JPH0624217B2 - Method of measuring specific resistance - Google Patents

Description

Detailed Description of the Related Art As a method for measuring the purity of a silicon (Si) -based gas, G
There are methods for directly analyzing gas such as C, IR, GC-MS, atomic absorption, and chemical analysis. Separately, a method of once manufacturing a single crystal thin film from a Si-based gas, measuring the purity of the single crystal thin film from its electrical properties, and indirectly evaluating the purity of the Si-based gas that is the raw material from the value of the purity is also available. is there.

With this electrical method, it is possible to perform ultra-trace analysis, which is far beyond direct gas analysis.

2. The dopant in the gas can be divided into P and N 2 types and measured in total.

3. Since part of the semiconductor manufacturing process is used, it is possible to make a judgment directly related to the product.

And so on.

As a method of epitaxially growing a single crystal thin film,
Generally, the CVD method is used.

As a method for measuring the electrical properties of a single crystal thin film, V
(Voltage), I (current) probe method and V (voltage),
There is a C-V method for measuring C (capacity).

The electrically measured purity of a single crystal thin film is an indication of the amount of electrically active components in the thin film, and generally the active components are called dopants. Dopants are classified into P type and N type according to the electric conduction type. The dominant side dopant determines the electrical conductivity type of the single crystal thin film.

For example, in the probe method, the amount of dopant in the single crystal thin film is
It is immediately calculated from the specific resistance value (Ωcm) calculated from the VI characteristics.

Therefore, generally, the electric conductivity type and the specific resistance value of the single crystal thin film are used as they are to indicate the purity of the electric silicon-based gas as a raw material.

Problems to be Solved by the Invention As a method for measuring the purity of a silicon-based gas, a single crystal thin film is formed from the silicon-based gas by epitaxial growth, and the purity of the silicon-based gas is calculated backward by the electrical properties of the thin film. The method to do is an essential rule for proper process control in the semiconductor industry.

However, the conventional measuring device is complicated and expensive, and
There is a problem that the pretreatment of the sample also requires a long process.

For example, the C to V method, which determines the purity of the thin film and hence the purity of the silicon-based gas from the relationship of V to C, requires complicated steps and expensive equipment for manufacturing the PN junction and the electrode as the pretreatment of the sample. In addition, the C-V characteristic measuring device is also expensive, and one sample needs to be measured for at least 3 days.

Further, as a method simpler than the CV method, there is a probe method for determining the purity of the thin film and hence the purity of the silicon-based gas from the above-mentioned V to I relationships.

Among them, the 2-probe method (spreading resistance method) and the 6-probe method are
The device is expensive and complicated to handle, and the simplest method, the four-probe method, is widely used. Now, in the two-probe method, the relationship between the electric conductivity of the single crystal thin film and the electric conductivity of the Si substrate is
Although there is no direct problem, in this four-probe method, in order to eliminate the influence of the substrate and measure the V to I characteristics of the single crystal thin film, the electrical conductivity types of the single crystal thin film and the substrate must be reversed. There is a problem.

For example, since a single crystal thin film generally made of monosilane gas is generally N type as an electric conduction type, it is necessary to always use a P type Si- substrate and epitaxially grow the single crystal thin film on it.

Now, in the case of growing epitaxially on a Si-substrate by the CVD method, the Si-substrate is usually heated to 1000 ° C or higher.
Then, a large amount of dopant (B as P type, P, As as N type, etc.) in the substrate escapes and mixes into the single crystal thin film during epitaxial growth.

Even if such a phenomenon of mixing of dopants occurred, when the monosilane gas had a low purity, the measurement could be performed without significantly affecting the specific resistance value.

However, purification technology has been developed and the purity is at least 100 Ωc.
When ultrahigh-purity monosilane of m, usually 1000 to 5000 Ωcm, must be targeted, it is necessary to consider that the amount of dopant in the single crystal thin film is very small and is generally N-type. Contamination by the N-type dopant merely lowers the specific resistance value and the phenomenon is simple. However, contamination by the P-type dopant has a faster escape rate for the P-type dopant than that for the N-type dopant, and a single crystal thin film. The phenomenon after contamination of the single crystal is also complicated, and the conduction type of the single crystal thin film becomes a mixed P, N type or a P inversion type, and the specific resistance value becomes very large even if it is not inverted.
It becomes impossible to measure by type. Therefore, P is the simplest method for the purity analysis of ultra-high purity silicon-based gas.
It was almost impossible to put a single crystal thin film into production using a type Si substrate and to put the method of measuring the specific resistance value by the four-probe method into practical use.

