Synthesis of ketones from α-oxocarboxylates and aryl bromides by Cu/Pd-catalyzed decarboxylative cross-coupling

Article abstract of DOI:10.1002/anie.200705127

(Chemical Equation Presented) The power of two metals: A Pd/Cu catalyst system mediates the in situ formation of acyl nucleophiles by decarboxylation of readily accessible and stable salts of α-oxocarboxylic acids and their cross-coupling with aryl or heteroaryl bromides to give ketones. The reaction may be usedin the presence of many functional groups and provides good yields.

Full text of DOI:10.1002/anie.200705127

Angewandte
Chemie
DOI: 10.1002/anie.200705127
Cross-Coupling
Synthesis of Ketones from a-Oxocarboxylates and Aryl Bromides by
Cu/Pd-Catalyzed Decarboxylative Cross-Coupling**
Lukas J. Gooßen,* Felix Rudolphi, Christoph Oppel, and Nuria Rodríguez
Dedicated to Dr. Nikolaus Müller
Aryl ketones are important structural elements in biologically
active compounds and functional materials.[1] Besides Frie-
del–Crafts acylations,[2] which yield these products mostly as
isomeric mixtures, the reaction of activated carboxylic acid
Scheme 1. Synthesis of ketones from a-oxocarboxylic acids.
derivatives with organometallic reagents,[3] e.g. of Weinreb
amides with Grignard compounds, is most often used for their
preparation.[4] Transition-metal catalysts increase the effi-
ciency of such cross-couplings such that even carbon nucle-
ophiles of low reactivity, e.g. organozinc compounds and
boronic acids, can be converted.[5–7] This results in substantial
improvement of functional group tolerance. Reaction var-
iants in which carboxylic acids are activated in situ, such as the
palladium-catalyzed synthesis of aryl ketones directly from
arylboronic acids and carboxylic acids in the presence of
anhydrides[8] or coupling reagents,[9] are particularly conven-
ient.
The reverse approach involving the coupling of acyl anion
equivalents with carbon electrophiles is used mainly in the
synthesis of alkyl ketones.[10] The required umpolung of the
carbonyl function is achieved by the conversion of aldehydes
into e.g. cyanohydrins, acetals, dithianes, or hydrazones.[11] In
contrast, there are only a few examples of the catalytic
arylation of acyl anion equivalents, including the coupling of
aryl bromides with N-tert-butylhydrazones described by
Hartwig et al.[12] Disadvantageous in all reactions of this
type are the additional derivatization and hydrolysis steps, as
well as the use of strong bases. Arylation of aldehydes by C–H
activation offers an atom-economical alternative but has
hitherto been possible only with a limited spectrum of
expensive aryl iodides.[13]
With our synthesis of biaryls from benzoate salts and aryl
halides, we demonstrated that decarboxylative cross-cou-
plings can constitute valuable alternatives to the correspond-
ing reactions with organometallic compounds.[14] In subse-
quent mechanistic investigations of the decarboxylation of
benzoic acids[15] we observed that a catalyst system consisting
of copper(I) oxide and 1,10-phenanthroline also mediated the
decarboxylation of 2-oxophenylacetic acid to benzalde-
hyde.[16] We conjectured that an acyl anion equivalent was
generated at the copper center and protonated to give the
aldehyde. Therefore, we decided to carry out this decarbox-
ylation in the absence of protons under basic conditions and
combine it with a palladium-mediated cross-coupling with
aryl halides to overall afford a decarboxylative ketone
synthesis (Scheme 1). The specific advantage of this strategy
is that the acyl nucleophiles are prepared in situ, without
protecting groups and in the absence of strong bases, from
readily accessible and stable salts of a-oxocarboxylic acids.[17]
The hypothetical mechanism of the planned transforma-
tion is illustrated in Scheme 2. The potassium a-oxocarbox-
ylate 1 initially reacts with the copper(I) complex a by ligand
metathesis to form the copper carboxylate b. Decarboxylation
of b affords the acyl copper species c, which then transfers its
aryl residue to the palladium(II) species e, itself derived from
the oxidative addition of the aryl halide 2 to the palladium(0)
catalyst d. The copper(I) halide complex a is released in the
transmetalation step, closing the catalyst cycle for the copper,
while the palladium catalyst d is regenerated from the acyl
aryl palladium species f by reductive elimination of the aryl
ketone 3.
