Method for evaluating shape of semiconductor substrate

Abstract

The present invention provides a novel technique capable of evaluating the shape of a semiconductor substrate. The present invention comprises a quantification step for quantifying a shape feature of a semiconductor substrate on the basis of: a main surface scattered light intensity measurement value obtained by measuring the intensity of scattered light as a result of scattering of laser light that entered from the main surface of the semiconductor substrate; and a back surface scattered light intensity measurement value obtained by measuring the intensity of scattered light as a result of scattering of laser light that entered from the back surface of the semiconductor substrate.

Description

Method for evaluating the shape of a semiconductor substrate

[0001] This invention relates to a technique for evaluating the shape of a semiconductor substrate.

[0002] In recent years, the introduction of new technologies and materials aimed at improving the performance and functionality of semiconductor devices has led to a surge in the manufacturing costs of semiconductor devices.

[0003] In response to these rising costs, efforts have been made to reduce costs by increasing the diameter of semiconductor substrates, which are the materials used in semiconductor devices.

[0004] On the other hand, with the introduction of new technologies and materials, the influence of substrate shape on the quality of semiconductor devices is increasing, and evaluation standards for substrate shape are becoming stricter year by year.

[0005] In particular, if a wafer is warped, it has a significant impact on the yield of semiconductor devices and the manufacturing process, so it is necessary to evaluate warping rigorously.

[0006] As an example of a method for evaluating warpage, Patent Document 1 describes a method for predicting the warpage of a SiC substrate by evaluating the internal stress of the SiC substrate using Raman scattering light.

[0007] Japanese Patent Publication No. 2015-59073

[0008] Evaluation methods that utilize Raman scattering require measuring Raman scattering, which is much weaker than elastic scattering such as Rayleigh scattering, and therefore take a considerable amount of time to perform.

[0009] Therefore, when introducing evaluation methods using Raman scattering light into the manufacturing process, there was a problem of reduced throughput.

[0010] In view of the above-mentioned problems, the problem that the present invention aims to solve is to provide a novel technology that can evaluate the shape of a semiconductor substrate.

[0011] The present invention, which solves the above problems, is based on the technical concepts shown in [1] to

[13] .

[0012] [1] A method for evaluating the shape of a semiconductor substrate, comprising a quantification step of quantifying the shape characteristics of the semiconductor substrate based on a main surface scattered light intensity measurement obtained by measuring the intensity of scattered light from laser light incident on the main surface of the semiconductor substrate and a back surface scattered light intensity measurement obtained by measuring the intensity of scattered light from laser light incident on the back surface of the semiconductor substrate.

[0013] [2] The semiconductor substrate shape evaluation method according to [1], wherein the quantification step quantifies the shape characteristics of the semiconductor substrate based on the ratio of the main surface scattered light intensity measurement value and the back surface scattered light intensity measurement value.

[0014] [3] The semiconductor substrate shape evaluation method according to [2], wherein the quantification step quantifies the shape characteristics of the semiconductor substrate using the ratio of the main surface scattered light intensity measurement value and the back surface scattered light intensity measurement value as a shape characteristic parameter indicating the shape characteristics of the semiconductor substrate.

[0015] [4] The shape characteristics of the semiconductor substrate include the amount of warping of the semiconductor substrate, as described in any of [1] to [3].

[0016] [5] A method for evaluating the shape of a semiconductor substrate according to any one of [1] to [4], wherein in the quantification step, the main surface scattered light intensity measurement is obtained by measuring the intensity of elastic scattered light obtained by elastically scattering laser light incident on the main surface of the semiconductor substrate, and the back surface scattered light intensity measurement is obtained by measuring the intensity of elastic scattered light obtained by elastically scattering laser light incident on the back surface of the semiconductor substrate.

[0017] [6] The semiconductor substrate shape evaluation method according to any one of [1] to [5], wherein the quantification step quantifies the shape characteristics of the semiconductor substrate based on the position parameter of the distribution of a plurality of main surface scattered light intensity measurements obtained by measuring the intensity of scattered light scattered by laser light incident on the main surface of the semiconductor substrate at a plurality of measurement points on the main surface of the semiconductor substrate, and the position parameter of the distribution of a plurality of back surface scattered light intensity measurements obtained by measuring the intensity of scattered light scattered by laser light incident on the back surface of the semiconductor substrate at a plurality of measurement points on the back surface of the semiconductor substrate.

[0018] [7] The semiconductor substrate shape evaluation method according to any one of [1] to [6], wherein the quantification step quantifies the shape characteristics of the semiconductor substrate based on image calculations of a distribution image of a plurality of main surface scattered light intensity measurements obtained by measuring the intensity of scattered light scattered by laser light incident on the main surface of the semiconductor substrate at a plurality of measurement points on the main surface of the semiconductor substrate, and a distribution image of a plurality of back surface scattered light intensity measurements obtained by measuring the intensity of scattered light scattered by laser light incident on the back surface of the semiconductor substrate at a plurality of measurement points on the back surface of the semiconductor substrate.

