Download Gas phase reactions of La with acetone: A density functional theory

Indian Journal of Chemistry
Vol. 51A, May 2012, pp. 669-675
Gas phase reactions of La+ with acetone: A density functional theory investigation
Gui-hua Chen* & Danxia Liang
School of Pharmaceutical and Chemical Engineering, Taizhou University, Linhai, 317000, PR China
Email: [email protected]
Received 20 September 2011; revised and accepted 10 April 2012
The gas phase reaction of La+ cation with acetone is investigated using density functional theory. Both ground and
excited state potential energy surfaces are investigated in detail. The present results show that the title reaction starts with
the formation of an O-attached complex (IM0). All possible pathways starting with C−O, C−H, and C−C activation have
been searched. These reactions can lead to four different products, (1LaO++C3H6, 2LaCH2COCH3++H, 1LaCOCH2++CH4 and
1
LaCH2++CH3CHO). The minimum energy reaction path is found to involve spin inversion after IM0 formation, and this
potential energy curve-crossing dramatically affects the reaction exothermic. As all the triplet intermediates, transition states
and products involved in the reaction lie below the ground reactants (3La++ CH3COCH3) after IM0 formation, the reaction is
expected to occur spontaneously over the singlet potential energy surface. The present results may be helpful in
understanding the mechanism of the title reaction and further experimental investigation of the reaction.
Keywords: Theoretical chemistry, Density functional calculations, Gas phase reactions, Potential energy surfaces,
Acetone, Lanthanum
Due to their great importance in catalytic and material
science1-6, transition metal ions and their oxides have
been widely used in many industrial processes. The
gas phase reactions of transition metal atoms and ions
with small organic molecules have aroused
considerable attention; many studies have been
reported on the investigation of the relevant
mechanism7-29. Gas phase reactions offer a unique
possibility to probe the intrinsic properties of reactive
organometallic species. The details of the activation
of C−H, C−C and C−O bonds in the initial steps are
of fundamental interest in catalysis. As one of the
most important industrial synthetic raw materials,
acetone is of particular interest in understanding C−H,
C−C and C−O bonds activation in small hydrocarbons
by transition metal atoms and ions, hence in the past
several years, the reactions of metal cations with
acetone have been extensively studied15,28-32.
Experimentally, it has been shown that the early
transition metal cations such as Sc+ and Ti+ can react
readily with acetone to give MO+ and C3H6 as
products, while for the later transition metal cations,
such as Fe+, Co+ and Ni+, the reaction gives
completely different products, i. e., MCO+ and C2H6.
Chen et al.28 have reported the potential energy
surfaces of the gas phase reaction of Ni+ with acetone.
Their theoretical study rationalizes experimental
findings and gives a deeper insight to the reaction
mechanism, leading to the loss of C2H6 and CO. For
the reaction between early transition metal ion with
acetone, Kim et al.29 investigated the detailed reaction
mechanism of Ti+ towards acetone, and the most
favorable reaction pathway leads to the formation of
MO+ and C3H6, which is quite different from that of
Ni+ with acetone. From the discussion above, one can
see that the earlier studies focused mainly on the
reactions between acetone and first row transition
metal cations such as Co+, Fe+, Cr+, Ti+, Ni+, etc.28-32.
The heavier metal atoms and cations have received
relatively less attention, and only a few theoretical
and experimental studies have been reported 15. Bayse
et al.15 have investigated the reaction between yttrium
atom and acetone in detail. Their studies based on
crossed molecular beams experiments predict CO
elimination to be the most feasible channel. It may be
noted that in their study no YO+C3H6 products
channel was observed, which is quite different from
the observations of reaction of the first row transition
metal with acetone.
From the previous experimental and theoretical
studies15,28-32 on the reactions between acetone and
first row cations or the larger transition metal Y, it is
clear that the change from first row transition metal to
metal Y leads to change in reaction mechanism, both
in the early and late stages of the reaction. Studies on
typical lanthanoids and acetone reactions has
670
INDIAN J CHEM, SEC A, MAY 2012
important theoretical and experimental significances.
To the best of our knowledge, few studies with typical
lanthanoid elements have been performed in this
reaction system. It is thus of interest to know if the
mechanism of the reaction of lanthanum, a typical
lanthanide, with acetone is similar to that of first or
second row early transition metal with acetone. We
have investigated the reactions of La+ cation with
acetone by using DFT methods. This is important
since there is no experimental study reported on the
reaction of La+ cation with acetone. The present work
is expected to predict further experimental findings
that could not be reached experimentally under the
considered conditions.
