Document Lgywz335bdVKV6ODbyGanpzeb

THIS PAPER WAS AWARDED FIRST PRIZE OF $750 IN THE 1974 ROON AWARDS COMPETITION Photo-Chemistry of Pigments Studies on the Mechanism of Chalking S. Pe t e r Pa p p as and Ric h a r d M. Fis c h er ** North Dakota State Universityt Studies are presented which demonstrate that* (1) ir radiation of titanium and zinc oxide pigments produces singlet oxygen; (2) irradiation of titanium dioxide pig ments in water yields hydrogen peroxide; and (3) the formation of singlet oxygen and hydrogen peroxide correlates with chalking tendencies of the pigments. These findings, together with the results of quenching studies, are interpreted In terms of a working hypo thesis, for the generation of reactive oxidants, which ties together previous work Into a unified scheme. The relative chalking rates of anatase and rutile titanium dioxide, as well as the improvement of chalk resis tance by surface treatment, are discussed within the framework of this scheme. The role of singlet oxygen in the chalking process, the importance of its presence with regard to the control of chalking, and possible mechanisms for its formation are also discussed. KEY WORDS: Chalking; Energy transfer; Hydrogen peroxide in chalking; Singlet oxygen in chalking; Pholo-chemistry; Photo-sensitization; Weathering. INTRODUCTION Ou r interest in chalking developed from a primary concern in the photochemistry or pigments. We were particularly intrigued by the prospect of pigments act ing as photo-sensitizers,1 and by the application of this concept to the utilization of UV radiation for curing pigmented coatings (UV curing). In view of the con siderable effort that has been devoted to the problem of chalking, it was inevitable that even a cursory view of the literature on pigment photo-chemistry would lead us to a consideration of the chalking phenomenon. This paper is based on R. M. Fischer's Ph D. Thesis, North Dakota State University, 1974. Presented by Dr. Pappas at the 52nd Aanual Meeting of the Federa tion of Societies for Coatings Technology in Atlanta. Ca., Nov. C-8. 1974. Present address: SM Co.,-3M Center, St. Paul, Minn. 55101. f Dept, of Polymers and Coatings, Fargo, N. D. 58102. In initiating a research project on chalking, we re ceived considerable encouragement and inspiration from Dr. Alfred E. Rheineck,2 to whose memory this paper is dedicated. Chalking is a common problem of exterior paints which involves degradation of the organic polymer part of the paint (be., the binder), thereby exposing the pigment, which appears on the surface as a chalk like substance. With regard to the mechanistic aspects of chalking, a recent review on the weatherability of Ti02-containing paints3 is particularly informative and presents two modem views: (1) that polymer degra dation is related to the formation of superoxide ion (or oxygen radical anion). Os*"", produced by electron transfer from excited-state Ti02 to molecular oxygen, and (2) that the degradative species is hydroxyl radi cal, HO*, postulated4 to arise by electron transfer from water to excited-state Ti02. Earlier studies had implicated oxygen atoms as the reactive oxidant.6 In view of our interest in photo-sensitization, we were particularly intrigued by the suggestion6 that elec tronically excited oxygen molecules may be produced on irradiation of the semi-conducting white pigments, by analogy with dye sensitization. In the way of background, two excited states of oxygen are of interest (both of which are singlet states in contrast to the triplet ground state of oxygen ): (1) the sigma singlet state, which possesses 37.5 keal/mol above the ground state and is predicted to have free radical reactivity, and (2) the delta singlet state, which possesses 22,5 keal/mol above the ground state and has been characterized as a reactive dienophile.7 Although higher in energy, the sigma state is consider ably shorter lived, and, in fact, rapidly deactivates to the delta state, which has been identified or implicated in many important chemical processes, including poly mer degradation8* and autoxidation of drying oils.10 Vo l . 