Reaction cavity radiation temperature measurement correction method and system, temperature measurement device, computer equipment and program product

Abstract

The application provides a reaction cavity radiation thermometry correction method and system, a temperature measuring device, computer equipment and a program product, and relates to the technical field of semiconductor detection. The application performs time domain alignment processing on the radiation signal and reflectivity signal of the wafer in the reaction cavity. Then, based on the temperature estimation result corresponding to the radiation signal, the delay variation law is determined. The reflectivity signal is dynamically compensated according to the delay variation law. The model parameters of the emissivity correction model are iteratively optimized according to the compensated reflectivity signal and the temperature estimation result. The radiation signal is corrected based on the optimized emissivity correction model. Therefore, the emissivity mismatch problem caused by the dynamic change of the delay of the reflectivity signal relative to the radiation signal with the change of temperature can be compensated. The temperature fluctuation caused by the dynamic misalignment of the signal and the deviation of the model parameters in the reflectivity correction process is reduced. The accuracy and stability of the radiation thermometry result in the reaction cavity are improved.

Description

Technical Field

[0001] This application relates to the field of semiconductor testing technology, and in particular to a method, system, temperature measuring device, computer equipment and program product for correcting radiation temperature measurement in a reaction chamber. Background Technology

[0002] In-cavity radiation thermometry typically involves acquiring the radiation signal from the wafer in the measurement wavelength range and retrieving the wafer temperature based on a calibration relationship between the radiation signal and temperature. However, the actual radiation intensity of the wafer is not only temperature-dependent but also affected by the wafer surface emissivity. Since the wafer's emissivity varies due to factors such as temperature, material state, surface roughness, and thin film deposition state, the corresponding radiation signal at the same temperature can change, affecting the accuracy of temperature retrieval. Therefore, existing technologies usually further acquire the wafer's reflectivity signal and use it to correct the emissivity of the radiation signal to improve temperature measurement accuracy.

[0003] Existing reflectivity correction methods typically assume that the reflectivity signal and the radiation signal are synchronized in time, or simply compensate for the time difference between them as a fixed delay. However, in actual reaction chamber operation, as the wafer temperature changes, the target surface state, material spectral characteristics, reflection response process, and the effective wavelength of the temperature measurement optical path may all change. This means that the delay of the reflectivity signal relative to the radiation signal is not a fixed value, but exhibits a dynamic characteristic that varies with temperature. This results in dynamic misalignment between the reflectivity signal and the radiation signal in different temperature ranges, thereby affecting the accuracy of reflectivity correction and the stability of temperature measurement. Summary of the Invention

[0004] The technical problem solved by this application is to provide a method, system, temperature measuring device, computer equipment and program product for radiation temperature measurement correction of a reaction cavity, so as to overcome the defects of the prior art that does not consider the dynamic change of the delay of the reflectivity signal relative to the radiation signal with temperature, resulting in low emissivity correction accuracy and poor stability of temperature measurement results.

[0005] In a first aspect, this application proposes a method for correcting radiation temperature measurement in a reaction cavity, the method comprising: Acquire the radiation and reflectivity signals of the wafer within the reaction chamber; The temporal offset relationship between the radiation signal and the reflectivity signal is determined based on the temporal offset relationship, and the reflectivity signal is aligned accordingly. The delay variation law is determined based on the temperature estimation result corresponding to the radiation signal, and the reflectivity signal after alignment is dynamically compensated according to the delay variation law to obtain the compensated reflectivity signal; the delay variation law is used to characterize the correspondence between the delay amount and temperature. Based on the compensated reflectivity signal and the temperature estimation result, the model parameters of the pre-built emissivity correction model are iteratively optimized; the emissivity correction model is used to characterize the correlation between the temperature estimation result, the reflectivity signal, and the emissivity. The radiation signal is corrected using an emissivity correction model with optimized model parameters to obtain the corrected radiation signal. The corrected temperature of the wafer is obtained based on the corrected radiation signal.

[0006] In one embodiment, the method further includes: The acquired radiation signal and reflectivity signal are preprocessed, and the time-domain offset relationship is determined based on the preprocessed radiation signal and reflectivity signal. The preprocessing methods include filtering and baseline correction. In one embodiment, determining the time-domain offset relationship between the radiation signal and the reflectivity signal, and aligning the reflectivity signal according to the time-domain offset relationship, includes: Correlation analysis was performed on the radiation signal and the reflectivity signal; Based on the correlation analysis results, the corresponding delay between the radiation signal and the reflectivity signal is determined, and the corresponding delay is used as the time-domain offset relationship. The reflectivity signal is time-shifted according to the corresponding delay amount to obtain the aligned reflectivity signal.

[0007] In one embodiment, the correlation analysis includes constructing a cross-correlation function between the radiation signal and the reflectivity signal, and determining the degree of correlation between the radiation signal and the reflectivity signal based on the function values ​​of the cross-correlation function under different delay conditions to obtain the cross-correlation analysis result; the corresponding delay amount is the delay amount corresponding to the cross-correlation function reaching its peak value.

[0008] In one embodiment, determining the delay variation law based on the temperature estimation result corresponding to the radiation signal includes: Based on the radiation signal, a temperature estimation sequence for the wafer at multiple sampling times is determined; Based on the temperature range corresponding to the temperature estimation sequence, the radiation signal and the reflectivity signal are divided into multiple segmented signals; For each segmented signal, determine the segment delay between the radiation signal and the reflectivity signal in that segmented signal; Linearly fit the segment temperature and segment delay corresponding to each segment signal to establish the delay variation law of the delay with temperature; the segment temperature is the temperature estimate within the temperature range of the segment signal.

[0009] In one embodiment, the step of dynamically compensating the aligned reflectivity signal according to the delay variation law to obtain a compensated reflectivity signal includes: Based on the temperature estimate of the wafer at the current sampling moment and the delay variation pattern, determine the dynamic delay amount corresponding to the current sampling moment; Based on the dynamic delay, determine the reflectance sampling position corresponding to the current sampling moment; Obtain the reflectance value corresponding to the reflectance sampling position from the aligned reflectance signal; Based on the reflectance values ​​corresponding to multiple sampling times, a compensated reflectance signal is generated. In one embodiment, the reflectance sampling position is a non-integer sampling position; The step of obtaining the reflectance value corresponding to the reflectance sampling position from the aligned reflectance signal includes: The reflectivity signal after alignment is interpolated to obtain the reflectivity value corresponding to the non-integer sampling position. In one embodiment, the model parameters include reflectivity correction coefficients and temperature-related correction coefficients; The step of iteratively optimizing the model parameters of the pre-built emissivity correction model based on the compensated reflectivity signal and the temperature estimation result includes: The current emissivity is determined based on the current reflectivity correction coefficient, the current temperature-related correction coefficient, the compensated reflectivity signal, and the temperature estimation result. Based on the current emissivity and the radiation signal, determine the current temperature sequence, and based on the current temperature sequence, determine the temperature fluctuation index; Based on the temperature fluctuation index, iteratively update the current reflectivity correction coefficient and the current temperature correlation correction coefficient; The current reflectivity correction coefficient and the current temperature correlation correction coefficient when the temperature fluctuation index meets the preset convergence condition are determined as the model parameters.