[0019] [8] A semiconductor substrate shape evaluation method according to any one of [1] to [7], wherein the quantification step includes an estimation step of estimating shape feature parameters indicating the shape characteristics of the target substrate from the main surface scattered light intensity measurement and the back surface scattered light intensity measurement of the target substrate using an estimation model created based on a dataset relating the main surface scattered light intensity measurement and the back surface scattered light intensity measurement of the reference substrate and shape feature parameters indicating the shape characteristics of the reference substrate.

[0020] [9] The semiconductor substrate shape evaluation method according to [8], wherein the estimation model is an estimation model created based on a dataset relating the ratio of the main surface scattered light intensity measurement value and the back surface scattered light intensity measurement value in the reference substrate to the shape feature parameters of the reference substrate, and the estimation step is to use the estimation model to estimate the shape feature parameters of the target substrate based on the ratio of the main surface scattered light intensity measurement value and the back surface scattered light intensity measurement value in the target substrate.

[0021]

[10] The shape characteristic parameter includes a warpage parameter indicating the amount of warpage of the semiconductor substrate, as described in [8] or [9].

[0022]

[11] The method for evaluating the shape of a semiconductor substrate according to any one of [8] to

[10] , wherein the laser light is S-polarized laser light, and the shape characteristic parameter includes a full-surface adsorption analysis parameter measured by adsorbing the entire back surface of the semiconductor substrate onto a vacuum chuck.

[0023]

[12] The method for evaluating the shape of a semiconductor substrate according to any one of [8] to

[10] , wherein the laser light is P-polarized laser light, and the shape characteristic parameters include non-adsorption analysis parameters measured by adsorbing three or one point on the back surface of the semiconductor substrate.

[0024]

[13] A method for creating an estimation model used to estimate the shape characteristics of a semiconductor substrate, based on a dataset that associates the ratio of a main surface scattered light intensity measurement obtained by measuring the intensity of scattered light from a laser beam incident on the main surface of a reference substrate with a back surface scattered light intensity measurement obtained by measuring the intensity of scattered light from a laser beam incident on the back surface of a reference substrate with shape characteristic parameters that indicate the shape characteristics of the reference substrate.

[0025] The present invention provides a novel technology that can evaluate the shape of a semiconductor substrate.

[0026] This figure shows an example of a scattered light intensity measuring device used for measuring scattered light intensity according to an embodiment of the present invention. This figure illustrates the relationship between the polarization of the laser light and the processed altered layer in the measurement of scattered light intensity according to an embodiment of the present invention. This figure shows the main surface scattered light intensity measurement value and the back surface scattered light intensity measurement value obtained by incident S-polarized light in the measurement of scattered light intensity according to an embodiment of the present invention. This figure shows the main surface scattered light intensity measurement value and the back surface scattered light intensity measurement value obtained by incident P-polarized light in the measurement of scattered light intensity according to an embodiment of the present invention. This is a data plot showing the relationship between GBIR and the ratio of the main surface scattered light intensity measurement value to the back surface scattered light intensity measurement value. This is a data plot showing the relationship between SORI and the ratio of the main surface scattered light intensity measurement value to the back surface scattered light intensity measurement value. This is a data plot showing the relationship between BOW and the ratio of the main surface scattered light intensity measurement value to the back surface scattered light intensity measurement value. This is a data plot showing the relationship between GBIR and the main surface scattered light intensity measurement value. This is a data plot showing the relationship between SORI and the main surface scattered light intensity measurement value. This is a data plot showing the relationship between BOW and the main surface scattered light intensity measurement value. This is a data plot showing the relationship between GBIR and back-surface scattered light intensity measurements. This is a data plot showing the relationship between SORI and back-surface scattered light intensity measurements. This is a data plot showing the relationship between BOW and back-surface scattered light intensity measurements. This is a distribution image mapping the ratio of main-surface scattered light intensity measurements to back-surface scattered light intensity measurements when S-polarized light is incident at the same point on both sides of SiC substrate A. This is a distribution image mapping the ratio of main-surface scattered light intensity measurements to back-surface scattered light intensity measurements when P-polarized light is incident at the same point on both sides of SiC substrate A. This is a 3D distribution image of GBIR, SORI, and BOW on SiC substrate A.

[0027] The following describes a method for evaluating the shape of a semiconductor substrate according to an embodiment of the present invention, using the attached drawings. Note that the embodiments shown below are examples of the present invention and the present invention is not limited to these embodiments.

