Reactions of a tungsten alkylidyne complex with mono-dentate phosphines: Thermodynamic and theoretical studies

Article abstract of DOI:10.1016/j.poly.2012.07.042

Addition of mono-dentate phosphines PMe₃ and PMe₂Ph to the W(VI) alkyl alkylidyne complex W(CH₂SiMe₃) ₃(≡CSiMe₃) (1) is reversible, each reaching equilibrium. Thermodynamic studies of the equilibria have been conducted, giving ΔH° = -10.0(1.1) kcal/mol and ΔS° = -23(4) eu for the addition of PMe₃ and ΔH°′ = -3.0 (0.7) kcal mol ^-1 and ΔS°′ = -6(3) eu for the addition of PMe ₂Ph, indicating that the addition is exothermic. The experimental measurement allows a benchmarking study to select a proper DFT method to describe the current system. Of the DFT methods tested, M06 has demonstrated superior performance in calculating binding energy of a bimolecular reaction. The calculated reaction pathways show that W(CH₂SiMe ₃)₃(≡CSiMe₃) (1) reacts with PR3 to form W(CH₂SiMe₃)₃(≡CSiMe₃)(PR ₃) (PR₃ = PMe₃, 3a; PMe₂Ph, 3b), and the adduct then undergoes α-H migration to form W(CH₂SiMe ₃)₂(=CHSiMe₃)₂(PR₃) (4a, 4b). 4a and 4b are found to be thermodynamically and kinetically stable intermediates. The calculations also suggest a pathway in the formation of the alkyl alkylidene alkylidyne complex W(CH₂SiMe₃)- (=CHSiMe₃)(≡CSiMe₃)(PR₃)₂ (5a).

Full text of DOI:10.1016/j.poly.2012.07.042

Polyhedron 58 (2013) 30–38
Contents lists available at SciVerse ScienceDirect
Polyhedron
journal homepage: www.elsevier.com/locate/poly
Reactions of a tungsten alkylidyne complex with mono-dentate phosphines:
Thermodynamic and theoretical studies
a,
b,
Ping Chen a, Brenda A. Dougan b, Xinhao Zhang a, , Yun-Dong Wu , Zi-Ling Xue
⇑
⇑
⇑
a Lab of Computational Chemistry and Drug Design, Laboratory of Chemical Genomics, Peking University Shenzhen Graduate School, Shenzhen 518055, PR China
b Department of Chemistry, University of Tennessee, Knoxville, TN 37996, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
Available online 23 July 2012
Addition of mono-dentate phosphines PMe3 and PMe2Ph to the W(VI) alkyl alkylidyne complex
W(CH2SiMe3)3(„CSiMe3) (1) is reversible, each reaching equilibrium. Thermodynamic studies of the equi-
libria have been conducted, giving
D
H° = ꢀ10.0(1.1) kcal/mol and S° = ꢀ23(4) eu for the addition of PMe3
D
Dedicated to the memory of Michelle Millar,
a scientist and teacher who gave so much to
the chemistry communities.
and
D
H°0 = ꢀ3.0 (0.7) kcal molꢀ1 and
D
S°0 = ꢀ6(3) eu for the addition of PMe2Ph, indicating that the addi-
tion is exothermic. The experimental measurement allows a benchmarking study to select a proper DFT
method to describe the current system. Of the DFT methods tested, M06 has demonstrated superior per-
formance in calculating binding energy of a bimolecular reaction. The calculated reaction pathways show
that W(CH2SiMe3)3(„CSiMe3) (1) reacts with PR3 to form W(CH2SiMe3)3(„CSiMe3)(PR3) (PR3 = PMe3, 3a;
Keywords:
Alkylidyne
PMe2Ph, 3b), and the adduct then undergoes a-H migration to form W(CH2SiMe3)2(@CHSiMe3)2(PR3) (4a,
Alkylidene
Phosphine adduct
Equilibrium
DFT calculations
4b). 4a and 4b are found to be thermodynamically and kinetically stable intermediates. The calculations
also suggest a pathway in the formation of the alkyl alkylidene alkylidyne complex W(CH2SiMe3)-
(@CHSiMe3)(„CSiMe3)(PR3)2 (5a).
a-H abstraction
Ó 2012 Elsevier Ltd. All rights reserved.
1. Introduction
(@CHSiMe3)2(DMPE-P) (4e). The tautomeric mixtures are in
equilibria, and undergo a-H abstraction to eliminate SiMe4 to form,
High-oxidation-state metal alkylidene and alkylidyne complexes
containing M@C and M„C bonds, respectively, often demonstrate
unique chemistry and have been extensively studied [1–4]. The
alkylidene and alkylidyne complexes usually contain ligands such
as neopentyl (–CH2CMe3) and trimethylsilylmethyl ligands
(–CH2SiMe3) that are free of b-H atoms [1–4]. Complexes containing
a –CH2SiMe3 ligand with a b-Si atom often show properties different
from those containing its neopentyl –CH2CMe3 analog. For example,
Clark and Schrock reported that the reactions of W(CH2CMe3)3-
(„CCMe3) with phosphines PMe3 and Me2PCH2CH2PMe2 (DMPE)
give the bis-phosphine/chelating phosphine products
W(CH2CMe3)(@CHCMe3)(„CCMe3)(PMe3)2 and W(CH2CMe3)-
by reacting with free PR3 and the free PMe2 group in DMPE-P
ligand, W(CH2SiMe3)(@CHSiMe3)(„CSiMe3)(PR3)2 (5a-b, 6a-b)
(Scheme 1) and W(CH2SiMe3)(@CHSiMe3)(„CSiMe3)(DMPE) (5e,
6e) (Scheme 2), respectively [6].
Another example of the difference between –CH2SiMe3 and
–CH2CMe3 ligands is the reactions shown in Scheme 3 [7].
Ta(@CHR)2(CH2SiMe3)(PMe3)2 (R = SiMe3, CMe3) reacts with
silanes to give metallocyclic products. In comparison, its neopentyl
analog, Ta(@CHCMe3)2(CH2CMe3)(PMe3)2, forms unknown species
with the silanes.