The present invention provides a method that eliminates the difficulties of the prior art.

That is, it is used to measure the specific resistance value of ultra-high purity silicon-based gas.
By heating the Si substrate in advance at a temperature higher than or equal to the maximum temperature used in all the epitaxial steps, more preferably at a temperature higher than that, the dopant existing near the surface of the Si substrate is allowed to escape in advance, and then the dry hydrochloric acid etching is performed. , The backside of the Si substrate is coated with ultra-high-purity Si-based gas poly-Si to thoroughly prevent the escape of the dopant from the Si substrate and to grow epitaxially a single-crystal thin film without contamination. According to the present invention, for the first time, ultrahigh-purity Si-based gas, which has been impossible by the conventional method, can be measured easily, inexpensively, and accurately.

OBJECTS OF THE INVENTION That is, an object of the present invention is to provide a simple, inexpensive and accurate method of measuring a specific resistance value of ultra-high purity silicon-based gas.

Means for Solving the Problems That is, in the present invention, in evaluating the purity of a silicon-based gas, the specific resistance value of a single crystal thin film epitaxially grown on a silicon substrate using the silicon-based gas is probed. In the method of measurement by the method, first, the silicon substrate is heat-treated at a temperature higher than the maximum temperature used in all the processes up to epitaxial growth, and the susceptor is coated with the above-mentioned silicon-based polysilicon, and the treatment is performed. The treated silicon substrate is etched with dry hydrochloric acid on a susceptor, and then a single crystal thin film is epitaxially grown on the silicon substrate using the silicon-based gas to measure the specific resistance of the single crystal thin film. A method for measuring a specific resistance value for evaluating the purity of a high-purity silicon-based gas, which is characterized in that It

The present invention will be described in more detail below.

In the present invention, the silicon-based gas means monosilane (SiH ₄ ), disilane (Si ₂ H ₆ ), trisilane (S
i ₃ H ₈ ) and other silicon hydrides; silicon tetrachloride (SiCl ₄ ), hexachlorodisilane (Si ₂ C
l ₆ ), dichlorosilane (SiH ₂ Cl ₂ ), trichlorosilane (SiH
Cl ₃₎ silicon chloride or the like; tetrafluorosilane (SiF _4), Jifurorushiran (SiH
Preferred examples include silicon fluorides such as ₂ F ₂ ), trifluorosilane (SiHF ₃ ), etc., but other types of silicon, such as silicon single crystal, can be obtained on the Si substrate by epitaxial growth. Any silicon-based gas may be used as long as it can be used as a raw material for forming a thin film.

Although referred to as a gas here, what is in a gaseous state at the time of use even if it is liquid at room temperature is included in the silicon-based gas.

The ultrahigh purity referred to in the present invention means that the value represented by the specific resistance value is at least about 100 Ωcm, and usually 1000 Ωcm or more.

In the present invention, an epitaxial growth method is used, but this epitaxial growth method is usually C
This is called the VD method.

The apparatus for growing the epitaxy may be a commonly used method. Examples of the heating method include high frequency heating and resistance heating. The reaction tube is preferably made of quartz, and may be vertical or horizontal. The susceptor is made of quartz, Si plate, or graphite, and a graphite susceptor coated with SiC is particularly preferable.

This susceptor is polysilicon-coated with a silicon-based gas which is used as a material for purity analysis.

As will be described later, a part of this polysilicon is deposited on the back surface of the Si substrate during etching with dry hydrochloric acid to cover the back surface of the Si substrate and prevent the escape of the dopant from the inside of the Si substrate and the contamination of the single crystal Si substrate. Acts to prevent.

The crystal type of the Si substrate may be (100) or (111). The electric conduction type may be either P type or N type in the case of the 2-probe method, but when the 4-probe method is used, it must be opposite to the conduction type of the single crystal thin film as described above. Therefore, it is necessary that the Si substrate and the epitaxially grown single crystal thin film on the Si substrate form a PN junction.

The specific resistance value may be any available value, but if it is too low, the heat treatment time of the Si substrate must be extended, and all the escaped dopant is exhausted and pollutes the reaction system. On the other hand, if it is too high, the contamination from the substrate is small, but the boundary surface between the single crystal thin film and the Si substrate is not clear, and it becomes difficult to measure the thickness of the single crystal thin film. For example, P-
The stain on the N-joint surface does not go well, and the film thickness measurement by the driller method becomes impossible.

