Theoretical and experimental investigation of palladium(II)-catalyzed decarboxylative addition of arenecarboxylic acid to nitrile

Article abstract of DOI:10.1021/om3009525

The reaction mechanism of palladium(II)-catalyzed decarboxylative addition of 2,6-dimethoxybenzoic acid to acetonitrile was investigated by means of density functional theory (DFT) calculations. Calculations of the free energy profile for decarboxylation and carbopalladation indicated carbopalladation as the rate-determining step of the reaction. Investigation of the free energy profile for a series of experimentally evaluated nitrogen-based bidentate palladium ligands revealed that higher energy is required for decarboxylation and carbopalladation employing the experimentally least efficient ligand. The DFT investigation also showed that the relative free energies of the transition states were lowered in polar solvent, and preparative experiments confirmed that a nonoptimal ligand could be greatly improved by addition of water to the reaction system.

Full text of DOI:10.1021/om3009525

Article
pubs.acs.org/Organometallics
Theoretical and Experimental Investigation of Palladium(II)-
Catalyzed Decarboxylative Addition of Arenecarboxylic Acid to
Nitrile
Fredrik Svensson, Rajendra S. Mane, Jonas Sav
̈
̈
marker, Mats Larhed, and Christian Skold*
Organic Pharmaceutical Chemistry, Department of Medicinal Chemistry, BMC, Uppsala University, P.O. Box 574, SE-751 23
Uppsala, Sweden
*
S Supporting Information
ABSTRACT: The reaction mechanism of palladium(II)-catalyzed decarboxylative addition of 2,6-dimethoxybenzoic acid to
acetonitrile was investigated by means of density functional theory (DFT) calculations. Calculations of the free energy profile for
decarboxylation and carbopalladation indicated carbopalladation as the rate-determining step of the reaction. Investigation of the
free energy profile for a series of experimentally evaluated nitrogen-based bidentate palladium ligands revealed that higher energy
is required for decarboxylation and carbopalladation employing the experimentally least efficient ligand. The DFT investigation
also showed that the relative free energies of the transition states were lowered in polar solvent, and preparative experiments
confirmed that a nonoptimal ligand could be greatly improved by addition of water to the reaction system.
1
6,18,28
18
and Ag, as well as for uncatalyzed systems.29−33 In
INTRODUCTION
Cu,
■
contrast, to the best of our knowledge no theoretical
mechanistic investigations of the formation of the ketimine
have been published.
Ketone moieties are common structural features in pharma-
ceuticals and natural products. There are a number of different
methods to produce ketones, and one elegant way of producing
aryl ketones is the carbon−transition metal 1,2-carbopallada-
tion of a nitrile moiety followed by hydrolysis of the formed
The focus of this study is the Pd(II)-catalyzed decarbox-
ylative addition of 2,6-dimethoxybenzoic acid (1) to acetonitrile
(MeCN) as reported by Lindh et al. (Scheme 1), which has
been used as a model reaction for the development of optimal
1
ketimine. Although this method was first reported in 1970, it
has recently received increased attention with the development
of protocols employing palladium catalysis by Larock and co-
9
reaction conditions to produce aryl ketones. Lindh et al.
2
,3
investigated the reaction by means of electrospray ionization
mass spectrometry (ESI/MS) to characterize intermediate
palladium complexes, and on the basis of mechanistic studies,
the reaction is believed to proceed by an initial decarboxylation
of 1, producing a Pd−aryl species and subsequent carbopalla-
dation of MeCN to form the ketimine 2. The ketimine is
readily hydrolyzed to form the corresponding ketone 3, with
workers. Several efficient protocols for synthesizing aryl
2
,3
4,5
ketones via an initial C−H activation, oxidative addition,
transmetalation,
now available.
