Inorganic/organic nanocomposite coatings: the next step in coating performance.

Inorganic/organic hybrid materials have considerable promise and are beginning to become a major area of research for many coating usages, including abrasion and corrosion resistance. Our primary approach is to prepare the inorganic phase in situ within the film formation process of the organic phase. The inorganic phase is introduced via sol-gel chemistry into a thermosetting organic phase. By this method, the size, periodicity, spatial positioning, and density of the inorganic phase can be controlled. An important aspect of the inorganic/organic hybrid materials is the coupling agent. The initial task of the coupling agent is to provide uniform mixing of the oligomeric organic phase with the sol-gel precursors, which are otherwise immiscible. UV-curable inorganic/organic hybrid systems have the advantages of a rapid cure and the ability to be used on heat sensitive substrates such as molded plastics. Also, it is possible to have better control of the growth of the inorganic phase using UV curing. It is our ultimate goal to completely separate the curing of inorganic and organic phases to gain complete control over the morphology, and hence optimization of "all" the coating properties. Thus far, it has been found that concomitant UV curing of the inorganic and organic phases using titanium sol-gel precursors afforded nanocomposite coatings which completely block the substrate from UV light while maintaining a transparent to visible light. Also, it has been found that the morphology of the inorganic phase is highly dependent on the concentration and reactivity of the coupling agent.

Keywords: Polyesters, polyurethanes, oils, UV, EB, radiation, aerospace, aircraft, automotive--OEM, automotive--refinish, industrial maintenance, urethane, aluminum, galvanized steel, alloy coated steel

**********

Inorganic/organic hybrid materials have been extensively studied. These hybrid materials provide unique properties such as improved physical, mechanical, and thermal properties, as well as gas-barrier and photonic properties. (1-4) The sol-gel technique of alkoxysilanes is one of the useful methods available for preparing inorganic/organic hybrid materials. The advantage of the sol-gel technique is that the reaction proceeds at ambient temperature to form ceramic materials, as compared to the traditional methods performed at high temperature. It is well known that when the inorganic component sizes reach nanometer scale, the hybrid materials often do not comply with estimates from the mixture because of interfacial interactions. Recently, there have been various approaches to constructing hybrid materials: e.g., Laine et al. and Choi et al. prepared an epoxy functional octahedral silsesquioxane reaction with diaminodiphenylmethane and formed a new inorganic/organic epoxy nanocomposites. (4-6) Matejka et al. reported modeling of inorganic/organic hybrid interpenetrating networks (IPN) based on epoxide-amine and a sol-gel precursor from tetraethoxysilane (TEOS). (2)

Soucek and coworkers have studied polyurea and polyurethane inorganic/organic films using different sol-gel precursors such as organofunctional alkoxysilanes or non-functional organoalkoxysilane. (7-10) The ceramer films exhibited enhanced adhesion and corrosion resistance properties via a self-assembly phase separation mechanism. In addition, the new inorganic/organic hybrid materials based on the plant oils were also investigated. The hybrid films showed the higher tensile strength and modulus, but the films were formed using high temperature cure. (11,12)

UV-curable thermosetting systems also have the advantages of rapid cure, low energy consumption, and low volatile organic contents (VOCs). This technology provides a number of applications for coatings, printing inks, adhesives, and dental relining. (13,14) The cationically photoinitiated polymerization has a longer living propagating species, which is referred to as "dark-cure." Epoxides, especially aliphatic epoxides, are one of the major materials for cationic initiated UV-curing technology. In our previous work, (15) a series of epoxynorbornene linseed oils (ENLO) was synthesized. The chemical structures were characterized using FTIR, nuclear magnetic resonance (NMR) ([.sup.1]H, [.sup.13]C-NMR), and matrix-assisted laser desorption/ionization time of flight (MALDI-TOF) mass spectra. The photoinitiated cationic polymerization and kinetic study of ENLO were also reported.

UV-radiation curing of coatings offers the benefit of ambient curing reactions, which is advantageous for temperature-sensitive substrates such as poly(vinyl chloride) and polycarbonate sheets. These plastic substrates require a protective barrier to curtail abrasion and sometimes to protect against photooxidation. For this reason, the current technology requires the addition of substantial quantities of a UV absorber for sufficient protection. However, competition for available photons between the photoinitiator and UV absorber limits the ability to capitalize on the benefits of UV cure.

