Temperature compensation method and system, semiconductor preparation method and semiconductor equipment

Abstract

The invention provides a temperature compensation method and system, a semiconductor preparation method and semiconductor equipment. According to the temperature compensation method, the thickness of a pipe wall film is adopted as a training characteristic value of a temperature compensation model, the influence factor, namely the thickness of the pipe wall film, of the process temperature serves as an independent variable of the compensation temperature, and the high-precision compensation temperature can be obtained easily. Moreover, according to the semiconductor preparation method, before the furnace tube process is executed, the temperature compensation model is adopted to obtain the compensation temperature, the preset temperature is adjusted according to the compensation temperature, the thin film with the target thickness is obtained, and the thin film has the good uniformity. According to the method, accurate compensation of the preset temperature is completed before the furnace tube process is executed, temperature changes in the production process are not intervened, the risk of dead goods caused by program crash can be reduced, and the application cost is lower.

Description

Technical Field

[0001] The present invention relates to the field of semiconductor manufacturing technology, and in particular to a temperature compensation method and system, a semiconductor manufacturing method, and semiconductor equipment. Background Art

[0002] The semiconductor furnace process is one of the important process links in the integrated circuit manufacturing process. It is used to grow or process wafers under specific temperature and atmosphere to achieve high-quality processing and thin film deposition on the wafer surface. Figure 1 As shown, during the film formation process, the reaction gas will adhere to the inner tube wall of the furnace tube 10 and accumulate to form a tube wall film 100. When these tube wall films 100 reach a certain thickness, they will affect the temperature detection and temperature control accuracy of the machine thermocouple (Inner Thermo couple), resulting in the film forming temperature not being the preset temperature, seriously affecting the deposition rate of the thin film, the uniformity of the film layer, and the accuracy of the film thickness adjustment. Figure 2 As shown, affected by the tube wall film 100, the parameter test line A on the surface of the wafer after the furnace tube process is significantly deviated from the specification center line B, and the process accuracy Ca is 3.74%, and the process capability index Cpk is only 3.24. In this regard, the existing process is to compensate for the film thickness and its uniformity by adjusting the constant temperature setting value of the furnace zone, but the accuracy of this compensation method is limited. In addition, the existing process can also set the temperature slope through the temperature control model to form a counteracting thermal gradient on the wafer to correct the radial change of the wafer surface temperature and optimize the uniformity of the film layer. Alternatively, on the basis of forming the thermal gradient, the deposition process is divided into multiple shorter deposition processes to achieve high-precision temperature slope setting. However, the method of forming the counteracting thermal gradient not only requires real-time changes in settings, but also has a complex operation process, and the cost of the application system is high, and it is also prone to dead goods due to program crashes.

[0003] Therefore, a new compensation method and system are urgently needed to solve the above technical problems. Summary of the invention

[0004] The object of the present invention is to provide a temperature compensation method and system, a semiconductor preparation method, and a semiconductor device to solve at least one of the problems of how to improve the accuracy of temperature compensation for a furnace tube process, how to improve the control accuracy of film thickness based on a furnace tube process, and how to improve the uniformity of thin film deposition.

[0005] In order to solve the above technical problems, the present invention provides a temperature compensation method, which is applied to a furnace tube process. During the furnace tube process, a reaction gas is passed through the furnace tube, and the reaction gas forms a tube wall film on the tube wall of the furnace tube. The temperature compensation method includes: Providing a plurality of furnace tubes, and placing a plurality of test pieces in each of the furnace tubes; Collecting the initial thickness of the tube wall film of each furnace tube; Using each of the furnace tubes, performing a furnace tube process with a different preset temperature on each of the test pieces; and, Collecting the thickness of the tube wall film of each of the furnace tubes and the thickness of the thin film formed on each of the test pieces at the different preset temperatures; A temperature compensation model is formed by training at least the preset temperature, the initial thickness of the tube wall film, the thickness of the tube wall film and the thickness of the film as characteristic values.

[0006] Optionally, in the temperature compensation method, while using each of the furnace tubes to perform furnace tube processes at different preset temperatures on the respective test pieces, the furnace tube processes are also performed on the test pieces for different durations at each preset temperature.

[0007] Optionally, in the temperature compensation method, the thickness of the tube wall film includes the thickness of the tube wall film of the corresponding furnace tube after each execution of the furnace tube process.

