Isomerisation following the platination of [7,8-Ph2-9,11-I 2-7,8-nido-C2B9H7]2-

Article abstract of DOI:10.1016/j.poly.2005.10.006

The synthesis, by electrophilic substitution, and characterisation of two new B-substituted nido diphenylcarboranes is described. [HNMe ₃][7,8-Ph₂-9-I-7,8-nido-C₂B₉H ₉] (1) and [HNMe₃][7,8-Ph₂-9,11-I ₂-7,8-nido-C₂B₉H₈] (2) were characterised spectroscopically with the structure of the latter additionally confirmed by a crystallographic study. [7,8-Ph₂-9,11-I ₂-7,8-nido-C₂B₉H₈]^2- was treated with a source of {Pt(PMe₂Ph)₂}²⁺ and three new platinacarboranes isolated. All were characterised by a combination of ¹H, ¹¹B and ³¹P NMR spectroscopies and by single crystal X-ray diffraction, and all were found to have undergone a 1,2 → 1,7 cage C atom isomerisation, relative to the presumed first product of the metallation. The major product 3 is 1,8-Ph₂-2,2-(PMe ₂Ph)₂-6,7-I₂-2,1,8-closo-PtC₂B ₉H₇ with both (iodide) substituted boron vertices in the upper part of the icosahedron. The minor products 4 and 5 are, respectively, 1,8-Ph₂-2,2-(PMe₂Ph)₂-6,12-I ₂-2,1,8-closo-PtC₂B₉H₇ and 1,8-Ph₂-2,2-(PMe₂Ph)₂-10,12-I ₂-2,1,8-closo-PtC₂B₉H₇, having one and two substituted boron vertices in the lower pentagonal belt. These structural results allow speculation on the mechanism of isomerisation of crowded platinacarboranes.

Full text of DOI:10.1016/j.poly.2005.10.006

Polyhedron 25 (2006) 915–922
www.elsevier.com/locate/poly
Isomerisation following the platination of
[7,8-Ph2-9,11-I2-7,8-nido-C2B9H7]2ꢀ q
*
*
David Ellis , Rhona M. Garrioch, Georgina M. Rosair, Alan J. Welch
Department of Chemistry, School of Engineering and Physical Sciences, Heriot-Watt University, Edinburgh EH14 4AS, UK
Received 23 August 2005; accepted 16 October 2005
Available online 1 December 2005
Warmly dedicated to Professor Michael B. Hursthouse, a mentor and friend to A.J.W. for more than 35 years, on the occasion of his 65th birthday.
Abstract
The synthesis, by electrophilic substitution, and characterisation of two new B-substituted nido diphenylcarboranes is described.
[HNMe3][7,8-Ph2-9-I-7,8-nido-C2B9H9] (1) and [HNMe3][7,8-Ph2-9,11-I2-7,8-nido-C2B9H8] (2) were characterised spectroscopically with
the structure of the latter additionally confirmed by a crystallographic study. [7,8-Ph2-9,11-I2-7,8-nido-C2B9H8]2ꢀ was treated with a
source of {Pt(PMe2Ph)2}2+ and three new platinacarboranes isolated. All were characterised by a combination of 1H, 11B and 31P
NMR spectroscopies and by single crystal X-ray diffraction, and all were found to have undergone a 1,2 ! 1,7 cage C atom isomerisa-
tion, relative to the presumed first product of the metallation. The major product 3 is 1,8-Ph2-2,2-(PMe2Ph)2-6,7-I2-2,1,8-closo-PtC2B9H7
with both (iodide) substituted boron vertices in the upper part of the icosahedron. The minor products 4 and 5 are, respectively, 1,8-Ph2-
2,2-(PMe2Ph)2-6,12-I2-2,1,8-closo-PtC2B9H7 and 1,8-Ph2-2,2-(PMe2Ph)2-10,12-I2-2,1,8-closo-PtC2B9H7, having one and two substituted
boron vertices in the lower pentagonal belt. These structural results allow speculation on the mechanism of isomerisation of crowded
platinacarboranes.
ꢀ 2005 Elsevier Ltd. All rights reserved.
Keywords: Metallacarborane; Synthesis; NMR spectroscopy; X-ray crystallography; Isomerisation mechanisms
1. Introduction
presumably via a highly sterically hindered, transient,
3,1,2-MC2B9 species (Scheme 1, Step 1). Under the benign
conditions of these reactions one could be confident that
For many years there has been significant interest in the
mechanism(s) of isomerisation of icosahedral carboranes
[1]. Some attempts have been made to study the phenome-
non experimentally using B-substituted species [2] but the
high temperatures generally required to induce cage reor-
ganisation have inevitably led to doubts about the integrity
of the B–X bonds under such conditions [3].
vertex-substituent
bonds
would
remain
intact.
Moreover, with certain metal fragments, specifically
fMoðg-C3H5ÞðCOÞ2þg, several intermediate species have
actually been isolated [5] and resemble structures predicted
[6] by ab initio computational studies of the isomerisation
of 1,2-closo-C2B10H12.
We have shown [4] that metallation of [7,8-Ph2-7,8-nido-
C2B9H9]2ꢀ with a sufficiently bulky metal–ligand fragment
leads to isomerisation to 2,1,8-MC2B9 products (effectively
a 1,2 ! 1,7 C-atom isomerisation) at room temperature,
Thus, we subsequently embarked on a programme to
map, experimentally, the mechanism(s) of carborane isom-
erisation via the synthesis and subsequent metallation of a
series of B-substituted carboranes, identifying the ultimate
products by crystallographic study [7]. Using {ML2}2+
fragments two general themes developed: addition of bulky
{NiL2}2+ generally yields products with 4,1,2-NiC2B9
architectures, representing ‘‘1,2 ! 1,2’’ C atom isomerisa-
tion [7c,7d,8], although occasionally 2,1,8-NiC2B9 species
q
Steric effects in heteroboranes. Part 31. For part 30, see [7d].
