以自差分模式操作砷化銦鎵/砷化銦鋁單光子雪崩二極體實現光子數解析之研究;Study of Photon Number Resolving Detection with InGaAs/InAlAs Single-Photon Avalanche Diodes under self-differencing mode

NCUIR > College of Electrical Engineering & Computer Science > Graduate Institute of Electrical Engineering > Electronic Thesis & Dissertation > Item 987654321/93626

Please use this identifier to cite or link to this item: http://ir.lib.ncu.edu.tw/handle/987654321/93626

Title:	以自差分模式操作砷化銦鎵/砷化銦鋁單光子雪崩二極體實現光子數解析之研究;Study of Photon Number Resolving Detection with InGaAs/InAlAs Single-Photon Avalanche Diodes under self-differencing mode
Authors:	甘偉宏;Kan, Wei-Hong
Contributors:	電機工程學系
Keywords:	單光子雪崩二極體;自差分電路;光子數解析;Single Photon Avalanche Diode;Self-differencing;Photon Number Resolving
Date:	2024-01-25
Issue Date:	2024-09-19 17:22:52 (UTC+8)
Publisher:	國立中央大學
Abstract:	具有光子數解析能力的單光子雪崩二極體能操作在近常溫，且保持較低水平的暗計數率，已成為量子領域中不可或缺的角色。以單光子光達為例，憑藉著對於單光子高靈敏度的特性，實現較遠目標的偵測，但這也導致單光子雪崩二極體易受環境光等非理想因素影響，增加數據處理的難度。如果能進一步將先進技術的光子數解析應用在SPAD上，就能將因環境光造成的微弱崩潰訊號從數據中去除，大幅提高數據的準確性。在上述的應用將其操作頻率提高，代表在相同時間下能偵測的資料量更多，有效增加數據的完整性，但在提升頻率的同時，也會提高一些非理想因素造成的崩潰訊號，增加後期數據處理的難度，因此要將元件操作在高頻且保持低水平的暗計數率，對於元件結構設計與外部電路是個很大的挑戰。在傳統的閘控模式，元件的崩潰訊號會被內部接面電容產生的電容訊號埋沒，導致須提高操作偏壓才能成功計數到，此時元件可能因過大的電流而飽和，無法解析出光子數。目前已有許多文獻提出成功從電容訊號中萃取出崩潰訊號的方式，如自差分電路和正弦波閘控模式搭配濾波器等，這些消除電容訊號的技術，成功將電容訊號消除讓元件操作在較低的偏壓下，避免元件因過大的雪崩電流達到飽和，以成功實現光子數解析。由於實驗室之前所使用的自差分電路操作頻率為9.9 MHz，需使用20公尺的延遲線進行延遲，導致訊號經過受到嚴重的衰減，無法將電容訊號完全消除，造成需施加較大的偏壓才能判讀到崩潰訊號，導致暗計數上升且不利於光子數解析量測。為了改善上述問題，本文重新設計適用於高頻的自差分電路，操作頻率從9.9 MHz提升至104.731 MHz，延遲線的長度縮短至3公尺，成功將電容訊號消除至更低的水平，最終成功判讀到更微弱的崩潰訊號，減輕元件受暗計數的影響，對於光子數解析量測有相當大的幫助。本文利用閘控模式搭配重新設計的自差分電路，利用延遲線較短的優勢，成功將電容訊號水平降至3 mV，雜訊抑制比與先前使用的自差分電路相比，從25 dB提升至37.6 dB。在200 K下，單光子偵測效率達到52 %，並在不同溫度下檢驗光子數解析的性能。為了進一步探討操作頻率對於元件性能的影響，本文分別比較元件在高頻、低頻和不同元件增益對於光子數解析性能的影響，分析元件在高頻與高增益時所會面臨的問題，並提出有效的解決辦法。 ;Single-photon avalanche diodes(SPADs) with photon-number-resolving (PNR) capability can operate at near-room temperature and maintain a lower level of dark count rate, making them indispensable in the field of quantum research. Taking single-photon lidar(SPL) as an example, these devices utilize their high sensitivity to single photon to enable the detection of distant targets. However, this sensitivity also makes SPADs susceptible to non-ideal factors such as ambient light, which increases the difficulty of data processing. If further equipping SPADs the capability of PNR with an advanced technique, it becomes possible to eliminate the weak avalanche signals caused by ambient light from the data, significantly improving data accuracy. In the above applications, increasing the operating frequency of SPADs means being able to detect a larger amount of data within the same time, effectively enhancing data integrity. However, while increasing the frequency, it also amplifies the avalanche signals caused by non-ideal factors, thereby increasing the difficulty of post-processing the data. Therefore, operating the device at high frequencies while maintaining a low level of dark count rates poses a significant challenge in terms of device structure design and external circuitry. In the traditional gated-mode, the avalanche signals of the device can be buried by the capacitance signal generated by the internal junction capacitance. This requires increasing the excess bias to successfully count the signals. However, at higher excess bias, the device may saturate due to excessive current, making it unable to resolve the photon number. Currently, there are many studies that have proposed successful methods to extract the avalanche signals from the capacitive signal. Techniques such as self-differencing(SD) circuits and sinusoidal gate-mode with filters have been used. These techniques eliminate the capacitance signal and allow the device to operate at lower excess bias, avoiding saturation caused by excessive avalanche current and enabling successful PNR. Due to the 9.9 MHz operating frequency of the SD circuit used in the laboratory, a 20 meters delay line was required for delay, resulting in severe signal attenuation. The capacitance signal could not be completely eliminated, necessitating a higher bias voltage for the detection of breakdown signals. This led to an increase in dark counts and posed challenges for PNR measurements. To address these issues, this study redesigned a SD circuit suitable for high frequency, increasing the operating frequency from 9.9 MHz to 104.731 MHz. The length of the delay line was shortened to 3 meters, successfully reducing the capacitance signal to a lower level. As a result, weaker breakdown signals could be detected, mitigating the impact of dark counts on the component and significantly aiding in PNR measurements. This study employed gated-mode with a redesigned SD circuit, taking advantage of the shorter delay line to successfully reduce the capacitance signal to 3 mV. The noise suppression ratio increased from 25 dB to 37.6 dB compared to the previously used SD circuit. At 200 K, the single-photon detection efficiency(SPDE) reached 52 %, and the performance of PNR was evaluated at different temperatures. To further explore the impact of operating frequency on the component′s performance, the study compared the effects at high and low frequencies, as well as the influence of different component gains on PNR. The analysis addressed challenges faced by the component at both high frequencies and high gains, proposing effective solutions.
Appears in Collections:	[Graduate Institute of Electrical Engineering] Electronic Thesis & Dissertation

Files in This Item:

File	Description	Size	Format
index.html		0Kb	HTML	52	View/Open

社群 sharing

Loading...