1.Engineering Research Center of Internet of Things Technology Applications (Ministry of Education), Department of Electronic Engineering, Jiangnan University, Wuxi 214122, China 2.Beijing Engineering Research Center of High-reliability IC with Power Industrial Grade, Beijing Smart-Chip Microelectronics Technology Co., Ltd., Beijing 102200, China 3.State Key Discipline Laboratory of Wide Band-gap Semiconductor Techonology, Xidian University, Xi’an 710071, China
Fund Project:Project supported by the National Natural Science Foundation of China (Grant Nos. 61504050, 11604124, 51607022) and the Laboratory Open Fund of Beijing Smart-chip Microelectronics Technology Co., Ltd., China
Received Date:18 August 2020
Accepted Date:17 November 2020
Available Online:19 March 2021
Published Online:05 April 2021
Abstract:In this paper, the temperature-dependent current-voltage (T-I-V) characteristics of lattice-matched InAlN/GaN heterostructure Schottky contact in a reverse direction are measured, and the voltage dependence and temperature dependence of the leakage current are studied. The obtained results are as follows.1) The reverse current is a strong function of voltage and temperature, and the saturation current is much larger than the theoretical value, which cannot be explained by the classical thermionic emission (TE) model. 2) In the low-bias region, the $ \ln(I/E)\text{-}E^{1/2} $ data points obey a good linear relationship, whose current slope and corresponding activation energy are close to the values predicted by the Frenkel-Poole (FP) model, indicating the dominant role of the FP emission mechanism. 3) In the high-bias region, the $ \ln(I/E^2)\text{-}E^{-1} $data points also follow a linear dependence, but the current slope is a weak function of temperature, indicating that the Fowler-Nordheim tunneling mechanism should be mainly responsible for the leakage current. 4) The current barrier height is extracted to be about 0.60 eV, which is much lower than the value of 2.91 eV obtained from the TE model, confirming the primary leakage path of the conductive dislocations, where the localized barrier is significantly reduced due to the ionization of shallow donor-like traps. Keywords:reverse leakage current/ bias and temperature/ conductive dislocations/ shallow donor state
3.结果与讨论图2(a)所示为晶格匹配InAlN/GaN异质结肖特基接触反向T-I-V和C-V特性曲线. 随着反向偏压的增大, 电流先快速增加然后增长速度变得缓慢, 并出现负温度系数. 在–4 — –6.5 V区域, 电容随着反向电压的增大迅速减小, 表明此时二维电子气(2DEG)浓度急剧减小, 在V = –6.5 V时沟道夹断, Vpin-off = –6.5 V. 2DEG面浓度(n2DEG)和反向偏压的关系可由以下关系式获得 图 2 (a) 不同温度下InAlN/GaN肖特基势垒二极管的I-V特性和常温下典型的C-V特性曲线; (b)电场与电压的函数关系图和n2 DEG对电压的依赖关系 Figure2. (a) The reverse I-V characteristics of InAlN / GaN Schottky diode measured at various temperatures and typical C-V curve at room temperature; (b)electric field and n2 DEG as a function of gate voltage.
式中, σFP为FP导电系数, qφFP是零电场时的发射势垒高度. (5a)式—(5b)式表明, 若FP发射电流占主导, 则ln(IFP/E)与E1/2数据应满足线性关系, 且斜率对应一固定的热激活能, q(q/πεε0)1/2~2.53 × 10–4 eV. 图5(a)和图5(b)所示分别为不同温度下ln(I/E)-E1/2关系以及相应的电流斜率. 可以看出, ln(I/E)-E1/2数据满足很好的线性依赖关系, 相应的电流斜率与理论值非常接近, 其激活能约为2.64 × 10–4 eV, 与理论值一致. 因此, 在低偏压下的电流应主要为FP发射电流. 图 5 (a)在不同温度下ln(I/E)和E1/2的关系; (b)相应的斜率βFP Figure5. (a) The relationship between ln(I/E) and E1/2 at different temperatures; (b) the corresponding current slope βFP.
实验表明, 位错是GaN器件反向漏电流的主要输运通道[20,21]. 图6(a)所示为可导位错的物理模型: 1)位错整体为电中性; 2) Ga空位居于位错中心, 为深能级受主态, 捕获电子后为负电性; 3)位错周围为ON缺陷, 为潜能级施主态, 电离后为正电性; 4)受主态获得电子后势垒抬高, 施主态失去电子后势垒降低, 成为有效的局域导电通道. 图6(b)所示为FP发射过程示意图: 费米能级附近, 界面施主态内的电子, 在电场辅助作用下向位错最低导带热发射过程. (5c)式表明, FP电流的截距具有固定的热激活能, 其数值即为零电场下的势垒高度. 图7所示为测试获得的c(T)-q/kT关系, 数据遵循很好的线性依赖特性, 与FP输运模型一致. 通过线性拟合获得FP电流的有效势垒高度约为0.59 eV. 该值远低于热发射电流势垒高度2.91 eV, 进一步证明了位错应是反向电流的主要输运通道. 图 6 (a) GaN中可导位错模型示意图; (b) FP 发射电流输运过程示意图 Figure6. (a) Schematic band diagram of the conductive dislocations in GaN; (b) schematic transport process of the FP emission current
图 7c(T)和q/kT的关系 Figure7. The relationship between c(T) and q/kT.
图8(a)所示为变温的高频C-V曲线. 可以看到, 随T从300 K升高至500 K, 器件的夹断电压向正方向移动, 由–6.4 V逐渐增大至–6 V. 如图8(b)黑色数据. 此时, 所对应的饱和电场可近似表示为 图 8 (a)在不同温度下的C-V曲线; (b)不同夹断电压所对应的电场关系 Figure8. (a) The C-V curves measured at different tempratures; (b) the corresponding electric field with different pinch-off voltages.
式中 mn* 为电子有效质量, qφFN为有效隧穿势垒高度, h为普朗克常量. 图9(a)所示为不同温度下的ln(I/E2)–E–1关系. 可以看到, 数据满足很好的线性关系, 且电流斜率基本不随温度改变, 与FN隧穿特性一致. 根据(8b)式, 有效势垒高度与电流斜率的关系可写为 图 9 (a)不同温度下ln(I/E2)和E–1关系图; (b) FN隧穿有效势垒高度随温度的变化 Figure9. (a) The relationship between ln(I/E2) and E–1 at different temperatures; (b) the effective barrier height extracted based on FN tunneling model.
图 1 (a) InAlN/GaN肖特基势垒二极管横截面结构示意图; (b)俯视图
图 2 (a) 不同温度下InAlN/GaN肖特基势垒二极管的I-V特性和常温下典型的C-V特性曲线; (b)电场与电压的函数关系图和n2 DEG对电压的依赖关系
图 3 不同温度下InAlN/GaN肖特基势垒二极管的反向饱和电流的理论值和实验值
图 4 (a)不同温度下lnI和E1/2关系图; (b)电流斜率βSC
图 5 (a)在不同温度下ln(I/E)和E1/2的关系; (b)相应的斜率βFP
图 6 (a) GaN中可导位错模型示意图; (b) FP 发射电流输运过程示意图
图 7 c(T)和q/kT的关系
图 8 (a)在不同温度下的C-V曲线; (b)不同夹断电压所对应的电场关系
图 9 (a)不同温度下ln(I/E2)和E–1关系图; (b) FN隧穿有效势垒高度随温度的变化
图 10 (a) FN 隧穿电流输运示意图; (b) EL mapping图(V = –14 V)