P(CH)P pincer rhodium(I) complexes: The key role of electron-poor imidazoliophosphine extremities

Article abstract of DOI:10.1021/ic3006508

The coordination chemistry of a potentially pincer-type dicationic meta-phenylene-bis(imidazoliophosphine) ligand 3 to neutral and cationic carbonylrhodium(I) centers has been investigated. Similarly to what was observed previously for its ortho-phenylene counterpart, 3 was found to bind to the RhCl(CO) fragment in a trans-chelating manner that makes possible a weak Rh-C(H) interaction, inferred from the nonbonding but relatively short Rh-C and Rh-H contacts observed in the solid state structure of the dicationic adduct (3)RhCl(CO) (5). Formation of the target PCP-type pincer complex could not be triggered despite multiple attempts to deprotonate the central C-H moiety in the initial dicationic adduct 5, or in the tricationic species [(3)Rh(CO)] ⁺ (8) generated by abstraction of the chloride ion from 5. Complex 8 was identified on the basis of NMR and IR analyses as a Rh(I)-stabilized P(CH)P-intermediate en route to the anticipated classical PCP-type pincer complex. Analysis of the electronic structure of this intermediate computed at the density functional level of theory (DFT level) revealed a bonding overlap between a Rh d-orbital and π-orbitals of the m-phenylene ring. NBO analyses and calculated Wiberg indices confirm that this interaction comprises an η¹-C-Rh bonding mode, with only secondary contributions from the geminal C and H atoms. Although the target PCP-type pincer complex could not be generated, treatment of the tricationic intermediate with methanol induced a P-CN₂ bond cleavage at both imidazoliophosphine moieties, resulting in the formation of a dicationic open pincer species, that is, a nonchelated bis((MeO)PPh₂)-stabilized aryl-Rhodium complex that is the κC-only analogue of the putative κP,κC,κP-PCP complex sought initially. Theoretical studies at the DFT level of experimental or putative species relevant to the final C-H activation process ruled out the oxidative addition pathway. Two alternative pathways are proposed to explain the formation of the open pincer complex, one based on an organometallic α-elimination step, the other based on an organic aromatization-driven β-elimination process.

Full text of DOI:10.1021/ic3006508

Article
pubs.acs.org/IC
P(CH)P Pincer Rhodium(I) Complexes: The Key Role of Electron-Poor
Imidazoliophosphine Extremities
Cecile Barthes,†,‡ Christine Lepetit,†,‡ Yves Canac,*,†,‡ Carine Duhayon,†,‡ Davit Zargarian,§
́
and Remi Chauvin*,†,‡
†CNRS, LCC (Laboratoire de Chimie de Coordination), 205, route de Narbonne, F-31077 Toulouse, France
‡Universite
§Dep
artement de Chimie, Universite
́
de Toulouse, UPS, INP, LCC, F-31077 Toulouse, France
de Montreal, Montreal, Quebec H3C 3J7, Canada
́
́
́
́
́
S
* Supporting Information
ABSTRACT: The coordination chemistry of a potentially pincer-
type dicationic meta-phenylene-bis(imidazoliophosphine)
ligand 3 to neutral and cationic carbonylrhodium(I) centers
has been investigated. Similarly to what was observed previously
for its ortho-phenylene counterpart, 3 was found to bind to the
RhCl(CO) fragment in a trans-chelating manner that makes
possible a weak Rh−C(H) interaction, inferred from the non-
bonding but relatively short Rh−C and Rh−H contacts observed in the solid state structure of the dicationic adduct
(3)RhCl(CO) (5). Formation of the target PCP-type pincer complex could not be triggered despite multiple attempts to
deprotonate the central C−H moiety in the initial dicationic adduct 5, or in the tricationic species [(3)Rh(CO)]+ (8) generated
by abstraction of the chloride ion from 5. Complex 8 was identified on the basis of NMR and IR analyses as a Rh(I)-stabilized
P(CH)P-intermediate en route to the anticipated classical PCP-type pincer complex. Analysis of the electronic structure of this
intermediate computed at the density functional level of theory (DFT level) revealed a bonding overlap between a Rh d-orbital
and π-orbitals of the m-phenylene ring. NBO analyses and calculated Wiberg indices confirm that this interaction comprises an
η1-C−Rh bonding mode, with only secondary contributions from the geminal C and H atoms. Although the target PCP-type
pincer complex could not be generated, treatment of the tricationic intermediate with methanol induced a P−CN2 bond cleavage
at both imidazoliophosphine moieties, resulting in the formation of a dicationic “open pincer” species, that is, a nonchelated
bis((MeO)PPh2)-stabilized aryl-Rhodium complex that is the κC-only analogue of the putative κP,κC,κP-PCP complex sought
initially. Theoretical studies at the DFT level of experimental or putative species relevant to the final C−H activation process ruled
out the oxidative addition pathway. Two alternative pathways are proposed to explain the formation of the “open pincer” complex,
one based on an organometallic α-elimination step, the other based on an organic aromatization-driven β-elimination process.
Scheme 1. Typology of Transition Metal Pincer Complexes,
Featuring Either an Electron-Rich Tridentate P-Csp2-P
Ligand A (left), or an Electron-Poor Tridentate P-Csp2-P
Ligand B (right)
1. INTRODUCTION
Rigid tripodal ligands and their transition metal complexes
are deserving of the generic appellation “pincer” owing to their
unique robustness, enhanced catalytic properties, and diversity.1
Strictly speaking, the term “pincer” currently refers to a tridentate
ligand of the LXL′ type (in the Green formalism), that binds to
metals in a double chelating mer fashion, which offers added
protection against hydrolytic or oxidative cleavage. Their attrac-
tive reactivity and material properties apart, the synthesis of pincer
complexes is of topical interest, in particular the key-process of
X-metal bond formation when the X center is the central carbon atom
of a meta-L,L′-disubstituted phenyl core. The kinetic and thermo-
dynamic stability of such LCL′-type pincer complexes is determined
not only by the tripodal chelating mode and the sp2 hybridization
state of the anionic C atom, but also by the nature of L and L′,
which may be P-, N-, S-, Se-, or C-centered donor fragments with
various steric and electronic demands. Among the latter,
electron-donating trialkyl- phosphinyl arms are the most popular
constituents of “electron-rich PCP pincers” A (Scheme 1).2
More recently, the complementary category of “electron-poor
PCP pincers” B has been exemplified by the introduction of
strongly π-accepting phosphanyl arms, like fluoroalkyl-3 and aryl-
phosphanes,4 N-pyrrolylphosphanes,5 phosphites,6 and, to a
lesser extent, phosphoramidites7 and phosphinites.8
In the design of optimal ligands for catalysis, a single type of co-
ordinating moiety with extreme donating character (strong or weak)
Received: March 28, 2012
© XXXX American Chemical Society
A
dx.doi.org/10.1021/ic3006508 | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Scheme 2. Selective Sequential Phosphinylation and Methylation of 1,3-Bis(imidazolyl)benzene 1
Figure 1. Molecular views of the X-ray crystal structures of the neutral and dicationic diphosphines 2 (left) and 3 (right), with thermal ellipsoids drawn at
the 30% probability level (for clarity, the triflate anions and the H atoms are omitted). Selected bond distances (Å) and angles (deg). 2: C1−P1
1.8266(18), C10−P2 1.8222(19), N1−C1 1.319(2), N2−C1 1.382(2), N1−C1−N2 110.39(15), N2−C1−P1 124.51(13). 3: C1−P1 1.833(4), N1−C1
1.360(5), N2−C1 1.368(4), N1−C1−N2 104.5(3), N2−C1−P1 121.5(3).
might be too limiting because transition metal centers must
satisfy varying criteria for achieving selectivity and activity
over all the steps in a given catalytic cycle.9 Along this direc-
tion, PCP pincers of type B might exhibit not only particular
catalytic properties, but also peculiar photophysical properties
resulting from a possible C→M→L2 charge transfer of the
“LMLCT” type. Perfluoroalkyl-P-substituted PCP pincers stand
as the most π-accepting examples reported hitherto,3 but since,
in a very broad sense, cationic centers are more electronegative
than fluorine atoms (basically because the range of the second
ionization potentials is strictly higher than the range of the
first ionization potentials), the introduction of imidazolium
P-substituents could generate an attractive new familly of stable
pincers of the P+CP+ variety. Imidazoliophosphanes are indeed
more accurately described as NHC→phosphenium adducts,10
where the positive charge formally located on the Lewis acceptor
P center (instead of the α-position as in generic carbeniophos-
phanes11) illustrates the decrease of the electron resource and
thus of the donor ability of the P ligand. Moreover, because of the
rigid frame of the pincer structure, the expected trans- arrange-
ment of the two P+ moieties should be compatible with the
formal electrostatic repulsion between the positive charges.
