Catalytic non-conventional trans-hydroboration: A theoretical and experimental perspective

Article abstract of DOI:10.1002/chem.201102729

We have studied the nonconventional trans-hydroboration reaction of alkynes both experimentally and theoretically. A catalytic system based on the in situ mixture of [{Rh- (cod)Cl}₂]/PCy₃ (cod=1,5-cyclooctadiene, Cy=cyclohexyl) has been able to activate pinacolborane and catecholborane and transfer boryl and hydride groups onto the same unhindered carbon atom of the terminal alkynes. The presence of a base (Et₃N) favored the non-conventional trans-hydroboration over the traditional cis-hydroboration. Varying the substrate had a significant influence on the reaction, with up to 99% conversion and 94% regioselectivity observed for para-methyl- phenylacetylene. Both DFT and quantum mechanical/molecular mechanical ONIOM calculations were carried out on the [RhCl(PR₃)₂] system. To explain the selectivity towards the (Z)-alkenylboronate we explored several alternative mechanisms to the traditional cishydroboration, using propyne as a model alkyne. The proposed mechanism can be divided into four stages: 1) isomerization of the alkyne into the vinylidene, 2) oxidative addition of the borane reagent, 3) vinylidene insertion into the Rh-H bond, and finally 4) reductive elimination of the C-B bond to yield the 1-alkenylboronate. Calculations indicated that the vinylidene insertion is the selectivity- determining step. This result was consistent with the observed Z selectivity when the sterically demanding phosphine groups, such as PCy₃ and PiPr₃, were introduced. Finally, we theoretically analyzed the effect of the substrate on the selectivity; we identified several factors that contribute to the preference for aryl alkynes over aliphatic alkynes for the Z isomer. The intrinsic electronic properties of aryl substituents favored the Z-pathway over the E-pathway, and the aryl groups containing electron donating substituents favored the occurrence of the vinylidene reaction channel.

Full text of DOI:10.1002/chem.201102729

DOI: 10.1002/chem.201102729
Catalytic Non-Conventional trans-Hydroboration: A Theoretical and
Experimental Perspective
Jessica Cid, Jorge J. Carbꢀ,* and Elena Fernꢁndez*[a]
Abstract: We have studied the non-
conventional trans-hydroboration reac-
tion of alkynes both experimentally
and theoretically. A catalytic system
based on the in situ mixture of [{Rh-
mechanical/molecular
ONIOM calculations were carried out
on the [RhCl(PR3)2] system. To explain
the selectivity towards the (Z)-alkenyl-
boronate we explored several alterna-
tive mechanisms to the traditional cis-
mechanical
yield the 1-alkenylboronate. Calcula-
tions indicated that the vinylidene in-
sertion is the selectivity-determining
step. This result was consistent with the
observed Z selectivity when the steri-
cally demanding phosphine groups,
such as PCy3 and PiPr3, were intro-
duced. Finally, we theoretically ana-
lyzed the effect of the substrate on the
selectivity; we identified several factors
that contribute to the preference for
aryl alkynes over aliphatic alkynes for
the Z isomer. The intrinsic electronic
properties of aryl substituents favored
the Z-pathway over the E-pathway,
and the aryl groups containing electron
donating substituents favored the oc-
currence of the vinylidene reaction
channel.
AHCTUNGTRENNUNG
A
(cod=1,5-cycloocta-
diene, Cy=cyclohexyl) has been able
to activate pinacolborane and catechol-
borane and transfer boryl and hydride
groups onto the same unhindered
carbon atom of the terminal alkynes.
The presence of a base (Et3N) favored
the non-conventional trans-hydrobora-
tion over the traditional cis-hydrobora-
tion. Varying the substrate had a signif-
icant influence on the reaction, with up
to 99% conversion and 94% regiose-
lectivity observed for para-methyl-phe-
nylacetylene. Both DFT and quantum
hydroboration, using propyne as a
model alkyne. The proposed mecha-
nism can be divided into four stages:
1) isomerization of the alkyne into the
vinylidene, 2) oxidative addition of the
borane reagent, 3) vinylidene insertion
À
into the Rh H bond, and finally 4) re-
À
ductive elimination of the C B bond to
Keywords: boranes · density func-
tional calculations · homogeneous
catalysis · hydroboration · rhodium
Introduction
actions, but always with the formation of cis-hydroboration
products.[6] However, Miyaura and co-workers reported the
first rhodium- and iridium-catalyzed non-conventional trans-
hydroboration reaction to yield (Z)-1-alkenylboronates.[7]
They postulated that at-least two dominant factors could re-
verse the conventional cis-hydroboration to prefer the trans-
hydroboration: 1) the presence of Et3N as an additive and
2) the use of an excess of the alkyne over the borane re-
agent, (Scheme 1). The lack of further studies on the non-
conventional trans-hydroboration approach prompted us to
study its mechanism from an experimental and a theoretical
point of view in order to establish the basis of the desired
selectivity in the trans-hydroboration reaction. Thus, we ex-
plored the influence of the ligand in modifying the metal
species and the influence of reaction conditions in order to
make general the methodology. Furthermore, the mecha-
nism proposed by Miyaura and co-workers has been investi-
gated using density functional theory (DFT) calculations,
which focused on the origin of the trans-selectivity.
The hydroboration of alkynes is a useful method for the syn-
thesis of 1-alkenylboronate compounds, which are versatile
intermediates in organic synthesis.[1] The conventional cis-
hydroboration reaction of alkynes with catecholborane
(HBcat),[2] pinacolborane (HBpin),[3] and 4,4,6-trimethyl-
1,3,2-dioxaborinane[4] to yield (E)-1-alkenylboronates was
first studied through stoichiometric syn-addition approaches.
The activation of dialkoxyboranes by transition-metal com-
plexes opened up a new perspective on hydroboration reac-
tions when Kono et al.[5] demonstrated that catecholborane
could be oxidatively added to the rhodium(I) center in
[RhCl
A
N
ACHTUNGTREN(NUNG III) complex [RhClH-
boration has been shown to alter the chemo-, regio-, and
diastereoselectivity of the stoichiometric hydroboration re-
[a] J. Cid, Dr. J. J. Carbꢁ, Dr. E. Fernꢂndez
Departament de Quꢃmica Fꢃsica i Inorgꢄnica
Universitat Rovira i Virgili
C/Marcel.lꢃ Domingo s/n, 43007, Tarragona (Spain)
Fax : (+34)977-559563
Results and Discussion
E-mail: j.carbo@urv.cat
The first catalytic cis-hydroboration reaction was observed
on alkenes by Mꢀnnig and Nçth.[6a] They proposed a mecha-
nism for the rhodium-catalyzed hydroboration reaction that
mariaelena.fernandez@urv.cat
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.201102729.
