OF CALIFORNIA LIBRARY OF THE UNIVERSITY OF CALIFORNIA 
 
 3 = 6 
 
 
 
 9 
 
 T OF CALIFORNIA 
 
 Qj/AkD 
 LIBRARY OF THE UNIVERSITY OF CALIFORNIA 
 
 OF CALIFORNIA 
 ?=f 
 
 LIBRARY OF THE UNIVERSITY OF CALIFORNIA 
 
E UNIVERSITY OF CALIFORNIA 
 
 LIBRARY OF THE UNIVERSITY OF 
 
 (5\\_yft> x* 
 
 E UNIVERSITY OF CALIFORNIA LIBRARY OF THE UNIVERSITY OF 
 
 iE UNIVERSITY OF CALIFORNIA LIBRARY OF THE UNIVERSITY OF 
 
THE RISE AND DEVELOPMENT 
 
 OF 
 
 THE LIQUEFACTION OF GASES 
 
THE 
 
 RISE AND DEVELOPMENT 
 
 OF THE 
 
 LIQUEFACTION OF GASES 
 
 BY 
 
 WILLETT L. HARDIN, PH.D. 
 
 HARRISON SENIOR FELLOW IN CHEMISTRY IN THB 
 UNIVERSITY OF PENNSYLVANIA 
 
 THE MACMILLAN COMPANY 
 
 LONDON : MACMILLAN & CO., LTD. 
 
 !95 
 
 ^// rights reserved 
 
COPYRIGHT, 1899, 
 BY THE MACMILLAN COMPANY. 
 
 Set up and electrotyped. Published August, 1899. Reprinted 
 March, 1905. 
 
PREFACE 
 
 RECENT developments in the liquefaction of air 
 and the recent liquefaction of hydrogen have 
 added considerable interest to the whole subject 
 of the liquefaction of gases. The literature on 
 this subject is scattered, for the most part, in 
 foreign journals, and is inaccessible to a majority 
 of those who are interested in scientific work. 
 
 The object of this little volume is to present 
 a complete history of the development of the 
 methods employed in the liquefaction of gases. 
 Sufficient theory has been given to enable the 
 popular reader to understand the principles in- 
 volved. While the book has been written in a 
 popular-science style, an effort has been made to 
 make it of value to those who are especially 
 interested in the subject by giving the references 
 to the original literature. 
 
 The first intention was to include a complete 
 account of researches at low temperatures and 
 of the industrial applications of liquefied gases. 
 
 V 
 
 224539 
 
vi PREFACE 
 
 This, however, would make the work unduly 
 large, and for that reason these subjects are con- 
 sidered only in a general way. 
 
 Full credit has been given throughout the book 
 to the various sources of information. In conclu- 
 sion I desire to express my obligations to Mr. 
 Louis M. Thorn for the care which he has taken 
 in the preparation of the drawings, and also to 
 Professor Edgar F. Smith for his kindness in 
 reading the manuscript and for many valuable 
 suggestions. 
 
 W. L. HARDIN. 
 
 APRIL, 1899. 
 
CONTENTS 
 
 INTRODUCTION 
 
 CHAPTER I 
 
 Introduction of the term "gas" Distinction between 
 gases and vapors Guericke's air-pump Experi- 
 ments of Boyle and Mariotte on the compression of 
 gases Influence of temperature on gaseous volume 
 Absolute temperature Early experiments on 
 the liquefaction of gases Vaporization of liquids 
 in closed tubes by Cagnaird de la Tour ... 5 
 
 CHAPTER II 
 
 Liquefaction of gases by Faraday Attempt to liquefy 
 air by Perkins Experiments of Bussy Evapora- 
 tion of liquids as a means of lowering the tempera- 
 ture Heat of vaporization Attempt to liquefy 
 air by Colladon Liquefaction and solidification 
 of carbonic acid by Thilorier Application of solid 
 carbonic acid and ether to the production of low 
 temperatures Sinking of gases to great depths in 
 the ocean Second series of experiments by Fara- 
 day Experiments of Natterer Observations of 
 Berthelot, Drion, and Mendeleeff Exceptions to 
 Boyle's law ........ 22 
 
 CHAPTER III 
 
 Experiments of Andrews on carbonic acid Introduc- 
 tion of critical constants Determination of critical 
 vii 
 
viii CONTENTS 
 
 constants Critical constants of gaseous mixtures 
 Continuity of the gaseous and liquid states of 
 matter Condition of matter at the critical point 
 Exceptions to Boyle's law Kinetic theory of gases 
 Relation between the gaseous and liquid states 
 as expressed by the equation of Van der Waals . 70 
 
 CHAPTER IV 
 
 Liquefaction of the so-called permanent gases by Caille- 
 tet and Pictet Liquefaction of ozone Liquefac- 
 tion and determination of boiling points and critical 
 constants of oxygen, air, nitrogen, etc., by Wroblew- 
 ski, Olszewski, and Dewar Solidification of the 
 so-called permanent gases Dewar vacuum bulbs 
 
 Researches at low temperatures Experiments 
 of Kamerlingh-Onnes History of the regenera- 
 tive method of refrigeration Application of this 
 method in the liquefaction of oxygen, air, nitrogen, 
 etc., by Linde, Hampson, Dewar, and Tripler 
 Theory of the regenerative method of refrigeration 
 
 Liquefaction and solidification of argon by 
 Olszewski Attempt to liquefy helium Lique- 
 faction of fluorine by Moissan and Dewar Lique- 
 faction of hydrogen and helium by Dewar 
 Experiments on krypton, metargon, and neon 
 The supposed new gas etherion Table of physi- 
 cal constants . -113 
 
 CONCLUSION 
 
 The three states of matter Industrial application of 
 liquefied gases Physiological action at low tem- 
 peratures Properties of matter at low temperatures 23 1 
 
"CONSIDER for a moment what would happen to the dif- 
 ferent substances which compose the globe, if its temperature 
 should be suddenly changed. Suppose, for instance, that the 
 earth . . . should be suddenly placed in a very cold region, 
 ... the water which at present forms our rivers and seas, 
 and probably a majority of the liquids which are known, 
 would be transformed into solid mountains. On this suppo- 
 sition the air, or at least a part of the aeriform substances 
 which compose it, would doubtless cease to exist in the state 
 of an invisible fluid, for want of a sufficient degree of heat ; 
 it would return to the liquid state, and this change would 
 produce new liquids of which we have no knowledge." 
 
 LAVOISIER (1784). 
 
LIQUEFACTION OF GASES 
 
 INTRODUCTION 
 
 PROBABLY no line of scientific research has been 
 more productive of ingenious experiments or more 
 prolific of results than that of the liquefaction of 
 gases. To convert ordinary air into a liquid or 
 even a solid is to-day a matter of little difficulty. 
 So great an achievement, however, is not the work 
 of a single individual or the product of a single 
 generation. It is a result of the combined efforts 
 of numerous experimenters, and represents the 
 progress of a century. 
 
 The development of the methods has been ac- 
 companied by many fruitless and discouraging 
 observations. The experiments have been of such 
 a nature as to require the application of consider- 
 able pressure. Vessels of exceedingly brittle glass 
 were used in most cases, and many explosions 
 resulted. There was also a great expense attached 
 to experiments of this nature. Within recent years 
 various metals and alloys have been substituted 
 for glass in the construction of apparatus. Pictet 
 
 B I 
 
OF GASES 
 
 Vays 1 t&afc 'modern* rhetariurgy has greatly aided in 
 the construction of pressure-vessels. The present 
 high grade of steel is almost indispensable. 
 
 Notwithstanding the fact that the difficulties 
 to be overcome were enormous, and that the ex- 
 perimenters were subjected to considerable danger, 
 the investigations have been carried on with 
 untiring zeal. The history of the liquefaction of 
 gases, like that of other lines of scientific research, 
 is marked with periods of unusual activity and 
 rapid progress. These periods of enthusiasm were 
 sometimes followed by a few years of quiet, un- 
 eventful observations. The announcement of a 
 new discovery, however, was always sufficient to 
 give a new impetus to the work. 
 
 In tracing the development of the methods em- 
 ployed in the liquefaction of gases, it is perhaps 
 advisable to divide the work into four periods. 
 The four chapters of this volume correspond to 
 these periods. 
 
 The first chapter is concerned with the early 
 history of the subject. In this period some of the 
 earlier observations on the compression of gases, 
 and likewise a few disconnected experiments on 
 the liquefaction of gases, are considered. These 
 investigations are only of historical interest, and 
 hence are only briefly outlined. 
 
 The second chapter begins with the work of 
 
INTRODUCTION 3 
 
 Faraday (1823), and embraces nearly half a century 
 of fruitful observations. The fundamental methods 
 for obtaining high pressures and low temperatures 
 were developed to a comparatively high degree 
 during this period. Carbonic acid, nitrous oxide, 
 ammonia, etc., were liquefied and solidified dur- 
 ing this time. 
 
 The third chapter is devoted to critical con- 
 stants, and the continuity of the gaseous and 
 liquid states of matter. The experiments of An- 
 drews form the first and most important division 
 of this chapter. These observations mark the 
 beginning of a new epoch in the liquefaction of 
 gases. A brief account is also given of some 
 experiments which have been made with a view of 
 determining the condition of matter at the critical 
 point. The equation of Van der Waals is likewise 
 briefly considered. 
 
 The fourth period begins with the liquefaction 
 of the so-called permanent gases by Cailletet and 
 Pictet (1877), and extends to the present time. 
 During this period the apparatus employed in the 
 liquefaction of gases has been perfected to a very 
 high degree. The chapter is divided into four 
 sections as follows : 
 
 i. The pioneer experiments of Cailletet and 
 Pictet on the liquefaction of the so-called perma- 
 nent gases (1877-1882). 
 
4 LIQUEFACTION OF GASES 
 
 2. The experiments of Wroblewski, Olszewski, 
 and Dewar from 1883 to 1895. This work may 
 be considered as supplementary to that of Caille- 
 tet and Pictet. 
 
 3. Liquefaction of gases by the regenerative 
 method. 
 
 4. Liquefaction of argon, hydrogen, helium, 
 etc. The section closes with a table of physical 
 constants. 
 
 This chapter, which extends over a period of 
 only two decades, occupies more than one half of 
 the present volume. 
 
 In the conclusion the three states of matter are 
 briefly compared, and- their similarities pointed 
 out. Reference is also made to various industrial 
 applications of liquefied gases. Great advance in 
 this direction seems imminent from the recent 
 progress in the liquefaction of gases. A short 
 account is next given of physiological action, and 
 the properties of matter at low temperatures. 
 Here again we may look forward to great achieve- 
 ments. The high pressures and low temperatures 
 which can be obtained by means of liquid air, etc., 
 have opened up new fields of research in every 
 branch of natural and physical science. 
 
CHAPTER I 
 
 EARLY HISTORY 
 
 THE fact has long been known that many liquids 
 can be converted into vapors which are similar, in 
 most respects, to ordinary gases. It has been 
 equally well known that the vapors thus produced 
 can be condensed to the original liquids by lower- 
 ing the temperature. These phenomena at once 
 suggest an intimate relation between the gaseous 
 and liquid states of matter. Some believed that 
 the possibility of condensation applied, not only to 
 certain vapors, but to all gases. Before the com- 
 position of the atmosphere was known, attempts 
 were made to condense it to a liquid. 
 
 Some of the experiments which are considered 
 in this chapter were made, not with a view of con- 
 densing the gas, but for the purpose of studying 
 the influence of temperature and pressure on gas- 
 eous volume. The results in many cases, however, 
 have an important bearing on the liquefaction of 
 gases. 
 
 Experiments on the pneumatic applications of 
 compressed air were made before the Christian 
 
 5 
 
6 LIQUEFACTION OF GASES 
 
 era. These observations, and many other experi- 
 ments on compressed gases, may be omitted in a 
 work of this nature. 
 
 For the present purpose we may begin with the 
 observations of Van Helmont 1 in the latter part of 
 the sixteenth and the early part of the seventeenth 
 century. He made the first great advance in the 
 study of gases. He introduced the term "gas," 
 and applied it to bodies which are similar to or- 
 dinary air. 'It is evident that the problem of con- 
 densing gases to the liquid state also occurred to 
 him, for he was the first to distinguish between 
 gases and vapors ; the latter, he said, can be con- 
 densed to the liquid state, while the former cannot. 
 This distinction remained unquestioned for nearly 
 two centuries. When we consider that experi- 
 ments on distillations occupied fifty years of Van 
 Helmont's life, 2 we can understand these state- 
 ments, which, in reality, are far in advance of the 
 age in which he lived. 
 
 About the middle of the seventeenth century 
 Guericke 3 constructed an air-pump. This ap- 
 paratus, in its simplest form, consisted of a ver- 
 tical brass cylinder, which was provided with a 
 movable piston. By means of this pump, he 
 
 1 Kopp, Geschichte der Chemie, I, pp. 121-122. 
 
 2 Boerhaave, Elements of Chemistry, Eng. ed., I, p. 16. 
 8 Haiiy, Nat, Philos., Eng. ed., p. 220. 
 
EARLY HISTORY 7 
 
 studied the influence of pressure on the volume 
 of a given quantity of air. The principle of 
 Guericke's air-pump has been applied by numer- 
 ous experimenters in their study of compressed 
 gases. 
 
 The first systematic observations on the influence 
 of pressure on the volume of a gas are those of 
 Boyle. In 1662 he announced an important law, 
 which bears his name. He stated that the volume 
 of a gas varies inversely as the pressure. In other 
 words, the product of the pressure and volume is 
 a constant. The law is usually expressed by the 
 equation 
 
 pv = c, 
 
 where c is a constant. The same law was an- 
 nounced by Mariotte a few years later. The con- 
 ditions under which gases deviate from this law 
 will be considered in a subsequent chapter. 
 
 From the writings of Boerhaave (1731), it is 
 evident that he had considered the problem of 
 liquefying and of solidifying ordinary air. In the 
 English edition of his Elements of Chemistry, pp. 
 249, 250, we find the following statement : 
 
 " The first property, then, of air which offers 
 itself to our consideration is its fluidity. This is 
 so natural to it that I do not remember ever to 
 have heard of any experiment by which air could 
 be deprived of it. It is evident to every one's 
 
8 LIQUEFACTION OF GASES 
 
 observation that even in the sharpest frost, when 
 almost everything is congealed, the air still re- 
 mains fluid ; nay, in an artificial cold, 40 greater 
 than nature has ever been known to produce, the 
 air still retained its fluidity. ... If you compress 
 the air, with ever so great weight and force, into 
 the utmost density, it does not then become solid 
 by concretion, but remains equally fluid as before. 
 ... I have never yet met with a single experiment 
 by which it appeared that air was coagulated into 
 a solid mass. I confess that, one noon in frosty 
 weather, when the air was very serene, I observed 
 some very small corpuscles floating about in it. ... 
 But, after a careful observation, I discovered that 
 these were nothing but little globules of water, 
 which were congealed, and which appeared in the 
 form of a very subtle hoar-frost." 
 
 In speaking of the elasticity of the air, the same 
 writer, pp. 263-264, says : 
 
 " Another law which we find to hold true is, that 
 the elasticity of the air cannot be destroyed. . 
 In whatsoever manner the air has been com- 
 pressed by the utmost power of weights, it has 
 always remained very fluid ; for, after it has been 
 contracted into the greatest density, it has con- 
 stantly restored itself again into all its particles, 
 so as to fill up exactly the former space ; all parti- 
 cles retreating with the same ease with which they 
 
EARLY HISTORY 9 
 
 came together. . . . We may fairly assert that 
 the fluidity of the air in all the large compass, 
 from the most rarefied to the most compressed, 
 remains without alteration ; and that therefore it is 
 neither capable of being solidified by the intensest 
 cold, nor the greatest degree of compression." 
 
 On page 266 of the same book we find that 
 Boerhaave was familiar with the fact that heat 
 expands the air. The following are his words : 
 
 " By the application of fire the air becomes so 
 rare that neither the measure nor limit of its dila- 
 tion has yet been discovered. . . . Air of unequal 
 masses, but of the same density, is always ex- 
 panded in the same measure by the same degree 
 of fire : so that these expansions in the same 
 density of air are, by a constant law of nature, 
 always proportional to the augmentations of 
 heat." 
 
 As early as 1 702 Amontons l studied the effect 
 of temperature on the elastic force of the air. He 
 says : " If equal or unequal masses of air are 
 charged with equal weights, their elastic forces 
 will be equally increased by equal degrees of 
 heat." 
 
 In 1787 Charles 2 called attention to the fact 
 
 1 Haiiy, Nat. Philos., Eng. ed., pp. 255-260. 
 
 2 Charles communicated his results to Gay Lussac and did not 
 publish them. 
 
TO LIQUEFACTION OF GASES 
 
 that all gases expand equally with the same in- 
 crease in temperature. Dalton 1 made similar 
 observations in 1801. He is also the author of the 
 following statement : 2 " There can scarcely be 
 a doubt entertained respecting the reducibility of 
 all elastic fluids of whatever kind into liquids ; and 
 we ought not to despair of effecting it in low 
 temperatures, and by strong pressures exerted 
 upon the unmixed gases." Dalton, just as La- 
 voisier, foresaw the result of subjecting gases to 
 intense cold. 
 
 The relation of temperature to the volume of 
 a gas was more thoroughly investigated by Gay 
 Lussac in 1802. From his results, as well as 
 from those of Dalton and Charles, we find that an 
 increase of i in temperature 3 increases the vol- 
 ume of a gas by about %]% of the volume at o, 
 provided the pressure remains constant. In other 
 words, the volume v of a gas at the temperature / 
 is given by the equation, 
 
 where V Q represents the volume at o. In various 
 
 1 Manchester Memoirs, V, p. 535, 1801. Ref. Roscoe and Schor- 
 lemmer, Chemistry, I, p. 60, 1895. 
 
 * Ref. Dewar, Proc. Roy. /;/., 8, p. 657, 1878. 
 
 8 All temperatures will be given in Centigrade degrees unless 
 otherwise expressed. 
 
EARLY HISTORY 11 
 
 text-books, this law is referred to as the law of 
 Charles, the law of Dalton, or the law of Gay 
 Lussac, depending upon the author of the book. 
 According to this law all gases have the same 
 coefficient of expansion with regard to tempera- 
 ture. In this respect the gaseous state of matter 
 stands alone ; the coefficients of expansion of 
 liquids and solids show considerable variation. 
 
 Absolute Temperature. 
 
 Throughout the discussion of the experiments 
 on the liquefaction of gases, reference will fre- 
 quently be made to absolute temperature. The 
 significance of this term may be best understood 
 by considering the law of Gay Lussac. Suppose 
 we start with a given volume of gas at o. Let 
 the pressure remain constant. If the tempera- 
 ture be increased to 273, the volume will be 
 doubled. If, on the other hand, the temperature 
 be lowered, the volume will be decreased by ^3 
 of the original volume for each degree of tem- 
 perature change. If the law remains valid for 
 all temperatures, an important conclusion fol- 
 lows; namely, at the temperature of 273 the 
 volume of the gas will become equal to zero. 
 Of course, it is not likely that such a condition 
 will ever be attained, for all gases, so far as 
 
12 LIQUEFACTION OF GASES 
 
 known, deviate from this law before the tem- 
 perature of 273 is reached. The temperature 
 at which the volume of a perfect gas becomes 
 equal to zero ( 273 C. or 460 F.) is called 
 
 the ABSOLUTE ZERO OF TEMPERATURE. Absolute 
 
 temperatures are measured from this point. The 
 absolute zero may also be defined as that tem- 
 perature at which the kinetic energy of the 
 molecules is equal to zero. Thomson (now Lord 
 Kelvin) has also proposed a definition which is 
 based entirely upon energy considerations. 1 This 
 definition, however, need not be discussed here. 
 He obtained the value 273. 7 for the absolute 
 zero. 
 
 Both Dalton and Gay Lussac made observa- 
 tions on the change of temperature which ac- 
 companies the expansion or compression of a 
 gas. Dalton 2 found that "about 50 of heat 
 are evolved when air is compressed to one half 
 of its original volume, and that, on the other 
 hand, 50 are absorbed by the corresponding 
 rarefaction." Gay Lussac 3 employed two large 
 flasks, one of which contained the gas, while the 
 
 1 Phil. Mag., Oct., 1848. 
 
 2 Memoirs of the Literary and Philosophical Society of Man- 
 chester, V, Part II, pp. 251-525. 
 
 8 See original memoir in The Free Expansion of Gases, by 
 Ames, pp. 3-13. 
 
EARLY HISTORY 13 
 
 other was exhausted. Thermometers were placed 
 in each of these flasks, and the two were then 
 connected. This allowed the gas to expand in 
 one flask, while it was being compressed in the 
 other. In this manner, Gay Lussac experimented 
 with ordinary air, hydrogen, carbon dioxide, and 
 oxygen, and observed the accompanying changes 
 of temperature. This subject has been more 
 thoroughly investigated by Joule and Thomson. 
 Their work will be referred to in a subsequent 
 chapter. This method, in recent years, has be- 
 come of great importance in the production of 
 low temperatures. 
 
 It has been considered probable by some that 
 Count Rumford, 1 in his experiments to deter- 
 mine the explosive force of gunpowder, may 
 have liquefied carbonic acid gas. He exploded 
 the powder in cylinders closed with a weighted, 
 movable valve. The following are his words : 
 "When the force of the generated elastic vapor 
 was sufficient to raise the weight, the explosion 
 was attended by a very sharp and surprisingly 
 loud report ; but when the weight was not raised, 
 as also when it was only a little moved, but not 
 sufficiently for the elastic vapor to make its 
 escape, the report was scarcely audible at the 
 
 1 Phil. Trans., 1797, p. 222. See also Alembic Club Reprints, 
 No. 12, p. 20, 
 
14 LIQUEFACTION OF GASES 
 
 distance of a few paces. ... In many of the 
 experiments, in which the elastic vapor was 
 confined, this feeble report attending the ex- 
 plosion of the powder was immediately followed 
 by another noise totally different from it, which 
 appeared to be occasioned by the falling back 
 of the weight." He also calls attention to the 
 small degree of expansive force of the confined 
 elastic vapor when it was allowed to escape. 
 Instead of rushing out with a loud report, the 
 gas escaped with a hissing noise. The aqueous 
 vapor produced in these experiments was evi- 
 dently condensed to a liquid, but, in the light 
 of the researches of Andrews, it seems that the 
 temperature produced by the explosion would 
 be higher than 31, the critical temperature 
 of carbon dioxide. If this is true, the carbonic 
 acid, of course, would not have been condensed 
 to the liquid state. 
 
 In 1799 Van Marum 1 published an account of 
 some experiments which were performed several 
 years earlier. The main object of these observa- 
 tions was to study the effect of low pressures on 
 different liquids. Some of the liquids, he stated, 
 were entirely converted to vapors by this method. 
 When the pressure was increased, these vapors 
 
 1 Gilbert's Ann., I, p. 145. 
 
EARLY HISTORY 15 
 
 condensed again to the liquid state. By means of 
 an air-pump, he compressed ammonia gas, and 
 found that, as the pressure increased, the volume 
 did not decrease in accordance with Boyle's law. 
 At a pressure of three atmospheres, he says, drops 
 of liquid ammonia were formed. It is very proba- 
 ble, however, that the gas used in this experiment 
 contained a small quantity of aqueous vapor. 
 
 Fourcroy and Vanquelin 1 made a series of ex- 
 periments, in which they subjected certain gases 
 to low temperatures. In Accum's Chemistry, I, 
 P- 337> we nn d the statement that ammonia gas 
 was liquefied in these experiments. It seems, how- 
 ever, that aqueous ammonia was used in these ob- 
 servations, and that the pure gas was not liquefied. 
 These experimenters also subjected hydrochloric 
 acid gas, sulphuretted hydrogen, and sulphurous 
 acid gas to low temperatures. They did not suc- 
 ceed, however, in liquefying any of these gases. 
 The authors state that they obtained a temperature 
 of - 40. 
 
 During this same year Guyton de Morveau 2 
 experimented with ammonia gas at low tempera- 
 tures. It is probable here, as in the observations 
 of Fourcroy and Vanquelin, that the gas contained 
 moisture. The method of drying was to pass the 
 
 1 Ann. de Chim., 29, p. 281. 
 
 2 Ann. de Chim., 29, pp. 290, 297. 
 
16 LIQUEFACTION OF GASES 
 
 gas into a glass vessel, the temperature of which 
 was 21. 25. The object of this was to convert 
 the aqueous vapor into ice. From this balloon 
 the uncondensed gas passed to a second glass 
 vessel, where it was subjected to a temperature 
 of 43. 25. At this temperature drops of liquid 
 formed on the interior surfaces of the containing 
 vessel. The author states that ammonia, dried 
 as completely as possible by subjecting to a tem- 
 perature of 21, condenses to a liquid at 48. 
 It is not likely that the gas was rendered com- 
 pletely dry by this method ; and the author calls 
 attention to the fact that the liquid, produced 
 at 48, probably contained a small quantity of 
 water. 
 
 In 1805 Stromeyer subjected arsine to low tem- 
 peratures. Thenard 1 says that the gas was con- 
 densed to a liquid in these experiments. Another 
 reference 2 states that "Professor Stromeyer con- 
 densed the gas so far as to reduce it in part to a 
 liquid, by immersing it in a mixture of snow and 
 muriate of lime, in which several pounds of quick- 
 silver had been frozen in the course of a few 
 minutes." Faraday remarks that, "from the cir- 
 cumstance of its being reduced only in part to a 
 liquid, we may be led to suspect that it was rather 
 
 1 Traite de Ckimie, I, p. 373. 
 
 2 Nicholson's your., 19, p. 382. 
 
EARLY HISTORY 17 
 
 the moisture of the gas that was condensed than 
 the gas itself." 
 
 Numerous text-books refer to the liquefaction of 
 sulphurous acid gas by Monge and Clouet. Thom- 
 son, in his System of Chemistry, 6th ed., II, p. 
 1 1 8, says the gas was condensed when exposed 
 to a temperature of 18. In Accum's Chemis- 
 try, I, p. 345, and Murray's Chemistry, II, p. 405, 
 we find the statement that the condensation of 
 sulphurous acid was effected by the application of 
 strong pressure and intense cold. 
 
 The experiments of Northmore 1 in 1805-1806 
 are among the most important of the earlier obser- 
 vations. His object was to determine the effect 
 of pressure on the affinities of gases. His results, 
 however, show that some of the gases were lique- 
 fied. The apparatus consisted of an exhausting 
 syringe, a condensing pump, a connecting spring- 
 valve, and glass receivers. When chlorine was 
 compressed, he observed the formation of a yel- 
 low, extremely volatile, fluid. This appears to be 
 the first reliable account of the liquefaction of 
 chlorine. He also liquefied hydrochloric acid gas 
 and sulphurous acid gas. The latter experiment, 
 he says, corroborates the statement of Monge and 
 Clouet, that sulphurous acid is condensed to a 
 
 1 Nicholson's Jour., 12. p. 368, and 13, p. 233. Also Alembic 
 Club Reprints, No. 12, pp. 69-79. 
 
i8 LIQUEFACTION OF GASES 
 
 liquid by the simultaneous application of strong 
 pressure and low temperature. In the experi- 
 ments on carbonic acid, Northmore remarks that 
 the receiver very unexpectedly bursts with violence. 
 
 These observations were very successful, consid- 
 ering the period in which they were made. It is 
 difficult to understand why the results obtained by 
 Northmore were no incentive to further investi- 
 gation. The fact remains, however, and these 
 experiments have only an historical interest. 
 
 In 1822, Cagnaird de la Tour 1 made a rather 
 extended series of experiments by heating liquids 
 in sealed glass tubes. This work is of considera- 
 ble importance, in as much as it is a forerunner 
 of the researches of Andrews on critical constants. 
 Tour called attention to the following facts : 
 
 1. Alcohol, naphtha, and ether, when subjected 
 to heat and pressure, are converted into vapors in 
 a space slightly greater than double that of each 
 liquid. 
 
 2. The increased pressure, due to the presence 
 of air in the tube, did not prevent the evaporation 
 of the liquid in the same space. The expansion, 
 however, was more regular. 
 
 3. The experiments on water were unsatisfac- 
 tory, owing to imperfections in the apparatus. 
 
 1 Ann. de Chim. et de Phys., 21, pp. 127, 178. 
 
EARLY HISTORY 
 
 Ether is completely vaporized in a space some- 
 what less than double that of the liquid, at a tem- 
 perature of about 1 60. The pressure exerted by 
 ether at this temperature is from 37 to 38 atmos- 
 pheres. At a temperature of 200, alcohol vaporizes 
 in a space less than three times that of the liquid^. 
 The pressure in this case is about 1 19 atmospheres. 
 
 The measurements on ether are given in the 
 following tables : l 
 
 Volume of Liquid = 7 Volume of Vapor = 20 
 
 Temperature in 
 Reaumur Degrees 
 
 Pressures in Atmos- 
 pheres 
 
 Differences 
 
 80 
 
 5 
 
 * 
 
 9 
 
 7 
 
 2 
 
 100 
 
 10 
 
 3 
 
 110 
 
 12 
 
 2 
 
 120 
 
 18 
 
 6 
 
 I 3 
 140 
 
 22 
 28 
 
 4 
 6 
 
 State of Vapor I 50 
 1 60 
 
 37 
 
 48 
 
 9 
 ii 
 
 170 
 1 80 
 IQO 
 200 
 
 59 
 68 
 
 78 
 86 
 
 ii 
 9 
 
 10 
 
 8 
 
 260 
 
 130 
 
 
 
 1 Ann. de Chim. et de Phys., 22, pp. 411, 412. 
 
20 
 
 LIQUEFACTION OF GASES 
 
 The results obtained by heating one half of this 
 quantity of liquid in a tube of the same volume 
 were as follows : 
 
 Volume of Liquid = 35 Volume of Vapor = 20 
 
 Temperature in 
 Reaumur Degrees 
 
 Pressures in Atmos- 
 pheres 
 
 Differences 
 
 100 
 
 14 
 
 
 
 no 
 
 17 
 
 3 
 
 120 
 
 22 
 
 5 
 
 I 3 
 
 28 
 
 6 
 
 I4O 
 
 35 
 
 7 
 
 State of Vapor 1 50 
 
 42 
 
 7 
 
 1 60 
 
 5 
 
 8 
 
 170 
 
 58 
 
 8 
 
 1 80 
 
 63 
 
 5 
 
 I 9 
 
 66 
 
 3 
 
 200 
 
 70 
 
 4 
 
 
 
 
 
 4 
 
 
 
 
 
 4 
 
 
 
 
 
 3 
 
 260 
 
 94 
 
 4 
 
 The important part of these observations is, 
 that the liquid disappeared at the same tempera- 
 ture in both cases. This was an indication that, 
 regardless of the pressure, the substance cannot 
 exist in the liquid state when heated above a cer- 
 tain temperature. These results exerted but little 
 influence until after the work of Andrews nearly 
 fifty years later. 
 
EARLY HISTORY 21 
 
 So far the investigations are only of historical 
 interest. The results which are recorded on the 
 preceding pages have been of little value to the 
 later experimenters. From this point the work 
 becomes more systematic. The interesting obser- 
 vations of Cagnaird de la Tour may be said to 
 close the first chapter in the liquefaction of gases. 
 
CHAPTER II 
 
 HERE, as in many other lines of scientific re- 
 search, we must turn to Faraday for the first sys- 
 tematic observations. This great experimenter has 
 taken the initiative step in many lines of scientific 
 investigation, and is the author of numerous and 
 important discoveries. In 1823, Faraday acci- 
 dentally liquefied chlorine. Although an accident, 
 he perceived the full significance of the result, 
 and followed the observation with two elaborate 
 series of experiments, in which he exhausted all 
 his resources in the endeavor to liquefy the so- 
 called permanent gases. 
 
 At the suggestion of Davy, Faraday experi- 
 mented with the hydrate of chlorine and studied 
 the effect of heating it in a closed glass tube. 
 When placed in water at 60, he observed no 
 change; but when placed in water at 100, a yel- 
 low gas was set free. On cooling, this gas con- 
 densed to a yellow liquid which resembled the 
 chloride of nitrogen. During the progress of this 
 experiment, Dr. Paris 1 happened to enter the lab- 
 
 1 Tyndall, Faraday as a Discoverer, p. 14. 
 22 
 
EXPERIMENTS OF FARADAY 23 
 
 oratory. " Seeing the oily liquid in the tube, he 
 rallied the young chemist for his carelessness in 
 employing soiled vessels. . . . Early the next morn- 
 ing, Dr. Paris received the following note : 
 
 " ' DEAR SIR, The oil you noticed yesterday turns out 
 to be liquid chlorine. 
 
 " ' Yours faithfully, 
 
 <"M. FARADAY.'" 
 
 The chlorine had been liquefied by pressure. 
 When the tube was opened, the contents ex- 
 ploded. 
 
 Faraday made a number of observations to 
 prove that this yellow liquid was chlorine. 
 Among other experiments, he subjected dry 
 chlorine gas to pressure, and obtained a liquid 
 which was similar in all respects to that obtained 
 from the hydrate. He also made some ap- 
 proximate determinations of the density of liquid 
 chlorine. 
 
 Almost simultaneous with this work of Fara- 
 day, Davy 1 liquefied hydrochloric acid gas by 
 treating ammonium chloride with sulphuric acid 
 in a closed tube. He then substituted ammonium 
 carbonate for the chloride, and endeavored to 
 liquefy carbon dioxide. Only one experiment was 
 
 1 Note to Faraday's article on " Liquefaction of Chlorine." 
 
24 LIQUEFACTION OF GASES 
 
 made, and, in that case, the tube burst. At the 
 suggestion of Davy, Faraday continued the work 
 alone. 
 
 In these experiments, a number of gases were 
 subjected to pressure, and the results were pub- 
 lished in one article 1 in 1823. 
 
 Sulphurous Acid 
 
 Mercury and concentrated sulphuric acid were 
 placed in a closed, bent glass tube which was 
 afterward sealed (Fig. i). The end of the tube 
 
 containing the reacting 
 substances was heated, 
 while the other end was 
 kept cool by means of 
 moistened paper. Sul- 
 phurous acid gas was gen- 
 erated, and, after the 
 FIG. i. 
 
 sulphuric acid had become 
 
 saturated, it was evolved, and condensed to a liquid 
 in the cold end of the tube. Faraday remarks 
 that sulphurous acid forms a limpid, colorless, 
 highly fluid liquid, which does not solidify at o F. 
 When the tube was opened, a portion of* the liquid 
 evaporated rapidly. This lowered the tempera- 
 
 1 Phil. Trans., 113, pp. 189-198. See also Akmbic Club 
 Reprints, No. 12, pp. 10-19. 
 
EXPERIMENTS OF FARADAY 25 
 
 ture of the remaining portion, so that it re- 
 mained a liquid 'for some time at the ordinary 
 atmospheric pressure. A piece of ice thrown 
 into the liquid caused rapid boiling. Faraday 
 then took dry sulphurous acid gas, and subjected 
 it to pressure in a glass tube by means of a 
 syringe. When the tube was cooled to o F., the 
 gas condensed to a liquid which was similar to 
 that obtained from the mercury and sulphuric acid. 
 
 Sulphuretted Hydrogen 
 
 In this experiment the sealed glass tube con- 
 tained hydrochloric acid and sulphuret of iron. 
 On warming the contents, sulphuretted hydrogen 
 was generated and condensed in the longer end 
 of the tube, which was surrounded by a mixture of 
 ice and salt. The resulting liquid was colorless, 
 limpid, and excessively fluid. By the introduction 
 of a small gauge intOj the tube, he found the pres- 
 sure to be nearly 17 atmospheres at the tempera- 
 ture of 50 F. 
 
 Carbonic Acid 
 
 The materials used in this case were ammonium 
 carbonate and concentrated sulphuric acid. The 
 glass tubes, however, were much stronger than 
 those used in the previous experiments. Even 
 then a number of violent explosions resulted. 
 
26 LIQUEFACTIOiN OF GASES 
 
 The carbonic acid was condensed to a color- 
 less, extremely fluid liquid. The vapor pressure 
 at 32 F. was measured, and found to be 36 
 atmospheres. 
 
 Euchlorine 
 
 This gas was generated from a mixture of potas- 
 sium chlorate and sulphuric acid in a closed tube. 
 On cooling, it condensed to a yellow, transparent 
 liquid. 
 
 Nitrous Oxide 
 
 Ammonium nitrate, dried by heating in the air to 
 partial decomposition, was heated in a closed bent 
 tube. When the gases were cooled, two liquids 
 separated. One was found to be water containing 
 a little nitrous oxide in solution, while the other 
 was nitrous oxide. The appearance of the liquid 
 was similar to that of carbonic acid. The vapor 
 pressure at 45 F. was found t to be 50 atmospheres. 
 These experiments were accompanied by a number 
 of violent explosions. 
 
 Cyanogen 
 
 Pure, dry, mercuric cyanide was treated in a 
 manner similar to that of ammonium nitrate. The 
 resulting cyanogen was condensed in the cold 
 end of the tube to a limpid, colorless, liquid. 
 
EXPERIMENTS OF PERKINS 27 
 
 The vapor pressure at 45 F. was found to be 
 3.6 atmospheres. 
 
 Ammonia 
 
 Dry silver chloride was allowed to absorb pure, 
 dry ammonia gas. The substance was then heated 
 in a closed tube. The gas was again set free, and 
 was condensed to a colorless, transparent liquid in 
 the longer arm of the tube, which was cooled by 
 means of ice and water. The vapor pressure at 
 50 F. was about 6.5 atmospheres. 
 
 Faraday repeated the experiments of Davy on 
 hydrochloric acid. He also determined a number 
 of physical constants of these various liquids. He 
 attempted to liquefy hydrogen, oxygen, etc., but 
 without success. Twenty-one years later, he pub- 
 lished another series of experiments on the lique- 
 faction of gases. These observations will be 
 considered in their proper place. Meanwhile we 
 may turn our attention to the work of other 
 investigators. 
 
 EXPERIMENTS OF PERKINS 
 
 In a note, 1 published in 1823, Perkins calls 
 attention to a series of experiments on compressed 
 air. The details of the work were not published 
 until a few years later. The primary object of the 
 
 1 Annals of Phil, VI, p. 66. 
 
28 LIQUEFACTION OF GASES 
 
 investigation was to study the effect of high press- 
 ure on liquids. The author states, however, that 
 when ordinary air was subjected to a pressure of 
 1000 atmospheres, drops of liquid began to form, 
 and that at 1200 atmospheres, a beautiful trans- 
 parent liquid could be seen on the surface of the 
 mercury. The liquid remained in the tube after 
 the pressure had been removed. Perkins sup- 
 posed that the liquid thus formed was condensed 
 air. From the researches of later experimenters, 
 however, it is evident that ordinary air cannot be 
 liquefied under such conditions. The liquid which 
 appeared at the surface of the mercury was evi- 
 dently formed by the condensation of aqueous 
 vapor. Perkins also remarks that he liquefied car- 
 buretted hydrogen at a pressure of 1200 atmos- 
 pheres. He developed a method for obtaining 
 extremely high pressures. 
 