Corresponding authors. Tel.: +44 131 451 3217; fax: +44 131 451 3180.
*
E-mail addresses: d.ellis@hw.ac.uk (D. Ellis), a.j.welch@hw.ac.uk
(A.J. Welch).
0277-5387/$ - see front matter ꢀ 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.poly.2005.10.006

916
D. Ellis et al. / Polyhedron 25 (2006) 915–922
Scheme 1. Possible formation of platinacarboranes 3–5 by triangle face rotations (tfrs) starting from an implied transient species I. In each tfr the face
highlighted in bold is assumed to rotate by 120ꢁ. In Steps 20 and 200 the B3 and CB2 tfrs could be sequential or synchronous. P = PMe2Ph.
are also formed [7c,7d], whereas addition of bulky
{PtL2}2+ always gives 2,1,8-PtC2B9 products [4,7b,7d],
although in one case [7b] we were also able to isolate a
3,1,2-PtC2B9 species and show that it represented an inter-
mediate by its isomerisation to 2,1,8-PtC2B9 on gentle
warming.
In this paper, we extend these studies by describing the
synthesis of new mono- and disubstituted nido-diphenyl-
carboranes and the results of subsequent metallation of
the latter with a {PtL2}2+ fragment.
glass plates. NMR spectra at 200.13 (1H), 400.13 MHz
(1H), 128.38 (11B) or 161.98 MHz (31P) were recorded on
Bruker AC 200 or DPX 400 spectrometers from CDCl3
or (CD3)2CO solutions at ambient temperature, chemical
shifts being recorded relative to SiMe4 (1H), BF3 Æ OEt2
(11B) or H3PO4 (31P). IR spectra were recorded from KBr
disks on a Perkin–Elmer Spectrum RX FT-IR spectropho-
tometer. Elemental analyses were determined by the EPS
School service. The starting materials 1,2-Ph2-1,2-closo-
C2B10H10 [9] and cis-PtCl2(PMe2Ph)2 [10] were prepared
by literature methods or slight variants thereof. All other
reagents and solvents were supplied commercially and were
used as received.
2. Experimental
2.1. Synthetic and spectroscopic studies
2.1.1. Synthesis of [HNMe3][7,8-Ph2-9-I-nido-C2B9H9]
(1)
Experiments were performed under dry, oxygen-free N2
using standard Schlenk techniques, with some subsequent
manipulation in the open laboratory. Solvents were freshly
distilled over CaH2 (CH2Cl2) or Na wire (THF, Et2O, 40–
60 petroleum ether), or were stored over 4 A molecular
sieves [CDCl3, (CD3)2CO]. Preparative thin layer chroma-
1,2-Ph2-C2B10H10 (1.000 g, 3.38 mmol) and KOH
(0.430 g; 8.45 mmol) were heated to reflux in EtOH
(50 ml) for 72 h. Excess KOH was removed by reaction with
CO2 and filtering off the K2CO3 formed. To the clear col-
ourless solution of K[7,8-Ph2-nido-7,8-C2B9H10] was added
I2 (0.870 g, 3.41 mmol) in ethanol (30 ml). The reaction
˚
tography (TLC) employed 20 · 20 cm Kieselgel 60 F254

D. Ellis et al. / Polyhedron 25 (2006) 915–922
917
mixture was heated to reflux for 2 h, during which time the
solution decolourised. The volume of solution was reduced
to ca. 10 ml in vacuo and distilled water (50 ml) added. Fil-
tration and addition of aqueous HNMe3Cl (0.380 g,
4.00 mmol) gave a white precipitate which was filtered off
and dried in vacuo. An alternative procedure involves
extraction of the precipitate into CH2Cl2 (3 · 30 ml), drying
over MgSO4 and subsequent removal of the drying agents
and solvents. An oily solid was recovered which was recrys-
tallised from CH2Cl2/40–60 petroleum ether to give a white
crystalline solid. Yield 1.000 g, (63% based on diphenylcar-
borane). Anal. Calc. for C17H29B9IN: C, 43.3; H, 6.21; N,
2.97. Found: C, 43.3; H, 6.39; N, 2.56%. IR mmax (cmꢀ1):
2537 (B–H str). 1H NMR d (ppm): 7.2–6.8 (m, 10H,
C6H5), 3.1 (s, 9H, NCH3). 11B–{1H} NMR d (ppm):
ꢀ5.70 (1B), ꢀ14.42 (4B), ꢀ20.13 (1B), ꢀ23.96 (1B),
ꢀ29.73 (1B), ꢀ35.68 (1B).
to dryness to yield a yellow solid. The product was purified
by TLC, eluting with 50/50 CH2Cl2/40–60 petroleum ether
to give three yellow bands, respectively, compounds 3
(Rf = 0.65), 4 (Rf = 0.56) and 5 (Rf = 0.45).
3: yield 0.060 g (20%). Anal. Calc. for C30H39B9I2P2Pt:
C, 35.8; H, 3.91. Found: C, 34.9; H, 3.88%. IR mmax
1
(cmꢀ1): 2575 (B–H str). H NMR d (ppm): 7.9–6.9 (m,
20H, C6H5), 2.2–1.2 (m, 12H, P–CH3). 11B–{1H} NMR d
(ppm): ꢀ3.69 (1B), ꢀ8.61 (1B), ꢀ11.70 (1B), ꢀ15.44 (3B),
ꢀ19.75 (2B), ꢀ23.93 (1B). 31P–{1H} NMR d (ppm):
1
1
ꢀ15.80 (s, 1P, JPt–P = 3284 Hz), ꢀ23.95 (s, 1P, JPt–P
=
3227 Hz).
4: yield 0.030 g (10%). Anal. Calc. for C30H39B9I2P2Pt:
C, 35.8; H, 3.91. Found: C, 35.1; H, 3.96%. IR mmax
1
(cmꢀ1): 2544 (B–H str). H NMR d (ppm): 8.0–6.9 (m,
20H, C6H5), 2.1–1.2 (m, 12H, P–CH3). 11B–{1H} NMR d
(ppm): ꢀ3.80 (2B), ꢀ11.38 (1B), ꢀ15.39 (4B), ꢀ18.09
(1B), ꢀ23.76 (1B). 31P–{1H} NMR d (ppm): ꢀ12.03
1
1
(s, 1P, JPt–P = 3306 Hz), ꢀ21.36 (s, 1P, JPt–P = 3217 Hz).