46, No. 599, Dec ember 1974 65 N41535 $. P. PAPPAS and R. M. FISCHER Thus, it appeared worthy of study to investigate whether singlet oxygen might be involved in the chalk ing process. EXPERIMENTAL Microwave Synthesis of 2-Methoxy-5-oxo-2, 5dihydrofuran (1) Singlet oxygen was produced in a flow system with a Raytheon microwave power generator,8 model PGM-10, 2450 me, as an excitation source. With the system under a vacuum of less than 1 mm, the micro wave discharge was ignited with a Tesla coil and the oxygen flow was increased until the system pressure was 3 mm. Furan vapor was released into the system until a pressure of 5 mm was attained. A total of 30 ml of furan was introduced into the system over a period of 6 hr. The reaction products and unreacted furan were collected in a liquid nitrogen cold trap. Vanadium pentoxide (1 g) and 80 ml of methanol were added, and the mixture was refrigerated for 36 hr. After fil tration of the vanadium pentoxide, the methanol and furan were removed from the higher boiling products by rotary evaporation at room temperature. The resi due (0.4 g) was decolorized with charcoal and fractionated by gas-liquid chromatography (glc), using a Varian-Aerograph 1740 instrument with flame ionization detector and disc integrator. The major product (73%) was collected and identified as 2methoxy-5-oxo-2, 5-dihydrofuran11 by infrared and mass spectral analysis. The infrared spectrum, ob tained in chloroform, exhibited the characteristic car bonyl absorbances at 5.57 and 5.68 /*.. The mass spec trum, obtained on a Hitachi Perkin-Elmer, RMU-6E, spectrometer, exhibited a prominent molecular ion at m/e 114, and fragment ions corresponding to losses of H, CH3, CHaO, and C02. Irradiation of the Furan-Pigment System Irradiations were carried out in a 2000-ml beaker modified with a condenser, sampling port and oxygen inlet A General Electric Par-3S, H100PSP38-4, 100-w black light source was mounted above the beaker and the light was filtered through a 2-mm Pyrex cover glass, to eliminate wavelengths shorter than 300 nra, Titanox A-MO and Titanox 2005, obtained from N L Industries, Inc., Ti-Pure R-901, obtained from E. I. du Pont de Nemours & Co., Inc,, and zinc oxide XX-602, obtained from the New Jersey Zinc Co., were utilized individually as sensitizers. Characterization of the Ti02 pigments is presented in Table I. In each case, the pigment (20 g) was allowed 24 hr to settle on the bottom of the reactor, which contained 400 ml of reagent-grade methanol. Furan (4.8 g) and ethy lene glycol (100 mg), the internal standard for glc analysis, were added to the reaction mixture just prior to irradiation. Oxygen was passed over the system at Table T--Characterization of the Titan!um Dioxides Titanium Dioxide Type Chalking Ability Surface Treatment Titanox A-MO` Titanox 2005" Ti-Pure R-901* Anatase Rutile Rutile High Medium Low Light Minimum Treated (a) ASTM D4TG-70, Type I. (b) Type II. (c> Type HI. a flow rate of 10-15 ml/min. The reaction temperature was maintained at about 15C by placing the beaker in a water bath with a coil for the continuous passage of tap water. Aliquots (20 ml) were removed every 24 hr. These were rotary evaporated, diluted with 0.25 ml of acetonitrile, and analyzed by glc. The de sired product, 2-methoxy-5-oxo-2, 5-dihydrofuran, was identified in the glc traces by peak matching with authentic material from the microwave synthesis. The peak matching was performed under five different sets of conditions, including the utilization of three different column-packing materials: 5% Carbowax 1500 on 80/90 Anakrom U, 5% isooctylphthalate on 60/ 80 Chromosorb G, and 1.5% Versamid 900 on 80/90 Chromosorb G. The system of choice was 5% Carbo wax 1500 on 80/90 Anakrom U in an 8 ft X & in. column at 128 C, with carrier gas flow rate of 10 ml/ min. Under these conditions, the retention time of the desired product was 11.1 min. This product was not formed in the absence of pigment over a 5-day ir radiation period. Preliminary studies had demon strated that the product is formed on sensitization with the dye, Eosin Y. Representative results with the pigments are pre sented in Table 2 and Figure 1. Studies were also carried out which demonstrated that neither the singlet oxygen product nor the internal standard, ethylene glycol, are differentially adsorbed on the Ti02 pigments examined. A sample of known product ratio was added to a suspension of the pigment in methanol and allowed to equilibrate for 24 hr. The pigment was removed by filtration through a bed of Celite 545, the methanol was removed by rotary evaporation, and the resulting sample was re-analyzed Table 2--Formation of 2-Methoxy-5-oxo-2, 5-dihydrofuran (I) Irradiation Product/Internal Standard* Time Titanox Titanox Ti-Pure (days) A-MO 2005 R-901 Zinc Oxide XX-602 0 1 0.028 Trace 2 0.09 0.022 - _ 0.017 0.028 3 0.105 0.038 0.035 