The "catalytic nitrosyl effect": NO bending boosting the efficiency of rhenium based alkene hydrogenations

Article abstract of DOI:10.1021/ja400135d

Diiodo Re(I) complexes [ReI₂(NO)(PR₃)₂(L)] (3, L = H₂O; 4, L = H₂; R = iPr a, Cy b) were prepared and found to exhibit in the presence of "hydrosilane/B(C₆F ₅)₃" co-catalytic systems excellent activities and longevities in the hydrogenation of terminal and internal alkenes. Comprehensive mechanistic studies showed an inverse kinetic isotope effect, fast H ₂/D₂ scrambling and slow alkene isomerizations pointing to an Osborn type hydrogenation cycle with rate determining reductive elimination of the alkane. In the catalysts' activation stage phosphonium borates [R ₃PH][HB(C₆F₅)₃] (6, R = iPr a, Cy b) are formed. VT ²⁹Si- and ¹⁵N NMR experiments, and dispersion corrected DFT calculations verified the following facts: (1) Coordination of the silylium cation to the O_NO atom facilitates nitrosyl bending; (2) The bent nitrosyl promotes the heterolytic cleavage of the H-H bond and protonation of a phosphine ligand; (3) H₂ adds in a bifunctional manner across the Re-N bond. Nitrosyl bending and phosphine loss help to create two vacant sites, thus triggering the high hydrogenation activities of the formed "superelectrophilic" rhenium centers.

Full text of DOI:10.1021/ja400135d

Article
pubs.acs.org/JACS
The “Catalytic Nitrosyl Effect”: NO Bending Boosting the Efficiency of
Rhenium Based Alkene Hydrogenations
Yanfeng Jiang,† Birgitta Schirmer,‡ Olivier Blacque,† Thomas Fox,† Stefan Grimme,§ and Heinz Berke*,†
†Anorganisch-Chemisches Institut, Universitat Zurich, Winterthurerstrasse 190, CH-8037, Zurich, Switzerland
̈
̈
̈
‡Organisch-Chemisches Institut, Universitat Munster, Corrensstrasse 40, 48149 Munster, Germany
̈
̈
̈
§Mulliken Center for Theoretical Chemistry, Institut fur Physikalische und Theoretische Chemie, Universitat Bonn, Beringstrasse 4,
̈
̈
D-53115 Bonn, Germany
S
* Supporting Information
ABSTRACT: Diiodo Re(I) complexes [ReI2(NO)(PR3)2(L)]
(3, L = H2O; 4 , L = H2; R = iPr a, Cy b) were prepared and
found to exhibit in the presence of “hydrosilane/B(C6F5)3” co-
catalytic systems excellent activities and longevities in the
hydrogenation of terminal and internal alkenes. Comprehen-
sive mechanistic studies showed an inverse kinetic isotope
effect, fast H2/D2 scrambling and slow alkene isomerizations pointing to an Osborn type hydrogenation cycle with rate
determining reductive elimination of the alkane. In the catalysts’ activation stage phosphonium borates [R3PH][HB(C6F5)3] (6,
R = iPr a, Cy b) are formed. VT 29Si- and 15N NMR experiments, and dispersion corrected DFT calculations verified the
following facts: (1) Coordination of the silylium cation to the ONO atom facilitates nitrosyl bending; (2) The bent nitrosyl
promotes the heterolytic cleavage of the H−H bond and protonation of a phosphine ligand; (3) H2 adds in a bifunctional
manner across the Re−N bond. Nitrosyl bending and phosphine loss help to create two vacant sites, thus triggering the high
hydrogenation activities of the formed “superelectrophilic” rhenium centers.
1. INTRODUCTION
The nitrosyl (NO) ligand exhibits noninnocent properties able
to switch its binding to metal centers from the linear (three-
electron donor) to the bent (one-electron donor) binding
mode.1 The bending of the nitrosyl ligand creates a vacant site
Our group has systematically studied rhenium nitrosyl
accompanied by a formal 2e− oxidation of the metal center
chemistry, particularly aimed to be applied in reductive
without the requirement of preceding ligand dissociation.2 Such
catalyses.7 In continuation of attempts to use ligand tuning to
redox change with NO bending, which was termed “stereo-
achieve improved catalytic performance, we demonstrated
chemical control of valence” by Enemark and Feltham,3 gave us
the notion that the NO ligand could be used to generate a
vacant site “on demand” by bending, which for rhenium as a
middle transition element is normally a difficult process, since
recently that the use of boron Lewis acids as co-catalysts can
greatly enhance alkene hydrogenations applying trans-diphos-
phine Re(I) bromo hydrides as catalysts comparable to the
performance of platinum group metal complexes.8 In these
rhenium shows a great tendency to establish stable 18 electron
cases, reversible bromide abstractions assisted by the external
complexes. Moreover, nitrosyl ligands at electron rich metal
centers possess high Lewis basicity at the ONO atom.4
Interaction with Lewis acids could subsequently increase the
metal to NO π-back-donation, which in principle facilitates the
NO bending. With regard to Negishi’s idea of catalytic
enhancement by formation of “super Lewis acids” or “super-
electrophiles”,5 the Lewis acid promoted NO bending together
with a vacant site generated in the catalytic reaction could be
regarded as a “double Lewis acid”. However, following this idea,
it remained challenging to find an appropriate transition metal
nitrosyl system to cooperate with a suitable Lewis acid for
amplification of the “nitrosyl effect” and to bring about efficient
catalyses. It deserves special mentioning that hydrogenations6
ranking highest next to C−C coupling catalyses were up to now
not documented based on the “catalytic nitrosyl effect”.
boron Lewis acid were the key activation step. The nitrosyl
ligand, on the other hand, functioned in these systems merely
as a strongly bound ancillary ligand with strong π-acceptor
properties endowing additionally a trans-effect or trans-
influence. Could the NO ligand attribute a still more
sophisticated functional role to improve catalysis? This
question challenged us to explore the ways of how to enforce
NO bending and to utilize it for catalysis.
In this article, we would like to disclose the first example of a
“catalytic nitrosyl effect” triggering efficient hydrogenations of
alkenes. Our explorations were based on systems of trans-
diphosphine diiodo Re(I) mononitrosyl complexes modified by
Received: January 9, 2013
Published: February 5, 2013
© 2013 American Chemical Society
4088
dx.doi.org/10.1021/ja400135d | J. Am. Chem. Soc. 2013, 135, 4088−4102

Journal of the American Chemical Society
Article
reactions with silylium cations as Lewis acids (Scheme 1). This
work was inspired by two underlying facts: (1) The
octahedral geometry of the rhenium center with the weak aqua
ligand trans to the strong NO ligand (Figure 1). Two solvent
Scheme 1
nucleophilicity of metal bound halogens decreases in the
order of F > Cl > Br > I. The weakest nucleophile iodide
exhibits least tendency toward attack from an external Lewis
9
acid; (2) The silylium species R′3Si+ generated in situ from a
R3′SiH/B(C6F5)3 mixture is one of the strongest oxophiles, as
exemplified by the work of Piers10 and Oestreich11 in
hydrosilylations of carbonyl groups. Comprehensive mecha-
nistic studies were carried out to accomplish eventually silylium
facilitated nitrosyl bending and of phosphine loss as the key
activation steps. Quantum chemical calculations were employed
to establish plausibility for the experimentally derived reaction
pathways.
Figure 1. Molecular structure of [ReI2(NO)(PCy3)2(H2O)] (3b) with
two THF molecules with 50% probability displacement ellipsoids. All
hydrogen atoms were omitted for clarity except for the H2O ligand.
Selected bond lengths (Å) and angles (deg): P(1)−Re(1), 2.5389(10);
P(2)−Re(1), 2.5198(10); I(1)−Re(1), 2.7589(3); I(2)−Re(1),
2.7653(2); N(1)−Re(1), 1.7335(34); N(1)−O(1), 1.1994(42);
O(2)−Re(1), 2.1733(24); N(1)−Re(1)−O(2), 178.98(11); P(2)−
Re(1)−P(1), 179.23(3); I(1)−Re(1)−I(2), 169.390(9); O(1)−
N(1)−Re(1), 179.3(3).
2. RESULTS AND DISCUSSION
2.1. Synthesis and Characterization of Re(I) Mono-
nitrosyl Diiodo Complexes. The Re(I) iodo complexes were
prepared either by a one-pot synthesis route starting from
rhenium metal or via a Finkelstein type bromide/iodide
exchange reaction, as depicted in Scheme 2. Treatment of Re
THF molecules are fixed in the crystal lattice showing
intermolecular hydrogen bonding to the aqua ligand possessing
protons acidified by coordination of the H2O molecule. The
two iodo ligands are located trans revealing a slight bending
toward the aqua ligand (I1−Re1−I2 = 169.390(9)°). A
relatively high electron density at the rhenium center, which
is enhanced by the strong π donation of the iodide ligands,
causes a low NO IR stretching frequency at 1664 cm−1. The
low overall yield of 3b is assumed to be caused by
decomposition of the hydroiodic acid in the reduction step.
To overcome this disadvantage, an alternative Re(II)
bromonitrosyl precursor was employed applying a Finkelstein
type exchange reaction to introduce the iodo ligands.
Treatment of [NEt4]2[ReBr5(NO)] with excess of NaI in
THF afforded at 90 °C within 24 h a crude deep blue Re(II)
product. Single crystals were obtained from THF/Et2O and X-
ray diffraction revealed a [NEt4][trans-ReI4(NO)(THF)]
structure (see Supporting Information). Treatment of this
crude precursor with excess of PCy3 in a mixture of ethanol and
water furnished the formation of pure 3b in an overall yield of
67% (Scheme 2).
Scheme 2. Synthesis of Re(I) Mononitrosyl Diiodo
Complexes
We found that the above approach could not be applied to
the PiPr3 congener as large amounts of [HPiPr3]I were formed
in the last synthetic step. Therefore, we resorted to another
route using the Re(I) dibromo dihydrogen complex [ReBr2-
(NO)(PiPr3)2(η2-H2)] (1a) as a starting material. In THF
solution, the η2-H2 ligand of 1a could be replaced by H2O at 90
°C within 5 min affording the aqua complex [ReBr2(NO)-
(PiPr3)2(H2O)] (2a) in quantitative yield. The IR spectrum
showed a broad ν(OH) absorption at 3428 cm−1 and a strong
ν(NO) band at 1667 cm−1. The 1H NMR spectrum displayed a
singlet resonance at 5.72 ppm in THF-d8 for the aqua ligand.
Like for the iodo analogue, a single-crystal X-ray diffraction
metal with H2O2 followed by reduction with H3PO2/NO in the
presence of NEt4I/HI afforded a crude deep blue Re(II)
iodonitrosyl compound, which was further reacted with excess
of PCy3 in ethanol giving within 15 h at 90 °C a diiodo aqua
Re(I) complex [ReI2(NO)(PCy3)2(H2O)] (3b) in 20% overall
yield. The presence of an aqua ligand could be assured by the
appearance of a broad OH absorption at 3566 cm−1 in the IR
spectrum and a OH singlet at 5.20 ppm (THF-d8) which
1
disappeared in the H NMR spectrum upon addition of D2O.
The X-ray crystallographic analysis of 3b revealed a pseudo-
4089
dx.doi.org/10.1021/ja400135d | J. Am. Chem. Soc. 2013, 135, 4088−4102

