Kinetics and mechanism of the nucleophilic substitution of tellurium(II) dialkanethiolates, Te(SR1)2 with thiols, HSR2

Article abstract of DOI:10.1080/10426500590906337

The equilibrium reaction between tellurium(II) dithiolates and thiols, Te(SR¹)₂ + 2 HSR² ? Te(SR ²)₂ + 2 HSR¹ was studied by means of ¹H- and ¹²⁵Te NMR spectroscopy and ab initio quantum chemical methods. It was found that the reaction is catalyzed by Bronsted acids and bases, the catalytic activity corresponding to the strength of the respective acid or base. Investigation of the initial step of the reaction, Te(SR¹)₂ + HSR² ? Te(SR ¹)(SR²) + HSR¹, showed it to proceed according to first order kinetics for Te(SR¹)₂, HSR² and for the catalyst. Ab initio geometry optimizations and frequency calculations suggest [Te(SR¹)(HSR¹)(HSR²)]⁺ and [Te(SR¹)₂(SR²)]^- to be stable intermediates and not transition states in the acid and base catalyzed reactions, respectively. The reaction hence proceeds via an additional elimination rather than an S_N2 mechanism. The catalytic activity displayed by acids and bases can be applied to reduce the temperature in synthesis of thermally labile tellurium(II) dithiolates. Copyright Taylor & Francis Inc.

Full text of DOI:10.1080/10426500590906337

Phosphorus, Sulfur, and Silicon, 180:815–825, 2005
Copyright © Taylor & Francis Inc.
ISSN: 1042-6507 print / 1563-5325 online
DOI: 10.1080/10426500590906337
Kinetics and Mechanism of the Nucleophilic Substitution
of Tellurium(II) Dialkanethiolates, Te(SR1)2 with Thiols,
HSR2
Holger Fleischer
Institut fu¨r Anorganische Chemie und Analytische Chemie, Johannes
Gutenberg Universita¨t Mainz, Mainz, Germany
The equilibrium reaction between tellurium(II) dithiolates and thiols, Te(SR1)2
+
2
2 HSR2
Te(SR ) + 2 HSR1 was studied by means of 1H- and 125Te NMR spec-
ꢀ
ꢁ
2
troscopy and ab initio quantum chemical methods. It was found that the reaction
is catalyzed by Brønsted acids and bases, the catalytic activity corresponding to the
strength of the respective acid or base. Investigation of the initial step of the reac-
1
2
1
tion, Te(SR1)2 + HSR2
Te(SR )(SR ) + HSR , showed it to proceed according
ꢀ
ꢁ
to first order kinetics for Te(SR1)2, HSR2 and for the catalyst. Ab initio geometry
optimizations and frequency calculations suggest [Te(SR1)(HSR1)(HSR2)]+ and
[Te(SR1)2(SR2)]− to be stable intermediates and not transition states in the acid
and base catalyzed reactions, respectively. The reaction hence proceeds via an addi-
tional elimination rather than an SN2 mechanism. The catalytic activity displayed
by acids and bases can be applied to reduce the temperature in synthesis of ther-
mally labile tellurium(II) dithiolates.
Keywords Ab initio; tellurium; 125Te NMR; thiolate
INTRODUCTION
Nucleophilic substitution is a fundamental reaction and played a cen-
tral role in the development of modern physical organic chemistry. In-
terest in extending this strategy to heteroatoms is growing.1 There are
indications that the mechanism is different between first- and second-
row atoms. For example, a classic SN2 mechanism is evident for nu-
cleophilic substitution at carbon,2 nitrogen,3−5 and oxygen,6,7 but an
addition-elimination pathway occurs for reactions at phosphorus8,9 and
is implied for reactions at silicon.10
In the present work, kinetics and mechanism of the exchange reac-
tion of tellurium(II) dithiolates, Te(SR1)2, with thiols, HSR2, according
Received January 20, 2004; accepted October 12, 2004.
Address correspondence to Holger Fleischer, Institut fu¨r Anorganische Chemie und
Analytische Chemie, Johannes Gutenberg Universita¨t Mainz, Duesbergweg 10-14, Mainz
D-55099, Germany. E-mail: fleische@mail.zdv.uni-mainz.de
815