Catalytic decarboxylation of benzoic acid analogues provides
a highly interesting transformation that utilizes cheap and
available carboxylic acids as aryl sources. Decarboxylation of
benzoic acid and substituted benzoic acids is well known for
uncatalyzed systems and has also been explored for Cu,
Pd, Ag, Au, Rh, and mixed Pd/Cu catalysis.
and co-workers utilized Pd(II)-catalyzed decarboxylation of
6
−8
9
10−12
decarboxylation, or desulfination
are
13
3
,9
the exception of sterically hindered mesityl ketimines.
14
15,16
The efficiency of the Pd(II)-catalyzed addition reaction was
strongly affected by the ligand used. Within a series of related
nitrogen-based bidentate ligands L1−L7 (Table 1) the yields
ranged from 16 to 73% under the same reaction conditions. In
1
7
18
19
20
21−25
Larhed
benzoic acid analogues in order to generate the active Pd−aryl
intermediate that precedes carbopalladation. Mechanistic
9
investigations of decarboxylation using DFT calculations have
been performed for metal-catalyzed systems employing Pd,
Received: October 11, 2012
Published: January 11, 2013
2
6,27
©
2013 American Chemical Society
490
dx.doi.org/10.1021/om3009525 | Organometallics 2013, 32, 490−497

Organometallics
Article
Scheme 1. Reaction Scheme
Table 1. Nitrogen Ligands and Isolated Yields of 3 from 1
and MeCN (Scheme 1) As Reported by Lindh et al.
PBF solvation model with default settings in Jaguar for water or
4
3,44
9
acetonitrile where applicable.
Dispersion correction was calculated
4
5
using the DFT-D3 program. Frequencies and thermodynamic
contribution to the free energy were calculated for the optimized
geometries in the gas phase at 373.15 K. The reported free energies
were calculated by adding the thermodynamic correction, including
zero-point energy, and the dispersion correction to the single-point
solution phase energy.
The presented transition states (TSs) were verified to have only one
imaginary frequency and were verified to be connected to the
presented succeeding and preceding energy minima, respectively. All
stationary minima were verified to have no imaginary frequencies.
4
6
Figures 2 and 4 were prepared using XYZ-viewer.
Chemistry. Reaction Conditions for Ligand Investigation (Table
9
1
) As Reported by Lindh et al. A 5 mL microwave vial was charged
with 2,6-dimethoxybenzoic acid (0.5 mmol), Pd(O CCF ) (0.02
2
3 2
mmol), ligand (0.024 mmol), H O (200 μL), and MeCN (2 mL). The
2
vial was capped in air and exposed to microwave heating for 30 min at
1
30 °C.
Reaction Conditions for Water Content Investigation (Table 3). A
mL process vial was charged with 2,6-dimethoxybenzoic acid (91 mg,
.5 mmol), Pd(O CCF ) (6.6 mg, 0.02 mmol), 2,2′-bipyridine (3.9
5
0
2
3 2
mg, 0.025 mmol), H O (200−1600 μL), and MeCN (600−2000 μL)
2
in a combination resulting in the same total volume. The vial was
capped in air and exposed to microwave heating for 30 min at 130 °C.
RESULTS AND DISCUSSION
■
The calculations were set up using Pd(OAc) as the palladium
2
source, acetonitrile as the solvent, and reactant and calculating
thermochemical properties at 373.15 K, conditions known to
yield product. Ligand L3 was chosen as the reference ligand,
since it has been reported to give the best yields and we chose
to focus our investigation on the reaction path indicated by the
a
b
Isolated yields. Pd(OAc) was used instead of Pd(O CCF ) .
2
2
3 2
9
general the methyl-substituted bipyridines or phenanthrolines
L3, L4, L6, and L7 were more efficient ligands in the reaction
in comparison to the nonsubstituted analogues L1, L2, and L5.
The beneficial effect from the methyl groups in phenanthrolines
and corresponding bipyridines could possibly be attributed to
destabilization of square-planar complexes in the reaction
reported ESI/MS studies.
The energies for dissociation and association between
different complexes considered were not calculated but have
been estimated to be diffusion limited with a energy
−1
47
requirement in the magnitude of 20 kJ mol . A dashed
line in the free energy profiles indicates that the pathway for the
connection was not investigated.
