面向金刚石薄膜生长的MPCVD谐振腔多物理场耦合仿真与结构优化

doi:10.16553/j.cnki.issn1000-985x.2025.0261

摘要/Abstract

摘要： 采用微波等离子体化学气相沉积（MPCVD）技术生长金刚石薄膜时，工艺参数与沉积装置是决定薄膜质量的核心因素。本文以公司MPCVD设备为研究对象，基于SolidWorks与COMSOL Multiphysics平台构建谐振腔仿真模型，对比发现二维轴对称模型与三维模型的电场分布特征高度一致。基于轴对称模型，开展不同功率、压强及钼台高度工况下的等离子体仿真，结果表明：功率升高会使等离子体分布趋于扁平，且沉积平台上方激发区域显著扩大；压强升高虽能提升等离子体密度，但会缩小其团簇体积并降低分布均匀性；钼台高度较高时，等离子体的分布均匀性亦能得到有效改善。为提升等离子体激发效率，降低散热压力，首先对谐振腔整体及内部结构进行单因素优化。优化结果显示，沉积台上方中心场强达473 276 V/m，较优化前提升26.4%，同轴段次生电场强度下降11.5%。以单因素优化结果为基准，采用Box-Behnken design（BBD）响应面法对10个参数初步优化，参考方差分析结果，筛选4个关键因素进行进一步优化，最终沉积台上方中心场强较单因素优化进一步提升5.5%，验证了多变量协同优化的有效性。本研究为MPCVD设备的参数调控、结构改进及工艺优化提供了理论支撑。

关键词: 微波等离子体化学气相沉积; 谐振腔; 数值模拟; 均匀性; 响应面分析

Abstract: Microwave plasma chemical vapor deposition （MPCVD） grown diamond thin film quality is dominated by process parameters and deposition apparatus. Taking a company’s MPCVD equipment as the research object， this study constructed resonant cavity simulation models via SolidWorks and COMSOL Multiphysics. Comparative analysis shows consistent electric field distribution between the two-dimensional axisymmetric and three-dimensional models. Based on the axisymmetric model， systematic plasma simulations were performed under varying power， pressure， and molybdenum stage height. Results show that increased input power flattens plasma distribution and significantly expands the excitation region above the deposition platform； higher pressure enhances plasma density but reduces cluster volume and distribution uniformity； a higher molybdenum stage also effectively improves plasma distribution uniformity. In this study， single-factor optimization was performed on the resonant cavity （overall and internal structure） to enhance plasma excitation efficiency and reduce heat dissipation pressure. This optimization results in an increase of the central electric field strength above the deposition stage to 473 276 V/m （26.4% higher than the pre-optimization value） and an 11.5% reduction in the coaxial segment’s secondary electric field strength. Based on single-factor optimization results， ten parameters were optimized via Box-Behnken design （BBD） and response surface methodology in this study， followed by a secondary optimization of four key factors. This multi-variable collaborative optimization further improves the central electric field strength by 5.5% compared to the single-factor optimization outcome， verifying the effectiveness of multi-variable collaborative optimization. This study provides theoretical support for MPCVD equipment parameter regulation， structural improvement， and process optimization.

Key words: microwave plasma chemical vapor deposition; resonant cavity; numerical simulation; uniformity; response surface methodology

中图分类号:

O782

刘本学, 陈明明, 李霞, 王光辉, 樊永豪, 荣津悦. 面向金刚石薄膜生长的MPCVD谐振腔多物理场耦合仿真与结构优化[J]. 人工晶体学报, 2026, 55(5): 715-727.

LIU Benxue, CHEN Mingming, LI Xia, WANG Guanghui, FAN Yonghao, RONG Jinyue. Multiphysics Coupling Simulation and Structural Optimization of MPCVD Resonant Cavity for Diamond Thin Film Growth[J]. Journal of Synthetic Crystals, 2026, 55(5): 715-727.

