Document a1BzBGq21n8NX4mnmaG9rqwVN

d^ E. I. DU PONT DE NEMOURS & COMPANY 256 VANDERPOOL STREET NEWARK, NEW JERSEY Serial No. 71-14 Copy No. 6 NEWARK PLANT PIGMENT COLOR RESEARCH REPORT Quinazoline Pigments Period Covered September 1969 to December 1970 (Part Time) FILE: 223 DATE: 6/22/71 rO 4^ CO h i -IU ~ i'= -s. " KN--71-14 Copy No. 6 X. Numerical File 2. Research Office File 223 3. Newark Library File 223 4. M.Hunt, Pigments, Wilmington , 5. w. s. s truve/J , Jacksonyte. R .Hanke ' 6* " R.H.Wetzel, Newport (circ. and File) 7. E. F. Klenke/B. H. Perkins/Library 8. P.J.Monahan, Newark (Vital Records), 9. N.6.Fisher, Central Research Dept.VWilmington 10. Newark Library - for Central Report Index 11. Newark Library - for central Report Index 12. j. Auerbach 13. E.E.Jaffe 14. Extra 15. Extra 16. Extra 17. Extra 18. Extra NEWARK PLANT PIGMENT COLOR RESEARCH REPORT TITLE t Quinazoline Pigments no (Part Time) J^tte Submitted; 5/3/71 Date Released; 6/22/71 Several yellow pigments, derivatives of 2-o-hydroxypheny1quinazoline, and their non-hydrogen bonded phenyl analogs were synthesized. These internally hydrogen bonded molecules were expected to show greater photostability than their non--hydrogen bonded phenyl analogs and comparable photostability to the photostable parent heterocycle 2-o-hydroxyphenyIquinazoline. The hydrogen bonded quinazoline pigments possessed broad visible absorption spectra, which made them tinctorially inferior to the commercial pigments with which they were compared. In terms of lightfastness, the hydrogen bonded quinazoline pigments showed only moderate photostability, a factor which is to date hot well understood. In an attempt to circumvent some of the drawbacks of the quinazoline pigments several related hydrogen bonded quinoline pigments were prepared and evaluated. In one case an improvement in color and photostability was observed. In an area related to the theory of proton transfer in hydrogen bonded molecules several new internally hydrogen bonded quinazoline molecules were synthesized for evaluation as commercial UV absorbers. DUP050082678 OBJECT PERIOD COVERED PATENT STATUS BACKGROUND SYNTHESIS DISCUSSION CONCLUSIONS EXPERIMENTAL PART ANALYTICAL DATA REFERENCES TABLE OF CONTENTS Page 1 Page I Page 1 Page 1 Page 7 Page li Page 21 Page 21 Page 23 Page 25 DUP050082679 OUINAZQLIME PIGMENTS PROGRESS REPORT I. OBJECT To synthesize yellow organic pigments possessing the o-hydroxyphenylquinazoline moiety. The incorporation of this group ing into pigment molecules was expected to produce compounds with comparable photostability to the photostable parent heterocycle 2-o-hydroxyphenylquinazoline. II. PERIOD COVERED! September 1969 to December 1970 (Part Time) III. PATENT STATUS The results of this investigation did not warrant the preparation of a patent application IV. BACKGROUND It has been known for some time that an improvement in photostability results by attaching a hydroxyl group in the ortho position of benzophenone. There has also been observed an improvement in lightfastness in heterocyclic nitrogen compounds by introducing an o-hydroxyphenyl group so that an intramolecularly hydrogen bonded molecule resulted. The increase in photostabi1ity of these molecules is attributed to the following "simplified" energy dissipating photoenolization mechanism; Benzophenones O-hydroxyphenyl quinazoline A large measure of understanding of this mechanism of stabilization came from the theoretical work of J.E, Otterstedt. Simply stated, he postulated that light energy absorbed by the ground state (enol) converted it to its first excited singlet state. Proton transfer from this singlet state could then occur to produce the first excited singlet state of the keto isomer, which could then dissipate its energy in the form of heat by ultimately reverting to its enol ground state. To substantiate this theory Very weak red DUP050082680 -2 shifted fluorescence has been observed from intramoleculary hydrogen bonded benzophenones, triazolines, pyrimidines/ quinolines/ and quinazolines. The observed red shifted fluorescence was then correlated with the Si (ketc?