Journal of the American Chemical Society
Article
study of 2a revealed trans alignment of the aqua molecule and
the nitrosyl ligand (see Supporting Information). The trans-
formation from 1a to 2a was reversible. In H2 atmosphere, 1a
was recovered instantaneously in quantitative yield from 2a in
benzene solution. The double cis labilization effect of the two
bromides and the trans effect of NO were expected to labilize
the aqua ligand so much that for instance double halogen anion
exchange would become accessible via exchange with the labile
water ligand position. Indeed, treatment of 2a with excess of
NaI in THF afforded within 2 h at room temperature the
diiodo aqua Re(I) congener [ReI2(NO)(PiPr3)2(H2O)] (3a) in
94% isolated yield. The IR spectra showed a broad ν(OH)
absorption at 3434 cm−1. The NO stretching frequency is
observed at 1674 cm−1, which is close to those of 2a and 3b. In
thermodynamically most stable 4a,b species. The assumed
kinetic product with H2 trans to the nitrosyl could not be
observed by NMR spectroscopy presumably due to the fact that
the π acceptor H2 is electronically too activated trans to NO.
Interestingly, the reaction of 3a,b or 4a,b with CO allows the
trapping of kinetic products. Treatment of either 3a,b or 4a,b
with 1 bar of CO in benzene-d6 at 23 °C afforded
instantaneously the kinetic product [ReI2(PR3)2-trans-(NO)-
(CO)] (5′, R = iPr a, Cy b) in 100% in situ yield originating
from CO attack between the two iodides of the 16e−
[ReI2(NO)(PR3)2] species. The IR spectra showed strong
ν(CO) absorptions at unexpectedly high wavenumbers (2082
for 5a′ and 2074 cm−1 for 5b′) indicating a high CO
electrophilicity. A singlet resonance at δ −16.7 ppm (5a′) or
−26.3 (5b′) was observed in the 31P{1H}-NMR spectra. 5a′,b′
is unstable mainly due to the strong π-acceptors CO and NO
trans to each other. Therefore, CO attack between iodide and
NO is taking place which is apparently kinetically hindered and
slow, leading to the thermodynamically stable complex
[ReI2(NO)(CO)(PR3)2] (5, R =iPr a, Cy b) within 15 h at
23 °C. In comparison with 5a′,b′, the 31P{1H}-NMR signals
were shifted low-field to δ −6.5 (5a) and −14.5 ppm (5b) and
the CO stretching frequencies became red-shifted to 1969 (5a)
and 1970 cm−1 (5b) with respect to the ν(NO) absorption at
1723 cm−1 for 5a and 1712 cm−1 for 5b. An X-ray diffraction
study of 5b confirmed the molecular structure with two cis-
iodides, and the NO and CO ligands cis positioned (Figure 2).
1
the H NMR spectra the aqua ligand appeared as a singlet at
5.43 ppm in THF-d8. The 31P{1H}-NMR spectrum exhibited a
singlet at −10.7 ppm shifted relatively high-field.
The aqua ligand in 3a,b is labile due to the strong cis
labilization effect of the two iodides. Under 1 bar of H2 in
benzene, the aqua ligand could be instantaneously replaced by
η2-H2 at 23 °C to afford the dihydrogen complexes [ReI2-
(NO)(PR3)2(η2-H2)] (4, R = iPr a, Cy b,) in 100% in situ yields
(Scheme 3). The 31P{1H}-NMR spectra displayed a singlet at δ
Scheme 3. Reaction of 3 with H2 and CO
1
3.06 (4a) or −4.94 ppm (4b). The H NMR spectra exhibited
2
triplet signals for the η2-H2 ligand at 0.64 ppm (4a, J(HP) = 20
Hz, T1 = 98 ms, THF-d8) and 0.80 ppm (4b, 2J(HP) = 21 Hz, T1
= 42 ms, benzene-d6), which are both shifted high-field
compared to those of 1a and 1b implying a strong electron
shielding effect. The relatively long T1 time of 4a,b indicates an
elongated H2 ligand approaching a dihydride structure. This
verifies the geometry of 4a,b bearing cis-aligned H2 and NO
ligands. A η2-H2 trans to NO is expected to exhibit a T1 time
less than 10 ms corresponding to a classical dihydrogen
character due to the competition between the σ*HH and π*NO
orbitals for back-bonding. Pure solid 4a,b could be obtained in
74% (4a) and 76% (4b) yield, respectively, from the same
reaction, but required the presence of anhydrous MgSO4, which
helped to trap the liberated H2O and prevented the back-
reaction to 3a,b during solvent evaporation. The IR spectra
showed strong ν(NO) absorptions at 1719 (4a) and 1710 cm−1
(4b). The reaction course from 3 to 4 can be interpreted in
terms of the initial formation of the 16e intermediate
[ReI2(NO)(PR3)2], which is assumed to be stabilized by the
two strong π-donating and strongly cis-labilizing iodides.
Dihydrogen, as a weak σ-donor and a good π-acceptor, places
itself preferentially between I and NO leading to the
Figure 2. Molecular structure of [ReI2(NO)(PCy3)2(CO)] (5b) with
50% probability displacement ellipsoids. Hydrogen atoms have been
omitted for clarity. Selected bond lengths (Å) and angles (deg): P(1)−
Re(1), 2.5343(4); P(2)−Re(1), 2.5384(4); I(1)−Re(1), 2.81659(13);
I(2)−Re(1), 2.78172(15); N(1)−O(2), 1.1805(19). P(1)−Re(1)−
P(2), 172.613(13); Re(1)−N(1)−O(2), 173.25(15); I(1)−Re(1)−
I(2), 92.52(1).
The coordination geometry of the kinetic and thermody-
namic product from the reaction of 3a,b with H2 were
investigated further by modeling ReI2(NO)(PMe3)2(η2-H2)
with dispersion corrected DFT. For this purpose the structures
were optimized at TPSS-D3/def2-TZVP level.12−14 Subsequent
single-point calculations with the more accurate B2PLYP-D315
double hybrid functional and the same basis set yielded the
energy values show in Figure 3 (for further details see the
Experimental Section and Supporting Information). As shown
in Figure 3, the thermodynamic product II reveals a η2-H2
4090
dx.doi.org/10.1021/ja400135d | J. Am. Chem. Soc. 2013, 135, 4088−4102

Journal of the American Chemical Society
Article
92% within 15 min corresponding to a TON of 9200 and TOF
of 3.7 × 104 h−1. Full conversion could be reached after 30 min.
Further increase of the S/C ratio to 40 000 afforded a TON of
2.0 × 104 and a TOF of 2.7 × 104 h−1 within 25 min. In
comparison, the two-component systems “3a/Et3SiH” and “3a/
B(C6F5)3” proved to be not catalytically active under the same
conditions. Similarly, applying “3b/Et3SiH/B(C6F5)3” a TON
of 9100 and a TOF of 3.6 × 104 h−1 within 15 min were
reached corresponding to a 91% conversion. The longevity of
the current system is dependent on both the kind of the Lewis
acid and of the hydrosilane. With either a “3a/Et3SiH/BPh3” or
a “3a/Et3SiH/AlF3” system, decrease of the catalytic perform-
ance was observed showing low TONs after 5 min. The
influence of the hydrosilane on the catalytic performance was
studied using the “3a/B(C6F5)3” system.19 In the case of
sterically more congested hydrosilanes, such as Ph3SiH and
iPr3SiH, hydrogenation could not be observed. With the
medium-sized hydrosilane MePh2SiH, a moderate activity with
a TOF of 7750 h−1 was seen within 1 h. In the case of the
sterically less hindered Me2PhSiH, which is known to be a
better hydride donor than Et3SiH, improved performance could
be achieved. The “3a/Me2PhSiH/B(C6F5)3” system brought
about a 90% conversion within 7.5 min corresponding to a
TON of 9040 and a TOF of 7.2 × 104 h−1. The catalytic system
of “3b/Me2PhSiH/B(C6F5)3” was even more efficient affording
a 93% conversion within 6.5 min corresponding to a TOF of
8.6 × 104 h−1. Noteworthy is the fact that under 40 bar of H2 at
23 °C the hydrogenations of 1-hexene catalyzed by “3a,b/
Me2PhSiH/B(C6F5)3” were extremely efficient that over 99%
conversions were achieved within 1 min affording the up-to-
date highest TOFs of up to 6.00 × 105 h−1 (entries 42 and 43 in
Table 1).
In general, the catalytic systems of “3a,b/Me2PhSiH/
B(C6F5)3” were found to be superior to the previously reported
“Re(I) bromo hydride/B(C6F5)3” system with respect to both
activity and longevity. Additionally, it should be pointed out
that short induction periods were invariably seen with the
“3a,b/hydrosilane/B(C6F5)3” systems. In comparison, when the
dihydrogen derivatives 4a,b were tested under the same
conditions, slightly improved catalytic results were obtained.
The “4a/Me2PhSiH/B(C6F5)3” system afforded a 97%
conversion within 7 min corresponding to a TON of 9700
and a TOF of 8.3 × 104 h−1. With the “4b/Me2PhSiH/
B(C6F5)3” system, a conversion of 89% was found within 6 min
giving a high TOF of 8.8 × 104 h−1, and full conversion was
reached within 25 min. Remarkably, induction periods were not
noticeable in the 4a,b cases, which not only verified the initial
formation of 4a,b in “3a,b/hydrosilane/B(C6F5)3”-catalyzed
hydrogenations, but also indicated the negative influence of the
aqua molecule in the generation of catalytically active species.
Differences in the performance of 3a,b and 4a,b became more
apparent when “Et3SiH/B(C6F5)3” was applied as a cocatalyst.
While a reaction time of 15 min was needed for a 92%
conversion based on the “3a/Et3SiH/B(C6F5)3” system, the
same reaction catalyzed by the “4a/Et3SiH/B(C6F5)3” system
afforded a conversion of 96% within 10 min corresponding to a
TON of 9640 and a TOF of 5.8 × 104 h−1 showing again no
induction period. Similarly, the hydrogenation of 1-hexene co-
catalyzed by “MePh2SiH/B(C6F5)3” could also be improved
with 4a giving a TOF of 1.3 × 104 h−1 within 30 min.
Figure 3. Kinetic and thermodynamic coordination isomers of
[ReI2(NO)(PMe3)2(η2-H2)] as calculated with B2PLYP-D3/def2-
TZVP//TPSS-D3/def2-TZVP. I: H−H eclipses I−Re−I axis with
trans-diiodide. II: H−H eclipses P−Re−P axis with cis-diiodide ligands.
All carbon and hydrogen atoms except for the H2 ligand are omitted
for clarity.
ligand parallel to the P−Re−P axis with a H−H distance of
1.527 Å emphasizing a “compressed” dihydride structure.16 In
comparison, the η2-H2 ligand in the kinetic product I, which
turned out to be a local energy minimum, was found to be
perpendicular to the P−Re−P axis and in-plane with the I−
Re−I axis exhibiting a H−H bond distance of 0.790 Å
corresponding to classical dihydrogen ligand. Importantly, the
energy difference between II and I was determined to be 18.0
kcal/mol in favor of II confirming the instability of the
geometry of the two π-acceptors of H2 and NO in trans
position.17 The Re−H distance of 2.005 Å in I indicates
weakness of the H2 binding to the rhenium center. In contrast,
the Re−H distance of 1.662 Å in II shows strong bonding
between the H atoms and the rhenium. The trans-alignment of
the strong iodo π-donor and the π-acceptor NO is preferred
due to the π “push-pull” interaction.18
2.2. Hydrogenation of Alkenes Catalyzed by the “Re(I)
Diiodide/Hydrosilane/B(C6F5)3” System. The potential of
3a,b in the catalytic hydrogenation of alkenes was explored. At
90 °C under 10 bar of H2, 3a,b showed reasonable activities in
the hydrogenation of 1-hexene giving a TON of 1.0 × 104
within 120 min. We reckoned that the catalytic performance of
3a,b could still be improved by finding ways to facilitate the
generation of open sites. Indeed, under milder conditions such
as 23 °C, hydrogenation did not occur at all, and this we
attributed to the unavailability of open sites. Therefore, utilizing
the strong oxophilicity of silicon,9−11 we envisaged to employ
the hydrosilane and B(C6F5)3 system as a cocatalyst to generate
in situ silylium ions expected to attach to the ONO atoms. These
NO derivatized species were thought to show facile bending at
the N atom with opening of a coordination site.
Indeed, the “Re(I) iodo complex/hydrosilane/B(C6F5)3”
system was found to exhibit up-to-date the best catalytic
performance in hydrogenations under mild conditions ever
studied before. To compare the efficiency of different catalytic
systems, the prototypic 1-hexene (2.5 mL) hydrogenations
were carried out under 10 bar H2 at 23 °C with a substrate-to-
catalyst ratio (S/C) of 10 000. The molar ratio of the rhenium
catalyst and the co-catalyst was set to 1:5 with a 1:1 molar ratio
of hydrosilane and B(C6F5)3. The results are listed in Table 1.
The “3a/Et3SiH/B(C6F5)3” system revealed a conversion of
With terminal alkenes, such as 1-octene, an excellent
conversion of 98% was achieved by addition of only 0.0125
mol % of “3a/Me2PhSiH/B(C6F5)3” at 23 °C requiring a
4091
dx.doi.org/10.1021/ja400135d | J. Am. Chem. Soc. 2013, 135, 4088−4102

