JPH0982768A - Evaluating method for semiconductor wafer - Google Patents

Abstract

PROBLEM TO BE SOLVED: To make it possible to evaluate the structure and characteristics of a semiconductor wafer by measuring the change of the resistivity caused by a thermal doner generated from interlattice oxygen in the case of annealing the wafer. SOLUTION: The change of the resistivity caused by a thermal doner generated from interlattice oxygen in the case of annealing a semiconductor wafer is measured. For example, an epitaxial wafer that a p-type epitaxial layer is vapor grown on a crystal substrate (100) made of p-type single crystal silicon manufactured by a CZ method is resistance-annealed at 450 deg.C for 60 hours to generate the thermal doner in the substrate. Since the excess thermal doner is generated by the low-temperature annealing, the substrate is inverted from the p-type to the n-type. The resistivity distribution becomes as shown, and the resistivity has a point of inflection (maximum value) at the boundary between the substrate and the epitaxial layer. Accordingly, the thickness of the epitaxial layer can be obtained by knowing the distance from the surface of the epitaxial layer to the point of inflection (maximum value).

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a semiconductor wafer evaluation technique, and more particularly to a technique effective when applied to, for example, film thickness measurement of an epitaxial layer in an epitaxial wafer and oxygen concentration distribution measurement of a silicon wafer.

[0002]

2. Description of the Related Art In a single crystal silicon wafer manufactured by the Czochralski (CZ) method, excess oxygen dissolved from a quartz crucible is taken into the lattice of silicon crystals in the process of pulling up molten silicon in the quartz crucible.
Its concentration is about 1 × 10 ¹⁸ / cm ³ . The oxygen taken into the wafer affects the electrical characteristics of the semiconductor element, such as the breakdown voltage of the gate oxide film of MISFET and the junction leakage current. In particular, oxygen near the surface of the wafer has a great influence on the electrical characteristics. Therefore, the cross section in the wafer
Evaluating the oxygen concentration distribution in the (depth) direction is an important issue in the semiconductor manufacturing process.

To measure the oxygen concentration of a wafer, a Fourier transform infrared spectrophotometer (Fourier transform infrared spectrophotometer) for measuring the absorption spectrum of infrared rays transmitted through the wafer is used.
FTIR) or a secondary ion mass spectrometer that measures secondary ions generated when a wafer is irradiated with high-energy primary ions with a mass spectrometer.
s spectroscopy; SIMS) is used.

On the other hand, in a recent miniaturized MOS device manufacturing process, as a measure for suppressing a latch-up phenomenon peculiar to CMOS, an epitaxial wafer in which an epitaxial layer of single crystal silicon is vapor-phase grown on a silicon wafer is adopted. Is being promoted. When using an epitaxial wafer, it is important to accurately evaluate the film thickness and resistivity of the epitaxial layer.

The known techniques for evaluating an epitaxial layer found by the present inventor are as follows.

JP-A-52-98467 discloses that when the resistivity of an epitaxial layer formed on a silicon wafer is measured by the 4-probe method, the epitaxial layer is sequentially etched by an appropriate thickness by alkali etching. Along with removing
A method for obtaining the resistivity corresponding to each thickness portion and calculating the resistivity of the epitaxial layer based on this result is described.

Japanese Unexamined Patent Publication No. 2-143543 discloses a method of measuring the thickness of an epitaxial layer formed on a silicon wafer by infrared Michelson interferometry.

Japanese Unexamined Patent Publication (Kokai) No. 6-232230 discloses a method of measuring the resistivity of a lower epitaxial layer of two epitaxial layers having different resistivity formed on a silicon wafer.

[0009]

Among the methods for measuring the oxygen concentration of a wafer described above, a Fourier transform infrared spectrophotometer (F
The method using TIR) has a drawback in that the oxygen concentration of the entire wafer can be measured, but the oxygen concentration distribution in the cross-section (depth) direction cannot be measured. Also, the method using a secondary ion mass spectrometer (SIMS) cannot be said to be a simple method because it requires time and skill for analysis, and moreover, compared with the measurement accuracy in the wafer surface, the method in the cross-section (depth) direction Has the disadvantage of low measurement accuracy.

