Palladium-Catalyzed Arylation of Trimethylsilyl Enolates of Esters and Imides. High Functional Group Tolerance and Stereoselective Synthesis of α-Aryl Carboxylic Acid Derivatives

Article abstract of DOI:10.1021/ja031544e

A general procedure for the palladium-catalyzed arylation of trimethylsilyl enolates of esters and imides is reported. In the presence of ZnF₂ or Zn(O-t-Bu)₂ as an additive, the trimethylsilyl enolates of esters, including those bearing α-alkoxy derivatives, underwent arylation in high yield with high functional group tolerance. This arylation chemistry was extended to ester derivatives bearing chiral auxiliaries to form new tertiary stereocenters. The arylation of imides bearing the Evans auxiliary proceeded with selectivities up to 90% de. Further, the arylation of the ketal developed by Ley provided α-aryl glycolates with excellent diastereoselectivities (90 to >98% de). This reaction provides a convenient route to the synthesis of enantiopure α-aryl-α-hydroxy esters. Reactions conducted with Zn(O-t-Bu)₂ as an additive occurred at room temperature to give enhanced diastereoselectivities with both chiral reagents. Mechanistic studies showed that the reaction conditions are neutral enough that the observed diastereomeric ratios reflect kinetic selectivities.

Full text of DOI:10.1021/ja031544e

Published on Web 04/03/2004
Palladium-Catalyzed Arylation of Trimethylsilyl Enolates of
Esters and Imides. High Functional Group Tolerance and
Stereoselective Synthesis of r-Aryl Carboxylic Acid
Derivatives
Xiaoxiang Liu and John F. Hartwig*
Contribution from the Department of Chemistry, Yale UniVersity,
New HaVen, Connecticut 06510
Received December 5, 2003; E-mail: john.hartwig@yale.edu
Abstract: A general procedure for the palladium-catalyzed arylation of trimethylsilyl enolates of esters and
imides is reported. In the presence of ZnF2 or Zn(O-t-Bu)2 as an additive, the trimethylsilyl enolates of
esters, including those bearing R-alkoxy derivatives, underwent arylation in high yield with high functional
group tolerance. This arylation chemistry was extended to ester derivatives bearing chiral auxiliaries to
form new tertiary stereocenters. The arylation of imides bearing the Evans auxiliary proceeded with
selectivities up to 90% de. Further, the arylation of the ketal developed by Ley provided R-aryl glycolates
with excellent diastereoselectivities (90 to >98% de). This reaction provides a convenient route to the
synthesis of enantiopure R-aryl-R-hydroxy esters. Reactions conducted with Zn(O-t-Bu)2 as an additive
occurred at room temperature to give enhanced diastereoselectivities with both chiral reagents. Mechanistic
studies showed that the reaction conditions are neutral enough that the observed diastereomeric ratios
reflect kinetic selectivities.
Introduction
Although mild in temperature and efficient in catalyst
loadings,15,18 these reactions are constrained by the stability and
functional group tolerance of alkali metal enolates at room
temperature and above. Because the enolate must be stable
enough to resist decomposition and condensation on the time
scale of the catalytic chemistry, the arylation of esters15,18 has
been limited to enolates such as those of tert-butyl acetate, tert-
butyl propionate, or methyl isobutyrate that are stable at room
temperature.21,22 Moreover, other functional groups contained
in the coupling partners must be stable to the basic medium at
room temperature or above. Palladium-catalyzed reactions of
alkali metal enolates of esters with aryl halides that contain
highly electrophilic functionality such as acetyl and nitro groups
have not been reported. Finally, products from reactions of aryl
halides with alkali metal enolates that possess new tertiary
stereocenters and reagents that contain remote stereocenters that
are sensitive to base are likely to undergo racemization in the
presence of the alkali metal enolates.23,24 Thus, more neutral
reaction conditions would expand the scope and applications
of the palladium-catalyzed arylation of enolates.25,26
The reaction of an electrophile with an enolate is a funda-
mental organic transformation. Though the scope of uncatalyzed
reactions of enolates with alkyl halide electrophiles is broad,
the scope of similar reactions of enolates with aryl halides is
limited. Much effort has been spent to render the arylations of
enolates as general as alkylations. Some progress has been made
with main group arylating reagents such as organobismuth and
aryllead compounds.1-5 However, the arylations of enolates
became more general with the recent development of palladium-
catalyzed procedures6-8 for the R-arylation of ketones,9-12
esters13-18 and amides.19,20
(1) Elliott, G. I.; Konopelski, J. P. Tetrahedron 2001, 57, 5683.
(2) Abramovitch, R. A.; Barton, D. H. R.; Finet, J. P. Tetrahedron 1988, 44,
3039.
(3) Barton, D. H. R.; Blazejewski, J. C.; Charpiot, B.; Finet, J. P.; Motherwell,
W. B.; Papoula, M. T. B.; Stanforth, S. P. J. Chem. Soc., Perkin Trans. 1
1985, 2667.
(4) Rossi, R. A.; Alonso, R. A. J. Org. Chem. 1980, 45, 1239.
(5) Millard, A. A.; Rathke, M. W. J. Am. Chem. Soc. 1977, 99, 4833.
(6) Culkin, D. A.; Hartwig, J. F. Acc. Chem. Res. 2003, 36, 234.
(7) Miura, M.; Nomura, M. Top. Curr. Chem. 2002, 219, 211.
(8) Sole, D.; Vallverdu, L.; Peidro, E.; Bonjoch, J. Chem. Commun. 2001, 1888.
(9) Fox, J. M.; Huang, X.; Chieffi, A.; Buchwald, S. L. J. Am. Chem. Soc.
2000, 122, 1360.
(10) Hamann, B. C.; Hartwig, J. F. J. Am. Chem. Soc. 1997, 119, 12382.
(11) Kawatsura, M.; Hartwig, J. F. J. Am. Chem. Soc. 1999, 121, 1473.
(12) Ehrentraut, A.; Zapf, A.; Beller, M. AdV. Synth. Catal. 2002, 344, 209.
(13) Beare, N. A.; Hartwig, J. F. J. Org. Chem. 2002, 67, 541.
(14) Goossen, L. J. Chem. Commun. 2001, 669.
(15) Jorgensen, M.; Lee, S.; Liu, X.; Wolkowski, J. P.; Hartwig, J. F. J. Am.
Chem. Soc. 2002, 124, 12557.
(16) Lee, S.; Beare, N. A.; Hartwig, J. F. J. Am. Chem. Soc. 2001, 123, 8410.