[0008] Optionally, in the temperature compensation method, the process of training and forming a temperature compensation model by taking at least the preset temperature, the initial thickness of the tube wall film, the thickness of the tube wall film and the thickness of the film as characteristic values ​​comprises: According to the initial thickness of the tube wall film of each furnace tube, the different preset temperatures, the different time periods and the corresponding thickness of the tube wall film, the growth rate of the tube wall film at the different preset temperatures is obtained; and the difference between the thickness of the film and the target thickness is obtained; The temperature compensation model is formed by training using the preset temperature, the thickness of the tube wall film, the growth rate of the tube wall film at different preset temperatures, the thickness of the film and the difference as characteristic values.

[0009] Optionally, in the temperature compensation method, after acquiring the temperature compensation model, the temperature compensation method further includes: acquiring, according to the temperature compensation model, the compensation temperature of each furnace tube process executed corresponding to each furnace tube.

[0010] Optionally, in the temperature compensation method, after obtaining the compensation temperature of each furnace tube process executed by each furnace tube according to the temperature compensation model, the temperature compensation method further includes: The corresponding preset temperature is adjusted using the compensation temperature, and the adjusted preset temperature is used as the process temperature to perform the furnace tube process for different durations, so as to obtain the thickness of the tube wall film and the thickness of the thin film formed on the test piece accordingly; Determine whether the difference between the thickness of the film and the target thickness is within a threshold range; if so, the compensation temperature meets the compensation requirements; if not, use the adjusted preset temperature, the different time lengths, the initial thickness of the tube wall film, the thickness of the tube wall film and the thickness of the film as feature values ​​to train the temperature compensation model until the compensation temperature output by the temperature compensation model meets the compensation requirements.

[0011] Optionally, in the temperature compensation method, before using each of the furnace tubes to perform furnace tube processes with different preset temperatures on the respective test pieces, the temperature compensation method further includes: collecting an initial thickness of each of the test pieces; and, In the process of collecting the thickness of the tube wall film of each of the furnace tubes and the thickness of the thin film formed on each of the test pieces at the different preset temperatures, the thickness of the thin film is obtained based on the initial thickness of each of the test pieces and the thickness of each of the test pieces after each execution of the furnace tube process.

[0012] Based on the same inventive concept, the present invention also provides a temperature compensation system for executing the temperature compensation method, the temperature compensation system comprising: a data acquisition unit and a model building unit; The data acquisition unit is used to collect and store the thickness of the tube wall film of each furnace tube and the thickness of the film formed on each test piece at different preset temperatures; The model building unit is used to train and form a temperature compensation model by taking at least the preset temperature, the initial thickness of the tube wall film, the thickness of the tube wall film and the thickness of the film as characteristic values.

[0013] Based on the same inventive concept, the present invention also provides a semiconductor manufacturing method, comprising: providing a wafer; Acquire compensation temperature by using the temperature compensation model obtained by the temperature compensation method; The preset temperature is adjusted using the compensation temperature to serve as the process temperature; A furnace process is performed on the wafer at the process temperature.

[0014] Based on the same inventive concept, the present invention also provides a semiconductor device including the temperature compensation system.

[0015] In summary, the present invention provides a temperature compensation method and system, a semiconductor preparation method, and a semiconductor device. Among them, the temperature compensation method provided by the present invention collects the thickness of the tube wall film of each of the furnace tubes and the thickness of the film formed on each of the test pieces at the different preset temperatures; and then uses at least the preset temperature, the initial thickness of the tube wall film, the thickness of the tube wall film, and the thickness of the film as characteristic values ​​to train and form a temperature compensation model, and the compensation temperature can be obtained according to the temperature compensation model. Compared with the prior art, the training characteristic values ​​of the temperature compensation model include the thickness of the tube wall film, which is intended to use the thickness of the tube wall film, an influencing factor of the process temperature, as an independent variable of the compensation temperature, which is conducive to obtaining a better accuracy compensation temperature and achieving the control accuracy of the film thickness of the furnace tube process.

