Robust operation of mesoporous antireflective coatings under variable ambient conditions

Antireflective coatings (ARCs) are important components in optical applications due to their ability to reduce reflection and maximize transmission of light across interfaces. Generating mesoporous films with adequate film thickness and refractive index is a common method to achieve amplitude and phase matching in low-cost single- layer interference-based ARCs. For high surface energy materials, pores on the 2 - 50 nm, i.e. the sub-wavelength scale, are subject to capillary condensation, and pore filling by surrounding gas phase water molecules at size-characteristic humidity values hampers their functioning. In this work, we examine the effect of relative humidity on mesoporous ARCs and present a simple method for the preparation of ARCs with robust operation under variable conditions. The materials route is based on the generation of well-defined porous aluminosilicate networks by block copolymer co- assembly with poly(isobutylene)block-poly(ethylene oxide) and post-synthesis grafting of trichloro(octyl)silane molecules to the pore walls. The functionalized films exhibited a maximum transmittance value of 99.8% with an average transmittance of 99.1% in the visible wavelength range from 400 nm to 700 nm. Crucially, the AR performance was maintained at high humidity values with an average transmittance decrease of only 0.2% and maximum values maintained at 99.7 %, which compared to maximum and average losses of 3.6% and 2.7 %, respectively, for non-functionalized reference samples. The ARCs were shown to withstand at least 50 humidity cycles, indicating long-term stability against fluctuating environmental conditions. Abstract Antireﬂective coatings (ARCs) are important components in optical applications due to their ability to reduce reﬂection and maximize transmission of light across interfaces. Generating mesoporous ﬁlms with adequate ﬁlm thickness and refractive index is a common method to achieve amplitude and phase matching in low-cost single-layer interference-based ARCs. For high surface energy materials, pores on the 2 50 nm, i.e. the sub-wavelength scale, are subject to capillary condensation, and pore ﬁlling by surrounding gas phase water molecules at size-characteristic humidity values hampers their functioning. In this work, we examine the eﬀect of relative humidity on mesoporous ARCs and present a simple method for the preparation of ARCs with robust operation under variable conditions. The materials route is based on the generation of well-deﬁned porous aluminosilicate networks by block copolymer co-assembly with poly(isobutylene) block -poly(ethylene oxide) and post-synthesis grafting of trichloro(octyl)silane molecules to the pore walls. The functionalized ﬁlms exhibited a maximum transmittance value of 99.8 % with an average transmittance of 99.1 % in the visible wavelength range from 400 nm to 700 nm. Crucially, the AR performance was maintained at high humidity values with an average transmittance decrease of only 0.2 % and maximum values maintained at 99.7 %, which compared to maximum and average losses of 3.6 % and 2.7 %, respectively, for non-functionalized reference samples. The ARCs were shown to withstand at least 50 humidity cycles, indicating long-term stability against ﬂuctuating environmental conditions.


Introduction
Antireflective coatings (ARCs) are a common component in fields such as eye glasses, display technology, solar cells and windows. 1 Light reflection at optical interfaces is a consequence of a contrast in refractive index, and thus one of the most appealing strategies for ARCs involves the generation of gradient structures on the wavelength scale. 2 Examples include biomimetic moth eye-type features, which consist of sub-wavelength nanopillars deposited or etched directly on a surface. 3 Top-down lithographical methods were first utilized to replicate these structures. 4 Nowadays, substrate etching methods for glass involve masks made of polystyrene colloids, 5 metallic nanoparticles 6 or block copolymers. 7 However, these methods are typically challenged by convoluted processes with numerous fabrication steps and issues related to their scalability. 3 An alternative approach are interference-based ARCs. A single-layer coating must fulfill two requirements: 8 1) the optical path length of the film n f ilm × d must be one quarter of the wavelength of the incident light (λ/4) and 2) the refractive index (RI) of the material needs to equate to the square root of the product substrate RI (n ARC = √ n air n substrate ). Given that ARCs are typically coated on glass substrates with RI of around 1.5, a thin film fulfilling anti-reflection (AR) conditions should have a target RI value ≈ 1.22-1. 23. Dense solids do not exhibit an RI value below 1.4, and thus, in order to achieve interference-based AR in a single layer, the introduction of porosity on the sub-wavelength scale is required.
Spin-coated polymer blends produced sub-nanometre structures via selective solvent etching of one block. 9 Other deposition methods include convective self-assembly 10 and electrostatic deposition 11 of silica nanoparticle-based materials. Pre-formed silica nanoparticles have been combined with sol-gel based silica "binder" to obtain AR coatings, albeit at higher than optimum RI values. 12 Recently, coatings with AR properties were produced from a sequential infiltration method which introduced gas phase alumina into swollen polymer films using atomic layer deposition. 13 Broadband functioning is facilitated by multilayer assemblies. 14 Structure-directing agents such as surfactants and block copolymers (BCPs) have been extensively studied for the formation of a variety of inorganic mesoporous architectures with closely controlled porosity and pore size. 15 Well established sol-gel chemistry methods, when combined with solution-based BCPs, allow for the facile preparation of mesoporous inorganic films via conventional liquid deposition processes. 16 influences the hydrolysis and condensation processes. 29 The potential loss of organic functionality during high temperature template removal 30  and glass (Pilkington Microwhite) substrates (1500rpm, 20 seconds). All substrates were first treated with oxygen plasma for 5 minutes followed by cleaning with a Snowjet CO 2 cleaner (Megatech Ltd.) immediately before deposition. The as-made hybrid films were annealed on a hot plate to a temperature of 130 • C (1 • C per min ramp, 30 minute dwell time) followed by calcination in a furnace at 450 • C for 1 hour.