Methodology
The potential energy surface for the title reaction has
been considered in detail. All molecular geometries
(reactants, intermediates, transition states and products)
were optimized by employing the UB3LYP density
functional theory method33. In all the calculations, the
basis set used consisted of the quasi relativistic
effective core potential (ECP) of Stuttgart on La. The
5d and 6s in La were treated explicitly by a (7s6p5d)
Gaussian basis set contracted to (5s4p3d)34. For
nonmetal atoms, a standardized 6−311++G** basis set
was used. For all the species involved in the reaction,
the enthalpy at 0 K is discussed in the present study,
and this energy is used to construct the profiles of the
PES. Harmonic vibration analyses were performed at
the same level of theory for all the optimized stationary
points to determine their properties (minimum or first
order saddle point) and to evaluate the zero point
vibrational energies (ZPEs). To verify whether the
located transition states connected the expected
minima, intrinsic reaction coordinate (IRC)
calculations were carried out for each transition state at
the same level35. All calculations in the present study
were performed using the Gaussian 03 program36.
Results and Discussion
The optimized geometries of the stationary points
over the PESs for the title reaction are depicted in
Fig. 1 and the profiles of the PES are shown in Fig. 2.
We also inspected the values of <S2> for all species
involved in the reaction, and found that the deviation
of <S2> is less than 5 %. This shows that spin
contamination is small in all the calculations.
As shown in Fig. 2, the title reaction starts with the
formation of an O-attached complex (IM0). From
Fig. 1, one can see that in 3IM0 or 1IM0, the
lanthanum cation bonds with the oxygen, with the
La−O distance being 2.223 (triplet) and 2.247
(singlet) Å. Owing to electron donation from the
oxygen lone pair orbitals to the La center, the C−O
bond is weakened, and is lengthened by 0.087 and
0.058 Å in 3IM0 and 1IM0 respectively. The
weakened C−O bonds are expected to be easily
activated by the metal atom to produce different
products. Energetically, 3IM0 is -48.93 kcal/mol
lower than 3La+ + CH3COCH3. For 1IM0, it is
computed to be 8.04 kcal/mol higher in energy than
the corresponding triplet initial complex 3IM0. It
should be pointed out that although several trials were
undertaken to search for possible transition states that
connect the reactants and IM0, no such transition
states were obtained. Obviously, the formation of IM0
is a barrier-free exothermic reaction. Starting from
IM0, the oxidative addition of acetone toward the La+
cation accounts for IM1 formation which stabilizes
the system by 13.62 kcal/mol over singlet PES, while
the analogue in triplet is 4.54 kcal/mol higher than
1
IM0. Obviously, the intersystem crossing occurs
between IM0 and IM1. From Fig. 1, one can see that
IM1 is a η2-CH3COCH3-metal encounter complex
with the lanthanum cation binding with the carbon
and oxygen simultaneously. The electron donation
from the oxygen lone pair orbitals to the lanthanum
center stretches the C−O bond of carbonyl in acetone
and thus weakens this bond. Once the encounter
complex IM1 is formed, three reaction pathways are
possible, i. e., C−O, C−H, and C−C bonds activation,
each of which will be discussed in detail over singlet
and triplet PESs.
First, we will discuss the singlet PES. The first
product generated from the reaction between
lanthanum cation and acetone is P1(1LaO++C3H6). As
shown in Fig. 2, this reaction channel starts with the
formation of the η2-CH3COCH3-metal complex 1IM1.
