Ind. Eng. Chem. Prod. Res. Dev. 1985, 24, 113-120
creases the free energy of the reaction while the membrane itself separates reactants from products. The knowledge accumulated in micellar and macromolecular catalysis (Fendler and Fendler, 1975) could be profitably transferred to membrane reactor applications. Appropriately constructed polymeric membranes can also find important applications in controlled release technologies and target direct drug deliveries. Within the foreseeable future, we shall witness the emergence of major new technologies based on polymeric membranes.
113
Gregoriadis, G. "Drug Carriers in Biology and Medicine"; Academic Press: London. 1979. Hupfer, B:; Ringsdorf, H.; Schupp, H. Makromol. Chem. 1981, 182, 247-253. Ishiwatari, T.; Fendler, J. J. J. Am. Chem. SOC. 1984, 106, 1908-1912. Johnston, D. S.;McLean, L. R.; Whittam, M. A,; Clark, A. D.; Chapman, D. Biochemistry 1983, 2 2 , 3194-3202. Kavanau, L. "Structure and Function in Biological Membranes"; Holden-Day: San Francisco, 1964. Kesting, R. E. "Synthetic Polymeric Membranes"; McGraw-Hill: New York, 1971. Kimelberg, H. K.; Mayhew, E. G. CRC Crlt. Rev. Toxlcol. 1978, 6, 25-79. Kippenberger, D. J.; Rosenquist, K.; Odberg, L.; Tundo, P.; Fendler. J. H. J. Am. Chem. SOC. 1983, 105, 1129-1135. Koch, H.; Ringsdorf, H. Makromol. Chem. 1981, 182, 255-262. Kuhn, H.; Mobius, D.; Bucher, H. I n "Physical Methods for Chemistry", Vol. 1, Part I11 B, Weissberger, A.; Rossiter, B. W.. Ed.; Wiley-Interscience: New York, 1972; pp 577-701. Kunitake, T.; Sakamoto, T. J. Am. Chem. SOC. 1978, 100, 4615-4617. Kunitake, T.; Okahata, Y.; Audo, R.; Shinka, S.; Hirakawa, S. J. Am. Chem. SOC. 1980, 102, 7877-81. Kunitake, T.; Ihara, H.; Okahata, Y. J. Am. Chem. SOC. 1983, 105, 6070-6078. Kurihara, K.; Fendler, J. H. J. Am. Chem. SOC. 1983, 105, 6152-6153. Lieser. G.; Tieke, B.; Wegner, G. Thin Solid Films 1980, 68, 77-85. Loeb, S.; Sourirajan, S.;UCLA Report, 1960, p 60. Maoz, R.; Sagiv, J. J. Colloid Interface Sci. 1984, in press. Mobius, D. Acc. lChem. Res. 1981, 14, 63-68. Moss, R. A.; Bizzigotti, G. 0. Tetrahedron Lett. 1982, 2 3 , 5235-5238. Moss. R. A.; Schreck, R. P. J. Am. Chem. SOC. 1983, 105, 6767-6768. Moss, R. A.; Shin, J. J. J. Chem. SOC., Chem. Commun. 1983, 1027-1028. Moss, R. A.; Ihara, Y.; Bizzigotti, G. 0. J. Am. Chem. SOC. 1982, 104, 7476-7478. Naegele, D.; Lando, J. B.; Ringsdorf, H. Macromolecules 1977, 10, 1339-1344. Netzer, L.; Iscovici, R.; Sagiv, J. Thin Solid Films 1983, 9 9 , 235-241. Netzer, L.; Sagiv, J. J. Am. Chem. SOC. 1983, 105, 674-676. Porter, G. Proc. R . SOC.London Ser. A . 1978, 362, 281-303. Reed, W.; Guterman, L.; Tundo, P.; Fendler, J. H. J. Am. Chem. SOC.1984, in press. Regen, S. L.;Singh, A.; Oehme, G.; Singh, M. J. Am. Chem. SOC. 1982, 104, 791-795. Singer, S.J.; Nicholson, G. L. Science 1972, 175, 720-731. Tieke, B.; Wegner, G. Makromol. Chem. 1978, 179, 1639-1642. Tundo, P.; Kippenberger, D. J.; Politi, M. J.; Klahn, P.; Fendler, J. H. J. Am. Chem. SOC. 1982, 104, 5352-5358. Tundo, P.; Kurihara, K.; Kippenberger, D. J.; Politi, M.; Fendler, J. H. Angew. Chem., Int. Ed. Engl. 1982, 2 1 , 81-82. Tyrrell, D. A., Heath, T. D., Colley, C. M.; Ryman, B. E. Blochim. Blophys. Acta 1978, 457, 259-302. Whitten, D. G. Angew. Chem., I n t . Ed. Engl. 1979, 18, 440-405.