Hydrophobic functionalization
Samples were allowed to cool post-calcination and subsequently placed in a covered glass Petri dish containing trichloro(octyl)silane (TOS, 97 %, Sigma Aldrich) diluted to 0.1mM in heptane (≥ 99.9 %, Sigma Aldrich) for 5 hours. Following removal, all samples were washed thoroughly using heptane and dried under a stream of nitrogen.

Material Characterization
Spectroscopic ellipsometry (SE) and environmental water-based ellipsometric porosimetry (EP) measurements were performed using a spectroscopic ellipsometer (Semilab SE2000) and analyzed with the software of the manufacturer (SEA). EP measurements with toluene as sorbent were performed under vacuum in a dedicated chamber using a Semilab PS2000 ellipsometer. All measurements were carried out at an incident angle of 73 • on a silicon substrate. Data fitting was performed using a standard Cauchy dispersion law. The Lorentz-Lorentz effective medium approximation model was used to calculate the relative volume of adsorbed water (or toluene) based on the polarizability and molar volume of the sorbent. 38 Prior to EP measurements, samples were heated on a hotplate at 120 • C for several minutes to ensure that exclude residual water in the pores. Transmittance measurements were performed using a Zeiss Axioscope A1 with 5x magnification coupled with 400 µm optical fibre (Ocean Optics) and a high resolution spectrometer (Ocean Optics, QE65000). Humidity was controlled during in situ transmittance measurements using mass flow controllers 6 (Bronkhorst, F-201CV-100) connected to an enclosed microscope stage (Linkam, LTS120) which was modified with the integration of a humidity sensor. All RH values were reported with an accuracy of ±5 %. Measurements were recorded 1 hour after stabilization of the desired relative humidity value. Contact angle measurements were performed on a drop shape analyzer (Kruss, DSA100). SEM images were carried out on a Carl Zeiss Orion Nanofab at an accelerating voltage of 25kV. Samples were sputter coated with platinum before imaging. the isotherm shape was found to be consistent with a type IV IUPAC classification associated to mesoporous materials. 39 The broadness of the loop and gradual slope of the adsorption curve is in line with a type H2 hysteresis attributed to "ink bottle" type pore interconnections. In this geometry, which is consistent with the observed inverse opal-type morphology, the emptying of non-surface pores during desorption is limited by the pore necking. 40,41 Based on Figure 2b, it is evident that ARC 1:3 samples undergo capillary condensation in the range of 45-60 % relative humidity (RH), thus no longer fulfilling the requirements for AR. The structure formation principle via dense packing of sol-loaded block copolymer micelles allows to tune the RI over a wide range by adjusting the O:I ratio. 42 The measured porosity and RI values compare favorably with similar samples that were prepared from higher molecular weight poly(isoprene)-block -poly(ethylene oxide) (PI-b-PEO) block copolymers. 18 These low RI values provide the potential for incorporating functional materials into the porous network while still maintaining AR conditions. Here, ARC 1:1nf samples were placed in a solution of trichloro(octyl)silane (TOS, RI ≈ 1.45 43 ) to functionalize the surface with non-polar long chain hydrocarbons via the reaction of the chlorosilane with exposed hydroxyl groups on the surface of the porous network (see also Figure 1). Trichlorosilanes were chosen as they exhibited greater surface coverage on silica materials when compared to monochloro or dichloro moieties. 44 EP was also used to monitor the stability of the functionalized coatings when subjected to repeated dry and wet air cycles. This approach provides a viable alternative for the testing of environmental durability of coatings compared to the standard tests which involve exposing the sample to constant RH values for extended periods of time followed by measurement of the material properties ex situ. 17,45 In earlier work on silica-based networks, repeated cycles of water adsorption and desorption demonstrated an impact on the mechanical integrity of the porous skeleton as well as hydrophobicity via capillary stresses and irreversible adsorption of water molecules within the pores. [46][47][48] In the present study, the ARC 1:1f sample maintained both its hydrophobicity and AR capabilities after 50 EP cycles. The average RI at 97 % RH was ≈1.22 with none of the RI measurements exceeding 1.23 ( Figure S3). Previous studies determined the effect of adsorption and desorption of water by varying temperature. 49 To our knowledge, however, this is the first reported work on the effects of humidity cycling on ARC stability.

Results and discussion
When light passes through float glass, intensity is lost from reflections at each air/glass interface. Typically, this results in the loss of ≈4 % of transmitted light at each interface.
In the present study, uncoated Pilkington Microwhite exhibited an average transmittance of 92.4 % between wavelengths of 400 to 700 nm with a maximum value of 92.6 % ( Figure   S4). When

Conclusions
In conclusion, we report in this work on materials and characterisation aspects of mesoporous ARCs for robust operation under variable ambient conditions. Ellipsometric porosimetry and transmission experiments with controlled relative humidity enabled to study in situ the impact of humidity-induced capillary condensation on the optical properties of ARCs.
The use of an amphiphilic PIB-b-PEO block copolymer as structure-directing agent for the co-assembly of aluminosilicate material derived from sol-gel chemistry allowed to deposit mesoporous optical coatings with broad porosity and thus refractive index tuning.
ARCs were therefore fabricated that met amplitude-and phase-matching conditions after grafting of a hydrophobic agent to the pore walls, here trichloro(octyl)silane (TOS). As a result, near-optimum optical performance was achieved under variable atmospheric conditions, with a maximum transmittance value for double-sided substrates of 99.8 % under dry and 99.7 % near saturation and an average transmittance in the wavelength range from 400 nm to 700 nm of 99.1 % and 98.9 %, respectively. The ARCs were shown to withstand repeated humidity cycles and offer a reliable route to low-cost solution-processable coatings on glass and other transparent substrates.    Figure S4: Refractive index values of ARC 1:1f samples after multiple EP cycles