NOTES.
219
Without this control cements would be used that were not suited for the purpose designed, and much damage would be occasioned in different cities, which damage would niuch cripple the asphalt paving industry. Cargoes of cement are shipped, one requirement of the cement being that it shall have a certain penetration number previously determined by the contracting parties. Since the introduction of this testing apparatus many qualities of cement hare been studied and others discovered. By it we have learned the influence of mild or sudden and marked changes of temperatures, and the consequence of severe cold. I t enables us to learn the influence of hardening and softening agents, a matter of great importance in the practical uses of asphalt cement. Thus it is manifest that this instrument has an important and serious part to perform in a great industry, and that its use is far from being an idle pastime. SCHOOL OF S f I N E S , C O L ~ - M E I ACOLLEGE, N. Y .CITY, January 23. I&,.
NOTES. Argon : A 'Vew Constituent of the Atmosphere.-At the meeting of the Royal Society, held on January 31, the long-expected paper by Lord Rayleigh and Professor Ramsay was read, and. a full report has just been received by way of The Cheiuicnl .?'ews, of February I , from nhich the following condensed summary is taken : A careful comparison of the nitrogen from urea, amnloniunl nitrite, and from nitrous atid nitric oxidchemical nitrogen" with atmospheric nitrogen was made, and gave a weight per liter for Chemical nitrogen of.. 1.2505 grams Atmospheric " 1.2572 " I'
............ ..............
Nitrogen which had been extracted from the air by means of magnesium was separated also and gave a figure differing inappreciably from that recorded above for " chemical nitrogen."' The nitrogen contained in magnesium nitride was also converted into ammonium chloride a n d this was found to contaiii exactly the proportion of chloriue contaliled in ordinary animonium chloride. From this it was concluded t h a t red-hot magnesium withdraws from " atmospheric nitrogen " no suhatance other than nitrogen capable of forminga hasic compound with hydrogen. l f t e r having endeavored in every way possible to detect i n " atmospheric nitrogen" known gases to account for the difference in specific gravity t h e authors finally repeated Cavendish's experiment. I t will he renieinbered that Cavendish found that when air to which oxygen was adcted,tlie electric spark passed for several days and the nitrous and uitric acids 1
See also page
2x1
of this issue.
KOTES.
22 0
removed by soap-lees and finally the oxygen removed by liver of sulphur that “ only a small bubble of air remained unahsorbed which certainly was not more than the i j o of the hulk of the phlogisticated air let u p into the tube ; so that if there is any part of the plilogisticated air of our atmosphere which differs froni the rest and can not be reilucetl to nitrous acid, we may safely conclude that it is iiot more than i i s part of the whole.” Improving the apparatus of Caveiitlish 60 as to shorten tlle duration of tlie experiment Kayleigh aiitl Ramsay founri that fifty cc. of air left 0 . 3 2 cc. of gas. O n adding this residue to a fresh fifty cc. of air arid repeating tlie operation the residue now amounted to 0.76 cc. By passing nitrogen over magnesium turnings contained i i i a heated tube, 1,500 cc. unabsorbed gas was obtained. and after passing this for several days over soda-liiiie, phosphoric anhydride, tiiagnesiuiii a t a red heat, and copper oxide, the gas \vas reduced to ZM cc. aiid its density found to he 16.I . After further absorption the density was increased to 19.09. On passing sparks for several hours through a mixture of a srriall quantity of this gas with oxygen, its ~ o l u l n ewas still further reduced. Assuming that this reduction was tluc to t h e further eliniinatioxi of nitrogen the cleiisity of the reiriaiiiiiig gas w a s calculated to he z o . o . By means of atinolysis usiiig long clay pipe-stems the atnouiit of argori was increased in tlie air, which finally passed from the pipes so that the nitrogen i n it weighed i n a total weight of approxjtnately two arid three-tenths grams a mean of o.oorS7 gram over that of the same volu m e of atriiospheric nitrogen.” By increasing tllc efficiency of the apparatus the excess \vas in mean o.oojj gr:irn. Experiments tlieii niatle to p r o w the ;il)sence or prc-sence of argon it1 chemical iiitrogen showed that three liters of chciiiical nitrogen from aiririioiiiuiii nitrite left three aud three-teiitlis cc. of argon, a part of which is accountetl for l)y ail accident. 1x1 a second experiment 5,660 cc. of tlie same nitrogen left three and five-tenths c r . argon. The source of this \vas fount1 i n tlic water used for confining the gases \vhich absorbs argon from tlie air in considerable amount. and again gives it u p to the confined gases. T h e amount of argon ohtained frotn the chemical nitrogen %-as less than 2fi of the iiorrnal aniount from atmospheric nitrogeii. The following quotations from the paper of Kayleigh and Kaiiisay and from those of Crookes ant1 Olszewski, which followed it, are scarcely susceptible of condensation : I ’
Separafiorz of Aygon on a L a r g e .G.alt,. To prepare argon on a large scale, air is freed from oxygen by nieans of red-hot copper. T h e residue is tlieii passed froni a gas-holder through a combustion tube, heatetl in a fiirnace, aiitl contaiiiing copper, in order t o remove all traces of oxygen : the issuing gas is then dried hy passage over soda-lime a n d phosphorus pentoxide, after passage through a small tube containing sulphuric acid, to indicate the rate of flow. It tlieri enters a combustion tube packed tightly with iriagnesium turiiings, a d heated
u
NOTES.