816
H. Fleischer
to (1) (R1, R2 = alkyl, alkenyl)
2
1
Te(SR1)2 + 2 HSR2
Te(SR ) + 2 HSR
(1)
ꢀ
ꢁ
2
was studied by means of NMR-spectroscopy and ab initio methods.
Eq. (1) was recently found to be a straightforward approach towards
novel tellurium(II) dithiolates, e.g. Te(SC6H4(o-NH2))2.11 This route
proved to be an efficient alternative to the already known method of
reductive elimination according to (Eqs. (2) and (3)):12,13
TeO2 + 4 HSR → Te(SR)2 + RSSR + 2 H2O
(2)
(3)
Te(OiPr)4 + 4 HSR → Te(SR)2 + RSSR + 4 HOiPr
In both reactions, only half of the thiol is converted to the tellurium(II)
dithiolate, while the other half is used to form the disulfide. Forma-
tion of the latter is a disadvantage, if the tellurium(II) dithiolate and
the disulfide are difficult to separate from each other or if the thiol
is rather valuable. In that respect, Te(SMe)2 and Te(StBu)2 are suit-
able reactants for preparative reactions according to Eq. (1). The thi-
ols, HSMe and HStBu, generated from those tellurium(II) dithiolates,
are very volatile compounds and the equilibrium according to Eq. (1)
can be shifted to the right side by distilling the thiols from reaction
mixture.
Equations (1)–(3) are not only of synthetic use, but also have physi-
ological importance. TeIV compounds react with the cysteine residue in
glutathione peroxidase by reductive elimination which leads to a deac-
tivation of the latter. It was reported that reactivation could be achieved
by addition of cysteine,14 and it is highly probable that the re-activation
is based on an exchange reaction analogous to Eq. (1).
In the present work, kinetics of Eq. (1) and appropriate mechanisms
are presented and compared to analogous reactions at two-valent sulfur
atoms.15,16
RESULTS AND DISCUSSION
Chemical Equilibria
For a variety of combinations of Te(SR1)2 and HSR2, dissolved in C6D6,
the equilibrium according to Eq. (1), was investigated by means of 1H-
and 125Te NMR spectra. Table I gives the equilibrium constants for
several equilibria according to (1), with
c[Te(SR2)2] · c[HSR1]2
K(1) =
c[Te(SR1)2] · c[HSR2]2

Te(SR1)2 with HSR2
817
TABLE I Data for Equilibrium According to Eq. (1) for
Several Combinations of Tellurium(II) Dithiolates,
Te(SR1)2 and Thiols, HSR2,a
R1
R2
c0[Te(SR1)2]a
c0(HSR2)a
K(1)
ꢂG298
Me
Me
Me
Me
tBu
tBu
Et
Allyl
Ph
0.41
0.59
0.35
0.18
0.22
0.88
0.91
0.37
0.24
0.62
2.1
1.8
2.0
2.0
1.0
−1.8
−1.5
−1.7
−1.7
0
Et
aSubstochiometric amounts of NMI or pyridine were used to
catalyze the exchange reactions.
bInitial concentrations, c0 are molalities, given in mol kg−1
.
An attempt to study equilibrium and kinetics of the reaction of
Te(SMe)2 with HSC6F5 failed, since the tellurium(II) dithiolates rapidly
were decomposed to elemental tellurium and disulfides.
The data suggests a slight shift of the equilibrium towards the right
side, if R2 = Me is replaced by R2 = Allyl, Et, tBu, or Ph. Uncertainties
of the equilibrium concentrations, due to errors in their determination
from NMR intensities, are rather large. It can nevertheless be stated
that none of the investigated ligands shifts the equilibrium entirely to
one side. From a purely statistical distribution of SR1 and SR2 to TeII
and H, an equilibrium constant K(1) = 1.0 would result. The experi-
mentally obtained constants are close to these values, and it is hence
inferred that for equilibrium Eq. (1) ꢂH298 ≈ 0.
Kinetic and Catalysis of Ligand Exchange
Equation (1) has to proceed in two steps
1
2
1
Te(SR1)2 + HSR2
Te(SR )(SR ) + HSR
(1a)
(1b)
ꢀ
ꢁ
2
2
1
Te(SR1)(SR2)2 + HSR2
Te(SR ) + HSR .
ꢀ
ꢁ
2
Compared to the ligand exchange between Te(OR1)4 and HOR2,17,18
reactions according to Eqs. (1a) and (1b) proceed very slowly. It took
about two weeks for a solution of Te(SiPr)2 and HStBu in CDCl3 to
reach the equilibrium state.13 It was observed that Eq. (1a) proceeds
quicker than Eq. (1b) for the exchange of Te(SMe)2 and HStBu, a fact
attributed to the steric requirement of the tert-butyl group.
The rates of reaction of Eqs. (1a) and (1b) are tremendously enhanced
by Brønsted acids and bases. The change of HSMe concentration with
time was monitored for the reaction of Te(SMe)2 with HStBu in benzene

