紫外探測(cè)器的理論仿真設(shè)計(jì)及制備研究
發(fā)布時(shí)間：2018-07-26 18:54
【摘要】：我國(guó)的海洋衛(wèi)星紫外成像儀將使用Ⅲ-N半導(dǎo)體光電探測(cè)器進(jìn)行成像,為了制備高響應(yīng)率、高探測(cè)率的紫外探測(cè)器,本文結(jié)合InGaN半導(dǎo)體材料的光電學(xué)性質(zhì),采用理論和實(shí)驗(yàn)相結(jié)合方法,研究了背照式InGaN紫外探測(cè)器的光電學(xué)性能。此外,還對(duì)日盲波段AlGaN紫外探測(cè)器及SAM型APD器件研究中遇到的問題進(jìn)行了研究。主要研究?jī)?nèi)容如下:首先,介紹了探測(cè)器性能表征參數(shù)及仿真平臺(tái)Silvaco TCAD,對(duì)數(shù)值模擬中應(yīng)用到的半導(dǎo)體基本方程和物理模型進(jìn)行了總結(jié)。具體研究InGaN紫外探測(cè)器中的n-GaN、i-InGa N和p-Ga N厚度,極化效應(yīng),載流子SRH復(fù)合壽命等參數(shù)對(duì)器件光譜響應(yīng)的影響。理論計(jì)算發(fā)現(xiàn):當(dāng)少數(shù)載流子SRH復(fù)合壽命在0.01-0.1ns時(shí),仿真的響應(yīng)光譜與實(shí)驗(yàn)測(cè)試曲線吻合較好。其中n-GaN會(huì)影響探測(cè)器的短波抑制比,但對(duì)峰值響應(yīng)率影響較小。在保持n-GaN、p-GaN厚度一致情況下,隨著i-InGaN層厚度的增加,探測(cè)器響應(yīng)率會(huì)先逐漸增大,達(dá)到最大值后逐漸減小,當(dāng)i-InGaN層厚度為400nm時(shí)響應(yīng)率達(dá)到最大值。由于In摩爾組分較低,InGaN/GaN界面極化電荷密度較小,極化強(qiáng)度對(duì)探測(cè)器的響應(yīng)率無明顯影響。此外,還數(shù)值模擬了側(cè)表面鈍化層厚度,刻蝕損傷深度,p電極寬度及臺(tái)面寬度對(duì)InGaN/GaN異質(zhì)結(jié)探測(cè)器暗電流的影響。觀察到暗電流隨側(cè)表面鈍化層厚度減小而增大,隨刻蝕損傷深度增大而增大。探測(cè)器的暗電流會(huì)隨臺(tái)面寬度增大而增大,與探測(cè)器p電極寬度無明顯關(guān)系。當(dāng)側(cè)面鈍化層SiO2厚度為50nm,刻蝕損傷厚度為20nm時(shí),暗電流的仿真值與實(shí)驗(yàn)測(cè)試值吻合較好。同時(shí),采用標(biāo)準(zhǔn)Ⅲ-N臺(tái)面器件制備工藝制備了In_(0.03)Ga_(0.97)N/GaN異質(zhì)結(jié)探測(cè)器,光敏面半徑為30μm,器件的動(dòng)態(tài)零偏電壓電阻約為2.00×10~(12)Ω,優(yōu)質(zhì)因子R0A=5.66×10~7Ω?cm~2,在371nm處峰值響應(yīng)率R=0.215A/W,對(duì)應(yīng)的峰值探測(cè)率D*=2.34×10~(13)cm?Hz1/2?W-1。此外,在制備工藝中研究了n電極退火條件對(duì)p電極歐姆接觸的影響,還對(duì)實(shí)際探測(cè)器暗電流特性進(jìn)行了簡(jiǎn)單分析。當(dāng)樣品在氮?dú)鈿夥障峦嘶饡r(shí),在保證n電極形成歐姆接觸的情況下,在一定范圍內(nèi),退火時(shí)間越短、退火溫度越低對(duì)p電極歐姆接觸的影響越小。對(duì)于探測(cè)器的暗電流有如下結(jié)果:(1)臺(tái)面寬度一定時(shí),p電極寬度越小,器件反向暗電流越小;(2)p電極一定時(shí),臺(tái)面寬度越小,器件反向暗電流越小;(3)加厚電極采用電極引出線形式的反向暗電流較直接覆蓋臺(tái)面的方法小約3個(gè)數(shù)量級(jí)。最后,引入帶尾效應(yīng)光學(xué)模型模擬AlGaN紫外探測(cè)器的響應(yīng)光譜。當(dāng)n1層厚度為20nm,載流子濃度為1.0×10~(18)cm~(-3)時(shí),SAM型結(jié)構(gòu)APD器件實(shí)現(xiàn)了加速區(qū)與倍增區(qū)分離。另外對(duì)SAM型APD器件提前擊穿現(xiàn)象進(jìn)行分析,數(shù)值模擬了場(chǎng)板結(jié)構(gòu)、微臺(tái)面結(jié)構(gòu)及斜臺(tái)面結(jié)構(gòu)在空氣中的器件電場(chǎng)分布情況。在一定范圍內(nèi)隨著場(chǎng)板下鈍化層厚度增加和場(chǎng)板長(zhǎng)度加長(zhǎng),p電極邊緣處電場(chǎng)強(qiáng)度逐漸減小并趨于穩(wěn)定,可調(diào)節(jié)場(chǎng)板下鈍化層厚度及場(chǎng)板長(zhǎng)度使得p電極邊緣處和場(chǎng)板邊緣處電場(chǎng)強(qiáng)度大致相等,此時(shí)器件被提前擊穿的可能性最小。在-100V高壓反向偏置下,45°傾角時(shí)斜臺(tái)面區(qū)有較大范圍的低電場(chǎng)分布。
[Abstract]:In order to prepare the ultraviolet detector with high response rate and high detection rate, the marine satellite ultraviolet imager of our country will use the -N semiconductor photodetector. In this paper, the photoelectric properties of the backilluminated InGaN ultraviolet detector are studied in combination with the photoelectric properties of the InGaN semiconductor materials, and the theoretical and experimental methods are used to study the photoelectric properties of the backlighting type ultraviolet detector. The problems encountered in the study of AlGaN UV detectors and SAM APD devices in the daily blind are studied. The main contents are as follows: first, the parameters of the detector performance and the simulation platform Silvaco TCAD are introduced, and the basic semiconductor equations and physical models applied in the numerical simulation are summarized. The InGaN UV detection is studied. The influence of n-GaN, i-InGa N and p-Ga N thickness, polarization effect and carrier SRH composite life on the spectral response of the device. The theoretical calculation shows that the response spectrum of the simulation is in good agreement