Alkyne hydroarylations with chelating dicarbene palladium(II) complex catalysts: Improved and unexpected reactivity patterns disclosed upon additive screening

Article abstract of DOI:10.1002/ejoc.200900321

Palladium(II) complexes with a ligand set made from a chelating N-heterocyclic dicarbene ligand and two weakly coordinating anions (generally introduced in situ upon addition of 2 equiv. of a suitable silver salt) were found to be very active and selectiv

Full text of DOI:10.1002/ejoc.200900321

FULL PAPER
DOI: 10.1002/ejoc.200900321
Alkyne Hydroarylations with Chelating Dicarbene Palladium(II) Complex
Catalysts: Improved and Unexpected Reactivity Patterns Disclosed Upon
Additive Screening[‡]
Andrea Biffis,*[a] Luca Gazzola,[a] Pierangelo Gobbo,[a] Gabriella Buscemi,[a]
Cristina Tubaro,[a] and Marino Basato[a]
Keywords: C–H activation / Palladium / Hydroarylation / Alkynes / Carbenes / N-heterocyclic carbenes
Palladium(II) complexes with a ligand set made from a che-
lating N-heterocyclic dicarbene ligand and two weakly coor-
dinating anions (generally introduced in situ upon addition
of 2 equiv. of a suitable silver salt) were found to be very
active and selective catalysts for the room-temperature hy-
droarylation of alkynes at low catalyst loading (0.1 mol-%).
Moreover, the screening of various strong acids as reaction
promoters revealed that both the strength of the acid and
the coordinating ability of its conjugated base influence the
catalytic performance. Most remarkably, the use of HBF4 to-
gether with a dicarbene Pd complex catalyst provides a dra-
matic change in the selectivity of the reaction, with the prev-
alent formation of a product stemming from the insertion of
two molecules of alkyne into the aromatic C–H bond. The
results presented herein highlight the fact that the dicarbene
ligand, apart from stabilising the catalyst, is also able to en-
hance catalytic activity and, most notably, to steer the reac-
tion selectivity towards novel products.
(© Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim,
Germany, 2009)
suggest
a Friedel–Crafts-type alkenylation mechanism
Introduction
(Scheme 1, path a)[8] whereas other data indicate a reaction
pathway that occurs by arene metallation (Scheme 1,
path b).[7b,9] The reaction is characterised by a high and
quite unusual regio- and stereoselectivity: in fact, it is re-
markable that the thermodynamically less favoured cis-aryl-
alkenes are the major products. Other noble metal centres,
such as platinum(II),[10] gold(I) and gold(III),[11] as well as
non-noble, electrophilic metal centres (group 3, 4, 13 and
15 elements, lanthanides)[12] have been successfully em-
ployed as alternative catalysts, but their efficiency appears
to be lower and/or their reaction scope narrower than that
of palladium(II).
Very recently we[13] reported palladium(II) complexes
with a ligand set made out of a chelating N-heterocyclic
dicarbene[14] and two halide ligands, which exhibit in this
reaction a superior performance compared with all other
previously described catalysts: alkyne hydroarylation can be
performed in a few hours at 80 °C with only 0.1 mol-% cat-
alyst and an equimolar amount of arene and alkyne to yield
the trans-hydroarylation product in high yields and with
good selectivity.
Aromatic C–H bond functionalisation reactions repre-
sent green and economic alternatives to the more classic
coupling reactions involving prefunctionalised aromatic
substrates, for example, Heck and cross-coupling reactions,
which are nowadays widely used for the conversion of aryl
halides into more complex organic molecules.[1] Several ex-
amples of such reactions have been reported in the recent
literature.[2,3] They are based on i) chelate-assisted oxidative
addition of the C–H bond to metal centres in a low oxi-
dation state,[4] ii) arene metallation by metal centres, which
attack the aromatic ring by electrophilic aromatic substitu-
tion or σ-bond metathesis[5] and finally iii) Friedel–Crafts-
type reactions promoted by metal centres that upon coordi-
nation electrophilically activate organic molecules towards
attack at the aromatic ring.[6]
Palladium(II) salts or complexes in a trifluoroacetic acid
environment are, for example, known to promote the cou-
pling reaction of arenes with alkynes.[7] In this case, prod-
ucts of the formal trans-hydroarylation of the triple bond
are formed. The mechanism of this reaction is still debated:
some theoretical calculations and experimental evidence
The reported reaction protocol is quite general with re-
spect to the alkyne, but limited to electron-rich arenes, al-
though this is a drawback common to all the catalytic sys-
tems described in the literature up to now. Furthermore,
the reaction conditions involve the use of a relatively high
reaction temperature and, most notably, of a great excess of
a strong acid. Both these requirements potentially limit the
[‡] Reactivity of Chelating Dicarbene Metal Complex Catalysts,
IV. Part III: Ref.[13a]
[a] Dipartimento di Scienze Chimiche, Università degli Studi di
Padova,
Via Marzolo 1, 35131 Padova, Italy
Fax: +39-049-8275223
E-mail: andrea.biffis@unipd.it
Eur. J. Org. Chem. 2009, 3189–3198
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
3189

A. Biffis, L. Gazzola, P. Gobbo, G. Buscemi, C. Tubaro, M. Basato
FULL PAPER
Scheme 1. Proposed reaction mechanisms for the hydroarylation of alkynes: a) Friedel–Crafts-type alkenylation; b) arene metallation.
application of this synthetic method as additional undesired identical catalytic performance within the margin of experi-
reactions, such as double-bond isomerisation, hydrolysis, al- mental error. Consequently, it appeared reasonable to as-
kyne hydration and/or polymerisation, are promoted.[13]
sume that under the adopted reaction conditions [1 equiv.
