|UFDC Home||myUFDC Home | Help|
This item has the following downloads:
1 DEVELOPMENT OF A SCALABLE BOTTOM UP NANOFABRICATION PLATFORM By CHIH HUNG SUN A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2009
2 2009 Chih Hung Sun
3 To my family in Taiwan, who, always support all my efforts especially for my PhD (To my Mom is a typical dedication)
4 ACKNOWLEDGMENTS Foremost, I would like to express my sincere gratitude to my advisor Pro fessor Peng Jiang for the continuous support of my Ph D study and research, for his patience, motivation, enthusiasm, and immense knowledge. His guidance helped me in all the time of resea rch and writing of this thesis. I also thank my committee members, Professor Kirk Ziegler Professor Oscar Crisalle and Professor Huiaki Xie for thei r contribution to my proposal presentation and dissertation defense. I thank our group members Nicholas Linn, Wei Lun Min Hongt a Yang, In Kook Jun Tzung Hua Lin and Wei Han Huang visiting professors Dr. Xuefeng Liu at Jiangnan University and Dr. Satos hi Watanabe at Kyoto University for giving me lo ts of suggestions and assisting to my research. I was very fortunate to be a member of this group. Finally, and most importantly, I express my thanks and gratitude to my parents and friends in Taiwan, who hav e always supported m e in spirit along this journey. My parents and friends gave me the courage to pursue my dreams, and encouraged me to be a successful researcher with an aggressive attitude in life and work. I am always going to be grateful for having them by my side.
5 TABLE OF CONTENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ ...... 4 LIST OF TABLES ................................ ................................ ................................ ................ 7 LIST OF FIGURES ................................ ................................ ................................ .............. 8 CHAPTER 1 INTRODUCTION ................................ ................................ ................................ ........ 13 1.1 Top Down and Bottom Up Fabrication ................................ ................................ 13 1.2 Colloidal Self Assembly ................................ ................................ ........................ 14 1.3 Antireflection Coatings ................................ ................................ ......................... 15 1.4 Surface Plasmons and Surface Enhanced Raman Scattering ........................... 15 2 SELF ASSEMBLY OF COLLOIDAL CRYSTALS BY SPIN COATING .................... 20 2.1 Experimental Procedure ................................ ................................ ....................... 20 2.2 Results and Discussion ................................ ................................ ........................ 21 3 TEMPLATED FABRICATION OF ANTIREFLECTION COATINGS ......................... 28 3.1 Templated Antireflection Coatings on Glass Sub strates ................................ ..... 28 3.1.1 Templated Moth Eye Antireflection Coatings ................................ ............ 28 3.1.2 Templated Fabrication of Binary Arrays as Antireflection Coatings .......... 31 3.2 Templating Antireflection Coatings on Semiconductor ................................ ....... 37 3.2.1 Surface Texturing Silicon Wafers as Antireflection Coatings .................... 37 3.2.2 Broadband Antireflection Coatings on Silicon Wafer ................................ 42 3.2.3 Biomimetic Subwavelength Antireflective Gratings on GaAs .................... 46 4 TEMPLATED FABRICATION OF SERS ACTIVE SUBSTRATES ........................... 62 4.1 Templated Fabrication of Periodic Metallic Nanopyramid Arrays ........................ 62 4.2.1 Experimental Procedure ................................ ................................ ............. 64 4.2.2 Results and Discussion ................................ ................................ .............. 65 4.2 Wafer Scale SERS Substrates with Highly Reproducible Enhancement ........... 71 4.2.1 Experimental Procedure ................................ ................................ ............. 71 4. 2.2 Results and Discussion ................................ ................................ .............. 73 5 TEMPLATED FABRICATION OF SUB100NM PERIODIC NANOSTRUCTURES .. 90 5.1 Experimental Procedure ................................ ................................ ....................... 91 5.2 Results and Discussion ................................ ................................ ........................ 94
6 6 CONCLUSIONS AND RECOMMENDATIONS ................................ ......................... 98 6.1 Antireflection Coatings ................................ ................................ ......................... 98 6.2 SERS Active Substrates ................................ ................................ ...................... 98 6.3 Sub 100nm Nanostructures ................................ ................................ ................. 99 6.4 Recommendations ................................ ................................ ................................ 99 LIST OF REFERENSCES ................................ ................................ ............................... 101 BIOGRAPHICAL SKETCH ................................ ................................ .............................. 111
7 LIST OF TABLES Table page 4 1 Assignment of SERS p eaks and c orresponding a verage c ounts ......................... 86 4 2 Assignment of SERS p eaks and R aman e nhancement f actors ........................... 87
8 LIST OF FIGURES Figure page 1 1 A flowchart describes the self assembly approach. ................................ .............. 18 1 2 The dispersion curve for a SP mode ................................ ................................ .... 18 1 3 Schematic diagram describes Raman Scattering. ................................ ................ 19 1 4 A typical energy level diagram for a molecule adsorbed on a metal surface. ...... 19 2 1 Spin coated silica polymer nanocomposites with long range ordering. ............... 25 2 2 Precise control over the nanocomposite thickness by spin coating ................... 25 2 3 Non close packed colloidal crystals after removing polymer matrix. ................... 26 2 4 Spin coated monolayer non close packed col loidal crystal with metastable square lattice. ................................ ................................ ................................ ......... 26 2 5 Interparticle spacing of spin coated silica/ETPTA dispersion for different volume fraction. ................................ ................................ ................................ ...... 27 2 6 SEM images of spin coated silica monomer dispersion with different volume fraction ................................ ................................ ................................ ................... 27 3 1 Schematic illustration of the templating procedure from making moth eye antireflection coatings. ................................ ................................ ........................... 51 3 2 AFM images and corresponding depth profiles of replicated nipple arrays at different RIE etching times. ................................ ................................ .................... 52 3 3 Experimental and calculated specular optical reflectivity at normal incidence. ... 52 3 4 Dependence of the normal incidence optical reflectivity at 600nm vs. RIE etching time. ................................ ................................ ................................ ........... 53 3 5 Comparison of the change of calculated effective refractive index. .................... 53 3 6 SEM images illustrating the progressive fabrications procedures for binary dimple nipple antireflection coatings. ................................ ................................ .... 54 3 7 Experimental (solid) and RCWA simulated (dotted) normal incidence specular reflection. ................................ ................................ ................................ 54 3 8 Sche matic illustration of the templating procedures for fabricating subwavelength structured antireflection pyramid gratings. ................................ .. 55
9 3 9 Re plicated inverted pyramids arrays in silicon. ................................ ..................... 55 3 10 Experimental and RCWA modeled specular optical reflectivity at normal incidence ................................ ................................ ................................ ............. 56 3 11 Comparison of the change of calculated effective refractive index at =600 nm. ................................ ................................ ................................ .......................... 56 3 12 RCWA simulated normal incidence optical reflection at =600 nm vs. inverted pyramid size. The 320 nm silica spheres are used as templates. ......... 57 3 13 Schematic illustration of the templating procedures for fabrication moth eye ARCs on silicon substrates. ................................ ................................ ................... 57 3 14 SEM images of biomimetic antireflection coatings on silicon substrates. ............ 58 3 15 Experimental (solid) and RCWA simulated (dotted) specular optical reflection at normal incidence ................................ ................................ ............................... 58 3 16 Comparison between the calculated optical reflection at normal incidence by the RCWA model (red) and the thin film multilayer model (blue) ........................ 59 3 17 SEM images of silica nanoparticles on GaAs. ................................ ...................... 59 3 18 SEM images of templated nanacones on GaAs substrates. .............................. 60 3 19 Experimental and RCWA simulated specular reflection at normal incidence. ..... 60 3 20 Change of the calculated effective refractive index at =700 nm. ........................ 61 4 1 Schematic illustration of the templating procedures for making periodic metallic nanopyramid arrays. ................................ ................................ ................. 81 4 2 Photograph of a gold coated colloidal crystal polymer nanocomposite. .............. 82 4 3 Inverted pyramid arrays anisotropically etched in (100) silicon wafer. ................. 82 4 4 SEM image of particle precipitation on KOH etched silicon. ................................ 83 4 5 Periodic gold nanopyramid array. ................................ ................................ ......... 83 4 6 Raman spectra of benzenethiol adsorbed on a gold nanopyramid arrays. ......... 84 4 7 Photograph of a gold coated colloidal crystal polymer nanocomposite consisting of 320 nm silica s pheres on a 4 in. silicon wafer ................................ 84 4 8 Sc anning electro n microscope images of the gold coated nanocomposite. ........ 85
10 4 9 Tapping mo de atomic force microscope images and corresponding depth profiles. ................................ ................................ ................................ ................... 85 4 10 SER spectra obtained on a gold coated nanocomposite consisting of 400 nm silica spheres and a flat gold control sample on glass ................................ ......... 86 4 11 Arithmetically averaged SER spectra recorded for benzenethiol molecules adsorbed on five areas (R0 R4) of 4 in. nanocomposites. ................................ ... 87 4 12 Normal incidence reflection spectra obtained at eight locations on a 4 in. metalized nanocomposite consisting of 400nm silica spheres. ............................ 88 4 13 Simulated Raman enhancement around gold semispherical protrusions ................................ ............ 89 5 1 Schematic illustration of the fabrication procedures for making periodic arrays of metal nanoholes with tunable size. ................................ ................................ ... 95 5 2 Templated polymer posts with adjustable size. ................................ .................... 96 5 3 SEM image of an array of gold nanoholes with 89 nm diameter templated from a monolayer nanocomposite consisting of 360 nm silica spheres. .............. 96 5 4 SEM images of micro candles. ................................ ................................ .............. 97
11 Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy DEVELOPMENT OF A SCALABLE BOTTOM UP NANOFABRICATION PLATFORM IN ALL CAPITAL LETTERS, EXACTLY AS ON TITLE PAGE By Chih Hung Sun December 2009 Chair: Peng Jiang Major: Chemical Engineering This study covers three important research topics related to the application of self assembly colloidal crystal by spin coating. The first topic describes the antireflection coatings on glass and semiconductor mater ials. Inspir ed by nature, the protruding nipples on moth eye can effectively reduce light reflection. Creating sub 300nm features with c urrent lithography based fabrication techniques ( e.g., photolithography or interference lithography ) i s costly and is limited by either low resolution or small sample size The spin coating technology that combines the simplicity and cost b enefit of bottom up self assembly with the scalability and compatibility of standard top down microfabrication is used in creating non close packed moth eye antireflection coatings The second topic describes the templating fabrication of surface enhanced Raman scattering (SERS) active substrates. Wafer scale metallic pyramid arrays with sharp nanoti ps can be replicated from silicon templates by a simple adhesive peeling process. The resulting nanopyramid arrays show stable surface enhance ments for Raman scattering from benzenethiol molecules adsorbed at the metal sur faces. To further improve the reproducibility, we report a simple colloidal templating approach for fab ricating surface enhanced Rama n scatterin g (SERS) substrates with highly
12 reproducible enha ncement over wafer sized areas. The substrates are produce d by evaporating a thin layer of gold on highly ordered silica colloidal crystal polymer nanocomposites created by a scalable spin coating technology. The ordered inorganic organic composite structu re induces the formation of island type gold films. Systematic measur ements and statistical analysis show that the substrates exhibit high SERS reproducibility with less than 28% standard deviation over a 4 in. wafer surface. Finite eleme nt electromagnetic modeling has also been employed to simulate SERS enhancement from these structured substrates. The third topic presented is the templated fabrication of sub 100nm nanostructures. Periodic polymer nanoposts and metal nanohole arrays with tunable size have been fabricated by templating from spin coated two dimensional non close packed colloidal crystal polymer nanocomposites.
13 CHAPTER 1 INTRODUCTION 1.1 Top Down and Bottom Up Fabrication Top down fabrication ha s been widely used in the semiconductor industry to fabricate cheap a nd fast electronic devices which are the hallmark of modern life and technology. Photolithography is the most commonly used top down technology to fabricate three dimensional components. Due to the light diffraction limit, feature sizes smaller than the wavelength of light cannot be achieved by conventional photolithography. 1 L ithography techniques such as e xtreme ultraviolet light or soft X rays lithography, electron beam lithography, and dip pen litho graphy have been used to generate smaller patterns. 2 These high cost and low throughput techniques cannot be used for m ass production and are only lab oriented processes. Another challenge of top down fabrication is the choice of materials. The use of short wavelength as the exposure light source and smaller features size requires materials with suitable optical and mechan ical properties. Aside from top down engineering technique, bo ttom up nanofabrication offers a new paradigm for material fabrication. Complex structures found in nature are assembled with building blocks like atoms, protein, RNA and DNA. The self assembl y building blocks can be made by organic, inorganic, polymeric and hybrid materials. Building blocks can be assembled into spheres and cubes, sheets and discs, wires and tubes, rings and spiral s with nm to cm dimensions. A flowchart describing the ideas of self as sembly is shown in F igure 1 1.
14 1.2 Colloidal Self Assembly The ability to synthesize monodisperse latex and silica nanoparticles by heterogeneous emulsion polymerizations and sol gel chemistry, respectively with diameters in the range of tens of nanometer to several micrometers provides a new class of materials with a range of application s in chemistry and physics. A variety of methods, such as gravit ational sedimentation 3 4 template assisted as sembly, 5 7 ele ctrostatic repulsion 8 10 electric field induced assembly, 11 12 robotic manipulation 13 L angmuir Blodgett deposition 14 spin co ating and capillary forces i nduced convective self assembly 15 17 have been developed to create both 3D and 2D colloidal crystals. The application of the colloidal crystal has been widely exploited recently. One of the most appealing properties of these periodic dielectric lattices is 3D photonic crystals behavior. 18 20 Colloidal crys tals can function as the optical analog of the electronic semiconductor. The 3D colloidal crystal can also be used as a template for preparation of three dimensionally ordered porous replicas, which are known as inverse opals A wide variety of macroporou s materials based on colloidal template s have been synthesized. The applications of these inverse opals include chemical and biological sensors 21 23 simultaneous chromatographic sepa ration, 24 optical sensing, power storage and generation 25 low threshold laser, 26 and full color displays. 27 Monolayer colloidal crystal can be used for lithographic patterning of surfaces, projection lithography, micromolds and surface relief patterns 28 Nanostrucures such as nanodots, nanorings, na no prisms, nanoholes, nanobowls, nanocrescent and hollow spheres nanostructures can be fabricated by two dimensional lithography. Applications include optical antennas, catalysts, chemical and biological sensors. 29 32 The sizes, periodicities, and materials of these nanostructures can be controlled by adjusting the
15 size of the colloidal templates and the filling materials from organic, inorganic to polymeric mat erials. 1.3 Antireflection Coatings Antireflection coatings (ARCs) are used in optical elements to reduce light reflection and enhance optical performance. A reduction in surface reflection is typically accomplished by the application of an antireflection coating of a quarter wavelength optical thickness. When the index of refraction of the coating ( ) equals the square root of the product of the indices of the surrounding medium ( 0 ) and the substrate ( ), that is, = 0 0 5 reflections are s uppressed at wavelengths near the quarter wavelength optical thickness. However, there are some limitations and problems when applying quarter wavelength optical thickness as antireflection coatings. To achieve the effective antireflection coatings for gl ass, the refractive index must equal 1.2 while material s with the lowest r efractive indices available are ~1.35. 33 As for semiconductors, a single layer film can only reduce light reflection at a single wavelength. A stack of alternate high and low refractive index films, all one quarter wavelength thick is used to achieve broadband antireflection coatings. To achieve broadband antireflections requires very precise control of the thickness of each layer. Additionally, the potential mismatch of physical properties (such as thermal expansion coefficient) of each layer reduces the stability of the coating layer s. By colloidal crystal templating large area and cost effective broadband an tireflection nanostructures can be generated. 1.4 Surface Plasmons and Surface Enhance d Raman Scattering Surface plasmons are electromagnetic waves propagating at the interface between a dielectric and a conductor, evanescently confined in the perpendicular
16 direction. These electromagnetic surface waves arise via the coupling of the electromagnetic fields to oscillations of the con 34 The surface pla smon s F igure 1 2. In order to couple light to surface plasmons, the momentum mismatch must be overcome. A prism is typically used to excite surface plasmons on flat metal surfaces Periodic metal nanostructures are also used to generate surface plasmons. 35 Recent a dvances in nanotechnology allow metal to be struc tured to tailor surface plasmon properties. Applications of surface plasmons have been exploited in optics 36 solar cells, 37 biological and chemical sensors 35,38 39 and surface enhanced Raman scattering. 40 Raman spectroscopy is a spectroscopic technique used to study vibrational and other low frequen cy modes in a system. It relies on inelastic scattering of monochromatic light, usually from a laser in the visible, near infrared, or near ultraviolet range. The incident photons are scattered inelastically by a molecul ar bond and shifted in frequency by the energy of its characteristic molecular vibrations. Two types of Raman scattering events, called Stokes and anti Stokes scattering may result, depend ing up on whether the incident light interact s with a molecule in the vibrational ground state or an excited vibrational state as shown in F igure 1 3 Spontaneous Raman scattering is typically very weak, and as a result the main difficulty of Raman s pectroscopy is separating the weak inelastically scat tered light from the intense Rayleigh scattered laser light. Surface enhancement Raman scattering was first discovered by Fleischmann in 1974 but the prevailing explanation for the nature of the
17 single enhancement was proposed by V an Duyne in 1977. 41 The mechanisms of SERS are sorted into two cl asses called chemical and electromagnetic enhancement. 42 The chemical enhancement can be explained by the new electronic states which arise from molecular metal ch emisorptions Modified electronic states serve as resona nt intermediate states in Raman scattering. When the highest occupied molecular orbital (HOMO) and the lowest unoccupied molecular orbital (LUMO) of the adsorbate symmetrically fall about the Fermi le vel of the metal surface, charge transfer excitations can occur at about half the energy of the intrinsic intramolecular excitations of the adsorbate as shown in Figure 1 4 42 The electromagnetic field of the light at the adsorption surface can be greatly enhanced if surface plasmons are excited The amplification of both incident laser field and scattered Raman field through their interaction with the surface consti tute the electromagnetic SERS mechanism. The electromagnetic contribution to the Raman signals enhancement is the fourth power of the local electric field enhancement. The most common types of SERS active substrates are clusters of colloidal si lver or gol d nanoparticles in solution or dry on a surface. 43 44 A lthough metal colloid s exhibit the largest enhancement factor, the reproducibility of this type of SERS substrate remai ns a problem. The interparticle spacing which generates hot spots for SERS is difficult to control due to the stochastic distribution of the nanospheres on a substrate. By colloidal crystal templating fabrication, large area and reproducible SERS active substrate s can be easily fabricated.