(17) Lloyd-Jones, G. C. Angew. Chem., Int. Ed. 2002, 41, 953.
(18) Moradi, W. A.; Buchwald, S. L. J. Am. Chem. Soc. 2001, 123, 7996.
(19) Shaughnessy, K. H.; Hamann, B. C.; Hartwig, J. F. J. Org. Chem. 1998,
63, 6546.
We have recently investigated reactions of trimethylsilyl
enolates and zinc enolates. Reactions of Reformatsky re-
agents25,26 with aryl halides occurred with broader scope than
(20) Lee, S.; Hartwig, J. F. J. Org. Chem. 2001, 66, 3402.
(21) Rathke, M. W.; Lindert, A. J. Am. Chem. Soc. 1971, 93, 2318.
(22) Rathke, M. W.; Sullivan, D. F. J. Am. Chem. Soc. 1973, 95, 3050.
(23) Spielvogel, D. J.; Buchwald, S. L. J. Am. Chem. Soc. 2002, 124, 3500.
(24) Hamada, T.; Chieffi, A.; Ahman, J.; Buchwald, S. L. J. Am. Chem. Soc.
2002, 124, 1261.
(25) Cossy, J.; De Filippis, A.; Pardo, D. G. Org. Lett. 2003, 5, 3037.
(26) Hama, T.; Liu, X.; Culkin, D. A.; Hartwig, J. F. J. Am. Chem. Soc. 2003,
125, 11176.
9
5182
J. AM. CHEM. SOC. 2004, 126, 5182-5191
10.1021/ja031544e CCC: $27.50 © 2004 American Chemical Society

Synthesis of R-Aryl Carboxylic Acid Derivatives
A R T I C L E S
Table 1. Effect of Additives on the Arylation of Trimethylsilyl
Ketene Acetal
reactions of alkali metal enolates, but certain functionalities
remained intolerant of the conditions, and tert-butyl esters were
required to observe clean reactivity from acetate and propionate
Reformatsky reagents. If the silyl enolates reacted on their own
under palladium-catalyzed conditions or with an additive that
did not simply generate high concentrations of alkali metal
enolates, then reactions of these reagents would be expected to
show high functional group tolerance. Moreover, trimethylsilyl
enolates can be synthesized directly from esters or amides and
are relatively stable to air and moisture.
entry
additives
solvent
dioxane
DMF
dioxane
dioxane
DME
DMF/DME (1:1)
product:ArBra
yieldb
1
2
3
4
5
6
7
8
9
none
none
0:100
24:76
0:100
0:100
0:100
44:56
75:25
49:51
97:3
0%c,d
nd
ZnCl2
0%c,d
0%c,d
0%
CuCl
ZnCl2 (0.5 equiv)
ZnCl2 (0.5 equiv)
ZnCl2 (0.25 equiv) DMF/DME (1:1)
ZnBr2 (0.25 equiv) DMF/DME (1:1)
ZnF2 (0.5 equiv)
ZnF2 (0.5 equiv)
ZnF2 (0.25 equiv)
LiCl (1.0 equiv)
Previously, palladium-catalyzed coupling of these reagents
was reported in the presence of stoichiometric amounts of tin
fluoride additive to generate the tin enolates27,28 or in the
presence of suprastoichiometric amounts of copper fluoride29
under conditions that involved an excess of enolate with respect
to aryl halide. Both of these protocols have drawbacks. Tin
halides are toxic, and only the acetate enolate was shown to be
reactive in the presence of the copper fluoride additive. One
specific case of the reaction of an activated aryl triflate with a
silyl ketene acetal in the presence of ZnCl2 additive has been
reported.30
nd
66%
38%
95%
∼100%
93%
nd
DMF/DME (1:1)
DMF
DMF
10
11
12
100:0
100:0
83:17
DMF
a Determined by GC. b GC yield; nd ) not determined. c p-tButyl-PhBr
was used. d T ) 100 °C.
The scope of the arylation of silyl ketene acetals with aryl
bromides in the presence of 0.5 equiv of ZnF2 as a cocatalyst
and the combination of P(dba)2 and P(t-Bu)3 as catalyst is
illustrated in Table 2. Reaction of 1.5 equiv of the trimethylsilyl
ketene acetal with PhBr in the presence of 0.5 equiv of ZnF2
and 1 mol % palladium catalyst in DMF at 80 °C generated the
coupled product in 91% isolated yield (entry 1). Most important,
these reactions occurred in similarly high yields with aryl
bromides that contain a variety of functional groups not tolerated
by the coupling of alkali metal enolates.
For example, 4-bromonitrobenzene reacted with the silyl
ketene acetal of methyl isobutyrate and methyl propionate to
give the arylated products in 98% (entry 2) and 76% yield (entry
3). Ester and (entries 4, 5) nitrile (entries 6, 7) functionalities
were also compatible with the conditions for coupling of silicon
enolates in the presence of zinc additive. Different substitution
patterns on the aryl bromide were tolerated; both 2-cyano- (entry
6) and 3-cyanobromobenzene (entry 7) reacted with the silyl
ketene acetal of methyl propionate to give the arylated product
in 75 and 67% yields, respectively. Bromobenzophenone reacted
with the enolate of methyl isobutyrate in 95% yield (entry 8).
In contrast to reactions of alkali metal and zinc enolates,
4-bromoacetophenone reacted with the silyl enolates of methyl
isobutyrate and tert-butyl propionate (entries 9, 10) to provide
the coupled products in high yield. These results suggest that
the activated main group enolate undergoes transmetalation with
the palladium intermediate faster than potential nucleophilic
attack on the ketone and enolization of the acidic proton.
Electron-rich aryl bromides (entry 11) also underwent the
coupling reaction in high yield.
Considering the many reports of Lewis acid-promoted aldol
reactions of trimethylsilyl ketene acetals,31 we evaluated reac-
tions of the combination silicon enolates and Lewis acidic
reagents to observe the coupling of this type of enolate. We
recently reported preliminary results on the arylation of trim-
ethylsilyl ketene acetals.26 Herein, we report the discovery of
two zinc additives that allow palladium complexes to catalyze
the arylation of trimethylsilyl enolates and the use of these
additives with enolates bearing chiral auxiliaries to form
optically active R-aryl carboxylic acid derivatives.
Results and Discussion
Reactions of trimethylsilyl ketene acetal with phenyl bromide
in the presence and absence of various additives are summarized
in Table 1. Reactions of trimethylsilyl ketene acetal and PhBr
in dioxane in the presence of Pd(dba)2 and P(t-Bu)3 as catalysts
at 100 °C for 12 h without additive led to no product. The same
reaction in DMF at 80 °C occurred to partial completion.
