JP2761055B2 - Silicon wafer and inspection method of silicon wafer - Google Patents

Description

The present invention relates to a silicon wafer and a method for inspecting a silicon wafer.

(Prior Art) In recent years, with the progress of high integration and high miniaturization of LSI, the performance required for a silicon wafer as a substrate has been
It is becoming more and more severe. For this reason, at the stage of pulling a silicon single crystal, reduction of the amount of impurities, optimization of the interstitial oxygen concentration, etc. are performed, and at the stage of processing, various improvements such as improvement of wafer flatness are performed. Some improvement in product yield has been achieved.

However, even when a semiconductor product is manufactured using a wafer to which these improvements have been added, a variation in the yield which is considered to be caused by the wafer occurs, and the overall yield does not tend to be improved at all. It is known that this tendency is remarkable in an N-type wafer.

(Problems to be Solved by the Invention) One of the causes of the variation in the yield due to the wafer is a stacking fault generated when the wafer is oxidized at a high temperature. Among stacking faults, those that occur particularly on the wafer surface are OSF (oxidation-induce).
d stacking fault), which is known to have a direct adverse effect on device characteristics.

Therefore, in order to reduce the variation in yield due to the wafer and to achieve a high yield as a whole,
It is necessary to measure the OSF density (units / cm ² ), which indicates how much OSF is present, and eliminate wafers with high OSF density.

Conventionally, OSF density is measured by performing selective chemical etching on the wafer surface after performing oxidation treatment. Since linear pits (OSF) appear on the wafer surface, the number of linear pits in a predetermined area is determined using an optical microscope. I was going by counting. However, since this measurement method is a destructive evaluation method that requires etching, and requires a visual inspection with extremely fatigue using an optical microscope, the OSF density is measured without etching or visual inspection. Was desired.

The object of the present invention is to propose a method for determining the OSF density without directly measuring the OSF density by a destructive evaluation method requiring etching and a visual inspection using an optical microscope. The purpose of the present invention is to reduce the variation in the yield that occurs when a semiconductor product is manufactured by using the method, and to achieve a high yield as a whole.

[Structure of the Invention] (Means for Solving the Problems) In order to solve the above problems, the present invention provides a method in which a silicon wafer is formed from a block of a silicon single crystal ingot.
And taking out the second test sample and placing the first test sample in an oxygen atmosphere at a temperature of 950 to 1100 ° C. for 15 to 20 minutes.
The interstitial oxygen concentration before the first heat treatment at the time of _performing the first heat treatment for a time is [Oi] _1ini ,
The interstitial oxygen concentration after the heat treatment is [Oi] _1af ,
750-800 in the second test sample in an oxygen atmosphere
After the second heat treatment at a temperature of 3 ° C. for 3 to 5 hours, the second test sample is successively placed in an oxygen atmosphere at 950 to 1
When the third heat treatment is performed at a temperature of 100 ° C. for 15 to 20 hours, the interstitial oxygen concentration before the second and third heat treatments is [Oi] _2ini, and after the second and third heat treatments When the interstitial oxygen concentration at is [Oi] _2af , (Means 1).

Further, according to the present invention, as a silicon wafer inspection method, first and second inspection samples are taken out from a block of a silicon single crystal ingot, and the interstitial oxygen concentration [Oi] _{1ini of} the first inspection sample and the first inspection sample are extracted. The interstitial oxygen concentration [Oi] _2ini of the second test sample was measured, and after _performing the first heat treatment on the first test sample in an oxygen atmosphere, the interstitial oxygen concentration of the first test sample was measured. Oxygen concentration [Oi]
_1af and measure the second in an oxygen atmosphere.
After the second heat treatment is performed on the test sample of the above, the third heat treatment is continuously performed on the second test sample in an oxygen atmosphere, and then the interstitial oxygen concentration [O
i] _{Measure 2af} and use the formula By calculating the [Oi] change rate defined by the above, a measure is taken to determine the quality of the silicon wafer obtained from the block of the silicon single crystal ingot (claim 2).

