HETEROCYCLES

Paper | Special issue | Vol. 79, No. 1, 2009, pp. 865-872
Received, 1st October, 2008, Accepted, 8th December, 2008, Published online, 10th December, 2008.
DOI: 10.3987/COM-08-S(D)57

■ Effect of Aryl Substituents on Intramolecular Cyclization of 2,2’-Biphenoquinones

Naoto Hayashi,* Akifumi Kanda, Taku Kamoto, Hiroyuki Higuchi, and Takeyuki Akita*

Department of Chemistry, Faculty of Science, Toyama University, Gofuku 3190, Toyama 930, Japan

Abstract

Effect of aryl substituents on intramolecular cyclizations of 3,3',5,5'-tetraaryl-2,2'-biphenoquinones (Ar = phenyl (1a) and 4-methoxyphenyl (1b)) has been studied. In benzene, 1a gave 2,4,6,8-tetraphenyldibenzofuran-1-ol (10) gradually as a main product, indicating the phenyl substituents preferred to stabilize the intermediate by delocalization of the negative charge rather than that of the positive one. In contrast, the reaction of 1b occurred spontaneously in order to give a complex mixture, which should be due to 4-methoxyphenyl substituent at the 3 position.

INTRODUCTION
2,2'-Biphenoquinone (I) is a more reactive isomer of 4,4'-biphenoquinone so as to undergo thermal cyclization facilely to give either dibenzofuran (II) or oxepino[2,3-b]benzofuran (III).1-4 According to Wan et al.,1 the mode of reaction depends upon the substituent (R) at the 5 (and 5') position. When R is an electron-donating substituent, such as methoxy and methyl groups,1a-b III is yielded preferably. This is explained in terms of the substituent effect of the R, which stabilizes the intermediate V by resonanc

and/or inductive effects. In contrast, 2,2'-biphenoquinones bearing an electron-withdrawing substituent, such as chloro group,1a-b give II rather than III, because the R stabilizes the intermediate IV also by resonance and/or inductive effects (Scheme 1). Based on the resonance theory, these explanations should be reasonably applicable to substituents at the 3 and 3' positions. Effect of aryl substituents at the 5 and 3 positions on the cyclization of 2,2'-biphenoquinones appears to be an interesting issue, because they can stabilize both IV and V by delocalization of the negative and positive charges, respectively; however those compounds have been rarely studied. Thus, in the present study, 3,3',5,5'-tetraaryl-2,2'-biphenoquinones have been synthesized and their cyclization reactions studied. As the stabilization effect on the intermediate was expected to vary depending on the nature of aryl substituent, phenyl (1a) and 4-methoxyphenyl (1b) substituents were employed. It should be noted that bulkiness of the aryl group at 3 position would not prevent formation of the epoxide ring in a steric manner (Scheme 1), because 3,3',5,5'-tetra-tert-butyl-2,2'-biphenoquinone was known to give oxepino[2,3-b]benzofuran in a good yield.3a

RESULTS AND DISCUSSION
Synthesis of 1a-b was depicted in Scheme 2. 3,3',5,5'-Tetrabromo-2,2'-dimethoxybiphenyl (3) was subjected to Suzuki-Miyaura coupling5 with phenylboronic acid to give tetraphenylbiphenyl 4, which was then deprotected in the presence of BBr3 to yield tetraphenyl-2,2'-biphenol 5a. On the other hand,

tetrabromo-2,2'-bis(methoxymethoxy)biphenyl (6) underwent Suzuki-Miyaura coupling with 4-methoxyphenylboronic acids to yield tetraarylbiphenyl (7), which was subjected to acid-catalyzed deprotection to give tetraaryl-2,2'-biphenol 5b.
When an ethereal solution of 5a was shaken with an aqueous solution of excess amount of potassium hexacyanoferrate(III) and sodium hydroxide for ten minutes, deep green solids were precipitated.6 Deep purple solids were obtained from 5b in a similar manner. Although NMR spectra of these solids could not be recorded due to instability, IR analysis revealed that the obtained solids were composed exclusively of 1a and 1b, respectively: for example, most peaks of IR spectra of the green solid were found to be very close to those of 5a in position and intensity except for broad peak at 3600-3200 cm-1 (i.e., stretching vibrations of O-H bond), which was absent in the former. The IR spectra also indicated that the obtained solid was not contaminated by monoradical 8. Absence of 8 was further confirmed by EPR spectra, where no peak was assigned to 8. EPR spectra also indicated that 1a should be represented by a quinonoid canonical form essentially, and contribution of the biradical one (9) be negligible. This is similar to the behavior of binaphthoquinone, which was also ESR inactive in solution7 unlike Bourdon and Calvin’s hindered 4,4'-biphenoquinones.8

