Novel, water-based fluorinated polymers with excellent antigraffiti properties.

In this article we describe novel, water-based, crosslinkable fluorinated polymers that form coatings with excellent antigraffiti properties. The synthesis of the binders and the surface and bulk properties of their coatings are discussed. The surface properties of these coatings are characterized in terms of their surface-free energy, as calculated from static contact angle measurements. The distribution of the fluorine atoms throughout the coating is measured by X-ray photoelectron spectroscopy (XPS). The bulk properties are studied by determining the crosslink density through dynamic mechanical thermal analysis (DMTA), and the effect of the crosslinking conditions on the crosslink density and the antigraffiti properties is discussed. The results indicate that a combined action of surface and bulk properties gives these coatings their excellent antigraffiti properties. The applicability of these polymers as protective coatings for metal and concrete surfaces are demonstrated.

Keywords: Stratification, crosslinking, cure, acrylics, fluorinated polymers, latexes, colloids, emulsions, antigraffiti, water-based, crosslinked, emulsion polymerization 

**********

Fluorinated polymers have very interesting properties that primarily result from the unique properties of the fluorine atom and the C-F bond. The fluorine atoms give these polymers their characteristic low surface free energy that makes them water- and oil-repellent. These properties make fluorinated polymers very useful as protective coatings and, as such, they have been used for antifouling and antisoiling applications. In addition, the C-F bond is very stable towards visible and UV-light, making fluorinated polymers resistant to degradation in outdoor applications.

One area where fluorinated polymers can be very useful is as antigraffiti coatings. In general, graffiti is perceived as aesthetically unappealing and the cost of graffiti removal is immense. It is reported that in Berlin alone an estimated 25,000 spray cans are used every day. (1) The Deutsche Bahn AG reported spending a three-digit million Euro every year to remove graffiti. Also, the Dutch Railways spent several million Euro a year and was cleaning up to 150 [m.sup.2] a day (2001), while in the city of London the cost for removing graffiti was reported to be as high as 140 million Euro in 2002!

Graffiti resistance is generally believed to be a surface property associated with the water- and oil-repellent character of fluorinated polymer coatings. The development of coatings having antigraffiti properties has been ongoing for many years in the coatings industry. In general, three approaches have been pursued. The first approach aims at using coatings with a very low surface free energy by using fluoro- or silicone-based binders or combinations thereof. A low surface free energy would make it easy to remove any applied graffiti due to a poor wetting of the coating by the graffiti. The second approach aims at making coatings that can resist the rigorous cleaning procedures and aggressive cleaning agents that are used to remove graffiti. In this way graffiti can be removed without affecting the coating. A third approach uses a topcoat that can be removed when soiled with graffiti. The first approach has by far received the most attention because it has the potential to significantly reduce the cost of removing graffiti and expand the lifetime of coatings, since this will be much less affected by graffiti in the first place. Also, from an environmental point of view, the use of less extensive cleaning methods and less aggressive cleaning agents is preferred.

Due to their very low surface free energy, fluorinated components migrate to the surface of a coating. This stratification process is reported to produce coatings having a very low surface free energy due to the selective enrichment at the surface with fluoro moieties, despite using low concentrations of fluorine (1-2 wt%). (2) Another way to achieve stratification is by blending fluorinated polymer latex particles with nonfluoro latex particles. (3) Both these approaches result in a thin surface layer enriched with fluoro components. However, such a thin surface layer is likely to erode in time, exposing the underlying coating, which will be deficient in fluoro compounds and have a much higher surface free energy. The protective properties will then be lost. Furthermore, in practice, crosslinking is generally required to obtain proper outdoor durability and chemical resistances, such as resistances against various cleaning agents. Crosslinking, however, counteracts the stratification process, resulting in a lower density of fluoro components at the coating surface. (4-5)

Another problem for antigraffiti systems is recoatability. A second layer of the same coating cannot be applied over the first layer. As well for silicone-based technology as for perfluorinated polymer technology, the coating that is formed has very low surface free energy properties resulting in a nonstick and difficult-to-wet surface. If a second layer of coating needs to be applied, this will also have wetting and adhesion problems and, most commonly, to such an extent that recoating is not possible. Recoatability is, however, desired, since the frequent removal of graffiti by (aggressive) solvents will affect the antigraffiti coating itself. This manifests itself in a loss of gloss or, in case of slow removal of the fluorine or silicone enriched top layer, a loss of antigraffiti properties. Currently the only solution is to remove the original antigraffiti coating and apply a new one. This is very elaborate because these antigraffiti systems are usually highly resistant and crosslinked, which make them very difficult to remove.

Another important requirement is that the antigraffiti coating should be able to resist several exposures to graffiti and the following cleaning procedures. When, after only a few exposures to graffiti, the coating would become dull or would lose its protective character, its use would be limited. The coating should maintain its protective character over a long period of time. Finally, it should retain its protective character after outdoor and UV exposures.

[FIGURE 1 OMITTED]

In this article we report novel, water-based fluorinated polymers with excellent antigraffiti properties, (6-7) which overcome the aforementioned problems of loss of protective properties and recoatability and which are crosslinked and suitable for outdoor applications.

