JPS60195437A - Evaluating method of crystal - Google Patents

Abstract

PURPOSE:To measure quickly and easily a distribution of a transition density in a crystal without breaking down the crystal, by measuring a difference of transmittivity between plural points, as a function of a moving distance of a light spot. CONSTITUTION:Light 1 which can transmit through in a sample 5 is made incident vertically on the surface of the sample 5 through a lens 3 and a mask 4, and vibrating incident light spots 16, 17 are given onto the sample by rotating a chopper 2. Also, light which has transmitted through the sample 5 is detected by a photodetector 6' of the end part of an optical fiber 6. Subsequently, an output of the photodetector 6' is provided to a synchronous rectifying amplifier 11, and a difference of a transmitting light intensity between two points on the surface of the sample 5 is derived, and recorded by a recorder 14 as a distribution of a transition density of the crystal 5.

Description

[Detailed Description of the Invention] [Technical Field] The present invention relates to a method for evaluating crystals, for example, semi-i4 < 4 <4'
; To fabricate a crystal device, the dislocation density can be quickly and easily reduced by A1 to /111 without destroying the crystal.
This relates to a method for determining

[Prior art] Currently produced soil/conductor crystals contain many dislocations, for example, CaASm crystals have 10' to 105 dislocations.
It is known that it contains dislocations (linear defects) on the order of cIl/cm3.

The presence of these dislocations creates traps in the crystal, lowers the mobility of carriers, and deteriorates the light emitting characteristics and life of light emitting diodes and laser diodes. - influence;

Furthermore, in general, the degree of dislocation vfI differs from wafer to wafer, and is also non-uniform within the wafer surface! Often takes 1.

Conventionally, the density (17) in a crystal could not be measured electrically, so it was considered to be a first-order Café density.

(1) Etch pits are removed. This method utilizes a solution reaction, and for each conductor crystal, a suitable corrosive solution, for example 1) For δAsrp crystals, sulfuric acid and hydrogen peroxide (H) are added. ) Noze: Using a water solution, the crystal is immersed in this solution, and the places where dislocations intersect with the crystal surface are dissolved to create etch pits. I will do it.

This method creates depressions on the crystal surface, so
This method has the disadvantage that the determined crystal can only be used as is for manufacturing devices. Furthermore, there is a great possibility that the components of the corrosive liquid will enter the crystal, especially when it comes to ultra-thin 1-mock crystals prepared by molecular beam shrimp taxi-rMBlj.
It cannot be applied to

(2) Etch Hillock Method In this method, unlike the method (1) above, corrosion or heat treatment is performed so that the locations where dislocations intersect with the crystal surface are raised.

For example, a material that generates a special oxide by causing a thermochemical reaction with the intersection of dislocations is applied to the crystal surface, and heat treatment is performed to generate the oxide. Next, dissolve the material on the surface where there is no blemish, leave a bump of oxide at the place where it intersects with the surface, and use optical wj, 0, etc. Count by 1.

This method also produces crystals with a high temperature of 11. There are also cases where it is necessary to heat-treat at tl, which has the disadvantage that the fine crystals are damaged.

(3) Deflation method There is an extremely strong force at the location where dislocations exist, and when impurity atoms are diffused into such a crystal at high temperature, the impurity atoms will be located near the Jr of the dislocation. If the host crystal is transparent to visible light and the impurity atoms have absorption in the visible region, then when such a crystal is irradiated with visible light and observed using an optical W4 microscope, Dislocations appear colored. Using this method.

Observation of the steric configuration of dislocations is the turning part.

However, this method also involves heat treatment at high temperatures, so it is a type of destructive inspection.

(4) X-ray diffraction a1 micromethod In general, perfect crystals have multihair reflections, so the diffraction intensity of X-rays is weak. 1. However, if there are dislocations in the crystal, scattering due to multihair reflection does not occur, so the X from such a crystal
The line diffraction intensity becomes stronger. Therefore, dislocations can be imaged by using an X-ray film, a turning point, a television camera sensitive to X-rays, or the like.

However, although the X-ray diffraction WIm method is widely used for silicon crystals, semiconductors such as GaAs have an absorption coefficient for X-rays 104@f1 degrees larger than that of S].
It takes a long time to capture an image. Furthermore, although it is possible to perform imaging in a short time using powerful X-rays, the equipment becomes extremely expensive. Therefore, conventionally, GaAs has been mainly evaluated based on the etch pit shape, although this has been lacking.

[Object of the Invention] Therefore, the object of the present invention is to solve these drawbacks and speed up dislocation in crystals such as semiconductor crystals. An object of the present invention is to provide a crystal evaluation method that quickly and easily determines -II+ 0 without destroying the crystal.