3
4
pathway.
Mechanistic studies of the reaction may provide valuable
insights into the catalytic system and aid in the development of
improved protocols. Herein we investigate the two main
reaction steps, i.e., decarboxylation and carbopalladation, by
means of DFT calculations and explore the effect of ligands and
solvents on the energy profile of the reaction.
Decarboxylation. The free energy profile for decarbox-
ylation is shown in Figure 1. The starting point for the
investigation was the ligated diacetate Pd(II) complex I.
Replacement of an acetate with 1 gave complex II (Figure
2), which is 6 kJ mol− lower in energy and resulted in the
lowest energy complex found preceding decarboxylation. This
low-energy complex II was chosen as reference, and all other
energies are reported relative to this. Palladium(II)-catalyzed
decarboxylation of benzoic acid has been investigated by others
1
EXPERIMENTAL DETAILS
■
Computational Details. All DFT calculations were carried out
3
5
26,27
using Jaguar version 7.8 employing the B3LYP hybrid func-
by means of DFT calculations.
In those investigations the
36−38
tional
and LACVP* basis set, which uses an effective core
decarboxylation TS is higher in energy than any of the
complexes preceding the TS. On the basis of those results it is
reasonable to assume that this is also the case in our
investigation, and because numerous pathways are possible,
the specific pathway from II to the decarboxylation TS was not
investigated in great detail. Dissociation of the second acetate
produced the cationic complex III with deprotonated 1 in a
39
potential for Pd and 6-31G* for elements H−Ar. The B3LYP
functional has for many years been the most widely used density
40
functional and has been used together with variants of the 6-31G
basis set for examining reaction profiles of Pd(II)-catalyzed
2
7,41,42
decarboxylation and carbopalladation.
The complexes were
optimized in the gas phase, and the optimized structures were
subsequently subjected to a single-point energy calculation using the
4
91
dx.doi.org/10.1021/om3009525 | Organometallics 2013, 32, 490−497

Organometallics
Article
Figure 1. Free energy profile of Pd(II)-catalyzed decarboxylation of 1 using L3 as ligand.
complex IV, which is connected to TS-I, in agreement with
previous DFT investigations, which have reported similar TSs
2
6,27
and prereactive minima.
In TS-I (Figure 2) the plane of the
carboxylic acid moiety is orthogonal to the plane of the aryl
group in 1. Furthermore, the aryl is located trans to the 6-
methyl-substituted pyridine moiety in L3, which corresponds to
the less sterically hindered pathway. Overall, the free energy
required from the preceding lowest energy complex II to TS-I
−1
is calculated to be 110 kJ mol , which is similar to that
reported for other investigations of Pd(II)-catalyzed decarbox-
26,27
ylation of benzoic acid analogues, albeit via neutral TSs.
Comparison between neutral and charged complexes and TSs
is a challenging task and requires extra care to be taken, since
such a comparison is highly dependent on the use of an
48
adequate solvation model. The continuum solvation model
used in this study has been shown to give a fair representation
of the relative energies, but solvation of charged molecules is
associated with a higher margin of error, which has been
−1
49
estimated for small anions to be in the range of 25 kJ mol .
After the decarboxylation product is formed in complex V,
dissociation of carbon dioxide gave complex VI. From VI
association of acetonitrile formed complex VIII, which is ready
for carbopalladation of the nitrile. However, in the reaction
mixture there are several potential ligands that may give low-
energy complexes preceding carbopalladation. It was observed
that the neutral complexes were energetically favored and the
neutral complex VII (Figure 4) was identified as the lowest
energy minimum preceding carbopalladation and thus was a
suitable starting point for carbopalladation.
Carbopalladation. Figure 3 shows the free energy profile
for carbopalladation. Dissociation of deprotonated 1 from
complex VII followed by association of MeCN gave the
cationic precarbopalladation complex VIII. In comparison to
Figure 2. Optimized geometries for II and TS-I. Bond lengths (Å)
between Pd and coordinated atoms and for the breaking carbon−
carbon bond in TS-I are depicted.