Although 1a was stable at ambient temperature in solid, it underwent reactions gradually in solution. The decay of 1a in benzene was followed by monitoring relative intensity of the electronic absorption peak at 708 nm (Figure 1), which became to be zero in one hour. A similar rate of decay was observed in toluene,

indicating that 1a did not behave as biradical species again. The main product was 2,4,6,8-tetraphenyldibenzofuran-1-ol 10 (42%), and no oxepino[2,3-b]benzofuran was obtained. This indicates that the reaction of 1a proceeded mainly along path a in Scheme 3, in which the phenyl substituents stabilized the intermediate VIa by means of delocalization of the negative charge. Unlike 3,3', 5,5'-tetraphenyl-4,4'-biphenoquinones,9 a structural isomer of 1a, no reaction product via intramolecular attack of oxygen atom to the phenyl substituent at the 3 position (path c in Scheme 3) was obtained.
In contrast to 1a, 1b was expected to give oxepino[2,3-b]benzofuran preferably because electron-donating character of 4-methoxyphenyl substituents in 1b should stabilize VIb to less extent and VIIb to more extent. Moreover, 3,3'-di-tert-butyl-5,5'-bis(4-tert-butylphenyl)-2,2'-biphenoquinones (11), which was generated from 2-tert-butyl-4-(4-tert-butylphenyl)phenol, was reported3a to give oxepino[2,3-b]benzofuran (12) in moderate yield (Scheme 4). Nevertheless, in various organic solvents such as benzene, DMSO, DMF, ethanol, CH3CN, acetone, THF, CH2Cl2, CHCl3, and n-hexane, 1b decayed rapidly to give a complex mixture, from which neither oxepino[2,3-b]benzofuran nor dibenzofuran product was detected. Although no mechanistic studies could be performed due to difficulty of the product analysis, comparison of molecular structure of 1b with that of 11 indicates that 4-methoxyphenyl substituent at the 3 position (not 5) should be responsible for this behavior.
In conclusion, this paper describes synthesis and reactivity of 3,3',5,5'-tetraaryl-2,2'-biphenoquinones 1a and 1b. EPR and UV/Vis study revealed that 1a should be represented by a quinonoid canonical form essentially, being similar to binaphthoquinones and unlike Bourdon and Calvin’s hindered

4,4'-biphenoquinones. The phenyl substituents in 1a were found to stabilize the intermediate by delocalization of the negative charge to give dibenzofuran as a main product in a moderate yield. In contrast, the reactions of 1b afforded a complex mixture probably because of the 4-methoxyphenyl substituent at the 3 (not 5) position.

EXPERIMENTAL
General. All commercially available chemicals were used without further purification. Melting points were determined on microscopic thermometer without correction. 1H and 13C NMR spectra were recorded on a JEOL JNM-ECP600 (600 MHz for 1H and 150 MHz for 13C) with tetramethylsilane as internal reference. Mass spectra were conducted on a JEOL MStation JMS-700 (EI) and a JEOL JMS-SX102A (HRMS/EI). Infrared spectra were measured on a JASCO FT/IR-6100. EPR spectra were recorded on a Brucker EMX EPR Spectrometer.

Synthesis of 3,3',5,5'-tetraphenyl-2,2'-dimethoxybiphenyl (4). A bi-layer solution of 3,3',5,5'-tetrabromo-2,2'-dimethoxybiphenyl (3)10 (1.53 g, 2.87 mmol) and phenylboronic acid (1.57 g, 12.9 mmol) in THF (30 mL) and 1 mol L-1 aqueous Na2CO3 (20 mL) was degassed with argon. After tetrakis(triphenylphosphine)palladium (0.664 g, 0.575 mmol) was added, the reaction mixture was refluxed for overnight. After cooling, organic layer was separated, and aqueous layer was extracted with Et2O. Combined organic phase was washed with brine, dried over MgSO4, and evaporated to dryness. From the crude product, 4 (1.38 g, 93%) was isolated by preparative column chromatography (SiO2, n-hexane/EtOAc 10:3) as a white powder.

Mp 81-86 ºC. 1H NMR (CDCl3): δ = 7.70-7.63 (12H, m, Ar), 7.46-7.40 (8H, m, Ar), 7.36 (2H, t, J = 7.3 Hz, Ar), 7.31 (2H, t, J = 7.3 Hz, Ar), 3.35 (6H, s, OMe). 13C NMR (CDCl3): δ = 154.94, 140.37, 138.81, 136.52, 135.35, 133.15, 129.44, 129.35, 129.24, 128.73, 128.27, 127.21, 127.11, 126.99, 60.74. MS: m/z = 518 (M+). HRMS (m/z): 518.2247 (M+, calcd. 518.2246 for C38H30O2).

