Air pressure control method for microwave plasma chemical vapor deposition diamond
Through real-time spectral monitoring and machine learning prediction combined with multi-stage air pressure regulation, the air pressure control problem in microwave plasma chemical vapor deposition technology is solved, precise regulation and stability of diamond growth process is achieved, and production efficiency and equipment reliability are improved.
Patent Information
- Application Number
- CN202510663918.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-05-22
- Publication Date
- 2025-08-29
AI Technical Summary
In microwave plasma chemical vapor deposition technology, air pressure control is difficult to accurately regulate, and traditional methods are difficult to match the air pressure requirements of different growth stages of diamond in real time. The plasma state during the reaction process is complicated, making it difficult to comprehensively monitor and insufficient research on parameter coupling relationships.
Real-time spectral monitoring and machine learning prediction combined with multi-stage air pressure regulation, optical emission spectrometer monitors plasma status, predicts the optimal air pressure value through long-term memory network models, and combines multi-parameter coordinated control and air pressure closed-loop control to achieve accurate air pressure regulation.
It realizes precise air pressure control in the diamond growth process, improves the crystal core density and surface flatness, improves growth speed and production efficiency, reduces production costs and equipment failure rates, and is suitable for large-scale industrial production.
Smart Images

Figure CN120560360A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of microwave plasma chemical vapor deposition, and more particularly to a gas pressure control method for microwave plasma chemical vapor deposition of diamond. Background Art
[0002] In the field of microwave plasma chemical vapor deposition technology, diamond has great application value in many fields such as semiconductors, quantum technology, and optics due to its excellent performance. For example, it is an ideal heat dissipation material for semiconductor devices, a research hotspot in quantum information technology, and an excellent laser crystal material. Microwave plasma chemical vapor deposition technology has become an ideal way to prepare large-size, high-quality single-crystal diamonds due to its advantages such as large and uniform deposition area and no electrode pollution.
[0003] However, microwave plasma chemical vapor deposition technology also faces many challenges, such as: 1. From the perspective of process parameter control, gas pressure is a key parameter, and its precise control is difficult. Traditional methods cannot accurately match the stringent gas pressure requirements of different diamond growth stages in real time. During the nucleation stage, a specific gas pressure range and gas flow ratio are required to create an environment conducive to nucleation and increase the nucleation density. During the coarsening and finishing stages, gas pressure and other parameters also need to be precisely adjusted to ensure grain growth quality and surface flatness. Existing technologies are difficult to achieve such precision. 2. From the perspective of monitoring and understanding the reaction process, the plasma state in the reaction chamber is complex, and it is difficult to obtain its information comprehensively and in real time by relying solely on conventional means. Atomic hydrogen and methyl groups in the plasma participate in key reactions, and their state changes affect diamond growth, which are difficult to capture with conventional monitoring. At the same time, there has been insufficient research on the coupling relationship between various process parameters. For example, when the gas pressure fluctuates, there is a lack of effective strategies for the coordinated adjustment of parameters such as microwave power, gas flow rate and substrate temperature.
[0004] In view of the above situation, the present invention provides a gas pressure control method for microwave plasma chemical vapor deposition of diamond. Summary of the Invention
[0005] In order to overcome the above-mentioned defects of the prior art, the present invention provides a gas pressure control method for microwave plasma chemical vapor deposition of diamond to solve the problems raised in the above-mentioned background technology.
