The formation of aldehydes from the photochemically activated reaction of Cp*Ir(CO)(Cl)(CH2R) complexes with water

Article abstract of DOI:10.1016/j.jorganchem.2012.11.030

The iridium complexes Cp*Ir(CO)(Cl)(CH₂R), which do not contain any β-hydrogen atoms, react with water to form aldehydes. This reaction was found to be photochemically activated, and a possible pathway involving radical intermediates has been proposed and studied experimentally and computationally.

Full text of DOI:10.1016/j.jorganchem.2012.11.030

Journal of Organometallic Chemistry 724 (2013) 275e280
Contents lists available at SciVerse ScienceDirect
Journal of Organometallic Chemistry
journal homepage: www.elsevier.com/locate/jorganchem
The formation of aldehydes from the photochemically activated reaction
of Cp*Ir(CO)(Cl)(CH2R) complexes with water
,
b
a,
Jia Ying Lee a, Wai Yip Fan b , Kar Hang Garvin Mak , Weng Kee Leong
*
*
a Division of Chemistry and Biological Chemistry, Nanyang Technological University, 21 Nanyang Link, Singapore 637371, Singapore
b Department of Chemistry, National University of Singapore, 3 Science Drive 3, Singapore 117543, Singapore
a r t i c l e i n f o
a b s t r a c t
Article history:
The iridium complexes Cp*Ir(CO)(Cl)(CH2R), which do not contain any b-hydrogen atoms, react with
Received 10 September 2012
Received in revised form
19 November 2012
water to form aldehydes. This reaction was found to be photochemically activated, and a possible
pathway involving radical intermediates has been proposed and studied experimentally and
computationally.
Accepted 20 November 2012
Ó 2012 Elsevier B.V. All rights reserved.
Keywords:
Transition metal radical
Iridium
Photochemical activation
Aldehyde
1. Introduction
under photochemical activation via homolytic IreCl bond fission
and radical abstraction. The details of our study into this reaction
The dimeric iridium complex [Cp*IrCl2]2 (1) reacts with terminal
alkynes (2) in the presence of water to form the alkyl carbonyl
complexes Cp*Ir(CO)(Cl)(CH2R) (3) via a C^C bond cleavage reac-
tion (Scheme 1). This reaction proceeds with a variety of aromatic
and aliphatic alkynes, and the possible reaction pathway has been
studied experimentally and computationally [1].
are presented.
2. Results and discussion
As mentioned above, small amounts of 4 and 5a were observed
to be formed during the synthesis of 3a. The same products were
obtained if purified 3a was allowed to stand in chloroform. The
identity of 4 was confirmed by an independent synthesis involving
stirring 1 under a CO atmosphere, as well as by a comparison of the
spectroscopic data with literature values [6], while the identity of
5a was confirmed by comparison of its NMR and GCeMS spectra
with an authentic sample.
When R contains a
b-hydrogen, the complexes 3 decompose via
b
-hydrogen elimination to give Cp*Ir(CO)(Cl)2 (4). Such a decom-
position pathway is well-known in transition metal alkyl chem-
istry, and it is equally well-known that if this pathway is shut off,
then the complexes are likely to be stable. Transition metal alkyl
complexes are also known to decompose via a-H elimination and
abstraction, but this mode of decomposition is generally not
common for late transition metal complexes. In the case of iridium
complexes, there are very few reports of a-H elimination and, in
general, they appear to occur for sterically encumbered iridium
centres [2e5].
