We describe low-index porous alumina thin films deposited from aqueous solutions containing flat-Al13 clusters and the nonionic block copolymer Pluronic F127. The films exhibit surface roughness as low as 0.5 nm, a characteristic useful for the fabrication of multilayer optical devices. We have investigated the chemical, optical, and structural properties of the films and their formation by Fourier transform infrared spectroscopy, temperature programmed desorption, spectroscopic ellipsometry, transmission electron microscopy, atomic force microscopy, and X-ray diffraction. By varying the Pluronic F127 concentration, film porosities can be tuned to produce refractive indexes between n = 1.26 and 1.54 (λ = 550 nm).
© 2016 Optical Society of America
Herein, we describe low-index Al2O3 films with surfaces approaching atomic smoothness that are deposited via a facile solution method. The combination of low refractive indexes and smooth surfaces satisfies conditions for fabrication of multilayer structures, optical resonators, and photonic crystals, where high refractive-index contrast, i.e., the difference in refractive index between high-index and low-index materials, enables high figures of merit . The low-index materials and methods also make possible simple approaches to fabricate anti-reflection coatings .
MgF2 exhibits the lowest refractive index (n = 1.37 @ λ = 550 nm) among all conventional dense thin films [3–5]. Oxide films with porous morphologies and attendant low densities display lower indexes. For example, oblique-angle deposition of SiO2 by electron-beam evaporation produces an array of oriented SiO2 nanorods with a porous microstructure . These films may exhibit a refractive index as low as 1.08. ALD integrated with nanolithography enables deposition of low-index films comprising thin shells of Al2O3 and ZnO. These films may exhibit a refractive index as low as 1.025 .
Solution deposition is an alternate method to the above vapor methods that provides potential to deposit a wider range of film compositions for scientific investigations and to coat large-area substrates. Researchers often modify conventional, metal-organic sol-gel techniques to produce porous oxide thin films. These methods, however, promote nanoparticle formation due to the highly reactive nature of the precursors. Consequently, they yield films with textured surfaces characterized by surface roughness commonly in the range of 5-10 nm [8–12]. Recently, we have shown aqueous oxo-hydroxo metal clusters in place of these sol-gel precursors produce very high-quality continuous and smooth films (surface roughness < 1 nm) [13–17]. The cluster precursor Al13(OH)24(NO3)15 (flat-Al13), for example, produces dense and atomically smooth Al2O3 films . When the flat-Al13 precursor is combined with the nonionic block copolymer, Pluronic F127, solution deposition and processing yield a porous microstructure and reduced refractive indexes while preserving the smooth surface of the dense congener. By varying the Pluronic F127 concentration in the solution, refractive indexes can be tuned from 1.56 to 1.26. In this way, we exploit the chemical and thermal stability of Al2O3 to realize a new, simple approach to robust low-refractive-index thin films.
2. Results and discussion
We have previously reported that aqueous solutions of flat-Al13 clusters produce high quality, dense amorphous films via spin coating . Here, we anticipated porous films could be deposited directly from a micellar solution of these clusters formed by adding a soluble poloxamer. We chose the poloxamer Pluronic F127. Up to 20% w/v F127 dissolves in a flat-Al13 solution that is 0.6 M in Al. This solution, however, is too viscous for spin coating, as it produces a non-uniform coating. Viscosity decreases with lower F127 concentrations. At lower loadings of 5 and 10% F127, dynamic light scattering (DLS) shows the presence of micelles with hydrodynamic radii of 33.5(4) and 33.3(3) nm, respectively. The DLS signal for a 1% solution is below the detection limit of the instrument, suggesting this F127 concentration is below the critical micelle concentration. Thus, we focused our studies on the 5 and 10% F127 solutions.
The micellar solutions containing the clusters and F127 were deposited on SiO2/Si substrates by spin coating. Figure 1(a) shows Fourier transform infrared (FTIR) spectra for a 10% F127 film baked at 200 and 500 °C. In the 200-°C spectrum, absorption bands at 2968, 2933, and 2872 cm−1 are characteristic of the –CH3 and –CH2 groups in F127; the intense peak at 1109 cm−1 corresponds to the ether stretch . The signals centered at 3475 and 1606 cm−1 are consistent with H2O stretching and bending modes. Above 200 °C in air, the F127 begins to oxidize and evolve from the film. The m/z signals 29-88 in the temperature programmed desorption (TPD) spectrum (Fig. 1(b)) shows most of the surfactant reacts and desorbs below 300 °C. From these data, we determined a 500-°C anneal ensures substantial, if not complete, removal the surfactant. Figure 1(a) also shows the absence of F127 absorption bands after a 500-°C anneal, confirming the burnout. The broad signal appearing at 904 cm−1 after the 500-°C anneal is consistent with an Al-O phonon mode . The peaks at 1253 and 1065 cm−1 are attributable to the SiO2 layer between the Si and porous alumina film .
