萬本儒臺灣大學：化學工程學研究所丁致遠Ting, Chih-YuanChih-YuanTing2007-11-262018-06-282007-11-262018-06-282004http://ntur.lib.ntu.edu.tw//handle/246246/52254本論文中以旋轉塗佈的方式製備中孔洞二氧化矽薄膜，並將其應用在低介電薄膜及抗反射薄膜上。薄膜中的孔洞，是先使用表面活性劑的微胞做為模版，之後再將模版移除所產生；而薄膜的製備是採用旋轉塗佈的方式。在論文中探討了薄膜厚度的控制，旋轉塗佈的影響，模版分子的效應，薄膜性質與薄膜結構的關連，以及超低介電係數材料與抗反射膜材料的開發研究。 旋塗薄膜的厚度跟旋塗溶液組成及旋轉塗佈配方相關，為了控制薄膜厚度，期望能找出顯著影響厚度的因素。論文中整理了文獻上旋塗薄膜的厚度理論分析，除了期望瞭解及整理影響薄膜厚度的變因外，主要期望評估是否可開發一理論厚度模式，來描述論文中的實驗系統。然而，從文獻整理中歸納的結論是，相較於推導理論模式，開發經驗關連式應是較實用且簡單的方法。因此在論文中探討了溶液組成，旋轉塗佈配方，以及鍛燒升溫方式對厚度的影響。研究中發現，溶液中TEOS的濃度影響薄膜厚度最大。 論文中同時探討旋轉塗佈製程以及鍛燒程序對中孔洞薄膜的影響，所探討的轉速為2600rpm，時間為0.5分鐘。由於同一薄膜在不同位置的折射率幾乎相同，推論旋塗薄膜之孔隙度在不同位置是相近的。研究中顯示，旋塗薄膜結構的規則性較未旋塗薄膜差，但是兩者d-spacing值的差異很小。另外，在旋塗過程中，部分模版從旋塗溶液中析出的現象（相分離）是較值得注意的。當以區塊共聚高分子P123為模版製備薄膜時，較濃的P123塗佈溶液（P123/TEOS=41wt%）將導致部分P123分子在旋轉塗佈的過程中發生相分離現象而析出，且會使薄膜產生龜裂及孔洞體積較少（與以較稀之P123旋塗溶液，P123/TEOS=20wt%，所製備之薄膜相比）的現象。關於鍛燒程序對中孔洞薄膜的影響，研究中發現只有當使用C16TMABr為模版所製備的薄膜，在鍛燒後薄膜結構有劇烈收縮且規則性變差的現象；而研究中也發現，相較於以其他模版分子所製備的薄膜，該薄膜中的孔壁厚度較薄。因此推論在以C16TMABr為模版的薄膜中，所觀察到的結構劇烈收縮及結構規則性變差的現象，應是肇因於該薄膜中較薄的孔壁厚度。 論文中也探討薄膜性質與薄膜結構的關係。薄膜的折射率及介電係數可輕易地藉由孔隙度來調整，但是隨著薄膜孔隙度增加，薄膜的漏電流呈指數攀升，且薄膜的機械強度也大幅減弱；因此如何在增加薄膜孔隙度的同時，抑制漏電流的攀升及維持薄膜的機械強度將是未來應用上非常重要的課題。另外研究中也發現，以離子性表面活性劑C16TMABr為模版所得薄膜之漏電流值因為太高而超過了儀器所能量測的範圍。同時由電容電壓曲線中發現，該薄膜的平板電壓為正值，這顯示了有釵h負電荷（應為Br－離子）在薄膜中。這些殘存的離子可能導致了該薄膜極高的漏電流值；由於離子型表面活性劑易產生殘餘離子，易導致極高的漏電流，因此並不適合作為低介電薄膜。若以非離子性的模版分子製備薄膜，薄膜之平板電壓均為負值，這顯示薄膜中有釵h正電荷的存在；另外隨著薄膜孔隙度的增加，薄膜平板電壓的絕對值越大，這顯示了當薄膜孔隙度增加的時候，薄膜中的正電荷也越多。這些正電荷應是由薄膜結構中氫氧基的氫原子所貢獻，因此也顯示在孔隙度較高的薄膜中，氫氧基也越多；而薄膜中氫氧基的增加，將導致薄膜漏電流的增加，恰與研究中漏電流隨著薄膜孔隙度的增加而呈現指數攀升的實驗現象相互印證。 針對低介電薄膜的應用，超低介電係數（1.47）可藉由調整塗佈溶液的組成及反應時間來達到。論文中使用Tween80為模版。藉由調整溶膠凝膠程序中反應物濃度（水），催化劑濃度（鹽酸），以及反應時間，可加速溶膠凝膠程序的水解及縮合反應，而研究中顯示，當水解及縮合反應較完整時，所得薄膜的介電係數也越低。推論當水解及縮合反應較完整時，薄膜結構較為堅固，因此能形成孔隙度較高的薄膜，也促成了介電係數的下降。除此之外，由於縮合反應較為完整，薄膜中易被極化的氫氧基也較少，這也是可能促成薄膜介電係數下降的原因之一。薄膜之漏電流密度一般在10-7 A/cm2或更小的範圍。另外，在一般大氣環境下，薄膜的介電係數在一個月後依舊維持恆定。 針對抗反射膜的應用，在波長590nm下，高穿透率（99%）的效果也成氐ル拲惆謏雂炷g膜的折射率及厚度而達成。研究中所使用的模版分子為Tween80。本研究以塗佈溶液中模版分子與TEOS的比例來控制折射率，以溶液中乙醇的濃度及轉速來控制膜厚，由於本研究中的薄膜折射率及厚度易於控制，因此不但適合作為抗反射膜的材料，也可應用於其他相關的濾光片材料。The preparation of spin-on mesoporous silica films and their applications as low dielectric constant (low-k) films and anti-reflection (AR) films were studied in this dissertation. The pores were introduced via the surfactant-templated method, and the film was prepared via the spin-coating process. The film thickness control, the effects of the spin-coating process, the effects of the templates, the correlations between the film architecture and the film properties, and the development of the ultra low-k materials along with the AR materials were investigated. The thickness of the spin-on films is related to the solution composition or the spin recipe. In order to control the film thickness, it was expected to find out the significant factors influencing the thickness. To understand more about the spin-coating, the theoretical models developed in the literatures were reviewed. Moreover, it was hoped to evaluate the possibility for developing a theoretical model that was capable of describing the relationship between the film thickness and the spin-coating conditions for the system in this dissertation. However, it was concluded from this research that to obtain an empirical correlation between the thickness and the operating conditions (including the coating solution composition