Epitaxial control method for DFB laser chip

By utilizing TACD simulation tools and reverse simulation technology during the epitaxial growth process of DFB laser chips, growth intervals were divided and growth parameters were monitored and adjusted in real time. This solved the problems of insufficient precision and uniformity in epitaxial growth control, and improved the quality of the epitaxial layer and the performance of the DFB laser chip.

CN118979302AActive Publication Date: 2024-11-19JIANGSU ETERN
View PDF 5 Cites 0 Cited by

Patent Information

Application Number
CN202411469613.X
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-10-21
Publication Date
2024-11-19
Estimated Expiration
2044-10-21

AI Technical Summary

Technical Problem

Existing epitaxial growth control methods for DFB laser chips cannot quickly and accurately adjust growth control parameters according to the growth state of the epitaxial layer during wafer epitaxial growth, resulting in poor accuracy and uniformity of epitaxial growth control and affecting the quality of epitaxial layer growth.

Method used

By reading wafer property information and expected performance indicators of DFB laser chips, reverse simulation is performed using TACD simulation tools to determine epitaxial growth control parameters. The growth cycle is divided into multiple growth intervals, and growth indicators are monitored and adjusted in real time, iterating until the predetermined growth cycle is completed.

Benefits of technology

This technology enables rapid and precise adjustment of growth control parameters based on the growth state of the epitaxial layer on the wafer, improving the growth quality of the epitaxial layer and ensuring the performance stability and consistency of the DFB laser chip.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118979302A_ABST
    Figure CN118979302A_ABST
Patent Text Reader

Abstract

The invention provides an epitaxy control method for a DFB laser chip, and relates to the technical field of wafer epitaxy growth, and the method comprises the steps: carrying out the reverse simulation under the constraint of meeting an expected epitaxy attribute characteristic and a preset epitaxy growth period, and determining an epitaxy growth control parameter; dividing and determining a plurality of growth intervals, and performing growth index setting on the plurality of growth intervals according to the expected epitaxial attribute characteristics to obtain a plurality of expected growth indexes; and executing epitaxial growth control according to the epitaxial growth control parameters, the plurality of growth intervals and the plurality of expected growth indexes. According to the method, the technical problem that the growth quality of the epitaxial layer is affected due to poor epitaxial growth control accuracy and uniformity caused by the fact that growth control parameters cannot be quickly and accurately adjusted according to the growth state of the epitaxial layer of the wafer in the epitaxial growth process of the wafer in the prior art can be solved, the growth quality of the epitaxial layer can be improved, and the yield of the epitaxial layer is improved. And the stability and the consistency of the performance of the DFB laser chip are ensured.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application relates to the technical field of wafer epitaxial growth, and in particular to an epitaxial control method for a DFB laser chip. Background Art

[0002] In the field of modern optical communications, DFB (distributed feedback) laser chips are widely used in optical fiber communication systems as one of the key devices. The performance of DFB lasers is highly dependent on the quality of their epitaxial layers, especially in the chip manufacturing process, where the accuracy and uniformity of epitaxial growth control are crucial. The material composition, thickness, and crystal surface profile of the epitaxial layer directly determine the laser's emission wavelength, output power, modulation rate, and other key performance indicators. Therefore, the control of the epitaxial growth process is crucial to ensure the performance of DFB laser chips.

[0003] At present, epitaxial growth technology usually adopts molecular beam epitaxy (MBE) or metal organic chemical vapor deposition (MOCVD) processes. In these processes, epitaxial growth control relies on preset growth parameters, such as temperature, gas flow rate and growth rate. However, the existing epitaxial growth control methods have certain limitations because the growth control parameters cannot be adaptively adjusted according to the monitored state of the epitaxial layer during the growth process.

[0004] In summary, the existing epitaxial growth control methods for DFB laser chips have technical problems such as the inability to quickly and accurately adjust the growth control parameters according to the growth state of the epitaxial layer of the wafer during the epitaxial growth process of the wafer, resulting in poor accuracy and uniformity of epitaxial growth control, which affects the quality of epitaxial layer growth. Summary of the invention

[0005] The purpose of the present application is to provide an epitaxial growth control method for a DFB laser chip, so as to solve the technical problem that the epitaxial growth control accuracy and uniformity are poor, which affects the quality of epitaxial layer growth, due to the inability to quickly and accurately adjust the growth control parameters according to the growth state of the epitaxial layer of the wafer during the epitaxial growth of the wafer in the existing epitaxial growth control method for DFB laser chips.