We thus expected that those complexes not containing an R
group with a b-hydrogen, for instance, Cp*Ir(CO)(Cl)(CH2Ph) (3a),
should be stable. Instead, we invariably obtained small amounts of
Cp*Ir(CO)Cl2 (4) and benzaldehyde (5a) from its synthesis accord-
ing to Scheme 1. We have now found that 3 reacted with water
As the formation of 3a generates HCl, our initial conjecture was
that 4 resulted from the reaction of HCl with 3a. Indeed a mixture
containing 3a and 4 in roughly equal amounts was obtained when
3a was stirred with 1 M aqueous HCl in acetonitrile. Monitoring this
reaction by NMR over a period of 6 days showed the gradual
formation of 4 and 5a, and the subsequent decarbonylation of 4 to
regenerate 1 (Scheme 2). The amount of 5a formed was slightly less
than stoichiometric compared to 4. The rate and yield of the reac-
tion was not increased by heating, but UV irradiation afforded
significant amounts of 5a in 20 min, while visible light irradiation
also gave a better yield of 5a albeit over a longer reaction time
(w20 h to reach completion); no reaction occurred when light was
completely excluded. Both irradiation methods also gave rise to
small amounts of different unidentified side products. In contrast to
* Corresponding authors.
E-mail address: chmlwk@ntu.edu.sg (W.K. Leong).
0022-328X/$ e see front matter Ó 2012 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.jorganchem.2012.11.030

276
J.Y. Lee et al. / Journal of Organometallic Chemistry 724 (2013) 275e280
Scheme 1.
this, an attempted synthesis of the C^C bond cleavage product
from the reaction of the bromo analogue of 1, viz., [Cp*IrBr2]2 (10),
with 2a under identical conditions to that above gave Cp*Ir(-
CO)(Br)2 (40) as the major product. Resonances which could be
attributed to Cp*Ir(CO)(Br)(CH2Ph) (3a0) and 5a were also observed,
and there was no significant improvement to the yield of 3a0 when
the reaction was carried out in the dark.
more facile than that of 3a to 4, and may be attributed to bromide
being a better leaving group; presumably, HBr liberated from the
formation of 3a0 is responsible for the formation of 40 in this case.
Visible light irradiation of a mixture of 3a and aqueous [Et4N]Cl
in acetonitrile showed the formation of 5a but no 4. The NMR
spectrum also showed a major species containing a Cp* resonance,
and the presence of acetamide. The formation of acetamide was
mostly likely to be due to the partial hydrolysis of acetonitrile. ESI-
MS analysis suggested that the iridium species also contained
acetamide ligands but we have otherwise not been able to confirm
its identity. A similar result was obtained with aqueous [Et4N]Br,
and when only water was used. These indicated that although an
acidic environment is not required for the formation of 5a, it is
required for the formation of 4.
An isotopic labelling experiment in which 3a which has been
deuterated at the CH2 protons (d2-3a), afforded 5a which was
deuterated at the aldehyde moiety. This indicated that the CH2 was
transformed into the aldehyde functionality. Similarly, the GCeMS
of a reaction using 3d in 10 M HCl which has been diluted with
H218O, showed peaks at m/z 186 and 188 in the mass spectrum,
indicating the presence of 18O-5d, and thus showing that the source
of the oxygen atom in the aldehyde was water.
On the basis of the above observations, we would like to propose
a reaction pathway for the reaction from 3a to 4 and 5a as shown
(Scheme 4). We have investigated this pathway computationally
using density functional theory at the B3LYP level using the
LANL2DZ basis set augmented with f polarization functions for the
Ir atom and the 6-311 þ G(2d,p) basis set for all other atoms; the
reaction free energy changes, with acetonitrile as the reaction
solvent, are also shown in Scheme 4.
