在電子元件微縮化的趨勢下，更小的單一元件尺寸能夠提高單位面積內的元件數量，單光子崩潰二極體SPAD已經能夠被使用在光達上做為光偵測器，透過台積電的0.18微米高壓製程，能夠將光偵測器的元件尺寸縮小至微米的尺度，但在光達的應用上，通常會將發光元件與光偵測元件獨立架設，因此在本論文中，嘗試將此製程所製造的SPAD與同樣製程下所製造的發光二極體做整合，達到發光元件與光偵測元件一體化的目的。
在SPAD的設計佈局中，透過參考文獻 cite{基於矽之單光子偵測器及其若干設計} 的最佳操作效率結構，嘗試透過增加主動接面的濃度，經過TCAD模擬內部衝擊游離發生區域分佈以及電場分佈，初步檢視優化的可能性，透過縮小空乏區寬度，增加衝擊游離的發生機率，以達到增加光偵測效率、降低時基誤差以及減少反應時間的目的，但量測結果顯示，在接面濃度增加的情況下，其增加的濃度並未相差太多，以至於整體光偵測效率量測曲線大同小異，然而衝擊游離發生機率的增加，反而造成暗計數上升，降低了信噪比，同時也減少偵測的精確度。
在LED的設計佈局中，因台積電的CMOS高壓製程之半導體材料-矽，為間接能隙之半導體材料，其能隙特性導致給予正偏時的非輻射躍遷數量遠大於輻射躍遷，發光效率較低，但透過累增效應以其穿隧效應的電致發光，在給予逆偏電壓的操作下，有效提高發光效率。透過將主動區接面的濃度加深以至於其發生穿隧效應的崩潰，與參考文獻 cite{以標準CMOS製程實現逆偏式矽發光二極體與元件設計} 中的LED結構比較，量測穿隧效應與累增效應所帶來的發光效率，再透過增加垂直接面的接觸面積，來嘗試增加發光效率的可能性。
再來將SPAD與LED嘗試兩種整合排列方式，一為將光偵測與發光結構並排排列，另一種為將光偵測結構包圍再發光結構中央，再量測兩種結構相互的影響。兩者的影響主要有熱能以及串擾的影響，串擾又分為電學串擾以及光學串擾，在量測後，可以觀察到在沒有任何光學隔絕的措施下，光學串擾嚴重的影響感光元件的偵測功能，導致SPAD無法截止，整體感光元件失效。
最後嘗試將發光元件與光偵測元件透過不同晶片，同向固定再PCB中央，經由聚光透鏡將光子聚光為平行光線後發射，再經由物體反射回聚光透鏡並聚焦再SPAD的主動區內，但整體效果不如預期，既使將兩元件從晶片中分開來，在約1mm左右的距離下仍然會有光學串擾現象的發生，足可看出光學串擾所帶來的嚴重性。

In the trend of miniaturization of electronic components, smaller size of single component can improve the number of components per unit area. Single photon avalanche diode (SPAD) has been used as a light detector on the LIDAR. Through the $0.18mu m$ high voltage process of the TSMC, the component size of the light detector can be reduced to micron scale. However, in the application of the LIDAR, the light emitting component and the light detecting component are usually set up independently. Therefore, this thesis try to integrate the SPAD manufactured by this process with the LED produced by the same process to achieve the purpose of all-in-one of the light emitting component and the light detecting component.

In the design and layout of SPAD, through the optimal operation efficiency structure in Reference cite{基於矽之單光子偵測器及其若干設計}, trying to increase the concentration of junction of active region, simulate the distribution of electric field and distribution area of impact ionization through TCAD, and preliminarily examine the possibility of optimization. By reducing the width of depletion region and increasing the occurrence probability of impact ionization to expect increase the photon detection efficiency, reduce the timing jitter and reduce the response time. However, the measurement results show that when the concentration of junction of active region increases, the increased concentration is not much different, so that the measurement curves of the overall photon detection efficiency are almost the same, but the occurrence probability of impact ionization increases,lead to the dark count increases, reduce the signal-to-noise ratio, and also reduce the accuracy of detection.

In the design and layout of LED, silicon, the semiconductor material of CMOS high-voltage process, is an indirect energy gap semiconductor material, its energy gap characteristics lead to the number of non-radiative transitions is much larger than the radiation transition when the positive bias is applied, and the luminous efficiency is low. However, the electroluminescence through the avalanche effect and tunnelling effect can effectively improve the luminous efficiency under the operation of the reverse bias applied. Increasing the concentration of junction of active region, make the breakdown occurred by mechanism of tunnelling effect, and then compared with the LED structure in Reference cite{以標準CMOS製程實現逆偏式矽發光二極體與元件設計}. The luminous efficiency caused by tunneling effect and avalanche effect is measured, and the possibility of increasing the luminous efficiency is tried by increasing the contact area of the junction in vertical direct.

