JP2021038995A - Impurity analysis method of silicon substrate surface - Google Patents

Abstract

To provide a silicon substrate surface impurity analysis method in which impurities on the surface of a silicon substrate are analyzed by total reflection fluorescent X-ray analysis.SOLUTION: A silicon substrate surface impurity analysis method is a method in which a vapor-phase cracking is performed by making a hydrofluoric acid of 20-30% volume concentration and vapors rising from a hydrogen peroxide solution of 12.4-18.6% volume concentration contact with the surface of the silicon substrate for 15 minutes or more; and subsequently impurities on the surface of the silicon substrate are assessed using a total reflection fluorescent X-ray analysis method.SELECTED DRAWING: Figure 3

Description

The present invention relates to a method for analyzing impurities on the surface of a silicon substrate, and particularly to a method for analyzing impurities on the surface of a silicon substrate, which analyzes impurities on the surface of the silicon substrate by total internal reflection fluorescent X-ray analysis.

Wafer cleanliness control is important in the semiconductor device manufacturing process, and WSA (Wafer Surface Analysis) is used as a wafer surface impurity analysis method.
In WSA, the wafer surface is exposed to acid vapor, the natural oxide film on the wafer surface is dissolved, and then the wafer surface is scanned with an acid solution to recover impurities on the wafer surface including the natural oxide film and induce the recovered acid solution. It is measured with a coupled plasma mass analyzer or the like. This method is characterized in that high-sensitivity analysis can be performed by recovering and concentrating impurities on the wafer surface, but has a drawback that the position information of impurities on the wafer surface is lost.

On the other hand, as a method capable of easily analyzing wafer surface impurities, there is a total reflection X-Ray Fluorescence analysis method (hereinafter referred to as TXRF method). The TXRF method is a method capable of detecting wafer surface impurities with high sensitivity by exciting wafer surface impurities with X-rays incident under total reflection conditions and detecting generated fluorescent X-rays.
FIG. 1 shows a principle diagram of the TXRF method. The TXRF method is non-destructive and can analyze the in-plane distribution of impurities, which is effective in detecting local contamination. On the other hand, there is also a problem that the detection sensitivity is inferior to that of chemical analysis such as WSA.

Therefore, in recent years, the TXRF method has been made more sensitive by combining the WSA and the TXRF method, drying the acid solution from which impurities have been recovered on the wafer, and performing TXRF analysis on the drying marks. , The advantage of the TXRF method is that the position information of impurities is lost.
Therefore, by drying the wafer surface after vapor phase decomposition, impurities on the wafer surface are aggregated into particles (Fig. 2), and by performing TXRF analysis in that state, the position information of the impurities in the wafer surface is maintained. , A method utilizing the effect of increasing the detection intensity is also used.
FIG. 3 shows a flow chart of the gas phase decomposition-TXRF method (Patent Document 1).

Patent No. 3690484

However, although hydrofluoric acid (hereinafter abbreviated as HF) is generally used as the gas phase decomposition condition for the morphological change of wafer surface impurities into particles, the optimum values of concentration and gas phase decomposition time are reached. It was not mentioned.
For this reason, it has been found that proper gas phase decomposition is not performed, the morphological change into particles is insufficient, and as a result, it leads to an increase in variation.

Therefore, the present invention provides TXRF analysis with higher sensitivity while maintaining the position information of impurities in the wafer plane by providing a particle-like morphological change according to the vapor phase decomposition conditions and appropriate vapor phase decomposition conditions. The purpose is to provide a way to do it.

The present invention has been made to solve the above-mentioned problems, and is a method for analyzing impurities on the surface of a silicon substrate, wherein 20 to 30% volume concentration of hydrogen peroxide and 12.4 to 18. Silicon characterized in that vapors generated from a 6% volume concentration hydrogen peroxide solution are brought into contact with each other for 15 minutes or more to perform vapor phase decomposition, and then impurities on the surface of the silicon substrate are evaluated by a total reflected fluorescent X-ray analysis method. A method for analyzing impurities on the surface of a substrate is provided.

HF concentration is 20 to _30%, X-ray intensity by TXRF analysis after vapor phase decomposition of the concentration of H 2 _{O 2} was used mixed acid solution at a concentration of 12.4 to 18.6% is, HF only (50% ) Is increased by about 4 times from the X-ray intensity by TXRF analysis after the gas phase decomposition, and the maximum effect is exhibited when the gas phase decomposition time is 15 minutes or more.

