Preparation of Water-Resistant Antifog Hard Coatings on Plastic

Nov 22, 2012 - (1, 2) In applications where high solar energy input is demanded, such as solar cells, fogging could reduce the light transmittance and...
0 downloads 6 Views 2MB Size
Article pubs.acs.org/Langmuir

Preparation of Water-Resistant Antifog Hard Coatings on Plastic Substrate Chao-Ching Chang,†,‡ Feng-Hsi Huang,† Hsu-Hsien Chang,† Trong-Ming Don,†,‡ Ching-Chung Chen,‡ and Liao-Ping Cheng*,†,‡ †

Department of Chemical and Materials Engineering, and ‡Energy and Opto-Electronic Materials Research Center, Tamkang University, New Taipei City, Taiwan, 25137 S Supporting Information *

ABSTRACT: A novel water resistant antifog (AF) coating for plastic substrates was developed, which has a special hydrophilic/hydrophobic bilayer structure. The bottom layer, acting both as a mechanical support and a hydrophobic barrier against water penetration, is an organic−inorganic composite comprising colloidal silica embedded in a cross-linked network of dipentaethritol hexaacrylate (DPHA). Atop this layer, an AF coating is applied, which incorporates a superhydrophilic species synthesized from Tween-20 (surfactant), isophorone diisocyanate (coupling agent), and 2-hydroxyethyl methacrylate (monomer). Various methods, e.g., FTIR, SEM, AFM, contact angle, and steam test, were employed to characterize the prepared AF coatings. The results indicated that the size and the continuity of the hydrophilic domains on the top surface increased with increasing added amount of T20, however, at the expense of hardness, adhesiveness, and water resistivity. The optimal T20 content was found to be 10 wt %, at which capacity the resultant AF coating was transparent and wearable (5H, hardness) and could be soaked in water for 7 days at 25 °C without downgrading of its AF capability.



interest.7−10 For example, Howarter et al. grafted hydrophilic species, poly(ethylene glycol) (PEG), onto a glass surface through the use of a silane-type coupling agent that bridged PEG and glass. Both good adhesion and hydrophilicity were attained via this method. However, the treated surface became very soft due to the presence of the hydrophilic agent. Even small mechanical impacts can give rise to visible scratching marks. Furthermore, because complex chemical processes were involved, it is impractical to apply this method to substrates of large surface area,8,9 such as building windows. III. Hydrophilic components are incorporated into the coating formula, which is then photo or thermal-cured to form an AF layer on the surface. Unlike method I, the hydrophilic species are chemically cross-linked to the matrix in this case. Although method III is applicable to virtually any kind of substrate materials (plastics are of particular interest), the interface between coating layer and substrate is susceptible to water invasion due to absorption by hydrophilic groups. The AF layer may swell and detach from the substrate in very humid environments.11 Alternatively, on glass substrates, an AF layer may be formed consisting of inorganic oxides such as TiO2, Cd2O3, ZnO, or ZrO2. Although it has the benefits of good adhesion and surface hardness, the high temperature process associated with oxide formation

INTRODUCTION Fog forms when saturated water vapor condenses in the form of small droplets capable of scattering visible light on a surface with a temperature lower than the dew point of the vapor. This undesirable phenomenon occurs frequently on bathroom mirrors, eyeglasses, swimming goggles, windshields, camera lens, etc., items which are closely related to our everyday life. Surface fog can reduce the precision of optical and analytical instruments, such as infrared microscopes or bronchoscopes.1,2 In applications where high solar energy input is demanded, such as solar cells, fogging could reduce the light transmittance and bring down the efficiency of energy usage.3 The basic concept of antifog is to create a hydrophilic surface such that arriving water droplets would spread and naturally form a continuous or nearly continuous water film on the surface, so that light can transmit directly free of interfering scattering from water microdroplets.4,5 Preparation of a hydrophilic surface generally falls into three categories: I. Hydrophilic agents are physically introduced into the polymer matrix without chemical bond formation. Usually, a simple process such as solution or melt blending is good enough to yield an effective AF surface. However, the hydrophilic agents may come off the coating surface during cleaning, and thus, stable long-term wettibility cannot be assured. As a result, this approach is only suited to products of short life cycles, such as food packaging.6 II. Chemical modification was performed directly on the surface of © 2012 American Chemical Society