Figure 2 shows how F127 concentration and temperature affect film thickness. Each film was deposited from a solution 0.6 M in Al; an Al2O3 film was deposited without F127 as the control. For a 100-°C soft bake alone, Fig. 2 shows a significant increase in film thickness in step with an increasing F127 solution concentration, from 63 nm with no F127 to 354 nm with 10% F127. These film thicknesses are consistent with the FTIR data and F127’s presence in the films at low temperatures. The thickness of each film decreases substantially after an anneal at 500 °C for 1 h. For the 0, 5, and 10% F127-derived films, film thickness decreases by 70, 63, and 68%, respectively. Again, these changes reflect the concentration of F127 in the precursor. The large thickness decrease and volume change after heating to 500 °C do not adversely affect film morphology. The films remain continuous with no visible cracks, and retention of their reflective character indicates the surfaces stay smooth.
Figure 3 shows dispersion curves for the 500-°C annealed films derived from ellipsometric data. Higher F127 precursor loadings produce smaller refractive indexes. Sellmeier equations for each film are given as Eqs. (1)–(3):
At 550 nm, the control Al2O3 film exhibits a refractive index n = 1.54. Processes with the 5 and 10% F127 precursors reduce n to 1.32 and 1.26, respectively. These results are consistent with the film thickness data (Fig. 2), and they suggest F127 effects pore formation.
To evaluate porosity from the observed refractive indexes, we applied the Bruggeman effective medium model, Eq. (4),21,22]. The derived results show the 5% F127 film has a porosity of 39% and the 10% F127, a porosity of 51%. Hence, doubling the F127 concentration from 5 to 10% increases film porosity by 31%.
To determine the effect of the alumina precursor concentration on film thickness and refractive indexes, we varied the concentration of the flat-Al13 precursor in a fixed concentration of 10% w/v F127. Figure 4 shows the resulting linear relationship between Al concentration and film thickness. The film thickness, t, may be estimated from Eq. (5) and the molar concentration, c, of the Al solution:
All films from the 10% w/v F127 solutions exhibit the same refractive index (1.26 ± 0.02), independent of Al concentration. This result suggests the micellar solutions produce a constant compressive force during drying that is balanced by the stiffness of the developing aluminum oxide hydroxide framework. Consequently, film thickness may be readily tuned at a fixed refractive index by changing the Al concentration alone. This tuning should be useful to deposit precise thicknesses for multilayer optical structures.
Figure 5 shows cross-sectional transmission electron microscope (TEM) images for an Al2O3 film deposited from a 0.6-M Al solution with 10% F127. Images of dense Al2O3 films processed without F127 appear uniform and free of contrast. The high contrast, i.e., light and dark regions in the images, is consistent with pore formation. We observe no obvious ordering of the pores. The STEM image (Fig. 5, right) shows a range of pore diameters with the larger ones having diameters of 10 to 12 nm. The TEM image also shows the Al2O3 film and carbon topcoat to be discrete layers. This result suggests the porous Al2O3 film can be integrated into multilayer optical structures with smooth and sharp interfaces.
The surface smoothness was further confirmed by AFM measurements, which reveal an rms roughness of 0.5 nm. This smoothness may be unexpected, considering pores form in the bulk of the film and may extend to the surface. The smoothness likely arises from the preferred drying and condensation of the surface relative to the film body, a common trait of films deposited with oxo-hydroxo metal clusters [23,24]. The F127 likely migrates to the wet interior of the film during initial drying and condensation processes, thereby becoming trapped as the film body condenses. It later burns out to leave pores in the bulk of the film.
We annealed a bulk sample of a 0.6-M Al, 10% F127 precursor to 500 °C, then measured the surface area of the resultant powder by the Brunauer–Emmett–Teller (BET) method. The BET data are consistent with the images in Fig. 5 and previous findings that show the Barrett-Joyner-Halenda (BJH) pore size for similar powders to be 14 nm .
We tested the thermal stability of a porous film at 900 °C to determine if the thin walls of the framework might collapse. From ellipsometric data, the index of refraction increases from n = 1.26 (500 °C anneal) to 1.30, well below the control-film value of 1.67 at 900 °C. This increase is consistent with a slight densification and preservation of the porous structure. Hence, the solution-deposited, porous Al2O3 film is thermally robust to temperatures as high as 900 °C.
