High-Prequeney Spectra of the Elements. 108 LXXX. The High-Frequency Spectra of the Elements, Part Tl, By Ht G., Mostuey, MA” 'HE first part + of this paper dealt with a method of "TD photogcdpting Kray ofotun se incad the spots ofa dozen elements. More than thirty other elemonta have now been investigated, and simple laws have been found which govern the results, and make it possible to predict ‘with confidence the position of the principal linos in the spectrum of any element from aluminium to gold. ‘The resent contribution is @ general preliminary survey, which Elnime neither to be completa nor vory acourta, ‘A somewhat differont method of photographing these spectra hs been developed independently by do Broglie? and by Hervog §. The latter closely confirms the angles given by Moseley and Darwin | for reflexion of Pt rays from selenite, De Broglie Gnds less satisfactory agreement. for the reflexion from rocksalt. De Broglie has also examined tho spectra of Wand Au, and has obiained for Cu and Fe results similar to those given in Part I. ‘The general experimental method has remained unaltered, and need not be again described. ‘The same crystal of assium Ferrocyunide bus been used as analyser throughout. Tho sharpness ‘of the lines of short wavelongel ‘has '* Communicated by the Author. 4 Motley, PL Mig avi. 1004 (1013) 17 Nov, 22 Dec, 1818, 10 Jen, 2 Feb, 2 March, wld, § Herveg, Verh. dD. Phys, Ges. xvi. p. 78, Jan. 1914, il Moseley & Darwin, Phil Mag. xavip. 10 (1013). BAL 19196 704 Mr. H. G. J. Moseley on the been much improved by reducing the breadth of the de- fining slit to about 0-2 mm. The ‘most convenient type of X-ray tube is drawn to scale in fig. 1. Tho aluminium trolley which carries tho targets oan be drawn to and fro by means of silk fishing-line wound on brass bobbin: An iron screen 8 fastened to the rails is furnished with a fine vertical sl ich defies the Xoray beam, Tho sit should actly opposite the foous-spot of the eathode-stream, Ahongh a slight rer can be remedicd by defleotng the xtbods rays with a-magnet. ‘The X rays escape by a side-tube 2} om. diameter closed by an aluminium window 0-022 mm. thick.’ The X-ray tube, which has a capacity of over 3 litres, was exhausted with a Gaede mereury-pump, for the loan of which Tam indebted to Balliol College. Fig. The radiations of long wave-length cannot penetrate an alominium window or more than a centimetre of two of ir. ‘The photographs bad therefore in this case to be taken inside an exhausted spectrometer. Fig. 2 gives a vertical section to scale of the X-ray tube and spectrometer. The former consists of a bulb containing the cathode, joined by a very large glass T-piece to a long tube of 4 eni. diameter, in which are the rails R and the carriage C.” 8 is the defining-alit and W a window of goldbeaters’ skin which separates the tube from the spectrometer. This material, which is usually air-tight, though sometimes it may require varnishing, is extremely transparent to X rays. A‘circular window of 2 cm. diameter will casily with. stand the pressure of the atmosphere if left undisturbed, Tn these experiments, however, the pressure was relievedHigh-Frequency Spectra of the Elements. 705 time the spectrometer was exhausted, and under such conditions the goldbeaters’ skin had frequently to be renewed. The spectrometer, which was specially id for this work, consists of a strong cirealat iron box o inside diameter and 8 em. high, closed by a lid which, when the flange in greased, makes an air-tight joint.” Two concentric grooves are cut in the floor of the box. The table A, which carries the plato-holder, rests on three steel balls, of which two run in the outer groove, while the third rests on the floor of the box. The position of the erysial- table B is controlled in like manner by the inner groove, ‘This geometrical construction for a spectrometer is well Fig. 2. Known, The scales are fixed to the box and the verniors to the tables. For these very soft rays the absorption by the black paper front of the plate-holder became serious, and ‘two sheots of black tissue-paper wore used instead. Lumps ‘of the pure elements, usually several millimetres thick, were ‘used as targets in ‘the caso of Mg, Al, Si, Mo, Ru, Pd, Ag, Sb, Ts, Foils such as Rh, W, Au wore oithor silver: tolled or brazed onto copper” 05 was uted inthe forms of a thin chemical deposit on copper. The alloys used wero ZxNi (10 per cent.), WFe (50 per cent.), NbTa (50 per cent.), and SoMn (50 per cent,). KCI and the oxides of the ‘rare-earth elements were rubbed onto the surface of 197706 