Macropolyhedral boron-containing cluster chemistry. A synthetic approach via the auto-fusion of [6,9-(SMe2)2-arachno-B 10H12]

Article abstract of DOI:10.1039/b604295c

In an attempt to build up borane-based multicluster assemblies, thermolysis of [6,9-(SMe₂)₂-arachnoB₁₀H ₁₂] 1 in inert hydrocarbon solution, followed by chromatographic separation, has resulted in the isolation not only of the previously established single-cluster product from this reaction, [5-(SMe ₂)-nido-B₁₀H₁₂] 2 (30%), but also the two two-cluster species [6,9-(SMe₂)₂-arachno-B ₁₀H₁₁-l-(6′-nido-B₁₀H₁₃)] 3 (20%) and [l,6′-(nido-B₁₀H₁₃)₂] 6 (ca. 0.5%) and the two three-cluster species [6,9-(SMe₂) ₂-arachno-B₁₀H₁₀-l,5-(6′-nido-B ₁₀H₁₃)₂] 4 (5%), characterized crystallographically, and [6,9-(SMe₂)₂-arachno-B ₁₀H₁₀-l,3-(6′-nido-B₁₀H ₁₃)₂] 5 (< 1%), identified by NMR spectroscopy. An improved crystallographic investigation of [5-(SMe₂)-nido-B ₁₀H₁₂] 2 is also presented. The feasibility of the stability of species resulting from multiple adjacent substitution of nido-decaboranyl units on the [6,9-(SMe₂)₂-arachno-B ₁₀H₁₂] skeleton is tested by DFT calculations. In an extension, to attempt the use of pre-a-linked two-cluster compounds as starting substrates, two-cluster [5-(SMe₂)-4-(2′-nido-B ₁₀H₁₃)-nido-B₁₀H₁₁] 7 (0.6%) has been isolated from the reaction of SMe₂ with [1,5′-(nido-B ₁₀H₁₃)₂], other identified products being compound 1 (39%) and compound 3 (10.5%). The Royal Society of Chemistry 2006.

Full text of DOI:10.1039/b604295c

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/dalton | Dalton Transactions
Macropolyhedral boron-containing cluster chemistry. A synthetic approach
via the auto-fusion of [6,9-(SMe2)2-arachno-B10H12]†
Jonathan Bould,a,b Udo Do¨rfler,a Nigam P. Rath,b Lawrence Barton,b Colin A. Kilner,a
Michael G. S. Londesborough,a Daniel L. Ormsbya and John D. Kennedya
Received 23rd March 2006, Accepted 22nd May 2006
First published as an Advance Article on the web 17th July 2006
DOI: 10.1039/b604295c
In an attempt to build up borane-based multicluster assemblies, thermolysis of [6,9-(SMe2)2-arachno-
B10H12] 1 in inert hydrocarbon solution, followed by chromatographic separation, has resulted in the
isolation not only of the previously established single-cluster product from this reaction, [5-(SMe2)-nido-
B10H12] 2 (30%), but also the two two-cluster species [6,9-(SMe2)2-arachno-B10H11-1-(6ꢀ-nido-B10H13)] 3
(20%) and [1,6ꢀ-(nido-B10H13)2] 6 (ca. 0.5%) and the two three-cluster species [6,9-(SMe2)2-arachno-
B10H10-1,5-(6ꢀ-nido-B10H13)2] 4 (5%), characterized crystallographically, and [6,9-(SMe2)2-arachno-
B10H10-1,3-(6ꢀ-nido-B10H13)2] 5 (<1%), identified by NMR spectroscopy. An improved crystallographic
investigation of [5-(SMe2)-nido-B10H12] 2 is also presented. The feasibility of the stability of species
resulting from multiple adjacent substitution of nido-decaboranyl units on the [6,9-(SMe2)2-arachno-
B10H12] skeleton is tested by DFT calculations. In an extension, to attempt the use of pre-r-linked
two-cluster compounds as starting substrates, two-cluster [5-(SMe2)-4-(2ꢀ-nido-B10H13)-nido-B10H11] 7
(0.6%) has been isolated from the reaction of SMe2 with [1,5ꢀ-(nido-B10H13)2], other identified products
being compound 1 (39%) and compound 3 (10.5%).
There is a fascination about the formation of new ‘borons-only’
Introduction
clusters because they have an aura of fundamental elegance. In
Extension of polyhedral boron-containing cluster chemistry to
terms of borons-only macropolyhedral species, the nine-vertex
arachno cluster compounds [(SMe2)B9H13] and [(OnBu2)B9H13]
include contiguous clusters of more than about twelve vertices
requires the intimate fusion of smaller clusters.1,2 Compounds that
have been found to yield the anti isomer of B18H22 upon simple
thermolysis; this is long known.16 In these reactions, the ready
dissociation of the weak SMe2 and OnBu2 ligands to generate
the known reactive {B9H13} species,17 or possibly a related
intermediate, is probably very significant. With this in mind, we
have been interested to see whether the related18 arachno species
[6,9-(SMe2)2-arachno-B10H12] (compound 1, schematic structure
I, L = SMe2), with ten rather than nine boron vertices, could also
similarly be induced readily to undergo cluster fusion. We have
been encouraged to examine for this, not only in the light of the
[(SMe2)B9H13] cluster-fusion reaction, but also because, in previ-
ous work, compound 1 has been reasonably supposed to give a ‘re-
active’ isomer of {(SMe2)B10H12} in its thermolytic reactions with
acetylenes to give {C2B10} dicarbaboranes.19,20 Thence we surmised
that this reactive intermediate might be active towards homofusion
in the absence of alternative substrates. Here we also noted
that the thermolysis of [6,9-(SMe2)2-arachno-B10H12] (compound
1), to give the ‘stable’ isomer of [(SMe2)B10H12], viz. [5-(SMe2)-
nido-B10H12] (compound 2, Fig. 1, schematic structure II, L =
SMe2),21–30 is also known to yield higher, though unspecified,
‘polymeric’ by-products,21 which might well be uncharacterised
multi-cluster species. Additionally, in the reaction of 1-octyne
with compound 1, a by-product has been tentatively identified
as containing three decaboranyl residues per molecule.18 In the
context of these considerations, we here report results on some
higher borane products from a preliminary investigation of the
simple thermolysis of [6,9-(SMe2)2-arachno-B10H12] (compound 1)
have two or more clusters intimately fused together so that they
have two or more framework atoms held in common between
individual pairs of clusters have been dubbed ‘macropolyhedral’
species. Although redox fusion, mediated by transition-element
centres, has been advocated as a general method of ‘designed’
fusion to achieve inter-cluster fusion,3 so far this has been
reported to be successful only to an upper limit of about fourteen
vertices. Meanwhile, there is merit in exploring for other routes
to macropolyhedral boron-containing cluster species. Of such
other routes, thermolyses of arachno nine-vertex boron-containing
cluster compounds have, to date, proved to be one of the
most successful general routes for achieving intercluster fusion.
Although there is as yet no consistent pattern to the products,
several interesting macropolyhedral boranes, heteroboranes and
metallaboranes of between eighteen and twenty-six vertices have
been obtained by this general method.1,2,4–15
aThe School of Chemistry of the University of Leeds, Leeds, England,
UK LS2 9JT. E-mail: j.d.kennedy@leeds.ac.uk; Fax: +44 113 343 6565;
Tel: +44 113 343 6414
bThe Department of Chemistry of the University of Missouri-St. Louis,
St. Louis, MO, 63121, USA
† IUPAC nomenclature for crystallographically determined species:
compound 2, 5-dimethyl sulfide-nido-decaborane; compound 3, 6,9-
bis(dimethyl sulfide)-5-(6ꢀ-nido-decaboranyl)-arachno-decaborane; com-
pound 4, 6,9-bis(dimethyl sulfide)-1,5-bis(6ꢀ-nido-decaboranyl)-arachno-
decaborane; and compound 7, 5-dimethyl sulfide-4-(2ꢀ-nido-decaboranyl)-
nido-decaborane.
3752 | Dalton Trans., 2006, 3752–3765
This journal is
The Royal Society of Chemistry 2006
©

