Monocrystalline silicon polished wafer-based flow guide cylinder metal content prediction method

Abstract

A flow guide cylinder metal content prediction method based on a monocrystalline silicon polished wafer relates to the field of metal content prediction, and comprises the following steps: S1, drawing a monocrystalline silicon rod to be detected by using a flow guide cylinder to be detected, processing the monocrystalline silicon rod to be detected into a polished wafer to be detected, and taking out the polished wafer to be detected; detecting the diameter of the to-be-detected polished section so as to calculate a distance parameter between the to-be-detected polished section and the guide cylinder; s2, detecting the metal content of the monocrystalline silicon polished wafer to be detected; and S3, predicting the metal content of the to-be-detected guide cylinder through the guide cylinder prediction model. The high sensitivity of the monocrystalline silicon polished wafer to the metal content is utilized, the metal content of the guide cylinder is accurately calculated through the metal content of the monocrystalline silicon polished wafer, nondestructive detection of the guide cylinder is achieved, and it is ensured that the detection result is accurate and reliable.

Description

Technical Field

[0001] The invention relates to the field of metal content prediction, in particular to a method for predicting the metal content of a guide tube based on a single crystal silicon polishing sheet. Background Art

[0002] As a key raw material in the semiconductor industry, the development trend of single crystal silicon polishing wafers is characterized by high-end, intelligent and green. With the rapid development of emerging technologies such as 5G and artificial intelligence, the demand for high-performance and high-purity single crystal silicon polishing wafers is growing. The control of impurities in silicon is particularly important, especially the control of metal impurity content. For example, in high-end application fields such as high-performance integrated circuits and solar cells, strict control of metal impurities is one of the key factors to ensure device performance.

[0003] As the most important metal impurity, the control of iron content is the key to metal control. The iron content in monocrystalline silicon polished wafers has a significant impact on the device. Iron impurities will reduce the minority carrier lifetime, affect the efficiency of MOS devices, and may cause device reliability to decrease. In terms of batteries, iron impurities will reduce the open circuit voltage, short circuit current and quantum efficiency of monocrystalline silicon cells, thereby affecting battery performance. Therefore, in the production process of monocrystalline silicon polished wafers, it is very important to strictly control the iron content.

[0004] In the production process of silicon polishing sheets, the introduction of metallic iron mainly comes from the graphite parts of the single crystal furnace. The guide tube, as the main component of the graphite thermal field of the single crystal furnace, is particularly important for the crystal pulling process. The guide tube is closest to the crystal rod. Under high temperature, through thermal radiation diffusion, the metal impurity iron inside it will gradually and evenly penetrate into the crystal rod, affecting the quality of the crystal rod. In addition, the feed amount in the current single crystal silicon pulling process continues to increase, and the cost of single crystal pulling continues to rise. Once the metal content of the guide tube exceeds the standard, the crystal rod will be contaminated with metal, which will cause huge cost losses.

[0005] At present, the metal content in the guide tube is usually detected by GDMS (glow discharge mass spectrometry), but this method is a destructive test. Therefore, graphite parts manufacturers often use the furnace samples that undergo the purification process together with the guide tube to implement GDMS testing. However, since the purification effect is greatly affected by factors such as the placement position in the purification furnace and the airflow guidance, the metal detection results of the furnace samples are difficult to accurately reflect the metal content of the guide tube itself, and accurate control and testing cannot be achieved.

[0006] In addition, for the manufacturers of monocrystalline silicon polishing sheets, they also need to test the metal content of the guide tube after purchasing the guide tube graphite parts to evaluate whether the purchased guide tube can meet the production needs and material specifications. Detecting the accurate metal content of the guide tube and feeding it back to the guide tube manufacturer can serve as a warning to the guide tube manufacturer whose metal content exceeds the standard. Summary of the invention

[0007] In order to solve the problem that the metal content detection result of the guide tube in the prior art is not accurate enough, the present invention provides a method for predicting the metal content of the guide tube based on a single crystal silicon polishing wafer.

