Polyhedral metallathiaborane chemistry: Synthesis and characterisation of metallathiaboranes based on the twelve-vertex icosahedral closo-{MSB10H10} unit, where M is Rh or Ir

Article abstract of DOI:10.1016/j.jorganchem.2007.11.014

The reaction of [nido-7-SB₁₀H₁₂] with [RhCl(PPh₃)₃] in the presence of N,N,N′N′-tetramethylnaphthalene-1,8-diamine (tmnd) in CH₂Cl₂ gives twelve-vertex [2,2-(PPh₃)₂-2-H-closo-2,1-RhSB₁₀H₁₀] (1) and eleven-vertex [8,8-(PPh₃)₂-nido-8,7-RhSB₉H₁₀] (2), as major products, plus the dimeric species [{(PPh₃)-closo-RhSB₁₀H₁₀}₂] (3) as a minor product. Reaction of 1 with PMe₂Ph in CH₂Cl₂ results in phosphine exchange and hydride substitution, affording the chloro analogue of 1, [2,2-(PMe₂Ph)₂-2-Cl-closo-2,1-RhSB₁₀H₁₀] (4). By contrast, reaction between [IrCl(PPh₃)₃] and [nido-7-SB₁₀H₁₂] in CH₂Cl₂ with tmnd affords only one product, twelve-vertex [2,2-(PPh₃)₂-2-H-closo-2,1-IrSB₁₀H₁₀] (5). [RhCl₂(η⁵-C₅Me₅)]₂ with [nido-7-SB₁₀H₁₂] under the same conditions gives twelve-vertex [2-(η⁵-C₅Me₅)-closo-2,1-RhSB₁₀H₁₀] (6). All the compounds are characterised by NMR spectroscopy, and by mass spectrometry, and the molecular structure of [2,2-(PMe₂Ph)₂-2-Cl-closo-2,1-RhSB₁₀H₁₀] (4) was established by single-crystal X-ray diffraction analysis. This last rhodathiaborane 4 is fluxional in solution through a process that involves a reversible partial rotation of the {RhCl(PMe₂Ph)₂} unit above the {SB₄} pentagonal face of the {SB₁₀H₁₀} fragment.

Full text of DOI:10.1016/j.jorganchem.2007.11.014

Available online at www.sciencedirect.com
Journal of Organometallic Chemistry 693 (2008) 435–445
www.elsevier.com/locate/jorganchem
Polyhedral metallathiaborane chemistry: Synthesis and characterisation
of metallathiaboranes based on the twelve-vertex icosahedral
closo-{MSB10H10} unit, where M is Rh or Ir
a,b,
a
c
d
Ramon Macıas *, Mark Thornton-Pett , Josef Holub , Trevor R. Spalding ,
´
´
d
c
a
ˇ
´
Yousafzai Faridoon , Bohumil Stıbr , John D. Kennedy
a School of Chemistry, University of Leeds, Leeds, West Yorkshire UK LS2 9JT, United Kingdom
b Instituto Universitario de Cata´lisis Homoge´nea, Facultad de Ciencias, Edificio D (4a planta), Pedro Cerbuna 12, 50009 Zaragoza, Spain
c
ˇ
Institute of Inorganic Chemistry of the Academy of Sciences of the Czech Republic, 25068 Rezˇ u Prahy, Czech Republic
d Department of Chemistry, University College Cork, National University of Ireland, Cork, Ireland
Received 4 September 2007; received in revised form 8 November 2007; accepted 9 November 2007
Available online 17 November 2007
Abstract
The reaction of [nido-7-SB10H12] with [RhCl(PPh3)3] in the presence of N,N,N0N0-tetramethylnaphthalene-1,8-diamine (tmnd) in
CH2Cl2 gives twelve-vertex [2,2-(PPh3)2-2-H-closo-2,1-RhSB10H10] (1) and eleven-vertex [8,8-(PPh3)2-nido-8,7-RhSB9H10] (2), as major
products, plus the dimeric species [{(PPh3)-closo-RhSB10H10}2] (3) as a minor product. Reaction of 1 with PMe2Ph in CH2Cl2 results
in phosphine exchange and hydride substitution, affording the chloro analogue of 1, [2,2-(PMe2Ph)2-2-Cl-closo-2,1-RhSB10H10] (4).
By contrast, reaction between [IrCl(PPh3)3] and [nido-7-SB10H12] in CH2Cl2 with tmnd affords only one product, twelve-vertex [2,2-
(PPh3)2-2-H-closo-2,1-IrSB10H10] (5). [RhCl2(g5-C5Me5)]2 with [nido-7-SB10H12] under the same conditions gives twelve-vertex [2-(g5-
C5Me5)-closo-2,1-RhSB10H10] (6). All the compounds are characterised by NMR spectroscopy, and by mass spectrometry, and the
molecular structure of [2,2-(PMe2Ph)2-2-Cl-closo-2,1-RhSB10H10] (4) was established by single-crystal X-ray diffraction analysis. This
last rhodathiaborane 4 is fluxional in solution through a process that involves a reversible partial rotation of the {RhCl(PMe2Ph)2} unit
above the {SB4} pentagonal face of the {SB10H10} fragment.
Ó 2007 Elsevier B.V. All rights reserved.
Keywords: Metallathiaborane; Polyhedral chemistry
1. Introduction
metallaheteroborane compounds prepared were ‘mixed-
sandwich’ compounds [(g5-C5H5)MC2B9H10], obtained
from three-component reactions between simple transi-
tion-element chlorides with Group 1 salts of the [g5-
C5H5]ꢀ hydrocarbon anion and the nido eleven-vertex
[C2B9H12]ꢀ or [C2B9H11]2ꢀ carbaborane anions [1–3].
Routes to {MEB10} species extend from this type of pro-
cess by the reactions of nido eleven-vertex anions based
on {EB10} units with transition-element complexes that
contain metal halide bonds, and this route has been the
one predominantly used for the synthesis of a wide range
of twelve-vertex metallaheteroborane species [13–16]. Here
we report the synthesis of some {MSB10} species starting
from the nido eleven-vertex {SB10} system.
Contiguous metallaheteroborane cluster compounds
based on the icosahedral twelve-vertex {2,1-MEB10H10}
unit (schematic cluster structure I) have been reported for
a variety of transition-elements M, where M is Cr, Mo,
W, Fe, Co, Ni, Pd, or Pt, and for a variety of heteroatoms
E, where E is C, Si, N, P, As, S, Se or Te [1–18]. The syn-
thetic procedures developed for the preparation of these
species are numerous. The first twelve-vertex icosahedral
*
Corresponding author.
´
E-mail address: rmacia@unizar.es (R. Macıas).
0022-328X/$ - see front matter Ó 2007 Elsevier B.V. All rights reserved.
doi:10.1016/j.jorganchem.2007.11.014

436
R. Macı´as et al. / Journal of Organometallic Chemistry 693 (2008) 435–445
ter compound 3 (schematic cluster structure V) was charac-
terised as such by comparison of its NMR properties with
those of the previously reported and crystallographically
determined [27] direct tellurium analogue [{(PPh3)-closo-
RhTeB10H10}2] (8) as well as by mass spectrometry. In
addition to this first-order synthesis work, and in an
attempt to effect phosphine–ligand replacement, we treated
the rhodathiaborane 1 with PMe2Ph in CH2Cl2, resulting
in the synthesis of the chloro species [2,2-(PMe2Ph)2-2-Cl-
closo-2,1-RhSB10H10] (4) (schematic cluster structure IV,
where X = Cl), isolated in 20% yield.
Eleven-vertex [nido-7-SB10H12] (schematic cluster struc-
ture II) was originally reported by Muetterties and co-
workers [19], and successively improved preparations have
subsequently been reported by Sneddon and co-workers
[20]. Although its metallathiaborane derivatives [2,2-
(PEt3)2-closo-2,1-PtSB10H10] [12] and [2,2-(PPh3)2-2-H-
closo-2,1-RhSB10H10] [15] were among the earliest
examples of reported twelve-vertex metallaheteroboranes,
the chemistry of {MSB10} cluster species has not been very
well developed compared with their isoelectronic metalladi-
carbaborane analogues. For example, there are only two
crystallographic studies of such closo twelve-vertex metalla-
thiaboranes: one is the ferrathiaborane [2-(g6-C6H5Me)-
closo-2,1-FeSB10H10] [18] and the other is the platinum
derivative [2,2-(PMe2Ph)2-closo-2,1-PtSB10H10] (referred
to below as compound 7) [21]. In the crystallographic anal-
ysis of [2-(g6-C6H5Me)-closo-2,1-FeSB10H10] the sulfur
atom was found to be disordered, precluding accurate
structural comparisons with analogous species. Moreover,
available NMR data on previous metallathiaboranes are
largely based on low-dispersion spectroscopy with little
assignment work. Therefore, as part of our general interest
in heteroboranes [22,23] and their metal derivatives
[11,13,14,16,21,24–29], herein we report synthetic, struc-
tural and NMR studies on some new twelve-vertex rho-
dium and iridium metallathiaboranes incorporating the
{SB10H10} ligand fragment.
