Arrested B-H activation en route to installation of a PBP pincer ligand on ruthenium and osmium

Article abstract of DOI:10.1021/om5001106

The reaction of HB(NCH₂PPh₂)₂C ₆H₄-1,2 with MCl₂(PPh₃)₃ (M = Ru, Os) affords the six-coordinate σ-borane complexes [MCl ₂(PPh₃){σ-BH-κ²

Full text of DOI:10.1021/om5001106

Article
pubs.acs.org/Organometallics
Arrested B−H Activation en Route to Installation of a PBP Pincer
Ligand on Ruthenium and Osmium
Anthony F. Hill* and Caitlin M. A. McQueen
Research School of Chemistry, The Australian National University, Canberra, ACT 0200, Australia
S
* Supporting Information
ABSTRACT: The reaction of HB(NCH2PPh2)2C6H4-1,2
with [MCl2(PPh3)3] (M = Ru, Os) affords the six-coordinate
σ-borane complexes [MCl2(PPh3){σ-BH-κ2-P,P′-HB-
(NCH2PPh2)2-C6H4}], in which the B−H bond remains
intact while coordinated to the metal center. Replacement of
the unique phosphine by π-acceptor ligands, e.g., CO and
CNC6H2Me3, induces B−H activation followed by sponta-
neous dehydrochlorination with the formation of the boryl
pincer complexes [RuCl(CA)2{B(NCH2PPh2)2C6H4}] (A =
O, NC6H2Me3-2,4,6).
ated B−H activation in many cases,3 prior coordination of a
INTRODUCTION
Chelate-assisted activation of B−H bonds has proven to be a
versatile route for the construction of polydentate metal-
■
pendant donor can provide kinetic impetus to assist the
progress of the reaction. Pincer ligands based on conventional
laboratrane1,2 and boryl pincer assemblies (Scheme 1).2 Thus,
classical donors (O, N, P, S, C) have enjoyed intense study in
recent times,4 in particular in the field of catalytic applications.5
while nonchelating boranes themselves undergo metal-medi-
Those involving more electropositive donors, e.g., silicon6 or
boron,2 have been less studied but might offer specific
Scheme 1. Representative Examples of Metallaboratrane and
advantages by virtue of the strong trans influence and trans
Boryl Pincer Complexes Arising from Chelate-Assisted B−H
effect associated with metal boryl, boratrane, and silyl linkages.7
Activation: (i) [IrCl(CO)(PPh3)2];2a (ii)
[IrCl(CO){P(C6H4Me-4)3}2];2b (c) [Pd(NCMe)4](BF4)2,
[Bu4N]Cl2e
t
The boranes HB(NCH2PR2)2C6H4 (R = Ph, Cy, Bu)
described by Yamashita2b,c are effective pro-ligands for the
installation of LXL-PBP8 boron pincers onto iridium,2b
rhodium,2c ruthenium,2i,j cobalt,2k and platinum2f centers. For
the majority of group 9 examples simple insertion of the metal
into the B−H bond occurs, increasing the valency9 of the metal
by two units, and while in the ruthenium and platinum
examples this is implicit, subsequent loss of the putative hydride
in combination with a co-ligand (chloride for platinum, phenyl
for ruthenium) returns the metal to its original valency. The
intimate mechanism of chelate-assisted B−H activation evokes
the possible intermediacy of σ-borane complexes in which
three-center, two-electron (3c-2e) B−H−metal interactions
occur. The geometry of the iridium complex [IrHCl{B-
(NCH2PtBu2)2C6H4}]2c (B−Ir−H = 63(2)°, B···H = 1.89(6)
Å) might at first glance point toward some σ-borane compared
with boryl hydride character; however this type of Y-
bipyramidal geometry is increasingly observed for five-
coordinate d6 complexes devoid of π-acidic ligands.10 Eisenstein
has computationally investigated the geometric preferences for
d6-trans-ML3(PR3)2 complexes,11 concluding that when a single
π-donor is present (Cl in this case), the angle between the
remaining two pseudoequatorial ligands will contract to afford a
Received: January 29, 2014
Published: April 9, 2014
Published 2014 by the American Chemical
Society
1977
dx.doi.org/10.1021/om5001106 | Organometallics 2014, 33, 1977−1985

Organometallics
Article
Y-shaped distorted trigonal bipyramidal geometry. A similar
situation has been examined in considerable detail, both
experimentally and computationally, for the complex [RhHCl-
(BO2C2Me4)(PiPr3)2] (B−Rh−H = 70.0(8)°, B···H = 2.02(3)
Å).12
Scheme 3. Synthesis of Metallaboratranes via B−H
Activation: (i) Na[H2B(mt)2];2a (ii) Na[HB(mt)3]16
Herein we wish to report the isolation of examples of six-
electron chelated σ-borane complexes (LLL-P(BH)P pincers),
which represent intercepted intermediates in the B−H
activation process, including rare structural data for an osmium
σ-borane complex.13,14 These may be converted via π-acid-
induced B−H activation to five-electron LXL-PBP pincer
complexes.
RESULTS AND DISCUSSION
■
We have previously described the facile reaction between
[Ru(C6H5)Cl(CO)(PPh3)2] and Na[HB(mt)3] (mt = methi-
mazol-1-yl) or HB(NCH2PPh2)2C6H4 to afford, respectively,
1a,15
the metallaboratrane [Ru(CO)(PPh3){B(mt)3}](Ru→B)8
or the σ-boryl pincer complex [RuCl(CO)(PPh3){κ3-B,P,P′-
B(NCH2PPh2)2C6H4}].2i A similar strategy employing the
silane HSiPh(NCH2PPh2)2C6H4 affords the σ-sily pincer
c o m p l e x [ R u C l ( C O ) ( P P h 3 ) { κ 3 - P , S i , P ′ - S i P h -
(NCH2PPh2)2C6H4}] (Scheme 2).6v
We have therefore considered the reaction of HB-
(NCH2PPh2)2C6H4 with substrates that might be less
predisposed to B−H activation and/or subsequent loss of the
hydrogen atom once transferred to the metal, so as to intercept
a putative σ-borane complex intermediate. The complexes
[MCl2(PPh3)3] (M = Ru,18 Os19) were found to serve this
purpose: A clean reaction ensues between each of these and
HB(NCH2PPh2)2C6H4 to afford complexes of composition
[MCl2(PPh3){HB(NCH2PPh2)2C6H4}] (M = Ru 1, Os 2,
Scheme 4) based on elemental microanalytical and accurate
mass spectrometric data.