--^ SQ (keto) transition in the emission spectra of these compounds. The relative stability of o-hydroxypheny1 nitrogen heterocycles has been found to be proportional to the change in resonance energy in going from the Si (enol) to Si (keto) states. Thus it has been found that the photostability of various o-hydroxyphenyl heterocycles can be put in the following order. quinazoline y quinoline^ triazine^ pyrimidine pyridine This order is reasonable in view of the increasing resonance stabilization of the molecules shown above. Of the various heterocycles tested the 2- (o-hydroxyphenyllquinazoline system possessed the highest photostability. This finding formed the basis of the attempt to synthesize pigment molecules possessing this photostabilizing heterocyclic moiety. The theory as advanced by Otterstedt, has found practical application in the synthesis of quinazoline UV absorbers. Ideally UV absorbers are colorless materials which absorb strongly in the UV but which are themselves photostable. By the technique of over^coating the UV absorber it has been DUP050082681 demonstrated that these compounds could afford significant protection against harmful radiation and protect otherwise insufficiently stable pigments. One drawback to the quinazoline UV absorbers has been their slight yellow coloration. This is due to the tail end of their UV absorption band. It occurred to us that if the yellowness observed in low molecular weight o-hydroxyphenylquinazolines could be enhanced through conjugation then this might make the basis for a light stable yellow pigment. V. SYNTHESIZED COMPOUNDS There are a number of synthetic routes to the quinazoline nucleus. However, in the course of the work synthetic limitations to their usefulness were encountered. It was discovered quite fortuitously that the reaction between o-hydroxybenzamides and aromatic aminoketones produces the quinazoline nucleus by simple thermal condensation. Below is a proposed reaction mechanism which accounts for the ready interaction of the amide and amine and rationalizes the fact that unsubstituted benzamide reacts only to produce simple amides. The synthesis of the valuable UV absorber 9. is presented DUP050082682 -4 - The probable long lifetime of intermediate A allows the amide - NHs to attach the -C*0 of the benzophenone leading to quinazoline ring formation, whereas in simple benzamides such a stabilized intermediate is not possible. In contrast the strong negative charge initially placed on the -0*0 of bensamide leads to ejection of ammonia and simple amide formation. For example: the reaction of banzamide with 1,5-diarainoanthraquinone yields 1,5-dibenzoylaminanthra- quinone. Under comparable conditions salicylamide yields the heterocycle 3b. The following yellow quinazoline pigments have been prepared. Their structures are assigned on the basis of elemental analysis, IR, and UV spectroscopy. Additional evidence is presented where alternative synthetic routes were successful to substantiate the proposed structure. RR 1 a)R=CH3 b)R*phenyl c)Rortho-hydroxy phenyl (OHP) 3 a)R=*phenyl " b)R*=OHP c)R2-hydroxy-3,5 dichlorophenyl R 4 a.) R*CH3 b)R*phenyl Rx 5. R=OHP Ra*=benzoyl 6 R=OHP DUP050082683 2 R<=OHP I a) Rsphenyl ' b) R*=OHP Essentially colorless compounds prepared to be tested as UV absorbers are shown below: 9 R=phenyl ~ Ri=OHP DUP050082684 6- In the course of the work other related compounds were prepared and are presented helows 11 * R-OSP i 14 R=OHP Ri=benzoyl DUP050082685 Synthesis A search of the literature revealed that compound la and lb were prepared by Plumpe, Belg. pat- G59016. They were reported to be yellows, compound lb was reported to be in fact a strong yellow. Cr207= The synthesis of lc depended on the isolation of naphthalenetetramine. However, difficulty was encountered in preparing the material described in the literature. in order to confirm our isolation of this tetramine, it was converted to the lendwn compound JJ. DUP050082686 Synthesis of 2 Synthesis of 3 3a ^.Renzaldehyfle Synthesis of 4 -8 - Salicvlamide "\7 --o Synthesis of 5 Salif-yl amide ^5 Synthesis of 6 Dilute HgSOA. *7 DUP050082687 Synthesis of 7 9- +6 Nitrobenzene v 7 J'Heat 7" Benzonitrile/ff1' n , 8a saliovlamide/2500 ^ ^ 2 Hvgroxv-benzQnltxiJ?e 8b H^/Heat 7 (Ciba) Chromophthal Red 3B Synthesis of 10 Synthesis of 11 .........i\ jQ reflux ' ~" ^iqlvfne reflux AH. DUP050082688 Synthesis of 12 - 10 - 4 12 12 Synthesis of 14 Synthesis of 15 0 / 'v "7 14 Reflux-----} 15 s<?