In this contribution, we demonstrate that chelating dicar- alkyne, 1 equiv. arene, 0.001 equiv. catalyst, 5 mL
bene palladium(II) complexes bearing weakly coordinating CF3COOH/1,2-dichloroethane (4:1), approximate reagent
anionic ligands are more efficient catalysts that can be con- concentration 2.1 , 80 °C, 5 h] a double M–X/CF3COO–
veniently employed at room temperature. Furthermore, the H exchange occurs, as predicted on the basis of DFT calcu-
use of these catalysts with acids stronger than trifluoro- lations by Strassner et al.[15]
acetic acid allows the amount of acid employed as additive
to be significantly reduced. Most remarkably, we show that
the nature of the employed acid, and in particular its
strength and the coordinating ability of its conjugated base,
Effect of the Addition of a Silver Salt
has a significant effect on the reaction outcome with dicar-
To confirm that the real catalytically active species con-
bene palladium(II) catalysts, producing not only differences
tains two trifluoroacetate ligands in the coordination sphere
in catalytic activity but also dramatic changes in reaction
we performed a standard hydroarylation reaction between
selectivity.
pentamethylbenzene and ethyl propiolate using complex 1
(Figure 1, Scheme 2) in the presence of 2 equiv. with respect
to the catalyst of silver trifluoroacetate.
Results and Discussion
In our previous work[13b] we showed that dicarbene com-
plexes with coordination sets differing only in the halide
ligands (e.g., complexes 3–5, Figure 1) exhibited an almost
Scheme 2. Hydroarylation reaction between pentamethylbenzene
and ethyl propiolate.
The conversion curve (Figure 2, open circles) clearly
shows that the reaction proceeds slightly more rapidly than
in the absence of the silver salt (Figure 2, filled circles). We
repeated the same reaction at room temperature and grati-
fyingly after 23 h we obtained a good conversion (74%; Fig-
ure 2, open squares); for comparison, analogous reactions
performed at room temperature with complexes containing
bromide or iodide ligands and in the absence of silver salts
Figure 1. Complexes 1–8.
3190
afforded only a few percent yield after 24 h.[13]
www.eurjoc.org
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2009, 3189–3198

Alkyne Hydroarylations with Dicarbene Palladium Catalysts
Figure 3. Conversion vs. time diagram for the reaction of penta-
methylbenzene and ethyl propiolate catalysed by complexes 3–5 at
room temperature, with or without added silver trifluoroacetate.
Figure 2. Conversion vs. time diagram for the reaction of penta-
methylbenzene and ethyl propiolate catalysed by complexes 1 and
2 at room temperature and 80 °C, with or without added silver
trifluoroacetate.
Catalyst Screening
We then turned our attention to a comparison of the
catalytic efficiency at room temperature of a series of palla-
dium complexes with different dicarbene ligands (Figure 1).
It has already been observed that the performance of these
catalysts at 80 °C was influenced by the type of dicarbene
ligand, without, however, clear evidence of the predomi-
nance of steric or electronic effects. Therefore we checked
the catalytic efficiency of the palladium(II) complexes 1, 4
and 6–8 in the presence of silver trifluoroacetate ([Ag]/[Pd]
2:1), using the standard reaction between pentamethylben-
zene and ethyl propiolate. Conversion curves (Figure 4) re-
vealed that all the employed complexes were active at room
The reaction yield achieved at room temperature after
24 h was comparable to that observed at 80 °C in the ab-
sence of added silver salt (83%), although the initial rate
was lower. Moreover, by performing the reaction at room
temperature, we could avoid the occurrence of side reac-
tions like hydrolysis of the ester function and Z/E isomeri-
sation of the olefin, thereby increasing the selectivity
towards the desired Z product (96 vs. 77% at 80 °C).
As a further confirmation of our hypothesis, we per-
formed the reaction with preformed complex 2 bearing two
coordinated trifluoroacetates (Figure 2, filled squares). As
expected, the conversion curve was found to be almost coin-
cident with that of the complex generated in situ by adding
silver trifluoroacetate to complex 1.
At room temperature the Pd–X/Ag–TFA exchange reac-
tion is fast and not influenced by the nature of the coordi-
nated halides, as illustrated by the absence of any induction
period and by the practically overlapping conversion curves
when complexes 4 and 5 were used as catalysts in the pres-
ence of 0.002 equiv. of silver trifluoroacetate (Figure 3,
open circles and squares). In contrast, we could observe
a long induction period in a test reaction performed with
complex 3 bearing chloride ligands in the absence of the
silver co-catalyst (Figure 3, filled triangles), which is clearly
necessary to generate the bis(trifluoroacetate) catalyst by
Pd–X/H–TFA exchange. As stated above, under the same
experimental conditions, complexes bearing bromide or io-
dide ligands, which are expected to be more strongly
bonded to the metal[16] and therefore not so easily displaced
by trifluoroacetate, are nearly inactive.
From these data it can be concluded that the halide/tri-
fluoroacetate exchange is a crucial step in the formation of
the competent catalytic species, which can be conveniently
accelerated by using a silver trifluoroacetate co-catalyst.
Figure 4. Conversion vs. time diagram for the reaction of penta-
methylbenzene and ethyl propiolate catalysed by complexes 1, 4
and 6–8 at room temperature and with added AgTFA.