[0006] To achieve the above object, the present invention provides the following technical solution: a method for controlling gas pressure in microwave plasma chemical vapor deposition of diamond, specifically comprising the following steps: S1. Real-time spectrum monitoring: Using an optical emission spectrometer, the emission spectrum of the plasma in the reaction chamber is continuously collected in the wavelength range of 200-800 nm. From the collected spectral data, the characteristic spectral line intensities of carbon species and hydrogen atoms, as well as the spectral line broadening parameters, are extracted. The spectral line broadening parameters are used to reflect the plasma state. S2. Machine Learning Prediction: Using a long-short-term memory network model, a correlation model between gas pressure and spectral characteristics was constructed. Using the current spectral parameters, microwave power, gas flow rate, and substrate temperature as input data, the optimal gas pressure value was predicted with an accuracy of ±0.5 kPa, providing a basis for gas pressure control. S3, multi-stage air pressure control: During the nucleation stage, the pressure in the reaction chamber is set at 20-30 kPa, and the flow ratio of hydrogen to methane is controlled at 95:5 to create an environment suitable for the formation of diamond nuclei and increase the nucleation density. During the coarsening stage, the gas pressure was reduced to 15-20 kPa, and the hydrogen to methane flow ratio was adjusted to 90:10 to accelerate the lateral growth of the grains and reduce the deposition of non-diamond carbon. During the finishing stage, the gas pressure is raised to 25-30kPa and modulated with a microwave power of 500-800W to achieve fine processing of the diamond surface and improve surface flatness and crystal quality. The stage switching condition is monitored by Raman spectroscopy. When the intensity ratio of the diamond characteristic peak to the non-diamond carbon characteristic peak is greater than 95%, the switching operation of the deposition stage is triggered; S4. Multi-parameter coordinated control: Gas pressure-power coupling: when the gas pressure fluctuation exceeds ±1 kPa, the microwave power is adjusted immediately and synchronously. The adjustment amplitude is calculated according to , so as to maintain the stability of the plasma. Where, is the gas pressure adjustment amount, and is the microwave power adjustment amount; Temperature compensation: If the substrate temperature deviation exceeds 5°C, the gas pressure setting value is corrected to eliminate the effect of temperature change on the deposition process. Where, is the gas pressure adjustment amount and is the temperature deviation amount; Gas flow coordination: When the gas pressure changes, the hydrogen and methane flow ratio is dynamically optimized to ensure the smooth progress of the reaction. Where, is the gas flow change and is the gas pressure adjustment; S5, air pressure closed-loop control, accurately adjusts the exhaust flow through a high-speed electric needle valve, and combines with a mass flow controller to stabilize the air intake to form a closed-loop control system to achieve dynamic and precise adjustment of the air pressure.
[0007] Preferably, in step S1, the spectral characteristic parameters include the hydrogen to methane intensity ratio and the Hα Stark broadening width, and these parameters are subjected to dimensionality reduction processing by principal component analysis to improve data processing efficiency and accuracy.
[0008] Preferably, in step S2, the machine learning model is trained using historical deposition data, and its input parameters also include substrate material type, microwave frequency of 2.45 GHz, and chamber temperature.
[0009] Preferably, in step S3, the gas pressure fluctuation range in each stage is controlled within ±1 kPa, and the gas flow rate adjustment accuracy can reach ±0.5 sccm.
[0010] Preferably, a negative bias enhancement technique is used in the nucleation stage of step S3, with the bias voltage ranging from -50V to -100V, to further increase the diamond nucleation density.
[0011] Preferably, oxygen doping is introduced during the finishing stage of step S2 so that the ratio of hydrogen to methane is 0.1-0.5%, thereby suppressing non-diamond carbon deposition and improving surface flatness.
[0012] The present invention also provides a gas pressure control system for executing the above-mentioned gas pressure control method for microwave plasma chemical vapor deposition of diamond, comprising a spectral feedback module, a machine learning module, a multi-stage control module, a multi-parameter coordination module and an actuator, wherein; The spectrum feedback module consists of a quartz window, a fiber optic spectrometer, and a data processing unit, and is used to collect plasma spectra in real time and extract corresponding characteristic parameters; The machine learning module uses a long short-term memory network algorithm to build an air pressure prediction model and is integrated into the controller; The multi-stage control module automatically switches the deposition stage according to the Raman spectrum monitoring results and adjusts the gas pressure and gas flow parameters at the same time; The multi-parameter collaborative module is used to establish a coupled control model of gas pressure, microwave power, and substrate temperature to achieve linkage adjustment among various parameters; The actuator includes a high-speed electric needle valve with a response time of less than 100ms, a molecular pump with a pumping speed of 500L / s, a mechanical pump with a pumping speed of 30L / s, a gas distribution system of a porous spray plate, and a reaction chamber.
[0013] Preferably, the reaction chamber adopts a double-layer water-cooling structure, the inner layer is quartz glass with a dielectric loss of less than 0.0002, the outer layer is made of stainless steel, and a pressure sensor with an accuracy of ±0.1kPa is configured to monitor the air pressure in real time.
[0014] Preferably, the porous spray plate of the gas distribution system has an aperture of 0.5 mm and an opening rate of 30%, and is combined with a spiral guide ring to achieve a gas flow rate deviation of less than 5%.