Reactions of these substrates with mild Lewis acids such as
ZnCl2 and CuCl in dioxane or DME in the presence of Pd(dba)2
and P(t-Bu)3 as catalyst at the same temperatures and times gave
no product, as determined by GC. However, reactions with 0.5
equiv of ZnF2 as an additive at 80 °C in DMF solvent in the
presence of the combination of Pd(dba)2 and P(t-Bu)3 as catalyst
led to reaction of the trimethylsilyl ketene acetal with PhBr.
Reaction with 0.25 equiv of ZnF2 also occurred to completion
and formed the coupled product in 93% yield as determined by
GC. This result suggests that ZnF2 serves as a cocatalyst rather
than a stoichiometric reagent. Interestingly, 1.0 equiv of LiCl
also promoted reaction with the same catalyst, but only about
80% of the PhBr was converted to product after 12 h at 80 °C
in this case.
The scope of reactions of chloroarenes with the silicon
enolates was narrower than that of reactions of bromoarenes,
and higher catalyst loadings were required. As shown in Table
2, activated chloroarenes such as p-chloro nitrobenzene (entry
12) and p-chloro methylbenzoate (entry 13) coupled with the
trimethylsilyl ketene acetal to provide the desired product in
excellent yields in the presence of 5 mol % catalyst.
R-Arylation of Imides Bearing the Evans Auxiliary.
Encouraged by the success of the arylation of the silyl enolates
of simple esters, we sought to extend this methodology to
reactions of enolates that would exploit further the low basicity
(27) Kosugi, M.; Hagiwara, I.; Sumiya, T.; Migita, T. Bull. Chem. Soc. 1984,
57, 242.
(28) Kuwajima, I.; Urabe, H. J. Am. Chem. Soc. 1982, 104, 6831.
(29) Agnelli, F.; Sulikowski, G. A. Tetrahedron Lett. 1998, 39, 8807.
(30) Koch, K.; Melvin, L. S.; Reiter, L. A.; Biggers, M. S.; Showell, H. J.;
Griffiths, R. J.; Pettipher, E. R.; Cheng, J. B.; Milici, A. J.; Breselow, R.;
Conklyn, M. J.; Smith, M. A.; Hackman, B. C.; Doherty, N. S.; Salter, E.;
Farrell, C. A.; Shulte, G. J. Med. Chem. 1994, 37, 3197.
(31) Nelson, S. G. Tetrahedron: Asymmetry 1998, 9, 357.
9
J. AM. CHEM. SOC. VOL. 126, NO. 16, 2004 5183

A R T I C L E S
Liu and Hartwig
Table 2. Coupling of Silylketene Acetals with Aryl Halides
a Isolated yields of reactions on 1.0 mmol scale. Values are an average of two runs. b Performed with 5% catalyst.
of the reaction medium. Carboxylic acid derivatives bearing
chiral auxiliaries are often used to prepare optically active
carbonyl derivatives.32-35 R-Arylation and subsequent cleavage
of the auxiliary would provide an enantioselective synthesis of
R-aryl carbonyl compounds.
Such a reaction sequence initiated with alkali metal enolates
would be limited in two ways. First, many of the alkali metal
enolates of ester derivatives bearing common auxiliaries are
unstable at the temperatures required for catalysis. Second, the
arylated product is more acidic than the reactant, and a base
that is strong enough to generate the enolate would epimerize
the R-stereocenter of the product. Thus, the reactions must be
conducted under conditions that do not generate alkali metal
enolate and that are neutral enough to prevent epimerization of
the R-stereocenter of the product. The conditions for the
R-arylation of silicon enolates with zinc additives seemed likely
to fit these criteria.
Among many chiral auxiliaries available for controlling the
stereochemistry of ester derivatives,32-35 the Evans auxiliary32
was chosen for initial study. The products that would result from
this arylation process are capable of being readily transformed
into carboxylic esters, ketones, aldehydes, and alcohols by
known methods.32,36-38 Results from arylation of enolates
bearing the Evans auxiliary are presented in Table 3. Heating
of the trimethylsilyl enolate of the Evans auxiliary with PhBr
in DMF at 80 °C in the presence of ZnF2 and the combination
of 5% Pd(dba)2 and 10% P(tBu)3 led to formation of the
diastereomeric coupled products in an 88:12 ratio. The major
(32) Evans, D. A.; Ennis, M. D.; Mathre, D. J. J. Am. Chem. Soc. 1982, 104,
1737.
(33) Schoellkopf, U. Tetrahedron 1983, 39, 2085.
(34) Seebach, D.; Sting, A. R.; Hoffmann, M. Angew. Chem., Int. Ed. 1997,
35, 2708.
(35) Job, A.; Janeck, C. F.; Bettray, W.; Peters, R.; Enders, D. Tetrahedron
2002, 58, 2253.
9
5184 J. AM. CHEM. SOC. VOL. 126, NO. 16, 2004

Synthesis of R-Aryl Carboxylic Acid Derivatives
A R T I C L E S
Table 3. Diastereoselective Reactions of Silyl Ketenimides Bearing the Evans Auxiliary with Aryl Bromides
a The sterochemistry of the diastereomeric products in entry 1 was established by comparison of the NMR spectra to those of literature values for 3 and
epi-3. The relative sterochemistry of the other products were established by NMR data that were similar to those of 3 and epi-3. b Isolated yield of reactions
on 0.5 to 1.0 mmol scales; average of two runs. The catalyst loading was not optimized. c Determined by integration of crude 1H NMR spectra. d Commercial
zinc t-butoxide contained roughly 50 wt % water, as determined by NMR spectroscopy. The quantity of Zn(OtBu)2 added was calculated to be 0.25 equiv,
when considering that the commercial material is 50% Zn(OtBu)2 by weight. e Combined yield of two isomers. f Yield determined by NMR spectroscopy.
diastereomer was isolated in 67% yield (entry 1). Reaction of
the more sterically demanding auxiliary derived from tert-
leucinol, instead of valinol, formed the diastereomeric products
in a 92:8 ratio (entry 3).
65% isolated yield. The diastereomeric ratio (77:23) was lower
than that obtained with more electron-rich aryl bromides. The
more hindered imide bearing a tert-butylacetyl group (entry 9)
also underwent reaction with phenyl bromide to provide the
arylated product with a diastereomeric ratio of 89:11, but a lower
35% yield of the major diastereomer was formed, as determined
by NMR spectroscopy.