Eur. J. Org. Chem. 2009, 3189–3198
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
3191

A. Biffis, L. Gazzola, P. Gobbo, G. Buscemi, C. Tubaro, M. Basato
FULL PAPER
temperature and that their performances, within the first Table 2. Catalyst screening in the hydroarylation of ethyl phenyl-
propiolate (or of diphenylacetylene) with pentamethylbenzene, [Pd]
= 0.5 mol-%.
5 h, were quite different. The order of decreasing activity (6
Ͼ 4 ≈ 7 Ͼ 1 Ͼ 8) is comparable to that already observed
Entry
1
Catalyst
Time [h]
Yield [%][a]
at 80 °C and, in particular, complex 6 with the more hinder-
ing NHC ligand showed the highest activity. The conversion
after 24 h with the benzimidazolin-2-ylidene complex 1 and
the imidazolin-2-ylidene complexes 4, 6 and 7 were similar
and quite satisfactory, whereas that of complex 8 was signif-
icantly lower, probably because of catalyst decomposition
after the first 7 h of reaction. These results suggest that the
steric hindrance imposed by the chelating dicarbene ligand
at the metal centre is an important factor in determining
the reactivity of the catalyst. Further support for this hy-
pothesis comes from preliminary data of electrochemical
studies,[17] which have demonstrated that the reduction po-
tentials of the metal in the employed catalysts, which is a
measure of the electron density on the metal, cannot be
correlated with their observed catalytic activity.
1
5
13
33
45
6
16
29
4
12
17
1
24
48
5
24
48
5
24
48
5
24
48
5
2
4
4
6
8
3[b]
4
9
19
19
29
30
5
24
48
[a] The yield was determined by GC–MS and/or 1H NMR spec-
troscopy. [b] Alkyne = diphenylacetylene. Reaction conditions: see
the Exp. Sect.
Arene and Alkyne Screening
Table 3. Arene screening in the hydroarylation of ethyl propiolate
catalysed by complex 6.
By using complex 6, which turned out to be the most
active in the previous screening at room temperature, we set
out to evaluate the generality of our catalytic system with
respect to both the alkyne and the arene. Two sets of stan-
dard reagents were employed under the same reaction con-
ditions without attempting to optimise the reaction yield.
The results are reported in Tables 1, 2, and 3.
Table 1. Alkyne screening in hydroarylation reactions with penta-
methylbenzene catalysed by complex 6 with added AgTFA.
[a] The conversion was based on the arene. The yield was deter-
1
mined by GC–MS and/or H NMR spectroscopy. Reaction condi-
tions: see the Exp. Sect.
room temperature. For example, by using complex 1 at
80 °C, phenylacetylene gave only 20% yield of the desired
product after 5 h because of competing alkyne hydration
and polymerisation.[13b] On the other hand, with complex 6
[a] The yield was determined by GC–MS and/or 1H NMR spec-
troscopy. [b] Reaction performed using complex 4 as catalyst. Reac-
tion conditions: see the Exp. Sect.
The highest activities and selectivities were observed with at room temperature it was possible to obtain a yield of
electron-poor terminal alkynes, whereas internal alkynes 52% after 48 h. The results with arylalkynes reported herein
were converted to a minor extent (Table 1). Remarkably, the should be, however, treated with some caution because it
selectivities for the desired product were in all cases signifi- has recently been demonstrated that trifluoroacetic acid
cantly improved in comparison with those obtained at alone can catalyse the hydroarylation of these substrates at
80 °C because competitive reactions were minimised at temperatures close to room temperature.[18]
3192
www.eurjoc.org
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2009, 3189–3198

Alkyne Hydroarylations with Dicarbene Palladium Catalysts
Quite unexpectedly, by using catalyst 4 in the reaction
The screening of arenes was performed by reacting ethyl
between pentamethylbenzene and ethyl phenylpropiolate, a propiolate with different methyl-substituted benzenes using
higher yield was obtained than with complex 6 (17% yield our best catalyst 6. The results reported in Table 3 show
vs. 9%) (Table 1, entries 3 and 2, respectively), in contrast that all the substrates were converted irrespective of the
with the order of reactivity observed in the previous screen- number of substituents, although the yield recorded with p-
ing. To clarify this aspect we performed some additional xylene was low. The distribution of the products (see
tests on this reaction by using other catalysts in 0.5 mol-% Scheme 3) did not change significantly on going from
amounts instead of the usual 0.1 mol-% because we also pentamethylbenzene to p-xylene; the main product was al-
wanted to increase the otherwise rather low reaction rate ways the trans-hydroarylation product, that is, the Z olefin.
with this alkyne. The results are summarised in Table 2.
An exception to this trend is the reaction with 1,2,4,5-tet-
The order of reactivity observed after 5 h in the reaction ramethylbenzene, which yields as a major byproduct the
with ethyl phenylpropiolate (8 Ͼ 1 Ͼ 4 Ͼ 6) was completely double hydroarylation product d (Scheme 3). This is proba-
reversed with respect to that obtained with ethyl propiolate bly a consequence of the comparatively low solubility of the
and illustrated in Figure 4 (6 Ͼ 4 Ͼ 1 Ͼ 8). In both cases, starting arene, which is always a shortcoming in this system
at longer reaction times, complex 8 was found to be deacti- and favours the reaction of a second molecule of the alkyne
vated, presumably because of the comparatively low sta- with the arene ring.
