C-C bond-forming reductive elimination of ketones, esters, and amides from isolated arylpalladium (II) enolates [2]

Full text of DOI:10.1021/ja015732l

5816
J. Am. Chem. Soc. 2001, 123, 5816-5817
Scheme 1
C-C Bond-Forming Reductive Elimination of
Ketones, Esters, and Amides from Isolated
Arylpalladium(II) Enolates
Darcy A. Culkin and John F. Hartwig*
Department of Chemistry, Yale UniVersity
P.O. Box 208107, New HaVen, Connecticut 06520-8107
ReceiVed February 28, 2001
ReVised Manuscript ReceiVed April 20, 2001
Metal enolates provide the foundation for many synthetic
methods. We and others have recently developed a palladium-
catalyzed process for the direct arylation of ketone enolates.1 The
reaction displays a high degree of regioselectivity and functional
group tolerance. It now encompasses reactions of amides,2
malonates,3 and diketones3b with both bromo- and chloroarene3a,b
electrophiles and has been conducted enantioselectively.4
The key step of this reaction is a C-C bond-forming reductive
elimination from an arylpalladium enolate. On the surface, this
reaction resembles C-C reductive eliminations of palladium
dimethyl or arylpalladium methyl complexes that occur with
nonpolar transition states.5 However, the pKb values of the
enolates6 are more similar to those of amides than of alkyls. Thus,
C-C reductive elimination from enolate complexes may more
closely resemble C-N bond-forming reductive eliminations of
amines,7 which show rates that depend strongly on the electronic
properties of the amide. The effect of carbon-bound ligand
properties on reductive elimination has been evaluated theoreti-
cally,8 but rarely experimentally.9
We report the first examples of arylpalladium enolates that are
sufficiently stable to isolate in pure form, but sufficiently reactive
to undergo reductive elimination of R-aryl carbonyl compounds
in high yields. Using these complexes, we have evaluated the
effect of enolate steric and electronic properties on geometry,
thermodynamic stability, and reductive elimination rates.
Previously, we generated arylpalladium enolate complexes at
-78 °C3a and as mixtures with Pd(0),1a but modification of these
systems to generate stable arylpalladium enolates that underwent
reductive elimination was not straightforward. Experiments
involving a series of arylpalladium enolate complexes with
different ligands showed that 1,2-bis(diphenylphosphino)benzene
(DPPBz) provided the appropriate balance of small bite angle,
backbone stability, and modest electron donation to create a
spectrum of isolable enolate complexes that undergo reductive
elimination upon heating. Diphenylethylphosphine complexes also
showed suitable stability and reactivity.
DPPBz- and PPh2Et-ligated arylpalladium enolate complexes
were prepared as analytically pure solids in 44-81% yield, as
shown in Scheme 1. Several coordination modes are possible for
isolated palladium enolate complexes,10 and both the enolate and
phosphine structure affected the connectivity in 1-16. With the
exception of benzyl phenyl ketone enolate 14, enolate complexes
from ketones with R-methyl or methylene protons and DPPBz
as phosphine were C-bound. The enolate of 2-butanone was bound
solely through the former methyl, instead of ethyl, group.
However, enolate complexes from ketones with R-methine protons
were O-bound, and 14 was a mixture of C- and O-bound isomers.
Complexes with PPh2Et as ligand displayed a trans geometry and
showed significantly greater preference for the O-bound form.
Complex 16 was exclusively O-bound, while 15 was a mixture
of O- and C-bound isomers in a 17:1 ratio.
The enolate connectivity was determined by NMR spectro-
scopic methods. For example, C-bound 1 displayed a single
methylene 1H NMR signal at δ 3.88, which was split by the two
inequivalent phosphine ligands (JH-P ) 10.3, 6.9 Hz). In addition,
the 13C NMR spectrum displayed a doublet of doublets for the
palladium-bound methylene carbon and a triplet at δ 202.7 (JH-P
1
) 4.1 Hz) for the carbonyl. In contrast, the H NMR spectrum
for O-bound 15 displayed two singlets at δ 4.90 and 4.99, and
the 13C NMR spectrum contained a singlet vinyl C-O resonance
at δ 168.9 and a second vinyl resonance at δ 77.9. The
connectivities of C-bound 4 and O-bound 16 were confirmed by
X-ray analysis.