 / 
 
 EXPERIMENTS OF BUSSY 
 
 In 1824, Bussy 1 published a series of observa- 
 tions on the liquefaction of gases. His method of 
 procedure was entirely different from that of Fara- 
 day. He subjected the gases to low temperatures, 
 but did not increase the pressure. He condensed 
 
 1 Ann. de Ckim. et de Phys., XXVI, p. 63; Pogg. Ann., I, p. 237, 
 1824. 
 
EXPERIMENTS OF BUSSY 29 
 
 sulphurous acid to a colorless, transparent liquid, 
 at the ordinary atmospheric pressure, by subject- 
 ing it to a temperature of 18 to 20. Bussy 
 observed that if the liquid sulphurous acid was 
 allowed to evaporate in the air, the temperature 
 sank to - 57, and that if evaporated under 
 reduced pressure, the temperature sank to 65. 
 His work was a great advance in the liquefaction 
 of gases, inasmuch as he made use of this method 
 to produce low temperatures. By means of liquid 
 sulphurous acid, he liquefied chlorine and ammo- 
 nia, and in a similar manner he liquefied and 
 solidified cyanogen. This method of lowering the 
 temperature soon came into general use for both 
 scientific and industrial purposes.^ 
 
 Heat of Vaporization 
 
 Inasmuch as the evaporation of liquids becomes, 
 at this period, an important method for the pro- 
 duction of low temperatures, we may briefly con- 
 sider, at this point, the -thermal change which 
 accompanies the passage of a substance from the 
 liquid to the gaseous state. 
 
 When a liquid changes to the gaseous state, a 
 certain quantity of heat is absorbed. The heat 
 required to change one gram of a liquid to a 
 vapor, at the same temperature and pressure, is 
 
30 LIQUEFACTION OF GASES 
 
 called the heat of vaporization. The heat absorbed 
 is used : 
 
 1. To increase the internal energy of the sub- 
 stance ; i.e. the energy of the particles must be 
 sufficiently increased to overcome the force of 
 cohesion. 
 
 2. To perform external work ; i.e. the volume of 
 the substance must be increased against a definite 
 pressure. 
 
 Let E represent the energy of the vapor, and E 
 the energy of the liquid, then the increase in the 
 internal energy is 
 
 E-EX 
 
 If Fand V\ represent the volumes of the vapor 
 and liquid, and p is the pressure, which remains 
 constant, then the external work performed is 
 equal to 
 
 P(V-VI). 
 
 The total energy and work required to vaporize a 
 liquid, therefore, is 
 
 Knowing the mechanical equivalent of heat, this 
 expression may be represented in thermal units ; 
 in this latter form it represents the heat of vapori- 
 zation. The numerical value of the heat of vapori- 
 zation depends upon the temperature. 
 
 Inasmuch as heat is absorbed during evaporation, 
 
EXPERIMENTS OF COLLADON 31 
 
 it is evident that a liquid cannot maintain a con- 
 stant temperature during the process of evapora- 
 tion, unless it receives heat from some external 
 source. Ordinarily, there is a constant flow of 
 heat from the surrounding bodies to the liquid 
 during evaporation. 
 
 Suppose, however, that this supply of heat from 
 the surrounding bodies is cut off by insulation, then 
 the heat required in the evaporation must be drawn 
 from the liquid itself. The temperature, of course, 
 will be lowered by this process. If the evapora- 
 tion be hastened, e.g. by reducing the pressure, the 
 temperature will be reduced more rapidly and to 
 a much greater degree. Leslie has shown that 
 water can be frozen by means of its own evapora- 
 tion under reduced pressure. 
 
 EXPERIMENTS OF COLLADON 
 
 In 1828 Colladon 1 constructed an apparatus 
 for the purpose of liquefying ordinary air. The 
 general plan of the apparatus is shown in figure 2. 
 
 The pressure was produced by means of an 
 hydraulic pump, and transmitted through the tube 
 Cc to the interior of a strong steel cylinder B, 
 which was partially filled with mercury. In this 
 cylinder was placed a glass tube T which was 
 
 1 Pictet, Ann. de Chim. et de Phys. [5], XIII, p. 226. 
 
LIQUEFACTION OF GASES 
 
 open at the 
 lower end, but 
 fused at the 
 upper end to 
 the thick- 
 walled, closed 
 glass tube /, 
 the interior 
 diameter of 
 which was from 1.5 
 to 2 mm. This tube 
 t projected from the 
 cylinder through the 
 elongated cover A, 
 above which it was 
 bent downward to 
 as shown in the 
 figure. The bent- 
 down portion of the 
 tube was placed in 
 a freezing mixture. 
 In case of condensation the 
 liquid, of course, would col- 
 lect in the lower end of the 
 tube /'. 
 
 Colladon experimented at 
 a temperature of 30 and 
 with pressures as high as 
 
 FIG. 3. 
 
EXPERIMENTS OF THILORIER 
 
 33 
 
 400 atmospheres. The results, however, were all 
 negative. 
 
 EXPERIMENTS OF THILORIER 
 
 In 1834 Thilorier l liquefied carbonic acid on a 
 much larger scale than had hitherto been done. 
 
 r\ 
 
 
 T 
 
 MG. 3. 
 
 The general plan of the apparatus 2 employed is 
 shown in figures 3 and 4. The gas is generated 
 
 1 Ulnstitut, 58, p. 197 ; Ann. de Chim. et ae Phys., 60, pp. 427, 432. 
 
 2 Lieb. Ann., 30, p. 122. 
 
 D 
 
34 LIQUEFACTION OF GASES 
 
 in the wrought-iron vessel A, and compressed by 
 means of its own pressure in the wrought-iron re- 
 ceiver F. These two vessels are similarly con- 
 structed. B represents a cross-section of the 
 cylinder A. This vessel is supported by means 
 of two pivots so that it can be rotated. Eighteen 
 hundred grams of bicarbonate of sodium, and 4^ 
 litres of water are placed in the generator. The 
 vessel D is then filled with concentrated sulphuric 
 acid, and placed in the cylinder with the sodium 
 bicarbonate and water. The stopper c is then 
 screwed tightly into the top of the generator, and 
 connection made with the receiver by means of 
 the copper tube E. On rotating the vessel A, the 
 sulphuric acid in D comes in contact with the car- 
 bonate. The carbonic acid gas which is evolved 
 in the reaction produces a very high pressure in 
 the generator. After several minutes the stop- 
 cocks mm are opened. The gas rushes immedi- 
 ately into the receiver until the pressures in the 
 two vessels are equal. A portion of the gas is 
 liquefied by pressure. The quantity of liquid pro- 
 duced may be largely increased by surrounding 
 the receiver with a freezing mixture. When equi- 
 librium has been established between the two ves- 
 sels A and F y the stop-cocks ;;/ and m are closed. 
 The pressure in the generator is then relieved, the 
 contents are removed, and a new charge is intro- 
 
EXPERIMENTS OF THILORIER 
 
 35 
 
 duced. The process is repeated in this manner 
 until from 2 to 3 litres of liquid carbonic acid have 
 
 FIG. 4. 
 
 collected in the receiver. From five to ten repeti- 
 tions are usually required. 
 
 In this way, Thilorier obtained liquid carbonic 
 
36 LIQUEFACTION OF GASES 
 
 acid in rather large quantities, and made a series 
 of observations on its physical properties. He 
 measured the vapor pressure at different temper- 
 atures and found it to be 36 atmospheres at o, 
 and 73 atmospheres at 30. He also determined 
 the specific gravity, and made observations on the 
 thermoscopic effect of the liquid. The condensed 
 gas was found to be insoluble in water and fat 
 oils, but soluble in all proportions in alcohol, ether, 
 naphtha, turpentine, and carbon disulphide. It 
 reacts with metallic potassium, but produces no 
 effect on lead, tin, copper, iron, etc. The abnor- 
 mal expansion of the liquid carbonic acid was also 
 investigated. 
 
 Thilorier made some important observations on 
 the production of low temperatures by the evapo- 
 ration of liquid carbonic acid. When a jet of the 
 liquid was directed upon the bulb of an alcohol ther- 
 mometer, the temperature sank rapidly to 90. 
 The sphere of action, however, in this as well as 
 in other cases was limited almost entirely to the 
 point of contact. The congelation of mercury was 
 confined to small portions of it ; and when the 
 hand was exposed to a jet of the liquid, a burning 
 sensation was felt, but the effect was confined to 
 very narrow limits. 
 
 The author explained this limited sphere of 
 action by the low conductivity and small capacity 
 
EXPERIMENTS OF THILORIER 37 
 
 for heat of carbonic acid. He says : " If gases 
 have little effect in the production of cold, it is not 
 so with vapors whose conductivity and capacity for 
 heat are much greater. I have, therefore, thought 
 that if a permanent liquid ether, for example 
 could be placed under the same conditions of 
 expansibility as liquefied gases, we might obtain 
 a frigorific effect much greater than that produced 
 by liquid carbonic acid. To accomplish this, ether 
 must be rendered explosible, and this I have easily 
 effected by mixing it with liquid carbonic acid. 
 In this intimate combination of the two liquids, 
 which dissolve each other in all proportions, ether 
 ceases to be a permanent liquid at the ordinary 
 atmospheric pressure ; it becomes expansible like 
 a condensed gas, still preserving its properties of 
 a vapor; namely, its conductivity and capacity 
 for heat." 
 
 The effects of a jet of explosible ether are more 
 pronounced and extend over a much greater area 
 than those produced by a jet of liquid carbonic 
 acid. Fifty grams of mercury can be solidified in 
 a few seconds by means of this mixture. 
 
 Solidification of Carbonic Acid 
 
 Thilorier not only liquefied carbonic acid, but 
 succeeded in converting it into a solid. The 
 
LIQUEFACTION OF GASES 
 
 receiver which contained the liquid carbonic acid 
 was connected with a vessel formed of two sepa- 
 rate halves (Fig. 5) by means of a small tube. 
 The sudden expansion produced a very low tem- 
 perature, and a portion of the gas solidified. The 
 author thought, however, that the solid particles 
 were composed of ordinary ice which resulted from 
 the freezing of aqueous vapor in the air. The 
 experiment was repeated before Arago, Thenard, 
 
 \ 
 
 FIG. 5. 
 
 and Dulong. The solid portions were then found 
 to be carbonic acid. " Gaseous at the ordinary 
 temperature and pressure, and liquid at o under a 
 pressure of 36 atmospheres, carbonic acid becomes 
 solid at a temperature of about 100 below zero, 
 and retains this new condition for several minutes 
 in the open air, without the necessity of any com- 
 pression." 
 
 The author calls attention also to the remarkable 
 difference in the behavior of the liquid and solid 
 
EXPERIMENTS OF ADDAMS 39 
 
 acid. While in the liquid state, he says, the 
 elastic force is exceedingly great, being equal 
 in explosive power to an equal weight of gun- 
 powder. The solid, on the other hand, disappears 
 insensibly by slow evaporation. A fragment of 
 solid carbonic acid, he adds, slightly touched by 
 the ringer, glides rapidly over a polished surface, 
 as if sustained by the gaseous atmosphere with 
 which it is constantly surrounded, until it is entirely 
 dissipated. The vaporization of solid carbonic 
 acid is complete. It leaves but rarely a slight 
 humidity, which may be attributed to the action of 
 the air on a cold body, the temperature of which 
 is far below that of freezing mercury. 
 
 The solidification of large quantities of carbonic 
 acid has been of great value to later experimenters 
 for the production of low temperatures. The solid 
 acid moistened with ordinary ether is known as 
 Thilorier's Mixture. A temperature of 
 
 -110=- i66F. 
 can be obtained by means of this mixture. 
 
 Modification of Thilorier's Apparatus 
 
 In a paper read before the British Association 
 for the Advancement of Science, 1 Addams calls 
 attention to three forms of apparatus which had 
 
 1 Proceedings for 1838, p. 70. 
 
40 LIQUEFACTION OF GASES 
 
 been used by him to liquefy and solidify carbonic 
 acid on a large scale. The first method is me- 
 chanical, in which powerful hydraulic pumps are 
 used to force the gas from one vessel into a 
 second, by filling the first with water, saline solu- 
 tions, oil, or mercury. The second form of appa- 
 ratus is a modification of that invented and used 
 by Thilorier. The third includes the mechanical 
 and the chemical methods, by means of which 
 much of the acid formed in the generator is pre- 
 served ; whereas by the arrangement of Thilorier, 
 two parts in three rush into the atmosphere and 
 are lost. The apparatus is provided with two 
 gauges, one to determine when the generator is 
 filled with water, and the other to indicate the 
 quantity of liquid acid in the receiver. By means 
 of this apparatus, Addams prepared liquid carbonic 
 acid in considerable quantity and measured the 
 elastic force at different temperatures. The solid 
 carbonic acid used by Faraday in his second series 
 of experiments on the liquefaction of gases was 
 prepared by Addams. 
 
 EXPERIMENTS ON HYDROGEN AND OXYGEN 
 
 In these experiments Maugham a attempted to 
 liquefy hydrogen and oxygen by means of press- 
 
 1 Proc. Brit. Assoc.for Adv. of Sci. (1838), p. 73. 
 
EXPERIMENTS OF TORREY 41 
 
 ure. The pressure was obtained by the elec- 
 trolysis of acidulated water in closed, strong glass 
 U-tubes. In this manner, the author was able to 
 burst tubes of the strongest glass. He suggested 
 that if the experiments were carried out at low 
 temperatures, the hydrogen and oxygen would 
 probably condense to the liquid state. This 
 method for obtaining high pressures has been used 
 by some of the later experimenters. 
 
 
 OBSERVATIONS OF TORREY 
 
 These experiments were not published by the 
 author, but the results were communicated in 
 private letters to Dr. Silliman, who published them 
 two years later. 1 The editors make the following 
 statement : 
 
 " We have been, from time to time, informed by 
 letters from Prof. John Torrey of New York, of 
 his progress in the condensation of gases, and 
 although not intended for publication, we now 
 take the liberty to give some citations from his 
 letters." 
 
 April 11, 1837, Torrey writes that, although 
 he was unable to procure a suitable form of 
 apparatus in which to condense carbonic acid, he 
 had successfully carried out numerous experiments 
 
 1 Silliman' 1 s Jour., 35, p. 374, 1839. 
 
42 LIQUEFACTION OF GASES 
 
 with glass tubes. A month later he forwarded a 
 tube containing liquid carbonic acid to Dr. Silli- 
 man. Torrey also liquefied sulphurous acid and 
 chlorochromic acid. Perfectly dry phosphorus, 
 he says, is not inflamed by liquid chlorochromic 
 acid, but if moist in the slightest degree, it will 
 burn with a loud explosion, requiring particular 
 precautions. In another letter, he remarks, " I 
 have been shooting with an air-gun, using liquid 
 carbonic acid for throwing the balls, and I hope 
 soon to emulate Perkins' steam gun." 
 
 EXPERIMENTS OF MITCHELL 
 
 In 1839 Mitchell 1 modified the apparatus of 
 Thilorier somewhat, and obtained both liquid and 
 solid carbonic acid. When the liquefied gas is 
 allowed to escape into an iron receiver, he says, 
 " a large portion of the liquid is instantly expanded 
 into gas which escapes through a tube. The 
 coldness consequent on the enormous expansion 
 freezes another portion of the liquid, which then 
 falls to the bottom of the receiver. About one 
 drachm of solid matter is thus formed for each 
 ounce of liquid." 
 
 Mitchell also made a series of observations on 
 the physical constants of liquid and solid carbonic 
 
 1 Sillimatfs Jour., 35, p. 346. 
 
EXPERIMENTS OF AIME 43 
 
 acid. By means of the solid acid he solidified 
 mercury, and experimented with the solid metal. 
 Liquid sulphurous acid freezes at about 80, 
 while ordinary ether, he says, when subjected to 
 a temperature of 100 is not, in the slightest 
 degree, altered. He further adds that, "when a 
 piece of solid carbonic acid is pressed against a 
 living animal surface, it drives off the circulating 
 fluids, and produces a ghastly white spot. If held 
 for fifteen seconds it raises a blister, and if the 
 application be continued for two minutes, a deep 
 white depression with an elevated margin is per- 
 ceived ; the part is killed, and a slough is in time 
 the consequence. I have thus produced both 
 blisters and sloughs, by means nearly as prompt 
 as fire, but much less alarming to my patients." 
 Experiments were also made with a number of 
 liquids and metals, by placing them in liquid 
 carbonic acid. 
 
 EXPERIMENTS OF AIME (I843) 1 
 
 In these experiments, ethylene, nitric oxide, 
 nitrogen, hydrogen, carbon monoxide, and oxygen 
 were subjected to very high pressures. The press- 
 ure was obtained by sinking the gases in suitable 
 vessels to great depths in the ocean. The method, 
 
 1 Ann. de Chim. et de P/iys., 8, p. 275, 1843. 
 
44 
 
 LIQUEFACTION OF GASES 
 
 of course, is unsatisfactory, inasmuch as the obser- 
 vations cannot be made at the time when the 
 pressure is applied. 
 
 The following table will show the relative de- 
 crease in volume for the different gases when 
 subjected to pressure : 
 
 Substance 
 
 Pressure 
 
 Ratio of Original Volume 
 to Final Volume 
 
 Oxygen 
 
 83 atmospheres 
 
 9 0: 
 
 Ethvlene 
 
 124 
 
 356: 
 
 Nitric oxide 
 
 165 
 
 251: 
 
 Carbon monoxide 
 
 165 
 
 180: 
 
 Oxygen 
 
 165 
 
 160: 
 
 Fluosilicon 
 
 105 " 
 
 350: 
 
 Aime thought that the ethylene and fluosilicon 
 were condensed to the liquid state. The evi- 
 dence, however, is not conclusive. Hydrogen and 
 nitrogen were subjected to a pressure of 220 
 atmospheres, without showing any indications of 
 liquefaction. The observations seem to show de- 
 viations from Boyle's law, but the results obtained 
 for oxygen are somewhat contradictory. 
 
 EXPERIMENTS OF FARADAY 
 
 In 1845 Faraday 1 published a second series of 
 observations. In the introduction, he says: "The 
 
 1 Phil. Trans., 135, p. 155; also, Alembic Club Reprints, No. 12, 
 P- 33- 
 
EXPERIMENTS OF FARADAY 45 
 
 experiments formerly made on the liquefaction of 
 gases, and the results which from time to time 
 have been added to this branch of knowledge, 
 especially by Thilorier, have left a constant de- 
 sire on my mind to renew the investigation. This, 
 with considerations arising out of the apparent 
 simplicity and unity of the molecular constitution 
 of all bodies when in the gaseous or vaporous state, 
 which may be expected, according to the indica- 
 tions given by the experiments of Cagnaird de la 
 Tour, to pass by some simple law into the liquid 
 state, and also the hope of seeing nitrogen, oxy- 
 gen, and hydrogen, either as liquid or solid bodies, 
 and the latter probably as metal, have lately in- 
 duced me to make many experiments on the sub- 
 ject." 
 
 The gases, in these experiments, were subjected 
 to the simultaneous influence of high pressure and 
 low temperature. The pressure was obtained by 
 the use of two air-pumps. "The first pump had 
 a piston of an inch in diameter, and the second 
 a piston of only half an inch in diameter, and 
 these were so associated by a connecting pipe, 
 that the first pump forced the gas into and through 
 the valves of the second, and then the second 
 could be employed to throw forward this gas, 
 already compressed to 10, 15, or 20 atmospheres, 
 into its final recipient at a much higher pressure." 
 
46 LIQUEFACTION OF GASES 
 
 The pressure exerted by certain gases when gen- 
 erated in closed, strong glass vessels was also 
 employed as a means of compression. 
 
 The low temperatures were produced by means 
 of a bath of Thilorier's mixture of solid carbonic 
 acid and ether. This mixture was allowed to 
 evaporate, at the ordinary atmospheric pressure, 
 in an open earthenware dish which rested in a 
 second larger vessel; the space between the two 
 being filled with dry flannel. Faraday modified 
 the method, however, and obtained a much lower 
 temperature by evaporating the mixture under 
 reduced pressure. 
 
 The influence of pressure is shown by the fol- 
 lowing table : 
 
 Pressure in Inches of Mercury 
 
 Temperature of Mixture 
 
 28.4 
 
 - 70 
 
 19.4 
 
 - 80.3 
 
 9.4 
 
 - 8 5 
 
 7-4 
 
 - 88.3 
 
 5-4 
 
 - 90.6 
 
 3-4 
 
 - 95 
 
 2.4 
 
 - 98.3 
 
 1.4 
 
 106.6 
 
 1.2 
 
 -no = - i66F. 
 
 The temperatures were measured by means of 
 an alcohol thermometer, and Faraday remarks that 
 
EXPERIMENTS OF FARADAY 47 
 
 the actual temperatures were probably from 5 to 
 6 lower than those recorded. 
 
 The gases were condensed in tubes, the forms 
 of which are shown in figures 6 and 7. "These 
 tubes were of green bottle glass, being from J to ^ 
 
 FIG. 6. 
 
 of an inch in external diameter, and from -fa to -fa 
 of an inch in thickness. They were chiefly of two 
 kinds, about 1 1 and 9 inches in length, the one, 
 when horizontal, having a curve downward near 
 the end to dip into the cold bath, and 
 the other, being in form like an inverted <Q{) 
 
 siphon, could have the bend cooled in 
 the same manner when necessary. In- 
 to the straight part of the horizontal 
 tube, and the longer leg of the siphon 
 tube, pressure-gauges were introduced 
 when required. Caps, stop-cocks, and 
 connecting pieces were employed to 
 attach the glass tubes to the pumps, 
 and these, being of brass, were of the 
 usual character of those employed for 
 operations with gases, except that they were small 
 and carefully made." 
 
 These tubes were tested by means of a hydro- 
 
48 LIQUEFACTION OF GASES 
 
 static press. One of them sustained a pressure of 
 118 atmospheres without breaking, while another 
 burst at a pressure of 67 atmospheres. 
 
 The pressures were measured by means of a 
 small glass tube which contained a movable cylin- 
 der of mercury, one end of the tube being closed. 
 A pressure of 10 or 20 atmospheres would reduce 
 the volume of air in the gauge to ^ or ^ of its 
 volume at the ordinary atmospheric pressure. 
 After the tube had once been calibrated, it was 
 a simple matter to measure the various pressures. 
 
 The solid carbonic acid was preserved for ex- 
 perimental purposes by placing it in a glass vessel, 
 which rested in the middle of three concentric glass 
 jars, separated from each other by dry jackets of 
 woollen cloth. This method proved to be very 
 effectual. The results obtained are as follows : - 
 
 Olefiant Gas 
 
 This gas was condensed to a clear, colorless, 
 transparent liquid, but could not be solidified at 
 the lowest temperature obtainable by means of the 
 carbonic acid mixture. The author made a large 
 series of observations on the vapor pressures of 
 this liquid at different temperatures. He also sug- 
 gested condensation and evaporation as a method 
 of purifying the gas. 
 
EXPERIMENTS OF FARADAY 49 
 
 Hydriodic Acid 
 
 This gas was easily condensed by means of the 
 carbonic acid bath. At a temperature of about 
 51 it became a clear, transparent, colorless 
 solid, which exhibits fissures or cracks similar to 
 those in ordinary ice. 
 
 Hydrobromic Acid 
 
 This' gas became liquid at a temperature of 
 about 73. The fluid was colorless and trans- 
 parent. At and below 87 it is a solid, transpar- 
 ent, crystalline body. The liquid does not freeze 
 until reduced much lower than the above-mentioned 
 temperature, but being frozen by the carbonic acid 
 bath, it remains a solid until the temperature, in 
 rising, attains to 87. 
 
 Flnosilicon 
 
 At a pressure of 9 atmospheres and a tempera- 
 ture of - 107, this substance condensed to a 
 liquid. In this condition it was clear, transparent, 
 colorless, and very fluid, like hot ether. It could 
 not be solidified at the lowest temperature em- 
 ployed in these experiments. 
 
 Phosphine and Arsine 
 
 Phosphine was liquefied, but could not be solidi- 
 fied. Faraday remarks that in this experiment 
 
50 LIQUEFACTION OF GASES 
 
 he was able to condense only a portion of the gas, 
 and suggests the presence of another gas which 
 could not be condensed under the conditions of 
 the experiment. This gas, he says, is probably 
 another phosphuretted hydrogen, or hydrogen 
 itself. 
 
 In regard to arsine, he says, " this body, lique- 
 fied by Dumas and Soubeiran, did not solidify at 
 the lowest temperature to which I could submit it, 
 i.e. not at 110 below zero." 
 
 Fluoboron 
 
 This substance was prepared from fluorspar, 
 fused boracic acid, and strong sulphuric acid, in 
 a tube generator, and conducted into a condensing 
 tube under the generating pressure. The ordi- 
 nary carbonic acid bath did not condense it, but 
 the application of one cooled under the air-pump 
 produced liquefaction. Fluoboron then appeared 
 as a very limpid, colorless, clear liquid, showing 
 no signs of solidification. At the lowest tempera- 
 ture it appeared as mobile as hot ether. 
 
 Muriatic, Sulphurous, and Carbonic Acids 
 
 All attempts to solidify muriatic acid were un- 
 successful. 
 
 Sulphurous acid, at 76, became a crystalline, 
 
EXPERIMENTS OF FARADAY 51 
 
 transparent, colorless solid, which sank freely into 
 the liquid. 
 
 In regard to carbonic acid, Faraday says, " The 
 solidification of carbonic acid by Thilorier is one 
 of the most beautiful experimental results of 
 modern times. When the acid is melted and re- 
 solidified by a bath of low temperature, it appears 
 as a clear, transparent, crystalline, colorless body, 
 like ice ; so clear, indeed, that at times it was 
 doubtful to the eye whether anything was in the 
 tube." 
 
 Euchlorine and Sulphuretted Hydrogen 
 
 Euchlorine was easily condensed to a crystalline 
 solid, the color and appearance of which were 
 similar to those of potassium bichromate ; it was 
 moderately hard, brittle, and translucent. The 
 solid melted at a temperature of 60. In at- 
 tempting to resolidify the substance, a peculiar 
 phenomenon was observed. The author remarks : 
 " Some hours after, wishing to solidify the same 
 portion of euchloriae which was then in a liquid 
 state, I placed the tube in a bath at 80, but 
 could not succeed either by continuance of the 
 tube in the bath, or shaking the fluid in the tube ; 
 but when the liquid euchlorine was touched by a 
 platinum wire it instantly became solid, and exhib- 
 ited all the properties before described. There are 
 
52 LIQUEFACTION OF GASES 
 
 many similar instances among ordinary substances, 
 but the effect in this case makes me hesitate in 
 concluding that all gases which as yet have re- 
 fused to solidify at temperatures as low as 110, 
 cannot acquire the solid state at such a tem- 
 perature." 
 
 Sulphuretted hydrogen, at a temperature of 
 85, solidified to a white, crystalline, translucent 
 substance, which was made up of a confused mass 
 of crystals. 
 
 Ammonia and Cyanogen 
 
 Ammonia was condensed to a white, translucent, 
 crystalline solid, which melted at a temperature of 
 -85. 
 
 Cyanogen was also obtained as a transparent 
 crystalline solid, the specific gravity of which was 
 nearly the same as that of the liquid. 
 
 Nitrous Oxide 
 
 This gas was solidified, at a temperature of 
 -i 00, to a beautiful, clear, crystalline, colorless 
 body. The pressure of the vapor rising from the 
 solid was less than one atmosphere. Hence " it 
 was concluded that liquid nitrous oxide could not 
 freeze itself by evaporation into the atmosphere, 
 as carbonic acid does; and this was found to be 
 
EXPERIMENTS OF FARADAY 
 
 53 
 
 true." Natterer observed the same fact a year 
 earlier. Faraday called attention to the fact that 
 very low temperatures could be obtained by the 
 evaporation of nitrous oxide. He says : " It is 
 probable that this substance may be used in cer- 
 tain cases, instead of carbonic acid, to produce 
 degrees of cold far below those which the latter 
 body can supply. . . . There is no doubt that the 
 substance placed in vacuo would acquire a tem- 
 perature lower than any as yet known." 
 
 Faraday made a large series of observations on 
 the vapor pressures of these condensed gases at 
 different temperatures. The following table con- 
 tains one result from each of the different sub- 
 stances : 
 
 Substance 
 
 Temperature 
 
 Pressure 
 
 
 
 
 43 4 atmospheres 
 
 
 
 
 Hydriodic acid .... 
 
 
 
 3-97 " 
 
 Fluosilicon 
 
 
 
 7O O " 
 
 Fluoboron 
 
 
 
 IO O " 
 
 
 J 3 
 
 
 Muriatic acid 
 
 
 
 26 2 " 
 
 Sulphurous acid .... 
 
 
 
 1.53 
 
 Sulphuretted hydrogen . . 
 
 
 
 10. " 
 
 Carbonic acid 
 
 
 
 38.5 
 
 Cyanogen 
 
 
 
 2.37 
 
 Ammonia 
 
 
 
 4.44 
 
 Arsine 
 
 
 
 8 95 " 
 
 Nitrous oxide 
 
 
 
 32.0 " 
 
54 LIQUEFACTION OF GASES 
 
 The following substances could not be solidified 
 at a temperature of 110: chlorine, ether, alco- 
 hol, carbon disulphide, caoutchoucine, camphine, 
 and oil of turpentine. 
 
 Faraday also attempted to liquefy the so-called 
 permanent gases by subjecting them to low tem- 
 peratures and high pressures, but without success. 
 At the lowest temperature obtainable by means of 
 the carbonic acid mixture, these gases were sub- 
 jected to the pressures expressed in the following 
 table: 
 
 
 Atmospheres 
 
 Hydrogen .... . 
 
 27 
 
 Oxygen 
 
 eg 
 
 Nitrogen ... . 
 
 3 
 CQ 
 
 Nitric oxide 
 
 j^ 
 CQ 
 
 Carbon monoxide 
 
 4.O 
 
 Coal gas 
 
 7,2 
 
 
 
 In no case were there any signs of liquefaction. 
 The author says : " Thus, though as yet I have 
 not condensed oxygen, hydrogen, or nitrogen, the 
 original objects of my pursuit, I have added six 
 substances, usually gaseous, to the list of those 
 that could previously be shown in the liquid state, 
 and have reduced seven to the solid form." 
 
 Although Faraday was unsuccessful in his at- 
 
EXPERIMENTS OF FARADAY 55 
 
 tempts to liquefy the so-called permanent gases, 
 yet he predicted that if the temperature could be 
 sufficiently reduced, these gases would pass into 
 the liquid or solid state. The behavior of gases 
 and vapors at different temperatures, he says, 
 " gives great encouragement to the continuance 
 of those efforts which are directed to the conden- 
 sation of oxygen, hydrogen, and nitrogen, by the 
 attainment and application of lower temperatures 
 than those yet applied. If to reduce carbonic 
 acid from the pressure of two atmospheres to that 
 of one, we require, then, to abstract only about 
 half the number of degrees that is necessary to 
 produce the same effect with sulphurous acid, it 
 is to be expected that a far less abstraction will 
 suffice to produce the same effect with nitrogen or 
 hydrogen, so that further diminution of tempera- 
 ture, and improved apparatus for pressure, may 
 very well be expected to give us these bodies in 
 the liquid or solid state." 
 
 After such an elaborate series of observations, it 
 is not surprising to find that Faraday anticipated 
 the work of Andrews on critical constants. He 
 makes the following statement : " M. Cagnaird de 
 la Tour has shown that at a certain temperature 
 a liquid, under sufficient pressure, becomes clear, 
 transparent vapor, or gas, having the same bulk 
 as the liquid. At this temperature, or one a little 
 
56 LIQUEFACTION OF GASES 
 
 higher, it is not likely that any increase of pressure, 
 except perhaps one exceedingly great, would con- 
 vert the gas into a liquid. Now the temperature 
 of 110 below zero, low as it is, is probably above 
 this point of temperature for hydrogen, and per- 
 haps for nitrogen and oxygen, and then no com- 
 pression, without the conjoint application of a 
 degree of cold below that we have as yet obtained, 
 can be expected to take from them their gaseous 
 state. Further, as ether assumes this state before 
 the pressure of its vapor has acquired 38 atmos- 
 pheres, it is more than probable that gases which 
 can resist the pressure of from 27 to 57 atmos- 
 pheres at a temperature of 110 could never 
 appear as liquids, or be made to lose their gaseous 
 state at ordinary temperatures. They may proba- 
 bly be brought into the state of very condensed 
 gases, but not liquefied." 
 
 EXPERIMENTS OF NATTERER 
 
 In 1844 Natterer 1 published an article on the 
 liquefaction and solidification of carbonic acid, 
 nitrous oxide, etc. Historically considered, this 
 work should precede the experiments of Faraday, 
 which have just been outlined. The observations 
 of Natterer, however, extend over a period of a 
 
 1 Jour. Prakt. Chem., 31, p. 375; Pogg, Ann., 62, p. 132, 1844. 
 
EXPERIMENTS OF NATTERER 57 
 
LIQUEFACTION OF GASES 
 
 number of years, 
 and for that reason 
 are taken up at 
 this point. 
 
 The methods 
 employed in the 
 liquefaction of 
 gases were mate- 
 rially advanced by 
 this investigator. 
 He placed in the 
 hands of the ex- 
 perimenters a 
 method for obtain- 
 ing extremely high 
 pressures. He also 
 introduced a con- 
 venient vessel for 
 preserving lique- 
 fied gases. 
 
 A sketch of the 
 complete appa- 
 ratus l employed 
 by Natterer is 
 shown in figure 8. 
 The details can be 
 
 1 Jour. Prakt. Chem., 
 35, p. 169. 
 
EXPERIMENTS OF NATTERER 59 
 
 seen from figures 9 and 10. The tube A (Fig. 9) 
 is provided on one end with a screw. This end 
 of the tube is screwed air-tight into an opening 
 in the cylinder B of the pump, and is fastened 
 in that position by means of the ring A'. The 
 other end of the tube A is provided with notches, 
 so that it can be fastened to a flexible leather 
 or rubber sack. This sack is connected with a 
 receiver, in which the gas to be compressed is 
 kept. The cylinder B of the pump is an iron 
 tube, about 20 inches in length and 0.6 inches in 
 diameter. It consists of two tubes screwed into 
 a larger short section B' . In the interior of the 
 cylinder is a piston C (Fig. 10), provided with a 
 leather cap, and fastened to the piston rod D. On 
 turning the contrivance G', the piston is moved up 
 and down. At the lowest position the piston is 
 below the inlet tube A. The wrought-iron vessel 
 G is provided at the bottom with a valve which 
 opens inward, and at the top with an outlet cock g. 
 The vessel G has a capacity of about 20 litres. A 
 freezing mixture is placed in the chamber H. The 
 receivers were always tested, before use, to a press- 
 ure of 1 50 atmospheres. 1 
 
 1 The apparatus of Natterer has been improved by Bianchi in 
 Paris, and later by Ritchie in Boston. 
 
60 LIQUEFACTION OF GASES 
 
 Condensation of Carbonic Acid 
 
 The gas to be liquefied was first dried by pass- 
 ing through a bottle containing calcium chloride. 
 The pump was then started, and the cock g 
 opened. When all the air had been removed from 
 the pump and receiver, the cock was closed and 
 the compression begun. The cooling was accom- 
 plished by means of ice water in the vessel H. 
 The operation required from one to one and a half 
 hours. In this way Natterer obtained liquid car- 
 bonic acid in considerable quantity. The liquid 
 could be kept in the receiver for several months. 
 He also obtained large quantities of solid carbonic 
 acid and studied its properties. 
 
 Nitrous Oxide 
 
 Natterer was the first to obtain large quantities 
 of liquid nitrous oxide. He found the boiling 
 point to be 105. He also solidified the gas and 
 filtered the solid out of the liquid. The freezing 
 point was found to be 1 15. 
 
 Experiments on Air, Oxygen, Hydrogen, etc. 1 
 
 Natterer subjected carbon monoxide to a press^ 
 ure of 150 atmospheres, but observed no indi- 
 
 1 Wien. Ber., 5, p. 351; 6, pp. 557 and 570; 12, p. 199. 
 
EXPERIMENTS OF BERTHELOT 61 
 
 cations of liquefaction. This induced him to 
 construct an apparatus, by means of which a 
 pressure of 1000 atmospheres could be obtained. 
 With this apparatus he hoped to condense the so- 
 called permanent gases to the liquid state. This 
 hope, however, was not realized. Only negative 
 results were obtained. He then modified the 
 apparatus, and subjected the gases to pressures 
 as high as 3600 atmospheres (54,000 pounds per 
 square inch), and concluded that with pressure 
 alone these gases could not be liquefied. He next 
 endeavored to combine the high pressure with low 
 temperature, but without success. The low tem- 
 perature was obtained by means of solid carbonic 
 acid and ether. 
 
 The experiments of Natterer, although unsuc- 
 cessful as far as the liquefaction of the permanent 
 gases is concerned, show that these gases under 
 high pressures deviate considerably from Boyle's 
 law. Further reference to these variations will 
 be made at the end of the chapter. 
 
 EXPERIMENTS OF BERTHELOT 
 
 In 1850 Berthelot 1 constructed an apparatus, 
 by means of which gases could be liquefied with- 
 out danger. The pressure was obtained by the 
 
 Compt. rend., 30, p. 666. 
 
62 LIQUEFACTION OF GASES 
 
 dilation of a liquid. The apparatus consisted 
 simply of a thick-walled glass tube which was 
 closed at one end. The tube was filled with pure 
 dry mercury free from air, and then drawn out 
 into a capillary .at the open end. The large end 
 of the tube was placed in a water bath, and the 
 capillary portion in a current of the gas to be 
 compressed. The water bath was then heated, 
 and a portion of the mercury expelled. On cool- 
 ing, the mercury contracted, and the portion which 
 had been expelled was replaced by the gas. The 
 capillary portion of the tube was then sealed, and 
 the water bath again heated. 
 
 Berthelot succeeded, with tubes of this kind, in 
 condensing chlorine, ammonia, and carbonic acid 
 to the liquid state. He observed the abnormal ex- 
 pansion of liquid carbonic acid, which was pointed 
 out by Thilorier. Berthelot also endeavored to 
 liquefy nitric oxide, carbon monoxide, and oxy- 
 gen. The results in these cases, however, were 
 negative. The pressure obtained in the experi- 
 ment on oxygen was about 780 atmospheres. 
 The temperature of the gas in this case was 
 lowered by means of solid carbonic acid. Ber- 
 thelot came to the conclusion that pressure would 
 not produce liquefaction under all conditions of 
 temperature. 
 
