Research Article www.acsami.org
UV-Curable Antismudge Coatings Chao Zheng, Guojun Liu,* and Heng Hu Department of Chemistry, Queen’s University, 90 Bader Lane, Kingston, Ontario Canada K7L 3N6
Downloaded via UNIV OF LOUISIANA AT LAFAYETTE on January 20, 2019 at 11:09:12 (UTC). See https://pubs.acs.org/sharingguidelines for options on how to legitimately share published articles.
S Supporting Information *
ABSTRACT: Current antismudge coatings that bear nano-pools of a grafted liquid ingredient for dewetting enablement (NP-GLIDE) are cured at high temperatures, which are undesirable for application on heat-sensitive substrates. Reported herein is the development of a NP-GLIDE coating that can be photocured at room temperature. Of the various formulations that have been tested, robust coatings were obtained from one recipe consisting of a photoinitiator, a trifunctional monomer that bears three double bonds, and a graft copolymer. The last bears pendent double bonds and a poly(dimethylsiloxane) (PDMS) side chain as the antismudge agent. Coatings were prepared by casting films from a solution containing these three components and then photolyzing the resultant films. A systematic study revealed that the liquid sliding property developed on the coating at a lower cross-linking density than that required for ink to contract. Further, retaining the ability to contract ink traces after many writing and erasing cycles was the most demanding of the three antismudge tests. For our optimized formulation, only 5 min of irradiation was required to yield a transparent coating with superior antismudge properties. Moreover, irradiating selected regions and then removing, with a solvent, reagents in the nonirradiated regions can yield a surface with patterned wettability. These advantageous properties of the new photocurable coating facilitate its applications. KEYWORDS: coatings, UV curable, antismudge, patterned wettability, graft copolymer, and self-assembly
■
INTRODUCTION Grafting a low-surface-tension liquid polymer, such as poly(dimehtylsiloxane)1−3 or a perfluorinated polyether,4,5 onto one component and curing the resultant compound with the other components of a PU or an expoxy formulation yielded an NP-GLIDE PU or epoxy coating. Here, the term NP-GLIDE derives from the observation that many test liquids have no problem to glide cleanly down such a coating at substrate tilt angles of less than 5° and that the coating contains nano-pools of a grafted liquid ingredient for dewetting enablement. These coatings are antismudge because marker inks and paints also readily contract on them and the contracted ink patches are easily removed. Additionally, the coatings are transparent because the nanopools of the antismudge agent on the surfaces and in the matrices of the coatings are much smaller than the wavelength of light and do not scatter light effectively. These advantageous features of the NP-GLIDE coatings facilitate their applications. However, the PU and epoxy formulations that we have worked with would cure thermally at high temperatures. While thermal curing is convenient for many applications, it would be problematic if one cures in an oven an NP-GLIDE coating that has been applied onto a heat-sensitive substrate, such as the touchscreen on a cellphone. Thus, the development of a UVcurable coating that can be rapidly cured at room temperature would broaden the applications of NP-GLIDE coatings. This paper reports a novel UV-curable NP-GLIDE coating. While various reactions such as a ring-opening polymerization initiated by photogenerated cations may have been utilized to cure a NP-GLIDE coating,6,7 we subjected vinyl © 2017 American Chemical Society
units to photoinitiated free radical polymerization for this purpose. In identifying a working system, we immediately thought of derivatizing P0-g-PDMS, where P0 denotes a polyol and P0-g-PDMS denotes P0 bearing a PDMS side chain, as a suitable candidate for this project (Scheme 1a) because of our previous use of P0-g-PDMS in the preparation of the NPGLIDE PU coating. We reasoned that double bonds could be introduced to the P0 backbone by reacting the hydroxyl groups of P0 with 2-isocyanatoethyl methacrylate (Scheme 1b) to yield P0II-g-PDMS, where P0II denotes P0 after double bond introduction. We further envisioned that a film could be cast from a mixture consisting of a photoinitiator, P0II, and P0II-gPDMS and that subsequent photolysis would yield a UV-cured NP-GLIDE coating. While a pristine coating from the above design could indeed contract ink, the ink contraction ability diminished on regions from which ink had been previously wiped off. Robust UV-cured NP-GLIDE coatings that could withstand at least 30 writing and ink removal cycles were obtained only after P0II was replaced with a liquid trifunctional monomer (TM) that was prepared from reacting 3 mol of 2hydroxyethyl methacrylate (HEMA) with 1 mol of a hexamethylene diisocyanate trimer (HDIT, Scheme 1c). We note that fluorinated compounds have been previously incorporated into UV-curable formulations to enhance the chemical resistance of the resultant coatings.8−12 However, most prior studies failed to go beyond reporting the static CAs Received: April 24, 2017 Accepted: July 10, 2017 Published: July 10, 2017 25623
DOI: 10.1021/acsami.7b05732 ACS Appl. Mater. Interfaces 2017, 9, 25623−25630
Research Article
ACS Applied Materials & Interfaces
Scheme 1. Chemical Structures of (a) P0-g-PDMS, (b) 2-isocyanatoethyl methacrylate, (c) HDIT, and (d) PDMS−OH
