EVALUATION METHOD OF 3C-SiC SINGLE CRYSTAL SUBSTRATE

Abstract

To provide a method capable of widely evaluating the presence / absence of a vacancy on a boundary between a SiC substrate and a 3C-SiC single crystal film in a nondestructive manner and in a short-time by a Raman spectrum.SOLUTION: An evaluation method of a 3C-SiC single crystal substrate in which a 3C-SiC single crystal film is grown on a Si substrate evaluates the presence / absence of a vacancy on a boundary between the Si substrate and the 3C-SiC single crystal film with a Raman spectrum obtained by a Raman spectroscopy in which a laser beam is irradiated to a surface of the 3C-SiC single crystal film of the 3C-SiC single crystal substrate.SELECTED DRAWING: Figure 1

Description

Technical Field

[0001] The present invention relates to a method for evaluating a 3C-SiC single crystal substrate.

Background Art

[0002] Compared with the bandgap of Si of 1.1 eV, SiC has a wide bandgap of 2.2 to 3.3 eV, so it has a high breakdown strength and a high thermal conductivity, and is thus a material expected to be used as a semiconductor material for various semiconductor devices such as power devices and high-frequency devices.

[0003] In addition, although its use as a platform for GaN growth (for example, Patent Document 1 and Non-Patent Document 1) is also being advanced, on the other hand, SiC wafers are mainly of small diameter, and there is a demand for increasing the diameter for power devices and high-frequency devices. If a high-quality 3C-SiC single crystal film can be formed on a large-diameter substrate, in addition to the use of the 3C-SiC single crystal film itself, it becomes possible to fabricate a heteroepitaxial wafer having a large-diameter and high-quality GaN layer.

[0004] Therefore, as a method for increasing the diameter, epitaxial growth on a Si substrate with good compatibility with the device process has been studied (for example, Patent Documents 1 and 2). These patent documents disclose that a 3C-SiC single crystal film can be grown on a Si substrate, and that a 3C-SiC single crystal film can be grown on a large-diameter substrate such as 300 mm in diameter by selecting the type of reactor.

[0005] In addition, it is disclosed that when 3C-SiC is grown on a Si substrate under conditions where the pressure is 1333 Pa or less, voids are formed directly under the 3C-SiC single crystal film (for example, Patent Document 3). When a material of Group III-V such as GaN is heteroepitaxially grown on a 3C-SiC single crystal substrate having voids formed directly under such a 3C-SiC single crystal film, an effect of relaxing the stress of the entire heteroepitaxial layer is expected.

Prior Art Documents

Patent Document

[0006]

Patent Document 1

Patent Document 2

Patent Document 3

Patent Document 4

Non-Patent Document

[0007]

Non-Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0008] On the other hand, the formation of a 3C-SiC single crystal film under such reduced pressure causes a decrease in the growth rate. Therefore, it is necessary to study the optimal growth conditions while confirming the presence or absence of vacancies.

[0009] As an evaluation of the presence or absence of these vacancies, cross-sectional TEM from the 3C-SiC surface has been performed. However, cross-sectional TEM is a destructive test that requires a long time and is a local analysis, and a non-destructive and short-time method for evaluating a wide area has been demanded.

[0010] In Patent Document 4, for each of a Si substrate and a wafer on which a 3C-SiC layer is formed, P-polarized light is adjusted to the Brewster angle of Si or the angle of P-polarized light is adjusted so that the interference fringes become minimum and then incident, and the infrared transmission spectra of each are measured, and the transmittance or absorbance spectrum is calculated from the ratio of both to obtain at least the wave number of the Lo mode. A method for measuring the phonon wave number of a 3C-SiC layer on a Si substrate is disclosed. However, the evaluation of the presence or absence of vacancies formed directly under the 3C-SiC single crystal film is not mentioned.

[0011] The present invention has been made to solve the above problems, and an object thereof is to provide a method capable of evaluating the presence or absence of vacancies at the interface between a Si substrate and a 3C-SiC single crystal film in a non-destructive manner, in a short time, and over a wide range.

