Durable broadband antireflection (AR) coatings remain an ongoing challenge for plastic optics used in a wide range of applications. Here, we show that glancing angle deposition of a commercial fluoropolymer can be used to fabricate extremely durable ultralow index AR coatings that reduce the solar spectrum-averaged () reflectance of acrylic and polycarbonate plastic to over a wide range of incidence angles up to . The coatings feature strong adhesion and exhibit outstanding resistance to heat, humidity, dirt, ultraviolet light, outdoor exposure, solvents, acids, bases, abrasion, and repeated bend/compression cycling. They are successfully applied to f/1 curved lens surfaces as well as acrylic Fresnel lenses, where coating both sides increases the solar spectrum-averaged transmittance from 92% to 98%. These results represent a significant development for plastic optics commonly used in solar concentrators as well as more generally for broadband AR applications that demand extreme environmental, chemical, and mechanical durability.
© 2017 Optical Society of America
Broadband antireflection (AR) coatings play an important role in applications ranging from optical lenses and imaging systems to displays, solar concentrators, and photovoltaics. Graded refractive index coatings and textures fabricated using a wide variety of methods [1–11] represent a common impedance-matching AR strategy that is simultaneously capable of suppressing reflection over a broad range of incident angles and wavelengths. However, combining this performance with the extreme environmental, mechanical, and chemical durability demanded in, for example, solar concentrator optics remains challenging, particularly for those made of acrylic or polycarbonate plastic.
Traditional inorganic AR coating strategies are not usually well-suited for plastic optics owing to poor adhesion, large thermal expansion mismatch, and inherent limitations in coating process temperature . Ophthalmic multilayer AR coatings developed for plastic eyeglass lenses are perhaps the most successful and durable solution to date; however, bandwidth is usually limited to the visible spectral range and the complexity (most involve four or more layers) represents an obstacle to more widespread application [13,14]. Single layer organic fluoropolymer AR coatings deposited from solution are also used in many commercial applications, but they are similarly limited in AR bandwidth . Recent efforts have sought improved AR performance by nanostructuring the polymer surface to grade the refractive index through a series of low pressure plasma etching steps [15,16], though this does not confer any of the durability-related benefits that are also typically desired from an AR coating.
Here, we introduce multilayer graded index AR coatings fabricated via glancing angle deposition (GLAD) [17–21] of the commercial fluoropolymer Teflon AF. Simple bilayer coatings adhere strongly to plastics such as acrylic and polycarbonate and reduce the solar spectrum-averaged () reflectance to over a wide range of incidence angles. The coatings are abrasion resistant and survive repeated mechanical bend and compression cycles on substrates flexed to a 1 cm radius. They are hydrophobic, with a water contact angle that supports anti-fogging behavior, and they are impervious to most organic solvents, acids, and bases. They are unaffected by prolonged ultraviolet light exposure and exhibit no deterioration in AR performance after ten days of damp heat testing ( and ) or after one month of continuous rooftop outdoor exposure. The coatings are successfully applied to f/1 curved lens surfaces as well as to both sides of an f/2 acrylic Fresnel lens, which increases the solar spectrum-averaged transmittance from approximately 92% to 98%. The combination of broadband, omnidirectional AR performance, extreme durability, and curved surface compatibility of the coatings introduced here should find widespread use in the plastic optics industry.
Figure 1(a) depicts the basic GLAD geometry, where the substrate is tilted at angle relative to the vapor flux incident from the evaporation source. In general, GLAD is a well-established technique for creating nanoporous films due to the self-shadowing effect [22,23], which enables the refractive index to be controlled over a wide range and has been exploited in numerous graded index AR coatings based on inorganic materials such as , , and [7,17–21]. Figure 1(b) shows that GLAD also works for thermally evaporated films of Teflon AF 1600 (amorphous fluoropolymer powder available from Chemours Co.), where the refractive index dispersion measured via spectroscopic ellipsometry varies from to for and , respectively. The films are deposited at a nominal rate of from a chamber base pressure of using a temperature-controlled substrate chuck set at and a source-to-substrate distance of approximately 30 cm (which achieves deposition uniformity across a 100 mm wafer). In general, a combination of high deposition rate and low substrate temperature is important for achieving the lowest refractive index at high deposition angles. All the films are specular by eye and transparent deep into the ultraviolet range for wavelengths . Figure 1(c) shows a cross-sectional scanning electron micrograph of a 305 nm thick film deposited on Si at , which exhibits the usual tilted nanostructure  and confirms that the film porosity is deeply subwavelength.
