Please use this identifier to cite or link to this item:
http://tdr.lib.ntu.edu.tw/jspui/handle/123456789/97162| Title: | 磁性穿隧元件之薄膜調整與電性研究 Modulation of Thin Film and Research of Electrical Property for Magnetic Tunnel Junction |
| Authors: | 黃文翰 Wen-Han Huang |
| Advisor: | 白奇峰 Chi-Feng Pai |
| Keyword: | 磁阻式隨機存取記憶體,磁穿隧結,自旋軌道耦合,磁阻比,鈷鐵硼/氧化鎂/鈷鐵硼系列, Magnetoresistive Random Access Memory (MRAM),Magnetic tunnel junction (MTJ) MTJ,Spin-orbit-torque (SOT),Tunneling magnetoresistance (TMR),CoFeB/MgO/CoFeB series, |
| Publication Year : | 2024 |
| Degree: | 碩士 |
| Abstract: | 磁阻式隨機存取記憶體是未來潛在的一種記憶體技術,其優勢在於非揮發性。磁穿隧結是磁阻式隨機存取記憶體的基本單元。在磁穿隧結的所有機制中,自旋軌道耦合型-磁穿隧結和磁穿隧阻比型-磁穿隧結是製造最有效率磁穿隧結的兩個關鍵元素。自旋軌道耦合型-磁穿隧結提供了一種低能耗的寫入方式,使用鎢作為自旋電子來源層。另一方面,磁穿隧阻比型-磁穿隧結理論發現鈷鐵硼/氧化鎂/鈷鐵硼系列能夠提供最高的磁阻比,使得不同狀態可以輕易被區分。基於這兩個關鍵元素,我們堆疊了 鉭/鎢/鈷鐵硼/氧化鎂/鈷鐵硼/鎢/鉭 的薄膜結構,並調整到最佳參數。為了在低功耗和高耐久性之間取得平衡,合理的矯頑場對於磁穿隧結的兩個磁性層(自由層和固定層)至關重要。在這篇論文中,我們展示了擁有垂直異向性和水平異向性的磁穿隧結膜厚調整,以及基於這些調整的矯頑場變化。此外,我們還展示了磁穿隧結的製造過程。未來,這些裝置不僅可以作為記憶體使用,還有可能用於模擬人工智能領域中生物神經元和突觸的行為。 Magnetoresistive Random Access Memory (MRAM) is a potential memory in the future, with its advantage on non-volatility. Magnetic tunnel junctions (MTJs) are the basic unit in MRAM. Among all mechanisms of MTJs, spin-orbit-torque MTJ (SOT-MTJ) and tunneling magnetoresistance (TMR) - based MTJ are two key elements in producing effective MTJs. SOT-MTJ provides a low-energy-consuming way to write in MTJs, using tungsten as its spin source layer. On the other hand, TMR-based MTJ suggests that CoFeB/MgO/CoFeB series provide the highest TMR ratio, where different states can easily be distinguished. Based on the two key elements, we deposit a stack of Ta/W/CoFeB/MgO/CoFeB/W/Ta thin film and modulate it to the best parameters. In order to strike the balance between low power consumption and high endurance, it is crucial to acquire a reasonable coercivity field (Hc) for the two magnetic layers, free layer and fix layer, in MTJs. In this thesis, we demonstrated the thickness modulations of PMA and IMA MTJs, and the Hc change based on these modulations. Also, we demonstrated the process of MTJ fabrication. In the future, these devices can not only be memories, but potentially also be used in simulating behaviors of biological neurons and synapses in artificial intelligence field. |
| URI: | http://tdr.lib.ntu.edu.tw/jspui/handle/123456789/97162 |
| DOI: | 10.6342/NTU202404513 |
| Fulltext Rights: | 同意授權(全球公開) |
| metadata.dc.date.embargo-lift: | 2025-02-28 |
| Appears in Collections: | 材料科學與工程學系 |
Files in This Item:
| File | Size | Format | |
|---|---|---|---|
| ntu-113-1.pdf | 4.94 MB | Adobe PDF | View/Open |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.
