JP2023161361A - Method for producing standard sample for surface metal contamination evaluation - Google Patents

Abstract

To provide a method for producing a standard sample for surface metal contamination evaluation, which suppresses generation of an oxide film or attachment of an organic substance such as hydrocarbons on a surface, and is excellent in repeatability and quantitativity.SOLUTION: A method for producing a standard sample quantitatively contaminated with predetermined metal impurities used in surface metal contamination analysis, includes a wafer surface cleaning step of removing the surface impurities of a silicon wafer for the standard sample, an oxidation reduction potential adjustment step of dissolving the metal impurities in an HF aqueous solution to reduce oxidation reduction potential until the metal impurities come into a neutral atom state, then a step of immersing the silicon wafer after the wafer surface cleaning step in the HF aqueous solution, and then a rinse step of removing the HF aqueous solution attached to the silicon wafer, and a step of drying the silicon wafer. The method for producing the standard sample for surface metal contamination evaluation is for producing the standard sample quantitatively contaminated with the predetermined metal impurities on the surface of the silicon wafer for the standard sample.SELECTED DRAWING: Figure 1

Description

The present invention relates to a method for producing a standard sample (standard wafer) that is quantitatively contaminated with a predetermined metal species (metal impurity) used in surface metal contamination analysis (surface metal contamination evaluation) of silicon wafers.

With the miniaturization of semiconductor devices, requirements for the cleanliness of silicon wafer surfaces are becoming stricter. In particular, metal contamination is a cause of device malfunction and deterioration of element characteristics, and is therefore strictly controlled.
On the other hand, when quantitatively analyzing metal contamination present on the surface of a silicon wafer, a standard sample in which the wafer surface is contaminated in advance with a predetermined amount of metal impurities is required.

For this reason, conventional methods for preparing standard samples include immersing a wafer in an alkaline hydrogen peroxide aqueous solution containing a certain concentration of metal impurities (Patent Document 1), and dipping the wafer surface in an ammonia-hydrogen peroxide aqueous solution → hydrofluoric acid. A method in which the surface is made hydrophilic by treatment with an aqueous solution → hydrochloric acid-hydrogen peroxide solution, a solution containing a certain concentration of metal impurities is dropped onto the surface, and after holding for a certain period of time, the wafer is rotated at high speed to remove the solution remaining on the wafer. (Patent Document 2), a method of contaminating a wafer surface with a predetermined amount of a metal impurity solution containing volatile alcohol, and a method of performing oxidation, reduction, chemical reaction treatment, etc. on the contaminated metal impurities (Patent Document 3), A method (Patent Document 4) has been disclosed in which the amount of metal impurities attached to the surface is controlled by pH adjustment based on the solubility product of metal oxides.

Japanese Patent Application Publication No. 6-249764 Japanese Unexamined Patent Publication No. 2-156636 Japanese Patent Application Publication No. 2000-243798 Japanese Patent Application Publication No. 10-178077

When analyzing metal contamination on the wafer surface using a secondary ion mass spectrometer, especially when using a time-of-flight secondary ion mass spectrometer (hereinafter referred to as TOF-SIMS), it is important to check the uniformity of the contamination concentration. The surface condition of the contaminated surface and its stability during analysis are extremely important. Specifically, if the sample surface is covered with an insulating material such as an oxide film, charge-up (electrification) will cause deterioration in mass resolution and variation in detection intensity, and if a large amount of organic substances such as hydrocarbons are present, There are problems with false detection and variations in detection strength due to mass interference. Therefore, a metal contamination standard sample that is free from the influence of oxide films and hydrocarbons is required.

However, according to the method of Patent Document 1, an oxide film is generated on the wafer surface because the treatment is performed with an alkaline hydrogen peroxide aqueous solution, and metal impurities are contained in the oxide film generated as oxides and hydroxides. There are hidden issues. Furthermore, according to the method of Patent Document 2, an oxide film is generated on the wafer surface by treating the wafer with an ammonia-hydrogen peroxide aqueous solution → a hydrofluoric acid aqueous solution → a hydrochloric acid-hydrogen peroxide aqueous solution, and a metal This is a state in which impurities are adsorbed. In both of the methods of Patent Documents 1 and 2, since an oxide film exists on the wafer surface, there is a problem that they are affected by the above-mentioned problems. Furthermore, according to the method of Patent Document 3, there is a risk that a high concentration of hydrocarbons will remain on the wafer surface due to the step of contaminating the wafer surface with a metal impurity solution containing volatile alcohol, and the reduction process may cause oxidation. Even if the film is removed, the concentration of fast-diffusing metals such as Ni and Cu on the sample surface will be different from the expected concentration due to the diffusion of impurities into the wafer bulk due to processing conditions (temperature, time). It is considered unsuitable as a standard sample. Furthermore, according to the method of Patent Document 4, the pH is adjusted by using one or more of hydrochloric acid, nitric acid, ammonium hydroxide, and hydrogen peroxide as solvents, but in any case, an oxide film is formed during the treatment. is generated, and there is a problem that the metal oxide is embedded in the oxide film or adsorbed onto the oxide film.