221
t o redness :n a second furnace. From this tube it passes through a second index-tube, and enters a small gas-holder capable of containing three or four liters. A single tube of magnesium will absorb from seven to eight liters of nitrogen. The temperature must be nearly t h a t of t h e fusion of t h e glass, and t h e curreut of gas must be carefully regulated, else t h e heat developed by t h e union of t h e magnesium with nitrogen will fuse t h e tube. Having collected t h e residue from roo to 150 liters of atmospheric nitrogen, which may amount to four or five liters, it is transferred to a small gas-holder connected with an apparatus, whereby, 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 half filled with soda-lime and half with phosphorus pentoxide ; it then passes a reservoir of about 300 cc. capacity, from which, by raising a mercury reservoir, it can be expelled into a small gas-holder. Next it passes through a tube containing magnesium turnings heated to bright redness. T h e gas is thus freed from any possible contamination with oxygen, hydrogen, or hydrocarbons, and nitrogen is gradually absorbed. As t h e amount of gas in t h e tubes and reservoir diminishes in volume it draws supplies from t h e gas-holder, and, finally, the circulating system is full of argon in a pure state. The circulating system of tubes is connected with a mercury pump, so that, in changing t h e magnesium tube, no gas may be lost. Before ceasing to heat t h e magnesium tube t h e system is pumped empty, and t h e collected gas is restored to t h e gas-holder; finally, all the argon is transferred from t h e mercury reservoir to t h e 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 t h e end, but circulation for a couple of days usually effects it. The principal objection to t h e oxygen method of isolating argon, as hitherto described, is the extreme slowness of t h e operation. I n extending the scale we had t h e great advantage of t h e advice of Mr. Crookes, who not long since called attention t o t h e flame rising from platinum terminals. which convey a high tension alternating electric discharge, and pointed out its dependence upon combustion of t h e nitrogen and oxygen of t h e air.’ The plant consists of a De Meritens alternator, actuated by a gas engine, and t h e currents are transformed t o a high potential by means of a Ruhmkorff or other suitable induction coil. T h e highest rate of absorption of t h e mixed gases yet attained is three liters per hour, about 3,000 times t h a t of Cavendish. I t is necessary to keep t h e apparatus cool, and from this and other causes a good many difficulties have been encountered. I n one experiment of this kind, t h e total air led i n after seven days’ working, amounted to 7,925 cc., and of oxygen (prepared from potassium 1 Chemzcal News,65, 301~ 1892.
NOTES.
222
chlorate), 9x37 cc. On the eighth and ninth days oxygen alone was added, of which about joo cc. was consun~erl,while there retiiaiiied about 700 c c . in the flask. Hence the proportion in which the air am1 cxygen coinhinecl was as 79 : 96. T h e progress of t h e removal of tlie iiitrogen was examined from time to time with the spectroscope, and 1)ecanie ultirnately very slotv. .At last the yellow line disappeared, the contraction having apparently stopped for two hours. It is worthy of notice :hat with the removal of t h e nitrogen, t h e arc discharge changes great:) i n appearance, hccorning narrower and hlue.rather than greenish ill x l o r . T h e final treatment of t h e residual 700 cc. of gas was 011 t h e riiotlel of the small scale operations, already tlescrihed. Oxygen or hydrogen coiild be supplied, a t pleasure, froin an electrolytic apparatus, but in no way coiilti the volunic he reduced below sixty-five cc. This residue' refusetl oxidation and showed no trace of the yellow line of nitrogen, even under favorable conditions \'+%en the gas stood for some days over water, the nitrogen line reasserted itself iri t h e spectrum, and iiiaiiy hour's sparking with a little oxygeii was reqnired again t o get rid of it. Intentional additions of air to gas free from nitrogeii showed that about one and one-half per cent. was clearly, and ahout three per cent. was conspicuously, .risible. About tlie saiiie n u ~ n h e r sapply to t h e visibility of nitrogen in oxygen when sparked untler these coiiditions, that is, a t atniospheric pressure, and with a jar coniiecteil to the secoiidary teriniiials.