On the other hand, among the above-mentioned methods for evaluating the epitaxial layer, the method for measuring the resistivity described in JP-A-52-98467 determines the interface between the silicon wafer and the epitaxial layer from the reflected light. To obtain a clear interface, it is necessary to change the impurity concentration or conductivity type of the silicon wafer and the epitaxial layer. Therefore, if the silicon wafer and the epitaxial layer have the same conductivity type and the same impurity concentration, the measurement is difficult or impossible.

Further, the method of measuring the thickness of the epitaxial layer (infrared Michelson interferometry) described in Japanese Patent Laid-Open No. 2-143543 discloses that infrared rays are reflected at the interface between the silicon wafer and the epitaxial layer. Since the two conditions that incident / reflected light are efficiently transmitted through the epitaxial layer are required, the impurity concentration of the silicon wafer is 1
In × 10 ¹⁸ / cm ³ or more, and the impurity concentration of the epitaxial layer can not be measured only when it is 1 × 10 ¹⁷ / cm ³ or more.

An object of the present invention is to provide a technique capable of evaluating the structure and characteristics of a semiconductor wafer.

The above and other objects and novel features of the present invention will become apparent from the description of the specification and the accompanying drawings.

[0014]

SUMMARY OF THE INVENTION Among the inventions disclosed in the present application, the outline of a representative one will be briefly described.
It is as follows.

The semiconductor wafer evaluation method of the present invention evaluates the structure of a wafer by measuring the change in resistivity due to a thermal donor generated from interstitial oxygen when the silicon wafer is annealed.

A specific description will be given of the case where the above-described evaluation method of the present invention is applied to the measurement of the thickness of an epitaxial layer in an epitaxial wafer.

About 1 × 10 ¹⁸ / cm ^{3 of} oxygen dissolved out from the quartz crucible when pulling the single crystal is taken in between the crystal lattices of the silicon wafer manufactured by the CZ method, and before being put into the wafer process. The interstitial oxygen is distributed almost uniformly in the wafer. On the other hand, ideally, the epitaxial layer formed on the wafer does not contain oxygen. However, in an actual epitaxial wafer, oxygen in the wafer springs out due to heat during epitaxial growth, so that a small amount of oxygen exists in the epitaxial layer near the interface with the wafer. However, its concentration is two orders of magnitude lower than the oxygen concentration in the wafer.

In the epitaxial wafer as described above, when the wafer and the epitaxial layer have the same conductivity type and the same concentration of impurities constituting the donor (or acceptor), the wafer and the epitaxial layer are separated from each other. Since the only difference is the difference in oxygen concentration, the film thickness of the epitaxial layer cannot be accurately known by the conventional measurement method.

By the way, when a silicon wafer manufactured by the CZ method is annealed at a low temperature near 450 ° C., several oxygen atoms gather to emit one electron (thermal donor) (Kaiser, W., Frisch, HL and Reiss, H. "Mech
anism of the formation of donor states in heat-trea
ted silicon ", Phys. Rev., 112, pp.1546-1554 (Dec.
1958)). The production amount of this thermal donor increases in proportion to the annealing time of the wafer, but the annealing temperature is 60
It is known to disappear at temperatures above 0 ° C.

When the above thermal donor is generated in the silicon wafer, the resistivity of the n-type wafer apparently decreases. On the other hand, the resistivity of the p-type wafer apparently increases, and it may be inverted to the n-type when the annealing time is lengthened. However, since the thermal donor is not generated in the epitaxial layer containing no oxygen, the resistivity does not change before and after the low temperature annealing.

For example, when an epitaxial wafer in which an n-type epitaxial layer is formed on an n-type wafer is annealed at a low temperature, the resistivity of the epitaxial layer on the surface is the same as that before annealing, whereas the resistivity of the inner wafer is Less than before annealing. Therefore, if a decrease in resistivity is observed from a certain depth (x) from the surface when the resistivity is measured along the cross-sectional direction of the epitaxial wafer after low temperature annealing, the epitaxial layer of this wafer is It can be seen that the film thickness is x.