Further, the present invention provides a method of inspecting a silicon wafer, wherein the first heat treatment is performed at a temperature of 950 to 1100 ° C.
To 20 hours, the conditions of the second heat treatment are 750 to 800 ° C. for 3 to 5 hours, and the conditions of the third heat treatment are 950
At a temperature of 11100 ° C. for 15 to 20 hours, a measure was taken such that the silicon wafer obtained from the block of the silicon single crystal ingot having the [Oi] change rate of 20 or more was determined to be good. (Claim 3).

(Example) Hereinafter, an example of the present invention will be described in detail.

First, for example, N-type, which is raised and created by the CZ method,
Specific resistance 1-5 [Ωcm], interstitial oxygen concentration 14.0-18.0 ×
10 ¹⁷ [atoms / cm ³ ] (former ASTM display), 10 silicon single crystal ingots (A to J) with various improvements of 5 inches in diameter are prepared. Next, two arbitrary portions are cut out from each ingot with a width of about 30 [cm] to form 20 silicon single crystal blocks (A1, A2 to J1, J2).

In addition, 25 single-sided mirror-surfaced silicon wafers are prepared from each cut block. Further, one of the 25 single-sided mirror-faced silicon wafers to be a measurement sample (sample) is taken out and cleaved into two half-split wafers. Then, these two measurement samples are used for two kinds of wafer lattice oxygen concentration measurement for providing the silicon wafer according to the present invention, and the following two kinds of measurement are performed.
Used for OSF (Oxidation-Induced Stacking Fault) measurement.

As the first measurement of the interstitial oxygen concentration of the wafer, first, for example, FT-IR
The interstitial oxygen concentration is measured using a (Fourier transform infrared absorption spectrometer) (old ASTM display, for example, measuring one point near the center of the wafer). In an oxygen atmosphere, about 1000 ° C, about 16
After performing the oxidation heat treatment (first heat treatment) for a long time, for example,
The interstitial oxygen concentration after the oxidation heat treatment is measured using FT-IR. The interstitial oxygen concentration of the wafer before the oxidizing heat treatment was set to [Oi] _1ini, and the interstitial oxygen concentration of the wafer after the oxidizing heat treatment was set to [Oi] _1af
And

Further, as a first OSF measurement, an oxide film formed on the surface of the silicon wafer subjected to the oxidizing heat treatment is removed using hydrofluoric acid. Further, the silicon wafer is subjected to overall etching (wright etching) by about 2 μm. After this,
Using an optical microscope, measure the OSF density of the silicon wafer (hereinafter, the OSF density thus measured is referred to as “O
SF density 1 ”. ).

Next, as a second wafer interstitial oxygen concentration measurement, first, for example, the other of the two prepared measurement samples, for example,
The interstitial oxygen concentration is measured using FT-IR. After performing an oxidation heat treatment (second heat treatment) at about 780 ° C. for about 3 hours in an oxygen atmosphere, the heat treatment is continuously performed in an oxygen atmosphere for about 10 hours.
An oxidation heat treatment (third heat treatment) is performed at 00 ° C. for about 16 hours.
Then, the interstitial oxygen concentration after the second and third heat treatments is measured using, for example, FT-IR. The interstitial oxygen concentration of the wafer before the second and third heat treatments is [Oi] _2ini, and the interstitial oxygen concentration of the wafer after the second and third heat treatments is [Oi]. _2af .

As a second OSF measurement, an oxide film formed on the surface of the silicon wafer subjected to the second and third heat treatments is removed using hydrofluoric acid. Further, the silicon wafer is subjected to overall etching (wright etching) by about 2 μm. Then, using an optical microscope, the silicon wafer
The OSF density is measured (hereinafter, the O
The SF density is called “OSF density 2”. ).

