Ultrasonic-assisted chemical mechanical polishing(UA-CMP)can greatly improve the sapphire material removal and surface quality,but its polishing mechanism is still unclear.This paper proposed a novel model of material...Ultrasonic-assisted chemical mechanical polishing(UA-CMP)can greatly improve the sapphire material removal and surface quality,but its polishing mechanism is still unclear.This paper proposed a novel model of material removal rate(MRR)to explore the mechanism of sapphire UA-CMP.It contains two modes,namely two-body wear and abrasive-impact.Furthermore,the atomic force microscopy(AFM)in-situ study,computational fluid dynamics(CFD)simulation,and polishing experiments were conducted to verify the model and reveal the polishing mechanism.In the AFM in-situ studies,the tip scratched the reaction layer on the sapphire surface.The pit with a 0.22 nm depth is the evidence of two-body wear.The CFD simulation showed that abrasives could be driven by the ultrasonic vibration to impact the sapphire surface at high frequencies.The maximum total velocity and the air volume fraction(AVF)in the central area increased from 0.26 to 0.55 m/s and 20%to 49%,respectively,with the rising amplitudes of 1–3μm.However,the maximum total velocity rose slightly from 0.33 to 0.42 m/s,and the AVF was nearly unchanged under 40–80 r/min.It indicated that the ultrasonic energy has great effects on the abrasive-impact mode.The UA-CMP experimental results exhibited that there was 63.7%improvement in MRR when the polishing velocities rose from 40 to 80 r/min.The roughness of the polished sapphire surface was R_(a)=0.07 nm.It identified that the higher speed achieved greater MRR mainly through the two-body wear mode.This study is beneficial to further understanding the UA-CMP mechanism and promoting the development of UA-CMP technology.展开更多
Material removal mechanism under non-contact condition between the pad and the wafer in the chemical mechanical polishing (CMP) process is investigated. Based on the assumption that almost all effective material remov...Material removal mechanism under non-contact condition between the pad and the wafer in the chemical mechanical polishing (CMP) process is investigated. Based on the assumption that almost all effective material removals take place due to the active abrasives which cut material through the plowing effects. A novel model is developed to predict the material removal rate (MRR) under non-contact condition between the pad and the wafer in CMP. Validated by the experimental data, the model is proved to be able to predict the change of MRR under non-contact condition. Numerical simulation of the model shows: the relative velocity u between the pad and the wafer and fluid viscosity η are the most important factors which impact MRR under non-contact condition; load changes of wafer also affects the MRR, but the effect is not as obvious as the relative velocity and fluid viscosity; when the radius of abrasive is not less than 50nm, the impact of MRR alone with the changes in the size of the abrasive can be ignored.展开更多
Physics equation-based semiconductor device modeling is accurate but time and money consuming.The need for studying new material and devices is increasing so that there has to be an efficient and accurate device model...Physics equation-based semiconductor device modeling is accurate but time and money consuming.The need for studying new material and devices is increasing so that there has to be an efficient and accurate device modeling method. In this paper, two methods based on multivariate rational regression（MRR） for device modeling are proposed. They are single-pole MRR and double-pole MRR. The two MRR methods are proved to be powerful in nonlinear curve fitting and have good numerical stability. Two methods are compared with OLS and LASSO by fitting the SMIC 40 nm MOS-FET I–V characteristic curve and the normalized mean square error of Single-pole MRR is 3.02 × 10^-8 which is 4 magnitudes less than an ordinary least square. The I–V characteristics of CNT-FET and performance indicators（noise factor, gain, power） of a low noise amplifier are also modeled by using MRR methods. The results show MRR methods are very powerful methods for semiconductor device modeling and have a strong nonlinear curve fitting ability.展开更多
基金This work was supported by the National Natural Science Foundation of China(Nos.51865030 and 52165025).
文摘Ultrasonic-assisted chemical mechanical polishing(UA-CMP)can greatly improve the sapphire material removal and surface quality,but its polishing mechanism is still unclear.This paper proposed a novel model of material removal rate(MRR)to explore the mechanism of sapphire UA-CMP.It contains two modes,namely two-body wear and abrasive-impact.Furthermore,the atomic force microscopy(AFM)in-situ study,computational fluid dynamics(CFD)simulation,and polishing experiments were conducted to verify the model and reveal the polishing mechanism.In the AFM in-situ studies,the tip scratched the reaction layer on the sapphire surface.The pit with a 0.22 nm depth is the evidence of two-body wear.The CFD simulation showed that abrasives could be driven by the ultrasonic vibration to impact the sapphire surface at high frequencies.The maximum total velocity and the air volume fraction(AVF)in the central area increased from 0.26 to 0.55 m/s and 20%to 49%,respectively,with the rising amplitudes of 1–3μm.However,the maximum total velocity rose slightly from 0.33 to 0.42 m/s,and the AVF was nearly unchanged under 40–80 r/min.It indicated that the ultrasonic energy has great effects on the abrasive-impact mode.The UA-CMP experimental results exhibited that there was 63.7%improvement in MRR when the polishing velocities rose from 40 to 80 r/min.The roughness of the polished sapphire surface was R_(a)=0.07 nm.It identified that the higher speed achieved greater MRR mainly through the two-body wear mode.This study is beneficial to further understanding the UA-CMP mechanism and promoting the development of UA-CMP technology.
基金National Natural Science Foundation of China (Grant No. 50705006)
文摘Material removal mechanism under non-contact condition between the pad and the wafer in the chemical mechanical polishing (CMP) process is investigated. Based on the assumption that almost all effective material removals take place due to the active abrasives which cut material through the plowing effects. A novel model is developed to predict the material removal rate (MRR) under non-contact condition between the pad and the wafer in CMP. Validated by the experimental data, the model is proved to be able to predict the change of MRR under non-contact condition. Numerical simulation of the model shows: the relative velocity u between the pad and the wafer and fluid viscosity η are the most important factors which impact MRR under non-contact condition; load changes of wafer also affects the MRR, but the effect is not as obvious as the relative velocity and fluid viscosity; when the radius of abrasive is not less than 50nm, the impact of MRR alone with the changes in the size of the abrasive can be ignored.
文摘Physics equation-based semiconductor device modeling is accurate but time and money consuming.The need for studying new material and devices is increasing so that there has to be an efficient and accurate device modeling method. In this paper, two methods based on multivariate rational regression（MRR） for device modeling are proposed. They are single-pole MRR and double-pole MRR. The two MRR methods are proved to be powerful in nonlinear curve fitting and have good numerical stability. Two methods are compared with OLS and LASSO by fitting the SMIC 40 nm MOS-FET I–V characteristic curve and the normalized mean square error of Single-pole MRR is 3.02 × 10^-8 which is 4 magnitudes less than an ordinary least square. The I–V characteristics of CNT-FET and performance indicators（noise factor, gain, power） of a low noise amplifier are also modeled by using MRR methods. The results show MRR methods are very powerful methods for semiconductor device modeling and have a strong nonlinear curve fitting ability.