In the present study, a UV-curable inorganic/organic hybrid film based on epoxidized linseed oil and ENLO were prepared using the in-situ sol-gel method. The morphology and phase structures of these hybrid films were characterized using atomic force microscopy (AFM), transmission electron microscopy (TEM), and small angle light scattering (SALS). Molecular structures between the inorganic and organic regions were probed using [.sup.29]Si solid-state NMR spectroscopy. For the epoxidized linseed oil/titanium system, the extent of cure and UV screening were investigated. In addition, physical properties were evaluated by dynamic mechanical thermal analysis, small angle X-ray scattering, and hardness.

EXPERIMENTAL

Materials

Tetraethylorthosilane (TEOS), hydrochloric acid (37%), and ethanol (100%) were purchased from Aldrich. Epoxynorbornene linseed oils (ENLO) were prepared as previously reported. (15,16) The photoinitiators, (4-octyloxyphenyl) phenyl iodonium hexafluoroantimonate (OPPI), were obtained from UCB. Titanium (IV) isopropoxide (TIP) was purchased from Aldrich. All chemicals were used as received except for the epoxidized linseed oil. The air-sensitive chemicals were weighed and transferred in a dry box under argon atmosphere. Epoxidized linseed oil (Flexol[TM] plasticizer LOE) and mixed triarylsulfonium hexafluoroantimonate salt (UVI-6974) photoinitiator were supplied by Dow Chemical Company. Characterized LOE contains the following: unsaturation = 2.4 iodine number; acid number = 0.2 mg/g; hydroxyl value = 9.0 mg/g; and total epoxide = 9.6 wt%. LOE was purified by azeotroping with anhydrous methanol under vacuum to 0.1 mm Hg while increasing the temperature from 19[degrees]C to 100[degrees]C. Titanium dioxide (MT-500-SA, 41 nm) was donated by Daicolor-Pope, Inc.

Instruments

[.sup.1]H-NMR spectra were recorded on a Gemini-300 spectrometer (Varian) in CD[Cl.sub.3] solvent with tetramethylsilane (TMS) as references. The [.sup.29]Si-NMR spectra were recorded on a Gemini-400 spectrometer (Varian) in CD[Cl.sub.3] solvent with TMS as references in a 10-mm NMR tube. Solid-state [.sup.29]Si NMR spectra were obtained at 4.7 T using a Varian Inova 200 MHz ([.sup.1]H frequency) spectrometer with a Doty Scientific supersonic MAS probe. Solid-state [.sup.13]C NMR spectra were obtained at 17.6 T using a Varian Inova 750 MHz ([.sup.1]H frequency) spectrometer with a Doty Scientific HXY probe. AFM was performed using a multimode scanning microscope (Digital Instrument Nanoscope III a) in a tapping mode with silicon tip. TEM experiments were carried out on a JEOL 1200 EX II using an accelerating voltage of 120 kV. The thin slices for the observation of nanosized phase segregation were obtained using a Reichert Ultracut S Cryomicrotome at room temperature.

Small angle X-ray spectroscopy was performed using a Rigaku 12 kW rotating anode with a Cu source and point focus, Frank Mirrors, and a Siemens Hi-Star multiwire area detector. The UV spectroscopy was completed with a Varian CARY 500 scan UV-Vis-NIR spectrophotometer directly on films removed from the substrate. These films ranged in thickness from 65 [micro]m to 90 [micro]m. The spectrum was measured from 280 nm to 400 nm.

Preparation of TEOS Oligomers

Tetraethoxysilane (0.24 mol) and ethanol (0.96 mol) were charged into a four-necked flask equipped with condenser, stirrer, and [N.sub.2] inlet. Distilled water (0.24 mol) was added to the mixture. Hydrochloric acid (0.99 mL, 37 wt%) dissolved in 10 mL of ethanol was added dropwise for one hour. The mixture was mechanically stirred for 24 hr at ambient temperature. The ethanol in the mixture was removed in vacuum (0.2 mm Hg) at 50[degrees]C to afford TEOS oligomers (27.9 g). The products were characterized using [.sup.1]H, [.sup.29]Si NMR, and MALDI-TOF-MS, as previously described.