We collected X-ray diffraction patterns for the porous films at 500 and 900 °C to determine if they were crystalline. The diffraction pattern of the film annealed at 500 °C exhibits only peaks arising from the substrate; the Al2O3 film is X-ray amorphous. After annealing to 900 °C, the X-ray pattern shows diffraction peaks consistent with crystalline γ-Al2O3 (Fig. 6). Interestingly, the film surface remains smooth, increasing from as rms roughness of 0.5 nm at 500 °C to 0.8 nm at 900 °C. From ellipsometry data, the 1000-°C anneal increases n to 1.43 (λ = 550 nm). At this temperature, sintering takes over and the porous structure begins to collapse.
In this contribution, we have demonstrated a simple method to produce near-atomically smooth, porous alumina thin films. By spin coating an aqueous precursor containing controlled concentrations of F127 surfactant and annealing in air, the refractive indexes of alumina thin films are tuned from 1.56 to 1.26. Heating the films in air above 300 °C removes the F127. The F127 concentration in the precursor solution controls film porosity, which in turn sets the refractive index. The porous framework remains intact up to 900 °C, even as the Al2O3 framework crystallizes to the γ phase. Film thickness is readily tuned and controlled by adjusting the Al concentration in the solution. This control, coupled with the near-atomic smooth surfaces, should allow precursors and processes to be tailored for fabrication of anti-reflection coatings and multilayered optical structures on a wide variety of substrates. The lowest refractive index film should be an effective anti-reflection coating for borosilicate glasses and other materials with indexes near 1.5.
Flat-Al13 powder was synthesized by a previously reported method , then dissolved in water to produce a stock solution 0.6 M in Al. Pluronic-F127 (Aldrich) was dissolved without purification into the stock solution at concentrations of 1, 5, 10, and 20% w/v. In some cases, the stock solution was diluted prior to addition of F127.
Dynamic light scattering measurements on these solutions were made with a Brookhaven Instruments ZetaPALS instrument.
In preparation for film deposition, substrates were cleaned with a Plasma Etch PE-50 system. Films were spin-coated at 3000 rpm for 30 s on highly doped p-Si/SiO2 for the TEM experiment and on doubled-side polished lightly doped p-Si/SiO2 for FTIR analysis. After deposition, films were baked at 100 °C for 1 min, then at 200 °C for 1 min. To remove F127, films were heated at 20 °C/min to 500 °C in a Neytech Qex furnace, then soaked at that temperature for 1 h. As a control, a single coat of the flat-Al13 precursor (without surfactant) was deposited and processed with the same conditions as the films containing F127.
Spectroscopic ellipsometry measurements were made with a J. A. Woollam M-2000 instrument to determine film thickness and refractive index; data were modeled with the CompleteEASE software package . The thin SiO2 substrate layer was included in all models.
FTIR spectra were obtained using a nitrogen-purged Thermo Nicolet 6700 infrared spectrometer, equipped with a room temperature pyroelectric detector. A single-pass transmission at Brewster incidence (74°) was used to minimize the substrate phonon absorption in the low frequency region (<1000 cm–1). The spectral range 400 - 4000 cm−1 was investigated with 4 cm−1 resolution. Three loops of 500 single beam spectral scans each were obtained for each sample and averaged to obtain a spectrum. The spectrum of the clean Si substrate served to establish the baseline for the spectra of the porous alumina films.
Temperature programmed desorption (TPD) data were collected from 1 x 1 cm2 thin-film samples on a Hiden Analytical TPD workstation equipped with a quadrupole (3F PIC) to analyze gas-phase products. Measurements were made under ultra-high vacuum 5 x 10−9 Torr with 70-eV mass-analyzer ionization potential and 20-µA emission current.
TEM micrographs were obtained with an FEI Titan 80-200 TEM/STEM transmission electron microscope operated at 200 kV. Porous alumina samples were coated first with carbon by evaporation, second with chromium, and third with platinum. Thin cross sections were then selectively fabricated with a focused gallium-ion beam on an FEI Quanta 3D Dual Beam FIB. The samples were welded to a copper TEM grid and thinned to approximately 100 nm with the ion beam.
XRD patterns were collected with a Rigaku Ultima IV multipurpose X-ray diffraction system equipped with Cu Kα radiation. AFM measurements were made with an Asylum Research MFP-3D atomic force microscope by acquiring images over 1 x 1 µm2 areas.