Herein we present a new strategy for the synthesis of aryl
ketones in which a-oxocarboxylic acid salts are converted into
acyl copper species by decarboxylation on a copper catalyst
and then arylated with aryl halides on a palladium catalyst.
[*] Prof. Dr. L. J. Gooßen, F. Rudolphi, C. Oppel, Dr. N. Rodríguez
FB Chemie – Organische Chemie, TU Kaiserslautern
Erwin-Schrödinger-Strasse, Geb. 54
67663 Kaiserslautern (Germany)
Fax: (+49)631-205-3921
E-mail: goossen@chemie.uni-kl.de
Homepage: http://www.chemie.uni-kl.de/goossen
[**] We thank the Deutsche Forschungsgemeinschaft and Saltigo
GmbH for financial support and the FCI, the Studienstiftung des
deutschen Volkes, and the Alexander-von-Humboldt-Stiftung for
fellowships.
Supporting information for this article is available on the WWW
under http://www.angewandte.com or from the author.
Scheme 2. Postulated catalyst cycle for the aryl ketone synthesis.
Angew. Chem. Int. Ed. 2008, 47, 3043 –3045
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
3043

Communications
To develop a feasible ketone synthesis based on these
to)bis[o-(tolylphosphanyl)benzyl]dipalladium(II) led to
results (entry 17, Table 1) comparable to those obtained
with the palladium catalyst generated in situ.[18] Of the
solvents tested, quinoline and NMP/quinoline mixtures
proved particularly suitable. Using pyridine instead of
quinoline makes it possible to adjust the reflux temperature
to 1708C, which is particularly useful for larger scale reactions
since residual moisture can then be removed continuously by
azeotropic distillation.
To ensure that the catalyst system is also suitable for less
reactive a-oxocarboxylic acids as substrates, we carried out a
number of experiments with potassium 3,3,3-trimethylpyru-
vate (1b). Since even this sterically highly shielded derivative
reacted in good yields within 36 h (entry 21, Table 1), we used
the optimized catalyst system without modification for the
reactions of a variety of aryl bromides with different a-
oxocarboxylic acids (Table 2). On the one hand, we found that
potassium oxophenylacetate reacted with electron-rich and
electron-deficient aryl and heteroaryl bromides to give the
phenyl ketones in good yields, and that a wide range of
functional groups including esters, ketones, and nitrile groups
are tolerated. On the other hand, 4-bromotoluene was
likewise coupled in good yields with diverse alkyl-, aryl- and
heteroaryl-substituted a-oxocarboxylic acids. The limits of
the current catalyst system were reached only with thermally
labile (3ia, vinylic a-oxocarboxylic acids) and sterically
extremely hindered substrates (3ao, 3la).
mechanistic considerations, we investigated different combi-
nations of copper and palladium catalysts for their efficacy in
the test reaction of potassium oxophenylacetate (1a) with 4-
bromotoluene (2a) (Table 1). We were pleased to find that a
catalyst system that had already proved to be effective in the
decarboxylative biaryl synthesis (15 mol% copper(I) iodide/
1,10-phenanthroline and 1 mol% palladium(II) acetylaceto-
nate in N-methylpyrrolidone (NMP)/quinoline at 1708C)
afforded modest yields of phenyl 4-tolyl ketone (entry 1,
Table 1). The observation that neither copper nor palladium
alone were active as catalysts (entries 2and 3, Table 1)
substantiates our proposed reaction path, while alternative
mechanisms of the Ullmann or Heck type appear improbable.