Received: October 22, 2012 Revised: November 21, 2012 Published: November 22, 2012 17193

dx.doi.org/10.1021/la304176k | Langmuir 2012, 28, 17193−17201

Langmuir

Article

Scheme 1. Preparation of (a) Silica Sol and (b) MSiO2 Sol and Polyacrylate-Silica Thin Film (Bottom Layer)



precludes it from being applied on plastic materials.12−18 A comprehensive review of the articles for preparation and design of such hydrophilic coatings has been carried out by Feng et al.19 In recent years, biomimetic, raspberry-like, or nanoporous antireflective/antifogging films composed by silica (SiO2) and titania (TiO2) nanospheres have been demonstrated.20−28 However, to improve the mechanical durability, robustness, and adhesion of the films, the resultant films have to be calcinated at 500 °C. As a matter of fact, there are several hurdles to overcome in order to prepare a robust, long-life, antifog coating on plastic substrates. First, if direct surface modification is to be implemented, one should consider the fact that plastics are sensitive to both temperature and solvent damaging. Second, the coating layer should adhere strongly to the substrate and should not deform or peel off when used in highly humid environments or when cleaning is required. Third, hardness of the coating should comply with the application criterion. In the current research, a new approach is adopted in light of the above concerns. The prepared antifog coating has a special bilayer configuration, for which the bottom layer (primer) is a cross-linked network of poly(acrylate) embedded with nanosized silica that provides mechanical strength and adhesiveness. On top of the primer is lain the AF layer, in which surface active agent, Tween 20, is modified and covalently incorporated into the polymer host. Because the top and bottom layers are interlinked by UV-cured poly(acrylate), these two layers are mechanically inseparable and appear transparent as if they were a uniform layer. Furthermore, the coating can be rinsed in water and remain integrality after drying. Thanks to the hydrophobic feature of the primer, water molecules cannot penetrate through the coating and detach it from the substrate. The detailed synthesis and characterization of the developed antifog coating is demonstrated in the following sections.

EXPERIMENTAL SECTION

Chemicals. 3-(Trimethoxysilyl) propyl methacrylate (MSMA, Degussa), tetraethoxysilane (TEOS, Fluka), sorbitan monolaurate (Tween 20, Mw = 1227.5, Aldrich), isophorone diisocyanate (IPDI, Aldrich), 2-hydroxyethyl methacrylate (2HEMA, Aldrich), dipentaethritol hexaacrylate (DPHA, Aldrich), Darocure 1173 (Ciba-Geigy), methyl ethyl ketone (MEK, Fluka), 2-propanol (IPA, Aldrich) were at regent grade or higher, and used as received. Synthesis of Surface-Modified Silica. The silica sol for the bottom layer was synthesized by a modified sol−gel method, cf. Scheme 1.29−32 An appropriate amount of TEOS was mixed with 2-propanol to form a homogeneous solution. To this solution, the catalyst HCl (aq) (pH 1.2) was added to induce hydrolysis−condensation of TEOS. The molar ratio of H2O/TEOS was set to be equal to four. The mixture was stirred for 3 h at room temperature to yield a silica sol, and then additional HCl (aq) (pH 1.2) and coupling agent, MSMA, with a molar ratio of H2O/MSMA = 3 and TEOS/MSMA = 4.5, were slowly dropped into the silica sol, cf. Scheme 1b. After reaction for another 3 h, the surface-modified silica (termed MSiO2, containing CC on their surface) was obtained. Preparation of the Antifog Coating. Photocurable coating sol was prepared by adding 10 g of the multifunctional cross-linking agent (DPHA), 0.59 g of the photoinitiator (Darocure 1173), and 17.68 g of 2-propanol into 20 g of the assynthesized MSiO2 sol with a total solid content (theoretical) adjusted to 30 wt %. The sol was spin-coated on a poly(methyl methacrylate) (PMMA) substrate, and then prebaked at 80 °C for 40 s, followed by UV-irradiation (broadband, 250 mJ/cm2) to obtain a cured film. The thickness of the film was measured to be ∼1.2 μm by interferometry. The radiation power was carefully chosen such that DPHA was only partly cross-linked, allowing for subsequent reaction with the top AF layer during UV-curing of the latter. 17194

dx.doi.org/10.1021/la304176k | Langmuir 2012, 28, 17193−17201

Langmuir

Article

Scheme 2. (a) Reaction of IPDI and 2-HEMA to Form the Intermediate Molecule 2-HEMA/IPDI and (b) Proposed Reaction Mechanism for the Synthesis of UV Curable Hydrophilic Agent T20

