摘要:

118 ELECTRICAL THEORY OF ADBORPTTON The writer considers the double layer as consisting of a swface of rigidly fixed atoms under continuous bombardment of positively and negatively charged ions, any particular point on the rigid surface becoming in turn negative, neutral and positive, these conditions arisdg in any order. The observed contact difference is the average effect of these conditions. Where several kinds of atoms are present in the solution the average number of any one of them at the surface will depend on their concentbration, valency and mobility. The variation of contact Werence from negative to neutral and positive was observed with cotton and aluminium sulphate near the neutral point. These variations occurred during the same experiment, the readings being direct measurements of E.1I.F.s developed by filtration under pressure.This point would be covered by putting n2 = 1 and = 2 or 3 in Mukherjee’s equation No. 13.118 ELECTRICAL THEORY OF ADBORPTTON The writer considers the double layer as consisting of a swface of rigidly fixed atoms under continuous bombardment of positively and negatively charged ions, any particular point on the rigid surface becoming in turn negative, neutral and positive, these conditions arisdg in any order. The observed contact difference is the average effect of these conditions. Where several kinds of atoms are present in the solution the average number of any one of them at the surface will depend on their concentbration, valency and mobility. The variation of contact Werence from negative to neutral and positive was observed with cotton and aluminium sulphate near the neutral point.These variations occurred during the same experiment, the readings being direct measurements of E.1I.F.s developed by filtration under pressure. This point would be covered by putting n2 = 1 and = 2 or 3 in Mukherjee’s equation No. 13. The Fnvnday Socitty a's not responsible foy opinions expressed bcfore it by Authors OY Speakers. Cransactions OF $arabap 5ociety. FOUNDED 1903. T O PROMOTE THE STUDY OF ELECTROCHEMISTRY, ELECTROMETALLURGY, CHEMICAL PHYSICS, METALLOGRAPHY, AN0 KINDRED SUBJECTS. VOL. XVIII. DECEMBER, I 922. PART 2 . THE GENERATION AND UTILISATION OF A GEiV-ER A I, D?SC USSION. COLD. GENERAL DISCUSSIOK on " THE GEKERATION AND UTILIS.4- TION OF COLD " was held jointly by The Faraday Society and The British Cold Storage and Ice Association on Monday, October I 6th, 1922, in the Hall of the Institution of Electrical Engineers, Victoria Embankment, London.The meetii;g consisted of the following three sessions :- (i) 2.30 to 4. I j p.m. ; presided over by Professor Alfred W. Porter, D.Sc., F.R.S., F.Inst.P., President of the Faraday Society.. ( i i ) 4.45 to 6. I j p.in. ; presided over by Mr. George Gosdsir, J.P., F.R.S.S., President of the British Cold Storage and Ice Association. (iii) 7.45 to 9.30 p.m. ; presided over by Mr. James Swinburne, F. R.S., M.Inst.C.E., F. Inst.P., Past- President of The Faraday Society. During the intervals demonstrations were given by Messrs 1 he subject was treated under two headings :- L i q u i d Air (Limited). r 7 I37138 THE GENERATION AND UTILISATION OF COLD (i) Laboratory Methods of Liquefaction. (ii) Industrial Methods of Liquefaction and Practical Applica- tions of Low Temperatures. Professor Alfred W. Porter opened ,the proceedings with the following general introduction.

ISSN:0014-7672    

DOI:10.1039/TF9221800137

出版商:RSC    

年代:1922

数据来源: RSC

摘要:

118 ELECTRICAL THEORY OF ADBORPTTON The writer considers the double layer as consisting of a swface of rigidly fixed atoms under continuous bombardment of positively and negatively charged ions, any particular point on the rigid surface becoming in turn negative, neutral and positive, these conditions arisdg in any order. The observed contact difference is the average effect of these conditions. Where several kinds of atoms are present in the solution the average number of any one of them at the surface will depend on their concentbration, valency and mobility. The variation of contact Werence from negative to neutral and positive was observed with cotton and aluminium sulphate near the neutral point. These variations occurred during the same experiment, the readings being direct measurements of E.1I.F.s developed by filtration under pressure.This point would be covered by putting n2 = 1 and = 2 or 3 in Mukherjee’s equation No. 13.118 ELECTRICAL THEORY OF ADBORPTTON The writer considers the double layer as consisting of a swface of rigidly fixed atoms under continuous bombardment of positively and negatively charged ions, any particular point on the rigid surface becoming in turn negative, neutral and positive, these conditions arisdg in any order. The observed contact difference is the average effect of these conditions. Where several kinds of atoms are present in the solution the average number of any one of them at the surface will depend on their concentbration, valency and mobility. The variation of contact Werence from negative to neutral and positive was observed with cotton and aluminium sulphate near the neutral point.These variations occurred during the same experiment, the readings being direct measurements of E.1I.F.s developed by filtration under pressure. This point would be covered by putting n2 = 1 and = 2 or 3 in Mukherjee’s equation No. 13.118 ELECTRICAL THEORY OF ADBORPTTON The writer considers the double layer as consisting of a swface of rigidly fixed atoms under continuous bombardment of positively and negatively charged ions, any particular point on the rigid surface becoming in turn negative, neutral and positive, these conditions arisdg in any order. The observed contact difference is the average effect of these conditions. Where several kinds of atoms are present in the solution the average number of any one of them at the surface will depend on their concentbration, valency and mobility.The variation of contact Werence from negative to neutral and positive was observed with cotton and aluminium sulphate near the neutral point. These variations occurred during the same experiment, the readings being direct measurements of E.1I.F.s developed by filtration under pressure. This point would be covered by putting n2 = 1 and = 2 or 3 in Mukherjee’s equation No. 13.118 ELECTRICAL THEORY OF ADBORPTTON The writer considers the double layer as consisting of a swface of rigidly fixed atoms under continuous bombardment of positively and negatively charged ions, any particular point on the rigid surface becoming in turn negative, neutral and positive, these conditions arisdg in any order.The observed contact difference is the average effect of these conditions. Where several kinds of atoms are present in the solution the average number of any one of them at the surface will depend on their concentbration, valency and mobility. The variation of contact Werence from negative to neutral and positive was observed with cotton and aluminium sulphate near the neutral point. These variations occurred during the same experiment, the readings being direct measurements of E.1I.F.s developed by filtration under pressure. This point would be covered by putting n2 = 1 and = 2 or 3 in Mukherjee’s equation No. 13.118 ELECTRICAL THEORY OF ADBORPTTON The writer considers the double layer as consisting of a swface of rigidly fixed atoms under continuous bombardment of positively and negatively charged ions, any particular point on the rigid surface becoming in turn negative, neutral and positive, these conditions arisdg in any order.The observed contact difference is the average effect of these conditions. Where several kinds of atoms are present in the solution the average number of any one of them at the surface will depend on their concentbration, valency and mobility. The variation of contact Werence from negative to neutral and positive was observed with cotton and aluminium sulphate near the neutral point. These variations occurred during the same experiment, the readings being direct measurements of E.1I.F.s developed by filtration under pressure.This point would be covered by putting n2 = 1 and = 2 or 3 in Mukherjee’s equation No. 13. THE GENERATION AND UTIIJISA'TION OF COT,DD. A GENERAL DISCUSSION OPENING REMARKS BY THE PRESIDENT OF THE F!&AU)L41r SOCIETY, ALFRED TV. PORTER, Il.Sc., F. R.S., F.Inst. P. We have met together to discuss the questions of the production and utilisation of cold. This is the thirt;-wcond General Discussion which has been arranged under the auspices of the Faraday Society, often in con- junction with others. On this occasion we have associated with us the British Cold Storage and Ice Association, and as President of the Faraday Society I welcome its members to our meeting. We had looked forward to having with us Professor Kamerlingh Onnes and Professor J.1'. Kuenen of 1,eyden. On the eve of the completion of the arrangements for the meeting, however, science suffered a great loss by the sudden and unexpected death of Professor Kuenen. He was well known to many in England, and therefore his loss is the more personal. I remember him when he worked at University College with Sir William Kamsay. He xias afterwards appointed to be Professor of Physics at Dundee University, nnd in thebe connections he made many friends in England. H e was best known scientifically for his work on binary mixtures of liquids and vapour>. He nab taking an active interest in the preparations for this meeting when his untimely death occurred. Professor Onnes, owing to delicate health, is also prevented from coming. 'The duty of representing the Leyden Cryogenic Laboratory has therefore fallen upon Professor Crommelin W ~ G I I I we are delighted and honoured in welcoming here.M. Georges Claude of I'aris was also expected, but is unable to be present. H e has, however, sent a communication which is in print and this will be read for him. This is not the place to go into detail in regard to the principles of mechanical refrigeratioh. They are all based upon the fundamental fact that heat will not flow spontaneously from a cold to a hot body : the transference can be effected only by the performance of mechanical work. The most theoretically perfect way of doing this is by a contrivance which passes the working substance through a reversed Carnot cycle of changes. In such a cycle the heat Qz can be removed from a refrigerated tnaterial at temperature T,, and the amount Q1 passed out of the system into a condenser at a higher temperature T, by the aid of the work, I?', required to effect the changes, where and 139140 THE GENERATION AND UTILISATION OF COLD So that the coefficient of performance = T, - --I__ Q2 Q2 -=-- 1%’ Qj - Qz TI - ‘I.’?’ This value of the coefficient of performance represents, however, the un- attainable limit: in practice, even if the Carnot cycle is aimed at, ad- ditional work has to be done owing to frictional and other losses and the coefficient of performance is always reduced thereby.The Carnot cycle requires that the compression and expansion shall take place in a cylinder in much the same way as in a steam engine; but in the reversed sense.I n mechanical refrigeration it is found to be more convenient to replace the cylinder expansion by expansion through a throttle or valve. I n this expansion less external work is done by the system than in cylinder expansion, and therefore to effect the same change in other respects there is a greater call for supply of external work from elsewhere. If the pressure and volume change fromp, and v, top, and v:, in the expansion the work done is onlyp,~, - plvl, whereas in a cylinder expansion it would have been The difference is FIG. I. I am accustomed to give the name “shirk” to this work that fails to be done; that IS, work actually done + shirk = work that might have been done. internal energy + pv = constant, or E -t pv == constant.The quantity E +pv is what is usually known in England as Totai Heat or Heat Contents. I submit that these names are not satisfactory, because the quantity is not heat in general; and the presence here of a member of the Leyden Laboratory encourages me to press the claims of the name proposed by Kamerlingh Onnes, viz., Enthalpy-a name which I have used for some years. We can denote it by H , which can stand either for capital h (Heat Contents) or for Greek E (Enthalpy) at the option’ of the reader. The outline scheme in a CO, or NH, machine is as seen in the The equation characteristic of the expansion through a throttle is thatTHE GENERATION AN11 UTII,ISATION OF COLL) 141 ._ --I I _ _ _ _ _ _ - I AB 1 HB - H A BC I Hc - H, CD ~ 0 DA 1 H, - H D diagram. expansion in the throttle is froin C to D.in this expansion; neither does it change in any complete cycle. can write down the following specification of the various changes :- l'he fluid is kept circulating by a compressor pump. The The enthalpy does not change We _ _ - 1 _ _ _ _ ~ _ _ _ __ - 0 ' Cylinder compressioii. 0 1 Throttle expansion (adi;ibatic). H, - H, HA - H, 1 Constant pressure cooling. ' Refrigerating eva?oration. 'CVIT1-I THROTTLE EXPANSION. Change of Path. I Enthaipy. Path. 1 Change of Enthalpy. 1 Heat Entry. 1 Characteristics. Heat Entry. AB 1 H, - HA 1 0 BC HC - HB Hc - H, CDO H D U - HC 0 D,A HA - HDo HA - H," = Qz Hence, Total work done on system = - Total heat entry = - H A + HD - Hc + HB = HB - HA [became Hc = HD]. Cylinder compression (const.entropy). Const. press, cooling. Cylinder expansion (const. entropy). Refrigerating evaporation. The heat removed from the refrigerated material at the low temperature is HA - H D . Hence the coefficient of performance is- HA - H D H B - HA' With cylinder expansion the data are :- WITH CYLINDER EXPANSION. Characteristics. --- -- -I I I _____--- The coefficient of performance is :- Q2 HA - HD,, IV - = HB - HA + H D , - HD' Because HD > HD,, and Q2>W this expression is greater than for throttle expansion. The adoption of the latter is governed by convenience. I have given these results to show thc importance of enthalpy in dealing with refrigeration problems. Throttle expansion is adopted for convenience, but it is well known that at sufficiently high temperatures it produces heating instead of cooling.This is so for hydrogen and helium at moderate pressures at ordinary temperatures. In such cases pre-cooling is necessary. I n the following account we will suppose that the flow is strictly adiabatic. I wish to emphasise, what is not so generally known, that there is in most cases, and probably in all cases, also a minimum temperature at which cooling takes place. Such a point has already been experimentally determined142 THE GENERATION AND UTILISATION OF COLD for CO, by Professors Jenkin and Pye.' The position found directly lay between - 20.7~ and - 31' C. By plotting their data for volume and temperature I calculated in 1913 that its value must lie at a temperature not much removed from - 24' C. when the pressure is about half the critical.This result is remarkable, not only owing to its bearing upon the possible use of a throttle at low temperatures but also because it shows that liquid CO, is, in this region, behaving very nearly as a perfect gas-its volume being nearly proportional to the absolute temperature. Moreover, I had previously shown that all the well-known equations of state concur in indicating that such points should exist. The con- clusions to be derived from the. various equations are that at any one pressure there are two inversion temperatures if any at all; but that above a certain pressure heating in passing through a throttle is the universal rule at all temperatures. These results are shown in Fig. 2 on which FIG. 2. Curve A. van der Waals.,, B. Dieterici (modified). ,, C. Clausius. x x x Inversion points for Nitrogen from experiment. 0 0 0 9 7 7 , ,, co, } the ordinates are reduced temperatures and the abscissae are reduced pressures, The curves shown represent the boundary between the heating and cooling regions on various assumptions and from experiment. The change of pressure in going through the throttle is supposed to be small. Curve A is according to van der Waals' equation of state. I t would indicate that for a pressure greater than nine times the critical the gas would undergo heating at all temperatures; also that for a reduced pressure of 5 (for example) the gas would cool only if the temperature lay between 5.4 and 1-3 times the critical. I t does not appear that any substance follows van der Wads' equation however. Curve C is, according to Clausius' equation, Phil.Trans., 1913, A, No. 499. Pvoc. Roy. SOL, A. Vol. LXXXIX., 377. 3 Phil. Mag., April, 1906, June, 1910.THE GENERATION AND UTILISATION OF COLD 143 where a, /3, y are the reduced pressure, volume and temperature respectively. Curve B is, according to Dieterici’s form of equation, modified as regards the power of y in the index as follows :- vhere 12 is taken as j / 3 . On the same curve are shown inversion points for nitrogen (crosses) and for carbon dioxide (small circles) obtained from Amagat’s data of p1 vl T for these gases by making use of the fact that at an inversion temperature (at which heating changes to cooling or vzce versa) ?‘- - v = 0. These experimental values lie m h o s t precisely on thc cwve calculated Jiroin Dieterici’s epuutiun with its ncodz9ed index.The cooling and heating regions for these real cases are labelled in the diagram. These results can be trusted for points above the critical point, but for lower temperatures and high pressures, Dieterici’s equation is not valid. I t will be perfectly clear froin the above that no such law as the inverse square temperature law for the cooliiig coefficient can possibly be valid in general. The true law may change sign twice or not at all. These facts may not have any importance in connection with food re- frigeration, because the working substances selected will always be such as at ordinary temperatures are either below or at least in the neighbour- hood of their critical temperatures; but they do impose a limit to the lowest temperature attainable by throttle expansion. A further word of explanation may be useful. The fall of temperature in throttle expansion is to be reckoned between any two points (one on each side of the throttle) at which the velocity of flow is small. In the throttle valve itself very high velocities may be reached, and since a considerable amount of energy is then present as kinetic energy the temperature may fall i~i the throttle below the final temperature. Ordinary hydrodynamics shows that if there were no friction in the gas the quantity - 1 vd’ would be transformed into kinetic energy, ultimately being re- trznsformed as the passage gets wider and the kinetic energy disappears. In the actual fluid, friction is never absent, and is continually generating heat at the expense of the kinetic energy; and the temperature even in the narruws never falls as low as it otherwise would. There may be condensation into the liquid or even the solid phase in the narrows, but unless the liquid or solid can be trapped it will pass on with the current as a mist or sleet, gradually melting by the friction it esperiences until in the wider part o l the passage it will have a temperature greater or less than at start as indicated in the above considerations and shown in Fig. i. > ZI 3‘2‘

ISSN:0014-7672    

DOI:10.1039/TF9221800139

出版商:RSC    

年代:1922

数据来源: RSC

摘要:

