Photostimulated reactions of phenylacetic acid dianions with aryl halides. Influence of the metallic cation on the regiochemistry of arylation.

Article abstract of DOI:10.1021/ol006177f

[reaction: see text]Phenylacetic acid dianions react via what appears to be an S(RN)1 process with aryl halides under photostimulation to afford aryl substitution products 5 and 6. When the counterion is K+, only 4-biphenylacetic acids 5 are obtained. Both alpha- and para-coupling occurs with Na+ to give a mixture of 5 and 6, while exclusive formation of diphenylacetic acids 6 is observed with the dilithio salt of 1.

Full text of DOI:10.1021/ol006177f

ORGANIC
LETTERS
2000
Vol. 2, No. 17
2643-2646
Photostimulated Reactions of
Phenylacetic Acid Dianions with Aryl
Halides. Influence of the Metallic Cation
on the Regiochemistry of Arylation
Godson C. Nwokogu,† Jim-Wah Wong, Thomas D. Greenwood, and
James F. Wolfe*
Department of Chemistry and the HarVey W. Peters Research Center for the Study of
Parkinson’s Disease and Disorders of the Central NerVous System,
Virginia Polytechnic Institute and State UniVersity, Blacksburg, Virginia 24061
jwolfe@chemserVer.chem.Vt.edu
Received June 8, 2000
ABSTRACT
Phenylacetic acid dianions react via what appears to be an SRN1 process with aryl halides under photostimulation to afford aryl substitution
products 5 and 6. When the counterion is K+, only 4-biphenylacetic acids 5 are obtained. Both r- and para-coupling occurs with Na+ to give
a mixture of 5 and 6, while exclusive formation of diphenylacetic acids 6 is observed with the dilithio salt of 1.
As part of a continuing study of radical-chain nucleophilic
aromatic substitution reactions, we had discovered previously
that photoinduced reactions of aryl and hetaryl halides with
phenyl-stabilized carbanions such as the potassium enolate
of ethyl phenylacetate (1a)1,2 and potassiophenylacetonitrile
(1b)3-5 proceed by the radical-chain SRN16 mechanism to
form mainly products 2a,b, derived from coupling of aryl
radicals at the R-position (Scheme 1). However, while ester
enolate 1a reacted to give only R-substituted product 2a,
potassionitrile 1b afforded small amounts (<10% total) of
ortho- and para-substituted products 3 and 4, respectively.
These results, which represent rare examples of aryl-to-aryl
Scheme 1
† Hampton University, Department of Chemistry, Hampton, VA 23668.
(1) Wong, J.-W.; Natalie, K. J.; Nwokogu, G. C.; Pisipati, J. S.; Flaherty,
P. T.; Greenwood, T. D.; Wolfe, J. F. J. Org. Chem. 1997, 62, 6152-
6159.
(2) Wong, J.-W. Ph.D. Dissertation, Virginia Polytechnic Institute and
State University, 1981.
(3) Hermann, C. K. F.; Sachdeva, Y. P.; Wolfe, J. F. J. Heterocycl. Chem.
1984, 24, 1061-1065.
(4) Hermann, C. K. F. Ph.D. Dissertation, Virginia Polytechnic Institute
and State University, 1984.
(5) Bunnett, J. F.; Gloor, B. F. J. Org. Chem. 1973, 38, 4156-4163.
(6) For reviews, see: (a) Rossi, R. A.; de Rossi, R. H. Aromatic
Substitution by the SRN1 Mechanism; ACS Monograph 178; American
Chemical Society: Washington, DC, 1983. (b) Norris, R. K. Compr. Org.
Synth. 1991, 4, 451-482. (c) Rossi, R. A.; Pierini, A. B.; Santiago, A. N.
Org. React. 1999, 54, 1-271.
10.1021/ol006177f CCC: $19.00 © 2000 American Chemical Society
Published on Web 07/26/2000