[0008] In order to achieve the above object, the specific scheme adopted by the present invention is: a method for predicting the metal content of a guide tube based on a single crystal silicon polishing sheet, comprising the following steps:

[0009] S1, using the guide tube to be tested to draw the single crystal silicon rod to be tested, and calculating the distance between the single crystal silicon polishing sheet to be tested and the guide tube to be tested during the drawing process;

[0010] S2, processing the single crystal silicon rod to be tested into a single crystal silicon polished wafer to be tested, and detecting the metal content of the single crystal silicon polished wafer to be tested;

[0011] S3, predicting the metal content of the guide tube to be tested by using the guide tube prediction model;

[0012] Among them, the metal content to be predicted is the iron content, and the guide tube prediction model is G=1.23+0.264*L-0.0983*H+0.0866*L*H, G is the metal content of the guide tube to be tested, L is the distance parameter between the single crystal silicon polishing sheet to be tested and the guide tube to be tested, and H is the metal content of the single crystal silicon polishing sheet to be tested.

[0013] As a further optimization of the above technical solution, the distance parameter between the single crystal silicon polishing sheet to be tested and the guide tube to be tested is L, L=D1-D2, D1 is the diameter of the small diameter end of the guide tube to be tested, and D2 is the diameter of the single crystal silicon polishing sheet to be tested.

[0014] As a further optimization of the above technical solution, the diameter difference between the single crystal silicon rod to be tested and the single crystal silicon polishing wafer to be tested is no more than 5 mm.

[0015] As a further optimization of the above technical solution, the method for detecting the metal content of the single crystal silicon polished sheet to be tested is the SPV surface photovoltage method.

[0016] As a further optimization of the above technical solution, the method for detecting the metal content of the single crystal silicon polished wafer to be tested is:

[0017] S201, selecting a standard guide tube with a metal content to be predicted ≤ 0.5 ppmw to draw a standard single crystal silicon rod, wherein the diameter of the standard single crystal silicon rod and the single crystal silicon rod to be tested are the same;

[0018] S202, processing the standard single crystal silicon rod into a standard single crystal silicon polished wafer;

[0019] S203, using the SPV method to detect the standard single crystal silicon polished wafer to obtain a 49-point reference spectrum of metals in the standard single crystal silicon polished wafer, and using the SPV method to detect the single crystal silicon polished wafer to be tested to obtain a 49-point test spectrum of metals in the single crystal silicon polished wafer to be tested;

[0020] S204, based on the 49-point test spectrum and the 49-point reference spectrum, the metal content H of the single crystal silicon polished wafer to be tested is obtained, where H={(H s1 -H J1 )+(H s2 -H J2 )+(H s3 -H J3 )+.......(H s49 -H J49 )} / 49,H sx is the metal content of the xth point in the 49-point test spectrum, H Jx It is the metal content of the x-th point in the 49-point reference spectrum, where x is any natural number from 1 to 49.

[0021] As a further optimization of the above technical solution, the method for establishing the guide tube prediction model is as follows: first, n guide tubes with different metal contents and the same size specifications are selected, where n is a natural number and n≥3; secondly, the n guide tubes are used to draw lightly boron-doped single crystal silicon rods of different diameters; thirdly, the lightly boron-doped single crystal silicon rods are processed into polished sheets, and the metal content of the polished sheets is tested; finally, sample blocks are taken from the small diameter end of the guide tube after drawing the lightly boron-doped single crystal silicon rods, and the metal content of the guide tube is detected by GDMS, and the guide tube prediction model is obtained by regression analysis using mathematical statistical tools.

[0022] As a further optimization of the above technical solution, the raw materials of the lightly boron-doped single crystal silicon rod are polycrystalline silicon and boron master alloy.

[0023] As a further optimization of the above technical solution, the method for establishing the guide tube prediction model is specifically as follows: first, three guide tubes with different metal contents and the same size specifications are selected, and three lightly doped single crystal silicon rods with different diameters are drawn using each guide tube. The lightly doped single crystal silicon rods are then processed into polished wafers, and the metal content of the polished wafers is tested using the SPV surface photovoltage method.

[0024] Compared with the prior art, the present invention has the following beneficial effects:

[0025] The present invention utilizes the high sensitivity of the single crystal silicon polishing sheet to the metal content, accurately infers the metal content of the guide tube through the metal content of the single crystal silicon polishing sheet, realizes non-destructive testing of the guide tube, and ensures that the test result is accurate and reliable.

