Novel synthesis of desymmetrized resorcinol derivatives: Aryl fluoride displacement on deactivated substrates

Article abstract of DOI:10.1021/jo0523868

A short, high-yielding synthesis of differentially substituted resorcinol derivatives has been developed that utilizes 1,3-difluorobenzene as the starting material and employs sequential nucleophilic aromatic substitution (S _NAr) reactions to generate desymmetrized products. The scope and limitations of the second S_NAr reaction on the deactivated 1-alkoxy-3-fluorobenzene intermediates have been investigated. This methodology has also been employed in the synthesis of desymmetrized catechol derivatives from 1,2-difluorobenzene.

Full text of DOI:10.1021/jo0523868

commercially available 3-methoxyphenol;5 however, highly
substituted aryl ethers cannot be synthesized by this method.
Metal-catalyzed coupling reactions have been carried out on
3-haloanisoles to generate 3-methoxy-substituted desymmetrized
resorcinols. Although many of these coupling reactions are
limited to tertiary alcohols,6 examples of metal-catalyzed
couplings of primary and secondary alcohols and 1-alkoxy-3-
halobenzenes have been described.1b,7
Given this background, we felt that the investigation of a
new route of entry into differentially alkylated resorcinols was
warranted. Commercially available 1,3-difluorobenzene ap-
peared to be an ideal starting material to access these types of
molecules. We reasoned that the first fluoride displacement by
an alkoxide should generate a desymmetrized 1-alkoxy-3-
fluorobenzene intermediate which is deactivated toward further
alkoxide displacement because of the electron-donating ability
of the 1-alkoxy substituent. Subjecting this intermediate to more
vigorous displacement conditions could then provide the desired
resorcinol products.
Novel Synthesis of Desymmetrized Resorcinol
Derivatives: Aryl Fluoride Displacement on
Deactivated Substrates
Aujin Kim,† Jeremiah D. Powers, and Jennifer F. Toczko*
Synthetic Chemistry, Chemical DeVelopment, GlaxoSmithKline,
FiVe Moore DriVe, P.O. Box 13398, Research Triangle Park,
North Carolina 27709
jennifer.f.toczko@gsk.com
ReceiVed NoVember 18, 2005
Although the displacement of aryl fluorides on electron-
deficient aromatic rings is well documented, the second fluoride
displacement, which will take place on an electron-rich aromatic
ring, is less well-known.8 Current literature methods of fluoride
displacement by an alkoxide on deactivated substrates include
activation of the alkoxy fluorobenzene by complexation with
chromium(III) prior to fluoride displacement, followed by
oxidative removal of the metal,9 and photochemical conditions.10
Seeing an opportunity to develop a short, efficient route to
differentially substituted resorcinols, we set about exploring their
synthesis from 1,3-difluorobenzene (1).
Desymmetrization of 1 with a selective SNAr reaction using
benzyl alcohol or methanol generates 2a,b in high yield.
Over the course of our work, we observed that the addition
of the polar, aprotic cosolvent 1,3-dimethyl-3,4,5,6-tetrahydro-
2(1H)-pyrimidinone (DMPU) consistently resulted in the con-
sumption of the starting material and decreased the reaction time,
allowing for better control over the formation of the bis-addition
byproducts. Using the potassium alkoxide generated from
KOtBu (2.4-2.5 equiv) and alcohol (3 equiv) in a toluene/
DMPU or a DME/DMPU solvent mixture provided the best
results (Table 1). DME was chosen as the cosolvent in the
3-fluoroanisole (2b) synthesis because of the fact that the lower
boiling point allowed for easier isolation of the product.
With 2a,b in hand, we next turned our attention to developing
conditions to accomplish the second nucleophilic aromatic
A short, high-yielding synthesis of differentially substituted
resorcinol derivatives has been developed that utilizes 1,3-
difluorobenzene as the starting material and employs se-
quential nucleophilic aromatic substitution (SNAr) reactions
to generate desymmetrized products. The scope and limita-
tions of the second SNAr reaction on the deactivated
1-alkoxy-3-fluorobenzene intermediates have been investi-
gated. This methodology has also been employed in the
synthesis of desymmetrized catechol derivatives from 1,2-
difluorobenzene.
