小分子在二氧化鈦表面上吸附及反應之理論計算

Abstract

本論文使用以投影擴充波（PAW）的密度泛函理論（DFT）為基礎的第一原理，計算各種不同小分子在二氧化鈦表面上之吸附及反應，其使用之位勢法則考慮以一般化梯度近似（GGA）為基礎的不同泛函。為改良太陽能電池應用中的異質接面問題，而達到較高的效率，我們考慮利用一些無機連接分子，以供半導體量子點在二氧化鈦分子上成長，論文亦預測出相關的反應物、生成物、中間產物、過渡態、吸附能、反應位能面、態密度、投影態密度、Bader電荷分析、反應速率常數，在本論文中，共研究五種連接分子，並分成三類討論。

(1)硫化氫於潔淨二氧化鈦表面（包含金紅石相(110)面及銳鈦礦相(101)面）之反應，結果顯示，在兩個表面上，伴隨水分子生成之氧↔硫交換之二氧化鈦表面皆為主要產物，但由預測出的反應位能面顯示，此產物在銳鈦礦相(101)面上又較在金紅石相(110)面上易形成，態密度及投影態密度的結果則顯示，硫表面摻雜之二氧化鈦其價帶可被修飾，進而達到縮小二氧化鈦能隙之目的。

(2)過氧化氫於潔淨二氧化鈦表面（包含金紅石相(110)面及銳鈦礦相(101)面）之反應，本論文預測出完整的吸附、反應機制、反應速率常數，由反應位能面及反應速率常數的結果顯示，主要產物為伴隨水分子產生之氧覆蓋二氧化鈦表面，而非如前人所預測，為由H2O2(a) □ 2HO(a)反應產生之雙羥基產物，此結果有助於仰賴二氧化鈦表面氧位置之太陽能電池製備及由水分子產物引起的水裂解產氫。

(3)以含矽的連接分子為矽量子點於二氧化鈦表面成長之系列研究，首先考慮矽甲烷於潔淨二氧化鈦表面（包含金紅石相(110)面及銳鈦礦相(101)面）之反應，然而，預測之反應位能面顯示SiHx(a)□ SiHx-1(a) + H(a)分裂反應並不容易發生，於是本論文接著研究四氯化矽於羥基覆蓋之二氧化鈦表面（銳鈦礦相(101)面，其為太陽能電池應用之主要相）之反應，預測出的反應位能面顯示其較矽甲烷於潔淨表面反應之優越性，其主產物Cl2Si-OO(a)可經由放熱的無能障反應而得，提供矽量子點於羥基覆蓋之二氧化鈦表面成長之穩定連接分子，最後，本論文經由SiHx與Cl2Si-OO(a)之反應來探討矽－矽鍵之形成，預測出的反應位能面顯示矽－矽鍵之形成與x強烈之關聯性，大體上而言，隨著x變小，矽－矽鍵之形成也隨之更易形成，故具鍵結能力的實驗（例如：微波裝置）將有助於由SiHx與Cl2Si-OO(a)反應之矽－矽鍵形成。

This thesis studies the adsorptions and reactions of some small molecules on the TiO2 surfaces by first-principles calculations based on the density functional theory (DFT) in conjunction with the projected augmented wave (PAW) approach, using different generalized gradient approximation (GGA) functionals. To improve the heterogeneous interface and achieve higher efficiencies for solar cell application, some inorganic linkers are employed for semiconductor quantum-dot growth on the TiO2 surface. We investigate the stability of those molecular linkers on TiO2. The corresponding reactants, products, intermediates, transition states, adsorption energies, potential energy profiles, density of states, projected density of states, Bader atomic charge analyses, and rate constants are predicted. In this thesis, five linkers, classified into three types, are studied.

(1) The reactions of H2S and its fragments on the clean TiO2 rutile (110) and anatase (101) surfaces are investigated for InS quantum dot (QD) growth. The results of calculations show that the O ↔ S exchanged TiO2 surface and H2O formation, S(v-O2c) + H2O(g), are the main products on both surfaces, of which the H2O formation reactions can occur more readily on the TiO2 anatase (101) surface than on the TiO2 rutile (110) surface based on the predicted potential energy profiles. The DOS and PDOS results show that the S-surface-doped TiO2 can modify the TiO2 valence band, lowing the TiO2 band gap significantly.

(2) The reactions of H2O2 and its fragments on the clean TiO2 rutile (110) and anatase (101) surfaces are studied. The complete adsorptions, reaction mechanism, and reaction rate constants of are predicted. The results of potential energy profiles and rate constants show that the main products are H2O(g) + O(a), instead of the two HO radicals formed by the reactions of H2O2(a) □ 2HO(a) as proposed in a previous study. The results are helpful for the solar cell fabrication application with O-TiO2(a) which may be more reactive for QD growth.

(3) A series of studies on Si-containing linkers are considered for Si-QD growth on the TiO2 surface. SiHx reactions on the clean rutile (110) and anatase (101) TiO2 surface are investigated first. However, the predicted potential energy profiles show that SiHx(a) □ SiHx-1(a) + H(a) decomposition reactions cannot take place easily. Hence, we consider SiClx dehydrochlorination reactions on the hydroxylated anatase (101) TiO2 surface, which is the main phase considered for the solar cell application. The predicted potential energy profiles show the facility of these processes, comparing with the SiHx reactions on the clean surface. Cl2Si-OO(a) is the main product formed barrierlessly and exothermically, which can be considered as a stable linker for Si-QD growth on the hydroxylated surfaces. The final study of SiHx reactions with Cl2Si-OO(a) are studied to understand the Si-Si bond formation. The predicted potential energy profiles show the strong relationship between the Si-Si bond formation and “x”. Roughly speaking, the Si-Si bond formation processes become easier when the x is smaller. Hence an experiment with a bond breaking technique such as microwave discharge should be helpful for the Si-Si bond formation by utilizing the SiHx and SiCl2 reactions.