Along this reaction pathway, the lanthanum cation
can insert into the C−O bond via a transition state
1
TS12 with the energy barrier of 18.88 kcal/mol. From
Fig. 1, one can see that the distance between the
carbon and oxygen atoms in carbonyl in 1TS12 is
lengthened from 1.451 to 2.724 Å, which indicates
that the C−O bond is activated. The corresponding
normal mode of imaginary frequency corresponds to
the rupture of the C−O bond with the La cation being
inserted into it. The C−O bond activation
intermediate 1IM2 is 36.58 kcal/mol more stable in
energy than 3La+ + CH3COCH3. As shown in Fig. 1,
CHEN & LIANG: DFT STUDIES ON GAS PHASE REACTION OF La+ + ACETONE
671
1.932
2.095
1.212
2.247
2.163
1.517
2.875
1.451
1.089
1.089
1.517
1.270
1.470
1.513
1.488
1.470
1.488
1
CH3COCH3
1
IM0
2.367
1.903
1.255
1.230
22
3.2
1
IM1
IM2
2.484
2.056
2.418
2.228
2.144
1.477
2.458
3.018
2.658
1.513
1.483
1.331
1.380
1.345
1.493
1IM3
1IM4
1IM5
2.129
1IM6
1.972
1.397
2.238
2.189
2.304
1.331
2.808
2.965
2.015
1.334
1.291
1.492
2.251
1.491
2.546
2.038
1.471
1.492
1.469
1TS01 (75.9i)
1IM7
1TS12 (109.8i)
1TS14 (392.1i)
2.161
2.197
2.204
1.914
2.132
1.382
2.430
2.342 2.498
1.518
1.613
1.261 1.307
1.381
2.764
1.266
1.380
2.573
2.383
1.334
1.368
1.408
1.454
1.386
1.504
1TS15 (948.4i)
1TS16 (973.8i)
1TS23 (1040.0i)
1TS47 (1474.8i)
2.216
2.232
2.223
1.315
1.281
1.491
1.498
1.498
1.299
1.495
1.495
1.491
3
IM0
3IM1
3
TS01 (70.8i)
Fig. 1Optimized geometries for the various stationary points located on the La++CH3COCH3 potential energy surfaces
(distances in angstroms).
672
INDIAN J CHEM, SEC A, MAY 2012
Fig. 2Potential energy surface profiles for the reaction of
La++CH3COCH3. [The relative energies include zero point energy
correction (at the harmonic oscillator level)].
the C−O distance in 1IM2 is lengthened to 2.875 Å,
which means that this bond has ruptured thoroughly.
The calculated La−C and La−O bonds are 2.658 and
2.875 Å respectively. Along this reaction coordinate,
the C−O bond activation is followed by a H-shift to
form 1IM3, with an energy barrier of 10.35 kcal/mol.
From Fig. 1, one can see that via the transition state
1
TS23, there is a methyl H migration to the middle
carbon centre. In the structure of 1IM3, the La cation
is weakly coordinated to two carbon atoms of
CH3CHCH2 simultaneously. This intermediate is
determined to be the deepest energy well along the
reaction coordinate, lying 76.04 kcal/mol below the
ground reactants. In 1IM3, the geometry parameters of
the CH3CHCH2 unit are similar to those of propene
except for the C=C bond, which is a little longer than
the one in propene (1.345 vs 1.331 Å). NBO analysis
indicates that there exists a strong interaction between
the 5d orbital of lanthanum and the π* orbital of C=C
bond, which can account for the high dissociation
energy of 18.66 kcal/mol between 1LaO+ and C3H6 in
1
IM3. From Fig. 2, the reaction energy of the
3
La++CH3COCH3 → 1LaO++C3H6 reaction is
calculated to be -57.38 kcal/mol at the chosen level.
In addition, as all the critical points along this C−O
bond activation pathway are below the reactants,
obviously the reaction leading to 1LaO++C3H6
formation is spontaneous and a highly exothermic
process. Davis et al.15 have examined the reaction
between yttrium atom and CH3COCH3, both
experimentally and theoretically. They did not
observe any YO+C3H6 species, or YO+C3H6
formation channel. Though no relevant experimental
studies on reaction of lanthanum with acetone are
available, from the discussion above we propose that
even at low energies the MO++C3H6 products can be
observed in the reaction of lanthanum cation with
acetone. As is known, the metal Y can usually be
categorized as a lanthanide analogous metal because
of chemical properties similar to these of the
lanthanides. From the above discussion, we can see
that as representative of early transition metals, the
typical lanthanide (such as La) is more effective in
capturing propene from acetone than the analogous
lanthanide (such as Y). Obviously, the reaction
mechanism of Y atom with CH3COCH3 can not be
applicable to lanthanides such as the representative La.
Now we will discuss the C−H bond activation
channel, which may lead to product P2
(2LaCH2COCH3++H) formation. It is very clear from
Fig. 2 that this route also starts with the formation of
η2-CH3COCH3-metal encounter complex, 1IM1.
Along the reaction channel, the next step corresponds
to the metal-mediated methyl H migration followed
by the non-reactive-dissociation. 1IM1 and 1IM6 are
connected through a direct, one-step methyl C−H
activation occuring via saddle point 1TS16.
