L. _mom AN INVESTIGATION OF NEW METHODS FOR THE SYNTHESIS OF BENZYNE Thesis for 1h. Degree of M. S. MICHIGAN STATE UNIVERSITY Monica H. Narang I966 ABSTRACT AN INVESTIGATION OF NEW METHODS FOR THE SYNTHESIS OF BENZYNE by Monica H. Narang New synthetic routes to benzyne were examined in this investigation. An attempt was made to generate di—tert. butylbenzyne, in order to study the steric effect of the tert -butyl groups on the reactivity of benzyne. In the course of this investigation, three different types of com- pounds were examined as possible precursors of benzyne. The first group of compounds prepared was the phos- phazines, o-quinonetriphenylphosphazine(V) and 2,4—di-tert. butyl-o—quinone-tris(dimethylamino)phosphazine(VIII). It was predicted that these phosphazines, on thermal decomposi- tion, would form benzyne together with nitrogen and a tertiary phosphine oxide. However, it was found that benzyne was not a major product in the decomposition of these phos— phazines. Next, betaine)nHII and XXII were examined as possible precursors of benzyne. These betaines, on thermal decomposi- tion were expected to yield benzyne and a tertiary phosphine oxide. Preparation of betaine}GHII was attempted but unsuccessful. However betaine XXII was prepared, but on thermal decomposition gave phenol and triphenylphosphine oxide. Monica H. Narang Lastly, 1,4-di—tert. butyl-Z-bromobenzene(XXV) was prepared and treated with a strong base, potassium tertiary butoxide, to generate benzyne. Tetracyclone was added to the reaction mixture in the hopes of trapping benzyne by a Diels-Alder addition. The tertiary butoxide ion was Observed to attack tetracyclone; instead of XXV, to form an aqgafi unsaturated ketone. AN INVESTIGATION OF NEW METHODS FOR THE SYNTHESIS OF BENZYNE BY Monica H. Narang A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE Department of Chemistry 1966 ACKNOWLEDGMENTS The author wishes to express her sincere appreciation to Dr. Eugene LeGoff for his guidance and counsel during the course of this investigation. ii TABLE OF CONTENTS INTRODUCTION . . . . . . . . . . . . . . . . . . . . HISTORY AND FORMULATION OF BENZYNE . . . . . . . . . DISCUSSION .. . . . . . . . . . . . . . . . . . . EXPERIMENTAL . . . . . . . . . . . . . . . . . . . . A. General Procedures and Apparatus . . . . B. Preparation of (2—hydroxyphenylhydrazino)- triphenylphosphonium chloride VII . . . . C Attempted Preparation of Triptycene Using o-QuinonetriphenylphOSphazine V . . . . . D. Preparation of 2,4—Di-tert. butyl-o- quinone—tris(dimethylamino)phosphazine VIII 2,4-Di-tert. butyl—pyrocatechol XII. . . . 2,4—Di—tert. butyl—o-benzoquinone XIII . . 2, 4-Di- tert. butyl-o-quinone diazide IX . Tris(dimethylamino)phosphine X . . . . . . 2, 4-Di- tert. butyl-o-quinone— —tris(di- methylamino)phosphazine VIII . . . . . . E. Attempted Preparation of Triptycene Using 2,4—Di-tert. butyl-o-quinone-tris(di- methylamino)phosphazine VIII . . . . . . . F Decomposition of 2, 4-Di- tert. butyl- o- quinone- -tris(dimethylamino)phosphazine VIII G. Attempted Preparation of Betaine XVIII. . . Page 24 24 24 25 25 25 26 27 27 27 Adduct (Betaine) from Triphenylphosphine and p-Benzoquinone XX . . . . . . . . . . . (2-Ethoxy, 5-hydroxyphenyl-triphenylphos- phonium iodide XXI . . . . . . . . (2, 5-Diethoxyphenyl)triphenylphosphonium iodide XIX . . . . . . . . . . . . . . . Betaine XVIII - attempted-.. . . . . . . . iii 29 29 29 TABLE OF CONTENTS (Cont.) Page H. Preparation of Betaine XXII . . . . . . . 30 (o-Methoxyphenyl)triphenylphosphonium Iodide XXIII . . . . . . . . . . . . . 30 Betaine XXII . . . . . . . . . . . . . . 31 I. Attempted Preparation of Triptycene Using Betaine XXII . . . . . . . . . . . . . . 32 J. Decomposition of Betaine XXII . . . . . . 32 K. Preparation of 1,4-Di-tert. butyl—Z- bromobenzene XXV . . . . . . . . . . . . 33 1,4—Di—tert. butyl-Z-nitrobenzene XXVII . 33 1,4-Di-tert. butyl-2-aminobenzene XXVIII. 33 1,4-Divtert. butyl-Z—bromobenzene XXV . . 33 L. Attempted Preparation of Benzyne from 1,4-Di-tert. butyl-2-bromobenzene XXV . . 34 REFERENCES . . . . . . . . . . . . . . . . . . . . 36 iv INTRODUCTION Benzyne, C6H4, is the parent of the family of arynes, not one of which has ever been isolated. Indeed, even physical detection of the short—lived C5H4 Species was not reported till 1960. Notwithstanding, the chemistry of benzyne and its derivatives is a flourishing branch of organic chemistry, laden with interest for both the synthetic and mechanisms chemist. The majority of reactions interpreted in terms of the benzyne intermediate involved the prsence of a halogen atom; however, such an atom is not essential to the mechanism. Any group which is sufficiently electronegative can give rise to benzyne. Thus, reactions of perchloroylaromatic compounds with strong neucleophiles1 and the conversion of aryl sulfonates to phenates have been interpreted in terms of benzyne type intermediates. The present investigation was undertaken to find new synthetic routes to benzyne. Some phosphazines, betaines and 1,4-di—tert. butyl-2-bromobenzene were examined as possible precursors of benzyne. An attempt was made to synthesize di—tert.