Acknowledgment Support of this work by the National Science Foundation is gratefully acknowledged.
Literature Cited Ackerman, R.; Inacher, 0.; Ringsdorf, H. Kollold Z.Z . Polym. 1971, 249, 1118- 1126. Akimoto, A.; Dorn, K.; Gros, L.; Ringsdorf, H.; Schupp, H. Angew. Chem. Eng. 1981, 2 0 , 90-91. Aibrecht, 0.; Johnston, D. S.; Vilhverde, C.; Chapman, D. Biochim. Biophys. Acta 1982, 687, 165-169. Barrand, A,; Rosilio, C.; Raudel-Teixier, N. Thin Solid Films 1980, 68, 7-12, 91-98, 99-106. Bubeck, C.; Tieke, B.; Wegner, G. Ber. Busenges. Phys. Chem. 1982, 8 6 , 495-498. Calvin, M. Acc. Chem. Res. 1978, 1 1 , 369-374. Day, D.; Ringsdorf, H. J. Polym. Sci. Polym. Lett. 1978, 16, 205-210. Day, D.; Lando, J. B. Makromolecules 1980, 13, 1478-1483, 1483-1487. Day, D.; Lando, J. B. J. Appl. Polym. Scl. 1981, 2 6 , 1605-1612. Day, D.; Ringsdorf, H. Macromol. Chem. 1979, 180, 1059-1063. Fendler, J. H. Acc. Chem. Res. 1980, 13, 7-13. Fendler, J. H. J. Phys. Chem. 1980, 84, 1485-1491. Fendler, J. H. Pure Appl. Chem. 1982, 54, 1809-1819. Fendler, J. H. "Membrane Mimetic Chemistry", Wiley-Interscience: New York, 1982. Fendler, J. H. Chem. Eng. News 1984, 25(1),25-38. Fendler, J. H. I n "Surfactants in Solution", Mial, K. L.; Lindman, B., Ed.; Plenum Press: New York, 1984. Fendler. J. H.; Fendler, E. J. "Catalysis in Micellar and Macromolecular Chemistry", Academic Press: New York, 1975. Fendler, J. H.;Romero, A. Life Sci. 1977, 2 0 , 1109-1120. Fendler, J. H.; Tundo, P. Acc. Chem. Res. 1984, 17, 3-8. Fuhrhop, J. H.; Batsch, H.; Fritsch, P. Angew. Chem. I n t . Ed. Engl. 1981, 2 0 , 804-805. Fuhrhop, J. H.; Mathieu, J. J. Chem. Soc., Chem. Commun. 1983, 144-145. Gaines, G. L., Jr. "Insoluble Monolayers at Liquid-Gas interfaces", Interscience: New York, 1966. Gratzel, M. Acc. Chem. 1981, 14, 376-384.
Received for review March 8, 1984 Revised manuscript received August 16, 1984 Accepted October 5, 1984
Ultraviolet Radiation Curable Paints A n n e M. Grosset' and Wei-Fang A. Su Westinghouse RbD Center, Pittsburgh, Pennsylvania
15235
In product finishing lines, ultraviolet radiation curing of paints on prefabricated structures could be more energy efficient than curing in natural gas-fired ovens. Diffuse ultraviolet light cures paints on three-dimensional metal parts. The spectral output of radiation sources must complement the absorption spectra of pigments and photoactive agents so that highly pigmented (>35% by weight) thick (>25 pm) films c a n be fully cured by UV radiation. Photosensitive compounds such as thioxanthones are used to photoinitiate unsaturated resins such as acrylated polyurethanes in paints cured by a free-radical mechanism. Cationic photoinitiators such as sulfonium or iodonium salts of complex metal halide anions are used in the polymerization of epoxy paints by ultraviolet radiation.
