XI b RAR.Y 
 
 OF THE 
 UN1VLR.SITY 
 Of ILLINOIS 
 
 666 
 If 6 
 He. 16-23 
 
 J
 
 f»]|^gELiSl 
 
 CENTRAL CIRCULATION AND BOOKSTACKS 
 
 The person borrowing this material is re- 
 sponsible for its renewal or return before 
 the Latest Date stamped below. You may 
 be charged a minimum fee of $75.00 for 
 each non-returned or lost item. 
 
 Th«ft, mulilaHen, or defacement of library maferlali con be 
 cauMt for $fudenf dUcIpllnary ocHon. All maferiali owned by 
 th« Unlvenily of lllinoli library are the property of the State 
 of Illinois and are protected by ArtleU 16B of llllnoit Criminol 
 law and Procedure. 
 
 TO RENEW, CALL (217) 333-8400. 
 University of Illinois Library at Urbana-Champaign 
 
 MAY 2 9 2001 
 
 When renewing by phone, write new due date 
 below previous due date. L162
 
 UNIVERSITY OF ILLINOIS BULLETIN 
 
 Vol. X. 5LPTLMBLR 23. 1912. No. 4 
 
 [E-ntered February 14, 1902, at Urbana, Illinois, as second-class matter under 
 Act of Congress of July 16, 1894.] 
 
 BULLLTINNo. 18 
 DLPARTMLNT OF CLRAMIC5 
 
 A. V. BLLININGLR. Director 
 
 A THERMAL 5TUDY OF BORIC ACID-51LICA 
 
 MIXTURES 
 
 BY 
 A. V. BLLININGLR AND PAUL TLLTOR 
 
 THE REPLACEMENT OF TIN OXIDE BY ANTIMONY 
 OXIDE IN ENAMEL5 FOR CA5T IRON 
 
 BY 
 
 R. L. BROWN 
 
 191 1-1912 
 
 PUBLISHED rORTNIGHTLY BY (HL UNIVERSITY
 
 [Reprinted from Transactions American Ceramic Society, Vol. XIV, 
 BY Permission.] 
 
 A THERMAL STUDY OF BORIC ACID-SILICA MIXTURES. 
 
 By A. V. Bleixixger and Pali. Teetok, Urbana, 111. 
 
 The subject of possible chemical combinations of silica 
 and boric acid has received some attention in our Transactions,'-^ 
 and the question raised is interesting inasmuch as such mixtures 
 possess most decidedly the character of glasses or solid solutions. 
 Thermal analysis thus does not promise a fruitful field of in- 
 vestigation. However, of the two methods com])rising thermal 
 analysis, the determinations of the softening temperatures is of 
 some interest in itself, since it gives us the general character 
 of the fusion curve of the two components involved. 
 
 A thermal lag is not to be expected either in the heating 
 or cooling curves. In the present work, a search was made, how- 
 ever, for such a point based on the statement of Binns, Trans. 
 
 A. C. S., X, p. 158, in which he records a temperature increase 
 upon the fusion of a mixture of boric acid and silica, due to 
 some exothermal change. The present research deals, (a) with the 
 determination of the softening points of Si02-B203 mixtures be- 
 tween the limits Bp.,-B.O.j.3Si02, (b) with the determination 
 of heating and cooling curves and (c) with an investigatio^i of the 
 solubility of the fused glasses in water. The reagents used were 
 chemically pure hydrous boric acid and silica, the latter being a 
 
 B. & A. preparation which unfortunately contained several per 
 cent, of sodium chloride and water. In the latter part of the 
 series, fusions were made also with fiint which had been passed 
 through a 200 mesh sieve. The calculation of the mixtures 
 was based upon the analyzed silica content, practically 97 per 
 cent. The boric acid was fused, cooled rapidly and kept in a 
 desiccator. It was crushed in a porcelain and pulverized in an 
 agate mortar. Similarly, the silica was ignited and kept in a 
 desiccator. The mixtures were ground together in the agate 
 mortar and fused over the blast lamp in a 10 cc. platinum crucible 
 kept covered during the heating. After some time, the yellow 
 color of the mass disappeared, which seemed to be a measure 
 of the completeness of the fusion. The cooled mass had an opaque 
 
 > Binns, Trans. A. C. S., Vol. X, p. 158. 
 2 Singer, Trans. A. C S., Vol. XI. p. 676.
 
 4 A THERMAL STUDY OF BORIC ACID-SILICA MIXTURES. 
 
 but glassy appearance. The fused mixture was easily removed 
 from the crucible by inserting a platinum rod and quickly cooling 
 in cold water. Then the fusion was pulverized and screened 
 through So and 150 mesh screens. The portion passing the 80 
 but remaining on the 150 mesh screen was used for the solubility 
 samples. This was done in order that no great variations in 
 the surface factor might affect the solubility of the several mix- 
 tures. 
 
 SOFTENING POINT DETERMINATION. 
 
 For this purpose, the fused mixtures of SiO^ and B2O3, 
 ground to a fine powder, were made up with a little water into 
 small cones, and placed in an electric resistance furnace. The 
 specimens were kept in position by means of platinum foil. Since 
 in glasses practically no other criterion is available than the 
 deformation point, the temperature at which the cones bent was 
 taken to represent the softening point. Care was taken to raise 
 the heat at a regular rate by rheostat regulation, and the tempera- 
 ture readings were made by means of a Pt-PtRd thermo-couple, 
 the electromotive force of which was determined by the method 
 of balancing against a standard cell by means of a potentiometer 
 indicator. 
 
 Owing to the fact that by mistake water was used in making 
 up the mixtures, some anhydrous boric acid reverted to the 
 hydrous form. This, of course, made it troublesome to determine 
 the deformation point of B2O3 owing to the evolution of steam 
 and the resulting bubbling. With the addition of o.i SiOj, 
 the cones seemed to stand up apparently in good shape. The 
 heat given off on adding water to the 6203.1.58102 mixture 
 was so great that the crucible could not be held in the hand. 
 At the same time very little heat was evolved by the 620.5.1.48102 
 and the 6203.1.68102 glasses. On fusing the 6303.1.58102 mix- 
 ture, it assumed a pink color. 
 
 It was soon observed that these glasses were quite viscous. 
 This was illustrated by the fact that a twisted platinum wire on 
 being lowered into the fused mass and again raised was found to 
 draw out a ribbon of glass. It is not surprising, therefore, that 
 the softening points could not be checked, in spite of the fact 
 that the same rate of heating was followed as closely as possible.
 
 A THERMAL STTDY OF IJOKIC ACID-SIl.ICA MIXTl'RUS. 5 
 
 /^ca 
 
 Tf?y4/VS.AM CE/i'.SOC. ^OL.X/^^ /='/0. /. 3L£/M/VO£/?& TEETO/^ 
 
 ^^ 
 
 o o.a ^.^ 0.6 0.3 /.o /.e /.^ /.<$ /ff ^^ 
 
 A^OLECaLES 0ES/02
 
 6 A THERMAL STUDY OF BORIC ACID-SILICA MIXTURES. 
 
 In going over this part of the work four times, the results shown 
 in Fig. I were obtained. The softening points of the mixtures 
 beyond BjOj.aSiOj are not plotted since the divergence in this 
 part of the series is still greater. 
 
 Softening point determinations were also made upon rods 
 drawn from all of the fusions but these likewise gave extremely 
 variable results, considerably lower than those obtained for the 
 cones. 
 