Scheme 2. Chelate-Assisted B−H and Si−H Bond Activation
with [Ru(C6H5)Cl(CO)L2] (i) Na[HB(mt)3];1 (ii)
HB(NCH2PPh2)2C6H4;2i and (iii)
6v
HSiPh(NCH2PPh2)2C6H4
Scheme 4. Synthesis of σ-Borane Pincer Complexes of
Ruthenium and Osmium
In each case, the presumed mechanism involves (i) initial
coordination of a pendant donor, (ii) formation of a σ-B−H−
Ru or σ-Si−H−Ru complex, (iii) B−H or Si−H activation, and
(iv) reductive elimination of benzene. This mechanistic
proposal is supported by the isolation of the complex
[RuH(CO)(PPh3){κ3-H,S,S′-HB(mt)3}] from the reaction of
[RuHCl(CO)(PPh3)3] with Na[HB(mt)3], wherein reductive
elimination of dihydrogen (cf. benzene) is less favored,
although it may be thermally induced (Scheme 3).16 In a
similar manner, the reaction of [RuHCl(PPh3)3] with Na-
[H2B(mt)2] affords a stable complex [RuH(PPh3)2{κ3-H,S,S′-
H2B(mt)2}], in which the 3c-2e B−H−Ru bond remains
intact.17
The gross geometry follows from the 31P{1H} NMR spectra,
which each comprise a doublet and triplet resonance
integrating for two and one phosphorus environments,
2
respectively, with JPP of the order typical of a meridional
PAPBPA arrangement of the A2B spin system nuclei (1: 23; 2: 12
Hz). The data of particular interest, however, relate to the B−H
linkage and whether it remains intact, and it must be said that
the 1H NMR data are equivocal. Had B−H activation occurred
to afford a cis-hydrido-boryl linkage, a high field double triplet
resonance would be expected, with the magnitude of the
1978
dx.doi.org/10.1021/om5001106 | Organometallics 2014, 33, 1977−1985

Organometallics
Article
doublet coupling being indicative of whether the hydride ligand
resided cis or pseudo-trans to the unique phosphine. In practice,
a broad resonance is observed for both (1: δH = −14.04; 2: δH
= −13.72), from which in the case of 2 one might discern dt
fine structure clouded by a broadening suggestive of interaction
with quadrupolar boron nuclei. Notably, the chemical shift of
the resonance for 1 is invariant over the temperature range −70
to 60 °C, arguing against this reflecting an equilibrium (1H
NMR time scale) between borane and boryl hydride isomers.
Inspection of the infrared spectrum of 1 reveals a weak
absorption (KBr: 2259 cm−1) suggestive of a B−H−M
interaction rather than discrete terminal B−H or M−H
absorptions. The complex [RuH(PPh3)2{κ3-H,S,S′-H2B(mt)2}]
provides a benchmark with νBH, νBHRu, and νRuH absorptions at
2379, 2120, and 1950 cm−1 respectively,17 while the free ligand
has νBH = 2583 cm−1 (KBr, see Supporting Information). For
complex 2 the corresponding infrared absorption appears too
weak to be unambiguously identified with confidence.
analysis. Such interactions have been noted for arylphosphines
coordinated to conventional ruthenium hydride ligands.16,20
The only structural data available21 for another osmium σ-
borane adduct relate to the metallacyclic complex [OsH{HB-
(C8H14)CH2PMe2}(PMe3)3] (3, Chart 1) obtained by Baker
Chart 1. Known Osmium Borane Complexes: (a) 3c-2e B−
H−Os;13 (b) 2c-2e Os→B Metallaboratrane;14 (c) 3c-2e B−
H−Os14
The identities of 1 and 2 were unequivocally established
through X-ray crystallographic studies of both as toluene
hemisolvates. While the data for 1 (R = 0.089) afforded a less
precise structural model than for 2 (R = 0.022), both were
isomorphous, and given the comparable covalent radii for the
two metals, it is prudent to discuss the more precise model for
2 as summarized in Figure 1.
and Marder from the reaction of [OsH(CH2PMe2)(PMe3)3]
with 9-borabicyclononane (H-BBN).13 Again, as with 2, the B−
H−Os interaction is part of a chelate assembly with B−H =
1.43(3), Os−H = 1.84(3), and Os−B = 3.005 Å and B−H−Os
= 133(2)°, compared with 2, which has B1−H1 = 1.48(3),
Os1−H1 = 1.56(3), and Os1−B1 = 2.107 Å and B1−H1−Os1
= 87.8(2)°. Thus, it would appear that the most striking
differences between the two complexes are that while 3 involves
a five-membered OsHBCP metallacycle compared with the six-
membered rings of the OsHB(NCP)2 metallabicycle in 2, it is
the former that displays a much larger Os−H−B angle with
little if any direct Os−B interaction. In contrast, the boron in 2
closely approaches the osmium, with the Os1···B1 separation
being shorter than observed for the osmaboratrane [Os(CO)-
(PPh3){B(mt)3}] (2.172 Å, four-coordinate boron)14 and
indeed more comparable to Os−B separations between
octahedral osmium and three-coordinate boron, which to
date21 span the range 2.05−2.26 Å.22 Furthermore, the
geometry at B1 is marginally pyramidalized with a B(N,N′,H)
angle sum of 346.2° and B(N,N′,Os) angle sum of 357.7° in the
opposite sense. Notably, this pyramidalization brings the boron
close to Cl1, and while B1, bearing two positively mesomeric
amino groups, would not be expected to be especially
electrophilic, the B1···Cl1 separation of 2.661 Å is well within
the sum of the van der Waals radii (rvdW(Cl) + rvdW(B) = 3.67
Å).21,23 Three-center, two-electron B−H−metal interactions
are remarkably variable in terms of the B−H−M angle and M···
B separation. While few data are available for (noncluster)
osmium complexes, many have been recorded for B−H−Ru
interactions,21,24 with B−H−Ru angles spanning the wide range
84−155°, loosely correlating (a geometric corollary) with Ru···
B separations in the range 2.12−2.94 Å. Correlation between
Ru···B and B−H distances is not convincingly evident, but this
is to be expected given the low precision of the H-position
modeling. The short Os1···B1 distance coupled with the acute
Os−H1−B1 angle therefore places 2 at the extreme “side-on”
end of the spectrum of 3c-2e B−H−metal interaction such that
little further distortion would be required to sever the B−H
bond with formation of discrete boryl and hydrido ligands (vide
infra). Chart 2 depicts three cases25−27 along this continuum,
with the examples chosen all sharing the common “CpMn-
(CO)2” fragment to emphasize the subtle balance.
Figure 1. (a) Molecular geometry of 2 in a crystal of 2·(C6H5Me)0.5
(70% displacement ellipsoids, aryl H atoms and solvent omitted).
Selected bond lengths (Å) and angles (deg): Os1−Cl1 2.4761(6),
Os1−Cl2 2.4644(5), Os1−P1 2.4018(6), Os1−P2 2.3884(6), Os1−
P3 2.3460(6), Os1−B1 2.107(2), Os1−H1 1.56(3), N1−B1 1.430(3),
N2−B1 1.433(3), B1−H1 1.48(3), P1−Os1−P2 155.19(2), Cl1−
Os1−Cl2 82.46(2), P1−Os1−B1 77.77(7), P2−Os1−B1 77.63(7),
Os1−H1−B1 87.8(2). (b) Inset shows view along the P1···P2 vector
illustrating the impact of secondary interligand interactions upon σ-
borane bonding.
The functionality of interest is the 3c-2e B−H−Os
interaction, and while this is clearly established, the poor
precision associated with modeling hydrogen atom positions in
the vicinity of heavy metals (2: μ = 2.93 mm−1), coupled with
the constraints imposed by chelation, call for caution in
overinterpreting geometrical parameters. Furthermore, incipi-
ent dihydrogen bonding between the borane hydrogen (δ−)
and an ortho-hydrogen atom (δ+) of one phenyl group (H1···
H721 2.38 Å, 2rvdW(H) = 2.40 Å) further discombobulates
1979
dx.doi.org/10.1021/om5001106 | Organometallics 2014, 33, 1977−1985

Organometallics
Article
(yellow and pale blue) between phenyl groups of the unique
phosphine and those of the pincer ligand contribute to a
favorable manifold of intramolecular interactions accounting for
the stability of the complex, which might be lost were B−H
activation to ensue.