1512
ꢅ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2012, 18, 1512 – 1521

FULL PAPER
alkynes. Elucidation by quantum mechanical methods com-
plemented the mechanistic studies on the rhodium-mediated
cis-hydroboration reaction, although, to the best of our
knowledge, the theoretical studies reported so far have only
considered alkene substrates.[6d,8] When Miyaura and co-
workers developed the first trans-hydroboration reaction of
alkynes, they observed that the b-hydrogen atom in the (Z)-
1-alkenylboronate unexpectedly did not derive from the
borane reagent, thus representing an example of non-con-
ventional trans-hydroboration. By means of a rhodium-
mediated hydroboration of [1-D]-1-octyne, they observed
that the deuterium-labeled terminal-carbon atom selectively
shifted onto the b-carbon atom. On the basis of this interest-
ing observation, the authors suggested a plausible mecha-
nism for both the acetylenic hydrogen migration and the
À
gem-addition of the B H bond. A rhodium complex modi-
fied with an electron-donating ligand could favor the oxida-
À
tive addition of the terminal C H bond on the substrate and
the resulting stable vinylidene complex might be the key in-
termediate. This process would be followed by the oxidative
addition of the borane and 1,2-migration of the boryl group
onto the a-carbon atom, thus providing the (Z)-1-alkenyl-
boronate via reductive elimination (Scheme 2, on bottom).
The presence of Et3N might suppress the cis-hydroboration
of the borane because a parallel experiment demonstrated
Scheme 1. Transition-metal-catalyzed cis- and trans-hydroboration reac-
tions.
À
involved the oxidative addition of a B H bond to the coor-
dinatively unsaturated metal center followed by alkene coor-
dination, insertion, and hydride migration onto the coordi-
nated alkene, with subsequent reductive elimination to form
that the treatment of [RhClHACHTUNGRTENNUG(Bcat)AHCTUNGTREN(NNGU PiPr3)2] with Et3N led
to the complete reductive elimination of catecholborane.[7]
Conventional trans-hydrosilylation of alkynes has been ex-
plained by the formation of carbene-type metal species as
the key intermediates.[9] To justify the observed unique ste-
reoselectivity, Ojima and co-workers[9a] postulated a silicon
shift onto the acetylenic bond and a zwitterionic carbene–
rhodium complex that undergoes isomerization (Scheme 3),
whereas Tanke and Crabtree[9b,c] proposed that the key in-
termediate is actually the closely related h2-vinyl complex
(Scheme 4). However, to the best of our knowledge, there
have been no reported examples of the conventional trans-
hydroboration of alkynes. To gain more insight into the
dominant mechanism in the cis- and non-conventional trans-
hydroboration reactions, we performed a series of rhodium-
and iridium-mediated hydroboration reactions of alkynes in
À
the B C bond. Scheme 2 (top) has adapted the catalytic
cycle of the cis-hydroboration reaction for the reaction with
Scheme 2. Suggested mechanistic cycles for cis-hydroboration and the
non-conventional trans-hydroboration reaction.
Scheme 3. Suggested catalytic cycle for the conventional rhodium-cata-
lyzed trans-hydrosilylation reaction.
Chem. Eur. J. 2012, 18, 1512 – 1521
ꢅ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
1513

J. J. Carbꢁ, E. Fernꢂndez et al.
gous catalytic system [{IrACHTNUTRGNE(UNG cod)Cl}2]/PCy3 was used instead, a
lower percentage of isomer 2 was achieved, to the benefit of
isomer 4 (Table 1, entries 7 and 8). Next, we examined the
influence of the phosphine ligands at a fixed substrate/
borane ratio of 1:1.2 in tetrahydrofuran. Under these reac-
tion conditions, several monophosphines, such as PACHTUNGTRENNUNG(nBu)3,
PMe3, and PPh3, and diphosphines 1,4-bis(diphenylphosphi-
no)butane (dppb) and 1,1’-bis(diphenylphosphino)ferrocene
(dppf) favored the formation of cis-hydroboration-products
3 and 4 more-strongly than monophosphine PCy3, with
isomer 3 being the major product (Table 1, entries 9–13).
When diphosphines were used in the catalytic system, the
activity was also lower, particularly when dppf was used as
the bidentate ligand (Table 1, entry 13). Because the modifi-
cation of the rhodium complex with monophosphine PCy3
Scheme 4. Suggested catalytic cycle for the conventional iridium-cata-
lyzed trans-hydrosilylation reaction.
provided the highest regioselectivity towards isomer
2
(Table 1, entry 3), we explored the influence of the Rh/PCy3
ratio and temperature on the reaction outcome. Figure 1 il-
lustrates the inverse relationship between activity and regio-
selectivity towards isomer 2 when the ratio of Rh/PCy3
changed from 1:1 to 1:2 to 1:4. The highest percentage of
product achieved for the non-conventional trans-hydrobora-
tion reaction was obtained at a Rh/PCy3 ratio of 1:4 and was
independently formed at 08C, 25 8C, and 708C. With all
these data in mind, we performed a study of the scope of
the substrates, taking into consideration different steric and
electronic properties. We first extrapolated the best reaction
conditions to promote the non-conventional trans-hydrobo-
ration reaction with similar aliphatic alkynes such as 1-hep-
tyne and 1-pentyne. In both cases, although the catalytic ac-
tivity was improved, the regioselectivity of isomer 2 de-
creased slightly (Table 2, entries 1 and 2). Steric demands on
the alkyne substituent favored the formation of the
parallel to DFT analysis. A catalytic system based on the in
situ mixture of [{RhACTHNUTRGNEUGN(cod)Cl}2]/PCy3 (cod=1,5-cycloocta-
diene, Cy=cyclohexyl) transformed the model substrate 1-
octyne into a mixture of isomers 2–4, depending on the sub-
strate/borane ratio. Isomer 2 was expected to be formed via
the non-conventional trans-hydroboration pathway, whilst
isomers 3 and 4 are the two regioisomers formed from the
cis-hydroboration reaction. The alkyne/borane ratio seems
to play a role in determining the regioselectivity for isomer
2 (Table 1, entries 1–3). In this case, the optimal result was
obtained when the pinacolborane reagent (1a) was present
in a slight excess. Apart from tetrahydrofuran, other sol-
vents like cyclohexane and dichloromethane were explored
without any significant improvement in the regioselectivity
towards product 2, (Table 1, entries 4–6). When the analo-
desired product in the case of tert-butyl acetylene
but not in the case of cyclohexyl acetylene (Table 2,
Table 1. Base-assisted metal-catalyzed hydroboration of 1-octyne.[a]
entries 3 and 4). The influence of electronic effects
on the rhodium-catalyzed hydroboration reaction
was principally observed when electron-withdraw-
ing and electron-donating phenyl acetylenes were
transformed into the alkenyl boronate products.
Entry
Catalytic system
Solvent
Substrate/
borane
Conv.
