Experimental and theoretical studies on the course of CO insertion into Pt-C and Pd-C bonds in neutral and cationic complexes, [MR(Cl){P(CH3)3}2] and [MR{P(CH3)3}2(s)]+BF 4<

Article abstract of DOI:10.1016/S0022-328X(00)00916-5

Behavior of neutral and cationic cis- and trans-monoorganoplatinum complexes has been examined and compared with that of the corresponding monoorganopalladium complexes. The cis- and trans-monoorganoplatinum complexes having two trimethylphosphine ligands

Full text of DOI:10.1016/S0022-328X(00)00916-5

www.elsevier.nl/locate/jorganchem
Journal of Organometallic Chemistry 622 (2001) 199–209
Experimental and theoretical studies on the course of CO insertion
into Pt–C and Pd–C bonds in neutral and cationic complexes,
[MR(Cl){P(CH3)3}2] and [MR{P(CH3)3}2(s)]+BF4– (M=Pt, Pd,
R=CH3, C6H5, s=coordinated solvent)
Yoshihito Kayaki a, Hirokazu Tsukamoto a, Masaki Kaneko a, Isao Shimizu a,
Akio Yamamoto a,*, Masanori Tachikawa b, Takahito Nakajima c
a Department of Applied Chemistry, Graduate School of Science and Engineering, and Ad6anced Research Institute for Science and Engineering,
Waseda Uni6ersity, Ohkubo, Shinjuku-ku, Tokyo 169-8552, Japan
b The Institute of Physical and Chemical Research (RIKEN), Wako, Saitama 351-0198, Japan
c Department of Applied Chemistry, Graduate School of Engineering, The Uni6ersity of Tokyo, Hongo, Bunkyo-ku, Tokyo 113-8656, Japan
Received 26 October 2000; received in revised form 24 November 2000; accepted 25 November 2000
Dedicated to Martin Bennett on the occasion of his retirement in recognition of his outstanding contribution to organometallic chemistry.
Abstract
Behavior of neutral and cationic cis- and trans-monoorganoplatinum complexes has been examined and compared with that of
the corresponding monoorganopalladium complexes. The cis- and trans-monoorganoplatinum complexes having two
trimethylphosphine ligands, [PtR(acetone){P(CH3)3}2]+BF4– (R=CH3 and C6H5) (4 and 5), have been prepared by removal of the
chloride ligand in neutral complexes, [PdR(Cl){P(CH3)3}2] (2 and 3) with one equivalent of AgBF4. The cationic platinum
complexes retaining the cis configuration of the parent neutral complexes, when prepared at low temperature, were found to be
isomerized into the trans complexes 5 above −10°C for the methyl complex 5a and above −30°C for the phenyl complex 5b.
Treatment of the cationic complexes 4 and 5 with CO gave no CO insertion product but afforded only the CO-coordinated
trans-monoorganoplatinum complexes. The reluctance of the organoplatinum complexes toward CO insertion stands in contrast
with the ease for the CO insertion of the corresponding organopalladium complexes. For clarifying the reasons of the marked
difference between the behavior of the two Group 10 metal complexes the ab initio molecular orbital calculations with the MP2
level have been performed. We have found the transition states for the cis to trans isomerization of the three-coordinated cationic
monoalkylplatinum complexes and for the CO insertion into the Pt–CH3 bond. The calculated barriers for the isomerization and
CO insertion in the cationic organoplatinum complex are higher than those for the corresponding organopalladium complex in
agreement with the experimental results showing the poor reactivity of the solvent-coordinated cationic organoplatinum
complexes toward CO insertion compared to the organopalladium complexes. The reason for the difference in the reactivity
between the palladium and platinum complexes can be ascribed to the relativistic effect of the platinum complexes. © 2001
Elsevier Science B.V. All rights reserved.
Keywords: CO insertion; Cis–trans isomerization; Palladium; Platinum; Monoalkyl complexes; Ab initio calculations
1. Introduction
organic compounds [1–3]. Besides cobalt and rhodium
complexes, palladium complexes have been utilized
most extensively in the carbonylations of aryl and vinyl
halides as well as of olefins. Examination of the behav-
ior of the monoorganopalladium complexes has pro-
vided important information regarding the elementary
steps in the catalytic reactions [4]. In the palladium-cat-
alyzed carbonylation and in olefin arylation (Mi-zoroki–
The CO insertion into a transition metal–carbon
bond constitutes one of the most essential elementary
processes in transition metal-catalyzed carbonylation of
* Corresponding author. Fax: +81-3-3204-6274.
E-mail address: akioyama@mn.waseda.ac.jp (A. Yamamoto).
0022-328X/01/$ - see front matter © 2001 Elsevier Science B.V. All rights reserved.
PII: S0022-328X(00)00916-5

200
Y. Kayaki et al. / Journal of Organometallic Chemistry 622 (2001) 199–209
Heck process [5]) as well as in olefin polymerization [6]
and copolymerization of olefin and CO [7], generation
of a cationic catalytic species has been often found to
enhance the catalytic activities. In our previous studies
[8] as well as in the studies by others [9–12] many
examples of the enhancement of the reactivities in the
catalytic systems by generating cationic mono-
organopalladium complexes from neutral mono-
organopalladium complexes have been presented.
Scheme 1 illustrates that the first order rate constant of
the CO insertion process into the neutral methyl-
(chloro)palladium complex is increased by 100 times by
converting the neutral complex into the cationic sol-
vent-coordinated complex by removal of the chloride
ligand with AgBF4 [8d,13].
We have also confirmed that the availability of a
coordination site for the incoming substrate such as CO
and olefin is quite important for the insertion process to
proceed. For a monoorganopalladium complex having
two monodentate tertiary phosphine ligands to undergo
a concerted insertion process, the monoorganopalla-
dium complex is required to possess a coordination site
available adjacent to the hydrocarbyl ligand for the
incoming substrate. In fact the reactivity of cationic
organopalladium complexes having a chelating diter-
tiary phosphine ligand such as 1,2-bis(dimethylphos-
phino)ethane (dmpe) and 1,2-bis(diphenylphosphino)-
ethane (dppe) toward CO insertion (Eq. (1)) was three
times greater than that of the cationic monoorganopal-
ladium complexes having two trimethylphosphine lig-
ands in mutually trans positions [8d,13].
um(II) complexes, cis-[PtR(X)L2], are known [17–21]
and their isomerization processes have been well docu-
mented [22]. However, precedents of the CO insertion
processes into neutral monoorganoplatinum complexes
have been quite limited [23–25]. The results of these
studies indicated that the CO insertion into the Pt–C
bond is less favorable than that into the Pd–C bond. A
fundamental question raised there is whether the differ-
ence in the reactivity of the organopalladium and
organoplatinum complexes for the CO insertion is due
to the higher barrier for the migratory CO insertion for
an organoplatinum complex or due to the presence of a
higher barrier for the trans to cis isomerization in the
monoorganoplatinum complex. Another question is
whether the CO insertion into the cationic complexes
proceeds under the constraint of the square planar
geometry or if there is another route through a five-co-
ordinate intermediate.
