J. Coat. Technol. Res., 8 (4) 497503, 2011 DOI 10.

1007/s11998-011-9324-3

Investigating the antigrafti properties of a polyurethane clearcoat containing a silicone polyacrylate additive
A. Mohammad Rabea, M. Mohseni, S. M. Mirabedini

ACA and OCCA 2011 Abstract In this study, a polyurethane clearcoat was modied using a silicone polyacrylate antigrafti additive and its surface free energy and mechanical properties were evaluated by various analytical techniques. The results showed that surface free energy decreased by replacing the polyol resin with the additive up to 10 mol%. Dynamic mechanical thermal analysis (DMTA) showed that the presence of the additive increased the storage modulus of the samples at ambient temperature. Evaluation of stainability of the coatings containing the additive revealed that a spray paint would be repelled from the surface showing a proper antigrafti behavior. The additive did not affect the surface roughness of the lms, indicating that the antigrafti properties resulted from changes in surface chemistry due to migration of the additive to the surface. Keywords Antigrafti, Stainability, Polyurethane, Silicone additive, Surface free energy of coatings have been performed intensively for the last few years. In general, three approaches have been recently mentioned in the literature to prevent the effects of grafti.3 The rst approach is based on coatings with a very low surface free energy using uoro- or silicone-based binders or their combinations.3 The second aims at making coatings that resist the rigorous cleaning procedures and aggressive agents used to remove grafti. The third one, however, uses a wax-based topcoat that will be cleaned by washing when soiled with grafti. Today, the chemistry of antigrafti coatings includes: acrylic or polyurethane uoropolymers, acrylic emulsions, solvent- or waterbased silicones and silanes, waxes, and sacricial polysaccharides.4 Because of superior mechanical properties and outstanding weathering resistance, polyurethanes are used as the main permanent antigrafti coatings.5 When grafti contact the coating they tend to wet the surface and then penetrate to the coating. Therefore, a balance of low surface energy and low free volume (high modulus) can prevent staining the surface and results in antigrafti properties. A review on the chemistry, degradation, and biodegradation of silicones and silanes is discussed by Graiver and coworkers.6 They showed that linear silicone polymers and cyclosiloxanes are readily degraded in the environment by natural processes. Tarnowski and coworkers4 investigated biodeterioration and performance of different systems including a uoropolymer, a silicone elastomer and a silane compound as antigrafti coatings on sandstone and marble. When the results of the antigrafti coatings were compared, the silane-based coating showed the best performance. The uoropolymer, however, revealed the worst result, and silicone elastomer stood in the middle. Feng et al.7 investigated the surface properties of the polyurethanes modied by dihydroxybutylterminated polydimethyl siloxane (DHPDMS). They showed that when DHPDMS content was about

Introduction
Whether it is considered as an art or vandalism, grafti has been targeted by different technologies. It has been reported that the cost of grafti abatement exceeds $12 billion annually in the U.S. alone.1,2 The research and development to improve the antigrafti properties
A. M. Rabea, M. Mohseni (&) Department of Polymer Engineering and Color Technology, Amirkabir University of Technology, P.O. Box 15875-4413, Tehran, Iran e-mail: [email protected] S. M. Mirabedini Iran Polymer and Petrochemical Institute, P.O. Box 14185/458, Tehran, Iran

497

J. Coat. Technol. Res., 8 (4) 497503, 2011

5 wt%, good surface and mechanical properties was achieved. No evaluation, however, was reported on antigrafti properties of this system. In this work, a series of silicone containing polyurethane coatings were prepared using an OH-functional silicone-modied polyacrylate as a stir-in additive. The effects of additive content on the bulk and surface properties were investigated using different analytical techniques. Also, colorimetric studies were employed to investigate the performance of the coatings against grafti.

Experimental
BYK-Silclean3700 (a solution of 25 wt% OH-functional silicone-modied polyacrylate in methoxypropylacetate), was obtained from BYK Altana group. This additive contains a polysiloxane backbone with organic modication at two ends with pendant OH-functional groups. The acrylic polyol, Uracrone CY433 (a 60 wt% solution in xylene/butylacetate mixture), was purchased from DSM. Desmodure N75 (a 75 wt% solution in a xylene/propylene glycol mono methyl ether acetate mixture) was supplied by Bayer Corporation. The mixing ratio OH:NCO was 1:1.2 of molar. All materials were used as received, without any further purication. Coatings were prepared with substitution of 5, 10, 15, and 100 mol% of Uracrone CY433 with additive, followed by mixing for 15 min under moderate agitation. The reason to replace some parts or the whole polyol is because this additive is an OH-functional compound enabling it to take part in the reaction with isocyanate groups. PU0% represents polyurethane without any antigrafti additive and PU100% means that the whole polyol has been substituted with the additive. The samples were then labeled according to the replaced polyol mol%. For example, PU5% represents 5 mol% of polyol substituted by the additive. Also, pure polyurethane without any additive (PU0%) was used as a reference. Various formulations of prepared antigrafti coatings are listed in Table 1.