[Structure of the Invention] In order to achieve the above object, the present invention uses a crystal sample! - Alternately irradiate multiple points at a certain period and detect the transmitted light from the crystal? [F'J in the first issue? 1: Then, the electrical signal is subjected to phase Wi sensing detection in synchronization with the synchronous Iha signal, and the transmission between each two of the multiple points is detected. Detecting the difference in V as a signal, then moving a plurality of light spots irradiated onto the crystal sample in the measurement direction and stopping, and determining the difference in transmittance at the stopped position, By sequentially repeating movement and measurement, the difference in transmittance between a plurality of points is measured as a function of the moving distance of the light point.

[Actual Afa Example] The drawings are shown below. (1) A detailed explanation of the present invention will be given below.

Figure 1 shows l! used in the measuring method of the present invention. This is an example of a lit constant device. In FIG. 1, l is, for example, near-infrared monochromatic light that can be transmitted through the sample 15, and this light is passed through a two-stage slit 911'L, and the sunken Nayoppa 2 transmits the light to 1- A single beam of light is generated which intersects r1' in the downward direction. This ray 1 is transmitted through a lens 3 and a mask 4.
An electric current is made incident on the surface of the sample 5 through the sample 5. Here, the optical mask 4 is for shielding stray light, and the optical mask 4 is for shielding the sample 5.
installed in front of the Incident light on the surface of sample 5
=18. +7 is the frequency gQ].

~to' Hz and amplitude d (approximately 1+am) below h &
Rotate chopper 2 so that it moves. Note that this frequency can be improved by the response characteristics of the light-receiving element, which will be described later. The light 15 that has passed through the sample 5 is made to approach the sample 5 and enter an optical fiber 6 installed with a gap of, for example, about 0.5 mm. The light 15 is detected by a PbS light-receiving element 6' installed on the end face of the optical fiber 6 and converted into a signal lO. The signal 10 is routed to the amplifier 11.

On the other hand, the light emitting diode 7 and the phototransistor 8 are arranged to face each other so that the light beam passes through either the outer slit or the center slit of the chopper 2. The phototransistor 8 receives the light and converts it into an electrical signal. This signal 9 is used as a synchronization signal to synchronize 1 with 1R number 10.
ff, input to the amplifier 11. Here, by narrowing down the bandwidth and performing phase sensitive detection, a synchronous rectification output for No. 46 10 is obtained, and this output is recorded on the recorder 14.

In this way, the difference in transmitted light intensity between the two points 16 and 17 on the sample surface can be determined.

In order to measure the difference in transmitted light intensity between two other points on the surface of the sample 5, as shown in FIG. Distance between #
It moves discontinuously by a distance #X that is an integral multiple of d=approximately III11). That is, the sample 5 is stopped after moving about 1 mm, and the difference in transmitted light intensity between the two points is determined. This -11
When the 1 value is determined, the distance is further moved by about 1 aun, and the difference in transmitted light intensity between the two points is determined by 4111 times. In this way, the desired measurement can be performed by moving the sample 5 so that the light hits the position where the dislocation density is desired and repeating the following measurements.

Next, the measurement procedure according to the method of the present invention will be explained.

First, the chopper 2 is moved up and down to adjust the intensity of the light incident on the two light beams 1 to be equal to each other, and the chopper 2 is fixed at that position. Then, the intensity of the incident light 1o'Jj is measured.

Next, a test tube 45 is inserted into the optical path as shown in FIG. 1 to determine the transmitted light intensity I relative to the incident light intensity IO. The position of the light spot at this time is defined as the origin CX=O). As a result of this measurement, as shown in FIG. 2, the light transmittance at the origin is T(0) = I/I. or can be obtained.

Next, the scanning distance X of the light spot is multiplied by the amplitude d of the light spot ad(
ll1m112+31"' I n ) and fix the sample at that location, and measure the intensity of transmitted light between the two points using the method described above. At this time, the change in transmitted intensity at each scanning distance #K is respectively Δ■1.ΔI2゜ΔI
3. ...Let it be ΔIn. If you change, ■, then only x-d is 1
11. First, Δ■ is measured with a neutral force, and then Δr is measured by moving J1 by x−d again. The same 4111 constant is repeated thereafter. Here, ΔIi/Io (i=l,
2, 3. ...-, n) is the light h iM between the two points x = (i-1) d and id, which is increased by the amplitude d.
This is equal to the difference of 10, and this is set as the change in light transmittance at each scan t/heel, ΔTi, that is, 6 pieces 1 = ΔIi/
Io (i = 1.2s..., n), and when defined in this way, the light iA at a point nd apart from the origin
4/ is given by T(nd) = T(o) ten, ΣΔT1 +=1 = T(o)+(3'ΔIi) / To (+)11. Δ11 in this formula includes the sign.

