Large Area Fabrication of Periodic TiO2 Nanopillars Using

Subscriber access provided by UB + Fachbibliothek Chemie | (FU-Bibliothekssystem)

Article

Large area fabrication of periodic TiO2 nanopillars using microsphere photolithography on a photopatternable sol-gel film Olga Shavdina, Loic Berthod, Thomas Kämpfe, Stéphanie Reynaud, Colette Veillas, Isabelle Verrier, Michel Langlet, Francis Vocanson, Pascal Fugier, Yves Jourlin, and Olivier Dellea Langmuir, Just Accepted Manuscript • DOI: 10.1021/acs.langmuir.5b01191 • Publication Date (Web): 23 Jun 2015 Downloaded from http://pubs.acs.org on June 25, 2015

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Langmuir is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 26

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

Large area fabrication of periodic TiO2 nanopillars using

microsphere

photolithography

on

a

photopatternable sol-gel film O. Shavdina a,d, L. Berthod a,b,c, T. Kämpfe a, S. Reynaud a, C. Veillas a, I. Verrier a, M. Langlet b,c, F. Vocanson a, P. Fugier d, Y. Jourlin a* , O. Dellea d a

Laboratoire Hubert Curien - UMR CNRS 5516, Université de Lyon F-42000 Saint-Etienne,

France b

Univ. Grenoble Alpes, LMGP, F-38000 Grenoble, France

c

CNRS, LMGP, F-38000 Grenoble, France

d

L2CE, Laboratoire des Composants pour le Conversion de l’Energie, CEA/LITEN, Laboratoire

d’Innovation pour les Technologies des Energies Nouvelles et des nanomatériaux, F-38054 Grenoble, France ABSTRACT: The authors demonstrate a unique low cost process to print 2D, submicron size, and high refractive index nano-pillars using a direct colloidal-photolithography process. A well collimated i-line source emitting at 365 nm wavelength illuminates a mono layer of silica microspheres of 1 µm diameter deposited on a photosensitive TiO2-based sol-gel layer. No etching process is needed since this layer is directly UV photo patternable like a negative photoresist. Furthermore this thin layer offers interesting optical properties (high refractive index and

ACS Paragon Plus Environment

1

Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 26

optical transparency) and good mechanical and chemical stability and thus can be directly used as a functional microstructure (for PV or sensor applications for example). The paper describes the modeling of the electric field distribution below the spheres during the illumination process, the photochemistry of the TiO2 sol-gel layer process, and preliminary results of TiO2 nano-pillars of around 200 nm in diameter fabricated on a three-inch substrate.

1. Introduction The micro-texturing of surfaces at the nanometer and micrometer size scales has attracted much attention for applications in various domains.1,2,3 For example, in the field of optics, the texturing of silicon solar cells can increase their efficiency : by using nanopillars which can be extended or described as 2D gratings and 2D photonic crystals, it is possible to reduce reflection and increase the capacity of absorption of the incident light.4,5 This light trapping is an important factor for achieving higher energy conversion efficiency.6,7 Another well-known example is the enhancement of the natural hydrophobicity of surfaces by micro-texturing with a forest of micropillars, leading to very useful properties such as protecting surfaces from rain and self-cleaning windows. Microstructures in the range of the micrometer scale are also used for low friction devices in microfluidics8,9 and tribology.10,11 Nanostructures are expected to find applications for increasing the efficiency and luminance of LEDs 12, which are widely used on smartphones and portable electronics as well as becoming increasingly common in lighting.13 Recently, a uniform coverage of nanopillars was fabricated through a combination of Langmuir–Blodgett assembly and reactive ion etching by changing the shape of the pillars from having vertical to tapered sidewalls with sharp tips.14 For the fabrication of nanopillars several research groups concentrated their efforts on nanoimprint lithography.15 However, this method presents several


2

Page 3 of 26

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

troubles like difficult molding conditions, the effect of thermal expansion through the process and surface sticking problems between resist layer and stamp. Otherwise, microsphere photolithography is a promising inexpensive fabrication tool for producing regular and homogenous arrays of micro/nano structures with different sizes. By combining the selfassembly and focalizing properties of microspheres, it is possible to use sphere arrays as masks to produce large areas of nanopillars.16,17,18,19,20 Being irradiated by an extended beam, each sphere is acting as a near-field microlens. In other words, when illuminated by a plane wave, focusing of the incident light arising from self-assembled spherical beads generates a field enhancement in a small area underneath their exit surface, called waist photonic nanojet,21 that propagates into the underlying photoresist layer. Research in the field of colloidalphotolithography was done by using positive, as well as negative photoresists, which lead to the formation of holes in the photoresist layer or pillars, respectively. 22 In these cases, photoresist layer acts as a mask for post processes such as etching process (RIE, ICP) or layer depositions. In our study, we intend to obtain nanopillars using the method of colloidal-photolithography combined with a titanium oxide (TiO2)-based sol-gel layer

23,24

, which acts as a negative

photoresist. This approach has been previously demonstrated using interferential lithography (or holography)25, or phase mask process.26,27,28 TiO2 thin films show a large interest in the field of optics due to their high transparency in the visible spectrum, their high refractive index, as well as their photocatalytic activity and photo-induced superhydrophilicity. Besides, the use of sol-gel resists is also interesting for the manufacturing process, since the created structures can be used directly without post-process29, which is one of the main interests. Thus, it can cover a wide number of application fields since it can be cost-effective and compatible to large substrates29.