Further, when an epitaxial wafer in which a p-type epitaxial layer is formed on a p-type wafer is annealed at a low temperature, the resistivity of the epitaxial layer on the surface is the same as that before annealing, whereas the resistivity of the inner wafer is low. Is greater than before annealing. Therefore, if resistivity increase is observed from a certain depth (y) from the surface when the resistivity is measured along the cross-sectional direction of the epitaxial wafer after the low temperature annealing, if the resistivity of the epitaxial layer of this wafer is It can be seen that the film thickness is y.

Further, a p-type (or n-type) is formed on the p-type wafer.
When an epitaxial wafer having a (type) epitaxial layer formed thereon is annealed at a low temperature for a long time to invert a p-type wafer into an n-type, the resistivity becomes maximum near the interface between the epitaxial layer and the wafer. Therefore, if the maximum value of the resistivity is observed at a certain depth (z) from the surface when the resistivity is measured along the cross-sectional direction, the film thickness of the epitaxial layer of this wafer is z. I understand.

In an actual epitaxial wafer,
Since a small amount of oxygen also exists in the epitaxial layer near the interface with the wafer due to the oxygen outflow during the epitaxial growth described above, the position of the interface between the epitaxial layer and the wafer and the resistivity change as described above. There is a slight deviation from the location. Therefore, by accurately measuring the concentration distribution of oxygen bubbling in the epitaxial layer and correcting the above error based on this measured value, the accurate film thickness of the epitaxial layer can be known.

To measure the resistivity along the cross-sectional direction of the wafer, for example, the resistivity in the vicinity of the contact surface is determined based on the voltage drop generated when a current is applied to the wafer through a probe that is in point contact with the cross section of the wafer. Spread resistance method (S
preading Resistance Method) etc.

As described above, according to the wafer evaluation method of the present invention, not only the thickness of the epitaxial layer having the conductivity type and the impurity concentration different from that of the wafer but also the conductivity type and the impurity concentration of the wafer are the same as those of the wafer. The thickness of the epitaxial layer, which was difficult or impossible to measure by the method, can be easily and accurately known. It is also possible to estimate the amount of oxygen bubbling into the epitaxial layer from the distribution of resistivity along the cross-sectional direction of the wafer.

The above is an example in the case of being applied to the film thickness measurement of an epitaxial layer in an epitaxial wafer, but the evaluation method of the present invention is not limited to this, and for example, DZ-IG (Denuded Zone-Intrinsic Getter).
ing) The oxygen concentration such as estimating the width of the DZ layer (low oxygen-free region) in the wafer and checking whether the inside of the wafer has an oxygen concentration distribution that can obtain desired electrical characteristics. It can be used to evaluate any wafer structure associated with the.

[0028]

Embodiments of the present invention will now be described in more detail with reference to the drawings.

(Example 1) This example is applied to measurement of the film thickness of an epitaxial layer in an epitaxial wafer.

First, an epitaxial wafer (P on P) obtained by vapor-phase growth of a p-type epitaxial layer (Epi layer) on a (100) crystal plane substrate (wafer) made of p-type single crystal silicon manufactured by the CZ method. ) Is prepared (FIG. 1). The resistivity of both the substrate and the epitaxial layer is 10 Ω · cm, and the oxygen concentration of the substrate is 1 × 10 ¹⁸ / cm ³ . The film thickness of this epitaxial layer is 10 μm, but it is unknown before the measurement.

The above oxygen concentration is determined by JEIDA (Jap
Oxygen concentration conversion factor [(3.03 ± 0.02) x 1 by an Electronic Industry Development Association)
0 ¹⁷ ]] ("Determination of Conversio
n Factor for Infrared Measurement of Oxygen in Sil
icon "J.Electrochem.Soc .; SOLID-STATE SCIENCE AND
TECHNOLOGY, Vol 132, No.7, July 1985.p1707-p171
2).