-butylbenzyne. It was proposed that the tertiary butyl groups would hinder reaction at the triple bond of benzyne, thus making di-tert. butylbenzyne more stable than unsubstituted benzyne. HISTORY AND FORMULATION OF BENZYNE Benzyne chemistry is only 13 years old, though the ob- servation of rearrangements 5J1 nucleophilic aromatic sub- stitutions, which have been interpreted to give benzyne- type intermediates, can be traced back for 85 years. It was during the sixties of the last century that Kekule and Wurtz discovered the transformation of sodium benzenesul— fonate to phenol with molten alkali.2r3 By the seventies the preferred formation of resorcinol in alkaline fusion of benzene-1,4-disolfonate or 4—bromobenzenesulfonate also became known.4:5 $03K ‘SOaK $03K \\\\\\\ E////// $03K 1 ”OH . r OH \\\\\\\ Br —Br $03K 803K Aryne intermediates were proposed as early as 1902 by Stoermer and Kahtert.6 They were the first to suggest that nucleophilic aromatic substitution proceded via a primary elimination to give an aryne intermediate. They accounted 2 3 for their unsuccessful attempts to isolate the aryne inter— mediate in terms of a pronounced Bayer strain in the inter- mediate. This explanation was not generalized, and therefore, sank into oblivion. Bachmann and Clarke in 1927,7 accounted for the tri- phenylene formed in the reaction of chlorobenzene with sodium, as a trimerization of "phenylene radicals" C3H4. Wittig in 1942 advanced the aryne hypothesis.3:9 He postulated the formation of O—lithiophenyl from fluorobenzene and phenyllithium via the "dipolar phenylene" intermediate (Ia). He considered this intermediate to be permanently polarized. Although lb was also mentioned in this con- nection; Ia and lb were not regarded as being identical. Ia Ib Morton gt, _I. also used the dipolar phenylene to interpret the reaction of chlorobenzene with amyl sodium.10 Another decade passed before John D. Roberts, in 1953,11 published experiments which put the benzyne hypothesis on on the road to general acceptance. Roberts investigated the isotOpe distribution in aniline obtained from chloro- benzene—(1—14C) with potassium amide in liquid ammonia. He postulated that the reaction pnmaxfied through a benzyne 4 intermediate. The name benzyne was coined by Roberts. Husigen and Rist in Munich treated o- and m-fluoroanisole with phenyllithium and isolated, after carbonation prac-v tically identical mixtures of isomeric methoxybiphenyl— carboxylic acids.12 The benzyne intermediate with its very reactive triple bond is a good dienophile. The formation of the benzyne intermediate in any reaction is demonstrated by trapping the benzyne with dienes like furan or anthracene. Wittig showed that o-bromofluorobenzene reacts with lithium metal in furan solution to form naphthalene-1,4-endoxide in 76% F .F L1> .__> + LiF furan> Br \Li yield.13 Similar reactions occur with other cyclic dienophiles; the reaction with anthracene is useful for preparation of the unique hydrocarbon triptycene which has the geometry of a three-blade water wheel all three benzene rings being equiva— F [:::]: Mg > [:::] anthracene Br The fact that non—cyclic dienes do not give such adducts H speaks for a one-step mechanism,16 a true DielSeAlder re- action. Unlike most dienophiles, the short-lived benzyne intermediate cannot wait for open-chain dienes to assume the requisite cis-like conformation which is guaranteed in cyclic dienes. In the absence of a diene the benzyne inter- mediate dimerizes to form biphenylene. l l // \\ \\ // Benzyne chemistry has developed very rapidly since its general acceptance in 1953. To date,a number of reactions which give rise to benzyne intermediates have been reported. Muller demonstrated the formation of benzyne from o-dihalo- benzene on treatment with two moles of tetraphenyl-disodio— ethane. (thNaC)2.17 In the early sixties of this century Scott reported that perchloroylaromatic compounds on treatment with strong nucleophiles gave benzyne type intermediates.1 While Wittig and Ebel reported that bis-[o—iodophenyl]-mercury forms 64% biphenylene on pyrolysis at 600°.18 6 However, the long sought goal of preparing benzyne in the absence of metal cations, halide ions or strong bases was first achieved by M. Stiles.19 He isolated benzene- diazonium-Z-carboxylate(II), the zwitterionic diazonium salt from anthranilic acid. N2+ ‘\c02' II This salt, on mild pyrolysis at 35° to 600 gave benzyne which was trapped with furan antracene, carboxylic acids and tert. butyl alcohol. This reaction was further perfected by L. Friedman,20 when he formed benzyne via aprotic diazo- tization of anthranilic acid, with amyl nitrite. NH2 amyl nitrite \COZH E. LeGoff has prepared benzyne by the decomposition of diphenyliodonium—Z—carboxylate(III) in diglyme.21 The benzyne intermediate formed was trapped with tetraphenyl- cyc10pentadienone and anthracene to yield 1,2,3,4—tetra— phenyl naphthalene and triptycene respectively. Biphenylene was obtained from compound III by flash photolysis at 325°. ¢ ¢ Tetracyclone ¢ I+¢> ¢ anthracene . - > Triptycene C02 III 325° > / I \ \\ ‘ // Two years later, Beringer studied the effects of sub- stituents on the rearrangement of 2' aryliodoniobenzoates (IV) and their cleavage to benzyne.22 As expected, the introduction of the electron withdrawing nitro group facili- tated the formation of ester via intramolecular nucleophilic substitution. The formation of benzyne was affected in- directly, as ester formation provided a competitive mode of reaction of the betaines. ,.I Ar \ c02- IV Ar = phenyl, mesityl, p—cyclohexylphenyl, 2,4—dimethoxyphenyl, m—nitrOphenyl, and 2-methyl-5-nitrophenyl. Within the short span of four years, from 1962 to 1965, a number of reactions which formed benzyne intermediates were reported.23,24,25,26,27,28,29,30,31,32,33 Some inter— esting reactions reported were: the decomposition of 8 1.2,3-benzothiadiazole-1,1-dioxide at 60° or at 10° in solution to give benzyne, nitrogen and sulphur dioxide;23 N -—---> + N2 + 802 Cadogon's decomposition of p-tert.