 The evidence thus far collected makes it apparent that any 
 reaction taking place under these conditions would be greatly 
 hindered by the internal molecular friction. 
 
 HEATING AND COOLING CURVES. 
 A considerable number of heating and cooling curves were 
 determined with special reference to the 6203.28100 mixture. 
 The latter was prepared from fused B2O3 and prepared Si02, and 
 from fused boric acid and flint, passed through the 200 mesh 
 sieve. In no instance was there a temperature acceleration or 
 lag observed, and, hence, the observation of Binns was not checked. 
 In Fig. 2, the heating curve in which the couple readings were 
 made by means of a potentiometer indicator is shown. The 
 junction was kept at 0° C. by means of ice. In Figs. 3 and 4, 
 both the heating and cooling curves for prepared silica and flint 
 mixtures as indicated by a Siemens and Halske recorder, making 
 a contact every 16 seconds, are presented. It was observed 
 that on fusing an}' mixture of B^Oj and SiO^, imthout previous 
 fritting, some vapor was expelled suddenly, carrying evidently 
 a certain amount of boron. This happened also when both the 
 boric acid and silica had been ignited separately to constant 
 weight before mixing. Since Professor Binns used an optical 
 pyrometer, it is quite possible that by focusing upon this vapor 
 the readings were changed as observed by him. 
 
 SOLUBILITY DETERMINATIONS. 
 The diflerent mixtures were fused and pulverized on cooling. 
 The resulting powder was screened through the 80 and 150 mesh 
 sieve. All material coarser than the 80 and finer than the 150 
 mesh was rejected. One gram samples were then weighed and 
 put in stoppered 250 cc. Erlenmeyer flasks. These were placed
 
 A THERMAL STUDY OF BORIC ACID-SILICA MIXTURES. 7 
 
 
 T^.4/VS. y^M CS/f. SOC. yVL 
 
 ^/^ 
 
 E/G.a. 
 
 
 ffI.£/Af//VO£ff A T£6 
 
 'ro/? 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 / 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 J 
 
 / 
 
 / 
 
 
 M/xr(/fP£ U5/A/0 PVrEA/r/OMETSR 
 
 'O 
 
 
 
 
 / 
 
 / 
 
 
 
 
 / 
 
 r 
 
 / 
 
 / 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 J 
 
 / 
 
 / 
 
 
 
 
 
 
 
 
 
 
 
 
 
 / 
 
 / 
 
 
 
 
 d 
 
 
 
 
 
 
 
 
 
 J 
 
 J 
 
 / 
 
 
 
 
 
 
 
 
 
 
 
 
 / 
 
 / 
 
 / 
 
 
 
 
 
 / 
 
 
 
 
 
 
 
 
 / 
 
 / 
 
 f 
 
 
 
 
 
 
 
 
 
 
 
 
 / 
 
 r 
 
 / 
 
 
 
 
 
 
 
 
 
 
 
 
 / 
 
 / 
 
 / 
 
 
 
 
 
 
 
 
 
 
 
 
 
 A 
 
 A 
 
 
 
 
 
 
 
 
 r 
 
 
 
 
 
 / 
 
 / 
 
 / 
 
 
 
 
 
 
 
 
 
 
 
 
 i 
 
 / 
 
 / 
 
 
 
 
 
 
 
 
 
 ^ 
 
 
 
 
 / 
 
 / 
 
 
 
 
 
 
 
 
 
 
 
 
 
 / 
 
 t- 
 
 / 
 
 / 
 
 
 
 
 
 
 
 
 
 
 J 
 
 
 1 
 
 / 
 
 / 
 
 
 
 
 
 
 
 
 
 
 
 
 / 
 
 / 
 
 
 
 
 
 
 
 
 
 
 
 c^ 
 
 / 
 
 f 
 
 f 
 
 
 
 
 
 
 
 
 
 
 
 
 
 /. 
 
 -r 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 n 
 
 / 
 
 
 
 
 ^/7 
 
 . 
 
 10 
 
 A 
 
 t<y 
 
 
 TC 
 
 t 
 
 ?^7 
 
 ro 
 
 T/A^£ //V M/NC/TSS
 
 ! A THERMAL STUDY OF BORIC ACID-SILICA MIXTURES. 
 
 T/fAAfSAM.Ce/i'.SOC.VOL.X/^ r/CP.3. BLEW/NOER&TEEW/? 
 
 Tff^A/S. AM. C£:/?.SOC \/OL X/y /7^ <? 0L£//V/A^(5£/r'A T££m/?
 
 A THERMAL STUDY oF IloKIC ACIDSIl.ICA MIXTl'RES. 9 
 
 in a shaking machine together with 200 cc. of distilled water. 
 After shaking for 10 hours and standing over night, the liquid 
 was filtered olT and the residue washed. Then 200 cc. of water 
 was again added (including the wash water used) and the flasks 
 were again shaken for one hour. The residue was again filtered 
 and washed. After drying, the residues and paper were ignited 
 and the weights determined. The insoluble matter in each case 
 was brushed off and the pajier burnt separately. 
 
 The weights of the residues are shown in the curve of Fig. 5. 
 
 
 rfi'/^//S.^A^. C£/? ^OC. ^^^.X/l^ 
 
 /^-/c^.^. 
 
 Bi.£/A//A/Gs/ps. rsEro/? 
 
 
 \ 
 
 s^ 
 
 ^ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 -A— 
 
 \ 
 \ 
 
 
 
 ^ 
 
 
 
 
 
 
 
 
 
 
 
 
 1- 
 
 
 \ 
 
 
 
 \ 
 
 
 
 
 
 
 
 
 
 
 
 i 
 
 
 
 
 
 \ 
 
 \ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 •^ 
 
 ^^ 
 
 
 ^^ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 '-C~^ 
 
 ^ 
 
 ■"=*-> 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 ^ 
 
 "^ 
 
 K' 
 
 ~-~ 
 
 - 
 
 
 
 
 
 
 
 
 
 
 
 
 
 \ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 \ 
 
 ^-^ 
 
 V 
 
 X 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 > 
 
 
 
 
 2 
 
 ^ £> 
 
 6 
 
 e / 
 
 / 
 
 a /.. 
 
 <t / 
 
 ? / 
 
 ^. 
 
 c ^., 
 
 ? ^ 
 
 ^ ^. 
 
 <s ^. 
 
 a ^o 
 
 It is seen that all of the boric acid was dissolved in the first of 
 the series as well as some of the silica. At a point close to 
 B.,03.i.8SiO., however, the solubility curve crosses the line 
 indicating the percentage content of B/)., (shown by the dotted 
 line) which proves that some BjO^ has been rendered insoluble. 
 This tendency increases with the silica content. At the molecular 
 ratio 1:2a decided drop occurs indicating the rapid formation of 
 an insoluble glass. As to the cause of this sudden change, we 
 can only conjecture at the present time. To establish the fact 
 that a chemical combination has taken place would require the
 
 lO A THERMAL STUDY OF BORIC ACID-SILICA MIXTURES. 
 
 rimning of a parallel series using another criterion such as the 
 specific gravity of the powder. At the present time, however, 
 this evidence might be used to support the claim that we are 
 dealing here with a solid solution representing a chemical union 
 between the silica and the boric acid. 
 
 CONCLUSIONS. 
 
 Fused boric acid-silica mixtures are typical glasses pos- 
 sessing no definite deformation temperatures. 
 
 No thermal phenomena were observed, i. e., there was no 
 absorption or liberation of heat throughout the series B^Og to 
 BA-3SiO,. 
 
