A Transparent and Wear-Resistant Anti-Smudge Coating Based on Polydimethylsiloxane

Abstract

Conventional liquid-like anti-smudge coatings often suffer from poor wear resistance, failing to meet the requirements of daily use in practical applications. In this work, anti-fingerprint coatings were fabricated from monovinyl-terminated polydimethylsiloxane (^ViPDMS), divinyl-terminated PDMS (^ViPDMS^Vi) and polymethylhydrosiloxane (PMHS), with octavinyl polyhedral oligomeric silsesquioxane (VPOSS) incorporated to enhance wear resistance, through hydrosilylation reaction. The one-pot reaction was carried out in two steps to suppress the aggregation of POSS moieties, resulting in a clear coating with 95% transmittance in the visible region even at a high POSS content of 15.4 wt %. The effects of PDMS composition, molecular weight of ^ViPDMS, and VPOSS content on coating properties were investigated, and the optimal formulation was determined. This formulation exhibited excellent resistance to staining by fingerprints, coffee, milk, and dye, and withstood 1000 cycles of abrasion with no discernible surface damage, demonstrating good anti-smudge and wear-resistant properties. In addition, the coating was readily deposited on various substrates, indicating its general applicability. The simplicity of the fabrication method and the good properties of the coating make it a promising material for anti-fingerprint applications.

1. Introduction

The pursuit of visually comfortable and clean appearance has stimulated tremendous interest in anti-fingerprint surfaces for various fields, such as displays, packaging, automotive interiors, handheld electronic devices, and optical lenses. − To address the problem of fingerprint stains, several approaches have been adopted to prepare anti-smudge coatings. − One popular method involves creating micro-/nano-scale rough structures followed by fluorination treatment. − Coatings fabricated by this method exhibit good anti-smudge properties, with water contact angles exceeding 150°. , However, the rough structures lead to relatively low transparency and poor wear resistance of the coatings, , hindering their long-term performance in anti-fingerprint applications. Another method involves slippery liquid-infused porous surfaces (SLIPS), where the infused lubricant provides anti-smudge property. − Nevertheless, the lubricant in such coatings can be depleted over time, resulting in loss of contaminant-repelling properties. , In comparison, liquid-like coatings are more promising candidates for anti-fingerprint coatings. In these coatings, a low-surface-energy polymer is grafted onto the substrate to form polymer brushes, which function as a liquid-like layer, providing excellent liquid-sliding properties and facilitating contaminant removal. − Such coatings involve no rough structures or pores and therefore exhibit high optical transparency. Moreover, the liquid-like polymer chains are covalently attached to the substrate, avoiding lubricant leakage or depletion.

Among typical low-surface-energy materials, polydimethylsiloxane (PDMS) is often selected for anti-smudge coating fabrication due to its low cost and non-toxic nature. − For example, Liu et al. integrated PDMS with a polyurethane-based matrix to make a fluorine-free coating applicable to various substrates that exhibited good resistance to contamination. Luo et al. grafted hydroxy-terminated PDMS onto a waterborne polyurethane, which was further cross-linked to construct a smooth omniphobic coating with excellent transparency and remarkable self-cleaning ability. Zhao et al. reacted double-terminally functionalized PDMS with an acrylic polyol resin and then cross-linked it to prepare an anti-smudge coating with high economic viability. While these coatings exhibit excellent resistance to various liquid contaminants, their wear resistance, which is crucial for long-term usage, remains inadequate. ,,

Polyhedral oligomeric silsesquioxane (POSS) is a three-dimensional organic/inorganic hybrid material that has been utilized as an additive to improve the mechanical properties of coatings in recent years. − By designing the type of substituent groups attached to the silicon atoms, the chemical reactivity and compatibility of POSS can be readily regulated. − For instance, Liu et al. combined 3-glycidyloxypropyl POSS with perfluoroalkyl acids to produce hard and flexible omniphobic coatings with excellent hardness, modulus, and wear resistance. Wheeler et al. incorporated various POSS species into aromatic and aliphatic resins and found that the moduli of the resins were significantly improved. Although POSS additives can enhance the mechanical properties of composites, their aggregation is difficult to avoid, which can be an issue for coatings where optical clarity is required. ,,

In this study, we fabricated an anti-fingerprint coating from octavinyl POSS (VPOSS), divinyl-terminated PDMS (^ViPDMS^Vi), monovinyl-terminated PDMS (^ViPDMS) and polymethylhydrosiloxane (PMHS) via a one-pot hydrosilylation reaction to form a dense network structure (Figure a). The PDMS components, in particular the tethered PDMS chains from ^ViPDMS, provide the coating with liquid-repelling properties, whereas the POSS component enhances its wear resistance. By carrying out the reaction in two distinct steps, the aggregation of POSS moieties was suppressed, and a clear and wear-resistant anti-fingerprint coating was produced.