The first and second wafer inter-lattice oxygen concentrations were similarly measured for the measurement samples relating to all the blocks (A1, A2 to J1, J2), and [Oi] _1ini , [Oi] _1af , [ O
i] _2ini and [Oi] _2af are measured. From this measurement result, the [Oi] change rate of each block ( _{ブ ロ ッ ク} ([Oi] _2ini −
[Oi] _2af ) / ([Oi] _1ini- [Oi] _1af )) is calculated. Further, the first and second OSF measurements are performed, and the OSF density 1 and the OSF density 2 are measured.

FIG. 1 shows these measurement results and calculation results collectively.

In the measurement results under the first OSF measurement conditions, the OSF density 1 is a small value for all wafers, and at first glance, the effect of improving the silicon wafer in the stage of pulling up the silicon single crystal and the stage of processing is recognized. It seems. However, the measurement results under the conditions of the second OSF measurement show that the OSF density 2 varies considerably from a large value (78.2 pieces / cm ² ) to a small value (0.7 pieces / cm ² ). . This means that since the inspection of the silicon wafer is not complete with only the first OSF measurement, there is a sufficient possibility that crystal defects such as dislocations and OSFs will occur in the actual manufacturing stage.

FIG. 2 shows the relationship between the [Oi] change rate of the silicon wafer and the OSF density 2.

According to the figure, [Oi] change rate (however, [Oi] _1ini =
[Oi] In the case of _1af , it is + ∞. ) As a parameter, the OSF density (OSF density) under the condition of the second OSF measurement
Density 2) can be estimated. According to this, the OSF density of silicon wafers with [Oi] change rate of 20 or more 2
However, it can be seen that it was 5 or less / cm ² . Also, OSF
The density 2 sharply increases below the [Oi] change rate 20 and becomes unfavorable for a silicon wafer.

That is, the following can be understood from the results shown in FIG. 1 and FIG.

First, since the [Oi] change rate and the OSF density 2 have a relationship as shown in FIG. 2, for example, using the FT-IR which is a non-destructive evaluation method, only the [Oi] change rate is used. If you ask for the OS
The OSF density 2 can be grasped without directly obtaining the F density 2. Second, OSF density 2 from the second OSF measurement
Increases rapidly when the [Oi] change rate as a parameter falls below the boundary of 20. For this reason, by using a silicon wafer having a [Oi] change rate of 20 or more in actual manufacturing, the variation in yield that occurs during the manufacture of semiconductor products is reduced, and high yield is expected as a whole. A silicon wafer created from the same block of a silicon single crystal ingot (a block of about 30 cm width that is arbitrarily cut from a silicon single crystal ingot) has substantially the same measurement results as the measurement sample. May be considered to be obtained.

Next, the silicon wafer (Oi) having a [Oi] change rate of 20 or more prepared in this manner (24 wafers excluding one measurement sample) was used.
4 megabit mask ROM using 10 sets (NO.1 to NO.10)
(Product of the present invention) was prepared. In addition, a silicon wafer based on only the first OSF measurement (24 wafers excluding one measurement sample)
4 bit mask ROM using 10 sets (NO.1 to NO.10)
(Comparative product) was prepared, and the difference in yield from the product of the present invention was examined.

FIG. 3 is a yield comparison diagram showing the yield of the product of the present invention as a relative yield to a comparative product. Here, the scale 1 on the vertical axis indicates the average yield of the comparative product.

According to the figure, it can be confirmed that the yield of the product of the present invention is remarkably stable near the maximum yield of the comparative product, while the yield of the comparative product varies. In addition, it can be seen indirectly that the OSF density (OSF density 2) obtained from the [Oi] change rate is always kept close to the actual value.

By the way, in the embodiment, in order to perform the first and second inter-lattice oxygen concentration measurements and the first and second OSF measurements, one silicon wafer is divided into half and two measurements are performed. A sample for use was prepared, but two silicon wafers prepared from silicon single crystal blocks were subjected to the first and second interstitial oxygen concentration measurements and the first and second OSF measurements, respectively. It may be a sample for use.