EXPERIMENTAL

Polymerizations

All polymer emulsions were prepared by semibatch emulsion polymerization. Of the total monomer feed, 5% was precharged at 85[degrees]C and an emulsified monomer feed was added during 90 min at 85[degrees]C using ammonium persulphate as an initiator. After completion of the monomer feed, the reactor content was kept at 85[degrees]C for another 30 min. Then the pH was adjusted to 6.5-7.5 using a 25% dimethyl aminoethanol solution. Finally, the reactor contents were cooled to room temperature, filtered, and collected. The specifications are shown in Table 1. All polymer emulsions were prepared with theoretical solids content of 30%. The amount of coagulum for all emulsions was below 0.1 wt%. Methyl methacrylate (MMA) and n-butyl methacrylate (nBA) were used to set the [T.sub.g] of the polymers using the Fox equation. The theoretical [T.sub.g] of the polymers was around 40[degrees]C. The weight average molecular weight of the polymers was comparable. Trifluoroethyl methacrylate was used as fluoromonomer.

Determination of the Antigraffiti Properties

The fluorinated coatings were prepared by mixing the fluorinated hydroxyl functional binders with commercially available isocyanate crosslinkers like Bayhydur 3100, available from Bayer (with a NCO:OH ratio of 1.5:1). Films (100 [micro]m wet) were cast on Leneta Cards using a 100-[micro]m wirerod. One set of films was dried at ambient conditions for 16 hr, one set for seven nights at ambient conditions, and one set was dried for four hours at ambient conditions followed by 16 hr at 50[degrees]C. Edding 3000 permanent markers, supplied by Edding AG (black, brown, yellow, green, orange, red, and blue), were used to mimic graffiti on the dried films. The ink was left on the film for 16 hr after which a paper tissue wetted with isopropyl alcohol (IPA) or 2-butanone (MEK) was used to remove the graffiti from the coating surface. To clean the coating surface, Dekontaminol (DEK), available from Bernd Neumann Malermeister GmbH, Berlin, was applied on the graffiti and left for five minutes, after which it was removed using a paper tissue. The removal was judged visually and rated from 5 (no visible mark left and no damage to the film) to 0 (coating removed or significant graffiti left) for each color applied. The maximum total rating was 35 (i.e., cumulative total for the seven markers). A similar procedure was used for the tests on metal and concrete. The coatings on concrete were dried at ambient conditions only.

[FIGURE 2 OMITTED]

Analytical Methods

The viscosity of the polymer emulsion was determined with a Brookfield viscometer at 23[degrees]C using spindle I at 60 rpm. Free monomer levels were measured using gas chromatography equipped with a FID detector. Particle sizes were measured using a Malvern 3000 HS apparatus using demineralized water as medium.

Gradient polymer elution chromatography (GPEC) was performed on a Waters Alliance 2690 system with a Waters 996 photodiode array detector and an evaporative light scattering detector of Polymer Laboratories (PL-ELS1000). The concentration of the samples was 25 mg/ml and the injection volume was 10 [micro]l. The used eluents were tetrahydrofuran (THF), acetonitril (ACN), and water. Trifluoroacetic acid, in a 0.1 vol% amount, was added to the eluents. A microsphere C18 column (Chrompack Varian) was used at 40[degrees]C. The gradient was comprised of water (100%)-ACN (100%)-THF (100%). The flow rate was 0.5 ml/min.

Dynamical thermal properties were assessed with dynamic mechanical thermal analysis (DMTA). Films were cast on a torsional braid and dried for 16 hr at ambient conditions or four hours at ambient conditions followed by 16 hr at 50[degrees]C. DMTA plots were recorded using a DMTA V apparatus from Rheometric Scientific over a temperature range of -40[degrees]C to +160[degrees]C at a frequency of 1 Hz and a heating rate of 4[degrees]C/min. The temperature range covered widely spans of actual [T.sub.g]s of the coatings. For all coatings a NCO:OH ratio of 1.5:1 was used.

For static contact angle measurements, films were cast onto Q-panels and allowed to dry for four hours at ambient conditions followed by 16 hr at 50[degrees]C. A NCO:OH ratio of 1.5:1 was used. A section of the coated panels was placed on a horizontal surface inside a small chamber equipped with a lid, within the field of view of Kruss Type G1 contact angle measuring microscope. An 8-[micro]l droplet of demineralized water or diiodomethane was placed on the coated surface of the panel using a syringe and the lid was closed. The droplet was allowed to spread out across the coating and the contact angle was measured after one minute.