-butyl-N—nitroso—acet- anilide in benzene to form tert. butylbenzyne, which was trapped by anthracene to form tert. butyl triptycene;31 \§x N anthracene 3 (O-Ac > o H tert. butyl—triptycene. fl Fields' pyrolysis of phthalic anhydride at 640° to form benzyne.32 Fields also reported that phthalic anhydride on electron impact loses carbon monoxide and carbon dioxide to give benzyne. This year three papers, reporting new methods for the preparation of benzyne, were published. Levy has demonstra— ted the formation of benzyne by the diazotization of o-amine— phenylboronic acid.34 Meyerson and Fields have reported that o-sulphobenzoic anhydride,35 like phthalic anhydride, on pyrolysis at 690° yields, benzyne, sulphur dioxide and carbon dioxide. The latest paper is that of Franck and Yanagi.36 They have been successful in making di-tert. butyl- benzene by the diazotization of 2,5-di—tert. butylaniline. DISCUSSION In the course of this investigation, different phosphorus compounds were prepared and tested as possible benzyne pre- cursors. The first part of this investigation involved the preapration and study of o—quinonetriphenylphosphazine (V). It was proposed that this compound could be thermally decomposed to give benzyne by the elimination of nitrogen and triphenylphosphine oxide. Both nitrogen and triphenyl— phOSphine oxide are good leaving groups. 0 + P493 °N=‘N// V This proposal was also justified on the grounds that both nitrogen and triphenylphosphine oxide have large heats of formation. Thus the above reaction would yield a new syn- thetic route to benzyne, if this method was successful. The next step would be the preparation of 2,5-di-tert. butyl- phOSphazine, which should decompose to 2,5-di-tert. butyl- benzyne. This benzyne derivative should not be very reactive, because of the two tert. butyl groups substituted on either side of the triple bond. The steric effect of the tert.- butyl groups may stabilize the benzyne intermediate long enough, so that the nature of the triple bond may be investi- gated. 10 However, the first step in the preparation of the phos— phazine required the preparation of o-benzoquinone diazide (VI). Diazide VI is extremely unstable and decomposes even at room temperature. All attempts to prepare VI from 0- amino phenol were unsuccessful. // O + VI The procedure was modified and the reaction was repeated. The diazonium salt from o-amino phenol was generated in solution and triphenylphosphine was added to it. (2-Hydroxy— phenylhydrazino)triphenyl phosphonium chloride (VII), m.p. 164—1650 was isolated from the reaction mixture isoamyl ¢3P + nitrite> + - > //P¢3 NH2 N2 Cl N=N VII A mixed melting point of o—amino phenol and VII showed that VII is not o-amino phenol. This was further confirmed by the I.R. spectrum of VII. Two bands at 3300 and 3400 cm-1 corresponding to the amino group in the I.R. of o-amino phenol were not observed in the spectrum of VI[. Although, a new band at 705 cm.1 which may be due to the phosphorous, nitrogen bond was observed. The preparation of the o—quinonephosphazine (V) was attempted by neutralizing VII with different bases. With 11 ammonium hydroxide a pink precipitate m.p. 160-1610, probably V, was obtained. This precipitate on standing turned orange m.p. 148-1550. c1“ NH4OH ’0 _______ + N=N /P¢3 N=N/P¢3 VII V The I.R. Spectrum of V does not show a band at 3100 cm-1 for the -OH group, which is observed in the spectrum of VII. Change in the m.p. of V, on standing, indicated that this compound may be undergoing decomposition. To study the products of decomposition, V was thermally decomposed by addition to molten anthracene. Anthracene melts at 213°. This temperature is high enough to decompose V. If benzyne should form during the decomposition process, it would give triptycene, by a Diels-Alder addition with anthra- cene. The reaction mixture was worked up as reported in Organic Synthesis for the preparation of triptycene.14 How- ever, no triptycene was detected. This strongly suggested that benzyne is not a major product of decomposition of V. 2,4-Di-tert. butyl-o-quinone-tris(dimethylamino)phos- phazine (VIII) was next tested as a possible benzyne pre- cursor . o A _+ + __ _ N=N/P(NMe2)3 > + N2 + o P(NMe2)3 VIII XI 12 Compound VIII was prepared instead of the 2,5—di—tert. butyl phosphazine, because the 2,4-ditert.butyl-o-benzoquinone diazide (IX) which was used to prepare VIII has been reported in the literature.37 0 O + $< > IX X VIII Further, tris(dimethylamino)phosphine (X) was used in lieu of triphenylphosphine,38 as it is more basic and thus forms a more stable phosphazine. Also,because of the more basic nature of X, the phosphorous atom should have a greater affinity for the ortho—substituted oxygen. This would en— courage the formation of tris(dimethylamino)phosphine oxide (XI). Besides, XI is water soluble and thus can be easily separated from the reaction products. 2,4-Di-tert. butyl—o-quinone-tris(dimethylamino)phos- phazine (VIII) was prepared in four steps. The first step involved the preparation of 2,4—di-tert. butyl catechol (XII) by the alkylation of 2-tert. butyl catechol,39 which was readily available. The second step involved the preparation of 2,4-di—tert. butyl o-benzoquinone (XIII) by the oxidation of XII with silver oxide.4° OH / 4O t-BuOH > A920 > OH \\ so XII XIII 13 The tosyl hydrazone (XIV) was prepared by treating the quinone (XIII) with 4-toluenesulfonylhydrazine (XV). Hydrazone XIV was chromatographed over a column of basic alumina to give quantitative yields of 2,4-di-tert. butyl- o-benzoquinone diazide (IX). This was a new one—step pro- cedure discovered to make pure quinone diazide (IX), from the hydrazone. + CH3 ‘C6H4 ‘SOzNHNHZ > so ‘\~ nygH—SOZ-C6H4-CH3 XIII XV XIV (it Lastly, IX was coupled with tris(dimethylamino)ph08phine go 4—> N=.l\1-SOZ-=C6H4-CH3 fin—fi—soz -C6H4 -CH3 (X) to give 2,4-di-tert. butyl—o-quinone—tris(dimethylamino)- phosphazine (VIII m.p. 151-155°.41 b + P(NMez )3 > + NEN N: N N/P(NM92) IX .X VIII PhOSphazine VIII is a new compound and its structure was confirmed by the nmr spectrum. The two tert.