[0016] The semiconductor preparation method provided by the present invention is to use the temperature compensation model to obtain the compensation temperature before executing the furnace tube process, and adjust the preset temperature accordingly to obtain a film of target thickness, and the film has better uniformity. Among them, the present invention completes the accurate compensation of the preset temperature before executing the furnace tube process, does not interfere with the temperature change in the production process, can reduce the risk of dead goods caused by program crashes, and has lower application costs. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Those skilled in the art will appreciate that the drawings are provided for a better understanding of the present invention, but do not constitute any limitation on the scope of the present invention.

[0018] Figure 1 It is a structural schematic diagram of a furnace tube machine in the prior art.

[0019] Figure 2 It is a schematic diagram of wafer quality inspection after performing a furnace tube process without temperature compensation in the prior art.

[0020] Figure 3 4 is a flow chart of a temperature compensation method in an embodiment of the present invention.

[0021] Figure 4 Schematic diagram of the structure of the temperature compensation system in the embodiment of the present invention.

[0022] Figure 5 It is a flow chart of a semiconductor manufacturing method in an embodiment of the present invention.

[0023] Figure 6 It is a schematic diagram of the process nodes of the semiconductor manufacturing method in an embodiment of the present invention.

[0024] Figure 7 It is a schematic diagram of wafer quality detection after executing a semiconductor manufacturing process in an embodiment of the present invention.

[0025] And, in the attached drawings: 10-furnace tube; 100-tube wall film; 20-data acquisition unit; 21-model building unit; 22-judgment processing unit; A-parameter test line; B-specification center line. DETAILED DESCRIPTION

[0026] In order to make the purpose, advantages and features of the present invention clearer, the present invention is further described in detail below in conjunction with the accompanying drawings and specific embodiments. It should be noted that the drawings are all in a very simplified form and are not drawn to scale, and are only used to conveniently and clearly assist in explaining the purpose of the embodiments of the present invention. In addition, the structure shown in the drawings is often a part of the actual structure. In particular, the emphasis of each drawing is different, and sometimes different proportions are used. It should also be understood that, unless otherwise specified or indicated, the terms "first", "second", "third", etc. in the specification are only used to distinguish between the various components, elements, steps, etc. in the specification, and are not used to represent the logical relationship or sequential relationship between the various components, elements, steps, etc.

[0027] During the furnace tube process, the reaction gas introduced into the furnace tube will adhere to the tube wall of the furnace tube and form a tube wall film. The thickness of the tube wall film will affect the detection accuracy of the reaction temperature in the furnace tube, and when the preset temperature is greatly different from the actual reaction temperature in the furnace tube, the film growth rate on the wafer surface will be too high or too low, and the film cannot be accurately adjusted.

[0028] Therefore, please refer to Figure 3 This embodiment provides a temperature compensation method for compensating the preset temperature of the furnace process, thereby compensating the film growth rate on the wafer surface, so that the film thickness meets the target thickness and also meets the film uniformity requirements. The temperature compensation method includes: Step 1 S11: providing a plurality of furnace tubes, and placing a plurality of test pieces in each of the furnace tubes; Step 2 S12: collecting the initial thickness of the tube wall film of each furnace tube; Step 3 S13: using each of the furnace tubes to perform furnace tube processes with different preset temperatures on the respective test pieces; and Step 4 S14: collecting the thickness of the tube wall film of each of the furnace tubes and the thickness of the thin film formed on each of the test pieces at the different preset temperatures; Step five S15: training a temperature compensation model by taking at least the preset temperature, the initial thickness of the tube wall film, the thickness of the tube wall film and the thickness of the film as characteristic values.

[0029] It can be seen that the temperature compensation method provided in this embodiment uses the thickness of the tube wall film as training data when training the temperature compensation model, and intends to use the thickness of the tube wall film, an influencing factor of the process temperature, as an independent variable of the compensation temperature, which is conducive to obtaining a compensation temperature with better accuracy and achieving the control accuracy of the film thickness of the furnace tube process.

[0030] The following is combined with Figures 3 to 7 , specifically illustrate the temperature compensation method provided in this embodiment.

[0031] Step 1 S11: providing a plurality of furnace tubes, and placing a plurality of test pieces in each of the furnace tubes.

[0032] Step 2 S12: collecting the initial thickness of the tube wall film of each furnace tube.

