Modulation of the C-C distance in disubstituted 1,2-R2-o-carboranes. Crystal structure of closo 1,2-(SPh)2-1,2-C2B10H10

Article abstract of DOI:10.1016/S0022-328X(02)01637-6

1,2-C_c-substituted o-carboranes, where C_c represents the cluster carbon atoms, display a range of experimental/computed C_c?C_c distances from 1.629/1.624 ? for the unsubstituted parent compound 1,2-C₂B₁₀H₁₂ (1) to 1.798/1.818 ? for 1,2-(SPh)₂-1,2-C₂B₁₀H₁₀ (5)¹ or 1.858(5)/1.826(5) for elucidated by single crystal X-ray diffraction. Purely alkyl substituents alter the C_c?C_c distance of the parent compound very little, which implies that steric effects, although relevant, are not the major cause of the lengthening. In contrast, substituents with lone pairs alter the C_c?C_c distance substantially. Our calculations suggest that the cause of the elongation is the transfer of electron density from the available lone pairs on the substituents to the Ψ* low-lying virtual orbitals mainly distributed around C_c, producing a decrease in the C_c?C_c bond order and, thereby, an increase in the C_c?C_c distance. A Bader analysis shows that the electron density at the bond-critical point, which is found at the mid-point of the C_c?C_c distance, decreases considerably with the presence of lone pairs of the sulfur atoms bound to the C_c atoms of the carborane cage.

Full text of DOI:10.1016/S0022-328X(02)01637-6

Journal of Organometallic Chemistry 657 (2002) 232ꢂ238
/
www.elsevier.com/locate/jorganchem
Modulation of the CÃ
/
C distance in disubstituted l,2-R2-o-carboranes.
Crystal structure of closo l,2-(SPh)2-l,2-C2B10H10
Jordi Llop a, Clara Vinas a,*, Josep M. Oliva a, Francesc Teixidor a, Miquel
˜
Angel Flores b, Raikko Kivekas c, Reijo Sillanpaa d
¨ ¨
a Institut de Cie`ncia de Materials de Barcelona, CSIC, Campus de Bellaterra, Cerdanyola, 08193 Barcelona, Spain
b Metallocens Department, REPSOL, Iꢀ
D, Embajadores 183, 28045 Madrid, Spain
/
c Inorganic Chemistry Laboratory, Box 55, University of Helsinki, FIN-00014 Helsinki, Finland
d Department of Chemistry, University of Turku, FIN-20014 Turku, Finland
Received 30 July 2001; accepted 6 June 2002
Abstract
1,2-Cc-substituted o-carboranes, where Cc represents the cluster carbon atoms, display a range of experimental/computed CcÁ Á ÁCc
˚
˚
distances from 1.629/1.624 A for the unsubstituted parent compound 1,2-C2B10H12 (1) to 1.798/1.818 A for l,2-(SPh)2-l,2-C2B10H10
(5)1 or 1.858(5)/1.826(5) for l,2-m-SCH2(CH2OCH2)2CH2S-1,2-C2B10H10 (7). Different CcÁ Á ÁCc distances can be achieved by
modifying the substituents on the carbon of the cluster compound. The crystal structure of a new disubstituted o-carborane 1,2-
(SPh)2-l,2-C2B10H10 (5) was elucidated by single crystal X-ray diffraction. Purely alkyl substituents alter the CcÁ Á ÁCc distance of the
parent compound very little, which implies that steric effects, although relevant, are not the major cause of the lengthening. In
contrast, substituents with lone pairs alter the CcÁ Á ÁCc distance substantially. Our calculations suggest that the cause of the
elongation is the transfer of electron density from the available lone pairs on the substituents to the C* low-lying virtual orbitals
mainly distributed around Cc, producing a decrease in the CcÁ Á ÁCc bond order and, thereby, an increase in the CcÁ Á ÁCc distance. A
Bader analysis shows that the electron density at the bond-critical point, which is found at the mid-point of the CcÁ Á ÁCc distance,
decreases considerably with the presence of lone pairs of the sulfur atoms bound to the Cc atoms of the carborane cage. # 2002
Elsevier Science B.V. All rights reserved.
Keywords: Bond-critical point; Bond modulation, closo-carboranes
1. Introduction
isolated cases, the possibility to tune up the CÃ
/C
distance in conventional organic compounds is rare.