Means for Solving the Problems

[0012] The present invention has been made to achieve the above object, and is an evaluation method for a 3C-SiC single crystal substrate in which a 3C-SiC single crystal film is grown on a Si substrate. Laser light is irradiated onto the surface of the 3C-SiC single crystal film of the 3C-SiC single crystal substrate, and the presence or absence of vacancies at the interface between the Si substrate and the 3C-SiC single crystal film is evaluated based on the Raman spectrum obtained by Raman spectroscopy. An evaluation method for a 3C-SiC single crystal substrate is provided.

[0013] According to such an evaluation method for a 3C-SiC single crystal substrate, the presence or absence of vacancies at the interface between the Si substrate and the 3C-SiC single crystal film can be evaluated in a non-destructive manner, in a short time, and over a wide range.

[0014] At this time, the ratio (I Si ) of the signal intensity I of the To mode of 3C-SiC to the Si Raman signal intensity I obtained from the Raman spectrum. To Or the ratio (I To / I Si ) of the signal intensity I of the Lo mode of 3C-SiC to the Si Raman signal intensity I Si Or the ratio (I Lo ) of the signal intensity I of the Lo mode of 3C-SiC to the Si Raman signal intensity I Lo / ISi ) is used as a threshold value, and from this threshold value, the presence or absence of the vacancies existing directly under the 3C-SiC single crystal film can be determined. Thereby, the presence or absence of the vacancies can be determined more reliably.

[0015] When measuring the Raman spectrum, it is preferable to adjust the focus of the laser beam so that the intensity I of the To mode signal of the 3C-SiC To and the intensity I of the Lo mode signal of the 3C-SiC Lo become maximum. Thereby, the presence or absence of the vacancies can be determined with higher accuracy and stability.

[0016] In this case, when the value of I To / I Si is 0.25 or more or the value of I Lo / I Si is 0.11 or more, it can be determined that the vacancies are present. Thereby, the presence or absence of the vacancies can be determined even more reliably.

[0017] Also, when the Raman signal intensity at 417 cm -1 is observed, it can also be determined that the vacancies are present. Even with such a method, the presence or absence of the vacancies can be determined reliably.

Advantages of the Invention

[0018] As described above, according to the method for evaluating a 3C-SiC single crystal substrate of the present invention, the presence or absence of vacancies at the interface between the Si substrate and the 3C-SiC single crystal film can be evaluated non-destructively, in a short time, and over a wide range by Raman spectroscopy.

Brief Description of the Drawings

[0019]

Figure 1

Figure 2

Figure 3

Embodiments for Carrying Out the Invention

[0020] Hereinafter, the present invention will be described in detail, but the present invention is not limited thereto.

[0021] As described above, there has been a demand for a method capable of non-destructively, quickly, and widely evaluating the presence or absence of voids at the interface between an Si substrate and a 3C-SiC single crystal film.

[0022] As a result of intensive studies on the above problems, the present inventors have found an evaluation method for a 3C-SiC single crystal substrate on which a 3C-SiC single crystal film is grown on an Si substrate. By irradiating the surface of the 3C-SiC single crystal film of the 3C-SiC single crystal substrate with laser light and evaluating the presence or absence of voids at the interface between the Si substrate and the 3C-SiC single crystal film based on the Raman spectrum obtained by Raman spectroscopy, it is possible to non-destructively, quickly, and widely evaluate the presence or absence of voids at the interface between the Si substrate and the 3C-SiC single crystal film by using the Raman spectrum, and thus the present invention has been completed.

[0023] [First Evaluation Means] Hereinafter, the first evaluation means according to the first embodiment of the present invention will be described.

[0024] The evaluation method of the 3C-SiC single crystal substrate of the present invention irradiates the surface of the 3C-SiC single crystal film of the 3C-SiC single crystal substrate with laser light, and evaluates the presence or absence of voids at the interface between the Si substrate and the 3C-SiC single crystal film based on the Raman spectrum obtained by Raman spectroscopy.

[0025] In the first evaluation means, the ratio (I Si of the signal intensity I of the To mode of 3C-SiC to the signal intensity I of the Si Raman signal obtained from the Raman spectrum To ofTo / I Si ) or the Si Raman signal intensity I Si The intensity I of the Lo mode signal of 3C-SiC with respect to Lo (I Lo / I Si ) is used as a threshold value, and the presence or absence of vacancies existing directly under the 3C-SiC single crystal film is determined from this threshold value. Thereby, the presence or absence of the vacancies can be surely determined.