Preparation of LOE/Titanium Isoproxide (TIP) Hybrid Coating or Ti[O.sub.2] Formulations

Potential reactive sites on the glassware were reduced by rinsing with a 4% dimethyldimethoxysilane, 6% water, 90% ethanol solution with the pH adjusted to 5 by the addition of acetic acid. Resins were mixed under argon in opaque 20-mL vials to prevent exposure to water and UV light. The concentration of TIP was varied from zero to 25% by weight relative to LOE. Photoinitiator was added to the resin, comprising 3% of the total solution. Under a nitrogen blanket, the resins were cast using a 50 micron (2 mil) drawdown bar on aluminum panels and immediately placed in an ultraviolet processor (RPC, 300 W/in medium-pressure mercury UV lamp at 16 fpm) and evaluated for dryness to touch. Ti[O.sub.2] was mixed with 2 wt% of LOE and dispersed using a high speed impeller disk combined with glass beads in the resin. To aid the dispersion process, 1% BYK-Chemie Antitera-207 was added. Following UV cure, the panels were stored for two weeks in a sealed humidity-controlled chamber held constant between 92% and 98% relative humidity via a saturated solution of aqueous potassium sulfate. The chamber was vented with fresh air for one hour every three days to remove the isopropanol byproducts.

Preparation of ENLO TEOS Hybrid Film

The UV-curable coatings were formulated using ENLO, photoinitiator, and TEOS oligomers. The formulations are shown in Tables 1-3. The formulations are numerically named to reflect the percentage of norbornylization and amount of TEOS oligomers, for example: ENLO-100-1 represents the ENLO with the maximum amount of norbornylization and 1 wt% of TEOS oligomers. The mixture was mixed on a roller mill for 24 hr. The aluminum panels were degreased with acetone before casting a formulation. A thin film of formulation was cast on aluminum panels (7.6 cm x 15 cm from Q-Panel) and glass plates using a drawdown bar #3. The wet thin film was cured using a Fusion UV-System processor (P300) with an H-bulb. The belt speed was 3.05 m/min.

Sample Preparation for AFM and TEM

The surfaces of the cured hybrid films were observed directly. For the cross-section observation, a piece of the same hybrid film was removed from the substrate. The clear epoxy resin and cure agent were filled in a flat embedding mold #70900 with cavities measuring 14 mm (L) x 5 mm (W) x 4 mm (D) (Electron Microscopy Science, Inc.). The hybrid film was placed in the middle and allowed to cure at room temperature. After being fully cured, the sample was cut using microtome with a diamond knife. The tapping mode was used to obtain both height and phase images and a 100-[micro]m scanner was selected. The force used by the cantilever was light enough to limit damage to the sample. The scanning rate was controlled to be 1-2 Hz for the low magnification images and the data were collected in 512 x 512 pixels per images.

Small Angle Light Scattering

Scattering patterns of hybrid films were obtained using SALS with a 2-mW He-Ne laser (632.8 nm), developed by Uniphase, and a beam size of 1-mm polarized monochromatic light. Middle region of C-Cut samples were measured, since the structural gradient on this plane was minimum, thus eliminating multiple scattering problems. The hybrid film scattered a linearly polarized light and the scattering was examined after the light passed through a polarized analyzer placed with its plane normal to the incident beam. Hv patterns were obtained when the analyzer was perpendicular to the polarizer. Vv patterns were obtained when the polarizer in the incident beam and the analyzer in the scattered beam were both vertical. The patterns were captured using an instant Polaroid camera and 16-bit CCD camera software (photometrics-cooled CCD) operated by IPLab spectrum.

Since there was no deformation for the inorganic particles, the equivalent radius ([R.sub.0]) of the spherulites was estimated using equation (1):

[R.sub.0] = [lambda] / [[pi] sin([[theta].sub.max] / 2)] (1)

Solid-State NMR Spectroscopy

All [.sup.13]C spectra were acquired using a RAMP CP/MASi (cross polarization with magic angle spinning) experiment with a MAS speed of 15 kHz, 3-sec relaxation delay, 1-ms contact time, 4.7-[micro]s 90[degrees] pulse width, and a 80-kHz decoupling field. Samples were packed into 4-mm silicon nitride rotors with aurum end caps. The [.sup.13]C chemical shifts were corrected by using hexamethylbenzene ([[delta].sub.Me] = 17.3 ppm) as an external reference. The [.sup.29]Si spectra were acquired using the standard CP/MAS experiment with a MAS speed of 5-kHz, 2-sec relaxation delay, 5-ms contact time, 5.6-[micro]s 90[degrees] pulse width, and a 55-kHz decoupling field. Samples were packed into 7-mm silicon nitride rotors with Kel-F end caps. The [.sup.29]Si chemical shifts were corrected by using tetramethylsilane (TMS [[delta].sub.Si] = 0 ppm) as an external reference.

