Method, system and device for controlling epitaxial growth of semiconductor and related products

Abstract

The invention provides a semiconductor epitaxial growth control method and system, a control device and a related product, and relates to the technical field of semiconductor detection.The semiconductor epitaxial growth control method comprises the steps that reflection spectrum data is obtained in real time and analyzed to obtain an optical feature set, then the optical feature set is input into an optical transmission model for iterative optimization, and a predicted growth rate is obtained; and comparing the predicted growth rate with the target growth rate, generating deviation information, and outputting a process adjustment control instruction, so that the main control system performs closed-loop correction on the process parameters in the next process period. According to the invention, the optical feature set can dynamically represent the influence of the interface transition layer on the reflection spectrum, so that the inversion precision and robustness of the real epitaxial structure state are improved; closed-loop correction is performed on the process parameters based on the predicted growth rate, so that the growth rate drift can be inhibited in time, the process disturbance can be adaptively compensated, the manual calibration and intervention requirements are reduced, and the production stability, repeatability and batch-to-batch consistency are improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of semiconductor detection, and in particular to a control method and system for semiconductor epitaxial growth, a control device and related products. BACKGROUND

[0002] In the epitaxial growth process of compound semiconductor devices such as vertical cavity surface emitting lasers (VCSEL), the thickness, composition and interface quality of the epitaxial layer directly determine the key performances such as the optical characteristics of the resonant cavity, the threshold, efficiency and consistency of the device. Therefore, in the metal organic chemical vapor deposition (MOCVD) epitaxial growth process, the growth state is monitored in-situ, and closed-loop control is realized accordingly, which is a common technical path to improve product yield and batch consistency. At present, the industry generally uses in-situ reflectometry to measure the change of the reflection spectrum of the growth surface with time, and according to the periodic oscillation characteristics of the reflection spectrum, the film thickness and growth rate are deduced to realize online monitoring of the epitaxial process.

[0003] The traditional in-situ spectral reflectometry monitoring scheme usually performs parameter inversion based on fixed optical constants and ideal layer structure model, for example, assuming that the refractive index and extinction coefficient (n, k) of each material layer are constant, the interface between each layer is an ideal abrupt interface, and the growth parameters, which usually include thickness and average growth rate, are calculated by fitting the reflection spectrum oscillation period or selecting the phase change of the characteristic wavelength.

[0004] However, in actual MOCVD epitaxial growth, due to memory effect and high temperature diffusion, there is a gradual change zone at the interface, and the optical constants of the material will change with the composition and temperature, so the fixed model parameters cannot characterize these dynamic physical effects, resulting in systematic errors in the inverted growth parameters, which further affects the process determination and control accuracy. SUMMARY

[0005] The purpose of the present application is to provide a control method, system and device for semiconductor epitaxial growth and related products to overcome the defects in the prior art that the fixed model parameters cannot characterize the dynamic physical effects in the epitaxial growth process, resulting in systematic errors in the inverted growth parameters, which further affects the process determination and control accuracy.

[0006] In a first aspect, the present application provides a control method for semiconductor epitaxial growth, comprising: obtaining reflection spectrum data; wherein the reflection spectrum data is generated based on the reflection light signal generated in the semiconductor epitaxial growth process; analyzing the reflectance spectrum data to obtain an optical feature set, wherein the optical feature set comprises at least one of a stop band center wavelength, a stop band width, a Fabry-Perot cavity mode center wavelength, a cavity mode depth, a cavity mode full width at half maximum, and an absolute reflectance value at a preset wavelength point of the distributed Bragg reflector; inputting the optical feature set into a preset optical transfer model, and iteratively optimizing a variable physical parameter set of the optical transfer model until a preset condition is met to obtain a target parameter set; generating a predicted growth rate based on the target parameter set, and comparing the predicted growth rate with a target growth rate to obtain deviation information; generating a process adjustment control instruction according to the deviation information, so that the main control system executes the epitaxial growth process according to the adjusted process parameters in the next process cycle.

[0007] In one of the embodiments, The analyzing the reflectance spectrum data to obtain an optical feature set comprises: performing spectrum feature recognition on the reflectance spectrum data to determine a stop band of the distributed Bragg reflector, and extracting a center wavelength and / or a stop band width of the stop band; and / or performing cavity mode feature extraction on the reflectance spectrum data to determine a Fabry-Perot cavity mode, and extracting a center wavelength, a cavity mode depth and / or a cavity mode full width at half maximum of the Fabry-Perot cavity mode; and / or selecting at least one preset wavelength point in the reflectance spectrum data, and extracting an absolute reflectance value at the preset wavelength point.

[0008] In one of the embodiments, the variable physical parameter set comprises a layer thickness of a current layer, a layer thickness, a component and an optical constant of an interface transition layer, and the optical constant comprises a refractive index and an extinction coefficient; The inputting the optical feature set into a preset optical transfer model, and iteratively optimizing a variable physical parameter set of the optical transfer model until a preset condition is met to obtain a target parameter set comprises: constructing an objective function; performing forward calculation on the preset optical transfer model according to the current variable physical parameter set to obtain predicted reflectance spectrum data; analyzing the predicted reflectance spectrum data to obtain a predicted feature set, calculating a difference between the predicted feature set and the optical feature set according to the objective function, iteratively optimizing the current variable physical parameter set according to the difference until a preset condition is met, and taking the variable physical parameter set at this time as the target parameter set.

[0009] In one of the embodiments, the objective function is a square of a two-norm difference between the predicted feature set and the optical feature set. the difference between the predicted feature set and the optical feature set according to the objective function, iteratively optimizing the current variable physical parameter set according to the difference until a preset condition is met, comprising: differencing the predicted feature set and the optical feature set according to the objective function to obtain a feature residual, and calculating a sensitivity matrix of the feature residual with respect to the current variable physical parameter set; based on the feature residual and the sensitivity matrix, using an iterative optimization algorithm to solve a parameter increment, and synchronously updating all parameter values in the variable physical parameter set based on the parameter increment until the objective function converges to less than a preset threshold.

[0010] In one of the embodiments, the generating a predicted growth rate based on the target parameter set comprises: performing an iterative optimization process at each of the plurality of continuous sampling time points to obtain a target parameter set corresponding to each sampling time point, and obtaining a target layer thickness of a current layer in each target parameter set; for a current sampling time point, obtaining a target layer thickness corresponding to a previous sampling time point adjacent thereto, and determining a time interval between the adjacent sampling time points; calculating a difference value of the target layer thickness and dividing the difference value by the time interval to obtain a predicted growth rate of the current sampling time point. In one of the embodiments, the generating a process adjustment control instruction based on the deviation information comprises: based on the deviation information, calculating a control amount using a preset control algorithm, wherein the preset control algorithm comprises a proportional-integral-derivative control algorithm and / or a model predictive control algorithm; generating a process adjustment control instruction for adjusting an epitaxial growth process parameter according to the control amount, the process parameter comprising a flow parameter of a metal organic precursor and / or a growth temperature parameter.