 Boric acid when fused with silica first dissolves some of 
 the latter. The amotmt of matter solulple in water decreases 
 somewhat more rapidly than the B^Og content. Between 2 and 
 2.2 SiO, a decided drop in the solubility of the glasses occurs, 
 indicating that some B2O3 has been rendered insoluble. It is 
 quite probable, although not proven, that a chemical combina- 
 tion might take place at this point. The gradual decrease in 
 solubility might thus be ascribed to the formation of some of 
 this combination at an earlier stage. At the point mentioned, 
 a more rapid enrichment would thus take place. 
 
 DISCUSSION. 
 
 Professor Binns: I am interested in the results secured by 
 Professor Bleininger. When I made the first experiments of 
 this kind I expressly disclaimed any pretension to the ability 
 called for by such work, and I based my chief claim as to the 
 action of boron upon the practical issues as expressed in glaze 
 composition. From this position, which has been confirmed 
 by Dr. Singer and ^Ir. Stull, I have not had cause to retreat.
 
 [Reprintkd from Transactions American Cbramic Societv, \\)L. XIV. 
 
 nv Pkrmission ] 
 
 THE REPLACEMENT OF TIN OXIDE BY ANTIMONY OXIDE 
 IN ENAMELS FOR CAST IRON.' 
 
 By R. E. Brown, Mt. Savage, Md. 
 
 INTRODUCTION. 
 
 Classification of Opacifiers. — Opacifiers for the purposes of 
 this work are divided into two classes: (i) partial opacifiers, 
 and (2) absolute opacifiers. 
 
 In the first class are included bone ash, fluorite, cryolite, 
 and silica. Bone ash is rarely employed in enamels, but the re- 
 maining three, especially the silica, are invariably used. Fluorite 
 and cr^'olite are advantageous, commercially, both from the 
 standpoint of their low fusibility and their fluorine content. 
 The latter gives them the property of acting as weak opacifiers, 
 thereby decreasing the amount of absolute opacifier needed.^ 
 The silica, as is showTi by the following work, has no opacifying 
 tendencies in itself in this type of glasses but emphasizes and in- 
 creases the opacity brought about by certain opacifiers proper. 
 
 In the second class, the opacifiers, per sc, are arsenious oxide, 
 zirconium oxide, tin oxide and antimony oxide. Arsenious 
 oxide finds a very limited use, being employed only for decorative 
 work on jewelry and art ware. Zirconium oxide is not widely 
 used; while some consider it too expensive for commercial 
 work, others regard it a cheap opacifier as a substitute for tin 
 oxide. ^ Tin oxide is by far the most widely used of the opacifiers, 
 and is employed not only in the enameling of sheet iron and cast 
 iron, but in the enameling of clay products as well. 
 
 Antimony oxide has had, to date, only an extremely limited 
 use as an opacifier. It has been used to some extent in Ger- 
 many in conjunction with zinc oxide as a substitute for tin oxide.* 
 It has been used in this country to some extent, and in one case 
 its use in conjunction with other ingredients is patented as "a 
 substitute for tin oxide. "^ In a "Note on White Antimony 
 
 ' Abstract of a thesis fulfiUins part of the requirements for the detrree of Hachelor of 
 Science in Ceramics, University of Illinois. 
 
 - .Mayer and Havas. Sprechsaal, XLII. 460-461. 
 
 3 La Ceramique. 11, 100-101. Keram Rundschau, XVI. 89-91. 135-139. 
 
 * Ph. Eyer, Stahl und Eisen. XVIII, 1097. 1099. 
 
 5 U. S. Patent, 932,839.
 
 12 REPLACEMENT OF TIN OXIDE IN ENAMELS FOR CAST IRON. 
 
 Enamels," Bock points out the dangers of the use of antimony 
 oxide in enamels for cooking vessels, but no mention is made of 
 its employment in enamels for cast iron." 
 
 The most expensive constituent of the ordinary commercial 
 enamel is the opacifying agent, tin oxide, both by reason of the 
 high market price and the quantity employed. The prices of 
 antimony oxide and tin oxide in barrel lots are at present 10V2 
 and 45 cents per pound, respectively. 
 
 Object of Work. — It was with a view to using the cheaper 
 opacifier for cast iron enamels that this work was undertaken. 
 The work attempts to determine, in a practical way, the con- 
 ditions under which antimony oxide may be used in enamels. 
 
 EXECUTION OF WORK. 
 
 Preparation of the Enamels. — In carrying out a series the 
 ingredients of the two extremes of a series were weighed up and 
 mixed thoroughly by passing 5 or 6 times through a 20 mesh 
 screen. The batch was next put into a small fare clay crucible 
 (capacity about 250 grams of the raw batch) and fritted in a pot 
 furnace, fired by a blast lamp using artificial gas and compressed 
 air as fuel. After melting and becoming relatively free from 
 bubbles the contents were poured into water. It was then dried, 
 again put into the crucibles and refritted, care being taken in 
 this second fritting to prolong the heating to such length of time 
 that no bubbles were given off. As soon as bubbles had ceased 
 to form, the contents were again quenched by pouring into water. 
 The shattered glass was dried and then ground in 8 inch, porcelain, 
 ball mills so as to pass a 200 mesh sieve. 
 
 The two extremes of a series, now in a powdered form, 
 were weighed out in the proper proportions to form the inter- 
 mediate enamels. These were now mixed by rubbing 5 or 6 
 times through a 60 mesh sieve. 
 
 The Ingredients. — ^The raw ingredients employed in intro- 
 ducing the following constituents with the molecular weights 
 used were as follows : 
 
 ZnO: Introduced as zinc oxide (81). 
 
 PbO : Brought in as red lead Fhfi, (685) . 
 
 schem. Ztg., XXXII, 516-517.
 
 REPLACEMENT OF TIN OXIDE IN ENAMELS FOR CAST IRON. 1 3 
 
 BaO: Where employed, was brought in by barium car- 
 bonate (197). 
 
 CaO: Brought in by lluorite (78), whiting (100), and 
 hydrated lime (74). 
 
 Na.,0: Introduced as sodium carbonate (io6), borax (382), 
 and cryolite (210). 
 
 K^O: Brought in by potassium nitrate (loi) and potash 
 feldspar (557). 
 
 MgO: Brought in by magnesium carbonate (84) and 
 magnesium oxide (40). 
 
 AI2O3: Potash spar (557) was used as the main source of 
 alumina. Small amounts of cryolite, NaaAlFg (210) were also 
 used. 
 
 B2O3; This was brought in by using borax (3S2) in the cover 
 enamels and as borax and boric acid (62) in the ground coats. 
 
 SnO, : Brought in as tin oxide (150). 
 
 Sb^Og: Introduced as white oxide of antimony (288). 
 
 SiOa: Brought in as potash feldspar (557), and as flint (60). 
 
 In addition to the above ingredients, ammonium carbonate 
 was used in some of the enamels of the later series. This 
 volatilizes readily and serves as a clarifier of bubbles in the glass 
 during fritting. 
 
 Trial Pieces. — The trial pieces were small circular discs 
 Vs" thick and 2" in diameter with a raised center. The iron 
 used for casting these trials did not prove to be of a very satisfac- 
 tory grade as it frequently produced large bubbles or blisters in 
 the enamels, probably due to the sulphur content of the iron. 
 
