隨著科技的發展及進步，物聯網(Internet of things, IoT)的應用隨之廣泛，並在未來將和人們的日常生活變的密不可分，據報告指出，2020年全球IoT的結點裝置(node)將大幅增加至280億個，顯現出IoT在商業市場上的重要性以及未來的發展性。其中記憶元件的革新將主導著物聯網的發展，在眾多新世代記憶元件當中，電阻式記憶體(RRAM)因具有低功耗與高性能的特點，非常有潛力成為新世代的主流記憶體。而目前RRAM面臨最大的問題，就是在物理機制方面沒有統一且明確的理論，尤其是在電阻切換的動態過程最難以釐清。因此釐清RRAM電阻切換過程的物理機制，將是電阻式記憶體能否商品化之關鍵因素。
RRAM電阻切換行為包含寫入(Set)，及抹除(Reset)。在Set過程中，Set電壓的分佈是元件量產的瓶頸之一，而分佈的原因與HRS電阻值的分佈直接相關，本論文透過介面效應分析HRS電阻分佈的原因，是由於氧離子反應不完全而造成切換層氧化程度不一，可透過電漿處理製作儲氧槽以改善HRS電阻值的分佈。在Reset部分，本論文首次發現Reset行為與元件串聯電阻大小有很強的相關性，當元件LRS電阻與串聯電阻大小相近時，將發生電壓正回饋效應，使Reset過程在極短的時間內結束。此外本論文也首次發現Reset的程度與消耗的能量，並沒有直接相關，而是與電場的有效控制範圍有關，也透過電場模擬釐清物理機制，並提出較有效率且節能的Reset方式。
在元件的材料方面，本論文提出利用銦錫氧化物(ITO)作為元件的電極，並具有優良的切換特性、可靠度、及自我限流的特性。此外也首次發現元件中氧離子的濃度梯度是電阻切換的驅動力之一，並提出機制模型解釋。

With the continuous advancement of technology, applications in the Internet of things (IoT) will become widespread, and even integrated to many people’s daily lives. According to predictions, by 2020, global IoT nodes will expand to 28 billion, indicating the importance of IoT in the commercial markets and its potential future development. This development will be dominated by the evolution of memory devices, and among the next-generation memory devices, resistive random access memory (RRAM) has the most potential to become the mainstream next-generation memory due to its advantages of low power-consumption and high performance. However, a major obstacle is the non-uniform physical mechanism in its resistance switching process. Therefore, the key factor in putting RRAM into mass production is to clarify the switching physical mechanism.
The resistive switching process includes the set and reset processes. In the set process, the distribution of the set voltage is one of the most important issue which hinders the mass production of RRAM, and the reason for this voltage distribution is considered to be highly related to the resistance distribution of HRS. This is the first time analyzing the phenomenon using the interface effect, and we determine that the reason is the different degrees of oxidation of the switching layer, which is caused by an incomplete reaction of oxygen ions. In this article, an additional oxygen reservoir is also built by plasma treatment to inhibit the distribution situation. During the reset process, when the series resistance value is close to the LRS resistance value, voltage positive feedback effect occurs in the process, causing the reset process to end in a very short period of time. In addition, this is the first study which finds that the degree of reset is dominated by the effective electrical field region rather than the energy consumed, and a supportive physical model is established by an electrical field simulation. Further, a high efficiency, energy-saving operation method is also proposed in this research.
In terms of the device material, ITO is proposed to act as the device electrode, and exhibits outstanding performance, reliability, and a self-compliance characteristic. Moreover, the concentration gradient of oxygen ions is considered to be one of the driving forces of the resistance switching process, and a model is also proposed for explanation of the mechanism.

