February 7, 1895] 



NA TURE 



351 



VIII. Separation of Ar^on on a Large Scale. 



To prepare argon or. a large scale, air is freed from osygen 

 ■by means of red-hot copper. The residue is then pased from 

 a gas-holder through a combuslion tube, heated in a furnace, 

 and containing copper, in order to rem ive all traces of oxyjjen ; 

 the issuing gas is then dried by passage over soda-lime and 

 phosphorus pentoxide, after passage through a small (J lube 

 containing sulphuric acid, to indicate the rate of How. li then 

 «Dters a combustion-tube packed tighily with magnesium turn- 

 ings, and healed to redness in a >econd furnace. P'rom this 

 tube it passes through a sec(»nd index tube, and enters a small 

 gas-holder capable of containing 3 or 4 litres. A single tube 

 of magnesium will absorb from 7 to 8 litres of nitrogen. The 

 temperalure must be nearly that of the fusion of the glass, and 

 the current of gas must be carefully regulated, else the heat 

 developed by the union of the magnesium with nitrogen will 

 fuse the tube. 



Having collected the residue from 100 or 150 litres of 

 atmospheric nitrou'en, which may amount to 4 or 5 litres, it is 

 transferred to a small gas-holder connected wiih an ajjparatus, 

 wherel-y, by means of a species of a self-acting Sprengel's pump, 

 the gas is caused to circulate through a tube half filled with 

 copper and half with copper oxide ; it then traverses a tube 

 ilalf filled with soda lime and half with phosphorus pentoxide ; 

 it then passes a reservoir of aboui 300 c.c. capacity, from which, 

 by raising a mercury reservoir, it can be expelled into a small 

 gas-holder. Next it passes through a lube containing mag- 

 nesium turnings heated to bright redness. The gas is thus 

 freed from any possible contamination with oxygen, hydrogen, 

 or hydrocarbons, and nitrogen is gradually absorbed. As the 

 amount of gas in the tubes and reservoir diminishes in volume, 

 it draws supplies from the gas-holder, and finally, the circulat- 

 ing system is full of argon in a pure state. The circulating 

 system ol tubes is connecied with a mercury pump, so that, in 

 changing the magnesium tube, no gas may be lost. Before 

 ceasing to heat the magnesium tube the system is pumped 

 empty, and the collected gas is restored to the gas-holder ; 

 finally, all the argon is transferred from the mercury reservoir 

 to the second small gas-holder, which should preferably be filled 

 with water saturated with argon, so as to prevent contamination 

 from oxygen or nitrogen ; or, if preferred, a mercury gas-holder 

 may be employed. The complete removal of nitrogen from 

 argon is very slow towards the end, but circulation for a couple 

 of days usually effecis it. 



The principal objection to the oxygen method of isolating 

 argon, as hitherto descri'ied, is the extreme slowness of the 

 operation. In cxtenHing the scale we had the great advantage 

 of the .advice ol Mr. Crookes, who not long since called .ntten- 

 tion to the flame rising from platinum terminals, which convey 

 a high tension alterna'ing electric discharge, and pointed cut its 

 dependence upon combustion of the nitrogen and oxygen of the 

 air.' The plant consists of a De Mcritens alternator, actuated 

 by a gas engine, and the currents ate transformed to a high 

 potential by means of a RuhmkorlT or other suitable indue ion 

 coil. The highest rate of absorption of the mixed gasc; yet 

 attained is 3 litres per hour, about 3CCK) times that of Caven- 

 dish. It is necessary to keep the ajiparatus cool, and from this 

 and other causes a good many difficulties have been en- 

 countered. 



In one experiment of this kind, the total air led in after 

 seven days' working, amounted to 7925 c.c, and of oxygen 

 {prepared from chlorate of potash), 9137 c.c. On the eighth 

 and ninth days oxygen alone was added, of which about 

 500 c.c. was consumed, while there remained about 700 c.c. in 

 the flask. Hence the proportion in which the air and oxygen 

 combined was as 79 : go. The progress of the removal of the 

 nitrogen was examined from time to time with the spectro- 

 scope, and became ultimately very slow. .Vt last the yellow 

 line disappeared, the contraction having apparently stopped for 

 two hours. It is worthy of notice that, with ihe removal of the 

 nitrogen, the arc discharge changes '.greatly in appearance, 

 becoming narrower and blue rather than greenish in colour. 