bility of this catalyst under the reaction conditions. These
results highlight the importance of the steric properties of
the alkyne in determining the efficiency of the catalytic sys-
tem. A rationalisation of the different behaviour of the
Effect of Acid and Conjugated Base
complexes with the two alkynes requires knowledge of the
Another important parameter to be investigated was the
reaction mechanism. If we assume a Friedel–Crafts-type role played by the acid on the reaction outcome. Up to now
mechanism (Scheme 1, path a),[8] which at this stage seems the standard acid employed in alkyne hydroarylation reac-
more plausible for our system, then the observed reversal tions was trifluoroacetic acid, which has to be added in very
in the order of catalytic efficiency can be explained by con- large amounts (arene/acid molar ratio of ca. 1:4) to pro-
sidering that an increased steric bulk of the complex en- mote reasonable catalytic activity. We have investigated the
hances the overall catalytic activity (e.g., by favouring pro- effect of changes in the nature and amount of added acid,
tonolysis of the Pd–vinyl intermediate), but negatively af- varying both its strength and the coordinating properties of
fects alkyne coordination to the metal, shifting the equilib- the conjugated base with the aim of decreasing the amount
rium towards the free alkyne. This last effect is particularly of acid while preserving catalytic activity. In a series of ex-
important in the case of hindered internal alkynes such that periments, we used complex 1 in various strong acids HX,
the required coordination for electrophilic attack to the ar- namely HTFA (pKa = –0.25), HBF4 (pKa = –4.9) or HOTf
ene is strongly inhibited.
(pKa = –14), adding to the system 2 equiv. of the corre-
sponding silver salt AgX with respect to 1. Moreover, we
reduced the quantity of the acid to an equimolar amount
with respect to the arene. The conversion curves in Figure 5
show that the catalytic efficiency increased using stronger
acids: for example, after 23 h, the arene conversion obtained
with triflic acid was 65%, compared with 25% in trifluoro-
acetic acid. Quite unexpectedly though, the reaction was
fastest in HBF4, which is a significantly weaker acid than
triflic acid; furthermore, with HBF4, the conversion of the
arene stopped abruptly at 50%. Analysis of the reaction
products led to the unexpected conclusion that this was due
to the prevalent formation of the double insertion product
c (Scheme 3), which consumed two molecules of alkyne for
each arene. Repeating the reaction with an arene/alkyne
molar ratio of 1:2 resulted in 86% conversion with 86%
selectivity towards product c. Remarkably, when simple
Pd(OAc)2 was employed as the catalyst under the same re-
action conditions, a much lower catalytic activity was ob-
served and it failed to show this peculiar selectivity, provid-
ing the conventional trans-hydroarylation product in 57%
yield after 24 h.
Comparison of the selectivities for the three systems at
23 h (Figure 6) showed that the main product in trifluoro-
acetic acid was the Z olefin (selectivity 89%), whereas in
Scheme 3. Possible products of the alkyne hydroarylation reaction. tetrafluoroboric acid the double insertion product was pre-
Eur. J. Org. Chem. 2009, 3189–3198
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
3193

A. Biffis, L. Gazzola, P. Gobbo, G. Buscemi, C. Tubaro, M. Basato
FULL PAPER
Figure 6. Selectivity at 23 h for the reaction of pentamethylbenzene
and ethyl propiolate catalysed by complex 1 in the presence of dif-
ferent acids; Ar/Al indicates the arene/alkyne molar ratio.
Figure 5. Arene conversion vs. time diagram for the reaction of
pentamethylbenzene and ethyl propiolate catalysed by complex 1
in the presence of different HX acids and of the corresponding
AgX salt; Ar/Al indicates the arene/alkyne molar ratio.
quently, it is reasonable to assume that c is formed directly
by insertion of a second alkyne molecule into the Pd–vinyl
dominant, as mentioned above (selectivity 86%). The cata- intermediate (Scheme 1), as already proposed by Fujiwara
lytic system in triflic acid was not very selective (70% selec- and co-workers.[7b]
tivity towards the Z olefin), the major byproducts being the
In a subsequent experiment the quantity of tetrafluo-
E olefin, hydrolysed olefins and a double insertion product roboric acid was reduced to an arene/acid molar ratio of
cЈ (Scheme 3) with Z,E configuration instead of the usual 1:0.1; the system was still very active in the hydroarylation
E,E (isomer c). We have verified that the formation of these of alkynes, reaching 59% conversion after 25 h compared
byproducts is triggered by the strong acidic environment: with 86% with an equimolar amount of acid (Figure 7).
for example, pure product c dissolved in a mixture HOTf/ Moreover, the selectivity of the system was not affected by
1,2-dichloroethane isomerises to cЈ within a few hours. the use of a reduced amount of acid as it remained constant
Moreover, triflic acid was also found to promote the hy- at 86% in product c. Interestingly, the same reaction run
droarylation reaction to some extent in the absence of Pd with the bis-trifluoroacetato complex 2 in HBF4 (Figure 7,
complex catalysts.[19] A conversion of 14% with 64% selec- full squares) was very slow (14% conversion after 25 h) and
tivity for the Z product was recorded after 24 h. In contrast, also the selectivity of the system towards the double inser-
neither tetrafluoroboric acid nor trifluoroacetic acid pro- tion product reached only 67% (compared with 86% with
moted the formation of hydroarylation products with ethyl complex 1 and AgBF4).
propiolate in the absence of a Pd catalyst.
These data confirm that the presence of a strong acid is
We also ran an experiment using oleum 20% as acid and necessary for the success of the reaction; the high activity
silver tetrafluoroborate as co-catalyst. The activity of this observed with acids stronger than the standard trifluoro-
system was superior to that of trifluoroacetic acid, but in- acetic acid is attributable to the faster protonolysis of the
ferior to that of tetrafluoroboric and triflic acid. Further- Pd–vinyl intermediate, which is generally proposed as the
more, complex product mixtures were formed, including rate-determining step of the catalytic cycle, particularly in
several hydrolysis and isomerisation products, so tests with the case of the Friedel–Crafts-type mechanism
this acid were not pursued further.