components of P0 are the hydroxyl-containing monomers, which have a molar fraction of ∼40%. Since the repeat unit number of the polymer is ∼20, an average of 8 hydroxyl groups exist in each P0 chain. To prepare P0II-g-PDMS, P0-g-PDMS was first synthesized.1 The protocol involved reacting PDMS−OH, PDMS bearing a terminal hydroxyl group (Scheme 1d), with excess oxalyl chloride, to yield PDMS-COCl (PDMS bearing a terminal acid chloride functionality). The excess oxalyl chloride was then removed via vacuum distillation at 60 °C before PDMS-COCl was added into a P0 solution in anhydrous chloroform to eventually yield P0-g-PDMS. While Figure S1a compares the 1 H NMR spectra of P0 and P0-g-PDMS, the SEC traces of the two polymers are compared in Figure S1b. Our quantitative analysis of the NMR spectra concluded that the P0-g-PDMS sample consisted of an average of 1.0 PDMS chain grafted onto each P0 chain and the PDMS weight fraction in the graft copolymer was 33 wt %. The second step involved reacting the residual hydroxyl groups of P0 with 2-isocyanatoethyl methacrylate at a molar ratio of 1.00/0.97. While the 1H NMR spectra of P0-g-PDMS and the produced P0II-g-PDMS are compared in Figure S1a, the FTIR spectra of the two are compared in Figure S1c. The data show that all of the hydroxyl groups of P0-g-PDMS were consumed during this process. The preparation of P0II-gPDMS was also accompanied by an anticipated shift to the higher molecular weight side for the SEC peak of P0II-g-PDMS relative to that of P0-g-PDMS (Figure S1b). TM Synthesis. The trifunctional monomer was synthesized by reacting HEMA with HDIT at a molar ratio of 3.21 using dibutyltin dilaurate as the catalyst.32 The consumption of isocyanate under these conditions has been confirmed by a FTIR study of samples collected from the reaction mixture at different times. The FTIR data in Figure S2 suggest that the isocyanate groups were consumed 2.0 h after the reactants were mixed. The TM used for our coating was prepared by reacting HDIT with HEMA for at least 2.0 h. Synthesis of Mixtures of TM and DM-PDMS. DMPDMS denotes a molecule that bears two double bonds and a PDMS arm. To prepare mixtures that consisted of DM-PDMS and TM, PDMS−OH and HDIT at hydroxyl-to-isocyanate molar ratios ranging between 1/99 and 1/3 were initially heated at 45 °C for 2.0 h with the catalyst dibutyltin dilaurate.
of several liquids in the characterization of the wetting properties of the final products. The contraction of ink and paint traces, which is more challenging to achieve than liquid sliding but deemed an essential qualification for an antismudge coating, were seldom reported. When PDMS was used,13−17 this polymer was bound to the coating matrix by both ends, which did not facilitate the formation of an enriched PDMS brush layer.18 Thus, no antismudge properties were reported for those systems despite the report of improved water repellency.14 Our NP-GLIDE coatings have no problems with gliding different test liquids because the PDMS chains were grafted via only one end to a coating surface and helped convert a cross-linked solid coating surface into a slippery liquid-like one. Our NP-DLIDE coatings were inspired by prior reports of grafted liquid-like monolayers19−22 and of slippery liquidinfused porous surfaces (SLIPS).23−30 In the former case, a monolayer of poly(dimethylsiloxane) (PDMS)19−21 or highly mobile organic molecules22 are grafted onto flat solid substrates to provide the dewetting capabilities. We think that NP-GLIDE coatings should be more wear-tolerant than monolayer coatings because polymers do wear and the NP-GLIDE coatings are much thicker than monoayer coatings. Our previous study has demonstrated that the NP-GLIDE coatings maintained their antismudge properties even after their surfaces were much abraded.1,2,4 We attributed this to the ability for the NP-GLIDE surfaces to self-regenerate: when the surface layer was worn away, nanopools that were initially embedded underneath were ruptured and chains of the freshly released liquid polymer replenished the worn surface. In the case of a SLIPS, the pores of a premade porous surface are filled with a low-surfacetension liquid.23−31 An NP-GLIDE coating differs from a SLIPS because the liquid antismudge agent is covalently attached to the coating matrix and a grafted antismudge agent should be less prone to loss by evaporation or leaching.3
■
RESULTS AND DISCUSSION P0II-g-PDMS Synthesis and Characterization. For relevance, we used a commercial polyol, P0, for this project. This polymer has been meticulously characterized previously via various 2-dimensional 1H NMR techniques.4 The polymer is unduly complex because various monomers had been introduced by the manufacturer to provide different properties of the coating solution and the final coating. However, the key 25624
DOI: 10.1021/acsami.7b05732 ACS Appl. Mater. Interfaces 2017, 9, 25623−25630
Research Article
ACS Applied Materials & Interfaces Sufficient HEMA was then added to react with the residual isocyanate groups. Identification of a Robust NP-GLIDE Formulation. Our prior studies suggest that a NP-GLIDE coating could be prepared by grafting PDMS to a multifunctional reactive component, such as a polyol, and then curing the resultant compound with other components of a coating formulation.1,2,4 Bearing this in mind, we experimented with three formulations as illustrated in Scheme 2. While formulation 1 or F1 consisted
Table 1. Comparison of the Water and Hexadecane SAs and CAs on Different Coatings hexadecane
water
sample
T% at 500 nm
CA (deg)
SA (deg)
CA (deg)
SA (deg)
F0 F2 F3
99 ± 1 98 ± 1 98 ± 1