View Article Online
in refluxing hydrocarbon solvents. Preliminary reports on some
aspects of this and closely related work have been made.25,26
experiments and assessments of TLC characteristics of some of
these additional products were therefore conducted.
Thence, a thermolytic reaction at lower temperature, using a
ca. 0.022 M solution of compound 1 in 50 : 50 benzene–hexane
at reflux for 6 h, followed by TLC separation, gave an isolated
yield of the bis(decaboranyl) species [6,9-(SMe2)2-arachno-B10H11-
1-(6ꢀ-nido-B10H13)] (compound 3, schematic structure III, Fig. 2)
in yields of up to 20% as a pale yellow solid. Compound 3,
previously unreported, was characterized by a single-crystal X-
ray diffraction analysis and by NMR spectroscopy as discussed
below. This linked cluster species is the first compound yet reported
that has the classical nido and arachno decaborane {B10} skeletons
r-bound. The monocluster species 2 was again isolated, now
in the smaller yield of 11%. Additional products, which mass
spectrometry indicated were still higher borane species, were again
apparent, but in our hands could not readily be separated by TLC
work. Further experiments were therefore conducted with HPLC
separation in mind.
Results and discussion
The heating for 6 h of a sample of [6,9-(SMe2)2-arachno-B10H12]
(compound 1) at reflux in toluene solution at a concentration of
ca. 0.020 M, followed by thin-layer chromatographic separation
(TLC), resulted in the isolation of [5-(SMe2)-nido-B10H12] (com-
pound 2) in 30% yield as a white air-stable solid. This was identified
by single-crystal X-ray diffraction analysis (Fig. 1) and NMR
spectroscopy. Selected interatomic dimensions for this compound
and the other species reported here are listed in Table 1. NMR and
single-crystal X-ray diffraction work on compound 2 has been
previously reported,21–23 but we were able to glean a somewhat
better crystallographic data set in this present work. Results
from this latter X-ray work, and additional NMR and structural
considerations with regard to compound 2 that arise from our
present study, are briefly discussed below. Other products were also
apparent from this reaction, and mass spectrometric assessment
of these indicated that they were higher boranes. Additional siting
Thus, the heating at reflux of 10 ml of a ca. 0.04 M solution
of compound 1 in benzene for 6 h, followed by repeated TLC
Fig. 1 Left-hand diagram: the crystallographically determined molecular structure of [5-(SMe2)-nido-B10H12] (compound 2; compare also references
23 and 24). Right-hand diagram: Lowest energy DFT-minimized structure for compound 2. Note the good conformational approximation to the
crystallographically determined structure in the right-hand diagram: clearly for this compound packing forces do not overcome the ca. 10 kJ mol−1 energy
barrier to the contra-rotation of the SMe2 unit vs. the cluster unit about the S(5)–B(5) bond (see Fig. 6, below).
˚
Table 1 Selected interatomic distances (A) with s.u.s in parentheses for compounds 2, 3, 4, 6 and 7
3
4
6
7
2
Primed
Unprimed
a
b
c
Primed
Unprimed
Primed
Unprimed
B(5)–B(10)
B(7)–B(8)
B(5)–B(6)
B(6)–B(7)
B(8)–B(9)
B(9)–B(10)
S–B
2.034(3)
1.830(3)
1.771(3)
1.654(3)
1.791(3)
1.785(3)
1.883(2)
1.955(6)
1.984(6)
1.845(6)
1.834(6)
1.788(6)
1.813(7)
1.943(4)
1.898(6)
1.890(6)
1.867(5)
1.889(5)
1.881(6)
1.852(5)
1.874(5)
1.861(5)
1.876(5)
1.858(5)
1.853(5)
1.860(5)
1.924(3) [S6]
1.927(3) [S9]
1.969(5)
1.993(5)
1.804(5)
1.810(5)
1.776(6)
1.803(6)
1.980(6)
1.979(6)
1.806(5)
1.806(5)
1.776(7)
1.789(3)
1.972(3)
1.996(3)
1.801(2)
1.802(3)
1.785(3)
1.794(3)
1.981(3)
1.989(3)
1.796(2)
1.796(3)
1.792(3)
1.793(3)
1.986(3)
1.984(3)
1.783(3)
1.781(3)
1.791(3)
1.794(3)
1.840(2)
2.038(3)
1.660(2)
1.783(3)
1.779(3)
1.790(3)
1.8890(18) [S5]
S–B
B–B
1.693(5)
1.688(5)
1.676(5)
1.686(2)
1.695(2)
conjuncto
This journal is
The Royal Society of Chemistry 2006
Dalton Trans., 2006, 3752–3765 | 3753
©

View Article Online
resonance NMR spectroscopy. Compound 5 was identified by
11B and 1H NMR spectroscopy, allied with DFT calculations
of structure and thence of 11B nuclear shielding, all in general
comparison with compound 4. Compounds 4 and 5 extend the
concept embodied in the structure of compound 3, in that the
structure of compound 3 has one nido decaboranyl unit bound
to an arachno decaboranyl unit, whereas compounds 4 and 5
have two such nido units bound to the arachno decaboranyl unit.
Exhaustive investigation by HPLC revealed trace amounts of other
two-cluster species both with and without bound SMe2 groups
but we have not been able to demonstrate, in this or other related
product mixtures resulting from the thermolysis of compound 1,
the presence of any four-cluster species, consisting, for example,
of three nido decaboranyl units bound to a central decaboranyl
substrate. The steric feasibility of having more than two nido-
decaboranyl units bound in adjacent positions on a central nido-
decaboranyl skeleton is discussed below.
Fig. 2 The crystallographically determined molecular structure of
[6,9-(SMe2)2-arachno-B10H11-1-(6ꢀ-nido-B10H13)] (compound 3).
and thence HPLC separation, yielded, in addition to the mon-
ocluster compound 2 and the two-cluster compound 3, small
quantities of identifiable higher borane species. We have isolated
and identified two of these as the isomeric three-cluster com-
pounds [6,9-(SMe2)2-arachno-B10H10-1,5-(6ꢀ-nido-B10H13)2] (com-
pound 4, schematic structure IV, Fig. 3, 13% yield) and
[6,9-(SMe2)2-arachno-B10H10-1,3-(6ꢀ-nido-B10H13)2] (compound 5,
schematic structure V, Fig. 4, ca. 0.5% yield). A repeat reaction
with two added equivalents of B10H14 showed a small amount
of [1,6ꢀ-(nido-B10H13)2] (compound 6, Fig. 5, ca. 2% yield).
Compound 4 was identified by a single-crystal X-ray diffraction
1
analysis allied with 11B and H single-resonance and multiple-
The crystallographically determined molecular structure of [5-
(SMe2)-nido-B10H12] (compound 2) is in Fig. 1. It is of classical nido
ten-vertex constitution as adequately discussed elsewhere.19,21,23
Formally, it is related to the [nido-B10H13]− anion27,28 by the
−
replacement of a formally anionic exo-terminal {:H} unit by a
neutral :SMe2 ligand. Starting from [6,9-(SMe2)2-arachno-B10H12]
(compound 1), there has been a formal loss of the two-electron
ligand SMe2 to give a classical arachno → nido two-electron
conversion. The SMe2 unit and some of the hydrogen atoms
have concomitantly repositioned themselves about the open face
of the classical nido/arachno boat-shaped skeleton. It has been
demonstrated that there is probably no re-arrangement of the
boron atom positions within the skeleton during this process.24
The crystallographic determination of compound 2 has pre-
viously been reported.23 In this previous report, however, it was
only possible to solve the structure as a racemic twin in space
group P212121. In our present work, we were able to get good
Fig. 3 The crystallographically determined molecular structure of
[6,9-(SMe2)2-arachno-B10H10-1,5-(6ꢀ-nido-B10H13)2] (compound 4).
3754 | Dalton Trans., 2006, 3752–3765
This journal is
The Royal Society of Chemistry 2006
©

View Article Online
gave a rotational energy barrier DG† of ca. 23 kJ mol−1, with
the minimum energy conformer (schematic VI) corresponding
approximately to the crystallographically determined configura-
tion (Fig. 1). This low DG† value implies that any differentiation
of individual rotamers would not be detectable by NMR spec-
troscopy down to less than −100 ◦C, in accord with observation.
The various energetic maxima from this study correspond to
conformations in which the methyl groups of the SMe2 moiety
approximately eclipse the adjacent BH(1), BH(2), BH(6) and
BH(10) positions as in schematics VII, VIII and IX. Because we
have become interested in how rotamers about exo-cluster linkages
may affect cluster geometry and cluster electronic structure, and
thence how they may affect cluster NMR shielding behaviour,27
we have repeated this calculation using the DFT approach at
the B3LYP/6-31G* level, and have calculated the 11B nuclear
shieldings by the GIAO approach.
Fig. 4 The molecular structure of [6,9-(SMe2)2-arachno-B10H10-1,3-
(6ꢀ-nido-B10H13)2] (compound 5), as determined by DFT calculations.
The results gave energy maxima in positions similar to those
found in the earlier extended-Hu¨ckel study, which is reasonable.
The maxima are approximately represented by structures VII, VIII
and IX, respectively corresponding to calculated energies at ca.
9.8, 2.8 and 7.6 kJ mol−1 relative to the lowest energy conformer
VI. The geometries arising from the calculation showed that there
was little effect on cluster geometry associated with the rotation
about the boron–sulfur link. There was also little effect on the
CSC interatomic angle of ca. 101◦ as the SMe2 ligand was rotated.
The principal change was in the boron-to-sulfur distance, which
maximized in the eclipsed positions VII, VIII and IX at 1.941,
Fig. 5 The crystallographically determined structure of [1,6ꢀ-(nido-
B10H13)2] compound 6.
˚
1.923 and 1.937 A, respectively. However, these extensions over
˚
the value of 1.915 A for the ground-state conformation VII are
single-crystal X-ray diffraction data at 150 K. Using anomalous
dispersion we were then able to solve and refine the absolute
structure with improved molecular dimensions, a selection of
very small. They are presumably induced principally by the steric
effects of the eclipsing.
Corresponding to these minimal geometrical changes, there is
also little variation in the calculated 11B nuclear shieldings of
the cluster as the SMe2 group is rotated relative to the cluster
(Fig. 6). The maximum variations are those for 11B(1) and 11B(10),
with variations over ranges of 5.7 and 5.4 ppm, respectively.
These small variations suggest only minor intra-cluster electronic
changes associated with the rotation; they are similar to, but
somewhat less than, those calculated as arising from P-phenyl
rotation about the phosphorus–carbon linkage in [7-Ph-nido-
7,8,10-PC2B8H10], the only other species examined for this type
of phenomenon, where the variations range over more than
6 ppm.30 For compound 2, there is reasonable correlation between
calculated and experimental 11B chemical shifts. This correlation
adds confidence to the use of this DFT-GIAO procedure for
the definitions of the structures of related compounds for which
crystallographically established structures are not available.31,32
This consideration applies particularly for compounds 4 and 5
in this present work as discussed below.
which is given in Table 1. Of these dimensions, the shorter
23
˚
distance B(6)–B(7) of 1.654(3) A, noted before, again is worth
emphasizing, because it leads to the suggestion that the broad
appearance of the two 11B NMR resonances arising from these
two sites derives from a large interboron internuclear coupling
1
constant J(11B–11B) between these two positions. Otherwise, the
11B spectrum itself has previously been assigned and discussed,
and its similarity to that of the [nido-B10H13]− anion noted.23 As
part of this present work we now have additionally measured and
1
assigned the H NMR chemical shifts of the cluster hydrogen
atoms. However, there is nothing exceptional about these within
1
the context of the generally expected parallels between the H
and 11B chemical shifts for directly bound BH(exo)} units within
a particular compound, with {BH(endo)} and {BHB} bridging
proton resonances lying to higher shielding of this general trend.29
Because of the chiral nature of compound 2, the two methyl
groups of the SMe2 ligand are chemically inequivalent and have
different chemical shifts,20 even though there may be relatively free
rotation about the boron-sulfur bond. This rotation has previously
been examined at the extended-Hu¨ckel level.20 This last study
The crystallographically determined structure of [6,9-(SMe2)2-
arachno-B10H11-1-(6ꢀ-nido-B10H13)] (compound 3, schematic III)
is in Fig. 2. This is seen to have the arachno ten-vertex
This journal is
The Royal Society of Chemistry 2006
Dalton Trans., 2006, 3752–3765 | 3755
©