18 Figure 1 1. A flowchart describ es the self assembly approach. Figure 1 2 The dispersion curve for a SP mode shows the momentum mismatch problem that must be overcome in order to couple light and SP modes to gether, with the SP mode always lying beyond the light line, that is, it has greater momentum ( k SP ) than a free space photon ( k 0 ) of the same frequency
19 Figure 1 3. Schematic diagram describes Raman Scattering. Figure 1 4. A t ypical energy level diagram for a molecule adsorbed on a metal surface. The occupied and unoccupied molecular orbitals are broadened into resonances by their interaction with the metal states; orbital occupancy is determined by the Fermi energy. Possible charge transfer excitations are shown
20 CHAPTER 2 SELF ASSEMBLY O F COLLOIDAL CRYSTAL S BY SPIN COATING Spin coating is a well established technique in microfabrication for the formation of highly uniform thin films with adjustable thickness over large area s 45 It has already been utilized in making colloidal crystals, such as colloidal masks for nanosphere lithography 46 47 However, traditional spin coating methods utilize volatile solvents (e.g. water and alcohol) to disperse colloidal particles. The rapid evaporation of solvents during spin coating processes leads to polycrystalline sam ples with poor qualities 48 By eliminating solvent evaporation, particles would have sufficient time to be organized into energetically favorable states during spin coating. After colloidal self assembly, the dispersing solvents need to be removed to release aligned colloidal crystals. The removal process should be gentle enough that does not affect the ordering of the colloidal arrays. Dry etching is a good candidate as it does not involve any wet processing that might torture the colloidal crystals with strong capillary force s. To satisfy the vacuum requirements of dry etching processes, the solvents need to be solidified before etching. Considering all these requirements, non volatile monomers stand out from other possible solvents as they can be easily polymerized to trap th e assembled colloidal arrays, and can subsequently be removed by standard plasma etching thereby releas ing the colloidal crystals. 2 .1 Experimental Procedure A volume fraction of ca. 20% s ilica colloids with diameters of ~ 30 nm to ~ 2 m are dispersed in ethoxylated trimethylolpropane triacrylate ( ETPTA ) monomer to form tran sparent and stable suspensions. In a typical spin coating experiment, the colloidal suspension is dispensed on a substrate and then spin coated with a standard spin
21 coater. The substrate is accelerated in 2 seconds to 200 rpm and kept at this speed for 2 3 minutes, then accelerated to a desired speed at a typical acceleration rate of 200rpm/s with continue d spin ning for a specific time needed to achieve a target thickness T he substrate is then transferred to a high intensity UV lamp to polymerize the monomer. 2 2 Results and Discussion The self assembled silica colloidal crystal polymer nanocomposite films exhibit bright monoch romatic six arm diffraction (Figure 2 1 ), indicating the formation of hexagonally packed spheres parallel to the wafer surface. 8,49 50 The long range hexagonal ordering is confirmed by top view scanning electron microscope (SEM) images of two samples consisting of 325 nm (Fig ure 2 1B and C ) and 1320 nm (Fig ure 2 1 D ) silica sphe res. The ordering perpendicular to the substrate surface is apparent in the cross sectional images shown in Fig ure 2 2 Most interesting ly the spin coated colloidal crystals exhibit unusual non close packed (ncp) structures as revealed by magnified SEM im a ges (Figure 2 2 C and D). This is more apparent after the polymer matrix been selectively removed by oxygen plasma etching (Fig ure 2 3A and C). For multilayer samples, spheres of the top layer only fill in the triangularly arranged crevices made by the non touching spheres of the second layer (Fig ure 2 3 C ). The center to center separation between adjacent spheres for all sa mples assembled using different size particles and with different thickness is determined to be ~ 1.41D, where D is the diameter of sil ica spheres, by the first peak of the pair correlation function (PCF, (Fig ure 2 3 B ) ) calculated from SEM images similar to Fig ure 2 3 A and C The spin coated nanocomposite films exhibit excellent thickness uniformity with variation of less th a n 4% over a four inch diameter wafer. The film thickness can be
22 controlled easily by changing the spin speed and time. It is inversely proportional to the final spin speed and the square root of the final spin time. Fig ure 2 2 shows cross sectional images of four c rystals of monolayer, 2 layers, 5 layers, and 41 layers made at different spin coating conditions. Even thicker samples can be assembled by multiple coatings. New colloidal multilayer s can be spin coated on the top of the original ones and the process can be repeated many times with the thickness increasing linearly after each coating. The modulated top surface of the pre c eding layer (Fig ure 2 2 C) functions as a template which aids in aligning the crystalline orientation of the subsequent multilayer. As dis cussed above, the polymer matrix need s to be removed to release the embedded silica colloidal crystals that can be used as templates in creating inverted photonic crystals with high refractive index contrast. Standard oxygen plasma etching is a better method than calcination in removing ETPTA polymer matrix, as it hardly affects the silica spheres and no def ects, such as cracks, are intro duced. Fig ure s 2 3 A, C and D show top and side view SEM images of monolayer and multilayer colloidal crystals after polymer matrix is removed l. The preservation of the hexagonal long range ordering and the center to center separation of the original nanocomposites throughout the plasma etching process are clearly evident. We notice a significant difference in the result ing crystalline quality between monolayer and multilayer colloidal crystals prepared by the sam e spin coating process The typical domain size of monolayer samples is only several hundred microns, much smaller tha n that of multilayer samples (~ cm). When t he spin coat ing speed is low (6000 rpm) only six arm diffraction patterns with exact 60 angles between neighboring arms are observed indicating the formation
23 of hexagonally ordered colloidal crystals. 8 Unexpected results occur when the spi n speed is higher than 6000 rpm. The alternat ing formation of hexagonal and square diffraction patterns is observed when the thickness of the colloidal crystals is gradually reduced during spin coating. The spin coating process is stopped once a strong four arm diffraction pattern is formed on the wafer surface and the ETPTA monomers are then photopolymerized by exposure to UV radiation. The solidified ETPTA matrix can finally be removed by oxygen plasma etching to release the embedded colloidal arrays. Figure 2 4 A shows a photograph of a 4 in. diameter colloidal monolayer sample made from 380 nm silica spheres and spin coated at 8000 rpm for 150 s. The sample exhibits a distinctive four arm diffraction pattern under white light illumination, and the angl es between the neighboring diffraction arms are 90. This pattern is a characteristic of long range square ordering. This is confirmed by the SEM image in Fig ure 2 4 B and is further evidenced by the squarely arranged peaks in the Fourier transform of a low magnification SEM image, as sh own in the inset of Figure 2 4 B Further SEM characterization shows that the squarely arranged arrays cover the whole waf er surface and the crystal is polycrystalline with typical domain size of several tens of micrometers. T o obtain colloidal monolayer with larger single crystalline domains, the laye r by layer thinning technology can be employed to gradually reduce the thickness of the spin coated multilayer crystals, which usually exhibit better ordering than shear aligned m onolayer arrays. The center to center nanoparticle distance can be controlled by adjusting the volume fractio n of silica monomer dispersion as shown in Figure 2 5. As the volume fraction of the silica spheres increase, interparticle spacing decreases. Figu re 2 6
24 shows the SEM images of top and side view of spin coated colloidal crystal with volume fraction s of 30% and 40%. Although increas ing volume fraction can reduce interparticle spacing, it also in creases the viscosity of silica ETPTA dispersion. For 50% volume fraction, the viscosity increases drastically and thus the silica ETPTA composite film is not uniform. We attribute the unusual formation of non close packed colloidal crystals during spin coating to the normal pressures produced by shear flows and monomer polymerization that squeeze shear aligned hexagonal layers of particles against substrates. Considering the low dielectric constant of the monomer (~ 3.0 at optical frequencies) and the small salt effect on the interparticle separation, t he electrostatic repulsion between charged spheres plays only a minor role in determining the resulting microstructures. The shear forces created by spin coating are crucial for aligning colloidal particles into hexagonally ordered crystals. The interactio ns of applied hydrodynamic, Brownian, and the colloidal forces determine the resulting microstructures. In sharp contrast to traditional rotational rocking cuvette and parallel plate shear cells, the flow profile in the spin coating process is not a unifor m shear, with the shear maximum occurring at the substrate and rapidly decaying to a zero value at the free surface. Although shear aligned colloidal crystallization has been extensi vely studied 51 54 the effect of non uniform shear on the formation of aligned microstructures has received little or no attentio n. Therefore, a detailed study on the underlying mechanisms of colloidal crystallization during spin coating, which has yet to be fully understood, will provide new insights into shear induced crystallization, melting and relaxation
25 Figure 2 1. Spin coated silica polyme r nanocomposites with long range ordering. A) Photograph of a 4 in. sample consisting of 325nm silica spheres illuminated with white light. B) Top view SEM image and its Fourier transform (insert) of the sample in A. C) Magnified SEM image of B. D) A sample made from 1320nm silica spheres. Figure 2 2. Precise control over the nanocomposite thickness by spin coating. A) Monolayer, B) 2 layer s C) 5 layers and D) 41 layer s A B C D A B C D
26 Figure 2 3. Non close packed colloidal crystals after removing polymer matrix. A) Top view SEM image of a monolayer sample. B) PCF calculated from a low magnification SEM image. C) Top view SEM image of a multilayer sample. D) Cross sectional SEM ima ge. Figure 2 4. Spin coated monolayer non close packed colloidal crystal with metastable square lattice. A) Photograph of a 4 in. sample illuminated with white light. B) SEM image of the sample in A. A B C D A B
27 Figure 2 5 Interparticle spacing of spin coat ed silica/ETPTA dispersion for different volume fraction. Figure 2 6 SEM images of spin coated silica monomer dispersion with different volume fraction. A) Top view SEM image of 30% volume fraction, B) side view of A. C) 40% volume, D) side view of C A B C D Interparticle Spacing (nm) Volume Fraction (%)
28 CHAPTER 3 TEMPLATED FABRICATIO N OF ANTIREFLECTION COATINGS 3.1 Templa ted Antireflection Coatings on Glass Substrates 3.1.1 Templated Moth Eye Antireflection Coatings Periodic optical microstructures are abundant in biological systems and have provided enormous inspiration for scientists to mimic natural structures for practical applications. 55 59 To name just a few, Morpho butterflies use multiple layers of cuticle and air as natural photonic crystals to produce striking blue color, 56 inspiring the development of chemical sensors for detecting trace amount of vapors. 58 Some nocturnal insects e.g., moths use arrays of non close packed nipples with sub 300 nm s ize as antireflection coatings (ARCs) to reduce reflectivity from their compound eyes. 60 Artificial ARCs are widely used in monitors, car dashboards, optical components, and solar cells. 33,61 63 Existing antireflection technologies, suc h as quarter wavelength multilayer films and nanoporous coatings e.g., phase separated polymers and nanoparticle a nd polyelectrolyte multilayers often perform suboptimally or are expensive to implement. 33,61 64 Inspired by the natural photonic structures, moth eye ARCs with subwavelength protrusion arrays have been widely explored. 60,65 However, c urrent lithography based fabrication techniques ( e.g., photolithography or interference lithography ) in creating sub 300 nm features are costly and are limited by either low resolution or small sample size. Self assembly in synthetic materials provides an inexpensive, simple to implement, inherently parallel, and high throughput alternative to lithography in creating periodic microstructures. 28 Unfortunately, most of the traditional self assembly techniques are not compatible with standard microfabrication, impeding scale up to an industrial scale
29 fabrication. 64,66 67 Additionally, conventional self assembly is limited to the creation of close packed structures, whe reas natural moth eye ARCs exhibit non close packing characteristics. 60 The spin coating technology that combines the simplicity and cost benefits of bottom up self assembly with the scalability and compatibility of standard to p down microfabrication is used in creating non close packed moth eye antireflection coatings. Experimental proc e dure A schematic outline of the templating procedures is shown in Fig ure 3 1. The fabrication of wafer scale, non close packed colloidal crystals embedded in poly(ethoxylated trimethylolpropane triacrylate) (PETPTA) matrix is performed by spin coating technology The resulting crystals can be easily and reproducibly created over arbitrarily large areas in minutes. The long range periodic surface protrusions of the shear aligned crystals similar to Fig ure 3 2 can be easily transferred to a poly( dimethylsiloxane) (PDMS Sylgard 184, Dow Corning) mold. The solidified PDMS mold can then be peeled off and put on top of ETPTA monomer supported by a glass slide with spacers (double stick tape, thickness of ~0.1 mm) in between. ETPTA monomer is polymerized for 2 s using a Xenon pulsed UV curing system. After peeling off PDMS mold, ETPTA nipple arrays coated glass slide can be ma de. The flexible PDMS mold enables the creation of microstructured coatings on both planar and curved substrates. The protrusion depth (~50 nm) of a replicated PETPTA array from the spin coated nanocomposite film is shallow compared to the radius of templa ting silica spheres(180 nm), as revealed by the atomic force microscope AFM image (Fig ure 3 2 A ), and its corresponding depth profile (bottom curve) is shown in Fig ure 3 2 D. The polymer matrix
30 of spin coated nanocomposites can be plasma etched using an Una xis Shuttlelock reactive ion etcher RIE inductively coupled plasma ICP operating at 40 mTorr oxygen pressure, 40 SCCM (SCCM denotes cubic centimeter per minute at STP flow rate), and 100 W to adjust the height of the protruded portions of silica spheres, r esulting in good control over the depth of replicated nipples. Figure 3 2 B and C show PETPTA nipple arrays replicated from the same nanocomposite sample, as shown in Fig ure 3 2A after 20s and 45 s RIE etching, respectively. The shape of nipples in the lat ter sample is close to hemispherical, as revealed by the depth profile (top curve in Fig ure 3 2 D.) The specular optical reflectivity of the replicated nipple arrays are evaluated using visible near IR reflectivity measu rement at normal incidence. An Ocean Optics HR4000 high resolution fiber optic UV visible near IR spectrometer with a reflection probe is used for reflectance measurements. 68 The resulting reflectivity is calibrated using a STAN SSL low reflectivity specular reflectance standard Ocean Optics. Figures 3 3 and Figure 3 4 illustrate the excellent control over the antireflection performance of the replicated nipple arrays by simply adjusting the RIE etching time (i.e., height of nipples). The hemispherical like nipple arrays (dark gray curve in Fig ure 3 3 ) sho w significantly smaller reflectivity than that of a flat control sample (black curve) for the whole visible spectrum. Results and discussion A thin film multilayer model has also been developed arrays and then compared with experimental spectra. 60,69 The nipple lattice is assumed to be hexagonal and the distance between the center s of the neighboring nipples is 2 D 70 Since the distance of the ni pples is small with respect to the wavelength of light, light propagation is governed by the effective refractive index of the nipple array,
31 which can be calculated from effective medium theory. 69 We first divide th e whole hemisphere cap layer in 100 layers and calculate the effective refractive index for each layer. The reflectance of each layer can then be calculated using a m atrix multiplication procedure for a stack of thin layers 69 Figure 3 3 shows that the simulated spectra (dotted lines) match the experimental spectra (solid line) and the reflection from the PETPTA/glass interface due to index mismatch might contribute to the small discrepancy between them. The good and adjustable antireflection performance of the replicated nipple arrays can be explained by mapping the effective refractive index ac ross the height of nipples (Figure 3 5 ). For featureless polymer film, the refractive index is sharply changed from 1.0 to 1.46 across the air polymer interface, leading to high reflectance. For nipp les with 50 nm height cross that are replicated directly from spin coated nanocomposite without RIE etching, the effective refractive index is continuously changed from 1.0 nipple peaks to 1.09 nipple troughs, then sharply increased to 1.46, resulti ng in r educed reflectivity ( Figure 3 4 ) ; while for hemispherical nipples (unfilled circle), the final step is moderate from 1.20 to 1.46, thus leading to minimal reflectivity. This simple yet scalable self assembly approach for fabricating efficient moth eye ant ireflection coatings with adjustable reflectivity and non close packed microstructures, which are not easily available by traditional self assembly approaches. The specular reflection of nipple arrays matches the theoretical prediction using a thin film multilayer model. 3 .1. 2 Templated Fabrication of Binary Arrays as Antireflection Coatings Facile production of complex periodic nanostructures over a large area is of great technological importance in developing novel electronic, optical, and magnetic