1.

Preparation of (a) the prepolymer, and (b) the coating.

2. Experimental Section

2.1. Materials

Octavinyl polyhedral oligomeric silsesquioxane (VPOSS, 95%), polymethylhydrosiloxane (PMHS, 1.55% H content), diiodomethane (98%), lactic acid (2-hydroxypropionic acid, GB, 98%), deuterated chloroform (CDCl₃, 99.8%) and 1-methoxy-2-propanol (99%) were purchased from Energy Chemical. Three monovinyl-terminated PDMS (^ViPDMS, M _n = 2000, 3500, and 6400 g/mol, respectively) and divinyl-terminated PDMS (^ViPDMS^Vi, M _n = 2000 g/mol) were synthesized in our laboratory. Vinyltrimethoxysilane (VTMS, 98%) was obtained from InnoChem. Karstedt catalyst (2000 ppm) was provided by Heraeus. Ethanol (99.7%), n-hexane (97%), acetone (99.5%) and toluene (99.5%) were supplied by Xilong Science. Potassium bromide (99%) was purchased from MERCK. Hydroxyl-terminated polydimethylsiloxane (HTPDMS, M _n = 550 g/mol) was obtained from Macklin. Glacial acetic acid (99.5%), sodium chloride (99.5%), and sodium hydrogen phosphate (99%) were supplied by Beijing Chemical Plant Co., Ltd. All solvents and reagents were used as received without further purification. Deionized water used in all experiments was purified with a PGeneral GWA-UN4 purification system (18.2 MΩ·cm resistivity).

2.2. Preparation of the Prepolymers

The prepolymers were synthesized following a literature procedure. First, VPOSS was dispersed in toluene, into which Karstedt catalyst (2.5% of the total mass of all reagents and the solvent) was added. The mixture was stirred at 300 rpm at room temperature for 30 min to ensure complete dissolution of VPOSS. Subsequently, ^ViPDMS, ^ViPDMS^Vi, PMHS, and VTMS (1% of the total mass of all reactants) were added. The total mass concentration of the mixture was 0.1 g/mL. The mixture was heated at 80 °C for 48 h under a nitrogen atmosphere with 300 rpm stirring. The ratios of the reagents and the compositions of the prepolymers prepared are summarized in Table . The synthesis of the 0 wt %VPOSS prepolymer is described in the Supporting Information. The reaction temperatures for the 20.0 wt %VPOSS and OD prepolymers were 90 °C and 70 °C, respectively.

1. Compositions of the Prepolymers .

prepolymer	VPOSS content (wt %)	molar ratio of M/D	M n of ViPDMS
M2D1	15.4	2:1	2000 g/mol
M1D2	15.4	1:2	2000 g/mol
OD	15.4	0:1	-
3.5kPDMS	15.4	1:2	3500 g/mol
6.4kPDMS	15.4	1:2	6400 g/mol
0 wt %VPOSS	0	1:2	2000 g/mol
7.7 wt %VPOSS	7.7	1:2	2000 g/mol
20.0 wt %VPOSS	20.0	1:2	2000 g/mol

^aNote: M denotes ^ViPDMS, D denotes ^ViPDMS^Vi; for all prepolymers, M _n of ^ViPDMS^Vi was 2000 g/mol, and the molar ratio of Si–H to total C=C was 1.2:1.

2.3. Preparation of the Coatings

Glass substrates (2.5 cm × 2.5 cm) were cleaned sequentially with n-hexane, acetone, ethanol, and deionized water, and then treated with oxygen plasma (CPC–F plus, CIF International Group Co., Ltd.) under vacuum (30 Pa) at a power of 120 W for 3 min prior to use. Toluene in the prepolymers was removed using a rotary evaporator. Then, 200 μL of the prepolymer was spin-coated onto the glass substrate at 1000 rpm for 10 s and then at 2000 rpm for 30 s, and the coating was cured in an oven at 80 °C for 8 h (Figure b). The thickness of the coating thus prepared was approximately 20–30 μm.

