WAVE MECHANICS Quantisation as a Problem of Proper Values (Part I) (Annalen der Physik (4), vol. 79, 1926) { 1, Ix this pa I wish to consider, first, the simple case of the drogen atom non-relativistic and unperturbed), and show vena we customary quantum itions can replaced or in which tho notion of “whole numbere™* marely as suah ta bot antes, duced. Rather when integralness does appear, it arises in the same Peearal Tay, fe doce in. the cave af the node-nursbere of a vibrating string. new conception is cay generalisation, strikes, I believe, very deepl; at the true nature of the juantam rules, The usual form of the latter is connected wi the Hamilton-Jacobi differential equation, @ Hs, )-z A solution of this equation is sought such as can be represented as the sum of fonctions, each being « function of one only of the independent variables g. Here we now put for S a new unknown ¥ such that it will appear 88 9 product of related functions of the single co-ordinates, #.e. we put @) S=K log yp. The constant K must be introduced from considerations of dimensions ; it has those of action. Hence we get a (5H) -2. Now we do not look for a solution of equation (1’), but proceed as follows. epee nebo the relativitio variation of mass, equation (1) can always be transformed so as to become a quadratic form (of ¢ ite Smt derivatives) equated to zero, (For one-electron problem 12 WAVE MECHANICS this holds even when mase-variation is not neglected, 8 function ¥, such thet for any arbitrary eae ct itt the ae of the said quadratic form, taken over the whole co-ordinate 7 is stationary, ¢ being everywhere real, single-valued, finite, and con tinuously differentiable up to the second order. The quantum conditions motion, and show that can be so chosen for all positive, but only for 8 disorete set of negative values of B. That is, the above variation Problem has s discrete and a continuous spectrum of proper values, continent energies of the hy sate oe Fa ot agreement K must have the value a © orbit, Form = . choice of co-ordinates in the formation of the variati tions being arbitrary, Tot un take rectangular Carteclene Thee (ty a) QQ Bale tyne; e=charge, m mass of an electron, r!=at +91 +21, Our variation problem then reads ™ tye 0) w-sffficsal + BoC) -Fla-2- we teers! being taken over all space. From this we find in the 0) B= fapapBh | ffacdy dcdul op 2M(2+2)g]-0. Therefore we must have, firstly, : (6) vi Sa(E +2 )y=0, and secondly, (6) faravtt-o. df is an element of the infinite closed surface over which the integral (It will turn out later that thie last condition requires supplement our problem by a to a to the behaviour of Sy at infinity, in to ensure the existence of the above-mentioned continuous spectrum of proper values. See later.) The solution of (5) can be effected, for evample, in polar co-ordinates, r, 4, f, if 9 be written as the product of three functions, each only of 1, of 8, or of g. The method is sufficiently well known. The function of the angles turns out to be » surface harmonic, and if that of + be called x, we get easily the differentia] equation, 2 Tam aware this formulation is not entirely unambiguous. QUANTISATION AND PROPER VALUES—I 8 2 K Ime _nv(n+1) ” i re a xn 0. n=0,1,2,3... The limitation of n to integral values is necessary so that the surface harmonic may be sing We require solutions of (7) that will remain finite for non-negative real values of r. Now? equation (7) has two si ities in complex r-plane, at r=0 and aoe ee a it aotinite Pom’ (@bentealsingulacty) of all integrals, but the first on the contrary is not (for any integral). ‘These two singularities form exactly the bounding points of our real eines a on eee era mor tat ba tae i x at ing points is equivalent to a ‘condition. The equation has in general no sntegral which remains finite at both end pointa; such an integral exists only for certain special values of the constants in the equation. It is now a question defining these special values. This is the jumping-off point of the whole investigation.* Let us examine first the singularity at r—0. The so-called indicial equation which defines the behaviour of the integral at this point, is . & pip ~1)+2p—n(n +1) =0, with roote 6) Pmt, pam —(n +1). The two canonical integrals at this point have therefore the ex- ponents » and —(# +1). Since 1 is not negative, only the first of these ip of use to us. Since it belongs to the greater exponent, it can be re- presented by an ordinary power series, which begins with 1". (The other integral, which does not interest us, can contain a logarithm, since the difference between the indices is an integer .) The next singularit gue ity is at infinity, so above power series is always convergent re] xh indoor i We therefore ba sents a tranecendental integral function, ve established. it The required solution é: constant factor) a single-valued define Te dota taped Lee eT Liam Sonne to the n. ‘We must now investigate