High-frequency enhancement-mode millimeterwave AlGaN/GaN HEMT with an fT/fmax over 100 GHz/200 GHz

Ultra-thin barrier (UTB) 4-nm-AlGaN/GaN normally-off high electron mobility transistors (HEMTs) having a high current gain cut-off frequency (f_T) are demonstrated by the stress-engineered compressive SiN trench technology. The compressive in-situ SiN guarantees the UTB-AlGaN/GaN heterostructure can operate a high electron density of 1.27×10¹³cm^-2, a high uniform sheet resistance of 312.8 Ω /□, but a negative threshold for the short-gate devices fabricated on it. With the lateral stress-engineering by full removing in-situ SiN in the 600-nm SiN trench, the short-gated (70 nm) devices obtain a threshold of 0.2 V, achieving the devices operating at enhancement-mode (E-mode). Meanwhile, the novel device also can operate a large current of 610 mA/mm and a high transconductance of 394 mS/mm for the E-mode devices. Most of all, a high f_T/f_max of 128 GHz/255 GHz is obtained, which is the highest value among the reported E-mode AlGaN/GaN HEMTs. Besides, being together with the 211 GHz/346 GHz of f_T/f_max for the D-mode HEMTs fabricated on the same materials, this design of E/D-mode with the realization of f_max over 200 GHz in this work is the first one that can be used in Q-band mixed-signal application with further optimization. And the minimized processing difference between the E- and D-mode designs the addition of the SiN trench, will promise an enormous competitive advantage in the fabricating costs.     

GaN/GaAs(1 0 0) superlattices grown by metalorganic vapor phase epitaxy using dimethylhydrazine precursor

Superlattices of cubic gallium nitride (GaN) and gallium arsenide (GaAs) were grown on GaAs(1 0 0) substrates using metalorganic vapor phase epitaxy (MOVPE) with dimethylhydrazine (DMHy) as nitrogen source. Structures grown at low temperatures with varying layer thicknesses were characterized using high resolution X-ray diffraction and atomic force microscopy. Several growth modes of GaAs on GaN were observed: step-edge, layer-by-layer 2D, and 3D island growth. A two-temperature growth process was found to yield good crystal quality and atomically flat surfaces. The results suggest that MOVPE-grown thin GaN layers may be applicable to novel GaAs heterostructure devices.     

具有高开启/关断电流比的Al2O3/AlGaN/GaN金属氧化物半导体高电子迁移率晶体管

采用原子层淀积(ALD)方法,制备了Al₂O₃为栅介质的高性能AlGaN/GaN金属氧化物半导体高电子迁移率晶体管(MOS-HEMT)。在栅压为-20 V时,MOS-HEMT的栅漏电比Schottky-gate HEMT的栅漏电低4个数量级以上。在栅压为+2 V时,Schottky-gate HEMT的栅漏电为191μA;在栅压为+20 V时,MOS-HEMT的栅漏电仅为23.6 nA,比同样尺寸的Schottky-gate HEMT的栅漏电低将近7个数量级。AlGaN/GaN MOS-HEMT的栅压摆幅达到了±20 V。在栅压V_gs=0 V时, MOS-HEMT的饱和电流密度达到了646 mA/mm,相比Schottky-gate HEMT的饱和电流密度(277 mA/mm)提高了133%。栅漏间距为10μm的AlGaN/GaN MOS-HEMT器件在栅压为+3 V时的最大饱和输出电流达到680 mA/mm,特征导通电阻为1.47 mΩ·cm²。Schottky-gate HEMT的开启与关断电流比仅为10⁵,MOS-HEMT的开启与关断电流比超过了10⁹,超出了Schottky-gate HEMT器件4个数量级,原因是栅漏电的降低提高了MOS-HEMT的开启与关断电流比。在V_gs=-14 V时,栅漏间距为10μm的AlGaN/GaN MOS-HEMT的关断击穿电压为640 V,关断泄露电流为27μA/mm。     

具有高阈值电压和大栅压摆幅的常关型槽栅AlGaN/GaN金属氧化物半导体高电子迁移率晶体管

介绍了一种具有高阈值电压和大栅压摆幅的常关型槽栅AlGaN/GaN金属氧化物半导体高电子迁移率晶体管。采用原子层淀积(ALD)方法实现Al₂O₃栅介质的沉积。槽栅常关型AlGaN/GaN MOS-HEMT的栅长(L_g)为2 μm,栅宽(W_g)为0.9 mm(0.45 mm×2),栅极和源极(L_gs)之间的距离为5 μm,栅极和漏极(L_gd)之间的距离为10 μm。在栅压为－20 V时,槽栅常关型AlGaN/GaN MOS-HEMT的栅漏电仅为0.65 nA。在栅压为+12 V时,槽栅常关型AlGaN/GaN MOS-HEMT的栅漏电为225 nA。器件的栅压摆幅为－20~+12 V。在栅压V_gs=+10 V时,槽栅常关型AlGaN/GaN MOS-HEMT电流和饱和电流密度分别达到了98 mA和108 mA/mm (W_g=0.9 mm), 特征导通电阻为4 mΩ·cm²。槽栅常关型AlGaN/GaN MOS-HEMT的阈值电压为+4.6 V,开启与关断电流比达到了5×10⁸。当V_ds=7 V时,器件的峰值跨导为42 mS/mm (W_g=0.9 mm,V_gs=+10 V)。在V_gs=0 V时,栅漏间距为10 μm的槽栅常关型AlGaN/GaN MOS-HEMT的关断击穿电压为450 V,关断泄露电流为0.025 mA/mm。     