[%][b]
2[c]
3[c]
4[c]
Regioselectivities of up to 94% for isomer 2 were
achieved when electron-rich alkynes were involved
(Table 2, entries 5–7). On the contrary, despite
being electron poor, para-trifluoromethyl-phenyla-
cetylene was converted into the product in almost
quantitative yield, but the cis-hydroboration was
very competitive. Other electron-rich alkynes with
high steric demands were also mainly converted
into isomer 2 (Table 2, entries 8 and 9). It is impor-
tant to note that despite the successful hydrobora-
tion of the para-trifluoromethyl-phenylacetylene, all
of the other substrates provided negligible or no
formation of isomer 4, owing to the constrained ad-
dition of the boryl moiety onto the most-hindered
carbon in the cis-hydroboration reaction. Interest-
ingly enough, when the rhodium-mediated hydrobo-
ration of para-methyl-phenyl acetylene was carried
1
2
3
4
5
6
7
8
[{Rh
[{Rh
[{Rh
[{Rh
[{Rh
[{Rh
N
THF
THF
THF
C6H12
C6H12
CH2Cl2
THF
THF
THF
THF
THF
THF
THF
1.2:1
1:1
1:1.2
1:1
1:1.2
1:1
1:1
1:1.2
1:1.2
1:1.2
1:1.2
1:1.2
1:1.2
47
87
73
99
88
71
98
63
87
98
89
64
39
38
60
66
48
65
58
53
58
30
18
9
47
34
30
46
28
34
37
33
47
56
62
57
47
15
6
4
6
7
8
9
9
23
26
29
22
20
[{Ir
[{Ir
U
9
[{Rh
[{Rh
[{Rh
[{Rh
[{Rh
N
10
11
12
13
ACHTUNGTRENNUNG
21
33
[a] Standard conditions: [{MACTHNURTGNE(NUG cod)Cl}2]/L (M=Rh or Ir, 0.015 mmol), monophosphine
(0.06 mmol), diphosphine (0.03 mmol), Et3N (5 mmol), pinacolborane (1.2 mmol), 1-
octyne (1 mmol) for substrate/borane ratio=1:1.2. Solvent (3 mL), 258C, 4 h. [b] Con-
version determined by GC analysis of the consumption of 1-octyne. [c] Percentage of
isomeric ratio, determined by GC analysis.
1514
www.chemeurj.org
ꢅ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2012, 18, 1512 – 1521

Catalytic Non-Conventional trans-Hydroboration
FULL PAPER
sults from real systems in order
to asses several key variables,
such as ligand and substrate
structure. We also assumed that
the two phosphine ligands coor-
dinated trans to each other, as
observed in the X-ray structure
of the key vinylidene complex
[trans-ACTHNUGRTNENUG(PiPr)2RhClACHTUNTGREN(NUGN =C=
CHMe)].[11] Initially, we exam-
ined the mechanism proposed
by Miyaura and co-workers
(Scheme 2, bottom); the first
step involved the isomerization
of the terminal alkyne group
into the vinylidene in the coor-
dination sphere of rhodium
Figure 1. Influence of the Rh/PCy3 ratio and temperature on the catalytic hydroboration of 1-octyne with pina-
colborane.
complex. This process has been
observed previously[11,12] and
extensively studied by computa-
tional methods.[13] The precise mechanism of iso-
merization had been the subject of some controver-
sy, but in the end all evidence pointed toward a un-
imolecular process involving the formation of an al-
kynyl hydride intermediate (Scheme 6). The second
step, 1,3-hydrogen shift to yield the vinylidene, was
postulated to be the rate-limiting step.[14] Angelis
et al.[13c] and Grotjahn et al.[13d] have demonstrated
that electron-donating ligands favor the formation
of both hydrido-alkynyl and vinylidene species. We
have also observed this trend in other transition-
metal complexes,[15] adding that the steric hindrance
of bulky ligands has an influence through the desta-
bilization of the alkyne.[15a] We noted that the non-
conventional trans-hydroboration was observed for
bulky and strongly electron-donating ligands, such
as PCy3 and PiPr3. Thus, in this first part of the dis-
cussion, we will assume that isomerization takes
place and we will focus on the addition of borane
to the vinylidene complex. Scheme 7 shows the
computed energy profile for the Z- and E-pathways
of the borane addition to the vinylidene group,
starting with boryl migration (mechanism A). The
vinylidene complex (7) was calculated to be 1.7 kcal
molÀ1 lower in energy than the alkyne complex
Table 2. Rhodium-catalyzed hydroboration of alkynes with pinacolborane.[a]
Entry
Substrate
Conv.
[%][b]
2[c]
3[c]
4[c]
1
2
80
65
65
67
33
31
2
2
3
4
5
6
7
84
88
95
75
76
74
51
48
90
94
26
49
32
10
6
0
0
18
0
0
8
9
72
44
77
78
23
22
0
0
[a] Standard conditions: [{RhACHTUNRGTNEUNG(cod)Cl}2]/L (0.015 mmol), PCy3 (0.12 mmol), Et3N
(5 mmol), pinacolborane (1.2 mmol), alkyne (1 mmol), substrate/borane ratio=1:1.2,
THF (3 mL), 258C, 4 h. [b] Conversion determined by GC analysis of the consumption
of the alkyne. [c] Percentage of isomeric ratio, determined by GC analysis.
out with catecholborane (1b) as the borane reagent, the re-
gioselectivity towards isomer 2 remained very high whilst
the conversion increased significantly (Scheme 5). It is
known that borane reagents derived from catechol units are
more reactive owing to their enhanced Lewis acid proper-
ties.[10] In order to gain a greater insight into the mechanistic
details of the trans-hydroboration reaction, a detailed theo-
retical study of the rhodium-mediated reaction of alkynes
was performed. In our first approach to the mechanism, we
used simplified model systems: PH3 for phosphine ligands,
[RhClACHTGNUTRENU(NG PH3)2ACHTUNGTREN(NUNG HCCMe)] (5). However, at a similar computa-
tional level, the preference for the vinylidene group was re-
HB
ACHTUNGTRENNUNG(O2C2H4) for the pinacolborane reagent, and the sim-
Scheme 5. Optimized trans-hydroboration reaction of alkynes with cate-
ꢀ
plest alkyne substrate, HC CMe. We also considered the re-
cholborane.
Chem. Eur. J. 2012, 18, 1512 – 1521
ꢅ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
1515

J. J. Carbꢁ, E. Fernꢂndez et al.
complex, in which not all of the conceivable hydrido-boryl
species and none of the transition states for oxidative addi-
tion were located.[8] We located the transition state connect-
ing intermediates 8A to vinyl complexes 9A (TSA8–9), which
corresponded to the 1,2-boryl migration onto the a-carbon
atom of the vinylidene group. This migration occurs within
Scheme 6. Mechanism for alkyne–vinylidene isomerization in electron-
rich metal (e.g., rhodium) centers. Relative energies shown are in kcal
À
the vinylidene molecular plane, and the distance of the B C
molÀ1
.
forming bond is 1.89 ꢆ at the pro-Z transition state. The dis-
tance between the boron atom and the hydride moiety is
À
relatively long (2.58 ꢆ), thereby showing that the B H bond
is cleaved when boryl migration takes place rather than the
concerted addition of borane to the vinylidene fragment.
