JP2019124483A - Carbon concentration evaluation method - Google Patents

Abstract

To provide a method for evaluating a carbon concentration in a silicon single crystal simply and with high sensitivity.SOLUTION: A carbon concentration evaluation method for evaluating a carbon concentration in a silicon single crystal with a PL method or a CL method comprises: irradiating the silicon single crystal with an electron beam at an irradiation level of 1.0×10(electrons/cm) to measure an intensity of a G-line and an intensity of a C-line at a liquid helium temperature and determining an intensity ratio between the G-line and the C-line (Ci-Cs intensity/Ci-Oi intensity); and evaluating the carbon concentration [Cs] by substituting the intensity ratio (Ci-Cs intensity/Ci-Oi intensity) and an oxygen concentration [Oi] in the Si single crystal into the following formula: [Cs]=(4.45×10)*(Ci-Cs intensity/Ci-Oi intensity)*[Oi].SELECTED DRAWING: Figure 1

Description

  The present invention relates to a method of evaluating the carbon concentration in a silicon single crystal.

The carbon concentration evaluation method by the low temperature photoluminescence (PL) method irradiates the sample with an ion beam such as an electron beam or a carbon ion or an oxygen ion to generate a composite defect, and measures the luminescence intensity caused by the composite defect; It is a method of quantifying carbon concentration from the strength (for example, nonpatent literature 1, patent documents 1), and is a method which can quantify carbon concentration with high sensitivity rather than FT-IR method and SIMS method.
Further, Patent Literatures 2 to 4 disclose methods of preparing a calibration curve of composite defect strength and carbon concentration in a silicon single crystal, and quantifying the carbon concentration.

In Patent Document 2, a luminescence spectrum (W line) derived from interstitial silicon (I) introduced by irradiating a silicon single crystal with an electron beam and a value standardized by a light emission line (TO line) derived from silicon There is disclosed a method of preparing a calibration curve between carbon concentrations in a silicon single crystal and quantifying the carbon concentration from the W line / TO line obtained by the luminescence method.
Patent Document 3 discloses that interstitial carbon or G line (Ci-Cs) or C line (Ci-Oi) generated by implanting ions other than carbon and oxygen into silicon single crystal There is disclosed a method of preparing a calibration curve between carbon concentrations, and quantifying the carbon concentration from the spectral intensity of carbon-related complex defects.
In Patent Document 4, a calibration curve is created between the intensity ratio of G and C lines generated by irradiating a silicon single crystal with an electron beam and the concentration ratio of carbon and oxygen in the silicon single crystal. There is disclosed a method of measuring a carbon concentration from the oxygen concentration in a silicon single crystal and the intensity ratio of G-line and C-line.

Japanese Patent Application Laid-Open No. 04-344443 JP, 2015-111615, A JP, 2015-156420, A Unexamined-Japanese-Patent No. 2015-101529

M. Nakamura et al. , J. Electochem. Soc. 141, 3576 (1994)

As described above, many methods for evaluating carbon concentration by the PL method are disclosed. However, a plurality of silicon single crystals each having a known carbon concentration and different carbon concentrations are prepared, the luminescence intensity of the complex defect is measured, and a calibration curve with the carbon concentration is finally prepared. The concentration of carbon in the silicon single crystal is quantified by measuring the luminescence intensity of the complex defect of and applying this to the calibration curve.
That is, in the case of determining the carbon concentration in a silicon single crystal, there is a problem that carbon concentration can not be determined without preparing a plurality of samples with known carbon concentrations and performing these measurements.

  The present invention has been made in view of the problems of the prior art, and it is an object of the present invention to provide a method for evaluating carbon concentration in a silicon single crystal simply and at high sensitivity.

