Metallaboratranes: Tris(methimazolyl)borane complexes of rhodium(I)

Article abstract of DOI:10.1021/om050772+

The syntheses and reactivity of the first rhodaboratranes, [RhX(PPh ₃){B(mt)₃}] (X = Cl, H) and [Rh-(η⁴-C ₈H₁₂){B(mt)₃}]Cl, are described in detail together with preliminary investigations of the mechanistic processes involved. The subsequent exploitation and circumvention of the lability of [RhCl(PPh ₃){B(mt)₃}] in the synthesis of a range of isonitrile, [Rh(CNR)(PPh₃){B(mt)₃}]Cl (R = ^tBu, C ₆H₃Me₂-2,6, C₆H₂Me ₃-2,4,6), phosphine, [Rh(PMe₃)_n(PPh ₃)_2-n{B(mt)₃}]Cl (n = 0, 1, 2), and dialkyldithiocarbamate, [Rh(S₂-NEt₂){B(mt) ₃}]Cl, complexes is described, along with the attempted synthesis of [Rh(CN^tBu)₂{B(mt)₃}]Cl from [Rh(η⁴-C₈H₁₂){B(mt)₃}]Cl. Single-crystal X-ray structure determinations of [Rh(L)(L′){B(mt) ₃}]-Cl (L = CN^tBu, CN(C₆H₃Me ₂-2,6), L′ = PPh₃; L = L′ = PMe₃) are reported.

Full text of DOI:10.1021/om050772+

Organometallics 2006, 25, 289-299
289
Metallaboratranes: Tris(methimazolyl)borane Complexes of
Rhodium(I)
Ian R. Crossley, Anthony F. Hill,* and Anthony C. Willis
Research School of Chemistry, Institute of AdVanced Studies, Australian National UniVersity,
Canberra, Australian Capital Territory, Australia
ReceiVed September 6, 2005
The syntheses and reactivity of the first rhodaboratranes, [RhX(PPh3){B(mt)3}] (X ) Cl, H) and [Rh-
(η4-C8H12){B(mt)3}]Cl, are described in detail together with preliminary investigations of the mechanistic
processes involved. The subsequent exploitation and circumvention of the lability of [RhCl(PPh3){B(mt)3}]
in the synthesis of a range of isonitrile, [Rh(CNR)(PPh3){B(mt)3}]Cl (R ) tBu, C6H3Me2-2,6, C6H2Me3-
2,4,6), phosphine, [Rh(PMe3)n(PPh3)2-n{B(mt)3}]Cl (n ) 0, 1, 2), and dialkyldithiocarbamate, [Rh(S2-
NEt2){B(mt)3}]Cl, complexes is described, along with the attempted synthesis of [Rh(CNtBu)2{B(mt)3}]Cl
from [Rh(η4-C8H12){B(mt)3}]Cl. Single-crystal X-ray structure determinations of [Rh(L)(L′){B(mt)3}]-
Cl (L ) CNtBu, CN(C6H3Me2-2,6), L′ ) PPh3; L ) L′ ) PMe3) are reported.
driven in no small part by the apparent utility of this “tamed-
thiolate” ligand as a model for sulfur donors in metalloenzymes.4
In contrast, the pursuit of complexes in which the transition
metal is in a high oxidation state has been largely neglected,
presumably due to a perceived incompatibility of the “soft”
sulfur donor with “hard”, high-valent metals, although this
avenue is now being addressed by studies on the metals of
groups 5 (NbV, TaV)5 and 10 (PtIV).6
In an organometallic context, significant developments per-
taining to this versatile ligand have also begun to emerge. The
larger eight-membered chelate rings adopted in HB(mt)3MLn
complexes (cf. six for HB(pz)3MLn) and the variable hybridiza-
tion at sulfur (cf. triganol pyrazolyl nitrogen donors) impart
additional flexibility to the coordinated ligand, thus availing a
range of novel bonding scenarios and structural motifs. Sig-
nificant among these is the κ3-H,S,S′ binding mode3m,p,7
incorporating a three-center two-electron (3c-2e) B-H-M
linkage, which exhibits an apparent propensity, in some
instances, to undergo B-H activation. Such a process is
Introduction
The report of the hydrotris(methimazolyl)borate ligand by
Reglinski1 heralded a new chapter in the study of poly(azolyl)-
borate compounds, in the fields of both organometallic and
coordination chemistry. The HB(mt)3 ligand has been mooted
as a “soft analogue” of Trofimenko’s ubiquitous hydrotris-
(pyrazolyl)borate HB(pz)3,2 an analogy borne out to some extent
by its now established affinity for the low-valent metals of
groups 6-12.3,4 The exploration of this aspect of poly-
(methimazolyl)borate coordination chemistry continues apace,
* Corresponding author. E-mail: a.hill@anu.edu.au.
(1) (a) Garner, M.; Reglinski, J.; Spicer, M. D.; Kennedy, A. R. Chem.
Commun. 1996, 1975. (b) Reglinski, J.; Carner, M.; Cassidy, I. D.; Slavin,
A.; Spicer, M. D.; Armstrong, D. R. J. Chem. Soc., Dalton Trans. 1999,
2119.
(2) Trofimenko, S. Scorpionates: The Coordination Chemistry of Poly-
(pyrazolyl)borate Ligands; Imperial College Press: London, 1999.
(3) (a) Abernethy, R. J.; Hill, A. F.; Neumann, H.; Willis, A. C. Inorg.
Chim. Acta 2005, 358, 1605. (b) Crossley, I. R.; Hill, A. F.; Humphrey, E.
R.; Smith, M. K.; Tshabang, N.; Willis, A. C. Chem. Commun. 2004, 1878.
(c) Foreman, M. R. St.-J.; Hill, A. F.; White, A. J. P.; Williams, D. J.
Organometallics 2003, 22, 3831. (d) Dodds, C. A.; Lehmann, M.-A.; Ojo,
J. F.; Reglinski, J.; Spicer, M. D. Inorg. Chem. 2004, 43, 4927. (e) Garner,
M.; Lewinski, K.; Pattek-Janczyk, A.; Reglinski, J.; Sieklucka, B.; Spicer,
M. D.; Szaleniec, M. Dalton Trans. 2003, 1181. (f) Garner, M.; Lehmann,
M.-A.; Reglinski, J.; Spicer, M. D. Organometallics 2001, 20, 5233. (g)
Paulo, A.; Correia, J. D. G.; Campello, M. P. C.; Santos, I. Polyhedron
2004, 23, 331. (h) Garcia, R.; Paulo, A.; Domingos, A.; Santos, I. Dalton
Trans. 2003, 2757. (i) Garcia, R.; Xing, Y.-H.; Paulo, A.; Domingos, A.;
Santos, I. J. Chem. Soc., Dalton Trans. 2002, 4236. (j) Garcia, R.; Domingos,
A.; Paulo, A.; Santos, I.; Alberto, R. Inorg. Chem. 2002, 41, 2422. (k)
Garcia, R.; Paulo, A.; Domingos, A.; Santos, I. J. Organomet. Chem. 2001,
632, 41. (l) Ibrahim, M. M.; Shu, M.; Vahrenkamp, H. Eur. J. Inorg. Chem.
2005, 1388. (m) Mihalcik, D. J.; White, J. L.; Tanski, J. M.; Zakharov, L.
N.; Yap, G. P. A.; Incarvito, C. D.; Rheingold, A. L.; Rabinovich, D. Dalton
Trans. 2004, 1626. (n) Graham, L. A.; Fout, A. R.; Kuehne, K. R.; White,
J. L.; Mookherji, B.; Marks, F. M.; Yap, G. P. A.; Zakharov, L. N.;
Rheingold, A. L.; Rabinovich, D. Dalton Trans. 2005, 171. (o) Patel, D.
V.; Mihalcik, D. J.; Kreisel, K. A.; Yap, G. P. A.; Zakharov, L. N.; Kassel,
W. S.; Rheingold, A. L.; Rabinovich, D. Dalton Trans. 2005, 2410. (p)
Alvarez, H. M.; Tanski, J. M.; Rabinovich, D. Polyhedron 2004, 23, 395.
(q) Philson, L. A.; Alyounes, D. M.; Zakharov, L. N.; Rheingold, A. L.;
Rabinovich, D. Polyhedron 2003, 22, 3461. (r) Alvarez, H. M.; Tran, T.
B.; Richter, M. A.; Alyounes, D. M.; Rabinovich, D.; Tanski, J. M.; Krawiec,
M. Inorg. Chem. 2003, 42, 2149. (s) White, J. L.; Tanski, J. M.; Rabinovich,
D. J. Chem. Soc., Dalton Trans. 2002, 2987. (t) Bailey, P. J.; Lorono-
Gonzales, D. J.; McCormack, C.; Parsons, S.; Price, M. Inorg. Chim. Acta
2003, 354, 61.
(4) (a) Melnick, J. G.; Docrat, A.; Parkin, G. Chem. Commun. 2004,
2870. (b) Morlok, M. M.; Docrat, A.; Janak, K. E.; Tanski, J. M.; Parkin,
G. Dalton Trans. 2004, 3448. (c) Docrat, A.; Morlok, M. M.; Bridgewater,
B. M.; Churchill, D. G.; Parkin, G. Polyhedron 2004, 23, 481. (d)
Bridgewater, B. M.; Fillebeen, T.; Friesner, R. A.; Parkin, G. J. Chem.
Soc., Dalton Trans. 2000, 4494. (e) Bridgewater, B. M.; Parkin, G. Inorg.