[0026] In order to set the threshold value in advance, laser light is irradiated onto the surface of the 3C-SiC single crystal film of a 3C-SiC single crystal substrate (sample for preliminary investigation) whose presence or absence of vacancies is unknown by Raman spectroscopic measurement as shown in FIG. 3, the generated Raman scattered light is spectroscopically analyzed by a diffraction grating, detected for each wavelength by a CCD detector, and the signal obtained by the CCD detector is converted into a Raman spectrum by a PC to obtain a Raman spectrum caused by the To mode and Lo mode (I To / I Si、 I Lo / I Si ) of Si and 3C-SiC.

[0027] When measuring the Raman spectrum, the focus of the laser light is adjusted so that the intensity I of the To mode signal of 3C-SiC To and the intensity I of the Lo mode signal of 3C-SiC Lo become maximum. Thereby, the presence or absence of the vacancies can be determined more accurately and stably.

[0028] The To mode and Lo mode indicate the transverse optical mode and longitudinal optical mode, respectively. The Si (519 cm -1 ) Raman signal intensity is considered to decrease due to thinning of the Si layer thickness by the presence of vacancies. Therefore, from the Raman spectrum obtained by Raman spectroscopy, the low position (hereinafter, also simply referred to as the "low position") and the high position (hereinafter, also simply referred to as the "high position") of the Raman signal intensity of Si with respect to the Raman signal intensities of the To mode and Lo mode of 3C-SiC are clarified.

[0029] When the positions with low and high Raman signal intensities of Si were identified from the obtained Raman spectrum, the values of I To / I Si、 I Lo / I Si were determined. I To and I Lo and I Si represent the To mode, Lo mode of 3C-SiC, and the Raman signal intensity (peak intensity) of Si, respectively. Multiple locations (for example, five points each) were measured at the low and high positions, and the average value and standard deviation were derived.

[0030] As a result of measuring five points each at the low and high positions, the average values were, respectively, I To / I Si = 0.33 and I Lo / I Si = 0.15 at the low position, and I To / I Si = 0.16 and I Lo / I Si = 0.07 at the high position.

[0031] At this time, the standard deviation σ was I To / I Si = 0.011 and I Lo / I Si = 0.009 at the low position, and I To / I Si = 0.003 and I Lo / I Si = 0.003 at the high position.

[0032] Also, the average values of the measurement results at the low and high positions were I To / I Si = 0.25 and I Lo / I Si = 0.11.

[0033] Next, FIB processing was performed on the surfaces at the low and high positions, and a cross-sectional sample of the 3C-SiC single crystal substrate was taken out. As a result, vacancies were observed at the 3C-SiC / Si interface at the low position. On the other hand, no vacancies were observed at the high position.

[0034] That is, it was confirmed that at the position where the Raman signal intensity of Si is low with respect to the Raman signal intensity in the To mode and Lo mode of 3C-SiC, vacancies exist, and at the position where it is high, no vacancies exist.

[0035] When measured at a position with vacancies and at a position without vacancies, the I To / I Si and I Lo / I Si average values were respectively I To / I Si = 0.25, I Lo / I Si = 0.11. This average value is outside the range of ±3σ of the values of I To / I Si and I Lo / I Si measured at a position with vacancies, and I To / I Si and I Lo / I Si measured at a position without vacancies. By using the average value as a threshold, the presence or absence of vacancies can be evaluated based on the threshold.

[0036] That is, it can be determined that vacancies exist when the value of I To / I Si is 0.25 or more or the value of I Lo / I Si is 0.11 or more. Thereby, the presence or absence of the vacancies can be reliably determined with objective numerical values.

[0037] I To / I Si = 0.17 to 0.30, I Lo / I Si = 0.08 to 0.12 range is outside the range of ±3σ of the values of I To / I Si and I Lo / I Si measured at a position with vacancies and at a position without vacancies. It is not necessary to fix it to the average value, and the above range can also be used as a threshold.