Cluster boron chemistry, in particular the o-carboranes,
The single, double, and triple bond distances are well
defined in conventional organic compounds although
important exceptions exist. Relevant examples are
cationic norbornenyl derivatives, on which extensive
theoretical studies have been conducted to evaluate the
existence of nonclassical bonding. As an example, in
(C5Me5)2Sm(O2C7Me5), in which the seven carbon
provides the possibility to modulate the CÃ/C distance in
an almost continuous way within the same family of
compounds. The CcÁ Á ÁCc distance in metallacarborane
cluster derivatives of [7,8-C2B9H11]2ꢁ can be regarded
as highly plastic. Thus, examples of CcÁ Á ÁCc elongation
have been found with different metal ions, such as
Ru(II), [3]; Rh(I), [4]; Rh(III), [5]; Mo (IV), [6] or
Co(III) [7]. In conventional cluster terminology, species
which experience a substantial CcÁ Á ÁCc elongation are
named pseudocloso [8]. Although steric effects of bulky
substituents on Cc atoms have been described to be
responsible for CcÁ Á ÁCc elongation [9], other data from
less bulky substituents [10] suggest that an electronic
contribution may also be relevant. In a recent paper
atoms in O2C7Me5 adopt a norbornadiene structure,
˚
C2 distance equals 1.876 A much longer than
[1] the C7Ã
/
˚
two times the C single bond radii (1.53 A) [2]. Besides
* Corresponding author
E-mail address: teixidor@icmab.es (C. Vinas).
˜
1
A longer CcÁ Á ÁCc computed distance has been found for 1,2-
(SCH3)2-1,2-C2B10H10 (6) but no experimental data are available.
0022-328X/02/$ - see front matter # 2002 Elsevier Science B.V. All rights reserved.
PII: S 0 0 2 2 - 3 2 8 X ( 0 2 ) 0 1 6 3 7 - 6

J. Llop et al. / Journal of Organometallic Chemistry 657 (2002) 232ꢂ
/238
233
concerning mixed sandwich pyrrolylꢂ
/
dicarbollide cobalt
Table 1
Crystallographic data and structure refinement details for compound 5
complexes we have shown that electronic effects appear
to be fundamental in CcÁ Á ÁCc modulation.
Chemical formula
Formula weight
Temperature (K)
C14H20B10S2
360.52
The closo character of mixed sandwich pyrrolylꢂ
/
dicarbollide complexes and the particularly equal
˚
294(2)
0.71069
˚
Wavelength (A)
CcÁ Á ÁCc distance, 1.64 A, in both the parent closo [3-
Co(h5-NC4H4)-l,2-C2B9H11] and o-carborane, led us to
hypothesize that the same explanation concerning
CcÁ Á ÁCc modulation was valid for [3-Co(h5-NC4H4)-
l,2-C2B9H11] and o-carborane derivatives. To this aim,
we decided to study the electronic influence of sub-
stituents on Cc atoms on the CcÁ Á ÁCc distance in the
latter. As with the mixed cobalt complexes, structural
data from X-ray diffraction analysis were compared
with calculated distances obtained with theoretical
methods based on density functional theory. The good
agreement between experimental and calculated CcÁ Á ÁCc
distances allowed us to explain the reasons of CcÁ Á ÁCc
elongation in terms of electronic effects based on
theoretical calculations.
Crystal system
Space group
Monoclinic
C2/c
˚
a (A)
20.0769(16)
8.7908(13)
11.7143(12)
111.127(7)
1928.5(4)
4
˚
b (A)
˚
c (A)
b (8)
Volume (A )
3
˚
Z
r(g cmꢁ3
)
1.242
0.270
744
Absorption coefficient (mmꢁ1
F(000)
Index ranges
)
05h 523, 05k 510,
ꢁ135l 513
1748/1697 [Rintꢃ0.014]
1425
Reflections collected/unique
Reflections observed
[I ꢀ2s(I)]
Refinement method
Data/restraints/parameters
Goodness-of-fit on F2
Final R indices [I ꢀ2s(I)]
R indices (all data)
Full-matrix least-squares on F2
1697/0/159
1.060
R1ꢃ0.0341, wR2ꢃ 0.0851
R1ꢃ0.0447, wR2ꢃ 0.0900
0.206, ꢁ0.201
2. Results and discussion
ꢁ3
˚
Dr (e A
)
In order to study the electronic influence of exo-
cluster carbon substitutents on the CcÁ Á ÁCc distance, a
series of disubstituted l,2-R2-l,2-C2B10H10 (RꢃH, (1);
/
Table 2
Et, (2); Ph, (3); SH, (4); SPh, (5) and SMe, (6)) o-
carborane derivatives was chosen.