[0006] In view of the above problems, the present application provides an epitaxial control method for a DFB laser chip, comprising: reading wafer attribute information of a wafer to be processed and expected performance indicators of a DFB laser chip, with the purpose of meeting the expected performance indicators, and obtaining expected epitaxial attribute characteristics of an epitaxial layer according to the wafer attribute information analysis; obtaining a predetermined epitaxial growth cycle, and using a TACD simulation tool to perform reverse simulation to determine epitaxial growth control parameters based on constraints of meeting the expected epitaxial attribute characteristics and the predetermined epitaxial growth cycle; determining a plurality of growth intervals based on the predetermined epitaxial growth cycle, setting growth indicators for the plurality of growth intervals according to the expected epitaxial attribute characteristics, and obtaining a plurality of expected growth indicators; and performing epitaxial growth control of the wafer to be processed according to the epitaxial growth control parameters, the plurality of growth intervals, and the plurality of expected growth indicators.

[0007] One or more technical solutions provided in this application have at least the following technical effects or advantages: By reading the wafer property information of the wafer to be processed and the expected performance indicators of the DFB laser chip, with the purpose of meeting the expected performance indicators, the expected epitaxial property characteristics of the epitaxial layer are obtained according to the wafer property information analysis; then a predetermined epitaxial growth cycle is obtained, and the expected epitaxial property characteristics and the predetermined epitaxial growth cycle are constrained, and a TACD simulation tool is used for reverse simulation to determine the epitaxial growth control parameters; on the other hand, a plurality of growth intervals are determined based on the predetermined epitaxial growth cycle, and growth indicators are set for the plurality of growth intervals according to the expected epitaxial property characteristics to obtain a plurality of expected growth indicators; finally, the epitaxial growth control of the wafer to be processed is performed according to the epitaxial growth control parameters, the plurality of growth intervals and the plurality of expected growth indicators; wherein the method for performing epitaxial growth control comprises: S1: selecting a first growth interval and a second growth interval, and a second expected growth indicator for the second growth interval, wherein the first growth interval is the first growth interval among the plurality of growth intervals, and the second growth interval is the first growth interval. Adjacent growth intervals of growth intervals; S2: according to the epitaxial growth control parameters, performing epitaxial growth of the first growth interval on the wafer to be processed, wherein the epitaxial growth control parameters include growth temperature, reaction gas flow rate and reaction gas ratio; S3: using reflection high-energy electron diffraction fixed-point monitoring to obtain a crystal growth data set, wherein the crystal growth data set includes an epitaxial layer thickness set and a crystal surface profile set; S4: predicting crystal growth within a predetermined window according to the epitaxial layer thickness set and the crystal surface profile set, and obtaining first predicted epitaxial property data of the end node of the first growth interval; S5: performing deviation calculation on the second expected growth index and the first predicted epitaxial property data, and determining a second growth deviation; S6: using the TACD simulation tool to perform reverse simulation in order to meet the second growth deviation and the second growth interval as constraints, and determining the second epitaxial growth control parameters; S7: performing epitaxial growth control of the second growth interval according to the second epitaxial growth control parameters; iterating loops S1 to S7 until the predetermined epitaxial growth cycle is completed. The above method can achieve the technical goal of quickly and accurately adjusting the growth control parameters according to the growth state of the wafer epitaxial layer, effectively solving the problem of insufficient accuracy and uniformity of epitaxial growth, thereby improving the growth quality of the epitaxial layer and ensuring the stability and consistency of the DFB laser chip performance.

[0008] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented according to the contents of the specification, and in order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are specifically cited below. It should be understood that the content described in this section is not intended to identify the key or important features of the embodiments of the present application, nor is it intended to limit the scope of the present application. Other features of the present application will become easy to understand through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] In order to more clearly illustrate the technical solutions in the present application or the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings in the following description are only exemplary, and for ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying any creative work.

[0010] Figure 1 A schematic flow chart of an epitaxy control method for a DFB laser chip according to the present application; Figure 2 A schematic diagram of a flow chart of performing epitaxial growth control of a wafer to be processed in an epitaxial growth control method for a DFB laser chip of the present application. DETAILED DESCRIPTION

[0011] The present application provides an epitaxial growth control method for DFB laser chips, which solves the technical problem that the epitaxial growth control method for DFB laser chips cannot quickly and accurately adjust the growth control parameters according to the growth state of the epitaxial layer of the wafer during the epitaxial growth of the wafer, resulting in poor accuracy and uniformity of epitaxial growth control and affecting the quality of epitaxial layer growth. The technical goal of quickly and accurately adjusting the growth control parameters according to the growth state of the epitaxial layer of the wafer can be achieved, effectively solving the problem of insufficient accuracy and uniformity of epitaxial growth, thereby improving the growth quality of the epitaxial layer and ensuring the stability and consistency of the performance of the DFB laser chip.

[0012] Below, the technical solutions in the present application will be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all of the embodiments of the present application. It should be understood that the present application is not limited to the example embodiments described herein. Based on the embodiments of the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present application. It should also be noted that, for the convenience of description, only the parts related to the present application are shown in the accompanying drawings, rather than all of them.