<fo,um Acetate i6 Acetic Acid 7 Reflux DUP050082689 11 Discussion Compounds la,b,c were prepared to examine their absorption spectra and photostabilities. Compound la shows its highest X max at 381 nm ( . 6,677) and its longest wavelength absorption at 400 ran ( , 398), in DMP solution. Comparing la to the hydrocarbon pyrene is revealing in that the hydrocarbons longest wavelength maximum in DMF solution occurs at 337 ran (c 44,673)^ while in the spectrum of la one does not see any absorption above 300 ran of an extinction coefficient greater then 6,70a Therefore, in comparing la to pyrene one sees a marked drop in extinction of la relative to pyrene and also a bathochromic shift of the longest wavelength maximum. In DMF solution compound lb Rphenyl shows a reasonably sharp absorption at Xmax 421 nm (- 3,136), whereas compound lc displays very broad absorption in the visible, its highest absorption occurring at 430 nm {C 2,043), Comparing compound lb to la one sees a shift to longer wavelength due to additional phenyl conjugation but still a reasonably low extinction, of the longest wavelength maximum. It is noteworthy that lc (R=*0HP) shows a broad absorption in the visible spectrum. This broadness has been noticed in the visible absorption spectra of other ortho- hydroxyphenyl quinazoline pigments. Two reasonable explanations might be advanced to explain this phenomenon. Pirstly, the introduction of the hydroxyl in the ortho position extends the conjugation of the system by forcing the phenyl moiety to lie in the plane of the tetrazapyrene ring. Secondly, due to the possibility of the existence of two distinct conformational arrangements as a consequence of the two sites of hydrogen bonding available to each (OHP) group, we may be seeing the visable spectra of two different molecules in solution. lc That intramolecular hydrogen bonding is very likely present in lc is deduced from the bathochromic shift of lc relative to lb in analogy with the bathochromic shift observed in the spectrum of the hydroxy vs. the methoxy compound shown below. DUP050082690 hmx 316 run (Methyl pentane) A Max 334 nm (Hexane ) Whether the broadness of absorption in the spectrum of lc is due to the presence of an equilibrium between rotomers or is inherent in this type of intramolecularly hydrogen bonded molecules cannot be decided unequivocally. This point will, however, be clarified When the spectra of J5 and 14 are compared. in addition, other evidence which bears on the co-planarity of lc comes from a comparison of the melting point of lb (334C) with that of lc (430C) . The higher melting point of lc is consistent with it being hydrogen bonded, and thus completely planar and more symmetrical. In summary, the introduction of hydrogen bonding in compound lg. results in a broadening of the visible absorption spectrum of this compound relative to the phenyl derivative. The broadness of absorption is an important consideration in that it adversely affects the intensity of color of these types of compounds. The lightfastness of compound lb and lc were evaluated by Fadeometer expostire. Both lb and lc were dispersion milled and both tints (1/25) and masstones were compared against Anthrapyrimidine Yellow. After 300 hrs. exposure lb showed a marked darkening in masstone and fading in tint. Whereas lc showed only moderate darkening in masstone and a slight darkening in tint. By comparison the standard Anthrapyrimidine Yellow showed only a slight fading both in tint and masstone after 300 hours exposure. These effects were more pronounced when both lb and lc were acid pasted and flushed and compared in the usual way by Fadeometer exposure. At this point it was of interest to prepare other quinazolines possessing more intense color and better photostability. Compound 2_ a benz-fused pyrene displayed enhanced longest wavelength absorption relative to lc. Although the absorption was also broad, it attained its highest value at 445 nm ( 3884) in DMF solution. The broadness of the band is indicated by its absorption at 490 nm (. 