The first rhodaboratrane: [RhCl(PPh3){B(mt)3}] (Rh→B) (mt = methimazolyl)

Article abstract of DOI:10.1039/b413305f

The reaction of [Rh(C₆H₅)Cl₂(PPh ₃)₂] with Na[HB(mt)₃] (mt = methimazolyl) provides [RhCl(PPh₃){B(mt)₃}](Rh→B) the first authentic example of a compound with a rhodium-boron dative bond.

Full text of DOI:10.1039/b413305f

View Article Online / Journal Homepage / Table of Contents for this issue
COMMUNICATION
www.rsc.org/chemcomm | ChemComm
The first rhodaboratrane: [RhCl(PPh3){B(mt)3}](RhAB)
(mt 5 methimazolyl){
Ian R. Crossley,a Mark R. St.-J. Foreman,b Anthony F. Hill,*a Andrew J. P. Whiteb and David J. Williamsb
Received (in Cambridge, UK) 1st September 2004, Accepted 28th September 2004
First published as an Advance Article on the web 25th November 2004
DOI: 10.1039/b413305f
s-Lewis acidity. The question of metal s-basicity is less
straightforward, being a function of metal, oxidation state, charge
and co-ligands. Thus the electron rich metal(0) centres in 1 and 2
would seem ideally suited for entering into such interactions. In
this paper we wish to report the isolation of the first
rhodaboratrane which whilst adopting a geometry and d-occu-
pancy analogous to those of 1 and 2, has the metal in a positive
oxidation state.
The reaction of [Rh(C6H5)Cl2(PPh3)2] with Na[HB(mt)3]
(mt 5 methimazolyl) provides [RhCl(PPh3){B(mt)3}](RhAB)
the first authentic example of a compound with a rhodium–
boron dative bond.
Metal–boron dative bonding has long been hypothesised1 but
more recently called into question.2 It was only with the isolation
of the metallaboratranes [M(CO)(PPh3){B(mt)3}](MAB) (Fig. 1,
mt 5 methimazolyl, M 5 Ru 1,3,4 Os 25) that such interactions
were unambiguously authenticated. In these cage structures a tran-
sannular bond between the formally zerovalent (d8) group 8 metal
and boron(III) is supported by three methimazolyl buttresses.
Our view of the bonding in such compounds supposes that a d8-
ML5 fragment constrained to adopt a monovacant octahedral
geometry would have a pair of electrons housed in an orbital of s
symmetry with respect to the vacant site. Under normal
circumstances a C4v d8-ML5 fragment would either relax to a
trigonal bipyramidal geometry (e.g., Fe(CO)5) or dissociate a
ligand to achieve the electronically favourable d8-ML4 square
planar geometry (Fig. 2). Neither of these options is available to
the geometrically constrained [M(CO)(PPh3){B(mt)3}] complexes
and accordingly this electron pair is accepted by the trivalent
bridgehead boron Lewis acid.
The synthesis of the group 8 metallaboratranes discussed above
proceeded via the reactions of [M(R)Cl(CO)(PPh3)2] (R 5 aryl,
vinyl, hydrido) with Na[HB(mt)3]. The s-organyl (or hydride)
ligand in these substrates serves as a hydrogen acceptor and
accordingly an isoelectronic rhodium(III) s-organyl complex was
required. The complex [Rh(C6H5)Cl2(PPh3)2] 3 has been pre-
viously obtained via a multi-step procedure, the phenyl ligand
arising from fragmentation of an SbPh3 ligand.6 We find that the
same complex may be more conveniently obtained via the reaction
of Wilkinson’s catalyst with phenylmercuric chloride (Scheme 1).
In addition to ease, the advantage of this approach is that it is in
principle extendable to a range of organomercurial halides
although we have not yet explored this avenue.
Treating
a dichloromethane–ethanol solution of 3 with
Na[HB(mt)3] does indeed proceed to the formation of the desired
rhodaboratrane [RhCl(PPh3){B(mt)3}](RhAB) 4 in 65% yield.§
When the reaction was carried out in CDCl3 the formation of
benzene was confirmed spectroscopically (1H NMR).
We are therefore currently concerned with exploring the limits
of such interactions with respect to two issues: d-occupancy and
metal s-basicity. In the case of d-occupancy, clearly for C4v-ML5
with less than a d8 configuration, the s orbital in question is empty
and therefore unable to act in the role of a Lewis base.{ Indeed,
the reactivity of C4v-d6-ML5 complexes is dominated by their
The mechanism of formation is presumably as suggested in
Scheme 2 by analogy with that proposed for 1 and 2 via (i)
bidentate chelation of the HB(mt)3 ligand; (ii) dissociation of a
phosphine; (iii) formation of an agostic B–H–Rh interaction; (iv)
B–H activation" and benzene reductive elimination and finally (v)
coordination of the pendant mt group.
The characterisation of
4 included a crystal structure
determination,I however the formulation also followed from
Fig. 1 Metallaboratranes.
{ Electronic supplementary information (ESI) available: Table 1:
Comparative bond lengths and angles for molecules A and B of
compound 4. Fig. S1: Molecular structure of one (B) of the two
independent molecules in crystals of 4. Fig. S2: Overlay of the two
independent molecules (A and B) in crystals of 4. Fig. S3: Molecular
structure (ORTEP) of one (A) of the two independent molecules in crystals
of 4. Fig. S4: Molecular structure (ORTEP) of one (B) of the two
independent molecules in crystals of 4. See http://www.rsc.org/suppdata/cc/
b4/b413305f/
*a.hill@anu.edu.au
Fig. 2 Various options for an 18-valence electron C4v-d8 ML5 complex.
Chem. Commun., 2005, 221–223 | 221
This journal is ß The Royal Society of Chemistry 2005