In the course of our work, it became necessary to synthesize
a variety of nonsymmetrical resorcinol derivatives. A survey
of the literature showed that this is a nontrivial problem,1 and
traditional methods of differentiating the alcohol moieties on
resorcinol have numerous drawbacks. A classical monoalkyla-
tion approach requires a large excess of resorcinol to avoid
overalkylation.1b,2 When nearly stoichiometric amounts of
resorcinol and electrophile are used, the alkylation reaction is
unselective and low yielding, resulting in a mixture of resorcinol,
mono-, and bis-alkylated products.3
Desymmetrization can be achieved by the selective depro-
tection of several bis-protected resorcinols;4 however, syntheses
of differentiated 1,3-alkoxybenzenes from these starting materi-
als require numerous protecting group manipulations. Differenti-
ated resorcinols have also been generated by the alkylation of
(5) Punna, S.; Meunier, S.; Finn, M. G. Org. Lett. 2004, 6, 2777.
(6) (a) Shelby, Q.; Kataoka, N.; Mann, G.; Hartwig, J. J. Am. Chem.
Soc. 2000, 122, 10718. (b) Parrish, C. A.; Buchwald, S. L. J. Org. Chem.
2001, 66, 2498. (c) Watanabe, M.; Nishiyama, M.; Koie, Y. Tetrahedron
Lett. 1999, 40, 8837.
(7) (a) Torraca, K. E.; Huang, X.; Parrish, C. A.; Buchwald, S. L. J.
Am. Chem. Soc. 2001, 123, 10770. (b) Wolter, M.; Nordmann, G.; Job, G.
E.; Buchwald, S. L. Org. Lett. 2002, 4, 973.
(8) (a) Kumazawa, K.; Ishihara, K.; Yamamoto, H. Org. Lett. 2004, 6,
2551. Although these reaction conditions generate the desired desymmetrized
product, they are unacceptable from a scale up perspective because of safety
concerns over the potential interaction between NaH and DMF (see Chem.
Eng. News 1982, 60, 5, 43). (b) Orjales, A.; Mosquera, R.; Toledo, A.;
Pumar, M. C.; Garc´ıa, N.; Cortizo, L.; Labeaga, L.; Innera´rity, A. J. Med.
Chem. 2003, 46, 5512.
(9) Brenner, E.; Baldwin, R. M.; Tamagnan, G. Org. Lett. 2005, 7, 937.
(10) (a) Zhang, G.; Wan, P. Chem. Commun. 1994, 19. (b) Zhang, B.;
Zhang, J.; Yang, D.-D. H.; Yang, N. C. J. Org. Chem. 1996, 61, 3236.
† GlaxoSmithKline Summer Intern, 2005.
(1) (a) Hoarau, C.; Pettus, T. R. R. Synlett 2003, 1, 127. (b) Boxhall, J.
Y.; Page, P. C. B.; Chan, Y.; Hayman, C. M.; Heaney, H.; McGrath, M. J.
Synlett 2003, 7, 997.
(2) Wittig, T. W.; Decker, M.; Lehmann, J. J. Med. Chem. 2004, 47,
4155.
(3) (a) Prasad, K.; Xu, D. D.; Tempkin, O.; Villhauer, E. B.; Repie`, O.
Org. Process Res. DeV. 2003, 7, 743. (b) Simas, A. B. C.; Furtado, L. F.
O.; Costa, P. R. R. Tetrahedron Lett. 2002, 43, 6893. (c) Srinivasan, M.;
Sankararaman, S.; Hopf, H.; Dix, I.; Jones, P. G. J. Org. Chem. 2001, 66,
4299.
(4) (a) Loubinoux, B.; Coudert, G.; Guillaumet, G. Synthesis 1980, 638.
(b) Zaugg, H. E. J. Org. Chem. 1976, 41, 3419.
10.1021/jo0523868 CCC: $33.50 © 2006 American Chemical Society
Published on Web 02/03/2006
2170
J. Org. Chem. 2006, 71, 2170-2172

TABLE 1. Selective Fluoride Displacement on 1,3-Difluorobenzene
TABLE 3. Selective Fluoride Displacement on 1,2- and
1,4-Difluorobenzene
entry
solvent
time (h)
yield (%)
entry
difluoride
time (h)
yield (%)
1
2
toluene/DMPU
DME/DMPU
3.5
3
2a, 89
2b, 80
1
2
8
9
3
5
10, 77a
11, 25b
a Reaction conditions: KOtBu (2.5 equiv), toluene (5 vol), DMPU (2
vol), BnOH (3 equiv), 80 °C, 3 h. b Reaction conditions: KOtBu (1 equiv),
toluene (5 vol), DMPU (2 vol), BnOH (1.2 equiv), 80 °C, 5 h.