Chart 2. Bonding Continuum for B−H−M Interactions with
Selected Examples: (a) 3c-2e Pseudoagostic;25 (b) 3c-2e σ-
Borane;26 (c) 3c-4e Boryl-Hydrido (i.e., “Oxidative”
Addition)27
1
Notably, H NMR data for the complex 3 comprise a single
broad resonance at δH = −10.78 accounting for both the
terminal Os−H and B−H groups.13 This exchange, which is
arrested at −80 °C, was interpreted equivocally as involving
either dissociation of the borane, leaving behind two terminal
Os−H groups, or, alternatively, equilibration via a symmetrical
Os(μ-H)2B arrangement. Given that this work predated the
large number of authenticated examples of metal−boron dative
bonding15,29 for late transition metals, perhaps one might now
also consider a third possibility involving B−H activation to
afford the metallaboratranes [Os(σ-H2)(PMe3)3{κ2-B,P-(BBN)-
CH2PMe2)](Os→B)8 or [OsH2(PMe3)3{κ2-B,P-(BBN)-
CH2PMe2)](Os→B)6 depending on whether equilibration
might be via a dihydrogen or cis-dihydride complex. Such an
interpretation would be consistent with the lower activation
barrier observed for the osmium complex compared with the
ruthenium analogue, given that B−H and C−H activation
processes are especially facile for 5d versus 4d metals.
*
Resonance for exchanging B−H−Mn and BH2 sites.
The formation of 1 is complete after 22 h at room
temperature. However, 31P{1H} and 1H NMR spectra
measured during the course of the reaction indicate the
formation and disappearance of an intermediate formulated as
shown in Scheme 4. Although this species could not be isolated
due to its evolution into 1, the spectroscopic data are consistent
with the formation of a positional isomer. Specifically, the high-
A further useful comparison between the compounds 2 and 3
relates to the trans influence of the borane ligand. It is to be
expected that a B−H “ligand” coordinating in a 3c-2e manner
would have a lesser trans influence than a conventional 2c-2e
metal−ligand bond. In the case of 3 (and its ruthenium
analogue) it was noted that the PMe3 ligand trans to the borane
had an Os−P bond length (2.2657(8) Å) somewhat contracted
relative to the two cis PMe3 ligands (2.3375(8), 2.3298(8) Å).
Trimethylphosphine is essentially a pure σ-donor ligand, while
in the case of 2 chloride is also a potential π-donor capable of
supplementing reduced electron density at the metal center,
resulting from a weak-field trans ligand.
This is indeed what is observed for 2 wherein Os1−Cl2
(2.4644(5) Å) is significantly (23 esd) contracted relative to
Os1−Cl1 (2.4761(6) Å) despite being subjected to more steric
congestion and a set of three ortho-aryl C−H···Cl inter-
mediate28 intraligand hydrogen-bonding interactions (average
CH···Cl2: 2.62 Å, rvdW(Cl) + rvdW(H) = 2.95 Å), which would
be expected to elongate the Os−Cl2 bond. The steric features
of 1 are presented in a space-filling depiction in Figure 2, from
which it is also apparent that two π-stacking interactions
1
field H NMR spectrum includes a signal at δH = −7.21, the
broadness of which suggests retention of the B−H connectivity.
This peak was sufficiently thermally decoupled from the
quadrupolar boron nucleus at −45 °C to observe phosphorus−
2
proton coupling, giving a broad doublet with a JPH of
approximately 45 Hz, although coupling to the second 31P
environment was not resolved. The large doublet coupling
suggests a trans arrangement of the triphenylphosphine and σ-
borane groups. The 31P NMR spectrum further supports this
stereochemistry, showing characteristic resonances for an AB2
spin system (2JAB = 25 Hz) with a slightly broadened triplet
resonance. Thus, while the “B(NCH2PPh2)2C6H4” pincer is
well disposed toward meridional coordination by virtue of the
trigonal boron donor, the corresponding borane presumably
offers more flexibility in the geometry of the B−H−M linkage,
possibly allowing the albeit strained occupation of a
pseudofacial set of coordination sites.
The formation of 1 and 2 involves liberation of two
equivalents of PPh3, and the 31P{1H} NMR spectra of both
crude reaction mixtures include a peak due to free PPh3. Given
the sharpness of this resonance, we may conclude that
phosphine dissociation from 1 and 2 is not apparent on this
NMR time scale. Nevertheless, on the chemical time scale it is
to be expected that steric pressures may well labilize the
phosphine, and this is suggested in the case of 1 by the slow (18
h, 60 °C) reaction with carbon monoxide and the more rapid
reaction with CNC6H2Me3-2,4,6 (66 h, 25 °C). The reaction of
1 with CO results in the formation of a compound that is
devoid of any resonance in the 1H NMR attributable to either a
B−H or Ru−H group. Infrared and NMR spectroscopic data
are consistent with the formation of the boryl pincer complex
[RuCl(CO)2{B(NCH2PPh2)2C6H4}] (4), which has been
previously observed to arise for the reversible reaction of
Figure 2. Space-filling representation for 1 depicting steric protection
afforded to the N2BHRu linkage (N = dark blue, H = gray, B = pink, π-
stacking interactions indicated in yellow and pale blue).
1980
dx.doi.org/10.1021/om5001106 | Organometallics 2014, 33, 1977−1985

Organometallics
Article
[RuCl(CO)(PPh3){B(NCH2PPh2)2C6H4}] with carbon mon-
oxide.2i We attribute the lack of reversibility in the present case
to sequestering of liberated PPh3 by eliminated HCl. A similar
reaction with CNC6H2Me3-2,4,6 is complete within 3 days at
room temperature to afford the cis-bis(isonitrile) complex
[RuCl(CNC6H2Me3-2,4,6)2{B(NCH2PPh2)2C6H4}] (5), the
1H NMR spectrum of which is again devoid of high-field
resonances attributable to B−H or Ru−H groups. The
appearance of two νCN absorbances in the IR spectrum
(CH2Cl2: 2104, 2061 cm−1) indicates a mutual cis disposition of
the isonitriles. This inference is supported by two resonances
(δC = 168.4, 171.4) in the 13C{1H} NMR spectrum, which also
includes a virtual triple resonance at δC = 51.4 (JPC = 23 Hz)
consistent with the trans arrangement of the PCH2 groups.
The 11B{1H} NMR spectrum of 5 comprises a single
resonance at δB = 57.3 shifted some 8 ppm downfield from that
of 1 (δB = 49.2) and in a region typical of three-coordinate
boron in a σ-boryl ligand. It would therefore appear that δB is
not an entirely definitive parameter for distinguishing σ-borane
and σ-boryl coordination modes. The characterization of 5
therefore included a crystallographic study, the results of which
are summarized in Figure 3 and confirm the formation of the σ-
by trans π-acid ligands had been previously demonstrated for
dioxaboryl ligands, e.g., [OsI(BO2C6H4)(CO)2(PPh3)2] (cct
isomer 2.145(5); ttt isomer 2.090(3) Å)22g and for [OsCl-
(BO2C2H4)(CO)n(PPh3)2] (n = 1, 2.043(4); n = 2, 2.179(7)
Å).22a Reflecting perhaps the constraints of chelation, the
ruthenium is slightly displaced from the N1−B1−N2−C4−C3
plane, and while the angle sum at B1 is close to ideal (358°),
those for N1 and N2 each reveal a slight pyramidalization
(352°).
