Method for detecting semiconductor monocrystalline silicon ingot before slicing

Abstract

The invention provides a method for detecting a semiconductor monocrystalline silicon ingot before slicing, and relates to the technical field of semiconductor cutting methods, before cutting the ingot, cutting a detection sample from the ingot, performing product specification detection on the detection sample, and judging whether the detection sample meets the product specification requirement or not through the product specification detection; carrying out in-plane distribution detection on the detection sample wafer if the detection sample wafer meets the product specification requirement, judging that the in-plane distribution of the single crystal is abnormal and stopping slicing processing in advance if the detection sample wafer meets the in-plane distribution in the limit sample, and judging that the single crystal is normal and continuing slicing processing if the detection sample wafer does not meet the in-plane distribution in the limit sample; the state of the impurities in the crystal ingot body is detected before slicing, and when the impurities in the crystal ingot body are not uniform, slicing is stopped, so that waste of processing cost is avoided.

Description

Technical Field

[0001] The present invention belongs to the technical field of semiconductor multi-wire cutting methods, and in particular relates to a method for detecting semiconductor single crystal silicon ingots before slicing. Background Art

[0002] Silicon wafer warpage is a key quality issue in the semiconductor manufacturing and photovoltaic industries. Warpage occurs during the single crystal silicon growth stage and during wafer slicing. In existing technologies, a crystal ingot is first pulled, then cut into ingots. The ingot quality is then inspected to ensure that the pulled ingot meets the target product specifications. The ingots that meet the product specifications are then cut into silicon wafers. Some silicon wafers may exhibit abnormalities such as cracking, poor morphology, or excessive warpage. Because impurity distribution testing is not performed during the ingot quality inspection, abnormalities may occur in the silicon wafers after slicing. It is impossible to determine whether the problem is with the single crystal or the slicing, resulting in a certain amount of waste in processing costs. Summary of the Invention

[0003] In view of this, the present invention provides a method for detecting semiconductor single crystal silicon ingots before slicing, so as to know in advance the uneven state of impurities in the ingot body and avoid wasting processing costs.

[0004] The technical solution adopted by the present invention to solve its technical problem is: A method for detecting a semiconductor single crystal silicon ingot before slicing, comprising the following steps: Step S1: before cutting the ingot, cutting a test sample from the ingot; Step S2: Perform product specification testing on the test sample to determine whether the test sample meets the product specification requirements; Step S3: If the test sample meets the product specification requirements, perform in-plane distribution detection on the test sample. If the test sample meets the in-plane distribution of the limit sample, it is determined that the in-plane distribution of the single crystal is abnormal, and the slicing process is stopped in advance; If the test sample does not meet the in-plane distribution of the limit sample, the in-plane distribution of the single crystal is judged to be normal and the slicing process is continued.

[0005] Preferably, in step S3, the limit sample is obtained by the following steps: S31: Cut samples from several ingots of the same product specification, perform product specification testing on the samples, and obtain an in-plane distribution map; S32: After the test is completed, several ingots of the same product specification are sliced ​​to obtain several experimental silicon wafers; S33: Warp and morphology inspection are performed on several experimental silicon wafers, and sample silicon wafers are obtained through screening; S34: Check the in-plane distribution map of the sample silicon wafer to obtain a limit sample.

[0006] Preferably, the S33 is specifically as follows: Step 1: performing Warp testing on several experimental silicon wafers, and obtaining an initial sample silicon wafer when the Warp of the experimental silicon wafer is greater than the product specification requirement; Step 2: Performing morphology detection on the initial sample silicon wafer to obtain a morphology map of the initial sample silicon wafer. When the middle portion of the morphology map of the initial sample silicon wafer is raised, a sample silicon wafer is obtained.

[0007] Preferably, the in-plane distribution is one or more of the in-plane distribution of resistivity, the in-plane distribution of carbon content, and the in-plane distribution of oxygen content.

[0008] Preferably, in step S3, the conformity of the detection sample to the in-plane distribution of the limit sample is specifically that the similarity between the in-plane distribution diagram of the detection sample and the in-plane distribution diagram of the limit sample is greater than 90%.

[0009] Preferably, when the in-plane distribution is the in-plane distribution of oxygen content, in step S3, the detection sample conforms to the in-plane distribution in the limit sample, specifically, the similarity between the central area of ​​the in-plane distribution diagram of the detection sample and the central area of ​​the in-plane distribution diagram of the limit sample is greater than 90%.

[0010] Preferably, the central area is 30%-50% of the central area.

[0011] Preferably, the conventional tests include resistivity, oxygen content, carbon content, and microscopic defects.

[0012] Compared with the prior art, the present invention has the following beneficial effects: The present invention provides a method for detecting semiconductor single crystal silicon ingots before slicing. Before cutting the crystal ingot, a detection sample is cut from the crystal ingot, and the detection sample is subjected to product specification detection. The product specification detection is used to determine whether the detection sample meets the product specification requirements; if the detection sample meets the product specification requirements, an in-plane distribution detection is performed on the detection sample. If the detection sample meets the in-plane distribution in the limit sample, it is determined that the in-plane distribution of the single crystal is abnormal, and the slicing process is stopped in advance. If the detection sample does not meet the in-plane distribution in the limit sample, it is determined that the single crystal is normal, and the slicing process is continued, so as to detect the status of impurities in the crystal ingot before slicing. When the impurities in the crystal ingot are uneven, the slicing is stopped to avoid wasting processing costs. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] Figure 1 The oxygen content distribution diagram is given as an example.