coupling in SRN1 reactions involving phenyl-stabilized
carbanions,7-9 prompted us to investigate the possibility that
other carbanions possessing a stabilizing phenyl substituent
might exhibit analogous reactivity and regiochemistry.
In the present study, the dipotassio salt of phenylacetic
acid dianion (1c) was chosen as the representative nucleo-
phile to examine the characteristics of photoinduced aryla-
tions. The choice of 1c was based on the fact that even
though there appear to be no published examples of car-
boxylic acid dianions participating as nucleophiles in single
electron transfer (SET)-initiated reactions of the SRN1 type,
the electrochemically induced oxidative R , para-dimerization
of certain aryl carboxylic acid dianions has been demon-
strated.10 Furthermore, empirical calculations of electron
Table 1. Reactions of Aryl Halides with Dianions 1c-ea
entry
aryl halide
dianion
product (yield, %)
1
2
3
4
5
6
7
8
PhI
PhI
1c
1e
1c
1c
1c
1c
1d
1e
5a (73)
6a (77)
5b (31)
5c (47)
5d (49)
5e (64)
5e (25), 6b (37)
6b (61)
2-MePhBr
3-MePhBr
4-MePhBr
4-MeOPhBr
4-MeOPhBr
4-MeOPhBr
a All reactions were performed using 2-3 equiv of phenylacetic acid
dianion relative to the aryl halide in liquid NH3 solution and irradiation for
5 h with 350 nm lamps in a Rayonet RPR-240 photochemical reactor.
density distributions in phenylacetic acid dianions using 13
C
chemical shift data suggest that the para-position of potassio
salt 1c possesses ca. 40% as much negative charge as the
R-position,11 which could make the former a possible site
for reaction with aryl radicals.
We now wish to report that photostimulated reactions of
dianion 1c with a variety of aryl halides results in regio-
specific arylation to afford exclusively para-coupled products
5a-e12 in yields ranging from 31% to 73% (eq 1,Table 1).
ortho-substituents in the aryl halide resulted in either
diminished product yields (entry 3) or no reaction at all, as
in the case of bromomesitylene, 4-iodonitrobenzene, and
4-bromo-N,N-diethylaniline.
To assess the influence of phenyl substituents on the
reactivity of dianion 1c, photostimulated reactions of dipo-
tassio salts 7a,b were carried out in the presence of
4-bromoanisole. As shown in Scheme 2, the dipotassio salt
Scheme 2
For example, when a solution of dianion 1c, generated by
deprotonation of phenylacetic acid using 2 equiv of KNH2
in liquid NH3,13 was treated with 4-bromotoluene under near-
UV irradiation for 5 h,14 4-biphenylacetic acid 5d was
obtained in 49% isolated yield (entry 5).15 The presence of
(7) The photoassisted R- and para-arylation of triphenylmethyl anion is
reported to proceed by a combination of radical coupling and SRN1
mechanisms; see: Tolbert, L. M.; Martone, D. P. J. Org. Chem. 1983, 48,
1185-1190.
(8) For examples of SRN1 reactions involving ortho and para coupling
of nitranions derived from aromatic amines, see: Kim, J. K.; Bunnett, J. F.
J. Am. Chem. Soc. 1970, 92, 7464-7466. (b) Pierini, A. B.; Baumgartner,
M. T.; Rossi, R. A. Tetrahedron Lett. 1987, 28, 4653-4656.
(9) Numerous examples of ortho- and para-coupling reactions of
phenoxides and naphthoxides proceeding by the SRN1 mechanism have been
reported; see: Rossi, R. A.; Pierini, A. B.; Santiago, A. N. Org. React.
1999, 54, 137-157.
(10) Renaud, P.; Fox, M. A. J. Org. Chem. 1988, 53, 3745-3752.
(11) Lambert, J. B.; Wharry, S. M. J. Am. Chem. Soc. 1982, 104, 5857-
5862.
(12) The absence of the characteristic methine resonance for diarylacetic
acids at ca. 5.0 δ in the 1H NMR spectrum of reaction mixtures precluded
arylation at the R-carbon of 1c.
(13) (a) Hauser, C. R.; Chambers, W. J. J. Am. Chem. Soc. 1956, 78,
4942-4944. (b) Meyer, R. B.; Hauser, C. R. J. Org. Chem. 1961, 26, 3296-
3698.
of 3-methoxyphenylacetic acid (7a) reacted about as ef-
ficiently with 4-bromoanisole as the unsubstituted dianion
1c (entry 6, Table 1), while the 2-methyl-substituted dianion
7b afforded only a low yield of para-coupled product 8b.
the addition was complete, irradiation was continued for 5 h and the reaction
mixture was poured slowly onto NH4Cl (6.42 g, 120 mmol) contained in a
1-L beaker. The ammonia and ether were evaporated, 50 mL each of water
and hexane were added to the solid residue, and the mixture was vigorously
stirred. An insoluble solid that remained was filtered and dissolved in 100
mL of hot water, and the solution was acidified to pH ) 1 with 2 M HCl.
The resulting mixture was cooled to O °C, and the precipitated solid was
filtered, washed with water, and air-dried to obtain 1.56 g (64%) of 5e, mp
182-185 °C. Recrystallization from CH2Cl2/ether gave 5e as colorless
needles, mp 187-188 °C, lit. mp 188-188.5 °C, see Linnell, W. H.; Smith,
H. J. J. Chem. Soc. 1959, 557-559. 1H NMR (CDCl3) δ 3.68 (s, 2H), 3.84
(s, 3H), 6.96 (d, 2H), 7.32 (d, 2H), 7.50 (m, 4H). The aqueous layer from
the original filtration was acidified with 2 M HCl, and the resulting oily
mixture was extracted with 2 × 50 mL portions of ether. The ethereal
solutions were combined, dried (MgSO4), and concentrated to a pale yellow
oil that solidified on standing; 2.47 g of essentially pure phenylacetic acid.
Evaporation of the hexane solution under vacuum left 0.15 g of 4-bro-
moanisole.
(14) Shorter reaction times resulted in lower product yields, e.g., after a
2 h reaction period only 22% of 5d was obtained.
(15) Representative Experimental Procedure. To a solution of KNH2
prepared from potassium (2.35 g, 60 mmol) in 300 mL of liquid ammonia
was added slowly a solution of phenylacetic acid (4.08 g, 30 mmol) in 30
mL of anhydrous ether, and the resulting pale green solution was stirred
for 30 min. The ammonia solution was then irradiated as 4-bromoanisole
(1.87 g, 10 mmol) in 20 mL of ether was added dropwise via syringe. After
2644
Org. Lett., Vol. 2, No. 17, 2000