Cross section (depth) of this epitaxial wafer
The initial oxygen concentration distribution along the direction is shown in FIG. Ideally, there is no oxygen in the epitaxial layer, so it does not appear in the figure. This epitaxial wafer was polished at a predetermined angle from the surface of the epitaxial layer to expose the cross section, and the resistivity was measured using the spreading resistance method.
Shown in Since the substrate and the epitaxial layer have the same conductivity type and the same impurity concentration, the resistivity from the epitaxial layer to the substrate is constant.

Next, this epitaxial wafer is 450
Low temperature annealing is performed at 60 ° C. for 60 hours to generate a thermal donor in the substrate. By performing low temperature annealing for a long time,
The substrate inverts from p-type to n-type as excess thermal donor is produced. Fig. 4 shows the resistivity distribution after low temperature annealing.
The oxygen concentration distribution is shown in FIG. Oxygen in the substrate flows out to the epitaxial layer, so that oxygen also exists in the epitaxial layer near the interface between the two, and the oxygen concentration in the substrate near the interface decreases accordingly. Therefore, thermal donors are also generated in the epitaxial layer near the interface.
Further, in the substrate near the interface, the production amount of thermal donors is reduced due to the reduced oxygen concentration. As a result, the resistivity has an inflection point (maximum value) at the interface between them (Fig. 4). Therefore, the film thickness of the epitaxial layer can be obtained by knowing the distance from the surface of the epitaxial layer to the inflection point (maximum value).

(Example 2) This example is applied to the film thickness measurement of an epitaxial wafer (N on N) in which an n-type epitaxial layer is vapor-deposited on an n-type substrate manufactured by the CZ method. Is. The resistivity of the substrate and the epitaxial layer, the oxygen concentration distribution before the low temperature annealing, and the film thickness of the epitaxial layer are the epitaxial wafer (P on
Same as P).

FIG. 6 shows the oxygen concentration distribution along the cross-sectional (depth) direction of the epitaxial wafer after the low temperature annealing.
Shown in The resistivity of the n-type substrate after the low temperature annealing is reduced due to the generation of the thermal donor. However, since it does not have the inflection point (maximum value) as in the case of the epitaxial wafer (P on P), it is difficult to know the interface between the epitaxial layer and the substrate accurately. It is possible to estimate the thickness.

(Embodiment 3) This embodiment is applied to the measurement of the oxygen concentration distribution along the cross-sectional (depth) direction of the wafer manufactured by the CZ method.

The amount of thermal donors produced by the low temperature annealing depends on the oxygen concentration in the wafer. If low temperature annealing is performed for a sufficient time, the amount of thermal donors generated is uniquely determined by the oxygen concentration, regardless of the annealing temperature and atmosphere. However, since the generation rate of the thermal donor depends on the annealing temperature, the generation amount differs in a short time.

In such a case, the correlation between the annealing temperature and time and the amount of thermal donors generated is obtained in advance using a wafer of known oxygen concentration (W. Kaiser; Electric).
al and Optical Properties of Heat-Treated Silicon
: Phys. Review., 105 (1957) 1751 shows this correlation).

(Embodiment 4) This embodiment is applied to the measurement of oxygen concentration distribution along the cross-section (depth) direction of a CZ wafer which has been annealed at a high temperature of hydrogen, for example, in a hydrogen atmosphere at 1200 ° C. for 1 hour. It was done.

When the wafer is subjected to high-temperature hydrogen annealing, oxygen in the vicinity of the surface is diffused outward, and the oxygen concentration is reduced by about one digit or more as compared with the inside. FIG. 7 schematically shows the cross-sectional structure of a wafer that has been subjected to high temperature hydrogen annealing. 8 shows the oxygen concentration distribution in the cross-sectional direction. Assuming that the oxygen concentration before high temperature hydrogen annealing is 1 × 10 18 / cm 3, after annealing,
The oxygen concentration decreases in the region from the surface to a depth of several 10 μm (50 μm or more). Since the outward diffusion of oxygen occurs on both sides of the wafer, the decrease in oxygen concentration is also seen on the back side of the wafer.