In this context, a systematic investigation of the coordination
chemistry of P+,P+-chelating bis(imidazoliophosphine) ligands
has been initiated. A recent report has documented how such a
ligand based on the o-phenylene bridge can behave as a versatile
cis- or trans-chelating ligand for Rh(I) centers depending on
the electronic endowment of the latter: trans-chelating at a soft
neutral RhCl(CO) center, cis-chelating at a harder cationic
[Rh(MeCN)2]+ center.12 As a follow-up to these ongoing studies
and with the ultimate goal of developing the afore-proposed
category of elusive P+CP+ pincer ligands, the coordination
chemistry of a P+,P+-chelating bis(imidazoliophosphine) ligand
based on the m-phenylene bridge has been examined. The present
contribution addresses the coordinating behavior of this a
priori less rigid ligand toward Rh(I) centers using both experi-
mental and theoretical investigation tools.
2. RESULTS AND DISCUSSION
2.1. Experimental Results. Double deprotonation of
the readily available 1,3-bis(imidazolyl)benzene 113 with 2 equiv
of n-BuLi in tetrahydrofuran (THF), followed by addition of a
stoichiometric amount of chlorodiphenylphosphine afforded the
neutral diphosphine 2 in 83% yield. Upon subsequent treatment
with 2 equiv of methyl triflate (MeOTf) in CH2Cl2, the targeted
dicationic diphosphine 3 was readily obtained in 95% yield
(Scheme 2). The ionic structure of 3 was reflected in its low
solubility in nonpolar solvents. The symmetric structures of 2
and 3 were indicated by the singlet resonances in their 31P NMR
spectra (δp = −30.2 ppm and δp = −21.0 ppm, respectively).
1
A singlet resonance appearing at +3.52 ppm in the H NMR
spectrum of 3 was assigned to the N−CH3 protons. The solid
state structures of 2 and 3 were determined by X-ray diffraction
analysis of colorless single crystals obtained from recrystallization
in CH2Cl2/toluene (2) or CH3CN/Et2O (3). Molecular views
and selected structural parameters are shown in Figure 1.14
The bis(imidazoliophosphine) 3 was then reacted with 0.5
equiv of [RhCl(cod)]2 in CH2Cl2 (cod = 1,5-cyclooctadiene).
The 31P NMR spectrum of the reaction mixture showed the
complete disappearance of 3, and the appearance in a 1:4 ratio of
two resonances at lower field (d, δp = +27.4 ppm, JPRh = 149.9 Hz;
d, δp = +26.3 ppm, JPRh = 151.9 Hz): these signals were tenta-
tively attributed to the formation of the expected 18-electron
Rh(I) complex 4 (Scheme 3) that may adopt two geometrical
configurations allowed by the chelating mode of the cod ligand:
equatorial−equatorial or axial−equatorial. In each isomer, the
observed equivalence of the two phosphorus atoms implies that
they occupy two axial or two equatorial equivalent positions, while
the cod ligand would thus be chelating in either equatorial−
equatorial or axial−equatorial mode, respectively. The reluctance
of the dicationic diphosphine 3 for the axial−equatorial chelating
B
dx.doi.org/10.1021/ic3006508 | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Scheme 3. Preparation of Dicationic Rhodium Complexes 4−5 of the Bis(amidiniophosphine) Ligand 3
mode could indeed be dictated by the formal +/+ electrostatic
repulsion, as recently invoked to explain the trans-chelating
preference of the o-phenylene-bridged isomer of 3 in the
palladium series.12b Complex 4 was isolated and characterized by
spectroscopy, HRMS, and by derivatization to a CO adduct.
Thus, bubbling CO through a CH2Cl2 solution of the isomeric
mixture of 4 followed by NMR analysis indicated the conversion
of the precursor 31P resonances to a unique signal at +18.7 ppm
(d, JPRh = 135.7 Hz). In addition, a unique 13C NMR signal was
observed at δC = +182.9 ppm (td, JRhC = 74.2 Hz, JPC = 15.1 Hz),
indicating a single type of CO ligand. These observations are in
agreement with the expected symmetrical structure of complex 5
(Scheme 3). This new derivative was isolated in 76% yield and
fully characterized, including by X-ray diffraction analysis
performed on yellow crystals deposited from a MeCN/Et2O
mixture (Figure 2).14
of coligands,15 and the measured value for 5 at 2005 cm−1 is of
the same order of magnitude as those reported for trans-
RhCl(CO)[P(OMe)3]2 (2011 cm−1),16 and for the closely
related trans-bis(imidazoliophosphine) complexes 6 (2003
cm−1),17 and 7 (2007 cm−1).12a These results indicate that the
σ-donor vs π-acceptor character of imidazoliophosphine ligands
is similar to that of phosphite ligands, whatever the bridge
constraint (Scheme 4). Since, the o- and m-phenylene bridges,
Scheme 4. IR Stretching Frequencies vCO (cm−1) of
Isostructural trans-Coordinated Rhodiumcarbonyl
Complexes of Trimethylphosphite, and Monodentate or
Chelating Bis(imidazolio-diphenylphosphine) Ligands of 6, 5,
and 7, Respectively
occurring in 7 and 5 respectively, have no influence on the donor
ability of the imidazoliophosphine moieties, they can thus be
considered as electronically insulating. Finally, while the trans-
chelation observed in 7 could have been partly attributed to a
steric repulsion between the closely facing N−Me groups,12b the
same trans-coordination observed in the less constrained com-
plex 5 (and all the more in 6) is only attributed to a steric
repulsion between the two donor ends of the ligands.
Figure 2. Molecular view of the X-ray crystal structure of the Rh(I)
complex 5, with thermal ellipsoids drawn at the 30% probability level
(for clarity, the triflate anions and the H atoms are omitted). Selected
bond distances (Å) and angles (deg). 5: C1−P1 1.8472(17), C10−P2
1.8402(16), N1−C1 1.3420(19), N2−C1 1.358(2), P1−Rh1,
2.3107(4) P2−Rh1, 2.2998(4), Cl1−Rh1, 2.3700(4), C1−P1−Rh1
102.80(5), N1−C1−N2 106.30(14), P1−Rh1−P2 176.80(16), Cl1−
Rh1−O1 178.83(5).
Having concluded that derivatives 4 and 5 would not proceed
spontaneously to form the target pincer complex, the requisite
metalation step was attempted by aiding the abstraction of the
central proton (H1) of the m-phenylene ring in 5. However,
complex 5 remained impervious to all attempts using different
bases (Et3N, n-BuLi, tert-BuOK, NaH, AgOAc) and various
conditions (solvents, temperature). Since H1 thus appeared less
acidic than anticipated for such a dicationic complex, we under-
took to abstract the chloride ligand, thereby increasing the overall
positive charge of the complex.
The Rh(I) atom in the solid state structure of 5 was found at
the center of a quasi-square-planar environment, where the trans-
arrangement of the two imidazoliophosphine moieties is
confirmed by the P−Rh−P and CO−Rh−Cl angles of 176.80°
and 178.83°, respectively. It is also worth noting here that while
the Rh···C1 and Rh···H1 distances of about 2.77 Å are definitely
nonbonding, they are shorter than the corresponding sums of
van der Waals radii (see Section 2.2, Figure 3 and Table 1).