5: yield 0.030 g (10%). Anal. Calc. for C30H39B9I2P2Pt:
C, 35.8; H, 3.91. Found: C, 35.6; H, 3.85%. IR mmax
2.1.2. Synthesis of [HNMe3][7,8-Ph2-9,11-I2-7,8-nido-
C2B9H8] (2)
To
a
solution of K[7,8-Ph2-nido-7,8-C2B9H10]
1
(cmꢀ1): 2581 (B–H str). H NMR d (ppm): 7.5–7.0 (m,
(1.69 mmol) in EtOH (50 ml) prepared as above, was added
I2 (1.080 g, 4.23 mmol) in ethanol (30 ml). The reaction
mixture was heated to reflux for 18 h, during which time
the solution decolourised. The volume of solution was
reduced to ca. 10 ml in vacuo and distilled water (50 ml)
added. Filtration and addition of aqueous HNMe3Cl
(0.173 g, 1.80 mmol) gave a white precipitate which was fil-
tered off and dried in vacuo. Again, an alternative work-up
involves extraction of the precipitate into CH2Cl2
(3 · 30 ml), drying over MgSO4 and subsequent removal
of the drying agents and solvents. A white solid was recov-
ered which was recrystallised by evaporation of a dichloro-
methane solution to give colourless crystals. Yield 0.230 g
(23% based on diphenylcarborane). Anal. Calc. for
C17H28B9I2N: C, 34.2; H, 4.73; N, 2.34. Found: C, 34.1;
20H, C6H5), 1.8–1.1 (m, 12H, P–CH3). 11B–{1H} NMR d
(ppm): ꢀ6.74 (3B), ꢀ13.28 (3B), ꢀ14.91 (2B), ꢀ31.11
1
(1B). 31P–{1H} NMR d (ppm): ꢀ13.93 (s, 1P, JPt–P
=
1
3290 Hz), ꢀ15.32 (s, 1P, JPt–P = 3347 Hz).
2.2. Crystallographic studies
Single crystals suitable for X-ray diffraction were grown
by evaporation of a CH2Cl2 solution (salt 2), from diffusion
of a CH2Cl2 solution and 40–60 petroleum ether at ꢀ30 ꢁC
(compounds 3 and 5) or from diffusion of a CH2Cl2 solu-
tion and diethylether at ꢀ30 ꢁC (compound 4). Salt 2 crys-
tallises with 1 12 equivalents of CH2Cl2 and compound 4
with 1 equivalent of Et2O. Data were collected at 100 K
(cold N2 gas) on a Bruker AXS P4 diffractometer produc-
ing graphite-monochromated Mo Ka radiation (k =
1
H, 4.76; N, 2.27%. IR mmax (cmꢀ1): 2546 (B–H str). H
NMR d (ppm): 7.2–6.8 (m, 10H, C6H5), 3.2 (s, 9H,
NCH3). 11B–{1H} NMR d (ppm): ꢀ13.26 (3B), ꢀ16.55
(2B), ꢀ18.43 (2B, B9,11), ꢀ28.17 (1B), ꢀ35.30 (1B).
˚
0.71069 A) using x scans to hmax = 25ꢁ, and were corrected
for absorption by psi scans. Structures were solved by
direct and difference Fourier methods and refined [11] by
full-matrix least-squares against F2. Refinement was com-
pleted with all non-hydrogen atoms assigned anisotropic
displacement parameters. For the anion of 2 Hexo atoms
were refined subject to a restrained B–H distance of
2.1.3. Syntheses of 1,8-Ph2-2,2-(PMe2Ph)2-6,7-I2-2,1,8-
PtC2B9H7 (3), 1,8-Ph2-2,2-(PMe2Ph)2-6,12-I2-2,1,8-
PtC2B9H7 (4) and 1,8-Ph2-2,2-(PMe2Ph)2-10,12-I2-2,1,8-
PtC2B9H7 (5)
˚
1.10(2) A, whilst the Hendo atom was freely refined. For
Salt 2 (0.180 g; 0.30 mmol) was suspended in dry ether
(30 ml) at 0 ꢁC and BuLi (0.24 ml of a 2.5 M solution in
3–5 cage H atoms were allowed to ride on their bound
boron with B–H constrained to 1.10 A (1.12 A for 3). In
n
˚
˚
hexanes, 0.60 mmol) was added dropwise. The solution
was stirred at room temperature for 18 h. All volatiles were
removed and THF (30 ml) added. The THF solution of
Li2[7,8-Ph2-9,11-I2-7,8-nido-C2B9H7] was added to a frozen
suspension of cis-PtCl2(PMe2Ph)2 (0.163 g, 0.30 mmol) in
THF (25 ml) at ꢀ196 ꢁC. The mixture was refrozen, then
slowly warmed to room temperature and stirred for 18 h.
All volatiles were removed in vacuo and CH2Cl2 (20 ml)
added. The yellow suspension was filtered and evacuated
all structures CH atoms were similarly constrained, with
˚
C–H = 0.95 (phenyl), 0.98 (CH3) or 0.99 (CH2) A. H atoms
except methyl H atoms were assigned displacement param-
eters set at 1.2 times that of the Ueq of their bound atom,
with methyl H atoms assigned displacement parameters
calculated as 1.5 times the bound carbon atom Ueq. H
atoms were included in all molecules of solvation. For 3
the maximum and minimum residual e-density are signifi-
˚
cant but both are within 1 A of Pt2 and thus chemically

918
D. Ellis et al. / Polyhedron 25 (2006) 915–922
meaningless. Table 1 contains details of the crystallo-
graphic studies.