The index range in Fig. 1(b) enables ideal single layer and near-ideal double layer quarter wave AR coatings for acrylic (polymethylmethacrylate, ) and polycarbonate substrates () [24,25]. Focusing on the double layer coating as a practical compromise between complexity and performance for solar concentrator applications, Fig. 2(a) shows the solar spectrum-averaged reflectivity ( at normal incidence) predicted for a coating composed of () and () GLAD layers with varying thickness on an acrylic substrate. Targeting the broad reflectivity minimum displayed in this contour plot, a bilayer coating was subsequently deposited on an acrylic substrate with thicknesses and determined via ellipsometry. Figure 2(b) shows the resulting single side reflectance spectrum measured at incident angle by roughening the back side of the substrate and painting it black. This result agrees well with the model prediction (blue dashed line) and leads to , which is nearly an order of magnitude lower than for the bare acrylic surface (solid black line). Model predictions for ideal single and triple layer AR coatings (red and green dashed lines, respectively) are included to convey the diminishing marginal improvement in AR performance with increasing layer number that motivates the bilayer design choice.
Figures 2(c) and 2(d) show the angle dependence of for bilayer AR coatings applied to acrylic and polycarbonate substrates, respectively. In both cases, the average reflectivity is strongly suppressed relative to the bare substrate at all angles, maintaining up to incidence angle and up to . To our knowledge, this is the best broadband, omnidirectional AR performance reported for acrylic and polycarbonate to date.
The naturally hydrophobic nature of fluoropolymers  combined with the nanoporosity of the GLAD films leads to a high water contact angle of approximately 140° on both acrylic and polycarbonate, as shown in Fig. 3(a). This in turn leads to anti-fogging behavior, as demonstrated in the photograph of Fig. 3(b), by cooling a partially coated sheet of acrylic below the ambient dew point. The GLAD AR coatings also inherit the outstanding chemical compatibility of Teflon. We find that bilayer AR-coated acrylic samples can withstand pools of harsh organic solvents such as toluene, xylene, and chlorobenzene on the surface for up to 5 min with no deterioration.
Strong adhesion to acrylic and polycarbonate is inferred from the resistance of coatings to multiple “sharp pull” Scotch tape tests ; they remain intact with no degradation in AR performance after sonicating in water and isopropyl alcohol for one hour. Abrasion resistance is evaluated by pulling a cheesecloth across a bilayer AR-coated acrylic surface, weighted to deliver a set pressure as outlined in MIL-SPEC and ISO standards [28–30]. At pressures less than , only slight surface scratching is evident under the microscope and there is a negligible increase in solar spectrum-averaged reflectivity (). Surface damage and reflectivity change becomes evident for pressures . The coatings are also flexible and withstand the stress and strain of extensive bend/compression cycling, as shown in Fig. 3(c). There, no statistically significant change in reflectance was observed for a bilayer AR-coated, 75 μm thick acrylic sheet flexed in both tension and compression around a 1 cm radius rod over the course of 500 cycles.
Environmental stability is evaluated through indoor damp heat testing, ultraviolet exposure, and outdoor rooftop testing. Damp heat testing at 85°C and 85% relative humidity for ten days revealed no change in AR performance on acrylic substrates. Similarly, indoor exposure to intense ultraviolet light from a Xe lamp with a power density of in the 275–375 nm wavelength range (the ultraviolet equivalent of approximately 19 suns) produced no measurable change in reflectivity after ten days of continuous exposure. Figure 3(d) shows the reflectance change recorded over the course of one month for bilayer-coated acrylic samples ( and ) placed on the rooftop of the Penn State Electrical Engineering East building facing south at latitude tilt during late summer. The slight increase in , shown in the inset, results mainly from the buildup of surface contaminants over time, as samples were not washed prior to each reflectance measurement. Cleaning with water and isopropanol at the conclusion of the experiment restores to its pristine value.
Given the GLAD layer sensitivity to deposition angle, it is important to understand the extent to which these AR coatings can be successfully applied to curved lens surfaces. Thickness and refractive index uniformity are evaluated by depositing a single GLAD layer onto a narrow strip of Ag-coated Kapton tape draped over the curved side of an f/2 planoconvex lens (25.8 mm radius of curvature) while rotating it azimuthally [i.e., about the direction in Fig. 1(a)] at a rate of 30 revolutions per minute. Figure 4(a) shows the thickness and refractive index measured via ellipsometry at seven locations along the strip corresponding to the lens surface locations shown in the inset. Both the thickness and refractive index vary by less than 4% relative to the mean, which is expected to have a negligible impact on AR performance based on the large tolerance associated with the reflectivity minimum in Fig. 2(a).