[0011] In a second aspect, the present application provides a control system for semiconductor epitaxial growth, which is arranged in a control device, and the control device is partially arranged in a semiconductor equipment reaction chamber; the system comprises: an acquisition module configured to acquire reflectance spectrum data; wherein the reflectance spectrum data is generated based on a reflected light signal generated during a semiconductor epitaxial growth process; an analysis module configured to analyze the reflectance spectrum data to obtain an optical feature set; wherein the optical feature set is associated with a layer thickness of an interface transition layer between adjacent material layers, an aluminum component, and an interface change state; a prediction module configured to input the set of optical characteristics into a preset optical transmission model, and iteratively optimize a set of variable physical parameters of the optical transmission model until a preset condition is met, to obtain a target parameter set; a processing module configured to generate a predicted growth rate based on the target parameter set, compare the predicted growth rate with a target growth rate to obtain deviation information, and generate a process adjustment control instruction according to the deviation information, so that the main control system executes the epitaxial growth process according to the adjusted process parameters in the next process cycle. In a third aspect, the present application further provides a control device, part of which is arranged in a semiconductor equipment reaction chamber; the control device comprises: a target light source configured to output a target light signal during the epitaxial growth of the semiconductor; a sensing module, part of which is arranged in the semiconductor equipment reaction chamber, configured to guide the target light signal to the wafer surface in the reaction chamber, and continuously collect a reflected light signal generated based on the target light signal on the wafer surface according to a preset frequency, and generate reflected spectrum data based on the reflected light signal; The control system for epitaxial growth of the semiconductor as described in the second aspect is configured to calculate deviation information of the predicted growth rate and the target growth rate according to the reflected spectrum data, and generate a process adjustment control instruction according to the deviation information, so that the main control system executes the epitaxial growth process according to the adjusted process parameters in the next process cycle.

[0012] In one of the embodiments, the sensing module comprises a fiber probe; the device further comprises a cleaning module, the cleaning module comprising a cooling assembly and a purge assembly; The cooling assembly is configured to actively cool the fiber probe, and the active cooling process comprises water cooling and / or air cooling; The purge assembly comprises a purge gas path configured to output purge gas to the lens end of the fiber probe, so as to reduce the deposition of pollutants on the lens during the collection process.

[0013] In one of the embodiments, the number of the fiber probes is multiple.

[0014] In one of the embodiments, the device further comprises an interface module, the interface module comprising a plurality of industrial standard communication interfaces, and each of the industrial standard communication interfaces is configured to be communicatively connected to the main control system of the epitaxial growth equipment having different communication protocols.

[0015] In a fourth aspect, the present application further provides a computer device. The computer device comprises a memory and a processor, the memory stores a computer program, and the processor implements the method steps in the first aspect when executing the computer program.

[0016] In a fifth aspect, the present application also provides a computer readable storage medium. The computer readable storage medium has a computer program stored thereon, and the computer program, when executed by a processor, implements the method steps in the first aspect.

[0017] The semiconductor epitaxial growth control method, system, control device and related products have at least the following advantages: The present application obtains the reflectance spectrum data in the epitaxial growth process in real time, and analyzes to obtain the optical feature set related to the interface transition layer. Then, the optical feature set is input into a preset optical transfer model, and the variable physical parameter set of the optical transfer model is iteratively optimized to obtain a predicted growth rate. The predicted growth rate is compared with a target growth rate, and deviation information is generated, and then a process adjustment control instruction is output. The main control system performs closed-loop correction on the process parameters in the next process cycle. By using the above scheme, the optical feature set of the present application can dynamically characterize the influence of the layer thickness, composition and interface change state of the interface transition layer on the reflectance spectrum without relying on fixed optical constants and ideal abrupt interface assumptions, thereby improving the inversion accuracy and robustness of the real epitaxial structure state. Further, the process parameters are corrected based on the predicted growth rate obtained by iterative optimization, which can realize timely suppression of growth rate drift and adaptive compensation of process disturbance, reduce the need for manual calibration and intervention, and improve the stability, repeatability and batch consistency of the epitaxial growth process. BRIEF DESCRIPTION OF DRAWINGS

[0018] Figure 1 The structure block diagram of the control device in one embodiment is shown in FIG. 1; Figure 2 The flowchart of the semiconductor epitaxial growth control method in one embodiment is shown in FIG. 2; Figure 3 The flowchart of the step of obtaining the optical feature set in one embodiment is shown in FIG. 3; Figure 4 The flowchart of the step of obtaining the target parameter set in one embodiment is shown in FIG. 4; Figure 5 The structure block diagram of the semiconductor epitaxial growth control system in one embodiment is shown in FIG. 5; Figure 6 The internal structure diagram of the computer device in one embodiment is shown in FIG. 6. DETAILED DESCRIPTION

[0019] Following, the embodiments of the present application are described through specific, concrete examples, and those skilled in the art can easily understand other advantages and effects of the present application from the disclosure of the present specification. The present application can also be implemented or applied through other different specific embodiments, and various modifications or changes can be made to the details in the present specification based on different views and applications without departing from the spirit of the present application. It should be noted that the following examples and features in the examples can be combined with each other without conflict.

[0020] Some example embodiments of the present application are described for illustrative purposes, and it should be understood that the present application can be implemented in other ways not specifically shown in the drawings.

[0021] Please refer to Figure 1 In an example embodiment, the present embodiment provides a control device, which is partially arranged in a semiconductor device reaction chamber. The reaction chamber is provided with a susceptor for placing a wafer, the susceptor is made of high-purity graphite and is supported by a rotating shaft driven by a motor, and can be continuously rotated during semiconductor epitaxial growth. Under the reaction conditions of the reaction chamber, such as the introduction of gas-phase precursors and the provision of appropriate temperature and pressure, the semiconductor material is deposited on the wafer surface in a layer-by-layer direction by epitaxial growth methods such as atomic layer epitaxy (ALE) or vapor phase epitaxy (VPE), to form an epitaxial film with controllable thickness.

[0022] Specifically, the control device includes a target light source, a sensing module, and a control system for semiconductor epitaxial growth.

[0023] The target light source is used to output a target light signal during semiconductor epitaxial growth. In an example embodiment, the target light source is a wide-spectrum halogen tungsten lamp source, which outputs a wavelength range of 600-1100 nm, and is equipped with constant current driving and temperature control to output a stable target light signal.

[0024] The sensing module is used to guide the target light signal to the wafer surface in the reaction chamber, continuously collect the reflected light signal generated based on the target light signal on the wafer surface at a preset frequency, and generate reflected spectrum data based on the reflected light signal. Specifically, the sensing module includes an optical probe and a detector.

[0025] The optical probe is partially arranged in the semiconductor device reaction chamber and is used to receive and conduct the target light signal. In an example embodiment, the optical probe is made of high-purity quartz or sapphire.

[0026] The detector is arranged outside the reaction chamber and is used to continuously collect the reflected light signal at a preset frequency, and then convert the reflected light signal into reflected spectrum data. In an example embodiment, the preset frequency is not less than 1 Hz.

[0027] It should be understood that the above-mentioned sensing module further comprises a light splitting device for splitting the received reflected light signal into a plurality of different specific wave bands. Exemplarily, the light splitting device can adopt a light grating or a filter wheel. At this time, the detector measures the reflected light signal under each wave band synchronously or high-speed switching.