1776). A lightfastness determination on 2. was performed with crude material. Tint (1/50) and masstone were prepared and compared against Irgazin 3 RLT In masstone 2. displayed comparable lightfastness to Irgazin after 225 hours Fadeometer exposure, but DUP050082691 - 13 in tint if faded at a rate slightly faster than Irgazin. The main drawbacks of 2 are its dullness and relatively low strength# The synthesis of compound 3g was accomplished via the route outlined in the literature. The synthesis of 3b has been accomplished via the reaction between salicylamide and 1,5--diaroinoanthraquinone as outlined earlier. The synthesis of has been accomplished earlier by P.S.Dhaliwal by reacting 1,5-disalicoylaminoanthraquinone under high pressure with ammonia giving a low yield of the desired IR comparison between the material prepared by us and Dhaliwal proved them to be identical. Compound 3b does not melt up to 450 c while 3a shows a melting point of 345C. This result was not unexpected based on our experience with JLc. Below are tabulated the UV spectra of some pertinent compounds. Perylene (DMF) 1,3,7,9 Tetraazapervlene (TCB) 3a (DMF) 3b (DMF) Broad Absorption A max 389 nm v (.12,166) ^ max 402 A max 398, (12,800) A max 395 (7443) A390 . (<L 13,110) (26,863) X max 428 (35,762) Xmax 415 ( 21,800) >.max 434. (24,000) ?> max 400 (7527) Xmax 444 (16,234) A 410 (e 13,110) A 470 ( 5,768) Once again in going from the hydrocarbon perylene to the aza aromatic tetrazaperylene one sees a drop in extinction of the long wavelength absorption and a concomitant bathocromic shift, Similarity the di phenyl tetraazaperylene 3a shows a bathochromic shift and a further drop in extinction vs. tetraazaperylene. Lastly, on going to the di-orthohydroxy perylene 3b one sees an additional drop in extinction and only broad visible absorption. This is a direct parallel to the observation made in the series of compounds la.lb.lc. The lightfastness of yellow compounds 3a, and 3b were determined on materials which were acid pasted and flushed into lithographic varnish. Tint (1/100) and masstones exhibits were prepared and compared against IRgazin 3 RLT. After 71 hours the tint of 3a had completely faded whereas the tint of 3b was only slightly changed. The masstones of 3a., and 3b, however, showed fading of comparable degree, in both cases the standard Irgazin appeared to be superior. The lightfastness of Compound 3b was somewhat improved when non particle size reduced material was used in place of acid pasted material However , in neither case was the 1ightfastness as good as that of compound lc or 2. During the course of our work it was noticed that several pigmentary quinazolines derived from anthraguinone derivatives could undergo apparent chemical reduction with the conventional reagent sodium dithionite. It occurred to us that the chemical reduction observed in solution might also be manifest on exposure to light. DUP050082692 To test this ideal,r a presumably mote difficultly reducible compound 13b was synthesised by the condensation of 1,5-diaminoanthraguinone with o-hydroxyacetophenone. The following is a proposed mechanism for the series of reactions leading to 13; + ,-2H20 *7 13 The new yellow compound J;3: was prepared in high yield and did not melt Up to 425C. Its structure is based on correct elemental analysis and IR Spectrum which is similar to that of 3b. The lightfdstness of this compound was tested on non-particle size re duced material prepared by reerystallization from -chloronaphthalene. Both tint U/50) and masstone exhibits were prepared and compared against Irgazin 3 RLT. After 72 hours exposure this material showed some fading in masstone and a slight break in tint. Whereas Irgazin showed very little change. It is not yet established whether compound 13: shows better lightfastness than 3b. The solution spectrum of compound 13: is not yet available but by rubout it appears to have a DUP050082693 - 15 - sharper spectrum relative to compound 3b. The apparent; sharpening in the spectrum of 13 is probably due to the fact that a single conformation is available to the molecule, whereas in 3b isomerism analogous to that in 1c is present. In the expectation of obtaining a high strength chrpmophore we synthesized the