PART I. - LABORATORY METHODS O F LIQUIWACTION. On the lowest temperature yet obtained BY Prof. Dr. H. KAMSRLINGH ONNES. Mr. President and Members of the Faraday Society and thc British Cold Storage and Ice Association. It is with deep sorrow that I bring to remembrance here the great loss science has suffered by the death of my dear friend and colleague Prof. KUENEN, a not less heavy personal loss for myself. On account of his many-sided work in experimental and theoretical thermo- dynamics, Prof. K U ~ E S was highly interested in the present meeting. His most beautiful work, such as the discovery of retrograde condensation, lies in the domain of the equilibrium of the liquid and gaseous phases i n which mixtures of gases separate, The part he would have taken in the discussion on industrial processes dealing with such mixtures would surely have been an iniportant one.I n the meantime he looked forward with great sympathy t o rny having the honour to address you. If it should prove that I could not read my paper personally he proniiscd to do so. The day before his unexpected death I saw him bright and in full vigour and it is a deep sorrow now to remember that the happy conversation we had the last hour we spent together had its starting point in arrangements f o r the present discussion. § 1. Introduction. - W-hen we approach the discussion of generation and utilisation of cold from the point of view of the application of liquefied gases in scientific laboratories and leave aside the study of tile properties of matter at low temperatures, including the principles of thermometry i n this domain, there are two points which come to the front.The first is to construct appliances which allow the making of measurements over all146 the range of temperatures that have Fecoiiic accessible. The second to descend to a lower temperature t,liaii hitherto has been done. As to the first point the paper of Dr. C'RO~IMEIJN gives a full description of the equipment of the Leiden cryogenic laboratory. The paper is in your hands, so 1 have not to dwell on the details of this equipment. But I hope t o be allowed to emphasize that, the true characteristic of the Leiden laboratory is to be found in its having a staff and a body of helpem as could only be formed in the long r u n of time. By constant contact with precision research in different departments of experimental physics, as well as with the historical development of the means to carry over these researches to always lower ranges of temperature, this staff has become the possessor of traditions which allow of the application of all experience gained in mastering the successive cryogenic difficulties to the attack of new problems of most varied kinds.In this way the laboratory with its staff realises in fact what has become more and more a n international desideratum of science. Science asks now in many departments for specialisation. An in- vestigator working in a special line and desiring to do any of his experiments at low temperatures, say a t those obtainable with liquid hydrogen, will evidently find great gain if in ar- ranging his special apparatus for work at these temperatures he can dispose of the help of a well trained staff to make this arrangement as efficient as possible.And when he has t o perform with this apparatus experiments which offer many difficulties in themselves, the help of this staff will allow him t o concentrate all his attention on the experiments, the mani- pulation with hydrogen going on so to say as if with water. The number of problems to be treated in this manner is increasing every day. As I decided some 40 years ago to undertake work a t low tcrnperatures, of course I had the conviction that it would give results of value for our understanding of the properties of matter. But the extension and importance which the work in this direction has attained, has widely surpassed any anticipation of mine.During the time that moderate quantities of liquid147 air had bcconic available foil tlic cxperiiiicuts which first occupied my iiiind, the necessity of some time making nicasiirements with liquid hydrogcn made itself felt. Before hydrogen was liqwficd, hcliuni was discovered, which subst aiiw afterwards pi*ovccl to ext piid enormously the domain opened by hydrogen. And again, before hcliuni was liquefied the discovery of PL~NCIC’S quarituiri lent a totally iicw aspect, to low tciiipcrature work. After the liquefaction of lieliuni superconductivity added an entirely unexpccted field of research. So there is an ever iii- creasing number of problems requiring work such as can hc dolie in the Tlcidcii laboratory.While 011 the oiic hand I shall shortly have the great satisfaction to be able to welcome tlie liquefying of helium in newly start cd cryogenic laboratories, on the other hand t.hc interiiational interest of the Leiden laboratory continuing its historical line of research becomes the morc prominent. Therc is one problciii in this line which, as I said before, comes naturally to thc Eront in the discussion of today. It is the extension to lower teniperatuyes of the range in which researches can be carried out. Allid so I beg to be d l o w d to give a preliminary account of experiments leading to the realisation of the lowest temperature yet attained. 4 2. First experiments. - As soon as tlie efforts to liyncfy hcliuiii had succeedcd, it was of course invcstigated whethcr it could 1 x 1 solidified also.This was even tried 011 the samc day on which licliniii had f o r the first time been seen as a liquid. The method used -\\-as that, of cvaporatioii under rcduccd pressurc Fig. 1 &ows diagrairiiiiatically the apparatus in which fQr t h e first time liquid helium has been seen and the circulation apparatus that served t o obtain it. After having been preli- minarily cooled in a n appropriate way the compressed helium passes through the spiral in the refrigerator in which it is cooled down t o tlic temperature of the hydrogen evaporating a t the air puiiip. Having been reduced to this extremely low tcmpcraturc it, ciitcrs tlic regenerator-spi~al, the end of which provided with a iiozzlc. IIew it cspaiids: part of it liquefies +lie LrsDi+process aiid tlw Yaponr rctnnis between the148 coils of the regenerator-spiral, the cscapiiig helium vapour being received in a gasholder and being compressed again by the pumps.The liquid helium collects a t the bottom of the vacuum- glass of the liquefier, where its accumulation can be seen: the bottom part of the liquefier glass was uiisilvered ; the surroun- ding vacuum glasses, the first one containing liquid hydrogen, the second onc liquid air, were also transparent. The glass with liquid air 'c1w protected against coiidensatioii of nioist-iwe from the air by a glass in which alcohol, kept at atmospheric ^"... -... . - 2.58. ---* -- 268.5 temperature, circulat cd. At the beginning of the experiment that was intended to produce refrigeration by evapor.stion of the liquid itself, thc nozzle was closed and the punips werc put out of action. The vapour emanating from the helium at the bottom part of the glass of the liquefier passed u i d e r atnios- pheric pressure to the gasholder.The heliurn was sccii boiling q uie t 1 y . In order to follow the evaporation of helium at reduced149 pressurc, it was only necessary to close the coiincction to the gasholder and to put into action the compressors now acting as vactmm pumps. The capacity of the pumps appeared to he grcat enough to remove the vapour so quickly that the temperature decreased considerably .and, when their action was supported by a powerful vacuum pump, it was found that the pressure conld decrease eveii to one ceiitimetre, the helium still not ceasing to b i b a liquid.At that time it was not possible to go further, as only an extemporized coiiiiectioii to the said p o ~ ~ w f u l vacuum pump conld lie established. I n 1909 the cxperimciit was repeated after proper prcparation to this end ; the pump mentioned, a RURKHARDT pump with (7, capacity of 360 &I2. per hour, could develop its full power. Then it appeared that, c\-eii when the pressurc decreased to two mms., the helium did not become solid. Though, at this Ion- temperature, the liquid lost its peculiarity of having its surface standing sharply defined, like the cdgc of a knife, against the wall and now showred the ordinary propcrties of capillarity, yet its striking mobility remained. With regard to the obtaining of solid helium, this rcsult was disappointing. However, it appeared from this result that the rcgioii of tcrnperatures i n which the properties of substances could be investigated by nieans of baths of liquid hclinin estciidcd farther than might have been hoped for by analogy with the other gases of low critical ternpe- rature, and so f a r the result was gratifying.For it is very difficult to obtain by rncaiis of some definite substance constant and homogeneous temperatures below the melting point of that substance. With helium the difficulties ~vould be so great that the temperature of solidification might be considered a limit, below which it could not be nsefully employed. Therefore in absence of a more volatile substance a limit would here have been put to science.Each time, when with further lowering of the vapour pressure it is found that thc helium remains liquid, this failure wit11 regard to the solidification of helium ineans a gain: a new region of tcmperatnre, which on account of its extreme situation is especially important, has proved to be accessible to us.150 4 3. Improvements in the helium-cryostat and cycle '). - Happily it has been possible to obtain a considerably lower temperature than has been mentioned above and we have arrived near the point, at which the low density of the vapours brings in a new limit, below which we caiiiiot descend. To day I will deal with the cxperinients by which this lowering of the lrnown limit of thc liquid condition of hcliurn has been obl ained. This advance, however, has gone on gradually hand in hand with the regula~ developnieiits of the laboratory and the last steps have only been possible after considerable improvement in various directions of the appliaiices for research in the region of helium temperatures.An important impyovement was obtained when we succeeded in transferring the liquid helium from the apparatus in which it had been liquefied to a cryostat, in which like other liquefied gases it can scrve as a bath for definite temperatures in the ordinary way. The cryostat C is still closely connected t o the liquefier L (comp. fig. 2) but the room in the cryostat that is available for experiiiients is iiow no longer blocked a t thc top by thc regenerator-spiral as in thc first appam- tus.The necessary appliances can be brought freely from the top o into the cryostat (comp. fig. 4b), whilst the bath surrounding the apparatus may be obtained by syphoiiing the helium from the liquefier into the cryostat. Such cryostats have been used already in several researches on the properties of substances at very low temperatures, especially in the domain of electricity and magnetism. Fig. 3 2, shows the apparatus together with a diagram of the improved helium circulation, in which such a cryostat has been inserted. The helium, that liquefies on leaving the nozzle k of the regenerator-spiral, collects in the bottom part of the lique- fier vacuum glass as in the first liqucfactioir experiments. This glass, however, is not closed at the bottom but ends in a double walled silvered syphon tube s (fig.2 ) , provided with a valve v. I ) I gratefully acknowledge licre a considerable augmentation of the stock of helium by the. splendid gifts of the American Navy, 30 Ma., and of Prof. Mc LENNAN, 6 MS. 2) Partly scliematical, partly with thc single objects to scale.151 When this valve is kept tight, a supply of liquid helium can he col- lected in the liquefier glass. By opening the valve (comp. v0, fig. 2 and fig. 5) liquid helium can pass to the actual cryostat. This cryostat itself consists of an unsilvered vacuum glass, in which the liquid helium is collected. This glass is surrounded by a second C S ' vacuum glass. I n silvering it two vertical slits (comp. fig. 5, p ) at opposite sides were been left clear to render visible what happens in the helium glass.The second vacuuni glass is surrounded by a third one, silvered in the same way and con- taining liquid air. The vapours arising from the cryostat and VOL. XVIIT-T3*152 i those coming from the liquefier are received in the pumps PI, P (fig. 3) used for the following purpose ; when the circulation is in action these vapours are driven hack by the pumps to the liquefier, there to renew the supply of liquid helium. With the arrangement set u p the process of liquefying helium may be continued, whilst the actual experiments are performed with G Fig. 3. the aid of the helium bath in the cryostat. As soon as too much helium has evaporated from the cryostat, the supply in the liquefier may be used to provide again the desired quantity.The iinprovement in dealing with liquid helium, obtained by the separation of cryostat and liquefier, was accompanied by an improvement of the liquefier itself that led t o a more rapid preparation of liquid helium and a t the same time a more economical use of liquid hydrogen. The main part of the modi- fication consists of a mom efficient USC of the available cold.153 The hcliurn, after being coimpressed to 30 atniospheres is divided (fig. 3) between two spirals b and b, that unite again, bring the helium through a tube containing charcoal cooled in liquid air and are again divided into c and c,. b and c are cooled by cold hydrogen vapour arising from the liquid hydrogen in the refrige- rator, b’ and c’ by cold helinm vapour arising irom the liquefier glass.Both spirals are much longer than in the first liquefier; they unite to form the spiral n and pass, just as with the first liquefier, after cooliiig bj- hydrogen vapour surrounding a and by liquid hydrogen surrounding e, into thc rcgeiierator-spiral f , which is provided with a nozzle k I). Besides obtaining in this way by better regeiieratioii larger efficiency of the liquefactor, the capacity of the circulation has been increased by insertion of larger compressors P, PI, which at the same titme can act as vacuum pumps. So in the new cryostats one had for hours a t one’s disposal a bath of say 500 em3. of liquid helium evaporating at a pressure of 3 mms. It was possible to perform extended experiments a t temperatures whose attainment could hardly have been demonstrated in the early experiments with liquid helium.Such a cryostat with a bath of already extremely low temperature has been used with nincli advantage in the experiments oil cooling helium still further by its own evaporation. The vapori- satioii apparatus (fig. 4), in which helium has been reduced t o the lowest teniperaturc that has beeii attained, consists maiiily of a small double-walled vacuum flask a (fig, 4u), containing the helium when it is cooled as much as possible, and of a wide outlet tube 2, for the gas formed by the vaporisation. The vaporisation flask is immersed (fig. 4b) in the iiiteiisely cooled helium bath of the cryostat C ; the gas escaping from the outlet b leaves the cryostat through the lid 72 which is connected to powerful vacuum pumps I.’ (fig.5). I t is evident, that those pumps not only must cause a high vacuum, but at the same time must have an extremely high capacity at the pressure of this vacuum, since the gas foriiied by the vaporisation will occupy a large volume at atmospheric temperature and a t the low vaporisation pressure. ’) For further particulars see Leiden Comm. No. 158 and Suppl. NO. 45.154 X a I z b C Fig. 4.Similarly it is seen at once, that tlic aim can l)c attained only if the cryostat affords opportunity to these large volumes of gas to escape by a wide outlet, as indeed it did in the apparatus used. If the tubes in which the gas moves are not very wide or are not a t very low ‘temperatures, the movement of the gas at tlie low pressures under consideration requires differences of prcssurc of the same order as that at which the gas is pumped away.These differences of pressure might form a coiisidcrable part of the pressure a t which the helinni evaporates. § 4. Temporary set-back. - These two coiisideratioiis show alrcady that great cleiiiaiids must bc made OH the appliances to be used for the experiments and cspecially on the capacity of the vacuum pumps, if descent to a very low pressure is aimed at. h’ow in 1910 an experiment had heen performed i n this direction on the same principle as has been followed now, but with iiisufficicnt appliances. Notwit hstandbig that the crayosta t was deficiently conceivecl, 11s a fortunate accident a lower limit for the vapour pressure was I-eached than that obtained in 1909.Though this result could not be obtained again on repetition of the experiment with apparatus arranged in the same way as wlieii the expciiment succedod, pet t liere was left no doubt that the yapour pressure could dcscciid far below 2,2 mnm without solidification of the helium and that cvcn at a vapour prcssurc of 0,2 mm. it would probably not solidify. This meant, however, that in eider to ensure a quite certain progress? much higher doniaiids had to be mrzdc on the expcrinieiits than when one stood only at 2,2 rims. as the loivcst limit. Thus niaiiy years elapsed before oiie could tliiiilr of lowering the limit below that acciden- tally found as 0,2 mm. and t,liis problem had t o be set aside and tlie treatment of various problems, more important f o r the moment mid ‘more in accordance with the gradual develop- nient of the appliances was first proceeded with; e.g. those concerning the superconductirit!- and the threshold value of the magnetic field by which ordinary i*esistance is generated i n super- conductors. The lowest temperature obtained reilzaiiied that which corresponds to 0,2 mm. v-apxw-pressure. I l i d estiiuatecl this156 temperature at l0,15 K. Taking into account the uncertainty of this estimation it would have been better to have said that in descending one had upproached t o rtarZy lo K. As the state of the helium work progressed it became more and more necessary that a limit for the vapour-pressure more in accordalzce with its new feature than 0,2 mm. had to be given and especially i t had to be established whether it would be possible to penctrate below lo K.At last liowevcr in 1919, this problem could he attacked, when the difficulties of the years of war and of crisis had been overcome. 0 5 . New attack. - F o r the removal of the helium from the evaporation apparatus a large BURCKHARDT vacuum pump VBi of 360 M3. per hour capacity could then be used, coupled in series with one of 18 M3. capacity T7B2 and a Srmms pump 17, of 2 &!I3. capacity. The arrangement and treatment of the large high- vacuum pump VBi were such, that there was no possibility of gas being given of€ from the lubricating oil into the vacuum. At the beginning the valves, which previously had been kept shut by means of springs until the gas by its own overpressure flowed from the pressure side of the cylinder into the outlet, were now opened and shut by a mechanical arrangement at the correct moment for the equalising of the pressure in both spaces.Later on the valves on the pressure side were taken away and the latter connected simply to the suction side of the auxiliary pump. With this pumping arrangement which had a limiting suction pressure of 0,04 mm. (in the best case 0,025 mm.) and with an evaporation glass which, though not so good as but in the main set up like that which served for tlic expcrimcnts which I shall describe, a volume of 2,'i litres of gas per hour (measured at N. T. P.) could be removed and thc pressure of suction a t the top of the cryostat reduced to 0,l mm. It was concluded that the vaporisation presszcre was again smaller than had been obtained in 1910, perhaps it may he estimated to have been reduced t o 0.15 mm.When the pressure had fallen to this value, therc occurred 110 further change in the evaporation, equilibrium between the heat received and the cooling by evaporation being attained.157 The heliurtl again did not solidify and tlic limit for the pressure liad again been somewhat lowerecl, whicli again led to higher demands 011 later experiments. $ 6. Battery of condensation vacuum pumps. -- Real progress could be made only by executing a long cherished i t v, I .,* I P c~ M 4 plan, namely : the construction of a vaeuuiii pump-complex of veyy large capacity with extremely low suction pressure. This was intended t o consist of a large number of LANGNUIR conden- sation punips connected in parallel.In 1920 the first step was inade in the realisatioii of this 1)laii and since then the lmttery VOL. SVIII-T8*-157155 of pumps has been regularly eiilarged l). In the esperimciits now described the battery had already gron-11 (fig. 5) into Fig. 6. one of 12 glass (V,) arid 6 iron (V,) LANGMUIR pump, coiinected I) My best thanks a r c duo t o P m L I P s ’ Ziicaiiclcscciit Lain11 Works, Eiiid- hoven, f o r prescnting us, at thc beginning of our attciupts, with ii iiuniber of theso pumps.159 in parallel into one single complex. The EURCKHARDT pumps connected as before in series served as the auxiliary pump. This is shown diagrammatically i n fig. 5 I). The battery of twelve glass pumps (see fig.6) consists of three series of four pumps, each series having (as later appeared super- fluous) a LANGMUIR pump as auxiliary (fig. 5) before being con- nected to the actual auxiliary consisting of the complex of BURCKHARDT pumps. I n this battery the mercury of the conden- sation pumps is heated by means of gas. Since flames are not permissible in such a room as the department for helium experi- ments, the battery is built into a small separate room in the department. The room containing the battery is ventilated with outside air by means of a fan, which maintains a blast of air through the nicks in the walls into the helium department 2). All connections to be found on the glass pumps mere made by sealing together. The different series were cemented to the copper suction-tubes which united in a main D, of twelve crxis diameter.As far as the copper tubes were not soldered to each other, they were cemented together just as the main is to the cap of the evaporation apparatus. The only connection by rubber was that of the tubes coming from the auxiliary LASG~~KIIZ pumps to the suction tube of the Burckhardt V b l . Ii’urther, all the copper used is varnished. In general the iron pumps are not so suitable for the highest vacuum as the glass ones, but they are quite good enough t o procure the high vacua with which we are concerned herc. They are heated elect rically and attention must be paid continually i n order that one of the cemented connectioiis does not give way. But I will not further dwell 011 these details, nor 011 the adver- sities and disappointments which necessarily accompanied tlie experiments which I am describing now, before they were concluded satisfactorily :!).’) Partly schematical, partly with tlie single objects to scale. One should 2) I n case of emergency all flames can be put out at once. 3) We mnst point oat that, if one of the p u m p suddenly burst, a large quantity of air wliould have t o be received by the BUECKHARDT-pump. Besides take into account the historical development.By the aid of the battery with which I am dealing now, on a suction-pre,wure of 0,005 mm. at the evaporation apparatus, a capacity of one litre (at N. T. P.) of removed gas per hour was finally obtained, corresponding with an evaporated quantity of 1,25 em3. per hour of liquid helium at 2 O K.0 7. Minimising the heat conveyed to the evaporating helium. - The very large capacity of the pumping arrangement may be utilised the better, the more the helium which has to be vapo- rised is protected against conveyance of heat and the more the frictional resistances between the evaporating helium surface and the pumps are diminished. Both requirements are, however, very difficult to combine in the construction of the evaporation apparatus: a wider outlet tube will diminish the frictional resistance, but at the same time it will give rise to a larger heat conveyance through the glass walls and especially by conduction in the column of gaseous helium in the tube. Both requirements have been satisfied as much as possible in the construction of the evaporation apparatus showm in fig.4, which was used in the last experiments in 1920 and 1921. (For the meaning of some details of the construction see also 5 5 9 and 10). Besides the small evaporation flask n with double-walled evacuated interspace and the similarly double-walled outlet tube b , the interspace of which can be evacuated through a tap I ) , we notice a single-walled part g . Advantage is taken of the latter to introduce, by means of an artifice, liquid helium into the evaporation-flask. To bring this about, helium is introduced into the helium space connected with the high-vacuum pumps, i. e. the evaporation apparatus, the connecting tube to the pumps and the pumps themselves, the battery of condensation pumps and the auxiliary pumps being out of action.The pressure is thus allowed to rise above the vapour pressure occurring in the bath of the cryostat. Inside, the liquid helium then flows down along the walls and so fills the bottom part of the apparatus. By again putting into safeties to the Burckhardt itself a shutting-slide has been placed in the suction-tube. This slide could be shut on the first signal. I) Any small lowerings of the vacuum consequent on the presence of soldered metal can be remedied by these means whenever necessary.161 ac&ion the auxiliary pumps evaporation under decreasing pressure is brought about until only the required quantity of liquid helium is present in the evaporation glass, in which, protected against heat absorption, it cools further until, to go lower, the high-vacuum pumps have to be put into action and the actual experiment can be commenced.The pressure under which suction occurs a t the cap of the evaporation apparatus is measured with a McLeod manometer H, fig. 5. Special care has been taken to prevent heat penetrating by conduction or by radi- ation into the helium in the evaporation flask. It is necessary to pay attention also to the radiation given out by parts of the appa- ratus remaining at ordinary temperatures, e. g. that coming from the cbp above the outlet tube through which the evaporised helium is led away. The order of magnitude of this radiation may be estimated by comparison with the' total black radiation from a plane surface (4,s. 10-9. T4 gram-calories per hour and per em".). Substituting the value of the ordinary temperature in this expression one arrives at 30 gr.cal., which, on account of thc small heat of vaporisation of helium, about 6 cal., suffices to vapo- rise a quantity of liquid which occupies some 30 litres at N. T. P. when in the gaseous state. The pump-complex is, however, calcula- ted to remove at thr pressures under consideration only a quantity of gas corresponding to one litre at N. T. P. per hour. The radia- tion towards thc evaporation flask should be received as much as possible by opaque metallic screens which are cooled down to low temperatures, preferably to the temperature of the helium bath in the cryostat. The radiation from screens cooled to this degree, on accouiit of its dependency on thc fourth power of the tempe- rature, is so small as to be negligible.Protection against the radia- tion falling sideways on the walls of the cvaporation flask may bc especially easily realised : the entire lower part of the evaporation glass is surrounded by a metal bowl, the upper side of which extends beyond the surface of the liquid helium in the crj-ostat. I n this bowl two slits are left open in order to render the evaporation flask visible through the unsilverecl strips of the vaciiiini glasses. As a rule the slits are kept shut by means of two screens, which may be rotated around the bowl. They are removed only when the level of the liquid mnst be observed;162 for illumination, use is made of a metal filament lamp, placed behind an alum solution. In order to shut off heat conveyance from above by radiation into the evaporation glass an arrangement has been constructed, by which the whole - made by Mr.KESSELRING, chief of the glassblower department - became a masterpiece of the glassblower’s art. Above the evaporation flask a there is sealed into the glass a double-walled cap x, the interspace of which is connected to the outside of the vacuum tube. The helium in the cryostat flows through the annular space of the cap, the upper side of which is blackened and the lower silvered. Radi- ation from above can only penetrate by reflection along the walls of this cap. The heat transference from above has been further minimised by narrowing down the single-walled middle part as much as the strength of the apparatus and the quantity of escaping vapour would allow.Further, screens ZJ cooled by the ascending gas and by other means to which we will refer later, were so placed in the outlet, that they did not hinder the free escape of the vapour. The inner walls of the outlet were blackened with a mixture of soot and celluloid solution in order t o diminish $he reflecting power. Finally, there serves to the same end the spiral t7p that has been introduced at the top, through which liquid hydrogen is forced and which removes part of the heat which otherwise would have been conducted below by the walls. In another respect also advan- tage has been taken of the glassblower’s art in the construction of this evaporation glass. As we have seen, the outlet tube has also a double-walled upper part, silvered and evacuated between the double walls.The stresses which originate on account of the great difference of temperature between the inner and outer walls, are taken by a metal case n soldered to the glass and serving as a spring. The heat transferred from above to the lower parts through the walls across the narrowing is taken up by the bath of the cryostat I), the level of the liquid in the cryostat being always kept above the narrow portion g . I n this ‘) The heat thus deviated to the bath has no perceptible influence on the rate of evaporation and hence on the time during which the experiment can be continued.way it is tried t o ciisurc that the temperature of the helium gas above the evaporation flask is iiot appreciably higher than that of the bath.The neck of the evaporation flask in which the evaporation under very low pressure occurs was long and iiarrow, firstly to reduce the heat condnction through the glass as much as possible, the small radius allowing the inner wall to be made cry thin and secondly to make, the velocity, with it-hich the vapour escapes, gryat enough to carry away the heat which otherwise would enter by conduction along the column of helium gas. In all this attention has been paid to the limit at which disadvantageous frictional resistances would make their ap- pearance. All the above iriciitioiied precautions having been taken t o protect the helium in the evaporation flask as much cts possiblc against heat transference, the Traporisat ion appeared t o have been reduced to 0,9 litres of normally measured gas.The suction-pressure produced by the high-vacuum pumps at the cap of the evaporation apparatus was slio~vii to bc 0,0055 mm. by the McImd-gauge. In the hope of reducing the quantity of glass which finally had to be cooled down by the evaporating helium, an especially thin-walled small Vacuum glass c was placed at the bottom of the evaporation f layk It was thought that. on continued pumping the liquid in thc small glass would continue to evaporate after the liquid surroundiiig the glass had evaporated, so that the circumstances for cooling the helium would become more favourable, as less glass had to be cooled and less heat would be conveyed along the wall. It will be seen presently that peculiarities in the course of the evaporation brought it about that the liquid level inside and outside the small glass sank at the same rate.As regards further progress in reaching lower temperatures, the bowl did not fulfil expectations. $ 8. Minimising the frictional resistance on the way from evaporating helium to pumps. - I will 1101~ proceed to discuss the rriinimisiiig of the frictional resistance of the vaporised heliiun in the evaporation apparatus on its way from the liquid surface164 t o the high-vacuum punips and hence using the low-suction pressure of the pumps to the best advantage at the evaporating surface. The width of the outlet tube could be increased only as far as the size of the top of the cryostat o (comp. fig. 4 b ) allowed. The opening in the latter could not be widened without rebuilding the whole cryostat I).With the greatest width a t present available for the outlet tube, the gas would, as a consequence of receiving heat in ascending, acquire so great a velocity, that a frictional resis- tance rnnch to large for our experiments would occur, unless special incans were taken to prevent this. For this purpose there has been placed in the top of the outlet tube a lining that can be cooled strongly by external means. This lining consists of the copper spiral Sp through which liquid hydrogen can be led. The spiral is connected to a Dewar flask of liquid hydrogen and the liquid is forced under a sniall overpressure through the former ; the supply being regulated by a flow-meter showing the quantity of evaporised hydrogen. Not only does this lining, through being thus cooled, reduce considerably the heating of the vapour in its ascent through the outlet tube and contribute to the prevention of radiation by the cooling of the various screens, but it also takes up, as we have said, part of the heat penetrating from above along the glass walls of the evaporation glass.By means of a small resistance thernioniet er placed undw the lowest turn of the coil it can be ascertained whether the arrangement is working properly; in the experiments that suweeded best the tempe- rature underneath the spiral decreased to - 2 O O O C, Then the loss of pressure due to frictional resistance, as we shall see, is reduced to only 0,Ol niiii. 0 9. Determination of pressure. - I n determining the pressure in the space immediately above the level of the evaporating helium use was made of a resistance manometer.Pressures such as those which occur above the surface of the evaporating helium arc too small t o be measured by iiieaiis of a suitable mercury ') The extensive work acccssary f o r thc construction of a €urtiier cryostat with a wider opening at the top is in hand.manometer except with a yery rigid arrangement such as would be very troublesome in this case. From this point of view a resistance rnanometer is already preferable. The nianoineter tube further may have very small dimensions and the manometer can he calibrated very well for pressures between 5 and 20 bars. What- ever manometer is used, if the space of the manometer is a t ordinary temperatures and is connected to the space a t the lower temperature by a narrow tube, the pressure in the inanonietei+ space will not be equal to that which has to be measured. *It the low pressure a t which the helium evaporates, the mean free path of the gas molecules, except where the connecting tube is a t a, w r y low temperature. is probably many times greater than the diameter of the tubc.Between the space at low and that at ordinary temperatures there occurs a pressure difference equal to the thermal molecular pressurc. The hope I) that this diffi- d t y inhereii t in measuring pressures a t low temperatures roulcl be avoided 11-y thc iisc of the resistance manometer, as it was here possible to keep the manometer tube itself a t a low tcinpcraturc, has happily been confirmed by a series of experiments carried out in collaboration with Mr.VAN GULIK. So the pressure below in the eyaporation flask has been deter- mined by means o€ a resistance inanonieter, the manometc~ tube of which was kept a t a temperature but little above that of the evaporat iiig helium by being immersed in the helium bath outside the c\-nporation flask. In fig. 4b and c is shown how the mano- iiicter tube is sealed to the lower part of the evaporatioll glass; in iig. 5 the arrangeinelits W for the measuying of the resistance arc sliowii t~iagrrammatirally. A t first sight it seems doubtful whether the principle on which the resistance manometer is based, yiz. the change o i resistance of the manometer wire with tempera- ture, can be applied a t a temperature so low that the resistance of the wire has not only fallen to a very small value but also does not change with temperature, as is the case with a platinum wire at the temperature of liquid helium.Hut it appears that the resistance remaining a t this temperature, viz. the additive resistance, is still sufficient (if a still serviceable small current flows through thc ’) Cf. Leiden Comm. Suppl. No. 34n.166 wire) to heat the wirc until the temperature is reached at which the resistance begins to increase distinctly and so the influence of the pressure on the loss of heat from the wire becomes obscr- vable through the difference of current necessary to rnaiiitain the same resistance. Though the measuring apparat us, if used in this way, becomes a means rather of indicating than of mea- suring the pressure, yet by calibration a t known pressures the desired aim is attained.Iii this calibration the apparatus was filled with gaseous helium at rest and with its lower end immersed in the same way as in the experiments in tlie helium bath. The upper part protruding into the cap of the cryostat reniained at ordinary temperatures, The tube 2, is not wide enough to render quite superfluous a correction for the thermal inolecular pressure between the upper and lower parts. The accuracy 01 the values of the pressures, which will be given presently, will be increased therefore, when the uncertainty resulting iroiri the fact that the correction has as yet only been caEciclutetZ, will be removed by new experiments. I will not dwell, however, at present on this correction, which is only 0,003 mm.Also we iieglect the difference which niay still exist between the prcssurc in the manometer tube and that at the surface of the helium itself. $ 10. Stirring arrangement. - Lastly we have t o mention the small stirrer r introduced into the evaporation flask a. With the arrangement of the evaporation glass shown in fig, 4 it consisted of a horizontal glass disk attached to a vertical glass rod. It can be moved up and down by means of a wire x , attached to a rod, which passes a glass tube in the cap; the tube is closed by means of a packing gland l). § 11. The final experiments. - For the succe,ss of the experi- mcnt with the cornplicat,ed arrangement which has been described ') I n an earlier experiment a spring was inserted bctwccn the wire and tlie rod.A diamond was suspeiided from the wire instead of tlic disk. I f tlic liclium had solidified and the diamond had cncountercd resistanw, the spring would have bccn stretchcil in its inovemcnt up and down.167 (and of which fig. 7 gives a view), it is necessary that numerous operations should be performed each in the time allotted to it and in regular order, the success of each of the operations them- selves depending upon careful preparation. A small amount of coiideiisatioii o n one of the glass walls through which the evaporation of the helium must be watched suffices to render observation impossible : on walls cooled in liquid hydrogen a gas rendered impure by traces of air gives coii- densation. On reflecting on what is required t o keep the glass walls through which it is necessary to see, perfectly transparent for hours after liquid lielium has been introduced f o r Fig.7. the first time into the cryostat, it will be understood that I am greatly indebted to Mr. FLIM, the chief of the technical depart- ment, for his devotion to the work. Thanks to him everything went according to plan. Early in the morning the preparation of 24 litres of liquid hydrogen was commenced, the previous day having been spent, on the one hand, in evacuating the apparatus and further putting it in working order a i d , on the other hand, in preparing a snf- ficient quantity (more than 50 litres) of liquid air. l'v1eanwhile the v o L . x~III--'~ s *-I 67following preparations were carried out : the helium circulatioii was further put in order; the pump, which had to remove t h e hydrogen from the helium liquefier at reduced pressure, was started; the space reserved for the liquid air used in cooling the hydrogen was next filled and the liquid hydrogen space filled, after having been first cooled with dry cold hydrogen gas.At 12 o’clock the liquid helium could be spphoned over into the cryostat? after which we proceeded to cool this bath further by evaporation and refilling by means of the helium circulation. At 1 o’dock the condensation of the helium into the evaporation apparatus could be commenced and the bottom part of the eva- poration glass was filled up t o somewhat above the double-walled cap mentioned earlier. At about 3 o’clock this helium had eva- porated so far as t o occupy only the lower part of the evaporation flask, the evaporation taking place first under the action of the auxiliary pump complex, later on under that of the combined high-vacuum and auxiliary pumps, which serve for the removal of the helium from the evaporation apparatus.The evaporation was further observed alternately with the naked eye and with the telescope of a cathetometer, the screens around the evaporation flask being kept shut as long as possible. Neither by means of the stirrer nor with the naked eye or with the telescope could anything be observed that pointed to the solidification of the helium even at the lowest vapour pressure observed ; the liquid retained its great mobility throughout. $ 12. Evaporation at different levels.- Further, it was observed that contrary to the expectation that the layer outside the small glass would evaporate first and then the helium inside the glms, both liquid levels fell at the same rate, so that they remained in the same horizontal plane. If (comp. fig. 8) by means of the lid shaped stirrer r (fig. 4a) liquid was thrown from the inside t o the outside, the outside level fell rapidly while that of the liquid inside rose until they were again in the same plane. If, by removing the screens and allowing the radiation from a lamp to fall on the evaporation flask, the outer layer was caused to evaporate, after turning the screens the outer layer was re-formed a t the expense of169 the helium within the sniall glass, and increased until both levels were a t the same height, after which both again fell at the same rate.The speed of readjust- ment by this distillation was striking. A correct judgement on this pheiiomeiioii will only become possible, when the deterniinations we haye in view concer- ning the heat conductivity of glass, of helium vapour and of liquid lieliuiii liave been carried out, whilst a knowledge of t h e latent heat of evaporation axid of the specific heat of liquid helium and of glass and of the Yiscosity of gaseous helinm is also desirable. The property of a maxiniuni density shown by helium has of coursc great influence 011 the obserred evaporation. Observations in 1911 had brought this property to light, but it n-as not suffi- Fig. 8. c+iently established whether the density approached a limi- ting value or whether it decreased a t still lower temperatures.That the latter is the case has been established by a repetition of the experiments undertaken in collaboration with Mr. BOKS after the completion of the experiments with which we are dealing now. This confirrnatioii holds only so far as the possibility of some peculiarity in thc expansion of glass is excluded. On cooling the surface of helium below 2 , 2 O K., the coldest layers of the liquid remain at the top. While in other cases in working with baths at reduced pressure care has been taken to stir vigorously, in this case stirring has been omitted, as this would have made a claim on the already small amount of available space at the top of the eryostat. The presence of a stirrer in the outside bath would probably have made the heat transference to the evaporation flask still smaller than it actually was.It was hoped that the imam which have been applied would have reduced the heat transference to one half of what it seemed to be assuming that there is no special change in the latent heat of vaporisation.170 3 13. Lowest limit of evaporation pressure. - However this may be, when the level of the evaporating helium had fallen t o the bottom of the narrowest part of the evaporation flask and to about half the height in the small glass, it appeared that the lowest rapour pressure was reached that could be obtained with this apparatus. Neglecting the small corrections mentioned above, the pressure a t the surface was 0,012 to 0,014 mm., mean 0,013 mm.In the cap there was a vacuum of in the mean 0,005 mm. The pressure difference due to the friction of the rapidly esca- ping vapours of low density was in the mean 0,008 mrn. This value agrees sufficiently well with the result of a special deter- mination of the frictional pressure experienced by helium moving through the apparatus with the same velocity. In this control experiment the evaporation flask was substituted by a tube through which helium, cooled to liquid hydrogen temperature and of the same density as in the actual experiment, flowed with the same velocity. This control experiment gave 0,009 mm. The observed frictional pressure also agreed with an estimate based on the probable distribution of temperaturc along the column of ascending helium.I n round numbers and allowing for the exis- ting uncertainties we may say that the limit for the evaporation pressure has been brought below 1/50 mni. and that we have progressed ten times as far as in the experiments of 1910 on which was based the estimate of the temperature mentioned as the lowest one then reached. In temperature difference as measured by K&vm degrees this means, as we will see in 4 14, only a very small range. Returning to the question of the solidification of helium we come to the following conclusion: as there is provisionally no doubt that helium has a maximum density (see $ 12) and as it is even not solidified a t a temperature below the half of that of the maximum density we cannot escape the question whether helium will not remain perhaps liquid even if it is cooled t o t h e absolute zero.4 14. Determinations of temperature. - It still remains to consider the question of what temperature corresponds to the evaporation pressure found. For the latter pressure an experi-171 niciitally established value (mi as wc have done be given. apart from a few pyohahly small corrections which demand further study; tlie same cannot be said of the temperature. In the deter- mination of tlie extremely low temperatures, a t which even the L- I,,,,? Fig. 9. lieliurii gas thermometer can iio 1 oiiger be use€ullj- employed, we ciit CL' a rcgioii the study of which has oiily comincixcd. T l i e i ~ f o ~ c we fall lm:k on the law of corresponding states of v m DEK WAAI~S to guidc us now in extrapolating the law con-172 iiecting temperature than those for which latter has been done and vapour pressure to lower temperatures it has been established experimentally. The f o r the range in wliicli it has becii possiblc till iiow to use the lieliuni gas thcrrriorneter.In this "ase the constnxtion of the thermomet cr had t o be adapted to the pressure, which introduced complications. In fact the pressure must be very low if the simple gas laws are to be appli- cable and even if only liquefaction is to be prevented. When the pressure becomes small, the correction for the thermal niole- cular pressure (the pressure diff ereiice between the theimo- meter reservoir a t low and the manometer space at ordinary tcmperatxre) in this caw also has to be applied, But notwith- standing thme difficultics we succeeded in measuring vapour pressures of helium down to 1 , 5 O K., in 1911 and 1913 with a thermometer with a mercury micromanometer and in 1917 with two thermometers with hot wire manometers.The results are plotted in the accompanying fig. 9, in which the abscissae are the recipro- cals of the reduced temperatures and the ordinates the logarithms of the reduced pressures. The point is how to extrapolate the line which passes through the observed points. For this purpose the vapour pressure curves of ether, mercury, argon, neon and hydrogen are shown on the same graph eliding in the triple-point of each substance. That for mercury has a still somewhat lower reduced temperature than has heen reached with helium.All the curves agree in that the curvature is only very small; the greatest curvature with helium occurs at the higher temperatures. They deviate from each other in that the slope differs for the various substances. In the application of the law of corresponding states to normal substances account has to be taken of such systematic changes of the parameters of the laws for liquids: for substances with low critical point the slope of the line in the diagram decreases as the critical temperature decreases. I n consideration of this it will be seen from the figure that helium satisfies this law in its generalised form ; in particular the slope is compatiblc with that of Ar, A7e and H , the small curvaturc a t higher reduced tempera- tures also falling into line.An extrapolation that can be accepted as a probable one, is that which is obtained by assuming the taiigctnt t o the cxpcrinieiital CUITC a t the point where this curveends, as its continuation. This has been drawn in the figure. It gives for the temperature correspoiidiiig to the limit of the eva- poration pressure reached in 1910 the value l0,15 K. and f o r that dealt with liere, which is the lowest ternpersturc yet attained, 0°,82 K, Taking iiit o account the uncertainty of the extrapolation it will be better to say that the lowcst temperattire yet attaijied is sonte h m d w d t h s of Q: degree below 0.O9 K. 0 15. Conclusion. - The qucstioii put above, wlictlicr we could desceiid below lo IC., is answered positively by this result. In round iiuinbcrs we have pi*ogi~essed of a degree a i d oiie may say that if we could have gone further oiily I//, of a degree, we sho~zld have arrived a t the h i t obtainable in the ordinary way with lieli~un. A better idea than given by these small numbers is obtained perhaps, wheii ~ v e express the lowering of temperature by the ratio in which we have decreased the absolute temperature. While the passing from ordinary tempc- rature to that. of heliurii evaporating at 0,2 m m . nieaiis 3 \owering in the ratio 250 t o 1 and from the rrieltiiig point of hydrogen to the heliuin tcmperatnre mentioned, one of 13 to 1, the present 1o~veriiig is only oiic of 1,4 to 1 and a further reduction of 1,2 to 1 would be nearly the limit of what could be doiie with liquid heliuni. If it is considered that, our Irnow- ledge of atomic stmcture reiiclers iiiiprolxhle that another substance could be discomred, o r obtained in another way, rnore volatile than lieliuiii, then the limit iiiilicateci, from which we are separated by 0711y such a sriinll aiiio~mt, would seem an absolute one sct to us i n the obtaining of yet lower tempera- tures. We caiiiiot accept such a liinit otlienvise than as a pwwisioiial one. There am even now def iiiite problems which require to be treated in the domain beyond thc seemingly impenetrable barrier. A simple example is the question whether a metal such as gold caii be made snpciwmductive by cooliiig it more than we have been able to do, This kind of problem reminds 11s of the problem of tlie liquefying of the periiiaiient gases. They with- stood the effoiats of the great clrpei~iniciitci- whose glorions mine is174 attached to your Socicty. Half a century later thc liqucfaction of hydrogen, the most incoGrcible gas with which FARADAY had operated was the brilliant achievement of the latest of his successors in office a t the Royal Institution: Sir JAMES DEWAX We may feel sure that the difficulty which has now arisen in our way will be overcome also and that the first thing needed is long and patient investigation of thc properties of matter at the lowest temperature we can reach.