[0028] A control system for semiconductor epitaxial growth, configured to acquire reflectance spectrum data, wherein the reflectance spectrum data is generated based on a reflected light signal generated during the semiconductor epitaxial growth process; analyze the reflectance spectrum data to obtain an optical feature set, wherein the optical feature set is associated with a layer thickness, a composition, and an interface variation state of an interface transition layer located between adjacent material layers; input the optical feature set into a preset optical transfer model, iteratively optimize a variable physical parameter set of the optical transfer model until a preset condition is met, and obtain a predicted growth rate; compare the predicted growth rate with a target growth rate to obtain deviation information; and generate a process adjustment control instruction based on the deviation information, so that a main control system executes an epitaxial growth process according to an adjusted process parameter in a next process cycle.

[0029] Optionally, the above-mentioned control device further comprises a cleaning module, and the cleaning module comprises a cooling assembly and a purge assembly.

[0030] The cooling assembly is configured to actively cool the optical fiber probe, wherein the active cooling process comprises water cooling and / or air cooling.

[0031] The purge assembly comprises a purge gas path configured to output purge gas to the lens end of the optical fiber probe, so as to reduce the deposition of pollutants on the lens during the acquisition process and ensure long-term stable operation in a harsh reaction chamber environment.

[0032] Optionally, the number of the above-mentioned optical fiber probes is multiple, so as to realize synchronous monitoring of different positions of the wafer and monitor uniformity. In the case of multiple optical fiber probes, a switching switch can be configured to sequentially send the acquisition signals of each probe into the detector according to channel polling.

[0033] Optionally, the above-mentioned control device further comprises an interface module, and the interface module comprises a plurality of industrial standard communication interfaces, each of which is configured to be communicatively connected to a main control system of an epitaxial growth device having a different communication protocol. By using the above-mentioned scheme, the control device of the present application has compatibility for multiple models of epitaxial growth devices, avoiding customized development and modification costs caused by inconsistent protocols. Exemplarily, the interface module of the present application supports multiple communication protocols, ensuring low-latency and high-reliability transmission of control instructions; wherein the communication protocols include real-time Ethernet protocols such as EtherCAT and TCP / IP.

[0034] Optionally, the interface module further has safety interlocking logic to prevent damage to the expensive main device from a misoperation. Specifically, the interface module performs a state check, an instruction whitelist detection, a step consistency detection, a handshake confirmation, and the like before sending any process control instruction, to perform an interlocking determination.

[0035] The control device outputs a wide-spectrum target light signal from a target light source, and a sensing module guides the target light signal to a wafer surface in a reaction chamber and continuously collects a reflected light signal at a preset frequency to generate reflected spectrum data, so that in-situ and real-time spectrum monitoring can be achieved during semiconductor epitaxial growth. Further, a control system for semiconductor epitaxial growth acquires reflected spectrum data during epitaxial growth in real time, and analyzes the reflected spectrum data to obtain an optical feature set associated with an interface transition layer. The optical feature set is input into a preset optical transfer model, and variable physical parameter sets of the optical transfer model are iteratively optimized to obtain a predicted growth rate. The predicted growth rate is compared with a target growth rate, and deviation information is generated, and then a process adjustment control instruction is output, so that the main control system performs closed-loop correction on process parameters in the next process cycle. By using the above scheme, the optical feature set of the present application can dynamically characterize the influence of the layer thickness, composition, and interface change state of the interface transition layer on the reflected spectrum without relying on fixed optical constants and ideal abrupt interface assumptions, thereby improving the inversion accuracy and robustness of the real epitaxial structure state. Further, the process parameters are corrected based on the predicted growth rate obtained by iterative optimization, so that the growth rate drift can be timely inhibited and the process disturbance can be adaptively compensated, the need for manual calibration and intervention is reduced, and the stability, repeatability, and batch consistency of the epitaxial growth process are improved.

[0036] Referring to Figure 2 In an exemplary embodiment, the present application provides a control method for semiconductor epitaxial growth, comprising: Step 202: acquiring reflected spectrum data; wherein the reflected spectrum data is generated based on a reflected light signal generated during semiconductor epitaxial growth.

[0037] Specifically, the reflected spectrum data refers to time-series reflected spectrum data composed of multiple frames of reflected spectrum data collected by a detector at a preset frequency, after a wide-spectrum target light signal output from a target light source is incident on a wafer surface in a reaction chamber during semiconductor epitaxial growth, and the reflected light signal reflected by the wafer surface and its epitaxial layer structure returns, and is used to characterize the dynamic change of the reflected spectrum with time during epitaxial growth.

[0038] Step 204: analyzing the reflected spectrum data to obtain an optical feature set; wherein the optical feature set is associated with the layer thickness, composition, and interface change state of the interface transition layer located between adjacent material layers.

[0039] Specifically, the epitaxial growth forms a multi-layer film structure on the wafer surface in the reaction cavity, which mainly includes DBR period layer, cavity layer, transition layer. After the wide spectrum light is incident, reflection and transmission will occur at the interface of each layer, and phase accumulation will occur in the layer. Multiple reflected lights are coherent superposition at the exit end, and finally form a reflection spectrum that changes with wavelength. Therefore, the reflection intensity of each wavelength point in the reflection spectrum contains the relevant information of the interface transition layer.

[0040] Based on the optical characteristic set obtained by analyzing the reflection spectrum, the set of multiple characteristic quantities for characterizing the optical response of the epitaxial structure can reflect the changes of the optical interference and optical resonance characteristics of the epitaxial layer and its interface transition layer with the growth process. Exemplarily, the DBR stop band features include stop band center wavelength and stop band width, which are mainly sensitive to the changes of layer thickness and average refractive index related to composition, and are the basis for reversing the growth rate and average composition; the Fabry-Perot (FP) cavity mode features include center wavelength, cavity mode depth, and cavity mode half-width, which are extremely sensitive to the optical thickness of the cavity, and their depth and shape strongly depend on the interface sharpness of the graded layer and the optical constants (n, k) of the material, which are the key basis for reversing the interface graded thickness and real-time optical constants; the reflectivity at a specific wavelength point provides an absolute intensity calibration reference and contains information about material absorption (k value).

[0041] Exemplarily, the changes in layer thickness and material composition of the interface transition layer, as well as the interface change state such as interface grading degree, diffusion degree or roughness, will cause changes in refractive index and extinction coefficient, and further cause the stop band position and width, the cavity mode position and linearity, and the reflectivity amplitude at the characteristic wavelength in the reflection spectrum to drift or deform. Therefore, the above optical characteristic set can be used as a comprehensive characterization input of the interface transition layer thickness, composition and interface change state for subsequent parameter inversion and growth rate prediction of the optical transmission model. The optical characteristic set includes at least one of the stop band center wavelength, the stop band width, the Fabry-Perot cavity mode center wavelength, the cavity mode depth, the cavity mode half-width, and the absolute reflectivity value at a preset wavelength point of the distributed Bragg reflector. In step 206, the optical characteristic set is input into a preset optical transmission model, and the variable physical parameter set of the optical transmission model is iteratively optimized until a preset condition is met, to obtain a target parameter set.

[0042] Specifically, the optical transmission model refers to a physical model for describing how a wide spectrum target light signal incident to the epitaxial multi-layer structure of a semiconductor forms a reflection spectrum. Exemplarily, the optical transmission model of the present embodiment can use the transfer matrix method (TMM) or the rigorous coupled wave analysis (RCWA).