Introduction The Westinghouse Electric Corp. has completed a program to develop ultraviolet light curable paints for application on three-dimensional objects representing typical industrial items. This program was sponsored by the U.S. 0196-4321/85/1224-0113$01.50/0
Department of Energy in an effort to evaluate a more energy-efficient technique than direct Tied gas ovens which traditionally have been used for curing paint films on metal substrates such as appliances, metal furniture, automobiles, and other fabricated metal parts. The goal of this program 0 1985 American Chemical Society
114
Ind. Eng.
Chem. Prod. Res. Dev., Vol.
24, No. 1, 1985
was to demonstrate the feasibility of using UV radiation as a replacement for natural gas in paint curing; it was not to develop paint for specific products. High quality, durable paint finishes are generally achieved by high-temperature baking cycles, e.g., 20 to 30 min at 175 "C. A high-temperature cure frequently results in a more highly cross-linked polymer. Only about 0.2% of the heat input is actually used in the curing of the paint film in conventional ovens (Grosset et al., 1981). Since most manufactured products are painted for decorative and protective purposes, industrial finishes are used in large volumes. In 1976, approximately 81.24 billion ft3 of natural gas was used to bake paints in product finishing lines in the United States ( U S . Department of Commerce Bureau of Census, 1976). However, the energy shortage of the 1970's and 1980's has led to renewed interest in more efficient alternative curing methods using energy sources other than oil or natural gas. Alternative systems based on electricity, which can be generated from a variety of sources, present the opportunity for meeting the needs of an efficient and economical curing system. One of the most promising and most widely accepted paths for translating electricity into chemically usable energy is radiation, especially ultraviolet (UV) radiation. UV curing, by itself or followed by some degree of thermal curing, consumes only one-third to one-fourth the energy used in natural gas direct fired oven curing because most of the energy is used for curing the resin and not in heating the painted part. Also, in the UV curing tunnel, no heated air mass is needed for curing coatings and for diluting solvent vapors, because the coatings are solvent-free and completely reactive. Since UV light has great specificity in exciting photopolymerization initiators, UV curing is more energy efficient than thermal (Hulme, 1975). Furthermore, there is no need for incineration or other pollution controls since there is practically no volatile organic emission, due to the absence of solvent in the coating. This process allows very rapid cure (sometimes in the order of 3 to 30 s) as opposed to the longer time (usually 3 to 30 min) required for a thermal cure so that throughput is increased; it also saves plant space since UV curing does not require the huge ovens necessary for conventional thermal curing processes. As a further benefit, UV coatings can be cured on heat-sensitive substrates. With electron beam irradiation, an inert atmosphere is required in the curing chamber, which is not necessary for UV curing. Also, shielding is less of a problem with UV than with electron beam curing. Furthermore, the capital equipment costs of UV are considerably lower. Despite higher material costs at the present time, the overall costs are lower for UV curing of paints than for curing by direct gas-fired ovens because of decreased energy consumption (Miller, 1974). Operating experience with UV light curing processes has indicated an overall operating cost which is approximately one-fourth that of natural gas. Although UV curing of clear and lightly pigmented coatings, or thin pigmented coatings such as inks, is well established in many applications from flooring to metal decorating, UV-curable industrial quality paints had not been developed previously. There are two obstacles to the use of heavily pigmented coatings on three-dimensional objects: one consists of the difficulty for the radiation to penetrate through a largely opaque coating for a through cure, and the other is the difficulty in assuring that the three-dimensional object receives uniform irradiation. The starting point of this program was the matching of resins, photoinitiators, and pigments which resulted in coatings that cured by UV radiation. The end point was
Figure 1. Ultraviolet light tunnel.