[0033] As can be seen from the above, based on the characteristics of the furnace tube process, it is inevitable that a tube wall film will be deposited on the side wall of the furnace tube during the process. Although the tube wall film will be cleaned regularly during the preventive maintenance (PM) of the equipment, the tube wall film will continue to be formed in the actual process. These tube wall films will interfere with the temperature detection of the furnace tube and affect the temperature control accuracy during the process. Therefore, in order to accurately obtain the degree of influence of the thickness of the tube wall film on the preset temperature, this embodiment uses multiple furnace tubes for data collection, and pre-detects the thickness of the initial tube wall film in each of the furnace tubes. Preferably, the thickness of the initial tube wall film in each of the furnace tubes is different, or partially the same. Exemplarily, the thickness of the initial tube wall film of five furnace tubes is collected, and they are 5nm, 10nm, 20nm, 18nm and 35nm respectively.

[0034] And, the test piece is a wafer specifically used for process detection. And the wafer includes but is not limited to a silicon-on-insulator (SOI) substrate, a bulk silicon substrate, a germanium substrate, a germanium silicon substrate, an indium phosphide (InP) substrate, a gallium arsenide (GaAs) substrate or a germanium-on-insulator substrate. Among them, this embodiment does not limit the specific number of the test pieces placed in each of the furnace tubes during each execution of the furnace tube process. Exemplarily, before each execution of the furnace tube process, three test pieces are placed in each of the furnace tubes. Preferably, before executing the subsequent furnace tube process, it is necessary to pre-measure the initial thickness of each test piece to facilitate the detection of the thickness of the thin film growth of each test piece after the subsequent furnace tube process is implemented.

[0035] Step three S13: using each of the furnace tubes to perform furnace tube processes with different preset temperatures on each of the test pieces.

[0036] Step 4 S14: collecting the thickness of the tube wall film of each of the furnace tubes and the thickness of the thin film formed on each of the test pieces at the different preset temperatures.

[0037] Specifically, before executing the step 3 S13, a plurality of different preset temperatures and other process parameters are set. For example, six preset temperatures of 600°C, 700°C, 800°C, 900°C, 1000°C and 1100°C are set, and in order to follow the single variable principle, other process parameters of each furnace tube are the same, such as the flow rate and flow velocity of the reaction gas introduced into each furnace tube, the type of reaction gas, and the vacuum environment pressure.

[0038] After setting the process parameters, the furnace tube processes of different preset temperatures are respectively performed on the test pieces in each of the furnace tubes. And, at each of the preset temperatures, the furnace tube processes of different durations are performed on the test pieces. Exemplarily, when the preset temperature is 600°C, each of the furnace tubes performs process times of 15min, 30min, 45min, 60min, 75min and 90min, and after each process time is completed, the thickness of the tube wall film corresponding to each of the furnace tubes and the thickness of the film formed on the corresponding test piece are collected. Similarly, when the preset temperature is 700°C, each of the furnace tubes performs process times of 15min, 30min, 45min, 60min, 75min and 90min, and after each process time is completed, the thickness of the tube wall film corresponding to each of the furnace tubes and the thickness of the film formed on the corresponding test piece are collected. Similarly, the furnace tube processes of different durations corresponding to the preset temperatures of 800°C, 900°C, 1000°C and 1100°C are completed respectively, and the thickness of the tube wall film corresponding to each of the furnace tubes under the corresponding temperature and duration conditions and the thickness of the thin film formed on the corresponding test piece are obtained.

[0039] Based on this, each furnace tube can collect the data collection table of tube wall film thickness and film thickness as shown in Table 1 below. i is the thickness of the tube wall film after each furnace tube process, D i is the thickness of the thin film formed on the test piece after each furnace tube process, and i is the sequence number of the process performed by each furnace tube.

[0040] Table 1 Data collection table of tube wall film thickness and film thickness

[0041] It should be noted that the thickness of the thin film formed on the test piece is the difference between the thickness of the test piece after the furnace tube process and the initial thickness of the test piece.

[0042] Step five S15: training a temperature compensation model by taking at least the preset temperature, the initial thickness of the tube wall film, the thickness of the tube wall film and the thickness of the film as characteristic values.