pero-perylene structure 4a*4b. Since the compound R=o-hydroxyphenyl has not been prepared, the interest in 4, has been restricted to an examination of its visible spectrum, compound 4a Ramethyl displays,'! max 410(6 24,650) andslmax 435(4 33/200) in DMP solution. Lightfastness of these compounds have not been determined pending the synthesis of the o-hydroxyphenyl compound for comparison. The synthesis of a relatively more intense yellow was achieved by the preparation of Compound 5. This material snowed *A max 435 nm ( 11,032) in DMP solution. "Purified 5 was acid pasted and flushed into lithographic varnish. After 300 hrs. Fadeometer exposure the tint and masstone of 5 showed nearly comparable lightfastness to irgazin 3 RLT. Although this material displayed reasonably good lightfastness its main drawback is its dull color. The spectrum of this material reveals fairly broad absorption extending as far as 520 nm. In order to see whether the lightfastness of this compound could be further improved and its absorption bands narrowed and shifted to shorter wavelength the related compound 14 has been prepared. This compound shows the expected IR spectrum which is strikingly similar to that of compound 5. previously prepared and characterized. The lightfastness of this compound has been evaluated by Fadeometer exposure. The recrystallized crude material was used to prepare tint (1/100) and masstone exhibits which were compared against irgazin 3 RLT and Anthrapyrimidine yellow. In tint the crude material showed strength comparable to the standards. The new compound is a relatively bright yellow. After 500 hours Fadeometer exposure it showed comparable lightfastness to the standards. The Visible spectrum of this material was determined and found to show a Xmax at 429 nm (C 19,500). The visible absorption spectrum of this material provides possible evidence that the broadness of absorption seen in o-hydroxyphenylquinazo1ine molecules may be due to the super imposition of the absorption spectra of the rotomeric isomers. Below is a table of pertinent data. Spectrum of 14 in DMF Solution Spectrum of 5. in DMF Solution Max 429(6 19,450) " 460(C 9,725) " 480(c 2,650) " 500 (. 90) " 520 (c 0) Max 435 (C. 11,032) " 460 (c 9,193) " 480 (c 4,413) 500 (c 1,323) " 520 (e 514) As the table shows compound J5 has approximately 1/2 * the extinction of compound 14 at its maximum. Thus possibly implying that compound 5. is existing in two distinct rotomeric forms. The sharper drop in extinction in compound 14 also supports this point since this is what would be expected of a molecule confined to only one conformation i.e. it shows a sharper absorption maximum. DUP050082694 16 - In order to test the idea that an ortho-hydroxyphenylquinazoline moiety could improve the lightfastness of an otherwise marginally lightfast pigment/ a direct comparison between structure X and its commercial counterpart Anthrapyrimidine yellow was desirable. The synthesis of X started with the hyrolysis of yellow compound X to give the red amine 6. Although 6 was not obtained in completely pure form its structure is supported by its IR spectrum which shows an amine doublet and a conjugated carbonyl band. Conclusive proof of structure was achieved by its conversion to Compound 3b by simple heating with salicylamide. Compound 7. is an insoluble dull yellow which did not analyze within acceptable analytical limits after acid recrystallization, However, its IR spectrum is very similar to its analog Anthrapyrimidine yellow. The lightfastness of X was determined on crude acid recrystallized material. Tints (1/100) and masstone exhibits were prepared and compared against Anthrapyrimidine yellow. After 72 hrs. Fadeometer exposure this material showed significant darkening in masstone and tint whereas, Anthrapyrimidine yellow remained unchanged. The results were disappointing and did not demonstrate that an o-hydroxyphenylquinazoline moiety could improve the lightfastness of an otherwise insufficiently lightfast pigment. The availability of Ciba chromophthal Red 3B enabled us to prepare yellow compounds 8a,8b. In (DMF) solution the visible spectrum of compound a showed )s max 430 nm (. 