FTIR Spectra. Fourier transform infrared (FTIR) absorption spectra of the formed MSiO2 sol and the cured coatings were obtained using a Nicolet 550 spectrometer. Cured thin films were ground with KBr (1:50) and pressed to form a disc for FTIR scanning. Liquid samples were prepared by dropping appropriate amount of the sol onto a KBr disc, and then the solvent was evaporated at 40 °C for 10 min in a vacuum oven. Particle Size Determination. The size and size distribution of particles in the sols were determined by the dynamic light scattering (DLS) analysis using a Zetasizer (Malvern, DTS 1060) at 25 °C. The instrument was equipped with a monochromatic coherent helium neon laser (633 nm) as the light source. A 4 mL sample was injected into the quartz cuvette secured on the holder, and then the scattered light was recorded at an angle of 173° with respect to the incident beam. Film Surface Observation. The nanoscale morphology of the cured film was observed using a Leo 1530 field emission scanning electron microscope (FE-SEM). The samples were vacuum-dried and then coated with a thin layer (ca. 1.0 nm) of a Pt−Pd alloy with a sputter coater equipped with a quartz crystal microbalance thickness controller. The samples were imaged at high magnifications (e.g., ×100k) under the acceleration voltage of 15 kV via an in-lens detector. Atomic force microscopic (AFM) imaging of the bottom and the AF layers were performed with a Nano Scope scanning probe microscope (CP-II, Veeco). Tapping mode was used to track the surface of the sample via a single crystal silicon probe (Olympus tapping mode etched silicon probe). The spring constant and resonance frequency of the probe were 42−80 N/ m and 50−100 kHz, respectively. The scanning frequency was 1 Hz. Both topographic diagram and phase contrast diagram

To prepare the coating sol for the AF layer, hydrophilic agents were modified first, cf. Scheme 2. Equal number of moles of 2-HEMA and IPDI were stirred in a glass reactor under temperature control. When the temperature reached 50 °C, dibutyltin dilaurate (0.1% of the total weight of 2-HEMA and IPDI) acting as the catalyst was added, and the reaction was allowed to proceed for 2 h at 50 °C. Subsequently, hydrophilic agent (Tween 20) was added and reacted for another 2 h. The molar ratio of IPDI/Tween20 was set to 1. The modified Tween 20 was called T20, hereinafter. The AF coating sols for the top layer were prepared by mixing T20 with the coating sol for the bottom layer, however, with some adjustment of the DPHA content to give 30 wt % silica in the coating, cf. Table 1. The formed sol was spin-coated on the partly cured bottom layer, followed by predrying (80 °C, 40 s) and UV-curing (500 mJ/cm2) to obtain an AF layer of ∼1 μm thick. Table 1. Chemical Species for Preparing the Top AF Layer of the Coating sample name AF2 AF4 AF6 AF8 AF10 AF12 AF15 AF96

MSiO sol 2

20 20 20 20 20 20 20 2

(g)

T20 (g)

IPA (g)

DPHA (g)

1173 (g)

