . 37 to 42 30 to 40 45 to 55 IV V VI Silver 33-3 34 40 Nickel . 8'6 8 4-6 Copper 41-8 42 44-6 Zinc 16-3 16 10-8 Nos. IV and V are specially designed for rolled, pressed, or drawn silver work ; No. Ill for cast work ; and No. VI for jewellery. Mousse? s silver alloy contains 59 J copper, 27 J silver, 9 1 zinc, and 3j nickel per cent. SILVER ALLOYS 407 Swiss monetary alloys are composed of : 20 Centimes. 10 Centimes. 5 Centimes. Silver 15 10 5 Copper . 50 55 60 Nickel . 25 25 25 Zinc 10 10 10 Mr. A. Parkes of Birmingham patented in 1844 the following alloys : Silver Nickel Copper Zinc I 20 20 60 100 II 10 10 60 20 100 1 Pelligot prepared the following alloys, which are said to be whiter and more malleable than alloys of silver and copper containing the same proportions of silver : Silver Copper Zinc . 90 5 5 II 80 10 10 III 83-5 9-3 7-2 The following alloys have been compiled from various sources : I II III IV V VI VII VIII IX Silver Copper . Nickel . Zinc 40 33 26 36-4 36-4 27-2 31 40 29 27 42 31 25 44 31 20 50 30 40 32 21 7 36 34 22 8 33 37 25 5 Comptes rendus, 1864, p. 645. 408 MIXED METALS CHAP. Silver, Copper, and Cadmium. Brannt states that cadmium imparts to silver alloys great flexibility and duc- tility without impairing their white colour, and gives the following as the more important alloys : i II III IV V VI VII Silver . 980 950 900 800 666 667 500 Copper Cadmium . 15 5 15 35 18 82 20 180 25 309 50 283 50 450 SILVER SOLDERS 161. A solder must be more fusible than the articles to be soldered, should run well when fused, and blend perfectly with the edges of surfaces to be united, otherwise strong and invisible connections cannot be effected ; hence solders of different qualities are required for different proportions of the metals in the alloys. The best junctions are made when the metal and solder are, as nearly as possible, equal as regards hardness, malleability, and fusibility. When the fusibility of the solder is nearly the same as that of the metal, greater skill is required on the part of the workman to prevent the metal melting, but the union will be more perfect. As a metal approaches its melting-point it becomes more and more expanded, i.e. more porous in structure, and the molecules of the solder, when melted, will more easily penetrate the meshes of the molecular network of the metal, and form a more intimate relationship. 1 Gee says, " the best solders we have found to be those mixed with a little zinc. These may be laminated, rolled, or filed into dust ; if the latter, it should be finely done, and this is better for every purpose. Too much zinc has a ten- dency to eat itself away during wear, thus destroying the 1 Silversmiths' Handbook, p. 76. x SILVER SOLDERS 409 article after a certain time. Such solders also volatilise considerably during soldering, so that the soldering operation has often to be repeated, thus wasting time and solder." The hardest solders, i.e. those most difficult to melt, are those made of silver and copper only ; medium solders contain zinc in addition to copper and silver ; and the easiest or softest solders are prepared with the addition of a certain proportion of tin. Arsenic is sometimes added to produce greater fusibility. For two metal surfaces to be joined together by soldering it is necessary that they should be clean. But during heat- ing oxidation takes place, unless the metal is protected from the oxygen of the air, so that a flux is added with a view of protection, and also for the purpose of dissolving the oxide formed, and leaving the parts to be joined perfectly clean. The flux used for the above purpose is borax. English Silver Solders 1 1. Fine silver 4 parts by weight, copper 1 part. II. Standard silver 3 parts by weight, brass (2 copper, to 1 zinc) 1 part. III. Fine silver 2 parts by weight, brass 1 part. Mr. Edward Matthey states that No. Ill is a very nice white solder and is used for ordinary plate-work. Gee says that the above solders are not the best for ordinary silver work The first is too difficult to melt The second and third are liable to contain lead, which burns away and corrodes the metal. The same remarks apply to No. III. When brass is used in making solder care must be taken that the spelter used in its composition is free from lead. In order to test the suitability of different compositions for soldering articles made of standard silver, the author performed the following experiments. A sample of solder, used for this purpose, was obtained from a large manu- facturer, and on analysis gave : 1 Percy's Gold and Silver, p. 166. 410 MIXED METALS CHAP. Silver . . . 64 '15 Copper . . . 22*85 Zinc 13-00 100 As the goods manufactured by the above firm require to be Hall-marked, it is necessary to make the metal better than standard in order to compensate for the low standard of the solder. The object of the author was, therefore, to obtain a solder as near as possible to standard silver, and thus avoid the necessity of alloying the metal so high. It was thought that by reducing the quantity of copper and replacing it by zinc the object might be attained. I II III IV Silver 925 925 925 925 Copper 50 38 25 Zinc 25 37 50 75 The above solders were tested by an experienced workman and all pronounced too hard, and not sufficiently liquid when melted to run well. No. II was considered the best of the series. The following were then tried : I II in Silver 800 800 800 Copper 100 50 25 Zinc 100 150 175 No. I was found to be the hardest and No. Ill the easiest to melt, which ran as readily as the solder given in the above analysis, containing 64 per cent of silver, and appeared to answer the purpose equally well. The following solder is mentioned by Tenner as suitable for jewellery : Parts by weight, oz. dwt. gr. Silver . . . 63 "3 or 12 16 Copper . . . 3-4 ,, 16 Brass . . . 33'3 ,,0 6 16 100 100 SILVER SOLDERS 411 French Silver Solders. The following solder is used for soldering silver wares of the standard 950 : Silver Copper Zinc. Parts by weight. 66-6 23-3 10 99-9 In making this solder, it is recommended to previously alloy the zinc with twice its weight of copper, when the following proportions are used : Silver Copper 1 Guettier gives the following : 66-6 30 3-3 99-9 I II III IV V VI Silver .... 5 3 2 4 2 1 Brass .... 1 1 1 Bronze (90 copper, 10 tin) Arsenic .... 