View Article Online
from compound 1 of one SMe2 unit, say that at the 6-position.
This dissociation would give either an incipient or intermediate
‘reactive’ isomer of compound 2 with a {9-(SMe2)B10H12} formu-
lation; compared to compound 1 it would now have a reactive
site at the 6-position instead of SMe2, (schematic X). This 6-
position, having lost SMe2, would be electrophilic, and would
attack the 1-position of a second [(SMe2)2B10H12] molecule via
B–H addition. There would be a consequent rearrangement of
hydrogen atoms around the open face together with concomitant
loss of SMe2 from the 9-position (schematic X). This would be
consistent with the general supposition that the 1,3 positions of
the arachno decaboranyl cluster have a high electron density,40 and
be thereby most susceptible to electrophilic attack by the vacant
6-position of the reactive {9-(SMe2)B10H12} moiety. However, as
we note below, isomerisations of the nido ten-vertex cluster are
also feasible in this type of system, such that positions initially
in the open face of the cluster may end in basal positions. In this
regard it is noted that any general rearrangement tendencies for
the ostensibly rigid nido-decaboranyl unit are far from thoroughly
investigated or established.
The process of additive attack by such a {9-(SMe2)B10H12}
intermediate should, in principle, be repeatable, to give multi-
cluster species. The isolation and characterisation of the two three-
cluster isomers [6,9-(SMe2)2-arachno-B10H10-1,5-(6ꢀ-nido-B10H13)2]
(compound 4, schematic structure IV) and [6,9-(SMe2)2-arachno-
B10H10-1,3-(6ꢀ-nido-B10H13)2] (compound 5, schematic structure V)
from the same system that yields the two-cluster compound 3
thence suggest that this is indeed the case. Of these two three-
cluster compounds, we have been able to establish the structure
of [6,9-(SMe2)2-arachno-B10H10-1,5-(6ꢀ-nido-B10H13)2] (compound
4) by an X-ray diffraction analysis (Fig. 3 above). Obtainable
crystals under our conditions were only very small fibres of ca.
0.08 × 0.02 mm cross section, and required the use of synchrotron-
generated X-irradiation for sufficient diffraction intensity for the
analysis, as reported in a preliminary note.25 The diffraction
analysis revealed the arachno {(SMe2)2B10} configuration as in
compounds 1 and 3, but now with two pendant open decaboranyl
groups rather than just the one of compound 3. These are on
the B(1) and B(5) positions, and are established as {nido-B10H13}
moieties both by their bridging hydrogen-atom positions and by
the general nido-type molecular dimensions established by the
single-crystal X-ray work. Their nido character was also apparent
from the results of 11B and 1H single-resonance and double-
resonance NMR spectroscopy. These results show that each {nido-
B10H13} unit is bound by its 6-position, as in compound 3 described
above. The NMR data were also consistent with the arachno char-
acter of the central {(SMe2)2B10H10} unit. DFT/GIAO molecular
geometry and nuclear shielding calculations on the molecular
structure of compound 4 gave d(11B) values that closely corre-
sponded to those measured experimentally, which gave confidence
in the use of this method for the establishment of the structure
of its isomeric compound 5 (see next paragraph) for which
Fig. 6 Variation in the molecular energy, given as DH in kJ mol−1 (upper
diagram), and in the calculated 11B NMR shielding, given as d(11B) in
ppm (lower diagram), at the various cluster sites in [5-(SMe2)-nido-B10H12]
(compound 2), as determined by DFT/GIAO calculations, as the SMe2
group and the {B10H12} moiety mutually contra-rotate about the S(5)–B(5)
bond.
6,9-(SMe2)2-substituted constitution of the starting species [6,9-
(SMe2)2-arachno-B10H12] (compound 1), but with a pendant nido-
decaboranyl {B10H13} group r-bound by its open-face 6-position
to the apical 1-position of the arachno-decaboranyl unit. The
ꢀ
˚
intercluster r bond B(1)–B(6 ) at 1.693(5) A is comparable to the
intercluster linkage in, for example, the [(nido-B10H13)2] isomers,
33–35
˚
in which it ranges from 1.679(3)–1.704(3) A,
and is similar
to the other r-linked interboron distances listed in Table 1 for
the other new compounds reported here. The individual arachno
and nido ten-boron clusters have straightforward geometries, with
dimensions corresponding closely to those of unsubstituted [6,9-
(SMe2)2-arachno-B10H12] (compound 1) itself,36 and to those of
unsubstituted B10H14 itself,37 respectively. As far as we are aware,
compound 3 is the first reported species that links together the
classical38 and otherwise well-examined nido and arachno ten-
boron clusters. Linked {ten-vertex closo : ten-vertex closo} and
{ten-vertex nido : ten-vertex nido} bis(decaboranyl) systems are,
by contrast, well recognized.31–35,39
The formation of compound 3 from compound 1 is of interest,
because a nido-decaboranyl {B10H13} cluster is generated from an
arachno [(SMe2)2B10H12] unit. This implies inter-cluster transfer of
a hydrogen atom as well as loss of the two SMe2 ligands. In view
of the knowledge of the existence of a ‘reactive’ {(SMe2)B10H12}
isomer that is readily formed from compound 1 under mild ther-
molytic conditions,29,30 it is reasonable to postulate a dissociation
3756 | Dalton Trans., 2006, 3752–3765
This journal is
The Royal Society of Chemistry 2006
©

View Article Online
single-crystal X-ray work was not feasible. A comparison diagram
of the experimental and calculated spectrum for compound 4 is
shown in Fig. 7.
and 5. In the platinaborane these are bound at the (2,4) positions
on the arachno central cluster, rather than the (1,5) and (1,3)
positions as in compounds 4 and 5, respectively. These different
positionings could suggest different mechanisms of cluster
conjunction. This would not be unexpected, because the method
of synthesis of the [(PMe2Ph)2(PMe2C6H4)2Pt2B8H6(B10H13)2]
species, from the reaction of the nine-vertex arachno platinaborane
[(PMe2Ph2)2PtB8H12] with and in molten B10H14, used as a reacting
◦
solvent, at +128 C,42 is quite different from the procedure with
compound 2 in this present work, although it is of interest to note
that all of the pendant nido-decaboranyl clusters resulting from
these fusion reactions are joined by the (6)-positioned vertices of
the nido-decaboranyl cluster.
A small amount of the two-cluster species [1,6ꢀ-(nido-B10H13)2]
(compound 6, Fig. 5, ca. 0.5%), identified by single-crystal X-
ray diffraction analysis, was also isolated from the mixture of
products from this reaction, together with small amounts of other
unidentified two-cluster compounds. The nido:nido compound 6
exhibits the same (1,6ꢀ) cluster linkage as found in the arachno:nido
compound 3. However, given the small yield, no definitive mechan-
istic conclusions may be drawn here, although it should be noted
in this context that, as borane cluster-substituent chemistry is
being developed and thence effective labelling becoming available,
it is becoming increasingly apparent that in many cases the
products ostensibly resulting from reactions on rigid borane cages
in fact involve borane-cluster rearrangements.43 The (1,6ꢀ)-linked
compound 6 is the eighth characterised [(B10H13)2] isomer out of
the eleven possible geometric isomers, the other seven being the
(1,1ꢀ), (1,2ꢀ), (1,5ꢀ), (2,2ꢀ), (2,5ꢀ), (2,6ꢀ) and (5,5ꢀ) analogues.33–35
It is of interest in terms of approaches to new fused macro-
polyhedral assemblies that, in both compounds 4 and 5, the
two nido-decaboranyl substituents are in adjacent positions on
the central arachno decaboranyl nucleus. In prognostications for
possible structures of large very condensed boron-containing
cluster species,1,2,42,44–46 it has been proposed that large glob-
ular boron–hydride species can consist of borons-only cores
surrounded by outer boron–hydride skins.1,2,42 One fundamental
architectural principle visualised here is that the central boron
cores of such species will themselves be based on known simpler
boron–hydride species, but with the exo-terminal boron-hydrogen
bonds formally replaced by interboron bonds to outer recognisable
boron–hydride cluster units. Thus, for example, the framework
of the computed species B84H54 (schematic XII),1,2,26 based on
a fundamental structural fragment of b-rhombohedral boron
with peripheral valencies bound by hydrogen, is seen to consist
of a central {B12} core with a nido-type pentagonal pyramidal
six-vertex {B6} unit bound to each of the twelve boron atoms,
with the twelve {B6} units mutually fused to generate the sixty-
boron-atom basis of the outer boron–hydride skin. This structural
perception suggests a general strategy for the synthesis of such
Fig. 7 NMR spectra for [6,9-(SMe2)2-arachno-B10H10-1,5-(6ꢀ-nido-B10H13)2]
(compound 4): upper trace, the 11B spectrum as calculated by DFT B3LYP/
6-31G* with no 1H coupling; lower trace, the measured room-temperature
1
128 MHz 11B–{ H(broad-band noise)} spectrum (CDCl3). Selective
line-broadening has been applied to appropriate resonance lines in
the calculated spectrum in order to mimic experimental linewidths.
As just mentioned, crystals of compound 5 suitable for single-
crystal X-ray work have not yet been obtained. However, NMR
spectroscopy shows a symmetrical bis(nido-decaboranyl)-arachno-
decaborane constitution for this species. Detailed 11B and 1H
single and multiple resonance work readily establishes the (1,3)
substituent pattern as in schematic cluster structure V above,
giving the overall formulation [6,9-(SMe2)2-arachno-B10H10-1,3-
(6ꢀ-nido-B10H13)2]. This structural conclusion was also supported
by DFT/GIAO structural and nuclear shielding calculations,
which gave d(11B) values corresponding very reasonably with those
measured and assigned experimentally (see Experimental section).
Compounds 4 and 5 are the first tridecaboranyls definitively
identified. The resultant {B30} units, as far as we are aware,
constitute the biggest molecular all-boron skeletal assemblies
yet unequivocally characterized. Although a sample of B30H38,
bis(nido-decaboranyl)-nido-decaborane, [B10H12(B10H13)2], was
reported a quarter of a century ago to be isolated from the products
from the thermolysis in refluxing toluene solution of nido-B10H14
in the presence of 0.03 mol% [6,9-(SC4H8)2-arachno-B10H12] as
catalyst, this was an unspecified and unspecifiable mixture of
isomers,41 of which some 546 are in principle possible. A closer
structural parallel for compounds 4 and 5 is perhaps the bis(nido-
decaboranyl)-arachno-diplatinadecaborane, [6,9-(PMe2Ph)2 -
l(5C,6P),l(8C,9P)-(PMe2 -ortho-C6H4 )2 -arachno-6,9-Pt2B8H8 -2,4-
(6ꢀ-nido-B10H13)2] [schematic cluster structure XI, in which M is
{Pt(PMe2Ph) (PMe2C6H4)}].42 This last species has two pendant
nido-decaboranyl units also bound by their 6-positions, but they
are now bound to a central arachno ten-vertex {Pt2B8} unit, rather
than to a central arachno ten-vertex {B10} unit as in compounds 4
This journal is
The Royal Society of Chemistry 2006
Dalton Trans., 2006, 3752–3765 | 3757
©