The present invention has been made in view of the above circumstances, and provides a standard for surface metal contamination evaluation that suppresses the formation of oxide films on the surface and the adhesion of organic substances such as hydrocarbons, and has excellent reproducibility and quantitative properties. The purpose is to provide a method for preparing samples.

In order to achieve the above object, the present invention provides a method for preparing a standard sample that is quantitatively contaminated with predetermined metal impurities used in surface metal contamination analysis of silicon wafers, comprising:
a wafer surface cleaning step of removing surface impurities from the silicon wafer prepared for the standard sample;
A redox potential adjustment step of dissolving the predetermined metal impurity in a hydrofluoric acid aqueous solution and lowering the redox potential until the predetermined metal impurity in the hydrofluoric acid aqueous solution becomes a neutral atom state;
A step of immersing the silicon wafer after the wafer surface cleaning step in the hydrofluoric acid aqueous solution after the oxidation-reduction potential adjustment step;
a rinsing step for removing the hydrofluoric acid aqueous solution adhering to the silicon wafer after the immersion treatment step;
A step of drying the silicon wafer after the rinsing step,
A method for preparing a standard sample for surface metal contamination evaluation is provided, which comprises preparing a standard sample in which the surface of a silicon wafer for use as a standard sample is contaminated with a quantitative amount of the predetermined metal impurity.

The method for preparing a standard sample for evaluating surface metal contamination of the present invention allows for high reproducibility, without forming an oxide film on the surface of a silicon wafer, and suppressing the adhesion of organic substances such as hydrocarbon ions. A standard sample for evaluating surface metal contamination with excellent quantitative properties can be provided. In particular, even when subjected to accurate mass spectrometry using a time-of-flight secondary ion mass spectrometer, the formation of oxide films and the adhesion of organic substances on the surface are absent (or suppressed), making it difficult to perform analysis using such methods. The influence is suppressed, and a constant amount of standard sample can be obtained without variation in the analysis value of surface metal impurities.

At this time, in the oxidation-reduction potential adjustment step, the oxidation-reduction potential can be lowered by bubbling hydrogen gas into the aqueous hydrofluoric acid solution by blowing hydrogen gas to reduce the aqueous hydrofluoric acid solution.

In this way, the oxidation-reduction potential in the hydrofluoric acid aqueous solution can be effectively lowered with almost no pH fluctuation.

Further, in the redox potential adjustment step, the predetermined metal impurity dissolved in the hydrofluoric acid aqueous solution can be any one or more of Ni, Co, and Sn.

Ni and Co cause leak failure during standby, and Sn causes poor oxide film properties, so these are metal impurities that need to be quantified by quantitative analysis to investigate the contamination state.

Further, the predetermined metal impurities on the surface of the silicon wafer after the drying step can be quantified using a secondary ion mass spectrometer, and the quantified silicon wafer can be used as a standard sample for the secondary ion mass spectrometer.

In this way, it can be used as a standard sample for quantitative analysis in a secondary ion mass spectrometer.

At this time, the secondary ion mass spectrometer can be a time-of-flight secondary ion mass spectrometer.

In this way, the standard sample can be particularly suitable for quantitative analysis using a time-of-flight secondary ion mass spectrometer capable of accurate mass spectrometry.

The method for preparing a standard sample for surface metal contamination evaluation of the present invention enables surface analysis with excellent reproducibility and quantitative properties, without forming an oxide film on the silicon wafer surface and suppressing the adhesion of hydrocarbon ions. We can provide standard metal samples for