A rationale for the presence of a single CO ligand in trans
position to the chloride ligand is a competition between the
strong π-accepting character of the former and the electron-poor
character of the “perpendicular” trans-chelating ligand 3, both
disfavoring spontaneous displacement of the Cl− ligand from a
Rh+ center. The single vCO IR stretching frequency of RhCl(CO)
complexes is commonly used for estimating the donor character
Treating 5 with a stoichiometric amount of AgOTf in MeCN
at room temperature, resulted in the formation of a white
precipitate (AgCl). The 31P NMR spectrum of the crude reaction
mixture displayed a new signal at δp = +12.4 ppm (d, JPRh = 127.6
Hz), which was attributed to the rhodium complex 8, finally
isolated in 94% yield (Scheme 5). Abstraction of the chloride
ligand in 5 was also manifest in the new signals observed in 13
C
C
dx.doi.org/10.1021/ic3006508 | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Scheme 5. Sequential Formation of the Tricationic Bis(imidazoliophosphine) Complex 8 and Dicationic Diphosphinite Complex
9, from the Rhodium(I) Precursor 5
NMR (δCO = +187.7 ppm, broad d, JCRh = 70.4 Hz) and IR (νCO
=
of the original ligand 3, respectively. The presence of the CO
ligand was further confirmed by IR spectroscopy (νCO = 2021
cm−1). Finally, multinuclear 1D-2D NMR spectroscopy and
mass spectroscopy [m/z = 951.1229 (M − TfO−)+] allowed
the structure 9 to be assigned to the new product (Scheme 5),
which could be isolated in 90% yield. Noteworthy, two sets of
2046 cm−1) spectra of 8, while its tricationic character (Scheme 5)
was evidenced by ESI MS and HRMS data [m/z = 1037.0303
(M − TfO−)+]. That this new species is not the target PCP
pincer complex was evident from the similar 31P chemical shifts
and JPRh values of 5 and 8 (δp = +18.7 vs +12.4 ppm; JPRh = 136 vs
128 Hz) which suggest only minor modifications in the elec-
tronic environment of the Rh(I) center.
1
13C and H NMR signals are observed for the N-methyl-
imidazolium nuclei of 9 (δ(N2CH) = 135.8 and 135.4 ppm,
δ(N2CH) = 9.59 and 9.51 ppm, δ(NCH) 36.8 and 36.7 ppm,
The formal 14-electron count at the Rh(I) center in 8 ob-
viously requires additional stabilization: this can be achieved
by either labile coordination of a MeCN solvent molecule (as
previously observed in a related tricationic rhodium complex),12a
or intramolecular weak interaction of the Rh+ center with the
vicinal C1−H1 bond of the m-phenylene bridge of the ligand 3.
The latter scenario is supported by the observation of a broad
1
δ(NCH) = 4.13 and 4.09 ppm): as other H, 13C, and 31P
signals are not split, this suggests diastereotopic conforma-
tional environments for the two imidazolium moieties (see
Supporting Information).
Formation of the “open pincer” complex 9, featuring a non-
chelated aryl ligand is consistent with a nucleophilic displace-
ment of the phosphenium centers from the NHC fragments by
methanol molecules. A similar reactivity of chloride anions as
alternative weak nucleophiles was indeed observed in [Pd(η2-
BIMIONAP)Cl2], a complex of the BIMIONAP imidazolio-
phosphine ligand.10c−e
2.2. Theoretical Investigations. Geometries and NMR
spectra of the initially targeted PCP pincer complex 10, chlo-
rinated precursor 5, P(CH)P pincer complex 8, and mono-
dentate aryl complex 9 were calculated at the PCM-B3PW91/
6-31G**/LANL2DZ*(Rh) level (Figure 3). The triflate
counterions were not included in the calculations, but the
acetonitrile solvent was implicitly accounted for as a dielectric
continuum (ε = 35.688).
1
deshielded H NMR signal at +9.80 ppm, coupled to a 13C
nucleus in the aromatic region (δC = +111.4 ppm, broad s, in
the 13C−1H HMQC spectrum). The width of the 13C−1H
resonances is also consistent with a weak interaction with the
103Rh nucleus (JRh,H and JRh,C ≤ 5 Hz). Since all attempts at
growing high-quality single crystals of 8 were unsuccessful, the
exact nature of the Rh···(C−H) interaction in this P(CH)P
complex was investigated on the basis of theoretical calculations
(see Section 2.2). The acronym “P(CH)P” is coined here for
such conjugate acid precursors of classical PCP pincer com-
plexes, the corresponding transformation formally consisting in
oxidative addition and deprotonation.
The observed Rh···(C−H) interaction in tricationic 8
suggested that direct deprotonation of the C−H moiety might
be possible. However, all attempts with various bases (Et3N,
n-BuLi, tert-BuOK, KHMDS, NaH, AgOAc) proved futile, and
depending on the nature of the base, unreacted starting material
or decomposition products were recovered. This paradoxical
nonselective acidic behavior of the C−H bond in the vicinity of
three formal positive charges is addressed by theoretical studies
in Section 2.2. Multiple unsuccessful attempts at growing single
crystals of 8 failed to allow unequivocal identification of this
species. An unexpected transformation was observed during one
such attempt when dry methanol was used as solvent. Monitor-
ing the reaction by 31P NMR spectroscopy at room temperature
showed the complete disappearance of the signal of 8 (δp =
+12.4 ppm, d, JPRh = 127.6 Hz) after 12 h, and the concomitant
emergence of a new signal at δp = +124.9 ppm (d, JPRh = 143.8 Hz).
The significantly different chemical shift of the new product
indicates a major change in the chemical environment of the Rh
center, but the key information for identifying the product
was provided by 13C NMR spectroscopy, which showed two
quaternary carbon atoms resonating at low field as triplets of
doublets: δC = +190.7 ppm (td, JCRh = 57.9 Hz, JCP = 16.3 Hz),
δC = +177.0 ppm (td, JCRh = 45.2 Hz, JCP = 16.3 Hz). These 13C
chemical shifts were ultimately attributed to a remaining CO
ligand and to the central sp2-C1 atom of the m-phenylene bridge
The calculation method is validated by a good agreement
between the calculated Cs-symmetric structure of 5 and the
corresponding XRD data (Table 1). As previously reported for
related complexes at this calculation level,10c only the Rh−P and
P−C bonds are found overestimated by about 0.05 Å. The
Rh···Ci distances (i = 1−3), ranging from 2.8 to 3.3 Å, are
indicative of weak Rh-Ci interactions: they are indeed sig-
nificantly shorter than the corresponding sum of van der Waals
radii (3.70 Å) but much longer than the sum of their covalent
radii (2.10 Å).18 The Rh···H1 distance of 2.85 Å (vs 2.78 Å from
the XRD structure determination) is also slightly shorter than the
corresponding sum of van der Waals radii (3.10 Å)20 but the
dominating Rh···(C1−H1) interaction is rather centered at the
C1 atom. The calculated solution structure of 5 was further
confirmed by the agreement between the experimentally
1
measured 13C, 31P, and H NMR chemical shifts and the ones
calculated at the B3PW91/6-31+G**/LANL2DZ*(Rh) level on
the static optimized geometries (see the Supporting Information,
Table S1).
The calculated structure for 8, which was obtained from
chloride abstraction from 5, deviates slightly from Cs-symmetry
(Figure 3). The Rh···Ci distances (i = 1−3) in 8 are significantly
shortened (up to 0.30 Å for the Rh···C1 distance), whereas the
C1−C2 and C1−C3 bonds of the m-phenylene ring are slightly
D
dx.doi.org/10.1021/ic3006508 | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Figure 3. Calculated structures of the polycationic rhodium complexes 5 and 8−10 at the PCM-B3PW91/6-31G**/LANL2DZ*(Rh) level
(acetonitrile solvent). Representative atom labeling is shown on complex 8.
a
Table 1. Calculated vs Experimental Bond Distances (Å) and Angles (deg) for the Complex Ions 5, 8−10 and Complexes A−B
b
Shown in Scheme 6 (ref 19)
5
8
B
A
9
10
calcd
exptl.