spreads 28 ± 1 28 ± 1
spreads 2.4 ± 1 2.8 ± 1
73 ± 1 101 ± 1 102 ± 1
>85 66 ± 3 66 ± 4
The high transparency of the F2 coatings must be due to the ability of the large P0II backbone to prevent the grafted PDMS side chains from undergoing macrophase separation and due to the compatibility between P0II and the P0II backbone of P0IIg-PDMS. P0II-g-PDMS did not undergo macrophase separation from the cross-linked TM in the F3 coatings probably partially because of the moderate compatibility between P0II and the cross-linked TM and partially because of the rapid cross-linking between TM and the P0II backbone under photolysis. The clear F2 and F3 coatings were further characterized by measuring the static CAs of 5 μL water and hexadecane droplets, as well as the sliding angles (SAs) of 5 μL-hexadecane and 20 μL-water droplets on these surfaces. The CAs of hexadecane and water were higher on the F2 and F3 coatings than on the F0 coating (Table 1). This increase should be due to the enrichment of the surfaces of the latter two coatings by the low-surface-tension PDMS chains. As the surface tension γC of the coating surface decreased, test liquids spread less on the coating and their CA θ increased according to Young’s equation γ − γLC cos θ = C γ (1)
Scheme 2. Illustrations of the Formulations Tested
of a DM-PDMS and TM mixture, formulation 2 (F2) consisted of P0II and P0II-g-PDMS. Meanwhile, F3 consisted of TM and P0II-g-PDMS. Lastly, a reference coating F0 was prepared from TM alone together with the photoinitiator. To prepare a coating, a particular formulation was diluted with dimethyl carbonate (DMC) and N,N-dimethylformamide (DMF) before the photoinitiator 2-hydroxy-2-methylpropiophenone was added. DMC and DMF which were selectively poor for PDMS were used to facilitate the formation of micelles from DM-PDMS or P0II-g-PDMS.1,4 Micelle formation helped hide PDMS in the cores and has been shown to improve the compatibility of the system during coating formation and to facilitate the uniform distribution of the PDMS nanopools throughout the coating matrix.1 The solvents were subsequently left to evaporate in an oven under a gentle nitrogen flow overnight before the coatings were cured via UV irradiation for 5.0 min. The complete consumption of the double bonds for formulations F0, F2, and F3 after 300 s of irradiation is confirmed by comparing FTIR spectra of the samples before and after their photolysis with the data shown in Figures S3 and S4. For comparison purposes, all coatings were 32 μm thick and contained 5.0 wt % of the photoinitiator and 2.5 wt % of PDMS. The F1 formulation after casting on a glass plate and solvent evaporation was opaque and the coating remained so after photolysis, suggesting that macrophase separation occurred in this system. Macrophase separation occurred probably because the HEMA-bearing arms of the dimer were not sufficiently thick or large to isolate PDMS into individual nanopools but allowed the different PDMS nanodomains to overlap. Because of sample opacity, the F1 coatings were not further characterized. The F2 and F3 coatings appeared clear both before and after sample photocuring. The clarity of the cured coatings is seen from the high transmittance values for 500 nm light (Table 1).
where γ is the tension of the test liquid and γLC is the interfacial tension between the test liquid and the coating. While neither hexadecane nor water glided down the F0 coating as intact droplets, these liquids slid down the F2 and F3 coatings cleanly. In the case of 5 μL hexadecane droplets, the measured sliding angles (SAs) were lower than 3°. Thus, hexadecane sliding was facile on these coatings. Water droplets with a volume of 20 μL also slid on the substrate at a tilt angle of 66 ± 4°. Therefore, PDMS incorporation enhanced the liquid sliding properties of the base coatings. The F2 and F3 coatings were also subjected to an ink contraction test. While a red ink containing ethyl alcohol and ethylene glycol monobutyl ether as the solvent left a continuous mark on the F0 coating (Figure 1), the ink shrank on both the F2 and F3 coatings. Additionally, the ink traces could be readily wiped off with a Kimwipes tissue from the F2 and F3 coatings. However, the traces could not be removed under identical conditions from the F0 coating. Thus, both F2 and F3 exhibit antismudge behavior. We repeated the ink contraction experiment on a previously marked region on the F2 and F3 coatings. Surprisingly, the ink contracting ability on the F2 coating diminished. On the other hand, the ink contracting ability remained on the F3 coating after 30 writing and erasing cycles. Thus, only the F3 formulation afforded a robust NP-GLIDE coating. The surprising difference between the F2 and F3 coatings suggests that the formulation of a robust NP-GLIDE coating is not as straightforward as we had originally envisioned. The ink contracting performance noticeably diminished on the F2 coating after writing erasure, probably because the exposed PDMS was mostly wiped off the surface together with the ink during the ink removal step. This behavior suggested that the 25625
DOI: 10.1021/acsami.7b05732 ACS Appl. Mater. Interfaces 2017, 9, 25623−25630
Research Article
ACS Applied Materials & Interfaces