[0028] In the present invention, the scattered light intensity measurement may be expressed as numerical data such as statistical values, including a position parameter (e.g., including one or more selected from mean (arithmetic, geometric, harmonic), median (quantile, ordinal statistic), mode, class value, etc.) and a scale parameter (e.g., including one or more selected from variance, standard score, standard deviation, mean deviation, median absolute deviation, range, full width at half maximum, etc.), obtained by statistically processing a plurality of measurement values ​​measured over the entire surface of the main or back surface of the substrate. Alternatively, the scattered light intensity measurement may be expressed as image data such as a distribution image obtained by mapping a plurality of measurement values ​​measured over the entire surface of the main or back surface of the substrate to the substrate, or a histogram of those plurality of measurement values.

[0029] In the present invention, the shape characteristics of a semiconductor substrate or the shape characteristic parameters obtained by quantifying the same include the amount of warpage of the semiconductor substrate or the warpage parameters obtained by quantifying the same, specifically GF3R (TIR), GF3D (FPD), GFLR (NTV), GFLD (NTD), GBIR (TTV), TAPER, TILTANGLE, SF3R (LTIR(3)), SF3D (LFPD(3)), SFLR (LTIR This indicates one or more parameters selected from (BF), SFLD (LFPD(BF)), LTIR(S3), LFPD(S3), SFQR(LTIR(SBF)), SFQD(LFPD(SBF)), SBIR(LTV), SBID(LDOF), WARP(GF3YFER, GF3NFER), BOW(GF3YFCD, GF3NFCD), SORI(GFLYFER, GFLNFER), etc.

[0030] Of the above parameters, GF3R(TIR), GF3D(FPD), GFLR(NTV), GFLD(NTD), GBIR(TTV), TAPER, TILTANGLE, SF3R(LTIR(3)), SF3D(LFPD(3)), SFLR(LTIR(BF)), SFLD(LFPD(BF)), LTIR(S3), LFPD(S3), SFQR(LTIR(SBF)), SFQD(LFPD(SBF)), SBIR(LTV), and SBID(LDOF) are parameters measured by adsorbing the entire back surface of the semiconductor substrate onto a vacuum chuck. In this invention, these parameters are referred to as full-surface adsorption analysis parameters.

[0031] Of the above parameters, WARP (GF3YFER, GF3NFER), BOW (GF3YFCD, GF3NFCD), and SORI (GFLYFER, GFLNFER) are parameters measured by adsorbing three or one point on the back surface of the semiconductor substrate to a vacuum chuck. In this invention, these parameters are referred to as non-adsorption analysis parameters.

[0032] <Scattered Light Intensity Measurement Device> In measuring the scattered light intensity according to the embodiment of the present invention, a measurement device equipped with a light-emitting system 10 and a light-receiving system 20 as shown in Figure 1 is preferably used.

[0033] The light projection system 10 includes a laser output unit 11 and a waveplate 12. The light projection system 10 is mounted so that the laser light L1 is incident on the semiconductor substrate 100 at an incident angle θ that is inclined with respect to the normal N of the surface. The light projection system 10 may be mounted so that the incident angle θ can be adjusted.

[0034] The laser output unit 11 is the source of the laser light L1, and can employ, for example, a gas laser such as a He-Ne laser, a semiconductor laser, or a solid-state laser such as a YAG laser.

[0035] The waveplate 12 performs adjustments to the laser light L1 generated by the laser output unit 11, including wavelength separation, fine adjustment, ellipticity adjustment, and polarization rotation. This waveplate 12 can switch between S-polarization and P-polarization of the laser light L1, and can control the penetration characteristics of the laser light L1 without changing measurement conditions such as the incident angle θ and wavelength λ. The appropriate material and configuration of the waveplate 12 are selected according to the material and conditions of the semiconductor substrate 100 being measured.

[0036] Furthermore, the light-receiving system 20 includes an objective lens 21, an imaging lens 22, a beam splitter 23, a light-receiving sensor 24, and a slit 25. It is desirable that the light-receiving system 20 be mounted at a measurement angle φ different from the incident angle θ so that reflected light L2 does not enter. The light-receiving system 20 may also be mounted so that the measurement angle φ can be adjusted.

[0037] It is desirable to employ a cube-type beam splitter as the beam splitter 23. As a configuration capable of converting the intensity of the scattered light L4 into an electrical signal, a photomultiplier tube, a photodiode, or the like can be adopted as the light receiving sensor 24. The slit 25 is arranged to determine the light receiving range of the scattered light L4.

[0038] <Measurement of Scattered Light Intensity> FIG. 1 is an explanatory diagram for explaining the measurement of scattered light intensity according to an embodiment. The measurement of the scattered light intensity according to the present invention measures the intensity of the scattered light L4 scattered by the processed and modified layer 102 (front surface side processed and modified layer 102a or back surface side processed and modified layer 102b) existing deeper than the surface 101 (front main surface 101a or back surface 101b) of the semiconductor substrate 100. Specifically, the laser light L1 irradiated from the light projecting system 10 is made to enter the semiconductor substrate 100 from the surface 101, and the scattered light L4 scattered by the incident light L3 that has entered the semiconductor substrate 100 in the processed and modified layer 102 is measured by the light receiving system 20. FIG. 1 is a diagram for obtaining the main surface scattered light intensity measurement value by measuring the intensity of the scattered light L4 scattered by the laser light L1 incident from the front main surface 101a in the front surface side processed and modified layer 102a. However, by inverting the semiconductor substrate 100, it is also possible to obtain the back surface scattered light intensity measurement value by measuring the intensity of the scattered light L4 scattered by the laser light L1 incident from the back surface 101b in the back surface side processed and modified layer 102b.