Variation of the zinc additive led to faster reaction rates and
improved diastereoselectivities over those obtained with ZnF2
as an additive. Reaction of PhBr in the presence of Zn(O-t-
Bu)2 with the silyl enolate of the auxiliary derived from valinol
occurred at room temperature to generate a ratio of the two
diastereomers in the crude reaction mixture of 91:9 (entry 2)
and the major diastereomer in 70% isolated yield. The reaction
rate may be faster with this zinc additive because it is more
The scope of the aryl halide that underwent reaction with
this silyl ketimine acetal was broad. As shown in Table 3,
3-bromoacetophenone (entry 5) coupled with the Evans auxiliary
to give the arylated product in 75% yield with a diastereomeric
ratio of 84:16. The ortho-substituted o-bromotoluene (entry 6),
reacted with the silyl enolate to afford the major diastereomer
in 78% yield with a diastereomeric ratio of 92:8. The electron-
neutral, para-substituted aryl bromide p-t-butylbromobenzene
(entry 7) afforded the major isomer in 57% yield with a
diastereomeric ratio of 83:17. The electron-poor 4-cyanophe-
nylbromide (entry 8) reacted to give the major diastereomer in
9
J. AM. CHEM. SOC. VOL. 126, NO. 16, 2004 5185

A R T I C L E S
Liu and Hartwig
Scheme 1
reaction of palladium with the (E)-silicon enolate instead of the
(Z)-lithium enolate. Alternatively, the weaker ability of silicon
to chelate the carbonyl groups could make the conformation of
the silicon enolate that reacts with palladium different from the
conformation of the lithium enolates that react with alkyl halides.
This conformation may involve a roughly 180° rotation about
the C-N bond to place the carbonyl dipoles in opposite
directions, and this opposite orientation of the auxiliary would
lead to formation of the opposite diastereomer.
r-Arylation of Acyclic r-Alkoxy Esters. Hydroxy or alkoxy
esters are important biological compounds,41 and much effort
has been spent to develop methods to prepare them.42-47 R-Aryl,
R-hydroxy esters are an important subset of these compounds,
and their synthesis remains challenging.42 We envisioned the
R-arylation of a hydroxy ester derivative as a rapid route to
these compounds. The success in the diastereoselective arylation
of the Evans auxiliary suggested that arylation of an R-hydroxy
ester derivative bearing an appropriate chiral auxiliary could
create an enantioselective synthesis of R-aryl, R-hydroxy esters.
soluble than ZnF2. Reactions of aryl bromides with the imide
containing the auxiliary from valinol in the presence of
Zn(OtBu)2 at the higher temperature of 80 °C generated slightly
lower diastereomeric excess than did reactions in the presence
of ZnF2 at the same temperature. This result indicates that the
higher diastereoselectivity results from lower temperatures and
not from a direct effect of the zinc additive on the step that
controls diastereoselectivity. The faster rates also led to full
conversion of PhBr at room temperature upon reaction of the
silyl enolate bearing the oxazolidinones from tert-leucinol (entry
4). In this case, the diastereoselectivity was 95:5, and the major
diastereomer was isolated in 61% yield.
To determine whether the ratio of diastereomers obtained
from these reactions was the kinetic ratio, we conducted the
arylation of the silicon enolate 2 in the presence of a diaste-
reomerically pure R-aryl imide 1 and in the presence of pure
epi-1. As shown in Scheme 1, the reaction formed the same
88:12 ratio of diastereomers as the reaction conducted without
added R-aryl imide. Most important, the R-aryl imides 1 and
epi-1 showed no discernible decay of their diastereomeric excess
during the reaction. This result clearly shows that the coupling
process creates kinetic ratios and that the conditions for the
arylation of silicon enolates are neutral enough to prevent the
epimerization of the base-sensitive stereocenters.
The origin of diastereoselection of the catalytic arylation of
the silicon enolates was different from the origin of diaste-
reoselection of the uncatalyzed alkylation of lithium enolates.
If the palladium intermediate were to approach the silicon
enolate in the same fashion that an alkyl halide approaches the
lithium enolate,39,40 then one would expect to obtain epi-3 as
the major diastereomer after reductive elimination from pal-
ladium with retention of configuration.32 However, diastereomer
3 was the major isomer obtained from the arylation reaction.
The formation of the opposite diastereomer to that expected
from the geometry of the lithium enolates could result from
Because the enolates of R-alkoxy esters decompose at room
temperature,48,49 attempts to conduct the R-arylation of the alkali
metal enolates of R-alkoxy esters were unsuccessful. As shown
in Scheme 2, reactions of PhBr with the lithium or sodium
enolate of benzyloxyl tert-butyl acetate 4 with Pd(dba)2 as a
catalyst and P(t-Bu)3 or a hindered heterocyclic carbene as the
ligand provided the arylated product 5 in less than 20% yield,
as determined by GC. The reaction between PhBr and the
lithium or sodium enolate of the less hindered methoxy ethyl
acetate under similar conditions gave ester 7 in even lower yield
(<5%).
Silicon enolates of these esters are more stable thermally.
This thermal stability, in combination with a method to induce
coupling chemistry from silicon enolates, allowed the develop-
ment of methods to generate R-aryl, R-alkoxy esters in good
yields. Results of the coupling of two representative R-alkoxy
esters, benzyloxyl tert-butyl acetate 4 and methoxy ethyl acetate
6, with various aryl halides is summarized in Table 4. Reactions
of the silyl enolates 4 and 6 occurred with aryl halides with
diverse electronic properties (entries 1-3). Aryl halides, such
as 4-bromonitrobenzene (entry 2), that are sensitive to base, as
well as aryl halides, such as methyl-4-bromobenzoate (entry
6), that are sensitive to nucleophilic reagents, underwent reaction
with the R-alkoxy esters in good yields. Moreover, these
reactions occurred with enolates that form products with new
quaternary centers (entry 4).
r-Arylation Cyclic Esters Bearing the Ley Auxiliary:
Diastereoselective Synthesis of r-Aryl, r-Hydroxy Esters.
With conditions to conduct the arylation of achiral alkoxy esters,
we sought to develop arylations of R-alkoxy esters bearing a
chiral auxiliary that would allow for a stereoselective synthesis
of R-aryl, R-hydroxy esters. Among the auxiliaries that could
(41) Coppola, G. M.; Schuster, H. F. alpha-Hydroxy Acids in EnantioselectiVe
Synthesis; VCH: Weinheim, 1997.
(42) Davis, F. A.; Chen, B. C. Chem. ReV. 1992, 92, 919.
(36) Amigoni, S.; Le Floc’h, Y. Tetrahedron: Asymmetry 1997, 8, 2827.