The geometry of 5 affords an opportunity to assess the
relative trans influences of chloride versus a σ-boryl upon two
otherwise identical π-acidic CNC6H2Me3 ligands, and the
potency of the σ-boryl ligand in this respect is patently clear
with the Ru1−C60 bond length (2.048(8) Å) being some 33
esd longer than that for Ru1−C50 (1.876(8) Å). A similar
differential trans influence on the part of more conventional σ-
boryl ligands has been demonstrated in the complexes cis,trans-
[Os(BO2C6H4)(CO)(NCMe)2(PPh3)2]+ for σ-donor nitri-
les,22c [Os(BO2C6H4)(S2CNEt2)(CO)(PPh3)2] for the sulfur
σ+π donors of a dithiocarbamate chelate,22c and [Os-
(BO2C6H4)X(CO)2(PPh3)2] (X = C6H4Me-2,22e I22g) for σ-
donor/π-acid CO ligands. Against these disparate co-ligand
sets, the recurring feature is the demonstration in each case of a
potent trans influence for σ-boryl ligands. Lin and Marder have
computationally investigated the origin of the trans effect of
boryl ligands using the Pt−Cl bond length of trans-[PtXCl-
(PMe3)2] as a reporter, leading to the conclusion that this far
exceeds that of σ-organyls and hydride ligands and that
diazaboryls are among the strongest trans influential σ-boryls
(exceeded by BH2 and BMe2).7
The facile conversion of the σ-borane complex 1 into the σ-
boryl complexes 4 and 5 invites mechanistic conjecture, given
that no intermediates were spectroscopically observed.7
Spectroscopic data for 1 neither suggest nor exclude the
operation of a B−H activation tautomerism to afford seven-
coordinated 6 (Scheme 5), from which phosphine might more
easily be liberated, given the potent (pseudo)trans effect of the
resulting σ-boryl ligand. As noted above, this does not appear to
operate on the 31P NMR time scale. In general, π-acceptor
ligands are expected to make C−H or B−H activation
processes less favorable, as these ligands are less able to
stabilize an increase in metal valency or oxidation state.
Coordination of these ligands would, however, be expected to
increase the acidity of any resulting metal hydride by stabilizing
the conjugate base through retrodonation. Late transition metal
hydride ligands are typically protic in nature, while B−H bonds
show hydridic character; that is, the proton is most likely lost
via the metal, rather than directly from boron. We are therefore
inclined to suspect that upon substitution of the unique
phosphine by CO, the resulting complex [RuHCl2(CO){B-
(NCH2PPh2)2C6H4}] (7a) is prone to spontaneous dehydro-
chlorination, possibly assisted by the weak base PPh3. The
resulting coordinatively unsaturated complex [RuCl(CO){B-
(NCH2PPh2)2C6H4}] (8) would then rapidly capture extra-
neous CO to afford 4, a reaction implicit in the previously
reported conversion of [RuCl(CO)(PPh3 ){B-
(NCH2PPh2)2C6H4}] to 4 upon exposure to CO.2i A similar
sequence would be expected to operate in the conversion of 1
to 5, although in that case, while isonitriles are typically more
nucleophilic than CO, their reduced π-acidity would make
dehydrochlorination of [RuHCl2(CNC6H2Me3-2,4,6){B-
(NCH2PPh2)2C6H4}] (7b) less facile. We consider B−H/M−
Figure 3. Molecular structure of 5 in a crystal (70% displacement
ellipsoids, aryl H atoms omitted, aryl groups simplified). Selected bond
lengths (Å) and angles (deg): Ru1−Cl1 2.4924(18), Ru1−P1
2.3307(18), Ru1−P2 2.3298(18), Ru1−B1 2.062(8), N1−B1
1.456(8), N2−B1 1.441(9), C60−Ru1 2.048(8), C50−Ru1
1.876(8), C60−Ru1−C50 93.5(3), C60−Ru1−P1 104.9(2), C60−
Ru1−P2 101.6(2), P1−Ru1−P2 153.51(7), P1−Ru1−B1 76.2(2),
P2−Ru1−B1 77.3(2).
boryl pincer framework including a conventional 2c-2e Ru1−
B1 bond length of 2.062(8) Å. The only structural data for a
diazaboryl complex of ruthenium are from the related pincer
complex [RuCl(CO)(PPh3){B(NCH2PPh2)2C6H4}]2i and the
recently reported hydrido complex [RuH(CO)2{B-
(NCH2PtBu2)2C6H4}].2j The crystal structure of the former
comprises two independent molecules with Ru−B bond lengths
of 2.051(15) and 2.088(15) Å,2i while the latter has an
intermediate Ru−B bond length of 2.072(8) Å.2j More
structural data are available for diazaboryl complexes of
osmium,22,30 which span the range 2.055−2.082 Å for Os−B
separations with the exception of the anomalous complex
[Os{B(NHC6H4Me-4)2}(CCPh)(CO)2(PiPr3)2],30a for
which an unusually long Os−B bond (2.256(4) Å) was
observed and suggested to arise from the trans coordination of
a π-acid (CO). The phenomenon of Os−B bond lengthening
1981
dx.doi.org/10.1021/om5001106 | Organometallics 2014, 33, 1977−1985

Organometallics
Article
Scheme 5. Mechanistic Conjecture Concerning π-Acid
(CA)-Activated σ-Borane/σ-Boryl Interconversion
Scheme 6. Hydrogen Umpolung: Migration of Hydrogen
from Metals to Boron across (a) M→BR3;31 (b) M−BR2;2k
and (c) MBR33 Bonds
disordered Co−H site approaches coplanarity with the
benzodiazaborolyl unit compared with 1 and 2, wherein the
metal is significantly displaced from the plane.
We have previously shown that a hydride ligand may migrate
from a late transition metal to the boron of a metal-
laboratrane,31,32 while Sabo-Etienne has reported the reversible
addition of dihydrogen across the RuB multiple bond of a
ruthenium borylene complex (Scheme 6).33 In the context of
metal-mediated dehydrocoupling catalysis, the ability of a
hydrogen atom to change its protic/hydridic nature (“umpo-
lung”) by migrating between a metal (BM−Hδ+) and boron
(M−BHδ−) provides an intriguing and potentially useful
dimension to the chemistry of metal−boron linkages.
Cl σ-metathesis pathways unlikely due to the anti-periplanar
H−B−M−Cl geometry of 1 and 2.
CONCLUSIONS
The proposed intermediacy of σ-borane complexes in the
installation of Yamashita’s boryl pincer ligands “B-
■
t
(NCH2PR2)2C6H4” (R = Ph, Cy, Bu) has been substantiated
by the isolation of the σ-borane complexes [MCl2(PPh3){HB-
(NCH2PPh2)2C6H4}] (M = Ru 1, Os 2). Furthermore, the
subsequent conversion to boryl pincer ligands via chelate-
assisted B−H activation could be shown to be induced by
varying the nature of the co-ligands, specifically the
introduction of π-acceptor ligands. It seems most likely that
these processes are induced not by favoring the B−H cleavage
reaction itself, but rather by increasing the Brønstead acidity of
the resulting metal hydride, favoring subsequent and
spontaneous, albeit slow, reductive dehydrohalogenation of
the metal center. A recent report by Peters2k documents a two-
step process (Scheme 6b) that is effectively the reverse of the
borane−boryl interconversion, i.e., addition of dihydrogen
across the 2c-2e B−Co bond of a cobalt complex of Yamashita’s
pincer.
