Trends in ortho-carboranes 1-X-2-R-1,2-C2B10H10 (R = Ph, Me) bearing an exo-CN-bonded substituent group (X = NO, N{double bond, long}NR′ or NHR′′)

Article abstract of DOI:10.1016/j.poly.2009.04.041

The preparation and crystal structures of four ortho-carboranyl-nitrogen compounds, PhCb^oN{double bond, long}N(C₆H₄Me-4) (1), PhCb^oNHNH(C₆H₄Me-4) (2), MeCb^oNHNHPh (3) and PhCb

Full text of DOI:10.1016/j.poly.2009.04.041

Polyhedron 28 (2009) 2359–2370
Contents lists available at ScienceDirect
Polyhedron
journal homepage: www.elsevier.com/locate/poly
Trends in ortho-carboranes 1-X-2-R-1,2-C2B10H10 (R = Ph, Me) bearing an
exo-CN-bonded substituent group (X = NO, N@NR0 or NHR00)
a
b
Mark A. Fox a, , Richard J. Peace , William Clegg , Mark R.J. Elsegood b,1, Kenneth Wade a,
*
*
a Chemistry Department, Durham University Science Laboratories, South Road, Durham DH1 3LE, UK
b School of Chemistry, Newcastle University, Newcastle Upon Tyne NE1 7RU, UK
a r t i c l e i n f o
a b s t r a c t
Article history:
The preparation and crystal structures of four ortho-carboranyl-nitrogen compounds,
PhCboN@N(C6H4Me-4) (1), PhCboNHNH(C6H4Me-4) (2), MeCboNHNHPh (3) and PhCboNHOH (4)
(Cbo = 1,2-C2B10H10; nitrogen groups at cage carbon C1, Ph or Me at C2), the last as a 1,4-dioxane solvate,
are reported. Comparisons of their structures with those of other ortho-carboranyl-nitrogen systems
studied earlier reveal further correlations between their cage C–C and exo-C–N bond distances and bond
orders. Substituent orientations and bond distances (cage C1–C2, exo-C1–N) in RCboNHR00 systems (R = Ph
Received 19 December 2008
Accepted 30 April 2009
Available online 8 May 2009
Keywords:
Carborane
Amine
Nitroso
Benzene
Cluster
or Me at C2) are consistent with dative
otherwise responsible for cage C1–C2
p
-bonding from a nitrogen lone pair into the cage carbon p-AO
r
bonding. Their C1–C2 bond distances are remarkably sensitive
to the planar (sp2) or pyramidal (sp3) nature of the NHR00 group. The N@O and N@NR0 residues in RCboX
prefer to be orientated in plane with the cage C1–C2 in contrast to the RCboNHR00 systems. Correlations
between their cage C–C and exo-C–N bond distances and the 11B NMR chemical shifts of their antipodal
p-Bonding
boron atoms reflect the
p
-bonding characteristics of the nitrogen substituent.
Ó 2009 Elsevier Ltd. All rights reserved.
1. Introduction
The extent to which the cage C1–C2 bond is affected by exo p-
bonding will therefore depend on the orientation of the substituent
in the exo-CN systems explored here. Though many ortho-carbo-
rane derivatives with exo-C–N bonds are known, only three struc-
tural studies have been carried out elsewhere to our knowledge:
The remarkable capacity of
p-donor substituents, attached to
carbon atoms of an ortho-carborane C2B10 cage, to influence cage
C1–C2 bond lengths was first detected some two decades ago [1]
with the structural characterisation of the proton sponge salt of
the anion PhCboOꢀ, (Cbo = 1,2-C2B10H10; Oꢀ at cage carbon C1, Ph
at C2) followed by many related studies [2,3–7] on ortho-carbor-
anes containing thiolato and phosphino groups. However, this area
has only recently been documented systematically by experimen-
tal and computational studies on systems RCboX and XCboX in
i
the first was on a rhenium complex PrCboN2Re(CO)4 [10], which
contains a 6-membered –Re–N@N–C–B–H– ring in which the me-
tal atom is attached to one nitrogen atom and to a boron-attached
hydrogen; the second was on a hydrazocarborane, HCboNRNHR
t
(R = CO2 Bu) [11], which contains an intramolecular cage C–Hꢁ ꢁ ꢁO
hydrogen bond [12]; and the third was on a zirconium complex,
PhN3CboZrCp2 [13], which contains a 3-membered ZrN2 ring. All
three systems thus contain intramolecular interactions that influ-
ence the orientation about nitrogen at C1 with respect to the cage
C1–C2 bond.
which X is a potential
rived therefrom by deprotonation [8,9]. Exo-C@X
p
-donor such as NH2, OH, SH or anions de-
-bonding in
p
these systems between the cage carbon atom and substituent X in-
volves a tangentially oriented p-AO on carbon that would other-
wise be involved in cage bonding, and C1–C2 bond lengthening
will occur if the p-AO used for exo-C@X p-bonding is the p-AO that
in ortho-carborane itself is involved in C1–C2
In 2004, we reported the crystal structures of PhCboNH2 and the
adduct PhCboNH2ꢁOP(NMe2)3 which revealed six independent mol-
ecules with C1–C2 bond distances ranging between 1.74 and
1.85 Å [8]. We also reported improved syntheses of ortho-carbo-
rane nitroso derivatives RCboNO and dicarboranylamines
(RCbo)2NH (R = Ph, Me) and discussed their structures, which in
the case of the secondary amines (and amides [(RCbo)2N]ꢀ derived
therefrom) showed significant cage distortion (C1–C2 bond length-
r-bonding (AO = a-
tomic orbital; Fig. 1).
* Corresponding authors.
E-mail addresses: m.a.fox@durham.ac.uk (M.A. Fox), kenneth.wade@durham.
ac.uk (K. Wade).
Present address: Chemistry Department, Loughborough University, Loughbor-
ough, Leicestershire LE11 3TU, UK.
ening) attributable to exo-C@N p-bonding [14–16]. To supplement
these studies, we have carried out a synthetic, spectroscopic,
structural and computational investigation of the compounds
1
0277-5387/$ - see front matter Ó 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.poly.2009.04.041

2360
M.A. Fox et al. / Polyhedron 28 (2009) 2359–2370
2.1. Synthetic aspects
X
Scheme 1 summarises the experimental procedures used for the
syntheses of the carboranes investigated in this definitive study.
Synthetic methods that have not been reported in our earlier pa-
pers [8,14] are described in detail in Section 4. The azocarboranes
were synthesised using a reported literature procedure [17]. The
reductions of the nitroso-carboranes RCboNO with hydrogen using
a palladium/carbon catalyst gave the known [18,19] hydroxylam-
ines RCboNHOH in high yields. High-yield reductions of the azocar-
boranes RCboNNAr to the hydrazines RCboNHNHAr were carried
out here using the reducing agents LiAlH4 and Zn/HCl.
C1
C2R
Fig. 1. Orbitals involved in the exo
group and R is not a donor group.
p p-donating
-bonding for RCboX where X is a
RCboX (R = Ph, Me; X = NNR0, NHNHR0, NHOH; R0 = Ph or C6H4Me-
4), and further spectroscopic and computational studies on
systems with X = NH2, NO, NHCboR and [NCboR]ꢀ, which have
revealed hitherto unremarked characteristics and trends in ortho-
carborane systems RCboX containing exo-CN units, which we
report here.
2.2. Structural aspects: new experimentally determined structures
2.2.1. The azo-carborane PhCboN2(C6H4Me-4) (1) and hydrazo-
carboranes PhCboNHNH(C6H4Me-4) (2) and MeCboNHNHPh (3)
The crystal structures of these three compounds were deter-
mined by X-ray diffraction (for details, see Section 4). Their molec-
ular structures are illustrated in Figs. 2 and 3.
The structures of the azo-carborane (1) and its hydrazo ana-
logue (2) (Fig. 2) are similar, differing significantly, and as ex-
pected, only in the bond lengths and angles in the region of their
–N@N– and –N(H)N(H)– units. Their NN links, at 1.250(2) in (1)
and 1.409(2) Å in (2), are of normal length for double and single
bonds, respectively, between nitrogen atoms. The CNN bond an-
gles, at both ends of the N(1)–N(2) links, appear not to differ be-
tween the two ends (implying that the link to the carboranyl
residue resembles that to the aryl group in both 1 and 2) nor be-
tween 1 and 2, suggesting both 1 and 2 contain sp2-hybridised
2. Results and discussion
In this section, we outline the synthetic procedures used to pre-
pare the new compounds, and describe their structures. We then
explore the structural, bonding and spectroscopic characteristics
of ortho-carboranyl-nitrogen systems RCboN@O, RCboN@NR0 and
RCboNHR00 in general, including both the new systems and those
previously characterised [8,14]. We also compare the structural
and bonding relationships of these 3D pseudoaromatic cages with
2D aromatic ring analogues.
KOtBu, 18C6
(RCbo)2NH
[K(18C6)][(RCbo)2N]
dioxane
R = Me 63%
R = Ph 53%
R = Me 86%
R = Ph 77%
NOCl,
Et2O,
-
MeCboNHNHPh
DME
40oC
LiAlH4
82%
ArN2BF4
Et2O
Et2O
RCboNNAr
RCboLi
R = Me, Ar = Ph 50%
R = Ph, Ar = 4-MeC6H4 82%
Et2O,
-
40oC
NOCl,
pentane
EtOH
Zn, HCl
PhCboNHNHC6H4Me
89%
H2, Pd/C
dioxane
RCboNHOH
RCboNO
R = Me 69%
R = Ph 61%
R = Me 95%
R = Ph 87%
Sn, HCl
DME
Cbo = -1,2-CB10H10C-
RCboNH2
R = Me 79%
R = Ph 57%
Scheme 1. Routes to carboranyl-nitrogen derivatives.

