Early, Douglas Dale2024-11-122024-11-121964https://hdl.handle.net/10919/121605It has been known for some time that the experimentally determined energies of the excited states of donors in silicon agree favorably with the values predicted by the effective mass theory due to Kohn (3). However, the experimentally determined energies of the ground states in silicon differ from the predicted value by at least 3.5 mev (2). This has been attributed to the neglect of effects near the impurity ion. By making corrections to the potential for small distances from the impurity ion and using symmetry argument, it has been shown (4) that the ground state in reality splits into a non-degenerate level, is 1s (A₁), which is the ground state, and doubly and triply degenerate 1s levels, 1s(E) and 1s(T₁) respectively. Recent work in this area has shown evidence of the 1s state split in antimony-, arsenic-, and phosphorus-doped silicon (1). At liquid helium temperature the ground state is the only state significantly populated so that, by using infrared radiation, for example, transitions may be induced from the ground state to the excited states and continuum. Thus the energies of the excited states relative to the ground state may be determined. Our data shows evidence that the 1s state split occurs for bismuth-doped silicon with the upper 1s levels having energies of-31.4 mev and -36.5 mev relative to the bottom of the conduction band. This shows that the lowest-lying upper 1s level is separated from the ground state by 34.1 mev and the two upper 1s levels are separated by 5.1 mev. The latter agrees very favorably with the data of Aggarwal (1), however, we are unable to determine the ordering of these states from our data. That the former is larger than the values found Aggarwal is to be expected because of the larger donor ionization energy of bismuth in silicon.1 volume (various pagings)application/pdfenIn CopyrightLD5655.V855 1964.E274BismuthSemiconductorsSiliconThesis