EXPERIMENTAL SECTION
■
General Considerations. All manipulations of air-sensitive
compounds were carried out under a dry and oxygen-free nitrogen
atmosphere using standard Schlenk and vacuum line techniques, with
dry and degassed solvents. NMR spectra were recorded at 25 °C on a
Varian Mercury 300 (1H at 300.1 MHz, 31P at 121.5 MHz) and Varian
Inova 300 (1H at 299.9 MHz, 13C at 75.42 MHz, 31P at 121.4 MHz,
11B at 96.23 MHz) spectrometers. The chemical shifts (δ) for H and
1
13C spectra are given in ppm relative to residual signals of the solvent,
11B and 31P relative to external references (BF3OEt2, H3PO4). Virtual
triplet resonances are indicated by tv. Low- and high-resolution mass
spectra were obtained on a ZAB-SEQ4F spectrometer by positive ion
ESI techniques using an acetonitrile matrix by the mass spectrometry
service of the Australian National University. Assignments were made
relative to M, where M is the molecular cation. Assignments were
verified by simulation of isotopic composition for both low- and high-
resolution levels. Elemental microanalysis was performed by the
microanalytical service of the Australian National University. Data for
X-ray crystallography were collected with a Nonius Kappa CCD
Thus, [Co(N2){B(NCH2PtBu2)2C6H4}] was shown to react
with an amine borane Me2HNBH3 to afford a borane pincer
complex [Co(BH4){σ-BH-κ2-P,P′-HB(NCH2PtBu2)2C6H4}].
The borane pincer is clearly related to that found in 1 and 2
with the exception that one cobalt of the each positionally
1982
dx.doi.org/10.1021/om5001106 | Organometallics 2014, 33, 1977−1985

Organometallics
Article
diffractometer. The compounds [RuCl2(PPh3)2],18 [OsCl2(PPh3)3],19
and HB(NCH2PPh2)2C6H4 were prepared according to published
procedures. The complex [Ru{B(NCH2PPh2)2C6H4}Cl(CO)2] (4)
has been reported previously,2i having been obtained via an alternative
route to that described below. Other reagents were used as received
from commercial suppliers.
H, 4.46; N, 2.58 (the presence of 0.5 equiv of toluene was
c r y s t al l og r a ph i c a l l y co nfi r m e d ) . Cr ys t a l da t a for
C50H44BCl2N2OsP3.0.5(C7H8): Mr = 1083.82, monoclinic, P21/c, a =
12.4233(1) Å, b = 20.3953(2) Å, c = 18.8159(2) Å, β = 92.0176(6)°, V
= 4764.56(8) Å3, Z = 4, Dx = 1.511 Mg m−3, μ(Mo Kα) = 2.93 mm−1,
T = 200(2) K, brown block, 0.31 × 0.18 × 0.17 mm, 10 913
independent reflections. F2 refinement, R = 0.022, wR = 0.051 for 9626
reflections (I >2σ(I), 2θmax =55°), 598 parameters, CCDC 982298.
Synthesis of [Ru{B(NCH2PPh2)2C6H4}Cl(CO)2] (4). In a J. Young
NMR tube, a sample of [Ru{κ3-HB(NCH2PPh2)2C6H4}Cl2(PPh3)]
(1) in C6D6 under an atmosphere of argon was frozen using dry ice/
ethanol. The atmosphere in the tube was then evacuated and replaced
with carbon monoxide (1 atm, three times), and the solvent was
allowed warm to ambient temperature. The mixture was then was
heated to 60 °C for 18 h and then analyzed by NMR spectroscopy.
The solvent was then removed and the residue dissolved in
dichloromethane to obtain IR data. Both NMR and infrared data for
the resulting product were consistent with the formation of the
previously reported complex [Ru{B(NCH2PPh2)2C6H4}Cl(CO)2]
(4).2 IR: CH2Cl2: 2035, 1972 cm−1; KBr: 2027, 1967 cm−1. NMR
2b
Synthesis of [Ru{κ3-HB(NCH2PPh2)2C6H4}Cl2(PPh3)] (1). A
solution of [RuCl2(PPh3)3]18 (1.000 g, 1.04 mmol) and HB-
(NCH2PPh2)2C6H4 (0.536 g, 1.04 mmol) in benzene (50 mL) was
stirred for 22 h. The solvent was concentrated under reduced pressure
to approximately 10 mL, and 15 mL of n-hexane was then added to
afford an orange precipitate. The supernatant was removed via cannula
filtration, and the precipitate washed with n-hexane (2 × 15 mL) and
dried in vacuo. Yield: 0.887 g (90%). Orange crystals of the toluene
hemisolvate 1·(C7H8)0.5 suitable for elemental and crystallograhic
analysis were obtained by slow diffusion of n-hexane into a toluene
solution of 1. IR (KBr, cm−1): 3047 νaromCH; 2921, 2846 νCH; 2259
1
νBHRu, 1478, 1435 νaromCC. NMR (C6D6, 298 K) H: δH −14.04 (br,
1H, RuHB, hhw = 0.2 ppm), 4.31 (m, 4H, PCH2), 6.45−8.72 (m, 39
H, N2C6H4 and C6H5). 1H{11B}: δH −14.30 (dt, 2JPH = 19.2 d, 13.2 t).
1
2,4
2
(C6D6, 25 °C) H: δH 4.27 (d tv, 2 H, PP′CHB,
J
= 2.7, JAB =
1,3
13C{1H}: 54.9 (tv, PCH2,
J
= 21), 111.2, 120.3 [C2−5(N2C6H4)],
PB
PC
2,4
2
12.3), 4.61 (d tv, 2 H, PP′CHA,
J
= 3.8, JAB = 12.3), 6.88−7.19,
129.3, 131.2 (C6H5), 131.6 (tv, C6H5, JPC = 5 Hz), 131.9, 132.2, 132.5,
PA
7.38, 7.96 (m, 24 H, PC6H5 and N2C6H4). 31P{1H}: δP = 57.91 (s).
Synthesis of [Ru{B(NCH2PPh2)2C6H4}Cl(CNC6H2Me3-2,4,6)2]
(5). A solution of [Ru{κ3-HB(NCH2PPh2)2C6H4}Cl2(PPh3)] (1:
0.100 g, 0.105 mmol) and mesityl isocyanide (0.032 g, 0.22 mmol)
in THF (10 mL) was stirred for 66 h. The solvent was removed under
reduced pressure. The residue was redissolved in THF (2 mL) and
then diluted with n-hexane (5 mL). The supernatant was separated
from the precipitate via cannula filtration, and the filtrate was freed of
volatiles under reduced pressure. The resulting residue was washed
with n-hexane (5 + 2 mL) and then diethyl ether (5 + 2 + 2 mL) to
give the product as a colorless solid. Despite this extensive washing, 1H
and 13C{1H} (but not 31P{1H}) NMR spectra consistently revealed
the presence of trace impurities, which we attribute to isonitrile
polymerization (see Supporting Information). Accordingly, satisfactory
elemental microanalytical data were not obtained. X-ray quality crystals
were obtained by slow evaporation of a DCM/n-hexane solution of the
product. Yield: 0.033 g (33%). IR (KBr, cm−1): 3047 νaromCH; 2918,
2842 νCH; 2109, 2073 νCN; 1475, 1434 νaromCC. IR (DCM, cm−1):
2104, 2061 νCN. NMR (CDCl3, 298 K) 1H: δH 1.47 (s, 6H, CH3), 2.08
(s, 3H, CH3), 2.32 (s, 3H, CH3), 2.42 (s, 6H, CH3), 4.42 (d.tv, 2H,
135.3 (C6H5), 135.6 (d, C6H5, JPC = 9 Hz), 135.9 (C6H5), 136.5 (tv,
C6H5, JPC = 6), 140.0 (tv, C6H5,
J
= 17), 144.7 [tv, C1,6(C6H4N2),
1,3
PC
3,4
J
= 7 Hz] (some aryl resonance assignments equivocal or obscured
PC
due to C6D6 overlap). 11B{1H}: δB 49.2 {1H}: δP 32.1 (d, PPh2, 2JPP
=
2
22 Hz), 49.2 (t, PPh3, JPP = 23 Hz). ESI-MS (+ve ion, MeCN): m/z
913.6 [M − Cl]+. Accurate mass: found 949.1307 [M + H]+, calcd for
C50H4511B35Cl214N231P3102Ru 949.1309. Anal. Found: C, 64.37; H,
5.00; N, 3.08. Calcd for C50H44BCl2N2RuP3.0.5(C7H8): C, 64.60; H,
4.86; N, 2.82 (the presence of 0.5 equiv of toluene was crystallo-
graphically confirmed but due to a high degree of disorder was
eliminated employing the PLATON SQUEEZE protocol). Crystal
data for C50H44BCl2N2P3Ru: Mr = 948.62, monoclinic, P21/c, a =
12.4171(8) Å, b = 20.3980(15) Å, c = 18.7584(13) Å, β = 91.922(4)°,
V = 4748.5(6) Å3, Z = 4, Dx = 1.327 Mg m−3, μ(Mo Kα) = 0.58 mm−1,
T = 200(2) K, orange block, 0.09 × 0.08 × 0.06 mm, 6213
independent reflections. F2 refinement, R = 0.089, wR = 0.014 for 3786
reflections (I >2σ(I), 2θmax = 45°), 535 parameters, CCDC 982299.