In order to achieve the above object, the present invention is a carbon concentration evaluation method for evaluating carbon concentration in a silicon single crystal by a photoluminescence method or a cathode luminescence method,
A composite of interstitial carbon Ci and substitutional carbon Cs generated in the silicon single crystal by irradiating the silicon single crystal with an electron beam at a dose of 1.0 × 10 ¹⁵ (electrons / cm ² ) Intensity (au) of emission line (G line) derived from defect Ci-Cs and intensity (a line) derived from complex defect Ci-Oi of interstitial carbon Ci and interstitial oxygen Oi U.) Is measured at liquid helium temperature, and the intensity ratio (Ci-Cs intensity / Ci-Oi intensity) of the G line and the C line is determined.
The carbon concentration [Cs] (atoms / cm ³ ),
The strength ratio (Ci-Cs strength / Ci-Oi strength) and the oxygen concentration [Oi] (ppma-JEITA) in the silicon single crystal can be expressed by the following formula [Cs] = (4.45 × 10 ¹⁴ ) · (Ci− Cs intensity / Ci-Oi intensity) [Oi]
The present invention provides a carbon concentration evaluation method characterized by substituting into.

The present invention is also a carbon concentration evaluation method for evaluating the carbon concentration in a silicon single crystal by a photoluminescence method or a cathode luminescence method,
A composite of interstitial carbon Ci and substitutional carbon Cs generated in the silicon single crystal by irradiating the silicon single crystal with an electron beam at a dose of 1.0 × 10 ¹⁵ (electrons / cm ² ) Intensity (au) of emission line (G line) derived from defect Ci-Cs and intensity (a line) derived from complex defect Ci-Oi of interstitial carbon Ci and interstitial oxygen Oi U.) Is measured at liquid nitrogen temperature, and the intensity ratio (Ci-Cs intensity / Ci-Oi intensity) of the G line and the C line is determined.
The carbon concentration [Cs] (atoms / cm ³ ),
The strength ratio (Ci-Cs strength / Ci-Oi strength) and the oxygen concentration [Oi] (ppma-JEITA) in the silicon single crystal can be expressed by the following formula [Cs] = (2.82 × 10 ¹⁴ ) · (Ci− Cs intensity / Ci-Oi intensity) [Oi]
The present invention provides a carbon concentration evaluation method characterized by substituting into.

  In the carbon concentration evaluation method according to the photoluminescence method or the cathode luminescence (CL) method of the present invention as described above, oxygen can be added to the above equation (hereinafter referred to as carbon concentration relationship equation or simply relationship equation) for obtaining carbon concentration. Since it is only necessary to substitute the concentration and the intensity ratio of G-line and C-line, when determining the carbon concentration in a silicon single crystal, prepare a silicon single crystal whose carbon concentration is known as in the past and prepare a calibration curve. It is not necessary to create it, and the carbon concentration can be easily obtained. Moreover, the carbon concentration can be quantified with higher sensitivity than the FT-IR method or the SIMS method.

  At this time, the oxygen concentration in the silicon single crystal can be 17 (ppma-JEITA) or less.

  In evaluating the carbon concentration, the oxygen concentration of the silicon single crystal to be evaluated is obtained and substituted into the above equation, but by setting the oxygen concentration of the silicon single crystal to 17 (ppma-JEITA) or less It is possible to prevent the S / N ratio from becoming worse. That is, at high oxygen concentration, since the formation of the C line is prioritized, the amount of Ci used to form the G line is reduced, and the intensity of the G line tends to be weak. However, by setting the above-described upper limit, it is possible to prevent the intensity of the G line from becoming too weak and to more accurately obtain the G line intensity.

  As described above, according to the carbon concentration evaluation method of the present invention, the carbon concentration can be evaluated easily and at high sensitivity without preparing a calibration curve as in the conventional method.

It is a flowchart which shows an example of the process of the carbon concentration evaluation method of this invention. It is a graph which shows the relationship between carbon concentration in a silicon single crystal, and proportionality constant alpha in case sample temperature is liquid helium temperature. It is a graph which shows the relationship of the carbon concentration evaluated by the Example and the comparative example.