Then, SPAD and LED are tried to be arranged in two ways. One is to arrange the light detection component and light emitting component in parallel, and the other is the light detection structure is in the center and surrounded by the light emitting structure, and then measure the mutual influence of the two structures. The main influence of the two are thermal energy and crosstalk, crosstalk is classified as electrical crosstalk and optical crosstalk. After measurement, it can be observed that in the condition of without any optical isolation, optical crosstalk seriously affects the detection function of the light detection component, SPAD cannot be quenched and result in light detection component malfunction.

Finally, the light emitting component and the light detection component are tried to separate into different chips and fix to the center of the PCB with the same direction. The photons are concentrated into parallel light by the convex lens and emitted, then reflected back by object to the convex lens and focused on the active region of SPAD. However, the overall effect is not as good as expected. Even if the two elements are separated from the chip, there will still be optical crosstalk at a distance of about $1mm$, which is sufficient to see the severity of optical crosstalk.

[1] I Brouk, Yael Nemirovsky, Stefan Lachowicz, EA Gluszak, Steven Hinckley, and Kamran Eshraghian. Characterization of crosstalk between cmos photodiodes. Solid-State Electronics, 46(1):53–59, 2002.
[2] Leigh Canham. Gaining light from silicon. Nature, 408(6811):411–412, 2000.
[3] CR Crowell and SM Sze. Temperature dependence of avalanche multiplication in semiconductors. Applied Physics Letters, 9(6):242–244, 1966.
[4] Gian-Franco Dalla Betta. Advances in photodiodes. BoD–Books on Demand, 2011.
[5] Henri Dautet, Pierre Deschamps, Bruno Dion, Andrew D MacGregor, Darleene Mac-Sween, Robert J McIntyre, Claude Trottier, and Paul P Webb. Photon counting techniques with silicon avalanche photodiodes. Applied optics, 32(21):3894–3900, 1993.
[6] Zan Dong, Wei Wang, Beiju Huang, Xu Zhang, Ning Guan, Jin Chen, Yun Gui, Haijun Liu, and Hongda Chen. Silicon-based led display array in standard cmos technology. In 7th IEEE International Conference on Group IV Photonics, pages 332–334. IEEE, 2010.
[7] Andrea Gallivanoni, Ivan Rech, and Massimo Ghioni. Progress in quenching circuits for single photon avalanche diodes. IEEE Transactions on nuclear science, 57(6):3815–3826, 2010.
[8] Massimo Ghioni, Angelo Gulinatti, Ivan Rech, Franco Zappa, and Sergio Cova.
Progress in silicon single-photon avalanche diodes. IEEE Journal of selected topics
in quantum electronics, 13(4):852–862, 2007.
[9] Vitali Lazarevich Ginzburg and Vadim Nikolaevich Tsytovich. Transition radiation and transition scattering-some questions regarding the theory. Moscow Izdatel Nauka, 1984.
[10] Alexander O Goushcha and Bernd Tabbert. On response time of semiconductor photodiodes. Optical Engineering, 56(9):097101, 2017.
[11] Yooli Kang, HX Lu, Y-H Lo, DS Bethune, and WP Risk. Dark count probability
and quantum efficiency of avalanche photodiodes for single-photon detection. Applied Physics Letters, 83(14):2955–2957, 2003.
[12] Kimberly E Kolb. Signal-to-noise ratio of geiger-mode avalanche photodiode singlephoton counting detectors. Optical Engineering, 53(8):081904, 2014.
[13] Wolfgang Lotz. Electron-impact ionization cross-sections and ionization rate coefficients for atoms and ions. The Astrophysical Journal Supplement Series, 14:207, 1967.
[14] H Melchior. Demodulation and photodetection techniques. Laser Handbook, 1:725–835, 1972.
[15] Yury Petrov. Optometrika. https://github.com/caiuspetronius/Optometrika/, 2022. Optometrika MATLAB library implements analytical and iterative ray tracing approximation to optical image formation using Snell＇s and Fresne.
[16] Dieter K Schroder. Semiconductor material and device characterization. John Wiley & Sons, 2015.
[17] NA Sobolev, AM Emel＇yanov, and KF Shtel＇makh. Avalanche breakdown-related electroluminescence in single crystal si: Er: O. Applied physics letters, 71(14):1930–1932, 1997.
[18] Simon M Sze and Kwok K Ng. Physics of semiconductor devices. John wiley & sons,2006.
[19] 陳冠廷. 基於矽之單光子偵測器及其若干設計. Master’s thesis, National Cheng
Kung University, 2020.
[20] 陳振元. 以標準cmos 製程實現逆偏式矽發光二極體與元件設計. Master’s thesis,
National Cheng Kung University, 2021.