Method for rapidly evaluating dislocation density of silicon carbide substrate plane by using stress
By evaluating the dislocation density (BPD) of silicon carbide substrates using stress detection and nonlinear curve fitting functions, this method solves the problems of sample damage and insufficient imaging capability in existing detection methods, and achieves non-destructive, low-cost, and efficient dislocation density testing.
Patent Information
- Application Number
- CN202411114755.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Priority Date
- 2024-07-11
- Filing Date
- 2024-08-14
- Publication Date
- 2024-11-08
AI Technical Summary
Existing technologies for detecting the dislocation density in the basal plane of silicon carbide substrates suffer from the problems of destructive detection methods causing damage to the sample and non-destructive detection methods having insufficient defect imaging capabilities.
Stress detection combined with nonlinear curve fitting function is used to obtain the average stress value through a wafer stress detector. The BPD density in a local area of silicon carbide substrate is evaluated using functional relationship. The BPD density is then nondestructively evaluated using KOH etching and dislocation defect detector.
This technology enables non-destructive and low-cost planar dislocation density testing on silicon carbide substrates, improving testing efficiency and accuracy.
Smart Images

Figure CN118914073A_ABST
Abstract
Description
Technical Field
[0001] The invention belongs to the field of semiconductor technology, and in particular relates to a method for rapidly evaluating the dislocation density of a silicon carbide substrate basal plane by utilizing stress. Background Art
[0002] In recent years, silicon carbide (SiC) has become one of the most popular wide bandgap semiconductor materials due to its excellent performance characteristics such as high breakdown electric field strength, high electron mobility and high thermal conductivity, especially in the market with growing demand for power electronic devices. Although the single crystal growth process of SiC crystals has been significantly optimized and crystals with low micropipe density have been successfully prepared, there are still a large number of structural defects, such as basal plane dislocations (BPDs), screw dislocations (TSDs), edge dislocations (TEDs) and mixed dislocations (TMDs). These defects seriously affect the performance and service life of the device. For example, the presence of BPDs will increase the on-resistance of the device, thereby reducing the reliability of MOSFET (metal-oxide semiconductor field-effect transistor) and bipolar devices. Bipolar PiN power devices prepared on substrates with high BPD density may have unstable forward voltage drift when operating in forward conduction mode for a long time.
[0003] At present, the technologies used for dislocation defect density detection are generally divided into two categories: destructive and non-destructive. The wafer is corroded in a high-temperature molten alkaline solution, and then the corrosion pits on the surface are observed through an optical microscope to obtain the dislocation defects on the surface of the wafer. This is a destructive detection method. The pattern formed after alkaline corrosion depends on experimental conditions such as corrosion time and corrosion liquid temperature. Although the alkaline corrosion method can detect all defects under the surface of the silicon carbide sample at one time, and the sample preparation is easy and the cost is low, alkaline corrosion is an irreversible process and will cause permanent damage to the sample. The non-destructive (non-destructive) detection method is mainly X-ray topography, whose biggest feature is that it requires a larger X-ray emitter, and its defect imaging capability still needs to be further improved. Summary of the invention
[0004] In order to solve the deficiencies in the above-mentioned prior art, the present invention provides a method for quickly evaluating the dislocation density of the basal plane of a silicon carbide substrate using stress. This method can simultaneously take into account the advantages of non-destructive and low-cost in traditional detection methods, thereby simplifying the testing process and improving the efficiency of experimental testing.
[0005] To achieve the above object, the technical solution of the present invention is as follows: A method for rapidly evaluating the dislocation density of a silicon carbide substrate basal plane by using stress, comprising: (1) Perform stress detection on the silicon carbide substrate to be tested, select any area, and obtain the average stress value (MPa) X of the substrate to be tested; (2) Evaluation of the BPD density (cm) in a local area of the SiC substrate -2 )Y, the functional relationship is as follows: .