To prepare thin solid lms, the required amounts of the polyisocyanate component were added to the mixtures and applied on glass substrates using a lm applicator (BYK-Gardner). The lms were then left to cure for at least 1 week in ambient. The dried lms had thicknesses of approximately 35 5 lm. Free-standing lms were also prepared by immersing the glass coated samples in de-ionized water for about 30 min and then dried at room temperature for 1 week. The wettability of surface is a key parameter to control the effect of grafti on the surface. Contact angle measurement can help in calculating surface energy and therefore the wettability condition. This was performed using a Kruss K14 tensiometer. The measurement is based on a dynamic method known as Wilhelmy approach. The liquids used to characterize free energies were water and formamide, the surface tensions of which are shown in Table 2. Mechanical properties for evaluation of the storage modulus of the coatings were tested with a Triton Tritec Model 2000 dynamic mechanical thermal analysis (DMTA) at a temperature range between 0 and 150C at a frequency of 1 Hz. The mode of the analysis was tension and heating rate was 5C/min (according to ASTM E1640-04). For evaluation of antigrafti properties, a black alkyd-based spray paint and a black permanent marker (Staedtler-Germany) were used. An area of 4 9 4 cm of the coatings was stained with the paint and the marker separately, and after 1 day in ambient condition, cleaned by a dry paper tissue. Visual evaluation was performed for painted samples. The effect of marker on the color change of the coatings was evaluated by measuring color attributes (L a* b*) before any grafti and after removing the marker. The measurement was conducted on the glass coated samples using a white calibrated specimen by a colorimeter model Mini Scan XE plus (Average CIE LAB 10/D65). Effect of inclusion of additive on the topography of coatings was studied by atomic force microscopy (AFM) imaging using a Dual scope/Rasterscope C26, DME, Denmark.

Results and discussion

BYK-SILCLEAN Uracrone 3700 (mol%) CY433 (mol%) 0.05 0.1 0.15 1 1 0.95 0.9 0.85

Desmodure N75 (mol%) 1.2 1.2 1.2 1.2 1.2

Effects of additive concentrations on the advancing and receding contact angles using water and formamide are presented in Table 3.
Table 2: Surface tensions of test liquids at 20C Probe liquids cD lv Water Formamide 21.8 39.5 Surface tension (mN/m) cP lv 51.0 18.7 clv 72.8 58.2

The number shows the mol% of polyol, which is substituted with the additive

498

J. Coat. Technol. Res., 8 (4) 497503, 2011

Contact angles of the probe liquids increased with increasing additive concentration, meaning that the surfaces had become hydrophobic. In real surfaces, the equilibrium contact angle is not exactly the Youngs contact angle, but changes around it. As surfaces are not perfectly smooth or chemically homogeneous, the contact angle lies between two

Table 3: Advancing (had) and receding (hrec) contact angles of water and formamide on the lms Samples had PU0% PU5% PU10% PU15% PU100% 74.6 90.9 102.9 103.7 104.8 Water hrec 49.7 74.7 84.1 84.2 88.4 Formamide had 65 83.9 95 94.9 96.1 hrec 20.4 45.8 53.7 58.2 64.3