We have been interested in the tautomerizations between alkyl
alkylidynes (3a, 3b, 3e) and bis-alkylidenes (4a, 4b, 4e) in Schemes
1 and 2 because these were among the few reported direct obser-
(@CHCMe3)(„CCMe3)(DMPE) through
a-H abstraction [3b]. The
structure of the latter was determined by Churchill and Youngs
[5]. In comparison, we have found that W(CH2SiMe3)3(„CSiMe3)
(1) reacts with PMe3, PMe2Ph, and DMPE, forming phosphine ad-
ducts W(CH2SiMe3)3(„CSiMe3)(PR3) (PR3 = PMe3, 3a; PMe2Ph, 3b)
(Scheme 1) and W(CH2SiMe3)3(„CSiMe3)(DMPE-P) (3e) (Scheme 2)
containing one free PMe2 group [6]. These adducts undergo
vation of exchanges through a-H migration [1j,m], although such
exchanges had been implicated in several systems [8–12]. The tau-
tomerization of silyl alkylidyne (Me3CCH2)2W(„CCMe3)(SiButPh2)
with bis-alkylidene (Me3CCH2)W(@CHCMe3)2(SiButPh2) was, to
our knowledge, the first directly observed exchange [13].
In the current work, we studied the thermodynamics of the
reversible addition of PMe3 and PMe2Ph to 1 forming 3a and 3b,
respectively, in order to probe the unusual chemistry of 1 in detail.
Density functional theory (DFT) studies have been performed to:
(1) understand the thermodynamics and kinetics of the transfor-
mation from 1 to 4; (2) explain why 4 is a relatively stable interme-
diate along the pathway to 5.
a-H migration to form their bis-alkylidene tautomers
W(CH2SiMe3)2(@CHSiMe3)2(PR3) (4a, 4b) and W(CH2SiMe3)2-
⇑
Corresponding authors.
E-mail addresses: zhangxh@pkusz.edu.cn (X. Zhang), chydwu@ust.hk (Y.-D. Wu),
xue@ion.chem.utk.edu (Z.-L. Xue).
0277-5387/$ - see front matter Ó 2012 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.poly.2012.07.042

P. Chen et al. / Polyhedron 58 (2013) 30–38
31
SiMe3
SiMe3
W SiMe3
Keq
kf
SiMe3
Me3Si
Me3Si
Me
Me
3 Si
W
W
PR3
+
Me3Si
Me3Si
SiMe3
3 Si
SiMe3
kr
PR3
PR3
1
3a 3b
,
4a 4b
,
heat
PR
k
+
3
-SiMe4
3a, 4a, 5a 6a
PR3 =PMe3,
PMe2Ph,
,
3b, 4b 5b 6b
,
,
PR3
W
PR3
W
SiMe3
SiMe3
H
SiMe3
H
Me3Si
Me3Si
+
SiMe3
PR3
PR3
5a 5b
,
syn -
anti-6a,6b
Scheme 1. Reactions of mono-dentate phosphines with 1 [6a,b,d].
Keq
SiMe3
SiMe3
SiMe3
Me3Si
kf'
Me3Si
Me3Si
W
W
W
DMPE
SiMe3
PMe2
+
Me3Si
Me3Si
kr'
SiMe3
PMe2
SiMe3
Me3Si
Me2P
Me2P
1
3e
4e
heat
- SiMe4
k'
SiMe3
SiMe3
W
Me2
P
Me2
P
SiMe3
SiMe3
SiMe3
H
+
W
H
Me2P
PMe2
DMPE =
P
P
Me2
Me2
SiMe3
5e
syn-
6e
anti-
Scheme 2. Reaction of the bidentate phosphine DMPE with 1 [6c].
potassium/benzophenone ketyl. Toluene-d8 was dried over acti-
vated molecular sieves and stored under N2. WCl6 was freshly sub-
limed under vacuum. W(CH2SiMe3)3(„CSiMe3) (1) [3d] was
prepared from (MeO)3WCl3 and six equivalents of Me3SiCH2MgCl
by
a procedure similar to that used in the preparation of
W(CH2CMe3)3(„CCMe3) [14]. 1H NMR spectra were recorded on a
Bruker AMX-400 spectrometer.
For the thermodynamic studies, the equilibrium constants Keq1
were obtained from at least five separate experiments at a given
temperature, and their averages are listed in Table 1. The maxi-
mum random uncertainty in the equilibrium constants were com-
bined with the estimated systematic uncertainty of ca. 5%. The
total uncertainties in the equilibrium constants were used in the
ln Keq1 versus 1000/T plot in Fig. 1 and the error propagation calcu-
lations. The estimated uncertainty in the temperature measure-
ments for an NMR probe was 1 K. The enthalpy (
entropy ( S°) changes were calculated from an unweighted non-
linear least-squares procedure. The uncertainties in H° and S°
were computed from the following error propagation formulas,
S° [13].
DH°) and
D
Scheme 3. Reactions of PhR0SiH2 (R0 = Me, Ph, CH2SiH2Ph) with phosphine alkyl-
idene complexes.
D
D
which were derived from ꢀRT ln Keq1
=
D
H° ꢀ TD
ꢀ
ꢁ
R2 T2maxT4min þ Tm2 inT4max
ꢄ
ꢂ
ꢃꢅ ꢂ
ꢃ
2
2
2. Material, methods and computational details
Keq1ðmaxÞ
Keq1ðminÞ
r
T
T
2
ðr
D
HꢁÞ ¼
ln
4
ðTmax ꢀ Tmin
Þ
r
All manipulations were performed under
a dry nitrogen
ꢂ
ꢃ
2
2R2T2maxT2min
Keq1
atmosphere with the use of either a dry box or standard Schlenk
techniques. Solvents were purified by distillation from
þ
ð1Þ
2
Keq1
ðTmax ꢀ Tmin
Þ

32
P. Chen et al. / Polyhedron 58 (2013) 30–38
Table 1
Equilibrium constants (Keq1 and K0 ) of 1 + PR3 ꢀ 3a/4a and 3b/4ba.
eq1
c
d
T (K)b
Keq1 for 1 + PMe3 ꢀ 3a/4a
K0eq1 for 1 + PMe2Ph ꢀ 3b/4b
263(1)
273(1)
278(1)
283(1)
288(1)
293(1)
298(1)
303(1)
16.4(1.5)
12.0(1.7)
10.8(0.4) ꢂ 102
7.1(0.8) ꢂ 102
5.3(0.3) ꢂ 102
3.6 (0.3) ꢂ 102
2.83(0.08) ꢂ 102
2.23(0.03) ꢂ 102
1.66(0.07) ꢂ 102
10.3(0.4)
8.8(0.6)
7.5(0.5)
a
Solvent: toluene-d8
b
The relatively small temperature ranges of 30–40 K for the exchanges, 1 + PMe3 ꢀ 3a/4a and 1 + PMe2Ph ꢀ 3b/4b, respectively, lead to relatively large uncertainties in
H° and S°) as the error calculations in the experimental section show.