[0010] In one embodiment, the emissivity correction model is expressed as:

[0011] in, This represents the emissivity at sampling time t. This represents the compensated reflectivity signal corresponding to sampling time t. This represents the temperature estimation result corresponding to sampling time t. This represents the reflectivity correction factor. This represents the temperature-related correction factor.

[0012] In one embodiment, the emissivity correction model with optimized model parameters is used to correct the radiation signal to obtain a corrected radiation signal, including: Based on the determined model parameters, the temperature estimation results, and the compensated reflectivity signal, the emissivity corresponding to the wafer is determined; The effective radiation signal is obtained by subtracting the background radiation signal from the radiation signal. The effective radiation signal is normalized and corrected based on the emissivity to obtain a corrected radiation signal, which satisfies the following expression:

[0013] in, This represents the corrected radiation signal. Indicates a radiated signal. Indicates background radiation signal, Indicates the emissivity.

[0014] In one embodiment, obtaining the corrected temperature of the wafer based on the corrected radiation signal includes: Based on the pre-established blackbody calibration relationship, the corrected radiation signal is converted into a wafer-corrected temperature.

[0015] In one embodiment, the method further includes: The temperature after wafer correction is used as the temperature estimate result after the radiation signal is updated. Based on the updated temperature estimation results, the dynamic compensation is re-executed and the model parameters of the emissivity correction model are determined, and the wafer-corrected temperature is updated based on the re-determined model parameters; If the preset update conditions are met, the update is stopped, and the temperature at which the update stops is taken as the final temperature of the wafer.

[0016] In one embodiment, the method further includes: Determine whether temperature-related drift exists based on the degree of change in the aforementioned delayed change pattern; In the presence of temperature-dependent drift, the aligned reflectivity signal is dynamically compensated; otherwise, the aligned reflectivity signal and the temperature estimation result are used to iteratively optimize the model parameters of the pre-built emissivity correction model.

[0017] Secondly, this application proposes a reaction cavity radiation temperature measurement correction system, the system comprising: The acquisition module is used to acquire the radiation and reflectivity signals of the wafer inside the reaction chamber; The delay detection module is used to determine the time-domain offset relationship between the radiation signal and the reflectivity signal, and to perform alignment processing on the reflectivity signal according to the time-domain offset relationship; The compensation module is used to determine the delay variation law based on the temperature estimation result corresponding to the radiation signal, and to dynamically compensate the aligned reflectivity signal according to the delay variation law to obtain the compensated reflectivity signal; the delay variation law is used to characterize the correspondence between the delay amount and temperature. An optimization module is used to iteratively optimize the model parameters of a pre-built emissivity correction model based on the compensated reflectivity signal and the temperature estimation result; the emissivity correction model is used to characterize the correlation between the temperature estimation result, the reflectivity signal, and the emissivity. The correction module is used to correct the radiation signal using an emissivity correction model with optimized model parameters to obtain a corrected radiation signal; and to obtain the corrected temperature of the wafer based on the corrected radiation signal.

[0018] Thirdly, this application proposes a temperature measuring device, the device comprising: The light source module is used to generate detection light and irradiate the wafer surface inside the reaction chamber with the detection light; An optical module is used to receive thermal radiation from the surface of the wafer and reflected light formed on the surface of the wafer based on the detection light, and to perform optical path separation between the thermal radiation and the reflected light; The acquisition module is used to sample the thermal radiation and the reflected light after optical path separation, and convert them into a first electrical signal and a second electrical signal, respectively; The signal conditioning module is used to process the first electrical signal and the second electrical signal respectively to obtain the radiation signal and the reflectivity signal; The reaction cavity radiation temperature measurement and correction system as described in the second aspect is used to obtain the corrected temperature of the wafer based on the radiation signal and the reflectivity signal.

[0019] Fourthly, this application also provides a computer device. The computer device includes a memory and a processor, the memory storing a computer program, and the processor executing the computer program to implement the method of the first aspect.

[0020] Fifthly, this application also provides a computer program product, including a computer program that, when executed by a processor, implements the method in the first aspect.

[0021] The above-mentioned reaction chamber radiation temperature measurement correction method, system, temperature measuring device, computer equipment and program products have at least the following advantages: This application first performs time-domain alignment processing on the radiation and reflectivity signals of the wafer within the reaction cavity. Then, based on the temperature estimation results corresponding to the radiation signal, it determines the delay variation law of the reflectivity signal relative to the radiation signal and dynamically compensates the reflectivity signal according to the delay variation law. Based on the compensated reflectivity signal and temperature estimation results, iteratively optimizes the model parameters of the emissivity correction model and corrects the radiation signal based on the optimized emissivity correction model. This can compensate for the emissivity mismatch problem caused by the dynamic change of the delay of the reflectivity signal relative to the radiation signal with temperature, and enable the emissivity correction process to adapt to changes in wafer temperature and radiation characteristics. It reduces temperature fluctuations caused by dynamic signal misalignment and model parameter deviations during the reflectivity correction process, thereby improving the accuracy and stability of the radiation temperature measurement results within the reaction cavity. Attached Figure Description

[0022] Figure 1 This is a structural block diagram of the temperature measuring device in one embodiment; Figure 2 This is a schematic flowchart of a reaction chamber radiation temperature measurement correction method in one embodiment; Figure 3 This is a structural block diagram of a reaction chamber radiation temperature measurement correction system in one embodiment; Figure 4 This is an internal structural diagram of a computer device in one embodiment. Detailed Implementation

[0023] The following specific examples illustrate the implementation of this application. Those skilled in the art can easily understand other advantages and effects of this application from the content disclosed in this specification. This application can also be implemented or applied through other different specific embodiments, and various details in this specification can also be modified or changed based on different viewpoints and applications without departing from the spirit of this application. It should be noted that, unless otherwise specified, the following embodiments and features in the embodiments can be combined with each other.

[0024] Some exemplary embodiments of this application have been described for illustrative purposes. It should be understood that this application may be implemented in other ways not specifically shown in the accompanying drawings.

[0025] Please see Figure 1 In one exemplary embodiment, this application provides a temperature measuring device, including: a light source module, an optical module, a data acquisition module, a signal conditioning module, and a reaction chamber radiation temperature measurement and correction system.

[0026] The light source module generates detection light and illuminates the wafer surface within the reaction chamber. The wafer, located within the reaction chamber, is the object under test for epitaxial growth, thin film deposition, or thermal processing, and radiates infrared energy during temperature changes. Exemplarily, the light source module employs a light-emitting device with the same or partially overlapping wavelength band as the radiometric temperature measurement, ensuring that the reflectivity measurement corresponds to the same or approximately the same spectral range as the radiometric measurement, thereby reducing the impact of wavelength differences on subsequent emissivity correction. Furthermore, the light source module can output detection light with preset modulation characteristics to facilitate the extraction of reflection information corresponding to the detection light from the received signal.