Intermediate Complex. In the above reaction, consumption of the
starting materials occurred within 15 min, with a second complex
observed in the reaction mixture that gradually converted to the final
PCH2, 2JHH = 12, 2,4JPH = 2), 4.69 (d tv, 2H, PCH2, 2JHH = 12, 2,4JPH
=
1
product. NMR (C7D8, 298 K) H: δH −7.21 (br, 1H, RuH), 4.20 (m,
4 Hz), 6.39 (s, 2H, C6H2), 6.76−8.05 (m, 26H, N2C6H4, C6H2 and
PC6H5). 13C{1H}: δ 18.2, 19.4 (o-MesCH3), 21.0, 21.3 (p-MesCH3),
4H, PCH2N, assignment equivocal due to overlap with final product,
1,3
51.4 (tv, PCH2,
J
= 23), 109.0, 117.9 [C2−5(N2C6H4)], 127.3
PC
aromatic peaks could not be unambiguously assigned due to
2
2
impurities). 31P{1H}: δP 26.5 (t, PPh3, JPP = 24), 34.5 (d, PPh2, JPP
(C6H5), 127.8 (tv, C6H5, JPC = 5), 128.2 (tv, C6H5, JPC = 5), 128.3,
128.7, 130.1 (C6H5), 130.6 (tv, C6H5, JPC = 5 Hz), 132.9, 134.5, 134.6,
134.7 (C6H5), 134.9 (tv, C6H5), JPC = 6 Hz), 137.9 (C6H5), 139.6 [tv,
= 25 Hz).
Synthesis of [Os{κ3-HB(NCH2PPh2)2C6H4}Cl2(PPh3)] (2). A
solution of [OsCl2(PPh3)3]19 (0.100 g, 0.0954 mmol) and HB-
(NCH2PPh2)2C6H4 (0.049 g, 0.0953 mmol) in benzene (5 mL) was
stirred for 30 min. The solvent was removed under reduced pressure,
and the resulting dark brown-black residue was washed with ethanol (7
mL) and n-hexane (5 mL) and dried in vacuo. Brown crystals of the
toluene hemisolvate 2·(C7H8)0.5 suitable for elemental and crystallog-
rahic analysis were obtained by slow diffusion of n-hexane into a
toluene solution of 2. Yield: 0.092 g (93%). IR (KBr, cm−1): 3051
νaromCH; 2954, 2924, 2854 νCH; 1463, 1436 νaromCC. NMR (C6D6, 298
K) 1H: δH −13.72 (m, 1H, OsH), 4.31 (m, 4H, PCH2), 6.46−8.68 (m,
39H, N2C6H4 and C6H5). 13C{1H}: 55.3 (tv, PCH2, 1,3JPC = 24), 111.2,
119.9 [C2−5(N2C6H4)], 130.7, 131.2 (C6H5), 131.3 (d, C6H5, JPC = 3
C1(C6H5),
J
= 19 Hz], 141.4 [tv, C1,6(N2C6H4),
J
= 8], 168.4
1,3
3,4
PC
PC
(RuCN), 171.4 (t, RuCN, JPC = 14 Hz). 11B{1H}: δB 57.3.
31P{1H}: δP 61.2. ESI-MS (+ve ion, MeCN): m/z 945.4 [M − Cl +
MeCN]+, 905.5 [M − Cl]+, 760.3 [M − Cl − CNC6H2Me3]+.
2
Accurate mass: found 905.2648 [M
−
Cl]+, calcd for
C52H5011B14N431P2102Ru 905.2647. Crystal data for C52H50BClN4P2Ru:
M = 940.28, triclinic, P1 (No. 2), a = 11.2613(4) Å, b = 11.7881(4) Å,
̅
r
c = 20.5180(8), α = 100.1449(16)°, β = 95.749(2)°, γ =
118.1335(17)°, V = 2311.12(15) Å3, Z = 2, Dx = 1.351 Mg m−3,
μ(Mo Kα) = 0.51 mm−1, T = 200(2) K, colorless prism, 0.18 × 0.07 ×
0.04 mm, 8093 independent reflections. F2 refinement, R = 0.075, wR
= 0.010 for 3924 reflections (I >2σ(I), 2θmax = 50°), 550 parameters,
CCDC 982300.
Hz), 132.1 (d, C6H5, JPC = 9), 135.0 (d, C6H5, JPC = 9), 136.0 (tv,
1,3
C6H5, JPC = 5), 140.0 [tv, C1(C6H5),
J
= 21 Hz], 144.2
PC
[C1,6(C6H4N2)] (some aromatic peaks obscured due to C6D6 overlap).
ASSOCIATED CONTENT
2
■
11B{1H}: δB 53.2. 31P{1H}: δP 0.3 (d, PPh2, JPP = 12), 7.4 (t, PPh3,
S
* Supporting Information
2JPP = 12 Hz). ESI-MS (+ve ion, MeCN): m/z 1077.7 [M + K]+,
1061.5 [M + Na]+, 1003.6 [M − Cl]+, 967.5 [M − 2Cl]+. Accurate
Crystallographic data for 1 (CCDC 982299), 2·(C7H8)0.5
(CCDC 982298), and 5 (CCDC 982300) in CIF format.
This material is available free of charge via the Internet at
http://pubs.acs.org.
m a s s : f o u n d 1 0 6 1 . 1 6 9 7 [ M
+
N a ] + , c a l c d f o r
C50H4411B35Cl214N223Na31P3192Os 1061.1700. Anal. Found: C, 59.71;
H, 4.38; N, 2.20. Calcd for C50H44BCl2N2OsP3·0.5(C7H8): C, 59.27;
1983
dx.doi.org/10.1021/om5001106 | Organometallics 2014, 33, 1977−1985

Organometallics
Article
Y.-H.; Zhang, Y.; Ding, X.-H. Inorg. Chem. Commun. 2011, 14, 1306−
1310. (q) Fang, H.; Choe, Y.-K; Li, Y.; Shimada, S. Chem.Asian J.
2011, 6, 2512. (r) Garcia-Camprubi, A.; Martin, M.; Sola, E. Inorg.
Chem. 2010, 49, 10649−10657. (s) Sola, E.; Garcia-Camprubi, A.;
Andres, J. L.; Martin, M.; Plou, P. J. Am. Chem. Soc. 2010, 132, 9111−
9121. (t) Hill, A. F.; Neumann, H.; Wagler, J. Organometallics 2010,
29, 1026−1031. (u) Dixon, L. H. S.; Hill, A. F.; Sinha, A.; Ward, J. S.