2
carbopalladation of alkynes starting from the η binding
41
1
bidentate binding mode, which was the lowest energy
minimum found for the cationic predecarboxylation complexes.
Changing the binding mode of deprotonated 1 resulted in
mode, the η -MeCN approaches the TS by Pd−N−C angle
bending up to TS-II (Figure 4). This makes the C−N bond,
just as in the case with alkynes, parallel to the aryl−Pd bond in
4
92
dx.doi.org/10.1021/om3009525 | Organometallics 2013, 32, 490−497

Organometallics
Article
Figure 3. Free energy profile of carbopalladation of MeCN using L3 as ligand.
the TS and with the plane of the aryl group orthogonal to the
plane of the forming ketimine moiety. The energy difference
between the carbopalladation TS of the alkyne and the
−1
prereactive complex is reported to be 85 kJ mol , which is
in the same range as the corresponding energy required for
carbopalladation of acetonitrile from VIII, which is calculated
−1
to be 77 kJ mol . The free energy required for the full
carbopalladation step starting from VII is calculated to be 132
−1
kJ mol .
From TS-II the formation and release of the imine product
starts with complex IX. Subsequent association of deprotonated
gives the neutral complex X, which is lower in energy.
1
Replacing the carbopalladation product 2 in complex X for
acetate regenerates the starting complex II, and hydrolysis
yields the target ketone 3, which lowers the energy to the final
level of the catalytic cycle. Overall the free energy profile shows
that the reaction is exergonic and suggests carbopalladation as
the rate-determining step. A detailed catalytic cycle based on
the computational results is shown in Scheme 2.
Ligand Effect. After the reaction energy profile was
charaterized using L3 as ligand, the effect on the reaction
energy profile from the rest of the ligands L1−L7 was
investigated, and the results are presented in Figure 5. The
relative energies of the presented complexes are shown in Table
2
. Complex II was identified as the lowest energy minima
preceding decarboxylation for all ligands. The cationic complex
III showed a clear trend in relative free energy, where the
ligands with methyl substituents (L3, L4, L6, and L7) gave
complexes lower in energy in comparison to ligands devoid of
methyl substituents (L1, L2, and L5). This trend can be
explained by the release of steric crowding in complex II for L3,
2
6,34
L4, L6 ,and L7, in agreement with investigations by others.
Figure 4. Optimized geometries for VII and TS-II. Bond lengths (Å)
between palladium and coordinated atoms and for the forming
carbon−carbon bond in TS-II are shown.
However, this effect was reduced when progressing to complex
IV. Overall, the free energy required for the decarboxylation
process from complex II to the transition state TS-I varied in
−1
the range 109−120 kJ mol .
4
93
dx.doi.org/10.1021/om3009525 | Organometallics 2013, 32, 490−497

Organometallics
Article
Scheme 2. Detailed Catalytic Cycle Based on the Computational Results
Figure 5. Free energy profile of Pd(II)-catalyzed decarboxylative addition of 2,6-dimethoxybenzoic acid to MeCN for different nitrogen-based
ligands (Table 1): L1, black; L2, turquoise; L3, magenta; L4, gold; L5, red; L6, blue; L7, green. Complex structures are the same as in Figures 1 and
2, with the exception that in the low-energy path for L4, L6, and L7 2,6-dimethoxybenzoic acid in complex X has been replaced with acetate.
4
94
dx.doi.org/10.1021/om3009525 | Organometallics 2013, 32, 490−497

Organometallics
Article
−1
Table 2. Relative Energies (kJ mol ) of the Investigated
Complexes for All Ligands
this method may be used to predict the efficacy of related
bidentate, nitrogen-based ligands such as BIAN-type ligands
50
or bipyridine- and phenanthroline-based ligands with sub-
stituents other than methyl to induce steric hindrance and
destabilize the neutral square-planar Pd(II) complexes. Thus,
ligands perhaps even more efficient than those considered in
this work could be identified.