[0013] For examples, please see the attached Figure 1 The present application provides an epitaxy control method for a DFB laser chip, which specifically comprises the following steps: Step 1: Read the wafer property information of the wafer to be processed and the expected performance index of the DFB laser chip, and obtain the expected epitaxial property characteristics of the epitaxial layer according to the wafer property information for the purpose of meeting the expected performance index.

[0014] Specifically, in the manufacturing process of DFB laser chips, the quality of epitaxial growth is crucial to the performance of the chip. In order to ensure that the chip reaches the expected performance indicators, the epitaxial growth process needs to be controlled based on the attribute information of the wafer and the expected performance indicators of the chip. First, the wafer attribute information of the wafer to be processed is read, where the wafer attribute information directly affects the growth process and quality of the epitaxial layer. The wafer attribute information includes wafer material, wafer lattice constant, surface roughness and wafer thickness. On the other hand, the expected performance indicators of the DFB laser chip are read, and the expected performance indicators include wavelength, power and modulation rate, which can be set according to the actual application scenarios and requirements of the chip.

[0015] Then, in order to meet the expected performance index for the purpose of wafer epitaxial growth, the epitaxial property characteristics of the epitaxial layer are analyzed according to the wafer property information. The property analysis model can be constructed based on the BP neural network, and the sample data is collected to train the property analysis model until convergence. The epitaxial property characteristics of the expected performance index are analyzed to obtain the expected epitaxial property characteristics, which include the expected material composition, the expected epitaxial layer thickness and the expected crystal surface profile. By obtaining the expected epitaxial property characteristics of the epitaxial layer, a basis is provided for guiding the adjustment of the growth control parameters during the epitaxial growth process, ensuring that the epitaxial growth quality of the final DFB laser chip meets the expected performance requirements.

[0016] Step 2: Obtain a predetermined epitaxial growth cycle, and use the TACD simulation tool to perform reverse simulation to determine the epitaxial growth control parameters, subject to the expected epitaxial property characteristics and the predetermined epitaxial growth cycle as constraints.

[0017] Specifically, a predetermined epitaxial growth cycle is obtained, wherein the epitaxial growth cycle is the entire process from the beginning to the end of the epitaxial growth and the expected growth time set. Those skilled in the art can set it according to the actual situation, such as setting the expected growth time to 12 hours. Then, in order to meet the expected epitaxial property characteristics and the predetermined epitaxial growth cycle as constraints, a reverse simulation is performed using a TACD simulation tool. First, a TCAD simulation tool suitable for the epitaxial growth process is selected, such as Synopsys Sentaurus Process. This type of tool can simulate the deposition, growth rate, surface morphology and other physical properties of the material; then, the expected epitaxial property characteristics (material composition, epitaxial layer thickness and crystal surface profile) and the predetermined epitaxial growth cycle are input into the TCAD simulation tool, and the epitaxial growth control parameters (growth temperature, reaction gas flow rate and reaction gas ratio) are automatically adjusted. Through iterative simulation, the growth control parameters are continuously adjusted until the thickness, material composition, surface morphology, etc. of the epitaxial growth reach the expected target value, and the epitaxial growth control parameters that meet the expected epitaxial property characteristics and the predetermined epitaxial growth cycle are output, and the epitaxial growth control parameters include growth temperature, reaction gas flow rate and reaction gas ratio. By obtaining the epitaxial growth control parameters, a basis is provided for the subsequent epitaxial growth control of the wafer.

[0018] Step three: Determine a plurality of growth intervals based on the predetermined epitaxial growth cycle, set growth indicators for the plurality of growth intervals according to the expected epitaxial property characteristics, and obtain a plurality of expected growth indicators.

[0019] Specifically, first, the predetermined epitaxial growth cycle is divided into multiple growth intervals according to a predetermined time step, and the predetermined time step can be set according to the actual situation. For example, the predetermined time step is set to 30 minutes to obtain multiple growth intervals, wherein the multiple growth intervals have a sequential identification. Then, the growth index of the multiple growth intervals is set according to the expected epitaxial property characteristics, that is, the expected growth index of each growth interval end node is configured, including the material composition, epitaxial layer thickness and crystal surface profile of each end node; by configuring the expected growth index for each growth interval end node, the precision of wafer epitaxial growth control can be improved, and the precise control of the epitaxial growth process can be ensured to achieve a high-quality epitaxial layer structure of the final DFB laser chip.

[0020] Step 4: Perform epitaxial growth control of the wafer to be processed according to the epitaxial growth control parameters, multiple growth intervals and multiple expected growth indicators.

