Biaryl and aryl ketone synthesis via Pd-catalyzed decarboxylase coupling of carboxylate salts with aryl triflates

Article abstract of DOI:10.1002/chem.200900892

A bimetallic catalyst system has been developed that for the first time allows the decarboxylative crosscoupling of aryl and acyl carboxylates with aryl triflates. In contrast to aryl halides, these electrophiles give rise to non-coordinating anions as byproducts, which do not interfere with the decarboxylation step that leads to the generation of the carbon nucleophilic crosscoupling partner. As a result, the scope of carboxylate substrates usable in this transformation was extended from ortho-substituted or otherwise activated derivatives to a broad range of ortho-, meta-, and para-substituted aromatic carboxylates. Two alternative protocols have been optimized, one involving heating the substrates in the presence of Cu^I/1,10- phenanthroline (10-15 mol %) and PdI₂/phosphine (23 mol%) in NMP for 1-24 h, the other involving Cu^I/l,10-phenanthroline (615mol%) and PdBr₂/Tol-BINAP (2 mol % ) in NMP using microwave heating for 5-10 min. While most products are accessible using standard heating, the use of microwave irradiation was found to be beneficial especially for the conversion of non-activated carboxylates with functionalized aryl triflates. The synthetic utility of the transformation is demonstrated with 48 examples showing the scope and limitations of both protocols. In mechanistic studies, the special role of microwave irradiation is elucidated, and further perspectives of decarboxylase crosscouplings are discussed.

Full text of DOI:10.1002/chem.200900892

DOI: 10.1002/chem.200900892
Biaryl and Aryl Ketone Synthesis via Pd-Catalyzed Decarboxylative
Coupling of Carboxylate Salts with Aryl Triflates
Lukas J. Goossen,* Christophe Linder, Nuria Rodrꢀguez, and Paul P. Lange[a]
9336
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 9336 – 9349

FULL PAPER
Abstract: A bimetallic catalyst system
has been developed that for the first
time allows the decarboxylative cross-
coupling of aryl and acyl carboxylates
with aryl triflates. In contrast to aryl
halides, these electrophiles give rise to
non-coordinating anions as byproducts,
which do not interfere with the decar-
boxylation step that leads to the gener-
ation of the carbon nucleophilic cross-
coupling partner. As a result, the scope
of carboxylate substrates usable in this
transformation was extended from
ortho-substituted or otherwise activat-
ed derivatives to a broad range of
ortho-, meta-, and para-substituted aro-
matic carboxylates. Two alternative
protocols have been optimized, one in-
volving heating the substrates in the
presence of CuI/1,10-phenanthroline
(10–15 mol%) and PdI2/phosphine (2–
3 mol%) in NMP for 1–24 h, the other
involving CuI/1,10-phenanthroline (6–
heating for 5–10 min. While most prod-
ucts are accessible using standard heat-
ing, the use of microwave irradiation
was found to be beneficial especially
for the conversion of non-activated car-
boxylates with functionalized aryl tri-
flates. The synthetic utility of the trans-
formation is demonstrated with 48 ex-
amples showing the scope and limita-
tions of both protocols. In mechanistic
studies, the special role of microwave
irradiation is elucidated, and further
perspectives of decarboxylative cross-
couplings are discussed.
15 mol%)
and
PdBr2/Tol-BINAP
(2 mol%) in NMP using microwave
Keywords: aryl triflates · carboxylic
acids · cross-coupling · homogene-
ous catalysis · palladium
Introduction
metal catalyst.[8] Examples include Heck reactions,[9] Sono-
gashira couplings,[10] a-arylations of carbonyl compounds,[11]
and couplings initiated by catalytic ortho-metallation steps
(Scheme 1).[12] Unfortunately, most of these transformations
proceed regiospecifically only for a limited range of sub-
strate types.
The development of new synthetic methods has always been
a key area of organic chemistry. The principal aim is to
refine and extend the existing network of synthetic transfor-
mations by adding new paths to access valuable products
from diverse starting points, by integrating renewable raw
materials, and by improving the performance and sustaina-
bility of available synthetic steps. As a result, an increasing
number of target molecules are becoming accessible in
fewer and safer steps, higher yields and with less waste,
starting from inexhaustible base chemicals. In the area of
À
carbon carbon bond formation, the regiospecificity, toler-
ance of functional groups and reduction of byproducts and
waste has steadily been improved by the discovery of and
constant progress in metal-catalyzed cross-coupling technol-
ogy,[1] and the ongoing development of better reaction var-
iants and catalysts. In these transformations, organometallic
reagents, for example, organoboron,[2] -tin,[3] -zinc,[4]
-copper,[5] -silicon[6] or -magnesium[7] compounds, are cou-
pled with organohalides or pseudohalides at positions prede-
fined by the two complementary leaving groups. Modern
protocols reach impressive performance levels in terms of
selectivity, functional group tolerance, and yield. However,
their inherent weakness lies in the necessity to generate stoi-
chiometric organometallic reagents in a separate step that
are to be used as the carbon nucleophiles.
À
Scheme 1. Strategies for aromatic carbon carbon bond formation. M=B,
Mg, Si, Zn, Cu, Sn, etc.; X=halogen, OSO2R’’, OMe, OC(O)R’’, etc.;
Y=carbon residue.
In the context of our work on the use of carboxylic acids
as substrates in homogeneous catalysis,[13] we developed de-
carboxylative cross-couplings as an alternative strategy for
À
C C bond formation. In this reaction type, the carbon nu-
cleophiles are generated in situ by a catalytic decarboxyla-
tion of carboxylate salts and cross-coupled with carbon elec-
trophiles.[14] It maintains the advantage presented by tradi-
tional cross-couplings of being able to predefine the position
This drawback can be overcome with alternative methods
À
for carbon carbon bond formation in which the carbon nu-
À
cleophile is generated under formal C H activation at a
À
of carbon carbon bond formation by leaving groups, while
obviating the need for stoichiometric organometallic re-
agents. Based on elemental steps also found in the pioneer-
ing work on decarboxylation reactions by Cohen,[15] Nils-
son,[16] and Myers,[17] we propose a reaction principle as illus-
trated in Scheme 2.
[a] Prof. Dr. L. J. Goossen, C. Linder, Dr. N. Rodrꢁguez, P. P. Lange
FB Chemie–Organische Chemie, TU Kaiserslautern
Erwin-Schroedinger-Strasse Geb. 54
67663 Kaiserslautern (Germany)
Fax : (+49)631-205-3921
E-mail: goossen@chemie.uni-kl.de
The reaction starts with the extrusion of CO2 from a
copper carboxylate II generated by salt exchange between a
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.200800892.
Chem. Eur. J. 2009, 15, 9336 – 9349
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
9337

L. J. Goossen et al.
our decarboxylative cross-coupling protocol. Still, the initial
procedures allowed the conversion only of ortho-substituted
benzoates, heteroaromatic or otherwise activated deriva-
tives. Subsequent mechanistic studies have indicated that
this restriction is not intrinsic but a consequence of the
strong affinity of the copper catalyst to the halide ions re-
leased in the cross-coupling step.[18,25]
Thus, the exchange of a halide ion for an arenecarboxy-
late at the copper center—required for catalytic turnover—
is unfavorable unless aided by coordinating substituents in
the vicinity of the carboxylate group. It will be a great chal-
lenge in future developments of the decarboxylation catalyst
to overcome this limitation by designing ligand systems that
induce a stronger preference for carboxylates over halide
ions at the copper(I) center. An analogous strategy appears
to be almost hopeless for silver-based decarboxylation sys-
tems considering the tremendous stability of silver halides.
We herein present a second strategy to surmount the
above restriction for both copper- and silver-based systems.
With newly designed palladium catalysts, we succeeded in
activating and cross-coupling aryl triflates as carbon electro-
philes with non-coordinating leaving groups. Because the tri-
flate ions released in the process are unable to block the car-
boxylates out of the coordination sphere of the decarboxyla-
tion catalyst regardless of their substitution patterns, a much
broader range of carboxylate substrates can be coupled with
aryl triflates compared to aryl bromides. In a preliminary
communication, we disclosed a first protocol with pertinent
examples to demonstrate the viability of this approach.[26] In
this full paper we give a detailed report of the catalyst de-
velopment and the exploration of scope and limitations of
the new transformation. Moreover, we present further stud-
ies leading to an improved protocol that proceeds within
several minutes rather than hours, gives better yields com-
pared to those reported with the initial procedure and
allows the synthesis of many new examples.
Scheme 2. Decarboxylative cross-coupling reactions; R=aryl, heteroaryl,
vinyl, acyl; X=I, Br, Cl, OTf.
potassium carboxylate 1 and a copper(I)/1,10-phenanthro-
line catalyst I. The resulting organocopper species III trans-
fers its carbon residue to an arylpalladium(II) complex V
generated by oxidative addition of an aryl electrophile 2 to
a palladium co-catalyst IV, giving rise to an organopalladium
species VI, and liberating the copper(I)–phenanthroline salt
I. The catalytic cycle for the palladium is closed by reductive
elimination of the cross-coupled product 3 and regeneration
of the initial palladium(0) species IV. The rate of the decar-
boxylation step strongly depends on the electronic nature of
the carboxylate substrate, and a good balance of the rates of
decarboxylation and cross-coupling is critical for minimizing
side reactions and achieving good yields of the desired prod-
uct.
In initial publications, the viability of this concept was
demonstrated with the successful coupling of various aro-
matic carboxylates with aryl halides using a catalyst system
consisting of 10 mol% of a CuIBr/1,10-phenanthroline cata-
lyst along with 3 mol% palladium bromide.[14,18] Subsequent-
ly, the catalyst system has undergone continuous improve-
ment, resulting, for example, in an extension of the substrate
scope from aryl bromides to aryl chlorides.[19] Moreover, a
novel aryl ketone synthesis was discovered based on an
analogous decarboxylative coupling of a-oxocarboxylates
with aryl halides.[20] In such decarboxylative cross-couplings,
copper and silver can be employed as mediators for the de-
carboxylation step.[18,21] Becht et al. showed that the use of
silver in overstoichiometric amounts is advantageous espe-
cially for methoxy-substituted arenecarboxylates.[22] Steg-
lich,[23] Forgione and Bilodeau[24] discovered a different cata-
lytic pathway for palladium-catalyzed decarboxylative cross-
couplings, which is applicable specifically to five-ring hetero-
arenes bearing a carboxylate group in the 2-position. In
their protocol, the carbopalladation presumably precedes
the decarboxylation, so that the position of the carboxylate
group is not the primary factor that defines the position of
Results and Discussion
Aryl triflates have successfully been used as coupling part-
ners in various transition metal-catalyzed reactions, for ex-
ample, cross-couplings,[27] aminations,[28] cyanations,[29] and
Heck-type couplings.[30] Their propensity to undergo oxida-
tive addition is high, at a level comparable to that of aryl
bromides.[31] As aryl triflates are generated from phenols,
which in turn are usually accessed via different synthetic
routes than aryl halides, they ideally complement the spec-
trum of substitution patterns available for the electrophilic
coupling partner. Unfortunately, aryl triflates are prone to
side reactions that are not observed for aryl halides, the
most important being a cleavage of the triflate ester by
water or nucleophiles under liberation of the corresponding
phenol or phenyl ester. To ensure that the results obtained
in the test reactions reflect only the catalyst performance
and are not overlaid by such uncatalyzed background reac-
tions, both model substrates for the desired decarboxylative
À
carbon carbon bond formation.
Based on the proposed mechanism (Scheme 2), one
would not expect to see such an inherent limitation regard-
ing the substitution pattern at the arenecarboxylate salt for
9338
www.chemeurj.org
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 9336 – 9349