32 devices. 71 73 For instance, subwavelength structured metal nanohole arrays exhibit extraordinary optical transmission, which is promising for realizing miniaturized devices for all optical integrated circuits. 36,74 75 Other than simple hexagonal and square arrays, complex nanohole geometries (e.g., binary arrays) have attracted great scientific and technological interest as they are promising structures for developing more efficient optical devices and biosensors, 76 77 as well as a fundamental understanding of surface plasmon enabled optical transmission. 78 79 Dielectric binary periodic nanostructures have also been explored for creating superhydrophobic and superoleophobic surfaces, 80 81 photonic crystals, 82 and antireflection c oatings. 83 However, creating periodic binary nanostructu res with subwavelength scale resolution is not a trivial task for the current nanofabrication techniques, such as electron beam lithography and focused ion beam. Although arbitrary nanostructures can be fabricated, attaining high throughput and large area fabrication in an inexpensive manner continues to be a major challenge with these top down techniques. 84 Bott om up self assembly and subsequent templating nanofabrication provides a much simpler and inexpensive alternative to nanolithography in fabricating periodic nanostructures. 28,46 47,85 86 Self assembled block copolymers and surfactants enable the creation of sub 50 nm periodic structures. 87 88 Colloidal lithography provides a versatile approach for making a large variety of periodic microstructures by using self assembled mono layer or multilayer colloidal crystals as deposition or etching masks. 46 47,85 86 Binary periodic nanostructures, such as metal nanopart icle arrays, have recently been demonstrated by using angle resolved colloidal lithography. 81,89 Unfortunately, most of the current colloidal assembly techniques suffer from low throughput. It usually takes
33 hours to days to assemble a centimeter sized crystal. In addition, current colloidal assembly techniques (e.g., convective self assembly, 15 16,90 gravitational deposition, 3 and physical confinement) are not compatible with standard microfabrication, limiting the on chip integration of practical devices Moreover, only close packed colloidal crystals are available through traditional self assembly. 12,91 But in many cases, non close packed crystals are preferred to expand the versatility of colloidal lithography. 92 94 Further more, in conventional colloidal lit hography, only self assembled colloidal spheres are used as structural templates. W e show the scalable fabrication of binary periodic nanostructures by utilizing double layer non close packed colloidal crystals as templat es. Contrary to traditional colloidal lithography and our previous templating approaches, the polymer matrix between the spin coated colloidal crystals has also been used as a sacrificial template to define the resulting periodic binary nanostructures, inc luding intercalated arrays of silica spheres and polymer posts, gold nanohole arrays with binary sizes, and dimple nipple antireflection coatings. We also show that binary structured antireflection coatings exhibit better antireflective properties than uni tary coatings Experimental procedure By controlling the spin coating condition, a bilayer silica ETPTA nanocomposite can be fabricated. The resulting bilayer nanocomposite was etched by oxygen plasma etch at 40 mTorr pressure, 40 SCCM oxygen flow rate, an d 100 W for 2 min to partially remove ETPTA polymer matrix for releasing the top layer silica spheres of the spin coated double layer colloidal crystal. Both the bottom and top layer silica spheres were removed by dissolution in 2% hydrofluoric acid aqueo us solution for 2 min. Poly(d imethylsiloxane) (PDMS, Sylgard 184, Dow Corning)
34 precursors were mixed and degassed and then poured over the resulting polymer post arrays with binary dimple nipple structures. After curing at 80 C for 30 min, the solidified PDMS mold was peeled off the polymer array and put on top of ETPTA monomer supported by a glass slide. The ETPTA monomer was photopolymerized for 4 s, using a Xenon pulsed UV curing system. The PDMS mold was finally removed to complete the fabrication proc ess. A HR4000 High Resolution Fiber Optic UV vis spectrometer with a reflection probe was used for reflectance measurements. A calibrated halogen light source was used to illuminate the sample. The beam spot size was about 3 mm on the sample surface. Measurements were performed at normal incidence and the cone angle of collection was less than 5. Absolute reflectivity was obtained as a ratio of the sample spectrum and the reference spectrum. The reference spectrum was the optical density obtained from an aluminum sputtered (1000 nm thickness) silicon wafer. The final value of absolute reflectivity was the average of several measurements obtained from different spots on the sample surface. In the current colloidal templating approach, we use double lay er, non close packed colloidal crystals prepared by the well established spin coating technique as structural templates. 68,70 The spin coating methodology is based on shear aligning concentrated colloidal suspensions in a nonvolatile monomer. Contrary to conventional colloidal self assemblies, spin coating enables rapid pro duction of wafer scale (up to 8 in. in diameter) colloidal crystals with remarkably large domain sizes. Another unique property of the spin coated colloidal crystals is their unusual non close packed crystalline structures. The spheres within each colloida l layer are hexagonally arranged
35 and are separated by 1.4 times the particle diameter. The thickness of the shear aligned colloidal crystals can be precisely controlled by adjusting the spin coating conditions (e.g., spin speed and time). Most important, the shear aligned colloidal crystals are embedded in a polymer matrix, which has a highly uniform thickness over wafer scale areas. Here we demonstrate a n approach that uses colloidal spheres as templates to generate periodic binary nanostructures that are not easily available by traditional top down and bottom up technologies. The current colloidal templating technique is taken one step further by demonstrating the production of binary dimple nipple antireflection coatings (ARCs) on glass substrates. Antireflection for flat panel displays, increasing the transmittance of optical lenses, reducing glaring from automobile dashboards, and enhancing the conversion efficiency of solar cells. 33,61,64 The binary antireflective structure is inspired by the microstructured corneas of moths, which consist of hexagonal arrays of non close packed subwavelength n ipples, similar to the protruded polymer posts as shown in Figure 3 2B. The height of the resulting nipples is at most the radius of the templating spheres, limiting the optical thickness and thus the antireflection performance of the unitary ARCs. By cont rast, the optical thickness of the binary ARCs can almost be doubled. We expect this will lead to improved antireflective performance. To generate binary ARCs, we first assemble double layer, non close packed colloidal crystals (Figure 3 6 A) by using the s pin coating technology. The polymer matrix is then partially removed and the exposed silica spheres are dissolved in a 2% hydrofluoric acid aqueous solution, leaving behind a polymer dimple nipple array as
36 shown by the top and side view SEM image in Figur e 3 6 B and C respectively The resulting periodic binary structure c an then be transferred to a poly(dimethylsiloxane) (PDMS) mold and then replicated to a thin layer of polymer (e.g., ETPTA) or sol gel glass on a glass substrate by the well established soft lithography technique. 95 Figure 3 6 D shows a side view SEM image of a replicated ETPTA binary array. The specular optical reflection of the template polymer binary ARCs is evaluated by using UV vis reflecti vity measurement at normal incidence. The solid lines in Figure 3 7 show the experimental reflection spectra from a flat glass substrate and a glass slide covered with a polymer dimple nipple ARC. For comparison with the new binary antireflective structure the normal incidence reflection from a glass slide coated with a unitary ETPTA hemispherical ARC as reported in our previous section is also shown (blue curve). The glass control sample exhibits 4% single surface reflection, while the binary ARC coated glass slide displays <0.5% reflectivity for the whole visible spectrum. It is also evident that the binary ARC shows improved antireflective performance over the unitary hemispherical ARC for most of the visible wavelengths. Result s and discussion The exp erimental reflection measurement is complemented by theoretical calculation by using a rigorous coupled wave analysis (RCWA) model. 96 A coordinate system is set up with the z axis perpendic ular to the templated dimple nipple surface so that the bottoms of the dimples are at = 0 and the nipple tops at = We divide the dimple nipple arrays into = 100 horizontal layers with thickness ( ) / ( ) The dimple and nipple lattices are assumed to be hexagonal and the distance between the centers of the neighboring dimples and nipples is defined as 1.41 D where D is the diameter of templating silica spheres. O ther structural
37 parameters (e.g., dimple size and depth, nipple depth, size, and shape) are obtained from the side view SEM image as shown in Figure 3 6 D. The fraction of ETPTA in each horizontal layer can then be determined by simple geometrical calculation. On the basis of the effective medium theory, the effective refractive index of the layer at height can be approximated by = + 1 w here = 1.46, = 1.0, and =2/3. 69,75 We finally calculate the reflectance of the whole system by solving the Maxwell equation to express the electromagnetic (EM) field in each layer and then matching EM boundary c onditions between neighboring layers for the determination of the reflectance of the system. The dotted lines in Figure 6 show the RCWA simulated reflection spectra from a flat glass substrate and a binary ETPTA ARC. The simulated spectra agree reasonably wel l with the experimental results. We show that the templated binary dimple nipple antireflection coatings exhibit improved performance over unitary hemispherical nipples. This technique is scalable and compatible with standard microfabrication, promisin g a wide range of applications ranging from plasmonic devices to antireflection coatings for flat panel displays and solar cells. 3.2 Templating Antireflection Coatings on S emiconductor 3.2.1 Surface Texturing Silicon Wafers as Antireflection Coatings Current production of solar cells is dominated by crystalline silicon modules; 97 98 however, due to the high refractive index of silicon, more than 30% of incident light is reflected back, which greatly reduces the conversion ef ficiency of photovoltaic devices. 64,99 100 To significantly suppress the reflective loss of solar cells, various antireflection techniques, such as quarter wavelength multilayer films and nanoporous
38 coatings, have been developed. 63,99,101 For crystalline silicon solar cells, silicon nitride films deposited by plasma enhanced chemical vapor deposition have become the industry standard for antireflection coatings (ARCs) 99 Unfortunately, these existing techniques often perform suboptimally or are expensive to implement, impeding development of solar cells that can be made truly economically competitive with fossil fuels. Inspired by the microstructured cornea of some nocturnal moths, 60,65 subwavelength structured gratings directly patterned on silicon subs trates have been extensively explored both experimentally 102 105 and theoretically 106 108 for developing broadband ARCs. These gratings with a period smaller than the wavelength of in cident light are intrinsically more stable and durable than multilayer ARCs since no foreign material is invol ved. E lectron beam lithography (EBL) is a common approach in fabricating subwavelength gratings. 109 Unfortunately, the low throughput and the high cost of EBL raise big concerns. Interference lithography 102 103 and nanoimprint lithography 110 enable the creation of subwavelength antireflection struc tures over large areas; however these techniques are still expensive to implement. Spin coating technology provides a simpler and cheaper self assembly approach in creating wafer scale subwavelength antireflection gratings on single crystalline silicon substrates. Contrary to most bottom up approaches, 111 112 which are favorable for low volume, laboratory scale production, this nonlithographic technique is compatible with standard microfabrication, enabling large scale production of subwavelength ARCs for solar collectors.
39 Experimental procedure A schematic illustration of the fabrication procedures for making wafer scale subwavelength inverted pyramid gratings on single crystalline silicon wafers is shown in Fig ure 3 8 We start to generate non c lose packed colloidal monolayer on a ( 100 ) silicon wafer ( test grade, n type, Montco Silicon Technologies ) by t he spin coating technology. The non close packed silica particles f unction as shadow masks during an electron beam evaporation process for depositing a 30 nm thick chromium layer. After lifting off the templating silica particles, a periodic array of nanoholes whose diameter is determined by the size of templating silica spheres can be formed. These circular nanoholes can then be used as etching masks during a KOH anisotropic etching process to create wafer scale inverted pyramid arrays in silicon substrates. We are able to control the pyramid size by simply adjusting the anisotropic etching conditions. Figure 3 9 shows scanning electron microscope (SEM) images of two inverted pyramid arrays templated from the same spin coated silica monolayer sample (320 nm particle diameter) The samples are etched in the same solution co ntaining 62.5 g KOH, 50 ml anhydrous 2 propanol, and 200 ml ultrapure water at 60 C for 90 and 210 s, respectively. It is apparent that longer etching leads to larger pyramids with well defined square bases, while the less etched samples have rounded corn ers. The size of the pyramids can be larger than that of the templating silica spheres due to the undercutting of silicon substrates underneath chromium nanoholes. The long range hexagonal ordering of the templated pyramidal pits is evident from the SEM im ages. Th e orthogonal crosses at the centers of the pits confirm the inverted pyramidal structures. 113
40 The specular optical reflectivity of the replicated pyramid arrays is evaluated using visible near IR reflectivity measurement at normal incidence. The sol id lines in Fig ure 3 10 show the measured normal incidence specular reflection from a polished (100) silicon wafer and the sample, shown in Fig ure 3 9 B with 360 nm pyramidal pits. The flat silicon substrate exhibits high reflection (~35%) for visible and near infrared wavelengths, while the subwavelength structured grati ngs show reduced reflection of ( 10% for long wavelengths >600 nm) The reflection is further reduced to ~ 2% for wavelengths around 400 nm. For smaller pyramids, the reflection progressively increases with decreased pyramidal pits. Results a nd discussion The templated silicon subwavelength gratings exhibit lower reflection than colloid based antireflection coatings on crystalline silicon solar cells. 64 Though the normal incidence reflection from the templ ated pyramid gratings is higher than other subwavelength structured ARCs made by lithographic techniq ues with typical reflection of ~ 2% 10%, 60,102,109 110 the cost benefit of this nonlithographic methodology is a major advantage. Additionally, optimization of the template structures will facilitate further improvement of the antireflection performance. The state of the art silicon nitride ARCs on crystalline silicon solar cells exhibit minimal (~2%) reflection around 600 nm, but t he reflection increases to more than 10% for near IR (>800 nm) and visible (<500 nm) wavelengths, which account for a significant portion of the solar spectrum. 64,99 By contrast, the template pyramid arrays show relatively low reflection at short wavelengths (Fig ure 3 10 ) A multilayer rigorous coupled wave analysis (RCWA) model 114 115 has also been developed to complement the optical measurement. Firstly, we divide the inverted
41 pyramid array into 100 horizontal layers with equal thickness. Since the KOH etched silicon pyramids have characteristic of 54.7 sidewalls, the depth of the anisotropic V sh ape pitches is determined by the base length of the pits. Based on the effective medium theory, 69 the effective refractive in dex of the layer at level can be approximated by = [ + ( 1 ] 1 / where is the fraction of silicon contained in the layer, = + is the complex refractive index of silicon ( and are optical consta nts) =1, and = 2 / 3 69 The optical constants of silicon which are functions of wavelengths are obtained from li terature. 116 Secondly, we calculate the reflectance of the whole system by solving the Maxwell equation to express the electromagnetic (EM) field in each layer and then match EM boundary conditions between neighboring layers for the determination of the reflectance of the system. The RCWA simulated reflection for a bare silicon substrate and an inverted pyramid array in silicon with 360 nm base length are shown by the dotted lines in Fig ure 3 10 It is apparent that the theoretical prediction for single crystalline silicon is close to the experimental spectrum, while for the subwavelength structured pyramid gratings, the modeling results only match with experimental data when the w avelength is large. The difference tends to be large when the wavelength becomes small. This is due to the limitation of the modeling where each layer is assumed to have a uniform refractive index, 106 which is, in fact, a two dimensional periodic function at each layer. Meanwhile, the effective refractive index formula is only accurate when the wavelength of light is larger than the period of the pyramid arrays. 69
42 To understand the improved antireflection performance for subwavelength structured pyramid arrays, we plot the calculated real part of t he complex refractive index at =600 nm versus th e depth of pyramids with 60 nm ( blue circles) and 360 nm (red circles) base lengths in Fig ure 3 1 1 For 60 nm size pyramids, the refractive index changes sharply at the interface between air and the structured silicon surface. The graded index which is desirable for suppressing the optical reflection, 60 is observed for 360 nm pyramids the refractive index changes from 1.0 to 1.7 at the air/silicon interface and then gradually changes to the bulk index of silicon. We further calculate the normal incidence reflectio n at =600 nm for pyramid array samples with different sizes and the results are shown in Fig ure 3 11 It is apparent that larger pyramids lead to lower reflection, matching our experimental observation. 3. 2.2 Broadband Antireflection Coatings on Silicon Wafer Subwavelength inverted pyramid arrays are anisotropically wet etched in single crystal silicon wafers Although this nonlithographic technique is much simpler and cheaper than PECVD deposition of SiN x ARCs, the antireflective performance of the templated ARCs is not as good as the commercial coatings. The normal incidence reflection is higher than 10% for most solar wavelengths due to the limited depth of the inverted pyramids. 60 Additionally, the technique requires va cuum deposition of chromium which significantly increases the complexity and the cost for manufacturing ARCs on single crystal silicon solar cells. We use a bio inspired templating technique for fabricating broadband moth eye ARCs on SC silicon substrates. The resulting subwavelength structured ARCs exhibit superior broadband antireflective performance than commercial SiN x coatings.
