SU915662A1 - Method for measuring rate of surface recombination of semiconductors - Google Patents

Description

The invention relates to semiconductor technology and can be used to control semiconductor wafers used for the manufacture of semiconductor devices.

A known method for determining the surface recombination rate of semiconductor heterostructures, [K], based on the measurement of the photoluminescence decay time.

The disadvantage of this method is its complexity, due to the need for irradiating the sample with monochromatic radiation.

The closest technical solution is a method for determining the surface recombination velocity in semiconductors, [2 [based on placing a flat sample in a magnetic field, the intensity vector of which is parallel to the sample plane, creating a pulsed electric field in the sample, the intensity vector of which is parallel to the sample plane and perpendicular to the magnetic field strength vector, measuring the dependence of the recombination luminescence intensity on the magnitude of the pulse electric field Skog fields of alternating polarity through the radiation detecting two opposite 30

2

surfaces of a flat sample and determining the surface recombination rate by calculation.

The disadvantages of this method are its complexity, low accuracy, and a narrow range of determined surface recombination rates.

The purpose of the invention is to simplify, improve accuracy and expand the range of 10 defined surface recombination rates.

This is achieved by measuring the dependences of the intensity of negative luminescence on the magnitude of the intensity of a pulsed electric field in the method for determining the surface recombination rate in semiconductors, and the surface recombination rates are calculated using the formula

where 5+ is the surface recombination rate on the sample surface;

- surface recombination rate on another sample surface;

y is a dimensionless field; β — dimensionless sample thickness;

915662

four

P _about - the maximum value of the intensity of negative luminescence;

P ~ is the intensity of negative luminescence in the initial part of the dependence P (E) n = cond81 ·

FIG. 1 shows a diagram of the device for implementing the method; in fig. 2 - field dependences of negative luminescence signals.

The device contains a semiconductor crystal 1, equipped with ohmic contacts and leads, electromagnet 2 poles, focusing lenses 3, radiation receivers 4, negative luminescence signal 5 from the front face, negative luminescence signal 6 from the back face.

The method is as follows.

In the absence of external fields, an equilibrium radiation flux P _o comes out of the semiconductor, the value of which is proportional to the equilibrium concentration of current carriers n2. this face becomes less than the equilibrium value. The intensity of the recombination radiation decreases by the corresponding samples, i.e., negative luminescence is observed. With a strong depletion of the near-surface part of the semiconductor, the modulation depth of the recombination radiation practically becomes equal to the value of P _o and saturation is observed on the field dependence of P (t, H). Thus, the maximum amplitude of negative luminescence is equal to the power of the equilibrium recombination radiation of free electron-hole pairs. The considered situation is typical for semiconductors with intrinsic conductivity.

With small deviations of the intensity of recombination radiation P from P _o , a straight line dependence of P (E, H) is observed, the slope of which in particular is determined by the surface recombination rate on the crystal face under study and does not depend on 5 on the other face. Since formula (1) includes the power ratio Ro / Po-P, it is sufficient to measure the luminescence signals in relative units, and for the unit to take the maximum negative luminescence signal in the saturation mode.

Thus, using the phenomenon of negative luminescence, it is possible to determine the surface recombination rate in low-resistance semiconductor crystals at high temperatures,

in which other methods, for example, based on the measurement of photoconductivity or photoluminescence, are not effective.

The intensity of the photoluminescence of a semiconductor placed in crossed E and H fields is influenced by two oppositely directed processes.

The first one, which leads to an increase in luminescence, is associated with a decrease in self-absorption losses of recombination radiation by the crystal thickness when the nonequilibrium carriers are pressed towards the surface under study, that is, the effect of increasing the photoluminescence is observed due to a decrease in the optical path of the outgoing rays.

The second one, which causes quenching of photoluminescence, is associated with a decrease in the total number of nonequilibrium carriers due to their nonradiative recombination on the surface.

With an increase in the field y ~ £ XI, both processes proceed simultaneously and the total effect of a change in the photoluminescence intensity depends on their partial contribution.

If the surface recombination rate is small, then in the region of small control fields y ~ £ X Я I, the luminescence increases to a certain, depending on the value of the field Tmax. (dominates the first process). With a further increase in the field y, the second process begins to dominate — a decrease in the total number of carriers and luminescence quenching occurs. So with umako. there is a maximum by which the value of 5 can be determined.