0.3 0.6 0.9 1.2 1.5 1.8 2.25 24

17.68 17.68 17.68 17.68 17.68 17.68 17.68 1.77

9.7 9.4 9.1 8.8 8.5 8.2 7.75 0.5

0.59 0.59 0.59 0.59 0.59 0.59 0.59 0.59 17195

dx.doi.org/10.1021/la304176k | Langmuir 2012, 28, 17193−17201

Langmuir

Article

employed, there was still ∼40% vinyl groups left unreacted (based on the peak area analysis). The free vinyl groups near the surface region would react with the curing agent, DPHA, in the top AF layer during the second stage curing process. The cured primer exhibited a hardness of 4H according to the pencil test, and it attached firmly to the PMMA substrate with a peeltest adhesion of 100%. FE-SEM imaging of the primer indicated a uniform surface morphology (Figure S1, Supporting Information), free of organic/inorganic phase domains at the resolution scale of 10 nm, agreeing with the fact that the coating was a highly transparent thin film. The 3-D topographic AFM diagram of the primer’s surface was extremely smooth with a measured average roughness as small as 1.4 nm (Figure S2, Supporting Information). Curing with a power lower than 250 mJ/cm2 has been tested to give higher residual CC concentration, however, at the sacrifice of the mechanical strength. In summary, 250 mJ/cm2 is appropriate for preparing a smooth coating surface possessing sufficient amount of vinyl groups for making bonds with the multifunctional monomer, DPHA, in the top AF layer during the second stage curing process. Preparation of the AF Layer. The surfactant, Tween-20, was modified to acquire UV photosensitivity by linking to 2HEMA via the spacer IPDI. The synthetic process involved two steps, as shown in Scheme 2. First, the −OH group of 2-HEMA and −NCO group of IPDI were reacted to form the urethane bond in the molecule, 2-HEMA/IPDI.33 During the course of this reaction, the −NH peak in the FTIR spectra (Figure S3, Supporting Information) stemming from the urethane linkage is found at 1529 cm−1 whose intensity increases gradually up to 1 h, and afterward it levels-off. Because 2-HEMA and IPDI were initially charged at equal molar amount, the reaction was considered to approach completeness in ca. 1 h. In the second step, the as-prepared molecule 2-HEMA/IPDI was reacted with Tween-20 to give a dual-functional molecule (T20), bearing both UV-curability and high hydrophilicity. The −NH peak of urethane at 1529 cm−1, signifying the presence of T20, rises progressively over the period of 2 h. In contrast, the -NCO signal at 2260 cm−1 declines with time and it vanishes virtually after reaction for 2 h, confirming that the second half of -NCO groups in IPDI has all been converted to urethane at the end of reaction (Figure S4, Supporting Information). It should be noted that the CC groups remain intact throughout the reaction. These groups can be used to cross-link with DPHA during UV curing of the top layer. Figure 2 shows the FTIR spectra of a typical bilayer coating, AF10. The Si−O−Si band at 1529 cm−1 indicates the presence of MSiO2 whereas CO at 1727 cm−1 is contributed by 2HEMA, Tween-20, and DPHA moieties. The small CC signal suggests that an effective UV-curing has been performed; thereby, a rather hard AF layer (4H) was created (details shown later). The amount of CC bond can further be reduced by postbaking of the sample at 100 °C. Apparently, baking for 1 h. is sufficient to consume most of the residual CC; prolonged baking is futile probably due to steric hindrance. The other effect of postbaking is for the compaction of silica particles. As is evident from the spectra, new Si−O−Si bonds are formed by alcohol condensation of Si−O−C and/or water-condensation of Si−OH groups. The bottom layer and bilayer coatings were highly transparent in the visible region. For example, the transmittance of the bilayer coating AF10 is 95−98% in the visible region (Figure S5).

were constructed. The former diagram provides information of surface roughness and domain size, whereas the later voltage distribution on the surface. Contact Angle, Adhesion, and Hardness Measurements. The contact angle between water and AF surface was measured by a FTA 125 contact angle/surface tension analyzer at room temperature. A 6 μL drop of water was placed onto the surface of the coating. The image was taken and the contact angle was measured from shape analysis of the sessile drop. Tape test (CNS 11684), also known as peel test, was carried out to evaluate the adhesion of the coatings on the substrate. The degree of adhesion was defined as the percentage of film residing on the plate after the peel test. The hardness of the cured films was examined by the industrial pencil hardness test (JIS K5400) with pencils of different hardness at a load of 765 g. Steam Tests. Steam tests of various coatings were carried out to see their AF performance. Boiling water was added into a beaker to about half full. Then, the sample was placed on the beaker with the coated surface facing down. Vapor condensed on the coating surface was observed and photographed.