8364). In contrast the visible spectrum of the ortho-hydroxypheny1 analog 8b showed only broad absorption, the maximum extending from 420 nm (.4263) to 440 nm ( 4263). Once again we see a broadness of absorption with a concomroitanf drop in extinction coefficient on going from the phenyl analog to the ortho-hydroxyphenyl derivative. The lightfastness of compounds 8a and 8b were determined, compound 8b X=0H showed somewhat better lightfastness than the phenyl derivative when it was evaluated in crude form. However, both compounds showed better lightfastness than compound 3b and comparable photostability to lc and 2, thus suggesting that a greater delocal ization of electrons over a more extended aromatic system is detri mental to the photostability of the quinazolines type compounds. Proton Transfer in Quinazolines In 2- ortho-hydroxypheny$quinazQline if is known that the molecule exits in the conformation C shown below. This is also the conformation from which a proton is transferred on photoexcitation. $ C DUP050082695 *- 17 The alternative to conformation C would he conformation D in which the proton would he transferred to the nitrogen at position three. R in c the loss in resonance energy of the heteroeycle on photo- excitation is smaller than in D. Conclusive proof that conformation C is the preferred one has come from NMR spectroscopy of various substituted quinazolines. It is Known that the photostahilities of 2-ortho-hydroxyphenylazaheterocycles, results from the photoenolization mechanism shown. The greater photostahilities ought to he associated with heterocycles undergoing the least change in resonance energy on proton transfer. 7 It appeared that this point could he demonstrated hy varying the nature of the phenolic portion of the 2-substituted quinazoline. Thus the predicted order of photostability for the following four compounds should be e11 > 9. > 10. DUP050082696 - is - > That is, the molecule which loses the least amount of resonance energy on excitation should he the most photostable It has also been shown that the position of the phenolic .OH signal in the MMR spectrum is proportional to the electron density on Nitrogen #1. This has been corroborated in a series of substituted 2-(orthohydroxyphenyiJquinazolines * Therefore, it is possible that the position of the OH signal in the NMR will be a function of the photostability of these compounds. Compounds Jyo and ll were prepared via the previously described synthesis. It is interesting that the preparation of E could not be achieved by this synthesis using l-hydroxy~2-naphthaiamide. The product of this reaction was DUP050082697 19 only the amide 17, shown below. 17 The position of the OH resonance of compounds 9,10,11,12 are tabulated below. COMPOUND 9 ltf 11 12 SlOH) CDCW(TMS) g (OH)CPCWfTMS) 13.16 ppm 13.66 very broad absorption above 8ppm 13.91 phenol 2.8 ppm $ Naphthol 3.2 The results indicate that the ordering of the signal positions are not as expected. It was expected that the signal for 10 would have appeared below f 13.16 since in this compound the acidity of the OH proton was expected to be lowest. In contrast# the position of the resonance for 12. may be reasonable in view of the fused benzene ring. The broad absorption observed for the OH signal in compound 11 is surprising. This broadness may indicate restricted rotation about the bond connecting the quinazoline and naphthalene portions* of the molecule. This point ought to be demons testable from the UV spectrum of this compound. In a paper by Porte et al# (j.A.Chem.Soc. 82,5057 {I960)),the chemical shifts at infinite dilution of various hydrogen bonded phenols were recorded. They observed that a linear correlation existed between the position of the OH resonance in the NMR and the C=0 frequency in the IR spectrum. Their results showed that internally hydrogen bonded phenols had the OH, signals quite downfield from the non-hydrogen bonded simple phenols. (8-llppm). So in comparing compounds 9# 10# 1.1# and 12 to phenol and {3 Napthol one sees the expected large downfield shift of the .OH resonance. DUP050082698 -- 20 -- The uv Spectra of Compounds 9,10/ and ii are given below. Compound 9 m ii UV Max* 330 nm Sh(350) 310 270 355 315 IjJL. 12,000 10,250 28,700 48,700 10,300 14,250 * in DMF