calcd
calcd
calcd
calcd
2.114
calcd
2.117
Rh−C1
Rh−H1
Rh−C2
Rh−C3
Rh−C7
Rh−P1
Rh−P2
C1−C2
C1−C3
C1−H1
C4−H4
C5−H5
2.813
2.850
3.224
3.304
1.830
2.351
2.359
1.396
1.392
1.084
1.086
1.084
1.078
77.1
2.766
2.782
3.117
3.299
1.826
2.300
2.311
1.392
1.387
2.516
2.636
3.067
3.056
1.818
2.362
2.364
1.405
1.405
1.089
1.084
1.084
1.078
71.8
2.514
2.345
2.256
1.911
3.082
3.166
1.879
2.362
2.368
1.412
1.410
3.123
3.123
1.878
2.311
2.311
1.406
1.406
1.873
2.382
2.386
1.826
2.340
2.340
1.103
1.137
1.085
1.085
1.084
1.085
1.085
1.084
C6−H6
Rh−H1−C1
Rh−C1−H1
Rh−C1−C2
Rh−C1−C3
C*−C1−H1
C*−C5−H5
C1−C3−C4−C5
C1−C2−C6−C5
P1−CN2
79.4
80.9
81.3
84.1
93.9
90.9
99.0
126.8
120.6
123.6
123.6
97.8
99.8
98.5
176.5
180.0
−0.1
−1.6
1.840
1.847
176.8
179.1
−2.3
0.9
169.7
179.9
0.40
173.0
179.9
180.0
−1.73
1.32
−0.44
1.840
1.843
1.856
1.859
1.845
1.845
P2−CN2
a
b
XRD data within esd errors. Representative atom labelling as per complex 8 (Figure 3). Calculations performed at the B3PW91/6-31G**/
LANL2DZ*(Rh) level.
elongated from 1.392 Å to 1.405 Å, thus revealing some increase
in the π···Rh interaction (Table 1). The main structural change,
however, is the bending of the C1−H1 bond by about 10° out of
the plane of the m-phenylene ring. This data indicates a definite
strengthening of the Rh···C1 interaction going from 5 to 8, and
complex 8 indeed appears as a P(CH)P pincer intermediate en
route to the putative PCP pincer 10. The optimized geometry of
10 was also calculated at the same level, and then compared to
that of the experimentally obtained “open pincer” aryl complex 9
(a nonchelated version of 10). Both 9 and 10 exhibit a C2-sym-
metry optimized structure, with very similar Rh−C and Rh−P
bond lengths (Table 1). Although no XRD analysis was avail-
able for 9 and 8 (see Section 2.1), their proposed structures
are confirmed by the good agreement of the recorded 13C, 31P,
and H NMR data with the calculated ones at the B3PW91/6-
1
31+G**/LANL2DZ*(Rh) level on the static optimized geo-
metries (see the Supporting Information, Table S1). Initially, the
C−H oxidative addition was anticipated to be the likely pathway
for the formation of the PCP pincer complex 10 from 5, but since
this reaction could not be induced under many experimental
conditions, the chemical bonding of the P(CH)P complex 8 was
analyzed in detail to shed some light on this phenomenon.
The structural and NMR features of 8 are actually very
similar to those of two related rhodium complexes described by
Milstein et al. (Scheme 6):19 (i) a η2-Cipso,H agostic complex A,
characterized experimentally, and (ii) a η1-Cipso intermediate B,
E
dx.doi.org/10.1021/ic3006508 | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
(iii) a significant positive NBO atomic charge of the agostic
hydrogen atom
Scheme 6. Structures of the Rhodium Complexes A and B
Described by Milstein et al.19
(iv) a significant stabilization energy (>1.2 kcal/mol) of the
interaction estimated via second order perturbation theory.21
The computed Wiberg bond indices of 8 are closer to those
of the η1-Cipso complex B relative to those of the CH-agostic com-
plex A (Scheme 6, Table 2). A weak interaction (1.78 kcal/mol)
Table 2. Wiberg Bond Indices Calculated at the B3PW91/6-
31+G**/DGDZVP(Rh)//B3PW91/6-31G**/
LANL2DZ*(Rh) Level for 8 (Figure 3) and Complexes A−B
Shown in Scheme 6 and Described in Ref 19
evidenced during computational investigations of the C−H
oxidative addition pathway from A.
The so-called “agostic” 2c→1c* motifs refer to a T-shaped
3-center-2-electron interaction between a two-center donating
bond (2c) and a one-center accepting orbital of a metal M (1c*)
that tends toward bond scission through metal insertion or
metallacycle formation as the interaction strengthens.20 In η2-
C,H agostic complexes, the M···H distance generally ranges from
1.8 to 2.3 Å, whereas the M···H···C angle lies in the range 90−
140°.21 The structural features of complex 8 fall outside of these
ranges, but are close to those of the η1-Cipso complex B (Table 1).
This is further supported by a bonding overlap, involving a
rhodium d orbital and a π orbital of the m-phenylene ring, that is
visible in the canonical MO # 160, stabilized by 0.082 Ha (51.2
kcal/mol) below the highest occupied molecular orbital
(HOMO; Figure 4).
Rh−C1
Rh−H1
C1−H1
Rh−C7
C7−O
A
B
8
0.134
0.067
0.088
0.054
0.008
0.011
0.740
0.808
0.859
0.977
0.741
1.021
2.078
2.051
2.182
is revealed by the NBO donation σC1−H → LP*(Rh) where
LP*(Rh) refers to the empty 5s orbital of the rhodium atom, is
estimated from NBO analysis. With a population of 1.964, the
σC−H is weakly depleted, whereas the empty 5s Rh orbital
acquires a significant occupancy of 0.460. The latter value may,
however, be related to stronger donor−acceptor interactions like
πC1−C2→LP*(Rh) (5.51 kcal/mol). Finally, it is noteworthy that
the NBO atomic charge of H1 (+0.31) is comparable to that of
the other aromatic hydrogen atoms (H4−H6) (+0.30).
All the NBO characteristics of complex 8 are thus consistent
with a weak η1-C1···Rh(I) interaction suggested by both struc-
tural data and MO analysis (see above). This interaction is
formally analogous to the η1-C bonding mode reported by
Pregosin et al. in Ru(II) and Pd(II) complexes.22
Complex 8 might be considered as a stabilized intermediate
on the way to the C−Rh−H oxidative addition process, which
appears as a likely key step in the formation of the PCP “open
pincer” complex 9. This stabilization may be attributed to
the strong electrophilic character of the Rh+ center in 8, and
ultimately of the ligand 3, preventing possible back-donation
from a filled Rh d orbital into the C−H σ* orbital. The latter
overlap is found in MO # 194, lying 0.148 Ha (92.9 kcal/mol)
above the lowest unoccupied molecular orbital (LUMO;
Figure 4). Electron enrichment of the metal center is expected
to lower the level of this empty orbital, and might thus be invoked
as the driving force for the formation of 9 when methanol is used
as a solvent. In accordance with the mechanism proposed for
nucleophile-induced [P−CN2]+ bond cleavage,10c methanol
molecules could indeed trigger such a cleavage in the highly
electrophilic complex 8. The latter would release a methyl
diphenylphosphinite ligand, thus inducing an increase of the
electronic density at the Rh(I) center, and favoring the C−H
oxidative addition step (Scheme 7).
The electronic enrichment was first computationally envisaged
through the coordination of the possibly released Ph2P(OMe)
ligand in the putative complex 11 (Scheme 7). The correspond-
ing two-component ligand “Ph2P(OMe) + monodentate
imidazolium-imidazoliophosphine 3′” derived from the meth-
anolysis of 3 (Scheme 7) is indeed a priori globally more donat-
ing than the parent ligand 3 (acting in a pseudotridentate manner
through the weak η1-C coordination). This was indeed checked
by calculation at the PBE/6-31G**/LANL2DZ*(Rh) level) of
the average IR CO stretching frequencies in the corresponding
[RhL2(CO)2]+ complexes serving as probes in a general scale of
the net electron-donating character of cis-coordinating ligands
Figure 4. Selected molecular orbitals (MOs) of complex 8 featuring
both the π-bonding overlap (MO # 160), and the dRh-σ*C−H preoxida-
tive addition overlap (MO # 194). Calculations performed at the PCM-
B3PW91/6-31G**/LANL2DZ*(Rh) level (acetonitrile solvent). En-
ergies are given in hartree.