The key geometric parameters at the corresponding pro-E
transition state only differ by less than 0.01 ꢆ. Accordingly,
we found that the two transition states for the Z- and E-
pathways lay 30.2 and 28.3 kcalmolÀ1 above compound 7, re-
spectively. These overall energy barriers seem somewhat
high for a process occurring at room temperature. The prod-
uct of boron migration (9A) is 21 kcalmolÀ1 more stable
than compound 3. Subsequent reductive elimination with
À
concomitant C H bond-formation leads to the borylated-
alkene product via transition state TSA9–10 with low energy
barriers (ca. 4 kcalmolÀ1). For this last part of the mecha-
nism, the energy differences between the Z and E paths are
within 1 kcalmolÀ1. Alternatively, the sequence of the reac-
tion could be inversed, that is, hydride migration followed
À
by reductive elimination of the C B bond (mechanism B).
Scheme 8 shows the computed energy profile for both the Z
and E pathways, and Figure 2 shows the structures and main
geometric parameters of selected intermediates and transi-
tion states. In this case, it was possible to locate the two hy-
drido-boryl intermediates (8B) and the transition states for
À
the oxidative addition of the B H bond to the rhodium
center (TSB7–8). The formation of complexes 8B is endother-
mic by 20 kcalmolÀ1 and proceeds via transition-state struc-
tures that have similar energies, leading to very-low reverse
energy barriers (<3 kcalmolÀ1). From intermediates 8B, the
forward energy barriers for hydride migration via TSB8–9 are
also very low (ca. 4 kcalmolÀ1). These low energy barriers
mean that the existence of intermediate 8 may depend on
the specific conditions. Among them, having a chloride
atom trans to the hydride (8B) instead of trans to the boryl
ligand (8A) stabilizes the intermediate. Nevertheless, the im-
portant fact is that the reaction goes uphill from compound
7 to reach the transition state for hydride migration (TSB8–
9), and the calculated overall barriers are 25.9 and 25.4 kcal
molÀ1 for the Z and E pathways, respectively. Interestingly,
these values are significantly lower than those corresponding
to mechanism A. In the TSB8–9 structures (Figure 2), the vi-
nylidene ligand bends away from its position in the reactants
(8B) to approach the hydride group (Cl-Rh-Ca bond angle
changes from 1138 in 8B to 1338 in TSB8–9 for both the Z
and E pathways), thereby ending up trans to the chloride
atom in the products (9B). In this migratory insertion, the
Scheme 7. A calculated potential-energy profile (kcalmolÀ1
mechanism suggested by Miyaura and co-workers involving an initial iso-
merization of the alkyne to the vinylidene (mechanism A). Hydrobora-
) for the
tion of propyne with HBACHTNURTGENN(UG O2C2H4) catalyzed by a [RhClAHCTUNGTREN(NUGN PH3)2] complex.
Solid lines correspond to the Z path (R1 =Me), dashed lines correspond
to the E path (R2 =Me).
ported to increase to up to approximately 13 kcalmolÀ1
using real PiPr3 instead of model PH3 phosphines.[13b] This
result further supports the postulation that the process is
thermodynamically favorable and essentially irreversible, as
demonstrated in previous experimental studies.[13a] This pro-
cess should be followed by oxidative addition of the borane
agent to compound 7 to yield a hydrido-boryl intermediate
(8A) with the boryl group cis to the vinylidene ligand. De-
spite all of our efforts, we could not locate either species 8A
nor the transition state for oxidative addition. Following the
previous discussion about the effect of strongly donating li-
gands,[13b,c,15] it is reasonable to think that the species 8A
would be stabilized by basic PCy3 phosphine groups and
that the same effect cannot be achieved by simplified PH3 li-
gands. A similar situation was faced in previous theoretical
À
formation of the C H bond is co-planar with the vinylidene
moiety. Once the transition state for migratory insertion is
reached, the reaction drops significantly in energy (ca.
studies on the hydroboration of alkenes using a [RhClACTHNUTRGNE(UNG PH3)]
1516
www.chemeurj.org
ꢅ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2012, 18, 1512 – 1521

Catalytic Non-Conventional trans-Hydroboration
FULL PAPER
substituent and the rhodium fragment trans to one other, is
1.4 kcalmolÀ1 thermodynamically more stable than the pro-
Z intermediate. Notably, the difference in energy between
the E and Z paths has increased on going from TSB8–9 to 9B
in favor of the E path. However, the high reverse barriers
for intermediates 9B prevent sterospecific product forma-
tion from being achieved thermodynamically, provided that
the two intermediates 9B do not interconvert (see below).
Finally, the reaction proceeds by reductive elimination of
À
the C B bond to give the product via TSB9–10. Again, the
barrier for the termination step was calculated to be very
low, only 2.4 kcalmolÀ1 for the Z pathway.[16] By analogy
with the proposed mechanisms for trans-hydrosilylation, we
explored a new mechanism that was consistent with the ob-
served trans-hydroboration products. The reaction could
À
proceed via migratory insertion of the alkyne into the Rh B
bond leading to a complex in which the boryl and alkyl sub-
stituents of the vinyl ligand were trans to one another. Then,
À
prior to the reductive C H coupling step, the pro-Z inter-
À
mediate could be formed via vinylic C C rotation
(Scheme 9). In addition, these calculations will evaluate the
feasibility of interconversion between the Z and E isomers
in intermediate 9B. The attempts to obtain the zwitterionic
carbene/rhodium complex separately, as proposed by Ojima
et al.[9a] (Scheme 3), and the related h2-vinyl complex pro-
posed by Crabtree and co-workers[9b–e] (Scheme 4) ended up
with the same structure, which resembles the proposal by
Crabtree and co-workers (Scheme 9). This species was char-
acterized as a transition state by a single imaginary frequen-
cy, whose normal mode corresponds to the rotation of the
Scheme 8. A calculated potential-energy profile (kcalmolÀ1) for a variant
on the mechanism suggested by Miyaura, in which hydrogen migration
onto the vinylidene carbon atom occurs first (mechanism B). Hydrobora-
tion of propyne with HBACHTNURTGENN(UG O2C2H4) catalyzed by a [RhClAHCTUNGTREN(NUGN PH3)2] complex.
Solid lines correspond to the Z path (R1 =Me), dashed lines correspond
to the E path (R2 =Me).
À
vinylic C C carbon bond. The computed energy barrier was
not very high (20.5 kcalmolÀ1), but it was significantly
higher (ca. 5 kcalmolÀ1) than of the reductive elimination of
À
the C H bond. Obviously, these results do not necessarily
preclude the proposed mechanism for the hydrosilylation of
À
alkynes, but they discard the C C rotation in the trans-hy-
droboration reaction catalyzed by phosphine-modified rho-
dium complexes. Moreover, they indicate that the Z and E
isomers of intermediates 9B would not readily interconvert
into one other, leading to the thermodynamically more-
stable E species. With all of these results in hand, we can
consider which could be the most-plausible selectivity-deter-
mining step. Although the energy profiles might be tuned by
the inclusion of real ligands, some features are already clear.