Density of A r g o n prepared by m e a m of #.qlgen. A first estimate of the density of argon prepared b y the oxygen method was founcletl upon t h e d a t a already recorded respecting t h e volume present ill air, on the assuiiiption that the accurately kiio\vu densities of atniospheric and of cheiiiical nitrogen differ on acco~iiitof the presence of argon i n the former, arid that during the treatment with oxygeil nothing is oxi-
dized except nitrogen. T h u s , if D = density of chemical nitrogen, I)'= " atmospheric nitrogen, cz' = ' ' argon, a = proportional voluine of argon in atniospheric nitrogen, t h e law of mixtures give~d ( I-u)I> = D', ord =D (D'-D)/a. In this formula D'- D and a are both small, but they are known with fair accuracy, Frorri the data already given' I
+
+
a
= -5-
0.79 x 7925 whence if (on an arbitrary scale of reckoning') D = 2.2990,D'= 2.3102,w e fiiid d = 3.378. T h u s if S, be 14, or 0, be 16, the density of argon is 20.6. .A direct deterniination by weighing is desirable, but hitherto it has not
NOTES.
223
been feasible to collect by t h i s means sufficient to fill t h e large globe employed for other gases. A mixture of about 400cc. of argon with pure oxygen, however, gave t h e weight 2 . 7 3 1 5 , 0.1045 in excess of t h e weight of oxygen, viz., 2.6270. Thus, if a he t h e ratio of t h e volume of argon to the whole volume, t h e number for argon will be2.6270 0.104g/a. T h e value of a , being involved only in t h e excess of weight above that of oxygen, does not require to be known very accurately. Sufficiently concordant analyses by two methods gave a = 0.1845; whence for t h e weight of the gas we get 3.193,so t h a t , if O,= 16, t h e density of t h e gas would he 19.45. An allowance for residual nitrogen, still visihle in t h e gas before admixtiire of oxygen, raises t h i s number to 19.7, which may be taken as t h e density of pure argon resulting from this determination.
+
Density of Arxon prepared by means of Magnesium. The density of t h e original sample of argon prepared has already been mentioned. I t was 19.09; and, after sparking with oxygen, it was calculated to be 20.0. The most reliable results of a number of determinations give it as 19.90. T h e difficulty in accurately determining t h e density is t o make sure t h a t all nitrogen has been removed. The sample of density 19.90 showed no spectrum of nitrogen when examined in a vacuum tube. I t is right, however, to remark that the highest density registered was 20.38. But there is some reason here to distrust t h e weighing of the vacuous globe. Spectrum of Argon. The spectrum of argon,seen in avacuum tube of about three mm.pressure consists of a great number of lines, distributedover aliiiost t h e whole visible field. Two lines are specially characteristic; they are less refratigihle than the red lines of hydrogen or lithium, and serve well to identify t h e gas, when examined in this way. Mr. Crookes, who will give a full account of t h e spectrum in a separate comniunication, has kindly furnished us with t h e accurate wave-lengths of these lines, RS well as of some others next t o be described; they are respectively 696.56 aiid 705.64, IO-^ min. Besides these red lines, a bright yellow line, more refrangible than t h e sodiuni line, occurs a t 603.84. A group of five bright green lines occurs next, besides a number of less intensity. Of t h e group of five, the second. which is perhaps the most brilliant, has t h e wave-length 561.00. There is next a blue or blue violet line of wave-length470.2 ; aiid last, i n t h e less easily visible part of t h e spectruni, there are five strong violet lilies, of which t h e fourth, which is the most brilliant, has t h e wave-length 420.0. Unfortunately, the red lines, which are not to be niistakeii for those of any other substance. are not easily seen when a jar discharge is passed through argon a t atmospheric pressure. The spectrum seen under these conditioiis has been examined by Professor Schuster. The most chardc-
224
NOTES.
teristic lines are perhaps those i n the neighborhood of F, and are very easily seen if there be not too much nitrogen, in spite of the presence of some oxygen and water vapor. The approximate wave-lengths are487.91,. .................. Strotig. [486.07] .................. F. 484.71 .................... S o t quite so stroug. 480.52 .................... Strong. 476.50.. .................. 473.53.. .................. Fairly strong characteristic triplet. 472.56.. ..................