[0006] Preferably, in step (1), stress detection is performed on the silicon carbide substrate to be tested to obtain an average stress value of the substrate to be tested, including: using a wafer stress detector to perform stress detection on the silicon carbide substrate to be tested.
[0007] The presence of BPD will cause the atoms around it to deviate from the equilibrium position, resulting in lattice distortion and the generation of stress fields. At the same time, the greater the stress, the more incomplete the arrangement of atoms inside the crystal, and the density of BPD may increase. Thermal stress is an important factor in the nucleation of BPD in silicon carbide. During the growth of silicon carbide single crystals, when the thermal stress exceeds a critical value, it will trigger the nucleation and proliferation of BPD and release the accumulated thermal stress. Although the formation of dislocations can release thermal stress, the atomic distortion at the core of the dislocation will cause the atoms at the core of the dislocation to be subjected to great stress. At the same time, when the thermal stress is large, BPD can proliferate through the process of linear motion and loop motion. The BPD density in the local area of the silicon carbide substrate can be quickly evaluated by the stress of the wafer.
[0008] The present invention has the following beneficial effects compared with the prior art: The present invention adopts an indirect testing method to convert BPD density detection into stress detection, while taking into account the advantages of non-destructive testing and low cost in traditional testing methods, thereby simplifying the testing process and improving the efficiency of experimental testing. BRIEF DESCRIPTION OF THE DRAWINGS
[0009] Figure 1 4 is a diagram showing stress detection results of a silicon carbide substrate in Example 1 of the present invention; Figure 2 : is a diagram showing the BPD density detection result of the silicon carbide substrate after etching in Example 1 of the present invention; Figure 3 The distribution diagram of the nine small areas selected in Example 1 of the present invention; Figure 4 Schematic diagram showing the trend correlation between the stress average value and the BPD density of 9 small areas in Example 1 of the present invention; Figure 5 This is a schematic diagram of the function fitting results provided in Example 1 of the present invention; Figure 6 A stress detection result diagram of a silicon carbide substrate and a position distribution diagram of nine selected small areas in Example 2 of the present invention; Figure 7 Graph showing the BPD density detection result of the silicon carbide substrate after etching in Example 2 of the present invention; Figure 8 The distribution diagram of the four small areas selected in Example 3 of the present invention; Fig. 9 A stress detection result diagram of a silicon carbide substrate and a position distribution diagram of five selected small areas in Example 4 of the present invention; Fig.10 4 is a diagram showing the BPD density detection result of the silicon carbide substrate after etching in Example 4 of the present invention; Fig.11 This is a distribution diagram of five small areas selected in Example 5 of the present invention. DETAILED DESCRIPTION The technical solutions in the embodiments of the present invention will be described clearly and completely below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments.
[0011] Terminology Note: Basal plane dislocation: BPD for short, is a dislocation defect in silicon carbide, which refers to the abnormal arrangement of atoms on a plane in the crystal lattice, in which some atoms are missing or replaced. Such dislocations are usually formed during crystal growth or processing, which may cause distortion of the crystal structure, generation of stress fields, and changes in the electrical properties of the crystal.
[0012] Stress: When a material is deformed, an equal but opposite reaction force is generated inside to resist the external force. The reaction force per unit area is defined as stress, with a dimension of Pa and a commonly used unit of MPa.
[0013] Thermal stress: stress caused by the constraint of thermal deformation due to temperature change without the action of external force.
[0014] Embodiment 1: Derivation of a method for rapidly evaluating the dislocation density of the basal plane of a silicon carbide substrate using stress (1) Perform stress detection on the silicon carbide substrate to be tested to obtain the stress condition of the substrate to be tested: The wafer stress tester is used to perform stress testing on the silicon carbide substrate to be tested. The test results are as follows: Figure 1 As shown in the test result graph, the whiter the color, the greater the stress, and the darker the color, the smaller the stress.