40

35 30 25 20 15 10 5 0 PU 0% PU 5% PU 10% PU 15% PU 100%

Sample
Surface free energy Disperse component Polar component

Fig. 1: Surface free energy of samples by the Wu method

values dening a hysteresis behavior: had > h > hrec, where had and hrec are the advancing and receding contact angles, respectively. Their difference is termed the contact angle hysteresis. The wetting properties of the solid are also modied by its roughness, so the contact angle is different from that of the same smooth surface. On rough surfaces, the water repellency is enhanced as the drop/substrate interaction decreases. For samples containing 5 mol% additive and more, the advancing contact angles of water exceed 90, which is the limit to consider the surface water repellent.8 The difference between receding and advancing contact angles shows a hysteresis effect. This hysteresis arises from chemical heterogeneity (Cassie effect) and/or surface roughness (Wenzel effect).9 The lower this difference, the lower is the roughness and greater is homogeneity upon increasing additive. The Wu method10 was employed to calculate surface free energy of samples containing different levels of additive as shown in Fig. 1. As shown in Fig. 1, surface free energies decreased with increasing additive concentration up to 10 mol%. This behavior probably is due to the migration of the low surface tension additive to the surface of the coatings. With 10 mol% of additive and greater, there may be the maximum presence of the additive on the surface. Also both disperse and polar components of surface energy decreased with increasing additive concentration. However, there is a strong reduction in polar component and a somewhat smaller reduction in disperse component. The migration of the low surface tension additive to the surface and its replacement with polar parts at the surface seems to be effective in reducing the surface energy of the coatings. The results of DMTA for different samples are shown in Figs. 2 and 3. Apart from surface energy, the extent of grafti effect on coating depends on the ability of grafti to penetrate into the lms. This can be evaluated using Tg
PU 10% PU 15% PU 100%

Surface free energy (mN/m)

5.0E+09
PU 10% PU 5%

Storage modulus (Pa)

PU 15%

5.0E+08

PU 0%

PU 100%

5.0E+07
PU 0% PU 10% PU 5% PU 15% PU 100%

5.0E+06 0 20 40 60 80 100 120 140

Temperature (C)
Fig. 2: Storage moduli vs temperature for different samples

499

J. Coat. Technol. Res., 8 (4) 497503, 2011

1.2

1.0

0.8

tan

0.6

0.4

0.2

0.0 40 50 60 70 80 90 100 110 120

Temperature (C)
Fig. 3: Tan d vs temperature for different samples

Table 4: Tg and crosslinking density of different samples Samples PU0% PU5% PU10% PU15% PU100% Tg (C) 78.2 84.9 78 71.8 73.3 Crosslinking density (mol/m3) 2.97 1.67 2.82 1.3 5.33 9 9 9 9 9 103 103 103 103 102

Table 5: Reaction of spray paint on the coatings and effectiveness of grafti removal Samples Reaction of spray paint on the coatings Not repelled Repelled Repelled Repelled Repelled Removal effectiveness Not removed Partially removed Completely removed Completely removed Completely removed

PU0% PU5% PU10% PU15% PU100%

and crosslinking measurement by DMTA. Figures 2 and 3 show that in samples in which polyol has been replaced by the additive (PU5%, PU10%, and PU15%), the storage moduli at ambient temperature are greater than that of PU0%. However, when the whole polyol is replaced by the additive (PU100%), a decrease in storage modulus can be seen. These suggest that the additive can be involved in the curing process of polyurethane. The Tan curves shown in Fig. 3 reveal that phase separation has occurred for some samples. For PU0% and PU5%, there may be no obvious phase separation as there can be seen only a single peak in tan curves. However, the existence of two maxima for PU10%, PU15%, and PU100% indicate the possibility of phase separation. These show that phase separation is happening for samples containing additive in excess of 10%. The origin of this phase separation might be because of differences in surface tension of the silicon part of additive and that of polyol.1113 Table 4 lists the Tg and crosslinking density obtained from data shown in Figs. 2 and 3. It can be seen that Tg of PU5% is higher than that of PU0%. However, for samples containing higher amounts of additive, Tg starts to decrease. The OH content of additive is 3.6, compared to 4.5 for acrylic polyol; the inclusion of this reactive additive will presumably result in a decrease in crosslinking density.

It is seen that while there is a decrease in crosslinking density in the presence of additive, storage modulus at ambient temperature has increased. This phenomenon has been previously discussed by Pu et al.14 They emphasized that phase separation of the hard and soft segments of a copolymer in a coating system, compatibility between two blocks or increase in crystallinity, due to the exibility of soft segments, can increase Tg and storage modulus. Hence, as the storage moduli for samples containing additive are greater than that of PU at ambient temperature, the inclusion of additive tends to decrease the penetration of grafti into the coating. Visual observations of the inuence of spray paint on the staining of the samples, as an indication of antigrafti behavior, are listed in Table 5. As can be seen from Table 5, spray paint has been repelled by the coatings containing additive. Also, it can be observed that the spray paint was completely removed from the surface of PU10%, PU15%, and PU100% (Fig. 4a). These are in agreement with the results of Fig. 1 and Table 3, conrming that the surface energies of the coatings were reduced by the incorporation of the additive. However, as it is shown in Fig. 1, no further reduction in the surface free energy can be observed for the samples containing 10 mol% and higher.