3 25 5 1 Copper 1 1 He further states that these solders ought to be melted several times. The metal is then laminated into thin bands, which are granulated into spangles, ready to be mixed with borax. When arsenic is used, it is not added until after the fusion of the other metals. Gee denies that there is any advantage in re-melting silver solder containing zinc several times, but that his experience is in direct opposition to this practice. Solders, into the 1 Guide Pratique, des Alliages Metalliques, pp. 324, 326. 412 MIXED METALS CHAP. composition of which volatile metals enter, become hard, brittle, and " drossy " by repeated re-meltings, and are best only melted once. In 1635 Bate published the following directions for preparing silver solders. " Take a quarter of an ounce of silver and 3 pennyweights of copper, melt them together, and it is done." 1 The following solders are recommended for special work : oz. dwt. gr. oz. dwt. gr. I. Fine silver 100 II. Fine silver 100 Shot copper 050 Shot copper 10 1 5 oz. dwt. gr. III. Fine silver 16 Shot copper Composition 50 3 12 12 1 oz. dwt. KT. V. Fine silver 1 Shot copper Pure spelter 12 3 1 15 1 10 IV. Fine silver Composition Pure tin oz. dwt. gr. 100 10 020 1 12 oz. dwt. gr. VI. Fine silver 100 Shot copper 030 Arsenic 020 oz. dwt. gr. oz. dwt. gr. VII. Fine silver 1 VIII. Fine silver 100 Composition 060 Composition 050 Arsenic 010 Tin 050 oz. dwt. gr. IX. Fine silver 100 Tin Arsenic 10 050 1 15 1 10 oz. dwt. gr. X. Fine silver 100 Composition 15 Arsenic 016 1 16 1 Gee's Silversmiths' Handbook, p. 84. x SILVER SOLDERS 413 Nos. I and II are recommended for work to be enamelled. No. III. Easy solder for filigree-work. No. IV. Easy solder for chains. No. V. Common easy solder. No. VIII. Easy silver solder. Noe. IX and X. Common easy solders. Silver solders are used for soldering other metals and alloys, such as cast-iron, steel, brass, German silver, gold alloys, etc. As already mentioned, the substance most commonly used as a flux in hard - soldering is borax. Powdered glass is occasionally used with very hard solders. In Vienna a substance is used termed " streu-borax," or "sprinkle-borax." It is composed of the following ingredients, which should be gently heated to expel the water of crystallisation, and the whole well pounded ready for use : Parts by weight. Calcined borax .... 87 Carbonate of soda .... 7 Common salt 5 100 The object of the mixture is to prevent the rising of the solder, and to facilitate its flushing. It, however, encumbers the work with more flux than when borax alone is used, and, if kept for some time after mixing in the wet state, turns the solder yellow. IMITATION SILVER ALLOYS 162. Clark's Patent Alloy. Copper 75, nickel 14-5, zinc 7*5, tin 1*5, cobalt 1'5 per cent. Baudoin's Alloy, Copper 72, nickel 16*6, cobalt 1*8, tin 2-5, zinc 7'1 per cent. About J per cent of aluminium may also be added. Parisian Alloy. Copper 69, nickel 19-5, zinc 6 '5, cadmium 5 per cent. White Alloy. Copper 64'5, tin 32, arsenic 3'5 per cent. Chinese Silver. Copper 58, zinc 17 '5, nickel 11*5, cobalt 11, silver 2 per cent. 414 MIXED METALS CHAP. Warned Alloy. Tin 37, nickel 26, bismuth 26, cobalt 1 1 per cent. Minargent. Copper 56, nickel 40, tungsten 3, aluminium 1 per cent. White Alloy. Copper 59, tin 31, brass 8, arsenic 2 per cent. See also the chapter on German silvers, and also the alloys of silver, zinc, copper, and nickel. The above alloys are used for cheap jewellery and electro- plated wares. Wires prepared from them may serve for stems of pins, brooch tongs, catches and joints, etc. They are harder and more difficult to work than ordinary silver alloys, but their hardness and tenacity adapt them for the purposes above mentioned. MIXING AND MELTING OP SILVER AND ITS ALLOYS 163. The crucibles best adapted for melting silver are made of clay and plumbago and known as "blacklead" and plumbago crucibles. They are capable of withstanding the action of heat and sudden changes of temperature much better than ordinary clay crucibles, and if previously annealed, can be used many times in succession without cracking or breaking. They also resist the corrosive action of slags and fluxes better than clay-pots. Alloys consisting of silver and copper only may be pre- pared by melting the two metals together, as previously stated, under a layer of charcoal powder. When melted the mixture should be well stirred with an iron rod, and the metal, if at the proper temperature, poured. When zinc is a constituent of the alloy, that metal must be heated and cautiously added after the silver and copper are melted, and the whole then vigorously stirred. It should be borne in mind that zinc is a volatile metal, and that the volatility increases as the temperature rises, so that it is advisable to add the zinc as soon as the other metals are melted. The x SILVER SOLDERS 415 inetal is generally poured into flat ingot-moulds, so as to produce a plate of silver suitable for rolling and for 'wire- drawing, and the same precautions apply as stated when treating of gold alloys. The charcoal powder employed should be of good quality, as defective alloys are sometimes produced by using bad charcoal. When tin is used, it should be added after melting the silver and copper, but this is not so necessary as in the case of zinc or arsenic. When scrap silver or silver alloys are added to the crucible alone, or along with new metal, certain impurities may also be admitted, and a flux will be necessary to remove them. In most cases carbonate of soda or borax is employed. The latter should be sparingly used. 164. Lemel, as the filings and turnings, etc., are termed, is purified by burning off organic matter, and then melting in a skittle-shaped crucible with suitable fluxes. The following may be taken as a general guide : Lemel 70 to 80 parts Carbonate of soda . . . 10 ,, 15 ,, Common salt 5 ,, Bisulphate of potash (sal -enixum) . . 2 ,, Saltpetre may be used instead of the sal-enixum, but either of these salts should be sparingly employed. The common salt should not be mixed with the silver and other fluxes, but kept as a covering for the mixture, as it prevents the mass rising too much, and overflowing the crucible. The crucible should not be more than half full to commence with. After the whole mass has become liquid, keep it in fusion for about half an hour. Then allow the contents gradually to cool, and break the pot to recover the lump of metal. Some manufacturers prefer to run down the " lemel " in an ordinary plumbago crucible, with carbonate of soda alone as the flux. When the metal is well fused, the mixture is well stirred with an iron rod from time to time, and the metal finally poured into an ingot-mould, ready for the refiner. 