ISSN:0014-7672    

DOI:10.1039/TF9221800145

出版商:RSC    

年代:1922

数据来源: RSC

摘要:

118 ELECTRICAL THEORY OF ADBORPTTON The writer considers the double layer as consisting of a swface of rigidly fixed atoms under continuous bombardment of positively and negatively charged ions, any particular point on the rigid surface becoming in turn negative, neutral and positive, these conditions arisdg in any order. The observed contact difference is the average effect of these conditions. Where several kinds of atoms are present in the solution the average number of any one of them at the surface will depend on their concentbration, valency and mobility. The variation of contact Werence from negative to neutral and positive was observed with cotton and aluminium sulphate near the neutral point. These variations occurred during the same experiment, the readings being direct measurements of E.1I.F.s developed by filtration under pressure.This point would be covered by putting n2 = 1 and = 2 or 3 in Mukherjee’s equation No. 13.118 ELECTRICAL THEORY OF ADBORPTTON The writer considers the double layer as consisting of a swface of rigidly fixed atoms under continuous bombardment of positively and negatively charged ions, any particular point on the rigid surface becoming in turn negative, neutral and positive, these conditions arisdg in any order. The observed contact difference is the average effect of these conditions. Where several kinds of atoms are present in the solution the average number of any one of them at the surface will depend on their concentbration, valency and mobility. The variation of contact Werence from negative to neutral and positive was observed with cotton and aluminium sulphate near the neutral point.These variations occurred during the same experiment, the readings being direct measurements of E.1I.F.s developed by filtration under pressure. This point would be covered by putting n2 = 1 and = 2 or 3 in Mukherjee’s equation No. 13.118 ELECTRICAL THEORY OF ADBORPTTON The writer considers the double layer as consisting of a swface of rigidly fixed atoms under continuous bombardment of positively and negatively charged ions, any particular point on the rigid surface becoming in turn negative, neutral and positive, these conditions arisdg in any order. The observed contact difference is the average effect of these conditions. Where several kinds of atoms are present in the solution the average number of any one of them at the surface will depend on their concentbration, valency and mobility.The variation of contact Werence from negative to neutral and positive was observed with cotton and aluminium sulphate near the neutral point. These variations occurred during the same experiment, the readings being direct measurements of E.1I.F.s developed by filtration under pressure. This point would be covered by putting n2 = 1 and = 2 or 3 in Mukherjee’s equation No. 13.118 ELECTRICAL THEORY OF ADBORPTTON The writer considers the double layer as consisting of a swface of rigidly fixed atoms under continuous bombardment of positively and negatively charged ions, any particular point on the rigid surface becoming in turn negative, neutral and positive, these conditions arisdg in any order.The observed contact difference is the average effect of these conditions. Where several kinds of atoms are present in the solution the average number of any one of them at the surface will depend on their concentbration, valency and mobility. The variation of contact Werence from negative to neutral and positive was observed with cotton and aluminium sulphate near the neutral point. These variations occurred during the same experiment, the readings being direct measurements of E.1I.F.s developed by filtration under pressure. This point would be covered by putting n2 = 1 and = 2 or 3 in Mukherjee’s equation No. 13.118 ELECTRICAL THEORY OF ADBORPTTON The writer considers the double layer as consisting of a swface of rigidly fixed atoms under continuous bombardment of positively and negatively charged ions, any particular point on the rigid surface becoming in turn negative, neutral and positive, these conditions arisdg in any order.The observed contact difference is the average effect of these conditions. Where several kinds of atoms are present in the solution the average number of any one of them at the surface will depend on their concentbration, valency and mobility. The variation of contact Werence from negative to neutral and positive was observed with cotton and aluminium sulphate near the neutral point. These variations occurred during the same experiment, the readings being direct measurements of E.1I.F.s developed by filtration under pressure.This point would be covered by putting n2 = 1 and = 2 or 3 in Mukherjee’s equation No. 13.118 ELECTRICAL THEORY OF ADBORPTTON The writer considers the double layer as consisting of a swface of rigidly fixed atoms under continuous bombardment of positively and negatively charged ions, any particular point on the rigid surface becoming in turn negative, neutral and positive, these conditions arisdg in any order. The observed contact difference is the average effect of these conditions. Where several kinds of atoms are present in the solution the average number of any one of them at the surface will depend on their concentbration, valency and mobility. The variation of contact Werence from negative to neutral and positive was observed with cotton and aluminium sulphate near the neutral point.These variations occurred during the same experiment, the readings being direct measurements of E.1I.F.s developed by filtration under pressure. This point would be covered by putting n2 = 1 and = 2 or 3 in Mukherjee’s equation No. 13.118 ELECTRICAL THEORY OF ADBORPTTON The writer considers the double layer as consisting of a swface of rigidly fixed atoms under continuous bombardment of positively and negatively charged ions, any particular point on the rigid surface becoming in turn negative, neutral and positive, these conditions arisdg in any order. The observed contact difference is the average effect of these conditions. Where several kinds of atoms are present in the solution the average number of any one of them at the surface will depend on their concentbration, valency and mobility.The variation of contact Werence from negative to neutral and positive was observed with cotton and aluminium sulphate near the neutral point. These variations occurred during the same experiment, the readings being direct measurements of E.1I.F.s developed by filtration under pressure. This point would be covered by putting n2 = 1 and = 2 or 3 in Mukherjee’s equation No. 13.118 ELECTRICAL THEORY OF ADBORPTTON The writer considers the double layer as consisting of a swface of rigidly fixed atoms under continuous bombardment of positively and negatively charged ions, any particular point on the rigid surface becoming in turn negative, neutral and positive, these conditions arisdg in any order.The observed contact difference is the average effect of these conditions. Where several kinds of atoms are present in the solution the average number of any one of them at the surface will depend on their concentbration, valency and mobility. The variation of contact Werence from negative to neutral and positive was observed with cotton and aluminium sulphate near the neutral point. These variations occurred during the same experiment, the readings being direct measurements of E.1I.F.s developed by filtration under pressure. This point would be covered by putting n2 = 1 and = 2 or 3 in Mukherjee’s equation No. 13.118 ELECTRICAL THEORY OF ADBORPTTON The writer considers the double layer as consisting of a swface of rigidly fixed atoms under continuous bombardment of positively and negatively charged ions, any particular point on the rigid surface becoming in turn negative, neutral and positive, these conditions arisdg in any order.The observed contact difference is the average effect of these conditions. Where several kinds of atoms are present in the solution the average number of any one of them at the surface will depend on their concentbration, valency and mobility. The variation of contact Werence from negative to neutral and positive was observed with cotton and aluminium sulphate near the neutral point. These variations occurred during the same experiment, the readings being direct measurements of E.1I.F.s developed by filtration under pressure. This point would be covered by putting n2 = 1 and = 2 or 3 in Mukherjee’s equation No.13.118 ELECTRICAL THEORY OF ADBORPTTON The writer considers the double layer as consisting of a swface of rigidly fixed atoms under continuous bombardment of positively and negatively charged ions, any particular point on the rigid surface becoming in turn negative, neutral and positive, these conditions arisdg in any order. The observed contact difference is the average effect of these conditions. Where several kinds of atoms are present in the solution the average number of any one of them at the surface will depend on their concentbration, valency and mobility. The variation of contact Werence from negative to neutral and positive was observed with cotton and aluminium sulphate near the neutral point. These variations occurred during the same experiment, the readings being direct measurements of E.1I.F.s developed by filtration under pressure.This point would be covered by putting n2 = 1 and = 2 or 3 in Mukherjee’s equation No. 13.118 ELECTRICAL THEORY OF ADBORPTTON The writer considers the double layer as consisting of a swface of rigidly fixed atoms under continuous bombardment of positively and negatively charged ions, any particular point on the rigid surface becoming in turn negative, neutral and positive, these conditions arisdg in any order. The observed contact difference is the average effect of these conditions. Where several kinds of atoms are present in the solution the average number of any one of them at the surface will depend on their concentbration, valency and mobility.The variation of contact Werence from negative to neutral and positive was observed with cotton and aluminium sulphate near the neutral point. These variations occurred during the same experiment, the readings being direct measurements of E.1I.F.s developed by filtration under pressure. This point would be covered by putting n2 = 1 and = 2 or 3 in Mukherjee’s equation No. 13.118 ELECTRICAL THEORY OF ADBORPTTON The writer considers the double layer as consisting of a swface of rigidly fixed atoms under continuous bombardment of positively and negatively charged ions, any particular point on the rigid surface becoming in turn negative, neutral and positive, these conditions arisdg in any order. The observed contact difference is the average effect of these conditions.Where several kinds of atoms are present in the solution the average number of any one of them at the surface will depend on their concentbration, valency and mobility. The variation of contact Werence from negative to neutral and positive was observed with cotton and aluminium sulphate near the neutral point. These variations occurred during the same experiment, the readings being direct measurements of E.1I.F.s developed by filtration under pressure. This point would be covered by putting n2 = 1 and = 2 or 3 in Mukherjee’s equation No. 13.118 ELECTRICAL THEORY OF ADBORPTTON The writer considers the double layer as consisting of a swface of rigidly fixed atoms under continuous bombardment of positively and negatively charged ions, any particular point on the rigid surface becoming in turn negative, neutral and positive, these conditions arisdg in any order.The observed contact difference is the average effect of these conditions. Where several kinds of atoms are present in the solution the average number of any one of them at the surface will depend on their concentbration, valency and mobility. The variation of contact Werence from negative to neutral and positive was observed with cotton and aluminium sulphate near the neutral point. These variations occurred during the same experiment, the readings being direct measurements of E.1I.F.s developed by filtration under pressure. This point would be covered by putting n2 = 1 and = 2 or 3 in Mukherjee’s equation No. 13.118 ELECTRICAL THEORY OF ADBORPTTON The writer considers the double layer as consisting of a swface of rigidly fixed atoms under continuous bombardment of positively and negatively charged ions, any particular point on the rigid surface becoming in turn negative, neutral and positive, these conditions arisdg in any order.The observed contact difference is the average effect of these conditions. Where several kinds of atoms are present in the solution the average number of any one of them at the surface will depend on their concentbration, valency and mobility. The variation of contact Werence from negative to neutral and positive was observed with cotton and aluminium sulphate near the neutral point. These variations occurred during the same experiment, the readings being direct measurements of E.1I.F.s developed by filtration under pressure.This point would be covered by putting n2 = 1 and = 2 or 3 in Mukherjee’s equation No. 13.118 ELECTRICAL THEORY OF ADBORPTTON The writer considers the double layer as consisting of a swface of rigidly fixed atoms under continuous bombardment of positively and negatively charged ions, any particular point on the rigid surface becoming in turn negative, neutral and positive, these conditions arisdg in any order. The observed contact difference is the average effect of these conditions. Where several kinds of atoms are present in the solution the average number of any one of them at the surface will depend on their concentbration, valency and mobility. The variation of contact Werence from negative to neutral and positive was observed with cotton and aluminium sulphate near the neutral point. These variations occurred during the same experiment, the readings being direct measurements of E.1I.F.s developed by filtration under pressure.This point would be covered by putting n2 = 1 and = 2 or 3 in Mukherjee’s equation No. 13.118 ELECTRICAL THEORY OF ADBORPTTON The writer considers the double layer as consisting of a swface of rigidly fixed atoms under continuous bombardment of positively and negatively charged ions, any particular point on the rigid surface becoming in turn negative, neutral and positive, these conditions arisdg in any order. The observed contact difference is the average effect of these conditions. Where several kinds of atoms are present in the solution the average number of any one of them at the surface will depend on their concentbration, valency and mobility.The variation of contact Werence from negative to neutral and positive was observed with cotton and aluminium sulphate near the neutral point. These variations occurred during the same experiment, the readings being direct measurements of E.1I.F.s developed by filtration under pressure. This point would be covered by putting n2 = 1 and = 2 or 3 in Mukherjee’s equation No. 13.118 ELECTRICAL THEORY OF ADBORPTTON The writer considers the double layer as consisting of a swface of rigidly fixed atoms under continuous bombardment of positively and negatively charged ions, any particular point on the rigid surface becoming in turn negative, neutral and positive, these conditions arisdg in any order.The observed contact difference is the average effect of these conditions. Where several kinds of atoms are present in the solution the average number of any one of them at the surface will depend on their concentbration, valency and mobility. The variation of contact Werence from negative to neutral and positive was observed with cotton and aluminium sulphate near the neutral point. These variations occurred during the same experiment, the readings being direct measurements of E.1I.F.s developed by filtration under pressure. This point would be covered by putting n2 = 1 and = 2 or 3 in Mukherjee’s equation No. 13.118 ELECTRICAL THEORY OF ADBORPTTON The writer considers the double layer as consisting of a swface of rigidly fixed atoms under continuous bombardment of positively and negatively charged ions, any particular point on the rigid surface becoming in turn negative, neutral and positive, these conditions arisdg in any order.The observed contact difference is the average effect of these conditions. Where several kinds of atoms are present in the solution the average number of any one of them at the surface will depend on their concentbration, valency and mobility. The variation of contact Werence from negative to neutral and positive was observed with cotton and aluminium sulphate near the neutral point. These variations occurred during the same experiment, the readings being direct measurements of E.1I.F.s developed by filtration under pressure.This point would be covered by putting n2 = 1 and = 2 or 3 in Mukherjee’s equation No. 13.118 ELECTRICAL THEORY OF ADBORPTTON The writer considers the double layer as consisting of a swface of rigidly fixed atoms under continuous bombardment of positively and negatively charged ions, any particular point on the rigid surface becoming in turn negative, neutral and positive, these conditions arisdg in any order. The observed contact difference is the average effect of these conditions. Where several kinds of atoms are present in the solution the average number of any one of them at the surface will depend on their concentbration, valency and mobility. The variation of contact Werence from negative to neutral and positive was observed with cotton and aluminium sulphate near the neutral point.These variations occurred during the same experiment, the readings being direct measurements of E.1I.F.s developed by filtration under pressure. This point would be covered by putting n2 = 1 and = 2 or 3 in Mukherjee’s equation No. 13.118 ELECTRICAL THEORY OF ADBORPTTON The writer considers the double layer as consisting of a swface of rigidly fixed atoms under continuous bombardment of positively and negatively charged ions, any particular point on the rigid surface becoming in turn negative, neutral and positive, these conditions arisdg in any order. The observed contact difference is the average effect of these conditions. Where several kinds of atoms are present in the solution the average number of any one of them at the surface will depend on their concentbration, valency and mobility.The variation of contact Werence from negative to neutral and positive was observed with cotton and aluminium sulphate near the neutral point. These variations occurred during the same experiment, the readings being direct measurements of E.1I.F.s developed by filtration under pressure. This point would be covered by putting n2 = 1 and = 2 or 3 in Mukherjee’s equation No. 13.118 ELECTRICAL THEORY OF ADBORPTTON The writer considers the double layer as consisting of a swface of rigidly fixed atoms under continuous bombardment of positively and negatively charged ions, any particular point on the rigid surface becoming in turn negative, neutral and positive, these conditions arisdg in any order.The observed contact difference is the average effect of these conditions. Where several kinds of atoms are present in the solution the average number of any one of them at the surface will depend on their concentbration, valency and mobility. The variation of contact Werence from negative to neutral and positive was observed with cotton and aluminium sulphate near the neutral point. These variations occurred during the same experiment, the readings being direct measurements of E.1I.F.s developed by filtration under pressure. This point would be covered by putting n2 = 1 and = 2 or 3 in Mukherjee’s equation No. 13.118 ELECTRICAL THEORY OF ADBORPTTON The writer considers the double layer as consisting of a swface of rigidly fixed atoms under continuous bombardment of positively and negatively charged ions, any particular point on the rigid surface becoming in turn negative, neutral and positive, these conditions arisdg in any order.The observed contact difference is the average effect of these conditions. Where several kinds of atoms are present in the solution the average number of any one of them at the surface will depend on their concentbration, valency and mobility. The variation of contact Werence from negative to neutral and positive was observed with cotton and aluminium sulphate near the neutral point. These variations occurred during the same experiment, the readings being direct measurements of E.1I.F.s developed by filtration under pressure. This point would be covered by putting n2 = 1 and = 2 or 3 in Mukherjee’s equation No.13. Apparatus and methods in the Leiden Cryogenic Laboratory BY Dr. C. A. CROMMELIN. 4 1. that it Introduction. -- Thc following gives a consistent description artiele is amanged so of the apparatus and methods used iii th I! Leiden cryogenic laboratory ; its purpose is t o make it easy to fiiid the exact places in the nunicrous commuiiicatioiis in which the different appliaiices aKe treated. We hare limited ourselves to a general outliiie; for the details, t lie original papers, to which the references refer, must ~ - J C consulted '1. Thc purpose of the Leiden cryogenic laboratory is aiid has always been to be able l o produce ciny temperature below zero degrees, in order to be able to make accurate physical measwe- meiits at those temperatures.And the claim is, for accuratc nieasnrenients at least, a constaiicp of 0 O . 0 1 during several hours. The best way to fulfill this claim is to make use of liquefied gases boiling, well stirred, nndw different pressurcs and keeping these pr(~ssurw exactly constant. In Leiden oiily pressures of one atmosphcr*c and loww (often as low as a few ni.m., and in the case of Iielinm a few tenths of a m.m.) aw used, as the coiistruction of cryostats for considerably higher pressures than one atrnosphere is difficult and may even present, danger. The yegioii of t emperatuiw covered by each substance ') For a description in full detail of some apparatus, which have not yct been published hitherto, especially the new hydrogen and helium plants and thc methyl chloride and ctliylenc boiling flaeks, see Comm.No. 158.is therefore tlie region between its boiling point and its trip12 point. In the following table is given a list of the substances used in Leiden, with the best values of their boiling and triple points; I have also added the triple point pressures, the critical temperatures aiid tlie critical pressures. For the better under- standing of the following pages it will be useful to have all these data collected in a table. Triple point SUBSTANCE. Triple point methyl chloride . nitrous oxide.. . . ethylene . . . . . . . . methane . . . . . . . . oxygen. . . . . . . . - nitrogen . . . . . . . . neon . . , . . . . . . . . hydrogen . . . . . . . helium . . . . . . . . . Boiling point .- 24fl - 89.8 - 103.72 - I d 1 37 - 182.95 - 195.78 - 245.92 - 25'3.76 - 268.83 0 - 103.6 - 102.4 -- 169 - 183.15 - 218.4 - 209.86 - 248.67 - 2h9.14 < - 272 - - - 7.0 9.64 32.35 5.07 < 0.002 - 3- 0.2 -.I____ Critical tempera- ture. $- 14gO + 36.50 + 9.50 - 82.85 - 118.82 .- 147.13 - 228.71 - 239.91 - 267.84 _____. -___ Critical pressure in atm. - -~ - -~ 65.98 71.65 50.65 45.60 49.713 33.490 26.86 12.80 2.26 In examining this table, one sccs that the ranges of tenipc- ratures from -%lo down t o -2218O C., from - 2 2 5 3 O to -259O C. and finally from -269O t o -272O C. arc attainable by baths of liquified gases; and that there are gaps between - 2 W and -2253O C., and between -2259O and -269O C. In the first gap there is still neon (from - 2 4 6 O t o - 249O C.), but this substaiice covers unfortunately only a small range.At the end of this paper the construction of the cryostats will be. discussed ; their construction depends of course on whether there is a boiling liquid available or not. Let 11s first cousidci* tlic cliffei~~nt i~ictliocls uscd in liqut~fying the vaiaious gases. $ 2. The regenerative cycles for the cascade : methyl chloride, ethylene and oxygen I). -- The principle of tlie I'IcTm c.aseade ') Comm. No. S7. See also Coinin. No, 14 $ 4-7.3 17i of cycles is so well known, that it need not bc explained here. The construction, however, of the Leideii regenerative cascade differs in iiiaiiy respects fro111 t lie original arraiigemciit of 1'Icwr as will bc leariicd from the following dcscriptioii.Thc first two steps in the cascade of temperatures are niadc by methyl chloride aiid ethylene. The nietliyl chloride, liquid at ordinary temperature under the moderate vapour pressure of about 4 atniospliercs (if pure, otherwise a little higher) is st.ored up iii a reservoir. I n starting tlic cycle, a certain quaiitity is allowed to flow into the so-called I)oiling flask, made of geriiiaii silver aiid coiistructed iii such a way that it can resist tlith external pressure of the atniosyliere aiid niay thus bc cvacuated. The liquid fills only a m a l l part of this flask. The boiling flask elids at the toy iii a very wide copper tube (nearly a11 the tubing iii the cryogeiiic laboratory is of copper) , leading to a powerful, fast running vacuum punip (from thc Burckhardt factory a t Uasels) with a displawiiieiit capacity of 360 31::.ail hour. The compression outlet of this punip iii connected with tlic vac~uiii side of a vacuuni and compression punip (of the origiiial COLLADOS t j y e used by PICTET from the Soci6t6 ghevoise at Gelleva) ; thesc two puttips arc thus connected i i i series. The outlet of this coinpressor is filially coniiected to tlic above nientioiied reservoir of the liquid. When the cycle is \\.orking aiid hot11 punips arc ruiiiiiiig at tlic proper speed, tlic state of tliiiigs is as follow: the liquid in the boiling flask boils, under the action of the vacuuni pump, at the greatly rcduced pressure of 1 to 1.5 c.111. a t a temperature of about - 85O to - 90° C. 14-01~ the vacu~iiii punip it goes to tlic coiiipressor, where it is coniprcssed a t 5 to G atmospheres (de- pciidiiig on its purity, the tenipci~atnre of the room, aiid that of the cooling water of the punip) , liquefied and finally retnrned t o the reservoir; thcu it goes again to thc boiling flask etc.Thc rate at which the methyl chloride is circulated is fiwln 18 to 20 M3. gas (under iioririal conditions) aii lionr. All thc valves being rcgulatcd properly, a stationary state is establislicd aiid we maintain, as long as we let the pumps ruii, a large bath at about - 90° c!. This is the first cyclc; for which, of4 178 course, must always be chosen a substance which is readily liquefied at ordinary temper a tu r e . We are now able to undertake the second step in the cascade. As. substance for the second cycle ethylene (recommended by CAILLETET, because of the large distance between boiling point and triple point and the niodcrate vapour pressure) is used in Leiden.As to the construction of the cycle, we may be brief, for it is nearly the same as in the case of the first. cycle, with two exceptions only. First that the ethylene, not being liquefiable a t the teiiiperatui~ of the room, cannot be stored up in a small reservoir (as in the case of the methyl chloride). Therefore a large vessel of 600 liters capacity has been inserted in the q-c.10. This container can hold a great part of the ethylene in thc gaseous stale. Secoiidly there has been inserted a condeiisatioii spiral of considerable length and surfact1 in the methyl chloride boiling flask, in order t o liquefy the ethylene.I?urther the cycle contains a Burckhardt vacuuni- pump and a compressor connected in series, exactly as in the case of the first cycle. When now both pimps are running a t the proper speed, we have the following state of things: the ethylene is compressed (but not liquefied) by the compressoy at a pressure of 5 t o 6 atmospheres and flows at this pressuw iiit o the condensation spiral, This spiral enters the met liyl chloride boiling flask at the top and emerges from it at the bottom, while the methyl chloride vapour flows of course in the opposite direction. Thus the ethylene is first cooled by the rnethyl chloride vapour, which is about to leave the boiling flask, afterwards by thc colder vapour in the middle of the flask and only finally by the liquid methyl chloride itself.The upper part of the boiling flask works as a regenerator of heat and therefore the cycle is called a regenerative czjcle. To assure the proper and economic working of such a boiling flask with condensatioii spiral care is to be taken that the rate of flow of the methyl chloride vapour and the ethylene, the length, the surfacc and the heat conductivity of the spiral and the heat insulating capacity of the walls of the flask must be such that: lo. the ~thylene is totally condeiiscd before reaching its oiztlct and 2O. tho mcthyl chloride rapour isallowed to ,,give off all its cold” (if 1 may cslwess niysell‘ in this may) to the ethylene. If in the construction of the boiling flask these coiiditioiis are fulfilled, the ethylene is completely liquefied and the nietliyl chloride v a p u r leavcs the boiling flask at about the temperature of the room, so thut no exterml sign of cold c m be see% on the copper tube which leads to the vacuum 1)uiiip.All external signs of cold. siiow and ice on tubes or apparatus, are strictly forbidden in a cryogenic lahoratorg ; thcg always mean ,,waste of cold”, i. e. waste of the thing we wish to get hold of and they are, as such, evidences that the working is iiot so ccoiioriiical a s it ought to be. It may perhaps he stated hem that iii thc Lcideii laborutoi-y cstcriial sigiis of cold are alinost never seen. The further working of the cthyleiic c3yclc iiiay iiow be tlcsc rilwd in a few words. ,\fter having heen liquefied, the etliyleiic enters the etliylene boiling flask at.the bottoin and boils thcrc under a pressure of about 2 c.m. at a temperature of about - 150° C’. under the action of the vacuuiii punip. The constructioii and (as will be seen afterwards) also the woi-king of this boiling flask is so rwcli the same as that of the methyl chloride boiling flask, that it is not necessary to dwell any longer upon it here. After having past the puiiip, the ethylene is led to the conipres- sol’, coinpressed to 5 to G atmospheres, tlieii liquefied etc. The rate of cir-cnlation is here too about 18 to 20 &I3. an hour. By this cycle we maintain a bath at about - 150° C. This ternpera- ture is coitsidera1)ly below tlic critical point o f oxygen ; and this gas can iio~v hc readily liquefied in the third cycle.The main apparatus in this cycle is the Brotherhood cornpressor ’) , with a capacity of 20 &I3. an hour. This punip is lubricated with a iiiixture of glycerine and water ; lubrication with oil presents danger, for the mixture of oil vapour and osy-gen is explosive. It compresses the oxygen in three steps to about 20 atmospheres. The compressed oxygen enters the coiidensatioii spiral, which is arranged in the ethylene boiling flask very much in the same way as thc ethylene coridevisatioii spiral in tlic methyl chloride boiling flask. The oxygen is liquefied and leaves the boiling flask ’) For an older type of Brotherhood compressor see Comm, NO. 51 $ 3. . _. - __ - VOL. XVIII-TO1 SO G at the bottom through a well insulated copper tube, which leads to the oxygen boiling flask. The oxygen boils there at atinos- pheric pressure, at a temperature of - 1 8 3 O C.It is not necessary t o lower the pressure in this case, as will be made clear afterwards. A vacuum pump thus is riot necessary in this cycle and the construction of the boiling flask is conse- quently much simpler than that of the ethylene and methyl chloride flasks. The oxygen evaporates in a rubber bag, from which it is sucked by the Brotherhood pump, compressed etc. 4 3. The liquefaction of air l) , nitrous oxide *> , methane and nitrogen 3 ) . General remarks. - The foiwtli operation is the liquefaction of air. The air is first freed from carbon dioxidc by letting it pass through a solution of caustic soda, then com- pressed by a Brotherhood pump of exactly the same type as the oxygen pump t o about 15 atmospheres, and finally liquefied in a condensation spiral, which is contained in the oxygen boiling flask.It is collected in a large cylindrical vacuum vessel, from which it is syphoned into the spherical vacuum vessels oE 3 t o 5 L., in which it is stored up. About 14 IJ. of liquid air an hour are prcpared in this way. A few general remarks. It is hardly necessary to say that this rather coniplicated plant of regenerative cycles arranged in cascade has not been built t o liquefy ail.. It has been built to work cryostats with methyl chloride, ethylene, oxygen and some other gases for phy- sical determinations at all the temperatures from - 2 4 O to - 2 1 7 O C. When the use of vacuum vessels became easy and reliable so that liquid air was the appropriate mean to store cold for liquefying and other purposes, the apparatus for the liyue- faction of air was added to the three existing cycles and this could be done in a exceedingly simple way, as was shown above.Although not less than 4 compressors and 2 big vacuum pumps are running whcn air is liquefied, the whole installation *) Comm. N". 04f X I I I ; 57, $ 4 aiid 5. ') Comm. NO. 83, IV. ') Comm. NO. 51, $ 4 and 5.IS1 7 works very ecoiioiiiicall;\-, requiring oiilj- 23 H. P., that is 1, 1,64 H. l’. per-- . The well known LINDE and CLAUDE instal- hour lations, although a good deal simpler in construction, require about the same airiount of power. But let us return to our subject. Besides the gases already inentioiled smie other gases are sometimes liquefied, as will be seen 011 looking at the table in the introduction.The liqucfactioil is readily effected by the cooling agents of the cycles described. generally in rather simple apparatus which will not be dcs- wibed here. Nitrous oxidc is used only whcii t e i ~ i p e r a t ~ u ~ s from - 90° to - 102O C. (bet~vcen methyl chloride and ethylene) arc. especially wanted; it is the liqnid, which takes the place of thc solid carbon dioxide; f o r with a solid it is difficult to get a. good coiistancp of temperatiire. In the same -way methane fits exactly in the rather consider- able gap between ethylene and oxygen, viz. from - 1 6 1 O to - 1 8 3 O C. In this region methane is ~ ~ g u l a i ~ l y used. Finally nitrogen is used chiefly in magnetic experiments, whcrc a inagnctical ly iiidif f ercnt liquicl is i.equii>ed.It is still worth while to draw attention t o the fact, that in the Leiden installation the inaiii substances have their own com- pressor, condensation spiral, boiling flask etc., which are never used for another snbstance. First this is a safety measure. Mixtures of sevei*al of the substances used would be explosive, and, in using the purnps for different substances alternately, tlir forniation of such dangei*ons mixtures could hardly cvcr h c avoided on account of the solubility of the substaiiccs in thc lubricating oil. The matter is so evident, that it need not be explained inore in detail. But evm if 110 explosive niistuim 15-onld be formed, the usc of one apparatus for only one substance is absolutely necessary to keep the plant always in working order.Continually evacu- ating the apparatus, and especially the pumps, would be an eiiormoiis waste of worlr aiid time a i d the productive capacity of the laboratory would be considerably smaller than it is now.182 8 5 4. The hydrogen plant I). - The cascadc iiictliocl of PICTRT, which can be so well adapted to the physics of low temperatures, fails below - 2 1 7 O (’. The lowest temperature which can bc obtained with oxygen, boiling undei* a pressure of a few milli- meters, is - 2 1 7 O C., but this temperature is of no avail for the liquefaction of the next gas, viz hydrogen, for this gas has a critical temperature of about - 240° C. (the use of neon will be discussed later).Another method, that of the IJIKDE process, making use of the JOULE-KELVIN effect, has to be applied (we will not dwell here upon the possibilities of other. inctliods). In making usc of the J~ULE-KELVTS effect it has to he taken into account that, at ordinary t emperaturcs. the effect gives a slight heating for hv- drogen and helium; so that hydrogen and helium have to be cooled below the inversion point before expanding. I will give now a short description of the hydrogen installation, as it now stands 2). The gaseous hydrogen is coinpressed up to 150 to 200 atm. by a set of four horizontal pumps obtained from thc Burckhardt factory at Basels. They run quite slowly (100 i*evolutions per minute) and compress the gas to the required pressurc in 5 stages.The first pump compresses 40 &I3. of gas an hour From 1 to 2 atni.; the secoiid pump works in 2 steps, the first one from 2 to 6, the second one from 6 to 25 alm. The third and the fourth pump are connected in parallel a i d each operates on half of the gas. They both work in 2 steps also, the first step compressing from 25 to 50, the second one from 50 to 250 atm., but generally a final pressurc of 150 t,o 200 atm. is used. Between the different steps the hydrogen is thoroughly cooled in spirals in cooling water. The compressed gas now conies into the liquefier, of which fig. I gives a schematical representation. The gas conies in a t A where the tube is divided iii 2 parallel tubes. One portion l) Comm. NO. 94f, X (the old plant), No.158 (the new plant). *) The original liquefier, of which a description in full dctail may be found in Comm. No. 94f, X, $ 2, is still preserved to be used, if necessary, as a reserve apparatus.9 183 of the gas goes through the spiral B (spiral wound tubes are re- presented by zigzag lines), the other oiic through 13'; tliesc spirals are uiiited again at C. The hydrogeii continues its course through the spirals D, E aiid F and finally cxpands to atliios- phcric pressure through the expansion valve K (handle 1"). I t is partly liquefied, the liquid is collected in the vacuuin glass uficl is syphoiiecl through the valves S and S' into the vacuuni I)ulbs 11 aiid 11'. In these bulbs thc hydrogen is transported to the diffcreiit i'ooi11s of the laboratory. As one easily sees, the construction of the apparatus is such as to cool the conipressed hydrogen flowing in the direction of thc cspansiou valve as efficiently as possible by the cold vapour, flowing in the opposite direction, according to thc regeneration principle. This principle is, as we saw, also applied in the construction of the Imiliiig flasks in the cycles.Moreover, cooljiig by iiieaiis ol' liquid air is absolutely neccssary in the case of hydrogen to bring it to the tcmperaturc wherc the JOUIIE-~LXIPI' effect gives cooling. Liquid air is used iis a refrigerating agent; it is poured through the valve L into the vacuuiii glass E' and it evaporates there uudei- a pressure of about 2 nim. maintained by the action of a power- f u l vacuuIii puiiip (displaccnieiit capacity 360 11;.an hour). 111 this way the liquefier is started. h'roiii the iiioiiiciit that the first sinail quantity of liquid is collected iii G - about half ail hour after the hegiiiiiiiig of the expansioii - the cold vapour of the liquid hydrogen does its work in the process of rege- iicratioii. And ~ \ * l i c w filially a stationary statc iq reacliclci, t h o regeneration acts so well that no external signs of cold are secii. 'Yhc tigure show clearly enough that the hydrogeii is cooled iii E by the liquid air itself, in D aiid B by the vapour of the air, in B', C' a i d P by the vapour of the hydrogen. Especi- ally interesting is t lie capacity of this liquefier to pl*oducc liquid : this is 13 1,. of liquid an hour in continuous working, i. c. lor as many eonscmtivc hours as is wanted. 4 5.The purification of the hydrogen. -- I t is hardly possiblc to operate a Iij~lrogcii liqucfier with comtiiercial hydrogcii 011184 Fig. 1.11 185 account of tlie imliuritics coiitaiiied iii the latter. As thcsc inipiivitivs v c i y often amount to 2 01' :3 a/o, tlicy block the i ~ a v ~ o w spirals and thz espaiisioii valve soiiic tinic aftei- tlic starting of the apparatus. I t is then inipossiblc to \voi*lr cv(w for c?. couplc of hours. Thoroughly pnrificd liydrogeii is absolutely iicces- sary. Wlieii iio liquid hydrogcii is available, a separator l), in ivliich the corlipi-esscd gas is coolcc1 to the tciiipcratnrc of liquid air, may bc used. n u t this apparatus does not deliver thc liydrogcii purc ciiougli to avoid 1)loclriiig ; foi* solid iiit rogcii has even at .- 2 1 7 O (".(lowest tenipci-atui-c of liquid osygcii) a still rather coiisiderable vapouy pressure. Generally after tlic preparation of about 4 liti-cs of liquid, tlic liqueficr has to lic stopped and hcatcd bcforc thc work can bc coiitinuccl. The evaporating liydrogcii, w1iic.h is of coiii*sc ~wrfcci 1:. p i i i ~ , is then carefully coll~ctcd. Aft cia soiiic time of ivorlriiig in this way, a sufficient quantity of purc gas will 11c stored up to kcep the apparatus 1-uniiiiig for n loiiger tiriic aiid to bc able to makc use of aii iniprovccl separator 2, which is in constant use now. Thc impurities arc renioyed i n this apparatus by cooliiig the gaseous liydrogeii down to tlic temperature of liquid hydrogen.The construction of this scparator is cxtremely siiiiplc : a wide copper spiral is niount cd in n silvtwd cyliiirliical \'acuuiii vessel ; liquid hydrogen is sprayed by siii all quantities uiiclcriicath this spiral; it evaporates a i d cools tlie tube iiitciwally to about - 253O C., the tciiipei'ature of hoiliiig Iiydrogcn, at which tempcraturc thc \-aj)oui* j)i*cssiii*~ of nitrogcii aiid osygcii arc entirely negligible. The gaseous hydrogen flows through thc vacuum glass, at tlic outside of tlic cold spiral; tlie iiiipurities are deposited on tlic spiral slid the liydrogcii lcavcs the separator entirely pure, if its spccd is I-egnlat ed properly. Once having got a stack of purc hydrogen, conipressed iii cylinders, this stock must be coiisidcrcd as a first eoiiditioii of working.This separatov is a sinall aiid handy apparatus, aiid delivers 5 MS. of perfectly pure hydrogeii an hour. As sooii as working l) Comm. N". 94f, XI. *) Comm. No. 109b.186 12 with this separator was started, the diffictultics with the iinpu- rities of the commercial hydrogen were quite overcome. S 6. Safety devices. - In describing the hyd~ogcii plant I have been rather schernatical and I haw not dwelt upon marly details and auxiliary apparatiis nlthongh they arc quite in- teresting from a technical point of yiew. They may be found in the original memoirs. 13nt I have to insist on the care that is bestowed 011 assuring safety in the cryogenic laboratory. Hydrogen being com bustible there is especially danger of explosion b;v mixing with air.Hvcry possibility of electric sparks, so easily obtained by thc friction of solid particles in the cold gas, must be avoided with the utmost care. The same applies to methyl chloride and ethylene. Ethylene has the ad- vantage over hydrogen that one can smell it, but it has about the same density as air so that it does not rise immediately to the ceiling like hydrogen. We had learned from our experience with methyl chloride and ethylene, which substances are used in large quantities and at rather high pressures, how to work with hydrogen. In fact, protection is now a strong tradition which has established itself by long practice in the laboratory, so strong, that the scientific and the techiiical pcrsonnel take it as a matter of course that in every experiment safety is given the first consideration.Work has often been very seriously handicapped for this reason but every sacrifice has to be made in this direction without sparing in the least time or trouble. 1)uiing sonic 35 years’ working with large quantities of com- bustible gases a serious accident has never happened in Leiden and no bystander has ever been hurt. 4 7. The helium pknt l ) . - The liquefier now in IISC has many advantages above the original one, which is kept to be used as reserve. It is built according to exactly the same principles as the hydrogen liquefier ; i. e. preliminary cooling (here with liquid hydrogen) bclow the inversion point l ) Cornm. No. 108 (tho old plant), No. 158 (the new plant).l.3 187 o f the JOULE-KELVIN effect, cspansion through an cs1)aiisioii valve; aiid in the stationary state cooling by helium vapour ac- cording to the regcncrative principle.We reproduce a schematical drawing of this liquefier (fig. 2). The gaseous helium compressed at about 30 atmospheres enters at A and is divided in 2 spirals B and B'. These spirals are united afterwards and then divided again in C and C'; B and C are cooled bj- cold hydrogen vapour, B' and C' by cold helium va1:our. Now these spirals are finally united in the spiral D, cooled in the coldest hydrogen vapour a i d in E cooled in thc liquid hydrogen itself. P , the last regeneration spiral ends in the expansion valve I<. Heye the helium is liquefied and runs through a yacuum tube into the cryostat, which is directly con- iiccted to the liquefier. We shall describe this cryostat after- wards.As to the stock of lieliun~, the first quantities TVCIT pi*cpared fiaoni monazite sand. T n later yeam Prof. OXSES received on several occasions preseii ts of heliuiu or of gases which contained more or less considerable quantities of helium, e. g. from Mr. GEORGES CLAUDE at Boulogne-sur-Seine and from the Wels- bach Light Company at Gloucester N.Y. Rut during the war there was no longer an]- chalice of such presents aiid at last the quantity available was hardly 300 litres so that, the situation began to become critical and a new preparation was considered, if nionazitc sand could be ohtained. Fortmiatoly this troublc was spared. The Anierieaii Navy in 1919 made Prof. ONNES a pimciit of iiot less than 30 M3.of helium prepai~d iii Tcsas during the war for the filling of airships. And Prof. Mc LENNAN of Toronto himself brought to Leiden in September 1921 a big cylinder, coiitaiiiiiig about 6 M3. of helium, which had heen prepared by him in Canada for. the same purpose, The stock of helium is now so plentihl, that it will certainly last for a great number of years. Akt t h e elid of this sketch of the Leideii heliuni iiistallation, I will once more einphasizc the fact that the plant is built as a circulation apparatus more or less in the. style of a cycle. of the PICTET cascade. The gas circulates many times before the appearance of the first drop of liquid. If %he gas could iiot be1 ‘4 Fig. 2.15 1 s9 compressed agaiii after tlic first expaiisioii and so forth (but had to be stoi.ed up c.y . ill a gasometer) tlicii a yuaiitity of heliuin, twenty tin-es as large as ire now have, would be neces- sary, just to start the licluefier. This csample makes it parti- cularly clear that ciiwdation is entirely indispensable in ibis kind of cryogenic worlr. 4 8. Liquid neon'). -- By mealis of the cycles of metliyl chloride, nitrous oxide, ethylene, methane, oxygen and hydrogen, the laboratory has available baths of liquids from - 2 4 O . down to - 2 5 9 O C., with the exception of the region between -217' (oxygen, boiling under a pressure of a few milli- metres) a i d - 253' ('. (hydrogen hoiliiig uiicler atinospheric pressure) aiid this rcgioii was for a lorig lime uiiattaiiial,lc fo:. accurate mtasurenients. The possibility o t' a bath of hydrogen boiling under the press~ii e of a few atmospheres has becii considered, but the c q ostat has not beeii coiistmcted.Urit1i n bath of liquid iieon (boiling point - 245O.92 C., triple point -248O.67 C.) a siiiall range of tciiiperature could be covered. The coiistructioii of sucli a bath has iiideccl bceii accoiiiplishc(l. Compressor, liquefier ct c. were not necessary, for neon is sold at --- 253" C., and iherc was ssufricieiit iiynid liyti~wgeii a\ dablo to obtain the desired Yesult with a yery compact and siinplt. apparatus. This apparatus is very much like the separator for the purificatioii of the hydrogen. It consists of a loiig wide copper spiral iii a. cylindrical vacuuiii vessel. Liquid liydrogeii ir, sprayed uiidcriieath ihis spiral, it evaporates aiid cools 11ic spiral internall3-, if necewaq- dovn to -253O C.The iieon flows through the vacuum glass at the outside of the spiral and coiideuses 011 it in solid 01' liquid state. At - 2 5 3 O C. iieoii is solid, but by proper regulation of the quantity of hydrogen, the spiral is brought to a somewhat higher temperature, so that thcb neon beco~nics liquid and cIi*ops off froni the spiral. I t finds its way through a vaciiuni tube at thc bottom of the vacuum glass to the ciayostat, a sccoiid vacuuni glass iiiinicdiittc1~- con- ') Comm. NO. 1 4 7 ~ .190 16 nected with the liquefier. In this way a bath of liquid neon was realised in a very simple way. A small range of temperatures could be covered by this bath and in 1915 it has actually been used f o r the determination of some physical constants of liquid neon.It has some advantages above a bath of liquid or gaseous hydrogen especially in researches in which electric sparks occur; but the main dis- advantage was the smallness of its range of temperatures. Therefore other methods were introduced, which could bring the whole range from - 2 2 1 7 O to - 2 2 5 3 O @. within the reach of accurate measurenients. These iiiethods will be discussed later on l). § 9. The purification of the neon. -- Working with a bath of liquid neon is only possible if a rather considerable quantity of neon is available. \Ye rrieiitioned already the present of Mr. GEORGES CLAUDE viz. a large quantity of gas, containing the most volatile distillation products of liquid air.This gas con- tained, besides lieliurri, about 30 G/c) of neon. In this way the laboratory came in posscssioii of a few hundred litres of neon. For a first purification, the gas was solidified by means of liquid hydrogen and the helium and hydrogen were pumped off; then the neon was heated up slowly and the evaporatinq gas chiefly consisting of neon was stored up separately; the less volatile impurities oxygen and nitrogen tjhen remained solid in the tube. This operation was repeated several times. Finally it was passed over. charcoal in liquid air according t o the well-known method of Professor DEWAR. Accurate measurements showed afterwards that this purified neon was not so pure as we thought. Therefore a more refined and accurate method of purificatioii was designed and applied by the author, with a view to determination of the critical point, vapour pressures and the rectilinear diaiiieter of CAILIXTET arid MATHIAS .First, the method described above was applied, but now iii an appa- ratus entirely rnade of glass, without any rubber tubing; only the middle fraction was used. Finally the neon was allowed *) 9 11 of this article.17 191 t o flow ~ c i y slou-ly tbi-oi~gh a \vide glass spiral iriiiiiri-sed iii liqnid hydrogcn ; the iicoii had to he k q ~ t bclon- its vapoiit* pi~essure at - 2 5 3 O ( I . (about 12 n i m . ) in oidci. t o avoid solidi- fication. Traces of less volatile impurities were deposited against the n-alls of the spiral. Finally it was f~actionatecl oiice more. In this way a wry pure product is obtained.10. The cryostats for liquids. - Hitlierlto, ~ v t hare only spokcn of the apparatus f o r the liquefaction aud tlic puiiifica- tioil of the different gases. To make accuratr physical dctcy- iniliatioris hom-ever, we need special apparatus i n which u’c map lrcep the tcriiper*aturc of the liquid constant and c\-cryvhci*e tho same, and which may coillain the substances of which we want to study tlic propci*ties at low tcmpei.atur.es. These tip- paratus, the so-called cq-osta ts, arc at the present riiorneiit for us even more interesting perhaps thaii the liquefyiiig apparatus. h’oi*iiierly the state of things was different ; then. t hc iiiaiii p~*ol)leiii waq the liquefaction itself and collectiiig :I fcw cubic centimeters of liquid oxygen 011 hydrogen was r? suvccss iii itself.Non- tlic i!!cltliods of Iic!nefping c~-cii thc most difficult liquefiable gases arc well developed and our attention is inore fixed 011 the possibilities of deteriiiiniiig accuratelv physical constants at low tcniperatnres. The coiistiwtioii of ciyostats with a constancy of 0.01 of a degrcc i~lprcsciits at the present iiionic~nt a most intercstiiig problwi. l u Leideii iiiany iiiodcls have bcckn in use iii thc past years. liiit as n7e are not giiviiig a historical sketch of the development, of the laboratory but only a (rather short) account of its 1)resent state, we confine ourselves to the description of the largest ~noclel for universal use with all liquid gases (except neon and helium). which is ~rnployed now I). Fig. 3, in which two euts of the appa ratus are given, may be understood without a long explanation.A few remarks will do. Thc largc vacuum vessel V (internal dia- meter 12 c.in.) contains a geriiiaii silver cylinder, in which the ’) W e will not dwell here upon the coastruction of cryostats f o r spwiatl purposes.1.8 192 liquid (avcragc 2\$ litrcs) is poured. Tlict glass is covercd by a cap in which are openings for the apparatus, for the inlet of the liquid and for the outlet of the vapour (the latter leading \. .JI Fig. 3. t o a gasometer or to a vacuum pump). Thc enveloped by liquid air in order to diminish the rate of eva- poration of the cryostat liquid and to improve coiistancy of the1.9 1 33 temperature. Generally, in the smaller types of cryostats I ) (which are built in exactly the same way as this largest one) the air is contained in a secoiid, larger vacuum vessel.But as 12 e m . is about the limit for cylindrical vacuuni vessels (at least wlien they are made OP ordinary ,,Thuringer ’ ? glass) the liquid air t~iiwlopiiig this cryostat is contained in a gwman silver cylindci*, iiisulated by nieaiis of thick walls, filled u p with cotton wool. The figure shows further (as an example of apparatus) the two reservoirs Th o f a differential gas thermometer, two resistaiice thermometers W and two stirrers R (see also the horizontal cut). These stirrers work with valves and literally pump the liquid ivund. Tlic sensitive oil manonietcr, by which very slight irreg- ularities in the Iwessure aye immediately seen, is not represen- ted in the drawing.The helium 2 , and iieoii ci-yostats 3, are in principle very inuc~h the same as the cryostat now described only their arrange- ment is different. Most of the liquids in use are ,,transportable ? ’, that, is to say, they are transported in vacuum bulbs from the liquefaction plants to the roo~ns where the cryost ats are installed. With helium and neon this transportation is not, or at least not yet, possible on accouiit of the costliness of the substances (little losses in transporting cannot he avoided) and, as to helium, on account of the volatility of the liquid helium. Therefore the helium and ncwii cryostats a3.c connected immediately to their liqucfiers in such a way that thc liquid flows froin the liquefier tlirwtly through a vacuuiii tube into the cryostat.Q 11. The hydrogen vapour cryostat ”). - Wc hare seen in the introduction, that theye are two gaps in the range of tcnipcratures which can not be attained by iiieaizs of a boiling liquid viz. from - 2 1 7 O to - 2 2 5 3 O C. (except for a small part l g - liquid neoii, this has been discussed already in 5 8) and from - 2 5 9 O to - 2 6 9 O C. To attain the teniperatures of the ‘) Comm. KO. 53, I11 mid l.Tr, 94c, 04d, !Mf, SIX. 3, Comii. 37”’ 119, 1%. ’1 Comm. NO. 1.17~. ‘) Comm. KO. 151a, 154c.194 20 first gap, a hydrogen yapour caryostat is now i n ixse in Tdeideii. OF this instrument, fig. 4 gives a cut in full detail. It is a rather complicated apparatus hut it, gives the required constancy o f temperature ( O O . 0 1 ) very well during several hours ; oiily the uniformity of temperature in the experimenting chamber is not yet exactly what i t ought to be. The principle is, that the experimenting chamber I: is kept at constant temperature by a current of cold hydrogen vapoui*. Before entering the cliainher, the hydrogen is heated up to the required temperature by an electric heating wire in which the current is regulated by a very delicate automatic arrangement. The vacuum vessel T' is half Filled with liquid hydrogen, the vessel XI contains the esperimcirtal chamber E. Both vessels are connccted 1)s the vacuum tubes ti,, and b , , , and are envelopcd by the big glasses B, avid V,, which contains liquid air. The gaseous hydrogen enters a t b,, at a prcssurc a little higher than atmospheric, is coiiducted to the bottom (b,) oC tlic vessel V and rises from there in bubbles to the surface of the liquid hydrogen. In this way we get a rather effective and adjustahle evaporation. of the liquid (average 80 litres an hour). Tlrc eva- porating hydrogen follows the arrows in the figure and arrives a t last at b, at a tcmpcrature oiilj- slightl;\- above - 253O C. in the heating chambers. There, it is heated to the requiiwl temperature and enters immediately the experimental chamber E. The heating chambers contain, besides the heating coils W,, the heliuiu tlicr~rionieter N,, connected by the steel capillary C to the niercury manometer, which regulates the heating current auto- matically by the contact a t X . If the hydrogen is slightly to warm, then the current is broken and the 'heating stops; if it is too cold then the niercuiy rises and the current is switched in. I suppose that this short account will give a sufficient idea of the working of this apparatus. Such a kind of cryostat is not yet available. for the second gap, from - 2 2 5 9 O to - 2 6 9 O C. Prom a theoretical point of view, the temperatures in question may be reached, either i n a helium vapour cryostat of the same model as the hydrogen vapour caryostat, or (those below the critical temperature of helium) in a cryostat with helium21 Fig. 4.boiling under increased pressure. Both constructions have beer1 considered but not yet carried out at the present date. A helium vapour cryostat of a construction different from that of the hydrogen vapour cryostat has been used recently f o r deter- 111 ilia t io 11s colic eimiii g s(supr.a (* o 11du c t om. A desv rip t ioii w ill b ( 3 published before loiig ').