For the equilibrium involving PMe3 and 3a/4a, the largest random uncertainty is Keq1(ran)/Keq1 = 0.84/7.1 = 12%. The total uncertainty
from Keq1(ran)/Keq1 = 12% and the estimated systematic uncertainty Keq1(sys)/Keq1 = 5% by Keq1/Keq1 = [( Keq1(ran)/Keq1
2 + ( Keq1(sys)/Keq1)2]1/2
K0eq1(ran)/K0eq1 = 1.7/12 = 14%. The total uncertainty
K0eq1(sys)/K0eq1 = 5% by Ke0 q1/Ke0 q1 = [( Ke0 q1(ran)/Ke0 q1 2 + (
thermodynamic (
D
D
c
r
rKeq1/Keq1 of 13% was calculated
r
r
r
r
)
r
.
d
For the equilibrium involving PMe2Ph and 3b/4b, the largest random uncertainty is
r
r
r
K0eq1/K0eq1 of 15% was calculated
from
Ke0 q1(ran)/K0eq1 = 14% and the estimated systematic uncertainty
r
r
r
)
r .
K0eq1(sys)/Ke0 q1)2]1/2
3.00
2.75
2.50
2.25
2.00
1.75
7.0
6.5
6.0
K
K
5.5
5.0
4.5
3.3
3.4
3.5
3.6
3.7
3.2
3.3
3.4
3.5
3.6
3.7
3.8
3.9
1000/T
1000/T
Fig. 1. Plots of ln Keq1 (and K0eq1) vs 1000/T of the equilibria: (Left) 1 + PMe3 ꢀ 3a/4a; (Right) 1 + PMe2Ph ꢀ 3b/4b.
Me3Si
Me3Si
SiMe3
H
W
SiMe3
W
Me3Si
Me3Si
SiMe3
H
Me3Si
- PR3
1
2
+
P
R
R
3
Ke
q
1
a
-
n
- PR3
Me3Si
+ PR3
Fast
+ PR3
Fast
P
d
Ke
3
Δ
q
G
1
'
,
1
Δ
H1
,
Δ
S
Me3Si
1
SiMe3
H
Me3Si
Me3Si
+ PR3 Keq2
W
SiMe3
W
- PR3 G , H ,
S
Δ
2
Δ
Δ
Me3Si
SiMe3
2
2
H
PR3
3a
PR3
4a, 4b
PR3 = PMe3,
PMe2Ph,
;
-SiMe4
+ PR3
3b
Heat
PR3
SiMe3
SiMe3
W
Me3Si
H
PR3
5a/6a, 5b/6b
Scheme 4. Equilibria from the reactions of 1 with PR3.

P. Chen et al. / Polyhedron 58 (2013) 30–38
33
Table 2
Relative enthalpies (kcal molꢀ1) involved in the equilibrium 1 + PMe3 ꢀ 4a.
Methods
1
2
3a
5.6
4a (
D
H1)
D
H2 = Hrel(4a) ꢀ Hrel(3a)
Gas-phase
B3LYP
M06
BP86
0.0
0.0
0.0
7.4
9.2
5.8
2.7
ꢀ12.1
ꢀ2.6
ꢀ2.9
ꢀ2.7
ꢀ5.2
ꢀ9.4
2.6
Toluene
B3LYP
M06
BP86
0.0
0.0
0.0
8.9
7.3
9.5
9.7
ꢀ5.8
6.6
6.7
ꢀ8.8
2.1
ꢀ3.0
ꢀ3.0
ꢀ4.5
Exp.
ꢀ10.0
ꢀ1.8
Table 3
Entropy (eu) involved in the equilibrium 1 + PMe3 ꢀ 4a.
Entropy
1
PMe3
3a
4a
D
S1 = S(4a) ꢀ [S(1) + S(PMe3)]
D
S2 = S(4a) ꢀ S(3a)
a
Gas-phase
STotal
STrans.
SRot.
241.1
44.7
35.9
77.8
38.9
25.4
13.5
267.2
45.1
36.4
262.8
45.1
36.4
ꢀ56.1
ꢀ4.4
SVib.
160.5
185.7
181.2
a
Toluene
Stotal
227.2
44.7
35.7
146.7
199.0
77.2
38.9
25.3
13.0
54.7
258.7
45.1
36.5
177.2
230.2
259.4
45.1
36.4
177.9
230.9
ꢀ45.0
0.7
Strans.
Srot.
Svib.
Scorr.
b
ꢀ22.8
0.7
Exp.
ꢀ23(4)
ꢀ1.5(1.7)
a
Stotal = Strans. + Srot + Svib.
Scorr. = 0.65 ꢂ (Strans. + Srot.) + Svib.
.
b
.
ꢄ
ꢂ
ꢂ
ꢃꢅ ꢂ
ꢃ
2
2
2R2T2minT2max
rT
T
Keq1ðmaxÞ
Keq1ðminÞ
2
SꢁÞ ¼
ln
Table 4
ðr
D
Reaction enthalpies (kcal molꢀ1), entropies (eu) and Gibbs free energies (kcal molꢀ1
)
4
ðTmax ꢀ Tmin
Þ
for the reaction 1 + PR3 ? 4 and WAP distance of 4 (in Å).
ꢃ
2
R2ðT2max þ T2min
Þ
r
Keq1
Keq1
a
b
D
H1
D
S1-corr
D
G1-corr
WAP
þ
ð2Þ
2
ðTmax ꢀ Tmin
Þ
PMe3 (4a)
PMe2Ph (4b)
PPh3 (4c)
ꢀ8.8/ꢀ10.0
ꢀ8.9/ꢀ3.0
ꢀ7.4/ꢀ
ꢀ22.8/ꢀ23
ꢀ28.6/ꢀ6
ꢀ29.7/ꢀ
ꢀ2.0/ꢀ3.1
ꢀ0.4/ꢀ1.2
1.5/–
2.597/2.514
2.609/–
2.730/–
Tmin and Tmax are the minimum and maximum temperatures in the
current studies; T is the mean temperature in the current studies.
Keq1(min) and Keq1(max) are the minimum and maximum equilibrium
PCy3 (4d)
ꢀ1.3/–
ꢀ29.4/–
7.5/–
2.861/–
a
Scorr = 0.65 ꢂ (Strans. + Srot.) + Svib.
.
constants, respectively.
r
Keq1/Keq1 is given in Table 1. Similar cal-
b
D
G1-Corr
=
D
H1 ꢀ 298.15 ꢂ
DS1-corr..
0
culations were performed for Keq1
.
