Ml ‘ 1 llllH‘Nlfl‘l‘llN‘M l x 126 658 THS THE PHOTOLYSIS OF 1-BROMO- 1.2,3,4,5- PENTAPHENYLCYCLOPENTAD {ENE AND 1,2;3,4,5- PENTAPHENYCYCLGWNTAD IEN - 1-0L IN THE ABSENCE AND PRESENCE OF OXYGEN Thesis {or Hm Degree of M. S. WCHIGAN STATE UNIVERSETY Douglas 0. Spry 1953 THESIS r) LIBRAR Y Michigan Stat: Univcrsity "\. C. -a-- MICHIGAN .QTATr.‘ ”NIVERSITY “z CHIGAN THE PHOTOLYSIS OF l-BROMO—l,2,3,4,5-PENTAPHENYLCYCLOPENTADIENE AND 1,2,3,4,5-PENTAPHENYLCYCLOPENTADIEN-l-Ol IN THE ABSENCE AND PRESENCE OF OXYGEN BY Douglas 0. Spry A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE Department of Chemistry 1963 ACKNOWLEDGMENTS The author wishes to express his sincere appreciation to Dr. G. J. Karabatsos for his guidance and counsel through- out the course of this investigation. Acknowledgment is also made to Floie M. Vane and Seymour Meyerson. ii ABSTRACT The possibility that the pentaphenylcyclopentadiene system might react photolytically in the absence of oxygen to yield the triphenylcyclopropenyl cation (I), thus leading to a relatively simple synthesis of the aromatic cyclopropenyl cation, was examined. In addition, the photolysis was studied in the presence of oxygen. ¢ Q ¢:>(¢ , Brigg , , Y Y a” g The photolysis of l—bromo—l,2,3,4,5—pentaphenylcyclo— 0—fl -9— Y Ye sat—0 I L pentadiene in ether, in the absence of oxygen, resulted in 1.2.3.4,5-pentaphenylcyclopentadiene (34.8%). Neither diphenylacetylene nor the triphenylcyclopropenyl cation were detected. Q : (D Q 0 Av: ether \ Q5 9 43 ¢ Br N2 Q H Douglas 0. Spry Under the same conditions l,2,3,4,5—pentaphenylcyclopentadien- l-ol did not lead to the triphenylcyclopropenyl cation. When the photolysis was carried out in anhydrous cyclohexane in the presence of oxygen l,2,3,4,5-pentaphenyl- cyclopentadien—l-ol reacted to give trans-dibenzoylstilbene and benzoic acid. Infrared evidence suggests that under the same conditions l-bromo-l,2,3,4,5—pentaphenylcyclopentadiene gives benzoyl bromide. ¢ ” o 4) OH (P/ _ \Q + \OH TABLE OF CONTENTS Page INTRODUCTION . . . . . . . . . . . . . . . . . . . . 1 RESULTS . . . . . . . . . . . . . . . . . . . . . . . 5 l. Photolysis of l—Bromo-l,2,3,4,5-Penta— phenylcyclopentadiene in Anhydrous Ethyl Ether in the Absence of Oxygen . 5 2. Photolysis of l-Bromo-l,2,3,4,5-Penta- phenylcycloPentadiene and l,2,3,4,- 5-Pentaphenylcyclopentadien—l-ol in Anhydrous Cyclohexane in the Absence of Oxygen . . . . . . . . . . . 6 3. Photolysis of l,2,3,4,5-Pentaphenyl- cyclopentadien—l-ol in Anhydrous Ethyl Ether in the Absence of Oxygen . . . . . . . . . . . . . . . . 6 4. Photolysis of l,2,3,4,5-Pentaphenyl- cyclopentadien—l-ol in Anhydrous Cyclohexane in the Presence of Oxygen . . . . . . . . . . . . . . . . 7 5. Photolysis of l-Bromo-l,2,3,4,5—Penta- phenylcyclopentadiene in Anhydrous Cyclohexane in the Presence of Oxygen . . . . . . . . . . . . . . . . 8 6. Lithium Aluminum Hydride Reduction of 1,2,3,4,5—Pentaphenylcyclopentadien- 1~ol ‘. . . . . . . . . . . . . . . . . 8 7. Attempted Preparation of the Tosylate of 1,2,3,4,5-Pentaphenylcyclo- ' pentadiene . . . . . . . . . . . . . . 8 iii DISCUSSION EXPERIMENTAL SUMMARY . REFERENCES iv Page 17 35 36 LIST OF FIGURES Figure Page 1. Jablonski Diagram . . . . . . . . . . . . . . lO 2. Infrared Spectrum of l,2,3,4,5-Pentaphenyl- cyclopentadien—l-ol . . . . . . . . . . . . 21 3. Infrared Spectrum of l—Bromo-l,2,3,4,S-Penta- phenylcyclopentadiene . . . . . . . . . . . 23 4. Infrared Spectrum of l,2,3,4,5-Pentaphenyl— cyclopentadiene . . . . . . . . . . . . . . 25 5. Hanovia Type S, Immersion Photolysis Apparatus . . . . . . . . . . . . . . . . . 26 6. Hanovia Type SH Photolysis Apparatus . . . . 27 INTRODUCTION Ziegler and Schnell (l) were the first to prepare 1, 2,3,4,5-pentaphenylcyclopentadien-l-ol (I) and l-bromo-l,2,3, 4,5—pentaphenylcyclopentadiene (II). S. ¢ . <1>