Solution in comparing 9.10. and ii we see a bathochromic shift in the long wavelength absorbtion of 10 and 11 relative to 9, This is reasonable since we are presumably extending the conjugation of the system relative to compound 9,. The lack of absorption bands in 11 vs.IP and the decrease of extinction of 11 is indicative of a decrease in conjugation of 11. However, this argument is not conclusive since, we do not have the compounds lacking the hydroxyl groups for comparison. The availability of the non-hydroxylated derivatives would enable us to differentiate between a substituent effect due to the hydroxyl group and the effects due to conformation. However, the evidence presented above does in fact suggest that compound 11 is twisted. ' Evaluation The following compounds have been incorporated into clear lacquer and drawn down over a clear transparent films 9.10,11.12, and 15. The material 15. was included so as to determine its photostability and relate it to that of the pigments based on this heterocycle such as 13, and 14. After 92 hours the following order of photostabilities were noted. The ratio of the initial absorption Ao at the maximum to the final absorption A are recorded. The order of photostabilities are as follows s Xmax 340 nm X^ax 330 12, A/Ao m 1 9t A/Ao m .97 Xmax 355 Xmax 355 A max 305 11,A/AO m .5 1, A/Ao .3 10,A/Ao .13 Since it was observed that experimental material 12, and 9 had comparable photostability, they were submitted for Fadeometer * exposure for a full 500 hours. The following results were noted. 12, A/Ao = .914 9, A/Ao .894 v Compound 12. actually shows better photostability than 9.. In comparing the photostabilities of 12. to 9. it would appear that free rotation of the phenyl ring in does not contribute to the deactivation of the excited singlet state of jj), Oton? results are somewhat disappointing in that the ordering of photostabilities are not what one could have expected based purely on resonance energy considerations. The fact that compound 10 DUP050082699 - 21 is least stable is reasonable but the fact that compound 11. is not the most stable is unexpected. The observation that compound 15 shows reasonable photostability correlated well with our observation of the related yellow pigment 14. Violet Perinone 16 The availability of 1,4,5,8 tetraaminonaphthalene prompted us to prepare violet perinone 16. The material was acid recrystallized presumably isolating one of the two possible isomers and evaluated for lightfastness. Tints and masstones were prepared and compared against RT-795-D. After 350 hours Fadeometer exposure 16 displayed comparable 1ightfastness to the guinacridone standard. Conclusions Various ortho-hydroxyphenylquinazoline pigments have been prepared. In most cases an improvement in lightfastness is observed with the introduction of an ortho-hydroxyphenyl moiety relative to a phenyl group. The lightfastness of certain derivatives (lc,2,58b) is improved over others. (3b). In terms of color many of these compounds show absorption spectra which are too broad to be of interest tinctorially. This result has been modified in the monoaza derivatives exemplified by compound 14. Experimental The pertinent technical data is to be found in laboratory Note booh 1926K. Synthesis of 3b - NB 1926/100 Into a 4 neck mechanically stirred reaction vessel was charged 78 grams {0.328 mole) commercial 1,5-diaminoanthraquinone and 650.0 grams salicylamide and the mixture blanketed with nitrogen. After the ingredients were well mixed, 6 ml of nitrobenzene was added to avoid plugging the reflux condenser. The reaction mixture Was melted and eventually brought to reflux (250c) . The mixture was homogenous and on continued heating began to evolve water. On heating for about 10 minutes crystals of compound 3b appeared to precipitate out of the melt. After the reaction was complete the mixture was cooled to 150C and 500 cc (DMF) was added. The mixture was refluxed for 15 minutes and the solid filtered hot. The yellow material was washed with DMF and then methanol, yielding 122 grams (84.6%) of 3b. The Solubility of this material is about lg in 2 liters boiling nitrobenzene. The highest quality material can be obtained by acid recrystallization from sulfuric acid followed by solvent recrystall ization . Synthesis of 13 - NB 1926/168 Into a 4 neck mechanically stirred reaction flask was placed 3g 1,5-diaminoanthraquinone and 25cc 2-hydroxyacetophenone. The mixture was blanketed with nitrogen and heated to reflux (220c),at which point the reaction mixture was homogenous. The mixture was refluxed for 24 hours. The heavy slurry of product was diluted with c\ -chloronaphthalene and maintained at 150 for IP minutes, and the DUP050082700 ..