A new practical ketone synthesis directly from carboxylic acids: First application of coupling reagents in palladium catalysis

Article abstract of DOI:10.1039/b106779f

A new palladium-catalyzed cross-coupling reaction between arylboronic acids and carboxylic acids, activated in situ for the oxidative addition to a tricyclohexylphosphine palladium(0) catalyst by treatment with di(N-succinimidyl) carbonate (DSC) is disclosed, which allows the high-yielding synthesis of various functionalized arylketones under mild conditions.

Full text of DOI:10.1039/b106779f

View Article Online / Journal Homepage / Table of Contents for this issue
A new practical ketone synthesis directly from carboxylic acids: first
application of coupling reagents in palladium catalysis
Lukas J. Gooßen* and Keya Ghosh
Max-Planck-Institut für Kohlenforschung, Kaiser-Wilhelm-Platz 1, D-45470 Mülheim an der Ruhr,
Germany. E-mail: goossen@mpi-muelheim.mpg.de; Fax: +49-208-306-2985; Tel: 49-208-306-2392
Received (in Cambridge, UK) 26th July 2001, Accepted 5th September 2001
First published as an Advance Article on the web 1st October 2001
A new palladium-catalyzed cross-coupling reaction between
arylboronic acids and carboxylic acids, activated in situ for
the oxidative addition to a tricyclohexylphosphine palla-
dium(0) catalyst by treatment with di(N-succinimidyl)
carbonate (DSC) is disclosed, which allows the high-yielding
synthesis of various functionalized arylketones under mild
conditions.
no literature precedent for the use of coupling reagents in
transition-metal catalysis.11
Due to the high stability and easy handling, we chose
phenylboronic acid (1a) and N-benzoyloxysuccinimide (4) as
our model substrates and screened various palladium catalysts
for activity in the desired conversion (Scheme 2). The catalysts
were prepared in situ from Pd(ii) precursors and two equivalents
of different phosphines. Some results are summarized in
Table 1.
The reaction of carboxylic acid derivatives, for example nitriles,
Weinreb-amides, anhydrides, or acid chlorides with carbon
nucleophiles to the corresponding ketones is an important C–C
bond-forming reaction that is commonly used in organic
synthesis.1 Various procedures have been developed to opti-
mize the yield of the ketones and avoid the formation of the
tertiary alcohols.2 Highly reactive acid chlorides can be
alkylated with a number of mild carbon nucleophiles, for
example organotin, -zinc and -copper compounds or boronic
acids.3,4 Furthermore, ketone synthesis from thioesters5 has
been reported. However, only a few protocols have been
disclosed for the highly desirable direct conversion of the
plethora of carboxylic acids into ketones and all of these require
aggressive lithium, magnesium, or aluminum reagents intoler-
ant of most functional groups.6 Recently, we reported a novel
palladium catalyzed synthesis of arylketones directly from
carboxylic acids and boronic acids using pivalic anhydride as an
activating agent.7 This transformation gives good yields for
many substrates, however, the reactivity of the pivalic anhy-
dride against basic groups and the inconvenience of the
separation of the products from the pivalic acid are still
disadvantageous. A milder and more practical ketone synthesis
from carboxylic acids and readily available boronic acids,8
which works under neutral conditions in the presence of many
functional groups would be of great value especially for
applications in combinatorial chemistry.
Whereas no reaction was observed for triarylphosphines,
with electron-rich trialkylphosphines moderate yields were
obtained (Entries 1–6). This may indicate that the oxidative
addition of the N-benzoyloxysuccinimide is the rate-determin-
ing step which should be facilitated by increasing the electron
density on the palladium. Sterically demanding tricyclohex-
ylphosphine gave best results while tri-tert-butylphosphine
complexes were not stable under the reaction conditions and
precipitation of palladium(0) was observed.
THF proved to be the most effective solvent (Entries 6–8).
The stability of the catalytic system was significantly enhanced
when a mild base was added (Entries 9–13). Best results were
obtained with Na2CO3 while the presence of soluble bases
appeared to inhibit the reaction. We also investigated the
Scheme 2 Cross coupling of N-benzoyloxysuccinimide.