FIG. 3 is a flowchart showing steps in the method for producing a standard sample for surface metal contamination evaluation of the present invention. FIG. 2 is a correlation diagram showing the relationship between the oxidation-reduction potential and pH of Ni in an aqueous solution at 25°C. 2 is a graph showing the surface Ni concentration (solution potential: −0.4 V, pH: fixed at 2.8) versus Ni concentration in a 1% HF aqueous solution (25° C.). It is a graph showing the surface Co concentration (solution potential: −0.42 V, pH: fixed at 2.8) versus the Co concentration in a 1% HF aqueous solution (25° C.). It is a graph showing the surface Sn concentration (solution potential: -0.4V, pH: fixed at 2.8) versus the Sn concentration in a 1% HF aqueous solution (25° C.). This is the measurement result of ⁵⁸ Ni mass spectrum (Example) by TOF-SIMS. This is the measurement result of ⁵⁹ Co mass spectrum (Example) by TOF-SIMS. This is the measurement result of ¹²⁰ Sn mass spectrum (Example) by TOF-SIMS. This is a measurement result of ⁵⁸ Ni mass spectrum (Comparative Example 1) by TOF-SIMS. It is a graph showing a comparison (Example vs. Comparative Example 1) of average values of integrated intensities when measuring 10 samples. It is a graph showing a comparison (Example vs. Comparative Example 1) of variations in the average value of integrated intensity when measuring 10 samples. This is the measurement result of ⁵⁹ Co mass spectrum (Comparative Example 2) by TOF-SIMS. This is the measurement result of ¹²⁰ Sn mass spectrum (Comparative Example 2) by TOF-SIMS. It is a graph showing a comparison (Example vs. Comparative Example 2) of average values of integrated intensities when measuring 10 samples. It is a graph showing a comparison of variations in the average value of integrated intensity when measuring 10 samples (Example vs. Comparative Example 2).

Embodiments of the present invention will be described below with reference to the drawings, but the present invention is not limited thereto.
FIG. 1 is a flow diagram showing an example of the steps of the method for producing a standard sample for surface metal contamination evaluation in the present invention. The process is broadly divided into a wafer surface cleaning process, an oxidation-reduction potential adjustment process (solution potential adjustment process), an immersion treatment process, a rinsing process, a drying process, and a process for evaluating metal impurities on the wafer surface.

(Wafer surface cleaning process)
To prepare a standard sample for surface metal contamination evaluation, first, a silicon wafer is prepared for the standard sample (standard sample wafer), and the silicon wafer is cleaned to make the surface free of impurities such as metals.
The silicon wafer used here is, for example, a mirror-finished silicon wafer that has been subjected to final cleaning in the silicon substrate manufacturing process, has no metal contamination on the wafer surface or inside, and has a resistivity of 0.1 Ω・cm. It is preferable to use the above p-type or n-type silicon wafer.

In order to make this standard sample wafer clean by cleaning, it is placed in an ammonia-hydrogen peroxide aqueous solution at 80°C to remove particles, and then immersed in a 1% hydrofluoric acid aqueous solution (hereinafter also referred to as HF aqueous solution). ) to remove the natural oxide film containing metal impurities, then put it in a hydrochloric acid-hydrogen peroxide aqueous solution at 80°C to remove the metal impurities, and finally remove the natural oxide film with a 1% HF aqueous solution. .

(Oxidation-reduction potential adjustment step)
Next, a contamination treatment is performed to contaminate the surface of the standard sample wafer with a predetermined amount of a predetermined metal impurity.
At this time, the metal impurities in the HF aqueous solution are dissolved in a separately prepared HF aqueous solution, and the HF aqueous solution is reduced by bubbling hydrogen gas into the HF aqueous solution. The redox potential is lowered until it becomes a atomic state. The method is not particularly limited as long as the redox potential of the HF aqueous solution can be lowered, but if hydrogen gas blowing is performed as described above, the redox potential can be effectively lowered with almost no pH change.

The above-mentioned predetermined metal impurity (ie, contaminating metal) is preferably one or more of Ni, Co, and Sn, for example. The reason is that Ni and Co cause leak failure during standby, and Sn causes poor oxide film properties, so quantitative analysis requires quantitatively contaminated wafers that are accurately contaminated with Ni, Co, and Sn. .
On the other hand, Ni, Co, and Sn are usually ionized in the HF aqueous solution, and in this state, Ni, Co, and Sn cannot be attached to the standard sample wafer in the next immersion process. By reducing the Ni, Co, and Sn ions in the HF aqueous solution to the state of neutral atoms, they can be attached to the wafer surface, and a wafer with quantitative contamination can be produced.

Here, the pH of the aqueous solution, the oxidation-reduction potential (solution potential), and the form of the metal will be explained using Ni as an example. FIG. 2 is an oxidation-reduction potential-pH diagram of an aqueous solution at 25° C. containing 1 ppm of Ni. FIG. 2 shows the form of Ni in an aqueous solution when the pH and solution potential of the aqueous solution are at certain values.