National Science Foundation (CHE-1102637) in support of the Center for Sustainable Materials Chemistry, an NSF Center for Chemical Innovation.
Undergraduates J. T. Arens, M. T. Gutierrez-Higgins, Y. R. Jones, J. I. Lopez, T. M. Rowe, and D. M. Whitehurst helped to acquire and analyze the data presented herein. They were participants in the Sustainable Materials Research Training (SMaRT) summer research program of the Center for Sustainable Materials Chemistry. The authors thank Shannon Boettcher for helpful discussions and insight and Peter Eschbach and Theresa Sawyer for assistance with SEM and TEM imaging in the OSU Electron Microscope Facility. The TEM was acquired with funds from the M.J. Murdock Charitable Trust, the Oregon Nanoscience and Microtechnologies Institute, and the U.S. National Science Foundation (Grant No. 1040588). The authors also thank Clement Bommier and Professor Xiulei (David) Ji for surface area and pore size measurements and Jeff Nason for assistance with the dynamic light scattering instrument.
References and links
1. J.-Q. Xi, M. F. Schubert, J. K. Kim, E. F. Schubert, M. Chen, S.-Y. Lin, W. Liu, and J. A. Smart, “Optical Thin-Film Materials with Low Refractive Index for Broadband Elimination of Fresnel Reflection,” Nat. Photonics 1(3), 176–179 (2007).
2. H. K. Raut, V. A. Ganesh, A. S. Nair, and S. Ramakrishna, “Anti-Reflective Coatings: A Critical, in-Depth Review,” Energy Environ. Sci. 4(10), 3779–3804 (2011). [CrossRef]
3. T. Pilvi, E. Puukilainen, U. Kreissig, M. Leskelä, and M. Ritala, “Atomic Layer Deposition of MgF2 Thin Films Using TaF5 as a Novel Fluorine Source,” Chem. Mater. 20(15), 5023–5028 (2008). [CrossRef]
4. J. Noack, K. Scheurell, E. Kemnitz, P. Garcia-Juan, H. Rau, M. Lacroix, J. Eicher, B. Lintner, T. Sontheimer, T. Hofmann, J. Hegmann, R. Jahn, and P. Löbmann, “MgF2 Antireflective Coatings by Sol–gel Processing: Film Preparation and Thermal Densification,” J. Mater. Chem. 22(35), 18535 (2012). [CrossRef]
5. G. Atanassov, J. Turlo, J. K. Fu, and Y. S. Dai, “Mechanical, Optical and Structural Properties of TiO2 and MgF2 Thin Films Deposited by Plasma Ion Assisted Deposition,” Thin Solid Films 342(1), 83–92 (1999). [CrossRef]
6. J. Q. Xi, J. K. Kim, E. E. Schubert, D. Ye, T. M. Lu, S.-Y. Lin, and J. S. Juneja, “Very Low-Refractive-Index Optical Thin Films Consisting of an Array of SiO2 Nanorods,” Opt. Lett. 31(5), 601–603 (2006). [CrossRef] [PubMed]
7. X. A. Zhang, A. Bagal, E. C. Dandley, J. Zhao, C. J. Oldham, B.-I. Wu, G. N. Parsons, and C.-H. Chang, “Ordered 3D Thin-Shell Nanolattice Materials with Near-Unity Refractive Indices,” Adv. Funct. Mater. 25(42), 6644–6649 (2015). [CrossRef]
8. B. Pal and M. Sharon, “Enhanced Photocatalytic Activity of Highly Porous ZnO Thin Films Prepared by Sol-Gel Process,” Mater. Chem. Phys. 76(1), 82–87 (2002). [CrossRef]
9. J. Yu, X. Zhao, and Q. Zhao, “Effect of Surface Structure on Photocatalytic Activity of TiO2 Thin Films Prepared by Sol-Gel Method,” Thin Solid Films 379(1–2), 7–14 (2000). [CrossRef]
10. K. Tadanaga, K. Kitamuro, A. Matsuda, and T. Minami, “Formation of Superhydrophobic Alumina Coating Films with High Transparency on Polymer Substrates by the Sol-Gel Method,” J. Sol-Gel Sci. Technol. 26(1–3), 705–708 (2003). [CrossRef]
11. S.-B. Jung, T.-J. Ha, and H.-H. Park, “Roughness and Pore Structure Control of Ordered Mesoporous Silica Films for the Enhancement of Electrical Properties,” J. Appl. Phys. 101(2), 024109 (2007). [CrossRef]
12. X. Zhang, M. Järn, J. Peltonen, V. Pore, T. Vuorinen, E. Levänen, and T. Mäntylä, “Analysis of Roughness Parameters to Specify Superhydrophobic Antireflective Boehmite Films Made by the Sol–gel Process,” J. Eur. Ceram. Soc. 28(11), 2177–2181 (2008). [CrossRef]
13. S. W. Smith, W. Wang, D. A. Keszler, and J. F. Conley, “Solution Based Prompt Inorganic Condensation and Atomic Layer Deposition of Al2O3 Films: A Side-by-Side Comparison,” J. Vac. Sci. Technol. A Vacuum, Surfaces. Film. 32(4), 041501 (2014).
