Highly regioselective palladium-catalyzed direct arylation of oxazole at C-2 or C-5 with aryl bromides, chlorides, and triflates

Article abstract of DOI:10.1021/ol1011778

Complementary palladium-catalyzed methods for direct arylation of oxazole with high regioselectivity (>100:1) at both C-5 and C-2 have been developed for a wide range of aryl and heteroaryl bromides, chlorides, iodides, and triflates. C-5 arylation is pre

Full text of DOI:10.1021/ol1011778

ORGANIC
LETTERS
2010
Vol. 12, No. 16
3578-3581
Highly Regioselective Palladium-Catalyzed
Direct Arylation of Oxazole at C-2 or C-5
with Aryl Bromides, Chlorides, and Triflates
Neil A. Strotman,*,† Harry R. Chobanian,*,‡ Yan Guo,‡ Jiafang He,‡ and
Jonathan E. Wilson‡
Department of Process Chemistry and Department of Medicinal Chemistry, Merck
Research Laboratories, Merck & Co., Inc., P.O. Box 2000, Rahway, New Jersey 07065
neil_strotman@merck.com; harry_chobanian@merck.com
Received May 21, 2010
ABSTRACT
Complementary palladium-catalyzed methods for direct arylation of oxazole with high regioselectivity (>100:1) at both C-5 and C-2 have been
developed for a wide range of aryl and heteroaryl bromides, chlorides, iodides, and triflates. C-5 arylation is preferred in polar solvents with
phosphines 5 or 6, whereas C-2 arylation is preferred by nonpolar solvents and phosphine 3. This represents the first general method for C-5
selective arylation of oxazole and should see broad applicability in the synthesis of biologically active molecules. Additionally, potential
mechanisms for these two competing arylation processes are proposed on the basis of mechanistic observations.
Oxazoles appear in numerous natural products and thera-
peutics, and these compounds exhibit biological activity over
a wide range of targets.1 For example, an oxazole ring is
contained in the antibiotic ostreogrycin A (1)2 as well as in
the immunosuppressant merimepodib (VX-497)3 (Figure 1).
Despite the existence of numerous approaches to the
construction and functionalization of oxazoles, there remains
† Department of Process Chemistry.
Figure 1
oxazoles.
. Examples of biologically active compounds containing
‡ Department of Medicinal Chemistry.
(1) For review articles of oxazole-containing natural products, see: (a)
Yeh, V. S. C. Tetrahedron 2004, 60, 11995–12042. (b) Jin, Z. Nat. Prod.
Rep. 2003, 20, 584–605.
(2) Also known as virginiamycin M1: (a) Delpierre, G. R.; Eastwood,
F. W.; Cream, G. E.; Kingston, D. G. I.; Sarin, P. S.; Todd, L.; Williams,
D. H. Tetrahedron Lett. 1966, 36, 9–372. (b) Delpierre, G. R.; Eastwood,
F. W.; Gream, G. E.; Kingston, D. G. I.; Sarin, P. S.; Todd, L.; Williams,
D. H. J. Chem. Soc. C 1966, 165, 3–1669.
a need for more efficient methods that grant chemists facile
access to this important class of compounds.
Recently, we disclosed methods for regioselective Suzuki
couplings of dihaloazoles, including 2,4-diiodooxazole, with a
variety of aryl and heteroaryl boronic acids.4 To address certain
(3) Looker, A. R.; Littler, B. J.; Blythe, T. A.; Snoonian, J. R.; Ansell,
G. K.; Jones, A. D.; Nyce, P.; Chen, M.; Neubert, B. J. Org. Process Res.
DeV. 2008, 12, 666–673.
10.1021/ol1011778  2010 American Chemical Society
Published on Web 07/19/2010

oxazole as well as at C-2 with aryl electrophiles, including
inexpensive and readily available aryl chlorides.