F3 coating was rapidly cured, which is typical for most UV curable formulations.34 As irradiation progressed, the hardness of the coating increased. Figure 3 shows how the hardness measured using a series of pencils invoking the ASTM D3363 protocol varied with irradiation time for the F3 coating samples. Below the photolysis time of 10 s, the coating was softer than the softest pencil with the hardness designation of 9B. A measurable hardness of 8B was seen only by the photolysis time of 30 s. The hardness eventually reached 3H by the photolysis time of 300 s. As the coating hardness increased, the rate of water CA decrease due to surface reconstruction diminished. Figure 4 plots how the water CA varied with their sitting time on coating samples that had been irradiated for various times. The water CAs on the sample irradiated for 5 s rapidly decreased to approach 73 ± 1°, the static water CA on a cross-linked F0 coating. This result suggests the ability for the more hydrophobic PDMS that was initially in contact with water to recede into the coating matrix, thus exposing the lightly crosslinked TM component. Surface reconstruction was also observed on the coating that was irradiated for 10 s. Meanwhile, no significant surface reconstruction was observed on the coatings that were irradiated for 30 and 300 s. Figure 3a plots the variation in the static or equilibrium water CAs, measured 5 min after droplet dispensing, as a function of time that was used to irradiate F3 coating samples. The CAs increased steeply with time between 5 and 30 s and plateaued after 60 s. These trends were consistent with those observed for the double bond conversion data obtained via FTIR characterization, suggesting that double bond cross-linking was essential for impeding surface reconstruction. Figure 3a also plots the variation in the SAs of 30-μL water droplets as a function of F3 irradiation time. The SA variation trend contradicted the CA variation trend, that is, the SA decreased as photolysis time increased and leveled off above the photolysis time of 120 s. The water CA variation trend also contrasted with that exhibited by the hexadecane CAs (Figure 3b). At shorter irradiation times, the hexadecane CAs were higher. They decreased with irradiation time initially and plateaued after 60 s. The higher CAs at shorter irradiation times might be due to the migration of the PDMS component to the surface under a hexadecane droplet, which should be more compatible with PDMS but helped increase hexadecane CAs. At irradiation times shorter than 30 s, hexadecane spread on the coatings. It started to glide down cleanly only on samples that were irradiated for 30 s and longer. The SA plateaued only after the irradiation time exceeded 120 s (Figure 3b). We also performed an ink contraction test on samples irradiated for different periods of time and the results are shown in Figure 5. No ink contraction was observed on samples that were irradiated for less than 30 s. The ink contraction capability further improved only beyond the irradiation time of 30 s and excelled at the irradiation time of 300 s. The data discussed above suggest that the CAs and SAs plateaued at shorter photolysis times than that required for the emergence of the ink shrinking behavior. Thus, ink shrinking is a more challenging test than the liquid CA and SA tests. While liquid CA and SA values have often been reported in the past,12−16 high CA values and clean liquid sliding do not automatically indicate that a coating possesses antismudge properties. This study suggests that a high degree of coating
Figure 1. Comparison of marker tests performed on the F0, F2, and F3 coatings: (a) Ink trace left on a pristine coating. (b) Status of the ink trace after cleaning with Kimpwipes tissue. (c) Ink trace left on the F2 and F3 coatings after writing on them the second time.
exposed PDMS chains were not tightly bound to the coating matrix in this case. Two factors could have contributed to the performance difference between the F2 and F3 coatings. First, the maximal possible cross-linking density should be higher in the F3 than in the F2 coatings. The F2 and F3 coatings consisted mostly of P0II and TM, respectively. A simple calculation revealed that the double bond contents in P0II and TM were 1.6 and 3.6 mmol/g, individually. Second, we suspect that the mobility and the spacing of the double bonds in the TM are larger than those in P0II. Thus, the extent of intrachain or intramolecular polymerization was probably higher in the F2 than in the F3 coating. Since the intramolecular polymerization did not contribute much to the binding between different molecules such as between P0II and P0II-g-PDMS, the P0II-g-PDMS chains were more tightly bound with the coating matrix in the F3 than in the F2 coating. F3 Curing. The wetting behavior of the F3 coating was studied as a function of its radiation time to establish the conditions for robust NP-GLIDE coatings. Figure 2 compares
Figure 2. Comparison of the FTIR spectra of a F3 coating that was irradiated for different periods of time.
the FTIR spectra of a F3 sample that was irradiated for different periods of time. While the marked peak at 1638 cm−1 in Figure 2 corresponds to the CC stretching of the HEMA double bonds,33 the peak at 1164 cm−1 arises from the stretching of the CO−O bond in the −OOC−C(CH3)CH2 group.33 The intensities of these two peaks decreased rapidly during the initial 30 s of irradiation. After 60 s, further decreases in the intensities of these peaks became barely noticeable. Thus, the 25626
DOI: 10.1021/acsami.7b05732 ACS Appl. Mater. Interfaces 2017, 9, 25623−25630
Research Article
ACS Applied Materials & Interfaces
Figure 3. Variation in the static CAs and SAs of (a) water and (b) hexadecane as a function of the F3 coating irradiation time. Also shown is the variation in the hardness of the coating as a function of irradiation time.