Hereinafter, the embodiment of the present invention will be described with reference to the drawings, but the present invention is not limited to this.
First, from the oxygen concentration in silicon single crystal and the intensity ratio (Ci-Cs intensity / Ci-Oi intensity) of Ci-Cs (G line) and Ci-Oi (C line) which are carbon related complex defects, silicon single The process which the present inventors derived | led-out the formula (carbon concentration relational expression) which can evaluate carbon concentration in a crystal | crystallization simply is demonstrated.

  According to Non-Patent Document 1, when a silicon single crystal substrate is irradiated with a high energy electron beam, silicon atoms at lattice positions are repelled out, and interstitial silicon (hereinafter referred to as I) and vacancies that are escaped from it. A pair (hereinafter referred to as V) is generated. Excessively generated I and V are unstable as a single substance, so they recombine (V + I → 0), I or V clusters, or react with impurities contained in the silicon substrate to form a complex. Form

  When substitutional carbon Cs is present in the silicon substrate, interstitial carbon Ci is generated as I generated by electron beam irradiation repels Cs. Furthermore, Ci reacts with other Cs to form Ci-Cs (G line), or reacts with interstitial oxygen Oi, which is another impurity contained in the silicon substrate, to form Ci-Oi (C line). Form).

The equilibrium constants in the generation of the G and C lines are defined as K ₁ and K ₂ (the following formulas (1) and (2), [4] and [5] in Non-Patent Document 1), and Since the formation and the generation of C line are competitive reactions, the following formulas (3) and (4) are obtained.
[Oi] is interstitial oxygen concentration (also referred to simply as oxygen concentration), [Cs] is substitutional carbon concentration (also referred to simply as carbon concentration), [Ci] is interstitial carbon concentration, and [CiCs] is G-line intensity (Ci-Cs intensity) and [CiOi] represent C-line intensity (Ci-Oi intensity).

  Here, the following equation (5) is obtained by putting together the similar terms in the equations (3) and (4) and solving for [Cs].

  Here, the proportional constant α is a constant which varies depending on the sample temperature and the electron beam irradiation condition. The reason is that if the sample temperature is different, the luminous efficiencies of the G and C lines change, and the G and C line intensities obtained change. In addition, when the electron beam irradiation conditions are different, since the generation ratio of the G ray and the C ray is different, the proportional constant α changes.

Next, a method of determining the proportional constant α of the equation (5) will be described.
Fifteen levels of silicon single crystal substrates (samples) different in carbon concentration and oxygen concentration were prepared by the CZ method.
The reason for preparing silicon single crystal substrates having different oxygen concentrations is to verify that the carbon concentration can be quantified from one relational expression even with different oxygen concentrations. In addition, in order to be able to detect the oxygen concentration more reliably by FT-IR, and in order to prevent the deterioration of the S / N ratio of the G line more reliably because the Ci used for forming the C line increases. The oxygen concentration range of the silicon single crystal substrate was prepared here at 0.2 to 17 (ppma-JEITA).

Moreover, in order to increase the accuracy of the derived proportionality constant α, at least five levels are desired. Here, 15 levels of samples are prepared.
Next, the carbon concentration and oxygen concentration of these samples were measured by the FT-IR method.
Thereafter, each silicon single crystal substrate is irradiated with an electron beam of 1.0 × 10 ¹⁵ (electrons / cm ² ) at an acceleration voltage of 2 (MV) by an electron beam irradiation apparatus, G line and C Lines were generated and their peak intensities were measured by photoluminescence. The sample temperature at this time was the liquid helium temperature.

In these silicon single crystal substrates, the obtained carbon concentration, oxygen concentration, G-line intensity, and C-line intensity are substituted into the above equation (5), the horizontal axis is the carbon concentration, and the vertical axis is the proportional constant obtained It was plotted on the graph as α. The results are shown in FIG.
Since the proportional constant α became a substantially constant value, 4.45 × 10 ¹⁴ was obtained as the average value of the above 15 levels of samples as α. Here, each unit has a carbon concentration of atoms / cm ³ , an oxygen concentration of ppma-JEITA, a G-line intensity, and a C-line intensity as a relative value (au).
From the above, the carbon concentration relational expression (6) in the present invention described below was obtained.