818
H. Fleischer
FIGURE 1 Catalytic activity of Brønsted acids and bases regarding equilib-
rium (1). (a): (+) 0.25 equivalents of NMI, ( ) 0.25 equivalents of pyridine, ( )
no base. (b) ( ) 0.5 equivalents of CF COOH, ( ) 8 equivalents of CH COOH.
ꢀ
•
3
3
All equivalents relative to Te(SMe)2.
solutions that contained acids and bases of different strengths. Results
are depicted in Figure 1.
Regarding the pKa and pKb values of the Brønsted acids and bases,
respectively (pKa(HOAc) = 4.75,19 pKa(CF3COOH) = 0.23,20 pKb(py) =
8.87,21 pKb(NMI) = 6.88,22 all in aqueous solution at 25◦C), their cat-
alytic activity with respect to Eq. (1) corresponds to their acid and base

Te(SR1)2 with HSR2
819
strength. Dimethylsulfoxide, which is a good Lewis base (donor num-
ber DN = 30 according to Gutmann,23 cf. pyridin, DN = 33), but a week
Brønsted base, does not act as a catalyst for this reaction. By plotting
the initial reaction rates against the initial concentrations of the respec-
tive reactant and keeping initial concentrations of the other reactants
constant, the reaction orders of the single species were obtained. For
the pyridine catalyzed exchange reaction of Te(SMe)2 with HStBu, or-
ders were Te(SMe)2: 0.86, HStBu: 1.05, and pyridine: 0.94. According
to these data, first order kinetics for each of the reactants is assumed,
and hence Eq. (1a) is a second order reaction.
Based on the kinetic data, mechanisms for the acid and base cat-
alyzed exchange reactions can be developed. Since the 3 J(H,H)-coupling
HSCH3, formed in (1) for R1 = Me, can be observed even in the presence
of pyridine of NMI, fast deprotonation of the thiol and protonation of
a thiolate anion can be excluded. If such an acid base reaction would
operate, the 3 J(H,H)-coupling could not be observed due to exchange de-
coupling. The equilibrium constant for the acid base reaction of HSMe
(pKs = 10.524) with NMI in aqueous solution is estimated as 4·10−5
.
In benzene, the solvent in which exchange reactions were investigated,
should be even smaller since SMe− and [NMIH]+ ions are less stabi-
lized. It is thus inferred that the thiol coordinates to the tellurium(II)
dithiolate before it interacts with the base. The acidity of the thiol is
increased by complexation of the tellurium(II) dithiolate, similar to the
enhanced acidity of H2O that coordinates to metal ions like Zn2+. This
hypothesis is corroborated by ab initio calculations (see further down).
The proposed mechanisms for the base and acid catalyzed thiolate ex-
change reactions are depicted in Scheme 1. The base catalyzed mecha-
nism consists of the following steps (B = Brønsted base), Eq. (4a)–(4f):
1
2
Te(SR1)2 + HSR2
[Te(SR ) (HSR )]
(4a)
(4b)
(4c)
(4d)
(4e)
(4f)
ꢀ
ꢁ
2
1
2
1
2
ꢀ
ꢁ
[Te(SR ) (HSR )] + B
[Te(SR ) (B · ·HSR )]
2
2
1
2
+
1
2
−
ꢀ
ꢁ
[Te(SR ) (B · · HSR )]
[BH] + [Te(SR ) (SR )]
2
2
1
2
1
[BH]+ + [Te(SR1)2(SR2)]−
[Te(SR )(SR )(B · · HSR )]
ꢀ
ꢁ
1
2
1
1
2
1
ꢀ
ꢁ
[Te(SR )(SR )(B · · HSR )]
[Te(SR )(SR )(HSR )] + B
1
2
1
1
2
1
ꢀ
ꢁ
[Te(SR )(SR )(HSR )]
Te(SR )(SR ) + HSR .
In experimental studies on the exchange reaction of disulfides, RSSR,
with thiols, HSR (R = CH2CH2NH2), in aqueous solution it was found
that the thiolate ion, RS− acts as nucleophile.15