Figure 6. (a) Photographs showing the clean sliding of hexadecane on a F3-coated glass slide in contrast with its spreading on an uncoated slide. (b) Photographs showing the contraction of a red paint on a F3coated glass slide but the consistent covering of an uncoated slide by this paint. Figure 4. Water CA variations as a function of time on F3 coatings that were irradiated for 5, 10, 30, and 300 s, respectively.
in the past and thus they should repel all liquids with surface tensions above ∼23 mN/m.2,3 We note that the SA of 66 ± 4° for 20-μL droplets on F3 containing 2.5 wt % of PDMS was substantially higher than those on the prior PU or epoxy-based NP-GLIDE coatings.1,2 To gain insight into this, we analyzed this coating by XPS and determined a Si atomic abundance of 6.2% on the coating surface (SI). This value was significantly lower than 11.7% reported for a NP-GLIDE PU coating that had a bulk PDMS content of 9.0 wt %1 or than 14.9% reported for a NP-GLIDE epoxy coating that had a bulk PDMS content of 7.4 wt %.2 Indeed, the F3 coating had a lower PDMS bulk content of 2.5 wt %. To examine the effect of increasing the bulk PDMS content, we prepared another F3 coating that contained 5.0 wt % of PDMS. The SA determined for 20 μL water droplets was still 67 ± 2°, which was the same as 66 ± 4° on the F3 coating containing 2.5 wt % of PDMS. Thus, we suspect that the surface PDMS content was already saturated by the bulk PDMS content of 2.5 wt % and did not further increase with bulk PDMS content beyond this value. We further speculate that the surface PDMS content difference between the current coating and the prior NP-GLIDE coatings probably arose from the different curing protocols used and the different coating compositions. The coating composition difference should affect the extent of PDMS surface stratification because we believe that a higher incompatibility between PDMS and the coating matrix should promote PDMS surface enrichment. The polyol P0 and isocyanate precursors of a NP-GLIDE formulation should be more polar than P0II and TM. Thus, the higher incompatibility between PDMS and the PU components probably rendered a
Figure 5. Ink traces left on F3 coatings that were irradiated for 2, 5, 10, 30, 60, 120, and 300 s.
cross-linking is essential for achieving a robust and full-fledged antismudge coating. NP-GLIDE Coating. F3 coatings that had been irradiated for 300 s were robust. Figure 6a compares photographs taken of dyed hexadecane droplets that were immediately and 4 s after their dispense on an uncoated and a F3-coated glass plate. By 4 s, the droplet had cleanly slid down the F3-coated plate. On the other hand, the droplet spread on the uncoated plate. Other organic solvents including diiodomethane, decane, and dodecane, which had surface tensions of 50.8, 23.8, and 25.4 mN/m also cleanly glided down these coatings, exhibiting respective SAs of 12 ± 2°, 2 ± 1°, and 2 ± 1° for 5 μL droplets. We suspect these coatings should be similar to the PDMScontaining PU- or epoxy-based coatings that we have examined 25627
DOI: 10.1021/acsami.7b05732 ACS Appl. Mater. Interfaces 2017, 9, 25623−25630
Research Article
ACS Applied Materials & Interfaces higher propensity for PDMS to segregate to the surface in a PU coating. We further speculate that the much faster room-temperature UV curing was not as effective as the thermal curing in promoting the surface stratification of PDMS. To check if this hypothesis was true, we subjected the F3 formulation containing 2.5 wt % PDMS to thermal curing at 170 °C for 2 h. On this thermally cured coating, the SA of 20 μL water droplets was decreased to 41 ± 1°, suggesting an increased extent of PDMS surface chain segregation. A more effective method to promote water sliding is to incorporate some free PDMS into a NP-GLIDE coating and to prepare a silicone-infused UV-curable NP-GLIDE coating.3 The infused free silicone increases the surface coverage by PDMS because it, unlike P0II-g-PDMS, does not contain a tethering P0II backbone that competes with PDMS for surface sites. Despite the water SA, the F3 coating readily contracted paint sprays. Figure 6b compares the behavior of a commercial red paint after its application onto a fully cured F3 coating and onto a glass glide. After 15 s, the paint was seen to have almost fully contracted on the F3-coated glass slide. However, the uncoated plate remained covered by the paint by this time and thereafter. The nanopool-containing structure of the F3 coating was confirmed via atomic force microscopy (AFM). Figure 7
Figure 8. (a) Variation in the transmittance at 500 nm for 32 μm cured F3 coatings as a function of the PDMS weight fraction in the coatings. (b) Variation of static CAs of water and hexadecane on the cured F3 coatings as a function of the PDMS weight fraction in the coatings.