and the process parameters) was more practical and easier than to derive a theoretical model. Therefore, the effects of the solution composition, the spin recipe, and the calcination on the thickness were examined and compared experimentally. From the results of this research, it was found that the most significant factor in the thickness control was the concentration of tetraethylorthosilicate (TEOS) in the colloid solution. The refractive index, the optical microscope, the X-ray diffraction and the nitrogen adsorption/desorption were used for the characterization of the effects of spin-coating and calcination on the mesoporous films made with different templates. The spin-on films were spin-coated at 2600 rpm for 0.5 min. The uniformity of the spin-on films was demonstrated by the uniform refractive indices at different positions. Moreover, the XRD results showed that the spin-coating only caused slightly poorer structural periodicity and sometimes minor variation of the d-spacing. However, the more notable effect on the films was the phase separation of some templates from the coating solution during the spin-coating. For the films templated with the block copolymer Pluronic P123, the phase separation of P123 was observed when the more concentrated P123 coating solution (P123/TEOS=41wt%) was used. The phase separation of P123 resulted in a crack film, whose porosity was less than that prepared with the less concentrated P123 solution (P123/TEOS=20wt%). For the effect of the calcination, on the other hand, it was observed that the calcination only caused the structure of the film templated with cetyltrimethylamonium bromide (C16TMABr) to collapse and shrink. For the films templated with non-ionic surfactants or copolymers, significant shrinkage of the structure was not observed. The thicker walls around the pores in the films templated with non-ionic templates may strengthen the structure and resist the shrinkage. The correlations between the film properties and the film architecture were also studied. It was concluded that the k value and the refractive index can be easily controlled by tuning the porosity of the film; however, both the larger leakage current density and the weaker mechanical strength for the more porous films were critical challenges for practical applications. For the film templated with C16TMABr, an ionic surfactant, the leakage current was too large to be measured. The flat band voltage of the film was positive. It indicated that there were a lot of negative charges (possibly residual Br－ ions from C16TMABr) within the film. These residual ions may result in the ultra high leakage current. Due to the ultrahigh leakage current exhibited by the film templated with C16TMABr, the ionic surfactants did not seem to be appropriate templates for the application as the low-k films. For the films prepared with different types of nonionic templates, the flat band voltages were negative. It indicated that there were positive charges within the film. Moreover, the flat band voltage of the more porous film shifted to a more negative value. It indicated that there were more positive charges within the more porous film. These positive charges may result from the protons of the silanol groups in the film structure. It suggested