1! If the interval sentence that moves the crystal in order to do In 'A-' does not match the amplitude d of the light change, then 9' T(x) = T(o) + (Q /d) Σ (ΔIi/
Io)-;1 (2) Calculate the light transmittance at a position of arbitrary scanning distance #X. Here, (ΔIi/Io) is the difference in light transmittance between two points separated by the amplitude d, as in equation (1). (2
) was derived by assuming that the difference in light transmittance between two points is proportional to the distance between the two points.

As described in 1- above, a light transmittance distribution curve along a predetermined scanning path is obtained as shown in the second factor.

FIG. 3 shows the results of measuring the distribution of signal intensity and light transmittance for a GaAs medium crystal wafer not doped with impurities using the measuring method according to the present invention.

Further, the dislocation density 4j measured by the etch bit method on the same scanning path is shown in FIG. 4 in comparison with FIG.

From the minute 1ti curves in Figures 3 and 4, each A ÷ scanning distance X
Regarding the point of light transmission II! When considering the relationship between the S ratio T(x) and the dislocation density ta°, the correlation curve in Figure 5 is f! ), and it was confirmed that there is a correlation between optical transmittance and dislocation density.

Figure X1IJ3 uses a lead sulfide PbS cell as a photodetector to irradiate a non-doped GaAs sample with semi-transparent wavelength of 1550 nm (0.8 eV), which has been chemically polished on both sides, to obtain a light spot amplitude of 1.015 mm and vibration frequency. It was measured at 100Hz.

Figure 4 shows the results of corroding the same sample by immersing it in a caustic potash KO) I solution at 420°C for 5 minutes, and measuring the density of the resulting Euchi pits using a reflective S-microscope. be.

According to the correlation curve in FIG. 5, the light transmittance changes inversely proportional to the dislocation density. This result can be interpreted qualitatively as follows.

When light with an energy smaller than the GaAs band gap of 1.4 eV passes through a single crystal, the four times that the light is blocked are due to non-uniform distribution of impurities and dislocations, that is, light convergence due to impurity atoms, or Diffuse reflection caused by light scattering due to dislocations is considered. Since impurities form a light-absorbing level, the transmission τF should have a wavelength dependence on the irradiated light, or such a phenomenon has not been observed. The uniformity can be interpreted as being due to light scattering due to dislocations.

In this way, by obtaining a characteristic curve representing the relationship between optical transmittance and dislocation density in advance, the dislocation density contained in the same type of semiconductor crystal with the same impurity content, thickness, and surface condition as the standard sample can be calculated. , was determined in advance by measuring the transmittance distribution of the crystal. It can be easily determined from the correlation between transmittance and dislocation density.

Further, if the purpose is simply to identify a location with a low dislocation density within the wafer surface, it is not necessarily necessary to obtain the above-mentioned characteristic curve in advance. In addition, for such purpose, it is not necessary to use a sample with both sides polished, and even if the sample is polished on one side or 11 sides, there is no problem in determining -Ill.

[Effects] As is clear from the above description, the present invention provides the following effects.

(1) Toru Il! High S subtle cooperation of low strength? ,・-1+1 'Ji'' can be done with 8 claws.

1'2) Using equipment that is inexpensive and can be handled throughout the fall,
Without destroying the crystal, +IN (play with ': 1α min 4)
Moreover, since it is a highly destructive measurement, the Al11-determined crystal can be used as it is for the production of de/heis.

(3) It takes less than 10 minutes to determine A+++ for one crystal weight/'%, and it is possible to determine Il+11 in ilL'+ compared to the case of one person.

(4) If it is possible to specify a rough location with a low dislocation density, the in-plane distribution of transmittance may be relative. In such cases, it is possible to use a wafer sample with one side polished and the other side sliced.

(5) The 411 constant of the dislocation density distribution of a sample with a wide surface Jh is 1-dimensional.

(6) The measuring force method of the present invention is suitable for automated measurement of crystal wafers, and scanning can be performed quickly.
- equal Sl of dislocation density by dimensional scanning;

Topological display can be performed using a computer etc.

(7) For the same reason, by using a laser beam with a wavelength that can pass through the crystal as a light source, and by using the signal as it is without undergoing addition processing, a microscope that displays minute dislocation 4J patterns can be developed. It is also possible to make one. Since the signal obtained by the method of the present invention has a differential feature, this feature is advantageous in increasing the contrast of the image in such a display.

(8) The measuring method of the present invention can be effectively applied not only to the measurement of the dislocation density distribution of semiconductor crystals, but also to the A111 constant of the dislocation density distribution of insulator crystals or metal crystals.

(9) In addition, in the measurement 'A:' method of the present invention, the wavelength of the Okawa light is changed to increase the transmitted light intensity by 1α゛! rI to I! By determining ll+, it is possible to measure not only the dislocation density distribution but also the single-dislocation density distribution.