-butyl groups (a) show sharp singlets at T 8.76 and T 8.67. The methyl groups (b) show a doublet at T 7.25 (J = 9 cps.) and the two 14 aromatic protons (c) show an AB quartet at T 2.27 (J = 3 cps). In order to study whether di-tert. butylbenzyne was produced on decomposition of VIII, the preparation of di— tert. butyl triptycene was attempted using VIII. No tripty- cene could be recovered from the reaction products. The products of decomposition were then studied. Compound VIII was decomposed in refluxing xylene. The decomposition product, tris(dimethylamino)phosphine oxide (XI), was removed by washing the xylene solution with water. The I.R. spectrum of XI matched the spectrum of a pure sample, thus exhibiting its formation. The residue from the xylene solution was chromatographed and the major product (unknown A) collected, using v.p.c. Unknown A was suspected to be tetra-tert. butyl biphenylene (XVI). The nmr supports this supposition. —L—a a a b b / \b / \ l I or I l a a \ / a \ /b b b a XVI The absorption of the tert.butyl groups (a) is seen as a doublet at T 8.60 (J = 7.5 cps) and the aromatic protons (b) absorb at T 2.83. The u.v. spectrum of biphenylene is recorded in the literature and shows two main regions of absorption, one of high intensity between 235-260 mu and the other at lower intensity in the region of 330—370 mu.42 The intensity of 15 the absorption at 250 mu is 10 times greater than at 350 mu. The u.v. spectrum of unknown A was taken. Two regions of absorption were observed, one at 250 mu and the other at 359 mu. A comparison of the two spectra indicated that unknown A absorped in the same regions as biphenylene and that the ratio between the intensity of absorptions in the two regions was very similar, (unknown A 31 /ek 9.53:1; 250 359 biphenylene ex /51 10:1). This further suggested 250 359 that unknown B was a biphenylene derivative XVI. A mass spectrum of VIII was taken. A very small peak at mass 188 corresponding to di-tert. butylbenzyne was observed. However, a large peak at mass 204 was observed. This peak corresponds to a fragment C14H200 and may possess a structure such as XVII. If this peak corresponds to _ _f 'O: ._ .J XVI I fragment XVII, then it strongly suggests that the carbon oxygen bond is difficult to break and is stronger than the carbon, nitrogen bond. If the carbon, oxygen bond is not broken when bombarded by an electron beam of an energy of 70 volts, it is very unlikely that this bond could be broken thermally when a smaller amount of energy is available. Therefore, unknown A is not likely to be tetra-tert. butyl biphenylene XVI. 16 However, no definite conclusion can be made on the basis of the mass spectrum, due to certain limitations of our mass spectrometer. For example, the long tube between the bulb containing the sample and the ionization chamber presents ample opportunity for rearrangement to take place, during the transfer of the sample to the ionization chamber. If benzyne was produced during the transfer, it would dimerize or react with another species. Thus, no peak for the benzyne species would be recorded in the mass spectrum. It was then decided to test betaine XVIII as a possible benzyne precursor. This compound has not been reported in the literature. (2,5—Diethoxyphenyl)triphenylphOSphonium p¢3 OEt XVIII iodide (XIX) was required for the preparation of the betaine. Compound XIX can be synthesized in three steps.43 The first step involves the preparation of adduct (betaine) XX of triphenylphosphine and p—benzoquinone. Betaine XX was alkylated to give (2-ethoxy, 5—hydroxy—phenyl)triphenyl— phosphonium iodide (XXI), Compound XXI was further alkylated to yield (2,5-diethoxyphenyl)phosphonium iodide XIX. This reaction yielded a tarry product. Suitable solvents were found to extract XIX from the tarry reaction mixture. But the 17 yields were poor. O O OEt u ' + I + P¢3 P¢3 ¢3P + > EtI > Etl > " 6H CH XX XXI b a (2 ”CH2 "CH3 P+¢3I' o-CH2—CH3 b a XIX An nmr spectrum of compound XIX was taken in chloro- form. The spectrum indicated that the two methyl groups (a) absorb in different regions. One methyl group shows a triplet at T 9.32 (J = 7.5 cps) and the other, a triplet at T 8.62 (J 7.5 cps). The methylene protons (b) show a quartet at T 5.92 (J = 7 cps) Since ethers can be cleaved by strong acids, the prepara— tion of betaine XVIII was attempted by treating XIX dis— solved in methanol with concentrated hydrochloric acid. OE t I-_ O I + I + P¢3 Hcl P ¢3 —'_> OEt OEt (attempted) XIX XVII I It was expected that the ethyl group, of the ethoxy substitu— ent, ortho to the phosphonium group would split off in 18 preference to the one in the meta position. This was ex- pected because the phenolate ion generated in the ortho position could be stabilized by resonance. 0 O I + II x/P‘I’a OEt OEt XVIII Red crystals (unknown B) were collected from the methanol solution, m.p. 145-1460. The nmr of unknown B showed two triplets of T 9.42 (J = 7.25 cps) and T 8.72 (J = 7.25 cps), and a quartet at T 6.02 (J = 7 cps). This nmr spectrum was very similar to that of XIX but the melting points of the two compounds differed by 84-85° and, while unknown B was bright red, XIX was white in color. Hence, unknown B could not be XIX. This is further indicated by the carbon, hydrogen analysis. 