416 MIXED METALS CHAP. HALL-MARKING AND ASSAY OFFICES 165. Standard silver is assayed and marked at certain duly authorised offices in various parts of the kingdom. The marks indicate the maker, quality of the standard, the place and year of assay, and the payment of duty. The name of the maker is indicated by his initials, the standard of 925 by a Lion Passant, the place of assay by heraldic arms, the year of assay by a letter, which is changed every year, and the payment of duty by the sovereign's head. There are seven assay offices, for which the arms are as follows : London, a leopard's head ; Birmingham, an anchor ; Chester, a sword between three garbs ; Exeter, a castle with three towers ; Sheffield, a crown ; Newcastle-upon-Tyne, three castles, with the addition of a leopard's head ; York, a cross and five lions, also with the addition of a leopard's head. There are two assay offices in Scotland, where the standard is indicated by the thistle. The distinctive marks are : Edinburgh, a castle ; Glasgow, a tree growing out of a mount, with a bell pendant on the sinister branch, and a bird on the top branch, over the trunk of a tree a salmon in fesse, in its mouth an annulet. In Ireland the assaying and marking is restricted to Dublin. The standard of 925 and place of assay are indicated by a harp crowned ; and the payment of duty by the figure of Hibernia, with an additional mark of the sovereign's head. POLISHING, FINISHING, ETC. 166. The beautiful lustre of silver is proverbial, as it is capable of receiving a most brilliant polish. The powders employed for polishing are Emery, pumice, rotten-stone, putty-powder (chiefly oxide of tin), crocus, rouge, etc., the latter being used for finishing. Scratches are often removed by a rather soft dark-grey stone, termed Water-of-Ayr stone. Emery and pumice are used for rough work to begin with, when the articles require much polishing. Oxide of iron, SILVER ALLOYS 417 sold under the names of colcothar of vitriol, crocus, tripel, or rouge, contains much impurity, which cannot be removed by washing. 1 Vogel recommends the use of ferric oxide, pre- pared by the calcination of ferrous oxalate, as preferable to the ordinary levigated colcothar. Mr. Ross communicated to the Sc. of Arts, May 1833, the following receipt: Crystals of ferrous sulphate are to be dissolved in water, and the solution filtered to remove siliceous matter. To the filtered solution a saturated solution of soda is added, and the precipitate repeatedly washed and then dried. It is then gradually heated to dull redness in a crucible, and poured into a dish. On cooling it absorbs oxygen, and acquires a beautiful dark-red colour, when it is fit for polishing silver and gold. To fit it for polishing harder substances it must be heated to bright redness, and kept so until it acquires a purple hue on exposure to air. After this treat- ment it must be nibbed with a wrought-iron spatula on a wrought-iron slab, and afterwards levigated in a very weak solution of gum-arabic. It is then almost impalpable, free from foreign matter, and eminently suited for polishing steel, glass, the softer gems, etc. No powder containing mercury should be used for polish- ing silver goods. The surfaces of silver articles are improved in appearance by a process of " whitening," by which a pure snow-white colour is imparted, after every other process of workmanship has been completed. This enrichment is most perfect when the metal is of good quality, and the finish, as regards smoothness and freedom from solder-marks, of a high order. Many different methods have been used. An old method is to dip the work in a thick solution of borax, then place it in a copper annealing pan, sprinkle it over with char- coal dust, and place the pan and its contents upon a clear fire. Heat until red-hot, then withdraw and allow to cool. The work is next boiled in dilute sulphuric acid, and if the right colour is not obtained the process is repeated one or 1 Chem. Gazette, 1854 (12), p. 410. 2 418 MIXED METALS CHAP. more times. The lower standards require five or six opera- tions to effect the proper degree of whiteness. Another plan is to dip the work in a mixture of 4 parts powdered charcoal and 1 part nitre, well mixed with water. The work is heated until the coating is thoroughly dry, when it is removed from the fire, allowed to cool, and boiled out in a solution of bisulphate of potash. After two or three operations a beautiful dead-white colour is the result It is then washed in soda and water containing a little soap, or " scratched " and burnished if required bright. The process is completed by drying in warm boxwood sawdust. 1 Gee's method of whitening consists of making the work red-hot, and boiling in dilute sulphuric acid (1 of acid to 40 of water). The process is repeated, if necessary, until the requisite colour is obtained. This method is not suitable for very common work, which requires a thin deposit of pure silver by the electro method, or by chemical decomposition of certain silver salts applied in the form of a paste, instead of subjecting it to the above whitening process. The articles may also be dipped in solutions containing silver, when silver is deposited on their surface. This is termed a " simple-immersion " process. 