the behaviour of this function at infinity on the positive real axis. To that end we simplify equation (7) by the substitution (9) x=, where a is so chosen that the term with 1/r* drops out. It is easy to verify that then « must have one of the two values », —(# +1). Equation (7) then takes the form, + Bor gudanoo ja the treatment of (7) I ono thanks to Herman Wert iat Beusitons (Oelscion Betcbert, No. 13, Goecheo, 1600, expedally chapters 3 snd 0).4 WAVE MECHANICS , 2U a+) aU 2m ") ae (+2 )0-0. Its integrals belong at r=0 to the its 0 and ~2a- the a-value, a =n, the first of these integrals, and for the soos wake a= -(n+1), the second of {hoes integrals in an integral function and leads, according to (9), to the desired solution, which is single-valued. (10) ann. Our solution U then, at r=0, belongs to the expon i (7’) is called Laplace’s equation. The general type is ent 0. Rquation (7") u +(8+2)o" +(a+2)0 =o. Here the constants have the values (1) Bon 0, 5.20041), Gyn 2BF, =e ‘This type of equation is comparatively si to handle is reason : The so-called ‘La lace’s trautformation, ¥ ich in janersd tee in sic A ea ee oe ft lutions to i The result only is given here. The atemal DY comples integrals. (12) = fee ee -ayn- tae L is a solution of (7*) for a path of integration L, for which as) [ Zlete—cyre-arieeno, The constants ¢,, cy a}, a, have the followi are the roots of the quadratic equation onowing values. ¢ and 0, (14) A+8y +e4=0, and (14) a, = 28 atdiqy, a ro In the case of equation (7') these become, using (11) and (10), 08) emt JO, an [aan me* me* "ae th 8" Haag tt). + + ven thePamnantation bythe integral (12) allows us ot onl) survey the asymptotic behaviour of the totality of solutions wien 2 Ct. Schlesinger, The theory is due to H. Poincaré and J, Horn, QUANTISATION AND PROPER VALUES—I 5 tends to infinity in a definite way, but also to give an account of this behaviour for one definite solution, which is always a much more difficult task. We shall at firet exclude the case where a, and a, are real integers. When this occurs, it occurs for both quantities simultaneously, and when, and only when, OO) met ne + Therefore we assume that (15) is not fulfilled. The behaviour of the totality of solutions when r tends to infinit in a definite manner—we think always of r becoming infinite throug] real positive values—is chi ised + by the behaviour of the two Uneasy independent solutions, which we will call U, and U,, and which are obtained by the following specialisations of the path of integration ZL. In case let z come from infinity and return there along the same path, in such a direction that (18) lim e* =0, po te. the real part of er is to become negative and infinite. In this way condition (13) is satisfied. In the one case let z make a circuit once round the Rot o, (solution U,), and in the other, round c, (solution U,). flow for very large real positive values of r, these two solutions are represented asymptotically (in the sense used by Poincaré) by an (oererr4 = Tse NT (aNer 4), Dy~ erro —1)r{etees— 1)T (a4) 04—23)"-3, in which we are content to take the first term of the asymptotic series of integral negative [Link] 7, We have now to distinguish between the two cases. 1. H>O, This guarantees the non-fulfilment of (15), as it makes the left hand a imaginary. Further, by (14*), o; and ¢, also become pure ii innries. ‘The exponential fanstions in (IT), etaee’r ws coal, are the re periodic functions which remain finite. The values of and a, from (14"} show that both U, and U, tend to zero like r-"-1. This must therefore be valéd for our ranatendental integral eolution U, whose behaviour we are investigating, however it may be linearly from U, and Uy. Purthes (oF and (10) show that the function x, i.e. the transcendental integral solution of the original equation (7), ware tends to zero like 1/r, as it arises from U maultiplication by ” Whe fine ites ed haa for every positive E, solutions, which are everywhere sii Sfianite, Sere aan ahhd wiih tnd No nore oak he etc nh it tinual oscillations. The surface condition (6) has yet to be discussed. x |, at least one two paths of integration described in the vn Sa ae tr et tin6 WAVE MECHANICS 2. B<0. In this case the possibility (15) is not eo ij excluded, yet we will maintain that exclusion Provisional. Then Py de") and (17), for ¢-> c ,U, grows beyond all limits, but '2 Vanishes exponentially. Our integral function D (and the same is trap for 9) vill Ren ey finite if, and ly if, U is identical with U,, save perhaps for a numerical factor. This, however, oan never be, as is proved thus: If a closed cirouit round both points c, and ¢, be chosen for the path L, thereby satiafying condition (13) since the circuit is closed on the Riemann surface of the integrand, on account of a, +a, bei an integer, then it is easy to show that the integral (12) re ts our is i U. (12) can be developed in a series of iti i converges, at all