View Article Online
geometry at rhodium is distorted octahedral with cis angles in the
range A: 80.14(5)–101.20(6)u [B: 81.01(5)–100.65(5)u]. The Rh–B
˚
bond [A: 2.132(6), B: 2.122(7) A], which is buttressed by three
methimazolyl bridges, is slightly shorter than the corresponding
Ru–B distance in 1, though longer than seen in rhodium boryls7
˚
(typically in the range 1.96–2.05 A); however it should be noted
Scheme 1 Synthesis of 3.
that the boron atoms in these latter species are three-coordinate,
and the majority also involve five-coordinate rhodium, both
factors expected to decrease the Rh–B separations. The geometry
at boron is slightly distorted tetrahedral with angles in the range A:
106.1(4)–115.4(4)u [B: 104.8(5)–114.6(5)u]. As was seen in the
structure of ruthenaboratrane 1, the geometric constraints of
accommodating adjacent, and bridged, tetrahedral (boron) and
octahedral (rhodium) centres causes distortions in the S(1)- and
S(2)-based chelate rings; whereas the fold angle between the
{S,C,N,B} and {S,Rh,B} planes of the S(3) five-membered
chelate ring is less than 1u for both molecules, the corresponding
folds for the S(1) and S(2) chelate rings are ca. A: 26 and 17u
[B: 27, 17u], respectively. The mt heterocycle trans to the phosphine
shows
a marginally longer Rh–S separation [Rh–S(3) A:
˚
2.3867(15), B: 2.3898(15) A] cf. those to S(1) and S(2) [A:
˚
2.3704(16), 2.3692(17), B: 2.3674(16), 2.3640(16) A respectively]. In
Scheme 2 Suggested mechanism for the formation of 4.
the case of the ruthenaboratrane 1 the phosphine coordinates trans
to the RuAB bond whilst in 4 it is the chloride that is trans to
boron. Since steric factors may be excluded, this preference might
be traced to a preference for the weaker p-acid to coordinate trans
to boron.
spectroscopic data. Of particular significance amongst these are (i)
the absence of a n(BH) infrared absorption; (ii) the existence of two
methimazolyl environments (1H NMR) and (iii) the appearance of
a signal in the 31P{1H} NMR spectrum showing coupling to
rhodium (d 28.8 ppm, 1JRhP 5 126 Hz). Notably, this signal shows
a slight broadening as a result of interaction with the quadrupolar
11B nucleus (half height width 5 8 Hz). Similar broadening was
noted in the case of 1 and 2.
The synthesis and structural characterisation of 4 confirms that
octahedral d8-metallaboratrane formation is not limited to the
electron rich zerovalent Os and Ru centres in 1 and 2. Rather,
rhodium(I) appears to also be sufficiently basic to enter into dative
bonding to boron(III). In this respect it is noteworthy that
Rabinovich has very recently obtained a five-coordinate cobalta-
boratrane [Co(PPh3){B(mt)3}]BPh4 as a minor side product from
the decomposition of [Co(PPh3){HB(mt)3}]BPh4.8 Since this is a
16-valence electron complex, the bonding does not fit into the
simple picture provided by Fig. 1.
The X-ray analysis of crystals of 4 showed there to be two
independent molecules (A and B) with essentially identical
conformations, the rms fit of all the non-hydrogen atoms being
˚
ca. 0.14 A (molecule A is shown in Fig. 3, molecule B in Fig S1 of
ESI,{ and an overlay of the two molecules in Fig. S2 of ESI{). The
We thank the EPSRC (UK) and the ARC (Australia) for
financial support and Professor G. Parkin for helpful discussion.
Ian R. Crossley,a Mark R. St.-J. Foreman,b Anthony F. Hill,*a
Andrew J. P. Whiteb and David J. Williamsb
aResearch School of Chemistry, Institute of Advanced Studies,
Australian National University, Canberra, ACT Australia.
E-mail: a.hill@anu.edu.au; Fax: +61 2 6125 3216; Tel: +61 2 6125 8577
bImperial College London, South Kensington, London, UK SW7 2AY
Notes and references
{ Nevertheless there are examples in the early literature claiming dative
interactions between d6-Cr(CO)5 fragments and boranes1 which may
however be discounted on simple electron counting grounds.
§ Data for 4. NMR (CDCl3, 25 uC) 1H (300 MHz): dH 3.46 (s, 6H, NMe),
3.67 (s, 3H, NMe), 6.76, 7.65 (br s 6 2, 1H 6 2, NCHLCH), 6.82, 8.10 (br
s 6 2, 2H 6 2, HCHLCH). 31P{1H} (121.4 MHz): dP 28.8 (d, JRhP 125.9
Hz). 11B{1H} (96.2 MHz): dB 1.7 (hhw 5 50 Hz). Anal. (Calc.) C 48.0
(47.98), H 4.12 (4.03), N 10.99 (11.19%).
" The term oxidative addition is in this instance inappropriate since the
formal oxidation state of rhodium remains + III.
I Crystal data for 4: C30H30BClN6PRhS3?3CH2Cl2, M 5 1005.70, mono-
clinic, P21/n (no. 14), a 5 19.1229(9), b 5 18.9722(6), c 5 25.0598(9) A,
˚
3
˚
b 5 110.156(4)u, V 5 8535.0(6) A , Z 5 8 (two independent molecules),
Dc 5 1.565 g cm23, m(Cu-Ka) 5 9.264 mm21, T 5 293 K, yellow prisms;
12 667 independent measured reflections, F2 refinement, R1 5 0.045,
Fig. 3 Molecular geometry of molecule A of 4.
222 | Chem. Commun., 2005, 221–223
This journal is ß The Royal Society of Chemistry 2005