The 100-mg-scale reaction of the iridium(I) complex
[IrCl(PPh3)3] under the same conditions as those used in
the preparation of the rhodium compound 1 above gave
only one predominant metallaheteroborane product, read-
ily identified by NMR spectroscopy and mass spectrometry
as the iridium analogue of the rhodium compound 1, i.e. as
[2,2-(PPh3)2-2-H-closo-2,1-IrSB10H10] (5), isolated in 12%
yield (Eq. (1)).
SB10H12 þ ½IrClðPPh3Þ3ꢁ þ tmnd
! ½ðPPh3Þ HIrSB10H10ꢁ5 þ ½tmndHꢁþ þ Clꢀ þ PPh3
2
ð1Þ
The pentamethylcyclopentadienyl rhodium analogue of 1,
i.e. [2-(g5-C5Me5)-closo-2,1-RhSB10H10] (6), was similarly
obtained in 25% yield from the reaction between
[RhCl2(g5-C5Me5)]2 and [nido-7-SB10H12] in the presence
of tmnd in CH2Cl2 solution at ambient temperature (Eq.
(2)). This colourless rhodathiaborane was also purified by
preparative TLC and characterised by NMR spectroscopy
and mass spectrometry.
2. Results and discussion
The rhodium(I) complex [RhCl(PPh3)3] was added to
[nido-7-SB10H12] in the presence of the non-nucleophilic
base N,N,N0,N0-tetramethylnaphthalene-1,8-diamine (tmnd)
in CH2Cl2 solution at room temperature. The resulting
reaction gave several products. After repeated prepara-
tive-scale thin-layer chromatography (TLC), we were able
to isolate and characterise three major components: yellow
[2,2-(PPh3)2-2-H-closo-2,1-RhSB10H10] (1) (26% yield),
orange [8,8-(PPh3)2-nido-8,7-RhSB9H10] (2) (17% yield)
and purple [{(PPh3)-closo-RhSB10H10}2] (3) (5% yield).
The nido eleven-vertex compound 2 (schematic cluster
structure III) was previously well characterised [25], and
is of a recognised structural type [28,29], but the character-
isation of the twelve-vertex closo compound 1 (schematic
cluster structure IV, where X = H), though previously
mentioned in the literature [15], was incomplete. To better
characterise compound 1, as part of this present work
NMR data for 1 have been gathered at higher field, and
are now assigned, and compared with reported data for
the other closo twelve-vertex species. The new double-clus-
2SB10H12 þ ½RhCl2ðC5Me5Þꢁ2 þ 4tmnd
! 2½ðC5Me5ÞRhSB10H10ꢁ6 þ 4½tmndHꢁþ þ 4Clꢀ
ð2Þ
The molecular structures of the metallathiaboranes in solu-
1
tion were readily established by means of 11B, 31P and H
multinuclear and multiple resonance NMR spectroscopy.
The data are summarised in Tables 1–3. For the four mono-
meric closo species [2,2-(PPh3)2-2-H-closo-2,1-RhSB10H10]
(1), [2,2-(PMe2Ph)2-2-Cl-closo-2,1-RhSB10H10] (4), [2,2-
(PPh3)2-2-H-closo-2,1-IrSB10H10] (5) and [2-(g5-C5Me5)-
1
closo-2,1-RhSB10H10] (6), the 11B and H resonances were
assigned on the reasoned basis of relative intensities and
1H–{11B(selective)} experiments, and thence by comparison
with other previously assigned closo twelve-vertex

R. Macı´as et al. / Journal of Organometallic Chemistry 693 (2008) 435–445
437
Table 1
11B and 1H NMR data for [2,2-(PPh3)2-2-H-closo-2,1-RhSB10H10] (1), [2,2-(PMe2Ph)2-2-Cl-closo-2,1-RhSB10H10] (4), [2,2-(PPh3)2-H-closo-2,1-IrSB10H10
(5) and [2-(g5-C5Me5)-closo-2,1-RhSB10H10] (6) at 284–297 K, together with those of [2,2-(PMe2Ph)2-2,1-PtSB10H10] (7) for comparison
]
Assignmenta
1b,d
4b
5c
6b,e,f
7b
d(11B)
d (1H)
d(11B)
d(1H)
d(11B)
d(1H)
d(11B)
d(1H)
d(11B)
d(1H)
12
9
(1B)
(1B)
(2B)
(2B)
(2B)
(2B)
(MH)
+14.2
+13.8
+7.7
+4.29
+3.55
+2.10
+3.51
+1.85
+1.79
ꢀ7.13i
+14.9
+13.4
+13.0
+9.3
+4.35
+3.00
+2.88
+3.59
+2.08
+1.75
+12.0
+13.2
+4.6
+4.03
+6.04
+2.28
+3.57
+2.02
+2.00
ꢀ8.93j
+9.1
+5.3h
+5.3h
+14.1
ꢀ18.1
ꢀ10.5
+3.32
+4.18
+2.23
+2.99
+1.86
+2.14
+15.6
+1.6
+3.70g
+5.07
+1.18
+3.14
+2.42
+1.59
3,6
7,11
4,5
8,10
2
ꢀ9.1
+7.4
+1.6
+3.3
ꢀ13.3
ꢀ10.3
ꢀ16.0
ꢀ13.6
ꢀ13.0
ꢀ14.2
ꢀ19.4
ꢀ22.5
a
See text.
b
CD2Cl2 solution, data from O’Connell et al. (Ref. [21]).
CDCl3 solution.
Mass spec.: m/zmax: 780;
Mass spec.: m/zmax: 388;
d(1H)(g5-C5Me5) +2.03.
4J(195Pt–1H) 62 Hz.
c
d
e
f
12
11
C
C
1H4131P2 32S103Rh requires 780.
1H25
B
36
10
10
12
11
B
32S103Rh requires 390.
10
g
h
i
Accidental coincidence of resonances.
Triplet [2 J(31P–1H) ca. 22 Hz] of doublets [1J(103Rh–1H) ca. 19 Hz].
Triplet [2J(31P–1H) ca. 23.5 Hz].
j
Table 2
Table 3
31P and additional 1H NMR data for [2,2-(PPh3)2-2-H-closo-2,1-
RhSB10H10] (1), [2,2-(PMe2Ph)2-2-Cl-closo-2,1-RhSB10H10] (4) and [2,2-
(PPh3)2-H-closo-2,1-IrSB10H10] (5)
NMR data for [{(PPh3)-closo-RhSB10H10}2] (3) and [{(PPh3)2RhTeB10
H10}2] (8)
-
Assignments
Compound 3c,g
d(11B)a
Compound 8d
d(11B)b
1a
4b
5c
d(1H)
d(1H)
d(31P)/ppm
+31.5
A +12.9d
+9.2
11 (1B)
+23.5
+11.7
+8.3
+3.2
ca. +10.0
ꢀ6.39e
+2.19
+3.91
+2.95
+2.68
+26.5
+17.0
+9.0
+6.0
ca. +11.1
ꢀ7.28f
B ꢀ0.8
+3.8
1J(103Rh–31P)/Hz
1J(103Rh–31PA)/Hz
1J(103Rh–31PB)/Hz
2J(31PA–31PB)/Hz
+106.2
7,9,12 (3B)
+4.66
+4.47
+3.02
121
95
37.5
3,6 (2B)
a
ca. +1.0
ꢀ8.5
+1.52
+1.97
+0.8
ꢀ9.7
+2.02
+2.75
CDCl3 solution, 220 K.
b
CD2Cl2 solution, 176 K; additional 1H NMR data: (CDCl3, 293 K)
d(1H)(CH3)A +1.90, [AX3]2 system, N = 2J(31P–1H) + 4J(31P–1H) =
10.4 Hz; d(1H)(CH3)B +1.52, [AX3]2 system, N = 2J(31P–1H) +
4J(31P–1H) = 11.3 Hz.
8,10 (2B)
ꢀ15.4
+1.53
+1.76
ꢀ16.0
+2.41
+1.92
ca. ꢀ13.5
ca ꢀ14.5
c
4,5 (2B)
CDCl3 solution, 223 K.
At higher temperatures these signals coalesce into one resonance
d
ꢀ19.3
+1.97
ca. ꢀ17.7
+1.88
[d(31P) +5.0, 1J(103Rh–31P) 110 Hz at 291 K]; coalescence temperature
a
CDCl3 at 294–303 K.
213 K at 100 MHz.
b
c
CD2Cl2 at 294–303 K.
d(31P) +24.8 ppm, 1J(103Rh–31P) 134 Hz, CDCl3 at 222 K.
d
Data from Ferguson, Lough et al. (Ref. [26]); additionally d(31P) +
29.1 ppm, 1J(103Rh–31P) 97 Hz, CD2Cl2 at 213 K.
e
{MEB10H10} species [13,14,17]. For each species, the cluster
atoms show 1:1:1:2:2:2 relative-intensity patterns in both
their 11B and 1H NMR spectra, and the absence of bridging
hydrogen atoms on the borane cage, which is consistent
with twelve-vertex closo clusters that have Cs symmetry in
Apparent 1:2:1 triplet [1J(103Rh–1H) + 2J(31P–1H)]/2 ca. 17 Hz (mean
value).
f
Apparent 1:2:1 triplet [1J(103Rh–1H) + 2J(31P–1H)]/2 ca. 17 Hz (mean
value).
g
1
11
Mass spec.: m/zmax: 1032; 12C36 H50
B
31P232S2103Rh2 requires 1034.