Also, in the first wafer interstitial oxygen concentration measurement,
The heat treatment temperature is not limited to about 1000C, but may be in the range of 950C to 1100C. In the second wafer interstitial oxygen concentration measurement, the temperature in the first oxidation heat treatment was about 78 ° C.
The temperature is not limited to 0 ° C, and may be in the range of 750 ° C to 800 ° C,
Further, the temperature in the second oxidation heat treatment is not limited to about 1000 ° C, but may be in the range of 950 ° C to 1100 ° C.

Furthermore, the first to third heat treatments in the first and second wafer interstitial oxygen concentration measurements were performed in an oxygen atmosphere, but there is no problem if performed in a nitrogen atmosphere.

[Effects of the Invention] As described above, according to the present invention, the following effects are produced.

Even if the OSF density is not directly measured by a destructive evaluation method requiring etching and a visual inspection using an optical microscope, for example, using the nondestructive evaluation method FT-IR [O
i] OSF density can be grasped by calculating the rate of change. Therefore, the quality of the silicon wafer obtained from the block of the silicon single crystal ingot can be determined without etching or visual inspection.

According to FIG. 2, when the rate of change of [Oi] as a parameter exceeds 20, the OSF density rapidly decreases to 5 pieces / cm ² or less. Therefore, the [Oi] rate of change was determined without directly measuring the OSF density.
If a semiconductor product is manufactured using the above-described silicon wafer, variations in the yield that occur during the manufacturing can be reduced, and a high yield can be achieved as a whole (claims 1 and 3).

[Brief description of the drawings]

FIG. 1 is a diagram collectively showing measurement results and calculation results relating to the first and second interstitial oxygen concentration measurements and the first and second OSF measurements, and FIG. FIG. 3 is a view showing the relationship between the change rate and the OSF density 2. FIG. 3 is a yield comparison diagram showing the yield of the product of the present invention as a relative yield to a comparative product.

Claims (3)

(57) [Claims] A first and a second test sample are taken out of a block of a silicon single crystal ingot, and the first and the second test samples are removed in an oxygen atmosphere.
When the first heat treatment was performed at a temperature of 0 ° C. for 15 to 20 hours, the interstitial oxygen concentration before the first heat treatment was changed to [O
i] _1ini , the interstitial oxygen concentration after the first heat treatment is [Oi] _1af, and the second test sample is 750 to 800 in an oxygen atmosphere.
After the second heat treatment at a temperature of 3 ° C. for 3 to 5 hours, the second test sample is successively placed in an oxygen atmosphere at 950 to 1
When the third heat treatment is performed at a temperature of 100 ° C. for 15 to 20 hours, the interstitial oxygen concentration before the second and third heat treatments is [Oi] _2ini, and after the second and third heat treatments When the interstitial oxygen concentration at is [Oi] _2af , A silicon wafer for manufacturing a semiconductor product, obtained from the block of the silicon single crystal ingot, which satisfies the condition shown by: 2. A first and a second test sample are taken out from a block of a silicon single crystal ingot, and the interstitial oxygen concentration [Oi] _{1ini of} the first test sample and the lattice of the second test sample are taken out. The interstitial oxygen concentration [Oi] _1af is measured after the first test sample is subjected to a first heat treatment in an oxygen atmosphere, and the interstitial oxygen concentration [Oi] _1af is measured. And performing a second heat treatment on the second test sample in an oxygen atmosphere, and then subjecting the second heat treatment to a third heat treatment on the second test sample in an oxygen atmosphere. The interstitial oxygen concentration [Oi] _2af of the test sample was measured, and the following formula was used. A silicon wafer inspection method for determining the quality of a silicon wafer obtained from the block of the silicon single crystal ingot by calculating the [Oi] change rate defined by: 3. The condition of the first heat treatment is a temperature of 950-1100 ° C. for 15-20 hours, and the condition of the second heat treatment is 750.
When the temperature of 800800 ° C. is 3 to 5 hours and the condition of the third heat treatment is 950 to 1100 ° C. for 15 to 20 hours, the silicon single crystal ingot having the [Oi] change rate of 20 or more 3. The method for inspecting a silicon wafer according to claim 2, wherein the silicon wafer obtained from said block is determined to be good.