基材偏壓對奈米碳管成長的影響與奈米碳管在有毒氣體感測之應用

Abstract

本論文可大致分成兩大部份。第一部份，研究合成奈米碳管(CNT)重要製程參數的影響；第二部份，製造奈米碳管輔助之毒性氣體感測技術。 本研究使用高基材偏壓的方法，取代原本於微波電漿化學氣相沉積法(MPCVD)製備觸媒輔助碳奈米材料中氫氣的功能，在不額外添加氫氣而使用純甲烷當反應氣體成功的合成出奈米碳管(CNTs)。本實驗一開始利用物理氣相沉積法(PVD)將所需觸媒沉積在矽晶片上，接著使用MPCVD氫氣電漿前處理的方式，在基材的表面形成均勻分散的觸媒，隨即便可進行數種奈米碳結構的合成；本實驗探討的主要製程參數包含觸媒的種類，有鐵、鈷與鎳三種、基材偏壓介於0伏特到-320伏特之間和沉積時間1分鐘到20分鐘之間；合成出來的碳奈米結構利用掃瞄式電子顯微鏡(SEM)、高解析穿透式電子顯微鏡(HRTEM) 、拉曼光譜儀(Raman spectroscopy) 與場效發射電性量測等方法來分析其結構與特性。結果顯示就算不額外添加氫氣，在高於-200伏特的偏壓下，便可得到不同的奈米結構。結果也證實文獻中氫氣在奈米碳管或其他奈米材料的合成機制上太過強調其重要性。在以鎳金屬為觸媒基材偏壓-320伏特的實驗中，隨著沉積時間從4分鐘到20分鐘的過程中，產生的奈米結構會由最初均勻分散的奈米碳管變成準直性的奈米碳管，最後形成義大利麵狀的奈米碳管；在數個實驗參數中，可以得到高管束密度、高純度與大小均一的準直性奈米碳管，其含有較低的啟動電壓5 V/μm與在10 V/μm電場下最大的電流密度高於35 mA/cm2等場效發射性質(儀器最高極限)。最後我們發現在無額外添加氫氣的奈米碳管製程中，提供一個數據來探討氫氣與積材偏壓在奈米碳管成長機制所扮演的角色。 在室溫下有毒氣體的感測實驗中，開發出以多壁奈米碳管(MWCNT)輔助而成包括30組檢測區塊(15個檢測元件各包含2組區塊)的系統晶片。待測氣體為模擬四種戰爭中危險的化學武器：二氯甲烷(dichloromethane), 乙腈(acetonitrile), 2-氯乙基乙基硫醚(2-chloroethyl ethyl sulfide)，和甲基膦酸二甲酯(dimethyl-methyl phosphonate (DMMP))，選擇14種帶有不同官能基特性的聚合物材料，接著以溶液液滴沉積的簡化製程方式將檢測材料堆疊沉積成以聚合物材料/MWCNTs/矽(001) 的形態之30種檢測區塊。這樣的檢測晶片堆疊方式可以有效防止MWCNTs直接接觸到待測氣體並與其反應，進而延長檢測晶片的壽命與維持靈敏度。本研究的檢測原理為偵測檢測晶片在接觸目標氣體後，各檢測區塊中電阻值的變化。結果顯示在每種氣體的檢測後分析都可得到如指紋般特殊的雷達圖，並且將得到的雷達圖利用3-D主成份數學分析軟體(PCA)可以在單一測試中很快速地分辨氣體的種類；結果也顯示各區塊的阻值變化與四種有毒氣體的濃度有著線性變化的相對關係，甚至對於DMMP這個有毒氣體，在製程仔細的監控下推論偵測極限可以達到43ppm甚至更低低濃度的檢測。

This dissertation could be basically divided into two parts. The first part was studying the important nanofabrication parameters of the carbon nanotubes (CNTs), the second part was to fabricate the the CNTs-assisted sensor array to detect the toxic gases. About the nanofabrication parameters, effect of replacing hydrogen gas with high substrate bias on formation of the catalyst (Fe, Co, Ni)- assisted carbon nanostructures by microwave plasma chemical vapor deposition (MPCVD) was examined, i.e., under pure CH4 as the only source gas and without additional H2 gas introduction. The Si wafers were first deposited with catalyst by physical vapor deposition (PVD), and then followed by H-plasma pretreatment to obtain the well-distributed catalyst particles on wafer. The pretreated specimens were then deposited in MPCVD to manipulate the various carbon nanostructures. The main process parameters include catalyst material (Co, Ni and Fe), the substrate bias (0 V to -320V), and deposition time (1 min to 20 min). The as-deposited nanostructures were characterized by SEM, HRTEM, Raman spectroscopy, and field emission I-V measurement. The results indicate that the various nanostructures on Si wafers can be manipulated under sufficient high negative substrate bias (> -200 V) without the additional hydrogen introduction. These results also improve our understanding on growth mechanisms of CNTs or other nanostructures, where the most of the proposed mechanisms in the literature were too much emphasis on the role of hydrogen. Based on the deposition conditions of -320V substrate bias with Ni as catalyst, it shows that the deposited nanostructures can be varied from the well-distributed CNTs at the initial growth stage, the well-aligned CNTs to become the spaghetti-like nanostructures by increasing the deposition time from 4 mins to 20 mins. Under the present deposition conditions, the best field emission properties are for the well-aligned CNTs nanostructures, in which the turn-on field strength can go down to 5 V/μm and the maximum current density > 35 mA/cm2 (which is our maximum instrument limitation) at field strength of 10 V/μm. In conclusion, the formation of CNTs without introducing additional hydrogen source provides the data base to clarify the roles of hydrogen and substrate bias in CNTs growth mechanism. About the room temperature toxic gas sensing, a system chip with a MWCNT (multiwalled carbon nanotube)-assisted array of 30 sensors (two sensors for each of 15 sensor types) was developed. Gases tested include four simulants of chemical warfare agents: dichloromethane, acetonitrile, 2-chloroethyl ethyl sulfide, and dimethyl-methyl phosphonate (DMMP). By selecting 14 different functional polymer materials, each composite sensor comprised of 30 sensing stacks (polymer/MWCNTs/Si(001), wafer) was fabricated by a solution droplet casting method to simplify the process. The principle of gas sensing is basically to measure the resistivity change of the composite sensor device upon contact with a target gas. One of the advantages of the sensing stack having a polymer overlayer above the MWCNT layer is to protect the MWCNT from direct interaction with the gas to improve sensor life and sensitivity. The results indicate that a fingerprint pattern of the sensor radar plot can be determined for each testing run, and that specificity can be achieved through a 3-D principal component analysis (PCA) of the radar plots. The results also show that a linear relationship between the resistance response and concentration is clearly evident for these four toxic gases. Therefore, it is believed that by using radar plots analyses and the PCA method the unknown toxic gases can be identified rapidly. By extrapolation and careful process monitoring, a sensitivity much lower than 43 ppm for DMMP vapor is likely.