Further, when the proportional constant α when the sample temperature is the liquid nitrogen temperature is determined in the same manner, it is 2.82 × 10 ¹⁴ .

Hereinafter, the carbon concentration evaluation method of this invention using the said relational expression (6) is demonstrated. FIG. 1 is a flow chart showing the steps of the evaluation method of the present invention.
First, a silicon single crystal substrate which is a sample for evaluation is prepared (step 1). The silicon single crystal substrate is not particularly limited, and may be, for example, a Czochralski method (CZ method) or a floating zone method (FZ method).
When evaluating the carbon concentration, the oxygen concentration is measured in a later step and substituted into the above relational expression, but at the time of the measurement, it is detected more reliably by FT-IR or SIMS, etc. As possible, for example, those of 0.02 (ppma-JEITA) or more are preferable. That is, by setting the concentration to 0.02 (ppma-JEITA) or more, the oxygen concentration can be determined more accurately, and therefore, the accuracy of the carbon concentration determined in the present invention can also be increased.
Moreover, in the intensity | strength measurement of G line, the thing of the range which is 17 (ppma-JEITA) or less of a range in which the S / N ratio of G line does not deteriorate, for example is preferable.
Further, the carbon concentration itself is not particularly limited.

  Next, the oxygen concentration (interstitial oxygen concentration) of the sample for evaluation is measured (step 2). For example, it can measure by FT-IR or SIMS. The method of measuring the oxygen concentration is not particularly limited as long as the oxygen concentration in the sample for evaluation can be determined.

Next, the intensity of the emission line (G line) derived from the composite defect Ci-Cs generated by irradiating the sample for evaluation with the electron beam by the PL method or the CL method, and the emission derived from the composite defect Ci-Oi The intensity of the line (C line) is measured, and the intensity ratio of the G line and the C line is determined (step 3). The peak intensities of each are measured to determine their ratio.
At this time, the temperature of the sample for evaluation is the liquid helium temperature or the liquid nitrogen temperature. At room temperature, the peaks of G-line and C-line become broad and the correct intensity can not be quantified.
Further, the electron beam irradiation amount is 1.0 × 10 ¹⁵ (electrons / cm ² ). By using this dose, it is possible to prevent the occurrence of problems such as G-line and C-line not being formed due to too small dose, and the silicon lattice is disturbed by too much, G-line and C-line. It is also possible to prevent the S / N ratio of the line from deteriorating.

  In addition, since the measurement depth can be changed by changing the wavelength of the laser light in the PL method by changing the acceleration voltage of electrons in the CL method, the desired depth from the sample surface can be adjusted by adjusting these conditions. Can be evaluated.

Then, the carbon concentration is evaluated by substituting the oxygen concentration and the intensity ratio of the G line and the C line into the carbon concentration relational expression (6) (step 4).
As described above, when the sample temperature at the time of measurement by the PL method or the CL method is the liquid helium temperature, the value of α in the equation (5) is set to 4.45 × 10 ¹⁴ as in the equation (6). However, when the sample temperature is liquid nitrogen temperature, the value of α is set to 2.82 × 10 ¹⁴ . This is because, as described above, when the sample temperature is different, the obtained G-line intensity and C-line intensity are different, and the proportional constant α is changed.

According to such an evaluation method of the present invention, in a silicon single crystal substrate whose carbon concentration is unknown, the oxygen concentration, G, without creating a calibration curve of the relationship between carbon concentration and luminescence intensity of complex defects as in the prior art. The carbon concentration can be quantified only by substituting the line intensity and the C-line intensity into the relational expression (6).
In other words, it is not necessary to prepare samples for preparing a calibration curve, and it is not necessary to measure the samples to prepare a calibration curve, so silicon single crystal is quicker, simpler and cheaper than before. It became possible to quantify the carbon concentration in the substrate. In addition, it can be evaluated at higher sensitivity than the FT-IR method or the SIMS method.