M.A. Fox et al. / Polyhedron 28 (2009) 2359–2370
2361
(1)
(2)
Fig. 2. Molecular structures of PhCboN2C6H4Me (1) and PhCboNHNHC6H4Me (2). Selected bond distances (Å) and angles (°) for (1): N(1)–N(2) 1.250(2); N(2)–C(9) 1.428(3);
C(2)–C(3) 1.509(3); N(1)–N(2)–C(9) 112.88(15); N(1)–N(2)–C(9)–C(14) 15.3(3). For (2): N(1)–N(2) 1.409(2); N(2)–C(9) 1.421(3); C(2)–C(3) 1.497(3); N(1)–N(2)–C(9)
115.73(16); N(1)–N(2)–C(9)–C(14) ꢀ25.7(3).
nitrogen atoms. In both 1 and 2, the phenyl substituent on C(2) lies
in a plane roughly perpendicular to the N(1)–C(1)–C(2) plane, as
would be expected not only on steric grounds (to keep it away
from the substituent on C(1)), but also as this is the orientation
The molecular structure of compound 3, MeCboNHNHPh, shown
in Fig. 3, differs from that of 2 in the orientation of the aryl-hydrazo
group, away from the methyl substituent on C(2). However, the lo-
cal geometry about N(1) resembles that in 2 in ensuring that the
lone pair on N(1) lies in the N(1)–C(1)–C(2) plane.
that optimises p-donation from ring to cage and causes ca. 0.03 Å
C1–C2 bond lengthening [20–23].
The orientations of the azo and hydrazo substituents on C(1),
however, differ significantly. In 1, the azo N(1)–N(2)–C(9) unit lies
in the N(1)–C(1)–C(2) plane, as in the nitroso carborane PhCboNO
studied previously [14]. In 2, the hydrazo N(1)–N(2)–C(9) unit is
twisted out of the C(1)–C(2)–C(3) plane, clearly not on steric
grounds, but to an extent that significantly ensures that the lone
pair on N(1) lies in that plane. The view of 2 in Fig. 2 shows the
2.2.2. The hydroxylamino-carborane PhCboNHOH (4)
The molecular structure of this compound (Fig. 4) was deter-
mined by single-crystal studies on a dioxane hemisolvate of com-
position PhCboNHOHꢁ0.5dioxane. The orientations of both
substituents on the carborane cage were found to resemble those
already discussed for compounds 2 and 3, and are in line with
those found previously for PhCboNH2 [8]. The phenyl group in 4 lies
roughly perpendicular to the N(1)–C(1)–C(2) plane, and the HNOH
ortho hydrogen atom on C(14) lying over, and attracted to, the
p-
cloud of the phenyl group on C(2), with an Hꢁ ꢁ ꢁcentroid distance
of 2.65 Å, and this attraction is likely to be responsible for the ori-
entation of the tolyl group.
Fig. 3. Molecular structure of MeCboNHNHPh (3). Selected bond distances (Å) and
angles (°): N(1)–N(2) 1.395(2), N(2)–C(9) 1.405(2), C(2)–C(3) 1.511(2), N(1)–N(2)–
C(9) 119.17(13), N(1)–N(2)–C(9)–C(14) ꢀ19.0(2).
Fig. 4. Molecular structure of PhCboNHOH (4). The dioxane molecule in the crystal
structure is not shown. Selected bond distances (Å): N(1)–O(1) 1.435(2); C(2)–C(3)
1.506(3).

2362
M.A. Fox et al. / Polyhedron 28 (2009) 2359–2370
Fig. 5. Supramolecular structure of PhCboNHOHꢁ0.5dioxane. Selected intermolecular distances (Å) and angles (°): O(1)ꢁ ꢁ ꢁO(2) 2.752(2), N(1)ꢁ ꢁ ꢁO(1A) 3.209(2), O(1)–
H(2)ꢁ ꢁ ꢁO(2) 177(3), N(1)–H(1)ꢁ ꢁ ꢁO(1A) 167(2). [O(1A) is generated from O(1) by screw-axis symmetry].
unit is oriented away from the phenyl group, like the HNNHR unit
in 2, implying that the lone pair on the nitrogen lies in the N(1)–
C(1)–C(2) plane. In the crystal structure, intermolecular N–Hꢁ ꢁ ꢁO
hydrogen-bonding interactions between the hydroxylamino units
link the molecules into (PhCboNHOH)n chains, which in turn are
interlinked by N–OHꢁ ꢁ ꢁO hydrogen bonds to both oxygen atoms
of the dioxane molecules (Fig. 5).
substituent R (Ph or Me) on the other cage carbon atom C(2). This
orientation not only minimises steric non-bonding repulsive inter-
actions between the groups on C(1) and C(2), but aligns the lone
pair of electrons on the exo-nitrogen atom on C(1) in the C(2)–
C(1)–N(1) plane for RCboNHR00, the alignment best suited for dative
p
-bonding from the nitrogen lone pair into the p-AO on C(1)
responsible for cage C(1)–C(2) -bonding, as shown in Fig. 6.
The extent to which exo-dative C@N -bonding donation from
the NHR00 group occurs (and whether it does so at the expense of
cage C(1)–C(2) -bonding) can be inferred from the experimental
r
p
2.3. Structural aspects: general characteristics of RCboN systems
r
The structures of the four compounds 1–4 described above, to-
gether with the four in our previous paper [14] [PhCboNO,
(PhCbo)2NH, (MeCbo)2NH and the anion (PhCbo)2Nꢀ] and the amine
PhCboNH2 characterised earlier [8], provide a useful data bank
from which to deduce some common characteristics of ortho-carb-
oranyl-nitrogen systems. These include preferred orientations of
the exo-nitrogen substituent with respect to the carborane cage,
and systematic complementary trends in their exo-C(1)–N(1) and
cage C(1)–C(2) bond distances; as the former shorten, the latter
lengthen.
The compounds we have studied are of two formula types,
RCboNHR00 and RCboN@Z. Their structures are shown in Figs. 6
and 7. The nitroso- and azo-carboranes PhCboN@O and
PhCboN@N(C6H4Me-4) are of the latter type [14]. All of the other
compounds are of the former type, including the anion (PhCbo)2Nꢀ
in which the lone pair left on deprotonation of the parent amine
can be regarded as occupying the site vacated by proton.
Figs. 6 and 7 show the NHR00 and N@Z units in their preferred
orientations about the exo-C(1)–N(1) bond with respect to the car-
borane cage, with the substituents R00 or Z leaning away from the
C(1)–C(2) and C(1)–N(1) distances in Table 1, in which compounds
are listed in the order of their increasing cage C(1)–C(2) distances,
which matches the sequence of decreasing C(1)–N(1) distances.
Table 1 includes also torsion angles C(2)–C(1)–N(1)–Z/R00, which
reveal that these compounds show some departures from the pre-
ferred orientations shown in Figs. 6 and 7; bond angles C(2)–C(1)–
N(1), which show expected deviations from a normal angle of
121.7° for an exo-bond on a regular icosahedron; and the C(1)–
N(1)–Z/R00 bond angles at nitrogen, which for species RCboNHR00
would be 120° for an ideal trigonal planar coordination at nitrogen
(sp2) and 109° for pyramidal nitrogen (sp3). The data in Table 1 in-
clude the four distinct molecules of PhCboNH2 in the asymmetric
unit of the crystal structure and for the two distinct molecules of
this same species in its hydrogen-bonded adduct with hexameth-
ylphosphoramide, PhCboNH2ꢁHMPA [8].
The range over which cage C(1)–C(2) bond distances vary in Ta-
ble 1 (1.677(2)–1.995(3) Å, i.e. 0.32 Å) is more than twice the range
over which exo-C@N distances vary (1.490(2)–1.345(4) Å i.e.
0.15 Å), because the former are fractional-order bonds becoming
progressively weaker as the table is descended, whereas the latter
sp 2 lone pair orbital
p lone pair orbital
Z
sp2 N
R''
H
sp3 N
sp3 N
R''
R''
H
Z
H
N
N
N
N
N
C
C
C
C
C
CR
C
R
CR
CR
CR
Fig. 7. Preferred orientation in RCboN@Z systems (Z = O, NR0). The cage may act as a
p-donor.
Fig. 6. Preferred orientations in RCboNHR00 systems. The cage acts as a
p-acceptor.