Comparing mechanism A and B, we observed the same en-
ergetic pattern: the reaction goes uphill in energy until the
boryl- and hydride-migration transition states, respectively.
The resulting rhodium/vinyl complex is low in energy, and
has a very high reverse barrier and a low barrier to give the
product. This pattern indicates that both boryl and hydride
migration are irreversible steps to give the product as de-
fined Z and E stereoisomers. Because we have shown that
the Z and E isomers do not interconvert easily, we can state
the either the boryl or hydride migration is the selectivity-
determining step. As a consequence, the selectivity of the
process is determined by kinetic control and can be de-
scribed by applying transition-state theory (TST). The hy-
Figure 2. Molecular structures and main geometric parameters of transi-
tion state TSB8–9 and intermediate 9B in the pro-E and pro-Z propyne-
hydroboration pathways using PH3 ligands. Distances in ꢆ and relative
energies in kcalmolÀ1
.
50 kcalmolÀ1) to give the rhodium/vinyl intermediates (9B),
which are 24–25 kcalmolÀ1 lower in energy than compound
7. The pro-E intermediate 9B, which contains the alkyne
Chem. Eur. J. 2012, 18, 1512 – 1521
ꢅ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
1517

J. J. Carbꢁ, E. Fernꢂndez et al.
transition state, points away
from the metal center. Thus, it
is reasonable to think that in-
creasing the bulk of the phos-
phine or alkyne substituent will
increase the preference for the
Z isomer by destabilizing the E
path. Next, we analyzed the
origin of substrate effects on
the reaction outcome. To under-
stand the higher selectivity ob-
served for aryl alkynes com-
pared with aliphatic alkynes, we
considered three different as-
pects: 1) the intrinsic prefer-
ence of the alkyne substituents
in the Z or E pathways, 2) the
higher sensitivity to the steric
hindrance of the phosphine
À
Scheme 9. Schematic representation of an alternative trans-hydroboration reaction through vinylic C C bond-
rotation; this mechanism is analogous to those proposed for the trans-hydrosilylation reaction. The transition
state involves an h2 interaction of the vinylic fragment with the rhodium center. The hydride reductive-elimi-
nation cis mechanism is shown for comparison. Energies shown are in kcalmolÀ1
.
dride migration in mechanism B has the lowest energy barri-
er, and consequently, it is the most likely. Moreover, as we
will discuss below, mechanism B is fully consistent with the
Table 3. Calculated relative energies of pro-E and pro-Z transition states
for different type of phosphines. Comparison with the experimental re-
sults.
Entry
Substrate
Phosphine
TSB8À9
Z/E[c]
expected selectivity. Thus, we focused on analyzing this
mechanism by determining the transition states for hydride
migration to the vinylidene group. To assess the effect of the
real ligands and experimental selectivity, we performed
hybrid quantum mechanics/molecular mechanics (ONIOM)
calculations on the key TSB8–9 transition states. The MM
region included the phosphine substituents and methyl
groups of pinacolborane.[17] We used ONIOM calculations
because they allowed us to screen several ligands and sub-
strates. Moreover, we felt that non-bonding ligand–substrate
interactions are key factors governing selectivity. In these
cases, DFT/MM methods can give superior results to pure
DFT calculations because dispersion forces are not properly
described for standard gradient-corrected density functional
approaches.[18] Table 3 compares the relative energies of the
key pro-Z and pro-E transition states for different ligands
and substrates with the experimentally observed selectivity.
For propyne and monophosphine ligands PH3 and PMe3
(Table 3, entries 1 and 2), the calculated energy difference
between the Z and E transition states is low, with the E
isomer slightly favored. Upon introduction of the steric ef-
fects of the bulky PCy3 and PiPr3 phosphine groups, the
trend inverts and the Z path becomes more favorable by 1.6
and 0.5 kcalmolÀ1, respectively (Table 3, entries 3 and 4).
These results are fully consistent with the observed major
products in our work and the work of Miyaura and co-work-
ers,[7] thus supporting mechanism B for non-conventional
trans-hydroboration reactions. If we take a closer look at the
geometry of the pro-E transition state with PCy3 (Figure 3),
we can observe that the alkyne substituent is pointing to-
wards the metal center, thereby establishing a repulsive
steric interaction with the auxiliary ligands. On the other
hand, the alkyne substituent in the corresponding pro-Z
pro-Z
pro-E
[kcalmolÀ1
]
[kcalmolÀ1
]
1
2
3
4
PH3
PMe3
PCy3
PiPr3
+0.5
+0.6
0.0
0.0
0.0
–
9:62(a)
66:30(a)
91:7(a,b)
+1.6
+0.5
0.0
5
6
7
8
9
PH3
PCy3
PiPr3
PH3
0.0
0.0
0.0
0.0
0.0
+0.2
+2.7
+1.1
+0.1
+0.4
–
90:10
97:2(b)
–
PH3
–
[a] Results for 1-octyne; [b] Values taken from the paper by Miyaura and
co-workers, see Ref [7]. [c] Ratio determined experimentally.
Figure 3. Molecular structures of the selectivity-determining pro-E and
pro-Z transition states for vinylidene insertion into the Rh H bond
À
during propyne hydroboration using PCy3 ligands at the B3LYP:UFF
level. Hydrogen atoms from the phosphine groups and the borane agent
are omitted for clarity. The atoms in the MM part are represented by
sticks. Relatives energies shown are in kcalmolÀ1
.
1518
www.chemeurj.org
ꢅ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2012, 18, 1512 – 1521

Catalytic Non-Conventional trans-Hydroboration
FULL PAPER
groups, and 3) the preference of the vinylidene path (non-
conventional trans) over the classical path (cis) that involves
direct addition of the borane group to the alkyne. In the ab-
sence of steric effects with the PH3 ligand, we observed a
clear difference on going from methyl- to phenylacetylene.
The relative energies of the pro-Z and pro-E transition
C6H4CH3) were 30.1, 29.1, 28.5, and 28.0 kcalmolÀ1, respec-
tively, which nicely correlated with the percentages of ob-
tained Z isomer: 48%, 66%, 90%, and 94%, respectively.
The lower the energy barrier, the more favored the non-con-
ventional trans reaction channel is, and consequently, the
higher the selectivity towards the Z-alkene isomer. We iden-
tified a linear relationship between the observed selectivity
for the Z isomer and the overall energy barriers for vinyli-
dene formation (Figure 5). Lynam and co-workers[13a] have
determined the rate constants for the isomerization of the
two-step process for n-butyl- and phenyl-substituted alkynes.