960 %H Compound XIX 54 4.52 calcd. Betaine XVIII 78.5 5.78 calcd. Unknown A A 41.61 3.45 determined That unknown B is not XIX is clearly evident. Inspection of the analytical results further indicated that unknown B could not be the betaine XVIII. At this stage it may be pointed out that the nmr spectrum of unknown B indicated that it had the same aliphatic groups 19 as XIX. Hence it was suggested that unknown B probably differed from XIX in the nature and number of phenyl groups on the phOSphorous atom. The Wittig reaction shows that betaines decompose to give an olefin and a tertiary phosphine oxide. Hence, to test this reaction for an aromatic betaine, XXII was pre- pared. It was proposed that triphenylphosphine oxide could be cleaved fromXXII to yield benzyne. Moreover, in this case, no nitrogen molecule intervenes between the phOSphorous atom and the aromatic ring as in the phosphazines. This should make the reaction much simpler. O I + O-P¢3 > // XXII To prepare betaine XXII,(o-methoxyphenyl)triphenyl- phOSphonium iodide (XXIII) was required. TriphenylphOSphine was added to the diazonium salt of o—anisidine, to yield OMe OMe - NH2 P¢: NaNOZ ->¢3P HCl NaI XXIII XXIII. OMe (o-Hydroxyphenyl)triphenylphosphonium iodide (XXIV)) was formed by splitting the methyl group of XXIII with pyridine 20 'hydrochloride and acetic acid. XXIV was neutralized with NaOH to give betaine XXII.44 OMe I“ OH O I +¢ I + _ I +¢ \ P 3 @NHCI) 4’31 NaOH > /P 3 ACOH // XXIII XXIV XXII The above compounds are not reported in the literature, although the para isomers are known.44 The most important evidence demonstrating the structure of XXII, XXIII and XXIV was the band at 3060 cm"1 for the -OH group in the I.R. Spectrum of XXIV which was not observed in the spectrum of XXII and XXIII. The analysis of XXII and XXIII were consistent with the structures proposed. The u.v. Spectra of the three compounds are recorded in the experimental section. To investigate if benzyne is a product of decomposition of betaine XXII, the preparation of triptycene was attempted using XXII. It was found that benzyne is not a major product of decomposition of XXII, as no triptycene was detected. To further investigate the formation of benzyne from betaine XXII, its mass spectrum was taken. However, no peak at mass 76, corresponding to benzyne, was observed. This may once again be due to the limitations of the mass spectrometer. To study the products of decomposition of betaine XXII, it was heated in a sealed,evacuated tube. Phenol and tri- phenylphosphine were the two main products recovered. It was suspected that moisture present in betaine XXII caused 21 its hydrolysis to form phenol. 0 OH ' 15¢ ' \ 3 H O - + Z > + o-P¢3 / Thus a study of phosphazines and betaines indicated that benzyne is not a major product of decomposition of these compounds. Since bromobenzene forms benzyne on treatment with a strong base like potassium tertiary butoxidefiian attempt was made to form 1,4-di-tert. butylbenzyne, by treating 1,4—di-tert. butyl-2-bromobenzene (XXV) with potassium tertiary butoxide. Compound XXV was obtained by the following sequence of reactions. p-Di-tert. butylbenzene (XXVI) was nitrated to give 1,4—di-tert.butyl—2-nitrobenzene (XXVII) which on reduction with iron and hydrochloric acid formed 1,4-di— tert. butyl-2—aminobenzene (XXVIII).46 The amino compound é? Compound XXV was then treated with potassium tertiary was diazotized and brominated to give XXV.47 WE XXVI XXVI I XXVI I I butoxide and tetracyclone (XXIX). Tetracyclone, a diene, was used in the hopes of trapping benzyne as a Diels-Alder 22 iadduct. A white compound was recovered from the solution. The carbon and hydrogen analysis indicated the formula of this compound to be C33H3002. Structures XXX and XXXI were prOposed for this compound. ¢ 0 ¢ I —— —-c—o tBu ¢\_1/ E [2 ¢ - H¢ ¢4 0 II xxx . tBu XXXI The nmr supports structure XXXI. A singlet at T 8.95 for the tert. butyl protons and a multiplet at 12.81 for the aromatic protons was observed in the nmr spectrum. The integration ratio between the two peaks is 1:2.3. The theoret- ical ratio should be 1:2.22. Thus the experimental ratio agrees with the theoretical ratio. However, it was the small peak observed at T 5.55 which helped to distinguish between structure XXX and XXXI. The vinylic proton in structure XXX can be compared to the vinylic protons in stilbene (XXXII). These protons are observed in the nmr spectrum at T 3.0.48 However, the benzylic proton of structure XXXI, lies adjacent to a carbonyl; such protons are observed between T 6.5 and T 5.0. For example, the benzylic protons in dibenzylketone (XXXIII) are observed at T 6.30.48 Thus, the peak at 15.55 strongly suggests o ¢H = H¢ ¢-CH2-C-CH2-¢ XXXII XXXIII that the structure of compound C33H3002 is XXXI. 23 The I.R. of this compound was also studied. Absorp- tions for the aromatic C-H stretch and C-C stretch were observed at 2594 cm—1 and 1600 cm.1 respectively. The carbonyl absorption was observed at 1695 cm—1. The I.R. data could not be used to differentiate between structure XXX and XXXI as O‘5 unsaturated esters and a-fi unsaturated ketones absorb in the same region of the I.R. spectrum. The formation of XXXI can be explained by a Michael reaction of tertiary butoxide ion with tetracyclone (XXIX). Cb ch :1: ¢ ¢\— ¢ \—_ _ \_/ o / O ”0 /: _ HOC(CH3)3 _ ’ —/ + oc (CH ) —-> W >/ /L_\ 3 3 A \ CD I ¢ ¢ (pt-,3 4’ o H 4’ u I XXIX tBu XXXI The Diels-Alder adduct of tetracyclone and benzyne was not detected. Therefore benzyne may not be a major product of the reaction. EXPERIMENTAL A. General Procedures andepparatus. All infrared spectra were obtained on a Perkin—Elmer 237B Grating Infrared Spectrophotometer. Ultraviolet spectra were determined in 1 cm quartz cells using a Beck- man DB Spectrophotometer. Nuclear magnetic resonance spectra were obtained on a Varian A-60 high resolution spectrometer, using tetra—methylsilane as a reference. Mass Spectra were obtained by a Consolidated Electrodynamics Corporation Mass Spectrometer Type 21-103C. Vapor phase chromatography was done on an Aerograph model A-90-P3 Gas Chromatograph. Melting points were determined on a Uni-melt capillary melting point apparatus and are uncorrected. Microanalysis determinations were performed by Spang Microanalytical Laboratory, Ann Arbor, Michigan and by Micro-Tech Laboratories, Inc., Skokie, Illinois. B. Preparation of (2-hydroxyphenylhydrazingfitriphenyl— phosphonium chloride VII. A solution of freshly sublimed o-amino phenol (4 g., 0.036 mole) in absolute alcohol (100 ml.) was saturated with hydrogen chloride gas. Iso-amyl nitrite (4 ml.) was added to the solution at 0°. The solution was then stir- red for 15 minutes. Triphenylphosphine (9ifl.g., 0.036 mole) was added dropwise with stirring. The white precipitate formed was filtered off, yield 12.55 g., m.p. 168°. 