that there were more silanol groups within the more porous film. The increase of the silanol groups should cause the elevation of the leakage current density, which was consistent with the higher leakage current density of the more porous film observed in this research. For the development of the ultra low-k films, an ultra low dielectric constant (1.47) at 1 MHz was achievable by tuning the parameters such as the colloid composition and the mixing time of the coating solution. Polyoxyethylene(20) sorbitan monooleatel, also known as Tween80, was used as the template. By controlling the concentration of the reactant (H2O), the concentration of the catalysts (HCl), and the reaction time, the hydrolysis and the polycondensation reactions of the sol-gel process can be facilitated. The dielectric constants of the films were found to be decreased with the extent of the hydrolysis and the polycondensation reactions within the sol-gel solution. There may be two reasons for the reduction of the k values. First, the film structure was stronger and thereby retained more pore volume if the hydrolysis and the polycondensation reactions were more complete. Second, the more polarizable silanol groups should be reduced when the polycondensation reactions were more complete. The leakage current densities can be of 10-7 A/cm2 order or lower under an electric field of 1 MV/cm. Moreover, the dielectric constants kept almost the same after the films were exposed in the atmosphere for one month. For the development of the AR films, a high transmittance (99%) at 590nm (wavelength) was achievable by tuning the refractive index and the film thickness of the AR films on the glass (the transmittance was originally about 92%). The advantage, that both the refractive index and the film thickness can be easily controlled, makes the surfactant-templated film quite attractive for being an optical filter material.Abstract Ⅰ Chinese Abstract Ⅱ List of Figures Ⅵ List of Tables Ⅹ Chapter 1 Introduction 1 1.1 Background and Motivation 1 1.2 Surfactant-Templated Method 5 1.2.1 Mechanism of surfactant-templated method 5 1.2.2 Features of mesoporous or mesostructured materials from surfactant-templated method 10 1.2.3 Applications of mesoporous or mesostructured materials 12 1.2.3.1 Semiconductor industry: future low-k films 12 1.2.3.2 Optical filters 13 1.2.3.3 Other optical applications 14 1.2.3.4 Sensors or adsorbents 15 1.2.3.5 Bio-related applications 17 1.2.3.6 Catalyst 18 1.2.4 Formation of surfactant-templated films 20 1.3 Objective and Scope of Dissertation 22 Chapter 2 Experimental 23 2.1 Chemicals 23 2.2 Apparatuses 24 2.3 General process for preparation 25 2.3.1 Cleaning substrates 25 2.3.2 Film preparation 25 2.4 Characterization 27 2.4.1 Structure Charaterization 27 2.4.1.1 X-ray diffraction 27 2.4.1.2 Nitrogen adsorption/desorption measurements 28 2.4.2 Property Characterization 30 2.4.2.1 Fabrication of MIS structure 30 2.4.2.2 Dielectric constant 30 2.4.2.3 Flat band voltage 31 2.4.2.4 Leakage current 31 2.4.2.5 Thickness 32 2.4.2.6 Refractive index 32 2.4.2.7 Mechanical strength 32 Chapter 3 Spin Coating Effects on Thickness 33 3.1 Introduction of Spin Coating Process 33 3.2 Review of Spin Coating 36 3.2.1 Theoretical Modeling-Review 36 3.2.1.1 Evaporation