820
H. Fleischer
SCHEME 1 Proposed mechanisms for the nucleophilic substitution reaction
of Te(SR1)2 with HSR2 catalyzed by (a) a Brønsted base, B and (b) a Brønsted
acid, HA.

Te(SR1)2 with HSR2
821
It is assumed that the acid-catalyzed mechanism also starts with
Eq. (4a) and ends with Eq. (4f), but Eqs. (4b) to (4e) are replaced by
Eqs. (5b) to (5e) (HA = Brønsted acid):
1
2
1
1
2
+
ꢀ
[Te(SR ) (HSR )] + HA
[Te(SR )(AH · · SR )(HSR )] (5b)
ꢁ
2
1
1
2
1
2
1
+
−
ꢀ
ꢁ
[Te(SR )(AH · ·SR )(HSR )]
[Te(SR )(HSR )(HSR )] + A (5c)
[Te(SR1)(HSR2)(HSR1)]+ + A−
[Te(SR )(HSR )(AH · · SR )] (5d)
1
1
2
ꢀ
ꢁ
1
1
2
1
1
2
ꢀ
ꢁ
[Te(SR )(HSR )(AH · · SR )]
[Te(SR )(HSR )(SR )] + HA. (5e)
Ab Initio-Studies
The mechanisms of the acid and base catalyzed identity reaction with
R1 = R2 = Me were investigated by ab initio geometry optimizations
and thermochemical calculations. Since in this case Eqs. (4d)–(4f), (5d),
and (5e) are exactly the reverse reactions of (4c), (4b), (4a), (5c), and
(5b), respectively. The latter had not to be considered. Thermochemi-
cal data are given in Table II, MP2/LANL2DZP optimized structures
of reaction intermediates [Te(SMe)2(HSMe)] (I), [Te(SMe)3]− (II), and
Te[(SMe)(HSMe)2]+ (III) are depicted in Figure 2. All three structures
represent true minima on the potential energy surface, at the HF level
as well as at the MP2 level, i.e. reactions proceed via an additional
elimination mechanism instead of an SN2 mechanism. This contrasts
to recent investigations on the substitution reaction (Eq. (6))
R1S− + R2SSR3 → R1S− + R2SSR3
TABLE II Free Standard Enthalpies, ∆G298, of Eqs.
(4a), (4b), (4c), (5b), and (5c) (R1 = R2 = Me, Values
are Given in kJ·mol–1). Different Bases, B, and
Different Acids, HA, Were Used (See Footnotes.)
Equation
ꢂG298
ꢂG298
(4a)
+20
(4b)a
(4b)b
(4c)a
(4c)b
+14 (5b)c
+14 (5b)d
+476 (5c)c
+460 (5c)d
+36
+23
+563
+457
aB = pyridine.
bB = imidazole (imidazol (pKb = 6.97) was used as model
compound for N-methylimidazol.
c
HA = CH3COOH,
d
HA = CF3COOH.

822
H. Fleischer
FIGURE 2 Ab-initio-optimized molecular structures of [Te(SMe)2(HSMe)] (a),
[Te(SMe)(HSMe)2]+ (b), and [Te(SMe)3]− (c). Internuclear distances in pm, bond
angles in ◦.
where an SN2 mechanism was found at the HF level and an additional
elimination mechanism at all levels including electron correlation.16
Formation of [Te(SMe)2(HSMe)] from Te(SMe)2 and HSMe was cal-
culated to be an endergonic process, i.e., the adduct only is formed
to a small fraction. Deprotonation Eq. (4c) and protonation Eq. (5c)
lead to charge separation and hence strongly are endergonic processes
(see Table II) for isolated molecules and molecular ions. Equations (4c)
and (5c), respectively, are decisive for the rates of base and acid cat-
alyzed ligand exchange reactions, as can be seen from the data given
in Table II. The thermochemical data was calculated for gas phase re-
actions but reflects nevertheless the experimentally observed trends.
Especially Eq. (4c) proceeds less endergonic for imidazole than for pyri-
dine and Eq. (5c) proceeds less endergonic for trifluoro acetic acid than
for acetic acid. If a linear-free enthalpy relation is assumed between
the free enthalpy of activation of Eqs. (4c) and (5c) (kinetic data), re-
spectively, and the associated free enthalpy of reaction (thermodynamic
data), the higher reaction rates with the stronger acids and bases can
be rationalized from the data given in Table II. Formation of a hydrogen