Chem. Commun. 2001, 4, 126. (f) Kimblin, C.; Bridgewater, B. M.;
Churchill, D. G.; Hascall, T.; Parkin, G. Inorg. Chem. 2000, 39, 4240. (g)
Kimblin, C.; Bridgewater, B. M.; Hascall, T.; Parkin, G. J. Chem. Soc.,
Dalton Trans. 2000, 891. (h) Kimblin, C.; Bridgewater, B. M.; Churchill,
D. G.; Parkin, G. Chem. Commun. 1999, 2301. (i) Bridgewater, B. M.;
Parkin, G. J. Am. Chem. Soc. 2000, 122, 7140. (j) Ji, M.; Benkmil, B.;
Vahrenkamp, H. Inorg. Chem. 2005, 44, 3518. Benkmil, B.; Ji, M.;
Vahrenkamp, H. Inorg. Chem. 2004, 43, 8212. (k) Schneider, A.; Vah-
renkamp, H. Z. Anorg. Allg. Chem. 2004, 630, 1059. (l) Shu, Mi; Walz,
R.; Wu, B.; Seebacher, J.; Vahrenkamp, H. Eur. J. Inorg. Chem. 2003,
2502. (m) Tesmer, M.; Shu, M.; Vahrenkamp, H. Inorg. Chem. 2001, 40,
4022.
(5) Hill, A. F.; Smith, M. K. Chem. Commun. 2005, 1920. (b) Hill, A.
F.; Rae, A. D.; Smith, M. K. Inorg. Chem. 2005, 7316.
(6) Crossley, I. R.; Hill, A. F.; Willis, A. C. Organometallics 2005, 24,
4889.
(7) (a) Hill, A. F.; Owen, G. R.; White, A. J. P.; Williams, D. J. Angew.
Chem., Int. Ed. 1999, 38, 2759. (b) Foreman, M. R. St.-J.; Hill, A. F.; Owen,
G. R.; White, A. J. P.; Williams, D. J. Organometallics 2003, 22, 4446. (c)
Kimblin, C.; Churchill, D. G.; Bridgewater, B. M.; Girard, S. A.; Quarless,
D. A.; Parkin, G. Polyhderon 2001, 20, 1891.
10.1021/om050772+ CCC: $33.50 © 2006 American Chemical Society
Publication on Web 11/30/2005

290 Organometallics, Vol. 25, No. 1, 2006
Crossley et al.
Chart 1. Metallaboratranes
Scheme 1. Formation of the Rhodaboratrane 4 (L ) PPh3)
implicated en route to the group 8 metallaboratrane complexes
[M(CO)(PPh3){κ4-B(mt)3}] (MfB, Chart 1, mt ) methimazolyl,
M ) Ru 1,7a Os 28), which comprise a dative metalfboron
linkage, housed within a cage structure. Significantly, these were
the first unequivocally authenticated examples of transition-
metal 2c-2e borane complexes, the existence of metalfboron
dative bonding, though long postulated,9,10 having fallen into
question with the discovery that early examples had been
incorrectly formulated.11,12
The structures of 1 and 2 are based upon tricyclo[3.3.3.0]
cages, in which the transannular MfB bond is subtended by
three methimazolyl buttresses. It has thus far been our inter-
pretation of the bonding in these complexes that the buttresses
impose a monovacant octahedral geometry at the d8 metal center,
precluding relaxation of the d8-ML5 fragment to more conven-
tional trigonal-bipyramidal or, via ligand dissociation, square
planar d8-ML4 structures. A “lone-pair” of electrons is thus
housed in a metal-based orbital of σ symmetry (sp3d2 type),
which is necessarily oriented toward, and accommodated by,
the Lewis acidic bridgehead boron(III) atom. One can thus
envisage that the propensity for attaining the metallaboratrane
motif is intrinsically linked not only to the d electron counts
since only for a minimum d8 configuration would such an orbital
be occupiedsbut also to the relative basicity of this metal lone-
pair. This basicity will in turn be a function of metal oxidation
state, overall charge, and the ancillary ligand set. Our exploration
of the first two of these influences has led to metallaboratranes
from groups 9 (RhI, IrI)13,14 and 10 (Pt0, PtII)15sincluding both
neutral and cationic examplessthereby illustrating a degree of
generality for the metallaboratrane motif with respect to d8 and
d10 metals, with apparently limited dependence upon oxidation
state and charge. However, the influence exerted by ancillary
ligands is a somewhat more complex issue that we are only
now beginning to explore.
metallaboratrane fragment and ancillary ligands. Aspects of this
work have formed the basis of a preliminary communication.13
Results and Discussion
The group 8 metallaboratranes 1 and 2 (Chart 1) were
originally obtained from the reaction between the hydrotris-
(methimazolyl)borate salt Na[HB(mt)3] and the organometallic
complexes [MRCl(CO)(PPh3)2,3] (M ) Ru, Os; R ) aryl, vinyl,
hydrido), wherein the σ-organyl or hydride ligands ultimately
serve as hydrogen acceptors.7a,8 The synthesis of a rhoda-
boratrane, isoelectronic with 1 and 2, thus required a rhodium-
(III) σ-organyl, to which end the complex [Rh(C6H5)Cl2(PPh3)2]
(3) appeared ideally suited. Since the preparation of 3 has
previously been achieved only via a multistep procedure, via
the curious fragmentation of an SbPh3 ligand to afford the
σ-phenyl function,16 we sought a more convenient route. This
was achieved by treating Wilkinson’s catalyst with 1 equiv of
phenylmercuric chloride in thf under reflux for 2-3 h. After
separation from the deposited elemental mercury and extraction
into dichloromethane, complex 3 was precipitated from the
concentrated solution by the addition of excess ethanol and so
obtained in high yields (60-80%) in analytically pure form.
Although not investigated, this approach offers the advantage
that, in principle, numerous σ-organyls might be obtained from
readily available Ar-HgCl reagents, precluding the need to
prepare the corresponding stibines, the analogous cleavage of
which is by no means guaranteed to proceed.
The reaction between complex 3 and Na[HB(mt)3] proceeds
readily in dichloromethane solution, at ambient temperature, to
afford after 1 h the anticipated rhodaboratrane complex [RhCl-
(PPh3){B(mt)3}] (4, Scheme 1), which precipitates from filtered
and concentrated dichloromethane solutions upon the addition
of diethyl ether. Samples so obtained are, however, frequently
contaminated by free triphenylphosphine, displaced from 3
during the reaction, which proved remarkably difficult to
remove. However, we have found that if the reaction is
performed in a 1:10 mixture of ethanol/dichloromethane,
although oxidation of the free phosphine becomes prevalent
under aerobic conditions, the isolated purity of 4 is significantly
These three factors are inextricably linked and might be
expected to influence both the propensity to attain a metal-
laboratrane motif and subsequently its relative stability and/or
reactivity. A full understanding of each, and the inherent
character of the metallaboratrane cage, thus calls for a systematic
investigation. Herein, we describe the first such study, with the
report of the synthesis of a family of rhodium-based metal-
laboratranes. Further, we discuss the factors involved in their
formation and their inherent reactivity and outline the early
indications in relation to the mutual influences between the
(8) Foreman, M. R. St.-J.; Hill, A. F.; White, A. J. P.; Williams, D. J.
Organometallics 2004, 23, 913.
(9) Shriver, D. F. J. Am. Chem. Soc. 1963, 85, 3509.
(10) (a) Parshall, G. W. J. Am. Chem. Soc. 1964, 86, 361. (b) Powell,
P.; Noth, H. J. Chem. Soc., Chem. Commun. 1966, 637. (c) Grobe, J.; Martin,
R. Z. Anorg. Allg. Chem. 1992, 607, 146.
(11) Braunschweig, H. Z. Nauturforsch. 1999, 54b, 839.
(12) Braunschweig, H. Angew. Chem., Int. Ed. 1998, 37, 1786.
(13) Crossley, I. R.; Foreman, M. R. St.-J.; Hill, A. F.; White, A. J. P.;
Williams, D. J. Chem. Commun. 2005, 221.
(14) Crossley, I. R.; Hill, A. F.; Willis, A. C. Organometallics 2005,
24, 1062.
(15) Crossley, I. R.; Hill, A. F. Organometallics 2004, 23, 5656.
(16) Cini, R.; Cavaglioni, A. Inorg. Chem. 1999, 38, 3751.

Metallaboratranes
Organometallics, Vol. 25, No. 1, 2006 291
Scheme 2. Proposed Formation of the Rhodaboratrane 4
enhanced, with Ph3PO proving more convenient to remove via
fractional crystallization.
from Wilkinson’s Catalyst (L ) PPh3)
As noted in a preliminary communication,13 the identity of 4
has been unequivocally established from analytical, spectro-
scopic, and crystallographic data. Most immediately indicative
of the metallaboratrane structural motif is the presence in the
1H NMR spectrum of resonances associated with two unique
methimazolyl environments, i.e., two NCH3 resonances and two
imidazolyl-derived NCHAdCHBN AB systems, which arise due
to the conformationally locked geometry imposed upon octa-
hedral coordination to the metal center. The retention of one
PPh3 ligand is apparent from integration of the aromatic proton
resonances in the 1H NMR spectrum and the observation in the
31P{1H} NMR spectrum of a single resonance (δP 28.8)
exhibiting doublet multiplicity (103Rh, I ) 1/2, 100%, JRhP 125.9
Hz). That the 31P NMR resonance is relatively sharp (hhw 8
Hz) implies that the phosphine ligand lies cis to the metal-
boron linkage, since a trans disposition routinely effects
significant broadening of the resonance due to an enhanced
interaction with the boron quadrupole. In the present case, this
precludes direct spectroscopic verification of the Rh-B interac-
tion, since the breadth of the 11B{1H} NMR resonance (hhw
50 Hz) prohibits resolution of 103Rh-11B spin-spin coupling,
while the paucity of metallaboratranes renders the chemical shift
(δB 1.7) uninformative in this respect, other than being sug-
gestive of four-coordinate boron. Notwithstanding, the Rh-B
linkage can be inferred from the absence in the infrared spectrum
of any absorption bands characteristic of either terminal (B-
H) or bridging (B-H-Rh) borohydride functionalities, evidence
stored under an inert atmosphere. Surprisingly, subsequent
preparations of this material, while universally successful, have
failed to allow for its isolation with complete exclusion of
intractable contaminants, the removal of which is precluded by
the intolerance of 5 toward conventional purification methods.