Geometrically, the activation C−H bond is elongated
to 1.613 Å and synchronously the distance between
lanthanum and hydrogen atoms is shortened to
2.132 Å. The imaginary frequency of 1TS16 is
973.8i cm-1, and the normal mode corresponds to the
rupture of methyl C−H bond with the lanthanum
cation binding with the H atom. One dissociation
channel of 1IM6 is direct rupture of the H─La bond to
account for the product P2 (2LaCH2COCH3++H).
However, this is a highly endothermic process with
dissociation energy of 61.72 kcal/mol. In addition, as
the rate determining step (1TS16) along this channel is
very low in energy, we propose that the species, 1IM6,
is rather abundant in title reaction.
With respect to the C−C bond activation
mechanism, one possible channel has been confirmed,
which
may
lead
to
formation
of
P3
1
+
1
( LaCOCH2 +CH4). Originating from IM1, the C−C
bond activation species, 1IM4, can be formed by the
CHEN & LIANG: DFT STUDIES ON GAS PHASE REACTION OF La+ + ACETONE
insertion of La cation into the C−C bond.
Energetically, the transition state (1TS14) is calculated
to be -34.80 kcal/mol below the energies of the
ground reactants. Compared with the PESs of the
C−O and C−H activation routes as discussed above,
we found that the C−C activation is most unfavorable
due to the relatively high energy barrier. As shown in
Fig. 1, the C−C bond in 1IM4 is ruptured thoroughly;
the two calculated La−C bonds and one La−O bond
are 2.458, 2.418, and 2.367 Å respectively.
Subsequently, originated from this C−C bond
activation species, 1IM4, the methyl-H shift could
convert it into 1IM7, over a barrier height of
32.41 kcal/mol. Through this step, the methyl
hydrogen atom is transfered toward the other terminal
methyl group, to form a CH4 molecule subsequently.
From Fig. 2, one can see that the second step, i.e., the
process of 1IM7 formation is the rate-determining one
along this path. In 1IM7, the methane moiety is η2
coordinated to the La cation. The binding energy
between CH4 and 1LaCOCH2+ is calculated to be
7.98 kcal/mol. Obviously, from Fig. 2, one can see
that this CH4 loss channel, 3La++CH3COCH3→
1
LaCOCH2++CH4, is also spontaneous in energy, and
highly exothermic by 38.17 kcal/mol. In earlier
theoretical studies on the gas phase reactions of
acetone with first-row transition metal cations and
some second-row transition metal atom (such as Ti+,
Ni+ and Y)15,28,29, no feasible channel for formation of
M(COCH2)+CH4 was determined. Though no
experimental researches on lanthanum with acetone
are available, from the discussion above, we found all
the species along this decarbonylation channel are
computed to be below 3La+ + CH3COCH3. We
propose that even at low energies, these
decarbonylation products can be observed in the
reaction of lanthanum cation with acetone. So,
compared with the metals, Ti+, Ni+ and yttrium, the
lanthanum cation may capture methane from acetone
more effectively.
Alternatively, the direct one-step H shift occurring
via saddle point 1TS15 may yield the complex 1IM5. A
striking feature of this new intermediate is the
dicoordination of La cation with the aldehyde and
methylene group. It should be pointed out such a
minimum was not observed in the previous studies of
M (or M+, M = Ti, Ni, and Y) +CH3COCH315,28,29.
When compared with the 1IM1 formation step, one
can see that this step, i.e., 1IM1→ 1IM5 isomerization
process, requires a high activation energy of
673
43.16 kcal/mol, but with the excess energy gained in
the formation of 1IM1, this isomerization process can
compete favorably. One dissociation channel of 1IM5
is direct rupture of the O─La bond to give rise to the
product P4 (1LaCH2++CH3CHO); this dissociation
process is computed to be endothermic by
31.82 kcal/mol. In the previous studies on the reaction
between M (or M+, M = Ti, Ni, and Y) and
acetone15,28,29, no acetaldehyde and MCH2 species
were detected, and no feasible MCH2+CH3CHO
formation channel was observed theoretically. From
the present theoretical work, it is obvious that in the
reaction between lanthanum cation and acetone, the
C−C activation which leads to CH3CHO formation is
also feasible and spontaneous in energy.
