Abstract

A new method is developed for solving the wave equation for two-electron atoms. The wave function is expanded into a triple orthogonal set in three perimetric coordinates. From the wave equation one obtains an explicit recursion relation for the coefficients in the expansion, and the vanishing of the determinant of these coefficients provides the condition for the energy eigenvalues and for the eigenvectors. The determinant was solved on WEIZAC for $Z=1 \mathrm{to} 10$, using an iteration method. Since the elements of the determinant are integers, and only an average of about 20 per row are nonvanishing, it has been possible to go to an order of 214 before exceeding the capacity of the fast memory of WEIZAC. The nonrelativistic energy eigenvalues obtained for the ground state are lower than any previously published for all $Z$ from 1 to 10. In the case of helium, our nonrelativistic energy value is accurate to within 0.01 ${\mathrm{cm}}^{\ensuremath{-}1}$ and is 0.40 ${\mathrm{cm}}^{\ensuremath{-}1}$ lower than the value computed by Kinoshita. From the wave functions obtained, the mass-polarization and the relativistic corrections were evaluated for $Z=1 \mathrm{to} 10$. Using the values of the Lamb shift computed by Kabir, Salpeter, and Sucher, we obtain an ionization potential for helium of 198 310.67 ${\mathrm{cm}}^{\ensuremath{-}1}$ as against Herzberg's value of 198 ${310.8}_{2}$\ifmmode\pm\else\textpm\fi{}0.15 ${\mathrm{cm}}^{\ensuremath{-}1}$. Comparison is also made with the available experimental data for the other values of $Z$. By the use of our magnetic tape storage, the accuracy of the nonrelativistic energy value for helium could be pushed to about 0.001 ${\mathrm{cm}}^{\ensuremath{-}1}$, should future improvements in the experimental values and in the computed radiative corrections warrant it.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call