图/表 18

图1 国内MPCVD设备等离子体分布图［8-10］

Fig.1 Plasma distributions of domestic MPCVD equipment［8-10］

图2 MPCVD设备图及结构示意图

Fig.2 Diagrams of the MPCVD equipment and its structure

图3 模型尺寸及边界条件设置

Fig.3 Schematic of model dimensions and boundary condition setup

图4 初始条件下电场分布图

Fig.4 Electric field distribution under initial conditions

图5 初始工况下腔体内关键参数分布图。（a）等离子体空间分布图；（b）等离子体存在时的电场分布图；（c）反应室温度分布图；（d）电子温度图

Fig.5 Distribution of key parameters in the cavity under initial operating conditions.（a） Spatial distribution diagram of plasma；（b） electric field distribution diagram in the presence of plasma；（c） temperature distribution diagram of the reaction chamber；（d） electron temperature diagram

图6 不同功率下的等离子体实验与仿真结果图

Fig.6 Plasma experimental and simulation results at different power levels

图7 功率梯度下等离子体密度曲线

Fig.7 Plasma density curves under power gradients

图8 不同压强下的等离子体实验与仿真结果图

Fig.8 Plasma experimental and simulation results at different pressures

图9 压强梯度下等离子体密度曲线

Fig.9 Plasma density curves under pressure gradients

图10 不同钼台高度下的等离子体分布图

Fig.10 Plasma distribution under different molybdenum stage heights

图11 钼台高度变化时等离子体密度曲线

Fig.11 Plasma density curves with varying molybdenum stage heights

图12 谐振腔本体结构参数单因素扫描结果

Fig.12 Single-factor scanning results of resonator body structural parameters

图13 谐振腔内部结构参数影响图。（a）石英玻璃高度GH；（b）引导装置半径dl；（c）沉积平台半径D；（d）沉积平台厚度h

Fig.13 Scanning diagrams of resonator internal structural parameters. （a） Height of quartz glassGH；（b） radius of guiding devicedl；（c） radius of deposition platformD；（d） thickness of deposition platformh

表1 BBD分析中4因素采样基准

Table 1 Levels of 4 factors for BBD sampling

Factor	Specific name	Reference value/mm	Minimum value/mm	Maximum value/mm
A	RadiusR₁	106.4	105.7	107.1
B	HeightH₂	102.0	101.8	102.2
C	HeightH₃	14.45	14.25	14.65
D	RadiusR₃	148.5	144.0	153.0

表2 4因素影响下拟合统计指标

Table 2 Fitted statistical metrics for 4 factors impact

Coefficient of determination （R²）	0.991 4
AdjustedR²	0.982 7
PredictedR²	0.950 2
Adequate precision	24.471 9

图14 各因素对响应量的交互作用图

Fig.14 Interaction effect diagrams of various factors on the response variable

表3 响应面优化方案

Table 3 Optimization scheme based on response surface method

Optimization scheme	1st	2nd	3rd
Radius，R₁/mm	106.584	106.581	106.590
Height，H₂/mm	102.022	102.024	102.020
Height，H₃/mm	14.250	14.250	14.250
Radius，R₃/mm	151.295	151.288	151.310
Predicted value/（V·m^-1）	491 229	491 229	491 228
Simulated value/（V·m^-1）	498 937	498 964	499 261
Error/%	1.57	1.57	1.64