Similar to that on the singlet PES, we have also
located possible activation channels on the triplet
PES. From the calculated result, we find that the
reaction mechanism on the triplet PES is in general
similar to that on the singlet one. The reaction starts
with a η2-CH3COCH3-metal complex followed by
three possible pathways: C−O, C−H and C−C
activation. Comparing all the relative energy in
Table S1 (Supplementary data) and channels shown in
Fig. 2, it is clear that the C−C insertion branch which
leads to product P3 (3LaCOCH2++ CH4) is most
feasible due to the relative low energy barrier. As all
the rate determining steps along these channels are
above 3La+ + CH3COCH3, these reaction pathways are
not spontaneous, which is quite different from that of
the singlet PES. As after IM1 formation, almost all
the intermediates, transition states and products
involved in the reaction on the singlet PES, lie below
the analogues on the triplet PES, obviously after IM1
formation, the singlet pathways are always
energetically preferred with respect to the
corresponding triplet routes. In addition, as the initial
O-attached complex 3IM0 is calculated to be
8.04 kcal/mol more stable than the singlet analogue
(see Fig. 2), we can speculate that the inter-system
singlet-triplet crossing occurs during the process of
3
IM0→ 1IM1. The aim of our following calculation is
to determine the region where the spin inversion
occurs, and to acquire the structure and energy of
crossing point between the two different potential
energy surfaces. We chose an approach suggested by
Yoshizawa et al.37 for locating the crossing points of
two PESs of different multiplicities approximately. The
potential energy profile of the singlet state from 1TS01
to 1IM1 is shown in Fig. 3. Along the IRC we find a
INDIAN J CHEM, SEC A, MAY 2012
674
channel because of the low energy barrier and
dissociation energy of exit-channel complex compared
with the C−C and C−H insertion reactions. The most
feasible channel may thus be described as:
3
La+ + CH3COCH3 → 3IM0 → CP → 1IM1 → 1TS12 →
1
IM2 → 1TS23 → 1IM3 → P1. The present theoretical
work adds new insight into the reaction between
transition metal M with acetone.
Supplementary data
Supplementary data, Table S1, associated with
this article, is available in the electronic from
at http://www.niscair.res.in/jinfo/ijca/IJCA 51A(05)
669-675_Suppl Data.pdf.
Fig. 3Potential energies from 3IM0 to 1IM1 along the triplet
IRC. [1, singlet; 2, triplet].
1
crossing point, CP, (after TS01) with energy of
-224.5607 Hartree. Thus, the reaction may jump from
the triplet PES to the singlet near the crossing point
CP without passing the transition state 3TS01. As can
be seen from Fig. 3, after passing point CP, the singlet
PES can provide a low energy reaction pathway
toward the η2-CH3COCH3-metal complex 1IM1.
Hence, the reaction is most likely to proceed through
the following steps: 3La+ + CH3COCH3 → 3IM0 →
CP → 1IM1, which is calculated to be exothermic by
54.51 kcal/mol. Actually, the reactions catalyzed by
metallic systems may often involve a change in the
spin states and proceed via a non-adiabatic path on
two or more potential energy surfaces, denoted as
“two state reactivity” (TSR)38-41, which has been
confirmed by experimental studies.
Conclusions
In the present study, the reaction mechanisms
between lanthanum cation and CH3COCH3 have been
investigated in detail, both on singlet and triplet PESs.
The reaction is expected to occur spontaneously over
the singlet PES after initial complex 1IM0 formation.
All possible pathways starting with C−O, C−H, and
C−C activation have been searched. The C3H6 loss of
acetone by La+ proceeds through the initial C−O
activation. The initial C−H activation accounts for the
H
elimination
products
(2LaCH2COCH3++H).
Originating from the intermediate complex 1IM1,
initial C−C bond activation may lead to two
decarbonylation products, 1LaCOCH2++ CH4 and
1
LaCH2++CH3CHO. Our calculations confirm the
initial carbonyl C−O activation pathway, which leads
to formation of 1LaO+ + C3H6. This serve as a major
Acknowledgement
This work was supported by the Zhejiang
Provincial Natural Science Foundation of China under
grant No. Y4090387.
References
1 Zhang Q & Bowers M T, J Phys Chem A, 108 (2004) 9755.
2 Halle L F, Crowe W E, Armentrout P B & Beauchamp J L,
Organometallics, 3 (1984) 1694.
3 Sonnenfroh D M & Farrar J M, J Am Chem Soc, 108 (1986)
3521.
4 Sievers M R, Jarvis L M & Armentrout P B, J Am Chem Soc,
120 (1998) 1891.