[0015] Preferably, the controller integrates PID algorithm and fuzzy control strategy to achieve an overshoot of less than 5% and an adjustment time of less than 30s during dynamic adjustment of air pressure.
[0016] Technical effects and advantages of the present invention: 1. Real-time spectral monitoring is used to obtain plasma state information, and the optimal gas pressure value is predicted by combining a machine learning model to achieve precise gas pressure control. The multi-stage gas pressure control strategy sets parameters based on the characteristics of each stage of diamond growth. For example, the parameters in the nucleation stage are optimized to increase the crystal nucleus density. The refinement stage combines microwave power modulation to improve surface flatness and crystal quality. Oxygen doping is introduced in the refinement stage, and negative bias enhancement technology is used in the nucleation stage to further suppress non-diamond carbon deposition and improve nucleation quality. 2. Through multi-stage gas pressure regulation and multi-parameter coordinated control, the reaction process is always in an optimized state, accelerating the diamond growth rate by more than 50% compared with traditional processes. The coordinated control of gas flow and the efficient gas pressure closed-loop control system ensure the full utilization of reaction gases, improve the utilization rate of precursors, reduce production costs, and improve overall production efficiency. 3. Through the closed-loop control of air pressure combined with a high-speed electric needle valve and a mass flow controller, dynamic and precise adjustment of air pressure is achieved, and the range of air pressure fluctuation is small. The special design of the reaction chamber and the precise coordination of the actuator ensure the stability of the deposition environment. Multi-parameter coordinated control responds to changes in air pressure, temperature, etc. in a timely manner, maintains plasma stability, reduces equipment failure rate and process scrap rate, improves the stability and reliability of the production process, and is conducive to industrial large-scale production. BRIEF DESCRIPTION OF THE DRAWINGS
[0017] Figure 1 The figure is a flow chart of the overall method of the present invention.
[0018] Figure 2 This is a diagram of the module connection system of the present invention. DETAILED DESCRIPTION
[0019] Example 1 This embodiment provides a method for controlling gas pressure in microwave plasma chemical vapor deposition of diamond, which specifically includes the following steps: S1. Real-time spectrum monitoring: An optical emission spectrometer is used to monitor the plasma in the reaction chamber in real time. During the deposition process, the emission spectrum in the wavelength range of 200-800nm is continuously collected. By analyzing the spectral data, the characteristic spectral line intensities and spectral line broadening parameters of carbon species and hydrogen atoms are extracted. These parameters can reflect the state of the plasma and provide an important basis for subsequent gas pressure regulation.
[0020] For example, in a certain experiment, spectral data collected by an optical emission spectrometer showed that at a specific deposition stage, the characteristic spectral line intensity of carbon species was relatively stable, while the Stark broadening width of Hα changed slightly, indicating that the electron density of the plasma may have changed slightly, and further attention needs to be paid to the adjustment of gas pressure.
[0021] S2. Machine Learning Prediction: With the help of the long short-term memory network model, a correlation model between gas pressure and spectral characteristics is constructed. The currently collected spectral parameters, microwave power, gas flow rate, substrate temperature and other data are used as input. The model is used to predict the optimal gas pressure value with an accuracy of up to ±0.5kPa.
[0022] In practical applications, a large amount of historical deposition data was collected, including spectral parameters, gas pressure values and corresponding diamond deposition quality under different process conditions. This data was used to train the long-short-term memory network model, enabling it to accurately learn the complex relationship between various parameters. For example, in a new deposition experiment, the model predicted the optimal gas pressure value of 22kPa based on the current spectral parameters and other input data, providing clear guidance for subsequent gas pressure regulation.
[0023] S3, multi-stage air pressure control: During the nucleation stage, the gas pressure in the reaction chamber is set in the range of 20-30kPa, and the hydrogen to methane flow ratio is maintained at 95:5. At this stage, such parameter settings create a favorable environment for the formation of diamond nuclei and effectively increase the nucleation density.
[0024] For example, in one experiment, the gas pressure was stabilized at 25 kPa during the nucleation stage, and the hydrogen to methane flow ratio was strictly controlled at 95:5. After 2 hours of deposition, scanning electron microscopy observations revealed that the diamond nuclei were evenly distributed on the substrate surface with a high density, laying a good foundation for the subsequent growth stage.