M.A. Fox et al. / Polyhedron 28 (2009) 2359–2370
2363
Table 1
X-ray-determined trends: the optimum torsion angle C(2)–C(1)–N(1)–Z/R00 to align the p-orbital lone pair of sp2-hybridised nitrogen with the cage C(1)–C(2) bond is 90° for
compounds RCboNHR00, and to align the sp2-orbital lone pair of sp2-hybridised nitrogen with the cage C(1)–C(2) bond is 180° for compounds RCboN@Z. HMPA = OP(NMe2)3.
X
C(1)–C(2) (Å)
C(1)–N(1) (Å)
C(2)–C(1)–N(1) (°)
C(1)–N(1)–Z/R00 (°)
C(2)–C(1)–N(1)–Z/R00 (°)
C(1)–C(2)–C(Ph)–C(Ph) (°)
Reference
PhCboX
NO
N@NC6H4Me
NHOH
1.677(2)
1.694(2)
1.737(3)
1.745(3)
1.765(3)
1.774(3)
1.785(3)
1.778(3)
1.794(3)
1.799(3)
1.818(8)
1.853(8)
1.980(3)
1.995(3)
1.490(2)
1.443(2)
1.423(3)
1.391(3)
1.403(3)
1.404(3)
1.392(2)
1.401(2)
1.404(2)
1.404(2)
1.360(8)
1.363(9)
1.355(4)
1.345(4)
112.1(2)
111.3(1)
115.7(2)
119.3(2)
115.9(2)
119.8(2)
118.5(2)
115.2(2)
117.3(2)
116.9(2)
116.1(6)
114.9(5)
118.8(2)
118.8(2)
113.0(2)
112.8(2)
110.7(2)
164.7
178.2
101.6
82.4
75.5
86.5
69.9
91.2
64.1
76.7
83.4
70.1
73.9
59.3
81.8
53.6
53.2
[14]
This work
This work
[8]
NH2 (4)
A
B
C
D
NHNHC6H4Me
NHCboPh
116.4(2)
132.0(2)
105.9
92.4
97.6
This work
[14]
NH2ꢁHMPA (2)
A
B
[8]
[NCboPh]ꢀ
127.0(2)
90.3
89.0
[14]
C(1)–C(2) (Å)
C(1)–N(1) (Å)
C(2)–C(1)–N(1) (°)
C(1)–N(1)–Z/R00 (°)
C(2)–C(1)–N(1)–Z/R00 (°)
Reference
MeCboX
NHCboMe
1.748(4)
1.752(4)
1.770(2)
1.409(4)
1.410(4)
1.388(2)
117.3(2)
117.0(2)
115.9(1)
131.1(2)
131.1(2)
118.8(1)
95.0
94.7
104.1
[14]
NHNHPh
This work
Table 2
Observed bond distances (Å) and calculated bond orders.
X
C(1)–C(2)
Bond order
C(1)–N(1)
Bond order
p-Bond order
PhCboX
NO
N@NAr
NHOH
NHNHAr
NHCboPh
NH2 (C)
NH2ꢁHMPA (av)
[NCboPh]ꢀ
1.677(2)
1.694(2)
1.737(3)
1.778(3)
1.798(3)
1.774(3)
1.835(8)
1.987(3)
0.605
0.590
0.539
0.485
0.453
0.456
0.396
0.235
1.490(2)
1.444(2)
1.423(3)
1.401(2)
1.404(2)
1.404(3)
1.362(8)
1.350(4)
0.921
0.998
1.025
1.044
1.038
1.102
1.146
1.295
0.056
0.070
0.073
0.104
0.124
0.143
0.185
0.357
MeCboX
NHCboMe (av)
NHNHPh
1.750(4)
1.770(2)
0.506
0.486
1.410(4)
1.387(2)
1.026
1.057
0.113
0.128
range in bond order from single to multiple. To aid comparison
within the series, AM1 calculations have been carried out using
the atomic coordinates determined by X-ray diffraction, without
further optimisation. Selected bond distances and orders, with cor-
as illustrated by the data in Table 3, which lists the experimental
lengths of all the 2-centre links in the immediate environment of
C(1); Fig. 8 illustrates that pentagonal pyramidal environment.
From Table 3 it is clear that, although there are minor variations
between compounds in the measured lengths of C(1)–B(3/6),
C(1)–B(4,5), C(2)–B(3,6), B(3,6)–B(4,5) and B(4)–B(5), these varia-
tions cannot be regarded as systematic or significant.
responding
r and p contributions, are given in Table 2.
The data in Table 2 show the correlation between the cage C(1)–
C(2) and exo-C(1)–N(1) bond orders expected from their lengths.
The bond orders of the cage Cꢁ ꢁ ꢁC bonds decrease as those of the
exo-C(1)–N(1) bonds increase. The range over which the C(1)–
C(2) bond orders vary (from 0.61 in PhCboNO to 0.24 in the anion
(PhCbo)2Nꢀ) is comparable to that over which the exo-C(1)–N bond
orders vary (from 0.92 to 1.30), and is itself worthy of comment.
Because icosahedral carborane clusters are held together by only
13 skeletal electron pairs spread around their 30 edge ‘bonds’,
the average cage edge bond order will be 0.43 (13/30). The cage
carbon–carbon bond in ortho-carborane HCboH itself has an order
some 50% in excess of this because the carbon atoms are more elec-
tronegative than their boron neighbours and so attract a greater
share of the electrons available. The data in Table 2 show that
the cage C(1)–C(2) bond order has been reduced to about this aver-
age icosahedral value of 0.43 in the amine PhCboNH2, and to
roughly half this value in the anion (PhCbo)2Nꢀ.
However, on closer inspection, the C1–C2 bond lengths deter-
mined cannot simply be explained by the orientation of the
p-
donating groups. For example, the two compounds 2 and 3 contain
similar C1–C2 bond lengths of ca. 1.77 Å even though it is generally
accepted that the phenyl group lengthens the bond by ca. 0.03–
0.05 Å compared to the methyl group due to the steric and/or elec-
tronic effect(s) of the former [24,25]. The difference of 0.04 Å for
the C1–C2 bond lengths in 2 and 4 is significant even though the
orientations of the nitrogen groups are similar. There are six dis-
tinct PhCboNH2 molecular geometries experimentally determined,
with C1–C2 bond lengths ranging from 1.745 to 1.853 Å (see Table
1) – a difference of 0.11 Å, which cannot be explained solely by the
orientation of the amine group.
2.4. Structural aspects: computational studies
In our discussion of bond lengths and bond order so far, we have
concentrated on the cage C(1)–C(2) and exo-C(1)–N(1) bonds. This
is because these are the only bonds in these systems PhCboX whose
lengths and orders change significantly and systematically with X,
Optimised geometries of carboranes at the MP2/6-31G* level of
theory have been shown to be in excellent agreement with exper-
imental geometries determined by gas-phase electron diffraction

2364
M.A. Fox et al. / Polyhedron 28 (2009) 2359–2370
Table 3
Interatomic distances (Å) for compounds PhCboX.
X
C1–C2
C1–N
C1–B3/6
C1–B4/5
C2–B3/6
B3/6–B4/5
B4–B5
H [26,27]
NO
NNAr
NHOH
NHNHAr
NH2 (av)
1.643(1)
1.677(2)
1.694(2)
1.737(3)
1.778(3)
1.767(3)
1.713(1)
1.706(2)
1.716(3)
1.721(3)
1.719(3)
1.715(4)
1.695(1)
1.703(2)
1.703(3)
1.701(3)
1.708(3)
1.698(4)
1.733(1)
1.740(2)
1.738(3)
1.732(3)
1.736(3)
1.722(4)
1.776(1)
1.786(3)
1.780(3)
1.779(3)
1.786(3)
1.780(4)
1.781(1)
1.779(3)
1.784(3)
1.781(3)
1.784(3)
1.775(4)
1.490(2)
1.443(2)
1.423(3)
1.401(2)
1.396(3)
[28–30] and X-ray crystallography [8,12,20,31,32]. Earlier, we had
tions of the nitrogen atom N1 attached to C1 for NHR00 groups were
also examined.
carried out computations on model geometries of HCboX with
p-
donor groups (X = OH, NHꢀ, NH2 and CH2ꢀ) to investigate the effect
of orientation of these groups on the C1–C2 bond length; in these
computations a planar configuration (sp2) was assumed for NH2
The model carboranes with P-acceptor groups, NO and NNH,
prefer to be oriented in plane (0° and 180°) with the C1–C2 bond,
as opposed to being oriented perpendicular to the C1–C2 bond,
with an energy difference of ca. 4 kcal molꢀ1 (Fig. 8). The reasons
for a slight preference for 0° orientations over 180° orientations
in these models are cage C–Hꢁ ꢁ ꢁX bond interactions [12]. Energet-
ically, the hydroxylamino (NHOH) group clearly favours the pyra-
midal form (sp3 N) over the planar form (sp2 N) irrespective of
the orientation. There is no strong preference for one orientation
over another for the NHOH group. The substituents NO, NNH and
NHOH (with sp3 N) have little orientational influence on the
ortho-carborane cage geometry.
[8]. The effect of a pyramidal form (sp3) at a nitrogen
p-donor
group on the C1–C2 bond or the effect of
p-acceptor groups on
the cage geometry had not been investigated previously. Here, data
for the systems HCboX where X = N@O, N@NH, NHOH, NH2 and
NHNH2 in various possible orientations (indicated by their torsion
angles) are listed in Table 4. The pyramidal or planar configura-
By contrast, the models containing the NH2 groups are similar
in energy irrespective of whether the nitrogen atom N1 is planar
(sp2) or pyramidal (sp3) in configuration (Fig. 7). However, the
90° orientations are energetically favourable and are expected for
Torsion
C2
Angle
B6
B3
a p
-donor group –NHR00 to be oriented to align the donor p-orbital
Plane of
substituent
N=Z or
with the C1–C2 bond for maximum overlap. The model HCboNH2
has four geometries within a range of 2.9 kcal molꢀ1 in energy
but their C1–C2 bond distances vary from 1.623 to 1.713 Å, a dif-
ference of 0.09 Å, attributable to the orientations and the sp2 or
sp3 configuration about the nitrogen atom. Similar observations
are found for the model carborane containing the NHNH2 group.
C1
NHR''
B5
B4
Fig. 8. The pentagonal pyramidal environment about C1.
Table 4
Selected bond lengths (Å) and energies of MP2-optimised geometries of HCboX where torsion angles for C2–C1–N–Z/R00 are fixed at 0, 90 and 180° (a = N–H away from C1, t = N–H
towards C1).
X
N1–Z/R00
C1–N1
C1–C2
C1–B3/6
C1–B4/5
Energy (au)
Relative E (kcal molꢀ1
)
NO
0
90
180
1.230
1.227
1.228
1.491
1.514
1.493
1.621
1.618
1.617
1.713
1.716
1.711
1.691
1.689
1.692
ꢀ459.79796
ꢀ459.79160
ꢀ459.79740
0.0
4.1
0.2
NNH
0
90
180
1.265
1.265
1.265
1.455
1.468
1.457
1.634
1.621
1.624
1.716
1.725
1.715
1.694
1.693
1.698
ꢀ439.98786
ꢀ439.98075
ꢀ439.98616
ꢀ460.99182
ꢀ460.99040
ꢀ460.98922
ꢀ460.98794
0.0
4.7
1.1
NHOH(sp3)
0
1.445
1.440
1.437
1.441
1.450
1.441
1.434
1.452
1.631
1.635
1.665
1.620
1.712
1.727
1.723
1.731
1.693
1.699
1.697
1.699
0.0
0.1
1.7
2.6
90(a)
90(t)
180
NHOH(sp2)
NH2(sp3)
0
90
180
1.396
1.397
1.394
1.397
1.377
1.396
1.609
1.739
1.607
1.749
1.720
1.756
1.704
1.700
1.706
ꢀ460.97107
ꢀ460.97564
ꢀ460.96728
ꢀ386.01371
ꢀ386.01982
ꢀ386.01982
ꢀ386.01650
ꢀ386.00656
ꢀ386.01552
ꢀ441.16414
ꢀ441.16998
ꢀ441.16752
ꢀ441.16800
13.0
10.0
15.6
0
1.019
1.017
1.015
1.020
1.430
1.422
1.412
1.434
1.640
1.651
1.675
1.623
1.731
1.722
1.725
1.732
1.700
1.702
1.700
1.701
4.1
0.0
0.0
2.2
90(a)
90(t)
180
NH2 (sp2)
0
90
1.007
1.008
1.396
1.383
1.612
1.713
1.752
1.718
1.705
1.703
8.9
2.9
NHNH2(sp3)
0
1.417
1.420
1.416
1.428
1.442
1.429
1.414
1.445
1.636
1.654
1.703
1.623
1.728
1.725
1.724
1.736
1.701
1.701
1.699
1.701
3.9
0.0
1.6
1.3
90(a)
90(t)
180
NHNH2(sp2)
0
90
180
1.389
1.399
1.395
1.406
1.386
1.403
1.612
1.731
1.611
1.748
1.721
1.755
1.705
1.701
1.706
ꢀ441.15795
ꢀ441.16639
ꢀ441.15887
8.0
2.4
7.4