L1
L2
L3
L4
L5
L6
10
L7
I
9
3
6
3
8
9
51
II
0
64
0
62
0
53
0
35
0
65
0
42
0
52
III
IV
60
59
58
64
58
70
69
Solvent Effect. In the calculated reaction route, the lowest
energy minima preceding both decarboxylation and carbopalla-
dation are neutral complexes, while the respective TSs are
cationic. Such charge separation should be facilitated in polar
solvents. Although a polar reaction medium using the reactant
acetonitrile as solvent was used, the reaction should be further
facilitated by using a solvent with increased dielectric constant.
This was investigated by calculating the reaction profile using
water as solvent. Since L1 was calculated to have the highest
energy requirements, this ligand was used when calculating the
effect from the solvent on the free energy profile.
TS-I
120
82
118
87
110
84
109
79
118
77
114
84
116
84
V
VI
77
64
59
59
61
63
76
VII
−48
26
−35
21
−34
22
−29
18
−39
19
−24
25
−26
27
VIII
TS-II
99
94
99
92
94
97
98
IX
52
47
60
53
48
56
68
X
16
22
18
19
10
21
36
II + 2
−26
−66
120
147
−26
−66
118
128
−26
−66
110
132
−26
−66
109
121
−26
−66
118
133
−26
−66
114
121
−26
−66
116
124
II + 3
ΔGdecarboxylation
ΔGcarbopalladation
A comparison of the reaction profiles using acetonitrile (ε =
52
3
7.5) and water (ε = 80.37) is shown in Figure 6. As
expected, the energy required for both decarboxylation and
carbopalladation were calculated to be lower when replacing
acetonitrile with the more polar solvent water in the solvation
model. The cationic complexes, including the TSs, are
stabilized relative to the neutral complexes, which resulted in
a reduced energy requirement for decarboxylation and
Following TS-I the neutral complex VII was identified as the
lowest energy complex for all ligands and, thus, the starting
point for carbopalladation. The ligand effect was found to be
significant in this complex, and L1 gave a significantly more
stable neutral complex in comparison to L3 and L4, which is in
line with the lack of sterically demanding methyl groups. The
same effect was seen with L2 and L5, albeit not as pronounced.
The required energy for carbopalladation was calculated to be
in the range 121−147 kJ mol and was calculated to be the
rate-determining catalytic step for all investigated ligands.
Overall the free energy required for carbopalladation, with the
exception of L3, seems to follow a trend where the higher
yielding ligands result in lower required energies for this
reaction step. Notably this trend is true also for the
decarboxylation, suggesting that both parts of the reaction are
favored by the same ligand features. These results suggest that
−1
carbopalladation by 23 and 34 kJ mol , respectively. Although
the energies for both TSs relative to the lowest preceding
complex are reduced, carbopalladation is still 26 kJ mol−
higher in energy, which should provide confidence that
carbopalladation is the more energetically demanding catalytic
step of the two.
1
−1
The results from the theoretical investigation suggest that the
solvent has a large effect on the transformation. From these
results we hypothesized that the poor experimental result when
employing L1 as ligand in the original reaction protocol could
Figure 6. Free energy profile for the reaction using different solvents and L1 as ligand. The black line represents the energy using acetonitrile as
reaction media, and the red line represents that using water. Complex structures are the same as in Figures 1 and 2, with the exception that, when
using water as solvent, acetate in complex II has been replaced with deprotonated 1 in the low-energy path.
4
95
dx.doi.org/10.1021/om3009525 | Organometallics 2013, 32, 490−497

Organometallics
Article
be the high energy requirement for carbopalladation (Table 1
and Figure 5) and that performing the process in solvent
systems with increased water concentration would be beneficial
under those reaction conditions.
In order to test this hypothesis and to support the
calculations, an experimental investigation was performed
with L1 as ligand in which the water concentration was varied.