View Article Online
wR2 5 0.100, 7916 independent observed absorption-corrected reflections
[|Fo| . 4s(|Fo|), 2hmax 5 120u], 709 parameters. CCDC 248523. See http://
www.rsc.org/suppdata/cc/b4/b413305f/ for crystallographic data in .cif or
other electronic format.
4 M. R. St.-J. Foreman, A. F. Hill, G. R. Owen, A. J. P. White and
D. J. Williams, Organometallics, 2003, 22, 4446.
5 M. R. St.-J. Foreman, A. F. Hill, A. J. P. White and D. J. Williams,
Organometallics, 2004, 23, 913.
6 R. Cini and A. Cavaglioni, Inorg. Chem., 1999, 38, 3751.
7 G. J. Irvine, M. J. G. Lesley, T. B. Marder, N. C. Norman, C. R. Rice,
E. G. Robins, W. R. Roper, G. R. Whittell and L. J. Wright, Chem.
Rev., 1998, 98, 2685.
8 D. J. Mihalcik, J. L. White, J. M. Tanski, L. N. Zakharov, G. P. A. Yap,
C. D. Incarvito, A. L. Rheingold and D. Rabinovich, Dalton Trans.,
2004, 1626.
1 K. B. Gilbert, S. K. Boocock, S. G. Shore, in Comprehensive
Organometallic Chemistry, ed. E. W. Abel, F. G. A. Stone and
G. Wilkinson, 1982, Pergamon, Oxford, vol. 6, pp. 880–886.
2 H. Braunschweig, Angew. Chem., Int. Ed., 1998, 37, 1786.
3 A. F. Hill, G. R. Owen, A. J. P. White and D. J. Williams, Angew.
Chem., Int. Ed., 1999, 38, 2759.
This journal is ß The Royal Society of Chemistry 2005
Chem. Commun., 2005, 221–223 | 223