10-Hydroxy-10,9-boroxarophenanthrenes: Versatile synthetic intermediates to 3,4-benzocoumarins and triaryls

Article abstract of DOI:10.1021/jo049343w

10-Hydroxy-10,9-boroxarophenanthrenes were obtained as unexpected major products upon BBr₃-mediated O-demethylation of 2-methoxybiaryls. The formation likely proceeds via intramolecular electrophilic aromatic cyclization of a reactive dibromoaryloxyborane intermediate. Essentially quantitative yields of 10-hydroxy-10,9-boroxarophenanthrenes were also obtained from 2-hydroxybiaryl and BCl₃/AlCl₃ with use of a modified literature procedure. As synthetic intermediates, 10-hydroxy-10,9- boroxarophenanthrenes were efficiently converted to 3,4-benzocoumarins and triaryls through Pd-catalyzed CO insertion and Suzuki reaction.

Full text of DOI:10.1021/jo049343w

SCHEME 1
10-Hyd r oxy-10,9-bor oxa r op h en a n th r en es:
Ver sa tile Syn th etic In ter m ed ia tes to
3,4-Ben zocou m a r in s a n d Tr ia r yls
Q. J ean Zhou,* Karin Worm, and Roland E. Dolle
Department of Chemistry, Adolor Corporation, 700
Pennsylvania Drive, Exton, Pennsylvania 19341-1127
jzhou@adolor.com
Received April 19, 2004
Abst r a ct : 10-Hydroxy-10,9-boroxarophenanthrenes were
obtained as unexpected major products upon BBr3-mediated
O-demethylation of 2-methoxybiaryls. The formation likely
proceeds via intramolecular electrophilic aromatic cyclization
of a reactive dibromoaryloxyborane intermediate. Essentially
quantitative yields of 10-hydroxy-10,9-boroxarophenan-
threnes were also obtained from 2-hydroxybiaryl and BCl3/
AlCl3 with use of a modified literature procedure. As
synthetic intermediates, 10-hydroxy-10,9-boroxarophen-
anthrenes were efficiently converted to 3,4-benzocoumarins
and triaryls through Pd-catalyzed CO insertion and Suzuki
reaction.
SCHEME 2
While carrying out the O-demethylation of 1 with BBr3
under standard reaction conditions, we anticipated its
clean conversion to cannabinoid mimetic 2 (Scheme 1).1
To our surprise, an unknown compound 3 was isolated
in 65% yield while the desired product 2 was isolated in
only ∼10% yield. Although the demethylation reaction
was later accomplished in high yield with use of TMSI,
we were intrigued by what had transpired during the
BBr3 treatment and by the structure of the anomalous
product.
LC/MS analysis revealed that the unknown product
was a single compound containing one boron atom. The
material was stable to both strong aqueous acid and base
at 50 °C. It was smoothly transformed via oxidative
hydrolysis to a second unknown (94% yield), which was
readily identified as 2,2′-dihydroxybiphenyl 4 by 1H
NMR, 13C NMR, and LC/MS analysis (Scheme 1). This
latter transformation indicated that the boron atom was
attached directly to an aryl carbon atom. From these and
spectroscopic studies, the unknown compound was as-
required to facilitate boroxarene formation. The reaction
mechanism is believed to proceed through a common
dibromophenoxyborane intermediate (6) which, in the
absence of the 3′-methyl group (6a ,b), hydrolyzes to give
phenol. In the presence of the 3′-methyl group (6c), an
alternative pathway ensues: intramolecular electrophilic
aromatic substitution, re-aromatization upon loss of HBr,
then hydrolysis (8 f 9 f 10). The strict electronic
requirement within the biaryl substrate may explain why
boroxarene formation has never been reported as a side
reaction during BBr3-mediated O-demethylation.