[0043] The set of variable physical parameters refers to a set of physical quantities that follow the dynamic changes of the process during the inversion process of the optical transmission model, and the target parameter set is the set of variable physical parameters when the preset conditions are met. Illustratively, the set of variable physical parameters in the embodiment includes the layer thickness of the current layer, the layer thickness of the interface transition layer, the composition, the optical constants of the material, and the optical constants include the refractive index n and the extinction coefficient k.

[0044] Under the constraint of a given set of variable physical parameters, the optical transmission model performs forward calculation based on the set of variable physical parameters, and outputs the corresponding predicted reflectance spectrum.

[0045] In step 208, a predicted growth rate is generated based on the target parameter set; a deviation information is obtained by comparing the predicted growth rate and the target growth rate; and a process adjustment control instruction is generated according to the deviation information, so that the main control system executes the epitaxial growth process according to the adjusted process parameters in the next process cycle.

[0046] Specifically, the growth rate is calculated based on the layer thickness of the current layer in the target parameter set, which represents the epitaxial thickness increment grown by the semiconductor per unit time. Generally speaking, in the process of semiconductor epitaxial growth, it is difficult to directly measure key process parameters such as precursor effective supply and surface reaction kinetics state, but they are all reflected in the growth rate; in addition, the rate deviation can be directly mapped to the input variable of the process parameter. Based on this, the growth rate is taken as the real-time index of the closed-loop feedback in the present application, which can quickly and effectively correct the deviation and improve the calibration accuracy.

[0047] Generally speaking, the target growth rate is determined based on the process recipe target and the process capability of the semiconductor equipment.

[0048] The main control system refers to the main control platform of the epitaxial growth equipment, which is used to execute the process recipe and control the execution units such as mass flow controller, heater, pressure control valve, etc., to complete the epitaxial growth process.

[0049] The control method for epitaxial growth of the semiconductor, real-time acquisition of reflection spectrum data in the epitaxial growth process, analysis of an optical feature set related to the interface transition layer, input of the optical feature set into a preset optical transfer model, iterative optimization of a variable physical parameter set of the optical transfer model, obtaining of a predicted growth rate, comparison of the predicted growth rate with a target growth rate, generation of deviation information, and output of a process adjustment control instruction to enable the main control system to perform closed-loop correction on the process parameters in the next process cycle. With the above scheme, the optical feature set of the present application can dynamically characterize the influence of the layer thickness, composition, and interface variation state of the interface transition layer on the reflection spectrum without relying on fixed optical constants and ideal abrupt interface assumptions, thereby improving the inversion accuracy and robustness of the real epitaxial structure state. Further, the process parameters are corrected based on the predicted growth rate obtained through iterative optimization, which can realize timely suppression of the growth rate drift and adaptive compensation of the process disturbance, reduce the need for manual calibration and intervention, and improve the stability, repeatability, and batch consistency of the epitaxial growth process.

[0050] Referring to Figure 3 Optionally, in the case that the optical feature set includes at least one of the stop band center wavelength, the stop band width, the Fabry-Perot cavity mode center wavelength, the cavity mode depth, the cavity mode full width at half maximum, and the absolute reflectivity value at the preset wavelength point of the distributed Bragg reflector, the reflection spectrum data is analyzed to obtain the optical feature set, including: Step 302, spectral feature recognition is performed on the reflection spectrum data to determine the stop band of the distributed Bragg reflector, and the center wavelength and / or the stop band width of the stop band are extracted.

[0051] Step 304, and / or, cavity mode feature extraction is performed on the reflection spectrum data to determine the Fabry-Perot cavity mode, and the center wavelength, the cavity mode depth, and / or the cavity mode full width at half maximum of the Fabry-Perot cavity mode are extracted.

[0052] Step 306, and / or, at least one preset wavelength point is selected in the reflection spectrum data, and the absolute reflectivity value at the preset wavelength point is extracted.

[0053] Optionally, spectral feature recognition is performed on the reflection spectrum data to determine the stop band of the distributed Bragg reflector, and the center wavelength and / or the stop band width of the stop band are extracted, including: The reflection spectrum data is preprocessed, wherein the preprocessing steps include smoothing filtering to suppress random noise and baseline correction to reduce the influence of the light source spectrum and link attenuation on the overall envelope of the spectrum. Optionally, the smoothing filtering includes sliding average filtering, Savitzky-Golay filtering, or median filtering.

[0054] In a preset wavelength search interval, a maximum reflectivity Rmax and an average reflectivity Rref of the preprocessed reflectance spectrum data are calculated, wherein the average reflectivity is an average of both ends of the wavelength search interval. A stop band determination threshold is constructed according to the following formula: Rth = Rref + a x (Rmax - Rref), wherein a is a preset proportion coefficient, which is usually set to 0.5-0.8. A wavelength range with a reflectivity greater than the stop band determination threshold and a continuous coverage greater than a preset minimum span is determined as a candidate stop band.

[0055] For the candidate stop band wavelength range, two wavelengths corresponding to the intersection of the reflectivity curve and the stop band determination threshold are determined as the short-wavelength boundary wavelength λL1 and the long-wavelength boundary wavelength λH1 of the stop band. When there are multiple candidate stop band wavelength ranges, the wavelength range that satisfies the highest average reflectivity or the largest span can be selected as the stop band. The center wavelength of the stop band is calculated as: λC = (λL1 + λH1) / 2; and the stop band width is calculated as: Δλ = λH1 - λL1.

[0056] Optionally, cavity mode features are extracted from the reflectance spectrum data to determine the Fabry-Perot cavity mode and extract the center wavelength, cavity mode depth and / or cavity mode full width at half maximum of the Fabry-Perot cavity mode, including: After identifying the DBR stop band, a preset sub-interval in the stop band interval is set as a cavity mode search interval. In the cavity mode search interval, extreme value detection is performed on the preprocessed reflectance spectrum data to locate a local peak or a local valley as a cavity mode candidate point. Exemplarily, the first derivative zero-crossing and second derivative sign discrimination are used to identify the local extreme point. It should be noted that the above-mentioned identification of the DBR stop band and the preprocessing of the reflectance spectrum data both use the method steps provided in the above-mentioned embodiments, and will not be described here to save space.

[0057] The candidate point that satisfies the preset cavity mode determination condition is selected as the FP cavity mode center wavelength λcav; wherein the preset cavity mode determination condition includes that the candidate point is located in the cavity mode search interval, the spectral shape width of the candidate feature falls within a preset range, and the reflectivity difference of the candidate point relative to the surrounding baseline satisfies a preset amplitude threshold.

[0058] A preset wavelength window is selected on both sides of the center wavelength, and the average value of the reflectivity in the window is calculated as the baseline reflectivity Rbase. The reflectivity at the center wavelength R(λcav) is taken, and then the cavity mode depth D = Rbase - R(λcav) is calculated. Based on the cavity mode depth and the baseline reflectivity, the half-height level is constructed as: Rhalf = Rbase - D / 2; the wavelength positions that satisfy the reflectivity equal to Rhalf are searched on both sides of the center wavelength to obtain the left intersection wavelength λL2 and the right intersection wavelength λH2; and the cavity mode full width at half maximum is calculated as: W = λH2 - λL2.