Pd-Catalyzed Decarboxylative Coupling
FULL PAPER
cross-coupling had to be chosen with great care. We decided
to use 4-tolyl triflate (2a) as the electrophilic coupling part-
ner, as this compound can be expected to be more hydrolyti-
cally stable than esters of more electron-deficient phenols.
Nitro-substituted benzoates appeared to be an advantageous
source of the nucleophilic coupling partner, because prior to
decarboxylation the nucleophilicity of these electron-defi-
cient carboxylates is relatively low, making them less likely
to attack the aryl triflates via nucleophilic addition–elimina-
tion reactions.
Thus, we based our search for an effective catalyst system
for the desired decarboxylative cross-coupling on the model
reaction of potassium 3-nitrobenzoate (1a) with 4-tolyl tri-
flate (2a), and evaluated various palladium cross-coupling
catalysts in combination with CuI/1,10-phenanthroline decar-
boxylation co-catalysts. The key findings are summarized in
Table 1, additional results having been disclosed in the pre-
liminary communication.[26]
suitable copper precursors, and while palladium(II) iodide
was the most effective palladium source, other palladium(II)
salts or palladium(0) precatalysts can also be used. With the
best protocol resulting from these investigations, 3-nitro-4’-
methylbiphenyl (3aa) was synthesized in 70% yield from
potassium 3-nitrobenzoate (1a) and excess 4-tolyl triflate
(2a) using 15 mol% Cu/1,10-phenanthroline and 3 mol%
PdI2/Tol-BINAP catalysts in the polar aprotic solvent NMP
(entry 8). Our hypothesis that there is no intrinsic limitation
of decarboxylative cross-couplings to 2-substituted carboxyl-
ate substrates is unambiguously validated by this result,
proving that the substrate restrictions originally observed
can be overcome when using aryl electrophiles with weakly
coordinating leaving groups instead of aryl halides. Howev-
er, the relatively harsh reaction conditions somewhat limit
the preparative utility of this initial protocol.
Inspired by reports on other transition-metal-catalyzed re-
actions[33] in which the reaction times could dramatically be
shortened using microwave irradiation instead of conven-
tional heating, for example, cross-couplings,[34] Buchwald–
Hartwig aminations,[35] Heck reactions,[36] and decarboxyla-
tive couplings of five-ring heteroarenes,[24] we sought for a
protocol that would allow decarboxylative cross-couplings to
similarly benefit from the efficient heating that microwave
irradiation provides. However, this proved to be a more
complex modification than expected. We attribute this to
the fact that in our decarboxylative coupling, two active
metal catalysts need to concurrently be formed in situ, and
their two interconnected catalytic cycles need to be micro-
wave-accelerated to a similar extent in order to remain
synchronized. Indeed, a closer inspection of relevant litera-
ture examples revealed that none of them involved two
metals present only in catalytic amounts. For example, the
seemingly related decarboxylative couplings that Forgione
and Bilodeau successfully performed in a microwave reac-
tor[24] proceed via a different mechanism that involves only
a palladium catalyst without a co-catalyst. Crabtree et al.
very recently disclosed a microwave-assisted protocol for an
Ag/Pd-mediated decarboxylative coupling, but they had to
use the decarboxylation catalyst in overstoichiometric
amounts to achieve reasonable turnover.[37]
Table 1. Development of the first effective catalyst system.[a]
Entry
Cu source
Pd source
Phosphine
Yield [%]
1
2
3
4
5
6
7
CuI
CuI
CuI
CuI
Cu2O
Cu2O
Cu2O
Cu2O
Cu2O
Pd
Pd
Pd
Pd
Pd
Pd
PdI2
PdI2
PdI2
N
P
P
G
15
39
38
47
52
31
58
70
59
BINAP
Tol-BINAP
Tol-BINAP
Tol-BINAP
Tol-BINAP
Tol-BINAP
Tol-BINAP
8[b]
9[c]
[a] Reaction conditions: 1 mmol of potassium 3-nitrobenzoate, 2 mmol
of 4-tolyl triflate, 15 mol% Cu source (7.5 mol% for Cu2O), 15 mol%
1,10-phenanthroline, 3 mol% Pd source, 9 mol% phosphine (4.5 mol%
for bidentate phosphines), 4 mL N-methylpyrrolidone (NMP), 1708C,
16 h. Yields determined by GC analysis using n-tetradecane as the inter-
nal standard. [b] 24 h reaction time. [c] Microwave heating: 1908C,
10 min (see Supporting Information).
The results illustrate that the choice of phosphine was the
decisive factor for the successful development of this reac-
tion (entries 1–4). This is not surprising as no coordinating
anions are present to stabilize the palladium, so that it can
only be held in solution if effectively stabilized by appropri-
ate ligands.[32] In contrast to other cross-coupling reactions
of triflates, in which the addition of excess halide salts is
often beneficial, this was not an option here as the halides
would have interfered with the decarboxylation step by co-
ordinating to the copper, as discussed above. The best re-
sults were achieved using the electron-rich, chelating phos-
phine Tol-BINAP (entry 4), and the less expensive mono-
After intricate development of the reaction procedure,
that is, limiting the microwave power to 150 W and preform-
ing the catalyst before submitting it to microwave irradia-
tion, we finally found a microwave-assisted protocol that al-
lowed us performing our model reaction within only minutes
at 1908C, rather than 16 h using conventional heating
(Table 1, entry 9). Still, the yield remained lower than that
achieved in the comparative experiment using conventional
heating (see entry 8).
While evaluating the scope of the initial reaction protocol,
we found that several substrate combinations gave unsatis-
factory yields, presumably caused by thermal decomposition
of the products. This prompted us to take another look at
the alternative microwave-assisted protocol. To exclude that
the limited success of our initial efforts was caused by an un-
fortunate choice of model substrates, we turned to potassi-
dentate triACHTUNGTRENNUNG(p-tolyl)phosphine was also reasonably effective
(entry 2). The copper and palladium precursors employed
were less critical for the reaction outcome (entries 5–7):
Copper(I) iodide and oxide were found to be almost equally
Chem. Eur. J. 2009, 15, 9336 – 9349
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
9339

L. J. Goossen et al.
um 2-nitrobenzoate (1b) as the nucleophilic coupling part-
ner this time, instead of the less reactive 3-nitrobenzoate
(1a). The results of these screening reactions are summar-
ized in Table 2.
the coupling of potassium 3-nitrobenzoate (1a) with 4-tolyl
triflate (2a) (see Table 1): In comparison to the thermal con-
ditions, the concentration was increased by a factor of four
to 1 mmolmLÀ1 to minimize the pressure buildup and still
ensure homogeneity of the mixture. Moreover, the tempera-
ture was raised to 1908C using at most 150 W of microwave
power, and the reaction time was reduced to 5 min. In order
to keep the rate of decarboxylation in balance with the rate
of cross-coupling, it was necessary to adjust the ratio of
copper to palladium from 5:1 to 1.5:1. In doing so, we were
able to reduce the amount of copper catalyst from 10 to
3 mol%, and still achieve an encouraging 62% yield of the
biaryl 3ba (entry 6). Careful reevaluation of the catalyst
Table 2. Development of a microwave-assisted reaction protocol.[a]
Entry
Cu Source
Pd Source
Phosphine
(p-Tol)3
Tol-BINAP
Yield [%]
1
2
3
4
Cu2O
Cu2O
Cu2O
Cu2O
Ag2CO3
Cu2O
Cu2O
Cu2O
Cu2O
Cu2O
Cu2O
Cu2O
Cu2O
CuI
PdI2
PdI2
PdBr2
P
G
83
74
77
64
52
62
9
44
52
61
78
63
80
79
51
81
80
system revealed that Tol-BINAP and PACTHNUTRGNEN(UG p-Tol)3 were similar-
P
P
ACHTUNGTRENNUNG
ly effective in conjunction with PdI2, and were superior to
all other phosphines (entries 6–10). Variation of the palladi-
um source led to the surprising discovery that under micro-
wave conditions and with Tol-BINAP as the ligand, PdI2
was no longer the optimal precursor, but that PdBr2 and in
Pd
N
ACHTUNGTRENNUNG
5[b]
6[c]
7[c]
8[c]
9[c]
10[c]
11[c]
12[c]
13[c]
14[c]
15[c]
16[c,d]
17[e]
PdCl
PdI2
PdI2
PdI2
PdI2
PdI2
N
PPh3
P
–
ACHTUNGTNER(NUNG p-Tol)3
PPh3
BINAP
particular Pd
copper source had a very limited influence on the reaction
outcome (entry 14). Interestingly, P(p-Tol)3, the most effec-
ACHTUNGTERNUN(NG acac)2 gave better yields (entries 11–13). The
Tol-BINAP
Tol-BINAP
Tol-BINAP
Tol-BINAP
Tol-BINAP
PACTHNGUTERNN(UG p-Tol)3
Tol-BINAP
Tol-BINAP
PdBr2
PdCl2
AHCTUNGTRENNUNG
Pd
Pd
Pd
Pd
ACHTUNGTRENNUNG
tive ligand under thermal conditions (entry 1), was now
clearly inferior to Tol-BINAP (entries 13, 15). Under the
best conditions, the reaction time could be reduced even fur-
ther to 2.5 min, and a yield of 81% was achieved (entry 16).
After this second round of catalyst optimization, the yields
obtained were almost identical to those for thermal condi-
tions despite the drastic reduction in both reaction time and
loading of the decarboxylation catalyst.
Based on these findings, we also reevaluated the coupling
of potassium 3-nitrobenzoate (1a) under microwave condi-
tions. Replacing PdACTHNUTRGNEUNG(acac)2 by PdBr2 and slightly increasing
the catalyst loading resulted in a very effective protocol for
this non-activated carboxylate. Yields substantially higher
than obtained with conventional heating were thus achieved
(entry 17).
AHCTUNGTRENNUNG
Cu2O
Cu2O
Cu2O
G
C
PdBr2
[a] Reaction conditions: thermal conditions: 1 mmol of potassium 2-ni-
trobenzoate, 2 mmol of 4-tolyl triflate, 5 mol% Cu2O, 10 mol% 1,10-phe-
nanthroline, 2 mol% Pd source, 6 mol% phosphine (3 mol% for biden-
tate phosphines), 4.0 mL NMP, 1708C, 1 h. Yields determined by GC
analysis using n-tetradecane as the internal standard. [b] 5 mol%
Ag2CO3, 3 mol% PdCl2ACHTUNGTRENNUNG(PPh3)2, 10 mol% PPh3, 4 mL NMP 1208C, 16 h.
[c] Microwave conditions: 3 mol% Cu source (1.5 mol% Cu2O), 3 mol%
1,10-phenanthroline, 2 mol% Pd source, 6 mol% phosphine (3 mol% for
bidentate phosphines), 1 mL NMP, 1908C/150 W/5 min. [d] 2.5 min reac-
tion time. [e] 0.5 mmol potassium 3-nitrobenzoate, 1 mmol 4-tolyl triflate,
7.5 mol% Cu2O, 15 mol% 1,10-phenanthroline, 2 mol% PdBr2, 3 mol%
Tol-BINAP, 3 mL NMP, 1908C/150 W/15 min.
After having identified such effective protocols, we inves-
tigated the scope of the cross-coupling of potassium 2-nitro-
benzoate (1b), an example representative of an activated
carboxylate substrate, with various aryl triflates (Table 3).
Under thermal conditions and using 10 mol% of copper and
2 mol% of palladium catalysts, all reactions were complete
within one hour, and most products were obtained in good
yields. A range of diversely functionalized substituents were
tolerated, including ether, ester, formyl, keto, and pyridyl
groups, fluorides, and chlorides. The yields were usually in
the same range as for the corresponding aryl bromide sub-
strates.[18]
In direct comparison, the microwave-assisted protocol was
more effective for most substrates, as higher yields can often
be obtained within only 5 min and using a lower loading of
the copper catalyst (3 mol%). The current performance
limit of both protocols is reached with ester, acyl, and nitro
groups in the 2- and 4-positions of the aryl triflate. For such
electron-deficient substrates, the thermal stability increas-
ingly became a problem, so that triflate esters of even more
Under thermal conditions (1708C), the best yields are ob-
tained using Cu2O/1,10-phenanthroline as the decarboxyla-
tion catalyst, and PdI2/PACTHNUTRGNE(UGN p-Tol)3 as cross-coupling catalyst
(entry 1). Remarkably, Tol-BINAP is slightly less effective
for this substrate combination (entry 2). All other palladium
and copper sources tested, including Pd
crease in yields (entries 3 and 4).
ACHTUNGTERN(NGNU acac)2, led to a de-
In the absence of halide ions, not only copper(I)- but also
silver(I)-based decarboxylation mediators were found to be
effective in catalytic amounts: After minimal optimization, a
reasonable 52% yield was achieved using 5 mol% silver car-
bonate as a co-catalyst (entry 5). This is the first example of
overstoichiometric turnover in a decarboxylative cross-cou-
pling with a silver/palladium system based on the AgI em-
ployed, and, thus, an important milestone in the develop-
ment of such alternative catalyst systems.
We then performed the reaction in the microwave with
the optimized copper-based catalyst system, using the instru-
ment settings and solvent amounts previously optimized for
9340
www.chemeurj.org
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 9336 – 9349