2.4. Characterizations

Fourier transform infrared (FTIR) spectra were collected on a Nicolet 6700 spectrometer equipped with a narrow-band MCT detector in transmission mode. Nuclear magnetic resonance (NMR) spectra were acquired on a Bruker Ascend-500 spectrometer using CDCl₃ as the solvent. The viscosity of the prepolymer was measured using an Anton Paar MCR702 rotational rheometer. A ramé-hart 200-F1 goniometer was employed to measure the water contact angles (WCA) and oil contact angles (OCA) of the coatings using 10 μL of deionized water or diiodomethane as the probing fluids. Each sample was measured at least three times, and the average value was reported. The surface roughness of the coatings was characterized using an Elite I optical profilometer. The transmittance in the visible range of 400–800 nm was analyzed on a double-beam UV–Vis spectrophotometer (TU-1901, Beijing PGeneral Instrument Co., Ltd.). Scanning electron microscopy (SEM) was performed on a Sigma 300 microscope (Zeiss).

2.5. Anti-Smudge Performance

An artificial fingerprint liquid was prepared following a reported method. , A rubber stamp or a brush was dipped into the artificial fingerprint liquid and then applied onto the coating surface, and the anti-smudge performance was evaluated by the residue and shrinkage of the liquid. In the second test, a thumb was pressed onto the coating surface with a constant force, and the fingerprint residue was observed under an optical microscope (XD-202, Jiangnan brand).

2.6. Wear Resistance Test

The wear resistance test was conducted using a wear resistance tester (A20-339, Dongguan Maike Instrument Equipment Co., Ltd.) with cotton cloth as the abrasion material under a pressure of 10 kPa. The test was performed at an abrasion speed of 30 mm/s for at least 100 cycles.

3. Results and Discussion

The kinetics of the hydrosilylation reaction between the Si–H bonds and the C=C bonds in the presence of Karstedt catalyst among the components was first investigated. Figure a presents the FTIR spectra of the reaction mixture for 7.7 wt % VPOSS collected at different reaction times, normalized with respect to the peak of the Si–CH₃ symmetric deformation at 1260 cm^–1. It can be observed that the intensity of the Si–H stretching band at 2167 cm^–1 decreases with time, and then barely changes after 36 h, indicating that the hydrosilylation reaction proceeded and reached a high degree of conversion before slowing down and almost stopping completely (Figure b). In the ¹H NMR spectra (Figure c), the resonance peaks assigned to the vinyl protons in the 5.6–6.1 ppm range and the Si–H peak at 4.65 ppm both decrease with time, while a new peak emerges at 0.4–0.5 ppm (Figure S1), which is attributed to the methylene groups formed by the reaction between Si–H and C=C bonds. The peak areas of Si–H and vinyl protons are plotted as functions of reaction time; they are consistent and both exhibit a similar trend to that revealed by FTIR (Figure d). The viscosity of the prepolymer solution in toluene thus prepared was 377 mPa·s. After removal of the solvent, the neat prepolymer exhibited a moderate viscosity of 870 mPa·s and showed shear-thinning behavior during spin coating, indicating good processability. Then, the prepolymer cast on the glass substrate was cured. The remaining Si–H and C=C groups in the prepolymer were found to disappear (Figure S2), indicating completion of the hydrosilylation reaction and the formation of the PDMS network.

2.

(a) FTIR spectra and (b) the Si–H peaks for the reaction mixture (for 7.7 wt % VPOSS) at different reaction times. (c) ¹H NMR spectra of the reaction mixture (for 7.7 wt % VPOSS). (d) The NMR peak areas of Si–H and vinyl protons as functions of the reaction time.

Previous studies have shown that POSS is not compatible with polymers and often agglomerates and crystallizes in the polymer matrix, resulting in phase separation and defects. ,, To alleviate this issue, we fabricated the coatings in two steps, as described above. In the first step, the prepolymers were prepared in the presence of toluene as the solvent, in which VPOSS was readily dissolved and well dispersed, reacting with PMHS via hydrosilylation without aggregation. Although the solvent was removed in the second step, the POSS moieties, now covalently attached to PMHS, became much more compatible with the PDMS matrix and unable to crystallize. In this way, optically clear coatings (see data below) with up to 15.4 wt % POSS incorporation could be obtained, more than double the amount achieved previously. By varying the ratios of the components, eight coatings of different compositions were prepared (Table ). The static water contact angles (WCAs) and oil contact angles (OCAs) for these coatings are listed in Table S1. The WCAs and OCAs were around 103° and 70°, respectively, typical values for PDMS surfaces, and much higher than that of bare glass, indicative of their liquid repellency.