Table 1. Optimization of C-5- and C-2-Selective Direct
Arylation of Oxazole with Bromobenzene
We initiated this study by examining the direct arylation of
oxazole using highly active ligands known to undergo facile
oxidative addition into aryl bromides, chlorides, and triflates
and looked for C-5 or C-2 selectivity. Arylation of oxazole with
bromobenzene was mediated by Pd and Cy3P to give moderate
C-2 selectivity (Table 2, entry 1). Changing the ligand to t-Bu3P
entrya
ligand
solvent
C5/C2 ratio
1
2
Cy3P-HBF4
t-Bu3P-HBF4
toluene
toluene
toluene
toluene
toluene
DEE
DMA
DMA
DMA
DMA
1:4.5
1:42
1:100
1:100
1:23
1:1.6
21:1
47:1
7.3:1
15:1
3
3
3
4
4
4
5
5
6
4b
5
Table 2. Direct Arylation of Oxazole with Phenyl Electrophiles
at C-5 and C-2
6
7
8
9b,c
10b
PhX
major
a Conditions: 1.5 equiv of oxazole, 0.1 M PhBr, 10% Pd(OAc)2, 20%
ligand, 3 equiv of K2CO3, 0.4 equiv of pivalic acid, 110 °C, 16 h. DEE )
1,2-diethoxyethane. b Used PhCl instead of PhBr. c <5% conversion.
entry (X )) methoda product C5:C2 mono/bis yieldb (%)
1
2
3
4
5
6
7
8c
Cl
Br
OTf
I
Cl
Br
OTf
I
A
B
B
B
C
C
C
C
2-C5
2-C5
2-C5
2-C5
2-C2
2-C2
2-C2
2-C2
15:1
58:1
31:1
29:1
1:100
1:100
1:27
1:14
12:1
30:1
49:1
43:1
25:1
18:1
24:1
17:1
56
83
64
73
77
75
76
82
limitations to this method, we investigated direct arylation of
oxazole, which provides several advantages. Oxazole is readily
available and inexpensive compared to dihaloazoles. Addition-
ally, direct arylation allows the installation of a vast array of
aryl and heteroaryl halides, which are far more numerous than
boronic acid coupling partners. Most importantly, direct ar-
ylation would be expected to couple at both C-2 and C-5, which
would be complementary to the 2,4-substitution pattern acces-
sible with our Suzuki chemistry.
a Method A: 2.0 equiv of oxazole, 0.2 M ArX in DMA, 5% Pd(OAc)2,
10% ligand 6, 3 equiv of K2CO3, 0.4 equiv of pivalic acid, 110 °C, 16 h.
Method B: 2.0 equiv of oxazole, 0.2 M ArX in DMA, 5% Pd(OAc)2, 10%
ligand 5, 3 equiv of K2CO3, 0.4 equiv of pivalic acid, 110 °C, 16 h. Method
C: 2.0 equiv of oxazole, 0.2 M ArX in toluene, 5% Pd(OAc)2, 10% ligand
3, 3 equiv of K2CO3, 0.4 equiv of pivalic acid, 110 °C, 16 h. b Isolated
yield of single isomer. c Used 3 equiv of Cs2CO3 instead of K2CO3.
Among all of the direct arylations of oxazoles, there is
but one report of selectivity at C-5 over C-2, and this was
limited to a chloropyrazine substrate.5,6 While C-5 and even
C-4 can be arylated if C-2 is blocked, there is currently no
method to couple selectively at either of these positions if
C-2 is not substituted, due to preferential C-2 arylation.7-10
This communication describes our efforts to develop general
complementary conditions for direct arylation at C-5 of
still led to preferential arylation at C-2, but with higher
selectivity (entry 2). Using RuPhos (3) (Figure 2) as the ligand,
(4) Strotman, N. A.; Chobanian, H. R.; He, J.; Guo, Y.; Dormer, P. G.;
Jones, C. M.; Steves, J. E. J. Org. Chem. 2010, 75, 1733–1739.
(5) For recent examples of benzoxazole arylation with Cu, Ni, and Pd,
respectively, see: (a) Do, H.-Q.; Daugulis, O. J. Am. Chem. Soc. 2007, 129,
12404–12405. (b) Hachiya, H.; Hirano, K.; Satoh, T.; Miura, M. Org. Lett.
2009, 11, 1737–1740. (c) Ackerman, L.; Barfu¨sser, S.; Pospech, J. Org.
Lett. 2010, 12, 724–726.
(6) Aoyagi, Y.; Inoue, A.; Koizumi, I.; Hashimoto, K. T.; Gohma, K.;
Komatsu, J.; Sekine, K.; Miyafuji, A.; Kunoh, J.; Honma, R.; Akita, Y.;
Ohta, A. Heterocycles 1992, 33, 257–272.
Figure 2. Ligands and precatalysts employed in this study.
(7) C-5 arylation when C-2 is substituted: (a) Ohnmacht, S. A.; Mamone,
P.; Culshaw, A. J.; Greaney, M. F. Chem. Commun. 2008, 124, 1–1243.