Natural Bond Orbital (NBO) analysis associated to agostic
η2-C,H interactions include the following main features:
(i) a noticeably depleted σC−H NBO, and a weak occupancy of
the empty accepting metal orbital
(ii) a significant bond order for the M−H and M−C bonds as
measured by Wiberg bond indices
F
dx.doi.org/10.1021/ic3006508 | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Scheme 7. Auxiliary Calculated Structures (See Text)
L2:15b
v
CO([Rh(3)(CO)2]+) = 2080.2 cm−1 > vCO([Rh-
Figure S1. With respect to 11 and 8, the Ph2P(OMe)-Rh-
enriched adduct 11a was thus found more stabilized than 8a,
both of them exhibiting square-pyramidal structures (18.4
kcal/mol vs 25.3 kcal/mol). Conversely, the kinetic barrier for the
transformation 11→11a (32.0 kcal/mol) was found to be higher than
the corresponding barrier for 8→8a (29.7 kcal/mol), and in accord-
ance with the Hammond postulate, 8aTS is a later transition
state (C1···H1 = 1.555 Å) than 11aTS (C1···H1 = 1.519 Å).
The reaction barrier is thus increased upon electronic enrich-
ment of the Rh center by the Ph2P(OMe) ligand in 11, while
preserving the dicationic character of the ligand 3 of 8 in the
ligand 3′ containing the C−H bond to be activated in 11.
In spite of the protic experimental conditions (methanol is
used as the solvent), the methanolysis of the trication 8 might
proceed sequentially through a transient dicationic complex 12
of the ligand 3′′ (Scheme 7) bearing a pendent imidazolylidene
moiety that could deprotonate the CH vertex either directly or
via the C−H oxidative adduct 12a. The latter is found to be
relatively more stabilized (by 17.8 kcal/mol vs 12) than 8a (25.3
kcal/mol vs 8), but less than 11a (18.4 kcal/mol vs 11, Figure 5).
The corresponding kinetic barrier through the transition state
12aTS was however found slightly higher (30.8 kcal/mol) than
that of the 8→8a process. The 12→12a barrier is also found to
be lower than the 11→11a barrier, in accordance with the
concomitant electronic enrichment of the monocationic ligand
3′′ with respect to the dicationic ligand 3′.
̃
̃
(3′)(Ph2P(OMe))(CO)2]+) = 2074.4 cm−1 > vCO([Rh(Ph2P-
̃
(OMe))2(CO)2]+) = 2072.6 cm−1.
The Rh−C distance in complex 11 (4.031 Å) indicates the
removal of the Rh−C interaction occurring in complex 8 (2.516 Å)
that was a priori expected to facilitate the oxidative addition
process. The energy profiles of the oxidative addition pathways
from 8 and 11 to their respective C−H oxidative adducts 8a and
11a (Scheme 7) are shown in Figure 5, and the calculated
The effect of electron enrichment was utimately investigated
in the [Rh(CO)(PMe3)2(C6H6)]+ model complex where the
dicationic diphosphine 3 is substituted by two strongly electron-
donating PMe3 ligands, but instead of the C1−H1 activa-
tion, only an η2-arene Rh(I) complex was obtained (Supporting
Information, Figure S2). These findings are therefore not in favor
of a C−H oxidative addition pathway for the formation of the
“open pincer” complex 9.
Keeping in mind that the medium of the formation of 9 is very
weakly basic, two alternative mechanisms could be envisaged
from the η1-C rhodium complex 8. The reacting MeOH mole-
cule may indeed first coordinate to the Rh center (Scheme 8, route a)
Figure 5. Oxidative addition energy profiles from the complexes 8
(in black), 11 (in blue), and 12 (in red) to the respective C−H oxidative
adducts 8a, 11a, and 12a (Scheme 7). Calculations performed at the
PCM-B3PW91/6-31G**/LANL2DZ*(Rh) level (acetonitrile solvent).
Relative energies, given in kcal/mol, are not corrected for zero-point
energy.
structures of the corresponding transition states and oxidative
addition products are given in the Supporting Information,
G
dx.doi.org/10.1021/ic3006508 | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Scheme 8. Possible Mechanisms for the Formation of the “Open Pincer” Complex 9 from the Precursor 8 in the Presence of
Methanol, Consistent with the Absence of Strong Base
were used as received from commercial sources. All reactions were
carried out under an argon atmosphere, using Schlenk and vacuum line
techniques. The following analytical instruments were used. 1H, 13C, and
31P NMR: Bruker ARX 250, DPX 300, and Avance 500. NMR chemical
shifts δ are in ppm, with positive values to high frequency relative to the
tetramethylsilane reference for 1H and 13C, and to the H3PO4 reference
for 31P; coupling constants J are in hertz (Hz). 103Rh chemical shifts are
given to high frequency from Ξ(103Rh) = 3.16 MHz. Elemental analyses:
Perkin-Elmer CHN 2400.
or add to the central phenylene ring by nucleophilic attack on
the activated α-positions of the C1 carbon atom (Scheme 8,
route b; Mulliken charges at C2 and C3: + 0.12. See Figure 3
for labeling). In both cases, the resulting highly acidic oxonium
would then be deprotonated by the weakly basic TfO− anion. In
route a, an α-elimination process (in the “organometallic sense”)
would ensue from the proximity of the Rh-coordinated methoxy
oxygen atom and the phenylene H1-atom, thus leading to formation
of the “open pincer” 9.23 In route b, β-elimination of methanol (in the
“organic sense”) would be dictated by the phenylene rearomatization
driving force.
2-(Diphenylphosphanyl)-1-{3-[2-(diphenylphosphanyl)-1H-
imidazol-1-yl]phenyl}-1H-imidazole (2). To a solution of 1,3-bis(N-
imidazolyl)benzene 1 (500 mg, 2.38 mmol) in THF (50 mL) cooled
to −78 °C was added n-butyllithium (2.5 M in hexane, 2.0 mL, 5.0 mmol).
The suspension was then warmed to −20 °C and stirred for 2 h. After
addition of chlorodiphenylphosphine (0.93 mL, 5.0 mmol) at −20 °C,
the solution was stirred at room temperature for 12 h. After evaporation
of the solvent under vacuum, purification by chromatography on silica
gel (CH2Cl2/pentane) of the residue gave 2 as a white solid (1.15 g,
83%). Recrystallization at −20 °C from CH2Cl2/toluene afforded 2 as
white crystals (mp 58−59 °C).
3. CONCLUSION
The use of the chelating and highly electron-poor bis-
(imidazoliophosphine) ligand 3 allowed to generate the Rh(I)
complexes 5, 8, and 9, that exhibit unprecedented structural
features. Complex 8 appears as a η1-C P(CH)P pincer inter-
mediate that is poised to form the putative PCP pincer complex
10. Although the nature of the bonding in complex 8, calling for
the contribution of a C-hypervalent Lewis structure in Scheme 5,
deserves further theoretical investigations, the long debated
C−H oxidative pathway in the formation of classical PCP pincer
complexes could be ruled out in the present case on the basis of
DFT calculation results. The inescapable findings of the com-
putational investigations suggest two alternative mechanisms
based on classical organometallic or organic key-processes (α-
elimination and aromaticity-driven β-elimination) and con-
sistent with the absence of strong base in the medium. Through
the formation of the aryl-Rh “open pincer” complex 9, the
electrophilic reactivity of the P−CN2 bond in the imidazolio-
phosphine ligands is shown to open new perspectives in the
investigation of selective reorganization processes within the
coordination sphere of transition metals such as Rh(I) centers.
Given the now demonstrated unique coordination properties for
the ortho- and meta-phenylene-bis(imidazoliophosphine) ligands,
efforts towards the comparative study of the para-phenylene-
bis(imidazoliophosphine) isomers deserve to be undertaken shortly.
1H NMR (CDCl3, 25 °C): δ = 7.28−7.50 (m, 25H, Har), 7.18 (brs,
2H, Har), 7.10 (brs, 1H, Har). 13C NMR (CDCl3, 25 °C): δ = 146.9 (d,
JCP = 8.6, C), 138.4 (s, C), 135.0 (d, JCP = 5.1, C), 133.9 (d, JCP = 20.8,
CHar), 131.6 (s, CHar), 129.6 (s, CHar), 129.3 (s, CHar), 128.6 (d, JCP
=
7.9, CHar), 128.2 (d, JCP = 4.8, CHar), 124.2 (t, JCP = 3.3, CHar), 123.6 (s,
CHar). 31P NMR (CDCl3, 25 °C): δ = −28.4. MS (ESI+): m/z = 579.2
[M + H]+; HRMS (ESI+): calcd. for C36H29N4P2 579.1867; found,
579.1880. Elemental analysis, calcd (%) for C36H28N4P2: C 74.73, H
4.88, N 9.68; found: C 73.93, H 4.98, N 9.29.