For example, the pH of a 1% HF aqueous solution is approximately 3.0, and the solution potential at that time is +0.6V, so the form of Ni in the aqueous solution is Ni ²⁺ , and it is present as Ni ²⁺ ions in the HF aqueous solution. shows that it exists. In other words, if Ni exists as ions in an aqueous solution, even if a silicon wafer (standard sample wafer) is present in the aqueous solution, Ni will remain in the aqueous solution without adhering to the silicon wafer. , it is not possible to prepare a quantitatively contaminated sample even if immersion treatment or the like is performed as it is.
In order to attach metal in an aqueous solution to a silicon wafer, it is necessary to change the form of Ni in the aqueous solution to a neutral atomic state. When HF is used as a solvent for an aqueous solution, the pH is in the acidic range. Therefore, in order to convert Ni ²⁺ in the HF aqueous solution to Ni ⁰ (neutral Ni atoms), the solution potential must be changed to Ni ²⁺ It is necessary to lower (reduce) the potential below the boundary potential between Ni0 and ^Ni0 .

The boundary potential between Ni ²⁺ and Ni ⁰ in an aqueous solution can be calculated from the Nernst equation below.
E=E0+(2.303×RT/nF)×logA
The solution potential E of Ni under certain conditions is expressed by the above formula,
E0: Standard redox potential of Ni (Ni ²⁺ ⇔ ^{Ni 0} ): -0.257V
R: gas constant (8.314J/K/mol)
T: Temperature (K)
F: Faraday constant (96485C/mol)
A: Concentration (mol/L)
n: Can be determined from the number of electron transfers.
For example, the solution potential E of a 1 ppm Ni HF aqueous solution at 25°C is -0.398V, and the HF aqueous solution must be reduced to lower the oxidation-reduction potential (solution potential) E below -0.398V as shown in Figure 2. For example, it is shown that Ni exists in the state of Ni ⁰ in an HF aqueous solution.

Therefore, while measuring the solution potential, the oxidation-reduction potential of the HF aqueous solution is lowered by bubbling hydrogen gas into the HF aqueous solution as described above. Although hydrogen gas is slightly dissolved in water, it causes almost no change in pH. On the other hand, since the solution potential can be lowered to about -0.45V, by using this method, it is possible to effectively quantitatively contaminate a silicon wafer with Ni even in an HF aqueous solution.
In addition, this method is not limited to Ni, but it has been shown that it is also possible to quantitatively contaminate other metals, such as Ni, if the boundary potential at the standard redox potential of ionized/neutral atoms is around -0.4V. . Furthermore, since the process is carried out in an HF aqueous solution, there is an advantage that no oxide film is formed on the standard sample wafer and no organic matter is attached to it.

In preparing the standard sample in the present invention, as mentioned above, since there is no organic matter (interfering ion) such as hydrocarbons attached to the wafer surface, these substances affect the spectrum of the elemental ions to be detected such as Ni mentioned above. interference and can prevent detection at a higher intensity than the original (actual amount of contamination).
Furthermore, since no oxide film is generated on the wafer surface when metal contamination occurs, it is possible to prevent the accuracy of analysis of metal impurities on the surface from being affected by charge-up (electrification). Therefore, it is possible to prevent variations in analysis values and to prevent an increase in the influence of interfering ions due to a decrease in mass resolution.

(Soaking treatment process)
In the above step, the oxidation-reduction potential is lowered until the metal impurities in the HF aqueous solution become neutral atoms, and then the standard sample wafer is immersed for a predetermined time. The immersion time is not particularly limited, but if the wafer is immersed for 3 to 5 minutes, metal impurities can be sufficiently attached to the wafer surface.

(rinsing process)
Next, the HF aqueous solution (contamination treatment solution) adhering to the surface of the standard sample wafer is washed away by rinsing using pure water. The rinsing method is not particularly limited, and can be performed by, for example, spun the wafer and pouring pure water over it.

(drying process)
Next, the wafer is dried.
Note that the drying method may be spin drying or natural drying, but it is preferable to dry without using an alcohol solvent. This is to prevent a large amount of hydrocarbons from adhering to the wafer surface.

(Evaluation of metal impurities on the wafer surface)
After the drying process, the standard sample wafer is subjected to quantitative determination of predetermined metal impurities using a secondary ion mass spectrometer (SIMS), and the quantified standard sample wafer is used as a standard sample for the secondary ion mass spectrometer. Among these, it can be quantified using a time-of-flight secondary ion mass spectrometer (TOF-SIMS) and used as a standard sample. In this way, a standard sample optimal for an accurate mass spectrometry system can be obtained.

In particular, when performing quantitative analysis of metal impurities on the wafer surface using TOF-SIMS, it is important that the influence of interfering secondary ions on the target metal secondary ions is small, and that the wafer surface is covered with an insulating material such as an oxide film, leading to charge build-up. This requires a standard sample with less deterioration in mass resolution and less variation in detection intensity. According to the present invention, it is possible to provide a standard sample with excellent reproducibility and quantitative performance, which has no oxide film or organic matter on the surface and is particularly suitable for quantitative analysis of wafer surface metals by TOF-SIMS.