XPS measurements were done in a VG-Escalab 200 spectrometer using an aluminum anode (Al K[alpha] = 1486.6 eV) operating at 510 VA with a background pressure of 2 X [10.sup.-9] mbar. As organofluoro compounds gradually lose some fluorine under the X-ray beam, the measurement time per sample is limited. Samples were measured sequentially at 0[degrees]C, 60[degrees], 75[degrees], and finally again at 0[degrees] normal to the surface. This corresponds to an information depth (95% of the signal) of 10, 5, 2.5, and finally again at 10 nm, assuming the effect of elastic scattering to be negligible. The fluorine over carbon atomic ratio (F/C) was determined from the integrated intensity ratio of the F1s ([I.sub.F1s]) versus the C1s ([I.sub.C1s]) emission according to:

[F]/[I] = [[[I.sub.F1s]]/[[I.sub.C1s]]] X [alpha]

The relative sensitivity factor ([alpha]=0.145) was determined from the sample containing 5.09 wt% fluorine and dried for one night at 50[degrees]C by comparing the intensity of the C-F component in the C1s spectrum to the intensity of the F1s emission.

[FIGURE 3 OMITTED]

RESULTS

Polymerizations (6)

The polymer emulsions were prepared by emulsion polymerization with variations in the concentration of fluorine (FP1-6) and hydroxyl groups (FP4, 7-10). FP1-6 all contained 1.31 wt% of hydroxyl groups; FP7-10 all contained 1.02 wt% fluorine. Fluorinated polymer 4, which contained 1.02 wt% of fluorine and 1.31 wt% of hydroxyl groups, was used as reference. The values for the amount of hydroxyl groups and fluorine were calculated on the basis of the polymer backbone compositions. The large particle size of polymers FP6 and FP10 was caused by flocculation of several particles. Despite this, however, no coagulum was formed. The specifications of the polymer emulsions are shown in Table 1.

The emulsion polymerization of FP4 was followed in some detail. The particle size development is shown in Figure 1, and the conversion of MMA, BA, and the fluorinated monomer is shown in Figure 2.

Figure 1 shows that the particle size increased rapidly during the first 20 min followed by a more gradual increase. During the first 20 min the precharged monomers, as well as the monomers that were added, were polymerized. This explains the more rapid increase in particle size during the initial stages. After 20 min, only the monomers that were fed were polymerized, resulting in a gradual and somewhat slower increase in particle size.

Figure 2 shows the free monomer levels during the polymerization of FP4.

During the initial stages the amount of free monomer increased to a plateau level of about 0.3 wt% MMA and nBA. This level was reached after approximately 40 min into the monomer feed. The free monomer level dropped quickly after the monomer feed ended after 90 min, indicating a rapid consumption of the remaining monomers. The level of free fluoromonomer throughout the reaction was low, indicating that transport of this monomer from the monomer droplets to the polymer particles poses no limitation. The final conversion was over 99.9%.

GPEC was used to study the chemical composition distribution of the polymers. Despite the large variations in concentration of fluorine, the chromatograms showed one identical peak with respect to the peak shape and broadness, indicating that the fluorine was incorporated homogeneously over this range. No indication for the presence of fluorine-rich polymer chains was found. When increasing the concentration of hydroxyl groups, the chromatograms shifted, which was due to the increase in concentration of hydroxyl functional monomer. No indications for the presence of fluorine-rich polymer chains were found. The GPEC results showed that irrespective of the variation of the concentration of fluorine or hydroxyl groups, the fluoromonomer was incorporated homogeneously.

Antigraffiti Results

Clearcoat and pigmented formulations based on the binders listed in Table 1 are shown in Table 2. Formulations had been developed for application by rolling, brushing, and spraying. For each of these application methods, the viscosity was adjusted with thickeners to such a value that defect-free films were applied. The ingredients were added in the order listed in Table 2.

The antigraffiti properties of the coatings prepared from these binders were determined as described in the Experimental Section. The results are shown in Table 3. The antigraffiti results indicate that the weight percentages of fluorine and hydroxyl groups both affect the antigraffiti properties. The graffiti was removed more readily with MEK and DEK than with IPA. Dekontaminol, which is a commercially used graffiti remover, was especially effective.

Table 3 shows that the antigraffiti properties are better when the films are dried for four hours at ambient conditions followed by 16 hr at 50[degrees]C. When dried at room temperature for one night, the antigraffiti properties were moderate but improved significantly when the coatings were allowed to dry for seven days. This is due to a further progressing of the isocyanate-hydroxyl crosslinking reaction. FTIR studies have showed that both at room temperature and at 50[degrees]C a gradual decrease in intensity of the NCO peak at 2270 [cm.sup.-1] is observed, indicating consumption of the isocyanate groups. This continued for over 50 hr after which, at room temperature, approximately 77% of the NCO groups were consumed while at 50[degrees]C approximately 90% of the NCO groups were consumed.

The contribution of each of the varied parameters is discussed in the following sections. This discussion is based on the antigraffiti results obtained with films dried for four hours at ambient conditions followed by 16 hr at 50[degrees]C. The same trends account for the films dried for one or seven nights at room temperature.

