JPS5843902B2 - Method for measuring resistivity of semiconductor silicon single crystal - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to three methods for measuring the resistivity of semiconductor silicon single crystals.

Generally, resistivity, which is a characteristic of semiconductor silicon single crystals, is measured and evaluated by a four-probe measurement method, a spreading resistance measurement method, or the like.

When attempting to measure the resistivity of a semiconductor silicon single crystal using such a spreading resistance measurement method, the measurement surface of the silicon single crystal to be measured should be chemically mechanically polished or chemically polished to make it a smooth surface before measurement. is common.

However, when measuring the resistivity of a silicon single crystal immediately after polishing, a constant value cannot be obtained.

In addition, since the resistivity of a silicon single crystal changes over a considerable period of time, in order to measure the stable resistivity of a silicon single crystal, it is necessary to open the silicon for 15 to 20 hours after processing the measurement surface of the silicon single crystal. There is a problem in that measurements can only be made after the single crystal has been left alone.

Particularly in epitaxial wafers, it is difficult to measure resistivity in a highly stable manner due to the above-mentioned change in resistivity over time.

The present inventors studied the change in resistivity over time and found that in the case of epitaxial wafers, depending on the resistivity of the vapor growth layer, the thickness of the vapor growth layer, and the atmosphere,
Although the speed and rate of change in resistivity over time immediately after vapor phase growth differs, the resistivity of the vapor phase grown layer of epitaxial wafers with the same characteristics eventually converges to a certain stable value. I found out that.

As an example of the above, an N-PB type (P-type Figure 1 shows the results of measuring the change in resistivity over time using the four-probe measurement method when the epitaxial wafer (substrate) was left indoors for a long time.

Curve A shown in Figure 1 shows the resistivity (ρ) of the vapor-phase grown layer: 10
．． The curve 8 is for an epitaxial wafer with a vapor growth layer thickness (t) of 7.0 μm and a vapor growth layer sheet resistance (Rs) of 16000Ω. .3Ω, thickness of vapor growth layer t) 5
．． The curve C is for an epitaxial wafer with a sheet resistance (Rs) of 8100 Ω for a vapor-phase growth layer of 3μ, and a resistivity (p) of the vapor-phase growth layer (p) of 0.96Ωα, and a thickness of the vapor-phase growth layer (t) 7. .2 μm, sheet resistance (R
s) For an epitaxial wafer of 1340 Ω, the 9th curve has a resistivity (ρ) of the vapor grown layer of 0.46.
Ωα, the thickness (t) of the vapor-phase grown layer is 7.0 μm, and the sheet resistance (Rs) of the vapor-phase grown layer is 657Ω.

Figure 2 shows the rate of change over time in the sheet resistance (Rs) of the vapor grown layer with respect to the thickness 1) of the vapor grown layer when the resistivity (ρ) of the vapor grown layer is used as a parameter, and A Curve 8 shows the resistivity of the vapor grown layer (ρ) 13Ω, curve 8 shows the resistivity of the vapor grown layer (ρ) 5Ω, and curve C shows the resistivity of the vapor grown layer (ρ) 5Ω.
J)) 1Ω, and the 9th curve is the resistivity (ρ) of the vapor growth layer of 0.
5Ω for each epitaxial wafer.

FIG. 5 shows the rate of change over time in the sheet resistance (Rs) of the vapor grown layer with respect to the resistivity (ρ) of the vapor grown layer when the thickness 1) of the vapor grown layer is used as a parameter. Curve A is the thickness of the vapor growth layer t) 3 μm, Curve 8 is the thickness of the vapor growth layer t) 5 μms Curve C is the thickness of the vapor growth layer t)
) 7 μm, 9 curve is the thickness of the vapor growth layer 1) 10 μm
, E curves are for each epitaxial wafer with a vapor growth layer thickness t) of 15 μm.

Above, we have explained the change in resistivity over time in an N+PB type epitaxial wafer as an example of an N type silicon single crystal.
In the case of B-type (N-type substrate) epitaxial wafers, the resistivity also changes over time.