By the way, as mentioned above, for example, the solution potential E of a 1 ppm Ni HF aqueous solution at 25°C is -0.398V, and the HF aqueous solution must be reduced to lower the oxidation-reduction potential (solution potential) E below -0.398V. For example, Ni becomes a neutral atom in an HF aqueous solution and can contaminate the immersed standard sample wafer, but other metal impurities also have a boundary potential of -0 in the standard redox potential of ionized/neutral atoms. If it is about .4V, these metal impurities can be quantitatively contaminated in the same way. This point will be further explained below using experimental examples in which the metal impurities were contaminated with Ni, Co, and Sn.
Furthermore, how to control the contamination concentration (the relationship between the concentration of metal impurities in the HF aqueous solution and the concentration of contamination) will also be explained using experimental examples.

(Experiment 1)
Hydrogen gas was bubbled into a 1% HF aqueous solution (25° C.) containing 1 ppm Ni by blowing hydrogen gas, and the solution potentials were adjusted so that the solution potentials were different. The solution potential and pH at this time were measured, and a clean p-type silicon wafer (resistivity: 10 Ω·cm) with a diameter of 150 mm was immersed therein. Then, it was washed with pure water for 10 minutes and spin-dried at 3000 rpm for 20 seconds.
After gas-phase decomposition of the wafer surface with HF gas, the entire surface of the wafer is scanned with a mixed aqueous solution of 2% HF and 2% H ₂ O ₂ to recover Ni on the wafer surface, and the Ni concentration in the recovered liquid is determined by inductively coupled plasma. The surface Ni concentration was obtained by analyzing with a mass spectrometer (ELEMENT 2 manufactured by Thermo Fisher Scientific, hereinafter referred to as ICP-MS).
Table 1 shows the average Ni concentration and variation (RSD) when 10 wafers were subjected to the above-mentioned processing and analysis.

As shown in Table 1, when the solution potential is −0.4 V or less, the amount of Ni deposited on the surface is approximately constant at around 3.60×10 ¹³ atoms/cm ² . Furthermore, it can be said that the variation in analytical values is relatively small. In this way, it can be seen that contamination can be achieved with a certain amount by adjusting the solution potential to the state of neutral atoms of metal impurities.

(Experiment 2)
Next, we investigated the surface Ni concentration when changing the Ni concentration in a 1% HF aqueous solution (25°C) with the solution potential fixed at -0.4 V and the pH fixed at 2.8. Figure 3 shows the results of recovery using a similar method and measurement using ICP-MS.
As shown in FIG. 3, as the Ni concentration in the HF aqueous solution increases, the surface Ni concentration can be increased. By changing the Ni concentration in the HF aqueous solution in this way, the surface Ni concentration can be controlled. When preparing a standard sample in the present invention, the Ni concentration in the HF aqueous solution may be appropriately determined depending on the desired contamination concentration.

(Experiments 3 and 4)
In a 1% HF aqueous solution (25°C) containing 1 ppm each of Co (Experiment 3) and Sn (Experiment 4) instead of Ni, hydrogen gas was bubbled by hydrogen gas blowing in the same manner as in Experiment 1, and the solution potentials were different. The solution potential was adjusted as follows. Note that the ionization/neutral atom boundary potential (25° C.) of Co and Sn is −0.417V for Co and −0.282V for Sn.
The solution potential and pH at this time were measured, and a clean p-type silicon wafer (resistivity: 10 Ω·cm) with a diameter of 150 mm was immersed therein. Then, it was washed with pure water for 10 minutes and spin-dried at 3000 rpm for 20 seconds.
Then, the wafer surface is subjected to vapor phase decomposition using HF gas, and the entire surface of the wafer is scanned with a mixed aqueous solution of 2% HF and 2% H ₂ O ₂ to recover Co and Sn on the wafer surface, and the Co in the recovery liquid. By analyzing the concentration and Sn concentration by ICP-MS, the surface Co concentration and the surface Sn concentration were obtained.
Tables 2 and 3 show the average concentrations and variations (RSD) of Co and Sn when the above-described processing and analysis were performed on 10 wafers.

As shown in Table 2, when the solution potential is −0.42 V or less, the amount of Co deposited on the surface is approximately constant at around 3.30×10 ¹³ atoms/cm ² and the variation is relatively small.
Further, as shown in Table 3, when the solution potential is −0.3 V or less, the surface Sn adhesion amount is approximately constant at around 5.00×10 ¹³ atoms/cm ² and the variation is relatively small.