TABLE 2. Nucleophilic Aromatic Substitution of 2a,b
TABLE 4. Nucleophilic Aromatic Substitution on 10
a Reaction conditions: KOtBu (2.5 equiv), ROH (3 equiv), toluene (10
vol), DMPU (2 vol), 100 °C, 18 h. b Reaction conditions: KOtBu (3.8
equiv), ROH (5 equiv), toluene (6 vol), DMPU (2 vol), 100 °C, 29 h.
c Reaction conditions: KOtBu (4 equiv), ROH (5 equiv), toluene (5 vol),
DMPU (2 vol), 100 °C, 44 h.
both phenol and 2,2,2-trifluoroethanol failed to undergo reaction
with both starting material 1 and the 1-alkoxy-3-fluorobenzene
substrates 2a,b.
After successfully demonstrating this methodology on 1,3
systems, we applied these conditions to 1,2- and 1,4-difluo-
robenzene (8 and 9). The first SNAr reaction with benzyl alcohol
and 8 proceeded in good yield to generate 10 in an analogous
manner to the 1,3-difluorobenzene substrate. However, the
reaction in the 1,4-difluoro system proceeded poorly and
provided 11 in only 25% yield (Table 3).
It is known that in aromatic systems halogens can destabilize
negative charges on the carbon to which they are attached.
Although this is contrary to the normal electron-withdrawing
ability of the halogens, in the case of aromatic systems, the
effect arises because the developing negative charge is in a
π-system. Moreover, the π-negative charge is greatest at the
position para to the point of nucleophilic attack.12 Therefore, a
fluoride atom para to the point of attack is substantially less
activating than fluoride atoms in meta and ortho positions. The
poor yield in the 1,4-difluoro system is likely due to this para-
fluoro deactivating effect. Nevertheless, with desymmetrized
intermediates 10 and 11 in hand, we turned our attention to the
second displacement.
displacement with a variety of alkoxides to generate differen-
tially substituted resorcinol products in good yield. The results
are summarized in Table 2.
Again, over the course of our investigations, we found that
the use of DMPU is key to this displacement as well. In the
absence of DMPU, we had difficulty driving the reaction to
completion, even with very high alkoxide loadings and extended
reaction times. Optimal conditions found are KOtBu (3.5-4
equiv) and alcohol (5 equiv) in a toluene/DMPU solvent mixture
at 100 °C. These operationally simple and straightforward
reaction conditions proved to be successful with a variety of
primary alcohols (entries 1-6) as well as with the more
sterically hindered neopentyl (entries 7 and 8) and secondary
(entries 9 and 10) examples.
Although alkoxides can engage in single electron transfer
(SET) processes,11 we believe that the reactions described in
this Note proceed through a SNAr and not a SET mechanism
because of the fact that we see no hydrodehalogenation or biaryl
coupled products. Additionally, the cyclopropyl group (entries
3 and 4) remains intact, providing evidence that a SET
mechanism is not taking place. Moreover, no cine substitution
products have been observed that would indicate the formation
of a benzyne intermediate.
The second displacement on 10 can be carried out in high
yield, using the same general reaction conditions as those
A limitation of this methodology is an intolerance of alcohols
that are electron deficient. For example, alkoxides derived from
(12) (a) Burdon, J. Tetrahedron 1965, 21, 3373. (b) Chambers, R. D.;
Close, D.; Williams, D. L. H. J. Chem. Soc., Perkin Trans. 2 1980, 778.
(11) Guthrie, R. D.; Nutter, D. E. J. Am. Chem. Soc. 1982, 104, 7478.
J. Org. Chem, Vol. 71, No. 5, 2006 2171

2917, 2931, 3034, 3067, 3090 cm-1. Anal. Calcd for C13H11OF:
C, 77.21; H, 5.48; O, 7.91; F, 9.1. Found: C, 77.34; H, 5.57; O,
8.06; F, 9.43.
employed in the 1,3 systems, using primary, secondary, and
neopentyl alcohols (Table 4) to generate the differentiated
catechol products 12a-c.