[0038] Since the ratio of the Raman signal intensities in the To mode and Lo mode of 3C-SiC measured at five points each at the positions with and without voids is constant, it is possible to evaluate the presence or absence of voids using only one of the To mode and Lo mode Raman signals.

[0039] At this time, the thickness of the 3C-SiC single crystal film can be, for example, 280 nm, and the wavelength of the laser light can be, for example, 325 nm, but it is not limited to this. However, since the penetration length of the laser light increases as the wavelength becomes longer, it is preferable to appropriately set the wavelength of the laser light according to the thickness of the 3C-SiC film.

[0040] Next, using the threshold value obtained above, the voids in the 3C-SiC single crystal substrate are evaluated. By Raman spectroscopic measurement as shown in FIG. 3, laser light is irradiated onto the surface of the 3C-SiC single crystal film of a 3C-SiC single crystal substrate (evaluation sample) whose presence or absence of voids to be evaluated is unknown, the generated Raman scattered light is dispersed by a diffraction grating, detected for each wavelength by a CCD detector, and the signal obtained by the CCD detector is converted into a Raman spectrum by a PC. Note that the focus of the laser light is adjusted so that the signal intensities in the To mode and Lo mode of 3C-SiC are maximized. For example, the focus of the Hd-Cd laser is adjusted by changing the height of the stage on which the evaluation sample is placed.

[0041] From the obtained Raman spectrum, the value of I To / I Si、 I Lo / I Si is calculated. I To , I Lo , I Si are the Raman signal intensities (peak intensities) in the To mode, Lo mode, and Si of 3C-SIC, respectively.

[0042] I To / I Si、 I Lo / I Si The above-mentioned threshold value of I To / I Si、 I Lo / ISi The presence or absence of voids is determined by comparing the values of I To / I Si、 I Lo / I Si Measurement points where the value is equal to or greater than each threshold value are determined to indicate the presence of vacancies directly below the 3C-SiC single crystal film.

[0043] The determination of the presence or absence of voids according to the present invention described above can be performed non-destructively and over the entire area of the substrate surface, so it is possible not only to determine the presence or absence of voids, but also to determine, for example, the void area and distribution.

[0044] [Second evaluation method] Next, a second evaluation means according to the second embodiment of the present invention will be described.

[0045] In the second evaluation method, in the obtained Raman spectrum, -1 When a Raman signal intensity of 0.05 or more is observed, it is determined that a vacancy exists. This method also makes it possible to reliably determine whether or not there are voids.

[0046] Using Raman spectroscopy as shown in Figure 3, laser light is irradiated onto the surface of the 3C-SiC single crystal film of the 3C-SiC single crystal substrate (evaluation sample) whose presence or absence of vacancies is unknown, and a Raman spectrum is obtained.

[0047] In the obtained Raman spectrum, 417 cm -1 When the Raman signal is confirmed, it is determined that there is a hole.

[0048] Measurements were taken five times at positions with and without holes, and the measurement was always 417 cm -1 A Raman peak of the following was observed.

[0049] 417cm at the hole location -1 The Raman signal at 417cm is thought to be due to the influence of the interface between the vacancies and the 3C-SiC or Si. -1The difference in the intensity of the Raman signals is thought to reflect the difference in the state of the interface between the vacancies and the 3C-SiC or Si. [Example]

[0050] The present invention will be specifically explained below with reference to examples, but the present invention is not limited to these examples.

[0051] (Examples 1 and 2) Four samples were used, each of which had a 280 nm thick 3C-SiC single crystal film heteroepitaxially grown on a 300 mm diameter Si single crystal substrate.

[0052] I of the first evaluation means To / I Si、 I Lo / I Si The threshold was set according to the threshold setting method of I. The wavelength of the laser light was 325 nm. Each threshold was calculated as the average value of 10 points, 5 high points and 5 low points. To / I Si =0.25, I Lo / I Si =0.11.

[0053] The laser Raman microscope used was the Nanophoton Laser Raman microscope RAMANdrive UV-VIS-NIR-SHK, and the measurement conditions were: excitation wavelength: 325 nm, diffraction grating: 3600 lines, resolution: 1.8 cm -1 It was decided.