Pd-Catalyzed Decarboxylative Coupling
FULL PAPER
Table 3. Scope for the coupling of 2-nitrobenzoate with electrophiles.[a]
Scheme 3. Side reaction observed for electron-poor aryl triflates.
Product
Yield [%] Product
Yield [%]
D: 91
mW: 84
D: 98
mW: 99
In order to further extend the scope to such electron-defi-
cient carbon electrophiles, more active catalysts need to be
developed that allow a conversion either at lower tempera-
tures or of more stable sulfonates (see Scheme 5).
The scope of the reaction with regard to the carboxylate
coupling partner was investigated next, using 4-tolyl triflate
(2a) as the carbon electrophilic substrate (Table 4).
D: 83
mW: 87
D: 79
mW: 93
D: 45
D: 30
ortho-Substituted benzoates and five-ring heteroarene-2-
carboxylates were cleanly coupled using both conventional
heating (16 h, 1708C) with 10 mol% copper and 2 mol%
palladium catalysts, and microwave irradiation (5 min,
1908C) with 5 mol% copper and 2 mol% palladium catalyst
(entries 1–8). Both electron-donating and electron-with-
drawing functional groups can be used as substituents in the
ortho-position, including methoxy, acyl, formyl, and cyano
groups. While the scope is reasonable for both protocols, it
became increasingly evident that the microwave protocol is
more effective: Even for particularly sensitive substrates
such as 3ea or 3sa, moderate yields were obtained with this
new protocol, whereas almost no conversion was observed
with the thermal conditions. Moreover, the use of modern
microwave technology, which involves the use of small, con-
tained vessels, certified for pressure reactions, that are easy
to degas and dry, is advantageous also from a practical
standpoint as it makes delicate-to-perform decarboxylative
cross-coupling reactions much simpler to carry out.
For the cross-coupling of the less reactive meta- and para-
substituted carboxylates, we applied the conditions previous-
ly optimized for potassium 3-nitrobenzoate (1a), that is,
conventional heating (24 h, 1708C) with 15 mol% copper
and 3 mol% palladium catalysts, and microwave irradiation
(15 min, 1908C) with 15 mol% copper and 2 mol% palladi-
um catalysts (entries 9–20). Both protocols proved to be ef-
fective for the coupling of a broad range of benzoates bear-
ing methyl, cyano, nitro, methoxy, and amide substituents,
and even for potassium nicotinate as an example of a six-
membered heterocycle. The successful coupling of potassium
3-thiophenecarboxylate illustrates that for this type of decar-
boxylative cross-coupling and in contrast to that of Bilo-
deau, five-membered heterocycles can be arylated not only
in the 2- but also in the 3-position, underlining once again
the mechanistic differences of both protocols.[24]
mW: 54
mW: 23
D: 37
mW: 23
D: traces
mW: 31
D: 99
D: 91
D: 91
D: 80
D: 79
D: 75
D: 76
D: 74
D: 64
D: 63
[a] Reaction conditions: thermal (D): 1 mmol potassium 2-nitrobenzoate,
2 mmol triflate, 5 mol% Cu2O, 10 mol% 1,10-phenanthroline, 2 mol%
PdI2, 6 mol% P
wave (mW): 1.5 mol% Cu2O, 3 mol% 1,10-phenanthroline, 2 mol% Pd-
ACHTUNGTRENNUNG(acac)2, 3 mol% Tol-BINAP, 1 mL NMP, 1908C/150 W/5 min.
ACHTUNGTRENNUNG(p-Tol)3, 4 mL NMP, 1708C, 1 h, isolated yields; micro-
electron-deficient phenols were too unstable to even survive
a few minutes of microwave heating. A control experiment
confirmed that indeed, it is this thermal instability that ac-
counts for the decreased yields, rather than the lack of cata-
lyst performance. Thus, when heating 4-nitrophenyl triflate
(2h) with potassium 2-nitrobenzoate (1b) above 1408C in
the absence of a catalyst, the formation of 2-nitrobenzoic
acid 4-nitrophenyl ester (4bh) is observed, which during
workup partially hydrolyzes under formation of 4-nitrophe-
nol (Scheme 3).
With the new microwave-based protocol, higher yields
were achieved than with that involving conventional heat-
ing, and only few limitations remain: For particularly elec-
tron-rich para-substituted benzoates, for example, potassium
4-anisate, the desired biaryl product was observed only in
trace quantities. Instead, the strong nucleophilicity of this
electron-rich benzoate derivative facilitates the transesterifi-
cation to such an extent that the resulting formation of 4-
Chem. Eur. J. 2009, 15, 9336 – 9349
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
9341

L. J. Goossen et al.
Table 4. Scope with regard to the carboxylate.[a]
or, alternatively, by employing more robust aryl sulfonates
(see Scheme 5).
Product
Yield [%]
Product
Yield [%]
D: 76
D: 72
mW: 73[b,c]
mW: 73[b]
Scheme 4. Decarboxylative cross-coupling of a-oxocarboxylates with aryl
triflates.
D: 30
D: 45
mW: 58[b]
mW: 54[b]
The above investigations revealed that whereas the cou-
pling of non-ortho-substituted benzoates is certainly feasible,
it is substantially more difficult than that of activated car-
boxylates. Beyond demonstrating future perspectives of the
concept of decarboxylative couplings, it was our goal to
probe the immediate synthetic value of the new transforma-
tion. Therefore, we decided to investigate to what extent
non-activated carboxylates, as exemplified by potassium 3-
nitrobenzoate (1a), can be coupled with triflate electro-
philes (Table 5). We were pleased to find that this test sub-
strate was successfully coupled with a sizeable number of
variously substituted aryl triflates.
D: 44
D: 40
mW: 50[b]
mW: 40[b]
D:75
D: 75
mW: 82[b,c]
mW: 75[b,c]
D: 72[d]
D: 68[d]
mW: 84[e]
mW: 81[e]
D: 52[d]
D: 58[d]
mW: 83[e]
mW: 76[e]
D: 44[d]
D: 49[d]
Table 5. Scope for the coupling of 3-nitrobenzoate with electrophiles.[a]
mW: 74[e]
mW: 71[e]
D: 62[d]
D: 53[d]
mW: 69[e]
mW: 59[e]
Product
Yield [%] Product
74
Yield [%]
61
D: 40[d]
D: 5[d,f]
mW: 59[e]
mW: 54[e]
D: 41[d]
D: 54[d]
mW: 50[e]
mW: 65[e]
34
69
[a] Reaction conditions: 1 mmol of potassium carboxylate, 2 mmol of 4-
tolyl triflate, 5 mol% Cu2O, 10 mol% 1,10-phenanthroline, 2 mol% PdI2,
40
64
49
40
6 mol% P
ACHTUNGTRENNUNG
Cu2O, 5 mol% 1,10-phenanthroline, 2 mol% PdACHTUNGTRENNUNG
BINAP, 1 mL NMP, 1908C/150 W/5 min. [c] 1 mmol of 4-tolyl triflate.
[d] 7.5 mol% Cu2O, 15 mol% 1,10-phenanthroline, 3 mol% PdI2,
4.5 mol% Tol-BINAP, 24 h. [e] 0.5 mmol of potassium carboxylate,
1 mmol of 4-tolyl triflate, 7.5 mol% Cu2O, 15 mol% 1,10-phenanthroline,
2 mol% PdBr2, 3 mol% Tol-BINAP, 3 mL NMP, 1908C/150 W/10 min.
[f] Yield determined by GC analysis using n-tetradecane as the internal
standard.
[a] Reaction conditions: 0.5 mmol of potassium 3-nitrobenzoate, 1 mmol
of aryl triflate, 7.5 mol% Cu2O, 15 mol% 1,10-phenanthroline, 2 mol%
PdBr2, 3 mol% Tol-BINAP, 3 mL NMP, 1908C/150 W/10 min, isolated
yields.
anisic acid 4-tolyl ester successfully competes with the de-
sired cross-coupling. For such strongly nucleophilic carboxy-
lates, this uncatalyzed background reaction becomes signifi-
cant starting at temperatures above 1408C (see Scheme 3).
Such remaining restrictions can probably be overcome with
new catalyst generations operating at lower temperatures
Longer reaction times are required for such non-activated
carboxylates, leading to increased thermal stress of the func-
tionalized triflates. Expectedly, the undesired transesterifica-
tion of the triflates was thus more strongly limiting, and sev-
eral triflates did not tolerate the high reaction temperatures
9342
www.chemeurj.org
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 9336 – 9349

