RU2059324C1 - Method for determining nonuniformity of doping semiconductor layers on insulating substrates - Google Patents

Abstract

FIELD: measurement of semiconductor material parameters. SUBSTANCE: specimen under test and standard specimen uniformly doped to same level are irradiated by γ-quantums forming structure defects. Temperature dependences of charge carrier concentration are measured in both sepcimens and measurement results are used to determine temperature intervals of structure defect ionization. Degree of uniformity of semiconductor layer doping is determined by extension of temperature interval of structure defect ionization in semiconductor layer under measurement relative to temperature interval of structure defect ionization in standard specimen. EFFECT: facilitated procedure. 2 dwg

Description

 The invention relates to a semiconductor measuring technique and can be used for non-destructive testing of uniformity of the doping level of semiconductor epitaxial, ion-implanted and diffusion layers on an insulating substrate.

A known method for determining the heterogeneity of the doping of semiconductors by measuring the dependence of the capacitance of a Schottky diode on voltage [1]
However, this method requires the formation of a test structure on the layer and allows measuring layer heterogeneity only on a low-resistance substrate.

A known method for determining the heterogeneity of doping, based on the measurement of the Hall effect with layer-by-layer removal of material [2]
However, this method is destructive.

Closest to the invention is a method for determining the heterogeneity of semiconductor materials, including irradiating the sample, measuring the same characteristics of the investigated and reference samples homogeneously alloyed to the same level, comparing the measured characteristics and determining the degree of alloying heterogeneity from the desired value [3]
The disadvantage of this method is that with its help it is possible to determine the heterogeneity of only high-resistance low-alloy materials. This is due to the fact that in low-impedance semiconductors (n> 1 · 10 ¹³ cm ^-3 ) the width of the space charge region near the rectifying contact when a blocking voltage close to the breakdown voltage is applied to it is insufficient for reliable detection of the photoresponse signal.

 The purpose of the invention is the expansion of the capabilities of the method in the field of highly alloyed materials.

 The goal is achieved by the fact that in the method for determining the inhomogeneity of doping of semiconductor layers on insulating substrates, in which the sample is irradiated, measuring the same characteristics of the test sample and reference samples uniformly doped to the same level, comparing the measured characteristics and determining the degree of uniformity of the doping of the layer from the comparison , according to the invention, the measured semiconductor layer simultaneously with the reference sample is irradiated with high energy particles, causing structural defects, measure the temperature dependence of the concentration of charge carriers in both samples in the range from 77 to 400 K, determine the temperature intervals of ionization of structural defects within this range, and the degree of inhomogeneity of the doping of the semiconductor layer is determined by the magnitude of the expansion of the temperature range of ionization of the structural defect in the measured semiconductor layer relative to the temperature range of ionization of a structural defect in the reference sample.

(1) где n₁ и n_o концентрации носителей заряда соответственно до и после перехода носителей заряда с энергетического уровня дефекта в зону основных носителей заряда;
N_c эффективная плотность состояний в зоне основных носителей заряда;
g фактор спинового вырождения легирующей примеси;
Δ Е энергетический зазор между энергетическим уровнем дефекта структуры и краем зоны основных носителей заряда;
k постоянная Больцмана.The essence of the proposed method is as follows. When semiconductors are irradiated with high-energy particles, stable radiation defects of the structure are formed in the crystal lattice. These defects introduce a local energy level into the band gap of the semiconductor, located in the interval between the middle of the band gap and the edge of the zone of the main charge carriers. The capture of charge carriers to this level leads to the appearance of a “step” in the dependence of the concentration of charge carriers on temperature in the temperature range of ionization of a structural defect. In this interval, the dependence of the concentration n of charge carriers on temperature T is determined by the following expression:
n

(1) where n ₁ and n _{o are the} concentrations of charge carriers, respectively, before and after the transition of charge carriers from the energy level of the defect to the zone of the main charge carriers;
N _{c is the} effective density of states in the region of the main charge carriers;
g is the spin degeneracy factor of the dopant;
Δ E is the energy gap between the energy level of the structural defect and the edge of the zone of the main charge carriers;
k Boltzmann constant.