The first step is photochemical homolytic fission of the IreCl
bond in 3a to give the radical intermediate A. The alternative
involving loss of the chloride ligand was excluded because per-
forming the reaction in the presence of AgBF4 in the absence of light
This reaction to form an aldehyde was tested out with a number
of other alkynes; the reaction of 1 with various alkynes to generate
the corresponding C^C bond cleavage product 3, followed by their
subsequent reaction to form 4 and 5 under the optimized condi-
tions, are given in Scheme 3. With the exception of 2-
ethynylbenzaldehyde, 2g, all of the aromatic alkynes contained
electron donating substituents. Conversion of complexes 3aed, and
3f, to their respective aldehydes was close to stoichiometric. For 3e,
significantly more 1 was present in the crude NMR, which affected
proper integration of the CHO resonance of 5e to the Cp* resonance
of 4. No analogous C^C bond cleavage product was observed for 2g,
but instead the aldehyde 5g was directly obtained from the reaction
of 1 with 2g. This reaction did not require visible light irradiation,
and irradiating a mixture of 1 and 2g gave some unidentified side
products. Both 2h and 2i failed to react to form 3h and 3i, respec-
tively, probably due to interference from the amino and hydroxyl
groups; the former, for instance, coordinates strongly to mono-
meric Cp*IrCl2 units [7]. In contrast, 3j showed no tendency
towards degradation to the aldehyde 5j.
Since 3a generated 4 upon standing in chloroform, it must
indicate that chloroform can act as a chloride source. This was
confirmed by a reaction of 3a with bromoform; the 1H NMR spec-
trum showed two Cp* resonances, one corresponding to 40 and
another which could be intrapolated to be that for Cp*Ir(CO)(Br)(Cl)
(400), in a 7:1 ratio. On the other hand, the NMR spectrum of
a reaction of 10 with 2a in chloroform showed only the presence of
3a0 and 40 but no 4. This indicates that the conversion of 3a0 to 40 is
Scheme 2.

J.Y. Lee et al. / Journal of Organometallic Chemistry 724 (2013) 275e280
277
R
Yield of 3 (%)
Integration ratio*
a
C6H5
70
40
80
72
50
58
-
1:19
1:19
1:20
1:19
1:5
b
c
p-FC6H4
p-ClC6H4
p-BrC6H4
p-CH3C6H4
p-CH3OC6H4
o-OHCC6H4
o-NH2C6H4
o-HOCH2C6H4
tBu
d
e
f
1:11
2:15
g**
h
i
-
-
j
27
*Integration ratio of the –CHO resonance of 5 to the Cp* signal of 4.
**5g obtained directly from the reaction of 1 with 2g; 3g not observed.
Scheme 3.
+ 257
+185
Ir
Ph
Ir
Ir
Ph
Cl
Ph
OC
OC
OC
Cl
HCl
H
B
3a
A
-44
H O
2
- 104
Ir
Ph
Ir
Ir
OC
OC
OC
Cl
H
H
Cl
H
PhCHO
5a
OH
4
4*
C
energies in kJ mol-1
Scheme 4.

278
J.Y. Lee et al. / Journal of Organometallic Chemistry 724 (2013) 275e280
resulted in the conversion of 3a to an unidentified compound with
no formation of 5a. On the other hand, irradiation of 3a in the
presence of TEMPO or acrylonitrile resulted in a mixture of
unidentified compounds which did not contain 4 or 5a, suggesting
the presence of radical intermediates. We have also verified that
TEMPO did not react with 3a in the absence of light.
form an organometallic radical species. a-H abstraction to form
a carbene and subsequent reaction of this with water led, eventu-
ally, to elimination of an aldehyde.
4. Experimental
The second step is hydrogen abstraction by the chlorine atom to
form a diradical which spin-pairs to produce the carbene inter-
4.1. General procedures
mediate B; the positive DG value associated with this step would be
All operations were carried out using standard Schlenk tech-
niques under an inert argon atmosphere unless otherwise stated.