Te(SR1)2 with HSR2
823
bridge between pyridine and [Te(SMe)2(HSMe)] according to Eq. (4b)
proceeds with ꢂG298 = +14 kJ·mol−1 and is thus thermodynamically
favored compared to a hydrogen bridge between pyridine and HSMe
(+17 kJ·mol−1). This corroborates the suggestion that coordination to
tellurium increases the acidity of HSR.
Parameters from ab-initio-optimized molecular structures show an
increased Te1–S2 and a decreased Te1–S4 distance on protonation as
well as deprotonation of [Te(SMe)2(HSMe)] (see Figure 2). Both struc-
tural changes lead to a transition of the reactant towards the products.
np(S4)-σ∗(Te1–S2) energy of interaction (NBO basis) was estimated
as 26 kJ·mol−1 for [Te(SMe)2(HSMe)], 619 kJ·mol−1 for [Te(SMe)3]−,
and 15 kJ·mol−1 for [Te(SMe)(HSMe)2]+. In the latter case, an ad-
ditional np(S4)-d(Te1) orbital interaction with 35 kJ·mol−1 strength-
ens the bond between the nucleophile and the Te atom. (According to
ab initio calculations, [Te(SR)3]− anions are stable towards dissociation
to Te(SR)2 and RS−. Attempts to isolated those anions with suitable
counter cations are currently undertaken in this lab). The Te1–S3 bond
in [Te(SMe)2(HSMe)] only is slightly influenced by protonation or de-
protonation. It is a bit shorter in the cation and longer in the anion,
compared to the neutral complex.
CONCLUSION
The ligand exchange at the Te atom in Te(SR1)2 by reaction with HSR2
proceeds via an SN2-mechanism. The reaction is catalyzed by Brønsted
acids and bases. The former destabilize the Te S bond to be cleaved
while the latter increase the nucleophily of the attacking thiol. The
catalytic activity of acids and bases allows the exchange of Eq. (1) to
proceed at lower temperatures and hence enables thermally labile tel-
lurium(II) dithiolates to be synthesized via this route.
EXPERIMENTAL PART
General Procedures
25
13
Te(SMe)2 and Te(StBu)2 were prepared according to literature
procedure, thiols and C6D6 were used as purchased. NMR: spectra
were recorded on a Bruker DRX 400, B (1H) = 400.0, B (125Te) =
1
1
126.387 MHz. Standard: TMS (1H) and Te(CH3)2 (125Te).
Kinetic Measurements
The amounts of reactants and solvent (C6D6) used in a given experi-
ment were weighted and the time of the experiment was started when