ISSN:0014-7672    

DOI:10.1039/TF9221800175

出版商:RSC    

年代:1922

数据来源: RSC

摘要:

118 ELECTRICAL THEORY OF ADBORPTTON The writer considers the double layer as consisting of a swface of rigidly fixed atoms under continuous bombardment of positively and negatively charged ions, any particular point on the rigid surface becoming in turn negative, neutral and positive, these conditions arisdg in any order. The observed contact difference is the average effect of these conditions. Where several kinds of atoms are present in the solution the average number of any one of them at the surface will depend on their concentbration, valency and mobility. The variation of contact Werence from negative to neutral and positive was observed with cotton and aluminium sulphate near the neutral point. These variations occurred during the same experiment, the readings being direct measurements of E.1I.F.s developed by filtration under pressure.This point would be covered by putting n2 = 1 and = 2 or 3 in Mukherjee’s equation No. 13.118 ELECTRICAL THEORY OF ADBORPTTON The writer considers the double layer as consisting of a swface of rigidly fixed atoms under continuous bombardment of positively and negatively charged ions, any particular point on the rigid surface becoming in turn negative, neutral and positive, these conditions arisdg in any order. The observed contact difference is the average effect of these conditions. Where several kinds of atoms are present in the solution the average number of any one of them at the surface will depend on their concentbration, valency and mobility. The variation of contact Werence from negative to neutral and positive was observed with cotton and aluminium sulphate near the neutral point.These variations occurred during the same experiment, the readings being direct measurements of E.1I.F.s developed by filtration under pressure. This point would be covered by putting n2 = 1 and = 2 or 3 in Mukherjee’s equation No. 13.118 ELECTRICAL THEORY OF ADBORPTTON The writer considers the double layer as consisting of a swface of rigidly fixed atoms under continuous bombardment of positively and negatively charged ions, any particular point on the rigid surface becoming in turn negative, neutral and positive, these conditions arisdg in any order. The observed contact difference is the average effect of these conditions. Where several kinds of atoms are present in the solution the average number of any one of them at the surface will depend on their concentbration, valency and mobility.The variation of contact Werence from negative to neutral and positive was observed with cotton and aluminium sulphate near the neutral point. These variations occurred during the same experiment, the readings being direct measurements of E.1I.F.s developed by filtration under pressure. This point would be covered by putting n2 = 1 and = 2 or 3 in Mukherjee’s equation No. 13. ETH1711 CHLORIDE (C.H,Cl). BY PROFESSOR C. F. JKNKIN, C.B.E., N.A. Ethyl chloride is a colourless liquid a little lighter than water ; as its boiling point at atmospheric pressure is IZ.;" C. it may bc handled in open vessels without much loss and may be stored in glass bottles or cylinders of quite moderate strength.I t is an anathetic, but must be inhaled in considerable strength before it has appreciable effect. I t is very inflammable arid considerable care must be exercised to avoid fires. I t is a very convenient material for refrigeration; the following data indicate its properties as a refrigerant :- Latent heat at 0" C. Specific heat of liquid = '348 at - 30° C. to '413 at + 40" C. = 93-7 cal. per gram, or C" lb. per lb. ,, ,, ,, gas = - 2 2 to '28. at - 30' is 1.035 C.C. per gram. Specific volume of liquid at + 3oo is I.I39 ,, 7 3 7 7 at + 20' is 297 C.C. per gram. 7 7 ( a t - 20' is 1305 ,, ,, ,, ,, ,, gas (saturated) The critical point is 190' C. and j4 atmospheres. The vapour pressure curve is shown in the accompanying figure.I t does not attack metals, but for some reason not yet fully explained, the liquid does contaminate mercury, so that mercury manometers can only be used for pressures less than the vapour pressure corresponding to the room tcrnperature-i.e. roughly for pressures less than atmospheric. It attacks india-rubber, but vutcanised rubber tubing and corks can be used, though they leak a little. They swell considerably, but return t o their original size when the ethyl chloride dries off. Pure rubber tape is softened almost to solution. Vulcanite swells moderately. I t causes most woods to swell, but lignurn vitz. can be used for valves, etc., when a non- conductor of heat is required. Bakelite is also useful as a non-conductor ; it only swells very slightly.Liquid ethyl chloride dissolves a small quantity of water, but can be dried by passing it through calcium chloride drying tubes. U7ith water it forms very remarkable ice crystals; if a shallow glass of ethyl chloride is exposed to the air, it dries up and rapidly falls in temperature and ice crystals form round the edge, condensed from the atmospheric moisture. These crystals appear to be tubular and grow rapidly, the ethyl chloride soaking up inside them as in a sponge. Similar crystals form at the throttle valve in a refrigerating machine if there is any water dissolved in the ethyl chloride and quickly choke the valve. I t is important, therefore, to keep the material dry in refrigerating plants. 1971 98 ETHYL CHLORIDE (C,H,Cl) Ethyl chloride is a convenient cooling material for use in laboratories ; when sprayed on to any object the temperature is quickly reduced to - 20' C.or lower. The vapour compression cycle is the one used. The pressures corresponding to the ranges of temperature commonly required are very low, for example for a range of k 20' C., the pressures are 19.4 and 3.5 lb. per sq. in. absolute ; the compressor therefore works with a moderate vacuum on the suction side. The type of compressor generally used is a rotary pump (of the Roots Blower type.) Oil being soluble in ethyl chloride cannot be used for lubrication, For mechanical refrigeration ethyl chloride is very suitable. Tunpmhm 'C FIG. I. so pure glycerin is employed ; this separates out by gravity without difficulty.The only serious difficulty in designing the plant is to make certain of avoid- ing leaks on the low pressure side ; all air leaking in accumulates in the con- denser and has to be purged from time to time. To avoid air leaks all valve spindles, compressor shafts, etc., have to be sealed with glycerin, so that any leakage is of glycerin not air. Owing to the low pressures the volume of the pump and the cross sections of the pipes are enormously larger than those in carbonic acid or ammonia plants. On the other hand the strength required is very little, which makes the machinery very safe. Reciprocating pumps are not suitable for compressors because the pistonETHYL CHLORIDE (C,H,Cl) I99 rod, which is lubricated with glycerin, carries moisture into the pump every stroke. The writer has had great difficulties with his research plant from this cause. To get over this various " non-flam ethyl chlorides " have been put on the market. These are mixtures of ethyl chloride with large proportions of methyl bromide. The mixture is not only non-inflammable, but may be used as a fire extinguisher. Its properties as a refrigerator are not very different from those of ethyl chloride, but as it is heavier than glycerin, special sepal ators are necessary. Impure ethyl chloride is supplied for refrigeration ; it contains a small proportion of methyl chloride which may be slightly advantageous, as it raises the vapour pressure. The writer is carrying out a complete investigation of the thermal properties of ethyl chloride for the Food Investigation Board (Department of Scientific and Industrial Research) and complete 8-+ and I-+ charts will shortly be published. The inflammability of ethyl chloride is its most serious drawback.