Solvents used were of AR grade. Preparative thin layer chroma-
tography was done using Merck 60 F254 (20 cm ꢁ 20 cm) silica gel
plates. 1H NMR spectra were recorded on a JEOL ECA400 or
ECA400SL 400 MHz NMR spectrometer. Chemical shifts were re-
ported with respect to the residual solvent peak in the respective
deuterated solvent. Solution IR spectra were obtained as dichloro-
methane (DCM) solutions on a Bruker Optik GmbH ALPHA FTIR
spectrometer at a resolution of 2 cmꢀ1 using a solution IR cell with
NaCl windows and pathlength of 0.1 mm. Mass spectra were ob-
tained on a Finnigan MAT95XL-T spectrometer in a 3NBA matrix
(FAB). GCeMS analyses were performed on a ThermoFinnigan DSQ
II instrument, equipped with Trace gas chromatograph and DSQ II
(EI) mass spectrometer, using a DB e5MScolumn. All UV and
tungsten lamp irradiation experiments were carried out using a Ace
Glass medium-pressure, Hanovia, mercury e vapor arc lamp
(approx. 254 nm, 450 Watts) and a Phillip N63 tungsten lamp
(200e400 nm, 60 Watts) respectively. UVevis absorbance data was
obtained using a Varian Cary 300 UVevis spectrophotometer.
Hydrogen gas was detected using a Balzer Prisma QMS 200 residual
mass analyzer. Complexes [Cp*IrCl2]2 (1), [Cp*IrBr2]2 (10) and
Cp*Ir(CO)(Cl)2 (4) were synthesized from their respective published
methods [10,11].
lowered considerably by hydration of the HCl formed. This subse-
quently undergoes nucleophilic attack by water at the carbene
carbon to eventually form the intermediate C. Although the
chemistry of iridium carbenes that are structurally similar to B have
been studied extensively [8], its expected behaviour with water
was uncertain. Thus although pathways involving alternatives to B
have been considered, that presented here appears to be the
simplest. There are different ways to obtain intermediate 4* from C,
but the most likely would be via a
for -H containing complexes of 3, thus generating 5a in the
process.
b-H elimination similar to that
b
The conversion of 4* to 4/40/400 in the presence of a Cl and/or Br
source is then expected to be very fast. It may be expected that this
step, or similar steps, would involve elimination of H2 and/or
formation of CH2Cl2 (from CHCl3) [9]. However, we have not been
able to detect either; analysis of the headspace gas of the reaction
for hydrogen using a residual gas analyzer did not generate any
conclusive results when compared to the reading from a blank
reaction, nor did an attempt at observing hydrogen gas or CH2Cl2
formation via NMR spectroscopy.
We have examined more closely the excited states of 3a. An
excited state computation with the TD method showed that exci-
tations into the first three excited states of 3a corresponded to
absorptions at 362, 302 and 261 nm. The solution (CH2Cl2) UVevis
spectrum of 3a showed two distinct absorption bands centred at
w370 and 320 nm (equivalent to w323 and 374 kJ molꢀ1, respec-
tively, in energy), in close accord with the first two computed
values. These values suggested that the irradiation could provide
sufficient energy for the endergonic step from 3a to A. Visualization
of the LUMO of 3a shows that it is anti-bonding with respect to the
IreCl bond (Fig. 1), suggesting that excitation into the LUMO would
weaken the IreCl bond.
4.2. Computational studies
The reaction energetics for the reaction of 3a to 4 and 5a was
studied by DFT theory utilising the Becke’s three parameter hybrid
function [12], and Lee-Yang-Parr’s gradient-corrected correlation
function (B3LYP) [13]. The LANL2DZ (Los Alamos Effective Core
Potential Double-z) basis set together with an f polarisation func-
tion [14], was employed for the Ir atom; the 6-311 þ G(2d,p) basis
set was used for all other atoms. Excited state calculations were
carried out with the TD method using the same basis sets [15]. For
calculations involving solvated species, the polarized continuum
model (PCM) was employed. Harmonic frequencies were calculated
at the optimized geometries to characterize stationary points as
equilibrium structures, with all real frequencies and to evaluate
zero-point energy (ZPE) correction. The optimised structure for B
still had one negative vibration frequency of about 11 cmꢀ1 which
3. Conclusion
Complexes of the general formula Cp*Ir(CO)(Cl)(CH2R) react
with water under photochemical activation to give the aldehydes
RCHO. A pathway for the reaction has been proposed on the basis of
experimental and computational studies, and it involves
photochemically-activated homolytic fission of the IreCl bond to
Fig. 1. Optimized structure (left) and LUMO (right) of 3a.