Further investigations showed that particularly effective
catalysts are formed from copper(I) bromide as the copper
source and palladium(II) bis(1,1,1,5,5,5-hexafluoroacetyl-
acetonate), [Pd(F6-acac)2] as the palladium source (entries 6
and 9, Table 1). The conversion benefits from the addition of
phosphane ligands; tris(o-tolyl)phosphane (P(o-Tol)3) pro-
vided the best results (entries 14 and 15, Table 1). This
optimized catalyst gave 70% yield after a reaction time of just
6 h and an almost quantitative yield after 16 h (entry 16,
Table 1). The stable preformed palladacycle trans-di(m-aceta-
Table 1: Development of the catalyst system.[a]
In summary, this decarboxylative cross-coupling reaction
constitutes a single-step synthesis of aryl ketones that is
broadly applicable, and unlike conventional ketone syntheses
requires no organometallic reagents. Instead, easy-to-handle,
readily accessible salts of carboxylic acids, some of which are
available on an industrial scale as intermediates in the
production of amino acids, are used as source of the acyl
nucleophiles. This reaction clearly demonstrates that the
concept of decarboxylative cross-couplings is by no means
restricted to the synthesis of biaryls but can serve as the basis
for the development of a broad spectrum of sustainable cross-
coupling reactions. An extension of the decarboxylative
couplings to aryl chloride substrates and the reduction of
the reaction temperatures by development of more active
decarboxylation catalysts are the subject of further investiga-
tions.
Entry Cu
Pd
Ligand
Solvent
Yield
[%]
source source
1
2
3
CuI
CuI
–
[Pd(acac)2]
–
[Pd(acac)2]
–
–
–
–
–
–
–
–
–
NMP/quin.[b]
NMP/quin.[b]
NMP/quin.[b]
NMP/quin.[b]
NMP/quin.[b]
NMP/quin.[b]
NMP/quin.[b]
NMP/quin.[b]
NMP/quin.[b]
NMP/quin.[b]
NMP/quin.[b]
NMP/quin.[b]
26
0
0
4
5
6
7
Cu2O [Pd(acac)2]
CuOAc [Pd(acac)2]
CuBr [Pd(acac)2]
CuBr [Pd2(dba)3][c]
CuBr PdCl2
20
35
37
31
34
38
26
31
33
62
58
70
99[e]
95[e]
8
9
CuBr [Pd(F6-acac)2]
10
11
12
13
14
15
16
17
18
19
20
CuBr [Pd(F6-acac)2] binap[c]
CuBr [Pd(F6-acac)2]
P P h
3
CuBr [Pd(F6-acac)2] dppf[c]
CuBr [Pd(F6-acac)2] P p(-MeOC6H4)3 NMP/quin.[b]
CuBr [Pd(F6-acac)2] P o(-Tol)3
CuBr [Pd(F6-acac)2] P o(-Tol)3
CuBr [Pd(F6-acac)2] P o(-Tol)3
CuBr [((o-Tol)2PC7H6)Pd(m-OAc)]2
CuBr [Pd(F6-acac)2] P o(-Tol)3
NMP/quin.[b]
NMP/quin.[b]
NMP/quin.[b]
NMP/quin.[b]
Experimental Section
[d]
[d]
3ba:
A mixture of potassium 3,3,3-trimethylpyruvate (5.05 g,
30.0 mmol),
(104.1 mg, 0.20 mmol), and copper(I) bromide (430.4 mg,
3.00 mmol) under nitrogen was treated with solution of 4-
palladium(II)
1,1,1,5,5,5-hexafluoroacetacetonate
mesitylene/quin.[b] 20[e]
[d]
a
[d]
[e]
CuBr [Pd(F6-acac)2] P o(-Tol)3
NMP56
bromotoluene (3.42 g, 2.46 mL, 20 mmol), tris(o-tolyl)phosphane
(182.6 mg, 0.6 mmol), and 1,10-phenanthroline (541 mg, 3.0 mmol)
in 36 mL NMP and 8.5 mL pyridine with the exclusion of air and
moisture. The reaction mixture was heated at reflux (1708C) for 36 h,
cooled, and filtered through Celite, and the filter cake was rinsed with
diethyl ether. The filtrate was washed with 1m hydrochloric acid (3 ”
20 mL), and the aqueous phases were extracted with diethyl ether (3 ”
20 mL). The combined organic phases were washed with 100 mL
saturated sodium chloride solution, dried over magnesium sulfate,
and filtered. After removal of the solvent and Kugelrohr distillation
92[e]
90[g]
[d]
CuBr [Pd(F6-acac)2] P o(-Tol)3
quin.