This is indeed the case found from direct transmission measurements on an f/1 acrylic planoconvex lens (12.5 mm radius of curvature) with a bilayer AR coating deposited on the curved surface. Figures 4(b) and 4(c), respectively, show the transmission spectra measured (using an integrating sphere detector) at the center and four edge locations of both the bare and AR-coated lenses. As highlighted in Fig. 4(d), the transmission difference between the AR and bare lenses is largely independent of the measurement location. Because the transmission difference is approximately equal to the reflectivity reduction of the AR-coated curved surface when absorption in the lens bulk is negligible (i.e., for ), the close overlay of the data in Fig. 4(d) confirms that the AR coating functions uniformly over the entire f/1 curved surface. Performing the same experiment on an f/4 lens (49.8 mm radius of curvature) results in a similar conclusion.
Figure 5 presents a direct application demonstration relevant to concentrating photovoltaics. In this case, the bilayer AR coating is deposited on both sides of a f/2 acrylic Fresnel lens with a groove density of . Figure 5(a) compares the transmission spectrum of a coated and uncoated lens, confirming an extraordinary broadband improvement that increases the solar spectrum-averaged transmittance from to . The AR performance is also visually apparent by eye, as shown in Fig. 5(b), where the reflection of fluorescent room lights in a selectively coated region of the Fresnel lens (denoted by the red dashed line) is strongly suppressed. Altogether, after AR coating more than 25 different lenses over the span of several months in a related solar concentrator effort, we find these results to be very reproducible.
The same AR coatings also perform well on glass, but the adhesion is much weaker and they are easily rubbed off when handled directly with gloves. It may be possible to address this issue by first treating the glass surface with a fluorosilane coupling agent to promote adhesion with the fluoropolymer; however, this hinges on the more basic question of what the final composition of the deposited film actually is. Previous investigations suggest that the polymer chains cleave into fragments between adjacent dioxole rings during the evaporation process and subsequently repolymerize on the target substrate [31–33]. This is consistent with an increase in background chamber pressure (from up to ) that we observe during each deposition and may underlie the strong adhesion of evaporated Teflon AF to other polymer substrates since these small molecule fragments may diffuse into the host chain network to some extent before repolymerizing. This notion is supported by our observation that AR coating adhesion is further improved by depositing the first layer at a low rate (<0.5 nm/s) onto a heated substrate, which would be expected to enhance diffusion and intermixing.
If this essentially mechanical mode of chain interlocking adhesion is indeed the case, it seems likely that the results achieved here for acrylic and polycarbonate will also apply more broadly to other optical polymers; preliminary results demonstrating strong adhesion to Zeonex 350R and TOPAS COC polymers support this conclusion. Moving forward, it will be important to better understand the chemical nature of GLAD Teflon AF films together with any such interfacial intermixing with the substrate, as this may provide a path to further improve the mechanical robustness of these coatings.
In conclusion, we have demonstrated broadband, ultralow index AR coatings for acrylic and polycarbonate plastic based on glancing angle-deposited Teflon AF that offers an extraordinary combination of mechanical, chemical, and environmental durability. Simple bilayer AR coatings are shown to reduce the solar spectrum-averaged reflectance of acrylic to less than 0.5% over a wide range of incidence angles and have been successfully applied to curved lens surfaces and Fresnel lenses alike. This AR strategy is compatible with standard commercial vacuum coating systems and should therefore find widespread use in the rapidly growing number of applications in which plastic optics are employed.
National Science Foundation (NSF) (CBET-1508968); Advanced Research Projects Agency-Energy (ARPA-E) (DE-AR0000626).
We thank Enrique Gomez for helpful discussions.