4 0.118 0.055 Trace 0.042 5 0.15 O.0G1 0.012 0.049 (a) Glc response ratio; internal standard: 4 raAf. 66 [o u r n a l o f Pa in t Te c h n o l o g y * DUP010001949 PHOTO-CHEMISTRY OF PIGMENTS Figure I--Graphic comparison of singlet oxygen product formation for Titanox A-MO and 2005 by glc. For each of the Ti02 pigments, the ratio of singlet oxygen product to internal standard remained unchanged, thereby eliminating selective adsorption as a possible explanation for the results. Several experiments were performed on the effect of p-benzoquinone (1.90 X 10'4 to 2.57 X 10"2 M) and triethylamine (4.0 X 10"2 M) on the formation of the singlet oxygen product, I, utilizing Titanox A-MO as sensitizer. In each case, the formation of singlet oxygen product was verified by glc prior to intro ducing the additive. The results of these studies are presented in Figures 2 and 5. Irradiation of the Water-Pigment System The apparatus was identical to that utilized in the furan system. Titanox A-MO, Titanox 2005, and Ti-Pure R-901 were utilized individually as sensitizers. In each case, the pigment (20 g) was dispersed with distilled water (300 ml) by magnetic stirring, and subsequently was allowed to settle on the bottom of the beaker. Unwetted pigment was skimmed off the surface prior to irradiation. During the irradiation, aliquots were removed periodically and filtered through a 5-mm bed of silicic acid to remove any sus pended pigment particles. The aliquots were assayed for hydrogen peroxide content by the following method.12 To 3.9-ml aliquots are added: (1) 0.5 ml of 0.3 M potassium phosphate buffer, pH 7.8, (2) 0.5 ml of 0.6 mM o-dianisidine dihydrochloride, and (3) 0.1 ml of a 0.013$ aqueous solution of horse radish peroxidase, buffered with potassium phosphate. The presence of hydrogen peroxide was qualitatively indicated by the formation of reddish color in the sample immediately after the addition of the peroxidase solution to the irradiated samples. Quantitative measurements were made with a Bausch and Lomb Spectronic 20 at 450 nm. Optical density at this wavelength was converted into hydrogen peroxide concentration by reference to a standard curve. The results are presented in Figure 3. Figure 2--Effect of Irieihyfamine on singlet oxygen product formation with Titanox A-MO The effect of p-benzoquinone (2.3 X 10'4 M) and triethylamine (4.0 X 10~2 M) on the stationary con centration level of hydrogen peroxide, utilizing Ti tanox A-MO as sensitizer, was investigated. The re sults are presented in Figures 4 and 6. Control experi ments were also carried out which demonstrated that the assay system is not adversely affected by the presence of the inhibitors at the concentrations uti lized. Furthermore, the stability of hydrogen peroxide toward triethylamine in the absence of pigment was demonstrated under the irradiation conditions, RESULTS Singlet Oxygen Production Our initial approach to the problem was to de termine whether delta singlet oxygen is produced on irradiation of TiOo, by exploiting its known specific reactivity with unsaturated compounds.13 After con siderable experimentation with several systems, we settled on the use of furan, which, in the presence of methanol, is known to produce 2-methoxy-5-oxo-2,5dihydrofuran (I) by dye-sensitized oxygenation.14 The established mechanism of this process is outlined in Scheme I, the criticaL feature being electronic energy Figure 3--Comparison of hydrogen peroxide formation for Titanox A-MO and 2005 Vo l . 46, No. 599, Dec ember 1974 67 DUPO10001950 S. P. PAPPAS and R. M. FISCHER IrrtdUtian Time (hour*) Figure 4--Effect of trlethylamme on hydrogen peroxide formation with Tftanox A-MO transfer (Equation 3) from the excited triplet-state dye (3D) to ground-state oxygen (302), to produce ground-state dye (D0) and excited singlet-state oxygen (102). 