JPH11297704A - Evaluation method for oxygen deposit density - Google Patents

Abstract

PROBLEM TO BE SOLVED: To judge the gettering capability of a wafer and to manufacture a device with high yield. SOLUTION: With this evaluation method, oxygen deposit density in a silicon wafer for a semiconductor device, for which specific resistance inside a substrate is equal to or less than 0.05 Ωand an oxygen density is (10-18)×10<17> atoms/cm<3> (old ASTM method) is evaluated. Even when a deposit size is extremely small in a low temperature process, by subjecting it to heat treatment for just the optimum time respectively in the temperature range of 900-1,100 deg.C and measuring the deposit density by a defect selective etching method, an infrared tomograph method and a transmission electron microscope(TEM) method, the deposit density after the low temperature process is evaluated with high accuracy.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a silicon epitaxial wafer used as a substrate of a circuit element such as an LSI (Large Scale Integrated Circuit) and a method of evaluating oxygen precipitates in a silicon wafer substrate. 11
After the heat treatment in the temperature range of 00 ° C. for the optimum time, the density of the precipitate is measured by the defect selective etching method, the infrared tomography method, and the transmission electron microscope (TEM) method. The present invention relates to a method for evaluating oxygen precipitates, which can evaluate a material density with high accuracy.

[0002]

2. Description of the Related Art High integration of silicon semiconductor devices is progressing rapidly, and characteristics required for silicon wafers are becoming more and more severe. In a highly integrated device, when a crystal defect or a metal impurity other than a dopant is contained in a device active region, a leak current of a P / N junction may be increased or a gate oxide film characteristic of a MOS device may be deteriorated. Are known.

Conventionally, such highly integrated devices include C
CZ-Si wafers grown by the Z method have been used. Supersaturated interstitial oxygen (1
0-18) × 10 ¹⁷ atoms / cm ³ , and it is well known that crystal defects such as oxygen precipitates, dislocations, and stacking faults are induced in the device process.

It is widely known that oxygen precipitates and crystal defects generated inside a wafer sufficiently distant from the device active region have a gettering effect on contaminating heavy metals. For this reason, the presence of oxygen deposits and crystal defects inside the wafer is indispensable for manufacturing devices with high yield.

On the other hand, in the so-called device active region near the wafer surface, during the high-temperature heat treatment at 1100 to 1200 ° C. in the formation of the LOCOS or WELL diffusion layer, the oxygen concentration is greatly reduced due to outward diffusion, and crystal defects are generated. Had been suppressed.

However, with the miniaturization of semiconductor devices,
High energy ion implantation has been used to form the WELL diffusion layer, and the device process has been performed at a low temperature of 1000 ° C. or less in order to make the junction depth shallower. Therefore, outward diffusion of oxygen becomes insufficient, and it becomes difficult to suppress generation of crystal defects in the device active region.

[0007] Under these circumstances, silicon epitaxial wafers in which a high-quality epitaxial layer substantially free of crystal defects is formed on a CZ-Si substrate have come to be widely used in today's highly integrated devices. Was.

[0008]

Boron is doped in the substrate of the invention Problems to be Solved Such epitaxial wafer, the resistivity of the substrate is less p / p ⁺ wafers 0.05Omucm (hereinafter, p / p ⁺
(Referred to as a wafer).

[0009] The gettering ability of contaminated heavy metals is as follows.
It is well known that generally, the higher the oxygen precipitate density, the greater.

However, as the temperature of the device process becomes lower, the size of the oxygen precipitate becomes very small, at most several tens of nm, and the defect selective etching (detection limit is 100 nm)
Degree) and infrared tomography (detection limit is about 30 nm)
It is also difficult to quantitatively evaluate the density of precipitates by the method. Therefore, there is a problem that the gettering ability of the wafer cannot be determined.

A method of judging the gettering ability of a p ^- wafer having a low boron concentration by performing a heat treatment at 1000 ° C. for 16 hours and measuring an oxygen precipitation amount by an infrared absorption method has been reported. (Hiroshi Takeno et al., Proceedings of the 58th Annual Conference of the Japan Society of Applied Physics, 2p-N-15 (19
97)).

However, in a p / p ⁺ wafer, the amount of precipitated oxygen cannot be measured by the infrared absorption method because infrared rays are absorbed by free carriers in the substrate.