[0039] The semiconductor substrate 100 is a semiconductor material manufactured in a process including at least one of a slicing process, a polishing process, and a grinding process. Examples of the semiconductor substrate 100 manufactured through these processes include compound semiconductor material substrates such as a silicon carbide (SiC) substrate, a gallium nitride (GaN) substrate, an aluminum nitride (AlN) substrate, a gallium oxide (Ga

[0040] O ３ ) substrate, a sapphire substrate, or a silicon (Si) substrate. In the slicing process, polishing process, and grinding process described above, the processed and modified layer 102, which is one of the factors of internal stress, can be introduced.

[0040] The semiconductor substrate 100 has a surface 101 flattened to a certain extent by mechanical polishing or the like, a processed and modified layer 102 introduced by a process including machining such as a slicing process, a grinding process, and a polishing process, and a bulk layer 103. Note that below the surface 101 of the semiconductor substrate 100, the layer boundaries are not clearly distinguishable. The processed and modified layer 102 with a smaller degree of strain towards the interior is continuously distributed in an ambiguous state at the boundary with the bulk layer 103. In other words, the processed and modified layer 102 is a concept referring to a layer with a relatively large amount of strain, and the bulk layer 103 is a concept referring to a layer with a relatively small or negligible amount of strain, and it can be said that there is no clear boundary separating the two.

[0041] FIG. 2 is an explanatory diagram for explaining the relationship between the polarization of the laser beam L1 and the processed and modified layer 102 in the measurement of the scattered light intensity. FIG. 2(a) is an explanatory diagram for explaining the case where the semiconductor substrate 100 is irradiated with the laser beam L1 of S polarization. FIG. 2(b) is an explanatory diagram for explaining the case where the semiconductor substrate 100 is irradiated with the laser beam L1 of P polarization.

[0042] The measurement of the scattered light intensity may include a step of making the laser beam L1 of S polarization incident from the surface 101 of the semiconductor substrate 100 having the processed and modified layer 102 and measuring the intensity of the scattered light L4 scattered under the surface 101 (see FIG. 2(a)). Note that for S polarization, the electric field of the laser beam L1 vibrates in a direction perpendicular to the incident plane and has a lower transmittance compared to the P polarization irradiated under the same conditions. Therefore, considering the total amount and energy of the incident light L3, etc., scattered light generated in a relatively shallow part is more easily detected for S polarization compared to P polarization, and S polarization is suitable for measuring the shallow part of the processed and modified layer 102.

[0043] Specifically, S polarization is suitable for measurement in the depth range of the substrate surface to 10 μm, to 8 μm, to 6 μm, or to 4 μm. That is, the measurement of the scattered light intensity using S polarization makes the laser beam L1 of S polarization incident from the surface 101 of the semiconductor substrate 100 having the processed and modified layer 102, and it is possible to obtain a measured value of the intensity of the scattered light L4 generated mainly in the depth range of about 10 μm below the surface 101.

[0044] Furthermore, the measurement of scattered light intensity may include a step of irradiating the semiconductor substrate 100 with P-polarized laser light L1 from the surface 101 and measuring the intensity of scattered light L4 scattered below the surface 101 (see Figure 2(b)). Note that in P-polarized light, the electric field of the laser light L1 vibrates in a direction parallel to the incident surface, and therefore has a higher transmittance compared to S-polarized light irradiated under the same conditions. For this reason, considering the total amount and energy of the incident light L3, scattered light generated in relatively deeper parts is easier to detect with P-polarized light compared to S-polarized light, and P-polarized light is suitable for measuring the deeper parts of the processed altered layer 102.

[0045] Specifically, P-polarized light is suitable for measurements in the range of ~47 μm, ~30 μm, ~20 μm, or ~10 μm from the substrate surface. That is, in the measurement step S22 using P-polarized light, P-polarized laser light L1 is incident from the surface 101 of the semiconductor substrate 100 having the processed altered layer 102, and the intensity measurement value of scattered light L4 generated in a depth range of approximately ~47 μm below the surface 101 can be obtained.

[0046] In the embodiment of the present invention, S-polarized light was incident on each of the seven SiC substrates A to G, and the scattered light intensity was measured at multiple measurement points on the substrate surface as described above, and the main surface scattered light intensity measurement values ​​and back surface scattered light intensity measurement values ​​(intensity map, intensity scale, mode) shown in Figure 3 were obtained.