RESULTS AND DISCUSSION

The original intent of this study was twofold: (1) demonstrate the concept of UV-curable inorganic/organic hybrid materials using modified seed oils, and (2) probe the structure of the inorganic phase under the rapid vitrification of the organic phase. Since the majority of the inorganic/organic hybrid work in our laboratory has been thermally cured systems, (7-12,15) it was proposed that UV-curing of the organic phase would result in smaller inorganic domains. This was based on the precept that an almost instantaneous cure speed of the organic phase would preclude sol-gel oligomers mobility. For a similar organic system with a titanium sol-gel precursor, it was shown that very small nano-domains of titanium oxides were formed. (17)

[FIGURE 1 OMITTED]

[FIGURE 2 OMITTED]

ENLO INORGANIC/ORGANIC COATINGS

The seed oils represent a biomass alternative technology which is in need of modification to be competitive with modern polymeric binders. The epoxynorbornene linseed oils (15,16) are all mixtures of oxiranes: fused norbornylene oxides and epoxidized linseed oil, as shown in Figure 1. The four modified linseed oils used in this study (ENLO, ENLO-25, ENLO-50, and ENLO-100) were chosen to elucidate the effect of the ratio of epoxide type to the film properties and material morphology. The cationic UV-curable system was chosen to concomitantly initiate chain reactions in both the inorganic and organic phases.

LOE/TITANIUM INORGANIC/ORGANIC COATINGS

The first attempts at UV cure with the titanium sol-gel precursor led to failure. No curing or incomplete curing were a consequence of premature sol-gel reaction with water. It was found that even a minimal of metal-oxo-bond formation, as shown in Scheme 1, competitively absorbed the UV light, thus preventing cure of the organic phase. Previously, Soucek et al. showed TIP to have a strong absorbance between 325 and 425 nm in n-hexane. (17,18) Although polarity of the medium is known to shift the range of absorbance, it was presumed that the titanium-oxo-clusters blocked the UV light at the wavelength needed for the UV photoinitiator to initiate polymerization of the organic phase. Upon introduction of water this absorbance band was broadened, presumably due to the rapid hydrolysis and condensation forming titanium-oxo-clusters and precipitated aggregates. The precursor, TIP, did not saturate the breadth of the UV spectrum, allowing absorbance by initiator that showed strong absorbance ranging from 260 to 370 nm. Thus, a rigorously controlled atmosphere including both the entire UV-curing processor and very pure sol-gel reagent was necessary for film formation. After UV irradiation, the films were exposed to high relative humidity conditions. It has been previously reported that some water is needed to facilitate proton transport for cationic photoinitiated curing of epoxide oligomers into films. (19)

[GRAPHIC OMITTED]

After UV curing, the sol-gel based ceramer film was optically transparent. These films had minimal roughness due to surface shrinkage that is commonly associated with UV-absorbing pigments. In contrast, the films based on thermally cured nanosized Ti[O.sub.2] particles were turbid and hazy. As previously stated, it is only with great difficulty that the nanosized titanium dioxide is adequately dispersed. Agglomeration of the particles yield the hazy appearance. Thermal cure of Ti[O.sub.2] films was necessary as attempts to UV cure led to surface shrinkage and incomplete through-cure.

The films were evaluated by small-angle x-ray scattering (SAXS) and showed no scattering other than for the organic phase. Although scattering was previously reported for the thermally cured titanium based ceramer films, this was not the case for the UV-cured ceramer films. It was assumed that the inorganic domain was below the detection limit of ~2 nm. The small inorganic domain was attributed to the speed of the cure of the organic phase, coupled with a lack of fugitive water for hydrolysis until after gelation of the organic phase. The UV-absorbance spectra of the cured ceramer films as a function of sol-gel precursor concentration is shown in Figure 2. The films appear to be dependant on concentration of TIP, and are comparable in absorbance to the films cured with titanium dioxide nanoparticles. The ELO film had a strong absorbance at 285 nm that could be attributed to carbonyl chromophores in the respective medium. Upon the addition of 5 wt% TIP, a second narrow peak appeared near 300 nm. When the percentage of TIP was increased from 5 to 10 wt% the peak widened, absorbing strongly between 280 and 360 nm. The broadening peak could be attributed to the sum of individual auxochromic shifts associated with a variety of structures formed during the organic titanium oxide oligomerization. At higher concentrations the same broad peak was observed with a proportional increase of intensity. Between 5 and 10 wt% the transition from a narrow absorbing to a broad absorbing peak was associated with the hydroxyl concentration of the ELO. At 5 wt% there was insufficient TIP to overcome the hydroxyl concentration resulting in the transesterification of TIP to form LOE-TIP bonds. At greater than 5 wt%, Ti-O-Ti bonds could form resulting in auxochromic shifts.