[0026] In the detection method described in the present invention, a standard single crystal silicon rod is drawn by selecting a standard guide tube with a predicted metal content ≤0.5ppmw, and a 49-point reference map of the metal to be tested is obtained. The metal content H of the single crystal silicon polishing sheet to be tested is obtained by combining the 49-point test map and the 49-point reference map. This can eliminate the error in the detection result caused by the metal contained in the raw materials for drawing the single crystal silicon rod, thereby improving the detection accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] Figure 1 It is a schematic diagram of the three-dimensional structure of the guide tube;

[0028] Figure 2 This is a schematic diagram of the positional relationship between the guide tube and the single crystal silicon rod during the single crystal silicon rod pulling process;

[0029] Figure 3 It is the 49-point reference spectrum detected by the SPV method;

[0030] Description of the drawings: 1. Guide tube, 2. Single crystal silicon rod. DETAILED DESCRIPTION

[0031] The technical solution of the present invention is further elaborated in detail below in conjunction with specific embodiments. The parts of the present invention that are not described and disclosed in detail in the following embodiments should be understood as the prior art known or should be known to those skilled in the art.

[0032] The present invention discloses a method for predicting the metal content of a guide tube based on a single crystal silicon polishing sheet, comprising the following steps:

[0033] S1, using the guide tube to be tested to draw the single crystal silicon rod to be tested, processing the single crystal silicon rod to be tested into the single crystal silicon polished sheet to be tested, and calculating the distance parameter between the single crystal silicon polished sheet to be tested and the guide tube to be tested during the drawing process. Specifically, the single crystal silicon rod is processed into the single crystal silicon polished sheet to be tested through normal processes of the prior art such as truncation, rolling, sticking, wire cutting, corrosion, grinding, polishing, etc.

[0034] The distance parameter between the single crystal silicon polishing sheet to be tested and the guide tube to be tested is L, L=D1-D2, D1 is the diameter of the small diameter end of the guide tube to be tested, and D2 is the diameter of the single crystal silicon polishing sheet to be tested.

[0035] When pulling the single crystal silicon rod to be tested, the rotary crystal pulling method of the existing technology is adopted, and the entire thermal field has a positioning setting. Before pulling the single crystal silicon rod to be tested, the seed crystal clamp above the guide tube is first aligned so that the single crystal silicon rod and the guide tube are coaxially distributed during the pulling process. Therefore, the pulled single crystal silicon rod to be tested is cylindrical, and the distance parameter L between the single crystal silicon polishing sheet to be tested and the guide tube to be tested is a fixed value in the same test.

[0036] S2, detecting the metal content of the single crystal silicon polished wafer to be tested.

[0037] The detection method for the metal content of the single crystal silicon polished sheet to be tested is the SPV method. The metal to be detected is iron, and the SPV method for detecting metallic iron is an existing technology. Specifically, referring to the current industry standard "YS / T 679-2018 Surface Photovoltage Method for Testing Diffusion Length of Minority Carriers in Extrinsic Semiconductors", the diffusion length of the silicon polished sheet is tested using an SPV (surface photovoltage) instrument. The attenuation change of the diffusion length measured before and after the iron-boron bond in silicon is decomposed by light pulses can accurately obtain the iron content in the silicon polished sheet. The process of converting the diffusion length to the corresponding metal will not be repeated here.

[0038] In order to avoid the influence of the metal elements to be measured in the basic components of the single crystal silicon rod when the single crystal silicon rod to be tested is pulled, thus affecting the prediction result of the metal content of the guide tube, when using the SPV method to detect the metal content of the single crystal silicon polishing sheet to be tested, a standard single crystal silicon rod is first pulled to eliminate the influence of the metal elements to be measured in the basic components of the single crystal silicon rod, as follows:

[0039] S201, selecting a standard guide tube with a metal content to be predicted ≤ 0.5 ppmw to draw a standard single crystal silicon rod, wherein the diameter of the standard single crystal silicon rod and the single crystal silicon rod to be tested are the same, and the diameter of the polished sheet after processing is also the same;

[0040] S202, processing the standard single crystal silicon rod into a standard single crystal silicon polished wafer;

[0041] S203, using the SPV method to detect the standard single crystal silicon polished wafer to obtain a 49-point reference spectrum of the metal to be tested, and using the SPV method to detect the single crystal silicon polished wafer to be tested to obtain a 49-point test spectrum of the metal to be tested; it should be noted that the 49-point reference spectrum and the 49-point test spectrum are both spectra obtained by using the SPV method to detect 49 points on the single crystal silicon polished wafer;

[0042] S204, based on the 49-point test spectrum and the 49-point reference spectrum, the metal content H of the single crystal silicon polished wafer to be tested is obtained, where H={(H s1 -H J1 )+(H s2 -H J2 )+(H s3 -H J3 )+.......(H s49 -H J49 )} / 49,H sx is the metal content of the xth point in the 49-point test spectrum, H Jx It is the metal content of the x-th point in the 49-point reference spectrum, where x is any natural number from 1 to 49.