1
I t is necessary to anticipate Mr. Crookes’ communication, and to state that when the current is passed froin the induction coil in one .direction, that end of the capillary tube next the positive pole appears of a redder, and that next the negative pole of a hluer hue. There are, in effect, t’rvo spectra, \vhich Mr. Crookes has succeeded in separating to a considerable extent. Mr. E. C. Baly,’ who has noticed a similar phenomenon, attrihutes it to the presence of two gases. He says : X’heu an electric currelit is passed through a mixture of two gases, one is separated from the other aiicl appears in the negative glow.” The couclusion would follow that what we have termed “argon” is in reality a mixture of two gases which have as yet not been separated. This conclusion, if true, is of great importance, and experiments are now in progress to test it by the use of other physical methods. The full bearing of this possibility will appear later. The presence of a small quantity of uitrogen interferesgreatly with the argon spectrum. But w e have found t h a t ill a tube with platinum electrodes, after the discharge has been passed for four hours, the spectrum of nitrogen disappears, and the argon spectrum niauifests itself in full purity. A specially constructed tube with rnagnesiuni electrodes, which we hoped would yield good results, reiiioved all traces of nitrogen, it is true ; but hydrogen was evolved from the magiiesiuin, anti showed its Characteristic lines very strongly. However, these are easily identified. The gas evolved on heating magnesium in uacuo, as proved by a separate experinient, consists entirely of hydrogeii. X r . Crookes has proved the identity of the chief lines of the spectrum of gas separated from air-nitrogen by aid of magnesium with that remaiuiiig after sparking the air-nitrogen with oxygen i n presence of caustic soda solution. Prof. Schuster has also found the principal lines ideutical in the spectra of the two gases, as observed by the jar discharge a t atmospheric pressure. Solubility of Avgon in Water. Determinations of the solubility in water of argon prepared by sparking, gave 3.94 volumes per 100 of water a t 1 2 ~ .T h e solubility,of gas preI ‘
1 Pi.oc. P h p . Soc.. 1x93, p, 147.
NOTES.
225
pared by means of magnesium was found t o be 4.05 volumes per 100 a t 1 3 . 9 ~ . The gas is therefore about two and one-half times as soluble as nitrogen, and possesses approximately t h e same solubility as oxygen. The fact t h a t argon is more soluble than nitrogen would lead us to expect it in increased proportion in t h e dissolved gases of rain-water. Experiment has confirmed this anticipation. ’(Nitrogen” prepared from t h e dissolved gases of wate? supplied from a rain-water cistern was weighed upon two occasions. The weights, correspondin‘g to those recorded * * were 2.3221 and 2.3227, showing an excess of twenty-four mgms. above t h e weight of true nitrogen. Since t h e corresponding excess for atmospheric nitrogen” is eleven mgnis. we conclude that t h e water “ nitrogen” i s relatively more t h a n twice as rich in argon. On the other hand, gas evolved from t h e hot spring a t Bath, and collected for us by Dr. A . Richardson, gave a residue after removal of oxygen and carbon dioxide, whose weight was only about midway between t h a t of true and atmospheric nitrogen.
Behavior a t Low Tewzjevatures.‘ Preliminary experiments, carried out to liquefy argon at a pressure of about 100 atmospheres, and a t a temperature of -O, failed. No appearance of liquefaction could be observed. Professor Charles Olszewski, of Cracow, t h e well-knowu authority on t h e constants of liquefied gases a t low temperatures, kindly offered to make experiments on t h e liquefaction of argon. His results are embodied in a separate communication, but it is allowable to state here t h a t the gas has a lower critical-point and a lower boiling-point than oxygen, and that h e has succeeded in solidifying argon to white crystals. The sample of gas h e experimented with was exceptionally pure, and had been prepared by help of magnesium. It showed no trace of nitrogen when examined in a vacuum tube.
Ratio of Specific Heats. I n order to decide regarding t h e elementary or compound nature of argon, experiments were made on the velocity of sound in it. I t will he remembered that, from the velocity of sound in a gas, t h e ratio of specific heat a t constant pressure to that at constant volume can be deduced by means of t h e equation-
when n is t h e frequency, t h e wave-length of sound, v its velocity, e the isothermal elasticity, d the density, ( I a t ) the temperature correction, C j t h e specific heat a t constant pressure, and CV that at constant voluiiie. I n comparing two gases a t t h e same temperature, each of which obeys Boyle’s law with sufficient approximation, and in using t h e same sound,
+
1 T h e arrangements for the experiments upon this branch of the subject were left entirely in Professor Ramsay’s hands.
NOTES.
226
many of these terms disappear, and the ratio of specific heats of one gas may be deduced from that of the other, if known, by means of the proportionA'd : A"d' : : 1.41 : A ' , where, for example. A and d refer t o air, of which the ratio is 1.41,according to observations by Kijntgeii, Wullrier3Kayser, and Jarnin and Richard. 'l'wo corripletely different series of observatioiis, one in a tube of about two ~ t i r i i .diatiieter, and one in oiie of eight i11111., made with entirely different samples of gas, gave, the first. 1 . 6 j as the ratio, and, the second, r .fir. Experiments made with the first tube. to test the accuracy of its working, gave for carbon dioxide tlie ratio p.276, instead of 1.288, the mean of all previous tleterriiiiiations ; and the half wave-length of sound in hydrogen \viis foalid to be 73.6, instead of 74. j, tlie iiieaii of those previously found. The ratio of the specific heats of Iiy(lrogel1 found was 1.39, instead of r . f o z . There ran br no douht, therefore. that argon gives practically the ratio of specific heats, aiz., I .66 proper to a gas i n which all the energy is transiatioiial. The only other gas which lias heel1 found to behave siniilarly is niercury gas, at a high teinperature.'
