原子層沉積技術與奈米結構於功能性薄膜之應用: 太陽能電池之吸光與載子傳輸層、透明導電薄膜及封裝

Abstract

本研究將原子層沉積技術及奈米結構應用於製備功能性薄膜，以解決現今太陽能電池等替代能源及減碳產業之關鍵議題。為有效解決全球暖化的危機，除了考量到技術的表現外，能否相容於現有的工業化製程以合理的成本進行大規模的生產也是問題的核心之一。本研究針對光電元件中常見的功能性薄膜進行開發，包括:太陽能電池中的吸光層與載子傳輸層、透明導電薄膜及封裝層，以解決綠能產業中效率、穩定性及成本高昂等重要議題。 薄膜的製程技術中，主要可分為真空製程及溶液態製程兩大類。為考量日後與可撓式軟性基板及捲對捲製程技術的整合性，真空製程中我們選用原子層沉積技術，溶液態製程則是以奈米材料的懸浮液進行塗膜。兩者相較於其他同類型的技術都有可低溫製程的優點，並各同時擁有獨特的性質，而有機會製備出具多功能性的薄膜。 針對太陽能電池的部分，我們以原子層沉積技術製備了電洞傳輸層，並嘗試開發矽奈米顆粒的溶液態製程以作為新穎的吸光層。藉由適當的調整製程參數，我們成功的開發出可精準控制厚度的氧化鎳製程，並以超薄的氧化鎳薄膜(4奈米)作為電洞傳輸層，取代不穩定的有機材料，達到與其相仿的光電轉化效率。我們亦開發出具共軛連結的改質技術，成功的提升矽奈米顆粒的穩定性。藉由分析光電特性，可推論出此共軛連結具有載子傳輸的特性，有利於未來吸光層之應用。溶液態製程的矽奈米顆粒薄膜經雷射退火後，可達到與真空製程的非晶矽相近的載子傳輸特性。 在透明導電薄膜的開發中，我們使用本實驗室所開發的混合層摻雜技術，搭配前驅物暴露製程，進一步的改善氧化鉿摻雜氧化鋅薄膜的導電度，使其電阻率低達4.5  10-4 Ω cm-1。我們亦嘗試以原子層沉積技術所開發的低溫銅薄膜製程，欲於氧化鉿摻雜氧化鋅中插入金屬材料，進而改善其導電性及可撓性。雖然我們成功的於120 °C的低溫下，製備出電阻率與銅塊相近，僅17奈米的銅膜，然而由於此技術缺乏自我限制的特性，因此無法達到原子層沉積技術該有的再現性。 最後，為了開發薄膜的上封裝層，我們驗證了以高分子單體混摻氧化石墨烯製備多功能阻氣薄膜的可行性。藉由選用可與氧化石墨烯有強作用力的單體，如具有苯環或羥基的丙烯酸酯，成功的將其均勻的分散在單體中。其中又以甲基丙烯酸羥乙酯具有較低的黏度，可提高超音波破碎的效率，進而得到脫層的氧化石墨烯。其低起始黏度也可容納較多的氧化石墨烯含量，在濃度達到三的重量百分比時，此分散液的外觀具有牛奶的光澤，顯示氧化石墨烯於溶液中具有一定程度的排整。在成膜時，此排整將有助於控制氧化石墨烯的排列，延長氣體的穿透路徑，提升阻擋氣體的能力。此漿料亦具有絕佳的存放及熱穩定性，除了有利於規模化生產外，適當的調整熱固化參數將有機會同時還原氧化石墨烯，製備出具有特殊導熱或導電性質的多功能阻氣薄膜。

Applications of atomic layer deposition (ALD) and nanostructures in functional thin films were studied to address the key issues of finding alternative energy and reducing carbon emission. In addition to the device performance, the compatibility of techniques with cost-effective mass production is also the heart of matter. In the study, we focused on the development of common functional films in optoelectronics, including charge transporting layer and absorbing layer of photovoltaics, transparent conductive film (TCF) and gas barrier, to solve the essential problems of performance, stability and high cost in green energy industry. Thin film fabrication can be divided into two categories: vacuum process and solution process. Considering the compatibility with the flexible plastic substrates and roll-to-roll process, we chose ALD and solution process from nanostructure dispersion for their low process temperature among other techniques. In addition, both of them possess unique characteristics, rendering the as-fabricated films with multifunctionality. In the part of photovoltaics, we deposited hole transporting layer (HTL) with ALD and tried to fabricate novel absorbing layer with solution process from silicon nanoparticles dispersion. Through properly adjusting process parameters, we successfully developed nickel oxide (NiO) process able to control film thickness with great precision. The power conversion efficiency (PCE) achieved with an ultrathin NiO film (4 nm) was comparable to that of devices with the commonly-used but instable organic HTL, (poly(3,4-ethylenedioxythiophene) polystyrene sulfonate (PEDOT:PSS). In the attempt to develop absorbing layer, stable Si nanoparticles dispersion was successfully fabricated with conjugated linkage between particles and functional groups. From the analysis of optical properties, we suggested the linkage had the capability of assisting charge transport between nanoparticles and functional groups, which might benefit the application in solar harvesting. Solution-processed Si film with electrical properties similar to that of amorphous Si film deposited by vacuum process was obtained from Si nanoparticles dispersion. In the field of TCF, the resistivity of hafnium-doped zinc oxide (Hf:ZnO), a good transparent conductive gas barrier process developed in our laboratory, was further improved to 0.00045 Ω/cm by an additional 2 s exposure after doping. To further improve the conductivity and flexibility, we tried to insert a metal layer in Hf:ZnO by a low temperature copper process. Although a relatively good conductivity close to the value of bulk Cu was obtained in a 17 nm Cu film at process temperature as low as 120 °C, the reproducibility was poor due to the lack of a self-limiting growth mechanism. In the last part of the dissertation, the feasibility of blending monomer with graphene oxide (GO) for fabricating multifunctional gas barrier was verified. Monomers with ligands which formed strong interaction with GO, e.g. phenyl or hydroxyl groups, were found to disperse GO well. Among the studied monomers, (Hydroxyethyl)methacrylate (HEMA) possessed low viscosity, able to obtain better sonication efficiency and therefore well-exfoliated GO sheets. In addition, the dispersion could accommodate GO concentration high enough to form liquid crystal (LC) without becoming too viscous. At 3 wt% GO in HEMA, the dispersion appeared milky, indicating certain degree of self-orientation of GO. The orientation was beneficial for controlling GO arrangement during film formation, increasing the length of gas permeation path and therefore gas barrier property. Besides, GO/HEMA dispersion showed great storage and thermal stability, which were compatible with industrial production and post reduction of GO for fabricating multifunctional gas barrier film.