2-(Diphenylphosphanyl)-1-{3-[2-(diphenylphosphanyl)-3-
methyl-1H-imidazol-3-ium-1 yl]phenyl}-3-methyl-1H-imidazol-
3-ium bistriflate (3). To a solution of 2 (1.10 g, 1.90 mmol) in CH2Cl2
(40 mL) cooled at −78 °C was added methyl-trifluoromethanesulfonate
(0.42 mL, 3.80 mmol). The solution was warmed to room temperature
and stirred for 2 h. After evaporation of the solvent under vacuum, the
residue was washed with Et2O (40 mL), affording a white solid (1.64 g,
95%). Recrystallization at −20 °C from MeCN/Et2O afforded 3 as white
crystals (mp: 56−57 °C).
1H NMR (CD3CN, 25 °C): δ = 7.83 (brs, 2H, Har), 7.72 (brs, 2H,
Har), 7.52−7.30 (m, 24H, Har), 3.52 (s, 6H, NCH3). 13C NMR
(CD3CN, 25 °C): δ = 144.7 (d, JCP = 58.3, C), 135.8 (s, C), 133.4 (d,
JCP = 21.6, CHar), 131.0 (s, CHar), 129.8 (d, JCP = 7.8, CHar), 129.5 (s,
CHar), 129.0 (s, CHar), 128.3 (s, CHar), 127.8 (s, C), 126.8 (s, CHar),
125.8 (t, JCP = 2.6, CHar), 125.3 (s, CHar), 121.2 (q, JCF = 320.8,
CF3SO3), 37.8 (CH3). 31P NMR (CD3CN, 25 °C):δ = −21.1. MS(ES+):
EXPERIMENTAL SECTION
General Remarks. THF, diethyl ether, and toluene were dried and
distilled over sodium/benzophenone, while pentane, dichloromethane,
acetonitrile, and methanol were dried over CaH2. All other reagents
■
−
m/z: 757.2 [M − CF3SO3 ]+; HRMS(ES+) calcd for C39H34N4O3F3P2S,
757.1779; found, 757.1785. Elemental analysis, calcd (%) for
H
dx.doi.org/10.1021/ic3006508 | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
C40H34F6N4O6P2S2: C 52.98, H 3.78, N 6.18; found: C 51.92, H 3.83,
N 5.85.
(t, JCP = 23.9, C), 133.5 (t, JCP = 23.9, C), 132.7 (s, CHar), 132.0 (s,
CHar), 131.9 (s, CHar), 131.7 (s, CHar), 131.6 (d, JCP = 7.5, CHar), 131.1
(t, JCP = 6.3, CHar), 130.8 (s, CHar), 129.1 (t, JCP = 5.0, CHar), 128.9 (t,
JCP = 5.0, CHar), 128.8 (s, CHar), 124.9 (s, CHar), 124.8 (s, CHar), 124.5
(s, CHar), 123.3 (s, CHar), 123.2 (s, CHar), 123.1 (s, CHar), 121.6 (s,
CHar), 121.4 (s, CHar), 120.8 (q, JCF = 321.8, CF3SO3), 55.0 (s, CH3O),
36.8 (s, CH3N), 36.7 (s, CH3N). 31P NMR (CD2Cl2, 25 °C):δ = +124.9
(d, JPRh = 143.8). 103Rh NMR (CD3CN, 25 °C): δ = −255 ppm.
Rhodium(I) Chloride Complexes of the Ligand 3 (4, 5). A
mixture of [RhCl(cod)]2 (163 mg, 0.33 mmol) and dication 3 (500 mg,
0.55 mmol) was dissolved in CH2Cl2 (20 mL) at −40 °C, and stirred at
room temperature for 12 h. After evaporation of the solvent, the crude
residue was washed with Et2O (30 mL) affording 4 as a yellow powder
(572 mg, 90%; mp 216−218 °C). Then complex 4 was dissolved in
CH2Cl2 (10 mL), and carbon monoxide was bubbled for 30 min at room
temperature. After evaporation of the solvent, and washing with
Et2O (20 mL), 5 was obtained as a yellow powder (404 mg, 76%).
Recrystallization from MeCN/Et2O at 0 °C afforded 5 as orange crystals
(mp 112−113 °C).
−
MS(ES+): m/z 951.1 [M − CF3SO3 ]+; HRMS(ES+) calcd for
C42H41N4O6F3P2SRh, 951.1229; found, 951.1255. Elemental analysis,
calcd (%) for C43H41F6N4O9P2RhS2: C 46.92, H 3.75, N 5.09; found: C
45.62, H 3.27, N 5.07.
Crystal Structure Determination of Compounds 2, 3, and 5.
Intensity data for compounds 2, 3, and 5 were collected at 180 K on an
Apex2 Bruker diffractometer using a graphite-monochromated Mo Kα
radiation source and equipped with an Oxford Cryosystems Cryostream
Cooler Device. Structures were solved using SIR9224 or SUPERFLIP,25
and refined by full-matrix least-squares procedures on F using the
programs of the PC version of CRYSTALS.26 Atomic scattering factors
were taken from the International tables for X-ray Crystallography.27
All non-hydrogen atoms were refined anisotropically. Hydrogen atoms
were refined using a riding model. Absorption corrections were
introduced using the program MULTISCAN.28 For 3, the crystal was
of poor quality (a phenyl group is disordered). Refinement of the used
model led to a satisfactory solution with a 0.5:0.5 occupancy ratio. It was
not possible to resolve diffuse electron-density residuals (enclosed
solvent molecules). Finally, treatment with the SQUEEZE29 facility
from PLATON30 resulted in a better refinement.
RhCl(cod)[3] Complex (4). 1H NMR (CDCl3, 25 °C): δ = 7.77−
7.16 (m, 28H, Har), 5.70 (br, 2H, CHcod), 4.55 (br, 2H, CHcod), 3.71 (s,
6H, NCH3), 2.43 (brs, 4H, CH2cod), 2.14 (brs, 4H, CH2cod). 13C NMR
(CDCl3, 25 °C): δ = 140.9 (d, JCP = 23.6, C), 135.3 (s, C), 132.9 (s,
CHar), 130.1 (s, CHar), 129.8 (s, CHar), 128.8−124.5 (m, C, CHar),
120.7 (q, JCF = 320.8, CF3SO3), 107.8 (s, CHcod), 73.0 (br, CHcod), 39.5
(CH3), 32.8 (s, CH2cod), 28.7 (s, CH2cod). 31P NMR (CDCl3, 25 °C):
δ = +26.4 (d, JPRh = 150.4). 103Rh NMR (CDCl3, 25 °C): δ = +379
−
ppm. MS(ES+): m/z 1003.1 [M-CF3SO3 ]+; HRMS(ES+) calcd for
C47H46ClN4O3F3P2SRh: 1003.1462; found: 1003.1458.
RhCl(CO)[3] Complex (5). 1H NMR (CD3CN, 25 °C): δ = 8.70 (s,
1H, Har), 8.10 (d, 2H, JHH = 5.0, Har), 7.87 (d, 2H, JHH = 5.0, Har), 7.83−
7.42 (m, 23H, Har), 3.30 (s, 6H, NCH3). 13C NMR (CD3CN, 25 °C):
δ = 182.9 (td, JCRh = 74.2, JCP = 15.1, CO), 140.6 (t, J = 13.8, C), 135.4
(s, CHar), 134.8 (s, C), 133.0 (s, CHar), 132.7 (d, JCP = 6.3, CHar), 132.6
(d, JCP = 6.3, CHar), 130.0 (s, CHar), 129.8 (t, JCP = 5.0, CHar), 129.6 (s,
CHar), 129.3 (t, JCP = 5.0, CHar), 129.1 (s, CHar), 127.7 (s, CHar), 126.7
Crystal data for 2: C36H28N4P2, M = 578.59 g·mol−1, Monoclinic, a =
10.1714(8), b = 20.5363(13), c = 15.1935(9) Å, β = 109.556(3)°, V =
2990.6(4) Å3, T = 180 K, space group P21/c, Z = 4, μ(Mo-Kα) = 0.178
mm−1, 33195 reflections measured, 7769 unique (Rint =0.043), 5067
reflections used in the calculations [I > 3σ(I)], 379 parameters, R1 =
0.0418, wR2 = 0.0446.