A ffr tnpts to Induce Chemical Coin6 i m t ion. Many attempts to induce argon to conibine will he described in full in the coriiplete paper. Suffice it to say here t h a t all such attempts have as yet proved abortive. Argon does not combine with oxygen in presence of alkali under the influence of the electric discharge, nor with hydrogen in presence of acid or alkali also when sparked ; nor with chlorine, dry or moist, when sparked ; nor with phosphorus a t a bright-red heat; nor with sulphur a t bright redness. Tellurium may be distilled in a current of tlie gas; so may sodium and potassium, their metallic luster reniaining unchanged. I t is unabsorbed by passing it over fused red-hot caustic soda, or soda-lime heated to bright redness ; it passes unaffected over fused and bright red-hot potassium nitrnte ; and red-hot sodium peroxide does not conibine with it. Persulphides of sodium and calcium are also without action a t a red heat. I'latiiiuui-black does not absorb it, nor does platiuuni sponge, and wet oxidiziiig and chlorinating agents, such as nitrohydrochloric acid, bromine water, bromine and. alkali, and hydrochloric acid and potassium permanganate, are eutirely without action. Experiments with fluorine are in contemplation, but the dificulty is g r e a t ; and an attempt will be made to produce a carbon arc in the gas. Mixtures of sodium and silica and of sodium and boracic anliydritle are also without action; hence it appears to resist attack b y iiascent silicon and by nascent boron. Ceneval Cbwclusions. I t remains, finally, to discuss the probable nature of the gas, or mix1 Kundt
arid Warburg. PO.^.., :I,itr..
135, 337 and 527.
NOTES.
227
t u r e of gases, which we have succeeded in separating from atmospheric air, aiid which we provisionally name argon. The presence of argon i n t h e atmosphere is proved by many lines of evidence. T h e high density of “atmospheric nitrogen,” t h e lower density of nitrogen from chemical sources, and t h e uniformity in t h e density of samples of chemical nitrogen prepared from different conlpounds, lead to the conclusion that t h e cause of the auomaly is the presence of a heavy gas in air. If that gas possess the density twenty compared with hydrogen, ‘‘ atmospheric ’’ nitrogen should contain of it approximately one per cent. This is, in fact, found to be t h e case. Moreover, as nitrogen is removed from air by nieans of red-hot magnesium, t h e density of t h e remaining gas rises proportionately t o t h e concentration of t h e heavier constituent. .Second.-This gas has been concentrated in t h e atmosphere by diffusion. I t is t r u e that it has not been freed from oxygen and nitrogen by diffusion, but t h e process of diffusion increases, relatively to nitrogen, t h e amount of argon in t h a t portion which does not pass through t h e porous walls. This has been proved by its increase in density. TIivd.-As the solubility of argon in water is relatively high, it is to be expected t h a t t h e density of t h e mixture of argon and nitrogen, pumped o u t of water along with oxygen, should, after t h e removal of the oxygen, be higher than t h a t of “ atmospheric ” nitrogen. Experiuient has shown that t h e density is considerably increased. Fourth.-It is in t h e highest degree improbable t h a t t w a processes, so different from each other, should manufacture the same prodnct. The explanation is simple if it be granted t h a t these processes merely eliniinate nitrogen from an “atniospheric” mixture. Moreover, as argon is an element, o r a mixture of elenleiits, its manufacture would meau its separation from one of t h e substances employed. T h e gas which can be removed from red-hot magnesium in a vacuum has been found to be wholly hydrogen. Nitrogen from chemical sources has been practically all absorbed by magnesium, and also when sparked in presence of oxyg e n ; hence, argon can not have resulted from the decomposition of nitrogen. That it is not produced from oxygen is sufficiently borne o u t by its preparation by means of magnesium. Other arguments could be adduced, but t h e above are sufficient to justify the conclusion t h a t argon is present in t h e atmosphere. T h e identity of t h e leading lines in the spectrum, the similar solubility and t h e similar density, appear to prove the identity of t h e argon prepared by both processes. Argon is an element, or a mixture of elements, for Clausius has shouii t h a t if K be the energy of translatorg motion of t h e molecules of a gas, and H their whole kinetic energy, then.K .. . - 3(cP=cv)
H ZCV Cp and CY denoting as usual the specific heat a t constant pressure and at
228
NOTES.