20
1
solution at room temperature. In addition, the H NMR
spectrum of 1 exhibits a triplet of doublets at d(1H) ꢀ7.13
and that of 5 exhibits a triplet at d(1H) ꢀ8.93, these reso-
nances corresponding to the exo-polyhedral hydride ligands
on the rhodium and iridium metal centres, respectively. The
1H spectrum of 4 at room temperature contains two
N(31P–1H) doublets centred at d +1.90 and +1.52, which
are assigned to the methyl protons of the PMe2Ph groups.
The 1H NMR spectrum of 6 exhibits a resonance of relative
intensity 15 at d +2.03 ppm, demonstrating the presence of
a {g5-C5Me5} fragment bound to the rhodium atom. The
corresponding datum for
[2-(g5-C5Me5)-closo-2,1-
RhTeB10H10] (9), the tellurium analogue of 6, for which
an X-ray diffraction study has been reported, is d(1H)
+2.01 ppm [13].
The NMR data of the dimer [{2-(PPh3)-closo-2,1-
RhSB10H10}2] (3) are likewise analogous to those for its
previously reported and crystallographically determined

438
R. Macı´as et al. / Journal of Organometallic Chemistry 693 (2008) 435–445
group-16
tellurium
congener
[{2-(PPh3)-closo-2,1-
gap between these higher-field and lower-field groupings
RhTeB10H10}2] (8) [26]. Each of 3 and 8 shows a triplet
in the high-field region between d(1H) ꢀ6 and ꢀ8 in its
1H NMR spectra (Table 3), which in each case is readily
assigned to the B–H–Rh bridging hydrogen atom position
(schematic cluster V above). The formulation of 3 is addi-
tionally supported by mass spectrometry that shows a
molecular envelope that is centred at m/e 1032 and that
matches the isotope pattern of an ion of formulation
[C36H50B20P2S2Rh2]+.
is markedly larger for the rhodium and iridium compounds
1, 4 and 6 (Fig. 1), with an average gap of ca. 20 ppm. By
contrast, the resonances of the {Pt(PMe2Ph)} species 7 are
much more closely grouped, and more evenly distributed
throughout the spectrum, although maintaining a similar
sequence of relative shielding. This difference, between 7
on one hand, and 1, 4 and 6 on the other hand, is probably
a consequence of differences in the nature of the metal-to-
cluster bonding as discussed below (compare also sche-
matic cluster structure IV with VII). The 11B(12) resonance
that is antipodal to the sulfur atom appears at lowest
shielding in all compounds, except for the {Rh(g5-
The 11B shielding pattern of the new twelve-vertex com-
pounds can be traced as in Fig. 1 to those of the thiaborane
starting material, [nido-7-SB10H12], and the previously
reported [21] closo twelve-vertex platinathiaborane, [2,2-
(PMe2Ph)2-2,1-PtSB10H10] (7). In all these closo metallathi-
aboranes, the groupings of the 11B(9), 11B(12), 11B(3,6) and
11B(7,11) resonances tend to be less shielded than the
groupings of the 11B(4,5) and 11B(8,10) resonances. The
1
C5Me5)} derivative 6. The H and 11B nuclear shielding
of the {BH(exo)} units follow a line of approximate slope
d(11B):d(1H) 11:1, which is in accord with previously
reported closo twelve-vertex {MEB10} relative shielding pat-
terns [16,26]. Noteworthy in this context is the exceptional
Fig. 1. Stick representations of the resonance positions and relative intensities in the 11B–{1H} NMR spectra of (a) [nido-7-SB10H12], (b) [2,2-(PMe2Ph)2-
2,1-PtSB10H10] (7) (data from O’Connell et al. [21]), (c) [2,2-(PMe2Ph)2-2-Cl-closo-2,1-RhSB10H10] (4), (d) [2-(g5-C5Me5)-closo-2,1-RhSB10H10] (6) and
(e) the [nido-7-SB10H12
vertex closo clusters differ.)
]
ꢀ anion. Hatched lines join equivalent positions. (Note that the numbering conventions for the eleven-vertex nido and closo twelve-

R. Macı´as et al. / Journal of Organometallic Chemistry 693 (2008) 435–445
439
1
lower H shielding for the BH(9) position when the antip-
ently greater stability of the nido eleven-vertex
{RhSB10H11} unit that has one bridging hydrogen atom
as opposed to less stable closo eleven-vertex configuration
VI. This preferential stability is also found in the eleven-
vertex rhodaazaborane [8,8-(PPh3)2-nido-8,7-RhNB9H11]
[29b], and in related manner in [8-(g4-HC5Me5)-nido-8,7-
RhSB9H9-l-9,10-(SMe)] which also adopts the nido config-
uration with a rhodium(I) centre (incidentally forcing the
unusual tetrahapto g4-HC5Me5 cyclopentadiene ligand)
rather than an alternative closed structure with a formal
rhodium(III) oxidation state and a more usual pentahapto
g5-C5Me5 cyclopentadienyl ligand [29a].
odal atom is iridium or platinum. This effect is previously
noted as an antipodal H shielding characteristic of closo
twelve-vertex compounds [13,14,16], and indeed other clus-
ter types [30], that contain third-row transition-element
centres as cluster constituents.
1
The reactions that produce compounds [2,2-(PPh3)2-2-
H-closo-2,1-RhSB10H10] (1), [2,2-(PPh3)2-2-H-closo-2,1-
IrSB10H10] (5) and [2-(g5-C5Me5)-closo-2,1-RhSB10H10]
(6) are presumably initiated by a deprotonation of [nido-
7-SB10H12] by the non-nucleophilic strong proton-abstract-
ing base tmnd, affording the [nido-7-SB10H11]ꢀ anion in
solution. Subsequent addition of the metal halide com-
plexes would then result in displacement of halide by the
nido-7-thiaborane anion to form intermediate metallathi-
aborane species. In the synthesis of the {Rh(C5Me5)} spe-
cies 6, the intermediate, of presumed formulation
{(C5Me5)ClRh(SB10H11)} with a low thiaborane-to-metal
coordination number, would then further eliminate HCl,
of which the H+ would become sequestered by tmnd in a
simple salt elimination, resulting in the final intimate pen-
tahapto coordination of the thiaborane ligand to the metal
centre to give the final [(C5Me5)RhSB10H10] formulation
for 6; overall the rhodium atom would maintain its initial
rhodium(III) oxidation state. In the formation of 1 and
5, by contrast, the intermediate would now be of presumed
formulation {(PPh3)2M(SB10H11)}, and the metal centre
would then add a hydrogen atom from the {SB10H11} unit
into its coordination sphere, resulting in a formal overall
oxidative addition of the metal centre into the cluster to
give the final [(PPh3)2HIrSB10H10] formulation. Thus, in
the formation of 1 and 5, the formal oxidation states of
the metal-atoms change from rhodium(I) and iridium(I)
in the starting [MCl(PPh3)2] complexes to rhodium(III)
and iridium(III) in the metallathiaborane products.
The formation of [8,8-(PPh3)2-nido-8,7-RhSB9H10] (2)
would appear to be more complicated. It may be possible
that the tmnd base causes the abstraction of a {BH} unit
from either [nido-7-SB10H12] or its [nido-7-SB10H11]ꢀ
monoanion and the resulting species, i.e. either neutral
[nido-7-SB9H11] or the [nido-7-SB9H10]ꢀ anion, reacts with
[RhCl(PPh3)3] to give a [(PPh3)2RhSB9H10] formulation for
2 directly. Certainly the ready formation of nido ten-vertex
species from real or incipient eleven-vertex species is a
well-recognised phenomenon in metal-atom-containing
borane cluster chemistry [31,32]. However, it is not clear
why the reaction of the corresponding iridium complex
[IrCl(PPh3)3] does not give an analogous [8,8-(PPh3)2-
nido-8,7-IrSB9H10] by-product under otherwise identical
conditions. Also, in view of the closed structures of the
twelve-vertex [2,2-(PPh3)2-2-X-closo-2,1-RhSB10H10] spe-
cies 1 and 4, it is of interest that this nido eleven-vertex spe-
cies 2 (schematic cluster structure III), a formal rhodium(I)
species [25], does not adopt the isomeric closed configura-
tion [1,1-(PPh3)2-1-H-closo-1,2-RhSB9H9] that would have
a rhodium(III) formulation (schematic cluster structure
VI). As recently published [29b], this arises from an inher-
The mechanism of formation of the purple dimeric spe-
cies [{2-(PPh3)-closo-2,1-RhSB10H10}2] (3) is also not clear.