View Article Online
species by the multiple substitution of {BH(exo)} hydrogen atoms
on starting borane substrates by boron–hydride units. These outer
moieties might then be induced to cross-link, perhaps by thermal
dihydrogen elimination, to generate a contiguous outer boron–
hydride skin (schematic XIII). Here the peralkylation of the closed
{C2B10} and related carbaborane skeletons47–49 offers an interesting
related paradigm: however, as far as we are aware, no attempts
have yet been made directly to interlink the outer carbon atoms
in these latter peralkylated {C2B10} species, although the results
of straightforward thermolysis experiments would be interesting.
In any event, it is clear that compounds 4 and 5, each with its two
adjacent {B10H13} substituents on a ten-boron core (Fig. 3 and
4, schematics IV and V), represent initial steps on the required
multi-cluster accumulation.
Fig. 8 The molecular structure of the as yet hypothetical species
[6,9-(SMe2)2-arachno-B10H8-1,2,3,4-(6ꢀ-nido-B10H13)4] as determined by
DFT calculations. For economy of computing time, the energetic minimum
was constrained to C2v symmetry, but the stereochemical feasibility of
having four adjacent nido-decaboranyl units bound to the four adjacent
sites on a typical borane four-boron deltahedral ‘diamond’ is nevertheless
readily demonstrated.
be extendable to bigger starting borane substrates. Here, in
preliminary siting experimentation, we find that the heating at
reflux of [1,5ꢀ-(nido-B10H13)2]33 in SMe2 solution for 5 h, followed
by TLC and HPLC separation, gave single-cluster [6,9-(SMe2)2-
arachno-B10H12)] (compound 1) and two-cluster [6,9-(SMe2)2-
arachno-B10H11-1-(6ꢀ-nido-B10H13)] (compound 3, described above,
schematic cluster structures III and XIV) as the principle products,
in isolated yields of 39 and 10.5%, respectively. No three-cluster
species were detected, but a small amount of the previously
unreported two-cluster species [5-(SMe2)-4-(2ꢀ-nido-B10H13)-nido-
B10H11] (compound 7, 0.6% yield) was also isolated. Compound
7 was characterised by a single-crystal X-ray diffraction analysis
(Fig. 9, schematic cluster structure XV) and by NMR spectroscopy
(see Experimental section below).
In this regard it is of interest to ascertain whether multiple
adjacent substitution is indeed a valid further target along this
strategic pathway. To assess this possibility, a DFT calculation of
a five-cluster tetrakis(nido-decaboranyl)-substituted {(SMe2)2B10}
arachno species, specifically [6,9-(SMe2)2-arachno-B10H8-1,2,3,4-
(6ꢀ-nido-B10H13)4], was carried out. This symmetrically substituted
(1,2,3,4) species was chosen for convenience of calculation. Also
for convenience of calculation, a C2v symmetry constraint was
adopted. This five-cluster system minimised energetically without
any disruption to the individual nido-decaboranyl and arachno-
decaboranyl geometries (Fig. 8), strongly supporting the concept
that multiple decaboranyl substitution on adjacent sites on
deltahedral borane clusters is in principle feasible, at least as far as
steric considerations in the product are concerned. We currently
examine for compounds that may induce or exhibit such multiple
substitution in this and related systems.
Fig. 9 The crystallographically determined molecular structure of
[5-(SMe2)-4-(2ꢀ-nido-B10H13)-nido-B10H11] (compound 7).
Compound 7 features a subcluster comparable to that found
in the single-cluster species, [5-(SMe2)-nido-B10H12] (compound 2)
and, although 7 is formed in modest yield, it nevertheless does
suggest the possible presence in the reaction mixture of reactive
In the general context of building bigger borane-based as-
semblies, there is also interest in determining whether this inter-
cluster coupling based on transient SMe2-containing species might
3758 | Dalton Trans., 2006, 3752–3765
This journal is
The Royal Society of Chemistry 2006
©

View Article Online
it may be noted that it is increasingly recognised that, under
some conditions, for example in the presence of Lewis base, open-
faced borane, metalladecaborane, and carbadecaborane clusters
do have a tendency to exhibit isomerisation. General mechanisms
for ten-vertex clusters are proposed to proceed through ‘vertex-
flip’ processes, in which a ‘prow’ B(6) or B(9) vertex may be
regarded as moving around the open face of the ten-vertex
cluster.43,50,51 Thus, for example, it is possible in principle to convert
the configuration of an initial [6,9-(SMe2)2-arachno-B10H11-1-
(5ꢀ-nido-B10H13)] product into the final (1,6ꢀ) configuration in
3, through a vertex flip on one nido-decaboranyl subcluster
(schematics XVII → XIX).
intermediates of the type {(SMe2)B10H11(B10H13)}, based on the
‘reactive’ isomer of [(SMe2)B10H12], but with an additional {nido-
B10H13} moiety appended. This augurs well for the isolation of
four-cluster species, as well as alternative three-cluster species,
from this type of reaction system.
The (1,6ꢀ)-r linkage in the major product [6,9-(SMe2)2-
arachno-B10H11-1-(6ꢀ-nido-B10H13)] (compound 3, schematic cluster
structure XIV) and the (4,2ꢀ)-r linkage in the minor product
[5-(SMe2)-4-(2ꢀ-nido-B10H13)-nido-B10H11] (compound 7, schematic
cluster structure XV) are clearly different to the starting config-
uration in [1,5ꢀ-(nido-B10H13)2] (schematic cluster structure XVI).
The formation of the single-cluster species [6,9-(SMe2)2-arachno-
B10H12] 1 (39%) shows that much cluster de-linking occurs in
the reaction system, possibly implying the formation of the two-
cluster compounds 3 and 7 via cluster fusion through a reactive
nido-type {(SMe2)B10H12} isomer, as seen for the formation of
compounds 4 and 5 as well as for compound 3 as reported above.
However, this reaction to give 3 and 7 from [1,5ꢀ-(nido-B10H13)2]
Two such cluster isomerisations, one for each sub-cluster, for
example as in schematics XX → XXI and XXII → XXIII, would be
required to afford the (4,2ꢀ) configuration observed for compound
7. However, deuterium-labelling experiments24 have been taken
to suggest that there is no cluster isomerisation either during the
formation of [(SMe2)2B10H12] 1 itself from nido-B10H14 in SMe2
solution at ambient temperature, or during the subsequent higher-
temperature formation of the proposed reactive {(SMe2)B10H12}
intermediate. That isomerisations occur in the present system
implies that the presence of an effective borane substituent on a
borane subcluster affects the course of the reaction in the two-
cluster compound. A precedent for this type of substituent effect
is that the heating of the chlorine-substituted decaboranyl cluster
species [2,4-Cl2-nido-B10H12] in SMe2 solution at reflux gives [6,9-
(SMe2)2-1,7-Cl2-arachno-B10H10] as one of the products, also via a
postulated ‘vertex flip’ during the course of the reaction.43,50 On
the other hand, the monohalogenated species [2-Br-nido-B10H13]
yields an apparently unrearranged species [6,9-(SMe2)2-2-Br-
arachno-B10H11].52 The high yield of the single-cluster product
[(SMe2)2B10H12] 1 here may be a consequence of cluster vertex
motility promoting cluster de-linking. This may impose a kinetic
limit on the number of decaboranyl clusters that may be conjoined
in this type of system. The overall reaction system is clearly
complex and contains competing processes: in the auto-reaction of
[6,9-(SMe2)2-arachno-B10H12] (compound 1) in refluxing benzene
◦
was carried out at lower temperature, in refluxing SMe2 (38 C),
in contrast to the reactions in refluxing benzene (80 ◦C) or toluene
(111 ◦C) to give 3, 4 and 5 as discussed above. Further, loss of
SMe2 from 1 to form the ‘reactive’ intermediate {(SMe2)B10H12}
has been shown19 to be retarded by excess SMe2, thus limiting
the likelihood of its formation in this system, which uses neat
refluxing SMe2 as solvent. Additionally, the production of the
two-cluster compound 3, which has only one appended borane
ligand, in significant quantity, in solution in neat refluxing SMe2,
suggests that this reaction of two-cluster [1,5ꢀ-(nido-B10H13)2]
to give compound 3 may proceed differently to that of the
single decaboranyl cluster discussed above (near schematic cluster
structures III, V and V). The observed products may therefore
result from direct attack by SMe2 on the original undissociated
double-clusters rather than through a cluster-coupling process.
Consequently, a succeeding cluster isomerisation would then be
required to yield the observed isomer composition. In this regard,
This journal is
The Royal Society of Chemistry 2006
Dalton Trans., 2006, 3752–3765 | 3759
©