First, we studied the effects of PDMS composition on the properties of the coatings. Compared with WCAs, receding contact angles (RCAs) are more relevant to the anti-smudge performance of the coatings, , as RCA characterizes the tendency of the liquid to contract on the surface. Figure a compares the RCAs for OD, M1D2, and M2D1 coatings, in which the molar ratios of ^ViPDMS to ^ViPDMS^Vi are 0:1, 1:2, and 2:1, respectively, with that of bare glass. The RCA on bare glass was close to 0° because the surface is rich in Si–OH groups and hydrophilic, whereas for the coating surfaces, the RCAs are all fairly high due to the hydrophobic nature of PDMS. More specifically, a clear trend is observed: the higher the ^ViPDMS content, the greater the RCA of the coating surface. When these surfaces were challenged with artificial fingerprint liquid using either a rubber stamp or a brush, as shown in Figure b, the residual liquid spread on the bare glass surface and formed a continuous adhesive layer, even after being wiped with a brush. The liquid behaved similarly on the OD surface, spreading on the surface and leaving a trace of liquid along the wiping direction, showing poor anti-smudge performance. For the coatings containing ^ViPDMS, especially the one with a higher content (M2D1), the artificial fingerprint liquid contracted significantly on the surfaces to occupy smaller areas, and the residual liquid along the wiping direction also contracted, exhibiting favorable anti-smudge properties. This trend is consistent with that revealed by RCA. This is because the PDMS chains derived from ^ViPDMS^Vi form the backbone of the crosslinked network, and their mobility is restricted. In contrast, the ^ViPDMS molecules are tethered to the network by only one end, forming pendant PDMS chains that are liquid-like and highly mobile, which can inhibit the adhesion of foreign species to the surface, achieving dynamic liquid-repellent properties. , Figure c shows the real fingerprints observed on these surfaces. On the bare glass surface, a large amount of fingerprint residue is left, and after wiping, the fingerprint trace is still observable. In comparison, on the OD surface, the fingerprint residue is significantly less, and on M2D1 and M1D2, the fingerprints are hardly discernible. No residual fingerprint is observed on these surfaces after wiping with tissue paper under a pressure of ∼2.5 kPa. While both M2D1 and M1D2 exhibited satisfactory anti-smudge properties, the latter has a higher content of ^ViPDMS^Vi, which forms a denser crosslinked network, beneficial to the stability of the coating. Therefore, the molar ratio of ^ViPDMS to ^ViPDMS^Vi was maintained at 1:2 for subsequent experiments.

3.

(a) RCAs measured on various surfaces. (b, c) Optical images of (b) artificial fingerprint liquid and (c) real fingerprint residue on coatings with different PDMS compositions (scale bars are 200 μm).

Next, we varied the molecular weight of ^ViPDMS in the coating. Figure compares the anti-smudge properties of the coatings prepared from ^ViPDMS of different molecular weights (2000, 3500, and 6400 g/mol). On these three surfaces, the artificial fingerprint liquid behaved similarly, contracting into droplets when the liquid was either pressed on with a rubber stamp or painted with a brush, showing comparable anti-smudge performance (Figure a). However, when challenged with real fingerprints, different results were found for these surfaces. As seen in Figure b, for the M1D2 coating (i.e. ^2kPDMS), the residue of the fingerprint pressed onto the surface is barely discernible, whereas on the surface of the ^6.4kPDMS coating, the residue is clearly observed, much more than that on the former. The fingerprint trace observed for the ^3.5kPDMS coating is intermediate between the two. In all three cases, the residual fingerprint can be completely removed by wiping with a paper towel. It is well known that fingerprint residue is a mixture of aqueous and oily substances. ,, Dynamic contact angle data indicate that the surface with shorter pendant PDMS chains (M1D2) is less hydrophobic than that with longer ones (^6.4kPMDS), as seen in Table S2, and at the same time is more oleophobic than the latter (Table S3). However, the surface tension of a mixture liquid is dominated by the component with lower surface energy, in this case the oily species. Therefore, M1D2 exhibits better overall resistance to fingerprint staining. Based on these results, M1D2, in which the molar ratio of ^ViPDMS to ^ViPDMS^Vi is 1:2 and the molecular weight of ^ViPDMS is 2000 g/mol, is the optimum formulation in terms of the anti-smudge performance of the coating.

4.

Optical images of (a) artificial fingerprint liquid and (b) real fingerprint residue on coatings prepared from ^ViPDMS of different molecular weights (scale bars are 200 μm).