A spontaneous intermolecular reaction of the rhodathiabo-
rane [2,2-(PPh3)2-2-H-closo-2,1-RhSB10H10] (1) is a possi-
ble route, although no double-cluster products were
observed when a sample of 1 was stirred in a CH2Cl2 solu-
tion at room temperature. In this context, it may be noted
that the group-16 tellurium congener [2,2-(PPh3)2-2-H-
closo-2,1-RhTeB10H10] (10) also did not give any bimetallic
rhodatelluraborane products when it was stirred in
CH2Cl2, C6H6 or thf solutions at room temperature, but
that very low yields of two purple compounds were isolat-
able from solutions of 10 that had undergone the more
forcing conditions of heating at reflux temperature in tolu-
ene or xylene for two days. An initial characterisation of
one of these two purple rhodatelluraborane products by
1
11B and H NMR spectroscopy has suggested the presence
1
of twenty {BH} units including two that exhibited H res-
onances at d(1H) ca. ꢀ8.0 and ꢀ7.0 ppm, respectively [33],
which correspond to Rh–H–B bridges and which compare
1
to the H resonance at d(1H) ꢀ6.39 observed for 3. Here it
can be noted that the molecular structures of two other
purple tellurium analogues, specifically the direct analogue
of 3, viz. [{2-(PPh3)-closo-2,1-RhTeB10H10}2] (8), and the
closely related mixed-ligand species [{2-(PPh3)-closo-2,1-
RhTeB10H10}{2-(CO)-closo-2,1-RhTeB10H10}] (11), have
been established by single-crystal X-ray diffraction meth-
ods [26]. These compounds were obtained from the reac-
tions of [2,2-(PPh3)2-2-H-closo-2,1-RhTeB10H10] (10) with
metal carbonyl complexes, i.e. either with [Mo(CO)6] to
give 8 (reaction in thf at ambient temperature for eighteen
hours) or with [Os3H2(CO)10] to give 11 (reaction in
CH2Cl2 heated at reflux for three days) [26]. Both com-
pounds 8 and 11 contained Rh–H–B bridges, exhibiting

440
R. Macı´as et al. / Journal of Organometallic Chemistry 693 (2008) 435–445
1H resonances at d(1H) ꢀ7.28 for the symmetrical
{Rh2(PPh3)2} species 8, and at d(1H) ꢀ7.65 and ꢀ6.39
for the asymmetric {Rh2(CO)(PPh3)} species 11, again
comparable to the d(1H) ꢀ6.39 value observed for 3.
Hence, a likely mechanism for the formation of 3 may
involve the thermolytic cleavage of a PPh3 ligand from 1
and the dimerisation of the ‘‘unsaturated” {(PPh3)H-
closo-RhSB10H10} intermediate species so formed, con-
comitant with a further elimination of dihydrogen to give
the stoichiometry of the observed product. The reaction
conditions used in the present study that here gave [{2-
(PPh3)-closo-2,1-RhSB10H10}2] (3) in 5% yield were not as
forcing as those used to prepare the {(RhTeB10H10)2} clus-
ter compounds 8 and 11, which were isolated in yields of
46% and 33%, respectively.
Table 4
˚
Selected interatomic distances (A) for [2,2-(PMe2Ph)2-2-Cl-closo-2,1-
RhSB10H10 (4) together with those for [2,2-(PMe2Ph)2-closo-2,1-
PtSB10H10] (7) (data from O’Connell et al. [21]) for comparison, with
]
estimated standard deviations (e.s.d.s) in parentheses
(a) From the metal-atom
Compound 4
Rh(2)–B(3)
Rh(2)–B(6)
Rh(2)–B(7)
Rh(2)–B(11)
Rh(2)–S(1)
Rh(2)–P(1)
Rh(2)–P(2)
Rh(2)–Cl(1)
Compound 7
Pt(2)–B(3)
Pt(2)–B(6)
Pt(2)–B(7)
Pt(2)–B(11)
Pt(2)–S(1)
Pt(2)–P(1)
Pt(2)–P(2)
2.269(2)
2.244(2)
2.258(2)
2.232(2)
2.3629(5)
2.3992(5)
2.3286(5)
2.4179(5)
2.318(3)
2.296(3)
2.230(3)
2.248(3)
2.6633(7)
2.2870(7)
2.3021(7)
(b) Sulfur–boron
Compound 4
S(1)–B(3)
S(1)–B(4)
S(1)–B(5)
Compound 7
S(1)–B(3)
S(1)–B(4)
S(1)–B(5)
S(1)–B(6)
2.101(2)
1.993(2)
1.993(2)
2.098(2)
2.034(3)
1.979(3)
1.973(4)
2.060(3)
The molecular structure of 4 was confirmed by a single-
crystal X-ray diffraction analysis (Fig. 2 and Tables 4 and
5). The crystals were monoclinic, space group P21/c, with
S(1)–B(6)
(c) Boron–boron
Compound 4
B(3)–B(4) (longest)
B(3)–B(7)
B(4)–B(5)
B(5)–B(6)
B(5)–B(9)
B(6)–B(11)
B(7)–B(11)
B(6)–B(10) (shortest)
Compound 7
B(3)–B(4)
B(3)–B(7)
1.920(3)
1.787(3)
1.851(3)
1.913(3)
1.764(3)
1.787(3)
1.837(3)
1.747(3)
1.919(5)
1.848(4)
1.840(5)
1.938(5)
1.750(5)
1.836(4)
1.821(4)
1.821(4)
B(4)–B(5)
B(5)–B(6) (longest)
B(5)–B(9) (shortest)
B(6)–B(11)
B(7)–B(11)
B(6)–B(1)
four molecules per unit cell. The compound is based on a
closed icosahedral {RhSB10H10} cluster unit. There is a
pentahapto interaction of the {SB4} face of the {SB10H10}
cluster fragment with the rhodium centre, and interboron
distances are within normal metallathiaborane ranges
[13,14,16]. The interatomic distances and angles within
the coordination sphere of the rhodium centre of 4 are of
interest when compared to those for platinum [21] in the
[2,2-(PMe2Ph)2-2,1-PtSB10H10] species 7. Thus, whereas
˚
the rhodium–sulfur distance of 2.3629(5) A in 4 is within
the bonding range found in other polyhedral rhodathiabo-
rane species [27–29], it is markedly shorter than the
˚
corresponding platinum–sulfur distance of 2.6633(7) A in
7. Thus, in the platinum compound 7, there is a significant
displacement of the platinum centre away from the sulfur
atom, which generates a distortion (schematic VII) from
closed icosahedral geometry as in schematic I. This
distortion is towards a more open ‘isonido’ type of geome-
try [34,35] that would have a four-membered {PtBSB}
open face: the metal centre thereby exhibits more of a tet-
rahapto g4 interaction with the {SB10H10} unit, rather than
g5pentahapto. In twelve-vertex closo-type 3,2,1-platinadi-
carbaboranes a related phenomenon generates a ‘slipped-
closo’ configuration as in schematic VIII [21,30,36,37].
Here, the ‘slipped-closo’ configuration VIII describes a dis-
tortion towards trihapto g3 and a nido-type twelve-vertex
configuration with a five-membered {PtBCCB} open face,
Fig. 2. ORTEP-3 type drawings [51] for [2,2-(PMe2Ph)2-2-Cl-closo-2,1-
RhSB10H10] (4) showing (a) the numbering scheme, and (b) the confor-
mation of the metal centre.

R. Macı´as et al. / Journal of Organometallic Chemistry 693 (2008) 435–445
441
Table 5
this ideal. As also mentioned above, there are significant
differences in the 11B nuclear shielding patterns when the
two compounds are compared (Fig. 1). A treatment of
these compounds from a metal-orientated viewpoint is per-
haps helpful to describe and rationalize the bonding differ-
ences implied by these geometrical and NMR differences.
In accord with such an approach, the rhodathiaborane 4
could thence be regarded as an eighteen-electron rho-
dium(III) complex between a formal {RhCl(PMe2Ph)2}2+
unit and a formal {SB10H10}2ꢀ fragment, the latter acting
as a tridentante pentahapto g5 ligand by means of its
{SB4} open face. By contrast, compound 7 could be
described as a sixteen-electron platinum(II) complex in
which a formal {Pt(PMe2Ph)2}2+ fragment is bound to a
{SB10H10}2ꢀ fragment that acts as a bidentate tetrahapto
g4 ligand by means of the {B4} unit of its {SB4} open face.