 The trials were cleaned by pickling for 20-30 minutes in a 
 dilute solution of hydrochloric acid so as to remove the scale 
 and any oxide present. After this they were washed and scrubbed 
 and then dipped in a dilute solution of sodium carbonate so as to 
 neutralize all of the acid. They were then scrubbed and washed 
 again, the surface water was wiped ofif, and the trials were put. 
 into a warm oven. Even with this seemingly thorough treat- 
 ment, the coat of carbon (left by dissolving the iron) was not; 
 entirely removed, and hence gave rise to bubbles during the burn- 
 ing process. Another method of cleaning the iron, used in the 
 latter part of the work, proved very effective. In this, the iron
 
 14 REPLACEMENT OF TIN OXIDE IN ENAMELS FOR CAST IRON. 
 
 was pickled as before and then put into a ball mill with sand and 
 water; thus all of the carbon was effectually removed with the 
 result that less trouble as regards bubbling was experienced. 
 
 The ground coat was applied to the trials by dipping, care 
 being taken to secure a thin coat that was as uniform as possible. 
 
 Composition of the Ground Coat. — The ground coat chosen 
 was of the composition shown : 
 
 0.30 K,0 ] 
 
 0.24 Na,0 I0.297AUO3 
 
 [-2.19 SiOa 
 0.31 BA j 
 
 o. 159 CaO 
 0.219 MgO 
 0.081 PbO 
 
 Batch weights: 15 flint; 30 potash feldspar; 10 boric acid; 
 5 KNO3; 5 PbgO,; 2.2 Ca(0H)2; 2.3 MgO; lo.o cryolite; i.o 
 fluorite. 
 
 The trial piece, after being slush coated and dried, was 
 placed in the furnace, which had been heated to the temperature 
 of burning. As soon as the iron had attained a sufficient heat 
 for fusing the ground enamel, it appears to "melt" not unlike 
 ^ coat of frost or snow. This commences at one spot and soon 
 lias extended over the whole trial. The trial now has a glossy 
 appearance. Almost coincident with this, innumerable bubbles 
 are formed which "break" at once. This continues for a short 
 time, after which the glossy coating or glass smooths down. 
 It is at this point, in the writer's judgment, that the burning of 
 the ground coat is complete. If the trial is not removed at once, 
 larger bubbles are formed, but the glass is then decidedly more 
 viscous than when the preliminary bubbles were formed. When 
 they break, if they break at all, they leave a rough slag on the 
 surface of the trial as is shown by cooling the trial piece. 
 
 Burning the Enamels. — ^The furnace used for the enameling 
 was of the open-fired type, i. e., without a muffle chamber, and 
 was fired by the use of artificial gas and compressed air as fuel. 
 Although this is not the type of furnace best suited for this kind 
 of work, it was the only one available and there was no time for 
 the construction of a more suitable one. The temperature of 
 burning was measured by a Le Chatelier pyrometer, the couple 
 of which was inserted so that its junction point was at the side
 
 REPLACEMEXT OF TIN OXIDE IN ENAMELS FoR CAST IRON. 15 
 
 of the trial. The holder for the trial piece consisted of a bar of 
 iron "upset" at one end and so shaped as to fit on the inside of 
 the trial piece. 
 
 Series I. 
 
 REPLACEMENT OF TIN OXIDE HV ANTIMONY OXIDE IN A TIN 
 
 ENAMEL. 
 
 This series was carried out by replacing the tin oxide in an 
 enamel similar to that given by Riddle in his "Types of Enamels 
 for Enameling Cast Iron Sanitary Ware," Trans. A. C. S., Vol. 
 IX, with antimony oxide, thus: 
 
 0.25 Na^O 1 
 
 o. 10 MgO I 
 
 0.15 PbO jo.isAIAl i.ooSiO. 
 
 0.15 BaO i 
 
 0.15 KjO 0.20 B2O3 J o. 15 SnO^-o.oys Sb^Oj 
 
 0.05 ZnO 
 
 0.15 CaO 
 
 Batch formulae (in equivalents). 
 
 cti 
 
 n 
 
 o 
 
 N 
 
 PQ 
 
 Ui 
 
 VI 
 
 tfl 
 
 2 
 
 3 
 4 
 5 
 
 0.15 
 0.15 
 0.15 
 0.15 
 0.15 
 
 0.05 
 0.05 
 0.05 
 0.05 
 0.05 
 
 0.15 
 0.15 
 0.15 
 
 0.15 
 0.15 
 0.15 
 
 0.150. 15 
 o. 150.15 
 
 0.05 
 0.05 
 0.05 
 0.05 
 0.05 
 
 O.IO 
 
 o. 10 
 o. 10 
 o. 10 
 o. 10 
 
 o. 10 
 o. 10 
 o. 10 
 o. 10 
 o. 10 
 
 0.15 
 0.15 
 
 0.15 
 0.15 
 0.15 
 
 o. 10 
 o. 10 
 o. 10 
 o. 10 
 o. 10 
 
 0.15 
 O.II25 
 0.075 
 0.0375 
 
 0.00 
 
 0.00 
 
 0.0188 
 
 0.0375 
 
 0.056 
 
 0.075 
 
 Description of Trials. — No. i is a good enamel and is a 
 typical tin enamel. All of the rest of the series, with the possible 
 exception of No. 2, are very poor. A peculiar "puckery" or 
 matte texture exists, the surface is rough and uneven, and the 
 enamel flies off in patches, resembling shivering of clay ware. 
 In the trials of enamels 3 and 4, the "puckery" effect is partially 
 overcome by laising the burning temperature. The shivering 
 is also lessened by this treatment. It is also evident in this 
 series as well as in the rest of the work that where an enamel is 
 applied, too thick shivering is more likely to occur. 
 
 Thinking perhaps that the "puckery" effect might be over- 
 come partially by making a more easily fusible enamel it was
 
 1 6 REPLACEMENT OF TIN OXIDE IN ENAMELS FOR CAST IRON. 
 
 decided to run a series with a more fusible RO combination. 
 It was also suggested that the "puckery" effect might be due 
 to the barium which reacted with the sulphur present in the 
 SbjOg.^ Some of the Sb203 was tested and found to contain 
 sulphur. 
 
 A series embodying the two above ideas was accordingly 
 carried out as given below. Although it was not conducted 
 strictly on a scientific basis it is sufficient to show in a practical 
 way the desired effect. 
 
 Series II. 
 
 VARIATION OF BARIUM OXIDE AND ITS EFFECT ON ANTIMONY 
 
 OXIDE. 
 0.25-0.54 NaoO 
 o. lo-o.oo MgO 
 
 o. 15-0'. 15 PbO o. 15-0. 16 AI2O3 ] i.ooSiOj 
 o . 1 5-0 . 1 6 KjO } } 
 
 0.05-0.05 ZnO I 0.20 B2O3 J 0.075 SbjOg 
 
 o. 15-0. 10 CaO 1 
 0.15-0.00 BaO J 
 
 Batch formulae (in equivalents). 
 
 d 
 
 "S 
 
 
 
 
 fa 
 
 SI 
 
 
 
 
 
 d 
 N 
 
 
 
 •z 
 
 4) 
 
 "o 
 
 >> 
 
 u 
 U 
 
 
 
 
 
 i 
 
 
 