5 Haynes C L, Fisher E R & Armentrout P B, J Am Chem Soc,
118 (1996) 3269.
6 Tolbert M A & Beauchamp J L, J Phys Chem, 90 (1986)
5015.
7 Zhao L M, Zhang R R, Guo W Y, Wu S J & Lu X Q, Chem
Phys Lett, 414 (2005) 28.
8 Zhao L M, Zhang R R, Guo W Y & Lu X Q, Chem Phys
Lett, 431 (2006) 56.
9 Zhao L M, Guo W Y, Zhang R R, Wu S J & Lu X Q,
ChemPhysChem, 7 (2006) 1345.
10 Dai G L, Wang C F, Wu J Y & Zhong A G, Comput Theor
Chem, 965 (2011) 60.
11 Chen X F, Guo W Y, Zhao L M, Fu Q T & Ma Y, J Phys
Chem A, 111 (2007) 3566.
12 Ma Y, Guo W Y, Zhao L M, Hu S Q, Zhang J, Fu Q T &
Chen X F, J Phys Chem A, 111 (2007) 6208.
13 Chen X F, Guo W Y, Yang T F & Lu X Q, J Phys Chem A,
112 (2008) 5312.
14 Bayse C A, J Phys Chem A, 106 (2002) 4226.
15 Schroden J J, Teo M & Davis H F, J Chem Phys, 117 (2002)
9258.
16 Schroden J J, Davis H F & Bayse C A, J Phys Chem A, 111
(2007) 11421.
17 Porembski M & Weisshaar J C, J Phys Chem A, 105 (2001)
6655.
18 Stauffer H U, Hinrichs R Z, Willis P A & Davis H F, J Chem
Phys, 111 (1999) 4101.
19 Stauffer H U, Hinrichs R Z, Schroden J J & Davis H F,
J Phys Chem A, 104 (2000) 1107.
CHEN & LIANG: DFT STUDIES ON GAS PHASE REACTION OF La+ + ACETONE
20 Willis P A, Stauffer H U, Hinrichs R Z & Davis H F, J Phys
Chem A, 103 (1999) 3706.
21 Hinrichs R Z, Schroden J J & Davis H F, J Am Chem Soc,
125 (2003) 860.
22 Schroden J J, Teo M & Davis H F, J Phys Chem A, 106
(2002) 11695.
23 Moc J & Gordon S, Theor Chem Acc, 120 (2008) 243.
24 Li T H, Cheng W X, Liu J S, Xie X G & Cao H, J Mol Struct
(Theochem), 761 (2006) 83.
25 Li T H, Wang C M, Xie X G & Jia Z M, J Mol Struct
(Theochem),959 (2010) 8.
26 Zhang F Y, Guo W Y, Zhao L M, Lin X Q, Zhang L Z,
Zhu H Y & Shan H H, J Phys Chem A, 113 (2009) 7103.
27 Dai G L, Chen H, Yan H, Wang C F & Zhong A G, Indian
J Chem, 49A (2010) 431.
28 Chen X F, Guo W Y, Zhao L M & Fu Q T, Chem Phys Lett,
432 (2006) 27.
29 Kim J, Kim T K & Ihee H, J Phys Chem A, 113 (2009)
11382.
675
30 Schilling J B & Beauchamp J L, J Am Chem Soc, 110 (1988) 15.
31 Allison J & Ridge D P, J Am Chem Soc, 100 (1978) 163.
32 Tolbert M A & Beauchamp J L, J Am Chem Soc, 106 (1984)
8117.
33 Becke A D, J Chem Phys, 98 (1993) 1372.
34 Dolg M, Stoll H, Savin A & Preuss H, Theor Chim Acta, 75
(1989) 173.
35 Fukui K, Acc Chem Res, 14 (1981) 363.
36 Gaussian 03, rev. B04, (Gaussian Inc, Pittsburgh PA, USA)
2003.
37 Yoshizawa K, Shiota Y & Yamabe T, J Chem Phys,111
(1999) 538.
38 Fiedler A, Schroder D, Shaik S & Schwarz H, J Am Chem
Soc, 116 (1994) 10734.
39 Harvey J N, Poli R & Smith K M, Coord Chem Rev, 238
(2003) 347.
40 Zhang G B, Li S H & Jiang Y S, Organometallics, 22 (2003)
3820.
41 Schroder D, Shaik S & Schwarz H, Acc Chem Res, 33 (2000) 139.