Typical Procedure for the Reaction of an Alkoxide with
1-Alkoxy-3-fluorobenzene: Preparation of 1-(Butyloxy)-3-(meth-
yloxy)benzene (3b). To a solution of KOtBu (3.5 g, 30.3 mmol)
and toluene (5 mL) was slowly added 1-butanol (3.8 mL, 42 mmol),
followed by DMPU (2 mL). The solution was heated at 100 °C for
30 min. 3-Fluoroanisole (2b, 0.90 mL, 7.9 mmol) was then added,
and the solution was heated for an additional 21 h at 100 °C. The
reaction mixture was cooled to room temperature and washed with
water, 10% brine (2×), and 6 N HCl. The combined aqueous layers
were extracted with EtOAc, and the combined organic layers were
dried and concentrated. The resulting oil was purified by flash
chromatography (2% EtOAc in hexanes) to produce 1.0 g (71%)
of 3b as a colorless oil.
Not surprisingly, 4-fluorophenyl phenylmethyl ether 11 fails
in the second displacement reaction. The difference in reactivi-
ties between 10 and 11 is likely due to the same reasoning put
forth for the reduced reactivity of difluoride 9. In this case, the
deactivating effect of the p-benzyloxy group is even greater than
that of the p-fluoride and no reaction takes place.
In conclusion, we have described an operationally simple and
straightforward method for the synthesis of desymmetrized
resorcinol and catechol derivatives using successive SNAr
reactions that employ aryl fluorides and a variety of alkoxide
nucleophiles.
1H NMR: δ 0.96 (t, 3H, J ) 7.4 Hz), 1.41-1.54 (m, 2H), 1.70-
1.80 (m, 2H), 3.78 (s, 3H), 3.93 (t, 2H, J ) 6.5 Hz) 6.45-6.51 (m,
3H), 7.16 (t, 1H, J ) 8 Hz). 13C NMR: δ 13.6, 18.7, 30.7, 55.0,
67.0, 100.6, 106.1, 106.5, 129.8, 159.9, 160.5. IR: 1042, 1148,
1199, 1264, 1286, 1453, 1492, 1590, 2835, 2872, 2935, 2985, 3030,
3067, 3090 cm-1. Anal. Calcd for C11H16O2: C, 73.30; H, 8.95.
Found: C, 73.23; H, 9.04.
Experimental Section
Typical Preparation of 1-Alkoxy-3-fluorobenzenes. Prepara-
tion of 3-Fluorophenyl Phenylmethyl Ether (2a). To a solution
of KOtBu (24.6 g, 219 mmol) and toluene (50 mL) was slowly
added benzyl alcohol (29 mL, 280 mmol), followed by DMPU (20
mL). The solution was heated at 80 °C for 30 min. 1,3-
Difluorobenzene (1, 9 mL, 92 mmol) was then added, and the
solution was heated for an additional 3.5 h at 80 °C. The reaction
mixture was cooled to room temperature, washed with water and
10% brine (2×), dried, and concentrated. The resulting oil was
purified by flash chromatography (1% EtOAc in hexanes) to
produce 16.7 g (89%) of 2a as a colorless oil.
Acknowledgment. Analytical chemistry support provided
by Joanne Anderson of GlaxoSmithKline is gratefully acknowl-
edged.
1H NMR: δ 5.05 (s, 2H), 6.64-6.78 (m, 3H), 7.23 (q, 1H, J )
8.5, 15.2 Hz), 7.31-7.45 (m, 5H). 13C NMR (125 MHz, CDCl3):
δ 70.2, 102.6 (d, J ) 24 Hz), 107.7 (d, J ) 21 Hz), 110.6, 127.5,
128.1, 128.6, 130.2, 136.5, 160.1 (d, J ) 11 Hz), 163.6 (d, J )
245 Hz). IR: 1025, 1164, 1262, 1277, 1488, 1592, 1610, 2873,
Supporting Information Available: Experimental details and
characterization data for all reported compounds. This material is
available free of charge via the Internet at http://pubs.acs.org.
JO0523868
2172 J. Org. Chem., Vol. 71, No. 5, 2006