[0027] An optical module receives thermal radiation from the wafer surface and reflected light formed on the wafer surface based on the reflected detection light. It then guides the thermal radiation and reflected light to corresponding acquisition paths to separate the optical paths. Exemplarily, the optical module includes a common objective lens and a beam splitter. The common objective lens collects both thermal radiation and reflected light, ensuring that the radiation measurement area and the reflectivity measurement area correspond to the same location on the wafer surface. The beam splitter separates the incoming thermal radiation and reflected light, allowing them to enter their respective acquisition paths for detection. This shared objective lens and beam splitter form a common optical path structure, ensuring that the radiation measurement area and the reflectivity measurement area are consistent, thereby reducing measurement errors caused by differences in the optical paths.

[0028] The acquisition module is used to sample the thermal radiation and reflected light after optical path separation, and convert them into a first electrical signal and a second electrical signal, respectively. Exemplarily, the acquisition module includes a radiation detector and a reflectivity detector. The radiation detector receives the thermal radiation generated by the wafer within the target temperature measurement band and outputs the corresponding first electrical signal; the reflectivity detector receives the reflected light from the wafer surface and outputs the corresponding second electrical signal. Specifically, in this embodiment, the acquisition module uses a narrowband InGaAs or Si detector.

[0029] The signal conditioning module processes the first electrical signal and the second electrical signal respectively to obtain a radiation signal and a reflectivity signal. For example, the signal conditioning module amplifies, filters, performs analog-to-digital conversion, and performs time synchronization processing on the first and second electrical signals respectively to obtain digitized radiation and reflectivity signals.

[0030] A reaction cavity radiation temperature measurement and correction system is used to obtain the corrected temperature of the wafer based on radiation and reflectivity signals. Specifically, the system acquires the radiation and reflectivity signals of the wafer within the reaction cavity; determines the time-domain offset relationship between the radiation and reflectivity signals, and aligns the reflectivity signal according to the time-domain offset relationship; determines the delay variation law based on the temperature estimation result corresponding to the radiation signal, and dynamically compensates the aligned reflectivity signal according to the delay variation law to obtain the compensated reflectivity signal; the delay variation law is used to characterize the correspondence between the delay and temperature; the system iteratively optimizes the model parameters of a pre-built emissivity correction model based on the compensated reflectivity signal and temperature estimation result; the emissivity correction model is used to characterize the correlation between the temperature estimation result, reflectivity signal, and emissivity; the radiation signal is corrected using the emissivity correction model with optimized model parameters to obtain the corrected radiation signal; and the corrected temperature of the wafer is obtained based on the corrected radiation signal.

[0031] The aforementioned temperature measurement device features a light source module that outputs detection light to the wafer within the reaction cavity. By ensuring that the corresponding wavelength band of the detection light is the same as or partially overlaps with the radiation temperature measurement band, the emissivity estimation bias introduced by wavelength differences is reduced, thus improving the accuracy of reflectivity correction. The optical module employs a common optical path acquisition method, ensuring that thermal radiation and reflected light have consistent spatial measurement positions and optical paths, reducing measurement errors caused by field-of-view deviations and optical path differences. The signal conditioning module synchronously conditions the two signals, improving the temporal consistency between the radiation signal and the reflectivity signal. The reaction cavity radiation temperature measurement correction system, based on the radiation signal and reflectivity signal, jointly corrects for time-domain offset, temperature-related delay changes, and emissivity model parameters. This compensates for the emissivity mismatch caused by the dynamic change in the delay of the reflectivity signal relative to the radiation signal with temperature variations, and enables the emissivity correction process to adapt to wafer temperature and radiation characteristic changes. This reduces temperature fluctuations caused by dynamic signal misalignment and model parameter deviations during reflectivity correction, improving the accuracy and stability of radiation temperature measurement results within the reaction cavity. The device is suitable for high-precision online temperature measurement scenarios with rapid temperature changes within the reaction cavity and complex process environments.

[0032] Please see Figure 2 In one exemplary embodiment, this application provides a method for correcting radiation temperature measurement in a reaction cavity, specifically including the following steps: Step 202: Obtain the radiation signal and reflectivity signal of the wafer inside the reaction chamber.

[0033] Specifically, the radiation signal is used to characterize the self-heating radiation characteristics of the wafer in the temperature measurement band and corresponds to the wafer temperature state, reflecting the current temperature change of the wafer.

[0034] The reflectivity signal is used to characterize the reflection response characteristics of a wafer to external incident detection light and is related to the optical properties of the wafer surface. It is used to reflect the wafer surface state and its changes with the process.

[0035] Step 204: Determine the time-domain offset relationship between the radiation signal and the reflectivity signal, and perform alignment processing on the reflectivity signal according to the time-domain offset relationship.

[0036] Specifically, the time-domain offset relationship is used to characterize the time correspondence between the radiation signal and the reflectivity signal, reflecting the time difference generated by the two signals during the acquisition process. Since the radiation signal and the reflectivity signal may reflect different physical states of the wafer at different time states, directly using the reflectivity signal for emissivity correction can easily introduce errors.

[0037] Based on this, the embodiments of this application determine the time-domain offset relationship between the two, and then perform alignment processing on the reflectivity signal according to the time-domain offset relationship, so that the aligned reflectivity signal and radiation signal have the same or similar state in the time dimension corresponding to the wafer.

[0038] Step 206: Determine the delay variation law based on the temperature estimation result corresponding to the radiation signal, and perform dynamic compensation on the aligned reflectivity signal according to the delay variation law to obtain the compensated reflectivity signal; the delay variation law is used to characterize the correspondence between the delay amount and temperature.

[0039] Specifically, the temperature estimation result is used to characterize the wafer temperature state determined based on the current radiation signal. For example, the temperature estimation result is a temperature result obtained from the radiation signal according to a pre-established radiation calibration relationship.

[0040] The delay parameter describes the time correspondence deviation between the reflectivity signal and the radiation signal, reflecting the dynamic relationship between the reflection response and radiation response of the wafer at different temperature states. The delay variation law characterizes the relationship between the time difference between the reflectivity signal and the radiation signal and temperature.

[0041] Since the surface state of the wafer, the spectral properties of the material, and the characteristics of the temperature-measuring optical path within the reaction chamber may all change with temperature, the delay of the reflectivity signal relative to the radiation signal is not a fixed value. Therefore, this embodiment determines the delay variation law based on the temperature estimation results and dynamically compensates the aligned reflectivity signal according to the delay variation law, ensuring that the reflectivity signal and the radiation signal maintain a consistent state correspondence during temperature changes.

[0042] Step 208: Based on the compensated reflectivity signal and temperature estimation results, iteratively optimize the model parameters of the pre-built emissivity correction model; the emissivity correction model is used to characterize the correlation between temperature estimation results, reflectivity signal and emissivity.

[0043] Specifically, emissivity characterizes the actual radiation capability of a wafer under its current state, and is used to establish the correspondence between the actual radiation and ideal radiation of the wafer. The model parameters of the emissivity correction model are used to characterize the adjustment variables in the emissivity correction model. By adjusting the values ​​of the model parameters, the degree of influence of reflectivity changes and temperature changes on the emissivity calculation results can be changed to adapt to the changes in wafer radiation characteristics under different process conditions.

[0044] Since fixed model parameters are difficult to accurately describe the emissivity variation process of wafers under different temperatures and process conditions, this embodiment iteratively optimizes the model parameters of the emissivity correction model based on the compensated reflectivity signal and temperature estimation results to improve the accuracy of emissivity calculation and provide more accurate emissivity parameters for subsequent radiation signal correction.