To examine these differences further, a simplistic frontier
orbital approach to the orbital interactions of a metal unit
with a thiaborane ligand unit can be adopted. Schematics X
and XI, respectively, represent the HOMO and SHOMO
orbitals of the formal {SB10H10}2ꢀ ligand, the size of the
circles approximating to the percentage components of
AO’s on those atomic positions. The metal-based molecu-
lar orbitals are LUMOs, thus the symmetry-determined
overlaps of these HOMO and SHOMO frontier orbitals
with the relevant LUMOs of the {RhCl(PMe2Ph)2}2+
and {Pt(PMe2Ph)2}2+ units would be expected to lead to
stronger overall interactions between the {RhCl-
(PMe2Ph)2}2+ unit and the {SB4} face of the {SB10H10}2ꢀ
ligand (schematics XII and XIII) than between the
{Pt(PMe2Ph)2}2+ unit and the {SB10H10}2ꢀ ligand (sche-
matics XIV and XV). This treatment of course ignores
any change in the energies of interacting orbitals as the
units bond. However, bearing in mind the that the angles
made by the non-cluster ligands about the rhodium atom
are close to 90°, it may be considered that three of the
atomic orbitals on the rhodium-based unit are involved
in bonding to the phosphine and chlorine ligands and three
others are symmetry-arranged to be directed towards the
{SB10H10}2ꢀ cluster ligand HOMO and SHOMO: this
interaction would give the rhodium centre a pseudo-octa-
hedral configuration, and result in increased electron
density along the Rh–S vector on cluster bonding. The
resulting configuration of the {RhCl(PMe2Ph)2} unit
would thence be as that observed in the solid-state struc-
ture. On the other hand, the {Pt(PMe2Ph)2}2+ fragment
does not have any suitably directed unfilled molecular orbi-
tal for overlap with the SHOMO of the {SB10H10}2ꢀ frag-
ment, and so the most efficient cluster bonding does not
involve a major interaction with the sulfur atom. Thus,
the platinum fragment in 7 is observed to engender a
‘slipped structure’ in which the platinum centre has a
pseudo-square-planar configuration. Such bonding
requirements of the {Pt(PMe2Ph)2}2+ and related frag-
ments, and how they may produce ostensibly anomalous
cluster structures and cluster electron counts, have been
discussed in adequate detail elsewhere [35,39].
Selected angles (°) between interatomic vectors for [2,2-(PMe2Ph)2-2-Cl-
closo-2,1-RhSB10H10] (4), together with those for [2,2-(PMe2Ph)2-closo-
2,1-PtSB10H10] (7) for comparison, with e.s.d.s in parentheses
(a) About the metal-atom
Compound 4
Compound 7
B(11)–Rh(2)–P(2)
B(6)–Rh(2)–P(2)
B(7)–Rh(2)–P(2)
B(3)–Rh(2)–P(2)
B(11)–Rh(2)–S(1)
B(6)–Rh(2)–S(1)
B(7)–Rh(2)–S(1)
B(3)–Rh(2)–S(1)
P(2)–Rh(2)–S(1)
B(11)–Rh(2)–P(1)
B(6)–Rh(2)–P(1)
B(7)–Rh(2)–P(1)
B(3)–Rh(2)–P(1)
P(2)–Rh(2)–P(1)
S(1)–Rh(2)–P(1)
P(2)–Rh(2)–Cl(1)
S(1)–Rh(2)–Cl(1)
P(1)–Rh(2)–Cl(1)
89.90(6)
131.56(6)
80.82(6)
112.29(6)
88.82(6)
54.11(6)
88.10(6)
53.91(6)
166.20(2)
109.75(6)
81.87(6)
157.71(6)
147.92(6)
97.37(2)
96.00(2)
84.83(2)
93.01(2)
84.67(2)
B(11)–Pt(2)–P(2)
B(6)–Pt(2)–P(2)
B(7)–Pt(2)–P(2)
B(3)–Pt(2)- P(2)
B(11)–Pt(2)–S(1)
B(6)–Pt(2)–S(1)
B(7)–Pt(2)–S(1)
B(3)–Pt(2)–S(1)
P(2)–Pt(2)–S(1)
B(11)–Pt(2)–P(1)
B(6)–Pt(2)–P(1)
B(7)–Pt(2)–P(1)
B(3)–Pt(2)–P(1)
P(2)–Pt(2)–P(1)
S(1)–Pt(2)- P(1)
106.70(8)
88.45(8)
151.82(8)
154.77(8)
84.26(8)
48.39(8)
83.81(8)
47.58(8)
108.50(3)
137.76(8)
170.87(8)
98.90(8)
93.54(8)
95.46(3)
122.48(2)
(b) About the sulfur atom
Compound 4
Compound 7
B(5)–S(1)–Rh(2)
B(4)–S(1)–Rh(2)
B(6)–S(1)–Rh(2)
B(3)–S(1)–Rh(2)
B(5)–S(1)–B(4)
B(5)–S(1)–B(6)
B(4)–S(1)–B(6)
B(5)–S(1)–B(3)
B(4)–S(1)–B(3)
B(6)–S(1)–B(3)
108.48(7)
B(5)–S(1)–Pt(2)
B(4)–S(1)–Pt(2)
B(6)–S(1)–Pt(2)
B(3)–S(1)–Pt(2)
B(5)–S(1)–B(4)
B(5)–S(1)–B(6)
B(4)–S(1)–B(6)
B(5)–S(1)–B(3)
B(4)–S(1)–B(3)
B(6)–S(1)–B(3)
106.01(10)
106.24(10)
56.44(9)
57.28(9)
55.5(2)
57.39(13)
97.53(14)
97.53(14)
57.11(14)
92.39(13)
108.90(7)
60.07(6)
60.76(6)
55.36(9)
55.71(9)
97.09(9)
97.32(9)
55.86(9)
95.51(8)
although it has been shown that the isonido-type tetrahapto
g4 configurations such as IX are energetically very similar
[21,30].
According to a simplistic application of cluster-geome-
try/electron-counting formalisms [38], [2,2-(PMe2Ph)2-2-
Cl-closo-2,1-RhSB10H10] (4) and [2,2-(PMe2Ph)2-2,1-
PtSB10H10] (7) could both be regarded as 26-electron
systems for which an icosahedral closo twelve-vertex struc-
ture would be predicted. However, the geometrical consid-
erations discussed above suggest that the bonding
interactions between the metal units and the thiaborane
residues in 4 and 7 are somewhat different from each other.
Thus, the twelve-vertex cluster unit of the rhodium com-
pound 4 is clearly icosahedral, whereas the platinum com-
pound 7 exhibits a marked and significant distortion from

442
R. Macı´as et al. / Journal of Organometallic Chemistry 693 (2008) 435–445
As mentioned above, the metal-to-thiaborane bonding
which exhibits two doublets of doublets. However, these
two 31P signals coalesce into one resonance at 213 K
(36.4 MHz spectrum), demonstrating that the metal-atom
environment of 4 is stereochemically non-rigid. We attri-
bute this fluxional behaviour to a reversible metal-to-hete-
roborane partial rotation that equilibrates the two
enantiomeric conformations of the metal centre above the
{SB4} face (schematics XVI and XVII). Variable-tempera-
ture NMR experiments give an activation energy DGà for
this fluxional process of 41 kJ molꢀ1 at 213 K.
differences between the rhodium and platinum metallathi-
aborane species are additionally manifested in their 11B
NMR shielding patterns (Fig. 1), which show that the res-
onances for the boron atoms flanking the metal–sulfur
linkage, i.e. for B(3) and B(6), shift about 17 ppm towards
higher field on changing the metal-containing fragment
from octahedral rhodium(III) or iridium(III) to square-pla-
nar platinum(II). The general 11B shielding pattern of [2,2-
(PMe2Ph)2-2,1-PtSB10H10] (7) [Fig. 1, diagram (b)] traces
readily to that of neutral [nido-7-SB10H12] [Fig. 1, diagram
(a)], implying intra-cluster bonding parallel between the
nido-shaped {SB10H10} fragments in each case. In this
regard a straightforward replacement of the two-electron
three-centre BHB bonds in [nido-7-SB10H12] by two-elec-
tron three-centre BPtB bonds in 7 can be envisaged. This
would be consistent with the frontier orbital approach
and an essentially square-planar platinum(II) centre as dis-
cussed above. On the other hand, the shielding pattern for
the rhodium species [2,2-(PMe2Ph)2-2-Cl-closo-2,1-
RhSB10H10] (4) [Fig. 1, diagram (c)], which also parallels
those for compounds 1 and 5, is markedly different from
that of [nido-7-SB10H12] [diagram (a)], consistent with a
much greater perturbation of the bonding within the
{SB10H10} unit, as expected from the permitted interac-
tions of the thiaborane ligand HOMO and SHOMO with
the octahedral rhodium(III) metal centre. Similar consider-
ations apply to the octahedral rhodium(III) species [2-(g5-
C5Me5)-closo-2,1-RhSB10H10] (6) [Fig. 1, diagram (d)]. The
11B shielding pattern for the [nido-7-SB10H11]ꢀ monoanion
[Fig. 1, diagram (e)] is much more fundamentally different,
with an essentially inverted shielding sequence compared to
the metal species, indicating few intra-cluster bonding sim-
ilarities with the metal complexes.
This ‘rocking’ mechanism contrasts to the fluxional
behaviour of a number of metallaheteroboranes containing
the {Pd(PMe2Ph)2} or {Pd(PMe2Ph)2} groups that exhibit
complete 360° contrarotational fluxionality associated with
the metal-to-heteroborane bonding [1,11,16,21]. The
reported activation energies DGà for these complete rota-
tional processes range from less than 30 up to ca.