24 25 C. Attempted Preparation of Triptycene Using ojguinone- triphenylphosphazine V. VII was treated with ammonium hydroxide. V was observed to form and was filtered off from the solution. (2.5 g., 0.0059 mole) of V was added slowly to molten anthracene (1 g., 0.0059 mole) in a pyrex test tube. The mixture was extracted with hot diglyme (40 ml.). Maleic anhydride (1 g., 0.010 mole) was added to the diglyme extract, and it was refluxed for half an hour. The reaction mixture was then set aside, at room temperature, for 2 hours. Maleic an- hydride, anthracene adduct was removed by filtration. The filtrate was refluxed with sodium hydroxide (15 ml., 1N) for 2 hours. When cool, the solution was poured into water (500 ml.). The brown residue, filtered from the water solution, was dissolved in carbon tetrachloride (40 ml.) and chromatographed on a column of acid-washed alumina (50 g.) using carbon tetrachloride (100 ml.) for elution. The carbon tetrachloride solution on evaporation gave a small amount of brown residue which was not triptycene. D. Preparation of 2,4—Di—tert. butyl-o-quinone—tris(di- methylamino)phosphazine VIII. 2,4—Di-tert. butyl-pyrocatechol XII. To 2-tert. butyl pyro- catechol (20 g.) dissolved in acetic acid (40 ml.), were added tert. butyl alcohol (40 ml.) and concentrated sul- phuric acid (8 ml.). The solution was stirred for 3 hours and then allowed to stand at room temperature for 30 hours. The reaction mixture was then stirred into 26 cold water (1 l.) and the crystals formed were filtered) off. Recrystallization from petroleum ether (b.p. 30—60°) gave a yield of 13.4 g. of product, m.p. 98-100°. The melting point reported in the literature is 98-1000. (39) 2,4—Di-tert. butyl-o-benzoquinone XIII. To 2,4—di-tert. butyl pyrocatechol (15 g.) in ether (225 ml.), after treat- ment with sodium sulphate (15 g.) and cooling to 0°, was added silver oxide (18 g.). The mixture was stirred for 3 hours, at 0°, and filtered. The residue was extracted with ether. The filtrate and ether extract were taken together in a large evaporating dish from which the ether was evapor— ated using an air blast. Quantitative yields of red needle- like crystals of 2,4-di-tert. butyl—o-benzoquinone were obtained, m.p. Ill-113°. The melting point recored in the literature is 113-1140. (40) 2,4-Di-tert. butyl-o-quinone diazide IX. p-Tosylhydrazide (11.4 g.) was suspended in methanol (60 ml.) and concentrated hydrochloric acid (6 ml.) and the solution was cooled to -8° to —12° 2,4—Di—tert. butyl-o—benzoquinone was dissolved in methanol (150 ml.) and added dropwise with stirring. Yellow orange crystals of the tosylhydrazone quickly formed; yield 14.2 g., m p. 89° with decomposition. The melting ‘point reported in the literature is the same. (37) The tosylhydrazone (15 g.) was dissolved in methylene chloride (15 ml.) and chromatographed on a column of basic 27 alumina (375 g.). Quantitative yields of 2,4—di-tert. butyl- o-benzoquinone diazide were obtained, m.p. 73°. The melting point reported in the literature is 76°. (37) Tris(dimethylamino)phosphine X. Phosphorous trichloride (43.5 ml., 68.5 g.) in anhydrous ether (200 ml.) was added dropwise to di-methyl amine (149 g.) in anhydrous ether (750 ml.) at -20°. The mixture was stirred for 2 hours at room temperature. The precipitate formed was filtered off and washed with ether. The combined ether solutions were fractionated and gave hexa—methyl phosphorous triamide (50. ml.); b.p.12 49°. The boiling point recorded in the litera- ture is b.p.12 49°. (38) 2,4—Di-tert. butyl-o-quinone-tris(dimethylamino)phosphazine VIII. Tris(dimethylamino)phosphine (X) (2 ml.) was added to 2,4— di—tert. butyl-o-quinone diazide (IX) (2.32 g., 0.1 mole) dissolved in a minimum amount of benzene. After a few minutes the solution was observed to take on a deep red color. On addition of pet. ether (b.p. 30—60°) and cooling, crys- tals were observed to separate our ot the solution. The crystals were filtered off, yield 3.4 g., m.p. 151—155°. E. Attempted Preparation of Triptycene Using 2,4—Di-tert. butyl-o-quinone—tris(gimeflnflamino)phosphazine VIII. The procedure followed was the same as has been outlined eaflier. However, no triptycene was obtained from the re- action. 28 F. Decomposition of 2,4—Di-tert. butyl-o-quinone—tris- (dimethylamino)phosphazine VIII. 2,4-Di-tert. butyl-o—quinone—tris(dimethylamino)- phosphazine (6 g.) was added to boiling xylene (40 ml.) in a round bottom flask. The xylene solution was washed with water. The water extract on evaporation gave an oily residue. The xylene solution gave a red colored residue on evaporation. This residue was dissolved in carbon tetra- chloride and chromatographed on a column of basic alumina. The residue from the first fraction obtained from the alumina column was redissolved in cyclohexane and chromatographed on a smaller column of basic alumina and eluted with cyclo— hexane. Two fractions were collected. Fraction A (1.0 g.) and fraction B (0.4 g.). Fraction A was dissolved in a small quantity of chloroform and put through the the gas chromatograph using column S.E.