3

Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 26

Using this type of resist in association with silica beads, we successfully obtained pillars at a nanometer scale. To deposit the layer of silica beads on a sol-gel film, we used a Langmuir Blodgett method owing to the relative simplicity with which the conditions of self-assembly can be achieved and the accurate control of the deposition process achieved even over large areas.30,31,32 The simulations using the rigorous method RCWA (Rigorous Coupled Wave Analysis)33 confirm well-known phenomena of focusing the light through the silica particles with formation of photonic nanojet, which propagates into the background medium from the shadowside surface of a plane-wave illuminated dielectric microsphere.34 2. Microspheres illumination modeling The modeling concerns the calculation of the electric field distribution during the illumination of a microsphere hexagonal 2D array with a plane wave. As will be explained in the Experimental section, before depositing the layer of silica beads and subsequent UV illumination, the sol-gel photoresist was coated with a PMMA layer in order to protect it from hydrolysis. The following simulations were therefore performed by taking into account the geometrical properties of all the optical mediums and thin layers, considering the wavelength and the microspheres parameters as illustrated in figure 1.


4

Page 5 of 26

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

Figure 1. Structure of the microsphere (being discretized) layer along zOx plane: ˄x ─ period along the axis Ox; n

media

─ refractive index of the optical media; n

part

─ refractive index of the

particle; nl.1 ─ refractive index of the PMMA layer; nl.2 ─ refractive index of the sol-gel layer; Ø part

─ diameter of the particle; ε ─ permittivity; hl - thickness of each particle layer; hl.1 ─

thickness of the PMMA layer; hl.2 ─ thickness of the sol-gel layer. In order to study the distribution of the electromagnetic field propagating through the assembly of the silica microspheres, the PMMA layer, and the TiO2 based sol-gel resist, we used the RCWA MC gratings method that was based on a lateral (i.e. in x- and y-direction) Fourier expansion of the dielectric function and of the electric and magnetic fields.35 Calculations were made by considering a periodic array of silica microspheres. Each micro-sphere was longitudinally discretized by dividing into 35 layers considering 1 µm diameter sphere with 1.45 refractive index. The choice of 35 layers is a good compromise to receive a quasi-exact value of the electromagnetic field in an acceptable calculation time. In addition, we considered an underneath PMMA/photoresist bi-layer coating, with a thickness of 55 nm and refractive index of 1.49 for the PMMA over-layer, and a thickness of 300 nm and refractive index of 1.65 for the TiO2 based photoresist. As a substrate for structuring, we used the glass BK 7 with a thickness of 1 mm and a refractive index of 1.5. A linearly polarized input wave, with the electric field vector


5

Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 6 of 26

in y-direction and a wavelength of 365 nm (a well collimated and filtered i-line source), was employed for the simulations. Figure 2 represents an elementary cell of the hexagonal arrangement of these spheres along the xOy plane with period along the y-axis ˄y = 1732 nm and along x-axis ˄x = 1000 nm. As explained before, when a microsphere is illuminated by a plane wave, a so-called photonic nanojet beam emerges from the sphere, having subwavelength transverse dimensions and low divergence36,37. The polarization of the input wave is not preserved through the microspheres which leads to a non-polarized photonic nanojet.

Figure 2. Elementary cell of the hexagonal arrangement of the spheres. In order to check the validity of the chosen number of the Fourier expansion orders and number of z-discretization layers, convergence tests had to be carried out. Figure 3 (a) represents the results of the y-component of the electric field distribution over an x-z plane located at midheight of the spheres in y-direction. The intensity of the photonic nanojet (defined as the square of the electric field amplitude) is about 16 times greater (red color) than the intensity of the incoming beam in the forward z-direction (blue color). This figure shows that the electric field distribution is modulated periodically inside the sphere and in the thin layers. It is explained by various patterns due to interferences between incoming beam and reflected beam at each