(2)
откуда

n_o-

(3)
Соотношение (3) показывает, что Т_и зависит как от концентрации дефектов структуры, которая равна (n_o-n₁), так и от уровня легирования полупроводника, равного n_o. При этом с ростом концентрации дефектов структуры значение Т_и уменьшается, а с увеличением уровня легирования, наоборот, увеличивается. В этом случае, если уровень легирования измеряемого полупроводникового слоя неоднороден, переход носителей заряда с энергетического уровня дефектов структуры в зону основных носителей заряда, происходящий с повышением температуры в интервале ионизации дефектов структуры, начинается в слаболегированных областях кристалла при более низких температурах, чем в сильнолегированных областях, а заканчивается в сильнолегированных областях, при более высоких температурах, чем в слаболегированных. Это приводит к расширению температурного интервала ионизации дефектов структуры в полупроводнике с неоднородным уровнем легирования. Связь величины расширения температурного интервала ионизации дефектов структуры со степенью неоднородности уровня легирования слоя можно определить расчетным путем с помощью анализа соотношения (1) для конкретных значений Δ Е и (n_o-n₁), а также экспериментальным путем с помощью измерения температурных интервалов ионизации дефектов структуры в наборе облученных калибровочных образцов, степень неоднородности уровня легирования которых известна.For definiteness, we denote by temperature T _and ionization of the structural defect such a temperature at which half of the structural defects are ionized. We obtain that the concentration of charge carriers at this temperature is n = 0.5 (n _o + n ₁ ). Substituting this value into relation (1), we obtain
0.5 (n _o + n ₁ )

(2)
where from

n _o -

(3)
Relation (3) shows that T _and depends on the concentration of structural defects, which is equal to (n _o -n ₁₎ and the semiconductor doping level equal to n _o. With increasing value of the concentration of structural defects _and T decreases, and with increasing doping level, on the contrary, increased. In this case, if the doping level of the measured semiconductor layer is inhomogeneous, the transition of charge carriers from the energy level of structural defects to the zone of main charge carriers, which occurs with increasing temperature in the interval of ionization of structural defects, begins in lightly doped regions of the crystal at lower temperatures than in heavily doped regions , but ends in heavily doped areas, at higher temperatures than in lightly doped ones. This leads to the expansion of the temperature range of ionization of structural defects in a semiconductor with an inhomogeneous doping level. The relationship between the expansion of the temperature interval of ionization of structural defects with the degree of heterogeneity of the layer doping level can be determined by calculation using analysis of relationship (1) for specific values of Δ E and (n _o -n ₁ ), as well as experimentally by measuring temperature intervals of ionization of defects structures in the set of irradiated calibration samples, the degree of heterogeneity of the doping level of which is known.

The essence of the proposed method is illustrated in FIG. 1, which shows the temperature dependences of the concentration of charge carriers in two uniformly doped semiconductor samples, in which the same number of radiation-induced structural defects are formed by irradiation with high-energy particles. The doping level of the first sample n _o (curve 1) is higher than the doping level of the second sample n ' _o (curve 2), therefore, as can be seen from FIG. 1 dependencies defects ionization structure temperature T 'in _the first sample is higher than the value T'_'and in the second sample. In FIG. 2 curve 1 shows the temperature dependence of the concentration of free charge carriers in a uniformly doped reference sample and curve 2 in a semiconductor layer, the doping level of which is inhomogeneous, but the concentration of free charge carriers n _o integrated over the layer (the average level of doping) is the same as in the reference sample. The reference sample and semiconductor layer, the characteristics of which are shown in FIG. 2, were simultaneously irradiated with the same stream of penetrating high-energy particles. In FIG. 2 it can be seen that the temperature interval T _o -T ' _{1 of} ionization of structural defects in a non-uniformly doped semiconductor layer is wider than the temperature range T _o -T _{1 of} ionization of the same structural defects in the reference sample.