with the experimental test curve when the minority carrier SRH composite life is in 0.01-0.1ns. When the thickness of the i-InGaN layer increases, the response rate of the detector will gradually increase with the thickness of the i-InGaN layer increasing, and the response rate decreases gradually with the thickness of the i-InGaN layer increasing. When the thickness of the i-InGaN layer is 400nm, the response rate reaches the maximum. Because the In mole component is low, the polarization charge density of the InGaN/GaN interface is smaller and the polarization is small. The polarization of the InGaN/GaN interface is small and the polarization is polarized. The intensity has no obvious effect on the response rate of the detector. In addition, the influence of the thickness of the passivation layer on the side surface, the depth of the etching damage, the width of the p electrode and the width of the mesa on the dark current of the InGaN/GaN heterojunction detector is also numerically simulated. The dark current increases with the width of the mesa, and has no obvious relation to the width of the detector p electrode. When the thickness of the side passivation layer is 50nm and the etching damage thickness is 20nm, the simulation value of the dark current is in good agreement with the experimental test. At the same time, the In_ (0.03) Ga_ (0.97) N/GaN heterojunction detection is prepared by the preparation process of standard III -N mesa device. The radius of the photosensitive surface is 30 mu m, the dynamic zero bias voltage resistance of the device is about 2 x 10~ (12) Omega, the high quality factor R0A=5.66 x 10~7 Omega cm~2, the peak response rate R=0.215A/W at 371nm, the corresponding peak detection rate D*=2.34 * 10~ (13) cm? Hz1/2? Besides, the influence of the annealing condition on the ohm contact of the electrode is studied in the preparation process, and the effect of the electrode annealing condition on the electrode ohm contact is also studied in the preparation process. The dark current characteristic of the actual detector is simply analyzed. When the sample is annealed in the nitrogen atmosphere, the shorter the annealing time is, the lower the annealing temperature has on the ohm contact of the p electrode when the N electrode is formed in the case of ohm contact in the nitrogen atmosphere. The dark current of the detector is as follows: (1) when the width of the mesa is certain, P The smaller the width of the electrode, the smaller the reverse dark current of the device; (2) when the p electrode is certain, the smaller the width of the mesa, the smaller the reverse dark current of the device; (3) the reverse dark current in the form of the electrode leading out line is approximately 3 orders of magnitude smaller than the method of directly covering the table. Finally, the tail effect optical model is introduced to simulate the response light of the UV detector. When the thickness of the N1 layer is 20nm and the carrier concentration is 1 * 10~ (18) cm~ (-3), the SAM type APD device separates the acceleration zone from the multiplier area. In addition, the early breakdown phenomenon of the SAM type APD device is analyzed. The field plate structure, the micro platform structure and the device electric field distribution in the air are numerically simulated. With the increase of the thickness of the passivation layer under the field plate and the lengthening of the field plate length, the electric