constant volume respectively. Hence if, as for mercury vapor ant1 for argon, the ratio of specific heats Cp : C u be 14, i t follows that K ;= 1 % . or t h a t the whole kinetic energy of the gas is accouiiteti for 1)y t h e translatory motion of its molecules. I n the case of mercury, t h e absence of interatomic energy is regaxled as proof of the nioiiato~iiic character of the vapor. and t h e conclusion holds equally good for argon. T h e only alternative is t o suppose that if argon molecules are di- or polyatomic, the atonis acquire n o relative motion, even of rotatiou, :1 conclusion exeeedingly improhahle in itself. arid one postulating t h e sphericity of such complex groups of atonis. N o w a monatomic gas can lie only an element, or a mixture of elements ; ant1 hence, it follows that argon is not of a cortipountl nature. From Avogadro’s law, the dcrisity of a gas is half its riiolecular weight ; and as t h e density of argon is approximately twenty. hence, its molecular weight rnust be forty. Rut its molecule is itlentical with its atoni : heiice, its atomic weight, or, i f i t be a mixture, the mean of t h e atomic weights of that mixture, taken for t h e proportion in which they are preseilt, rriust be forty. There is evidence both for and against the hypothesis that argon is a niixture ; for, owing to Mr Crooke’s ohservatious of t h e dual character of its spectrum ; against, because of Professor Olszewski’s statement that i t has a tlefiiiite melting-point, a definite boiling-point, and a definite critical temperature and pressure ; and because, on compressing t h e gas i n presence of its liquid, pressure remains sensibly constant until all gas has condeiiscd to liquid. The latter experiments are t h e well-knou-n criteria of a pure substance ; the former is not known with certainty to 1 x characteristic of a mixture. The conclusions which follow are, however, so startling that in our future experiinental work we shall entleavor to decitle t h e question by other means. For the present, however, t h e balauce of evidence seeins to point t o simplicity-. U‘e have. therefore, t o discuss the relations to other eleirieuts of an element of atomic weight forty. %‘e inclined for long to t h e view t h a t argon was possibly one or more than oiie of the elements which niiKlit be expected to follow fluorine in the periodic classification of the eleriieiits-elements which should have an atomic weight between niueteen, t h a t of fluorine, and twenty-three, t h a t of sodium. But this view is completely put out of court by t h e discovery of the xnonatoniic nature of its molecules. The series of elements possessing atomic weights near forty are35.5 Potassium. ..... .39.1 Calciuin.. ............................... 40.0 44.0 Scalldiurrl ................................ There can be no doubt t h a t potassium, calcium, and scandiurn follow legitimately their predecessors in the vertical columus, lithium, beryllium,
NOTES.
229
and boron, and t h a t they are in almost certain relation with rubidium, strontium, and ( b u t not so certainly) yttrium. If argon be a single element, then there is reason to doubt whether t h e periodic classification of t h e elements is complete; whether, in fact, elementsmaynot exist which can not be fitted among those of which it is composed. On the other hand, if argon be a mixture of two elements, they might find place in t h e eighth group, one after chlorine and one after bromine. Assuming thirty-seven ( t h e approximate mean between t h e atomic weights of chlorine and potassium) to be t h e atomic weight of t h e lighter element, and forty t h e mean atoniic weight found, and supposing t h a t t h e second element has an atomic weight between those of bromine, eighty, and rubidium, 85.5 ; v i z . , eighty-two, t h e mixture should consist of 93.3 per cent. of t h e lighter, and six and seven-tenths per cent. of t h e heavier element. But it appears improbable t h a t such a high percentage as six and seventenths of heavier element should have escaped detection during liquefaction. If it be supposed t h a t argon belongs to t h e eighth group, then its properties would fit fairly well with what might be anticipated. For t h e series, which contains-
Si,
I V . , py1. and V.
, SAI;;;
VI,
>
and C1,I. to VII.
I
might be expected to end with an element of monatomic molecules of no valency, i. e . , incapable of forming a compound, or if forming one, being a n octad ; and it would form a possible transition to potassium, with its monovalence, on t h e other hand. Such conceptions are, however, of a speculative nature; yet they may be, perhaps, excused, if they, in any way lead to experiments which tend to throw more light on t h e anomalies of this curious element. In conclusion, it need excite no astonishment t h a t argon is so indifferent to reagents. For mercury, although a monatomic element, forms compountls which are by no nieans stable at a high temperature in the gaseous state ; and attempts to produce compounds of argon may be likened to attempts to cause combination between mercury gas a t bo and other elements. As for t h e physical condition of argon, that of a gas, we possess no knowledge why carbon, with its low atoniic weight, should he a solid, while nitrogen is a gas, except in so far as we ascribe molecular complexity to t h e former and comparative molecular simplicity to t h e latter. Argon, with its comparatively low density and its molecular simplicity, might well be expected to r a n k among t h e gases. And its inertness, which has suggested its name, sufficiently explains why it has not previously been discovered as a constituent of conipouiid bodies. We would suggest for this element, assuming provisionally that it is not a mixture, t h e symbol A. We have to record our t h a n k s t o Messrs. Gordon, Kallas, and Matthews, who have materially assisted us in t h e prosecution of this research. Mr. Crookes (before reading his paper) said : Allow me, Mr. President, t o take this opportunity of striking t h e key-note of t h e chorus of applause
2\30
SOTES.