Chinese Abstract………………………..……….……………….……………..…..…iii
English Abstract.…………………..………………………...…………………...…...iv
Contents…………………….…………………………………………………..….….vi
Figure Captions………………………………………………………………………..ix
Table Captions……………………………………………………………….….....…xv

Chapter 1 Introduction……………………...…………………………………..…....1
1-1 Introduction of memory.………………..…….………………………...….……..1
1-2 Research Motivation…..……………………………………………..……….…..2

Chapter 2 Literature Review..………………………………...........………..…........3
2-1 Electrical Memory………….………..……………………………………….…..3
2-2 Advanced Memory...……………………..…………………………..…………..5
2-3 Carrier Conduction Mechanisms in RRAM…...………………….…...….……..10
2-4 Commercial Application and Global Development of RR AM……....….….…..16

Chapter 3 Set Voltage Distribution Stabilized by Constructing an Oxygen Reservoir in Resistive Random Access Memory.……………………………….…21
3-1 Motivation…………………………….…………………………………….…..21
3-2 Experimental Method…………………………………………………...…...….21
3-3 Results and Discussion…………………………………………..……...…...….22
3-4 Summary…………………………………………..…………………....…...….29


Chapter 4 Confirmation of Filament Dissolution Behavior by Analyzing Electrical Field Effect during Reset Process in Oxide-based RRAM…………....30
4-1 Motivation…………………………….…………………………………….…..30
4-2 Experimental Method…………………………………………………...…...….30
4-3 Results and Discussion…………………………………………..……...…...….31
4-4 Summary…………………………………………..…………………....…...….40

Chapter 5 Prediction of Reset Behavior Based on Voltage positive-Feedback Effect during the Switching Process in Oxide-Based RRAM………….……...….41
5-1 Motivation…………………………….…………………………………….…..41
5-2 Experimental Method…………………………………………………...…...….41
5-3 Results and Discussion…………………………………………..……...…...….42
5-4 Summary…………………………………………..…………………....…...….51

Chapter 6 Ultra-Low Power Resistance Random Access Memory Device and Oxygen Accumulation Mechanism in an Indium-Tin-Oxide Electrode…………52
6-1 Motivation…………………………….…………………………………….…..52
6-2 Experimental Method…………………………………………………...…...….53
6-3 Results and Discussion…………………………………………..……...…...….53
6-4 Summary…………………………………………..…………………....…...….66

Chapter 7 An Adjustable Built-in Resistor on Oxygen Vacancy-Rich electrode-capped Resistance Random Access Memory……………………………………...67
7-1 Motivation…………………………….…………………………………….…..67
7-2 Experimental Method…………………………………………………...…...….67
7-3 Results and Discussion…………………………………………..……...…...….68
7-4 Summary…………………………………………..…………………....…...….75

Chapter 8 Engineering Interface-type Resistance Switching Based on Forming Current Compliance in ITO/Ga2O3:ITO/TiN Resistance Random Access Memory: Conduction Mechanisms, Temperature Effects, and Electrode Influence……………...…………………………………………..………………….76
8-1 Motivation…………………………….…………………………………….…..76
8-2 Experimental Method…………………………………………………...…...….77
8-3 Results and Discussion…………………………………………..……...…...….79
8-4 Summary…………………………………………..…………………....…...….92