15,16 The singlet oxygen undergoes a DielsAlder reaction with furan to produce a peroxide (Equation 4), which may be isolated at low tempera ture, but reacts with methanol under ambient condi tions to produce I. Sc h e me I D0 -- 'D* -f 3Oa-> D0 -f 'OS (l) (2) 0) V) H ___k H + CH3OH -----------> <*/ (5) V"H xcr OCHj I It was our intent to determine whether Ti02 could function in place of the dye in the energy trans fer step. For the purpose of identifying and monitor ing the formation of I by glc, this compound was pre pared independently by reaction of furan with singlet oxygen generated in a microwave discharge,8 followed by work-up with methanol. Preliminary studies were also performed utilizing dye sensitization. Irradiations with pigments as sensitizers were carried out in a beaker with side arms for oxygen intake and exhaust. The pigment was deposited on the bottom of the beaker, which also contained the furan-methanol solu tion and ethylene glycol as an internal standard for glc analysis. The system was irradiated from the top through a Pyrex cover glass to exclude wavelengths shorter than 300 rim. The initial results with a chalking anatase, Titanox A-MO, were encouraging, as shown in Figure I, since no singlet oxygen product was ob- Figure 5--Effect of p-benzoqulnone on singlet oxygen product formaHon with Titanox A-MO tained in the absence of pigment. These findings prompted the investigation of a medium-chalking ru tile, Titanox 2005, which produced significandy less product (see Figure 1), and of a more chalk-resistant rutile, Ti-Pure R-90I, which generated only a trace of product after irradiation for 4 days. These pigments are described in Table 1. A zinc oxide pigment, XX602, which was also examined, produced the singlet oxygen product I to the same extent as did Titanox 2005. These results were found to be reproducible, and representative data are presented in Table 2. The formation of product I constitutes strong evi dence for the intermediacy of singlet oxygen. Further support was acquired by demonstrating that triethylamine (4 X 10~2 M), a known singlet oxygen quencher,17 effectively quenches the formation or I, as shown in Figure 2. The possibility that the relative extents of formation of I reflected differential adsorp tion of either product I or the internal standard on the pigments was also ruled out. Thus, the results demonstrate that: (1) irradiation of titanium and zinc oxide pigments produces singlet oxygen, and (2) the extent of formation of singlet oxygen correlates with the chalking tendencies' of the TiOo pigments ex amined. Figure 6--Effect of p-benzoqulnone on hydrogen peroxide formation with Titanox A-MO ,68 Jo u r n a l o f Pa in t Te c h n o i^* DUPO10001951 Hydrogen Peroxide Formation In view of the possible relationship of these re sults to chalking, consideration was given to the role of water, since there appears to be no question that moisture accelerates the chalking process.3 This is particularly emphasized in a recent study which clearly implicates the involvement of hydroxyl radicals (HO*).4 The formation of HO* was proposed to occur by oxidation of water adsorbed on the surface of ex cited TiOs with concomitant reduction of titanium to the +3 state. However, this mechanism has been questioned on thermodynamic grounds.3 An alterna tive source of HO* which should be considered is hy drogen peroxide. Photolysis of hydrogen peroxide is known to pro duce hydroxyl radicals, and this process appears to be catalyzed by Ti02.18 Furthermore, hydroxyl radi cals are produced in a redox reaction between hydro gen peroxide and Ti (III),19 which is known to be present in the Ti02 lattice.20 Thus, hydrogen peroxide represents an attractive precursor for hydroxyl radi cals. However, the formation of H202 on irradiation of TiOo pigments has not been clearly demonstrated and, consequently, its possible involvement in the chalking process has remained an open question.3 This situation is contrasted by the well-documented formation of hydrogen peroxide21 on irradiation of ZnO, and its recognized relationship with chalking in zinc oxide systems.22 For these reasons, it was decided that fur ther investigation on the possible formation of hydro gen peroxide with Ti02 pigments was warranted. Irradiations were carried out in the same ap paratus utilized for the singlet oxygen experiments, except that the methanol-furan solution was replaced by water. Anticipating that levels of H202 would, if produced, be low, a highly sensitive and specific en zymatic-spectroscopic assay method,12 rather than the conventional iodide titration, was utilized. The assay system is based on the reaction of hydrogen peroxide with the enzyme, horse radish peroxidase, which, in the presence of o-anisidine, may be quantitatively monitored at 450 nm. As illustrated in Figure 3, the formation of hydrogen peroxide was readily demon strated with the chalking anatase, Titanox A-MO, and lower levels were obtained with the medium-chalking rutile, Titanox 2005. Hydrogen peroxide could not be detected with Ti-Pure R-901. Thus, in accordance with the singlet oxygen results, the formation of