21[f] CuBr [Pd(F6-acac)2] P o(-Tol)3
NMP/quin.[b]
[d]
[a] Reaction conditions: 15 mol% Cu cat., 1 mol% Pd cat., 3 mol%
ligand (1.5 mol% with bidentate ligands), 15 mol% 1,10-phenanthroline,
2 mL solvent (quin.=quinoline), 1708C, 6 h. [b] 3:1 ratio. [c] dba=
dibenzylideneacetone,
binap=2,2’-bis(diphenylphosphanyl)-1,1’-
binaphthyl, dppf=1,1’-bis(diphenylphosphanyl)ferrocene. [d] 2 mol%
ligand. [e] After 16 h. [f] Potassium 3,3,3-trimethylpyruvate as substrate.
[g] After 36 h.
3044 www.angewandte.org
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2008, 47, 3043 –3045

Angewandte
Chemie
Table 2: Scope of the new ketone synthesis.[a]
The experiments in Table 2were carried out on a 1.00-mmol scale
in 20-mL septum-capped vessels. The products were purified by
column chromatography (SiO2, hexane/ethyl acetate gradient) after
aqueous workup.
Received: November 6, 2007
Published online: March 10, 2008
Product
Yield [%][a] Product
83
Yield [%][b]
90[c]
Keywords: aryl ketones · copper · cross-coupling ·
.
homogeneous catalysis · palladium
[1] a) P. J. Masson, D. Coup, J. Millet, N. L. Brown, J. Biol. Chem.
1994, 270, 2662 – 2668; b) K. R. Romines, G. A. Freeman, L. T.
Schaller, J. R. Cowan, S. S. Gonzales, J. H. Tidwell, C. W.
Andrews, D. K. Stammers, R. J. Hazen, R. G. Ferris, S. A.
Short, J. H. Chan, L. R. Boone, J. Med. Chem. 2006, 49, 727 – 739.
[2] C. W. Schellhammer in Methoden der Organischen Chemie
(Houben-Weyl), Volume VII/2a (Ed.: E. Müller), Thieme,
Stuttgart, 1973, p. 15 – 379.
[3] J. Cason, J. Am. Chem. Soc. 1942, 64, 1106 – 1110.
[4] S. Nahm, S. M. Weinreb, Tetrahedron Lett. 1981, 22, 3815 – 3818.
[5] E. Negishi, F. Liu in Metal-Catalyzed Cross-Coupling Reactions
(Eds.: F. Diederich, P. J. Stang), Wiley-VCH, Weinheim, 1998,
p. 18 – 19.
[6] a) E. Negishi, V. Bagheri, S. Chatterjee, F. T. Luo, J. A. Miller,
A. T. Stoll, Tetrahedron Lett. 1983, 24, 5181 – 5184; b) P. Kno-
chel, M. C. P. Yeh, S. C. Berk, J. Talbert, J. Org. Chem. 1988, 53,
2392 – 2394.
[7] M. Haddach, J. R. McCarthy, Tetrahedron Lett. 1999, 40, 3109 –
3112.