[0047] Similarly, P-polarized light was incident on each of the same seven SiC substrates A to G, and the scattered light intensity was measured at multiple measurement points on the substrate surface as described above, obtaining the main surface scattered light intensity measurements and back surface scattered light intensity measurements (intensity map, intensity scale, mode) shown in Figure 4.

[0048] Note that the mode obtained here may be replaced with other positional parameters (for example, one or more selected from the mean (arithmetic, geometric, harmonic), median (quantile, ordinal statistic), mode, class value, etc.).

[0049] <Measurement of Shape Feature Parameters> In the embodiment of the present invention, shape feature parameters (GBIR, SORI, BOW) were measured for each of the seven SiC substrates A to G, and the results shown in Table 1 were obtained. GBIR is known as the thickness unevenness based on the back surface of the wafer, and SORI and BOW are known as parameters that indicate the degree of the wafer's curvature (warp).

[0050] In this embodiment of the present invention, the shape characteristic parameters were measured using a flatness tester (FT-900) manufactured by NIDEK Corporation, but other well-known measuring devices and methods may also be used. Furthermore, the shape characteristic parameters may be replaced with other parameters.

[0051]

[0052] <Creation of Datasets and Estimation Models> By appropriately selecting, processing, and associating the data shown in Figures 3 and 4 and Table 1, a dataset like the one shown in Tables 2 and 3 can be obtained.

[0053] For example, Table 2 is a dataset that correlates the main surface scattered light intensity measurement (mode) and back surface scattered light intensity measurement (mode) obtained by incidenting S-polarized light on each of the SiC substrates A to G with the intensity ratio of these measurements, GBIR, SORI, and BOW.

[0054]

[0055] Table 3 is a dataset showing the main surface scattered light intensity measurement (mode) and back surface scattered light intensity measurement (mode) obtained by incidenting P-polarized light onto each of the SiC substrates A to G, along with the intensity ratio of these measurements, GBIR, SORI, and BOW, in relation to each of these values.

[0056]

[0057] Based on the datasets shown in Tables 2 and 3, data plots like those shown in Figures 5 to 7 can be obtained, with the horizontal axis representing the ratio of the principal surface scattered light intensity measurement to the back surface scattered light intensity measurement. The dashed lines in the figures represent linear approximation lines.

[0058] Figure 5 is a data plot with GBIR on the vertical axis and the ratio of the principal surface scattered light intensity measurement to the back surface scattered light intensity measurement on the horizontal axis. Figure 5(a) corresponds to the principal surface scattered light intensity measurement and back surface scattered light intensity measurement obtained when S-polarized light is incident, and Figure 5(b) corresponds to the principal surface scattered light intensity measurement and back surface scattered light intensity measurement obtained when P-polarized light is incident.

[0059] As shown in Figure 5, GBIR shows a certain level of correlation with the ratio of the principal surface scattered light intensity measurement to the back surface scattered light intensity measurement, regardless of the type of polarization. In particular, GBIR showed a stronger correlation with the ratio of the principal surface scattered light intensity measurement to the back surface scattered light intensity measurement obtained when S-polarized light was incident (R ２ = 0.8543).

[0060] Figure 6 is a data plot with SORI on the vertical axis and the ratio of the principal surface scattered light intensity measurement to the back surface scattered light intensity measurement on the horizontal axis. Figure 6(a) corresponds to the principal surface scattered light intensity measurement and back surface scattered light intensity measurement obtained when S-polarized light is incident, and Figure 6(b) corresponds to the principal surface scattered light intensity measurement and back surface scattered light intensity measurement obtained when P-polarized light is incident.

[0061] As shown in Figure 6, SORI shows a certain level of correlation with the ratio of the principal surface scattered light intensity to the back surface scattered light intensity, regardless of the type of polarization. In particular, SORI showed a stronger correlation with the ratio of the principal surface scattered light intensity to the back surface scattered light intensity obtained when P-polarized light was incident (R ２ = 0.8063).

[0062] Figure 7 is a data plot with the vertical axis representing the BOW (Blind-of-Wave) and the horizontal axis representing the ratio of the principal surface scattered light intensity to the back surface scattered light intensity. Figure 7(a) corresponds to the principal surface scattered light intensity measurements and back surface scattered light intensity measurements obtained when S-polarized light is incident, and Figure 7(b) corresponds to the principal surface scattered light intensity measurements and back surface scattered light intensity measurements obtained when P-polarized light is incident.

[0063] As shown in Figure 7, BOW shows a certain level of correlation with the ratio of the principal surface scattered light intensity measurement to the back surface scattered light intensity measurement, regardless of the type of polarization. In particular, BOW showed a stronger correlation with the ratio of the principal surface scattered light intensity measurement to the back surface scattered light intensity measurement obtained when P-polarized light was incident (R ２ = 0.718).