Pd-Catalyzed Decarboxylative Coupling
FULL PAPER
Table 6. Effects of additives on the Cu-catalyzed protodecarboxylation.[a]
was added, which has a lower
preference for amine ligands
(entry 4).
Tol-BINAP,
the
ligand used to stabilize the pal-
ladium cross-coupling catalyst,
also has an adverse effect on
the reaction turnover, presuma-
bly due to its coordination to
the copper catalyst in competi-
tion with 1,10-phenanthroline
(entry 5). The presence of a
combination of PdI2 and Tol-
BINAP leaves the rate of de-
carboxylation unaffected for
the 3-nitro derivative (7a), but
depletes the yield for 7b and
shuts down the reaction alto-
gether for 7c (entry 6). This ad-
verse effect of the palladium
co-catalyst on the decarboxyla-
tion is less pronounced when a
Pd0 precursor is employed, and
Entry
3-NO2- (8a)
4-CN- (8b)
Yield [%]
4-MeO (8c)
N Ligand
Pd Source
Pd Source
1
2
3
4
5
6
7
–
–
–
PdI2
–
–
–
–
28
50
49
54
35
59
49
70
59
10
89
51
86
13
19
40
99
42
0
1,10-Phen
1,10-Phen
1,10-Phen
1,10-Phen
1,10-Phen
1,10-Phen
1,10-Phen
1,10-Phen
39 (51)[b]
14
30
0
Pd
–
N
Tol-BINAP
Tol-BINAP
Tol-BINAP
–
PdI2
0
Pd
–
N
11
99
25
8[c]
9[c]
PdI2
Tol-BINAP
[a] Reaction conditions: 1 mmol of benzoic acid derivative, 7.5 mol% Cu2O, 15 mol% N-ligand, 3 mol% Pd
source, 4.5 mol% Tol-BINAP, 2 mL NMP, 1708C, 16 h. Yields were determined by GC analysis using n-tetra-
decane as the internal standard. [b] 15 mol% 4,7-diphenyl-1,10-phenanthroline instead of 1,10-phenanthroline.
[c] 1 mL NMP, 1908C/150 W/5 min.
of the protocol involving conventional heating long enough
to give satisfactory yields of the desired biaryl products. For-
tunately, the newly developed microwave-assisted protocol
was visibly more effective here, and a range of 3-nitro-sub-
stituted biaryls could be synthesized in moderate to good
yields, allowing the conversion even of sensitive triflates
bearing esters, fluoride, or keto groups.
The beneficial effect of microwave-induced heating was
so profound in these last examples that we decided to per-
form a series of photodecarboxylation experiments to eluci-
date the reason for the strong acceleration of the decarboxy-
lation step (Table 6).
All these protodecarboxylation reactions were performed
in parallel for three representative benzoic acids, 3-nitroben-
zoic acid (7a) as an electron-poor, non-activated example,
4-cyanobenzoic acid (7b) as an electron-poor derivative sub-
stituted with a problematic coordinating group, and 4-me-
thoxybenzoic acid (7c) as an electron-rich substrate. Upon
heating the benzoic acids to 1708C over 16 h with CuI oxide
in the absence of a nitrogen ligand, only low conversions
were achieved for the decarboxylation of electron-deficient
derivatives 7a and 7b, and the electron-rich compound 7c
showed no reaction at all (entry 1). When 1,10-phenanthro-
line was present in the mixture as a copper ligand, reasona-
ble turnovers were obtained for all three derivatives, al-
though the performance of especially developed protodecar-
boxylation catalysts was not quite reached (entry 2).[25] An
addition of palladium(II) iodide did not significantly affect
the turnover of 3-nitrobenzoic acid (7a), but had a strong in-
fluence on the decarboxylation of the less reactive deriva-
tives 7b and 7c (entry 3). We attribute this to palladium(II)
competing with copper(I) for the phenanthroline ligands,
leading to a partial loss in activity of the copper decarboxy-
lation catalyst. In contrast, no such effect was observed
can most likely again be attributed to ligand-exchange reac-
tions between CuI and PdII (entry 7). When performing a
similar series of experiments for the microwave-assisted pro-
todecarboxylation, such clear trends were no longer ob-
served. In the control experiments (entry 8), all three car-
boxylic acids were quantitatively converted, and the respec-
tive arenes were detected in high yields. The presence of a
Pd precursor and/or a phosphine ligand affected the decar-
boxylations, but to a much smaller extent than under ther-
mal conditions, so that even for the least reactive derivative
7c, the product, anisole (8c), was still detected in 25% yield
(entry 9). This may indicate that the protodecarboxylation is
strongly accelerated by microwave irradiation, whereas
ligand exchange is not. Another interpretation of these find-
ings is that under microwave irradiation, the palladium(II)
catalyst is almost instantaneously reduced to palladium(0),
which binds strongly to the Tol-BINAP ligand, so that much
less of either species is available to interfere with the copper
decarboxylation catalyst. Moreover, a decomposition of the
palladium catalyst under liberation of phosphine should be
less likely to occur in the short reaction time in the micro-
wave, which would offer an additional explanation for the
better yields achieved with the latter protocol.
We were especially pleased to find that the new catalyst
not only allows the use of aryl triflates in the cross-coupling
of aromatic carboxylates but also in our recently disclosed
decarboxylative aryl ketone synthesis. In this transforma-
tion, a-oxocarboxylate salts are decarboxylated under for-
mation of acyl nucleophiles,[20] which are then directly cou-
pled with the aryl triflates to give the corresponding aryl ke-
tones (Scheme 4). The representative examples illustrate
that both aromatic and an aliphatic a-oxocarboxylate salts
can be coupled in reasonable yields both using conventional
heating and microwave irradiation.
when instead of PdI2, the palladium(0) precursor PdACHTNUTRGNEUNG(dba)2
Chem. Eur. J. 2009, 15, 9336 – 9349
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
9343

L. J. Goossen et al.
Conclusions
represents a greater challenge, but substantial progress was
made over the last months. Since we have already succeeded
in performing protodecarboxylations at temperatures of
only 1208C with a silver-based catalyst (Table 2), we are op-
timistic that with future catalyst generations, decarboxyla-
tive cross-couplings can soon routinely be performed in re-
fluxing toluene.
Overall, the new protocol for the decarboxylative cross-cou-
pling of potassium carboxylates with aryl triflates presents a
powerful strategy for overcoming remaining substrate re-
strictions that had so far limited the practical applicability of
decarboxylative couplings. It allows performing decarboxy-
lative cross-couplings with a much broader range of aromat-
ic carboxylates regardless of their substitution pattern, and
thus gives additional evidence that previously observed re-
strictions of the substrate scope were due largely to the
presence of coordinating halide anions in the reaction mix-
ture. Moreover, it extends the scope of utilizable aryl elec-
trophiles to phenol-derived compounds, which themselves
are often accessed via fundamentally different synthetic
routes than aryl halides. Many biaryls were shown to be ac-
cessible using standard heating, and with the use of micro-
wave irradiation in combination with a specially adapted
catalyst system, an even broader range of non-activated car-
boxylates was successfully coupled with various functional-
ized aryl triflates. The new catalyst systems also allow the
coupling of a-oxocarboxylate salts with aryl triflates to give
the corresponding ketones, which represents a substantial
extension of this decarboxylative ketone synthesis.
Experimental Section
General methods: Reactions were performed in oven-dried glassware
under a nitrogen atmosphere containing a Teflon-coated stirring bar and
dry septum. All microwave irradiation experiments were carried out in
oven-dried appropriate microwave process tubes (10 mL) equipped with
Teflon-coated stirring bars and septum caps under an argon atmosphere.
For the exclusion of atmospheric oxygen from the reaction media, three
freeze-pump thaw cycles were preformed before the reagents were
mixed. Solvents were purified by standard procedures prior to use. All
reactions were monitored by GC using n-tetradecane as an internal stan-
dard. Response factors of the products with regard to n-tetradecane were
obtained experimentally by analyzing known quantities of the substances.
GC analyses were carried out using an HP-5 capillary column (phenyl
methyl siloxane 30 mꢂ320ꢂ0.25, 100/2.3–30–300/3) and a time program
beginning with 2 min at 608C followed by 308CminÀ1 ramp to 3008C,
then 3 min at this temperature. Column chromatography was performed
using a Combi Flash Companion-Chromatography-System (Isco-Systems)
and RediSep packed columns (12 g). NMR spectra were obtained on
Bruker AMX 400 or on Bruker Avance 600 systems using CDCl3,
[D4]MeOH and D2O as solvents, with proton and carbon resonances at
400/600 MHz and 101/151 MHz, respectively. Mass spectral data were ac-
quired on a GC-MS Saturn 2100 T (Varian). Microwave assisted reac-
tions were performed using the Discover LabMate with CEM’s Chem-
Driver reaction monitoring software.
As a next milestone towards establishing decarboxylative
cross-couplings as true alternatives to traditional carbon
À
carbon bond forming methods based on preformed organo-
metallic reagents, we see the development of improved cata-
lysts that will allow using more robust and less expensive
sulfonates, for example, mesylates or tosylates. These com-
pounds can be expected to possess an enhanced stability
against thermal decomposition and to be resistant against
nucleophilic attack by electron-rich carboxylates, and their
use could thus allow accessing the few product structures
still outside of the protocols disclosed herein. Although we
are a long way from having discovered a practical protocol
for the conversion of these derivatives, first experiments
have provided very encouraging results. For example, 23%
of 4-methyl-2’-nitrobiphenyl (3ba) was formed when cou-
pling potassium 2-nitrobenzoate (1b) with 4-tolyl tosylate
(9a) using 10 mol% CuI/1,10-phenanthroline and 5 mol%
1-Methyl-2-pyrrolidone (NMP) was dried by removing water as a toluene
azeotrope. Copper salts and palladium(II) iodide were dried in vacuo at
60 and 808C, respectively, prior to use. All potassium salts were dried for
2 h in vacuo at room temperature prior to use. All other compounds are
commercially available and were used without further purification.
General procedure for the synthesis of potassium salts of the carboxylic
acids (1a–u): A 250 mL, two-necked, round-bottomed flask was charged
with the carboxylic acid (20.0 mmol) and ethanol (20.0 mL). To this, a so-
lution of potassium tert-butoxide (2.24 g, 20.0 mmol) in ethanol (20.0 mL)
was added dropwise over 2 h. After complete addition, the reaction mix-
ture was stirred for another 1 h at room temperature. A gradual forma-
tion of a white precipitate was observed. The resulting solid was collected
by filtration through a 7 cm Bꢃchner funnel, washed sequentially with
ethanol (2ꢂ10.0 mL) and cold (08C) diethyl ether (10.0 mL), transferred
to a round-bottomed flask, and dried at 2ꢂ10À3 mmHg to provide the
corresponding potassium salts of the carboxylic acids 1a–u in 81–98%
yield.
PdACHTUNGTRENNUNG(OAc)2/XPhos catalysts in NMP at 1708C.
General procedure for the synthesis of aryl trifluoromethanesulfonates
(2a–r):
A solution of trifluoromethanesulfonic anhydride (4.00 mL,
24.0 mmol) in CH2Cl2 (10.0 mL) was added dropwise to a solution of the
corresponding phenol (20.0 mmol) and pyridine (3.23 mL, 40.0 mmol) in
anhydrous CH2Cl2 (20 mL) at 08C. After complete addition, the mixture
was warmed to room temperature and allowed to stir for 1 h. The mix-
ture was then diluted with Et2O (30 mL), quenched with 10% aq. HCl
and washed successively with sat. NaHCO3 and brine. After drying over
MgSO4, the solvent was removed under reduced pressure and the residue
was purified by Kugelrohr distillation to give the trifluoromethanesulfo-
nates 2a–r in 79–99% yield.
Scheme 5. First example of a successful coupling of an aryl tosylate.
Finally, the temperatures required by current catalyst sys-
tems for decarboxylative cross-couplings remain high.
Having now established that such transformations have a
much broader scope than originally perceived, the develop-
ment of decarboxylation catalysts capable of operating at
lower temperatures will be in the focus of our interest. This
General procedure for the biaryl synthesis under thermal conditions
Method A (activated carboxylates in Tables 3 and 4): An oven-dried, ni-
trogen-flushed 20 mL vessel was charged with the potassium carboxylate
1b–j (1.00 mmol), copper(I) oxide (7.2 mg, 0.05 mmol), palladium(II)
9344
www.chemeurj.org
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 9336 – 9349