[8] a) L. J. Gooßen, K. Ghosh, Angew. Chem. 2001, 113, 3566 – 3568;
Angew. Chem. Int. Ed. 2001, 40, 3458 – 3460; b) L. J. Gooßen, K.
Ghosh, Eur. J. Org. Chem. 2002, 3254 – 3267; c) L. J. Gooßen, L.
Winkel, A. Dꢀhring, K. Ghosh, J. Paetzold, Synlett 2002, 1237 –
1240.
[9] L. J. Gooßen, K. Ghosh, Chem. Commun. 2001, 2084 – 2085.
[10] E. J. Corey, D. Seebach, Angew. Chem. 1965, 77, 1134 – 1135;
Angew. Chem. Int. Ed. Engl. 1965, 4, 1075 – 1077.
[11] a) G. Stork, L. Maldonado, J. Am. Chem. Soc. 1971, 93, 5286 –
5287; b) R. E. Damon, R. H. Schlessinger, Tetrahedron Lett.
1975, 16, 4551 – 4554; c) D. Seebach, Angew. Chem. 1969, 81,
690 – 700; Angew. Chem. Int. Ed. Engl. 1969, 8, 639 – 649;
d) R. M. Adlington, J. E. Baldwin, J. C. Bottaro, M. W. D. Perry,
J. Chem. Soc. Chem. Commun. 1983, 1040 – 1041.
82
59
67
56
83
82
99
70
57
78
72
50
73
72
64
96
26
69[c]
59
[12] A. Takemiya, J. F. Hartwig, J. Am. Chem. Soc. 2006, 128, 14800 –
14801.
51
[13] a) Y. C. Huang, K. K. Majumdar, C. H. Cheng, J. Org. Chem.
2002, 67, 1682– 1684; b) S. Ko, B. Kang, S. Chang, Angew. Chem.
2005, 117, 459 – 461; Angew. Chem. Int. Ed. 2005, 44, 455 – 457.
[14] a) L. J. Gooßen, G. Deng, L. M. Levy, Science 2006, 313, 662–
664; b) L. J. Gooßen, N. Rodrꢁguez, B. Melzer, C. Linder, G.
Deng, L. M. Levy, J. Am. Chem. Soc. 2007, 129, 4824 – 4833.
[15] L. J. Gooßen, W. R. Thiel, N. Rodrꢁguez, C. Linder, B. Melzer,
Adv. Synth. Catal. 2007, 349, 2241 – 2246.
45
52
34[c]
5[d]
[16] Decarboxylation of a-oxocarboxylic acids: a) A. Claus, R.
Wollner, Ber. Dtsch. Chem. Ges. 1885, 18, 1856 – 1861; b) T.
Cohen, I. H. Song, J. Am. Chem. Soc. 1965, 87, 3780 – 3781.
[17] Synthesis of acylcuprates from organolithium compounds: D.
Seyferth, R. C. Hui, J. Am. Chem. Soc. 1985, 107, 4551 – 4553.
[18] W. A. Herrmann, C. Brossmer, K. ꢂfele, C. P. Reisinger, T.
Priermeier, M. Beller, H. Fischer, Angew. Chem. 1995, 107,
1989 – 1992; Angew. Chem. Int. Ed. Engl. 1995, 34, 1844 – 1848.
[a] Reaction conditions: 15 mol% CuBr, 1 mol% [Pd(F6-acac)2], 2 mol%
P(o-Tol)3, 15 mol% 1,10-phenanthroline, 2 mL NMP/quinoline (3:1),
1708C, 16 h. The substituent originating from the a-oxocarboxylic acid is
shown on the left side of the carbonyl group. [b] Yield of isolated product.
[c] After 36 h. [d] GC yield.
(1058C/4 ” 10À3 mbar), the product 3ba was obtained as a yellow oil
(3.17 g, 90% yield). Its spectroscopic data correspond to those of tert-
butyl 4-tolyl ketone (CAS 30314-44-4).
Angew. Chem. Int. Ed. 2008, 47, 3043 –3045
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
3045