PRI me R. The degree of heterogeneity of the doping level of the epitaxial layer of n-silicon 3, OKEP1, grown on a sapphire substrate is determined. For this, the integral concentration of charge carriers in the layer at room temperature is measured by the van der Pauw method. It is equal to n _o = 4,5 · 10 ¹⁵ cm ^-3 . A reference sample of silicon KEF1, uniformly doped with phosphorus to a level of 4.5 x ^{10 15} cm- ^{3, is} selected. These conditions are met, as a rule, by bulk samples of silicon cut from an ingot, uniformly alloyed from the melt. The measured layer and the reference sample are simultaneously irradiated with a flux of γ-rays of ⁶⁰ Co equal to ≈ 5 · 10 ¹⁸ cm ^-2 . Such irradiation leads to the formation of structural defects in irradiated materials, usually called A centers, which are complexes of an oxygen atom and a vacancy (i.e., an unoccupied silicon atom in the lattice). The energy ΔE and the spin degeneracy factor of these defects are known and are equal to 0.17 eV and 2, respectively. Then, the temperature (in the range of 77–400 K) dependence of the concentration of charge carriers is measured in the epitaxial layer and the reference sample. The measurement results are presented in graphical form (Fig. 2 curve 1) for the reference sample and curve 2 for the epitaxial layer. From the graph determine the temperature range of ionization of structural defects at which the values of n _o and n ₁ in the samples are equal to 4.1 · 10 ¹⁵ cm ^-3 and 10 ¹⁴ cm ^-3, respectively. The temperature range of ionization of structural defects in the reference sample is T _o -T ₁ = 250-91 = 159 K, and in the measured layer T ' _o -T' ₁ = 261-82 = 179 K. The degree of heterogeneity of the doping level of the reference sample is considered equal to unity. The degree of inhomogeneity of the doping of the measured semiconductor layer is determined by the magnitude of the expansion of the temperature range of ionization of structural defects, which was Δ179-159 = 20 K, and relations (1) - (3), according to which the obtained value Δ corresponds to a change in n _o from the size of the sample from 3, 6 · 10 ¹⁵ to 5.4 · 10 ¹⁵ cm ^-3 . The degree of heterogeneity of the doping level of the measured epitaxial layer turned out to be Δ n _o = 5.4 · 10 ¹⁵ / 3.6 · 10 ¹⁵ = 1.5. The determination of the heterogeneity of the doping of the same layer by the prototype did not allow to obtain the result due to the fact that with such a high initial electrical conductivity of the layer, the prototype was inoperative.

 Thus, the above example shows that when using the proposed technical solution, there is an expansion of the capabilities of the method in the field of highly alloyed materials.

The proposed method was used when sorting by the degree of inhomogeneity of doping on a batch of 40 plates of diffusion layers used as the base areas of bipolar transistors. In this case, all the plates of the party were randomly divided into two sub-parties. One subparty A went through all further transistor manufacturing operations using standard technology. In another sub-party B, the proposed method determined the degree of heterogeneity of the doping level of the layers and the sub-party was divided into three series B ₁ , B ₂ and B ₃ . The value of Δn _o in the series In ₁ was in the range of 1.0-1.3; in series B ₂ in the range of 1.3-1.5; and in series B ₃ 1.5-1.8. During the manufacturing operation of the emitter in the series B ₁ , B ₂ and B ₃ an adjustment was made for the value Δn _o ; for series B _{1, the} dose of implantable impurity was 800 μC, respectively (as in standard technology), for series B ₂ 900 μC, for series B ₃ 1000 μC. After that, the percent of output transistors in the A and B subparts was controlled, which amounted to 80% and 93%, respectively. This shows that a breakdown by the degree of heterogeneity of the doping level by the proposed method allows increasing the yield of suitable semiconductor devices.

Claims (1)

 METHOD FOR DETERMINING THE DEGREE OF HETEROGENEITY OF DOPING SEMICONDUCTOR LAYERS ON INSULATING SUBSTRATES, including irradiating a sample with electromagnetic radiation, measuring the same characteristics of a controlled and uniformly doped to the same concentration as a controlled, standard sample, comparing the measured characteristics and unevenness based on it layer, characterized in that, in order to provide the ability to measure highly doped semiconductor layers, semi the conductor layer simultaneously with the standard sample is irradiated with a flux of γ quanta, and the temperature dependences of the concentration of charge carriers on the temperature are measured in the range of 77 400 K, the temperature ranges of ionization of radiation defects of the structure are determined within this range, and the degree of inhomogeneity of the doping of the semiconductor layer is determined by the expansion temperature range of ionization of a structural defect in a controlled semiconductor layer relative to the temperature range of ionization and a structural defect in the standard sample.

Images