氮化鋁鎵/氮化鎵高電子遷移率電晶體之 電流暫態特性研究

NCUIR > College of Electrical Engineering & Computer Science > Graduate Institute of Electrical Engineering > Electronic Thesis & Dissertation > Item 987654321/77744

Please use this identifier to cite or link to this item: http://ir.lib.ncu.edu.tw/handle/987654321/77744

Title:	氮化鋁鎵/氮化鎵高電子遷移率電晶體之電流暫態特性研究
Authors:	林晏平;Lin, Yen-Pin
Contributors:	電機工程學系
Keywords:	氮化鋁鎵;氮化鎵;高電子遷移率電晶體;電流暫態特性
Date:	2018-08-07
Issue Date:	2018-08-31 14:54:43 (UTC+8)
Publisher:	國立中央大學
Abstract:	AlGaN/GaN HEMTs已被廣泛用於RF功率放大器，特別是應用於基地台功率發射器，因為以GaN為基底的寬能隙材料有高電子遷移率與高崩潰電場特性。然而在GaN HEMTs的缺陷中產生的電流的散射(dispersion)是很重要的議題。此論文中，在AlGaN/GaN HEMTs模擬元件中加入不同的缺陷進行暫態模擬分析，並將AlGaN/GaN HEMTs在開啟狀態、不同靜態工作點與不同基底溫度下施加應力，進行電流暫態測量分析。模擬元件為AlGaN/GaN HEMT結構，若在模擬元件的SiO2和AlGaN界面或GaN中加入會捕捉電子的缺陷(acceptor-like trap)，則在持續開啟狀態下，會有電流增加或減少現象，但若加入會釋放電子的缺陷(donor-like trap)，則會有電流增加趨勢。論文中測量的商用元件AlGaN/GaN HEMTs，元件尺寸是閘極長度(gate length)與閘極寬度(Gate width)分別為0.25 μm與125 μm，並且成長在4吋SiC基板上，以SiN為鈍化層而且有源耦合場板(source-coupled field-plate)。開啟狀態下的電流暫態測量包含兩個部分。一個是在固定靜態工作點進行100 s的電流暫態測量，另一個為由不同工作點切換回ID = 100 mA/mm、VDS = 28 V時的靜態工作點進行100 s的電流暫態測量。為了分析陷阱活化能(EA)，所有的電流暫態測量皆會在25°C、40°C、50°C、60°C，4個不同溫度下進行。元件在固定靜態工作點進行電流暫態測量觀察，若此工作點的 ID = 427 mA/mm，會有明顯的通道載子被捕捉效應(trapping effect)，而若工作點的 ID = 33~238 mA/mm，有通道載子被捕捉或釋放效應(trapping & detrapping effect)。但在25°C時，若工作點的ID = 33 ~ 345 mA/mm，會跟元件在高溫下(40°C、50°C、60°C)時有著不同的暫態特性。另外元件由不同偏壓狀態切換回ID = 100 mA/mm、VDS = 28 V狀態進行100 s的電流暫態測量。若切換前狀態的ID大於100 mA/mm，則100 s內有明顯的通道載子釋放效應；若切換前狀態的ID小於100 mA/mm，則100 s內有明顯的通道載子被捕捉效應。元件由ID = 100 mA/mm、VDS = 28 V 切換至不同偏壓狀態進行100 s的電流暫態測量。若切換後狀態的ID = 343 ~ 427 mA/mm或ID = 33 ~ 87 mA/mm，則100 s內有明顯的通道載子被捕捉效應；若切換後狀態的ID = 145 ~ 238 mA/mm，100 s內有明顯的通道載子被捕捉與釋放效應。論文中也在模擬元件中加入適當的缺陷去擬合量測的電流暫態結果，可以有相當高的吻合度。因此，藉由模擬可推測元件中缺陷的位置、能階與濃度。論文中所有的暫態電流分析，萃取出的陷阱活化能有些是負值，有些值偏小，由此可推論論文中通道載子被捕捉或釋放效應，與溫度不太相關，可能是穿隧行為(tunneling behaviors)所造成的。所有不同靜態工作點、不同元件基底溫度的持續開啟狀態下的電流暫態測量都顯示複雜的通道載子被捕捉或釋放效應。此研究觀察電流暫態結果跟GaN HEMT用在功率放大器的應用有密切相關，因此進一步的深入研究是必要的。 ;AlGaN/GaN HEMTs have been widely used in RF power amplifiers, especially in base station power transmitters, because GaN-based wide energy gap materials have high electron mobility and high breakdown electric field characteristics. However, the dispersion of current generated in the defects of GaN HEMTs is an important issue.In this paper, different defects were added to the AlGaN/GaN HEMTs analog device for transient simulation analysis, and the AlGaN/GaN HEMTs were subjected to stress in the on state, different static operating points and different substrate temperatures for current transient measurement analysis. The analog device is an AlGaN/GaN HEMT structure. If acceptor-like traps are added to the SiO2 and AlGaN interface or GaN of the analog device, current may increase or decrease in a continuously on state, but if add donor-like traps, there will be a tendency for the current to increase. The commercial device AlGaN/GaN HEMTs measured in the paper have a gate length and a gate width of 0.25 µm and 125 µm, respectively, and grow on a 4 inch SiC substrate, passivated by SiN layer and have source-coupled field-plate. The current transient measurement in the on state consists of two parts. One is a current transient measurement of 100 s at a fixed quiescent operating point, and the other is a current transient of 100 s from a quiescent operating point when switching from different operating points to ID = 100 mA/mm and VDS = 28 V. To analyze trap activation energy(Ea), all current transient measurements were performed at 25°C, 40°C, 50°C and 60°C. The device is subjected to current transient measurement at a fixed quiescent operating point. If the ID of the operating point is 427 mA/mm, there will be a significant trapping effect, and if the working point ID = 33~238 mA/mm, there will be trapping and detrapping effect. However, at 25°C, if the ID of the working point = 33 ~ 345 mA/mm, it will have different transient characteristics with the device at higher temperature(40°C, 50°C, 60°C). In addition, the devices are switched from different bias states to ID = 100 mA/mm and VDS = 28 V for 100 s current transient measurement. If the ID of the state before switching is greater than 100 mA/mm, there is a significant detrapping effect within 100 s, but if the ID of the state before switching is less than 100 mA/mm, there is a significant trapping effect within 100 s. The device is switched from ID = 100 mA/mm and VDS = 28 V to different bias states for 100 s current transient measurement. If the ID of the switched state is 343 ~ 427 mA/mm or ID = 33 ~ 87 mA/mm, there will be obvious trapping effect within 100 s, if the ID of the switch state is 145 ~ 238 mA/mm within 100 s, there are obvious trapping and detrapping effects. In this paper, the appropriate defects are also added to the analog device to fit the measured current transient results, which can have a fairly high degree of coincidence. Therefore, the position, energy level and concentration of defects in the device can be estimated by simulation. In all the transient current analysis in this paper, the trap activation energy extracted is somewhat negative, and some values are too small. It can be inferred that the trapping and detrapping effects in this paper, which is not related to temperature, but may be tunneling. Current transient measurements at all static operating points and continuously open states of different devices substrate temperatures show complex trapping and detrapping effects. This study observes that current transient results are closely related to the application of GaN HEMTs in power amplifiers, so further in-depth studies are necessary.
Appears in Collections:	[Graduate Institute of Electrical Engineering] Electronic Thesis & Dissertation

Files in This Item:

File	Description	Size	Format
index.html		0Kb	HTML	242	View/Open

社群 sharing

Loading...