Nobel prize to P. Kusch awarded in 1955 "for his precision determination of the magnetic moment of the electron''. Co-winner W. E. Lamb "for his discoveries concerning the fine structure of the hydrogen spectrum''
Foley, H.M.; Kusch, P.; On the Intrinsic Moment of the Electron
Phys. Rev. 73 (1948) 412;
Reprinted in Selected Papers on Quantum Electrodynamics, editor J. Schwinger, Dover Publications, Inc., New York (1958) 135.
In a previous letter1 we have reported the observation that the ratio of the gJ values of the 2P3/2 and 2P1/2 states of gallium has the value 2.00344; the value 2 for this ratio follows from Russell-Saunders Coupling and the conventional spin and orbital gyromagnetic ratios. If each of these states is exactly described by Russell-Saunders coupling, this observation can only be explained by setting (S
- 2L) = 0.00229 ± 0.00008, where the electron spin g value is gS = 2 + S, and the orbital momentum g value is gL = 1 + L. Since each of these atomic states may be separately subject to configuration interaction perturbations, the interpretation of this result was not entirely clear. A determination has now been made of the
ratio of the gJ values of Na in the 2S1/2 state and of Ga in the 2P1/2 state. The experimental procedure was similar to that previously described.1 The known hyperfine interaction constants of gallium2 and sodium3 were employed in the analysis of the data. We find for this ratio the value 3.00732 ± 0.00018 instead of the value 3. This result can be explained by making (S
- 2L) = 0.00244 ± 0.00006. The agreement between the values of (S - 2L) obtained by the two experiments makes it unlikely that one can account for the effect by perturbation of the states. The effect of configuration interaction on the gJ value of sodium is presumably negligible.4 To explain our observed effect without
modification of the conventional values of gS or gL introduces the rather unlikely requirement that both states of gallium be perturbed, and by amounts just great enough to give the agreement noted above. From any experiment in which the ratio of the gJ values of atomic states is determined, it is possible to determine only the quantity (S - 2L). If, on the
basis of the correspondence principle we set L equal to zero, we may state the result of our first experiment as gS = 2.00229 ± 0.00008 and that of our recent experiment as gS = 2.00244 ± 0.00006. It is not possible, at the present time, to state whether the apparent discrepancy between these values is real. It is conceivable that some small perturbation of the states would give rise to a discrepancy of the indicated magnitude.
These results are not in agreement with the rec ent suggestion by Breit6 as to the magnitude of the intrinsic moment of the electron.
scriptsize 1P. Kusch and H. M. Foley, Phys. Rev. 72 (1947) 1256. 2G. E. Becker and P. Kusch, to be published. 3S. Millman and P. Kusch. Phys. Rev. 58 (1940) 438. 4M. Phillips. Phys. Rev. 60 (1941) 100. 5Dr. J. Schwinger has very kindly informed
us in advance of publication of his conclusion from theoretical studies that L is zero whereas S may not vanish. 6G. Breit, Phys. Rev. 72 (1947) 984. .
Related references More (earlier) information appears in P. Kusch and H. M. Foley, Phys. Rev. 72 (1947) 1256;
See also G. Breit, Phys. Rev. 72 (1947) 984;
S. Millman and P. Kusch, Phys. Rev. 60 (1941) 100;
First measurement of g-2 for the electron.