[0014] Figure 2 This is a morphology diagram of the silicon wafer of the embodiment.

[0015] Figure 3 The following are sample images for implementation examples. DETAILED DESCRIPTION

[0016] The technical solutions and technical effects of the embodiments of the present invention are further elaborated below in conjunction with the accompanying drawings of the present invention.

[0017] A method for detecting a semiconductor single crystal silicon ingot before slicing, comprising the following steps: Step S1: before cutting the ingot, cutting a test sample from the ingot; Step S2: Subject the test sample to product specification testing to determine whether the test sample meets the product specification requirements through product specification testing; determine whether the single crystal silicon ingot is qualified through preliminary testing, which will not affect the judgment of the warp cause in the subsequent steps; the product specification requirements refer to the indicators of the customer's product requirements, such as resistivity less than 0.05Ω·cm, oxygen content less than 21ppma, carbon content <0.5ppma, micro defects, etc.

[0018] Step S3: If the test sample meets the product specification requirements, perform in-plane distribution detection on the test sample. If the test sample meets the in-plane distribution of the limit sample, it is determined that the in-plane distribution of the single crystal is abnormal, and the slicing process is stopped in advance; If the test sample does not meet the in-plane distribution of the limit sample, the in-plane distribution of the single crystal is judged to be normal and the slicing process is continued.

[0019] Compared with the prior art, the present invention has the following beneficial effects: The present invention provides a method for detecting semiconductor single crystal silicon ingots before slicing. Before cutting the crystal ingot, a detection sample is cut from the crystal ingot, and the detection sample is subjected to product specification detection. The product specification detection is used to determine whether the detection sample meets the product specification requirements; if the detection sample meets the product specification requirements, an in-plane distribution detection is performed on the detection sample. If the detection sample meets the in-plane distribution in the limit sample, it is determined that the in-plane distribution of the single crystal is abnormal, and the slicing process is stopped in advance. If the detection sample does not meet the in-plane distribution in the limit sample, it is determined that the single crystal is normal, and the slicing process is continued, so as to detect the status of impurities in the crystal ingot before slicing. When the impurities in the crystal ingot are uneven, the slicing is stopped to avoid wasting processing costs.

[0020] Furthermore, in step S3, the limit sample is obtained by the following steps: S31: Cut samples from several ingots of the same product specification, perform product specification testing on the samples, and obtain an in-plane distribution map; S32: After the test is completed, several ingots of the same product specification are sliced ​​to obtain several experimental silicon wafers; S33: Warp and morphology inspection are performed on several experimental silicon wafers, and sample silicon wafers are obtained through screening; S34: Check the in-plane distribution map of the sample silicon wafer to obtain a limit sample.

[0021] Among them: the limit samples under the same product specification are the same.

[0022] Furthermore, the S33 is specifically as follows: Step 1: performing warp testing on several experimental silicon wafers, and obtaining initial sample silicon wafers when the warp of the experimental silicon wafers is greater than the product specification requirement, such as the product specification requirement of warp < 10 μm or warp < 5 μm; Step 2: Perform morphology detection on the initial sample silicon wafer to obtain the morphology map of the initial sample silicon wafer. When the middle part of the morphology map of the initial sample silicon wafer is bulged, the sample silicon wafer is obtained. If only Warp is used to judge, it is impossible to know whether it is caused by uneven distribution within the crystal rod surface or caused by mechanical processing during the cutting process. Therefore, warp and morphology are judged at the same time to make the obtained limit sample accurate; the bulge in the middle of the morphology map refers to the bulge in the middle of the morphology map, and the color of the radial uneven distribution pattern changes gradually.

[0023] Furthermore, the in-plane distribution is one or more of the in-plane distribution of resistivity, in-plane distribution of carbon content, and in-plane distribution of oxygen content; for example, when the limit sample is the in-plane distribution of oxygen content, the graph is as follows: Figure 1 As shown in the figure, during the Czochralski growth process, the erosion of the quartz crucible causes oxygen to dissolve in the silicon melt and enter the lattice as the crystal grows, forming interstitial oxygen. The distribution of oxygen is affected by thermal field convection, crystal / crucible rotation rate, and pulling speed, usually showing a radial gradient. The edge oxygen content is more easily enriched due to melt convection and crucible contact, and the following regular distribution will appear: (1) Resistivity distribution within the surface: high in the middle and low at the edges; (2) Oi content distribution in the plane: high in the middle and low at the edge; (3) Cs content distribution in the plane: low in the middle and high at the edge; Because oxygen atoms occupy the interstitial positions of the silicon lattice, the local lattice expands. If the distribution within the surface is uneven, it will cause residual stress gradients. The stress release after cutting will cause warp abnormalities. By pre-detecting the distribution within the ingot surface, it is possible to know in advance whether the warp is abnormal during the single crystal growth stage, thereby avoiding waste of production costs due to poor slicing caused by uneven impurities in the ingot body.