43 Experimental procedure The schematic illustration of the templating procedures for making wafer scale silicon nipple arrays is shown in Fig ure 3 13 Monolayer silica colloidal crystals with non close packed structures are created by the established spin coating technique. The non close packed silica particles can be used as etching masks during a SF 6 reactive ion etching (RIE) process ( 40 mTorr chamber pressure, 26 SCCM SF 6 ,5 SCCM O 2 25 W to pattern arrays of nipple like structures directly on silicon wafers ( test grade, n type, ( 100 ) Montco Silicon Technologies) Under above RIE conditions, the etching selectivity between silicon and silica is more than 10:1. 111 Templating silica spheres are finally removed by a hydrofluoric acid wash ~2% aqueous solution for 2 min. Figure 3 14 A shows a scanning electron microscope (SEM) image of a monolayer colloidal crystal consisting of 360 nm diameter silica spheres prepared by the spin coating technique. The long range hexagonal arrangement and the non close packed structure of silica particles are evident. Figure 3 14 B shows a tilted SEM image of a sample after 9 min SF 6 etch. Due to the high etching selectivity between silicon and sili ca, the unprotected silicon is preferentially etched by reactive fluorine ions. The shrinkage of templating silica spheres is not significant. The RIE etching is isotropic, leading to the formation of nipple like microstructures whose sizes and aspect rati os are determined by the templating silica sphere diameter and the RIE etchi ng time. Figure 3 14 C and D show tilted SEM images of silicon nipple arrays after 6 and 9 min RIE etching. The preservation of the hexagonal ordering and the interparticle distance of the original silica colloidal crystal are apparent for the templated nipple arrays. Subwavelength scale nipples with aspect ratio s as high as ~ 5.0 can be easily fabricated
44 by this approach The specular optical reflectivity of the templated silicon nip ple arrays are evaluated using visible near IR reflectivity measurement at normal incidence. Figure 3 15 compares the normal incidence reflection from a 15 min RIE etched silicon nipple array sample, a bare SC silicon wafer, and a commercial SC silicon sol ar cell with PECVD deposited SiN x ARC (Plastecs,Webster, MA) The polished SC silicon wafer shows more than 35% reflection for wavelengths from 350 to 850 nm (black solid line) agreeing well with previous measurement. 83,105 The Plastecs SC silicon solar cell exhibits distinctive bluish color caused by the SiN x coating and presents the minimal reflection at around 700 nm ( blue solid line) The templated nipp le array shows excellent broadband antireflection property and the reflection is below 2.5% for a wide range of wavelengths (red solid line) Results and discussion The experimental reflection measurement is complemented by theoretical calculation using a rigorous coupled wave analysis RCWA model. 109 A coordinate system is set up with z axis perpendicular to the nipple surface so that the array troughs are at = 0 and the nipple peaks at = We divide the nipple arrays into = 100 horizontal layers with thickness / The z coordinate and the radius of each layer satisfy = 1 ( ) where 0 The nipple lattice is assumed to be hexa gonal and the distance between the centers of the nei ghboring troughs is defined as 1.41 D, where D is the diameter of templating silica spheres. 103,117 Therefore, the horizontal layer at level contains a fraction ( ) of silicon with complex refractive index = + and are optical constants obtained from literature 118 and a fraction 1 ( ) of air with refractive index =1. The hexagonal
45 structure of the nipple arrays leads to the formula = 2 3 2 Based on the effective medium theory, 104 the effective refractive index ( ) of the layer at height can be approximated by = [ + ( 1 ] 1 / where = 2 / 3 102,104 We finally calculate the reflectance of the whole system by solving the Maxwell equation to express the electromagnetic (EM) field in each layer and then matching EM boundary conditions between neighboring layers for the determination of the reflectance of the system. 109 The dotted lines in Fig ure 3 15 show the RCWA simulated reflection from a bare silicon wafer and a subwavelength nipple array with base radius of 210 nm and height of 800 nm templated from 360 nm silica spheres. It is evide nt that the simulated results agree reasonably well with the experimental spectra for both samples, though both the peak position and the reflection magnitude match with the RCWA calculation for flat silicon, while only magnitude matches with theoretical p rediction for periodic nipple arrays. This is due to the limitation of the RCWA modeling where each layer is assumed to have a uniform refractive index, 109 which is in fact a 2 D periodic function at each layer. We further conduct optical modeling for the same subwavelength nipple array using another common simulation method the thin film multilayer (TFM) model. 104 The TFM model employs similar layer slicing method as RCWA but a different matrix mu ltiplication procedure to calculate the reflection from a stack of thin layers. The simulation results obtained from both models are compared in Figure 3 15 It is apparent that these two models generate almost identical results. The resulting gratings exh ibit excellent broadband antireflection properties. The specular optical reflection from the templated moth eye ARCs matches with theoretical
46 predictions using RCWA and TFM models. These biomimetic coatings are of great technological importance in improvin g the conversion efficiencies of crystalline silicon solar cells. 3 .2.3 Biomimetic Subwavelength Antireflective Gratings on GaAs Gallium arsenide (GaAs) is a technologically important semiconductor that has been widely used in optoelectronics, such as vert ical cavity surface emitting lasers 119 near IR photodetectors 120 and highly efficient concentrator solar cells 121 122 However, owing to the high refractive index (RI) of GaAs ( =3.6 for visible wavelengths), more than 30% of incident light is reflected back from the substrate surface. This greatly re duces the efficiency of GaAs based optoelectronic devices. To suppress the unwanted reflective losses, vacuum deposited multilayer dielectric (e.g., MgF 2 /ZnS ) are used as antireflection coatings (ARCs) Unfortunately, these multilayer ARCs are expensive t o fabricate owing to the stringent requirement of high vacuum, material selection, and layer thickness control. Additionally, thermal mismatch induced lamination and material diffusion of the multilayer ARCs limit the device performance at high power densi ties. 121 Inspired by the broadband antireflection of microstructured cor neas of moths, which consist of non close packed arrays of sub 300 nm nipples 60 subwavelength antireflective structures have been extensively exploited 123 124 These structures can reduce reflection over a wider range of wavelengths and exhibit much improved thermal stability than conventional multilayer ARCs. However, scalable production of subwavelength ARCs is not a trivial task for the current top down nan olithography technologies (e.g., electron beam lithography and interference lithography) 100,109 Bottom up colloidal lithography, which uses self assembled colloidal crystals as
47 depositio n or etching masks to pattern periodic nanostructures 85,89 provides a much simpler and inexpensiv e alternative to nanolithography in creating subwavelength gratings 64,125 U nfortunately traditional colloidal assemblies suffer from low throughput, small areas, incompatibility with standard microfabrication, and limited close packed crystal structures have been developed. S ignificant obstacles have been encountere d when we employ this nonlithographic spin coat technique to GaAs substrates. Only disordered colloidal monolayer can be assembled on GaAs by spin coating. This greatly reduces the uniformity and reproducibility of the antireflective performance of templated ARCs. We use a generalized single layer reduction (SLD) technique that enables reproducible production of monolayer colloidal crystals on a variety of substrates that otherwise lead to the formation of disordered crystals by the traditional spin coating approach. We also demonstrate the scalable production of broadband ARCs on crystalline GaAs substrates, and their antireflective properties are characterized by both experimental measurements and theoretical simulation. Experimental procedure Figu re 3 17A shows a typical scanning electron microscope (SEM) ima ge of a colloidal monolayer con sisting of 320 nm silica spheres prepared on a crystalline GaAs substrate ( N type, (100), University Wafer) by the standard spin coating process. The particles ar e completely disordered as confirmed by the rings in the Fourier transform of the SEM image. Contrary to the direct formation of colloidal monolayer by spin coating, where the substrate plays a crucial role in determining the resulting crystalline quality, double layer colloidal crystals are f irstly assembled in SLD. Figure 3 17B shows a double layer colloidal crystal prepared by spin
48 coating colloidal silica suspensions (20 vol. % in nonvolatile ethoxylated trimethylolpropane triacrylate monomer) at 8000 r pm for 9 0 s. The non close packed hexagonal arrangement for both the top and bottom layer particles is clearly evident. After removing the top layer particles by sweeping using a clean room Q tip under flowing water, crystalline colloidal monolayer with non close packed structure are formed as shown in Fig ure 3 17C. The distinctive six arm Bragg diffra ction star formed on the 2 in. (~5.08 cm) GaAs wafer ( inset of Fi gure 3 17C) indicates the hexagonal ordering is over the whole wafer surface. The interpart icle distance of the non close packed colloidal crystal is determined to be 1.4 1 D where D is the diameter of silica spheres, by the first peak of the pair correlation function (PCF, Figure 3 17D) which is calculated as = 1 / ( + ) / ( + ) where counts the number of spheres that lie within a spherical shell, of radius from an arbitrary origin; = 2 for a particular radial distance r ; and is the average particle number density. Figure 3 17D also shows the positions of the oscillating PCF peaks agree well with those obtained from a perfect hexagonal lattice with 2 interparticle distance. Preliminary results show the SLD technology can be easily and reproducibly applied to other important su bstrates, such as Ni, Cr, glass, and poly(methyl methacrylate). Only disordered monolayer is formed on these substrates by the traditional spin coating technique The periodically arranged silica particles can then be used as etching masks during a BCl 3 re active ion etch (RIE) process operating at 7.5 mTorr chamber pressure, 20 SCCM BCl 3 RIE=100 W, and inductively coupled plasma ICP =300 W on a Unaxis Shuttlelock reactive ion etcher to generate arrays of cone shaped nanopillars. Figure 3
49 18A and B show the templated GaAs structures after 9.5 min RIE and removing templating silica particles by a 2% hydrofluoric acid wash. Interestingly, we observe the formation of mushroom mushrooms are cone shaped and are caused by th e isotropic etching of GaAs by reactive chlorine ions. The polymer wetting layer between the templating silica spheres and the substrate 126 which protects the GaAs surface immediately under the particles from being etched, causes the formation of the flat has been used to develop superoleophob ic surface though it is not easily available by conventional top down and bottom up techniques. Longer dry etching leads to sharper conical stems and smaller caps as revealed by the SEM image of a 12 min etched sample in Fig ure 3 18C. The residual caps can finally be removed by a brief ultrasonication in acetone, resulting in the formation of conical GaAs gratings as shown in Fig ure 3 18D. The preservation of the long range hexagonal ordering and the interparticle distance of the original silica colloidal m onolayer throughout the templating process are clearly evident by comparing the SEM images of Fig ures 3 17 and 3 18. The specular reflectivity of the templated subwavelength gratings are evaluated using visible and near IR reflectivity measurement at norma l incidence by using an Ocean Optics HR4000 spectrometer. The solid curves in Fig ure 3 19 show the measured reflection from a bare GaAs wafer and the nanocone grating as shown in Fig ure 3 18D The wafer exhibits > 30% reflectivity, while much reduced reflec tion (~3%) is obtained for the templated subwavelength grating. Importantly, the resulting moth eye ARCs on GaAs are broadband, exhibiting consistent low reflection over both visible and near IR wavelengths.
50 Results and discussion The optical measurement is complemented by theoretical simulation using a rigorous coupled wave analysis (RCWA) model 106 A conical profile is used to simulate the te mplated nipples as shown in Figure 3 18D We divide the cone array into 100 horizontal circular layers. The in tercone dis tance is defined as 2 D by the PCF calculation shown in Figure 3 17D We can calculate the fraction of GaAs in each layer as = 2 / 3 2 where and are the coordinate and the bottom radius of the cone, respectively. Based on the effective me dium theory, the effective RI of the layer at height can be approximated by = [ + ( 1 ] 1 / where =2/3 69,123 We can then calculate the tot al reflectance of express the electromagnetic (EM) field in each layer and matching EM boundary conditions between neighboring layers. The complex RI of GaAs is used to calculate the resulting reflectance 127 The curves with circles in Fig ure 3 19 show the simulated reflection from a flat GaAs wafer and an array of conical nipples templated from 320 nm silica spheres. The calculated spectrum for the featureless substrate has the same shape as the experimental result. The simulated reflection from the subwavelength nanocone array also agrees reasonably well with the experimental spectrum. The excellent br oadband antireflection properties of the templated conical gratings can be understood by mapping the calculated effective RI at =700 nm across the height of 320 nm conical nipples (Fig ure 3 20 ). For bare wafer, the RI changes sharply from air (RI=1.0) to bulk GaAs (RI=3.772) while for templated nipples, the RI changes gradually from 1.0 to 3.41 and then to 3.772. As the reflection ( ) from an interface between two materials with RI
51 of 1 and 2 is governed by equation = 1 2 1 + 2 2 the RI gradient exhibited by the conical nipples thus leads to very low reflection over a wide range of wavelengths. The technique is scalable and compatible with standard microfabrication, promising for developing more economic, efficient, and reliable GaAs based optoelectronic devices. Figure 3 1. Schematic illustration of the templating procedure from making moth eye antireflection coatings.
52 Figure 3 2. AFM images and corresponding depth profiles of replicated nipple arrays at different RIE etching times. A) 0 s B) 20 s. C) 45 s. 360 nm diameter silica spheres are used as templates Figure 3 3 Experimental ( solid ) and calculated ( dotted ) specular optical reflectivity at normal incidence. Black: flat PETPTA. Light gray: 20 s RIE etch, 110 nm spherical cap. Dark gray: 45 s RIE etch, 180 nm hemispherical cap. A B C D Wavelength (nm)
53 Figure 3 4 Dependence of the normal inc idence optical reflectivity at 600nm vs. RIE etching time performed on spin co ating colloidal crystal polymer nanocomposite. Figure 3 5 Comparison of the change of calculated effective refractive index from the nipp le peaks (n eff =n air =1) to the nipple trough s ( n eff = n PETPTA =1.46) between hemispherical (unfilled circle) and 50nm height spherical cap (cross) nipple arras. The diameter of colloidal particles is 360nm. Reflectivity Etch Time (s) Refractive Index Distance (nm)
54 Figure 3 6 SEM images illustrating the progressive fabrications procedures for binary dimple nipple antireflection coatings: A) top view image of a double layer colloidal crystal template c onsisting of 340nm silica spheres, B) top view image after removing silica spheres by wet etch, C) side view image of the sample in panel B and D) replicated ETPTA binary array from the sample in panel B. Figure 3 7. Experimental (solid) and RCWA simulated (dotted) normal incidence specular reflection from a bare glass substrate (black), a glass slide co v er ed with a temp l ated ETPTA binary dimple nipple a r ray (red), and a glass slide coated with an ETPTA unitary hemispherical nipple array (blue). A B C D Reflection (%) Wavelength (nm)
55 Figure 3 8. Schematic illustration of the templating procedures for fabricating subwavelength structured antireflection pyramid grating s on single crystalline silicon Figure 3 9. Replicated inverted pyramids arrays in silicon. A) The 300nm size pyramids etched at 60C for 90s. B) The 360nm size pyramids etched at 60C for 210s. T he 320nm diameter silica spheres are used as templates. A B
56 Figure 3 10. Experimental (solid) and RCWA modeled (dotted) specular optical reflectivity at normal incidence. Black: bare (100) silicon wafer. Red: 360nm size pyramids template from 320nm silica spheres. Figure 3 11. Comparison of the change of calculated effective refractive index at =600 nm from the wafer surface ( depth=0 ) to the vertex plane of inverted pyramids between a 360 nm pyramid array ( red circles ) and a 60 nm pyramid array ( blue circles ) The diameter of templating silica spheres is 320 nm. Reflection (%) Wavelength (nm) Refractive Index Inverted Pyramid Depth (nm)
57 Figure 3 1 2 RCWA simulated normal incidence optical reflection at =600 nm vs. inverted pyramid size. The 320 nm silica spheres are used as templates. Figure 3 13. Schematic illustration of the templating procedures for fabrication moth eye ARCs on silicon substrates. Reflection (%) Inverted pyramid size (nm)
58 Figure 3 14. SEM images of biomimetic antireflection coatings on silicon substrates. A) SEM images of a spin coated monolayer non close packed colloidal crystal consisting of 360nm silica spheres. B) Tilted (20 o ) SEM image showing templating silica array and underneath silicon nipples etched by SF 6 R IE for 9min. C) Templ a ted silicon nipple array after 6min of SF 6 etch. D) Silicon nipples after 9min SF 6 etch. Figure 3 15. Experimental (solid) and RCWA simulated (dotted) specular optical reflection at normal incidence for a flat silicon substrate (black lines), a commercial multicrystalline silicon solar cell (blue lines), and a 15 min etched nipple array (red lines). A B C D Reflection (%) Wavelength (nm)
59 Figure 3 1 6 Comparison between the calculated optical reflection at normal incidence by the RCWA model ( red ) and the thin film multilayer model ( blue ) for an array of nipples with 210 nm base radius and 800 nm height. Figure 3 1 7 SEM images of silica nanoparticles on GaAs. A) Disordered colloidal monolayer formed on GaAs by traditional spin coating. Inset, Fourier transform of the image. B) Double layer colloidal crystals formed on GaAs. C) Monolayer colloidal crystal formed by SLD. Inset, photograph of a monolayer crystal f ormed on a 2 in. GaAs wafer illuminated with white light. D) Pair correlation function calculated from a low magnification SEM image. For comparison, the PCF for a n ideal hexagonal lattice with 2D inter particle distance is also shown (black lines). A B C D Reflection (%) Wavelength (nm)
60 Figure 3 1 8 SEM images of templated nanacones on GaAs substrates. A) Templated mushroom like GaAs microstructures after 9.5 min BCl 3 RIE. B) Higher magnification image of A. C) Conical nipples formed after 12 min BCl 3 RIE. D) Same sample as C after a brief ultrasonication in acetone. Figure 3 1 9 Experimental and RCWA simulated specular reflection at normal incidence from a bare GaAs wafer and the conical grating as shown in Figure 3 1 8 D. A B C D Reflection (%) Wavelength (nm) Bare GaAs GaAs Cones
61 Figure 3 20 Change of the calculated effective refractive index at =700 nm from the wafer surface (depth=0) to the bottom of templated GaAs nanocones with 320 nm height. The dotted line indicates the refractive index of bulk GaAs at the same wavelength. Refractive index Depth (nm)
62 CHAPTER 4 TEMPLATED FABRICATIO N OF SER S ACTIVE SUBSTRATES 4.1 Templated Fabrication of Periodic Metallic Nanopyramid Arrays Metallic substrates with periodic subwavelength structures exhibit unique surface plasmon properties that are of great scientific interest and considerable technological importance in developing nanophotonic devices, data storage, and biosensors 36,75, 128 130 Metallic nanotips are particularly interesting as surface enhanced Raman scattering (SERS) substrates because they can concentrate the electromagnetic field at the tip apex, resulting in a great enhancement of the Raman scattering intensity from molecules in the vicinity of the nanoscopic tips. 131 134 Scanning probe microscopy (SPM) tips coated with noble metals (e.g., Au and Ag) have been shown to enhance Raman scattering by 3 6 orders of magnitude. 133 134 The Raman enhancement factor of gold nanocrescent moons with sub 10 nm sharp edges, which are tem plated from sacrificial microspheres, i s estimated to be larger than 1x 10 10 making them promising for ultrasensitive biosensors. 7 Other asymmetric metallic nanostructures, such as half shells, 135 nanocups, 136 nanoprisms, 137 and nanorings, 138 have also been developed for potential SERS applications. Unfortunately, precise control over the position, orientation, and spacing of these aspheric nanomaterials is not trivial and impedes the exper imental reproducibility of SERS sensing. Planar nanotip arrays with high tip density and aligned tip orientation are promising SERS substrates that could resolve the reproducibility issue of current tip enhanced Raman spectroscopy. 132 Electron beam lithography (EBL) and focused ion bea m (FIB) lithography are two popular technologies used to create periodic metallic nanostructures with arbitrary patterns. 74,139 Unfortunately, the high cost and the low throughput greatly