167. State in which Silver is Imported. "Silver bullion arrives in this country from America in various forms. That from Nevada and Canada is in rectangular oblong bars of various sizes, which weigh from 1000 to 1500 ounces each, and differ in content of silver from 100 to 999 per 1000, and which may or may not be free from gold. If * dore ' (i.e. such as contain gold), the pro- portion of gold rarely exceeds one-third of the weight, the alloy or base metal present chiefly consisting of copper and lead. The weight, assay produce, name of assayer, and value in dollars are usually stamped upon each bar. " The metal is sold in the state in which it arrives, unless of low quality, when it is usually re-melted here by one of 1 Silversmiths 1 Handbook, p. 145. x IMPORTED SILVER 419 the Bank of England melters. A large amount of silver comes from Mexico in the form of coined dollars, and is usually sold in that form for the China market, where the dollars are received as currency. Some Mexican dollars contain sufficient gold to pay for its extraction ; and these are generally distinguished by certain letters upon them, specifying the mint in Mexico where they were coined, and they are assorted accordingly. Plata Pina is another form in which silver is imported into this country from Chili, Peru, and a few other localities ; and such silver having been obtained by the process of amalgamation is generally in cylindrical or polygonal masses, according to the form of the moulds into which the amalgam was pressed before removing the mercury. Plata Pina sometimes contains gold, and is melted previously to sale in this market, when there is a loss of from 2 to 10 per cent, due to impurities. After fusion the metal is often 999 per 1000 fine. " Silver is imported from South America in the state of coin, such as sols, Bolivian dollars, etc. ; and in bars weighing 2000 to 3000 ounces each. These bars are either half-cylin- drical or rectangular. The former are of good quality, varying from 995 to 999 per 1000 fine. The bars vary in fineness from 950 to 990, and are extremely hard, due to the presence of sulphur, arsenic, and antimony, which often renders them very brittle. Up to 1875 these bars were melted on arrival and refined, as a necessary protection to the purchaser. They are now sold in the state in which they are received in order to save the cost of re-melting." J 1 Percy's Gold and Silver, vol. i. pp. 301, 302. CHAPTER XI PLATINUM ALLOYS 168. Platinum unites with most metals to form alloys, but in consequence of its high melting point, great difficulty is experienced in obtaining bodies of a definite composition containing volatile constituents. It should also be re- membered that metals like silver and copper, which, under ordinary circumstances, do not vaporise, readily assume the gaseous state at the melting point of platinum. When a small quantity of platinum is heated with a large excess of a more fusible metal, the fusing point of the former is lowered sufficiently to enable it to melt and alloy with the latter at ordinary furnace temperatures. The alloys of platinum are, for the most part, much more fusible than platinum itself. When it is desired to alloy much platinum with another metal, a special arrangement, known as the oxy-hydrogen blowpipe furnace, is employed to effect the fusion. Two well-fitting lumps of quicklime are hollowed out so as to form a cavity, in which the metal to be melted is placed. The cavity in the lower block is deeper than that of the upper one, as it has to contain the molten metal. The upper block is perforated through the centre with a hole, through which an oxy-hydrogen blowpipe passes, and a side opening between the two blocks permits the escape of the products of combustion, and serves as an outlet for the molten metal The blowpipe is a double tube, the inner 420 CHAP, xi PLATINUM ALLOYS 421 one conveying oxygen, and the outer one conveying hydrogen or coal-gas. The intense heat, produced by the burning of hydrogen in oxygen, is sufficient to melt a considerable quantity of platinum or platinum-alloy. In preparing an alloy, the platinum is melted first, and the other metal or metals then added through an aperture in the top of the furnace, which is then closed with a lime-plate. The flame can be modified by means of stop -cocks to suit the conditions required. If the oxygen be in excess during the process of alloying with a base metal, the latter will be largely oxidised, so that it is advisable to have a slight excess of hydrogen to prevent this loss. The molten metal or the prepared molten alloy is cast into bars or ingots in moulds of lime, in suitable sizes for wire-drawing or rolling. 169. Platinum and Silver. These metals unite in several proportions, forming white or greyish-white alloys, which are harder and tougher than silver, and less fusible, malleable, and ductile, as the proportion of platinum is greater. There is a great tendency for the two metals to separate according to their specific gravities on cooling, the platinum settling to the bottom. Alloys with 17 to 35 per cent of platinum are used in dentistry, and known as platine- au-titre. These alloys are less readily tarnished than silver or ordinary silver alloys. Lewis in the last century prepared small quantities of the following platinum alloys : I II III IV Silver ... 1 2 3 7 Platinum 1 1 1 1 The mixtures in I, II, and III required a strong white-heat for fusion ; the products were hard and brittle in proportion to the contained platinum. The constituents of No. IV melted readily, and the alloy was harder, greyer, and of a coarser grain than silver ; it hammered tolerably well. He found a considerable separation of platinum at the bottom 422 MIXED METALS CHAP. of the moulds, unless the alloy solidified immediately on pouring. 1 Berthier states that an alloy with 7 per cent of platinum is brittle, but Mr. E. Matthey denies the truth of this statement. An alloy of 37*5 per cent silver and 62-5 per cent platinum is said to have a colour half-way between silver and platinum, to flatten under the hammer, but to crack in rolling. With alloys low in platinum, nitric acid dissolves a certain quantity of platinum along with the silver. An alloy containing only 5 per cent of platinum dissolves completely in nitric acid. With sulphuric acid the silver only dissolves. When the above nitric acid solution is heated with sulphuric acid, the platinum separates out. 2 Messrs. Johnson and Matthey prepare an extremely ductile alloy of 2 parts silver and 1 part platinum, as an article of commerce. This alloy has been adopted as a standard of electrical resistance. Platinum and Gold. See Gold Alloys. 170. Platinum and Copper. Alloys of these metals may be obtained by fusion of the constituents in all proportions. A high temperature is required for their pro- duction, the oxy-hydrogen furnace being required when the platinum is in excess. They possess considerable ductility, malleability, and tenacity ; are capable of forming a variety of shades of colour, and are less tarnished by the atmosphere than alloys of copper with base metals. As the proportion of platinum increases the alloys become harder, whiter, and more brittle. They are capable of a high polish and have been used for the reflectors of telescopes. When zinc is added in addition to copper, alloys are obtained nearly equal to gold in colour and lustre, superior in durability, and used in the manufacture of jewellery and ornaments. An alloy of 1 part platinum and 4 parts copper is hard, 1 Traite des Essais (2), p. 800. 