[0013] The present invention has been made in order to overcome the above problems.
A silicon epitaxial film is provided on the surface, the specific resistance inside the substrate is 0.05 Ωcm or less, and the oxygen concentration is (10 to 1).
8) × 10 ¹⁷ atoms / cm ³ (old ASTM
It is an object of the present invention to provide a method for evaluating the density of oxygen precipitates in a silicon wafer for a semiconductor device according to the above method, and to enable determination of the gettering ability of the wafer.

[0014]

SUMMARY OF THE INVENTION In order to evaluate the density of oxygen precipitates, the inventors first set the specific resistance of a substrate at 0.0%.
5ΩCm or less, oxygen concentration is (10-18) × 10 ¹⁷ at
oms / cm ³ (old ASTM method) p / p + wafers are subjected to low-temperature process heat treatment, and then subjected to additional heat treatments under various conditions to grow oxygen precipitates inside the wafers and to apply various evaluation methods for deposition. The possibility of quantitative evaluation of material density was investigated.

As a result of the above investigation, the inventors found that (1) 90
After performing heat treatment for 4 hours or more in a temperature range of 0 to 1100 ° C., measuring the precipitate density by a defect selective etching method; (2) performing heat treatment for 1 hour or more in a temperature range of 900 to 1100 ° C. Then, the precipitate density is measured by an infrared tomography method. (3) After performing a heat treatment for 0.5 hour or more in a temperature range of 900 to 1100 ° C., the precipitate density is measured by a transmission electron microscope (TEM) method. The present inventors have found that the density of oxygen precipitates after the low-temperature process can be evaluated with high accuracy by measuring the temperature, and completed the present invention.

That is, the present invention has a silicon epitaxial film on the surface and the specific resistance inside the substrate is 0.05.
Ωcm or less, oxygen concentration is (10-18) × 10 ¹⁷ atom
A method for evaluating the density of oxygen precipitates in a silicon wafer for semiconductor devices of ms / cm ³ (old ASTM method), wherein the heat treatment is performed in a temperature range of 900 to 1100 ° C. .

[0017]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS In the present invention, the characteristics of a silicon wafer to be evaluated are set as follows.
Hereinafter, when the oxygen concentration is (10-18) × 10 ¹⁷ atoms /
Limited to cm ³ (old ASTM method), such wafers have very small oxygen precipitates,
This cannot be detected by defect selective etching or infrared tomography, and the heat treatment of the present invention solves this problem.

In the present invention, prior to the evaluation, 900
The heat treatment in the temperature range of １1100 ° C. is for setting the size of the oxygen precipitate to a required size. However, the atmosphere may be oxygen, argon, or the like in addition to the nitrogen atmosphere of the example. However, it is not preferable because it takes a long time for the growth to be slow and the oxygen precipitates to be
If the temperature exceeds ℃, some of the oxygen precipitates form a solution, which makes it difficult to quantitatively evaluate the density.

In the present invention, the heat treatment is performed in a temperature range of 900 to 1100 ° C. In order to measure the precipitate density by the defect selective etching method, the size of the oxygen precipitate is 100 n.
m, a heat treatment of at least 4 hours is necessary, and the size of the oxygen precipitate needs to exceed 30 nm to measure the precipitate density by the infrared tomography method. Heat treatment is required, and in order to measure the precipitate density by a transmission electron microscope (TEM) method, the size of the oxygen precipitate needs to exceed 10 nm, and a heat treatment of at least 0.5 hour is required.

In the present invention, as the method for measuring the precipitate density, preferable conditions are 1000 ° C. for 8 hours for the defect selective etching method and 1 hour for the infrared tomography method.
At 000 ° C. for 4 hours, and in the case of a transmission electron microscope (TEM) method, at 1000 ° C. for 1 hour.

[0021]

EXAMPLES Comparative Example 1 A specific resistance of 10 mΩcm and an oxygen concentration of 15.5 × 10 ¹⁷ a
CZ-Si of toms / cm ³ (old ASTM method)
A wafer was prepared and a sample was prepared according to the following procedure. The wafer was baked in a hydrogen atmosphere at 1150 ° C. for 80 seconds in a lamp heating type horizontal CVD epitaxial apparatus, and then an epitaxial film was formed. The film formation process uses trichlorosilane as a source gas and 10
A 4 μm epitaxial film was formed at 80 ° C.