[0043] S3, predicting the metal content of the guide tube to be tested by using the guide tube prediction model.

[0044] The guide tube prediction model is G=1.23+0.264*L-0.0983*H+0.0866*L*H, G is the metal content of the guide tube to be tested, L is the distance parameter between the single crystal silicon polishing sheet to be tested and the guide tube to be tested, and H is the metal content of the single crystal silicon polishing sheet to be tested.

[0045] It should be noted that: by selecting a standard guide tube with a metal content to be tested ≤0.5ppmw to draw a standard single crystal silicon rod, and obtaining a 49-point benchmark map of the metal to be tested, combining the 49-point test map and the 49-point benchmark map to obtain the metal content increment of the single crystal silicon polished wafer to be tested, and inputting the metal content increment into the guide tube prediction model as the metal content of the single crystal silicon polished wafer to be tested, the error caused by the metal contained in the raw materials for drawing the single crystal silicon rod can be eliminated, thereby improving the detection accuracy.

[0046] The process of establishing the guide tube prediction model is as follows: first, select n guide tubes with different metal contents and the same size specifications, where n is a natural number and n≥3; second, use n guide tubes to pull lightly boron-doped single crystal silicon rods of different diameters; third, process the lightly boron-doped single crystal silicon rods into polished wafers, and use the SPV method to test the metal content of the polished wafers; finally, take samples from the small diameter end of the guide tube after pulling the lightly boron-doped single crystal silicon rod, detect the metal content of the guide tube by GDMS, and use mathematical statistical tools and regression analysis to obtain the guide tube prediction model.

[0047] Specifically, the process of establishing a guide tube prediction model for detecting the iron content in the guide tube includes the following steps:

[0048] First, three guide tubes 1 with different iron contents and the same size specifications were selected for cleaning and calcination, and the inner diameter of the small diameter end of the guide tube 1 was 250 mm;

[0049] Secondly, three guide tubes were used to draw lightly boron-doped single crystal silicon rods of different diameters; each guide tube was used to draw three lightly boron-doped single crystal silicon rods, and a total of nine lightly boron-doped single crystal silicon rods were drawn; the raw materials of the lightly boron-doped single crystal silicon rods were polycrystalline silicon and boron master alloy. Since the iron content of lightly boron-doped single crystal silicon rods can be accurately tested, the trace iron elements in the silicon rods can be accurately reflected. In contrast, if heavily boron-doped single crystal silicon rods are used, the test cannot be carried out because the impurity content is too high, resulting in inaccurate test results. In this case, the high concentration of iron and other impurities will interfere with the normal operation of the test instrument, making the test data lose its reference value.

[0050] Next, the lightly doped single crystal silicon rods were processed into polished wafers, and the iron content of the polished wafers was tested using the SPV method. When testing the iron content of the polished wafers, 49 points were taken on the polished wafers.

[0051] During the process of pulling single crystal silicon rods, due to the heat radiation effect of the guide tube at high temperature, the iron content of the guide tube will be evenly radiated to the single crystal silicon rod through the rotating single crystal silicon rod.

[0052] Depend on Figure 1 , 2 It can be seen that the guide tube 1 is conical, with a large diameter end and a small diameter end. When in use, the large diameter end is upward and the small diameter end is downward. The distance parameter L between the guide tube 1 and the single crystal silicon polishing sheet is: the difference between the inner diameter of the small diameter end of the guide tube 1 and the diameter of the polishing sheet processed by the single crystal silicon rod 2.

[0053] Finally, samples were taken from the small diameter end of the guide tube after pulling the lightly boron-doped single crystal silicon rod, and the iron content was tested by GDMS. Mathematical statistical tools were used to perform regression analysis to obtain the guide tube prediction model.

[0054] Table 1 shows the SPV49 point full scan results obtained by pulling lightly boron-doped single crystal silicon rods of different diameters using guide tubes with different iron contents.