[0025] During the coarsening stage, reducing the gas pressure to 15-20 kPa and adjusting the hydrogen to methane flow ratio to 90:10 can help accelerate the lateral growth rate of grains while reducing the deposition of non-diamond carbon.
[0026] For example, in the coarsening stage, the gas pressure was adjusted to 18kPa, and the hydrogen to methane flow ratio was 90:10. After a period of deposition, scanning electron microscopy was used again to observe that the diamond grains began to grow significantly laterally, and the amount of non-diamond carbon deposited was small, ensuring the purity and quality of the diamond.
[0027] During the finishing stage, the gas pressure is raised to 25-30 kPa and modulated with a microwave power of 500-800 W. At this stage, the precise control of gas pressure and microwave power enables fine processing of the diamond surface, improving surface flatness and crystal quality.
[0028] For example, during the finishing stage, the gas pressure was set to 28 kPa and the microwave power was adjusted to 700 W. After several hours of deposition, the diamond surface was inspected using an atomic force microscope. The results showed that the surface flatness was significantly improved and the crystal quality was also significantly improved.
[0029] The phase switching condition is to monitor the peak intensity ratio of the diamond characteristic peak to the non-diamond carbon characteristic peak through Raman spectroscopy. When the ratio is greater than 95%, the switching operation of the deposition phase is triggered.
[0030] For example, in one experiment, when the Raman spectrum detected that the peak intensity ratio of the diamond characteristic peak to the non-diamond carbon characteristic peak reached 96%, the system automatically judged and switched to the next deposition stage, ensuring the smooth progress of the entire deposition process and the stability of the diamond quality.
[0031] S4. Multi-parameter coordinated control: Gas pressure-power coupling: when the gas pressure fluctuation exceeds ±1 kPa, the microwave power is adjusted synchronously. The adjustment amplitude is calculated according to the formula to maintain the stability of the plasma, where is the gas pressure adjustment amount and is the microwave power adjustment amount.
[0032] For example, during a certain deposition process, due to external interference, the gas pressure suddenly rose by 1.5kPa. According to the formula, the microwave power needs to be increased by 15W, that is, =15W, to maintain the stable state of the plasma and ensure that the diamond deposition process is not affected.
[0033] Temperature compensation: If the substrate temperature deviation exceeds 5°C, the gas pressure setting value is corrected according to the formula to eliminate the influence of temperature change on the deposition process, where is the gas pressure adjustment amount and is the temperature deviation amount.
[0034] For example, in one experiment, due to a brief failure in the cooling system, the substrate temperature rose by 8°C. According to the temperature compensation formula, the air pressure setting value needed to be corrected and increased by 0.4kPa, that is, =0.4kPa, thereby effectively compensating for the impact of temperature changes on the deposition process and ensuring the quality of diamonds.
[0035] Gas flow coordination: When the gas pressure changes, the hydrogen and methane flow ratio is dynamically optimized according to the formula to ensure the smooth progress of the reaction, where is the gas flow change and is the gas pressure adjustment.
[0036] For example, when the gas pressure drops by 2 kPa, according to the gas flow coordination formula, the hydrogen to methane flow ratio needs to be adjusted accordingly, that is, the hydrogen flow rate is reduced and the methane flow rate is appropriately increased to maintain the balance of the reaction and ensure the quality and efficiency of diamond deposition.
[0037] S5, air pressure closed-loop control, accurately adjusts the exhaust flow through a high-speed electric needle valve, and combines with a mass flow controller to stabilize the air intake, forming a closed-loop control system to achieve dynamic adjustment of the air pressure.
[0038] During the entire deposition process, the system monitors the air pressure in real time and compares it with the set value. When the air pressure deviates, the high-speed electric needle valve responds quickly and accurately adjusts the exhaust flow according to the deviation. At the same time, the mass flow controller stabilizes the air intake, so that the air pressure quickly returns to the set value, ensuring that the entire deposition process is carried out under stable air pressure conditions.