Table 1 Effects of the reaction conditions on the yielda
Ligand
Pd-source
Solvent Base
Product (%)b
The utilization of mild coupling reagents such as DCC–
HOBt, CDI (1,1A-carbonyldiimidazole) or DSC for the activa-
tion of the carboxylic acids could be the key to a much-
improved process (Scheme 1). These compounds are known to
cleanly convert carboxylic acids into stable intermediates,
which are just reactive enough to ensure smooth conversion to
the corresponding amides.9 Due to their easy handling, their
reliability and selectivity combined with an excellent tolerance
of functional groups, coupling reagents have become indis-
pensable for most applications in peptide synthesis.
In principle, a catalytic cycle for the palladium-catalyzed
reaction of carboxylic acids with boronic acids consisting of the
oxidative addition of an activated acid derivative producing an
acyl palladium complex, followed by transmetallation of a
boronic acid, and reductive elimination of the ketone appeared
to be feasible.10 However, to the best of our knowledge there is
1
2
3
4
5
6
7
8
9
10
11
12
13c
14d
15
16
17
18
19
20
PPh3
DPPF
P(o-Tol)3 Pd(OAc)2
Pd(OAc)2
Pd(OAc)2
THF
THF
THF
THF
THF
THF
DMF
Toluene
THF
THF
THF
THF
THF
THF
THF
THF
THF
THF
THF
—
—
—
—
—
—
—
—
< 2
< 2
< 2
< 2
19
33
6
12
48
< 2
40
41
64,e 76, 98f
74
48
P(tBu)3
P(nBu)3
PCy3
PCy3
PCy3
PCy3
PCy3
PCy3
PCy3
PCy3
PCy3
PCy3
PCy3
PCy3
PCy3
PCy3
PCy3
Pd(OAc)2
Pd(OAc)2
Pd(OAc)2
Pd(OAc)2
Pd(OAc)2
Pd(OAc)2
Pd(OAc)2
Pd(OAc)2
Pd(OAc)2
Pd(OAc)2
Pd(OAc)2
Pd(OAc)2
dba3Pd2
K2CO3
Cs2CO3
KF
NEt3
Na2CO3
Na2CO3
Na2CO3
Na2CO3
Na2CO3
Na2CO3
Na2CO3
Na2CO3
5,e 36, 45f
70,e 92, 97f
15,e 25, 26f
2,e 80, 98f
67,e 81, 97f
Pd(NO3)2
PdCl2
Pd(acac)2
Pd(F6-acac)2 THF
a Conditions: 1 mmol N-benzoyloxysuccinimide, 1.2 mmol phenylboronic
acid, 3 mol% Pd, 7 mol% ligand, 2 mmol base, 60 °C, 14 h. b Yields
determined by GC. c 1 mmol water added. d 10 mmol water added. e 3 h;
f 36 h.
Scheme 1 Cross-coupling of boronic acids and carboxylic acids.
2084
Chem. Commun., 2001, 2084–2085
This journal is © The Royal Society of Chemistry 2001
DOI: 10.1039/b106779f

View Article Online
influence of the counter ion on the palladium pre-catalyst and
observed faster reactions with weakly coordinating counter-
In summary, the introduction of easy-to-handle coupling
reagents as activating agents for acids in transition-metal
catalysis led to the discovery of a convenient new ketone
synthesis from carboxylic acids and boronic acids. Further
applications of this new concept, for example, in the reduction
of carboxylic acids to aldehydes or the Heck reaction of benzoic
acid derivatives are under current investigation.
2
ions such as NO3 or F6-acac than with stronger coordinating
Cl2 (Entries 16–20).10 Under the reaction conditions, the
boronic acids act as the reducing agents for the palladium so that
biaryls are formed in small quantities. With dba3Pd2 as a pre-
catalyst, lower yields were observed, which may be due to an
insufficient dissociation of the dba from Pd(0). The presence of
small quantities of water has little effect on the reaction
outcome so that it is not necessary to dry the reagents or
solvents, however, the presence of a large excess of water was
found to lower the yields (Entries 13–15).
We thank M. Rössig and L. Winkel for technical assistance
and Professor Dr M. T. Reetz for generous support and constant
encouragement.
