Mathematical Proceedings of  the Cambridge Philosophical  Society http://journals.cambridge.org/PSP Additional services for Mathematical  Proceedings of the Cambridge  Philosophical Society: Email alerts: Click here Subscriptions: Click here Commercial reprints: Click here Terms of use : Click here The calculation of atomic fields L. H. Thomas Mathematical Proceedings of the Cambridge Philosophical Society / Volume 23 /  Issue 05 / January 1927, pp 542 ­ 548 DOI: 10.1017/S0305004100011683, Published online: 24 October 2008 Link to this article: http://journals.cambridge.org/ abstract_S0305004100011683 How to cite this article: L. H. Thomas (1927). The calculation of atomic fields. Mathematical  Proceedings of the Cambridge Philosophical Society, 23, pp 542­548  doi:10.1017/S0305004100011683 Request Permissions : Click here Downloaded from http://journals.cambridge.org/PSP, IP address: 128.42.202.150 on 05 Dec 2012 542 Mr Thomas, The calculation of atomic fields The calculation of atomic fields. By L. H. THOMAS, B.A., Trinity College. [Received 6 November, read 22 November 1926.] The theoretical calculation of observable atomic constants is often only possible if the effective electric field inside the atom is known. Some fields have been calculated to fit observed data* but tor many elements no «uch fields are available. In the following paper a method is given by which approximate fields can easily be determined for heavy atoms from theoretical considerations alone. 1. Assumptions and the deduction from them of an equation. The following assumptions are made. (1) Relativity corrections can be neglected. (2) In the atom there is an effective field given by potential V, depending only on the distance r from the nucleus, such that V-+0 as r-f, Vr^-E, the nuclear charge, as r-*-0. (3) Electrons are distributed uniformly in the six-dimensional phase space for the motion of an electron at the rate of two for each h* of (six) volume. (This means one for each unit cell in the phase space of translation and rotation of a spinning electron.) The part of the phase space containing electrons is limited to that for which the orbits are closed. (4) The potential V is itself determined by the nuclear charge and this distribution of electrons. In reality the effective field at any point depends on whether the point is empty or occupied by a foreign electron or one or another atomic electron and on the circumstances of that occupation. These fields can only be expected to be sensibly the same or approximately calculable from the above assumptions if the density of electrons is large, that is, in the interior of heavy atoms. If e, m, p are the charge, mass and momentum of an electron, the Hamiltonian function for the electronic motion is ((1) and (2) above), 2-f-eV. ' D. B. Hartree, Proc. Gamb. Phil. Soc, 21, p. 625; E. Fues, Zeit. ftir. Pkys., 11, p. 369. Mr Thomas, The calculation of atomic fields 543 There are ((3) above) electrons at two for each h? of phase space for which i.e. at per unit of ordinary (coordinate) space. Thus ((4) above) with ((2) above) V-*- 0 as r -•- oo, Vr ~*-E as r -*• 0. Now express distance in terms of the ' radius of the normal orbit of the hydrogen atom,' a = hi /4nts me* = 53.10"9 cms., potential in terms of the potential of an electron at this distance, so A3 and equation 1*1 becomes 8. with i^r -*• 0 as p -»• QO , p^r -•• i\T, the atomic number, as p -*• 0. (It is useful to note that with ' a' as unit of length, the charge and mass of the electron as units of charge and mass, A=2TT, whence 1*2 is at once verified.) The ' effective nuclear charge ' at distance p is then given by p Putting -^r = Ya= , the equation for <£ is 36—2 544 Mr Thomas, The calculation of atomic Jidda 2. Discussion of the equation. Write log p = x, pA — w, and the equation becomes d2 w _ dw o a or, if dwfdx—p rip= «,(«>*-12) The maximum and minimum locus of this equation is P = ~W(W 7 ~ 1 2 ) The inflexion locus is 2w Tw'—12) p= = y ("0> 7 + (1 + 6w*)' and gives the direction in which the solutions cross the inflexion locus. There are two singular points, ^ = 0 , ^ = 0 ; w= 144,p = 0. At w, at «; give the form of the solutions, c being arbitrary. The dp/div discriminant gives p = 0, and w = 144 or = 144p~ as