EXPERIMENTS OF DRION 63 
 
 OBSERVATION OF DRION 
 
 In 1859 Drion 1 made some observations similar 
 to those made by Andrews a few years later. The 
 experiments, however, were not very elaborate. 
 Working with hydrochloric ether, nitrogen tetrox- 
 ide, and sulphurous acid, he found that the coef- 
 ficient of expansion of these substances in the 
 liquid state increases very rapidly with increasing 
 temperature. 
 
 He also observed that, when the liquids were 
 gradually heated in closed tubes to the tempera- 
 ture of complete vaporization, the surfaces lost 
 their definition and reflecting power, and became 
 nebulous in appearance. The nebulous zone in- 
 creased in size both upward and downward, and 
 at the same time became less distinct until the 
 tube appeared entirely empty. Similar appear- 
 ances were produced in the inverse order by 
 gradually cooling the tube. 
 
 When a capillary tube was partially immersed in 
 the liquid, the curvature of the meniscus and the 
 capillary elevation gradually decreased as the 
 temperature rose, and finally, just before complete 
 vaporization, the surface became plane, and the 
 level of the liquid within the tube was the same as 
 that of the liquid without the tube. 
 
 1 Ann. de Chim. et de Phys. [3], pp. 5 and 221. 
 
64 LIQUEFACTION OF GASES 
 
 Later, Drion and Loir l described a method for 
 liquefying gases in considerable quantity. They 
 lowered the temperature by the method of Bussy ; 
 i.e. by the evaporation of volatile liquids. These 
 experimenters, however, modified the method by 
 conducting finely divided currents of dry air 
 through the liquid during evaporation. In this 
 way Drion and Loir succeeded in liquefying a 
 number of gases, and studied their properties. 
 The low temperatures were measured by means of 
 an alcohol thermometer. 
 
 During the following year these experimenters 2 
 succeeded in obtaining a perfectly transparent 
 mass of solid carbonic acid. The temperature was 
 lowered by the evaporation of liquid sulphurous 
 acid and ammonia. The transparent mass of 
 carbonic acid, they state, consisted of cubical 
 crystals. 
 
 OBSERVATIONS OF MENDELEEFF 
 
 In 1861 Mendeleeff called attention to some 
 important relations between the gaseous and liquid 
 states of matter. In the concluding paragraph to 
 an article on " The Expansion of Liquids when 
 Heated above their Boiling Points," 3 he makes the 
 
 1 Phil. Mag. [4], 20, p. 202, 1860. 
 *Itrid. [4], 21, p. 495, 1861. 
 8 Lieb. Ann., 119, p. i, 1861. 
 
EXCEPTIONS TO BOYLE'S LAW 65 
 
 following statement : The absolute boiling point of 
 a liquid is that temperature at which the cohesion 
 and the heat of vaporization become equal to zero. 
 At this temperature the liquid changes to vapor, 
 regardless of the pressure and volume. The 
 absolute boiling point of alcohol he determined to 
 be 250, and that of water 580. The absolute 
 boiling point suggested by Mendeleeff corresponds 
 to the critical temperature introduced by Andrews 
 (see next chapter). 
 
 EXCEPTIONS TO BOYLE'S LAW 
 
 Before closing the present chapter, reference 
 should be made to some observations on the rela- 
 tion of pressure to the volume of a gas. These 
 experiments were made for most part to test the 
 validity of Boyle's law. Van Marum, in 1799(566 
 p. 15), called attention to the fact that ammonia gas 
 at high pressures does not obey this law. Other 
 isolated experiments have been recorded, but for 
 the present purpose we may begin with the obser- 
 vations of Despretz 1 in 1827. He determined the 
 relative compressibilities of a number of gases. 
 The volumes of the different gases at the begin- 
 ning of the experiment were equal to each other, 
 but as the pressure was increased, this equality of 
 
 1 Ann. de Chini. et de Phys. [2], 34, p. 335. 
 F 
 
66 LIQUEFACTION OF GASES 
 
 volumes was destroyed. Carbonic acid and am- 
 monia were found to be more compressible than 
 air. At pressures above 15 atmospheres, he 
 noticed that the volume of air was slightly less than 
 that of hydrogen. Similar observations were made 
 by Pouillet. In 1830 Arago and Dulong 1 experi- 
 mented with air at pressures ranging from i to 27 
 atmospheres. In these experiments the observed 
 volume was usually somewhat less than that calcu- 
 lated from Boyle's law. 
 
 In 1854 Natterer studied the effect of exceed- 
 ingly high pressures on oxygen, nitrogen, air, and 
 hydrogen. He subjected these gases to pressures 
 of more than 3000 atmospheres. The apparatus 
 consisted simply of a compression-pump and a 
 receiver, which was provided with a manometer. 
 The quantity of gas which the receiver contained 
 at the ordinary atmospheric pressure was increased 
 10, 20, 50, and 100 fold, and so on, and the corre- 
 sponding pressures measured. The following table 2 
 contains some of the results. The numbers in the 
 first column under each gas represent the volumes 
 which were forced into the receiver, while those 
 in the second column represent the corresponding 
 pressures in atmospheres. 
 
 1 Ann. de Chim. et de Phys. [2], 43, p. 74. 
 z Wien. Ber., 12, pp. 204-206, 1854. 
 
EXCEPTIONS TO BOYLE'S LAW 
 
 HYDROGEN 
 
 OXYGEN 
 
 NITROGEN 
 
 Volume 
 
 Pressure 
 
 Volume Pressure 
 
 Volume 
 
 Pressure 
 
 8 
 
 8 
 
 7 
 
 7 
 
 5 
 
 5 
 
 18 
 
 18 
 
 17 
 
 17 
 
 15 
 
 15 
 
 28 
 
 28 
 
 27 
 
 27 
 
 25 
 
 2 5 
 
 38 
 
 38 
 
 37 
 
 37 
 
 35 
 
 35 
 
 48 
 
 48 
 
 47 
 
 47 
 
 45 
 
 45 
 
 58 
 
 58 
 
 57 
 
 57 
 
 55 
 
 55 
 
 68 
 
 68 
 
 67 
 
 67 
 
 65 
 
 65 
 
 78 
 
 78 
 
 77 
 
 77 
 
 75 
 
 75 
 
 88 
 
 89 
 
 87 
 
 87 
 
 85 
 
 85 
 
 98 
 
 100 
 
 97 
 
 97 
 
 95 
 
 96 
 
 198 
 
 222 
 
 117 
 
 117 
 
 195 
 
 206 
 
 298 
 
 35 2 
 
 147 
 
 147 
 
 295 
 
 336 
 
 398 
 
 505 
 
 177 
 
 177 
 
 395 
 
 542 
 
 598 
 
 930 
 
 187 
 
 1 88 
 
 495 
 
 882 
 
 798 
 
 1584 
 
 207 
 
 210 
 
 595 
 
 1546 
 
 928 
 
 2154 
 
 507 
 
 670 
 
 705 
 
 2790 
 
 1008 
 
 2790 
 
 657 
 
 1354 
 
 
 
 A glance at this table will show that the meas- 
 urements were only approximations. There can 
 be no doubt from the results, however, that these 
 gases under high pressures deviate from Boyle's 
 law. The relative deviations increase with increas- 
 ing pressure. 
 
 Regnault 1 investigated this subject more thor- 
 oughly. He introduced the gas into an accurately 
 
 1 Mem. de I'Acad., 21, p 329. 
 
68 
 
 LIQUEFACTION OF GASES 
 
 calibrated glass tube, which was kept at a uniform 
 temperature. The gas was then compressed to a 
 certain fraction of the initial volume by means of 
 mercury, and the pressure again measured. If the 
 initial pressure and volume be represented by / 
 and Vty and the final pressure and volume by / 
 and v, the following table l will show that the prod- 
 uct of the pressure and volume does not remain 
 constant. 
 
 In every case, except hydrogen, the compressi- 
 
 AIR 
 
 NITROGEN 
 
 A 
 
 Pv 
 
 A 
 
 ^ 
 
 mm. 
 
 
 mm. 
 
 
 738.72 
 
 1.001414 
 
 75346 
 
 1.000988 
 
 2112.53 
 
 1.002765 
 
 4953-92 
 
 1.002952 
 
 4140.82 
 
 1.003253 
 
 8628.54 
 
 1 .004768 
 
 9336.41 
 
 1.006366 
 
 10981.42 
 
 1 .006456 
 
 CARBON DIOXIDE 
 
 HYDROGEN 
 
 A 
 
 
 
 A 
 
 Pv 
 
 mm. 
 
 
 mm. 
 
 
 764.03 
 
 1.007597 
 
 . . . 
 
 . . . 
 
 3186.13 
 
 1.028698 
 
 22II.I8 
 
 0998584 
 
 4879 77 
 
 I 045625 
 
 5845.18 
 
 0.996121 
 
 9619.97 
 
 1.155865 
 
 9176.50 
 
 0.992923 
 
 1 Preston, Theory of Heat, p 402. 
 
EXCEPTIONS TO BOYLE'S LAW 69 
 
 bility was found to be greater than that calculated 
 from Boyle's law. The compressibility of hydro- 
 gen was found to be less than the calculated value. 
 Owing to this peculiar behavior, Regnault distin- 
 guished hydrogen as a gaz plus que parfait. 
 
 During the last thirty years this subject has 
 been very thoroughly investigated by Cailletet, 
 Amagat, and others. The results of these obser- 
 vations will be considered in the next chapter. 
 
CHAPTER III 
 
 SECTION I 
 Conception of Critical Constants 
 
 WHILE the investigations on Boyle's law were 
 being carried, out, Andrews was considering the 
 problem of the liquefaction of gases. With his 
 observations a new era begins. Maxwell 1 says, 
 "The experiments of Andrews on carbonic acid 
 furnish the most complete view hitherto given of 
 the relation between the gaseous and liquid states 
 of matter, and of the mode in which the properties 
 of a gas may be continuously and imperceptibly 
 changed into those of a liquid." The question as 
 to the relative value of temperature and pressure in 
 the liquefaction of gases had long been discussed. 
 The interesting observations of Caignard de la 
 Tour (p. 18), of Drion (p. 63), and of Mendeleeff 
 (p. 64), all point toward the conclusion that for 
 every gas there is a definite temperature above 
 which it cannot be liquefied. These results, how- 
 ever, remained unheeded until Andrews published 
 
 1 Theory of Heat, p. 124. 
 70 
 
CONCEPTION OF CRITICAL CONSTANTS 71 
 
 his exhaustive researches in this field. From his 
 observations it is evident that for every gas there 
 is a definite temperature above which it cannot 
 be condensed to the liquid state. This point 
 Andrews called the Critical Temperature. The 
 Caignard de la Tour point, to which Faraday 
 frequently referred, and the absolute boiling point 
 suggested by Mendeleeff, correspond to this same 
 temperature. 
 
 Previous to the work on critical constants, 
 Andrews 1 studied the effect of high pressures 
 and low temperatures on the volumes of various 
 gases. He experimented at temperatures varying 
 from 75 to --110. Ordinary air was com- 
 pressed to gi^ of its original volume ; in which 
 state the density was but little inferior to that of 
 water. Oxygen was reduced to ^^, and hydrogen 
 to -g-Jfl- of the original volumes. Experiments were 
 also made on carbon monoxide, nitrogen, and nitric 
 oxide. None of the gases showed any indications 
 of liquefaction. 
 
 The results obtained in the preceding experi- 
 ments led Andrews to make a careful study of 
 the conditions under which carbonic acid can be 
 liquefied. The first account of this work was 
 communicated by the author to Dr. Miller, who 
 
 1 Chem. News, 4, p. 158, 1861. 
 
72 LIQUEFACTION OF GASES 
 
 published a short notice of the results in I863. 1 
 The note reads as follows : 
 
 " On partially liquefying carbonic acid by press- 
 ure alone, and gradually raising at the same time 
 the temperature to 88 F., the surface of demarka- 
 tion between the liquid and gas became fainter, 
 lost its curvature, and finally disappeared. The 
 space was then occupied by a homogeneous fluid, 
 which exhibited, when the pressure was suddenly 
 diminished or the temperature slightly lowered, a 
 peculiar appearance of moving or flickering striae 
 throughout the entire mass. At temperatures 
 above 88 no apparent liquefaction of carbonic 
 acid, or separation into two distinct forms of mat- 
 ter, could be effected, even when a pressure of 300 
 or 400 atmospheres was applied. Nitrous oxide 
 gave analogous results." 
 
 A complete account of the first series of obser- 
 vations was published by the author in i869. 2 
 The apparatus employed is shown in figures u, 
 12, and 13. The gas to be liquefied was com- 
 pressed in glass tubes of the form represented 
 in figure u. The upper portion, from a to b, is 
 a capillary tube. From b to c the diameter is 
 about 2.5 mm., and from c to f about 1.25 mm. 
 Before use the tubes were carefully calibrated by 
 
 1 Miller, Chemical Physics, 3d. ed., p. 328. 
 
 2 Bakerian Lecture, Trans. Roy. Soc., 1869, Part 2, p. 575. 
 
CONCEPTION OF CRITICAL CONSTANTS 73 
 
 means of mercury. The pure, dry, carbonic acid 
 was conducted through the tube for several hours, 
 to remove the air. The capillary end was then 
 hermetically sealed, and 
 the other end tempora- 
 rily closed until the tube 
 was placed in an upright 
 position with this end 
 below the surface of 
 mercury, as shown in 
 the third tube (Fig. 11). 
 Heat was then applied 
 to the tube until a small 
 quantity of the gas was 
 expelled through the 
 mercury. On cooling, a 
 short column of mer- 
 cury rose in the tube. 
 The open end, while 
 under the surface of the 
 mercury, was then con- 
 nected with the receiver 
 of an air-pump, and the tube partially exhausted. 
 In this way the mercury could be made to rise any 
 desired height. The tube being previously cali- 
 brated, it was a simple matter then to accurately 
 calculate the volume of the gas at o and 760 mm 
 pressure. 
 
 'c 
 e|n 
 
 FIG. ii. 
 
74 
 
 LIQUEFACTION OF GASES 
 
 The tube containing the gas, with a small quan- 
 tity of mercury, was placed in a strong, cold-drawn 
 copper tube, provided at each end with a massive 
 brass flange. A longitudinal section of this ar- 
 rangement is shown in figure 12. Two brass end- 
 
 ! 
 
 FIG. 12. 
 
 pieces were firmly bolted to the flanges, and the 
 connection made air-tight by means of leather 
 washers. The lower end of the tube was pro- 
 vided with a steel screw 180 mm. in length and 
 4 mm. in diameter. By means of this screw a 
 
CONCEPTION OF CRITICAL CONSTANTS 75 
 
 pressure of 400 atmospheres could be obtained. 
 The portion of the copper tube not occupied by 
 the glass tube was filled with water. When the 
 screw was inserted, the pressure gradually in- 
 creased ; the mercury rose in the glass tube and 
 forced the gas up into the capillary portion which 
 had been calibrated. The complete apparatus is 
 represented in figure 13. One of the tubes contains 
 the carbonic acid, while the other contains air and 
 serves as a comparison tube. A communication 
 is established between the two tubes through a and 
 b, below the mercury, so that the pressures will be 
 equal. The pressures at different intervals were 
 calculated from the relative volumes of the com- 
 pressed air. 
 
 At a temperature of 13.!, the carbonic acid 
 began to liquefy under a pressure of 48.89 atmos- 
 pheres. This point could not be observed directly, 
 inasmuch as the first visible traces of liquid repre- 
 sent a column two or three millimetres in length. 
 It was determined indirectly, however, by observ- 
 ing the volume of the gas at a few tenths of a 
 degree above the point of liquefaction, and then 
 calculating the contraction of the gas in cooling to 
 the point where liquefaction begins. The results 
 obtained at the temperatures 13.! and 21. 5 are 
 given in the following table, in which S represents 
 the ratios of the volumes of air before and after 
 
7 6 
 
 LIQUEFACTION OF GASES 
 
 compression, e represents the corresponding ratios 
 for carbonic acid, t is the temperature of the air, 
 and /' the temperature of the carbonic acid : 
 
 CARBONIC ACID AT 13.! 
 
 CARBONIC ACID AT 21. 5 
 
 5 
 
 / 
 
 e 
 
 /' 
 
 & 
 
 / 
 
 e 
 
 t' 
 
 i 
 47-5 
 
 io. 7 5 
 
 I 
 76.16 
 
 I 3 .i8 
 
 i 
 46.70 
 
 8.63 
 
 I 
 67.26 
 
 21. 46 
 
 4^89" 
 
 10 .86 
 
 i 
 80.90 
 
 13 .09 
 
 60.05 
 
 8 .70 
 
 I 
 114.7 
 
 21 .46 
 
 i 
 49.00 
 
 10 .86 
 
 i 
 105.9 
 
 13 .09 
 
 i 
 60.29 
 
 8 .70 
 
 i 
 I74~8 
 
 21 .46 
 
 49-28 
 
 10 .86 
 
 268.8 
 
 13 .09 
 
 60.55 
 
 8 .70 
 
 240.5 
 
 21 .46 
 
 i 
 50-15 
 
 10 .86 
 
 i 
 462.9 
 
 13 .09 
 
 i 
 61 oo 
 
 8 .70 
 
 i 
 367-7 
 
 21 .46 
 
 i 
 76.61 
 
 10 .86 
 
 I 
 500.7 
 
 13 .09 
 
 i 
 
 62 21 
 
 8 .70 
 
 I 
 440.0 
 
 21 .46 
 
 i 
 90-43 
 
 10 .86 
 
 I 
 5 IO -7 
 
 13 .09 
 
 I 
 62.50 
 
 8 .70 
 
 I 
 443-3 
 
 21 .46 
 
 In both series of results there is an abrupt 
 change of volume at the point where liquefaction 
 begins. The small increase in pressure, which 
 was necessary to continue the liquefaction, was 
 attributed by Andrews to a small quantity of air 
 in the tube containing the carbonic acid. After 
 the condensation to liquid was complete, the press- 
 ure increased rapidly. The high coefficient of 
 expansion of liquid carbonic acid, which had been 
 observed by Thilorier, was confirmed by Andrews. 
 
 The following table gives the results obtained 
 at the temperatures 31. i and 32. 5 : 
 
CONCEPTION OF CRITICAL CONSTANTS 77 
 
 CARBONIC ACID AT 31.! 
 
 CARBONIC ACID AT 32. 5 
 
 I 
 
 t 
 
 
 
 '' 
 
 S 
 
 t 
 
 e 
 
 t' 
 
 i 
 54-79 
 
 110.59 
 
 I 
 80.55 
 
 3I--I7 
 
 i 
 
 12. 10 
 
 i 
 859 
 
 32'. 5 o 
 
 i 
 
 II .55 
 
 I 
 90.04 
 
 3i -19 
 
 i 
 71-52 
 
 12 .15 
 
 i 
 I 4-3 
 
 32 -34 
 
 62.67 
 
 II .44 
 
 I 
 103.1 
 
 3i -19 
 
 73.60 
 
 12 .30 
 
 i 
 156.0 
 
 32 -45 
 
 i 
 67.60 
 
 II .63 
 
 124.4 
 
 31 .15 
 
 i 
 7402 
 
 12 .30 
 
 i 
 159-9 
 
 32 .46 
 
 i 
 73.26 
 
 II .45 
 
 I 
 169.0 
 
 31 .09 
 
 i 
 76.25 
 
 12 .40 
 
 19 .7 
 
 32-38 
 
 i 
 73^83 
 
 13 .00 
 
 I 
 174-4 
 
 31 .08 
 
 i 
 78^52 
 
 12 .50 
 
 ^-g 
 
 32 -48 
 
 i 
 75-40 
 
 II .62 
 
 I 
 
 31 .06 
 
 i 
 79-77 
 
 12 -35 
 
 351-3 
 
 32 -54 
 
 i 
 79.92 
 
 II .16 
 
 I 
 
 31 .10 
 
 i 
 84.90 
 
 12 -35 
 
 i 
 3878 
 
 32 -75 
 
 i 
 85-*9 
 
 II .45 
 
 I 
 
 45 5 
 
 3 1 -5 
 
 
 
 
 
 In these experiments there was no evidence 
 that liquefaction had taken place. There was a 
 sudden decrease in volume at a definite pressure 
 in each series, just as in the preceding observa- 
 tions, but the fluid in this case did not separate 
 into two distinct layers. 
 
 Observations were also made at temperatures of 
 3 5. 5 and 48. i. In each case the abruptness of 
 the change in volume becomes less marked as the 
 temperature increases. The results may be best 
 compared by reference to the curves (Fig. 14). 
 The ordinates represent the pressures in atmos- 
 
LIQUEFACTION OF GASES 
 
 pheres, while the abscissas represent the corre- 
 ponding volumes. 
 
 The curve for 13.! shows that at low pressures 
 
 -105 
 '100 
 95 
 90 
 
 as 
 ao 
 73 
 
 -70 
 -65 
 -60 
 -55 
 -SO 
 
 FIG. 14. 
 
 the carbonic acid approximately obeys the laws of 
 gaseous expansion. At a pressure of about 50 
 atmospheres, the curve becomes almost parallel 
 to the axis of abscissas ; i. c. the volume decreases 
 rapidly while the pressure remains practically con- 
 
CONCEPTION OF CRITICAL CONSTANTS 79 
 
 stant. This represents the change from the gas- 
 eous to the liquid state. When this change is 
 complete, the pressure increases rapidly, and the 
 curve becomes almost parallel to the axis of ordi- 
 nates. The curve for 21. 5 is very similar to that 
 for 1 3. i. The abruptness of the change, how- 
 ever, is not quite so pronounced. The curves for 
 31. i and 32. 5 also show abrupt changes, after 
 which the contraction in volume corresponds 
 approximately to that of liquid carbonic acid. 
 However, liquefaction does not take place at these 
 temperatures. At 48. i the curve shows no abrupt 
 changes. Yet at high pressures the gas deviates 
 considerably from Boyle's law, as can be seen by 
 comparison with the air curves in the same figure. 
 From these results it became evident to An- 
 drews that carbonic acid can be liquefied at or 
 below the temperature of 21. 5, but cannot be 
 liquefied at or above the temperature of 31.!. He 
 then made a series of careful experiments to 
 determine accurately the highest temperature at 
 which carbonic acid can be liquefied. This was 
 found to be 30. 92 C, or 87. 7 F. Below this tem- 
 perature two distinct layers of the fluid, separated 
 by a well-defined meniscus, could be observed, 
 while at higher temperatures no such separation 
 could be produced. This point is the critical 
 temperature of carbonic acid. The pressure re- 
 
8o LIQUEFACTION OF GASES 
 
 quired to liquefy carbonic acid at this temperature 
 is the critical pressure. 
 
 The change of state at the critical temperature 
 is gradual and imperceptible. The author says, 
 " I have frequently exposed carbonic acid to much 
 higher pressures, and have made it pass, without 
 break or interruption, from what is regarded by 
 every one as the gaseous state, to what is, in like 
 manner, universally regarded as the liquid state. 
 Take, for example, a given volume of carbonic 
 acid gas at 50, or at a higher temperature, and 
 expose it to increasing pressure until 150 atmos- 
 pheres have been reached. When the full press- 
 ure has been applied, let the temperature be 
 allowed to fall till the carbonic acid has reached 
 the ordinary temperature of the atmosphere. Dur- 
 ing the whole of this operation no breach of con- 
 tinuity has occurred. It begins with a gas, and 
 by a series of gradual changes, presenting no- 
 where any abrupt alteration of volume or sudden 
 evolution of heat, it ends with a liquid. The 
 closest observation fails to discover anywhere indi- 
 cations of a change of condition in the carbonic 
 acid, or evidence at any period of the process of 
 part of it being in one physical state, and part 
 in another. That the gas has actually changed 
 into a liquid would, indeed, never have been sus- 
 pected, had it not shown itself to be so changed 
 
CONCEPTION OF CRITICAL CONSTANTS 81 
 
 by entering into ebullition on the removal of the 
 pressure." 
 
 Andrews extended his observations to ammo- 
 nia, ether, carbon disulphide, nitrous oxide, and 
 hydrochloric acid. These substances exhibited 
 properties similar to those of carbonic acid. Their 
 critical temperatures were measured, and found, in 
 some cases, to be above 100. 
 
 Andrews also suggested a distinction between 
 gases and vapors which has met with general 
 approval. This distinction is based upon the 
 critical temperature. Many of the properties of 
 vapors, he says, depend on the gas and liquid 
 being present together. This can happen only at 
 temperatures below the critical point. A vapor, 
 therefore, may be defined as a gas below the 
 critical temperature. A vapor can be condensed 
 to a liquid by pressure alone, while a gas cannot. 
 Carbonic acid, then, is a vapor below 31, and 
 a gas at higher temperatures. Below 200 ether 
 is a vapor ; above that temperature it is a gas. 1 
 
 Determination of Critical Constants 
 
 After the introduction of critical constants their 
 measurement became a matter of considerable im- 
 
 1 In 1876, Andrews extended his observations to much higher 
 temperatures, but with similar results. Trans. Roy. Soc., 1876, 
 Part 2, p. 421. 
 G 
 
82 LIQUEFACTION OF GASES 
 
 portance. Most of the experimenters on the lique- 
 faction of gases have determined critical constants 
 by various methods, and have introduced various 
 definitions for the critical state. The simplest 
 method for determining the critical temperature 
 and critical pressure is that employed by Andrews. 
 The substance is placed in a thick-walled glass 
 tube, which is sealed at one end. The highest 
 temperature at which the liquid meniscus can be 
 formed by pressure is taken as the critical tempera- 
 ture, and the corresponding pressure as the critical 
 pressure. 
 
 An ingenious method has been introduced by 
 Cailletet and Colardeau 1 for the determination of 
 the critical constants of substances, such as water, 
 which attack glass at high temperature. The 
 operation is carried out in a steel tube. Numer- 
 ous methods have also been introduced by other 
 experimenters. 2 
 
 Critical Constants of Gaseous Mixtures 
 
 Andrews called attention to the fact that the 
 presence of the so-called permanent gases lowered 
 the critical temperature of such gases as carbonic 
 acid. He was unable to liquefy a mixture of three 
 
 1 Ann. de Chim. et de Phys. [6], 25, p. 519. 
 
 2 See Heilborn for compilation of literature on critical constants. 
 Zeit. phys. Chem n 7, p. 601. 
 
CONCEPTION OF CRITICAL CONSTANTS 83 
 
 parts of carbonic acid with four parts of nitrogen 
 at temperatures above 20. The critical tem- 
 perature of carbonic acid containing one-tenth of 
 its volume of air or nitrogen was found to be con- 
 siderably lower than that of the pure gas. 
 
 Dewar l made a large series of observations on 
 the liquefaction of mixtures of carbonic acid with 
 various substances. He thought that the carbonic 
 acid probably united with the substance to form 
 a definite compound which decomposed when the 
 pressure was released. 
 
 Pawlewski 2 worked on organic compounds, and 
 concluded that, for the same class of compounds, 
 the critical point of the mixture lay between 
 those of the two constituents ; and he thought 
 that the interval was divided proportionally to the 
 quantities of the compounds present. Ansdell, 3 
 on the other hand, maintained that this law does 
 not hold true. 
 
 During recent years a large amount of work has 
 been done in this field. The references to these 
 investigations, however, will be sufficient for the 
 present purpose. See especially Dr. Kuenen, 
 Zeit. phys. CJiem., n, p. 33; and Proc. Phys. Soc., 
 Lond., 13, p. 523, 1895, and 15, p. 235, 1897; 
 
 1 Proc. Roy. Soc., 1880. 
 
 2 Ber. d. d. chem. Ges., 1882, p. 460. 
 8 Proc. Roy. Soc., 34, p. 113, 1882. 
 
84 LIQUEFACTION OF GASES 
 
 Duhem, Jour. Phys. Chem., I, p. 273, 1897. The- 
 oretical, Van der Waals, Zeit. phys. Ckern., 5, 
 P- 133- 
 
 SECTION II 
 
 Continuity of the Gaseous and Liquid States of 
 Matter 
 
 Since the introduction of critical constants by 
 Andrews much discussion has arisen concerning 
 the condition of matter at or near the critical 
 point. Andrews was the first to call attention to 
 this problem. " What is the condition of carbonic 
 acid," he asks, " when it passes at temperatures 
 above 31, from the gaseous state down to the 
 volume of the liquid, without giving evidence at 
 any part of the process of liquefaction having 
 occurred ? Does it continue in the gaseous state, 
 or does it liquefy, or have we to deal with a new 
 condition of matter ? If the experiment were made 
 at 1 00, or at a higher temperature, the probable 
 answer would be that the gas preserves its gaseous 
 condition during the compression. On the other 
 hand, when the experiment is made at tempera- 
 tures a little above 31, the great fall which occurs 
 at one period of the process would lead to the con- 
 jecture that liquefaction had actually taken place, 
 although optical tests carefully applied failed at 
 
CONTINUITY OF STATE 85 
 
 any time to discover the presence of liquid in con- 
 tact with a gas. . . . Under many of the condi- 
 tions which I have described it would be vain to 
 attempt to assign carbonic acid to the liquid rather 
 than the gaseous state. Carbonic acid, at the tem- 
 perature of 3 5. 5, and under a pressure of 108 
 atmospheres, is reduced to -^^ of the volume 
 occupied under a pressure of one atmosphere ; but 
 if any one ask whether it is now in the gaseous or 
 liquid state, the question does not, I believe, admit 
 of a positive reply. The substance in this condi- 
 tion stands nearly midway between the gas and the 
 liquid." 
 
 Numerous observations have been made by later 
 experimenters to show that the liquid state per- 
 sists beyond the critical point, while other investi- 
 gators have endeavored to show that this point is 
 the limit of the liquid state. 
 
 Avenarius 1 made a series of observations on 
 "The Causes which Determine the Critical Point," 
 and concluded that the volume of the saturated 
 vapor at the critical point is not equal to that of 
 the liquid at the same temperature. He experi- 
 mented with ether. 
 
 Ansdell, 2 on the other hand, experimenting with 
 
 J Mem. Acad. Sci., St. Petersburg, 1876-1877; Ref., Ansdell, 
 Proc, Roy. Soc., 30, p. 120. 
 
 2 Proc. Roy. Soc., 30, p. 117, 1880. 
 
86 
 
 LIQUEFACTION OF GASES 
 
 hydrochloric acid, says, " The volumes of the satu. 
 rated vapor and liquid gradually approach each 
 other as the temperature nears the critical point, 
 and would undoubtedly become identical if the 
 experiments could be carried on up to the critical 
 
 - 750. 
 
 - 120. 
 
 10 
 
 20 50 
 
 FIG. 15. 
 
 4-Q 
 
 50 
 
 point." The results of his experiments are shown 
 by the curves in figure 15. The volumes are rep- 
 resented by ordinates, and the temperatures by 
 abscissas. The curve a represents vapor volumes, 
 
CONTINUITY OF STATE 87 
 
 and the curve b liquid volumes. At a temperature 
 slightly above 50, the curves show considerable 
 change in direction. Both approach the critical 
 point x> but the experiments could not be success- 
 fully carried to this temperature. 1 
 
 Ramsay 2 says, "The critical point is that point 
 at which the liquid, owing to expansion, and the 
 gas, owing to compression, acquire the same spe- 
 cific gravity, and consequently mix with each 
 other." 
 
 During this same year, Clark 3 made a series 
 of experiments with sulphurous acid in the two 
 branches of a U-tube. He found that when the 
 temperature was near the critical point, the disap- 
 pearance of the meniscus in one branch did not 
 affect the level of the liquid in the other. From 
 this he concluded that the density of the vapor at 
 the critical point is equal to that of the liquid. 
 
 This theory seems to have been generally ac- 
 cepted. Thiesen 4 says, "A substance is to be 
 considered a liquid, or gas, according as the den- 
 sity is greater or less than the critical density." 
 
 1 Amagat has investigated this subject more thoroughly, and 
 says the density of the vapor rapidly approaches that of the liquid 
 at the critical temperature. Compt. rend., 114, p. 1093. See also 
 Cailletet and Mathias, Jour, de Phys. [2], 5, p. 549, 1886. 
 
 2 Proc. Roy. Soc., 30, p. 323, 1880. 
 
 3 Proc. Phys. Sec., Lond., 4, p. 41, 1880. 
 
 4 Zeit. compr. undfltiss. Case, I, p. 87, 1897. 
 
88 LIQUEFACTION OF GASES 
 
 Quite recently, however, von P. de Heen-Liittich 
 has published an article entitled, " Uber die ange- 
 blichen Anomalieen in der Nahe des kritischen 
 Punktes," 1 in which he states that there are two 
 critical densities, the critical density of the liquid 
 and the critical density of the vapor. The latter, 
 he says, is equal to one-half of the former. A 
 large number of observations were made, and the 
 results obtained seem to contradict some of the 
 work of previous experimenters. 2 
 
 The condition of a substance at or near the 
 critical point, however, has been more thoroughly 
 investigated from a different standpoint. Hannay 
 and Hogarth 3 examined the solvent properties of 
 some fluids for non-volatile solids during the pas- 
 sage of the solvent through the critical point. 
 A precipitation of the solid, on passing through 
 this point, they thought, would be conclusive evi- 
 dence of a change from liquid to gas. If, on the 
 other hand, the solid remained in solution, it would 
 be an evidence of the perfect continuity of the 
 gaseous and liquid states. 
 
 In the first experiment potassium iodide was 
 dissolved in alcohol, and the solution heated in a 
 closed glass tube. There was no indication of the 
 
 1 Zeit. compr. undfliiss. Case, 2, pp. 97, 113, and 134, 1898. 
 
 2 See also M. Gouy, Compt. rend. 115, p. 720, 1892. 
 8 Proc. Roy. Soc., 30, p. 178, 1880. 
 
CONTINUITY OF STATE 89 
 
 separation of a solid, even at a temperature of 
 350, which is nearly 100 above the critical point 
 of alcohol. 
 
 The next experiment consisted in dissolving the 
 crystals of potassium iodide in alcohol vapors. 
 The temperature of the alcohol was never lower 
 than 65 above the critical point. When the press- 
 ure was strongly increased, the potassium iodide 
 dissolved. By releasing the pressure rapidly, fine 
 crystals of the salt separated out as a film on the 
 glass, and in some cases as 'a cloud of fine particles, 
 which floated about in the menstruum. 
 
 Other solutions were then examined. Calcium 
 chloride in alcohol, and sulphur in carbon disul- 
 phide, showed no indications of precipitation at 
 temperatures above the critical points of the 
 solvents. The blue solution of cobaltous chloride 
 in alcohol preserved its color at temperatures far 
 above the critical point of alcohol. The absorption 
 spectrum showed no change on passing through 
 the critical point. The authors accepted these 
 results as evidence of the solubility of solids in 
 gases, and of the perfect continuity of the gaseous 
 and liquid states. Ramsay considers the last of 
 these experiments as an evidence that the liquid 
 state persists beyond the critical point. He says : 
 " Messrs. Hannay and Hogarth found that the 
 absorption spectrum of colored salts remains un- 
 
90 LIQUEFACTION OF GASES 
 
 altered even when the liquid in which they are 
 dissolved loses its meniscus. Surely no clearer 
 proof is needed to show that the solids are not 
 present as gases, but are simply solutions in a 
 liquid medium." l 
 
 Ramsay 2 has determined the condition of a 
 fluorescent solution at temperatures above the 
 critical point of the solvent. Eosine in alcoholic 
 solution failed to show the phenomenon of fluo- 
 rescence after the meniscus had disappeared. 
 The author showed that benzene, at temperatures 
 slightly above the critical point, expands and con- 
 tracts, with regard to pressure, approximately the 
 same as a liquid. The gaseous state, he thinks, 
 depends entirely upon the mean velocity of the 
 molecules and not upon the mean free path. He 
 considered a gas as a substance consisting of simple 
 molecules, and a liquid as a substance consisting 
 of aggregates -of gaseous molecules. Above the 
 critical point, he says, a substance may be wholly 
 liquid or gas, depending upon the pressure. 
 
 Three years later Jamiu 3 announced a similar 
 theory. He saw no reason why the liquid state 
 should cease at the critical point. He says : " I 
 believe that gases are liquefiable at any tem- 
 perature when the pressure is sufficient, but that 
 
 1 Proc. Roy. Soc., 30, p. 327. 2 Ibid., 31, p. 194. 
 Phil. Mag. [5], 1 6, p. 71, 1883. 
 
CONTINUITY OF STATE 91 
 
 an unperceived circumstance has prevented the 
 liquefaction from being seen." He looked upon 
 the disappearance of the meniscus only as an evi- 
 dence that the density of the liquid was equal to 
 that of the vapor, a condition which could be 
 brought about by means of pressure alone. In 
 support of this theory the following experiment 
 of Cailletet 1 is quoted : A mixture of one part of 
 air and five parts of carbonic acid was compressed 
 until a portion of the latter gas was liquefied. The 
 pressure was then increased to 200 atmospheres, 
 when the meniscus disappeared. Jamin explains 
 this phenomenon on the assumption that the den- 
 sity of the gas at that pressure was equal to that of 
 the liquid. 
 
 Cailletet and Colardeau 2 have also endeavored 
 to prove that the liquid state persists at tempera- 
 tures above the critical point. They experimented 
 with a solution of iodine in liquid carbonic acid. 
 The colored solution was heated, in a closed glass 
 tube, to the critical point of the solvent. After 
 the meniscus disappeared the portion of the tube 
 which had contained the liquid remained colored, 
 while the upper portion was colorless. The ab- 
 sorption spectrum of the substance in this condi- 
 tion showed that the iodine existed in solution, and 
 
 1 Compt. rend., 90, p. 210, 1880. 
 
 2 Ann. de Chim. et de Phys. [6], 18, p. 269, 1889. 
 
92 LIQUEFACTION OF GASES 
 
 not as a vapor. The authors concluded then that 
 the liquid state still persisted in the lower portion 
 of the tube. Villard, 1 however, has shown that 
 under certain conditions iodine is soluble in gaseous 
 carbonic acid, and that this does not serve as a test 
 for the liquid state. 
 
 The theory that the critical point is the limit of 
 the liquid state has been ably defended by Hannay. 
 Reference has already been made to his work on 
 the solubility of solids in gases. Later he investi- 
 gated the phenomenon of capillarity 2 for carbonic 
 acid, ammonia, sulphurous acid, nitrous oxide, car- 
 bon disulphide, chlorine, and various alcohols, and 
 found that capillarity disappears at or near the 
 critical point. 3 He looked upon the solvent power 
 of a gas as depending upon two conditions, the 
 molecular closeness and the vis viva. At con- 
 stant density, he says, the solvent power increases 
 with the kinetic energy of the molecules. This 
 statement, however, requires some modification. 
 Altschul 4 has carefully studied the influence of tem- 
 perature on the solvent power of a gas. He filled 
 a strong glass tube with a solution of I gram 
 
 1 Ann. de Chim. et de P/iys., 10, p. 387, 1897. 
 
 2 Proc. Roy. Soc., 30, p. 478, 1880. 
 
 8 Similar observations were also made by Clark. Proc. Phys. Soc. t 
 Lond., 4, p. 41. 
 
 * Zeit. compr. undfliiss. Case, I, p. 207, 1897. 
 