M.A. Fox et al. / Polyhedron 28 (2009) 2359–2370
2365
The geometries of PhCboX and MeCboX were also computed at
the MP2/6-31G* level of theory. Comparisons shown in Table 6 re-
veal very good agreements between the selected geometric param-
eters for optimised and experimental geometries in all cases. The
NHR00 groups, as discussed for the model carboranes, are particu-
larly intriguing: the sp3 N form is more stable than the sp2 N form
in these groups. However, for the NH2 group the energy difference
between the geometries containing sp3 N and sp2 N groups is small
(less than 3 kcal molꢀ1), which indicates that both forms could co-
exist in the solid and solution states where intermolecular interac-
tions come into play. The computed C1–C2 distances of 1.743 and
1.813 Å for the sp3 N and sp2 N PhCboNH2 geometries, respectively,
and their energies, are in broad agreement with the observation of
six distinct PhCboNH2 molecules with C(1)–C(2) bond distances of
1.745–1.853 Å (see Table 1) in the crystal structure.
sp2 N
sp3 N
R''
Z
N
H
R''
H
N
N
Fig. 9. Preferred orientations in PhNHR00 and PhN@Z systems.
For compounds with NHNHR0 groups, 2 and 3, the relative ener-
gies between their sp3 N and sp2 N geometries are slightly larger at
ca. 4.0 kcal molꢀ1, but comparison of the bond parameters of
experimental geometries with sp3 N optimised geometries are
poor. The experimental geometry for 3 is in much better agree-
ment with the sp2 N optimised geometry of MeCboNHNHPh,
whereas the experimental geometry for 2 lies between the two
optimised geometries (sp2 N and sp3 N) of PhCboNHNHC6H4Me.
As for PhCboNH2, these geometries depend on intermolecular
interactions such as crystal packing forces. This subtle difference
in the sp2/sp3 N character results in similar C1–C2 bond distances
found experimentally for 2 and 3. The pyramidal sp3 N form is
clearly favoured in energy for the PhCboNHOH geometry and in ac-
cord with the experimental geometry found for 4.
The optimised and experimental geometries for the dicarbora-
nylamines (PhCbo)2NH and (MeCbo)2NH are in excellent agreement
and their planar nitrogen atoms are clearly sp2 in character. Opti-
mised geometries for some methyl analogues, where no experi-
mental structures were determined, reveal geometries and trends
similar to the experimental and optimised geometries for the phe-
nyl analogues.
There are parallels here between carborane chemistry and aro-
matic ring chemistry, where substituents that can act as
or -acceptors adopt orientations that maximise interaction with
the -system of the aromatic ring (Fig. 9) provided that steric fac-
p-donors
p
p
tors do not rule out the preferred orientations of substituents NO,
NNH, NHOH, NH2 and NHNH2 when attached to a benzene ring. Se-
lected bond distances and relative energies from computational
data on these benzene derivatives for the two orientations, planar
(0°) and perpendicular (90°), at the MP2/6-31G* level of theory are
listed in Table 5. For the acceptor groups NO and NNH, the coplanar
form is energetically preferred over the perpendicular form by 8.9
and 5.1 kcal molꢀ1, respectively, but the orientation has little influ-
ence on the ring geometries [33]. The NHOH group strongly fa-
vours the pyramidal (sp3) configuration over the planar (sp2)
form energetically, but small energy and geometry differences be-
tween the two orientations are found for the sp3 form. The amino
NH2 group does not have a strong energetic preference for the
sp2 or sp3 form but has a notable orientational effect on the C–N
bond, with distances of 1.381 Å for the sp2 N geometry in plane
with the C1–C2 bond (0°) and 1.435 Å for the sp3 N geometry per-
pendicular to the C1–C2 bond (90°) [34]. The ring geometry, how-
ever, remains largely unaffected by the C–N bond variations. The
orientation of the NHNH2 group has a similar effect on the geom-
etry and energies as the orientation of the NH2 group. Preferred
orientations determined computationally for PhNNH and
PhNHNH2 are in accord with the orientations of the aromatic rings
in the experimental structures for 1, 2 and 3.
2.5. Experimental and computed NMR trends
The 11B NMR spectra of polyhedral borane clusters such as car-
boranes provide a rich source of information provided that signals
can be assigned with confidence. At the simplest level, the number
of different boron sites in a molecule, deduced from the number of
resonances, helps identify isomers. Of more importance in the
present context is the ‘antipodal effect’ in icosahedral carborane
chemistry [35–39], whereby the NMR shift of the boron atom di-
rectly opposite a substituted cage carbon atom is sensitive to the
nature of the substituent. In ortho-carborane derivatives RCboX,
bearing substituents X and R on carbon atoms 1 and 2, respectively,
the resonances of the atoms opposite (B12 and B9, respectively) re-
spond to the corresponding substituent, particularly if that substi-
Table 5
Selected bond lengths (Å) and energies of MP2-optimised geometries of PhX where
torsion angles for C2–C1–N–Z/R00 are fixed at 0 and 90°.
X
N1–Z/R00
C1–N1
C1–C2
Relative energy (kcal molꢀ1
)
NO
0
90
1.244
1.243
1.443
1.460
1.399
1.394
0.0
8.9
tuent is a
RCboX, where X is a nitrogen
pected to be related to the degree of exo
exo-C1–N1 and cage C1–C2 bond lengths.
p
-donor [2,8]. For the present series of compounds
-donor, the antipodal shift is ex-
-bonding, and hence to
NNH
0
90
1.272
1.270
1.433
1.441
1.399
1.395
0.0
5.1
p
p
NHOH(sp3)
NHOH(sp2)
NH2(sp3)
0
90
1.440
1.465
1.432
1.434
1.399
1.398
0.0
1.5
For such a comparison to be made the 11B NMR shift for the
antipodal atom (11B d-B12) must be reliably assigned. Although
the two peaks corresponding to B9 and B12 are readily identified
from the peak intensities it is impossible to assign each unambig-
uously without assumptions being made. One of these peaks ap-
pears at approximately the same shift in all compounds. This
therefore can be assigned to B9, antipodal to the Ph or Me group
in PhCboX and MeCboX, respectively. This peak would be expected
to change little in these series of compounds as the substituent on
C2, i.e. the Me or Ph group, remains unchanged. 11B NMR shifts for
B9 and B12 have been assigned on this basis (see Table 7).
0
90
1.395
1.412
1.378
1.408
1.403
1.403
7.5
16.9
0
90
1.014
1.019
1.409
1.435
1.403
1.401
0.0
2.4
NH2(sp2)
0
90
1.007
1.005
1.381
1.416
1.405
1.403
1.2
9.2
NHNH2(sp3)
NHNH2(sp2)
0
90
1.410
1.449
1.404
1.432
1.404
1.401
0.0
1.8
0
90
1.396
1.403
1.386
1.415
1.404
1.402
2.4
12.4