inset shown in Figure 8a where coatings containing different PDMS weight fractions were placed onto written text. One can crisply discern the text beneath the coatings regardless of the PDMS content. We also measured the CAs of water and hexadecane on F3 coatings at a fixed thickness of 32 μm but different PDMS contents or at a fixed PDMS content of 2.5 wt % but different thickness and the results were plotted in Figures 8b and S6b, respectively. The CAs did not change significantly with coating thickness as anticipated. More interestingly, they did not change with PDMS bulk content when it exceeded 0.50 wt %. In contrast, the contact angles of these two test liquids on a series of NP-GLIDE epoxy coatings did not plateau until the bulk PDMS content surpassed 4.0 wt %.2 Thus, PDMS readily self-enriched on the coating surface in the current UV-curable system. However, the equilibrium or saturated surface PDMS content as revealed from the contact angle values and XPS data was not as high as in the thermally cured epoxy system. Surfaces with Patterned Wettability. An anticipated advantage of a UV-curable NP-GLIDE formulation is the creation of surfaces with patterned wettability. In demonstrating this, we cut the letters Q and U, as well as star and moon symbols, into aluminum foils. These masks were then placed between a UV lamp and the films and the samples were irradiated. After 5 min of irradiation, the samples were then rinsed with acetone to remove the nonirradiated regions. Upon drying, the irradiated region and its surrounding areas were scribbled on multiple times with a marker. This treatment yielded the results shown in Figure 9. Figure 9 clearly shows that antismudge patterns were created on the glass plates. The reagents in the irradiated regions underwent cross-linking and were not rinsed away by acetone. In contrast with the surrounding regions that were not irradiated, these cross-linked regions exhibited antismudge behavior.
Figure 7. AFM phase images of (a) the cross-section and (b) the upper surface of a F3 coating cured for 300 s.
compares the phase images of a cross section and an upper surface of a cured F3 coating containing 5.0 wt % of PDMS, where the cross-section was obtained by microtoming a cured sample. Also shown in Figure S5 are a topography and a phase image for a cross-sectional surface of a F0 coating. A comparison between Figure 7a and Figure S5b allows the immediate conclusion of the existence of nanopools in the F3 but not in the F0 coating matrix. However, the nanopool structure of Figure 7b on the surface of the coating was not as evident as that observed on a previous epoxy NP-GLIDE coating.2 Nonetheless, the surface had a biphasic structure and one component had to be PDMS, because Si was seen on the surface by XPS and the surface demonstrated antismudge properties. The small diameters of the PDMS nanopools dictated that the coatings did not scatter much light and were highly transparent. While Figure 8a plots how the transmittance of 500 nm light through 32 μm coatings varied as a function of PDMS content, Figure S6a shows how the average transmittance of F3 samples at 2.5 wt % PDMS varied as a function of coating thickness up to the coating thickness of 83 μm. The transmittance was high (above 96%) for all the samples examined. The optical clarity can also be appreciated from the
■
CONCLUSIONS Reacting a limiting amount of PDMS−OH and HEMA with HDIT yielded a mixture of DM-PDMS and TM. Mostly TM was prepared by reacting HEMA with HDIT at a molar ratio of 3.21/1.00. Furthermore, P0II or P0II-g-PDMS was prepared by reacting a polyol P0 with 2-cyanatoethyl methacrylate or with PDMS-COCl and then 2-cyanatoethyl methacrylate, respectively. Formulations consisting of DM-PDMS and TM (F1), P0II-g-PDMS and P0II (F2), or P0II-g-PDMS and TM (F3) were then cast along with 2-hydroxy-2-methylpropiophenone to yield films on glass plates. These films were subsequently 25628
DOI: 10.1021/acsami.7b05732 ACS Appl. Mater. Interfaces 2017, 9, 25623−25630
ACS Applied Materials & Interfaces
■
Research Article
AUTHOR INFORMATION
Corresponding Author
*E-mail:
[email protected]. ORCID
Guojun Liu: 0000-0002-9643-5070 Notes
The authors declare no competing financial interest.
■
ACKNOWLEDGMENTS The Lorama Group, Inc., NSERC, and OCE are acknowledged for sponsoring this research. Liu thanks the CRC program for a Canada Research Chair position in Materials Science. We also thank Dr. Kunzhi Shen for taking videos of the wettability tests, Dr. Igor Kozin for useful discussion on FTIR analysis, Jane Wang for taking the AFM images, and Dr. Gabrielle Schatte for obtaining the XPS data.
■
Figure 9. Antismudge patterns generated on glass plates after their coatings were irradiated with light passing through the photomasks (shown as inserts) and the coating precursor in the nonirradiated regions were removed with acetone. The patterns were revealed by multiple marker stokes that eventually covered the nonirradiated regions but contracted in the cross-linked regions.