Journal of the American Chemical Society
Article
a
Table 1. Hydrogenation of Alkenes Catalyzed by “[Re]/Hydrosilane/B(C6F5)3” Systems
entry
alkene (mmol)
1-hexene (20)
[Re]
co-catalyst
temp (°C)
T (min)
conv (%)
TON
TOF (h−1
)
1
2
3
4
5
3a
3a
3b
3a
3a
3a
3a
3a
3a
3b
3a
3a
3a
3a
3a
3b
3a
3a
3a
3a
3a
3a
3b
3a
3a
3b
3a
3b
4a
4b
4a
4b
4b
4a
4b
4b
4a
4a
4a
4a
4a
3a
3b
-
23
90
90
23
23
23
23
23
23
23
23
23
23
23
23
23
23
23
90
90
23
90
90
23
90
90
90
90
90
90
23
23
23
23
23
23
23
23
23
23
90
23
23
60
120
70
15
30
25
20
20
3
--
--
--
1-hexene (20)
1-hexene (20)
1-hexene (20)
1-hexene (20)
1-hexene (40)
1-hexene (20)
1-hexene (20)
1-hexene (20)
1-hexene (20)
1-hexene (20)
1-hexene (20)
1-hexene (20)
1-hexene (20)
1-hexene (20)
1-hexene (20)
1-octene (16)
styrene (9)
--
--
100
100
92
100
50
--
1.0 × 104
1.0 × 104
9200
1.0 × 104
2.0 × 104
--
5000
8571
Et3SiH/B(C6F5)3
Et3SiH/B(C6F5)3
Et3SiH/B(C6F5)3
Et3SiH
3.7 × 104
2.0 × 104
4.9 × 104
--
b
6
7
8
B(C6F5)3
--
--
--
9
Et3SiH/BPh3
5
513
1.0 × 104
1.3 × 104
3.0 × 104
--
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
Et3SiH/BPh3
3
6
658
Et3SiH/AlF3
5
25
--
2478
--
iPr3SiH/B(C6F5)3
Ph3SiH/B(C6F5)3
MePh2SiH/B(C6F5)3
MePh2SiH/B(C6F5)3
MePh2SiH/B(C6F5)3
MePh2SiH/B(C6F5)3
MePh2SiH/B(C6F5)3
MePh2SiH/B(C6F5)3
MePh2SiH/B(C6F5)3
MePh2SiH/B(C6F5)3
MePh2SiH/B(C6F5)3
MePh2SiH/B(C6F5)3
MePh2SiH/B(C6F5)3
MePh2SiH/B(C6F5)3
MePh2SiH/B(C6F5)3
MePh2SiH/B(C6F5)3
MePh2SiH/B(C6F5)3
--
60
40
60
75
65
90
20
8
--
--
--
78
90
93
98
28
59
7750
9040
9330
7812
1272
7750
7.2 × 104
8.6 × 104
5208
3816
c
1,7-octadiene (7)
allylcydohexene (6.5)
cyclooctene (8)
cydooctene (8)
cyclooctene (8)
cyclohexene (20)
cydohexene (20)
cyclohexene (20)
1,5-cyclooctadiene (8)
1,5-cyclooctadiene (8)
1-hexene (20)
1-hexene (20)
1-hexene (20)
1-hexene (20)
1-hexene (20)
1-hexene (20)
1-hexene (20)
1-hexene (20)
1-hexene (20)
1-octene (16)
styrene (9)
4107
3.1 × 104
182
d
900
3
84
2730
402
10
65
38
3
8036
3
2611
1540
312
5.2 × 104
1.5 × 104
6250
6
3
30
20
30
10
110
65
10
12.5
25
7
74
29
84
20
97
99
96
92
100
97
89
100
66
50
42
9
7346
2931
1.5 × 104
8793
1.3 × 104
9776
c
6696
c
1629
9710
5296
--
9940
9175
Et3SiH/B(C6F5)3
Et3SiH/B(C6F5)3
Et3SiH/B(C6F5)3
Me2PhSiH/B(C6F5)3
Me2PhSiH/B(C6F5)3
Me2PhSiH/B(C6F5)3
Me2PhSiH/B(C6F5)3
Me2PhSiH/B(C6F5)3
Me2PhSiH/B(C6F5)3
Me2PhSiH/B(C6F5)3
Me2PhSiH/B(C6F5)3
Me2PhSiH/B(C6F5)3
Me2PhSiH/B(C6F5)3
9640
5.8 × 104
4.4 × 104
2.4 × 104
8.3 × 104
8.8 × 104
2.4 × 104
1.3 × 104
1.2 × 104
3786
9241
1.0 × 104
9700
6
8840
25
30
20
30
3
1.0 × 104
6652
3982
1893
cyclooctene (8)
cyclooctene (8)
1-hexene (20)
1-hexene (20)
325
6500
4
40
100
99
1623
2.4 × 104
6.0 × 105
5.9 × 105
e
42
e
1
1.0 × 104
9900
43
1
a
Reactions were performed using 0.002 mmol [Re], 0.01 mmol of hydrosilane, 0.01 mmol of Lewis acid under 10 bar H2 unless otherwise stated and
b
c
monitored by a Buechi Pressflow controller. A total of 5 mL of 1-hexene with 0.001 mmol 3a. Based on the consumption of H2, as 2 equiv of H2
was needed for full hydrogenation. Determined by GC−MS spectroscopy. Under 40 bar of H2.
d
e
reaction time of 90 min. The hydrogenation of styrene also
worked well under the same conditions affording a TON of
1272 within 20 min. In the case of internal alkenes, such as
cyclohexene, cyclooctene (COE), and 1,5-cyclooctadiene
(COD), the hydrogenations at 23 °C showed poorer
performance presumably due to a relative faster deactivation
of the catalysts. For instance, in the hydrogenation of COE
catalyzed by 0.025 mol % of “3a/Me2PhSiH/B(C6F5)3”
deactivation of the catalyst occurred within 3 min giving a
TON of 402. Increasing the temperature to 90 °C resulted in
improved TONs and TOFs. For instance, while the hydro-
genation of cyclohexene catalyzed by the “3a/Me2PhSiH/
B(C6F5)3” system afforded a low TON value of 312 at 23 °C,
the same reaction carried out at 90 °C afforded a TON of 7366
and a TOF of 1.5 × 104 h−1 within 30 min. It is interesting to
compare the performance of COE with COD as substrates in
hydrogenations under 10 bar of H2 at 90 °C. In terms of
activity, the catalysis of COE hydrogenation was found superior
with respect to COD giving a higher TOF of 5.2 × 104 h−1
within 3 min. In terms of longevity of the catalyst, however,
4092
dx.doi.org/10.1021/ja400135d | J. Am. Chem. Soc. 2013, 135, 4088−4102

Journal of the American Chemical Society
Article
COD is a better substrate than COE affording a high TON of
6696 within 30 min. This result not only excluded the alkene
coordination step as rate determining, but also suggested the
existence of an active species bearing two vacant sites that could
be more efficiently stabilized by the bidentated ligand COD
than the monodentate COE.
(C6F5)3”, the formation of 6a or 6b was also recognized during
and after the hydrogenation experiments.
To establish the pathway to 6 in more detail, various
additional experiments were carried out, as depicted in Scheme
4: (1) The reaction of free PiPr3 or PCy3 with B(C6F5)3 under
Scheme 4. Investigation on the course of the formation of 6
2.3. Mechanistic Studies. 2.3.1. Removal of a PR3 Ligand
via Formation of [R3PH][HB(C6F5)3] for Catalyst Activation. It
is interesting to note that after catalytic reactions of Table 1 oils
had formed. To identify the composition of these oils, we
exemplarily scaled up the catalyst loading of “3a/Et3SiH/
B(C6F5)3” in the hydrogenation of 1-hexene. A precipitate
could be isolated and identified by NMR spectroscopy, which
turned out to be a mixture with the phosphonium borate
1
[iPr3PH][HB(C6F5)3] (6a) as the main component. In the H
NMR spectra, a doublet of quartet signal was observed at 5.62
1
ppm (3JHH = 4 Hz, JHP = 458 Hz), which was assigned to the
[iPr3PH]+ proton. The 11B NMR spectra exhibited a doublet at
−25.4 ppm (1JBH = 92 Hz), which could be attributed to the
[HB(C6F5)3]− anion. The 31P{1H}-NMR spectrum showed a
unique singlet at 43.9 ppm. In the 19F NMR spectra, the
[HB(C6F5)3]− anion was identified by a set of resonances at
1
−133.40 (m, 6F, ortho-C6F5), −163.54 (t, JCF = 19 Hz, 3F,
para-C6F5) and −166.64 ppm (m, 6F, meta-C6F5). The
presence of free B(C6F5)3 was also observed. ESI-MS
spectroscopy confirmed the formation of the [iPr3PH]+ cation
(m/z 161.0) and of the [HB(C6F5)3]− anion (m/z 513.0). It is
important to note that the related salt [HPiPr3]I, which could
be distinguished from 6a by its lower solubility in aprotic
solvents, was not formed in the resultant solution. Remarkably,
when the same hydrogenation reaction was terminated by
removing the H2 atmosphere shortly after the induction period,
6a was present, as well. The supernatant solution was
additionally examined by 31P NMR spectroscopy indicating
formation of a minor, yet unknown phosphorus-containing
species. By addition of 1-hexene to this solution, hydrogenation
could not be initiated suggesting decomposition of the
catalytically active species. Similar results were obtained with
the catalytic system of “3b/Et3SiH/B(C6F5)3” confirming the
formation of [Cy3PH][HB(C6F5)3] (6b) as the major
phosphorus-containing species after workup of the hydro-
genation catalysis. An X-ray crystallographic study of 6b
confirmed the proposed phosphonium borate structure, as
shown in Figure 4. Examining other active systems, such as
“3a(b)/Me2PhSiH/B(C6F5)3” and “4a(b)/MePh2SiH/B-
10 bar H2 in 1-hexene afforded at 23 °C within 10 min the
fluoride migrated product [(R3P)(C6F4)BF(C6F5)2] (R =iPr,
Cy), which was reported by Stephan et al. to be also accessible
via the direct mixing of PR3 (R = iPr, Cy) with B(C6F5)3 in
absence of H2 and alkene.20 As 6a,b could not be observed
from free phosphine and B(C6F5)3, a mechanism involving
initial dissociation of PR3 from rhenium and subsequent
heterolytic splitting of H2 by FLPs was excluded.21 (2) In the
absence of 3a, a mixture of PiPr3 and B(C6F5)3 reacts with
Et3SiH in pentane under 10 bar of H2. However, 6a was not
formed under these circumstances. (3) In absence of Et3SiH,
the reaction of 3a with B(C6F5)3 and H2 pressure did not afford
any precipitate. Likewise, the hydrogenation of 1-hexene
catalyzed by 3a,b with no co-catalysts added, did not lead to
formation of 6a,b. (4) In the case of an inactive catalytic
system, such as “3a/B(C6F5)3” with either iPr3SiH or Ph3SiH,
6a was also not formed. These results implied that the
phosphonium borate was generated in the course of the
catalyst’s activation. (5) The alkene proved to be not an
essential reaction component, since 6a could still be generated
along with the formation of 4a from a mixture of 3a, B(C6F5)3
and Et3SiH in pentane under 10 bar of H2 at ambient
temperature. (6) In the absence of both H2 and alkene, the
reaction of 3a with excess of B(C6F5)3 and Me2PhSiH in either
benzene-d6 or THF-d8 afforded over 85% of 4a and a small
amount of 6a (<10%), as well as an unknown species (<5%)
showing a resonance at 34 ppm in the 31P NMR spectrum. The
η2-H2 ligand in 4a must stem from the hydrolysis reaction
between Me2PhSiH and the aqua ligand of 3a. (7) The FLP
type phosphonium borate 6 proved to be inert toward catalysis,
since hydrogenation of 1-hexene was not observed at 90 °C
under 10 bar H2 using 0.05 mol % of 6b. All these results
pointed to the formation of 6a,b via intramolecular heterolytic
cleavage of H2 ligand by a neighboring phosphine ligand.22
Figure 4. Molecular structure of [Cy3PH][HB(C6F5)3] (6b) with 30%
probability displacement ellipsoids. All hydrogen atoms have been
omitted for clarity except for the HB and HP.
4093
dx.doi.org/10.1021/ja400135d | J. Am. Chem. Soc. 2013, 135, 4088−4102