 
 BO 
 
 % 
 
 1^ 
 
 a 
 
 S 
 
 w 
 
 I 
 
 2 
 
 3 
 
 4 
 
 0.15 
 0.21 
 0.29 
 
 0-35 
 
 0.05 
 0.05 
 0.05 
 0.05 
 
 0.15 
 0. 10 
 0.05 
 0.00 
 
 0.15 
 
 0. 10 
 
 0.05 
 0.00 
 
 0.05 
 0.05 
 0.05 
 0.05 
 
 0.00 
 0.02 
 0.04 
 0.06 
 
 0.00 
 0.02 
 0.04 
 0.06 
 
 0.00 
 0.03 
 0.07 
 0. 10 
 
 O.IO 
 
 0. 10 
 0. 10 
 
 O.IO 
 
 O.IO 
 
 0.07 
 
 0.03 
 0.00 
 
 0.15 
 0.148 
 0.136 
 0.13 
 
 0. 10 
 0. 14 
 0.18 
 0.22 
 
 0.075 
 0.075 
 0.075 
 0.075 
 
 Description of Trials. — The "puckery" effect has decreased 
 toward the end of the series which contains no barium and in 
 No. 4 is not present at all. This enamel is a fair enamel which 
 adheres well. No. i is somewhat shivered. This series shows 
 from a practical standpoint that barium should not be used to 
 any very large extent in an enamel where there is a contact with 
 sulphur gases. Its use, however, in enamels where tin is used as 
 an opacifier is very much desired, owing to its ability to decrease 
 shivering. 
 
 '' Sb203 is prepared from stibnite, Sb2S3, by roasting in air, hence sulphates are formed 
 which, if not entirely removed, would combine with the barium compounds.
 
 REPLACEMENT OF TIN OXIDE IN ENAMELS FOR CAST IRON. 1 7 
 
 Series III. 
 
 VARIATION OF THE SILICA CONTENT. 
 This series was varied between the limits of i.oo and 2.00 
 •equivalents of silica as shown : 
 
 o.ioCaO 
 o . 54 NaoO 
 0.15 Pbb 
 0.16 K,0 
 0.05 ZnO 
 
 0.16 Al A 
 
 0.20 B,Oq 
 
 1 .00-2 .00 SiOj 
 0.075 vSboOj 
 
 
 
 
 Batch formulae (in 
 
 squivalents). 
 
 
 
 
 6 
 
 2: 
 
 
 
 a 
 Z 
 
 1 ' i 
 
 
 
 V 
 
 >• 
 
 Wi 
 
 
 
 
 
 
 
 i 
 
 a 
 
 c 
 S 
 
 
 1 
 
 I 
 
 2 
 
 3 
 
 4 
 
 5 
 
 6 
 
 7 
 
 8 
 
 9 
 
 0.35 
 0.35 
 0.35 
 0.35 
 0.35 
 0.35 
 0-35 
 0.35 
 0.35 
 
 0.05 
 0.05 
 0.05 
 0.05 
 0.05 
 0.05 
 0.05 
 0.05 
 0.05 
 
 0.05 
 
 0.05 
 0.05 
 0.05 
 0.05 
 0.05 
 0.05 
 0.05 
 0.05 
 
 0.06 
 0.06 
 0.06 
 0.06 
 0.06 
 0.06 
 0.06 
 0.06 
 0.06 
 
 0.06 
 
 0.06 
 0.06 
 0.06 
 0.06 
 0.06 
 0.06 
 0.06 
 0.06 
 
 0. ID 
 
 0. 10 
 0. 10 
 0. 10 
 0. 10 
 0. 10 
 0. 10 
 0. 10 
 0. 10 
 
 O.IO 
 
 0. 10 
 0. 10 
 0. 10 
 0. 10 
 0. 10 
 0. 10 
 0. 10 
 0. 10 
 
 0.13 
 0.13 
 0.13 
 0.13 
 0.13 
 0.13 
 0.13 
 0.13 
 0.13 
 
 0.22 
 
 0.345 
 0.47 
 
 0.595 
 0.72 
 0.845 
 0.97 
 I 095 
 1 .22 
 
 0.075 
 0.075 
 0.075 
 0.075 
 0.075 
 0.075 
 0.075 
 0.075 
 0.075 
 
 Description of Trials. — Nos. i, 2 and 3 have an egg shell- 
 like texture but otherwise are fair enamels. The trials of enamels 
 Nos. 4 and 5 are better and do not show the above texture to such 
 a degree. No. 6 is a fair enamel but is a trifle dull. No. 7 is a 
 good enamel and adheres well. It is whiter and has a better 
 gloss than the average commercial enamel. Enamel No. 8 is 
 whiter than No. 7 and has a better gloss. A few of the trials 
 shiver somewhat, showing that the silica is a trifle too high. 
 Enamel No. 9 has shivered still more, but on the trials where 
 it held, it is the whitest and most brilliant of the series. Enamels 
 Nos. 8 and 9 have an exceptionally white color and are more than 
 the equal of the average tin enamel in this respect. 
 
 The result of this series seems to show that the last two 
 enamels are too high in silica and also that a silica content of 
 over 1.85 equivalents is conducive in shivering. The burning 
 temperature rises as the silica content increases; but this heat,
 
 1 8 REPLACEMENT OF TIN OXIDE IN ENAMELS FOR CAST IRON. 
 
 even with the enamels containing 2.0 SiOj, did not cause the iron 
 to deteriorate to any visible extent. As silica increases, the 
 whiteness is increased, and it is evident that a sacrifice must be 
 made of part of the whiteness in order to obtain enamels that 
 do not shiver. 
 
 Series IV. 
 
 VARIATION OF ALUMINA. 
 
 This series was run between the limits of o.i and 0.2 equiv- 
 alent of AI2O3. To bring in the AUOg in combined form, i. e., 
 as spar, it was necessary to change the RO with respect to K2O 
 and NajO thus: 
 
 o. 16-0.20 K^O 
 
 0.15 PbO 
 
 o. 10 CaO 
 
 o . 54-0 . 50 Na20 
 
 0.05 ZnO 
 
 o. 10-0.20 ALO, 1 1.80 SiO, 
 
 o . 20 B,0, 
 
 0.075 SbPa 
 
 
 
 
 Batch formulae (in equi 
 
 valents). 
 
 
 
 
 6 
 
 
 
 •a 
 
 
 
 a 
 N 
 
 
 
 
 IZ 
 
 V 
 
 
 
 g 
 
 
 
 
 
 1 
 
 I. . . . 
 
 3.... 
 
 4 
 
 5. ... 
 6 
 
 0.35 
 0.342 
 
 0-334 
 0.326 
 0.318 
 0.31 
 
 0.05 
 0.05 
 0.05 
 0.05 
 0.05 
 0.05 
 
 0.05 
 0.05 
 0.05 
 0.05 
 0.05 
 0.05 
 
 0.06 
 
 0.049 
 
 0.038 
 
 0.025 
 
 0.013 
 
 0.00 
 
 0.06 
 0.06 
 0.06 
 0.06 
 0.06 
 0.06 
 
 0.06 
 0.06 
 0.06 
 0.06 
 0.06 
 0.06 
 
 O.IO 
 O.IO 
 
 0.16 
 
 O.IO 
 O.IO 
 O.IO 
 
 O.IO 
 O.IO 
 O.IO 
 O.IO 
 O.IO 
 O.IO 
 
 0.07 
 0.09 
 
 O.I I 
 
 0.13 
 0.15 
 0.17 
 
 1.38 
 1.26 
 1. 14 
 
 1.02 
 0.90 
 
 0.78 
 
 0.075 
 0.075 
 0.075 
 0.075 
 0.075 
 0.075 
 
 Description of Trials. — All enamels of the series are good 
 enamels with whiteness increasing toward No. 6, i. e., with in- 
 crease of AI2O3. The temperature required for maturing increases, 
 however, with the AI2O3. The best enamel of the series, taking 
 burning temperature, whiteness, gloss, and adhesive properties 
 into consideration, is No. 4 containing 0.16 AI2O3. 
 