˚
Selected bond lengths (A) and angles (8) for compound 5
Bond lengths
The synthesis of 1,2-(SPh)2-l,2-C2B10H10 was success-
fully achieved by reacting o-carborane, n-butyllithium
and phenyldisulphide in the ratio 1:2:2. Scheme 1 shows
the synthesis reaction. This generated compound was
characterized by elemental analysis and NMR spectro-
scopies and is in agreement with the proposed formula.
Good quality crystals for X-ray diffraction studies
were grown by slow evaporation of a solution of 5 in
SÃC(1)
1.776(2)
1.783(2)
1.799(3)
C(1)ÃB(2)
C(1)ÃB(3)
C(1)ÃB(4)
1.725(2)
1.700(2)
1.692(3)
SÃC(7)
C(1)ÃC(1)a
Bond angles
C(1)ÃSÃC(7)
104.66(8) SÃC(1)ÃB(3)
118.09(5) SÃC(1)ÃB(4)
121.4(1)
112.8(1)
10.6(2)
119.7(1)
125.4(1)
120.2(2)
119.5(2)
95.4(2)
SÃC(1)ÃC(1)a
SÃC(1)ÃB(2)
SÃC(7)ÃC(8)
SÃC(7)ÃC(12)
SÃC(1)ÃB(2)a
SÃC(1)ÃC(1)aÃSa
C(1)ÃSÃC(7)ÃC(8)
C(1)ÃSÃC(7)ÃC(12)
CH2Cl2ꢂC6H14 solutions. Crystallographic data and
/
ꢁ89.5(2)
B(2)aÃC(1)ÃSÃC(7)
160.6(1)
selected bond parameters for 5 are listed in Tables 1
and 2. Fig. 1 shows a drawing of the molecule with the
labeling system in. The structure is formed of discrete
molecules having crystallographic C2 symmetry with the
aEquivalent position ꢁxꢀ1, y, ꢁzꢀ1/2.
expected position towards the B(2)a atom, thus decreas-
ing the SÃC(1)Ã C(1)Ã
B(2)a angle and opening the SÃ
C(1)a, SÃ
C(1)ÃB(2) and SÃC(1)ÃB(4) angles. This
operation also increases the distance between the phenyl
group and the carborane cage, and opens the SaÃC(1)aÃ
C(1)ÃS torsion angle to 10.6(2)8.
The SÁ Á ÁS distance of 3.483(1) A is only ca. 0.12 A
shorter than the sum of the corresponding van der
Waals radii, [11] and the free electron pairs of the S
atoms are orientated away from each other to minimize
repulsions.
two fold axis going through the mid-points of C(1)Ã
C(1)a and B(6)ÃB(6)a bonds (aꢃ
xꢀ1, y, ꢁzꢀ1/2).
The SÃC bond lengths are equal within experimental
/
/
/
/
/
/
/
ꢁ
/
/
/
/
/
/
/
/
/
errors and the angles around C(1) vary from 112.8(1) to
125.4(1)8. As a result the S atom is moved away from its
/
/
/
a
˚
˚
Comparison of 5 with l,2-dithia-l,2-dicarba-closo-
dodecaboranes incorporating a macrocyclic fragment
chain between the S atoms, 1,2-(1?,10?-dithia-4?,7?-diox-
Scheme 1. Reaction proposed for the formation of 5.

234
J. Llop et al. / Journal of Organometallic Chemistry 657 (2002) 232ꢂ238
/
from Table 3 are the large CcÁ Á ÁCc distances in thio-Cc-
substituted o-carboranes 4ꢂ6. Moreover, the phenyl
/
group in 5 produces a similar effect on the CcÁ Á ÁCc
distance to that of the simple thiol. Another result to be
discussed is the longer CcÁ Á ÁCc distance in 6 than in 5: a
methyl group bound to sulfur provides a longer CcÁ Á ÁCc
distance than a phenyl group. As we will show below,
not only steric but electronic effects must be taken into
account in order to understand these results. Leaving
aside the packing forces, we consider that three factors
may influence the CcÁ Á ÁCc distance: (i) steric (repulsive)
effects due to substituting one or both of the contiguous
carbon atoms with those containing bulky groups or
those containing lone pairs, (ii) an electronic effect due
to the electronegativity differences between hydrogen
atoms and substituting groups and (iii) a p-electronic
back-donation from the substituting groups into C*
orbitals of the cluster, mostly sited on the CcÁ Á ÁCc
region. Steric crowding would produce repulsive forces
that would tend to move the contiguous carbon atoms
apart; an enhanced electronegativity of the substituent
would deplete electron density between Cc(1) and Cc(2)
and would lengthen the distance, and p-electronic back-
donation to the LUMO (C*) orbital, mostly sited on the
two carbon atoms, would weaken the CcÁ Á ÁCc bond
order. It is then apparent that although influential, the
Fig. 1. ORTEP plot of l,2-(SPh)2-l,2-C2B10H10 (5) and the atomic
numbering scheme. Displacement ellipsoids are shown at 30% prob-
ability levels, except for H atoms which are drawn as small circles of
arbitrary radii.