[0015] (2) Carry out molten alkali etching on the silicon carbide substrate: The substrate after stress detection in step (1) is corroded using a KOH etching furnace. The purity specification of the KOH used for etching is analytical grade (AR), the liquid surface temperature is 450°C, and the etching time is 45 minutes.
[0016] (3) After cleaning the silicon carbide substrate after etching in step (2), place it under a dislocation defect detector for detection to obtain the BPD density of the substrate to be tested: rinse the substrate surface with deionized water to remove residual KOH; use an ultrasonic cleaner to perform anhydrous ethanol ultrasonic cleaning on the substrate rinsed with deionized water for 30 minutes, and then dry and wipe the substrate after ultrasonic cleaning; use the dislocation defect detector to set the color block corresponding to the BPD density from small to large as follows: BPD density <1000 / cm 2 When the corresponding color block is blue, the BPD density is greater than 5000 / cm 2 The corresponding color block is red and is located at 5000 / cm 2 and 1000 / cm 2 The BPD density between the two gradually transitions from red to blue. The bluer the color, the smaller the BPD density, and the redder the color, the larger the BPD density. Figure 2 The BPD density test was performed on the etched silicon carbide substrate, and the test results are shown in Figure 2 shown.
[0017] Compare and analyze the test results of step (3) with the test results of step (1): Nine small areas were randomly selected on the silicon carbide substrate, and their positions are distributed as follows: Figure 3 As shown, the specific values of stress averages and BPD density of the nine small areas are listed in Table 1.
[0018] Table 1 Average stress values and BPD density data of 9 small areas
[0019] The trend diagram of the stress average value and BPD density of 9 small areas is shown in the figure below. Figure 4 As shown, it can be seen that the BPD density is positively correlated with the average stress value. The average stress value (MPa) is defined as X, and the BPD density (cm -2 ) is defined as Y. Through data fitting, the nonlinear curve fitting function relationship is obtained as follows (the fitting curve is as follows Figure 5 shown): ; This functional relationship is the nonlinear curve fitting relationship between the BPD density and the stress average value in the local area of the silicon carbide substrate. The fitting results are tested. The test results are as follows: Figure 5 As shown, it can be seen that the adjusted R 2 The value reaches 0.98253, which has a good fitting effect.
[0020] Embodiment 2: A method for rapidly evaluating the dislocation density of the basal plane of a silicon carbide substrate using stress Another silicon carbide substrate was selected and subjected to stress testing using a wafer stress tester. The test results are as follows: Figure 6 As shown, 9 small areas are randomly selected on it, and the location distribution diagram of the 9 small areas is as follows Figure 6 As shown, the average stress values of these nine small areas are listed in Table 2. Substituting the average stress values of the nine small areas into the functional relationship The BPD density assessment values of 9 small areas were obtained and the specific data are listed in Table 2.
[0021] The silicon carbide substrate was subjected to KOH molten alkali etching, the liquid surface temperature during etching was 450°C, and the etching time was 45 minutes; after etching, the substrate surface was rinsed with deionized water to remove residual KOH; an ultrasonic cleaning machine was used to perform anhydrous ethanol ultrasonic cleaning on the substrate after deionized water rinsing for 30 minutes, and then the ultrasonic substrate was dried and wiped; a dislocation defect detector was used to set the color block colors corresponding to the BPD density from small to large to the form in Example 1, and the BPD density of the etched silicon carbide substrate was detected, and the detection results were as follows: Figure 7 As shown, the measured values of BPD density of 9 small areas selected therefrom are statistically shown in Table 2. By comparing the BPD density assessment values and the measured values of BPD density of the 9 small areas, the applicability and accuracy of the above functional relationship can be proved.
