RU2145128C1 - Method for producing n-type nuclear-doped silicon (options) - Google Patents

Abstract

FIELD: electronic engineering. SUBSTANCE: method involves enrichment of source silicon ingot with N-times stable isotope ³⁰Si and its irradiation with neutrons. According to main nuclear reaction (n, γ), desired concentration of donor ³⁰Si is obtained on stable isotope ³¹P contained in natural mixture of silicon isotopes. Then ingot is decontaminated, subjected to radiometric check-up and to high-temperature annealing for eliminating essential portion of radiation defects and for stabilizing electrophysical properties of material. According to first option, fluence rating of neutrons is reduced N times thereby reducing concentration of radiation defects to the same degree. According to second option, fluence rating specified for irradiating silicon with natural isotope mixture is maintained, N times higher concentration of donor ³¹P and, hence, concentration of charge carriers, is obtained. For reducing amount of complex radiation defects that cannot be eliminated by post-irradiation high- temperature annealing when using second method, proportion of high-energy neutrons is reduced by means of neutron spectrum moderator. EFFECT: reduced amount of complex radiation defects and improved electrophysical properties of silicon. 2 cl, 1 dwg

Description

 The invention relates to the field of nuclear technology for the production of poly- and single-crystal n-type silicon and has the goal of dramatically expanding the control capabilities of the structure and, through it, the electrophysical properties of semiconductor silicon and devices based on it.

где n - нейтрон, γ - гамма-излучение при захвате нейтрона; β^- - бета-частица с энергией 2,5 МэВ, испускаемая со временем полураспада 157,3 мин компаунд-ядром ³¹Si, ³¹P - легирующая примесь - фосфор, являющийся продуктом ядерной реакции (И.М. Греськов, С.П. Соловьев, В.А. Харченко. "Ядерное легирование полупроводников. Обзорная информация" НИИТЭХИМ, Москва, 1982 г., с. 19; В.А. Харченко, С.П. Соловьев - Известия АН СССР, Сер. Неорганические материалы, 7, N 12, 2137, 1971 г.).At present, n-type silicon doping is carried out by neutron irradiation of a silicon ingot of the required dimensions and quality in a nuclear reactor or other nuclear-physical installation, when the main nuclear reaction is a reaction on a stable ³⁰ Si isotope contained in an amount of 3 natural isotopes ,12%:

where n is the neutron, γ is gamma radiation during neutron capture; β ^- - beta particle with an energy of 2.5 MeV, emitted with a half-life of 157.3 min compound nucleus ³¹ Si, ³¹ P - dopant - phosphorus, which is the product of a nuclear reaction (I.M. Greskov, S.P. Soloviev, V. A. Kharchenko. "Nuclear Doping of Semiconductors. Overview" NIITEKHIM, Moscow, 1982, p. 19; V. A. Harchenko, S. P. Soloviev - Izvestiya AN SSSR, Ser. Inorganic Materials, 7 N 12, 2137, 1971).

 The main disadvantage of this prototype is the limitation of the neutron fluence rate and, therefore, the degree of doping with phosphorus due to the accumulation of complex radiation defects that are not eliminated by the post-radiation thermal annealing of the irradiated ingot and significantly reduce the electrophysical properties of silicon. The fluence rate is also limited in connection with the inevitability of induced ingot radioactivity for nuclear technology.

The essence of the proposed method lies in the fact that the starting silicon raw material is enriched with a stable ³⁰ Si isotope N times, an ingot of the required dimensions and quality is obtained from the enriched raw material in a known manner, and the ingot is doped in a nuclear reactor or other nuclear-physical installation by reaction (1) . In this case, two variants of the method for producing nuclear-doped silicon of n-type can be provided, for which the post-radiation operations — decontamination of the ingot, its dosimetric control, annealing of radiation defects, improving and stabilizing the electrophysical properties of the ingot, are carried out in the same way as provided by the prototype. In this case, both variants of the method ensure the achievement of the same technical result — a decrease in the concentration of complex radiation defects and, accordingly, an improvement in the electrophysical properties of silicon.

 Option 1.

 Reducing the neutron fluence by a factor of N in comparison with the above-mentioned prototype, a concentration of residual complex radiation defects and a corresponding increase in the electrophysical characteristics of silicon are obtained by a factor of N in freshly irradiated material.