2 ffandworterbuch der Ghemie (7), p. 958. XI PLATINUM ALLOYS 423 ductile, of a yellow-pink colour, and susceptible of a high polish. An alloy of equal parts by weight of copper and platinum, according to Clarke, is yellow, having the colour and specific gravity of gold, extensible, easily worked by the file, and tarnished by exposure to air. An alloy of 4 parts platinum and 96 parts copper is malleable, rose-coloured, and exhibits a fine-grained fracture. An alloy of 3 parts platinum and 2 parts copper is nearly white, very hard, and brittle. 171. The following alloys have a golden-yellow colour. No. IV, known as Cooper's gold, is malleable, ductile, and closely resembles 18-carat gold : I II III IV V VI VII VIII Platinum Copper Zinc . Silver 18'2 45-5 9 5 5 29-3 667 4 1875 81-25 577 38-5 3-8 667 29-1 4-2 29-1 667 4-2 19 81 Brass . 18-3 60 Nickel Q 30 Cooper's Mirror Metal Copper 57 '85, platinum 9'49, zinc 3'51, tin 27'49, arsenic 1-66. The inventor claims for this alloy that it is indifferent to the weather, and takes a beautiful polish. Cooper's Pen Metal. The above alloy is said to be suit- able for pens. Another alloy consists of copper 13 parts, platinum 50 parts, and silver 36 parts. The hardness and non-corrosive character of Cooper's alloys render them suit- able for the manufacture of mathematical instruments and for chronometer wheels. 172. Platinum and Iridium. These metals unite in different proportions, but the intense heat of the oxy- 424 MIXED METALS CHAP. hydrogen blowpipe is necessary to inelt them and bring them into union. The alloy consisting of 9 parts platinum and 1 part iridiurn is used as a standard metal bar for the metric system. It is extremely hard, as elastic as steel, more difficultly fusible than platinum, perfectly unalter- able in air, and capable of taking an exceedingly beautiful polish. In the year 1870 Messrs. Johnson, Matthey, and Co. prepared a standard bar of the above alloy for the Parisian Commission of the International Metrical System, and after it had been subjected to every possible test which could be suggested in competition with other materials, it was, after two years' trial, pronounced the best, and adopted as the material for the manufacture of all the standard weights and measures. The following alloys have been prepared by Deville and Debray, and their specific gravities determined : Platinum. Iridium. Specific gravity. 90 ... 10 ... 21-615 85 ... 15 ... 21-618 66-67 . . . 33-33 . . . 21-874 5 ... 95 ... 22-384 Specific gravity of platinum = 21 '504 ; of iridium = 2 2 '421. According to Deville and Debray, an alloy of 90 per cent platinum and 10 iridium has the same coefficient of expansion as the original metre preserved in the French Archives, which is known to have been made of impure platinum. The 90 per cent platinum alloy is not attacked by aqua- regia. Alloys with 20 per cent iridium are malleable and capable of being worked. An alloy of equal parts of the two metals is brittle, but capable of welding to some extent. An alloy of 1 part iridium and 10 parts platinum, when laid on copper, serves for metallic mirrors. A native alloy of platinum and iridium is found in the Ural Mountains and in Brazil. xi PLATINUM ALLOYS 425 Platinum vessels for chemical operations generally contain iridium, which makes them stronger and harder. 173. Alloys of Platinum with easily fusible Metals. Platinum readily unites with arsenic and antimony, the combination being attended with vivid incandescence, form- ing brittle and easily fusible alloys. Tin unites with platinum when the metals are fused together in equal parts, forming a hard, dark-coloured, some- what fusible, brittle, and coarse-grained alloy. Zinc unites with platinum forming an alloy of similar, properties to the above-mentioned tin alloy. Platinum and Lead readily unite, and very little lead makes platinum brittle. When molten lead is poured upon platinum, a portion of the latter is fused and dissolved in the lead. The alloys are hard, brittle, and granular. Platinum and Bismuth form brittle alloys. Mr. Lewis found that the alloys ranging from 1 to 24 parts of bismuth to 1 of platinum are brittle, easily fusible, and have a laminar fracture. By contact with air they acquire a purple or violet tint. When moderately heated some of the alloys undergo liquation, the bismuth partially separat- ing out. When strongly heated in air the bismuth largely burns off, forming bismuth oxide. Platinum heated with Cadmium till the excess of the latter is volatilised forms a silver-white, very brittle, fine- grained alloy, refractory in the fire, and containing 46 per cent of platinum. 174. Platinum and Nickel. According to Lampadius, equal parts of nickel and platinum unite to form a pale yellowish-white alloy, perfectly malleable, susceptible of a high polish, equal to copper in fusibility and to nickel in magnetic power. 175. Platinor. This is a name given to certain alloys containing platinum, of a golden-yellow colour, and consist- 426 MIXED METALS CHAP. XI ing of platinum, copper, silver, zinc, and nickel. An alloy of the colour of gold, and said to be quite constant in air, is prepared as follows : Melt 10 parts of silver with 45 parts of copper, then add 1 8 parts of brass and 9 parts of nickel. The temperature must then be raised to the highest pitch, and 18 parts of platinum-black added. 176. Platinum-Bronze. Several alloys of platinum, of a comparatively inexpensive nature, have been manufactured under the above name ; and it has been claimed for them that they are indifferent to the action of the air and water. They admit of a high polish, and retain their lustre for a long time. The following table shows their composition and uses : Parts. Uses Nickel. Platinum. Tin. Silver. Brass. For table utensils . 90 9 9 bells . . 81-5 8 16 17 ,, articles of 86-5 5 13 ... luxury For tubes for tele- 71 14-5 14-5 scopes, etc. For ornaments 31-6 3-2 ... ... 