The parent boroxarene 13 was previously prepared by
Dewar3 in 1960 upon treatment of 2-phenylphenol 7a
with BCl3 gas in hexane followed by the addition of AlCl3
(Scheme 3). Bridger4 subsequently found that the step-
wise addition of reagents was unnecessary. The yield in
both reports was ca. 65%. Boroxarenes have found limited
application in the literature, including their use as
antioxidant lubricant additives,5 Lewis acids for the aldol
condensation,6 and fungicides.7 However, there are no
reports employing boroxarenes as synthetic intermedi-
ates.
signed structure 3,
a substituted 10-hydroxy-10,9-
boroxarophenanthrene.2
To investigate the unexpected formation of 3 further,
we prepared 2-methoxybiphenyls 5a -c (Scheme 2), and
subjected them to the same reaction conditions as com-
pound 1. Interestingly, 5a ,b gave the anticipated phenols
7a ,b (100% yield), while 5c furnished 10 in quantitative
yield. It was clear that a 3′-methyl substituent was
(1) (a) Worm, K. I.; Zhou, Q. J .; Dolle, R. E. Solid-Phase Approach
To Biaryl Cannabinoid Mimetics; Abstracts of the 224th National
Meeting of the American Chemical Society, Boston, MA; American
Chemical Society: Washington, DC, 2002; ORGN 163. (b) Gareau, Y.;
Dufresne, C.; Gallant, M.; Rochette, C.; Sawyer, N.; Slipetz, D. M.;
Tremblay, N.; Weech, P. K.; Metters, K. M.; Labelle, M. Bioorg. Med.
Chem. Lett. 1996, 6, 189-194.
(3) Dewar, M. J . S.; Dietz, R. J . J . Chem. Soc. 1960, 1344.
(4) Bridger, R. F. U.S. Patent 4 210 599, 1980.
(5) Braid, M. U.S. Patent 4 353 807, 1982.
(6) Davis, F. A.; Dewar, M. J . S. J . Org. Chem. 1968, 8, 3324.
(7) Kohn, G. K.; McMurtry, R. J . U.S. Patent 3 686 398, 1972.
(2) This class of compound is collectively referred to herein as
boroxarenes.
10.1021/jo049343w CCC: $27.50 © 2004 American Chemical Society
Published on Web 06/25/2004
J . Org. Chem. 2004, 69, 5147-5149
5147

SCHEME 3
SCHEME 4
TABLE 1. Resu lts of Bor oxa r en e F or m a tion
yield
(%)
boroxarene
R1
R2
R3
electrophilic substitution occurs has a strong negative
effect on the cyclization process, resulting in either
marginal or no production of boroxarene (18, 19, 20). The
presence of an electron-donating group (EDG) para to the
site of electrophilic attack dramatically counteracts the
influence of an EWG group as observed for the formation
of 21 (80% versus 0% for 19). In contrast, the presence
of an EWG or an EDG in the phenolic ring has no
influence on boroxarene synthesis (16, 17). The failure
to isolate bisboroxarene 22 (recovered starting material
only) may result from unfavorable steric interactions of
the putative 2,2′-bis(dichloroboronite)biphenyl and/or
lack of electron-donating substituents as a driving force.
Finally, the exclusive formation of products 14 and 218
serves to demonstrate the regioselectivity of the intra-
molecular cyclization (less hindered sp2 carbon atom).
Scheme 4 displays three chemical transformations of
boroxarene 3. The oxidative hydrolysis, which was men-
tioned in Scheme 1, converted 3 into 2,2′-dihydroxy
compound 4 with 94% yield. Suzuki coupling with
3-methoxyphenyl iodide gave triaryl derivative 24 in 63%
yield, despite the potential for steric hindrance. This
particular reaction was carried out in the microwave;
however, we did not find a dramatic difference between
conventional heating methods and microwave conditions.
Palladium-catalyzed carbonylation occurred in good yield
as evidenced by isolation of the tricyclic lactone 25 (87%
yield). None of the putative intermediate methyl ester
was observed, suggesting facile conversion to the benzo-
coumarin ring system under the reaction conditions.
This preliminary success in using boroxarene 3 as a
synthetic intermediate encouraged us to evaluate similar
transformations for the boroxarenes listed in Table 1.