图15 优化前、后电场强度对比

Fig.15 Comparison of electric field intensity before and after optimization

参考文献 23

[1]	LI L J， AN K， XU G Y， et al. Twins and dark features in MPCVD diamond films［J］. Applied Surface Science， 2025， 701： 163222.
[2]	HUANG H Z， HUANG C， DENG L， et al. Microstructure and wear resistance of the CuSn₁₉Ti₁₀/diamond composite coatings on copper substrate by laser cladding［J］. Diamond and Related Materials， 2024， 150： 111667.
[3]	权　乐. CVD金刚石膜研究进展［J］. 真空科学与技术学报， 2024， 44（10）： 841-852.
	QUAN L. Research progress of CVD diamond film［J］. Chinese Journal of Vacuum Science and Technology， 2024， 44（10）： 841-852 （in Chinese）.
[4]	KAMO M， SATO Y， MATSUMOTO S， et al. Diamond synthesis from gas phase in microwave plasma［J］. Journal of Crystal Growth， 1983， 62（3）： 642-644.
[5]	GUO Z F， GUO B， ZHANG J F， et al. CVD diamond processing tools： a review［J］. Journal of Advanced Research， 2025， 74： 333-358.
[6]	闫新胜. 20 kW/915 MHz微波等离子体装置的设计与研究［D］. 合肥：中国科学技术大学， 2021.
	YAN X S. Design and research of a 20 kW/915 MHz microwave plasma device［D］. Hefei： University of Science and Technology of China， 2021 （in Chinese）.
[7]	LI X J， ZHENG S Q， ZHAO B R， et al. Design and numerical simulation of novel reentrant microwave cavity［J］. Physics Procedia， 2011， 22： 101-106.
[8]	SU J J， LI Y F， DING M H， et al. A dome-shaped cavity type microwave plasma chemical vapor deposition reactor for diamond films deposition［J］. Vacuum， 2014， 107： 51-55.
[9]	LI Y F， SU J J， LIU Y Q， et al. Design of a new TM₀₂₁ mode cavity type MPCVD reactor for diamond film deposition［J］. Diamond and Related Materials， 2014， 44： 88-94.
[10]	WANG Q J， WU G， LIU S， et al. Simulation-based development of a new cylindrical-cavity microwave-plasma reactor for diamond-film synthesis［J］. Crystals， 2019， 9（6）： 320.
[11]	LOMBARDI G， HASSOUNI K， STANCU G D， et al. Modeling of microwave discharges of H₂ admixed with CH₄ for diamond deposition［J］. Journal of Applied Physics， 2005， 98（5）： 053303.
[12]	FÜNER M， WILD C， KOIDL P. Simulation and development of optimized microwave plasma reactors for diamond deposition［J］. Surface and Coatings Technology， 1999， 116： 853-862.
[13]	PLEULER E， WILD C， FÜNER M， et al. The CAP-reactor， a novel microwave CVD system for diamond deposition［J］. Diamond and Related Materials， 2002， 11（3/4/5/6）： 467-471.
[14]	HASSOUNI K， SILVA F， GICQUEL A. Modelling of diamond deposition microwave cavity generated plasmas［J］. Journal of Physics D： Applied Physics， 2010， 43（15）： 153001.
[15]	HAO X B， LI Y C， KONG X， et al. Optimizing MPCVD systems for diamond growth through advanced microwave transmission theory［J］. Journal of Crystal Growth， 2025， 651： 128008.
[16]	GICQUEL A， CHENEVIER M， HASSOUNI K， et al. Validation of actinometry for estimating relative hydrogen atom densities and electron energy evolution in plasma assisted diamond deposition reactors［J］. Journal of Applied Physics， 1998， 83（12）： 7504-7521.
[17]	ACHARD J， TALLAIRE A， SUSSMANN R， et al. The control of growth parameters in the synthesis of high-quality single crystalline diamond by CVD［J］. Journal of Crystal Growth， 2005， 284（3/4）： 396-405.
[18]	张一卓. 新型MPCVD金刚石膜沉积装置模拟及实验研究［D］. 太原：太原理工大学， 2022.
	ZHANG Y Z. Simulation and experimental study on a novel MPCVD diamond film deposition device［D］. Taiyuan： Taiyuan University of Technology， 2022 （in Chinese）.
[19]	MU C Y， LI G Z， LV X Y， et al. Effect of hydrogen flow rate on the plasma state and quality of diamond crystals grown by MPCVD［J］. Materials Science in Semiconductor Processing， 2025， 185： 108884.
[20]	徐　磊. MPCVD装置的多物理场仿真及优化研究［D］. 成都：电子科技大学， 2025.
	XU L. Research on multi-physics simulation and optimization of MPCVD equipment［D］. Chengdu： University of Electronic Science and Technology of China， 2025 （in Chinese）.
[21]	MARQUES L， JOLLY J， ALVES L L. Capacitively coupled radio-frequency hydrogen discharges： the role of kinetics［J］. Journal of Applied Physics， 2007， 102（6）： 063305.
[22]	YANG Z L， AN K， FENG X R， et al. Explore the growth mechanism of high-quality diamond under high average power density in the MPCVD reactor［J］. Materials Science and Engineering： B， 2024， 302： 117248.
[23]	LIANG X B， WANG L， ZHU H L， et al. Effect of pressure on nanocrystalline diamond films deposition by hot filament CVD technique from CH₄/H₂ gas mixture［J］. Surface and Coatings Technology， 2007， 202（2）： 261-267.