2366
M.A. Fox et al. / Polyhedron 28 (2009) 2359–2370
Table 6
Comparison of selected computed and experimental geometric parameters (Å and °) for RCboX systems. Experimental values are shown in italics.
X
C1–C2
C1–N1
C2–C1–N1
C1–N1–X/R00
Angle sum at N
Relative energy (kcal molꢀ1
)
PhCboX
H
1.636
1.643(1)
1.671
NO
1.490
112.1
111.6
1.677(2)
1.679
1.694(2)
1.719
1.490(2)
1.441
1.444(2)
1.430
112.1(2)
111.1
111.3(1)
113.5
113.0(2)
111.3
112.8(2)
110.6
N@NC6H4Me
NHOH
sp3
sp2
322.4
360.0
324.7
360.0
360.0
335.4
360.0
342.4
332.7
360.0
344.3
0.0
8.3
1.828
1.367
116.2
120.9
1.737(3)
1.790
1.798(3)
1.727
1.830
1.778(3)
1.743
1.813
1.767(3)
1.977
1.423(3)
1.401
1.404(2)
1.424
1.379
1.401(2)
1.411
1.374
1.396(3)
1.353
115.7(2)
115.9
117.3(2)
116.4
119.1
115.2(2)
114.5
117.0
118.4(2)
118.7
110.7(2)
132.3
132.0(2)
117.4
120.5
116.4(2)
NHCboPh
NHNHC6H4Me
sp3
sp2
0.0
4.2
NH2
sp3
sp2
0.0
2.6
[NCboPh]ꢀ
126.2
1.987(3)
1.355(4)
118.8(2)
127.0
MeCboX
H
NO
N@NPh
NHOH
1.630
1.638
1.644
1.696
1.774
1.748
1.750(4)
1.737
1.764
1.770(2)
1.683
1.770
1.488
1.444
1.427
1.374
1.406
1.410(4)
1.448
1.384
1.387(2)
1.420
1.379
112.5
112.2
115.4
118.2
117.3
117.2(2)
114.5
117.9
115.9(1)
114.9
118.0
111.9
110.9
113.2
120.7
132.3
131.1(2)
119.9
121.6
118.8(1)
sp3
sp2
0.0
6.0
NHCboMe
NHNHPh
360.0
359.8
338.9
360.0
355.4
330.0
360.0
sp3
sp2
0.0
3.8
NH2
sp3
sp2
0.0
2.7
[NCboMe]ꢀ
1.886
1.358
118.7
126.2
Table 7
Comparison of selected experimental (in CDCl3) and computed NMR shifts (ppm) for RCboX systems. Calculated values are in italics.
X
d(B12)
d(B12H)
d(C1)
d(C2)
PhCboX
H [8]
NO
ꢀ1.2
ꢀ2.0
ꢀ4.2
ꢀ5.5
ꢀ5.8
ꢀ6.8
ꢀ8.3
ꢀ12.6
ꢀ1.2
0.2
2.46
2.67
2.57
2.38
2.28
2.32
2.18
1.72
3.09
3.34
3.18
3.00
2.86
2.85
2.80
2.08
60.1
114.1
98.8
98.4
94.4
102.3
96.3
129.5
57.7
125.1
103.8
101.0
95.7
105.3
98.1
125.4
76.5
81.3
81.7
86.5
90.3
90.0
87.7
89.8
78.5
84.1
83.0
87.8
92.5
91.3
92.2
95.5
N@NC6H4Me
ꢀ2.4
ꢀ4.5
ꢀ5.1
ꢀ6.0
ꢀ7.8
ꢀ13.6
NHOHa
NHCboPh
NHNHC6H4Meb
b
NH2
[NCboPh]ꢀc
MeCboX
H [40]
NO
ꢀ1.7
ꢀ2.4
ꢀ4.2
ꢀ6.1
ꢀ6.4
ꢀ7.1
ꢀ9.4
ꢀ13.7
0.0
0.4
2.31
2.50
2.40
2.24
2.22
2.16
1.98
1.59
3.05
3.29
3.07
2.83
2.78
2.81
2.90
1.99
61.5
111.4
95.7
94.5
90.9
96.7
91.1
120.0
58.6
120.2
97.9
96.6
93.0
98.9
92.5
119.1
70.4
73.3
73.9
78.6
81.8
80.9
78.9
87.0
70.5
74.6
73.8
81.8
84.0
82.2
81.0
90.8
N@NPh
NHOHa
NHCboMe
NHNHPhb
ꢀ2.6
ꢀ5.8
ꢀ6.0
ꢀ6.5
ꢀ8.4
ꢀ14.4
b
NH2
[NCboMe]ꢀc
a
Computed shifts for sp3 geometry.
b
c
Computed shifts averaged for both sp2 and sp3 geometries.
Observed values in CD3CN.
The antipodal shift increases as
p
-bonding increases, displayed
The shifts of the BH protons revealed by 1H{11B} NMR can be as-
signed using 1H–11B HETCOR or 1H{11B selective} experiments. The
1H shifts of the hydrogen atoms bonded to the antipodal boron
show the same relationship with the C–N bond length as the 11B
shifts of B12, decreasing frequency of the antipodal hydrogen with
by the graph of the B12 shift vs experimental C–N bond lengths
(Fig. 10). This is in agreement with previous work showing that
such shifts are related to electron donation to the cluster. Hence
the 11B NMR data give an indication of degree of exo
p-bonding
and this can be used to assess such effects in compounds which
have not been structurally characterised.
increased
p-bonding to the cage suggesting increased shielding.
The 13C shifts of the cage carbon atoms C1 and C2, however, are

M.A. Fox et al. / Polyhedron 28 (2009) 2359–2370
2367
0
-2
PhCboNO
PhCboN2Ar
PhCboNHOH
(MeCbo)2NH
-4
(PhCbo)2NH
11B
antipodal
shift
MeCboNHNHPh
PhCboNHNHAr
-6
-8
PhCboNH2.HMPA
PhCboNH2
-10
-12
-14
(PhCbo)2N-
1.34
1.36
1.38
1.40
1.42
1.44
1.46
1.48
1.50
C-N bond length
Fig. 10. Correlation between the observed 11B shift of B12 (ppm) and the experimental C(1)–N(1) bond length (Å).
not simply related to the exo-C–N
NMR chemical shifts on MP2-optimised geometries are in very
good agreement with observed shifts and trends as shown in Table
7.
p
-bonding effects. Calculated
Melting points were measured in capillary tubes with an Elec-
trothermal 9200 heating block. Infrared spectra were recorded
from KBr discs on Perkin–Elmer 1600 series FTIR or Perkin–Elmer
1720X FTIR spectrometers and ultraviolet spectra with a Shimadzu
UV 1201. Elemental carbon, hydrogen and nitrogen analyses were
performed using Exeter Analytical CE-440 or Carlo Erba Strument-
azione EA Model 1106 instruments. Mass spectra (MS) were re-
corded on a VG Micromass 7070E instrument under EI conditions
at 70 eV. Values of M show the isotope range 10Bn to 11Bn including
3. Conclusions
In this paper we have reported the crystallographically derived
molecular structures and 11B, 13C and 1H NMR spectra of four new
ortho-carboranyl-nitrogen compounds PhCboN@N(p-tolyl) (1),
PhCboNHNH(p-tolyl) (2), MeCboNHNHPh (3) and PhCboNHOH (4)
(Cbo = 1,2-C2B10H10). Together with other carboranyl-nitrogen sys-
tems RCboX reported earlier, these provide a useful data bank from
which the general structural, bonding and NMR characteristics of
such systems can be discerned. Their structures show how dative
exo-C@N p-bonding from the substituent to the cage is reflected
in shortening of the exo-C–N bond (as it gains multiple character)
and lengthening of the cage C1–C2 bond (as it loses bond order) to
a
13C contribution if observed. NMR spectra were measured using
Varian Unity-300 (1H, 11B, 13C), Bruker AM250 (1H, 13C), Bruker
Avance 400 (1H, 11B, 13C) and/or Varian Inova 500 (1H, 11B) instru-
ments. All chemical shifts are reported in d (ppm) and coupling
constants in Hz. 1H NMR spectra were referenced to residual protio
impurity in the solvent (CDCl3, 7.26 ppm). 13C NMR spectra were
referenced to the solvent resonance (CDCl3, 77.0 ppm). 11B NMR
spectra were referenced externally to Et2OꢁBF3, d = 0.0 ppm. Peak
assignments of cage boron and hydrogen atoms were determined
with the aid of 2D 11B{1H}–11B{1H} COSY, selective 1H{11B} and
1H–11B correlation spectra.
an extent that reflects the
p-donor power of the substituent, the
orientation of the CN group and the sp2 or sp3 character of the
nitrogen atom for the NHR00 group, which are seen from computa-
tional studies to be important. The 11B NMR chemical shift of the
boron atom antipodal to the substituent also provides a guide to
4.1. Preparation of azocarboranes (modification of a reported [17]
method)
the p-donor power of X.
A solution of the substituted lithiocarborane was prepared by
the addition of butyllithium (5.05 ml, 2.5 M in hexanes) to a solu-
tion of the starting carborane (0.0127 mol) in diethyl ether
(100 ml) at 0 °C. The solution was warmed to ambient temperature
with stirring for 30 min. The diazonium salt was added as a solid
over a period of 30 min and stirred overnight to give a cloudy
red solution. Water was added and the organic layer was separated
and washed with water. The organic layer was dried and evapo-
rated to leave a brown residue. This was recrystallised from hexane
to give orange crystals of the azocarborane.
1-Methyl-2-phenylazo-ortho-carborane, 1.78 g (50%), M.p.
122.5–123.5 °C (lit. [17] 121–122 °C). Anal. Calc. for C9H18B10N2:
C, 41.2; H, 6.9; N, 10.7. Found: C, 41.1; H, 7.0; N, 10.6%. IR mmax
(KBr) [cmꢀ1]: 3062w (aryl CH), 2926w (methyl CH), 2640m,
2615s, 2582s, 2574s, 2550s (BH), 1497s, 1451s, 1205s, 1157s,
1072s, 1019s, 768s, 728s, 683s, 419s. 1H{11B} NMR (CDCl3), d:
4. Experimental
All air-sensitive manipulations were carried out under dry, oxy-
gen-free N2. Stirring refers to use of a magnetic stirrer. Hexanes
were distilled over Na. 1,2-Dimethoxyethane (DME) was dried by
reflux and distillation over potassium; ether refers to diethyl ether
dried, where appropriate, over sodium. Ether solutions were dried
over magnesium sulfate and evaporated near room temperature. 1-
Methyl-ortho-carborane [41] and 1-phenyl-ortho-carborane [42]
were prepared by literature methods and dried by sublimation at
0.01 mm Hg. 1-Nitroso-2-methyl-ortho-carborane and 1-nitroso-
2-phenyl-ortho-carborane were made as described elsewhere
[14].
Benzenediazonium
tetrafluoroborate
and
4-meth-
ylbenzenediazonium tetrafluoroborate salts were made using a
general literature method [43].