[0045] Step 210: The radiation signal is corrected using the emissivity correction model with optimized model parameters to obtain the corrected radiation signal; the corrected temperature of the wafer is obtained based on the corrected radiation signal.

[0046] Specifically, the corrected radiation signal characterizes the equivalent radiation features after eliminating the influence of wafer emissivity, and can more accurately reflect the true temperature state of the wafer compared to the original radiation signal.

[0047] The corrected temperature of the wafer can be obtained by combining the corrected radiation signal with a pre-established radiation-temperature correspondence. Since the corrected radiation signal can more accurately characterize the actual radiation capability of the wafer, the measured temperature obtained has higher accuracy and stability compared to the temperature result obtained directly based on the original radiation signal.

[0048] The aforementioned method for correcting radiation temperature measurement in a reaction cavity first performs time-domain alignment processing on the radiation and reflectivity signals of the wafer within the reaction cavity. Then, based on the temperature estimation results corresponding to the radiation signal, it determines the delay variation law of the reflectivity signal relative to the radiation signal and dynamically compensates for the reflectivity signal according to the delay variation law. Based on the compensated reflectivity signal and temperature estimation results, iteratively optimizes the model parameters of the emissivity correction model and corrects the radiation signal based on the optimized emissivity correction model. This method can compensate for the emissivity mismatch problem caused by the dynamic change of the delay of the reflectivity signal relative to the radiation signal with temperature changes, and enables the emissivity correction process to adapt to changes in wafer temperature and radiation characteristics. It reduces temperature fluctuations caused by dynamic signal misalignment and model parameter deviations during the reflectivity correction process, thereby improving the accuracy and stability of the radiation temperature measurement results within the reaction cavity.

[0049] Optionally, the above-mentioned reaction chamber radiation temperature measurement correction method further includes: The acquired radiation and reflectivity signals are preprocessed, and the time-domain offset relationship is determined based on the preprocessed radiation and reflectivity signals. The preprocessing methods include filtering and baseline correction.

[0050] Specifically, filtering is used to reduce high-frequency noise introduced during the acquisition process. For example, a moving average filter is used to smooth the radiation signal and the reflectivity signal respectively, where the moving window length L is determined according to the sampling rate and noise level, and can be L=5 to 11.

[0051] Baseline correction processing is used to eliminate background offset and low-frequency drift components in the signal. For example, high-pass filtering is applied to the radiated and reflective signals after moving average filtering to remove low-frequency background components and slow-changing trends, ensuring the processed signal varies around a uniform reference baseline, thereby mitigating the effects of equipment and environmental variations.

[0052] By employing the above scheme, noise reduction of radiation and reflectivity signals is achieved through a moving average filter, combined with high-pass filtering for baseline correction. This approach balances signal smoothing with dynamic response capability, reduces the impact of high-frequency noise and low-frequency drift on time-domain offset analysis, minimizes subsequent dynamic compensation and emissivity correction errors, and improves the stability and accuracy of the reaction cavity radiation temperature measurement results. Optionally, the temporal offset relationship between the radiation signal and the reflectivity signal is determined based on the temporal offset relationship, and the reflectivity signal is aligned according to the temporal offset relationship, including: Correlation analysis is performed on the radiation signal and the reflectivity signal; the corresponding delay between the radiation signal and the reflectivity signal is determined based on the correlation analysis results, and the corresponding delay is used as the time-domain offset relationship; the reflectivity signal is time-shifted based on the corresponding delay to obtain the aligned reflectivity signal.

[0053] Optionally, the correlation analysis includes constructing a cross-correlation function between the radiation signal and the reflectivity signal, and determining the degree of correlation between the radiation signal and the reflectivity signal based on the function value of the cross-correlation function under different delay conditions, thereby obtaining the cross-correlation analysis results; the corresponding delay is the delay amount that makes the cross-correlation function reach its peak value.

[0054] Specifically, normalized cross-correlation is performed on the radiation signal and the reflectivity signal to obtain the correlation between the two signals under different numbers of delayed sampling points; the expression for the normalized cross-correlation function is as follows:

[0055] in, Indicates the number of delayed sampling points. Normalized cross-correlation results between the time-radiation signal and the reflectivity signal; This represents the preprocessed radiation signal; This represents the preprocessed reflectivity signal; Indicates the amount of time delay; , These represent the mean values ​​of the radiation signal and the reflectivity signal, respectively. This indicates the number of delayed sampling points.

[0056] Furthermore, in this embodiment, by traversing different numbers of delayed sampling points, the number of delayed sampling points corresponding to the point where the normalized cross-correlation function reaches its maximum value is determined, and this number of delayed sampling points is determined as the corresponding delay between the radiation signal and the reflectivity signal. This corresponding delay can characterize the time-domain offset relationship between the two signals, and its expression is: .

[0057] After determining the corresponding delay amount, the reflectivity signal is time-shifted according to the corresponding delay amount. If This indicates that the reflectivity signal lags behind the radiation signal, so the reflectivity signal is shifted forward. sampling points; if Then, the reflectivity signal is shifted backward by the corresponding sampling point. The expression for the aligned reflectivity signal is:

[0058] in, This represents the aligned reflectivity signal. For data positions that exceed the original time range after time shift, null values ​​are used for padding, and subsequent processing only uses the effective overlap time interval between the radiation signal and the aligned reflectivity signal.

[0059] By employing the above scheme, the corresponding delay between the radiation signal and the reflectivity signal is determined through normalized cross-correlation calculation. This reduces the impact of the amplitude difference between the two signals on the correlation analysis results and improves the stability and accuracy of the time-domain offset relationship determination. Furthermore, by performing a time shift on the reflectivity signal based on the corresponding delay, the reflectivity signal and the radiation signal correspond to the same or approximately the same physical state of the wafer in the time dimension. This reduces time-domain misalignment and improves the accuracy and stability of subsequent dynamic compensation, emissivity correction, and temperature calculation.

[0060] Optionally, the delay variation law is determined based on the temperature estimation result corresponding to the radiation signal, including: The temperature estimation sequence of the wafer at multiple sampling times is determined based on the radiation signal; the radiation signal and reflectivity signal are divided into multiple segmented signals according to the temperature range corresponding to the temperature estimation sequence; for each segmented signal, the segmented delay between the radiation signal and reflectivity signal in that segmented signal is determined; the segmented temperature and segmented delay corresponding to each segmented signal are linearly fitted to establish the delay variation law of the delay with temperature; the segmented temperature is the temperature estimate within the temperature range of the segmented signal.

[0061] Specifically, in this embodiment, the temperature estimation sequence of the wafer is first obtained based on the radiation signal. Exemplarily, the uncorrected radiation signal is combined with a pre-established blackbody calibration relationship to estimate the temperature estimates corresponding to multiple sampling times, thereby forming a temperature estimation sequence. This initial temperature estimation sequence is used to characterize the temperature change process of the wafer within the sampling time range.