[0039] Example 2 This embodiment provides a gas pressure control system for executing the gas pressure control method for microwave plasma chemical vapor deposition of diamond, including: The spectral feedback module consists of a quartz window, a fiber optic spectrometer, and a data processing unit. The quartz window is installed at a suitable position in the reaction chamber, allowing the spectrum emitted by the plasma to pass through and be transmitted to the fiber optic spectrometer. The fiber optic spectrometer collects the plasma spectrum in real time and transmits it to the data processing unit. The data processing unit uses the principal component analysis method to process the collected spectral data, extracting spectral characteristic parameters such as the hydrogen-to-methane intensity ratio and the Hα Stark broadening width. After processing, these parameters are transmitted to the machine learning module to provide data support for air pressure prediction.
[0040] For example, in one experiment, the spectral feedback module collected the plasma spectrum in real time. After analysis by the principal component analysis method, the data processing unit accurately extracted the hydrogen to methane intensity ratio of 3.5 and the Hα Stark broadening width of 0.3nm, and transmitted these data to the machine learning module in a timely manner.
[0041] The machine learning module uses a long short-term memory network algorithm to build a gas pressure prediction model and integrates the model into the controller. The machine learning module receives spectral characteristic parameters from the spectral feedback module, as well as real-time data such as microwave power, gas flow rate, substrate temperature, substrate material type, 2.45 GHz microwave frequency, and chamber temperature; The model uses these input data for calculations, predicts the optimal air pressure value through the trained association model, and transmits the prediction results to the multi-stage control module and the multi-parameter collaborative module to provide a basis for air pressure regulation.
[0042] For example, in a certain experiment, the machine learning module predicted that the optimal air pressure value for the current stage was 23kPa based on various input data, and promptly fed back the result to other modules to guide the adjustment of the air pressure.
[0043] The multi-stage control module automatically switches between deposition stages based on Raman spectroscopy monitoring results, adjusting gas pressure and flow parameters accordingly. The module monitors the intensity ratio of diamond-specific peaks to non-diamond carbon-specific peaks in the Raman spectrum in real time. When this ratio reaches a set switching condition greater than 95%, the deposition stage is automatically switched. At the same time, according to different deposition stages, the multi-stage control module adjusts the gas pressure and gas flow according to preset parameters. For example, in the nucleation stage, the gas pressure is set in the range of 20-30kPa, and the hydrogen to methane flow ratio is controlled to 95:5. In the coarsening stage, the gas pressure is adjusted to 15-20kPa, and the hydrogen to methane flow ratio is adjusted to 90:10.
[0044] For example, in one experiment, the multi-stage control module monitored the Raman spectrum in real time. When it detected that the intensity ratio of the diamond characteristic peak to the non-diamond carbon characteristic peak reached 96%, it automatically switched the deposition stage from the coarsening stage to the fine-tuning stage, and adjusted the gas pressure and gas flow parameters accordingly to ensure the smooth progress of the deposition process.
[0045] The multi-parameter collaborative module is used to establish a coupled control model of gas pressure, microwave power, and substrate temperature to achieve linkage regulation among various parameters. When the gas pressure fluctuates, the module collaboratively adjusts the microwave power and gas flow according to preset formulas and rules to maintain the stability of the plasma and the smooth progress of the reaction. For example, when the gas pressure fluctuation exceeds ±1kPa, the multi-parameter collaborative module calculates and adjusts the microwave power according to the gas pressure-power coupling formula. When the substrate temperature deviation exceeds 5°C, the gas pressure set value is corrected according to the temperature compensation formula. When the gas pressure changes, the hydrogen and methane flow ratio is dynamically optimized according to the gas flow collaborative formula.
[0046] For example, in one experiment, the gas pressure suddenly rose by 1.2 kPa for some reason. The multi-parameter collaborative module quickly calculated according to the formula and synchronously increased the microwave power by 12 W, while adjusting the hydrogen and methane flow ratio, effectively maintaining the stability of the plasma and the normal progress of the reaction.