Me3Si
SiMe3
Me3Si
H
W
Me3Si
Me3Si
Me3Si
SiMe3
W
SiMe3
+
H
PMe3
+
PMe3
3a-frag
4a-frag
25.4
ΔEdeform = 13.7
20.3
Me3Si
ΔEdeform = 20.3
ΔEint = -28.4
SiMe3
11.7
ΔEint = -35.5
Me3Si
Me3Si
Me3Si
Me3Si
SiMe3
H
W
SiMe3
H
0.0
Me3Si
-8.1
W
W
-10.1
PMe3
Me3Si
SiMe3
SiMe3
W
SiMe3
H
H
Me3Si
Me3Si
Me3Si
PMe3
3a
2
+ PMe3
1
+ PMe3
4a
Fig. 2. Energy decomposition analysis of the binding complex 3a and 4a (energies in kcal molꢀ1).

34
P. Chen et al. / Polyhedron 58 (2013) 30–38
2.1. Thermodynamic study of the equilibrium between
W(CH2SiMe3)3(„CSiMe3) (1), PMe3 and the 3a ꢀ 4a tautomeric
mixture
4). The addition was fast and essentially completed by the time
NMR spectra were recorded. After the addition, the solution in tol-
uene-d8 was kept at room temperature for 48 h to ensure that
equilibrium was established. Variable-temperature NMR spectra
of the mixtures were studied, and the equilibrium constants,
Keq1 = [4a]/([1] [PMe3]), measured with a variable amount of
PMe3 between 273 and 303 K are listed in Table 1. The peaks of
3a were very small and ignored for the calculations of Keq. For
the 3a ꢀ 4a tautomerization, 4a dominates by a ratio of ca.
9.4–12.3 [6a,b]. A plot of ln Keq1 versus 1000/T (Fig. 1-Left) gave
At least five experiments were conducted.
A mixture of
W(CH2SiMe3)3(„CSiMe3) (1, 20.7 mg, 0.04 mmol), 4,40-dimethyl-
phenyl (internal standard), and toluene-d8 (ca. 0.65 mL) in a J.R. Youngs
NMR tube was added a ca. 0.7–1.8 equiv PMe3 via syringe. A weighed
amount of 4,40-dimethylphenyl was used, and its concentration in
the solution was typically in the 0.02–0.03 M range. The sample was
kept at room temperature for 48 h to ensure that equilibrium was
established. The NMR probe was pre-cooled or pre-heated to the set
temperature. After the NMR tube was inserted into the probe, a 1H
NMR spectrum was taken after the temperature was stabilized.
Keq1 = [4a]/([1] [PMe3]) were calculated from the integration of 1,
PMe3, and 4a. The Me peak in 4,40-dimethylphenyl, –CH2– peak in 1,
and the PMe3 peak in 4a were used in the integration and calculation
D
H° = ꢀ10.0 (1.1) kcal molꢀ1 and
D
S° = ꢀ23(4) eu, indicating that
the addition is exothermic. The large, negative entropy of the reac-
tion is not surprising, as in the addition, two molecules give one
adduct molecule. The equilibrium constants (Keq1) range from
10.8 ꢂ 102 at 273 K to 1.7 ꢂ 102 at 303 K, indicating that the
3a ꢀ 4a tautomers are strongly favored. Hence, PMe3 strongly
binds to 1 and its dissociation is disfavored. At higher tempera-
tures, the equilibrium shifts towards 1 + PMe3.
of Keq1
.
2.2. Thermodynamic study of the equilibrium between
W(CH2SiMe3)3(„CSiMe3) (1), PMe2Ph and the 3b ꢀ 4b tautomeric
mixture
3.2. Thermodynamic study of the addition of PMe2Ph to 1
PMe2Ph was utilized for comparison with PMe3. W(CH2SiMe3)3-
(„CSiMe3) (1) in toluene-d8 was added a various amount of ca.
1.0–1.5 equiv of PMe2Ph via syringe. The addition to 1 is much
slower than the one involving PMe3, and takes about 2–3 days at
room temperature to reach the equilibrium. The sample was kept
at room temperature for 5 days to ensure that equilibrium was
established.
At least three experiments were conducted. A mixture of
W(CH2SiMe3)3(„CSiMe3) (1, 29.1 mg, 0.06 mmol), 4,40-dimethyl-
phenyl (internal standard), and toluene-d8 (ca. 0.53 mL) in a J.R.
Youngs NMR tube was added a ca. 1.0–1.5 equiv PMe2Ph via syr-
inge. A weighed amount of 4,40-dimethylphenyl was used, and its
concentration in the solution was typically in the 0.02–0.03 M
range. The sample was kept at room temperature for five days to en-
sure that equilibrium was established. The NMR probe was pre-
cooled or pre-heated to the set temperature. After the NMR tube
was inserted into the probe, the 1H NMR spectrum was taken after
the temperature was stabilized. After the sample was placed at the
set temperature for 30 min, no change in the NMR spectrum was
observed. The sample was, however, kept for two hours at each
temperature, during which NMR spectra were taken every 30 min
0
to ensure no further change was observed. Keq1 = [4b]/([1]
[PMe2Ph]) were calculated from the integration of 1, PMe2Ph, and
4b. The Me peak in 4,40-dimethylphenyl, –CH2– peak in 1, Me peak
in PMe2Ph, and the PMe2Ph peak in 4b were used in the integration
0
and calculation of Keq1
.
2.3. Computational methods
All calculations were carried out using Gaussian 09 [15]. The
performance of the density functional theory (DFT) methods
B3LYP [16,17], M06 [18], and BP86 [19] has been examined.
Triple-f basis sets 6-311G(d, p) [20–22] were used for H, C, Si, and
P; and def2-TZVP [23,24] with effective core potentials (ECP) [25]
were employed for tungsten. The basis set was found to be excellent
in describing metal–carbon and metal–oxygen bonds [26,27].
Geometry optimizations and vibrational frequency calculations
were performed both in gas phase and solution. Solvation effects
were treated with the continuum solvent models SMD [28]. Ther-
mal corrections are calculated at standard conditions (298.15 K
and 1 atm).
3. Results and discussion
3.1. Thermodynamic study of the addition of PMe3 to 1
Alkyl alkylidyne W(CH2SiMe3)3(„CSiMe3) (1) was treated with
a various amount of PMe3 (0.7–1.8 equiv) to probe the addition
reaction of phosphine and the position of the equilibrium (Scheme
Fig. 3. Energy profile along the WAP distance.

P. Chen et al. / Polyhedron 58 (2013) 30–38
35
Variable-temperature NMR spectra of the exchange of 1 and
PMe2Ph with 3b/4b were studied, and the equilibrium constants,
Keq1 = [4b]/([1] [PMe2Ph]), measured between 263 and 303 K with
Overestimation of the entropy for the process which involves
molarity change by gas-phase calculation is well-known [34–36].
