臺灣博碩士論文加值系統

本論文主要的目的為針對國研院台灣半導體研究中心提供學術界使用的TSMC 0.35 μm mixed-signal 2P4M (D35)以及TSMC 0.18 μm Mixed Signal 1P6M (T18) CMOS-MEMS標準製程中的導電材料進行特性量測，以提供熱型感測元件之設計參考。量測的項目包含電阻溫度係數(TCR)、熱導率、席貝克係數和本質雜訊等，待測的導電材料層為D35標準製程中的Poly1、Poly2、METAL1、METAL2以及T18標準製程中的Poly1、METAL1與METAL6。本研究利用台灣半導體研究中心所提供之D35與T18 CMOS-MEMS教育性晶片進行量測元件之設計與製作，熱導率與席貝克係數的量測微元件具有隔熱良好之懸浮結構，利用懸浮結構上加熱器形成待測材料的溫度梯度，透過施加給加熱器的功率與量測元件上溫度感測器所量測的材料兩端溫度差，即可計算出其熱導率與席貝克係數。材料TCR則是將元件放入加熱的恆溫水槽，並將數位電表連接晶片上的待測薄膜進行量測。最後，本論文設計一熱電式紅外線感測元件，並以ANSYS有限元素模擬軟體針對量測結果進行熱穩態模擬，藉以評估此CMOS-MEMS標準製程應用於熱型感測元件之可行性。
本論文成功製作出D35及T18導電材料特性量測晶片，量測結果顯示D35 Poly1、D35 Poly2及T18 Poly1之片電阻分別為7.75 Ω/sq、44.3 Ω/sq、7.76 Ω/sq，TCR則分別為0.083 %/°C、0.068 %/°C、0.3 %/°C，可以發現雖然兩種製程的Poly1多晶矽上方皆成長了一層金屬矽化物，導致片電阻值較低，但T18的Poly1具有較大的TCR，因此可推斷兩者的金屬矽化物成分並不相同。材料的熱導率以及席貝克係數也被成功量測，D35 Poly1、D35 Poly2、T18 Poly1的熱導率量測結果分別為40 Wm-1K-1、40 Wm-1K-1、28.9 Wm-1K-1，而席貝克係數的量測元件分別由D35 Poly1 & METAL1、D35 Poly2 & METAL1、T18 Poly1 & METAL1組成三個不同元件中的待測材料熱電偶對，席貝克係數的量測結果則分別為7.73 μV/K、112.6 μV/K、6.18 μV/K。最後，設計一熱電式紅外線感測元件，並以ANSYS模擬軟體針對量測結果進行熱穩態模擬，透過模擬結果推算紅外線感測元件的雜訊等效溫差以評估感測性能，而由D35 Poly1 & Poly2所組成熱電對的熱電式紅外線感測元件模型具有最低的雜訊等效溫差(NETD) 156.86 mK。
若根據上述D35與T18製程材料特性的量測與模擬結果來選擇熱型感測器的材料，T18製程的Poly1因為具有較低的熱導率以及較高的TCR，所以做為熱阻式感測器的感測電阻材料，能有最佳的感測性能，而D35製程的Poly1與Poly2因為具有較低的熱導率，且兩者的席貝克係數相差較大，所以較適合做為熱電式感測器的熱電偶材料。

The purpose of this study is to measure the characteristics of the electrical conductive materials used in TSMC 0.35 μm mixed-signal 2P4M (D35) and TSMC 0.18 μm Mixed Signal 1P6M (T18) CMOS-MEMS standard processes provided by Taiwan Semiconductor Research Institute (TSRI) for the development of thermal sensors. The measured characteristics of the conductor films include temperature coefficient of resistance (TCR), thermal conductivity, Seebeck coefficient and intrinsic noise. The electrical conductive layers to be measured in D35 process are Poly1, Poly2, METAL1, METAL2, and those in T18 process are Poly1, METAL1, and METAL6. In this study, D35 and T18 CMOS-MEMS educational chips provided by TSRI were used to design and fabricate the measurement devices. The micro-devices for measurement of thermal conductivity and Seebeck coefficient both have a suspended structure with good heat insulation, and the suspended structure was heated by a heater inside it to create the temperature gradient of a conductor material to be measured. The thermal conductivity and Seebeck coefficient of the material can be calculated by the input heating power and the temperatures at two ends of the conductor measured by built-in mico temperature sensors. The TCR of the materials were measured by using a temperature controlled water bath tank. Eventually, a thermopile infrared sensor modeled by the measured parameters was designed and characterized by performing ANSYS FEA thermal steady state simulation.