field strength at the edge of the p electrode gradually decreases and tends to be stable. The thickness of the passivation layer under the field plate and the length of the field plate can make the electric field intensity at the edge of the p electrode and the edge of the field plate approximately equal. At this time, the possibility of early breakdown of the device is the smallest. The reverse bias of the -100V high pressure is reversed. Under the 45 degree angle, there is a wide range of low electric field distribution on the inclined mesa.
【學(xué)位授予單位】：中國(guó)科學(xué)院大學(xué)(中國(guó)科學(xué)院上海技術(shù)物理研究所)
【學(xué)位級(jí)別】：碩士
【學(xué)位授予年份】：2017
【分類號(hào)】：TN23
本文編號(hào)：2146972
[Abstract]:In order to prepare the ultraviolet detector with high response rate and high detection rate, the marine satellite ultraviolet imager of our country will use the -N semiconductor photodetector. In this paper, the photoelectric properties of the backilluminated InGaN ultraviolet detector are studied in combination with the photoelectric properties of the InGaN semiconductor materials, and the theoretical and experimental methods are used to study the photoelectric properties of the backlighting type ultraviolet detector. The problems encountered in the study of AlGaN UV detectors and SAM APD devices in the daily blind are studied. The main contents are as follows: first, the parameters of the detector performance and the simulation platform Silvaco TCAD are introduced, and the basic semiconductor equations and physical models applied in the numerical simulation are summarized. The InGaN UV detection is studied. The influence of n-GaN, i-InGa N and p-Ga N thickness, polarization effect and carrier SRH composite life on the spectral response of the device. The theoretical calculation shows that the response spectrum of the simulation is in good agreement with the experimental test curve when the minority carrier SRH composite life is in 0.01-0.1ns. When the thickness of the i-InGaN layer increases, the response rate of the detector will gradually increase with the thickness of the i-InGaN layer increasing, and the response rate decreases gradually with the thickness of the i-InGaN layer increasing. When the thickness of the i-InGaN layer is 400nm, the response rate reaches the maximum. Because the In mole component is low, the polarization charge density of the InGaN/GaN interface is smaller and the polarization is small. The polarization of the InGaN/GaN interface is small and the polarization is polarized. The intensity has no obvious effect on the response rate of the detector. In addition, the influence of the thickness of the passivation layer on the side surface, the depth of the etching damage, the width of the p electrode and the width of the mesa on the dark current of the InGaN/GaN heterojunction detector is also numerically simulated. The dark current increases with the width of the mesa, and has no obvious relation to the width of the detector p electrode. When the thickness of the side passivation layer is 50nm and the etching damage thickness is 20nm, the simulation value of the dark current is in good agreement with the experimental test. At the same time, the In_ (0.03) Ga_ (0.97) N/GaN heterojunction detection is prepared by the preparation process of standard III -N mesa device. The radius of the photosensitive surface is 30 mu m, the dynamic zero bias voltage resistance of the device is about 2 x 10~ (12) Omega, the