Pd-Catalyzed Decarboxylative Coupling
FULL PAPER
iodide (7.2 mg, 0.02 mmol), 1,10-phenanthroline (18.0 mg, 0.10 mmol)
Compound 3aa was also prepared in 84% yield (179 mg) using the same
and P
(p-Tol)3 (18.0 mg, 0.06 mmol). Under an atmosphere of nitrogen, a
amounts of starting materials via two 0.5 mmol batches following Method
D.
degassed solution of the aryl triflate (2a–r) (2.00 mmol) and the internal
standard n-tetradecane (50 mL) in NMP (4 mL) was added via syringe.
The resulting mixture was stirred at 1708C for 1–16 h depending on the
reactivity of the substrate. After the reaction was complete, the mixture
was cooled to room temperature, diluted with aqueous HCl (1N, 10 mL)
and extracted with ethyl acetate (3ꢂ20.0 mL). The combined organic
layers were washed with water and brine, dried over MgSO4, filtered, and
the volatiles were removed in vacuo. The residue was purified by column
chromatography (SiO2, ethyl acetate/hexane gradient) yielding the corre-
sponding biaryl.
4’-Methyl-2-nitrobiphenyl (3ba): Compound 3ba was prepared following
Method A from potassium 2-nitrobenzoate (1b) (205 mg, 1.00 mmol) and
4-methylphenyl trifluoromethanesulfonate (2a) (480 mg, 2.00 mmol),
yielding 3ba after 1 h reaction time as a yellow oil (193 mg, 91%). The
spectroscopic data (NMR, GC-MS) matched those reported in the litera-
ture for 4’-methyl-2-nitrobiphenyl [CAS: 70680-21-6]. Using the same
amounts, compound 3ba was also prepared following Method C in 84%
yield (178 mg).
2-(2-Nitrophenyl)naphthalene (3bb): Compound 3bb was prepared fol-
Method B (non-activated carboxylates in Table 4): Method B is analo-
gous to Method A, but with a higher loading of a modified catalyst and a
reaction time of 24 h. The following amounts were used: potassium car-
boxylate (1a, 1k–u) (1.00 mmol), copper(I) oxide (10.7 mg, 0.075 mmol),
palladium(II) iodide (10.8 mg, 0.03 mmol), 1,10-phenanthroline (27.0 mg,
0.15 mmol), Tol-BINAP (30.5 mg, 0.045 mmol), 4-tolyl triflate (2a)
(480 mg, 2.00 mmol), n-tetradecane (50 mL), and NMP (4 mL).
lowing Method
1.00 mmol) and 2-naphthyl trifluoromethanesulfonate (2b) (552 mg,
2.00 mmol), yielding 3bb after 1 h reaction time as yellow solid
(243 mg, 98%). M.p. 98–1008C. The spectroscopic data (NMR, GC-MS)
matched those reported in the literature for 2-(2-nitrophenyl)naphtha-
lene [CAS: 94064-83-2]. Using the same amounts, compound 3bb was
also prepared following Method C in 99% yield (245 mg).
A from potassium 2-nitrobenzoate (1b) (205 mg,
a
General procedure for the biaryl synthesis under microwave-assisted con-
ditions
4’-Methoxy-2-nitrobiphenyl (3bc): Compound 3bc was prepared follow-
ing Method A from potassium 2-nitrobenzoate (1b) (205 mg, 1.00 mmol)
Method C (activated carboxylates in Table 3): An oven-dried, argon
flushed 10 mL microwave vessel was charged with potassium 2-nitroben-
zoate (1b) (1.00 mmol), copper(I) oxide (1.10 mg, 0.015 mmol), palladiu-
m(II) acetylacetonate (6.10 mg, 0.02 mmol), 1,10-phenanthroline
(5.40 mg, 0.03 mmol) and Tol-BINAP (20.4 mg, 0.03 mmol). A degassed
solution of the aryl trifluoromethanesulfonate (2a–h) (2.00 mmol) and
the internal standard n-tetradecane (50 mL) in NMP (1 mL) was added
via syringe. The resulting solution was then stirred for 5 min at 508C fol-
lowed by microwave irradiation at 1908C for 5 min at a maximum power
of 150 W and cooled afterwards with air jet cooling. The maximal pres-
sure noted was 4.5 bar. The mixture was then diluted with aqueous HCl
(1n, 10 mL) and extracted with ethyl acetate (3ꢂ20.0 mL). The com-
bined organic layers were washed with water and brine, dried over
MgSO4, filtered, and the volatiles were removed in vacuo. The residue
was purified by column chromatography (SiO2, ethyl acetate/hexane gra-
dient), yielding the corresponding biaryl.
and
4-methoxyphenyl
trifluoromethanesulfonate
(2c)
(512 mg,
2.00 mmol), yielding 3bc after 1 h reaction time as
a
yellow solid
(189 mg, 83%). M.p. 60–618C. The spectroscopic data (NMR, GC-MS)
matched those reported in the literature for 4’-methoxy-2-nitrobiphenyl
[CAS: 20013-55-2]. Using the same amounts, compound 3bc was also
prepared following Method C in 87% yield (198 mg).
2’-Methyl-2-nitrobiphenyl (3bd): Compound 3bd was prepared following
Method A from potassium 2-nitrobenzoate (1b) (205 mg, 1.00 mmol) and
2-methylphenyl trifluoromethanesulfonate (2d) (480 mg, 2.00 mmol),
yielding 3bd after 1 h reaction time as a yellow oil (166 mg, 79%). The
spectroscopic data (NMR, GC-MS) matched those reported in the litera-
ture for 2’-methyl-2-nitrobiphenyl [CAS: 67992-12-5]. Using the same
amounts, compound 3bd was also prepared following Method C in 93%
yield (198 mg).
1-(2’-Nitrobiphenyl-4-yl)-propan-1-one (3be): Compound 3be was pre-
pared following Method
A from potassium 2-nitrobenzoate (1b)
Method C’ (Table 4): Method C’ is based on Method C, but with a higher
loading of the copper(I)/phenanthroline catalyst. The following amounts
were used: potassium carboxylate (1c–j) (1.00 mmol), copper(I) oxide
(205 mg, 1.00 mmol) and 4-propionylphenyl trifluoromethanesulfonate
(2e) (564 mg, 2.00 mmol), yielding 3be after 1 h reaction time as a
yellow solid (116 mg, 45%). M.p. 83–848C; 1H NMR (600 MHz, CDCl3):
d=8.02 (d, J=8.4 Hz, 2H), 7.92 (dd, J=8.1, 0.9 Hz, 1H), 7.65 (td, J=
7.6, 1.0 Hz, 1H), 7.53 (td, J=7.8, 1.3 Hz, 1H), 7.43 (dd, J=7.7, 1.0 Hz,
1H), 7.40 (d, J=8.4 Hz, 2H), 3.03 ppm (q, J=7.2 Hz, 2H), 1.24 ppm (t,
J=7.2 Hz, 3H); 13C NMR (151 MHz, CDCl3): d=200.1, 148.9, 142.0,
136.4, 135.5, 132.6, 131.7, 128.8, 128.3, 128.2, 124.4, 31.9, 8.2 ppm; elemen-
tal analysis calcd (%) for C15H13NO3: C 70.6, H 5.13, N 5.49; found: C
70.7, H 5.3, N, 5.49; MS (ion trap, EI, 70 eV): m/z (%): 255 (3) [M+], 227
(19), 226 (100), 180 (11), 152 (12); using the same amounts, compound
3be was also prepared following Method C in 54% yield (139 mg).
(3.6 mg,
0.025 mmol),
palladium(II)
acetylacetonate
(6.10 mg,
0.02 mmol), 1,10-phenanthroline (9.0 mg, 0.05 mmol), Tol-BINAP
(20.4 mg, 0.03 mmol), 4-tolyl triflate (2a) (480 mg, 2.00 mmol), n-tetrade-
cane (50 mL), and NMP (1 mL).
Method D (non-activated carboxylates in Tables 4 and 5): Method D is
based on Method C, but using higher loading of copper catalyst, a higher
volume of NMP and a longer reaction time. The following amounts were
used: potassium salt of the carboxylic acid (1a, 1k–u) (0.50 mmol), cop-
per(I) oxide (5.37 mg, 0.0375 mmol), palladium(II) bromide (2.66 mg,
0.01 mmol), 1,10-phenanthroline (13.5 mg, 0.075 mmol), Tol-BINAP
(10.2 mg, 0.015 mmol) and a solution of the trifluoromethanesulfonate
(2a, d, g, i, m–p) (1.00 mmol) and the internal standard n-tetradecane
(50 mL) in NMP (3 mL). The resulting solution was then stirred for 5 min
at 508C followed by microwave irradiation at 1908C for 10 min at a maxi-
mum power of 150 W and cooled afterwards with air jet cooling. The iso-
lated yield was determined by combining two identical 0.5 mmol-scale re-
actions. The resulting mixture was then diluted with aqueous HCl (1n,
10 mL) and extracted with ethyl acetate (3ꢂ20.0 mL). The combined or-
ganic layers were washed with water and brine, dried over MgSO4, fil-
tered, and the volatiles were removed in vacuo. The residue was purified
by column chromatography (SiO2, ethyl acetate/hexane gradient), yield-
ing the corresponding biaryl.
4-Acyl-2’-nitrobiphenyl (3bf): Compound 3bf was prepared following
Method A from potassium 2-nitrobenzoate (1b) (206 mg, 1.00 mmol) and
4-acylphenyl trifluoromethanesulfonate (2 f) (536 mg, 2.00 mmol) yielding
3bf after 1 h reaction time as a yellow solid (72 mg, 30%). M.p. 100–
1028C. The spectroscopic data (NMR, GC-MS) matched those reported
in the literature for 4-acyl-2’-nitrobiphenyl [CAS: 5730-96-1]. Using the
same amounts, compound 3bf was also prepared following Method C in
23% yield (56 mg).
Ethyl 2’-nitrobiphenyl-2-carboxylate (3bg): Compound 3bg was prepared
following Method
A from potassium 2-nitrobenzoate (1b) (205 mg,
1.00 mmol) and ethyl 2-(trifluoromethylsulfonyloxy)benzoate (2g)
(596 mg, 2.00 mmol), yielding 3bg after 1 h reaction time as a yellow oil
(100 mg, 37%). The spectroscopic data (NMR, GC-MS) matched those
reported in the literature for ethyl 2’-nitrobiphenyl-2-carboxylate [CAS:
72256-33-8]. Using the same amounts, compound 3bg was also prepared
following Method C in 23% yield (63 mg).
4’-Methyl-3-nitrobiphenyl (3aa): Compound 3aa was prepared following
Method B from potassium 3-nitrobenzoate (1a) (205 mg, 1.00 mmol) and
4-methylphenyl trifluoromethanesulfonate (2a) (480 mg, 2.00 mmol) in
NMP (8 mL), yielding 3aa as a pale yellow solid (153 mg, 72%). M.p.
65–678C. The spectroscopic data (NMR, GC-MS) matched those report-
ed in the literature for 4’-methyl-3-nitrobiphenyl [CAS: 53812-68-3].
4-Nitro-2’-nitrobiphenyl (3bh): Compound 3bh was prepared following
Method C from potassium 2-nitrobenzoate (1b) (206 mg, 1.00 mmol) and
Chem. Eur. J. 2009, 15, 9336 – 9349
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
9345