This is due to the fact that translational and rotational degrees of
freedom in gas phase are reduced in solution. Several approxima-
tions have been applied to account for this effect in different sys-
tems [37–39]. As shown in Table 3, the calculated entropy loss
0
0
a variable amount of PMe2Ph are listed in Table 1. A plot of ln Keq1
versus 1000/T (Fig. 1-Right) gave
D
H°0 = ꢀ3.0 (0.7) kcal molꢀ1 and
D
S°0 = ꢀ6(3) eu. The equilibrium constants (Keq10) range from
16.4(1.5) at 263 K to 7.5(0.5) at 303 K. The much smaller equilibrium
constants, in comparison to those of addition of PMe3 to 1 (Table 1),
suggest that, although the 3b ꢀ 4b tautomer and the bis-alkylidene
(4b) are favored, the binding ofPMe2Ph to 1 is not as strongly favored
D
S1 (ꢀ56.1 eu) for the bimolecular process (1 + PMe3 ꢀ 4a) is sig-
nificantly higher than the experimental value (ꢀ23 eu). This prob-
lem cannot be addressed by performing full optimization and
frequency calculation in toluene with SMD. On the other extreme,
one can estimate the entropy by neglecting the translational and
rotational contribution but only taking the vibrational contribution
into account [40]. However, this will underestimate the entropy
simply because translation and rotation are in reality not com-
pletely suppressed. Therefore, one solution is to fit a scaling factor
between 1 and 0 for the rotational and translational entropy con-
tribution (Strans. + Srot.) to reproduce the experimental results. For
the reaction studied here, a factor of 0.65 was found to be suitable
for scaling the entropy in solution. The corrected entropy calcu-
lated by Eq. (3) (ꢀ22.8 eu) is closed to the experimental measured
value [ꢀ23(4) eu].
as the binding of PMe3. This is also reflected in smaller D
H°0 and thus
the release of much less energy during the addition of PMe2Ph to 1.
The reaction entropy is also less negative. This is presumably due
to the fact that PMe3 is a stronger
r-donor than PMe2Ph, and the
sterics involving more bulky phenyl group versus Me in PMe3 may
also be a factor. With decreasing temperature, the equilibrium shifts
0
towards 1. The 1 + PMe2Ph ꢀ 3b/4b equilibrium [Keq1 = 16.4(1.5)
at 273 K] is shifted more to the left (1 + PMe2Ph) than the
1 + PMe3 ꢀ 3a/4a equilibrium [Keq1 = 10.8(0.4) ꢂ 102 at 273 K].
3.3. Calculations – benchmarking
Scorr: ¼ 0:65 ꢂ ðStrans: þ Srot:Þ þ Svib:
ð3Þ
To evaluate the performance of computational methods, the
thermodynamics of two processes, the equilibrium 1 + PMe3 ꢀ 4a
and the equilibrium 3a ꢀ 4a, were calculated and compared with
the experimentally measured values.
The M06/(6-311G(d, p), def2-TZVP with ECP) level of theory
with SMD solvent corrections was chosen for the calculation in this
work while the entropy values in the following discussion were
corrected as outlined above.
Relative enthalpies calculated using a series of DFT methods are
listed in Table 2. All of the relative enthalpies (DH2) of the equilib-
rium 3a ꢀ 4a calculated at B3LYP, M06 and BP86 agree with the
experimental measured value reasonably well [6a]. However, nei-
ther B3LYP nor BP86 level of theory can reproduce the reaction en-
3.4. Formation of 4: theoretical studies of the thermodynamics
The calculated values for 1 + PMe3 ? 4a are consistent with the
measured values (Table 4). The equilibrium favors 4a. In the case of
PMe2Ph (4b), the free energy is close to 0 kcal molꢀ1, implying that
the concentrations of 1 and 4 are comparable in the equilibrium
thalpy of 1 and PMe3 (
various DFT methods has been reported recently [29–33]. The
D
H1). Underestimation of binding energy by
H1
D
calculated by M06 is consistent with the experimental result. The
experimental results (1 + PMe3 ꢀ 4a) identify M06 as an appro-
priate method for the system. Solvent effect does not affect the rel-
ative stabilities of 3a and 4a much, but decrease the binding
mixture. The positive free energy of binding
DG1 and the signifi-
cant longer WAP distances of 4c and 4d imply that such processes
are unfavorable, agreeing well with the experimental observation
that bulky PPh3 and PCy3 cannot form complexes 4.
enthalpies by ca. 4 kcal molꢀ1
.
Fig. 4. Potential free energy surface for different tautomerization pathways with geometries of the transition states 1TS2, 3TS4-a, 3TS4-b (distances in Å, angles in degree,
free energies in kcal molꢀ1).

36
P. Chen et al. / Polyhedron 58 (2013) 30–38
Fig. 5. Potential free energy surface for the formation of 8a from 1 + PMe3.
The binding preference of PR3 ligand with the alkylidyne (1) or
of 3TS4-a and 3TS4-b have a trigonal bipyramidal geometry. Be-
cause the smaller distortion of 3TS4-a and 3TS4-b from their pre-
cursor intermediates, the free energy barriers for ligated
bis(alkylidene) (2) have already been analyzed in a molecular orbi-
tal diagram [6a]. Here, we use an energy-decomposition analysis to
probe the binding energy [41,42]. This strategy decomposes the
binding energy into two parts, deformation energy and interaction
energy. The deformation energy is the energy induced by the dis-
tortion of binding partners into the geometries adopted in the cor-
responding adduct. As shown in Fig. 2, the relative energies of 3a-
frag and 4a-frag were calculated by summating the single point
energy of the tungsten fragment and the PMe3 fragment of 3a
and 4a, respectively. The interaction energy is the inherent energy
tautomerization
(3TS4-a:
28.1 kcal molꢀ1
and
3TS4-b:
29.0 kcal molꢀ1) are much lower than that of ligandless tautomer-
ization (1TS2: 38.7 kcal molꢀ1). Since the binding of PR3 ligand is a
barrierless process, it suggests that the formation of 4 undergoes
the pathway along 1 ? 3 ? 4, instead of 1 ? 2 ? 4.
3.6. Transformation of 4 to the alkyl alkylidene alkylidyne complexes 5
and 6
to separate two fragments. The interaction energy of
DEint for 4a is
higher than that for 3a, consistent with the previous MO analysis
[6a]. The deformation energy for 2 is less than that of 1, reflecting
that the relative minor geometrical distortion of 2 to 4a compared
to that of 1 to 3a.