[0062] Furthermore, the radiation and reflectivity signals are segmented according to the temperature range corresponding to the temperature estimation sequence. For example, the entire measurement interval is divided into multiple temperature segments, such as multiple segments covering a temperature range of 50°C. Each segment includes a radiation signal subsequence and a reflectivity signal subsequence at the corresponding sampling time, serving as the segment signal. Since the wafer temperature variation range within the same segment is relatively small, the delay between the radiation and reflectivity signals can be approximated as relatively stable within that segment.

[0063] For each segmented signal, the segment delay between the radiation signal and the reflectivity signal in that segment is determined. For example, for the k-th segmented signal, the radiation signal subsequence and the reflectivity signal subsequence within that segment are obtained, and correlation analysis is performed on them to obtain the segment delay corresponding to that segment. The segmented delay It is used to characterize the time offset of the reflectivity signal relative to the radiation signal within this temperature range.

[0064] Further, the segment temperature corresponding to each segment signal is determined. The segment temperature is used to characterize the temperature state of the corresponding segment signal. For example, the segment temperature can be an estimated temperature value within the temperature range of the segment signal, such as the midpoint temperature of the temperature range, or the average of the estimated temperature values ​​corresponding to multiple sampling times within the segment. Taking the k-th segment signal as an example, its segment temperature is denoted as... The segmented delay is denoted as .

[0065] After obtaining the segmented temperature and segmented delay corresponding to each segmented signal, a linear fit is performed on multiple segmented temperatures and multiple segmented delays to establish the delay variation law of the delay amount with temperature. For example, the delay variation law can be expressed as: ;in, This represents the delay at temperature T; T represents the estimated temperature; a represents the coefficient of variation of the delay with temperature; and b represents the bias term of the delay.

[0066] It should be noted that in other embodiments, when there is a nonlinear trend between the segmented temperature and the segmented delay, a delay variation law can be established by using high-order polynomial fitting or spline fitting based on the data variation characteristics, so as to improve the ability of the delay variation law to characterize the delay variation trend in different temperature ranges.

[0067] After completing the linear fitting, the delay variation law is obtained to characterize the relationship between the delay and temperature. This delay variation law reflects the trend of the delay of the reflectivity signal relative to the radiation signal changing with the wafer temperature, providing a basis for subsequent determination of the dynamic delay based on the current temperature estimate and dynamic compensation of the reflectivity signal.

[0068] By adopting the above scheme, the radiation signal and reflectivity signal are segmented according to the temperature range of the temperature estimation sequence, and the segment delay amount in each temperature segment is determined. Then, linear fitting is performed based on the segment temperature and the segment delay amount. This can effectively identify and quantify the delay changes caused by wavelength drift or physical lag, and avoid the problem of signal misalignment in different temperature ranges caused by using only a fixed delay amount. This improves the accuracy of subsequent dynamic compensation and emissivity correction.

[0069] Optionally, the reflectivity signal after alignment is dynamically compensated according to the delay variation law to obtain the compensated reflectivity signal, including: Based on the temperature estimate and delay variation of the wafer at the current sampling moment, determine the dynamic delay amount at the current sampling moment; based on the dynamic delay amount, determine the reflectivity sampling position at the current sampling moment; obtain the reflectivity value corresponding to the reflectivity sampling position from the aligned reflectivity signal; and generate a compensated reflectivity signal based on the reflectivity values ​​at multiple sampling moments.

[0070] Specifically, in this embodiment, the reflectance value used at the current sampling time is not directly taken from the input reflectance signal at the sampling time. Instead of taking the value at a given location, the corresponding sampling position is found in the input reflectivity signal based on the dynamic delay.

[0071] For each sampling point Use the temperature estimate corresponding to the current sampling time. Substituting the delay variation pattern, determine the dynamic delay corresponding to the current sampling time. Specifically, during the initial dynamic compensation execution... This is based on the initial temperature estimate obtained from the uncorrected radiation signal; during iterative dynamic compensation, This is the temperature result obtained after the last calibration.

[0072] For example, when the delay variation follows a linear relationship, the dynamic delay corresponding to the current sampling time is expressed as: ;in, Indicates the sampling time The corresponding dynamic delay; Indicates the sampling time The corresponding temperature estimate; This represents the coefficient of variation of the delay with temperature. This represents the bias term indicating the delay. After obtaining the dynamic delay, it is determined based on the current sampling time. and dynamic delay Determine the reflectance sampling position corresponding to the current sampling time.

[0073] Optionally, when the reflectance sampling position is a non-integer sampling position, the reflectance value corresponding to the reflectance sampling position is obtained from the aligned reflectance signal, including: Interpolation is performed on the aligned reflectivity signal to obtain the reflectivity value corresponding to the non-integer sampling position.

[0074] Specifically, due to The number of sampling points may not be an integer, and directly shifting according to an integer number of sampling points would result in compensation errors. Therefore, in this embodiment, interpolation is used to resample the input reflectivity signal to obtain the compensated reflectivity value corresponding to the current sampling time. For example, the expression for the compensated reflectivity signal is:

[0075] in, Indicates the sampling time The corresponding compensated reflectivity signal; Indicates the input reflectivity signal at the sampling position The value at; Indicates based on the current sampling time and dynamic delay Determined reflectance sampling location; This indicates interpolation processing.

[0076] For example, the interpolation process employs cubic spline interpolation or linear interpolation. When the reflectance sampling location... When the sampling position is an integer, the reflectance value corresponding to that sampling position in the input reflectance signal is directly read; when the reflectance sampling position is an integer, the reflectance value is read directly. When the sampling position is a non-integer sampling position, the reflectance value corresponding to the non-integer sampling position is obtained through interpolation, thereby achieving precise alignment at the sub-sampling level.

[0077] Furthermore, before interpolation, this embodiment of the application also performs boundary expansion processing on the input reflectivity signal. For example, the boundary expansion processing includes mirror extension or constant extension to avoid failing to obtain a valid reflectivity value when the reflectivity sampling position exceeds the original sampling range. Following the above method, a dynamic delay amount is determined for each of the multiple sampling times, and the reflectivity value corresponding to each sampling time is obtained. The reflectivity values ​​corresponding to the multiple sampling times are then arranged in chronological order to form a compensated reflectivity signal. .

[0078] Optionally, the above-mentioned reaction chamber radiation temperature measurement correction method further includes: The presence of temperature-dependent drift is determined based on the degree of change in the delay variation law; if temperature-dependent drift exists, dynamic compensation is performed on the aligned reflectivity signal; otherwise, the model parameters of the pre-built emissivity correction model are iteratively optimized using the aligned reflectivity signal and temperature estimation results.

[0079] Specifically, the aforementioned dynamic compensation action is only performed when temperature drift exists. For example, determining the existence of temperature-related drift includes: in the expression of the delayed change law... In the middle, the fitted slope If the absolute value of the delay is greater than a preset slope threshold and the coefficient of determination is greater than a preset fitting threshold, then temperature-dependent drift is determined to exist, and dynamic compensation is performed. Otherwise, it is determined that only a fixed delay exists, and the aligned reflectivity signal is directly used in the subsequent emissivity correction model calculation. The coefficient of determination characterizes the degree of consistency between the temperature of each segment and the delay of each segment; the larger the coefficient of determination, the stronger the regularity of the delay change with temperature. In this embodiment, the preset slope threshold is set to 0.01, and the preset fitting threshold is set to 0.8.