[0024] Furthermore, in step S3, the conformity of the detection sample to the in-plane distribution of the limit sample is specifically that the similarity between the in-plane distribution diagram of the detection sample and the in-plane distribution diagram of the limit sample is greater than 90%.

[0025] Furthermore, when the in-plane distribution is the in-plane distribution of oxygen content, in step S3, the detection sample meets the in-plane distribution in the limit sample, specifically, the similarity between the central area of ​​the in-plane distribution diagram of the detection sample and the central area of ​​the in-plane distribution diagram of the limit sample is greater than 90%.

[0026] Furthermore, the central area is 30%-50% of the central area, wherein the central area refers to the ratio of the area formed by a circle with the central point as the center and a certain length as the radius to the entire surface distribution map.

[0027] In one embodiment, taking the in-plane distribution uniformity of oxygen content as an example, a limit sample is obtained by the following steps: a 12-inch heavily doped red Ph (phosphorus) single crystal is pulled, and its product specifications are: resistivity specification 0.001~0.005Ω·cm; oxygen content <21ppma; the pulled single crystal is cut into ingots, and samples are cut from the ingots and subjected to 61-point oxygen testing. Based on the test data, an in-plane distribution map of oxygen is drawn, the ingots are sliced, and the silicon wafers are subjected to warp testing. The warp is 40μm, and its morphology is as follows: Figure 2 As shown, the color is uneven and the middle part is bulging, so the oxygen content distribution diagram of this limit sample is as follows Figure 3 shown.

[0028] Furthermore, the product specification testing includes resistivity, oxygen content, carbon content, and microscopic defects.

[0029] The above disclosure is only a preferred embodiment of the present invention, and it is certainly not intended to limit the scope of the present invention. A person skilled in the art can understand that all or part of the processes of the above embodiment and equivalent changes made in accordance with the claims of the present invention are still within the scope of the invention.

Claims

1. A method for detecting semiconductor single crystal silicon ingots before slicing, characterized in that: The following steps are involved: Step S1: before cutting the ingot, cutting a test sample from the ingot; Step S2: Perform product specification testing on the test sample to determine whether the test sample meets the product specification requirements; Step S3: If the test sample meets the product specification requirements, perform in-plane distribution detection on the test sample. If the test sample meets the in-plane distribution of the limit sample, it is determined that the in-plane distribution of the single crystal is abnormal, and the slicing process is stopped in advance; If the test sample does not meet the in-plane distribution of the limit sample, the in-plane distribution of the single crystal is judged to be normal and the slicing process is continued.

2. The method for detecting semiconductor single crystal silicon ingots before slicing according to claim 1, wherein: In step S3, the limit sample is obtained by the following steps: S31: Cut samples from several ingots of the same product specification, perform product specification testing on the samples, and obtain an in-plane distribution map; S32: After the test is completed, several ingots of the same product specification are sliced ​​to obtain several experimental silicon wafers; S33: Warp and morphology inspection are performed on several experimental silicon wafers, and sample silicon wafers are obtained through screening; S34: Check the in-plane distribution map of the sample silicon wafer to obtain a limit sample.

3. The method for detecting semiconductor single crystal silicon ingots before slicing according to claim 2, wherein: The S33 is specifically as follows: Step 1: performing a Warp test on several experimental silicon wafers, and obtaining an initial sample silicon wafer when the Warp of the experimental silicon wafer is greater than the product specification requirement; Step 2: Performing morphology detection on the initial sample silicon wafer to obtain a morphology map of the initial sample silicon wafer. When the middle portion of the morphology map of the initial sample silicon wafer is raised, a sample silicon wafer is obtained.

4. The method for detecting semiconductor single crystal silicon ingots before slicing according to claim 1 or 3, wherein: The in-plane distribution is one or more of the in-plane distribution of resistivity, the in-plane distribution of carbon content, and the in-plane distribution of oxygen content.

5. The method for detecting semiconductor single crystal silicon ingots before slicing according to claim 4, wherein: In step S3, the detection sample conforms to the in-plane distribution of the limit sample, specifically, the similarity between the in-plane distribution diagram of the detection sample and the in-plane distribution diagram of the limit sample is greater than 90%.

6. The method for detecting semiconductor single crystal silicon ingots before slicing according to claim 4, wherein: When the in-plane distribution is the in-plane distribution of oxygen content, in step S3, the detection sample meets the in-plane distribution in the limit sample, specifically, the similarity between the central area of ​​the in-plane distribution diagram of the detection sample and the central area of ​​the in-plane distribution diagram of the limit sample is greater than 90%.

7. The method for detecting semiconductor single crystal silicon ingots before slicing according to claim 6, wherein: The central area is 30%-50% of the central area.

8. The method for detecting semiconductor single crystal silicon ingots before slicing according to claim 1 or 2, wherein: The product specification tests include resistivity, oxygen content, carbon content, and microscopic defects.

Images