(Experiments 5 and 6)
In addition, the concentration of Co (Experiment 5) in a 1% HF aqueous solution (25°C) with a solution potential of -0.42V and a pH of 2.8, and a solution potential of -0.4V and a pH of 2. The surface Co concentration and the surface Sn concentration were collected using the same method as Experiments 3 and 4 when changing the concentration of Sn (Experiment 6) in a 1% HF aqueous solution (25 °C) fixed at .8. , the results measured by ICP-MS are shown in FIGS. 4 and 5.
As shown in FIGS. 4 and 5, similarly to Ni in Experiment 2, the surface Co concentration and surface Sn concentration can be controlled by changing the Co concentration and Sn concentration in the HF aqueous solution.

EXAMPLES Hereinafter, the present invention will be explained in more detail by showing Examples and Comparative Examples of the present invention, but the present invention is not limited thereto.
(Example)
As an example, a method for preparing a standard sample for evaluating surface metal contamination of Ni, Co, and Sn in the present invention and results of surface metal analysis by TOF-SIMS will be described.
The wafer used as a standard sample for surface metal contamination evaluation was a p-type PW wafer with a diameter of 150 mm and a resistivity of 9 to 12 Ω·cm that had been subjected to final cleaning in the silicon substrate manufacturing process.
(Wafer surface cleaning process)
In order to make this wafer clean, it is first placed in an ammonia-hydrogen peroxide aqueous solution at 80°C to remove particles, etc., then the native oxide film containing metal impurities is removed in a 1% HF aqueous solution, and then After removing metal impurities by placing it in a hydrochloric acid-hydrogen peroxide aqueous solution at 10°C, the natural oxide film is finally removed with a 1% HF aqueous solution.

(Oxidation-reduction potential adjustment process ~ immersion treatment process)
Apart from the above washing step, the following HF aqueous solution containing a certain amount of metal impurities was prepared.
Ni(NO ₃ ) ₂ , CoCl ₂ , and SnCl ₄ in 1000 ppm standard solution manufactured by Fuji Film Wako Pure Chemical Industries were adjusted to 10 ppb each in a 1% HF aqueous solution at 25°C, and hydrogen gas was added at a flow rate of 1 L/min. The solution potential and pH were measured while blowing and bubbling into the HF aqueous solution and adjusting the solution potential by stirring. The solution potential and pH were measured using a multi-digital water quality meter WQ-300 manufactured by HORIBA. Hydrogen gas blowing was continued until the solution potential became -0.4 V for the HF aqueous solution containing Ni and Sn, and -0.42 V for the HF aqueous solution containing Co, and then the wafer was immersed for 3 minutes.

(Rinse process - drying process)
The wafer subjected to the above immersion treatment was washed with ultrapure water for 10 minutes, and finally spin-dried at 3000 rpm for 20 seconds.
The Ni, Co, and Sn concentrations on the surfaces of these samples are as shown in the respective solution concentrations of 0.01 ppm in FIGS. 3 to 5 described above.
Ten of the above samples were produced.

(Metal impurity evaluation process)
Next, the prepared sample was introduced into TOF-SIMS (nano-TOF II manufactured by ULVAC-PHI), and positive secondary ions were analyzed using a Bi ion source. The analysis was performed on Ni (mass number 58), Co (mass number 59), and Sn (mass number 120), and mass spectra were obtained for each. The results are shown in FIGS. 6 to 8.

What is important when acquiring a mass spectrum is whether the sample surface is conductive or not, and whether there are interfering ions near the target ions. When ions are present and the interfering ion intensity is high, each mass peak in the mass spectrum becomes broad due to mass interference and fluctuations in sample potential, resulting in a problem of poor quantitative performance. Conventional methods that cause formation of oxide films and adhesion of organic substances have such problems.
However, in the present invention, since HF treatment is performed, there is no oxide film on the sample surface, and since no organic solvent is used, excessive organic components are not detected, and target ions (Ni, Co, Sn) and interfering ions ( ²⁹ Si ₂ , ²⁸ SiCH ₃ O, C ₆ O ₃ ) were properly separated, indicating that there was no problem in quantitative performance. Furthermore, as will be described later, the variation in analysis values is smaller than in Comparative Examples 1 and 2, and it can be said that the reproducibility is sufficiently high.

By using the wafer thus obtained as a standard sample for TOF-SIMS, highly accurate impurity evaluation is possible when actually evaluating metal contamination of a silicon wafer to be evaluated.