6

Page 7 of 26

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

interface (inside the sphere and air / layers interface). Figure 3(b) illustrates the intensity distribution for a cross section in x-direction, taken at the z-position of the sphere / PMMA (z = 1000 nm), PMMA / sol-gel (z = 1055 nm), and sol-gel / BK7 substrate (z = 1355 nm) interfaces. This figure shows that the full width at half maximum (FWHM) value is 242 nm at any interfaces. However, figures 3(a) and (b) also indicate that slight changes in the field intensity occur when the nanojet propagates from the PMMA / sol-gel layer interface toward the sol-gel layer / BK7 substrate one. At the latter interface, the electric field profile is slightly broader (in the low intensity range; light blue in Figure 3 (a)) than at the outer surface of the sol-gel layer. Besides, the maximal field intensity slightly decreases over nanojet propagation through the solgel layer. In other words, the maximal intensity (red color in Figure 3(a)) exhibits some kind of inverse trapeze profile. Figure 3 (c) shows the electric field distribution for a cross section direction taken along the diameter of the microsphere. This figure confirms the well-known fact that the nanojet waist propagates from the inside toward the outside of the sphere on a maximum propagation distance of ≈ 2.5 λ (here λ = 365 nm) and leads to a subwavelength focal spot modulated by interferences of the electric fields. All these results demonstrate that a 1 µm microsphere diameter is efficient to form a subwavelength focal spot based on constructive interferences of the photonic nanojet fields. The modeling also showed that changing the substrate refractive index from 1.4 to 1.7 did not affect strongly the profile of the electrical field distribution, meaning that i/ the substrate has little effect on the nanojet propagation, and ii/ a large range of substrates can be used for the fabrication of nanopillars. It should be noted that, in our modelling, we have not taken into account the absorbance/transmission of the sol-gel layer. Thus, the simulation is not fully rigorous. However, the measured difference of transmission at 365 nm wavelength between the illuminated and the not illuminated sol-gel layer is only 2% and


7

Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 8 of 26

the refractive index before and after UVA treatment is, respectively, 1.63 and 1.62 (ellipsometric measurements), which suggests that no significant photobleaching, densification, and/or refractive index modification occurs over illumination. Thus, optical properties of the sol-gel layer remain nearly the same and, in our modeling, we assumed for simplification purpose that the nanojet does not evolve over illumination.

Figure 3. (a) Electric field amplitude distributions (y-component) over an x-z plane situated at mid-height of the spheres of 1µm diameter. The white lines represent the microsphere profiles. The color scale illustrates the electric field intensity. (b) RCWA calculated cross sectional intensity of electric field profile taken at z = 1000 nm, z = 1055 nm, z = 1355 nm (in normal incidence). (c) photonic nanojet observed for a z cut positioned at the middle of the sphere in xand y-direction. The dashed line represents the microsphere profile.


8

Page 9 of 26

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

3. Experimental 3.1 Sol-gel layer processing The whole thin film processing is schematized in Figure 4.

Figure 4. The different steps of the thin film processing: a) Layers deposition (TiO2 based solgel layer + PMMA layer); b) Microspheres deposition; c) Exposition under UV light; d) Development. Specific sol-gel formulation and procedures were used to deposit the TiO2-based photoresist. A sol was prepared from titanium ispropoxyde orthotitanate (TIPT) complexed by benzoyl acetone (BzAc), using a mixture of two different sols. The first one (S1) was fabricated by mixing TIPT with BzAc in methanol (MeOH), using a TIPT/BzAc/MeOH molar composition of 1/0.75/20.4 23

. The second sol (S2) was prepared by mixing TIPT, deionized water, HCl, and the solvent


9

Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

butanol

(BuOH).

For

this

sol,

the

TIPT/H2O/HCl/BuOH

Page 10 of 26

molar

composition

was

1/0.82/0.13/23.9. Before being mixed with S1, S2 was aged for two days at room temperature. Once mixed, the two sols formed a final sol of 0.5 mol.l-1 TIPT concentration with a BzAc/TIPT molar ratio of 0.6. This final sol was deposited on a three-inch glass sample by spin-coating at 7000 rpm (Figure 4 (a)). Let us note that the whole procedure described hereafter is compatible with treatments on larger size substrates. The obtained sol-gel layer was then naturally dried at room temperature and subsequently heated at 110°C during 90 min, leading to a so-called xerogel film, i.e. an inorganic polymer film constituted of Ti-O-Ti chains with organic chain-end groups arising from the sol formulation, mainly TIPT-BzAc complexed species. This xerogel film is soluble in different solvents (alcohol, chloroform, acetone, etc) as far as BzAc stays complexed with TIPT. The figure 5 represents the complexation between these two species.

Figure 5. Molecular structure of the BzAc chelating agent (left), TIPT precursor (center), and resulting TIPT-BzAc complex (right). The main interest of this protocol relies on the properties of BzAc, which make the film soluble in a solvent while being sensitive to UVA light. Under UVA illumination, the TIPT-BzAc complex partially degrades in insoluble titanium-based species such as carbonates and/or carboxylates. Therefore, it creates a contrast of solubility between illuminated and non-


10

Page 11 of 26

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

illuminated areas when selectively exposed to UVA light. After illumination a heating step of 8 min at 110°C was finally used to increase solubility contrasts. These conditions led to a ~ 350 nm thick TiO2-based photoresist, as deduced from profilometer measurements and SEM imaging.

However, the xerogel film is very sensitive to water and ambient humidity.