(b) Pivsa-Art, S.; Satoh, T.; Kawamura, Y.; Miura, M.; Nomura, M. Bull.
Chem. Soc. Jpn. 1998, 71, 467–473.
selectivities of >100:1 could be obtained for C-2 arylation (entry
3). These same conditions could be applied to arylation with
chlorobenzene while still maintaining this high level of C-2
selectivity (entry 4). While the reaction with X-Phos (4) was
C-2 selective in toluene (entry 5), we were pleasantly surprised
to find that the regiochemical outcome switched to unselective
in 1,2-diethoxyethane (DEE) (entry 6) and to highly C-5
selective in DMA (entry 7).11 Subsequent optimization gave
47:1 C-5/C-2 selectivity using 3,4,5,6-tetramethyl-t-Bu-X-Phos
(5) in DMA (entry 8). Disappointingly, 5 gave only moderate
(8) C-4 arylation when both C-2 and C-5 are substituted: Liegault, B.; Lapointe,
D.; Caron, L.; Vlassova, A.; Fagnou, K. J. Org. Chem. 2009, 74, 1826–1834.
(9) There are several examples of selective C-2 arylation of oxazoles even when
C-5 is not substituted, for example: (a) Flegeau, E. F.; Popkin, M. E.; Greaney,
M. F. Org. Lett. 2008, 10, 2717–2720. (b) Verrier, C.; Martin, T.; Hoarau,
C.; Marsais, F. J. Org. Chem. 2008, 73, 7383–7386. (c) Derridj, F.; Djebbar,
S.; Benali-Baitich, O.; Doucet, H. J. Organomet. Chem. 2008, 693, 135–
144. (d) Hoarau, C.; Du Fou de Kerdaniel, A.; Bracq, N.; Grandclaudon,
P.; Couture, A.; Marsais, F. Tetrahedron Lett. 2005, 46, 8573–8577.
(10) For a general review of direct arylation including C-2/C-5 selective
arylation of thiazole and N-alkylimidazoles, see: (a) Alberico, D.; Scott,
M. E.; Lautens, M. Chem. ReV. 2007, 107, 174–238. (b) Bellina, F.; Rossi,
R. Tetrahedron 2009, 65, 10269–10310.
Org. Lett., Vol. 12, No. 16, 2010
3579

When we applied these direct arylation conditions to other
phenyl electrophiles, we were pleased to find that phenyl triflate
and iodobenzene reacted under the same conditions as bro-
mobenzene to give both C-5 (method B (5 in DMA)) and C-2
(method C (3 in toluene)) products with high selectivity (Table
2, entries 3, 4, 7, and 8).13,14 Arylation with chlorobenzene
(method A (6 in DMA) or C) and bromobenzene (method B
or C) could be carried out with high C-5 or C-2 selectivity
and high isolated yields (Table 2, entries 1, 2, 5, and 6).
Table 3. Direct Arylation of Oxazole with Various Aryl and
Heteroaryl Halides
With these three sets of optimized conditions (methods
A-C) in hand, we began to look at direct arylation of
oxazole with a variety of aryl and heteroaryl chlorides and
bromides (Table 3). It was quickly found that method B was
sufficiently reactive with bromobenzene or electron-deficient
aryl bromides but that method A was superior for electron-
rich or sterically hindered aryl bromides.15 C-5 arylation of
oxazole with aryl bromides, chlorides, and several heteroaryl
chlorides was accomplished with high selectivity and in good
yield using methods A or B (entries 1-6).
Method C was effective for the regioselective arylation
of oxazole at C-2 for a variety of aryl and heteroaryl
bromides and chlorides in high selectivities and good yields
(entries 7-12). Although the C-2 selectivity was low for
2-bromo-5-fluoropyridine (and other 2-halopyridines) under
our standard conditions (2.5:1 C-2/C-5), changing the base
to Cs2CO3 led to high C-2 selectivity (entry 13).
Over the course of this study, we made several observa-
tions that suggest possible mechanisms for these competing
arylation processes. Our initial assumption had been that both
C-5 arylation and C-2 arylation occurred through concerted
metalation-deprotonation (CMD) pathways, as proposed by
Fagnou for direct arylation of other heterocycles (Figure 3,
a Method A: 2.0 equiv of oxazole, 0.2 M ArX in DMA, 5% Pd(OAc)2,
10% ligand 6, 3 equiv of K2CO3, 0.4 equiv of pivalic acid, 110 °C, 16 h.