View Article Online
dealt with earlier in this account, the presence of the reactive inter-
mediate promotes cluster oligomerisation; on the other hand, the
reactions in SMe2 as solvent suggest that free SMe2 will promote
cluster isomerisation and de-linking of inter-cluster r linkages.
We hope to investigate these differential types of processes
further.
visually with ambient or ultra-violet irradiation. HPLC was
performed on a silica-gel column (Lichosorb SI 60, 25 × 2.12 cm).
Mass spectrometry was carried out on a VG Autospec instrument
using electron-impact ionisation at 70 eV. NMR spectroscopy was
performed at ca. 5.9 and 11.8 T (fields corresponding to 250
1
and 500 MHz H frequencies, respectively) using commercially
available instrumentation and using techniques and procedures
as adequately described and enunciated elsewhere.29,57–59 Chemical
shifts d are given in ppm relative to N = 100 MHz for d(1H)
( 0.05 ppm) (nominally TMS) and N = 32.083972 MHz for d(11B)
( 0.5 ppm) (nominally [F3BOEt2] in CDCl3).29 N is as defined by
McFarlane.60
Conclusion
One initial objective, of achieving intimate intercluster fusion to
give contiguous new borane skeletons of more then twenty vertices
with more than one vertex in common, so that the multicentre
bonding extends over both subclusters, was not observed in this
approach. Nevertheless, the achieved results of two or more
clusters conjoined by two-electron two-centre r linkages, in good
yield for a thermolytic conjoining reaction, has some interesting
auguries. First, the ready attack of the ‘reactive’ {(SMe2)B10H12}
isomer at the electrophilic site of [6,9-(SMe2)2-arachno-B10H12]
(compound 1) suggests that the method could be of more general
use in co-thermolysis reactions of compound 1 with other boranes
in order to link the versatile nido-decaboranyl residue to other
borane residues. Reaction with the [closo-B12H12]2− dianion to
give [B12H11(B10H13)]2−, which might be precursive, via further
oxidation, to the relatively recently reported53,54 very interesting
fused-cluster species [B22H22]2−, would be an attractive target
here. Second, the multiple addition of nido-decaboranyl residues
in adjacent sites is of potential use in the synthesis of higher
more condensed boranes. Here, the close proximity of the nido-
decaboranyl residues on the central borane substrate might imply
that coupling reactions among them, with dihydrogen loss, may
be readily effected (schematic structure XIII above). The close
proximity also suggests that the adjacent open nido-decaboranyl
clusters may be linkable by bonding to a common heteroatom,
such as a transition-element centre. We are currently devising
experiments that may generate larger quantities of the adjacently
substituted bis(decaboranyl) species 4 and 5 in order to test these
hypotheses. In more general cluster construction terms, these
could be steps towards the feasibility of building up big filled-
ball globular species based on the concept of borons-only cores
surrounded by boron hydride sheaths (schematic structures XII
and XIII above).1,2,45 These latter ‘filled-ball’ architectures may be
those that ultimately typify big borane chemistry. Their ‘filled-ball’
constitution2,42,44,45 contrasts to the ‘hollow-ball’ architectures of
the fullerenes.
Formation of [5-(SMe2)-nido-B10H12] (compound 2) and [1-(6ꢀ-
nido-B10H13)-6,9-(SMe2)2-arachno-B10H11] (compound 3). A sam-
ple of [(SMe2)2B10H12] (compound 1; 100 mg; 410 lmol) was
dissolved in toluene (20 ml) and heated at reflux for 3 h. After
cooling to room temperature, the more volatile components were
removed under reduced pressure, the solid residue redissolved in
CH2Cl2 (ca. 5 ml) and the products separated and purified by
repeated preparative TLC, development using 30 : 70 hexane–
CH2Cl2 as the liquid phase giving crude [7-(SMe2)-nido-B10H12]
(compound 2) as a white solid (RF 0.81). After recrystallisation by
diffusion of hexane into a solution in CH2Cl2 at room temperature,
pure compound 2 was isolated in 30% yield (23 mg, 122 lmol).
By using a mixture of cyclohexane (10 ml) and benzene (8 ml)
as solvent for compound 1 (100 mg; 410 lmol), and heating at
reflux for 6 h, two compounds were obtained. Both were separated
and purified by repeated preparative TLC, final development with
hexane–CH2Cl2 (3 : 7) giving [5-(SMe2)-nido-B10H12] (compound
2; RF 0.81; 8 mg, 11%) as a white solid, and [1-(6ꢀ-nido-B10H13)-
6,9-(SMe2)2-arachno-B10H11] (compound 3; RF 0.50; 15 mg, 20%)
as a yellow solid. Measured NMR data for compound 2 (CDCl3,
294–297 K), ordered as d(11B) [d(1H) of directly bound H], are:
B(1) +5.5 [+2.64], B(2) −27.5 [+0.62], B(3), −3.1 [+3.03], B(4)
−41.7 [+0.41], B(5) −5.1 [+3.03], B(6) +21.9 [+4.22], B(7) +0.5
(1J(11B, 11B) 110 Hz) [no H], B(8) −10.5 [+2.22], B(9) −5.1 [+2.93]
and B(10) −11.0 [+3.30]; additionally d(1H){l-H(5,6)} −4.09,
{l-H(8.9)} −2.65, {l-H(9,10)} −2.63 and d(1H) (SMe2) +2.68.
Calculated d (11B) values (DFT/GIAO) for the minimum energy
rotamer (schematic configuration VI above) as follows: B(1) +7.6,
B(2) −28.0, B(3) −2.7, B(4) −41.8, B(5) +0.4, B(6) +17.2, B(7)
−5.2, B(8) +5.2, B(9) −7.4 and B(10) −10.3. Measured NMR
data for compound 3 (CDCl3, 294–297 K), ordered as d(11B)
[d(1H) of directly bound H], are: B(1) −40.2 (no H), B(2ꢀ,4ꢀ) −37.9
[+0.45, +0.38], B(3) −32.3 [+0.67], B(6,9) −23.1 [−0.20], B(7,8)
−19.1 [+1.63], B(5,10) −17.9 [+1.66], B(2,4) −2.8 [+2.23], B(5ꢀ,7ꢀ)
−0.4 [+3.10], B(8ꢀ,10ꢀ) +0.3 [+3.00], B(9ꢀ) +7.7 [+3.69], B(1ꢀ,3ꢀ)
+11.4 [+3.56], B(6ꢀ) +32.5 [1 (conjuncto position, no terminal
H)]; additionally d(1H) {l-H(5,10)} −4.42, {l-H(7,8)} −4.53, {l-
H(5ꢀ,6ꢀ and 6ꢀ,7ꢀ)} −1.36, {l-H(8ꢀ,9ꢀ and 9ꢀ,10ꢀ)} −2.24 and d(1H)
(SMe2) +2.44.
Experimental
General
Reactions were carried out in dry solvents (toluene, benzene,
hexane, cyclohexane and dichloromethane) under dry dinitrogen,
but subsequent manipulatory and separatory procedures were
carried out in air. [6,9-(SMe2)2-arachno-B10H12] (compound 1)55,56
and [1,5ꢀ-(nido-B10H13)2]31–35 was prepared essentially as in litera-
ture methods. Preparative thin-layer chromatography (TLC) was
carried out using 1 mm layers of silica gel G (Fluka, type GF254)
made from water slurries on glass plates of dimensions 20 × 20 cm2,
followed by drying in air at 80 ◦C; components were located
Isolation of [6,9-(SMe2)2-arachno-B10H10-1,5-(6ꢀ-nido-B10H13)2]
(compound 4), [6,9-(SMe2)2-arachno-B10H10-1,3-(6ꢀ-nido-B10H13)2]
(compound 5) and [1,6ꢀ-(nido-B10H13)2 (compound 6). In a slightly
modified repetition of the above procedure, a solution of [6,9-
(SMe2)2-arachno-B10H12] (compound 1; 107 mg, 440 lmol) in
3760 | Dalton Trans., 2006, 3752–3765
This journal is
The Royal Society of Chemistry 2006
©