However, the identity of 5 is further supported by the observa-
tion of the parent ion (m/z 716.1) and characteristic fragments
in the ESI+ mass spectrum. The solution behavior of 5 in
chlorinated solvents was also convincingly observed during
prolonged NMR acquisitions. Over several hours, the 1H NMR
signals associated with 5 were replaced, inter alia, with
resonances characteristic of the chloro-rhodaboratrane complex
4. The complex nature of this product mixture and the trace
amounts of DCl presumably present in commercial deutero-
chloroform are both consistent with the low yields of 4 that
have been obtained in this way and, while inconclusive, do
support our suggested mechanism. Moreover, we have subse-
quently observed a further example of B-H activation at a
rhodium(I) center that is apparently followed by hydride/chloride
exchange, in our reported, high-yield, synthesis of the rhod-
aboratrane salt [Rh(η4-C8H12){B(mt)3}]Cl (6‚Cl),17 from [RhCl-
(η4-C8H12)]2.
A trait common to each of the rhodaboratranes 4-6+ is their
propensity to decompose in solution, yielding largely intractable
mixtures. This is least pronounced in the case of the salt 6‚Cl,
which survives for up to one week, depending upon the solvent
used, while both 4 and 5 become completely intractable within
a matter of hours. The processes involved in this remain unclear,
although it is apparent that the lability of the ancillary ligandss
rather than of the boratrane fragmentsis of primary significance,
the progressive loss of the phosphine from both 4 and 5 being
readily observed by 31P NMR spectroscopy. Where spectro-
scopic samples are prepared without the rigorous exclusion of
air and moisture, free triphenylphosphine is rarely observed.
Rather, triphenylphosphine oxide is the predominant, often
exclusive, phosphorus-containing species. This would seem to
imply that the [Rh{B(mt)3}]+ cation is an effective catalyst for
the oxidation of PPh3, given the typical reluctance of this
process. Indeed, catalytic oxidation by late transition metals has
been widely observed, and its kinetics thoroughly explored.18
1
for which is similarly absent from the H NMR spectrum.
It is interesting to note that while the synthesis of 4 has only
been reliably achieved via the outlined procedure, we have
encountered small amounts during attempts to prepare the
hydrotris(methimazolyl)borate complex [Rh(PPh3)2{HB(mt)3}]
via the reaction between Wilkinson’s catalyst and Na[HB(mt)3].
While the product mixture was essentially intractable, crystals
of 4, identified both from X-ray diffraction and microanalytical
data, were observed to form in modest quantities from NMR
samples. Although initially a perplexing occurrence, our sub-
sequent demonstrations14,15 that the HB(mt)3 ligand can undergo
B-H activation even in the absence of a hydrogen acceptor
co-ligandsthat is, insertion of the metal into the B-H linkage
to afford a dative M f B interaction and metal hydridesdo
allow us to speculate as to the origin of 4 in this instance. We
suspect that the reaction between Wilkinson’s catalyst and Na-
[HB(mt)3] (Scheme 2) affords initially a complex of the type
[Rh(PPh3)2{κ3-S,S′,H-HB(mt)3}] (i), which is unstable toward
activation of the bridging B-H linkage (ii), thus generating
[RhH(PPh3){B(mt)3}] (5), which is isoelectronic with [PtH-
(PPh3){B(mt)3}]+.15 Hydride-chloride exchange, behavior typi-
cal of hydrido complexes in chlorinated solvents, can then afford
4, though whether this occurs via a classical radical process or
is induced by adventitious HCl is unclear.
In order to verify this postulate, we sought to prepare and
isolate the putative complex 5. This goal was approached by
the reaction between [RhCl(η4-C8H12)(PPh3)] and 1 equiv of
Na[HB(mt)3] in dichloromethane, which is complete after 4 h.
The complex 5 is a moderately air-stable yellow solid, identified
on the basis of spectroscopic data (Table 1, Experimental
Section), which reflect the characteristic signatures of a met-
allaboratrane, in addition to confirming the presence of the
hydride and phosphine ligands. The acquisition of microana-
lytical data for 5 was, however, confounded by its slow, solid-
state, decomposition over the course of several hours, even when
(17) Crossley, I. R.; Hill, A. F.; Humphrey, E. R.; Willis, A. C.
Organometallics 2005, 24, 4083.

292 Organometallics, Vol. 25, No. 1, 2006
Table 1. Phosphorus-31 and Hydrogen-1 NMR Data for Rhodaboratrane Complexesa
δP (JRhP) [hhw] δH
28.8 (125.9)
Crossley et al.
complex
[RhCl(PPh3){B(mt)3}] (4)
3.46 (s, 6H, NCH3), 3.67 (s, 3H, NCH3), 6.76, 7.65 (d × 2,
1H × 2, 3JHH 2, NCHdCH), 6.82, 8.10 (d × 2, 2H × 2, 3JHH
2 Hz, NCHdCH), 7.1-7.5 (m, 15 H, C6H5)
[RhH(PPh3){B(mt)3}] (5)
22.9 [220]
-16.5 (dd, 1H, 1JRhH 16.8, 2JPH 3.5, RhH), 3.36 (s, 6H,
NCH3), 3.46 (s, 3H, NCH3); 6.43, 6.56 (d × 2, 1H × 2, 3JHH
2, NCHdCH), 6.69, 6.92 (d × 2, 2H × 2, 3JHH 2 Hz, NCHdCH),
7.24-7.37 (m, 9H, C6H5), 7.59-7.60 (m, 6H, C6H5)
0.9 (s, 9H, CCH3), 3.44 (s, 6H, NCH3), 3.45 (s, 3H, NCH3),
6.68, 7.61 (d × 2, 1H × 2, 3JHH 2, NCHdCH), 6.95, 8.83
(d × 2, 2H × 2, 3JHH 2 Hz, NCHdCH), 7.35-7.58 (m, 15H, C6H5)
2.27 (s, 6H, CCH3-2,6), 2.88 (s, 6H, NCH3), 3.48 (s, 3H,
NCH3), 6.70, 7.36 (d × 2, 1H × 2, 3JHH 2, NCHdCH),
6.65, 8.34 (d × 2, 2H × 2, 3JHH 2 Hz, NCHdCH), 7.07 (m,
1H, C6H3), 7.09 (m, 2H, C6H3), 7.20-7.40 (m, 15 H, C6H5)
1.66 (s, 6H, CCH3-2,6), 3.44 (s, 6H, NCH3), 3.51 (s, 3H,
NCH3), 6.72, 7.64 (d × 2, 1H × 2, 3JHH 2, NCHdCH),
6.93, 8.74 (d × 2, 2H × 2, 3JHH 2 Hz, NCHdCH), 6.88
(m, 1H, C6H3), 6.90 (m, 2H, C6H3); 7.27-7.31 (m, 9H,
C6H5), 7.52-7.60 (m, 6H, C6H5)
1.60 (s, 6H, CCH3-2,6), 2.21 (s, 3H, CCH3-4), 3.49 (s, 6H,
NCH3), 3.51 (s, 3H, NCH3), 6.68, 8.81 (s × 2, 1H × 2,
NCHdCH), 6.63, 8.41 (s × 2, 2H × 2, NCHdCH), 6.69 (s,
2H, C6H2), 7.23-7.41 (m, 15 H, C6H5)
1.60 (s, 6H, CCH3-2,6), 2.21 (s, 3H, CCH3-4), 3.42 (s, 6H,
NCH3), 3.50 (s, 3H, NCH3); 6.71, 7.63 (2 × d, 1H × 2, 3JHH
2, NCHdCH), 6.92, 8.71 (d × 2, 2H × 2, 3JHH 2 Hz,
NCHdCH), 6.69 (s, 2H, C6H2), 7.27-7.32 (m, 9H, C6H5),
7.51-7.60 (m, 6H, C6H5)
[Rh(PPh3)(CNtBu){B(mt)3}]+ (8+)
[Rh(PPh3)(CNXyl){B(mt)3}]+ (9a+)
9.8 [260]
33.0 [112]
[Rh(PPh3)(CNXyl){B(mt)3}]+ (9b+)
9.1 [270]
[Rh(PPh3)(CNMes){B(mt)3}]+ (10a+)
[Rh(PPh3)(CNMes){B(mt)3}]+ (10b+)
33.3 (113.2)
9.2 [260]
[Rh(PPh3)2{B(mt)3}]+ (11+)
[Rh(S2CNEt2){B(mt)3}] (12)
[Rh(PMe3)(PPh3){B(mt)3}]+ (13+)
30.9 (127.5), 11.1 [61]b
2.73 (s, 6H, NCH3), 3.16 (s, 3H, NCH3), 6.64 (s, 2H,
NCHdCH), 6.76 (s, 1H, NCHdCH), 7.46-7.78 (m, 36H,
NCHdCH and C6H5).
1.18-1.34 (m, 6H, CCH3), 3.48 (s, 6H, NCH3), 3.49 (s, 3H,
NCH3), 6.43, 6.63 (d × 2, 1H × 2, 3JHH 2, NCHdCH),
6.73, 6.91 (d × 2, 2H × 2, 3JHH 2 Hz, NCHdCH).