events, for r sufficiently small, and since it satisfies equation (7'}, it must coincide with the series for U. Therefore U is Hpronmnted y (12) if L be a closed cirouit round both points ¢, and cy. is closed circuit can be so distorted, however, as to make it a additively combined from the two paths, considered above, which belonged to U, and U,; and the factors are non-vanishing, 1 and ozp Therefore U cannot coincide with U,, but must contain also U;. Our integral function U, which alone of the solutions of (7) is considered for our blem, is therefore not finite for r large, on the above hypothesis, ing meanwhile the question of completeness, te. the ing that our treatment allows us to find all the linearh ind it solutions of the problem, then we may state : ‘or negative values of which do not satisfy condition (15) our variation problem has no solution. We have now only to investi te that discrete set of negative E-values which satisfy condition (15). , and a, are then both integers, The first of the integration paths, which previo gave us the funda- mental values U, and U,, must now. undoubtedly be modified s0 as to give & non-vanishing result, For, since a, ~1 is certainly positive, the point ¢, is neither a point nor a pole of the integrand, but an ordinary zero, The point c, can also become regular if a, — I is also not ive. In every case, however, two suitable Paths are readily found the integration effected completely in terms of known functions, 80 thet the behaviour of the solutions can be fully investigated. a me* . (18°) KJ waae IH 3,4.., Then from (14") we have a4’) @-le=lt+n, a,-l=—-l+n, ‘Two cases have to be distinguished: Len and i>n. (a) ten. Then ¢, and c, lose every singular character, but instead become starting-points or end-poit of the path of integration, in order to fulfil condition (18), A third characteristic point here is at infinity (negative and real). Every path between two of these three points yields a solution, and of theae three solutions there are two linearly in- QUANTISATION AND PROPER VALUES—I 7 lependent, as is easily confirmed if the integrals are calculated out. In Seen fe gonecodentl integral solution is given by the path from 0, to og. That shis integral remains regular at r=0 can be seen at once without calculating it. I emphasize this ib, 88 the actual calculation is apt to obscure it. However, the ulation does show that 16 integral becomes indefinitely great for positive, infinitely t values of r, One of the other two integrals remains finite for r , but it becomes infinite for r=0. . Therefore when 1n, Then from (14), o, is a zero and c.a of the firat order atleast of the integrand. Two independent int aro then obtained one from the path which leads from z= ~ co to the zero, intentionally avoiding the pole; and the other from the residue at the pole. The latter is the integral function. We will give its calculated palue, but multiplied by r®, so that we obtain, according to (9) and (10), the solution y of the original equation (7). (The multiplying constant is arbitrary.) We find =m Fea — 22/1 (18) xm): feymaren 3 FG na -2) i t this is a solution that can be utilised, since it remains nite forall weal non-negative values of r. In addition, it satisfies the surface condition (6) because ot its vanishing exponentially at infinity. lecting then the results for # negative : , , Oona negative, our variation problem has solutions if, and only ¢, E satisfies condition (15). cole oat) cen be gion to She inter at least one such at our di oun be given to the integer n, ich denotes the order of the surface harmonic a; ‘in the equation. The Of the solution ing on r is given by (18). ‘aking into account the constants in the surface harmonic (known to be 2n+ Lin number), it i further found that 2 for any The discovered solution has exactly arbitrary constants permizeible ible (n, 1) combination ; and therefore for a presoribed value of t P orbit We have thus confirmed the main points of the statements originally made about the proper-value spectram of our variation problem, but there lencies. ae aire information as to the completeness of the collected ystem of proper functions indicated above, but I will not concern myself with that in this paper. From experience of cases, it may be supposed that no value has escaped us. Secondly, it must be rememl that the proper functions, ascertained for EF positive, do not solve the variation problem, ¢ originally tod, because they only tend to zero at ji ity as 1/r, an Dhetaline Oper only tends tp 2exo on an iafinite ephere aa Ife Hence the surface integral (6) is still of the same order as dy at infinity. If it is desired therefore to obtain the continuous spectrum, another condition must be added to the problem, viz. that 5) is to vanish at8 WAVE MECHANICS infinity, or at least, that it tends to = constant i di . s = . fini t val ind fi has is onuse i eat zi te lta ce . i § 2. Condition (15) yields (19) ~ Bn ey Therefore the well-known Bobr A i Baler tay, We cea ay Seen orepending to the fr roar of ta the Soarant Kntodued into (2) (20) x-2, from which comes * (i9’) - Hate , Our I is the principal quantum number, +1 is analogous azimuthal juantum number, The splitting up of this numhes though 8 closer definition of the surface harmonic i resolution of the azimuthal quantum int : ‘polar ” quantum. These numbers here define system thee £ the of node- Gives the number of the “node-spheres”, for it i eetablished Hh Sanction J) in (18) haa exaciy I~» =I potivg the hyperbolic orbits, to which gne may abe a a oerain emtam,of radial fact i : It is interesting to note that the inside whic i of (18) differ sensi iy from sero, and ouietie chen Tae seg anes suo he Mee ane the general order of magnitude of the major axis of the ipse in case, The factor, multiplied by which the radius Vector enters as the i of the . Fs naturally—the reci of a length, end ent tree, function Sis— (21) Ky V~3mB ~ met Taimat "1 Where a;= the semi-axis of the Ith elliptic orbit, (The equations follow from (19) pls the inown relation B= 5 . The quantity (21) gives the onder of i roots when and ® are small; for then Sara ber gots sas, of Zeota of (2) are of the order of unity. That is naturally no longer 2 are large numbers. At present REE uantum number a a. SO . corresponding to this is the proceeding to infinity, under condinual oscillations, of the functions iz : QUANTISATION AND PROPER VALUES—i 9 §3. It is, of course, strongly suggested that we should to connect the function y with some vibrefion process in the atom, whi would more nearly a; reality than the electronic orbits, the real existence of which is being very much questioned to-day. I originally intended to found the new quantum conditions in this more intuitive manner, but finally gave them the above neutral mathematical form, because it brings more clearly to light what is really essential. The essential thing seama to me to be, that the tion of “ whole numbers” no enters into the quantum rules mysteriously, but that we have, traced the matter.» step further back, and found the “ integralness ” to have its origin in the fini and single-valuedness of a certain space function. _, 1 do not wish to discuss further the possible representations of the vibration. before more complicated cases have been calculated toncofliy om the new stand pint Tt is not decided that the results will merely re-eoho those of the usual quantum theory. For gzample, if the relativistio Kepler problem be worked out, it s found to in » remarkable manner to Aalf-integral partial quanta (radial and azimuthal). Still, a few remarks on the representation of the vibration may be permitted. Eee eg a penetra er a iberations in place suggestive papers . Louis de Broglie,! and by reflecting over the space distribution of those “ phase Pace of eh he ba chon eked ihe always a whole wennber Hpmarared along the path, present on sac period or ques par of greasive waves, while we are led to stationary vibrations if we interpret our formulae as rey ting vibrations. I have lately shown * that the Einstein gas can be based on the considera- tion of such stationary proper vibrations, to which the dispersion law of de Broglie’s phase waves has been applied. The above reflections on the atom could have been represented as a generalisation from those on the gas model. Té we take the separate functions (18), multiplied by = surface harmonic of order n, as the description of proper vibration pro- cesses, then the quantity-F must have so: ing to do with the the purctatee (arty valod 3) being ‘proportional tothe square “ parameter ” ly ) bei portic to the square of the freq: . “However, in the fir place, such # statement in our case would to imaginary frequencio asncies for the negative K-values, and, secondly, instinct leads us to believe that the energy must be Proportional to the frequency itaslf and not to ite square, contradiction is explained thus. There has been no natural zero level laid down for the “‘ parameter” # of the variation equation (6), especially as the unknown function ¢ appears multiplied by a function of r, which can be changed by a constant to meet a corresponding Ihde Asn, de PI 1 . 22, . 19%. ‘Broglic, Pe el) Pie tose, Tem Poin )10 WAVE MECHANICS n y how eoy grt compaced wih a Seren r very com with all the admissibl tive Be (which are already limited by (i5)). Then fiwtly, dhe fequensiee will become real, and second! , since our B-values correspond to only relatively amall equaney diftrne, they will actually be very approxi- mately Proportional to tl eae frequency aiferenoea. This, again, is all vis nd xed can require, as long as the zero level of e view that the frequency of the vibration Pproceas is given by 2) y-OVOTE O04 oR +... where C is a constant very great compared with all the E’ has still another very appreciable advantage. It permits an understanding i the Bohr frequency condition. 