and congratulations which will follow from all cheniists present on this most valuable a n d important paper. T h e difficulties i n aresearch of this kind are peculiar. Here we have a new cheiiiical element, t h e principal properties of which seem to be t h e liegation of all chemical properties. Cheiiiists will understand h o w clifficult it is to deal wit11 aiiytliirig which foriiis 110 conipouncls ant1 uniteswith nothing. The discovery com~iieiice~l h y a prediction, followed after an interval b y realization. TXscoveries of this kind are more important and takv ii higher r a n k , than (Iiscoveries wliicli, more or less. come in a Iiaphazaril sort of \yay. The pretlictiori atid discovery of argon are only equalled Iiy t h e few iliscoveries of eleiiieiits which have been made i n cheiiiistry 1 ) t~h e careful stucly of tlie periodic law, and to surpass it we must go back to t h e prelictetl existelice arid subsequent discovery of an unknown planet l)y .%ciains ancl 1,everrier. 0)i f h c Spt>c-tix of A1.go17. By 1i.ilIiaiti Crookes. 1;. R . S., etc. (Alistract.) Through the kintltiess of Lord Kayleigh atid Professor Ranisay I have been enahled to examine t h e spectrum of this gas in a very accurate spectroscope. a i d also to take photographs of its spectra iu a spcctrograpli fitted with a coniplete quartz train. .lrgoii resembles nitrogen in t h a t it gives two distinct spectra, accorilin): t o t h e strength of the induction current ernployetl. Iiut while t h e two spectra of nitrogen are different in character, one showing fluted bands and t h e other s h a r p lines, t h e argon spectra hoth consist of s h a r p lilies. I t is, however, very difficult t o get argon so free from nitrogeu t h a t it \\ill tiot a t first show tlie nitrogen flutings superposed on its own special system of lines. I have usecl argon prepareti 1)y Lord Rayleigh, Professor Ranisay, arid myself, ant1 however free it was supposed to he from iiitrogeti, I could always detect t h e nitrogen hatids i n its spectriiiii. Tliesc, however, soon disappear w h e n the iiiiluction spark is passecl tliruugli the t u b e for some tiine, varying froin a few iiiitiutes to a few hours. T h e v a c u i i i i i tu1)es best adapted for shorting the spectra a r e of tlie ortlinary I'liicker form, having a capillary tube i n t h e rtiiiltlle. For photographing the higher rays which are c u t off hy glass I have used ii similar tube, " eiid on," having a quartz window a t one end. T h e pressure of argon giving t h e greatest luniinosity aiid most brilliant spectrutii is three iiini. At this poiut t h e color of t h e tlischargc is ail orange-retl, and t h e spectrum is rich in red rays, two heing especially prominent at wave-lengths 696.56 and 705.64. On passing t h e current t h e traces of nitrogen bands soon disappear, and t h e argon spectrum is seen i n a state of purity. At this pressure t h e platinum from t h e poles spatters over t h e glass of t h e bulbs, owing t o what I have called electrical evaporation,"' and I t h i n k t h e residual nitrogen is alm)rlxil b y the iiiiely-ilivitletl metal. Similar absorptions are freqtiently noticed b y those who work much with vacuum tuhes. If t h e pressure is further reduced, aut1 a Leyden jar intercalateil in t h e "
1 Kuy. Sor. Pvoc., I , 68, Jubc, 2891.
NOTES.
