Deposition and Characteristics of TiN, TiNxOy, and N-doped TiO2 Thin Films by Sputtering Using Air as a Reactive Gas

本研究主要是在濺鍍法中以空氣做為反應性氣體製備氮化鈦(TiN)、氮氧化鈦(TiNxOy)與氮摻雜二氧化鈦(N-doped TiO2)薄膜並探討其特性。這三種薄膜具有優異的物理與化學性質是具相當重要工業應用之材料，而傳統上利用濺鍍法製備上述三種薄膜幾均以純氮氣或氮/氧混合氣體做為反應性氣體。若以空氣做為反應性氣體，可在高背景壓力即低真空下製備薄膜，此製程方法可大幅縮短抽真空所耗費的時間，具極大之工業應用價值。 實驗之主要變數為空氣/氬氣流量比值，範圍控制在0.05到2.00，其他製程參數則固定在工作壓力0.13 Pa、濺鍍功率400 W及基板偏壓為50V，以探討其對薄膜特性之影響。由於利用空氣做為反應性氣體，即不須在高真空，而可在較高背景壓力1.3×10-2 Pa (低真空)下操作。另有些部份會利用N2做為反應性氣體製備TiN以及空氣做為反應性氣體製備TiNxOy薄膜，此二者均在高真空下操作，以做對照。 當空氣/氬氣流量比值為0.10~0.15時，所得薄膜為岩鹽型TiN結構，N/Ti比為0.80–0.83而氧含量為9–13 at%、電阻率則為110–130 μΩ•cm、與硬度為26–27 GPa，此與文獻所報導的氮化鈦性質比對後，證實以空氣做為反應性氣體能成功製備出氮化鈦薄膜。本研究顯示在空氣/氬氣流量比值較低時，可製備出氮化鈦薄膜之原因是動力學反應主導所致。由於電漿中氧的解離能比氮的高，使氮較易解離且增加其在低流量比值時之碰撞機率。另推測氮化鈦中固溶少量的氧可以穩定氮化物的結構。 隨著空氣/氬氣流量比值增加到0.18–0.40，薄膜亦具有岩鹽型結構，然其繞射峰往高角度偏移，是為結晶之氮氧化鈦薄膜，其氧含量則增加到22–34 at%、電阻率為230–1460 μΩ•cm、硬度則為25–31 GPa。當空氣/氬氣流量比值超過0.50，結果顯示此時薄膜為非晶質結構，氧含量則增加至39–43 at%、電阻率為2×104–1×106 μΩ•cm。 當空氣/氬氣流量比值增加到1.20時，生成之薄膜為氮摻雜Ti3O5，隨著流量比達到1.40，薄膜轉變為氮摻雜Ti3O5和銳鈦鑛相TiO2混合相。而在流量比值增加到1.80時可得到氮摻雜銳鈦鑛相TiO2。XPS結果顯示氮摻雜銳鈦鑛TiO2氧含量可達7.5 at%。氮摻雜銳鈦鑛TiO2能隙為3.05-3.11 eV，且其能隙隨氮含量增加呈線性下降趨勢。利用可見光可降解亞甲基藍染劑之實驗，顯示氮摻雜銳鈦鑛二氧化鈦具有極佳可見光光觸媒效應。 CASTEP 理論計算不同氮摻雜在二氧化鈦之位置與不同氮摻雜濃度和氧空孔對二氧化鈦能帶結構之影響。結果顯示，氮取代氧的能隙比氮間隙原子之摻雜導致能隙要小，而氧空孔會使二氧化鈦能隙增加。當氧空孔和不同濃度的氮取代氧同時存在時，隨著氮摻雜濃度增加，能隙會下降，此結果可驗證我們之實驗結果。 本研究單純控制不同空氣/氬氣流量比，不僅可製備氮化鈦、氮氧化鈦及氮摻雜二氧化鈦單層薄膜，亦可製備Ti/TiN/TiNxOy/N-doped TiO2 等之多層膜。此製程研究方法簡便且可大幅縮短製程時間，更可應用至其他類似氮化物、氮氧化物與氮摻雜氧化物之薄膜系統，具有極廣泛的工業用途。This study focuses on the preparation and characterization of TiN, TiNxOy, and N-doped TiO2 thin films by reactive sputtering using air as a reactive gas. The films exhibiting many superior physical and chemical properties are of technologically important materials and conventionally prepared by reactive sputtering using nitrogen and/or oxygen as reactive gases. Replacing nitrogen and/or oxygen with air as a reactive gas allows the process to be conducted at high base pressures (low vacuum), which can then reduce drastically the processing time and lead to many industrial applications. The prime deposition variable was simply the air/Ar flow ratio that was varied in a wide range of 0.05-2.00. The deposition parameters of the working pressure (= 0.13 Pa), power (= 400 W), and substrate bias (= 50 V) were fixed throughout the study. Since air was used as a reactive gas, high vacuum was not needed and the base pressure of 1.3×10-2 Pa (low vacuum) was conducted. For comparison in some cases, N2 was also used as a reactive gas for making TiN films at a low base pressure of 6.6×10-4 Pa (high vacuum), whereas TiNxOy films were also prepared by using air as a reactive gas in such a base pressure. As the air/Ar flow ratio reached 0.10-0.15, the films revealed the characteristic rock-salt structured TiN. The N/Ti of the films was 0.80-0.83 with 9-13 at% of oxygen. The determined resistivities and hardnesses of the films were in the range of 110-130 μΩ∙cm and 26-27 GPa, respectively. All obtained data fulfill the characteristics of TiN films. Kinetic formation apparently prevails over thermodynamic predictions at low air/Ar ratios. In such plasma, the dissociation energy of oxygen is much larger than that of nitrogen, which enhances the impingement rate of nitrogen. The dissolution of oxygen in the TiN films seems also to stabilize the nitride structure. When the air/Ar ratio was in the range of 0.18-0.40, the films had the rock-salt structure with the diffraction peaks shifting to higher angles, characterized as crystalline TiNxOy. As the ratio exceeded 0.50, the films were amorphous. The oxygen content in the TiNxOy films increased from 22-34 at% (crystalline) to 39-43 at% (amorphous). The resistivity was about 230-1460 μΩ•cm in the crystalline regime and increased rapidly up to 2×104-1×106 μΩ•cm as the films were amorphous. The