adecane-1?,10?-diyl)-l,2-dicarba-closo-dodecaborane (7)
[12] and 1,2-(1?,7?-dithia-4?-oxaheptane-1?,7?-diyl)-1,2-
dicarba-closo-dodecaborane (8) [13] reveals some minor
differences. The C(1)Ã
/
C(1)a or CcÁ Á ÁCc distance in 5 is
˚
quite short, 1.799(3) A, being significantly shorter than
those in the two non-equivalent molecules of l,2-(1?,10?-
dithia-4?,7?-dioxadecane-1?,10?-diyl)-l,2-dicarba-closo-
˚
dodecaborane(1.858(5) and 1.826(5) A). The value of the
CcÁ Á ÁCc bond in l,2-(1?,7?-dithia-4?-oxaheptaneane-1?,7?-
˚
diyl)-l,2-dicarba-closo-dodecaborane is 1.816 (6) A and,
steric crowding or lone pairꢂ
/
lone pair repulsive forces is
although longer than in 5, the difference is not as
significant. In 2-methyl-1-sulfido-l,2-dicarba-closo-do-
˚
decaborane anion [14] the distance is 1.792(5) A.
not sufficient to explain the CcÁ Á ÁCc lengthening in 4ꢂ
/
6.
As a consequence, electronic factors should be also
considered, as we pointed out earlier [10] when it was
proposed that the CcÁ Á ÁCc distances could be estimated
Therefore, the contribution of methyl and sulfido
substituents to the lengthening of CcÁ Á ÁCc bond is
comparable to that of two SPh groups.
with the empirical equation: dꢃ
/
d0ꢀ
/
a aiꢀa Si, where
/
d is the CcÁ Á ÁCc distance in the o-carborane derivative,
d0 the CcÁ Á ÁCc distance in the o-carborane, ai the
summed electronic contributions and Si the summed
steric contributions.
The computed and experimental CcÁ Á ÁCc distances of
compounds 1ꢂ6 are displayed in Table 3. As it is shown
/
there, binding of ethyl groups to Cc does not change
significantly the CcÁ Á ÁCc distance. The presence of an
electron-withdrawing and bulkier group, such as phenyl
in 3 does increase this distance more. Theory and
The Ã
suitable probes to determine the major factors influen-
cing the CcÁ Á ÁCc lengthening. In fact, since ÃSPh is
larger than ÃSMe it should facilitate CcÁ Á ÁCc elongation
for reasons of steric crowding. Furthermore, ÃSPh is
/
SMe and Ã/SPh substituents were selected as
/
˚
experiment does deviate about 0.03 A in giving the
/
CcÁ Á ÁCc distance. However, the most striking results
/
Table 3
a
6, and electron density (e Aꢁ3) at the
˚
˚
Experimental (where available), B3LYP/6-31G computed CcÁ Á ÁCc distances (A) in ethane and compounds 1ꢂ
/
BCP at the mid-point between the carbon atoms of the cluster, also computed at the B3LYP/6-31G a level of theory for compounds ethane, 1 and 6
Compound
Experimental CcÃCc distance
Computed CcÃCc distance
r(rBCP)
CH3ÃCH3
1,2-C2B10H12 (1)
1.525 [20]
1.629 [21]
1.530
1.624
1.687
1.764
1.803
1.818
1.836
1.900
1.64
1.26
1,2-Et2-1,2-C2B10H10 (2)
1,2-Ph2-l,2-C2B10H10 (3)
1,2-(SH)2-1,2-C2B10H10 (4)
1,2-(SPh)2-1,2-C2B10H10 (5)
l,2-(SMe)2-l,2-C2B10H10 (6)
1,2-m-SCH2(CH2OCH2)2CH2S-1,2-C2B10H10 (7)
a
1.733(4)/1.720(4) [22]
1.798(3)
0.85
a
1.858(5)/1.826(5) [12]
a
Two molecules in the asymmetric unit.