Organometallics 2014, 23, 653−658. (v) Wu, S.; Li, X.; Xiong, Z.; Xu,
W.; Lu, Y.; Sun, H. Organometallics 2013, 22, 3227−3237. (w) Bernal,
M. J.; Torres, O.; Martín, M.; Sola, E. J. Am. Chem. Soc. 2013, 135,
19008−19015.
(7) Zhu, J.; Lin, Z.; Marder, T. B. Inorg. Chem. 2005, 44, 9384.
(8) Herein, we will employ the covalent bond classification (Z, X, L
etc.) recommended by Green to indicate the nature of the LaxLeqLax
donor set: Green, M. L. H. J. Organomet. Chem. 1995, 500, 127−148.
Metallaboratranes such as [IrH(CO)(PPh3){BH(mt)2}] (mt =
methimazolyl) would accordingly be designated as LZL-SBS pincers.
(9) Notwithstanding the multitude of definitions for “valency”, use of
the terms oxidative addition or oxidation state in this context is of
questionable utility given that both boron and hydrogen have
electronegativities lower than or comparable to those of the platinum
group metals.
AUTHOR INFORMATION
Corresponding Author
*E-mail: a.hill@anu.edu.au.
Notes
■
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
This work was supported by the Australian Research Council
(DP110101611, DP130102598). The assistance of Dr.
Anthony C. Willis in the acquisition and interpretation of
crystallographic data is gratefully acknowledged.
REFERENCES
■
(1) For recent reviews covering metal−boron dative bonding see:
(a) Bouhadir, G.; Amgoune, A.; Bourissou, D. Adv. Organomet. Chem.
2010, 58, 1−107. (b) Braunschweig, H.; Dewhurst, R. D. Dalton Trans.
2011, 40, 549−558. (c) Braunschweig, H.; Dewhurst, R. D.;
Schneider, A. Chem. Rev. 2010, 110, 3924−3957.
(2) (a) Crossley, I. R.; Hill, A. F.; Willis, A. C. Organometallics 2005,
24, 1062−1064. (b) Segawa, Y.; Yamashita, M.; Nozaki, K.
Organometallics 2009, 28, 6234−6242. (c) Segawa, Y.; Yamashita,
M.; Nozaki, K. J. Am. Chem. Soc. 2009, 131, 9201−9203. (d) Hasegawa,
M.; Segawa, Y.; Yamashita, M.; Nozaki, K. Angew. Chem., Int. Ed. 2012,
51, 5956−6960. (e) Masuda, Y.; Hasegawa, M.; Yamashita, M.;
Nozaki, K.; Ishida, N.; Murakami, M. J. Am. Chem. Soc. 2013, 135,
7142−7145. (f) Ogawa, H.; Yamashita, M. Dalton Trans. 2013, 42,
625−629. (g) Spokoyny, A. M.; Reuter, M. G.; Stern, C. L.; Ratner, M.
A.; Seideman, T.; Mirkin, C. A. J. Am. Chem. Soc. 2009, 131, 9482−
9483. (h) El-Zaria, M. E.; Arii, H.; Nakamura, H. Inorg. Chem. 2011,
50, 4149−4161. (i) Hill, A. F.; Lee, S. B.; Park, J.; Shang, R.; Willis, A.
C. Organometallics 2010, 29, 5661−5669. (j) Miyada, T.; Yamashita,
M. Organometallics 2013, 22, 5281−5284. (k) Lin, T.-P.; Peters, J. C. J.
Am. Chem. Soc. 2013, 135, 15310−15313.
(3) (a) Irvine, G. J.; Lesley, M. J. G.; Marder, T. B.; Norman, N. C.;
Rice, C. R.; Robins, E. G.; Roper, W. R.; Whittell, G. R.; Wright, L. J.
Chem. Rev. 1998, 98, 2685−2722. (b) Smith, M. R., III. Prog. Inorg.
Chem. 1999, 48, 505−567. (c) Aldridge, S.; Coombs, D. L. Coord.
Chem. Rev. 2004, 248, 535−559. (d) Braunschweig, H.; Colling, M.
Coord. Chem. Rev. 2001, 223, 1−51.
(10) In contrast to the trigonal bipyramidal geometry, which is
preferred for d8-ML5 complexes, the corresponding singlet d6-ML5
system is prone to a Jahn−Teller distortion away from the idealized
tbp geometry, for a which an electronic degeneracy (dxy compared
with dx −y ) would exist. For complexes of the form trans-ML3(PR3)2
2
2
some relaxation of this requirement might be expected for a
heteroleptic ML3 equaltorial ligand set.
(11) (a) Riehl, J. F.; Jean, Y.; Eisenstein, O.; Pelissier, M.
Organometallics 1992, 11, 729−737. (b) Jean, Y.; Eisenstein, O.
Polyhedron 1988, 7, 405−407.
(12) Lam, W. H.; Shimada, S.; Batsanov, A. S.; Lin, L.; Marder, T. B.;
Cowan, J. A.; Howard, J. A. K.; Mason, S. A.; McIntyre, G. J.
Organometallics 2003, 22, 4557−4568. This complex was also
characterized by neutron diffraction and computationally interrogated;
however for consistency the X-ray-derived metrical parameters are
given here.
(13) As far as we can ascertain, the only structural data for an
osmium σ-borane held in the Cambridge Crystallographic Data Base
pertain to the metallacyclic complex [OsH{H(BBN)CH2PMe2}
(PMe3)3] obtained from the reaction of [OsH(CH2PMe2)(PMe3)3]
with (HBBN)2: Baker, R. T.; Calabrese, J. C.; Westcott, S. A.; Marder,
T. B. J. Am. Chem. Soc. 1995, 117, 8777−8784.
(4) (a) Morales-Morales, D.; Jensen, C. G. M. The Chemistry of Pincer
Compounds; Elsevier Science: 2007. (b) Peris, E.; Crabtree, R. H.
Coord. Chem. Rev. 2004, 248, 2239−2246.
(5) Chase, P. A.; Koten, G. V. The Pincer Ligand: Its Chemistry and
Applications (Catalytic Science), 1 ed.; Imperial College Press: London,
2010.
(14) A (non σ-) borane complex [Os(CO)(PPh3){B(mt)3}] (mt =
methimazolyl) has been structurally characterized but has the
metallaboratrane structure with a direct 2c-2e Os→B interaction, i.e.,
without a 3c-2e B−H−Os connectivity. The isolated but soultion
unstable σ-borane complex [Os(C6H5)(CO)(PPh3){κ3-H,S,S′-
HB(mt)3}] has been proposed as an intermediate en route to the
formation of [Os(CO)(PPh3){B(mt)3}]: Foreman, F. R. St.-J; Hill, A.
F.; White, A. J. P.; Williams, D. J. Organometallics 2004, 23, 913−916.
(15) For metallaboratranes, The (M→B)n formalism denotes n metal
valence electrons including the pair associated with the dative M→B
bond: Hill, A. F. Organometallics 2006, 25, 4741−4743.
(16) Foreman, M. R. St.-J.; Hill, A. F.; Owen, G. R.; White, A. J. P.;
Williams, D. J. Organometallics 2003, 22, 4446−4450.
(6) (a) Stobart, S. R.; Zhou, X.; Cea-Olivares, R.; Toscano, A.