In the present paper we wish to present the results of
our attempts addressed to the above questions through
experimental as well as ab initio molecular orbital
theoretical studies.
2. Results and discussion
2.1. Synthesis and isomerization beha6ior of cationic
organoplatinum(II) complexes ha6ing two monodentate
tertiary phosphines
Although there are ample examples of cationic cis-
monoalkylplatinum complexes having bidentate phos-
phine ligands [26] as well as the cationic trans
complexes, trans-[PtRL2]+ [27], less studies have been
made on properties of the cis cationic organoplatinum
complexes having monodentate phosphine ligands [28–
30]. Such cationic three-coordinated monoalkylplat-
inum complexes, cis-[PtRL2]+, have been proposed as
key intermediates in CO insertion [31], olefin insertion
[32], b-hydrogen elimination [28], and cis–trans isomer-
ization [21]. For example, there are studies reporting
that the isomerization of neutral cis complexes occurs
via cationic three-coordinate platinum species with dis-
sociation of an anionic ligand of the neutral complexes
in a protic solvent [33,34] and some cationic cis
methanol-coordinated complexes can be formed by
protonolysis of cis-dialkylplatinum complexes [27,28].
We prepared first the neutral cis-monoorgano-
(1)
In spite of the obvious importance of the cationic cis
complexes in insertion steps, few examples of cis-
monoorganopalladium(II) bearing two monodentate
phosphine ligands have been reported [14–16]. On the
other hand, a variety of neutral cis-monoorganoplatin-
(chloro)platinum complexes
2
by treating cis-
diorganoplatinum complexes 1 with an equimolar
amount of HCl in ether at −78°C. The cis complexes
2 were isomerized into the trans complexes 3 by warm-
ing under reflux condition [35] (Eq. (2)). The cationic
cis-organoplatinum complexes 4a and 4b were gener-
ated by treatment of the neutral monoorganoplatinum
Scheme 1.

Y. Kayaki et al. / Journal of Organometallic Chemistry 622 (2001) 199–209
201
Scheme 2.
chloride complexes cis-[PtR(Cl){P(CH3)3}2] (R=CH3
and C6H5) (2a and 2b) with an equivalent each of
AgBF4 in acetone at −50°C as shown in Eq. (3). The
cationic cis complexes 4 were found to retain the cis
configurations of the starting neutral complexes 2 be-
low −30°C as confirmed by observation of the AB
pattern signals of their trimethylphosphine ligands with
(5)
2JPP coupling constant (14.7 Hz) in the 31P-NMR
2.2. Reaction of cationic organoplatinum complexes
with CO
a
1
spectra. The low J
coupling constants (1687 and
a
Pt–
1632 Hz for 4a and 4b) Pare typical of phosphorus atoms
a
Treatment of the cationic trans complexes 5 with
atmospheric pressure of CO afforded CO-coordinated
trans complexes [PtR(CO){P(CH3)3}2]+BF4– (R=CH3;
6a, C6H5; 6b) at room temperature without forming any
acyl complexes produced by CO insertion into the
Pt–C bond (Eq. (6)). The cationic cis complexes 4 did
not undergo the CO insertion but simply isomerized
rapidly to the trans carbonyl complexes 6 even at
−50°C, despite that they have an available coordina-
tion site cis to the methyl or phenyl group (Scheme 2).
in PMe3 trans to the carbon atom in platinum com-
1
plexes, while the values J
=4140 and 4132 Hz are
b
Pt–
consistent with the presence oPf a very weak trans donor
ligand, acetone. The cationic cis complexes 4, thus
obtained, are thermally unstable in solution. Following
the reaction course with H- and 31P-NMR indicated
1
that the complexes 4a and 4b are isomerized readily at
−10 and −30°C, respectively, to the corresponding
trans complexes 5a and 5b (Eq. (4)). The cationic
trans-monoorganoplatinum complexes 5a and 5b were
independently obtained by treatment of the neutral
trans-organoplatinum chloride complexes 3a and 3b
with an equimolar amount of AgBF4 in acetone at
−30°C (Eq. (5)).
The results on the isomerization behavior of
monoorganoplatinum complexes having two monoden-
tate phosphine ligands indicate that the cis to trans
isomerization takes place readily in the cationic
organoplatinum complexes while the trans to cis iso-
merization was not observed.
(6)
The CO insertion into the Pt–C bond of the cationic
complexes 4 and 5 bearing two trimethylphosphine
(2)
ligands was not observed even under 11 atm of CO at
50°C in acetone, whereas CO insertion into cationic
phenylplatinum complex with a bidentate phosphine
ligand was confirmed [26c]. In this case, retaining the
cis configuration with the bidentate ligand hinders the
isomerization, keeping the site cis to the phenyl ligand
available for the CO ligand to allow the migratory
insertion.
(3)
(4)
In contrast to the cationic monoorganoplatinum sys-
tem that is reluctant to undergo the CO insertion, the

202
Y. Kayaki et al. / Journal of Organometallic Chemistry 622 (2001) 199–209
monomethylpalladium
complex,
trans-[PdCH3-
2.3. Theoretical studies on the reactions of cationic
monomethyl complexes [M(CH3)(PH3)2]+ (M=Pt and
Pd) with CO
(acetone){P(CH3)3}2]+BF4–, undergoes ready CO inser-
tion to form the CO-coordinated acetyl complex, trans-
[Pd{CH3C(O)}(CO){P(CH3)3}2]+BF4–
(7),
under
atmospheric CO at −20°C (Scheme 1). The cationic
solvent-coordinated acetylpalladium complex, trans-
[Pd{CH3C(O)}(acetone){P(CH3)3}2]+BF4–, can be also
prepared from the neutral acetylpalladium complex,
trans-[Pd{CH3C(O)}(Cl){P(CH3)3}2], with one equiva-
lent of AgBF4 without causing the decarbonylation of
the acetyl ligand (Eq. (7)). On the contrary, decarbony-
lation of the neutral acylplatinum complexes, trans-
[Pt{RC(O)}(Cl){P(C6H5)3}2] (R=alkyl and aryl), was
found to be promoted by AgPF6 to give cationic
The above experimental results have shown that the
cationic solvent-coordinated monoorganoplatinum
complexes are inert to CO insertion in contrast to the
corresponding monoorganopalladium complexes that
readily undergo the CO insertion. In order to clarify the
dominant factors influencing the CO insertion processes
and the reason for the different behavior between the
cationic organopalladium and organoplatinum com-
plexes toward CO, we carried out ab initio molecular
orbital calculation with MP2 level taking into account
electron-correlation effects. The calculations were car-
ried out using PH3 for P(CH3)3. We optimized the
structures of the reactants, intermediates, transition
states (TSs), and products. The calculated reaction
paths of the isomerization and CO insertion for the
platinum and palladium complexes are shown in
Scheme 3. In the left part of the Scheme 3 are shown
the cis–trans isomerization courses and in the right part
the CO insertion courses. The optimized bond distances
and bond angles are shown in Figs. 1 and 2. The
potential energy profiles of the isomerization, insertion,
and decarbonylation paths for palladium and platinum
complexes are shown in the Figs. 3 and 4, respectively.