They obtained values for the 1,3-hydrogen shift that were
somewhat lower (22.8 and 22.1 kcalmolÀ1 for R=nBu and
Ph, respectively) than ours (29.1 and 28.5 kcalmolÀ1 for R=
CH3 and Ph, respectively) owing to our use of the simplified
states invert, with the
Z isomer becoming preferred
(Table 3, entries 1 and 5). Nevertheless, the energy differ-
ence between the Z and E isomers for phenylacetylene is
also small (ca. 0.2 kcalmolÀ1), and consequently subtle ef-
fects are expected to be responsible for them. Figure 4 col-
lects the structures of the pro-Z and pro-E transition states
for the phenylacetylene substrate. When comparing the two
isomers for each substrate type, we observed very similar
geometric and electronic parameters (Figures 2 and 4). The
Figure 4. Molecular structures and main geometric parameters of the se-
lectivity-determining pro-E and pro-Z transition states for vinylidene in-
À
sertion into the Rh H bond during phenylacetylene hydroboration with
PH3 ligands. Distances in ꢆ and relative energies shown are in kcal
Figure 5. Correlation between the observed Z isomer (%) and the overall
activation barriers for vinylidene formation.
molÀ1
.
inverse trend should be related to the balance between the
cis and trans disposition of the alkyne substituents and be-
tween the cis and trans stabilization of the developing nega-
tive charge at the vinylidene a-carbon atom. Other aryl-sub-
stituted alkyne substrates show similar relative energies of
the transition states for PH3 ligand (Table 3, entries 8 and
9). Thus, the intrinsic electronic properties of phenyl acety-
lene have a positive influence on the selectivity towards the
Z product as compared with alkyl acetylenes. For phenylace-
tylene, addition of the steric hindrance of PCy3 ligand
causes the energetic preference of the Z isomer to increase
by 2.5 kcalmolÀ1 (Table 3, entries 5 and 6). The latter value
is similar to that observed for methyl-acetylene (+2.1 kcal
molÀ1), thus showing that both types of substrates have simi-
lar sensitivities to the steric effect exerted by the phosphine
ligands. The last aspect that we considered was whether the
vinylidene formation becomes more favored than the cis-hy-
droboration channel for phenylacetylene. Thus, assuming
that the 1,3-shift of the hydrogen onto the alkynyl ligand is
the rate-determining step in vinylidene formation,[14] we first
calculated the energy barriers for different substrates with
the PH3 ligand. The overall energy barriers from the vinyli-
less-basic PH3 ligands.[13a] However, the trend in substituent
effects observed by the same authors was reflected in the
calculations on our model systems, that is, that isomerization
is easier for aryl than for alkyl substituents. These results
also support the idea that the higher selectivity for the Z
isomer in aryl-substituted alkynes is also related to the pre-
dominance of vinylidene over the cis-hydroboration reaction
channel. We note that the rigorous analysis of this aspect
would require full characterization of both reaction chan-
nels, thereby accounting for the electronic effects of PR3
ligand and the role of the base. Detailed studies on these as-
pects are in progress in our laboratory, but we believed that
it is worthwhile reporting the first results on the simplified
model systems.
Conclusions
We can conclude that the catalytic system based on the in
situ mixture of [{RhACHTNUGTRNEUNG(cod)Cl}2] and basic and bulky phos-
phine groups, such as PCy3, favored the non-conventional
trans-hydroboration over the cis-hydroboration in the pres-
ence of Et3N. The optimized reaction conditions for the hy-
ꢀ
dene complex for HC CR (R=C6H4CF3, Me, Ph, and
Chem. Eur. J. 2012, 18, 1512 – 1521
ꢅ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
1519

J. J. Carbꢁ, E. Fernꢂndez et al.
to reach complete dissolution and complete formation of the catalytic
complex in situ. Next, freshly distilled catecholborane (1.2 mmol) or pi-
nacolborane (1.2 mmol) was added to the solution of catalyst followed by
the substrate (1 mmol). The mixture was stirred at RT for 4 h. The prod-
ucts were characterized by 1H NMR spectroscopy and GC to determine
the degree of conversion and the selectivity obtained.
droboration of 1-octyne showed that the most-successful
substrate/borane ratio was 1:1.2, and tetrahydrofuran
seemed to be the solvent of choice. The highest percentage
of non-conventional trans-hydroborated product was ach-
ieved with a ratio of [Rh]/PCy3 =1:4 and was successfully
formed at 08C, 258C, and 708C. Subtle changes in the
nature of the substrate indicated that electron-rich alkynes
with high steric demands were mainly converted into the de-
sired organoboron isomers. We have computationally char-
acterized a plausible reaction mechanism for the non-con-
ventional trans-hydroboration reaction through an initial
alkyne to vinylidene isomerization, following the suggestion
of Miyaura and co-workers based on their deuterium-la-
beled experiment. Unlike their mechanism, we propose a se-
Computational details: All calculations were performed using the Gaussi-
an09 series of programs.[19] Full quantum mechanical calculations on
model systems were performed within the framework of density function-
al theory (DFT)[20] using the B3LYP functional.[21] A quasi-relativistic ef-
fective-core potential operator was used to represent the 28 innermost
electrons of the Rh atom, as well as the 10 innermost electrons of the P
atoms.[22] The basis set for Rh and P atoms was that associated with the
pseudopotential,[22] with a standard double-x LANL2DZ contraction,[19]
and, in the case of P atoms, supplemented by a d shell.[23] The C, H, O,
Cl, and B atoms were represented by means of the 6–31GACHTUNGTRENNUNG(d,p) basis
set.[24] All geometry optimizations were full, with no restrictions. Station-
ary points located in the potential-energy hypersurface were character-
ized as true minima through vibrational analysis. Transition states located
in the potential-energy hypersurface were characterized through vibra-
tional analysis, having one and only one imaginary frequency, whose
normal mode corresponded to the expected motion. For the hybrid quan-
tum mechanics/molecular mechanics (QM/MM) calculations, we applied
the ONIOM method as implemented in the Gaussian 09 package.[25] The
À
quence of vinylidene insertion into the Rh H bond followed
À
by reductive elimination of the C B bond. Thus, the mecha-
nism can be divided into four stages: 1) isomerization of the
alkyne into the vinylidene to yield a rhodium/vinylidene
complex, 2) oxidative addition of the borane reagent, 3) vi-
À
nylidene insertion into the Rh H bond, and finally 4) reduc-
QM region included the [RhCl
ACHTUNGTRENNUNG
À
tive elimination of the C B bond to yield the 1-alkenylboro-
acetylene substrates, and the HBAHCNUTGTRENNUNG
nate. Calculations indicated that the insertion of vinylidene
stituted of the methyl substituents of borane and the substituents (Me,
iPr, and Cy3) of the phosphines. The QM level was the same as men-
tioned above. UFF force field[26] was used as implemented in Gaussian 09
to describe the atoms included in the MM part.
À
into the Rh H bond is the selectivity-determining step. In-
troducing the steric effects of real ligands, we were able to
reproduce the experimental outcome, thus supporting the
consistency of the proposed mechanism. Our calculations
also indicated that bulky ligands are required to selectively
obtain (Z)-1-alkenylboronates and that increasing the steric
hindrance of the ligands causes an increase in selectivity of
the Z isomer (PCy3 >PiPr3 >PMe3 >PH3). The higher selec-
tivity observed for aryl alkynes compared with aliphatic al-
kynes can be explained by the analysis of different factors.