[0021] Specifically, the epitaxial growth control of the wafer to be processed is finally performed according to the epitaxial growth control parameters, multiple growth intervals and multiple expected growth indicators, that is, the epitaxial growth of the wafer is monitored and feedback-adjusted in real time according to the order of the growth intervals and multiple expected growth indicators to ensure that the growth quality at any stage of the epitaxial growth process meets the expected requirements, and ultimately achieve high-quality epitaxial layer growth to meet the performance indicators of the DFB laser chip.

[0022] By providing an epitaxial growth control method for DFB laser chips, the technical problem that the epitaxial growth control method for DFB laser chips cannot quickly and accurately adjust the growth control parameters according to the growth state of the epitaxial layer of the wafer during the epitaxial growth of the wafer, resulting in poor accuracy and uniformity of epitaxial growth control, which affects the quality of epitaxial layer growth, can be solved. By dividing a predetermined epitaxial growth cycle into multiple growth intervals, growth indicators are set for the multiple growth intervals according to expected epitaxial property characteristics to obtain multiple expected growth indicators; then the first growth interval and the second growth interval are selected, as well as the second expected growth indicator for the second growth interval; according to the epitaxial growth control parameters, epitaxial growth of the first growth interval is performed on the wafer to be processed, and the crystal growth data set is obtained by fixed-point monitoring using reflected high-energy electron diffraction; further, the crystal growth within the predetermined window is predicted according to the epitaxial layer thickness set and the crystal surface contour set, and the first predicted epitaxial property data of the end node of the first growth interval is obtained; then the second expected growth indicator is set for the second growth interval. The deviation is calculated from the first predicted epitaxial property data to determine the second growth deviation; and with the second growth deviation and the second growth interval as constraints, the TACD simulation tool is used for reverse simulation to determine the second epitaxial growth control parameters; finally, the epitaxial growth control of the second growth interval is executed according to the second epitaxial growth control parameters, and the above steps are iterated and cycled until the predetermined epitaxial growth cycle is completed; the technical goal of quickly and accurately adjusting the growth control parameters according to the growth state of the epitaxial layer of the wafer can be achieved, and the problem of insufficient accuracy and uniformity of epitaxial growth can be effectively solved, thereby improving the growth quality of the epitaxial layer and ensuring the stability and consistency of the performance of the DFB laser chip.

[0023] Furthermore, the present application includes: The wafer attribute information includes wafer material, wafer lattice constant, surface roughness and wafer thickness, and the expected performance indicators include wavelength, power and modulation rate.

[0024] Specifically, the wafer attribute information includes wafer material, wafer lattice constant, surface roughness and wafer thickness, wherein the wafer material is the basis of the epitaxial layer to be processed, and directly affects the lattice matching and growth behavior of the epitaxial layer; the wafer lattice constant is the spatial periodic constant of the atomic arrangement in the crystal structure. When the epitaxial layer grows, the lattice constant of the epitaxial material should be matched with the lattice constant of the base wafer as much as possible to reduce the stress and defects caused by lattice mismatch; surface roughness refers to the flatness of the wafer surface; the wafer thickness is the physical thickness of the wafer material, which affects the thermal conductivity and mechanical strength of the epitaxial growth.

[0025] The expected performance indicators include wavelength, power and modulation rate, where wavelength is a characteristic parameter of the light emitted by a DFB laser and is crucial to meeting communication standards. The wavelength of the laser is determined by the material composition and thickness of the epitaxial layer; power is the output energy of the laser, and the power of the laser is determined by the material quality, thickness and electrical properties of the epitaxial layer. A high-quality epitaxial layer can improve the photoelectric conversion efficiency of the laser, thereby increasing the output power; modulation rate is the rate at which the laser modulates the optical signal, usually in GHz (gigahertz), and refers to the speed at which the laser can modulate the signal. The modulation rate is closely related to the electrical properties of the epitaxial layer (such as carrier mobility, injection efficiency, etc.), especially the epitaxial thickness of the active region.

[0026] Furthermore, the present application includes: Taking the wafer attribute information as a constraint and wafer epitaxial growth as a guide, a sample data set is retrieved and acquired, wherein the sample data includes sample performance indicators and sample epitaxial attribute characteristics, and the sample epitaxial attribute characteristics include material composition, epitaxial layer thickness and crystal surface profile; the sample data set is used to perform supervised training on a BP neural network to acquire an attribute analyzer; the expected performance indicators are parsed by the attribute analyzer, and the expected epitaxial attribute characteristics are output.

[0027] Specifically, first, the wafer property information and the expected epitaxial growth process are used as search constraints, wherein the expected epitaxial growth process can be selected according to actual conditions, including molecular beam epitaxy (MBE), metal organic chemical vapor deposition (MOCVD) process, etc., and wafer epitaxial growth is used as a search guide word, and sample data retrieval is performed based on big data technology to obtain a sample data set, wherein the sample data includes sample performance indicators and sample epitaxial property characteristics, and the sample epitaxial property characteristics include material composition, epitaxial layer thickness and crystal surface profile.