monoorgano(carbonyl)platinum
complexes,
trans-
[PtR(CO){P(C6H5)3}2]+PF6– [36] (Eq. (8)). These results
imply that the cationic acetylplatinum species weakly
bound with a solvent molecule are thermodynamically
unstable relative to the corresponding acetylpalladium
complexes. A cationic benzoylplatinum complex, cis-
[Pt{C6H5C(O)}(CO){P(C6H5)3}2]+BF4–, on the other
hand, was isolated as a stable complex in the decar-
bonylation of an a-keto benzoylplatinum complex [37]
(Eq. (9)).
(7)
2.3.1. Cis–trans isomerization of the cationic
monomethyl complexes
The stable structure of cationic monomethylpalla-
dium complex having the two phosphine ligands takes
the T-shaped trans and cis forms, I-trans and I-cis,
where the I-trans form is more stable than the I-cis
form by 7.2 kcal mol−1 (Fig. 3). Due to the trans
influence of the methyl ligand the bond lengths of the
Pd–P and Pt–P bonds (trans to the methyl) in I-cis
are longer than those of the Pd–P and Pt–P in I-trans.
(8)
(9)
Scheme 3.

Y. Kayaki et al. / Journal of Organometallic Chemistry 622 (2001) 199–209
203
,
Fig. 1. Optimized geometries (in A and °) of intermediates and
transition states for CO insertion into [PdCH3(PH3)2]+
.
,
Fig. 2. Optimized geometries (in A and °) of intermediates and
The cis complex isomerizes to the trans one with swing-
ing of the PH3 ligand via the Y-shaped transition states.
The energy of the transition state TS1 having the Y
shape is 2.7 kcal mol−1 higher than that of the I-cis for
the palladium complex. Although a quite similar profile
is found for the platinum analog, we can find the
difference that the energy barrier for the cis to trans
isomerization is smaller for the platinum complex than
for the palladium analog and the cis to trans isomeriza-
tion is more exothermic (Fig. 4).
transition states for CO insertion into [PtCH3(PH3)2]+
.
It should be noted that for the migratory CO inser-
tion into I-trans to take place the reverse process, i.e.
isomerization from the I-trans to I-cis is required as the
initial step. The barriers for the isomerization from the
I-trans to I-cis are calculated as 14.4 and 9.9 kcal
mol−1 for the platinum and palladium complexes, re-
spectively. The higher barrier for the isomerization in
the methylplatinum complex than that for the corre-
sponding methylpalladium complex partly accounts for
the poor reactivity of the solvent-coordinated cationic
organoplatinum complexes toward CO insertion.
Fig. 3. Potential–energy profile of cis–trans isomerization and CO
insertion into [PdCH3(PH3)2]+, in kcal mol−1, relative to I-trans.
2.3.2. CO insertion into the metal–carbon bond 6ia
four-coordinate intermediates
Stabilization of the T-shaped complexes is observed
in coordination of the CO ligand with the cationic
Fig. 4. Potential–energy profile of cis–trans isomerization and CO
insertion into [PtCH3(PH3)2]+, in kcal mol−1, relative to I-trans.

204
Y. Kayaki et al. / Journal of Organometallic Chemistry 622 (2001) 199–209
complexes to give II-cis and II-trans species both for
the palladium and platinum analogs. The CO insertion
into the Pd–CH3 bond takes the path as shown in
Scheme 3 from II-cis where the methyl and CO ligands
are situated in mutually adjacent positions.
imental studies on CO insertion into [Pd(CH3)2L2] and
[Pd(C2H5)2L2] complexes (L=monodentate tertiary
phosphine ligands) where the product distributions are
compatible with the alkyl migration processes [42].
Theoretical study by Sakaki et al. [43] on the CO
insertion into [Pt(CH3)F(CO)(PH3)] was also compat-
ible with the alkyl migration scheme.
The Pd–CO bond in II-cis is considerably shorter
,
than in the II-trans by 0.531 A (Fig. 1). The short
Pd–CO distance due to the smaller trans effect of the
phosphine ligand relative to the methyl ligand may
facilitate the process to approach the transition state. In
the insertion transition state TS2 the methyl group is
closer to the carbonyl group with elongation of the
Pd–CH3 bond from 2.026 to 2.181 with forming CH3–
The disagreement of the present result with the previ-
ous studies may reflect the difference in the complexes
employed. The present study deals with the cationic
monomethyl complexes having two phosphine ligands
whereas the CO insertion into the neutral dialkylpalla-
dium complexes having one phosphine ligand was stud-
ied in the previous work. In the neutral dialkyl
complexes the presence of the alkyl ligand trans to the
other alkyl ligand (modeled by H ligand) may serve to
labilize the alkyl–palladium bonds by higher trans ef-
fect. The presence of the two phosphine ligands may
also influence the course of the insertion process.
Comparison of the energy diagrams between the
palladium and the platinum systems (Figs. 3 and 4)
shows that the CO insertion steps from II-cis to III-cis
are slightly endothermic for both palladium and plat-
inum systems. The activation barrier of 20.4 kcal
mol−1 from the platinum II-cis to TS2 is much higher
than 6.6 kcal mol−1 calculated for the corresponding
palladium complex. The significantly higher activation
barrier for the CO insertion into the methylplatinum
complex than into the methylpalladium complex ac-
counts for the poor reactivity of the cationic platinum
complexes.