,3, 4» £? (I) (II) 100% 100% They treated the bromo compound (II) with metallic silver in benzene and obtained a violet colored solution. The authors felt that this might be the free radical (III). (III) Molecular weight determination indicated no dimerization. Similarly a colored solution was obtained when Ziegler (1) treated the alcohol with concentrated sulfuric acid. On addition of water to the colored solution he obtained the bispentaphenylcyclopentadienyl ether (V). d’fia __._,_. Q n 96%HSO OH b e’ (I) (IV) (V) From spectroscopic studies Bloom and Krapcho (2) concluded that the colored species was indeed the cation and that possibly it was stabilized by delocalization around the cyclopentadienyl ring. (VI) In View of the fact that both the radical and the cation do not conform to Huckel's rule and, therefore, would not be particularly stable, Breslow (3) reviewed the work of Bloom and Krapcho and found that the following reaction took place. He concluded that the spectrum for the cation was simply a mixture of the spectra of (VII) and (VIII). In the photolysis of 1,2,3,4,5-pentaphenylcyclopenta- diene system it was thought that the following reaction might be preferred over those leading to the formation of the cation or the free radical. ¢ L @b 1 a) Y (x) (XI) 0—181 Kl + V 55-(3 Breslow (4) first synthesized the triphenylcyclopropenyl cation by reacting diphenylacetylene with phenyldiazoacetoni— trile to obtain 1,2,3-triphenylcyclopropenyl cyanide. From the cyanide he then prepared various derivatives, e.g. the methyl ether, the bromide, the alcohol and the carboxylic acid. It was of interest to study the photolysis of (X) in the presence of oxygen. In the photolysis of tetracyclone Bikalis and Backer (5) observed the following: - _ \(PDCIII) (XIV) 9. 0' ¢ ’/,> 0‘ \. .+_ C:<) 4) o o ~o . J (XV) (XII) was suggested as the intermediate. The object of this research was to identify the products of the photolysis of l-bromo-l,2,3,4,5-pentapheny1- cyclopentadiene and 1,2,3,4,S-pentaphenylcyclopentadien-l-ol both in the presence and absence of oxygen, and to suggest mechanisms for their formation. RESULTS Photolysis of l—Bromo-1,2,3L4,5-Pentaphenylcyclopentadiene in Anhydrous Ethyl Ether in the Absence of Oxygen. Two compounds besides recovered starting material resulted from the photolysis of l-bromo-1,2,3,4,5- pentaphenylcyclopentadiene in anhydrous ethyl ether in the absence of oxygen. The first compound was identified by I.R., melting point, carbon-hydrogen analysis and mass spectroscopy to be l,2,3,4,5-pentaphenylcyclopentadiene (VII). The melting point was 250-2520; the literature value (3) is 250°. Three different runs gave hydrocarbon yields of 34.0%, 34.8% and 38.8%. Synthesis of the compound yia the Grignard reaction on 1-bromo-l,2,3,4,5- pentaphenylcyclopentadiene followed by mixed melting point indicated no depression. The second compound was identi- fied by I.R., n.m.r. and mass spectroscopy to be o-di(2-ethylhexyl)phthalate (XVI), a plasticizer origin- ating from the tygon tubing connector used in the gas disperser. Photolysis of an ether solution containing cuttings of tygon tubing gave o—di(2-ethylhexyl) phthalate. O CH -CH 2 3 CH - I I 3 fHZ E-O-CH -CH-(CH2)3—CH 2 3 CH3 (CH2)3-CH-CH -O- \O 2 (XVI) (II) (VII) Efforts to identify alph -bromoethyl ether or other bromo ethyl ether products failed. Photolysis of l-Bromo—l,2,3,4,5-Pentaphenylcyclopentadiene and 1L2,3,4,S—Pentaphenylcyclopentadien—l-ol in Anhydrous Cyclohexane in the Absence of Oxygen. Both compounds were unreactive under these conditions; however, a small amount of o-di(2-ethylhexyl) phthalate was present due to the tygon tubing. Photolysis of 1,2,3,4,5-Pentaphenylcyclopentadien—l-olgin Anhydrous Ethyl Ether in the Absence of Oxygen. Photolysis of the alcohol under these conditions resulted mainly in o-di(2-ethylhexyl)phthalate which again originates from the tygon tubing. Lithium aluminum hydride reduction of an ether solution of the impure o-di(2-ethylhexyl)phthalate resulted in two products. The first compound was identified by I.R. and mixed melting point to be 1,2,3,4,S—pentaphenylcyclopentadiene (VII). The second compound was identified by melting point and mass spectroscopy to be 2,3,4,5-tetraphenylfuran (XVII). 