J. Llop et al. / Journal of Organometallic Chemistry 657 (2002) 232ꢂ
/238
235
also more electron-withdrawing than Ã
/
SMe, therefore,
its stronger capability for electron polarization or
greater electron-depletion in the CcÁ Á ÁCc connectivity
should promote CcÁ Á ÁCc elongation.
If we restrict our discussion to the structural data
obtained by X-ray diffraction we observe that the
thioether species 5 (see Table 3) have a long CcÁ Á ÁCc
distance, almost 20% longer than that of a single CÃ
/C
bond. This raises some doubts about whether a sig-
nificant degree of bonding exists at all. However, as seen
in Table 3, a continuous range of CcÁ Á ÁCc distances is
Fig. 2. HOMO of compound 1 calculated at the B3LYP/6-31G* level
of theory.
˚
observed, ranging from 1.62 to 1.84 A, indicating that a
smooth elongation and not an abrupt break takes place.
This implies that the CcÁ Á ÁCc is a bond of gradually
decreasing strength. As noted above, our interpretation
is that steric crowding, or lone pairꢂlone pair repulsion,
/
although partly responsibles for the elongation, are not
the major factor responsible for the weakening of the
bond. Electronic effects are very relevant, too. We go
on, now, to discuss on electronic effects of the H/H (4),
SPh/SPh (5) and Sme/SMe (6) substituents on the
CcÁ Á ÁCc distance. The Pauling electronegativity of sulfur
(2.5) is identical with that of carbon. However, the
inductive character can be modulated by substituents on
S. The methyl group will make the sulfur atom less
electron-withdrawing than the phenyl group. If electro-
negativity, or electron polarization towards sulfur, was
the reason for the CcÁ Á ÁCc elongation, the SPh/SPh
substituents should produce the longer CcÁ Á ÁCc distance,
due to electron density depletion. This conclusion is in
contradiction with our calculations, since the longer
distance is found with Sme/SMe (6). In earlier work, the
CcÁ Á ÁCc elongation produced by phosphorus in l-PPh2-
2-Me-1,2-C2B10H10 was interpreted in terms of overlap
of the tangential p orbitals in the cluster with p or d-
orbitals on the phosphorus [15]. At that time, we did not
pursue this explanation. Our interpretation of the
CcÁ Á ÁCc distance in the present series of l,2-Cc-substi-
tuted o-carboranes is that p-back donation from an
occupied orbital on sulfur, the lone pair, is partially
transferred to the LUMO (C*) on the cluster with
important orbital contributions for the carbon atoms
Cc. Thus the LUMO antibonding orbital on the carbon
atoms is partially filled, decreasing the CcÁ Á ÁCc bond
order. This hypothesis is supported by our calculations
Fig. 3. Schematic energy level diagram for combination of LUMO of
compound 1 and HOMO of SR groups in 4ꢂ6.
/
antibonding (Fig. 4). This transfer of electronic density
into the C* orbitals on Cc(1) and Cc(2) is one of the
causes of bond lengthening when hydrogen atoms are
replaced by groups with lone pairs. The HOMO is then
the result of mixing one filled orbital with one empty
one, the LUMO, and the global result of this HOMO is
a partial transfer of electron density into the CÃ
/C
antibonding orbital fragment [16]. As mentioned above,
the extent of CcÁ Á ÁCc bonding at the distances exhibited
by the compounds studied in this work is not clear. A
performed on compounds 1ꢂ
of compound (1) calculated at the B3LYP/6-31G* level
of theory. As can be seen, no nodes exist in the SÃCc
/
6. Fig. 2 shows the HOMO
/
connectivity, although there is one between the two Cc’s.
This can be interpreted considering the interaction of
two orbitals coming from different fragments: the
HOMO on the S,S fragment that combines with the
LUMO on the unsubstituted o-carborane (1) to form
two new molecular orbitals (see Fig. 3). One of them, the
occupied one corresponding to the HOMO in 5 has an
Fig. 4. NHOMO of compound 5 calculated at the B3LYP/6-31G*
important original LUMO contribution, that is CÃ
/C
level of theory showing that SÃS orbitals overlap.