The final treatment of the le idual 700 c.c. of uas was on the 

 model of the small scale operations already described. Oxygen 

 or hydrogen could be .supplied at pleasure from an electrolytic 

 apparatus, hut in no way could the volume be reduced below 

 65 c.c. This re idue refused oxidation, and showed no trace 

 of the yellow line of nitrogen, even under favourable condi- 

 tions. 



' Clicmtcal Xc-vs^ vol. Iw. p. 301, iSjr. 



When the gas stood for some days over water, the nitrogen 

 line reasserted itself in the speclmm, and many hours' sparking 

 with a little oxj'gen was rei|uired again to get rid of it. Inten- 

 tional additions of air to gas free from nitrogen showed that 

 about li percent, was clearly, and about 3 percent, was con- 

 spicuously, visible. About the same numb rs apply to the 

 visibility of nitrogen in oxygen when sparked under these 

 conditions, that is, at atmospheric pressure, and with a jar 

 connected to the secondary terminals. 



IX. Density of Argon prepared by means of Oxygen. 



A first estimate of the density of argon prepared by the 

 oxygen method was founded upon the data already recorded 

 respecting the volume present in air, on the assumption that 

 the accurately known densities of atmospheric and of chemical 

 nitrogen differ on account of the presence of argon in the former, 

 and ihat during the treatment with oxygen nothing is oxidised 

 except nitrogen. Thus, if 



D = density of chemical nitrogen. 

 D' = ,, atmospheric nitrogen, 



tl = „ argon, 



a = proportional volume of argon in atmospheric 

 nitrogen, 



the law of mixtures give 



ar/-f (I- a)D = r)', 

 or d=D~{D'-D)/a. 



In this formula D' - D and a are both small, but they are 

 known with fair accuracy. From the data already given 



079 -7925 



whence if (on an arbitrary scale of reckoning) D = 2'2990, 

 D' = 23i02, we find </=3'378. Thus if N^ be 14, or O; be 16, 

 the density of argon is 20'6. 



-\ direct determination by weighing is desirable, but hitherto 

 it has not been feasible to collect by this means sufficient to 

 fill the large globe employed for other gases. .\ mixture of 

 about 400 c.c. of argon with pure oxygen, however, gave the 

 weight 27315, o'l045 in excess of the weight of oxygen, viz. 

 20270. Thus, if a he the ratio of the volume of argon to the 

 whole volume, the number for argon will be 



2'627o - O'i045/o. 



The value of a, being involved only in the excess of weight 

 above that of oxygen, doss not require to be known very ac- 

 curately. .Sufficiently concordant analyses by two methods gave 

 a — o'i845 ; whence for the weight of the gas we g^t 3 193, so 

 that, if O., = 16, the density ol the gas would be I9'45. ^^ 

 allowance for re-idual nitrogen, still visible in the gas before 

 admixture of oxjgen, raises this number to 197, which may bs 

 t.aken as the density of pure argon resulting from this deter- 

 mination. 



X. Ilensity of .Argon prepared by means of .Magnesium. 



The density of the original sample of argon prepared has 

 already been mentioned. It was 1909 ; and, afier sparking 

 with oxygen, it was calculated to be 200. The most reliable 

 results of a number of determinations give it as I9'90. The 

 difficulty in accurately determining the density is to make sure 

 that all nitrogen has been removed. The sample of density 

 1990 showed no spectrum of nitrogen when examined in a 

 vacuum tube. It is right, however, to remark that the highest 

 density registered was 20'3S. But there is some reason here to 

 distruit the weighing ot the vacuous globe. 



XI. Spectrum of Argon. 



The spectrum of argon, seen in a vacuum tube of about 

 3 mm. pressure, consists of a great number of lines, dislributed 

 over almost the whole visible field. Two lines are specially 

 characteristic ; they are less refrangible than the red lines of 

 hydrogen or lithium, and serve well to identify the gas, when 

 examined in this w.ay. Mr. Crookes, who will give a full ac- 

 count of the spectrum in a separate communicatiim, has kindly 

 furnished us with Ihe accurate wave-lengths of these line-', as 

 well as of some others next 10 be described ; they are respec- 

 tively 696 56 and 70s '64, 10 ■'■ mm. 



licsides these red lines, a bright yellow line, more refrangible 

 than the sodium line, occurs at 603 S4. .\ group of five bright 



NO. I319, VOL. 51] 