Further, in performing the gas phase decomposition, a chemical solution is injected into a container having an opening, a closed space is formed by covering the opening of the container with the silicon substrate, and the vapor of the chemical solution makes a surface on the closed space. The surface of the silicon substrate can be vapor-phase decomposed.

In this way, the surface of the silicon substrate facing the closed space can be exposed to the vapor of the chemical solution that fills the closed space of the container, and the surface of the silicon substrate can be vapor-phase decomposed.

Further, after the vapor phase decomposition, the surface of the silicon substrate can be dried, and then impurities on the surface of the silicon substrate can be evaluated by total reflection fluorescent X-ray analysis.

In this way, after the vapor phase decomposition, the surface of the silicon substrate is dried to aggregate the impurities on the surface of the silicon substrate into particles, and in that state, the impurities on the surface of the silicon substrate are positioned by the total reflection fluorescent X-ray analysis method. It can be evaluated while preserving the information.

Further, the surface of the silicon substrate can be dried by heating a container in which the silicon substrate is vapor-phase-decomposed.

In this way, by heating the container in which the silicon substrate is vapor-phase-decomposed, drying of the surface of the silicon substrate is easily promoted.

Further, in the present invention, the surface of the silicon substrate can be dried by irradiating the silicon substrate with infrared rays.

In this way, the silicon substrate can be directly heated by infrared rays to accelerate the drying of the surface of the silicon substrate.

If impurity analysis method of the silicon substrate surface of the present invention, HF concentration of _20~30%, H _{2 O} 2 concentration after gas phase decomposition using mixed acid solution at a concentration of 12.4 to 18.6% The X-ray intensity by TXRF analysis is about 4 times higher than the X-ray intensity by TXRF analysis after gas phase decomposition of HF only (50%), and the maximum effect is exhibited when the gas phase decomposition time is 15 minutes or more. Can be done. Therefore, it is possible to analyze with extremely high sensitivity.
Furthermore, by drying the surface of the silicon substrate after vapor phase decomposition, impurities on the surface of the silicon substrate can be aggregated in the form of particles, so that the position information of the impurities in the wafer surface can be maintained and the sensitivity can be further increased. TXRF analysis can be performed.

It is a principle diagram of the TXRF method. It is a figure which shows the morphological change into a particle in a gas phase decomposition-drying process. It is a flow chart of a gas phase decomposition-TXRF method. It is sectional drawing which shows an example of a gas phase decomposition container. The HF + H ₂ O ₂ gas phase decomposition time and the increase in X-ray intensity after the gas phase decomposition are shown. The HF + H ₂ O ₂ vapor phase decomposition time and the lower limit of TXRF detection are shown. The results of the HF + H ₂ O ₂ composition and the X-ray intensity increase ratio before and after the gas phase decomposition are shown. The results of the HF gas phase decomposition time and the X-ray intensity increase ratio after the gas phase decomposition are shown. The results of the HF vapor phase decomposition time and the lower limit of TXRF detection are shown.

Hereinafter, the present invention will be described in detail, but the present invention is not limited thereto.

According to the TXRF method, it is possible to analyze the in-plane distribution of impurities in a non-destructive manner, which is effective in detecting local contamination, but the detection sensitivity is higher than that of chemical analysis such as WSA. There is also the problem of being inferior.

Although the TXRF method has been made more sensitive by combining the WSA and the TXRF method, there is a drawback that the position information of impurities, which is an advantage of the TXRF method, is lost.
Therefore, by drying the wafer surface after vapor phase decomposition, impurities on the wafer surface are aggregated into particles, and by performing TXRF analysis in that state, the detection intensity can be increased while maintaining the position information of the impurities on the wafer surface. It was also increased.

However, it has been found that the appropriate vapor phase decomposition conditions are unknown, and the morphological change into particles is insufficient, resulting in an increase in variation.

Therefore, the present inventor provides TXRF analysis with higher sensitivity while maintaining the position information of impurities in the wafer surface by providing the morphological change into particles according to the gas phase decomposition conditions and the appropriate gas phase decomposition conditions. In order to provide a method for performing the above, it was found that the vapor phase decomposition by a mixed acid of hydrofluoric acid (HF) and hydrogen peroxide solution (H ₂ O _{2) exerts a morphological change into particles suitable for TXRF analysis.} It was. Furthermore, we _{also found the relationship between the concentration of HF and H 2} O ₂ and the gas phase decomposition time, and found that TXRF analysis can be performed with higher sensitivity than before.
That is, the present invention is a method for analyzing impurities on the surface of a silicon substrate, from 20 to 30% volume concentration of hydrofluoric acid and 12.4 to 18.6% volume concentration of hydrogen peroxide solution on the surface of the silicon substrate. This is a method for analyzing impurities on the surface of a silicon substrate, which comprises contacting the generated steam for 15 minutes or more to perform gas phase decomposition, and then evaluating impurities on the surface of the silicon substrate by a total reflected fluorescent X-ray analysis method.
The X-ray intensity by TXRF analysis after gas phase decomposition using an acid solution mixed at a concentration of HF volume concentration of 20 to 30% and H ₂ O _{2 volume concentration of 12.4 to 18.6% is HF only (} 50%) is increased by about 4 times from the X-ray intensity by TXRF analysis after gas phase decomposition, and the maximum effect is exhibited when the gas phase decomposition time is 15 minutes or more.