When this wafer is annealed at a low temperature (450 ° C.), a thermal donor proportional to the oxygen concentration is generated, so that the resistivity along the cross-sectional direction was initially uniform, but after low temperature annealing, The pattern wafer changes as shown in FIG. 9A, and the n-type wafer changes as shown in FIG. 9B. Therefore, from this result, the oxygen concentration distribution along the cross-sectional direction can be easily obtained. In the case of a p-type wafer, it may be inverted to n-type by long-time annealing and the apparent resistivity may decrease, so it is necessary to set an appropriate annealing time.

Since the oxygen concentration of the wafer has a great influence on the electrical characteristics of the semiconductor element, it is possible to improve the reliability and manufacturing yield of the LSI by accurately knowing the oxygen concentration distribution along the sectional direction. Further, it is possible to easily confirm whether or not the manufactured wafer has a desired oxygen concentration distribution.

Furthermore, it is not possible to distinguish visually whether the wafer obtained from the outside is an epitaxial wafer or a wafer subjected to high temperature hydrogen annealing. However, in the case of an epitaxial wafer, the low oxygen concentration region exists only on the front surface side, whereas in the case of a wafer subjected to high temperature hydrogen annealing, it also exists on the back surface side. Further, in the case of an epitaxial wafer, the width of the low oxygen concentration region is practically about 20 μm (50 μm or less), whereas in the case of a wafer subjected to high temperature hydrogen annealing, it exists in a width of 50 μm or more. Therefore, if these wafers are annealed at a low temperature, a thermal donor is generated, and the resistivity is measured, it is determined from the oxygen concentration distribution corresponding to the resistivity whether the wafer is an epitaxial wafer or a wafer subjected to high temperature hydrogen annealing. Can be easily identified.

According to this embodiment, the width of the low oxygen concentration region can be easily measured. Further, the oxygen concentration distribution in the wafer surface can be measured by using the four-point probe method or the like.

(Embodiment 5) This embodiment is applied to the measurement of the oxygen concentration distribution of a DZ-IG wafer.

The DZ-IG wafer has a high temperature (1050 ° C.) in an inert gas (nitrogen etc.) atmosphere for forming the DZ layer.
Degree) A wafer obtained by combining annealing and low temperature annealing for forming the IG layer, or by performing these individually.

FIG. 10 shows the resistivity distribution along the cross-sectional direction of the DZ-IG wafer after the low temperature annealing. The DZ-IG wafer has a smaller decrease in oxygen concentration near the surface than a wafer subjected to high temperature hydrogen annealing. That is, the difference in oxygen concentration between the inside and the surface of the wafer is smaller than that of the wafer subjected to the high temperature hydrogen annealing, and the depth thereof is shallow. Therefore,
By measuring the decrease in oxygen concentration on the backside of the wafer, DZ
It is possible to determine whether the wafer is an IG wafer or a wafer subjected to high temperature hydrogen annealing.

In some cases, these wafers cannot be easily distinguished. In that case, it may be determined by comparing with a third wafer in which it is known in advance whether the wafer is a DZ-IG wafer or a wafer subjected to high temperature hydrogen annealing.

According to the present embodiment, the width of the DZ layer can be easily measured, which makes it possible to estimate the gettering effect that affects the electrical characteristics of the semiconductor element. In addition, by using the four-point probe method,
It is also possible to measure the oxygen concentration distribution within the wafer surface.

As described above, the invention made by the inventor has been specifically described based on the embodiments. However, the present invention is not limited to the above-described embodiments, and various modifications can be made without departing from the gist of the invention. Needless to say.

[0051]

Advantageous effects obtained by typical ones of the inventions disclosed in the present application will be briefly described.
It is as follows.

According to the present invention, by measuring a change in resistivity due to a thermal donor generated from interstitial oxygen when a silicon wafer is annealed at a low temperature, for example, a wafer structure such as a film thickness of an epitaxial layer in an epitaxial wafer is measured. Can be evaluated easily and accurately.

