標題: 二氧化鉿電阻式記憶體之探討與隨機電報雜訊分析
The Study of Device Characteristics and the Random Telegraph Noise Analysis in HfO2-based Resistive Random Access Memory
作者: 陳敬翰
Chen, Ching-Han
莊紹勳
Chung, Steve S.
電子工程學系 電子研究所
關鍵字: 電阻式記憶體;隨機電報雜訊;操作;氧離子;RRAM;RTN;Operation;Oxygen atom
公開日期: 2013
摘要: 傳統非揮發性記憶體如FLASH 和SONOS 等快閃記憶體有著許多先天上的微縮限制,如隨機摻雜擾動(Random Dopant Fluctuation)、隨機電報雜訊(Random Telegraph Noise)和最小穿隧氧化層厚度。為了解決這些的問題,電阻式記憶體近年來成為非揮發性記憶體的熱門主題,乃因其擁有高密度、低成本、結構簡單、製程容易及具有極佳的微縮特性等優點。 至於電阻式記憶體的氧化物材料有很多選擇,像是二氧化鋅、氧化鎳、二氧化鉿等。 而在本論文中,主要探討以二氧化鉿為基底的電阻式記憶體。 首先,我們採用了掃描(sweep)、脈衝(pulse)及直接操作(non-forming)三種操作方式來探討電阻式記憶體。除此之外,也探討不同氧化層厚度對於電阻式記憶體特性的影響。而在多位元存取方面,我們成功地在掃描操作及脈衝操作中產生四個可辨別的狀態,且相鄰的兩個狀態至少存在100 k(ohm)的差別,且均表現出極佳的資料保存(retention)特性。 其次,我們運用隨機電報雜訊分析,來分析路徑產生對記憶體的判讀產生的影響。藉由分析捕捉時間(capture time)與發射時間(emission time),可以計算出缺陷位置。結合隨機電報雜訊電流趨勢及缺陷位置這兩種概念,我們藉由簡單的電訊量測即可估算出軟性崩潰路徑的截面積大小。除此之外,在隨機電報雜訊量測中,我們首次觀察到一種沒有規律出現的異常雜訊 (abnormal noise)。這種異常雜訊可能是由氧離子移動造成,與隨機電報雜訊是由缺陷捕捉電子或釋放電子的行為不同。
Charge-trapping memory devices (e.g., FLASH, SONOS etc.) have several inherent scaling limits, such as random dopant fluctuation, random telegraph noise, and minimum tunnel oxide thickness. In order to solve the problems, Resistance-change Random Access Memory (RRAM) has recently received much more attention as a result of high-density, low-cost, simple structure (Metal-Insulator-Metal), easy fabrication, and excellent potential of scaling. As to the oxide materials, there are many choices such as ZrO2, NiO, HfO2, etc. In this thesis, we will focus on the HfO2-based RRAM devices. First, three operation methods are used for the comparison of the RRAM devices including sweep operation, pulse operation, and non-forming operation. Besides, a series of experiments have also been carried out to understand how the oxide thickness affects the characteristics of RRAM. As to multi-level storage, we have successfully demonstrated four distinct states not only with sweep operation but also pulse operation and the two nearest states are separated by a window of 100 k () at least which show good data retention in 10,000 seconds. Next, the RTN analysis has been used to examine the effects of soft breakdown paths in RRAM devices. By analyzing the capture and emission time, the trap position could be identified. Combining the two concepts of RTN current variation and the trap position extraction, we can roughly estimate the cross-section area of soft breakdown (SBD) paths simply by the electrical measurements. Furthermore, in the RTN measurements, for the first time, an abnormal noise was found which appeared irregularly. The abnormal noise might be influenced by the oxygen atoms migration which is different from the behavior of RTN signals which are governed by the electron trapping and de-trapping.
URI: http://140.113.39.130/cdrfb3/record/nctu/#GT070050137
http://hdl.handle.net/11536/73016
顯示於類別:畢業論文