Journal of the American Chemical Society
Article
2.3.2. Deuterium Isotope Studies. Deuterium isotope
kinetics was pursued in the hydrogenation of 1-hexene at 23
°C under 10 bar of H2 or D2. In the case of “3a/Me2PhSiH/
B(C6F5)3” and “3a/Et3SiH/B(C6F5)3” catalytic systems, inverse
KIE values of 0.62 and 0.61 were determined, respectively. This
suggested that the irreversible H−H splitting is not rate
determining.23 Under the catalytic condition applying D2, the
formed phosphonium borate was identified to be the deuterium
derivative [DPiPr3][HB(C6F5)3] (100%) confirming a hydride
transfer process from hydrosilane to B(C6F5)3. The deuterium
incorporation into the hexane starting from 1-hexene was
2
examined via H NMR revealing two singlets at 2.11 and 1.70
ppm. This confirmed the formation of 1,2-deuterated hexane
expected for Wilkinson or Osborn type hydrogenations.6c−e
The reaction thus went regio-specific. In the absence of H2, 1-
hexene could be isomerized with 10 mol % of 3a and 50 mol %
of “Me2PhSiH/B(C6F5)3” leading to (Z/E)-2-hexene in 30%
conversion at 23 °C after 48 h. Apparently this alkene
isomerization is comparatively slow and in no way competitive
with the hydrogenation. A mechanism is suggested to involve a
crucial hydride α-olefin rhenium complex, which via olefin
insertion into a Re−H bond forms a secondary rhenium alkyl
species. β-hydride elimination produces 2-hexene. The hydro-
genation system was also found to be active in H2/D2
scrambling. When the mixture of 3a, Me2PhSiH and B(C6F5)3
was treated in toluene with H2/D2 (550 mbar for each), the
characteristic HD triplet signal at 4.50 ppm (JHD = 44 Hz) was
Figure 5. 29Si NMR spectra recorded in toluene-d8. (Top) Mixture of
Et3SiH and B(C6F5)3 in 1:1 ratio at 233 K. (Middle) After addition of
4a (in a ratio of 1:3 to Et3SiH) into the mixture at 233 K within 10 s.
(Bottom) After warming up to 293 K and immediate measurement.
dihydrogen complex possessing an electron-deficient Re center
coordinated by two phosphine ligands.
1
observed in the H NMR spectra at 23 °C within 10 min.
2.3.4. 15N NMR Evidence for the Silylium Cation Induced
Nitrosyl Activation. 15N NMR spectroscopy was utilized to
provide further evidence for the nitrosyl activation pathway, in
particular the facilitation of the nitrosyl bending.25 A 98% 15N-
enriched rhenium diiodo dihydrogen complex [ReI2(15NO)-
(PCy3)2(η2-H2)] (4b-15N) was prepared starting from Re metal
and 15NO in a 7% overall yield using the synthetic method
described in Scheme 2. The 15NO gas was generated from
Na15NO2 via the reaction with FeSO4 and H2O.25 The 15N
NMR spectrum of 4b-15N was recorded in toluene-d8 at 23 °C
using CH3NO2 as external reference. It revealed a triplet at δ
−49.4 ppm (2JN−P = 5 Hz) due to coupling with the Re bound
phosphorus ligand. When ca. 5 equiv. of Me2PhSiH/B(C6F5)3
(1:1) were added to the toluene solution of 4b-15N, the triplet
signal immediately disappeared concomitant with the appear-
ance of two new resonances observed as a doublet at δ 105.9
ppm (2JN−P = 5 Hz) and a singlet at δ 90.8 ppm corresponding
to a monophosphine and a phosphine-free species, respectively.
The large downfield chemical shift in the 15N NMR upon
addition of the cocatalytic system would be in accord with the
reported observation of chemical shift changes when linear
nitrosyls transform into bent ones.26 Together with the 29Si
NMR spectroscopic observations, this is interpreted in terms of
attachment of the silylium cation to the ONO atom supporting
the NO bending.
2.3.3. VT-29Si NMR Evidence for the Silylium Cation
Induced Nitrosyl Activation. Variable temperature NMR
experiments were carried out in order to further shed light
on the possible intermediates. The 29Si NMR spectrum of the
mixture of Et3SiH and B(C6F5)3 in toluene-d8 at −40 °C
showed a new resonance at δ 9.8 ppm, which correlated with a
1
triplet at 0.97 ppm and a quartet at 0.52 ppm in the H NMR
spectrum attributed to the formation of the H transferred
intermediate possessing B−H···Si structures (Figure 5).
Addition of 0.33 equiv of 4a to the mixture at −40 °C revealed
a new singlet at 30.3 ppm in the 29Si NMR implying the
coordination of a silyl cation to a Lewis basic atom.24 The
resonances of the ethyl group of the Re-NO-SiEt3 moiety were
observed as a triplet at 0.81 and a quartet at 0.63 ppm, as
confirmed by 29Si, 1H correlation spectra. A Si−I bond, which is
supposed to appear at ca. 8.6 ppm in 29Si NMR spectrum, was
not observed.24 This again excluded that the activation course
went via silylium induced iodo dissociation. The phosphonium
borate 6a was not formed at this stage as revealed by the
1
absence of respective signals in both the H and the 31P NMR
spectra. The trapped intermediate is unstable at low temper-
ature and after warming to 23 °C it fully evolves into two
species showing in the 29Si NMR spectrum singlets at 29.2 and
34.8 ppm with complete consumption of Et3SiH. In
comparison, when 4a was added to the mixture of Et3SiH/
B(C6F5)3 at 23 °C, one more species was observed showing a
singlet at 44.2 ppm in the 29Si NMR spectrum. The 31P NMR
spectrum indicated the formation of 6a along with three other
new species at 60.2, 49.0, and 35.1 ppm. It was noticed that the
singlet at 35.1 ppm was found to correlate with a triplet signal
at 2.44 ppm (H2, J = 26 Hz) and two multiplets at 2.49 and
2.3.5. DFT Investigation of Silylium Facilitated NO
Bending. Four model systems I, II, I-SiMe3 and II-SiMe3 (-1
and -2) derived from the complex [ReI2(NO)(PMe3)2(η2-H2)]
bearing either trans- or cis-diiodide ligands with or without
SiMe3+ Lewis acid attached, were investigated in more detail by
DFT calculations. Their geometries were optimized at TPSS
level12 employing the large Gaussian AO-basis set def2-TZVP13
and the D3 dispersion correction.14 Subsequent single-point
calculations with the double-hybrid functional B2PLYP-D315
and the same basis set were carried out to yield the energy
1
1.19 ppm in the H NMR spectrum. The downfield chemical
shift of H2 in comparison to that of 4 and the size of coupling
constant of the triplet signal suggested the presence of a new
4094
dx.doi.org/10.1021/ja400135d | J. Am. Chem. Soc. 2013, 135, 4088−4102

Journal of the American Chemical Society
Article
Figure 6. Sketches of local energy minimum structures of five rhenium dihydrogen diiodide systems (with linear and bent nitrosyl ligands Re−N−O
= 135°). I: H−H eclipses I−Re−I axis. II: H−H eclipses P−Re−P axis. I-SiMe3: H−H eclipses I−Re−I axis with Si atom on ONO. II-SiMe3-1: H−H
eclipses P−Re−P axis with Si atom orientated cisoid to H2. II-SiMe3-2: H−H eclipses P−Re−P axis with the Si atom orientated transoid to H2. All
the carbon and hydrogen atoms except for the H2 ligand are omitted for clarity.
Table 2. Relative Energies and Geometry Parameters of the Local Minima Obtained upon the Bending of Nitrosyl Ligand
I
II
I-SiMe3
134.5
II-SiMe3-1
II-SiMe3-2
Re−N−O (o)
180.0
135.0
178.1
135.0
175.5
180.0
135.0
120.5
176.6
135.0
120.0
ΔE (kcal/mol)
I1−Re−N (deg)
I2−Re−N (deg)
Re−N (Å)
N−O (Å)
H−H (Å)
0.0
101.7
101.2
1.767
1.185
0.790
2.008
2.006
+23.0
90.2
0.0
173.4
92.1
+21.5
156.9
102.0
1.806
1.227
1.494
1.657
1.657
0.0
102.5
102.5
1.728
1.287
0.787
2.075
2.076
+14.9
96.3
0.0
168.8
96.8
+8.7
155.8
102.3
1.777
1.311
1.512
1.656
1.657
+20.4
152.5
99.5
0.0
174.0
91.0
+9.8
176.9
84.5
26.5
178.0
86.6
110.3
1.788
1.232
0.785
2.078
2.097
109.0
1.777
1.183
1.527
1.662
1.662
1.763
1.748
1.269
1.560
1.660
1.660
1.777
1.311
1.525
1.655
1.656
1.737
1.153
1.512
1.662
1.662
1.787
1.306
1.533
1.668
1.671
1.787
1.307
1.483
1.670
1.673
1.339
0.783
2.076
2.114
Re−H1 (Å)
Re−H2 (Å)
values discussed below. These values possess an estimated
accuracy of about 1 −2 kcal/mol. All calculated structures are
calculated to be local energy minima. Selected examples with
linear NO and bent NO at 135° are displayed in Figure 6. The
geometric parameters obtained from the molecules with NO
bending are listed in Table 2. For the type I and II complexes,
relative energies of 23.0 and 21.5 kcal/mol, respectively, were
calculated for the bending of the Re−N−O angle to 135°,
demonstrating that both conformers prefer a linear NO
arrangement and that a considerable energetic demand would
be necessary to achieve even small bending angles. The type I
complex with the H−H axis eclipsing the I−Re−I axis shows a
lower preference for the NO bending than the type II complex
with H−H eclipsing the P−Re−P axis. In both cases, the iodide
ligand cis to the nitrogen atom is slightly bent away from the
NO ligand and the I2−Re−N angle increases along with
bending of the NO ligand. At the same time, the Re−N and
N−O bond distances increase with the bending of the Re−N−
O angle. The enhanced π-back-donation from the Re center to
NO leads to decreased bond orders.
complexes. For instance, a smaller energy of 14.9 kcal/mol was
required to achieve a bending angle of 135° for I-SiMe3 in
comparison to 23.0 kcal/mol for I. Further bending to 130°
resulted in full dissociation of the dihydrogen ligand trans to the
bent NO group. In the case of II-SiMe3, two isomers are
present: II-SiMe3-1 with coordinated silylium cisoid to the
dihydrogen ligand and II-SiMe3-2 with coordinated silylium
transoid to the dihydrogen ligand. Comparing the two
conformers with linear Re−N−O angles, the II-SiMe3-1 is
found to be 7.8 kcal/mol more stable than II-SiMe3-2. Such an
orientational preference of the silylium residue can only be
interpreted in one way - steric repulsion between the H2 and
+
SiMe3+ units is much smaller than between the iodo and SiMe3
moieties. It is worth mentioning that the large energy difference
between I and II (18 kcal/mol) decreased to 9.6 kcal/mol
between I-SiMe3 and II-SiMe3-2. This can be explained by the
fact that the trans influence of the nitrosyl ligand, which is the
main cause for the instability of I, is weakened by coordination
of the silylium ion to the ONO atom. That way the effect that
the Lewis acid facilitates NO bending is more pronounced in
the case of II-SiMe3 (-1 and -2). The energies for bending the
Re−N−O angle to 135° are rather small with only 8.7 kcal/mol
+
Upon attachment of the SiMe3 Lewis acid to the nitrosyl
ligand, NO bending was facilitated in the I-SiMe3 and II-SiMe3
4095
dx.doi.org/10.1021/ja400135d | J. Am. Chem. Soc. 2013, 135, 4088−4102

Journal of the American Chemical Society
Article
Scheme 5. Proposed Nitrosyl Activation Pathway and an Osborn-type Catalytic Cycle
for II-SiMe3-1 and 9.8 kcal/mol for II-SiMe3-2. The angle
between the two iodide ligands in II-SiMe3-1 widens up to
102.3°, which emphasizes the additional potential for
This activates the silicon group for R′3Si+ transfer to the
Lewis basic nitrosyl oxygen atom of 4 via the TS-1 transition
state accompanied by full hydride transfer from the hydrosilane
to B(C6F5)3. As verified by the DFT calculations, intermediate
I-1 with the Si atom oriented toward the H2 is favored over its
isomer I-1′ with the Si atom oriented toward the iodo ligand.
Subsequently, nitrosyl bending occurs affording the 16e−
Re(III) species I-2, in which the lone pair electron of the
bent nitrosyl moiety is anti to the H2 ligand. A vacant site is
expected to be generated forming a seven-coordinate Re(III)
species. However, this is apparently prevented by the cis-
labilizing effect of the two iodides in I-2 promoting the six-
coordinate geometry. In I-2, the Re center is more electron-
deficient turning the η2-H2 ligand acidic, or in other words,
highly polarized toward Hδ+-Hδ‑.16,27 The crypto-H+ is readily
transferred to the neighboring in-plane cis-phosphorus atom
affording the hydride I-3 accompanied by formation of the
phosphonium borate 6.22
coordination of a ligand. At 135° II-SiMe3-1 could be
calculated to be 8.9 kcal/mol more stable than II-SiMe3-2.
The lone pair of the NNO atom on the bent nitrosyl moiety of
II-SiMe3-1 repels the neighboring iodo ligand resulting in a N−
Re−I2 angle of 101.9° and N−Re−I1 angle of 155.7°.
Compared to this, repulsion of the dihydrogen ligand by the
lone pair of the NNO atom in the bent II-SiMe3-2 is less
effective showing a reduced N−Re−I2 angle of 84.5°. Further
bending of the Re−N−O angle to 120.5° in II-SiMe3-1
requires 20.4 kcal/mol relative to the linear nitrosyl case.
Similarly, the bond orders of the Re−N and N−O bonds
decrease along with the extent of NO bending.
2.3.6. Proposed Activation Pathway and Catalytic Cycle.
The activation pathway of the rhenium catalysts by the
“hydrosilane/B(C6F5)3” co-catalytic systems is expected to go
along with the generation of phosphonium borate, as depicted
in Scheme 5. It is most likely that in the initiation stage
reversible coordination of the B(C6F5)3 Lewis acid to the Si−H
bond of the hydrosilane occurs affording an intermediate
bearing hyperconjugative B···H−Si and B−H···Si resonance
structures.10,11
The “superelectrophilic” 14e−, Re(III) hydride species I-3
serves as the basic catalytic intermediate driving the hydro-
genation cycle along an Osborn-type scheme with alkene before
H2 addition.6d,e Importantly, the vacant site in I-3 can open up
4096
dx.doi.org/10.1021/ja400135d | J. Am. Chem. Soc. 2013, 135, 4088−4102