In addition, although the carbon concentration measurement by FT-IR method is standardized by JEITA (JEITA EM-3503), the detection lower limit is 2 × 10 ¹⁵ atoms / cm ³ , but according to the present invention, it is detected by FT-IR It has also become possible to quantify low carbon concentrations that can not be achieved.

Hereinafter, the present invention will be more specifically described with reference to examples of the present invention and comparative examples, but the present invention is not limited to these.
(Example)
A silicon single crystal substrate was cut out from a silicon single crystal ingot having different carbon concentrations and oxygen concentrations pulled up by the CZ method or FZ method, and 46 samples (hereinafter, referred to as samples for evaluating carbon concentration) were produced.

  Next, the oxygen concentration of the sample for carbon concentration evaluation was measured by the FT-IR apparatus. As a result, the oxygen concentration of 46 samples was in the range of 0.26 to 16.4 (ppma-JEITA).

Subsequently, the carbon concentration evaluation sample was irradiated with an electron beam by an electron beam irradiation apparatus (acceleration voltage 2 (MV), irradiation amount 1.0 × 10 ¹⁵ (electrons / cm ² )). With respect to the sample for carbon concentration evaluation irradiated with the electron beam, the emission spectrum is measured with the sample temperature as liquid helium temperature using a photoluminescence measuring device, G-line intensity and C-line intensity are obtained, G-line and C-line The intensity ratio of These values were substituted into the said relational expression (6), and the carbon concentration of each sample for carbon concentration evaluation was computed.

  In addition, the carbon concentration calculated here was compared with the value calculated | required by the conventional method, and the reliability of the evaluation method of this invention was confirmed. Therefore, a comparative example according to the conventional method will be described next.

(Comparative example)
The carbon concentration of the sample for carbon concentration evaluation was calculated by the conventional carbon concentration evaluation method by low temperature PL method (patent document 4). Details will be described below.
First, eight samples of silicon single crystal substrates (hereinafter, referred to as calibration curve samples) having different carbon concentrations and oxygen concentrations were prepared. Thereafter, the carbon concentration and the oxygen concentration of the calibration curve sample were measured by the FT-IR method, and the ratio of the carbon concentration to the oxygen concentration (hereinafter referred to as [Cs] / [Oi]) was calculated. Since the influence of the carbon concentration of the reference can not be ignored at low carbon concentration in FT-IR, a sample with a carbon concentration of 1.0 × 10 ¹⁵ (atoms / cm ³ ) or higher, which is a high concentration, was prepared. Moreover, the oxygen concentration prepared in the range of 3-12 (ppma-JEITA).

Next, the sample for calibration curve was irradiated with an electron beam by an electron beam irradiation apparatus (acceleration voltage 2 (MV), irradiation dose 1.0 × 10 ¹⁵ (electrons / cm ² ). Calibration irradiated with electron beam The emission spectrum of the sample for line was measured using a photoluminescence measurement apparatus to obtain G-line intensity and C-line intensity, and the sample temperature at this time was the liquid helium temperature as in the example. The ratio of the line intensity to the C-line intensity (hereinafter referred to as G / C) was calculated.
The obtained [Cs] / [Oi] and G / C were plotted for each sample, and a calibration curve was created between [Cs] / [Oi] and G / C.

  Subsequently, G-line intensities and C-line intensities are similarly determined for the carbon concentration evaluation samples used in the examples, and the ratio of G / C, which is the ratio between them, is applied to the calibration curve to obtain [Cs] / [Oi ] Was calculated. Furthermore, the carbon concentration was determined from the calculated value of [Cs] / [Oi] and the oxygen concentration of the measured carbon concentration evaluation sample.

