In this paper, a novel high voltage superjunction lateral double diffused MOSFETs (SJ-LDMOS) using a variable high permittivity (VHK) dielectric trench is presented. A relatively high HK dielectric is in the upper trench, which is connected with the drain electrode to suppress the high electric field (E-field) peak under the drain by the dielectric reduced surface field (RESURF) effect. In addition, a relatively low HK dielectric is at the bottom of the trench. On the one hand, the substrate is effectively depleted by a suitable HK dielectric layer, and the vertical depletion region of the substrate is greatly expanded. On the other hand, the overall vertical bulk E-field distribution is modulated by the E-field peaks generated at the position of varying K dielectric. A more uniform bulk E-field distribution is obtained for VHK SJ-LDMOS, leading to a high breakdown voltage (BV). Compared to the conventional SJ-LDMOS, the blocking voltage per micron of the drift region of VHK SJ-LDMOS has increased by 41.2%. Besides, compared with the SJ-LDMOS with a uniform-K, the BV of VHK SJ-LDMOS is improved by about 9.5%. The condition of the optimal range of the variable high permittivity is also presented. Meanwhile, the proposed VHK SJ-LDMOS has good conduction characteristics and heat dissipation
This paper presents a machine learning-based figure of merit model for superjunction (SJ) U-MOSFET (SSJ-UMOS) with a modulated drift region utilizing semi-insulating poly-crystalline silicon (SIPOS) pillars. This SJ drift region modulation is achieved through SIPOS pillars beneath the trench gate, focusing on optimizing the tradeoff between breakdown voltage (BV) and specific ON-resistance (RON,sp). This analytical model considers the effects of electric field modulation, charge-coupling, and majority carrier accumulation due to additional SIPOS pillars. Gaussian process regression is employed for the figure of merit (FOM = BV2/RON,sp) prediction and hyperparameter optimization, ensuring a reasonable and accurate model. A methodology is devised to determine the optimal BV-RON,sp tradeoff, surpassing the SJ silicon limit. The paper also delves into a discussion of optimal structural parameters for drift region, oxide thickness, and electric field modulation coefficients within the analytical model. The validity of the proposed model is robustly confirmed through comprehensive verification against TCAD simulation results.
We report a GaN-on-Si AlGaN/GaN high-electron-mobility transistor (HEMT) incorporating a drain-side Pb(Zr,Ti)O3 (PZT) trench for electrostatic field management. The mechanism is formulated using the displacement-field boundary conditions and evaluated by two-dimensional technology computer aided design with a field-independent relative-permittivity sweep from 3.9 to 500. The calculations reveal the nonmonotonic dependence of terminal voltage on permittivity and a broader drain-side high-field distribution for the nominal εr=300 case. Experimentally, a representative PZT-trench HEMT reaches 1405 V at LGD=9 μm, compared with 783 V for a co-fabricated conventional device and 957 V for a previously reported hybrid Schottky–ohmic drain HEMT, corresponding to breakdown-voltage improvements of 79% and 47%, respectively, while retaining a specific ON-resistance of 1.47 mΩ cm2. The dynamic-to-static ON-resistance ratios, RON,dyn/RON,static, are 1.6× and 2.5× after 900 and 1200 V off-state stress, respectively, at a 3 μs extraction delay. These results demonstrate simultaneous enhancement of voltage handling and dynamic performance through drain-side dielectric-boundary engineering.