(Scheme 1).[8a] However, no simple direct correlation is ob-
Considering the peculiar reaction selectivity obtained served between the strength of the acid and the efficiency
with tetrafluoroboric acid, we set out to further investigate of the catalytic process. We interpret this by considering
this system. First, we ran a reaction with an arene/alkyne that the coordinating ability of the conjugated base
–
–
molar ratio of 1:3 to verify if the formation of the double (CF3COO– Ͼ CF3SO3 Ͼ BF4 ) also influences the cata-
insertion product c took place after the trans-hydroaryl- lytic activity: a weaker acid like HBF4 (pKa = –4.9) per-
ation reaction and to evaluate the selectivity of the system forms better than a stronger acid like HOTf (pKa = –14)
–
–
towards multiple insertions. The selectivity towards product because its conjugated base BF4 compared with CF3SO3
c (86% after 24 h) was the same as that obtained with an leaves two coordination sites on the metal centre almost free
arene/alkyne molar ratio of 1:2. Moreover, we did not ob- and capable of accommodating the reagents. Similarly, the
serve a decrease in the amount of trans-hydroarylation presence of more strongly coordinating trifluoroacetate li-
product a in favour of product c. This indicates that the gands on the starting complex strongly decreases the cata-
synthesis of c is not sequential to the formation of a. Conse- lytic activity in tetrafluoroboric acid. The same considera-
3194
www.eurjoc.org
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2009, 3189–3198

Alkyne Hydroarylations with Dicarbene Palladium Catalysts
Table 4. Arene screening in the hydroarylation of ethyl propiolate
catalysed by complex 1 in HBF4.
Figure 7. Arene conversion vs. time diagram for the reaction of
pentamethylbenzene and ethyl propiolate catalysed by complexes 1
and 2 in HBF4; [arene]/[alkyne] = 1:2.
tions could be invoked to explain the dramatic change in
the reaction selectivity observed with HBF4 as in this case
the metal centre should better coordinate a second molecule
of alkyne thereby preferentially affording the double inser-
tion product c. However, the selectivity maintained by using
simple Pd(OAc)2 in HBF4 makes it apparent that this expla-
nation is not fully satisfactory. Furthermore, very recently
Oyamada and Kitamura showed that some bidentate phos-
phane ligands, most notably diphenylphosphanylethane
(dppe), were able to steer the reaction selectivity towards
product c, even when using trifluoroacetic acid as a cosol-
vent.[20] Therefore, the high preference for the formation of
the double insertion product c appears at present to be a
peculiarity of selected palladium(II) complex catalysts. We
are currently investigating whether ligand features deter-
mine this selectivity change.
[a] The conversion was based on the arene. The conversion based
on the alkyne is presented in parentheses. The yield was determined
by GC–MS and/or 1H NMR spectroscopy. [b] Reaction performed
by using an arene/alkyne molar ratio of 1:2. [c] Additional yield in
the product resulting from double insertion at both the 2- and 6-
positions. [d] Yields of the double insertion product at the 2-posi-
tion or, in parentheses, at the 5-position. Reaction conditions: see
the Exp. Sect.
Substrate Screening in HBF4
The generality of the preferred reaction pathway with tet-
rafluoroboric acid leading to the double insertion product
has been evaluated with respect to both the alkyne and the an electron-withdrawing group into the arene, as in 2-
arene (Tables 4 and 5). Complex 1 (0.1 mol-%) in the pres- bromo-1,3,5-trimethylbenzene (Table 4, entry 8). The al-
ence of AgBF4 (0.2 mol-%) was used under standard reac- kyne/arene molar ratio was found to influence the product
tion conditions (arene/alkyne molar ratio of 1:1, arene/ distribution (Table 4, entries 4 and 5); the main product was
HBF4 molar ratio of 1:1, 1,2-dichloroethane, room tem- always the double insertion product, but at a lower arene/
perature).
alkyne ratio the double trans-hydroarylation product d was
The results reported in Table 4 demonstrate that the ef- not present.
fect of the addition of HBF4/AgBF4 on the selectivity of
The position of the methyl substituents on the ring also
the reaction is general: the double insertion product c was played an important role in determining the reactivity of
the major product with all the arene substrates employed. the substrates; for example, p-xylene showed a lower reac-
The recorded yield in c after 24 h depends on the nature of tivity than m-xylene (Table 2, entries 6 and 7) and this can
the arene and follows the already observed trends for Pd- be attributed to the activation of both the 2- and 5-posi-
catalysed hydroarylation reactions,[7,13] that is, it decreases tions of the ring in the latter case. Moreover, with m-xylene
for less methyl-substituted arenes or upon introduction of a 2:1 arene/alkyne adduct has been isolated, which is pre-
Eur. J. Org. Chem. 2009, 3189–3198
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
3195

A. Biffis, L. Gazzola, P. Gobbo, G. Buscemi, C. Tubaro, M. Basato
FULL PAPER
Table 5. Alkyne screening in hydroarylations with pentamethylben- Note also that in the case of arylalkynes the hydroarylation
zene catalysed by complex 1 in HBF4.
reaction was promoted to a certain degree by HBF4 alone
(Table 6), which clearly complicates the evaluation of the
catalytic performance of the Pd complex. The acid also pro-
motes isomerisation processes, particularly marked in the
case of diphenylacetylene (Table 5, entry 6): at 5 h the only
observed product is the Z olefin, whereas at 48 h it is the E
isomer.