First, the effect on yield was investigated by performing four
reactions with increasing concentrations of water correspond-
ing to an increasing water/acetonitrile ratio in the reaction
vessel, shown in Table 3. The results show a clear trend of
increasing yield with increased concentration of water up to
reaction media as well as the limited solubility of the benzoic
acid at lower temperatures in more aqueous solutions. After this
initial point, however, the reaction with higher water
concentration clearly proceeded at a higher rate than the
other, which supports a reduced energy requirement in the rate-
determining step.
CONCLUSIONS
■
In summary, palladium(II)-catalyzed decarboxylative addition
of 2,6-dimethoxybenzoic acid to acetonitrile was investigated by
means of DFT calculations. The results suggest that
carbopalladation is the rate-determining step of the reaction.
The beneficial effect from the methyl groups in dimethylated
ligands was attributed to destabilization of the neutral square-
planar complexes preceding the decarboxylation and carbo-
palladation TSs, with a more marked effect in the latter reaction
step. A significant solvent effect was seen in the calculations,
which predicted reduced energy requirements for both
decarboxylation and carbopalladation in water and experimental
investigations were performed to support the calculations. The
experimental results show that increased concentration of water
indeed increases the effectiveness and rate of the reaction when
employing 2,2′-bipyridine and thus partially reduces the
dependency on steric destabilization from the ligands in the
reaction.
49% yield, which is a yield 3 times higher than that of the
original protocol.
Table 3. Effect of Increased Water Concentrations on the
Yield of the Reaction
a
entry
amt of H O (μL)
yield (%)
2
1
2
3
4
200
600
15
27
34
49
1100
1600
a
Isolated yields.
Furthermore, the predicted lower energy requirement of the
rate-determining carbopalladation when using water as solvent
should correspond to an increased reaction rate. This led us to
conduct a basic kinetics experiment, measuring the conversion
into product over time using two different reaction setups with
different amounts of water in the reaction media. For this
experiment the two extremes from Table 3, entries 1 and 4,
were selected and the total process time divided into six time
points. See the Supporting Information for experimental details.
The results are shown in Figure 7. The normalized GC-MS
response is plotted against reaction time, not including time for
heating to the desired temperature. Thus, the higher
concentration of product after 5 min for the reaction carried
out in a lower water concentration can possibly be attributed to
the longer heating time before reaching the set point of that
ASSOCIATED CONTENT
■
*
S
Supporting Information
Text, figures, and tables giving geometries and energies of the
presented complexes, experimental procedures for synthesis
and kinetic studies, and heating profiles. This material is
available free of charge via the Internet at http://pubs.acs.org.
AUTHOR INFORMATION
■
*
Corresponding Author
E-mail: Christian.Skold@orgfarm.uu.se.
Notes
The authors declare no competing financial interest.
Figure 7. Results from the kinetic study of the reaction using different amounts of water, 200 μL (Table 2, entry 1) in plot A and 1600 μL (Table 2,
entry 4) in plot B, at 130 °C for 5, 10, 15, 20, 25, and 30 min, respectively. The GC-MS response is normalized with the response of the internal
standard (2-methylnaphthalene, IS) plotted against time (heating time not included). The desired methyl ketone is denoted by solid lines and the
decarboxylated byproduct 1,3-dimethoxybenzene by dotted lines.
4
96
dx.doi.org/10.1021/om3009525 | Organometallics 2013, 32, 490−497

Organometallics
Article
(
(
(
(
(
̈
35) Jaguar; Schrodinger LLC, New York, 2011.
ACKNOWLEDGMENTS
■
36) Becke, A. D. J. Chem. Phys. 1993, 98, 5648−5652.
37) Becke, A. D. J. Chem. Phys. 1993, 98, 1372−1377.
We thank Prof. Per-Ola Norrby and Dr. Luke Odell for fruitful
discussions and Dr. Wesley Schaal for linguistic revision. M.L.
and R.S.M. are grateful to Erasmus Mundus (EXPERTS
Programme) for a Postdoctorate Fellowship. M.L. thanks the
Swedish Research Council.