In the carbon concentration evaluation sample, the relationship between the carbon concentration quantified by the above-mentioned conventional method (Cs concentration in the comparative example) and the carbon concentration quantified by the method of the present invention (Cs concentration in the example) is plotted in FIG. did.
From FIG. 3, it was confirmed that the present invention is effective because the carbon concentration (example) according to the method of the present invention matches very well the carbon concentration (comparative example) according to the conventional method.

  In the conventional carbon concentration evaluation method, a sample for calibration curve is prepared in addition to the silicon single crystal for which the carbon concentration is to be quantified, and the carbon concentration can not be quantified unless the calibration curve is prepared. As described above, it is a very time consuming task. Furthermore, in the conventional carbon concentration evaluation method, the carbon concentration can not be quantified without preparing a calibration curve each time the carbon concentration in the silicon single crystal is determined.

On the other hand, according to the present invention, if the electron beam irradiation conditions and the sample temperature at the time of measurement by the PL method or the CL method are the same, G line intensity, C line intensity and oxygen concentration are obtained and The carbon concentration can be determined simply by substituting.
Therefore, the present method can quantify the carbon concentration more quickly and easily than the conventional carbon concentration evaluation method.

Also, the carbon concentration was determined in the same manner as in the example except that the sample temperature at the time of measurement by the PL method was the liquid nitrogen temperature, and the proportional constant of the equation for determining the carbon concentration was changed to 2.82 × 10 ¹⁴ . Similar results to 3 were obtained.

  The present invention is not limited to the above embodiment. The above embodiment is an exemplification, and it has substantially the same configuration as the technical idea described in the claims of the present invention, and any one having the same function and effect can be used. It is included in the technical scope of the invention.

Claims (3)

A carbon concentration evaluation method for evaluating a carbon concentration in a silicon single crystal by a photoluminescence method or a cathode luminescence method,
A composite of interstitial carbon Ci and substitutional carbon Cs generated in the silicon single crystal by irradiating the silicon single crystal with an electron beam at a dose of 1.0 × 10 ¹⁵ (electrons / cm ² ) Intensity (au) of emission line (G line) derived from defect Ci-Cs and intensity (a line) derived from complex defect Ci-Oi of interstitial carbon Ci and interstitial oxygen Oi U.) Is measured at liquid helium temperature, and the intensity ratio (Ci-Cs intensity / Ci-Oi intensity) of the G line and the C line is determined.
The carbon concentration [Cs] (atoms / cm ³ ),
The strength ratio (Ci-Cs strength / Ci-Oi strength) and the oxygen concentration [Oi] (ppma-JEITA) in the silicon single crystal can be expressed by the following formula [Cs] = (4.45 × 10 ¹⁴ ) · (Ci− Cs intensity / Ci-Oi intensity) [Oi]
The carbon concentration evaluation method characterized by substituting into. A carbon concentration evaluation method for evaluating a carbon concentration in a silicon single crystal by a photoluminescence method or a cathode luminescence method,
A composite of interstitial carbon Ci and substitutional carbon Cs generated in the silicon single crystal by irradiating the silicon single crystal with an electron beam at a dose of 1.0 × 10 ¹⁵ (electrons / cm ² ) Intensity (au) of emission line (G line) derived from defect Ci-Cs and intensity (a line) derived from complex defect Ci-Oi of interstitial carbon Ci and interstitial oxygen Oi U.) Is measured at liquid nitrogen temperature, and the intensity ratio (Ci-Cs intensity / Ci-Oi intensity) of the G line and the C line is determined.
The carbon concentration [Cs] (atoms / cm ³ ),
The strength ratio (Ci-Cs strength / Ci-Oi strength) and the oxygen concentration [Oi] (ppma-JEITA) in the silicon single crystal can be expressed by the following formula [Cs] = (2.82 × 10 ¹⁴ ) · (Ci− Cs intensity / Ci-Oi intensity) [Oi]
The carbon concentration evaluation method characterized by substituting into.   The oxygen concentration in the said silicon single crystal shall be 17 (ppma-JEITA) or less, The carbon concentration evaluation method of Claim 1 or Claim 2 characterized by the above-mentioned.

Images