ISSN:0014-7672    

DOI:10.1039/TF9221800197

出版商:RSC    

年代:1922

数据来源: RSC

摘要:

118 ELECTRICAL THEORY OF ADBORPTTON The writer considers the double layer as consisting of a swface of rigidly fixed atoms under continuous bombardment of positively and negatively charged ions, any particular point on the rigid surface becoming in turn negative, neutral and positive, these conditions arisdg in any order. The observed contact difference is the average effect of these conditions. Where several kinds of atoms are present in the solution the average number of any one of them at the surface will depend on their concentbration, valency and mobility. The variation of contact Werence from negative to neutral and positive was observed with cotton and aluminium sulphate near the neutral point. These variations occurred during the same experiment, the readings being direct measurements of E.1I.F.s developed by filtration under pressure.This point would be covered by putting n2 = 1 and = 2 or 3 in Mukherjee’s equation No. 13.118 ELECTRICAL THEORY OF ADBORPTTON The writer considers the double layer as consisting of a swface of rigidly fixed atoms under continuous bombardment of positively and negatively charged ions, any particular point on the rigid surface becoming in turn negative, neutral and positive, these conditions arisdg in any order. The observed contact difference is the average effect of these conditions. Where several kinds of atoms are present in the solution the average number of any one of them at the surface will depend on their concentbration, valency and mobility. The variation of contact Werence from negative to neutral and positive was observed with cotton and aluminium sulphate near the neutral point.These variations occurred during the same experiment, the readings being direct measurements of E.1I.F.s developed by filtration under pressure. This point would be covered by putting n2 = 1 and = 2 or 3 in Mukherjee’s equation No. 13.118 ELECTRICAL THEORY OF ADBORPTTON The writer considers the double layer as consisting of a swface of rigidly fixed atoms under continuous bombardment of positively and negatively charged ions, any particular point on the rigid surface becoming in turn negative, neutral and positive, these conditions arisdg in any order. The observed contact difference is the average effect of these conditions. Where several kinds of atoms are present in the solution the average number of any one of them at the surface will depend on their concentbration, valency and mobility.The variation of contact Werence from negative to neutral and positive was observed with cotton and aluminium sulphate near the neutral point. These variations occurred during the same experiment, the readings being direct measurements of E.1I.F.s developed by filtration under pressure. This point would be covered by putting n2 = 1 and = 2 or 3 in Mukherjee’s equation No. 13.118 ELECTRICAL THEORY OF ADBORPTTON The writer considers the double layer as consisting of a swface of rigidly fixed atoms under continuous bombardment of positively and negatively charged ions, any particular point on the rigid surface becoming in turn negative, neutral and positive, these conditions arisdg in any order.The observed contact difference is the average effect of these conditions. Where several kinds of atoms are present in the solution the average number of any one of them at the surface will depend on their concentbration, valency and mobility. The variation of contact Werence from negative to neutral and positive was observed with cotton and aluminium sulphate near the neutral point. These variations occurred during the same experiment, the readings being direct measurements of E.1I.F.s developed by filtration under pressure. This point would be covered by putting n2 = 1 and = 2 or 3 in Mukherjee’s equation No. 13.118 ELECTRICAL THEORY OF ADBORPTTON The writer considers the double layer as consisting of a swface of rigidly fixed atoms under continuous bombardment of positively and negatively charged ions, any particular point on the rigid surface becoming in turn negative, neutral and positive, these conditions arisdg in any order.The observed contact difference is the average effect of these conditions. Where several kinds of atoms are present in the solution the average number of any one of them at the surface will depend on their concentbration, valency and mobility. The variation of contact Werence from negative to neutral and positive was observed with cotton and aluminium sulphate near the neutral point. These variations occurred during the same experiment, the readings being direct measurements of E.1I.F.s developed by filtration under pressure.This point would be covered by putting n2 = 1 and = 2 or 3 in Mukherjee’s equation No. 13. ETHYL CHLORIDE (C,H,Cl) I99 DISCUSSION ON ‘‘ LABORATORY METHODS OF LIQUEFACTION.” Professor C. F. Jenkin: May I ask Dr. Crommelin the number of benches that can be accommodated with apparatus of this sort. I think it would be very interesting to know how many people can be doing work at these low temperatures a t the Leyden Laboratory. Is it only possible for one or say twenty? Dr. Crornrnelin: I t is very hard to say a definite number, but it would be ten or twelve research students, or something like that. I t de- pends entirely upon the kind of investigation. Dr. M. W. Travers: Like many others I have been profoundIy interested in Professor Crommelin’s account of the work of the Leyden Laboratory, which I had the honour of visiting twenty years ago in company with Sir William Ramsay.I n those days we were interested in low temperature research, but I am afraid if I showed on the screen pictures of our plant they would raise a smile. Our interest in liquid air and liquid hydrogen was the outcome of investigations connected with the rare gases of the atmosphere. We began our work without a liquefying plant, obtaining liquid air through the kind offices of Dr. Hampson and Mr. I(. S. Murray, of what is now the British Oxygen Company, who was then commencing experiments, the development from which he has told us about this evening. Small quantities of liquid air, brought to our laboratory at seven o’clock in the evening, were our only supplies for the first nine months that we were engaged on this work.Early in the year 1899, Mr. Rose Innes presented the laboratory with a Hampson liquid air machine, and Dr. Ludwig Blond provided the funds for the purchase of a compressor, making it possible for us to carry out our work under reasonable conditions. Later in this investigation it became clear that it would only be possible to obtain pure neon by separating it from argon and helium by means of liquid hydrogen. The problem of liquefying hydrogen without spending much more than on the work fell to my lot. ‘The liquefier was constructed from stock sizes of brass tube, punched2 00 DISCUSSION blanks, etc., with only a single casting. The hydrogen was made in a beer barrel.An exhaust pump which formed part of the plant was a blowing pump, made by one of the students in his own workshop, with the valves reversed, and it was driven by a gas engine of a primitive type, borrowed from the engineering museum. However, the plant worked satisfactorily, and gave us sufficient liquid hydrogen for our purpose. Later, I reconstructed it, and carried out some researches at temperatures down to 14' K. That was in 1903, and since then my work has lain in other directions. I want to ask one question, which arises out of Professor Porter's remarks. I always found that I got the best results with my hydrogen liquefier when the pressure at the top of the first cooling coil was about 1 2 0 atmospheres, or less. The pressure at the jet must have been lower.This was the pressure which was generally maintained at the gauge while the liquefaction of the hydrogen was actually proceeding. I may add that, on one occasion, when giving a demonstration of the liquefaction of hydrogen in Berlin, the gas was obtained from cylinders, and was successfully liquefied at a much lower gauge pressure. We tried working at higher pressures, but I never remember seeing the pressure indicated by the gauge standing at over 120 atmospheres while the liquid was running into the receiver. I should like to ask Professor Crommelin if he has observed an upper limit of pressure above which the cooling on expansion appears to diminish? I believe that he told us that the helium plant was run at a low pressure. Dr. Crommelin: 25 to 30 atmospheres with the helium plant and 150 to 160 with the hydrogen plant.Dr. Travers : Do you find that at 180 atmospheres you get a better or a worse result'? Dr. Crommelin: There is not much difference, even at 2 2 0 atmos- pheres the producing capacity is about the same. Professor Porter: I do not expect hydrogen to lower at all in temperature in throttle expansion at any temperature if the pressure is higher than 240 atmospheres. I was, at the beginning, very much afraid that the condensation of even small quantities of air would lead to blocking of the pipes and stoppage of the apparatus. Though my hydrogen undoubtedly contained traces of air, it gave me no trouble. On the other hand, when a block occurred, it was always traced to oil from the plant. Is that your experience, Professor Crommelin ? Dr.Crommelin: Blockings may occur on account of impurities in the gas or of oil vapour ; we use oil separators, so that there is no trouble with the oil. Dr. Travers: My next point is of practical importance. Dr. Travers : You take some precautions, however. Dr. Crommelin: Yes, there are oil separators, but I am not quite aware of the details of their construction. There was in the beginning a good deal of trouble with the liquid hydrogen apparatus. If you worked the plant for half an hour it was all right. If you worked for a longer time then you had trouble. At the Reichsanstalt they had trouble when working for long periods, and they had to put in a heating spiral, so as to be able to heat the apparatus up, clear it, and then cool down again.Dr. Travers: Professor Crommelin speaks of 13 litres of liquid hydrogen an hour. We never obtained more than 6 a litre, which was, in those days, a very large quantity.DISCUSSION 201 Dr. J. A. Harker: I would like to say one or two words to endorse your expression of thanks to Dr. Crommelin for coming here this afternoon and giving us these very interesting papers. I was able, some ten years ago, to visit Leyden and see some of the things which he has de- scribed to us. I remember that on that occasion Professor Onnes and Dr. Crommelin were both present and I had the extreme pleasure of their own exposition of some of the things they were doing. One thing I. re- member-and it remained with me-which Dr. Crommelin has perhaps not sufficiently emphasised to-day, was the marvellous dependence of the epoch-making work that has been done at Leyden on extremely delicate and difficult apparatus-building, and more particularly on apparatus made of glass.One of the things that impressed me most at that time was something which Dr. Crommelin referred to this afternoon as a more or less ordinary matter-the use of vacuum vessels of extremely complicated construction, without which many of the results they have obtained could never have been reached. When at Leyden I was shown a piece of the helium plant, which was a vacuum vessel about as thick as one’s little finger, bent round at a right angle. This was not a single vacuum vessel but it enclosed another concentric with it, the tubes of both being of egg- shell thickness. Anyone reasonably skilled can make a vacuum vessel of sorts, but it is quite another matter to make one vacuum vessel inside another.Professor A. W. Porter: There is only one point with which I want to deal, and that is in connection with the measurement of the extremely low temperatures to which we have been introduced to-night. Professor Onnes has attempted to obtain the value of these extremely low temperatures (below I’ K) by plotting a curve of the reciprocals of the reduced temperatures against the logarithms of the reduced vapour pressures as far as the directly determinable values go and then extrapolat- ing by means of a linear law. I wish to suggest that it may be possible to obtain a truer extrapolation by means of one of the various theoretical curves which are adopted for representing vapour pressures.Approximately, for many substances at any rate, we can write, b - T = aye 7 , where T and y are reduced pressure and temperature respectively ; or since 7r = I when y = I, log 7~ = log the slope will be, b’ Y - - + b’. On a curve with the same co-ordinates as Kamerlingh Onnes uses hence it is greatest at the origin and diminishes as 5 increases-ultimately, but gradually, becoming straight. I have determined the constant b‘ in such an equation taking as fixed point the experimental point for = 3, so as to be well within the experimental region. The value of 6’ is 8.46 and the curve is shown in the figure by a continuous line; Kamerlingh Onnes’ curve is shown Y Y202 DISCUSSSION dotted.From the curve I have read off the temperatures corresponding to various pressures and tabulate them below along with values from K. Onnes' curve for comparison. These remarks are only intended to indicate that it would be worth while specially to study the general form of the vapour pressure curve in the non-extrapolated region and to employ the information so gained to extend our knowledge of temperature values. I Temperature from Pressure mm. Mercury. I<. Onnes. I 0'1 '05 '0 2 '0 I '005 '002 '001 I' K 0.89 -82 '79 '74 '70 '65 Porter. I' K 0'94 '87 '785 -825 '74 -71 Professor Kamerlingh Onnes (rqh communicated, December I 6fh, 1922) : I quite agree with Professor Porter that it is of importance to study experimentally the vapour pressure curve in the non-extrapolated region to get better information on the temperature values in the lower one.There is a new investigation of this difficult subject in hand andDISCUSSION 203 I trust we shall succeed in extending it to lower temperatures and in obtaining more accurate data than are availakle at present. The present data seemed compatible with a curve that for lower reduced temperatures becomes finally rectilinear, in analogy with the cases given in Fig. 9, 14. When this was the case, then by drawing the tangent to the experimental curve the values of temperature obtained would at least not be too low. By extrapolation according to a theoretical formula of the form of that given by Professor Porter lower temperatures, and not as Professor Porter finds higher temperatures, were found. Professor Verschaffelt who calculated the formula which was used in this extrapolation had emphasised the theoretical meaning of by introducing into it for A the reduced chemical constant for helium, 0'59, and for C the theoretical value for monatomic gases, 2.5, and then fitting B to the observations. Assuming log,,, ;r = - 2-18 for y = 8, according to the plotting, one finds R = 0 - 5 j ; so we should have, below y = 8. The temperatures read from the curve represented by this formula, viz. : 0'1 0.94 '05 -8 7 '02 *SO ' 0 I '75 so05 ' 7 0 '002 '65 '00 I *6 2 are even lower than those found by rectilinear extrapolation. So the results, as I said, indicate that with the linear extrapolation we were on the safe side. Differences of some hundredths of a degree such as occur between the results of Professor Porter and of myself are after all to be considered as within the limit of uncertainty of the experimental data from which we have to start.

ISSN:0014-7672    

DOI:10.1039/TF9221800199

出版商:RSC    

年代:1922

数据来源: RSC

摘要:

118 ELECTRICAL THEORY OF ADBORPTTON The writer considers the double layer as consisting of a swface of rigidly fixed atoms under continuous bombardment of positively and negatively charged ions, any particular point on the rigid surface becoming in turn negative, neutral and positive, these conditions arisdg in any order. The observed contact difference is the average effect of these conditions. Where several kinds of atoms are present in the solution the average number of any one of them at the surface will depend on their concentbration, valency and mobility. The variation of contact Werence from negative to neutral and positive was observed with cotton and aluminium sulphate near the neutral point. These variations occurred during the same experiment, the readings being direct measurements of E.1I.F.s developed by filtration under pressure.This point would be covered by putting n2 = 1 and = 2 or 3 in Mukherjee’s equation No. 13. PART 11. - INDUSTRIAL METHODS OF LIQUEFAC- TION AND YRAC'TICAL APPLICATIONS OF LOW TEMPERATURES. Mr. George Goodsir, President of the British Cold Storage and Ice Association, opened this section of the proceedings with the following remarks :- Before calling on the authors who are going to give us their papers, I want to say a word or two on behalf of the Society which I represent. First, I would like to express their best thanks to the Faraday Society for the honour they have done us in inviting us to meet with you and confer upon the subject of low temperatures. The Society which I represent is, in name, quite new, because this is practically the first official gathering under the name of the British Cold Storage and Ice Association.I t is an amalgamation of two Societies which have been in existence, the one since 1910, and the other since ten years later. Their members are largely engaged in practical engineering, ship-owning and general refrigerating business and store keeping of various kinds, and its interests do not, therefore compete in any way with the Faraday Society which is inter- ested in the scientific side of refrigeration. I am afraid that when the discussion comes to deal with liquid air and such matters, we may find ourselves a little out of our depth. The temperatures which you handle quite unconcernedly are far beyond our wildest imaginations in the nature of cold, and our lowest temperatures are mere summer coolness beside what you are accustomed to speak about. Some day, perhaps, liquid air and other recent products of science may be made available for the ordinary work of the preservation of perishable food, but I do not think that is just yet. Meantime, this discussion which is to take place to-day may, perhaps, open our eyes to the possibilities in that direction, and although our members may not, many of them, be able to assist you very practically in the discussion, we will listen and learn, I am sure, and we will do what we can to add to the knowledge on the subject which may spring from our practical experience. With these few preliminary remarks I have pleasure in calling upon Mr. K. S. Murray, a gentleman well acquainted with the practical liquefaction of air for many years past, to open this section of the proceedings. 204

ISSN:0014-7672    

DOI:10.1039/TF9221800204

出版商:RSC    

年代:1922

数据来源: RSC

摘要:

118 ELECTRICAL THEORY OF ADBORPTTON The writer considers the double layer as consisting of a swface of rigidly fixed atoms under continuous bombardment of positively and negatively charged ions, any particular point on the rigid surface becoming in turn negative, neutral and positive, these conditions arisdg in any order. The observed contact difference is the average effect of these conditions. Where several kinds of atoms are present in the solution the average number of any one of them at the surface will depend on their concentbration, valency and mobility. The variation of contact Werence from negative to neutral and positive was observed with cotton and aluminium sulphate near the neutral point. These variations occurred during the same experiment, the readings being direct measurements of E.1I.F.s developed by filtration under pressure.This point would be covered by putting n2 = 1 and = 2 or 3 in Mukherjee’s equation No. 13.118 ELECTRICAL THEORY OF ADBORPTTON The writer considers the double layer as consisting of a swface of rigidly fixed atoms under continuous bombardment of positively and negatively charged ions, any particular point on the rigid surface becoming in turn negative, neutral and positive, these conditions arisdg in any order. The observed contact difference is the average effect of these conditions. Where several kinds of atoms are present in the solution the average number of any one of them at the surface will depend on their concentbration, valency and mobility. The variation of contact Werence from negative to neutral and positive was observed with cotton and aluminium sulphate near the neutral point.These variations occurred during the same experiment, the readings being direct measurements of E.1I.F.s developed by filtration under pressure. This point would be covered by putting n2 = 1 and = 2 or 3 in Mukherjee’s equation No. 13.118 ELECTRICAL THEORY OF ADBORPTTON The writer considers the double layer as consisting of a swface of rigidly fixed atoms under continuous bombardment of positively and negatively charged ions, any particular point on the rigid surface becoming in turn negative, neutral and positive, these conditions arisdg in any order. The observed contact difference is the average effect of these conditions. Where several kinds of atoms are present in the solution the average number of any one of them at the surface will depend on their concentbration, valency and mobility.The variation of contact Werence from negative to neutral and positive was observed with cotton and aluminium sulphate near the neutral point. These variations occurred during the same experiment, the readings being direct measurements of E.1I.F.s developed by filtration under pressure. This point would be covered by putting n2 = 1 and = 2 or 3 in Mukherjee’s equation No. 13.118 ELECTRICAL THEORY OF ADBORPTTON The writer considers the double layer as consisting of a swface of rigidly fixed atoms under continuous bombardment of positively and negatively charged ions, any particular point on the rigid surface becoming in turn negative, neutral and positive, these conditions arisdg in any order.The observed contact difference is the average effect of these conditions. Where several kinds of atoms are present in the solution the average number of any one of them at the surface will depend on their concentbration, valency and mobility. The variation of contact Werence from negative to neutral and positive was observed with cotton and aluminium sulphate near the neutral point. These variations occurred during the same experiment, the readings being direct measurements of E.1I.F.s developed by filtration under pressure. This point would be covered by putting n2 = 1 and = 2 or 3 in Mukherjee’s equation No. 13.118 ELECTRICAL THEORY OF ADBORPTTON The writer considers the double layer as consisting of a swface of rigidly fixed atoms under continuous bombardment of positively and negatively charged ions, any particular point on the rigid surface becoming in turn negative, neutral and positive, these conditions arisdg in any order.The observed contact difference is the average effect of these conditions. Where several kinds of atoms are present in the solution the average number of any one of them at the surface will depend on their concentbration, valency and mobility. The variation of contact Werence from negative to neutral and positive was observed with cotton and aluminium sulphate near the neutral point. These variations occurred during the same experiment, the readings being direct measurements of E.1I.F.s developed by filtration under pressure.This point would be covered by putting n2 = 1 and = 2 or 3 in Mukherjee’s equation No. 13.118 ELECTRICAL THEORY OF ADBORPTTON The writer considers the double layer as consisting of a swface of rigidly fixed atoms under continuous bombardment of positively and negatively charged ions, any particular point on the rigid surface becoming in turn negative, neutral and positive, these conditions arisdg in any order. The observed contact difference is the average effect of these conditions. Where several kinds of atoms are present in the solution the average number of any one of them at the surface will depend on their concentbration, valency and mobility. The variation of contact Werence from negative to neutral and positive was observed with cotton and aluminium sulphate near the neutral point.These variations occurred during the same experiment, the readings being direct measurements of E.1I.F.s developed by filtration under pressure. This point would be covered by putting n2 = 1 and = 2 or 3 in Mukherjee’s equation No. 13.118 ELECTRICAL THEORY OF ADBORPTTON The writer considers the double layer as consisting of a swface of rigidly fixed atoms under continuous bombardment of positively and negatively charged ions, any particular point on the rigid surface becoming in turn negative, neutral and positive, these conditions arisdg in any order. The observed contact difference is the average effect of these conditions. Where several kinds of atoms are present in the solution the average number of any one of them at the surface will depend on their concentbration, valency and mobility.The variation of contact Werence from negative to neutral and positive was observed with cotton and aluminium sulphate near the neutral point. These variations occurred during the same experiment, the readings being direct measurements of E.1I.F.s developed by filtration under pressure. This point would be covered by putting n2 = 1 and = 2 or 3 in Mukherjee’s equation No. 13.118 ELECTRICAL THEORY OF ADBORPTTON The writer considers the double layer as consisting of a swface of rigidly fixed atoms under continuous bombardment of positively and negatively charged ions, any particular point on the rigid surface becoming in turn negative, neutral and positive, these conditions arisdg in any order.The observed contact difference is the average effect of these conditions. Where several kinds of atoms are present in the solution the average number of any one of them at the surface will depend on their concentbration, valency and mobility. The variation of contact Werence from negative to neutral and positive was observed with cotton and aluminium sulphate near the neutral point. These variations occurred during the same experiment, the readings being direct measurements of E.1I.F.s developed by filtration under pressure. This point would be covered by putting n2 = 1 and = 2 or 3 in Mukherjee’s equation No. 13.118 ELECTRICAL THEORY OF ADBORPTTON The writer considers the double layer as consisting of a swface of rigidly fixed atoms under continuous bombardment of positively and negatively charged ions, any particular point on the rigid surface becoming in turn negative, neutral and positive, these conditions arisdg in any order.The observed contact difference is the average effect of these conditions. Where several kinds of atoms are present in the solution the average number of any one of them at the surface will depend on their concentbration, valency and mobility. The variation of contact Werence from negative to neutral and positive was observed with cotton and aluminium sulphate near the neutral point. These variations occurred during the same experiment, the readings being direct measurements of E.1I.F.s developed by filtration under pressure. This point would be covered by putting n2 = 1 and = 2 or 3 in Mukherjee’s equation No.13.118 ELECTRICAL THEORY OF ADBORPTTON The writer considers the double layer as consisting of a swface of rigidly fixed atoms under continuous bombardment of positively and negatively charged ions, any particular point on the rigid surface becoming in turn negative, neutral and positive, these conditions arisdg in any order. The observed contact difference is the average effect of these conditions. Where several kinds of atoms are present in the solution the average number of any one of them at the surface will depend on their concentbration, valency and mobility. The variation of contact Werence from negative to neutral and positive was observed with cotton and aluminium sulphate near the neutral point. These variations occurred during the same experiment, the readings being direct measurements of E.1I.F.s developed by filtration under pressure.This point would be covered by putting n2 = 1 and = 2 or 3 in Mukherjee’s equation No. 13.118 ELECTRICAL THEORY OF ADBORPTTON The writer considers the double layer as consisting of a swface of rigidly fixed atoms under continuous bombardment of positively and negatively charged ions, any particular point on the rigid surface becoming in turn negative, neutral and positive, these conditions arisdg in any order. The observed contact difference is the average effect of these conditions. Where several kinds of atoms are present in the solution the average number of any one of them at the surface will depend on their concentbration, valency and mobility.The variation of contact Werence from negative to neutral and positive was observed with cotton and aluminium sulphate near the neutral point. These variations occurred during the same experiment, the readings being direct measurements of E.1I.F.s developed by filtration under pressure. This point would be covered by putting n2 = 1 and = 2 or 3 in Mukherjee’s equation No. 13.118 ELECTRICAL THEORY OF ADBORPTTON The writer considers the double layer as consisting of a swface of rigidly fixed atoms under continuous bombardment of positively and negatively charged ions, any particular point on the rigid surface becoming in turn negative, neutral and positive, these conditions arisdg in any order. The observed contact difference is the average effect of these conditions.Where several kinds of atoms are present in the solution the average number of any one of them at the surface will depend on their concentbration, valency and mobility. The variation of contact Werence from negative to neutral and positive was observed with cotton and aluminium sulphate near the neutral point. These variations occurred during the same experiment, the readings being direct measurements of E.1I.F.s developed by filtration under pressure. This point would be covered by putting n2 = 1 and = 2 or 3 in Mukherjee’s equation No. 13.118 ELECTRICAL THEORY OF ADBORPTTON The writer considers the double layer as consisting of a swface of rigidly fixed atoms under continuous bombardment of positively and negatively charged ions, any particular point on the rigid surface becoming in turn negative, neutral and positive, these conditions arisdg in any order.The observed contact difference is the average effect of these conditions. Where several kinds of atoms are present in the solution the average number of any one of them at the surface will depend on their concentbration, valency and mobility. The variation of contact Werence from negative to neutral and positive was observed with cotton and aluminium sulphate near the neutral point. These variations occurred during the same experiment, the readings being direct measurements of E.1I.F.s developed by filtration under pressure. This point would be covered by putting n2 = 1 and = 2 or 3 in Mukherjee’s equation No. 13.118 ELECTRICAL THEORY OF ADBORPTTON The writer considers the double layer as consisting of a swface of rigidly fixed atoms under continuous bombardment of positively and negatively charged ions, any particular point on the rigid surface becoming in turn negative, neutral and positive, these conditions arisdg in any order.The observed contact difference is the average effect of these conditions. Where several kinds of atoms are present in the solution the average number of any one of them at the surface will depend on their concentbration, valency and mobility. The variation of contact Werence from negative to neutral and positive was observed with cotton and aluminium sulphate near the neutral point. These variations occurred during the same experiment, the readings being direct measurements of E.1I.F.s developed by filtration under pressure.This point would be covered by putting n2 = 1 and = 2 or 3 in Mukherjee’s equation No. 13.118 ELECTRICAL THEORY OF ADBORPTTON The writer considers the double layer as consisting of a swface of rigidly fixed atoms under continuous bombardment of positively and negatively charged ions, any particular point on the rigid surface becoming in turn negative, neutral and positive, these conditions arisdg in any order. The observed contact difference is the average effect of these conditions. Where several kinds of atoms are present in the solution the average number of any one of them at the surface will depend on their concentbration, valency and mobility. The variation of contact Werence from negative to neutral and positive was observed with cotton and aluminium sulphate near the neutral point. These variations occurred during the same experiment, the readings being direct measurements of E.1I.F.s developed by filtration under pressure.This point would be covered by putting n2 = 1 and = 2 or 3 in Mukherjee’s equation No. 13. INI>USTRZA I, METHOlIS OF L1QUElJL%CTION AND PRACTICAL APPi,ICATIONS 01; LOIV TEMPERATURES. BY K. S. MUKRAY, M.I.MEcI-I.E. 'The title of this paper was not selected by myself, IVere I to deal with every subject which might legitimately be collected under such a comprehensive umbrella, I should require far more time than has been allotted to me. I have therefore decided to deal only with the subject of the separation of the constituents of air, with special reference to low temperature liquefaction.When people talk of separating the constituents of the air, nine out of every ten have in view only the extraction of oxygen from the atmosphere, and, as I have been engaged in extracting vastly more than my fair share for the last thirty-five years, the subject is one on which I should be able to speak with some degree of authority. I n fact, the history of the British Oxygen Company, with which I have been associated for the whole of that period, is so intimately bound up with the industrial development of oxygen that it is impossible, at any rate for me, to consider these subjects apart. Therefore, for historical accuracy, before dealing with low temperature extraction of oxygen from the atmosphere, I propose to refer very briefly to the barium oxide process for effecting the same purpose, as it was un- doubtedly by that process, in the hands of the British Oxygen Company, that the oxygen industry was founded.The British Oxygen Company was established in 1886 as the Brins Oxygen Company, to take over an alleged process for the production of oxygen, patented by two French brothers whose name was adopted as the title of the Company. The Brin process was based on Bous- singault's discovery in 1851, that at a temperature of about 540' C. the monoxide of the metal barium would absorb oxygen readily from the at- mosphere with the resulting formation of dioxide, and that at a higher temperature of about 870" C. the oxygen thus absorbed would be given off again and the barium restored to the monoxide condition ready for the cycle to be repeated.As a natural consequence of Boussingault's discovery many efforts had been made to establish a commercial process for the production of oxygen based on this apparently unalterable property of barium oxide. In spite, however, of its chemical simplicity, many practical difficulties arose which remained unsurmounted until the advent of the Brim Oxygen Company. Even then it was only after prolongedand costly experimental work which resulted in the Brin process being practically thrown overboard, that the barium method of abstracting oxygen from the atmosphere became an established success. I described the process fully in a paper I read before the Institution of Mechanical Engineers on January 31s1, 1890, and I ZOj206 INDUSTRIAL METHODS OF LIQUEFACTION AND cannot do better than refer to that paper anyone who may be still interested in the process.Here it is sufficient to say that the fundamental changes which we effected were : (I) vvorking at a constant furnace temperature of about 650" C., change of pressure being relied upon for detertnining the respective Frti. ~,-Hampsoii's Air Liquefier. phases of oxidation and deoxidation ; (2) short operations lasting only a few minutes, and (3) the automatic reversal of the various valve gears involved in operating the process. 'The process was exclusively employed by the British Oxygen Company for nearly twenty years in their London, Birmingham, Manchester, and Glasgow factories, and a considerable number of plants, varying in capacityFIG. westmin minster Works, B.O.Co., 1906. [To face Page 207.PRACTICAL APPLICATIONS OF LOW TEMPERATURES 2 0 7 from 5000 to 30,000 cubic feet of oxygen per day, were also erected by them to provide oxygen for industrial applications. A fact which is especially worthy of mention is that the first plants for the commercial production of oxygen in Paris, Berlin, and New York, were of the barium oxide type and were erected by the British Oxygen Company. I am afraid not a single barium plant remains in existence to-day although in point of economy the process does not compare unfavourably with liquid oxygen plants of the smaller type, and it was abandoned mainly because it was incapable of producing oxygen of equal purity.As early as 1896 the British Oxygen Company, as the makers of Harnpson’s apparatus for the liquefaction of air, became connected with low temperature research. The Hampson apparatus (Fig. I), which is based on a nozzle, or free internal, expansion, is so efficient in its tempera- ture interchange, that when expanding about 700 cubic feet of air from a pressure of 150 atmospheres to atmospheric pressure, it begins to produce liquid in about 6 minutes with a yield of about I litre of liquid per hour. For the production of liquid air on a commercial scale, this apparatus is too small to be of any value. On the other hand it is specially designed for the quick production of liquid for research purposes and it has been adopted for that purpose in many educational establishments and research laboratories throughout the world.The success which attended the introduction of this apparatus, coupled with the still more important work of Professor Carl von Linde in the liquefaction ef air, pointed to a method for the separation of its constituents which might develop into a rival to the barium process ; consequently, the British Oxygen Company, like other concerns equipped with facilities for the production of liquid air, were quietly experimenting between 1896 and 1903 with various methods for separating the constituents of liquid air by partial evaporation. I t must, however, be admitted that the problem remained entirely unsolved until the process of rectification was disclosed by Professor Linde in the latter year.From that date onwards rectification has been the fundamental principle underlying every patented method for separating the constituents of liquid air. The British Oxygen Company were not long thereafter in realising that their barium process for producing oxygen was at length seriously threatened by the introduction of rectification for separating the constituents of liquid air. We therefore concluded an arrangement with Professor Linde under which we acquired his British and British Colonial rights in rectification. I n 1906 we erected in our Westminister \\’orla 3 Linde plant capable of producing about 500 cubic feet of oxygen per hour. I t is illustrated in the photograph, Fig. 2 . This photograph is of some general interest, because it shows on the left-hand side the first liquid air plant erected for the production of oxygen in this country, and on the right-hand side a portion of the last, and best, of many barium oxide plants erected by the Company.Professor Lincie himself was so struck with the efficiency and mechanical simplicity of this latter plant that, largely on his recommendation, the two plants were worked side by side for many months before it was finally decided to substitute liquid air for barium oxide plants in the production of oxygen throughout the Company’s factories. I n 1908, as the outcome of prolonged litigation, which established Linde’s claim to the rectification of liquid air, the Company entered into an agreement with the Sociitk 1’Air Liquide of France under which weZOS INDUSTRIAL METHODS OF LIQUEFACTION -4ND acquired the Claude British and British Colonial patents (with the excep- tion of Canada), for the separation of oxygen and nitrogen from liquid air.I n many of our factories we have to-day Linde and Claude plants working side by side. We therefore claim to have had exceptional facilities for comparing the two systems on which the important oxygen industry of to- day has been developed in all parts of the world. I now propose to describe and illustrate diagrammatically the essential features of the Linde and Claude systems, and as I am dealing with the subject from the industrial rather than the scientific point of view, I shall confine myself mainly to facts and figures which are based on practical experience.I must therefore assume the production of liquid air to have reached the stage where Linde’s adaptation of the Joule-Thomson effect and Claude’s expansion engine enabled that liquid to be produced in com- FIG. 3.-Linde’s single column type separator. mercial quantities. - I must also as- sume the physical laws involved in its production to be generally under- stood. The Linde plant shown on the left of Fig. z represents the earliest type employed for the separation of oxygen from liquid air. Such plants consist essentially of an air compressor; a suitable purifier for eliminating carbon dioxide; a drier for removing mois- ture from the air; a fore-cooler, which is kept cold by means of an ammonia or a carbon dioxide ma- chine; and finally, the separator containing the counter current heat interchanger and the rectification column.The working of such a plant is as follows: Air is drawn by the com- pressor through a lime or caustic soda purifier in order to remove carbonic acid. It is then compressed, under conditions as nearly isothermal as possible, to a pressure of about 135 atmospheres and passed through a drier containing calcium chloride in order to remove moisture. It then enters the fore-cooler where it is cooled to a temperature of about - soo C., any remaining traces of moisture being deposited there as ice, and finally passes to the separator of which a sectional elevation is shown diagram- matically in Fig. 3. The main pipe D conveying this compressed air to the apparatus is split up into three small pipes d at the point where it enters the counter-current interchanger coil C.I t passes down these pipes, which are reunited into a single pipe dl when they leave the interchanger and this pipe dl which is coiled in the vaporiser B as shown, is attached to the inlet of the valve G. At this point, by the adjustment of the regulating valve spindle H, the air is caused-to expand from a high to a low pressure and is discharged at this low pressure through the rose-ended pipe d, into the top of the rectification column A. The expanded air fills the column and then flows through the only two possible outlet pipes c and e which surround thePRACTICAL APPLICATIONS OF LOW TEMPERATURES 209 small pipes d thus forming a counter-current interchange of heat, as the outgoing expanded air ascends the interchanger coil through which the high pressure air is descending in the small pipes d.The expanded cold gases, having thus abstracted heat from the incoming compressed air, leave the separator at C and E as indicated in the diagram, nearly at atmospheric temperature. I t is at the valve G that Linde obtains the Joule-Thomson cooling effect and this is made cumulative by means of the counter-current heat interchanger. The apparatus is completely enclosed in a wooden casing and all spaces are packed with suitable insulating material, consequently, after starting the plant with everything at atmospheric temperature, the separator gradually cools down until some of the expanded air begins to liquefy and collect round the coils in the vaporiser B.The quantity of liquid thus collected is registered outside the apparatus by means of an ordinary coloured liquid which is contained in a glass tube b enclosed in a pressure equalising circuit as indicated. Once this stage is reached liquid begins to accumulate rapidly in the vaporiser B, and the initial pressure of the compressed air may be gradually reduced by increasing the opening of the valve G. As the liquid begins to accumulate round the coils in the vaporiser B, the compressed air transmits some of its latent heat to the liquid. The latter is thereby evaporated whilst the compressed air is itself liquefied in proportion to the amount of heat thus extracted. The vapours thus produced flow upwards through the rectification column A whilst the liquid formed in the compressed air pipe coil dl ascends through the pipe d2 and is discharged through the rose-end at the top of the rectification column. The temperature gradient necessary for efficient rectification then rapidly becomes established in the column, and the final stage is reached under which the separation of oxygen is egected.The pipes C and E of the counter-current interchanger then begin to fulfil their proper functions, the nitrogen vapours being delivered through tile former from the top of the column and the oxygen vapours through the latter from the top of the vaporiser. The separated gases leave the apparatus at C and E respectively at or near atmospheric temperature, having abstracted heat from the incoming compressed air whilst passing up tile counter-current interchanger.'l'he action which takes place in this, the simplest type of rectification column, must now be briefly explained as exactly the same principle is involved in all others. Linde relies for his temperature gradient on the difference between the boiling point of nitrogen and oxygen. The former is - 196" C., aiid the latter - 183" C. Liquid air is discharged into the top of the coluiili1 at an intermediate temperature of about - 194' C. As it passes do\\n nitrogen, being the more volatile component, immediately begins t o evaporate off and the liquid gradually becomes richer in oxygen, with a somewhat higher temperature. As this rich oxygen liquid accumulates in the vaporiser it is evaporated by the compressed air passing through the coil until it is practically pure oxygen with a temperature very nearly - 18j" C. There is thus a temperature gradient of about I I O C.established between the top and the bottom of the column. Liquid air passing dowi the column is brought into intimate contact with the rising vapours of oxygen and an exchange of material takes place. At each stage some of the rising oxygen is condensed and some of the nitrogen in the descend- ing liquid is evaporated, whilst the liquid gradually gains in tempera Lure2 I 0 INDUSTRIAL METHODS OF LIQUEFACTION AND until by the time it reaches the vaporiser its composition is that of practi- cally pure oxygen. The gas on the other hand which passes off from the top of the column through pipe C is mainly nitrogen at a tempera- ture about 11’ C.lower than that of the oxygen at the bottom of the column. The oxygen which rises in the column to effect the material exchange with the nitrogen of the descending liquid, is carried back to the vaporiser together with most of the oxygen contained in the original liquid; thus there is a constant accumulation of oxygen which passes off as vapour through the pipe E. The gases taken separately from the top and bottom of the apparatus through the counter-current interchanger, are made to abstract heat from the incoming compressed air on its way to be liquefied. When normal oxygen-producing conditions are established, the air supply from the compressor becomes reduced from 135 atmospheres to about 55 atmospheres, the latter pressure being found sufficient to make good all thermal losses due to friction, leakage of heat from outside, and imperfect interchange in the counter-current system.This type of separator is easy to operate and from an economical point of view has been found satisfactory hitherto for use in relatively small oxygen producing plants. A defect, however, of rectification in this form is that although oxygen of high purity can be produced, at least 7 per cent. of oxygen is contained in the waste nitrogen. Bayly’s well-known experi- ments in 1900 demonstrated that the vapour in equilibrium with any liquid mixture of oxygen and nitrogen always contained more nitrogen than the liquid and that when the evaporating liquid is air, the proportion of oxygen present in the vapour is 7 per cent. Therefore, in any rectification column, such as this early Linde type, where liquid air enters at the top, a loss, amounting to about one-third of the oxygen in the air, is inevitable. With large plants this becomes a serious matter, and I believe I am correct in stating that Claude was the first person to design a separator which obviates this loss.I will, therefore, now, in order to maintain chronological accuracy, briefly describe a type of Claude plant capable of producing 4,000 cubic feet of oxygen per hour, which is largely employed by the British Oxygen Company at the present time in nearly all of their factories in this country. The outstanding features of Claude’s plmts are : (I) Cooling by means of an expansion engine. (2) A selective system of rectification rendered possible by a preliminary partial separation of oxygen and nitrogen.(3) A combined system of heat interchange and drying which renders chemical abstraction of moisture from the air unnecessary. The working of the plant is as follows : Air is drawn by a compressor through purifiers to yemove carbon dioxide as in the Linde plant. I t is then compressed to a maximum pressure of about 33 atmospheres and passed through two vertical tubular heat interchangers in series. In pass- ing through the second of these it meets in counter-current the cold separated oxygen and nitrogen passing outwards from the separator and a very complete interchange of heat takes place. Here also most of the moisture is thrown down as water and drawn off at intervals, the remainder being arrested in the form of ice round the tubes at the top of the inter- changer.Meanwhile the first interchanger which had previously per- formed the same function, is being thawed out by the air passing through it, in order to be ready for subsequent use. Thus these interchangers are operated as such alternately, being changed over at intervals of 8 hours. The compressed air leaves the second interchanger at a tempera-PRAC'1'ICAL AYPLICATIONS OF LOW TEMPERATURES 2 I I ture of about - 100' C. 60 to 70 per cent. ofthe cooled air goes direct to the expansion engine where, in expanding to about 4 atmospheres, it per- forms external work against a dynamo brake and issues from the engine a t a temperature not far removed from its point of liquefaction. The remainder of the air passes under full pressure into an interchanger, termed the liquefier, where it is liquefied by meeting in counter-current all $he cold oxygen and nitrogen passing from the separator to the second, or operating, interchanger previously described.The cold vapour from the expansion engine and the liquid from the liquefier both enter the outer Compartment A of the liquid collecting vessel shown in the diagram (Fig. 4) at the bottom of the vaporiser, the liquid supply being controlled by the valve shown, which performs the same function as the valve G in the Linde apparatus already described. Bv this skilful combination of internal and external expansion, Claude , is able to obtain a high efficiency from his expansion engine. For this latter purpose it is obviously desirable to keep the compressed air as warm as possible.By passing the cold gases through a liquefier a twofold purpose is served. First, a moderate supply of liquid air for making good cold losses is assured by very effective heat inter- change, coupled with the Linde method of internal expansion. Second, the heat usefully abstracted by the cold gases from the compressed air pass- ing in counter - current through the liquefier, raises their temperature to a t point which prevents them abstracting, at a later stage, too much heat from the sion engine. C- D- BI, compressed air passing through the main interchanger to feed the expan- Reverting to the diagram, it will I hope now be obvious that a regular and adjustable supply of liquid air and intensely cold vapour is fed to the -+-= -+* FIG.4.-Claude's separator. outer compartment A below the vapor- iser. The vapour thus ascends the vertical nest of tubes B (leading from the top of this compartment), which are immersed in baths of liquid oxygen C and D. I n these tubes the va- pour undergoes progressive liquefaction and oxygen being the less volatile constituent, it is the more readily condensed and flows back as liquid into the compartment. A partial scrubbing effect, or rectification, also takes place, consequently the vapour passing out at the top of the tubes is very rich in nitrogen. This rich nitrogen vapour travels down a similar nest of tubes leading to the inner compartment E of the bottom vessel which it reaches partly in the liquid condition.Thus in practice liquid containing about 40 per cent. oxygen is obtained in the outer compartment and liquid containing about 4 per cent. oxygen is obtained in the inner one. A pressure of 4 atmospheres is necessary in these compartments, and the tubes communicating with them, in order to raise the liquefying point of the oxygen-nitrogen mixtures above - 183" C., the temperature of the liquid VOL. XVIII-TI02 I z INDUSTRIAL METHODS OF LIQUEFACTION AND oxygen bath, so that liquefaction may take place and that latent heat may pass to the liquid oxygen and evaporate it to a corresponding extent. Above the vaporiser, as in the Linde apparatus, is the rectification column. The liquid, rich in nitrogen, is conveyed from the inner compartment E to the top of the column at H.The liquid containing about 40 per cent. oxygen is conveyed from the outer compartment A to the column at the point K where the composition of the ascending gases should correspond to that of a gas mixture in equilibrium with a 40 per cent. oxygen liquid. Rectification with the correct temperature gradient then proceeds as already described in connection with the Linde separator (Fig. 3) with however the important difference that, as rich liquid nitrogen replaces liquid air at the top of the column, a correspondingly lower temperature is maintained there, thus extending rectification, with the result that nearly all the oxygen is abstracted from the air. The waste nitrogen is taken away to the liquefier interchanger from the top of the column and oxygen through the pipe G above the lower of the two communicating oxygen baths, which latter are so arranged that a uniform supply of good quality is assured. An interesting addition to the separator is a small nest of tubes in the centre of the vaporiser through which a quantity of rich nitrogen from the inner compartment E can be withdrawn as indicated.After this additional scrubbing a relatively small, but useful, quantity of gaseous nitrogen of 99-5 to IOO per cent. purity, can be obtained as a by-product. All parts of the apparatus subject to temperatures below that of the atmosphere, are efficiently insulated and when working conditions arc established, air is delivered by the compressor at a pressure of 26 atmos- pheres. 33 atmospheres, already mentioned as the maximum pressure, is the initial pressure employed to hasten cooling down.This is less than a quarter of the initial pressure required by Linde and other plants, which is not only a fine tribute to the efficiency of Claude’s expansion engine, but this relatively low maximum pressure enables lighter material and more efficient design to be employed in the interchangers and elsewhere. A striking feature of the expansion engine is that it works without any cylinder lubrication. Specially treated leather, from which water and fatty matter have been removed, is employed as piston packing and it is fcund that, at the low temperatures involved, leather in this condition remains pliable and the engine give3 no more trouble than an oil lubricated machine. We find that these plants under daily working conditions produce oxygen of 98-5 per cent.purity for an expenditure of about 185 B.H.P. per hour or 46 B.H.P. per 1000 cubic feet of oxygen. The labour required for working the plant is no more than is required for a plant one-tenth the size. We do not employ larger plants than these as we find them to be a unit of convenient size for our oxygen factories. We do not employ Linde plants of larger capacity than 2000 cubic feet of oxygen per hour, but so far as the separator is concerned, these and all larger Linde plants are now similar in most respects to the Claude design. In producing cold with these larger plants, Linde does not employ an expansion engine doing external work but still relies on the Joule-Thomson effect produced by internal expansion of the air.On the other hand he economises power by compressing only one-third of the air used to a constant pressure of about 135 atmospheres and the remaining two-thirds to 4 atmospheres. Thus like Claude he admits liquid and gaseous air to his separator and appears to obtain similar results with very little morePRACTICAL APPLICATIONS OF rAow TEMPERATURES 2 13 expenditure of power. Fig. 5 is a diagranmatic illustration of Linde’s separator. This consists of two rectification columns, one above the other. The lower, like the condenser tubes in Claude’s apparatus, works under a pressure of 4 atmospheres and the upper under atmospheric pressure. The high pressure air at 135 atmospheres enters the bottom vaporiser a t C and is liquefied in exactly the same manner as in the single apparatus illustrated in Fig.3, entering the lower column, after expansion, through the rose- end shown. This supply of liquid air, like that of the Claude liquefier, makes good all cold losses. The low pressure gaseous air enters the bottom vaporiser at B and passing through a separate coil is discharged into the lower column about the centre, as shown, from which point it rises to a nest of tubes D surrounded by a bath of liquid oxygen, as in the Claude apparatus. Here partial liquefaction takes place, and the liquid falls back over the plates with a scrubbing effect on the rising vapours. Most of the oxygen is thus removed and accumu- lates as liquid in the lower vaporiser. The re- maining gas, practically pure nitrogen, passes over to be condensed in the external tubes as indicated.Thus liquids, rich in nitrogen and oxygen respectively, are supplied as in the Claude apparatus to the upper column, in the positions shown, for final rectification. Both nitrogen and oxygen are drawn off as indicated through a counter-current coil interchanger 0 similar in construction to that shown in Fig. 3, in order to abstract heat from the incoming high and low pressure supplies of air, the whole being similarly insulated. I n these plants, as in the smaller, Linde emplojs an auxiliary fore-cooler which is kept cold by an ammonia or carbon dioxide machine, but he also em- ploys the fore-cooler for separating out mois- ture from the air. I am not able to speak from personal ex- perience of a Linde plant of greater capacity than 2 0 0 0 cubic feet of oxygen per hour.This latter plant, hOM ever, absorbs about column type separator. 106 B.H.P. per hour, or j 3 B.H.P. per 1000 cubic feet of oxygen produced. Both Linde and Claude oxygen separators have been manufactured up to a capacity of about goo0 cubic feet of oxygen per hour. ll’ith such plants the power consumed per 1000 cubic feet of oxygen I)roduced, is about 33 B.H.P. I believe the Linde Company have also manufactured larger units, for which they claim still lower consumption of power, but the design and construction of efficient heLt interchangers for large oxygen plants is likely to put a limit to their size and I venture to d ;ubt 11 hether a plant with much lower consumption of power than 30 B.H.P. per I ooo cubic feet of oxygen is possible with due regird to economy in other directions. Whilst on the subject of large plant units, reference may appropriately be made to the well-advertised Jeffries Norton process which we have repeatedly been assured would produce oxygen at 6d. or less 1000 cubic feet. FIG. 5.-Linde9s double2 14 INDUSTRIAL METHODS OF LIQUEFACTION AND The process appears to be based on the fact that heat interchange should take place between high pressure gases, in order to recover power by expansion more effectively than is the case in existing systems, where heat is transmitted from a high to a low pressure gas. In Fig. 6 I give a very free diagrammatic illustration of the process as I understand it. Air at a pressure of say 2 0 atmospheres enters the system at A and on its way to the rectification column B (which is operated at only a slightly lower pressure) it is made to pass through two inter- changers, as diagrammatically illustrated, in order to heat up outgoing nitrogen and oxygen.The former of these first passes through its inter- changer under full pressure. It is then split up as shown, about 60 per cent. being heated by suitable external means at D to a temperature of about 550' C. This hot and fully compressed gas then passes to a motor E, which it drives, generating FIG. 6.-Diagram of the Jeffries Norton process. sufficient power, it is claimeld, to wori; the whole system. The remaining 40 per cent. of nitrogen, in order to help in counteracting heat influx, is ex- panded in the engine C, the expanded gas on its way out of the system being again employed as shown to ab- stract heat from the incoming com- pressed air.The separated oxygen leaving the rectification column under pressure passes as shown, to the ex- pansion engine P, and the expanded oxygen on its way out of the system, is employed to abstract heat as indi- cated from the incoming compressed air. I have already shown that in order to maintain the correct temperature gradient in rectification columns, it is necessary to have oxygen vapours at the bottom and nitrogen vapours at the top, but when the pressure in the column is nearly the same as that of the compressed air entering it, then the condensation of the latter in liquid oxygen cannot be effected and normal rectification cannot be set up.In order to get over this difficulty and transfer heat from the top to t6e bottom of his column, Norton places a coil K in the liquid oxygen bath at the bottom of the column through which an independent supply of oxygen, raised by the compressor F to a pressure above that prevailing in the column, is caused to flow after being first cooled by the outgoing oxygen vapour in G. The circulating oxygen is then condensed in the oxygen bath and afterwards expanded through the valve H to a temperature low enough to condense nitrogen. I t is then drawn through the coil M to be recompressed by the compressor F in the enclosed circuit. The Norton process has had the advantage of Government support in the U.S.A. where a large plant for the separation of helium from natural gases was erected in Petrolia, Texas. The plant was never successful and I understand it has now been abandoned, a plant erectedPRACTICAL APPLICATIONS OF LOW TEMPERATURES 2 15 by the Linde Air Products Company of the U.S.A.during the war at Fort IVorth, Texas, being employed instead. I have not heard of any Norton plant being successfully operated for the production of oxygen and although the process has clever features and is refreshingly original, 1 am afraid most engineers who have had experience with very low temperatures, will regard it as somewhat impracticable. There are many other makers of oxygen plants to-day, mostly of small size. Nearly all are based on the original Linde apparatus (Fig. 3) and I am not aware of any possessing novelty in design.Only one maker that I know of employs the Claude expansion engine. I n that case, some of the air from the main compressor at a very high pressure and without previous cooling below atmospheric temperature, is sent through the ex- pander and then through a heat interchanger to the centre of the column. ‘This has certain advantages in the production of liquid air but I do not think it can be economically adapted to the production of oxygen. I t is apparently claimed as an advantage of the system that ordinary oil can be used for lubricating the cylinder of the expansion engine. I t is, however, a greater advantage of the Claude system that oil need not be used at all. Cylinder lubricating oil is a source of trouble in liquid oxygen plants in two respects.(I) Traces are always liable to be carried forward and tend to foul up both interchanger pipes and the column, thus necessitating periodical and difficult cleanmg. ( 2 ) The other, and more serious trouble, is in connection with the air compressor cylinders. The heat generated in each stage of compression, in spite of very efficient cooling, necessitates the use of special high flash point oils and precautions to ensure that the flash point temperature is not exceeded. I t has been proved, however, that any lubricating oil heated in the presence of compressed air beyond its flash point, breaks down into simpler substances including free carbon and acetylene. These substances, particularly acetylene, are liable to be carried forward in the air and to accumulate as solid particles in liquid oxygen baths, thus producing conditions which have given rise to numerous ex- plosions which have been almost invariably traced to the presence of acetylene.I n our oxygen factories, until the cause became obvious, these incidents were amongst the most disquieting of our experience, but since we have got to the root of the trouble, we have greatly reduced the risk by introducing suitable safeguards on our air compressors. We have also taken precautions not to store carbide or use acetylene within IOO yards of our air intake. Thus we are forced to be very inhospitable in our own factories to a gas which hss otherwise proved a valuable business ally. Hitherto I have dealt only with plants producing a gaseous oxygen product.Liquid oxygen can always be drawn from such plants but they are not the most convenient construction for the purpose. I t is obvious that when some of thc oxygen is drawn away in the liquid state its value in the interchangers for abstracting heat from the incoming air is lost. This has to be compensated for by increased pressure. Hitherto the demand for liquid oxygen has been practically restricted to rescue work and aircraft uses, but its employment for explosives, developed by the Germans during the war, is likely to extend and create a demand for plants primarily designed to produce liquid oxygen. The Sociktk 1’Air Liquide recently supplied a plant for this purpose to certain mines in Lorraine where German plants had hitherto been used. I t works at a pressure of 60 atmospheres and produces 7 j litres of iiquid oxygen per hour for an expenditure of 140 B.H.P.; less than 2 B.H.P.per litre. Il’e are also supplying to South Africa n much smaller plant working at a higher pressure2 I 6 INDUSTRIAL METHODS OF LIQUEFACTION AND which will produce liquid oxygen for an expenditure in power of 2-5 B.H.P. per litre and will alternatively produce gaseous oxygen. These results compare favourably with any others I have seen, but I am hopeful that we shall still further improve upon them in the near future. The price of oxygen in the liquid state must always be more than in the gaseous owing to the greater expenditure of power involved in its production and losses due to evaporation in its use. We sell the liquid at infrequent intervals and small quantities, as drawn from the oxygen plants in our factories, at prices down to IS.4d. per lb. in Dewar vessels of small capacity, but to a regular and large consumer very much better terms could now be offered. In Lorraine, where I have seen liquid oxygen extensively used in the iron mines for blasting purposes I was told that it cost I fr. 5 0 per litre, which, at the rate of exchange prevailing at that time, corresponded to about 7d. per litre in this country. Either of the liquid oxygen plants to which I have just referred would be easily capable of producing the liquid at this figure. In recent years, thanks to improved methods of constructing metallic Dewar flasks, the transport of liquid oxygen has become much simplified.I n flasks of 25 litres capacity and upwards the loss due to ordinary evapora- tion I am told should not exceed 5 per cent. in 2 4 hours, although in the mines a much larger percentage is of course lost by evaporation whilst the cartridges are being impregnated with the liquid immediately before they are used. As regards the cost of gaseous oxygen, this varies with the size of plant employed to a larger extent than in the case of the liquid. As evidence of the extent to which this occurs it is only necessary to compare the cost of producing gaseous oxygen by means of present day liquid air plants of maximum and minimum sizes. For this purpose a plant producing 500 cubic feet of oxygen per hour may be taken as the smallest practicable unit, and for the other extreme I will take a Claude plant producing goo0 cubic feet of oxygen per hour, for that is the largest size of which I can speak with adequate personal knowledge.Assuming power in each case to cost Id. per B.H.P. (a somewhat high figure for the larger plant) then working continuously to their full capacity, with adequate power, labour, material and a fair propcrtion of all the usual standing charges, including interest on capital and depreciation, the cost of oxygen produced by the small plant would be 19s. and by the large one 4s. 6d. per 1000 cubic feet. If these plants were worked intermittently so as to produce annually only half of their full output, the standing charges would tell greatly in favour of the large plant. I estimate that the cost of oxygen woulcl then be with the small plant 27s.and with the large 6s. 3d. per 1000 cubic feet. In the case of oxygen factories supplying the gas in cylinders it is obvious that with requisite supervision, heavier overhead charges, relatively smail plants in operation and a fluctuating demand the cost of oxygen must always be very appreci- ably higher than in works where the gds can be used direct from the holder. Throughout their factoiies in this country the British Oxygen Company have now plant capable of producing an aggregate of z million cubic feet of oxygen per day. Our present sales fluctuate between 4 and 5 million cubic feet per week, which is less than half the total capacity of our plants. Our present average price for oxygen in cylinders is about 38s. per 1000 cubic feet, although to many large consumers the price is substantially lower.These costs are for gas into the gasholder.This price I believe compares favourably with that prevailing in other countries with the probable exception of Germany, where in any case the collapse of the mark renders comparison difficult. The secret of cheap oxygen is large liquid air units worked to their full capacity. By having increased the size of our factory units since the war we are able now to sell oxygen below our pre-war price. The oxygen business has become a good barometer of trade. I t has suffered severely during the prolonged industrial depression, but the demand has been perceptibly increasing within the last two months. If this continues, as everybody must hope will be the case-for it will indicate improved general conditions in trade-then it will be obvious, from the figures I have given, that with an increased output still further reductions in the price of oxygen in cylinders should become possible.The amount of oxygen sold in this country for medical purposes of every kind probably does not exceed I per cent. of the total output in cylinders. About 3 per cent. is used for limelight and experimental pur- poses. Of the industrial uses for compressed osygen the most important is now metal cutting. This must represent over 5 0 per cent. of the total demand, and is about equally divided between scrap-cutting, or destructive work, and much important constructive work. The latter is perhaps one of the most promising outlets for oxygen.When the oxygen cutter is employed in conjunction with self-feeding profiling machines, now constantly being designed for various classes of work, it is found to operate almost with the precision of a saw and with far greater rapi- dit}.. Oxy-acetylene welding ranks next to metal cutting, amongst the iridustrial applications of oxygen. About 25 per cent. of the gas sold in cylinders iii employed for that purpose and it is becoming obvious that oq.-acetylene welding can easily hold its own in a great variety of work against electric welding and other processes. The remainder of the oxygen sold in cylinders is mostly employed for lead burning purposes and for obtaining high temperatures in special metallurgical operations. I n the wider fields of industry where the use of oxygen has often been suggested, as for instance in the enrichment of air for blast furnaces, and in conjunction with steam, for the continuous gasification of fuel, it is doubtful whether the stage has yet been reached when oxygen can be pro- duced cheaply enough even in large plant units to be economically em- ployed. I n Germany oxygen is used in the manufacture of nitric acid and acetic acid, but otherwise neither in that country nor any other, so far as I am aware, has the use of oxygen yet developed to any appreciable extent in the chemical industry. I t must be admitted that industrial development throughout the world is still mainly confined to the cylinder trade. I n this country, in spite of trade depression, the quantity of oxygen distributed in cylinders at the present time must approach 300 million cubic feet per annum (of which upwards of 2 2 5 million are supplied by the British Oxygen Company). In France I believe the annual output is about the same as here, but it is stated to be three times as much in the United States of America and more than twice as much in Germany. At the outset of this paper I expressed the inte:ition of limiting myself to an interpretation of its title which would permit me to deal exclusively with the separation of the constituents of the air. I wish, however, to ex- plain that I had not advanced far with my subject before I realised that I should have to limit myself still further and abandon the consideration of all constituents except oxygen if I attempted any historical review of the All the rest goes to industry.2 18 INDUSTRIAL METHODS OF LOW TEMPERATURES industrial production and use of that gas. I have therefore confined myself to the subject on which I am best informed and have devoted this paper ex- clusively to oxygen. I offer no apology for this because I believe that the industrial importance of oxygen to-day warrants independent consideration. During the war the supply of oxygen was described by a prominent member of the present Government as a matter of paramount national importance. This was certainly no exaggeration at that time and since the war the economic value of oxygen in the industrial applications I have mentioned, has been recognised to the full. Unfortunately exaggerated claims and statements are very apt to breed misconceptions in these difficult days of abnormal exchanges and keen international competition. I am therefore grateful to this important society for affording me an opportunity of placing on record what I believe to be a reliable statement of the present position in the oxygen industry.