Effect of the Concentration of Fluorine

An increase in the concentration of fluorine from 0 to 1.02 wt% resulted in an increase in the antigraffiti properties followed by a reduction when the concentration was increased further. Thus, already at low concentrations of fluorine very good antigraffiti properties were obtained (35 out of a possible 35) after one night at 50[degrees]C. It was expected that upon increasing the concentration of fluorine, the surface free energy of the resulting coatings would be reduced. To test this, the polar and dispersive components of the surface free energy were calculated using the geometric mean equation derived by Owens and Wendt, (8) based upon the static contact angles of water and diiodomethane on the coatings. (9) The results are shown in Table 4 for films dried for four hours at ambient conditions followed by 16 hr at 50[degrees]C. (10)

As expected, the overall surface free energy ([gamma]) decreased when the concentration of fluorine was increased. This is caused by a strong reduction of the polar component ([[gamma].sup.p]) accompanied by a somewhat smaller reduction of the dispersive component ([[gamma].sup.d]) (Figure 3).

[FIGURE 4 OMITTED]

The polymer containing 5.09 wt% of the fluorine, PF6, had a very low surface free energy with a very small polar component. However, the antigraffiti properties were not good (Table 3). Apparently, there is not a direct relationship between the surface free energy of these coatings and the antigraffiti properties. This is also evident from Figure 3.

When the concentration of fluorine was increased from 0 wt% to 1.02 wt%, the antigraffiti properties increased. A further increase in fluorine resulted in a decrease. The surface free energy, however, showed a continuous decrease with increasing concentration of fluorine. This indicates again that there is no direct relationship between the antigraffiti properties and the surface free energy of these coatings. The coating with the lowest surface free energy was FP6, which contained 5.09 wt% of fluorine. This resulted in a very poor wetting by the applied graffiti, which caused beads of ink on the surface. This so-called "pearl" effect can be seen in Figure 4 (left). The left half of this picture shows the situation after application of the graffiti by the Edding markers, while the right half shows the results after cleaning with Dekontaminol.

As can be seen in the right half of Figure 4, the poor wetting did not result in good antigraffiti properties, since after removal of the graffiti, traces were left behind that are clearly visible.

[FIGURE 5 OMITTED]

[FIGURE 6 OMITTED]

Figure 5 shows the results for FP4. The left shows the situation after application of the graffiti and the right shows the situation after cleaning with DEK. In contrast to FP6, the polymer with 1.02 wt% fluorine, FP4, did not show the "pearl" effect (Figure 5, left) but the graffiti was removed without leaving a trace (Figure 5, right). These results again indicate that antigraffiti is not just a surface property.

Effect of the Amount of Crosslinking

Crosslinking is achieved by the reaction between the hydroxyl groups of the fluorinated binders and the isocyanate groups of the crosslinker. Crosslinking is necessary to obtain resistances against graffiti, and also against various aggressive cleaning agents. Figure 6 shows that the concentration of hydroxyl groups, and hence the amount of crosslinking, has a significant effect on the antigraffiti properties.

When no crosslinking was present, the coatings were seriously affected by the cleaning agents. Cleaning resulted in a significant loss of gloss and was accompanied by partial dissolution of the coatings. An increase in the concentration of hydroxyl groups from 0 wt% to 1.31 wt% resulted in a strong improvement in antigraffiti properties. An increase beyond this concentration resulted in a decrease in antigraffiti. The polymer with 2.18 wt% of crosslinking monomer showed a "pearl" effect, as is also observed for the polymer having 5.09 wt% of fluorine, but the antigraffiti properties were slightly inferior to the polymer having only 1.31 wt% of hydroxyl groups.

The effect of crosslinking was studied in detail by calculating the number of moles of elastically effective network chains per cubic centimeter of coating, [[nu].sub.e], using DMTA measurements. This term can be seen as a measure for the crosslink density and is calculated following [[nu].sub.e] = E'/3RT, in which R is the gas constant and T the temperature in Kelvin. (11) The value for E' at T = 393 K (120[degrees]C) was used for the calculation, Table 5 summarizes the results.

As expected, an increase in concentration of hydroxyl groups from 0 wt% to 1.31 wt% resulted in an increase in crosslink density. A further increase to 2.18 wt% hydroxyl groups, however, did not result in a further increase in crosslink density, but rather led to a slight decrease. Possibly when a certain level of crosslinking is achieved, the molecular mobility of both the crosslinker and polymeric chains is so restricted that no further crosslinking reactions can take place due to limited mixing of crosslinker and polymer chains. It seems that 1.31 wt% of hydroxyl groups represents an optimum between the extent of crosslinking reaction and the molecular mobility within the coating. The improvement in antigraffiti properties with increase in crosslink density can be explained by the fact that a more crosslinked coating is also more molecularly dense and, therefore, contains less free volume. As such, it is difficult for the graffiti to penetrate the coating.

A similar trend was observed for the concentration of fluorine. Going from 0 wt% to 1.02 wt% of fluorine, the crosslink density increased followed by a decrease when more fluorine was introduced. Fluorine atoms have a high molar mass and an increase in concentration of fluorine may reduce the mobility of the fluorinated segments, which at a certain concentration of fluorine may limit the crosslinking reactions. Alternatively, the reduction in surface free energy caused by increased levels of fluorine in the coating may also lead to a reduced level of compatibility between the fluorinated polymer and the isocyanate crosslinker. Apparently, this did not occur at 1.02 wt% fluorine, but at 5.09 wt% fluorine this did affect the crosslinking. Table 5 further shows that a high crosslink density is accompanied with very good antigraffiti properties.