[0054] A laser beam was irradiated near the center of the surface of the 3C-SiC single crystal film on the 3C-SiC single crystal substrate, and the resulting Raman scattered light was dispersed by a diffraction grating and detected for each wavelength by a CCD detector. The signal obtained by the CCD detector was converted into a Raman spectrum by a PC. The focus of the laser beam was adjusted to maximize the signal intensity of the To mode and Lo mode of 3C-SiC.

[0055] Two of the four samples showed low Raman signal intensities of Si. These were designated as samples in Examples 1 and 2.

[0056] For the samples of Examples 1 and 2, I was obtained from the Raman spectra. To / I Si and I To / I Si The values of I were calculated. To / I Si =0.34, I Lo / I Si =0.16, Example 2 is I To / I Si =0.34, I Lo / I Si =0.15.

[0057] The calculated I To / I Si、 I Lo / I Si Since the value of was equal to or greater than the threshold value, Examples 1 and 2 were determined to have voids.

[0058] To confirm the results of the evaluation, cross-sectional TEM observation was performed, and voids were confirmed directly below the 3C-SiC single crystal film.

[0059] (Examples 3 and 4) Of the four samples, the remaining two, which showed high Raman signal intensities of Si, were used as samples of Examples 3 and 4.

[0060] Examples 3 and 4 I To / I Si、 I Lo / I Si When the values of I To / I Si =0.16 、 I Lo / I Si =0.07.

[0061] The calculated I To / I Si、 I Lo / I Si Since the value of was equal to or less than the threshold value, Examples 3 and 4 were determined to be free of voids.

[0062] Cross-sectional TEM observation was performed to confirm the results of the evaluation, and no voids were found at the interface between the Si substrate and the 3C-SiC single crystal film.

[0063] I in Examples 1 to 4 To / I Si、 I Lo / I Si The calculated values, the results of the determination of the presence or absence of voids, and the results of cross-sectional TEM observations carried out to confirm the determination results are shown in Table 1.

[0064] [Table 1]

[0065] (Examples 5 and 6) In Examples 5 and 6, from the Raman spectra obtained in Examples 1 and 2, -1 The presence or absence of Raman signals was determined.

[0066] As a result, in both Examples 5 and 6, -1 Since a Raman signal of the above was confirmed, it was determined that voids were present in both Examples 5 and 6. The determination results and the cross-sectional TEM observation results (voids present) in Examples 1 and 2 were consistent.

[0067] (Examples 7 and 8) As Examples 7 and 8, from the Raman spectra obtained in Examples 3 and 4, 417 cm -1 The presence or absence of Raman signals was determined.

[0068] As a result, in both Examples 7 and 8, -1 Since no Raman signal was confirmed, it was determined that there were no pores in both Examples 7 and 8. The determination results and the cross-sectional TEM observation results (no pores) in Examples 3 and 4 were consistent.

[0069] 417 cm in Examples 5 to 8 -1 Table 2 shows the results of the determination of the presence or absence of Raman signals and voids, as well as the results of cross-sectional TEM observations carried out to confirm the determination results.

[0070]

Table 2

[0071] Fig. 1 shows the Raman spectrum of the surface of a 3C-SiC single crystal substrate determined to have pores, and Fig. 2 shows the Raman spectrum of the surface of a 3C-SiC single crystal substrate determined to have no pores.

[0072] In the Raman spectrum of Fig. 1, signals of Si: 519 cm -1 , To: 792 cm -1 , Lo: 966 cm -1 were detected. Furthermore, a Raman signal of 417 cm -1 was observed. When observing this position by cross-sectional TEM, it was confirmed that pores were formed directly below the 3C-SiC single crystal film.

[0073] In the Raman spectrum of Fig. 2, signals of Si: 519 cm -1 , To: 792 cm -1 , Lo: 966 cm -1 were detected, but the Raman signal of 417 cm -1 was not observed. When observing this position by cross-sectional TEM, no pores were confirmed directly below the 3C-SiC single crystal film.

[0074] (Comparative Example) For the 3C-SiC single crystal substrates determined to have pores in Example 1 and Example 2, cross-sectional TEM observations were performed at arbitrary locations. However, the locations where pores were formed were partial, and it was difficult to reliably capture the formation of pores.