65-2 176A. Ehodium Alloys. Rhodium has by some experi- menters been confounded with iridium. Neither of these metals forms real alloys with silver. On treating the mix- ture in nitric acid, rhodium separates out either in crystals or in the amorphous state. Iridium acts in a similar way. Rhodium and gold form a true alloy. Bismuth, tin, and antimony form alloys with rhodium. CHAPTER XII IRON AND STEEL ALLOYS 177. The general impression in the past has been that alloys of iron are of little importance, which was due to an imperfect acquaintance with their nature and properties ; but at the present time there is more light being thrown on the subject, and if we expand the idea of an alloy so as to include those compounds in which very small quantities of other metals, such as aluminium, are present, then the alloys of iron may be considered of very great importance. As a rule, iron may be alloyed with most metals; but the combination is somewhat difficult to effect, and, in the majority of cases, only those alloys with a small quantity of iron, or iron with a small quantity of other metals, have been found to have useful applications. Iron, added to other metals or alloys, imparts new and sometimes im- portant properties, such as increased hardness, elasticity, and tenacity. Many of the more recently discovered metals have of late years been added to iron with more or less marked alteration in its qualities, and such combinations are now ordinary commercial articles. In the following description of iron alloys, the metals known as "malleable-iron," "cast-iron," and " steel" are not treated as separate metals, but considered as different varieties of iron. 178. Iron and Manganese. Our knowledge of the 427 428 > MIXED METALS CHAP. alloys formed by these metals has been greatly increased within recent years by the labours of Mr. K. A. Had field of Sheffield. Iron readily unites with manganese, and when the proportion of the latter metal is considerable, the alloy is very hard, whiter, more fusible, and much more brittle in character than iron. In small quantity, up to '4 or '5 per cent, manganese is highly beneficial in steel, and some of the very best steel is that containing a little manganese. Steel containing from 2^ to 7 per cent of manganese is brittle and comparatively worthless, but when the amount exceeds 7 per cent, alloys possessing very great strength and toughness are obtained. 1 The weakness of the low percentage alloys may be understood ,from the following tests made by Mr. Hadfield. Cast-bars 2| inches square and 30 inches long, supported on bearings 2 feet apart, were broken by hydraulic pressure. One specimen containing -37 per cent of carbon and 4*45 per cent of manganese was fractured by a pressure of 3| tons, whilst a bar of ordinary cast-iron stood 1 2 tons, and bars containing 17 to 20 per cent manganese stood a pressure of 29^ and 38 tons respectively. A cast-bar containing 4*73 per cent manganese, when dropped from a height of 3 or 4 feet on to a cast-iron floor, broke in two or three places. A sample con- taining '48 per cent carbon and 4'9 per cent manganese, though very ductile while hot, could be reduced to powder by a hand -hammer when cold, little or no cohesion seeming to exist between the particles. On the other hand, a specimen of forged material, containing 13-75 per cent manganese and '85 per cent carbon, when water toughened had a tensile strength of 65 tons per square inch, with 50 per cent elongation ; another specimen had a strength of 69 tons, and 46 per cent elongation. The strongest alloy contains about 14 per cent of manganese. When manganese-steel is plunged into water no hardening effect takes place like that of ordinary steel, but the metal 1 Hadfield. Paper on Manganese Alloys. I. C. Eng. 28th February 1888. xn IRON ALLOYS 429 containing upwards of 7 per cent of manganese acquires increased tenacity and toughness. From a large number of tests it has been found that the higher the temperature to which the alloy is raised, and the more suddenly the cool- ing takes place, the higher is its breaking stress, and the greater its toughness and elongation. The influence of water quenching on manganese-steel is strikingly shown in alloys required to be drawn into wire. If it be attempted to draw hammered or rolled rods into wire without a previous heating and quenching it will not draw at all, and ordinary annealing makes no difference to it. If, however, it be raised to a yellow heat and then plunged into cold water it can be readily drawn into wire. After reducing the wire two numbers of the gauge, the metal is again heated and plunged into water. By this means it may be drawn to any reasonable degree of fineness. The density of manganese-steel is a little higher than that of ordinary steel. In its ordinary condition it is very hard and easily scratches steel that is not highly tempered. When the manganese exceeds 20 per cent the alloy is practically non-magnetic. Dr. Hopkinson found that the maximum magnetisation of wrought -iron and manganese steel (with 12*36 per cent of manganese) is as 258 to 1. It shows no elongation under the magnetic influence. Manganese-steel does not exhibit the anomalous expan- sion and "after-glow," termed re - calescence, which takes place in magnetic metals when they cool to a certain critical temperature, after being heated to whiteness. A Sheffield firm reported that in rolling a considerable length of manganese-steel, the finer it became the more it retained its heat, in fact, it appeared to gather heat in the process. Ferro-Manganese is a variety of metal specially manu- factured in a blast furnace from ores rich in oxide of manganese, and is very extensively used in the manufacture of mild steel. When the pig-iron contains less than about 20 per cent manganese, its fracture shows large crystalline 430 MIXED METALS CHAP. cleavage planes, and it is then termed spiegel-eisen. The variety known as ferro-manganese is a hard, crystalline body, but the fractured surface does not present the large cleavage planes so characteristic of spiegel-eisen. It contains from 20 to 85 per cent manganese. Mr. Hadfield has patented non-magnetic alloys suitable for dynamo brackets, or similar purposes. They are made by melting together iron and carbon, commonly about 3 J per cent, with other materials to give a composition of 7 to 30 per cent manganese and either 2^ to 8 per cent silicon, or 2 to 8 per cent of aluminium, or 1 ^ to 8 per cent each of aluminium and silicon. Appropriate alloys may be made from No. 1 hematite pig, siliceous pig, ferro-manganese, and aluminium. 