Journal of the American Chemical Society
Article
Figure 7. B2PLYP-D3/def2-TZVP calculated ΔE (kcal/mol) of the structurally optimized model complexes along the catalyst activation course.
on-demand, eventually stabilized by interaction with the lone
pair of the ONO atom. This neighboring group effect provides
protection to the unsaturated species and might account for the
longevity of the catalyst. It should also be emphasized here that
although the pKa value of the bound H2 in the intermediate I-2
cannot be determined due to its high instability, the phosphine
ligand is considered to be basic enough (for instance with a
pKa(aq) of 9.70 for PCy3)28 to abstract a proton from the
polarized H−H ligand supposedly in the transition state of its
elimination. The basicity of PiPr3 is slightly weaker than that of
PCy3, but still in the range that the deprotonation can happen.
This higher basicity of PCy3 might account for the higher
activity of the catalytic system with 3b due to a more efficient
intramolecular heterolytic H2 cleavage and faster phosphine
liberation producing higher concentrations of the I-3 type
species.
catalysts,29 H−H heterolytic cleavage of the polarized H2 ligand
across the Re−N bond could occur affording the intermediate
I-3′. This step would demonstrate a novel cooperative function
of the bent nitrosyl ligand especially suited for H2 catalyzes with
polar reactivity characteristics.30 The N−H moiety in I-3′ is
then thought to be deprotonated by a labile cis-positioned
phosphine ligand affording the 14e− Re(III) hydride I-4′, which
is an isomer of I-3, but more activated due to the lack of
stabilization. I-4′ can enter the Osborn-type hydrogenation
cycle, or eventually isomerize to I-3.
2.3.7. DFT Calculations in Support of the Catalyst
Activation Path. To support the proposed activation pathways,
dispersion corrected DFT (B2PLYP-D3/def2-TZVP//TPSS-
D3/def2-TZVP) calculations were carried out. The free energy
profiles connecting the most plausible species are depicted in
+
Figure 7. Coordination of the SiMe3 moiety to the ONO atom
Parallel to this, another pathway is proposed based on the
thermodynamically less favored isomer I-2′, which can be
derived from either nitrosyl bending of I-1′ or isomerization of
I-2. The fact that several singlets in a close chemical shift range
evolved in the 29Si NMR spectra of stoichiometric reactions
with the Et3SiH/B(C6F5)3 reagent (Figure 4) supports the idea
that structurally related species exist which all bear Re-NO-Si
moieties. I-2′ is quite interesting due to the cis-alignment of the
H2 ligand and the lone pair of the NNO atom available by
bending. In a similar fashion to the Noyori-Morris type
of the model PMe3 complex is highly exothermic with an
energy of −51.5 and −59.2 kcal/mol forming I-1(PMe3) and I-
1′(PMe3), respectively (reactants are included as structure I-0
in Figure 7). A bending angle of 135° of the Re−N−O unit
leads to I-2(PMe3) and I-2′(PMe3) positioned at somewhat
elevated energies of 8.7 and 9.8 kcal/mol relative to I-1(PMe3)
and I-1′(PMe3), respectively. Heterolytic cleavage of the H−H
bond by the basic nitrogen atom of I-2(PMe3) would require
an additional energy of 13.2 kcal/mol to form I-3′(PMe3)(cis)
bearing syn-positioned Re−H and N−H bonds at an energy of
4097
dx.doi.org/10.1021/ja400135d | J. Am. Chem. Soc. 2013, 135, 4088−4102

Journal of the American Chemical Society
Article
30.8 kcal/mol relative to the reference conformer. The
formation of the alternative isomer possessing trans-Re−H
and N−H bonds is at a slightly lower energy of 28.6 kcal/mol.
The H−H bond cleavage in I-2(PMe3) leads to another isomer
bearing anti-Re−H and N−H bonds at an energy of 10.4 kcal/
mol. However, the transition state for the H−H splitting
process could not be located. The deprotonation of the N−H
bond by a phosphine ligand demands an energy of 9.3 kcal/mol
starting from I-3′(PMe3)(cis). Alternatively, the energy of the
monophosphine isomer I-3(PMe3) with the lone pair on the
NNO atom trans to the Re−H was calculated to be 39.9 kcal/
mol relative to the start conformer. Since this reaction course is
highly endothermic, it is kinetically not very probable.
Coordination of either olefin or H2 to I-4′(PMe3) or I-
3(PMe3) drastically decreases the energy of the newly formed
species, and would therefore render the overall hydrogenation
reaction thermodynamically feasible. The high energy level of
the 14e− hydride I-4′(PMe3) or 1-3(PMe3) with bent NO
ligands also implies that the 16e− hydride isomers bearing a
linear NO ligand2c,d might also be involved in the catalytic
cycle.
2.3.8. Exclusion of the Formation of an Initial Iodo
Hydride Complex from 4a,b and the Hydrosilane. In an
earlier publication, we discovered the pentacoordinate bromo
complex of the type [ReBr(H)(NO)(PR3)2] to show high
activities in the hydrogenation of olefins when boron Lewis
acids were applied as sole co-catalysts.8 In contrast to the
bromo case where the bromo hydride could quantitatively be
obtained from the dibromo precursor, the analogous reaction of
the [ReI2(NO)(PCy3)2(η2-H2)] (4b) with Et3SiH at 100 °C
afforded only a complex mixture and a large amount of
unreacted 4b. The much lower tendency of the iodo complex
to react with hydrosilanes might be attributed to the fact that
the Si−I bond (about 56 kcal/mol) is of much lower
thermodynamic strength than the Si−H bond (about 76
kcal/mol), while the Si−Br bond (74 kcal/mol) is close in
energy to the Si−H bond.31 Nevertheless, in order to exclude
the pathway involving initial formation of an iodo hydride
complex from 3 and Et3SiH followed by B(C6F5)3 activation,
the five coordinate iodo hydride 8b was exemplarily prepared
by a route as depicted in Scheme 6. Starting from [ReBr2(H)-
Figure 8. Molecular structure of [ReI2(H)(NO)(PCy3)2] (7b) with
50% probability displacement ellipsoids. Hydrogen atoms have been
omitted for clarity. Selected bond lengths (Å) and angles (deg): I(1A)-
Re(1), 2.7520(2); I(2A)-Re(1), 2.7498(3); P(1)−Re(1), 2.4794(6);
P(2)−Re(1), 2.4865(6); N(1A)-O(1A), 1.195(3); P(1)−Re(1)−P(2),
148.40(2); P(1)−Re(1)−I(2A), 105.825(15); P(2)−Re(1)−I(2A),
105.770(16); P(1)−Re(1)−H, 73.4(12); P(2)−Re(1)−H, 75.1(12).
isolated in 52% yield by extraction with pentane. The IR
spectrum showed a strong ν(NO) absorption at 1657 cm−1.
The 1H NMR spectrum in benzene-d6 exhibited a broad
unresolved triplet at δ −15.9 ppm for the hydride ligand. The
31P{1H}-NMR spectrum displayed a singlet at δ 31.2 ppm.
Finally, the catalytic performance of 8b in the hydrogenation of
1-hexene was tested in the presence of B(C6F5)3. At 23 °C
under 10 bar of H2, a TON of 1830 was obtained within 30 min
corresponding to a conversion of 18% and a TOF of 3661 h−1,
which is by far not comparable to those TOFs obtained with
the “3b/Et3SiH/B(C6F5)3” system. On the basis of these
experiments, we excluded the formation of iodo hydride
complexes of type 8b in the course of the activation process of
the “3b/Et3SiH/B(C6F5)3” catalytic system.
3. CONCLUSIONS
In this article, the first example of a “catalytic nitrosyl effect”
with activation by NO bending was demonstrated. This
bending generated an open site, which contributed to the
reaction course of highly efficient hydrogenation catalyzes. The
“Re(I) diiodide/hydrosilane/B(C6F5)3” co-catalytic systems
could be tuned to eventually furnish the most efficient rhenium
catalyst so far known for the hydrogenation of alkenes. Both the
activity and longevity are superior to previously reported
catalytic systems. The remarkable catalytic performance is
boosted by the bending of the ReNOSi moiety assisted by the
in situ attached silylium ions followed by intramolecular
heterolytic dihydrogen cleavage in the catalyst’s activation
course. The coordination of a Lewis acid to NO is vital to
accomplish excellent catalytic performance. DFT calculations
could show this mechanism to be plausible and proof the
importance of the silyl-coordination for the reactions course.
This work demonstrates ways of utilization of middle transition
metals via a tuning of ligand effects to replace platinum group
metals in highly efficient catalytic applications. A new protocol
and strategy to improve the catalytic performance of nitrosyl
transition metal complexes could be established. Further
exploration and exploitation in catalysis of the Lewis acid
induced nitrosyl bending, denoted as the “catalytic nitrosyl
effect”, is currently under investigation in our group.
Scheme 6. Synthesis of the Re(I) Iodo Hydride 8b
(NO)(PCy3)2],32 halide exchange with NaI in THF afforded at
23 °C within 15 h the Re(II) iodo hydride [ReI2(H)(NO)-
(PCy3)2] (7b) in 82% yield. The IR spectra showed a strong
ν(Re−H) absorption at 2006 cm−1 and a ν(NO) absorption at
1684 cm−1. The molecular structure of 7b was established by a
X-ray diffraction study, as depicted in Figure 8. Similar to the
bromo derivative, 7b adopted a pseudo-octahedral geometry
around the rhenium center. The two trans-phosphine ligands
are bending over strongly toward the hydride ligand with a P1−
Re1−P2 angle of 148.40(2)°.
Reduction of 7b with Zn in THF at 23 °C for 15 h afforded a
deep purple solution, from which the five-coordinate Re(I)
iodo hydride complex [ReI(H)(NO)(PCy3)2] (8b) could be
4098
dx.doi.org/10.1021/ja400135d | J. Am. Chem. Soc. 2013, 135, 4088−4102