View Article Online
benzene (10 ml) was heated under reflux for 6 h. The products were
then subjected to separation by TLC, using 30 : 70 hexane–CH2Cl2
as liquid medium. Just behind the most mobile band at RF 0.8
{identified as [5-(SMe2)-nido-B10H12] (compound 2)} was a faint
yellow band, which was subjected to further separation by HPLC
(57 : 43 hexane–CH2Cl2, 20 ml min−1). Three components of
interest were obtained: A, RT 9.6, B RT 11.3 and C RT 12.1 min.
Multi-element NMR spectroscopy indicated that A contained
a r-linked arachno–nido cluster species. It has eluded further
characterisation. NMR analysis of B together with DFT/GAIO
11B chemical shift calculations enabled us reasonably to iden-
tify the compound as [6,9-(SMe2)2-arachno-B10H10-1,3-(6ꢀ-nido-
B10H13)2] (compound 5; 1 mg, 0.5%). NMR data for compound
5 (CDCl3, 294–297 K), ordered as measured value of d(11B)
{calculated d(11B) value by DFT/GIAO} [d(1H) of directly bound
H], are as follows: B(1,3) −38.5 {−37.0} [conjuncto site, no H],
B(4ꢀ,4ꢀꢀ) −37.4 {−38.7} [+0.35], B(2ꢀ,2ꢀꢀ) −32.4 {−32.1} [+0.17],
B(6,9) −23.0 {−24.8} [−0.13], B(5,7,8,10) −16.9 {−14.8} [+1.62],
B(2,4) −2.0 {5.0} [+2.31], B(8ꢀ,10ꢀ,8ꢀꢀ,10ꢀꢀ) −0.3 {−3.4} [+2.87],
B(5ꢀ,7ꢀ,5ꢀꢀ,7ꢀꢀ) +0.0 {+1.1} [+2.94], B(9ꢀ,9ꢀꢀ) +8.3 {+3.9} [+3.66],
B(1ꢀ,3ꢀ,1ꢀꢀ,3ꢀꢀ) +11.4 {+10.2} [+3.47] and B(6ꢀ,6ꢀꢀ) +32.1 {+29.9}
[conjuncto site, no H]; additionally d(1H) {l-H(5,10;7,8)} −4.33,
d(1H) {l-H(5ꢀ,6ꢀ and 6ꢀ,7ꢀ; 5ꢀꢀ,6ꢀꢀ and 6ꢀꢀ,7ꢀꢀ)} −1.41, d(1H) {l-
H(8ꢀ,9ꢀ and 9ꢀ,10ꢀ; 8ꢀꢀ,9ꢀꢀ and 9ꢀꢀ,10ꢀꢀ)} −1.41 and d(1H) (SMe2)
+2.66. Unfortunately, the compound was not amenable to mass
spectrometric analysis. Peak C was identified as [6,9-(SMe2)2-
arachno-B10H10-1,5-(6ꢀ-nido-B10H13)2] (compound 4). Diffusion of
hexane into a CDCl3 solution of compound 4 yielded small
yellow crystals which required a synchrotron-generated X-ray
radiation source to obtain sufficient diffraction intensity for a
crystallographic analysis. Selected geometrical data for 4 are
listed in Table 1 above. Measured NMR data for compound 4
(CDCl3, 294–297 K), ordered as d(11B) [d(1H) of directly bound
H], are: B(1) −40.5 [conjuncto position, no terminal H], B(3, 4ꢀ)
−37.5 [+0.44, +0.31], B(4ꢀ) −36.2 [+0.51], B(2ꢀꢀ) −32.8 [−0.27],
B(2ꢀ) −31.7 [+0.68], B(9) −22.9 [+0.14], B(6,7) −19.6 [+1.72,
+1.58], B(7,8) −17.5 [+1.79], B(5) −14.1 [conjuncto position, no
terminal H], B(2) −3.2 [+2.30], B(1ꢀꢀ) −2.1 [+2.22], B(5ꢀ,5ꢀꢀ, 10ꢀ)
−1.1 [+3.00{2H} and +2.87], B(8ꢀ,8ꢀꢀ,10ꢀ) +0.5 [+3.04], B(9ꢀ,9ꢀꢀ)
+8.4 [+3.77], B(1ꢀ,3ꢀ,1ꢀꢀ,3ꢀꢀ) +11.6 [+3.61{2H} and +3.52{2H}],
B(6ꢀ or 6ꢀꢀ) +26.8 [conjuncto position, no terminal H], B(6ꢀꢀ or
6ꢀ) +32.2 [conjuncto position, no terminal H]; additionally d(1H)
{l-H(5,10)} −4.17, {l-H(7,8)} −4.26, {l-H(5ꢀ,6ꢀ and 6ꢀ,7ꢀ; 5ꢀꢀ,6ꢀꢀ
and 6ꢀꢀ,7ꢀꢀ)} −1.29, −1.52(2H), −1.62, {l-H(8ꢀ,9ꢀ and 9ꢀ,10ꢀ; 8ꢀꢀ,9ꢀꢀ
and 9ꢀꢀ,10ꢀꢀ)} −2.15, −2.24, −2.31(2H) and d(1H) (SMe2) +2.46].
Calculated d(11B) values (DFT/GIAO) are shown in Table 2 and a
comparison diagram of the measured and calculated d(11B) values
is shown in Fig. 7. Mass spectrometry showed an isotopomer
envelope centred at m/z 484 corresponding to that calculated
+
for {C4H48B30S2} . Repeat reactions were carried out similarly
but with two equivalents of nido-B10H14 present. Yields varied
with each reaction but the following is a typical procedure: [6,9-
(SMe2)2-arachno-B10H12] (compound 1; 129 mg, 0.53 mmol) and
nido-B10H14 (0.138 g, 1.1 mmol) in benzene (20 ml) were heated
under reflux for 3 h. The solvent was then removed using a
rotary evaporator (water-bath, ca. 30 ◦C) and the decaborane
removed by overnight ambient temperature vacuum sublimation
(ca. 0.01 mm Hg). Preparative TLC (70 : 30 CH2Cl2–hexane) of
the residue showed a characteristic pattern of three main bands:
B (RF 0.5, yellow), C (RF 0.4, yellow, compound 3, 27.5 mg,
14%) and D (RF 0.2, colourless, detected under UV irradiation,
compound 1, 10 mg). Further TLC separation of B using 50 : 50
CH2Cl2–hexane as liquid medium revealed the presence of a
faint UV-active band above the main yellow band. The yellow
band was identified as [6,9-(SMe2)2-arachno-B10H10-1,5-(6ꢀ-nido-
B10H13)2] (compound 4; 33 mg, 0.68 mmol, 13%). Further HPLC
separation (30 : 70 CH2Cl2:hexane, 20 ml min−1) of the UV-
active band gave three components, E, F and G, with RT 5.5,
7.4 and 8.4 min, respectively. The first two components, E
and F, appeared to be isomeric [(B10H13)2] species but there
was insufficient material for further characterisation. Vacuum
sublimation of component G (sealed glass tube with one end
placed on a warm surface) gave colourless crystals that en-
abled its characterisation by single-crystal X-ray diffraction
analysis as [1,6ꢀ-(nido-B10H13)2] (Compound 6, 2.5 mg, 10 lmol
0.5%). Selected geometrical data for 6 are listed in Table 1
above.
Isolation of [5-(SMe2)-4-(2ꢀ-nido-B10H13)-nido-B10H11] (com-
pound 7). A solution of [1,5ꢀ-(nido-B10H13)2] (98.7 mg, 0.41 mmol)
in SMe2 (5 ml) was heated at reflux for 5 h, after which time
the SMe2 was removed using a rotary-film evaporator (water-
bath ca. 30 ◦C). The resulting solid was redissolved in CH2Cl2,
precipitated with diethyl ether, and filtered off. The off-white
precipitate thus obtained was identified by 11B NMR spectroscopy
as [6,9-(SMe2)-arachno-B10H12] (compound 1; 62 mg, 30%). The
filtrate was reduced in volume, applied to a preparative TLC
plate and developed using 100% CH2Cl2 as liquid medium. Four
bands between RF 0.6 and RF 0.4 were identified under UV and
Table 2 DFT-GIAO//B3LYP/6-31G* predicted shifts for [6,9-(SMe2)2-arachno-B10H10-1,5-(6ꢀ-nido-B10H13)2] (compound 4) in ppm
Central arachno
sub-cluster
nido-Decaboranyl
nido-Decaboranyl
Position
sub-cluster attached at 1ꢀ
sub-cluster attached at 5ꢀ
1
2
3
4
5
6
7
8
9
−36.80
−2.65
+10.23
−33.31
+10.00
−37.90
−1.21
+28.43
−1.95
+0.25
+5.65
+0.94
+10.16
−32.33
+9.91
−36.54
−1.89
+23.19
−1.74
−0.31
+7.41
−0.68
−37.18
+0.29
−13.30
−17.90
−18.18
−16.61
−20.35
−17.30
10
This journal is
The Royal Society of Chemistry 2006
Dalton Trans., 2006, 3752–3765 | 3761
©

View Article Online
3762 | Dalton Trans., 2006, 3752–3765
This journal is
The Royal Society of Chemistry 2006
©