As shown in Table 3, the antigraffiti properties depend on the curing temperature and time. The best results were obtained when the coatings were dried for four hours at ambient conditions followed by 16 hr at 50[degrees]C. The effect of the curing conditions on the crosslink density is shown in Table 6.

Curing at 50[degrees]C clearly resulted in a higher crosslink density compared to curing at room temperature, irrespective of the time of room temperature curing. This was not unexpected since the mobility of the crosslinker and polymer chains will be higher at 50[degrees]C and vitrification is postponed. Prolonged curing at room temperature results in a slight increase in the crosslink density and the coating density and this can explain the improved antigraffiti properties when dried at room temperature for a longer time. The effectiveness of the crosslinking reaction in FP4 is further demonstrated by comparing the crosslink density of the FP4 films cured at room temperature (Table 6) with the values for the polymers having the same amount of OH groups but cured at 50[degrees]C (FP1 to 3 and FP5 to 6 in Table 5). Despite the lower curing temperature, the crosslink density of the room-temperature-cured films of FP4 were always higher than those of all modifications with the same amount of OH groups cured at 50[degrees]C.

The high crosslink density gave polymer FP4 certain barrier properties. When a water droplet containing a dye was placed on top of the coating, no dye was absorbed by the coating during at least 48 hr of exposure. Even when the coating was damaged with a cross-cut tester no absorption was observed. Alternatively, when a water-soluble dye was incorporated in the coating and a water drop was placed on top of the coating, no leaching of the dye to the water drop was observed for at least 48 hr. Again, when the coating was damaged using a cross-cut tester no leaching of the dye was observed. These simple experiments revealed two things: First, these coatings are efficient barriers against dyes, also during prolonged exposure. Second, when a dye has penetrated the coating, removal is very difficult. This implies that it is essential to prevent the graffiti from penetrating a coating since otherwise removal may be difficult.

DISTRIBUTION OF FLUORINE

The distribution of fluorine atoms in the coatings of the polymers having 0.0-5.09 wt% of fluorine was determined by XPS. Figure 7 shows that there is a near linear relationship between the F/C ratio near the surface and the wt% of fluorine for the coatings dried for four hours at ambient conditions followed by 16 hr at 50[degrees]C. The distribution of the fluorine throughout the coatings was measured from a depth of 10 nm, which is considered to be representative for the bulk of the coating, to 2.5 nm, which is considered to be representative for the surface of the coating. Figure 7 shows that the amount of fluorine at any depth increases when more fluorine is incorporated.

Independent of the concentration of fluorine, all coatings showed an enrichment of fluorine near the surface. The F/C ratio for the polymer containing 5.09 wt% of fluorine was 0.107 at a depth of 10 nm and 0.203 at a depth of 2.5 nm. This gave an enrichment factor of 1.9. Analogous, the enrichment factors for the polymers with 2.03, 1.02, 0.76, and 0.25 wt% fluorine were, respectively, 2.1, 2.4, 2.5, and 2.9. Thus, the enrichment near the surface was larger when less fluorine was present. These values, however, are much lower compared to literature where enrichment factors of 20-80 have been reported. (2,12) The high crosslinking density (Table 5) in these coatings possibly explains the lower enrichment of fluorine that was observed. The lower enrichment has two distinct advantages. First of all, when the top layer is damaged or erodes in time, the underlying layers will still have significant amounts of fluorine and, therefore, will still be able to offer protection against graffiti. Hence, the protective character of these coatings will not be compromised by erosion or damages. Secondly, due to the lower enrichment the surface free energy of these coatings is relatively high, making them recoatable.

RECOATABILITY AND COATING PROPERTIES

The recoatability of FP4 was tested in two ways. First, a second layer of paint was applied on top of a first layer that was not exposed to graffiti. The second layer can be applied on top of the first layer as soon as 30 min after application of the first layer. Second, a second layer was applied on top of a paint layer that had gone through five cycles of graffiti removal with Dekontaminol. In both cases, the second layer formed a coating with excellent appearance and good wet and dry adhesion to the first layer. This is quite unique considering the excellent antigraffiti properties that are achieved by this system and the known problems with recoatability of systems based on the perfluorinated or silicone technology. A likely explanation for this is the relatively high surface free energy of FP4, especially when compared to alternative technologies. As is also evident from Figure 7, FP4 does not have a surface highly enriched with fluorine, hence its good recoatability.

[FIGURE 7 OMITTED]

The presented fluorinated polymers adhered well to various substrates such as metal and concrete. Also, the adhesion to various waterborne primers and solventborne two-component primers was good. Table 7 lists some of the properties of a commercial system based on the previously described fluorinated polymers on metal and concrete.