In the embodiment described above, light is irradiated alternately at two points on the sample 1- at a constant cycle, but the present invention is not limited to this example, and the chopper slits can be arranged in any number of stages, such as three or four stages. , By placing ^1 over multiple desired stages, light is sequentially applied to multiple points on the sample 1-! 1 may be set. In that case, the difference in transmittance between the outputs adjacent to Ij: among the synchronous rectified outputs of l1ll'l & is calculated as i! ).

Furthermore, the moving direction of the sample, that is, the scanning direction of the light spot, does not need to be fixed in a fixed direction as shown in FIG. It is not necessary to make the intervals constant; if a synchronization signal is generated in synchronization with the scanning, scanning can be performed at variable intervals.

Furthermore, light scattering tomography is a crystal evaluation method that utilizes light scattering at dislocation sites, similar to the method of the present invention, which has the convenience of being able to measure the crystallinity of a sample wafer without cutting or processing it. In this method, a thin light beam 1 parallel to the wave 7X is incident, and a scattered light component perpendicular to the surface is detected. However, in this method, it is necessary to cut the wafer into a rectangle several mm wide and optically polish the cross section, and furthermore, it is difficult to measure thin wafers with a thickness of 1 mm or less. In contrast, the 4111 method of the present invention does not have such drawbacks.

[Brief explanation of the drawing]

FIG. 1 is a block diagram showing an embodiment of an apparatus for measuring dislocation density distribution by carrying out the method of the present invention, and FIG. Diagram for explaining the procedure. Figure wIJ3 is a characteristic curve diagram showing the distribution of signal and transmittance with respect to the scanning distance. FIG. 4 is a characteristic curve diagram showing the distribution of dislocation density with respect to scanning distance X, and pl'% Figure 5 is a correlation curve diagram that reduces the correlation of j large excess light with respect to one piece of dislocation density. 1... Near-infrared tIt colored light rays, 2... Chopper, 3... Lens. 4... Optical mask, 5... Crystal sample, 6... Optical fiber, 6'... Light receiving terminal, 7... Light emitting group, 8... Phototransistor, 8... Period 48 No., lO...Detection Iha No., 11...Synchronous rectification amplifier, 13...Scanning direction, 14...Record A1. 15... Trial 1 transmitted light, 18, 17... Incident light spot. Procedural amendment sound ← J9 year 3n3o1] Commissioner of the Japan Patent Office Kazuo Wakasugi 1, evaluation method of display crystals in the case 3, relationship with the person making the amendment Patent Deseto (435) Japan Broadcasting Corporation (1) Specifications Correct "transmitted intensity" in the λth line of page 1 of the book to "transmitted light intensity". (2) Line 13 on page 13 to line 7θ on page 14I are corrected as follows. The reason why light with an energy smaller than the band gap of 1.4 eV of 1-GaAs is prevented from passing through a single crystal is thought to be due to light scattering by impurity elements precipitated near dislocations. As a result, the light transmittance is low at locations where the dislocation density is high, and conversely, the light transmittance is high at locations where the dislocation density is low. (3) ``1 transmittance dislocation density'' on the 16th line of the same page 1 is corrected to ``passage rate vs. dislocation density''. (4) Page 17, line 2 to line j [(9) Also... ] Delete. that's all.

Claims (1)

[Scope of Claims] l) A light beam with a wavelength transmitted through the crystal sample 4 is transmitted to a plurality of points on the crystal sample +tfi with a constant period/l to 11(
(the light transmitted from the nfl crystal is detected and converted into an electric signal, and a synchronization signal is generated at the constant period 11), and the synchronization signal is A phase-sensitive detection is performed in synchronization with the signal to detect the difference in transmittance between each two of the plurality of points as a signal, and then the above-mentioned light beam is irradiated onto the 111'li+''. Move the plurality of light spots in the /I11'Ii:' direction and stop, and at the stop +1- position l1i
By measuring the difference in transmittance t and sequentially repeating the movement and the If4 constant, the difference in transmittance between the plurality of points is determined as If4 as the interval of the moving distance of the light spot. How to evaluate crystals. 2) The signal detected by performing the above measurement is converted to the signal detected at the position before the movement. +! 2. The method for evaluating a crystal according to claim 1, characterized in that by planting the light spot, the transmittance uM =ij is set as a function of the moving distance of the light spot by /Ill:+i2. 3) For a standard sample having the same chemical composition, impurity content, thickness, and surface condition as the crystal and having a known dislocation density distribution, 4 rate difference from the standard sample l
- to determine the relationship between the dislocation density and the transmittance in advance, then measure the transmittance of the crystal along the same path as the predetermined path, and from the measured value, 43. The method for evaluating a black crystal as set forth in claim 42, wherein a dislocation density distribution is measured using the relationship between the dislocation density and the transmittance. (Margin below)