hydrogen peroxide correlated with chalking tendencies of the pigments. Further Quenching Studies The production of a steady-state concentration of hydrogen peroxide is also characteristic of the zinc oxide system, for which it has been demonstrated by isotopic labeling that both oxygens of the hydrogen peroxide are derived from molecular oxygen.23 With the hope of providing further insights into the possible PHOTOCHEMISTRY OF PIGMENTS relationships between singlet oxygen and hydrogen peroxide formation, some common quenching experi ments were also carried out. Of particular significance, the addition of 4 X I0~2 M triethylamine, which total ly quenched singlet oxygen product formation with Titanox A-MO, reduced the corresponding stationary level of hydrogen peroxide by about 50$, as shown in Figure 4. Appropriate control experiments were per formed which demonstrated that hydrogen peroxide solutions were stable toward triethylamine in the ab sence of pigment under the irradiation conditions. The addition of 2 X 10"4 M p-benzoquinone efficiently quenched both the formation of singlet oxygen prod uct and hydrogen peroxide, as shown in Figures 5 and 6. It is interesting to note in Figure 5 that singlet oxygen production is not quenched by 10'2 M p-benzoquinone. This apparent anomaly arises from the production of singlet oxygen, not from the pigment, but by energy transfer from p-benzoquinone which exhibits substantial absorption in the 300 to 400 nm region at concentrations above 10'3 M. This result is presented to illustrate one of the complexities of the quenching studies. DISCUSSION Generation of the Reactive Oxidants Our first reaction to the results is the basic sim ilarity in the photo-chemistry of titanium and zinc oxide pigments, as measured by singlet oxygen and hydrogen peroxide formation. Thus, Scheme II, which is offered as a working hypothesis for relating the present findings to previous results, may apply, basi cally, to both pigment types, although specific refer ence is made to TiOs- s c h e me II TiO, 4. 02 h T Ti02+*-------Or n (6) II Tib* 4 KV (7) II + H,0 ^ TiOa 4 HO. 4 H03. (8) 2H02* -> HaOa' 4 Os (9) H502 4 Ti (III) -> Ti (IV) 4 HO- 4 HO. (10) hr H2Oa -----> 2HOTiO, (11) II 4 RaCHOH TiOa 4 R.CrrO 4 H=0* (12) Irradiation of titanium and zinc oxide pigments in the 300 to 400 nm range is known to promote chemisorption of oxygen, presumably by electron trans fer from the pigment, since the oxygen radical anion (superoxide) has been detected (Equation 6).24 27 The production of singlet oxygen may occur by re versal of electron transfer (ion-annihilation) with the excess energy stored in the oxygen molecule (Equa tion 7). This step is preceded by the formation of singlet oxygen on ion-annihilation of superoxide and \Vor.. '16. No. 599, De c e mb e r 1974 69 DUPO10001952 S. P. PAPPAS and R. M. FISCHER ferricenium ions generated electro-cbemically.-8 Equa tions 6 and 7 constitute net electronic energy transfer from the pigment to molecular oxygen. This mode of energy transfer represents a particularly attractive mechanism for photo-sensitization by photo-conduc tive pigments. In the presence of water. Complex II may accept a proton (H*) and an electron to produce perhydroxyl (HCV) and hydroxyl (HO-) radicals, respectively, with the regeneration of Ti02 (Equation 8j. Perhy droxyl radicals are known to disproportionate into hy drogen peroxide and oxygen at diffusional rates (Equa tion 9).29 This mode of hydrogen peroxide formation conforms to the known reactivity of superoxide,30 and is consistent with the finding that both oxygen atoms of hydrogen peroxide are derived from molecular oxy gen, as demonstrated with zinc oxide.23 Analogous schemes for the production of hydrogen peroxide have been suggested.3*23 The stationary state plateau of hydrogen peroxide may be attributed to redox decomposition by Ti (III) ions (Equation 10),19 as well as to direct photolysis to hydroxyl ions, which may be catalyzed by Ti02 (Equation 11).18 Thus, the known effect of donor sites, such as Ti (III), in promoting chalking may be attributed, in part, to enhancement of hydroxyl radical formation. Donor sites are also expected to facilitate electron-transfer to molecular oxygen.3 Equation 8 depicts the oxidation of water by Complex II. Alternatively, Complex II may participate in the direct oxidation of organic functionality, e. g., alcohols to ketones (Equation 12). Such pigmentcatalyzed photo-oxidations have been interpreted in terms of oxidation