Table 6. Alkyne screening in hydroarylations with pentamethylben-
zene in HBF4 without metal catalyst.
[a] The yield was determined by GC–MS and/or 1H NMR spec-
troscopy. Reaction conditions: see the Exp. Sect.
[a] The yield was determined by GC–MS and/or 1H NMR spec-
troscopy. [b] The reaction was performed by using an arene/alkyne
molar ratio of 1:2. Reaction conditions: see the Exp. Sect.
Conclusions
We have been able to significantly improve our original
sumably formed by a hydroarylation reaction of olefin a. system for the hydroarylation of alkynes with chelating di-
Altogether, the conversions are remarkably good and com- carbene palladium(II) catalysts by the appropriate choice of
parable to those observed in trifluoroacetic acid, despite the additives such as a strong acid and a silver salt. By using a
smaller quantity of acid used.
catalyst with weakly coordinating anionic ligands, gener-
Concerning the behaviour of this catalytic system in the ated in situ upon addition of 2 equiv. of the corresponding
reaction of pentamethylbenzene with differently substituted silver salt, the reaction can be performed at room tempera-
alkynes (Table 5), terminal alkynes were found once more ture, thus minimising side reactions, like hydration and
to be more reactive than internal ones. The selectivity of polymerisation of the alkyne and hydrolysis of the ester
the system strongly depends on the nature of the alkyne. function. The use of acids stronger than trifluoroacetic
Terminal alkynes conjugated with electron-withdrawing acid, like HBF4 or HOTf, allows a significant reduction in
groups, like ethyl propiolate, predominantly give the double the quantity of the acid used while maintaining the catalytic
insertion product c. A double insertion product is also pro- activity. Most notably, use of the dicarbene Pd catalyst 1
duced as a minor product with 3-butyn-2-one; in this case, in HBF4 with 2 equiv. of AgBF4 produced a highly active
the Pd–vinyl intermediate undergoes rapid isomerisation to catalytic system able to direct the selectivity of the reaction
yield the cis-hydroarylation product. In this case, however, towards the formation of the double insertion product. The
different stereoisomers of the double insertion product can applicability of this reaction protocol is quite general with
form so that additional work is needed for their identifica- respect to electron-rich arenes.
tion and quantification. Finally, no double insertion prod-
It is important to note that in the light of the findings
uct was observed with terminal or internal arylalkynes. presented herein, the role of the chelating dicarbene ligand
Possibly, in this case the coordination of a second molecule is not limited to the stabilisation of the metal in its +II
of alkyne to the Pd–vinyl intermediate is unfavoured for oxidation state: the ligand is also able to promote catalytic
steric reasons, thus preventing the formation of product c. activity and to control the reaction selectivity. Given the
3196
www.eurjoc.org
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2009, 3189–3198

Alkyne Hydroarylations with Dicarbene Palladium Catalysts
and the arene (if liquid) were then added and the resulting solution
was stirred at 25 °C for 5 min. Finally the alkyne (13.2 mmol) was
added and the reaction mixture was stirred at 25 °C for the time
indicated in the tables. Portions of the solution (0.2 mL) were
drawn off from the reaction mixture and analysed by H NMR or
GC–MS. The same conditions were used for the tests reported in
nature of these ligands, this is further confirmation that the
combination of a strongly coordinating and electron-donat-
ing ligand, together with the presence of positive charge on
the metal, may yield complexes that in spite of the nature
of the ligand exhibit excellent performance as electrophilic
catalysts.[21]
1
Table 6, however, without the use of a catalyst and silver salt.
Work currently in progress aims at expanding the scope
of the reaction to different classes of aromatic systems, at
finding novel applications for electrophilic metal carbene
complexes as well as at developing still less-impacting reac-
tion media.
Determination of Conversion Curves Reported in Figure 7: Penta-
methylbenzene (13.2 mmol), the palladium(II) complex (1.32 µmol)
and AgBF4 (2.64 µmol) were placed in a 100 mL round-bottomed
flask previously evacuated and filled with argon. HBF4 (13.2 mmol
or 1.32 mmol) and 1,2-dichloroethane (the quantity necessary to
reach a total volume of 6.3 mL) were then added and the resulting
solution was stirred at 25 °C for 5 min. Finally, ethyl propiolate
(13.2 mmol) was added and the reaction mixture was stirred at
25 °C. Portions of the solution (0.2 mL) were drawn off from the
Experimental Section
General Remarks: All manipulations were carried out using stan-
dard Schlenk techniques under argon or dinitrogen. The reagents
were purchased by Aldrich as high-purity products and generally
used as received. Complexes 1,[13b] 3,[22] 4,[23] 5,[24] 6,[25] 7[26] and
8,[27] were prepared according to literature procedures. All solvents
were used as received as technical grade solvents. NMR spectra
were recorded with a Bruker Avance 300 spectrometer (300.1 MHz
1
reaction mixture and analysed by H NMR or GC–MS.
[1] A. de Meijere, F. Diederichs (Eds.), Metal-catalyzed cross-cou-
pling reactions, 2nd ed., Wiley-VCH, Weinheim, 2004.
[2] K. I. Goldberg, A. S. Goldman (Eds.), Activation and Function-
alization of C–H Bonds, ACS Symposium Series 885, American
Chemical Society, Washington, DC, 2004.