[FIGURE 3 OMITTED]

This research provides a method to form nanosized titanium-oxo-clusters with cluster sizes < 2 nm without measurable agglomeration. This is the first time, to the best of our knowledge, that a titanium sol-gel precursor has been photo-cured to form a UV-blocking/absorbing ceramer film. The co-continuous inorganic/organic network maintained physical permanence, improved toughness, and optical clarity (visible light). Amorphous ceramer composites that are transparent to visible light and block UV light, as depicted in Figure 3 and shown over a penny in Figure 4, can exhibit an array of applications such as a UV filter for optics. Also, control of titaniumoxo-cluster size can lead to exploration of the edge of Ti[O.sub.2] photonic properties.

Microscopy, Spectroscopy, Morphology, and Models of Structure

The morphology of the hybrid coating surface was observed using AFM and TEM, as shown in Figures 5 and 6, respectively. The cured neat ENLO films were uniform films (Figure 5a), while the bright spots dispersed in the cured hybrid coating films were observed (Figure 5b). The bright spots are silica which could be estimated to have an average size of 38.2 nm. In order to clarify the sizes and their distribution of the siloxane clusters in three-dimensional scales, the cross-section of the hybrid film was observed (Figure 5c). The AFM imaging indicated that the hybrid coating films had microphase separation and the silica gels were highly dispersed at the nanoscale level within the organic polymer phase. The TEM of the hybrid film with 20 wt% TEOS oligomers was observed. The dark aggregates with various shapes were observed over the entire area of the hybrid films. The results indicated that silica gels have excellent dispersion in ENLO films on a nanoscale. On the other hand, larger size silica gels, e.g., 200 nm, were also observed. More importantly, a phase percolation between silica gels and ENLO was observed. This result suggested that a "tether" structure existed and may explain the increased crosslink density in these hybrid films.

[FIGURE 4 OMITTED]

The Hv and Vv scattering results from SALS are shown in Figures 7a and 7b, respectively. A two-fold symmetry Vv scattering and four-fold symmetry Hv scattering pattern were observed. These results indicated that a heterogeneous structure existed in these hybrid films. As corresponded to morphology by AFM and TEM, the silica gels were distributed throughout the hybrid films with aggregation sizes ranging from nanometer to microscale. The size of the silica gels at microscale could be estimated using equation (1). The particle sizes were from 0.59 to 2.18 microns. These results were in agreement with the AFM and TEM data.

[FIGURE 5 OMITTED]

The [.sup.29]Si solid-state NMR spectra can be used to probe the substitution of the ROSi(OR)[.sub.4-n] units. The spectra of hybrid films resulting from in-situ TEOS oligomers and UV-curable ENLO are shown in Figure 8. Compared with the [.sup.29]Si-NMR spectra for liquid samples, there are no four-substituted ([Q.sub.4]) structures for pure TEOS oligomers in the solid state. The spectrum of TEOS oligomers displays signals corresponding to [delta] -89.045 ppm ([Q.sub.1]), -95.435 ppm, -96.348 ppm ([Q.sub.2]), and -101.643 ppm, -102.80 ppm ([Q.sub.3]), which are agreeable with the signals of solution of TEOS oligomers. However, the resonances around [delta] -110 ppm corresponding to [Q.sub.4] appeared for the hybrid films. (25) The results indicated [Q.sub.1], [Q.sub.2], or [Q.sub.3] signals converted to [Q.sub.4] signals after the reactive groups such as [equivalent to] Si-OEt and [equivalent to] Si-OH distributed reacted with organic components in ring-opening polymerization; which suggested a formation of "tether" structure of organic regions and nanostructured siloxane networks.