(1) (a) Hamann, B. C.; Hartwig, J. F. J. Am. Chem. Soc. 1997, 119, 12382-
12383. (b) Palucki, M.; Buckwald, S. L. J. Am. Chem. Soc. 1997, 119, 11108-
11109. (c) Satoh, T.; Kawamura, Y.; Miura, M.; Nomura, M. Angew. Chem.,
Int. Ed. Engl. 1997, 36, 1740-1742.
Thus, the C-bound isomer is favored electronically in these
systems if the enolate is located trans to a phosphine, but the
(2) Shaughnessy, K. H.; Hamann, B. C.; Hartwig, J. F. J. Org. Chem. 1998,
63, 6546-6553. Honda, T.; Namiki, H.; Satoh, F. Org. Lett. 2001, 631-633.
(3) (a) Kawatsura, M.; Hartwig, J. F. J. Am. Chem. Soc. 1999, 121, 1473-
1478. (b)Fox, J. M.; Huang, X.; Chieffi, A.; Buchwald, S. L. J. Am. Chem.
Soc. 2000, 122, 1360-1370. (c) Djakovitch, L.; Kohler, K. J. Organomet.
Chem. 2000, 606, 101-107.
(10) For examples of C-bound enolates, see: Albeniz, A. C.; Catalina, N.
M.; Espinet, P.; Redon, R. Organometallics 1999, 18, 5571-5576. Vicente,
J.; Abad, J. A.; Chicote, M.-T.; Abrisqueta, M.-D.; Lorca, J.-A.; Arellano,
M. C. R. d. Organometallics 1998, 17, 1564-1568. Veya, P.; Floriani, C.
Organometallics 1993, 12, 4899-4907. Suzuki, K.; Yamamoto, H. Inorg.
Chim. Acta 1993, 208, 225-229. Byers, P. K.; Canty, A. J.; Skelton, B. W.;
Traill, P. R.; Watson, A. A.; White, A. H. Organometallics 1992, 11, 3085-
3088. Burkhardt, E. R.; Bergman, R. G.; Heathcock, C. H. Organometallics
1990, 9, 30-44. Wanat, R. A.; Collum, D. B. Organometallics 1986, 5, 120-
127. Bertani, R.; Castellani, C. B.; Crociani, B. J. Organomet. Chem 1984,
269, C15-C18. For examples of O-bound enolates, see: Sodeoka, M.;
Tokunoh, R.; Miyazaki, F.; Hagiwara, E.; Shibasaki, M. Synlett 1997, 463-
466. Bouaoud, S.-E.; Braunstein, P.; Grandjean, D.; Matt, D.; Nobel, D. Inorg.
Chem. 1988, 27, 2279-2286. Bouaoud, S.-E.; Braunstein, P.; Grandjean, D.;
Matt, D.; Nobel, D. J. Chem. Soc., Chem. Commun. 1987, 7, 488-490. Ito,
Y.; Nakatsuka, M.; Kise, N.; Saegusa, T. Tetrahedron Lett. 1980, 21, 2873-
2876. For examples of oxallyl enolates, see: Lemke, F. R.; Kubiak, C. P. J.
Organomet. Chem. 1989, 373, 391-400. Yanase, N.; Nakamura, Y.; Kawagu-
chi, S. Inorg. Chem. 1980, 19, 1575-1581. Yanase, N.; Nakamura, Y.;
Kawaguchi, S. Chem. Lett. 1979, 5, 591-594. Yoshimura, N.; Murahashi,
S.-I.; Moritani, I. J. Organmet. Chem. 1973, 52, C58-C60.
(4) Ahman, J.; Wolfe, J. P.; Troutman, M. V.; Palucki, M.; Buchwald, S.
L. J. Am. Chem. Soc. 1998, 120, 1918-1919.
(5) Ozawa, F.; Ito, T.; Nakamura, Y.; Yamamoto, A. Bull. Chem. Soc. Jpn.
1981, 54, 1868-1880. Ozawa, F.; Kurihara, K.; Fujimori, M.; Hidaka, T.;
Toyoshima, T.; Yamamoto, A. Organometallics 1989, 8, 180-188. Gillie,
A.; Stille, J. K. J. Am. Chem. Soc. 1980, 102, 4933-4941. Moravskiy, A.;
Stille, J. K. J. Am. Chem. Soc. 1981, 103, 4182-4186.