Method B: 2.0 equiv of oxazole, 0.2 M ArX in DMA, 5% Pd(OAc)2, 10%
ligand 5, 3 equiv of K2CO3, 0.4 equiv of pivalic acid, 110 °C, 16 h. Method
C: 2.0 equiv of oxazole, 0.2 M ArX in toluene, 5% Pd(OAc)2, 10% ligand
3, 3 equiv of K2CO3, 0.4 equiv of pivalic acid, 110 °C, 16 h. b Isolated
yield of single isomer. c Required 10% Pd and 20% ligand. d Used 3 equiv
of Cs2CO3 instead of K2CO3.
C-5 selectivity for arylation with chlorobenzene (entry 9),
although high selectivity could again be obtained by employing
CataCXium A (6) (entry 10).12 These trends where more polar
solvents led to greater C-5 arylation and where the use of aryl
chlorides instead of bromides led to greater C-2 arylation were
observed across all of the ligands we investigated.
Figure 3. Proposed catalytic cycles for direct oxazole arylation
through CMD and deprotonation pathways.
structure A).10,16 However, the dependence of the selectivity
on the nature of the base (Cs2CO3 vs K2CO3) and aryl halide
(PhCl vs PhBr) seemed inconsistent with solely a CMD
pathway in which neither of these factors should be involved
in the selectivity-determining step.
(11) For related solvent effects with a 3-carboalkoxy furan and thiophene,
see: Glover, B.; Harvey, K. A.; Liu, B.; Sharp, M. J.; Tymoschenko, M. F.
Org. Lett. 2003, 5, 301–304.
(12) Except for CataCXium A, every other ligand that we examined
showed much lower C-5/C-2 selectivity when arylated with chlorobenzene
rather than with bromobenzene.
3580
Org. Lett., Vol. 12, No. 16, 2010

In several head-to-head experiments in DMA with a
variety of catalysts, we demonstrated that the C-5/C-2 ratio
increased with higher concentrations of PivOH. This sup-
ported the hypothesis that C-5 arylation was proceeding
through a CMD pathway, which is thought to require the
intermediacy of a Pd-carboxylate complex (Figure 3,
structure A), and that C-2 arylation was occurring through a
different pathway. The involvement of a CMD pathway for
arylation at C-5 is consistent with experimental10 and
computational17 results for thiazole and imidazoles. However,
even with no PivOH, C-5 arylation, while considerably
slower, was still dominant in some cases. We realized that
our results were confounded by the use of Pd(OAc)2, which
could also facilitate a CMD pathway. Moreover, even the
carbonate bases that we were employing could theoretically
coordinate to Pd and mediate a CMD pathway.
following catalytic cycle to account for the differences in
regioselectivity that we observed with different bases,20 with
different halide leaving groups, and with or without PivOH
(Figure 3). Both catalytic cycles begin with oxidative addition
of the aryl halide to a Pd(0) complex.21 In the presence of
PivOH and weak bases, the ArPdX species would form an
ArPd(OPiv) intermediate which could undergo CMD with
oxazole at C-5. Reductive elimination of the ArPd(5-oxazoyl)
intermediate would lead to the C-5 arylated product and
regenerate Pd(0). Alternatively, when strong base is em-
ployed, a potassium oxazole species (or ring-opened tau-
tomer) may directly attack ArPdX, forming an ArPd(2-
oxazole) intermediate.22 After reductive elimination, this
would regenerate Pd(0) and form the C-2-arylated product.
We believe that the solvent effects that we observe may be
related to greater stabilization of a polar CMD transtition state by
DMA than toluene, leading to greater C-5 selectivity. Additionally,
the rate at which poorly nucleophilic KOPiv displaces halide from
the ArPdX intermediate should be substantially slower in less polar
solvents, allowing for a more C-2 selective reaction. Moreover,
the higher C-2 selectivity (lower C-5 selectivity) that we observe
with PhCl vs PhBr could also be explained by slower displacement
of the chloride from ArPdCl by KOPiv, allowing the deprotonation
pathway to dominate.
In conclusion, we have developed general conditions for
highly selective direct arylation of oxazole at both C-5 and C-2.
These methods are applicable to a variety of aryl and heteroaryl
electrophiles, including bromides, chlorides, and triflates. Ad-
ditionally, this represents the first general method for C-5
selective direct arylation of oxazole, which is a common
structural motif in natural products and pharmaceuticals.