(37) Vanderlei, J. M. d. L.; Coelho, F.; Almeida, W. P. Tetrahedron: Asymmetry
1997, 8, 2781.
(43) Yu, H.; Ballard, C. E.; Boyle, P. D.; Wang, B. Tetrahedron 2002, 58, 7663.
(44) Diez, E.; Dixon, D. J.; Ley, S. V. Angew. Chem., Int. Ed. 2001, 40, 2906.
(45) Zhu, Y.; Shu, L.; Tu, Y.; Shi, Y. J. Org. Chem. 2001, 66, 1818.
(46) Zhu, Y.; Tu, Y.; Yu, H.; Shi, Y. Tetrahedron Lett. 1998, 39, 7819.
(47) Adam, W.; Fell, R. T.; Stegmann, V. R.; Saha-Moeller, C. R. J. Am. Chem.
Soc. 1998, 120, 708.
(38) Takacs, J. M.; Jaber, M. R.; Vellekoop, A. S. J. Org. Chem. 1998, 63,
2742.
(39) Bull, S. D.; Davies, S. G.; Key, M.-S.; Nicholson, R. L.; Savory, E. D.
Chem. Commun. 2000, 1721.
(48) Yu, H.; Ballard, C. E.; Wang, B. Tetrahedron Lett. 2001, 42, 1835.
(49) Kunz, H.; Mohr, J. J. J. Chem. Soc., Chem. Commun. 1988, 1315.
(40) Hintermann, T.; Seebach, D. HelV. Chim. Acta 1998, 81, 2093.
9
5186 J. AM. CHEM. SOC. VOL. 126, NO. 16, 2004

Synthesis of R-Aryl Carboxylic Acid Derivatives
A R T I C L E S
Table 4. Arylation of Silicon Enolates of Alkoxy Esters
Scheme 2
and conversion of the aryl bromide. For example, reactions
conducted with 0.5 equiv of Zn(OtBu)2 instead of 0.25 equiv
afforded 95% instead of 80% yield (entry 5 vs entry 4). This
effect of the amount of zinc reagent was observed with reactions
of several bromoarenes. Coupling between 3-nitrobromobenzene
and the silyl enolate of 8 proceeded in 78% yield with 0.25
equiv of Zn(OtBu)2 (entry 6) but occurred in a higher 89% yield
in the presence of 0.5 equiv of Zn(OtBu)2 (entry 7).
a Isolated yield of reactions on a 1.0 mmol scale, average of two runs.
As expected from results of other coupling reactions of silyl
enolates, the reaction of the silicon enolate of dioxanone 8
occurred with a variety of electronically and sterically distinct
aryl halides, including those containing electrophilic functional
groups. Reactions of the electron-rich 4-bromoanisole (entry 9)
produced the coupled product in 58% yield with a diastereomeric
ratio greater than 50:1. Ortho-substituted aryl bromides also
coupled with the silicon enolate of 8 in good yield and with
high stereoselectivity (entries 11, 12). Moreover, aryl bromides
containing potentially reactive electron-withdrawing groups such
as acetyl, nitro, and ester groups, as well as an aryl halide
containing a chloro substituent (entries 4, 6, 9, 11), reacted with
this enolate with high stereoselectivity. For example, methyl
4-bromobenzoate coupled with a combination of the silyl enolate
of 8 and Zn(OtBu)2 additive at room temperature to form a single
diastereomer by NMR spectroscopy in 80% yield (entry 4). The
equally high selectivity of the reactions of the enolate of 8 with
activated and unactivated aryl halides contrasts with the lower
diastereoselectivities obtained from reactions of the silyl enolate
of the Evans auxiliary with activated aryl halides.
be suitable for these studies44,50 we chose to investigate the
cyclic ketal 8 designed by Ley et al.44 This auxiliary has been
shown previously to undergo highly diastereoselective alkylation
and aldol condensation reactions.44,51-54 We sought, in part, to
compare the diastereoselection of the arylation process with this
rigid cyclic structure to that we obtained with the Evans
auxiliary.
As shown in Scheme 2, reactions of the dioxanone 8 with
aryl bromides in the presence of alkali metal base and the
combination of 5 mol % Pd(dba)2 and 10 mol % P(t-Bu)3 as a
catalyst occurred in yields that were as low as those from
reactions of acyclic R-alkoxy enolates with aryl bromides.
However, the combination of the silicon enolate of 8 and ZnF2
reacted at 80 °C with phenyl bromide in the presence of Pd-
(dba)2 and P(t-Bu)3 as a catalyst to provide the coupled product
in good yield and with high diastereoselectivity. Moreover,
analogous reactions of aryl bromides with the combination of
the silicon enolate of 8 and Zn(OtBu)2 occurred at room
temperature.
Results on the arylation of the silyl enolate of Ley’s
dioxanone 8 are summarized in Table 5. Phenyl bromide coupled
with this enolate in the presence of 0.5 equiv of ZnF2 (entry 1)
to provide a single diastereomer, as determined by NMR
spectroscopy, in 67% isolated yield. Reactions conducted at
room temperature with Zn(OtBu)2 (entry 2) as an additive also
occurred with high diastereoselectivity, and the reaction yields
were slightly higher (73%) when this zinc additive was used.
The amount of zinc additive influenced the yield of product
Zinc t-butoxide from our commercial source contained
roughly 50% water by weight. Therefore, the molar amount of
zinc is roughly half the amount that would be added if the
reagent contained only zinc tert-butoxide. Thus, a reaction
conducted with 0.5 equiv of zinc tert-butoxide that was dried
by refluxing in benzene with a Dean-Stark apparatus gave the
coupled product in a yield determined by NMR spectroscopy
that was comparable to that obtained with twice the amount of
the commercial “zinc tert-butoxide”. Because this reaction
tolerated the presence of water, a reaction conducted by
weighing all the reagents except the ligand in air, followed by
addition of the ligand and degassed DMF by syringe occurred
with rates and yields that were similar to those of reactions
conducted with more rigorously dry conditions.
(50) Swindell, C. S.; Tao, M. J. Org. Chem. 1993, 58, 5889.
(51) Michel, P.; Ley, S. V. Synthesis 2003, 1598.
(52) Diez, E.; Dixon, D. J.; Ley, S. V.; Polara, A.; Rodriguez, F. Synlett 2003,
1186.
(53) Michel, P.; Ley, S. V. Angew. Chem., Int. Ed. 2002, 41, 3898.
(54) Dixon, D. J.; Guarna, A.; Ley, S. V.; Polara, A.; Rodriguez, F. Synthesis
2002, 1973.