the application of these coatings to prefabricated metal structures to evaluate the viability of this technique in producing commercially acceptable painted products. These paints produced films that were durable, adherent, and opaque at a nominal thickness of 25 pm. Experimental Section The major compositions of the UV-curable paints described herein contain a photosensitizer system (photoinitiator and photosensitizer), a resin binder system (resins and reactive diluents), and suitable pigments and modifiers (extenders, flow control agents, surfactants). The ingredients were mixed well and then ground on a three-roll mill to thoroughly disperse the pigment. For the preliminary screening tests, paints were applied on three different cold rolled steel panel substrates: iron phosphated, zinc phosphated, and untreated. Later studies involved the application of UV-curable paints onto nonferrous substrates such as plastics and other metals. The paints were applied at a nominal film thickness of 25 pm with a Bird film applicator (draw down bar). The coated panels were passed through the UV-curing processor where they were irradiated with UV radiation. The initial screening tests conducted were for crosshatch adhesion and pencil hardness. Coatings passing these tests were evaluated by more intensive screening using Taber abrasion, conical mandrel bend, and impact resistance tests. A more complete evaluation of UV-curable paints was conducted by subjecting them to salt spray, humidity, Weather-Ometer, and Fade-Ometer tests. To simulate the shapes, angles and configurations of industrial painted parts, 7.62 X 7.62 X 12.7 cm3 steel file boxes were chosen for the pilot plant demonstration. The exteriors and interiors of the file boxes were spray-painted on a laboratory scale finishing line which was set up to demonstrate the practicality of the UV light curing system and to generate operating and economic data. To simulate a typical industrial paint line, an overhead conveyor moving at 2.44 m/min took spray-painted parts through a 1.52 m long tunnel which had six Fusion Systems Corp. Model F440 UV lamps arranged radially around the travel path. The UV-curing tunnel is illustrated in Figure I.
Results and Discussion Paint Development. In the conventional development of a paint, the resin binder is chosen first; then, suitable pigments are added, followed by various modifiers such as extenders and flow-control agents. In the development of UV-curable paints the procedure is more complicated than for conventional paints. Due to the interrelationship
Ind. Eng. Chem. Prod. Res. Dev., Vol. 24, No. 1, 1985
0 372
300
3ta
3a
4w
4n
MI
Wavelength. nm
Figure 2. UV absorption spectrum of TiOz
of the resin binder, pigments, photoinitiators, and other agents, the absorption spectra of each component must be selected to complement the output spectra of the UV radiation source. The selection of each component in the paint formulations is discussed in the following. Pigments. During the first phase of the program, suitable photosensitive materials were identified and formulated into paints. The pigments, which largely determined the spectral absorptivity of the coatings, were selected first. An important criterion in the choice of pigments is their effect on the penetration of UV light through the depth of the pigmented film. High opacity at a cured thickness of about 25 pm was desired. The hiding power of a pigment is determined by its ability to scatter and absorb visible radiation. For white pigments, hiding is accomplished primarily by light scattering, which depends on the difference in refractive index between pigment and binder. Pappas and Kuhhirt (1975) have examined the pigments under consideration for their effect on the ability of UV to cure pigmented films. In the pigment selection process the amounts of pigment and the film thickness were adjusted to provide films with approximately the same hiding power. As expected, films with greater hiding power allow less penetration of UV radiation. Cure times are similar for formulations using the following three grades of titanium dioxide: an anatase TiOz (American Cyanamid Co., Unitane 0-310), rutile TiOz (Tioxide America, Inc., Tioxide RHDGX), and rutile