(Comparative example 1)
As Comparative Example 1, the method described in Patent Document 1 will be described.
The same silicon wafer as in the example was used as a wafer as a standard sample for surface metal contamination evaluation.
To clean this wafer, it was first placed in an ammonia-hydrogen peroxide aqueous solution at 80°C for 10 minutes to remove particles, and then placed in a 1% HF aqueous solution for 5 minutes to remove the natural oxide film containing metal impurities. The wafer was then placed in a hydrochloric acid-hydrogen peroxide aqueous solution at 80° C. for 10 minutes to remove metal impurities and form a natural oxide film on the wafer surface.

Apart from the above treatment, Ni(NO ₃ ) ₂ was adjusted to 10 ppb in an alkaline hydrogen peroxide aqueous solution (mixed aqueous solution of 29% ammonia water and 31% hydrogen peroxide water) containing a certain amount of metal impurities. was heated to 80°C. The wafer was placed in a heated solution and immersed for 10 minutes.
The immersed wafer was taken out, washed in pure water for 10 minutes, and then spin-dried at 3000 rpm for 20 seconds.
Ten of the above samples were prepared and analyzed by TOF-SIMS in the same manner as in the example. The results are shown in FIG. Furthermore, regarding the analysis value (integrated intensity), the results of comparing the average value of 10 sheets and its variation between Example and Comparative Example 1 are shown in FIGS. 10 and 11.

In the sample in Comparative Example 1, contaminant metal exists in the oxide film or on the surface of the oxide film, but in TOF-SIMS analysis, due to the presence of the oxide film (insulating film), the sample potential is affected by charge-up, and the mass Each mass peak in the spectrum becomes broad and causes mass interference (see FIG. 9). Therefore, the analysis value is higher than the original value (Example) (see FIG. 10). Furthermore, for the same reason, it is considered that the analytical values are more dispersed than in the case of the example in which no oxide film is present (see FIG. 11).

(Comparative example 2)
Next, as Comparative Example 2, the method described in Patent Document 3 will be described.
The same silicon wafer as in the example was used as a wafer as a standard sample for surface metal contamination evaluation.
To make this wafer clean, first, it was immersed in a 1% HF aqueous solution at 25°C for 5 minutes, then rinsed with pure water for 10 minutes, and then rinsed in a 65°C ammonia-hydrogen peroxide solution (1% ammonia/5% After the wafer was immersed in hydrogen peroxide (hydrogen peroxide) for 10 minutes, the wafer was rinsed with pure water for 10 minutes, and then air-dried.

Next, 1 mL of an isopropanol-containing solution containing 10 ppb of each of CoCl ₂ solution and SnCl ₄ solution was dropped onto the wafer surface and air-dried, and then the wafer was placed in a hydrogen gas atmosphere at 150°C. Reduction processing was performed.
Furthermore, ten of the above samples were prepared and analyzed by TOF-SIMS in the same manner as in the example. The results are shown in FIGS. 12 to 15.

It is presumed from the same document that in the sample of Comparative Example 2, the natural oxide film on the wafer surface disappeared due to the reduction treatment, and each metal chloride changed to a neutral metal. On the other hand, the organic components constituting isopropanol are thought to have volatilized due to the reduction treatment, but the results of TOF-SIMS measurements show that SiCH ₃ O, which is an interfering molecular ion in Co analysis, and C ₆ O, which is an interfering molecular ion in Sn analysis, ₃ was detected at a higher concentration than in the example, although there were variations, causing mass interference with each of Co and Sn (see FIGS. 12 and 13). This is considered to be the cause of the analysis value being higher than the original value (example) (see FIG. 14) and the cause of variation in the analysis value (see FIG. 15).

The specification includes the following aspects.
[1]: A method for preparing a standard sample that is quantitatively contaminated with a predetermined metal impurity used in surface metal contamination analysis of silicon wafers, the method comprising:
a wafer surface cleaning step of removing surface impurities from the silicon wafer prepared for the standard sample;
A redox potential adjustment step of dissolving the predetermined metal impurity in a hydrofluoric acid aqueous solution and lowering the redox potential until the predetermined metal impurity in the hydrofluoric acid aqueous solution becomes a neutral atom state;
A step of immersing the silicon wafer after the wafer surface cleaning step in the hydrofluoric acid aqueous solution after the oxidation-reduction potential adjustment step;
a rinsing step for removing the hydrofluoric acid aqueous solution adhering to the silicon wafer after the immersion treatment step;
A step of drying the silicon wafer after the rinsing step,
A method for producing a standard sample for surface metal contamination evaluation, comprising producing a standard sample in which the surface of a silicon wafer for the standard sample is contaminated with a quantitative amount of the predetermined metal impurity.
[2]: In the redox potential adjustment step, the surface metal of [1] above, in which hydrogen gas is bubbled into the hydrofluoric acid aqueous solution by hydrogen gas blowing to reduce the hydrofluoric acid aqueous solution and lower the redox potential. Method for preparing standard samples for contamination evaluation.
[3]: In the redox potential adjustment step, the predetermined metal impurity dissolved in the hydrofluoric acid aqueous solution is any one or more of Ni, Co, and Sn. ] Method for preparing standard samples for surface metal contamination evaluation.
[4]: The predetermined metal impurity on the surface of the silicon wafer after the drying step is quantified by a secondary ion mass spectrometer, and the quantified silicon wafer is used as a standard sample for the secondary ion mass spectrometer. The method for preparing a standard sample for surface metal contamination evaluation according to any one of [1] to [3] above.
[5]: The method for preparing a standard sample for surface metal contamination evaluation according to [4] above, wherein the secondary ion mass spectrometer is a time-of-flight secondary ion mass spectrometer.