by pigment holes, represented by Ti02%.31 The application of semi-conductor theory to the photo-catalvtic activity of pigments is currently receiving considerable attention 3 The observed quenching of both singlet oxygen and hydrogen peroxide by low concentrations of pbenzoquinone may be attributed to its high electron affinity (approximately 1.40 as compared to 0.87 eV for oxygen).82-33 Specifically, p-benzoquinone is ex pected to undergo charge-transfer complexation with excited Ti02 more effectively than oxygen. In fact, the corresponding radical anion of p-benzoquinone has been detected on irradiated surfaces of titanium and zinc oxide.34 In this sense, p-benzoquinone may be considered to function as a very effective acceptor site. Such sites, in general, are expected to retard com plexation and electron-transfer to oxygen. This con cept provides a rationale for improvement of chalk resistance by surface treatment of Ti02 pigments with potential electron-acceptors, such as Zn (II) and AI (III).35 Scheme II is based on a common precursor, the ion-pair complex II, from which are generated the reactive oxidants which initiate chalking. The critical role of this complex is particularly appreciated when one considers that anatase photosorbs oxygen more efficiently than does rutile Ti02r5 even though rutile has higher absorptivity in the 300 to 400 nm range.30 Thus, higher chalking rates with anatase, as well as improvement of chalk resistance by surface treatment with potential electron-acceptors, are explicable with in the framework of this scheme. Surface treatment effects on the formation and reactivity of hydroxyl radicals have previously been suggested.4 Singlet Oxygen In Chalking The production of singlet oxygen on irradiation of the pigments raises the question of its role in the chalking process, particularly in view of the correla tion of its formation with the inherent chalking ten dencies of the pigments. This question relates to the oxidative degradation of the polymeric binder result ing in the physical exposure of pigment at the coating surface, which may be considered as the macroscopic aspect of chalking. In this regard, recent evidence has clearly implicated singlet oxygen as a precursor of hydroperoxides in hydrocarbon polymers37*38 and fatty acid derivatives.10 These results are particularly sig nificant since hydroperoxides occupy a key position in autoxidation: increasing in concentration during the induction period, and triggering the accelerated stage, which is self-catalyzing.38*40 The formation of hydroperoxides is based on the reaction of singlet oxy gen with systems containing allylic hydrogens, and oc curs concertedly with double-bond migration (Equa tion 13).18 RCH2CH=CHR' -f xOa* OOH RCH=Ch 1)HR' (IS) In accord with Equation 13, a model system has been studied which demonstrates that singlet oxygen may initiate oxidation at unsaturated polymer ends, terminated by disproportionation.41 The formation of hydroperoxide has also been reported in the reaction of a saturated hydrocarbon and singlet oxygen 41 This reaction may involve hydrogen abstraction by the sig ma singlet state, which is higher in energy and is pre dicted to have free-radical reactivity, although its lifetime in solution is exceedingly short (about 1010 sec).16 However, the hydrocarbons were adsorbed on alumina in these studies, and the lifetime of the sigma state may be extended in a solid matrix due to re duced mobility. Accordingly, sigma singlet oxygen may also be involved in chalking, which occurs in a polymer-pigment matrix. The production of singlet oxygen provides a new concept to consider with regard to the control of chalking. In addition to the direct degradative effects, the possibility also remains that singlet oxygen may indirectly contribute to chalking. The results, reported herein, of reduced stationary levels of hydrogen per oxide in the presence of triethylamine, an effective singlet oxygen quencher, are suggestive of a possible relationship between singlet oxygen and hydrogen 70 Jo u r n a l o p Pa in t Te c h n o l o g y DUPO10001953 Dr. S. Peter Pappas, Professor In the Polymers and Coatings Dept, at Norlh Dakota State University, received his A.8. Degree in Chemistry from Dart mouth College in 1958, and his Ph.D. Degree from the University of Wiscon sin in 1962. Prior to joining North Dakota State in 1968, he did post doctoral work at the University of Wis consin and Brandels University, and served on the chemistry