Moreover, these methods should be highly valuable to phar-
maceutical chemists wishing to install one aryl group at either
C2 or C5 and then build a library of analogues functionalized
at the other position. Finally, we have proposed a catalytic cycle
that rationalizes the complementary selectivities that we obtained
under different reaction conditions.
Using X-Phos and RuPhos precatalysts (7 and 8, Figure 2),
we examined direct arylation of oxazole with bromobenzene
using a variety of bases, with or without PivOH. We were
surprised to find that with strong bases like KOH or KO-t-Bu,
reactions gave >100:1 C-2/C-5 selectivity regardless of the
catalyst, solvent, or presence of PivOH. The 100:1 ratios
favoring C-2 obtained with KOH or KO-t-Bu and 7 in DMA
are in stark contrast to the 15:1 ratio favoring C-5 obtained under
otherwise identical conditions with K2CO3 as the base (Scheme
1). The use of these stronger bases resulted in a 1500 fold
increase in the relative rate of C-2 arylation vs C-5 arylation.
Scheme 1
.
Effect of Base on Selectivity for Direct Arylation of
Oxazole
Acknowledgment. We thank Dr. Gavin Jones (MIT) for
helpful computational discussions.
Given the reasonable acidity of the C-2 proton of ox-
azole,17 it seems plausible that reaction at this center may
occur through formal deprotonation.18 Either the potassium
oxazole species or the dominant ring-opened enolate tau-
tomer19 could react with an ArPdX species. We propose the
Supporting Information Available: Experimental details,
characterization data, and NMR spectra of all coupling
products. This material is available free of charge via the
Internet at http://pubs.acs.org.
OL1011778
(13) Replacing K2CO3 with Cs2CO3 for arylation with PhI using C-5-
selective method B led to low regioselectivity (1.5:1 C2/C5). On the other
hand, C-2-selective method C gave poor reactivity and selectivity (28%
conversion, 1.8:1 C2/C5) with PhI using K2CO3, but the rate and
regioselectivity were improved by switching to Cs2CO3.
(19) It has been shown through trapping experiments and spectroscopi-
cally that 2-lithiooxazoles and 2-oxazole magnesiates exist predominantly
in the ring-opened form: (a) Hodges, J. C.; Patt, W. C.; Connolly, C. J. J.
Org. Chem. 1991, 56, 449–452. (b) Whitney, S. E.; Rickborn, B. J. Org.
Chem. 1991, 56, 3058–3063. (c) Bayh, O.; Awad, H.; Mongin, F.; Hoarau,
C.; Bischoff, L.; Tre´court, F.; Que´guiner, G.; Marsais, F.; Blanco, F.; Abarca,
B.; Ballesteros, R. J. Org. Chem. 2005, 70, 5190–5196.
(14) Phenyl tosylate was very poorly reactive.
(15) 2-Bromotoluene, 2-bromomesitylene, and 4-bromoanisole all showed
incomplete conversion with method B but full conversion with method A.
(16) Gorelsky, S. I.; Lapointe, D.; Fagnou, K. J. Am. Chem. Soc. 2008,
130, 10848–10849.
(20) Although Cs2CO3 is not an appreciably stronger base than K2CO3,
it has a higher solubility in toluene, making the base concentration much
higher when employing Cs2CO3 instead of K2CO3.
(21) It is worth noting that Pd(0) is probably generated from Pd(OAc)2
through formation of a Pd(oxazole)2 intermediate followed by reductive
elimination. In reactions that are C-5 selective, we have isolated 5,5′-
bis-oxazole.
(22) Formation of an ArPd(2-oxazole) intermediate may also involve
coordination of oxazole to Pd through N or O and subsequent deprotona-
tion.
(17) H/D exchange of the C-2 proton of oxazole in 1.43 M NaOMe in
MeOD proceeded at room temperature with a t1/2 of 17 min: Brown, D. J.;
Ghosh, P. B. J. Chem. Soc. B 1969, 270–276.
(18) The formation of Cu-2-thiazoyl and Cu-2-N-arylimidazoyl
intermediates has been implicated in the C-2-selective direct arylation of
thiazole and imidazoles: (a) Bellina, F.; Cauteruccio, S.; Mannina, L.; Rossi,
R.; Viel, S. Eur. J. Org. Chem. 2006, 693. (b) Mori, A.; Sekiguchi, A.;
Masui, K.; Shimada, T.; Horie, M.; Osakada, K.; Kawamoto, M.; Ikeda, T.
J. Am. Chem. Soc. 2003, 125, 1700.
Org. Lett., Vol. 12, No. 16, 2010
3581