11B NMR spectra are more informative. In the 11B–{1H}
spectrum of 1 a 1:4:1:1:1:1 pattern, from high to low fre-
quency, is observed, although the anticipated C1 symmetry
of the product requires the integral-4 signal to be a
1 + 1 + 1 + 1 coincidence. In the 11B spectrum the uncou-
pled resonance due to the B atom bearing the iodine substi-
tuent is not obvious, and is presumed to lie within the
envelope of these coincident signals. Compound 2 has a
3:2:2:1:1 pattern, its anticipated Cs symmetry requiring
the highest frequency resonance to be a 2 + 1 coincidence.
This time, however, the signal due to the two, symmetry
equivalent, substituted B atoms is clear in the 11B spectrum
at d ꢀ18.43.
3. Results and discussion
3.1. B-substituted carboranes
Electrophilic iodination of K[7,8-Ph2-nido-7,8-C2B9H10]
with elemental I2 proceeds smoothly in refluxing ethanol
to generate, after metathesis with HNMe3Cl, the new
B-substituted nido diphenylcarboranes [HNMe3][7,8-Ph2-
9-I-nido-7,8-C2B9H9] (1) and [HNMe3][7,8-Ph2-9,11-I2-
7,8-nido-C2B9H8] (2).
Use of one equivalent of I2 affords a good yield of the
monosubstituted 9-I species with <5% of the disubstituted
9,11-I2 carborane also formed, as observed by 11B spectros-
copy. Addition of two equivalents of I2 gives pure disubsti-
tuted compound, albeit in somewhat less good yield.
Further substitution is not observed even with large
excesses of I2. The synthesis of salt 1 is analogous to that
used by Hawthorne and Olsen to prepare the related spe-
cies [NMe4][9-I,-7,8-nido-C2B9H11], although this early
paper [12] does not report further substitution to the diiodo
product. Both 1 and 2 are appreciably soluble in acetone
and acetonitrile and sparingly soluble in halogenated sol-
vents, but insoluble in 40/60 petroleum ether and toluene.
The compounds 1 and 2 were fully characterised by
microanalysis, NMR and IR spectroscopies. In both cases
the 1H NMR spectra reveal signals due to cage phenyl and
N-methyl hydrogen atoms in the correct relative integrals.
Electrophilic attack on a nido carborane is expected to
occur at the open face, but, to establish an unequivocal
structure, a crystallographic determination of salt 2, as its
1 12 CH2Cl2 solvate, was undertaken.
Fig. 1 shows a perspective view of a single anion and
Table 2 lists selected interatomic distances and interbond
angles. The 9,11-disubstitution by iodine is confirmed.
The endo H atom H10a is semi-bridging the B10–B11 edge,
˚
˚
B10–H10a 1.17(5) A, B11–H10a 1.56(5) A, consistent with
the optimised structure of [7,8-nido-C2B9H12]ꢀ [13]. As
anticipated [14], the (H-bridged) B10–B11 edge,
˚
1.842(7) A, is consequently somewhat longer that the
unbridged but otherwise equivalent B9–B10 edge,
˚
˚
1.805(8) A. The C7–C8 distance in 2 is 1.610(6) A, almost
identical to that in the unsubstituted analogue [7,8-Ph2-
+
7,8-nido-C2B9H10]ꢀ (1.590(5) A for the [HNEt3] salt and
˚
+
˚
1.602(3) A in the [C6H5CH2NMe3] salt) [15] and to that
Table 1
Crystallographic data for 2–5a
Compound
2
3
4
5
Formula
Mr
C
724.88
0.60 · 0.50 · 0.30
triclinic
17H28B9I2N ꢁ 1 12 CH2Cl2
C30H39B9I2PPt
1007.73
0.42 · 0.22 · 0.20
monoclinic
P21/n
12.003(4)
17.308(6)
17.020(11)
90
93.17(4)
90
3530(3)
4
C30H39B9I2PPt Æ C2H5O
C30H39B9I2PPt
1007.73
0.40 · 0.60 · 0.50
monoclinic
P21/c
10.9980(10)
25.352(3)
12.761(2)
90
95.950(10)
90
3538.9(8)
4
1081.85
0.30 · 0.50 · 0.40
monoclinic
P21/n
15.820(2)
13.462(2)
20.210(2)
90
110.440(10)
90
4033.1(9)
4
Crystal size (mm)
Crystal system
Space group
ꢀ
P1
˚
a (A)
8.7277(7)
11.0465(13)
16.0255(19)
83.165(10)
84.375(11)
70.348(10)
1441.9(3)
2
˚
b (A)
˚
c (A)
a (ꢁ)
b (ꢁ)
c (ꢁ)
3
˚
U (A )
Z
F(000)
702
1912
2080
1912
l (Mo Ka) (mmꢀ1
Data measured
Unique data, n
Rint
)
2.47
6088
5022
0.0648
0.0431
0.0945
1.074
5.84
7669
6203
0.0404
0.0482
0.1056
1.063
5.12
8597
7058
0.0427
0.0436
0.1067
0.735
5.82
7598
6209
0.0340
0.0378
0.0956
0.953
R (all data)
wR2 (all data)
S (all data)
Variables,ꢀp3
332
1.663
ꢀ1.467
397
3.213
442
0.880
ꢀ1.118
398
0.847
ꢀ1.178
˚
Emax (e Aꢀ3
)
˚
Emin (e A
)
ꢀ3.252
P
P
P
P
2
2 1=2
2
2
a
R ¼ kF o j ꢀ j F ck= jF oj; wR2 ¼ ½ ½wðF 2o ꢀ F c2Þ ꢂ= wðF o2Þ ꢂ ; ðwꢀ1 ¼ ½r2ðF oÞ þ ðaPÞ þ bPꢂ where
a
and
b
are variables and P ¼
P
½0:333ðF oÞ þ 0:667ðF cÞ ꢂÞ; S ¼ ½ ½wðF 2o ꢀ F c2Þ =ðn ꢀ pÞꢂ1=2, (where n is the number of data and p the number of parameters).