0 I, 0 ¢\ (XVII) It is not known whether the 2,3,4,5—tetraphenylfuran resulted directly from the photolysis of the alcohol in anhydrous ethyl ether or from the reduction of some other photolysis product. Photolysis of l,2y3y4,5-Pentaphenylcyclopentadien-l-ol in Anhydrous Cyclohexane in the Presence of Oxygen. Two compounds resulted from the photolysis of 1,2,3,4,5-pentaphenylcyclopentadien—l—ol in anhydrous cyclohexane in the presence of oxygen. The first compound was identified by its melting point and I.R. as trans— dibenzoylstilbene (XIII). The melting point was 229.5- 230.0°; the literature value (5) is 229.6-230.0°. The I.R. was similar to the I.R. reported (5). The second compound was identified as benzoic acid from the I.R. and mixed melting point. Photoly§is of l—Bromo—l,2,3,4,5-Pentaphenylcyclopentadiene in Anhydrous Cyclohexane in the Presence of Oxygen. There was insufficient sample for a thorough product study; I.R. studies indicate the presence of benzoyl bromide (See experimental). None of these reactions gave diphenyl acetylene or the salt of the triphenylcyclopropenyl cation. Lithium Aluminum Hydride Reduction of l,2,3,4,5-Penta— phenylcyclopentadien-l—ol. The lithium aluminum hydride reduction of an ether solution of the alcohol resulted in l,2,3,4,5-pentaphenyl- cyclopentadiene (VII). Attempted Preparation of the Tosylate of 1L2L3L4,5-Penta- phenylcyclopentadiene. The preparation of the tosylate of l,2,3,4,5-penta- phenylcyclopentadiene yia the tosyl chloride according to the procedure of Krapcho and Bensen (6) failed. DISCUSSION The initial Step in the photolysis involves the absorption by the pentaphenylcyclopentadiene of radiant energy. From the normal ground state singlet the molecule can be promoted either to an excited singlet state, that is singlet-singlet transition, or to a triplet state, that is singlet-triplet transition. An excited molecule may undergo (7) the following deactivating secondary processes: Transfer some or all of its energy to another molecule. Become an ion through some electron-transfer process. Split into two fragments or radicals, in whiCh the electron spin momentum is not zero, but the charge is. React in its excited state with another molecule. Re-emit the absorbed quantum as fluorescence or phosphorescence. The possible paths for absorbing and emitting energy are summarized in Jablonski's diagram (7). It is possible that the blood red color formed in the photolysis of the bromo compound in anhydrous diethyl ether and anhydrous cyclohexane under nitrogen is due to the presence of either the cyclopentadienyl radical,or molecular 9 10 7L T3 S 2 It T2 F G S1 _ T. l A JABLONSKI—DIAGRAM singlet— singlet B. Fluorescence, internal conversion solvent singlet- triplet quenching excited singlet -triplet C. Phosphorescence, internal transition conversion, solvent quenching triplet- triplet G. Internal conversion ll bromine, or an excited state of the molecule. The usually fast reaction of molecular bromine with solvents having active hydrogens available in the presence of light discounts the possibility that the color is due to molecular bromine. It could not be due to fluorescence of some excited molecular species because of the persistence of the color after the reaction