(t, JCP = 25.8, C), 124.8 (s, CHar), 122.9 (t, JCP = 23.2 C), 121.1 (q, JCF
=
320.8, CF3SO3), 39.2 (CH3). 31P NMR (CD3CN, 25 °C): δ = +18.7
(d, JPRh = 135.7). 103Rh NMR (CD3CN, 25 °C): δ = −75 ppm.
MS(ES+): m/z 923.0 [M − CF3SO3−]+; HRMS(ES+) calcd for
C40H34N4O4F3P2SClRh, 923.0472; found, 923.0491. Elemental
analysis, calcd (%) for C41H34ClF6N4O7P2RhS2: C 45.89, H 3.19,
N 5.22; found: C 45.00, H 3.20, N 5.10.
Crystal data for 3: C40H34F6N4O6P2S2, M = 906.80 g·mol−1, mono-
clinic, a = 15.633(3), b = 22.942(5), c = 14.911(3) Å, β = 116.15(3)°,
V = 4800(2) Å3, T = 180 K, space group C2/c, Z = 4, μ(Mo-Kα) = 0.246
mm−1, 22578 reflections measured, 7462 unique (Rint =0.0659), 4905
reflections used in the calculations [I > 3σ(I)], 266 parameters, R1 =
0.1134, wR2 = 0.1073.
Rhodium(I)carbonyl Triflate Complex of the Ligand 3 (8). A
mixture of 5 (280 mg, 0.26 mmol) and silver trifluoromethanesulfonate
(74 mg, 0.29 mmol) cooled to −20 °C, was dissolved in MeCN
(10 mL). After warming up to room temperature, the solution was
stirred for 12 h. The solid was then filtered off, and the solution
evaporated under vacuum. After addition of THF (10 mL) and filtration
of the remaining solid, the solvent was removed under vacuum, affording
8 as a yellow solid (291 mg, 94%).
Crystal data for 5: C41H34ClF6N4O7P2RhS2, M = 1073.17 g·mol−1,
triclinic, a = 12.7519(12), b = 13.7106(13), c = 13.8297(13) Å, α =
77.390(4), β = 79.713(4), γ = 68.869(4)°, V = 2187.7(4) Å3, T = 180 K,
space group P1, Z = 2, μ(Mo-Kα) = 0.700 mm−1, 55848 reflections
̅
measured, 13792 unique (Rint = 0.021), 11019 reflections used in the
1H NMR (CD3CN, 25 °C): δ = 9.81 (br, 1H, Har), 8.26 (s, 2H, Har),
8.05 (s, 2H, Har), 7.77−7.38 (m, 23H, Har), 3.28 (s, 6H, NCH3). 13C
NMR (CD3CN, 25 °C): δ = 187.7 (d, JCRh = 70.4, CO), 138.5 (peudo-t,
J = 17.6, C), 135.9 (d, J = 46.5, C), 135.5 (s, C), 133.4 (s, CHar), 133.1 (s,
CHar), 132.0 (s, CHar), 131.7 (s, CHar), 130.8 (s, CHar), 130.4−130.3
(m, CHar), 129.5 (s, CHar), 124.3 (br, C), 121.0 (q, JCF = 320.8,
CF3SO3), 120.8 (br, C), 111.4 (s, CHar), 39.4 (CH3). 31P NMR
(CD3CN, 25 °C):δ = +12.4 (d, JPRh = 125.6). 103Rh NMR (CD3CN,
calculations [I > 3σ(I)], 577 parameters, R1 = 0.0299, wR2 = 0.0306.
COMPUTATIONAL DETAILS
■
Geometries were fully optimized at the PCM-B3PW91/6-31G**/
LANL2DZ*(Rh) level of calculation using Gaussian09.31 LANL2DZ*-
(Rh) means that f-polarization functions derived by Ehlers et al.32 for Rh
have been added to the LANL2DZ(Rh) basis set. Vibrational analysis
was performed at the same level as the geometry optimization. Solvent
effects were included using the polarizable continuum model (PCM)
implemented in Gaussian09 for acetonitrile (ε = 35.688). From these
optimized structures, the potential energy surface was explored along
the reaction coordinate (C1−H1 distance) at the same calculation level.
The intrinsic reaction coordinate was followed using the IRC technique
implemented in Gaussian 09 for all located transition states.
−
25 °C): δ = −138 ppm. MS(ES+):m/z 1037.0 [M − CF3SO3 ]+;
HRMS(ES+) calcd for C41H34N4O7F6P2RhS2, 1037.0303; found,
1037.0319. Elemental analysis, calcd (%) for C42H34F9N4O10P2RhS3:
C 42.51, H 2.89, N 4.72; found: C 37.90, H 2.84, N 4.76 (the low value
found for the carbon element might be due to residual silver).
Rhodium(I)carbonyl Aryl Complex 9. Complex 8 (80 mg, 0.067
mmol) was dissolved in dry methanol (2 mL). After stirring overnight
at room temperature, the solvent was then removed under vacuum.
After washing by Et2O (10 mL), 9 was obtained as a brown solid
(67 mg, 90%).
The magnetic shielding tensor was calculated at the B3PW91/6-
31+G**/LANL2DZ*(Rh) using the GIAO (gauge-independent atomic
orbital) method implemented in Gaussian09.31 The 31P NMR chemical
shifts were estimated with respect to the usual H3PO4 reference.
1H NMR (CD2Cl2, 25 °C): 9.59 (s, 1H, CH), 9.51 (s, 1H, CH), 8.35
(d, 1H, JHH = 5.0, Har), 8.11 (s, 1H, Har), 7.93 (brs, 2H, Har), 7.82−7.71
(m, 4H, Har), 7.65−7.63 (m, 3H, Har), 7.59−7.41 (m, 15H, Har), 6.84 (d,
1H, JHH = 1.9, Har), 4.13 (s, 3H, CH3N), 4.09 (s, 3H, CH3N), 3.47 (s,
NBO analysis33 was performed using NBO 5.G.34 In NBO analysis,
deviations from idealized Lewis structures are interpreted in terms of
donor−acceptor interactions between filled localized bonds or lone-
pairs (donors) and empty antibonds (acceptors). They are referred to as
“delocalization” corrections to the zeroth-order natural Lewis structure
and estimated by the second-order perturbation theory. The stabilization
6H, CH3O). 13C NMR (CD2Cl2, 25 °C): δ = 190.7 (td, JCRh = 57.8, JCP
=
16.3, CO), 177.0 (td, JCRh = 44.0, JCP = 17.6, C), 140.0 (s, C), 135.9 (s,
C), 135.8 (broad s, N2CH), 135.4 (broad s, N2CH), 135.3 (s, C), 133.6
I
dx.doi.org/10.1021/ic3006508 | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
energy E(2) associated with delocalization is given by E(2) = n ⟨Fij⟩2/εij, where
Fij is the off-diagonal NBO Fock matrix element (and can be taken as
proportional to the overlap between those orbitals), εij is the energy
difference between the donor NBO i and the acceptor NBO j, and n is
the occupancy of the NBO i. Since perturbation theory is valid for small
perturbations only, absolute stabilizations become less reliable as their
value increase. General trends are therefore more meaningfull than the
absolute values of the energetic contributions. Molecular orbitals were
plotted using the GABEDIT program.35
(9) (a) Scholl, M.; Ding, S.; Lee, C. W.; Grubbs, R. H. Org. Lett. 1999,
1, 953. (b) Debono, N.; Canac, Y.; Duhayon, C.; Chauvin, R. Eur. J.
Inorg. Chem. 2008, 2991. (c) Abdellah, I.; Debono, N.; Canac, Y.;
Vendier, L.; Chauvin, R. Chem.Asian. J. 2010, 5, 1225.
(10) (a) Kuhn, N.; Fahl, J.; Blaser, D.; Boese, R. Z. Anorg. Allg. Chem.
̈
1999, 625, 729. (b) Azouri, M.; Andrieu, J.; Picquet, M.; Richard, P.;
Hanquet, B.; Tkatchenko, I. Eur. J. Inorg. Chem. 2007, 4877.
(c) Abdellah, I.; Lepetit, C.; Canac, Y.; Duhayon, C.; Chauvin, R.