The change in resistivity over time in the P-NB type epitaxial wafer is higher than that in the N-PB type epitaxial wafer, whereas the change in resistivity over time in the P-NB type epitaxial wafer is higher than that immediately after vapor phase growth. It has a characteristic that the resistivity changes in a direction lower than that of the resistivity of the N-PB type, and the rate of change is smaller than that of the N-PB type with the same resistivity.

However, just as in the case of N-PB epitaxial wafers, the resistivity of P-NB type epitaxial wafers changes over time, and the resistivity of epitaxial wafers with the same characteristics eventually converges to a certain stable value (as shown in the figure). omission).

When measuring the resistivity of epitaxial wafers whose resistivity changes over time as described above, it is important to have a good understanding of the characteristics of each type of epitaxial wafer and the process of change over time due to the atmosphere in which it is left, and then measure the Otherwise, the resistivity could not be easily measured, and measurement accuracy also became a problem.

The present invention solves the above problems, and conducted various experimental studies to directly stabilize the resistivity (hereinafter referred to as aging) in a short time and with good reproducibility in accordance with the characteristics of the silicon single crystal to be measured. This was based on the discovery that the resistivity could be stabilized in a short time by treating the measurement surface of the silicon single crystal to be measured with an oxidizing aqueous solution or an oxidizing atmosphere.

In the present invention, a silicon single crystal that has been subjected to chemical mechanical polishing or etching, or an epitaxial wafer that has just been vapor-phase grown, or the single crystal that is undergoing aging, is heated in an oxidizing aqueous solution or in an oxidizing aqueous solution for a short time (within about 1 hour). After performing an aging process to stabilize the resistivity of the surface to be measured, the resistivity is measured by the spreading resistance method or the four-probe measurement method.

The oxidizing aqueous solution used in the present invention is H2O (deionized water), which can easily obtain high purity quality.

HNO3, H2O2, H2SO4+H20, Na2Cr
An aqueous solution such as 20□ can be used.

In particular, H2O (deionized water) is advantageous because it can easily obtain high-purity quality, and it is also advantageous in terms of workability, waste liquid treatment (pollution control), etc.

When processing silicon single crystals in an oxidizing aqueous solution,
H2O, HNO3, H2O2゜When using H2SO4+H20 aqueous solution, boil it and remove Na2Cr2o
7 When an aqueous solution is used, the treatment is performed at room temperature (20 to 25°C).

Further, as the oxidizing atmosphere used in the present invention, the steam and ozone released when an aqueous solution of HNO3, H2O2, etc. is heated to boiling can be used, since it is easy to obtain as high a purity as possible.

Next, the present invention will be explained in detail using one embodiment of the present invention.

Example 1 On a P-type silicon substrate with a diameter of 501n11L, N-type vapor-phase growth layers of various resistivities and thicknesses were formed using the SiH4 vapor-phase elongation method, and two N+P-type epitaxial layers were formed under the same conditions. The sheet resistance of each epitaxial wafer was left indoors for a long time (5 to 40 days) and the sheet resistance when the aging process was completed, and that of each other epitaxial wafer was calculated from 5 to 40 days. Table 1 shows the results of measuring the cytoresistance using the four-probe measurement method immediately after boiling (aging) with H20 (deionized water) for 10 minutes.

As shown in Table 1, the sheet resistance when the epitaxial wafer was left indoors for a long time and the sheet resistance of the epitaxial wafer after the aging process matched within the measurement error range.

It was confirmed that even if each aging-treated epitaxial wafer was left indoors for several tens of days, there was almost no change in sheet resistance over time as shown in Table 2.

Agein using H2O (deionized water) as in this example.
In g treatment, epitaxial wafers with low sheet resistance showed a very stable rate of change when boiled for about 1 minute or more, but when the sheet resistance was high, the rate of change reached its maximum after boiling for 5 to 10 minutes. , and there was a tendency for the rate of change to become somewhat lower around 5 to 10 minutes.

