標題: 互補式金屬氧化物半導體製程單光子累崩光偵測器特性與其應用
Characteristic and Application of Single Photon Avalanche Photodiode with Complementary Metal-Oxide-Semiconductor Process
作者: 巫朝陽
Wu, Jau-Yang
林聖廸
Lin, Sheng-Di
電子工程學系 電子研究所
關鍵字: 偵測器;累崩;單光子;detector;avalanche;single photon
公開日期: 2015
摘要: 單光子累崩偵測器具有極高的光靈敏度與時間解析度,利用其特性在生物螢光生命解析影像與測距影像的應用是近來熱門的研究。在本論文中利用TCAD模擬元件內部特性下,透過台積電不同製程設計單光子累崩偵測器元件,再透過自行架設的量測系統定義出元件的特性。單光子累崩偵測器有幾個重要的特性定義參數,分別為暗計數(Dark Count Rate)、光偵測率(Photon Detection Efficiency)、時間抖動(Jitter)、二次崩潰(After-pulsing),本論文首先介紹為了量測以上參數所自行架設的自動化量測系統,並說明如何準確地獲知元件的優越與差別。 在台積電標準0.18 um 互補式金屬氧化物半導體高頻製程中,利用其三個不同載子類型與濃度的佈值濃度井設計並比較了兩種崩潰電壓分別為10 V與20 V的元件。另外透過元件設計的技巧來降低元件暗計數,與嘗試著設計陣列中能有較高填充率(Fill Factor)的元件佈局。另外利用有較多不同的佈值濃度井的0.18 um 互補式金屬氧化物半導體高壓製程,設計了具有不同崩潰電壓的元件。並且,我們亦設計具有不同主動區深度的元件,使其能有不同響應特性;依據元件崩潰電壓與暗計數的關係可做穿隧效應在其中扮演的重要性的探討。 由於單光子累崩偵測器操作在極強的電場下,其主動區的電場需有均勻的分佈如此在實際應用中才能有一致的表現。 本論文中利用自行架設的二維光計數掃描量測,針對兩個分別具有高崩潰電壓與低崩潰電壓元件的電場分佈特性進行分析探討。本論文發現高崩潰電壓元件存在著與深度相關的電場分佈均勻度,而低崩潰電壓的元件保持相當均勻的電場分佈。 最後,本論文以低暗計數互補式金屬氧化物半導體單光子累崩偵測器與市售的砷化鎵銦單光子累崩偵測器作為輻射溫度計,並比較了分別以兩種偵測器量測輻射溫度之實驗與理論結果。根據實驗結果,發現單光子累崩偵測器作為輻射溫度計感測器比以往所用的電子耦合元件具有更高的靈敏性。將互補式金屬氧化物半導體累崩偵測器元件特性增強並配合電路設計,其所開發出的影像陣列期望能作為具有大溫度範圍與同時利用元件高時間解析度特性測得偵測物之距離的熱影像照相機。
The single photon avalanche photodiode (SPAD) exhibits ultra high photon sensitivity and timing resolution, hence it is currently being studied and used for the applications of biological fluorescence lifetime imaging microscopy and range-finder imaging. With this aim, in this thesis we use the Technology Computer Aided Design (TCAD) to simulate the characteristics of SPAD, design structures of SPAD with different processes in TSMC, and characterize them by our measurement systems. We characterize SPADs with several important parameters, including the dark count rate (DCR), photon detection efficiency (PDE), timing jitter, and after-pulsing. This thesis introduces the measurement setups that can automatically acquire above parameters for evaluating the performance of various devices. We design and compare two SPAD devices with breakdown voltage (VBre) of 10 V and 20 V by three different wells with different types of carriers and doping concentrations in TSMC 0.18 um complementary metal-oxide-semiconductor (CMOS) RF process. We design several structures to reduce the sources of DCR of SPAD device and modify the device layout for improving the fill factor of an array. Furthermore, we design different device structures with various VBre by the use of TSMC 0.18 um CMOS High Voltage process that has more choices of well with different doping concentrations. Among them, there are devices with different depths for the active region, resulting in different photon spectral response. The tunneling effect for devices with various VBre can be further analyzed in characterizing the DCR of these devices. While SPAD is operated based on the ultra high electric field induced avalanche, an uniform electric field distribution is demanded for a consistent response in practical applications. In this thesis, we use the setup of 2D photon count mapping measurement system for examining the uniformity of electric field distribution in two SPADs with respectively high and low VBre. The SPAD with higher VBre has distinct depth-dependent non-uniformity of electric field distribution, but the SPAD with lower VBre has better uniformity of electric field distribution at all depths. In the last part of this thesis, we explore both CMOS SPAD with low DCR and a commercial InGaAs SPAD as a radio thermometer and compare the measurement results to the theoretical calculations. According to the measurement results, the SPAD, as being a radio thermometer, has higher sensitivity than CCD. It is expected that the imaging array based on CMOS SPAD with enhanced performance and specially designed circuit will be a prospective thermal image camera which at the same time can capture the distance information of sensed object due to the high timing resolution of SPAD.
URI: http://140.113.39.130/cdrfb3/record/nctu/#GT079711815
http://hdl.handle.net/11536/125835
Appears in Collections:Thesis