/

236
J. Llop et al. / Journal of Organometallic Chemistry 657 (2002) 232ꢂ238
/
Bader analysis [17] shows that there is a bond critical
point (BCP) at the mid-point of CcÁ Á ÁCc in all com-
pounds. At the BCP, the electron density is a maximum
in the plane defined by the two vectors of negative
curvature of the density, and a minimum along the axis
ethane and compounds 1, 2 and 6. According to these
results the density at the BCP is 76 and 52% smaller,
respectively, in compounds 1 and 6 as compared with
ethane. Nevertheless, the density at the BCP is not zero,
and therefore, certain extent of CcÁ Á ÁCc bonding is
predicted, depending on the density at the BCP.
of positive curvature*
/
in this case the CcÁ Á ÁCc axis. Fig.
5 shows a contour map of electron density for com-
pounds 1 and 6 in the plane perpendicular to the CcÁ Á ÁCc
line. The BCP passes through at the mid-point of this
line, where the maximum of the density is depicted as the
core of the contour density circles at the mid-left site of
the contour maps.
3. Conclusion
To summarize, and as it has been shown by the above
results, the CcÁ Á ÁCc distance in 1,2-Cc-substituted o-
carboranes can be modeled as function of the substi-
tuents. Purely alkyl substituents do not substantially
alter the CcÁ Á ÁCc distance of the parent compound,
implying that steric effects, although relevant, are not
the major cause of the lengthening. In contrast, sub-
stituents with lone pairs alter the CcÁ Á ÁCc distance
In order to estimate the ‘degree’ of bonding in these
compounds we chose ethane as a model for a single CÃ
/C
bond, and compounds 1, 2 and 6 for comparative
purposes. Table 3 shows in the last column the electron
density at the BCP sited at the mid-point of CcÁ Á ÁCc in
substantially. This cannot be explained by lone pairꢂ
/
lone pair repulsion or by the effect of electron-polarizing
groups. The more plausible explanation is the transfer of
electron density from the available lone pairs on the
carbon substituents to the C* orbitals, producing a
decrease in the CcÁ Á ÁCc bond order and, thereby, an
increase in the CcÁ Á ÁCc distance.
4. Experimental
4.1. Instrumentation
Elemental analyses were performed using a Carlo
Erba EA1108 microanalyzer. IR spectra were recorded
with KBr pellets on a Nicolet 710 FT spectrophot-
ometer, 1H- and 1H{11B}-NMR (300.13 MHz),
13C{1H}-NMR (75.47 MHz) and11B-NMR (96.29
MHz) spectra were recorded with a Bruker ARX 300
instrument equipped with the appropriate decoupling
accessories. Chemical shift values for 11B-NMR spectra
were referenced to external BF3×
/
OEt2 and those for1H-,
1H{11B}- and 13C{1H}-NMR spectra were referenced to
SiMe4. Chemical shifts are reported in units of parts per
million downfield from tetramethylsilane, and all cou-
pling constants are reported in Hz.
4.2. Materials
Unless otherwise noted, all manipulations were car-
ried out under a dinitrogen atmosphere using standard
vacuum line techniques. Diethyl ether was distilled from
sodium benzophenone prior to use and the rest of the
solvents were of reagent grade quality and were used
without further purification. Commercial o-carborane
was sublimed under high vacuum prior use. A 1.4 M
solution of n-butyllithium in hexane (Fluka), phenylsul-
Fig. 5. Contour map of the electron density for compounds 1 (a) and 6
(b) in the plane perpendicular to the CcÁ Á ÁCc line and containing the
BCP at the midpoint between the two carbon atoms.

J. Llop et al. / Journal of Organometallic Chemistry 657 (2002) 232ꢂ
/238
237
fide, KCl and MgSO4 purchased from commercial
sources and used as received.
4.5. Computational approach
All calculations in this work were performed with the
GAUSSIAN-98 suite of programs [19] using the B3LYP/6-
31G* model, which is based on density functional
theory and uses a double-z basis set, with an additional
set of polarization functions on all atoms except
hydrogen. The molecular geometries were all optimized
at the same level of theory and the energy minima were
checked through frequency calculations in compounds
1, 2, 4 and 6. The molecular geometries of compounds 3
and 6 were optimized starting from the X-ray geometry,
obtaining null forces on nuclei. No frequency calcula-
tions were performed on these molecules due to the size
of the system.