The dianions of 2-methoxy- and 3-chlorophenylacetic acids
did not undergo para-coupling with 4-bromoanisole, and the
para-blocked compound, 4-bromophenylacetic acid, failed
to yield any of the alternative R- or ortho-coupled products.
1-Bromo- and 1-iodonaphthalene underwent para-coupling
reactions with dianion 1c to give 4-(1-naphthyl)phenylacetic
acid (9) in yields of 34% and 48%, respectively. Similarly,
2-bromonaphthalene afforded a 32% yield of 4-(2-naphthyl)-
phenylacetic acid (10). All of the reactions that produced
para-coupled products were characterized by essentially
complete consumption of the respective aryl halides after 5
h of irradiation, and reductive dehalogenation of the substrate
proved to be a significant side reaction in all cases.
reaction. The similar nature of the photoinduced R-arylations
involving dilithio salt 1e (5 h) including their resistance to
inhibition by up to 20 mol % of DTBN, is consistent with
operation of the same recalcitrant mechanism. The proposed
mechanism for the para-coupling is illustrated in steps 1-5
of Scheme 3. The radical-chain reaction is initiated by SET
Scheme 3
Varying the alkali metal counterion of phenylacetic acid
dianion led to a marked influence on the regioselectivity of
the photoinduced arylation reaction. Thus, while the dipo-
tassio salt, 1c, gave exclusively the para-coupled product
5e with 4-bromoanisole (entry 6, Table 1), when the
counterion was changed to Na+, as in dianion 1d, a 3:2
mixture of R : para-arylation products 6b and 5e, respec-
tively, was obtained (entry 7, Table 1). With Li+ as the
counterion, substitution occurred exclusively at the R-position
of dianion 1e to give a 61% yield of diphenylacetic acid
derivative 6b (entry 8, Table 1). R-Coupling was also the
only reaction observed between dilithio salt 1e and iodo-
benzene (entry 2, Table 1). This surprising counterion effect
may be rationalized in terms of differences in electron
densities at the para-positions of dianions 1c and 1e. The
large, soft K+ ions of 1c bond less tightly to the oxygens of
the carboxylate group than the small, hard Li+ ions of 1e,
resulting in a greater delocalization of charge to the para-
position of the former. The disodium salt 1d represents an
intermediate case.
Some mechanistic aspects of the coupling process became
evident with the finding that reaction of 4-bromotoluene with
dianion 1c failed to occur without near-UV irradiation. When
an irradiated reaction involving these same reactants was
carried out for 2 h in the presence of 20 mol % of the radical
scavenger, di-tert-butyl nitroxide (DTBN),16 the yield of
para-coupled product 5d was not diminished compared with
that obtained in a similar uninhibited photoreaction.14 The
necessity for photostimulation strongly suggests the involve-
ment of a SET mechanism in the arylation reaction. The
relatively long reaction times required17 and the lack of
inhibition by a catalytic amount of DTBN18 imply that the
substitution process is an example of a rather inefficient SRN1
from dianion 11 to ArX under the influence of near-UV
irradiation (step 1). The resulting aryl radical anion dissoci-
ates to give Ar• (step 2), which then undergoes para-coupling
reaction with dianion 11 affording product dianion radical
13 (step 3). Electron transfer from 13 to ArX (step 4)
generates product anion 14 and the aryl radical anion, which
continues the propagating cycle (steps 2-4). The substitution
process then concludes with deprotonation-isomerization of
14 by excess dianion 11 to yield product dianion 15 (step
5). An alternative nonchain mechanism involving the cou-
pling of substrate aryl radicals (Ar•) with radical anion 1219
seems unlikely in light of both the sluggishness of the
arylation reaction and the observed metallic cation effect.
As part of our mechanistic investigation, we considered
possible termination steps involving dianion radical 13 that
would be consistent with both the lethargic nature of the
substitution reaction and reduction of the aryl halide. One
such chain interrupting reaction might involve para hydrogen
atom abstraction from 13 by Ar• to generate product dianion
15 and ArH (step 6). This possibility was subsequently
discounted by the reaction of dipotassio 4-deuteriophenyl-
acetate with 4-bromotoluene, which failed to yield any
4-deuteriotoluene. Another conceivable chain-breaking option
(16) (a) Hoffman, A. K.; Feldman, A. M.; Geblum, E.; Hodgson, W. G.
J. Am. Chem. Soc. 1964, 86, 639-646. (b) Nelsen, S. F.; Bartlett, P. D. J.
Am. Chem. Soc. 1966, 88, 143-149. (c) Carver, D. R.; Hubbard, J. S.;
Wolfe, J. F. J. Org. Chem. 1982, 47, 1036-1040.
(17) By comparison, monoanion 1a reacts with iodobenzene under
essentially identical conditions to give 90% of 2a (ArdPh) after 1 h, see
ref 2.
(18) Two molecular equivalents of DTBN (based on 4-bromotoluene)
were required to completely quench the 2 h reaction.
(19) Galli, C. Tetrahedron 1988, 44, 5205-5208.
Org. Lett., Vol. 2, No. 17, 2000
2645