[0028] Then, an initial attribute analyzer is constructed based on the BP neural network. The initial attribute analyzer is a neural network model that can be iteratively optimized in machine learning, including an input layer, multiple hidden layers and an output layer, wherein the input data of the input layer is a performance indicator, and the output data is an extensional attribute feature. Then, the sample data set is divided into a training set and a validation set according to a predetermined ratio, usually 80% of the training set and 20% of the validation set; the initial attribute analyzer is further supervised and trained using the training set, that is, supervised learning is performed using the loss function and the gradient descent algorithm, and the weights and biases in the model are optimized and corrected to obtain an initial attribute analyzer in a convergent state; then the initial attribute analyzer is verified and trained using the validation set until an attribute analyzer that meets the expected conditions is obtained. By constructing an attribute analyzer based on the BP neural network to analyze extensional attribute features, the efficiency and accuracy of obtaining extensional attribute features can be improved.

[0029] Finally, the expected performance index is input into the attribute analyzer for parsing, and the expected extension attribute feature is output.

[0030] Further, as attached Figure 2 As shown, this application includes: S1: Select a first growth interval and a second growth interval, and a second expected growth index for the second growth interval, wherein the first growth interval is the first growth interval among the multiple growth intervals, and the second growth interval is an adjacent growth interval to the first growth interval; S2: According to the epitaxial growth control parameters, perform epitaxial growth of the first growth interval on the wafer to be processed, wherein the epitaxial growth control parameters include growth temperature, reaction gas flow rate and reaction gas ratio; S3: Use reflected high-energy electron diffraction fixed-point monitoring to obtain a crystal growth data set, wherein the crystal growth data set includes an epitaxial layer thickness set and a crystal surface profile set; S4: According to the epitaxial layer thickness set and the crystal surface profile set, perform crystal growth prediction within a predetermined window to obtain first predicted epitaxial property data at the end node of the first growth interval; S5: Perform deviation calculation on the second expected growth index and the first predicted epitaxial property data to determine the second growth deviation; S6: Using the TACD simulation tool to perform reverse simulation to meet the second growth deviation and the second growth interval as constraints to determine the second epitaxial growth control parameters; S7: Perform epitaxial growth control of the second growth interval according to the second epitaxial growth control parameters.

[0031] Specifically, first, a first growth interval and a second growth interval are selected from the multiple growth intervals, wherein the first growth interval is the first growth interval to start among the multiple growth intervals, and the second growth interval is the adjacent growth interval of the first growth interval, and a second expected growth index of the second growth interval is obtained. Then, according to the epitaxial growth control parameters, the epitaxial growth of the first growth interval is performed on the wafer to be processed, wherein the epitaxial growth control parameters include growth temperature, reaction gas flow rate and reaction gas ratio; a predetermined monitoring node is configured, wherein the time intervals of adjacent monitoring nodes are the same, which can be set according to demand, such as monitoring once every 1 minute, and during the epitaxial growth process in the first growth interval, the wafer is monitored at a predetermined monitoring node using reflection high energy electron diffraction (RHEED), for example, the thickness of the epitaxial layer is monitored in real time every minute through the periodic changes of the RHEED stripes, the change period of the stripes is inversely proportional to the thickness of the epitaxial layer, and the growth thickness is calculated by analyzing the stripe period; the intensity and clarity of the RHEED pattern reflect the surface roughness and the step structure of the crystal. By analyzing these patterns, the change of surface roughness is recorded; and a crystal growth data set is obtained, wherein the crystal growth data set includes an epitaxial layer thickness set and a crystal surface profile set.

[0032] Further, the crystal growth prediction within the predetermined window is performed based on the epitaxial layer thickness set and the crystal surface profile set, wherein the predetermined window is the sum of the monitoring data delay, the prediction data delay and the parameter control delay, that is, in order to solve the hysteresis of the monitoring data, the wafer growth state at the end node of the first growth interval is predicted based on the monitoring data to obtain the first predicted epitaxial property data at the end node of the first growth interval. Then, the deviation of the second expected growth index and the first predicted epitaxial property data is calculated to obtain the second growth deviation, which includes the material composition deviation, the epitaxial layer thickness deviation and the crystal surface profile deviation.

[0033] Then, in order to satisfy the second growth deviation and the second growth interval as constraints, the TACD simulation tool is used to perform reverse simulation, output the second epitaxial growth control parameters, and perform epitaxial growth control in the second growth interval according to the second epitaxial growth control parameters. This can improve the fineness of wafer epitaxial growth control and achieve the technical goal of quickly and accurately adjusting the growth control parameters according to the growth state of the wafer epitaxial layer.