was stopped. Although we cannot rule out entirely the possibility of a triplet diradical species, the fact that the color is absent in the photolysis of the alcohol argues against such a species. We ascribe, therefore, the color to the pentaphenylcyclopentadienyl radical. In the case of the bromo compound the weakness of the carbon-bromine bond and the relative stability of the bromo radical result in rupture of the bond. In the case of the alcohol the carbon-oxygen bond is strong enough to prevent its rupture under the experi- mental conditions. Once the pentaphenylcyclopentadienyl radical is formed it can abstract irreversibly (the penta- phenylcyclopentadiene is insoluble in ether) a hydrogen from the alpha position of diethyl ether. The alpha hydrogen of ethers are known to be susceptible to free radical attack. 12 Excited State ———-s Q + Br- ,,O \ CH2_ CH3 The pentaphenylcyclopentadienyl radical apparently is fairly stable because of the existence of the blood red color for three to five minutes after the reaction is stopped. Molecular orbital calculations (8) give 2.473, 1.855; and 1.248 for the cyclopentadienyl anion, radical and cation respectively. The mechanism of the photo-oxidation of condensed ring hydrocarbons is still a subject of controversy. Since oxygen is paramagnetic, it was suggested by Terenin (9) that it could effect a spin reversal in the singlet excited state, to yield the triplet state. Porter and Windsor (10) have shown that oxygen quenches triplet anthracene very strongly in hexane solution; however, the initial yield is independent of the oxygen concentration thus indicating initial l3 formation of the triplet state without the aid of oxygen. If this is true in our studies the excited state leading to the pentaphenylcyclopentadienyl radical could be a triplet state (XVIII). The reaction of the triplet state molecule with oxygen could form an intermediate diradical (XIX) which cyclises to yield the transannular peroxide (XX). This peroxide is analogous to the one (XII) suggested by Bikalis and Backer (5). o 0 I (XIX) l A (XX) 14 The peroxide then decomposes to yield_gi§-dibenzoylstilbene (XIV) and the carbene (XXI). Apparently under the experi— mental conditions the.gig-dibenzoylstilbene isomerizes to the Eggng-dibenzoylstilbene (XIII). The carbene could react with oxygen to yield the peroxy-diradical which then decomposes to give the corresponding benzoic acid and benzoyl bromide. / I) ‘ 0 \Y \0' (Br) J‘ (0H) \ ,/Y 6? Trevay and Brown (11) were the first to report the hydrogenolysis of the carbon-oxygen bond by the addition of lithium aluminum hydride at the reflux temperatures of ether. Sonntag, Linden, Backer and Spoerri (12) later gave the following results on the reduction of tetracyclone: 15 OH Tetracyclone was added to excess LiAlH4 at 0-100. Excess LiAlH4 was added to tetra- cyclone at room temperature. Tetracyclone was added to excess LiAlH4 and refluxed. The authors claim that the hydrocarbon can result from two different sources. 1. The allylic ketone can be reduced to an allylic alcohol intermediate which is dehydrated to the hydrocarbon during acid hydrolysis. 2. Hydrogenloysis of 2,3,4,5—tetraphenylcyclopentadien— l-ol.‘ :é—A . " <:E.:\/::}: The reaction of 1,2,3,4,5-pentaphenylcyclopentadien— l-ol by excess lithium aluminum hydride at reflux temperatures of anhydrous ether led to l,2,3,4,5-pentaphenylcyclopentadiene. The results indicate that the second mechanism proposed by Sonntag, Linden, Backer and Spoerri (12) is valid, although they do not exclude the first. I. EXPERIMENTAL Materials A. Ethyl Ether Anhydrous, C.P. diethyl ether (Merck) was allowed to stand over sodium wire for two days. It was then distilled from lithium aluminum hydride directly into the reaction flask. Cyclohexane