Resistivity test adaptation method for single crystal rod feeding
By cutting samples from the head and tail of single crystal rods, setting up a dense test point map, and calculating the average resistivity value, the problem of resistivity non-uniformity caused by uneven single crystal growth interface was solved. This achieved the matching of the four-probe method and the eddy current method, reducing feeding errors and waste.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- SHANDONG GRINM SEMICON MATERIALS CO LTD
- Filing Date
- 2025-12-11
- Publication Date
- 2026-04-24
AI Technical Summary
The unevenness of the growth interface during single crystal growth leads to non-uniform resistivity, resulting in different substrate resistivity test results. The four-probe method has large testing errors and it is difficult to match the feeding range with the factory output range.
Multiple samples were cut from both ends of a single crystal rod using the four-probe method. A dense test point map was set up, the equipment was calibrated, and the average resistivity value was obtained. Outliers were removed, and the results were matched with those from the eddy current method.
This reduces the testing error of the four-probe method, achieves the matching of the resistivity of the four-probe feeding method with the resistivity of the eddy current method, and reduces feeding waste.
Smart Images
Figure CN121917844A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to a resistivity testing adaptation method for single crystal rod feeding, used for resistivity testing of feeding materials in semiconductor material production. Background Technology
[0002] Generally speaking, the electrical properties of semiconductor materials are closely related to factors such as impurity type, impurity concentration, temperature, and crystal defects. However, due to the uneven growth interface during single crystal growth, the resistivity of single crystal slices grown by Czochralski or zone melting is not uniform, which leads to different results in the substrate resistivity test due to different test point locations.
[0003] The resistivity of semiconductor materials is the reciprocal of their conductivity and is one of the important electrical parameters. It reflects the compensated impurity concentration and is directly related to the carrier concentration. Undoped intrinsic silicon carriers are generated by the accidental breaking of covalent bonds in silicon atoms due to lattice thermal vibrations.
[0004] The four-probe method is generally used to test the resistivity of semiconductors. This method is simple in principle, but its drawback is that it can damage the surface. However, to meet the requirements of online testing in automated production lines, the eddy current method, a non-destructive testing method, is typically used in subsequent processes. In substrate factories, the four-probe method is generally used to test the resistivity of single-crystal segments cut at both ends to control the approximate resistivity range of the single-crystal segment. Based on the four-probe test results, materials are fed into the substrate, and after processing into substrate wafers, the resistivity of the substrate wafers is tested using the eddy current method.
[0005] During the production process, the resistivity range of the single crystal segment was defined using the four-probe method, but after feeding the material, there were still unqualified products whose slice resistivity test exceeded the specification range. Summary of the Invention
[0006] To address the aforementioned problems in the existing technology, the purpose of this invention is to propose a resistivity testing adaptation method for single crystal rod feeding. Using this method can reduce resistivity testing errors during single crystal rod feeding and achieve matching between the feeding range and the factory range.
[0007] To achieve the above objectives, the present invention adopts the following technical solution: A resistivity testing adaptation method for single crystal rod feeding includes the following steps: (1) Cut off both ends of the single crystal rod to obtain multiple test samples; (2) Set up a map of the four-probe dense test points; (3) After calibrating the four-probe device with the standard sample, place the sample on the four-probe stage and retrieve the dense test point map for testing; (4) The average resistivity values of the obtained dense test points are used as the true resistivity value of the sample to obtain the resistivity feeding range of the single crystal in this section.
[0008] In this invention, in step (1), the thickness of the test sample is 1-2 mm to reduce the error of the four probes; the TTV (Total Thickness Variation) of the sample surface is less than 10, and the height difference of the cutting line marks on the surface is less than 20 μm.
[0009] In this invention, in step (2), the four-probe test point map is set in a circular shape and extends to the edge in a radial pattern, with the distribution radius of the points ranging from 1 to 30 mm.
[0010] In this invention, in step (2), the number of points in the dense point test point map is 7-50.
[0011] In this invention, in step (3), after the four-probe device is calibrated with the standard sample, if the resistivity is maintained at <0.3Ω·cm, the 100mA range is used for measurement; if the resistivity is maintained at 0.3~3Ω·cm, the 10mA current range is used for measurement; if the resistivity is maintained at 3~30Ω·cm, the 1mA current range is used for measurement; if the resistivity is maintained at 30Ω·cm or above, the 0.1mA current range is used for measurement.
[0012] In this invention, the formula for averaging the resistivity values at dense test points is as follows: ,in Let be the resistivity measured by the four probes at point i, and when If the measurement value at that point is removed, the average value is recalculated.
[0013] The advantages of this invention are: This invention provides a method for matching the resistivity of silicon wafers tested by four probes with the resistivity tested by eddy current method. This method can achieve matching between the resistivity of the four probes and the resistivity of the eddy current method at the factory, thereby reducing material waste caused by testing errors and minimizing testing deviations. Attached Figure Description
[0014] Figure 1 This is a flowchart of the testing process for the present invention.
[0015] Figure 2 This is a map of the four-probe dense point test locations in an embodiment of the present invention.
