錳摻雜硒化鉛奈米陣列間不同耦合強度對其元件電性之影響

Abstract

利用奈米晶體(Nanocrystals)製作的元件，其電性會因為晶體間的耦合強度不同而改變，文獻中已有球狀硒化鉛奈米晶體自組裝薄膜，與奈米晶體陣列元件之報導。但在奈米晶體組裝之準一維線狀結構，卻較少有電性方面的報導。本篇論文為探討此準一維線狀陣列系統之電性，並觀察線狀陣列間耦合強度對其元件電性的影響。實驗所用之材料為摻雜錳之硒化鉛奈米線狀陣列，其直徑為80 nm，長度在500 nm-1 μm間。利用標準的電子束微影製程，可製造出彼此間距為250-350 nm的兩個鈦金電極，再藉由介電泳力(Dielectrophoresis Force)將奈米陣列跨接於兩鈦金電極間，並可藉由調變介電泳參數，來控制奈米線狀陣列的並聯根數，與排列緊密程度，量測其對元件電性之影響。論文中分析的元件，皆經過物理或化學方法降低其室溫電阻，即減小接觸電阻問題，之後再以變溫方式量測元件之電流對電壓在不同溫度下(範圍從300 K至100 K)之特性。所有元件之電流對電壓行為，皆為線性關係，然而其電阻對溫度的行為卻不同，我們依據掃描式電子顯微鏡圖像(Scanning Electron Microscope Image)，把元件區分成兩類。第一類之特性為在降溫過程中，電阻變化有一轉折點，轉折溫度以下的電阻維持定值；另一類元件則沒有轉折點，維持負溫度係數之關係，我們用fluctuation-induced tunneling模型來解釋實驗數據。

There are several studies on electrical coupling between colloidal PbSe nanocrystals. The ability to tune electrical properties of devices has received large attention. Here we present a study of the electrical coupling between quasi one-dimensional Pb1-xMnxSe nanoarrays. These nanoarrays are 80 nm in width and 500 nm-1 μm in length. We use standard electron-beam lithography to fabricate two Ti/Au electrodes with a gap of 250-350 nm. The ac electric fields, inducing movement of polarizable nanoarrays, are applied to deposit the nanoarrays into the nanoscale gap. We have prepared devices with various different parallel 1D nanoarrays for electrical characterizations. All the devices are thermally treated or chemically treated (with hydrazine) to reduce the contact resistances. The temperature dependent resistances have been obtained at temperatures from 300 K to 100 K. We found that our devices could be categorized into two types, according to their resistance-temperature behaviors. The first type of devices exhibits a transition temperature below which the resistance is independent of temperature. The second type of devices maintains negative temperature dependence in the whole temperature range. Comparing the different electrical properties of the two types, we concluded that the interchain coupling between quasi 1D nanoarrays results in a variation of resistance-temperature behavior. Moreover, the data can be analyzed in line with the fluctuation-induced tunneling model.