The intrinsic electronic properties of aryl substituents are
more favorable for the Z pathway than alkyl substituents.
For electron-donating substituents, the formation of the vi-
nylidene complex is favored, which seems to facilitate the
occurrence of the vinylidene reaction channel over the clas-
sical cis-hydroboration pathway.
Acknowledgements
The authors are grateful for financial support from the MEC of Spain
(CTQ2010–16226, CTQ2011–29054-C02–01, CTQ2005–08351), from the
Consolider Ingenio 2010 (CSD2006–0003), and from the Direcciꢁ Gener-
al de Recerca (DGR) of the Autonomous Government of Catalonia
(2009SGR462 and XRQTC). J.C. thanks the URV for a grant.
[1] a) A. Pelter, K. Smith, H. C. Brown in, Borane Reagents: Best Syn-
thetic Methods (Eds.: A. R. Katritzky, O. MethCohn, C. W. Rees),
Academic Press, London, 1988, p. 194.
[2] a) H. C. Brown, S. K. Gupta, J. Am. Chem. Soc. 1971, 93, 1816;
b) H. C. Brown, S. K. Gupta, J. Am. Chem. Soc. 1975, 97, 5249.
[3] a) Ch. E. Tucker, J. Davidson, P. Knochel, J. Org. Chem. 1992, 57,
3482; b) I. Pergament, M. Srebnik, Org. Lett. 2001, 3, 217.
Experimental Section
[4] W. G. Woods, Ph. L. Strong, J. Am. Chem. Soc. 1966, 88, 4667.
[5] H. Kono, K. Ito, Y. Nagai, Chem. Lett. 1975, 1095.
General: All reactions and manipulations were conducted using standard
vacuum-line techniques under an atmosphere of dry nitrogen. All organic
solvents were distilled over dehydrating agents and degassed with nitro-
[6] a) D. Mannig, H. Nçth, Angew. Chemie, 1985, 97, 854–855; Angew.
Chem. Int. Ed. Engl. 1985, 24, 878; b) R. T. Baker, J. C. Calabrese,
S. A. Westcott, P. Nguyen, T. B. Marder, J. Am. Chem. Soc. 1993,
115, 4367; c) B. P. Cleary, R. Eisenberg, Organometallics 1995, 14,
4525; d) A. E. Dorigo, P. V. R. Schleyer, Angew. Chem. 1995, 107,
108; Angew. Chem. Int. Ed. Engl. 1995, 34, 115; e) J. R. Knorr, J. S.
Merola, Organometallics 1990, 9, 3008; f) C. Crudden, D. Edwards,
Eur. J. Org. Chem. 2003, 4695; g) I. Beletskaya, A. Pelter, Tetrahe-
dron 1997, 53, 4957; h) M. J. Burgess, M. J. Ohlmeyer, J. Org. Chem.
1988, 53, 5178; i) J. M. Brown, H. Doucet, E. Fernꢂndez, H. E.
Heeres, M. W. Hooper, D. I. Hulmes, F. I. Knight, T. P. Layzell, G. C.
Lloyd-Jones in Transition Metal Catalysed Reactions (Chemistry for
the 21st Century) (Eds.: S. I. Murahashi, S. G. Davies), Blackwell Sci-
ence Oxford, UK, 1999, 465; j) J. M. Brown, in Modern Rhodium-
Catalyzed Organic Reactions, (Eds: P. A. Evans), Wiley-VCH, Wein-
heim, 2004; k) C. M. Vogels, S. A. Westcott, Curr. Org. Chem. 2005,
9, 687; l) M. J. Burgess, M. J. Ohlmeyer, Chem. Rev. 1991, 91, 1179.
gen before use. [{RhACHTUNGTRENNUNG(cod)Cl}2], [{IrACHTUNGTNER(NUGN cod)Cl}2], and all the substrates were
used as purchased from Sigma Aldrich. Deuterated solvents for routine
NMR measurements were used as purchased from SDS. NMR spectra
were obtained on either a Varian Goku 400 or a Varian Mercury 400
1
spectrometer. H NMR and 13C {1H} NMR chemical shifts are reported in
ppm (d) relative to tetramethylsilane. 11B {1H} NMR chemical shifts are
reported in ppm (d) relative to BF3·Et2O. GC analysis was performed on
an Agilent Technologies 6850 apparatus with a flame-ionization detector
equipped with an achiral column HP-5 (30 m, 0.25 mm ID, 0.25 mm thick-
ness) using H2 as the carrier gas.
General procedure for the rhodium-catalyzed trans-hydroboration of al-
kynes: Catalyst precursor ([{RhACTHNUGTERN(NUGN m-Cl)ACHUTGTNREN(NGU cod)}2] or [{IrAHCUTNGERT(GNNNU m-Cl)ACHTNGURTEN(NUGN cod)}2]=
0.015 mmol) and the ligand (0.06 mmol) were introduced into a previous-
ly purged Schlenk tube under a nitrogen atmosphere and dissolved in
THF (3 mL) and NEt3 (3 mL, 5 mmol). The mixture was stirred for 5 min
1520
www.chemeurj.org
ꢅ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2012, 18, 1512 – 1521

Catalytic Non-Conventional trans-Hydroboration
FULL PAPER
[7] T. Ohmura, Y. Yamamoto, N. Miyaura, J. Am. Chem. Soc. 2000, 122,
4990.
[18] For example, see: a) J. J. Carbꢁ, A. Lledꢁs, D. Vogt, C. Bo, Chem.
Eur. J. 2006, 12, 1457; b) W. M. C. Sameera, F. Maseras, Phys. Chem.
Chem. Phys. 2011, 13, 10520.
[8] a) C. Widauer, H. Grꢇtzmacher, T. Ziegler, Organometallics 2000,
19, 2097; b) D. G. Musaev, A. M. Mebel, K. Morokuma, J. Am.
Chem. Soc. 1994, 116, 10693; c) X. Huang, Z. Lin, in Computational
Modeling of Homogeneous Catalysis (Catalysis by Metal Com-
plexes), (Eds.: F. Maseras, A. Lledꢁs), Kluwer, Springer US, New
York, 2002, 189.
[9] a) I. Ojima, N. Clos, R. J. Donovan, P. Ingallina, Organometallics
1990, 9, 3127; b) R. S. Tanke, R. H. Crabtree, J. Am. Chem. Soc.
1990, 112, 7984; c) C.-H. Jun, R. H. Crabtree, J. Organomet. Chem.
1993, 447, 177; d) Y. Maruyama, K. Yamamura, I. Nakayama, K.
Yoshiuchi, F. Ozawa, J. Am. Chem. Soc. 1998, 120, 1421; e) R. H.
Crabtree, New J. Chem. 2003, 27, 771.