(1) Rabnawaz, M.; Liu, G.; Hu, H. Fluorine-Free Anti-Smudge Polyurethane Coatings. Angew. Chem., Int. Ed. 2015, 54, 12722− 12727. (2) Hu, H.; Liu, G.; Wang, J. Clear and Durable Epoxy Coatings That Exhibit Dynamic Omniphobicity. Adv. Mater. Interfaces 2016, 3, 1600001. (3) Hu, H.; Wang, J.; Wang, Y.; Gee, E.; Liu, G. Silicone-Infused Antismudge Nanocoatings. ACS Appl. Mater. Interfaces 2017, 9, 9029− 9037. (4) Rabnawaz, M.; Liu, G. Graft-Copolymer-Based Approach to Clear, Durable, and Anti-Smudge Polyurethane Coatings. Angew. Chem., Int. Ed. 2015, 54, 6516−6520. (5) Wu, X.; Liu, M. H.; Zhong, X. M.; Liu, G.; Wyman, I.; Wang, Z. P.; Wu, Y. Q.; Yang, H.; Wang, J. B. Smooth Water-Based Antismudge Coatings for Various Substrates. ACS Sustainable Chem. Eng. 2017, 5, 2605−2613. (6) Bulut, U.; Crivello, J. V. Investigation of the Reactivity of Epoxide Monomers in Photoinitiated Cationic Polymerization. Macromolecules 2005, 38, 3584−3595. (7) Sangermano, M.; Razza, N.; Crivello, J. V. Cationic Uv-Curing: Technology and Applications. Macromol. Mater. Eng. 2014, 299, 775− 793. (8) Miao, H.; Bao, F. F.; Cheng, L. L.; Shi, W. F. Fluorinated Modification of Hyperbranched Polyesters Used for Improving the Surface Property of Uv Curing Coatings. J. Fluorine Chem. 2010, 131, 1356−1361. (9) Miao, H.; Cheng, L. L.; Shi, W. F. Fluorinated Hyperbranched Polyester Acrylate Used as an Additive for Uv Curing Coatings. Prog. Org. Coat. 2009, 65, 71−76. (10) Tan, J. Q.; Liu, W. Q.; Wang, H. L.; Sun, Y.; Wang, S. J. A Novel Mono-Methacryloyloxy Terminated Fluorinated Macromonomer Used for the Modification of Uv Curable Acrylic Copolymers. J. Appl. Polym. Sci. 2016, 133, 43116. (11) Tan, J. Q.; Liu, W. Q.; Wang, H. L.; Sun, Y.; Wang, S. J. Preparation and Properties of Uv-Curable Waterborne Comb-Like (Meth)Acrylate Copolymers with Long Fluorinated Side Chains. Prog. Org. Coat. 2016, 94, 62−72. (12) Tan, J. Q.; Liu, W. Q.; Wang, Z. F. Waterborne Uv-Curable Comb-Shaped (Meth) Acrylate Graft Copolymer Containing Long Fluorinated and/or Polysiloxane Side Chains. RSC Adv. 2016, 6, 34364−34375. (13) Bai, C. Y.; Zhang, X. Y.; Dai, J. B. Synthesis and Characterization of Pdms Modified Uv-Curable Waterborne Polyurethane Dispersions for Soft Tact Layers. Prog. Org. Coat. 2007, 60, 63−68. (14) Chen, W. H.; Chen, P. C.; Wang, S. C.; Yeh, J. T.; Huang, C. Y.; Chen, K. N. Uv-Curable Pdms-Containing Pu System for Hydrophobic Textile Surface Treatment. J. Polym. Res. 2009, 16, 601−610.
vitrified via photolysis. The F1 coatings were cloudy probably because of the inability of the small DM head groups to prevent different PDMS domains from overlapping and forming PDMS-rich macrodomains. Meanwhile, the F2 coatings were optically clear but readily lost their ink shrinking ability after a single writing and erasing cycle. Only the F3 formulation yielded transparent robust antismudge coatings that withstood more than 30 writing and erasing cycles, suggesting that the bonding between P0II-g-PDMS and the cross-linked TM matrix was stronger than that between P0II-g-PDMS and the cross-linked P0II matrix. A F3 formulation containing 2.5 wt % PDMS was photocross-linked to different degrees and the evolution of its wetting performance was monitored as a function of irradiation time. This systematic study revealed that the hexadecane and water CA and SA values plateaued at irradiation times longer than 60 s. However, full-fledged ink contraction ability developed only by irradiation time of ∼300 s. Thus, the ink contraction test is more stringent than the CA and SA tests. The P0II-g-PDMS and TM coating possessed the transparency, internal structure, as well as ink and paint contraction capabilities that are desired of an NP-GLIDE coating. More importantly, the coating could be irradiated under masks and then developed (via removal of the un-cross-linked polymer) to yield cross-linked domains with patterned wettability. We anticipate practical applications for the developed UV-curable antismudge coating and believe in the guidance value of lessons learned from this study.