Therefore, usually 10 to 500 Ωcm is preferable.

The conditions for producing a single crystal thin film by CVD may be any known method, for example, diluting a silicon-based gas to about 0.01 to 0.5% with a diluting gas such as H ₂ and passing it on a Si substrate heated by a susceptor. To add. The temperature of the Si substrate at this time is usually about 1000 to 1100 ° C.

The epitaxially grown single-crystal thin film should of course have no defects or abnormal precipitation, but it may be present in a very small amount so as not to affect the analysis value.

The film thickness of the single crystal thin film is not particularly limited, but is preferably in the range of 5 to 50 μ in view of the production speed, the size and number of defect shapes.

As the probe method for measuring the single crystal thin film obtained by the method of the present invention, the 4-probe method is simple and most preferable, but the 2-probe method does not cause any problem even if the measurement is performed.

Needless to say, in order to produce a complete single crystal thin film by epitaxial growth, the operating conditions of the CVD apparatus must be set to the highest level, but pretreatment of the Si substrate is also important.

The Si substrate is preferably cleaned with a chemical agent by using a combination of sulfuric acid, nitric acid, hydrofluoric acid, hydrochloric acid, ammonia, hydrogen peroxide and the like.

Thereafter, the Si substrate is a feature of the present invention installed in a quartz reaction tube, heat treatment at a temperature higher than the maximum temperature used in all steps up to epitaxial growth, more preferably above this temperature, and then, Etching with dry hydrochloric acid is performed.

Although the surface of the Si substrate is etched in this etching process, the poly-Si layer previously coated on the SiC-coated graphite susceptor is transferred to the back surface of the substrate, resulting in the escape of the dopant from the back surface of the Si substrate. The contamination of the single crystal thin film is effectively prevented.

Immediately thereafter, a single crystal thin film is formed on the substrate by a CVD apparatus.

Changing the order of this operation is not preferable. That is, S
When the i substrate is dry hydrochloric acid etched and then heat-treated, the dry hydrochloric acid etching has no effect.

Further, it is preferable to carry out epitaxial growth immediately after etching with dry hydrochloric acid in order not to increase defects.

As a result, the most preferable result can be obtained by performing the heat treatment on the Si substrate, dry hydrochloric acid etching, and epitaxial growth in this order.

Now, the operation effect of the heat treatment, which is one of the features of the present invention,
The dopant (BP, As, etc.) contained in the Si-substrate surface and its vicinity is vaporized and escaped in advance so as not to be mixed into the single crystal thin film during epitaxial growth. For that purpose, the temperature of the above-mentioned heat treatment is preferably higher than at least the temperature during the subsequent dry hydrochloric acid etching or the epitaxial growth operation. Specifically, the temperature during dry hydrochloric acid etching is usually 1060 ° C to 11 ° C.
It is about 50 ℃, and the growth temperature of epitaxy is usually 100.
Since the temperature is about 0 ° C. to 1100 ° C., it is desirable to perform the above-mentioned heat treatment at a temperature higher than the temperature of these steps, and the temperature of the heat treatment is higher than that, that is, 1150 ° C. or higher. 115
If the temperature is less than 0 ° C, even if the temperature is higher than the maximum temperature of the subsequent process, the heat treatment time must be very long.
Contamination by dopants from the Si substrate is unavoidable and not very practical.

Further, if the heat treatment temperature is too high, defects are rather generated in the epitaxially grown single crystal thin film, which is not preferable.

If the heat treatment temperature is 1150 ° C or higher,
About 20 minutes or more is enough.

By performing the heat treatment under the above conditions, the Si-substrate surface dopant-reduced layer produced, withstands a slight decrease in the thickness of the dopant-reduced layer due to the next dry hydrochloric acid etching treatment,
Si in a single crystal thin film in the next epitaxial growth process
-It seems to play a role in preventing contamination by the dopant leaching from the inside of the substrate.

In the present invention, the susceptor is poly-Si coated using a silicon-based gas, which is a material for which the purity is to be evaluated, but this poly-Si coating method may be any method that is usually performed. For example, the silicon-based gas used as a sample is usually diluted with hydrogen to a concentration of 0.01 to 0.5%.
Circulate for ~ 60 minutes.

The temperature of the susceptor is preferably about 1000 to 1100 ° C,
The lower the better.