Synthesis of 3,3',5,5'-tetraphenyl-2,2'-biphenol (5a). To a solution of 4 (1.38 g, 2.66 mmol) in CHCl3 (20 mL) was dropwised boron tribromide (0.57 mL, 6.1 mmol) at 0 ºC. After stirring for 3 h, the reaction mixture was quenched with MeOH (5 mL) then water (5 mL), and organic substances were extracted into CHCl3. The organic solution was washed with water and brine, dried over Na2SO4, and the solvent was removed under vacuum leaving 5a (1.24 g, 95%) as a white solid. Since the obtained 5a was practically pure, it was used in the next reaction without further purification.
Mp 189-192 ºC (lit.,2a 189-192 ºC). 1H NMR (CDCl3): δ = 7.65 (2H, d, J = 2.4 Hz, Ar), 7.64-7.62 (10H, m, Ar), 7.50 (4H, t, J = 7.2 Hz, Ar), 7.44-7.40 (6H, m, Ar), 7.34-7.31 (2H, m, Ar), 5.94 (2H, s, OH).

Synthesis of 3,3',5,5'-tetrakis(4-methoxyphenyl)-2,2'-dimethoxymethoxybiphenyl (7). A bi-layer solution of 3,3',5,5'-tetrabromo-2,2'-bis(methoxymethoxy)biphenyl (6)11 (1.96 g, 3.33 mmol) and 4-methoxyphenylboronic acid (2.28 g, 15.0 mmol) in DME (120 mL) and 1 mol L-1 aqueous Na2CO3 (80 mL) was degassed with argon. After tetrakis(triphenylphosphine)palladium (0.77 g, 6.7 mmol) was added, the reaction mixture was refluxed for overnight. After cooling, organic layer was separated, and aqueous layer was extracted with EtOAc. Combined organic phase was washed with brine, dried over MgSO4, and evaporated to dryness. From the crude product, 7 (2.06 g, 88%) was isolated by preparative column chromatography (SiO2, n-hexane/EtOAc 2:1) as a white powder.
Mp 76-84 ºC. 1H NMR (CDCl3): δ = 7.66 (2H, d, J = 2.6 Hz, Ar), 7.62 (4H, d, J = 8.8 Hz, Ar), 7.58 (4H, d, J = 8.8 Hz, Ar), 7.53 (2H, d, J = 2.4 Hz, Ar), 6.99 (4H, d, J = 8.8 Hz, Ar), 6.97 (4H, d, J = 8.8 Hz, Ar), 4.53 (4H, s, -OCH2OMe), 3.85 (6H, s, ArOMe), 3.83 (6H, s, ArOMe), 2.76 (6H, s, -OCH2OMe). 13C NMR (CDCl3): δ = 159.06, 158.86, 151.44, 136.57, 135.53, 133.96, 132.95, 131.41, 130.65, 129.11, 128.51, 127.97, 114.19, 113.69, 98.79, 56.33, 55.30, 55.25. MS: m/z = 698 (M+). HRMS (m/z): 698.2883 (M+, calcd. 698.2880 for C44H42O8).

Synthesis of 3,3',5,5'-tetrakis(4-methoxyphenyl)-2,2'-biphenol (5b). To a solution of 7 (0.086 g, 0.12 mmol) in DME (40 mL) was added 3 mol L-1 hydrochloric acid (6.0 mL, 18 mmol) at ambient temperature. After refluxing for 3 h, the reaction mixture was cooled, and organic substances were extracted into Et2O. The solution was washed with water and brine, dried, and the solvent was removed under vacuum leaving 5b (0.068 g, 90%) as a white solid. Since the obtained 5b was practically pure, it was used in the next reaction without further purification.
Mp 99-101 ºC. 1H NMR (CDCl3): δ = 7.56-7.53 (12H, m, Ar), 7.02 (4H, d, J = 8.8 Hz, Ar), 6.96 (4H, d, J = 8.8 Hz, Ar), 5.88 (2H, s, OH), 3.86 (6H, s, OMe), 3.83 (6H, s, OMe). 13C NMR (CDCl3): δ = 159.28, 158.93, 148.87, 134.18, 132.97, 130.55, 129.66, 129.57, 128.90, 128.84, 127.84, 125.44, 114.31, 114.23, 55.35, 55.34. MS: m/z = 610(M+). HRMS (m/z): 610.2353 (M+, calcd. 610.2355 for C40H34O6).