View Article Online
visible light, of which two yielded identifiable products. A very pale
yellow band at RF ca. 0.5 was identified as [6,9-(SMe2)-arachno-
B10H12] (compound 1; 19 mg, giving a total yield of 39%). A pale
yellow band at RF ca. 0.55 was subject to further separation by
preparative HPLC (43 : 57 CH2Cl2–hexane, 20 ml min−1), giving
[1-(6ꢀ-nido-B10H13)-6,9-(SMe2)2-arachno-B10H11] (compound 3; RT
22.0 min; 16.0 mg, 10.5%) and [5-(SMe2)-4-(2ꢀ-nido-B10H13)-nido-
B10H11] (compound 7; RT 28.6 min; 0.8 mg, 0.6%). NMR data for
compound 7 (CDCl3, 294–297 K), ordered as measured value
of d(11B) {calculated d(11B) value by DFT/GIAO} [d(1H) of
directly bound H], are: B(4) −39.6 {−40.2} [conjuncto site, no
H], B(4ꢀ) −35.4 {−36.7} [+0.61], B(2ꢀ) −30.5 {−31.4} [conjuncto
site, no H], B(2) −27.2 {−26.6} [−0.61], B(10) −9.8 {−12.2}
[+2.05], B(7 and 9) −4.2 {−2.4 and −6.0} [+2.96(2H)], B(3) −2.0
{−3.4} [+2.56], B(5ꢀ,7ꢀ, 8ꢀ,10ꢀ) +1.4 {+0.71, +1.47, +1.81, +0.90}
[+3.13(2H) and +3.09(2H)], B(5) +1.4 {+0.5} [site of SMe2
substituent, no H], B(8) +3.8 {+4.9} [+3.21], B(9ꢀ) +7.1 {+6.7}
[+2.98] and B(1) +8.5 {+7.3} [+3.75], B(6ꢀ) +10.9 {+8.6} [+4.13],
B(1ꢀ, 3ꢀ) +13.8 {+13.0 and +13.4} [+3.74 and 3.62], B(6) +20.2
{+19.1} [+4.13]; additionally d(1H) {l-H(6,7)} −4.31, d(1H) {l-
(5ꢀ6ꢀ,6ꢀ7ꢀ)} −1.85 (2H), {l-H(8,9, 9ꢀ10ꢀ))} −2.06 (2H), {l-H(89,
910)} −2.44 (2H) and d(1H) (SMe2) +2.66. Mass spectrometry
showed an isotopomer envelope centred at m/z 302 corresponding
˚
(k = 0.71073 A); data for compound 4 were measured with
˚
synchrotron radiation (k = 0.6874 A), using a Bruker AXS
SMART CCD diffractometer.25,26,68–70 In each case, the crystal
was cooled with a nitrogen gas stream. Absorption corrections
were made by semi-empirical methods. The structures were
solved by automatic direct methods and refined by least squares
methods on all measured F2 values, with a weighting scheme
w−1 = r2(Fo2) + (aP)2 + (bP), where P = (Fo + 2 Fc )/3.
2
2
ꢀ
ꢀ
Residuals were defined by R1
=
||Fo| − |Fc||/ |Fo|,
ꢀ
ꢀ
wR2 = [ w(Fo − Fc )2/ w(Fo2)2]1/2. Programs consisted of
standard control software for the diffractometers, members of the
SHELX family,71 and local programs. There were two molecules
in the asymmetric unit of compound 4 and of these, one of
the decaboranyl subclusters contained disorder due to a pseudo-
plane of symmetry through B(1)B(4)B(6). This was resolved in
terms of the partial occupancy of the B(7)/B(5) positions. The
crystallographically determined structures of all the compounds
were consistent with their assigned 11B and 1H NMR spectra, con-
firming that in each case the crystals were representative of the bulk
samples. Drawings of molecular structures are by the ORTEP-
3 program,72 and atoms are represented by 50% probability
ellipsoids.
2
2
CCDC reference numbers 280344 (2), 280345 (3), 162053 (4),
280346 (6) and 280347 (7).
For crystallographic data in CIF or other electronic format see
DOI: 10.1039/b604295c
+
to that calculated for {C2H30B20S1} . 11B NMR spectroscopy
indicated that the remaining two bands also contained r-linked
arachno–nido cluster species but we have been unable to fully
characterise them. Hexane diffusion into a CDCl3 solution of 7
gave colourless needles suitable for the single-crystal X-ray diffrac-
tion analysis, from which selected geometrical data are listed in
Table 1 above.
Acknowledgements
We thank the Deutsche Forschungsgemeinschaft for support to
U. D. and acknowledge the UK EPSRC for previous instru-
mental grants and for a partial contribution to a studentship to
D. L. O. Part of this work was self-financed by J. B., who thanks the
School of Chemistry of the University of Leeds for facilities. We
also thank Dr C. van Wu¨llen for providing a copy of the DIGLO
software and Prof. J. B. C. Whitaker for assistance with computer-
software configuration, Dr D. A. Waller and M. Holmes for
computer-hardware assistance, S. A. Barrett and Y. M. McKinnes
for technical assistance, and Dr M. Thornton-Pett for crystallo-
graphic advice in some of the earlier stages of this work.
Computational methods
For the IGLO calculations, structures were initially optimised with
the 3-21G basis set and the final optimization was performed using
the 6-31G* basis set and the Gaussian 92 package.61 The resultant
geometry was used to predict the NMR chemical shifts using the
IGLO package62–66 and the DZ basis set incorporated therein. For
the DFT calculations, both structures were initially optimised with
the STO-3G and 6-31G* basis sets using standard methods. The
final optimizations, frequency analyses (to confirm true minima)
and GIAO NMR predictions were performed using the 6-31G*
basis set and B3LYP methodology as incorporated in the Gaussian
98 package.67 Gas-phase NMR predictions were calibrated via
a prediction at the same level of theory on diborane, which is
taken to have an NMR shift d(11B) of −16.6 ppm with respect to
BF3OEt2.
References
1 J. Bould, W. Clegg, S. J. Teat, L. Barton, N. P. Rath, M. Thornton-
Pett and J. D. Kennedy, Boron Chemistry at the Millennium: Special
Edition, Inorg. Chim. Acta, 1999, 289, 95.
2 H.-J. Yao, C.-H. Hu, J. Sun, R.-S. Jin, P.-J. Zheng, J. Bould, R. Greatrex,
J. D. Kennedy, D. L. Ormsby and M. Thornton-Pett, Collect. Czech.
Chem. Commun., 1999, 64, 927.
3 See, for example: R. N. Grimes, Acc. Chem. Res., 1983, 16, 22, (Adv.
Inorg. Chem. Radiochem., 1983, 26, 55); and in Recent Advances in
Metallacarborane Sandwich Chemistry: Controlled Linking, Stacking,
and Cool Fusion, ch. 13 in The Borane–Carborane–Carbocation
Continuum, ed. J. Casanova, Wiley, New York, 1998, pp. 307–332, and
references cited therein.
4 M. A. Beckett, J. E. Crook, N. N. Greenwood, J. D. Kennedy and W. S.
McDonald, J. Chem. Soc., Chem. Commun., 1982, 552.
5 M. A. Beckett, J. E. Crook, N. N. Greenwood and J. D. Kennedy,
J. Chem. Soc., Chem. Commun., 1983, 1228.
6 M. A. Beckett, J. E. Crook, N. N. Greenwood and J. D. Kennedy,
J. Chem. Soc., Dalton Trans., 1986, 1879.
X-Ray crystallography
Table 3 gives information for the structures of compounds 2,
3, 4, 6 and 7. Data for compound 4 were reported elsewhere,25
but are included here for convenience of reference. Data for
compound 2 were collected on a Stoe STADI 4 diffractome-
ter with a conventional sealed-tube X-ray source for Cu-Ka
˚
radiation (1.541 84 A). Data for compounds 3, 6 and 7 were
collected on a pre-Bruker Nonius KappaCCD diffractometer with
a conventional sealed-tube X-ray source for Mo-Ka radiation
This journal is
The Royal Society of Chemistry 2006
Dalton Trans., 2006, 3752–3765 | 3763
©