L. J. Goossen et al.
4-nitrophenyl trifluoromethanesulfonate (2h) (546 mg, 2.00 mmol) yield-
ing 3bh as an orange solid (76 mg, 31%). M.p. 87–898C. The spectro-
scopic data (NMR, GC-MS) matched those reported in the literature for
4-nitro-2’-nitrobiphenyl [CAS: 606-81-5].
7.44 ppm (dd, J=7.6, 1.4 Hz, 1H); 13C NMR (151 MHz, CDCl3) d=
191.7, 148.7, 138.6, 136.6, 135.0, 133.7, 132.7, 131.9, 129.4, 129.2, 128.9,
128.9, 124.4 ppm; elemental analysis calcd (%) for C13H9NO3: C 68.7, H
4.0, N 6.2; found: C 68.3, H 3.7, N 6.1; MS (ion trap, EI, 70 eV): m/z
(%): 226 (13) [M+], 210 (55), 199 (85), 182 (100), 154 (100), 152 (90), 115
(100).
3’,5’-Dimethyl-2-nitrobiphenyl (3bi): Compound 3bi was prepared fol-
lowing Method
A from potassium 2-nitrobenzoate (1b) (205 mg,
1.00 mmol) and 3,5-dimethylphenyl trifluoromethanesulfonate (2i)
(508 mg, 2.00 mmol), yielding 3bi after 1 h reaction time as a yellow oil
(225 mg, 99%). The spectroscopic data (NMR, GC-MS) matched those
reported in the literature for 3’,5’-dimethyl-2-nitrobiphenyl [CAS: 51839-
09-9].
2-Methyl-8-(2’-nitrophenyl)quinoline (3br): Compound 3br was prepared
following Method
1.00 mmol) and 8-quinolinyl trifluoromethanesulfonate (2r) (554 mg,
2.00 mmol), yielding 3br after 1 h reaction time as yellow solid
A from potassium 2-nitrobenzoate (1b) (205 mg,
a
(200 mg, 80%). M.p. 129–1318C. The spectroscopic data (NMR, GC-MS)
matched those reported in the literature for 8-(2’-nitrophenyl)quinoline
[CAS: 108530-08-1].
2-Nitrobiphenyl (3bj): Compound 3bj was prepared following Method A
from potassium 2-nitrobenzoate (1b) (205 mg, 1.00 mmol) and phenyl tri-
fluoromethanesulfonate (2j) (452 mg, 2.00 mmol), yielding 3bj after 1 h
reaction time as a yellow oil (181 mg, 91%). The spectroscopic data
(NMR, GC-MS) matched those reported in the literature for 2-nitrobi-
phenyl [CAS: 86-00-0].
2-Fluoro-4’-methylbiphenyl (3ca): Compound 3ca was prepared follow-
ing Method
A
from potassium 2-fluorobenzoate (1c) (178 mg,
1.00 mmol) and 4-methylphenyl trifluoromethanesulfonate (2a) (480 mg,
2.00 mmol), yielding 3ca after 16 h reaction time as a colorless oil
(141 mg, 76%). The spectroscopic data (NMR, GC-MS) matched those
reported in the literature for 2-fluoro-4’-methylbiphenyl [CAS: 72093-41-
5]. Compound 3ca was also prepared following Method C’, only using
1.00 mmol of 4-methylphenyl trifluoromethanesulfonate (2a) (240 mg), in
73% yield (135 mg).
4-Chloro-2’-nitrobiphenyl (3bk): Compound 3bk was prepared following
Method A from potassium 2-nitrobenzoate (1b) (205 mg, 1.00 mmol) and
4-chlorophenyl trifluoromethanesulfonate (2k) (521 mg, 2.00 mmol),
yielding 3bk after 1 h reaction time as a yellow solid (211.5 mg, 91%).
M.p. 60–628C. The spectroscopic data (NMR, GC-MS) matched those re-
ported in the literature for 4-chloro-2’-nitrobiphenyl [CAS: 6271-80-3].
4’,5-Dimethyl-2-nitrobiphenyl (3da): Compound 3da was prepared fol-
lowing Method
A from potassium 5-methyl-2-nitrobenzoate (1d)
8-(2’-Nitrophenyl)quinoline (3bl): Compound 3bl was prepared following
Method A from potassium 2-nitrobenzoate (1b) (205 mg, 1.00 mmol) and
8-quinolinyl trifluoromethanesulfonate (2l) (554 mg, 2.00 mmol), yielding
3bl after 1 h reaction time as a yellow solid (200 mg, 80%). M.p. 129–
1318C. The spectroscopic data (NMR, GC-MS) matched those reported
in the literature for 8-(2’-nitrophenyl)quinoline [CAS: 108530-08-1].
(219 mg, 1.00 mmol) and 4-methylphenyl trifluoromethanesulfonate (2a)
(480 mg, 2.00 mmol), yielding 3da after 16 h reaction time as a pale
yellow solid (164 mg, 72%). M.p. 60–628C. The spectroscopic data
(NMR, GC-MS) matched those reported in the literature for 4’,5-dimeth-
yl-2-nitrobiphenyl [CAS: 70689-98-4]. Using the same amounts, com-
pound 3da was also prepared following Method C’ in 73% yield
(166 mg).
Ethyl 2’-nitrobiphenyl-3-carboxylate (3bm): Compound 3bm was pre-
pared following Method
A from potassium 2-nitrobenzoate (1b)
Isopropyl 4’-methylbiphenyl-2-carboxylate (3ea): Compound 3ea was
prepared following Method A from potassium 2-(isopropoxycarbonyl)-
benzoate (1e) (246 mg, 1.00 mmol) and 4-methylphenyl trifluorometha-
nesulfonate (2a) (480 mg, 2.00 mmol), yielding 3ea after 16 h reaction
time as a white solid (76 mg, 30%). M.p. 130–1318C. The spectroscopic
data (NMR, GC-MS) matched those reported in the literature for isopro-
pyl 4’-methylbiphenyl-2-carboxylate [CAS: 937166-54-6]. Using the same
amounts, compound 3ea was also prepared following Method C’ in 58%
yield (147 mg).
(205 mg, 1.00 mmol) and ethyl 3-(trifluoromethylsulfonyloxy)benzoate
(2m) (596 mg, 2.00 mmol), yielding 3bm after 1 h reaction time as a
yellow oil (214 mg, 79%). The spectroscopic data (NMR, GC-MS)
matched those reported in the literature for ethyl 2’-nitrobiphenyl-3-car-
boxylate [CAS: 236102-71-9].
3’-Acetyl-2-nitrobiphenyl (3bn): Compound 3bn was prepared following
Method A from potassium 2-nitrobenzoate (1b) (205 mg, 1.00 mmol) and
3-acetylphenyl trifluoromethanesulfonate (2n) (536 mg, 2.00 mmol),
yielding 3bn after 1 h reaction time as a yellow solid (183 mg, 76%).
M.p. 95–988C; 1H NMR (600 MHz, CDCl3): d=7.95 (dt, J=7.1, 1.8 Hz,
1H), 7.89–7.90 (m, 1H), 7.88 (dd, J=8.1, 0.9 Hz, 1H), 7.61 (td, J=7.6,
1.0 Hz, 1H), 7.44–7.51 (m, 3H), 7.41 (dd, J=7.7, 1.3 Hz, 1H), 2.58 ppm
(s, 3H); 13C NMR (151 MHz, CDCl3) d=197.5, 148.7, 138.0, 137.2, 135.3,
132.5, 132.4, 131.8, 128.8, 128.6, 128.0, 127.6, 124.2, 26.5 ppm; elemental
analysis calcd (%) for C14H11NO3: C 69.7, H 4.6, N 5.8; found: C 69.5, H
4.5, N 5.5; MS (ion trap, EI, 70 eV): m/z (%): 241 (15) [M+], 226 (100),
199 (12), 182 (16), 153 (15), 152 (13), 115 (12).
4’-Methylbiphenyl-2-carbaldehyde (3 fa): Compound 3 fa was prepared
following Method A from potassium 2-formylbenzoate (1 f) (188 mg,
1.00 mmol) and 4-methylphenyl trifluoromethanesulfonate (2a) (480 mg,
2.00 mmol), yielding 3 fa after 16 h reaction time as a yellow oil (88.3 mg,
45%). The spectroscopic data (NMR, GC-MS) matched those reported
in the literature for 4’-methylbiphenyl-2-carbaldehyde [CAS: 16191-28-
9]. Using the same amounts, compound 3 fa was also prepared following
Method C’ in 54% yield (106 mg).
4’-Methylbiphenyl-2-carbonitrile (3ga): Compound 3ga was prepared fol-
4’-Fluoro-2-nitrobiphenyl (3bo): Compound 3bo was prepared following
Method A from potassium 2-nitrobenzoate (1b) (205 mg, 1.00 mmol) and
4-fluorophenyl trifluoromethanesulfonate (2o) (488 mg, 2.00 mmol),
yielding 3bo after 1 h reaction time as a yellow oil (163 mg, 75%). The
spectroscopic data (NMR, GC-MS) matched those reported in the litera-
ture for 4’-fluoro-2-nitrobiphenyl [CAS: 390-38-5].
lowing Method
1.00 mmol) and 4-methylphenyl trifluoromethanesulfonate (2a) (480 mg,
2.00 mmol), yielding 3ga after 16 h reaction time as white solid
A from potassium 2-cyanobenzoate (1g) (185 mg,
a
(85.4 mg, 44%). M.p. 43–488C. The spectroscopic data (NMR, GC-MS)
matched those reported in the literature for 4’-methylbiphenyl-2-carboni-
trile [CAS: 114772-53-1]. Using the same amounts, compound 3ga was
also prepared following Method C’ in 50% yield (97.0 mg).
3’-Methoxy-2-nitrobiphenyl (3bp): Compound 3bp was prepared follow-
ing Method A from potassium 2-nitrobenzoate (1b) (205 mg, 1.00 mmol)
2-Methoxy-4’-methylbiphenyl (3ha): Compound 3ha was prepared fol-
lowing Method A from potassium 2-methoxybenzoate (1h) (190 mg,
1.00 mmol) and 4-methylphenyl trifluoromethanesulfonate (2a) (480 mg,
2.00 mmol), yielding 3ha after 16 h reaction time as a pale yellow solid
(80.5 mg, 40%). M.p. 81–838C. The spectroscopic data (NMR, GC-MS)
matched those reported in the literature for 2-methoxy-4’-methylbiphenyl
[CAS: 92495-53-9]. Using the same amounts, compound 3ha was also
prepared following Method C’ in 40% yield (80.5 mg).
and
3-methoxyphenyl
trifluoromethanesulfonate
(2p)
(512 mg,
2.00 mmol), yielding 3bp after 1 h reaction time as a yellow oil (170 mg,
74%). The spectroscopic data (NMR, GC-MS) matched those reported
in the literature for 3’-methoxy-2-nitrobiphenyl [CAS: 92017-95-3].
2’-Nitrobiphenyl-3-carbaldehyde (3bq): Compound 3bq was prepared
following Method
1.00 mmol) and 3-formylphenyl trifluoromethanesulfonate (2q) (508 mg,
A from potassium 2-nitrobenzoate (1b) (205 mg,
2.00 mmol), yielding 3bq after 1 h reaction time as
a colorless oil
(145 mg, 64%). 1H NMR (600 MHz, CDCl3): d=10.03 (s, 1H), 7.94 (dd,
J=8.2, 1.2 Hz, 1H), 7.91 (dt, J=7.2, 1.6 Hz, 1H), 7.83 (t, J=1.7 Hz, 1H),
7.66 (td, J=7.6, 1.3 Hz, 1H), 7.59 (t, J=7.4 Hz, 1H), 7.52–7.57 (m, 2H),
2-(4-Methylphenyl)thiophene (3ia): Compound 3ia was prepared follow-
ing Method A from potassium thiophene-2-carboxylate (1i) (166 mg,
1.00 mmol) and 4-methylphenyl trifluoromethanesulfonate (2a) (480 mg,
9346
www.chemeurj.org
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 9336 – 9349