14. K. Jiang, J. T. Anderson, K. Hoshino, D. Li, J. F. Wager, and D. A. Keszler, “Low-Energy Path to Dense HfO2 Thin Films with Aqueous Precursor,” Chem. Mater. 23(4), 945–952 (2011). [CrossRef]
15. L. B. Fullmer, R. H. Mansergh, L. N. Zakharov, D. A. Keszler, and M. Nyman, “Nb2O5 and Ta2O5 Thin Films from Polyoxometalate Precursors: A Single Proton Makes a Difference,” Cryst. Growth Des. 15(8), 3885–3892 (2015). [CrossRef]
16. S. T. Meyers, J. T. Anderson, D. Hong, C. M. Hung, J. F. Wager, and D. A. Keszler, “Solution-Processed Aluminum Oxide Phosphate Thin-Film Dielectrics,” Chem. Mater. 19(16), 4023–4029 (2007). [CrossRef]
17. R. H. Mansergh, L. B. Fullmer, D.-H. Park, M. Nyman, and D. A. Keszler, “Reaction Pathway: Aqueous Hexatantalate Clusters to High-Density Tantalum Oxide Nanofilms,” Chem. Mater. 28(5), 1553–1558 (2016). [CrossRef]
19. D. N. Goldstein, J. A. Mccormick, and S. M. George, “Al2O3 Atomic Layer Deposition with Trimethylaluminum and Ozone Studied by in Situ Transmission FTIR Spectroscopy and Quadrupole Mass Spectrometry,” J. Phys. Chem. C 112(49), 19530–19539 (2008). [CrossRef]
20. J. Kwon, M. Dai, M. D. Halls, E. Langereis, Y. J. Chabal, and R. G. Gordon, “In Situ Infrared Characterization during Atomic Layer Deposition of Lanthanum Oxide,” J. Phys. Chem. C 113(2), 654–660 (2009). [CrossRef]
21. M. J. Sailor, Porous Silicon in Practice: Preparation, Characterization and Applications (Wiley-VCH, 2012), Chap. 5.
22. C. Pickering, M. I. J. Beale, D. J. Robbins, P. J. Pearson, and R. Greef, “Optical studies of the structure of porous silicon films formed in p-type degenerate and non-degenerate silicon,” J. Phys. C Solid State Phys. 17(35), 6535–6552 (1984). [CrossRef]
23. J. T. Anderson, W. Wang, K. Jiang, T. Gustafsson, C. Xu, E. L. Garfunkel, and D. A. Keszler, “Chemically Amplified Dehydration of Thin Oxide Films,” ACS Sustain. Chem.& Eng. 3(6), 1081–1085 (2015). [CrossRef]
24. K. C. Fairley, D. R. Merrill, K. N. Woods, J. Ditto, C. Xu, R. P. Oleksak, T. Gustafsson, D. W. Johnson, E. L. Garfunkel, G. S. Herman, D. C. Johnson, and C. J. Page, “Non-Uniform Composition Profiles in Inorganic Thin Films from Aqueous Solutions,” ACS Appl. Mater. Interfaces 8(1), 667–672 (2016). [CrossRef] [PubMed]
25. W. Cai, J. Yu, and M. Jaroniec, “Effect of Nonionic Structure-Directing Agents on Adsorption and Structural Properties of Mesoporous Alumina,” J. Mater. Chem. 21(25), 9066–9072 (2011). [CrossRef]
26. W. Wang, K. M. Wentz, S. E. Hayes, D. W. Johnson, and D. A. Keszler, “Synthesis of the Hydroxide Cluster [Al13(μ3-OH)6(μ-OH)18(H2O)24]15+ from an Aqueous Solution,” Inorg. Chem. 50(11), 4683–4685 (2011). [CrossRef] [PubMed]
27. J. Hale and B. Johs, CompleteEASE Data Analysis Manual (J. A. Wollam Inc., 2012)