[0064] Furthermore, for comparison, based on the datasets shown in Tables 2 and 3, data plots such as those shown in Figures 8 to 13 can be obtained, with the horizontal axis representing the principal surface scattered light intensity measurement or the back surface scattered light intensity measurement itself. Note that in Figures 8 to 13, the horizontal axis is logarithmic because of the large variability in the principal surface scattered light intensity measurement or the back surface scattered light intensity measurement. The dashed lines in the figures represent linear approximation curves.

[0065] Figure 8 is a data plot with GBIR on the vertical axis and principal surface scattered light intensity measurements on the horizontal axis. Figure 8(a) corresponds to the principal surface scattered light intensity measurements obtained when S-polarized light is incident, and Figure 8(b) corresponds to the principal surface scattered light intensity measurements obtained when P-polarized light is incident.

[0066] As shown in Figure 8, GBIR did not show any correlation with the principal surface scattered light intensity measurement, regardless of the type of polarization. (R ２ <0.1).

[0067] Figure 9 is a data plot with SORI on the vertical axis and the principal surface scattered light intensity measurement on the horizontal axis. Figure 9(a) corresponds to the principal surface scattered light intensity measurement obtained when S-polarized light is incident, and Figure 9(b) corresponds to the principal surface scattered light intensity measurement obtained when P-polarized light is incident.

[0068] As shown in Figure 9, SORI did not show any correlation with the principal surface scattered light intensity measurement, regardless of the type of polarization. (R ２ <0.2).

[0069] Figure 10 is a data plot with the vertical axis representing the BOW and the horizontal axis representing the principal surface scattered light intensity measurement. Figure 10(a) corresponds to the principal surface scattered light intensity measurement obtained when S-polarized light is incident, and Figure 10(b) corresponds to the principal surface scattered light intensity measurement obtained when P-polarized light is incident.

[0070] As shown in Figure 10, BOW did not show any correlation with the principal surface scattered light intensity measurement, regardless of the type of polarization. (R ２ <0.1).

[0071] FIG. 11 is a data plot with the vertical axis being GBIR and the horizontal axis being the measured value of the back surface scattered light intensity. FIG. 11(a) corresponds to the measured value of the back surface scattered light intensity obtained by incident S-polarized light, and FIG. 11(b) corresponds to the measured value of the back surface scattered light intensity obtained by incident P-polarized light, respectively.

[0072] As shown in FIG. 11, GBIR did not show a correlation with the measured value of the back surface scattered light intensity regardless of the type of polarization. (R ２ <0.1).

[0073] FIG. 12 is a data plot with the vertical axis being SORI and the horizontal axis being the measured value of the back surface scattered light intensity. FIG. 12(a) corresponds to the measured value of the back surface scattered light intensity obtained by incident S-polarized light, and FIG. 12(b) corresponds to the measured value of the back surface scattered light intensity obtained by incident P-polarized light, respectively.

[0074] As shown in FIG. 12, SORI did not show a correlation with the measured value of the back surface scattered light intensity regardless of the type of polarization. (R ２ <0.1).

[0075] FIG. 13 is a data plot with the vertical axis being BOW and the horizontal axis being the measured value of the back surface scattered light intensity. FIG. 13(a) corresponds to the measured value of the back surface scattered light intensity obtained by incident S-polarized light, and FIG. 13(b) corresponds to the measured value of the back surface scattered light intensity obtained by incident P-polarized light, respectively.

[0076] As shown in FIG. 13, BOW did not show a correlation with the measured value of the main surface scattered light intensity regardless of the type of polarization. (R ２ <0.1).

[0077] As described above, the shape characteristics of the semiconductor substrate did not show a correlation with the measured value of the main surface scattered light intensity or the measured value of the back surface scattered light intensity itself, and a certain degree of correlation was observed regardless of the type of polarization with the ratio of the measured value of the main surface scattered light intensity to the measured value of the back surface scattered light intensity.

[0078] <Shape Evaluation of Semiconductor Substrates> (1) Quantification and Evaluation of Substrate Shape Characteristics Using Estimation Models As described above, there is a certain level of correlation between the shape characteristics of semiconductor substrates and the ratio of the main surface scattered light intensity measurement value to the back surface scattered light intensity measurement value, regardless of the type of polarization. Therefore, by defining the above SiC substrates A to G as reference substrates, the approximate straight line based on the data plots shown in Figures 5 to 7 can be used as a kind of estimation model to estimate / quantify and evaluate the shape characteristics of the target substrate.

[0079] As shown in Figures 5 to 7, GBIR shows a stronger correlation with the ratio of the main surface scattered light intensity value to the back surface scattered light intensity value obtained by incident S-polarized light, while SORI and BOW show a stronger correlation with the ratio of the main surface scattered light intensity value to the back surface scattered light intensity value obtained by incident P-polarized light. Furthermore, GBIR is known to be a full-surface adsorption analysis parameter, while SORI and BOW are known to be non-adsorption analysis parameters. Therefore, it is preferable that the shape features indicated by the full-surface adsorption analysis parameters, including GBIR, be estimated / quantified based on the main surface scattered light intensity value and back surface scattered light intensity value obtained by incident S-polarized light, and that the shape features indicated by the non-adsorption analysis parameters, including SORI and BOW, be estimated / quantified based on the main surface scattered light intensity value and back surface scattered light intensity value obtained by incident P-polarized light.