1.17 (d, 9 H, 2JPH 5.3, PCH3), 2,98 (s, 6H, NCH3), 3.46 (s, 3H, NCH3),
6.65, 7.35 (d × 2, 1H × 2, 3JHH 2, NCHdCH),
30.8 [30] (121)c-39.7 [300]
0.2 (117.5)d -38.6 [300]
6.59, 8.32 (d × 2, 2H × 2, 3JHH 2 Hz, NCHdCH), 7.26-7.37
(m, 15 H, C6H5)
[Rh(PMe3)2{B(mt)3}]+ (14+)
1.16 (d, 9H, 2JPH 9.6, PCH3), 1.41 (d, 9H, 2JPH 5.5,
PCH3), 3,47 (s, 3H, NCH3), 3.49 (s, 6H, NCH3), 6.74, 7.23
(d × 2, 1H × 2, 3JHH 2, NCHdCH), 7.16, 8.41 (d × 2, 2H ×
2, 3JHH 2 Hz, NCHdCH)
c
d
a Measured in CDCl3 at 25 °C, J and half-height-width [hhw] values given in Hz. b As PF6 salt: δP ) -143.1 (sep., JPF 712 Hz). JPP not resolved. JPP
21.5 Hz.
While the product mixtures have not been comprehensively
the metal center. In the presence of the electrophilic [Rh-
{B(mt)3}]+ fragment (or its adducts, vide supra), chelation of
the pendant methimazolyl donor would follow, thus initiating
a bimolecular transfer of the HB(mt)3 ligand. The neutral [Rh-
{κ2-HB(mt)3}{B(mt)3}] complex so generated would then be
susceptible to electrophilic attack by a further equivalent of [Rh-
{B(mt)3}]+, leading to the formation of 7‚Cl.
characterized, it has been found that when chloroform solutions
of 4, 5, or 6‚Cl are allowed to stand, the enigmatic bimetallic
rhodaboratrane salt [Rh2{B(mt)3}2{κ2-S,S′-HB(mt)3}]Cl (7‚Cl,
RhfB, Chart 2) crystallizes from solution as the hexachloroform
solvate. As we have reported recently,17 this material is a trace
component of decomposition mixtures, but can be prepared in
isolable, albeit modest, yields from 6‚Cl and an excess of Na-
{HB(mt)3]. With respect to the decomposition process, the
stability of the “Rh{B(mt)3}” fragment is clear, although one
must question the origin of the terminal HB(mt)3 ligand, since
the presence of free [HB(mt)3]- in admixture with any of 4 to
6 should be evident from their in situ NMR spectra. Since this
is not the case, one must presume the in situ generation of ligated
HB(mt)3 from a boratrane and some hydrogen source. While
we have not pursued this process, one can conjecture (Scheme
3) that when dissolved in deuterochloroform, which was not
rigorously purified, a rhodaboratane such as 4 might be
susceptible to hydrochlorination by trace HCl, thus affording
[RhCl2{κ2-HB(mt)3}(PPh3)], i.e., initial metal protonation fol-
lowed by B-H elimination and coordination of the choride to
The isolation of complex 4 was of some significance to
metallaboratrane chemistry, not least in providing an example
of a metallaboratrane sustained at a metal center in a nonzero
oxidation state.15,3m Moreover, the presence of both a labile
ligand (PPh3) and metal-halide linkage would seem to render
it an ideal substrate for further derivatization. However, the
instability of this compound, both in solution and the solid state,
constitutes a significant obstacle in this respect. In addressing
this problem, we sought to exploit the lability of the PPh3 ligand,
by treating samples of 4, generated in situ, with an excess of a
more potent σ-donor, in an attempt to trap the 16-electron
“[RhCl{B(mt)3}]+” fragment as a less labile derivative. In the
interests of obviating any difficulties associated with the
presence of free phosphine, the initial ligands of choice were
t
the isonitriles CNR (R ) Bu, C6H3Me2-2,6 (Xyl), C6H2Me3-
(18) See for example: (a) Graham, B. W.; Laing, K. R.; O’Connor, C.
J.; Roper, W. R. J. Chem. Soc., Dalton Trans. 1972, 1237. (b) Graham, B.
W.; Laing, K. R.; O’Connor, C. J.; Roper, W. R. J. Chem. Soc., Chem.
Commun. 1970, 1272, and references therein.
2,4,6 (Mes)). Surprisingly, when samples of 4, generated as
described in dichloromethane, are stirred with excess of an
isonitrile, the isolated products are not those arising from

Metallaboratranes
Organometallics, Vol. 25, No. 1, 2006 293
Table 2. Selected Bond Distances (Å) and Angles (deg) for Complexes 8+, 9+, and 14+ (two molecules in the asymmetric unit),
with Estimated Standard Uncertainties (esd’s) in Parentheses
Complex 8+
Rh(1)-S(1)
Rh(1)-S(2)
Rh(1)-S(3)
Rh(1)-C(1)
Rh(1)-B(1)
Rh(1)-P(1)
N(12)-B(1)
N(22)-B(1)
N(32)-B(1)
2.368(2)
2.373(2)
2.396(2)
1.922(7)
2.155(7)
2.523(2)
1.549(9)
1.538(9)
1.565(8)
S(1)-Rh(1)-S(3)
S(2)-Rh(1)-S(3)
S(1)-Rh(1)-S(2)
S(1)-Rh(1)-C(1)
S(2)-Rh(1)-C(1)
Rh(1)-C(1)-N(1)
C(1)-N(1)-C(2)
N(12)-B(1)-N(22)
N(12)-B(1)-N(32)
N(22)-B(1)-N(32)
92.58(7)
92.30(7)
167.72(6)
87.9(2)
85.9(2)
175.5(5)
172.5(7)
114.4(5)
107.5(5)
107.7(5)
Complex 9+
Rh(1)-S(1)
Rh(1)-S(2
Rh(1)-S(3)
Rh(1)-C(1)
Rh(1)-B(1)
Rh(1)-P(1)
N(12)-B(1)
N(22)-B(1)
N(32)-B(1)
2.366(1)
2.370(1)
2.408(1)
1.917(3)
2.146(3)
2.505(1)
1.556(3)
1.561(3)
1.556(3)
S(1)-Rh(1)-S(3)
S(2)-Rh(1)-S(3)
S(1)-Rh(1)-S(2)
S(1)-Rh(1)-C(1)
S(2)-Rh(1)-C(1)
Rh(1)-C(1)-N(1)
C(1)-N(1)-C(2)
N(12)-B(1)-N(22)
N(12)-B(1)-N(32)
N(22)-B(1)-N(32)
92.94(3)
94.08(2)
167.50(2)
85.31(7)
86.65(7)
179.5(2)
174.1(3)
113.5(2)
108.1(2)
108.0(2)
Complex 14+
molecule A
molelcule B
molecule A
molecule B
Rh-S(1)
2.377(2)
2.359(2)
2.418(2)
2.459(3)
2.293(3)
2.153(11)
1.559(12)
1.551(12)
1.576(12)
2.358(2)
2.368(2)
2.432(3)
2.471(3)
2.292(3)
2.148(10)
1.561(12)
1.552(11)
1.540(13)
S(1)-Rh-S(3)
91.20(8)
92.26(9)
167.30(9)
95.96(9)
96.42(9)
90.45(9)
85.72(10)
114.9(8)
105.9(7)
107.7(6)
90.11(9)
92.22(8)
168.18(9)
93.31(9)
98.32(9)
91.37(9)
85.73(9)
114.5(7)
107.2(7)
106.1(7)
Rh-S(2)
Rh-S(3)
Rh-P(1)
Rh-P(2)
Rh-B
S(2)-Rh-S(3)
S(1)-Rh-S(2)
S(1)-Rh-P(1)
S(2)-Rh-P(1)
S(1)-Rh-P(2)
S(2)-Rh-P(2)
N(12)-B-N(22)
N(12)-B-N(32)
N(22)-B-N(32)
N(12)-B
N(22)-B
N(32)-B
Chart 2. Dirhodaboratrane Cation in the Salt
consistent with a trans disposition for phosphorus and boron, a
feature that has been verified crystallographically (Figure 1, vide
infra). That the phosphine disposition of 8 differs from that of
the parent rhodaboratrane 4 would seem to imply a thermody-
namic preference for trans P-RhfB in the former, relative to
the situation in 4. Indeed, it has become a pragmatic observation
that metallaboratranes exhibit a thermodynamic preference for
the dative metalfboron linkage to lie trans to the weaker π-acid.
This is reinforced by the fact that the less π-acidic aromatic
isonitriles CNXyl and CNMes afford, under comparable condi-
tions, mixtures of both isomers; that is, with cis (a) and trans
(b) dispositions of the phosphine and borane ligands. The cis
arrangement has been confirmed as the kinetic product, since
the relative proportions of 9a‚Cl and 10a‚Cl present decrease
as the mixing times are prolonged and are the sole products
when the reactions are quenched after just 1 h. Once isolated,
both 9a‚Cl and 10a‚Cl isomerize in solution to afford, after 3
days (ambient conditions), exclusively the thermodynamic
isomers 9b‚Cl (Figure 2) and 10b‚Cl, a process conveniently
monitored by infrared spectroscopy. This process can be
accelerated by heating isolated 9a‚Cl or 10a‚Cl in dichlo-
romethane under reflux and is thereby complete within 12 h.
Alternatively 9b‚Cl and 10b‚Cl may be prepared directly under
these conditions.