3,4-Benzocoumarin 26 and its derivatives are naturally
occurring biologically active agents possessing valuable
pharmaceutical properties.9 Previous synthetic approaches
to 3,4-benzocoumarins employed either strong oxidative
conditions10 or elaborate preconstruction of appropriately
functionalized precursors in such a way that the final
step (benzocoumarin formation) could be carried out in
a mild and selective fashion.11 As summarized in Table
13
14
15
16
17
18
19
20
21
22
H
H
CH3
H
H
Cl
F
H
H
CH3
H
H
H
H
F
CH3
CH3
H
H
H
CH3
F
H
H
H
H
99a,b
98a
98a
96a
99a
15c
0c
0c
F
80c
0c,d
23
80e
a
b
Isolated yield. Identical yield was obtained from 2-methoxy-
biphenyl with 3 equiv of BCl3. c Determined by LC/MS. 3 equiv
d
of BCl3, 8 mol % of AlCl3, 90 °C. e From 4-(1,1′-dimethylheptyl)-
3′-methoxy-biphenyl-2-ol, using 3 equiv of BCl3.
In an attempt to have broader access to boroxarenes
to evaluate their potential utility in synthesis, we rein-
vestigated the original literature procedure for the
preparation of 13. After a thorough study, we developed
an optimized procedure. The literature’s one-pot protocol
slowly passed BCl3 gas into a solution of 7a in hexane at
room temperature (Scheme 3). Under these conditions
BCl3 is initially the limiting reagent and the putative
intermediate 2-biphenyl chloroboronite 11 may react with
the starting material to form a boronate dimer/trimer
(the major side product of the reaction). The undesired
dimer/trimer cannot undergo intramolecular cyclization
to form boroxarene, but rather is hydrolyzed upon aque-
ous workup regenerating starting phenol. We reasoned
that by keeping BCl3 in excess throughout the reaction,
this may minimize dimer/trimer formation and poten-
tially improve the reaction yield. When this small modi-
fication in protocol was carried out, i.e., the addition of
the phenol 7a into a dilute solution of BCl3 in hexane
(ca. 0.1 M) followed by addition of AlCl3 (∼4 mol %), a
quantitative yield of 13 was obtained. Strict control of
anhydrous conditions as well as the use of fresh reagents
was essential for a successful reaction. The optimized
reaction conditions were then applied to a range of
2-arylphenols.
(8) Structures were confirmed by 1H NMR, 13C NMR, and MS. See
Supporting Information.
(9) (a) Gringauz, A. J . Pharm. Sci. 1976, 65, 291. (b) Liu, H.;
Santostefano, M.; Lu, Y.; Safe, S. Organohalogen Compd. 1993, 13.
(c) Weidner-Wells, M. A.; Altom, J .; Fernandez, J .; Fraga-Spano, S.
A.; Hilliard, J .; Ohemeng, K.; Barrett, J . F. Bioorg. Med. Chem. Lett.
1998, 8, 97. (d) Douros, J . D., J r.; McNelis, E. J . U.S. Patent 3 248 286,
1966.
Table 1 displays the substrates and the yields of the
corresponding boroxarenes. The presence of an electron-
withdrawing group (EWG) in the pendent aryl ring where
5148 J . Org. Chem., Vol. 69, No. 15, 2004

TABLE 2. Resu lts of Ca r bon yla tion
SCHEME 5
product
R1
R2
R3
yield (%)
26
27
28
29
30
31
H
H
CH3
H
H
H
CH3
H
H
H
H
H
H
CH3
F
90a
100a,b
96a
a
2-3 equiv of BCl3, 4-8 mol % of AlCl3, 50-80 °C for 16 h.
b(1) 10 mol % of Pd(OAc)2 , 20 mol % of DPPF, Et3N; (2) 1 equiv
of Pd(OAc)2. Both conditions gave similar results. cWhen R ) Me,
heating in acetic acid at 50 °C for 16 h resulted in 49% yield.