500

J. Coat. Technol. Res., 8 (4) 497503, 2011

(a)

Before

After

(b)
Before

After

Fig. 4: Surface appearance of the samples containing different mol% additive, before and after spray paint (a) and marker (b) removal

Table 6: Color attributes of lms before any grafti and after marker removal Samples Before any grafti After removing grafti L PU0% PU5% PU10% PU15% PU100% 88.43 88.43 88.54 88.63 88.49 a* 2.73 2.67 2.73 2.66 2.55 b* 1.48 1.25 1.32 1.25 1.24 L 62.93 81.52 80.26 80.96 74.13 a* 6.06 3.12 3.24 3.21 3.37 b* 7.13 27.12 1.34 7.39 1.82 8.87 1.65 8.22 4.25 15.4 DE

In a further approach, a permanent black marker was used to make some grafti on the surfaces of the samples. For quantitative comparison of the coatings appearance upon exposure to grafti marking, the color attributes, L a* b* (numbers are the average of three different measurements) and color differences, DE, between the lms before any marking and after the grafti removal were recorded as listed in Fig. 4(b) and Table 6.

Table 6 shows that the incorporation of additive does not impart any color to the samples. Moreover, the highest DE after grafti removal is related to PU0% (27.12). This is because of a greater reduction in lightness and less reduction in a* and b*. By comparing the PU0% with other samples, it can also be seen that these changes are lower in L and b* and negligible in a*. This means that the color of samples changes from slightly yellow-green to faint blue-green. For PU100%, color change is slightly greater. L and b* reduction are more than those of coatings with 5, 10, and 15 mol% of additive. This can possibly be explained due to the low storage modulus of this sample among coatings containing additive at ambient temperature as shown by DMTA. As the most color change is related to lightness, the lightness of the coatings before any grafti and after removing permanent black marker is shown in Fig. 5. A little lightness difference (DL) between the coatings containing 5, 10, and 15 mol% of additive reveals the effect of additive to maintain the surface energy as low as possible to make the lms resist against marking. The increase in difference for

501

J. Coat. Technol. Res., 8 (4) 497503, 2011

PU100%, however, may be ascribed to the increase in the rate of grafti penetration to the sample as discussed above. Hence, storage modulus may be another important factor in surface cleanability, because higher storage modulus causes a lower penetration of grafti to the coating, improving the antigrafti behavior. In a similar work, Scheerder and coworkers3 synthesized uorinated polymers with antigrafti properties. They observed that with increase in uorine content, in spite of reduction in surface energy, antigrafti properties became worse. They assigned the increase in uorine content to crosslinking density reduction, so grafti could permeate to the coating more easily. Therefore, low surface free energy and high storage modulus together are the two detrimental parameters to obtain a proper antigrafti coating. Figure 6 shows AFM topographic images of PU0%, PU10%, and PU100% samples. The images reveal that surfaces are relatively smooth. The calculation of Z value range in the specied area (Sy), the range from the mean height of the ve tallest summits to the mean depth of the ve deepest valleys (Sz), the mean of the deviation from the area mean Z value (Sa), and

a measure of the dispersion of the data from the area mean (Sq) are listed in Table 7. The data presented in Table 7 show that inclusion of additive has no signicant effect on the surface roughness of coatings. Hence, the difference between the advancing and receding contact angles should not arise from the roughness and may be solely related to the change in surface chemistry. It means that in Wenzel equation (Coshi = iCosh0),15 when the h0 is less than 90, the higher the i (roughness factor or rogosity value) the greater is the chance of surface wetting. However, as the contact angles climb to values greater than 90, the rougher the surface, the greater hi becomes. This tends to cause the coatings to show a more hydrophobic behavior.