RESULTS AND DISCUSSION Preparation and Characterization of the Primer. The primer is an organic−inorganic hybrid material composed of nanosized silica dispersed uniformly in the polymer host.31 The sol−gel process for the synthesis of surface-modified silica, MSiO2, involves hydrolysis and condensation of TEOS and MSMA. The detailed chemical analyses (FTIR and 29Si NMR) of this reaction have been documented.23 Particle size of the modified silica was measured to be ∼7 nm, consistent with those reported in the literature.29 The formed MSiO2 sol was mixed with DPHA, photoinitiator, and 2-propanol and then UV-cured to yield an organic−inorganic hybrid hard coating on PMMA plate.28 Figure 1 shows the effect of UV power on the curing efficiency

Figure 1. FTIR spectra of the UV-cured coatings for the bottom layer, showing the effect of irradiation power.

of a typical sample. A monotonous decrease of the CC peak signal at 1634 cm−1 is indicated with increasing UV irradiation intensity. When the power was set to 500 mJ/cm2, the peak was very small, implying that most of the vinyl groups were converted to C−C bonds during the UV-induced cross-linking reaction. In case that only half of the power 250 mJ/cm2 was 17196

dx.doi.org/10.1021/la304176k | Langmuir 2012, 28, 17193−17201

Langmuir

Article

soft domains on the surface.34,35 For example, the primer is a very hard material composed of highly cross-linked DPHA and MSiO2, and thus high scanning voltage is imposed. On the other hand, a relatively low voltage is delivered to the surface of AF96 due to the presence of the soft hydrophilic agent, T20. In this context, by analyzing the voltage contrast diagram, one can differentiate the hydrophilic (soft) from the hydrophobic (hard) zones on a flat AF coating surface. Three typical cases are demonstrated in Figure 4 for comparison. The T20 contents for these coatings are 4, 12, and 15 wt %, respectively. From the topographic diagrams (right part), the measured average surface roughness is found to increase just slightly from 1.0 to 3.6 nm with a large increase of T20 dosage from 4 to 15 wt %. On the other hand, the voltage contrast diagrams (left part) use gray scale to manifest areas of different voltages; the light-gray depicts high voltage (hard) regions, whereas the darkgray, low voltage (soft) ones. As is expected, the dark area increases as the T20 content is raised. To estimate the distribution of soft/hard domains on the AF surface, the voltage contrast diagrams are further processed to give black and white bicolor patterns, as shown in Figure 5, with 5 V being taken as the border value. This voltage is selected based on two facts: (i) it is close to the arithmetic mean of the highest voltage of the primer’s surface and the lowest voltage of AF96 (Figure 4); (ii) as the lowest voltage of the primer’s surface is 6 V, below 5 V contribution from the soft segments (T20) is expected to outweigh that from the hard segments (cured DPHA and MSiO2). In fact, bicolor patterns with various cutting-edge voltages over the range 4.5−5.5 V have been created, and from them a similar conclusion can be inferred, considering the effect of soft/hard pattern on the AF performance discussed below. From Figure 5, variation of soft/hard domains with respect to T20 content is clearly illustrated. For the sample AF4, only small separate black dots of ca. 10−35 nm are present, and for

Figure 2. FTIR spectra of the bilayer AF coating AF10.

AFM was employed to disclose the surface morphology as well as the distribution of T20 within the surface region of various prepared coatings. Figure 3 shows the topographic and voltage contrast diagrams of the surfaces of the primer and sample AF96. The former consists of DPHA and MSiO2, whereas the latter contains mostly T20 (96%). The topographic diagrams (right part) for these two samples are similar; both surfaces are extremely smooth with measured average roughness of ∼1 nm. However, their voltage contrast diagrams (left part) are distinctively different. For the primer’s surface, the scanning voltage is above 7 V over the entire surface, except for some sporadic spots. In contrast, for the sample AF96, a small scanning voltage, 1.41 V, covers ∼90% (based on image analysis) of the scanning area. Generally, the voltage value depends both on the roughness and the rigidity of the surface. For a very smooth surface having roughness 100 nm, and in case that the growing water droplets contact a hydrophilic area, they will spread to reduce their surface tension. Therefore, AF effect can be achieved as long as the hydrophilic area on the coating surface can form a pattern that prohibits the growth of water droplets beyond ∼100 nm, which is manifested in Figure 5. For the sample AF4, water droplets of size as large as ∼160 nm (cf. the white area) may rest independently on the coating surface; hence, fog is expected to form on this surface. In contrast, for both AF10 and AF15, the sizes of the hydrophobic domains are all less than 100 nm (average 63 nm for AF10). As arriving water droplets tend to spread into a continuous film, scattering

does not occur, and these two coatings will exhibit AF characteristic. Antifog and Hardness Tests. Hardness, contact angle, adhesion, and AF tests of various prepared coatings were performed and the results are summarized in Table 2. The plastic substrate PMMA is relatively soft with a pencil hardness