[0022] Table 2 Average stress values and BPD density data of 9 small areas
[0023] Embodiment 3: A method for rapidly evaluating the dislocation density of the basal plane of a silicon carbide substrate using stress Four small areas were randomly selected again on the silicon carbide substrate in Example 2. The position distribution diagram of the four small areas is as follows: Figure 8 As shown, according to Figure 6 The test results show the average stress values of the four small areas in Table 3. Substituting the average stress values of the four small areas into the functional relationship , the BPD density evaluation values of the four small areas were obtained and the specific data are listed in Table 3. Figure 7 The test results show the measured values of BPD density in the four small areas in Table 3. By comparing the BPD density assessment values and the measured values of BPD density in the four small areas, the applicability and accuracy of the above functional relationship can be proved.
[0024] Table 3 Average stress values and BPD density data of four small areas
[0025] Embodiment 4: A method for rapidly evaluating the dislocation density of the basal plane of a silicon carbide substrate using stress Another silicon carbide substrate was selected and subjected to stress testing using a wafer stress tester. The test results are as follows: Fig. 9 As shown, 5 small areas are randomly selected on it, and the location distribution diagram of the 5 small areas is as follows Fig. 9 As shown, the average stress values of the five small areas are statistically listed in Table 4. Substituting the average stress values of the five small areas into the functional relationship The BPD density assessment values of 5 small areas were obtained and the specific data are listed in Table 4.
[0026] The silicon carbide sample was subjected to molten alkali etching, the liquid surface temperature during etching was 450°C, and the etching time was 45 minutes; after etching, the substrate surface was rinsed with deionized water to remove residual KOH; an ultrasonic cleaning machine was used to perform anhydrous ethanol ultrasonic cleaning on the substrate rinsed with deionized water for 30 minutes, and then the ultrasonic substrate was dried and wiped; a dislocation defect detector was used to set the color block colors corresponding to the BPD density from small to large to the form in Example 1, and the detection results are shown as follows: Fig.10 As shown, the measured values of BPD density of 5 small areas selected therefrom are statistically shown in Table 4. By comparing the BPD density assessment values and the measured values of BPD density of the 5 small areas, the applicability and accuracy of the above functional relationship can be proved.
[0027] Table 4 Average stress values and BPD density data of five small areas
[0028] Embodiment 5: A method for rapidly evaluating the dislocation density of the basal plane of a silicon carbide substrate using stress Five other small areas were randomly selected on the silicon carbide substrate in Example 4. The position distribution diagram of the five small areas is as follows: Fig.11 As shown, according to Fig. 9 The test results show the average stress values of the five small areas in Table 5. Substituting the average stress values of the five small areas into the functional relationship , the BPD density evaluation values of the five small areas were obtained and the specific data are listed in Table 5. Fig.10 The test results show the measured values of BPD density in these five small areas in Table 5. By comparing the BPD density assessment values and the measured values of BPD density in the five small areas, the applicability and accuracy of the above functional relationship can be proved.
[0029] Table 5 Average stress values and BPD density data of five small areas
[0030] By comparing the BPD density assessment value and the BPD density measured value, it can be found that although there is an error between the assessment value and the measured value, the error is within the error range of the above functional relationship (see Figure 5 ), and by adjusting the parameter size within the error range of each parameter, the evaluation value can be adjusted to be completely consistent with the measured value, thus proving the accuracy of the above functional relationship.
Claims
1. A method for rapidly evaluating the dislocation density of a silicon carbide substrate basal plane using stress, characterized in that: The following steps are involved: (1) Perform stress detection on the silicon carbide substrate to be tested, select any area, and obtain the average stress value (MPa) X of the substrate to be tested; (2) Evaluation of the BPD density (cm) in a local area of the SiC substrate -2 )Y, the functional relationship is as follows: 。 2. The method according to claim 1, characterized in that: In step (1), stress detection is performed on the silicon carbide substrate to be tested, and an arbitrary area is selected to obtain the average stress value of the substrate to be tested, including: using a wafer stress detector to perform stress detection on the silicon carbide substrate to be tested.