Cyclohexane, C.P. grade, was purified by shaking with concentrated sulfuric acid and by washing with water several times. It was dried over anhydrous sodium sulfate and then stored over sodium wire. The material had a transmittance over 98% at 230m” . It was distilled from lithium aluminum hydride directly into the reaction flask. C. Oxygen Oxygen gas was dried by passing it through concen- trated sulfuric acid, potassium hydroxide pellets, and Drierite. l7 18 D. Nitrogen Prepurified nitrogen was bubbled through a gas train consisting of the following (13): 1. Two solutions of Fieser's reagent (l4). 2. A solution of saturated lead acetate. 3. Concentrated sulfuric acid. 4. Potassium hydroxide pellets. 5. Drierite. After passing through the six gas trains the nitrogen was assumed to be dry and free of oxygen. 2,3,4,5-Tetraphenylcyclopentadien—l-one (Tetracyclone) C.P. grade tetraphenylcyclopentadien—l-one was obtained from the Aldrich Chemical Company. It was deep purple in color and melted at 219—2200, the value cited by Ziegler (1). It had the following spectra: I.R. in carbon tetrachloride, 3.28m, 5.12w, 5.32w, 5.35w, 5.79s, 6.70m, 6.29m, 7.38m, 7.68m, 9.15bs, 9.73m, 10.9lw, 11.18w, 14.165, 14.463. 1 max 95% ETOH 263.04..” (620,596), 336.54.”..(é 6,883). N.m.r., singlet atr =3.9 in carbon tetrachloride. l9 l,2,3,4,5-Pentaphenylcyclopentadien-l—ol 1,2,3,4,5-Pentaphenylcyclopentadien-l—ol was prepared by reacting tetracyclone with phenyl magnesium bromide. A seven-fold molar excess of the Grignard reagent over the ketone was required. On recrystal- lization from ethyl ether— ligroin yellow crystals melting at 176-1780 were obtained in 53% yield. The following is a typical run: In a two liter, three neck reaction flask was placed 6.19 grams of magnesium turnings equipped with a calcium chloride drying tube. The flask was then heated with a flame to dryness. After cooling to room temperature 45 ml. of anhydrous ethyl ether and 27.2 ml. of bromobenzene was added. After the re— action had started 75 ml. of anhydrous ethyl ether was added and the reaction was cooled occasionally. When the bubbling ceased the solution was refluxed for one hour. After cooling, 14.00 grams of tetra— cyclone, dissolved in anhydrous ethyl ether, was 2O slowly added. After a mild reflux of ten minutes the yellow solution was allowed to stand for two hours. Following the normal.work up of the Grignard reaction the crude alcohol was recrystallized four different times from ether Iigrdin. 1,2,3,4,5-Pentaphenylcyclopentadien—l—ol had the following spectra (Figure 2 shows the I.R. 95% ETOH .. spectrum): A max 248.34.,” (6 29,063), 355.0mu cyclohexane (é 7,648), shoulder at 26947.“, A 249.041)» max (6 31,697), 356.5mv.(6'6,705), shoulder at 271.7~mM- N.m.r. in carbon tetrachloride:f' = 3.00 (singlet), 3.71 (broad singlet); the ratio of the two peaks was 24.4/l.0 respectively. G. 1-Bromo-1,213,4,5-Pentaphenylcyclopentadiene In 5.0 ml. of glacial acetic acid was placed 2.0 grams of 1,2,3,4,5-pentaphenylcyclopentadien-l-ol. Anhydrous hydrogen bromide was bubbled through the solution until the formation of an orange precipitate. 21 .mpfluoanomnuwu :oQHmo CH HOIH Icmflpmucmmoaomuahcmgmmucomlm.¢.m.N.H mo Esuuommm UwumumcHll.m musmflm 22 The solution was allowed to stand for ten minutes at room temperature and then it was cooled, filtered, and washed with ligroin. A quantitative yield of the bromide (2.3 grams) melting at 186-1870 resulted. It had the following spectra (Figure 3 shows the I.R. 95% ETOH I spectrum): A. 249.34.,“ (6 28,490), 350.573M(é5,609), max cyclohexane )1 max250.04%,(6 27,146), 352.0mv4(é 2,914). N.m.r. in carbon tetrachloride:r~ =3.02(singlet). l,2,3,415-Pentaphenylcyclopentadiene In a 500 m1., three neck reaction flask, equipped with a calcium