faculty at Emory University. Dr. Richard M. Fischer is a Senior Re search Chemist in the Research and New Products Group of the 3M Co/s Adhesives, Coatings and Seaiers Div. He received his B.S. Degree in Chem istry and Physics from North Dakota State University, and his Ph.D, Degree in Polymers and Coatings from the same university in 1974. PHOTO-CHEMISTRY OF PIGMENTS of singlet oxygen product I was found to correlate with known chalking tendencies of the Ti02 pigments. The formation of hydrogen peroxide on Irradiation of the Ti02 pigments was also investigated and shown to correlate with their chalking tendencies. These results point up basic similarities in the photo-chemical be havior of titanium and zinc oxide pigments. A scheme is presented for the generation of reactive oxidants, which relates the present findings to previous work on chalking. The relative chalking rates of anatase and rutile titanium dioxides, as well as the improvement of chalk resistance by surface treatment, are discussed within the framework of this scheme. The importance of singlet oxygen as a reactive oxidant, as well as an entity to consider in the control of chalking, is discussed. The production of singlet oxygen is attributed to electronic energy-transfer from excited-state pigments, via electron-transfer and ionannihilation processes, although alternative hypotheses are also considered. peroxide formation. Since hydroxyl radicals (HO*), which have been implicated in chalking.4 are probably formed from hydrogen peroxide, the quenching of singlet oxygen may reduce HO* levels as well. An ob vious approach is the incorporation of singlet oxygen quenchers, such as Ni (II) chelates,42 which have been utilized effectively as stabilizers against photo degradation of polymers.43 The generation of singlet oxygen is reasonably at tributed to electronic energy transfer from the pig ments, via complex II, as shown in Scheme II, Equa tion 7. While this represents an attractive hypothesis, alternative pathways are conceivable. Two possibilities which we have considered are: (1) formation of oxy gen in the excited singlet state in the disproportiona tion of perhydroxyl radicals (Scheme II, Equation 9); and (2) singlet oxygen production from hydrogen per oxide, by analogy with hypochlorite oxidation.44 Both of these possibilities may be distinguished from energy transfer by demonstrating hydrogen peroxide quench ing with an agent that does not quench the singlet oxygen product We plan to pursue these and related studies on pigment photo-sensitization. SUMMARY Studies on the photo-chemistry of a zinc oxide and three representative TiOo pigments are presented. Irradiation of these pigments in the presence of a furan-methanol solution yields 2-metho.\y-5-oxo-2, 5dihvdrofuran (I). These findings constitute strong evi dence for the intermediacy of singlet oxygen. Fur ther support for singlet oxygen production was ob tained bv quenching studies. The extent of formation ACKNOWLEDGMENT The authors are pleased to acknowledge the Na tional Science Foundation Traineeship Program, Sher win-Williams Co., and DeSoto, Inc. for financial sup port, and also wish to thank Dr. Allan Fischer, Bio chemistry Dept., North Dakota State University, for assistance with the hydrogen peroxide assay, References (1) For a pertinent review on energy transfer and photo sensitization, see Knowles, A., Chem. Ind. (London), 17, 1058 (1978). (2) Chairman, Dept, of Polymers and Coatings, North Dakota State University, 1958-1971; Deceased, Aug. 1971. (3) Sullivan, W. F., Progr. Org. Coatings, l, 157 (1972). (4) Volz, H. G., Kaempf, G., and Fitzky, H. G., Farbe und Lack, 78, 1037 (1972); and Xth FATIPEC Congress, 1970, p 107. (5) Weyl, W, A., and Fortand. T., Ind. Eng. Chem., 42, 257 (1950). (6) Egerton, G. S., and Morgan. A. G.. J. Soc. Dyers Colourists, 268 (1971). (7) Khan, A. U., and KearAs. D. R., "Advances in Chemistry Series, No. 77/' American Chemical Soc.. Washington. D. C., 1968, p 143. (8) For example, see Kaplan, M. L., and Kelleher, P. G.. Science, 169, 1206 (1970) . (9) Trozzolo, A. M., in "Polymer Stabilization." Wiley-Interscience, New York, 1972, p 159. (10) Rawls, H. R., and Van Santen, P, J., J. Am. Oil Chemists Soc., 47, 121 (1970). (11) Schroeter, S. H., Appel, R., Brammer, R., and Schenck. G. O., Ann. Chem., 697, 42 (1966). (12) "Worthington Enzyme Manual," Worthington Biochemical Co., Freehold, N. J., 1972. (13) Higgins, R., Foote, C. S., and Cheng, H., "Advances in Chemistry Series, No. 77," American Chemical Soc., Wash ington, D. C., 1968, p 102. (14) Schenck, G. O., Ann. Chem., $84, 156 (1953) ; also, sec Gollnick, K,, and Schenck, G. p., in "1,4-Cycloaddition Reactions," Academic Press, New York, 1967, p 255. (15) Kautsky, H., Trans. Faraday Soc., $$, 216 (1939). (16) Kearns. D. R.. Chem. Rev., 71, 395 (1971) . (17) Ogryzlo, E. A., and Tang, C. W., /. Am. Chem. Soc., 92, 5034 (1970). Vo l . 