2368
M.A. Fox et al. / Polyhedron 28 (2009) 2359–2370
7.81 (d, 2H, ortho-phenyl CH), 7.58 (t, 1H, para-phenyl CH), 7.52 (d,
2H, meta-phenyl CH), 2.50 (s, 2H, H8,10), 2.40 (s, 5H, BH including
H12), 2.30 (s, 3H, CH3), 2.20 (s, 3H, BH); 11B{1H} NMR (CDCl3), d:
ꢀ4.2 (1B, B12), ꢀ6.4 (1B, B9), ꢀ10.6 (8B); 13C NMR (CDCl3), d:
ꢀ3.9 (1B, B9), ꢀ6.8 (1B, B12), ꢀ10.6 (4B), ꢀ12.6 (4B); 13C{1H}
NMR (CDCl3), d: 144.9 (C–CH3), 131.8, 129.5, 128.8 (aryl CH);
131.0 (ipso C6H5), 130.5 (para C6H5), 130.1 (C–N), 112.5 (CHCN),
102.3 (C1), 90.0 (C2), 20.4 (CH3).
1
2
151.2 (C6H5 ipso), 133.5 (d of t, JCH 161 Hz, JCH 8 Hz, C6H5 para),
1
2
129.4 (d of d, JCH 161 Hz, JCH 8 Hz, C6H5 ortho), 123.6 (d of t,
4.3. Preparation of carboranylhydroxylamines
1JCH 161 Hz, JCH 6 Hz, C6H5 meta), 95.7 (C1), 73.9 (C2), 22.5 (q,
2
1JCH 132 Hz, CH3).
The methyl nitrosocarborane (0.22 g) was dissolved in 10 ml of
p-dioxane, 5% Pd/C (40 mg) added and the solution degassed by a
freeze-pump–thaw process. Hydrogen was admitted and its uptake
measured using a standard hydrogenation apparatus. Twenty-five
milliliters of hydrogen was consumed over a period of 6 h. The
solution was filtered and evaporated to leave a white solid
(210 mg, 95%), M.p. 253–255 °C (lit. [19] 256–258 °C). Anal. Calc.
for C15H24B10N2: C, 19.1; H, 7.9; N, 7.4. Found: C, 19.4; H, 8.1; N,
6.2%. MS (EI+, m/z) [M]+ 185–192; 189 (100); IR mmax (KBr)
[cmꢀ1]: 3340, 3290 (NH,OH), 2914 (methyl CH), 2579s (BH),
1254s, 1116s, 1080m, 870s. 1H{11B} NMR (CDCl3), d: 5.94 (s, 1H,
NH), 5.39 (s, 1H, OH), 2.46 (s, 2H, H3,6), 2.24 (s, 3H, BH), 2.22 (s,
1H, BH), 2.18 (s, 2H, BH), 2.05 (s, 2H, H8,10), 2.02 (s, 3H, CH3);
11B{1H} NMR (CDCl3), d: ꢀ6.1 (2B, B9,12), ꢀ10.9 (4B), ꢀ11.7 (4B);
13C{1H} NMR (CDCl3), d: 94.5 (C1), 78.6 (C2), 21.6 (CH3).
A larger-scale preparation was used for the phenyl analogue to
afford 3.37 g (87%) from 3.86 g of the nitroso compound, M.p. 95–
96 °C, (lit. [18] 98–99 °C). Anal. Calc. for C15H24B10N2: C, 38.2; H,
7.2; N, 5.4. Found: C, 38.3; H, 6.8; N, 5.6%. MS (EI+, m/z) [M]+
247–255; 251 (100); IR mmax (KBr) [cmꢀ1]: 3532, 3463, 3280
(NH,OH), 3061 (phenyl CH), 2574s (BH), 1493m, 1446s, 1071s,
1003s, 689s. 1H{11B} NMR (CDCl3), d: 7.70 (d, 2H, ortho C6H5),
7.47 (t, 1H, para C6H5), 7.40 (t, 2H, meta C6H5), 5.69 (s, 1H, NH),
5.09 (s, 1H, OH), 2.83 (s, 2H, H3,6), 2.53 (s, 2H, BH), 2.42 (s, 3H,
BH), 2.38 (s, 1H, H12), 2.17 (s, 2H, H8,10); 11B{1H} NMR (CDCl3),
d: ꢀ3.8 (1B, B9), ꢀ5.7 (1B, B12), ꢀ10.6 (4B), ꢀ12.1 (4B); 13C{1H}
NMR (CDCl3), d: 131.3, 130.8 (para C), 130.0 (ipso C), 129.1, 98.4
(C1), 86.5 (C2).
1-Phenyl-2-(4-methylphenyl)azo-ortho-carborane, 2.77 g (82%),
M.p. 90–91 °C. Anal. Calc. for C15H22B10N2: C, 53.3; H, 6.5; N, 8.3.
Found: C, 53.1; H, 6.8; N, 8.1%. MS (EI+, m/z) 91 (C6H4Me); 216–
221 [PhCb]+; IR mmax (KBr) [cmꢀ1]: 3052w (aryl CH), 2922w
(methyl CH), 2642m, 2603s, 2566s (BH), 1600s, 1500s, 1489s,
1472s, 1446s, 1157s, 1071s, 1026s, 827s, 810s, 802s, 752s, 688s.
3
1H{11B} NMR (CDCl3), d: 7.70 (d, 2H, ortho C6H5), 7.42 (t, JHH
3
8 Hz, 1H, para C6H5), 7.34 (t, JHH 8 Hz, 2H, meta C6H5), 7.35 (d,
3
3JHH 8 Hz, 2H, meta C6H4Me), 7.17 (d, JHH 8 Hz, 2H, ortho
C6H4Me), 3.00 (s, 2H, BH), 2.63 (s, 2H, BH), 2.57 (s, 3H, BH), 2.44
(s, 1H, BH), 2.37 (s, 3H, CH3), 2.34 (s, 2H, H9,12); 11B{1H} NMR
(CDCl3), d: ꢀ4.2 (2B, B9,12), ꢀ11.0 (8B); 13C NMR (CDCl3), d:
149.1 (C–N), 144.4 (C–CH3), 131.1 (ortho C6H5), 130.5 (ipso C6H5),
130.2 (para C6H5), 129.8, 128.2 (aryl CH); 123.5 (CHCN), 98.8
(C1), 81.7 (C2), 21.6 (CH3).
4.2. Preparation of hydrazocarboranes
Lithium aluminium hydride, LiAlH4, (0.3 g, 8 mmol), was added
to a solution of 1-methyl-2-phenylazo-ortho-carborane (0.50 g,
1.91 mmol) in diethyl ether (20 ml) and the mixture stirred for
20 h. Wet diethyl ether was added, followed by water (10 ml)
and dilute HCl until the cloudy solution became clear. The organic
layer was washed with water (2 ꢂ 50 ml), dried and evaporated to
leave a pale yellow solid. This was recrystallised from hot hexane
to yield colourless crystals of 1-methyl-2-phenylhydrazo-ortho-
carborane (0.41 g, 82%). M.p. 142–144 °C (lit. [17] 143–144 °C).
Anal. Calc. for C9H20B10N2: C, 40.9; H, 7.6; N, 10.3. Found: C, 40.7;
H, 7.8; N, 10.3%. MS (EI+, m/z) [M]+ 260–267; 264 (100); IR mmax
(KBr) [cmꢀ1]: 3360, 3322 (NH), 3125 (phenyl CH), 2933 (methyl
CH), 2615s, 2580s, 2559s (BH), 1601, 1497, 1454m, 1251m,
4.4. Preparation of methyl-ortho-carboranyl amine
The compound 1-nitroso-2-methyl-ortho-carborane (1.50 g)
was dissolved in dimethoxyethane (25 ml) and tin powder
(1.50 g) was added. Concentrated HCl (25 ml) was added drop-
wise and the solution stirred for 15 min, after which time the
blue colour had disappeared. The solution was heated to reflux
for 3 h, cooled to room temperature and diluted with diethyl
ether (100 ml). The solution was washed with water
(3 ꢂ 50 ml), dried and evaporated. The white residue was sub-
limed to yield 1-amino-2-methyl-ortho-carborane (1.10 g, 79%),
M.p. 301–302 °C (lit. [19] 302–303 °C). Anal. Calc. for
C3H15B10N: C, 20.8; H, 8.7; N, 8.1. Found C, 20.9; H, 9.0; N,
7.1%. MS (EI): M, 169–176 (C3H15B10N = 173). I.R. (cmꢀ1): 3306,
3219br (NH), 2939w (methyl CH), 2671s, 2634s, 2602s, 2580s
(BH), 1491m, 1451m, 1221m, 1079s, 1026m, 1002m, 864m,
809m, 764s, 700s. 1H{11B} NMR (CDCl3): 3.00 (2H, s, H4,5),
2.99 (2H, NH2), 2.41 (2H, H7,11), 2.20 (1H, H9), 2.04 (3H, CH3),
2.02 (2H, H3,6), 1.98 (1H, H12), 1.95 (2H, H8,10), 11B NMR
(CDCl3): ꢀ5.5 (1B, d, B9), ꢀ9.4 (1B, d, B12), ꢀ9.8 (2B, d, B4,5),
ꢀ10.6 (4B, d, B3,6,7,11), ꢀ12.5 (2B, d, B8,10). 13C{1H} NMR
(CDCl3): 91.1 (C1), 78.9 (C2), 21.1 (CH3).
3
1019m, 754s, 696 m. 1H{11B} NMR (CDCl3), d: 7.24 (t, JHH 7 Hz,
3
3
2H, meta C6H5), 6.90 (t, JHH 7 Hz, 1H, para C6H5), 6.85 (d, JHH
8 Hz, 2H, ortho C6H5), 5.68 (s, 1H, NH), 4.89 (s, 1H, NH), 2.52 (s,
2H, H8,10), 2.23 (s, 3H, BH), 2.16 (s, 3H, BH), 2.09 (s, 3H, CH3),
2.05 (s, 2H, BH); 11B{1H} NMR (CDCl3), d: ꢀ6.1 (1B, B9), ꢀ7.1 (1B,
B12), ꢀ11.0 (4B), ꢀ11.8 (4B); 13C NMR (CDCl3), d: 147.3 (C6H5 ipso),
129.3 (d, 1JCH 158 Hz, C6H5), 121.0 (d, 1JCH 160 Hz, C6H5), 113.0 (d of
t, 1JCH 156 Hz, C6H5), 96.7 (C1), 80.9 (C2), 22.1 (q, 1JCH 130 Hz, CH3).
The compound 1-phenyl-2-(4-methylphenyl)azo-ortho-carbo-
rane (0.73 g) was dissolved in 20 ml of ethanol. Zinc dust (3.5 g)
was added and 20 ml conc HCl added dropwise. The solution was
stirred overnight, becoming colourless. It was poured into water
(200 ml), giving a white precipitate. The solid was extracted with
diethyl ether (3 ꢂ 50 ml). The organic layer was washed with
water, dried and evaporated. The residue was recrystallised from
hexane to give pale yellow crystals of 1-phenyl-2-(4-methyl-
phenyl)hydrazo-ortho-carborane (0.65 g, 89%). M.p. 148–149.5 °C.
Anal. Calc. for C15H24B10N2: C, 52.8; H, 7.0; N, 8.2. Found: C, 53.0;
H, 7.0; N, 8.1%. MS (EI+, m/z) [M]+ 336–343; 340 (100); IR mmax
(KBr) [cmꢀ1]: 3312s (NH), 3120, 3105 (aryl CH), 2950 (methyl
CH), 2663s, 2628s, 2550s (BH), 1514s, 1262m, 814s, 689s.
4.5. Crystal structure determinations
3
1H{11B} NMR (CDCl3), d: 7.69 (d, JHH 8 Hz, 2H, ortho C6H5), 7.53
3
3
(t, JHH 8 Hz, 1H, para C6H5), 7.40 (t, JHH 8 Hz, 2H, meta C6H5),
Crystals of the compounds 1–4 were examined on Bruker
3
3
6.82 (d, JHH 8 Hz, 2H, meta C6H4Me), 6.17 (d, JHH 8 Hz, 2H, ortho
C6H4Me), 5.28 (br,s, 1H, NH), 4.83 (s, 1H, NHAr), 2.83 (s, 2H,
H3,6), 2.59 (s, 2H, BH), 2.41 (s, 3H, BH incl H9), 2.32 (s, 1H, H12),
2.21 (s, 3H, CH3), 2.14 (s, 2H, H8,10); 11B{1H} NMR (CDCl3), d:
SMART (3) and Stoe STADI4 (1, 2, 4) diffractometers with Mo K
a
radiation (k = 0.71073 Å; Cu K with k = 1.54184 Å for 1) at
a
160 K. Crystal data and other information are given in Table 8.
Standard methods and software were employed, including refine-