2
2
2

D. Ellis et al. / Polyhedron 25 (2006) 915–922
919
the Ph rings stand effectively perpendicular to the nido face
˚
of the carborane. The B–I bonds in 2 are 2.210(5) A for
˚
B9–I9 and 2.195(5) A for B11–I11, essentially the same as
that in the anionic [5-I-7,8-Ph2-7,8-nido-C2B9H9]ꢀ, 2.211
˚
(7) A [7a], but slightly longer than those in the neutral com-
˚
pounds 9,12-I2-1,2-closo-C2B10H10, average 2.178(5) A
˚
[17], 1,2-Ph2-9,12-I2-1,2-closo-C2B10H8, 2.175(5) A [7a],
˚
1,2-Ph2-9-I-1,2-closo-C2B10H9, 2.178(4) A [18], and 9,10-
˚
I2-1,7-closo-C2B10H10, average 2.153(14) A [19].
3.2. Platincarboranes
We have previously noted that, as is logical, metallation
of asymmetrically-substituted carboranes leads to a greater
number of isomerised products than does metallation of
symmetrically-substituted species. Thus, we have chosen
to platinate the 9,11-disubstituted carborane 2 to promote
isomerisation and track its mechanism. Deprotonation of 2
was effected with nBuLi in diethyl ether to produce Li2[7,8-
Ph2-9,11-I2-7,8-nido-C2B9H7]. Addition of a THF solution
of this to a suspension of cis-PtCl2(PMe2Ph)2 in the same
solvent at ꢀ196 ꢁC, followed by warming to room temper-
ature and stirring for 18 h afforded a yellow solution. On
work-up by TLC three products were isolated. The fastest
moving band 3 transpired to be the major component,
being approximately twice as abundant as each of the
two slower bands 4 and 5.
Fig. 1. Perspective view of the anion of [HNMe3][7,8-Ph2-9,11-I2-7,8-nido-
C2B9H10] (2). Atoms are drawn as 50% probability ellipsoids except for H
atoms.
Table 2
˚
Selected interatomic distances (A) and interbond angles (ꢁ) in 2
B1–B2
B1–B3
B1–B4
B1–B5
B1–B6
B2–B3
B2–C7
B2–B11
B2–B6
B3–B4
B3–C8
B3–C7
B4–C8
B4–B9
B4–B5
B5–B9
B5–B10
B5–B6
B6–B10
1.748(8)
1.790(8)
1.776(7)
1.779(8)
1.794(8)
1.754(8)
1.734(7)
1.811(8)
1.791(8)
1.780(8)
1.745(7)
1.733(7)
1.745(7)
1.774(8)
1.762(8)
1.741(7)
1.805(8)
1.823(8)
1.800(8)
B6–B11
C7–C8
C7–B11
C8–B9
1.770(7)
1.610(6)
1.620(7)
1.610(7)
1.805(8)
1.842(7)
1.510(6)
1.509(6)
2.210(5)
2.195(5)
1.17(5)
B9–B10
B10–B11
C7–C71
C8–C81
B9–I9
B11–I11
B10–H10a
B11–H10a
N1–C11
N1–C12
N1–C13
C1S–C11
C1S–C12
C2S–C13
C2S–C13#1
Compound 3 was initially characterised by microanaly-
sis and IR and multinuclear NMR spectroscopy. As is typ-
1
ical of such compounds the H NMR spectrum proved
relatively uninformative in terms of detailed structural
information, simply revealing signals consistent with the
presence of phenyl and methyl H atoms with the correct
relative integrals. 11B–{1H} NMR spectroscopy was also
of limited use: although the chemical shift range observed,
d ꢀ4 to ꢀ24, indicates a closo species (more consistent with
an isomerised 2,1,8-PtC2B9 architecture that with a slipped
3,1,2-PtC2B9 cage [7b,7d]), the relative broadness of the
signals and the presence of significant signal overlap pre-
cludes definitive interpretation and, specifically, prevents
identification of signals which remain uncoupled in the
11B spectrum. The 31P–{1H} NMR spectrum reveals the
1.56(5)
1.488(7)
1.492(7)
1.491(6)
1.730(6)
1.764(7)
1.661(11)
1.743(11)
C71–C7–B11
C71–C7–B2
C71–C7–B3
C71–C7–C8
C81–C8–C7
C81–C8–B3
C81–C8–B4
C81–C8–B9
121.2(4)
I9–B9–C8
I9–B9–B4
I9–B9–B5
I9–B9–B10
I11–B11–B10
I11–B11–B6
I11–B11–B2
I11–B11–C7
121.7(3)
121.6(4)
116.3(4)
117.6(4)
116.4(4)
115.2(4)
120.2(4)
121.6(4)
117.5(3)
121.2(3)
121.1(3)
123.2(3)
120.5(3)
116.8(3)
121.2(3)
presence of inequivalent phosphorus nuclei, each signal
1
showing the expected 195Pt satellites (average JPt–P
=
2
3256 Hz). Careful analysis also reveals residual JP–P cou-
pling, however the value is too small to quote accurately.
The structure of 3 was established by crystallographic
study. Fig. 2 shows a perspective view of one molecule while
Table 3 lists selected interatomic distances and interbond
angles. The compound is thus revealed to be 1,8-Ph2-2,2-
(PMe2Ph)2-6,7-I2-2,1,8-closo-PtC2B9H7, the cage having
undergone a 1,2 ! 1,7 C atom isomerisation (confirming
the interpretation of the 11B chemical shift range). The
substituted (iodine-bearing) boron atoms are B6 and B7.