[Rh2{B(mt)3}2{K2-S,S′-HB(mt)3}]Cl (7‚Cl)
displacement of PPh3, but rather the rhodaboratrane salts [Rh-
t
(PPh3)(CNR){B(mt)3}]Cl (R ) Bu 8‚Cl, Xyl 9‚Cl, Mes 10‚
Cl), resulting from displacement of chloride (Scheme 4). In each
case, the product identity is apparent from spectroscopic data
(Table 1), with NMR signals associated with both the coordi-
nated isonitriles (additionally verified by characteristic CtN
stretching modes in the infrared spectra) and PPh3 being
1
apparent in the H NMR spectrum, along with those of the
borane ligand. The bulk compositions were further confirmed
by observation of the parent cations in the positive mode ESI
mass spectrum (m/z 798 8+, 846 9+, 860 10+) and largely
consistent elemental microanalytical data.19 Phosphine retention
is also apparent from the 31P{1H} NMR spectra, which are
devoid of resonances associated with either free PPh3 or its
oxide. In the case of 8‚Cl, the 31P{1H} NMR spectrum also
suggests that the isonitrile lies cis to the RhfB linkage; the
single resonance appears appreciably broadened (hhw 260 Hz),
Given the ease with which isonitriles would seem to displace
the chloride ligand of 4, the inherent lability of the triphen-
ylphosphine ligand, and an emerging propensity for the forma-
tion of cationic rhodaboratranes with even weak donors, we
reasoned that the “in situ” approach should be equally applicable
to a range of neutral and anionic ligands, e.g., PR3/AgX (PR3
) PPh3, dppe; X ) BF4, PF6), CO, -S2CNR2 (R ) Et, Me).
This approach has surprisingly met with only limited success,
(19) We note that while all other data (spectroscopic and microanalytical)
unequivocally confirm the identity and high-purity of compounds 8‚Cl to
10‚Cl, carbon analyses have routinely been found to be significantly in
error. Given the reproducibility of this error and the lack of deviation for
other elemental compositions (H, N, S, Cl), we must conclude this to be a
peculiarity of the rhodaboratrane isonitrile salts for which we are currently
unable to account.

294 Organometallics, Vol. 25, No. 1, 2006
Crossley et al.
Scheme 3. Proposed Generation and Bimolecular Transfer
of a HB(mt)3 Ligand during Formation of 7‚Cl (L ) PPh3)
Scheme 4. Ligand Substitution Reactions of
Rhodaboratranes (L ) PPh3)a
more typically affording complex and often intractable mixtures,
although with the complete consumption of 4, thus implying
either inherent instability of the products or prohibitively slow
reaction rates (i.e., kreact. , kdecomp.). Where more selective
outcomes prevailed, purification proved impracticable and the
reproducibility questionable, though the identity of the major
products could be convincingly inferred on the basis of
spectroscopic (1H and 31P NMR) data. The formation, in
quantity, of both [Rh(PPh3)2{B(mt)3}]PF6 (11‚PF6) and [Rh(S2-
NEt2){B(mt)3}] (12) can thus be concluded (Table 1), despite
the persistence of intractable contaminants, the formulation of
both being further supported by the observation of parent ions
(m/z 978 11+; 601 12) and characteristic fragments in the ESI+
mass spectra.
Greater success followed from the reaction between 4 and 1
equiv of PMe3, which affords quantitatively, after 1 h, the
complex salt [Rh(PMe3)(PPh3){B(mt)3}]Cl (13‚Cl, RhfB).
Once again, the identity of 13‚Cl was established from key
spectroscopic data (Table 1), with the 31P{1H} NMR spectrum
verifying both the retention of the triphenylphosphine ligand
and that its disposition to the borane has been maintained (δP
a
t
(i) Na[HB(mt)3]; (ii) CHCl3/HCl; (iii) CNR (R ) Bu, C6H3Me2-
2,6, C6H2Me3-2,4,6); (iv) PMe3; (v) Na[S2CNEt2].
surprising persistence of OPMe3, which is apparently generated
during synthesis, or workup from the excess of PMe3 required
to effect useful conversion rates. This oxidized species was
initially presumed to arise due to the use of dichloromethane/
ethanol solvent mixtures, but continued to plague the synthetic
protocol even in its absence, despite attempts at the rigorous
exclusion of air/moisture.
Taken together with the isomerization of 9a and 10a (kinetic)
to 9b and 10b (thermodynamic), which presumably occurs via
a dissociative process, the formation of 14‚Cl from 13‚Cl would
seem to emphasize the heightened lability of the weaker σ-donor
PPh3, when disposed cis to the dative RhfB linkage, relative
to a trans arrangement as in 9b‚Cl and 10b‚Cl. It is, however,
interesting to note that on no occasion have we observed the
replacement of PPh3 within 8‚Cl, 9‚Cl, or 10‚Cl by a second
1
30.8, d, JRhP 121 Hz), in addition to the presence of a single
PMe3 ligand lying trans to boron (δP -39.7, hhw 300 Hz). The
solution stability of 13‚Cl is, however, limited, gradual loss of
the PPh3 ligand being apparent, leading ultimately to an
intractable mixture. In the presence of excess PMe3, this process
instead leads to replacement of PPh3 by the more potent σ-donor,
to afford after several days the bis(phosphine) complex [Rh-
(PMe3)2{B(mt)3}]Cl (14‚Cl), which can alternatively be gener-
ated within 12 h in dichloromethane under reflux. Once formed,
14‚Cl exhibits appreciable solution stability, which has allowed
its characterization by single-crystal X-ray diffraction (Figure
3), in support of spectroscopic data. The acquisition of reliable
microanalytical data has, however, been complicated by the

Metallaboratranes
Organometallics, Vol. 25, No. 1, 2006 295
Figure 1. Molecular geometry of the cation [Rh(PPh3)(CNtBu)-
{B(mt)3}]+ (8+) in crystals of nonstoichiometrically solvated [8]‚
Cl. Hydrogen atoms and phosphine substituents are omitted for
clarity (50% displacement ellipsoids). The tertiary butyl group is
rotationally disordered and refined across two sites with isotropic
displacement parameters (one orientation shown).
Figure 3. Molecular geometry of the cationic complex [Rh(PMe3)2-
{B(mt)3}]+ (14+) in crystals of solvated [14]‚Cl. Hydrogen atoms
and phosphine substituents are omitted for clarity (50% displace-
ment ellipsoids). The asymmetric unit comprises two unique
molecules (one shown) with comparable internal geometries.
Chart 3. Canonical Forms for (a) Metallaboratrane and (b)
Base-Stabilized σ-Boryl Complex
is not clearly distinguishable, perhaps implying some degree
of fluxionality. Interpretation of the data is further confounded
by the presence of intractable and inseparable contaminants,
and accordingly we have not further pursued the isolation of
[Rh(CNtBu)2{B(mt)3}]Cl.
Figure 2. Molecular geometry of the cation [Rh(PPh3)(CNXyl)-
{B(mt)3}]+ (9+) in crystals of [9]‚Cl‚CHCl3. Hydrogen atoms and
phosphine substituents omitted for clarity (50% displacement
ellipsoids).
isonitrile ligand, despite the use of excess CNR (R ) tBu, Xyl,
Mes). Thus steric factors may also play a role. In the interests
of completeness, we therefore sought to prepare the complex
[Rh(CNtBu)2{B(mt)3}]Cl, commencing from the salt 6‚Cl, [Rh-
(η4-C8H12){B(mt)3}]Cl (vide supra), in which the lability of the
cyclooctadiene ligand has been inferred. However, when 6‚Cl
is treated with an excess (>2 equiv) of CNtBu, the rhodabo-
ratrane fragment is surprisingly destroyed, affording a largely
intractable mixture with a predominance of the [Rh(CNtBu)4]+
cation presumably present as the chloride salt, which was
identified on the basis of 1H NMR and infrared data.20 A more
controlled reaction ensues when an exact stoichiometric ratio
(2 equiv of CNtBu) is employed, such that the formation of
[Rh(CNtBu)2{B(mt)3}]Cl would seem possible on the basis of
ESI+ mass spectrometric data (m/z 619.1 [M]+, 536.1 [M -
CNtBu]+, 453.0 [M - 2 CNtBu]+), although the 1H NMR data
are ambiguous, apparently indicating the presence of three
magnetically distinct methimazolyl units, which would seem
inconsistent with the simplest formulation. Moreover, the
magnetic and chemical inequivalence of the tert-butyl protons
Solid-State Structures. The molecular cations 8+, 9+, and
14+ each adopt a distorted octahedral geometry about rhodium,
with cis angles at the metal in the ranges 83.40(2)-96.54(6)°
for 8+, 84.83(7)-97.88(2)° for 9+, and 83.00(3)-96.42(9)° for
14+ A [B: 83.2(3)-98.32(9)°]. These data are largely compa-
rable to those reported for the neutral rhodaboratrane complex
[RhCl(PPh3){B(mt)3}] (4) [two crystallographically independent
molecules: A 80.14(5)-101.20(6)°; B 81.01(5)-100.65(5)°],13
but more closely to those of the cationic bimetallic complex
[Rh2{B(mt)3}2{κ2-S,S′-HB(mt)3}]+ (7+) [81.23(4)-97.30(4)°],17
which also has predominantly neutral donors about rhodium.
The Rh-B distances of 8+, 9+, and 14+ are comparable within
2σ, but are marginally longer (2σ-3σ) than those observed for
4, which is consistent with the reduced donor capacity of a
cationic metal center relative to the neutral case. This confirms
that the anomalously short Rh-B separations observed for the
bimetallic 7+ [2.098(6) and 2.091(5) Å]17 are not intrinsic to
its cationic character, which might have implied a significant
contribution from the σ-boryl canonical form of the B(mt)3Rh
cage (Chart 3, vide infra). Rather, it seems more likely that they
are inherent from the yet to be quantified influence of the
additional (π-dative) sulfur donors within the coordination
sphere. We have previously noted that the influence of the ligand
(20) (a) Branson, P. R.; Green, M. J. Chem. Soc., Dalton Trans. 1972,
1303. (b) Dart, J. W.; Lloyd, M. K.; Mason, R.; McCleverty, J. A. J. Chem.
Soc., Dalton Trans. 1973, 2039.

