自發有序排列之金屬奈米顆粒的熱穩定性及相關研究：藉由程式控溫即時AES、LEED和熱脫附切入

Thermostability and Related Issues of Self-Aligned Metal Nanoparticles: Research by Temperature Programmed Real-Time AES, LEED, and Desorption

劉冠德(Kuan-Te Liu)

指導教授 ： 林敏聰

共同指導教授 ： 宋克嘉(Ker-Jar Song)

國立臺灣大學/理學院/物理研究所/碩士(2007年)

https://doi.org/10.6342/NTU.2007.01610

摘要

我所屬實驗團隊之前將各種金屬原子蒸鍍在Al2O3/NiAl(100)基板上，製作出具有以下優點的金屬奈米顆粒(nanoparticles; NPs)：有序的排列、均勻的尺寸分佈(~2.7nm)以及高熱穩定性。而這些優點歸因於氧化鋁層表面產生獨特的週期性一維長條結構，其間距約4 nm。以上結果主要由STM得出，而本論文實驗則利用標題所述方法來做包括熱穩定性在內的相關研究，主要包含二部分：
第一部分是找出大幅提升該系統熱穩定性的方法。之前實驗[3]對STM資料的解釋是Co NPs的熱穩定性甚高，1.5 ML可於長時間高溫退火（800-1090 K）處理下仍保持約260/104 nm2的粒子密度。但在本獨立實驗中利用程式控溫即時Auger電子能譜（TPA）和熱脫附（TPD）卻得到和上述矛盾的結果：Co並不會以任何形式脫附，而於大約600 K開始明顯穿過氧化層向NiAl內部擴散，而於1000 K時近乎所有的Co皆全部擴散進NiAl內部。我們檢討數據和實驗記錄後找出合理的解釋並加以證實： Co NPs的熱穩定性會因曝氧而提升近400 K。導致此結果的機制有二：I. 氧氣可鈍化Co NPs的表層。II. Co NPs 擔任催化劑的角色來催化CoOx和NiAl基底之間的氧化還原反應而加厚其間的氧化鋁層。此兩機制合併延緩NPs向下擴散和相互燒結合併的過程。此法可能可以應用在多種不同的金屬NPs上。這個在STM實驗[3]中未預期的曝氧效果其可能的發生原因為較長的實驗時間（樣品放置過夜）、STM樣品架於昇溫時吐氣和較少的Co覆蓋量。我們也嘗試了CO的影響，而類似的Fe NPs的實驗結果也包含在其中。
第二部分是藉由改變氧化鋁層和基材間的晶格錯位關系來影響氧化鋁層表面週期性一維長條結構的間隔大小。改變法是調整NiAl中表層Ni和Al的比例以及摻入Co和Fe以期於表層形成不同晶格大小的穩定合金相。雖有了部分成果，但需以更高表面解析能力的工具如SPA-LEED和STM作進一步分析。
在本論文最後的附錄中包含對氧化鋁層本身的研究，包括不同溫度曝氧和不同加熱速率下的脫附行為的實驗，以及本文實驗資料補遺。
對於如何製造熱穩定性更好、具特定大小、密度及排列方式的磁性奈米粒子，本實驗的結果可說是指引出一個新的方向。第一部分恰似化學家們處理溶液中的微粒所常用的方法，同時此CocoreCoOshell NP的FM金屬核和AFM氧化外殼之間的磁交換耦合（exchange coupling）可望增加額外的異向性（anisotropy）而使其FM金屬核的磁距的熱穩定性（blocking temperature; TB）增加，最後突破NP的超順磁極限（superparamagnetic limit）[39~41]。而第二部分也極具發展性，有待後人作更進一步研究。

關鍵字

自發有序排列 ； 奈米顆粒 ； 熱穩定性 ； 程式控溫即時AES ； 晶格錯位

並列摘要

A previous series of studies of our lab[1~3] on vapor depositing several kinds of metals on Al2O3/NiAl(100) had created metal nanopar-ticles(NPs) with following characteristics: well-ordered alignment, self-limiting size distribution with average size of ~2.7nm, and high thermostability. And these features can be attributed to peculiar one-dimensional long stripes with ~4 nm interdistance on the surface of the ultrathin Al2O3 template. Above conclusions are results of STM re-search. This thesis includes researches on thermostability and related issues of the same system studied by methods mentioned in thesis title. It contains two major parts:
The first part is about finding the method which can greatly en-hance thermostability of NP assembly. One of the conclusions of STM experiments of a previous study[3] is that Co NPs have high thermosta-bility, e.g. 1.5ML Co NP assembly sustains the density of ~260/104 nm2 even after 800 – 1090 K annealing. But in this independent experiment, We employed Temperature Programmed real-time Auger (TPA) and Temperature Programmed thermal Desorption (TPD) and found that pure Co NP assembly would not desorb in any form but start diffusing into the NiAl bulk at ~600K and it’s auger signal disappears almost completely at about 1000K. This disagrees with the previous conclu-sion[3]. After reviewing data and records of previous experiment, we found rational explanation and proved it: thermostability of Co NPs can be enhanced nearly 400 K by exposing them to oxygen. The final result can be attributed to two mechanisms: I. Oxygen can passivate the boundary of each Co NP. II. Co NPs work as catalyst to catalyze a redox reaction between part of CoOx and NiAl substrate, which increases the thickness of the Al2O3 film between them. Combining these two mechanisms can prevent Co NPs from sintering with each other and diffusing into substrate. This method may be applied on several differ-ent kinds of metal NPs. The unexpected oxygen surfactant in previous STM experiment[3] possibly come from longer experimental time (overnight), STM sample holder outgasing, and smaller Co coverage. We also show the effect of CO as surfactant and TPA results of Fe NPs here.
The second part is about changing the interdistance of peculiar one-dimensional long strips on the surface of the single-crystalline Al2O3/NiAl(100) by altering lattice mismatch between alumina over-layer and substrate. The idea is to vary the ratio of Ni to Al in surface layer of NiAl(100) and alloying it with Co or Fe then we expect some new stable alloy phase with lattice constant differing from original would form on surface layer of NiAl(100). Although we got some achievements here, they need to be analyzed further by instruments with higher surface analyzing abilities like SPA-LEED and STM.
There is an appendix in the final of this thesis. It contains some supplementary data from experiments of alumina layer on NiAl(100). The research in alumina layer includes exposing to oxygen at different temperatures and desorbing behaviors under different heating rates.
Results in this thesis may open a new door to create more thermal stable magnetic metal NPs of special size, density, and alignment in UHV. First part just like what chemists usually do to isolate particles from each other in solution. And 2nd part also has lots of possibilities waiting for further studying.

並列關鍵字

self-aligned ； nanoparticle ； thermostability ； temperature programmed real-time AES ； TPA ； lattice mismatch

參考文獻

[30] 林君岳，臺灣大學物理系碩士論文，2006年

[16] 林文欽，臺灣大學物理系博士論文，2006年

[32] 黃伯群，臺灣大學物理系碩士論文，2005年

[42] Shangjr Gwo, Chung-Pin Chou, Chung-Lin Wu, Yi-Jen Ye, Shu-Ju Tsai, Wen-Chin Lin, and Minn-Tsong Lin, Phys Rev. Lett. 90, 185506-1 (2003)

[3] W. C. Lin, S. S. Wong, P. C. Huang, C. B. Wu, B. R. Xu, C. T. Chiang, H. Y. Yen, and Minn-Tsong Lin, Appl. Phys. Lett. 89, 153111 (2006).