We then set out to combine the activation of the carboxylic
acid with di(N-succinimidyl) carbonate and the transformation
to the ketone into a convenient one-pot procedure (Scheme 3).†
Since the activation of the acid was found to proceed smoothly
within a few minutes in THF in the presence of Na2CO3, this
was easily accomplished: in a typical procedure, all reagents
except the boronic acid are dissolved in THF and the mixture is
stirred until the CO2 evolution has ceased. Then, the boronic
acid is added and the reaction is heated until complete
conversion. Under these conditions F6-acac gave best results,
and it is beneficial to add a slight excess of the phosphine to
make up for losses during the first reaction step. The water-
soluble by-products N-hydroxysuccinimide and boric acid are
easily removed by a single water wash.
In order to test the generality of this protocol, we applied it to
a variety of different substrates. Table 2 shows the broad scope
of the new transformation. Both electron-rich and electron-poor
alkyl, aryl and even heteroaryl carboxylic acids work equally
well, many functional groups such as keto, cyano, ester, nitro,
amido and even hydroxy groups are tolerated, and no side
products arising from enolisation of the keto groups were
observed in significant quantities.
Notes and references
† Synthesis of 2-(4A-methoxyphenyl) ethylphenyl ketone (3e): a 100 mL flask
was charged with palladium hexafluoroacetylacetonate (156 mg, 0.30
mmol), tricyclohexylphosphine (252 mg, 0.90 mmol), 3-phenylpropionic
acid (2e) (1.50 g, 10.0 mmol), Na2CO3 (2.08 g, 20.0 mmol), and di(N-
succinimidyl) carbonate (3.33 g, 13.0 mmol). The reaction vessel was
purged with argon and degassed THF (30 mL) was added. The yellow
mixture was stirred at 60 °C for a few minutes until the gas evolution had
ceased. Then, the solution was cooled down to RT, a solution of
4-methoxyphenylboronic acid (1a) (1.82 g, 12.0 mmol) in THF (30 mL)
was added and the purple reaction mixture was stirred at 60 °C overnight.
The reaction slurry was then poured into water (300 mL) and extracted 3
times with 100 mL portions of ethyl acetate. The combined organic layers
were dried over MgSO4, filtered, and the volatiles were removed in vacuo.
The residue was adsorbed on a plug of Al2O3. Nonpolar impurities such as
the phosphine or the biaryl were removed by elution with hexane. The
product (2.19 g, 91%) was then eluted with 20% ethyl acetate in hexane. 1H
NMR (300 MHz, CDCl3, 25 °C, TMS): d = 7.93 (d, 3J (H,H) = 9 Hz, 2H),
7.23–7.19 (m, 5H), 6.92 (d, 3J (H,H) = 9 Hz, 2H), 3.85 (s, 3H), 3.24 (t, 3J
(H,H) = 7 Hz, 2H), 3.06 (t, 3J (H,H) = 7 Hz, 2H) ppm; 13C NMR (75 MHz,
CDCl3, 25 °C, TMS): d = 197.8, 163.5, 141.5, 130.3, 130.0, 128.5, 128.4,
126.1, 113.7, 55.5, 40.1, 30.3 ppm; MS (70 eV): m/z (%): 240 (33) [M+],
135 (100), 121 (2), 107 (6), 92 (8), 77 (13); HRMS: calcd. for C16H16O2
[M+]: 240.115029; found: 240.115132; anal. calcd. for C16H16O2 (240.30):
C, 79.97; H, 6.71; N, 0.0; found: C, 80.12; H, 6.78; N, 0.0. The reactions in
Table 1 and Table 2 were performed on a 1 mmol scale using 50 mg
tetradecane as an internal GC standard. The products were isolated by
column chromatography (Al2O3 hexane–ethyl acetate 10:1) and charac-
terized by means of 1H and 13C NMR as well as by GC-MS and HRMS.
1 (a) J. March, Advanced Organic Chemistry, Wiley, New York, 3rd
Edition, 1985, 433–435, 824–827; (b) R. C. Larock, Comprehensive
Organic Transformations, VCH, New York, 1989, 685–702; (c) D. A.
Shirley, Organic Reactions, Wiley, New York, 1954, Vol. 8, 28–58; (d)
B. T. O’Neill, in Comprehensive Organic Synthesis, ed. B. Trost and I.
Fleming, Pergamon, Oxford, 1991, Vol. 1, 397–458.
Scheme 3 One-pot synthesis of arylketones from carboxylic acids.
Table 2 Pd-catalyzed synthesis of arylketonesa
Comp.