a singular solution. There is an approximate particular solution, \ (2-21), 144 ( which satisfies 3 rip _ 12 3 w(wl-12) 5\ dw ~ 5 X p The solutions of 22 lie roughly as in the sketch (Fig. 1), the arrows give the direction of increase of p. The only solutions for which -*-0 as p -*• and =0 (l//>) as p -*• 0 correspond to the solution through 0 and A in the sketch—2'21 is an approximation to this solution*. Different values of the nuclear charge * It is only at A that p becomes infinite. Mr Thomas, The calculation of atomic fields 545 correspond to the replacement of x by x + c, which does not affect 2"1, so that if the equation is integrated numerically, starting from an initial position with w and p near A and any value of x, all the required solutions can be deduced. Pig. i 3. The numerical integration. For the initial values put in 2*1, obtaining w = 144 (u + 1)~4 , V—-J-, ax dv o' = \(M+au1 ), + M8 (a {- 7V - 7X,} + 2a2 \») - 3a*\*u* «J {35 V73 - 292 + a (134 - 14 V73)} + cm" {7 (4 V73 - 34) + a (61 - 7 V73)} - 3a2 \J u4 . 546 Mr Thomas, The calculation of atomic fields For X = - 1 (V73 - 7) = - -77200 and u > 0. G < 0 for a = - (35 V73 - 292)/(134 - 14 V73) = - 0027900, G > 0 for a = 0, from which it can be shown that for u > 0 - -772OO«< ^ < - -77200(M-0027900M") ...(31). For the actual numerical integration it is convenient to put x = X\oge10, w/=144.10y , so that 2'1 becomes -25-(3-5-g)} ...(3-2), while ^ = ^144.10r -4 <*+ c ) (3-31), (3-33), where c is to be determined from the atomic number. Z is the effective nuclear charge. If 7=-l, w = 1 0 i - 1 =-77828, and 1-3515 > ^ > 13489 (from 31). Starting with numerical integration was carried out by steps of '1 to X = — 3 by the aid of the formulae (dy\ (dy\ fdt y\ 1 . /d% u\ 5 Ao/cPv\ — I — I —-) = I —- I + - A I —— I H A2 1 —a I + dxjn+1 \da;Jn \da?Jn 2 Vcte3 /^, 12 V.ds'/n-B " " * See Whittaker and Bobinson, The Calcului of Observations, p. 866. Mr Thomas, The calculation of atomic fields 547 For X = — 3 it appears that 35 - ^ = 508, log10144 + F = 74385 so equation 3"32 gives Z = ^£.. 1-008. i.e. c =-7 since here closely enough Z = iV the atomic number. e.g. for N = 55 (caesium), c=1810. 4. Numerical results. The following table gives the values of 35 — -Ty and logi0144 + Y found by numerical integration and the corresponding values of p, Z, yfr for caesium. The former may be in error by about 10 in the last decimal place. For pQ < "006, the field is sensibly a Coulomb field. For p0 > l- 5, the approximate formula 2% 21 is an accurate enough solution of the differential equation, but this equation is not an accurate representation of the facts. For the element of atomic number N the corresponding values are given by /55\* [) \55/ The values Zx are (unpublished) values calculated by Mr Hartree for caesium from the observed levels and which he has very kindly allowed me to include for comparison. In conclusion, I wish to thank Professor Bohr and Professor Kramers for their encouragement when I was carrying out the numerical integration last March. 548 Mr Thomast The calculation of atomic fields -X 0 •1 •2 •3 •4 •5 •6 •7 •8 •9 1 0 1 1 1-2 1-3 1-4 1-5 1-6 1-7 1-8 1-9 2-0 2 1 2-2 2-3 2-4 2-5 2-6 2-7 2-8 2-9 3 0 ••«-£ dX 2150 2-015 1-880 1-746 1-615 1-489 1-371 1-261 1160 1-069 •987 •914 •851 •795 •747 •706 •671 •642 •614 •595 •577 •564 •552 •542 •534 •527 •521 •517 •513 •510 •508 logl0144+r L-1584 1-0167 •8614 •6927 •5105 •3156 1086 1-8901 1-6611 1-4225 T-1752 2-9202 2-6584 2-3906 21176 3-8402 35590 3-2746 4-9875 4-6979 4-4064 4-1134 58191 5-5238 5-2276 6-9306 6-6330 63349 6-0364 7-7376 7-4385 Po 1-517 1-205 •9572 •7603 •6040 •4800 •3811 •3027 •2404 •1910 •1517 •1205 •09572 07603 •06040 •04800 •03811 •03027 •02404 •01910 •01517 •01205 •009572 007603 •006040 •004800 •003811 •003027 •002404 •001910 •001517 76 10-4 13-7 175 21-6 25-8 301 342 38-0 41-3 442 47-7 48-7 50-3 51-5 52-5 53-2 53-8 54-0 54-4 54-6 54-8 54-9 54-9 550 55-0 55-0 55-0 55-0 550 55-0 1-887 3-412 6-008 10-23 1690 27-10 42-26 6418 9515 138-0 1961 2739 376-4 510-4 683-8 906-8 1198 1556 2018 2601 3340 4273 5450 6936 8809 11170 14140 17870 22580 28570 35960 Zi 9-9 12-5 16-0 19-7 24-3 29-0 33-4 36-6 39-5 42-2 44-7 46-6 47-8 48-4 49-3 50-6 51-6 52-4 53-4 53-9 54-1 54-4 54-6 54-7 54-8