[0059] Optionally, at least one preset wavelength point is selected in the reflection spectrum data, and an absolute reflectivity value at the preset wavelength point is extracted, including: According to the design center wavelength of the to-be-monitored epitaxial structure, the DBR stop band position, and the optimal interval of the signal-to-noise ratio of the spectrometer, one or more preset wavelength points λp are selected in advance. Exemplarily, each preset wavelength point can be arranged in the DBR stop band, near the edge of the stop band, and near the cavity mode, so as to improve the sensitivity to the changes in the layer thickness, the composition, or the interface state.

[0060] The reflection spectrum data includes the measured light intensity data of multiple wavelength channels, the measured light intensity at the preset wavelength point is normalized with the reference light intensity at the same wavelength point of the reference spectrum, to obtain the absolute reflectivity value R(λp, t) at the preset wavelength point; wherein the reference spectrum is obtained by collecting under the condition of a standard reflector or a known reflectivity reference sheet.

[0061] By using the above scheme, the key information in the original reflection spectrum can be structured and expressed without adding additional hardware sensors, by performing hierarchical spectrum analysis on the reflection spectrum data and extracting multiple types of optical feature sets with clear physical meanings. Further, the optical feature set can be quickly and accurately locked and quantified from the complex reflection spectrum, to improve the real-time performance and repeatability of feature extraction. Specifically, the center wavelength and / or the stop band width of the extracted DBR stop band can represent the reflection band drift and deformation caused by the changes in the DBR period thickness, the refractive index contrast, and the interface quality, so as to facilitate the identification of process abnormal trends such as composition drift and interface gradual change; the center wavelength, the cavity mode depth, and / or the cavity mode half-height width of the extracted FP cavity mode can represent the changes in the cavity length and the cavity loss, and has higher resolution capability for the changes in the epitaxial layer thickness evolution, scattering enhancement, and interface state; the absolute reflectivity value extracted at the preset wavelength point can form a fast monitoring index for the spectrum amplitude, and can be used to reflect the link attenuation, the overall change in the reflection intensity, and the local optical response change. The multi-dimensional optical feature set thus formed can significantly improve the observability and robustness of the subsequent optical transmission model inversion, reduce the system error caused by the fitting of only fixed model parameters, provide more stable and quantifiable inputs for the growth rate prediction and closed-loop process adjustment, and thus improve the epitaxial growth process monitoring precision, the process consistency, and the yield.

[0062] Please refer to Figure 4 Optionally, in the case that the set of variable physical parameters includes the layer thickness of the current layer, the layer thickness of the interface transition layer, the composition, and the optical constant of the material, the optical constant including the refractive index and the extinction coefficient, the optical feature set is input into a preset optical transmission model, the set of variable physical parameters of the optical transmission model is iteratively optimized until a preset condition is met, to obtain a target parameter set, including: Step 402, constructing an objective function.

[0063] Step 404, performing forward calculation on the preset optical transmission model according to the current variable physical parameter set to obtain predicted reflectance spectrum data.

[0064] Step 406, analyzing the predicted reflectance spectrum data to obtain a predicted feature set, calculating the difference between the predicted feature set and the optical feature set according to the objective function, and iteratively optimizing the current variable physical parameter set according to the difference until a preset condition is met, and taking the variable physical parameter set at this time as the target parameter set.

[0065] Specifically, the variable physical parameter set to be inverted θ = [d, x, σ, n, k], wherein d is the layer thickness of the current layer, σ is the layer thickness of the interface transition layer, x is the composition, n is the refractive index, and k is the extinction coefficient. The optical feature set extracted by the method of steps 302-306 is denoted as F_meas.

[0066] Optionally, the objective function is the square of the two-norm of the difference between the predicted feature set and the optical feature set, and its expression is: L(θ) = ||F_pred(θ) - F_meas||², wherein F_pred(θ) is the predicted feature set.

[0067] Optionally, calculating the difference between the predicted feature set and the optical feature set according to the objective function, and iteratively optimizing the current variable physical parameter set according to the difference until a preset condition is met, includes: Using the objective function, the feature residual is obtained by differentiating the predicted feature set and the optical feature set, and the sensitivity matrix of the feature residual with respect to the current variable physical parameter set is calculated; based on the feature residual and the sensitivity matrix, the parameter increment is solved by using the iterative optimization algorithm, and all parameter values in the variable physical parameter set are updated synchronously based on the parameter increment until the objective function converges to less than a preset threshold.

[0068] Specifically, the purpose of iteration is to find a set of variable physical parameters θ such that the above objective function L(θ) converges to meet the preset condition, and in this embodiment, the preset condition is that it converges to less than a preset threshold. Illustratively, the iterative optimization algorithm in this embodiment can optionally use a nonlinear least squares optimization algorithm (Levenberg–Marquardt, L-M) or a Bayesian optimization algorithm based on a proxy model. The following takes the L-M algorithm as an example to illustrate the iterative optimization algorithm.

[0069] In each iteration, a forward calculation is performed by the optical transfer model based on the current variable physical parameter set θ to obtain predicted reflectance spectrum data. The predicted feature set F_pred(θ) is extracted from the predicted reflectance spectrum data using the method of steps 302-306. The difference between the predicted feature set F_pred(θ) and the optical feature set F_meas is calculated to obtain the feature residual δF, which is taken as the reference residual vector.

[0070] A preset perturbation amount is applied to each parameter in the current variable physical parameter set to form a corresponding perturbed parameter, and only one parameter is changed at each perturbation, and the remaining parameters remain unchanged as reference values. For each perturbed parameter, the above forward calculation and feature extraction are repeatedly performed to obtain a perturbed predicted feature set, the difference between the perturbed predicted feature set and the reference residual vector is calculated, and the difference is ratio-operated with the perturbation amount of the parameter to obtain the influence degree of the parameter change on each feature residual. The influence degree is taken as a row or a column of the sensitivity matrix, and finally the sensitivity matrix J of all parameters in the variable physical parameter set is obtained.

[0071] Solving the iterative optimization algorithm equation to obtain the parameter increment Δθ, and adjusting all parameters in the variable physical parameter set θ according to the parameter increment. Wherein, T is a transpose, μ is a damping coefficient, and I is a unit matrix.

[0072] The above steps are repeated until the absolute value of the difference between the objective function values of two adjacent iterations is less than a preset threshold, and it is considered that the preset condition is met. The value of the variable physical parameter set at this time is taken as the value of the target parameter set.

[0073] Optionally, the application supports an online learning mode, which can fine-tune the initial parameters of the optical transfer model based on historical successful inversion data to realize self-optimization.

[0074] By using the above scheme, the difference between the measured optical feature set and the predicted feature set calculated based on the optical transfer model is constructed as an objective function, so that the epitaxial structure parameter inversion process is converted from an empirical parameter adjustment to a quantifiable and convergent optimization solving process. Further, by iteratively calculating the feature residual and its sensitivity relationship with respect to the variable physical parameter set, and simultaneously updating multiple coupled parameters in the variable physical parameter set, adaptive correction can be realized under process dynamic drift and non-ideal interface conditions, systematic errors caused by fixed parameter models are reduced, and the accuracy and robustness of the inversion result are improved. At the same time, the target parameter set obtained by optimization has a clear physical meaning and can be used to output more reliable growth rates in real time, providing stable input for subsequent closed-loop control, thereby improving the stability, repeatability and batch consistency of the epitaxial growth process, and reducing the cost of manual calibration and downtime maintenance.