(6) Bordwell, F. G.; Fried, H. E. J. Org. Chem. 1981, 46, 4327-4331.
Bordwell, F. G. Acc. Chem. Res. 1988, 21, 456-463. Bordwell, F. G.; John
A. Harrelson, J. Can. J. Chem. 1990, 68, 1714-1718. Bordwell, F. G.;
Cornforth, F. J. J. Org. Chem. 1978, 43, 1763-1768.
(7) Driver, M. S.; Hartwig, J. F. J. Am. Chem. Soc. 1997, 119, 8232-
8245.
(8) Tatsumi, K.; Hoffmann, R.; Yamamoto, A.; Stille, J. K. Bull. Chem.
Soc. Jpn. 1981, 54, 1857-1867. Low, J. J.; Goddard, W. A., III J. Am. Chem.
Soc. 1984, 106, 6928-6937.
(9) Brune, H. A.; Stapp, B.; Schmidtberg, G. J. Organomet. Chem. 1986,
307, 129-137.
10.1021/ja015732l CCC: $20.00 © 2001 American Chemical Society
Published on Web 05/25/2001

Communications to the Editor
J. Am. Chem. Soc., Vol. 123, No. 24, 2001 5817
Table 2. Rates and Yields of Reductive Elimination for
[Pd(DPPBz)(C6H4-2-Me)(R)] at 90 °C
Scheme 2
complex
R
pKaa
kobs × 104 b yieldc
7
8
9
10
11
12
13
18
CH2C(O)CH2CH3
CH2C(O)C(CH3)3
CH2C(O)C6H4-4-OMe 25.7
CH2C(O)C6H4-4-CF3
CH2C(O)OC(CH3)3
CH2C(O)NMe2
CH2C(O)NEt2
Me
24.4
27.7
2.7 ( 0.1
4.5 ( 0.2
3.0 ( 0.2
3.6 ( 0.4
2.6 ( 0.1
3.7 ( 0.3
6.3 ( 0.1
>20
96
96
95
85
93
98
96
96
22.7
ca. 31
34-35
34-35
56
Table 1. Rates and Yields of Reductive Elimination for C- and
O-Bound Enolates of [Pd(DPPBz)(Ar)(R)] at 90 °C
a Values are from ref 6 in DMSO solvent. b Units: s-1; reactions
conducted with added DPPBz. c Units: %; reactions conducted with
added PPh3.
complex
Ar
R
kobs × 104 a yieldb
1
2
3
4
5
6
C6H4-4-t-Bu CH2C(O)C6H4-4-Me 3.0 ( 0.2
C6H4-4-t-Bu CHCH3C(O)C6H5
C6H4-4-t-Bu OC(dCMe2)C6H5
C6H4-2-Me
C6H4-2-Me
C6H4-2-Me
87
57
of diarylamines, anilines, and alkylamines. For a set of arylpal-
ladium amido complexes derived from these amines, reductive
elimination occurred at 65 °C for diarylamides and below room
temperature for alkylamides.7 In contrast, the reductive elimination
of enolates 7-13 showed little difference in rate constant as a
function of enolate electronic properties. These complexes reacted
with rate constants that varied by less than a factor of 3 and
without any correlation with pKa, as shown most dramatically by
comparing kobs values for 10 and 12.
It is well established that reductive elimination can occur from
both three- and four-coordinate palladium complexes.7,8 Although
elimination after partial dissociation of the rigid DPPBz is
unlikely, rate-limiting dechelation is consistent with the small
electronic effect. Thus, we evaluated reductive elimination from
[(DPPE)Pd(o-Tol)(CH2C(O)(C6H4-4-Me)] (17), which is identical
to 4 except for containing the more flexible bis(diphenylphos-
phino)ethane ligand. This complex reacted roughly 2-fold slower
(kobs ) (1.6 ( 0.1) × 10-4 s-1) than did 4, implying that reductive
elimination occurs directly from the four-coordinate DPPBz
complex.
6.5 ( 0.4
82
CH2C(O)C6H4-4-Me 3.7 ( 0.4
99
98
CHCH3C(O)C6H5
5.8 ( 0.4
OC(dCMe2)C6H5
<10
a Units: s-1; reactions conducted with added DPPBz; b Units: %;
reactions conducted with added PPh3.