Conclusion
Coatings with antigrafti properties have been produced by incorporating a silicone-modied polyacrylate additive into a polyurethane lm. The results showed that surface free energy of the samples decreased notably in the presence of additive up to 10 mol%. This led to a decrease in the wetting of the surface by grafti. The grafti was then successfully repelled. DMTA revealed that the presence of silicone additive could increase the storage modulus at ambient temperature and gave rise to decrease in penetration of grafti to the coatings. The mechanism of action of surface energy reduction was assumed to be migration of silicone additive chains to the surface. Colorimetric
Table 7: Roughness parameters of PU0%, PU10%, and PU100%

100 90 80 70 60 50 40 30 20 10 0 PU 0% PU 5%
Without graffiti

Lightness (%)

PU 10%

PU 15%

PU 100%

Sample
Removed graffiti after 1 day

Samples PU0% PU10% PU100%

Sy (nm) 4.33 5.75 3.45

Sz (nm) 4.05 3.89 3.02

Sa (nm) 0.359 0.373 0.321

Sq (nm) 0.484 0.477 0.403

Fig. 5: Lightness of the coatings before any grafti and after removing marker

[nm] 3.0

[nm]

[nm] 5

(a)

[3.66 nm]4.33 nm

(b)

[3.74 nm]5.75 nm

(c)

[2.55 nm]3.45 nm

3.5 3.0 2.5 2.0 1.5

2.5

4 3
50

2.0 1.5

50

0n

0 50

nm

50

0n

50

m 0n

0n

50

0n

1.0

Fig. 6: AFM topographic images of (a) PU0%, (b) PU10%, and (c) PU100%

1.0 0.5 0

0.5

1 0

502

J. Coat. Technol. Res., 8 (4) 497503, 2011

analysis also showed that the greatest color change after removing a permanent marker corresponded to coating without any additive. Also, it was concluded that low surface free energy and high storage modulus were most likely the key parameters to obtain coatings with antigrafti properties.

References
1. Maty, J, Anti-Gratti Coatings: Taking Aim Against Tagging. J. Arch. Coat., 4 1826 (2007) 2. Zielink, A, Durability: Rules of the Gratti Game. J. Arch. Coat., 4 2934 (2007) 3. Scheerder, J, Visscher, N, Nabuurs, T, Overbeek, A, Novel, Water-Based Fluorinated Polymers With Excellent Antigrafti Properties. JCT Res., 2 617625 (2005) 4. Tarnowski, A, Zhang, X, McNamara, C, Martin, ST, Mitchell, R, Biodeterioration and Performance of AntiGrafti Coatings on Sandstone and Marble. J. ACCR, 32 316 (2007) 5. Wu, X, Rosen, RP, Development of Low-VOC 2K Waterborne Polyurethane Anti-Graft Coatings. JCT Coatings Tech., 5 6672 (2008) 6. Graiver, D, Farminer, KW, Narayan, R, A Review of the Fate and Effects of Silicones in the Environment. J. Polym. Environ., 11 129136 (2003)

7. Feng, L, Zhang, X, Dai, J, Ge, Z, Chao, J, Bai, C, Synthesis and Surface Properties of Polyurethane Modied by Polysiloxane. Front. Chem. China, 3 15 (2008) 8. Adamson, AW, Physical Chemistry of Surfaces. Wiley, New York (1990) 9. Hikita, M, Tanaka, K, Nakamura, T, Kajiyama, T, Takahara, A, Super-Liquid-Repellent Surfaces Prepared by Colloidal Silica Nanoparticles Covered With Fluoroalkyl Groups. Langmuir, 21 72997302 (2005) 10. Wu, W, Nancollas, GH, Determination of Interfacial Tension from Crystallization and Dissolution Data: A Comparison with Other Methods. Adv. Colloid Interface Sci., 79 229279 (1999) 11. Tyagi, D, Yilgor, I, McGrath, JE, Wilkes, GL, Segmented Organosiloxane Copolymers: 2 Thermal and Mechanical Properties of SiloxaneUrea Copolymers. Polymer, 25 18071816 (1984) 12. Majumdar, P, Webster, DC, Surface Microtopography in Siloxaneepolyurethane Thermosets: The Inuence of Siloxane and Extent of Reaction. Polymer, 48 74997509 (2007) 13. Majumdar, P, Webster, DC, Preparation of SiloxaneUrethane Coatings Having Spontaneously Formed Stable Biphasic Microtopograpical Surfaces. Macromolecules, 38 58575859 (2005) 14. Pu, Z, Mark, JE, Li, Z, Zhu, J, Some Block Copolymers Illustrating the Effects of Siloxane and Silane Units on the Properties of TerephthalateGlycol Thermoplastics. Polymer, 40 46954701 (1999) 15. Wenzel, RN, Resistance of Solid Surfaces to Wetting by Water. Ind. Eng. Chem., 28 988994 (1936)

503