[0075] Optionally, the predicted growth rate is generated based on the target parameter set, comprising: The iterative optimization process is performed at each of the plurality of continuous sampling time points to obtain a target parameter set corresponding to each sampling time point, and a target layer thickness of a current layer in each target parameter set is obtained; for a current sampling time point, a target layer thickness corresponding to a previous sampling time point adjacent to the current sampling time point is obtained, and a time interval between the adjacent sampling time points is determined; a difference value of the target layer thickness is calculated and divided by the time interval to obtain a predicted growth rate of the current sampling time point.

[0076] Specifically, the target parameter set obtained by the inversion is {d_opt, x_opt,σ_opt, n_opt, k_opt}. The expression for calculating the predicted growth rate is: r(t) = (d_opt(t) - d_opt(t-Δt)) / Δt, where d_opt(t) and d_opt(t-Δt) are the target layer thicknesses of the current layer at the current sampling time point and the previous sampling time point, respectively; and Δt is the time interval. It should be understood that the predicted growth rate is an instantaneous value.

[0077] With the above scheme, since the layer thickness is obtained by inversion under the condition of simultaneously considering the composition, interface gradual change and material optical constant change, the predicted growth rate calculated based on the layer thickness has stronger robustness and consistency with respect to process drift, non-ideal interface and optical constant change, and can realize real-time and continuous tracking of the growth rate, thereby providing more reliable feedback basis for subsequent closed-loop adjustment of process parameters such as precursor flow, temperature or pressure, and improving the epitaxial thickness control precision and batch-to-batch repeatability.

[0078] Optionally, the process adjustment control instruction is generated according to the deviation information, comprising: Based on the deviation information, a control amount is calculated by using a preset control algorithm, wherein the preset control algorithm includes a proportional-integral-derivative control algorithm and / or a model predictive control algorithm; The process adjustment control instruction for adjusting the epitaxial growth process parameters is generated according to the control amount, and the process parameters include the flow parameter of the metal organic precursor and / or the growth temperature parameter. Specifically, in the case where the preset control algorithm includes a proportional-integral-derivative control algorithm, the proportional-integral-derivative control algorithm calculates the control amount according to the current value, cumulative value and change trend of the deviation information; in the case where the preset control algorithm includes a model predictive control algorithm, the model predictive control algorithm predicts the future growth rate according to the deviation information and process state quantity and solves the control amount under the constraint condition. The constraint condition refers to the safety boundary of the process and equipment, and the control amount is a parameter increment and / or a target set value for adjusting the process parameter.

[0079] The main control system receives the process adjustment control instruction, adjusts the process parameters according to the control quantity, and executes the epitaxial growth process according to the adjusted process parameters in the next process cycle. Exemplarily, the metal organic precursor includes trimethylaluminum (TMAl) or trimethylgallium (TMGa).

[0080] By converting the growth rate deviation into an executable control quantity, the above-mentioned scheme realizes closed-loop feedback adjustment of the epitaxial growth process, can timely correct the deviation under conditions such as source bottle efficiency attenuation, mass transfer condition fluctuation, and cavity state drift, suppresses the accumulation of the rate deviation, improves the layer thickness control precision and batch-to-batch consistency, and further, the control instruction can directly act on the metal organic precursor flow parameter and / or the growth temperature parameter, so that the process adjustment path is clear and the execution efficiency is high, thereby improving the real-time controllability of epitaxial growth and reducing the probability of manual intervention and rework.

[0081] The above-mentioned control method of semiconductor epitaxial growth, real-time acquisition of reflectance spectrum data in the epitaxial growth process, and analysis of the optical feature set related to the interface transition layer, and then inputting the optical feature set into the preset optical transfer model to iteratively optimize the variable physical parameter set of the optical transfer model to obtain the predicted growth rate; comparing the predicted growth rate with the target growth rate and generating deviation information, and then outputting a process adjustment control instruction to make the main control system perform closed-loop correction on the process parameters in the next process cycle. By using the above-mentioned scheme, the optical feature set of the present application can dynamically characterize the influence of the layer thickness, composition, and interface change state of the interface transition layer on the reflectance spectrum without relying on fixed optical constants and ideal abrupt interface assumptions, thereby improving the inversion accuracy and robustness of the real epitaxial structure state. Further, based on the predicted growth rate obtained by iterative optimization, the process parameters are corrected in a closed loop, which can timely suppress the growth rate drift and adaptively compensate for process disturbances, reduce the need for manual calibration and intervention, and improve the stability, repeatability, and batch-to-batch consistency of the epitaxial growth process.

[0082] It should be understood that, although each step in the flowchart involved in each embodiment as described above is displayed in sequence according to the arrow, these steps are not necessarily executed in the order indicated by the arrow. Unless otherwise specified herein, the execution of these steps is not strictly limited in sequence, and these steps can be executed in other orders. Moreover, at least part of the steps in the flowchart involved in each embodiment as described above can include multiple steps or stages, which are not necessarily executed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily sequential, but can be alternately or alternately executed with at least part of other steps or steps or stages in other steps.

[0083] Based on the same inventive concept, the embodiments of the present application also provide a control system for semiconductor epitaxial growth, which is suitable for the control method for semiconductor epitaxial growth. The implementation scheme for solving the problem provided by the system is similar to the implementation scheme described in the above method. Therefore, the specific limitations in one or more system embodiments provided below can refer to the limitations of the method described above, and will not be repeated here.

[0084] Please refer to Figure 5 In one embodiment, a control system for semiconductor epitaxial growth is provided, which is arranged in a control device, and the control device is partially arranged in a semiconductor device reaction cavity. The control system for semiconductor epitaxial growth comprises an acquisition module, an analysis module, a prediction module and a processing module.

[0085] The acquisition module is configured to acquire reflectance spectrum data, wherein the reflectance spectrum data is generated based on a reflected light signal generated during the semiconductor epitaxial growth process.

[0086] The analysis module is configured to analyze the reflectance spectrum data to obtain an optical feature set, wherein the optical feature set is associated with a layer thickness of an interface transition layer between adjacent material layers, an aluminum component and an interface change state.

[0087] The prediction module is configured to input the optical feature set into a preset optical transfer model, iteratively optimize a variable physical parameter set of the optical transfer model until a preset condition is met, and obtain a target parameter set.

[0088] The processing module is configured to generate a predicted growth rate based on the target parameter set, compare the predicted growth rate with a target growth rate to obtain deviation information, and generate a process adjustment control instruction according to the deviation information, so that the main control system executes the epitaxial growth process according to the adjusted process parameters in the next process cycle.