[0055] Table 1: Test No. Actual metal content measured outside the guide tube G / ppmw Diameter of lightly boron-doped single crystal silicon polishing wafer D / mm Distance parameter between lightly boron-doped single crystal silicon polishing sheet and guide tube *L / cm <![CDATA[SPV49 point result H*E9 / (atoms / cm 3 )]]> 1 3.2 157 9.3 0.9 2 3.2 193 5.7 4 3 3.2 218 3.2 10 4 7.5 155 9.5 3.8 5 7.5 195 5.5 6.9 6 7.5 213 3.7 23 7 15.3 150 10 15 8 15.3 189 6.1 29.8 9 15.3 227 2.3 133.1

[0056] The data in Table 1 were analyzed by MINITAB to establish a regression equation to obtain the guide tube prediction model. Through mathematical statistics and regression equations, the R value (goodness of fit) was obtained to be 96%, and the P value was <0.001, indicating a significant correlation.

[0057] The guide tube prediction model is obtained: G=1.23+0.264*L-0.0983*H+0.0866*L*H

[0058] G is the actual metal content of the guide tube, L is the distance parameter between the single crystal silicon polishing sheet and the guide tube, and H is the SPV 49-point test result.

[0059] It should be noted that: 1) the method of taking 49 points on the polishing sheet and using the SPV method to test the metal content of the polishing sheet is referred to as SPV49 points, and the specific selection of the 49 points is common knowledge in the art;

[0060] 2) According to actual experiments, it takes thousands of hours for the iron content of the guide tube to diffuse and then be slightly reduced, so the iron content loss in pulling a furnace of lightly doped single crystal silicon rods can be ignored. After drawing the lightly boron-doped single crystal silicon rods, the iron content of the guide tube can be characterized by testing the iron content using GDMS. The trace iron radiated by the heat of the guide tube acts on the lightly boron-doped single crystal silicon rods, which can be significantly reflected on the lightly boron-doped single crystal silicon rods. Therefore, the present invention utilizes the high sensitivity of the single crystal silicon polishing sheet to the metal content, and accurately infers the metal content of the guide tube through the metal content of the polishing sheet, thereby realizing non-destructive testing of the guide tube and ensuring that the test results are accurate and reliable.

[0061] During the production of monocrystalline silicon, the metal content of the monocrystalline silicon polishing sheet and the guide tube is at different orders of magnitude. Assuming that the metal content of the guide tube is 1PPMW, the corresponding metal content concentration is about 2.5E16. When the guide tube is used to pull a crystal rod, the guide tube will radiate metal to the monocrystalline silicon rod. At this time, the iron content concentration in the polishing sheet is about 5E10, which is about 1 million times different from the previous one. By detecting the iron content of the monocrystalline silicon polishing sheet and analyzing the iron content of the guide tube, the guide tube can be non-destructively tested and accurate test results can be obtained.

[0062] Application Example 1

[0063] S1: Use the guide tube to be tested to draw the single crystal silicon rod to be tested, process the single crystal silicon rod to be tested into the single crystal silicon polishing sheet to be tested, and calculate the distance parameter L between the single crystal silicon polishing sheet to be tested and the guide tube to be tested.

[0064] The diameter of the single crystal silicon polishing sheet to be tested drawn by the guide tube to be tested is D2=189mm, the diameter of the small diameter end of the guide tube to be tested is D1=250mm, and the distance parameter L between the single crystal silicon polishing sheet to be tested and the guide tube to be tested is 6.1cm.

[0065] S2: Detect the metallic iron content of the single crystal silicon polished wafer to be tested.

[0066] A standard guide tube with a metallic iron content of ≤0.5ppmw is selected to draw a standard single crystal silicon rod, and the diameter of the standard single crystal silicon rod and the single crystal silicon rod to be tested are the same;

[0067] Processing standard single crystal silicon rods into standard single crystal silicon polished wafers;

[0068] The SPV method is used to detect the standard single crystal silicon polished wafer to obtain a 49-point reference spectrum of the metal iron to be tested, and the SPV method is used to detect the single crystal silicon polished wafer to be tested to obtain a 49-point test spectrum of the metal iron to be tested; Figure 3 It is a 49-point reference map detected by the SPV method. The test points of the 49-point test map detected by the SPV method are the same as those of the reference map.

[0069] Table 2 below shows the specific location and map data of each test point of SPV49 (taking 6 o'clock as an example).