,
,
CO bond of 1.887 A that is longer than 1.494 A in the
product III-cis. The increase in the AP–Pd–CO and
BP–Pd–CH3 angles of 102.3 and 103.5° in TS2 indi-
cates that both the methyl and the CO ligands move
toward each other for the C–C bond formation. In the
subsequently formed immediate insertion product III-
cis, substantial interaction between the palladium and
the oxygen atom in the acetyl group can be seen. The
Pd–C–O angle of 88.5° and the Pd–O distance of
,
2.196 A are close to the previous calculated values of
Pd–C–O angle of 88.7° and the Pd–O distance of 2.30
2
,
A for the h type interaction in the acetyl complex
having ethanediimine ligand [38]. Notably, the P–Pd–P
angle of 103.6° in III-cis is greater than the value of
about 84° calculated for CO insertion in a system with
PH2CHꢀCHPH2 ligand [39]. The wider P–Pd–P angle
may be taken as an indication that the CO insertion
may be coupled with the phosphine migration to the
trans form III-trans. A similar feature in the P–Pt–P
angle was also observed in propene insertion in a
cationic hydridoplatinum system having monodentate
phosphine ligands [40]. From experimental results of an
acceleration of CO insertion into cationic methylpalla-
dium complexes in the order of dppeꢀdppfBdppp#
2.3.3. CO insertion into the metal–carbon bond 6ia
fi6e-coordinate intermediates
Following the examination of a CO insertion course
through a four-coordinate intermediate as discussed
above we also examined a path through a five-coordi-
nate intermediate. In a kinetic study concerning the
carbonylation Garrou and Heck proposed a mechanism
through a five-coordinate intermediate [44]. Involve-
ment of five-coordinate species was also suggested in
carbonylation with bidentate [45] and tridentate [46]
ligands. However, examination of coordination course
of CO to II-trans or II-cis gave no local minima corre-
sponding to [PdCH3(CO)2(PH3)2]+.
Interactive hybrid fragment orbitals of [PdCH3-
(CO)(PH3)2]+ and CO to form trigonal bipyramidal
structures (a) and (b) and a square planar fragment
with CO to form a square pyramidal complex (c) in
Chart do not show significant overlap between these
fragments. A direct insertion of CO into palladium–
carbon bond without coordination to the palladium
center [47] as depicted in (d) is also found to be
electronically inadequate. It can be concluded that the
migratory insertion mechanism via four-coordinate spe-
cies in cis form is more favorable than the other possi-
dppb
(dppf=1,1%-bis(diphenylphosphino)ferrocene;
dppb=
dppp=1,3-bis(diphenylphosphino)propane;
1,4-bis(diphenylphosphino)butane) it was argued that a
large P–Pd–P angle will push the methyl group toward
the coordinated CO ligand, possibly lowering the acti-
vation barrier for the ligand migration [12].
Although the comparison of the initial II-cis and the
product III-cis indicates that the insertion proceeds as
overall CO migration, the conformation of TS2 may be
regarded as the transition state showing a concerted
migration of alkyl and CO ligands. The III-cis complex
was found isomerized readily to III-trans and is further
stabilized by coordination of CO to give the CO-coor-
dinated acetyl complex IV. The present result is differ-
ent from those in the previous theoretical studies
dealing with the CO insertion into the Pd–CH3 and
Pt–CH3 bonds in [M(CH3)(H)(CO)(PH3)] modeling the
CO insertion into [Pd(CH3)2(PH3)2] with MP2 calcula-
tions by Koga and Morokuma [41] and from the exper-

Y. Kayaki et al. / Journal of Organometallic Chemistry 622 (2001) 199–209
205
Table 1
generated by replacement of one of the two PH3 ligands
by CO. Energy barriers for the CO insertion in two
possible geometries with the methyl and the carbonyl
ligands in mutually cis positions have been computed as
shown in the second and third rows in Table 1. The
energy barriers for CO insertions into the M–CH3
bonds for the neutral palladium and platinum com-
plexes were found greater than those in the correspond-
ing cationic systems in agreement with the experimental
results where the cationic complexes were found to be
more susceptible to CO insertion. The higher energy
barriers for CO insertion in neutral platinum systems
than in the corresponding palladium systems were also
computed.
Energy barriers for the CO insertion into the M–CH3 bond in
cationic and neutral complexes (M=Pd, Pt) (in kcal mol−1
)
2.3.5. E6aluation of relati6istic effect on the CO
insertion into Pt–CH3 bond
The difference in the energetics of the palladium and
platinum complexes for the CO insertion can be ex-
plained by taking relativistic effect into account as
discussed by Koga and Morokuma [41]. We investi-
gated the relativistic effect on the CO insertion into the
cationic platinum complex. For this purpose, the rela-
tivistic scheme by eliminating small-components of the
Dirac equation (RESC) theory [48] was employed. The
RESC method is one of quasi-relativistic all-electron
methods, and it is known that this method yields a
satisfactory result for valence properties of heavy ele-
ment compounds [48]. As the usual effective core po-
tentials (ECPs) have already included the relativistic
effect, it is difficult to compare the difference between
relativistic and non-relativistic results directly. In order
to discuss the relativistic and the non-relativistic results
all-electron calculation was used in the RESC approach
under the HF approximation. The MINI basis set was
employed for platinum [49] and the optimized struc-
tures of TS2 shown in Fig. 2 were used.
Table 2
Non-relativistic (NR) and relativistic (Rel.) results of activation en-
ergy and heat of formation in the CO insertion into the Pt complex
(in kcal mol−1
)
Reactant
TS
Product
NR
Rel.
0.0
0.0
20.6
33.2
−1.5
20.9
ble routes involving the five-coordinate transition struc-
tures for the reaction of the cationic complex with CO.
The contribution of the five-coordinate intermediates
for the CO insertion is considered negligible in contrast
to the system bearing nitrogen donor ligands [47,48].
Table 2 shows the activation energy and the heat of
formation in the CO migratory insertion from the
platinum complex II-cis into III-cis with the non-rela-
tivistic and relativistic results. By taking the relativistic
effect into account the activation energy increases by
12.6 kcal mol−1 and the reaction becomes 20.9 kcal
mol−1 endothermic. The result clearly shows that the
relativistic effect acts adversely to the CO insertion into
the Pt–CH3 bond to take place than into the Pd–CH3
bond. The theoretical result is in agreement with the
experimental observation. The influence of the relativis-
tic effect conspicuously appears in the transition state
of this reaction. Fig. 5 shows the difference between the
relativistic and non-relativistic electron densities in the
transition state. The density around the platinum atom
is changed spherically by the relativistic effect due to
the well-known contraction of the s atomic orbital with
the relativity. Furthermore, the density difference map
Chart 1.