179. Iron and Nickel. These metals unite together to form a series of alloys which have lately received a good deal of attention ; and although the most recent productions contain small quantities of other metals, such as manganese, they will be considered under this heading. Certain native alloys, occurring in aerolites, contain from 3 to 10 per cent of nickel. Faraday and Stodard prepared alloys of iron and nickel containing 3 and 10 per cent of nickel respectively. The former appeared to be as malleable, and could be as easily worked as pure iron. The latter was semi-ductile, very tenacious, with a granular fracture, and little affected by the atmosphere. They are capable of a high polish. Bergmann states that iron and nickel combine in all pro- portions. Lampadius states that an alloy of 5 parts nickel and 2 parts iron is moderately hard, easily malleable, and has the colour of steel. Mr. James Riley of Glasgow and Mr. Hall of Sheffield have independently given- considerable attention to iron- nickel alloys, and in a paper read by the former gentleman before the Iron and Steel Institute in May 1889, the following facts were stated : " The alloys can be made in crucibles or on a large scale in the Siemens's Open Hearth furnace, where a charge can be worked off in about seven xii IRON ALLOYS 431 hours. Its working demands no special care, and the com- position of the resulting steel is easily and definitely con- trolled. If the charge is properly worked, nearly all the nickel is found in the steel, and almost none in the slag. The metal sets steadily in the mould ; it is more fluid than ordinary steel when melted ; it sets more rapidly, and appears thoroughly homogeneous. The ingots are clean and smooth in appearance on the outside, but those richest in nickel are a little more piped than are ingots of ordinary mild steel. There is less liquation of the metalloids in these ingots, therefore the liability to serious troubles from this cause is much reduced. The scrap produced in hammer- ing, rolling, shearing, etc., can be re- melted in making another charge without loss of nickel. " If the steel has been properly made, and is of correct composition, it will hammer and roll well, whether it contains little or much nickel. It must be remembered, however, that in nickel -steel we have present nickel, manganese, and iron ; with carbon, silicon, phosphorus, and sulphur, and that a difference in quantity of some of these will influence the character of the alloy ; in other words, the degree of purity of the nickel and iron employed will modify the result, as in the case of ordinary iron and steel" The following table embodies the results of tests on various alloys made by Mr. Riley : [TABLE Teats as Rolled and Annealed. as Cast led. i jnao jad on jo uowoiu:juo3 s 004 ui ssaa^s jo uot^oj}aoo Extension per cent in inches. ^U30 J9d jo Extens per ce in inch suaj ui assays ^nao aad TOJ Extension per cent in inches. 8HO> U} 883^8 ooo s " s PO (NiOi-H r-l i-l T < M O OO r-l Risers, 87 precipitation of, 382 Roberts -Austen, 39, 42, 58, 92, 94 vessels, 425 115, 364, 399 Plumbago crucibles, 78. 381 muffles, 380 Robey, 263, 280 Rose's alloy, 342 Plumbers' solder, 329 Royal Mint, 364, 374 Plumbum nigrum, 3 Poling, 16 Rudberg, 324 Rndeloff, 273 Polishing, 62 Ruolz, 305 Porpez, 372 Positive, 4 Safety plugs, 336 Potassium, 35 Sagged, 263 amalgam, 360 Sal-ammoniac, 379 bitartrate, 72 San Francisco, 380 chlorate, 72 Sand, 10, 193 chloride, 382 Sankey and Smith, 448 cyanide, 72 Schneider, 437 nitrate, 72 Scrap-metal, 84 Potosi silver, 303 Pouring gate, 279 Screw-propeller, 270 Sea-water on metals, 146, 294 Powder cases, 198 Seal rings, 37S Preece, 262 Seared, 198 Prince's metal, 153 Protosilicates, 67 Segregation, 154 Shaku-do, 289 Pugin, 108 Pulverised, 317 Shepherd, 109, 224 Sheet-brass, 140 Pure gold, preparation of, 382 corrosion of, 144 silver, preparation of, 391 Shibu-ichi, 289 Pyrolusite, 29, 266 Pyromorphite, 19 Ship bolts, 142 Shot metal, 349 tower, 350 Quartz, 75 Sideraphite, 452 Queen's metal, 334 Sieward, 43S Quenching, 89 Silica, 10, 73 Quicksilver, 17 bricks, 80 Silicates, 67, 69, 73 Raoult, 45 Silicon, 5, 10 Realgar. 22 -bronze, 261 Recalescence, 429 -copper, 264 Red brass, 152 fluoride, 70, 72 gold, 373 Silver, 1-14 lead, 71 alloys, 391 -short, 25 amalgams, 358 Reducing agent, 8 and aluminium, 398 atmosphere, 302 antimony, 3t>4 Refined gold, 382 arsenic, 394 Refractory materials, 75 bismuth, 395 Regenerators, 180 cadmium, 408 Rcgulus of antimony, 21 copper, 399 Re-heating, 115 iron, 397 Resilience, 269 lead, 3i<7 Resistance, 282 nickel, 397 Reverberatory furnace, 75, 181, 188 phosphorus, 406 468 MIXED METALS Silver and tin, 395 zinc, 396 bullion, 418 chloride, 15 coins, lead in, 403 -copper alloys, 399, 401 finishing and polishing, 416 how imported, 398 in steel, 397 melting of, 314 scrap, 415 solders, 408 standards, 408 tarnish on, 210 toughening of, 383 Silverite, 303 Silveroid, 303 Siruilor, 99, 151 Simmerstach, 433 Skim-gate, 280 Slag, 67, 82 clean, 68 Slitting rolls, 189 Smaltine, 31 Smith, 298, 349 Societe Industrielle de Mulhouse, 448 Soda ash, 382 Sodium, 35 amalgam, 35, 44, 359 carbonate, 71 chloride, 71 nitrate, 72 sulphide, 349 Soft solder, 328 Soldering, 287 Solders, brass, 167 for gold alloys, 384 Solidus, 110 Solutions, 44, 46, 50 Solvent, 93, 121 Sorby, til Sorel's alloy, 166 Sound castings, 302 Sovereign, 374 Specific gravity, 6 heat, 6 Special bronzes, 253 Speculum metal, 246 Spelter, 105 Sperry, 141 Spiegel-eisen, 430 Spills, 87 Spillyness, 185, 186 Spiral, 448 Spitting, 392 Spongy gold, 382 Spontaneous annealing, 89 Spring's experiments. 96 Spuey metal, 185 Stahl and Eisen, 294 Standard gold, 374, 380 silver, 402 Stansbie, 389 Stas, 392 Statue bronze, 248 Stead, 26, 48, 63 Steel, 3, 26, 427 and aluminium,'443 cleaning of, 211 and nickel, 430 -nickel alloys, tests of, 432 conductivity of, 433 magnetism of, 434 Stereotype, 348 Sterling silver, 402 Sterro-metal, 158 Stibnite, 21 Stirling, 240, 205 Stirrers, 380 Stoddard, 397 Storer, 118, 404 Stourbridge bricks, 80 Strip castings, 184 Stroh, 439 Strontia, 34 Strontium, 34 Sullage, 246 Sulphides, 70 Sulphur. 