[0080] By adopting the above scheme, the estimated temperature value corresponding to the current sampling time can be substituted into the delay variation law, and the dynamic delay amount that changes with temperature can be determined point by point. This avoids the problem that the reflectivity signal and radiation signal are still misaligned in different temperature ranges when only a fixed delay amount is used. By resampling the reflectivity signal through interpolation, compensation can be made when the dynamic delay amount is a non-integer number of sampling points, thereby improving the timing correspondence accuracy between the reflectivity signal and the radiation signal, and thus improving the accuracy and stability of subsequent emissivity correction and wafer temperature measurement results.

[0081] Optionally, when the model parameters include reflectivity correction coefficients and temperature-related correction coefficients, the model parameters of the pre-built emissivity correction model are iteratively optimized based on the compensated reflectivity signal and temperature estimation results, including: Based on the current reflectivity correction coefficient, the current temperature correlation correction coefficient, the compensated reflectivity signal, and the temperature estimation result, determine the current emissivity; based on the current emissivity and radiation signal, determine the current temperature sequence, and determine the temperature fluctuation index based on the current temperature sequence; based on the temperature fluctuation index, iteratively update the current reflectivity correction coefficient and the current temperature correlation correction coefficient; determine the current reflectivity correction coefficient and the current temperature correlation correction coefficient when the temperature fluctuation index meets the preset convergence condition as model parameters.

[0082] Specifically, the emissivity correction model is used to determine the emissivity of the wafer based on its current temperature and reflectivity.

[0083] Optionally, the expression for the emissivity correction model is:

[0084] in, This represents the emissivity at sampling time t. This represents the compensated reflectivity signal corresponding to sampling time t. This represents the temperature estimation result corresponding to sampling time t. This represents the reflectivity correction factor. This represents the temperature-dependent correction coefficient. By introducing the temperature-dependent correction coefficient, this application can effectively compensate for the nonlinear deviation in emissivity-reflectivity caused by wavelength drift or changes in the spectral properties of materials with temperature, which is superior to the fixed coefficient model in traditional technologies.

[0085] when At that time, the emissivity correction model degenerates into a fixed coefficient model, i.e. This indicates that the emissivity of the fixed-coefficient model is determined solely by the reflectivity signal and the fixed reflectivity correction coefficients, without considering the effect of temperature changes on emissivity. and At that time, the emissivity correction model further degenerates into the Kirchhoff ideal model, that is: This indicates that under ideal opaque conditions, the emissivity and reflectivity of the wafer surface satisfy a complementary relationship, with emissivity equal to 1 minus reflectivity. Using the above scheme, without considering dynamic delay and temperature coefficient, it can degenerate into a fixed-coefficient model or Kirchhoff's ideal model used in traditional techniques, covering the models used in traditional techniques and offering better backward compatibility and flexibility.

[0086] Temperature fluctuation indexes are used to evaluate the temperature correction effect corresponding to the current model parameters. For example, the temperature fluctuation index includes at least one of the standard deviation, peak value, and variance of the current temperature series. For example, the temperature fluctuation index is defined as the temperature standard deviation over the entire time series or the effective overlapping time interval, and its expression is: ;in, Indicates temperature fluctuation index; Indicates the calculation of standard deviation; Indicates the sampling time The corresponding calibrated measured temperature.

[0087] A smaller temperature fluctuation index indicates a more stable corrected temperature sequence and a better match between the current model parameters and the current radiation characteristics of the wafer. Based on this, the embodiments of this application optimize the reflectivity correction coefficient by minimizing the temperature fluctuation index. Temperature-related correction factor Iterative optimization is performed. For example, the Nelder-Mead simplex method or particle swarm optimization algorithm is used to solve for... Minimum optimal parameter combination To ensure the physical rationality of the model parameters, constraints are placed on the range of parameter values, for example: , By employing the above constraints, the parameters can be prevented from deviating from a reasonable physical range during the iteration process, thereby improving the stability and reliability of the emissivity calculation results. Using the above scheme, with the standard deviation of the corrected temperature sequence as the optimization objective function, the undetermined parameters in the emissivity model are automatically solved without manual calibration or prior knowledge, adapting to different surface materials and measurement conditions.

[0088] If the current temperature fluctuation index does not meet the preset convergence condition, the current reflectivity correction coefficient and the current temperature correlation correction coefficient are iteratively updated according to the temperature fluctuation index, and the emissivity, temperature sequence and temperature fluctuation index are recalculated based on the updated current reflectivity correction coefficient and the current temperature correlation correction coefficient.

[0089] If the current temperature fluctuation index meets the preset convergence conditions, the current reflectivity correction coefficient and the current temperature-related correction coefficient are determined as the model parameters of the emissivity correction model. For example, the preset convergence conditions include the temperature fluctuation index being less than a preset fluctuation threshold, or the change in the temperature fluctuation index obtained in two adjacent iterations being less than a preset change threshold.

[0090] Optionally, an emissivity correction model with optimized model parameters is used to correct the radiation signal, resulting in a corrected radiation signal, including: Based on the determined model parameters, temperature estimation results, and compensated reflectivity signal, the emissivity corresponding to the wafer is determined; the background radiation signal is subtracted from the radiation signal to obtain the effective radiation signal; the effective radiation signal is normalized and corrected according to the emissivity to obtain the corrected radiation signal, which satisfies the following expression:

[0091] in, This represents the corrected radiation signal. Indicates a radiated signal. Indicates background radiation signal, Indicates the emissivity.

[0092] Specifically, the background radiation signal is used to characterize the signal components caused by detector dark current, ambient background radiation, or system bias. The background radiation signal can be obtained through pre-measurement or estimated based on background sampling results during temperature measurement. The effective radiation signal obtained after subtracting the background radiation signal is then corrected by emissivity normalization to obtain a calibrated radiation signal that more closely approximates the blackbody radiation state. Furthermore, based on the calibrated radiation signal and a pre-established radiation temperature calibration relationship, the wafer calibration temperature is determined.

[0093] By adopting the above scheme and introducing reflectivity correction coefficients and temperature-related correction coefficients, the emissivity correction model can not only characterize the influence of wafer surface reflection characteristics on emissivity based on the compensated reflectivity signal, but also characterize the influence of emissivity on temperature changes based on temperature estimation results. Iterative optimization of model parameters based on temperature fluctuation indices enables the model parameters to adapt to the current wafer surface growth state and process temperature changes, reducing emissivity estimation bias caused by fixed model parameters. Simultaneously, by subtracting background radiation signals and normalizing the effective radiation signal based on emissivity, the influence of background interference and emissivity changes on radiation temperature measurement results can be reduced, improving the accuracy and stability of wafer temperature measurement.

[0094] Optionally, the corrected temperature of the wafer is obtained based on the corrected radiation signal, including: Based on the pre-established blackbody calibration relationship, the corrected radiation signal is converted into the wafer-corrected temperature.

[0095] Specifically, the blackbody calibration relationship refers to the correspondence between the calibrated radiation signal and temperature established using a standard blackbody furnace. The blackbody calibration formula in this embodiment adopts the Wien approximation calibration formula, and its expression is:

[0096] in, Indicates the sampling time The corresponding wafer measurement temperature; Indicates the sampling time The corresponding corrected radiation signal; , , These are calibration coefficients obtained beforehand using a blackbody furnace; Represents the natural logarithm; Used to convert Kelvin temperature to Celsius temperature.