38) Lee, C.; Yang, W.; Parr, R. G. Phys. Rev. B 1988, 37, 785−789.
39) Hay, P. J.; Wadt, W. R. J. Chem. Phys. 1985, 82, 299−310.
(40) Sousa, S. F.; Fernandes, P. A.; Ramos, M. J. J. Phys. Chem. A
2007, 111, 10439−10452.
(41) Ahlquist, M.; Fabrizi, G.; Cacchi, S.; Norrby, P.-O. J. Am. Chem.
Soc. 2006, 128, 12785−12793.
42) Henriksen, S. T.; Tanner, D.; Skrydstrup, T.; Norrby, P.-O.
Chem. Eur. J. 2010, 16, 9494−9501.
43) Marten, B.; Kim, K.; Cortis, C.; Friesner, R. A.; Murphy, R. B.;
Ringnalda, M. N.; Sitkoff, D.; Honig, B. J. Phys. Chem. 1996, 100,
1775−11788.
44) Tannor, D. J.; Marten, B.; Murphy, R.; Friesner, R. A.; Sitkoff,
REFERENCES
1) Garves, K. J. Org. Chem. 1970, 35, 3273−3275.
2) Zhou, C.; Larock, R. C. J. Am. Chem. Soc. 2004, 126, 2302−2303.
3) Zhou, C.; Larock, R. C. J. Org. Chem. 2006, 71, 3551−3558.
4) Hsieh, J.-C.; Chen, Y.-C.; Cheng, A.-Y.; Tseng, H.-C. Org. Lett.
■
(
(
(
(
(
(
1
(
2
(
012, 14, 1282−1285.
5) Goossen, L. J.; Rudolphi, F.; Oppel, C.; Rodríguez, N. Angew.
Chem., Int. Ed. 2008, 47, 3043−3045.
6) Ueura, K.; Miyamura, S.; Satoh, T.; Miura, M. J. Organomet.
Chem. 2006, 691, 2821−2826.
7) Wong, Y.-C.; Parthasarathy, K.; Cheng, C.-H. Org. Lett. 2010, 12,
736−1739.
8) Zhao, B.; Lu, X. Tetrahedron Lett. 2006, 47, 6765−6768.
9) Lindh, J.; Sjoberg, P. J. R.; Larhed, M. Angew. Chem., Int. Ed.
010, 49, 7733−7737.
10) Behrends, M.; Sav
Catal. 2011, 1, 1455−1459.
11) Liu, J.; Zhou, X.; Rao, H.; Xiao, F.; Li, C.-J.; Deng, G.-J. Chem.
D.; Nicholls, A.; Honig, B.; Ringnalda, M.; Goddard, W. A. J. Am.
Chem. Soc. 1994, 116, 11875−11882.
(
(45) Grimme, S.; Antony, J.; Ehrlich, S.; Krieg, H. J. Chem. Phys.
2
(
010, 132, 154104−19.
(
1
(
(
2
(
46) Marothy, S. A. d. XYZ-viewer, version 0.970, Stockholm, 2010.
(47) McMullin, C. L.; Jover, J.; Harvey, J. N.; Fey, N. Dalton Trans.
2
010, 39, 10833−10836.
48) Fristrup, P.; Ahlquist, M.; Tanner, D.; Norrby, P.-O. J. Phys.
Chem. A 2008, 112, 12862−12867.
49) Ahlquist, M.; Kozuch, S.; Shaik, S.; Tanner, D.; Norrby, P.-O.
Organometallics 2005, 25, 45−47.
50) Van Asselt, R.; Elsevier, C. J. Organometallics 1992, 11, 1999−
001.
51) Sjo
Cucciolito, M. E.; Vitagliano, A. Organometallics 1992, 11, 3954−3964.
52) Values as used in Jaguar 7.8.