Mr. H. G. J. Moseley on the nickel plates roughened with coarse emery-powder. The only serious difficulty in the experiments is caused by the heat produced by the cathode ray bombardment, and the consejuent liberation of gas and destruction ofthe surface of the target. This makes it necessary to uso the element in a Form which is not too volatile and prevents the employ nent of a very powerful discharge. The total time of an exposure, including rent, varied from three maiauts for a submane such as rathenium, which could safely be heated, to thirty ‘minutes for the rare earth oxides. ‘The importance of using ‘an efficient high-tension valve may again be mentioned. ‘The oxides of Sa, Bu, Gd, Er were given me by Sir Wil- liam Crookes, O.M., to whom I wish to express my sincere ratitude. For the loan of the Os and a button of Ru I am indebted to Messrs. Johnson Matthey. The alloys were obtained from the Metallic Compositions Ov., and the oxi of La, Ce, Pr, Nd, and Er from Dr. Schuchardt, of Gérlitz. Almost every line was photographed in two different orders, and the double angles of reflexion measured as before to within 0°1 and sometimes 0°05, In some sets of experi iments n small error caused by the crystal surface not bong exactly on the spectrometer-axis gave rise to a systematic Giscrepancy in the results obtained from reflexion in different orders, It was found that this error, which never changed the reflexion-angle by more than 0°05, could be measured: more accurately from the amount of the discrepancy than thy dr olearatn of ec, re ty rection was necaary ‘rhen lag. the Tong’ wave-ln apparatus. In this case the sii and photograph are Bot tuidstant from the crystal and the positon of tne spectra lines on the plate is no longer independent of the anglo at which the crystal is sot. ‘The necessary corrections were Caleulated geometiclly, and verifed by photographing the same line for both right-handed and IctUhended efeSions and with the crystal set at various angles. nthe work on tho very short wave-engths, the reflexion af tho genoral heterogenous radiation gave some trouble. This is‘always an important part of the radiation from an x tbe, but witha hard tube its anelysed by rellesion mainly into constituents of very short wave-length, and so ‘usually does not interfere with ‘the line-spectra. It is only with an extremely soft tube, combined with precautions against absorption by the air, that constituents. reflected at large angles become prominent. When examining such a spectrum as that of Ag in the K series, the general reflexion| High-Frequency Spectra of the Elements. 07 cannot be avoided. Unfortunately, when photographed it takes the form of irregular fringes, which effectually hide spectrumlines. A change of target, withthe postion of slitand crystal unaltered, does not aifect the appearance ‘of the fringes, a fact which proves that they are due to the general heterogenoous radiation, Tt is easy to show that the Fringes are merely a very foreshortened pattern of patches ‘on the erystal surface which reflect exceptionally well. The way in which they move and spread out laterally as the crystal, is turned. provides a proof of this, and 30 does Barkla’s* observation that when the erystal is moved side- ‘ays the fringes move with it. Tt is eaay to devise methods for getting rid of the fringes. In the first place, narrowing the slit or increasing the distance from the crystal will diminish their intensity compared. with that of the line-spectrum. In the second place, turning the crystal will move and blur the fringes, but leave the sharpness of tho lines unaffected Provided the slit and photograph are equidistant from the Feflecting surface f. "The quantitative measurements of Moseley ‘and Darwin t on tho reflexion of the general radiation must have been little affected by these fringes, as the incident. beam was restricted to a very narrow pencil which always impinged on the same part of the crystal. ‘The results obtained for radiations belonging to Barkla’s K series are given in Table I, and for convenience the figures already given in Part Iv are included. The wave Jength 2 has been calculated from the glancing angle of reflexion @ by means of the relation nA=2d sin8, where d has been taken to be 8454%10-* om. As before, the strongest lino is called « and the next lino 8. ‘The square root of the frequency of each line is plotted in fig. 3, and the wave-lengths can be read off with the help of the scale at the top of the diagram. ‘The spectrum of Al was photographed in the first order only.” ‘The very light cements give several othr fantor Jines, which have not yet been fully investigated, while the esults for Mg and Na are quite complicated, and apparently depart from the simple relations which connect the spectra of theother elements. In the spectra from yttrium onwards only the a line has so far been measured, and further results in these directions will be given in « later paper. ‘The { Burka and Mastys, rv, Phys. Soe; London (1819) 4 Money oe 1035" Seu also W. Hl a We Brag, Pre. ‘ay, Boe, fy Iexsvlp. 