[0016] Figure 3 This is a comparison chart of resistivity tests in an embodiment of the present invention. Detailed Implementation
[0017] The present invention will be further described below with reference to the accompanying drawings and embodiments, but this does not imply a limitation on the scope of protection of the present invention.
[0018] like Figure 1As shown, this invention provides a resistivity testing adaptation method for single crystal rod feeding, and the specific testing process includes the following steps: (1) Cut multiple sample pieces with a thickness of 1-2 mm from both ends of the single crystal rod. The sample surface TTV < 10 and the surface is flat with a cutting line height difference < 20 μm. (2) Set up a map of four probe dense test points. The radius of the map ranges from 1 to 30 mm, and the number of points is 7 to 50. (3) After calibrating the four-probe device with the standard sample, place the sample on the four-probe stage and retrieve the dense test point map for testing. During the test, if the resistivity is <0.3Ω·cm, use the 100mA range for measurement; if the resistivity is 0.3~3Ω·cm, use the 10mA current range for measurement; if the resistivity is 3~30Ω·cm, use the 1mA current range for measurement; if the resistivity is above 30Ω·cm, use the 0.1mA current range for measurement.
[0019] (4) The average resistivity values obtained from the dense test points are used as the true resistivity value of the sample to obtain the resistivity feeding range of this section of single crystal. The formula for calculating the average value in this step is: ,in Let be the resistivity measured by the four probes at point i, and when If the measurement value at that point is removed, the average value is recalculated.
[0020] Example In this embodiment, the ADE7200 device is used for the non-contact resistivity measurement method, and the RT3000 four-probe device is used for the four-probe resistivity measurement method. The resistivity testing adaptation method steps in this embodiment are as follows: (1) Cut off the head and tail ends of the single crystal rod to obtain multiple sample pieces with a thickness of 2mm. Select 12 sample pieces to be tested cut from the single crystal rod. The surface TTV of the sample pieces is <10, and the surface is flat with a cutting line height difference of <20μm. (2) The four-probe testing equipment is first calibrated according to the test standard to achieve a test deviation within ±3% of the test standard sample; the probe diameter D of the non-contact resistivity testing equipment is then used to set up a map of the four probe center dense point test points based on the diameter D; such as Figure 2 As shown, 41 points are set in the area, with the outermost circle having a radius of 6mm. A circle is set every 1mm inwards, for a total of 5 circles. Eight test points are set at equal intervals on each circle, with one point at the center.
[0021] (3) Place the sample on the four-probe stage to retrieve the dense test point map for testing. After starting the test, 41 resistivity data points can be obtained for each sample.
[0022] (4) The average value of the resistivity data of 41 points in each sample is taken as the true resistivity value of the sample and recorded to obtain the resistivity feeding range of the single crystal in this section.
[0023] The above 12 samples were fabricated into conventional silicon wafers with a thickness of approximately 740μm. The resistivity was tested using the eddy current method with an ADE7200 instrument and the resistivity was tested using a four-probe single-point instrument with an RT3000 four-probe instrument.
[0024] Plot the ADE test data, the four-probe dense point test data, and the four-probe point test data into line graphs. The results are as follows: Figure 3 As shown, the four-probe map dense point test method of this invention is well-matched with the ADE test results, with an error of less than 1%, while the four-probe single-point test results have a larger error. Verification shows that this method can achieve matching between the resistivity of the single crystal feed and the resistivity at the factory.
Claims
1. A resistivity testing adaptation method for single crystal rod feeding, characterized in that, The method includes the following steps: (1) Cut off both ends of the single crystal rod to obtain multiple sample pieces; (2) Set up a map of the four-probe dense test points; (3) After calibrating the four-probe device with the standard sample, place the sample on the four-probe stage and retrieve the dense test point map for testing; (4) The average resistivity values of the obtained dense test points are used as the true resistivity value of the sample to obtain the resistivity feeding range of the single crystal in this section.
2. The resistivity testing adaptation method for single crystal rod feeding according to claim 1, characterized in that, In step (1), the thickness of the cut test sample is 1-2 mm, the surface TTV is <10, and the surface is flat with a cutting line height difference of <20 μm.
3. The resistivity testing adaptation method for single crystal rod feeding according to claim 1, characterized in that, In step (2), the number of points in the dense test point map is 7-50.
4. The resistivity testing adaptation method for single crystal rod feeding according to claim 1, characterized in that, In step (2), the four-probe test point map is set in a circle, extending radially to the edge, with a point distribution radius ranging from 1 to 30 mm.
5. The resistivity testing adaptation method for single crystal rod feeding according to claim 1, characterized in that, In step (3), after the four-probe device is calibrated with the standard sample, if the resistivity is <0.3Ω·cm, use the 100mA range for measurement; if the resistivity is 0.3~3Ω·cm, use the 10mA current range for measurement; if the resistivity is 3~30Ω·cm, use the 1mA current range for measurement; if the resistivity is above 30Ω·cm, use the 0.1mA current range for measurement.
6. The resistivity testing adaptation method for single crystal rod feeding according to claim 1, characterized in that, In step (4), the formula for averaging the resistivity values at dense test points is as follows: ,in Let be the resistivity measured by the four probes at point i, and when If the measurement value at that point is removed, the average value is recalculated.