As reported previously, heating the tautomeric equilibrium
mixtures of 3 ꢀ 4 in the presence of phosphines leads to the inter-
esting tungsten alkyl alkylidene alkylidyne complexes
W(CH2SiMe3)(@CHSiMe3)(„CSiMe3)(PR3)2 (R3 = Me3, 5a; Me2Ph,
5b) (Scheme 4) [6d]. The mechanism of the transformation of
4a,b to the tungsten alkyl alkylidene alkylidyne complexes is not
clear. Therefore, DFT calculations have been performed here to ex-
plore the reaction pathways. As shown in Scheme 4, the
3a ꢀ 4a ? 5a conversion includes two elementary steps, i.e.,
SiMe4 elimination and binding of a second equivalent of PR3. The
previous experimental kinetics study suggests that SiMe4 elimina-
tion occurs prior to the addition of a second PR3. That allows us to
focus on the step of SiMe4 elimination.
3.5. Formation of 4: theoretical studies of the kinetics
After discussing the thermodynamics of the formation of 4 from
1 and PR3, we turned to the mechanism and kinetics of this pro-
cess. As shown in Scheme 4, the transformation of 1 + PR3 to 4
may undergo two possible pathways, i.e. 1 ? 2 ? 4 or 1 ? 3 ? 4.
The difference in the two pathways is a reversal of the sequence
of ligand binding and tautomerization. It is hypothesized that the
ligand binding is a fast step. To investigate the binding process,
an energy surface scan along the WAP distance was carried out.
Fig. 3 shows the relative energy (Erel) versus WAP distance. The
association energy curves revealed that the PR3 binding steps are
indeed barrierless.
The potential energy surface of the formation of 8a, which then
binds PMe3 to give the bisphosphine product 5a, from 1 is shown
in Fig. 5. The coordination of the phosphine ligand PMe3 and the
equilibrium between 3a and 4a have been discussed above. 4a is
a thermodynamically stable intermediate. At higher temperatures,
Theoretical studies then focused on the tautomerization of the
-H transfer, which will be the rate-determining step for the for-
mation of 4. The structures and the free energy relative to refer-
ence (1 + PR3) of transition states of 1TS2, 3TS4-a and 3TS4-b are
shown in Fig. 4. 1TS2 adopts a tetrahedral geometry, while both
3a and 4a may undergo an
give the monophosphine and alkyl alkylidene alkylidyne interme-
diate 8a. For 3TS6-a, the -H transfer takes place from one alkyl
group to another alkyl group, while the -H transfer occurs from
alkylidene to alkyl group in 4TS6-a. The relative free energy of
a-H abstraction to eliminate SiMe4 and
a
a
a

P. Chen et al. / Polyhedron 58 (2013) 30–38
37
4TS6-a is 34.7 kcal molꢀ1, 5.5 kcal molꢀ1 higher than 3TS6-a, sug-
gesting that the pathway 4a ? 3a ? 8a is more favorable for the
formation of the tungsten alkyl alkylidene alkylidyne complex
5a. The overall enthalpy and free energy barrier from 4a to 8a
are calculated to be 32.2 and 31.2 kcal molꢀ1, in a reasonable
agreement with the experimental measured values (29 and
28 kcal mol-1). The slightly higher barrier of transformation from
4a to 8a than that of the transformation from 1 to 4a, indicating
that 4a is a kinetically stable intermediate.
[2] See, e.g., the following recent papers: (a) S. Sarkar, K.P. McGowan, S.
Kuppuswamy, I. Ghiviriga, K.A. Abboud, A.S. Veige, J. Am. Chem. Soc. 134
(2012) 4509;
(b) A.M. Geyer, M.J. Holland, R.L. Gdula, J.E. Goodman, M.J.A. Johnson, J.W.
Kampf, J. Organomet. Chem. 708–709 (2012) 1;
(c) J. Heppekausen, R. Stade, R. Goddard, A. Fuerstner, J. Am. Chem. Soc. 134
(2012) 11045;
(d) J.G. Andino, H. Fan, A.R. Fout, B.C. Bailey, M.-H. Baik, D.J. Mindiola, J.
Organomet. Chem. 696 (2011) 4138;
(e) S. Aguado-Ullate, J.J. Carbo, O. Gonzalez-del Moral, A. Martin, M. Mena, J.-
M. Poblet, C. Santamaria, Inorg. Chem. 50 (2011) 6269;
(f) X. Wu, C.G. Daniliuc, C.G. Hrib, M. Tamm, J. Organomet. Chem. 696 (2011)
4147;
(g) B.A. Dougan, Z.-L. Xue, Sci. China Chem. 54 (2011) 1903;
(h) D. Zhang, Y. Zhang, K. Gao, C.-S. Wang, Chin. J. Inorg. Chem. 24 (2008) 861;
(i) S. Chen, M.H. Chisholm, Inorg. Chem. 48 (2009) 10358;
(j) X. Solans-Monfort, C. Coperet, O. Eisenstein, J. Am. Chem. Soc. 132 (2010)
7750;
(k) M. Yu, I. Ibrahem, M. Hasegawa, R.R. Schrock, A.H. Hoveyda, J. Am. Chem.
Soc. 134 (2012) 2788;
(l) J. Zhu, G. Jia, Z. Lin, Organometallics 25 (2006) 1812;
(m) B.C. Bailey, R.R. Schrock, S. Kundu, A.S. Goldman, Z. Huang, M. Brookhart,
Organometallics 28 (2009) 355;
(n) W. Zhang, J. Yamada, K. Nomura, Organometallics 27 (2008) 5353;
(o) N. Merle, J. Trebosc, A. Baudouin, I.D. Rosal, L. Maron, K. Szeto, M. Genelot,
A. Mortreux, M. Taoufik, L. Delevoye, J. Am. Chem. Soc. 134 (2012) 9263;
(p) Y.P. Barinova, Y.E. Begantsova, N.E. Stolyarova, I.K. Grigorieva, A.V.
Cherkasov, G.K. Fukin, Y.A. Kurskii, L.N. Bochkarev, G.A. Abakumov, Inorg.
Chim. Acta 363 (2010) 2313.
4. Concluding remarks
Addition of two mono-dentate phosphines (PR3 = PMe3 and
PMe2Ph) to W(CH2SiMe3)3(„CSiMe3) (1), forming W(CH2SiMe3)3-
(„CSiMe3)(PR3) (3a, 3b) and their bis-alkylidene tautomers
W(CH2SiMe3)2(@CHSiMe3)2(PR3) (4a, 4b), has been found to be
reversible. It should be noted that the reaction of excess
W(CH2SiMe3)3(„CSiMe3) (1) and Me2PCH2CH2PMe2 (DMPE) ap-
pears to be more complicated than the reaction of 1 with the
mono-dentate phosphines, and is not pursued in the current work.