65 kJ molꢀ1 [21]. In compound 4, however, the complete
rotation of the {RhCl(PMe2Ph)2} unit above the pentago-
nal {SB4} face is not achieved, even at higher temperatures
(100 °C in C6D6CD3 solution), implying a lower limit for
DGà of ca. 80 kJ molꢀ1 for any such complete contrarota-
1
tion. This is demonstrated by the higher temperature H
NMR spectra that show two distinct sets of P-methyl res-
onances, which in a complete rotation would coalesce into
one resonance set. Related hindered rotational fluxionali-
ties in metal-to-heteroborane coordination are common,
and, for example, have been previously reported for a large
series of closo twelve-vertex rhodium and iridium dicarb-
aboranes of general formulation [L2HM{carb}], where M
is Rh or Ir, L is PPh3, PEt3, or PMe2Ph and {carb} is
{1,2-C2B9H10R}, {1,7-C2B9H10R}, or {1,12-C2B9H10R},
with R being H, Me, Ph, or n-Bu [40]. Conversely, fluxio-
nality involving a complete contrarotation of the heterob-
orane and non-heteroborane and ligand spheres about
the metal centre is a general behaviour for closo twelve-ver-
tex metallaheteroboranes of palladium and platinum (e.g.
compound 7). The intimate mechanism of the complete
rotation of {Pt(PR3)2} or {Pd(PR3)2} units versus the hete-
roborane ligand faces has been reported to involve the
The crystallographically determined molecular structure
of [2,2-(PMe2Ph)2-2-Cl-closo-2,1-RhSB10H10] (4) (Fig. 2)
also reveals an asymmetric conformation of the
{RhCl(PMe2Ph)2} vertex above the {SB4} pentagonal face
of the thiaborane ligand. This configuration is fully in
accord with the 31P NMR spectrum at 176 K (Table 2),

R. Macı´as et al. / Journal of Organometallic Chemistry 693 (2008) 435–445
443
shifting of these fragments via various g4- and g3-bonded
coordinations (e.g. schematics VIII and IX above) with a
concomitant twisting of the {M(PR3)2} units about an axis
passing approximately through the metal-atom and its
antipodal boron atom [21]. Here it is of interest that the less
intimate two-orbital metal-to-cluster bonding of the
pseudo-square-planar palladium and platinum species
gives a freer fluxionality than the more intimate, and more
symmetric, three-orbital involvement in the cluster bonding
shown by the octahedral {RhX(PR3)2} and {IrX(PR3)2}
species. Finally, in terms of ligand-sphere fluxionality, it
should be mentioned that the less localised rhodium-
to-hydrocarbon bonding in [2-(g5-C5Me5)-closo-2,1-
RhSB10H10] (6) will also of course enable fluxionality,
specifically a very ready rotation of the symmetrically
bound {C5Me5} unit relative to a cluster axis approxi-
mately defined by the Rh(2)–B(9) vector, so that all five
{CMe} units of the {C5Me5} moiety are rendered on
time-average chemically equivalent in solution.
isotopomer envelope indicating ready dihydrogen loss.
The relatively hydrogen-rich nido compound 2 also showed
significant fragmentation with dihydrogen loss. All NMR
spectra clearly showed that pure single compounds were
present. Compound identification was assisted by compar-
ison with established NMR data [13,14,17] for selenium-
and tellurium-containing analogues of compounds 1, 3, 4,
5, and 6. In the case of compound 4 the structure was con-
firmed in the solid-state by a single-crystal-X-ray diffrac-
tion analysis.
2.2. Synthesis of [2,2-(PPh3)2-2-H-closo-2,1-RhSB10H10]
(compound 1) and [{(PPh3)-closo-RhSB10H10}2]
(compound 3)
[RhCl(PPh3)3] (100 mg; 120 lmol) was added to a solu-
tion of [nido-7-SB10H12] (20 mg; 130 lmol) and tmnd
(26 mg; 120 lmol) in dry CH2Cl2 (ca. 20 ml) at room tem-
perature. The reaction mixture was stirred under dinitrogen
for 2 h. The resulting yellow solution was filtered off
through silica. The filtrate was concentrated (rotary evapo-
rator, 30 °C, water-pump pressure) and applied to prepara-
tive TLC plates. Elution with 3:2 CH2Cl2/hexane gave
three different bands: (a) purple, RF = 0.9, (b) yellow,
RF = 0.8, and (c) orange, RF = 0.7. These were removed
from the plates and the compounds extracted with CH2Cl2.
After recrystallisation from CH2Cl2/hexane, the compo-
nents of these bands were thence, respectively, character-
ised as the purple dimer [{(PPh3)-closo-RhSB10H10}2] (3)
(6 mg; 3 lmol; 5%), yellow [2,2-(PPh3)2-2-H-closo-2,1-
RhSB10H10] (1) (24 mg; 35 lmol; 26%) and orange [8,8-
(PPh3)2-nido-8,7-RhSB9H11] (2) (16 mg; 21 lmol; 17%).
2.1. Experimental
2.1.1. General
The reactants [cis-RhCl(PPh3)3] [41], [cis-IrCl(PPh3)3]
[42], [{RhCl2(g5-C5Me5)}2] [43] and [nido-7-SB10H12]
[19,20] were prepared by standard literature methods.
The N,N,N0N0-tetramethylnaphthalene-1,8-diamine (tmnd),
PPh3 and PMe2Ph were obtained from commercial sources.
The reactions were carried out under an inert atmosphere
of dry dinitrogen, although subsequent manipulations were
carried out in air. Dried and degassed solvents were used
throughout. 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
pates of dimensions 20 ꢂ 20 cm followed by drying in air
at 80 °C. NMR spectroscopy was performed at 2.4 and/
or 9.4 T on commercially available instrumentation, with
the general techniques being essentially as detailed in other
paper from our laboratories [16c,44,45], and using the gen-
eral approach as summarised, for example, by Reed [46].
2.3. Synthesis of [2,2-(PMe2Ph)2-2-Cl-closo-2,1-
RhSB10H10] (compound 4)
PMe2Ph (0.4 ml of a 0.1 M solution in CH2Cl2) was
added to a yellow solution of [2,2-(PPh3)2-2-H-closo-
RhSB10H10] (1) (prepared as above; 20 mg; 26 lmol) in
CH2Cl2 (30 ml) at room temperature. The reaction mixture
was stirred for 4 h under dinitrogen, then filtered through
silica, reduced in volume as above, and applied to a pre-
parative TLC plate. The chromatogram was developed
with 1:1 CH2Cl2/hexane, leading to the isolation of a yel-
low component of RF = 0.43 that was characterised as
Chemical shifts
d are given in ppm. relative to
N = 100 MHz for d(1H) (quoted 0.05 ppm) (nominally
SiMe4), N = 40.480 730 MHz for d(31P) (quoted
0.5 ppm) (nominally 85% H3PO4), and N = 32.083
971 MHz for d(11B) [47] (quoted 0.5 ppm) (nominally
[47] F3BOEt2 in CDCl3); N is as defined by McFarlane
[48]. Mass spectrometry was carried out on a VG AUTO-
SPEC instrument using either 70 eV EI (electron-impact)
ionization and the solid-sample introduction probe, or
using FAB ionisation (fast-atom bombardment; Cs+-ion
bombardment of a target solution in 3-nitrobenzyl alcohol;
gas-phase pressure 10ꢀ6 mm Hg), according to the charac-
teristics of the sample. Compounds were characterised with
multinuclear NMR spectra and mass spectrometry. Mass-
spectrometric parent ions consistent with the proposed
molecular formulations were observed in all cases, except
for dimeric compound 3, which showed an {Mꢀ2}+
[2,2-(PMe2Ph)2-2-Cl-closo-2,1-RhSB10H10]
(4)
(3 mg;
5 lmol; 20%). Crystals suitable for X-ray diffraction analy-
sis were grown by diffusion of hexane into a solution of 4 in
CH2Cl2 using the overlayering technique.
2.4. Synthesis of [2,2-(PPh3)2-2-H-closo-2,1-IrSB10H10]
(compound 5)
Using the same procedure as in the synthesis of com-
pound
3 above, the reaction between [IrCl(PPh3)3]
(100 mg; 100 lmol) and nido-7-SB10H12 (17 mg; 108 lmol)
in the presence of tmnd afforded a pale-yellow product of

444
R. Macı´as et al. / Journal of Organometallic Chemistry 693 (2008) 435–445
RF = 0.43 (eluent mixture 3:2 CH2Cl2/hexane), which was
recrystallised from CH2Cl2/hexane and identified as [2,2-
(PPh3)2-H-closo-2,1-IrSB10H10] (5; 10 mg; 12 lmol; 12%).
Academy of Sciences, together with the Royal Society of
Chemistry Overseas Author’s fund, for support and assis-
tance with reciprocal visits.
2.5. Synthesis of [2-(g5-C5Me5)-closo-2,1- RhSB10H10]
(compound 6)
Appendix A. Supplementary material
CCDC 280332 contains the supplementary crystallo-
graphic data for 4. These data can be obtained free of
charge from The Cambridge Crystallographic Data Centre
via www.ccdc.cam.ac.uk/data_request/cif. Supplementary
data associated with this article can be found, in the online
version, at doi:10.1016/j.jorganchem.2007.11.014.