TiOZ (NL Industries, Inc., Titanox 2020), which were selected for pigmenting white top coats and one-coat enamels in free-radical and cationic systems. The rutile form of titanium dioxide pigment was selected for a top coat or one-coat enamel because it provides superior hiding power at lower pigment volume concentration and thickness, and it has less chalking than the anatase form. However, it can be seen in Figure 2 (Hulme and Marron, 1974,1977;Hulme et al., 1976a, 1976b; McGinniss, 1978) that the spectral absorbance of rutile titanium dioxide screens light through most of the ultraviolet and visible ranges, leaving only a small window through which the UV radiation can penetrate (Pappas and Kuhhirt, 1975). Therefore, it was necessary to select photosensitizers or photoinitiators which absorb strongly in this spectral region and also to select lamps which have a large output in this area, such as iron iodide lamps, other metal halide lamps, xenon lamps, and the Fusion Systems Corp. nonmercury D or V bulbs (Matthews, 1980). On the other hand, use of a medium pressure mercury lamp resulted solely in surface cure. The nonmercury Fusion Systems Corp. D and V bulbs were chosen and mounted on a conveyorized UV processor modified to be usable with several different lamp systems. Pigments other than TiOz were also investigated Thalo Green, Thalo Blue, Brown Iron Oxide, Fast Red, Red Iron Oxide, Chrome Yellow, Carbizol Violet, Raw Umber, Chromium Oxide, Hansa Yellow, and Yellow Oxide. A
115
pigmentation study was conducted to establish typical pigment loadings for white, black, and gray paints with complete hiding power. Although the loading has not been optimized, 40% by weight of Ti02 gives complete hiding in a white-pigmented coating. The 40% by weight level of Mapico Black gives complete hiding for a black pigmented coating and the 40% level of a mixture of 95% Ti02, and 5% Mapico Black yields a pigmented coating whose color closely approximates ANSI 61 light gray machinery enamel. Paints of several different colors were formulated based on a Ti02-pigmented acrylated urethane white paint photoinitiated with a 0.5% 2-chlorothioxanthane (CTX) and 1%N-methyldiethanolamine (MDEA) system. These paints cured in 3 to 30 s, with faster curing paints usually resulting in a smooth coating. Yellows and reds are particularly slow to cure. The colors of all the paints in this pigment study were clear and bright, with no yellow discoloration. Primers were formulated with several different pigments. Red iron oxide or zinc chromate was added as a corrosion-inhibiting pigment. Halox BW-111 was used as an additional inhibitor. Fibrene C-400 served as a pigment extender. Free-Radical Polymerization. The selection of photosensitizer systems and resin binders was based on two photoinduced polymerizations: the free-radical mechanism and the cationic mechanism. In any given system, freeradical production is dependent upon the chemical characteristics and reaction pathways of the individual components. A free-radical mechanism photoinitiator system contains one or more photoinitiators and may contain a photosensitizer or co-initiator. Co-initiators would provide a synergistic effect with the photoinitiator serving as hydrogen donors or forming a donor acceptor complex with an excited carbonyl group and participating in a chargetransfer or electron-transfer process. Co-initiators could also be photoinitiators which react by a different mechanism, as in hybrid systems. The most promising photoinitiators for the free-radical system, 2-chlorothioxanthone (CTX) and benzil, absorb in the spectral range which is not screened by Ti02. The effects of concentration on cure time and physical properties of films photoinitiated with 2-chlorothioxanthone (CTX), benzil, 2methylanthraquinone (MEAQ), and 2-ethylanthraquinone (EAQ) (Hulme, 1975; Hulme and Marron, 1974, 1977; Hulme et al., 1976, 1977; Murov, 1973; Berner et al., 1979; Ciba-Geigy, 1980; Dufour, 1979; UCB, a,b; Delzenne, 1979) were examined. In each case, cure times and physical properties of films which were cured using Fusion Systems Corporation D bulbs (Matthews, 1980) reached an optimum and then declined with increasing photoinitiator concentration. Co-initiators