[0080] For example, consider the case of estimating the GBIR of a target substrate using the data plot and approximation line shown in Figure 5(a). First, the scattered light intensity is measured on the target substrate using S-polarized (or P-polarized) light, and the measured values ​​of the main surface scattered light intensity and the back surface scattered light intensity are obtained, and their ratio is calculated. If the ratio of the measured values ​​of the main surface scattered light intensity and the back surface scattered light intensity on the target substrate is calculated to be approximately 2, then the GBIR of the target substrate is estimated to be approximately 6 μm based on the data plot and approximation line shown in Figure 5(a).

[0081] For example, consider the case where the SORI of the target substrate is estimated using the data plot and approximation line shown in Figure 6(b). First, the scattered light intensity is measured on the target substrate using P-polarized (or S-polarized) light, and the measured values ​​of the main surface scattered light intensity and the back surface scattered light intensity are obtained, and their ratio is calculated. If the ratio of the measured values ​​of the main surface scattered light intensity and the back surface scattered light intensity on the target substrate is calculated to be approximately 0.5, then the SORI of the target substrate is estimated to be approximately 100 μm based on the data plot and approximation line shown in Figure 6(b).

[0082] For example, consider the case where the BOW of the target substrate is estimated using the data plot and approximation line shown in Figure 7(b). First, the scattered light intensity is measured on the target substrate using P-polarized (or S-polarized) light as described above, and the measured values ​​of the main surface scattered light intensity and the back surface scattered light intensity are obtained, and their ratio is calculated. If the ratio of the measured values ​​of the main surface scattered light intensity and the back surface scattered light intensity on the target substrate is calculated to be approximately 0.75, then the BOW is estimated to be approximately 60 μm based on the data plot and approximation line shown in Figure 7(b).

[0083] (2) The shape characteristics of the semiconductor substrate may be quantified and evaluated by numerical calculation, or by using the calculated values ​​based on the main surface scattered light intensity measurement and the back surface scattered light intensity measurement as shape characteristic parameters. In this case as well, it is preferable that the shape characteristics shown by the full-surface adsorption analysis parameters including GBIR are estimated / quantified based on the main surface scattered light intensity values ​​and back surface scattered light intensity values ​​obtained by incident S-polarized light, and it is preferable that the shape characteristics shown by the non-adsorption analysis parameters including SORI and BOW are estimated / quantified based on the main surface scattered light intensity values ​​and back surface scattered light intensity values ​​obtained by incident P-polarized light.

[0084] (3) The shape characteristics of the semiconductor substrate can be quantified and evaluated visually based on image calculations, or based on image calculations using the intensity maps (distribution images of main surface scattered light intensity measurements and back surface scattered light intensity measurements within the substrate surface) described in Table 1 or Table 2.

[0085] For example, Figure 14 shows the ratio of the distribution image of the main surface scattered light intensity measurement and the distribution image of the back surface scattered light intensity measurement when S-polarized light is incident on the SiC substrate A described above. That is, Figure 14 is a distribution image mapping the ratio of the main surface scattered light intensity measurement and the back surface scattered light intensity measurement when S-polarized light is incident at the same point on both sides of the SiC substrate A.

[0086] For example, Figure 15 shows the ratio of the distribution image of the main surface scattered light intensity measurement values ​​to the distribution image of the back surface scattered light intensity measurement values ​​when P-polarized light is incident on the SiC substrate A described above. That is, Figure 15 is a distribution image mapping the ratio of the main surface scattered light intensity measurement values ​​to the back surface scattered light intensity measurement values ​​when P-polarized light is incident at the same point on both sides of the SiC substrate A.

[0087] Figures 16(a) to 16(c) are 3D mapped images of the GBIR, SORI, and BOW of the SiC substrate A described above, measured with a flatness tester (FT-900) manufactured by NIDEK Corporation.

[0088] Comparing Figures 14 to 16, it can be seen that the image calculation results shown in Figures 14 and 15 reflect and quantify to some extent the various shape features shown in Figure 16.

[0089] 10 Light projection system 11 Laser output unit 12 Waveplate 20 Light receiving system 21 Objective lens 22 Imaging lens 23 Beam splitter 24 Light receiving sensor 25 Slit 100 Semiconductor substrate 101 Front surface 101a Main surface 101b Back surface 102 Processed and altered layer 102a Processed and altered layer on the main surface side 102b Processed and altered layer on the back surface side 103 Bulk layer L1 Laser light L2 Reflected light L3 Incident light L4 Scattered light N Normal θ Incident angle φ Measurement angle

Claims

1. A method for evaluating the shape characteristics of a semiconductor substrate, comprising a quantification step of quantifying the shape characteristics of the semiconductor substrate based on a main surface scattered light intensity measurement obtained by measuring the intensity of scattered light from laser light incident on the main surface of the semiconductor substrate, and a back surface scattered light intensity measurement obtained by measuring the intensity of scattered light from laser light incident on the back surface of the semiconductor substrate.