Journal of the American Chemical Society
4. EXPERIMENTAL SECTION
Article
4.2.2. [ReI2(NO)(PiPr3)2(H2O)] (3a). In a 20 mL vial in glovebox, 2a
(36 mg, 0.05 mmol) and excess of NaI (105 mg, 0.7 mmol) were
mixed in 3 mL of THF. A yellow-brown solution immediately formed,
and after stirring at room temperature for 2 h, a unique organometallic
species is present in solution as indicated by 31P NMR spectra. The
mixture was filtered through a glass sintered funnel and the filtrate was
dried in vacuo. The residue was extracted with benzene (2 × 2 mL)
and dried in vacuo. The obtained brown solid was washed again with
pentane (2 × 2 mL), dried in vacuo giving a brown solid. Yield: 38 mg,
0.047 mmol, 94%. IR (KBr, cm−1): ν(O−H) 3435 (br), ν(C−H)
4.1. General Experimental. All manipulations were carried out
under an atmosphere of dry nitrogen using standard Schlenk
techniques or in a glovebox (M. Braun 150B-G-II) filled with dry
nitrogen. Solvents were freshly distilled under N2 by employing
standard procedures and were degassed by freeze−thaw cycles prior to
use. The deuterated solvents were dried with sodium/benzophenone
(toluene-d8, benzene-d6, THF-d8) or P2O5 (chlorobenzene-d5) and
hvacuum transferred for storage in Schlenk flasks fitted with Teflon
1
valves. H NMR, 13C{1H}-NMR, 31P{1H}-NMR, 19F NMR and 11B
1
2961, 2926, 2871, 1458, ν(NO) 1674. H NMR (300.08 MHz, THF-
NMR data were recorded on a Varian Gemini-300, Varian Mercury
200, or Bruker DRX 500 spectrometers using 5 mm diameter NMR
tubes equipped with Teflon valves, which allow degassing and further
introduction of gases into the probe. Chemical shifts are expressed in
parts per million (ppm). 1H and 13C{1H}-NMR spectra were
referenced to the residual proton or 13C resonances of the deuterated
solvent. All chemical shifts for the 31P{1H}-NMR data are reported
downfield in ppm relative to external 85% H3PO4 at 0.0 ppm. Signal
patterns are reported as follows: s, singlet; d, doublet; t, triplet; m,
multiplet. IR spectra were obtained by using KBr pellets or ATR
methods with a Bio-Rad FTS-45 FTIR spectrometer. Microanalyses
were carried out at the Anorganisch-Chemisches Institut of the
University of Zurich. Complexes 1 were prepared according to
reported procedures.4b B(C6F5)3, Na15NO2, AlF3 and diverse alkenes
were purchased from Aldrich and used without further purification. 1-
Hexene, 1-octene, cyclooctene, and 1,5-cyclooctadiene were purified
by distillation.
4.2. General Procedure for Hydrogenation of Alkenes
Catalyzed by the “[Re]/Hydrosilane/B(C6F5)3” System. In a 30
mL steel autoclave equipped with a stirring bar, certain amounts of the
substrate alkene (e.g., 1-hexene, 2.5 mL, 20 mmol; 1-octene, 2.5 mL,
16 mmol; cyclooctene, 1 mL, 8 mmol; 1,5-cyclooctadiene, 1 mL, 8
mmol; cyclohexene, 2 mL, 20 mmol; styrene, 1 mL, 9 mmol; 1,7-
octadiene, 1 mL, 7 mmol), appropriate amount of rhenium catalyst
(3a, 1.5 mg, 0.002 mmol; 3b, 2.0 mg, 0.002 mmol; 4a, 1.6 mg, 0.002
mmol; 4b, 2.1 mg, 0.002 mmol; 7a, 1.1 mg, 0.002 mmol; 7b, 1.6 mg,
0.002 mmol), 5 equiv of B(C6F5)3 (5.2 mg, 0.01 mmol) and 5 equiv of
hydrosilane (Et3SiH, 1.6 μL, 0.01 mmol; Me2PhSiH, 1.6 μL, 0.01
mmol; MePh2SiH, 2.0 μL, 0.01 mmol; iPr3SiH, 2.0 μL, 0.01 mmol;
Ph3SiH, 2.6 mg, 0.01 mmol) were mixed. After being flushed with 3.7
bar of H2 thrice, the system was charged with 10 bar of H2 and kept
stirring at either ambient temperature or 90 °C. The progress of the
reactions was monitored by a Buechi Pressflow Gas Controller and the
conversion of the hydrogenations was calculated based on the
consumption of H2 in the system. When 40 bar of H2 was employed,
the reaction course was monitored by the decreased pressure of
autoclave. The hydrogenation products were characterized by 1H
NMR spectroscopy in CDCl3. The TONs and TOFs values were
determined based on the consumption of H2.
d8, ppm): δ 5.43 (s, 2H, H2O), 3.40 (m, 6H, PCH(CH3)2), 1.33−1.43
(m, 36H, PCH(CH3)2). 13C{1H}-NMR (75.47 MHz, THF-d8, ppm):
δ 27.5 (t, J(PC) = 10 Hz, P-CH), 20.3 (s, PCH(CH3)2). 31P{1H}-NMR
(121.47 MHz, THF-d8, ppm): δ −10.7. Anal. Calcd for
C18H44I2NO2P2Re (809,05): C, 26.74; H, 5.49; N, 1.73. Found: C,
26.87; H, 5.36; N, 1.65.
4.2.3. [ReI2(NO)(PCy3)2(H2O)] (3b). Method 1: In a 100 mL round-
bottom flask, H2O2 (5 mL, 30%) was added dropwise to Re powder
(1.5 g, 8.0 mmol) at 0 °C. NEt4I (2.5 g, 9.5 mmol) was then added
and the mixture was stirred for 3 h at 90 °C. After that, additional part
of NEt4I (2.5 g, 9.5 mmol) was added, and the mixture was dissolved
in 30 mL of HI (57%) and 3 mL of H3PO2 (50%). NO gas was
bubbled through the solution at 110 °C. Twenty hours later, the
reaction mixture was cooled down to room temperature and was
filtered. The collected dark residue was extracted with MeOH (10 × 5
mL) and the solvent was evaporated in vacuo. The obtained residue
was further washed with H2O (5 × 10 mL) and diethyl ether (5 × 10
mL), dried in vacuo affording a dark blue solid. Yield: 1.8 g. A mixture
of this crude compound (180 mg) and PCy3 (170 mg, 0.61 mmol)
suspended in ethanol (5 mL) was stirred at 90 °C for 15 h. During this
reaction time, a brown precipitate gradually formed. The reaction
mixture was cooled to room temperature and filtered. The residue was
washed with ethanol (2 × 2 mL) and dried in vacuo giving a brown
solid. Yield: 183 mg, 20% (overall). IR (ATR, cm−1): ν(O−H) 3566
(br), ν(C−H) 2916, 2848, 1444, ν(NO) 1644. 1H NMR (300.08
MHz, THF-d8, ppm): δ 5.20 (s, 2H, H2O), 3.16 (br, 6H, P-CH),
1.24−2.13 (m, 60H, CH2). 13C{1H}-NMR (75.47 MHz, THF-d8,
ppm): δ 36.3, 29.1, 27.8, 26.7. 31P{1H}-NMR (121.47 MHz, THF-d8,
ppm): δ −19.1. Anal. Calcd for C36H68I2NO2P2Re (1048,89): C,
41.22; H, 6.53; N, 1.34. Found: C, 41.09; H, 6.70; N, 1.39. Method 2:
[NEt4]2[ReBr5(NO)] (1.0 g, 1.14 mmol) was treated with excess of
NaI (10.0 g, 7.33 mmol) in 20 mL of THF. The mixture was kept
stirring at 90 °C for 24 h. The resultant deep blue solution was filtered
through a glass sintered funnel to remove NaBr. The collected solution
was dried in vacuo, and further washed with H2O (3 × 5 mL) to afford
a crude Re(II) iodonitrosyl product. Single-crystals were formed from
layered solutions of THF and diethyl ether, and were shown by X-ray
diffraction studies to be the [NEt4][ReI4(NO)(THF)] complex. IR
(ATR, cm−1): ν(C−H) 2969, 2921, 2852, ν(NO) 1734. Anal. Calcd
for C12H28I4N2O2Re (926.19): C, 15.56; H, 3.05; N, 3.02. Found: C,
15.29; H, 3.10; N, 3.15. As a matter of fact, purification of the crude
product is not necessary as the impurity could be removed in the
subsequent step. The reaction of the crude Re(II) compound (500
mg) with PCy3 (650 mg, 2.32 mmol) in 10 mL of ethanol and 10
drops of H2O afforded at 90 °C within 15 h a brown precipitate, which
was isolated and washed with ethanol (4 × 5 mL) affording the
compound 3b. Yield: 351 mg, 67% (overall).
4.2.1. [ReBr2(NO)(PiPr3)2(H2O)] (2a). In a 3 mL Young-NMR-Tube,
1a (70 mg, 0.10 mmol) was mixed with degassed H2O (90 μL) and
THF (0.5 mL). The solution was kept at 90 °C for 5 min resulting in
the formation of an orange solution. 31P NMR spectra indicated the
full conversion of the starting material. The reaction mixture was
cooled to room temperature and filtered through Celite. The solvent
was evaporated and the residue was further washed with pentane (2 ×
2 mL), dried in vacuo giving an orange solid: 69 mg; yield, 96%. 2a: IR
(KBr, cm−1): ν(O−H) 3428 (br), ν(C−H) 2963, 2924, 2873, 1460,
4.2.4. [ReI2(NO)(PR3)2(η2-H2)] (4, R = iPr a, Cy b). In a 50 mL
Young-Schlenk tube, 3a (34 mg, 0.04 mmol) or 3b (54 mg, 0.05
mmol) was mixed with excess of anhydrous MgSO4 in 3 mL of
benzene. The N2 atmosphere was replaced with 1.3 bar of H2 by using
a freeze−pump−thaw cycle. After warming up to room temperature,
NMR spectroscopy indicated the instantaneous formation of the
hydrogen coordinated rhenium diiodo complex 4a or 4b in 100% in
situ yield. The reaction was kept at room temperature for overnight,
the mixture was filtered through a glass sintered funnel and the filtrate
was dried in vacuo. 4a: 25 mg. Yield: 74%. IR (ATR, cm−1): ν(C−H)
2960, 2922, 2873, 1456, ν(NO) 1719. 1H NMR (300.08 MHz,
benzene-d6, ppm): δ 3.06 (m, 6H, PCH(CH3)2), 1.28−1.39 (m, 18H,
1
ν(NO) 1667. H NMR (300.08 MHz, THF-d8, ppm): δ 5.72 (s, 2H,
H2O), 3.06 (m, 6H, PCH(CH3)2), 1.32−1.38 (m, 36H, PCH(CH3)2).
13C{1H}-NMR (75.47 MHz, THF-d8, ppm): δ 25.6 (t, J(PC) = 10 Hz,
P-CH), 20.0 (s, PCH(CH3)2). 31P{1H}-NMR (121.47 MHz, THF-d8,
1
ppm): δ −5.9. H NMR (300.08 MHz, benzene-d6, ppm): δ 4.90
(broad, 2H, H2O), 3.07 (m, 6H, PCH(CH3)2), 1.26−1.36 (m, 36H,
PCH(CH3)2). 13C{1H}-NMR (75.47 MHz, benzene-d6, ppm): δ 25.1
(t, J(PC) = 10 Hz, P-CH), 20.0 (s, PCH(CH3)2). 31P{1H}-NMR
(121.47 MHz, benzene-d6, ppm): δ −4.7. Anal. Calcd for
C18H44Br2NO2P2Re (713,08): C, 30.26; H, 6.21; N, 1.96. Found: C,
30.60; H, 6.11; N, 1.72.
4099
dx.doi.org/10.1021/ja400135d | J. Am. Chem. Soc. 2013, 135, 4088−4102