[0089] Optionally, in the case where the optical feature set includes at least one of a stop band center wavelength, a stop band width, a Fabry-Perot cavity mode center wavelength, a cavity mode depth, a cavity mode full width at half maximum, and an absolute reflectance value at a preset wavelength point of a distributed Bragg reflector, the analysis module analyzes the reflectance spectrum data to obtain the optical feature set, including: performing spectral feature recognition on the reflectance spectrum data to determine the stop band of the distributed Bragg reflector, and extracting the center wavelength and / or the stop band width of the stop band; and / or performing cavity mode feature extraction on the reflectance spectrum data to determine the Fabry-Perot cavity mode, and extracting the center wavelength, the cavity mode depth and / or the cavity mode full width at half maximum of the Fabry-Perot cavity mode; and / or selecting at least one preset wavelength point in the reflectance spectrum data, and extracting the absolute reflectance value at the preset wavelength point.

[0090] Optionally, in the case that the set of variable physical parameters includes the layer thickness of the current layer, the layer thickness of the interface transition layer, the composition, the optical constants of the material, the optical constants including the refractive index and the extinction coefficient, the prediction module inputs the set of optical characteristics into a preset optical transfer model, iteratively optimizes the set of variable physical parameters of the optical transfer model until a preset condition is met, and obtains a target parameter set, including: constructing a target function; performing forward calculation on the preset optical transfer model according to the current set of variable physical parameters to obtain predicted reflectance spectrum data; analyzing the predicted reflectance spectrum data to obtain a predicted characteristic set, calculating the difference between the predicted characteristic set and the optical characteristic set according to the target function, and iteratively optimizing the current set of variable physical parameters according to the difference until the preset condition is met, and taking the set of variable physical parameters at this time as the target parameter set.

[0091] Optionally, the prediction module calculates the difference between the predicted characteristic set and the optical characteristic set according to the target function, iteratively optimizes the current set of variable physical parameters according to the difference until the preset condition is met, including: using the target function to difference the predicted characteristic set and the optical characteristic set to obtain a characteristic residual; and calculating a sensitivity matrix of the characteristic residual with respect to the current set of variable physical parameters; based on the characteristic residual and the sensitivity matrix, using an iterative optimization algorithm to solve a parameter increment, and synchronously updating all parameter values in the set of variable physical parameters based on the parameter increment until the target function converges to less than a preset threshold.

[0092] Optionally, the processing module generates a predicted growth rate based on the target parameter set, including: performing an iterative optimization process at each of a plurality of consecutive sampling times to obtain a target parameter set corresponding to each sampling time, and obtaining a target layer thickness of the current layer in each target parameter set; for the current sampling time, obtaining a target layer thickness corresponding to the previous sampling time adjacent thereto, and determining a time interval between the adjacent sampling times; calculating the difference of the target layer thickness and dividing by the time interval to obtain the predicted growth rate of the current sampling time.

[0093] Optionally, the processing module generates a process adjustment control instruction according to the deviation information, including: based on the deviation information, a control amount is calculated using a preset control algorithm, wherein the preset control algorithm includes a proportional-integral-derivative control algorithm and / or a model predictive control algorithm; generating a process adjustment control instruction for adjusting the epitaxial growth process parameters according to the control amount, the process parameters including the flow parameter of the metal organic precursor and / or the growth temperature parameter.

[0094] The control system of the semiconductor epitaxial growth can obtain reflection spectrum data in real time during the epitaxial growth process, analyze and obtain a set of optical characteristics related to the interface transition layer, input the set of optical characteristics into a preset optical transfer model, iteratively optimize a set of variable physical parameters of the optical transfer model, and obtain a predicted growth rate. The predicted growth rate is compared with a target growth rate, deviation information is generated, and a process adjustment control instruction is output. The main control system performs closed-loop correction on the process parameters in the next process cycle. According to the above scheme, the set of optical characteristics can dynamically characterize the influence of the layer thickness, composition and interface change state of the interface transition layer on the reflection spectrum without relying on fixed optical constants and ideal abrupt interface assumptions, thereby improving the inversion accuracy and robustness of the real epitaxial structure state. Further, the process parameters are corrected based on the predicted growth rate obtained by iterative optimization, which can realize timely suppression of the growth rate drift and adaptive compensation of the process disturbance, reduce the need for manual calibration and intervention, and improve the stability, repeatability and batch consistency of the epitaxial growth process. Each module in the control system of the semiconductor epitaxial growth can be realized by software, hardware and a combination thereof in whole or in part. Each module can be embedded in or independent of a processor in a computer device in hardware form, or can be stored in a memory in the computer device in software form, so as to be called and executed by the processor to perform the operations corresponding to each module.

[0095] In a feasible embodiment, a computer device, which can be a terminal, has an internal structure diagram as shown in Figure 6The computer device includes a processor, a memory, an input / output interface, a communication interface, a display unit and an input device. The processor, the memory and the input / output interface are connected through a system bus. The communication interface, the display unit and the input device are connected to the system bus through the input / output interface. The processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for running the operating system and the computer program in the non-volatile storage medium. The input / output interface of the computer device is configured to exchange information between the processor and external devices. The communication interface of the computer device is configured to perform wired or wireless communication with external terminals. The wireless communication can be achieved through WIFI, mobile cellular network, NFC (Near Field Communication) or other technologies. The computer program is executed by the processor to implement the control method for semiconductor epitaxial growth. The display unit of the computer device is configured to form a visually visible picture, which can be a display screen, a projection device or a virtual reality imaging device. The display screen can be a liquid crystal display screen or an electronic ink display screen. The input device of the computer device can be a touch layer overlaid on the display screen, or a key, trackball or touchpad arranged on the shell of the computer device, or an external keyboard, touchpad or mouse, etc.

[0096] Those skilled in the art can understand that, Figure 6 The structure shown in the figure is only a block diagram of part of the structure related to the scheme of the present application, and does not constitute a limitation on the computer device to which the scheme of the present application is applied. The specific computer device can include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0097] In a feasible embodiment, a computer device is provided, which includes a memory and a processor. The memory stores a computer program. The processor executes the computer program to implement the method steps in the control method for semiconductor epitaxial growth.

[0098] In a feasible embodiment, a computer readable storage medium is provided, which stores a computer program. The computer program is executed by a processor to implement the method steps in the control method for semiconductor epitaxial growth.

[0099] In a feasible embodiment, a computer program product is provided, which includes a computer program. The computer program is executed by a processor to implement the method steps in the control method for semiconductor epitaxial growth.

[0100] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a non-volatile computer readable storage medium, and when the computer program is executed, the processes of the above-mentioned embodiments of the methods can be included. Any reference to memory, database or other medium used in the embodiments provided in the present application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical storage, high-density embedded non-volatile memory, resistive memory (ReRAM), magnetoresistive random access memory (MRAM), ferroelectric memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. As an illustration but not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM), etc. The database involved in the embodiments provided in the present application can include at least one of a relational database and a non-relational database. The non-relational database can include a distributed database based on a block chain, etc., without being limited thereto. The processor involved in the embodiments provided in the present application can be a general-purpose processor, a central processing unit, a graphics processing unit, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, etc., without being limited thereto.

[0101] Any combination of the technical features of the above embodiments can be made. In order to make the description simple, all possible combinations of the technical features in the above embodiments are not described, however, as long as the combination of the technical features does not exist, it should be considered as the scope of the present application.