2.3.4. Comparison of energy barriers for CO insertion
into the Pd–CH3 or Pt–CH3 bond in cationic systems
with those in neutral systems
The results of calculation of the energy barriers (TS2)
for the CO insertion into the cationic and neutral
methyl–palladium and platinum complexes (II-cis) are
summarized in Table 1. The results show that the
activation barriers for the CO insertion into the Pt–
CH3 bond in the cationic complexes are higher by 13.8
kcal mol−1 than into the Pd–CH3 bond.
As another possible route for the CO insertion into
monomethylpalladium and platinum complexes
[MCH3(Cl)(PH3)2] (M=Pd, Pt) remains the course in-
volving square planar complexes [MCH3(Cl)(CO)(PH3)]

206
Y. Kayaki et al. / Journal of Organometallic Chemistry 622 (2001) 199–209
shows that it becomes unfavorable to form the C–C
bonding with the relativity. By considering the relativis-
tic effect the density of the platinum side of the methyl
carbon increases, whereas the density of the methyl side
of the carbonyl carbon decreases. Since the relativistic
effect is much larger for platinum complexes than for
the palladium complexes, all the metal–carbon bonds
in the platinum complexes should be stronger than in
the palladium complexes. Cleavage of the strong plat-
inum–carbon bond in the methyl(carbonyl)platinum
complex in the CO insertion process will require a
larger activation energy.
insertion into the Pt–CH3 bond is significantly higher
than that into the Pd–CH3 bond in agreement with the
experimental results. The feasibility of CO insertion
routes involving five-coordinate intermediates having
two CO ligands has been theoretically explored but was
refuted as unlikely. The route involving replacement of
one of the two phosphine ligands by CO in the neutral
system, [MCH3(Cl)(PH3)2] (M=Pd, Pt), was revealed
to require higher activation barriers than the route
through the cationic system, [MCH3(CO)(PH3)2]+. The
lower reactivity of the methylplatinum complex than
the methylpalladium complex for CO insertion was
clarified in terms of greater relativistic effect in the
platinum complex with stronger Pt–CH3 bond and less
favorable transition state than in the methylpalladium
complex.
3. Conclusion
The overall feature of experimental results on cis–
trans isomerization and CO insertion into methyl–pal-
ladium and methyl–platinum complexes having two
monotertiary phosphine ligands is in agreement with
the results of theoretical studies. For trans-methylpalla-
dium and methyplatinum complexes having two ter-
tiary phosphine ligands the trans to cis isomerization is
required to bring the methyl and CO ligands in adja-
cent positions suitable for CO insertion to take place.
The theoretical study reveals that the trans to cis iso-
merization barrier for the palladium complex is lower
than that for the platinum analog. The theoretical study
also indicates that the activation barrier for the CO
4. Experimental
4.1. General procedures
All the manipulations were performed under argon
atmosphere by using Schlenk techniques. The neutral
monomethyl complexes, cis-[PtCH3(Cl){P(CH3)3}2] (2a)
and trans-[PtCH3(Cl){P(CH3)3}2] (3a), were synthesized
by the literature method [34]. The diphenyl complex,
cis-[Pt(C6H5)2{P(CH3)3}2] (1b) was prepared from cis-
[PtCl2{P(CH3)3}2] with C6H5Li (two equivalents) ac-
cording to the literature method [18a]. All the reagents
were used as received from commercial suppliers. Sol-
1
vents were dried, distilled, and stored under argon. H-
(270 MHz, referenced to SiMe4 via residual solvent
protons), 13C{1H}- (67.9 MHz, referenced to SiMe4 via
the solvent resonance), and 31P{1H} (109.4 MHz, refer-
enced to 85% H3PO4 as an external standard) NMR
were recorded on a JEOL EX-270 spectrometer. Cou-
pling constants (J values) are given in hertz (Hz), and
spin multiplicities are indicated as follows: d (doublet),
t (triplet), m (multiplet), vt (virtual triplet), and br
(broad). IR spectra were measured by a Hitachi I-2000
spectrometer. Elemental analyses were carried out using
a Yanako MT-3.
4.2. Preparation of cis- and
trans-phenyl(chloro)palladium complexes;
cis-[PtC6H5(Cl){P(CH3)3}2] and
trans-[PtC6H5(Cl){P(CH3)3}2] (2b and 3b)
The neutral monophenylplatinum complexes were
prepared using the general procedure described by
Chatt [16]. Thus, an ethereal solution of dry HCl (6.0
ml, 2.48 mmol) was added to the approximately equiva-
lent amount of cis-[Pd(C6H5)2{P(CH3)3}2] (1.216 g, 2.43
mmol) in ether at −78°C. Stirring the reaction mixture
for 30 min gave a colorless solution. After evaporation
Fig. 5. Difference between the relativistic and non-relativistic densities
in the transition states of the CO insertion into the methylplatinum
complex (TS2). Solid lines indicate higher density of the relativistic
result than that of the non-relativistic one. Lower density is repre-
sented by dashed lines.

Y. Kayaki et al. / Journal of Organometallic Chemistry 622 (2001) 199–209
207
of the solvent at −30°C, the residue was crystallized
from acetone–petroleum ether to give the cis complex
2b in 71% yield. The trans complex 3b was obtained in
90% yield by a similar procedure under a reflux
condition.
4.3.4. trans-[PtC6H5{(CD3)2CO}{P(CH3)3}2]+BF–4 (5b)
1H-NMR (acetone-d6, −30°C) l 6.78–7.89 (5H,
PtC6H5, m), 1.23 (18H, P(CH3)3, vt, JPH=4.0 Hz,
3JPtH=32.1 Hz); 31P{1H}-NMR (acetone-d6, −30°C) l
−12.7 (1JPtP=2717 Hz).
4.2.1. cis-[PtC6H5(Cl){P(CH3)3}2] (2b)
4.4. Preparation of the cationic
monoorgano(carbonyl)platinum complexes (6)
1H-NMR (acetone-d6, −30°C) l 6.61–7.39 (5H,
PtC6H5, m), 1.55 (9H, P(CH3)3, d, 2JPH=8.8 Hz,
3JPtH=16.1 Hz), 1.43 (9H, P(CH3)3, d, JPH=11.0 Hz,
2
3JPtH=46.2 Hz); 31P{1H}-NMR (acetone-d6, −30°C) l
4.4.1. trans-[PtCH3(CO){P(CH3)3}2]+BF–4 (6a)
1
19.2 (2JPP=14.6 Hz, JPtP=1634 Hz), −27.3 (2JPP
=
A solution of AgBF4 (271 mg, 1.39 mmol) in acetone
(5 ml) was added dropwise to an acetone (10 ml)
solution of trans-[PtCH3(Cl){P(CH3)3}2] (558 mg, 1.40
mmol) at −40°C. After the white precipitate immedi-
ately formed was removed by filtration, CO was bub-
bled into the colorless solution at −40°C. A small
amount of Pt black was removed by filtration. The
solution was evaporated to one-fourth of its original
volume under reduced pressure. Ether was added into
the concentrate until a white precipitate was formed.