Using a comparison with the experimental thermodynamic data,
benchmarking work has been carried out and M06 is found to be
an appropriate method for this system. Consequently, a detailed
computational study has been performed to investigate the reac-
tion of W(CH2SiMe3)3(„CSiMe3) (1) with PR3. Because the binding
[3] (a) R.R. Schrock, J.D. Fellmann, J. Am. Chem. Soc. 100 (1978) 3359;
(b) D.N. Clark, R.R. Schrock, J. Am. Chem. Soc. 100 (1978) 6774;
(c) W. Mowat, G. Wilkinson, J. Chem. Soc., Dalton Trans. (1973) 1120;
(d) R.A. Andersen, M.H. Chisholm, J.F. Gibson, W.W. Reichert, I.P. Rothwell, G.
Wilkinson, Inorg. Chem. 20 (1981) 3934;
(e) L.-T. Li, M. Hung, Z.-L. Xue, J. Am. Chem. Soc. 117 (1995) 12746;
(f) J.K.C. Abbott, L.-T. Li, Z.-L. Xue, J. Am. Chem. Soc. 131 (2009) 8246.
[4] (a) Z.-L. Xue, L.-T. Li, L.K. Hoyt, J.B. Diminnie, J.L. Pollitte, J. Am. Chem. Soc. 116
(1994) 2169;
of PR3 ligand is barrierless and the PR3 facilitates the
a-H migra-
tion, the formation of 4 undergoes the pathway along 1 ? 3 ? 4,
instead of 1 ? 2 ? 4. Heating the tautomeric equilibrium mixtures
of 3 ꢀ 4 in the presence of the phosphine gives the tungsten
alkyl alkylidene alkylidyne complexes W(CH2SiMe3)(@CHSiMe3)-
(„CSiMe3)(PR3)2 (PR3 = PMe3, 5a; PMe2Ph, 5b). The potential
energy surface of the formation of 8a indicates that the relative
free energy of 4TS6-a is higher than 3TS6-a. This suggests the
formation of 8a along the pathway 4a ? 3a ? 8a, followed by
binding of 8a to PMe3 to give the bisphosphine product 5a. This
study provides a missing piece to understand the role of 4 plays
along the pathway of the reaction of 1 and PR3. 4 is found to be a
thermodynamically and kinetically stable intermediate.
(b) Y.-D. Wu, K.W.K. Chan, Z.-L. Xue, J. Am. Chem. Soc. 117 (1995) 9259;
(c) L.-T. Li, Z.-L. Xue, G.P.A. Yap, A.L. Rheingold, Organometallics 14 (1995)
4992;
(d) L.-T. Li, J.B. Diminnie, X.-Z. Liu, J.L. Pollitte, Z.-L. Xue, Organometallics 15
(1996) 3520;
(e) J.B. Diminnie, H.D. Hall, Z.-L. Xue, Chem. Comm. (1996) 2383;
(f) X.-Z. Liu, L.-T. Li, J.B. Diminnie, G.P.A. Yap, A.L. Rheingold, Z.-L. Xue,
Organometallics 17 (1998) 4597.
[5] M.R. Churchill, W.J. Youngs, Inorg. Chem. 18 (1979) 2454.
[6] (a) L.A. Morton, X.-H. Zhang, R. Wang, Z. Lin, Y.-D. Wu, Z.-L. Xue, J. Am. Chem.
Soc. 126 (2004) 10208;
(b) L.A. Morton, R. Wang, X. Yu, C.F. Campana, I.A. Guzei, G.P.A. Yap, Z.-L. Xue,
Organometallics 25 (2006) 427;
(c) B.A. Dougan, Z.-L. Xue, Organometallics 28 (2009) 1295;
(d) L.A. Morton, S. Chen, H. Qiu, Z.-L. Xue, J. Am. Chem. Soc. 129 (2007) 7277.
[7] (a) J.B. Diminnie, Z.-L. Xue, J. Am. Chem. Soc. 119 (1997) 12657;
(b) J.B. Diminnie, J.R. Blanton, H. Cai, K.T. Quisenberry, Z.-L. Xue,
Organometallics 20 (2001) 1504;
(c) J.R. Blanton, T.-N. Chen, J.B. Diminnie, H. Cai, Z.-Z. Wu, L.-T. Li, K.R.
Sorasaenee, K.T. Quisenberry, H.-J. Pan, C.-S. Wang, S.-H. Choi, Y.-D. Wu, Z.-Y.
Lin, I.A. Guzei, A.L. Rheingold, Z.-L. Xue, J. Mol. Catal. A 190 (2002) 101;
(d) J.B. Diminnie, X.-Z. Liu, H. Cai, Z.-Z. Wu, J.R. Blanton, A.A. Tuinman, K.T.
Quisenberry, C.E. Vallet, R.A. Zuhr, D.B. Beach, Z.-H. Peng, Y.-D. Wu, T.E.
Concolino, A.L. Rheingold, Z.-L. Xue, Pure Appl. Chem. 73 (2001) 331.
[8] A.M. LaPointe, R.R. Schrock, W.M. Davis, J. Am. Chem. Soc. 117 (1995) 4802.
[9] K.G. Caulton, M.H. Chisholm, W.E. Streib, Z.-L. Xue, J. Am. Chem. Soc. 113
(1991) 6082.
Acknowledgments
Financial support from the National Science Foundation of Chi-
na (21133002), US National Science Foundation (CHE-1012173),
the Shenzhen Peacock Program, and Peking University Shenzhen
Graduate School is acknowledged. We would like to thank Prof.
Olaf Wiest for helpful discussion and the Shenzhen Supercomputer
Center for computational resources.
References
[10] Z.-L. Xue, K.G. Caulton, M.H. Chisholm, Chem. Mater. 3 (1991) 384.
[11] Z.-L. Xue, S.-H. Chuang, K.G. Caulton, M.H. Chisholm, Chem. Mater. 10 (1998)
2365.
[1] See, e.g., the following reviews and references therein: (a) R.R. Schrock, Chem.
Rev. 102 (2002) 145;
[12] K.G. Caulton, R.H. Cayton, M.H. Chisholm, J.C. Huffman, E.B. Lobkovsky, Z.-L.
Xue, Organometallics 11 (1992) 321.
(b) R.H. Grubbs, Tetrahedron 60 (2004) 7117;
(c) R.R. Schrock, Chem. Rev. 109 (2009) 3211;
[13] (a) T.-N. Chen, Z.-Z. Wu, L.-T. Li, K.R. Sorasaenee, J.B. Diminnie, H.-J. Pan, I.A.
Guzei, A.L. Rheingold, Z.-L. Xue, J. Am. Chem. Soc. 120 (1998) 13519;
(b) T.-N. Chen, X.-H. Zhang, C.-S. Wang, S.-J. Chen, Z.-Z. Wu, L.-T. Li, K.R.