General prodecedure of preparation of 3,3',5,5'-tetraaryl-2,2'-biphenoquinones. In a separatory funnel was placed a solution of 0.10 g of 3,3',5,5'-tetraaryl-2,2'-biphenol in 10 mL of Et2O or EtOAc. To this solution was added a solution of 0.85g (2.6 mmol) of potassium hexacyanoferrate(III) and 0.20 g (5.0 mmol) sodium hydroxide in water (12 mL), and the resulting mixture was vigorously shaken for about 10 min. The precipitate was isolated by suction filtration, washed with several portion of water, and dried in a vacuum desiccator. Yield 1a, 80%; 1b, 32%.

Synthesis of 2,4,6,8-tetraphenyldibenzofuran-1-ol (10). A solution of 100 mg (0.20 mmol) of 1a in 100 mL of benzene was stirred at ambient temperature. When the starting material disappeared on TLC, the solvent was removed under vacuum leaving a reaction mixture, which was subject to preparative column chromatography (SiO2, n-hexane/CHCl3 1:1) to isolate 10 (42 mg, 42%) as a white powder.
Mp 235-239 ºC. 1H NMR (CDCl3): δ = 8.41 (1H, d, J = 1.8 Hz, Ar), 8.03 (2H, dd, J = 8.4, 1.1 Hz, Ar), 7.96 (2H, dd, J = 8.4, 1.1 Hz, Ar), 7.88 (1H, d, J = 1.8 Hz, Ar), 7.77 (2H, dd, J = 8.2, 1.3 Hz, Ar), 7.60-7.58 (5H, m, Ar), 7.55-7.53 (2H, m, Ar), 7.51-7.47 (5H, m, Ar), 7.45-7.42 (1H, m, Ar), 7.40-7.36 (2H, m, Ar), 5.99 (1H, s, OH). 13C NMR (CDCl3): δ = 154.45, 152.66, 147.79, 141.51, 137.31, 136.45, 136.28, 136.05, 129.78, 129.46, 128.83, 128.73, 128.67, 128.62, 128.47, 128.24, 128.15, 127.86, 127.57, 127.22, 127.09, 125.58, 125.18, 124.88, 122.81, 120.65, 118.49, 113.27. MS: m/z = 488(M+). HRMS (m/z): 488.1776 (M+, calcd. 488.1776 for C36H24O2).

ACKNOWLEDGMENTS
This work was financially supported by Saneyoshi Scholarship Fundation, Japan.

References

1. (a) B. Guan and P. Wan, J. Chem. Soc., Chem. Commun., 1993, 409; CrossRef (b) B. Guan and P. Wan, J. Photochem. Photobiol. A: Chem., 1994, 80, 199; CrossRef (c) S. Rayne, R. Sasaki, and P. Wan, Photochem. Photobiol. Sci., 2005, 4, 876. CrossRef
2. (a) H.-D. Becker, J. Org. Chem., 1969, 34, 2027; CrossRef (b) H.-D. Becker and K. Gustafsson, Tetrahedron Lett., 1976, 4883; CrossRef (c) H.-D. Becker and H. Lingnert, J. Org. Chem., 1982, 47, 1095. CrossRef
3. (a) F. R. Hewgill and G. B. Howie, Aust. J. Chem., 1978, 31, 1061; (b) F. R. Hewgill, Tetrahedron, 1978, 34, 1595; CrossRef (c) F. R. Hewgill, D. G. Hewitt, G. B. Howie, C. L. Raston, R. J. Webb, and A. H. White, J. Chem. Soc., Perkin Trans. 1, 1979, 290; CrossRef (d) F. R. Hewgill, B. La Greca, F. Legge, and P. E. Roga, J. Chem. Soc., Perkin Trans. 1, 1983, 131. CrossRef
4. H. Meier, H.-P. Schneider, A. Rieker, and P. B. Hitchcock, Angew. Chem., Int. Ed. Engl., 1978, 17, 121. CrossRef
5. N. Miyaura and A. Suzuki, Chem. Rev., 1995, 95, 2457. CrossRef
6. When one equivalent of K3Fe(CN)6 and NaOH were used, 8 was not obtained probably because of rapid decay.
7. Y. Matsunaga, Can. J. Chem., 1959, 37, 1003. CrossRef
8. J. Bourdon and M. Calvin, J. Org. Chem., 1957, 22, 101. CrossRef
9. A. S. Hay, J. Org. Chem., 1971, 36, 218. CrossRef
10. H. Gilman, J. Swiss, and L. C. Cheney, J. Am. Chem. Soc., 1940, 62, 1963; CrossRef N. Thasana, S. Pisutjaroenpong, and S. Ruchirawat, Synlett, 2006, 1080.
11. A. Alexakis, D. Polet, S. Rosset, and M. S. March, J. Org. Chem., 2004, 69, 5660. CrossRef