Accordingly, water causes hydrolysis and subsequent fast polymerization of inorganic chains, which become longer and intermingled with each other’s, and result in a chemically stable xerogel film. This latter can then no longer dissolve while being immersed in a solvent for developing post process. So, for the following steps, it was necessary to coat the sol-gel resist with an additional thin layer of PMMA intended to protect it from hydrolysis caused by the silica beads-containing aqueous solution. This PMMA layer was deposited by spin-coating (5000 rpm) using a solution of PMMA (350 Kg/mol) dissolved in anisol with a mass concentration of 4%. The solvent is dried at room temperature for 1h to finally yield a PMMA film of about 55 nm in thickness deposited on the TiO2-based photoresist (Figure 4 (b)). Let us note that this PMMA film is highly transparent to UVA light and thus compatible with the UV illumination. 3.2 Silica beads deposition The Langmuir Blodgett (LB) technique38 was employed to deposit 1 µm silica sphere arrays using a Boostream® device (Fig. 4(b))39,40,41. Figure 6 (c) shows a SEM image of a multilayer assembly constituted, from bottom to top, of a xerogel thin layer (≈ 350 nm thick), an additional protection layer of PMMA (55 nm thick), and a monolayer of silica microsphere of 1 µm diameter. Figure 6 (b) illustrates a uniform microsphere hexagonal pattern covering the sol-gel photoresist. The crystallographic orientation of this hexagonal pattern is kept constant on large areas of several hundred of micrometers. Crystallographic defects, such as dislocations, cause a change of orientation between adjacent areas but the whole substrate is uniformly coated with


11

Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 12 of 26

hexagonally arranged areas. Actually, the uniform deposition of self-organized silica beads is compatible with substrates of large size. It is exemplified in figure 6 (a) for a substrate of 29x25cm2, where observed colors illustrate light diffraction effects induced by silica beads covering the whole substrate surface. We have shown that this uniform deposition is also compatible with substrates of complex shape, such as lenses or cylinders.

Figure 6. Photography of silica beads deposited on a large substrate of 29x25cm2 a), typical SEM images of a SiO2 sphere / PMMA layer / TiO2 xerogel multilayer assembly before UV treatment b) in top view, and c) in tilted view.

3.3 UV illumination process All the samples were illuminated between 8 and 20 min under collimated light from a gas discharge lamp mercury vapor with a power of 26 mW.cm-2 (Fig. 4(c)). Then a short heating step of 8 min at 110°C was necessary, as described in the previous section. The microspheres were removed by ultrasonication for 1 min in deionized water. The final development was then performed using a three-step procedure based on successive washings in chloroform, ethanol and deionized water, during 1 min each. Indeed, chloroform dissolves easily PMMA and the nontreated xerogel layer, ethanol washes the sample, and water stabilizes nanopillars. The whole protocol finally leads to periodic nanopillars derived from the photosensitive xerogel film (Fig. 4


12

Page 13 of 26

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

(d)). The samples were then baked for 120 min at temperatures ranging from 100°C to 500°C in order to assess the influence of this post bake process on the film transformation and on the nanopillar dimensions.

4. Nanopillars analysis Figure 7 shows a typical result after post-illumination development of the xerogel layer, which reveals a large area hexagonal array of periodically organized nanopillars according to the microspheres layout (figure 7(a)). The periodicity of about 1 µm depends exclusively on the diameter of the microspheres in contact to each other’s, as illustrated in Fig 6. Besides, the nanopillars exhibit an inverse trapeze profile, with a base narrower than the top (figure 7(b)). The photosensitive xerogel film acts as a negative photoresist with a nearly linear photosensitivity response, which is different from the majority of traditional positive or negative all-organic commercial photoresists. In other words, the nanopillar shape is expected to fit the intensity profile of nanojet. According to the figures 3 (a) and (b), the electric field intensity slightly decreases as the nanojet propagates through the photoresist, showing some kind of inverse trapeze profile. It is therefore inferred that, in the first stages of UV illumination, the photoresist is more efficiently stabilized at its outer surface compared to the photoresist / substrate interface, which would explain the inverse trapeze shape of derived nanopillars.


13

Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 14 of 26

Figure 7. Typical SEM images of printed nanopillars a) in top view and b) in tilted view. Figure 8 illustrates the nanopillars obtained from a xerogel thin layer illuminated during 8 min and subsequently heat-treated for two hours at 100°C, 300°C, or 500°C. In any cases, the SEM images show nanopillars with an inverse trapeze shape. Besides, an additional heating step at increasing temperature concomitantly leads to a gradual reduction in the section and height of the nanopillars, i.e. in their volume, which depicts a thermally activated densification mechanism. This densification mechanism was confirmed by ellipsometric measurements of the refractive index at 633 nm performed on non-imprinted xerogel films heat-treated at various temperatures. Accordingly, the nanopillar volume decreases when increasing the temperature while the film refractive index rises (Figure 9). After heating at 500°C, the nanopillars lose about 65% of their initial volume (as deduced from SEM image analyses) and corresponding films have a refractive index of 2.25 (close to the value of TiO2 anatase phase) compared to a refractive index of 1.65 for the xerogel film after heat-treatment at 100°C. Besides, we have previously shown that a heat-treatment at 500°C not only promotes densification of the photoimprinted structures, but also allows an amorphous to crystalline transition leading to photocatalytically active imprinted TiO2 arrays23.