chloride drying tube was placed 1.0 grams of magnesium turnings. The flask was heated with a flame to dryness. After cooling to room temperature 5 m1. of anhydrous tetrahydrofuran, .5 g of 1-bromo-1,2,3,4,5-pentapheny1cyclopentadiene and three drops of ethylene bromide was added. After the reaction had started 8 m1. of anhydrous tetrahydro- furan was added. When the bubbling ceased the solution was refluxed for one hour. Upon cooling the solution was hydrolyzed with 10% sulfuric acid. Purification gave .40 grams of l,2,3,4,5—pentaphenyl— cyclopentadiene (94.2%) melting at 250—2520. It had 23 ma HH (anhoahcmnmmucmmlm.v.m.N.HIOEOHQIH mo Esuuommm UmHMHMCHII.m onsmflm I h m _ . .mpflmasmflp conumo GA wcmflpmucmm 24 the following spectra (Figure 4 shows the I.R. )1 cyclohexane spectrum): 245.64MA(é 31,863) , 341.047,,“ max (6 9,698). I. Preparation of the Tosylate of 112L3L4,5-Pentaphenyl- _gyclopentadiene. To .27 grams of pure tosyl chloride dissolved in 3 m1. of anhydrous pyridine was added .5 grams of 1,2,3,4,S-pentaphenylcyclopentadien—l-ol. The solution was allowed to stand at 00 for five days. It was then treated with ice water followed by ether extraction. The ether solution was then treated with 5% HCl, diltue NaHC03, water and anhydrous sodium sulfate. 1,2,3,4,5-Pentaphenylcyclopentadien-l-ol (.45 grams) was recovered. 2.Apparatus The apparatus used in the photolysis experiments are shown in figures 5 and 6. 3. Procedure The sample was placed in the apparatus and the apparatus was dried in an oven at 1100 for two hours. It was then removed and connected to the gas train and purged for ten minutes with nitrogen. The gas 25 ea A .HmM EH ocmflpmucomoHohoHmconmmvcomlm.¢.m.N.H mo Eduuommm pmnmumcHll.v musmflm 26 r— 1 <:::>_ MERCURY.LAMP PYREX FILTER < > h,“ H 0 /" *‘\ 2 \\0 k~_ -J_____ v” H 0 GAS VYCOR COOLING SYSTEM \\u // / GAS I \\ / TYGON CONNECTOR - “no.-. ‘ ea \EE: ;:::> A GAS DISPERSER .3 / STIRRING BAR Figure 5.--Hanovia type S, 200 watt, water cooled, vycor immersion apparatus equipped with a quartz filter and a magnetic stirrer. 27 H20 , .5 H20 (-——.:Jw ) \:::::D DRYING TUBE ti:fi C5 :9 GAS PYREX FLASK SOLUTION MERCURY LAMP Figure 6.-—Hanovia type _SH 100 watt. 28 was disconnected and the solvent was then distilled from lithium aluminum hydride directly into the reaction flask. Gas was then bubbled through the solution for ten minutes, the reaction was started, and a positive pressure was maintained throughout the photolysis. 4. Photolysis Experiments A. Photolysis of l-Bromo—l,2,3,4,5—Pentaphenylcyclo— pentadiene in Anhydrous Ethyl Ether under Nitrogen. A 1.67 X 10-2 molar solution of the bromo com- pound in anhydrous diethyl ether under nitrogen was photolyzed using the immersion apparatus shown in figure 5. A color change of yellow to blood red resulted after three minutes of photolysis. On stopping the reaction the color changed from blood red back to yellow after approximately three to five minutes. I.R. indicated no detectable reaction pro- ducts. After one hour of photolysis white crystals began to precipitate. After nineteen hours the solution was still blood red, but the reaction was stopped and the solution was filtered. The precipitate (34.8%) was recrystallized from benzene and 95% ethanol to give white crystals, m.p. 250-2520, parent peak in 29 mass spectrum at 447. The compound was identified as 1,2,3,4,5—pentapheny1cyclopentadiene. Anal. caICd' for 935H26: c, 94.12; H, 5.88- F0und= c, 93.64; H, 6426. The ether was evaporated under vacuum from the remaining solution and a brown viscous material re- sulted. Chromatographic separation of this material using neutral alumina (Woelm neutral alumina obtained from Alupharm Chemicals, New Orleans, La.) gave three products: unreacted l-bromo-l,2,3,4,5-pentaphenyl— cyclopentadiene (eluted with carbon tetrachloride), l,2,3,4,5-pentaphenylcyclopentadiene (eluted with anhydrous ether), and a carbonyl compound (eluted with methanol). The carbonyl compound was a viscous yellow liquid having the following spectra: I.R. in carbon tetrachloride, 3.44m, 5.803, 6.87w, 7.27w, 7.923, 8.93m, 9.37w, in carbon disulfide, 3.423, 5.833, 7.27m, 7.93, 8.963, 9.373, 9.63w, 10.5bw, 13.55bm, 14.3bw; smear, 3.413, 5.813, 6.22w, 6.30w, 6.81m, 7.24m, 7.863, 8.943, 9.343, 9.6lw, 13.51bm, 14.24bm. cyclohexane ),max 274'5*U*‘ N.m.r. in carbon tetrachloride: r'=2.5 (A2B system), 5.87 (Doublet, J = 4.8 cps), 2 30 9.0 (multiplet). The carbonyl compound was identified as o—di(2-ethylhexyl)phthalate. Photolysis of l-Bromo—l,2,3,4,5-Pentaphenylcyglo- pentadiene in Anhydrous Cyclohexane under Nitroggg. A 1.57 X 10.3 molar solution of 1-bromo-l,2,3,4,5— pentaphenylcyclopentadiene in anhydrous cyclohexane under nitrogen was photolyzed for twenty-two hours using the Immersion apparatus. The blood red color developed very slowly and remained three to five minutes after the photolysis was stOpped. The cyclo- hexane was evaporated under vacuum. A reddish solid compound was obtained which upon chromatographing yielded mainly unreacted bromide and a small amount of o-di(2-ethy1hexyl)phthalate; I.R. in carbon tetra— chloride, 3.303, 5.17w, 5.37w, 5.82m, 6.27bm, 6.743, 6.943, 7.30w, 7.97bs, 9.38m, 9.75m, 11.00m, 14.6bs. Photolysis of l,2,3,4,5-Pentaphenylcyclopentadien-l—ol in Anhydrous Cyclohexane under Nitrogen. A 3.28 X 10-3 molar solution of l,2,3,4,5—penta- phenylcyclopentadien-l—ol in anhydrous cyclohexane under nitrogen was photolyzed for fourteen hours. No color change was observed. The solvent was evaporated under vacuum. I.R. indicated the presence 31 of unreacted alcohol and o-di(2-ethy1hexyl)phthalate; I.R.‘in carbon tetrachloride, 2.85m, 3.303, 3.46w, ‘5.18w, 5.33w, 5.60w, 5.81m, 6.05w, 6.283, 6.733, 6.943, 8.0bw, 8.93bw, 9.43bw, 9.78m, 10.30w, 10.93bm. 14.35bs. Photolysis of ly2L344y5—Pentaphenylcyclopentadien—l—ol in Anhydrous Ethyl Ether under Nitrogen. A 2.02 X 10—2 molar solution of 1,2,3,4,5—penta— phenylcyclopentadien—l-ol in anhydrous ethyl ether nitrogen was photolyzed for fifteen hours using the immersion apparatus. No color change resulted. The solvent was evaporated under vacuum. The I.R. indi— cated the presence of unreactive alcohol and o-di(2-ethylhexyl)phthalate; I.R. in carbon tetra- chloride, 2.93m, 3.053, 5.18w, 5.37w, 5.833, 6.26m, 6.35m, 6.90bs, 9.5bs, 10.5m, 10.9m, 14.53, N.m.r. in carbon tetrachloride; P'= 2.5(A2B system), 3.0(pheny1), 2 5.87(doublet, J:4.8cps), 9.0(multip1et). A chromatographic separation using both neutral alumina and charcoal was unsuccessful. Solvent ex— traction using pentane resulted in a fair separation of o-di(2—ethylhexy1)phtha1ate from 1,2,3,4,5-penta- phenylcyclopentadien—l-ol. A lithium aluminum hydride 32 reduction of the pentane extract, that is the o-di(2-ethylhexyl)phthalate contaminated with 1,2,3,- 4,5-pentaphenylcyclopentadien-l-ol, resulted in white crystals melting at 250-2520 (1,2,3,4,5-pentaphenyl- cyclopentadiendfand an oil. White crystals, m.p.l72— 1730, parent peak in mass spectrum at 372, resulted from treatment of the oil with 95% ethanol. The com- pound was identified as 2,3,4,5-tetraphenylfuran. Anal. Calcd. for C28H200: C, 90.40; h, 5.43; 0, 4.30. Found: C, 86.57; H, 5.99; 0, 7.44. 