63 limit the implementation of these advanced lithographic techniques in practical application. Templated synthesis provides a much cheaper, simpler, and faster alternative to complex nanolithography in creating periodic metallic nanostructures. 28 Examples include an array of metallic pyramids with nanoscale tips made by templated deposition of metals on anisotropically et ched silicon pits, which are defined by phase shifting photolithography, 140 142 and an array of subm icrometer scale particles with uniform periodic spacing and nearly uniform size and crystallographic orientations prepared by solid state dewetting of gold thin films on oxidized silicon surfaces patterned with pyramidal pits by interference lithography. 143 Other templates, such a block copolymers, 144 colloids, 86,145 146 and even biological ma terials, 147 have also been u tilized in fabricating periodic metallic nanostructures with length scales ranging from nanometer to micrometer. However, most of these templating approaches either still require some sort of lithographic patterning or are not compatible with standard microfabrication, thereby impeding the cost efficiency and scale up of these unconventional methodologies. We use metallic nanohole arrays templated from spin coated two dimensional colloidal crystal s to devel op a new templating technique for fabricating wafer scale, periodic metallic pyramid arrays with nanoscale tips and high tip density. A nisotropic KOH wet etching is utilized to transfer the periodic circular nanoholes into inverted pyramid arrays in (100) silicon wafers. These inverted pyramids can then be used as third generation templates for replicating large area metallic nanopyramid arrays by simple physical vapor deposition and adhesive peeling. This new nanofabrication approach is based solely on sel f asse mbly and templated fabrication and no
64 lithographic patterning is involved in any step. Yet this technology is still scalable and compatible with standard microfabrication enabling large scale produc tion of metallic nanostructures for potential SERS applications. 4.2.1 Experimental Procedure The fabrication of wafer scale, monolayer and non close packed colloidal crystal polymer nanocomposites is performed according to Chapter 2 In short, monodispersed silica colloids are dispersed in ETPTA to make f inal particle volume fraction of ~ 20%; 2 wt % Darocur 1173 is added as photoinitiator. The silica ETPTA dispersion is dispensed on a 3 acryloxypropyl trichlorosilane (APTCS) primed (100) silicon wafer and spin coated at 8000 rpm for 6 min on a standard spi n coater, yielding a hexagonally ordered colloidal monolayer. The monomer is then photopolymerized for 4 s using a Pulsed UV Curing System. The polymer matrix is fully removed using a reactive ion etcher operating at 40 mTorr oxygen pressure, 40 SCCM flow rate, and 100 W for 4 min. A 30 nm mask of chromium is deposited on the wafer using sputtering deposition at a deposition rate of 1.6 /s. The wafer is then rinsed in deionized water and rubbed with a cleanroom Q tip to remove templating silica microsphere s. Templating silica particles can also be removed by dissolving them in a 2 vol % hydrofluoric acid aqueous solution for 2 3 min. The removal of the particles creates a visible color change. The (100) silicon wafer covered by arrays of chromium nanoholes is then wet etched in a freshly prepared solution of 62.5 g KOH, 50 mL of anhydrous 2 propanol, and 200 mL of ultrapure water at 60 C for various durations. The wafer is rinsed with deionized water and then wet etched with a chromium etchant (type 1020, T ransene) to remove the chromium template. The etched wafers show iridescence under white light illumination. To create a nanopyramid array in gold, we sputtered the wafer with 500 nm of gold at a
65 deposition rate of ~ 5 /s. The layer of gold on the surface of the wafer can be easily peeled off with Scotch tape (3M), yielding a non close packed nanopyramid array in gold. To separate the metallic nanopyramid arrays from the silicon templates in a more reliable and reproducible way, we applied a thin layer of p olyurethane adhesive (NOA 60, Norland Products) between the metalized wafer and a glass substrate. The adhesive is then polymerized by exposure to ultraviolet radiation. The silicon wafer templates can finally be peeled off, resulting in the formation of w afer scale nanopyramid arrays supported on glass substrates. Gold nanopyramid array samples are placed in a 5 mM solution of benzenethiol and 200 proof ethanol for 45 min and then rinsed in roughly 25 mL of 200 proof ethanol for several minutes. The samples are allowed to dry in air for 20 min, after which the Raman spectra are measured. A flat gold film sputtering deposited on a glass slide using the same deposition condition is used as the control sample for Raman spectra measurements. Raman spectra are measured with a Renishaw inVia confocal Raman microscope using a 785 nm diode laser at 15 mW with an integration time of 10 s and a 40 m 2 spot size. 4.2.2 Results and Discussion In contrast with previous lithographic approaches to fabricating nanopyramid arrays, 132,140 our method is simply based on colloidal self assembly and templated sy nthesis. A schematic outline of the fabrication procedures is shown in Figure 4 1. We start from spin coating concentrated silica ETPTA monomer dispersions using a standard spin coater. Wafer scale, monolayer, hexagonally ordered colloidal arrays can be r eproducibly made in minutes by controlling the spin speed and time. Following a
66 rapid photopolymerization of ETPTA monomer to immobilize silica particles, the polymer matrix are removed by a brief oxygen plasma etch. The resulting colloidal monolayer exhib its non close packed arrangement of particles with interparticle distance of ~ 1.4 D where D is the diameter of the silica spheres (see Figure 4 2A). Using this simple spin coating method, silica particles with a wide range of sizes from ~ 100 nm to larger t han 1 m have been asse mbled into monolayer crystals. Manipulation of silica particle size enables control of the size an d separation of nanopyramids in the templated arrays. These shear aligned, non close packed silica particles can then be utilized as deposition masks during conventional physical vapor deposition (e.g., sputtering, thermal evaporation, or electron beam evaporation). The deposited metals, such as Cr or Ti/Au, fill the interstitials between silica spheres and accumulate on the top halves of particles as well. Because silica particles are loosely attached to the substrate, they can be easily removed by gentle rubbing with a cleanroom swab, leaving behind a metallic nanohole arrays as sh own by the SEM image in Figure 4 2 B. Templating silica spheres can also be removed by dissolving in a 2% hydrofluoric acid aqueous solution to lift off metals. We found that a thin layer of chromium (20 30 nm) is sufficient to sustain the KOH wet etching in the following step. These circular nanoholes retain t he size and spacing of the templating silica spheres as well as their hexagonal long range ordering. Under white ligh t illumination, these templated nanohole arrays function as diffraction gratings, exhibiting strong iridescence. Shape and edge roughness o f the templated nanoholes determine the qualities of the resulting inverted pyramids in silicon, 141 thus pre cautions need to be taken to ensure circular shapes and smooth edges of the metallic nanoholes. Slower PVD deposition rate helps to reduce the grain size and the
67 roughness of edges, whereas thermal or EB evaporation, which is more unidirectional in deposit ion, is better than sputtering in maintaining the circular shapes of nanoholes. 45 We then use templa ted chromium nanohole arrays as second generation etching masks to create inverted pyramid arrays in (100) silicon wafers through anisotropic etching in an aqueous solution containing KOH and 2 propanol. It is well known that KOH is a wet etchant that attacks silicon preferentially in the < 100 > plane, producing characteristic anisotropic V shape pitches with 54.7 sidewalls. 45 Images A and B in Figure 4 3 show SEM images of pyramidal pits that are templated from 320 nm silica spheres and etched at 60 C for 120 and 420 s, respectively. The long range hexagonal ordering of these pits is obvious from the SEM images and is further confirmed by the hexagonally arranged dots in the fast Fourier transform (FFT) of these images (inse ts of images A and B in Figure 4 3 ). Another interesting feature of th e calculated FFTs is the two sets of four arm stars with exact 90 angles between neighboring arms surrounding the central dots, characteristic of square pyramidal pits. The orthogonal cro sses at the centers of the pits also verify the inverted pyramidal s tructures. The spacing between neighboring pits is the same as that of the original non close packed colloidal arrays (Figure 4 2A), whereas the pit size (252 28 nm) for the 120 s sample is smaller than the size of the nanoholes ( ~ 320 nm). This indicates that the etching reaction starts from the center of the nanoholes and then propagates to the edges; otherwise, the spacing between neighboring nanoholes could not be retained in the templated pyramidal pits. For longer etching duration, undercutting of si licon underneath the chromium nanoholes occurs. This leads to larger inverted pyramids (Figure 4 3B) with well defined square base s. For shorter etching time ( Figure 4 3A), the corners of the square bases
68 are not as sharp as those of over etched ones. We also notice that there are some rectangular shaped pyramids in the anisotropic etched samples that are replicated from noncircular (e.g., oblate) nanoholes. 141 The size and depth of the inverted pyramidal pits can be easily controlled by adjusting the wet etching duration. Figure 4 3C shows the dependence of the size of the pyramids vs. different etching time at 60 C. More than 100 pyramids are measured using SEM to arrive at the reported size and size distribution of each sample. If we rise the reaction temperature to 80 C, which is commonly used in anisotropic etching of silicon for micromachining, 45 the etching reaction becomes so vigorous that the precise control over the di mensions of the pyramids is more difficult. For certain KOH etched sampl es, we find unwanted particle precipitation on the surface of the silicon wafer as well as in the pyramidal pits. Figure 4 4 shows a typical SEM image of such particles precipitated on an anisotropically etched n type silicon wafer. Though these particles are sparsely distributed on the wafer surface, they affect the uniformity of the resulting gold nanopyramid arrays by creating random defects in the pyramids. Previous study shows that these particles are iron oxide precipitated from the reaction of iron impurities in KOH pellets with hydroxide ions. A brief etching (1 min) in 2 M HCl aqueous solution at room temperature can easily remove these unwanted particles. 148 We then use the inverted silicon pyramids as the third generation templates to replicate metallic nanopyramid arrays. Conventional PVD deposition is carried out to deposit various metals in the silicon pits and form continuous metal films on the surface of the silicon wa fer. For metals with weak adhesion to silicon, such as Au, Ag, Pt, and
69 Pd, the deposited films can be adhered onto a glass substrate using a thin layer of polyurethane adhesive, followed by peeling off the silicon templates. The resulting wafer scale nanop yramid arrays exhibit striking diffractive colors and a characteristic six arm diffractive star (Figure 4 5A). The adjacent arms of the diffraction star form exact 60 angles, a result indicating the formation of long range hexagonal ordering. 8,49 Figure 4 5B shows a typical top view SEM image of a replicated Au nanopyramid array. The hexagonal ordering of nanopyramids is clearly evid ent from the image. However, the periodic nanopyramids are polycrystalline, as is clear from the SEM images. The typical domain size is several hundred micrometers, limited by the single crystal domain sizes of the original spin coated monolayer colloidal crystal. Our previous results show that spin coated monolayer crystals have much smaller single crystal domains than multilayer crystals made by the same spin coating process. The gold pyramids are faithful replica of the original inverted silicon template s, indicating that no breaking of sharp tips occurred during the film peeling off procedure. Most of the pyramids have sub 10 nm sharp tips revealed by the magnified SEM image shown in the insets of images B and C in Figure 4 5 and the side view SEM image in Figure 4 5C. The spacing between neighboring nanopyramids measured using SEM is the same as the original non close packed colloidal arrays. By a simple geometrical calculation, we estimate the nanotip density is about 6x 10 8 tips cm 2 f or 300 nm templating silica spheres. For metals with strong adhesion to silicon, such as Cr, Ni, and Al, the surface of silicon pits can be modified by octadecyl silane (ODS) to reduce the metal adhesion and facilitate the film peeling, 149 or a thin film of Au can be predeposited as a sacrificial layer before the deposition of adhesive metals. KOH/IPA wet etching can also be used
70 to dissolve the silicon templates to release the metallic nanopyramid array s provided the metals can survive the chemical etching. 141 We evaluate the performance of our gold nanopyram id arrays as SERS substrates using benzenethiol as a model compound (Figure 4 6). In addition to the excellent affinity to gold surfaces, benzenethiol molecules have a large Raman cross section, presumably as a result of chemical enhancement alongside elec tromagnetic enhancement. 128 The templated gold nanopyramid array sample (red curve) gives a strong Raman signal of adsorb ed benzenethiol molecules. The positions of Raman peaks agree well with those in the literature for benzenethiol on gold substrates. 128,150 151 In control experiments, no SERS spectra are observed for benzenethiol molecules adsorbed on the flat sputtered gold films deposited under the same conditions on glass substrates (blue curve). The SERS enhancement factor for the gold nanopyramid substrate is estimated to be ~7x 10 5 using the method described in the literature by comparing the Raman intensity for two peaks at 1080 and 1581 cm 1 150 151 Previous results show that the greatest SER S enhancements occur when localized plasmon resonances on the structured metallic surfaces are present at both the excitation wavelength and Raman scattered wavelength. 151 Here the structural parameters of the gold nanopyramid arrays (e.g., pyrami d size, sep aration, and height), which greatly affect the plasmon resonances, have not been optimized yet. We believe the SERS enhancement factor of our periodic substrates can be further improved by tailoring the structures of the templated nanopyramid ar rays to mat ch the optimal SERS requirements
71 4.2 Wafer Scale SERS Substrates with Highly Reproducible Enhancement Using spin coated monolayer colloidal crystals as templates, we have fabricated periodic gold nanopyramid arrays with nanoscale sharp tips. They have also been demonstrated as SERS active substrates with a high enhancement factor (up to 10 8 ). 152 153 However, the fabrica tion process involves multiple steps (e.g., metal deposition, mask liftoff, and KOH anisotropic etch), and each step could introduce some defects and uncertainties. The accumulation of defects results in poor reproducibility and large variation (>100%) in SERS enhancement over wafer scale areas. Here, we report a much simpler templating technology that only requires a single metal deposition step for creating SERS substrates with high reproducibility over large areas. The new methodology leverages the hig h uniformity of the spin coated colloidal arrays and well established physical vapor deposition techniques. 4 .2. 1 Experimental Procedure The fabrication of wafer scale, non close d packed silica colloida l crystal polymer nanocomposite is performed by followin g the established spin coating procedures. In short, St o ber silica colloids are first dispersed in an ETPTA monomer (with 2 wt % Darocur 1173 photoinitiator) to make a final particle volume fraction of 20%. The colloidal suspension is disposed on an APTCS primed silicon wafer and spin coated at 300 rpm for 1 min, 800 rpm for 1 min, 1500 rpm for 20 s, 3000 rpm for 20 s, and 7000 rpm for 2 min. The ETPTA monomer is rapidly polymerized for 12 s by using a pulsed UV curing system. Two nanometer chromium a nd 18 nm gold are finally deposited on the surface of the colloidal crystal polymer nanoc omposite by electron beam evaporation at a typical deposition rate of 0.1 nm/s from graphite crucibles at 2.5x 10 6 mbar.
72 A calibrated halogen light source is used to i lluminate the sample for reflection measurement The beam spot size is about 3 mm on the sample surface. Measurements are performed at normal incidence, and the cone angle of collection is less than 5. Absolute reflectivity is obtained as a ratio of the s ample spectrum and reference spectrum. The reference spectrum is the optical density obtained from an aluminum sputtered (1000 nm thickness) silicon wafer. The gold coated nanocomposite is immersed in a 5 mM solution of benzenethiol in 200 proof ethanol fo r 2 days and then dried in air for 20 min. Raman spectra are obtained using a 50 X objective and a 785 nm diode laser at 0.5 mW with an integration time of 10 s and a 40 m 2 spot size. In the finite element method (FEM) model, 154 we suppose that the gold coated nanocomposite is placed horizontally, so that the interface between the substrate and the medium (air) is parallel to the xz plane while the hemispherical protrusions are along the y axis. The total electromagnetic fields = + and = + and scatter (gold) domains 2 = 0 2 = 0 where and are domain dependent permittivity and permeability, respectively, and is the light frequency. Continuity of tangential components of and across the interface between air and gold leads to the following boundary conditions with n as the interfacial normal vector 1 2 = 0 1 2 = 0 (2) (1)
73 We then employ FEM under Comsol Multiphysics environment to solve eqs 1 and 2 for each substance (air and gold). Ten boundary layers around the medium and scatter domains are artificially construct (PML) boundary approach is then used to define the boundary conditions for the bounded domains in order to obtain high resolution numerical solutions. 155 The electronic and magnetic conductivity of each boundary layer can be set artificially so that little or no electromagnetic radiation will be refl ected back into the domain of scatter. To simulate electromagnetic fields in the newly augmented domains, we solved Maxwell eq 1 in all subdomains. The boundary condition (eq 2) still holds for all internal interfaces. As to the outer boundaries of the PML layers, a low reflection boundary condition (eq 3) is used to minimize residual reflectio n and attenuate the wave quickly within the layers = 0 After solving Maxwell eq 1 together with boundary conditions (eqs 2 and 3), the two di mensional electric field can be used to calculate the order of magnitude of the Raman enhancement factor [ log ] as 156 = 0 4 whe re and 0 are the electric field amplitude at locat ion and the incident field, respectively. 4.2.2 Results and Discussion Contrary to traditional colloidal self assembly technologies, which usually take days or even weeks to assemble centimeter sized colloidal crystals as templates for making SERS su bstrates, the spin coating technology is rapid and scalable. We have demonstrated that wafer sized (up to 8 in. diameter) colloidal crystals can be fabricated (3)
74 in minutes. In this methodology, monodispersed silica particles with diameter ranging from tens o f nanometers to several micrometers are dispersed in a nonvolatile ETPTA monomer and then shear aligned to form highly ordered colloidal crystals by using standard spin coating equipment. After photopolymerization of ETPTA monomers, the colloidal arrays ar e embedded in a polymer matrix, and the spheres of the top layer protrude out of the film, forming a periodic surface. The resulting colloidal crystal polymer nanocomposites exhibit highly uniform thickness and remarkably large single crystalline domain si ze over the whole wafer area. Another unique property of the shear aligned colloidal crystal is the unusual non close d packed structure. The interparticle distance between neighboring colloids is fixed at 1.4 times of the particle diameter. The SERS active substrate is fabricated by subsequent deposition of a 2 nm layer of chromium and an 18 nm layer of gold on the spin coated nanocomposite by electron beam evaporation. Figure 4 7 shows a photograph of a metalized nanocomposite consisting of 320 nm silica spheres on a 4 in. silicon wafer illuminated with white light. The nanocomposite film is prepared by spin coating a colloidal suspension at 7000 rpm for 2 min, having a thickness of 3 monolayers. The metalized sample exhibits a brilliant six arm diffraction star with exact 60 angles between neighboring arms. This pattern is caused by Bragg diffraction of visible light by the hexagonally ordered colloidal arrays. Some defects (e.g., comets) caused by large airborne solid particles are apparent on the wafer. Many of these defects can be avoided by conducting spin coating in a cleanroom.