With an increase in the surface recombination rate, the photoluminescence maximum shifts to lower-field y, and then completely disappears, and only y quenching of the photoluminescence is observed with increasing y. Therefore, in this way, one can measure 5 only in a limited range of values at which the maximum dependence of the photoluminescence field is observed.

To observe negative luminescence, there is no need to use an external excitation source, since in this case the source of charge carriers under non-equilibrium conditions is the semiconductor surface itself, which has a certain recombination rate.

Determining the surface recombination rate from an analysis of the field dependences of negative luminescence induced in a semiconductor using crossed £ and полей fields made it possible to reduce the focusing operation of the exciting radiation on the sample and determine its absolute value, i.e., simplify the method and increase its accuracy,

915662

five

Since in the proposed method of determining 8 there is no need to measure the radiation power · in absolute values, the accuracy of this method is significantly increased.

When fields and Λ are crossed in a semiconductor, electron-hole pairs drift in the direction normal to the faces under study and a decrease in the carrier concentration occurs in the near-surface region of the semiconductor, which is accompanied by negative luminescence, which determines the surface generation rate from the field dependencies.

Under conditions of quasineutrality, the rates of generation and recombination on the surface are equal to each other. In principle, in practice, one can always choose the values of external fields such that negative luminescence is observed. For this reason, the range of measured values of 5 using the proposed method is much wider than that of the prototype and has no strict limitations.

Example. Samples were made of undoped p-1p5b with a concentration of uncompensated donors Νά - Ν _α = = 10 ¹⁴ cm ^-3 . Plate thickness 10 ^-2 cm.

Ohmic contacts are applied by fusing indium with tellurium. Using the holder, the samples are placed between the poles of an electromagnet, after which a pulsed electric field of alternating polarity is applied to the contacts of the crystal. The resulting modulated recombination radiation is focused by means of lenses on photodetectors and then recorded with an oscilloscope. According to the measurement data of the negative lu-minescence signals, the field dependences are plotted from both faces, shown in FIG. 2. Measurements are carried out at a constant magnetic field of 5 kgf.

The depth of modulation of the recombination radiation in the region of saturation of the field dependences P (E) corresponds to the value P _o . The density of equilibrium radiation fluxes is the same for both faces, therefore the saturation signals measured by different receivers with unequal sensitivity can be equated to each other and taken as a unit of reference.

By selecting any points on the straight line field dependencies (points A and B), they find the corresponding values of the fields E at a known constant N. Point A corresponds to E = 7 V / cm, and B — 5 V / cm at H 5 kgf. Substituting the calculated values of β and y in the formula (1), we find the values of the surface recombination rates on the opposite faces of the crystal. For μ _π = 7.7 · 10 ⁴ , μ _ρ = 7 · 10 ² and τ = 10 ^-8 with the surface recombination rate of co6

responsibly equal to 5a = 3.4-10 ⁴ cm / s and 5+ = 2.2-10 ⁴ cm / s.

The proposed method of measuring the speed of surface recombination, while maintaining the technical and economic advantages of the prototype method, has a reduction in the number of measurement operations; the simplification and cheapening of measuring equipment that implements the method by eliminating lasers and monochromators; expansion of the range of measured surface recombination rates; decrease in labor input both at measurements, and when carrying out calculations; improving the accuracy of determining the rate of surface recombination.

Claims (1)

Claim A method for determining surface recombination rates in semiconductors based on placing a flat sample in a magnetic field whose intensity vector is parallel to the sample plane, creating a pulsed electric field in the sample whose intensity vector is parallel to the sample plane and perpendicular to the magnetic field strength vector, and measuring the dependences of the recombination luminescence intensity on the intensity of the pulsed electric field of alternating polarity by registering radiation from two opposite surfaces of a flat sample and determining the surface recombination rate by calculation, characterized in that, in order to simplify, improve accuracy and expand the range of surface recombination rates determined, measure the dependences of the negative luminescence intensity on the magnitude of the pulsed electric field, and the surface recombination rate calculated by the formula where 5+ is the surface recombination rate on the sample surface; five-. - surface recombination rate on another sample surface; γ is a dimensionless field; β — dimensionless sample thickness; P _about - the maximum value of the intensity of negative luminescence; Р_ is the intensity of negative luminescence in the initial part of the dependence P (Е) _n = const