On metal, these fluorinated polymers formed hard, glossy films with excellent chemical resistances. On concrete, despite its basic nature and porosity, these coatings provided excellent protection. Table 7 shows that on both substrates, in addition to the excellent antigraffiti properties, the general resistances and the resistances to Petrol, Diesel, and various oils were also very good. The very high level of resistances was further demonstrated by the excellent resistance against Skydrol, which is a brake fluid used in the airplane industry, and Dot 3, another type of brake fluid. These aggressive substances significantly affect many coating systems.

Outdoor exposure is continuing. Clearcoat formulations exposed for as long as three years in Waalwijk, The Netherlands, showed no reduction in gloss. In addition, the antigraffiti properties of a panel exposed for two years were not deteriorated. These clearcoat panels also showed a low dirt pick-up. As mentioned in the Introduction, the ability to withstand repeated exposure to graffiti and its removal is an important property. Two durability tests were performed, one on metal and one on concrete substrates. Table 8 shows the results of both cases as compared with commercially available, two-component, waterborne polyurethane antigraffiti systems.

On metal, the test was performed according to the specifications from the Deutsche Bahn using a harder version of FP4. In this test, test panels were dried for 14 days at room temperature followed by application of graffiti by felt pens. The stained films were placed in an oven at 50[degrees]C for two days, after which the graffiti was removed with DEK. The films were allowed to recover for three days, after which this procedure is repeated up to 10 cycles. To pass this test successfully, in addition to maintaining the antigraffiti properties, the gloss values must be >50 at 20[degrees] after the 10 cycles.

On concrete, the antigraffiti properties of white pigmented and clearcoats were tested both as a matt and as a glossy coating. The antigraffiti properties of the pigmented coatings were slightly less compared to those of the clear coatings, while no differences were observed between the matt and glossy coatings. In this test three layers of FP4 were applied, allowing for four hours of drying at ambient conditions between each layer. An important result is that no difference in antigraffiti properties was observed when the final coating was allowed to dry for two weeks, or just for two days, prior to the application of the graffiti.

The results clearly show that the performance both on metal and on concrete upon repeated exposures to graffiti and repeated cleaning actions remains at a very high level. On metal, the Deutsche Bahn test was passed successfully. The market reference "2K WB PU1" showed poor test results after only one exposure to graffiti and was not tested further. On concrete, the antigraffiti property of FP4 remained excellent throughout the 10 cycles. Of the competitor product, the initial level of protection was reasonably high but this rapidly dropped during the test.

To assess the performance of these fluorinated coatings on concrete more thoroughly, a series of tests were performed (Table 9). The results shown are for a clearcoat based on FP4 using a NCO:OH ratio of 1.2:1, and in which the isocyanate was mixed in by Heidolph. However, no big differences were observed between the performance of clear and white pigmented coatings when the isocyanate was mixed in by hand or by Heidolph, or when the NCO:OH ratio was varied from 1:1 to 2:1.

The coating based on FP4 was not affected by the various weathering conditions to which it was exposed. In the Weiss Global test the coating was exposed to 50 cycles, each lasting 14 hr, and each consisting of three hours of rain at 20[degrees]C, one hour cooling down to -20[degrees]C and remaining at this temperature for three hours, followed by heating to 50[degrees]C at 50% relative humidity in one hour and keeping the coating under these conditions for three hours, followed finally by cooling down to 25[degrees]C in 20 min and keeping the coating at this temperature for two hours 40 min while exposing it to UV. No deterioration of the coating was observed. In the freeze/thaw test, the coating was exposed to 20 cycles of an eight-hour exposure to salt water at -35[degrees]C followed by a 16-hr room temperature exposure to water. Again, no deterioration was observed. This indicates that the flexibility of this coating is quite high since it was able to withstand strong variations in both temperature and humidity. In addition, the hot tire resistance and the adhesion to concrete and concrete primed with epoxy coatings were very good. These results demonstrate the high level of protection these fluorinated coatings offer, not only against graffiti and many other aggressive substances, but also against various substrates and under various outdoor conditions. Finally, the excellent protective character of these systems was further demonstrated by exposing the coatings to UV. After a coating based on FP4 was exposed to UV for 1000 hr, the antigraffiti properties were completely retained.

CONCLUSIONS

We have described novel water-based fluorinated coatings with excellent antigraffiti properties. We have shown that the crosslinking density, as among others influenced by the amount of fluorine, has a prime impact on antigraffiti properties. An increase in the concentration of fluorine did lower the surface free energy of the coatings but the antigraffiti properties clearly showed a maximum at lower concentrations of fluorine used. This is strong evidence that antigraffiti is not strictly a surface property but that the bulk properties play an important role as well. The reduced mobility, because of the high crosslink density, prevented the graffiti from penetrating the coating while the relative low surface free energy promoted easy removal of the graffiti. This combined action of surface and bulk properties gave these coatings excellent antigraffiti properties.

These coatings have a very high level of chemical resistance against a wide variety of fluids, ranging from acids and bases to various oils and aggressive substances like Skydrol and Dot 3. In addition, the recoatability of these coating is excellent. Moreover, these systems were able to withstand repeated exposures to graffiti and cleaning actions without losing their protective character. Finally, the coatings were able to withstand strong variations in temperature and humidity. This combination of excellent antigraffiti properties, the high level of chemical resistances, good outdoor durability, and the recoatability makes these fluorinated polymers excellent protective coatings.