The metal atom is bound to a CB4 face which is folded into
the usual envelope conformation (dihedral angle between
B3C1B6 and B6B11B7B3 planes 14.0ꢁ) and, moreover,
#1atom generated by the operation 1 ꢀ x, ꢀy, 1 + z.
in the substituted species [C6H5CH2NMe3][5-I-7,8-Ph2-7,8-
˚
nido-C2B9H9] (1.609(10) A) [7a]. In C-adjacent diphenyl-
carborane derivatives we conveniently describe the confor-
mation of the Ph rings with respect to the cage by
parameter hPh, the modulus of the average Ccage–Ccage
–
CPh–CPh torsion angle [16]. The measured values in 2 are
10.0ꢁ for the ring on C7 and 8.0ꢁ for the ring on C8, these
low values quantifying the clear conclusion from Fig. 1 that

920
D. Ellis et al. / Polyhedron 25 (2006) 915–922
Table 3
˚
Selected interatomic distances (A) and interbond angles (ꢁ) in 3–5
3
4
5
Pt2–C1
Pt2–B3
Pt2–B7
Pt2–B11
Pt2–B6
Pt2–P1
Pt2–P2
C1–B3
C1–B4
C1–B5
C1–B6
C1–C11
B3–B4
B3–C8
B3–B7
B4–B5
B4–C8
B4–B9
B5–B6
2.666(7)
2.217(8)
2.202(7)
2.245(8)
2.365(7)
2.326(2)
2.298(2)
1.705(10)
1.647(10)
1.707(10)
1.666(10)
1.505(10)
1.821(11)
1.772(10)
1.864(11)
1.738(11)
1.747(11)
1.759(11)
1.844(11)
1.743(11)
1.775(11)
1.789(1)
2.671(5)
2.234(7)
2.190(6)
2.247(7)
2.355(7)
2.3003(15)
2.3105(15)
1.705(9)
1.655(9)
1.689(9)
1.682(9)
1.507(8)
1.829(9)
1.761(8)
1.896(9)
1.733(9)
1.725(9)
1.765(9)
1.845(10)
1.742(9)
1.785(9)
1.810(9)
1.886(9)
1.710(9)
1.752(9)
1.792(9)
1.774(9)
1.745(9)
1.518(8)
1.792(9)
1.758(9)
1.783(9)
1.760(9)
1.757(9)
2.609(5)
2.251(6)
2.217(6)
2.246(6)
2.266(6)
2.3055(14)
2.2887(14)
1.686(8)
1.657(8)
1.661(8)
1.733(8)
1.495(7)
1.831(8)
1.764(8)
1.871(8)
1.749(8)
1.731(8)
1.761(8)
1.829(8)
1.776(9)
1.799(9)
1.813(9)
1.879(9)
1.721(8)
1.748(8)
1.807(9)
1.767(8)
1.715(8)
1.511(7)
1.792(9)
1.788(9)
1.751(8)
1.774(9)
1.745(9)
B5–B9
B5–B10
B6–B10
B6–B11
B7–C8
B7–B12
B7–B11
C8–B9
C8–B12
C8–C81
B9–B10
B9–B12
B10–B11
B10–B12
B11–B12
1.875(11)
1.716(9)
Fig. 2. Perspective view of 1,8-Ph2-2,2-(PMe2Ph)2-6,7-I2-2,1,8-closo-
PtC2B9H7 (3). Atoms are drawn as 50% probability ellipsoids except for
H atoms.
1.790(10)
1.817(10)
1.744(10)
1.713(10)
1.496(9)
1.780(11)
1.741(11)
1.795(10)
1.774(11)
1.774(10)
˚
the Pt atom is slipped away from C1 (D = 0.47 A) resulting
˚
in a long Pt2–C1 distance, 2.666(7) A, emphasised by its
non-inclusion in Fig. 2. The (electronic) reasons underlying
these slipping and folding distortions are well documented
[20] but we believe that in the present case the slipping is
enhanced by a (steric) contribution arising from the orienta-
tion of the Ph ring on C1. The presence of an adjacent I
atom, I6, requires the plane of this Ph ring to lie perpendic-
ular to the metal-bonded CB4 plane (a convenient measure
of this is f, the modulus of the average Pt2–C1–C11–C tor-
sion angles, f = 66.3ꢁ) pushing back the {Pt(PMe2Ph)2}
fragment.
B–I
B6–I6
B7–I7
B6–I6
B12–I12
B10–I10
B12–I12
2.231(8)
2.173(7)
2.215(7)
2.165(6)
2.181(6)
2.169(6)
Compound 4 was similarly initially characterised by
microanalysis and IR and multinuclear NMR spectros-
copy, and ultimately by X-ray diffraction. The H NMR
P1–Pt1–P2
93.05(7)
96.71(5)
96.61(5)
1
spectrum was broadly similar to that of 3, showing Ph
and Me resonances with the correct relative integrals.
The 11B–{1H} NMR spectrum again revealed resonances
between d ꢀ4 and ꢀ24, indicative of a 2,1,8-closo-PtC2B9
cage. Once more the resonance(s) due to B–I could not
clearly be discerned in the corresponding 11B spectrum.
In the 31P–{1H} spectrum once more were two signals, each
the Ph group on C1 subtends a high f angle, 84.8ꢁ, thus
extending the slip distortion of the {Pt(PMe2Ph)2} frag-
˚
˚
ment to D = 0.48 A, Pt2–C1 = 2.666(7) A.
The final product of the metallation, compound 5,
1
moves slowest on the TLC plate. H NMR spectroscopy
revealed the expected phenyl and methyl resonances and
the 11B–{1H} NMR spectrum showed only four peaks, d
ꢀ6 to ꢀ32, presumably reflecting significant coincidences.