Organometallics 2001, 20, 4766−4768. (b) Bushnell, G. W.; Casado,
M. A.; Stobart, S. R. Organometallics 2001, 20, 601−603. (c) Zhou, X.;
Stobart, S. R. Organometallics 2001, 20, 1898−1900. (d) Brost, R. D.;
Bruce, G. C.; Joslin, F. L.; Stobart, S. R. Organometallics 1997, 16,
5669−5680. (e) Tilley, T. D.; Sangtrirutnugul, P. Organometallics
2008, 27, 2223−2230. (f) Kwok, W.-H.; Lu, G.-L.; Rickard, C. E. F.;
Roper, W. R.; Wright, L. J. J. Organomet. Chem. 2004, 689, 2511−
2522. (g) Ruddy, A. J.; Mitton, S. J.; McDonald, R.; Turculet, L. Chem.
Commun. 2012, 48, 1159−1161. (h) MacInnis, M. C.; McDonald, R.;
Ferguson, M. J.; Tobisch, S.; Turculet, L. J. Am. Chem. Soc. 2011, 133,
13622−13633. (i) Mitton, S. J.; McDonald, R.; Turculet, L. Angew.
Chem., Int. Ed. 2009, 48, 8568−8571. (j) Morgan, E.; MacLean, D. F.;
McDonald, R.; Turculet, L. J. Am. Chem. Soc. 2009, 131, 14234−
14236. (k) Mitton, S. J.; McDonald, R.; Turculet, L. Organometallics
2009, 28, 5122−5136. (l) MacLean, D. F.; McDonald, R.; Ferguson,
M. J.; Caddell, A. J.; Turculet, L. Chem. Commun. 2008, 5146−5148.
(m) MacInnis, M. C.; MacLean, D. F.; Lundgren, R. J.; McDonald, R.;
Turculet, L. Organometallics 2007, 26, 6522−6525. (n) Korchin, E. E.;
Leitus, G.; Shimon, L. J. W.; Konstantinovski, L.; Milstein, D. Inorg.
Chem. 2008, 47, 7177−7189. (o) Li, Y.-H.; Ding, X.-H.; Zhang, Y.; He,
W.-R.; Huang, W. Inorg. Chem. Commun. 2012, 15, 194−197. (p) Li,
(17) Abernethy, R. J.; Hill, A. F.; Tshabang, N.; Willis, A. C.; Young,
R. D. Organometallics 2009, 28, 488−492.
(18) Hallman, P. S.; Stephenson, T. A.; Wilkinson, G. Inorg. Synth.
1970, 12, 237−240.
(19) Elliott, G. P.; McAuley, N. M.; Roper, W. R.; Shapley, P. A.
Inorg. Synth. 2009, 26, 184−185.
(20) Abernethy, R. J.; Hill, A. F.; Smith, M. K.; Willis, A. C.
Organometallics 2009, 28, 6152−6159.
(21) ConQuest, Version 1.15, Cambridge Crystallographic Data
Centre, February 2014 release.
(22) (a) Rickard, C. E. F.; Roper, W. R.; Williamson, A.; Wright, L. J.
J. Organomet. Chem. 2004, 689, 1609−1616. (b) Rickard, C. E. F.;
1984
dx.doi.org/10.1021/om5001106 | Organometallics 2014, 33, 1977−1985

Organometallics
Article
Roper, W. R.; Williamson, A.; Wright, L. J. Organometallics 1998, 17,
4869−4874. (c) Rickard, C. E. F.; Roper, W. R.; Williamson, A.;
Wright, L. J. Organometallics 2002, 21, 4862−4872. (d) Rickard, C. E.
F.; Roper, W. R.; Williamson, A.; Wright, L. J. Angew. Chem., Int. Ed.
1999, 38, 1110−1113. (e) Irvine, G. J.; Rickard, C. E. F.; Roper, W. R.;
Williamson, A.; Wright, L. J. Angew. Chem., Int. Ed. 2000, 39, 948−950.
(f) Rickard, C. E. F.; Roper, W. R.; Williamson, A.; Wright, L. J.
Organometallics 2002, 21, 1714−1718. (g) Rickard, C. E. F.; Roper, W.
R.; Williamson, A.; Wright, L. J. Organometallics 2000, 19, 4344−4355.
(23) A search of the CCDC21 for the N2HB···Cl connectivity about
boron in the range 2.0 < B···Cl < 6.0 Å revealed only four compounds,
each of which had conventional covalent B−Cl bonds that fell within
the narrow range 1.84−1.97 Å: (a) Mohlen, M.; Neumuller, B.;
Dashti-Mommertz, A.; Muller, C.; Massa, W.; Dehnicke, K. Z. Anorg.
Allg. Chem. 1999, 625, 1631. (b) Maringgele, W.; Noltemeyer, M.;
Schmidt, H.-G.; Meller, A. Main Group Met. Chem. 1999, 22, 715.
(c) Hill, A. F.; Smith, M. K. Chem. Commun. 2005, 1920. (d)
A
titanocene derivative has been described in which one cyclo-
pentadienyl ligand bears a bis(pentafluorphenyl)boryl substituent,
which interacts with a titanium-bound chloride (B···Cl = 2.007 Å);
however the Ti−Cl bond is substantially elongated (2.483 Å)
compared with the terminal Ti−Cl ligand (2.304 Å): Lancaster, S.
J.; Al-Benna, S.; Thornton-Pett, M.; Bochmann, M. Organometallics
2000, 19, 1599.
(24) For the purposes of this discussion, metallapolyborane clusters
and complexes with Ru(μ-H)2B geometries are omitted.
(25) Kakizawa, T.; Kawano, Y.; Shimoi, M. Organometallics 2001, 20,
3211−3213.
(26) Schlecht, S.; Hartwig, J. F. J. Am. Chem. Soc. 2000, 122, 9435−
9443.
(27) Braunschweig, H.; Colling, M.; Kollann, C.; Englert, U. J. Chem.
Soc., Dalton Trans. 2002, 2289−2296.
(28) Orpen has suggested demarcating regimes for short (≤2.52 Å),
intermediate (2.52−2.95 Å), and long (2.95−3.15 Å) H···Cl−M
́
hydrogen bonding to chloride ligands: Aullon, G.; Bellamy, D.;
Brammer, L.; Burton, E. A.; Orpen, A. G. Chem. Commun. 1998, 653.
(29) Although metallaboratranes based on d6-metal centers have yet
to be isolated, simple electron-counting arguments suggest a seven-
coordinate MX2L4Z system would have the appropriate orbital
occupancy to sustain M→B bonding if the remaining X and ligands
were sufficiently σ-basic.
(30) (a) Buil, M. L.; Esteruelas, M. A.; Garces, K.; Onate, E. J. Am.
Chem. Soc. 2011, 133, 2250−2263. (b) Clark, G. R.; Irvine, G. J.;
Roper, W. R.; Wright, L. J. J. Organomet. Chem. 2003, 680, 81−88.
(31) Crossley, I. R.; Hill, A. F. Dalton Trans. 2008, 201−203.
(32) For further examples see: (a) Tsoureas, N.; Kuo, Y.-Y.; Haddow,
M. F.; Owen, G. R. Chem. Commun. 2011, 47, 484−486. (b) Harman,
W. H.; Peters, J. C. J. Am. Chem. Soc. 2012, 134, 5080−5082. (c) Fong,
H.; Moret, M.-E.; Lee, Y.; Peters, J. C. Organometallics 2013, 32,
3053−3062.
(33) Alcaraz, G.; Helmstedt, U.; Clot, E.; Vendier, L.; Sabo-Etienne,
S. J. Am. Chem. Soc. 2008, 130, 12878−12879.
1985
dx.doi.org/10.1021/om5001106 | Organometallics 2014, 33, 1977−1985