Finally, the VPOSS content in M1D2 (which contains 15.4 wt %) was adjusted, and the wear resistance of the coatings was investigated using a linear wear method. Figure a presents optical images of the coatings containing different amounts of VPOSS after 100 abrasion cycles. Severe damage was observed on the coating containing no VPOSS. At a VPOSS content of 7.7 wt %, the damage to the coating by abrasion was greatly alleviated, with much fewer and lighter scratches observed on the surface. Then, with the VPOSS content increasing to 15.4 wt %, the coating became more resistant to abrasion, showing no discernible surface damage after the abrasion. The results clearly demonstrate that VPOSS is dispersed in the PDMS matrix as a nanofiller to form an organic–inorganic crosslinked network, increasing the crosslinking density and mechanical strength, and the more POSS added, the more pronounced the effect. However, when the VPOSS content was increased further to 20.0 wt %, the wear resistance of the coating deteriorated, with an increased number of scratches observed on the coating surface. Significant aggregation of VPOSS particles was observed in this coating with an excessively high VPOSS content, resulting in a significantly higher surface roughness of 167.5 nm (Figure S3). In contrast, the M1D2 coating showed no sign of POSS aggregation and hence a much lower surface roughness of 29.2 nm. This dramatic rise in roughness leads to defects and stress concentration points, adversely affecting the wear resistance of the coating.

5.

Optical images of (a) the coatings with different VPOSS contents after 100 cycles of abrasion, and (b) the M1D2 (i.e. 15.4 wt %VPOSS) coating before and after 1000 cycles of abrasion. (Scale bars are 200 μm).

The optimum coating, M1D2, with a VPOSS content of 15.4 wt %, was further challenged with 1000 cycles of the abrasion test. No discernible damage was observed on the coating surface (Figure b), and the RCA measured on the surface varied within 10° (Figure S4). In addition, the coating maintained resistance against staining by artificial fingerprint liquid and real fingerprints, showing no observable sign of deterioration (Figure S5). The optical clarity of the coating was assessed by UV–vis absorption spectroscopy. POSS units tend to aggregate and crystallize in polymer matrices, forming particles larger than the wavelength of visible light, resulting in light scattering and reduced optical clarity of the composites. In this work, VPOSS was introduced into the composites in the presence of toluene, so that POSS agglomeration was largely suppressed. As a result, few POSS aggregates of micron size were observed (Figure S3a). As illustrated in Figure a, the coating exhibits a high transmittance of 95% as prepared. After 1000 abrasion cycles, the coating remained clear with a high transmittance of >92%, with the underlying text clearly legible (Figure a inset). These findings demonstrate the excellent clarity and long-term wear resistance of the coating.

6.

(a) UV–vis spectra of bare glass and the M1D2 coating before and after 1000 abrasion cycles (inset is an optical image of the coating after abrasion). (b) Optical images of the M1D2 coating deposited on various substrates.

In addition, the M1D2 coating could be readily fabricated on various substrates, such as PET, PVC, and PC (Figure b). The coatings on these substrates were uniform and transparent. Moreover, the coating was resistant to staining by various liquids encountered in daily life, such as dyes, milk, and coffee (Figure S6). Finally, the M1D2 coating was wiped 100 times with ethanol, acetone, and n-hexane, respectively, and the static and dynamic water contact angles were found to remain unchanged (Table S4), showing its stability against typical polar and nonpolar organic solvents. This is attributed to the chemical inertness of PDMS. ,

4. Conclusion

We have demonstrated a transparent and wear-resistant anti-smudge coating fabricated from PDMS, incorporating a high content of POSS building blocks. By carrying out the one-pot hydrosilylation reaction in two steps, the aggregation of POSS moieties was suppressed, resulting in coatings with high optical clarity. The effects of PDMS composition, molecular weight of the tethered PDMS chains, and POSS content on coating performance were investigated, and the optimal formulation was determined. In this formulation, the molar ratio of mono-functional PDMS to di-functional PDMS is 1:2, the molecular weights of both are 2000 g/mol, and the VPOSS content is 15.4 wt %. The coating thus prepared readily repels staining by fingerprints and other common liquids, exhibits a high transmittance of >95% in the visible region, and withstands 1000 abrasion cycles, showing excellent long-term wear resistance. The coating can be easily deposited on various substrates, indicating general applicability. Based on these good properties, the coating is expected to be an excellent candidate for anti-fingerprint applications.

The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/polymscitech.6c00053.

• The preparation procedure for the 0 wt %VPOSS coating, additional static and dynamic contact angle data, ¹H NMR spectra, SEM images and surface profiles of 15.4 wt %VPOSS and 20.0 wt %VPOSS coatings, and stain resistance data (PDF)

The authors declare no competing financial interest.