Hereinafter, the method for analyzing impurities on the surface of a silicon substrate of the present invention will be described with reference to the drawings.
In the method for analyzing impurities on the surface of a silicon substrate of the present invention, the surface of the silicon substrate is typically subjected to gas phase decomposition by vapor of an acid solution of a mixture of hydrofluoric acid and hydrogen peroxide solution. The surface of the vapor-decomposed silicon substrate is dried, and the dried silicon substrate surface is irradiated with X-rays at an incident angle smaller than the critical angle of total reflection conditions, and the reflected fluorescent X-rays are used to irradiate the surface of the silicon substrate. Perform total internal reflection X-ray fluorescence analysis to analyze impurities in.

As the silicon substrate used here as an analysis sample, a silicon substrate having a flat surface and a mirror finish is used in order to perform analysis by utilizing total reflection of X-rays. The total reflection fluorescent X-ray analysis method (TXRF method) is shown in FIG. 1 and has been described above, and is omitted here.

In performing the gas phase decomposition, as shown in FIG. 4, a container 2 having an opening 2a is used.
The opening 2a in the container 2 is sealed in an airtight state by covering it with a silicon substrate 1 as an analysis sample, and a closed space 2b is formed in the internal space. The chemical solution is injected into the closed space 2b by an injection means (not shown).
Chemical solution, here, HF volume concentration of _20~30%, H _{2 O} 2 volume concentration is used mixed acid solution at 12.4 to 18.6%.
When the chemical solution is injected into the closed space 2b in the container 2, it becomes a part of steam due to the atmospheric temperature in the closed space 2b at that time, and the steam fills the closed space 2b in the container 2. The surface of the silicon substrate 1 facing the closed space 2b in the container 2 is exposed and gas phase decomposition is performed. In this case, the vapor generated from HF having a volume concentration of 20 to 30% and H ₂ O ₂ having a volume concentration of 12.4 to 18.6% can be brought into contact with each other for 15 minutes or more.
The above-mentioned container 2 was used in consideration of the contact efficiency between the surface of the silicon substrate 1 and the vapor of the chemical solution, but the present invention is not limited to the method using the container 2.

Next, a method of drying the surface of the vapor-phase-decomposed silicon substrate 1 will be described.
It should be noted that heating does not have to be performed for drying, but the heating means for heating is appropriate. That is, the heating means may heat the container 2 with a heater or the like at the time of vapor phase decomposition, or may irradiate an infrared lamp or the like from the upper part of the silicon substrate 1, and water droplets aggregated on the surface of the silicon substrate during the vapor phase decomposition may be generated. It is necessary not to grow too large. Here, the ambient temperature in the closed space 2b was 23 ° C., and the heating was performed without heating the heater or the like.

In the experiment, the change in X-ray intensity during the gas phase decomposition time was clarified not only for the vapor during the gas phase decomposition but also _{for the steam generated from the mixed acid of HF + H 2} O _2. As a result, the X-ray intensity after gas phase decomposition is about 3 times higher than that before gas phase decomposition with HF alone, whereas it is about 12 times higher with _{HF + H 2} O _2. Found to be seen.

This is because the _{addition of H 2} O ₂ increases the redox potential of the water droplets aggregated on the surface of the silicon substrate 1 and increases the oxidizing power, so that the aggregation of impurities is higher than the gas phase decomposition performed by HF alone. It is probable that the effect was improved.
As a result, the incident X-rays slightly penetrate into the aggregate (particles), and fluorescent X-rays are emitted from the surface and the inside of the aggregate, so that it is considered that the X-ray detection intensity is increased. In addition, as shown in Tables 1 and 2 described later, the X-ray intensity at the gas phase decomposition time reaches the maximum in about 15 minutes, and thereafter, the X-ray intensity does not increase even if the gas phase decomposition time is extended. I also found that I couldn't.
However, when the wafer to be evaluated is a silicon wafer, these effects are not so effective for elements such as Cu, which have a lower ionization tendency than silicon, and even in the present invention, the X-ray intensity before and after gas phase decomposition is not so effective. Little increase was seen.

