为了更精确预测基于 0.18 μm工艺的 40 V NLDMOS器件实际应用条件下的热载流子寿命,介绍了一种通过可靠性测试和计算机辅助数学解析相结合的热载流子可靠性寿命预计方法.该方法基于实际直流状态下的热载流注入测试数据,结合TCAD仿真,对测试结果进行了线性函数、多次函数、幂函数、指数函数及Dreesen R函数拟合;通过分析当前业界LDMOS器件的热载流子注入测试主流模型预测精度的局限性,找出了最优热载流子模型,提出了适合Python语言编程的改进型Dreesen R模型;通过数学解析推导方法以及基于Python语言的计算机辅助编程计算,得出了栅极以及漏极全工作电压范围内的热载流子参数退化曲线;通过模拟工作波形不同上升沿及下降沿的函数曲线、上升及下降时间以及不同占空比,得出随着时间变化的交直流转换因子曲线.最终新的测试项目可以通过不同电压下的直流状态下测试结果以及已经得到的交直流转换因子曲线,来直接获取工作场景交流状态的热载流子寿命.该评估方法解决了采用直流状态下的测试来解决现场复杂应用波形的热载流子寿命评估难题,较大节省了测试时间,提高了寿命预测精度.
Study on hot carrier lifetime of a 40V NLDMOS device
In order to predict the hot carrier lifetime of 40 V NLDMOS device based on 0.18 μm process more accurately,a hot carrier reliability lifetime prediction method combining reliability testing and computer-aided mathematical analysis is introduced in this paper.The method is based on the actual DC hot carrier injection test data,combined with TCAD simulation,and the test results are fitted with linear function,multiple function,power function,exponential function and Dreesen R function.By analyzing the limitation of the prediction accuracy of the mainstream model of hot carrier injection test of LDMOS devices in the current industry.The optimal hot carrier model is found out,and an improved Dreesen R model suitable for Python programming is proposed.The degradation curves of hot carrier parameters in the full voltage range of gate and drain are obtained by mathematical analysis and computer-aided programming based on Python language.By simulating the function curve of different rising edge and falling edge,rising and falling time and different duty cycle,the AC/DC conversion factor curve with time change is obtained.Finally,the new test project can directly obtain the hot carrier lifetime of AC state in the working scene through the test results of DC state under different voltages and the obtained AC/DC conversion factor curve.The evaluation method solves the difficult problem of hot carrier lifetime evaluation of complex waveforms in field by testing under DC state,greatly saves the test time and improves the life prediction accuracy.
40 V NLDMOSHCIreliabilityAC/DC conversion factorlifetime predictionKirk effect
NETL
NSTL
万方数据