CONTINUITY OF STATE 93 
 
 of potassium iodide in 150 grams of alcohol, and 
 carefully heated the substance to the critical 
 point of the solvent. At this temperature the 
 substance remained in solution, but when gradu- 
 ally heated to a temperature of about 356 a por- 
 tion of the tube became filled with very fine 
 glistening crystals, which could be seen with the 
 naked eye. These crystals did not settle to the 
 bottom, but floated about throughout the whole 
 space of the tube. From this observation, it is 
 evident that the statement of Hannay that, "re- 
 taining the volume the same, the higher the tem- 
 perature the greater the solvent power of K 
 a gas," is true only within certain limits. 
 
 Hannay also made use of an entirely 
 different method for studying the condi- 
 tion of a fluid at the critical point. The 
 apparatus employed is shown in 
 figure 1 6. The liquid was en- 
 closed in the upper part 
 of the tube A by 
 means of the mer- 
 cury B. The lower 
 
 ^>^^ FIG. 16. 
 
 part of the appa- 
 ratus C was filled with nitrogen. By compressing 
 the mercury, and gently tapping the tube, a small 
 bubble of the gas could be made to pass up 
 through the column of mercury into the liquid. 
 
94 LIQUEFACTION OF GASES 
 
 At temperatures below the critical point, the 
 bubble of nitrogen showed a distinct meniscus, 
 and passed up through the column of liquid. At 
 temperatures above the critical point, the bubble 
 diffused immediately throughout the whole space. 
 Hannay looked upon this experiment as decisive 
 evidence that the liquid state ceases at the critical 
 temperature. He says : " There can be no liquid 
 above the critical point, as this is the termination 
 of all properties which distinguish a liquid from a 
 gas." 
 
 We come now to the question originally asked 
 by Andrews. "What is the condition of matter, 
 at temperatures slightly above the critical point?" 
 Does it belong to the liquid, or the gaseous state ? 
 The answer to this question depends entirely 
 upon the definitions of the terms liquid and gas. 
 Hannay l suggests four states of matter, the 
 gas, vapor, liquid, and solid. The vapor, he says, 
 is a distinct state of matter. The fact, however, 
 that some of the properties of vapors differ from 
 those of gases under ordinary conditions, is not 
 necessarily an evidence of a different state of 
 matter, in the sense in which that term is usually 
 employed. Crookes 2 has shown that the proper- 
 ties of gases under extremely low pressures differ 
 
 1 Proc. Roy. Sot., 31, p. 520. 
 
 2 Ibid., 30, p. 469, 1880. 
 
CONTINUITY OF STATE 95 
 
 essentially from those of gases under the ordinary 
 pressure, and he has suggested the term radiant 
 matter for such conditions. Three states of 
 matter, then, apart from the liquid and solid, have 
 already been suggested, namely, the vapor, the 
 gas, and radiant matter. It is evident, from the 
 above suggestions, that the state which shall be 
 assigned to matter under definite conditions de- 
 pends entirely upon the definitions of the various 
 terms. If we limit the states of matter to the 
 solid, liquid, and gas, which is probably advisable, 
 and accept the ordinary definitions, namely, that 
 a gas has neither form nor volume, but tends to 
 expand indefinitely, and that a liquid has a defi- 
 nite volume, but assumes the form of the vessel 
 in which it is contained, then it seems that a fluid 
 above the critical point belongs to the gaseous 
 state. Matter in that condition tends to expand 
 indefinitely without showing any signs of ebulli- 
 tion. If a small bubble of an indifferent gas is 
 allowed to enter the fluid, it shows no meniscus, 
 but diffuses imperceptibly throughout the whole 
 space. 
 
 The gaseous and liquid states, however, gradu- 
 ally approach each other at the critical point. The 
 change from a gas to a liquid at this point seems 
 to be a continuous process, and shows no abrupt 
 evolution of heat. The intimate relation which 
 
96 LIQUEFACTION OF GASES 
 
 exists between the two states has been admirably 
 expressed by Andrews. He says : " The ordinary 
 gaseous and ordinary liquid states of matter may 
 be made to pass into one another by a series of 
 gradations so gentle that the passage shall nowhere 
 present any interruption or breach of continuity. 
 From carbonic acid as a perfect gas to carbonic 
 acid as a perfect liquid, the transition we have 
 seen may be accomplished by a continuous pro- 
 cess, and the gas and liquid are only distant stages 
 of a long series of continuous physical changes." 
 
 SECTION III 
 
 Relation between the Gaseous and Liquid States, as 
 expressed by the Equation of Van der Waals 
 
 According to Boyle's law (p. 7), the product of 
 the pressure and volume of a gas is a constant 
 quantity. This relation is expressed by the 
 
 equation, 
 
 pv = c. 
 
 According to the law of Charles-Gay Lussac (p. 10), 
 the volume v of a gas at any temperature t, and 
 under the constant / , is given by the equation, 
 
 where V Q is the volume of the gas at o, and a is 
 the coefficient of expansion (= 
 
EQUATION OF VAN DER WAALS 97 
 
 If the pressure in the last equation should change 
 from / to any pressure /, the resulting volume #, 
 according to Boyle's law, is, 
 
 hence, pv = p^(\ + at). 
 
 If T represents the absolute temperature, then 
 tT 273. The coefficient of expansion a being 
 equal to ^ J3> we have, 
 
 or, pv 
 
 273 
 
 The expression / ?> represents the product of the 
 pressure and volume at o and 760 mm. pressure ; 
 
 hence, is a constant. Calling this term R, 
 
 we have p v =RT. I 
 
 This is the equation usually employed to express 
 the relation of the pressure and volume to the 
 temperature of a gas. It represents a combina- 
 tion of the law of Boyle with that of Charles -Gay 
 Lussac. 
 
 Reference has already been made to the fact 
 that gases under high pressures deviate from 
 Boyle's law (p. 65). The most elaborate series of 
 
93 LIQUEFACTION OF GASES 
 
 experiments in this direction is that of Amagat. 1 
 He took into account both pressure and tempera- 
 ture, and hence his results apply directly to 
 equation I. 
 
 The results obtained by Amagat are represented 
 by the curves in figures 17, 18, 19, and 20. Fig- 
 
 -4O <SO <30 100 120 I4O /60 I8O 00 ISO 
 FIG. I 7 . 
 
 ure 17 shows the compressibility of hydrogen. 
 The abscissas represent the pressures in atmos- 
 pheres, while the ordinates represent the corre- 
 sponding values of the product pv. If the equation 
 pv = R T should remain true under all conditions, 
 the curves would become straight lines parallel to 
 
 1 Ann. de Chim. et de Phys. [4], 29; [5], 23, p. 358; Compt. 
 rend., 73, p. 183; 75, p. 479. 
 
EQUATION OF VAN DER WAALS 
 
 99 
 
 the axis of abscissas. The curves for hydrogen 
 are approximately straight lines under the condi- 
 tions of the experiments, but the value of pv in- 
 creases, after a certain point has been reached, 
 with increasing pressure. 
 
 The curves for nitrogen are shown in figure 18. 
 In this case the product of pv decreases slightly 
 
 120 140 J6O 1QO SOO 20 -40 
 
 FIG. 1 8. 
 
 ZOO 3OO JKO 
 
 at first, and then increases with increasing press- 
 ure. As the temperature is lowered the curves 
 for both hydrogen and nitrogen show greater 
 divergences from a straight line. 
 
 Figure 19 represents the curves for ethylene. 
 In this case the divergence from a straight line 
 becomes more pronounced. The value of pv de- 
 creases rapidly at first, and then increases. The 
 
100 
 
 LIQUEFACTION OF GASES 
 
 change in the direction of the curve becomes more 
 abrupt as the temperature approaches the critical 
 point. The form of the curves for extremely high 
 
 \ 
 
 \ 
 
 \ 
 
 to 40 OO nr. 
 
 140 .if)!) tna 20 no ttn fen Joo 
 
 FIG. 19. 
 
 temperatures would be similar to those of nitrogen 
 and hydrogen. 
 
 The curves for carbonic acid (Fig. 20) are very 
 
EQUATION OF VAN 
 
 similar to those for ethylene. Near the critical 
 point the change in the direction of the curve 
 becomes very abrupt. The compressibility at 35, 
 
 \ 
 
 \ 
 
 \\ 
 
 \ 
 
 \ 
 
 \ 
 
 ISO }40 '60 IQO 
 
 FIG. 20. 
 
 SAO JOO 320 
 
 and above 80 atmospheres pressure, is approxi- 
 mately the same as for liquid carbonic acid. 
 
 In 1888 Amagat 1 extended his observations on 
 
 1 Compt. rend. t 107, p. 523. 
 
K-V^V"? J LIQUEFACTION OF GASES 
 
 air, nitrogen, oxygen, and hydrogen to pressures 
 as high as 3000 atmospheres, and found that the 
 compressibility of these gases under extremely 
 high pressures is approximately the same as for 
 liquids. The density of oxygen at a pressure of 
 3000 atmospheres was found to be greater than 
 that of water. 
 
 Reference might be made to other experiments 
 of Amagat in this field, as well as to those of 
 other investigators. The preceding examples, how- 
 ever, are sufficient to show that the equation pv 
 = RT, while approximately true under certain 
 conditions of temperature and pressure, cannot 
 be applied to gases under all conditions. 
 
 Numerous attempts have been made to deduce 
 a formula which will express the mutual relation 
 of the pressure, volume, and temperature of gases 
 under all conditions. The most complete equa- 
 tions are those of Van der Waals and Clausius. 
 Both of these equations are based upon the kinetic 
 theory of gases, according to which the molecules 
 of all gases are in a state of motion. The press- 
 ure of a gas, according to this theory, is due to 
 molecular impacts against the walls of the con- 
 taining vessel. 
 
 The kinetic theory of gases involves the assump- 
 tion that gaseous molecules are perfectly elastic. 
 Granting this assumption, and supposing each 
 
EQUATION OF VAN DER WAALS 103 
 
 molecule, between any two successive collisions 
 against other molecules or against the walls of the 
 vessel, to move in a straight line, then the pressure 
 of a gas, in terms of molecular velocity, etc., can 
 be calculated. Some of the calculations in this 
 direction are very elaborate, and involve higher 
 mathematics. The more elementary calculation, 
 which can be found in numerous text-books, is 
 sufficient for the present purpose. 
 
 Suppose the gas to be contained in a cube, the 
 side of which is / units in length. Let n represent 
 the number of molecules, each of which is moving 
 with the velocity #, and m the mass of each mole- 
 cule. The velocity u of a molecule moving in any 
 direction can be resolved into the three compo- 
 nents, x, y, and z t at right angles respectively to 
 three sides of the cube, provided the relation 
 x 1 + y 2 + ^ 2 = ^ 2 is preserved. 
 
 Suppose we consider a single molecule moving 
 with the velocity x perpendicular to one side of 
 the cube. The molecule, being perfectly elastic, 
 rebounds from the wall of the vessel with a ve- 
 locity equal to that with which it approaches. Con- 
 sequently the impulse given to the wall by each 
 collision will be equal to 2 mx, i.e. twice the 
 momentum of the molecule. The collision against 
 
 the two parallel faces of the cube will take place - 
 
i64 LIQUEFACTION OF GASES 
 
 times per unit of time. Hence, the total action of 
 a molecule per unit of time is - . The same 
 
 is true for the other two components, and we have, 
 for the total action of a molecule against all sides 
 of the vessel per unit of time, the value, 
 
 The total impulse, then, of all the molecules per 
 
 o 7/7 //7/ 
 
 unit of time is equal to -- . This value rep- 
 
 resents the total pressure on all sides of the vessel. 
 Dividing this expression by the area of the cube 
 (6/ 2 ), we obtain the pressure/ per unit of area; 
 
 i mgm p = 2 y *** 2 . The volume of the vessel, of 
 
 course, is equal to / 3 . Representing this by v, 
 and simplifying, we have 
 
 or 
 
 pv= ;#. 
 
 According to this expression, the product of the 
 pressure and volume of a gas is equal to two-thirds 
 of the kinetic energy of the molecules. Inasmuch 
 as the temperature of a gas varies directly with 
 
EQUATION OF VAN DER WAALS 105 
 
 the kinetic energy of the molecules, this equation 
 is similar in every respect to the equation, 
 
 The above calculations are based upon the as- 
 sumption that the molecules are free to move 
 throughout the entire space of the containing 
 vessel, and that the mutual attraction between the 
 molecules is inappreciable. It is evident, how- 
 ever, that the molecules are free to move only in 
 the space which is not occupied by the molecules 
 themselves. A correction must be made, there- 
 fore, for the actual volume of the molecules. At 
 low pressures this correction is very small, but at 
 high pressures, where the molecules are crowded 
 together more closely, it becomes appreciable. At 
 high pressures, also, the mutual attraction between 
 the molecules must be taken into account. 
 
 In the equation developed by Van der Waals l 
 the influence due to the actual volumes of the 
 molecules, and to the mutual attraction between 
 the molecules, is taken into consideration. The 
 complete equation is 
 
 in which />, v, R, and T represent the same values 
 
 1 Die Continuit'dt des gasformigen und fliissigen Zustandes. 
 
io6 LIQUEFACTION OF GASES 
 
 as in equation I (p. 97), a is a function of the 
 mutual attraction between the molecules, and b is 
 proportional to the actual space occupied by the 
 molecules. 
 
 The development of this equation represents an 
 elaborate mathematical calculation. For the pres- 
 ent purpose, however, we will consider only the 
 application. Taking the atmospheric pressure as 
 unit pressure, and the volume of one gram of gas 
 at o and a pressure of one atmosphere as unit 
 volume, Van der Waals calculated the values of a 
 and b, from the compressibility of carbonic acid 
 gas, and found them to be 0.00874 and 0.0023 re " 
 spectively. Substituting these values, we have 
 
 w _ 0.0023) =RT. 
 
 Placing p and v each equal to unity, the corre- 
 sponding temperature is o, or 7^=273; and 
 
 hence, 
 
 R = 1.00646 
 
 273 
 
 If a and b remain constant under all conditions, we 
 have as a general equation for carbonic acid 
 
 The values of / and v obtained from this equa- 
 tion for different values of T are represented by 
 
EQUATION OF VAN DER WAALS 
 
 107 
 
 the curves 1 in figure 21. The volumes are rep- 
 resented by abscissas, and the corresponding press- 
 ures by ordinates. The curve AD corresponds to 
 
 FIG. 21. 
 
 a temperature of 13.!. The curve, instead of 
 passing from B to C in a straight line, as was 
 
 1 Ostvvald, Outlines of Gen. Chem., p. 86. 
 
io8 LIQUEFACTION OF GASES 
 
 observed by Andrews (p. 78), is continuous, and 
 passes through the points a and 7. In 1871 
 J. Thomson 1 suggested that the curves given by 
 Andrews could probably be obtained as continuous 
 curves. The curves represented in figure 21 are 
 very similar to those suggested by Thomson. In 
 each of the curves I, 2, and 3 it will be noticed that 
 there are three volumes corresponding to one press- 
 ure. In curve I the three volumes corresponding 
 to a single pressure occur only between the limits 
 of pressure represented by y and a. As the tem- 
 perature increases, the three points on the curves 
 representing these three different volumes gradu- 
 ally approach each other, and finally meet at the 
 point k. 
 
 The point k, where the three volumes become 
 equal to each other, was called by Van der Waals 
 the critical point. For carbonic acid .this point 
 corresponds to a temperature of 32. 5. Above 
 this temperature there is but one volume corre- 
 sponding to a definite pressure. This tempera- 
 ture is about one and a half degrees higher than 
 the critical temperature obtained experimentally 
 by Andrews. The pressure corresponding to the 
 point k represents the critical pressure. In the 
 case of carbonic acid the calculated value is 61 
 
 1 Proc. Roy. Soc., 1871. 
 
EQUATION OF VAN DER WAALS 109 
 
 atmospheres, while the experiments of Andrews 
 show a value of about 70 atmospheres. The 
 volume corresponding to the point k represents 
 the critical volume. 
 
 The same results may be obtained by solving 
 the equation algebraically. If the equation be 
 arranged according to the descending powers of 
 v, we have 
 
 o (y . RT\ . a ab l 
 tr (b + -}v* + -v -- = o. 1 
 
 \PJPP 
 
 This equation is of the third degree with respect 
 to v, and, of course, has three roots. A solution 
 of the equation will show that the three roots are 
 real, or that one is real and the other two imagi- 
 nary. That is, for any pressure there are either 
 three values or one value of v. 
 
 This fact is also shown by the graphic represen- 
 tation in figure 2 1 . These curves show that, within 
 a certain interval of pressure for each temperature 
 below 32. 5, there are three corresponding values 
 
 1 The equation as given by Van der Waals is 
 
 '+*)(' "*)('+ 
 
 In this formula, we have, instead of R, the expression ( I -f a) ( I b} . 
 This value is obtained by placing p and v each equal to unity in 
 the original equation of Van der Waals. Substituting R for the 
 expression (i + ar)(i ), and introducing the absolute tempera- 
 ture T, we obtain the equation given above, 
 
/ 
 
 ( 
 
 \ 
 
 no LIQUEFACTION OF^ GASES 
 
 for the volume; while above 32. 5, each pressure 
 corresponds to a definite volume. At the critical 
 point /, the three volumes are equal to each other. 
 If we assume, then, that at the critical point, 
 the three values of v, or the three roots of the 
 equation 
 
 RT\ 9 , a ab 
 
 \ir + -v -- = o 
 pjpp 
 
 are each equal to <, the value of $ must be such 
 that 
 
 o -\- RT .9 ci JJQ ttb 
 3<#>=- Z r , 34> 2 =-, and< 3 = . 
 P P P 
 
 Simplifying these expressions, and representing 
 the critical temperature by 0, and the critical press- 
 ure by TT, we obtain, 
 
 <j> 3 b = critical volume. 
 
 TT = a = critical pressure. 
 27 b z 
 
 = -- - = critical temperature. 
 27 Rb 
 
 Having once determined the values of a and b 
 for a gas, the critical constants can be calculated 
 from the above formulas. The critical constants 
 of carbonic acid calculated in this manner, and the 
 corresponding values obtained experimentally by 
 Andrews, are as follows : 
 
EQUATION OF VAN DER WAALS in 
 
 
 Calculated 
 
 Experimental 
 
 Critical temperature . . . 
 Critical pressure .... 
 Critical volume 
 
 32-5 
 61 atm. 
 o 0069 
 
 3o-92 
 70 atm. 
 0.0066 
 
 
 
 
 The equation of Van der Waals applies to a 
 homogeneous gas or a homogeneous liquid. It 
 does not, however, apply to a liquid and its vapor 
 in mutual contact. During the last twenty years 
 this equation has been the subject of considerable 
 discussion. The very elaborate series of experi- 
 ments by Young 1 show that the generalizations of 
 Van der Waals can be regarded only as approxi- 
 mations. 
 
 In 1880 Clausius 2 called attention to some devia- 
 tions from the equation of Van der Waals. He 
 regarded the assumption of Van der Waals, that 
 the mutual attraction of the molecules is indepen- 
 dent of the temperature, and is a function only of 
 the volume, as true only for a perfect gas. The 
 assumption that the mutual attraction of the mole- 
 cules is inversely proportional to the square of the 
 volume was regarded by Clausius only as a close 
 approximation under certain conditions. Clausius 
 finally suggested the following equation to express 
 
 1 Proc. Phys. Soc., Lond., p. 233, 1897. 
 
 2 Phil. Mag. [5], 9, p. 393. 
 
112 LIQUEFACTION OF GASES 
 
 the relation of the pressure and volume to the 
 temperature of a gas : 
 
 . _ r> T c 
 
 P ' v-a T(v-0yf 
 
 in which R, c, a, and are constants. In some 
 respects this equation is a closer approximation 
 than that of Van der Waals. In 1882 Sarrau 1 
 investigated this equation very thoroughly with 
 reference to the experimental results obtained by 
 Amagat. He found that, in the case of hydrogen 
 and nitrogen, the equation expresses very closely 
 the actual relation of the pressure and volume to 
 the temperature. With certain other gases, how- 
 ever, some deviations were observed. 
 
 From the experimental results it is evident that 
 the equations of Van der Waals and Clausius, as 
 well as similar equations of other experimenters, 
 can be regarded only as close approximations. 2 
 Inasmuch as all of these equations are based upon 
 the kinetic theory of gases, it may be remarked 
 that within recent years considerable opposition 
 has been offered against this theory. 
 
 1 Compt. rend., 94, pp. 639, 718, and 845. 
 
 2 As further reference to this subject see, among others, Ran- 
 kine, J^rans. Roy. Soc., 1854, p. 336; Joule and Thomson, ibid., 
 1862, p. 579; Recknagel, Pogg. Ann., 145, p. 469, 1872; Him, 
 Theorie Mechanique de la Chaleur, 3d ed., II, p. 21 1 ; Amagat, 
 Ann. de Chim. et de Phys. [5], 28, p. 500, 1883; and Zeit. fur 
 Compr. und fl'iiss. Case, 2, p. 178, 1898; Violi, Phil. Mag., 27, 
 p. 527, 1889; J- Traube, Wied. Ann., 61, 2, pp. 380-400, 1897. 
 
CHAPTER IV 
 
 LIQUEFACTION OF THE SO-CALLED PERMANENT 
 GASES 
 
 SECTION I 
 
 "THE meeting of the French Academy on the 
 twenty-fourth of December, 1877, was a memor- 
 able one. On that day the members were told 
 that Cailletet had succeeded in liquefying both 
 oxygen and carbon monoxide at his works at 
 Chatillon-sur-Seine, and that the former gas had 
 also been liquefied by Raoul Pictet in Geneva." 
 The following letter, 1 addressed by Cailletet to 
 Sainte-Claire Deville, was read by Dumas : 
 
 " I have to tell you first, and without losing a moment, 
 that I have just this day liquefied oxygen and carbon 
 monoxide. 
 
 " I am, perhaps, doing wrong to say liquefied, for at the 
 temperature obtained by the evaporation of sulphurous 
 acid, about 29, and at a pressure of 300 atmospheres, 
 I see no liquid, but a mist so dense that I infer the pres- 
 ence of a vapor very near its point of liquefaction. 
 
 " I write to-day to M. Deleuil for some protoxide of 
 nitrogen, by means of which I shall be able, without doubt, 
 to see oxygen and carbon monoxide flow. 
 
 1 Compt. rend., 85, p. 1217. 
 
U4 LIQUEFACTION OF GASES 
 
 " P. S. I have just made an experiment which sets 
 my mind greatly at ease. I compressed hydrogen to 
 300 atmospheres, and after cooling down to 28, I re- 
 leased it suddenly. There was not a trace of mist in the 
 tube. My gases, carbon monoxide and oxygen, are 
 therefore about to liquefy, as this mist is produced only 
 with vapors which are on the verge of liquefaction. The 
 prediction of M. Berthelot has been completely realized. 
 
 " Louis CAILLETET. 
 "December 2, 1877." 
 
 Deville added also that Cailletet had successfully 
 repeated his experiments on the condensation of 
 oxygen, on Sunday, December sixteenth, in the 
 laboratory of the Normal School. 
 
 The following telegram l was then read : 
 
 " GENEVA, December 22. 
 
 " To-day I liquefied oxygen at a pressure of 320 atmos- 
 pheres, and a temperature of 140, obtained by means 
 of sulphurous and carbonic acids. 
 " Signed, 
 
 " RAOUL PICTET." 
 
 During the session of the Academy a second 
 telegram 2 was received from Pictet. 
 
 "GENEVA, December 24, 4.15 P.M. 
 
 "A second experiment, performed before numerous 
 assistants, has thoroughly confirmed the results which I 
 communicated to M. Dumas last Saturday. 
 
 " PICTET." 
 
 1 Compt. rend., 85, p. 1214. ' 2 Ibid., p. 1220. 
 
EXPERIMENTS OF CAILLETET 115 
 
 The researches of Andrews had already shown 
 that all gases under the proper conditions of tem- 
 perature and pressure would pass into the liquid 
 state. The experimental evidence, however, for 
 the so-called permanent gases, was left for these 
 two simultaneous, but independent experimenters. 
 Roscoe and Schorlemmer, in their account of these 
 observations, say : " It is difficult, on reading the 
 descriptions of these experiments, to know which 
 to admire most, the ingenious and well-adapted 
 arrangement of the apparatus employed by Pictet, 
 or the singular simplicity of that used by Cailletet. 
 The latter gentleman is one of the greatest of 
 French ironmasters, whilst the former is largely 
 engaged as a manufacturer of ice-making machin- 
 ery, and the experience and practical knowledge 
 gained by each in his own profession have materi- 
 ally assisted to bring about one of the most inter- 
 esting results in the annals of scientific discovery." 
 
 EXPERIMENTS OF CAILLETET 
 
 Previous to the experiments with oxygen, Caille- 
 tet succeeded in liquefying a number of gases. 
 He condensed acetylene 1 to a colorless, extremely 
 mobile liquid, and found its vapor-pressure to be 
 48 atmospheres at a temperature of i, and 103 
 
 1 Compt. rend., 85, p. 851. 
 
LIQUEFACTION OF GASES 
 
 atmospheres at 31. He also liquefied nitric oxide 
 and methane. 1 The former gas condensed at a 
 temperature of 11, when subjected to a press- 
 ure of 104 atmospheres. At a temperature of 
 + 8 there was no sign of liquefaction, even 
 
 with a pressure of 270 
 atmospheres. He con- 
 cluded that the critical 
 temperature was be- 
 tween -f 8 and 1 1. 
 The methane was ob- 
 tained only in the form 
 of a mist. The pure 
 gas was subjected to 
 a pressure of 180 at- 
 mospheres, and a tem- 
 perature of 7. By 
 releasing the pressure 
 suddenly, the gas was 
 condensed to a fine 
 mist. These results 
 were communicated to 
 the French Academy 
 by Berthelot, who stated 
 
 at the same meeting that oxygen and similar gases 
 could probably be liquefied by the new method 
 which had been introduced by Cailletet. 
 
 77 
 
 FIG. 22. 
 
 1 Compt. rend,, 85, p. 1016. 
 
EXPERIMENTS OF CAILLETET 117 
 
 The apparatus 1 employed by Cailletet in the 
 liquefaction of oxygen and carbon monoxide was 
 similar to that used in the previous experiments. 
 A longitudinal section of the condensing apparatus 
 is shown in figure 22. The glass tube TT is en- 
 tirely filled with the gas to be compressed. This 
 tube can be removed from the apparatus as shown 
 in figure 23. When the air in the tube has been 
 completely displaced by the gas, the upper end of 
 
 T. 
 
 FIG. 23. 
 
 the tube is hermetically sealed. The lower end is 
 closed with the finger, and the tube introduced 
 into the strong, wrought-iron cylinder B, which is 
 partially filled with mercury. The tube is fastened 
 in the receiver by means of the bronze screw A. 
 The upper portion of the tube is surrounded by a 
 glass cylinder M, containing liquid nitrous oxide, 
 which in turn is surrounded by a safety bell-jar 
 C. n and n' represent gauges for measuring the 
 pressure. 
 
 The method of operation is very simple. Water 
 is forced through the tube // into the cylinder B, 
 
 1 Ann. de Chim. et de Phys. [5], 15, p. 132. 
 
n8 LIQUEFACTION OF GASES 
 
 by means of an hydraulic pump. The mercury 
 which is displaced by the water passes into the 
 lower end of the tube T. As the pressure in- 
 creases, the mercury gradually rises, and com- 
 presses the gas into a very small space near the 
 top of the tube. 
 
 With this apparatus, Cailletet 1 subjected oxygen 
 and carbon monoxide to a pressure of 300 atmos- 
 pheres and a temperature of 29. The tem- 
 perature was reduced in these experiments by 
 means of liquid sulphurous acid. When the heat, 
 due to the compression, had been removed from 
 the gas, and the temperature had become con- 
 stant, the pressure was suddenly released. In both 
 instances a mist was formed, showing that a par- 
 tial condensation had taken place. These observa- 
 tions are really the first experimental evidence of 
 the liquefaction of the so-called permanent gases. 
 The condensation of oxygen by Pictet was accom- 
 plished about three weeks later. Cailletet's work, 
 however, was not known to the latter experimenter 
 at that time. The methods employed by the two 
 investigators were entirely different ; and both 
 series of observations may be considered as pio- 
 neer experiments. 
 
 1 Compt. rend., 85, p. 1213. 
 
EXPERIMENTS OF CAILLETET 119 
 
 Liquefaction of Nitrogen 
 
 After condensing carbon monoxide and oxygen 
 to the liquid state, Cailletet extended his researches 
 to other gases which had not yet been liquefied. 
 Pure dry nitrogen, 1 when subjected to a pressure 
 of 200 atmospheres and a temperature of 1 3, and 
 then suddenly released, formed drops of liquid, of 
 an appreciable volume, which remained in the tube 
 for a period of three seconds. The first experiment 
 was made at a temperature of 29. The author 
 says there can be no doubt that the nitrogen was 
 actually liquefied in these experiments. 
 
 Liquefaction of Air 
 
 After liquefying the gases which compose the 
 atmosphere, Cailletet proceeded to condense ordi- 
 nary air 2 to the liquid state. He says, "Having 
 liquefied nitrogen and oxygen, the liquefaction of 
 air follows as a matter of course ; but I thought it 
 would be interesting to make a direct experiment, 
 which, of course, succeeded perfectly." 
 
 Experiments with Hydrogen 
 
 Cailletet also extended his experiments to hydro- 
 gen. 3 To obtain evidence of liquefaction with this 
 
 1 Compt. rend., 85, p. 1270. 2 Ibid., p. 1271. 3 Ibid., p. 1270. 
 
120 LIQUEFACTION OF GASES 
 
 gas was a matter of considerable difficulty. In 
 the first experiment no peculiarities were observed. 
 The observation was repeated many times. When 
 the gas was subjected to a pressure of 280 atmos- 
 pheres, and then suddenly released, an exceedingly 
 fine and subtle mist was formed throughout the 
 length of the tube. The duration of the phenome- 
 non was extremely short. The experiment was 
 successfully repeated a number of times in the 
 presence of Berthelot, Deville, and Mascart. 
 
 EXPERIMENTS OF PiCTET 1 
 
 Pictet's work on the liquefaction of oxygen fur- 
 nishes one of the most brilliant experiments of 
 modern science. It was the culmination of a long 
 experience in the liquefaction of gases and the 
 production of low temperatures. The many diffi- 
 culties which accompany experiments of this 
 nature were foreseen and provided for. In the 
 introduction to the memoir on these observations, 
 we find that, previous to the experimental part, 
 Pictet made a careful study of the problem. He 
 looked upon cohesion as a property common to the 
 molecules of all forms of matter ; hence, he says, 
 all forms of matter can exist in the gaseous, liquid, 
 or solid state. He recognized clearly that press- 
 
 1 Ann. de Chim. et de Phys. [5], 13, p. 145; Archives des 
 Sciences Physiques et Naturelles, 1878, p. 1 6. 
 
EXPERIMENTS OF PICTET 121 
 
 ure alone would not suffice for the condensation 
 of the so-called permanent gases, but he thought 
 that if the temperature could be sufficiently re- 
 duced, the molecular cohesion of these gases would 
 condense them into liquids or solids. 
 
 The following five conditions were considered 
 by Pictet to be essential in an experiment of this 
 nature : 
 
 1. The gas to be liquefied must be as pure as 
 possible. 
 
 2. The experimenter must be able to obtain 
 considerable pressure, and also be able to measure 
 this pressure accurately. 
 
 3. The method must be such as to enable the 
 operator not only to obtain a very low temperature, 
 but to maintain this temperature indefinitely. 
 
 4. The condensing surface exposed to the low 
 temperature should be as large as possible. 
 
 5. Provision should be made for the sudden 
 expansion of the gas which has been subjected to 
 this low temperature and high pressure. 
 
 These conditions were all fulfilled in the care- 
 fully constructed apparatus of Pictet. It is neces- 
 sary, in these experiments, that the heat be 
 removed from the gas as rapidly as possible, hence 
 metallic tubes were employed'instead of glass. 
 
 The apparatus provides for three distinct opera- 
 tions : 
 
122 
 
 LIQUEFACTION OF GASES 
 
 1. The circulation of 
 sulphurous acid, which 
 produces the first low- 
 ering of temperature. 
 
 2. The circulation of 
 carbonic acid or pro- 
 toxide of nitrogen, which 
 produces a second low- 
 ering of the tempera- 
 ture. 
 
 3. The generation of 
 oxygen in a closed 
 vessel provided with a 
 long, narrow tube which 
 is completely sur- 
 rounded by the carbonic 
 acid. 
 
 Figure 24 represents 
 an elevation of this ar- 
 rangement. U and V 
 are longitudinal sec- 
 tions of two large drums. 
 The former is provided 
 with a large copper tube 
 R, which is about 12 cm. 
 in diameter, and i.iom. 
 in length. The tube 
 and drum are slightly 
 
EXPERIMENTS OF PICTET 123 
 
 inclined, one end being 12 cm. higher than the 
 other. 
 
 The liquid sulphurous acid is introduced into 
 this tube through the pipe s, which enters at the 
 lower end, and on the upper side of the tube. At 
 the upper end of the tube R is a stop-cock r which 
 is connected by means of a long tube, 25 mm. in 
 diameter, with the suction of the first pump P. 
 
 When the pump is operated, and a partial 
 vacuum is produced in the tube R, the sulphur- 
 ous acid will begin to evaporate immediately, and 
 the temperature of the liquid will fall rapidly. 
 The more perfect the vacuum, the lower will the 
 temperature fall. 
 
 In order to make the process continuous for a 
 period of several hours, it was necessary to make 
 use of a second pump, P' (Fig. 25), and arrange 
 the two so that the suction of P' corresponded to 
 the pressure of P. Experiment showed that, by 
 means of this arrangement, the temperature could 
 be reduced almost 20 farther than with one pump 
 alone. The pumps were made of cast iron. The 
 pistons were hollow, and provided with a circula- 
 tion of water. The valves were carefully con- 
 structed of steel. The speed of the pumps varied 
 from 80 to 100 strokes per minute. The pistons 
 of the two pumps were connected by means of a 
 metallic pipe. P' was connected with the copper 
 
124 LIQUEFACTION OF GASES 
 
EXPERIMENTS OF PICTET 125 
 
 condenser C, the tubes of which were traversed by 
 a current of water. The condenser C is also con- 
 nected with the tube z. 
 
 The action is as follows : The sulphurous acid 
 which is evaporated in the tube R passes through 
 the cock r into the first pump P, thence into the 
 second pump P', and finally into the condenser C, 
 where it is compressed to from i to 2 atmospheres 
 and condensed to the liquid state. From the con- 
 denser C, the sulphurous acid passes through the 
 pipe s, and again enters the tube R. The quantity 
 of acid which passes from c to R can be regulated 
 by means of the screw-valve q, so as to exactly 
 replace the vapor which is exhausted. The quan- 
 tity of liquid then in the tube R will remain con- 
 stant. 
 
 With this arrangement, the cycle of sulphurous 
 acid is complete ; and the fall of temperature 
 produced in the large tube R can be perma- 
 nently maintained. 
 
 The first circulation of sulphurous acid is only 
 an expedient to obtain a sufficient quantity of car- 
 bonic acid or protoxide of nitrogen in the liquid 
 state. To accomplish this a tube 5, 6 cm. in 
 diameter and 1.15 m. in length, is placed in the 
 tube R. The tube S, of course, is completely 
 immersed in the liquid sulphurous acid, and has 
 a temperature of 65. At this temperature, it 
 
126 LIQUEFACTION OF GASES 
 
 requires a pressure of only 4 to 6 atmospheres to 
 condense the carbonic acid. 
 
 The carbonic acid was prepared from Carrara 
 marble and hydrochloric acid ; it was then care- 
 fully washed and dried, and finally stored in the 
 oil gasometer G (Figs. 24 and 25). This vessel 
 was connected, by means of the tube c, with the 
 three-way cock K (Fig. 25). The gaseous car- 
 bonic acid was pumped from the vessel G, by 
 means of the pumps O and O 1 , which are similar 
 in every respect to P and P f , and compressed in 
 the tube S, where it immediately condensed to the 
 liquid state. 
 
 The next step was to utilize the liquid carbonic 
 acid in lowering the temperature of oxygen gas. 
 This was accomplished in the drum V, which is 
 similar to U. The two concentric copper tubes, 
 D and A, were placed in this drum, which was 
 slightly inclined, but in the opposite direction from 
 that of U. The tube D was 3.70 m. in length and 
 35 mm. in external diameter, while A was 4.16 m. 
 in length and 15 mm. in external diameter. The 
 arrangement is shown in the figures. The tube 
 D is connected with the tube S, which contains 
 the liquid carbonic acid, by means of the pipe t. 
 The flow of acid from 5 to D is regulated by 
 means of the screw-valve/. The pipe c" connects 
 the upper part of D with the three-way cock K. 
 
EXPERIMENTS OF PICTET 127 
 
 By turning the cock K, the suction of the pump 
 O is brought into connection with the tube D by 
 means of the pipes c 1 and c". The pump O' is 
 connected with the tube 5 by means of the pipe s. 
 
 With this arrangement the cycle of carbonic 
 acid is complete, just as in the case of sulphurous 
 acid. The directions in which the acids flow at 
 different parts of the apparatus are shown by the 
 arrows. By means of this double circulation a 
 temperature of from 120 to 140 could be 
 obtained. 
 
 The tube A is bent downward at one end and 
 connected with the large wrought-iron flask B. 
 This vessel, with a capacity of 1659 cc., was 
 forged with the utmost care, so as to insure 
 homogeneity throughout the metal. The walls 
 of the flask were 35 mm. in thickness. The ves- 
 sel B contains a definite quantity of a perfectly 
 dry mixture of potassium chlorate and potassium 
 chloride. On heating the flask, by means of the 
 gas burner b, oxygen gas is generated. In this 
 way a pressure of several hundred atmospheres 
 can easily be obtained. The oxygen, of course, 
 is compressed in the tube A as well as in the 
 flask B. 
 
 After the apparatus had once been set up, fifteen 
 days were spent in preliminary experiments, in 
 order to test separately the different parts of the 
 
128 LIQUEFACTION OF GASES 
 
 apparatus, and to determine the conditions under 
 which the best results could be obtained. A 
 thousand precautions, he says, are necessary to 
 insure the success of the experiment. After all 
 the conditions and details had been carefully 
 studied, Pictet began on the morning of Decem- 
 ber twenty-second to make a complete experiment. 
 The complete working of the apparatus can be 
 understood from the following details which were 
 given concerning the first experiment : 
 
 9.00 A.M. The sulphurous acid pumps are 
 started. The temperature in the tube R falls 
 rapidly. 
 
 9.30. The temperature is 55. The car- 
 bonic acid pumps are started. The gasometer 
 descends. The pressure of the carbonic acid is 
 6 atmospheres. During the working of the pumps 
 the pressure slowly increases to 8 atmospheres. 
 