[0102] The above embodiments only express several implementation manners of the present application, and the description is more specific and detailed, but it should not be understood as a limitation on the scope of the patent of the present application. It should be pointed out that for ordinary skilled in the art, without departing from the concept of the present application, a number of modifications and improvements can be made, which are all within the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the appended claims.

Claims

1. A method of controlling the epitaxial growth of a semiconductor, characterized by, The method comprises: acquiring reflectance spectrum data; wherein the reflectance spectrum data is generated based on a reflected light signal generated in a semiconductor epitaxial growth process; analyzing the reflectance spectrum data to obtain an optical feature set; wherein the optical feature set comprises at least one of a stop band center wavelength, a stop band width, a Fabry-Perot cavity mode center wavelength, a cavity mode depth, a cavity mode full width at half maximum, and an absolute reflectance value at a preset wavelength point of a distributed Bragg reflector; inputting the optical feature set into a preset optical transfer model, iteratively optimizing a variable physical parameter set of the optical transfer model until a preset condition is met, to obtain a target parameter set; generating a predicted growth rate based on the target parameter set; comparing the predicted growth rate with a target growth rate to obtain deviation information; generating a process adjustment control instruction according to the deviation information, so that a main control system executes an epitaxial growth process according to an adjusted process parameter in a next process cycle.

2. The method of claim 1, wherein, The analysis of the reflectance spectrum data to obtain an optical feature set comprises: performing spectral feature recognition on the reflectance spectrum data to determine a stop band of a distributed Bragg reflector, and extracting a center wavelength and / or a stop band width of the stop band; and / or, performing cavity mode feature extraction on the reflectance spectrum data to determine a Fabry-Perot cavity mode, and extracting a center wavelength, a cavity mode depth, and / or a cavity mode full width at half maximum of the Fabry-Perot cavity mode; and / or, selecting at least one preset wavelength point in the reflectance spectrum data, and extracting an absolute reflectance value at the preset wavelength point.

3. The method of claim 2, wherein, The variable physical parameter set comprises a layer thickness of a current layer, a layer thickness, a component, and an optical constant of an interface transition layer, the optical constant comprising a refractive index and an extinction coefficient; The inputting of the optical feature set into a preset optical transfer model, the iterative optimization of a variable physical parameter set of the optical transfer model until a preset condition is met to obtain a target parameter set comprises: constructing an objective function; performing forward calculation on the preset optical transfer model according to the current variable physical parameter set to obtain predicted reflectance spectrum data; analyzing the predicted reflectance spectrum data to obtain a predicted feature set, calculating a difference between the predicted feature set and the optical feature set according to the objective function, iteratively optimizing the current variable physical parameter set according to the difference until a preset condition is met, and taking the variable physical parameter set at this time as the target parameter set.

4. The method of claim 3, wherein, The objective function is a square of a two-norm difference between the predicted feature set and the optical feature set; The calculation of a difference between the predicted feature set and the optical feature set according to the objective function, and the iterative optimization of the current variable physical parameter set according to the difference until a preset condition is met comprises: performing difference between the predicted feature set and the optical feature set using the objective function to obtain a feature residual, and calculating a sensitivity matrix of the feature residual with respect to the current variable physical parameter set; Based on the feature residual and the sensitivity matrix, a parameter increment is solved by using an iterative optimization algorithm, and all parameter values in the set of variable physical parameters are synchronously updated based on the parameter increment until the objective function converges to be less than a preset threshold.

5. The method of claim 1, wherein, The generating a predicted growth rate based on the target parameter set comprises: The iterative optimization process is performed at each of the plurality of continuous sampling time points to obtain a target parameter set corresponding to each sampling time point, and a target layer thickness of a current layer in each target parameter set is obtained; For a current sampling time point, a target layer thickness corresponding to a previous sampling time point adjacent to the current sampling time point is obtained, and a time interval between the adjacent sampling time points is determined; A difference value of the target layer thickness is calculated and divided by the time interval to obtain a predicted growth rate of the current sampling time point.

6. The method of claim 1, wherein, The generating a process adjustment control instruction according to the deviation information comprises: Based on the deviation information, a control amount is calculated by using a preset control algorithm, wherein the preset control algorithm comprises a proportional-integral-derivative control algorithm and / or a model predictive control algorithm; A process adjustment control instruction for adjusting an epitaxial growth process parameter is generated according to the control amount, and the process parameter comprises a flow parameter of a metal organic precursor and / or a growth temperature parameter.

7. A control system for the epitaxial growth of semiconductors, characterised in that The control device is arranged in the control device, and the control device is partially arranged in the semiconductor equipment reaction cavity; the system comprises: An acquisition module is configured to acquire reflectance spectrum data, wherein the reflectance spectrum data is generated based on a reflected light signal generated during an epitaxial growth process of a semiconductor; An analysis module is configured to analyze the reflectance spectrum data to obtain an optical feature set, wherein the optical feature set is associated with a layer thickness of an interface transition layer between adjacent material layers, an aluminum component, and an interface change state; A prediction module is configured to input the optical feature set into a preset optical transmission model, iteratively optimize a set of variable physical parameters of the optical transmission model, and obtain a target parameter set until a preset condition is met; A processing module is configured to generate a predicted growth rate based on the target parameter set, compare the predicted growth rate with a target growth rate to obtain deviation information, and generate a process adjustment control instruction according to the deviation information, so that a main control system executes an epitaxial growth process according to adjusted process parameters in a next process cycle.

8. A control device characterized by comprising: Part of the control device is arranged in the semiconductor equipment reaction cavity, and the control device comprises: A target light source is configured to output a target light signal during an epitaxial growth process of a semiconductor; A sensing module is partially arranged in the semiconductor equipment reaction cavity and is configured to guide the target light signal to a wafer surface in the reaction cavity, and continuously collect a reflected light signal generated based on the target light signal on the wafer surface at a preset frequency, and generate reflectance spectrum data based on the reflected light signal; The control system for epitaxial growth of a semiconductor as claimed in claim 7 is configured to calculate deviation information of a predicted growth rate and a target growth rate based on the reflectance spectrum data, and generate a process adjustment control instruction according to the deviation information, so that a main control system executes an epitaxial growth process according to adjusted process parameters in a next process cycle.

9. The control device of claim 8, wherein The sensing module comprises a fiber probe; the device further comprises a cleaning module, the cleaning module comprising a cooling assembly and a purge assembly; The cooling assembly is configured to actively cool the fiber probe, and the active cooling process comprises water cooling and / or air cooling; The purge assembly comprises a purge gas path configured to output purge gas to the lens end of the fiber probe to reduce the deposition of contaminants on the lens during the acquisition process.

10. The control device of claim 9, wherein The number of the fiber probes is multiple.

11. The control device of claim 8, wherein The device further comprises an interface module, the interface module comprising multiple industrial standard communication interfaces, each of the industrial standard communication interfaces being configured to be communicatively connected to a main control system of an epitaxial growth apparatus having a different communication protocol.

12. A computer device comprising a memory and a processor, the memory storing a computer program, characterized in that, The processor, when executing the computer program, implements the steps of the method of any one of claims 1-6.

13. A computer readable storage medium having stored thereon a computer program, characterized in that The computer program, when executed by the processor, implements the steps of the method of any one of claims 1-6.

Images