9
J. AM. CHEM. SOC. VOL. 126, NO. 16, 2004 5187

A R T I C L E S
Liu and Hartwig
Table 5. Arylation of the Silicon Enolate of Ley’s Dioxanone
a Isolated yields of reactions on 0.5 mmol to 1.0 mmol scale; average of two runs. b Ratios determined by integration of 1H NMR spectra of crude
reactions. c Diastereomeric ratio was stated to be “>50:1” when only one isomer was observed by NMR spectroscopy of the crude reaction mixture.
d Commercial zinc t-butoxide contains roughly 50 wt % water, as determined by NMR spectroscopy. The number of equivalents of zinc alkoxide was
calculated assuming the commercial zinc t-butoxide is 50 wt %.
The diastereomeric ratios of the arylated products were
determined by H NMR spectroscopy. The most stable diaste-
Scheme 3
1
reomer has been established to contain the alkyl or aryl
substituents in the equatorial position and methoxy groups in
the axial position.44 The major isomer from the catalytic reaction
was shown to be this thermodynamically more stable isomer.
Spectral data on the minor isomeric product for comparison to
that of the major product were obtained from samples of the
two diastereomers generated by protonation of the lithium
enolate of the arylated product at low temperature53 with
t-butanol and acetic acid.
To ensure this arylation of Ley’s reagent would lead to an
enantioselective synthesis of R-hydroxy esters, the enantiopure
isomer was deprotected. As shown in Scheme 3, both diaster-
eomers were individually treated with TMSCl in methanol,44
and the optical purity of the resulting methyl mandelic ester
was determined to be greater than 99.5%. Thus, deprotection
of the arylated product preserves the stereochemistry created
in the coupling step as expected.44
thermodynamic factors. First, we monitored by 1H NMR
spectroscopy the reaction of PhBr with the silyl enolate of the
dioxanone 8 catalyzed by P(t-Bu)3 and Pd(dba)2 with zinc
t-butoxide as an additive. If the ratio was determined by the
relative thermodynamic stability of the two diastereomers, we
might observe the accumulation of the minor isomer during the
reaction, followed by epimerization of this material to the major
isomer. Only the major isomer was observed during the reaction.
Second, we tested for epimerization of the components of a
diastereomeric mixture of the arylated products. A 1:1.1 ratio
Two experiments were conducted to determine whether the
observed diastereomeric ratio was controlled by kinetic or
9
5188 J. AM. CHEM. SOC. VOL. 126, NO. 16, 2004

Synthesis of R-Aryl Carboxylic Acid Derivatives
A R T I C L E S
Scheme 4
of the two epimers of the R-aryl dioxanone 12 was added to
the reaction of the enolate of 8 with phenyl bromide in the
presence of Zn(OtBu)2 as an additive and Pd(dba)2 and
P(t-Bu)3 as catalysts, as shown in Scheme 4. The 1:1.1 mixture
of diastereomers 12 did not change measurably during this
reaction. A 1:1.4 mixture of diastereomers 12 changed to a 1:3.4
mixture that favored the major diastereomer of the catalytic
reaction when the same process was conducted with ZnF2 as
an additive. These data suggest that some epimerization could
occur with the less reactive system containing zinc fluoride,
but this small amount of epimerization cannot account for the
high diastereoselectivities observed from the coupling of the
silyl enolate of 8 with aryl halides.
The diastereoselectivity of the catalytic arylation indicates
that the arylpalladium halide approaches the silicon enolate from
the same direction that an alkyl halide approaches the lithium
enolate. The palladium would approach the enolate more
favorably from the less hindered equatorial site to avoid the
unfavorable 1,3-diaxial interaction with the methoxy groups.
Our selectivity compares well with the 60:1 ratio of diasate-
reomers reported by Ley and co-workers from reaction of allylic
bromides with the lithium enolate of 8,44 despite the temperature
of the catalytic arylation being much higher than that of the
uncatalyzed alkylation.
Possible Role of Zinc Additives. The mechanism by which
the zinc additives promote the coupling of the silicon enolates
is ambiguous, but we have obtained several lines of data that
argue against a complete transmetalation of silicon to zinc. One
might propose that reaction of the silicon enolates with ZnF2
would generate a zinc enolate because the F-Si bond is strong
and fluoride is commonly used to cleave silicon-oxygen bonds.
Moreover, we have recently shown that Reformatsky reagents
react with aryl halides to form R-aryl esters and amides.25,26
However, the scope of the reactions with the silicon enolate
and zinc halide is different from that of the Reformatsky
reagents. For example, 4-bromoacetophenone reacts in high yield
with the silyl enolate of tert-butyl propionate and zinc additive,
but this substrate suffers attack at the carbonyl group by the
Reformatsky reagents.
Reaction of the trimethylsilylenolate of methyl isobutyrate with
ZnF2 in DMF at 80 °C for 12 h led to decomposition of about
50% of the silyl ketene acetal to the ester without accumulation
of a new enolate, whereas the catalytic reaction occurred to
completion within this time. In addition, monitoring of the
reaction of Ley’s reagent with phenyl bromide in the presence
of the palladium catalyst and Zn(OtBu)2 as an additive showed
no evidence for the accumulation of a zinc enolate. Only the
decay of the silicon enolate and the accumulation of coupled
product and a small amount of ester from hydrolysis was
observed. Therefore, if a zinc enolate were generated as an
intermediate during this process, its formation must be endo-
ergic, and the resulting enolate must react quickly with the
palladium.
Furthermore, the use of only 0.25 equiv of ZnF2 provided
the coupled product from reaction of methyl trimethylsilyl
ketene acetal with PhBr in 93% yield as determined by GC.
Clearly, stoichiometric amounts of fluoride are not required for
the reaction to occur to completion. In addition, zinc bromide
or a mixed fluorozinc bromide would have to be an effective
promoter if fluoride-catalyzed cleavage of the silicon enolate
occurred because the transmetalation with an arylpalladium
bromide complex would generate a zinc bromide. ZnBr2 does
serve as a mild promoter of the arylation of the silyl ketene
acetals, but reaction of methyl trimethylsilyl ketene acetal with
phenyl bromide in the presence of the palladium catalyst and
ZnBr2 gave only 38% yield of the coupled product after 12 h
at 80 °C. Use of an admixture of ZnBr2 and ZnF2 as a promoter
also did not lead to full conversion of phenyl bromide under
the same reaction conditions.