 9.50. The temperature is 49, the pressure 
 8.5 atmospheres. 
 
 I stop the admission of carbonic acid to the 
 pumps. 
 
 10.20. Temperature 65, pressure 3.9 at- 
 mospheres. 
 
 A slight quantity of gas is again admitted. 
 
 10.40. Temperature 60, pressure 5 atmos- 
 pheres. 
 
EXPERIMENTS OF PICTET 129 
 
 800 litres of carbonic acid are now liquefied. 
 Hoar-frost covers the lower part of the manom- 
 eter m' . 
 
 10.50. The wrought-iron flask is screwed on 
 to the tube A. It is charged with a mixture of 
 700 grams of potassium chlorate and 250 grams 
 of potassium chloride powdered together in a 
 mortar, sifted and carefully dried. 
 
 i i.oo. The gas under the flask is lighted. 
 
 Carbonic acid is admitted more freely into the 
 pumps. The pressure increases to 10 atmos- 
 pheres ; the temperature is 48. The appear- 
 ance of frost on the pipe c" indicates that the 
 carbonic acid has passed over into the long 
 tube D. 
 
 11.15. The pipe c" is connected with the suc- 
 tion of the pumps. The temperature of the car- 
 bonic acid reaches a minimum, 130. 
 
 11.35. The manometer m 1 on the oxygen tube 
 records a pressure of 5 atmospheres. 
 
 The circulation of sulphurous and carbonic acids 
 is completely established. 
 
 12. 10 P.M. The oxygen manometer records a 
 pressure of 50 atmospheres. 
 
 12. 16. The pressure increases to 60 atmos- 
 pheres, and then gradually rises as follows : 
 
130 LIQUEFACTION OF GASES 
 
 12.23 pressure, 70 atmospheres. 
 
 12.29 " 8o 
 
 12.34 " 9 " 
 
 12.36 " 100 
 
 12.37 " 150 " 
 
 12.37.25" " 200 " 
 
 12.38 " 460 
 
 12.39 " 5 10 " 
 12.39.30" " 522 
 
 12.40 " 525 
 12.42 " 526 
 12.44 " 525 
 12.48 " 505 
 12.50 " 495 
 
 i. oo " 471 
 
 1.05. The pressure is 471 atmospheres. 
 
 The pressure is stationary ; hence all the chem- 
 ical and physical phenomena have terminated. 
 
 The condensation has produced the fall of press- 
 ure recorded by the manometer. The tube A is 
 filled with liquid oxygen. 
 
 There is evidently an excess of gas, which causes 
 a higher pressure than that corresponding to the 
 temperature of the liquid carbonic acid. 
 
 1. 10. The pressure is exactly 470 atmospheres. 
 The plug which closes the tube A is opened. 
 A liquid jet issues with great violence, and 
 
EXPERIMENTS OF PICTET 131 
 
 assumes the appearance of a brilliant white pencil. 
 A bluish halo surrounds the jet, especially the 
 lower part. The jet of liquid is from 10 to 12 
 cm. in length, and about 1.5 to 2 cm. in diameter. 
 It continues for a period of 3 or 4 seconds. 
 
 The regulating cock is closed. The pressure is 
 396 atmospheres, but falls rapidly to 352 atmos- 
 pheres, where it remains stationary for about 3 
 minutes. 
 
 1. 18. The tube is again opened; a second 
 liquid jet, similar to first, issues. But following 
 this, the gas escapes with a characteristic aeriform 
 appearance. 
 
 The gas, in expanding, produces a mist by par- 
 tial condensation. The appearance, however, is 
 quite different from that of the first jet. This 
 indicates an absence of liquid in the tube. 
 
 1.19. The pressure is 50 atmospheres. The 
 gas escapes as a bluish mist, but there is no evi- 
 dence of liquid being carried with it. 
 
 Live coals, when placed under the second jet, 
 blaze up instantly with great violence. 
 
 This experiment demonstrated clearly that oxy- 
 gen can be condensed to the liquid state. Pictet 
 calls special attention to the difference between 
 the mist produced by the expansion of the gas 
 alone, and that produced when there was liquid in 
 the tube. The change, he says, from one to the 
 
132 LIQUEFACTION OF GASES 
 
 other was so apparent that it was noticed by more 
 than twenty spectators at the same moment. 
 
 Five experiments were made, three with car- 
 bonic acid, and two with protoxide of nitrogen. 
 The same quantity of potassium chlorate and 
 potassium chloride was used in each of these 
 experiments. 
 
 The more important observations refer to five 
 successive stages, which the author designates as 
 follows : 
 
 1 . The maximum stationary pressure of the oxy- 
 gen before the first jet is allowed to escape. 
 
 This pressure is always stationary for at least 
 a quarter of an hour, and is always less than the 
 pressure recorded at the end of the chemical 
 reaction. 
 
 2. The pressure immediately after the escape 
 of the first jet, when it is distinctly observed that 
 the liquid has been replaced by a gaseous jet. 
 
 3. The stationary pressure after the first jet. 
 The fall of pressure in this operation is due to 
 
 the condensation of oxygen. 
 
 When the tube is filled a second time, condensa- 
 tion ceases, and the pressure becomes stationary. 
 
 4. The pressure immediately after the second jet. 
 
 5. The stationary pressure after the second jet. 
 The third jet was never complete ; it was always 
 
EXPERIMENTS OF PICTET 
 
 133 
 
 shorter than the first two. This indicated that the 
 condensation was not sufficient to fill the tube 
 three times. 
 
 These pressures for five different experiments 
 on the liquefaction of oxygen are shown in the 
 following table. The pressures are given in atmos- 
 pheres : 
 
 
 NUMBER OF EXPERIMENT 
 
 i 
 
 2 
 
 3 
 
 4 
 
 5 
 
 I. Maximum stationary pressure 
 before the first jet . 
 2. Pressure immediately after the 
 first jet ... 
 
 470 
 367 
 3 08 
 285 
 274 
 
 
 
 471 
 
 395 
 339 
 290 
 
 271 
 245 
 253 
 
 
 
 471 
 43 2 
 378 
 2 9 I 
 272 
 
 o 
 
 469 
 4OO 
 346 
 285 
 2 5 I 
 215 
 
 218 
 
 
 
 469 
 4l6 
 361 
 296 
 
 253 
 20 5 
 212 
 
 
 
 3. Stationary pressure after the 
 first jet 
 
 4. Pressure immediately after the 
 second iet 
 
 5. Stationary pressure after the 
 second jet 
 
 6. Pressure immediately after the 
 third jet 
 
 7. Stationary pressure after the 
 third jet .... 
 
 8. Pressure after the fourth jet. 
 In this case the jet was en- 
 tirely "aseous 
 
 
 Pictet endeavored to determine the density of 
 liquid oxygen. The method, of course, was 
 
134 LIQUEFACTION OF GASES 
 
 indirect, and involved numerous calculations. In 
 the original memoir, several pages are devoted to 
 this part of the experiment. He concluded that 
 the density of liquid oxygen was very nearly 
 equal to that of water. Later researches have 
 shown the density to be slightly greater than 
 that of water, but considering the data upon 
 which Pictet based his calculations, the result is 
 remarkably accurate. 
 
 Measurements on the vapor-pressure of liquid 
 oxygen showed it to be about 270 atmospheres at 
 the temperature of the liquid carbonic acid, and 
 about 250 atmospheres at the temperature obtained 
 by means of the protoxide of nitrogen. 
 
 Solidification of Oxygen 
 
 In the third and fourth experiments the escap- 
 ing jet was carefully examined by means of an 
 electric light. The jet appeared to consist of two 
 parts ; a central portion of 2 to 3 mm. in diameter, 
 and an outer envelope. It resembled two con- 
 centric cylinders, the inner of which was partially 
 transparent, while the outer appeared as snow- 
 white dust. The light reflected, by the dust, at 
 right angles to the incident ray was examined 
 by means of a polariscope, and found to be parti- 
 ally polarized. This examination was made by 
 
EXPERIMENTS OF PICTET 135 
 
 M. H. Dufour, Professor of Physics in the Acad- 
 emy of Lausanne. Although the experiment was 
 not conclusive evidence, it was thought by those 
 present that the outer portion of the jet consisted 
 of solid particles of oxygen. 
 
 Experiments with Hydrogen 
 
 " Having obtained the preceding results with 
 oxygen," Pictet says, "we were naturally induced 
 to treat hydrogen in a similar manner. The entire 
 mechanical apparatus used for the first gas can be 
 used for the second without alteration." 
 
 In this experiment the hydrogen was generated 
 from 1261 grams of potassium formate and 500 
 grams of caustic potash. The two were mixed 
 together, and carefully dried. 
 
 The temperature was lowered by means of 
 sulphurous acid and protoxide of nitrogen. The 
 method of procedure was exactly the same as in 
 the case of oxygen. 
 
 The following are some of the notes recorded in 
 the first experiment, which was begun at 7 P.M. 
 Jan. 10, 1878 : 
 
 The pumps were started at 7 o'clock, and 
 when the circulation of the two acids was com- 
 plete, the gas under the hydrogen generator was 
 lighted. 
 
136 LIQUEFACTION OF GASES 
 
 8.32 P.M. The pressure of hydrogen was 50 atmospheres. 
 
 8.47 " " " 100 " 
 
 9.00 " " " 200 " 
 
 9.04 " " " 300 
 
 9.06. i $" " " " 400 
 
 9.07 ' " " 500 
 
 9.08 550 
 9.10.30" " " " 640 " 
 9.11 " 650 
 9.11.30" " " 652 " 
 
 At this point the pressure became almost sta- 
 tionary, and the tube containing the hydrogen was 
 opened. An opaque jet of a bluish tint issued 
 from the orifice. The jet became intermittent, 
 and Pictet concluded that the hydrogen had solidi- 
 fied at the opening in the tube. 
 
 From the observations recorded in this experi- 
 ment, there can be but little doubt that, at the 
 moment when the pressure was relieved, the 
 hydrogen was partially liquefied and probably 
 solidified. 
 
 Cailletet and Pictet removed the last doubt as to 
 the possibility of liquefying such gases as oxygen, 
 nitrogen, and hydrogen. The idea of permanent 
 gases is no longer tenable. Since these experi- 
 ments were made, the apparatus for the liquefac- 
 tion of gases has been materially modified and 
 improved. The names of these two investigators, 
 while connected especially with the pioneer work 
 
EXPS. OF WROBLEWSKI AND OLSZEWSKI 137 
 
 in the liquefaction of the permanent gases, are 
 associated also with the names of the more recent 
 experimenters. 
 
 LIQUEFACTION OF OZONE BY HAUTEFEUILLE AND 
 CHAPPUIS 
 
 In 1882 Hautefeuille and Chappuis condensed 
 ozone to the liquid state. 1 The gas was com- 
 pressed in the Cailletet apparatus to a pressure of 
 125 atmospheres. The temperature was lowered 
 by means of a jet of liquid ethylene to about 
 100. Under these conditions the gas con- 
 densed to an indigo-blue liquid. The liquid re- 
 mained for a short time in a static condition at 
 the ordinary atmospheric pressure. The authors 
 state that the ozone probably contained some oxy- 
 gen. This gas has been liquefied by later expe- 
 rimenters, who have also determined its critical 
 constants and boiling point. 
 
 SECTION II 
 EXPERIMENTS OF WROBLEWSKI AND OLSZEWSKI 
 
 In 1883 the names of two brilliant scientists, 
 Wroblewski and Olszewski, were added to the list 
 
 1 Compt. rend., 94, p. 1249. 
 
138 LIQUEFACTION OF GASES 
 
 of experimenters on the liquefaction of gases. 
 These two observers, co-workers at times, and 
 independent investigators at other times, have 
 done much to bring about the present state of 
 perfection in the methods employed in the con- 
 densation of gases. 
 
 Their first work in this direction consisted in 
 the liquefaction of oxygen, nitrogen, and carbon 
 monoxide. 1 Acting upon the suggestion of Caille- 
 tet they employed liquid ethylene as a refrigerant. 
 
 The apparatus employed is represented in fig- 
 ures 26 and 27. The gas was compressed by 
 means of the contrivance 2 shown in figure 26. 
 The gas is introduced into the large glass tube i 
 which rests in the hollow iron cylinder a b. The 
 upper end of the cylinder is closed air-tight by 
 means of the brass contrivance d % which is held in 
 its position by means of the brass screw c. A 
 small steel tube extends from the upper end of 
 the tube i throughout the brass portion d. The 
 side tube / serves to connect the interior of the 
 apparatus with a manometer and a Cailletet pump. 
 The method of operation is similar to that of 
 Cailletet. The large tube i and the smaller tube 
 
 1 Wied. Ann., 20, p. 243, 1883. 
 
 2 This portion of the apparatus was constructed by Wroblewski 
 in 1882, for the purpose of studying surface tension. Compt. 
 rend., 95, pp. 284 and 342. 
 
EXPS. OF WROBLEWSKI AND OLSZEWSKI 139 
 
 FIG. 26. 
 
140 
 
 LIQUEFACTION OF GASES 
 
 above are filled with the gas to be compressed. 
 Mercury is then introduced through / until the de- 
 sired pressure is obtained. The resisting strength 
 of the apparatus was estimated to be about 500 
 atmospheres. 
 
 FIG. 27. 
 
EXPS. OF WROBLEWSKI AND OLSZEWSKI 141 
 
 The liquefying apparatus is represented in fig- 
 ure 27. The compression apparatus is shown in 
 the lower right-hand corner. In this case, how- 
 ever, the steel tube d is replaced by the thick- 
 walled capillary glass tube q which is fused into 
 the upper end of the tube i. The tube i, together 
 with the small tube q, has a capacity of about 200 
 cubic centimetres. These tubes are filled with the 
 gas to be liquefied by means of a Jolly mercury 
 pump. The tube q passes air-tight into the glass 
 vessel s which, in turn, is closed air-tight by means 
 of a rubber stopper, t is a hydrogen thermometer 
 constructed on the principle of the Jolly air ther- 
 mometer. The copper tube w, which passes 
 through the T^tube //, leads to the liquid ethylene 
 in the receiver x of a Natterer compression pump. 
 This receiver rests in the large zinc vessel s which 
 contains a freezing mixture. The spiral portion 
 of the tube w is surrounded by solid carbonic acid 
 and ether in the vessel b. The side tube of u is 
 connected with the lead tube i>, which leads to a 
 Bianchi exhaust pump. The liquid ethylene then 
 can be evaporated under reduced pressure. 
 
 Liquefaction of Oxygen 
 
 The oxygen gas was prepared from pure potas- 
 sium perchlorate, washed with caustic potash, and 
 dried by means of concentrated sulphuric acid. 
 
142 LIQUEFACTION OF GASES 
 
 The tubes i and q were then filled with the gas. 
 The tube q was cooled down to a temperature of 
 130. The pressure was then increased to more 
 than twenty atmospheres, when the gas was com- 
 pletely liquefied. The liquid oxygen collected in 
 the lower end of tube q, and showed a well-defined 
 meniscus. Measurements were made of the vapor- 
 pressure at different temperatures. 
 
 Liquefaction of Nitrogen and Carbon Monoxide 
 
 After the successful experiment on the liquefac- 
 tion of oxygen, the investigation was extended to 
 nitrogen and carbon monoxide. When subjected 
 to a temperature of 136, and a pressure of 150 
 atmospheres, both of these substances remained in 
 the gaseous condition. When the pressure was 
 suddenly released, however, there was evidence of 
 liquefaction in the tube. By relieving the press- 
 ure more slowly and keeping it always above 50 
 atmospheres, the nitrogen and carbon monoxide 
 were both obtained as liquids in a static condition. 
 The liquids were transparent and colorless, and 
 showed, in each case, a well-defined meniscus. 
 
 These observations were the beginning of a long 
 series of experiments by these investigators. Their 
 later researches were carried on independently, 
 and will be considered presently. 
 
EXPERIMENTS OF WROBLEWSKI 143 
 
 DEMONSTRATION OF THE LIQUEFACTION OF 
 OXYGEN BY DEWAR 
 
 During the next year after the liquefaction of 
 oxygen and nitrogen by Wroblewski and Ols- 
 zewski, James Dewar of London described an 
 apparatus 1 for demonstrating the liquefaction of 
 oxygen in the lecture room. Since that time he 
 has been a constant worker in the field of the 
 liquefaction of gases, and is the author of many 
 important contributions. 
 
 From this point it is probably advisable to con- 
 sider the work of Wroblewski, Olszewski, and 
 Dewar separately. The historical development of 
 the methods can be gathered from the dates of the 
 experiments. 
 
 EXPERIMENTS OF WROBLEWSKI 
 
 Soon after the experiments of Wroblewski and 
 Olszewski on the liquefaction of oxygen, air, nitro- 
 gen, etc., in 1883, the former experimenter made 
 some important observations in the same field. 
 Early the next year he subjected hydrogen 2 to a 
 pressure of 100 atmospheres in a small glass tube. 
 The temperature of the gas was lowered by the 
 evaporation of liquid oxygen which surrounded the 
 
 1 Proc. Roy. Inst., 1 1, p. 148; Phil. Mag., 18, p. 210, 1884. 
 
 2 Compt. rend., 98, p. 149, 1884. 
 
144 
 
 LIQUEFACTION OF GASES 
 
 liquefying tube. On suddenly releasing the press- 
 ure, signs of ebullition were observed in the tube, 
 showing that the hydrogen had been partially 
 liquefied. 
 
 In April of the same year Wroblewski published 
 an article l on the boiling points of oxygen, air, 
 nitrogen, and carbon monoxide. The boiling points 
 of oxygen under different pressures were as 
 follows : 
 
 Pressure 
 
 Boiling Point 
 
 50 atmospheres 
 27.02 " 
 
 24.40 
 
 22.20 " 
 
 II3 
 
 - 129.6 
 
 - 1334 
 
 - 135-8 
 
 - 184. 
 
 The last temperature represents the boiling 
 point of oxygen under the ordinary atmospheric 
 pressure. The following values were obtained for 
 the boiling points of air, carbon monoxide, and 
 nitrogen under a pressure of one atmosphere : 
 
 Substance 
 
 Boiling Point 
 
 Carbon monoxide 
 Air 
 Nitrogen 
 
 - 1 86 
 - 192.2 
 - 194-3 
 
 1 Compt. rend., 98, p. 982, 1884. 
 
EXPERIMENTS OF WROBLEWSKI 145 
 
 By evaporating liquid air and nitrogen under 
 reduced pressure, a temperature of 200 was 
 obtained. The author stated in this article that 
 atmospheric air will be the refrigerant of the 
 future. 
 
 A few months later the same experimenter 1 
 liquefied methane and studied its properties. The 
 critical temperature he found to be 73. 5, and 
 the critical pressure 56.8 atmospheres. The boil- 
 ing point, he says, is between 155 and 160. 
 By using liquid marsh-gas as a refrigerant, 2 the 
 author states that oxygen, air, carbon monoxide, 
 and nitrogen, may be liquefied at comparatively 
 low pressures. 
 
 In 1885 Wroblewski published a detailed ac- 
 count of the apparatus 3 employed by him in the 
 liquefaction of oxygen, carbon monoxide, air, nitro- 
 gen, etc. The plan of the apparatus is shown in 
 figure 28. 
 
 The two iron cylinders, a and b, previously 
 tested to 150 atmospheres pressure, are provided 
 at the ends with the screw-cocks c, d, e, and /. 
 The two vessels are connected with each other 
 and with the manometer h by means of the cop- 
 
 1 Compt. rend., 99, p. 136, 1884. 
 
 2 The use of liquid methane as a refrigerant was first suggested 
 by Dewar, Nature, 28, p. 551, 1883. 
 
 3 Wien. Ber., 91, p. 672, 1885; Wied. Ann. 25, p. 371, 1885. 
 
 L 
 
146 
 
 LIQUEFACTION OF GASES 
 
 per tube gg. This connection, however, can be 
 interrupted at any time by means of the screw- 
 cocks at the ends of the vessels. The cylinders 
 are fastened by means of the metal strip i in the 
 
 FIG. 28. 
 
 zinc chamber k, which can be filled, if desirable, 
 with a freezing mixture. The copper tube /, about 
 three metres in length, is connected with a Natterer 
 pump. The screw-cock e is connected by means 
 of the tube m with the steel contrivance ;/, which 
 
EXPERIMENTS OF WROBLEWSKI 147 
 
 in turn is connected with the liquefying apparatus 
 and the air manometer o. This contrivance is held 
 firmly in its position by means of the iron frame q. 
 The pressure in the vessel a is increased to the 
 liquefying pressure, not less than 40 atmospheres, 
 while in b it ranges from 100 to 120 atmospheres. 
 Each of the vessels, a and b, contain small cylin- 
 ders of calcium chloride and caustic potash, to 
 remove any moisture and carbon dioxide from the 
 gas. 
 
 The gas is liquefied in the thick-walled glass 
 tube r, 42 to 46 cm. in length, which is capable of 
 resisting a pressure of 60 atmospheres. The lower 
 end of the tube is sealed, while the upper end is 
 cemented into the brass flange s, and is closed air- 
 tight by means of the contrivance /. The opening 
 v is connected by means of the copper tube w with 
 n, which in turn is connected with the compression 
 cylinders. The opening x serves for the introduc- 
 tion of a thermometer which, in this case, was a 
 thermo-electric pair connected with a sensitive re- 
 flecting galvanometer. The tube x is placed in a 
 larger tube m f , which is connected with the cylin- 
 .der d' containing liquid ethylene. The tube m f , 
 in turn, rests in the larger tube y t and communi- 
 cates with it at the small opening n'. The vessel 
 r' contains solid carbonic acid and ether, through 
 which the liquid ethylene must pass before it 
 
148 LIQUEFACTION OF GASES 
 
 reaches the liquefying tube. The ethylene vapors 
 pass out of the apparatus through the tube f, 
 which is connected with an air-pump to reduce the 
 pressure. At a pressure of about 10 mm. of mer- 
 cury the temperature sank to 152. When the 
 temperature is thus lowered, the pressure in the 
 liquefying tube is increased to about 40 atmos 
 pheres, when the gas condenses to the liquid state 
 Considerable care is required to get the apparatus 
 filled with perfectly dry gas. 
 
 In this way Wroblewski liquefied oxygen in 
 considerable quantity, and kept it in a static con- 
 dition, at the ordinary atmospheric pressure, for a 
 period of a quarter of an hour. He also liquefied 
 nitrogen, air, and carbon monoxide, and studied 
 their properties. The following values were ob- 
 tained by him for the critical constants of these 
 gases : 
 
 Substance 
 
 Critical 
 Temperature 
 
 Critical Pressure 
 
 Oxygen 
 Carbon monoxide . . 
 Nitrosfen 
 
 -118 
 140.2 
 
 I AC C 
 
 50 atmospheres 
 39 
 
 1C " 
 
 
 1 T-->O 
 
 oy 
 
 The author also made a careful study of the 
 measurement of low temperatures. For this pur- 
 pose he made use of a copper-German-silver 
 
EXPERIMENTS OF WROBLEWSKI 149 
 
 thermo-electric pair and a sensitive reflecting gal- 
 vanometer. Previous to the measurements, the 
 apparatus was compared with a hydrogen ther- 
 mometer. By the evaporation of oxygen at dif- 
 ferent pressures, the following temperatures were 
 obtained : 
 
 Pressure 
 
 Temperature 
 
 74 cm. of mercury 
 
 - i8i. 5 
 
 1 6 
 
 - 190.0 
 
 10 " " 
 
 - 190.5 
 
 8 " " 
 
 - 191.98 
 
 6 " " 
 
 - 194-4 
 
 4 
 
 - 197-7 
 
 2 u " 
 
 200.4 
 
 The oxygen in these experiments was partially 
 solidified. 
 
 The following temperatures were obtained by 
 the evaporation of carbon monoxide : 
 
 Pressure (in cm. of mercury) : 
 
 73.5 16 10 64 
 
 Temperature : 
 
 -190 -197. 5 -I98.83 -2oi.5 -2oi.6 
 
 The carbon monoxide was completely solidified. 
 The temperatures obtained by the evaporation of 
 liquid nitrogen are as follows : 
 
 Pressures: 74 12 8 6 4.2 
 
 Temperatures: -193 201 201. 7 204 206 
 
150 LIQUEFACTION OF GASES 
 
 The nitrogen at the lowest temperature was 
 partially solidified. 
 
 The author also determined the vapor-pressures 
 of liquid oxygen, carbon monoxide, and nitrogen at 
 different temperatures. 
 
 Liquefaction of Air 
 
 In 1885 Wroblewski made an elaborate series of 
 experiments on the liquefaction of air. 1 The ap- 
 paratus employed was similar to that represented 
 in figure 28. The object of these experiments 
 was to separate liquid air into two distinct layers. 
 
 In a preliminary experiment, a mixture of five 
 parts of carbonic acid and one part of air, at a 
 temperature of o, was compressed until a portion 
 of the gas was condensed to the liquid state. In 
 this condition the gas and liquid in the tube were 
 separated by a well-defined meniscus. As the 
 pressure was increased, the meniscus became less 
 distinct, and finally disappeared at the instant 
 when the optical density of the gas became equal 
 to that of the liquid. By increasing the pressure 
 still further, a second meniscus was produced, 
 which occupied a much higher position than the 
 first. In this condition the tube contained two 
 liquids, separated by a sharp meniscus. After a 
 
 1 Wied. Ann., 26, p. 134. 
 
EXPERIMENTS OF WROBLEWSKI 151 
 
 time the liquids mixed completely, and formed a 
 homogeneous liquid. 
 
 In a similar manner liquid atmospheric air was 
 obtained in two distinct layers, separated by a 
 well-defined meniscus. The air was partially lique- 
 fied at a temperature of 142, which is above the 
 critical temperature of nitrogen. When the press- 
 ure was increased to 40 atmospheres, the meniscus 
 which separated the liquid from the gas disap- 
 peared. The temperature was then lowered, and, 
 at a pressure of 37.8 atmospheres, a second menis- 
 cus was formed, which occupied a higher position 
 than the first. The air in this condition consisted 
 of two layers of liquid separated by a sharp menis- 
 cus. Analysis showed that the lower layer con- 
 tained about three per cent more oxygen than 
 the upper layer. After a couple of minutes the 
 two liquids mixed and formed a homogeneous 
 fluid. 
 
 The author also made a large number of obser- 
 vations on the vapor-pressure of liquid air at dif- 
 ferent temperatures. He also studied carefully 
 the changes of temperature and pressure which 
 accompanied the mixture of the two layers of 
 liquid air. 
 
 In addition to the experiments which have been 
 outlined, Wroblewski made numerous observations 
 on the properties of matter at low temperatures. 
 
152 LIQUEFACTION OF GASES 
 
 He also made a study of the relation between the 
 gaseous and liquid states of matter. 1 At the early 
 age of forty years, just in the midst of his active 
 career, he met with a fatal accident. "While 
 working late at night in his laboratory, he fell 
 asleep, and in his sleep he overthrew a kerosene 
 lamp. His clothing began to burn, and the wounds 
 thus received resulted, four days later, in death." 
 "The value of Wroblewski's supplement to the 
 work of Cailletet and Pictet cannot be easily 
 estimated, and we deplore the occurrence of the 
 fatal accident which robs the scientific world of 
 such an able experimenter in this important 
 department of chemistry." 2 
 
 EXPERIMENTS OF OLSZEWSKI 
 
 During the first few. years of investigation in 
 the field of the liquefaction of gases, Olszewski's 
 experiments were somewhat similar to those of 
 Wroblewski. The work of these two observers 
 was carried on simultaneously until the death of 
 the latter in 1888. 
 
 Soon after the liquefaction of oxygen, nitrogen, 
 
 1 Wied. Ann., 29, p. 428, 1886. 
 
 2 Wroblewski's work on the liquefaction of gases has been pub- 
 lished in the form of a pamphlet entitled, Comment Vair a ete 
 Liquefie. Ref. Devvar, Chem. News, 73, p. 42. 
 
EXPERIMENTS OF OLSZEWSKI 153 
 
 etc., by Wroblewski and Olszewski in 1883, the 
 latter experimenter made use of liquid oxygen as 
 a refrigerant in the liquefaction of other gases. 
 In 1884 he subjected hydrogen * to a pressure of 
 nearly 200 atmospheres, and at the same time 
 cooled the gas to the temperature of liquid oxygen 
 (boiling under a pressure of 6 mm. of mercury). 
 Under these conditions, hydrogen showed no 
 meniscus, but when the pressure was suddenly re- 
 leased a momentary ebullition was observed. The 
 author concluded that, at the temperature obtained 
 by the evaporation of liquid oxygen under reduced 
 pressure, it is impossible to obtain liquid hydrogen 
 in a static condition. 
 
 The experiments on hydrogen were repeated 
 during the same year, with similar results. 2 In 
 this series of experiments the author determined 
 the critical pressure of nitrogen, and found it to be 
 39.2 atmospheres. 
 
 Olszewski also made a large series of experi- 
 ments on the production of low temperatures by 
 the evaporation of liquid oxygen, nitrogen, air, etc. 
 under reduced pressure. The apparatus 3 employed 
 by him in the earlier experiments on the liquefac- 
 tion of gases is represented in figure 29. 
 
 The gas is liquefied in the thick-walled glass tube 
 
 1 Compt. rend., 98, p. 365. 2 Ibid., p. 913. 
 
 3 Wied. Ann., 31, p. 58. 
 
'54 
 
 LIQUEFACTION OF GASES 
 
 a, which is 30 centimetres in length, and 14 milli- 
 metres in diameter. The lower end of this tube 
 
 is sealed off, while the upper end is slightly wi- 
 dened, and fastened in the brass flange b, which 
 
EXPERIMENTS OF OLSZEWSKI 155 
 
 screws into the brass contrivance c. The tube d 
 represents a hydrogen thermometer. The liquefy- 
 ing tube a is connected, by means of the copper 
 tube e, ist, with the manometer f, which serves for 
 the measurement of high pressures lower than I at- 
 mosphere ; 2d, with the air manometer g for the 
 measurement of high pressures ; 3d, with a vacuum- 
 pump; 4th, with the aspirator r\ 5th, with the 
 Natterer receiver /, by means of which a pressure 
 of 60 to 80 atmospheres may be obtained. The 
 tube a is placed in a system of concentric glass 
 tubes, which communicate with each other at the 
 top. The outer tube is connected with an exhaust- 
 pump by means of the pipe n. The inner tube of 
 the system is connected by means of a copper tube 
 with the Natterer receiver /, which contains liquid 
 ethylene. The temperature of the liquefying tube 
 is lowered by the evaporation of this liquid under 
 reduced pressure. The liquid ethylene, before 
 entering the system of tubes, passes through a cop- 
 per coil, surrounded by ether and solid carbonic 
 acid, in the chamber m. By the evaporation of the 
 liquid ethylene, under a pressure of 10 mm. of 
 mercury, the temperature sank below 1 50. 
 
 The working of the apparatus is very simple. 
 After the temperature has been lowered to about 
 1 50, the gas is compressed in the tube a by 
 means of the vessel z, which contains the gas 
 
I 5 6 
 
 LIQUEFACTION OF GASES 
 
 under high pressure. The condensation to liquid 
 begins in a very short time. 
 
 The experiments which have been carried out 
 with this apparatus have been published in numer- 
 ous articles. In I884 1 a series of observations 
 were made on the critical temperature and press- 
 ure of nitrogen, and on the temperature obtained 
 by the evaporation of liquid ethylene and nitrogen 
 under reduced pressure. The results obtained with 
 ethylene are as follows : 
 
 Pressure 
 
 Temperature 
 
 750 mm. of mercury 
 
 -I0 3 
 
 107 " 
 
 - II5-5 
 
 3i " " 
 
 - 139 
 
 9.8 " 
 
 - i5 -4 
 
 The temperatures obtained by the evaporation 
 of liquid nitrogen are : 
 
 Pressure 
 
 Temperature 
 
 35 atmospheres 
 
 -I 4 6 
 
 17 
 
 - 1 60 
 
 i " 
 
 - 194.4 (boiling point) 
 
 In vacuo 
 
 - 213 
 
 1 Compt. rend., 99, p. 133. 
 
EXPERIMENTS OF OLSZEWSKI 157 
 
 A few months later he made use of liquid air l 
 as a refrigerant, and measured the temperatures 
 obtained under different pressures. 
 
 Pressure 
 
 Temperature 
 
 39 atmospheres 
 14 
 
 4 
 
 i " 
 
 In vacuo 
 
 140 (critical point) 
 
 146 
 
 176 
 
 191.4 (boiling point) 
 
 205 
 
 Solidification of Gases 
 
 In carrying out the experiments at low temper- 
 atures, Olszewski succeeded in solidifying a num- 
 ber of gases, and in measuring their melting 
 points. In i884 2 he evaporated liquid carbon 
 monoxide under reduced pressure, and obtained a 
 very low temperature. At a pressure of one at- 
 mosphere the temperature was 190; this rep- 
 resents the boiling point of carbon monoxide. By 
 reducing the pressure as far as possible the tem- 
 perature sank to 211, and the substance solidi- 
 fied to an opaque mass. 
 
 Early the next year a series of similar experi- 
 ments were made with nitrogen. 3 At a tempera- 
 
 1 Compt. rend., 99, p. 184. 2 Ibid., p. 706. 
 
 3 Ibid., 100, p. 350, 
 
i 5 8 
 
 LIQUEFACTION OF GASES 
 
 ture of 214 the substance solidified to an 
 opaque mass. By evaporating the resulting solid 
 under a pressure of 4 mm. of mercury the tem- 
 perature sank to 225. 
 
 During the same year methane and nitric oxide 
 were evaporated under reduced pressure. 1 The 
 former gas solidified at a temperature of 185. 8. 
 At this temperature the vapor-pressure was 80 
 mm. of mercury. When the pressure was re- 
 duced to 5 mm. the temperature sank to 201. 5. 
 Nitric oxide solidified at a temperature of 167. 
 Under a pressure of 18 mm. of mercury a tem- 
 perature of i 76. 5 was obtained. 
 
 Substance 
 
 Freezing Point 
 
 Chlorine 
 
 Hydrochloric acid 
 
 Hydrofluoric acid 
 
 Phosphine 
 
 Arsine 
 
 Stibine 
 
 Ethylene 
 
 Silicon tetrafluoride 2 
 
 Hydrogen selenide 3 
 
 102 
 
 - 116 
 -9 2 -3 
 
 - 133 
 
 -91.5 
 -169 
 
 - 102 
 
 -68 
 
 1 Compt. rend., loo, p. 940. 
 
 2 Silicon tetrafluoride does not melt, but sublimes into a gas at 
 the ordinary pressure. 
 
 8 Phil. Mag., 39, p. 210. 
 
EXPERIMENTS OF OLSZEWSKI 159 
 
 Olszewski also solidified a number of other 
 gases 1 some of which had been liquefied many 
 years before. The freezing points are given in 
 the preceding table. 
 
 Ethane and Propane remained liquid at a tem- 
 perature of 151. 
 
 In 1885 Olszewski made use of liquid air as a 
 refrigerant. 2 By evaporating this liquid under 
 a very low pressure, a temperature of 220 was 
 obtained. An account is also given in this article 
 of the condensation of a mixture of two volumes 
 of hydrogen with one volume of oxygen to a color- 
 less liquid which formed in thin layers. 
 
 Densities and Boiling Points of Liquid Methane \ 
 
 Nitrogen, and Oxygen 
 
 In 1886 methane, oxygen, and nitrogen 3 were 
 liquefied by means of the apparatus described on 
 page 154. A series of observations were then 
 made on the densities of these liquids at their boil- 
 ing points. The following results were obtained : 
 
 Liquid 
 
 Density 
 
 Methane 
 Oxygen 
 
 Nitrogen 
 
 0.415 (mean of 3 determinations) 
 i.i24( "6 " ) 
 0.885 ( "4 " ) 
 
 1 Wien. Monatshefte fur Chem., 5, p. 127; Wien. Ber., 94, p. 209. 
 
 2 Compt. rend., 101, p. 238. 8 Wied. Ann., 31, p. 58. 
 
160 LIQUEFACTION OF GASES 
 
 The following boiling points were also deter- 
 mined : 
 
 Liquid 
 
 Boiling Point 
 
 Methane 
 
 -I6 4 
 
 Oxygen 
 
 - 181.4 
 
 Nitrogen 
 
 - 1944 
 
 Carbon monoxide 
 
 IQO 
 
 Nitric oxide 
 
 ~ 153 
 
 These temperatures were measured by means 
 of a hydrogen thermometer. 
 
 Liquefaction of Ozone 
 
 Olszewski repeated the experiments of Haute- 
 feuille and Chappuis on the liquefaction of ozone, 
 with a view of determining the boiling point. 1 The 
 ozone was prepared by -means of Siemens' appara- 
 tus. In order to obtain the liquid in a static con- 
 dition under a pressure of one atmosphere, liquid 
 oxygen was used as the refrigerant. The appa- 
 ratus previously described was employed in this 
 experiment. 
 
 At the temperature of boiling oxygen ( i8i.4), 
 the ozone condensed to a dark blue liquid. In 
 layers of two millimetres in thickness the liquid 
 
 1 Wied. Ann., 37, p. 337. 
 
EXPERIMENTS OF OLSZEWSKI 161 
 
 appeared opaque. The liquid oxygen which sur- 
 rounded the tube of ozone was evaporated under 
 reduced pressure with a view of solidifying the 
 ozone. The experiment, however, was unsuccess- 
 ful. The boiling point was determined, and found 
 to be 1 06. The liquid is somewhat explosive, 
 especially when brought into contact with liquid 
 ethylene. 
 
 Apparatus for the Liquefaction of Gases on a 
 Large Scale 
 
 According to the earlier observations of Olszew- 
 ski it appears that liquid oxygen is a better refrig- 
 erant than liquid air or nitrogen. In order that 
 this liquid might be used as a refrigerant in the 
 liquefaction of other gases, the author constructed 
 an apparatus by means of which oxygen can be 
 liquefied on a comparatively large scale. This 
 apparatus was constructed in 1890, and enlarged 
 during the same year. 1 A section of the enlarged 
 apparatus is shown in figure 30. 
 