Figure 8. SEM images of nanopillars formed after 8 min illumination under UVA for different post-development heat-treatments: no heat-treatment a), heat-treatment at 100°C b),


14

Page 15 of 26

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

300°C c) and 500°C d).

Figure 9. Graph of nanopillar volume (%) and corresponding film refractive index versus post bake temperature. Volume values have been normalized to the value measured after heattreatment at 100°C and the refractive index was determined at 633 nm wavelength. Figure 10 illustrates the nanopillars obtained from a xerogel thin layer insulated for 10, 14, and 20 min, before and after a heat-treatment during two hours at 500°C. This figure reveals many changes in the size and shape of derived nanopillars. Before heat-treatment, the mean diameter of nanopillars increases with longer exposure times, from about 200 nm after 10 min UVA treatment to about 400 nm after 20 min UVA treatment. This feature depicts an increasing UVA dose (the product of the light intensity by the illumination duration) on the edges of the nanojet, which enlarges the photoresist areas chemically stabilized over UV illumination. On the one hand, after a short UV illumination (10 min or less), the measured mean diameter is in fairly good agreement with the FWHM of the electric field intensity (242 nm) deduced from RCWA calculations (Figure 3). On the other hand, our modeling does not predict an enlargement of the


15

Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 16 of 26

nanopillars, probably because it does not account for weak changes in the optical properties (absorbance/transmission, refractive index) of the sol-gel layer over illumination.

Figure 10. SEM images of nonapillars formed after 10 min a), 14 min b) and 20 min c) of UVA illumination, and after 10 min d), 14 min e) and 20 min f) of UVA illumination and a subsequent heat-treatment at 500°C during 2h.

Figure 11 illustrates the derived rise of the nanopillar volume with UV exposure time. After a 20 min UV illumination, the volume is nearly doubled compared to that obtained after short UV illumination durations (8 or 10 min). Besides and as previously described, an additional heating step (here, 500°C for two hours) promotes a nanopillar densification depicted by their volume reduction. However, it is worthwhile noting that, after heating at 500°C, the influence of the UV illumination duration on the nanopillar volume is exactly the same as before heattreatment, i.e. after a 20 min UV illumination the volume is nearly doubled compared to that obtained after short UV illumination durations and post bake treatment. This means that, as illustrated in Figure 11, after the heat-treatment the nanopillars have lost around 65% of their


16

Page 17 of 26

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

initial volume irrespectively of the UV illumination duration effects. Finally, Figure 10 also shows that the global shape changes gradually over UV illumination. For a 10 min UV illumination, the nanopillars exhibit an inverse trapeze shape with a base narrower than the top, as previously illustrated in Figure 8. For a 14 min UV illumination, the edges become vertical, leading to cylindrical nanopillars. For a 20 min UV illumination, the trapezoidal form is reversed, the base becoming larger than the top.

Figure 11. Volume of nanopillars obtained from a xerogel thin layer insulated for 8, 10, 14, and 20 min, before and after a heat-treatment for two hours at 500°C, and volume values measured after heat-treatment at 500°C normalized to the value measured before heat-treatment. These shape evolutions can be discussed according to figure 12, which shows the UV/visible transmission spectra of a xerogel film heat-treated at 110°C before and after subsequent UV illumination. In the visible spectral range, these spectra exhibit transmission fluctuations due to interferences phenomena arising from multiple reflections at the film-air and film-substrate interfaces. Transmission minima depict reflection features arising from the


17

Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 18 of 26

relatively high refractive index of the xerogel film (1.62-1.63) compared to that of the glass substrate (1.50). The fact that these minima are not significantly modified after UV illumination supports ellipsometric measurements showing that the refractive index does not evolve significantly over UV illumination. Transmission maxima correspond to wavelengths where substrate-film-air interfacial interferences do not perturb the transmission spectrum. For these wavelengths, the fact that the film-on-glass transmission closely fits that of the bare substrate demonstrates that the photo-resist film does not present any optical loss, i.e. exhibits an excellent optical quality in the visible range before and after UV illumination. Important transmission losses are observed below 400 nm. These losses are firstly due to absorption inherent to the xerogel film, which intrinsically absorbs light in the UVA spectral range. A marked local transmission minimum (transmission inferior to 10%) is also observed around 365 nm, which depicts light absorption arising from the TIPT-BzAc complex. It is observed that the transmission at 365 nm slightly increases, by about 2%, after UV illumination. This increase illustrates the partial photolytic decomposition of the complex. In our modelling, we decided for simplification to omit the xerogel film absorption. It appears that, after a short illumination and despite this omission, the mean diameter of nanopillars is in fairly good agreement with the calculated FWHM of the electric field intensity, but it is clear that this omission is not fully rigorous. Thus, the experimentally observed nanopillar shape evolutions over UV illumination can eventually depict changes in the xerogel film absorption, which are not accounted for by the modelling. On the one hand, for short UV illumination durations (weak UV dose), it is likely that, beside the inverse trapeze profile of the field intensity, light absorption by the xerogel film also contributes to the inverse trapeze profile of nanopillars. In other words, the xerogel is quickly stabilized at the surface and hide the deeper layer from the UV light. On the