■
REFERENCES
ASSOCIATED CONTENT
S Supporting Information *
The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acsami.7b05732. Experimental section, polymer characterization results, coating hardness and adhesion test results, XPS data for a F3 coating, and AFM images of a F0 coating (PDF) 25629
DOI: 10.1021/acsami.7b05732 ACS Appl. Mater. Interfaces 2017, 9, 25623−25630
Research Article
ACS Applied Materials & Interfaces (15) Sangermano, M.; Bongiovanni, R.; Malucelli, G.; Roppolo, I.; Priola, A. Siloxane Additive as Modifier in Cationic Uv Curable Coatings. Prog. Org. Coat. 2006, 57, 44−49. (16) Pozos Vazquez, C.; Tayouo, R.; Joly-Duhamel, C.; Boutevin, B. Uv-Curable Bismaleimides Containing Poly(Dimethylsiloxane): Use as Hydrophobic Agent. J. Polym. Sci., Part A: Polym. Chem. 2010, 48, 2123−2134. (17) Zhang, S. W.; Chen, Z. D.; Xu, Z. M.; Gang, S.; Bai, H. Y.; Liu, X. Y. Hydrophobic, Transparent Waterborne Uv-Curable Polyurethane Nanocomposites Based on Polycarbonate and Pcl-Pdms-Pcl Reinforced with Colloidal Silica. J. Coat. Technol. Res. 2016, 13, 1021− 1033. (18) Wooh, S.; Vollmer, D. Silicone Brushes: Omniphobic Surfaces with Low Sliding Angles. Angew. Chem., Int. Ed. 2016, 55, 6822−6824. (19) Cheng, D. F.; Urata, C.; Yagihashi, M.; Hozumi, A. A Statically Oleophilic but Dynamically Oleophobic Smooth Nonperfluorinated Surface. Angew. Chem., Int. Ed. 2012, 51, 2956−2959. (20) Wang, L.; McCarthy, T. J. Covalently Attached Liquids: Instant Omniphobic Surfaces with Unprecedented Repellency. Angew. Chem., Int. Ed. 2016, 55, 244−248. (21) Cheng, D. F.; Urata, C.; Masheder, B.; Hozumi, A. A Physical Approach to Specifically Improve the Mobility of Alkane Liquid Drops. J. Am. Chem. Soc. 2012, 134, 10191−10199. (22) Chen, W.; Fadeev, A. Y.; Hsieh, M. C.; Oner, D.; Youngblood, J.; McCarthy, T. J. Ultrahydrophobic and Ultralyophobic Surfaces: Some Comments and Examples. Langmuir 1999, 15, 3395−3399. (23) Wong, T. S.; Kang, S. H.; Tang, S. K. Y.; Smythe, E. J.; Hatton, B. D.; Grinthal, A.; Aizenberg, J. Bioinspired Self-Repairing Slippery Surfaces with Pressure-Stable Omniphobicity. Nature 2011, 477, 443− 447. (24) Solomon, B. R.; Khalil, K. S.; Varanasi, K. K. Drag Reduction Using Lubricant-Impregnated Surfaces in Viscous Laminar Flow. Langmuir 2014, 30, 10970−10976. (25) Tesler, A. B.; Kim, P.; Kolle, S.; Howell, C.; Ahanotu, O.; Aizenberg, J. Extremely Durable Biofouling-Resistant Metallic Surfaces Based on Electrodeposited Nanoporous Tunstite Films on Steel. Nat. Commun. 2015, 6, 8649. (26) Lafuma, A.; Quere, D. Slippery Pre-Suffused Surfaces. EPL 2011, 96, 56001. (27) Ma, W.; Higaki, Y.; Otsuka, H.; Takahara, A. Perfluoropolyether-Infused Nano-Texture: A Versatile Approach to Omniphobic Coatings with Low Hysteresis and High Transparency. Chem. Commun. 2013, 49, 597−599. (28) Liu, H. L.; Zhang, P. C.; Liu, M. J.; Wang, S. T.; Jiang, L. Organogel-Based Thin Films for Self-Cleaning on Various Surfaces. Adv. Mater. 2013, 25, 4477−4481. (29) Urata, C.; Dunderdale, G. J.; England, M. W.; Hozumi, A. SelfLubricating Organogels (Slugs) with Exceptional Syneresis-Induced Anti-Sticking Properties against Viscous Emulsions and Ices. J. Mater. Chem. A 2015, 3, 12626−12630. (30) Yu, L.; Chen, G. Y.; Xu, H. L.; Liu, X. K. Substrate-Independent, Transparent Oil-Repellent Coatings with Self-Healing and Persistent Easy-Sliding Oil Repellency. ACS Nano 2016, 10, 1076−1085. (31) Milionis, A.; Dang, K.; Prato, M.; Loth, E.; Bayer, I. S. Liquid Repellent Nanocomposites Obtained from One-Step Water-Based Spray. J. Mater. Chem. A 2015, 3, 12880−12889. (32) Burel, F.; Lecamp, L.; Youssef, B.; Bunel, C.; Saiter, J. M. Synthesis and Photoinitiated Polymerization of a New Urethane Acrylate Monomer: Influence of Polymerization Temperature. Thermochim. Acta 1999, 326, 133−141. (33) Pavia, D. L.; Lampman, G. M.; Kriz, G. S.; Vyvyan, J. A. Introduction to Spectroscopy; Cengage Learning: 2008. (34) Schwalm, R. UV Coatings: Basics, Recent Developments and New Applications, 1st ed. ed.; Elsevier Science: Kidlington, 2006; p 322.
25630
DOI: 10.1021/acsami.7b05732 ACS Appl. Mater. Interfaces 2017, 9, 25623−25630