M.A. Fox et al. / Polyhedron 28 (2009) 2359–2370
2369
Table 8
Crystal data and refinement information for compounds 1–4.
Compound
1
2
3
4
Formula
M
C15H22B10N2
338.5
C15H24B10N2
340.5
C9H20B10N2
264.4
C8H17B10NOꢁ½C4H8O2
295.38
Crystal system
Space group
monoclinic
C2/c
monoclinic
P21/c
monoclinic
P21/n
monoclinic
P21/c
a (Å)
b (Å)
c (Å)
28.695(6)
7.071(2)
19.008(5)
96.34(3)
3833.2(17)
8
11.753(6)
7.617(4)
21.899(13)
104.74(6)
1895.9(18)
4
8.6912(11)
15.568(2)
11.5680(15)
96.576(3)
1554.9(3)
4
12.394(1)
6.714(1)
19.498(2)
94.30(2)
1617.9(3)
4
b (°)
V (Å3)
Z
Data collected
Unique data
Rint
6424
2992
0.046
4051
3329
0.049
8134
3099
0.056
4131
2846
0.031
Refined parameters
246
0.053
0.154
0.25, ꢀ0.23
252
0.049
0.137
0.24, ꢀ0.21
271
0.049
0.126
0.23, ꢀ0.20
217
0.048
0.134
0.25, ꢀ0.21
R (on F, F2 > 2
r)
Rw (on F2, all data)
Minimum, maximum electron density (e Åꢀ3
)
ment on all F2 values [44]; no absorption corrections were applied,
References
and no structural disorder was found.
[1] D.A. Brown, W. Clegg, H.M. Colquhoun, J.A. Daniels, I.R. Stephenson, K. Wade, J.
Chem. Soc., Chem. Commun. (1987) 889.
[2] R. Coult, M.A. Fox, W.R. Gill, K. Wade, W. Clegg, Polyhedron 11 (1992) 2717.
[3] R. Kivekäs, R. Sillanpää, F. Teixidor, C. Viñas, R. Nuñez, Acta Crystallogr., Sect. C
50 (1994) 2027.
5. Computational
AM1 [45,46] calculations were carried out on crystallographi-
cally determined geometries using MOPAC2002 (Version 2.40)
within the CACHE 6.1 program for Windows [47]. Ab initio computa-
tions were carried out with the GAUSSIAN 03 package [48]. All model
geometries of HCboX and PhX with fixed torsion angles for C2–C1–
N–R/X listed in Tables 4 and 5, respectively, were optimised
initially at the HF/6-31G* level of theory followed by the MP2/6-
31G* level of theory. The geometries of PhCboX and MeCboX listed
in Table 6 were optimised at the HF/6-31G* level of theory either
with no symmetry constraints (and confirmed by frequency calcu-
lations to be a true minimum) or with the nitrogen atom N1 con-
strained to a planar configuration. These geometries were then
optimised at the MP2/6-31G* level of theory.
[4] R. Kivekäs, R. Sillanpää, F. Teixidor, C. Viñas, R. Nuñez, Acta Crystallogr., Sect. C
51 (1995) 1864.
[5] R. Kivekäs, F. Teixidor, C. Viñas, R. Nuñez, Acta Crystallogr., Sect. C 51 (1995)
1868.
[6] J.M. Oliva, C. Viñas, J. Mol. Struct. 556 (2000) 33.
[7] J. Llop, C. Viñas, J.M. Oliva, F. Teixidor, M.A. Flores, R. Kivekäs, R. Sillanpää, J.
Organomet. Chem. 657 (2002) 232.
[8] L.A. Boyd, W. Clegg, R.C.B. Copley, M.G. Davidson, M.A. Fox, T.G. Hibbert, J.A.K.
Howard, A. Mackinnon, R.J. Peace, K. Wade, Dalton Trans. (2004) 2786.
[9] J.M. Oliva, N.L. Allan, P.v.R. Schleyer, C. Viñas, F. Teixidor, J. Am. Chem. Soc. 127
(2005) 13547.
[10] A.I. Yanovskii, Y.T. Struchkov, V.N. Kalinin, A.V. Usatov, L.I. Zakharkin, Koord.
Khim. 8 (1982) 240.
[11] J.F. Valliant, O.O. Sogbein, P. Morel, P. Schaffer, K.J. Guenther, A.D. Bain, Inorg.
Chem. 41 (2002) 2731.
[12] M.A. Fox, A.K. Hughes, Coord. Chem. Rev. 248 (2004) 457.
[13] L. Deng, H.S. Chan, Z. Xie, J. Am. Chem. Soc. 127 (2005) 13774.
[14] M.A. Fox, J.A.H. MacBride, R.J. Peace, W. Clegg, M.R.J. Elsegood, K. Wade,
Polyhedron 28 (2009) 789.
[15] M.A. Fox, J.A.H. MacBride, R.J. Peace, K. Wade, J. Chem. Soc., Dalton Trans.
(1998) 401.
[16] L.A. Boyd, H.M. Colquhoun, M.G. Davidson, M.A. Fox, W.R. Gill, P.L. Herbertson,
T.G. Hibbert, J.A.H. MacBride, R.J. Peace, G.P. Rivers, K. Wade, Spec. Publ. R. Soc.
Chem. 201 (1997) 289.
[17] V.N. Kalinin, G.G. Zhigareva, L.I. Zakharkin, Synth. Inorg. Met.-Org. Chem. 2
(1972) 105.
[18] L.I. Zakharkin, V.N. Kalinin, Dokl. Akad. Nauk SSSR 164 (1965) 577.
[19] L.I. Zakharkin, G.G. Zhigareva, Zh. Obshch. Khim. 45 (1975) 1293.
[20] E.S. Alekseyeva, M.A. Fox, J.A.K. Howard, J.A.H. MacBride, K. Wade, Appl.
Organomet. Chem. 17 (2003) 499.
[21] M. Tsuji, J. Org. Chem. 69 (2004) 4063.
[22] I.V. Glukhov, K.A. Lyssenko, A.A. Korlyukov, M.Y. Antipin, Faraday Discuss. 135
(2007) 203.
[23] I.V. Glukhov, M.Y. Antipin, K.A. Lyssenko, Eur. J. Inorg. Chem. 7 (2004) 1379.
[24] ZG. Lewis, A.J. Welch, Acta Crystallogr., Sect. C 49 (1993) 705.
[25] T.D. McGrath, A.J. Welch, Acta Crystallogr., Sect. C 51 (1995) 646.
[26] R.L. Thomas, G.M. Rosair, A.J. Welch, Acta Crystallogr., Sect. C 52 (1996) 1024.
[27] I.V. Glukhov, K.A. Lyssenko, A.A. Korlyukov, M.Y. Antipin, Russ. Chem. Bull. 54
(2005) 547.
Calculated NMR shifts at the GIAO-B3LYP/6-311G* level were
obtained from these MP2-optimised geometries. Theoretical 11B
chemical shifts at the GIAO-B3LYP/6-311G*//MP2/6-31G* level
were referenced to B2H6 (16.6 ppm [49]) and converted to the
usual BF3ꢁOEt2 scale: d(11B) = 102.83 ꢀ
r(
11B). The 13C and 1H
chemical shifts were referenced to TMS: d(13C) = 179.81 ꢀ
r(
13C);
d(1H) = 32.28 ꢀ
r
(1H). Agreements between observed and calcu-
lated (B3LYP/6-311G*//MP2/6-31G* level) 11B and 13C NMR shifts
generated from optimised geometries are generally very good for
carboranes [50–54]. Agreements between observed and calculated
1H NMR shifts in carboranes are often not as good due to a narrow
ppm range (ca. 12 ppm) and substantial solvent effects on 1H shift
measurements [40].
Supplementary data
CCDC 704893, 704894, 704895 and 704896 contain the supple-
mentary crystallographic data for 1, 2, 3 and 4. These data can be
obtained free of charge via http://www.ccdc.cam.ac.uk/conts/
retrieving.html, or from the Cambridge Crystallographic Data Cen-
tre, 12 Union Road, Cambridge CB2 1EZ, UK; fax: (+44) 1223-336-
033; or e-mail: deposit@ccdc.cam.ac.uk.
[28] D. Hnyk, J. Holub, M. Hofmann, P.v.R. Schleyer, H.E. Robertson, D.W.H. Rankin,
J. Chem. Soc., Dalton Trans. (2000) 4617.
[29] D. Hnyk, D.W.H. Rankin, Dalton Trans. (2009) 585.
[30] M.A. Fox, R. Greatrex, A. Nikrahi, P.T. Brain, M.J. Picton, D.W.H. Rankin, H.E.
Robertson, M. Bühl, L. Li, R.A. Beaudet, Inorg. Chem. 37 (1998) 2166.
[31] M. Buhl, P.v.R. Schleyer, J. Am. Chem. Soc. 114 (1992) 477.
[32] M.A. Fox, A.E. Goeta, A.K. Hughes, A.L. Johnson, J. Chem. Soc., Dalton Trans
(2002) 2132.
Acknowledgements
[33] D.A. Fletcher, B.G. Gowenlock, K.G. Orrell, J. Chem. Soc., Perkin Trans. 2 (1998)
798.
[34] M. Barfield, P. Fagerness, J. Am. Chem. Soc. 119 (1997) 8699.
EPSRC is gratefully acknowledged for funding (M.A.F., R.J.P.,
W.C.). We thank Durham University for access to its High Perfor-
mance Computing facility.
ˇ
[35] S. Hermánek, J. Plešek, V. Gregor, B. Štibr, J. Chem. Soc., Chem. Commun. (1977)
561.
ˇ
[36] S. Hermánek, D. Hnyk, Z. Havlas, J. Chem. Soc., Chem. Commun. (1989) 1859.