O-bound form is favored if located trans to an aryl group.
However, a C-bound enolate complex that would possess a
quaternary carbon bound to the metal is less stable in all cases
than its O-bound tautomer for steric reasons.
In addition to geometry, we determined the thermodynamic
stability of the enolate complexes relative to the corresponding
ketone, ester, or amide, as shown in eq 1. In some cases, catalytic
In fact, the rates for C-C bond-forming reductive elimination
from the DPPBz complexes were not independent of electronics
in all cases: arylpalladium methyl (DPPBz)Pd(Me)(o-tolyl) (18)
underwent reductive elimination at 90 °C with a half-life of less
than 5 min (kobs > 20 × 10-4 s-1), and a DPPBz-ligated
arylpalladium complex of a malonate anion did not produce any
arylmalonate product upon thermolysis.
amounts of potassium enolates were added to promote the rate
of equilibration. The relative stabilities were the following: 4, 7,
11, 13 ≈ 14 > 5 > 6 (see eq 1). Therefore, the stability was
controlled by the number of substituents at the R-carbon and not
by the pKa of the carbonyl compound, except when an aryl group
was directly bound to the R position (14 vs 4 and 5).
Thermolysis of 1-13 at 110 °C in the presence of added
PPh3 to bind the released Pd(0) fragment induced reductive
elimination of the corresponding R-aryl carbonyl compound.
Except for 6, they formed product in 57-99% yield by 1H NMR
spectroscopy with an internal standard (Scheme 2). Similar
thermolysis of PPh2Et-complexes 15 and 16 in the presence of
PPh2Et generated R-aryl ketones in 70 and 45% yield.
To determine if coordination mode controlled reaction rates,
we measured kobs values for reductive elimination from both O-
and C-bound DPPBz-ligated enolates (Table 1). Instead of a
simple reductive elimination, coupling could occur by migratory
insertion of the CdC unit of the O-bound enolate into the Pd-
Ar bond, by a pathway related to that for the Heck reaction.11 In
addition, isomerization of a C-bound enolate to its enol tautomer,
followed by favorable C(sp2)-C(sp2) coupling, could occur. In
the first case, O-bound enolates should react faster, and in the
second, isobutyrophenone enolates should not couple. The low
yield observed from O-bound 6 vs the high yields from 4 and 5
disfavors the first path, and the high yield from 3 argues against
a palladaenol intermediate. Simple reductive elimination occurs.
Values of kobs (Table 2) were measured at 90 °C by UV-vis
spectroscopy for reductive elimination from enolate complexes
derived from carbonyl compounds with pKa values ranging from
23 to 34 in DMSO. These values are essentially identical to those
The steric properties of the enolate influenced reaction rates
in a systematic fashion, but the magnitudes of the differences
were small. Reductive eliminations from the more crowded
arylpalladium propiophenone enolates 2 and 5 were faster than
those from the corresponding acetophenone enolates 1 and 4 (6.5
× 10-4 and 5.8 × 10-4 s-1 vs 3.0 × 10-4 and 3.7 × 10-4 s-1).
Likewise, diethylacetamide enolate 13 reacted faster than did
dimethylacetamide 12 (6.3 × 10-4 s-1 vs 3.7 × 10-4 s-1), and
pinacolone enolate 8 reacted faster than did 2-butanone enolate
7 (4.5 × 10-4 s-1 vs 2.7 × 10-4 s-1). These results demonstrate
that arylation of the less hindered carbon of a dialkyl ketone occurs
with high selectivity in the catalytic arylation of ketone enolates
because the less hindered enolate complex is the major tautomer,
not because of a large difference in rates for reductive elimination.
In summary, C-C bond-forming reductive elimination is less
sensitive to electronic perturbations than analogous C-N bond
formation, but is sensitive to large changes in electronics. We
will investigate carbon-bound ligands that possess other functional
groups in the R position to understand further the effect of
electronics on this fundamental reaction.
Acknowledgment. We thank the National Institutes of Health
(R01GM58108-02-04) for support of this work.
Supporting Information Available: Experimental procedures for
compound preparation and kinetic analysis (PDF). This material is
available free of charge via the Internet at http://pubs.acs.org.
(11) Beletskaya, I. P.; Cheprakov, A. V. Chem. ReV. 2000, 100, 3009-
3066.
JA015732L