4.3. Synthesis of closo 1,2-(SPh)2-1,2-C2B10H10 (5)
A solution containing diethyl ether (30 ml) and 1,2-
dicarba-closo-dodecaborane (0.500 g, 3.467 mmol) was
cooled to 0 8C during the addition (10 min, dropwise)
of n-buthyllitium (4.9 ml, 6.9 mmol). After stirring for
30 min at 0 8C, the mixture was left to stand at room
temperature (r.t.) for 30 min, then cooled again to 0 8C
before adding an diethyl ether solution (10 ml) of
phenyldisulphide (1.51 g, 3.47 mmol). The ice-bath
was removed and the mixture stirred 30 min at r.t.
The ether mixture was washed twice with a solution of
KCl in water (50 ml). The organic layer was dried
(anhydrous MgSO4, 24 h), and evaporated to yield the
analytically pure white solid, yield 0.95 g (76%).
5. Supporting information available
Crystallographic data for 5 have been deposited with
the Cambridge Crystallographic Data Centre in. cif
format, ref. No. CCDC No 136786.
Recrystalization from dichloromethaneꢂn-hexane (1/1)
gave pale yellow microcrystals. Anal. Calc. for
C14H20B10S2: C, 46.64; H, 5.59; S, 17.79. Found: C,
/
46.45; H, 5.67; S, 17.50%. IR: n (cmꢁ1)ꢃ
/
3066 (y Caryl
H), 1470, 1439,
H). H-NMR (CDC13,
7.41 (m, Haryl), 3.6ꢂ0.5 (br, BÃH).
1H{11B)-NMR (CD2C12, 25 8C): dꢃ
7.76ꢂ7.41 (m,
aryl, 10H), 2.45 (br s, BÃH, 2H), 2.38 (br s, BÃH,
4H), 2.02(br s, BÃH, 2H) and 1.80 (br s, BÃH, 2H).
13C{1H}-NMR (CDCl3, 25 8C): dꢃ
137.4, 131.3, 130.3,
129.4 (Caryl), 91.7 (BÃ
C). 11B-NMR (CDCl3, 25 8C):
dꢃ 147, 2B), ꢁ8.9 (4B), ꢁ9.9 (2B),
3.6 (d, 1J(B, H)ꢃ
11.7 (2B).
Ã
/
H), 2631, 2619, 2590, 2581, 2563 (y BÃ
/
1
746, 725, 705, 685, 496 (d Caryl
Ã
/
Acknowledgements
25 8C): dꢃ7.76ꢂ
/
/
/
/
/
/
This work was supported by CICYT (Project
MAT98-0921), Generalitat de Catalunya (Grant
1998F1 00724), and the Academy of Finland (project
41519, RK).
H
/
/
/
/
/
/
/
ꢁ
/
/
/
/
ꢁ
/
References
[1] W.J. Evans, K.J. Forrestal, J.W. Ziller, J. Am. Chem. Soc. 117
(1995) 12635 (and references therein).
4.4. X-ray crystallographic study
[2] N.W. Alcock, Bonding and Structure, Ellis Horwood, 1990.
[3] (a) Z.G. Lewis, A.J. Welch, J. Organomet. Chem. 430 (1992) C45;
(b) U. Gradler, A.S. Weller, A.J. Welch, D. Reed, J. Chem. Soc.
¨
The unit cell parameters for 1,2-(SPh)2-l,2C2B10H10
were calculated by least-squares refinement of 25 well-
Dalton Trans. (1996) 335;
centered reflections and the data were collected by v ꢂ
/
(c) F. Teixidor, C. Vinas, M.A. Flores, G.M. Rosair, A.J. Welch,
˜
A.S. Weller, Inorg. Chem. 37 (1998) 5394.
2U scan mode on a Rigaku AFC-5S diffractometer.
Intensities of three reflections monitored periodically
during the data collection exhibited no significant
variation. The data were corrected for Lorentz and
polarization effects, and for extinction. Also an empiri-
cal absorption correction was performed (C scans). The
structure was solved and refined according to routine
procedure using the SHELX-97 program package [18].
Approximate non-hydrogen positions were obtained by
direct methods, and after refinement of non-hydrogen
atoms, all hydrogen atom positions were picked from
subsequent Fourier map. In final refinement, the non-
hydrogen atoms were refined with anisotropic and
hydrogen atoms with isotropic displacement para-
meters.
[4] A.J. Welch, in: P. Braunstein, L.A. Oro, P.R. Raithby (Eds.),
Steric Effects in Metallacarboranes in Metal Clusters in Chem-
istry, Wiley-VCH, 1999, p. 69.
[5] F. Teixidor, M.A. Flores, C. Vinas, R. Sillanpaa, R. Kivekas, J.
