ELECTRON IRRADIATION OF GERMANIUM

ROBERT RICHARD PEPPER, Purdue University

Abstract

Changes in the electrical conductivity and Hall coefficient of germanium samples, irradiated with 4.5 Mev electrons, have been observed as a function of the total irradiation flux.Ambient temperatures for these experiments were obtained using liquid nitrogen, dry ice and acetone, and ice water as coolants. Intermediate temperatures were attained for annealing studies by means of an electric heater. A special cryostat was used so that the sample could be supported in a magnetic field inside the vacuum system of the Purdue linear accelerator, while still maintaining thermal contact with the coolant. Several samples, representing a wide range of initial carrier concentration were used in these studies.From measurements of the conductivity and Hall coefficient, the carrier concentration and mobility may be calculated. High energy electron irradiation results in the production of defects in the crystal lattice, which may act as donor or acceptor states and change the carrier concentration. In Ge, the carrier concentration changes in such a manner that the Fermi level moves toward a position 0.1 or 0.2- ev, depending on the temperature of irradiation, above the valence band. Thus for relatively impure samples the electron or hole density and the conductivity all decrease with irradiation. For very high purity p type samples, the hole concentration and conductivity may increase.An attempt has been made, following a method suggested by Fan, to determine the position of the energy levels in the forbidden band, which are associated with the defects produced by the irradiation. For sufficiently well separated levels, a plot of the sum of the changes in hole and electron concentrations per unit of incident flux as a function of the Fermi level should produce a series of steps which locate the energy levels and give their relative concentrations. Although the procedure followed is not so straightforward, the experiments do indicate that bombardment produces a set of levels at about 0.06 ev below the conduction band; another level or group of levels, whose exact position is undetermined, is introduced near the center of the forbidden gap.In determining the number of defects per cm3 which are produced, the James-Lark Horovitz model was employed. From the experimental results, as applied to this model, a cross-section was determined and compared to the theoretical calculations based on the theories of Rutherford, Mott, and Feshbach. The theoretical cross section using a threshold energy of 31 ev is 91 barns compared to 67 barns obtained in our experiments.The extent of ionization of the defects is also indicated by this model. Thus the changes in carrier mobility with irradiation could be calculated and compared to experiment. In this way it is confirmed that defects, such as the vacancies in n type Ge, can be doubly ionized.Data on annealing processes indicate very extensive healing (about 50$) of bombardment-introduced defects at temperatures between 0°C and 20°C, but they anneal only slightly at -70°C or lower. It is concluded that room temperature annealing is primarily vacancy-interstitial recombination.Non-equilibrium electronic effects were observed even at liquid nitrogen temperatures. The non-equilibrium processes are attributed to minority carrier trapping centers introduced during irradiation. Intense ionization produced by the electron beam keeps the traps filled with electrons during bombardment. When the beam is turned off, the traps empty very slowly at a rate depending on the ambient temperature. Such effects are usually observed only in p type Ge, both in initially p type material and in samples which have been converted by the irradiation to p type.

Degree

Ph.D.

Subject Area

Electrical engineering

COinS