Journal of the American Chemical Society
Article
2
PCH(CH3)2), 1.06−1.16 (m, 18H, PCH(CH3)2), 0.89 (t, J(HP) = 22
161.0 [iPr3PH]+, 513.0 [HB(C6F5)3]−. This product was invariably
observed in the resultant solution in each 3a or 2a catalyzed
hydrogenation of 1-hexene. Similarly, the formation of 6b was
observed during the hydrogenation courses of the “3b/hydrosilane/
B(C6F5)3” catalytic system. For instance, the hydrogenation of 1-
hexene (3 mL) with 3b (40 mg, 0.04 mmol), Me2PhSiH (16 μL, 0.10
mmol) and B(C6F5)3 (41 mg, 0.08 mmol) under 10 bar of H2 afforded
a brown precipitate, from which colorless single crystals of 6b were
1
Hz, 2H, H2). H NMR (300.08 MHz, THF-d8, ppm): δ 3.18 (m, 6H,
2
PCH(CH3)2), 1.34−1.43 (m, 36H, PCH(CH3)2), 0.64 (t, J(HP) = 20
Hz, 2H, H2). 13C{1H}-NMR (75.47 MHz, benzene-d6, ppm): δ 26.1
(t, J(PC) = 11 Hz., P-CH), 20.5 (s, PCH(CH3)2), 19.8 (s, PCH(CH3)2).
31P{1H}-NMR (121.47 MHz, benzene-d6, ppm): δ 3.09. Anal. Calcd
for C18H44I2NOP2Re (792.51): C, 27.28; H, 5.60; N, 1.77. Found: C,
26.99; H, 5.51; N, 1.72. 4b: 39 mg, 76%. IR (ATR, cm−1): ν(C−H)
1
1
obtained from a layered solution of pentane and chlorobenzene. H
2926, 2850, 1445, ν(NO) 1710. H NMR (300.08 MHz, benzene-d6,
3
ppm): δ 1.10−3.00 (m, 68H, P(C6H11)3, H2). 1H NMR (300.08 MHz,
THF-d8, ppm): δ 1.31−2.87 (m, 66H, P(C6H11)3), 0.80 (t, 2J(HP) = 21
Hz, 2H, H2). 13C{1H}-NMR (75.47 MHz, benzene-d6, ppm): δ 36.5
(t, J(PC) = 12 Hz, P−C), 30.9, 30.3, 27.4, 26.6. 31P{1 NMR (121.47
MHz, benzene-d6, ppm): δ −4.93. Anal. Calcd for C36H68I2NOP2Re
(1032.89): C, 41.86; H, 6.64; N, 1.36. Found: C, 41.69; H, 6.56; N,
1.30. In comparison, the same reaction carried out in THF-d8 afforded
only 48% (4a) or 75% (4b) of the hydrogen coordinated product,
which remained in an equilibrium with the starting material 3a(b) at
room temperature over 72 h.
NMR (300.08 MHz, C6D5Cl, ppm): δ 4.80 (dxq, 1H, JHH = 4 Hz,
1JHP = 446 Hz, PH), 1.04−2.08 (m, 66H, P(C6H11)3). 13C{1H}-NMR
(75.47 MHz, C6D5Cl, ppm): δ 171.47, 125.52, 46.72, 28.24, 26.70,
15.62. 31P{1H}-NMR (121.47 MHz, C6D5Cl, ppm): δ 32.9 (s). 11B
1
NMR (96.28 MHz, C6D5Cl, ppm): δ −27.29 (d, JHB = 69 Hz). 19F
NMR (282.33 MHz, C6D5Cl, ppm): δ −134.16 (m, 6F, ortho-C6F5),
1
−164.93 (t, JCF = 21 Hz, 3F, para-C6F5), −167.95 (m, 6F, meta-
C6F5). MS (ESI): m/z 281.2 [Cy3PH]+, 513.0 [HB(C6F5)3]−. Anal.
Calcd for C36H35BF15P (794.42): C, 54.43, H, 4.44. Found: C, 54.51;
H, 4.40.
4.2.7. Deuterium Isotope Effect of “3/Hydrosilane/B(C6F5)3”
Catalyzed Hydrogenation of 1-Hexene. In a 30 mL steel autoclave
equipped with a stirring bar, rhenium complex (3a, 1.5 mg, 0.0025
mmol; 3b, 2.0 mg, 0.0025 mmol), B(C6F5)3 (5.2 mg, 0.01 mmol) and
hydrosilane (Me2PhSiH, 1.6 μL, 0.01 mmol; Et3SiH, 1.6 μL, 0.01
mmol) were dissolved in 2.5 mL of 1-hexene. After being flushed with
3.7 bar of D2 thrice, the system was charged with 10 bar of D2 and kept
stirring at ambient temperature. After the full conversion was achieved,
the supernatant solution was separated from the precipitate and
further identified by 2H NMR spectroscopy as purely 1,2-hydro-
T1 measurements were carried out at room temperature in
benzene-d6 at an NMR field strength of 11.7 T.
4.2.5. [ReI2(NO)(PR3)2(CO)] (5, R = iPr a, Cy b). In a 3 mL Young-
NMR-Tube, 3a (17 mg, 0.02 mmol) or 3b (21 mg, 0.02 mmol) was
dissolved in 0.5 mL of benzene. The N2 atmosphere was replaced with
1 bar of CO by using a freeze−pump−thaw cycle. After being kept at
room temperature for 15 h, NMR spectroscopy indicated complete
formation of the carbon monoxide coordinated rhenium diiodo
complex 5a or 5b in 100% in situ yield. The mixture was filtered
through a glass sintered funnel and the filtrate was dried in vacuo. The
residue was washed with pentane (2 × 1 mL), dried affording a brown-
yellow solid. 5a: 13 mg, Yield: 79%. IR (ATR, cm−1): ν(C−H) 2962,
2932, 2874, 1453, ν(CO) 1969, ν(NO) 1723. 1H NMR (300.08 MHz,
benzene-d6, ppm): δ 3.00 (m, 6H, P-CH), 1.37−1.44 (m, 18 H, CH3),
1.10−1.17 (m, 18 H, CH3). 13C{1H}-NMR (75.47 MHz, CDCl3,
ppm): δ 26.65 (t, J(PC) = 12 Hz, P-CH), 21.05, 19.62. 31P{1H}-NMR
(121.47 MHz, CDCl3, ppm): δ −6.54 (s). Anal. Calcd for
C19H42I2NO2P2Re (818.51): C, 27.88; H, 5.17; N, 1.71. Found: C,
27.54; H, 5.05; N, 1.70. 5b: 12 mg, 56%. IR (ATR, cm−1): ν(C−H)
2919, 2849, 1444, ν(CO) 1970, ν(NO) 1712. 1H NMR (300.08 MHz,
benzene-d6, ppm): δ 1.22−2.92 (m, 66H, P(C6H11)3). 13C{1H}-NMR
(75.47 MHz, CDCl3, ppm): δ 37.00 (t, J(PC) = 11 Hz, P-CH), 31.24,
29.85, 27.76, 26.58. 31P{1H}-NMR (121.47 MHz, CDCl3, ppm): δ
−14.51 (s). Anal. Calcd for C37H66I2NO2P2Re (1058.89): C, 41.97; H,
6.28; N, 1.32. Found: C, 42.05; H, 6.55; N, 1.17. When the same
reaction was investigated by NMR and IR within 5 min, formation of
the carbonyl isomer 5′ was observed in quantitative yield. 5a′: IR
(ATR, cm−1): ν(C−H) 2962, 2930, 2873, 1457, ν(CO) 2082, ν(NO)
2
genated product. C4H9CHDCH2D: H NMR (46.06 MHz, toluene,
ppm): δ 2.11 (s, −CHD−), 1.70 (s, −CH2D). The precipitate was
isolated and further verified as [DPiPr3][HB(C6F5)3]. 31P{1H}-NMR
(121.47 MHz, THF-d8, ppm): δ 41.5 (s). 11B NMR (96.28 MHz,
1
THF-d8, ppm): δ −25.11 (d, JHB = 90 Hz).
4.2.8. H2/D2 Scrambling Experiments. In a 3 mL Young-NMR-
tube, 3a (4.5 mg, 0.006 mmol), B(C6F5)3 (10.4 mg, 0.02 mmol) and
Me2PhSiH (3.2 μL, 0.02 mmol) were dissolved in 0.5 mL of toluene-
d8. The nitrogen atmosphere was replaced with 1100 mbar of H2 and
D2 in a 1:1 ratio using a freeze−pump−thaw cycle. The solution was
kept at 23 °C for 10 min, and was investigated by NMR spectroscopy
showing the formation of HD along with 4a and the phosphonium
borate 6a. 1H NMR (199.95 MHz, toluene-d8, ppm, 296 K): δ 4.41 (t,
3
1
J = 43 Hz, HD), 4.51 (dxq, 1H, JHH = 4 Hz, JHP = 458 Hz, P−H).
31P{1H}-NMR (80.94 MHz, toluene-d8, ppm): δ 42.2 (6a), 3.33 (4a).
4.2.9. [ReI2(15NO)(PCy3)2(η2-H2)] (4b-15N). First the 15N-enriched
Re(II) precursor [NEt4]2[Re(15NO)Br5] was prepared according to
reported procedure by passing 15NO gas, which was produced from
the reaction of Na15NO2 with FeSO4·7H2O and H2O. 430 mg of the
Re(II) precursor was obtained from 1 g of Re and 1 g of Na15NO2
which gives a low yield of ca. 10% due to the inefficient reaction of
rhenium oxide with generated 15NO. By following the synthetic
procedure described for 3b and 4b, the fully 15N enriched complex
4b-15N was prepared in an overall yield of 7%. 15N NMR (50.69 MHz,
1
1708. H NMR (300.08 MHz, benzene-d6, ppm): δ 3.15 (m, 6H, P-
CH), 1.19−1.26 (m, 36 H, CH3). 31P{1H}-NMR (121.47 MHz,
CDCl3, ppm): δ −16.67 (s). 5b′: IR (ATR, cm−1): ν(C−H) 2924,
2852, 1443, ν(CO) 2074, ν(NO) 1701. 1H NMR (300.08 MHz,
benzene-d6, ppm): δ 1.29−3.21 (m, 66H, P(C6H11)3). 31P{1H}-NMR
(121.47 MHz, CDCl3, ppm): δ −26.29 (s).
2
toluene-d8, ppm): δ −49.4 (t, JNH = 5 Hz).
4.2.6. [R3PH][HB(C6F5)3] (6, R = iPr a, Cy b) Formed in the
Catalytic Hydrogenation of 1-Hexene by the “[Re]/Hydrosilane/
B(C6F5)3” Reagent. In a 30 mL steel autoclave equipped with a stirring
bar, 3a (4.5 mg, 0.0075 mmol), B(C6F5)3 (10.4 mg, 0.02 mmol) and
Et3SiH (5.0 μL, 0.03 mmol) were dissolved in 2.5 mL 1-hexene. After
charging with 10 bar of H2, hydrogenation occurred and the catalysis
was deliberately terminated by removing the H2 atmosphere after 10
min corresponding to an approximate half conversion. Precipitates
were observed in the reaction vessel and separated from the solution,
4.2.10. [ReI2(H)(NO)(PCy3)2] (7b). In a 3 mL Young-NMR-Tube,
[ReBr2(H)(NO)(PCy3)2] (65 mg, 0.07 mmol) and excess of NaI (230
mg, 1.53 mmol) were mixed in 2 mL of THF. The mixture was kept
stirring at 23 °C for 15 h to afford a dark-brown solution. The mixture
was filtered through a glass sintered funnel to remove the NaBr
byproduct. The filtrate was dried in vacuo and further extracted with
toluene (3 × 2 mL) and dried to afford the Re(II) diiodo hydride as a
black-brown solid. Yield: 59 mg, 82%. IR (ATR, cm−1): ν(C−H) 2921,
2855, 1437, ν(Re−H) 2006, ν(NO) 1684. Anal. Calcd for
C36H67I2NOP2Re (1031.89): C, 41.90; H, 6.54; N, 1.36. Found: C,
41.61; H, 6.79; N, 1.26.
1
dried in vacuo. 6a: H NMR (300.08 MHz, C6D5Cl, ppm): δ 5.62
3
1
(dxq, 1H, JHH = 4 Hz, JHP = 458 Hz, PH), 2.88 (m, 6H,
PCH(CH3)2), 1.37−1.49 (m, 36H, PCH(CH3)2). 13C{1H}-NMR
(75.47 MHz, C6D5Cl, ppm): δ 19.31, 18.79, 17.19, 17.15. 31P{1H}-
NMR (121.47 MHz, C6D5Cl, ppm): δ 43.9 (s). 11B NMR (96.28
4.2.11. [ReI(H)(NO)(PCy3)2] (8b). In a 20 mL vial in glovebox, 7b
(40 mg, 0.04 mmol) was treated with excess of zinc powder in THF
solution and the mixture was kept stirring at 23 °C (Cautions: reaction
at higher temperatures afforded only a complex mixture) for overnight
to afford a dark-purple solution. The excess of zinc was removed by
1
MHz, C6D5Cl, ppm): δ −25.44 (d, JHB = 92 Hz). 19F NMR (282.33
MHz, C6D5Cl, ppm): δ −133.40 (m, 6F, ortho-C6F5), −163.54 (t, 1JCF
= 19 Hz, 3F, para-C6F5), −166.64 (m, 6F, meta-C6F5). MS (ESI): m/z
4100
dx.doi.org/10.1021/ja400135d | J. Am. Chem. Soc. 2013, 135, 4088−4102

Journal of the American Chemical Society
Article
filtration and the filtrate was dried in vacuo. The resulted brown
residue was further extracted with pentane (5 × 2 mL) and dried in
vacuo giving a purple solid. Yield: 18 mg, 52%. IR (ATR, cm−1): ν(C−
H) 2922, 2851, 1442, ν(NO) 1657. 1H NMR (300.08 MHz, benzene-
d6, ppm): δ 1.25−2.79 (m, 66H, P(C6H11)3), −15.92 (br, 1H, Re−H).
13C{1H}-NMR (75.47 MHz, benzene-d6, ppm): δ 31.28, 30.66, 28.22,
27.05. 31P{1H}-NMR (121.47 MHz, benzene-d6, ppm): δ 31.2 (s).
Anal. Calcd for C36H67INOP2Re (904.98): C, 47.78; H, 7.46; N, 1.55.
Found: C, 47.51; H, 7.40; N, 1.51.
4.3. Computational Details. All calculations were carried out
with the TURBOMOLE 6.3 suite of programs.33 The structures were
optimized with the meta-GGA functional TPSS12 applying the D3-
dispersion correction with Becke-Johnson damping.14 Subsequent
single-point calculations were carried out at the more accurate double-
hybrid density functional B2PLYP-D3 level.15 For both calculations
the large Gaussian-AO basis set def2-TZVP13 and the RI
approximation34 were used. The final level of theory can therefore
be abbreviated as B2PLYP-D3/def2-TZVP//TPSS-D3/def2-TZVP
and has an estimated accuracy of about 1−2 kcal/mol. Further details
can be found in the Supporting Information.
REFERENCES
■
(1) (a) Richter-Addo., G. B.; Legzdins, P. Metal Nitrosyls; Oxford
University Press: New York, 1992; (b) Hayton, T. W.; Legzdins, P.;
Sharp, W. B. Chem. Rev. 2002, 102, 935. (c) Richter-Addo, G. B.;
Legzdins, P.; Burstyn, J. Chem. Rev. 2002, 102, 857. (d) Roncaroli, F.;
Videla, M.; Slep, L. D.; Olabe, J. A. Coord. Chem. Rev. 2007, 251, 1903.
(e) Coppens, P.; Novozhilova, I.; Kovalevsky, A. Chem. Rev. 2002, 102,
861. (f) Machura, B. Coord. Chem. Rev. 2005, 249, 2277.
(2) (a) Enemark, J. H.; Feltham, R. D. Coord. Chem. Rev. 1974, 13,
339. (b) Enemark, J. H.; Feltham, R. D. Top. Stereochem. 1981, 12,
155. (c) Song, J.; Hall, M. B. J. Am. Chem. Soc. 1993, 115, 327.
(d) Ogasawara, M.; Huang, D.; Streib, W. E.; Huffman, J. C.; Gallego-
Planas, N.; Maseras, F.; Eisenstein, O.; Caulton, K. G. J. Am. Chem.
Soc. 1997, 119, 8642. (e) Hodgson, D. J.; Payne, N. C.; McGinnety, J.
A.; Pearson, R. G.; Ibers, J. A. J. Am. Chem. Soc. 1968, 90, 4486.
(f) Pierpont, C. G.; Van Derveer, D. G.; Durlang, W.; Eisenberg, R. J.
Am. Chem. Soc. 1970, 92, 4760. (g) Colman, J. P.; Farnham, P. H.;
Dolcetti, G. J. Am. Chem. Soc. 1971, 93, 1788.
(3) Enemark, J. H.; Feltham, R. D. Proc. Natl. Acad. Sci. U.S.A. 1972,
4.4. X-ray Diffraction Analyses. Single-crystal X-ray diffraction
data were collected at 183(2) K on a Xcalibur diffractometer (Agilent
Technologies, Ruby CCD detector) for all compounds using a single
wavelength Enhance X-ray source with MoKα radiation (λ= 0.71073
Å).35 The selected suitable single crystals were mounted using
polybutene oil on the top of a glass fiber fixed on a goniometer head
and immediately transferred to the diffractometer. Pre-experiment,
69, 3534.
(4) (a) Llamazares, A.; Schmalle, H. W.; Berke, H. Organometallics
2001, 20, 5277. (b) Gusev, D.; Llamazares, A.; Artus, G.; Jacobsen, H.;
Berke, H. Organometallics 1999, 18, 75. (c) Bohmer, J.; Haselhorst, G.;
Wieghart, K.; Nuber, B. Angew. Chem., Int. Ed. Engl. 1994, 13, 1473.
(d) Legzdins, P.; Sayers, S. F. Chem.Eur. J. 1997, 3, 1579. (e) Crease,
A. E.; Legzdins, P. Chem. Commun. 1972, 268. (f) Crease, A. e.;
Legzdins, P. Dalton Trans. 1973, 1501. (g) Lee, K. E.; Arif, A. M.;
Gladysz, J. A. Chem. Ber. 1991, 124, 309. (h) Sharp, W. B.; Legzdins,
P.; Patrick, B. O. J. Am. Chem. Soc. 2001, 123, 8143.
data collection, data reduction and analytical absorption corrections36
35
were performed with the program suite CrysAlisPro
.
The crystal
structures were solved with SHELXS9735 using direct methods. The
structure refinements were performed by full-matrix least-squares on
F2 with SHELXL97.37 All programs used during the crystal structure
determination process are included in the WINGX software.38
PLATON39 was used to check the result of the X-ray analyses. For
more details, see the Crystallographic Information files (Supporting
Information). CCDC-907258 (for 2a), CCDC-907259 (for 3b),
CCDC-907260 (for 5b), CCDC-907261 (for 6b), and CCDC-907262
(for 7b) contain the supplementary crystallographic data (excluding
structure factors) for this paper. These data can be obtained free of
charge from The Cambridge Crystallographic Data Centre via www.
ccdc.cam.ac.uk/data_request/cif.
(5) Negishi, E. Chem.Eur. J. 1999, 5, 411.
(6) (a) de Vries, J. G.; Elsevier, C. J. The Handbook of Homogeneous
Hydrogenation; Wiley-VCH: Weinheim, 2007; For selective examples:
(b) Crabtree, R. H. Acc. Chem. Res. 1979, 12, 331. (c) Osborn, J. A.;
Jardine, F. H.; Young, J. F.; Wilkinson, G. J. Chem. Soc. A. 1966, 12,
1711. (d) Schrock, R. R.; Osborn, J. A. J. Am. Chem. Soc. 1976, 98,
2134. (e) Schrock, R. R.; Osborn, J. A. J. Am. Chem. Soc. 1976, 98,
4450. (f) Landis, C. R.; Halpern, J. J. Am. Chem. Soc. 1987, 109, 1746.
(g) Helmchen, G.; Pfaltz, A. Acc. Chem. Res. 2000, 6, 336.
(h) Roseblade, S. J.; Pfaltz, A. Acc. Chem. Res. 2007, 12, 1402.
(i) Yi, C. S.; Lee, D. W. Organometallics 1999, 18, 5152. (j) Noyori, R.;
Hashiguchi, S. Acc. Chem. Res. 1997, 30, 97.
ASSOCIATED CONTENT
(7) (a) Dudle, B.; Rajesh, K.; Blacque, O.; Berke, H. J. Am. Chem. Soc.
2011, 133, 8168. (b) Jiang, Y.; Blacque, O.; Fox, T.; Frech, C. M.;
Berke, H. Organometallics 2009, 28, 5493. (c) Choualeb, A.;
Maccaroni, E.; Blacque, O.; Schmalle, H. W.; Berke, H. Organometallics
2008, 27, 3474. (d) Jiang, Y.; Berke, H. Chem. Commun. 2007, 3571.
(e) Landwehr, A.; Dudle, B.; Fox, T.; Blacque, O.; Berke, H. Chem.
Eur. J. 2012, 18, 5701.
■
S
* Supporting Information
Experimental Section with detailed synthesis and character-
ization of complexes, the crystallographic details of 2a, 3b, 5b,
1
6b and 7b, various H-, 13C-, 31P-, 19F-, 29Si- and 15N NMR
spectra, representative hydrogenation profiles, the figure of
transformation course from 5a′ to 5a, further computational
details and additional numbers and structures. Crystallographic
information files. This material is available free of charge via the
Internet at http://pubs.acs.org.
(8) Jiang, Y.; Hess, J.; Fox, T.; Berke, H. J. Am. Chem. Soc. 2010, 132,
18233.
(9) (a) Pauling, L. Science 1994, 263, 983. (b) Kim, K. C.; Reed, C.
A.; Elliott, D. W.; Mueller, L. J.; Tham, F.; Lin, L.; Lambert, J. Science
2002, 297, 825. (c) Ree, C. A. Acc. Chem. Res. 1998, 31, 325. (d) Reed,
C. A. Acc. Chem. Res. 2010, 43, 121.
(10) (a) Blackwell, J. M.; Sonmor, E. R.; Scoccitti, T.; Piers, W. E.
Org. Lett. 2000, 2, 3921. (b) Parks, D. J.; Blackwell, J. M.; Piers, W. E.
J. Org. Chem. 2000, 65, 3090. (c) Berkefeld, A.; Piers, W. E.; Parvez, M.
J. Am. Chem. Soc. 2010, 132, 10660.
(11) (a) Rendler, S.; Oestreich, M. Angew. Chem., Int. Ed. 2008, 47,
5997. (b) Hog, D. T.; Oestreich, M. Eur. J. Org. Chem. 2009, 5047.
(c) Klare, H. F. T.; Oestreich, M.; Ito, J.-i.; Nishiyama, H.; Ohki, Y.;
Tatsumi, K. J. Am. Chem. Soc. 2011, 133, 3312. (d) Mewald, M.;
Froehlich, R.; Oestreich, M. Chem.Eur. J. 2011, 17, 9406.
(e) Muether, K.; Froehlich, R.; Mueck-Lichtenfeld, C.; Grimme, S.;
Oestreich, M. J. Am. Chem. Soc. 2011, 133, 12442.
AUTHOR INFORMATION
■
Corresponding Author
hberke@aci.uzh.ch
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
Financial support from the Swiss National Science Foundation,
Lanxess AG, Leverkusen, Germany, the Funds of the University
of Zurich, the DFG and SNF within the project ’Forscher-
gruppe 1175 - Unconventional Approaches to the Activation of
Dihydrogen’ are gratefully acknowledged.
(12) Tao, J.; Perdew, J. P.; Staroverov, V. N.; Scuseria, G. E. Phys.
Rev. Lett. 2003, 91, 146401.
(13) Weigend, F.; Ahlrichs, R. Phys. Chem. Chem. Phys. 2005, 7, 3297.
4101
dx.doi.org/10.1021/ja400135d | J. Am. Chem. Soc. 2013, 135, 4088−4102