18

Page 19 of 26

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

other hand, it is possible that the UV light transmission increase occurring over illumination favors a better photo-induced stabilization in the deeper layers of the xerogel film. In that case, not only an increase of the UV dose may induce an increase of the nanopillar volume, but an improved stabilization of the deeper layers may also induce a reverse change in the nanopillar trapeze shape. Accordingly, figure 3 shows that the electric field profile is slightly broader at the base (z=1355nm) than at the top (z=1055nm) of photopatterned nanopillars. In any case, our results indicate the possibility to finely monitor the shape of photo-imprinted nanopillars through a suitable control of the UV illumination duration and post bake treatment. Remarkably, Figure 10 indicates that the initially observed trapezoidal or cylindrical pillar shapes are preserved after the final heat-treatment at 500°C.

Figure 12. UV/visible transmission spectra of a bare glass substrate (a) and a glass substrate coated with a BzAc-TiO2 xerogel film heat-treated at 110°C (b), and subsequently exposed to UV light (c).


19

Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 20 of 26

5. Conclusion This paper shows for the first time the feasibility of periodic TiO2 nanopillars obtained from a fast and cost effective method based on unique combination of a sol-gel derived photopatternable resist and a direct colloidal-photolithography process. TiO2 nanopillars of around 200 nm diameter and 300 nm height have been imprinted uniformly on the whole surface of a three-inch glass substrate. The sol-gel derived photoresist, in combination with UV illumination duration and post bake heat-treatment effects, enables to control both the shape, volume, and refractive index of the nanopillars. This work opens the route toward new applications of nanostructured surfaces dealing, for instance, with solar cells or lighting devices sensors or others outdoors applications. Further developments concern geometry (diameter, spherical and non-spherical geometry) and material (silica, polystyrene…) microspheres modeling, allowing an optimized nanojet electric field distribution underneath the microspheres. In order to address numerous applications in the field of energy and sensors, the process could be currently adapted to nonconventional shape substrates such as lenses or cylinders. ASSOCIATED CONTENT Supporting Information. Experimental section including materials and methods. This material is available free of charge via the Internet at http://pubs.acs.org.

AUTHOR INFORMATION Corresponding Author


20

Page 21 of 26

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

*Email: [email protected] Notes The authors declare no competing financial interest.

ACKNOWLEDGMENTS The Authors would like to thank the French Region Rhône-Alpes for its financial support, in the framework of an ARC 4 Energies.

REFERENCES

(1) Wakuda, M.; Yamauchi, Y.; Kanzaki, S.; Yasuda, Y. Effect of surface texturing on friction reduction between ceramic and steel materials under lubricated sliding contact . Wear 2003, 254, 356─363. (2) Sopori, B. L.; Pryor, R.A. Design of antireflection coatings for textured silicon solar cells. Solar Cells 1983, 8, 249─261. (3) Dewan, R.; Marinkovic, M.; Noriega, R.; Phadke, S.; Salleo, A.; D. Knipp, D. Light trapping in thinfilm silicon solar cells with submicron surface texture. Opt. express 2009, 17, 23058─23065. (4) Ko, Y.H.; Yu, J.S. Design and fabrication of antireflective GaN subwavelength grating structures using periodic silica sphere monolayer array patterning. Appl. Phys. B 2013, 113, 567─573. (5) Fan, Z.; Kapadia, R.; Leu, P. W.; Zhang, X.; Chueh, Y.-L.; Takei, K.; Yu, K.; Jamshidi, A.; Rathore, A. A.; Ruebusch, D. J.; Wu, M.; Javey, A. Ordered Arrays of Dual-Diameter Nanopillars for Maximized Optical Absorption. Nano Lett., 2010, 10, 3823–3827. (6) Borraa, M.Z.; Güllüb, S. K.; Esa, F.; Demircioğlua, O.; Günövena, M.; Turana, R.; Bek, A. A feasibility study for controlling self-organized production of plasmonic enhancement interfaces for solar cells. Appl. Surf. Sci. 2014, 317, 43–50. (7) Lin,Q.; Hua, B.; Leung, S.; Duan, X.; Fan, Z. Efficient Light Absorption with Integrated Nanopillar/Nanowell Arrays for Three-Dimensional Thin-Film. Photovoltaic Applications 2013, 7, 2725– 2732. (8) Calliesa, M.; Chenb, Y.; Martyd, F.; Pépinb, A.; Quéréa, D. Microfabricated textured surfaces for super-hydrophobicity investigations. Microelectronic engineering 2005, 78─79, 100─105.