296 Organometallics, Vol. 25, No. 1, 2006
Crossley et al.
trans to the Rh-B linkage is also unquantified, owing to a dearth
of structural data. Although the compounds herein would seem
appropriate to this end, the data remain less than conclusive.
A feature common to each of the structurally characterized
rhodaboratranes is the apparent exertion of an appreciable trans
influence by the RhfB linkage, as determined by comparison
of the trans Rh-L distances against relevant examples deposited
in the Cambridge Crystallographic Data Center. No doubt this
contributes to the trans effect implicit in both the lability of
metallaboratranes and the stereochemistry of the kinetic products
of ligand substitution. This might appear counterintuitive, given
that an inverse trans influence would be anticipated for a
σ-retrodative bond. Nonetheless, the anomaly is confirmed by
the data for 14+ in which the presence of two equivalent PMe3
ligands lying cis [A 2.293(3); B 2.292(3) Å] and trans [A 2.459-
(3); B 2.471(3) Å] to boron provides an internal standard. It
might be argued that this is indicative of the bridgehead boron-
(III) atom behaving not as a σ-Lewis acid but rather as an
internally base-stabilized σ-boryl, for which a strong trans
influence would be expected.21 This would represent a coun-
terpoint to our favored “metallaboratrane” description, which
is however an extreme scenario, and contributions from such a
canonical form cannot be unequivocally discounted. Our as-
sertion that this linkage is predominantly, if not exclusively,
σ-retrodative in character is based largely upon electroneutrality
arguments, which would seemingly preclude oxidation of the
metal center by boron, especially for the formally isoelectronic
group 10 systems.15 Geometric data for the rhodium compounds
do, however, reveal flattening of two of the mt buttresses away
from local C3V-B(mt)3Rh symmetry (Figure 4), reflected also
in the NMR spectroscopic data (vide supra), which could be
argued to support a contribution from the boryl form. However,
while moderate deviations between the three B-N linkages are
in each case apparent [4 1.534(10)-1.567(9); 7+ 1.552(7)-
1.566(7); 8+ 1.538(9)-1.565(8); 9+ 1.556(3)-1.561(3); 14+ A
1.551(12)-1.576(12); B 1.540(13)-1.561(12) Å], these do not
correlate with the similarity noted for the Rh-S distances, when
variations might be anticipated for a base-stabilized boryl
thiolato complex. Moreover, variations in the C-S distances
are insignificant (e2σ) and offer no tangible evidence for
differing bond orders. We thus conclude that, as for group 8
metallaboratranes,7,8 the geometric deviations are inherent from
accommodating the constraints attending a tetrahedral boron
center in proximity to octahedral rhodium and, thus, serve only
to alleviate strain within the chelate rings.
Figure 4. Projections of the simplified structure of 8+ along (a)
B-Rh and (b) S1-S2 vectors illustrating distortion from local C3V-
B(mt)3Rh symmetry.
Figure 5. Space-filling models of (a) [Rh(PPh3)(CNtBu){B(mt)3}]
(8+) and (b) [Rh(PMe3)2{B(mt)3}] (14+) projected along the S2-
Rh-S1 axes. B(mt)3, light gray; isonitrile, white; phosphines, dark
gray.
The conformation adopted by the boratrane cage imposes
significant constraints upon the ancillary geometry about the
metal. Thus, while the complex cation 14+ readily accom-
modates two PMe3 ligands (Figure 5), within 8+ and 9+, where
the ligand trans to boron is the larger PPh3, the isonitrile
occupies an apparent “pocket” in the coordination sphere, which
is seemingly inadequate to easily accommodate a more sterically
demanding ligand. Indeed, this is consistent with the difficulties
encountered in cleanly isolating the salt [Rh(PPh3)2{B(mt)3}]-
PF6 (11‚PF6) and the lability ascribed to this salt and [Rh(PMe3)-
(PPh3){B(mt)3}]Cl (13‚Cl). Moreover, it might be argued that
the thermodynamic preference for trans disposition of boron
and phosphorus in 8+, 9+, and 10+ (vide supra) is, at least in
part, a function of the steric perturbation that would be incurred
by accommodating a PPh3 ligand within the cavity presented
by the S1- and S3-based heterocycles, rather than an intrinsic
predilection for the weaker π-acid to lie trans to boron. Indeed,
this perturbation is clearly apparent from the greater displace-
ment of the boron atom from the [P, Rh, S1, S2] mean plane
within 4 [BfRh-Cl, A 0.483, B 0.512 Å] as compared with
7+-9+ and 14+ [7 0.049, 0.078 Å; 8+ 0.204 Å; 9+ 0.193 Å;
14+ A 0.321, B 0.278 Å]. Nonetheless, the electronic argument
retains credence, since the observation is common to a range
of metallaboratranes, regardless of steric considerations.
Two stereochemical observations noted above call for com-
ment. First, there appears to be a developing trend in metal-
laboratrane chemistry that, under thermodynamic control, the
preferred geometry in octahedral complexes of the form
B(mt)3ML1L2 involves the weaker π-acid assuming the position
trans to the MfB interaction. The second observation is that
there appears to be a significant trans influence exerted by the
MfB group (confirmed for 14+ where L1 ) L2) and that this
(21) Irvine, G. J.; Lesley, M. J. G.; Marder, T. B.; Norman, N. C.; Rice,
C. R.; Robins, E. G.; Roper, W. R.; Whittel, G. R.; Wright, L. J. Chem.
ReV. 1998, 98, 2685.

Metallaboratranes
Organometallics, Vol. 25, No. 1, 2006 297
Chart 4. Interactions of Orbitals with π-Symmetry with
Respect to Metal-Ligand Axesa
Chart 5. Origins of trans Influence in Octahedral
d8-Metallaboratranes
account for the structurally authenticated stereochemistries of
known octahedral metallaboratranes in addition to six ruthenium-
based examples that we have yet to report. The exceptions to
this are the hydrido complexes 5 and [PtH(PPh3){B(mt)3}]+,
wherein the position of the hydride adjacent to boron is a
corollary of the mechanism of formation and the inability of
hydride ligands to exchange sites by heterolytic solvolysis (cf.
chloride).
a πS ) methimazolyl-based π-donor orbital; πLc ) acceptor orbitals
on a π-acid ligand positioned cis to the MfB bond; Lt ) ligand trans
to MfB bond.
presumably contributes to a trans effect implicit in the general
lability of metallaboratranes with respect to dissociative ligand
substitution (confirmed by the stereochemistries of 9a+/10a+
(kinetic) and 9b+/10b+ (thermodynamic)). These issues will be
considered separately, beginning with the preferred location of
π-acid ligands.
The second phenomenon, the operation of a trans influence
(linked to a trans effect), may be traced to the nature of the
bonding in the σ-framework. Conventional (σ) trans influences
are discussed in terms of competition between σ-donors;
however in the case of octahedral metallaboratranes, only one
of the two trans-disposed metal substituents (B and Lt) is a
Lewis base (Lt). Chart 5 depicts the σ orbitals on boron and the
The tris(methimazolyl)borate ligand has been suggested to
act as a particularly effective π-donor, a feature that has been
quantified with recourse to carbonyl-associated infrared data
for the alkylidyne complexes [W(tCC6H4Me-4)(CO)2(L)]
where L represents a wide range of the more commonly
encountered facially tridentate ligands.3c It might reasonably be
assumed that this superlative π-basicity is also a feature of tris-
(methimazolyl)borane coordination. The orbitals available for
interaction with a potential π-acid in metallaboratranes (dxy, dxz,
dyz) will be destabilized to varying extents by the number (1, 2,
or 3) of occupied methimazolyl π-donor orbitals (πS) with
which they interact. Defining the MfB axis as “z” and the
M-Sunique axis as “x” (Chart 4) the π-interactive metal orbitals
increase in energy (i.e., π-retrobasicity) in the order (dxz - πS)
< (dyz - 2 πS) < (dxy - 3 πS). A cylindrically symmetric
bifacial π-acid (CO, CNR, CR, NO) with a pair of orthogonal
acceptor orbitals (πL) will interact with two of these metal
orbitals. Coordination along the x-axis, i.e., Lc, cis to the MfB
bond, will involve interaction with dxz and dxy, affording greater
net stabilization (4πS) than would interaction with the dxz and
t
2
trans ligand L in combination with the dz orbital on the metal
in a manner that is bonding and antibonding with regard to M-B
and M-Lt, respectively. NB: In a homoleptic d8-ML6 complex
this combination has zero net overlap between metal and ligand
orbitals; rather, the conventional eg* orbitals, which are in all
directions M-L antibonding, are generally considered. For the
d8 case where one metal substituent is a σ-Lewis acid, population
of this orbital combination would be expected to strengthen the
M-B bond but weaken the M-Lt bond, accounting for the
observed trans influence in the ground-state structures observed
to date. In considering the observed trans effect, this M-Lt
ground-state destabilization may be one contributory factor;
however transition-state stabilization might also be considered.
The transition state for Lt departure might be viewed as akin to
the electronically favorable d8-square planar geometry, in which
the dz orbital is further stabilized by interaction with the BIII
2
d
yz combination (3πS) presented to a ligand coordinating trans
center.
to boron (Lt). Conversely, coordination of a π-donor (e.g., Cl
in 4), in which case the interaction will be repulsive, would be
least disfavorable were it to involve the orbitals dxz and dyz,
i.e., coordination along the z-axis, with Lt trans to the MfB
bond. Although these arguments are only qualititative, they
Conclusions
We have reported the first systematic study of a family of
metallaboratrane complexes, based on rhodium(I), with a

298 Organometallics, Vol. 25, No. 1, 2006
Crossley et al.
with ethanol (10 mL) and ether (10 mL), and then dried in vacuo.