Pd-Catalyzed Decarboxylative Coupling
FULL PAPER
2.00 mmol), yielding 3ia after 16 h reaction time as a white solid (131 mg,
75%). M.p. 60–628C. The spectroscopic data (NMR, GC-MS) matched
those reported in the literature for 2-(4-methylphenyl)thiophene [CAS:
16939-04-1]. Compound 3ia was also prepared following Method C’, only
using 1.00 mmol of 4-methylphenyl trifluoromethanesulfonate (2a)
(240 mg), in 82% yield (143 mg).
N-(4’-Methylbiphenyl-4-yl)acetamide (3qa): Compound 3qa was pre-
pared following Method B from potassium 4-acetamidobenzoate (1q)
(217 mg, 1.00 mmol) and 4-methylphenyl trifluoromethanesulfonate (2a)
(480 mg, 2.00 mmol), yielding 3qa as a white solid (119 mg, 53%). M.p.
165–1688C. The spectroscopic data (NMR, GC-MS) matched those re-
ported in the literature for N-(4’-methylbiphenyl-4-yl)acetamide [CAS:
1215-21-0]. Compound 3qa was also prepared following Method D in
59% yield (132 mg).
2-(4-Methylphenyl)furan (3ja): Compound 3ja was prepared following
Method A from potassium furan-2-carboxylate (1j) (150 mg, 1.00 mmol)
and 4-methylphenyl trifluoromethanesulfonate (2a) (480 mg, 2.00 mmol),
yielding 3ja after 16 h reaction time as a yellow oil (119 mg, 75%). The
spectroscopic data (NMR, GC-MS) matched those reported in the litera-
ture for 2-(4-methylphenyl)furan [CAS: 17113-32-5]. Using the same
amounts, compound 3ja was also prepared following Method C’, only
using 1.00 mmol of 4-methylphenyl trifluoromethanesulfonate (2a)
(240 mg), in 75% yield (119 mg).
3-Chloro-4’-methylbiphenyl (3ra): Compound 3ra was prepared follow-
ing Method
B
from potassium 3-chlorobenzoate (1r) (228 mg,
1.00 mmol) and 4-methylphenyl trifluoromethanesulfonate (2a) (480 mg,
2.00 mmol), yielding 3ra as colorless oil (80.0 mg, 40%). The spectro-
scopic data (NMR, GC-MS) matched those reported in the literature for
3-chloro-4’-methylbiphenyl [CAS: 19482-19-0]. Compound 3ra was also
prepared following Method D in 59% yield (118 mg).
3-Methoxy-4’-methylbiphenyl (3sa): Compound 3sa was prepared follow-
4-Methyl-4’-nitrobiphenyl (3ka): Compound 3ka was prepared following
Method B from potassium 4-nitrobenzoate (1k) (205 mg, 1.00 mmol) and
4-methylphenyl trifluoromethanesulfonate (2a) (480 mg, 2.00 mmol),
yielding 3ka as a white solid (146 mg, 68%). M.p. 116–1178C. The spec-
troscopic data (NMR, GC-MS) matched those reported in the literature
for 4-methyl-4’-nitrobiphenyl [CAS: 2143-88-6]. Using the same amounts,
compound 3ka was also prepared following Method D in 81% yield
(174 mg).
ing Method
D from potassium 3-methoxybenzoate (1s) (95 mg,
0.50 mmol) and 4-methylphenyl trifluoromethanesulfonate (2a) (240 mg,
1.00 mmol). After combining two identical 0.5 mmol scale-reactions com-
pound 3sa was isolated as a pale yellow solid (107 mg, 54%). M.p. 74–
768C. The spectroscopic data (NMR, GC-MS) matched those reported in
the literature for 3-methoxy-4’-methylbiphenyl [CAS: 24423-07-2].
3-(4-Methylphenyl)pyridine (3ta): Compound 3ta was prepared following
Method B from potassium nicotinate (1t) (161 mg, 1.00 mmol) and 4-
methylphenyl trifluoromethanesulfonate (2a) (480 mg, 2.00 mmol), yield-
ing 3ta as a colorless oil (69.1 mg, 41%). The spectroscopic data (NMR,
GC-MS) matched those reported in the literature for 3-(4-methylphenyl)-
pyridine [CAS: 4423-09-0]. Compound 3ta was also prepared following
Method D in 50% yield (84 mg).
4’-Methylbiphenyl-3-carbonitrile (3la): Compound 3la was prepared fol-
lowing Method
B from potassium 3-cyanobenzoate (1l) (185 mg,
1.00 mmol) and 4-methylphenyl trifluoromethanesulfonate (2a) (480 mg,
2.00 mmol), yielding 3la as a pale yellow solid (100 mg, 52%). M.p. 50–
518C. The spectroscopic data (NMR, GC-MS) matched those reported in
the literature for 4-methylbiphenyl-3’-carbonitrile [CAS: 133909-96-3].
Compound 3la was also prepared following Method D in 83% yield
(160 mg).
3-(4-Methylphenyl)thiophene (3ua): Compound 3ua was prepared fol-
lowing Method B from potassium thiophene-3-carboxylate (1u) (166 mg,
1.00 mmol) and 4-methylphenyl trifluoromethanesulfonate (2a) (480 mg,
2.00 mmol), yielding 3ua as a white solid (94 mg, 54%). M.p. 70–728C.
The spectroscopic data (NMR, GC-MS) matched those reported in the
literature for 3-(4-methylphenyl)thiophene [CAS: 16939-05-2]. Com-
4’-Methylbiphenyl-4-carbonitrile (3ma): Compound 3ma was prepared
following Method B from potassium 4-cyanobenzoate (1m) (185 mg,
1.00 mmol) and 4-methylphenyl trifluoromethanesulfonate (2a) (480 mg,
2.00 mmol), yielding 3ma as a pale yellow solid (112 mg, 58%). M.p. 89–
918C. The spectroscopic data (NMR, GC-MS) matched those reported in
the literature for 4-methylbiphenyl-4’-carbonitrile [CAS: 50670-50-3].
Compound 3ma was also prepared following Method D in 76% yield
(147 mg).
pound 3ua was also prepared following Method
D in 65% yield
(113 mg).
2-(3-Nitrophenyl)naphthalene (3ab): Compound 3ab was prepared fol-
lowing Method from potassium 3-nitrobenzoate (1a) (205 mg,
B
0.50 mmol) and 2-naphthyl trifluoromethanesulfonate (2b) (552 mg,
2.00 mmol) in NMP (8 mL), yielding 3ab as a yellow solid (154 mg,
62%). M.p. 110–1128C. The spectroscopic data (NMR, GC-MS) matched
those reported in the literature for 2-(3-nitrophenyl)-naphthalene [CAS:
94064-82-1]. Compound 3ab was also prepared in 74% yield (184 mg)
using the same amounts of starting materials via two 0.5 mmol batches
following Method D.
4’-Methyl-4-(trifluoromethyl)biphenyl (3na): Compound 3na was pre-
pared following Method B from potassium 4-(trifluoromethyl)benzoate
(1n) (228 mg, 1.00 mmol) and 4-methylphenyl trifluoromethanesulfonate
(2a) (480 mg, 2.00 mmol), yielding 3na as a white solid (80 mg, 44%).
M.p. 122–1248C. The spectroscopic data (NMR, GC-MS) matched those
reported in the literature for 4’-methyl-4-(trifluoromethyl)biphenyl
[CAS: 97067-18-0]. Compound 3na was also prepared following Method
D in 74% yield (135 mg).
2’-Methyl-3-nitrobiphenyl (3ad): Compound 3ad was prepared following
Method B from potassium 3-nitrobenzoate (1a) (205 mg, 1.00 mmol) and
2-methylphenyl trifluoromethanesulfonate (2d) (480 mg, 2.00 mmol) in
NMP (8 mL), yielding 3ad as an orange oil (87 mg, 41%). The spectro-
scopic data (NMR, GC-MS) matched those reported in the literature for
2’-methyl-3-nitrobiphenyl [CAS: 51264-60-9]. Compound 3ad was also
prepared in 61% yield (129 mg) using the same amounts of starting ma-
terials via two 0.5 mmol batches following Method D.
1-(4-Methylphenyl)naphthalene (3oa): Compound 3oa was prepared fol-
lowing Method
B
from potassium 1-naphthoate (1o) (210 mg,
1.00 mmol) and 4-methylphenyl trifluoromethanesulfonate (2a) (480 mg,
2.00 mmol), yielding 3oa as a colorless oil (107 mg, 49%). The spectro-
scopic data (NMR, GC-MS) matched those reported in the literature for
1-(4-methylphenyl)naphthalene [CAS: 24423-07-2]. Compound 3oa was
also prepared following Method D in 71% yield (155 mg).
Ethyl 3’-nitrobiphenyl-2-carboxylate (3ag): Compound 3ag was prepared
following Method
D from potassium 3-nitrobenzoate (1a) (103 mg,
3,4’-Dimethyl-4-nitrobiphenyl (3pa): Compound 3pa was prepared fol-
0.50 mmol) and ethyl 2-(trifluoromethylsulfonyloxy)benzoate (2g)
lowing Method
B from potassium 3-methyl-4-nitrobenzoate (1p)
(298 mg, 1.00 mmol). After combining two identical 0.5 mmol-scale reac-
1
(197 mg, 1.00 mmol) and 4-methylphenyl trifluoromethanesulfonate (2a)
(480 mg, 2.00 mmol), yielding 3pa as a yellow solid (142 mg, 62%. M.p.
56–578C; 1H NMR (600 MHz, CDCl3): d=8.05–8.08 (m, 1H), 7.50–7.52
(m, 2H), 7.49–7.50 (m, 2H), 7.28 (d, J=7.9 Hz, 2H), 2.67 (s, 3H),
2.41 ppm (s, 3H); 13C NMR (151 MHz, CDCl3): d=147.6, 145.9, 138.7,
135.8, 134.3, 131.0, 129.7, 127.1, 125.4, 125.1, 21.1, 21.0 ppm; elemental
analysis calcd (%) for C14H13NO2: C 74.0, H 5.8, N 6.16; found: C 73.8,
H 5.8, N 6.00; MS (ion trap, EI, 70 eV): m/z (%): 227 (100) [M+], 210
(92), 182 (35), 167 (38), 166 (22), 165 (49), 155 (25). Compound 3pa was
also prepared following Method D in 69% yield (158 mg).
tions, compound 3ag was isolated as colorless oil (92 mg, 34%). H NMR
(400 MHz, CDCl3): d=8.21 (ddd, J=8.0, 2.2, 1.0 Hz, 1H), 8.18 (t, J=
1.9 Hz, 1H), 7.96 (dd, J=7.8, 1.4 Hz, 1H), 7.62 (ddd, J=7.7, 1.4, 1.2 Hz,
1H), 7.53–7.60 (m, 2H), 7.49 (td, J=7.6, 1.2 Hz, 1H), 7.34 (dd, J=7.5,
1.0 Hz, 1H), 4.12 (q, J=7.2 Hz, 2H), 1.07 ppm (t, J=7.2 Hz, 3H);
13C NMR (101 MHz, CDCl3): d=167.4, 143.4, 140.4, 134.6, 131.7, 130.7,
130.5, 129.6, 128.7, 128.3, 123.5, 122.0, 121.4, 61.1, 13.8 ppm; elemental
analysis calcd (%) for C14H13NO2: C 66.4, H 4.8, N 5.2; found: C 66.6, H
4.7, N 5.0. MS (Ion trap, EI, 70 eV): m/z (%): 271 (45) [M+], 226 (100),
225 (25), 210 (28), 196 (32), 180 (31), 151 (27).
Chem. Eur. J. 2009, 15, 9336 – 9349
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
9347