Journal of the American Chemical Society
Article
(14) (a) Grimme, S.; Antony, J.; Ehrlich, S.; Krieg, H. J. Chem. Phys.
2010, 132, 154104. (b) Grimme, S.; Ehrlich, S.; Goerigk, L. J. Comput.
Chem. 2011, 32, 1456.
excellent perspective on redox non-innocent ligands, see: Lyaskovskyy,
V.; de Bruin, B. ACS Catal. 2012, 2, 270.
(31) Clayden, J.; Greeves, N.; Warren, S.; Wothers, P. Organic
Chemistry; Oxford University Press: New York, 2001.
(32) Jiang, Y.; Blacque, O.; Fox, T.; Frech, C. M.; Berke, H. Chem.
Eur. J. 2010, 16, 2240.
(15) (a) Grimme, S. J. Chem. Phys. 2006, 124, 034108. (b) Schwabe,
T.; Grimme, S. Phys. Chem. Chem. Phys. 2006, 8, 4398.
(16) (a) Kubas, G. J. Chem. Rev. 2007, 107, 4152. (b) Kubas, G. J.
Metal Dihydrogen and σ-Bond Complexes; Kluwer: New York, 2001;
(c) Kubas, G. J. Acc. Chem. Rev. 1988, 21, 120. (d) Kubas, G. J.; Ryan,
R. R.; Swanson, B. I.; Vergamini, P. J.; Wasserman, H. J. J. Am. Chem.
Soc. 1984, 106, 451. (e) Hoffmann, R. Am. Sci. 2012, 374. (f) Berke,
H. ChemPhysChem 2011, 9, 1837.
(33) Ahlrichs, R.; Armbruster, M. K.; Baer, M.; Baron, H.-P.;
Bauernschmitt, R.; Crawford, N.; Deglmann, P.; Ehrig, M.; Eichkorn,
K.; Elliott, S.; Furche, F.; Haase, F.; Haeser, M.; Haettig, C.; Hellweg,
A.; Horn, H.; Huber, C.; Huniar, U.; Kattannek, M.; Koelmel, C.;
Kollwitz, M.; May, K.; Nava, P.; Ochsenfeld, C.; Oehm, H.; Patzelt, H.;
Rappoport, D.; Rubner, O.; Schaefer, A.; Schneider, U.; Sierka, M.;
Treutler, O.; Unterreiner, B.; von Arnim, M.; Weigand, F.; Weis, P.;
TURBOMOLE, 2008, http://www.turbomole.com.
(17) Matthews, S. L.; Heinekey, D. M. J. Am. Chem. Soc. 2006, 128,
2615.
(18) (a) Balcells, D.; Carbo, J. J.; Maseras, F.; Eisenstein, O.
Organometallics 2004, 23, 6008. (b) Huang, D.; Streib, W. E.;
Eisenstein, O.; Kenneth, K. G. Organometallics 2000, 19, 1967.
(19) Mayr, H.; Basso, N.; Hagen, G. J. Am. Chem. Soc. 1992, 114,
3060.
(20) Welch, G. C.; Cabrera, L.; Chase, P. A.; Hollink, E.; Masuda, J.
D.; Wei, P. R.; Stephan, D. W. Dalton Trans. 2007, 3407.
(21) (a) Stephan, D. W.; Erker, G. Angew. Chem., Int. Ed. 2010, 49,
46. (b) Stephan, D. W. Org. Biomol. Chem. 2008, 6, 1535. (c) Stephan,
D. W. Dalton Trans. 2009, 3129. (d) Chase, P. A.; Stephan, D. W.
Angew. Chem., Int. Ed. 2008, 47, 7433. (e) Kenward, A. L.; Piers, W. E.
Angew. Chem., Int. Ed. 2008, 47, 38. (f) Welch, G. C.; Juan, R. R. S.;
Masuda, J. D.; Stephan, D. W. Science 2006, 314, 1124. (g) Spies, P.;
Schwendemann, S.; Lange, S.; Kehr, G.; Frohlich, R.; Erker, G. Angew.
Chem., Int. Ed. 2008, 47, 7543.
̈
(34) (a) Eichkorn, K.; Treutler, O.; Ohm, H.; Haser, M.; Ahlrichs, R.
̈
Chem. Phys. Lett. 1995, 240, 283. (b) Hattig, C.; Weigend, F. J. Chem.
̈
Phys. 2000, 113, 5154.
(35) Agilent Technologies (formerly Oxford Diffraction), Yarnton,
England, 2011.
(36) Clark, R. C.; Reid, J. S. Acta Crystallogr., Sect. A 1995, 51, 887.
(37) Sheldrick, G. M. Acta Crystallogr., Sect. A 2008, 64.
(38) Farrugia, L. J. J. Appl. Crystallogr. 1999, 32, 837.
(39) Spek, A. L. J. Appl. Crystallogr. 2003, 36, 7.
(22) (a) Ingleson, M. J.; Brayshaw, S. K.; Mahon, M. F.; Ruggiero, G.
D.; Weller, A. S. Inorg. Chem. 2005, 44, 3162. (b) Kubas, G. J. Adv.
Inorg. Chem. 2004, 56, 127. (c) Yi, C. S.; Lee, D. W.; He, Z.;
Rheingold, A. L.; Lam, K.-C.; Concolino, T. E. Organometallics 2000,
19, 2909.
(23) (a) Parkin, G. Acc. Chem. Res. 2009, 42, 315. (b) Iuliis, M. Z. D.;
Morris, R. H. J. Am. Chem. Soc. 2009, 131, 11263. (c) Jones, W. D. Acc.
Chem. Res. 2002, 36, 140. (d) Gould, G. L.; Heinekey, D. M. J. Am.
Chem. Soc. 1989, 111, 5502.
(24) Marsmann, H.; Uhlig, F.; Mikhova, B. Chemical Shifts and
Coupling Constants for Silicon-29, Springer: Berlin, Heidelberg, New
York, 2008.
(25) Mason, J.; Larkworthy, L. F.; Moore, E. A. Chem. Rev. 2002, 102,
913.
(26) (a) Berger, S.; Braun, S.; Kalinowski, H. O. NMR-Spektroskopie
von Nichtmetallen; Georg Thieme Verlag: Stuttgart, NY, 1992. (b) Bell,
L. K.; Mason, J.; Mingos, D. M. P.; Tew, D. G. Inorg. Chem. 1983, 22,
3497. (c) Botto, R. E.; Kolthammer, B. W. S.; Legzdins, P.; Roberts, J.
D. Inorg. Chem. 1979, 18, 2049. For other examples with high δN:
(d) Gaviglio, C.; Ben-David, Y.; Shimon, L. J. W.; Doctorovich, F.;
Milstein, D. Organometallics 2009, 28, 1917. (e) Duffin, P. A.;
Larkworthy, L. F.; Mason, J.; Stephens, A. N.; Thompson, R. M. Inorg.
Chem. 1987, 26, 2034. (f) Evans, D. H.; Mingos, D. M. P.; Mason, J.;
Richards, A. J. Organomet. Chem. 1983, 249, 293.
(27) (a) Jessop, P. G.; Morris, R. H. Coord. Chem. Rev. 1992, 155.
(b) Jia, G.; Morris, R. H. J. Am. Chem. Soc. 1991, 113, 875. (c) Jia, G.;
Lough, A. J.; Morris, R. H. Organometallics 1992, 1, 161.
(d) Cappellani, E. P.; Drouin, S. D.; Jia, G. C.; Maltby, P. A.;
Morris, R. H.; Schweitzer, C. T. J. Am. Chem. Soc. 1994, 116, 3375.
(28) Li, T.; Lough, A. J.; Morris, R. H. Chem.Eur. J. 2007, 13, 3796.
(29) (a) Noyori, R.; Ohkuma, T. Angew. Chem., Int. Ed. 2001, 40, 40.
(b) Noyori, R.; Yamakawa, M.; Hashiguchi, S. J. Org. Chem. 2001, 66,
7931. (c) Morris, R. H. Chem. Soc. Rev. 2009, 38, 2282. (d) Abdur-
Rashid, K.; Clapham, S. E.; Hadzovic, A.; Harvey, J. N.; Lough, A. J.;
Morris, R. H. J. Am. Chem. Soc. 2002, 124, 15104. (e) Morris, R. H.
Coord. Chem. Rev. 2008, 252, 2381. (f) Clapham, S. E.; Hadzovic, A.;
Morris, R. H. Coord. Chem. Rev. 2004, 248, 2201. (g) Maire, P.;
Buettner, T.; Breher, F.; Le Floch, P.; Gruetzmacher, H. Angew. Chem.,
Int. Ed. 2005, 44, 6318.
(30) For an excellent highlight on cooperating ligands in catalysis,
see: Gruetzmacher, H. Angew. Chem., Int. Ed. 2008, 47, 1814. For an
4102
dx.doi.org/10.1021/ja400135d | J. Am. Chem. Soc. 2013, 135, 4088−4102