View Article Online
7 M. A. Beckett, N. N. Greenwood, J. D. Kennedy, P. A. Salter and M.
39 See, for example: R. A. Watson-Clark, K. Shelley and M. F. Hawthorne,
Inorg. Chem., 2000, 39, 1901, and references therein.
40 J. H. Hall, D. S. Marynick and W. N. Lipscomb, J. Am. Chem. Soc.,
1974, 96, 770.
41 S. K. Boocock, N. N. Greenwood and J. D. Kennedy, J. Chem. Res.
(S), 1981, 50.
Thornton-Pett, J. Chem. Soc., Chem. Commun., 1986, 556.
ˇ
8 T. Jel´ınek, J. D. Kennedy and B. St´ıbr, J. Chem. Soc., Chem. Commun.,
1994, 1415.
ˇ
9 P. Kaur, J. Holub, N. P. Rath, J. Bould, L. Barton, B. St´ıbr and J. D.
Kennedy, Chem. Commun., 1996, 273.
10 L. Barton, J. Bould, J. D. Kennedy and N. P. Rath, J. Chem. Soc., Dalton
Trans., 1996, 3145.
11 J. Bould, W. Clegg, J. D. Kennedy, S. J. Teat and M. Thornton-Pett,
J. Chem. Soc., Dalton Trans., 1997, 2005.
12 J. Bould, J. D. Kennedy, L. Barton and N. P. Rath, Chem. Commun.,
1997, 2405.
42 J. Bould, D. L. Ormsby, H.-J. Yao, C.-H. Hu, J. Sun, R.-S. Jin,
S. L. Shea, W. Clegg, T. Jel´ınek, N. P. Rath, M. Thornton-Pett,
ˇ
R. Greatrex, Pei-Ju Zheng, L. Barton, B. St´ıbr and J. D. Kennedy,
in Macropolyhedral Boron-containing Cluster Chemistry. Further
Progress beyond the Icosahedron. July 1999, in Contemporary Boron
Chemistry, ed. M. Davidson, A. K. Hughes, T. B. Marder and K. Wade,
Royal Society of Chemistry, Cambridge, UK, 2000, pp. 171–174.
43 J. Bould, U. Do¨rfler, M. Thornton-Pett and J. D. Kennedy, Inorg. Chem.
Commun., 2001, 4, 544, and references therein.
44 E. D. Jemmis and B. Kiran, Curr. Sci., 1994, 66, 766; Elemental Boron
Route to Stuffed Fullerenes, ch. 5 in Computational Chemistry. Reviews
of Current Trends, vol. 1, ed. J. Leszczynski, World Scientific, Singapore,
1996, pp. 175–184.
13 J. Bould, W. Clegg, J. D. Kennedy and S. J. Teat, J. Chem. Soc., Dalton
Trans., 1998, 2777.
14 J. Bould, S. A. Barrett, L. Barton, N. P. Rath and J. D. Kennedy, Inorg.
Chem. Commun., 1998, 1, 365.
15 M. G. S. Londesborough, E. J. MacLean, S. J. Teat, M. Thornton-Pett
and J. D. Kennedy, Chem. Commun., 2005, 1584.
ˇ
16 J. Plesˇek, S. Herˇma´nek, B. St´ıbr and F. Hanousek, Collect. Czech. Chem.
Commun., 1967, 32, 1095;; J. Dobson, P. C. Keller and R. Schaeffer,
Inorg. Chem., 1968, 7, 399.
45 E. D. Jemmis and M. M. Balakrishnarajan, Bull. Mater. Sci., 1999, 22,
863.
17 L. C. Ardini and T. P. Fehlner, Inorg. Chem., 1973, 12, 798.
18 J. Plesˇek, S. Herˇma´nek and F. Hanousek, Collect. Czech. Chem.
Commun., 1968, 33, 699; J. Plesˇek, S. Herˇma´nek and F. Hanousek,
Collect. Czech. Chem. Commun., 1968, 33, 2177.
19 W. E. Hill, F. A. Johnson and R. W. Novak, Inorg. Chem., 1975, 14,
1244.
46 S. L. Shea, J. Bould, M. G. S. Londesborough, S. D. Perera, A. Franken,
ˇ
D. L. Ormsby, T. Jel´ınek, B. St´ıbr, J. Holub, C. A. Kilner, M. Thornton-
Pett and J. D. Kennedy, Pure Appl. Chem., 2003, 75, 1239.
47 T. Peymann, C. B. Knobler and M. F. Hawthorne, J. Am. Chem. Soc.,
1999, 121, 5601; T. Peymann, C. B. Knobler and M. F. Hawthorne,
Chem. Commun., 1999, 2039, and references cited therein.
48 B. T. King, Z. Janousˇek, B. Gru¨ner, M. Trammel, B. C. Noll and J.
Michl, J. Am. Chem. Soc., 1996, 118, 3313, and references cited therein.
49 C.-W. Tsang and Z. Xie, Chem. Commun., 2000, 1839, and references
cited therein.
20 E. I. Tolpin, E. Mizusawa, D. S. Becker and J. Venzel, Inorg. Chem.,
1980, 19, 1182; E. Mizusawa, S. E. Rudnich and K. Erics, Inorg. Chem.,
1980, 19, 1188.
21 W. Knoth and E. L. Muetterties, J. Inorg. Nucl. Chem., 1961, 20, 66.
ˇ
22 B. St´ıbr, J. Plesˇek and S. Herˇma´nek, Collect. Czech. Chem. Commun.,
50 J. Bould, J. E. Crook, N. N. Greenwood, J. D. Kennedy and M.
Thornton-Pett, J. Chem. Soc., Dalton Trans., 1990, 1441, and references
therein.
1972, 37, 2696.
23 A. N. Bridges, D. R. Powell, J. A. Dopke, J. M. Desper and D. F. Gaines,
Inorg. Chem., 1998, 37, 503.
ˇ
51 B. St´ıbr, J. Plesˇek and A. Zoba´cˇova´, Polyhedron, 1982, 1, 824.
24 H. Beall and D. F. Gaines, Inorg. Chem., 1998, 37, 1420.
25 J. Bould, U. Doerfler, W. Clegg, S. J. Teat, M. Thornton-Pett and J. D.
Kennedy, Chem. Commun., 2001, 1788.
26 J. Bould, M. G. S. Londesborough, D. L. Ormsby, J. A. H. MacBride,
K. Wade, C. A. Kilner, W. Clegg, S. J. Teat, M. Thornton-Pett,
R. Greatrex and J. D. Kennedy, J. Organomet. Chem., 2002, 657,
256.
52 C. Naar-Colin and T. L. Heying, Inorg. Chem., 1963, 2, 659.
53 O. Volkov, W. Dirk, U. Englert and P. Paetzold, Reactions of the
undecaboranate anion [B11H11]2− in Contemporary Boron Chemistry,
ed. M. Davidson, A. K. Hughes, T. B. Marder and K. Wade, Royal
Society of Chemistry, Cambridge, UK, 2000, pp. 159–162; O. Volkov,
W. Dirk, U. Englert and P. Paetzold, Z. Anorg. Allg. Chem., 1999, 625,
1193.
27 A. R. Siedle, G. M. Bodner and L. J. Todd, J. Inorg. Nucl. Chem., 1971,
54 N. S. Hosmane, A. Franken, G. Zhang, R. R. Srivastava, R. Y. Smith
and B. F. Spielvogel, Main Group Met. Chem., 1998, 21, 319.
55 R. Schaeffer, J. Am. Chem. Soc., 1957, 79, 1006.
56 B. M. Graybill, J. K. Ruff and M. F. Hawthorne, J. Am. Chem. Soc.,
1961, 83, 2669.
57 G. Ferguson, J. D. Kennedy, X. L. R. Fontaine, Faridoon and T. R.
Spalding, J. Chem. Soc., Dalton Trans., 1988, 2555.
58 M. A. Beckett, M. Bown, X. L. R. Fontaine, N. N. Greenwood, J. D.
Kennedy and M. Thornton-Pett, J. Chem. Soc., Dalton Trans., 1988,
1969.
59 See, for example: D. Reed, Chem. Soc. Rev., 1993, 22, 109.
60 W. McFarlane, Proc. R. Soc. London, Ser. A, 1968, 306, 185.
61 Gaussian 92, Revision D.3, M. J. Frisch, G. W. Trucks, M. Head-
Gordon, P. M. W. Gill, M. W. Wong, J. B. Foresman, B. G. Johnson,
H. B. Schlegel, M. A. Robb, E. S. Replogle, R. Gomperts, J. L. Andres,
K. Raghavachari, J. S. Binkley, C. Gonzalez, R. L. Martin, D. J. Fox,
D. J. Defrees, J. Baker, J. J. P. Stewart and J. A. Pople, Gaussian, Inc.,
Pittsburgh, PA, 1992.
33, 3671.
28 L. G. Sneddon, J. C. Huffman, R. O. Schaeffer and W. E. Streib,
J. Chem. Soc., Chem. Commun., 1972, 474.
29 J. D. Kennedy, in Boron, ch. 8 in Multinuclear NMR, ed. J. Mason,
Plenum, New York and London, 1987, pp. 221–258, and references
cited therein.
ˇ
30 D. L. Ormsby, R. Greatrex, B. St´ıbr and J. D. Kennedy, J. Organomet.
Chem., 2000, 614/615, 61.
31 M. J. Carr, M. G. S. Londesborough, J. Bould, I. C´ısarˇova´ and J. D.
Kennedy, Collect. Czech. Chem. Commun., 2005, 70, 430.
32 J. Bould, W. Clegg and J. D. Kennedy, (D. M. P. Mingos Special
Edition), Inorg. Chim. Acta, 2006, 359, 3723.
33 S. K. Boocock, N. N. Greenwood, J. D. Kennedy and D. Taylorson,
J. Chem. Soc., Chem. Commun., 1979, 16; N. N. Greenwood, J. D.
Kennedy, W. S. McDonald, J. Staves and D. Taylorson, J. Chem. Soc.,
Chem. Commun., 1979, 17.
34 G. M. Brown, J. W. Pinson and L. L. Ingram, Inorg. Chem., 1979, 18,
1951.
62 W. Kutzelnigg, Isr. J. Chem., 1980, 19, 193.
35 S. K. Boocock, J. Chem. Soc., Dalton Trans., 1980, 790; S. K. Boocock,
N. N. Greenwood, J. D. Kennedy, W. S. McDonald and J. Staves,
J. Chem. Soc., Dalton Trans., 1980, 790; S. K. Boocock, N. N.
Greenwood, J. D. Kennedy, W. S. McDonald and J. Staves, J. Chem.
Soc., Dalton Trans., 1980, 790; S. K. Boocock, Y. M. Cheek, N. N.
Greenwood and J. D. Kennedy, J. Chem. Soc., Dalton Trans., 1981,
1430; S. A. Barrett, N. N. Greenwood, J. D. Kennedy and M. Thornton-
Pett, Polyhedron, 1985, 4, 1981; J. Bould, J. D. Kennedy, W. Clegg and
S. J. Teat, Acta Crystallogr., Sect. C, 2001, 57, 779.
36 D. E. Sands and A. Zalkin, Acta Crystallogr., 1962, 15, 410.
37 A. Tippe and W. C. Hamilton, Inorg. Chem., 1969, 8, 464.
38 S. G. Shore, in Nido and Arachno Boron Hydrides, ch. 3 in BoronHy-
dride Chemistry, ed. E. L. Muetterties, Academic Press, New York, San
Francisco and London, 1975, pp. 79–174.
63 M. Schindler and W. Kutzelnigg, J. Chem. Phys., 1982, 76, 1919.
64 W. Kutzelnigg, U. Fleischer and M. Schindler, NMR. Basic Principles
and Progress, Springer, Berlin and Heidelberg, 1990, vol. 23,
pp. 165 ff.
65 U. Meier, C. van Wu¨llen and M. Schindler, J. Comput. Chem., 1992,
13, 551.
66 DIGLO software, version 25.09.95, personal electronic communication
from C. van Wu¨llen.
67 Gaussian 98, Revision A.7, M. J. Frisch, G. W. Trucks, H. B. Schlegel,
G. E. Scuseria, M. A. Robb, J. R. Cheeseman, V. G. Zakrzewski, J. A.
Montgomery, Jr., 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,
3764 | Dalton Trans., 2006, 3752–3765
This journal is
The Royal Society of Chemistry 2006
©

View Article Online
Q. Cui, K. Morokuma, D. K. Malick, A. D. Rabuck, K. Raghavachari,
J. B. Foresman, J. Cioslowski, J. V. Ortiz, A. G. Baboul, B. B.
Stefanov, 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.
Johnson, W. Chen, M. W. Wong, J. L. Andres, M. Head-Gordon, E. S.
Replogle and J. A. Pople, Gaussian Inc., Pittsburgh, PA, 1998.
68 J. Bould, W. Clegg, J. D. Kennedy, S. J. Teat and M. Thornton-Pett,
J. Chem. Soc., Dalton Trans., 1997, 2005.
69 R. J. Cernik, W. Clegg, C. R. A. Catlow, G. Bushnell-Wye, J. V. Flaherty,
G. N. Greaves, I. Burrows, D. J. Taylor, S. J. Teat and M. Hamichi,
J. Synchrotron Radiat., 1997, 4, 279.
70 W. Clegg, M. R. J. Elsegood, S. J. Teat, C. Redshaw and V. C. Gibson,
J. Chem. Soc., Dalton Trans., 1998, 3037.
71 G. M. Sheldrick, SHELX97, University of Go¨ttingen, Germany, 1997;
G. M. Sheldrick, SHELXTL Manual, Bruker AXS Inc., Madison, WI,
USA, 1994, 1998.
72 L. J. Farrugia, ORTEP-3, J. Appl. Crystallogr., 1997, 30, 565.
This journal is
The Royal Society of Chemistry 2006
Dalton Trans., 2006, 3752–3765 | 3765
©