The solvent was decanted and the residue was dried in
1
14.6 Hz, JPtP=4117 Hz); IR (KBr disc) wPt–P=408
cm−1, wP–C=950, 966 cm−1; Anal. Found: C, 31.47;
H, 5.11. Calc. for C12H23ClP2Pt: C, 31.35; H, 5.04%.
4.2.2. trans-[PtC6H5(Cl){P(CH3)3}2] (3b)
1H-NMR (acetone-d6, −30°C) l 6.85–7.85 (5H,
PtC6H5, m), 1.24 (18H, P(CH3)3, vt, JPH=4.0 Hz,
3JPtH=30.8 Hz); 31P{1H}-NMR (acetone–d6, −30°C)
l –13.2 (1JPtP=2738 Hz); IR (KBr disc) wPt–P=501
cm−1, wP–C=956 cm−1; Anal. Found: C, 31.17; H,
5.06. Calc. for C12H23ClP2Pt: C, 31.35; H, 5.04%.
1
vacuo to give 451 mg (67%) of white powder. H-NMR
(acetone-d6, −30°C) l 1.87 (18H, P(CH3)3, vt, JPH
=
4.3. Generation of the cationic complexes (4 and 5) in
NMR tube
3
3
4.2 Hz, JPtH=31.5 Hz), 0.68 (3H, PtCH3, t, JPH=8.8
Hz, 2JPtH=64.1 Hz); 31P{1H}-NMR (acetone-d6, −
30°C) l −13.6 (1JPtP=2299 Hz); 13C{1H}-NMR (ace-
tone-d6, −30°C) l 179.4 (PtCO, br), 13.9 (P(CH3)3, vt,
To a solution containing complex 1 in acetone-d6
(0.35 ml) was added a solution of one equivalent of
silver salts in acetone-d6 (0.10 ml) at −78°C in an
NMR tube. The NMR tube was shaken ten times
(white suspension of silver halide was formed immedi-
ately) and was kept until the upper layer became clear
before the NMR spectrum was measured.
J
PC=10.1 Hz, 2JPtC=35.1 Hz), −2.0 (PtCH3, t, JPC
=
7.3 Hz); IR (KBr disc) wC–O=2092 cm−1
.
4.4.2. trans-[PtC6H5(CO){P(CH3)3}2]+BF–4 (6b)
The phenyl complex can be obtained in 70% yield by
1
the similar procedure. H-NMR (acetone-d6, −30°C) l
4.3.1. cis-[PtCH3{(CD3)2CO}{P(CH3)3}2]+BF–4 (4a)
1H-NMR (acetone-d6, −50°C) l 1.76 (9H, P(CH3)3,
7.07–7.70 (5H, PtC6H5, m), 1.65 (18H, P(CH3)3, vt,
J
PH=4.3 Hz, 3JPtH=32.3 Hz); 31P{1H}-NMR (ace-
2
3
tone-d6, −30°C) l −14.7 (1JPtP=2321 Hz); IR (KBr
d, JPH=11.9 Hz, JPtH=53.6 Hz), 1.52 (9H, P(CH3)3,
2
3
disc) wC–O=2108 cm−1
.
d, JPH=9.0 Hz, JPtH=16.9 Hz), 0.35 (3H, PtCH3,
dd, 3JPH=2.2 Hz, 3JPH=7.3 Hz, 2JPtH=46.8 Hz);
The rapid formation of the carbonyl complexes 6a
and 6b from the cationic complexes 4 and 5 was
monitored by NMR. CO gas was bubbled into the
acetone-d6 (0.45 ml) solution of each cationic platinum
complex generated at −78°C in NMR tube as men-
tioned above.
31P{1H}-NMR (acetone-d6, −50°C) l −13.2 (2JPP
=
14.7 Hz, 1JPtP=1687 Hz), −25.2(2JPP=14.7 Hz,
1JPtP=4140 Hz).
4.3.2. cis-[PtC6H5{(CD3)2CO}{P(CH3)3}2]+BF–4 (4b)
1H-NMR (acetone-d6, −50°C) l 6.61–7.39 (5H,
PtC6H5, m), 1.58 (9H, P(CH3)3, d, 2JPH=11.7 Hz,
2
3JPtH=16.2 Hz), 1.56 (9H, P(CH3)3, d, JPH=9.2 Hz,
4.5. Computational details
3JPtH=53.5 Hz); 31P{1H}-NMR (acetone-d6, −50°C) l
1
19.2 (2JPP=14.7 Hz, JPtP=1632 Hz), −27.3 (2JPP
=
All geometries and transition states were optimized
with the second-order Møller–Plesset perturbation
(MP2) method in the GAUSSIAN-94 program package
[50]. We have employed the standard Huzinaga–Dun-
ning basis set [51] augmented by the polarization func-
tion for carbon and oxygen atoms, and the Hay–Wadt
effective core potential (ECP) [52] are used for palla-
dium, platinum, and phosphorus atoms.
1
14.7 Hz, JPtP=4132 Hz).
4.3.3. trans-[PtCH3{(CD3)2CO}{P(CH3)3}2]+BF–4 (5a)
1H-NMR (acetone-d6, −30°C) l 1.43 (18H, P(CH3)3,
3
vt, JPH=3.6 Hz, JPtH=30.1 Hz), 0.66 (3H, PtCH3, t,
3JPH=7.7 Hz, 2JPtH=89.4 Hz); 31P{1H}-NMR (ace-
tone-d6, −30°C) l −6.7 (1JPtP=2717 Hz).

208
Y. Kayaki et al. / Journal of Organometallic Chemistry 622 (2001) 199–209
Chem. Res. 17 (1984) 67. (d) G.K. Anderson, H.C. Clark, J.A.
Acknowledgements
Davies, Organometallics 1(1982) 64. (e) G.K. Anderson, G.J.
Lumetta, Organometallics 4 (1985) 1542.