Thus, three possible roles of the zinc additive are consistent
with our data. The zinc additive might function as a Lewis acid
to reduce the strength of the silicon-oxygen bond, and the
aprotic polar solvent DMF could function as a Lewis base to
further activate the silyl enolate. Second, the zinc additive might
abstract halide from palladium to generate a more positively
charged palladium center that would undergo faster transmeta-
lation. Third, the silylketene acetal could be transformed into a
zinc enolate in a thermodynamically unfavorable but kinetically
accessible step. The resulting enolate could then react rapidly
Moreover, no direct evidence for transmetalation was ob-
served upon reaction of the silicon enolates with zinc halides.
9
J. AM. CHEM. SOC. VOL. 126, NO. 16, 2004 5189

A R T I C L E S
Liu and Hartwig
auxiliary44 was used to determine diastereoselectivities and was prepared
by literature procedures.
with palladium, and the fluoride from silicon could be returned
to the resulting zinc bromide to regenerate the zinc fluoride.
Representive Procedure for the Arylation of Silyl Ketene Acetals.
Methyl phenylisobutyrate (Table 2, entry 1).15 To a screw-capped
vial containing PtBu3 (40 µL of a 0.500 M solution in toluene, 0.020
mmol), Pd(dba)2 (5.8 mg, 0.010 mmol), ZnF2 (52.0 mg, 0.500 mmol),
and phenyl bromide (157.0 mg, 1.000 mmol) was added tert-butyl
trimethylsilyl methyl ketene acetal (301.0 mg, 1.49 mmol), followed
by DMF (4.0 mL). The vial was sealed with a cap containing a PTFE
septum and removed from the drybox. The heterogeneous reaction
mixture was stirred at 80 °C for 12 h. The crude reaction was then
allowed to cool to room temperature and diluted with Et2O (50 mL).
The resulting solution was washed with H2O (5 × 20 mL). The organic
phase was dried over Na2SO4, filtered, and concentrated at reduced
pressure. The residue was then purified by chromatography on silica
gel, eluting with a 1-2% gradient of EtOAc in hexane to provide the
Conclusions
The arylation of the preformed silicon enolates in the presence
of zinc additives in DMF provides a process that is comple-
mentary to previously published arylation of alkali metal
enolates and Reformatsky reagents. This procedure tolerates
various functional groups, such as nitro, cyano, ester, and ketone
substituents, on the aryl bromide. Furthermore, the thermal
stability of the silicon enolates makes them suitable for the first
formation of R-aryl-R-alkoxy esters by coupling of the corre-
sponding enolates with aryl halides. On the other hand, these
reactions are slower than those of the alkali metal enolates and
Reformatsky reagents and require polar solvents. Thus, each
method displays advantages, and the combination of methods
provides a catalytic route to a variety of R-aryl ester derivatives.
1
title compound (162 mg) in 91% yield. H NMR (CDCl3): δ 7.27-
7.38 (5H, m), 3.69 (3H, s), 1.62 (6H, s). 13C NMR (CDCl3): δ 177.72,
145.08, 128.85, 127.15, 126.02, 52.64, 46.94, 26.98.
Arylation of Trimethylsilyl Ketenimines of the Evans Auxiliary.
(4S,2′S)-4-Isopropyl-3-(2′-phenyl-propanoyl)oxazolidin-2-one (Table
3, Entries 1, 2).58 To a screw-capped vial containing PtBu3 (0.500 M
solution in toluene, 200 µL, 0.010 mmol), Pd(dba)2 (29 mg, 0.050
mmol), ZnF2 (52 mg, 0.50 mmol), and phenyl bromide (157 mg, 1.00
mmol) were added the trimethylsilyl enolate 2 of the Evans imide (370.0
mg, 1.44 mmol), followed by DMF (10 mL). The vial was sealed with
a cap containing a PTFE septum and removed from the drybox. The
heterogeneous reaction mixture was stirred at 80 °C for 12 h. The crude
reaction was then allowed to cool to room temperature and diluted with
Et2O. The resulting solution was washed with H2O. The organic phase
was dried over Na2SO4, filtered, and concentrated at reduced pressure.
Purification of the crude material by flash chromatography, eluting with
The development of conditions for the arylation of silicon
ester enolates has allowed the preparation of products with
stereogenic centers R to the carbonyl group. We have developed
two processes with substrates bearing established chiral auxil-
iaries, and the diastereomeric ratios have been shown to reflect
kinetic selectivities. Significant diastereoselectivities have been
observed for the arylation of imides bearing the Evans auxiliary.
These reactions provide an enantioselective synthesis of tertiary
R-aryl acids. In addition, excellent diastereoselectivities were
achieved from the arylation of Ley’s auxiliary, and this reaction
provides an enantioselective synthesis of R-arylated R-hydroxy
esters. Successful diastereoselective arylation with these aux-
iliaries suggests that this methodology may be conducted with
silicon enolates of other carbonyl derivatives bearing chiral
auxiliaries and can be applied to the synthesis of structurally
complex products, even at the latter stages of a synthesis. A
related reaction was employed previously in the synthesis of a
leukotriene B4 antagonist.30 Use of the catalyst and additive of
the current work would increase the efficiency of the coupling
step of this synthesis and would simplify the experimental
procedures.
1
2% EtOAc in hexanes, gave 67% yield of the R-aryl imide. H NMR
(CDCl3): δ 7.27 (2H, d, J ) 7.4 Hz), 7.22 (2H, t, J ) 7.3 Hz), 7.15
(1H, t, J ) 7.3 Hz), 5.06 (1H, q, J ) 7.0 Hz), 4.25-4.28 (1H, m),
4.00-4.07 (2H, m), 2.33-2.37 (1H, m), 1.43 (3H, d, J ) 7.0 Hz),
0.84 (3H, d, J ) 6.6 Hz), 0.82 (3H, d, J ) 6.1 Hz). 13C NMR
(CDCl3): δ 175.04, 153.99, 140.72, 128.98, 128.56, 127.59, 63.50,
59.43, 43.45, 28.94, 20.09, 18.42, 15.11.
Experimental Procedure for the Evaluation of Epimerization
during the r-Arylation. To a screw-capped vial containing PtBu3
(0.500 M solution in toluene, 20 µL, 0.010 mmol), Pd(dba)2 (2.9 mg,
0.0050 mmol), ZnF2 (5.2 mg, 0.050 mmol), and phenyl bromide (15.7
mg, 0.100 mmol) were added the trimethylsilyl enolate 2 of the Evans
imide (37.0 mg, 0.144 mmol) and compound 1 (>90% de),58 followed
by DMF (1.0 mL). The vial was sealed with a cap containing a PTFE
septum and removed from the drybox. The heterogeneous reaction
mixture was stirred at 80 °C for 12 h. The crude reaction was then
allowed to cool to room temperature and diluted with Et2O (5.0 mL).
