標題: 利用濺鍍氮化鋁緩衝層在濕式蝕刻與乾式蝕刻圖形化藍寶石基板來改善氮化鎵品質
Improve GaN quality with sputter AlN buffer layer on wet-etched and dry-etched PSS
作者: 李振宏
吳耀銓
Li, Jhen-Hong
Wu, Yew-Chung
材料科學與工程學系所
關鍵字: 氮化鋁;緩衝層;圖型化藍寶石基板;AlN;buffer layer;PSS
公開日期: 2016
摘要: 發光二極體(light emitting diodes,LEDs)近年來逐漸取代傳統日熾燈泡主要由於LED的體積小、壽命長及效率高等優點,因此常見於汽車車頭燈、螢幕背光板、路燈等日常用品。而目前常用來做LED材料為氮化鎵(GaN),不過由於氮化鎵缺乏本身基板因此我們常利用藍寶石基板來磊晶GaN,不過由於兩者之間的晶格不匹配與熱膨脹系數差異,因此會造成許多缺陷使得內部量子效率(IQE)下降。因此我們常用的方法是讓藍寶石基板圖型化使得差排彎曲並增加表面粗糙度改變光線路徑,因而可以光取出率(LEE)與內部量子效率。 而在本實驗第一部分主要是利用濕式蝕刻藍寶石基板去製作凸型圖型化藍寶石基板與凹型圖型化藍寶石基板後,再去濺鍍氮化鋁緩衝層,磊晶氮化鎵並做分析,在實驗中發現凸型圖型化藍寶石基板的磊晶品質會比凹型圖型化藍寶石基板好,而發光功率凸型圖型化藍寶石基板更是明顯大於凹型圖型化藍寶石基板,這主是因為凸型的光取出率比凹型還要好。 在實驗第二部分則是利用乾式蝕刻圖型化藍寶石基板改變氮化鋁分佈的位置,分為氮化鋁在乾式蝕刻圖型化藍寶石基板(RPSS)、底部氮化鋁在乾式蝕刻圖型化藍寶石基板(BPSS)及頂部在乾式蝕刻圖型化藍寶石基板(TPSS),而在TPSS的部分因為GaN主要都在圓錐上造成無法磊平做分析。磊晶品質部分我們原本預期在BPSS比RPSS還要好,不過結果卻相反,而兩者的LEE幾乎相同。因此從發光功率得知RPSS比BPSS好也可以證實在磊晶品質部分確實是RPSS較好。
Light emitting diodes(LEDs) have been replaced incandescent-filament lamp due to their advantages of small size、long life and high efficiency. There are a lot of applications such as car headlight, screen backlight source and street light. The material of LEDs is GaN, but GaN is lack of own’s substrate. As a result, we usually use sapphire as substrate to epitaxy GaN. There are still many problems between sapphire and GaN such as lattice mismatch and thermal expansion coefficient difference, which will make defects and degrade internal quantum efficiency (IQE).So the solution we use is pattern sapphire substrate (PSS) which can make dislocation bending and change the optical path. It will improve IQE and light extraction efficiency(LEE). Our first experiment is using wet-etching to fabricate wet-etched convex-PSS (convex-PSS)and wet-etched concave-PSS(concave-PSS).Then, we sputter AlN on PSS to epitaxy GaN and analysis. In the experiment, we discover GaN quality of convex-PSS is better than concave-PSS and light output power(LOP) of convex-PSS is much better than concave-PSS. It is because the LEE of convex is better. Our second experiment is using dry-etching PSS and modifying AlN buffer layer. The GaN of Top AlN on PSS (TPSS) mainly growth on the cone and difficult to epitaxy smoothly which is difficult to fabricate LED chip. Our expectation of GaN quality is Bottom AlN on PSS(BPSS) is better than AlN on PSS(RPSS),but the truth is in contrast. Because of their almost same LEE, we can determine whose IQE is better by the LOP and we observe the LOP of RPSS is better than BPSS.
URI: http://etd.lib.nctu.edu.tw/cdrfb3/record/nctu/#GT070351532
http://hdl.handle.net/11536/139472
顯示於類別:畢業論文