 Option 2

где d₀ - диаметр дефекта, Δ - мода распределения (для нейтронов спектра деления

). Также убедительно показано, что при уменьшении энергии бомбардирующих частиц мода распределения резко сдвигается в сторону меньших размеров дефектов. Полностью термализованные нейтроны комплексных радиационных дефектов практически не создают.While maintaining the fluence norm established for the prototype, an increase in N times the content of ³⁰ Si isotope in silicon leads to an increase in N times the concentration of the donor impurity evenly distributed in silicon - ³² P, i.e. carrier concentration. This method requires taking measures to eliminate the high-energy part in the neutron spectrum, which is associated with the formation of complex radiation defects in matter. The required change in the neutron spectrum is achieved by using well-known moderators of beryllium or other materials with low atomic weight; moderators are installed in the irradiation channel. The physical justification for this pathway is electron microscopy and X-ray diffraction studies of the distribution of complex radiation defects in crystalline materials depending on the energy of the bombarding nuclear particles (K. Mercle-Report AEPE-R5269, Harwell, 1966; ML Jenkins, CA English-Journ. Nucl. Mat. 108 -109, 46, 1982; VV Kirsanov "Complexes of point defects in irradiated cubic metals, their size distribution", Preprint NIIAR, P-60, Melekess, 1970). This distribution is subject to Rayleigh law:

where d ₀ is the diameter of the defect, Δ is the distribution mode (for neutrons of the fission spectrum

) It has also been convincingly shown that, as the energy of the bombarding particles decreases, the distribution mode shifts sharply toward smaller defect sizes. Fully thermalized neutrons of complex radiation defects practically do not create.

Due to the fluence reserve ingot obtained for the ³⁰ Si isotope enriched, the fluence reserve allows the technologist to maneuver when implementing the first or second options. The quantitative side in determining the compromise between the first and second options is provided by a set of graphs or tables of the dependence of certain controlled parameters on the irradiation and annealing mode.

For both options (N 1 and N 2), the statement is true: since complex radiation defects act on a number of important physical properties of silicon as foreign impurities, the highest result during its nuclear doping is provided for maximum enrichment with the stable ³⁰ Si isotope. The transition from the polyisotopic composition of silicon ( ²⁸ Si - 92.18%, ²⁹ Si - 4.71%, ³⁰ Si - 3.12%) to the monoisotopic ³⁰ Si by itself, i.e. without neutron irradiation, it acts in the same direction, since in this case the frequencies due to the difference in the atomic weights of the isotopes are eliminated in the vibrational spectrum of the crystal.

 Example No. 1.

The proposed option N 1 for producing nuclear-doped n-type silicon was implemented on the experimental channel of a nuclear reactor using standard silicon ingots with a natural mixture of stable isotopes obtained by the crucible-free melting method, with a neutron fluence of 7 • 10 ²⁰ cm ^-2 , the concentration of charge carriers 2 was obtained. 5 • 10 ¹⁷ cm ^-3 ; at the same time, for a 20-gram silicon ingot with dimensions of about 60x12 mm, the content of which the stable ³⁰ Si isotope is increased from 3.12% to 15%, i.e. approximately 5 times, the same concentration of charge carriers was obtained after irradiation with a fluence of 1.5 • 10 ²⁰ cm ^-2 . A graph of the dependence of the concentration of charge carriers on the neutron fluence, where two experimental points are obtained for the enriched silicon ingot, is given in the drawing. Thus, the ratio of the fluence value is close to the reciprocal of the ratio of the content of the named isotope in the irradiated ingots taken for the experiment. This indicates the correctness of the technical solution and the practical value of the proposed method.

 Example No. 2.

Option N 2 for producing nuclear-doped n-type silicon was implemented in the same experiment as option N 1, the results of which are shown in the drawing. From the dependence of the concentration of charge carriers on the neutron fluence (for example, for the same value of Ф _Н = 10 ¹⁸ cm ^-2 ) it can be seen that this concentration is approximately 5 times higher for a silicon ingot enriched 5 times with the ³⁰ Si isotope, compared with an ingot with a natural mixture of isotopes. Thus, the enrichment of silicon N times with the ³⁰ Si isotope leads after its neutron irradiation to a higher (N times) degree of doping with phosphorus, i.e., option N 2 allows one to obtain high supersaturation of silicon with phosphorus with its uniform distribution in the bulk of the crystal. Moreover, the use of a moderator of a neutron spectrum moderator in this embodiment makes it possible to reduce the fraction of high-energy neutrons and, accordingly, reduces the number of complex radiation defects.

Claims (2)

1. A method for producing nuclear-doped silicon of n-type, in which the ingot of the starting silicon is irradiated with neutrons and the main concentration of the donor impurity ³¹ P is obtained on the stable ³⁰ Si isotope by the main nuclear reaction (n, γ), the ingot is deactivated, its dosimetric control, as well as to remove a substantial part of the radiation-induced defects and stabilize the material properties of its thermal annealing, characterized in that the silicon source before it is enriched in exposure times N stable isotope ³⁰ Si and lower in N times neutron fluence and, ca th, the concentration of complex radiation defects that can not be thermally annealed. 2. A method for producing nuclear-doped silicon of n-type, in which an ingot of the starting silicon is irradiated with neutrons and the desired concentration of donor impurity ³¹ P is obtained on the stable isotope ³⁰ Si on the stable isotope ³⁰ Si, the ingot is deactivated, its dosimetric control, as well as to remove a significant part of radiation defects and stabilize the properties of the material, its thermal annealing, characterized in that the initial silicon is enriched N times with the stable ³⁰ Si isotope before irradiation and while maintaining the neutron fluence with A moderator of the neutron spectrum reduces the fraction of high-energy neutrons responsible for the formation of complex radiation defects.