75 Long range hexagonal ordering and non close d packing of the metalized colloidal crystal are evident in the typical top view SEM image as shown in Figure 4 8 A. Exten sive SEM examination reveals that periodic colloidal arrays with similar crystalline structure and quality uniformly cover the whole wafer surface. Interestingly, the magnified SEM image in Figure 4 8 B illustrates that the gold coatin g on the nanocomposite is rough and gold islands of tens of nanometer scale size and sub 10 nm gaps are clearly evident. The side view SEM image in Figure 4 8 C further confirms the granular microstructure of the deposited gold layer, and the protrusion depths of the spheres are measured to be 80 nm. Although island type and discontinuous gold and silver films have been extensively studied for a wide range of applications such as molecular electronics and SERS 157 162 the formation of gold islands on the colloidal crystal poly mer nanocomposite is still somewhat unusual. As demonstrated in early studies, only when the metal film is thin (<10 nm nominal thickness for gold and <20 nm nominal thickness for silver), the surface condensation and nucleation of evaporated metal can ind uce the formation of island type films. 159,161 However, the thickness of the evaporated gold on the spin coated nanocomposite is 18 nm, which is above the threshold for the formation of discontinuous films. 163 Control experiments show that evaporated gold of similar thickness forms continuous films on flat glass substrates. It is known that a thin ( 80 nm) po lymer wetting layer uniformly covers the protruded spheres of the spin coated nanocomposite. To evaluate the effect of this polymer layer on the formation of gold islands, we selectively remove the polymer layer on the protruded silica spheres by a brief oxygen plasma etch (40 mTorr oxygen pressure, 40 SCCM flow rate, and 100 W
76 for 15, 30, and 45 s) and then deposit chromium and gold with similar thicknesses as the nanocomposite sample. AFM images show that the resulting metal films are continuous and smoo th. This indicates that the polymer wetting layer plays a crucial role in the formation of gold islands during the evaporation of a metal film. We have also compared the AFM images and corresponding depth profiles of the metalized nanocomposite and the sam e sample after wet etching the gold and chromium layers in Figure 4 9 The metalized sample exhibits a rough and granular surface, and the root mean squared surface roughness ( R rms ) is measured to be 3. 162 After removing the metal coating, the nanocomposite surface is smooth and the roughness is reduced to 1.05, which is almost identical to that of a newly spin coated sample. This suggests that the formation of gold islands is not because of the surface buckling of the polymer nanocomposite caused by the deposition of metals. The depth profiles in panels B and D of Figure 4 9 show that the protrusion depths of the silica sphere s retain after metal evaporation, indicating conformal deposition of metals on the surface of the nanocomposite. Although the underlying mechanism for the formation of gold islands during evaporation of relatively thick metals has yet to be fully understood and is still under investigation, the creation of discontinuous metal films with periodic microstructures over wafer scale areas could find important technological applications in nanoelectronics, electromechanical devices (e.g., strain gauge), and biosensors. 157 159 We focus on the application of these structured metal films as reproducible SERS substrate s. Figure 4 10 compares the SER spectra of benzenethiol molecules adsorbed on a continuous gold control sample and a n 18 nm gold coated nanocomposite consisting of 400 nm silica spheres. Benzenethiol is chosen as the model molecule
77 because of its ability t o assemble into dense monolayer on gold and its large Raman cross section. 151 The periodic, island type gold film shows strong and distinctive SERS peaks, whose positions and relative amplitude match with those in the literature for benzenethiol molecules ad sorbed on structured gold surfaces, 151 while the flat gold control sample does not display clear SERS signal. The assignment of the SERS peaks to different vibrational modes is shown in Table 4 1. The SERS enhancement factor, G is calculated using the metho d described in the literature by comparing the Raman intensity (after baseline subtraction) for various vibrational peaks obtained for a pure benzenethiol liquid ( ) and at the metalized nanocomposite surface ( ) as 164 = where is 9.8 M, is the gold surface area covered by benzenethiol molecules (0.45 nmolcm 2 ), 150 h is the confocal length of the spectrometer ( 88 m determined for the 50 X objective used in the experiment), and is the surface roughness ( 3.5 measured by AFM). We have systematically investigated the reproducibility of SERS enhancement over 4 in. diameter samples consisting of 320 and 400 nm silica spheres. Four concentric rings with radii of 0.5, 1.6, 2.7, and 3.8 cm separate the 4 in. wafers into five regions designated as R0, R1, R2, R3, and R4, respectively. In each region, we have randomly obtained at least 10 SER spectra, and the average Raman counts and corresponding standa rd deviation for different SERS peaks are listed in Table 1. From the Table 4 1, it is evident that the SERS enhancement is reproducible from place to place within each region. We have also calculated the standard deviation of the averaged Raman counts for each vibrational mode across the 4 in. wafer. The results
78 are 19.4% (1575 cm 1 ), 16.2% (1074 cm 1 ), 14.2% (1023 cm 1 ), 22.0% (1000 cm 1 ), 13.8% (695 cm 1 ), and 15.5% (419 cm 1 ) for the 320 nm sample and 22.1% (1575 cm 1 ), 20.9% (1074 cm 1 ), 25.9% (1023 cm 1 ), 23.0% (1000 cm 1 ), 24.7% (695 cm 1 ), and 27.6% (419 cm 1 ) for the 400 nm sample. We therefore conclude that the templated substrates exhibit high SERS reproducibility with less than 28% standard deviation over a 4 in. wafer surface. From the SERS mea surements, we have also found that the peak position for any specific vibrational mode is almost identical with less than 1 nm variation from place to place on the 320 and 400 nm samples. Therefore, we arithmetically average the SER spectra from the five r egions of the two 4 in. samples and show the results in Figure 4 11 We have also calculated the SERS enhancement factors using the averaged Raman intensity for different peaks, and the results are listed in Table 4 2. Because of a high scattering background for pure benzenethiol, the peaks at 1575 and 1074 cm 1 are beyond the detection limit of the spectrometer. Therefore, we only list SERS enhancement factors for other four peaks in Table 4 2. It is apparent from Table 4 2 that a SERS enhancement factor on the order of 10 7 can be achieved, and the 400 nm sample exhibits a slight higher enhancement. It is also interesting to notice that for 320 nm silica spheres the enhancement factor tends to increase from the center of the wafer (R0 region) to the edge (R4 region), while this trend is reversed for 400 nm silica spheres. We speculate that the crystalline parameters (e.g., interparticle separation, particle protrusion depth, and single crystalline domain size) could be slightly different from center to edge due to the nonuniform shear stress during the spin coating process.
79 It is well known that SERS enhancement is sensitive to the structu ral parameters of the samples. To help understand the enhancement mechanism, we have conducted optical reflection measurements at normal incidence and finite element electromagnetic modeling. It is well known that localized (Mie scattering based, surrounding metal nanoparticles) 43,165 and delocalized (Bragg scattering based, covering micrometer scale areas) 166 167 surface plasmons play important roles in determining the amplitude of SERS enhancement. Bartlett et al. have demonstrated that periodically structured metallic nanovoids prepared by electrochemical deposition exhibit strong and reproducible SERS enhancement, even though the metal films are not rough. 150 151 They have also shown that greatest SERS enhancement occurs when surface plas mon resonances on structured metallic surfaces are present at the excitation wavelength and Raman scattered wavelength. 150 151 Compared to other metal island SERS substrates 157 160,162 which t ypically only exhibit localized surface plasmon resonance, our periodic metal films could support localized and delocalized surface plasmons. The former is originated from the metal islands, while the latter is caused by the Bragg scattering from the perio dic structure whose lattice constant matches with the wavelength of operating light. To evaluate the surface plasmon resonance of metalized nanocomposites, we measured optical reflection at normal incidence. Figure 4 1 2 shows the reflection spectra obtaine d at eight random locations on a metalized nanocomposite consisting of 400 nm silica spheres. The position of the laser excitation wavelength, 785 nm, is also indicated by the dashed line. The absorbance valleys (peaked at 600 and 800 nm) in the reflection spectra could be attributed to the interference of the incident light with
80 delocalized and localized surface plasmons as well as the colloidal multilayers. 166 It i s evident that the position of the excitation laser almost coincides with the absorbance valley located at 800 nm. This could result in high SERS enhancement as shown in Figure 4 11 B. To further evaluate the contribution of the delocalized surface plasmo ns to the overall SERS enhancement, we calculated the electric field amplitude distribution and corresponding Raman enhancement factors surrounding arrays of gold hemispherical protrusions using the Comsol Multiphysics software. 154 Because the periodic array is symmetric, we construct a simplified two dimensional model which can be considered as sections through a three dimensional array at the point of maximal enhancement (Figure 4 1 3 A). Figure 4 1 3 A shows the calculated distribution of a SERS enhancement factor around two adjacent hemispherical protrusions with a templating sphere diameter of 320 nm and interprotrusion distance of 1.4 x 320 nm. The simulation results show that the maximal SERS e nhancement factors occur at the top of the semicircles. The spatial distribution of the enhancement factors around th e two semicircles is asymmetric, indicating strong electromagnetic interaction between the neighboring scatters. Figure 4 1 3 B illustrates that a larger array (12 semicircles) results in higher enhancement ( 10 4.6 ), and Figure 4 1 3 C demonstrates that the maximal enhancement factor reaches a plateau when the array has more than 12 scatters. In real SERS experiments, the laser spot ( 40 m 2 ) co vers 250 protrusions. It should be noted that the current electromagnetic modeling represents a significant simplification of the real case as the contributions from the localized surface plasmons caused by isolated gold islands and the charge transfer me chanism, which arises from the electronic interaction
81 between the adsorbed molecules and metal surface, 168 are not being considered. This could explain the large discrepancy between the experimental and calculated SERS enhancement. Indeed, the continuous and smooth gold films deposited on the oxygen plasm a etched nanocomposites exhibit a much lower SERS enhancement factor (<10 5 ) than that of the discontinuous films. We have developed a simple and scalable bottom up approach for fabricating periodically structured surfaces that can serve as substrates for depositing gold island films with reproducible SERS enhancement over wafer sized areas. Figure 4 1. Schematic illustration of the templating procedures for making periodic metallic nanopyramid arrays. Silica monomer Spin coat; UV cure RIE Sputter Cu Remove silica Wet etch; Cr etch Sputter Au Peel of Au
82 Figure 4 2. Photograph of a gold coated colloidal crystal polymer nanocomposite consisting of 320 nm silica spheres on a 4 in. silicon wafer illuminated with white light. Figure 4 3. Inverted pyramid arrays anisotropically etched in (100) silicon wafer. A) SEM image of a sample etch at 60 C for 120s. The inset showing a Fourier transform of the image. B) SEM image of a sample etch at 60 C for 420s. The inset showing a Fourier transform of the image. C) Average size and size distribution of inverted pyramids etched at 60 C for various durations. B A A B C A B A B C Pyramid Size (nm) Etch Time(s)
83 Figure 4 4. SEM image of particle s precipitation on KOH etched silicon. Figure 4 5 Pe riodic gold nanopyramid array. A) Photograph of a 4 in. gold nanopyramid array templated from 320 nm silica spheres illuminated with white light. B) Top view SEM image of a replicated gold nanopyramid array The inset showing a magnified SEM image of a sin gle nanopyramid. C) Side view SEM image of the sam e gold nanopyramid array as in B A magnified pyramid is shown in the inset. A B C
84 Figure 4 6 Raman spectra of benzenethiol adsorbed on a gold nanopyramid arrays (red curve) replicated from a silicon inverted pyramid array that is templated from 320 nm silica spheres and etched at 60 C for 210s and a flat gold control sample (blue curve). Figure 4 7 Photograph of a gold coated colloidal crystal polymer nanocomposite consisting of 320 nm silica spheres on a 4 in. silicon wafer illuminated with white light. SERS Intensity/Counts Raman Shift (cm 1 )
85 Figure 4 8 Scanning electron microscope (SEM) images of the gold coated nanoc omposite as shown in Figure 4 7 A) Top view. B) Magnified top view. C) Tilted view (45). Figure 4 9 Tapping mode atomic force microscope (AFM) images and corresponding depth profiles. (A and B) Metalized nanocomposite consisting of 320 nm silica spheres. (C and D) Same sample afte r dissolving the metal coating. A B C A B C D
86 Figure 4 10 SER spectra obtained on a gold coated nanocomposite consisting of 400 nm silica spheres (red) and a flat gold control sample on glass (black). The spectra were taken using a 785 nm diode laser at 0.5 mW with an integration time of 10s. Table 4 1. Assignment of SERS p eaks and c orresponding a verage c ounts a peak (cm 1 ) assignment region Raman counts 320nm 400nm 1575 a1, (C C C) R0 3099698 (22.5%) 74601799 (24.1%) R1 3606862 (23.9%) 70601542 (21.8%) R2 3370772 (22.9%) 5162776 (15%) R3 4499590 (13.1%) 5521901 (16.3%) R4 4853624 (12.9%) 4368715 (16.4%) 1074 a1, (C C C) and (C S) R0 55051208 (21.9%) 10850704 (6.5%) R1 64241282 (20%) 119691863 (15.6%) R2 68411135 (16.6%) 81351414 (17.4%) R3 8170999 (12.2%) 82801155 (13.9%) R4 81031241 (15.3%) 7489971 (13%) 1023 a1, (C H) R0 3480841 (24.2%) 7604850 (11.2%) R1 4016969 (24.1%) 77121261 (16.4%) R2 4270811 (19%) 51231032 (20.1%) R3 4939668 (13.5%) 5141904 (17.6%) R4 4890920 (18.8%) 4389729 (16.6%) 1000 a1, (C C C) R0 41001000 (24.4%) 9164897 (9.8%) R1 49101177 (24%) 88011397 (15.9%) R2 5183918 (17.7%) 63451123 (17.7%) R3 6222774 (12.4%) 63811004 (15.7%) R4 72211136 (15.7%) 5409792 (14.6%)
87 Table 4 1. Continued peak (cm 1 ) assignment region Raman counts 320nm 400nm 695 a1, (C C C) and (C S) R0 1112197 (17.7%) 2179458 (21%) R1 1207301 (24.9%) 2190552 (25.2%) R2 1180267 (22.6%) 1607343 (21.3%) R3 1337223 (16.7%) 1222213 (17.4%) R4 1560381 (24.4%) 1500330 (22%) 419 a1, (C C C) and (C S) R0 39701041 (26.2%) 81421815 (22.3%) R1 4972970 (19.5%) 80671135 (14.1%) R2 5354847 (15.8%) 57771312 (22.7%) R3 60761010 (16.6%) 3860888 (23%) R4 56971286 (22.6%) 6356749 (11.8%) a Peaks and Raman counts recorded for benzenethiol molecules adsorbed on five areas (R0 R4) of 4 in. nanocomposites consisting of 320 and 400nm silica spheres. Figure 4 11 Arithmetically averaged SER spectra recorded for benzenethiol molecules adsorbed on five areas (R0 R4 ) of 4 in nanocomposites consisting of A) 320 and B) 400 nm silica spheres. Spectra were taken using a 785 nm diode laser at 0.5 mW with an integration time of 10s. Table 4 2 Assignment of SERS p eaks and Raman e nhancement f actors peak/(cm 1 ) assignment region enhancement factor 320nm 400nm 1023 a1, (C H) R0 2.210 7 4.810 7 R1 2.610 7 4.910 7 R2 2.710 7 3.310 7 R3 3.210 7 3.310 7 R4 3.110 7 2.810 7 A B
88 Table 4.2 Continued. peak/(cm 1 ) assignment region enhancement factor 320nm 400nm 1023 a1, (C H) R0 2.210 7 4.810 7 1000 a1, (C C C) R0 6.510 6 1.510 7 R1 7.810 6 1.410 7 R2 8.210 6 1.010 7 R3 9.810 6 1.010 7 R4 1.110 7 8.610 6 695 a1, (C C C) and (C S) R0 1.110 7 2.110 7 R1 1.210 7 2.110 7 R2 1.110 7 1.5107 R3 1.310 7 1.2107 R4 1.510 7 1.4107 419 a1, (C C C) and (C S) R0 4.910 7 1.010 8 R1 6.210 7 1.010 8 R2 6.610 7 7.210 7 R3 7.510 7 4.810 7 R4 7.110 7 7.910 7 Peaks and Raman enhancement factors were calculated for benzenethiol molecules adsorbed on five areas (R0 R4) of 4 in. nanocomposites consisting of 320 and 400nm silica spheres. Figure 4 1 2 Normal incidence reflection spectra obtained at eight locations on a 4 in. metalized nanocomposite consisting of 400nm silica spheres. Reflection (%) Wavelength (nm)
89 Figure 4 1 3 Simulated Raman enhancement around gold semispherical protrusions 785 nm. A) 2 scatters. B) 12 scatters. C) Simulated order of magnitude of maximal SERS enhancement factor (log Gmax) versus number of scatters. C A B
90 CHAPTER 5 TEMPLATE D FABRICATION OF SUB 100NM PERIODIC NANOS TRUCT URES Subwavelength periodic met al nanohole arrays are of great scientific interest and considerable technological importance in developing miniaturized optical devices for all optical integrated circuits (e.g., optical waveguides, switches, and couplers), 36,169 171 information storage, 172 efficient light emitting diodes, 173 and biosensors, 77 as well as fundamental understanding of surface plasmon enabled optical transmission and diffraction. 74 Zero mode waveguiding through arrays of sub 100 nm nanoholes perforated in a metal film has also been demonstrated as a simple and highly parallel method for single molecule bioanalysis at high concentrations. 75,174 Unfortunately, these important applications have been greatly impeded by the expensive and complex nanofabrication techniques (e.g., electronbeam, focused ion beam, and interference lithography) in creating subwavelength nan ohole arrays. 74,84,174 Attaining high throughput and large area fab rication continues to be a major challenge with these top down techniques. Bottom up self assembly and templated nanofabrication provide a much simpler, faster, and inexpensive alternative to nanolithography in generating periodic nanostructures. 15,28,85,175 For instance, self assembled monolayer or multilayer colloidal crystals are used as templates to create a large variety of functional periodic nanostructures including nanohole arrays. 46 47,145,176 However, most of the available bottom up approaches suffer from low throughput and incompatibility with mature micr ofabrication, limiting mass production and on chip integration of practical devices. Furthermore, only close packed colloidal crystals are available through traditional self
91 assembly, whereas non close packed crystals are needed to template nanohole arrays 12,86,177 Although the spin coat technology has been demonstrated to generate wafer scale non close packed 2D and 3D arrays, 2D submicrometer scale particles ( <100 nm ) have not been achieved reproducibly in large area by spin coat technology. Here we report a flexibl e and scalable non lithographic approach for fabricating large area sub 100 nm periodic na nostructures using submicrometer sized particles as templates. Th e resulting nanofeature size can be easily adjusted by controlling the templating conditions. Sub 100 nm patterns, whose size is only ~ 20% of the original templating sphere diameter, can be fabricated over an 80 cm 2 area. 5.1 Experimental Procedure Figure 5 1 shows a schematic outline of the new templating procedure for making periodic metal nanohole arrays. The established spin coating technique is utilized to create a monolayer of non close packed silica colloidal crystals embedded in a thin layer of poly mer matrix as structural templates. In the spin coatin g technique, monodisperse silica particles are disper sed in non volatile ethoxylated trimethylolpropane triacrylate (ETPTA, Sartomer) monomers to form stable dispersions with a particle volume fraction of ~ 20%. The dispersions are then spin coated on planar substrates (e.g., glass and silicon wafer) using standard spin coating equipment. Both multilayer and monolayer colloidal crystals with non close packed structures can be formed by simply tuning the s pin coating speed and time. The ETPTA monomers can then be polymerized by exposure to ultraviolet radiation to immobilize the shear aligned particle arrays.