[25] P.J. Stang, Z. Zhong, A.M. Arif, Organometallics 11 (1992)
1017.
This work was supported by a grant-in-aid for Scien-
tific Research Priority Areas Research no. 05236106
from the Ministry of Education, Science, Sports and
Culture. We wish to express our gratitude to Nippon
Zeon Co. Ltd., for funding.
[26] (a) T.G. Appleton, M.A. Bennett, Inorg. Chem. 17 (1978) 738.
(b) L. Mole, J.L. Spencer, N. Carr, A.G. Orpen, Organometallics
10 (1991) 49. (c) P.W.N.M. van Leeuwen, C.F. Roobeek, H. van
der Heijden, J. Am. Chem. Soc. 116 (1994) 12117
[27] (a) F. Basolo, J. Chatt, H.B. Gray, R.G. Pearson, B.L. Shaw, J.
Chem. Soc. A (1961) 2207. (b) H.C. Clark, J.D. Ruddick, Inorg.
Chem. 9 (1970) 1226. (c) H.C. Clark, L.E. Manzer, Inorg. Chem.
10 (1971) 2699. (d) M.H. Chisholm, H.C. Clark, J. Am. Chem.
Soc. 94 (1972) 1532. (e) D.P. Arnold, M.A. Bennett, J.
Organomet. Chem. 199 (1980) 119. (f) D.P. Arnold, M.A. Ben-
nett, J. Organomet. Chem. 202 (1980) 107. (g) D.P. Arnold,
M.A. Bennett, Inorg. Chem. 23 (1984) 2110. (h) R.A. Michelin,
M. Mozzon, B. Vialetto, R. Bertani, G. Bandoli, R.J. Angelici,
Organometallics 17 (1998) 1220.
References
[1] J. Tsuji, Palladium Reagents and Catalysts Innovations in Or-
ganic Synthesis, Wiley, Chichester, 1995.
[2] R.F. Heck, Palladium Reagents in Organic Syntheses, Academic
Press, New York, 1985.
[3] M. Beller, in: B. Cornils, W.A. Herrmann (Eds.), Applied Ho-
mogeneous Catalysis with Organometallic Compounds, vol. 1,
VCH, Weinheim, 1996, p. 148.
[4] A. Yamamoto, J. Chem. Soc. Dalton Trans. (1999) 1027 and
refs. cited therein.
[28] (a) R. Romeo, M.R. Plutino, L.I. Elding, Inorg. Chem. 36 (1997)
5909. (b) G. Alibrandi, D. Minniti, L.M. Scolaro, R. Romeo,
Inorg. Chem. 27 (1988) 318.
[29] R. Romeo, G. Alibrandi, L.M. Scolaro, Inorg. Chem. 32 (1993)
4688.
[5] G.T. Crisp, Chem. Soc. Rev. 27 (1998) 427.
[6] (a) G.J.P. Britovsek, V.C. Gibson, D.F. Wass, Angew. Chem.
Int. Ed. Engl. 38 (1999) 428. (b) S.D. Ittel, L.K. Johnson, M.
Brookhart, Chem. Rev. 100 (2000) 1169.
[30] K. Siegmann, P.S. Pregosin, L.M. Venanzi, Organometallics 8
(1989) 2659.
[7] E. Drent, P.H.M. Budzelaar, Chem. Rev. 96 (1996) 663.
[8] (a) A. Yamamoto, J. Organomet. Chem. 500 (1995) 337. (b) F.
Ozawa, T. Hayashi, H. Koide, A. Yamamoto, J. Chem. Soc.
Chem. Commun. (1991) 1469. (c) F. Kawataka, Y. Kayaki, I.
Shimizu, A. Yamamoto, Organometallics 13 (1994) 3517. (d) F.
Kawataka, I. Shimizu, A. Yamamoto, Bull. Chem. Soc. Jpn. 68
(1995) 654. (e) Y. Kayaki, I. Shimizu, A. Yamamoto, Bull.
Chem. Soc. Jpn. 70 (1997) 917. (f) A. Yamamoto, Y. Kayaki, K.
Nagayama, I. Shimizu, Synlett (2000) 925.
[9] C.M. Killian, D.J. Tempel, L.K. Johnson, M. Brookhart, J. Am.
Chem. Soc. 118 (1996) 11664.
[10] F.C. Rix, M. Brookhart, P.S. White, J. Am. Chem. Soc. 118
(1996) 4746.
[11] J.S. Brumbaugh, R.R. Whittle, M. Parvez, A. Sen, Organometal-
lics 9 (1990) 1735.
[12] G.P.C.M. Dekker, C.J. Elsevier, K. Vrieze, P.W.N.M. van
Leeuwen, Organometallics 11 (1992) 1598.
[13] Y. Kayaki, F. Kawataka, I. Shimizu, A. Yamamoto, Chem.
Lett. (1994) 2171.
[14] Y. Kayaki, I. Shimizu, A. Yamamoto, Bull. Chem. Soc. Jpn. 70
(1997) 1141.
[15] H. Urata, M. Tanaka, T. Fuchikami, Chem. Lett. (1987) 751.
[16] A.L. Casado, P. Espinet, Organometallics 17 (1998) 954.
[17] (a) J. Chatt, B.L. Shaw, J. Chem. Soc. (1959) 705. (b) J. Chatt,
B.L. Shaw, J. Chem. Soc. (1959) 4020.
[18] F.H. Allen, A. Pidcock, J. Chem. Soc. A (1968) 2700.
[19] (a) R. Romeo, D. Minniti, S. Lanza, P. Uguagliati, U. Belluco,
Inorg. Chem. 17 (1978) 2813. (b) G. Alibrandi, D. Minniti, R.
Romeo, G. Cum, R. Gallo, J. Organomet. Chem. 291 (1985)
133. (c) G. Alibrandi, M. Cusumano, D. Minniti, L.M. Scolaro,
R. Romeo, Inorg. Chem. 28 (1989) 342. (d) G. Alibrandi, L.M.
Scolaro, D. Minniti, R. Romeo, Inorg. Chem. 29 (1990) 3467.
[20] A.L. Rieger, G.B. Carpenter, P.H. Rieger, Organometallics 12
(1993) 842.
[31] I. To´th, T. Ke´gl, C.J. Elsevier, L. Kolla´r, Inorg. Chem. 33 (1994)
5708.
[32] R. Romeo, D. Minniti, S. Lanza, Inorg. Chim. Acta 18 (1976)
L15.
[33] (a) G. Faraone, V. Ricevuto, R. Romeo, M. Trozzi, J. Chem.