[FIGURE 6 OMITTED]

The structures of inorganic/organic hybrids involve either a prereacted inorganic phase or a simultaneous polymerization of both inorganic and organic reactive groups. Relative to TEOS oligomers gel-process, UV-curing polymerization of ENLO is much faster. The simple structure of the hybrid films is that the crosslinked TEOS oligomers with various sizes are embedded in the organic (polyether) matrix. Since these are acid-catalyzed sol-gel reactions, there is a large fraction of [equivalent to] Si-OEt and [equivalent to] Si-OH reactive groups distributed throughout these inorganic particles. The ring-opening polymerization of epoxynorbornene linseed oils proceeds efficiently in the presence of a photogenerated protonic acid to form a polyether. It is proposed that the hydroxyl groups generated from the ring-opening polymerization of epoxy groups reacts with either the ethoxyl or silanols groups of the polysilicate network to form a "tethered" structure between the organic and inorganic phase.

A schematic model for the UV-curable hybrid films is proposed and shown in Figure 9. The scattered small dark regions represent the silicate-rich domain and the connecting lines represent the organic crosslinked polymers formed by the photopolymerization of ENLO. A similar model of such hybrid materials has been proposed by Wilkes. (20) Their model structure described the highly condensed silica "cluster" connecting with organic-rich matrix (tether structure), which may not truly illustrate the hybrid materials' structures. Since the UV-curing process is a very fast ring-opening polymerization, not all of the reactive groups such as ethoxyl and silanols would react with hydroxyl groups. Therefore, except for the "tethered" structure, the silica gel formed by highly self-condensed TEOS oligomers embedded into the polymer matrix (ENLO). The results from AFM, TEM, and SALS data supports this model of a moderately dense inorganic structure. However, other previously reported models include a more open structure with more intimate contact between the two phases.

[FIGURE 7 OMITTED]

The interactions between inorganic phase and organic phase, especially the "tether" structure, improve the compatibility of the silica gel and ENLO network and would result in the production of a higher crosslink density as compared to the cured ENLO film. Higher crosslink density should translate to higher hardness and modulus and lower elongation. The higher modulus of the hybrid films can be explained by poor flexibility and low fracture toughness. The results from tensile properties, fracture toughness, and DMTA indicated that hybrid films that have a smaller amount of TEOS oligomers (<6 wt%) would result in lower modulus; while the higher content of TEOS oligomers (>20%) would make the hybrid materials have higher modulus, but be brittle. The UV-curable hybrid coatings possess higher hardness, modulus, and thermal stability than the cured ENLO films.

The structure of the inorganic phase was unusual when compared to the previously reported thermally cured inorganic/organic hybrid systems. (7-12) For this system, a column or iceberg structure for the inorganic phase was observed. This could be attributed to the photoinitiator absorbing light on the surface of the film and the top layers of the film and curing first. As the photoinitiator was depleted near the surface, the lower layers were cured in a stratified fashion resulting in columns of inorganic phase perpendicular to the film substrate interface.

We have reported a number of different models and structures for inorganic/organic hybrid materials for different systems. For a thermally cured system with a relatively long ramped curing schedule (1 hr at 130, 180, and 210[degrees]C), and at a high cure temperature, the structure of the inorganic phase consisted of large diffuse regions that were almost blister-like. (21) The inorganic phase was not particulate with a discrete phase boundary, instead a large amount of organic phase was interspersed within the inorganic phase. Consequently, an open interpenetrating inorganic/organic hybrid model (22) was proposed for long reaction times and highly crosslinked organic matrix with potentially a high molar ratio of coupling groups. (23) For polyurethane-based inorganic/organic hybrid thermally cured systems where the inorganic phase is coupling agent starved, the inorganic phase forms particulate silicon-oxo-clusters with particle size being dependent on the [T.sub.g] of the polyol segment of the polyurethane, as shown in Figure 10. (24) The diffuse morphology is depicted in Figure 10a, and a particle-like morphology is depicted in Figure 10b. Finally, the structure of the UV curing based on titanium sol-gel precursors in a seed oil-based epoxy organic phase was observed to be ~2-5 nm. It was postulated that the size of the inorganic phase was controlled by the speed of the UV curing, hence the rapid vitrification of the organic phase.

The inorganic/organic system reported herein does not possess an aggressive coupling agent as a functional group (secondary hydroxyl), such as carboxyl or alkoxylsilane groups, even though there certainly is a large excess in stoichiometry favoring the coupling group. Also, the TEOS oligomers are considerably less reactive than the titanium-based sol-gel precursors, thus the relative rate of reactivity of the inorganic phase is not the controlling factor in this case. Therefore, the inorganic domain size and structure is not fully dependent on the relative reaction kinetics of the inorganic versus organic phases. The accumulated data seems to point to the concentration and aggressiveness (reactivity) of the coupling agent as the primary factor for determining the structure and size of the inorganic domains.