0-di(2—ethylhexyl) phthalate was finally purified by molecular distil— lation of the pentane extract. Photolysis of 1,2,3,4,SfPentaphenylcyclopentadien-l-ol in Anhydrous Cyclohexane under Oxygen. The Hanovia type SH_photolysi3 apparatus was used to photolyze l,2,3,4,S-pentaphenylcyclopentadien— l-ol. A 2.5 X 10-2 molar solution of l,2,3,4,5-penta— phenylcyclopentadien—1-ol was photolyzed at reflux temperature under oxygen for twenty hours. No color change was evident. The solvent was evaporated using vacuum and a viscous yellow material resulted; I.R. (smear); 2.95-4.40bs, 4.68w, 5.30w, 5.95bs, 6.283, 6.653, 6.933, 7.08w, 9.85w, 10.75w, 13.28bs, 14.25bs. 33 Crystallization from ethyl ether and ligroin resulted in white crystals, m.p. 229.5-230.0, I.R. (mull), 3.503, 6.04m, 6.37w, 6.94m, 7.33w, 7.86w, 8.02m, 9.15w, 9.82w, 13.09m, 13.31m, 14.17m, 14.38m. The compound was identified as Egagg-dibenzoylstilbene. Extraction of the resulting solution with 10% sodium hydroxide and ether, acidification then of the sodium hydroxide solution followed by ether extraction and recrystallization of the solid from water resulted in white crystals melting at 121.0—121.5. This compound was benzoic acid. Photolysis of l-Bromo-l,2,3,4,5-Pentaphenylcyclopenta- diene in Anhydrous Cyclohexane under Oxygen. 4.95 X 10-3 molar solution of 1—bromo—l,2,3,4,5- pentaphenylcyclopentadiene in anhydrous cyclohexane under oxygen was photolyzed for twenty—five hours using the Hanovia type SH apparatus. The photolysis was done at reflux temperature and no color change was observed. The cyclohexane was evaporated using vacuum and a brown oil resulted with a penetrating odor; I.R. (smear), 3.453, 5.873, 6.883, 7.33w, 7.60w, 7.98bm, 9.8Sbm, 9.79w, 10.45bm, 11.25w, 12.0w, 14.08bm. Hydrolysiscfifthe oil with 10% sodium hydroxide followed 34 by acidification resulted in another oil; I.R. in carbon tetrachloride, 3.5vb3, 5.903, 7.23m, 7.60m, 7.803, 8.53m, 9.0bm, 9.75m, 10.75b3, 12.5bm, 13.20bm, 14.203. I.R. of the ether extract (in carbon tetra— chloride), 3.433, 5.833, 6.893, 7.85bs, 9.55b3, 11.4w, 13.20bw. There was insufficient sample for further product study. SUMMARY It was experimentally determined that neither l,2,3,— 4,5—pentaphenylcyclopentadien—l-ol nor l-bromo-1,2,3,4,5— pentaphenylcyclopentadiene react photolytically in the absence of oxygen to yield the triphenylcyclopropenyl cation. The photolysis of l-bromo-l,2,3,4,S-pentaphenylcyclopentadiene in anhydrous ethyl ether in the absence of oxygen gave l,2,- 3,4,5-pentaphenylcyclopentadiene (34.8%). The product is explained by a free radical mechanism. The photolysis of 1,2,3,4,5-pentaphenylcyclopentadiene—l—ol under similar conditions did not result in the formation of l,2,3,4,5- pentaphenylcyclopentadiene. Both the alcohol and the bromo compound failed to react when the photolysis was carried out in anhydrous cyclohexane in the absence of oxygen. The photolysis of l,2,3,4,5-pentaphenylcyclopentadien- l—ol in anhydrous cyclohexane in the presence of oxygen gave itgagg-dibenzoylstilbene and benzoic acid. A free radical mechanism is suggested involving the formation of an inter- mediate transannular peroxide. 35 10. 11. 12. 13. 14. REFERENCES von Ziegler and B. Schnell, Ann, 445, 266-282 (1925). . M. Bloom and A. P. Krapcho, Chem. and Ind., 882 (1959). . Breslow and H. W. Chang, J. Am. Chem. Soc., 8;, 3727-3728 (1961). Breslow and C. Yuan, J. Am. Chem. Soc.,._Q, 5991— 5994 (1958). M. Bikales and E. T. Backer, J. Org. Chem., 21, 1405-1407 (1956). P. Krapcho and M. Bensen, J. Am. Chem. Soc., 84, 1036 (1962). . P. Simons, Quart. Rev., 1;, 3 (1959). D. Roberts, Notes on Molecular Orbital Calculatigns, p. 129, W. A. Benjamin, Inc., New York, 1962. Terenin, Acta Physicochim. U.R.S.S.,.1§, 210 (1943). Proter and M. W; Windsor, Proc. Roy. Soc., A, 245, 238 (1958). W. Trevay and W. G. Brown, J. Am. Chem. Soc., 11, 1675 (1949). O. Sonntag, S. Linder, F. I. Becker and P. E. Spoerri, , J. Am. Chem. Soc., 15, 2283 (1953). F. Fieser, Experiments in Organic Chemistry, Vol III, D. C. Heath and Company, Boston, 1957. . F. Fieser, J. Am. Chem. Soc.,_4§, 2639 (1924). 36 CHEMISTRY LIBRARY Ill IIIIIHIIIW