28 (013), ‘away abd Darv doa 199708 Mr. H. G. J. Moseley on the Taste TL. x. Anis SORBESSSRRSERERNESTES spectra both of K and of Ol wore obtained by means of a target of KCl, but it is very improbable that the observed ines have been attributed to th wrong elements, ‘The sine fr element fom Y onwards speared to consis of = very close donblet, an effect provionsly observed by Bragg * in the cate of thotiam, " The results obtained for the spectra of the L series are given in Table IT. and plotted in fig. 3. ‘These spectra con- tain five lines, « 8, 7, 8, ¢ reckoned in order of decreasing wave-length and decreasing intensity. ‘There is also always 2 faint companion «’ on the long’ wavelength side of a zathr faint line betwen and 1 for tho rare exeth elements at least, and a numbor of very faint lines of wavo- length greater than «. Of those, a, 6, ¢, and ry have been systematically measured with the object of finding out how the spectrum alters from one element to another. ‘The fact that often values are not given for all these lines merely. indicates the incompleteness of the work. ‘The spectra, 0 far as they bave been examined, are so entirely similar that without doubt «, 8, and ¢ at least alw Often was © Bragg, ‘Nature,’ March 29, 1914,sensi High-Frequency Spectra of the Elements. 709 ig. 8. WAVE LENGTH x lolom, puzasoonane: 7 ee BUREN eoRbereadonatospERnaunaae: 3 6 a SQUARE ROOT OF FREQUENCY x lo" 201| | 710 Mr. H. G. J. Moseley on the Tawee II. x. . Bins, | g tine, | ti S| atone | REG. | toe | Bee, Namber —| | we | 0 ae | 41 bes | 2 nian ger | a | Paladin wer | a6 2 o 1 a 8 50 @ | e | & 4 6 s 3 we a Fi n notincluded in the limited range of wave-length bbe photographed on one plate. Sometimes lines have not been measured, either on account of faintness or of the confusing proximity of lines due to impurities. Lines due to impurities were frequently present, but caused little trouble except in the rare earth group. Here two extreme cases occurred. ‘The X-ray spectrum of the rascodymia showed that it consisted roughly of 50 per cent. Yi'5p per cent. Ca, and 18 per cout. Pr.” Unfortanataly the position expected for the a line of Pr coincides with the known position of the 8 line of La, but the 8 line of Pr ‘was quite conspicuous, and had precisely the wave-length anticipated. Two specimens of erbia were used. ‘The speci- ‘men purchased contained 50 per cent, Er and 50 per cent. of another element, of which the X-ray, spectram coincides ‘with the spectrum caleulated for Ho. ‘The erbia given by Sir William Crookes was evidently nearly pure, but showed the @ and # lines of Ho quite faintly, and also faint lines agrecing with «and @ of Ds and a of TnI and of TmII. ‘The Na was free from La, Ce, and Pr, but contained fairas High-Frequency Spectra of the Blements. 12 roportion of Sm. The Sm, Eu, and Gd appeared to be Pure, I bope son to complete the examstion of the Spectra of this group. Conclusions, In fig. 3 the spectra of the elements are arranged on horizontal lines spaced at equal distances. The order chosen for the elements is the order of the atomic weights, except in the cases of A, Co, and Te, where this clashes with the order of the chemical properties.” Vacant lines have been left for an element between Mo and Ru, an element between Nd and Sa, and an element between W and Os, none of which are yet known, while Tm, which Welsbach has separated into two constituents, is given two lines. This is equivalent to assigning to successive clements a series of successive characteristic integers. On this principle the integer N for Al, the thirteenth element, has been taken to be 13, and the values of N then assumed by the other elements are given fon the left-hand sido of fig. 3. "This proceeding is justified by the fact that it introduces perfect regularity into the ray spectra. Examination of ig. sbows thal the valves of! for all the lines examined both in the K and the L series now fall on regular curves which upproximate to straight lines. ‘The game thing is shown more clearly by comparing. the values of N in