-30 at 230°. The sample (unknown A) corresponding to the two major peaks was col- lected. Twenty trial were run using 50 micro liters of chloroform solution at a time, to collect enough quantity of the sample for identification. The major peak was observed at a retention time of 5 minutes. At this point there were two peaks that overlapped. Lowering of the column tempera— ture did not bring about an effective separation. U.v. spectrum of unknown B ifizfiane 250 (e 7870), 359 (e 1139). (Based on molecular wt. of 376). The chromatogram for fraction B Showed the same major peaks, as fraction A, except that they were much smaller, 29 No significant quantity of the samples could be obtained from fraction B. G. Attempted Preparation of Betaine XVIII. Adduct (Betaine) from Triphenylphosphine and peBenzoquinone XX. Solutions of p-benzoquinone (2.5 g., 0.023 mole) and tri- phenylphosphine (6.08 g,, 0.023 mole) in anhydrous benzene were mixed in a dry nitrogen atmosphere (inert bag). There was no immediate change; gradually, after several minutes, a deep red color developed in the solution and a precipitate began to appear. After 20 hours at room temperature (in- ert bag) the precipitate was collected (7.8 g.); it consisted of fine, yellow—greenish crystals, m.p. 262-266°. The value for the melting point reported in the literature is 262-266°. (43) (2—Ethoxy, 5—hydroxyphenyl)triphenylphosphonium Iodide XXI. A solution of betaine (5.0 g.) in methanol was refluxed with excess of ethyl iodide (3.5 ml.), for 10 hours. Removal of solvent left a residue which was dissolved in chloroform and re-precipitated with ether; yield 6.6 g., m.p. 232- 235°. The value for the melting point reported in the literature is 224—231° (43) (2,5-Diethoxyphenyl)triphenylphosphonium Iodide XIXJ (2- Ethoxy,5-hydroxyphenyl)triphenylphosphonium iodide (3.5 g.) was added to a solution of sodium (0.15 g.) in methanol (35 ml.). Excess ethyl iodide (3.5 ml.) was added and the 30 solution was refluxed for 12 hours. The residue remaining after evaporation was taken up in chloroform and the chloro- form solution was washed with 5% aqueous sodium hydroxide and with water. Evaporation of the dried chloroform solu- tion left an oily residue. The residue was treated with ethyl acetate containing a few drops of methanol. The ethyl acetate solution dissolved the oil and left behind a light green residue; m.p. 212—215°. The value for the melting point reported in the literature is 212—2150. (43) On crystallization from methanol the product collected was white in color, m.p. 230°. Betaine XVIII — Attempted—. (2.5-Diethoxyphenyl)triphenyl- phosphonium iodide (0.5 g.) was dissolved in methanol (4 ml.) and the solution was treated with concentrated hydro- chloric acid (1 ml.). Red crystals (unknown B) were observed to form in the solution. The crystals were filtered off at 48 hour intervals, m.p. 145-1460. Analysis: Calcd. 78.5% C 5.78% H Found. 41.61% C 3.45% H. The above compound could not be identified. H. Preparation of Betaine XXII. (o-Methoxyphenyl)triphenylphosphonium Iodide XXIII. Sodium nitrite (3.45 g., 0.05 mole) in water (30 ml.) was added to o—anisidine (6.15 g., 0.05 mole) in concentrated hydro- chloric acid (10 ml. ~ 0.1 mole). The diazonium salt formed was buffered by adding sodium acetate (8.1 g., 0.15 mole). 31 Thus adding 3 mole equivalents per mole equivalent of excess acid. Triphenylphosphine (13.114 9, 0.05 mole) in ethyl acetate (200 ml.) was added dropwise with stirring to the diazonium salt. The solution was observed to turn red. The aqueous layer was extracted with ether and the organic layer with water. The combined aqueous solutions were treated with sodium iodide. VXXIII precipitated out of the solution. The compound was redissolved in methanol and re- precipitated with ether, m.p. 258—60°; infrared absorption spectrum of XXIII in chloroform, bands at: 2925 (s), 1600 (S), 1435 (s), 255 (m), 1110 (s). The u.v. absorption 95% EtOH Spectrum of XXIII at: xmax 292 (8 4690), 276 (E 4840), 269 (E 4410). Analysis: Calcd. for C25H220PI: 60.5% C; 4.44% H Found 61.07% C; 4.72% H. Betaine XXII. (o-Methoxyphenyl)triphenylphosphonium iodide (4.16 g., 0.0084 mole), pyridine hydrochloride (12.5 g.) and acetic acid (5 ml.) were heated for 6 hours with stir— ring at 190—200°. The solution was poured into water and on treatment with sodium iodide gave (o—hydroxyphenyl)tri— phenylphosphonium iodide (XXIV), yield 5.3 g., m.p. 264° (aqueous methanol); infrared absorption Spectrum of XXIV in chloroform at: 3060 (S), 1590 (s), 1440 (s), 1110 (s). The 95% EtOH max 296 (8 5390), u.v. absorption Spectrum of XXIV: I 276 (a 5330), 269(5 5050). Compound XXIV was dissolved in hot water and reprecipi- tated with sodium hydroxide to give betaine XXII, yield 32 2.46 g., m.p. 279-280o (aqueous methanol). Infrared absorp- tion Spectrum of XXII in chloroform at: 2925 (w), 1580 (s), 1430 (m), 1105 (m). The u.v. absorption spectrum of XXII: 95 EtOH xmax 241 (e 5310). Analysis Calcd. for C24H190P: 81.5% C 5.37% H Found 80.34% c 5.40% H. I. Attempted Preparation of Triptycene Using Betaine XXII. The procedure followed was the same as outlined before. However, no triptycene could be prepared using betaine XXII. J. Decomposition of Betaine XXII. (a) Betaine XXII (1 g.) was heated at 300° in a test tube. The test tube was hooked to an aspirator. The yellow product obtained was chromatographed over a column of 25 g. of silicic acid. The column was eluted with chloroform (200 ml.), ethyl acetate (100 ml.) and 50% methanol in ethyl acetate (100 ml.). Phenol (0.3624 g.) was obtained from the chloroform solution. Triphenylphosphine (0.0770 g.) was alSo obtained from the chloroform solution. Some start- ing material was obtained from the other solutions. (b) Betaine XXII (1 g.) was heated in a six inch pyrex test tube. The tube was evacuated (press 2 mm.). The products were chromatographed as before. Obtained were 0.2837 g. phenol, 0.2976 g. of triphenylphosphine oxide and 0.1986 g. of starting material. Therefore a total of 0.7799 g. of material was recovered. 33 K. Preparation of 1,4-Di-tert. butyl-Z—bromobenzene. .