2370
M.A. Fox et al. / Polyhedron 28 (2009) 2359–2370
ˇ
[37] M. Bühl, P.v.R. Schleyer, Z. Havlas, D. Hnyk, S. Hermánek, Inorg. Chem. 30
(1991) 3107.
R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H. Nakai, M.
Klene, X. Li, J.E. Knox, H.P. Hratchian, J.B. Cross, C. Adamo, J. Jaramillo, R.
Gomperts, R.E. Stratmann, O. Yazyev, A.J. Austin, R. Cammi, C. Pomelli, J.W.
Ochterski, P.Y. Ayala, K. Morokuma, G.A. Voth, P. Salvador, J.J. Dannenberg, V.G.
Zakrzewski, S. Dapprich, A.D. Daniels, M.C. Strain, O. Farkas, D.K. Malick, A.D.
Rabuck, K. Raghavachari, J.B. Foresman, J.V. Ortiz, Q. Cui, A.G. Baboul, S.
Clifford, J. Cioslowski, B.B. Stefanov, G. Liu, A. Liashenko, P. Piskorz, I.
Komaromi, R.L. Martin, D.J. Fox, T. Keith, M.A. Al-Laham, C.Y. Peng, A.
Nanayakkara, M. Challacombe, P.M.W. Gill, B. Johnson, W. Chen, M.W. Wong,
C. Gonzalez, J.A. Pople, GAUSSIAN, Inc., Wallingford, CT, 2004.
ˇ
[38] S. Hermánek, Chem. Rev. 92 (1992) 325.
ˇ
[39] S. Hermánek, Inorg. Chim. Acta 289 (1999) 20.
[40] E.S. Alekseyeva, A.S. Batsanov, L.A. Boyd, M.A. Fox, T.G. Hibbert, J.A.K. Howard,
J.A.H. MacBride, A. Mackinnon, K. Wade, Dalton Trans. (2003) 475.
[41] M.F. Hawthorne, T.D. Andrews, P.M. Garrett, E.P. Olsen, M. Reintjes, F.N. Tebbe,
L.F. Warren, P.A. Wegner, D.C. Young, Inorg. Synth. 10 (1967) 91.
[42] P.T. Brain, J. Cowie, D.J. Donohoe, D. Hnyk, D.W.H. Rankin, D. Reed, B.D. Reid,
H.E. Robertson, A.J. Welch, M. Hofmann, P.v.R. Schleyer, Inorg. Chem. 35 (1996)
1701.
[43] A. Roe, Organic Reactions, vol. 5, Wiley, New York, 1949.
[44] G.M. Sheldrick, Acta Crystallogr., Sect. A 64 (2008) 112.
[45] M.J.S. Dewar, E.G. Zoebisch, E.F. Healy, J.J.P. Stewart, J. Am. Chem. Soc. 107
(1985) 3902.
[49] T.P. Onak, H.L. Landesman, R.E. Williams, J. Phys. Chem. 63 (1959)
1533.
[50] M.A. Fox, A.K. Hughes, J.M. Malget, J. Chem. Soc., Dalton Trans. (2002) 3505.
[51] A.R. Turner, H.E. Robertson, K.B. Borisenko, D.W.H. Rankin, M.A. Fox, Dalton
Trans. (2005) 1310.
[46] M.J.S. Dewar, C. Jie, E.G. Zoebisch, Organometallics 7 (1988) 513.
[47] CACHE 6.1.1 for Windows, Fujitsu Ltd., Chiba, Japan, 2003.
[48] GAUSSIAN 03, Revision C.02, M.J. Frisch, G.W. Trucks, H.B. Schlegel, G.E. Scuseria,
M.A. Robb, J.R. Cheeseman, J.A. Montgomery Jr., T. Vreven, K.N. Kudin, J.C.
Burant, J.M. Millam, S.S. Iyengar, J. Tomasi, V. Barone, B. Mennucci, M. Cossi, G.
Scalmani, N. Rega, G.A. Petersson, H. Nakatsuji, M. Hada, M. Ehara, K. Toyota,
[52] P.v.R. Schleyer, J. Gauss, M. Bühl, R. Greatrex, M.A. Fox, J. Chem. Soc., Chem.
Commun. (1993) 1766.
[53] I.D. Mackie, H.E. Robertson, D.W.H. Rankin, M.A. Fox, J.M. Malget, Inorg. Chem.
43 (2004) 5387.
[54] A.S. Batsanov, M.A. Fox, A.E. Goeta, J.A.K. Howard, A.K. Hughes, J.M. Malget, J.
Chem. Soc., Dalton Trans. (2002) 2624.