92 The resulting colloidal crystal polymer nanocomposite has a thin polymer wetting layer ( ~ 100 nm thick) between the sil ica monolayer and the substrate. The current templating technique is aimed at utilizing this polymer wetting layer to transfer the hexagonal patterns of spin coated colloidal monolayers consisting of submicrometre sized particles into arrays of nanoholes with a sub 100 nm diameter (Fig ure 5 1). In this approach, conventional oxygen plasma etching is firstly applied to partially remove the polymer matrix between the non close packed colloidal arrays. Silica particles function as etching masks during the oxygen plasma etching process. They protect the polymer wetting layer immediately underneath them to form polymer posts whose size can be easily adjusted by controlling the etching conditions. The templating silica particles can then be r emoved by dissolving in a 2% hydrofluoric acid aqueous solution for 2 min. A thin layer of metal can finally be deposited between the exposed polymer posts, followed by lifting off polymer templates to form hexagonal arrays of metal nanoholes with tunable size. Figure 5 2A, B and C show top vi ew scanning electron microscope (SEM) images of the templated polymer posts after oxygen plasma etching operating at 40 mTorr pressure, 40 sccm flow rate, and 100W on a Unaxis Shuttlelock RIE/ICP reactive ion etcher fo r 2, 6, and 8 min, respectively, followed by hydrofluoric acid wash to remove the templating silica particles. It is evident that the resulting polymer posts retain the long range hexagonal ordering and the inter particle distance of the original non close packed colloidal monolayer consisting of 360 nm silica spheres. The cross sections of the posts are round shaped for all three above conditions, indicating uniform etching of ETPTA from the peripheries of the templated posts.
93 It is also apparent fro m Fig u re 5 2 A C that longer etching duration leads to smaller polymer posts. We determine the size and size distribution of the templated posts by averaging over 100 posts using image analysis software (Scion Image). Fi gure 5 2D shows the dependence of the size of the templated posts on the oxygen plasma etching time. The measured size standard deviation is around 1 0%, which is a little bit worse than that of the templating silica spheres ( ~ 5 7%). Importantly, much smaller polymer posts (as small as 55 nm) than the original templating silica spheres (360 nm) can be achieved. This represents ~ 85% shrinkage of the template size. When longer plasma etching duration (49 min) is applied, the polymer posts become too narrow to support the silica particles, resulting in the collapse of the particles and subsequent complete etching of the polymer posts. After removing the silica parti cles by hydrofluoric acid wash, the polymer posts can be used as sacrificial deposition masks to template metal nanohole arrays with tunable size. A thin layer of metal is deposited by traditional physical vapor deposition techniques (e.g., electron beam evaporation and sputtering) to fill up the spacing between the non close packed polymer posts. As the adhesion between the polymer and the su bstrate is quite low, the posts can be easily removed by a gentle sweeping using a cleanroom Q tip under flowing water, resulting in the formation of wafer scale periodic nanohole arrays. Fig ure 5 3 shows a SEM image of a templa ted gold nanohole array with 89 13 nm diameter. The nanoholes keep the same long range hexagonal ordering and inter particle distance as the original monolayer colloidal crystal template. Besides providing a much simple r approach than nanolithography for generating sub 100 nm nanoho le arrays, the new templating nanofabrication approach enables
94 creation of complex micropatterns that are not easily available by traditional top down techniques. Fig ure 5 4 A and B show top and side view SEM images of arrays of from spin coated colloidal monolayers. In this approach, non close packed silica particles are firstly used as etching masks during oxygen plasma etching to pattern polymer posts underneath them as described above. The released silica spheres can then be utilized as etching masks for the second time during chlorine reactive ion etching (RIE, 5 mTorr pressure, 20 SCCM chlorine flow rate, and 80 W) to generate ordered arrays of silicon pillars. The polymer posts are not affected by the chlorine RIE process, the templated silicon pillars after removing the templating silica particles by a brief hydrofluoric acid wash. 5.2 Results and Discussion The height of the polymer po sts can also affect the spatial distribution of the re active ions during the above particle masked RIE process, resulting in the formation of unusual (GaSb) substrate as shown in Fig ure 5 4 C and D. Due to different substrate wettability the polyme r wetting layer on a GaSb wa fer (~ 500 nm thick) is much thicker than that on a silicon or a glass substrate (~ 100 nm). The shadowing effect created by the templating silica spheres is then significantly affected by the different polymer spacer height. On a silicon wafer, the templating silica particles are close to the substrate surface, resulting in effective shadowing of the regions underneath the particles. This leads to the formation of pillars with vertical sidewalls and almost the same diameter as tha t of silica spheres. While for a GaSb substrate under similar chlorine RIE conditions, the tall polymer posts between the substrate and the templating particles change the spatial distribution of the reactive chlorine ions, leading to the formation of tape red pillars with flat circular caps
95 (Figure 5 4D). The cap size is similar to that of the templating silica spheres. GaSb pillars with a high aspect ratio and vertical sidewalls resembling the silicon pillars as shown in Figure 5 4B have previously been fa bricated using lithographically patterned photoresist as an etching mask under similar chlorine RIE conditions as described above. 178 This verifies the importance of the polymer posts in controlling the shape of the templated microstructures during the part icle masked dry etching process In summary, we have developed a simple yet scalable templating nanofabrication technique that enabl es the creation of periodic nanostructures with tunable feature size. Figure 5 1. Schematic illustration of the fabrication procedures for making periodic arrays of metal nanoholes with tuna ble size by using monolayer non close packed colloidal crysta l polymer nanocom posite s as templates
96 Figure 5 2. Templated polymer posts with adjustable size. A C) SEM images of polymer posts with diameters of 20210 nm, 787 nm, and 7010 nm after 2, 6, and 8 min oxygen plasma etching, respectively. D) Size dependence of the templated polymer posts on the oxygen plasma etching time. 360 nm silica spheres are used as the original templates. Figure 5 3. SEM image of an array of gold nanoholes with 89 nm diameter templated from a monol ayer nanocomposite consisting of 360 nm silica spheres. A B C D
97 Figure 5 4 SEM images of micro candles. A) and B) Top and side view SEM images of an array of silicon coated 2 D colloidal crystal. C) and D) Top and side view SEM images of a templated A B C D
98 CHAPTER 6 CONCLUSIONS AND RECO MMENDATIONS This dissertation covers three research topics which demonstrate the important application s of templating nanostructure by sp in coating technology. The conclusions associated with the antireflection coatings presented in Section 6. 1 The conclusions associated with the SERS active substrates are presented in Section 6. 2 The conclusion associated sub 100nm structures are presen ted in section 6. 3 The recommendations for further study are presented in section 6. 4 6.1 Antireflection Coatings Antireflection coatings templated from spin coated colloidal arrays are generated on glass substrates and semiconductor wafers. Bio inspired moth eye structures are generated by surface templating u sing PDMS to transfer nanostructure on glass substrates. Binary array antireflection coatings are fabricated using two layers silica/ETPTA nanocomposite. This dimple and nipple binary nanostructures can further reduced reflection to below 0.5% in the visible light range. By using 2 dimensional non close packed colloidal crystals as etching masks, nanostructures can be generated on Si and GaAs wafer surfaces that greatly reduced the reflection. These cost effective and large area broadband antireflection coatings on Si and GaAs are expected to enhance optical performances of optical devices. 6. 2 SERS Active Substrates Two types of SERS active substrates are fabricated by spin coating technology. Gold nanopyramid arrays with sharp tips that can concentrate electromagnetic waves are used as SERS active substrates with an enhance ment factor of ~7x10 5 Further
99 improvements are made by using gold coated polymer pyramids and electrochemical SERS to increase enhancement factor up to 10 8 A highly reproducible SERS active substrate is demonstrated It leverages the uniformity of spin coated colloidal arrays and conventional p hysical vapor deposition techniques. The formation of discontinuous, island type metal films with periodic microstructures over large areas could lead to important technological applications in nanoelectronics, electromechanical devices, and biosensors. 6.3 Sub 100nm Nanostructures Sub 100 nm metal nanohole arrays, which have important technological applications in integrated nanophotonics and bio microanalysis, can be easily templated over a large area. The polymer wetting layer between the spin coated c olloidal monolayer and the substra te also plays a crucial role in determining the shape of the templated microstructures that have broad applications ranging from plasmonics to broadband antireflection coatings for solar cells. 6.4 Recommendations Future w ork should focus on the colloidal assembly mechanism by shear align ing force. This cost effective spin coat technology is beneficial to both academ ic and industry because it can quickly ( in several minutes) assemble 2 and 3 dimension al colloidal crystal in large scale (up to 8 in. wafer) and is compatible with current microfabrication technology. A better understanding of the formation mechanism could lead to tuning the colloidal structures which could further extend its applications. Another future w ork is to fabricate real solar cells based on the antireflection coati ngs we have developed. Our work ha s shown that the reflection is greatly reduced in the visible range by templating from spin coated made colloidal crystals on both glass
100 and semiconduct or materials. This reduction of refl ection can increase the absorption of photons and hence increase the photon electron conversion efficiency. Light transmission measurement through templated nanohole arrays should also be conducted The combination of e xtraordinary transmissi on and high conductivity makes gold nanohole arrays a potential candidate for transparent electrode s
101 LIST OF REFERENSCES (1) Betzig, E.; Trautman, J. K. Science 1992 257 189. (2) Liu, Z. W.; Wei, Q. H.; Zhang, X. Nano Letters 2005 5 957. (3) Blanco, A.; Chomski, E.; Grabtchak, S.; Ibisate, M.; John, S.; Leonard, S. W.; Lopez, C.; Meseguer, F.; Miguez, H.; Mondia, J. P.; Ozin, G. A.; Toader, O.; van Driel, H. M. Nature 2 000 405 437. (4) Mayoral, R.; Requena, J.; Moya, J. S.; Lopez, C.; Cintas, A.; Miguez, H.; Meseguer, F.; Vazquez, L.; Holgado, M.; Blanco, A. Advanced Materials 1997 9 257. (5) vanBlaaderen, A.; Ruel, R.; Wiltzius, P. Nature 1997 385 321. (6) Yin, Y. D.; Lu, Y.; Gates, B.; Xia, Y. N. Journal of the American Chemical Society 2001 123 8718. (7) Ozin, G. A.; Yang, S. M. Advanced Functional Materials 2001 11 95. (8) Pieranski, P. Contemporary Physics 1983 24 25. (9) Weissman, J. M.; Sunkara, H. B.; Tse, A. S.; Asher, S. A. Science 1996 274 959. (10) Jethmalani, J. M.; Ford, W. T.; Beaucage, G. Langmuir 1997 13 3338. (11) Hayward, R. C.; Saville, D. A.; Aksay, I. A. Nature 2000 404 56. (12) Lumsdon, S. O.; Kaler, E. W.; Williams, J. P.; Velev, O. D. Appl. Phys. Lett. 2003 82 949. (13) Garcia Santamaria, F.; Miyazaki, H. T.; Urquia, A.; Ibisate, M.; Belmonte, M.; Shinya, N.; Meseguer, F.; Lopez, C. Advanced Materials 2002 14 1144. (14) Rec ulusa, S.; Ravaine, S. Chemistry of Materials 2003 15 598. (15) Vlasov, Y. A.; Bo, X. Z.; Sturm, J. C.; Norris, D. J. Nature 2001 414 289. (16) Wong, S.; Kitaev, V.; Ozin, G. A. Journal of the American Chemical Society 2003 125 15589. (17) Mihi, A.; Ocana, M.; Miguez, H. Advanced Materials 2006 18 2244. (18) John, S. Physical Review Letters 1987 58 2486. (19) Yablonovitch, E. Physical Review Letters 1987 58 2059.
102 (20) Joannopoulos, J. D.; Villeneuve, P. R.; Fan, S. H. Nature 1997 386 143. (21) Perpall, M. W.; Perera, K. P. U.; DiMaio, J.; Ballato, J.; Foulger, S. H.; Smith, D. W. Langmuir 2003 19 7153. (22) Scott, R. W. J.; Yang, S. M.; Chabanis, G.; Coombs, N.; Williams, D. E.; Ozin, G. A. Advanced Materials 2001 13 1468. (23) Cassagneau, T.; Caruso, F. Advanced Materials 2002 14 1837. (24) Chen, H. H.; Suzuki, H.; Sato, O.; Gu, Z. Z. Applied Physics a Materials Science & Processing 2005 81 1127. (25) Kamp, U.; Kitaev, V.; von Freymann, G.; Ozin, G. A.; Mabury, S. A. Advance d Materials 2005 17 438. (26) Colombelli, R.; Srinivasan, K.; Troccoli, M.; Painter, O.; Gmachl, C. F.; Tennant, D. M.; Sergent, A. M.; Sivco, D. L.; Cho, A. Y.; Capasso, F. Science 2003 302 1374. (27) Arsenault, A. C.; Puzzo, D. P.; Manners, I.; Ozin, G. A. Nature Photonics 2007 1 468. (28) Ozin, G. A.; Arsenault, A. C. Nanochemistry: A Chemical Approach to Nanomaterials; RSC, Cambridge 2005 (29) Sun, Z. Q.; Yang, B. Nanoscale Research Letters 2006 1 46. (30) Li, Y.; Cai, W.; Duan, G. Chemistry of Materials 2008 20 615. (31) Yan, F.; Goedel, W. A. Nano Letters 2004 4 1193. (32) Wang, X. D.; Lao, C. S.; Graugnard, E.; Summers, C. J.; Wang, Z. L. Nano Letters 2005 5 1784. (33) Walheim, S.; Schaffer, E.; Mlynek, J.; Steiner, U. Science 1999 283 520. (34) Maier, S. A. (35) Homola, J.; Koudela, I.; Yee, S. S. Sensors and Actuators B Chemical 1999 54 16. (36) Barnes, W. L.; Dereux, A.; Ebbesen, T. W. Nature 2003 424 824. (37) Reilly, T. H.; van de Lagemaat, J.; Tenent, R. C.; Morfa, A. J.; Row len, K. L. Appl. Phys. Lett. 2008 92 (38) Johnsson, B.; Lofas, S.; Lindquist, G. Analytical Biochemistry 1991 198 268.
103 (39) Lofas, S.; Johnsson, B. Journal of the Chemical Society Chemical Communications 1990 1526. (40) Gordon, R.; Sinton, D.; Kavanag h, K. L.; Brolo, A. G. Accounts of Chemical Research 2008 41 1049. (41) Jeanmaire, D. L.; Vanduyne, R. P. Journal of Electroanalytical Chemistry 1977 84 1. (42) Campion, A.; Kambhampati, P. Chemical Society Reviews 1998 27 241. (43) Kneipp, K.; Kneip p, H.; Kneipp, J. Accounts of Chemical Research 2006 39 443. (44) Vlckova, B.; Gu, X. J.; Tsai, D. P.; Moskovits, M. Journal of Physical Chemistry 1996 100 3169. (45) Madou, M. J. Fundamentals of Microfabrication: The science of Miniaturization; CRC, Boca Raton, FL, 2002 2002 (46) Hulteen, J. C.; Vanduyne, R. P. Journal of Vacuum Science & Technology a Vacuum Surfaces and Films 1995 13 1553. (47) Kosiorek, A.; Kandulski, W.; Chudzinski, P.; Kempa, K.; Giersig, M. Nano Letters 2004 4 1359. (48) Xu, Y.; Schneider, G. J.; Wetzel, E. D.; Prather, D. W. Journal of Microlithography Microfabrication and Microsystems 2004 3 168. (49) Hoffman, R. L. Transactions of the Society of Rheology 1972 16 155. (50) Ruhl, T.; Spahn, P.; Winkler, H.; Hellmann, G. P. Macromolecular Chemistry and Physics 2004 205 1385. (51) Ackerson, B. J. Journal of Physics Condensed Matter 1990 2 SA389. (52) Ackerson, B. J.; Clark, N. A. Physical Review Letters 1981 46 123. (53) Ackerson, B. J.; Hayter, J. B.; Clark, N. A.; Cotter, L. Journal of Chemical Physics 1986 84 2344. (54) Ackerson, B. J.; Pusey, P. N. Physical Review Letters 1988 61 1033. (55) Jeong, K. H.; Kim, J.; Lee, L. P. Science 2006 312 557. (56) Vukusic, P.; Sambles, J. R. Nature 2003 424 852. (57) Aizenberg, J.; Muller, D. A.; Grazul, J. L.; Hamann, D. R. Science 2003 299 1205.