˜
¨ ¨
Am. Chem. Soc. 122 (2000) 1963.
¨
[6] J.-H. Kim, M. Lamrani, J.-W. Hwang, Y. Do, Chem. Commun.
(1997) 1761.
[7] J. Llop, C. Vinas, F. Teixidor, L. Victori, R. Kivekas, R.
˜
¨
Sillanpaa, Organometallics 19 (2001) 4024.
¨ ¨
[8] M.P. Mingos, M.I. Forsyth, A.J. Welch, Chem. Commun. (1977)
605.
[9] (a) Z.G. Lewis, A.J. Welch, J. Organomet. Chem. 430 (1992) C45;
(b) U. Gradler, A.S. Weller, A.J. Welch, D. Reed, J. Chem. Soc.
¨
Dalton Trans. (1996) 335.
[10] R. Kivekas, R. Sillanpaa, F. Teixidor, C. Vinas, R. Nunez, M.
¨ ¨ ¨
˜ ˜
Abad, Acta Crystallogr. Sect. C 51 (1995) 1864.
[11] A. Bondi, J. Phys. Chem. 68 (1964) 441.

238
J. Llop et al. / Journal of Organometallic Chemistry 657 (2002) 232ꢂ238
/
[12] F. Teixidor, C. Vinas, J. Rius, C. Miravitlles, J. Casabo, Inorg.
˜
[19] M.J. Frisch, G.W. Trucks, H.B. Schlegel, G.E. Scuseria, M.A.
Robb, J.R. Cheeseman, V.G. Zakrzewski, J.A. Montgomery,
R.E. Stratmann, J.C. Burant, S. Dapprich, J.M. Millam, A.D.
Daniels, K.N. Kudin, M.C. Strain, O. Farkas, J. Tomasi, V.
Barone, M. Cossi, R. Cammi, B. Mennucci, C. Pomelli, C.
Adamo, S. Clifford J. Ochterski, G.A. Petersson, P.Y. Ayala, Q.
Cui, K. Morokuma, D.K. Malick, A.D. Rabuck, K. Raghava-
chari, J.B. Foresman, J. Cioslowski, J.V. Ortiz, B.B. Stefenov, G.
Liu, A. Liashenko, P. Piskorz, I. Komaromi, R. Gomperts, R.L.
Martin, D.J. Fox, T. Keith, M.A. Al-Laham, C.Y. Peng, A.
Nanayakkara, C. Gonzalez, M. Challacombe, P.M.W. Gill, B.G.
Johnson, W. Chen, M.W. Wong, J.L. Andre´s, M. Head-Gordon,
E.S. Replogle, J.A. Pople, GAUSSIAN-98, Gaussian, Inc, Pitts-
burgh, PA, 1998.
Chem. 29 (1990) 149.
[13] F. Teixidor, A. Romerosa, J. Rius, C Miravitlles, J. Casabo, C
Vin˜as, E. Sanchez, J. Chem. Soc Dalton Trans. (1990) 525.
[14] R. Kivekas, R. Benakki, C. Vinas, R. Sillanpaa, Acta Crystallogr.
˜
¨ ¨ ¨
Sect. C 55 (1999) 1581.
[15] R. Kivekas, R. Sillanpaa, F. Teixidor, C. Vin˜as, R. Nun˜ez, Acta
¨ ¨
¨
Crystallogr. Sect. C 50 (1994) 2027.
[16] As a reviewer has suggested the NHOMO may well be the non-
bonding orbital (*) shown in Fig. 3. The graphical representation
displayed in Fig. 4 seems to support this possibility. This lowering
in energy with regard to the HOMO can be the result of symmetry
allowed orbital mixing with filled and empty orbitals on the
original cluster. We thank the reviewer for his comments.
[20] CRC Handbook of Chemistry and Physics, 73rd ed., D.R. Lide,
Ed. In Chief, CRC Press Inc, Boca Raton, Florida, 1992.
[21] M.G. Davidson, T.G. Hibbert, J.A.K. Howard, A. Mackinnon,
K. Wade, Chem. Commun. 19 (1996) 2285.
[17] R.F.W. Bader, Atoms in Molecules:
Clarendon Press, Oxford, 1994.
A Quantum Theory,
[18] G.M. Sheldrick, SHELX-97. Program for Crystal Structure Refine-
ment, University of Gottingen, Germany, 1997.
¨
[22] Z.G. Lewis, A.J. Welch, Acta Crystallogr. Sect. C 49 (1993) 705.