Table I. with those of Q » dy » being the frequency of the « line and vp the fandamental Rydberg frequency. Tt is. hore plain that Qx very approximately, except for the radiations of very short wave- length which gradually diverge from thie relation. Again, in Table IT. a comparison of N with @ = Far where » is the frequeney of the Ts « line, shows. that Q, =N—T-4 approximately, although a systematic deviation clearly shows that the relation isnot accurately linear in this, ‘Now if either the elements were not characterized by these integers, or any mistake had been made in the order chosen or in the number of plages left for unknown elements, these » Welsbach, Monateh, xxx. p. 878 (1911) 203204 ne Mr. H. G. J. Moseley on the regularities would at once disappear. Wo, can, therefore conclude from the evidence of the X-ray spectra alone, with- out using any theory of atomic structure, that these integers aro really characteristic of the elements. Further, as itis improbable that two different stable elements should bave the same integer, three, and only three, more elements are likely to exist between Aland Au. As the X-ray spectra of those olements can be confidently predicted, they should not be diffeult to find. The examination of keltium * would bbe of exceptional interest, as no place has been assigned to this element. ‘Now Rutherfordt has proved that the most important con- stituent of an atom is its central positively charged nucleus, and van den Brock has put forward tho view that the charge carried by this nuclous is in all cases an integral multiple of the charge on the hydrogen nucleus. ‘There is every reason to suppose that the integer which controls the ‘X-ray spectrum is the same as the number of electrical units in the nucleus, and these experiments therefore give the strongest possible support to the hypothesisof van den Brock. Boddy § has pointed out that the-chemical, properties of radiovelemenis are strong evidenoo that this hypothesis la true for the elements from thallium to uranium, so that its general validity would now seem to be established. ‘From the approximate linear relation between »t and N for each line we obtain the general equation y= (N—0), where A and bare constants characteristic of each line. For the K a line A (S-3)m and oat. For the La line approximately Aa(b—2)n mat bo74 ‘The fact that the numbers and arrangement of the lines in the K and the L spectra are quite different, strongly suggests that they come from distinct vibrating systems, while the fact that b is much larger for the L lines than for the K lines 4 Urbain, Cop 241 QOD, + Rutherford, Phil flag. xe. p. Goo (1012), and xxvii. p 458 (1014). } Van den Brook, Bayes Ze vp. 82 (1018), and! Natur, Nov. 2, Doo. 25, 1018, March 6, 1014, ' Soddy, Jahrbuch Rad nd, lect x. p.198 (1918); ‘Natur Dec. et 18 (858).High-Frequency Spectra of the Elements. 118 suggests that tho I system is situated tho further from the nucleus. Tt was shown in Part I. * that the linear relation between # and N—5 was most naturally explained if the vibrating system was a ring of electrons rotating round the central uclens with an angular momentum which was the same for the different elements. ‘This view has been analysed and put in a more generalised form in a letter to “Nature's, written in answer to criticisms made by Lindemann 3. ‘Semmarys 1. Every element from aluminium to g by an integer N which determines its X-ray spectrum. Every deual in the speciram ofan eloment enn theres Fore be predicted from the spectra of its neighbours, 2 This integer N, the atomic number of the element, identified with tho number of positive units of electricity contained in the atomic nucleus. 3. The atomic nombers for all elements from Al to Au have been tabulated on the assumption that N for Al is 13. 4. Tho onder of tho tomo numbers isthe same as tat of the atomic weights, except where the latter disagrees ith the order of the chemical properties, Known elements correspond with all the numbers be- between 13 and 79 exoept three. There are here threo possible elements still undiscovered. 6. The frequency of any line in the X-ray spectrum is approximately proportional to A(N—8)*, where A and 6 aro constants. Id is charactorized wish to thank Prof. J. 8. Townsend, F.R.S., for providing amo with ovary facility for carrying on this work, which has been greatly assisted by @ grant from the Institut Inter. national de Physique Solvay. “Hlectrcal Laboratory, ‘Oxford. 1 Lae. cit. p. 1082. + Moseley, “Nature Jen. 16 (1914). 4} F.A. Lindemann, “Naturg, Jan i, Feb. 5, 1014 205