Ga空位对GaN:Gd体系磁性影响的第一性原理研究

采用基于密度泛函理论的第一性原理结合投影缀加平面波的方法,研究了GaN中Ga被稀土元素Gd替代以及与邻近N或Ga空位组成的缺陷复合体的晶格常数、磁矩、形成能以及电子结构等性质.结果发现,Gd掺杂GaN后禁带宽度变窄,由直接带隙半导体转为间接带隙半导体;单个Gd原子掺杂给体系引入大约7μB的磁矩;在Gd与Ga或N空位形成的缺陷复合体系中,N空位对引入磁矩贡献很小,大约0.1μB,Ga空位能引入约2μB的磁矩.随着Ga空位的增多,体系总磁矩增加,但增加量与Ga空位的位置分布密切相关.当Ga空位分布较为稀疏时,Gd单原子磁矩受影响较小,但当Ga空位距离较近且倾向于形成团簇时,Gd单原子磁矩明显增加,而且这种情况下空位形成能也最小.     

MOCVD生长的未掺杂GaN的结晶特性与补偿度关系的研究

对本实验室用ＭＯＣＶＤ方法生长的未故意掺杂的ＧａＮ单晶膜进行了结晶性能、电学性能研究。结果表明,室温时ＧａＮ的Ｘ射线双晶衍射半高宽与其补偿度有较强的依赖关系。高补偿的ＧａＮ的Ｘ射线双晶衍射半高宽较宽。低补偿的ＧａＮ的Ｘ射线双晶衍射半高宽较窄。     

三代微光像增强器GaAs负电子亲合势（NEA）光电阴极稳定性研究

针对三代管ＧａＡｓ阴极灵敏度下降问题,开展工艺过程质量研究,分析影响因素并经工艺改进解决阴极不稳定问题     

气源MBE外延自组装GeSi量子点光致荧光研究

利用气源分子束外延设备（MBE）制作了GeSi自组装量子点样品。利用原子力显微镜（AFM）和光致荧光（PL）光谱研究了量子点的形貌和光学性质。气源MBE在较低温度下生长的量子点材料具有较高的量子点覆盖度。200K以下载流子以局域激子形式束缚在量子点中,激子束缚能约为17meV。升温至200K,载流子的输运过程发生变化。对量子点PL积分强度与温度关系曲线进行拟合得到量子点中空穴跃迁至浸润层的热激活能为129meV。     

硅衬底生长的InGaN/GaN多层量子阱中δ型硅掺杂n-GaN层对载流子复合过程的调节作用

为了解决在单晶硅衬底上生长的InGaN/GaN多层量子阱发光二极管器件发光效率显著降低的问题,使用周期性δ型Si掺杂的GaN取代Si均匀掺杂的GaN作为n型层释放多层界面间的张应力。采用稳态荧光谱及时间分辨荧光谱测量,提取并分析了使用该方案前后的多层量子阱中辐射/非辐射复合速率随温度（10~300 K）的变化规律。实验结果表明引入δ-Si掺杂的n-GaN层后,非辐射复合平均激活能由（18±3）meV升高到（38±10）meV,对应非辐射复合速率随温度升高而上升的趋势变缓,室温下非辐射复合速率下降,体系中与阱宽涨落有关的浅能级复合中心浓度减小,PL峰位由531 nm左右红移至579 nm左右,样品PL效率随温度的衰减受到抑制。使用周期性δ型Si掺杂的GaN取代Si均匀掺杂的GaN作为生长在Si衬底上的InGaN/GaN多层量子阱LED器件n型层,由于应力释放,降低了多层量子阱与n-GaN界面、InGaN/GaN界面的缺陷密度,使得器件性能得到了改善。     

过渡金属掺杂的g-GaN吸附Cl2和CO气体分子的第一性原理研究

基于密度泛函理论的第一性原理计算, 系统研究了类石墨烯氮化镓(g-GaN)和掺杂过渡金属原子(TM)的 g-GaN 对 Cl₂和CO气体分子的吸附行为。结果表明, Cl₂和 CO在本征 g-GaN上的吸附均为物理吸附, 2个体系的吸附能均为正值, 表明体系不稳定。相反, Cl₂和 CO在 Fe和 Co掺杂的 g-GaN上吸附时的吸附能为负值, 且吸附能较小, 表明吸附体系稳定。通过分析态密度、电荷密度差和能带结构等性质, 可以得出结论:过渡金属原子的引入能有效增强气体分子与 g-GaN之间的相互作用。