* 22 p* solid filtered and washed free of solvent* yielding 5.25 gr of 13.. This material can he recrystallized from c^'chloronaphthalene. its 4e A-f Vm414rtf /^^hlornnaDhiilialene. DUP050082701 Analyses - All compounds the analyses of which are given below, appear in notebook 1926K and their page numbers are recorded below. P.62,86-lb calcd for Cs4HX4N4 :C,80,43; H,3,94; N,15,63 Pound: C/80.2; H,4,4; N,15.8 P.48-1C calcd for Ca4BX4N402: C,73,84,H,3,61,-II, 14,35 FoundT'c, 73, 9;H, 3, 8 ;tf, 14.5 P.94-1 calcd for C28H16N402: C,76,35; H,3.66? N,12,72 Found: C,76,3,; H,3.9,N,12.9 P.71-3a calcd for: C28H16N402; C,76.35, H,3.66;N, 12.72 Found: C,76.0; H,3.8; 12.7. P.100-3b calcd for: ; C, 76.35; H,3.66; N,12.72 Found: C, 76.0; B,3.8; 12.7 P. 163-3(2 calcd for: C^H^Cl^C^ ; C,58.2; H,2.095? N, 9.7; Cl, 24.5 Found: C,56.5; H,2.3; N,8.5; Cl,23.6 P.108-4a calcd for: Cs4HX4N4; C,80.43; H 3.94; N, 15.63 Found 80.5; H,4.1; N,15.1 Plll-4b calcd for; C34H18N4; C,84.63; H,3.76; N,11,61 Found: C,83.0?H,3.7; 8,11.5 P.93-5 calcd for: C28H17N302; C,75.84; H,3.86; N,9,48 Found: C,75.9; H,4.0; N, 9.6. P.116-Crude 6, calcd for: C21H13N3O2; C,74.33; H,3.86; N, 12.38 Found: C,68.3; H,3.9; N,11.5 P.152-7 calcd for: C37H19N504; C,74.37; H,3.2; N,11.72 Found: C,68.1; H,3.3; N,10.6. P.65-pa calcd for: C42H2sN402: C,82.6; H,3.61; N,9.12 Found: C,80.0; H,4.0; N,9.0 P.124--8b calcd for :C42H22lSr404 C,78.1; H,3.40; N,8,66 Found: C/77.2; H,3.5; B,8.2 Compound 9 - has been previously prepared and had acceptable analysis. P.115-10 calcd for; C24H16N20: C,82.74; H,4.63; N,8.04 Found: C,81.6; H,4.7; N,7.8 P.167-11 calcd for: C H 8o0: 0,82.74; 8,4.63; ,8.04 24 16 Found: C,82.8; H,4.7; N,8.05 DUP050082702 24 - P.96-i2 calcd for: C2()H12N20: C,81.077 8,4.08; N,9.45 Found: C,8l.O; H,4.3; N,9.5 P.168-13 calcd for: C--H, N 0o: C,82.18; H,4.14; N,6.39 -- j o 18 2 2 Pound C,82.1; H,4.3; 6.4 P,171-14 calcd for: C29HisN23 s c'78*72* H'4 10 N*6.33 Found: C.78.7, H,4.3; N,6.3 P.183-15 calcd for; C21H15NO:C#84.82; H,5.08; N,4.71 Found: C,84.4; 5.0; M,4.7 P.165-16 The following analysis pertains to either cis to trens. 16 obtained by acid recrystallization of the crude product. calcd for: c'45*45; H/ 0.582; N,8.16; Cl,41.2 Found: C,44.9; H,0.7; N, 8.2; Cl,39.4 P.114-17 calcd for? C H NO ; C, 78.46; H,4.66; N,3.8l X/ J Found: C,78.7; H,4.7; N,3.9 UV Spectra - See KB 1961 la UV Max (DMF) 300 rm (11,574),311(12,324),344(2,577),361(4,967), 375 sh(4,615),381(6,677). lb UV Max (DMF) 293 (81,124), 421(3,136 ) lc UV Max (DMF) 285(83,025) 358(37,739) UV Max (DMF) 254(33,484),276(56,014),343 sb (35,970),362(42,340) 3a UV (DMF) 260(47,785), 272(52,606),344(8,288),395(7,443),400(7,527), 418(12,686),444(16,239) 3b UV Max (DMF) 267 (70,794),320(33,561) 4a UV (DMF) 277 (51,000),305 sh (24,000) ,385 (20,600),410(24,650), 435(33,200) 5 UV (DMF) 265 (50,896); 435(11,032) 8a UV (DMF) 286(83,643), 351 sh (13,334) 8b UV (DMF) 268(32,401),288(38,796),315(34,106),358(14,779) 12. UV (DMF) 266(32,900), sh 272 (31,650), sh 300(18,500) ,sh 342(8760), sh 387(4,880) DUP050082703 * 25 REFERENCES 1. A General Study of Photostability, J. E.Otterstedt, OCR-65-1 2. A General Study of Photostability, J.E.Otterstedt, Orchem RD 65-199 3. Photostability Data For Quinazoline and Pyrimidine Compounds, J.E.Otterstedt, Orchem RD-67-46 4. Emission Studies of Photostable Compounds, J.E.Otterstedt, Orchem RD-67-158 5. Synthesis of Model UV Stabilizers I, R.Pater,Orchem, RD-65-315 6. Synthesis of Model OV Stabilizers II, R.Pater, Orchem,RD-66-49 7. Synthesis of Polymeric OV Photostabilizers Rased on 2-(2-Hydroxyphenyl )-Quinazoline System, R.J.Axelrod, Orchem, RD-68-81 8. Photostable o-Hydroxyphenylquinazolines, R.Pater,J.Heterocyclic Chemistry 1, 1113 (1970). DUP050082704