[0075] As described above, according to the examples of the present invention, for a 3C-SiC single crystal substrate on which a 3C-SiC single crystal film was grown on an Si substrate, the presence or absence of pores existing at the interface between the Si substrate and the 3C-SiC single crystal film could be determined from the Raman spectrum obtained by Raman spectroscopy.

[0076] In this way, the present invention can easily measure the formation of voids directly below a 3C-SiC single crystal film over a wide area in a short amount of time in a non-destructive manner, and can also reliably identify the location of void formation.

[0077] The present specification includes the following aspects. [1]: A method for evaluating a 3C-SiC single crystal substrate in which a 3C-SiC single crystal film is grown on a Si substrate, the method comprising: irradiating a laser beam onto the surface of the 3C-SiC single crystal film of the 3C-SiC single crystal substrate; and evaluating the presence or absence of voids at the interface between the Si substrate and the 3C-SiC single crystal film using a Raman spectrum obtained by Raman spectroscopy. [2]: Si Raman signal intensity I obtained from the Raman spectrum Si To mode signal intensity of 3C-SiC vs. I To The ratio (I To / I Si ) or the Si Raman signal intensity I Si Lo-mode signal intensity of 3C-SiC versus I Lo The ratio (I Lo / I Si ) as a threshold value, and determining the presence or absence of voids present directly below the 3C-SiC single crystal film from the threshold value. [3]: When measuring the Raman spectrum, the To mode signal intensity I To and the Lo mode signal intensity I of the 3C-SiC Lo The method for evaluating a 3C-SiC single crystal substrate according to [2] above, further comprising adjusting the focus of the laser light so that the [4]: I To / I Si The value of is 0.25 or more or Lo / I Si The method for evaluating a 3C-SiC single crystal substrate according to [2] or [3] above, further comprising determining that the vacancies are present when the value of is 0.11 or more. [5]:417cm -1The evaluation method of the 3C-SiC single crystal substrate of the above [1], the above [2], the above [3] or the above [4], which includes determining that the vacancy exists when the Raman signal intensity of

[0078] Note that the present invention is not limited to the above embodiment. The above embodiment is an example, and any configuration that has substantially the same configuration as the technical idea described in the claims of the present invention and exhibits the same operational effects is included in the technical scope of the present invention.

Claims

1. A method for evaluating a 3C-SiC single crystal substrate on which a 3C-SiC single crystal film is grown on an Si substrate, the method comprising irradiating a laser beam onto the surface of the 3C-SiC single crystal film of the 3C-SiC single crystal substrate and evaluating the presence or absence of vacancies at the interface between the Si substrate and the 3C-SiC single crystal film based on a Raman spectrum obtained by Raman spectroscopy. A method for evaluating a 3C-SiC single crystal substrate, characterized in that.

2. The Si Raman signal intensity I obtained from the Raman spectrum Ｓｉ to the 3C-SiC To mode signal intensity I Ｔｏ ratio (I Ｔｏ / I Ｓｉ ), or the 3C-SiC Lo mode signal intensity I Ｓｉ to the Si Raman signal intensity I Ｌｏ ratio (I Ｌｏ / I Ｓｉ ), and using this threshold value to determine the presence or absence of the vacancies existing directly under the 3C-SiC single crystal film. The evaluation method of the 3C-SiC single crystal substrate according to claim 1, characterized in that

3. When measuring the Raman spectrum, the focus of the laser beam is adjusted so that the TO mode signal intensity I of the 3C-SiC Ｔｏ and the LO mode signal intensity I of the 3C-SiC Ｌｏ become maximum. The method for evaluating a 3C-SiC single crystal substrate according to claim 2, characterized in that.

4. The above-mentioned I Ｔｏ / I Ｓｉ has a value of 0.25 or more, or when the value of the above-mentioned I Ｌｏ / I Ｓｉ has a value of 0.11 or more, it is determined that the pores are present. The method for evaluating a 3C-SiC single crystal substrate according to claim 2 or claim 3, characterized in that.

5. 417 cm －１ The method for evaluating a 3C-SiC single crystal substrate according to claim 1, wherein when the Raman signal intensity of －１ is observed, it is determined that the vacancy exists.

Images