[0070] Table 2: Serial number X Cooxd.(mm) Y Coord.(mm) <![CDATA[49-point basic map data H J > <![CDATA[49-point test spectrum data H s > <![CDATA[SPV 49 - point increment H z > 1 0 0 5.59E+09 9.46E+09 3.87E+09 2 22.46 0 5.87E+09 2.76E+10 2.18E+10 3 15.88 15.88 8.07E+09 1.51E+10 7.08E+09 4 0 22.46 4.94E+09 5.97E+09 1.03E+09 5 -15.88 15.88 8.35E+09 2.17E+10 1.34E+10 6 -22.46 0 5.70E+09 2.59E+10 2.02E+10 7 -15.88 -15.88 3.87E+09 5.00E+09 1.14E+09 8 0 -22.46 8.00E+09 2.54E+10 1.74E+10 9 15.88 -15.88 6.64E+09 3.04E+10 2.37E+10 10 44.92 0 6.28E+09 2.35E+10 1.72E+10 11 41.5 17.19 2.70E+09 3.76E+09 1.06E+09 12 31.76 31.76 5.69E+09 7.98E+10 7.41E+10 13 17.19 41.5 7.80E+09 2.17E+10 1.39E+10 14 0 44.92 6.00E+09 4.07E+10 3.47E+10 15 -17.19 41.5 8.25E+09 1.27E+10 4.42E+09 16 -31.76 31.76 6.86E+09 2.10E+10 1.42E+10 17 -41.5 17.19 8.99E+09 1.19E+11 1.10E+11 18 -44.92 0 5.34E+09 2.62E+10 2.09E+10 19 -41.5 -17.19 8.66E+09 1.42E+11 1.33E+11 20 -31.76 -31.76 8.81E+09 8.26E+10 7.37E+10 21 -17.19 -41.5 7.14E+09 2.17E+10 1.46E+10 22 0 -44.92 7.07E+09 6.47E+10 5.77E+10 23 17.19 -41.5 6.87E+09 1.31E+10 6.21E+09 24 31.76 -31.76 1.71E+09 2.60E+09 8.83E+08 25 41.5 -17.19 7.37E+09 1.41E+11 1.34E+11 26 67.38 0 5.61E+09 3.42E+11 3.37E+11 27 65.08 17.44 8.32E+09 2.91E+11 2.82E+11 28 58.35 33.69 6.16E+09 2.97E+11 2.91E+11 29 47.64 47.64 6.91E+09 2.45E+11 2.38E+11 30 33.69 58.35 8.64E+09 1.74E+11 1.65E+11 31 17.44 65.08 8.83E+09 1.10E+11 1.02E+11 32 0 67.38 8.75E+09 2.19E+11 2.10E+11 33 -17.44 65.08 8.25E+09 1.11E+11 1.03E+11 34 -33.69 58.35 6.55E+09 2.15E+11 2.09E+11 35 -47.64 47.64 5.84E+09 9.40E+10 8.82E+10 36 -58.35 33.69 8.82E+09 3.25E+11 3.16E+11 37 -65.08 17.44 5.97E+09 2.85E+11 2.79E+11 38 -67.38 0 7.12E+09 3.89E+11 3.82E+11 39 -65.08 -17.44 6.86E+09 1.74E+11 1.67E+11 40 -58.35 -33.69 7.70E+09 2.16E+11 2.09E+11 41 -47.64 -47.64 7.27E+09 2.33E+11 2.26E+11 42 -33.69 -58.35 7.50E+09 1.83E+11 1.75E+11 43 -17.44 -65.08 8.54E+09 6.47E+10 5.62E+10 44 0 -67.38 5.23E+09 2.87E+11 2.82E+11 45 17.44 -65.08 7.76E+09 1.71E+11 1.63E+11 46 33.69 -58.35 6.05E+09 2.22E+11 2.16E+11 47 47.64 -47.64 8.93E+09 2.32E+11 2.23E+11 48 58.35 -33.69 6.96E+09 2.28E+11 2.21E+11 49 65.08 -17.44 7.97E+09 1.34E+11 1.26E+11 Summary / / / / 1.20E+11 .

[0071] The iron content of the polishing sheet is H, H={(H s1 -H J1 )+(H s2 -H J2 )+(H s3 -H J3 )+.......(Hs49 -H J49 )} / 49=(H z1 +H z2 +H z3 +……H z49 ) / 49.

