High-resistivity Samples Research Articles

Nuclear Magnetic Resonance in Single Crystals of CdS

The ${\mathrm{Cd}}^{111}$ and the ${\mathrm{Cd}}^{113}$ NMR lines in CdS single crystals were studied at 4.2\ifmmode^\circ\else\textdegree\fi{}K. The spin-lattice relaxation time ${T}_{1}$ of high-resistivity samples is strongly affected by optical illumination. A typical value of ${T}_{1}$ for a sample cooled in the dark, 3\ifmmode\times\else\texttimes\fi{}${10}^{3}$ min, decreased to 30 min while the sample was being illuminated with white light. If the sample was kept at 4.2\ifmmode^\circ\else\textdegree\fi{}K, ${T}_{1}$ did not return to the original value but remained at about 4\ifmmode\times\else\texttimes\fi{}${10}^{2}$ min after the light was turned off. Bleaching with red light restored ${T}_{1}$ to 1.4\ifmmode\times\else\texttimes\fi{}${10}^{3}$ min. The lifetime ${T}_{1\ensuremath{\rho}}$ of the nuclear-spin polarization in the rotating frame was measured as a function of rf field strength ${H}_{1}$ under various illumination conditions. ${T}_{1\ensuremath{\rho}}$ decreases very rapidly when ${H}_{1}$ is decreased below 1 G. Light-created trapped paramagnetic centers are mainly responsible for these relaxation effects. In order to fit the data with nuclear-spin diffusion theory, it was necessary to consider at least three different types of trapped centers. The density of these centers under various illumination conditions was estimated. NMR signals of nuclei near the trapped centers were observed using a double-resonance technique in the rotary frame. A strong signal of ${\mathrm{S}}^{33}$ was also found using this technique.

Comparison of the Photoelectric Properties of Cleaved, Heated, and Sputtered Silicon Surfaces

The properly cleaved silicon surface exhibits highly reproducible photoelectric properties, and is known from electron diffraction results to be of high structural perfection. The surface-atom rearrangement that occurs upon heating the cleaved (111) surface to 1000°K in vacuum causes marked changes in the electrical properties of the surface. The work function of a high-resistivity p-type sample which has the energy bands flat up to the surface in the cleaved state drops from 4.8 to 4.6 eV and the photoelectric threshold drops from ∼5.15 to ∼4.6 eV. The kinetic-energy distribution of emitted electrons and the yield spectrum indicate at least two distinct groups of photoelectrons in both the cleaved and heated samples. The principal group, of lower energy, is attributed to a direct optical transition in both cases. In the heated samples, the group of high-energy electrons is further separated from the low group than in the cleaved samples, and originates from just below the Fermi level at the surface. This group is tentatively ascribed to emission from surface states that have moved up to the Fermi level in the forbidden gap. The yield spectrum for a sputtered and annealed (111) silicon surface is closely similar to that for the cleaved and heated samples, as expected from the fact that in low-energy electron diffraction both types of surfaces have the same atomic arrangement. Changes in photoemission during cleaning show that ∼95% of the photoelectrons are trapped by an oxide layer ∼20 Å thick.

Radiation-Induced Recombination and Trapping Centers in Germanium. I. The Nature of the Recombination Process

Extensive measurements on irradiated germanium indicate that previous analyses of the recombination process are incorrect. A model which explains the observations in both $n$- and $p$-type material is presented. According to this model, the recombination level lies 0.36 ev above the valence band in gamma-irradiated $n$-type germanium. Presumably due to the extensive perturbation caused by neutron irradiation, the level lies slightly lower in neutron-irradiated material. Trapping levels which are not present in antimony-doped germanium occur in arsenic-doped material \ensuremath{\sim}0.17 ev above the valence band. Other trapping levels are observed only in high-resistivity antimony-doped samples. It has not been possible, from the data presented, to determine the values of hole and electron capture cross sections associated with the recombination level; however, the ratio $\frac{{\ensuremath{\sigma}}_{p}}{{\ensuremath{\sigma}}_{n}}$ has been determined: Its value is \ensuremath{\sim}1000.

Hall Mobility of Optically Excited Carriers in Germanium

Steady-state Hall coefficient measurements of optically excited carriers have been made for high-resistivity gold-, iron-, cobalt-, and nickel-doped germanium crystals in the temperature range from 200\ifmmode^\circ\else\textdegree\fi{}K to 54\ifmmode^\circ\else\textdegree\fi{}K. In the spectral region of impurity absorption, ($h\ensuremath{\nu}<0.7$ ev), carriers generated in $p$-type samples are holes; carriers generated in $n$-type samples are electrons. In the region of intrinsic absorption, ($h\ensuremath{\nu}>0.7$ ev), Hall coefficient measurements indicate that photoconduction is due primarily to mobile electrons in both $p$- and $n$-type high-resistivity samples. These observations suggest that, in the crystals studied, hole trapping is primarily responsible for the intrinsic photosensitivity in both $n$- and $p$-type samples. These results are qualitatively consistent with the double-acceptor model, proposed to account for the two impurity levels introduced into Ge by each of the aforementioned impurities.