2. The semiconductor substrate shape evaluation method according to claim 1, wherein the quantification step quantifies the shape characteristics of the semiconductor substrate based on the ratio of the main surface scattered light intensity measurement value and the back surface scattered light intensity measurement value.

3. The method for evaluating the shape of a semiconductor substrate according to claim 2, wherein the quantification step quantifies the shape characteristics of the semiconductor substrate using the ratio of the main surface scattered light intensity measurement value to the back surface scattered light intensity measurement value as a shape characteristic parameter indicating the shape characteristics of the semiconductor substrate.

4. The method for evaluating the shape of a semiconductor substrate according to any one of claims 1 to 3, wherein the shape characteristics of the semiconductor substrate include the amount of warping of the semiconductor substrate.

5. The method for evaluating the shape of a semiconductor substrate according to claim 1, wherein in the quantification step, the main surface scattered light intensity measurement is obtained by measuring the intensity of elastic scattered light obtained by elastically scattering laser light incident on the main surface of the semiconductor substrate, and the back surface scattered light intensity measurement is obtained by measuring the intensity of elastic scattered light obtained by elastically scattering laser light incident on the back surface of the semiconductor substrate.

6. The method for evaluating the shape of a semiconductor substrate according to claim 1, wherein the quantification step quantifies the shape characteristics of the semiconductor substrate based on the position parameter of the distribution of a plurality of main surface scattered light intensity measurements obtained by measuring the intensity of scattered light scattered by laser light incident on the main surface of the semiconductor substrate at a plurality of measurement points on the main surface of the semiconductor substrate, and the position parameter of the distribution of a plurality of back surface scattered light intensity measurements obtained by measuring the intensity of scattered light scattered by laser light incident on the back surface of the semiconductor substrate at a plurality of measurement points on the back surface of the semiconductor substrate.

7. The semiconductor substrate shape evaluation method according to claim 1, wherein the quantification step quantifies the shape characteristics of the semiconductor substrate based on image calculations of a distribution image of a plurality of main surface scattered light intensity measurements obtained by measuring the intensity of scattered light scattered by laser light incident on the main surface of the semiconductor substrate at a plurality of measurement points on the main surface of the semiconductor substrate, and a distribution image of a plurality of back surface scattered light intensity measurements obtained by measuring the intensity of scattered light scattered by laser light incident on the back surface of the semiconductor substrate at a plurality of measurement points on the back surface of the semiconductor substrate.

8. The semiconductor substrate shape evaluation method according to claim 1, wherein the quantification step includes an estimation step of estimating shape feature parameters indicating the shape characteristics of a target substrate from the main surface scattered light intensity measurement and the back surface scattered light intensity measurement of a reference substrate, using an estimation model created based on a dataset relating the main surface scattered light intensity measurement and the back surface scattered light intensity measurement of a reference substrate with shape feature parameters indicating the shape characteristics of the reference substrate.

9. The semiconductor substrate shape evaluation method according to claim 8, wherein the estimation model is an estimation model created based on a dataset relating the ratio of the main surface scattered light intensity measurement value and the back surface scattered light intensity measurement value on the reference substrate to the shape feature parameters of the reference substrate, and the estimation step is to use the estimation model to estimate the shape feature parameters of the target substrate based on the ratio of the main surface scattered light intensity measurement value and the back surface scattered light intensity measurement value on the target substrate.

10. The method for evaluating the shape of a semiconductor substrate according to claim 8, wherein the shape characteristic parameter includes a warpage parameter indicating the amount of warpage of the semiconductor substrate.

11. The method for evaluating the shape of a semiconductor substrate according to claim 8, wherein the laser light is S-polarized laser light, and the shape characteristic parameters include surface adsorption analysis parameters measured by adsorbing the entire back surface of the semiconductor substrate onto a vacuum chuck.

12. The method for evaluating the shape of a semiconductor substrate according to claim 8, wherein the laser light is P-polarized laser light, and the shape characteristic parameters include non-adsorption analysis parameters measured by adsorbing three or one point on the back surface of the semiconductor substrate.

13. A method for creating an estimation model used to estimate the shape characteristics of a semiconductor substrate, based on a dataset that associates the ratio of a main surface scattered light intensity measurement obtained by measuring the intensity of scattered light from laser light incident on the main surface of a reference substrate with a back surface scattered light intensity measurement obtained by measuring the intensity of scattered light from laser light incident on the back surface of a reference substrate with shape characteristic parameters that indicate the shape characteristics of the reference substrate.

Images