[FIGURE 8 OMITTED]

The titanium-based inorganic/organic hybrid coatings demonstrate the following two concepts. First of all, the rapid cure of the organic phase via UV-initiation coupled with the fast reaction kinetics of the titanium tetraisopropoxide facilitates a nano-inorganic domain. Secondly, the titanium sol-gel precursors can produce in situ carboxylic acid coupling agents, which the TEOS oligomers cannot. The titanium sol-gel precursors are transesterification catalysts which react with the ester groups of the triglycerides to form oligomeric carboxylic acids. These acids are very aggressive coupling agents and serve to tether the titanium sol-gel precursor as hydrolysis occurs, limiting the inorganic domain to a few nanometers. In addition, the in situ formation of coupling agent serves to isotropically distribute the titanium-oxo-clusters in a way that is not possible for nanosized particulate Ti[O.sub.2].

The coatings industry is slow to change and new technologies seem to be even slower to develop and be accepted. Indeed, if there were not environmental regulations, one would suspect that the coating technology of the 1970s would be more dominant than it already is. Inorganic/organic hybrid coatings have great promise for the future, whether it is in the area of stimuli response coatings or something as important as a one-step corrosion resistant primer/metal pretreatment for steel or aluminum. Hybrid coatings are already used as abrasion, mar, and chemical resistant coatings in the eyeglass industry. The ability to understand and, hence, control the reactive processes which generate the morphology of these hybrid films is key to generating coating systems that are industrially viable.

[FIGURE 9 OMITTED]

[FIGURE 10 OMITTED]

CONCLUSIONS

A series of UV-curable hybrid films based on epoxidized and epoxynorbornene linseed oils (LOE and ENLO) were prepared using the in situ method of sol-gel precusors. The hybrid films had nanophase separation structures and the gels uniformly dispersed in the organic phase at nanoscale level. The coupling agent was an important factor in determining the inorganic phase structure. The relative reactivity of the coupling agent as well as the concentration were both salient factors in determining the periodicity and density of the inorganic phase. The iceberg or column-like inorganic structures were attributed to stratification caused by a stratification of concomitant cure of both the inorganic and organic phases from the surface of the film downward. For a uniform nano-inorganic domain, both the inorganic and organic phases must have fast curing kinetics.

ACKNOWLEDGMENTS

Financial support for this research was provided by a national research initiative competitive grant (USDA-NRI-2003-35504-12950). We would like to thank Ms. Kathleen Marie Wollyung, Department of Chemistry, for MALDI-TOF mass spectra measurements and Prof. Stephen D. Cheng, Department of Polymer Science, for allowing us to use AFM.

References

(1) Alexandre, M. and Dubois, P., Mater. Sci. Eng., 28, 1 (2000).

(2) Matejka, L., Dukh, O., and Kolarik, J., Polymer, 41, 1449 (2000).

(3) Yano, K., Usuki, A., and Okada, A., J. Polym. Sci. Part A: Polym. Chem., 35, 2289 (1997).

(4) Laine, R.M., Choi, J., and Lee, I., Adv. Mater., 13, 800 (2001).

(5) Choi, J., Harcup, J., Yee, A.F., Zhu, Q., and Laine, R.M., J. Am. Chem. Soc., 123, 11420 (2001).

(6) Choi, J., Yee, A.F., Lee, I., and Laine, R.M., Macromolecules, 36, 5666 (2003).

(7) Ni, H., Aaserud, D.J., Simonsick, W.J. Jr., and Soucek, M.D., Polymer, 41, 57-71 (2000).

(8) Ni, H., Simonsick, W.J. Jr., Skaja, A.D., Sailer, R.A., Williams, J.P., and Soucek, M.D., Prog. Org. Coat., 38, 97-110 (2000).

(9) Ni, H., Skaja, A.D., Sailer, R.A., and Soucek, M.D., Macromol. Chem. Phys., 201(6), 722-732 (2000).

(10) Wu, S., Sears, M.T., Soucek, M.D., and Simonsick, W.J. Jr., Polymer, 40, 5675-5686 (1999).

(11) Tuman, S.J. and Soucek, M.D. "Novel Inorganic/Organic Coatings Based on Linseed Oil and Sunflower Oil with Sol-Gel Precursors," J. COAT. TECHNOL., 68, No. 854, 73 (996).