[0034] Furthermore, the present application includes: After the epitaxial growth control of the second growth interval is performed according to the second epitaxial growth control parameter, the iterative cycle S1 to S7 is performed until the predetermined epitaxial growth cycle is completed.

[0035] Specifically, after executing the epitaxial growth control of the second growth interval according to the second epitaxial growth control parameters, the same method is used to iterate the loop S1 to S7 to continue to optimize and adjust the epitaxial growth control parameters of the third growth interval, the fourth growth interval and the subsequent growth intervals until the predetermined epitaxial growth cycle is completed. By performing iterative feedback control, the quality and performance of the epitaxial layer in the entire growth cycle can be ensured to be stable, and ultimately meet the predetermined design requirements.

[0036] Furthermore, the present application includes: Perform thickness growth trend analysis based on the epitaxial layer thickness set to determine the thickness growth rate; select the terminal epitaxial layer thickness from the epitaxial layer thickness set, predict crystal growth within a predetermined window based on the thickness growth rate and the terminal epitaxial layer thickness, and obtain a first predicted epitaxial layer thickness; perform profile change trend analysis based on the crystal surface contour set to determine profile change characteristics; select the terminal surface contour from the crystal surface contour set, predict crystal growth within a predetermined window based on the contour change characteristics and the terminal surface contour, and obtain a first predicted crystal surface contour; obtain the first predicted epitaxial property data based on the first predicted epitaxial layer thickness and the first predicted crystal surface contour.

[0037] Specifically, first, the thickness growth trend analysis is performed according to the epitaxial layer thickness set. By drawing a thickness growth trend graph, the horizontal axis represents time (minutes) and the vertical axis represents the thickness of the epitaxial layer (nanometers), the trend graph can show whether the epitaxial growth is uniform, or whether there are fluctuations or anomalies; then the thickness growth rate is calculated, and the thickness growth rate is the increase in thickness per unit time, which can be obtained by averaging the overall growth rate of the entire epitaxial growth process in the first growth interval. Then, the terminal epitaxial layer thickness is selected from the epitaxial layer thickness set, and the terminal epitaxial layer thickness is the epitaxial layer thickness of the terminal monitoring node; further, the crystal growth prediction within the predetermined window is performed according to the thickness growth rate and the terminal epitaxial layer thickness, that is, the predicted epitaxial layer thickness of the end node of the first growth interval is calculated to obtain the first predicted epitaxial layer thickness.

[0038] On the other hand, according to the crystal surface profile set, the profile change trend analysis can be performed by drawing a trend graph of surface roughness change over time and a graph of fringe intensity change. Through these graphs, the change pattern of the surface profile over time can be observed, and the profile change characteristics can be determined, wherein the profile change characteristics include the changes in surface roughness, fringe intensity and step structure. Then, the terminal surface profile is selected from the crystal surface profile set, and the terminal surface profile is the wafer surface profile of the terminal monitoring node. The crystal growth prediction within the predetermined window is further performed based on the profile change characteristics and the terminal surface profile to obtain the first predicted crystal surface profile. Finally, the first predicted material component is obtained based on the first predicted epitaxial layer thickness and the first predicted crystal surface profile analysis, and the first predicted epitaxial layer thickness, the first predicted crystal surface profile and the first predicted material component are set as the first predicted epitaxial attribute data.

[0039] Furthermore, the present application includes: The sum of the monitoring data delay, the prediction data delay and the parameter control delay is taken as the predetermined window.

[0040] Specifically, in the process of epitaxial growth control, monitoring data, prediction data and parameter control operations may produce different delays. In order to make more accurate crystal growth predictions, the sum of monitoring data delay, prediction data delay and parameter control delay can be used as a predetermined window, that is, the total duration of these delay factors is considered to set the prediction window, so as to make timely and accurate control adjustments. Monitoring data delay refers to the delay caused by the collected data being recorded and transmitted to the control system when RHEED or other equipment is used for real-time monitoring during epitaxial growth; prediction data delay refers to the delay in calculation and model processing when predicting crystal growth trends based on existing monitoring data; parameter control delay refers to the time required for adjusting growth parameters to have an actual impact on epitaxial layer growth during the process of controlling epitaxial growth. By setting a predetermined window for prediction, it is possible to ensure that appropriate adjustments are made during the delay period, which improves the accuracy and real-time performance of epitaxial growth control and ultimately helps to achieve high-quality epitaxial layer growth.