[10] a) L. Dang, H. Zhao, Z. Lin, T. B. Marder, Organometallics 2008,
27, 1178; b) J. Zhu, Z. Lin, T. B. Marder, Inorg. Chem. 2005, 44,
9384; c) H. Braunschweig, P. Brenner, A. Muller, K. Radacki, D.
Rais, K. Uttinger, Chem. Eur. J. 2007, 13, 7171; d) C. N. Iverson,
M. R. Smith III, Organometallics 1996, 15, 5155.
[19] Gaussian 09, Revision A.02, M. J. Frisch, G. W. Trucks, H. B. Schle-
gel, G. E. Scuseria, M. A. Robb, J. R. Cheeseman, G. Scalmani, V.
Barone, B. Mennucci, G. A. Petersson, H. Nakatsuji, M. Caricato, X.
Li, H. P. Hratchian, A. F. Izmaylov, J. Bloino, G. Zheng, J. L. Son-
nenberg, M. Hada, M. Ehara, K. Toyota, R. Fukuda, J. Hasegawa,
M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H. Nakai, T. Vreven,
J. A., Jr., Montgomery, J. E. Peralta, F. Ogliaro, M. Bearpark, J. J.
Heyd, E. Brothers, K. N. Kudin, V. N. Staroverov, R. Kobayashi, J.
Normand, K. Raghavachari, A. Rendell, J. C. Burant, S. S. Iyengar,
J. Tomasi, M. Cossi, N. Rega, J. M. Millam, M. Klene, J. E. Knox,
J. B. Cross, V. Bakken, C. Adamo, J. Jaramillo, R. Gomperts, R. E.
Stratmann, O. Yazyev, A. J. Austin, R. Cammi, C. Pomelli, J. W.
Ochterski, R. L. Martin, K. Morokuma, V. G. Zakrzewski, G. A.
Voth, P. Salvador, J. J. Dannenberg, S. Dapprich, A. D. Daniels, O.
Farkas, J. B. Foresman, J. V. Ortiz, J. Cioslowski, D. J. Fox, ; Gaussi-
an, Inc., Wallingford CT, 2009.
[11] F. J. G. Alonso, A. Hçhn, J. Wolf, H. Otto, W. Werner, Angew.
Chem. 1985, 97, 401; Angew. Chem. Int. Ed. Engl. 1985, 24, 406.
[12] a) D. B. Grotjahn, X. Zeng, A. L. Cooksy, J. Am. Chem. Soc. 2006,
128, 2798; b) T. Rappert, O. Nurnberg, N. Mahr, J. Wolf, H. Werner,
Organometallics 1992, 11, 4156; c) J. Wolf, H. Werner, O. Serhadli,
M. L. Ziegler, Angew. Chem. 1983, 95, 428; Angew. Chem. Int. Ed.
Engl. 1983, 22, 414.
[20] a) R. G. Parr, W. Yang, in Density Functional Theory of Atoms and
Molecules, Oxford University Press, Oxford, UK, 1989; b) T. Zie-
gler, Chem. Rev. 1991, 91, 651.
[21] a) C. Lee, W. Yang, R. G. Parr, Phys. Rev. B 1988, 37, 785; b) A. D.
Becke, J. Chem. Phys. 1993, 98, 5648; c) P. J. Stephens, F. J. Devlin,
C. F. Chabalowski, M. J. Frisch, J. Phys. Chem. 1994, 98, 11623.
[22] P. J. Hay, W. R. Wadt, J. Chem. Phys. 1985, 82, 299.
[13] a) M. J. Cowley, J. M. Lynam, J. M. Slattery, Dalton Trans. 2008,
4552; b) B. A. Vastine, M. B. Hall, Organometallics 2008, 27, 4325;
c) F. de Angelis, A. Sgamellotti, N. Re, Organometallics 2007, 26,
5285; d) D. B. Grothahn, X. Feng, A. L. Cooksy, W. S. Kassel, A. G.
di Pasquale, L. N. Zakharov, A. L. Rheingold, Organometallics 2007,
26, 3385; e) C. H. Suresh, N. Koga, J. Theor. Comput. Chem. 2005, 4,
59; f) Y. Wakatsuki, N. Koga, H. Werner, K. Morokuma, J. Am.
Chem. Soc. 1997, 119, 360.
[23] A. Hçllwarth, M. Bçhme, S. Dapprich, A. W. Ehlers, A. Gobbi, V.
Jonas, K. F. Kçhler, R. Stegmann, A. Veldkamp, G. Frenking, Chem.
Phys. Lett. 1993, 208, 237.
[24] a) M. M. Francl, W. J. Pietro, W. J. Hehre, J. S. Binkley, M. S.
Gordon, D. J. Defrees, J. A. Pople, J. Chem. Phys. 1982, 77, 3654;
b) W. J. Hehre, R. Ditchfield, J. A. Pople, J. Chem. Phys. 1972, 56,
2257; c) P. C. Hariharan, J. A. Pople, Theor. Chim. Acta 1973, 28,
213.
[14] For a recent review, see: J. M. Lynam, Chem. Eur. J. 2010, 16, 8238.
[15] a) E. Bustelo, J. J. Carbꢁ, A. Lledꢁs, K. Mereiter, M. C. Puerta, P.
Valerga, J. Am. Chem. Soc. 2003, 125, 3311; b) C. Garcꢃa-Yebra, C.
Lꢁpez-Mardomingo, M. Fajardo, A. AntiÇolo, A. Otero, A. Rodrꢃ-
guez, A. Vallat, D. Lucas, Y. Mugnier, J. J. Carbꢁ, A. Lledꢁs, C. Bo,
Organometallics 2000, 19, 1749.
[16] Despite our efforts, we were unable to locate the transition state
TSB9À10 for the E path. We expect a very low energy barrier for the
other three calculated transition states in the reductive elimination
step, and consequently, only a small influence in determining the se-
lectivity and activity of the reaction.
[25] a) F. Maseras, K. Morokuma, J. Comput. Chem. 1995, 16, 1170; b) S.
Humbel, S. Sieber, K. Morokuma, J. Chem. Phys. 1996, 105, 1959;
c) M. Svensson, S. Humbel, R. D. J. Froese, T. Matsubara, S. Sieber,
K. Morokuma, J. Phys. Chem. 1996, 100, 19357; d) M. Svensson, S.
Humbel, K. Morokuma, J. Chem. Phys. 1996, 105, 3654; e) S. Dap-
prich, I. Komꢂromi, K. S. Byun, K. Morokuma, M. J. Frisch, J. Mol.
Struct. 1999, 462, 1.
[26] A. K. Rappꢈ, C. J. Casewit, K. S. Colwell, W. A. Goddard III, W. M.
Skiff, J. Am. Chem. Soc. 1992, 114, 10024.
[17] Within this method and the QM/MM partition, we mainly accounted
for the steric effects of the phosphine ligands.
Received: August 31, 2011
Published online: December 30, 2011
Chem. Eur. J. 2012, 18, 1512 – 1521
ꢅ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
1521