1. L. Ye, Y. Zhang, X. Zhang, T. Hu, R. Ji, B. Ding, and B. Jiang, Sol. Energy Mater. Sol. Cells 111, 160 (2013). [CrossRef]
2. X. Zhang, S. Cai, D. You, L. H. Yan, H. B. Lv, X. D. Yuan, and B. Jiang, Adv. Funct. Mater. 23, 4361 (2013). [CrossRef]
3. H. Park, D. Shin, G. Kang, S. Baek, K. Kim, and W. J. Padilla, Adv. Mater. 23, 5796 (2011). [CrossRef]
4. B. Wang and P. W. Leu, Nanotechnology 23, 194003 (2012). [CrossRef]
5. L. Liu, X. Wang, M. Jing, S. Zhang, G. Zhang, S. Dou, and G. Wang, Adv. Mater. 24, 6318 (2012). [CrossRef]
6. B. Wang, E. Stevens, and P. W. Leu, Opt. Express 22, A386 (2014). [CrossRef]
7. J. Q. Xi, M. F. Schubert, J. K. Kim, E. F. Schubert, M. Chen, S. Y. Lin, W. Liu, and J. A. Smart, Nat. Photonics 1, 176 (2007).
8. E. R. Klobukowski, W. E. Tenhaeff, J. W. McCamy, C. S. Harris, and C. K. Narula, J. Mater. Chem. C 1, 6188 (2013). [CrossRef]
9. X. Li, J. Gao, L. Xue, and Y. Han, Adv. Funct. Mater. 20, 259 (2010). [CrossRef]
10. D. M. Sim, M. J. Choi, Y. H. Hur, B. Nam, G. Chae, J. H. Park, and Y. S. Jung, Adv. Opt. Mater. 1, 428 (2013). [CrossRef]
11. A. Rahman, A. Ashraf, H. Xin, X. Tong, P. Sutter, M. D. Eisaman, and C. T. Black, Nat. Commun. 6, 5963 (2015). [CrossRef]
12. U. Schulz, Handbook of Plastic Optics, S. Baumer, ed. (Wiley, 2010), pp. 161–195.
13. F. Samson, Surf. Coat. Technol. 81, 79 (1996). [CrossRef]
14. U. Schulz, U. B. Schallenberg, and N. Kaiser, Appl. Opt. 41, 3107 (2002). [CrossRef]
15. U. Schulz, F. Rickelt, P. Munzert, and N. Kaiser, Opt. Mater. Express 4, 568 (2014). [CrossRef]
16. U. Schulz, P. Munzert, R. Leitel, I. Wendling, N. Kaiser, and A. Tunnermann, Opt. Express 15, 13108 (2007). [CrossRef]
17. J. Xi, J. K. Kim, and E. F. Schubert, Nano Lett. 5, 1385 (2005). [CrossRef]
18. M. M. Hawkeye and M. J. Brett, J. Vac. Sci. Technol. A 25, 1317 (2007). [CrossRef]
19. A. Barranco, A. Borras, A. R. Gonzalez-Elipe, and A. Palmero, Prog. Mater. Sci. 76, 59 (2016). [CrossRef]
20. K. Robbie and M. Brett, J. Vac. Sci. Technol. A 15, 1460 (1997). [CrossRef]
21. S. R. Kennedy and M. J. Brett, Appl. Opt. 42, 4573 (2003). [CrossRef]
22. L. Abelmann and C. Lodder, Thin Solid Films 305, 1 (1997). [CrossRef]
23. K. Robbie, J. Sit, and M. J. Brett, J. Vac. Sci. Technol. B 16, 1115 (1998). [CrossRef]
24. J. Zhao and M. A. Green, IEEE Trans. Electron. Devices 38, 1925 (1991). [CrossRef]
25. H. A. Macleod, Thin-Film Optical Filters, 3rd ed. (IOP, 2001).
26. S. R. Coulson, I. Woodward, P. S. Badyal, S. A. Brewer, and C. Willis, J. Phys. Chem. B 104, 8836 (2000). [CrossRef]
27. D. S. Campbell, L. I. Mai, and R. Glang, Handbook of Thin Film Technology (McGraw-Hill, 1970), Chap. 12.3.
28. “Military specification:coating of glass optical elements (anti-reflection),” MIL-C-675C (U.S. GPO, 1980), Para. 126.96.36.199.
29. “Optics and optical instruments - Optical coatings - Part 4: Specific test methods,” DIN ISO 9211-4 (2008).
30. K. H. Guenther, Appl. Opt. 23, 3612 (1984). [CrossRef]
31. P. S. Ho, J. Leu, and W. W. Lee, Low Dielectric Constant Materials for IC Applications (Springer, 2003), p. 115.
32. S. L. Madorsky, V. E. Hart, S. Straus, and V. A. Sedlak, J. Res. Natl. Bur. Stand. 51, 327 (1953). [CrossRef]
33. T. C. Nason, J. A. Moore, and T. M. Lu, Appl. Phys. Lett. 60, 1866 (1992). [CrossRef]