[0097] For example, the blackbody furnace is set as multiple known temperature points, and the radiation response signal output by the temperature measurement channel at each known temperature point is collected. The calibration coefficient is obtained by fitting the corresponding radiation response signal at each known temperature point. , , During the temperature measurement phase, the corrected radiation signal... Substituting into the blackbody calibration formula above, the corresponding wafer measurement temperature can be obtained. .

[0098] By adopting the above scheme, the calibrated radiation signal is converted into the wafer calibrated temperature according to the blackbody calibration relationship, which can keep the measured temperature consistent with the blackbody furnace calibration system, thereby improving the accuracy of wafer radiation temperature measurement.

[0099] Optionally, the above-mentioned reaction chamber radiation temperature measurement correction method further includes: The temperature after wafer correction is used as the temperature estimate result after the radiation signal is updated; based on the updated temperature estimate result, dynamic compensation and model parameters of the emissivity correction model are re-executed, and the temperature after wafer correction is updated based on the re-determined model parameters; if the preset update conditions are met, the update is stopped, and the temperature corresponding to the time when the update stops is taken as the final temperature of the wafer.

[0100] Specifically, after completing a temperature calculation using the steps provided in the above embodiments, this application embodiment can further use the obtained measured temperature as an updated temperature estimation result, and re-execute the correction process based on the above steps to obtain a new measured temperature. The preset update conditions include the difference between the measured temperatures before and after the update being less than a preset temperature threshold, or the difference between the temperature fluctuation index before and after the update being less than a preset fluctuation threshold.

[0101] By adopting the above scheme, the model parameters are further modified based on the existing temperature results, thereby improving the timing matching accuracy between the reflectivity signal and the radiation signal, reducing the fluctuation of the measured temperature sequence, and improving the accuracy and stability of the final temperature measurement results.

[0102] It should be understood that although the steps in the flowcharts of the embodiments described above are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the embodiments described above may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages of other steps.

[0103] Based on the same inventive concept, this application also provides a reaction cavity radiation temperature measurement correction system. This system is applicable to the above-described reaction cavity radiation temperature measurement correction method. The solution provided by this system is similar to the solution described in the above-described method. Therefore, the specific limitations of one or more system embodiments provided below can be found in the limitations of the method above, and will not be repeated here.

[0104] Please see Figure 3 In one embodiment, the reaction cavity radiation temperature measurement and correction system includes: an acquisition module, a delay detection module, a compensation module, an optimization module, and a correction module.

[0105] The acquisition module is used to acquire the radiation and reflectivity signals of the wafer inside the reaction chamber.

[0106] The delay detection module is used to determine the time-domain offset relationship between the radiation signal and the reflectivity signal, and to align the reflectivity signal according to the time-domain offset relationship.

[0107] The compensation module is used to determine the delay variation law based on the temperature estimation result corresponding to the radiation signal, and to dynamically compensate the aligned reflectivity signal according to the delay variation law to obtain the compensated reflectivity signal; the delay variation law is used to characterize the correspondence between the delay amount and temperature.

[0108] The optimization module is used to iteratively optimize the model parameters of the pre-built emissivity correction model based on the compensated reflectivity signal and temperature estimation results; the emissivity correction model is used to characterize the correlation between temperature estimation results, reflectivity signal and emissivity.

[0109] The calibration module is used to correct the radiation signal using an emissivity correction model with optimized model parameters, and obtain the corrected radiation signal; based on the corrected radiation signal, the temperature of the wafer after calibration is obtained.

[0110] The aforementioned reaction cavity radiation temperature measurement correction system first performs time-domain alignment processing on the radiation and reflectivity signals of the wafer within the reaction cavity. Then, based on the temperature estimation results corresponding to the radiation signal, it determines the delay variation law of the reflectivity signal relative to the radiation signal and dynamically compensates for the reflectivity signal according to the delay variation law. Based on the compensated reflectivity signal and temperature estimation results, iteratively optimizes the model parameters of the emissivity correction model and corrects the radiation signal based on the optimized emissivity correction model. This system can compensate for the emissivity mismatch problem caused by the dynamic change of the delay of the reflectivity signal relative to the radiation signal with temperature changes, and enables the emissivity correction process to adapt to changes in wafer temperature and radiation characteristics. It reduces temperature fluctuations caused by dynamic signal misalignment and model parameter deviations during the reflectivity correction process, thereby improving the accuracy and stability of the radiation temperature measurement results within the reaction cavity. Each module in the aforementioned reaction chamber radiation temperature measurement and correction system can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in the processor of a computer device in hardware form or independent of it, or stored in the memory of a computer device in software form, so that the processor can call and execute the corresponding operations of each module.

[0111] In one feasible embodiment, a computer device is provided, which may be a terminal, and its internal structure diagram may be as follows: Figure 4As shown, the computer device includes a processor, memory, input / output interface, communication interface, display unit, and input device. The processor, memory, and input / output interface are connected via a system bus, and the communication interface, display unit, and input device are also connected to the system bus via the input / output interface. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage media. The input / output interface is used for exchanging information between the processor and external devices. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, mobile cellular networks, NFC (Near Field Communication), or other technologies. When the computer program is executed by the processor, it implements the aforementioned reaction chamber radiation temperature measurement correction method. The display unit is used to form a visually visible image and can be a display screen, projection device, or virtual reality imaging device. The display screen can be an LCD screen or an e-ink screen. The input device of the computer device can be a touch layer covering the display screen, or buttons, trackballs, or touchpads set on the casing of the computer device, or external keyboards, touchpads, or mice, etc.

[0112] Those skilled in the art will understand that Figure 4 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0113] In one feasible embodiment, a computer device is provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the above-described reaction chamber radiation temperature measurement correction method.

[0114] In one feasible embodiment, a computer program product is provided, including a computer program that, when executed by a processor, implements the above-described reaction cavity radiation temperature measurement correction method.

[0115] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0116] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.

Claims

1. A method for correcting radiation temperature measurement in a reaction chamber, characterized in that, The method includes: Acquire the radiation and reflectivity signals of the wafer within the reaction chamber; The temporal offset relationship between the radiation signal and the reflectivity signal is determined based on the temporal offset relationship, and the reflectivity signal is aligned accordingly. The delay variation law is determined based on the temperature estimation result corresponding to the radiation signal, and the reflectivity signal after alignment is dynamically compensated according to the delay variation law to obtain the compensated reflectivity signal. The delay variation law is used to characterize the correspondence between the delay amount and temperature; Based on the compensated reflectivity signal and the temperature estimation result, the model parameters of the pre-built emissivity correction model are iteratively optimized; the emissivity correction model is used to characterize the correlation between the temperature estimation result, the reflectivity signal, and the emissivity. The radiation signal is corrected using an emissivity correction model with optimized model parameters to obtain the corrected radiation signal. The corrected temperature of the wafer is obtained based on the corrected radiation signal.