Once more, it was not possible to assign those resonance(s)
due to the substituted B atoms. Asymmetrically disposed
phosphorus atoms were again evident from the observation
of two signals (with attendant Pt satellites) in the 31P–{1H}
1
with Pt satellites (average JPt–Pt = 3262 Hz) and showing
small residual JP–P coupling.
The crystallographic study of 4 (Fig. 3 and Table 3)
reveals a very similar molecular structure to that estab-
lished for 3 except that the second substituted B atom is
now B12 as opposed to B7. Thus, compound 4 is
1,8-Ph2-2,2-(PMe2Ph)2-6,12-I2-2,1,8-closo-PtC2B9H7. The
PtC2B9 cage of 4 resembles that of 3 in many respects:
the Pt-bonded CB4 face is envelope-folded by 15.6ꢁ, and
1
NMR spectrum, the average JPt–P being 3319 Hz.
The result of a crystallographic study of compound 5 is
shown in Fig. 4, and Table 3 again lists key molecular
parameters. This time both substituted B atoms (B10

D. Ellis et al. / Polyhedron 25 (2006) 915–922
921
last atom can adopt an orientation which simply mini-
mises steric repulsion between it and the phosphine
ligands. Thus, f is only 10.9ꢁ, as a result of which both
˚
the slip parameter D, 0.36 A, and the Pt2–C1 distance,
˚
2.609(5) A, are reduced in 5 relative to equivalent values
in 3 and 4.
In compounds 3–5 collectively, the six B–I distances span
˚
the range 2.165(6)–2.231(8) A. The two B6–I6 distances
(compounds 3 and 4) are the two longest, but it is not
obvious why this should be. Certainly the four distances
˚
measured here to B7, B10 and B12, 2.169(6)–2.181(6) A,
˚
correspond well with those, 2.169(9)–2.204(12) A, previ-
ously found in 1,8-Ph2-2,2-(PMe2Ph)2-X-I-2,1,8-closo-
PtC2B9H8, X = 7,10,12 [7d].
3.3. Possible mechanistic implications
In several recent contributions [7] we have interpreted
the results of both 1,2 ! 1,2 and 1,2 ! 1,7 C atom isomeri-
sations in B-substituted nickel- and platinacarboranes in
terms of either one or two triangle face rotations (tfrs),
starting from a crowded, transient, 3,1,2-MC2B9 species
(in one case [7b] this intermediate was actually isolated).
In Scheme 1 we show how the formation of the platina-
carboranes 3 could also be explained by a tfr. In Step 1 the
overcrowded, transient species I is formed by simple com-
plexation of the metal fragment and carborane ligand. To
relieve this overcrowding, a CB2 face in I undergoes a tfr
(Step 2) affording the major product 3. (note that since I
has Cs symmetry either one of two equivalent CB2 faces
could rotate).
Fig. 3. Perspective view of 1,8-Ph2-2,2-(PMe2Ph)2-6,12-I2-2,1,8-closo-
PtC2B9H7 (4). Atoms are drawn as 50% probability ellipsoids except for
H atoms.
and B12) are in the lower half of the molecule, which is
thus identified as 1,8-Ph2-2,2-(PMe2Ph)2-10,12-I2-2,1,8-
closo-PtC2B9H7. Although the upper CB4 face is still
envelope-folded, this time by 15.6ꢁ, the non-adjacency of
a substituted B atom to C1 means that the Ph ring on this
The formations of the minor products 4 and 5 cannot be
so simply rationalised, but two factors appear relevant
here. Firstly, we are fully confident that isomerisation only
n
occurs following metallation since BuLi deprotonation of
2 as described in Section 2.1.3 followed by reprotonation
on addition of aqueous HNMe3Cl regenerates only 2 as
shown by 11B NMR spectroscopy. Secondly, it is likely that
3–5 are formed via parallel rather than sequential isomeri-
sation steps given the mild conditions under which they are
prepared and our subsequent inability to convert from one
to another by heating (either nothing happens or we
observe substantial decomposition).
We thus suggest Step 20, as an alternative to Step 2, in
which one of two possible B3 triangle faces in I rotates to
position one iodine substituent antipodal to a cage C atom,
synchronous with, or followed immediately by, a CB2 tfr
analogous to that in Step 2, this overall process generating
product 4. Moreover, Step 200, with both iodine substituents
moving antipodal to cage C atoms, synchronous with or
followed by CB2 tfr, affords product 5.
We acknowledge that above mechanism is highly specu-
lative, and that it raises several issues. Firstly, we have pre-
viously [7d] considered rotation of a CB2 face involving
two B vertices from the lower pentagon (shown in Step
2) as less likely than one involving one upper and one lower
B atom. Secondly, with regard to Step 20, one could equally
Fig. 4. Perspective view of 1,8-Ph2-2,2-(PMe2Ph)2-10,12-I2-2,1,8-closo-
PtC2B9H7 (5). Atoms are drawn as 50% probability ellipsoids except for H
atoms.

922
D. Ellis et al. / Polyhedron 25 (2006) 915–922
(f) S. Wu, M. Jones, J. Am. Chem. Soc. 111 (1989) 5373;
(g) G.M. Edvenson, D.F. Gaines, Inorg. Chem. 29 (1990) 1210;
(h) B.M. Gimarc, D.S. Warren, J.J. Ott, C. Brown, Inorg. Chem. 30
(1991) 1598;
(i) Y.V. Roberts, B.F.G. Johnson, J. Chem. Soc., Dalton Trans.
(1994) 759;
(j) A.J. Welch, Steric effects in metallacarboranes, in: P. Braunstein,
L.A. Oro, P.R. Raithby (Eds.), Metal Clusters in Chemistry, Wiley/
VCH, New York/Weinheim, 1999, p. 69.
reasonably anticipate the combination of one of the tfrs
shown with the symmetry-equivalent of the other, affording
a
further product 1,8-Ph2-2,2-(PMe2Ph)2-7,10-I2-2,1,8-
closo-PtC2B9H7 (II) but this has not yet been identified.