Next, the p / p ⁺ wafer was subjected to standard low-temperature process heat treatment (maximum temperature 1000 ° C.) to obtain a sample. Immediately after low-temperature process, defect selective etching with light etchant, infrared tomography and TEM
The oxygen precipitate density was evaluated by the method. However, none of the methods confirmed the presence of oxygen precipitates.

Example 1 Next, an example of a method for evaluating the density of oxygen precipitates according to the present invention will be described. First, for the purpose of growing oxygen precipitates that cannot be observed immediately after the low-temperature process, 800
Heat treatment was performed at a temperature of 900C, 900C, 1000C, 1100C, and 1200C for 5 hours in a nitrogen atmosphere, and the precipitate density after the heat treatment was measured by a defect selective etching method. FIG. 1 shows the results.

Thus, the density of the deposit is 900 to 1100.
It can be seen that the temperature is almost constant up to ℃ and is about 1 × 10 ⁷ / cm ² , while the value at 800 ° C. and 1200 ° C. is several orders of magnitude smaller. It is considered that this is because the growth of the oxygen precipitate was slow at 800 ° C. and the oxygen precipitate was solutionized at 1200 ° C. Therefore, the temperature of the heat treatment for exposing oxygen precipitates is 900 to 1
It can be said that 100 ° C. is optimal. No oxygen precipitate is newly generated by the heat treatment in this temperature range.

Next, examples of the relationship between the heat treatment time and the quantitative evaluation by various evaluation methods at the heat treatment temperature included in this temperature range will be described. FIG. 2 shows the results of measuring the oxygen precipitate density by the defect selective etching method, the infrared tomography method, and the TEM method when the additional heat treatment time at 1000 ° C. was changed.

As shown in FIG. 2, the defect selective etching method is about 4
More than 1 hour in infrared tomography,
It can be seen that in the EM method, the precipitate density is saturated after about 0.5 hours or more, and the precipitate is hardly observed in each time less than about 0.5 hour. The measurement result of the defect selective etching method is converted into a volume density in consideration of an etching amount.

From the above results, the defect selective etching method requires 4 hours or more, and the infrared tomography method requires 1 hour or more.
In the TEM method, it can be said that the precipitate density can be measured with high accuracy by heat treatment for about 0.5 hours or more.

The same study was carried out for CZ-Si wafers and p / p-epi wafers having a low boron concentration. However, the oxygen precipitate density can be measured with high accuracy by the heat treatment conditions and the evaluation method according to the present invention. I knew there was.

[0029]

According to the method for evaluating the density of oxygen precipitates according to the present invention, even when the size of precipitates is very small in a low-temperature process, heat treatment is performed for an optimum time in a temperature range of 900 to 1100 ° C. After that, by measuring the precipitate density by the defect selective etching method, infrared tomography method, and transmission electron microscope (TEM) method, the precipitate density after the low temperature process can be evaluated with high accuracy, and the gettering ability of the wafer can be evaluated. It is possible to make a decision and manufacture a device with a high yield.

[Brief description of the drawings]

FIG. 1 is a graph showing a relationship between a heat treatment temperature after a low-temperature process and a precipitate density.

FIG. 2 is a graph showing a relationship between a heat treatment time after a low-temperature process and a precipitate density.

Claims (4)

[Claims] 1. A substrate having a silicon epitaxial film on its surface, a specific resistance inside the substrate of 0.05 Ωcm or less, and an oxygen concentration of (10 to 18) × 10 ¹⁷ atoms / cm ³ (old
ASTM method) for evaluating the density of oxygen precipitates in a silicon wafer for semiconductor devices.
A method for evaluating the density of oxygen precipitates which is heat-treated in a temperature range of 1 to 1100 ° C. 2. The method according to claim 1, wherein the temperature is 900 to 1100 ° C.
A heat treatment for 4 hours or more in the above temperature range, and then measuring the precipitate density by a defect selective etching method. 3. The method according to claim 1, wherein the temperature is 900 to 1100 ° C.
After performing heat treatment for 1 hour or more in the above temperature range, the density of oxygen precipitates is measured by infrared tomography. 4. The method according to claim 1, wherein the temperature is 900 to 1100 ° C.
After performing a heat treatment for 0.5 hour or more in the temperature range described above, the density of precipitates is measured by transmission electron microscopy.