 Series V. 
 
 VARIATION OF ANTIMONY OXIDE. 
 
 This series as well as the remaining two series was carried 
 out in two parts, A and B, the two parts being practically alike
 
 REPLACEMENT OF TIN OXIDE IN ENAMELS FOR CAST IRON. I9 
 
 -except for the silica content. Part B was carried out first and the 
 limits of Sb.O., were not high enough, hence these were changed 
 in A. 
 
 Series V, A. 
 
 0.16 K.O 1 
 
 0.05 ZnO 0.16 AUO3 1 1 .8 SiO. 
 
 o.ioCaO I I- 
 
 o.isPbO |o.2oB,03 Jo-o.i4Sb203 
 
 o . 54 NajO J 
 
 0.16K2O 
 0.05 ZnO 
 o. 10 CaO 
 0.15 PbO 
 0.54 NajO 
 
 Series V, B. 
 
 0.16 A1,0, 2.0 SiO,. 
 
 " \ 
 o . 20 B2O3 J 0-0 . 1 1 SbjO, 
 
 V, A. Batch formulae (in equivalents). 
 
 c 
 a: 
 
 ■0 
 
 0.35 
 
 0.05 
 
 0-35 
 
 0.05 
 
 0.35 
 
 0.05 
 
 0.35 
 
 0.05 
 
 0.35 
 
 0.05 
 
 0.35 
 
 0.05 
 
 0.35 
 
 0.05 
 
 0.35 
 
 0.05 
 
 N 
 
 O 
 
 a 
 
 O 
 
 CQ 
 
 'U, 
 
 I . 
 
 2 . 
 
 3- 
 4- 
 5- 
 6. 
 
 7 
 
 0.05 
 0.05 
 0.05 
 0.05 
 0.05 
 0.05 
 0.05 
 0.05 
 
 0.06 
 0.06 
 0.06 
 0.06 
 0.06 
 0.06 
 0.06 
 0.06 
 
 0.06 
 0.06 
 0.06 
 0.06 
 0.06 
 0.06 
 0.06 
 0.06 
 
 o. 10 
 o. 10 
 o. 10 
 o. 10 
 o. 10 
 o. 10 
 o. 10 
 o. 10 
 
 o. 10 
 o. 10 
 o. 10 
 o. 10 
 o. 10 
 o. 10 
 o. 10 
 o. 10 
 
 0.13 
 0.13 
 0.13 
 0.13 
 0.13 
 0.13 
 0.13 
 0.13 
 
 1 .02 
 1 .02 
 1 .02 
 1 .02 
 1 .02 
 1 .02 
 1 .02 
 1 .02 
 
 0.00 
 0.02 
 0.04 
 0.06 
 0.08 
 
 o. 10 
 
 O. 12 
 
 0.14 
 
 
 
 V, B 
 
 Batch formulae (in equi 
 
 valents 
 
 )■ 
 
 
 
 d 
 
 •z 
 
 
 
 « 1 ,5 
 
 c 1 c 2 
 
 u 
 
 ■x. 
 
 s £ 
 
 I 
 
 3 
 
 4 
 
 5 
 
 6 
 
 0.35 
 
 0-35 
 0.35 
 0.35 
 0.35 
 0.35 
 
 0.05 
 0.05 
 0.05 
 0.05 
 0.05 
 0.05 
 
 0.05 
 0.05 
 0.05 
 0.05 
 0.05 
 0.05 
 
 0.06 
 0.06 
 0.06 
 0.06 
 0.06 
 0.06 
 
 0.06 
 0.06 
 0.06 
 0.06 
 0.06 
 0.06 
 
 0. 10 
 0. 10 
 0. 10 
 0. 10 
 0. 10 
 0. 10 
 
 O.IO 
 O.IO 
 
 0. 10 
 0. 10 
 0. 10 
 0. 10 
 
 0.13 
 0.13 
 0.13 
 0.13 
 0.13 
 0.13 
 
 I .22 
 I .22 
 I .22 
 I .22 
 I .22 
 I .22 
 
 0.00 
 
 0.022 
 
 0.044 
 
 0.066 
 
 0.088 
 
 0. II 
 
 Description of Trials. — The trials of enamel i A have but
 
 20 REPLACEMENT OF TIN OXIDE IN ENAMELS FOR CAST IRON. 
 
 slight opacity. Xo. 2 A has a trifle more and so on up the series, 
 Enamel Xo. 3 A has a fair opacity, Xo. 4 A and 5 A are good 
 enamels, Xo. 5 A being the whitest. Xo. 6 A is a good enamel. 
 It is whiter than Xo. 5 but is not quite so glossy. Xo. 7 is a good 
 enamel and is a trifle "matte" in texture. Xo. S has a beautiful 
 matte texture and differs from all the rest of the series in this 
 respect. One of the trials, however, shows a tendency to shiver 
 but this may possibly be due to the mode of application. Enamel 
 Xo. 5 is the best of the A part of the series, taking gloss, finish, 
 and general appearance into consideration, while for a dull or 
 matte texture Xo. S is the best. Enamels Xo. 7 and 8 require 
 a higher temperature for burning, thus indicating that high 
 antimony decreases the fusibility. 
 
 With part B of the series shivering is more evident in every 
 case. The enamels which held are, however, of greater brilliancy 
 and opacity, enamel Xo. 2 of A being identical in appearance 
 with Xo. I of B. Enamels 2, 3 and 4 of part B are practically 
 the same as 3, 4 and 5 of part A respectively. From this we would 
 conclude that 0.016 equivalent of Sb,03. in this range of silica 
 content, has about the same opacifying effect as 0.02 equivalent 
 of silica. 
 
 Series VI. 
 
 REPLACEMENT OF ANTIMONY OXIDE BY TIN OXIDE IN AN ANTIMONY 
 
 ENAMEL. 
 
 This series, also using two different equivalents of silica., 
 was carried out as follows : 
 
 VI, A. 
 o.i6K,,0 
 
 0.05 ZnO 1 o. 16 AUO3 ' 1 .80 SiO. 
 o . 10 CaO [• } 
 
 0.15 PbO I 0.20 B2O3 J 0.075 Sb203-o.i5 SnO, 
 o . 54 Na^O J 
 
 VI, B. 
 0.16 K,0 1 
 
 0.05 ZnO I o. 16 AUO3 ~ 2.0 SiO., 
 o . 10 CaO > > 
 
 0.15 PbO I 0.20 B2O3 J 0.075 Sb203-o.i5 SnO, 
 0.54 Xa.P J
 
 REPLACEMENT ( )F TLN OXIDE L\ ENAMELS FOR CAST IRON. 2 1 
 
 Batch forimilae (in equivalents). 
 VI. A. 
 