Clearly a number of questions about the mechanism(s)
of isomerisation that afford(s) compounds 3–5 remain,
and studies designed to answer these are in hand. This is
entirely consistent with the approach to chemical crystal-
lography that Michael Hursthouse instilled in one of us
as a postgraduate student – that molecular structure deter-
mination by crystallography should not be regarded simply
as the end of one chemical investigation but, equally, based
upon what that structure may mean in terms of chemical
bonding or chemical reactivity, the beginning of another.
[2] e.g. H.D. Kaeze, R. Bau, H.A. Beall, W.N. Lipscomb, J. Am. Chem.
Soc. 89 (1967) 4218.
[3] L.I. Zakharkin, V.N. Kalinin, Izv. Akad. Nauk SSSR, Ser. Khim.
(1969) 607.
[4] D.R. Baghurst, R.C.B. Copley, H. Fleischer, D.M.P. Mingos, G.O.
Kyd, L.J. Yellowlees, A.J. Welch, T.R. Spalding, D. OꢀConnell, J.
Organomet. Chem. 447 (1993) C14.
[5] (a) S. Dunn, G.M. Rosair, Rh.Ll. Thomas, A.S. Weller, A.J. Welch,
Angew. Chem. Int. Ed. 36 (1997) 645;
(b) S. Dunn, G.M. Rosair, A.S. Weller, A.J. Welch, Chem. Commun.
(1998) 1065;
P
P
(c) S. Dunn, R.M. Garrioch, G.M. Rosair, L. Smith, A.J. Welch,
Collect. Czech. Chem. Commun. 64 (1999) 1013.
[6] D.J. Wales, J. Am. Chem. Soc. 115 (1993) 1557.
[7] (a) S. Robertson, D. Ellis, T.D. McGrath, G.M. Rosair, A.J. Welch,
Polyhedron 22 (2003) 1293;
Pt
Ph
I
C
I
C
(b) S. Robertson, D. Ellis, G.M. Rosair, A.J. Welch, J. Organomet.
Chem. 680 (2003) 286;
Ph
(c) S. Robertson, D. Ellis, G.M. Rosair, A.J. Welch, Appl. Organo-
met. Chem. 17 (2003) 516;
II
(d) S. Robertson, R.M. Garrioch, D. Ellis, T.D. McGrath, B.E.
Hodson, G.M. Rosair, A.J. Welch, Inorg. Chim. Acta 358 (2005)
1485.
4. Supplementary material
[8] R.M. Garrioch, P. Kuballa, K.S. Low, G.M. Rosair, A.J. Welch, J.
Organomet. Chem. 575 (1999) 57.
[9] M.M. Fein, J. Bobinski, N. Meyers, N. Schwartz, M.S. Cohen,
Inorg. Chem. 2 (1963) 1111.
[10] J.M. Jenkins, B.L. Shaw, J. Chem. Soc. A (1966) 770.
[11] G.M. Sheldrick, SHELXTL version 5.1, Bruker AXS Inc., Madison, WI,
1999.
[12] F.P. Olsen, M.F. Hawthorne, Inorg. Chem. 4 (1965) 1839.
[13] M.A. Fox, A.E. Goeta, J.A.K. Howard, A.K. Hughes, A.L. Johnson,
D.A. Keen, K. Wade, C.C. Wilson, Inorg. Chem. 40 (2001) 173.
[14] G.F. Mitchell, A.J. Welch, J. Chem. Soc., Dalton Trans. (1987)
1017.
Crystallographic data for the structure analyses have
been deposited with the Cambridge Crystallographic Data
Centre, CCDC Nos. 281522–281525 for 2–5, respectively.
Copies of this information may be obtained free of charge
from The Director, CCDC, 12 Union Road, Cambridge
CB2 1EZ, UK (fax: +44 1223 336 033; e-mail depos-
it@ccdc.cam.ac.uk or www: http://www.ccdc.cam.ac.uk).
Acknowledgements
[15] J. Cowie, D.J. Donohoe, N.L. Douek, A.J. Welch, Acta Crystallogr.
C 49 (1993) 710.
[16] J. Cowie, B.D. Reid, J.M.S. Watmough, A.J. Welch, J. Organomet.
Chem. 481 (1994) 283.
[17] A.S. Batsanov, M.A. Fox, J.A.K. Howard, A.K. Hughes, A.L.
Johnson, S.J. Martindale, Acta Crystallogr. C 59 (2003) o74.
[18] T.D. McGrath, M.A. Fox, A.J. Welch, Acta Crystallogr. C 56 (2000)
487.
We thank the Leverhulme Trust (D.E., R.M.G.) for sup-
port, C. Graham for microanalyses and Dr A.S.F. Boyd
for NMR spectra.
References
[1] (a) D. Grafstein, J. Dvorak, Inorg. Chem. 2 (1963) 1128;
(b) W.N. Lipscomb, Science 153 (1966) 373;
[19] Z. Zheng, W. Jiang, A.A. Zinn, C.B. Knobler, M.F. Hawthorne,
Inorg. Chem. 34 (1995) 2095.
(c) L.I. Zakharkin, V.N. Kalinin, Dokl. Akad. Nauk SSSR 169
(1966) 590;
[20] (a) D.M.P. Mingos, M.I. Forsyth, A.J. Welch, J. Chem. Soc., Chem.
Commun. (1977) 602;
(d) B.M. Gimarc, J.J. Ott, Inorg. Chem. 25 (1986) 83;
(e) D.J. Wales, A.J. Stone, Inorg. Chem. 26 (1987) 3845;
(b) D.M.P. Mingos, M.I. Forsyth, A.J. Welch, J. Chem. Soc., Dalton
Trans. (1978) 1363.