Ar
R
Yield (%)b
2 (a) G. M. Rubottom and C. Kim, J. Org. Chem., 1983, 48, 1550; (b) Y.
Ahn and T. Cohen, Tetrahedron Lett., 1994, 35, 203; (c) T. Fujisawa, S.
Iida, H. Uehara and T. Sato, Chem. Lett., 1983, 1267.
3 (a) R. K. Dieter, Tetrahedron, 1999, 55, 4177; (b) M. P. Sibi, Org. Prep.
Proced. Int., 1993, 25, 15; (c) V. Farina, V. Krishnamurthy and W.
Scott, in Organic Reactions, Wiley, New York, 1997, Vol 50, 1–652.
4 (a) M. Haddach and J. R. McCarthy, Tetrahedron Lett., 1999, 40, 3109;
(b) N. A. Bumagin and D. N. Korolev, Tetrahedron Lett., 1999, 40,
3057.
3a
3b
3c
3d
3e
3f
3g
3h
3i
Phenyl
Phenyl
Phenyl
90
95
90
90
91
56
88
55
51
42
42
2-Phenylethyl
2-Phenylethyl
2-Phenylethyl
2-Phenylethyl
2-Phenylethyl
2-Phenylethyl
2-Phenylethyl
2-Methoxycarbonylethyl
m-Acetoxyphenyl
4-Pyridyl
p-Methoxycarbonylphenyl 53
3-Thienyl
3-Furyl
p-Acetylphenyl
Cyclohexyl
p-Nitrophenyl
p-Methoxyphenyl
p-Cyanophenyl
p-Trifluoromethylphenyl
p-Acetamidophenyl
m-Cyanophenyl
HO-C11H22
o-Tolyl
1-Naphthyl
p-Methoxyphenyl
p-Acetylphenyl
3-Thienyl
2-Furyl
Phenyl
Phenyl
Phenyl
Phenyl
Phenyl
Phenyl
Phenyl
Phenyl
Phenyl
Phenyl
Phenyl
Phenyl
Phenyl
5 (a) L. Liebeskind and J. Srogl, J. Am. Chem. Soc., 2000, 122, 11260; (b)
P. Zurer, Chem. Eng. News, 2000, 78, 26.
3k
3l
6 (a) M. J. Jorgenson, Organic Reactions, Wiley, New York, 1970, Vol.
18, 1–97; (b) C. R. Iwanow, Hebd. Seances Acad. Sci., 1928, 186, 442;
(c) A. Meisters and T. Mole, Aust. J. Chem., 1974, 27, 1665.
7 L. J. Gooßen and K. Ghosh, Angew. Chem., 2001, 113, 3566.
8 (a) T. Ishiyama, M. Murata and N. Miyaura, J. Org. Chem., 1995, 60,
7508; (b) M. Murata, T. Oyama, S. Watanabe and Y. Masuda, J. Org.
Chem., 1997, 62, 6458; (c) M. Murata, T. Oyama, S. Watanabe and Y.
Masuda, J. Org. Chem., 2000, 65, 164.
9 M. Bodanszky and A. Bodanszky, The Practice of Peptide Synthesis, ed.
K. Hafner, J.-M. Lehn, C. W. Rees, P.v.R. Schleyer, B. M. Trost, R.
Zahradnik, Springer, Berlin, 1984.
10 C. Amatore and A. Jutand, Acc. Chem. Res., 2000, 33, 314.
11 Reactions of acylmidazolides with Grignard reagents have however
been reported, some of them being enhanced by addition of Cu(i): (a)
H. A. Staab, Angew. Chem., Int. Ed. Engl., 1962, 1, 351; (b) B. F.
Bonini, M. Comes-Franchini, M. Fochi, G. Mozzanti, A. Ricci and G.
Varchi, Synlett, 1998, 9, 1013.
3m
3n
3o
3p
3q
3r
3s
49
48
88
81
37
68
95
89
86
90
78
3t
3u
3v
3w
3x
Phenyl
Phenyl
a Conditions: 1 mmol carboxylic acid, 1.2 mmol boronic acid, 1.3 mmol
DSC, 3 mol% Pd(F6-acac)2, 9 mol% PCy3, 2 mmol Na2CO3, 60 °C, 20 h.
b Isolated yields.
Chem. Commun., 2001, 2084–2085
2085