Soc. A (1971) 1877. (b) R. Romeo, D. Minniti, M. Trozzi, Inorg.
Chim. Acta 14 (1975) L15. (c) R. Romeo, D. Minniti, M. Trozzi,
Inorg. Chem. 15 (1976) 1134. (d) R. Romeo, Inorg. Chem. 17
(1978) 2040. (e) R. Romeo, D. Minniti, S. Lanza, Inorg. Chem.
18 (1979) 2362. (f) R. Romeo, D. Minniti, S. Lanza, Inorg.
Chem. 19 (1980) 3663. (g) M.J. Blandamer, J. Burgess, R.
Romeo, Inorg. Chim. Acta 65 (1982) L179. (h) M.J. Blandamer,
J. Burgess, D. Minniti, R. Romeo, Inorg. Chim. Acta 96 (1985)
129. (i) G. Alibrandi, L.M. Scolaro, R. Romeo, Inorg. Chem. 30
(1991) 4007.
[34] (a) R. van Eldik, D.A. Palmer, H. Kelm, Inorg. Chem. 18 (1979)
572. (b) H. Kelm, W.J. Louw, D.A. Palmer, Inorg. Chem. 19
(1980) 843. (c) W.J. Louw, R. van Eldik, H. Kelm, Inorg. Chem.
19 (1980) 2878. (d) R. van Eldik, D.A. Palmer, H. Kelm, W.J.
Louw, Inorg. Chem. 19 (1980) 3551.
[35] D.M. Adams, J. Chatt, B.L. Shaw, J. Chem. Soc. (1960) 2047.
[36] M. Kubota, R.K. Rothrock, J. Geibel, J. Chem. Soc. Dalton
Trans. (1973) 1267.
[37] T.-M. Huang, J.-T. Chen, G.-H. Lee, Y. Wang, Organometallics
10 (1991) 175.
[38] P. Margl, T. Ziegler, Organometallics 15 (1996) 5519.
[39] P. Margl, T. Ziegler, J. Am. Chem. Soc. 118 (1996) 7337.
[40] S. Creve, H. Oevering, B. Coussens, Organometallics 18 (1999)
1967.
[41] N. Koga, K. Morokuma, J. Am. Chem. Soc. 108 (1986) 6136.
[42] F. Ozawa, A. Yamamoto, Chem. Lett. (1981) 289.
[43] S. Sakaki, K. Kitaura, K. Morokuma, K. Ohkubo, J. Am.
Chem. Soc. 105 (1983) 2280.
[44] P.E. Garrou, R.F. Heck, J. Am. Chem. Soc. 98 (1976) 4115.
[45] (a) K. Frankcombe, K.J. Cavell, R. Knott, B. Yates, J. Chem.
Soc. Chem. Commun. (1996) 781. (b) K.E. Frankcombe, K.J.
Cavell, B.F. Yates, R.B. Knott, Organometallics 16 (1997) 3199.
(c) M.J. Green, G.J.P. Britovsek, K.J. Cavell, F. Gerhards, B.F.
Yates, K.E. Frankcombe, B.W. Skelton, A.H. White, J. Chem.
Soc. Dalton Trans. (1998) 1137.
[21] A. Crispini, K.N. Harrison, A.G. Orpen, P.G. Pringle, J.R.
Wheatcroft, J. Chem. Soc. Dalton Trans. (1996) 1069.
[22] R. Romeo, Comments Inorg. Chem. 11 (1990) 21.
[23] R.W. Glyde, R.J. Mawby, Inorg. Chem. 10 (1971) 854.
[24] (a) G.K. Anderson, R.J. Cross, J. Chem. Soc. Dalton Trans.
(1979) 1246. (b) G.K. Anderson, R.J. Cross, J. Chem. Soc.
Dalton Trans. (1980) 1434. (c) G.K. Anderson, R.J. Cross, Acc.

Y. Kayaki et al. / Journal of Organometallic Chemistry 622 (2001) 199–209
209
[46] (a) B.A. Markies, P. Wijkens, J. Boersma, A.L. Spek, G. van
Koten, Rec. Trav. Chim. Pays-Bas 110 (1991) 113. (b) B.A.
Markies, P. Wijkens, A. Dedieu, J. Boersma, A.L. Spek, G. van
Koten, Organometallics 14 (1995) 5628. (c) J.H. Groen, A. de
Zwart, M.J.M. Vlaar, J.M. Ernsting, P.W.M.N. van Leeuwen,
K. Vrieze, H. Kooijman, W.J.J. Smeets, A.L. Spek, P.H.M.
Budzelaar, Q. Xiang, R.P. Thummel, Eur. J. Inorg. Chem.
(1998) 1129.
[47] M.R.A. Blomberg, C.A.M. Karlsson, P.E.M. Siegbahn, J. Phys.
Chem. 97 (1993) 9341.
[48] (a) T. Nakajima, K. Hirao, Chem. Phys. Lett. 302 (1999) 383. (b)
T. Nakajima, T. Suzumura, K. Hirao, Chem. Phys. Lett. 304
(1999) 271. (c) T. Suzumura, T. Nakajima, K. Hirao, Int. J.
Quantum Chem. Symp. 75 (1999) 757.
Y. Sakai, H. Tatewaki, Gaussian Basis Sets for Molecular
Calculations, Elsevier, Amsterdam, 1984.
[50] M.J. Frisch, G.W. Trucks, H.B. Schlegel, P.M.W. Gill, B.G.
Johnson, M.A. Robb, J.R. Cheeseman, T. Keith, G.A. Pe-
tersson, J.A. Montgomery, K. Raghavachari, M.A. Al-Laham,
V.G. Zakrzewski, J.V. Ortiz, J.B. Foresman, J. Cioslowski, B.B.
Stefanov, A. Nanayakkara, M. Challacombe, C.Y. Peng, P.Y.
Ayala, W. Chen, M.W. Wong, J.L. Andres, E.S. Replogle, R.
Gomperts, R.L. Martin, D.J. Fox, J.S. Binkley, D.J. Defrees, J.
Baker, J.P. Stewart, M. Head-Gordon, C. Gonzalez, J.A. Pople,
GAUSSIAN-94, Gaussian Inc., Pittsburgh PA, 1995.
[51] T.H. Dunning, P.J. Hay, in: H. Schaefer (Ed.), Methods
of Electronic Structure Theory, Plenum, New York, 1977,
p. 1.
[49] S. Huzinaga, J. Andzelm, M. Klobukowski, E. Radzio-Andzelm,
[52] P.J. Hay, W.R. Wadt, J. Chem. Phys. 82 (1985) 270.
.