2. The method according to claim 1, characterized in that, The method further includes: The acquired radiation signal and reflectivity signal are preprocessed, and the time-domain offset relationship is determined based on the preprocessed radiation signal and reflectivity signal. The preprocessing methods include filtering and baseline correction.

3. The method according to claim 1, characterized in that, The step of determining the time-domain offset relationship between the radiation signal and the reflectivity signal, and aligning the reflectivity signal according to the time-domain offset relationship, includes: Correlation analysis was performed on the radiation signal and the reflectivity signal; Based on the correlation analysis results, the corresponding delay between the radiation signal and the reflectivity signal is determined, and the corresponding delay is used as the time-domain offset relationship. The reflectivity signal is time-shifted according to the corresponding delay amount to obtain the aligned reflectivity signal.

4. The method according to claim 3, characterized in that: The correlation analysis includes constructing a cross-correlation function between the radiation signal and the reflectivity signal, and determining the degree of correlation between the radiation signal and the reflectivity signal based on the function value of the cross-correlation function under different delay conditions, thereby obtaining the cross-correlation analysis result; the corresponding delay amount is the delay amount corresponding to the cross-correlation function reaching its peak value.

5. The method according to claim 1, characterized in that, The step of determining the delay variation law based on the temperature estimation result corresponding to the radiation signal includes: Based on the radiation signal, a temperature estimation sequence for the wafer at multiple sampling times is determined; Based on the temperature range corresponding to the temperature estimation sequence, the radiation signal and the reflectivity signal are divided into multiple segmented signals; For each segmented signal, determine the segment delay between the radiation signal and the reflectivity signal in that segmented signal; Linearly fit the segment temperature and segment delay corresponding to each segment signal to establish the delay variation law of the delay with temperature; the segment temperature is the temperature estimate within the temperature range of the segment signal.

6. The method according to claim 5, characterized in that, The step of dynamically compensating the aligned reflectivity signal according to the delay variation law to obtain the compensated reflectivity signal includes: Based on the temperature estimate of the wafer at the current sampling moment and the delay variation pattern, determine the dynamic delay amount corresponding to the current sampling moment; Based on the dynamic delay, determine the reflectance sampling position corresponding to the current sampling moment; Obtain the reflectance value corresponding to the reflectance sampling position from the aligned reflectance signal; Based on the reflectance values ​​corresponding to multiple sampling times, a compensated reflectance signal is generated.

7. The method according to claim 6, characterized in that, The reflectance sampling position is a non-integer sampling position; The step of obtaining the reflectance value corresponding to the reflectance sampling position from the aligned reflectance signal includes: The reflectivity signal after alignment is interpolated to obtain the reflectivity value corresponding to the non-integer sampling position.

8. The method according to claim 1, characterized in that, The model parameters include reflectivity correction coefficients and temperature-related correction coefficients; The step of iteratively optimizing the model parameters of the pre-built emissivity correction model based on the compensated reflectivity signal and the temperature estimation result includes: The current emissivity is determined based on the current reflectivity correction coefficient, the current temperature-related correction coefficient, the compensated reflectivity signal, and the temperature estimation result. Based on the current emissivity and the radiation signal, determine the current temperature sequence, and based on the current temperature sequence, determine the temperature fluctuation index; Based on the temperature fluctuation index, iteratively update the current reflectivity correction coefficient and the current temperature correlation correction coefficient; The current reflectivity correction coefficient and the current temperature correlation correction coefficient when the temperature fluctuation index meets the preset convergence condition are determined as the model parameters.

9. The method according to claim 8, characterized in that, The expression for the emissivity correction model is: in, This represents the emissivity at sampling time t. This represents the compensated reflectivity signal corresponding to sampling time t. This represents the temperature estimation result corresponding to sampling time t. This represents the reflectivity correction factor. This represents the temperature-related correction factor.

10. The method according to claim 9, characterized in that, The emissivity correction model, with optimized model parameters, corrects the radiation signal to obtain the corrected radiation signal, including: Based on the determined model parameters, the temperature estimation results, and the compensated reflectivity signal, the emissivity corresponding to the wafer is determined; The effective radiation signal is obtained by subtracting the background radiation signal from the radiation signal. The effective radiation signal is normalized and corrected based on the emissivity to obtain a corrected radiation signal, which satisfies the following expression: in, This represents the corrected radiation signal. Indicates a radiated signal. Indicates background radiation signal, Indicates the emissivity.

11. The method according to claim 1, characterized in that, The step of obtaining the corrected temperature of the wafer based on the corrected radiation signal includes: Based on the pre-established blackbody calibration relationship, the corrected radiation signal is converted into a wafer-corrected temperature.

12. The method according to claim 1, characterized in that, The method further includes: The temperature after wafer correction is used as the temperature estimate result after the radiation signal is updated. Based on the updated temperature estimation results, the dynamic compensation is re-executed and the model parameters of the emissivity correction model are determined, and the wafer-corrected temperature is updated based on the re-determined model parameters; If the preset update conditions are met, the update is stopped, and the temperature at which the update stops is taken as the final temperature of the wafer.

13. The method according to claim 1, characterized in that, The method further includes: Determine whether temperature-related drift exists based on the degree of change in the aforementioned delayed change pattern; In the presence of temperature-dependent drift, the aligned reflectivity signal is dynamically compensated; otherwise, the aligned reflectivity signal and the temperature estimation result are used to iteratively optimize the model parameters of the pre-built emissivity correction model.

14. A reaction chamber radiation temperature measurement and correction system, characterized in that, The system includes: The acquisition module is used to acquire the radiation and reflectivity signals of the wafer inside the reaction chamber; The delay detection module is used to determine the time-domain offset relationship between the radiation signal and the reflectivity signal, and to perform alignment processing on the reflectivity signal according to the time-domain offset relationship; The compensation module is used to determine the delay variation law based on the temperature estimation result corresponding to the radiation signal, and to dynamically compensate the aligned reflectivity signal according to the delay variation law to obtain the compensated reflectivity signal; the delay variation law is used to characterize the correspondence between the delay amount and temperature. An optimization module is used to iteratively optimize the model parameters of a pre-built emissivity correction model based on the compensated reflectivity signal and the temperature estimation result; the emissivity correction model is used to characterize the correlation between the temperature estimation result, the reflectivity signal, and the emissivity. The correction module is used to correct the radiation signal using an emissivity correction model with optimized model parameters to obtain a corrected radiation signal; and to obtain the corrected temperature of the wafer based on the corrected radiation signal.

15. A temperature measuring device, characterized in that, The device includes: The light source module is used to generate detection light and irradiate the wafer surface inside the reaction chamber with the detection light; An optical module is used to receive thermal radiation from the surface of the wafer and reflected light formed on the surface of the wafer based on the detection light, and to perform optical path separation between the thermal radiation and the reflected light; The acquisition module is used to sample the thermal radiation and the reflected light after optical path separation, and convert them into a first electrical signal and a second electrical signal, respectively. The signal conditioning module is used to process the first electrical signal and the second electrical signal respectively to obtain the radiation signal and the reflectivity signal; The reaction cavity radiation temperature measurement and correction system as described in claim 14 is used to obtain the corrected temperature of the wafer based on the radiation signal and the reflectivity signal.

16. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the method of any one of claims 1-13.

17. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the method as described in any one of claims 1-13.

Images