[0047] The actuator includes a high-speed electric needle valve with a response time of less than 100ms, a molecular pump with a pumping speed of 500L / s, a mechanical pump with a pumping speed of 30L / s, a multi-hole spray plate gas distribution system, and a reaction chamber; The high-speed electric needle valve accurately adjusts the exhaust flow according to the system's control instructions, achieving rapid response and adjustment to the air pressure. The molecular pump and mechanical pump work together to ensure that the air pressure in the reaction chamber can meet the process requirements at different stages. The porous spray plate gas distribution system ensures uniform distribution of the reaction gas in the reaction chamber. The porous spray plate has a pore size of 0.5mm and an opening rate of 30%. In combination with the spiral guide ring, the gas flow rate deviation can be achieved to less than 5%. The reaction chamber adopts a double-layer water-cooling structure. The inner layer is quartz glass with a dielectric loss of less than 0.0002, and the outer layer is made of stainless steel. A pressure sensor with an accuracy of ±0.1kPa is configured to monitor the air pressure in real time. Its structural characteristics are compatible with the parameter monitoring and regulation process in the air pressure control method, and can provide a stable environment for diamond deposition.
[0048] For example, during the entire deposition process, the various components of the actuator work closely together. When the system detects that the air pressure needs to be adjusted, the high-speed electric needle valve moves quickly to accurately adjust the exhaust flow rate. The molecular pump and mechanical pump adjust the pumping speed as needed to maintain stable air pressure. The porous spray plate gas distribution system ensures that the reaction gas is evenly distributed. The double-layer water-cooling structure and pressure sensor of the reaction chamber ensure the stability of the chamber environment and accurate monitoring of the air pressure.
[0049] In summary, the present invention obtains plasma spectrum data through real-time spectral monitoring, extracts the intensity and broadening parameters of the characteristic spectral lines of carbon species and hydrogen atoms to reflect the plasma state, uses the long-short-term memory network model to construct a pressure-spectral feature correlation model, and combines parameters such as microwave power and gas flow to predict the optimal pressure value; During the nucleation, coarsening, and refinement stages of diamond deposition, a specific gas pressure and gas flow ratio is set, and the phase is switched based on the characteristic peak intensity ratio monitored by Raman spectroscopy. When the gas pressure fluctuates or the substrate temperature changes, multi-parameter coordinated control such as gas pressure-power coupling, temperature compensation, and gas flow coordination is used to maintain plasma stability and normal reaction progress. Finally, a closed-loop control system is formed with the help of a high-speed electric needle valve and a mass flow controller to achieve dynamic adjustment of the air pressure. The corresponding air pressure control system executes the above-mentioned air pressure control method through the coordinated operation of various functional modules such as the spectral feedback module and the machine learning module. The special design of the actuators such as the reaction chamber ensures the stability and efficiency of the entire deposition process.
[0050] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.
Claims
1. A method for controlling gas pressure in microwave plasma chemical vapor deposition of diamond, characterized in that: The specific steps include: S1. Real-time spectrum monitoring: Using an optical emission spectrometer, the emission spectrum of the plasma in the reaction chamber is continuously collected in the wavelength range of 200-800 nm. From the collected spectral data, the characteristic spectral line intensities of carbon species and hydrogen atoms, as well as the spectral line broadening parameters, are extracted. The spectral line broadening parameters are used to reflect the plasma state. S2. Machine Learning Prediction: Using a long-short-term memory network model, a correlation model between gas pressure and spectral characteristics was constructed. Using the current spectral parameters, microwave power, gas flow rate, and substrate temperature as input data, the optimal gas pressure value was predicted with an accuracy of ±0.5 kPa, providing a basis for gas pressure control. S3, multi-stage air pressure control: During the nucleation stage, the pressure in the reaction chamber is set at 20-30 kPa, and the flow ratio of hydrogen to methane is controlled at 95:5 to create an environment suitable for the formation of diamond nuclei and increase the nucleation density. During the coarsening stage, the gas pressure was reduced to 15-20 kPa, and the hydrogen to methane flow ratio was adjusted to 90:10 to accelerate the lateral growth of the grains and reduce the deposition of non-diamond carbon. During the finishing stage, the gas pressure is raised to 25-30kPa and modulated with a microwave power of 500-800W to achieve fine processing of the diamond surface and improve surface flatness and crystal quality. The stage switching condition is monitored by Raman spectroscopy. When the intensity ratio of the diamond characteristic peak to the non-diamond carbon characteristic peak is greater than 95%, the switching operation of the deposition stage is triggered; S4. Multi-parameter coordinated control: Gas pressure-power coupling: when the gas pressure fluctuation exceeds ±1 kPa, the microwave power is adjusted immediately and synchronously. The adjustment amplitude is calculated according to , so as to maintain the stability of the plasma. Where, is the gas pressure adjustment amount, and is the microwave power adjustment amount; Temperature compensation: If the substrate temperature deviation exceeds 5°C, the gas pressure setting value is corrected to eliminate the effect of temperature change on the deposition process. Where, is the gas pressure adjustment amount and is the temperature deviation amount; Gas flow coordination: When the gas pressure changes, the hydrogen and methane flow ratio is dynamically optimized to ensure the smooth progress of the reaction. Where, is the gas flow change and is the gas pressure adjustment; S5, air pressure closed-loop control, accurately adjusts the exhaust flow through a high-speed electric needle valve, and combines with a mass flow controller to stabilize the air intake to form a closed-loop control system to achieve dynamic and precise adjustment of the air pressure.