[0043] In the furnace tube process, due to the existence of the tube wall film, there is a gap between the preset temperature and the actual reaction temperature, which in turn affects the film deposition rate, resulting in the thickness of the film formed on the test piece being difficult to meet the target thickness. Therefore, the tube wall film thickness, preset temperature and process time are all factors that affect the film thickness. In order to obtain the relationship between the influence of these three factors on the film thickness, it is preferred to first obtain the relationship between the thickness of the tube wall film and the preset temperature and process time, that is, the growth rate of the tube wall film at different preset temperatures. Specifically, according to the data collected in Table 1, the growth rate of the tube wall film at different preset temperatures is calculated, and the calculation formula is: R i =(H i -H i-1 ) / T i Among them, R i is the growth rate of the tube wall film calculated based on the i-th furnace tube process data; H i is the thickness of the tube wall film after the i-th furnace tube process; H i-1 is the thickness of the tube wall film after the i-1th furnace tube process; T i is the process time of executing the i-th furnace tube process. For example, when the preset temperature shown in Table 1 is 600°C and the process time is 15min, R1=(H1-H0) / 15; wherein H0 is the thickness of the initial tube wall film.

[0044] Furthermore, according to the growth rate of the tube wall film at different preset temperatures, it can be known that the preset temperature has a direct impact on the growth rate of the tube wall film, and within a certain range, the higher the preset temperature, the greater the growth rate of the tube wall film. That is, the thicker the growth thickness of the tube wall film per unit time. However, the applicant has found through research that the thicker the thickness of the tube wall film, the more serious the interference with the reaction temperature, which leads to a large difference between the thickness of the film grown on the test piece and the target thickness. Therefore, after obtaining the growth rate of the tube wall film at different preset temperatures, the preset temperature, the thickness of the tube wall film, the growth rate of the tube wall film at different preset temperatures, the thickness of the film, and the difference between the thickness of the film and the target thickness are used as feature values ​​to train and establish a temperature compensation model.

[0045] It can be understood that the temperature compensation model is used to obtain the thickness of the tube wall film in different time periods at the set temperature based on the growth rate of the tube wall film; then based on the thickness of the tube wall film and the set temperature, the difference between the thickness of the film under the corresponding conditions and the target thickness is obtained, and then according to the difference, combined with the characteristic values ​​such as the preset temperature, the process time and the thickness range of the tube wall film, the deviation value of the preset temperature, that is, the compensation temperature, is reversely calculated. Preferably, the temperature compensation model can be trained based on models such as neural networks, Kriging interpolation, and linear regression.

[0046] Furthermore, after obtaining the temperature compensation model, the compensation temperature of each furnace tube process is calculated according to the data collected in step 4 S14. Based on this, each furnace tube process is re-executed with each compensation temperature obtained, and the growth thickness of the film of the test piece and the thickness of the tube wall film of the furnace tube are collected after each furnace tube process. And, it is determined whether the difference between the thickness of the film and the target thickness is within the threshold range; if so, the compensation temperature meets the compensation requirements; if not, the temperature compensation model is trained with the adjusted preset temperature, the different time lengths, the initial thickness of the tube wall film, the thickness of the tube wall film and the thickness of the film as feature values ​​until the compensation temperature output by the temperature compensation model meets the compensation requirements.

[0047] For example, when the preset temperature is 600°C, the process time is 15 minutes, the thickness of the tube wall film is 50nm, the thickness of the thin film grown on the test piece is 100nm, and it is 20nm less than the target thickness, the temperature compensation model is calculated and a compensation temperature of 80°C is obtained. Then, the compensation temperature is added to the original preset temperature as the new preset temperature, that is, 680°C. Under the same other process conditions, the furnace tube process is re-executed for 15 minutes, and the thickness of the tube wall film can be obtained to be 55nm, and the thickness of the thin film grown on the test piece is 115nm, and the difference with the target thickness is reduced. Based on this, the compensated characteristic values, such as the preset temperature of 680°C, the thickness of the tube wall film of 55nm and the thickness of the thin film of 115nm, can be used as a set of training data and added to the test set to continue training and optimizing the temperature compensation model until the compensation temperature output by the temperature compensation model makes the difference between the compensated thickness of the thin film and the target thickness within the threshold range, so as to improve the accuracy of the temperature compensation model in obtaining the compensation temperature.

[0048] After the temperature compensation model is trained, the temperature compensation model can be used to perform temperature compensation on the furnace tube process to alleviate the influence of the tube wall film on process temperature detection and optimize the film growth rate and uniformity.