The coating of the susceptor with polysilicon can be done in the usual manner with the device being the same device as the subsequent epitaxial growth, with the susceptor installed therein.

The etching operation using dry hydrochloric acid is carried out by placing the Si-substrate, which is also heat-treated as described above, on the susceptor which is poly-Si-coated as described above. The device in this case may be the same device used for the subsequent epitaxial growth. Dry hydrochloric acid with hydrogen usually
0.2-5% Dilute to a concentration of 0.5-3%, preferably 5-30 minutes
Distribute in minutes.

The temperature of the Si-substrate is preferably between 1060 ° C and 1150 ° C. If it is too low, there is no etching effect on the surface, if it is too high,
This is because the dopant diffuses from the inside to the Si substrate surface, and the effect of the heat treatment performed before this disappears.

SUMMARY OF THE INVENTION The features of the present invention described above will be briefly summarized here.

When determining the purity of ultra-high-purity silicon-based gas by measuring the specific resistance value with a probe, the dopant that escapes the epitaxially growing single-crystal thin film from the Si substrate based on that gas. It is necessary to thoroughly prevent the pollution by.

When using the same electrically conductive Si substrate as the silicon-based gas, the specific resistance value is always low due to contamination from the substrate. It is very uncomfortable that the measured value of ultra-high-purity silicon-based gas with a high specific resistance value is always low, but from the reliability of the measured value, the measured value is on the safe side. This is a favorable tendency. That is, the smaller the pollution, the higher the specific resistance value. In other words, the higher the specific resistance value, the closer to the true value, and the preferable result is obtained.

However, the probe method that can be used is 2-probe method (spreading resistor)
It is expensive and complicated to operate.

On the other hand, when using the simple and inexpensive four-probe method that is widely used, it is necessary to reverse the electric conduction types of the single crystal thin film and the Si substrate to form P and N junctions. That
The influence of the reverse type dopant escaping from the Si substrate on the specific resistance value is very large.

That is, the opposite type dopants compensate with each other in the single crystal thin film and apparently increase the specific resistance value.

Even if an ultrahigh-purity gas is measured and a high specific resistance value is obtained, its reliability will be decisively compromised.

Therefore, unless it is possible to thoroughly prevent the contamination of the Si substrate with the reverse type dopant, the application of the ultra-high purity gas of the four-probe method would not be practical at all, and in fact, there was no conventional method. .

Recently, we have succeeded in thoroughly preventing the escape of the dopant from the Si-substrate by the above-mentioned method, and as a result, the 2-probe that may use the Si substrate of the same electric conduction type as the source gas. Not only the method but also the simple and inexpensive 4-probe method, which is possible only by using the reverse type Si substrate, has been successfully applied to the purity measurement of ultra-high purity silicon-based gas.

One of the characteristic requirements is the heat treatment of the Si substrate.
Removal of the dopant from the surface layer of the Si substrate, and the other one is to remove the back surface of the Si substrate during dry hydrochloric acid etching.
It is to coat with polysilicon produced by using a silicon-based gas as a material whose purity is to be evaluated.

Effects of the Invention According to the method of the present invention, the purity of ultra-high purity silicon-based gas can be analyzed and evaluated using a simple and inexpensive probe method.

The demand for ultra-high-purity silicon-based gas has recently been rapidly increasing in the semiconductor industry, and therefore, the industrial applicability of the present invention for providing a simple and inexpensive measuring method for its purity must be said to be extremely large. .

Examples will be shown below, but it should be clearly understood here that this merely shows a preferred embodiment of the present invention and is not intended to limit the technical scope of the present invention. I have to.

Example 1 A. Si substrate 4 inch φ, (111) surface, P-type (B-doped), 50Ωcm single-sided mirror-polished Si wafer is divided into 10 equal parts, and each surface is washed with sulfuric acid and hydrogen peroxide solution, and then washed and dried with pure water. did. This was used as a Si substrate.

B. Equipment A SiC coated graphite susceptor is installed in the horizontal quartz reaction tube, and induction heating is performed from the outside with a high frequency of 400 KH _Z output 5 Kw.

The Si substrate is placed on the susceptor and heated.

From the inlet of the quartz reaction tube, hydrogen as a diluting gas, monosilane as a silicon-based gas of the material whose purity is to be evaluated, and dry hydrochloric acid as an etching gas can be blown through the flowmeter. The exhaust gas is led from the outlet to the treatment facility.