neglected 36 3.2.1.2 Evaporative effects 41 3.2.2 Prediction of Thickness by Theoretical Models 49 3.2.2.1 Polymer solution 49 3.2.2.2 Colloidal solution 50 3.2.2.3 Sol-gel solution 51 3.2.3 Radial Uniformity of Thickness 54 3.2.4 Summary 56 3.3 Controlling Thickness of Spin-on Mesoporous Films 58 3.3.1 Introduction 58 3.3.2 Experimental 59 3.3.3 Results 60 3.3.4 Discussions 65 3.3.4.1 Relationship between thicknesses and spin speeds 65 3.3.4.2 Relationship between thicknesses and solute concentrations 66 3.3.4.3 Radial uniformity 67 3.3.5 Conclusions 69 Chapter 4 Effects of Spin-Coating on Mesoporous Silica Films from Different templates 71 4.1 Introduction 71 4.2 Experimental 75 4.2.1 Sample preparation 75 4.2.2 Characterization 77 4.3 Results and Discussions 78 4.3.1 Uniformity of the films 78 4.3.2 Effects of spin-coating and calcination on pore structures of spin-on films made with different templates 80 4.3.2.1 C16TMABr as the templates 80 4.3.2.2 Brij-56 as the templates 84 4.3.2.3 Tween80 as the templates 88 4.3.2.4 P123 as the templates 94 4.3.2.5 PEG1450 as the templates 99 4.3.2.6 Discussions 102 4.4 Conclusions 106 Chapter 5 Correlations between Film Architecture and Film Properties 109 5.1 Introduction 109 5.2 Experimental 112 5.2.1 Sample preparation 112 5.2.2 Characterization 112 5.3 Results and Discussions 115 5.3.1 Refractive index and porosity 116 5.3.2 Dielectric constant and porosity 119 5.3.3 Refractive index and dielectric constant 123 5.3.4 Flat band voltage 125 5.3.5 Leakage current 128 5.3.6 Mechanical properties 132 5.4 Conclusions 137 Chapter 6 Mesoporous Silica Films as Low-k Materials 139 6.1 Introduction 139 6.1.1 Why low-k materials are needed 139 6.1.2 Relative permittivity (or dielectric constant) 140 6.1.3 Porous silica-based low-k materials 143 6.1.4 Mesoporous silica low-k materials by a surfactant- templated method 145 6.2 Experimental 148 6.3 Results and Discussions 149 6.3.1 Effects of surfactant amount in the coating solution 149 6.3.2 Effects of water amount in the coating solution 153 6.3.3 Effects of mixing time for coating solution 155 6.3.4 Effects of acid amount in the coating solution 157 6.4 Conclusions 159 Chapter 7 Mesoporous Silica Films as Anti-Reflection Films 161 7.1 Introduction 161 7.2 Theory and Design of Anti-Reflection Coatings 164 7.3 Development of AR Coating by a Surfactant- Templated Process 172 7.3.1 Introduction 172 7.3.2 Experimental 172 7.3.2.1 Sample preparation 172 7.3.2.2 Characterization 173 7.3.3 Results and Discussions 174 7.3.3.1 Controlling Refractive Indices and Film Thicknesses 174 7.3.3.2 Transmittance Enhancement 180 7.3.4 Conclusions 184 Chapter 8 Conclusions 185 Chapter 9 Suggestions and Future Studies 189 Acknowledgements 193 References 195 Appendix 209 A. Determine Film Capacitance and Flat Band Voltage 209 B. Leakage Current Densities on Electrodes with Different Areas 211 C. Comparison of Film Thicknesses Measured by SEM and n&k Analyzer 212 D. Measuring Mechanical Properties 2131851357 bytesapplication/pdfen-US低介電膜旋轉塗佈抗反射膜模版薄膜孔洞材料介孔中孔洞界面活性劑模版法spin coatingtemplatemesoporousporous materialthin filmlow-k filmanti-reflection filmsurfactant-templated methodlow dielectric constant filmthesishttp://ntur.lib.ntu.edu.tw/bitstream/246246/52254/1/ntu-93-D89524017-1.pdf