might be the reduction of Ar• by 13. However, as in step 6,
this entropically unfavorable encounter between two highly
reactive intermediates present in low concentrations seems
unlikely to play an important role in the arylation process.
Another way in which the chain carrying Ar• might be taken
out of the propagation cycle would involve reduction by
dianion 11 to aryl anions, which then undergo protonation
by liquid NH3 to generate ArH. This pathway also seems
improbable since no products expected from dimerization
of radical anion 1210 were detected in any of the arylation
reactions.
In conclusion, the reactions of dipotassio phenylacetic acid
salts 1c and 7a,b with aryl halides described in this study
represent unique examples of photostimulated arylation
occurring at an aromatic carbon remote from the site of
original carbanion generation. This regiochemistry contrasts
with exclusive R-carbon substitution observed in typical two-
electron SN2-type displacement reactions of phenylacetic
dianions.13 Moreover, the metallic cation effect on the
regioselectivity of substitution observed with dialkali salts
1c-e appears to be unprecedented in photoinduced arylations
of carbanionic species.20 Finally, with further optimization,
the para-coupling reactions described here could serve as a
convenient, one-step synthesis of certain 4-biphenylacetic
acids and their derivatives.
Acknowledgment. The authors are pleased to acknowl-
edge the generous support of this work by the Harvey W.
Peters Research Center for Parkinson’s Disease and Disorders
of the Central Nervous System Foundation and by the
National Science Foundation through grant CHE-8513216.
Supporting Information Available: 1H NMR data,
melting points, and literature references for known com-
pounds 5a, 5b, 5d, 5e, 6a, 6b, 8a, and 9b. Additionally,
elemental analyses or HRMS data for new compounds 5c,
8b, and 9a. This material is available free of charge via the
Internet at http://pubs.acs.org.
OL006177F
(20) For a report of the influence of metallic cation on the stereoselectivity
of SRN1 arylation of imide R-enolates, see: Lotz, G. A.; Palacios, S. M.;
Rossi, R. A. Tetrahedron Lett. 1994, 35, 7711-7714.
2646
Org. Lett., Vol. 2, No. 17, 2000