̈
(
̈ ̈
marker, J.; Sjoberg, P. J. R.; Larhed, M. ACS
(
(
(
2
Eur. J. 2011, 17, 7996−7999.
(
(
5
(
12) Miao, T.; Wang, G.-W. Chem. Commun. 2011, 47, 9501−9503.
13) Rodriguez, N.; Goossen, L. J. Chem. Soc. Rev. 2011, 40, 5030−
(
̈
gren, M.; Hansson, S.; Norrby, P. O.; Åkermark, B.;
048.
(
14) Smith, M. B.; March, J. March’s Advanced Organic Chemistry:
Reactions, Mechanisms, and Structure. 6th ed.; Wiley: Hoboken, NJ,
007; pp 745−747.
15) Gooßen, L. J.; Deng, G.; Levy, L. M. Science 2006, 313, 662−
64.
16) Gooßen, L. J.; Thiel, W. R.; Rodríguez, N.; Linder, C.; Melzer,
B. Adv. Synth. Catal. 2007, 349, 2241−2246.
17) Dickstein, J. S.; Mulrooney, C. A.; O’Brien, E. M.; Morgan, B. J.;
Kozlowski, M. C. Org. Lett. 2007, 9, 2441−2444.
18) Goossen, L. J.; Linder, C.; Rodriguez, N.; Lange, P. P.; Fromm,
A. Chem. Commun. 2009, 7173−7175.
19) Dupuy, S.; Lazreg, F.; Slawin, A. M. Z.; Cazin, C. S. J.; Nolan, S.
2
(
6
(
(
(
(
P. Chem. Commun. 2011, 47, 5455−5457.
(
20) Sun, Z.-M.; Zhang, J.; Zhao, P. Org. Lett. 2010, 12, 992−995.
(
21) Gooßen, L. J.; Zimmermann, B.; Knauber, T. Angew. Chem., Int.
Ed. 2008, 47, 7103−7106.
22) Gooßen, L. J.; Zimmermann, B.; Knauber, T. Angew. Chem.
008, 120, 7211−7214.
23) Gooßen, L. J.; Rodríguez, N.; Lange, P. P.; Linder, C. Angew.
Chem. 2010, 122, 1129−1132.
24) Goossen, L. J.; Rodríguez, N.; Melzer, B.; Linder, C.; Deng, G.;
Levy, L. M. J. Am. Chem. Soc. 2007, 129, 4824−4833.
25) Goossen, L. J.; Rodríguez, N.; Lange, P. P.; Linder, C. Angew.
(
2
(
(
(
Chem., Int. Ed. 2010, 49, 1111−1114.
(
(
26) Xue, L.; Su, W.; Lin, Z. Dalton Trans. 2010, 39, 9815−9822.
27) Zhang, S.-L.; Fu, Y.; Shang, R.; Guo, Q.-X.; Liu, L. J. Am. Chem.
Soc. 2009, 132, 638−646.
28) Sraj, L. O. C.; Khairallah, G. N.; da Silva, G.; O’Hair, R. A. J.
Organometallics 2012, 31, 1801−1807.
29) Chuchev, K.; BelBruno, J. J. J. Mol. Struct. (THEOCHEM) 2007,
(
(
8
(
07, 1−9.
30) Ruelle, P. J. Chem. Soc., Perkin Trans. 2 1986, 1953−1959.
31) Li, J.; Brill, T. B. J. Phys. Chem. A 2003, 107, 2667−2673.
32) Ruelle, P. J. Comput. Chem. 1987, 8, 158−169.
(
(
(
33) Shi, J.; Huang, X.-Y.; Wang, J.-P.; Li, R. J. Phys. Chem. A 2010,
1
(
14, 6263−6272.
34) Hansson, S.; Norrby, P. O.; Sjo
Cucciolito, M. E.; Giordano, F.; Vitagliano, A. Organometallics 1993,
2, 4940−4948.
̈
gren, M. P. T.; Åkermark, B.;
1
4
97
dx.doi.org/10.1021/om3009525 | Organometallics 2013, 32, 490−497