231
circuit, t h e color of the luminous discharge changes from red t o a rich steel blue, and t h e spectrum shows an almost entirely different set of lines. I t is not easy to obtain t h e blue color and spectrum entirely free from t h e red. The red is easily got by usiifg a large coil’ actuated with a current of three amperes aud six volts. There is then no tendency for it to burn blue. The hlue color may he obtained with t h e large coil hy actuating it with a current of 3.84 amperes and eleven volts, intercalating a jar of fifty square inches surface. The make-and-break must be screwed u p so as to vibrate as rapidly as possible. The red glow is produced by t h e positive spark, and t h e blue by t h e negative spark. I have taken photographs of the two spectra of argon partly superposed. I n t h i s way their dissimilarity is readily seen.* I n t h e spectrum of t h e blue glow I have counted 119 lines, and in that of t h e red glow eighty lines, niakirig 199 lines in a l l ; of these, twenty-six appear to be conimon to both spectra. I have said t h a t t h e residual nitrogen is removed by sparking t h e tube for some time when platinum terniihals are sealed in. This is not t h e o n l y way of purifying the argon. B y t h e kindness of Professor Ramsay I was allowed to take some vacuum tubes to his laboratory and there exhaust and fill them with some of his purest argon. On this occasion I simultaneously filled, exhausted, and sealed off two Plucker tubes. one h a r i n g platinutii and t h e other aluminum terminals. On testing t h e gas immediately after they were sealed off, each tube showed t h e argon spect r u m , contaminated by a trace of nitrogen bands. The next d a y t h e tube with platinum terminals was unchanged, but that having aluminum terminals showed t h e pure spectrum of argon, t h e faint nitrogen b:tntls having entirely disappeared during t h e night. After an hour’s sparkiiig and a few days’ rest the tube with platinum terniiiials likewise gave a pure argon spectrkm. \Vhen a mixture of argon with a very little nitrogen is sparked ill a tube made of pure fused quartz, without inside metallic terminals. the nitrogen hands do not disappear from t h e argon spect r u m , but the spectra of argou and nitrogen continue t o be seen siniultaneously. A vacuuiti tuhe was filled with pure argon and kept on t h e pump while observations were made on t h e spectrum of t h e gas as exhaustion proceeded. The large coil was used with a current of 8.84 amperes and eleveii volts; n o jar was interposed. At a pressure of three mill. t h e spectrum was that of t h e pure red glow. This persisted as t h e exhaustion rose, until a t a pressure of about half a ~nillimeterflashesof blue light made their appearance. At a quarter of a millimeter t h e color of t h e ignited gas was pure l)lue, and t h e spectrum showed no trace of t h e red glow. 1 ’The coil uked has ahorit sixty miles of sccoiidary wire, and when fully charged xives a torrent of sparks twenty-four inches long. T h e stnailer coil gives six-inch sparks when worked with six half-pint Grove cells. 2 Photographs of the different spectra of argon. and other gaseous spectra for ccinparisou, were projected on the screeu.
232
NOTES.
An experiment was now made to see if t h e small quantity of argon normally preseiit in t h e atmosphere c o u l ~ lie l detected without previous coiicentration. Nitrogen was prepared froin t h e atiiiosphere b y burning phosphorus, aiitl was purified in t h e usual manner. This gas, well clrieil over pliosphoric anhydride. was passed into a vacuum tube. t h e air waslied out 1)). t\vo fillings aiitl exh:iu.itioiis. ant1 he tube was filially se:iletl off a t a preqsure of fifty-two iniii. I t \vas useti for photogr:iplling tlie liaiitl spectrum of nitrogen on several occasions, aiitl altogether it was esposetl to t h e intiuctioii current from tlic large coil for eight h o u r s hcfore any cliaiige \vas noticed. The last time wheii photographing its spectrum tlifficulty \vas experienced in gettiiig the spark to pass, so I increased t h e curreiit ant1 iiitercalated a sinall jar. ?‘he color iiriineiliately cliaii~eilfroin Mie reddish yellow of iiitrogeii to t h e lilue of argoii, aiitl on applyin% t h e spectroscope t h e lilies of argon shone o u t with scarcely :iny admixture uf nitrogen bantls. IVith great tlifficulty aii(1 11) cniplo!:iiifi a v e r y small jar I \vas a1)le t o t a k e one photograph of its spectrtini and compare it with t h e spectruiii of argoii froin I’rofessol- Kaiii\ay, both heiiig taken oil t h e saitie plate, but t h e tube soon ljecaiiie iioii-conducting, and I coultl iiot then force a s p a r k tlirougli, except 1)y e m p l o y ing a tlaiigerousl). 1:trge ciirrent. \!’henever a flash passed it was of a (lee11liliie color. Assunling t h a t t h e atrnosplierc coiitaiiix oiie per cent. of argtiii, the three niui. of iiitrogeii origiiially i n t h e tulic \vould contairi 0.03 i i i i i i . of argon. After t h e nitrogcn liar1 lieen absorlietl 11y t h e spattereil platiiiuni, this pressure of argoii \ v o u l d be iiear tlie point of iioiicoiitluctioii. 111all cases wheii argon has been obtained i l l this manner t h e spectrum has lieen that of t h e blue-glowing gas. Very little of t h e retl rays can be been. The change from retl t o blue is chiefly dependent o i l t h e strength ant1 lieat of the spark : partly also on the Lllegree of exhaustion. I t is not iniprolialile, and I uiiderstaiitl t h a t inrlepeiitleiit observations h a v e alreatly led t h e discoverers to t h e satlie coi~clusioii-that t h e gas argon is not a siiii~ileI)ody. but is a mixture of a t least two eleiiieiits, one of tvhich g l o w s retl anti the other blue, each having its distinctive spectruiii. ‘1‘11~ tlicory that it is a simple lintly. has, however. support frorri tlic analogy of other gases. Tlius, iiitrogen has t\vo tlistiiict spcctr;i, one or tlie other k i n g pro(lucet1 1)y varying t h e pressure a i i r l inteiisity of t h e spark. I ?iavc iiiatle vacuuiii tubes containing rarefied iiitrogeu \vhich sliow either tlic fluterl Iiantl o r the sharp liiie spectiuiii by siiiiply turiiiiig t h e screw of the inal