0A a to the latter the ome x e ferences, and therefore from (22) also to the d ces of the proper frequencies » of those hypothetical vibration processes. But these Proper frequencies are all very great compared with the emission frequencies, and ey agree very _ closely ornong themselves, The emission frequencies a) therefore aa deep rence tones” of the proper vibrations ves. It is quite conceivable that on the transition of from one to another of the normal vibrations, something—I mean tho ight wave— with a frequency allied to each frequency difference, shout appearance. One only needs to imagine that the light wave is causally sted to the beats, which necessarily arise at each int of inne the transition ; ‘and that the frequency of the fight is defined secon i i cata per the intensity maximum of the ayia may be cbjected that rire conclusions are based on the relation f ‘orm (after expansion of the square root), from : whieh the Bol Sequency condition iteelf dooms to obtain the ‘ore of an approximation. » however, is merely apparently so, and it is wholly avoided when the relativistic theory is developed and makes & profounder insight possible. The large constant C is naturally very intimately connected with the rest-energy of the electron (me). Also the seemingly new and independent introduction of the constant 4 (already brought in by (20), into the frequency condition, is cleared tp. or rather avoided, by the relativistic theory. But unfortunately he correct establishment of the latter meets Tight away with certain didicalticn, which have been already alluded to. . is 'Y necessary to emphasize how much more congenial it would be to imagine that at a quantum transition the changes over from one form of vibration to another, than to thi * lated. Multi QUANTISATION AND PROPER VALUES—I u of a jumping electron. The changing of the vibration form can take place continuously in space and time, and it can readily last as Jong as the emiasion process lasts empirically (experiments on canal rays by W. Wien); nevertheless, if during this transition the atom is placed for a comparatively short time in an electric field which alters the frequencies, then the beat frequencies are immediately sympathetically, and for just as long as the field operates. Tt is known that this ‘imentally established fact has hitherto presented the greatest diffonlties, See the well-known attempt at a Oe forget, how: 1 aT tifcat progress us not , rever, in our ification over our in these matters, that the idea of only one proper vibration being excited whenever the atom doce not radiateif we must hold fast to this idea—is very far removed from the natural picture of a vibrating system. We know that a macroscopic system does not behave like that, but yields in & pot-pourri of its proper vibrations. But we should nob m © up our minds {00 icky on this point. A pot-pourri vibrations wor permisai T & single atom, anes ee there no beat frequencies could arise other than those which, ing to lence, the atom is ca) emitting i . The actual sending out of many of spectral lines simultaneously by the same atom does not contradict experience. It is thus conceivable that only in the normal state (and approximately in certain “‘ meta-stable ” states) the atom vibrates with one proper frequency and just for this reason does not radiate, namely, because no beats arise. The stimulation may consist of a simultaneous excitation of one or of several other proper frequencies, whereby beats originate and evoke emission of light. Under all circumstances, I believe, the woe functions, which belong to the same frequency, are in ‘al all simultaneously stimu- the proper values corresponds, namely, in the language of the previous theory to degeneration. To the reduction of the quantisation of degenerate systems probably corresponds the arbitrary partition of the energy among the functions ging to one proper value. Addition at the proof correction on 28.2,1926. Tn the case of conservative 5 in classical mechanics, the variation problem can be formulated in s neater way than was previously shown, and without express reference to the Hamilton-Jacobi differ- ential equation. Thus, let 7 (g, p) be the kinetic energy, expressed as a function of the co-ordinates and momenta, V the potential energy, and dr the volume element of the space, “ measured rationally ”, tc. it is not simply the product dg, dq, dq, .. . dga, but this divided by the square root of the discrimmant of the muadratic form 7 (a, 2). (CE. Gibbs’ Statistical Mechanics.) Then let ~ be such as to “Hamilton integral ”12 WAVE MECHANICS (23) for Rr(a, %) +7} stationary, while fulfilling the normalising, accessory condition 4) [¥irn. It is to be remarked that in the quantity a, of (14") we have essentially the well-known Sommerfeld expression ~yatve (CE. Atombau, 4th (German) od., p. 775.) Phyrical Institute of the University of Ziirich. (Received January 27, 1926.)