high quality factor R0A=5.66 x 10~7 Omega cm~2, the peak response rate R=0.215A/W at 371nm, the corresponding peak detection rate D*=2.34 * 10~ (13) cm? Hz1/2? Besides, the influence of the annealing condition on the ohm contact of the electrode is studied in the preparation process, and the effect of the electrode annealing condition on the electrode ohm contact is also studied in the preparation process. The dark current characteristic of the actual detector is simply analyzed. When the sample is annealed in the nitrogen atmosphere, the shorter the annealing time is, the lower the annealing temperature has on the ohm contact of the p electrode when the N electrode is formed in the case of ohm contact in the nitrogen atmosphere. The dark current of the detector is as follows: (1) when the width of the mesa is certain, P The smaller the width of the electrode, the smaller the reverse dark current of the device; (2) when the p electrode is certain, the smaller the width of the mesa, the smaller the reverse dark current of the device; (3) the reverse dark current in the form of the electrode leading out line is approximately 3 orders of magnitude smaller than the method of directly covering the table. Finally, the tail effect optical model is introduced to simulate the response light of the UV detector. When the thickness of the N1 layer is 20nm and the carrier concentration is 1 * 10~ (18) cm~ (-3), the SAM type APD device separates the acceleration zone from the multiplier area. In addition, the early breakdown phenomenon of the SAM type APD device is analyzed. The field plate structure, the micro platform structure and the device electric field distribution in the air are numerically simulated. With the increase of the thickness of the passivation layer under the field plate and the lengthening of the field plate length, the electric field strength at the edge of the p electrode gradually decreases and tends to be stable. The thickness of the passivation layer under the field plate and the length of the field plate can make the electric field intensity at the edge of the p electrode and the edge of the field plate approximately equal. At this time, the possibility of early breakdown of the device is the smallest. The reverse bias of the -100V high pressure is reversed. Under the 45 degree angle, there is a wide range of low electric field distribution on the inclined mesa.
【學(xué)位授予單位】：中國(guó)科學(xué)院大學(xué)(中國(guó)科學(xué)院上海技術(shù)物理研究所)
【學(xué)位級(jí)別】：碩士
【學(xué)位授予年份】：2017
【分類號(hào)】：TN23
【參考文獻(xiàn)】
相關(guān)期刊論文 前7條
1 尹達(dá)一;周青;黃小仙;張燕;李向陽;;海面溢油紫外推掃相機(jī)航空遙感監(jiān)測(cè)校飛結(jié)果分析[J];海洋科學(xué)進(jìn)展;2014年02期
2 張燕;儲(chǔ)開慧;邵秀梅;袁永剛;劉大福;陳新禹;李向陽;;GaN基512×1元紫外長(zhǎng)線列焦平面探測(cè)器組件[J];光學(xué)學(xué)報(bào);2009年12期
3 趙文伯;趙紅;葉嗣榮;黃烈云;唐遵烈;羅木昌;楊曉波;廖秀英;向勇軍;鄒澤亞;;Al_xGa_(1-x)N日盲紫外探測(cè)器及其焦平面陣列[J];半導(dǎo)體光電;2009年01期
4 張燕;龔海梅;白云;陳亮;許金通;湯英文;游達(dá);趙德剛;郭麗偉;李向陽;;空間用紫外探測(cè)及AlGaN探測(cè)器的研究進(jìn)展[J];激光與紅外;2006年11期
5 游達(dá);許金通;湯英文;何政;徐運(yùn)華;龔海梅;;p-GaN/Al_(0.35)Ga_(0.65)N/GaN應(yīng)變量子阱肖特基紫外探測(cè)器[J];半導(dǎo)體學(xué)報(bào);2006年10期
6 郝國(guó)強(qiáng),張永剛,劉天東,李愛珍;InGaAs PIN光電探測(cè)器的暗電流特性研究[J];半導(dǎo)體光電;2004年05期
7 邵慶輝,葉志鎮(zhèn),黃靖云;GaN基器件中的歐姆接觸[J];材料導(dǎo)報(bào);2003年03期
相關(guān)碩士學(xué)位論文 前2條
1 盧怡丹;InGaN紫外探測(cè)器性能研究[D];中國(guó)科學(xué)院研究生院（上海技術(shù)物理研究所）;2014年
2 李俊;HVPE生長(zhǎng)自支撐GaN技術(shù)研究[D];長(zhǎng)春理工大學(xué);2009年
，本文編號(hào)：2146972
本文鏈接：http://sikaile.net/kejilunwen/dianzigongchenglunwen/2146972.html
最近更新
教材專著