hardness of the crystalline TiNxOy films was in the range of 25-31 GPa. As the air/Ar flow ratio reached 1.20, the obtained films were N-doped Ti3O5 and transformed into N-doped Ti3O5 with N-doped TiO2 mixed phases at the air/Ar ratios of 1.40-1.60. N-doped anatase TiO2 films obtained at the air/Ar ratios of 1.80-2.00 could incorporate up to about 7.5 at% of substitutional nitrogen. The measured optical band gaps of the N-doped anatase TiO2 films varied from 3.05 to 3.11 eV. The band gap decreased almost linearly with the nitrogen concentration. N-doped TiO2 films exhibited superior visible-photocatalytic properties from the degradation of Methylene blue test. The band structures of the anatase TiO2 with oxygen vacancies and/or N dopants were calculated using the CASTEP program. The substitutional dopant (NO) was more photocatalytic-effective than the interstitial dopant NI(c), while the band gap increased with increasing the oxygen vacancies. The calculations for the anatase TiO2 with a vacancy and various N dopants concentration reveal that the band gap decreased with the concentration of N dopants, which verifies our above experimental results. Tailoring simply the air/Ar flow ratios in sputtering can yield not only single layers but multilayers consisting of TiN, TiNxOy, and N-doped TiO2 thin films. This green technique may be extended to make other similar nitride, oxynitride, and nitrogen-doped oxide thin film systems, which has great industrial applications.誌 謝 II Abstract III 摘要 V List of Tables IX List of Figures XI Chapter 1 Introduction 1 1.1 Titanium Nitrides, Oxynitrides, and Oxides Thin Films 1 1.2 Magnetron Sputtering Deposition 5 1.3 Motivation 6 1.4 Thesis Overview 6 Chapter 2 Theoretical Background and Literature Survey 9 2.1 Theoretical Background 9 2.1.1 Plasma versus Deposition 9 2.1.2 Structures and Deposition Parameters 13 2.2 Literature Survey 14 2.2.1 Deposition of TiN Films by Sputtering 14 2.2.2 Deposition of TiNxOy Films by Sputtering 22 2.2.3 Deposition of N-doped TiO2 Films by Sputtering 28 2.2.4 CASTEP Calculation of Electronic Structures in Pure and Doped TiO2 33 Chapter 3 Experiments 38 3.1 Experimental Flowchart 38 3.2 Deposition of TiN, TiNxOy, and N-doped TiO2 Films 38 3.2.1 Equipment 38 3.2.2 Deposition Procedures 38 3.3 Characterization 42 3.3.1 Color 42 3.3.2 Crystal Structure 42 3.3.3 Microstructure and Morphology 43 3.3.4 Chemical Composition 43 3.3.5 Electrical, Mechanical, Optic-electronic, and Photocatalytic Properties 44 3.4 CASTEP Calculation 45 Chapter 4 Results 51 4.1 Deposition and Characterization of TiN Films Using Air as a Reactive Gas 51 4.1.1 Color and Appearance 51 4.1.2 Crystal Structure 53 4.1.3 Morphology and Microstructure 56 4.1.4 Chemical Composition 60 4.1.5 Electrical Resistivity and Hardness 68 4.2 Deposition and Characterization of TiNxOy Films Using Air as a Reactive Gas 72 4.2.1 Color and Appearance 72 4.2.2 Crystal Structure 74 4.2.3 Morphology and Microstructure 77 4.2.4 Chemical Composition 81 4.2.5 Electrical Resistivity and Hardness 87 4.3 Deposition and Characterization of N-doped TiO2 Films Using Air as a Reactive Gas 91 4.3.1 Color and Appearance 91 4.3.2 Crystal Structure 92 4.3.3 Morphology and Microstructure 93 4.3.4 Chemical Composition 95 4.3.5 Optical Property 99 4.3.6 Photocatalytic Activity 103 Chapter 5 Discussion 106 5.1 Verification and Process Windows of TiN, TiNxOy, and N-doped TiO2 106 5.1.1 Characteristics of TiN, TiNxOy, and N-doped TiO2 106 5.1.2 Process Windows 118 5.2 Thermodynamically or Kinetically Controlled Formation 119 5.2.1 Thermodynamic Origins 119 5.2.2 Kinetic Reasons 120 5.3 Simulation and Experimental Studies of Electronic Structures 126 5.3.1 TiNx(O) 126 5.3.2 N-doped TiO2 130 5.4 Extended Aspects 137 Chapter 6 Conclusions 144 References 14