The monosilane gas used as the data was 1000 Ωcm as measured by the CV method.

C. Experimental Method The heat treatment of the Si substrate was carried out by heating in a hydrogen stream at 1150 ° C. for 30 minutes.

The susceptor poly-Si coating is monosilane 3 ml / m
It was diluted with in and passed over a susceptor heated to 1050 ° C. for 60 minutes.

The heat-treated Si substrate was set on a poly-Si coated susceptor and placed in a quartz reaction tube.

It was heated by high frequency and the Si substrate was kept at 1130 ° C. 40 ml / min of dry hydrochloric acid was diluted with 2.0 / min of hydrogen and circulated for 10 minutes to perform etching with dry hydrochloric acid. After that, the Si substrate was kept at 1030 ° C., 3 ml / min of monosilane was diluted to 4.5 / min of hydrogen and allowed to flow for 60 minutes to grow a single crystal thin film of Si epitaxially.

D. Result V (mv) / I (μA) measured using a 4-point probe was 86.5
Met.

The thickness of the Si single crystal thin film measured by the driller method is 25
It was μ.

The following equation was used as the relational expression for calculating the specific resistance value.

The electric conduction type was based on thermoelectromotive force.

The single crystal thin film obtained by this experiment is N type, 980Ω
The value was in cm, which was in good agreement with the value obtained by the CV method.

Comparative Example 1 An experiment was performed in the same manner as in Example 1 except that the Si substrate was not heat-treated.

The electric conduction type of the single crystal thin film became P type, which was different from the monosilane gas of the material.

Comparative Example 2 An experiment was performed in the same manner as in Example 1 except that etching with dry hydrochloric acid was not performed.

The electric conduction type of the single crystal thin film is a mixture of P type and N type,
It was impossible to measure the specific resistance value.

Comparative Example 3 An experiment was conducted in the same manner as in Example 1 except that the susceptor was not coated with poly-Si. The electric conduction type of the single crystal thin film became P type, which was different from the type of monosilane gas in the material.

Comparative Example 4 The order of the experimental method of Example 1 was changed, and first, polySi was used.
A Si substrate was placed on a C-coated graphite susceptor and etched with dry hydrochloric acid. Then, the Si substrate was heat-treated, and then a single crystal thin film of Si was produced by epitaxial growth.

The single crystal thin film is a mixture of P-type and N-type electric conduction types,
It was impossible to measure the specific resistance value.

Examples 2, 3 and 4 Comparative Examples 5 and 6 In Example 1, examination was conducted by changing the temperature of the heat treatment of the Si substrate.

The results are summarized in Table 1.

It can be seen that the heating temperature should be 1150 ° C or higher.

Examples 5, 6, 7 Monosilane gas (specific resistance value, 100, 500, 1000, 2000 Ωcm) measured by the CV method was used as a source gas, and the measurement was performed in the same manner as in Example 1. All were N type. The results are shown in Table 2.

Example 8 An experiment was conducted in the same manner as in Example 1 except that disilane was used as the silicon-based gas instead of monosilane, and the temperature of epitaxial growth was changed from 1030 ° C to 1000 ° C.

The result was N type 480 Ωcm.

Example 9 A. Si substrate A 4-inch φ (111) surface, N-type (P-doped) 70 Ωcm single-side mirror-polished Si wafer was used.

B. The device was the same as in Example 1.

C. Experimental method The Si substrate was heated at 1170 ° C for 30 minutes.

Others were the same as in Example 1.

D. Result: N type 950 Ωcm as a result of measurement using a 2-probe measuring instrument.
Met.

Comparative Example 7 Example 9 except that the Si substrate was not heat-treated.
I went the same way.

The result was N type 100 Ωcm.

Claims (1)

[Claims] 1. When evaluating the purity of a silicon-based gas, in the method of measuring the specific resistance value of a single crystal thin film epitaxially grown on a silicon substrate using the silicon-based gas by a probe method, first, The silicon substrate is heat-treated at a temperature higher than the maximum temperature used in all the steps up to epitaxial growth, the susceptor is coated with polysilicon of the above silicon-based gas, and the above-mentioned treatment is performed on the susceptor subjected to the treatment. After etching a silicon substrate with dry hydrochloric acid, a single crystal thin film is epitaxially grown on the silicon substrate using the above silicon-based gas, and the specific resistance of the single crystal thin film is measured. A method for measuring the specific resistance value for evaluating the purity of gas.