The complex could be recrystallized from a mixture of dichlo-
romethane and ethanol. Yield: 0.706 g (69%). 31P{1H} NMR: δP
21.58 (d, JRhP 103.2 Hz). The complex was characterized by
comparison of spectroscopic data with those previously published.16
selection of ancillary ligands encompassing a range of basicities,
π-acidities, and steric profiles. This has revealed appreciable
and generic lability for rhodaboratrane complexes that can be
moderated by the judicious use of strong σ-donor ligands and/
or salt formation. The application of a range of isonitriles (CNR,
R ) tBu, Xyl, Mes) to this end has further illustrated an apparent
thermodynamic preference for cis disposition of the MfB
linkage and the most π-acidic ligand, the formation of [Rh-
(PPh3)(CNR){B(mt)3}]Cl (trans-RNC-RhfB) as kinetic prod-
ucts being observed, with their subsequent isomerization to the
trans-P-RhfB isomer under thermodynamic control. This is
similarly reflected in the isolation of [Rh(PMe3)(PPh3)-
{B(mt)3}]Cl (trans-Me3P-RhfB), with no evidence for the cis
isomer, although in each case the relative contributions of
electronic and steric effects remain to be determined. Indeed,
crystallographic studies of [RhCl(PPh3){B(mt)3}], [Rh(PPh3)-
(CNR){B(mt)3}]Cl (R ) tBu, Xyl), [Rh2{B(mt)3}2{κ2-S,S′-HB-
(mt)3}]Cl, and [Rh(PMe3)2{B(mt)3}]Cl leave scope for debate
as to the definite covalent bonding scenario within the metall-
aboratrane motif and the extent to which geometric deformities
are inherent from electronic versus steric influences. Qualitative
arguments have been presented to account for these observations,
in anticipation of a more rigorous and quantitive theoretical
study. Clearly, a full understanding of metallaboratranes, in the
context of their formation, structure, and reactivity, will require
further substantial investigation.
1
Synthesis of [RhCl(PPh3){B(mt)3}] (trans-Cl-RhfB) (4). A
mixture of 3 (0.100 g, 0.129 mmol) and Na[HB(mt)3] (0.049 g,
0.131 mmol) in dichloromethane (20 mL) and ethanol (2 mL) was
stirred under an atmosphere of nitrogen for 1 h. After allowing it
to settle, the mixture was filtered via cannula to a second Schlenk
flask and concentrated under reduced pressure to approximately 2
mL volume. The addition of excess diethyl ether resulted in
precipitation of a pale orange solid. The solvent was removed via
filter-cannula and the solid dried in vacuo. Yield: 0.060 g (62%).
FAB-MS: m/z 715 [M - Cl]+, 453 [M - Cl - PPh3]+. NMR
(CDCl3): 11B δB 1.7 (s br, hhw 50 Hz). Anal. Found: C, 48.0; H,
4.12; N, 10.99. Calcd for C30H30BClN6PRhS3: C, 47.98; H, 4.03;
N, 11.19.
Synthesis of [RhH(PPh3){HB(mt)3}] (trans-P-RhfB) (5). A
mixture of [RhCl(PPh3)(η4-C8H12)] (0.200 g, 0.393 mmol) and Na-
[HB(mt)3] (0.150 g, 0.401 mmol) in dichloromethane (15 mL) was
stirred anaerobically for 4 h. The product was isolated in an
analogous manner to 4, as a yellow solid, dried in vacuo, and stored
under N2. Yield: 0.170 g (60%). ESI+ MS: m/z 716.1 [M]+, 715.1
[M - H]+, 453.1 [M - H - PPh3]+. NMR (CDCl3): 11B{1H} δB
2.1 (s br, hhw 55 Hz). Solution instability precluded acquisition of
satisfactory elemental data.
Synthesis of [Rh(CNtBu)(PPh3){B(mt)3}]Cl (trans-P-RhfB)
(8‚Cl). A solution of compound 4 was prepared, as described, from
3 (0.200 g, 0.258 mmol) and Na[HB(mt)3] (0.099 g, 0.265 mmol)
in a mixture of dichloromethane (20 mL) and ethanol (2 mL) over
1 h. The mixture was filtered anaerobically, and to the filtrate was
added an excess of CNtBu (0.1 mL, 0.9 mmol). After stirring for
12 h the golden solution was concentrated under reduced pressure
to ca. 3 mL volume and the product precipitated by the addition of
excess diethyl ether. The golden yellow solid was isolated by
filtration and dried in vacuo. The complex could be recrystallized
from a mixture of dichloromethane and diethyl ether. Yield: 0.160
g (74%). IR: νmax(CtN) 2186 cm-1 (KBr). ESI+ MS: m/z 797.6
[M]+, 714.8 [M - CNtBu]+, 535.9 [M - PPh3]+. NMR (CDCl3):
11B{1H}: δB ≈ 9.0 (s br, hhw 100 Hz). 13C{1H}: δC 29.6 [d, 4JRhC
14.0, CCH3], 33.6 (2C), 34.1 (1C) [NCH3], 124.1 (2C), 124.5 (1C)
Experimental Section
General Methods. Conventional Schlenk and vacuum line
techniques were routinely employed throughout for the exclusion
of air. Isolated products generally exhibited moderate air-stablity;
thus analytical samples were typically prepared in air. Solvents were
distilled under prepurified nitrogen from appropriate drying agents.
The compounds [RhCl(PPh3)3],22 [RhCl(η4-C8H12)]2,23 CNC6H2Me3-
2,4,6,24 and Na[HB(mt)3]7a were prepared according to published
procedures, while [RhCl(PPh3)(η4-C8H12)] was obtained by treat-
ment of [RhCl(η4-C8H12)]2 with 2 equiv of PPh3. All other materials
were obtained from commercial sources and used as supplied after
spectroscopic and analytical verification of purity. All NMR spectra
were recorded on a Varian Inova 300 instrument (1H, 299.945 MHz,
13C, 75.428 MHz, referenced to external SiMe4; 31P, 121.420 MHz,
ref external 85% H3PO4; 11B, 96.232 MHz, ref external BF3‚Et2O).
Carbon-13 NMR assignments were confirmed with recourse to 2-D
correlation (HMQC and HMBC) spectra. In general, multiplet (ddq)
resonances due to coordinated isonitrile carbon nuclei were not
identified. Elemental microanalytical19 and mass spectrometric data
were provided by the ANU analytical services.
Preparation of [RhCl2(C6H5)(PPh3)2] (3). Under an atmosphere
of prepurified nitrogen, a flask was charged with [RhCl(PPh3)3]
(1.226 g, 1.33 mmol) and PhHgCl (0.430 g, 1.37 mmol) in THF
(40 mL), then the mixture was brought to reflux. After 3 h, during
which time the mixture assumed an orange coloration, it was
allowed to cool to ambient temperature. Dichloromethane (40 mL)
was then added and the mixture filtered through diatomaceous earth
to remove metallic mercury. The filter pad was washed with a
further 40 mL of CH2Cl2, and then the combined filtrates were
concentrated under reduced pressure until an orange solid began
to form. The addition of ethanol (80 mL) completed precipitation
of the orange product, which was collected by filtration, washed
[NCHdCHNB], 119.3 (1C), 121.2 (2C) [NCHdCHNB], 128.5 [d,
2
3JCP 8.6, C3,5(C6H5)], 129.7 [C4(C6H5)], 133.5 [d, JCP 13.2, C2,6
-
(C6H5)], 134.3 [d, 1JCP 23.6 Hz, C1(C6H5)], 161.8 (1C), 163.5 (2C)
[CdS], CtN not identified. Anal. Found:19 C, 46.54; H, 4.76; N,
11.78; Cl, 4.27; S, 11.00. Calcd for C35H39BClN7PRhS3: C, 50.40;
H, 4.71; N, 11.76; Cl, 4.25; S, 11.53. Crystal data: C35H39BN7-
PRhS3‚Cl‚(CHCl3)2.53‚(H2O)0.3. Mw ) 1141.32, monoclinic, C2/c
(No. 15), a ) 42.9100(12) Å, b ) 9.8725(3) Å, c ) 27.4618(8) Å,
â ) 112.2130(10)°, V ) 10770.2(5) Å3, Z ) 8, Dc ) 1.408 Mg
m-3, µ(Mo KR) ) 0.923 mm-1, T ) 200(2) K, yellow needle, 9423
independent reflections, F refinement R1 ) 0.0684, wR2 ) 0.0649
for 5505 independent absorption-corrected reflections [I > 3σ(I);
2θmax ) 50°], 609 parameters, CCDC 281910.
Synthesis of [Rh(CNXyl)(PPh3){B(mt)3}]Cl (trans-C-RhfB)
(9a‚Cl). Prepared as 8‚Cl from 3 (0.100 g, 0.129 mmol), Na[HB-
(mt)3] (0.050 g, 0.13 mmol), and excess CNXyl (0.075 g, 0.57
mmol). Product isolated after 1 h. Yield: 0.072 g (63%). IR: νmax
-
(CtN) 2125 (KBr), 2133 cm-1 (CH2Cl2). ESI+ MS: m/z 846.1
[M]+, 715.1 [M - CNXyl]+, 535.9 [M - PPh3]+, 453.1 [M -
PPh3 - CNXyl]+. NMR (CDCl3): 11B{1H} δB ≈ 8.7 (s br, hhw
196 Hz). 13C{1H}: δC 18.9 [CCH3], 33.0 (2C), 34.2 (1C) [NCH3],
123.9 (2C), 124.2 (1C) [NCHdCHNB], 119.2 (1C), 120.0 (2C)
[NCHdCHNB], 127.6 [C4(C6H3)], 128.0 [C3,5(C6H3)], 127.2 [d,
(22) Osborn, J. A.; Wilkinson, G. Inorg. Synth. 1967, 10, 67.