 The gas is liquefied in the steel cylinder a, of 
 about 200 cc. capacity. The upper end of this cyl- 
 inder is connected by means of a small copper 
 tube with a metallic manometer b, and an iron 
 vessel c of 10 litres capacity. The vessel c con- 
 
 1 Phil. Mag. [5], 39, p. 192. 
 
162 
 
 LIQUEFACTION OF GASES 
 
 tains dry air or oxygen under a pressure of 100 
 atmospheres. The lower end of the cylinder a is 
 connected by means of a copper tube with the 
 screw-cock d, which serves as an outlet for the 
 
 FIG. 30. 
 
 liquid oxygen or air. The double- or triple-walled 
 glass vessel m which surrounds the cylinder a is 
 connected with the reservoir f, which contains 
 liquid ethylene, and with an exhaust-pump, by 
 means of the tube /. The liquid ethylene, before 
 
EXPERIMENTS OF OLSZEWSKI 163 
 
 entering the vessel m, passes through the chamber 
 g t which contains a mixture of solid carbonic acid 
 and ether. The temperature of this mixture is 
 lowered still further by connecting the tube n with 
 an exhaust-pump. The method of operation is as 
 follows : - 
 
 The pressure in the chamber g is first lowered 
 to 50 mm. of mercury. The vessel m is then 
 brought into communication with the exhaust- 
 pump and the reservoir of liquid ethylene. The 
 ethylene eva-porates rapidly at first, but finally col- 
 lects as a liquid in the vessel m. As soon as the 
 liquid ethylene completely surrounds the vessel a, 
 the supply is cut off. When the temperature is 
 lowered to the critical temperature of the gas con- 
 tained in the vessel c, communication is estab- 
 lished between this vessel and the cylinder a. 
 The gas enters the cooled cylinder under a press- 
 ure indicated by the manometer b, and soon 
 begins to liquefy. The manometer, during the 
 condensation, shows a constant fall, and becomes 
 stationary only when the cylinder a is completely 
 filled with liquid. When this is accomplished the 
 vessel c is closed, and the liquid oxygen or air is 
 allowed to pass out through the cock d into the 
 double- or triple-walled glass vessel e, after which 
 the process may be repeated. 
 
 The temperatures in these experiments were 
 
164 LIQUEFACTION OF GASES 
 
 not measured directly, but were calculated from 
 the pressure under which the liquid ethylene was 
 allowed to evaporate. This pressure was meas- 
 ured by means of the metallic vacuometer k. 
 
 With this apparatus Olszewski obtained 200 
 cubic centimetres of liquid air, and even larger 
 quantities of liquid oxygen. In the earlier experi- 
 ments it was thought that the liquid oxygen was 
 colorless, but when obtained in a larger quantity it 
 was found to be pale blue in color. 
 
 The Critical Constants and Boiling Point of 
 Hydrogen 
 
 In 1891 Olszewski 1 made another series of 
 experiments with hydrogen. The apparatus em- 
 ployed was a modification of that represented in 
 figure 30. Liquid oxygen and liquid air were 
 used as refrigerants in these experiments. The 
 author was unable to obtain any indications of a 
 meniscus of liquid hydrogen, but on allowing the 
 gas to expand at very low temperatures signs of 
 ebullition were observed. Moreover, when the 
 hydrogen was allowed to expand slowly from 
 pressures of 80, 90, 100, no, 120, and 140 atmos- 
 pheres, the phenomenon of ebullition always 
 appeared at a pressure of 20 atmospheres. From 
 
 1 Phil. Mag. [5], 39, p. 199. 
 
EXPERIMENTS OF OLSZEWSKI 165 
 
 these observations Olszewski concluded that 20 
 atmospheres represent the critical pressure of 
 hydrogen. 
 
 In order to test the accuracy of this method of 
 determining the critical pressure of a gas, the 
 experiments were extended to oxygen and ethy- 
 lene. The critical pressure of oxygen had already 
 been measured, and found to be 50.8 atmospheres. 
 To determine the same constant by the expansion 
 method, the gas was cooled to a temperature of 
 about 1 6. 3 above the critical point, and then 
 allowed to slowly expand. In every case the 
 meniscus and signs of ebullition appeared at a 
 pressure of about 5 1 atmospheres, provided the 
 initial pressure was not lower than 80 atmos- 
 pheres. The author concluded, from his ex- 
 periments on oxygen and ethylene, that critical 
 pressures can be determined very closely by 
 means of the expansion method. 
 
 A few years later 1 the author endeavored to 
 measure the temperature of the hydrogen at the 
 moment of expansion. For this purpose a very 
 sensitive platinum resistance thermometer was 
 employed. The thermometer was previously com- 
 pared with a hydrogen thermometer at temper- 
 atures varying from o to 208. 5. When this 
 
 1 Phil. Mag. [5], 40, p. 202. 
 
166 LIQUEFACTION OF GASES 
 
 comparison had once been made it was a sim- 
 ple matter to calculate temperatures lower than 
 208. 5. The following results were obtained: 
 
 Expansion of Hydrogen from a High 
 Pressure to 
 
 Temperature 
 
 20 atmospheres (critical pres.) 
 
 10 " 
 
 i 
 
 234. 5 (critical temperature) 
 
 -239 -7 
 -243 .5 (boiling point) 
 
 The experiments were then extended to oxygen, 
 which was allowed to expand at a temperature of 
 about 1 6 above the critical point. The critical 
 temperature of oxygen determined in this way was 
 found to be 118 to 119. 2, and the boiling 
 point i8i.4 to 182. 5. The corresponding 
 values measured directly with a hydrogen ther- 
 mometer are n8.8 and i8i.4 to 182.7, 
 respectively. 
 
 Olszewski concluded, from these observations, 
 that the critical temperature of hydrogen is about 
 - 234. 5, and the boiling point 243. 5. These 
 values agree very closely with those obtained theo- 
 retically by Natanson. 1 
 
 In connection with the experiments which have 
 been outlined, Olszewski also made numerous ex- 
 periments on the properties of liquefied gases, and 
 
 1 Phil. Mag. [5], 40, p. 272, 1895. 
 
EXPERIMENTS OF DEWAR 167 
 
 on the properties of matter in general at low tem- 
 peratures. These investigations were carried out, 
 for most part, in collaboration with his colleague 
 Witkowski. The observations of Olszewski on 
 argon and helium will be considered in Section IV 
 of this chapter. 
 
 EXPERIMENTS OF DEWAR 
 
 Reference has already been made to the experi- 
 ments of Dewar on the liquefaction of gaseous 
 mixtures, and the determination of their critical 
 constants. These observations were made in 1880. 
 Four years later he extended his experiments to 
 the so-called permanent gases, and constructed an 
 apparatus for demonstrating the liquefaction of 
 oxygen in the lecture room. 1 The essential parts 
 of the apparatus are shown in figure 31. The gas 
 to be liquefied is contained in the iron reservoir C 
 under a pressure of 150 atmospheres. This res- 
 ervoir is connected with the manometer D, and 
 also, by means of the small copper tube /, with 
 the liquefying tube F. The pressure is regulated 
 by means of the screw-cock A. The liquefying 
 tube is placed in the glass tube G, which con- 
 tains liquid ethylene, solid carbonic acid, or liquid 
 nitrous oxide. This tube rests in a larger glass 
 
 l Prot. Koy. Inst., 1884, p. 148 ; Phil. Mag. [5], 18, p. 2io. 
 
1 68 
 
 LIQUEFACTION OF GASES 
 
 tube, with which it communicates through the 
 opening E. The outer tube is connected with a 
 Bianchi exhaust-pump, so that the liquid ethylene, 
 etc., can be evaporated under reduced pressure. 
 
 FIG. 31. 
 
 The pressure under which the evaporation takes 
 place is measured by means of the gauge J. 
 
 The oxygen, at the temperature obtained by the 
 evaporation of liquid ethylene under a pressure of 
 25 millimetres of mercury, could easily be lique- 
 
EXPERIMENTS OF DEWAR 169 
 
 fied at a pressure of from 20 to 30 atmospheres. 
 When liquid nitrous oxide or solid carbonic acid 
 were used as refrigerants, it was necessary to 
 make use of the sudden expansion of the gas to 
 lower the temperature still further. This was 
 accomplished by increasing the pressure in the 
 liquefying tube to 80 atmospheres, and then open- 
 ing the screwcock B. 
 
 This method was thoroughly satisfactory for 
 demonstrating the liquefaction of oxygen. The 
 liquefying tube was only about five millimetres 
 in diameter, and hence only small quantities of 
 liquid could be obtained. The tube filled with 
 liquid oxygen (for projection) contained about 1.5 
 cubic centimetres. Some rough observations were 
 also made on the density of liquid oxygen. 
 
 In 1886 Dewar 1 published an account of an 
 apparatus by means of which oxygen, air, etc., 
 can be liquefied on a much larger scale. A sec- 
 tion of the apparatus is represented in figure 32. 
 The chamber b contains a mixture of solid car- 
 bonic acid and ether. Ethylene is conducted into 
 the tube a, where it is liquefied by means of th'e low 
 temperature produced by the carbonic acid. This 
 liquid passes from the coiled tube into the cham- 
 ber d, which is surrounded by a larger vessel con- 
 
 l Proc. Roy. Inst., 1886, p. 550. 
 
LIQUEFACTION OF GASES 
 
 ^ a taining solid carbonic 
 acid and ether. The 
 liquid ethylene evapo- 
 rates into the space 
 between the two 
 chambers. A contin- 
 uous copper tube, 
 about forty-five feet 
 in length, passes first 
 through the outer 
 vessel, and then 
 through the chamber con- 
 taining liquid ethylene. A 
 very low temperature can 
 be produced in this way. 
 
 When the temperature 
 has been lowered, as de- 
 scribed above, perfectly dry 
 oxygen gas is introduced 
 into the copper tube under 
 a pressure of about 75 at- 
 mospheres. The pressure 
 gauge (not represented in 
 the figure) soon indicates 
 the beginning of liquefaction 
 in the tube. The valve A is 
 then opened and the liquid 
 oxygen rushes out into the 
 
 FIG. 32. 
 
EXPERIMENTS OF DEWAR 171 
 
 glass tube g which is immersed in the liquid 
 ethylene contained in the tube /. In the first ex- 
 periments about 22 cubic centimetres of liquid 
 oxygen were obtained in the tube. In order to 
 reduce the temperature as far as possible, the 
 tubes g and i were each connected with an exhaust- 
 pump. A temperature of about 200 was thus 
 obtained. During this process a white deposit 
 was formed in the tube g y which was supposed, 
 then, to be solid oxygen. Later, however, the 
 author observed that these white particles of solid 
 matter were due to impurities. 
 
 Having constructed an apparatus for liquefying 
 oxygen, air, etc., in considerable quantity, Dewar 
 extended his observations on the properties of 
 liquefied gases, and the properties of matter in 
 general at low temperatures. In 1892 he pub- 
 lished an account of some observations on the 
 magnetic properties of liquid oxygen. 1 Becquerel 
 was the first to call attention to this property of 
 oxygen. He experimented with charcoal which 
 was saturated with oxygen gas. In 1849 Faraday 
 noticed that oxygen is strongly magnetic in com- 
 parison with other gases. Dewar observed that if 
 liquid oxygen is placed between and a little below 
 the poles of an electro-magnet, the liquid will rise 
 
 1 Proc. Roy. Inst., 13, p. 695. 
 
172 LIQUEFACTION OF GASES 
 
 to the poles when the circuit is completed. The 
 magnetic moment of liquid oxygen, he says, is 
 about 1000, when the magnetic moment of iron is 
 taken as 1,000,000. When liquid air was placed 
 between the poles of the magnet, all of the liquid 
 was drawn to the poles, and there was no separa- 
 tion of oxygen and nitrogen. 
 
 The absorption spectrum of liquid oxygen was 
 also examined, and found to be the same as 
 that of the gas. Both the liquid and the highly 
 compressed gas show .a series of five absorption 
 bands, situated respectively in the orange, yellow, 
 green, and blue of the spectrum. Some experi- 
 ments were also made on chemical action at low 
 temperatures. Reference will be made to these 
 observations in the conclusion at the end of the 
 volume. 
 
 During the next year Dewar announced that he 
 had succeeded in freezing ordinary air into a clear, 
 transparent solid. 1 This experiment is of consider- 
 able importance, inasmuch as neither oxygen nor 
 air had been solidified previous to this time. 
 
 The Dewar Vacuum Bulbs 
 
 In 1893, Dewar 2 made an important contribu- 
 tion to the researches at low temperatures by the 
 
 1 Chem. News, 67, p. 126. 2 Proc. Roy. Inst., 14, p. I. 
 
EXPERIMENTS OF DEWAR 
 
 173 
 
 introduction of a vessel for storing such volatile 
 fluids as liquid oxygen, liquid air, etc. These 
 liquids, and other similar liquids, which had been 
 obtained previous to this time, could be kept only 
 a few minutes in the open 
 air owing to the rapid 
 evaporation. Experimen- 
 ters had already retarded 
 the evaporation to some 
 extent by the use of 
 double-walled vessels. 
 Dewar conceived the idea 
 of exhausting the air from 
 the space between the 
 walls of such vessels. 
 
 Figure 33 represents 
 one of these bulbs. The 
 outer vessel is exhausted 
 to a very high degree, 
 and then sealed off at 
 the lower end. This pre- 
 vents, to a considerable 
 extent, the convective transference of 'heat from 
 the outer air to the liquid air in the inner vessel. 
 
 To test the efficiency of this form of vessel, 
 comparisons were made with similar vessels which 
 had not been exhausted. The comparisons were 
 based upon the amount of liquid oxygen or liquid 
 
 FIG. 33. 
 
174 
 
 LIQUEFACTION OF GASES 
 
 ethylene evaporated in a given time. The vessels 
 were placed in water which was kept at constant 
 temperature. The results were as follows : 
 
 Amount of Gas Evolved 
 
 Liquid oxygen surrounded by a vacuum 
 
 chamber 
 
 Liquid oxygen surrounded by an air 
 
 chamber 
 
 Liquid ethylene surrounded by a vacuum 
 
 chamber 
 
 Liquid ethylene surrounded by an air 
 
 chamber 
 
 170 cc. per mm. 
 840 
 56 " 
 250 
 
 From these observations it is evident that the 
 rate of evaporation from the air bulbs is about five 
 times that from the vacuum bulbs. By covering 
 the inner vessel with a thin deposit of silver, the 
 rate of evaporation is reduced to less than one half 
 of that given in the table. Liquid oxygen or liquid 
 air can be kept for several hours in such vessels 
 under the ordinary atmospheric pressure. 
 
 The next step was to construct a series of 
 vacuum vessels of different forms which could 
 be employed in experiments of different nature. 
 Figure 34 represents a few of the many forms of 
 these vessels which have been constructed. Dewar 
 and other experimenters have made use of these 
 
EXPERIMENTS OF DEWAR 
 
 175 
 
 bulbs for investigating the properties of matter at 
 low temperatures, and for determining the proper- 
 ties and physical constants of liquefied gases. 
 
 In 1894 Dewar determined the thermal trans- 
 parency of some liquefied gases for heat of high 
 
 FIG. 34 . 
 
 refrangibility. 1 Taking chloroform as the unit of 
 comparison, and correcting for differences in the 
 refractive indices, the results are as follows : 
 
 Chloroform i.o. 
 
 Carbon disulphide 1.6 
 
 Liquid oxygen 0.9 
 
 Liquid nitrous oxide 0.93 
 
 Liquid ethylene 0.60 
 
 Ether 0.50 
 
 1 Proc. Roy. Inst., 14, p. 393. 
 
176 LIQUEFACTION OF GASES 
 
 Notwithstanding the low temperatures, liquid 
 oxygen and nitrous oxide are very transparent to 
 high temperature radiation. 
 
 The author suggested in this article that, instead 
 of silvering the inner bulbs of the vacuum vessels, 
 it is better to leave a small quantity of mercury in 
 the vacuum chambers. When liquid air is intro- 
 duced into the inner bulbs, a thin coating of solid 
 mercury is found on the outer surface. 
 
 A series of observations were also made on 
 the breaking stress of metals at low tempera- 
 tures. The following table contains the breaking 
 stress, in pounds, for metallic wires of 0.098 in. in 
 
 diameter : 
 
 15 -182 
 
 Steel (soft) 420 700 
 
 Iron 320 670 
 
 Copper ...'.... 200 300 
 
 Brass 310 440 
 
 German silver 470 600 
 
 Gold 255 340 
 
 Silver 330 420 
 
 The breaking stress of wires is not changed 
 by cooling down to 182, and then allowing the 
 temperature to rise. 
 
 In 1895 Dewar made use of a different form 
 of apparatus in the liquefaction of gases. 1 This 
 apparatus involves the regenerative principle in 
 
 1 Proc. Roy. Inst., 15, p. 133. 
 
EXPERIMENTS OF DEWAR 177 
 
 the method of refrigeration, and belongs more 
 properly to the next section (see p. 197). 
 
 By means of this apparatus liquid air, etc., can 
 be obtained in a very short time, and at a com- 
 paratively small expense. By placing a litre of 
 liquid air in a globular vacuum bulb and subject- 
 ing it to exhaustion, the author states that as much 
 as half a litre of solid air can be obtained and 
 maintained in the solid condition for a period of 
 half an hour. An examination of the solid showed 
 it to be a " nitrogen- jelly " containing liquid oxygen. 
 
 After obtaining liquid air, oxygen, and nitrogen 
 in considerable quantities, Dewar made an elabo- 
 rate series of experiments on their densities. The 
 method consisted in weighing different substances 
 of known specific gravities in the liquids. More 
 than twenty substances were weighed in liquid 
 oxygen, and corrections made for the contraction 
 of the solids. The mean of the results gave a 
 value of 1.1375 for the density of liquid oxygen. 
 By weighing a large silver ball in liquid air and 
 liquid nitrogen, the densities of the liquids were 
 found to be 0.910 and 0.850 respectively. 
 
 In 1897 Dewar constructed an apparatus for the 
 examination of the least condensible portion of 
 air. 1 From the experiments with this apparatus 
 
 1 Chem. News, 76, p. 272. 
 
178 LIQUEFACTION OF GASES 
 
 the author concluded that every gas which occurs 
 in the atmosphere either condenses to the liquid 
 state, or is soluble in liquid air. 
 
 Numerous observations have been made by 
 Dewar and Fleming on the electric conductivity 
 of metals, and the dielectric constant of organic 
 liquids at low temperatures. 1 
 
 The experiments of Dewar on the liquefaction 
 of gases since 1895 will be considered in the next 
 two sections. 
 
 EXPERIMENTS OF KAMERLINGH-ONNES 
 
 During the last few years Kamerlingh-Onnes 
 has been experimenting on the liquefaction of 
 gases. The apparatus which is used by him at 
 present in the liquefaction of oxygen is somewhat 
 similar to that employed by Pictet in 1877. The 
 complete operation consists of three cycles, as 
 follows : 2 
 
 First Cycle. Liquid methyl chloride surrounds 
 a tube containing ethylene under pressure. The 
 vessel which contains the methyl chloride is con- 
 nected with an exhaust-pump, by means of which 
 the pressure is considerably reduced. The rapid 
 evaporation of the liquid lowers the temperature 
 
 1 A list of references to these observations has been compiled by 
 Dickson. Phil. Mag. [5], 45, p. 528, 1898. 
 
 2 Zeit. fur conipr. und fliiss. Gase, I, p. 169, 1898. 
 
EXPERIMENTS OF KAMERLINGH-ONNES 179 
 
 of the chamber, and hence, of the ethylene. The 
 vapors of methyl chloride which pass out of the 
 vessel are compressed, by means of a second 
 pump, in a chamber where they finally condense 
 to a liquid. This liquid passes back again to the 
 vessel which surrounds the ethylene tube, thus 
 making the cycle complete. 
 
 Second Cycle. This cycle is similar to the first. 
 The compressed ethylene is liquefied by means of 
 the low temperature which results from the evapo- 
 ration of the methyl chloride. The liquid ethylene 
 is conducted into a tube, where it surrounds a 
 smaller tube containing compressed oxygen gas. 
 The liquid ethylene is connected with an exhaust- 
 pump, and evaporated under reduced pressure. 
 The resulting vapors are recondensed, and the 
 cycle is made complete. 
 
 Third Cycle. In this cycle the oxygen is con- 
 densed to the liquid state. The oxygen tube is 
 connected with both an exhaust and a compression 
 pump. The liquid ethylene, boiling under reduced 
 pressure, lowers the temperature of the compressed 
 oxygen to -- 140, when it condenses to a liquid. 
 By means of the pumps the liquid oxygen is 
 removed to a carefully protected reservoir. The 
 operation consists of three complete cycles, and 
 hence is continuous. 
 
 A section of the apparatus is shown in the 
 
i8o LIQUEFACTION OF GASES 
 
 article to which reference has already been made. 
 The drawing, however, is rather complicated, and 
 for that reason has not been reproduced here. By 
 means of this apparatus the author has obtained 
 liquid oxygen in considerable quantity. 
 
 SECTION III 
 
 LIQUEFACTION OF GASES BY THE REGENERATIVE 
 METHOD 
 
 So far, the low temperatures necessary for the 
 liquefaction of gases have been obtained mainly 
 by the evaporation, especially under reduced press- 
 ure, of liquid carbonic acid, liquid nitrous oxide, 
 liquid ethylene, liquid air, etc. A different method 
 of lowering the temperature has become of consid- 
 erable importance in recent years. The gas which 
 is to be liquefied is compressed to a very high 
 pressure ; the heat due to the compression is re- 
 moved by means of water, and the gas is then 
 allowed to expand. In this way exceedingly low 
 temperatures can be obtained. With this method 
 the gas to be liquefied is the only substance, apart 
 from the apparatus, which is necessary for carry- 
 ing out the experiment. The process is frequently 
 referred to as the self-intensification^ or the regen- 
 erative, method. 
 
REGENERATIVE METHOD 181 
 
 The history of this method is rather interesting. 
 According to Joule, 1 Dr. Cullen and Dr. Darwin 
 were the first to observe that the temperature of a 
 gas is lowered by rarefaction and increased by 
 compression. They experimented with ordinary 
 air. Dalton succeeded in measuring this change 
 of temperature with some degree of accuracy (see 
 p. 12). In 1806 Gay Lussac (p. 12) made a series 
 of experiments on the changes of temperature 
 which accompany the compression and rarefaction 
 of gases. Thilorier (1834) observed that carbonic 
 acid could be solidified by means of its own expan- 
 sion (p. 38). In 1845 Joule, and later Joule and 
 William Thomson, made a very exhaustive study of 
 these phenomena, both experimentally and theo- 
 retically. Mayer, Rankine, and Clausius have also 
 contributed largely to this subject. Further refer- 
 ence to these investigations will be made in the theo- 
 retical considerations which will be taken up later. 
 
 In connection with these and many other similar 
 observations which were conducted solely for scien- 
 tific purposes, a number of experiments have also 
 been made with a view of applying this method of 
 lowering the temperature for refrigerating pur- 
 poses. In 1849 Dr. John Gorrie 2 constructed an 
 
 1 Phil. Mag. [3], 26, p. 369. 
 
 - Wallis-Tayler, Refrigerating and Ice-Making Machinery, 
 p. 116. 
 
1 82 LIQUEFACTION OF GASES 
 
 ice machine, based upon the compression and 
 expansion of air. In 1857 Siemens 1 constructed 
 a refrigerating machine, which he describee as 
 follows : 
 
 " The invention relates to freezing and refriger- 
 ating by the expansion of air or elastic fluid. The 
 air is first compressed by a cylinder, or by pumps 
 of any suitable construction, by which the temper- 
 ature is raised, and it is cooled while in the com- 
 pressed state, and is then allowed to expand in a 
 cylinder or engine of any suitable construction, by 
 which the temperature is lowered. The air thus 
 cooled is brought in contact with the articles to be 
 cooled or frozen, and is then conducted through an 
 interchanger, or apparatus, by which it is made to 
 cool the compressed air which enters the inter- 
 changer in the opposite direction. . . . The prin- 
 ciple of the invention is adapted to produce an 
 accumulated effect, or an indefinite reduction of 
 temperature." Kirk in 1863, Marchant in 1869, 
 and Giffard and Postle in 1873, constructed ice 
 machines in which the freezing was produced by 
 the expansion of air. 2 In 1885 Solvay 3 described 
 a method which is similar to that of Siemens. A 
 
 1 Linde, Engineer, 82, p. 486. 
 
 2 Wallis-Tayler, Refrigerating and Ice-Making Machinery, 
 p. 116 et seq. 
 
 3 Linde, Engineer, 82, p. 486. 
 
REGENERATIVE METHOD 183 
 
 very efficient refrigerating machine, based upon 
 this same principle, was constructed by Wind- 
 hausen. 1 Reference might be made to other ma- 
 chines of a similar nature, but it is not within the 
 scope of a book of this kind to consider the subject 
 of refrigerating machines. Those who are inter- 
 ested in that branch of the work can refer, in con- 
 nection with the various text-books and journals 
 on engineering, to Wallis-Tayler's Refrigerating 
 and Ice-Making Machinery, and especially to the 
 two journals Ice and Refrigeration, published in 
 New York and Chicago, and the Zeitschrift fiir die 
 gesammte Katie-Industrie, published in Leipzig. 
 
 The preceding examples, however, are sufficient 
 to show that the lowering of temperature, which 
 accompanies the expansion of a gas, was observed 
 a century ago, and that this method of lowering 
 the temperature has been applied technically for at 
 least fifty years. The method is not new. 
 
 In 1874 Edwin J. Houston 2 suggested a form of 
 apparatus which is very similar to the machines 
 employed at present in the liquefaction of air. 
 The following are his words : " The means of 
 obtaining exceedingly low temperatures seem at 
 last to have been fulfilled in the ' Windhausen Ice 
 and Refrigerating Machine.' Though introduced 
 
 1 Houston, Jour. Frank. Inst., 67, p. 10. 
 
 2 ibid., p. 9, 1874- 
 
184 LIQUEFACTION OF GASES 
 
 for practical purposes, mainly for the cheap pro- 
 duction of artificial ice, the machine contains latent 
 possibilities, which we hope will at once be utilized, 
 that open up the most promising field to the origi- 
 nal investigator, and bid fair to enrich science with 
 stores of new facts." 
 
 " . . . In the Windhausen process, a steam 
 engine is employed to compress the air to 2 or 3 
 atmospheres. The heat developed by the com- 
 pression is drawn off during the passage of the 
 condensed air through pipes in a series of cham- 
 bers, in which cold water is flowing. The cooled 
 air is then allowed to expand into a cylinder un- 
 der gradually diminishing pressure, the expansion 
 being attended with the development of great 
 cold." 
 
 ". . . The following modifications of the ap- 
 paratus would render its cold-producing power 
 almost unlimited : 
 
 "i. A communication between the expansion 
 cylinder and the chambers through which the con- 
 densed air is conducted before it is allowed to 
 expand. Supposing this outlet regulated by a 
 cock, a blast of very cold air could replace the run- 
 ning water, and reduce the compressed air to a 
 very low temperature. 
 
 " 2. The introduction of a second compressing 
 cylinder, with which the compressed air, after be- 
 
REGENERATIVE METHOD 185 
 
 ing cooled, could be still further compressed, again 
 cooled, and finally conducted into the expansion 
 cylinder. Under a pressure of, say, 60 atmos- 
 pheres, a considerable mass of air at the tempera- 
 ture of, say, i 00 F. would, in its expansion, 
 produce a reduction of temperature greater per- 
 haps than any yet obtained. . . . There would 
 appear to be no other limit to the reduction of 
 temperature save what would arise from the 
 strength of materials, or the liquefaction and sub- 
 sequent freezing of the nitrogen, or the oxygen of 
 the air, or of the air itself. 
 
 " Among the advantages that we may rationally 
 expect to accrue from the apparatus thus modified 
 are the following : 
 
 " i. The confirmation or otherwise of the ' abso- 
 lute zero ' as determined by the expansion or 
 contraction of gases, by heat or cold. 
 
 " 2. The liquefaction and subsequent solidifica- 
 tion of many of the incoercible gases, the determi- 
 nation of their physical peculiarities as liquids or 
 solids, together with their crystalline form. 
 
 "3. The action of intense cold on the chemical 
 affinities of certain gaseous compounds. 
 
 "4. The action of intense cold on the color of 
 certain chemical compounds." 
 
 These predictions have been thoroughly fulfilled. 
 In the apparatus suggested, Houston anticipated 
 
186 LIQUEFACTION OF GASES 
 
 the methods which were employed twenty years 
 later in the liquefaction of air by Linde, Hampson, 
 and Tripler. 
 
 In 1875 Coleman 1 constructed an apparatus for 
 the liquefaction, on a large scale, of some very 
 volatile hydrocarbons. The lowering of the tem- 
 perature was produced by the expansion of the 
 compressed gases. The machine involves : 
 
 " i. The pumping of the gas by steam power 
 into a system of tubes capable of being externally 
 cooled, and from which condensed liquids can be 
 drawn off by ball-cocks. 
 
 " 2. Employing the compressed gas, after being 
 deprived of its liquid, for working a second engine 
 coupled with, and parallel to, the first, thus receiv- 
 ing a portion of the force originally employed in 
 the compression. 
 
 " 3. Employing the expanded gas, after having 
 had its temperature reduced in the act of doing 
 the work of pumping, for supplying the necessary 
 cold for cooling a portion of the condenser pipes 
 to zero." 
 
 The cycle of operations was complete. The 
 temperature of the compressed gas was about 5, 
 while the minimum temperature after expansion 
 was 45. With this apparatus a continuous 
 
 1 Chem. News, 39, p. 87. 
 
EXPERIMENTS OF LINDE 187 
 
 stream of the liquefied gas could be produced. 
 About 250,000 gallons of the liquid hydrocarbons 
 were produced during the first three years. 
 
 Cailletet and Pictet have also made use of the 
 expansion of gases as a means of lowering the 
 temperature, but in their work the regenerative 
 principle was not used. Wroblewski, Olszewski, 
 and Dewar, previous to 1895, made experiments 
 of a similar nature. Reference has already been 
 made to these observations. In 1894 Kamerlingh- 
 Onnes made use of the regenerative principle in 
 the liquefaction of gases. 1 Omitting a number of 
 observations on the compression and expansion of 
 gases, which have but slight bearing on the conden- 
 sation of gases, we may proceed at once to the con- 
 sideration of some forms of apparatus which have 
 recently been constructed for the purpose of lique- 
 fying oxygen, nitrogen, and especially ordinary air. 
 
 APPARATUS EMPLOYED BY LINDE IN THE 
 LIQUEFACTION OF AIR 
 
 This apparatus is an outgrowth of a long expe- 
 rience in the construction of refrigerating ma- 
 chines. We may omit these devices, however, 
 and proceed at once to the consideration of the 
 apparatus employed in the liquefaction of air. 
 
 1 Ref. Dewar, Proc. Roy. InsL, 15, p. 133. 
 
188 LIQUEFACTION OF GASES 
 
 The apparatus was successfully operated in 
 May, 1895, and was constructed as follows : l The 
 general arrangement of the apparatus is shown in 
 figure 35. The air to be liquefied is brought by 
 means of the compressor C to a pressure of about 
 200 atmospheres (in the first few experiments the 
 pressure employed was about 65 atmospheres). R 
 is a water-cooler, which serves to remove the heat 
 of compression. The compressed air then passes 
 through the concentric tube-system H t which is 
 placed in a large, well-insulated chamber, to the 
 expansion valve r, where it is allowed to escape 
 into the receiver G. The sudden expansion of the 
 air produces a considerable lowering of tempera- 
 ture. The cold air rushes out of the receiver, and 
 passes up through the outer tube, thus lowering 
 the temperature of the compressed air in the inner 
 tube. As the process continues the temperature 
 gradually falls until the air begins to liquefy in the 
 receiver G. This takes place at the temperature 
 of boiling air ; consequently, the liquid air is ob- 
 tained in a static condition at the ordinary atmos- 
 pheric pressure. 
 
 In this way Linde obtained liquid air in consider- 
 able quantity. The resulting liquid was very rich in 
 oxygen, inasmuch as tfte boiling point of oxygen is 
 
 1 Engineer, 82, p. 4861 
 
EXPERIMENTS OF LINDE 
 
 189 
 
 higher than that of nitrogen. The author modified 
 the apparatus somewhat in order to obtain a more 
 complete separation of the oxygen and nitrogen. 1 
 The apparatus employed at present 2 differs 
 
 1 Engineer, 82, p. 509. The efficiency and theory of the appa- 
 ratus are also considered by Linde in this article. 
 
 2 Zeit. f. Eleklrochem., 4, p. 4, 1897 5 a ^ so Zeit. Compr. und 
 fttiss. Case, I, p. 117, 1897. 
 
190 
 
 LIQUEFACTION OF GASES 
 
 somewhat from that just described. The general 
 arrangement of this apparatus is shown in figure 
 36. The tube-system in this case consists of three 
 
 FIG. 36. 
 
 concentric copper tubes, arranged in the form of 
 a spiral. The coils are placed in a carefully in- 
 sulated chamber. By means of the compressor 
 d, the air is forced into the innermost tube of the 
 
EXPERIMENTS OF LINDE 191 
 
 system under a pressure of about 200 atmos- 
 pheres. The heat of compression is removed by 
 means of the water-cooler G. The small tube is 
 provided with an expansion valve at a, where the 
 air is allowed to expand into the space between 
 innermost and middle tubes. In this operation 
 the air passes from a pressure of 200 atmospheres 
 to a pressure of 16 atmospheres. The tempera- 
 ture, of course, is greatly reduced by the sudden 
 expansion of the air. The cold air passes up 
 through the middle tube, and finally back to the 
 compressor d, where it is again compressed to 
 200 atmospheres. This process can be seen from 
 the directions of the arrows. The compressor e 
 maintains a pressure of 16 atmospheres in the 
 middle tube. As the process continues, the tem- 
 perature gradually becomes lower. The middle 
 tube is provided with an expansion valve at b, 
 similar to that at a, through which the air is al- 
 lowed to expand a second time. In this case, the 
 air passes from a pressure of 16 atmospheres into 
 the space between the outer and middle tubes, 
 where the pressure is one atmosphere. This 
 operation lowers the temperature still further. 
 The cold air passes up through the outer tube of 
 the system, abstracting heat from the compressed 
 air, and finally escapes through the outlet at the 
 top of the chamber. The apparatus is so adjusted 
 
192 LIQUEFACTION OF GASES 
 
 that the compressor e supplies sufficient air to 
 exactly replace the small quantity which is lost. 
 The cycle is now complete, and the temperature 
 of the system gradually decreases. When the pro- 
 cess has continued for a period of from \\ to 2 
 hours, liquid air begins to collect in the bottom 
 of the chamber, and passes into the double-walled 
 Dewar bulb c. The liquid can be drawn from this 
 vessel by opening the stop-cock //. About one 
 litre of liquid air can be produced per hour, with 
 a three horse-power engine. With a larger ap- 
 paratus, and a correspondingly increased power, 
 the liquid, of course, could be obtained in much 
 larger quantities. Numerous observations have 
 also been made by this experimenter on the prop- 
 erties of matter at low temperatures. 
 
 Linde has in the process of construction, for 
 the Rhenania Chemical works at Aix-la-Chapelle, 
 a liquid air plant with a capacity of about 50 litres 
 of liquid air per hour, and -an expenditure of about 
 1 20 horse-power. 1 
 
 HAMPSON'S APPARATUS FOR THE LIQUEFACTION 
 OF GASES 
 
 The apparatus employed by Hampson in the 
 liquefaction of oxygen, air, etc., is based upon the 
 
 1 Ewing, Engineering, 65, p. 310, 1898. 
 
EXPERIMENTS OF HAMPSON 
 
 193 
 
 same principle as 
 that employed by 
 Linde, which has just 
 been described. Con- 
 siderable discussion 
 has arisen concern- 
 ing this form of ap- 
 paratus as to priority 
 of invention. Hamp- 
 son's patent is dated 
 May 23, 1895. With 
 this statement as to 
 the date of the inven- 
 tion, we may leave 
 the question of pri- 
 ority and proceed at 
 once with a descrip- 
 tion of the apparatus. 
 A longitudinal sec- 
 tion of the apparatus 1 
 is represented in 
 figure 37. The gas 
 to be liquefied (oxy- 
 gen, air, etc.) is in- 
 troduced through the 
 small tube at the 
 
 1 Engineer, 8l, p. 310, 
 1896. 
 
 o 
 
194 LIQUEFACTION OF GASES 
 
 upper end under a pressure of 120 atmospheres. 
 The gas enters the apparatus at the ordinary tem- 
 perature (the heat of compression having been 
 removed). The tube through which the gas is 
 introduced extends downward in the form of a 
 spiral around a central column. At the lower 
 end of the spiral is an expansion valve which 
 opens from above by means of a screw. The 
 expanded gas passes up through the apparatus 
 and out again at the upper end. The outer por- 
 tions of the apparatus are carefully packed with 
 felt. 
 
 The details of the spiral tube and expansion 
 valve are shown in figure 38. The coil A ter- 
 minates in the jet-piece D, which delivers the gas 
 against a flat plug on the screw C. A slightly 
 different form of valve is shown in the small 
 figure to the left. The gas on escaping through 
 this valve expands to a pressure of one atmos- 
 phere. The temperature of the gas is consider- 
 ably reduced in consequence of this expansion. 
 The cold gas passes up around the coil as shown 
 by the arrows in the figure, thus lowering the 
 temperature of the compressed gas. This makes 
 the process self-intensifying. As the operation 
 continues the temperature gradually falls, until 
 finally the gas begins to liquefy. Liquid oxygen 
 and liquid air can be obtained in this way in a 
 
EXPERIMENTS OF HAMPSON 
 
 195 
 
 static condition at the ordinary atmospheric press- 
 ure in a comparatively short time. The apparatus 
 was constructed on 
 rather a small scale, and 
 had a capacity of about 
 two cubic centimetres 
 of liquid oxygen per 
 minute. 
 
 During the next year 
 the apparatus was some- 
 what modified. 1 The air 
 in this case was intro- 
 duced under a pressure 
 of 87 atmospheres. The 
 jet of liquid air could 
 be seen within twenty- 
 five minutes, and the 
 liquid air began to col- 
 lect in the receiver 
 within thirty-three min- 
 utes from the beginning 
 of the operation. A 
 vacuum vessel was used FIG. 38. 
 
 as the receiver. 
 
 1 Engineer ; 83, p. 294. 
 
196 LIQUEFACTION OF GASES 
 
 APPARATUS EMPLOYED BY DEWAR IN THE 
 LIQUEFACTION OF GASES 
 
 On the 1 9th of December, 1895, Dewar read a 
 paper before the English Chemical Society on the 
 " Liquefaction of Air and Researches at Low 
 Temperature." 1 In this paper the author de- 
 scribes an apparatus in which the self-intensifica- 
 tion method of refrigeration is employed. The 
 compressed gas is cooled to about 80 by means 
 of liquid carbonic acid, and then allowed to ex- 
 pand under a regenerative coil. A section of the 
 apparatus is shown in figure 39. 
 
 The carbonic acid is introduced into the appara- 
 tus at B, and passes through the coiled tube re- 
 presented by the shaded circles. C is a carbonic 
 acid valve, and H the carbonic acid outlet. The 
 air or oxygen to be liquefied enters the apparatus 
 at A, under a pressure of 150 atmospheres. D is 
 a regenerative coil, and F an expansion valve 
 where the compressed gas is allowed to expand 
 into the vacuum vessel G. 
 