Example 2 On a P-type silicon substrate with a diameter of 50 mm, two N+-PB type epitaxial layers were grown using the SiH4 vapor phase growth method with various resistivities and thicknesses under the same conditions. The sheet resistance of each epitaxial wafer after fabrication and cracking was left indoors for several 10 days and the aging change was completed, and the sheet resistance of each other epitaxial wafer** was calculated using a concentration of 1i50%. After recovering the resistivity by immersing it in the HF solution for about 5 minutes (this treatment restores the resistivity to almost the value immediately after vapor phase growth), the temperature was about 23°C and the concentration was 1.
Table 3 shows the sheet resistance measured by the four-probe measurement method after being treated in a 0% Na2Cr2O7 aqueous solution for about 5 minutes.

As shown in Table 3, the sheet resistance and N a 2 Cr 20 when the epitaxial wafer is left indoors for a long time
7. The sheet resistance of the epitaxial wafer after the aging process using an aqueous solution was consistent within the measurement error range.

Furthermore, as in Example 1, it was confirmed that even if each epitaxial wafer subjected to the aging process in this example was left indoors for several tens of days, the sheet resistance changed over time almost unchanged as shown in Table 4. Ta.

Age using Na2Cr2O7 aqueous solution as in this example
In the ing treatment, the concentration of Na2Cr2O7, Na2Cr2
Since there are some differences in the rate of change in sheet resistance depending on conditions such as the temperature of the O7 aqueous solution and treatment time, Na 2 Cr
Preferably, the concentration of 207 is 5 to 30%, the temperature of the Na2Cr2O7 aqueous solution is 20 to 25 DEG C., and the treatment time is within about 60 minutes.

Example 3 In order to investigate the treatment effect in an oxidizing atmosphere, a mercury lamp was installed in a clean bench, and the same sample as in Example 1 was placed in the clean bench. However, the aging effect was observed.

Furthermore, in an experiment in which warm hydrogen peroxide or warm nitric acid and the same sample as in Example 1 were placed in a desicake, the same aging effect as in the above example was observed.

In each of the above embodiments, the resistivity of the N-type silicon single crystal is
Although we have explained that the measurement is performed after the P ing process,
It was confirmed that similar results were obtained when measuring the resistivity of silicon single crystals.

As detailed above, the sheet resistance measured for a silicon single crystal subjected to aging treatment is different from the true resistivity measured immediately after vapor phase growth or chemical mechanical polishing, but it is different from the true resistivity measured immediately after chemical mechanical polishing. Since it can be stabilized to a certain value according to , it can be put to practical use by making appropriate corrections.

According to the method of the present invention, a stable surface state can be formed on the silicon single crystal to be measured through a short-time aging process, so the resistivity of a high-purity silicon single crystal can be easily measured in a short time with good reproducibility. be.

[Brief explanation of the drawing]

Figure 1 is a graph showing the change over time in the resistivity of the vapor-grown layer of various N-PB type epitaxial wafers in terms of the rate of change in sheet resistance, and Figure 2 is a graph showing the vapor-grown layer of various N+PB type epitaxial wafers. This is a graph showing the rate of change in sheet resistance over time with respect to the vapor growth thickness when the resistivity of various types of N-PB type eviaxial wafers is used as a parameter. 5.
It is a graph showing the rate of change in sheet resistance over time with respect to the resistivity of a phase growth layer.

Claims (1)

[Claims] 1. When measuring the resistivity of a high-purity modified silicon single crystal,
A force method for measuring the resistivity of a semiconductor silicon single crystal, characterized in that the resistivity of the silicon single crystal to be measured is measured after the surface of the silicon single crystal to be measured is inactivated. 2 Claim 1 in which the surface of the silicon single crystal to be measured is inactivated in an oxidizing aqueous solution or an oxidizing atmosphere
Force method for measuring resistivity of semiconductor silicon single crystals as described in Section 1. 3. Resistivity measurement of a semiconductor silicon single crystal according to claim 1, in which the surface of the silicon single crystal to be measured is boiled with an aqueous solution of H2O (deionized water), H2O2, H2SO4 + H20, or treated with an aqueous solution of Na2Cr2O7. force law. 4. Cover the surface of the silicon single crystal to be measured with HNO3°H2O.
2. A method for measuring the resistivity of a semiconductor silicon single crystal according to claim 1, wherein the method is carried out in an ozone (03) atmosphere or in an ozone (03) atmosphere.