ACKNOWLEDGMENTS

The authors wish to thank Pascal van den Thillart, Miranda Moeling, Pieter Meulemans, Lucien de Koninck, Andre Harmsen, Czes Blotnicki, and Dennis van Opstal for their contribution to this work. Cilla Schoonenberg is acknowledged for the DMTA experiments, Paul Cools is acknowledged for the GPEC measurements, and Dr. Peter Thune of the Technical University of Eindhoven is acknowledged for the XPS experiments.

References

(1) Stadler, R., Maier, A., and Steidl, N., Eur. Coat. J., 7-8, 63 (2003).

(2) Ming, W., Tian, M., van de Grampel, R.D., Melis, F., Jia, X., Loos, J., and van der Linde, R., Macromolecules, 35, 6920 (2002).

(3) Thomas, R.R., Lloyd, K.G., Stika, K.M., Stephans, L.E., Magallanes, G.S., Dimonie, V.L., Sudol, E.D., and El-Aasser, M.S., Macromolecules, 33, 8828 (2002).

(4) Ming, W., Melis, F., van der Grampel, R.D., van Ravenstein, L., Tian, M., and van der Linde, R., Proc. of 2002 Athens Conference on Coating Science & Technology, 189, Athens, Greece, 2002.

(5) Geurts, J.M., "Latices with Intrinsic Crosslink Activity," Ph.D. Thesis, Technical University Eindhoven, 1997.

(6) WO 03/016412A1.

(7) Scheerder, J., Nabuurs, T., and Overbeek, A., Proc. of 2003 Athens Conference on Coating Science & Technology, 285, Athens, Greece, 2003.

(8) Owens, D.K. and Wendt, R.C., J. Appl. Polym. Sci., 13, 1741 (1969).

(9) Wu, S., Polymer Interface and Adhesion, Marcel Dekker, New York, Basel, p. 169, 1982.

(10) It is known that variations in surface roughness can greatly affect contact angle values. AFM studies of coating prepared from FP1-6 have indicated that the roughness of these coatings was comparable and small.

(11) Hill, L.W. and Kozlowski, K., "Crosslink Density of High Solids MF-Cured Coatings," JOURNAL OF COATINGS TECHNOLOGY, 59, No. 751, 63 (1987).

(12) Thomas, R.R., Anton, D.R., Graham, W.F., Darmon, M.J., Sauer, B.B., Stika, K.M., and Swartzfager, D.G., Macromolecules, 30, 2883 (1997).

Jurgen Scheerder, ([dagger]) Nico Visscher, Tijs Nabuurs, and Ad Overbeek -- DSM NeoResins*

Presented at the 82nd Annual Meeting of the Federation of Societies for Coatings Technology, October 27-29, 2004, in Chicago, IL.

* Sluisweg 12, P.O. Box 123, 5140 AC, Waalwijk, The Netherlands.

([dagger]) Author to whom correspondence should be addressed. Email: [e-mail].

Table 1 -- Specifications of Polymer Emulsions

Fluorinated   wt% F               Particle Size  Viscosity
Polymer (FP)  (/emulsion)         (nm)           (mPa.s.)   pH

 1            0                    95               8       6.64
 2            0.25                 97               8       6.60
 3            0.76                 95               8       6.53
 4            1.02                101               8       6.58
 5            2.03                158               8       6.39
 6            5.09                697 (a)        2220       7.23

              wt% OH (/emulsion)

 7            0                    55               9       7.10
 8            0.44                 54               9       6.86
 9            0.87                 59               8       6.59
 4            1.31                101               8       6.58
10            2.18                222              13       6.52

(a) Large particle size due to flocculation.

Table 2 -- Formulations

                Clearcoat

Binder            71.70
Demi water         7.85
Zonyl FSO          0.05     Wetting agent
Nuvis FX 1070      1.10     Thickener
Demi water         1.10
Drew 210-693       0.70     Defoamer
Bayhydur 3100     17.50     Crosslinker

                 White Pigmented

Binder                 6.70
Demi water             1.90
Disperse-Ayd W-33      1.40     Dispersing agent
Surfynol 104E          1.40     Wetting agent
Nuvis FX 1070          0.50     Thickener
Tioxide R-HD2         24.00     Pigment
Binder                48.70
Zonyl FSO              0.05     Wetting agent
Surfynol 104E          1.40     Wetting agent
Drew 210-693           0.50     Defoamer
Bayhydur 3100         13.45     Crosslinker

Table 3 -- Antigraffiti (AG) Results of the Fluorinated Coatings
(Clearcoat) (a,b)

                                   4 hr @ Ambient
              Drying               Conditions/
              Conditions        16 hr @ 50[degrees]C
Fluorinated   wt% F
Polymer (FP)  (/emulsion)  IPA          MEK             DEK

 1            0            17           21              24
 2            0.25         18           26              29
 3            0.76         30           24              15
 4            1.02         31           35              35
 5            2.03         26           28              33
 6            5.09         16           18              25

              wt% OH
              (/emulsion)