— 1,4-Di-tert. butyl-2-nitrobenzene XXVII. Ninety percent HNO3 (13.9 ml., 0.25 mole) was added to a solution of acetic acid (48 g., 0.8 mole) at 0°. p-Di-tert. butyl benzene (1.9 g., 0.1 mole) was added to this mixture at -10°. The resulting solution was quenched with ice and the pre- cipitate formed was filtered off. Quantitative yields of the nitro compound were obtained, m.p. 85-85.8°. The melt- ing point reported in the literature is 87-88°. (46) 1,4-Di-tert. butyl-2—aminobenzene XXVIII. Compound XXVII (7.05 g.) and iron filings (6 g., 40 mesh) were added to 50% ethanol (12 ml.) and hydrochloric acid (0.6 g.). The resulting solution was refluxed for 24 hours. The amine was extracted from the mixture by absolute alchol and re- crystallized from alcohol, m.p. 103—104°. The melting point reported in the literature is 103—1040. (46) 1,4-Di-tert. butyl—2-bromobenzene XXV. A mixture of XXVIII (25 g.), cuprouS bromide (36 g.), potassium bromide (30 g.) and 48% of hydrobromic acid (250 ml.) was cooled to -40°. Solid sodium nitrite (8.5 g.) was added to the mixture over 60 minutes. After Slow warming to room temperature, the reaction mixture was washed with petroleum ether (b.p. 40- 60°). The petroleum ether extract was chromatographed over a column of alumina. On elution of the column an orange colored solution was collected which was rechromatographed 34 over another alumina column. The yellow oil collected was distilled three times to obtain XXV (12.6 g.) as a light yellow colored liquid, b.p. 112-118°/l.5-1.75 mm. The boil- ing point reported in the literature is 154-156°/13 mm. (47) The bromo compound (0.2 ml.) was put through the gas chromato- graph using column S.E.-30 at 200° to obtain 106 m.g. of pure colorless XXV. The nmr Spectrum Shows the two tert. butyl groups at T 8.72 and T 8.52. The aromatic protons exhibit a typical ABC spectrum A 12.90, B T 2.80 and C T 2.52 (JAB = 6.9 cps., J = 1.5 cps., J = 0.4 cps.) BC AC L. AttempEed Preparation of Benzyne from 1,4—Di-tert. butyl- 2-bromobenzene XXV. Compound XXV (2 g., 0.00744 mole) and tetracyclone (2.775 g., 0.00744 mole) were dissolved in benzene. Potassium tertiary butoxide (1.214 g., 0.00744 mole; contains 38% tert. butyl alcohol) was added and the resulting solution was refluxed for 24 hours. Excess benzene was distilled off and the concentrated solution was chromatographed over a column of alumina using cyclohexane and benzene (1:1). Tetracyclone was obtained in the first fraction. The column was then eluted with benzene—cyclohexane which on evapora— tion gave 0.340 g. of white crystals, XXXI, m.p. 164—1650. Crystallization from benzene-ethanol gave 0.159 g. of crystals, m.p. 169-1700. Infrared absorption spectrum of XXXI in carbon tetrachloride at: 2594 (s), 1695 (s), 1600 (m)- 35 Analysis Calcd. for C33H3002: 86.5% C 6.55% H Found 87.63% c 6.76% H. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. F. REFERENCES L. Scott, and R. E. Oesterling, J. Am. Chem. Soc. g2, 5284 (1960). . Kekule, Zeitsehr. P. Chem., 300, 1867; Compt. rend. 6_4, 753 (1864) . Wurtz, Compt. rend. 64, 749 (1864); Ann. Chem. Pharm., 144,121 (1% 7). . C. Barth, E. D. Hughes and C. K. Ingold, J. Chem. Soc. 960 (1874). . Limpricht, Ber., l, 1349 (1874). . Stoermer, and B. Kahlert, Ber., 35, 1633 (1902) . E. Bachmann, and H. T. Clarke, J. Am. Chem. Soc., 22, 2089 (1927) . Wittig, Naturwissenschaften,§QJ 696 (1942) . Wittig, Angew. Chem., pg, 10 (1954) . A. Morton, J. B. Davidson, and B. L. Hakan, J. Am. Chem. Soc., 64, 2242 (1942) . D. Roberts, H. E. Simmons, L. A. Carlsmith, and C. W. Vaugham, J. Am. Chem. Soc., 15, 3290 (1953). . Huisgen, and H. Rist, Naturwissenschaften, 41, 358 (1954); Ann., 594, 137 (19557. . Wittig, and L. Pohmer, Angew. Chem. , 67, 348 (1955); Chem. Ber. , 89,1334 (I956). . Wittig, Organic Synthesis, 39, 75 (1959). . D. Bartlett, M. J. Ryan, and S. G. Cohen, J. Am. Chem. Soc., 64, 2649 (1942). M. Arnett, J. Org. Chem., 25, 324 (1960). . Mfiller and G. Rbscheisen, Chem. Ztg,, 89, 101 (1956) . Wittig and H. F. Ebel, Ann., 650, 20-34 (1961). . Stiles and R. G. Miller, J. Am. Chem. Soc., 82, 3802 (1960); J. Am. Chem. Soc., gg, 1792-7 (19687 36 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. 39. 40. Q I?! CC ('3 37 . Friedman and F. M. Loqulla, J. Am. Chem. Soc., 55, 1549 (1963). -- . LeGoff, J. Am. Chem. Soc., 51, 3786 (1962). . M. Beringer and S. J. Huang, J. Org, Chem., 29 6 , 1637-8 (1964). . Wittig and R. W. Hoffman, Ber., 55, 2718-28 (1962). . Wawzone and J. H. Wagenknecht, J. Electrochem. Soc., 110, 420-2 (1963). . H. S. Winkler and G. Wittig, J. Org. Chem., 28(7), 1733-40 (1963). . Guenther, Ber., 96 7 , 1801-9 (1963). . McNelis, J. Org. Chem., 28 11 , 3188-94 (1963). . Koebrich, Ber., 96 10 , 2544-9 (1963). . D. Campbell and C. W. Reevs, Pro. Chem. Soc., 269 (1964). R. W. Hoffmann, Ber., 97(10), 2763-71 (1964). 2 31 IE 2 . J. G. Cadogan and P. G. Hibbert, Proc. Chem. Soc., 338-9 (1964). . K. Fields and S. Meyerson, Chem. Commun-, 29, 474-6, (1965). . P. Buu-Hoi, P. Jacquiqnon, and Mangane, Chem. Commun., 52, 624-5 (1965). . Verbit, J. S. Levy, H. Robitz, and W. Kwalwasser, Tetrahedron Letters, lg, 1053-5, (1966). . Meyerson and E. K. Fields, Chem. Commun., 5, 275-7 (1966). . W. Franck and K. Yanagi, Tetrahedron Letters, 25, 2905 (1966). . Ried and R. Dietrich, Chem. Ber., 53, 387-391 (1961). . Oediger and K. Eiter, Ann. Chem., 682, 58-61 (1965). . Schulze and W. Flaig, Ann., 515, 231-41 (1952). . Flaig, J. Ploetz and H. Biergans, Ann., 595, 196— 213 (1955). 38 41. W. Ried and H. Appel, Ann., 679, 56 (1964). 42. E. P. Carr, L. W. Pickett and D. Voris, J. Am. Chem. Soc., §§, 3231 (1941). 43. F. Ramirez and S. Dershowitz, J. Am. Chem. Soc., 15, 5614 (1956). "' 44. H. Hoffman and D. Michael, chem. Ber., 95, 528-35 (1962). 45. J. I. G. Cadogan, Private Commun.; G. Wittig, Ang w. Chem., 15, (1966) - in press. 46. M. S. Carpenter, W. M. Easter and T. F. Wood, J. Org. Chem., 55, 586-617 (1951). 47. J. M. A. Baas and B. M. Wepster, Rec. Trav. Chim.,55, 457 (1966). “‘ 48. NMR Spectra Catalog of the Instrument Division of Varian Associates, Spectra No. 511, 514 and 638. WMHWSANSTATEIHHVFRSHY UBWQRWQ l i IIIII III ,I‘ I.) W», WI“ 3 174 4208 ‘ Hi 1 3" 7’ M l‘ 3 129 30