[0072] S3: Predict the metallic iron content of the guide tube to be tested through the guide tube prediction model G=1.23+0.264*L-0.0983*H+0.0866*L*H.

[0073] Finally, a sample from the small diameter end of the guide tube to be tested was taken for GDMS test to verify the accuracy of the metallic iron content predicted by the guide tube prediction model.

[0074] In this application example, the metal iron content of the guide tube to be tested obtained according to the guide tube prediction model, the metal iron content of the guide tube to be tested obtained by GDMS test, and the intermediate detection data are shown in Table 3 below:

[0075] Table 3: project <![CDATA[SPV 49 point result * E9 / atoms / cm 3 > Diameter of the single crystal silicon rod to be tested after rolling / mm Distance parameter between the single crystal silicon polishing sheet to be tested and the guide tube to be tested L / cm The metallic iron content in the guide tube to be tested obtained according to the guide tube prediction model / ppmw The metallic iron content in the guide tube to be tested obtained by GDMS / ppmw Deviation range / % Application Example 1 5.1 189 6.1 5.03 4.9 2.5% .

[0076] As shown in Table 3, the deviation between the metallic iron content obtained by the prediction method of the present invention and the metallic iron content obtained by GDMS is less than 5%, which is within the controllable range. This shows that the metal content of the guide tube can be accurately calculated based on the metal content of the polishing sheet, and non-destructive testing of the guide tube can be achieved, and the test results are accurate and reliable.

[0077] It should be noted that the diameter of the single crystal silicon rod to be tested after rolling in Table 3 is the diameter of the polishing sheet prepared from the single crystal silicon rod to be tested. Since the diameter of the single crystal silicon rod is not completely consistent, there may be tiny irregular shapes and slight defects on the surface. In order to avoid these problems, it is necessary to process it after rolling to obtain a standard circular polishing sheet. During the rolling process, the removal amount is maintained at no more than 5mm (that is, the diameter difference between the single crystal silicon rod to be tested and the single crystal silicon polishing sheet to be tested is no more than 5mm) to reduce the impact of the rolling operation on the metal content of the single crystal silicon rod. In addition, in other processes of the present invention for processing the single crystal silicon rod into a single crystal silicon polishing sheet, the diameter difference between the single crystal silicon rod and the corresponding polishing sheet is maintained at no more than 5mm.

[0078] Application Example 2

[0079] The overall method of Application Example 2 is the same as that of Application Example 1, with the main difference being that the selected guide tube to be tested and the diameter of the pulled single crystal silicon rod to be tested are different from those of Application Example 1.

[0080] In this application example 2, the metal iron content of the guide tube to be tested obtained according to the guide tube prediction model, the metal iron content of the guide tube to be tested obtained by GDMS test, and the intermediate detection data are shown in Table 4 below:

[0081] Table 4: project <![CDATA[SPV 49 point result * E9 / atoms / cm 3 > Diameter of the single crystal silicon rod to be tested after rolling / mm Distance parameter between the single crystal polishing sheet to be tested and the guide tube to be tested L / cm The metallic iron content in the guide tube to be tested obtained according to the guide tube prediction model / ppmw The metallic iron content in the guide tube to be tested obtained by GDMS / ppmw Deviation range / % Application Example 2 34 207 4.3 11.68 12.2 4.5%

[0082] As shown in Table 4, the deviation between the metallic iron content obtained by the prediction method of the present invention and the metallic iron content obtained by GDMS is less than 5%, which is within the controllable range. This indicates that the metal content of the guide tube can be accurately calculated based on the metal content of the polishing sheet, and non-destructive testing of the guide tube can be achieved, and the test results are accurate and reliable.

[0083] Comparative Example 1

[0084] GDMS was used to test the iron content of the furnace sample of the guide tube in Application Example 1. The iron content of the furnace sample of the guide tube in Application Example 1 was 4.0 ppmw, which had a deviation range of (4.9-4.0) / 4.9*100%=18.36% from the metal iron content in the guide tube to be tested by GDMS, which was much larger than the iron content of the guide tube obtained by the prediction method of the present invention.