 6 
 2: 
 
 § 
 
 d 
 2 
 
 R. Lead 
 
 C 
 
 s 
 
 C 
 12 
 
 >• 
 
 
 
 
 
 n 
 u 
 
 
 
 ca 
 
 u 
 X 
 
 b 
 
 c 
 
 i 
 
 
 
 c 
 
 I. . . . 
 
 0.35 
 
 0.05 
 
 0.05 
 
 0.03 
 
 0.06 
 
 O.IO 
 
 O.IO 
 
 0.13 
 
 1.02 
 
 0.075 
 
 0.00 
 
 
 0.35 
 
 0.05 
 
 0.05 
 
 0.03 
 
 0.06 
 
 O.IO 
 
 0.10 
 
 0.13 
 
 1.02 
 
 0.056 
 
 0.038 
 
 3 
 
 0.35 
 
 0.05 
 
 0.05 
 
 0.03 
 
 0.06 
 
 O.IO 
 
 O.IO 
 
 0.13 
 
 1.02 
 
 0.038 
 
 0.075 
 
 4 
 
 035 
 
 0.05 
 
 0.05 
 
 0.03 
 
 0.06 
 
 O.IO 
 
 O.IO 
 
 0.13 
 
 1.02 
 
 0.019 
 
 0.II25 
 
 5 
 
 0-35 
 
 0.05 
 
 0.05 
 
 0.03 
 
 0.06 
 
 O.IO 
 
 O.IO 
 
 0.13 
 
 1.02 
 
 0.00 
 
 0.15 
 
 VI, B. 
 
 I. . . . 
 
 0.35 
 
 0.05 
 
 0.05 
 
 0.03 
 
 0.06 
 
 O.IO 
 
 O.IO 
 
 0.13 
 
 1.22 
 
 0.075 
 
 0.00 
 
 2. . . . 
 
 0-35 
 
 0.05 
 
 0.05 
 
 0.03 
 
 0.06 
 
 O.IO 
 
 O.IO 
 
 0.13 
 
 1.22 
 
 0.056 
 
 0.038 
 
 3.... 
 
 0-35 
 
 0.05 
 
 0.05 
 
 0.03 
 
 0.06 
 
 O.IO 
 
 O.IO 
 
 0.13 
 
 1.22 
 
 0.038 
 
 0.075 
 
 4 
 
 0.35 
 
 0.05 
 
 0.05 
 
 0.03 
 
 0.06 
 
 O.IO 
 
 O.IO 
 
 0.13 
 
 1.22 
 
 0.019 
 
 O.II25 
 
 5. ... 
 
 0.35 
 
 0.05 
 
 0.05 
 
 0.0^ 
 
 0.06 
 
 O.IO 
 
 O.IO 
 
 0.13 
 
 1.22 
 
 0.00 
 
 0.15 
 
 Description of Trials. — All enamels of the A part of the series 
 adhere tenaciously and are good enamels. Enamel No. i has 
 more opacity and whiteness than No. 5, these two properties 
 decreasing uniformly between these extremes. The antimony 
 enamel requires a slightly higher temperature for maturing, but 
 not to such extent as to be detrimental to the iron. 
 
 In the B part of the series shivering is much in evidence, 
 due to the increased silica. Enamels Nos. i and 2 have good 
 opacity but Nos. 3 and 4 are much inferior in this respect. In 
 enamel No. 5 the silica has dissolved the SnOj almost entirely. 
 Taking the results of this series we would conclude that SiO, 
 at the higher limit is opposite in effect with regard to SboOg and 
 SnO,. In the case of Sb203 the opacity, whiteness and brilliancy 
 are increased, while with vSnO, these properties, notably the 
 opacity, are decreased. Shivering, however, is increased in 
 either case. The results obtained in part A of the series are not 
 in accord with those of Riddle whose high limit of silica was 1.23 
 equivalents. In enamel No. 5 part A as given above, a good 
 white enamel was obtained using 1.8 equivalents of silica. 
 
 It might be interesting to note also at this point, the be- 
 havior of the enamels on wTought iron. The enamels of part 
 A were applied to iron washers, and although they had not been
 
 22 REPLACEMENT OF TIN OXIDE IN ENAMELS FOR CAST IRON. 
 
 previously cleaned, the enamels held perfectly and were of good 
 whiteness, brilliancy and texture. 
 
 Series VII. 
 
 VARIATION OF BORIC OXIDE. 
 
 This series employs two equivalents of silica and the NajCOg 
 content is varied in order to reach the lower limit of B2O3 still 
 maintaining the same ratio. 
 
 0.16 KoO 
 0.05 ZnO 
 o. 10 CaO 
 0.15 PbO 
 0.54 NaoO 
 
 VII, A. 
 0.16 AI2O3 1 1.80 SiOa 
 
 0.10-0.40 B2O3 J 0.075 Sb203 
 
 VII, B. 
 
 1 2.00 SiO, 
 
 0.16K2O 1 
 
 0.05 ZnO I o. 16 AI2O3 
 
 o. 10 CaO } 
 
 0.15 PbO I 0.10-0.40 B2O3 J 0.075 Sb203 
 
 0.54 Nap J 
 
 VII, A. 
 Batch formulae (in equivalents) 
 
 1 
 
 
 
 
 (S 
 
 •d 
 >A 
 pi 
 
 
 
 a 
 
 
 
 Cryolite 
 Ca(OH)2 
 
 Borax 
 K. Spar 
 
 
 
 
 I 
 
 0.4 
 
 0-375 
 
 0.350 
 
 0.325 
 
 0.300 
 
 0.275 
 
 0.25 
 
 0.05 
 0.05 
 0.05 
 0.05 
 0.05 
 0.05 
 0.05 
 
 0.05 
 0.05 
 0.05 
 0.05 
 0.05 
 0.05 
 0.05 
 
 0.03 
 0.03 
 0.03 
 0.03 
 0.03 
 0.03 
 0.03 
 
 0.06 
 0.06 
 0.06 
 0.06 
 0.06 
 0.06 
 0.06 
 
 O.IO 
 O.IO 
 O.IO 
 O.IO 
 O.IO 
 O.IO 
 O.IO 
 
 0.05 
 
 0.075 
 
 O.IO 
 
 0.125 
 0.150 
 0.175 
 
 0.2 
 
 0.13 
 0.13 
 0.13 
 0.13 
 0.13 
 0.13 
 0.13 
 
 1.02 
 1.02 
 1.02 
 1.02 
 1.02 
 1. 02 
 1.02 
 
 0.075 
 0.075 
 
 3 
 
 4 
 
 5 
 
 6 
 
 7 
 
 0.075 
 0.075 
 0.075 
 0.075 
 0.075 
 
 VII, B. 
 
 0.4 
 
 0.05 
 
 0.05 
 
 0.03. 
 
 0.06 
 
 O.IO 
 
 0.05 
 
 0.13 
 
 1.22 
 
 0.075 
 
 0.375 
 
 0.05 
 
 0.05 
 
 0.03 
 
 0.06 
 
 O.IO 
 
 0.075 
 
 0.13 
 
 1.22 
 
 0.075 
 
 0-350 
 
 0.05 
 
 0.05 
 
 0.03 
 
 0.06 
 
 O.IO 
 
 O.IO 
 
 0.13 
 
 1.22 
 
 0.075 
 
 0.325 
 
 0.05 
 
 0.05 
 
 0.03 
 
 0.06 
 
 O.IO 
 
 0.125 
 
 0.13 
 
 1.22 
 
 0.075 
 
 0.300 
 
 0.05 
 
 0.05 
 
 0.03 
 
 0.06 
 
 O.IO 
 
 0.150 
 
 0.13 
 
 1.22 
 
 0.075 
 
 0.275 
 
 0.05 
 
 0.05 
 
 0.03 
 
 0.06 
 
 O.IO 
 
 0.175 
 
 0.13 
 
 1.22 
 
 0.075 
 
 0.250 
 
 0.05 
 
 0.05 
 
 0.03 
 
 0.06 
 
 O.IO 
 
 0.200 
 
 0.13 
 
 1.22 
 
 0.075
 
 REPLACEMENT OF TIN OXIDE IN ENAMELS FOR CAST IRON. 23 
 
 Description of Trials. — All enamels of part A adhere well 
 and are good white enamels up to No. 6. Nos. 6 and 7 liave a 
 yellowish cast and are not all desirable enamels. Bubbling is 
 also evident in the enamels of higher BJJ.^ content. Enamel No. 
 I is the whitest of the five enamels. 
 