Sorasaenee, J.B. Diminnie, H.-J. Pan, I.A. Guzei, A.L. Rheingold, Y.-D. Wu, Z.-L.
Xue, Organometallics 24 (2005) 1214.
[14] R.R. Schrock, J. Sancho, S.F. Pederson, Inorg. Synth. 26 (1989) 45.
[15] M.J. Frisch, G.W. Trucks, H.B. Schlegel, G.E. Scuseria, M.A. Robb, J.R. Cheeseman,
G. Scalmani, V. Barone, B. Mennucci, G.A. Petersson, H. Nakatsuji, M. Caricato,
X. Li, H.P. Hratchian, A.F. Izmaylov, J. Bloino, G. Zheng, J.L. Sonnenberg, M.
Hada, M. Ehara, K. Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y.
Honda, O. Kitao, H. Nakai, T. Vreven, J.A. Montgomery Jr., J.E. Peralta, F. Ogliaro,
M. Bearpark, J.J. Heyd, E. Brothers, K.N. Kudin, V.N. Staroverov, R. Kobayashi, J.
Normand, K. Raghavachari, A. Rendell, J.C. Burant, S.S. Iyengar, J. Tomasi, M.
Cossi, N. Rega, N.J. Millam, M. Klene, J.E. Knox, J.B. Cross, V. Bakken, C. Adamo, J.
Jaramillo, R. Gomperts, R.E. Stratmann, O. Yazyev, A.J. Austin, R. Cammi, C.
(d) X. Wu, M. Tamm, Beilstein J. Org. Chem. 7 (2011) 82;
(e) M.C. Haibach, S. Kundu, M. Brookhart, A.S. Goldman, Acc. Chem. Res. 45
(2012) 947;
(f) K. Nomura, J. Chin. Chem. Soc. 59 (2012) 139;
(g) F. Rascon, C. Coperet, J. Organomet. Chem. 696 (2011) 4121;
(h) R.R. Schrock, Dalton Trans. 40 (2011) 7484;
(i) A.L. Odom, Dalton Trans. 40 (2011) 2689;
(j) Z.-L. Xue, L.A. Morton, J. Organomet. Chem. 696 (2011) 3924;
(k) D.J. Mindiola, Acc. Chem. Res. 39 (2006) 813;
(l) R.E. Da Re, M.D. Hopkins, Coord. Chem. Rev. 249 (2005) 1396;
(m) X. Yu, L.A. Morton, Z.-L. Xue, Organometallics 23 (2004) 2210;
(n) J. Feldman, R.R. Schrock, Prog. Inorg. Chem. 39 (1991) 1;
(o) A. Mayr, H. Hoffmeister, Adv. Organomet. Chem. 32 (1991) 227;
(p) W.A. Nugent, J.M. Mayer, Metal-Ligand Multiple Bonds, Wiley, New York,
1988.

38
P. Chen et al. / Polyhedron 58 (2013) 30–38
Pomelli, J.W. Ochterski, R.L. Martin, K. Morokuma, V.G. Zakrzewski, G.A. Voth,
[27] X. Zhang, H. Schwarz, Theor. Chem. Acc. 129 (2011) 389.
[28] A.V. Marenich, C.J. Cramer, D.G. Truhlar, J. Phys. Chem. B 113 (2009) 6378.
[29] Y. Zhao, D.G. Truhlar, Acc. Chem. Res. 41 (2008) 157.
[30] Y. Zhao, D.G. Truhlar, Org. Lett. 9 (2007) 1967.
[31] Y. Zhao, D.G. Truhlar, J. Chem. Theor. Comput. 5 (2009) 324.
[32] H.-C. Yang, Y.-C. Huang, Y.-K. Lan, T.-Y. Luh, Y. Zhao, D.G. Truhlar,
Organometallics 30 (2011) 4196.
[33] Y. Zhao, D.G. Truhlar, Chem. Phys. Lett. 502 (2011) 1.
[34] D.H. Wertz, J. Am. Chem. Soc. 102 (1980) 5316.
[35] B.O. Leung, D.L. Reid, D.A. Armstrong, A. Rauk, J. Phys. Chem. A 108 (2004)
2720.
[36] D. Ardura, R. López, T.L. Sordo, J. Phys. Chem. B 109 (2005) 23618.
[37] J. Cooper, T. Ziegler, Inorg. Chem. 41 (2002) 6614.
[38] P.A. Dub, R. Poli, J. Mol. Catal. A 324 (2010) 89.
[39] M. Wang, T. Fan, Z. Lin, Organometallics 31 (2012) 560.
[40] M. Sumimoto, N. Iwane, T. Takahama, S. Sakaki, J. Am. Chem. Soc. 126 (2004)
10457.
P. Salvador, J.J. Dannenberg, S. Dapprich, A.D. Daniels, Ö. Farkas, J.B. Foresman,
J.V. Ortiz, J. Cioslowski, D.J. Fox, Gaussian 09, Revision A.1, Gaussian, Inc.,
Wallingford, CT, 2009.
[16] A.D. Becke, J. Chem. Phys. 98 (1993) 5648.
[17] C. Lee, W. Yang, R.G. Parr, Phys. Rev. B 37 (1988) 785.
[18] Y. Zhao, D.G. Truhlar, Theor. Chem. Acc. 120 (2008) 215.
[19] J.P. Perdew, Phys. Rev. B 33 (1986) 8822.
[20] A.D. McLean, G.S. Chandler, J. Chem. Phys. 72 (1980) 5639.
[21] K. Raghavachari, J.S. Binkley, R. Seeger, J.A. Pople, J. Chem. Phys. 72 (1980) 650.
[22] R.C. Binning Jr., L.A. Curtiss, J. Comp. Chem. 11 (1990) 1206.
[23] F. Weigend, R. Ahlrichs, Phys. Chem. Chem. Phys. 7 (2005) 3297.
[24] (a) The basis sets were obtained from the Gaussian Basis Set Library EMSL at
https://bse.pnl.gov/bse/portal.;
(b) D. Feller, J. Comput. Chem. 17 (1996) 1571;
(c) K.L. Schuchardt, B.T. Didier, T. Elsethagen, L. Sun, V. Gurumoorthi, J. Chase,
J. Li, T.L. Windus, J. Chem. Inf. Model 47 (2007) 1045.
[25] B. Metz, H. Stoll, M. Dolg, J. Chem. Phys. 113 (2000) 2563.
[26] X. Zhang, H. Schwarz, Chem. Eur. J. 16 (2010) 5882.
[41] T. Ziegler, A. Rauk, Theor. Chim. Acta 46 (1977) 1.
[42] K. Morokuma, Acc. Chem. Res. 10 (1977) 294.