[0085] The above description of the disclosed embodiments enables one skilled in the art to implement or use the present invention. Various modifications to these embodiments will be apparent to one skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but rather to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for predicting the metal content of a guide tube based on a single crystal silicon polishing sheet, characterized in that: The following steps are involved: S1, using the guide tube to be tested to draw the single crystal silicon rod to be tested, processing the single crystal silicon rod to be tested into the single crystal silicon polishing sheet to be tested, detecting the diameter of the single crystal silicon polishing sheet to be tested and calculating the distance parameter between the single crystal silicon polishing sheet to be tested and the guide tube to be tested; S2, detecting the metal content of the single crystal silicon polished wafer to be tested; S3, predicting the metal content of the guide tube to be tested by using the guide tube prediction model; Among them, the metal content to be predicted is the iron content, and the guide tube prediction model is G=1.23+0.264*L-0.0983*H+0.0866*L*H, G is the metal content of the guide tube to be tested, L is the distance parameter between the single crystal silicon polishing sheet to be tested and the guide tube to be tested, and H is the metal content of the single crystal silicon polishing sheet to be tested.

2. The method for predicting the metal content of a guide tube based on a single crystal silicon polishing sheet according to claim 1, characterized in that: The calculation method of the distance parameter between the single crystal silicon polishing sheet to be tested and the guide tube to be tested is: L= (D1-D2), D1 is the diameter of the small diameter end of the guide tube to be tested, and D2 is the diameter of the single crystal silicon polishing sheet to be tested.

3. The method for predicting the metal content of a guide tube based on a single crystal silicon polishing sheet according to claim 1, characterized in that: The diameter difference between the single crystal silicon rod to be tested and the single crystal silicon polished wafer to be tested is not greater than 5 mm.

4. The method for predicting the metal content of a guide tube based on a single crystal silicon polishing sheet according to claim 1, characterized in that: The method for detecting the metal content of the single crystal silicon polishing sheet to be tested is the SPV surface photovoltage method.

5. The method for predicting the metal content of a guide tube based on a single crystal silicon polishing sheet according to claim 1, characterized in that: The metal content detection method of the single crystal silicon polishing wafer to be tested is: S201, selecting a standard guide tube with a metal content to be predicted ≤ 0.5 ppmw to draw a standard single crystal silicon rod, and the diameter of the standard single crystal silicon rod and the single crystal silicon rod to be tested are the same; S202, processing the standard single crystal silicon rod into a standard single crystal silicon polished wafer; S203, using the SPV method to detect the standard single crystal silicon polished wafer to obtain a 49-point reference spectrum of metals in the standard single crystal silicon polished wafer, and using the SPV method to detect the single crystal silicon polished wafer to be tested to obtain a 49-point test spectrum of metals in the single crystal silicon polished wafer to be tested; S204, based on the 49-point test spectrum and the 49-point reference spectrum, the metal content H of the single crystal silicon polished wafer to be tested is obtained, where H={(H s1 -H J1 )+(H s2 -H J2 )+(H s3 -H J3 )+.......(H s49 -H J49 )} / 49,H sx is the metal content of the xth point in the 49-point test spectrum, H Jx It is the metal content of the x-th point in the 49-point reference spectrum, where x is any natural number from 1 to 49.

6. The method for predicting the metal content of a guide tube based on a single crystal silicon polishing sheet according to claim 5, characterized in that: The method for establishing the guide tube prediction model is as follows: first, n guide tubes with different metal contents and the same size are selected, where n is a natural number and n≥3; second, the n guide tubes are used to draw lightly boron-doped single crystal silicon rods of different diameters; third, the lightly boron-doped single crystal silicon rods are processed into polished sheets, and the metal content of the polished sheets is tested; finally, sample blocks are taken from the small diameter end of the guide tube after the lightly boron-doped single crystal silicon rods are drawn, the metal content of the guide tube is detected by GDMS, and the guide tube prediction model is obtained by regression analysis using mathematical statistical tools.

7. The method for predicting the metal content of a guide tube based on a single crystal silicon polishing sheet according to claim 5, characterized in that: The raw materials of lightly boron-doped single crystal silicon rods are polycrystalline silicon and boron master alloy.

8. The method for predicting the metal content of a guide tube based on a single crystal silicon polishing sheet according to claim 5, characterized in that: The method for establishing the guide tube prediction model is specifically as follows: first, three guide tubes with different metal contents and the same size are selected, and three lightly doped single crystal silicon rods with different diameters are drawn using each guide tube. The lightly doped single crystal silicon rods are then processed into polished wafers, and the metal content of the polished wafers is tested using the SPV surface photovoltage method.

Images