In this study, the chips for measuring the characteristics of electrical conductive materials were successfully produced. The measurement results show that the sheet resistances of D35 Poly1, D35 Poly2 and T18 Poly1 are 7.75 Ω/sq, 44.3 Ω/sq, 7.76 Ω/sq, respectively, and the TCR are 0.083 %/°C, 0.068 %/°C, 0.3 %/°C, respectively. The low sheet resistances of Poly1 layers of D35 and T18imply that the Poly1 films were covered by a silicide layer respectively. However, the TCR magnitudes of these Poly1 films are quite different. It indicates that the composition of the two metal silicide layers are different. The thermal conductivity and Seebeck coefficient of the materials have also been measured. The thermal conductivity measurement results of D35 Poly1, D35 Poly2, and T18 Poly1 are 40 Wm-1K-1, 40 Wm-1K-1, 28.9 Wm-1K-1, respectively. The Seebeck coefficient measurement results of D35 Poly1 & METAL1, D35 Poly2 & METAL1, T18 Poly1 & METAL1 thermocouple pairs are 7.73 μV/K, 112.6 μV/K, 6.18 μV/K, respectively. In this work, the performance of a thermopile infrared sensor modeled with the measurement results was estimated by adopting ANSYS thermal steady state simulation. Through the simulation, the noise equivalent temperature difference (NETD) of the thermopile infrared sensor model can be calculated to evaluate the sensing performance of it. The thermopile infrared sensor model with the thermocouple pair composed of D35 Poly1 & Poly2 has the lowest noise equivalent temperature difference (NETD) of 156.86 mK.
According to the measurement and simulation results of the materials in D35 and T18 processes, it can be inferred that T18 Poly1 is most suitable as the sensing resistor of thermal type altimeter, because it has lower thermal conductivity and higher TCR. Moreover, D35 Poly1 & Poly2 are most suitable as the thermopile materials of thermoelectric sensors, because they have lower thermal conductivity and a large difference in the value of Seebeck coefficient.

中文摘要 I
Abstract IIII
致謝 V
目錄 VI
圖目錄 VIII
表目錄 XI
第一章 緒論 1
1.1 研究動機 1
1.2 文獻回顧 4
1.3 論文架構 12
第二章 基礎理論介紹 13
2.1 熱導 13
2.1.1 固體熱導 13
2.1.2 氣體熱導 16
2.1.3 輻射熱導 17
2.1.4 總熱導 18
2.2 席貝克效應(Seebeck effect) 18
2.3 電阻溫度係數(Temperature coefficient of the resistance，TCR) 19
2.4 雜訊分析原理 20
2.4.1 熱雜訊(Thermal noise) 20
2.4.2 閃爍雜訊(Flicker noise) 22
2.4.3 溫度擾動雜訊(Temperature fluctuation noise) 22
2.4.4 散射雜訊 23
2.4.5 總雜訊 23
2.5 紅外線感測器性能參數 24
2.5.1 響應度(Responsivity) 24
2.5.2 雜訊等效功率(Noise equivalent power) 25
2.5.3 偵測度(D) 25
2.5.4 歸一化偵測度(

[1]Arvin Emadi, “Infrared Thermopile Detector Array for the integrated Microspectrometer,” IEEE Sensors 2007 Conference, pp. 435-438, Atlanta, Georgia, USA, 28-31 Oct, 2007.
[2]吳冠諭，3D 立體結構派藍尼真空計之設計、製作與量測，國立高雄應用科技大學光電與通訊工程研究所碩士論文，2013。
[3]Donald Neamen, “Semiconductor Physics and Devices: Basic Principles. 4th ed.,” New York, NY: McGraw-Hill, 2012.
[4]Martin von Arx, “Test structures to measure the Seebeck coefficient of CMOS IC polysilicon,” IEEE Transactions on Semiconductor Manufacturing, vol. 10, no. 2, pp. 201-208, 1997.
[5]Martin von Arx, “Process-dependent thin-film thermal conductivities for thermal CMOS MEMS,” Journal of Microelectromechanical Systems, vol. 9, no. 1, pp. 136-145, 2000.
[6]Huchuan Zhou, “Development of a thermopile infrared sensor using stacked double polycrystalline silicon layers based on the CMOS process,” Journal of Micromechanics and Microengineering, 2013.
[7]Moinuddin Ahmed, “The flicker noise in amorphous silicon based temperature sensors in flexible substrates,” 2011 21st International Conference on Noise and Fluctuations, Toronto, ON, pp. 154-157, 2011.
[8]劉靜，微奈米尺度熱傳學，五南文化，台北市，2004。
[9]Seung-Min Lee, David Cahill, “Heat transport in thin dielectric films,” Journal of Applied Physics, 1997.