21

Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 22 of 26

(9) Zhao, H.; Park, K.; Law, K. Effect of Surface Texturing on Superoleophobicity, Contact Angle Hysteresis, and “Robustness”. Langmuir 2012, 24, 14925─14934. (10) A Skarmoutsou, A. ; Charitidis, C.A. ; Gnanappa, A.K. ; Tserepi, A. ; Gogolides, E. Nanomechanical and nanotribological properties of plasma nanotextured superhydrophilic and superhydrophobicpolymeric surfaces. Nanotechnology 2012, 23, 505711─505722 (11) Pham, D.C.; Na, K.; Yang, S.; Kim, J.; Yoon, E.-S. Nanotribological properties of silicon nanopillars coated by a Z-DOL lubricating film. Journal of Mechanical Science and Technology 2010, 24, 59─65. (12) Shei, S.C. SILAR-based application of various nanopillars on GaN-Based LED to enhance lightextraction efficiency. Journal of Nanomaterials 2013, 653981─653987. (13) W. Ding; Y. Wang; Chen, H.; Chou, S.Y. Plasmonic nanocavity organic light-emitting diode with significantly enhanced light extraction, contrast, viewing angle, brightness, and low-glare. Adv. Funct. Mater 2014, 24, 6329─6339. (14) Hsu, C.-M.; Connor, S.T.; Tang, M.X.; Cui , Y. Wafer-scale silicon nanopillars and nanocones by Langmuir–Blodgett assembly and etching. Appl. Phys. Lett. 2008, 93, 1─3. (15) Zhou, W.; Min, G.; Zhang, J.; Liu , Y.; Wang, J.; Zhang, Y.; Sun, F. Nanoimprint Lithography: A Processing Technique for Nanofabrication Advancement. Nano-Micro Lett. 2011, 3, 135─140. (16) Wu; W.; Dey, O.; Memis, O.; Katsnelson, A.; Mohseni, H. Novel Self-Assembled and Maskless Technique for Highly Uniform Arrays of Nano-Holes And Nano-Pillars. Nanoscale Res. Lett. 2008, 3, 123─127. (17) Wu; W.; Dey, O.; Memis, O.; Katsnelson, A.; Mohseni, H. Fabrication of Large Area Periodic Nanostructures Using Nanosphere Photolithography. Nanoscale Res. Lett. 2008, 3, 351─354. (18) Szabó, Z.; Volk, J.; Fülöp, E.; Deáka, A.; Bársonya, I. Regular ZnO nanopillar arrays by nanosphere photolithography. Photonics and Nanostructures - Fundamentals and Applications 2013, 11, 1─7. (19) Chang, Y.-C.; Lu, S.-C.; Chung, H.-C.; Wang, S.-M.; Tsai, T.-D.; Guo, T.F. High-Throughput Nanofabrication of Infra-red and Chiral Metamaterials using Nanospherical-Lens Lithography. Scientific Reports 2013, 3, 1─7. (20) Wu, W.; Dey, D.; Katsnelson, A.; Memis, O.G.; Mohseni, H. Large areas of periodic nanoholes perforated in multistacked films produced by lift-off. J. Vac. Sci Technol. 2008, 26, 1745─1747 (21) Heifetz, A.; Kong, S.-C.; Sahakian, V.; Taflove, A.; Backman, V. Photonic Nanojets. J. Comput. Theor. Nanosci. 2009, 6, 1979–1992 (22) Asoh, H.; Fujihara, K.; Ono, S. Triangle pore arrays fabricated on Si (111) substrate by sphere lithography combined with metal-assisted chemical etching and anisotropic chemical etching. Nanoscale Res. Lett. 2013, 7, 1─8