such as tertiary amines and arylchloromethyl and chlorosulfonyl compounds were added to various formulations for hydrogen transfer, electron transfer, and energy transfer. CTX-photoinitiated films containing co-initiators exhibited yellowing. The film cure times depend upon the ratio of concentrations of CTX and amine. The following compounds exhibited reactivity in a photoinitiator system in both clear and pigmented coatings: N-methyldiethanolamine (MDEA) (McGinniss, 1979; Armbruster et al., 1979), diethylenetriamine (DETA), monoethanolamine (MEOA), diethanolamine (DEOA), triethanolamine (TEOA), N,N-dimethylaminoethyl methacrylate (DMEMA), phenylethylamine (PEA), triethylamine (TEA), dimethylethanolamine (DMEOA),
116
Ind. Eng. Chem. Prod. Res. Dev., Vol. 24, No. 1, 1985
morpholine (MORPH), ethyl-p-dimethylaminobenzoate (EPDMAB), quinoline sulfonyl chloride (QSC), Virginia Chemical Co. Uvercryl P101, P102, and P104, Stauffer Chemical Co. Vicure 30 (benzoin isopropyl ether, V-30), and Michler’s Ketone [4,4-bis(N,N-dimethylamine benzophenone)]. MEOA, MDEA, and DETA were particularly reactive in clear coatings. The most promising formulation had N-methyldiethanolamine (MDEA) as co-initiator. The formulations containing CTX and MDEA cured rapidly (3.2 s) and formed hard films, but adhesion was unsatisfactory. The formulations containing CTX and morpholine cured more slowly, were softer, and had better adhesion than formulations containing CTX and MDEA. Isopropylthioxanthone and methylthioxanthone exhibited properties similar to those of CTX. Benzil and tertiary amine photoinitiator systems were evaluated. Benzil/amine systems in general gave fast curing, soft, white films having poor adhesion. Cure times depend upon the ratio of the concentrations of benzil and amine (McGinniss, 1978). The cure rate first decreased rapidly and then increased slightly with an increasing photoinitiator/amine ratio. The minimum was generally around 1:1. The benzil/morpholine system had good properties, but it was extremely sensitive to thickness. Cured films containing benzil and morpholine had greater adhesion than films containing CTX. Their hardness and adhesion were better in thin coatings (1.25 pm). Combinations of CTX and benzil with tertiary amines resulted in increased adhesion with no decrease in cure speed. Although EAQ and MEAQ-initiated films were yellow and slower curing than CTX or benzil-containing films due to their lower absorptivity in the near-ultraviolet range, they were durable and adherent. In general, it appears that the addition of tertiary amines to EAQ-photoinitiated formulations had little effect on properties, although the photocleavage of 8-quinoline sulfonyl chloride decreased cure time. Pigmented coatings containing CGI 1184 (Ciba-Geigy) showed no improvement in whiteness over coatings containing CTX. Combinations of photoinitiators were tested. Coatings using combinations of photoinitiators which included benzil also exhibited inconsistent adhesion; i.e., they were particularly sensitive to film thickness. The use of CTX and amine resulted in coatings similar to coatings containing benzil and morpholine. EAQ was tested in conjunction with other photoinitiators. A more rapid cure and good adhesion were achieved with the addition of ethyl p-dimethylaminobenzoate as a co-initiator. Combinations of EAQ with other photoinitiators such as benzil might accelerate cure rate while retaining adhesion. Formulations containing EAQ appeared to most consistently produce satisfactory adhesion. As expected, an increase in length of time of exposure t o UV light radiation led to harder coatings. Photoinitiator systems containing MEAQ and EAQ had consistently good adhesion, except in combination with benzophenone. Coatings containing benzoin ethers and other photoinitiators, which were partially screened by the titanium dioxide, cured at a moderate rate and were soft. Several polyesters and acrylated urethanes were selected for their optimum properties of high reactivity, adherence, flexibility, and toughness. An acrylated urethane, Hughson Chemical Co. TS3577-11, was chosen as the base resin for the majority of the formulations. The reactive monomeric diluents tested, which included multifunctional acrylates, monofunctional acrylates, and