[Brief description of drawings]

FIG. 1 is a diagram schematically showing a cross-sectional structure of an epitaxial wafer used in Example 1 of the present invention.

FIG. 2 is a graph showing an initial oxygen concentration distribution of an epitaxial wafer used in Example 1 of the present invention.

FIG. 3 is a graph showing an initial resistivity distribution of the epitaxial wafer used in Example 1 of the present invention.

FIG. 4 is a graph showing a resistivity distribution after low temperature annealing of the epitaxial wafer used in Example 1 of the present invention.

FIG. 5 is a graph showing oxygen concentration distribution after low temperature annealing of the epitaxial wafer used in Example 1 of the present invention.

FIG. 6 is a graph showing an initial resistivity distribution of an epitaxial wafer used in Example 2 of the present invention.

FIG. 7 is a diagram schematically showing a cross-sectional structure of a heat-treated wafer used in Example 4 of the present invention.

FIG. 8 is a graph showing an initial oxygen concentration distribution of a heat-treated wafer used in Example 4 of the present invention.

9A is a graph showing the resistivity distribution of a p-type heat-treated wafer used in Example 4 of the present invention after low temperature annealing, FIG.
(B) is a graph showing the resistivity distribution of the n-type heat-treated wafer after low temperature annealing.

FIG. 10 is a graph showing a resistivity distribution of a DZ-IG wafer used in Example 5 of the present invention after low temperature annealing.

 ─────────────────────────────────────────────────── ─── Continuation of the front page (72) Norio Suzuki Nobuo Suzuki 5-20-1, Josuihoncho, Kodaira-shi, Tokyo Inside the Semiconductor Division, Hitachi, Ltd. (72) Hirofumi Shimizu 5 Josuimotocho, Kodaira-shi, Tokyo Hitachi Co., Ltd. Semiconductor Division, 20-21 Chome (72) Inventor Yushi Sugino 5-201-1, Josuihoncho, Kodaira-shi, Tokyo Inside Hitachi Semiconductor Division, Corporation

Claims (9)

[Claims] 1. A method for evaluating a semiconductor wafer, which comprises measuring a change in resistivity due to a thermal donor generated from interstitial oxygen when the semiconductor wafer is annealed. 2. The method for evaluating a semiconductor wafer according to claim 1, wherein the annealing temperature of the semiconductor wafer is 450.
A method for evaluating a semiconductor wafer, which is characterized in that the temperature is about ° C. 3. The method of evaluating a semiconductor wafer according to claim 1, wherein a change in resistivity along the cross-sectional direction of the semiconductor wafer is measured. 4. The method for evaluating a semiconductor wafer according to claim 1, wherein a change in resistivity in the plane of the semiconductor wafer is measured. 5. The method for evaluating a semiconductor wafer according to claim 1, wherein the semiconductor wafer is an epitaxial wafer. 6. The method for evaluating a semiconductor wafer according to claim 5, wherein the conductivity type and the impurity concentration of the wafer and the epitaxial layer formed on the surface thereof are the same. . 7. The method for evaluating a semiconductor wafer according to claim 1, wherein the semiconductor wafer is a wafer subjected to high temperature hydrogen annealing. 8. The method for evaluating a semiconductor wafer according to claim 1, wherein the semiconductor wafer is a DZ-IG wafer. 9. A (100) produced by the Czochralski method.
What is claimed is: 1. A method for evaluating a semiconductor wafer, comprising: measuring a film thickness of a p-type epitaxial layer grown on a surface of a p-type single crystal silicon wafer having a crystal plane, which comprises annealing a silicon wafer having the epitaxial layer at a predetermined temperature. By generating a thermal donor in the silicon wafer, the resistivity is changed from the surface of the epitaxial layer along the inside of the silicon wafer, and the distance from the surface of the epitaxial layer to the inflection point of the resistivity is obtained. Thus, the method for evaluating a semiconductor wafer is characterized in that the film thickness of the epitaxial layer is measured.