22

Page 23 of 26

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

(23) Briche, S.; Tebby, Z.; Riassetto, D.; Messaoud, M.; Gamet, E.; Pernot, E.;Roussel, H.; Dellea, O.; Jourlin, Y.; Langlet, M. New insight in photo-patternad sol-gel-derived TiO2 films, J. Mater. Sci. 2011, 46, 1474─1486 (24) Fallet, M.; Permpoon, S.; Deschanvres, J. L.; Langlet, M. Influence of physico- structural properties on the photocatalytic activity of sol-gel derived TiO2 thin films. J Mater Sci 2006, 41, 2915–2927. (25) Gâté, V. ; Bernaud, G. ; Veillas, C. ; Cazier, A. ; Vocanson, F., Y., M. Fast dynamic interferometric lithography for large submicrometric period diffraction gratings production. Opt. Eng. 2013, 52, 091712 (1─7) (26) Bourgin, Y. ; Jourlin, Y. ; Parriaux, O. ; Talneau, A. ; Tonchev, S. ; Veillas, C. ; Karvinen, P. ; Passilly, N. ; Md Zain, A. R. ; M. De La Rue, R. ; Van Erps, J. ; Troadec, D. 100 nm period grating by high-index phase-mask immersion lithography. Opt. express, 2010, 18, 10557─10566. (27) Stehlin, F.; Bourgin, Y.; Spangenberg, A.; Jourlin, Y.; Parriaux, O.; Reynaud, S.; Wieder, F.; Soppera, O. Direct nanopattering of 100 nm metal oxide periodic structures by Deep-UV immersion lithography. Opt. letters 2012, 37, 4651─4653. (28) Le Blanc, D. ; Pelissier, S. ; Saravanamuttu, K. ; Najafi, S. I. ; Andrews, M.P. Self-processing of surface-relief gratings in photosensitive hybrid sol-gel glasses. Adv. Mater. 1999, 11, 1508─1511. (29) Gâté, V.; Jourlin, Y.; Vocanson, F.; Dellea, O.; Vercasson, G. Sub-micrometric patterns written using a DIL method coupled to a TiO2 photo-resist. Opt. Mater. 2013, 35, 1706─1713. (30) Cabrera, E.-J.; Jaller, L. M. ; Amade, R.; Portal, S. M.; Pascual, E.; Bertran, E. Photonic Characteristics of Langmuir-Blodgett Self-Assembled Monolayers of Colloidal Silica Particles, Nanoscience and Nanotechnology Letters 2012, 4, 1─5. (31) Tsai, P.-S.; Yang, Y.-M. Fabrication of Hydrophobic Surfaces by Coupling of Langmuir - Blodgett Deposition and a Self-Assembled Monolayer. Langmuir 2006, 22, 5660─5665. (32) Marquez, M.; Grady, P. The Use of Surface Tension to Predict the Formation of 2D Arrays of Latex Spheres Formed via the Langmuir - Blodgett-Like Technique. Langmuir 2004, 20, 10998─11004. (33) Semenikhin, I.; Zanuccoli, M.; Benzi, M.; Vyurkov, V.; Sangiorgi, E.; Fiegna, C. Computational efficient RCWA method for simulation of thin film solar cells. Opt. Quant. Electron. 2012, 44, 149─154. (34) Jalali, T.; Erni, D. Highly confined photonic nanojet from elliptical particles. J. Mod. Optic. 2014, 61, 1069 ─ 1076. (35) Moharam, M. G.; Gaylord, T. K. Rigorous coupled-wave analysis of planar-grating diffraction. J. Opt. Soc. Am. 1981, 71, 811─818. (36) Itagi, A.V.; Challener, W.A. Optics of photonic nanojets. J. Opt. Soc. Am. A 2005, 22, 2847─2859. (37) Geints, Y.E.; Panina, E.K.; Zemlyanov, A.A. Control over parameters of photonic nanojets of dielectric microspheres. Opt. Commun. 2010, 283, 4775─4781.


23

Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 24 of 26

(38) Bardosova, M.; Pemble, M.E.; Povey, I. M.; Tredgold, R. H. The Langmuir-Blodgett Approach to Making Colloidal Photonic Crystals from Silica Spheres. Adv. Mater. 2010, 22, 3104─3124. (39) Delléa, O.; Shavdina, O.; Fugier, P.; Coronel, P.; Ollier, E.; Désage, S.F. Precision assembly technologies and systems control methods in microspheres precision assembly for colloidal lithography. IPAS 2014, 435, 107─117. (40) Delléa, O.; Coronel, P.; Fugier, P. Procede de transfert d'objets sur un substrat à l'aide d'un film compact de particules. Patent WO 2013, 102608, 1─31. (41) Delléa, O. Breakthrough at air/liquid interface: Conference on Organised Films. Cork, Ireland 2013.

BooStream© Technology. 13th European


24

Page 25 of 26

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

Abstract Graphic

Large area fabrication of periodic TiO2 nanopillars using microsphere photolithography on a photopatternable sol-gel film O. Shavdina a,d, L. Berthod a,b,c, T. Kämpfe a, S. Reynaud a, C. Veillas a, I. Verrier a, M. Langlet b,c, F. Vocanson a, P. Fugier d, O. Dellea d, Y. Jourlin a*

ABSTRACT: The authors demonstrate a unique low cost process to print 2D, submicron size, and high refractive index nano-pillars using a direct colloidal-photolithography process. A well collimated i-line source emitting at 365 nm wavelength illuminates a mono layer of silica microspheres of 1 µm diameter deposited on a photosensitive TiO2-based sol-gel layer. No etching process is needed since this layer is directly UV photo patternable like a negative photo-resist. Furthermore this thin layer offers interesting optical properties (high refractive


Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

index and optical transparency) and good mechanical and chemical stability and thus can be directly used as a functional microstructure (for PV or sensor applications for example). The paper describes the modeling of the electric field distribution below the spheres during the illumination process, the photochemistry of the TiO2 sol-gel layer process, and preliminary results of TiO2 nano-pillars of around 200 nm in diameter fabricated on a three-inch substrate.


Page 26 of 26

Large Area Fabrication of Periodic TiO2 Nanopillars Using

Recommend Documents