[3] a) D. Alberico, M. E. Scott, M. Lautens, Chem. Rev. 2007, 107,
174–238; b) L. A. Goj, T. B. Gunnoe, Curr. Org. Chem. 2005,
9, 671–685; c) O. Daugulis, V. G. Zaitsev, D. Shabashov, Q. N.
Pham, A. Lazareva, Synlett 2006, 3382–3388; d) A. R. Dick,
M. S. Sanford, Tetrahedron 2006, 62, 2439–2463; e) L. C. Cam-
peau, K. Fagnou, Chem. Commun. 2006, 1253–1264; f) F. Kak-
iuchi, N. Chatani, Adv. Synth. Catal. 2003, 345, 1077–1101; g)
V. Ritleng, C. Sirlin, M. Pfeffer, Chem. Rev. 2002, 102, 1731–
1769.
[4] a) S. Murai, F. Kakiuchi, S. Sekine, Y. Tanaka, A. Kamatani,
M. Sonoda, N. Chatani, Nature 1993, 366, 529–531; b) R. Mar-
tinez, R. Chevalier, S. Darses, J.-P. Genet, Angew. Chem. Int.
Ed. 2006, 45, 8232–8235.
[5] D. Garcia-Cuadrado, P. de Mendoza, A. A. C. Braga, F. Ma-
seras, A. M. Echavarren, J. Am. Chem. Soc. 2007, 129, 6880–
6886, and references cited therein.
[6] a) I. Iovel, K. Mertins, J. Kischel, A. Zapf, M. Beller, Angew.
Chem. Int. Ed. 2005, 44, 238–242; b) A. Dzudza, T. J. Marks,
J. Org. Chem. 2008, 73, 4004–4016, and references cited therein.
[7] a) C. Jia, D. Piao, J. Oyamada, W. Lu, T. Kitamura, Y. Fuji-
wara, Science 2000, 287, 1992–1995; b) C. Jia, W. Lu, J. Oyam-
ada, T. Kitamura, K. Matsuda, M. Irie, Y. Fujiwara, J. Am.
Chem. Soc. 2000, 122, 7252–7263; c) C. Jia, T. Kitamura, Y.
Fujiwara, Acc. Chem. Res. 2001, 34, 633–639; d) M. S. Viciu,
E. D. Stevens, J. L. Petersen, S. P. Nolan, Organometallics 2004,
23, 3752–3755.
1
for H and 75.5 for 13C). Chemical shifts (δ) are reported in units
of ppm relative to the residual solvent signals. Elemental analyses
were carried out by the microanalytical laboratory of our depart-
ment with a Fisons EA 1108 CHNS-O apparatus.
Synthesis of (1,1Ј-Dimethyl-3,3Ј-methylenedibenzimidazolin-2,2Ј-
ylidene)palladium(II) Bis(trifluoroacetate) (2): A suspension of com-
plex 1 (0.20 g, 0.25 mmol) and AgTFA (0.11 g, 0.50 mmol) in ace-
tonitrile (15 mL) was heated at 60 °C for 4 h and then filtered
through Celite to remove silver salts. The solvent was removed un-
der vacuum and the off-white residue was washed with diethyl ether
(2ϫ5 mL), filtered and dried under vacuum (132 mg, yield 76%).
C21H16F6N4O4Pd·0.5AgBr (702.7): calcd. C 35.90, H 2.30, N 7.97;
1
found C 35.63, H 1.79, N 7.88. H NMR ([D6]DMSO, 25 °C): δ =
4.02 (s, 3 H, CH3), 6.75 and 7.40 (2 d, 1 H, CH2), 7.40–7.60 (m, 4
H, Ar), 7.76 (d, 1 H, CH=CH), 8.25 (d, 1 H, CH=CH) ppm.
13C{1H} NMR ([D6]DMSO, 25 °C): δ = 34.0 (CH3), 57.1 (CH2),
111.3, 111.9, 124.5, 124.6, 132.3, 133.7 (CAr) ppm. Signals arising
from carbene carbon atoms and CF3COO groups were not de-
tected.
General Procedures for the Catalytic Tests
Tests Reported in Tables 1, 2, and 3. Determination of Conversion
Curves Reported in Figures 2, 3, and 4: The arene (13.2 mmol, if
solid), the palladium(II) complex (1.32 or 6.6 µmol for the tests
reported in Table 2) and silver trifluoroacetate (2.64 or 13.2 µmol
for the tests reported in Table 2) were placed in a 100 mL round-
bottomed flask previously evacuated and filled with argon. Tri-
fluoroacetic acid (4 mL), 1,2-dichloroethane (1 mL) and the arene
(if liquid) were then added and the resulting solution was stirred at
25 °C for 5 min. Finally the alkyne (13.2 mmol) was added and the
reaction mixture was stirred at 25 °C for the time indicated in the
tables. Portions of the solution (0.2 mL) were drawn off from the
[8] a) J. A. Tunge, L. N. Foresee, Organometallics 2005, 24, 6440–
6444; b) E. Soriano, J. Marco-Contelles, Organometallics 2006,
25, 4542–4553.
[9] N. Chernyak, V. Gevorgyan, J. Am. Chem. Soc. 2008, 130,
5636–5637.
[10] a) J. Oyamada, T. Kitamura, Chem. Lett. 2005, 34, 1430–1431;
b) J. Oyamada, T. Kitamura, Tetrahedron Lett. 2005, 46, 3823–
3827; c) J. Oyamada, T. Kitamura, Tetrahedron 2006, 62, 6918–
6925.