L. J. Goossen et al.
3’,5’-Dimethyl-3-nitrobiphenyl (3ai): Compound 3ai was prepared follow-
ing Method D from potassium 3-nitrobenzoate (1a) (103 mg, 0.50 mmol)
and 3,5-dimethylphenyl trifluoromethanesulfonate (2i) (254 mg,
1.00 mmol). After combining two identical 0.5 mmol-scale reactions, com-
pound 3ai was isolated as a yellow solid (157 mg, 69%). M.p. 65–668C.
The spectroscopic data (NMR, GC-MS) matched those reported in the
literature for 3’,5’-dimethyl-3-nitrobiphenyl [CAS: 337973-04-3].
or 0.03 mmol), see Table 6. Subsequently, a solution of n-tetradecane
(50 mL) in NMP (1 mL) was added via syringe. The resulting mixture was
then stirred at 1908C for 15 min at a maximum power of 150 W and
cooled afterwards with air jet cooling. The pressure noted at this temper-
ature was 4.5 bar. Then, the reaction mixture was diluted with ethyl ace-
tate (2 mL). An aliquot of the reaction mixture (0.25 mL) was dissolved
in ethyl acetate (2 mL), washed with HCl (5n, 2 mL), dried over MgSO4/
NaHCO3 and analyzed by GC.
Ethyl 3’-nitrobiphenyl-3-carboxylate (3an): Compound 3an was prepared
following Method D from potassium 3-nitrobenzoate (1a) (102.5 mg,
0.50 mmol) and ethyl 3-(trifluoromethylsulfonyloxy)benzoate (2i)
(298 mg, 1.00 mmol). After combining two identical 0.5 mmol-scale reac-
tions, compound 3an was isolated as yellow oil (108 mg, 40%). The spec-
troscopic data (NMR, GC-MS) matched those reported in the literature
for ethyl 3’-nitrobiphenyl-3-carboxylate [CAS: 942232-55-5].
Acknowledgements
We thank the Deutsche Forschungsgemeinschaft, Saltigo GmbH, and
NanoKat for funding, Umicore AG for the generous donation of catalysts
and the Alexander von Humboldt Foundation for a scholarship to N.R.
3’-Acetyl-3-nitrobiphenyl (3am): Compound 3am was prepared following
Method B from potassium 3-nitrobenzoate (1a) (205 mg, 1.00 mmol) and
3-acetylphenyl trifluoromethanesulfonate (2m) (536 mg, 2.00 mmol) in
NMP (8 mL), yielding 3am as a white solid (135 mg, 58%). M.p. 85–
868C. The spectroscopic data (NMR, GC-MS) matched those reported in
the literature for 3’-acetyl-3-nitrobiphenyl [CAS: 371157-19-6]. Com-
[1] Metal-Catalyzed Cross-Coupling Reactions, 2nd ed. (Eds.: A. de
Meijere, F. Diederich), Wiley-VCH, Weinheim, 2004.
pound 3ad was also prepared following Method
D in 49% yield
[2] a) N. Miyaura, Top. Curr. Chem. 2002, 219, 11–59; b) N. Miyaura,
A. Suzuki, Chem. Rev. 1995, 95, 2457–2483.
(114 mg).
4’-Fluoro-3-nitrobiphenyl (3ao): Compound 3ao was prepared following
Method D from potassium 3-nitrobenzoate (1a) (103 mg, 0.50 mmol) and
4-fluorophenyl trifluoromethanesulfonate (2o) (244 mg, 1.00 mmol).
After combining two identical 0.5 mmol-scale reactions, compound 3ao
was isolated as a yellow solid (139 mg, 64%). M.p. 81–848C. The spectro-
scopic data (NMR, GC-MS) matched those reported in the literature for
4’-fluoro-3-nitrobiphenyl [CAS: 10540-32-6].
[3] a) M. Kosugi, K. Sasazawa, Y. Shimizu, T. Migita, Chem. Lett. 1977,
301–302; b) J. Stille, Angew. Chem. 1986, 98, 504–519; Angew.
Chem. Int. Ed. Engl. 1986, 25, 508–524; c) D. Milstein, J. Stille, J.
Am. Chem. Soc. 1978, 100, 3636–3638; d) A. F. Littke, L. Schwartz,
G. C. Fu, J. Am. Chem. Soc. 2002, 124, 6343–6348.
[4] a) E. I. Negishi, Acc. Chem. Res. 1982, 15, 340–348; b) P. Knochel,
R. Singer, Chem. Rev. 1993, 93, 2117–2188.
[5] B. H. Lipshutz, K. Siegmann, E. Garcia, F. Kayser, J. Am. Chem.
Soc. 1993, 115, 9276–9282.
[6] a) T. Hiyama, Y. Hatanaka, Pure Appl. Chem. 1994, 66, 1471;
b) S. E. Denmark, R. F. Sweis, Acc. Chem. Res. 2002, 35, 835–846.
[7] a) K. Tamao, K. Sumitani, M. Kumada, J. Am. Chem. Soc. 1972, 94,
4374–4376; b) M. Kumada, Pure Appl. Chem. 1980, 52, 669–679;
c) J. A. Miller, Tetrahedron Lett. 2001, 42, 6991–6993; d) R. J. P.
Corriu, J. P. Masse, J. Chem. Soc. Chem. Commun. 1972, 144a;
e) A. H. Roy, J. F. Hartwig, J. Am. Chem. Soc. 2003, 125, 8704–
8705.
3’-Methoxy-3-nitrobiphenyl (3ap): Compound 3ap was prepared follow-
ing Method D from potassium 3-nitrobenzoate (1a) (103 mg, 0.50 mmol)
and
3-methoxyphenyl
trifluoromethanesulfonate
(2p)
(256 mg,
1.00 mmol) yielding 3ap as a yellow oil (91.6 mg, 40%). The spectroscop-
ic data (NMR, GC-MS) matched those reported in the literature for 3’-
methoxy-3-nitrobiphenyl [CAS: 128923-93-3].
4,4’-Dimethylbenzophenone (6aa) (see Scheme 4): Compound 6aa was
prepared following Method A heating 16 h, from potassium oxo-(4-meth-
ylphenyl)acetate (5a) (203 mg, 1.2 mmol) and 4-methylphenyl trifluoro-
methanesulfonate (2a) (480 mg, 2.00 mmol), yielding 6aa as
a light
yellow solid (179 mg, 85%). M.p. 67–688C. The spectroscopic data
(NMR, GC-MS) matched those reported in the literature for 4,4’-dime-
thylbenzophenone [CAS: 611-97-2]. Compound 6aa was also prepared
following Method C’ in 89% yield (187 mg).
[8] D. Alberico, M. E. Scott, M. Lautens, Chem. Rev. 2007, 107, 174–
238.
[9] I. V. Seregin, V. Gevorgyan, Chem. Soc. Rev. 2007, 36, 1173–1193.
[10] a) K. Sonogashira, Y. Tohda, N. Hagihara, Tetrahedron Lett. 1975,
16, 4467–4470; b) K. Sonogashira, in Metal-Catalyzed Cross-Cou-
pling Reactions (Eds.: F. Diederich, P. J. Stang), Wiley-VCH, Wein-
heim, 1998, Chapter 5; c) R. Chinchilla, C. Nꢄjera, Chem. Rev. 2007,
107, 874–922.
[11] a) M. Palucki, S. L. Buchwald, J. Am. Chem. Soc. 1977, 99, 11108–
11109; b) B. C. Hamann, J. F. Hartwig, J. Am. Chem. Soc. 1977, 99,
12382–12383; c) T. Satoh, Y. Kawamura, M. Miura, M. Nombra,
Angew. Chem. 1997, 109, 1820–1822; Angew. Chem. Int. Ed. Engl.
1997, 36, 1740–1742.
[12] a) V. Snieckus, Chem. Rev. 1990, 90, 879–933; b) E. J.-G. Anctil, V.
Snieckus in Metal-Catalyzed Cross-Coupling Reactions, 2nd ed.
(Eds.: A. de Meijere, F. Diederich), Wiley-VCH, Weinheim, 2004,
pp. 761–813.
[13] a) L. J. Goossen, N. Rodrꢁguez, K. Gooßen, Angew. Chem. 2008,
120, 3144–3164; Angew. Chem. Int. Ed. 2008, 47, 3100–3120;
b) L. J. Goossen, L. Winkel, A. Dçhring, K. Ghosh, J. Paetzold, Syn-
lett 2002, 1237–1240; c) L. J. Goossen, A. Dçhring, Adv. Synth.
Catal. 2003, 345, 943–947; d) L. J. Goossen, J. Paetzold, L. Winkel,
Synlett 2002, 1721–1723; e) L. J. Goossen, J. Paetzold, Angew.
Chem. 2004, 116, 1115–1118; Angew. Chem. Int. Ed. 2004, 43, 1095–
1098.
4’,2,2-Trimethylpropiophenone (6ba) (see Scheme 4): Compound 6ba
was prepared following Method A heating 16 h, from potassium 3,3,3-tri-
methylpyruvate (168 mg, 1.0 mmol) (5b) and 4-methylphenyl trifluoro-
methanesulfonate (2a) (480 mg, 2.00 mmol), yielding 6ba as a colorless
oil (136 mg, 77%). The spectroscopic data (NMR, GC-MS) matched
those reported in the literature for 4’,2,2-trimethylpropiophenone [CAS:
30314-44-4]. Compound 6ba was also prepared following Method C’ in
51% yield (88 mg).
General procedure for the decarboxylation study (Table 6)
Method A (thermal conditions): An oven-dried, nitrogen flushed vessel
was charged with the carboxylic acid (7a–c) (1.00 mmol), Cu2O (10.7 mg,
0.075 mmol) and the appropriate amount of phenanthroline (0 or
0.15 mmol), Tol-BINAP (0 or 0.045 mmol), PdI2 (0 or 0.03 mmol), Pd-
ACHTUNGTRENNUNG(dba)2 (0 or 0.03 mmol), KI (0 or 0.06 mmol), KOTf (0 or 1.00 mmol),
see Table 6. A degassed solution of n-tetradecane in NMP (2.0 mL) was
added via syringe and the resulting mixture was stirred at 1708C for 16 h.
Then, the reaction mixture was allowed to cool to room temperature and
was diluted with ethyl acetate (2 mL). An aliquot of the reaction mixture
(0.25 mL) was dissolved in ethyl acetate (2 mL), washed with HCl (5n,
2 mL), dried over MgSO4/NaHCO3 and analyzed by GC.
Method B (microwave-assisted conditions): An oven-dried, argon flushed
10 mL microwave vessel was charged with the carboxylic acid 7a–c
(1.00 mmol), Cu2O (10.7 mg, 0.075 mmol) and the appropriate amount of
phenanthroline (0 or 0.15 mmol), Tol-BINAP (0 or 0.045 mmol), PdI2 (0
[14] L. J. Goossen, G. Deng, L. M. Levy, Science 2006, 313, 662–664.
[15] a) T. Cohen, R. A. Schambach, J. Am. Chem. Soc. 1970, 92, 3189–
3190; b) T. Cohen, R. W. Berninger, J. T. Word, J. Org. Chem. 1978,
43, 837–848.
9348
www.chemeurj.org
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 9336 – 9349

Pd-Catalyzed Decarboxylative Coupling
FULL PAPER
[16] a) M. Nilsson, Acta Chem. Scand. 1966, 20, 423–426; b) M. Nilsson,
C. Ullenius, Acta Chem. Scand. 1968, 22, 1998–2002.
[17] A. G. Myers, D. Tanaka, M. R. Mannion, J. Am. Chem. Soc. 2002,
124, 11250–11251.
[27] a) K. Ritter, Synthesis 1993, 735–762; b) A. F. Littke, C. Dai, G. C.
Fu, J. Am. Chem. Soc. 2000, 122, 4020–4028.
[28] J. Louie, M. S. Driver, B. C. Hamann, J. F. Hartwig, J. Org. Chem.
1997, 62, 1268–1273.
[18] L. J. Goossen, N. Rodrꢁguez, B. Melzer, C. Linder, G. Deng, L. M.
Levy, J. Am. Chem. Soc. 2007, 129, 4824–4833.
[29] A. Zhang, J. L. Neumeyer, Org. Lett. 2003, 5, 201–203.
[30] A. B. Dounay, K. Hatanaka, J. J. Kodanko, M. Oestreich, L. E. Over-
man, L. A. Pfeifer, M. M. Weiss, J. Am. Chem. Soc. 2003, 125, 6261–
6271.
[31] a) T. Kamikawa, T. Hayashi, Tetrahedron Lett. 1997, 38, 7087; b) G.
Espino, A. Kurbangalieva, J. M. Brown, Chem. Commun. 2007,
1742–1744.
[32] A. Jutand, A. Mosleh, Organometallics 1995, 14, 1810–1817.
[33] a) P. Appukkuttan, E. Van der Eycken, Eur. J. Org. Chem. 2008,
1133–1155; b) C. O. Kappe, Angew. Chem. 2004, 116, 6408–6443;
Angew. Chem. Int. Ed. 2004, 43, 6250–6284.
[34] N. E. Leadbeater, M. Marco, Org. Lett. 2002, 4, 2973–2976.
[35] Y. Wan, M. Alterman, A. Hallberg, Synthesis 2002, 1597–1600.
[36] M. Larhed, A. Hallberg, J. Org. Chem. 1996, 61, 9582–9584.
[37] A. Voutchkova, A. Coplin, N. E. Leadbeater, R. H. Crabtree, Chem.
Commun. 2008, 6312–6314.
[19] L. J. Goossen, B. Zimmermann, T. Knauber, Angew. Chem. 2008,
120, 7211–7214; Angew. Chem. Int. Ed. 2008, 47, 7103–7106.
[20] L. J. Goossen, F. Rudolphi, C. Oppel, N. Rodrꢁguez, Angew. Chem.
2008, 120, 3085–3088; Angew. Chem. Int. Ed. 2008, 47, 3043–3045.
[21] L. J. Goossen, G. Deng, Eur. Pat. PCT/DE2006/001014, 2006.
[22] a) J.-M. Becht, C. Catala, L. D. Cedric, C. Le Drian, A. Wagner,
Org. Lett. 2007, 9, 1781–1783; b) J.-M. Becht, C. Le Drian, Org.
Lett. 2008, 10, 3161–3164.
[23] C. Peschko, C. Winklhofer, W. Steglich, Chem. Eur. J. 2000, 6, 1147–
1152.
[24] P. Forgione, M. C. Brochu, M. St-Onge, K. H. Thesen, M. D. Bailey,
F. Bilodeau, J. Am. Chem. Soc. 2006, 128, 11350–11351.
[25] L. J. Goossen, W. R. Thiel, N. Rodrꢁguez, C. Linder, B. Melzer, Adv.
Synth. Catal. 2007, 349, 2241–2246.
[26] L. J. Goossen, N. Rodrꢁguez, C. Linder, J. Am. Chem. Soc. 2008,
130, 15248–15249.
Received: April 3, 2009
Published online: August 28, 2009
Chem. Eur. J. 2009, 15, 9336 – 9349
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
9349