[0041] Furthermore, the present application includes: Using the TACD simulation tool, the growth simulation of the wafer to be processed is performed according to the first growth interval and epitaxial growth control parameters to obtain the first simulated epitaxial layer thickness, the first simulated crystal surface contour and the first simulated material composition; the deviation of the first predicted epitaxial layer thickness and the first simulated epitaxial layer thickness is calculated to obtain the first predicted thickness difference; the deviation of the first predicted crystal surface contour and the first simulated crystal surface contour is analyzed to determine the first predicted contour deviation characteristics; the first simulated material composition is correlated and corrected according to the first predicted thickness difference and the first predicted contour deviation characteristics to obtain the first predicted material composition; the first predicted epitaxial layer thickness, the first predicted crystal surface contour and the first predicted material composition are set as the first predicted epitaxial attribute data.

[0042] Specifically, first, the growth simulation of the wafer to be processed is performed using the TACD simulation tool according to the first growth interval and epitaxial growth control parameters to obtain the first simulated epitaxial layer thickness, the first simulated crystal surface profile and the first simulated material composition. Then, the deviation of the first predicted epitaxial layer thickness and the first simulated epitaxial layer thickness is calculated to obtain the first predicted thickness difference; the deviation of the first predicted crystal surface profile and the first simulated crystal surface profile is analyzed to determine the first predicted profile deviation characteristics, wherein the predicted profile deviation characteristics include surface roughness deviation, fringe intensity deviation and step structure deviation.

[0043] Then, the first simulated material component is corrected according to the first predicted thickness difference and the first predicted profile deviation characteristics, and a correction model can be constructed based on historical data analysis. For example, the expression of the correction model is: ;in, is the modified i-th material component, is the simulated material component of the i-th material, is the thickness correction factor, is the surface profile correction factor, To predict the thickness difference, To predict the profile deviation feature. Use the correction model to correct the first simulated material composition according to the thickness difference and the surface profile deviation, and obtain the first predicted material composition by adjusting the proportion of each component. For example, if the In content is 50% in the simulation, but due to the correction of the thickness deviation and the surface roughness, the predicted In content is 48%; obtain the first predicted material composition. Finally, the first predicted epitaxial layer thickness, the first predicted crystal surface profile and the first predicted material composition are set as the first predicted epitaxial property data.

[0044] In summary, the epitaxial control method for a DFB laser chip provided in this application has the following technical effects: The method divides a predetermined epitaxial growth cycle into a plurality of growth intervals, sets growth indicators for the plurality of growth intervals according to expected epitaxial property characteristics, and obtains a plurality of expected growth indicators; then selects a first growth interval and a second growth interval, and a second expected growth indicator for the second growth interval; performs epitaxial growth in the first growth interval on the wafer to be processed according to epitaxial growth control parameters, and obtains a crystal growth data set by using reflected high-energy electron diffraction fixed-point monitoring; further predicts crystal growth within a predetermined window according to an epitaxial layer thickness set and a crystal surface profile set, and obtains first predicted epitaxial property data at the end node of the first growth interval; then performs a quantitative analysis of the second expected growth indicator. The deviation is calculated from the first predicted epitaxial property data to determine the second growth deviation; and with the second growth deviation and the second growth interval as constraints, the TACD simulation tool is used for reverse simulation to determine the second epitaxial growth control parameters; finally, the epitaxial growth control of the second growth interval is executed according to the second epitaxial growth control parameters, and the above steps are iterated and cycled until the predetermined epitaxial growth cycle is completed; the technical goal of quickly and accurately adjusting the growth control parameters according to the growth state of the epitaxial layer of the wafer can be achieved, and the problem of insufficient accuracy and uniformity of epitaxial growth can be effectively solved, thereby improving the growth quality of the epitaxial layer and ensuring the stability and consistency of the performance of the DFB laser chip.

[0045] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present application. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to the embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.

[0046] Obviously, those skilled in the art can make various changes and modifications to the present application without departing from the spirit and scope of the present application. Thus, if these modifications and variations of the present application belong to the scope of the present application and its equivalent technology, the present application is also intended to include these modifications and variations.

Claims

1. A method for controlling epitaxy of a DFB laser chip, characterized in that: include: Reading wafer property information of the wafer to be processed and expected performance indicators of the DFB laser chip, with the purpose of meeting the expected performance indicators, and obtaining expected epitaxial property characteristics of the epitaxial layer according to the wafer property information analysis; Obtaining a predetermined epitaxial growth cycle, using the expected epitaxial property characteristics and the predetermined epitaxial growth cycle as constraints, performing reverse simulation using a TACD simulation tool, and determining epitaxial growth control parameters; Determine a plurality of growth intervals based on the predetermined epitaxial growth cycle, set growth indicators for the plurality of growth intervals according to the expected epitaxial property characteristics, and obtain a plurality of expected growth indicators; The epitaxial growth control of the wafer to be processed is performed according to the epitaxial growth control parameters, multiple growth intervals and multiple expected growth indicators.

2. The epitaxial control method for a DFB laser chip according to claim 1, characterized in that: The wafer attribute information includes wafer material, wafer lattice constant, surface roughness and wafer thickness, and the expected performance indicators include wavelength, power and modulation rate.