104 (58) Potyrailo, R. A.; Ghiradella, H.; Vertiatchikh, A.; Dovidenko, K.; Cournoyer, J. R.; Olson, E. Nature Photonics 2007 1 123. (59) Sun, T. L.; Feng, L.; Gao, X. F.; Jiang, L. Accounts of Chemical Research 2005 38 644. (60) Stavenga, D. G.; Foletti, S.; Palasantzas, G.; Arikawa, K. Proceedings of the Royal Society B Biological Sciences 2006 273 661. (61) Hiller, J.; Mendelsohn, J. D.; Rubner, M. F. Nature Materials 2002 1 59. (62) Ibn Elhaj, M.; Schadt, M. Nature 2001 410 796. (63) Xi, J. Q.; Schubert, M. F.; Kim, J. K.; Schubert, E. F.; Chen, M. F.; Lin, S. Y.; Liu, W.; Smart, J. A. Nature Photonics 2007 1 176. (64) Prevo, B. G.; Hon, E. W.; Velev, O. D. Journal of Materials Chemistry 2007 17 791. (65) Clapham, P. B.; Hutley, M. C. Nature 1973 244 281. (66) Shiu, J. Y.; Kuo, C. W.; Chen, P. L.; Mou, C. Y. Chemistry of Materials 2004 16 561. (67) Prevo, B. G.; Hwang, Y.; Velev, O. D. Chemistry of Materials 2005 17 3642. (68) Jiang, P.; Prasad, T.; McFarland, M. J.; Colvin, V. L. Appl. Phys. Lett. 2006 89 (69) Macleod, H. A. Thin Film Optical Filters 3rd ed; Institute of Physics, Bristol 2001 40. (70) Jiang, P.; McFarland, M. J. Journal of the American Chemical Society 2004 126 13778. (71) Kalsin, A. M.; Fialkowski, M.; Paszewski, M.; Smoukov, S. K.; Bishop, K. J. M.; Grzybowski, B. A. Science 2006 312 420. (72) Redl, F. X.; Cho, K. S.; Mur ray, C. B.; O'Brien, S. Nature 2003 423 968. (73) Zeng, H.; Li, J.; Liu, J. P.; Wang, Z. L.; Sun, S. H. Nature 2002 420 395. (74) Ebbesen, T. W.; Lezec, H. J.; Ghaemi, H. F.; Thio, T.; Wolff, P. A. Nature 1998 391 667. (75) Genet, C.; Ebbesen, T. W. Nature 2007 445 39. (76) Matsui, T.; Agrawal, A.; Nahata, A.; Vardeny, Z. V. Nature 2007 446 517. (77) Lesuffleur, A.; Im, H.; Lindquist, N. C.; Oh, S. H. Appl. Phys. Lett. 2007 90
105 (78) Liu, H. T.; Lalanne, P. Nature 2008 452 728. (79) Strelniker, Y. M. Physical Review B 2007 76 (80) Tuteja, A.; Choi, W.; Ma, M. L.; Mabry, J. M.; Mazzella, S. A.; Rutledge, G. C.; McKinley, G. H.; Cohen, R. E. Science 2007 318 1618. (81) Zhang, G.; Wang, D. Y.; Mohwald, H. Nano Letters 2007 7 3410. (82) Velikov, K. P.; Christova, C. G.; Dullens, R. P. A.; van Blaaderen, A. Science 2002 296 106. (83) Koo, H. Y.; Yi, D. K.; Yoo, S. J.; Kim, D. Y. Advanced Materials 2004 16 274. (84) Henzie, J.; Lee, M. H.; Odom, T. W. Nature Nanotechnology 2007 2 549. (85) Choi, D. G.; Yu, H. K.; Jang, S. G.; Yang, S. M. Journal of the American Chemical Society 2004 126 7019. (86) Jang, S. G.; Yu, H. K.; Choi, D. G.; Yang, S. M. Chemistry of Materials 2006 18 6103. (87) Kresge, C. T.; Leonowicz, M. E.; Roth, W J.; Vartuli, J. C.; Beck, J. S. Nature 1992 359 710. (88) Park, M.; Harrison, C.; Chaikin, P. M.; Register, R. A.; Adamson, D. H. Science 1997 276 1401. (89) Zhang, G.; Wang, D. Y. Journal of the American Chemical Society 2008 130 5616. (90) Jiang, P.; Bertone, J. F.; Hwang, K. S.; Colvin, V. L. Chemistry of Materials 1999 11 2132. (91) Aizenberg, J.; Braun, P. V.; Wiltzius, P. Physical Review Letters 2000 84 2997. (92) Jiang, P.; McFarland, M. J. Journal of the American Chemical Society 2005 1 27 3710. (93) McLellan, J. M.; Geissler, M.; Xia, Y. N. Journal of the American Chemical Society 2004 126 10830. (94) Hakanson, U.; Persson, J.; Persson, F.; Svensson, H.; Montelius, L.; Johansson, M. K. J. Nanotechnology 2003 14 675. (95) Xia, Y. N.; Whitesides, G. M. Annual Review of Materials Science 1998 28 153. (96) Moharam, M. G.; Pommet, D. A.; Grann, E. B.; Gaylord, T. K. Journal of the Optical Society of America a Optics Image Science and Vision 1995 12 1077.
10 6 (97) Change, US Dep artment of Energy Report of the Basic Energy Sciences Workshop on Solar Energy Utilization 2005 April, 18 21 (unpublished) (98) Luque, A.; Hegedus, S. 2003 268. (99) Doshi, P.; Jellison, G. E.; Rohatgi, A. Applied Optics 1997 36 7826. (100) Gombert, A .; Glaubitt, W.; Rose, K.; Dreibholz, J.; Blasi, B.; Heinzel, A.; Sporn, D.; Doll, W.; Wittwer, V. Thin Solid Films 1999 351 73. (101) Chen, D. G. Solar Energy Materials and Solar Cells 2001 68 313. (102) Lalanne, P.; Morris, G. M. Nanotechnology 1997 8 53. (103) Aydin, C.; Zaslavsky, A.; Sonek, G. J.; Goldstein, J. Appl. Phys. Lett. 2002 80 2242. (104) Striemer, C. C.; Fauchet, P. M. Appl. Phys. Lett. 2002 81 2980. (105) Kanamori, Y.; Roy, E.; Chen, Y. Microelectronic Engineering 2005 78 79 287. (106) Grann, E. B.; Moharam, M. G.; Pommet, D. A. Journal of the Optical Society of America a Optics Image Science and Vision 1995 12 333. (107) Southwell, W. H. Journal of the Optical Society of America a Optics Image Science and Vision 1991 8 54 9. (108) Zhao, J.; Green, M. A. Ieee Transactions on Electron Devices 1991 38 1925. (109) Kanamori, Y.; Sasaki, M.; Hane, K. Optics Letters 1999 24 1422. (110) Yu, Z. N.; Gao, H.; Wu, W.; Ge, H. X.; Chou, S. Y. Journal of Vacuum Science & Technology B 2003 21 2874. (111) Kanamori, Y.; Hane, K.; Sai, H.; Yugami, H. Appl. Phys. Lett. 2001 78 142. (112) Wu, C. T.; Ko, F. H.; Lin, C. H. Appl. Phys. Lett. 2007 90 (113) Kwak, E. S.; Henzie, J.; Chang, S. H.; Gray, S. K.; Schatz, G. C.; Odom, T. W. Nano Letters 2005 5 1963. (114) Moharam, M. G.; Gaylord, T. K. Journal of the Optical Society of America 1981 71 811. (115) Moharam, M. G.; Gaylord, T. K. Journal of the Optical Society of America a Optics Image Science and Vision 1986 3 1780. (116) Green, M. A.; Keevers, M. J. Progress in Photovoltaics 1995 3 189.
107 (117) Heine, C.; Morf, R. H. Applied Optics 1995 34 2476. (118) Nikolajeff, F.; Lofving, B.; Johansson, M.; Bengtsson, J.; Hard, S.; Heine, C. Applied Optics 2000 39 4842. (119) Smols ki, O. V.; O'Daniel, J. K.; Johnson, E. G. Ieee Photonics Technology Letters 2006 18 1931. (120) Wu, M. C.; Huang, Y. H.; Ho, C. L. Ieee Electron Device Letters 2007 28 797. (121) Oliva, E.; Dimroth, F.; Bett, A. W. Progress in Photovoltaics 2008 16 289. (122) Poortmans, J.; Arkhipov, V. 2006 xvii. (123) Wendt, J. R.; Vawter, G. A.; Smith, R. E.; Warren, M. E. Journal of Vacuum Science & Technology B 1996 14 4096. (124) Huang, Y. F.; Chattopadhyay, S.; Jen, Y. J.; Peng, C. Y.; Liu, T. A.; Hsu, Y. K .; Pan, C. L.; Lo, H. C.; Hsu, C. H.; Chang, Y. H.; Lee, C. S.; Chen, K. H.; Chen, L. C. Nature Nanotechnology 2007 2 770. (125) Lohmuller, T.; Helgert, M.; Sundermann, M.; Brunner, R.; Spatz, J. P. Nano Letters 2008 8 1429. (126) Sun, C. H.; Min, W. L.; Jiang, P. Chemical Communications 2008 3163. (127) Palik, E. D. Handbook of Optical Constants of Solids Academic 1991 (128) Dieringer, J. A.; McFarland, A. D.; Shah, N. C.; Stuart, D. A.; Whitney, A. V.; Yonzon, C. R.; Young, M. A.; Zhang, X. Y.; Van Duyne, R. P. Faraday Discussions 2006 132 9. (129) Baker, G. A.; Moore, D. S. Analytical and Bioanalytical Chemistry 2005 382 1751. (130) Xia, Y. N.; Halas, N. J. Mrs Bulletin 2005 30 338. (131) Lu, Y.; Liu, G. L.; Kim, J.; Mejia, Y. X.; Lee, L. P. Nano Letters 2005 5 119. (132) Chattopadhyay, S.; Chen, L. C.; Chen, K. H. Critical Review Solid State Material Science 2006 31 155418. (133) Watanabe, H.; Ishida, Y.; Hayazawa, N.; Inouye, Y.; Kawata, S. Physical Review B 2004 69 (134) Pettinger, B.; Ren, B.; Picardi, G.; Schuster, R.; Ertl, G. Physical Review Letters 2004 92
108 (135) Love, J. C.; Gates, B. D.; Wolfe, D. B.; Pa ul, K. E.; Whitesides, G. M. Nano Letters 2002 2 891. (136) Charnay, C.; Lee, A.; Man, S. Q.; Moran, C. E.; Radloff, C.; Bradley, R. K.; Halas, N. J. Journal of Physical Chemistry B 2003 107 7327. (137) Jin, R. C.; Cao, Y. C.; Hao, E. C.; Metraux, G. S.; Schatz, G. C.; Mirkin, C. A. Nature 2003 425 487. (138) Aizpurua, J.; Hanarp, P.; Sutherland, D. S.; Kall, M.; Bryant, G. W.; de Abajo, F. J. G. Physical Review Letters 2003 90 (139) Kahl, M.; Voges, E.; Kostrewa, S.; Viets, C.; Hill, W. Sensors and Actuators B Chemical 1998 51 285. (140) Henzie, J.; Barton, J. E.; Stender, C. L.; Odom, T. W. Accounts of Chemical Research 2006 39 249. (141) Henzie, J.; Kwak, E. S.; Odom, T. W. Nano Letters 2005 5 1199. (142) Henzie, J.; Shuford, K. L.; Kwak, E. S.; Schatz, G. C.; Odom, T. W. Journal of Physical Chemistry B 2006 110 14028. (143) Giermann, A. L.; Thompson, C. V. Appl. Phys. Lett. 2005 86 (144) Lu, J.; Chamberlin, D.; Rider, D. A.; Liu, M. Z.; Ma nners, I.; Russell, T. P. Nanotechnology 2006 17 5792. (145) Tessier, P. M.; Velev, O. D.; Kalambur, A. T.; Rabolt, J. F.; Lenhoff, A. M.; Kaler, E. W. Journal of the American Chemical Society 2000 122 9554. (146) Mahajan, S.; Abdelsalam, M.; Suguwara, Y.; Cintra, S.; Russell, A.; Baumberg, J.; Bartlett, P. Physical Chemistry Chemical Physics 2007 9 104. (147) Payne, E. K.; Rosi, N. L.; Xue, C.; Mirkin, C. A. Angewandte Chemie International Edition 2005 44 5064. (148) Nielsen, C. B.; Christensen, C.; Pedersen, C.; Thomsen, E. V. Journal of the Electrochemical Society 2004 151 G338. (149) Resch, R.; Meltzer, S.; Vallant, T.; Hoffmann, H.; Koel, B. E.; Madhukar, A.; Requicha, A. A. G.; Will, P. Langmuir 2001 17 5666. (150) Abdelsalam, M. E.; Mahajan, S.; Bartlett, P. N.; Baumberg, J. J.; Russell, A. E. Journal of the American Chemical Society 2007 129 7399. (151) Cintra, S.; Abdelsalam, M. E.; Bartlett, P. N.; Baumberg, J. J.; Kelf, T. A.; Sugawara, Y.; Russell, A. E Faraday Discussions 2006 132 191.
109 (152) Lin, T. H.; Linn, N. C.; Tarajano, L.; Jiang, B.; Jiang, P. Journal of Physical Chemistry C 2009 113 1367. (153) Linn, N. C.; Sun, C. H.; Arya, A.; Jiang, P.; Jiang, B. Nanotechnology 2009 20 225303. (154) Br own, R. J. C.; Wang, J.; Tantra, R.; Yardley, R. E.; Milton, M. J. T. Faraday Discussions 2006 132 201. (155) Jin, J. The Finite Element Method in Electromagnetics, 2 nd ed ; John Wiley and S ons: New York 2002 (156) Moskovits, M. Reviews of Modern Physics 1985 57 783. (157) Wei, H. Y.; Eilers, H. Journal of Physics and Chemistry of Solids 2009 70 459. (158) Bendikov, T. A.; Rabinkov, A.; Karakouz, T.; Vaskevich, A.; Rubinstein, I. Analytical Chemistry 2008 80 7487. (159) Seminario, J. M.; Ma, Y. F.; Agapito, L. A.; Yan, L. M.; Araujo, R. A.; Bingi, S.; Vadlamani, N. S.; Chagarlamudi, K.; Sudarshan, T. S.; Myrick, M. L.; Colavita, P. E.; Franzon, P. D.; Nackashi, D. P.; Cheng, L.; Yao, Y. X.; Tour, J. M. Journal of Nanoscience and Nanotechnology 2004 4 907. (160) Semin, D. J.; Rowlen, K. L. Analytical Chemistry 1994 66 4324. (161) Vanduyne, R. P.; Hulteen, J. C.; Treichel, D. A. Journal of Chemical Physics 1993 99 2101. (162) Schlegel, V. L.; Cotton, T. M. Analytical Chemistry 1991 63 241. (163) Maissel, L. I.; Glang, R. Handbook of Thin Film technology ; McTraw Hill: New York 1970 (164) Tian, Z. Q.; Ren, B.; Wu, D. Y. Journal of Physical Chemistry B 2002 106 9463. (165) Haynes, C. L.; Van Duyne, R. P. Journal of Physical Chemistry B 2001 105 5599. (166) Kelf, T. A.; Sugawara, Y.; Cole, R. M.; Baumberg, J. J.; Abdelsalam, M. E.; Cintra, S.; Mahajan, S.; Russell, A. E.; Bartlett, P. N. Physical Review B 2006 74 (167) Sugawara, Y.; Kelf, T. A.; Baumberg, J. J.; Abdelsalam, M. E.; Bartlett, P. N. Physical Review Letters 2006 97 (168) Campion, A.; Ivanecky, J. E.; Child, C. M.; Foster, M. Journal of the American Chemical Society 1995 117 11807.
110 (169) Hutter, E.; Fendler, J. H. Advanced Materials 2004 16 1685. (170) Altewischer, E.; van Exte r, M. P.; Woerdman, J. P. Nature 2002 418 304. (171) Andrew, P.; Barnes, W. L. Science 2004 306 1002. (172) Jenkins, D.; Clegg, W.; Windmill, J.; Edmund, S.; Davey, P.; Newman, D.; Wright, C. D.; Loze, M.; Armand, M.; Atkinson, R.; Hendren, B.; Nutter, P. Microsystem Technologies Micro and Nanosystems Information Storage and Processing Systems 2003 10 66. (173) Okamoto, K.; Niki, I.; Shvartser, A.; Narukawa, Y.; Mukai, T.; Scherer, A. Nature Materials 2004 3 601. (174) Levene, M. J.; Korlach, J.; Turner, S. W.; Foquet, M.; Craighead, H. G.; Webb, W. W. Science 2003 299 682. (175) Xia, Y. N.; Gates, B.; Yin, Y. D.; Lu, Y. Advanced Materials 2000 12 693. (176) Cole, R. M.; Baumberg, J. J.; Garcia de Abajo, F. J.; Mahaj an, S.; Abdelsalam, M.; Bartlett, P. N. Nano Letters 2007 7 2094. (177) Abdelsalam, M. E.; Bartlett, P. N.; Baumberg, J. J.; Coyle, S. Advanced Materials 2004 16 90. (178) Kanamori, Y.; Kobayashi, K.; Yugami, H.; Hane, K. Japanese Journal of Applied Ph ysics Part 1 Regular Papers Short Notes & Review Papers 2003 42 4020.
111 BIOGRAPHICAL SKETCH Chih Hung Sun grew up in Hsinchu, Taiwan. He received his egree at the National Tai wan University in 2001 and his m aster degree at National Tsing Hua University in 2003. He started hi s military service in July 2003 and he was discharged in January 2005. He began his graduate studies at the University of Florida in August al research group to pursue a doctorate degree. He graduated in the fall of 2009 after spending three and half years being educated in chemical engineering and material science.