Dimeric [{RhCl2(g5-C5Me5)}2] (50 mg; 83 lmol) was
added to
a colourless solution of [nido-7-SB10H12]
(28 mg; 180 lmol) and tmnd (71 mg; 330 lmol) in CH2Cl2
(50 ml) at room temperature. The resulting red solution
was stirred for 15 h, and monitored periodically by analyt-
ical TLC. The final reaction mixture was filtered through
silica gel, concentrated (rotary evaporator) and subjected
to preparative TLC on silica gel with CH2Cl2/hexane
(1:1) as eluting medium. The major band (RF = 0.5) was
located under UV irradiation, extracted into CH2Cl2,
recrystallised from CH2Cl2/hexane, and characterised by
mass spectrometry and NMR spectroscopy as colourless
[2-(g5-C5Me5)-closo-2,1-RhSB10H10] (6) (16 mg; 41 lmol;
25%).
References
[1] (a) D.E. Hyatt, J.L. Little, J.T. Moran, F.R. Scholer, L.J. Todd, J.
Am. Chem. Soc. 89 (1967) 3342;
(b) W.H. Knoth, J. Am. Chem. Soc. 89 (1967) 3342;
(c) W.H. Knoth, J.L. Little, L.J. Todd, Inorg. Synth. 11 (1968) 41;
(d) W.H. Knoth, Inorg. Chem. 10 (1971) 598;
(e) L. Wesemann, Y. Ramjoie, M. Trinkaus, B. Ganter, Inorg.
Chem. 36 (1997) 5192.
[2] R.R. Rietz, D.F. Dustin, M.F. Hawthorne, Inorg. Chem. 13 (1974)
1580.
[3] (a) R.N. Grimes, in: G. Wilkinson, E.W. Abel, F.G.A. Stone (Eds.),
Comprehensive Organometallic Chemistry I, vol. 1, Pergamon,
London, 1982, pp. 459–541;
2.6. Single-crystal X-ray diffraction analysis
Crystals of [2,2-(PMe2Ph)2-2-Cl-closo-2,1-RhSB10H10]
(4) suitable for the X-ray analysis were grown from concen-
trated CH2Cl2 solution by liquid–liquid diffusion from an
overlayer of hexane. The collection of data was carried
out using a Stoe STADI4 diffractometer operating in the
x–h scan mode using Mo Ka radiation. The structure
was solved by direct methods using SHELXS 86 [49] and
was refined by full-matrix least squares using SHELXL 93
[50]. All non-hydrogen atoms were refined with anisotropic
displacement parameters. The cluster-associated hydrogen
atoms were located on Fourier-difference syntheses and
were freely refined isotropically.
(b) R.N. Grimes, in: G. Wilkinson, E.W. Abel, F.G.A. Stone (Eds.),
Comprehensive Organometallic Chemistry II, vol. 1, Pergamon,
London, 1995, pp. 373–430;
(c) I.V. Pisareva, I.T. Chizhevsky, P.V. Petrovskii, E.V. Vorontzov,
V.I. Bregadze, F.M. Dolgushin, A.I. Yanovsky, Inorg. Chim. Acta
280 (1998) 233.
[4] C.G. Salentine, M.F. Hawthorne, J. Am. Chem. Soc. 97 (1975) 6382.
[5] W.E. Carrol, M. Green, F.G.A. Stone, A.J. Welch, J. Chem. Soc.,
Dalton Trans. (1975) 2263.
[6] P.A. Wegner, L.J. Guggenberger, E.L. Muetterties, J. Am. Chem.
Soc. 92 (1970) 3473.
[7] H.-P. Hansen, J. Muller, U. Englert, P. Paetzold, Angew. Chem., Int.
¨
Ed. Engl. 30 (1991) 1377.
Crystal data are as follows. [2,2-(PMe2Ph)2-2-Cl-closo-
2,1-RhSB10H10] (4): C16H32B10P2SClRh; M = 564.88, yel-
low prism, 0.46 ꢂ 0.42 ꢂ 0.38 mm, from CH2Cl2/hexane,
[8] X.L.R. Fontaine, N.N. Greenwood, J.D. Kennedy, P.I. MacKinnon,
I. Macpherson, J. Chem. Soc., Dalton Trans. (1987) 2385.
[9] (a) J.L. Little, A.C. Wong, J. Am. Chem. Soc. 93 (1971) 522;
(b) J.L. Little, S.S. Pao, Inorg. Chem. 17 (1978) 584;
(c) J.L. Little, Inorg. Chem. 15 (1976) 114;
(d) H.-P. Hansen, U. Englert, P. Paetzold, Z. Anorg. Allg. Chem. 621
(1995) 719.
[10] T.P. Hanusa, N. Roig de Parisi, J.G. Kester, A. Arafat, L.J. Todd,
Inorg. Chem. 26 (1987) 4100.
[11] S.R. Bunkhall, X.L.R. Fontaine, N.N. Greenwood, J.D. Kennedy,
M. Thornton-Pett, J. Chem. Soc., Dalton Trans. (1990) 73.
[12] W.R. Hertler, F. Klanberg, E.L. Muetterties, Inorg. Chem. 6 (1967)
1696.
monoclinic,
a = 11.8132(7),
b = 13.1857(12),
c =
3
˚
˚
16.4234(13) A, b = 98.582(6)°, U = 2529.6(6) A , Dcalc
=
1.483 Mg mꢀ3; k = 0.71073 A (Mo Ka), T = 173 K, space
group P21/c, Z = 4, l(Mo Ka) = 0.995 mmꢀ1, R1 =
0.0181 for 4058 reflections with I > 2r(I), and wR2 =
0.0473 for all 4444 unique reflections.
˚
Acknowledgements
[13] Faridoon, M. McGrath, T.R. Spalding, X.L.R. Fontaine, J.D.
Kennedy, M. Thornton-Pett, J. Chem. Soc., Dalton Trans. (1990)
1819.
[14] Faridoon, O. Ni Dhubhghaill, T.R. Spalding, G. Ferguson, B.
Kaitner, X.L.R. Fontaine, J.D. Kennedy, D. Reed, J. Chem. Soc.,
Dalton Trans. (1988) 2739.
ˇ
Contribution No. 113 from the Leeds-Rezˇ Anglo-Czech
Polyhedron Collaboration (ACPC). R.M. thanks the
Basque Government for a scholarship, and the Ministry
of Education and Science of Spain for support within the
[15] D.A. Thompson, R.W. Rudolph, J. Chem. Soc., Chem. Commun.
(1976) 770.
[16] (a) G. Ferguson, M. Parvez, J.A. MacCurtain, O. Ni Dhubhghaill,
T.R. Spalding, D. Reed, J. Chem. Soc., Dalton Trans. (1987) 699;
(b) G. Ferguson, J.D. Kennedy, X.L.R. Fontaine, Faridoon, T.R.
Spalding, J. Chem. Soc., Dalton Trans. (1988) 2555;
´
framework of the ‘‘Ramon y Cajal” program. We thank
the Ministry of Education of the Czech Republic (project
no. LC 523) for support and the UK EPSRC for equip-
ment grants (J/56929, L/49505, M/83360 and R/61949).
We also thank the Royal Society (London) and the Czech

R. Macı´as et al. / Journal of Organometallic Chemistry 693 (2008) 435–445
445
ˇ
´
´
(c) Faridoon, O. Ni Dhubhghaill, T.R. Spalding, G. Ferguson, B.
Kaiter, X.L.R. Fontaine, J.D. Kennedy, J. Chem. Soc., Dalton
Trans. (1989) 1657;
[29] (a) R. Macıas, J. Holub, J.D. Kennedy, B. Stıbr, M. Thornton-Pett,
W. Clegg, J. Chem. Soc., Dalton Trans. 149 (1997);
(b) R. Macıas, J. Bould, J. Holub, J.D. Kennedy, B. Stıbr, M.
ˇ
´
´
(d) G. Ferguson, J.F. Gallagher, M. McGrath, J.P. Sheehan, T.R.
Spalding, J.D. Kennedy, J. Chem. Soc., Dalton Trans. (1993) 27;
(e) G. Ferguson, Faridoon, T.R. Spalding, Acta Crystallogr., Sect. C
44 (1988) 1368.
Thornton-Pett, Dalton Trans. (2007) 2885.
[30] (a) J. Bould, J.D. Kennedy, N.N. Greenwood, J. Organomet. Chem.
249 (1983) 11;
(b) N.N. Greenwood, J.D. Kennedy, M. Thornton-Pett, J.D. Wool-
lins, J. Chem. Soc., Dalton Trans. (1985) 2397.
[17] S.-O. Kang, P.J. Carroll, L.G. Sneddon, Inorg. Chem. 28 (1989) 961.
[18] S.-O. Kang, P.J. Carroll, L.G. Sneddon, Organometallics 7 (1988)
772.
[19] W.R. Hertler, F. Klanberg, E.L. Muetterties, Inorg. Chem. 6 (1967)
1696.
[20] (a) S.-O. Kang, L.G. Sneddon, Inorg. Chem. 27 (1988) 3298;
(b) A.M. Shedlow, L.G. Sneddon, Inorg. Chem. 37 (1998) 5269.