824
H. Fleischer
the compounds were mixed with each other. The starting employed con-
centrations varied between 0.01 and 0.5 mol L−1. The concentrations
of reactants and products during the course of the reaction were ob-
tained from the initial concentrations and the intensities of the signals
in the 1H NMR spectra. The orders of Eq. (1a) with respect to HSR2 and
Te(SR1)2 were determined by using an excess of the other reactant, such
that its concentration was kept nearly constant throughout the exper-
iment (method of pseudo first order kinetic). Employing then different
amounts of HSR2 or Te(SR1)2, the orders of Eq. (1a) with respect to the
compounds were determined from the ratio of the initial reaction rates
(differential method).26 In a similar manner, the orders of Eq. (1a) with
respect to the catalytically active compounds acetic acid and pyridine
were determined.
Theoretical Methods
Quantum chemical ab initio investigations were performed with the
GAUSSIAN 94 program package.27 Optimization of molecular geome-
tries, calculation of vibrational frequencies and an analysis of the elec-
tronic structure in terms of natural orbitals28 were performed at the
MP2 level, with an effective core double zeta valence basis set,29 aug-
mented by appropriate polarization functions for Te, S (with exponents
according to Ho¨llwarth et al.),30 and C (exponent 0.75).
REFERENCES
[1] P. Beak, Acc. Chem. Res., 25, 215–222 (1992).
[2] S. Gronert, J. Am. Chem. Soc., 113, 6041–6048 (1991).
[3] P. Beak and J. Li, J. Am. Chem. Soc., 113, 2796 (1991).
[4] M. Bu¨hl and H. F. Schaefer III, J. Am. Chem. Soc., 115, 9143–9147 (1993).
[5] M. N. Glukhovtsev, A. Pross, and L. Radom, J. Am. Chem. Soc., 117, 9012–9018
(1995).
[6] P. Beak and D. Loo, J. Am. Chem. Soc., 108, 3834 (1986).
[7] S. M. Bachrach, J. Org. Chem., 55, 1016–1019 (1990).
[8] J. Li and P. Beak, J. Am. Chem. Soc., 114, 9206–9207 (1992).
[9] S. M. Bachrach and D. C. Mulhearn, J. Phys. Chem., 97, 12229–12231 (1993).
[10] R. Damrauer, C. H. DePuy, and V. M. Bierbaum, Organometallics, 1, 1553–1554
(1982).
[11] H. Fleischer and D. Schollmeyer, Inorg. Chem., 41, 4739–4747 (2002).
[12] W. Mazurek, A. G. Moritz, and M. J. O’Connor, Inorg. Chim. Acta, 113, 143–146
(1986).
[13] H. Fleischer, S. Stauf, and D. Schollmeyer, Inorg. Chem., 38, 3725–3729 (1999).
[14] A. Albeck, H. Weitman, B. Sredni, and M. Albeck, Inorg. Chem., 37, 1704–1712
(1998).
[15] J. C. Pleasants, W. Guo, and D. L. Rabenstein, J. Am. Chem. Soc., 111, 6553–6558
(1989).

Te(SR1)2 with HSR2
825
[16] S. M. Bachrach and D. C. Mulhearn, J. Phys. Chem., 100, 3535–3540 (1996).
[17] H. E. Gottlieb, S. Hoz, I. Elyashiv, and M. Albeck, Inorg. Chem., 33, 808–811 (1994).
[18] H. Fleischer, Y. Dienes, and D. Schollmeyer, Eur. J. Inorg. Chem., 2073–2081 (2002).
[19] A. D. Headley, S. D. Starnes, L. Y. Wilson, and G. R. Famini, J. Org. Chem., 59,
8040–8046 (1994).
[20] L. A. Reinhardt, K. A. Sacksteder, and W. W. Cleland, J. Am. Chem. Soc., 120,
13366–13369 (1998).
[21] M. S. K. Niazi and J. Mollin, Bull. Chem. Soc. Jpn., 60, 2605–2610 (1987).
[22] C. S. Cassidy, L. A. Reinhardt, W. W. Cleland, and P. A. Frey, J. Chem. Soc. Perkin
Trans. 2, 3, 635–642 (1999).
[23] V. Gutmann, Coord. Chem. Rev., 18, 225–255 (1976).
[24] A. P. Munro, D. L. H. Williams, J. Chem. Soc. Perkin Trans. 2, 9, 1794–1797 (2000).
[25] H. Fleischer, N. W. Mitzel, D. Schollmeyer, Eur. J. Inorg. Chem., 815–821 (2003).
[26] K. J. Laidler, Chemical Kinetics, (2nd ed.), McGraw-Hill, New York, St. Louis.
[27] 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. Petersson, 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, and J.
A. Pople, Gaussian 94, Revision E.1, Gaussian, Inc., Pittsburgh PA, 1995.
[28] A. E. Reed, L. A. Curtiss, and F. Weinhold, Chem. Rev., 88, 899–926 (1988).
[29] W. R. Wadt and P. J. Hay, J. Chem. Phys., 82, 284–298 (1985).
[30] A. Ho¨llwarth, M. Bo¨hme, S. Dapprich, A. W. Ehlers, A. Gobbi, V. Jonas, K. F. Ko¨hler,
R. Stegmann, A. Veldkamp, and G. Frenking, Chem. Phys. Lett., 208, 237–240 (1993).