80a
99a
Cl
H
H
95c
a
b
Isolated yield. Identical yield was obtained by using 10 mol
% of Pd(OAc)2, 20 mol % of DPPF, and 3 equiv of Et3N. c Deter-
mined by LC/MS.
10-hydroxy-10,9-borazarophenanthrenes 37 and 38
(Scheme 5). Quantitative conversion was achieved under
relatively forcing conditions (2-3 equiv of BCl3, 4-8 mol
% of AlCl3, 50-90 °C, 16 h). However, the borazarenes
so obtained showed aggregate characteristics: higher
masses present in the LC/MS and complex proton NMR
spectra. Carbonylation of these aggregates generated the
desired lactams 39 and 40 in only 17% yield as deter-
mined by LC/MS.
TABLE 3. Resu lts of Tr ia r yl F or m a tion
Harris et al. reportedly isolated monomer 37 by re-
crystallizing from 50% aqueous acetic acid.12 Considering
this, the carbonylation of 38 was carried out in acetic acid
at 50 °C overnight. The yield of lactam 40 significantly
increased to 49%. Lactam 40 is a natural product known
as phenaglydon that was subsequently isolated and its
structure was confirmed by comparing the 1H and 13C
NMR and MS of the synthetic material with the litera-
ture data.13
In summary, we demonstrated regioselective borox-
arene formation from 2-hydroxybiphenyl and 2-methoxy-
biphenyl under mild conditions. As synthetic intermedi-
ates, boroxarenes were converted to 3,4-benzocoumarins
through CO insertion and triaryls through Suzuki cou-
pling. Application of this chemistry to commercially
available 3′-methyl-biphenyl-2-ylamine resulted in an
efficient two-step synthesis of phenaglydon.
product
R
Ar
yield (%)
32
33
34
CH3
H
H
Ph
81a
65b
64b
m-CHO-Ph
p-MeO-Ph
a
b
Isolated yield. Determined by LC/MS.
2, the Pd-catalyzed CO insertion of boroxarenes provides
an alternative method for their construction under mild
conditions with high regioselectivity. In this reaction, 1
equiv of Pd(OAc)2 was used and the reactions were
essentially complete within 2-4 h, affording 3,4-benzo-
coumarins in excellent yields. Little or no purification
was needed other than a simple aqueous workup and
filtration. This reaction proceeds with equal ease and
efficiency with use of a catalytic amount of the Pd(OAc)2
(10 mol %) accompanied by DPPF (20 mol %).
The Pd-catalyzed Suzuki coupling of boroxarene was
likewise straightforward. Three additional examples 32-
34 were carried out under conventional conditions with
results virtually identical with those found for 24 as
summarized in Table 3.
Ack n ow led gm en t. We thank Clay Rutledge and
Bill Barker for their valuable NMR and LC/MS service.
Su p p or tin g In for m a tion Ava ila ble: Experimental de-
tails for compounds 3, 4, 7a , 7b, 10, 13-18, 21, 23-34, and
37-40. This material is available free of charge via the
Internet at http://pubs.acs.org.
The optimized reaction conditions were also applied to
2-aminobiphenyls (35 and 36), attempting to prepare the
J O049343W
(10) Migachev, G. I.; Andrievskii, A. M.; Poplavskii, A. N.; Dokuni-
khin, N. S. U.S. Patent 591 473, 1978. (b) Olah, G. A.; Wang, Q.;
Trivedi, N. J .; Prakash, G. K. S. Synthesis 1991, 9, 739.
(12) Harris, K. D. M.; Kariuki, B. M.; Lambropoulos, C.; Philp, D.;
Robinson, J . M. A. Tetrahedron 1997, 53, 8599.
(13) Wurz, G.; Hofer, O.; Greger, H. Nat. Prod. Lett. 1993, 3, 177.
(11) Harayama, T.; Yasuda, H. Heterocycles 1997, 46, 61.
J . Org. Chem, Vol. 69, No. 15, 2004 5149