J.Y. Lee et al. / Journal of Organometallic Chemistry 724 (2013) 275e280
279
corresponded to a molecular torsion. All calculations were per-
formed using the Gaussian 09 suite of programs [16].
400 (1.97 ppm, intrapolated from the values for 4 and 40) in an
integration ratio of 7:1.
4.3. Reaction of 1 with 2a
4.5.4. Reaction of 3a with aqueous tetraethylammonium chloride
under visible light irradiation
The reaction was modified from a reported procedure [1]. To 1
In a Wheaton vial, Et4NCl (4 mg, 0.02 mmol) and deionized
water (1 drop) were added to a solution of 3a (5 mg, 0.01 mmol) in
ACN (2 mL). The reaction mixture was irradiated with a tungsten
lamp and stirred at room temperature for 1 d. The colour of the
solution changed from yellow to green. After removal of all volatiles
in vacuo, a 1H NMR spectrum taken in CDCl3 showed a mixture with
an unidentified, major Cp* resonance at 1.61 ppm. Among the
mixture, acetamide and 5a were identified.
(25 mg, 0.03 mmol) and 2a (10
mL, 0.09 mmol) in DCM (1 mL), water
(25 L) was added. The orange coloured mixture was stirred at
m
room temperature for 1 d, after which all volatiles were removed in
vacuo. The residue was redissolved in minimal DCM and subjected
to preparative thin layer chromatography (TLC) using hexane/DCM
(1:1, v/v) as the eluent. Recovery of the major yellow band yielded
3a as a yellow solid.
The reactions of 1a with 2be2j, and 10 with 2a and 2j, were
performed using this same procedure.
Acetamide: 1H NMR (
NH), 5.74 (br, s, 1H, NH).
5a: 1H NMR (
d
, CHCl3) ¼ 2.01 (s, 3H, CH3), 5.37 (br, s, 1H,
d
, CHCl3) ¼ 7.54 (t, 2H, aromatic), 7.64 (t, 1H,
4.4. Synthesis of Cp*Ir(CO)(Br)2 (40)
aromatic), 7.88 (d, 2H, aromatic), 10.02 (s, 1H, CHO).
ESI-MS analysis showed a major set of signals at m/z ¼ w750 in
the positive region and MSeMS of the peak at 750 showed two
fragmentations to 714 (eCl) and 691 (eCH3CONH2).
The reaction was modified from the reported procedure for the
synthesis of 4 [11]. A solution of 10 (15 mg, 0.021 mmol) in DCM
(2 ml) was subjected to three cycles of freeze pump thaw in a carius
tube. Carbon monoxide (1 atm) was then introduced into the tube.
The orange solution turned yellow instantly and was left to stir for
a further 2 h.
4.5.5. NMR monitoring of the reaction of 3a in the presence of
TEMPO
In a NMR tube, TEMPO (5 mg, 0.03 mmol) was added to a solu-
tion of 3a (15 mg, 0.03 mmol) in d2-DCM (0.5 mL). The solution was
irradiated with a tungsten lamp at room temperature for 1 d. The
colour of the solution changed from yellow to orange. A comparison
of the 1H NMR spectrum collected at 1 d with that collected at 0 d
showed that 3a had completely reacted to form an unknown
mixture. A similar observation was made with the reaction was
1H NMR (
d
, CDCl3) ¼ 2.02 (s, 15H, Cp*). IR (DCM, cmꢀ1
)
nCO ¼ 2055 cmꢀ1 (vs).