Hereinafter, the present invention will be specifically described with reference to Examples and Comparative Examples, but this does not limit the present invention.

{Example, Comparative Example}
A forcibly contaminated wafer was used as the silicon substrate in order to analyze impurities on the surface of the silicon substrate. As the forced contamination wafer, a clean p-type PW wafer having a diameter of 200 mm is used, and a 1000 ppm atomic absorption standard solution (Cr, Fe, Ni, Cu) manufactured by Kanto Chemical Co., Ltd. is appropriately diluted, and ultra-high purity ethanol manufactured by Tama Chemical Co., Ltd. ( AA100) A solution prepared with a solvent was added dropwise to the center of the wafer in an amount of 20 μL, and the solution was air-dried to prepare wafers having a content of each element of 0.05 ng, 0.5 ng, 5 ng, and 50 ng in the added dropwise solution.
The total reflection fluorescence X-ray analyzer used was TREX630T manufactured by Technos. The analysis conditions were 40 kV, 40 mA, the X-ray incident angle was 0.05 degrees, and the measurement time was 300 seconds / point. Further, the silicon wafer was subjected to gas phase decomposition using the gas phase decomposition container 2 shown in FIG.

(Example)
_{A 1: 1 ratio (25% HF + 15.5% H 2} O ₂ ) of Stella Chemifa EL-grade hydrofluoric acid (50%) and Santoku Kagaku EL-grade hydrogen peroxide solution (31%) in a gas phase decomposition vessel. ), A total of 200 mL was injected, the silicon substrate was placed on a gas phase decomposition container so that the PW surface faced downward, and the gas phase decomposition was performed by sealing the _{HF + H 2} O _{2 gas.} The gas phase decomposition time was set to 6 levels of 3 minutes, 5 minutes, 10 minutes, 15 minutes, 20 minutes, and 30 minutes, and after the gas phase decomposition, it was naturally dried in a clean draft.

Table 1 shows _{the X-ray intensity (cps) before and after the vapor phase decomposition by HF + H 2} O ₂ vapor phase decomposition and the X-ray intensity after the vapor phase decomposition ÷ the X-ray intensity before the vapor phase decomposition (Ratio). It shows the relationship with time. The X-ray intensity increases after the gas phase decomposition as compared with that before the gas phase decomposition, and the increase in the X-ray intensity shows a tendency of increasing up to 15 minutes in the gas phase decomposition time. In fact, a strength improvement of about 12 times was observed. However, after the decomposition time is 15 minutes, no increase in X-ray intensity is observed even if the gas phase decomposition is performed further.

Further, FIG. 5 shows the HF + H ₂ O ₂ gas phase decomposition time and the X-ray intensity increase ratio after the gas phase decomposition. Each plot represents the average value at 0.05 ng, 0.5 ng, 5 ng, and 50 ng, and the error bars are the maximum value and the minimum value.

Similarly, the lower limit of detection (atoms / cm ² ) of each element in TXRF under each condition is shown in Table 2 and FIG. In Table 2, it was found that the lower limit of detection becomes smaller as the ratio of increase in X-ray intensity before and after gas phase decomposition increases. From this result, it is considered that the change in background intensity has almost no effect and the X-ray intensity is simply increased.

Further, Table 3 and FIG. 7 show the X-ray intensity and the X-ray intensity ratio before and after the gas phase decomposition in the mixing ratio of HF + H ₂ O _{2 at 15 minutes of the gas phase decomposition.}
From this result, in _{the mixed acid of HF and H 2} O ₂ , the effect of the X-ray intensity after vapor phase decomposition is reduced and the maximum effect is exhibited even if the HF concentration is too high or too low. This was the case when the ratio was 1: 1 (25% HF + 15.5% H ₂ O ₂ ). Further, the condition that the X-ray intensity ratio before and after the vapor phase decomposition exceeds 10 times is that H ₂ O ₂ is 12.4% to 18.6% when HF is 20% to 30%.

(Comparison example)
By injecting 200 mL of STELLA CHEMIFA EL-grade hydrofluoric acid (50%) into the gas phase decomposition container, placing the silicon substrate on the gas phase decomposition container 2 so that the PW side faces downward, and sealing the HF vapor. Gas phase decomposition was performed. The gas phase decomposition time was set to 6 levels of 3 minutes, 5 minutes, 10 minutes, 15 minutes, 20 minutes, and 30 minutes, and after the gas phase decomposition, it was naturally dried in a clean draft.