16, No. 599, De c e mb e r 1974 71 DUPO10001954 S. P. PAPPAS and R. M. FISCHER (18) Irick, G.. Jr., /. AppS. Polymer Sci., IS, 2387 (1972). (19) Pryor, W. A., "Free Radicals," McGraw-Hill, New York, 1966, p 138, (20) Gray, T. J.f McCain, C. C., and Masse, X, G., J. Phys. Chem., 61, 472 (1959) , (21) Clay. H. F., /, 0*7 4x Colour Chemists0 Assoc40, 935 (1957). (22) (a) Winter, G. and Whittem, R. N,, J. Oil & Colour Chemists* Assoc., 33, 477 (1950) ; (b) Morley-Smith, C. T., ibid., 33, 484 (1950). (23) Calvert, J. G., Thcurer. K., Rankin, G. T., and MacNevIn, W. U., J. Am. Chem. Soc., 76, 2575 (1954). (24) Kennedy, D. R.. Ritchie, M.. and Mackenzie, J., Trans. Faraday Soc., 34, 119 (1958). (25) Murley, R. D,, J. Oil r Colour Chemists* Assoc., 45, 16 (1962). (26) Fukurawa, S.. Sander, K. M., and Kwan, T., J. Catal., 21, 364 (1968). (27) Tanaka. K.. and Blyholder, G., J. Phys. Chem., 76, 3184 (1974). (3) Mayeda, E. A., and Bard. A. J., J. Am. Chem. Soc., 95, 6223 (1973). (29) Batilch, D. L.. Drysdale, D. D., and Lloyd, A. C., "Evalu ated Kinetic Data for High Temperature Reactions," Vol. 1, Buttenvorths. London, 1972. (30) Sehested, K., Rasmussen, O. L.. and Fricke, H., J. Phys. Chem., 72, 626 (1968); ako, see Durrant. P. J., and Durrant, B., "Introduction to Advanced Inorganic Chemistry/' 2nd Ed., John Wiley and Sons. Inc., New York. 1970. p 803. (31) Sander, K. M., and Morrison, S. R., Surface Set., 36, 622 (1973). (32) David, K. M. C., Hammond, P. R., and Peover, M. E., Trans. Faraday Soc., 61, 1516 (1965). (33) Vendeneyev, V. I., Gurvich, L. V., Kondrafyev, V, N., Medvedev, V. A., and Frankevich, Y. L., "Bond Energies, Ionization Potentials, and Electron Affinities," Edward Arnold Ltd., London, 1966, p 195. (34) Pimenov, Y. D., Kholraogorov, V. E-, and Terenin, A. N,, Dakl. Akad. Nauk. SSSR, 163, 935 (1965); see Chem. Abst., 63, 14081c (1965) . (35) Vondjidis, A. G., and Clark. W. C.. Nature, 198, 278 (1963). (36) Bevan, H., Dawes, S. V., and Ford. R. A., Spectrochim. Acta, 13, 43 (1958). (37) Kaplan, M. L., and Kcllcher, P. G., /. Polymer Sci., (A-l), 8, 3163 (1970). (38) Carlsson, D. J,, and Wiles, D. M., J. Polymer Sci., Polymer Lett. Ed., 21, 759 (1973) . (39) Ingold* K. U., Chem. Rev., 61, 563 (1961). (40) (a) Hawkins, W. L., in "Polymer Stabilization/' WileyInterscience, New York, 1972, p 2: (b) Shelton, J. R., ibid., L29. plan, M. L., and Kelleher, P. G., J. Polymer Set., (B), ' ' 9, 565 (1971). (42) Carlsson, D. J., Mendenhall, G. D., Suprunchuk, T., and Wiles. D. M.. /. Am. Chem. Soc., 94, 8960 (1972) . (43) Guillory, J. F., and Cook, C. F., /. Polymer Sci., Polymer Chem. Ed., 11, 1927 (1973) . (44) Foote, C. S.. Accounts Chem. Res., 1, 104 (1968). n -i ! i ii | i i i MILDEW DEFACEMENT OF ORGANIC COATINGS (A Sp e c ial Lit e r a t u r e Su r v e y ) la April 1971 chc Paint Research Institute of the Federation sponsored an encounter session between a group of microbiologists and a group of paint scientists on the subject, "Mildew Defacement of Organic Coatings." Die purpose of the meeting was to initiate a research program designed to uncover the fundamentals of the general subject. As a result of the encounter session, a contract was granted by the Paint Research Institute to Bottelle Memorial Institute to conduct a literature survey an dbte subject. Part 1 of the survey covers the biology of Aureobasidium puuvlam pnd contains information to which few paint scientists have had access. Part II, covering the Micro biological Deterioration of Coatings, summarizes the types of deterioration, bacterial and fungal; the microorganisms involved, as well as the painc materials; and control measures. Die survey cites from 709 references and organizes published information into lucid statements of the current understanding of this important area. 132 pages, 814 x 11. |25.00 per copy to Federation members; $35.00 per copy to non-members. Order From: Federation of Societies for Coatings Technology 121 South Broad Street Philadelphia, Pennsylvania 19107 j ] j i * i i 72 Jo u r n a l o f Pa in t Te c h n o l o g y DUP010001955