 The results obtained in part B are substantially the same as 
 those of part A. The enamels are whiter, however, than the ones 
 of the same B/)., content and the yellowish cast of enamels 6 
 and 7 of part A has disappeared in the corresponding enamels of 
 part B. Shivering is present to quite a large extent in part B, 
 due to the high silica. As in part A, bubbling is prominent 
 in the enamels of the higher B.Oj content. The dilTerence in 
 whiteness of the high and low B.O^ enamels in part A is not so 
 pronounced in this part of the series. Enamels B i and B 7 
 have very little difference in whiteness, B i being a little the 
 whitest. The difference in maturing temperature is however 
 quite large and the tendency to bubbling is more evident. 
 
 The results indicate that the lower the B^O^ the better and 
 whiter are the enamels. The limits for desirable enamels are 
 about 0.15-0.30 B2O3. 
 
 LIMITS OF THE INGREDIENTS. 
 
 The limits of the ingredients and their effects established 
 by this work are as follows : 
 
 SiO^: The effect of silica is to increase brilliancy, white- 
 ness, acid-resisting properties and gloss. If increased too high, 
 shivering takes place and the maturing temperature is too high. 
 The limits are about i. 65-1. 85 equivalents, those nearer the 
 higher limit being the preferable. 
 
 Al.,0.j : Increased AljO., increases the temperature for matur- 
 ing and gives whiter enamels. The high limit is around 0.18 
 equivalent. The low limit was not established but for com- 
 mercial enamels is probably about 0.13 equivalent. 
 
 Sb^O^: The effect of vSb._,0.j is to increase the maturing tem- 
 perature, and to increase the whiteness and opacity when em- 
 ployed between the limits of 0.0-0.09 equivalent Sbp^. If 
 used between the limits of 0.1-0.14 equivalent the enamels are 
 dull at the lower limits and matteness increases at the higher.
 
 24 REPLACEMENT OF TIN OXIDE IN ENAMELS FOR CAST IRON. 
 
 At the high Hmit, 0.14 equivalent, shivering is Hkely to occur. 
 
 For brilliant enamels of good opacity and texture the limits 
 are 0.06-0.09 equivalent, about 0.075 being preferable. 
 
 vSnO.: No variation of the SnO. content was made but a 
 good enamel was obtained using 0.15 equivalent of SnO,. 
 
 B,0., : The effect of increased B.Og is to lower the maturing 
 temperature, to increase the tendency to produce bubbles, to 
 decrease the whiteness when used above a certain limit, increase 
 o-loss and to increase the solubility of the enamel. The limits 
 are about 0.15-0.3 equivalent, those nearer the lower limit being 
 preferable. 
 
 BaO: The effect of BaO in Sb.O., enamels is to produce a 
 "puckery" or matte effect. This is no doubt due to the sulphur 
 arising from the Sbp.; and the fuel gases, which comes in contact 
 with the barium compounds. 
 
 The most likely enamel taking all points into consideration 
 
 is: 
 
 o.i6K,0 1 
 
 0.05 ZnO o.i6Al,,0, I i-SoSiO, 
 
 o. 10 CaO 
 
 0.15 PbO I 0.20 B2O3 J 0.075 SbPa 
 
 0.54 Na,0 J 
 
 • DISCUSSION. 
 
 Professor Staler: Why do you not include the fluorine in 
 your formula? No one will be able to calculate the batch from 
 the formula unless you do so. Moreover, it makes a vast differ- 
 ence whether an enamel contains a small or a large amount of 
 this element. 
 
 Mr. Brown: I do not think it is necessary. I introduced 
 it as cryolite, using 0.06 equivalent of cryolite throughout. 
 
 Professor Sialcy: Mr. Brown, I just want to ask one more 
 question. Did you get an absolutely pure white enamel, or was 
 it of a greenish or bluish tint? There have been many attempts 
 made to use antimony in place of tin oxide in cast iron enamels, 
 but it has never given a satisfactory white. They get a tint 
 they call white, but it is not a commercial white. Do you have 
 any idea of how to avoid getting that greenish, bluish white so 
 characteristic of antimony oxide?
 
 REPLACEMENT OF TIN OXIDE IN ENAMELS KOR CAST IRON. 25 
 
 Mr. Broiin: I did not carry on work to eliminate the cast 
 you speak of. The cast was not present to an a^^sravated extent 
 that I could see. A number of others said the same thing. Tliere 
 is a slight bluish cast or tint in some of the trials. 
 
 Professor Staler: In your ilnal enamel as well as in all the 
 others ? 
 
 Mr. Brown: It was not so pronounced in this case, but 
 more so in the enamels of higher silica content. 
 
 Mr. Burt: I noticed in speaking of the enameled iron 
 industry they always speak of dusting the enamel on and I would 
 like to get a little description of what the mechanical process is — 
 of what is involved in this dusting on of the glaze. 
 
 Professor Staley: In a paper (" The Manufacture of Enameled 
 Iron Sanitary Ware," Trans. A. C. S., \'ol. V'lII, p. 172) I pub- 
 lished several years ago, you can find a description of the ordinary 
 method of making a piece of enameled cast iron. The only 
 difference between the method described there and the method 
 used at present is the use of a mechanical agitator. 
 
 Mr. Burt: What mesh sieve do you use? 
 
 Professor Staley: The sieve is a fifty- or sixty-mesh sieve. 
 
 Mr. Brown: I would like to ask Professor Staley what his 
 opinion is of the lluorine in a fused enamel — ^whether it is 
 volatilized or whether it is retained in the enamel. I have read 
 of several instances where they analyzed for lluorine and found 
 it in the enamels in small quantities. 
 
 Professor Staley: That is all a matter, in my mind, of how- 
 hard, how long, and how hot you heat the enamel. You can 
 volatilize it all, ot you can have the larger portion of it stay in. 
 If it is all volatilized you have no opacifying effect from the use of 
 fluorides. In cast iron enamels that are heated or fritted in the 
 ordinary length of time, the large bulk of fluorine stays in. 
 
 NOTE PREPARED AFTER READING THE PAPER. 
 
 Professor Bleininger: It seems to me that Mr. Brown has 
 solved his problem satisfactorily. He has accomplished two 
 things, viz., the production of a white enamel wliicli compares 
 favorably with the best tin enamels, in the opinion of impartial 
 observers, and he likewise has shown clearly the kind of enamel
 
 26 REPLACEMENT OF TIN OXIDE IN ENAMELS FOR CAST IRON. 
 
 required for the use of antimony as an opacifier, which differs 
 somewhat from the common t^^pe. 
 
 As regards the poisonous quaUty of antimony compounds. 
 Rickmann, Sprechsaal, XL^^ 115-117, says that during an ex- 
 perience of ten years the use of metasodium antimonate has not 
 proven injurious. However, he points out that the antimony 
 oxide compounds (tartar emetic, etc.) are poisonous. For cast 
 iron enamels, therefore, the use of Na^Sb^Oa might be a perfectly 
 feasible solution.