Chapter 9 Conclusion…………………………………………..…….…………….93

References………………………………………………………..……....………….96

Appendix
Publication List……………………………………………...……………….……..99

[1] "Human Memory. Herman Ebbinghaus". Lecture notes in Psychology published at Dr. Bruce Abbott's homepage at IPFW
[2] T. C. Chang et al.,” Mater. Today, vol. 19, issue. 5, pp. 254-264, Jun. 2016.
[3] H. Akinaga, H. Shima, “Resistive Random Access Memory (ReRAM) Based on Metal Oxides”, Proceeding of the IEEE, vol.98, issue.12, 2010.
[4] H. S. Philip Wong et al., “Metal-Oxide RRAM”, Proceeding of the IEEE, vol.100, no.6, 2012.
[5] Y. Wang et al.,” Nanotechnology, vol. 21, no. 4, p. 045202, Jan. 2010.
[6] Ming Liu et al., “Multilevel resistive switching with ionic and metallic filaments,” Appl. Phys. Lett. Vol.94, issue, 23, page.233106, Jun. 2009.
[7] H. Sun et al., “Overcoming the dilemma between RESET current and data retention of RRAM by lateral dissolution of conducting filament,” IEEE Electron Device Lett., vol. 34, no. 7, pp. 873–875, Jul. 2013.
[8] T. M. Tsai et al., “Bipolar resistive RAM characteristics induced by nickel incorporated into silicon oxide dielectrics for IC applications,” IEEE Electron Device Lett., vol. 33, no. 12, pp. 1696–1698, Dec. 2012.
[9] K. C. Chang et al., “Physical and chemical mechanisms in oxide-based resistance random access memory,” Nanosc. Res. Lett., vol. 10, pp. 120-1–120-27, Mar. 2015.
[10] T. J. Chu et al., “Charge Quantity Influence on Resistance Switching Characteristic During Forming Process,” IEEE Electron Device Lett, vol. 34, no. 4, pp. 502-504, Apr. 2013.
[11] I. Hwang et al.,” The role of contact resistance in GeTe and Ge2Sb2Te5 nanowire phase change memory reset switching current,” App. Phys. Lett., vol. 106, issue. 19, pp. 193106, May. 2015.
[12] A. Gyanathan et al.,” Two-bit multi-level phase change random access memory with a triple phase change material stack structure,” J App Phys., Vol. 112, issue. 10, pp. 104504, Oct. 2012.
[13] D. Loke, et al., “Breaking the Speed Limits of Phase-Change Memory,” Science, vol. 336, issue, 6088, pp.1566, Jun. 2012.
[14] Gary A. Prinz, “Magnetoelectronics,” Science, vol. 282, issue. 5394, pp. 1660, Nov. 1998.
[15] R. Zhao, L.P. Shi et al., Non-Volatile Memory Technology Symposium, 2007. NVMTS 07, 36-39 (2007)
[16] D. N. Nguyen and L. Z. Scheick, Radiation Effects Data Workshop, 2001 IEEE 57-61 (2001)
[17] T. W. Hickmott, “Low-Frequency Negative Resistance in Thin Anodic Oxide Films,” J. Appl. Phys. vol. 33, issue. 9, pp. 2669, Jun. 2004.
[18] A. Beck, J. G. Bednorz et al., “Reproducible switching effect in thin oxide films for memory applications,” Appl. Phys. Lett. vol. 77, issue. 1, pp. 139, May. 2000.
[19] A. Sawa, “Resistive switching in transition metal oxides”, Mater. Today, vol. 11, issue. 6, pp. 28–36. Jun, 2008.
[20] S. M. Sze, Semiconductor devices physics and technology, Wiley, Hoboken, NJ, USA 2002.
[21] S. B. Long et al., “A Model for the Set Statistics of RRAM Inspired in the Percolation Model of Oxide Breakdown”, IEEE Electron Device Lett, vol. 34, no. 8, pp. 999-1001, Aug. 2013.
[22] T. J. Chu et al., “Ultra-high resistive switching mechanism induced by oxygen ion accumulation on nitrogen-doped resistive random access memory,” Appl. Phys. Lett. vol. 105, issue. 22, pp. 223514, Nov. 2014.
[23] C. Y. Lin et al., “Effects of varied negative stop voltages on current self-compliance in indium tin oxide resistance random access memory,” IEEE Electron Device Lett., vol. 36, no. 6, pp. 564–566, Jun. 2015.
[24] P. H. Chen et al., “Improving performance by doping gadolinium into the indium-tin-oxide electrode in HfO2-based resistive random access memory” IEEE Electron Device Lett., vol. 37, no. 5, pp. 584–587, May 2016.
[25] R. Zhang et al., “Characterization of oxygen accumulation in indium-tin-oxide for resistance random access memory,” IEEE Electron Device Lett., vol. 35, no. 6, pp. 630–632, Jun. 2014.
[26] P. H. Chen et al.,“Effects of erbium doping of indium tin oxide electrode in resistive random access memory”, Applied Physics Express, vol.9, no.3, pp.034202, 2016.
[27] P. H. Chen et al., “Bulk Oxygen-Ion Storage in Indium-Tin-Oxide Electrode for Improved Performance of HfO2-Based Resistive Random Access Memory”, Applied Physics Express, vol.9, no.3, pp.034202, 2016.
[28] J. Chen et al.,“Nitrogen buffering effect on oxygen in indium-tin-oxide-capped resistive random access memory with NH3 treatment,” IEEE Electron Device Lett., vol. 36, no. 11, pp. 1138–1141, Nov. 2015.
[29] C. C. Shih et al.,“Resistive switching modification by ultraviolet illumination in transparent electrode resistive random access memory,” IEEE Electron Device Lett., vol. 35, no. 6, pp. 633–635, Jun. 2014.