[11] a) M. T. Reetz, K. Sommer, Eur. J. Org. Chem. 2003, 3485–
3496; b) Z. Shi, C. He, J. Org. Chem. 2004, 69, 3669–3671.
[12] a) T. Tsuchimoto, T. Maeda, E. Shirakawa, Y. Kawakami,
Chem. Commun. 2000, 1573–1574; b) C. E. Song, D.-u. Jung,
S. Y. Choung, E. J. Roh, S.-g. Lee, Angew. Chem. Int. Ed. 2004,
43, 6183–6185; c) R. Li, S. R. Wang, W. Lu, Org. Lett. 2007,
9, 2219–2222; d) D. S. Choi, J. H. Kim, U. S. Shin, R. R.
Deshmukh, C. E. Song, Chem. Commun. 2007, 3482–3484; e)
M. Y. Yoon, J. H. Kim, D. S. Choi, U. S. Shin, J. Y. Lee, C. E.
Song, Adv. Synth. Catal. 2007, 349, 1725–1737.
1
reaction mixture and analysed by H NMR or GC–MS.
Tests Reported in Tables 4 and 5. Determination of the Conversion
Curves Reported in Figure 5: The arene (13.2 mmol, if solid), the
palladium(II) complex (1.32 µmol) and the silver salt (2.64 µmol,
AgTFA when HTFA was used, AgOTf for HOTf, AgBF4 for
HBF4) were placed in a 100 mL round-bottomed flask previously
evacuated and filled with argon. The acid (13.2 mmol), 1,2-dichlor-
oethane (the quantity necessary to reach a total volume of 6.3 mL)
Eur. J. Org. Chem. 2009, 3189–3198
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
3197

A. Biffis, L. Gazzola, P. Gobbo, G. Buscemi, C. Tubaro, M. Basato
FULL PAPER
[13] a) G. Buscemi, A. Biffis, C. Tubaro, M. Basato, Catal. Today
2009, 140, 84–89; b) A. Biffis, C. Tubaro, G. Buscemi, M. Bas-
ato, Adv. Synth. Catal. 2008, 350, 189–196.
Savechenkov, A. P. Rudenko, A. V. Vasil’ev, G. K. Fukin, Russ.
J. Org. Chem. 2005, 41, 1316–1328.
[20] J. Oyamada, T. Kitamura, Chem. Commun. 2008, 4992–4994.
[21] For other examples of N-heterocyclic carbene metal complexes
as electrophilic catalysts, see: M. Muehlhofer, T. Strassner,
W. A. Herrmann, Angew. Chem. Int. Ed. 2002, 41, 1745–1747
[14] For general references on N-heterocyclic carbene ligands, see:
a) F. E. Hahn, M. C. Jahnke, Angew. Chem. Int. Ed. 2008, 47,
3122–3172; b) F. Glorius (Ed.), N-Heterocyclic Carbenes in
Transition Metal Catalysis, Topics in Organometallic Chemis-
try, vol. 21, Springer, Heidelberg, 2007; S. P. Nolan (Ed.), N-
Heterocyclic Carbenes in Synthesis, Wiley-VCH, Weinheim,
2006; for reviews on chelating dicarbene metal complexes, see:
A. Mata, M. Poyatos, E. Peris, Coord. Chem. Rev. 2007, 251,
841–859; c) A. T. Normand, K. J. Cavell, Eur. J. Inorg. Chem.
2008, 2781–2800.
[15] T. Strassner, M. Muehlhofer, A. Zeller, E. Herdtweck, W. A.
Herrmann, J. Organomet. Chem. 2004, 689, 1418–1424.
[16] K. Fagnou, M. Lautens, Angew. Chem. Int. Ed. 2002, 41, 26–
47.
and ref.[15]
.
[22] S. Ahrens, A. Zeller, M. Taige, T. Strassner, Organometallics
2006, 25, 5409–5415.
[23] a) E. Herdtweck, M. Muehlhofer, T. Strassner, Acta Crys-
tallogr., Sect. E, Struct. Rep. Online 2003, m970–m971; b)
M. G. Gardiner, W. A. Herrmann, C.-P. Reisinger, J. Schwarz,
M. Spiegler, J. Organomet. Chem. 1999, 572, 239–247.
[24] W. A. Herrmann, C.-P. Reisinger, M. Spiegler, J. Organomet.
Chem. 1998, 557, 93–96.
[25] W. A. Herrmann, J. Schwarz, M. G. Gardiner, Organometallics
1999, 18, 4082–4089.
[17] G. Buscemi, A. Ahmed Isse, M. Basato, A. Gennaro, unpub-
lished results.
[18] M. A. Rahman, O. Ogawa, J. Oyamada, T. Kitamura, Synthesis
2008, 3755–3760.
[19] a) S. A. Aristov, A. V. Vasil’ev, G. K. Fukin, A. P. Rudenko,
Russ. J. Org. Chem. 2007, 43, 691–705; b) A. O. Shchukin, A. V.
Vasilyev, Appl. Catal. A: General 2008, 336, 140–147; c) P. Yu.
[26] C. Tubaro, A. Biffis, C. Gonzato, M. Zecca, M. Basato, J. Mol.
Catal. A 2006, 248, 93–98.
[27] A. M. Magill, D. S. McGuinness, K. J. Cavell, G. J. P. Britov-
sek, V. C. Gibson, A. J. P. White, D. J. Williams, A. H. White,
B. W. Skelton, J. Organomet. Chem. 2001, 617–618, 546–560.
Received: March 25, 2009
Published Online: May 14, 2009
3198
www.eurjoc.org
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2009, 3189–3198