 When the temperature of the system is lowered to 
 about 80, the gas is allowed to expand through 
 the valve F. This produces a further decrease 
 in the temperature. The cooled gas passes up 
 around the regenerative coil and lowers the tem- 
 
 1 Proc. Roy. Inst., 15, p. 133. 
 
EXPERIMENTS OF DEWAR 
 
 197 
 
 FIG. 39. 
 
198 LIQUEFACTION OF GASES 
 
 perature of the compressed gas. The temperature 
 of the system falls rapidly, and within fifteen 
 minutes from the time of starting, liquid air (or 
 whatever the substance may be) begins to drop 
 into the vacuum vessel G. By means of this 
 apparatus Dewar was able to liquefy oxygen, air, 
 etc., on a comparatively large scale. 
 
 Dewar also made a number of experiments on 
 " gas jets containing liquid." In these observa- 
 tions a small regenerative coil was placed in a long 
 vacuum vessel. Three different forms of these 
 vessels were constructed. The compressed gas 
 was allowed to expand from the coil near the 
 lower end of the vacuum tube. Oxygen under a 
 pressure of 100 atmospheres, and previously cooled 
 to 79, produced a visible jet of liquid. Dewar 
 suggests this method as a very rapid means of 
 obtaining low temperatures. 
 
 Similar experiments were made with hydrogen. 
 When cooled to 200 and allowed to expand from 
 a pressure of 140 atmospheres in one of these 
 tubes, no liquid jet could be seen. If the gas con- 
 tained a few per cent of oxygen the liquid jet be- 
 came visible. By allowing the pure hydrogen to 
 expand from the pressure of 200 atmospheres over 
 a longer regenerative coil, which has been previously 
 cooled to 200, the liquid jet becomes visible. 
 The liquid hydrogen, however, could not be ob- 
 
EXPERIMENTS OF TRIPLER 199 
 
 tained in a static condition. To obtain some idea 
 of the temperature of the liquid jet, the author 
 placed some liquid air and liquid oxygen in the 
 lower end of the tube. Within a few minutes 
 about fifty cubic centimetres of these liquids were 
 transformed into hard, white solids, resembling 
 avalanche snow. The solid oxygen had a pale 
 bluish color, and showed by reflection all the 
 absorption bands of the liquid. 
 
 TRIPLER'S APPARATUS FOR THE LIQUEFACTION 
 OF AIR 
 
 The method employed by Tripler in the lique- 
 faction of air is based upon the same principle as 
 that employed by Linde and Hampson, but is 
 operated on a much larger scale. Unfortunately 
 no thoroughly scientific account of this apparatus 
 has yet been published. The temperature is low- 
 ered by expanding highly compressed air in a 
 long tube under a regenerative coil. The general 
 plan of the apparatus is shown in figure 4O. 1 
 
 The steam boiler b supplies steam to the com- 
 pressor c under a pressure of about 85 pounds per 
 square inch. The compressor is provided with 
 three air cylinders arranged in tandem on the 
 same rod. These cylinders are cooled by means 
 
 1 Engineering A'eivs, 39, p. 246, 1898. 
 
200 
 
 LIQUEFACTION OF GASES 
 
EXPERIMENTS OF TRIPLER 201 
 
 of water-jackets. The first or low-pressure cylin- 
 der is 10^ inches in diameter, and compresses the 
 air to a pressure of about 4 atmospheres. The 
 second or intermediate cylinder is 6J inches in 
 diameter, and increases the pressure to about 25 
 atmospheres. The third or high-pressure cylinder 
 is 2| inches in diameter, and delivers the air at 
 a pressure of about 150 atmospheres. The air 
 is taken into the compressor from the outside 
 through the pipe a, which is provided with a dust- 
 separator at d. The heat of compression is re- 
 moved by means of the water in the tank/. Af- 
 ter leaving the cooler the compressed air passes 
 through the separator s, which removes the last 
 traces of moisture. 
 
 The air is now ready to be liquefied. This is 
 accomplished by means of the two liquefiers m 
 and 11. These liquefiers consist, in each case, of a 
 long coil of copper tubing with an expansion valve 
 at the lower end. The coils are carefully pro- 
 tected from the external heat by means of felt. 
 The air enters these liquefiers under a pressure 
 of 1 50 atmospheres, and is allowed to expand to a 
 pressure of one atmosphere. The temperature is 
 considerably reduced by this expansion ; the cold 
 air passes up around the coil and lowers the tem- 
 perature of the compressed air, and thus the influ- 
 ence becomes accumulative. Within about twenty 
 
202 LIQUEFACTION OF GASES 
 
 minutes the air begins to liquefy in the lower 
 end of the tube. The liquid air is removed by 
 means of the valves at the lower ends of the 
 liquefiers. 
 
 As operated at present this apparatus has a 
 capacity of from three to four gallons of liquid air 
 per hour. Tripler has obtained liquid air on a 
 much larger scale than have any of the other 
 experimenters. It is not an uncommon occur- 
 rence to ship liquid air from this plant (in ten- 
 gallon cans) to a distance of several hundred 
 miles. The liquid has been successfully trans- 
 ported from New York to Philadelphia, Wash- 
 ington, Boston, and other cities. This plant has 
 made it possible to experiment with liquid air in 
 both technical and scientific lines at a compara- 
 tively small expense. 
 
 THEORY OF THE SELF-INTENSIFICATION METHOD 
 OF REFRIGERATION 
 
 An elaborate discussion of this problem necessi- 
 tates the use of complicated mathematical formulae, 
 and would be out of place in a work of this nature. 
 For that reason the subject will be considered 
 only in a general way. 1 It has been known for 
 
 1 For a more complete discussion, see Joule and Thomson, 
 Trans. Roy. Soc., 144, p. 321, or some work on thermodynamics. 
 
THEORY OF REGENERATIVE METHOD 203 
 
 more than a century that when a gas is compressed, 
 the temperature rises, and, conversely, when the 
 compressed gas is allowed to expand, the tempera- 
 ture falls. In 1842 Mayer investigated the cause 
 of these thermal changes. "Whence comes the 
 heat," he asks, " generated during the compression 
 of a gas, and what becomes of the heat which van- 
 ishes when the gas expands ? " He considered the 
 problem in the light of the conservation of energy. 
 The work which is done in the compression of a 
 gas, he says, is changed into heat, and the work 
 which is done by the expanding gas, against the 
 external pressure, cannot spring into existence 
 from nothing, but comes from the heat which the 
 gas loses. Mayer's conception implies that heat 
 is a form of energy. He fully recognized the fact 
 that heat can be obtained from or transformed 
 into other forms of energy. 
 
 In order to make this subject clear, it will be 
 necessary to consider the two specific heats of 
 gases. By the specific heat of a substance is 
 meant the capacity of unit mass of the substance 
 for heat ; i.e. the ratio of the amount of heat sup- 
 plied to the body to the rise in temperature. 
 When a gas is heated, either the pressure or 
 volume is increased. If the pressure of a gas 
 is kept constant during a specific heat determina- 
 tion, the value obtained is greater than that ob- 
 
204 LIQUEFACTION OF GASES 
 
 tained at constant volume. In the former case the 
 gas expands against a definite pressure ; i.e. it per- 
 forms a certain amount of mechanical work. The 
 excess of heat required to raise the temperature of 
 a gas at constant pressure over that required at 
 constant volume is simply the amount of energy 
 required in the expansion of the gas. 
 
 According to this theory there should be no 
 temperature change when a gas is allowed to ex- 
 pand into a vacuum, as there is no external work 
 to overcome. The experiments of Gay Lussac 
 and those of later experimenters have shown this 
 to be true. 1 Moreover, if the theory of Mayer is 
 true, the difference between the quantities of heat 
 necessary to raise the temperature of one gram 
 of air through one degree under the two condi- 
 tions; i.e. at constant pressure and constant vol- 
 ume, should correspond to the work performed in 
 expanding one gram of air ^\^ of its volume at 
 o. Both of these values have been determined. 
 The first is 0.0692 calories of heat, and the second 
 2923.5 gram centimetres of work. If, then, 0.0692 
 calories of heat correspond to 2923.5 gram centi- 
 
 1 Strictly speaking, this statement is true only in the case of a 
 perfect gas. Although there is no external work to overcome when 
 ordinary gases expand into a vacuum, a small quantity of energy 
 is required to overcome the molecular attraction. This causes a 
 slight decrease in the temperature. 
 
THEORY OF REGENERATIVE METHOD 205 
 metres of work, one calorie of heat corresponds to 
 
 , = 42245 gram centimetres of work. This 
 0.0692 
 
 latter value, being equivalent to one calorie, is 
 called the mechanical or dynamical equivalent of 
 heat, and agrees very closely with the values 
 obtained by different methods by Joule, Rowlands, 
 and others. We may safely assume, then, that 
 the heat which is absorbed when a compressed 
 gas is allowed to expand is equivalent to the 
 external work performed. 
 
 In order to calculate the decrease in tempera- 
 ture which accompanies the expansion of a gas, it 
 is necessary to know the specific heat of the gas 
 at constant pressure and constant volume ; also 
 the initial and final pressures of the gas. 
 
 Suppose a small quantity of heat, dQ, be com- 
 municated to a gas at constant volume, and let the 
 rise in temperature be represented by dT. Then, 
 as no external work has been done, 
 
 where C v represents the specific heat of the gas at 
 constant volume. Equation (i) may be written 
 
 If the pressure p had remained constant when the 
 heat was communicated, there would have been a 
 
206 LIQUEFACTION OF GASES 
 
 slight increase in volume which we may represent 
 by dv. Then a certain amount of external work 
 (pdv) would have been performed. If this value 
 is represented in thermal units, the available heat 
 for increasing the temperature of the gas is dQ 
 pdv, and the above equation becomes 
 
 dQ-pdv=C v dT (2) 
 
 dQ r i Pdv 
 
 or 3^ =6 +^F* 
 
 dT dT 
 
 The term -^ in this case represents the specific 
 dl 
 
 heat of the gas at constant pressure, and hence 
 
 (3) 
 
 where C p is the specific heat at constant pressure. 
 By differentiating the equation pv R T with 
 respect to v, we obtain 
 
 pdv = RdT, 
 
 or ; - 
 
 hence C p = C v + R. 
 
 If the equation pv = RT be differentiated with 
 respect to all of the variables, we have 
 
 pdv + vdp = RdT. 
 
THEORY OF REGENERATIVE METHOD 207 
 
 Substituting for R its equivalent C p C v , this 
 equation becomes 
 
 pdv + vdp 
 
 By substituting this value of dT in 2, we obtain 
 dQ = ^pdv + ^vdp. (4) 
 
 If the gas is allowed to expand adiabatically, 1 
 dQ = o, and hence, from 4, we have 
 
 /& + # = <>. 
 
 Dividing this equation by -^, we obtain 
 
 The last equation may be written 
 
 where k represents the ratio of the specific heat of 
 the gas at constant pressure to that at constant 
 volume. If the gas be allowed to assume two 
 definite conditions with respect to pressure and 
 volume, p, v, and p lt v lt and we integrate equa- 
 
 J A gas is said to expand adiabatically when no heat is com- 
 municated from or given out to the external surroundings. 
 
208 LIQUEFACTION OF GASES 
 
 tion (5) between these limits, and transpose the 
 result, we obtain 
 
 log/ - log/, = k (log t>, - log z>), 
 
 i 
 
 i& 
 
 "(*) 
 
 / \ 
 
 10 ^) 
 
 Simplifying this, we have 
 
 (6) 
 
 Suppose now that a gas be allowed to expand 
 from the pressure / to the pressure / r Before 
 and after the expansion the gas must fulfill the 
 equations pv RT, and p^\ RT^ respectively. 
 Dividing the first of these by the second, we have 
 
 pv _ T 
 P\ v \ T i 
 
 If we substitute for its value from equa- 
 
 l 
 tion (6), the equation becomes 
 
 Knowing, then, the initial and final pressures, 
 and also the initial temperature, the final tempera- 
 ture can be calculated from equation (7). 
 
 In the case of air, k= 1.41. Substituting this 
 
THEORY OF REGENERATIVE METHOD 209 
 
 value, and assuming the initial temperature to be 
 zero centigrade and the final pressure to be one 
 atmosphere, the theoretical temperatures obtained 
 by the adiabatic expansion of air are given in the 
 following table : l 
 
 Initial Pressures 
 
 Final Temperatures 
 
 TOO atmospheres 
 
 - 20I.5 
 
 200 " 
 
 -214 .5 
 
 300 
 
 221 .0 
 
 400 " 
 
 -225 .1 
 
 500 " 
 
 - 228 .2 
 
 Applying these results to the liquefaction of 
 gases by means of the regenerative coil, it is evi- 
 dent that the expansion of the gas in the tube 
 lowers the temperature by an amount which corre- 
 sponds to equation (7). Further, the issuing jet 
 experiences a much greater decrease in tempera- 
 ture owing to the greater difference between the 
 initial and final pressures. Finally, the expanded 
 gas of very low temperature passes up around 
 the coil and lowers the temperature of the com- 
 pressed gas ; i.e. the initial temperature of the gas, 
 before expansion, gradually decreases. In this 
 way the influence becomes accumulative, and the 
 
 1 Ostwald's Outlines of General Chemistry t p. 83. 
 P 
 
210 LIQUEFACTION OF GASES 
 
 process has been termed the regenerative or self- 
 intensifying method. 
 
 In practice the efficiency of such an apparatus 
 is never equal to the theoretical. In the first 
 place, there is always an inflow of heat from the 
 outside. Careful insulation may reduce, but can 
 never entirely eliminate, this influence. Further- 
 more, the interchange of heat between the ex- 
 panded gas and the counter-current apparatus is 
 never complete. 1 
 
 Rayleigh suggests that the efficiency of the va- 
 rious forms of apparatus employed at present in 
 the liquefaction of air might be considerably in- 
 creased by the use of a turbine. 2 He says : " It 
 must not be overlooked, that to allow the work 
 of expansion to appear as heat at the very place 
 where the utmost cooling is desired, is very bad 
 thermodynamics. The work of expansion should 
 not be dissipated within, but be conducted to the 
 exterior. ... A turbine of some sort might be 
 used. This would occupy little space, and even if 
 of low efficiency, would still allow a considerable 
 fraction of the work of expansion to be conveyed 
 away. The worst turbine would be better than 
 none, and would probably allow the pressures to 
 
 1 For details in regard to the efficiency of the apparatus, see 
 Linde, Engineer, 82, pp. 485 and 509. 
 
 2 Nature, 58, p. 199, 1898. 
 
THEORY OF REGENERATIVE METHOD 211 
 
 be reduced. It should be understood that the ob- 
 ject is not so much to save the work, as to obviate 
 the very prejudicial heating arising from its dis- 
 sipation in the coldest part of the apparatus. It 
 seems to me that the future may bring great 
 developments in this direction, and that it may 
 thus be possible to liquefy even hydrogen at one 
 operation." 
 
 Since the regenerative principle has proved 
 successful in the liquefaction of air, it has been 
 suggested by some inventors that the immense 
 quantity of heat which is stored up in the earth's 
 atmosphere may be obtained as available energy 
 for doing work at a very small expense. Their 
 idea is that there is considerably more energy 
 stored up in the liquid air than is required to 
 produce the liquid by the regenerative method. 
 Without going into the theory of this suggestion, 
 we may dispose of it by quoting the following pas- 
 sage from Nernst : 
 
 " In the judgment of some inventors who are 
 completely permeated with the accuracy of the 
 law of the conservation of energy, it is by no 
 means regarded as impossible to construct a ma- 
 chine which should be able to furnish work as 
 desired and free of cost. External work and heat 
 are equivalent to each other. Moreover, energy 
 in the form of heat is in abundance, so that it only 
 
212 LIQUEFACTION OF GASES 
 
 needs an apparatus in which one shall apply it in 
 driving our machine, to use up the energy of its 
 environments. Such an apparatus, for example, 
 might be sunk into a great water reservoir, whose 
 enormous quantity of energy could be changed 
 into useful work ; it would, for example, make the 
 steam engines of our ocean steamers unnecessary, 
 and would keep the screw of a ship in motion as 
 long as desired, and at the cost of the immeasur- 
 able store of heat in the sea. Such an apparatus 
 would be in certain respects a perpetual motion, 
 and yet not contradictory to the first law of ther- 
 modynamics, since it would extract the heat of its 
 environments and give it back again as external 
 work which, as a result of the friction of the screw, 
 would change itself back again into heat, to enter 
 the cycle anew. 
 
 " Unfortunately such an apparatus, which would 
 make coal worthless as a source of energy, ap- 
 pears to be a chimera, exactly as was the per- 
 petual motion of the inventor of the last century ; 
 at least, many fruitless attempts have made this 
 more than probable. Thus, as we sum up the 
 numerous abortive endeavors, we come, in a way 
 analogous to that which led to the knowledge of 
 the conservation of energy, to the proposition that 
 an apparatus which could continually change the 
 heat of its environments into external work is a 
 
LIQUEFACTION OF ARGON 213 
 
 contradiction to a law of nature, and therefore an 
 impossibility. Although by recognizing this law 
 the human spirit of invention may be poorer by 
 one problem, yet natural investigation is com- 
 pensated for it by a principle of almost unlimited 
 application." 
 
 SECTION IV 
 
 LIQUEFACTION OF ARGON, HYDROGEN, HELIUM, 
 ETC. 
 
 There remain to be considered some special 
 observations in the liquefaction of gases. The 
 quantity of gas to be liquified, in some cases, was 
 very small, and hence an apparatus of special 
 construction was necessary. 
 
 Liquefaction and Solidification of Argon 
 
 In 1895 Olszewski 1 subjected argon to low tem- 
 peratures and high pressures. A sample of the 
 pure dry gas, about 300 cc., was obtained from 
 Ramsay. Four series of experiments were made ; 
 two with the object of determining the critical 
 temperature and pressure, and two for the pur- 
 pose of determining the boiling point and freezing 
 point under atmospheric pressure. 
 
 1 Trans. Roy. Soc., 186, p. 253, 1895. 
 
214 LIQUEFACTION OF GASES 
 
 In the first two experiments use was made of 
 an apparatus similar to that of Cailletet. Liquid 
 ethylene, boiling under reduced pressure, was used 
 as the refrigerant. At a temperature of 128. 6 
 and a pressure of 38 atmospheres the argon con- 
 densed to a colorless liquid. On slowly raising 
 the temperature the meniscus became less distinct 
 and finally disappeared. This was repeated sev- 
 eral times with a view of determining the critical 
 temperature and pressure. The mean of seven 
 observations gave 121 for the critical tem- 
 perature, and 50.6 atmospheres for the critical 
 pressure. The vapor pressure of argon at a 
 temperature of 139.! was found to be 23.7 
 atmospheres. 
 
 The experiments on the boiling and freezing 
 points were carried out by means of the apparatus 
 represented in figure 41. The argon was con- 
 tained in the glass burette b, closed at both ends 
 with glass stop-cocks. The lower end of the 
 burette was connected, by means of a flexible 
 tube, with the mercury reservoir a. By means of 
 the mercury, the gas could be transferred to the 
 liquefying tube d, which was immersed in the 
 liquid oxygen contained in the quadruple-walled 
 glass tube e. The tube i was connected with a 
 large air-pump, and the tube c with a mercury 
 air-pump. 
 
LIQUEFACTION OF ARGON 
 
 215 
 
 When the temperature of the tube d had be- 
 come equal to that of liquid oxygen boiling under 
 the ordinary atmospheric pressure, the argon was 
 
 FIG. 41. 
 
216 LIQUEFACTION OF GASES 
 
 admitted, but showed no signs of liquefaction. 
 This showed that the boiling point of argon is 
 lower than that of oxygen. The pressure of the 
 argon was then adjusted so as to remain equal to 
 that of the atmosphere. The tube containing the 
 liquid oxygen was slightly exhausted, and at a 
 temperature of 187 the argon began to liquefy. 
 As a mean of four experiments Olszewski gives 
 187 for the boiling point of argon. From the 
 volume of gas used and the volume of liquid ob- 
 tained, the density of liquid argon at the boiling 
 point was estimated to be about 1.5. 
 
 The temperature of the liquid oxygen was then 
 lowered by slow exhaustion to 191, when the 
 argon solidified to a crystalline mass resembling 
 ice. At lower temperatures the mass became 
 opaque. The substance was frozen and melted 
 four times. The mean value obtained for the 
 melting point by means of a hydrogen thermome- 
 ter is - i89.6. 
 
 Experiments ^vith Helium 
 
 In 1896 Olszewski 1 made an extensive series 
 of experiments with a view of liquefying helium. 
 The pressure was obtained by means of a Caille- 
 tet apparatus. The liquefying tube of this appa- 
 
 1 Wied. Ann., 59, p. 184 ; Nature, 54, p. 377. 
 
EXPERIMENTS WITH HELIUM 217 
 
 ratus was thoroughly exhausted by means of a 
 mercury pump, and then carefully filled with dry 
 helium which had been obtained from Ramsay. 
 In the first series of experiments liquid oxygen 
 was used as the refrigerant. By the evaporation 
 of this liquid under a pressure of 10 mm. of mer- 
 cury a temperature of 210 could be obtained. 
 At this temperature and under a pressure of 125 
 atmospheres helium showed no signs of liquefac- 
 tion. The gas was then allowed to expand until 
 the pressure had decreased to twenty atmos- 
 pheres, and in some cases to one atmosphere, but 
 there was no evidence of condensation to the 
 liquid state. 
 
 In the second series of experiments liquid air 
 was employed as the refrigerant. By the evapora- 
 tion of liquid air under a pressure of 10 mm. 
 of mercury the temperature was reduced to about 
 220. Even at this low temperature and with 
 a pressure of 140 atmospheres, the results were 
 all negative. Under these conditions the gas 
 was again allowed to expand, but there were no 
 indications of liquefaction. The author says : " In 
 every single instance I have obtained negative 
 results, and, as far as my experiments go, helium 
 remains a permanent gas, and apparently much 
 more difficult to liquefy than even hydrogen." 
 The experiments were carried out on a very small 
 
2i8 LIQUEFACTION OF GASES 
 
 scale, owing to the small quantity of gas at hand 
 (about 140 cc.). 
 
 The temperatures were not measured directly 
 in these experiments, but were calculated by 
 Olszewski from the Laplace-Poisson equation for 
 the change of temperature in a gas during adi- 
 abatic expansion. According to this equation the 
 temperature of the gas when allowed to expand, 
 as previously described, to a pressure of one 
 atmosphere, is about - 264. It seems more 
 probable, however, in the light of recent investi- 
 gations, that the temperature was somewhat above 
 this point. The author also constructed a helium 
 thermometer and compared it with a hydrogen 
 thermometer from the temperature of 182 to 
 -210. The results obtained by the two ther- 
 mometers agreed very closely. 
 
 Liquefaction of Fluorine 
 
 In 1895 Dewar remarked that " fluorine is the 
 only widely distributed element in nature which 
 has not been liquefied." Two years later, how- 
 ever, this gas was condensed to the liquid state by 
 Moissan and Dewar. 1 The latter experimenter had 
 previously subjected fluorine to low temperatures 
 with a view of liquefaction, but without success. 
 
 1 Compl. rend., 124, p. 1202 ; and 125, p. 505, 1897. 
 
LIQUEFACTION OF FLUORINE 219 
 
 The fluorine used by Moissan and Dewar was 
 prepared by the electrolysis of potassium fluoride 
 dissolved in anhydrous hydrofluoric acid. The 
 acid fumes were removed by conducting the gas 
 through a platinum coil, which was surrounded 
 by a mixture of solid carbonic acid and alcohol ; 
 after which the fluorine was conducted through 
 platinum tubes filled with perfectly dry sodium 
 fluoride. 
 
 The apparatus 1 employed in the liquefaction is 
 represented in figure 42. The glass bulb E is 
 fused to the platinum tube A, which surrounds a 
 smaller platinum tube D. Each of these tubes is 
 provided with a screw-cock so that the communica- 
 tion with the outer air or the current of fluorine 
 can be interrupted at pleasure. In the first series 
 of experiments the glass bulb E was immersed in 
 liquid oxygen, which was contained in a cylindrical 
 vacuum vessel. At the temperature of boiling air 
 (about 183) the fluorine passed through the 
 apparatus without showing any signs of liquefac- 
 tion. By evaporating the liquid oxygen under re- 
 duced pressure, liquid fluorine soon began to collect 
 in the apparatus. The outlet to the fluorine tube 
 was then closed, and the glass bulb soon became 
 filled with a clear yellow, extremely mobile liquid. 
 
 1 Chem. News, 76, p. 261. 
 
220 
 
 LIQUEFACTION OF GASES 
 
 FIG. 42. 
 
 At this temperature fluorine did 
 not attack the glass bulb. 
 
 The experiment was repeated, 
 using freshly prepared 
 liquid air as the refrig- 
 erant. At a tempera- 
 ture of - 190 the 
 fluorine condensed to 
 the liquid state. From the two 
 series of experiments the authors 
 calculated the boiling point to 
 be about 187, which is the 
 same temperature as that ob- 
 tained by Olszewski for the 
 boiling point of argon. The 
 critical temperature was esti- 
 mated to be about 120, and 
 the critical pressure about 40 
 atmospheres. The chemical 
 activity of fluorine was greatly 
 reduced at the temperature of 
 boiling oxygen. At this tem- 
 perature it does not replace 
 iodine, and is without action on 
 phosphorus, boron, silicon, iron, 
 etc. With hydrogen, turpentine, 
 and benzene it reacts with incan- 
 descence. 
 
LIQUEFACTION OF HYDROGEN 221 
 
 The density of liquid fluorine was determined 
 by placing solid substances of different specific 
 gravities in the liquid. The result of a number of 
 observations gave a density of 1.14. The liquid 
 is soluble in all proportions in liquid oxygen and 
 liquid air. 
 
 The authors endeavored to reduce the tempera- 
 ture to the freezing point of fluorine. When the 
 glass bulb was filled to about three-fourths of 
 its capacity, the valves were closed. The liquid 
 oxygen surrounding the glass bulb was then 
 evaporated under a very low pressure, and the 
 temperature sank to 210 ; but even at this low 
 temperature the fluorine retained its characteristic 
 mobility. In a second experiment the liquid flu- 
 orine was introduced into a glass tube, which was 
 afterward sealed and kept for some time at a tem- 
 perature of - 210, but there were no indications 
 of solidification. 
 
 Liquefaction of Hydrogen 
 
 Of the various gases known, hydrogen has pre- 
 'sented the most difficult problem to the experi- 
 menters on the liquefaction of gases. During the 
 last two decades numerous attempts have been 
 made to liquefy this gas. Reference has already 
 been made to the experiments of Cailletet and 
 Pictet, in which hydrogen was probably obtained 
 
222 LIQUEFACTION OF GASES 
 
 in the form of a very fine mist (pp. 1 19, 135). Wro- 
 blewski and Olszewski modified these experiments 
 somewhat by cooling the gas with liquid oxygen, 
 and then allowing the gas to expand (pp. 143, 164). 
 In these observations there can be no doubt as to 
 the formation of a mist of liquid hydrogen. Both 
 of these experimenters endeavored to determine 
 the critical constants and boiling point of this gas. 
 In 1894 Dewar 1 attacked the problem from a 
 different standpoint. Realizing that, with liquid 
 oxygen and liquid air as refrigerants, the tempera- 
 ture could not be sufficiently reduced for the lique- 
 faction of hydrogen, he endeavored to obtain a 
 liquid, the critical temperature and boiling point 
 of which are considerably lower than those of air 
 and oxygen. This he thought could be accom- 
 plished by liquefying a mixture of hydrogen and 
 nitrogen. In regard to the efficiency of this 
 method the author says : " One thing can, how- 
 ever, be proven by the use of the gaseous mixture 
 of hydrogen and nitrogen ; viz., that by subjecting 
 it to a high compression at a temperature of 
 - 200, and expanding the resulting liquid into' 
 the air, a much lower temperature than anything 
 that has yet been recorded up to the present time 
 can be reached. This is shown by the fact that 
 
 1 Chem. Neivs, 70, p. 115. 
 
LIQUEFACTION OF HYDROGEN 223 
 
 such a mixed gas gives, under the conditions, a 
 paste or jelly of solid nitrogen, evidently giving off 
 hydrogen, because the escaping gas burns fiercely. 
 Even when hydrogen containing from two to five 
 per cent of air is similarly treated, the result is a 
 white solid mass (solid air), along with a clear 
 liquid of low density which is so exceedingly vola- 
 tile that no known device for collecting it has 
 been successful." 
 
 During the next year Dewar l extended his obser- 
 vations on the liquefaction of hydrogen (p. 198). 
 He says : " Hydrogen, cooled to - 200, was 
 forced through a fine nozzle under a pressure of 
 140 atmospheres, and yet no liquid jet could be 
 seen. If, however, hydrogen, previously cooled by 
 a bath of boiling air, is allowed to expand from a 
 pressure of 200 atmospheres over a regenerative 
 coil, a liquid jet can be seen after the circulation 
 has continued for a few minutes, along with a 
 liquid which is in rapid rotation in the lower part 
 of the vacuum-vessel. The liquid did not accumu- 
 late, owing to its low specific gravity and the rapid 
 current of gas. These difficulties will doubtless 
 be overcome by the use of a differently shaped 
 vacuum-vessel, and by better isolation." 
 
 In May, 1898, Dewar obtained liquid hydrogen 
 
 1 Proc. Roy. Inst., 15, p. 146. 
 
224 LIQUEFACTION OF GASES 
 
 for the first time in a static condition. 1 The 
 method of procedure was similar to that which has 
 just been described. Hydrogen cooled to a tem- 
 perature of 205, and under a pressure of 180 
 atmospheres, was allowed to escape continuously 
 from the nozzle of a coil of pipe, at the rate of 
 from ten to fifteen cubic feet per minute, into a 
 doubly silvered vacuum-vessel of special construc- 
 tion. The space surrounding this vessel was kept 
 at a temperature below 200. Soon after the 
 operation was begun, liquid hydrogen began to 
 drop from this vacuum-vessel into a second vessel 
 which was doubly isolated, in that it was sur- 
 rounded by a third vacuum-vessel. Within five 
 minutes about twenty cubic centimetres of liquid 
 hydrogen had collected in the second bulb. The 
 experiment was interrupted at this point by the 
 solidification of air in the pipes. The liquid 
 obtained was clear and colorless, and showed a 
 well-defined meniscus. No provision was made in 
 this experiment for determining the boiling point of 
 liquid hydrogen, but the author calls attention to 
 an observation which shows that the temperature 
 of liquid hydrogen under atmospheric pressure is 
 extremely low. A long piece of glass tubing, 
 sealed at one end and open at the other, was 
 
 l Proc. Roy. Soc., 63, p. 256, 1898. 
 
LIQUEFACTION OF HYDROGEN 225 
 
 cooled by immersing the closed end in liquid 
 hydrogen. The portion of the tube immersed in 
 the liquid was immediately filled with solid air. 
 
 A few months later Dewar determined the boil- 
 ing point and density of liquid hydrogen. 1 The 
 boiling point was measured by means of a platinum 
 resistance thermometer, and was found to be about 
 - 238. This is a few degrees higher than the 
 values given by Wroblewski and Olszewski. The 
 author thinks that the critical temperature of 
 hydrogen is about 225, and the critical pressure 
 about fifteen atmospheres. The density of liquid 
 hydrogen was determined by measuring the vol- 
 ume of gas which is given off when ten cubic cen- 
 timetres of the liquid are allowed to evaporate. 
 The result of the experiment gave a density of 
 0.07 for liquid hydrogen, which is only about one- 
 fourteenth that of water. 
 
 During the present year Dewar has made a 
 series of observations on the temperature obtained 
 by evaporating liquid hydrogen under reduced 
 pressure. 2 When the liquid was evaporated under 
 a pressure of 25 millimetres of mercury, there 
 were no indications of solidification or loss of 
 mobility. The temperature according to the plati- 
 num resistance thermometer was 239. i, which 
 
 1 Chem. News, 77, pp. 261 and 282, 1898. 
 vibid.y 79, p. 61, 1899. 
 Q 
 
226 LIQUEFACTION OF GASES 
 
 is only one degree lower than the boiling point 
 The author says that the temperature thus obtained 
 should have been five or ten degrees below the 
 boiling point, and adds that the experiment will be 
 repeated with larger quantities of liquid hydrogen. 
 
 Quite recently Dewar has repeated the experi- 
 ments on the boiling point of hydrogen, 1 and ob- 
 tained a value somewhat lower than that obtained 
 in the previous experiments. The author pre- 
 pared 250 cubic centimetres of liquid hydrogen 
 for these observations. The temperature of hy- 
 drogen, boiling under the atmospheric pressure, 
 was determined by means of a rhodium-platinum 
 resistance thermometer and found to be 246. 
 This value is 8 lower than that obtained by 
 means of the platinum resistance thermometer. 
 
 In an addendum to this paper the author calls 
 attention to some measurements made by means 
 of a hydrogen thermometer under reduced press- 
 ure. This thermometer gave - 182. 5 for the 
 boiling point of oxygen, and 252 for the boil- 
 ing point of hydrogen. If this value is the true 
 boiling point of hydrogen, it is likely that, by evap- 
 orating liquid hydrogen under reduced pressure, 
 the temperature can be lowered to within ten or 
 twelve degrees of the absolute zero. 
 
 1 Chem. News, March 24, 1899, p. 133. 
 
LIQUEFACTION OF HELIUM 227 
 
 Liquefaction of Helium 
 
 After an elaborate series of experiments with a 
 view of liquefying this gas, Olszewski remarked, 
 " As far as my experiments go, helium remains a 
 permanent gas, and apparently is much more diffi- 
 cult to liquefy than hydrogen" (p. 217). After 
 obtaining liquid hydrogen in a static condition, 
 Dewar 1 placed a sealed glass tube containing 
 helium in the liquid hydrogen. A colorless liquid 
 immediately condensed on the sides of the tube. 
 By placing the same tube in liquid air, boiling 
 under reduced pressure, no condensation was ob- 
 served. The author concluded that the boiling 
 point of helium is very close to that of hydrogen. 
 
 Some Recently Discovered Gases 
 
 During the last year Ramsay and Travers have 
 reported three new gases in the atmosphere. The 
 first of these gases, 2 which the authors designated 
 as " krypton," was obtained by evaporating 750 
 cubic centimetres of liquid air, and collecting the 
 gas from the last ten cubic centimetres. After 
 removing the oxygen and nitrogen from this gas, 
 a residue of 26.2 cubic centimetres remained in 
 the vessel. The residual gas had a spectrum dif- 
 
 l Proc. Roy. Soc., 63, p. 257, 1898. 
 *Ibid., p. 405, 1898. 
 
228 LIQUEFACTION OF GASES 
 
 ferent from that of argon, and was considered by 
 the authors as an elementary substance. 
 
 The other two gases, which were called by the 
 authors " the companions of argon," were obtained 
 from liquid argon. 1 The first portions of gas 
 which escape when liquid argon is allowed to 
 evaporate were collected by Ramsay and Travers, 
 and sparked with oxygen gas ; after which the 
 excess of oxygen was removed. The residual gas 
 gave a spectrum different from that of argon and 
 krypton, and was called " neon." 
 
 During the evaporation of the liquid argon a 
 white solid separated. After the liquid had been 
 completely evaporated, the solid residue was vapor- 
 ized, and the gas collected. This gas was found 
 to be different from those just described, and was 
 called "metargon." 
 
 It is evident from the preceding account that 
 these gases, mixed with air, argon, etc., have all 
 been liquefied. The authors state that "while 
 metargon is a solid at the temperature of boiling 
 air, krypton is probably a liquid, and more volatile 
 at that temperature." Liquid neon is somewhat 
 more volatile than liquid argon. 
 
 During this same year Brush published an ac- 
 count of a supposed new gas. 2 The gas was 
 
 1 Proc. Roy. Soc., 63, p. 437, 1898. 
 
 2 Read before the American Association for the Advancement o* 
 Science, Aug. 23, 1898. 
 
TABLE OF PHYSICAL CONSTANTS 229 
 
 obtained by exhausting, to a high degree, pulver- 
 ized soda glass. The resulting gas was found to 
 be a much better conductor of heat than any other 
 known gas. At a pressure of 0.000096 atmos- 
 pheres the conducting capacity was twenty times 
 that of hydrogen. The author says : " Evidently a 
 new gas of enormous heat-conducting capacity was 
 present, mixed-with the last small traces of air." 
 From measurements of the conductivity he con- 
 cluded that the density of the new gas is only 
 about YO^O tnat f hydrogen. Owing to the ex- 
 tremely low density the author suggested the name 
 "etherion" for the new gas. In case subsequent 
 observation l should confirm the theory of Brush, 
 the new gas will furnish an interesting problem to 
 the experimenters on the liquefaction of gases. 
 
 TABLE OF PHYSICAL CONSTANTS 
 
 The following table contains the critical con- 
 stants, boiling points, melting points, etc., of sub- 
 stances which usually occur as gases, and of some 
 of the most common liquids. A few theoretical 
 results are also given for some of the heavier 
 metals. It frequently happens that the results 
 obtained by one experimenter do not agree with 
 those obtained by another. This is especially true 
 
 1 Crookes has already suggested that this gas may be highly 
 exhausted aqueous vapor. Chem. News, 78, p. 221, 1898. 
 
230 
 
 LIQUEFACTION OF GASES 
 
 in regard to the critical constants. The names in 
 the right-hand column are given as authority for 
 the critical constants chosen. 1 
 
 Substance 
 
 Critical 
 Tempera- 
 ture 
 
 Criti- 
 cal 
 Press- 
 ure 
 
 Boiling 
 Point 
 
 Freez- 
 ing 
 Point 
 
 Color of 
 Liquid 
 
 Observer 
 
 Acetone 
 
 + 2 37-5 
 
 60 
 
 + 56.5 
 
 
 Colorless 
 
 Sajotschewski 
 
 Acetylene . . 
 
 + 37-5 
 
 68 
 
 
 .. 
 
 Colorless 
 
 Ansdell 
 
 Air .... 
 
 
 _ 
 
 
 
 
 
 
 
 Alcohol. . . 
 
 +243^.6 
 
 62.7 
 
 + 78-3 
 
 -130 
 
 Colorless 
 
 Ramsay and Young 
 
 Ammonia . 
 
 + 130 
 
 
 33-7 
 
 - 75 
 
 Colorless 
 
 Dewar 
 
 Argon . . . 
 
 121 
 
 50.6 
 
 -187' 
 
 -1890.6 
 
 Colorless 
 
 Olszewski 
 
 Arsine . . . 
 
 
 
 55 
 
 119 
 
 Colorless 
 
 
 Carbon dioxide 
 
 + 31 
 
 75 
 
 
 -65 
 
 Colorless 
 
 Andrews 
 
 Carbon disul- 
 
 
 
 
 
 
 
 phide . . . 
 
 +2710.8 
 
 74-7 
 
 + 46 
 
 -110 
 
 Colorless 
 
 Sajotschewski 
 
 Carbon mon- 
 
 
 
 
 
 
 
 oxide . 
 Chlorine . . 
 