The resulting solution was washed with H2O (5 × 2.0 mL). The organic
phase was dried over Na2SO4, filtered, and concentrated at reduced
Experimental Section
General Methods. Reactions were conducted using standard drybox
1
techniques. H and 13C NMR spectra were recorded in CDCl3 on a
400 MHz spectrometer with tetramethylsilane or residual protiated
solvent used as a reference and coupling constants reported in hertz
(Hz). Chromatographic purifications were performed by flash chro-
matography using silica gel (200-400 mesh) or using an automated
chromatography system. Yields for final products in all tables refer to
isolated yields and are the average of two runs. Products that had been
reported previously were isolated in greater than 95% purity, as
1
pressure. The residue was then analyzed by H NMR spectroscopy.
Only one set of signals was observed for compound 1 and its
1
(2′)epimer.58 From integration of the H NMR signals, the diastereo-
1
determined by H NMR and capillary gas chromatography (GC). All
meric ratio of the isopropyl derivative formed from the R-arylation
was 88:12. The same reaction on a 1.00 mmol scale in the absence of
the added R-aryl imide provided the major diastereomer in 67% yield
after purification by flash chromatography eluting with 2% EtOAc in
hexanes.
13C NMR spectra were proton decoupled. GC analyses were obtained
with a DB-1301 narrow bore column for high-temperature ramp
applications (max 120 °C/min). Methyl trimethylsilyl ketene acetal,
ZnF2, and Zn(OtBu)2 were purchased from commercial suppliers. The
silyl ketene acetal of tert-butyl propionate55 and the trimethylsilyl ether
derivative of the Evans auxiliary56 were prepared according to literature
procedures. The racemic version57 of the available nonracemic Ley
Representative Procedures for the Arylation of the Ley Auxiliary.
5,6-Dimethoxy-5,6-dimethyl-3-phenyl-[1,4]dioxan-2-one (Table 5,
Entries 1, 2).57 5,6-Dimethoxy-5,6-dimethyl-[1,4]dioxan-2-one 844,57
(1.00 g, 5.26 mmol) was dissolved in anhydrous THF (20 mL), and
(55) Hoffman, R.; Kim, H. O. J. Org. Chem. 1988, 53, 3855.
(56) Fuentes, L. M.; Shinkai, I.; Salzmann, T. N. J. Am. Chem. Soc. 1986, 108,
4675.
(58) Fukuzawa, S.; Chino, Y.; Yokoyama, T. Tetrahedron: Asymmetry 2002,
13, 1645.
(57) Ley, S. V.; Michel, P. Synlett 2001, 11, 1793.
9
5190 J. AM. CHEM. SOC. VOL. 126, NO. 16, 2004

Synthesis of R-Aryl Carboxylic Acid Derivatives
A R T I C L E S
the mixture was cooled to -78 °C under N2. LDA (3.2 mL of a 2.0 M
solution in THF/heptane, 6.4 mmol) was then added dropwise. After
addition, the mixture was stirred for another 5 min before addition of
TMSCl (0.83 mL, 6.6 mmol). The resulting solution was then allowed
to warm to room-temperature overnight. The THF was then removed
under reduced pressure, and hexanes (20 mL) were added to the residue.
The resulting suspension was filtered through Celite, and the hexanes
were removed in vacuo to afford 1.40 g of the desired trimethylsilyl
ketene acetal (judged to be 90-95% pure from NMR). 1H NMR
(CDCl3): δ 5.55 (1H, s), 3.41 (3H, s), 3.26 (3H, s), 1.49 (3H, s), 1.42
(3H, s), 0.25 (9H, s). This material was directly employed in the
arylation step. In a drybox, to a screw-capped vial were added PhBr
(78.5 mg, 0.500 mmol), ZnF2 (26.0 mg, 0.0250 mmol), Pd(dba)2 (15.0
mg, 0.025 mmol), and P(tBu)3 (100 µL, 0.50 M solution in toluene,
0.050 mmol). The trimethylsilyl ketene acetal prepared above (170 mg
calculated according to the purity, 0.650 mmol) was added to the
mixture followed by 4.0 mL of DMF. The vial was then sealed and
heated at 80 °C for 12 h. The solution was then allowed to cool to
room temperature and partitioned between ethyl ether (50 mL) and
water (10 mL). The ether layer was washed with water and brine and
dried over Na2SO4. After the ether was removed, the residue was
subjected to automated flash chromatography using 8% EtOAc/hexanes
to provide the desired product (88.4 mg) in 66% yield. 1H NMR
(CDCl3): δ 7.59 (2H, d, J ) 6.8 Hz), 7.32-7.41 (3H, m), 5.21 (1H,
s), 3.47 (3H, s), 3.45 (3H, s), 1.58 (3H, s), 1.52 (3H, s). 13C NMR
(CDCl3): δ 168.85, 136.44, 129.01, 128.91, 128.22, 105.65, 98.98,
74.07, 50.46, 49.78, 18.40, 17.47.
Procedure for Conducting the Reactions without a Glovebox.
To a round-bottom flask were added Pd(dba)2 (2.9 mg, 0.0050 mmol),
Zn(OtBu)2 (21.0 mg of commercial material, which is about 50% H2O
by weight), the trimethylsilyl enolate of the Ley auxiliary (34.0 mg of
calculated according to the purity, 0.130 mmol) and 3-bromonitroben-
zene (20.2 mg, 0.100 mmol). The flask was purged with N2 for 5 min
before addition of P(tBu)3 (20 µL of 0.5 M solution in toluene, 0.010
mmol). DMF (1.0 mL) was then added, and the resulting mixture was
stirred for 12 h. Trimethoxybenzene as an internal standard was then
added to the solution, and the mixture was partitioned between Et2O
(5.0 mL) and H2O (2.0 mL). The ether layer was washed with water,
and the ether was evaporated in vacuo. The yield determined by NMR
spectroscopy was determined with an internal standard to be nearly
quantitative, which is similar to the yield observed when the reactions
are assembled in a drybox.
Acknowledgment. The authors thank the NIH (NIGMS
GM58108) for funding, Merck Research Laboratories for an
unrestricted gift, and Johnson-Matthey for palladium complexes.
Supporting Information Available: Detailed procedures and
spectral data for new compounds. This material is available free
of charge via the Internet at http://pubs.acs.org.
JA031544E
9
J. AM. CHEM. SOC. VOL. 126, NO. 16, 2004 5191