nonacrylic unsaturated esters (Rynby et al., 1978) which showed high reactivity when photoinitiated by use of isobutyl benzoin ether (Stauffer Chemical Co., Vicure V-lo), are pentaerythritol tetracrylate (PETTA), pentaerythritol triacrylate (PETA), trimethylolpropane triacrylate (TMPTA), tetraethyleneglycol diacrylate (TEGDA), diethyleneglycol diacrylate (DEGDA), 1,6-hexanediol diacrylate (HDDA), triethylene glycol dimethacrylate (TEGDMA), and 1,3-butylene dimethacrylate (BDMA). Selected paint formulations curing with a free radical mechanism are listed in Table I. The formulations shown are UV-curable acrylated urethane one-coat enamels. Because of their high reactivity, multifunctional acrylates such as 1,&hexanediol diacrylate, tripropyleneglycol diacrylate (TPGDA), and triethyleneglycol diacrylate (TEGDA) were chosen as viscosity reducing diluents. Monofunctional acrylates, such as 2-ethoxyethyl acrylate, were also added to reduce the viscosity of the resin system and to increase flexibility of the film. The viscosity and cure times of several systems which combined various resins and reactive diluents were determined and the properties of their films were evaluated by conical mandrel bend, impact, abrasion, and hardness tests. Cationic Polymerization. Newly developed UV-sensitive cationic photoinitiators (Pappas, 1978),the so-called “onium salts”, aryldiazonium, diaryliodonium, and triarylsulfonium salts of complex metal halide anions such as hexafluoroarsenate, hexafluorophosphate, and hexafluoroantimonate, were used to polymerize epoxy resin. Cationic photoinitiators, 3M’s FC-508 and FC-509, were used in the formulation. Coatings photoinitiated with FC-508 have a longer shelf life than those using FC-509, but the cure rate of FC-508 is slower than that of FC-509. Since the spectral sensitivity of cationic photoinitiators is relatively low (Pappas, 1978, 1980) and the white pigment, rutile Ti02,screened out the major absorption region of the photoinitiators, a photosensitizer is necessary in the formulation to produce an effective photoinduced reaction. The following photosensitizers were used: FC-510 (3M Co.), 2-~hlorothioxanthone,thioxanthone, benzanthrone, 1,3-diphenyl-2-propanone, triphenyl acetophenone, fluorenone, biphenyl, flavone, 4-acetyl biphenyl, diacetyl, fluorene, 9,10-phenanthrenequinone,benzophenone, and benzil. 2-Chlorothioxanthone, thioxanthone, and FC-510 had the same degree of effectiveness,whereas all the others were less effective. The cure rate is optimized at a 3 to 1 ratio of FC-509 to photosensitizer and a 4 to 1 ratio of FC-508 to photosensitizer. 2-Chlorothioxanthone and thioxanthone usually produced a slightly yellow (off-white) coating, which may be due to the absorption of the photosensitizer in the visible range. FC-510 also gave a slightly yellow coating right after the UV curing. However, the degree of yellowness diminished with time. A Fusion System Corp. nonmercury D bulb or V bulb was used to cure the epoxy paints, because the bulbs have strong emission spectra at the edge of, or outside, the range of the rutile TiOz absorption spectra (see Figure 3, the emission spectra of fusion bulbs). An epoxy resin ERL 4221 (Union Carbide) was used as a paint vehicle, because of its rapid cure speed (Crivello, 1977; Watt, 1978) and excellent film adhesion (Nylen and Sunderland, 1965). Reactive diluents, butyl glycidyl ether (Ciba Geigy RD-I), 1,4-butanediol diglycidyl ether (Ciba Geigy RD-2), 1-butanol, and limonene dioxide were used to lower viscosity, but they usually made the coating brittle. The flexibility of ERL 4221 can be modified by
Ind. Eng. Chem. Prod. Res. Dev., Vol. 24, No. 1, 1985
117
Table I. Properties of UV-Curable Acrylated Polyurethane Paints Cured with D Bulb formulation TS 3577-11 HDDA EEA CTX Titanox 2010
TS 3577-11 HDDA EEA benzil Titanox 2010
TS 3577-11 HDDA EEA EAQ Titanox 2010
TS 3577-11 HDDA EEA CTX MDEA Titanox 2010
TS 3577-11 HDDA EEA Benzil morpholine Titanox 2010
crosshatch test, %* Zn phos Fe phos steel
pencil hardness testo Zn phos Fe phos steel
wt, I
cure time
30.00 10.00 5.00 0.18 20.00
9.6 s 3.81 m/min 3.00-kW lamp output
100 100
100 100
12 10
HB HB
3B