2. The gas pressure control method for microwave plasma chemical vapor deposition of diamond according to claim 1, characterized in that: In step S1, the spectral characteristic parameters include the hydrogen to methane intensity ratio and the Hα Stark broadening width, and these parameters are reduced in dimension through principal component analysis to improve data processing efficiency and accuracy.
3. The gas pressure control method for microwave plasma chemical vapor deposition of diamond according to claim 2, characterized in that: In step S2, the machine learning model is trained using historical deposition data, with input parameters also including substrate material type, microwave frequency of 2.45 GHz, and chamber temperature.
4. The gas pressure control method for microwave plasma chemical vapor deposition of diamond according to claim 3, characterized in that: In step S3, the gas pressure fluctuation range in each stage is controlled within ±1kPa, and the gas flow adjustment accuracy can reach ±0.5sccm.
5. The gas pressure control method for microwave plasma chemical vapor deposition of diamond according to claim 4, characterized in that: In the nucleation stage of step S3, negative bias enhancement technology is used with a bias range of -50V to -100V to further increase the diamond nucleation density.
6. The gas pressure control method for microwave plasma chemical vapor deposition of diamond according to claim 5, characterized in that: Oxygen doping is introduced during the finishing stage of step S2, so that the ratio of hydrogen to methane is 0.1-0.5%, which inhibits non-diamond carbon deposition and improves surface flatness.
7. A gas pressure control system for executing the gas pressure control method for microwave plasma chemical vapor deposition of diamond according to claim 6, characterized in that: It includes a spectral feedback module, a machine learning module, a multi-stage control module, a multi-parameter collaborative module and an actuator, among which; The spectrum feedback module consists of a quartz window, a fiber optic spectrometer, and a data processing unit, and is used to collect plasma spectra in real time and extract corresponding characteristic parameters; The machine learning module uses a long short-term memory network algorithm to build an air pressure prediction model and is integrated into the controller; The multi-stage control module automatically switches the deposition stage according to the Raman spectrum monitoring results and adjusts the gas pressure and gas flow parameters at the same time; The multi-parameter collaborative module is used to establish a coupled control model of gas pressure, microwave power, and substrate temperature to achieve linkage adjustment among various parameters; The actuator includes a high-speed electric needle valve with a response time of less than 100ms, a molecular pump with a pumping speed of 500L / s, a mechanical pump with a pumping speed of 30L / s, a gas distribution system of a porous spray plate, and a reaction chamber.
8. The air pressure control system according to claim 7, characterized in that: The reaction chamber adopts a double-layer water-cooling structure, the inner layer is quartz glass with a dielectric loss of less than 0.0002, the outer layer is made of stainless steel, and a pressure sensor with an accuracy of ±0.1kPa is configured to monitor the air pressure in real time.
9. The air pressure control system according to claim 8, characterized in that: The porous spray plate of the gas distribution system has an aperture of 0.5 mm and an opening rate of 30%, and when combined with the spiral guide ring, it can achieve a gas flow rate deviation of less than 5%.
10. The air pressure control system according to claim 9, characterized in that: The controller integrates PID algorithm and fuzzy control strategy to achieve an overshoot of less than 5% and an adjustment time of less than 30 seconds during dynamic adjustment of air pressure.
Citation Information
Cited By
Multi-frequency plasma resonance and thermal gradient synergistic heteroepitaxy large carat color diamond preparation system and method
CN121110177A
Molecular beam epitaxy substrate temperature field control method and system based on artificial intelligence
CN121433396A
MPCVD parameter virtual optimization control system and method based on digital twinning
CN121785128A