In Table 4, the X-ray intensity (cps) before and after the gas phase decomposition by HF vapor decomposition and the X-ray intensity after the gas phase decomposition ÷ the X-ray intensity before the gas phase decomposition (Ratio) are calculated and calculated as the gas phase decomposition time. Represented the relationship. The X-ray intensity increases after the gas phase decomposition as compared with that before the gas phase decomposition, and the increase in the X-ray intensity shows a tendency of increasing until the gas phase decomposition time is 15 minutes. In fact, the strength was improved about 3 times. However, after the decomposition time is 15 minutes, no increase in X-ray intensity is observed even if the gas phase decomposition is performed further.

Further, FIG. 8 shows the HF gas phase decomposition time and the X-ray intensity increase ratio after the gas phase decomposition. Each plot represents the average value at 0.05 ng, 0.5 ng, 5 ng, and 50 ng, and the error bars are the maximum value and the minimum value. ^{Similarly, the lower limit of detection (atoms / cm 2} ) of each element in total internal reflection fluorescent X-ray under each condition is shown in Table 5 and FIG.

From the above results, HF + H ₂ O ₂ was used for gas phase decomposition, and the mixing ratios thereof were 20 to 30% for HF volume concentration and 12.4 to 18.6% for _{H 2} O _{2 volume concentration.} The standard to be used. Further, by setting the gas phase decomposition time to 15 minutes or more, the X-ray intensity after the gas phase decomposition becomes about 12 times the X-ray intensity before the gas phase decomposition, and the X after the gas phase decomposition is performed only by HF. An X-ray intensity four times higher than the line intensity can be obtained. In addition, the lower limit of detection is about 1/4 of that in the case of gas phase decomposition using only HF.

In summary, when analyzing metal impurities on the surface of a silicon wafer, elements with a higher ionization tendency than silicon or equivalent to silicon are four times more than when TXRF analysis is performed after vapor phase decomposition using only HF. Sensitivity can be improved.

If impurity analysis method of the silicon substrate surface of the present invention, HF concentration of _20~30%, H _{2 O} 2 concentration after gas phase decomposition using mixed acid solution at a concentration of 12.4 to 18.6% The X-ray intensity by TXRF analysis is about 4 times higher than the X-ray intensity by TXRF analysis after gas phase decomposition of HF only (50%), and the maximum effect is exhibited when the gas phase decomposition time is 15 minutes or more. Can be done.
Furthermore, by drying the surface of the silicon substrate after vapor phase decomposition, impurities on the surface of the silicon substrate can be aggregated in the form of particles, so that the position information of the impurities in the wafer surface can be maintained and the sensitivity can be further increased. TXRF analysis can be performed.

The present invention is not limited to the above embodiment. The above embodiment is an example, and any one having substantially the same configuration as the technical idea described in the claims of the present invention and exhibiting the same effect and effect is the present invention. It is included in the technical scope of the invention.

1 ... Silicon substrate, 2 ... Container, 2a ... Aperture, 2b ... Sealed space.

Claims (5)

This is a method for analyzing impurities on the surface of a silicon substrate. Steam generated from 20 to 30% volumetric acid fluoride and 12.4 to 18.6% volumetric hydrogen peroxide solution is applied to the surface of the silicon substrate for 15 minutes. A method for analyzing impurities on the surface of a silicon substrate, which comprises contacting the above to perform gas phase decomposition and then evaluating impurities on the surface of the silicon substrate by a total reflected fluorescent X-ray analysis method. In performing the gas phase decomposition, a chemical solution is injected into a container having an opening, a closed space is formed by covering the opening of the container with the silicon substrate, and the vapor of the chemical solution is used to face the closed space. The method for analyzing impurities on the surface of a silicon substrate according to claim 1, wherein the surface of the silicon substrate is vapor-phase decomposed. The silicon substrate surface according to claim 1 or 2, wherein after the vapor phase decomposition, the surface of the silicon substrate is dried, and then impurities on the surface of the silicon substrate are evaluated by a total reflection fluorescent X-ray analysis method. Impurity analysis method. The method for analyzing impurities on the surface of a silicon substrate according to claim 3, wherein the surface of the silicon substrate is dried by heating a container obtained by vapor-phase-decomposing the silicon substrate. The method for analyzing impurities on the surface of a silicon substrate according to claim 3, wherein the surface of the silicon substrate is dried by irradiating the silicon substrate with infrared rays.

Images