 -14.0 
 
 + 141 
 
 to 
 
 190 
 - 36. 6 
 
 -207 
 
 -102 
 
 Colorless 
 Yellow 
 
 Wroblewski 
 Dewar 
 
 Chloroform 
 
 +260 
 
 54 
 
 + 61 
 
 
 Colorless 
 
 Sajotschewski 
 
 Cyanogen . . 
 Ether . . . 
 
 + 195- 5 
 
 61.7 
 40 
 
 - 21 
 
 + 35 
 
 - 34-4 
 
 Colorless 
 Colorless 
 
 Dewar 
 Ramsay 
 
 Ethylene 
 Fluorine . . 
 
 + 10 D .I 
 I20(?) 
 
 4o(?) 
 
 102.5 
 
 187 
 
 -169 
 
 Colorless 
 Yellow 
 
 Dewar 
 Moissan and Dewar 
 
 Gold. . . . 
 
 +4300 
 
 
 
 + 1035 
 
 
 Calculated by Guldberg 
 
 Helium . . . 
 
 
 
 t f 
 
 
 Colorless 
 
 Dewar 
 
 Hydrochloric 
 
 
 
 
 
 
 
 acid . . . 
 
 +5i.25 
 
 86 
 
 35 
 
 n6 
 
 Colorless 
 
 Ansdell 
 
 Hydrogen . . 
 
 
 
 252 
 
 
 Colorless 
 
 Dewar 
 
 Hydrogen sul- 
 
 
 
 
 
 
 
 phide . . . 
 
 + 100. 2 
 
 92 
 
 6i.8 
 
 - 85 
 
 Colorless 
 
 Dewar 
 
 Iron .... 
 Methane 
 
 + 5200 
 
 95- 5 
 
 5 
 
 164 
 
 +1500 
 
 Colorless 
 
 Calculated by Guldberg 
 Dewar 
 
 Nitric oxide . 
 
 - 93-5 
 
 71.2 
 
 153-6 
 
 -167 
 
 Colorless 
 
 Olszewski 
 
 Nitrogen 
 
 -146 
 
 35 
 
 194-5 
 
 214- 
 
 Colorless 
 
 Olszewski 
 
 Nitrous oxide . 
 
 - 35-4 
 
 75 
 
 - 87-9 
 
 
 Colorless 
 
 Dewar 
 
 Oxygen 
 
 -118 
 
 5 
 
 183 
 
 
 Pale blue 
 
 Wroblewski 
 
 Ozone . . . 
 
 
 
 -125 
 
 
 Indigo blue 
 
 
 Phosphine . 
 
 
 
 85 
 
 T -~O 
 
 Colorless 
 
 
 Platinum . . 
 
 + 7000 
 
 
 
 + 1800 
 
 
 Calculated by Guldberg 
 
 Sulph. dioxide 
 
 
 78-9 
 
 8 
 
 . 
 
 Colorless 
 
 Sajotschewski 
 
 Water . . . 
 
 +358.1 
 
 
 -4-100 
 
 
 
 Colorless 
 
 Nadejdine 
 
 1 For further data and literature on critical constants, see Heil- 
 born, Zeit. Phys. Chem., 7, p. 602, 1891. 
 
CONCLUSION 
 
 I. The Three States of Matter 
 
 THE experiments which have been described 
 show that all gases have been condensed to the 
 liquid state. It has also been shown that, with 
 very few exceptions, all gases have been solidified. 
 The results leave no doubt that all of these sub- 
 stances can exist in the gaseous, liquid, or solid 
 state. At the high temperatures which have 
 been obtained by means of the electric furnace, 
 the densest solids have been liquefied and vola- 
 tilized. 
 
 Andrews says the liquid state of matter forms a 
 link between the solid and gaseous states. This 
 link, however, is frequently suppressed, and the 
 solid passes directly into the gaseous condition. 
 Iodine and arsenic are well-known examples of 
 solids which, at the ordinary pressure, sublime 
 directly to the gaseous state without assuming the 
 intermediate liquid condition. Solid carbonic acid 
 behaves in a similar manner. The melting points 
 of these substances are higher than their boiling 
 points. If iodine crystals are placed in a test-tube 
 
 231 
 
232 LIQUEFACTION OF GASES 
 
 under sulphuric acid, and the temperature gradu- 
 ally raised, the substance melts and does not vapor- 
 ize. Prytz l has shown that at a pressure of five 
 atmospheres solid carbonic acid does not sublime, 
 but passes directly into the liquid state. 
 
 Any substance can exist as a gas at a much 
 lower temperature than that at which it can exist 
 as a liquid. Below the temperature of zero de- 
 grees ice slowly sublimes. In the far northern 
 regions the atmosphere always contains aqueous 
 vapor. Pellat 2 has recently shown that iron sub- 
 limes very slightly at the ordinary temperature 
 and pressure. Under the proper conditions of 
 temperature and pressure all substances can be 
 made to assume the gaseous, liquid, or solid state. 
 The three states of matter are usually defined as 
 follows : 
 
 1. A gas has neither form nor volume, but 
 tends to expand indefinitely. 
 
 2. A liquid has a definite volume, but assumes 
 the form of the vessel in which it is contained. 
 
 3. A solid has a definite form and volume. 
 The relation between the gaseous and liquid 
 
 states has already been discussed (Chapter III). 
 The change from one condition to the other was 
 found to be gradual and imperceptible. There is 
 
 1 Phil. Mag., 39, p. 308, 1895. 
 
 a Zeit.fiir compr. undjluss., Case, 2, p., 95, 1898. 
 
CONCLUSION 233 
 
 no sharp dividing line. The same is true of the 
 solid and liquid states. Their properties, in many 
 cases, are very similar. As the pressure is in- 
 creased, solids tend more and more to assume the 
 form of the vessel in which they are contained. 
 Crystals which have been subjected to enormous 
 pressure in the crust of the earth are found to 
 be distorted into various shapes without being 
 fractured. The crystals seem to flow. The mole- 
 cules of solids are not rigidly fixed in definite 
 positions about which they vibrate, but in many 
 cases move about throughout the entire sub- 
 stance. This has been shown by the experi- 
 ments of Roberts-Austen on the diffusion of 
 metals. 1 He showed that, at a temperature of 
 250, and even at a temperature of 100, gold 
 diffuses throughout the length of solid cylinders 
 of lead. Similar results were obtained with silver 
 and gold at a temperature of 800. The rate 
 of diffusion in such cases is, of course, very low, 
 but the process is similar to the diffusion of gases 
 and liquids. 2 
 
 1 Trans. Roy. Soc., 187, p. 383, 1896. 
 
 2 Heydweiller has endeavored to find some evidence of critical 
 phenomena between the liquid and solid states. The change from 
 the transparent solid to the liquid, he said, appeared to be gradual, 
 and the melting point increased with the pressure. The experi- 
 ments were extended to pressures as high as 3500 atmospheres. 
 Wied. Ann., 64, p. 725, 1898. 
 
234 LIQUEFACTION OF GASES 
 
 2. Industrial Application of Liquefied Gases 
 
 Liquefied gases have been employed for techni- 
 cal purposes in various directions. The carbonic 
 acid industry is well known. The liquid is used 
 in the preparation of aerated waters, and in the 
 manufacture of salicylic acid. Enormous quanti- 
 ties of carbonic acid are liquefied annually by 
 various establishments. Liquid sulphurous acid 
 is now an ordinary product of commerce. When- 
 ever the gaseous product is desired for laboratory 
 use or technical purposes, it is usually obtained 
 from the liquid. At present, about 4,000,000 kilo- 
 grams of this liquid are being prepared annu- 
 ally. Liquid acetylene has been introduced for 
 illuminating purposes. Nitrous oxide is now lique- 
 fied on a large scale, and used as an anaesthetic 
 for minor surgical operations, especially in den- 
 tistry. Liquefied gases are also used in large 
 quantities for the purpose of refrigeration. The 
 ammonia ice-machine is now in operation in most 
 cities. In this process the temperature is lowered 
 and the water frozen by the evaporation of liquid 
 ammonia. Liquid sulphurous acid has also been 
 used for the same purpose. In 1885 Wroblewski 
 said, " Liquid air will be the refrigerant of the 
 future." This prediction, of course, has not yet 
 been fulfilled. 
 
CONCLUSION 235 
 
 Considerable has been said and written about 
 the use of liquefied gases as a motive power. 
 Numerous attempts have been made to introduce 
 engines or motors for this purpose, but no great 
 success has yet crowned these efforts. The ex- 
 tremely low temperatures which result from the 
 expansion of these liquids to gases at the ordinary 
 pressure are very objectionable in the application 
 of the liquids as a motive power. It is not likely 
 that these liquids will prove of any great service 
 where steam-power is practicable (see p. 211). 
 They may prove to be of considerable value, how- 
 ever, in cases where steam-power is impracticable. 
 It has already been suggested that liquid hydro- 
 gen and liquid air may furnish a solution to the 
 balloon problem. 
 
 The industry of liquefied gases is growing rap- 
 idly. Every year sees a wider application of these 
 liquids. A complete discussion of this subject, 
 however, does not fall within the scope of a work 
 of this nature. 1 
 
 3. Physiological Action at Low Temperatures 
 
 Some interesting observations have been made 
 in regard to the influence of very low temperatures 
 on living organisms. In 1870 Cohn 2 made use of 
 
 1 For further discussion of the subject, see references on p. 183. 
 2 Cohn's Beitrdge zur Biologic der Pftanzen, 1870, 2, p. 221. 
 
236 LIQUEFACTION OF GASES 
 
 freezing mixtures, and subjected bacteria, for a 
 period of 12 hours, to temperatures varying from 
 o to 1 8 without destroying their activity. Mel- 
 sens 1 used solid carbonic acid and exposed yeast 
 and vaccine lymph to a temperature of 78 with- 
 out destroying the life of the organisms. Pictet 
 and Yung 2 subjected various bacteria to low tem- 
 peratures. They reduced the temperature by the 
 evaporation of liquid sulphurous and carbonic 
 acids. The organisms were subjected for 20 
 hours to a temperature of 70; for 89 hours to 
 a temperature of 76; and finally for 20 hours 
 to a temperature of 130 (=202 F.). Yeast 
 ferment showed no alterations under the micro- 
 scope, but lost its power of fermentation. Bacillus 
 anthracis and several other micro-organisms re- 
 tained their virulence when injected into living 
 animals. 
 
 In 1885 a ver y elaborate series of experiments 
 were made by Coleman and McKendrick 3 in re- 
 gard to the effect of low temperatures on certain 
 bacteria. Thirty samples of fresh meat were 
 placed in two-ounce white glass phials. The 
 bottles were then carefully closed with corks 
 which had been steeped in mastic varnish, and the 
 
 1 Cotnpt. rend., 70, p. 629 ; and 71, p. 325. 
 
 *Ibid., 98, p. 747, 1884. 
 
 8 Proc. Roy. Inst., n, p. 309. 
 
CONCLUSION 
 
 237 
 
 necks of the corked bottles were immersed in mol- 
 ten sealing wax. The specimens were treated as 
 follows : 
 
 6 samples were exposed to a temperature of - 17 for 65 hours. 
 
 6 
 
 6 
 
 6 
 
 6 
 
 U it 
 
 it it 
 
 a 
 
 -29 
 
 " -34 
 
 -40 
 " -62 
 
 Within ten or twelve hours after removal to a 
 warm room, signs of putrefaction were visible in 
 all of the bottles, and in the course of a few days 
 the putrefactive process was fully established. 
 Other samples were then exposed to a temperature 
 of 83 for a period of 100 hours with similar 
 results. Samples of fresh milk, which had been 
 hermetically sealed and subjected to a temperature 
 of 62 for eight hours, curdled when kept in a 
 warm room. 
 
 The following observations were made by sub- 
 jecting a rabbit to low temperatures : 
 
 
 Temperature 
 
 Pulse 
 
 Respiration 
 
 Before the experiment 
 
 99 .2 F. 
 
 1 60 per min. 
 
 45 per min. 
 
 After 30 min. exposure 
 
 
 
 
 to - 93 . . . . 
 
 94.2 
 
 
 
 
 
 After 60 min. exposure 
 
 
 
 Scarcely 
 
 to 100 . . . 
 
 43.2 
 
 40 per min. 
 
 Perceptible 
 
238 LIQUEFACTION OF GASES 
 
 In this condition reflex action became almost 
 imperceptible. When placed in a warm room the 
 animal completely recovered. 
 
 The effect of low temperatures on cold-blooded 
 animals is entirely different from that on warm- 
 blooded animals. The author states that, at low 
 temperatures, a frog became as hard as a stone in 
 from ten to twenty minutes, while the warm-blooded 
 animal produced within itself sufficient heat to 
 enable it to remain soft and comparatively warm 
 during an exposure for one hour to a temperature 
 of 100 F. The production of heat, however, 
 was not equal to the loss, and the animal was con- 
 tinually losing ground ; the bodily temperature 
 having decreased 56 during the exposure. 
 
 In 1893 Pictet 1 made a second series of exper- 
 iments. He describes the struggle of nature 
 against external attacks as follows : 
 
 When a dog is placed in a copper receiver which 
 is cooled down to from 60 to 90, its temper- 
 ature rises about one-half degree during the first 
 ten minutes. After ninety minutes the tempera- 
 ture falls one degree. Then follows a point where 
 the struggle is given up; the temperature falls rap- 
 idly, and the animal dies suddenly. The author 
 made similar experiments by exposing his arm to 
 
 l Chemiker Zeitting, 1893, p. 1337; Chem. News, 68, p. 312. 
 
CONCLUSION 239 
 
 low temperatures. The only pain occurs within 
 the arm on the periosteum, the epidermis expe- 
 riencing no pain whatever. 
 
 All insects resist a temperature of 28, but 
 not 35. Myriapods resist a temperature of 
 -50, and snails 120. The eggs of birds lose 
 their vitality at 2 or 3, the eggs of ants at 
 o, and the eggs of the silkworm at 40. In- 
 fusoria died at 90, and bacteria retained their 
 virulence after an exposure to a temperature of 
 - 213. The temperature in the case of the bac- 
 teria was lowered by means of frozen atmospheric 
 air. 
 
 In 1894 Pictet 1 continued his observations, and 
 exposed himself to a temperature of 110. The 
 body was well protected by clothing during this 
 exposure, and the legs were kept in motion. Pictet 
 says that, for a number of years previous to this 
 exposure, he had been suffering from indigestion, 
 and then adds that the exposure to this low tem- 
 perature effected almost a complete cure; He 
 suggests that the influence of low temperatures 
 on physiological action may prove to be of great 
 therapeutic value. 
 
 Dewar 2 states that McKendrick tried the effect 
 of low temperatures on the spores of micro-organ- 
 
 1 Compt. rend., 119, p. 1016. 
 
 2 Chem. News, 67, p. 211, 1893. 
 
240 LIQUEFACTION OF GASES 
 
 isms. Samples of flesh, blood, milk, and similar 
 substances were sealed in glass tubes and exposed 
 for a period of one hour to a temperature of 
 - 182. On standing at the ordinary temperature 
 for several days these substances became putrid. 
 Seeds of different kinds germinated after being 
 exposed to a temperature of 182. Dewar con- 
 siders the experiments with seeds as an evidence 
 in favor of the possibility of Lord Kelvin's sug- 
 gestion, that life may have been brought to the 
 newly cooled earth upon a seed-bearing meteorite. 
 
 4. Properties of Matter at Low Temperatures 
 
 At the low temperatures obtained by the evap- 
 oration of liquid air, etc., the properties of most 
 substances are materially modified. All organic 
 compounds and, with very few exceptions, all 
 inorganic compounds, are solids at that tempera- 
 ture. Rubber placed in liquid air loses its elas- 
 ticity and becomes as brittle as glass, but regains 
 its original condition when the temperature is 
 allowed to rise. Tin and many other metals also 
 become very brittle. Kreutz l has shown that the 
 power of absorbing light is considerably modified 
 at low temperatures. Many substances (mercuric 
 
 1 Phil. Mag. [5], 39, p. 209. 
 
CONCLUSION 241 
 
 iodide, mercuric oxide, etc.) change color under 
 such conditions. 
 
 Pictet 1 and Dewar 2 have investigated the in- 
 fluence of low temperatures on the phenomenon 
 of phosphorescence. Many substances lose their 
 power of phosphorescence at low temperatures, 
 while other substances, which exhibit only a feeble 
 phosphorescence at the ordinary temperature, be- 
 come more phosphorescent. A temperature of 
 80 is sufficient to stop all sensible emission 
 from previously excited calcium sulphide, but it 
 does not prevent the unexcited calcium sulphide 
 from absorbing light-energy which can be evolved 
 at higher temperatures. 
 
 Zakrzewski 3 has shown that the specific heat of 
 silver changes about three per cent in the interval 
 from o to 100. Important results have also 
 been obtained by Dewar and his associates on the 
 resistance of metals at low temperatures. 4 The 
 resistance of any metal to the passage of the elec- 
 tric current decreases with decreasing temperature. 
 This law, however, does not hold true for alloys. 
 Dewar says : " The results point to the conclusion 
 that pure metals have no resistance near the abso- 
 
 1 Compt. rend., Sept. 24, 1894. 
 
 2 Proc. Roy. Inst., 14, p. 665, 1895. 
 
 3 Phil. Mag. [5], 39, p. 191. 
 
 4 See references on p. 178. 
 
242 LIQUEFACTION OF GASES 
 
 lute zero of temperature. With alloys there is 
 little change in resistance. In the case of carbon 
 the resistance decreases with increasing tempera- 
 ture. At the temperature of the electric arc, car- 
 bon appears to have no resistance." 
 
 Some interesting observations have been made 
 on chemical action at low temperatures. Dewar 
 says : " At a temperature of 200 the molecules 
 of matter seem to be drawing near to what might 
 be called the ' death of matter ' so far as chemical 
 action is concerned." At this temperature yellow 
 phosphorus and liquid oxygen show no signs of 
 reaction. In 1861 Loir and Drion 1 showed that 
 liquid ammonia in contact with concentrated sul- 
 phuric acid does not react at first. Most of the 
 more recent experimenters on the liquefaction of 
 gases have made observations on chemical action 
 at low temperatures. The references in most cases 
 have already been given. In general, chemical 
 action ceases under these conditions. Metallic 
 sodium and potassium can be thrown into liquid 
 oxygen without action. Pictet 2 has shown that 
 sodium does not react with aqueous hydrochloric 
 acid (15 % solution) at a temperature of 80. 
 
 In some cases, however, chemical action takes 
 place at very low temperatures. Liquid ethylene 
 
 1 Phil. Mag. [4], 20, p. 202. 
 
 2 Compt. rend., 115, p. 814, 1894. 
 
CONCLUSION 243 
 
 reacts with chlorine and bromine at a tempera- 
 ture of 102. 5. Dewar 1 has shown that photo- 
 graphic action takes place at a temperature of 
 - 1 80. The photographic action seems to be 
 reduced by about 80 per cent. The author states 
 that " it is certain that the Eastman film is fairly 
 sensitive to photographic action at a temperature 
 of 200. " Quite recently it has been shown that 
 liquid fluorine at a temperature of - 187 reacts, 
 with evolution of light and heat, with hydrogen, 
 benzene, turpentine, etc. 
 
 The preceding observations merely indicate the 
 various directions in which liquefied gases have 
 been employed. More than a century ago the great 
 Lavoiser predicted that, were the earth suddenly 
 placed in a very cold region, the atmosphere would 
 cease to exist as an invisible fluid, but would return 
 to the liquid state, and new liquids, of which we 
 have no knowledge, would be produced. This pre- 
 diction has been thoroughly confirmed. One by 
 one the various gases have been condensed to the 
 liquid state, and the term " permanent gas " has 
 lost its significance. Along with the development 
 of the methods employed in the liquefaction of 
 gases, new industries of great commercial value 
 have been opened up. The extremely low tem- 
 
 1 Proc. Roy. Inst., 14, p. 665. 
 
244 LIQUEFACTION OF GASES 
 
 peratures which can be obtained by means of these 
 liquids have broadened the range of scientific re- 
 search. Numerous and important observations in 
 this direction have already been made, yet the 
 investigations at low temperatures are only in their 
 infancy. For many years the scientific world has 
 been speculating in regard to the probable condition 
 and properties of matter at the absolute zero of 
 temperature, a temperature which experimen- 
 ters have sought in vain to reach. Every year, 
 however, shortens the distance to travel, and at 
 present only a few degrees separate us from the 
 desired goal. 
 
INDEX TO AUTHORS 
 
 Addams, 39-40. 
 
 Aime, 43-44. 
 
 Altschul, 92. 
 
 Amagat, 69, 87, 98-102, 112. 
 
 Am on tons, 9. 
 
 Andrews, 3, 14, 18, 20, 55, 63 
 
 70-82, 84, 94. 96, 108, 
 
 no, 115, 230, 231. 
 Ansdell, 83, 85, 230. 
 Arago, 38, 66. 
 Avenarius, 85. 
 
 Becquerel, 171. 
 
 Berthelot, 61-62, 114, 116, 120. 
 
 Bianchi, 59. 
 
 Boerhaave, 7-9. 
 
 Boyle, 7, 96. 
 
 Brush, 228, 229. 
 
 Bussy, 28-29, 64. 
 
 Cagnaird de la Tour, 18-21, 
 55, 70. 
 
 Cailletet, 3, 4, 69, 82, 87, 91, 
 114, 115-120, 136, 137, 
 152, 187, 214, 216, 221. 
 
 Chappuis, 137, 160. 
 
 Charles, 9, 10, n. 
 
 Clark, 87, 92. 
 
 Clausius, in, 112, 181. 
 
 Clouet, 17. 
 
 Cohn, 235. 
 
 Colardeau, 82, 91. 
 
 Coleman, 186-187,236. 
 
 Colladon, 31-32. 
 
 Crookes, 94, 229. 
 
 Cullen, 181. 
 
 Dalton, 10, n, 12, 181. 
 
 Darwin, 181. 
 
 Davy, 22, 23, 24, 27. 
 
 Deleuil, 113. 
 
 Despretz, 65. 
 65, Deville, 113, 120. 
 109, Dewar, 4, 10, 83, 143, 145, 152, 
 167-178, 187, 197-199, 218- 
 221, 221-227, 230, 239, 240, 
 241, 242, 243. 
 
 Dickson, 178. 
 
 Drion, 63-64, 70, 242. 
 
 Dufour, 134. 
 
 Duhem, 84. 
 
 Dulong, 38, 66. 
 
 Dumas, 50, 113, 114. 
 
 Ewing, 192. 
 
 Faraday, 3, 16, 22-27, 28, 40, 44- 
 45, 56,71,171. 
 
 Fleming, 178. 
 Fourcroy, 15. 
 
 "3. 
 
 138, 
 
 Gay Lussac, 10, n, 12, 13, 96, 181, 
 
 204. 
 
 Giffard, 182. 
 Gorrie, 181. 
 Gouy, 88. 
 Guericke, 6. 
 Guldberg, 230. 
 Guy ton de Morveau, 15. 
 
 Hampson, 186, 192-195, 199. 
 
 Hannay, 88, 89, 92, 93, 94. 
 Hautefeuille, 137, 160. 
 
 245 
 
246 
 
 INDEX TO AUTHORS 
 
 Heen-Liittich, 88, 
 Heilborn, 82, 230. 
 van Helmont, 6. 
 Heydweiller, 233. 
 Him, 112. 
 Hogarth, 88, 89. 
 Houston, 183, 184, 185. 
 
 Jamin, 90, 91. 
 
 Joule, 13, 112, 181, 202, 205. 
 
 Kamerlingh-Onnes, 178-180, 187. 
 Kelvin (Lord). See William 
 
 Thomson. 
 Kirk, 182. 
 Kreutz, 240. 
 Kuenen, 83. 
 
 Lavoisier, 10, 243. 
 
 Leslie, 31. 
 
 Linde, 182, 186, 187-192, 193, 
 
 199, 210. 
 
 Loir, 64, 242. 
 
 Marchant, 182. 
 
 Mariotte, 7. 
 
 van Marum, 14, 65. 
 
 Mascart, 120. 
 
 Mathias, 87. 
 
 Maugham, 40. 
 
 Maxwell, 70. 
 
 Mayer, 181, 203, 204. 
 
 McKendrick, 236, 239. 
 
 Melsens, 236. 
 
 Mendeleeft, 64-65, 70, 71. 
 
 Miller, 71, 72. 
 
 Mitchell, 42. 
 
 Moissan, 218-221, 230. 
 
 Monge, 17. 
 
 Nadejdine, 230. 
 Natanson, 166. 
 Natterer, 53, 56-61, 66, 67. 
 Nernst, 211. 
 Northmore, 17, 18. 
 
 Olszewski, 4, 137-142, 143, 152- 
 167, 187, 213-218, 220, 222, 
 225, 227, 230. 
 
 Paris, 23. 
 
 Pawlewski, 83. 
 
 Pellat, 232. 
 
 Perkins, 27, 28, 42. 
 
 Pictet, i, 3, 4, 31, 113, 114, 115, 
 118, 120-137, 152, 178, 221, 
 236, 238, 239, 241, 242. 
 
 Postle, 182. 
 
 Pouillet, 66. 
 
 Preston, 68. 
 
 Prytz, 232. 
 
 Ramsay, 87, 89, 90, 213, 217, 227, 
 
 228, 230. 
 
 Rankine, 112, 181. 
 Rayleigh, 210. 
 Recknagel, 112. 
 Regnault, 67, 69. 
 Ritchie, 59. 
 Roberts-Austen, 233. 
 Roscoe, 115. 
 Rowlands, 205. 
 Rumford (Count), 13. 
 
 Sajotschewski, 230. 
 Sarrau, 112. 
 Schorlemmer, 115. 
 Siemens, 182. 
 Silliman, 41, 42. 
 Solvay, 182. 
 Soubeiran, 50. 
 Stromeyer, 16. 
 
 Thenard, 16, 38. 
 
 Thiesen, 87. 
 
 Thilorier, 33-39, 40, 42, 45, 51, 62, 
 
 76, 181. 
 
 Thomson, J., 108. 
 Thomson, William, 12, 13, 112, 181, 
 
 202, 240. 
 Torrey, 41-42. 
 Traube, J., 112. 
 
INDEX TO AUTHORS 
 
 247 
 
 Travers, 227, 228. 
 Tripler, 186, 199-202. 
 Tyndall, 23. 
 
 Vanquelin, 15. 
 Villard, 92. 
 Violi, 112. 
 
 van der Waals, 84, 96, 105-112. 
 Wallis-Tayler, 181, 182, 183. 
 
 Windhausen, 183, 184. 
 
 Witkowski, 167. 
 
 Wroblewski, 4, 137-142, 143- 
 
 152, 153, 187, 222, 225, 230, 
 
 234- 
 
 Young, in, 230. 
 Yung, 236. 
 
 Zakrewski, 241. 
 
INDEX TO SUBJECTS 
 
 Acetylene, liquefaction of, 114. Carbon disulphide, 81. 
 
 vapor pressure of, 114, 
 Adiabatic expansion, 207, 209. 
 Air, attempts to liquefy, 27, 31, 60, 
 
 boiling point of, 144. 
 
 density of liquid, 177. 
 
 liquefaction of, 119, 148, 150, 156, 
 169, 177, 187-202. 
 
 solidification of, 172, 177, 225. 
 Alcohol, vaporization of, in closed 
 
 tubes, 18. 
 
 Ammonia, liquefaction of, 15, 27, 
 29, 62, 8 1. 
 
 solidification of, 52. 
 
 vapor pressure of, 53. 
 Argon, liquefaction of, 213-216. 
 
 boiling point of, 216. 
 
 solidification of, 216. 
 Arsenic, sublimation of, 231. 
 Arsine, liquefaction of, 16, 50. 
 
 solidification of, 158. 
 
 vapor pressure of, 53. 
 
 Benzene, expansion of liquid, 90. 
 Boiling point, table of, 230. 
 
 absolute, 65, 71. 
 Boyle's law, 7, 96. 
 exceptions to, 15, 44, 61, 65-69, 
 
 79, 97-102. 
 
 Breaking stress of metals at low 
 temperatures, 176. 
 
 Calcium chloride, solution of, in 
 
 alcohol vapor, 89. 
 Carbon dioxide, see Carbonic Acid. 
 
 thermal transparency of, 175. 
 Carbonic acid, liquefaction of, 13, 
 *7, 25, 33, 40, 41, 42, 56, 60, 62, 
 64, 72, 75, 80. 
 solidification of, 37, 40, 42 51 60, 
 
 64. 
 Carbon monoxide, attempts to 
 
 liquefy, 43, 54, 60, 62, 71. 
 liquefaction of, 113, 114, 118, 142, 
 
 148. 
 
 solidification of, 149, 157. 
 Chemical action at low tempera- 
 tures, 242-243. 
 Chlorine, liquefaction of, 17, 22, 23, 
 
 29, 62. 
 
 solidification of, 158. 
 Chlorochromic acid, liquefaction 
 
 of, 42. 
 
 action of phosphorus on liquid, 42. 
 Chloroform, thermal transparency 
 
 of liquid, 175. 
 
 Coal gas, attempt to liquefy, 54. 
 Cobaltous chloride, solution of, in 
 
 alcohol vapor, 89. 
 Critical point, condition of matter 
 
 at the, 84-96. 
 Critical pressure, 80, 81-83. 
 
 table of, 230. 
 Critical temperature, 65, 71, 81-83. 
 
 table of, 230. 
 
 Cyanogen, liquefaction of, 26, 27. 
 solidification of, 27, 52. 
 
 Dielectric constant, at low tempera- 
 tures, 178. 
 
 248 
 
INDEX TO SUBJECTS 
 
 249 
 
 Electric conductivity of metals at 
 
 low temperatures, 178. 
 Ethane, attempt to solidify, 159. 
 Ether, vaporization of, in closed 
 tubes, 18, 19, 20, 81. 
 
 thermal transparency of, 175. 
 Etherion, 229. 
 Ethylene, liquefaction of, 48. 
 
 solidification of, 158. 
 
 thermal transparency of liquid, 
 
 175- 
 Euchlorine, liquefaction of, 26, 51. 
 
 solidification of, 51. 
 Expansion of gases, influence of 
 
 temperature on, 12. 
 Explosible ether, 37. 
 
 Fluoboron, liquefaction of, 50. 
 
 Fluorescence, 90. 
 
 Fluorine, liquefaction of, 218-221. 
 
 attempt to solidify, 221. 
 
 reaction of, at low temperatures, 
 
 220, 221, 243. 
 Fluosilicon, liquefaction of, 49. 
 
 Gas, definition of, 6, 81, 95, 232. 
 
 kinetic theory of, 102-104. 
 
 relation to liquid, 84-112. 
 Gaseous mixtures, critical constants 
 of, 82. 
 
 Helium, attempts to liquefy, 216- 
 
 218. 
 
 liquefaction of, 227. 
 Hydriodic acid, liquefaction of, 49. 
 
 solidification of, 49. 
 Hydrobromic acid, liquefaction of, 
 
 49- 
 
 solidification of, 49. 
 Hydrocarbons, liquefaction of, 86, 
 
 187. 
 Hydrochloric acid, liquefaction of, 
 
 15, 17, 23, 27, 81. 
 solidification of, 158. 
 Hydrochloric ether, expansion of 
 liquid, 63. 
 
 Hydrofluoric acid, solidification of, 
 
 158- 
 Hydrogen, attempts to liquefy, 40, 
 
 43, 54. 60, 71, 114, 119, 135. 
 boiling point of, 164, 225, 226. 
 density of liquid, 225. 
 liquefaction of, 143, 153, 221- 
 
 226. 
 Hydrogen selenide, solidification 
 
 of, 158. 
 
 Hydrogen sulphide, see Sulphuret- 
 ted Hydrogen. 
 
 Ice machines, 181-183, 2 34- 
 
 Ice, sublimation of, 232. 
 
 Industrial application of liquefied 
 gases, 234, 235. 
 
 Iodine, solubility of, in liquid car- 
 bonic acid, 91, 92. 
 liquefaction of, 232. 
 
 Iron, sublimation of, 232. 
 
 Krypton, 227, 228. 
 
 Liquid, definition of, 95, 232. 
 
 Marsh gas, see Methane. 
 Matter, properties of, at low tem- 
 peratures, 240-244. 
 Mechanical equivalent of heat, 205. 
 Melting point, table of, 230. 
 Metargon, 228. 
 Methane, liquefaction of, 114, 145, 
 
 159- 
 
 density of liquid, 159. 
 solidification of, 158. 
 Muriatic acid, see Hydrochloric 
 Acid. 
 
 Naphtha, vaporization of, in closed 
 
 tubes, 18. 
 Neon, 228. 
 Nitric oxide, attempts to liquefy, 
 
 43, 54, 62, 71. 
 liquefaction of, 116. 
 solidification of, 158. 
 
2 5 
 
 INDEX TO SUBJECTS 
 
 Nitrogen, attempts to liquefy, 43, 
 
 54, 71- 
 
 density of liquid, 159, 177. 
 liquefaction of, 119, 142, 148, 159, 
 
 177. 
 
 solidification of, 150, 158. 
 Nitrogen tetroxide, expansion of 
 
 liquid, 63. 
 liquefaction of, 63. 
 Nitrous oxide, liquefaction of, 26, 
 
 52, 56, 60, 72, 8 1. 
 solidification of, 52, 60. 
 thermal transparency of liquid, 
 
 175- 
 
 Olefiant gas, see Ethylene. 
 Oxygen, attempts to liquefy, 40, 43, 
 
 54, 60, 62, 71. 
 
 absorption spectrum of, 172. 
 density of liquid, 159, 177- 
 liquefaction of, 113, 114, "8, 128- 
 132, 141, 143, 148, 159, l6 7, l6 9, 
 
 177, 179. I9 2 - I 99- 
 magnetic property of, 169. 
 solidification of, 134. 
 Ozone, liquefaction of, 137, 160. 
 attempt to solidify, 161. 
 
 Phosphine, liquefaction of, 49. 
 solidification of, 158. 
 
 Phosphorescence, at low tempera- 
 tures, 241. 
 
 Physical constants, table of, 230. 
 
 Physiological action at low tem- 
 peratures, 235-240. 
 
 Potassium iodide, solution of, in 
 alcohol vapor, 88, 89. 
 
 Propane, attempt to solidify, 159. 
 
 Radiant matter, 95. 
 Refrigerating machines, 182, 183. 
 Regenerative method of refrigera- 
 tion, 180-202. 
 theory of, 202-213. 
 
 Silicon tetrafluoride, solidification 
 
 of, 158. 
 
 Solid, definition of, 232. 
 Solids, diffusion of, 233. 
 solution of, in gases, 88-90. 
 solution of, in solids, 233. 
 Specific heat, of gases, 203. 
 of metals at low temperatures, 
 
 241. 
 
 Stibine, solidification of, 158. 
 Sulphur, solution of, in carbon di- 
 
 sulphide, 89. 
 Sulphur dioxide, see Sulphurous 
 
 Acid. 
 
 Sulphuretted hydrogen, liquefac- 
 tion of, 15, 25, 52. 
 solidification of, 52. 
 Sulphurous acid, liquefaction of, 
 
 15, 17, 24, 29, 42, 63. 
 solidification of, 43, 51. 
 
 Temperature, absolute, n. 
 
 influence of, on gaseous volume, 
 
 I, 10. 
 
 Thilorier's mixture, 39. 
 
 Vacuum bulbs, 172-175- 
 Vapor, definition of, 6, 81, 94. 
 density of saturated, 85-88. 
 Vaporization, heat of, 29. 
 
LIGHT VISIBLE AND INVISIBLE: 
 
 A SERIES OF LECTURES DELIVERED AT THE 
 ROYAL INSTITUTION OF GREAT BRITAIN, 
 
 BY 
 
 SYLVANUS P. THOMPSON, D.Sc., F.R.S., M.R.I., 
 
 Principal of and Professor of Physics in The City and Guild's Technical 
 College, Finsoury, London. 
 
 12mo. Cloth. $1.50. 
 
 Inter-Ocean : His demonstrations of the various points 
 of light phenomena that he takes up are admirably made, 
 and through requiring concentration of attention on the 
 part. of the reader, are full of the most delightful interest. 
 
 Outlook : The lectures are extremely popular in form, 
 cover all the important points involved, and include a 
 thorough and most readable chapter on the Roentgen 
 ray. While not a text-book, this might well serve as an 
 introduction to the systematic study of light phenomena. 
 
 Journal of Education : It is readable and up to date. 
 Here we have the results of many years' experience in 
 teaching and lectures on Optics. The book is fully illus- 
 trated, and will prove of great value to the student. 
 
 THE MACMILLAN COMPANY, 
 
 66 FIFTH AVENUE, NEW YORK. 
 
OUTLINES 
 
 OF 
 
 INDUSTRIAL CHEMISTRY, 
 
 A TEXT-BOOK FOR STUDENTS. 
 
 By FRANK HALL THORPE, Ph.D., 
 
 Instructor in Industrial Chemistry in the Massachusetts Institute 
 of Technology. 
 
 Cloth. 8vo. Price $3.50. 
 
 Education : The result is a text -book that will become a stand- 
 ard for use in colleges and technical schools. 
 
 Scientific American : We have long waited for a modern book 
 on this subject which would be strictly scientific, but which would also 
 give in plain, intelligible language the modern processes for making 
 of various chemicals and information relating to the carrying on of 
 various chemical industries. The need of a thoroughly modern 
 book in English on the subject has been very pronounced, and we 
 are happy to say that at last we have a book which, while possibly 
 not ideal, fills nearly all the conditions of a book of this kind. 
 The author has taken an extremely heterogeneous collection of 
 material, and has assorted and combined it with rare judgment. 
 The result is immensely satisfactory, which will place the book 
 among our standard works of reference. 
 
 Journal of Education : This treatise is all that the Massachu- 
 setts Institute of Technology stands for in scholarship, in science, 
 and in laboratory ideals. It is what the student seeks who would 
 master the elements of industrial chemistry, is all that the teacher 
 can ask, and meets the ideal of the specialist. 
 
 THE MACMILLAN COMPANY, 
 
 66 FIFTH AVENUE, NEW YORK. 
 
$ti*ll 
 
 7 DAY USE 
 
 AS. 
 
 RETURN TO DESK FROM WHICH BORROWED 
 
 PHYSICS LIBRARY 
 
 This publication is due on the LAST DATE 
 stamped below. 
 
 
 RB 17-60m-12,'57 
 
 General Library 
 
 University of California 
 
 Berkeley 
 
 LIBRARY OF THE UNIVERSITY OF CALIFORNIA 
 
ERSITY OF CALIFORNIA LIBRARY OF THE UNIVERSITY OF CALIFOI 
 
 ase 
 
 1 
 
 " 
 
 ERSITY OF CALIFORNIA LIBRARY OF THE UNIVERSITY OF CALIFOf 
 
 Q^\i> 
 iRSITY OF CALIFORNIA IIRRARY fif TUF imvcn CITY ns