3. The epitaxial control method for a DFB laser chip according to claim 2, characterized in that: Characterize the expected epitaxial properties of the epitaxial layer, including: Using the wafer property information as a constraint and wafer epitaxial growth as a guide, a sample data set is retrieved and acquired, wherein the sample data includes sample performance indicators and sample epitaxial property characteristics, and the sample epitaxial property characteristics include material composition, epitaxial layer thickness, and crystal surface profile; Using the sample data set to perform supervised training on the BP neural network to obtain an attribute analyzer; The expected performance indicator is analyzed by the attribute analyzer to output the expected extension attribute feature.

4. The epitaxial control method for a DFB laser chip according to claim 3, characterized in that: The epitaxial growth control of the wafer to be processed is performed according to the epitaxial growth control parameter, multiple growth intervals and multiple expected growth indicators, including: S1: Selecting a first growth interval and a second growth interval, and a second expected growth index of the second growth interval, wherein the first growth interval is the first growth interval among the multiple growth intervals, and the second growth interval is an adjacent growth interval to the first growth interval; S2: performing epitaxial growth in the first growth interval on the wafer to be processed according to the epitaxial growth control parameters, wherein the epitaxial growth control parameters include growth temperature, reaction gas flow rate and reaction gas ratio; S3: using reflection high energy electron diffraction fixed point monitoring to obtain a crystal growth data set, wherein the crystal growth data set includes an epitaxial layer thickness set and a crystal surface profile set; S4: Predicting crystal growth within a predetermined window according to the epitaxial layer thickness set and the crystal surface profile set, and obtaining first predicted epitaxial property data of an end node of a first growth interval; S5: performing deviation calculation on the second expected growth indicator and the first predicted extension attribute data to determine a second growth deviation; S6: using the TACD simulation tool to perform reverse simulation to determine the second epitaxial growth control parameters, with the second growth deviation and the second growth interval as constraints; S7: Perform epitaxial growth control in the second growth interval according to the second epitaxial growth control parameters.

5. The epitaxial control method for a DFB laser chip according to claim 4, characterized in that: After the epitaxial growth control of the second growth interval is performed according to the second epitaxial growth control parameter, the iterative cycle S1 to S7 is performed until the predetermined epitaxial growth cycle is completed.

6. The epitaxial control method for a DFB laser chip according to claim 4, characterized in that: Predicting crystal growth within a predetermined window according to the epitaxial layer thickness set and the crystal surface profile set includes: Performing thickness growth trend analysis based on the epitaxial layer thickness set to determine the thickness growth rate; Selecting a terminal epitaxial layer thickness from the epitaxial layer thickness set, predicting crystal growth within a predetermined window according to the thickness growth rate and the terminal epitaxial layer thickness, and obtaining a first predicted epitaxial layer thickness; Performing profile change trend analysis based on the crystal surface profile set to determine profile change characteristics; Selecting a terminal surface profile from the crystal surface profile set, predicting crystal growth within a predetermined window according to the profile change characteristics and the terminal surface profile, and obtaining a first predicted crystal surface profile; The first predicted epitaxial property data is obtained based on the first predicted epitaxial layer thickness and the first predicted crystal surface profile.

7. The epitaxial control method for a DFB laser chip according to claim 6, characterized in that: The sum of the monitoring data delay, the prediction data delay and the parameter control delay is taken as the predetermined window.

8. The epitaxial control method for a DFB laser chip according to claim 6, characterized in that: Obtaining the first predicted extension attribute data includes: Using a TACD simulation tool, the growth simulation of the wafer to be processed is performed according to the first growth interval and the epitaxial growth control parameters to obtain a first simulated epitaxial layer thickness, a first simulated crystal surface profile and a first simulated material composition; performing deviation calculation on the first predicted epitaxial layer thickness and the first simulated epitaxial layer thickness to obtain a first predicted thickness difference; Performing deviation analysis on the first predicted crystal surface profile and the first simulated crystal surface profile to determine a first predicted profile deviation feature; Performing correlation correction on the first simulated material component according to the first predicted thickness difference and the first predicted profile deviation feature to obtain a first predicted material component; The first predicted epitaxial layer thickness, the first predicted crystal surface profile, and the first predicted material composition are set as the first predicted epitaxial property data.

Citation Information

Patent Citations

  • Semiconductor film process parameter optimization system for neural network model prediction

    CN112016676A

  • Semiconductor technological process control method based on machine learning

    CN115458422A

  • Epitaxial wafer, manufacturing method and device thereof and storage medium

    CN116798898A

  • Wafer epitaxial growth control method based on machine learning

    CN118223119A

  • Thickness prediction network learning method, semiconductor device manufacturing method, and semiconductor material deposition equipment

    US20200193290A1