[0049] Based on the same concept, this embodiment also provides a temperature compensation system for executing the above temperature compensation method. Figure 4 The temperature compensation system includes: a data acquisition unit 20 and a model building unit 21; the data acquisition unit 20 is used to collect and store the thickness of the tube wall film of each furnace tube and the thickness of the film formed on each test piece at different preset temperatures; and to collect the initial thickness of each test piece. The model building unit 21 is used to obtain the growth rate of the tube wall film at different preset temperatures according to the initial thickness of the tube wall film of each furnace tube, the different preset temperatures, the different time lengths and the corresponding thickness of the tube wall film; and to obtain the difference between the thickness of the film and the target thickness; the preset temperature, the thickness of the tube wall film, the growth rate of the tube wall film at different preset temperatures, the thickness of the film and the difference are used as feature values ​​to train and form the temperature compensation model.

[0050] Preferably, the temperature compensation system further includes a judgment processing unit 22. The judgment processing unit 22 is used to judge whether the difference between the thickness of the film and the target thickness is within a threshold range; if so, the compensation temperature meets the compensation requirements; if not, the model building unit 21 is used to train the temperature compensation model with the adjusted preset temperature, the different time lengths, the initial thickness of the tube wall film, the thickness of the tube wall film and the thickness of the film as feature values ​​until the compensation temperature output by the temperature compensation model meets the compensation requirements.

[0051] Based on the same concept, this embodiment also provides a semiconductor manufacturing process. Figure 5 and Figure 6 , the semiconductor preparation process comprises: Step 1 S21: providing a wafer; Step 2 S22: using the temperature compensation model obtained by the temperature compensation method to obtain compensation temperature; Step 3 S23: using the compensation temperature to adjust the preset temperature as the process temperature; Step 4 S24: performing a furnace tube process on the wafer at the process temperature.

[0052] It is understandable that the semiconductor preparation process adopts a furnace tube process to implement thin film deposition, and before executing the furnace tube process, it is necessary to use an advanced process control module (Advanced Process Control, APC) to configure the process parameters. After APC pre-adjusts the various process parameters, the temperature compensation model obtained by the above-mentioned temperature compensation method is used to obtain the compensation temperature of the corresponding furnace tube process, and the preset temperature is adjusted with the compensation temperature. After the temperature compensation adjustment is completed, the furnace tube process is executed again, which can not only improve the film growth rate and obtain better process effects, but also does not interfere with temperature changes in the production process, can reduce the risk of dead goods caused by program crashes, and has lower application costs. Based on this, the applicant adopts the semiconductor preparation process to deposit a thin film and obtains Figure 7 The test results shown in the figure show that the process accuracy Ca is 1.62% and the process capability index Cpk is 8.45. Figure 2 The deviation of the parameter test line A on the surface of the wafer prepared by the semiconductor preparation process provided by this embodiment is significantly reduced compared with the specification center line B, and the process effect is effectively improved.

[0053] Based on the same concept, this embodiment also provides a semiconductor device, including the temperature compensation system.

[0054] In summary, the present embodiment provides a temperature compensation method and system, a semiconductor preparation method, and a semiconductor device. Among them, the temperature compensation method uses the thickness of the tube wall film as the training feature value of the temperature compensation model, and intends to use the thickness of the tube wall film, an influencing factor of the process temperature, as the independent variable of the compensation temperature, which is conducive to obtaining a compensation temperature with better accuracy. And, the semiconductor preparation method is to use the temperature compensation model to obtain the compensation temperature before executing the furnace tube process, and adjust the preset temperature accordingly to obtain a film of target thickness, and the film has better uniformity. Among them, the present invention completes the precise compensation of the preset temperature before executing the furnace tube process, does not interfere with the temperature changes in the production process, can reduce the risk of dead goods caused by program crashes, and has lower application costs.

[0055] In addition, it should be recognized that although the present invention has been disclosed as a preferred embodiment, the above embodiment is not intended to limit the present invention. For any technician familiar with the art, without departing from the scope of the technical solution of the present invention, the technical content disclosed above can be used to make many possible changes and modifications to the technical solution of the present invention, or modified into equivalent embodiments of equivalent changes. Therefore, any simple modification, equivalent change and modification made to the above embodiment according to the technical essence of the present invention without departing from the content of the technical solution of the present invention still belongs to the scope of protection of the technical solution of the present invention.