4.5. Synthesis of Cp*Ir(CO)(Cl)(CD2Ph) (d2-3a)
To 1 (25 mg, 0.03 mmol) and 2a (10
mL, 0.09 mmol) in dried 1,2-
dichloroethane (1 mL) was added D2O (25
m
L). The orange coloured
performed using acrylonitrile (2
0.02 mmol) in d2-DCM (0.5 mL).
Repeating this reaction with the addition of deionized water (1
drop) did not yield any 5a. Repeating this reaction using an
amberized NMR tube showed no reaction.
mL, 0.03 mmol) and 3a (10 mg,
mixture was stirred at room temperature for 1 d, after which all
volatiles were removed in vacuo. The residue was redissolved in the
minimum volume of DCM and subjected to preparative thin layer
chromatography (TLC) using hexane/DCM (1:1, v/v) as the eluent
Recovery of the major yellow band yielded Cp*Ir(CO)(Cl)(CD2Ph)
(d2-3a) as a yellow solid.
4.5.6. NMR monitoring of the reaction of 3a in the presence of AgBF4
In an amberized NMR tube, AgBF4 (4 mg, 0.02 mmol) and
deionized water (1 drop) were added to a solution of 3a (10 mg,
0.02 mmol) in d2-DCM (0.5 mL). The solution was left to stand at
room temperature for 1 d. The 1H NMR spectrum collected at 0 d
showed that 3a had reacted instantaneously to form an unknown
compound, and no further reaction occurred after 1 d. No 5a was
formed.
1H NMR (
aromatic), 7.13e7.26 (m, 4H, aromatic).
d
, CDCl3) ¼ 1.79 (s, 15H, Cp*), 6.94e6.99 (m, 1H,
4.5.1. Reaction of 3d in d3-ACN with H218O under visible light
irradiation
InanNMRtube,aqueousHCl(dilutedwithH218O)wasaddedto3d
(10 mg, 0.02 mmol) in d3eACN (0.75 mL). The reaction mixture was
irradiated with a tungsten lamp at room temperature for 16 h. Anal-
ysis of the reaction mixture showed the presence of 4 and 18O-5d.
Acknowledgement
ATReIR (DCM, cmꢀ1
18O-5d: 1H NMR (
CHO). GCMS: RT (min) ¼ 6.26; m/z ¼ 186 [M].
)
nCO ¼ 1651 cmꢀ1 (vs).
d
, d3-ACN) ¼ 7.74 (m, 4H, aromatic), 9.95 (s,1H,
This work was supported by Nanyang Technological University
and the Ministry of Education (Research Grant No. T208B1111) and
one of us (K.H.G.M.) thanks the National University of Singapore for
a Research Scholarship.
4.5.2. Reaction of 3a under various irradiation conditions
To each of two NMR tubes were placed 1 M aqueous HCl (1 drop),
3a (10 mg, 0.02 mmol), and d3-ACN (0.5 mL). They were then irra-
diated at room temperature with UV and tungsten lamps, respec-
tively, for1d. Analysisofbothreactionmixturesshowedthepresence
of complexes 4, 5a and some unidentified organic products.
Complexes 3bef and 3j were irradiated with a tungsten lamp
using this same procedure. Complex 3j showed no reaction.
Appendix A. Supplementary data
Supplementary data related to this article can be found at http://
dx.doi.org/10.1016/j.jorganchem.2012.11.030.
References
4.5.3. Reaction of 3a with bromoform
[1] V.S. Sridevi, W.Y. Fan, W.K. Leong, Organometallics 26 (2007) 1173e1177.
[2] X. Li, L.N. Appelhans, J.W. Faller, R.H. Crabtree, Organometallics 23 (2004)
3378e3387.