Note that the present invention is not limited to the above embodiments. The above-mentioned embodiments are illustrative, and any embodiment that has substantially the same configuration as the technical idea stated in the claims of the present invention and has similar effects is the present invention. within the technical scope of the invention.

Claims (12)

A method for preparing a standard sample contaminated with a predetermined metal impurity quantitatively for use in surface metal contamination analysis of silicon wafers, the method comprising:
a wafer surface cleaning step of removing surface impurities from the silicon wafer prepared for the standard sample;
A redox potential adjustment step of dissolving the predetermined metal impurity in a hydrofluoric acid aqueous solution and lowering the redox potential until the predetermined metal impurity in the hydrofluoric acid aqueous solution becomes a neutral atom state;
A step of immersing the silicon wafer after the wafer surface cleaning step in the hydrofluoric acid aqueous solution after the oxidation-reduction potential adjustment step;
a rinsing step for removing the hydrofluoric acid aqueous solution adhering to the silicon wafer after the immersion treatment step;
A step of drying the silicon wafer after the rinsing step,
A method for producing a standard sample for evaluating surface metal contamination, comprising producing a standard sample in which the surface of a silicon wafer for use as a standard sample is contaminated with a quantitative amount of the predetermined metal impurity. 2. The method according to claim 1, wherein in the redox potential adjustment step, hydrogen gas is bubbled into the hydrofluoric acid aqueous solution by hydrogen gas blowing to reduce the hydrofluoric acid aqueous solution and lower the redox potential. Method for preparing standard samples for surface metal contamination evaluation. The surface according to claim 1, wherein in the redox potential adjustment step, the predetermined metal impurity dissolved in the hydrofluoric acid aqueous solution is any one or more of Ni, Co, and Sn. Method for preparing standard samples for metal contamination evaluation. The surface according to claim 2, wherein in the redox potential adjustment step, the predetermined metal impurity dissolved in the hydrofluoric acid aqueous solution is any one or more of Ni, Co, and Sn. Method for preparing standard samples for metal contamination evaluation. A claim characterized in that the predetermined metal impurity on the surface of the silicon wafer after the drying step is quantified by a secondary ion mass spectrometer, and the quantified silicon wafer is used as a standard sample for the secondary ion mass spectrometer. A method for preparing a standard sample for surface metal contamination evaluation according to item 1. A claim characterized in that the predetermined metal impurity on the surface of the silicon wafer after the drying step is quantified by a secondary ion mass spectrometer, and the quantified silicon wafer is used as a standard sample for the secondary ion mass spectrometer. A method for preparing a standard sample for surface metal contamination evaluation according to item 2. A claim characterized in that the predetermined metal impurity on the surface of the silicon wafer after the drying step is quantified by a secondary ion mass spectrometer, and the quantified silicon wafer is used as a standard sample for the secondary ion mass spectrometer. A method for preparing a standard sample for surface metal contamination evaluation according to item 3. A claim characterized in that the predetermined metal impurity on the surface of the silicon wafer after the drying step is quantified by a secondary ion mass spectrometer, and the quantified silicon wafer is used as a standard sample for the secondary ion mass spectrometer. A method for preparing a standard sample for surface metal contamination evaluation according to item 4. 6. The method for preparing a standard sample for evaluating surface metal contamination according to claim 5, wherein the secondary ion mass spectrometer is a time-of-flight secondary ion mass spectrometer. 7. The method for preparing a standard sample for evaluating surface metal contamination according to claim 6, wherein the secondary ion mass spectrometer is a time-of-flight secondary ion mass spectrometer. 8. The method for preparing a standard sample for evaluating surface metal contamination according to claim 7, wherein the secondary ion mass spectrometer is a time-of-flight secondary ion mass spectrometer. 9. The method for preparing a standard sample for evaluating surface metal contamination according to claim 8, wherein the secondary ion mass spectrometer is a time-of-flight secondary ion mass spectrometer.