[10]郭竹安，熱電堆陣列元件之設計與分析，國立高雄應用科技大學光電與通訊工程研究所碩士論文，2017。
[11]Ting-Wei Shen, Kai-Chieh Chang, Chih-Ming Sun, and Weileun Fang, “Performance enhance of CMOS-MEMS thermoelectric infrared sensor by using sensing material and structure design,” Journal of Micromechanics and Microengineering, 2018.
[12]陳志軍，高靈敏度 CMOS-MEMS 派藍尼真空計 之設計、製作與量測，國立高雄應用科技大學光電與通訊工程研究所碩士論文，2015。
[13]John Gilbert Kaufman, “Fire Resistance of Aluminum and Aluminum Alloys and Measuring the Effects of Fire Exposure on the Properties of Aluminum Alloys,” ASM International, pp. 2 – 9, 2016.
[14]陳君豪，CMOS 製程相容熱阻型紅外線微感測器材料低頻雜訊分析，碩士論文，國立高雄應用科技大學光電與通訊工程研究所，2015。
[15]葉盈志，矽化鎳材料應用於熱阻型元件之研究，碩士論文，國立高雄應用科技大學光電與通訊工程研究所，2017。
[16]王信智，CMOS製程相容之量測熱導率微元件模擬、設計與製作，碩士論文，國立高雄應用科技大學光電與通訊工程研究所，2012。
[17]陳威竣，CMOS-MEMS 熱電堆紅外線微感測器之優化，碩士論文，國立高雄應用科技大學光電與通訊工程研究所，2014。
[18]游迪鈞，CMOS 矽化鎢材料席貝克係數之量測研究，碩士論文，國立高雄應用科技大學光電與通訊工程研究所，2014。
[19]Robert Johanson, “1/f noise in hydrogenated amorphous silicon-germanium alloys,” IEE Proceedings - Circuits, Devices and Systems, vol. 150, no. 4, 2003.
[20]Sheng-Lyang Jang, “A model of 1f noise in polysilicon resistors,” Solid-State Electronics, vol 33, issue 9, pp. 1155-1162, 1990.
[21]Friedemann Volklein, “A Microstructure For Measurement Of Thermal Conductivity Of Polysilicon Thin Films,” Journal of Microelectromechanical Systems, vol. 1, no. 4, pp. 193-196, 1992.
[22]Huchuan Zhou, Piotr Kropelnicki, and Chengkuo Lee, “CMOS Compatible Midinfrared Wavelength-Selective Thermopile for High Temperature Applications,” Journal of Microelectromechanical Systems, vol. 24, no. 1, pp. 144-154, 2015.
[23]鮑柏雯，CMOS 奈米元件閃爍雜訊與 RTS 之特性分析，碩士論文，國立交通大學理學院(應用科技)學程，2010。
[24]Ming-Chun Hsieh, “Design and fabrication of a novel crystal SiGeC far infrared sensor with wavelength 8-14 micrometer,” IEEE Sensors Journal, vol. 2, no. 4, pp. 360-365, 2002.
[25]Guanxiong Liu, Sergey Rumyantsev, Alexander Balandin, and Michael Shur, “Surface and volume 1/f noise in multi-layer graphene,” 2013 22nd International Conference on Noise and Fluctuations (ICNF), Montpellier, pp. 1-4, 2013.
[26]Mohamed Jamal Deen, Sergey Rumyantsev, and Orchard-Webb, “Low frequency noise in heavily doped polysilicon thin film resistors,” Journal of Vacuum Science & Technology B: Microelectronics and Nanometer Structures, 1998.
[27]Ralf Brederlow, Werner Weber, Claus Dahl, Doris Schmitt-Landsiedel, and Roland Thewes, “Low-frequency noise of integrated polysilicon resistors,” IEEE Transactions on Electron Devices, vol. 48, no. 6, pp. 1180-1187, 2001.
[28]Farid Taherkhani and Hamed Rezania, “Temperature and size dependency of thermal conductivity of aluminum nanocluster,” Journal of Nano Particle Research, 2012.
[29]Ziyang Wang, Vladimir Leonov, Paolo Fiorini, and Chris Van Hoof, “Realization of a wearable miniaturized thermoelectric generator for human body applications,” Sensors and Actuators A: Physical, Volume 156, Issue 1, 2009.
[30]Ali Shakouri, “Recent Developments in Semiconductor Thermoelectric Physics and Materials,” Annual Review of Materials Research 41:1, pp. 399-431, 2011.
[31]René Lenggenhager, Henry Baltes, Jon Peer, and Martin Forster, “Thermoelectric infrared sensors by CMOS technology. Electron Device Letters,” IEEE, 13, pp. 454-456, 1992.