[3] E. Clot, J. Chen, D.H. Lee, S.Y. Sung, L.N. Appelhans, J.W. Faller, R.H. Crabtree,
O. Eisenstein, J. Am. Chem. Soc. 126 (2004) 8795e8804.
Complex 3a (20 mg, 0.04 mmol) was stirred with bromoform
(1 ml) in a carius tube for 4 d. The yellow solution turned orange
after 1 d, and no further colour change was observed over the
course of the reaction. After removal of all volatiles in vacuo, a 1H
NMR spectrum in CDCl3 showed the presence of 40 (2.01 ppm) and
[4] M.J. Burk, M.P. McGrath, R.H. Crabtree, J. Am. Chem. Soc. 110 (1988) 620.
[5] E. Carmona, M. Panaque, M.L. Poveda, Dalton Trans. (2003) 4022e4029.

280
J.Y. Lee et al. / Journal of Organometallic Chemistry 724 (2013) 275e280
[6] P.M. Maitlis, J.W. Kang, J. Organomet. Chem. 26 (1971) 393e399.
[7] D.L. Davies, O. Al-Duaij, J. Fawcett, M. Giardello, S.T. Hilton, D.R. Russell, Dalton
Trans. (2003) 4132e4138.
[16] M.J. Frisch, G.W. Trucks, H.B. Schlegel, G.E. Scuseria, M.A. Robb,
J.R. Cheeseman, J.A. Montgomery Jr., T. Vreven, K.N. Kudin, J.C. Burant,
J.M. Millam, S.S. Iyengar, J. Tomasi, V. Barone, B. Mennucci, M. Cossi,
G. Scalmani, N. Rega, G.A. Petersson, H. Nakatsuji, M. Hada, M. Ehara,
K. Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao,
H. Nakai, M. Klene, X. Li, J.E. Knox, H.P. Hratchian, J.B. Cross, C. Adamo,
J. Jaramillo, R. Gomperts, R.E. Stratmann, O. Yazyev, A.J. Austin, R. Cammi,
C. Pomelli, J.W. Ochterski, P.Y. Ayala, K. Morokuma, G.A. Voth, P. Salvador,
J.J. Dannenberg, V.G. Zakrzewski, S. Dapprich, A.D. Daniels, M.C. Strain,
O. Farkas, D.K. Malick, A.D. Rabuck, K. Raghavachari, J.B. Foresman, J.V. Ortiz,
Q. Cui, A.G. Baboul, S. Clifford, J. Cioslowski, B.B. Stefanov, G. Liu, A. Liashenko,
P. Piskorz, I. Komaromi, R.L. Martin, D.J. Fox, T. Keith, M.A. Al-Laham, C.Y. Peng,
A. Nanayakkara, M. Challacombe, P.M.W. Gill, B. Johnson, W. Chen,
M.W. Wong, C. Gonzalez, J.A. Pople, Gaussian 03, Gaussian Inc., Wallingford
CT, 2004.
[8] H. Werner, Organometallics 24 (2005) 1036e1049.
[9] J. Campos, R. Peloso, E. Camona, Angew. Chem. Int. Ed. 51 (2012) 8255e8258.
[10] R.G. Ball, W.A. Graham, D.M. Heinekey, J.M. Hoyano, A.D. McMaster,
B.M. Mattson, S.T. Michel, Inorg. Chem. 29 (1990) 2023.
[11] D.S. Gill, P.M. Maitlis, J. Organomet. Chem. 87 (1975) 359e364.
[12] A.D. Becke, J. Chem. Phys. 98 (1993) 568e5652.
[13] C. Lee, W. Yang, R.G. Parr, Phys. Rev. B 37 (1998) 785e789.
[14] A.W. Ehlers, M. Böhme, S. Dapprich, A. Gobbi, A. Höllwarth, V. Jonas,
K.F. Köhler, R. Stegmann, A. Veldkamp, G. Frenking, Chem. Phys. Lett. 208
(1993) 111e114.
[15] J.B. Foresman, M. Head-Gordon, J.A. Pople, M.J. Frisch, J. Phys. Chem. 109
(1992) 8218e8224.