 7            0            Films removed by cleaning agents
 8            0.44         Films removed by cleaning agents
 9            0.87         25           29              30
 4            1.31         31           35              35
10            2.18         28           29              31

Fluorinated            1 night @ rt             7 days @ rt
Polymer (FP)      IPA       MEK     DEK     IPA     MEK     DEK

 1                14        14      14      14      20      27
 2                15        15      16      14      20      28
 3                16        16      14      20      26
 4                17        21      24      14      22      31
 5                15        14      16      15      20      27
 6                15        18      15      13      18      24

                wt% OH
                (/emulsion)

 7                        Films removed by cleaning agents
 8                        Films removed by cleaning agents
 9                 7 (c)    15 (c)  20 (c)  10 (c)  13 (c)  16 (c)
 4                17        21      24      14      22      31
10                 7        14      19      11      15      19

(a) The cumulative total of all seven colors are given. Max = 35.
(b) Antigraffiti ratings are [+ or -]1.
(c) Films affected by cleaning agent.

Table 4 -- Surface-Free Energy Data (a)

Fluorinated              [[gamma].sup.p]  [[gamma].sup.d]  [gamma]
Polymer      wt% F (b)   (mN/m)           (mN/m)           (mN/m)

1            0           19.7             37.5             57.2
2            0.25         7.5             37.0             54.5
3            0.76        14.8             36.5             51.3
4            1.02        13.1             36.5             49.6
5            2.03        12.8             35.9             48.7
6            5.09         5.0             26.9             31.9

(a) Measured at 22[degrees]C.
(b) On emulsion.

Table 5 -- Crosslink Density and Antigraffiti (AG) Results (a)

Fluorinated      wt% F            [[nu].sub.e] (* [10.sup.-2]
Polymer (FP)     (/emulsion)           mol/[cm.sup.3])         AG (b)

 1                  0                         2.81               24
 2                  0.25                      2.81               29
 3                  0.76                      3.44               33
 4                  1.02                      6.14               35
 5                  2.03                      3.53               33
 6                  5.09                      2.39               25

              wt% OH (/emulsion)

 7                  0                         0.27               --
 8                  0.44                      1.85               --
 9                  0.87                      3.08               30
 4                  1.31                      6.14               35
10                  2.18                      5.11               31

(a) Curing conditions: four hours at ambient conditions followed by 16
hr at 50[degrees]C.
(b) Graffiti removed with DEK.

Table 6 -- Effect of Curing Conditions on Crosslink Density and
Antigraffiti (AG) Properties of FP4

                               [[nu].sub.e] (* [10.sup.-2]
Curing Condition               mol/[cm.sup.3])               AG (a)

4 hr at ambient conditions/
  16 hr at 50[degrees]C        6.14                          35
1 night at room temperature    4.55                          24
7 days at room temperature     4.76                          31

(a) Graffiti was removed with DEK.

Table 7 -- Properties on Metal and Concrete Plates of a Commercial
Fluorinated Coating Based on the Fluorinated Polymers (Clearcoat) (a)

                       Metal (b)               Concrete (c)

Hardness (sec)                    196     Petrol             5
Gloss (20[degrees]/60[degrees])    80/90  Diesel             5
Dust free (min)                    10     Motor oil          5
Tack free (hr)                      4*    Skydrol            5
Petrol                              5     Dot 3              4
Diesel                              5     Spent oil          5
MEK (double rubs)                >200     MEK                5
H2SO4                               5     H2SO4              5
NaOH (10%)                          5     NaOH (10%)         5
                                          Ammonia (25%)      5
                                          DBP                5
Antigraffiti (d)                   35     Antigraffiti (d)  35

* 0 = poor; S = good.
(a) Formulated with an NCO:OH ratio of 1.5:1.
(b) Dried for four hours at ambient conditions followed by 16 hr at
50[degrees]C.
(c) Dried for one week at room temperature.
(d) Graffiti removed with DEK, maximum of 35.

Table 8 -- Antigraffiti and Gloss Values of Fluorinated Coatings Upon
Repeated Exposure (Clearcoats on Metal and Concrete)

                                 1 Cycle   3 Cycles  7 Cycles  10 Cycles

Metal
Gloss (20[degrees]/60[degrees])  75/90     --        --        60/85
Antigraffiti                     35        35        33        33
2K WB PU1                        24-31

Concrete
Antigraffiti                     35        35        35        35 (a)
2K WB PU2                        29        30        25        22

(a) After recovery. The results are based on an average of three
different test panels.

Table 9 -- Properties of FP4 on Concrete (Clearcoat)

Hot tire
1 hr                             5
16 hr                            5
Weiss Global test
After 10 cycles                  OK
After 50 cycles                  OK
Freeze/thaw
Salt water/water (20 cycles)     OK
QCT
500 hrs                          OK
1000 hrs                         OK
Adhesion (kg/[cm.sup.2])
Concrete                         20
Transparent epoxy                32
Pigmented epoxy                  24