[21] D. O’Connell, J.C. Patterson, T.R. Spalding, G. Ferguson, J.F.
[31] (a) X.L.R. Fontaine, H. Fowkes, N.N. Greenwood, J.D. Kennedy,
M. Thornton-Pett, J. Chem. Soc., Dalton Trans. (1987) 2417;
(b) M. Bown, X.L.R. Fontaine, N.N. Greenwood, J.D. Kennedy,
M. Thornton-Pett, J. Chem. Soc., Dalton Trans. (1988) 925;
(c) E.J. Ditzel, X.L.R. Fontaine, H. Fowkes, N.N. Greenwood, J.D.
Kennedy, P. MacKinnon, Zhu Sisan, M. Thornton-Pett, J. Chem.
Soc., Chem. Commun. (1990) 1692.
ˇ
´
´
´
Gallagher, Y. Li, J.D. Kennedy, R. Macıas, M. Thornton-Pett, J.
[32] (a) J. Holub, J.D. Kennedy, T. Jelınek, B. Stıbr, J. Chem. Soc.,
Dalton Trans. (1994) 1317;
Holub, J. Chem. Soc., Dalton Trans. (1996) 3323.
[22] (a) M. Bown, X.L.R. Fontaine, J.D. Kennedy, J. Chem. Soc., Dalton
Trans. (1988) 1467;
ˇ
´
´
(b) J. Plesˇek, B. Stıbr, X.L.R. Fontaine, T. Jelınek, M. Thornton-
Pett, S. Herˇma´nek, J.D. Kennedy, Inorg. Chem. 33 (1994) 2994;
ˇ
´ ´
(c) B. Stıbr, J. Holub, T. Jelınek, B. Gruner, J. Fusek, F. Teixidor, C.
¨
Vinas, J.D. Kennedy, Collect. Czech. Chem. Commun. 62 (1997) 1229.
˜
(b) C. Bauer, D. Gabel, T. Borrmann, C.A. Kilner, J.D. Kennedy,
M. Thornton-Pett, U. Do¨rfler, J. Organomet. Chem. 657 (2002) 205;
(c) A. Franken, C.A. Kilner, M. Thornton-Pett, J.D. Kennedy,
Collect. Czech. Chem. Commun. J. Plesˇek Special Edition 67 (2002)
869;
[33] Faridoon, PhD Thesis, University College, Cork, Ireland, 1988.
[34] (a) J. Bould, J.D. Kennedy, M. Thornton-Pett, J. Chem. Soc., Dalton
Trans. (1992) 563;
ˇ
´
´
´
(d) J. Bould, M. Bown, J.D. Kennedy, Collect. Czech. Chem.
Commun. 70 (2005) 410.
(b) B. Stıbr, J.D. Kennedy, E. Drdakova, M. Thornton-Pett, J.
Chem. Soc., Dalton Trans. (1994) 229.
ˇ
´
[23] (a) J.H. Jones, K. Basˇe, B. Stıbr, J.D. Kennedy, X.L.R. Fontaine,
N.N. Greenwood, M.G.H. Wallbridge, Polyhedron 8 (1989) 2089;
[35] J.D. Kennedy, in: J. Casanova (Ed.), The Borane–Carborane–
Carbocation Continuum, Wiley, New York, 1998, pp. 85–116.
[36] H.M. Colquhoun, T.J. Greenhough, M.G.H. Wallbridge, J. Chem.
Soc., Dalton Trans. (1979) 619.
ˇ
´
´
(b) T. Jelınek, J.D. Kennedy, B. Stıbr, M. Thornton-Pett, Inorg.
Chem. Commun. 1 (1998) 179;
ˇ
´
´
ˇ
´
(c) J. Plesˇek, B. Stıbr, D. Hnyk, T. Jelınek, S. Hermanek, J.D.
Kennedy, M. Hoffman, P.v.R. Schleyer, Inorg. Chem. 37 (1998)
3902;
[37] D. O‘Connell, T.R. Spalding, G. Ferguson, J.F. Gallagher, J.D.
Kennedy, J. Organomet. Chem. 503 (1995) C12.
[38] (a) K. Wade, J. Chem. Soc., Chem. Commun. (1971) 792;
(b) K. Wade, Adv. Inorg. Chem. Radiochem. 18 (1976) 1;
(c) R.E. Williams, Inorg. Chem. 1 (1971) 210;
ˇ
(d) T. Jel´ınek, I. Cisarˇova´, B. St´ıbr, J.D. Kennedy, M. Thornton-
Pett, J. Chem. Soc., Dalton Trans. (1998) 2965;
ˇ
´
(e) J. Holub, B. Stıbr, D.L. Ormsby, J.D. Kennedy, R. Greatrex,
(d) R.E. Williams, Adv. Inorg. Chem. Radiochem. 18 (1976) 67;
(e) D.M.P. Mingos, Nature (Phys. Sci.) 236 (1972) 99;
(f) D.M.P. Mingos, Acc. Chem. Res. 17 (1984) 311.
Inorg. Chem. Commun. 3 (2000) 178;
ˇ
´
´
(f) T. Jelınek, B. Stıbr, J.D. Kennedy, D. Hnyk, M. Buhl, Dalton
¨
Trans. (2003) 1326;
(g) P.K. Dosangh, J. Bould, M.G.S. Londesborough, T. Jelınek, M.
Thornton-Pett, B. Stıbr, J.D. Kennedy, J. Organomet. Chem. 680
(2003) 312.
[39] (a) J.D. Kennedy, Main Group Metal Chem. 12 (1989) 149;
(b) M.P. Murphy, T.R. Spalding, C. Cowey, J.D. Kennedy, M.
Thornton-Pett, J. Holub, J. Organomet. Chem. 550 (1998) 151;
(c) J. Bould, P.A. Cooke, U. Do¨rfler, J.D. Kennedy, L. Barton, N.P.
Rath, M. Thornton-Pett, Inorg. Chim. Acta 285 (1999) 290.
[40] See, for example: T.B. Marder, R.T. Baker, J.A. Long, J.A. Doi,
M.F. Hawthorne, J. Am. Chem. Soc. 103 (1981) 2988.
[41] J.A. Osborn, G. Wilkinson, Inorg. Synth. 10 (1967) 67.
[42] M.A. Bennett, J.L. Latten, Inorg. Synth. 26 (1989) 200.
[43] J.W. Kang, K. Moseley, P.M. Maitlis, J. Am. Chem. Soc. 91 (1969)
5970.
´
ˇ
´
[24] (a) Y.-H. Kim, A. Brownless, P.A. Cooke, R. Greatrex, J.D.
Kennedy, M. Thornton-Pett, Inorg. Chem. Commun. 1 (1998) 19;
ˇ
(b) S.L. Shea, T.D. McGrath, T. Jel´ınek, B. St´ıbr, M. Thornton-Pett,
J.D. Kennedy, Inorg. Chem. Commun. 1 (1998) 97;
(c) M. Bakardjiev, J. Holub, M.J. Carr, J.D. Kennedy, B. Stıbr,
Dalton Trans. (2005) 909.
ˇ
´
[25] G. Ferguson, M.C. Jennings, A.J. Lough, S. Coughlan, T.R.
ˇ
´
Spalding, J.D. Kennedy, X.L.R. Fontaine, B. Stıbr, J. Chem. Soc.,
Chem. Commun. (1990) 891.
[44] M. Bown, X.L.R. Fontaine, N.N. Greenwood, J.D. Kennedy, Z.
Anorg. Allg. Chem. 602 (1991) 17.
[26] G. Ferguson, A.J. Lough, Faridoon, M.N. McGrath, T.R. Spalding,
J.D. Kennedy, X.L.R. Fontaine, J. Chem. Soc., Dalton Trans. (1990)
1831.
[45] J.D. Kennedy, X.L.R. Fontaine, M. McGrath, T.R. Spalding, Magn.
Reson. Chem. 29 (1991) 711.
[46] D. Reed, Chem. Soc. Rev. (1993) 109.
[27] S. Coughlan, T.R. Spalding, G. Ferguson, J.F. Gallagher, A.J.
[47] J.D. Kennedy, in: J. Mason (Ed.), Multinuclear NMR, Plenum, New
York, 1987, pp. 221–254.
ˇ
Lough, X.L.R. Fontaine, J.D. Kennedy, B. St´ıbr, J. Chem. Soc.,
Dalton Trans. (1992) 2865.
[48] W. McFarlane, Proc. Roy. Soc.(London) Ser. A A306 (1986) 185.
[49] G.M. Sheldrick, Acta Crystallogr., Sect. A 46 (1990) 467.
[50] G.M. Sheldrick, SHELXL 93, Program for refinement of crystal
structures, University of Go¨ttingen, 1993.
´
[28] (a) R. Macıas, N.P. Rath, L. Barton, Organometallics 18 (1999)
3637;
´
(b) R. Macıas, N.P. Rath, L. Barton, Angew. Chem., Int. Ed. Engl.
38 (1999) 162.
[51] L.J. Farrugia, ORTEP-3, J. Appl. Crystallogr. 30 (1997) 565.