ISSN:0014-7672    

DOI:10.1039/TF9221800205

出版商:RSC    

年代:1922

数据来源: RSC

摘要:

118 ELECTRICAL THEORY OF ADBORPTTON The writer considers the double layer as consisting of a swface of rigidly fixed atoms under continuous bombardment of positively and negatively charged ions, any particular point on the rigid surface becoming in turn negative, neutral and positive, these conditions arisdg in any order. The observed contact difference is the average effect of these conditions. Where several kinds of atoms are present in the solution the average number of any one of them at the surface will depend on their concentbration, valency and mobility. The variation of contact Werence from negative to neutral and positive was observed with cotton and aluminium sulphate near the neutral point. These variations occurred during the same experiment, the readings being direct measurements of E.1I.F.s developed by filtration under pressure.This point would be covered by putting n2 = 1 and = 2 or 3 in Mukherjee’s equation No. 13.118 ELECTRICAL THEORY OF ADBORPTTON The writer considers the double layer as consisting of a swface of rigidly fixed atoms under continuous bombardment of positively and negatively charged ions, any particular point on the rigid surface becoming in turn negative, neutral and positive, these conditions arisdg in any order. The observed contact difference is the average effect of these conditions. Where several kinds of atoms are present in the solution the average number of any one of them at the surface will depend on their concentbration, valency and mobility. The variation of contact Werence from negative to neutral and positive was observed with cotton and aluminium sulphate near the neutral point.These variations occurred during the same experiment, the readings being direct measurements of E.1I.F.s developed by filtration under pressure. This point would be covered by putting n2 = 1 and = 2 or 3 in Mukherjee’s equation No. 13.118 ELECTRICAL THEORY OF ADBORPTTON The writer considers the double layer as consisting of a swface of rigidly fixed atoms under continuous bombardment of positively and negatively charged ions, any particular point on the rigid surface becoming in turn negative, neutral and positive, these conditions arisdg in any order. The observed contact difference is the average effect of these conditions. Where several kinds of atoms are present in the solution the average number of any one of them at the surface will depend on their concentbration, valency and mobility.The variation of contact Werence from negative to neutral and positive was observed with cotton and aluminium sulphate near the neutral point. These variations occurred during the same experiment, the readings being direct measurements of E.1I.F.s developed by filtration under pressure. This point would be covered by putting n2 = 1 and = 2 or 3 in Mukherjee’s equation No. 13.118 ELECTRICAL THEORY OF ADBORPTTON The writer considers the double layer as consisting of a swface of rigidly fixed atoms under continuous bombardment of positively and negatively charged ions, any particular point on the rigid surface becoming in turn negative, neutral and positive, these conditions arisdg in any order.The observed contact difference is the average effect of these conditions. Where several kinds of atoms are present in the solution the average number of any one of them at the surface will depend on their concentbration, valency and mobility. The variation of contact Werence from negative to neutral and positive was observed with cotton and aluminium sulphate near the neutral point. These variations occurred during the same experiment, the readings being direct measurements of E.1I.F.s developed by filtration under pressure. This point would be covered by putting n2 = 1 and = 2 or 3 in Mukherjee’s equation No. 13.118 ELECTRICAL THEORY OF ADBORPTTON The writer considers the double layer as consisting of a swface of rigidly fixed atoms under continuous bombardment of positively and negatively charged ions, any particular point on the rigid surface becoming in turn negative, neutral and positive, these conditions arisdg in any order.The observed contact difference is the average effect of these conditions. Where several kinds of atoms are present in the solution the average number of any one of them at the surface will depend on their concentbration, valency and mobility. The variation of contact Werence from negative to neutral and positive was observed with cotton and aluminium sulphate near the neutral point. These variations occurred during the same experiment, the readings being direct measurements of E.1I.F.s developed by filtration under pressure.This point would be covered by putting n2 = 1 and = 2 or 3 in Mukherjee’s equation No. 13. THE MANUFACTURE OF HYDROGEN BY THE PARTIAT, LIQUEFACTION OF wmm GAS AND COKE-OTEN GAS. BY GEORGES CLAUDE, OF PARIS. (Translated by H. RORNS.) For B number of years I have been engaged in the manukicture of hydrogen by the partial and direct liquefaction of water gas and its analogues. For this purpose water gas, compressed to a suitable degree, is in the first instance deprived of its carbon dioxide and is dried. l’he gas is then sent into a heat exchanger E, Fig. I, in which it is cooled by circulating in the opposite direction to the hydrogen and carbon monoxide which have already been separated. ‘ h e gas then enters by the collector C the bottom of a sheaf of vertical tubes F, the lower portion of which plunges into a bath of carbon monoxide boiling under atmospheric pressure.By the combined influences of the pressure and of the temperature of the liquid bath a large portion of the carbon monoxide of the ascending gases is liquefied; it flows back into the collector C. Owing to the pressure this liquid is forced out through the tube ‘l? and the cock R into the vnporiser V, where it replaces the liquid which is being evaporated. The remaining gas, containing hydrogen and a little of the residual CO, continues to rise in the sheaf of tubes F. There the gases encounter a temperature which is being more and more lowered by means which I shall presently explain. Under that influence the rest of the carbon monoxide is liquefied.T/zeoreticni&, therefore, it is a sensibly pure hydrogen which escapes from the top of the tubes, and it is this cold and campressed hydrogen which is expanded in the nozzle I> and further cooled by this expansion to be sent back around the sheaf F, in order there to produce the very low temperatures of which I have just spoken. I n working this process two sources of inconvenience were experienced which I have only recently been able to overcome. In the first instance, when the hydrogen is entering the expansion engine at extremely low teinperature as indicated, the frigorific efficiency of the expansion will be very low. There is in that case rz progressive loss of liquid, unless we go up to pressures unduly high from the point of view of energy expenditure, and this all the more so because none of the customary lubricants can be employed at temperatures below - 200’ C.Hence there is abnormal friction in the engine which will again lower the frigorific efficiency. I n the second instance, the calorific mass of the gases, leaving the tubes and after- wards the nozzle D and circulating about the rising gases, will be smaller than that of the latter by the whole mass of the cubon monoxide which had been liquified by their action. *4s a consequence the ascending gas cannot, even 2192 2 0 THE MANUFACTURE OF HYDROGEN BY THE PA~lKl'IL41J assuming a perfect heat exchange to take place, leave the sheaf of tubes at the temperature at Ivliich the expanded gases entered the tubes. The final temperature of the expansion of the hydrogen will, therefore, not be very low, and in addition this temperature will be badly utilised for the purification of the ascending gases.il -eO- c1 FIG. I . 'These various troubles have been overcome in the following manner :- I. Part of the compressed gas which is enterina the system is with- .? drawn from the heat exchanger E a t a point M, Big. 2, before it arrives in the cold portion; it is thus relatively IittIe cooled. I t is sent through the tube t into an auxiliary system of tubes G around which the hydrogen, leaving the bundle and passing on to the nozzle, is being circulated byi~ieniis of the pipes L and K. 'This relatively warm gas will then reheat the hydrogen before its expansion and will hence increase the efficiency of the expansion process.2. If we relied exclusively upon thii mettis, the final temperature of the expansion would remain unsatisfactory offing to the want of lubrication and p"" u M-?C F I G . 2 . to the eventual freezing out of the residual CO ( - 206' C.). \Ve must manage to lubricate the machinery by a substance which will remain liquid when the temperature sinks below - 206'. That body cannot be petrol ether, nor any similar substance since they would all be solidified at this tempera- ture. I have, therefore, made use of nitrogen which only solidifies at - 2 I I'2 2 2 THE MANUFACTURE OF HYDROGEN EU THE PARl'IXL and which can moreover be added to the hydrogen, before its expansion, without causing any complication, at least in cases when we are aiming at th2 synthesis of ammonia.current of compressed and dried nitrogen is cooled down to about its condensation point in the heat-exchanger E by means of the tube system XU ; it is then added to the hydrogen at Z, before the hydrogen enters into the nozzle. I n this way then we secure accept- able conditions for the lubrication. As a first consequence the temperature of the escaping gas is easily lowered down to - 208' and - 210' C., as soon as we lubricate by means of nitrogen. As a second consequence this addition of nitrogen to the gases leaving the tubular system iiicreases, to any extent that we desire, the calorific niass of the expanded gas. l'he latter gas may therefore equal or surpass in mass that of the gas ascending in the sheaf F, including the carbon monoxide which has been condensed in this part of the tubes.As a result the temperature of expansion is con- siderably improved, and it is moreover much better utilised. We can indeed easily eliminate by these means the carbon monoxide within I per cent., and the hydrogen finally isolated can without any difficulty be used i n my hypercomfrzssion process for the synthesis of ammonia. This process has been developed and put into practice in the works of the Grande Paroisse, near Montereau, where an apparatus for the pro- duction of joo nz3 of hydrogen per hour is in operation, feeding a unit for j tons of ammonia per day. The carbon monoxide, containing all the nitrogen of the water gas it should be noted, is discharged at a percentage which may easily be raised to 85.The amount of pure hydrogen obtained by this method is therefore very satisfactory. The carbon monoxide may be utilised, either for the manufacture of various chemical products, or for driving the internal combustion motors of the installation. The latter utilisation has been adopted in Montereau. The process which I have outlined requires the compression of the gases to degrees, varying with the size of the apparatus from 15 to 30 atmospheres. The first figure applies to apparatus of a capacity of 2000 ?a3 of hydrogen per hour. This process necessitates the command of a relatively high motive power, and one may draw attention to the advantages offered by processes based upon the catalysis of CO into C02, such as are employed in the Haber process.I have therefore only studied and tried this process with a view of its ultimate application to a particularly interesting case, that of the coke-oven fuymzces (and also that of town gas) in which the presence of a considerable portion of methane renders catalytic processes inoperative. The complexity of the gas mixtures we have to deal with in this case and the diversity of the freezing points of the constituents may make us fear that we should have to meet serious difficulties in the Lvorking of this process. As a matter of fact, however, an apparatus has already been constructed for this purpose and it has, after very short trials, been put successfully to service in the Bkthune mines. T h e essential cause of the success is the very high reciprocal solubility of the diverse condensable constituents.This first apparatus has a productive capacity of 350 m4 of hydrogen per hour, and it is operating with a com- pression of about 2 7 atmospheres. An installation for the utilisation of the hydrogen produced by means of this apparatus in the manufacture of ammonia by the application of hypercompression has already been erected and is actually being put in working order. I hope that this installation will be the point of departure for a much more important installation in which I intend to make use of apparatus for the production of 2000 m3 of hydrogen per hour.LIQUEE’ACTION OF U’ATEII GA4S AND COKE-OVEN GL4S 2 2 3 One of the essential reasons that make me count upon a development of this process in the coke-oven industry lies in the many indirect ad- vantages which.the process promises. On the one hand, since the gases to be treated must already be com- pressed for the extraction of the hydrogen, we are naturally led to effect the stripping of the gas of its benzol likewise under pressure. All the benzol which at present escapes when the process is carried out at atmospheric pressures, will then easily be retained; that will constitute a gain which in certain cases may come up to I kilogram of benzol per ton of coke. This improved recovery will moreover be attained by means of a greatly reduced amount of solvent and of heat energy, and at much diminished losses of solvent. The recovery will be effected in absorption and distillation ap- paratus of comparatively very small dirnensions.On the other hand the ethylene, this precious gas, the percentage of which in furnace gases is too low for its extraction with much success at atmospheric pressure, may easily be collected to a large extent in the course of the operations. For it will be con- densed almost alone in the preliminary cold-exchanger of the apparatus, and it can be extracted as a 40 or 50 per cent. mixture which will very readily be utilisable, either for the manufacture of alcohol or for use in autogeneous welding or similar uses. T o give an idea of the possible importance of this by-product I may state that, supposing the furnace-gas treated to contain I -5 per cent. of ethylene and that it can all be extracted, this would correspond to our obtaining zoo litres of alcohol per ton of ammonia in addition to other products. Finally I should remark that from the calorific standpoint hydrogen must be regarded as the very worst of all combustible gases. One cubic metre of hydrogen represents only about 2 600 calories (lower calorific power) against the 3000 calories of carbon monoxide and the 10,000 of methane. When therefore we extract the hydrogen from coke-oven gases we, in fact, enrich the gases in a true sense, and it is a considerable en- richment which they undergo, since they become fit for uses for which they would otherwise be unsuitable. When the gas is to be distributed it will be completely free of every trace of the condensable impurities that cause so much mischief in our actual gas distribution systems; when it is to be utilised as industrial gas, it will enable us to obtain extremely high temperatures and to combine, for instance, under specially interesting conditions, the production of nitric oxide by the Hausser process with the synthesis of ammonia. These are the principal advantages of the process which I. have the honour to describe before you. I should like to emphasise that one of the characteristic essential features of the process is the extreme smallness of the necessary apparatus. An apparatus for 1000 m3 of hydrogen per hour requires a sheaf of liquefaction tubes, 40 cm. in diameter and 3 m. in height.

ISSN:0014-7672    

DOI:10.1039/TF9221800219

出版商:RSC    

年代:1922

数据来源: RSC

摘要:

118 ELECTRICAL THEORY OF ADBORPTTON The writer considers the double layer as consisting of a swface of rigidly fixed atoms under continuous bombardment of positively and negatively charged ions, any particular point on the rigid surface becoming in turn negative, neutral and positive, these conditions arisdg in any order. The observed contact difference is the average effect of these conditions. Where several kinds of atoms are present in the solution the average number of any one of them at the surface will depend on their concentbration, valency and mobility. The variation of contact Werence from negative to neutral and positive was observed with cotton and aluminium sulphate near the neutral point. These variations occurred during the same experiment, the readings being direct measurements of E.1I.F.s developed by filtration under pressure.This point would be covered by putting n2 = 1 and = 2 or 3 in Mukherjee’s equation No. 13.118 ELECTRICAL THEORY OF ADBORPTTON The writer considers the double layer as consisting of a swface of rigidly fixed atoms under continuous bombardment of positively and negatively charged ions, any particular point on the rigid surface becoming in turn negative, neutral and positive, these conditions arisdg in any order. The observed contact difference is the average effect of these conditions. Where several kinds of atoms are present in the solution the average number of any one of them at the surface will depend on their concentbration, valency and mobility. The variation of contact Werence from negative to neutral and positive was observed with cotton and aluminium sulphate near the neutral point.These variations occurred during the same experiment, the readings being direct measurements of E.1I.F.s developed by filtration under pressure. This point would be covered by putting n2 = 1 and = 2 or 3 in Mukherjee’s equation No. 13.118 ELECTRICAL THEORY OF ADBORPTTON The writer considers the double layer as consisting of a swface of rigidly fixed atoms under continuous bombardment of positively and negatively charged ions, any particular point on the rigid surface becoming in turn negative, neutral and positive, these conditions arisdg in any order. The observed contact difference is the average effect of these conditions. Where several kinds of atoms are present in the solution the average number of any one of them at the surface will depend on their concentbration, valency and mobility.The variation of contact Werence from negative to neutral and positive was observed with cotton and aluminium sulphate near the neutral point. These variations occurred during the same experiment, the readings being direct measurements of E.1I.F.s developed by filtration under pressure. This point would be covered by putting n2 = 1 and = 2 or 3 in Mukherjee’s equation No. 13.118 ELECTRICAL THEORY OF ADBORPTTON The writer considers the double layer as consisting of a swface of rigidly fixed atoms under continuous bombardment of positively and negatively charged ions, any particular point on the rigid surface becoming in turn negative, neutral and positive, these conditions arisdg in any order.The observed contact difference is the average effect of these conditions. Where several kinds of atoms are present in the solution the average number of any one of them at the surface will depend on their concentbration, valency and mobility. The variation of contact Werence from negative to neutral and positive was observed with cotton and aluminium sulphate near the neutral point. These variations occurred during the same experiment, the readings being direct measurements of E.1I.F.s developed by filtration under pressure. This point would be covered by putting n2 = 1 and = 2 or 3 in Mukherjee’s equation No. 13.118 ELECTRICAL THEORY OF ADBORPTTON The writer considers the double layer as consisting of a swface of rigidly fixed atoms under continuous bombardment of positively and negatively charged ions, any particular point on the rigid surface becoming in turn negative, neutral and positive, these conditions arisdg in any order.The observed contact difference is the average effect of these conditions. Where several kinds of atoms are present in the solution the average number of any one of them at the surface will depend on their concentbration, valency and mobility. The variation of contact Werence from negative to neutral and positive was observed with cotton and aluminium sulphate near the neutral point. These variations occurred during the same experiment, the readings being direct measurements of E.1I.F.s developed by filtration under pressure.This point would be covered by putting n2 = 1 and = 2 or 3 in Mukherjee’s equation No. 13.118 ELECTRICAL THEORY OF ADBORPTTON The writer considers the double layer as consisting of a swface of rigidly fixed atoms under continuous bombardment of positively and negatively charged ions, any particular point on the rigid surface becoming in turn negative, neutral and positive, these conditions arisdg in any order. The observed contact difference is the average effect of these conditions. Where several kinds of atoms are present in the solution the average number of any one of them at the surface will depend on their concentbration, valency and mobility. The variation of contact Werence from negative to neutral and positive was observed with cotton and aluminium sulphate near the neutral point.These variations occurred during the same experiment, the readings being direct measurements of E.1I.F.s developed by filtration under pressure. This point would be covered by putting n2 = 1 and = 2 or 3 in Mukherjee’s equation No. 13.118 ELECTRICAL THEORY OF ADBORPTTON The writer considers the double layer as consisting of a swface of rigidly fixed atoms under continuous bombardment of positively and negatively charged ions, any particular point on the rigid surface becoming in turn negative, neutral and positive, these conditions arisdg in any order. The observed contact difference is the average effect of these conditions. Where several kinds of atoms are present in the solution the average number of any one of them at the surface will depend on their concentbration, valency and mobility.The variation of contact Werence from negative to neutral and positive was observed with cotton and aluminium sulphate near the neutral point. These variations occurred during the same experiment, the readings being direct measurements of E.1I.F.s developed by filtration under pressure. This point would be covered by putting n2 = 1 and = 2 or 3 in Mukherjee’s equation No. 13.118 ELECTRICAL THEORY OF ADBORPTTON The writer considers the double layer as consisting of a swface of rigidly fixed atoms under continuous bombardment of positively and negatively charged ions, any particular point on the rigid surface becoming in turn negative, neutral and positive, these conditions arisdg in any order.The observed contact difference is the average effect of these conditions. Where several kinds of atoms are present in the solution the average number of any one of them at the surface will depend on their concentbration, valency and mobility. The variation of contact Werence from negative to neutral and positive was observed with cotton and aluminium sulphate near the neutral point. These variations occurred during the same experiment, the readings being direct measurements of E.1I.F.s developed by filtration under pressure. This point would be covered by putting n2 = 1 and = 2 or 3 in Mukherjee’s equation No. 13.118 ELECTRICAL THEORY OF ADBORPTTON The writer considers the double layer as consisting of a swface of rigidly fixed atoms under continuous bombardment of positively and negatively charged ions, any particular point on the rigid surface becoming in turn negative, neutral and positive, these conditions arisdg in any order.The observed contact difference is the average effect of these conditions. Where several kinds of atoms are present in the solution the average number of any one of them at the surface will depend on their concentbration, valency and mobility. The variation of contact Werence from negative to neutral and positive was observed with cotton and aluminium sulphate near the neutral point. These variations occurred during the same experiment, the readings being direct measurements of E.1I.F.s developed by filtration under pressure. This point would be covered by putting n2 = 1 and = 2 or 3 in Mukherjee’s equation No.13.118 ELECTRICAL THEORY OF ADBORPTTON The writer considers the double layer as consisting of a swface of rigidly fixed atoms under continuous bombardment of positively and negatively charged ions, any particular point on the rigid surface becoming in turn negative, neutral and positive, these conditions arisdg in any order. The observed contact difference is the average effect of these conditions. Where several kinds of atoms are present in the solution the average number of any one of them at the surface will depend on their concentbration, valency and mobility. The variation of contact Werence from negative to neutral and positive was observed with cotton and aluminium sulphate near the neutral point. These variations occurred during the same experiment, the readings being direct measurements of E.1I.F.s developed by filtration under pressure.This point would be covered by putting n2 = 1 and = 2 or 3 in Mukherjee’s equation No. 13.118 ELECTRICAL THEORY OF ADBORPTTON The writer considers the double layer as consisting of a swface of rigidly fixed atoms under continuous bombardment of positively and negatively charged ions, any particular point on the rigid surface becoming in turn negative, neutral and positive, these conditions arisdg in any order. The observed contact difference is the average effect of these conditions. Where several kinds of atoms are present in the solution the average number of any one of them at the surface will depend on their concentbration, valency and mobility.The variation of contact Werence from negative to neutral and positive was observed with cotton and aluminium sulphate near the neutral point. These variations occurred during the same experiment, the readings being direct measurements of E.1I.F.s developed by filtration under pressure. This point would be covered by putting n2 = 1 and = 2 or 3 in Mukherjee’s equation No. 13.118 ELECTRICAL THEORY OF ADBORPTTON The writer considers the double layer as consisting of a swface of rigidly fixed atoms under continuous bombardment of positively and negatively charged ions, any particular point on the rigid surface becoming in turn negative, neutral and positive, these conditions arisdg in any order. The observed contact difference is the average effect of these conditions.Where several kinds of atoms are present in the solution the average number of any one of them at the surface will depend on their concentbration, valency and mobility. The variation of contact Werence from negative to neutral and positive was observed with cotton and aluminium sulphate near the neutral point. These variations occurred during the same experiment, the readings being direct measurements of E.1I.F.s developed by filtration under pressure. This point would be covered by putting n2 = 1 and = 2 or 3 in Mukherjee’s equation No. 13.118 ELECTRICAL THEORY OF ADBORPTTON The writer considers the double layer as consisting of a swface of rigidly fixed atoms under continuous bombardment of positively and negatively charged ions, any particular point on the rigid surface becoming in turn negative, neutral and positive, these conditions arisdg in any order.The observed contact difference is the average effect of these conditions. Where several kinds of atoms are present in the solution the average number of any one of them at the surface will depend on their concentbration, valency and mobility. The variation of contact Werence from negative to neutral and positive was observed with cotton and aluminium sulphate near the neutral point. These variations occurred during the same experiment, the readings being direct measurements of E.1I.F.s developed by filtration under pressure. This point would be covered by putting n2 = 1 and = 2 or 3 in Mukherjee’s equation No. 13.118 ELECTRICAL THEORY OF ADBORPTTON The writer considers the double layer as consisting of a swface of rigidly fixed atoms under continuous bombardment of positively and negatively charged ions, any particular point on the rigid surface becoming in turn negative, neutral and positive, these conditions arisdg in any order.The observed contact difference is the average effect of these conditions. Where several kinds of atoms are present in the solution the average number of any one of them at the surface will depend on their concentbration, valency and mobility. The variation of contact Werence from negative to neutral and positive was observed with cotton and aluminium sulphate near the neutral point. These variations occurred during the same experiment, the readings being direct measurements of E.1I.F.s developed by filtration under pressure.This point would be covered by putting n2 = 1 and = 2 or 3 in Mukherjee’s equation No. 13.118 ELECTRICAL THEORY OF ADBORPTTON The writer considers the double layer as consisting of a swface of rigidly fixed atoms under continuous bombardment of positively and negatively charged ions, any particular point on the rigid surface becoming in turn negative, neutral and positive, these conditions arisdg in any order. The observed contact difference is the average effect of these conditions. Where several kinds of atoms are present in the solution the average number of any one of them at the surface will depend on their concentbration, valency and mobility. The variation of contact Werence from negative to neutral and positive was observed with cotton and aluminium sulphate near the neutral point. These variations occurred during the same experiment, the readings being direct measurements of E.1I.F.s developed by filtration under pressure.This point would be covered by putting n2 = 1 and = 2 or 3 in Mukherjee’s equation No. 13.118 ELECTRICAL THEORY OF ADBORPTTON The writer considers the double layer as consisting of a swface of rigidly fixed atoms under continuous bombardment of positively and negatively charged ions, any particular point on the rigid surface becoming in turn negative, neutral and positive, these conditions arisdg in any order. The observed contact difference is the average effect of these conditions. Where several kinds of atoms are present in the solution the average number of any one of them at the surface will depend on their concentbration, valency and mobility.The variation of contact Werence from negative to neutral and positive was observed with cotton and aluminium sulphate near the neutral point. These variations occurred during the same experiment, the readings being direct measurements of E.1I.F.s developed by filtration under pressure. This point would be covered by putting n2 = 1 and = 2 or 3 in Mukherjee’s equation No. 13. THE PRODUCTION OF LIQUID OXYGEN FOR USE ON ,AIR- CRAFT. PART I. STORAGE AND TRAKSPORT OF LIQUID OXYGEN. Introduction. Containers insulated with materials of low thermal conductivity. Vacuum insulated containers. Glass and porcelain containers.Metal containers. Absorbent material for removing residual gases. (a) Charcoal. ( b ) Silica gel. Precautions to be observed in exhausting the interspace. Rate of loss from commercial vessels. Separation of the thermal leakage into its components. (a) Conduction down the neck. ( b ) conduction through residual gas. (c) Radiation transfer between the walls. ((I) Use of a radiation shield. (a) Heylandt’s Vaporiser. (6) Griffiths’ Vaporiser. Methods of generating gas from the liquid as desired. PART 11. PRODUCTION OF LIQUID OXYGHN. Description of laboratory plant. Liquid oxygen plant with expansion engine. Portable Claude type oxygen plant. PART I. STORAGE , 4 ~ n TRANSPORT OF LIQUID OXYGEN. fntroduction. The atmosphere is so attenuated at the altitudes at which flying must be carried out for certain military purposes that the human lungs are unable to obtain the requisite amount of oxygen for normal respira- tion.The quantity of oxygen absorbed per minute is about 3 to 4 litres out of a total volume of 60 litres of air measured at normal temperature and pressure. The weight of air passed through the lungs at zo,ooo feet is reduced to half and therefore needs to be considerably enriched in oxygen. Oxygen apparatus must supply about 4 normal litres of gas at these heights for comfort and safety. The oxygen in the early days was carried in cylinders under pressure, but the weight 01 such cylinders1 is considerable, and a reducing valve is also required. If liquid oxygen is substituted for high pressure gas the liquified 1 See Report of Gas Cylinders Committee, 1921.224PRODUC'lXON OF LIQ.UID OX1'I'GEN FOR AIRCRAFT 2 2 j gas is equivalent to compressed gas at IZ,OOO lbs. per sq. inch as against the 1800 lbs. which is the highest pressure that can be conveniently used in cylinders. The fundamental difficulty attendant upon the use of any liquified gas whose critical temperature is much below normal atmospheric temper- ature, is the maintenance of the containing vessel cold enough to store the liquid during transport and use. Thermal leakage must usually be paid for by evaporation of the liquid, and the evolved gas cannot, in general, be utilised. The employment of the liquid gases may therefore become very inefficient, unless due care is taken in selecting a proper container.Indealing with liquid oxygen, for example, a temperature of - 185" C. must be maintained and as the latent heat of oxygen is only about j o gram-calories a litre will be evaporated for every 7 2 calories of heat that leaks in. This quantity of heat is roughly equivalent to the con- duction into the vessel by a I cm. length of copper wire 0.004 sq. cms. in area in one minute for a temperature difference of about 200' C. between its ends. The heat required to raise the gas from its boiling point to the temperature at the exit from the neck is not usually obtained from the spherical portion of the vessel, but this heat absorption influences the conduction along the neck only. For the high order of insulation necessary in the storage of liquid air and oxygen, two classes of containers have been used; firstly, vessels insulated with layers of material of low conductivity, secondly, double- walled containers evacuated to provide a vacuum space around the inner vessel.(a> Contuiners 1fzsulatt.d with il//utericiis of Low Thermal Coizdzcrtiuity. -For ordinary size vessels the use of the first method is practically out of the question since the materials available have for the thermal conductivity at low temperature a value not appreciably different from that at room temperature, and the heat conduction through a moderate thickness is usually very considerable. For instance, an experimental vessel was made holding 60 lbs. of liquid oxygen with a thickness of 3 inches of magnesia ; this vessel evaporated the contents in one daj.. The robust nature of this type cf vessel would make it admirable for transport, but it is also open to several objections, namely :- I.I t is very difficult to maintain the insulation dry, in viciv of the unavoidable condensation of dew over the exterior surface, ~ i n d a state of dryness is vital to the efficiency of the insulation. 2. The vessels are bulky and heavy. 3. They evaporate a very considerable amount of liquid i n the cooling down of tne insulation. I n the case of vessels in constant use for storage purposes, the last objection would hardly apply. Insulated vessels, resembling " 'i'hermos " flasks, are made commercially for household use. They have welded steel inner and outer shells with the interspace packed with a form of natural graphite, and partially evacuated.Such vessels art: of but little use as liquid air containers, since the rate of loss is of the order of 340 grams per hour for a 44 Iitre capacity vessel, and slightly less for a 2 litre capacity vessel. This rate implies total evaporation of the liquid in about 16 hours. I t may.be reinarkecl that the high rate is partly due to the short neck (h) F ~ ~ S Z C U N Z ITJzsulmd C,~o~zt~~ii~e~s.--'i'he alternative type of vessel to and its large diameter-.226 'THE PRODUCTION OF LIQUID OXYGEN the lagged container, and the only practicable one at present, is the vacuum-walled container. Such vessels are made both of metal and of non-metals such as glass, quartz, and porcelain. (6) GZuss and PorceZmh Containers.-The advantage possessed by glass and similar materials for the construction is that the vessels can be made seamless and heated to a high temperature to free the walls of -~ FIG.I.-Metal vacuum walled vessel with charcoal pad. gas during evacuation. Hence no absorbent material is necessary in the interspace to remove residual gas. The efficiency of the glass type of vessel varies very considerably with different vessels, due presumably to variations in the degree of evacuation. In our experience, a half-litre vessel would be classed as good if its rate was only of the order of IIO grams of liquid oxygen evaporated per day: the corresponding figure for a litre vessel being 2 3 0 grams.10K USE ON AIKCRrW'l' 2 2 7 Of course a vessel will often show a rate many times this value and then re-evacuation produces a considerable improvement.Glass vessels with a capacity of 5 litres (one gallon) have been made, but their fragility renders them of little use outside the Iaboratoiy. In any case it is necessary to syphon the liquid out of the large sized vessels as the stress on the necks becomes excessive when an attempt is made to pour out the liquid. Large vessels are very difficult to construct. Efforts have been directed towards the development of porcelair? vacuum vessels of large size, since experiment indicated that this material was suitable for the purpose. However, the quantity production of such vessels has not been proceeded with, as the metal flasks show many advantages. In Germany porcelain vessels mere rnanufactured of large size and used for the transport of liquid oxygen, but the manufactured now seems to have been abandoned.(d) MetaZ Cuntnizers.-Metal vacuum vessels are now constructed commercially in various sizes, thc usual form is shown in Fig. I. The inner and outer shells are made up of hemispherical spinnings soldered tonether along circumferential seams. These seams are marked v v Fig. I , and are made by the use of a blow-pipe, each joint being s in pre- Ourer vessel d vessel Outer vessel vessel llihia Washer wih hole in centre Asbestos FLig Inner FIG. TWO methods of supporting inner vessel concentric with outer vessel. viously tinned. Care is necessary in the use of flux as spirits are apt to penetrate into the inner space and dull the polish, even rosin ;leaves a decided atmosphere in the flask.The bottom of the inner neck is flared out to carry the weight and also well soldered as it is here that fracture generally occurs. The top of the neck is finished with a brass collar which is grooved to receive the solder which is well melted in. The inner neck is the only metallic contact between the two vessels and its conductivity is therefore important ; for this reason the material should be as thin as possible and about I cm. in diameter. In practice tubing less than about 20 S.1V.G. is difficult to obtain, the metal is inclined to develop flaws in manufacture. I n large vessels some additional support is desirable for the inner vessel especially to prevent the inner vessel swinging about and breaking the neck. This may be prevented by a number of small asbestos blocks around the circumference, or any other insulator may be used which is capable of withstanding heating to about I O O O C.during evacuation without decomposition. An alternative method is also shoivn in Fig. 2 , where the inner vessel is provided with a projecting pin which fits into a hole in a mica washer held in a tube fixed to the outer vessel. 'l'his method has been found very satisfactory in practice.228 THE PRODUCTION OF LIQUID OXYGEN Absorbent Materiak f o r Removing ResiduaZ Gases. All commercial vacuum vessels now in use have a quantity of ab- sorbent material in the vacuum space which clears up the residual gas when cooled down by the filling of the vessel. Metal vessels can be made without an absorbent,l provided they are heated to at least 300° C.to free the walls from gas. Unfortunately the seams will not stand this heating unless hard soldered or welded, and the high temperature required for either of these processes dulls the polish on the walls and thus causes excessive radiation of heat. A very much lower degree of evacuation is required for a vessel containing an absorbent and in fact a Fleuss pump in good condition and dry will serve for evacuation. (a) CharcoaZ.-The best absorbent, according to Dewar,2 is charcoal made from shells or fruit stones, but almost any variety can be used; even coal absorbs a considerable amount of gas. The quantity of good charcoal required is very small and 4 grams for a volume of IOO C.C. interspace is ample. The inflammability of charcoal is a matter for serious consideration in warfare and its use should be avoided.Non-inflammable substances such as silica gel serve the same purpose. (6) SiZica Gel.-This substance is well known as an absorbent and Professor Henry Briggs, on behalf of the Oxygen Research Committee (Department of Scientific and Industrial Research), has made a thorough investigation of its properties, especially with regard to low temperature applications in vacuum vessels. Silica gel has only about half the ab- sorbing power of the best charcoal and it is also of lower density so that, when it is used in vessels constructed for charcoal, greater care is neces- sary in evacuation to ensure that the pressure is low enough not to leave too much air to be ab~orbed.~ Precautions to be Observed in Exhc ustirzg the Inters-ace.I n charcoal absorbent vessels the evacuation is usually of quite a low order, and several vessels have been found to have as much as I mm. of mercury gas pressure in the vacuum space at ordinary temperatures and have yet remained serviceable. The chief objection to this high pressure is the prolonged interval required by the charcoal to absorb all the air when the vessel is filled. This may take as long as half an hour. I n evacuating vessels by any method a fundamental point is heating the charcoal to redness immediately before inserting into the vessel, and also heating the entire vessel to as high a temperature as the seams will stand, which is usually about 100' C., and maintaining at this temperature until the vacuum ceases to fall of€ when the pump is stopped for a few moments.The chief object of this heating is to remove moisture. Kate 01. Loss froin Comntercial Vessels. The rate of evaporation of liquid oxygen from good comniercial vessels is of the following order of niagnitude :- 1 Atvial. dc Physique, 191s. 'J Pvocccdings qf thc Royal Iristitiition, 1906- €'roc-. R. S . , Edinburgh, 1920-1.FOR USE ON AIRCRAFT 2 2 9 Capacity in Litres Liquid. 1 Rate in Grams per Day. Percentage per Day. I 3 5 21 1 70 I I1 1290 I 4'5 I =5 840 5 40 660 The above values are averages relating to a large number of both German and English vessels that have been studied by the writer. A good glass vessel is therefore nearly twice as efficient as a metal vessel, but the average vessel, especially such as used for household purposes, is quite the reverse.Separatioiz of the Thermal Leakage into its Components. The sources of loss in a metal vessel can be divided into three classes, i, conduction down the neck tube, ii, gas coiiduction across the interspace, and iii, radiation between the two walls. The percentage rate is rate of loss calculated for a full vessel: the percentage calculated on the actual contents increases uniformly as the vessel empties since the rate of evaporation does not fall off as the volume of liquid decreases. ( a ) Condzrction down the iVeck.-'l'he magnitude of the heat leakage into the flask by conduction along the neck depends upon the length of the neck, its cross-sectional area, and the thermal conductivity of the material. Manganin with the low thermal conductivity of 0.035 C.G.S.units is a suitable material for the construction of the neck, but thin-walled tubes of this material are difficult to obtain. The nearest approach to manganin is German silver which is extensively used in Germany for the purpose, but it is difficult to obtain tubes in this country. The only alloy readily obtain- able is copper-nickel (German silver without zinc). 'This is of distinctly higher conductivity and is nioreover liable to the development of flaws in manufacture. For practical reasons connected with filling it is undesirable to use tubes smalier than I cni. bore and they should preferably be 14 cms. 'The thickness of wall is generally of the order of 0.8 to I mm. so that the area of metal is about 24 to 45 sq.mm. Some years ago it was pointed out by Swan in connection with Regnault's work on the specific heat of gases that the conduction of heat along a tube is influenced by the flow of gas i n the tube, and that this is a factor to be considered in calculating heat leakage has also been pointed out by Professor Hriggs. I t is not therefore sufficient to assume that the two end tempera- tures determine the thermal transfer, but the actual gradient at the junction of the neck with the inner vessel must be considered. The temperature distribution along the neck of a flask containing liquid oxygen is shown in Fig. 3. I t may be remarked that the gas-temperature distribution has already been published. Tne temperatures shown are actual neck temperatures obtained by means of a fine thermocouple in contact with the metal wall and not the distribution of temperature along the gas stream.The E.M.F. generated by the copper-constantan thermocouple was measured by a potentiometer and the necessary precautions taken. The upper curve refers to an alloy tube and the lower to a brass tube of The evaporation Prom t!ie one litre size is unduly high, but very few of these have been made and no doubt they are capable of considerable improvement.230 THE PRODUCTION OF LIQUID OXYGEN fairly high conductivity selected for contrast. The gas flow was about three-quarters of a litre in each case. I n all the flasks tested a gradient of temperature was evident at the bottom of the neck. (8) Conduction throzgh ResiduaZ Gas.-The second factor in the efficiency of the vessel, gas conduction, depends on the nature of the residual gas, the pressure and the area of the vessel.The gas present is mostly air with probably water vapour and hydrogen in small quantities. No direct determination of the pressure appear to have been published, but Dewar and Briggs have investigated the pressure indirectly by submitting a flask to various external temperatures and calculating the losses from as- sumptions based upon the fact that the total loss = a(T41 - T4J + b(T4, - T4 2) J V T + constant loss for neck, where T, and T, are the temperature of the inner and outer walls. +a- T e m be r ct r c' re C. - - FIG. 3. 10 8 6 --inches- 4 2 0 Curves showing distribution of temperature along inner tube of neck.Top curve The temperature gradient depends upon the rate of flow of the gas out through alloy, bottom brass. the neck. The first term refers to radiation transfer, the second to conduction. For a metal flask Briggs found that the conduction was approximately equal to the neck loss at room temperature, and each constituted 2 5 per cent. of the total. These experiments were carried out with great care on flasks which are probably specially constructed and well evacuated, for the writer has found that the method when applied to commercial vessels passing through a cycle of temperatures changes gave the result shown in Fig. 4. In this case gas is evidently evolved from the walls even at moderate temperature elevations, and although the flask may have been heated during evacuation (the pressure is, of course, much lower when liquid oxygen is present in the flask than during evacuation period), the evaporation rate does not completely recover and return to its normal value for a consider- able time.Hence the evolved gas is not readily absorbed by the charcoal;FOR USE ON AIRCRAFT 2 3 1 the gas may be hydrogen in which case the value of the conductivity would be appreciably different from that of air. The above method should therefore be applied with some caution except in the case of glass vessels. (c) Radiation Transfer between the WaZk-The factor of greatest importance is the radiation transfer. The heat transferred by this means depends upon the emissivity of the walls and the fourth power of the absolute temperatures.The selection of the best reflecting surface for a long wave-length radiation at low temperature is still a matter for research as vacuum flask tests are of limited help in view of the many variables involved. Copper, gilding metal (a brass alloy), nickel and silver plating give much the same results within the limits of the order of accuracy of the experiments. Iron appears to be inferior, for it is difficult to obtain a high polish on it. The degree of polish of the walls is of considerable importance. I t is probable that little improvement is possible in the efficiency of the best metals without radical changes in design. (d) Gse of n Radiation ShieZd-One promising line of investigation to 0 2j0 50" / 5 I ooo r -0 Temperature of outer wall in degrees C.FIG. +-Effect of a cycle of temperature on the rate of a flask. follow would be to effect a reduction in the radiation transfer by the use of radiation shields. The chances of improvement are greatest here as this effect is at present responsible for the largest loss. I t can be shown that the effect of reflecting surfaces is to reduce the radiation by 2- where n is the number of the intermediate surfaces, n + I assuming that they touch neither wall. 'The writer has constructed one vessel with a radiation shield as shown in Fig. 5. The intermediate wall is supported from the neck by a thin steel tube which is slotted to diminish the cross-sectional area, but which yet provides a stiff support. This vessel has been used for transport purposes for about two years, and the efficiency is about 15 per cent.better than a vessel of the same size with two walls. Nethods of Genera fing Gas from the Lipid as Desired. When liquid oxygen is used as a substitute for compressed gas the liquid must be evaporated at a definite and steady rate to suit the require- ments of the airman. ( a ) KeyZaandt's Vajoriser.-The method devised by Heylacdt is shown diagrammatically in Fig. 6.232 THE PRODUCTION OF LIQUID OXYGEN The liquid is carried in a vacuum vessel and ejected by the pressure due to its natural evaporation and the stream passes up the neck through a small tube into a flask boiler where it is evaporated. The gas passes through a trap to catch any drops of liquid and thence through coils of copper pipe to the control valve which is graduated two, three, or five litres of gas per minute. The pressure above the liquid in the container is shown by a small pressure gauge and regulated by a relief valve.The delivery of oxygen for FIG. 5. Metal vacuum vessel fitted with radiation shield. This shield is carried from a tube attached to top of neck. The tube is slotted away to reduce heat conduction along it. FIG. 6. IKsauilc PlWQL l K U A I E VALYL i I I b I I I I I I I I , I 6 I any given setting of the valve depends iipon several factors and varies from time to time for the same pressure reading. (b) Grifiths' V'poriser.-The writer has also designed an alternative form of vaporiser as shown in Fig. 7, the action of which depends upon the varying metallic contact between the walls of the vessel.A copper pad is carried by a flexible diaphragm on the outer wall and adjusted into contact with the inner vessel by a screw. The evaporation can then be controlled with accuracy. It is important that the diaphragm should be a good conductor such as silver or copper. The advantage of this form of vaporiser arises from the fact that the vessel may be non-spillable by absorbing the liquid intoFOR USE ON AIRCRAFT 233 asbestos or similar material. of the weight of the oxygen will suffice to absorb the liquid. A quantity of asbestos about 10 per cent. PART 11. PRODUCTION OF LIQUID OXYGEN. The entire output of oxygen in this country, apart from a small medical demand, was in the early days of the war absorbed by the welding industry. The only plants available for experimental work were small units of the form shown in Fig.8. Delivery pipe. Asbestos wool. Tube containing charcoal. Copper pad. Corrugated diaphragm of silver. Adjusting nut. FIG. 7. Descr$tion of Laboratory Pla7zt. The air is drawn in through a line of purifiers on the left, compressed to about 1,800 lbs. per sq. inch in a three-stage compressor and thence pas- sing through a water separator and caustic potash column into a Hampson liquefier. The lubrication of the compressor is largely effected by water and the separator deals with this moisture. The liquefier consists of flat wound spirals of 4 inch copper pipe; four tubes in parallel. At the bottom, these tubes merge into a valve which regulates the expansion. The output of these small plants is about one-third of a litre per horse- power hour of liquid air which must be rectified to obtain a high oxygen percentage.The plant is fairly reliable if fitted with direct electric motor drive234 THE PRODUCTION OF LIQUID OXYGEN running at not more than 250 revolutions per minute, but chain, gear, or belt drive is troublesome on small compressors, since they are not generally provided with sufficently large flywheels. The method of constructing the interchanger is open to improvement as caustic, if carried over, permanently chokes the tubes and the output suffers accordingly. s k I t is well known that the Joule-Thomson effect employed by Linde gives a cooling effect of the order of a quarter of a degree per atmosphere, whereas the Claude system of malting the air do external work gives a cool- P ing effect T, = 0.29 To 2, YeFOR USE ON AIRCRAFT 2.35 where To = initial temperature T, = final temperature Po = final pressure (atmos.) P, = pressure before expansion.This leads to the conclusion that the cooling effect is about three times as great with the expansion engine; but in practice the expansion engine becomes very troublesome near the liquefaction point and the Joule- Thomson effect increases so that a combination of the two methods is usually employed. There is, however, very little difference in the perform- ance in small plants with or without an expansion engine, but the out- standing feature which can be utilised is the fact that an expansion engine plant can be designed for about 40 to 5 0 atmospheres pressure a t the compressor.Liquid Oxygen PZant with Exjansion Eizgim. T o illustrate the general scheme of an expansion engine plant, one built by the Liquid Air and Rescue Syndicate will be described as it is the simplest, although it uses high compression. ,4ir is drawn by the com- pressor, Fig. g, through the scrubber in order to remove as much as pos- sible of the carbon dioxide. The air enters the tower at the bottom. The absorbing agent ( I z per cent. caustic soda solution) on the other hand flows from the top to the bottom through steel turnings and so meets the air and absorbs the COz. This purifier requires the circulation of the alkali by a pump which draws the liquid from the bottom and delivers it at the top. The air freed from CO, is drawn into the four stage coni- pressor and compressed to about zoo atmospheres (3000 Ibs.sq. inch). Between each stage is an intercooler. The air from the compressor passes through a water and oil separator to the high pressure drying bottle. Before entering the drying bottle the air passes through coils which are surrounded by the cold air leaving the liquefier and part of the water vapour is condensed. The remaining moisture is removed by the drying bottle. The dry air passes to the liquefying plant which consists of a liquefier and an expansion engine. Part of the air passes through air-cooled coils on the outside of the liquefier and then enters the cylinder of the expansion engine in which it is expanded down to a low pressure, thus becoming cooled. This cold air passes through the separator and then enters the liquefier near the bottom by the pipe and rises through the heat interchanger.I n this interchanger are many small pipes through which the other part of the compressed air passes downwards and gives up heat to the rising air from the expansion engine until this air finally reaches the expansion valve, cold but at high pressure. The air is expanded through the expansion valve to about atmospheric pressure and is thus further cooled. The cold air mixes with the expansion engine exhaust and helps to cool still further the incoming high pressure air. By this means the air reaches the expansion valve at a continuously reduced temperature until part of it is liquefied on expansion. The liquid formed collects at the bottom of the liquefier and can be drawn off by a cock.The enrichment of the air in oxygen depends on the different boiling-points of the oxygen and nitrogen. ‘The rectification column is situated at the bottom of the apparatus. The up-going gas meets the liquid flowing downwards and a gradual interchange takes place, the lower boiling-point nitrogen passing away as gas and osygen liquid collect in236 THE PRODUCTION OF LIQUID OXYGEN the bottom. The air leaving the interchanger passes through coils which cool the air passing to the expansion engine. The purity of this liquid is about 75 per cent. I I I c c a aJ .C W c x 0 &FOR USE ON AIRCRAFT 237 $%rtabZc CZaude Tyje oxygen pdnnt. One made by Sociktk L'Air Liquide Boulogne-sur-Seine, Paris, is carried normally on a railway truck.The general arrangement is shown diagrammatically in Fig. 10. 'The air is compressed in a two stage compressor A and delivered at a A plant using the Claude :;>'stem is more elaborate. pressure of about 45 atmospheres to the separator R where moisture absorbed from the injection water in the cylinders of the compressor is removed. The air then passes successively through the purifying chambers C (only one of which are shown), filled with caustic soda or caustic potash, and enters the liquefying column. The arrangement of the column and expansion engine are shown in Fig. I I . Air passes from the purifier by the pipe I into the temperature exchanger A, and traverses the nest of small238 THE PRODUCTION OF LIQUID OXYGEN tubes 2 which are cooled by expanded and cold gas flowing in the opposite direction.The cooled air leaves the temperature exchanger by the pipe 3 and part enters the expansion engine E, by the pipe 4, and the other part enters a nest of tubes 5 in the liquefier 6 by the pipe 10. The air enter- ing the expansion engine does ex- ternal work on the piston and is expanded and thus cooled. The compressed air in the nest of tubes in the liquefier is cooled by expanded and cold air flowing in the opposite direction and under the continued action of cooling and pressure liquefies, and passing through the needle valve 8, mixes in the pipe g with the cold air from the expansion engine. The - I cold air and liquid air enter the top of another nest of tubes 10 called the vapouriser, which is system.contained in the intermediate part of the rectification column. On leaving these tubes, the mixture of gas and liquid enters a sump 11 at the bottom of the column. From this sump it rises through the pipe 12 and regulating valve 13 to the top of the column. The top portion of the rectifying column contains a FIG. II.-Diagram of liquefier on Claude FIG. 12.-Portable liquid oxygen plan: working on Claude system. The expansion engine can be seen on the right, the column in the centre and the resistance frames for dissipatii-ig the energy developed in the dynamo attached to the expansion engine o n the right. The air compressor is situated behind the resistance grid. series of staggered and perforated trays 14 which are provided with caps 15. The liquid in the mixture descends through the trays and enters the vapouriser by means of the pipe 16 in the space surrounding the nest of tubes 10. The gas in the mixture, which contains a large quantity of nitrogen, since nitrogen has a lower boiling-point than oxygen, leaves the of the column by the pipe 1 7 and first enters the liquefier in the spaceFOR USE ON AIRCRAFl 239 surrounding the nest of tubes 5 , thus cooling the compressed air in the tubes. From there it enters the temperature exchanger by the pipe IS in the space surrounding the nest of tubes 2, and fulfils the same function of cooling the compressed air flowing in the opposite direction. I t leaves by the pipe 19, warmed but still at some degrees below the temperature of the incoming air. When the liquid air supply is sufficiently large for liquid oxygen to be produced, the valve 1 3 is slowly closed down but not completely shut. The exhaust pressure in the expansion engine then rises to the same pressure as that of the liquid air in the nest of tubes 10 in the vapouriser and causes the surrounding liquid to boil. The gas thus formed rises through the plates 14 while the liquid formed passes through the valve 13 and flows into the top of the columns. The liquid descends through the trays while the gas bubbles around each of the caps. The output of this plant is about 2 0 litres per hour of 96 per ccnt. purity oxygen. The horse-power by independent test of the compressor is slightly less than 60 or 3 H.P. per litre per hour. A photograph of the complete plant is shown in Fig. 12. The expansion engine on the extreme right drives a dynamo, the output of which is absorbed by a variable resistance. About 3 electrical horse-power is produced in normal running. This plant produces liquid in four hours, the average output being 43 lbs. per hour for each successive hour. The pressure at the compressor is 45 atmospheres at 250 K.P.M., and the expansion engine runs at 160 R.P.M., producing 2500 watts. The purity of the liquid oxygen is 96 per cent.

ISSN:0014-7672    

DOI:10.1039/TF9221800224

出版商:RSC    

年代:1922

数据来源: RSC