(23) Komiya, S.; Fukuoka, A. In Synthesis of Organometallic Com-
pounds: A Practical Guide; Koyima, S., Ed.; Wiley: New York, 1996; p
238.
(24) (a) Huffman, C. W. J. Org. Chem. 1958, 23, 727. (b) Obrecht, R.;
Huffmann, R.; Ugi, I. Synthesis 1985, 400.
3JCP 9.7, C3,5(C6H5)], 129.8 [C4(C6H5)], 133.4 [d, 2JCP 13.2 Hz, C2,6
-

Metallaboratranes
Organometallics, Vol. 25, No. 1, 2006 299
(C6H5)], 161.1 (1C), 163.2 (2C) [CdS], quaternary C1,2,6(C6H3),
C1(C6H5), and CtN not identified.
3 (0.100 g, 0.13 mmol) and Na[HB(mt)3] (0.050 g, 0.13 mmol) in
a mixture of dichloromethane (20 mL) and THF (2 mL). After
stirring for 1 h, 1.1 equiv of PMe3 (0.015 mL, 0.15 mmol) was
added directly to the mixture. After a further hour the mixture was
filtered to remove the inorganic halide byproducts, then concentrated
under reduced pressure to ca. 3 mL. The addition of excess diethyl
ether effected the precipitation of 13‚Cl as a pale orange solid,
which was isolated by anaerobic filtration and dried in vacuo.
Yield: 0.080 g (75%). ESI+ MS: m/z 715.1 [M - PMe3]+, 529.1
[M - PPh3]+, 453.0 [M - PPh3 - PMe3]+. NMR (CDCl3): 11B-
{1H} δB ≈ 8.8 (s br, hhw 250 Hz).
Synthesis of [Rh(CNXyl)(PPh3){B(mt)3}]Cl (trans-P-RhfB)
(9b‚Cl). As for 9a‚Cl; upon the addition of the isonitrile (CNXyl),
the mixture was heated under reflux for 12 h. Isolated as described
for 9a‚Cl. Yield: 0.075 g (65%). IR: νmax(CtN) 2155 (KBr), 2158
cm-1 (CH2Cl2). ESI+ MS: m/z 846.1 [M]+, 715.1 [M - CNXyl]+,
535.9 [M - PPh3]+, 453.1 [M - PPh3 - CNXyl]+. NMR
(CDCl3): 11B{1H} δB ≈ 8.9 (s br, hhw 172 Hz). 13C{1H}: δC 17.3
[CCH3], 33.6 (2C), 34.2 (1C) [NCH3], 124.1 (2C), 124.5 (1C)
[NCHdCHNB], 119.4 (1C) 121.4 (2C) [NCHdCHNB], 127.6 [C4
(C6H3)], 128.4 [C3,5(C6H3)], 133.8 [C2,6(C6H3)], 128.5 [d, 3JCP 8.8,
C3,5(C6H5)], 129.7 [C4(C6H5)], 133.4 [d, 2JCP 13.2 Hz, C2,6(C6H5)],
Synthesis of [Rh(PMe3)2{B(mt)3}]Cl (RhfB) (14‚Cl). A
solution of 4, prepared as described for 13‚Cl from 3 (0.200 g,
0.26 mmol), and Na[HB(mt)3] (0.050 g, 0.27 mmol) was filtered
to remove the inorganic halides, then treated with excess PMe3 (0.1
mL, 0.97 mmol), and the mixture was brought to reflux for 15 h.
Concentration of the golden yellow solution to ca. 3 mL, followed
by the addition of excess diethyl ether, led to the precipitation of
14‚Cl as a golden yellow solid, which was isolated by filtration
and dried in vacuo. Yield: 0.140 g (76%). ESI+ MS: m/z 605.0
[M]+, 529.1 [M - PMe3]+, 453.1 [M - 2 PMe3]+. NMR (CDCl3):
11B{1H} δB ≈ 9.5 (s br, hhw 250 Hz). 13C{1H}: δC 17.2, 17.3 [d
× 2, PCH3, 1JCP ≈7 - second order], 33.7 (2C), 34.2 (1C) [NCH3],
124.7 (2C), 124.1 (1C) [NCHdCHNB], 118.6 (1C), 120.2 (2C)
[NCHdCHNB], 164.5 (1C), 164.8 (2C) [CdS]. Occluded OPMe3
(1 equiv by NMR) is a persistent obstacle to obtaining microana-
lytical data from bulk samples, which also retain ill-defined solvent
traces. Anal. Found: C, 34.45; H, 5.17; N, 10.49; S, 11.21; Cl,
6.73. Calcd for C21H42BClN6OP3RhS3: C, 34.42; H, 5.78; N, 11.47;
S, 13.13; Cl, 4.84. Crystal data: C18H33BN6P2RhS3‚Cl‚H2O‚CHCl3‚
(C4H10O)0.25. Mw ) 1474.06, orthorhombic, P212121 (No. 19), a )
9.1802(2) Å, b ) 21.9173(5) Å, c ) 32.4110(8) Å, V ) 6521.3(3)
Å3, Z ) 4, Dc ) 1.502 Mg m-3, µ(Mo KR) ) 1.044 mm-1, T )
200(2) K, yellow needle, 10 400 independent reflections, F refine-
ment R1 ) 0.0405, wR2 ) 0.0402 for 6099 independent absorption-
corrected reflections [I > 2σ(I); 2θmax ) 48.4°], 660 parameters,
CCDC 281909. NB: The structure has a non-centrosymmetric space
group; the absolute structure of the crystal used was determined
by refinement of a Flack parameter (0.02(4), 4538 Friedel pairs).
1
134.2 [d, JCP 5.7, C1(C6H5)], 161.6 (1C), 163.0 (2C) [CdS],
quaternary C1(C6H3) and CtN not identified. Anal. Found:19 C,
47.22; H, 4.70; N, 10.13; Cl, 4.60; S, 10.00. Calcd for C39H39-
BClN7PRhS3: C, 53.10; H, 4.46; N, 11.12; Cl, 4.02; S, 10.90.
Crystal data: C39H39BN7PRhS3‚Cl‚3CHCl3. Mw ) 1240.26, tri-
clinic, P1h (No. 2), a ) 10.8109(2) Å, b ) 14.8035(3) Å, c )
18.0613(4) Å, R ) 86.3691(10)°, â ) 73.2647(12)°, γ ) 74.3865-
(13)°, V ) 2665.63(10) Å3, Z ) 2, Dc ) 1.545 Mg m-3, µ(Mo
KR) ) 1.008 mm-1, T ) 200(2) K, yellow plate, 12 247
independent reflections, F refinement R1 ) 0.0431, wR2 ) 0.0529
for 9728 independent absorption-corrected reflections [I > 3σ(I);
2θmax ) 55.3 °], 596 parameters, CCDC 281911.
Synthesis of [Rh(CNMes)(PPh3){B(mt)3}]Cl (trans-C-RhfB)
(10a‚Cl). Prepared from a solution of 4, generated in situ as
described for 9‚Cl, and excess CNMes (0.080 g, 0.55 mmol).
Product isolated after 1 h. Yield: 0.080 g (69%). IR: νmax(CtN)
2124 cm-1 (KBr). ESI+ MS: m/z 860.1 [M]+, 715.1 [M -
CNMes]+, 598.1 [M - PPh3]+, 453.1 [M - PPh3 - CNMes]+.
NMR (CDCl3): 11B{1H} δB ≈ -2.0 (s br, hhw 384 Hz).
Satisfactory 13C NMR data not obtained due to conversion to 10b‚
Cl during acquisition.
Synthesis of [Rh(CNMes)(PPh3){B(mt)3}]Cl (trans-P-RhfB)
(10b‚Cl). As described for 10a‚Cl, upon addition of the isonitrile
(CNMes), the mixture was heated under reflux for 15 h. Isolated
as for 10a‚Cl. Yield: 0.098 g (85%). IR: νmax(CtN) 2154 (KBr),
2165 cm-1 (CH2Cl2). ESI+ MS: m/z 860.1 [M]+, 715.1 [M -
CNMes]+, 598.1 [M - PPh3]+, 453.1 [M - PPh3 - CNMes]+.
NMR (CDCl3): 11B{1H} δB ≈ 9.1 (s br, hhw 212 Hz). 13C{1H}:
δC 17.2 [CCH3-2,6], 21.1 [CCH3-4], 33.5 (2C), 34.2 (1C) [NCH3],
124.4 (2C), 124.1 (1C) [NCHdCHNB], 119.4 (1C), 121.3 (2C)
[NCHdCHNB], 128.3 [C3,5(C6H2)], 133.9 [C2,6(C6H2)], 138.6 [C4-
Acknowledgment. We thank the Australian Research Coun-
cil (DP034270) for financial support and Ms. L. M. Caldwell
for the preparation of mesitylisocyanide.
3
(C6H3)], 128.4 [d, JCP 8.9, C3,5(C6H5)], 129.7 [C4(C6H5)], 133.4
Supporting Information Available: Full details of crystal-
lographic structure determinations for solvates of 8‚Cl (CCDC
281910), 9‚Cl (CCDC 281911), and 14‚Cl (CCDC 281909) in CIF
format. This material is available free of charge via the Internet at
http://pubs.acs.org.
[d, 2JCP 13.2, C2,6(C6H5)], 134.0 [s, 1JCP 24.0 Hz, C1(C6H5)], 161.7
(1C), 163.1 (2C) [CdS]. Anal. Found:19 C, 46.41; H, 4.62; N, 11.09;
Cl, 4.99; S, 10.54. Calcd for C40H41BClN7PRhS3: C, 53.61; H,
4.61; N, 10.94; Cl, 3.96; S, 10.73.
Synthesis of [Rh(PMe3)(PPh3){B(mt)3}]Cl (trans-Me3P-
RhfB) (13‚Cl). A solution of 4 was prepared as described from
OM050772+