(12) Teng, G., Wegner, J.R., Hurtt, G.J., and Soucek, M.D., Prog. Org. Coat., 42, 29-37 (2001).

(13) Fouassier, J.P., Photoinitiation, Photopolymerization and Photocuring, Fundamentals and Applications, Hanser Publications, pp. 246-365, 1995.

(14) Pappas, S.P. (Ed.), UV Curing: Science and Technology, Vol. 4, Technology Marketing Corporation, 1978.

(15) Zong, Z., Soucek, M.D., Hu, J., and Liu, Y., J. Polym. Sci. Part A: Polym. Chem., 41, 3440 (2003).

(16) Chen, J., Soucek, M.D., Simonsick, W.J. Jr., and Celikay, R.W., "Preparation and Photopolymerization of Norbornyl Epoxidized Linseed Oil," Polymer, 40(20), 5379 (2002).

(17) Johnson, A.H. and Soucek, M.D., "Preparation of Nanosized UV-Absorbing Titanium-Oxo-Clusters via a Photo-Curing Process," published online, J. Adv. Poly. Tech. (2005).

(18) Ballard, R.L., Williams, J.P., Njus, J.M., Kiland, B.R., and Soucek, M.D., "Inorganic-Organic Hybrid Coatings with Mixed Metal Oxides," Eur. Polym. J., 37, 381 (2001).

(19) (a) Chen, J. and Soucek, M.D., "UV-Curing Kinetics of Cyclohexyl Epoxides with Real-Time Fourier Transform Infrared Spectroscopy," J. Appl. Poly. Sci., Vol. 90, 2485 (2003). (b) Chen, J. and Soucek, M.D., "Model for the Effects of Water on the Cationic UV-Curing of Cyclohexyl Epoxides," J. COAT. TECHNOL., 75, No. 937, 49 (2003).

(20) Huang, H.H. and Wilkes, G.L., Polymer Bulletin, 18, 455 (1987).

(21) Ballard, R.L., Tuman, S.J., Fouquette, D.J., Stegmiller, W., Soucek, M.D., "Effects of an Acid Catalyst on the Inorganic Domain of Inorganic-Organic Hybrid Materials," Chem. Mater., Vol. 11, 726 (1999).

(22) Deffar, D., Teng, G., and Soucek, M.D., "Comparison of Titanium-Oxo-Clusters Derived from Sol-Gel Precursors with Ti[O.sub.2] Nanoparticles in Drying Oil Based Ceramer Coatings," Macromol. Mater. Eng., 286, 204 (2001).

(23) Wold, C.R., Ni, H., and Soucek, M.D. "Model Reaction Study on the Interaction between the Inorganic and Organic Phases in Drying Oil Based Ceramer Coatings," Chem. Mater., 13(9), 3032 (2001).

(24) Ni, H. and Soucek, M.D., "Nanostructured Polyurethane Ceramer Coatings for Aircraft," J. COAT. TECHNOL., 74, No. 933, 125 (2002).

(25) Zhong, Z., Soucek, M.D., and Xue, C., "Unusual Inorganic Phase Formation in Ultraviolet-Curable Organic-Inorganic Hybrid Films," J. Poly. Sci., Part A., 43, 8, 1607 (2005).

M.D. Soucek, ([dagger]) Z. Zong, and A.J. Johnson -- The University of Akron*

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

* Department of Polymer Engineering, Akron, OH 44325.

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

Table 1 -- Formulations of ENLO-100 with Different TEOS Oligomers

                                   wt (%)
             ENLO-100  ENLO-100-1  ENLO-100-5  ENLO-100-10  ENLO-100-20

ENLO-100     96        95          91          86           76
TEOS          0         1           5          10           20
  oligomers
OPPI          4         4           4           4            4

Table 2 -- Formulations of ENLO-50 with Different TEOS Oligomers

                                    wt (%)
                ENLO-50  ENLO-50-2  ENLO-50-6  ENLO-50-10  ENLO-50-20

ENLO-50         96       94         90         86          76
TEOS oligomers   0        2          6         10          20
OPPI             4        4          4          4           4

Table 3 -- Formulations of ENLO-25 with Different TEOS Oligomers

                                    wt (%)
                ENLO-25  ENLO-25-1  ENLO-25-5  ENLO-25-10  ENLO-25-20

ENLO-25         96       95         91         86          76
TEOS oligomers   0        1          5         10          20
OPPI             4        4          4          4           4