12 dioxide, 12 removal of, 70 trioxide, 12 Sulphurous, 3 acid, 381 Surface tension, 49 Sweating, 383 Symbols, 36 Syphon flue, 184 Table bells, 243 Talmi gold, 151 Tamman, 318 Tarnish, 120, 389 Taylor, 377 Telephone wire, 262 Temperature, 5 of solidification,) 45 Tempering, 29, 89, 326 alloys, 327 Tenacity, 6 Tenner, 410 Ternary alloys, 39, 439 Tetmajer, 162, 288 Thermometers, 353 Thomson, 434 Thurston, 135 Tiers-argent, 398 INDEX 469 Tilden, 144 Waldo Tin, 320 Warne's alloy, 414 alloys, 320 Watch springs, 399 amalgam, 353 cleaning of, 212 Water, 9 -of- Ayr stone, 416 and antimony, 331 Webster, 287 bismuth, 337 Wedding, 61 lead, 324 Wedding-rings, 363 zinc, 3-20 Weiller, 261 sulphide, 21 Tinmen, 329 Welding, 23 Wet colouring, 386, 389 Tinning, 310 White alloys, 413 Tinstone, 21 brass, 167 Tissier, 275, 399 Titaniferous iron ore, 438 White copper, 303 gold, 373 Tombac, 151 metal, 257, 296, 331 Tookey, 380 bearings, 300 Torching, 197 Whitening, 417 Tough gold, 382 Wiglev, 366 Toughness, 6 Willis', 441 Tourney's alloy, 151 Wingham, 441 Trial plate, 364, 384 Wire brass, 140 Tungsten, 28 strip, 184 steel, 439 Wood grain, 291 Turner, 82, 109 Wood's alloy, 323 Tutania, 334 Wortley bricks, 80 Type metal, 346 Wright, 45 Types, 348 Yellow metal, 142 Uchatius, 260 Union by agitation, 81 Zeiss, 456 United States Mint, 363 Zinc, 31 Uranium, 29 acid dips for, 209 aluminium, 456 Van der Ven, 262 amalgam, 353 Vanadium steel, 447 blende, 32 Vaporisation, 81 fnme, 183 Vegetating, 392 and arsenic, 32 Violle, 399 phosphorus, 32 Virginia silver, 303 Viscid slag, 383 Vogel, 417 Volatile metals, 81 oxide, 31, 74, 171, 184 sulphide, 32 -tin alloys, 320 Ziscon, 456 Volatilisation, 3, 22, 301 Zisium, 456 THE END Printed by R. & R. CLARK, LIMITED, Edinburgh. WORKS BY ARTHUR H. HIORNS A TEXT-BOOK OF ELEMESITABY MEIAL- LFBGY. Second Edition. Globe 8m 3*. PRACTICAL MEIALLUBGY AND ASSAYING. Second Edition. Globe 8m &L MIXED METALS OR METALLIC ALLDYSL Globe 8m fe. IBON AND STEEL MLOTUFACTURE. Gtobe 8ro. 3&6dL METAL OOLOHRIXG AND BfiOMZOTG. Globe 8ra 5& PBISTdPLES OF METALLUBGY. GfobeSva. 60. QUESTIONS ON METALLUBGY. Globe 8*0. Sewed. Is. METALLOGRAPHY. Globe 8ra 6s. STEEL A3SID IBOX FOB ADVANCED STUD- ENTS. Globe 8m 10s. 6d. AND CO, Lm, LOXDOX. 1 WORKS ON METALLURGY A HANDBOOK OF METALLUKGY. By Dr. CARL SCHNABEL. Translated by Professor H. Louis. Illustrated. Second Edition. Medium 8vo. Vol. I., 25s. net. Vol. II., 21s. net. A TKEATISE ON OEE DEPOSITS. By J. ARTHUR PHILLIPS, F.R.S. Illustrated. Second Edition. Revised by Prof essor H. Louis. 8vo. 28s. GOLDFIELDS OF AUSTEALASIA. By K. SCHMEISSER. Translated and edited by Professor H. Louis. Illustrated. Medium 8vo. 30s. net. THE WITWATEESEAND GOLDFIELDS BAN- KET AND MINING PEACTICE. By S. J. TRUSCOTT, F.G.S. With numerous Plans and Sections. Second Edition. Super Eoyal 8vo. 30s. net. GOLD MINES OF THE EAND. By FREDERICK H. HATCH (Mining Engineer) and J. A. CHALMERJS (Mining Engineer). With Maps, Plans, and Illus- trations. Super Eoyal 8vo. 17s. net. A HANDBOOK OF GOLD MILLING. By Professor H. Louis. Third Edition. Illustrated. Crown 8vo. 10s. net. CANADA'S METALS. A Lecture. By Sir W. C. EGBERTS- AUSTEN, F.E.S. 8vo. 2s. 6d. net. MACMILLAN AND CO., LTD., LONDON. 2 WORKS ON GEOLOGY AND MINERALOGY By Sir ARCHIBALD GEIKIE. TEXT-BOOK OF GEOLOGY. Illustrated. Fourth Edition. Two vols. Medium 8vo. 30s. net. CLASS-BOOK OF GEOLOGY. Illustrated. Fourth Edition. Crown 8vo. 5s. A PRIMER OF GEOLOGY. Illustrated. Pott 8vo. Is. GEOLOGICAL SKETCHES AT HOME AND ABROAD. Illustrated. 8vo. 10s. 6d. OUTLINES OF FIELD GEOLOGY. Illustrated. Globe 8vo. 3s. 6d. THE SCENERY OF SCOTLAND. Viewed in connection with its Physical Geology. Third Edition. Crown 8vo. 10s. net. THE ANCIENT VOLCANOES OF GREAT BRITAIN. Illustrated. Two vols. Super Royal 8 vo. 36s.net. THE FOUNDERS OF GEOLOGY. Second Edition. 8vo. 10s. net. MACMILLAN AND CO., LTD., LONDON. 3 WORKS ON GEOLOGY AND MINERALOGY THE SCENERY OF SWITZERLAND AND THE CAUSES TO WHICH IT IS DUE. With numerous Plans and Illustrations. By LORD AVEBURY. Fourth Edition. Crown 8vo. 6s. THE SCENERY OF ENGLAND AND THE CAUSES TO WHICH IT IS DUE. By LORD AVEBURY. Illustrated. Fifth Edition. Crown 8vo. 6s. THE GEOLOGY OF NOVA SCOTIA, NEW BRUNSWICK, AND PRINCE EDWARD ISLAND ; OR ACADIAN GEOLOGY. By Sir J. W. DAWSON, J.L.D., F.R.S. Fourth Edition. 8vo. 21s. GEOLOGY AND GEOGRAPHY OF NORTHERN NIGERIA. By J. D. FALCONER, D.Sc. With Notes by the late ARTHUR LONGBOTTOM, B.A., and an Appendix by HENRY WOODS, M.A. Illus- trated. 8vo. 10s. net. PHYSICS OF THE EARTH'S CRUST. By the Rev. OSMOND FISHER. Second Edition. 8vo. 12s. MACMILLAN AND CO., LTD., LONDON. 4 WORKS ON GEOLOGY AND MINERALOGY OBSERVATIONS OF A NATURALIST IN THE PACIFIC BETWEEN 1896 AND 1899. By H. P. GUPPY, M.B. Vol. I. Vanua Levu, Fiji : Its Physical and Geological Characters. Illustrated. 8vo. 15s. net. Vol. II. Plant Dispersal. 8vo. 21s. net. GEOLOGY OF SOUTH AFRICA. By Dr. F. H. HATCH and Dr. G. S. CORSTORPHINE. Illustrated. Second Edition. 8vo. 21s. net. CHARACTERISTICS OF EXISTING GLACIERS. By Prof. WILLIAM HERBERT HOBBS. Illustrated. 8vo. 13s. 6d. net. GEOLOGY AND GENERAL PHYSICS. By LORD KELVIN. Crown 8vo. 7s. 6d. A TREATISE ON ROCKS, ROCK-WEATHER- IXG, AND SOILS. By Prof. GEORGE P. MERRILL. 8vo. 17s. net. MINERALOGY. An Introduction to the Scientific Study of Minerals. By Sir H. A. MIERS, F.R.S. 8vo. 25s. net. MACMILLAN AND CO., LTD., LONDON. 5 WORKS ON GEOLOGY AND MINERALOGY THE AGE OF MAMMALS IN EUROPE, ASIA, AND NORTH AMERICA. By Prof. HENRY F. OSBORN, LL.D., D.Sc. Illustrated. 8vo. 19s. net. ECONOMIC GEOLOGY. With Special Reference to the United States. By Professor H. RIES. New and revised edition. Illustrated. 8vo. 15s. net. VOLCANOES OF NORTH AMERICA. By I. C- RUSSELL. 8vo. 17s. net. AN INTRODUCTION TO GEOLOGY. By Professor WILLIAM B. SCOTT. Second Edition. Crown 8vo. 11s. net. CRYSTALLOGRAPHY AND PRACTICAL CRYSTAL MEASUREMENT. By A. E. H. TUTTON, D.Sc., F.R.S. Illustrated. 8vo. 30s. net. ELEMENTS OF CRYSTALLOGRAPHY FOR STUDENTS OF CHEMISTRY, PHYSICS, AND MINERALOGY. By GEORGE HUNTINGDON WILLIAMS, Ph.D. Crown 8vo. 6s. TEXT-BOOK OF PALAEONTOLOGY. By KARL A. VON ZITTEL, Professor of Geology and Palaeon- tology at Munich. Translated and edited by CHARLES R. EASTMAN, Ph.D. Revised and enlarged from the German original. Medium 8vo. Vol. I. 25s. net. Vol. II. 10s. net. MACMILLAN AND CO., LTD., LONDON.