Claims

1. A temperature compensation method is applied to a furnace tube process, wherein a reaction gas is passed through the furnace tube during the furnace tube process, and the reaction gas forms a tube wall film on the tube wall of the furnace tube, characterized in that: The temperature compensation method comprises: Providing a plurality of furnace tubes, and placing a plurality of test pieces in each of the furnace tubes; Collecting the initial thickness of the tube wall film of each furnace tube; Using each of the furnace tubes, performing a furnace tube process with a different preset temperature on each of the test pieces; and, Collecting the thickness of the tube wall film of each of the furnace tubes and the thickness of the thin film formed on each of the test pieces at the different preset temperatures; A temperature compensation model is formed by training at least the preset temperature, the initial thickness of the tube wall film, the thickness of the tube wall film and the thickness of the film as characteristic values.

2. The temperature compensation method according to claim 1, characterized in that: In the process of using each of the furnace tubes to perform furnace tube processes at different preset temperatures on the respective test pieces, the furnace tube processes are also performed on the test pieces at each preset temperature for different durations.

3. The temperature compensation method according to claim 2, characterized in that: The thickness of the tube wall film includes the thickness of the tube wall film of the corresponding furnace tube after the furnace tube process is performed each time.

4. The temperature compensation method according to claim 2 or 3, characterized in that: The process of training and forming a temperature compensation model by taking at least the preset temperature, the initial thickness of the tube wall film, the thickness of the tube wall film and the thickness of the film as characteristic values ​​comprises: According to the initial thickness of the tube wall film of each furnace tube, the different preset temperatures, the different time periods and the corresponding thickness of the tube wall film, the growth rate of the tube wall film at the different preset temperatures is obtained; and the difference between the thickness of the film and the target thickness is obtained; The temperature compensation model is formed by training using the preset temperature, the thickness of the tube wall film, the growth rate of the tube wall film at different preset temperatures, the thickness of the film and the difference as characteristic values.

5. The temperature compensation method according to claim 1, characterized in that: After acquiring the temperature compensation model, the temperature compensation method further comprises: acquiring, according to the temperature compensation model, a compensation temperature of each furnace tube process executed corresponding to each furnace tube.

6. The temperature compensation method according to claim 5, characterized in that: After obtaining the compensation temperature of each furnace tube process executed by each furnace tube according to the temperature compensation model, the temperature compensation method further includes: The corresponding preset temperature is adjusted using the compensation temperature, and the adjusted preset temperature is used as the process temperature to perform the furnace tube process for different durations, so as to obtain the thickness of the tube wall film and the thickness of the thin film formed on the test piece accordingly; Determine whether the difference between the thickness of the film and the target thickness is within a threshold range; if so, the compensation temperature meets the compensation requirements; if not, use the adjusted preset temperature, the different time lengths, the initial thickness of the tube wall film, the thickness of the tube wall film and the thickness of the film as feature values ​​to train the temperature compensation model until the compensation temperature output by the temperature compensation model meets the compensation requirements.

7. The temperature compensation method according to claim 1, characterized in that: Before using each of the furnace tubes to perform furnace tube processes with different preset temperatures on the respective test pieces, the temperature compensation method further includes: collecting the initial thickness of each of the test pieces; and, In the process of collecting the thickness of the tube wall film of each of the furnace tubes and the thickness of the thin film formed on each of the test pieces at the different preset temperatures, the thickness of the thin film is obtained based on the initial thickness of each of the test pieces and the thickness of each of the test pieces after each execution of the furnace tube process.

8. A temperature compensation system, characterized in that: Used to perform the temperature compensation method according to any one of claims 1 to 7, the temperature compensation system comprises: a data acquisition unit and a model building unit; The data acquisition unit is used to collect and store the thickness of the tube wall film of each furnace tube and the thickness of the film formed on each test piece at different preset temperatures; The model building unit is used to train and form a temperature compensation model by taking at least the preset temperature, the initial thickness of the tube wall film, the thickness of the tube wall film and the thickness of the film as characteristic values.

9. A semiconductor manufacturing method, characterized in that: include: providing a wafer; Acquire compensation temperature using the temperature compensation model obtained by the temperature compensation method according to any one of claims 1 to 7; The preset temperature is adjusted using the compensation temperature to serve as the process temperature; A furnace process is performed on the wafer at the process temperature.

10. A semiconductor device, characterized in that: Comprising a temperature compensation system as claimed in claim 8.

Images