基于有限元法的超声辅助线锯单晶硅表面粗糙度影响规律

Influence of process parameters on surface roughness of ultrasonic-assisted wire sawed monocrystalline silicon based on finite element method

为研究超声振动辅助切割单晶硅对硅片的表面粗糙度的影响,利用ABAQUS软件建立单颗磨粒切割单晶硅的二维仿真模型,通过响应面法(RSM)分析仿真数据,获得单晶硅表面粗糙度与设定的线切割工艺参数之间的交互关系,研究单颗磨粒在不同工艺参数下对单晶硅表面粗糙度的影响。研究结果表明,通过建立评价表面粗糙度与工艺参数之间的回归模型,并对模型进行目标优化,得到最佳的工艺参数为:超声频率f=31 kHz,超声振幅A=10μm,磨粒速度vl=28 m/s,磨粒倾角θ=-15°,磨粒半锥角φ=60°。

In order to study the influence of ultrasonic vibration-assisted cutting of monocrystalline silicon on the surface roughness of silicon wafers,a two-dimensional simulation model of single abrasive grain cutting monocrystalline silicon was established by ABAQUS software,and the interaction between the surface roughness of monocrystalline silicon and the set wire-cutting process parameters was obtained by analyzing the simulation data by response surface method(RSM),and the influence of single abrasive grain on the surface roughness of monocrystalline silicon under different process parameters was studied.Results show that by establishing a regression model to evaluate the relationship between surface roughness and process parameters,and optimizing the model,the optimal process parameters are as follows:ultrasonic frequency f=31 kHz,ultrasonic amplitude A=10μm,abrasive velocity v l=28 m/s,abrasive inclination angleθ=-15°,abrasive half cone angleφ=60°.