Chem.Eur. J. 2010, 16, 13095. (d) Abdellah, I.; Boggio-Pasua, M.;
Canac, Y.; Lepetit, C.; Duhayon, C.; Chauvin, R. Chem.Eur. J. 2011,
17, 5110. (e) Maaliki, C.; Lepetit, C.; Canac, Y.; Bijani, C.; Duhayon, C.;
Chauvin, R. Chem.Eur. J. 2012, 18, 7705.
ASSOCIATED CONTENT
■
S
* Supporting Information
Complementary calculation details are available from the
authors. Crystallographic data is provided in CIF format. H,
(11) For a recent review: Canac, Y.; Maaliki, C.; Abdellah, I.; Chauvin,
R. New. J. Chem. 2012, 36, 17.
1
(12) (a) Canac, Y.; Debono, N.; Vendier, L.; Chauvin, R. Inorg. Chem.
2009, 48, 5562. (b) Canac, Y.; Debono, N.; Lepetit, C.; Duhayon, C.;
Chauvin, R. Inorg. Chem. 2011, 50, 10810.
and 31P NMR spectra are also provided. This material is available
free of charge via the Internet at http://pubs.acs.org.
(13) Vargas, V. C.; Rubio, R. J.; Hollis, T. K.; Salcido, M. E. Org. Lett.
2003, 5, 4847.
AUTHOR INFORMATION
■
(14) CCDC 873530 (2), CCDC 873528 (3), and CCDC 873529 (5)
contain the supplementary crystallographic data for this paper. These
data can be obtained free of charge from the Cambridge Crystallo-
graphic Data Centre via www.ccdc.cam.ac.uk/data_request/cif.
(15) (a) Chianese, A. R.; Li, X.; Janzen, M. C.; Faller, J. W.; Crabtree, R.
H. Organometallics 2003, 22, 1663. (b) Canac, Y.; Lepetit, C.; Abdalilah,
M.; Duhayon, C.; Chauvin, R. J. Am. Chem. Soc. 2008, 130, 8406.
(16) Wu, M. L.; Desmond, M. J.; Drago, R. S. Inorg. Chem. 1979, 18,
679.
Corresponding Author
*Fax: (+33)5 61 55 30 03. E-mail: yves.canac@lcc-toulouse.fr
(Y.C.), chauvin@lcc-toulouse.fr (R.C.).
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
(17) Andrieu, J.; Azouri, M.; Richard, P. Inorg. Chem. Commun. 2008,
11, 1401.
In addition to thanking the Minister
de la Recherche et de la Technologie and the Universite
Sabatier, the authors thank also the Centre National de la Recherche
Scientifique, and the ANR program (ANR-08-JCJC-0137-01).
The theoretical studies were performed using HPC resources
from CALMIP (Grant 2011 [0851]), and from GENCI-[CINES/
IDRIS] (Grant 2011 [085008]).
̀
e de l’Enseignement Super
́
ieur
́
Paul-
(18) http://www.ccdc.cam.ac.uk/products/csd/radii/table.php4.
(19) (a) Montang, M.; Schwartsburd, L.; Cohen, R.; Leitus, G.; Ben-
David, Y.; Martin, J. M. L.; Milstein, D. Angew. Chem., Int. Ed. 2007, 46,
1901. (b) Montang, M.; Efremenko, I.; Cohen, R.; Shimon, L. J. W.;
Leitus, G.; Diskin-Posner, Y.; Ben-David, Y.; Salem, H.; Martin, J. M. L.;
Milstein, D. Chem.Eur. J. 2010, 16, 328.
(20) Weinhold, F.; Landis, C. R. In Valency and Bonding: A Natural
Bond Orbital Donor-Acceptor Perspective; Cambridge University Press:
Cambridge, U.K., 2005; pp 480.
REFERENCES
■
(1) (a) Albrecht, M.; van Koten, G. Angew. Chem., Int. Ed. 2001, 40,
3750. (b) Lagt, M. Q.; van Zwieten, D. A. P.; Moerkerk, A. J. C. M.;
Gebbink, R. J. M. K.; van Koten, G. Coord. Chem. Rev. 2004, 248, 2275.
(c) Nishiyama, H. Chem. Soc. Rev. 2007, 36, 1133. (d) Morales-Morales,
D.; Jensen, C. M. The Chemistry of Pincer Compounds; Elsevier:
Amsterdam, The Netherlands, 2007. (e) Morales-Morales, D. Mini-Rev.
Org. Chem. 2008, 5, 141. (f) Benito-Garagorri, D.; Kirchner, K. Acc.
(21) Saßmannhausen, J. Dalton Trans. 2011, 40, 136.
(22) Pregosin, P. S. Chem. Commun. 2008, 4875.
(23) Wood, C. D.; McLain, S. J.; Schrock, R. R. J. Am. Chem. Soc. 1979,
101, 3210.
(24) Altomare, A.; Cascarano, G.; Giacovazzo, C.; Guagliardi, A.; Burla,
M. C.; Polidori, G.; Camalli, M. J. Appl. Crystallogr. 1994, 27, 435.
(25) Palatinus, L.; Chapuis, G. J. Appl. Crystallogr. 2007, 40, 786.
(26) Betteridge, P. W.; Carruthers, J. R.; Cooper, R. I.; Prout, K.;
Watkin, D. J. J. Appl. Crystallogr. 2003, 36, 1487.
(27) International Tables for X-ray Crystallography; Kynoch Press:
Birmingham, England, 1974; Vol. IV.
(28) Blessing, R. H. Acta Crystallogr. 1995, A51, 33.
(29) Sluis, P. v. d.; Spek, A. L. Acta Crystallogr. 1990, A46, 194.
Chem. Res. 2008, 41, 201. (g) Selander, N.; Szabo,
111, 2048.
́
K. J. Chem. Rev. 2011,
(2) (a) Rybtchinski, B.; Milstein, D. Angew. Chem., Int. Ed. 1999, 28,
870. (b) van der Boom, M. E.; Milstein, D. Chem. Rev. 2003, 103, 1759.
(3) (a) Adams, J. J.; Lau, A.; Arulsamy, N.; Roddick, D. M. Inorg. Chem.
2007, 46, 11328. (b) Adams, J. J.; Lau, A.; Arulsamy, N.; Roddick, D. M.
Organometallics 2011, 30, 689. (c) Adams, J. J.; Arulsamy, N.; Roddick,
D. M. Organometallics 2011, 30, 697.
(30) Spek, A. L. Acta Crystallogr. 2009, D65, 148.
(31) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb,
M. A.; Cheeseman, J. R.; Scalmani, G.; Barone, V.; Mennucci, B.;
Petersson, G. A.; Nakatsuji, H.; Caricato, M.; Li, X.; Hratchian, H. P.;
Izmaylov, A. F.; Bloino, J.; Zheng, G.; Sonnenberg, J. L.; Hada, M.;
Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa, J.; Ishida, M.; Nakajima,
T.; Honda, Y.; Kitao, O.; Nakai, H.; Vreven, T.; , Jr., Montgomery, J. A.;
Peralta, J. E.; Ogliaro, F.; Bearpark, M.; Heyd, J. J.; Brothers, E.; Kudin,
K. N.; Staroverov, V. N.; Kobayashi, R.; Normand, J.; Raghavachari, K.;
Rendell, A.; Burant, J. C.; Iyengar, S. S.; Tomasi, J.; Cossi, M.; Rega, N.;
Millam, N. J.; Klene, M.; Knox, J. E.; Cross, J. B.; Bakken, V.; Adamo, C.;
Jaramillo, J.; Gomperts, R.; Stratmann, R. E.; Yazyev, O.; Austin, A. J.;
Cammi, R.; Pomelli, C.; Ochterski, J. W.; Martin, R. L.; Morokuma, K.;
Zakrzewski, V. G.; Voth, G. A.; Salvador, P.; Dannenberg, J. J.; Dapprich,
S.; Daniels, A. D.; Farkas, O.; Foresman, J. B.; Ortiz, J. V.; Cioslowski, J.;
and Fox, D. J. Gaussian 09, Revision A.1; Gaussian, Inc.: Wallingford,
CT, 2009.
(4) Chase, P. A.; Gagliardo, M.; Lutz, M.; Spek, A. L.; van Klink, G. P.
M.; van Koten, G. Organometallics 2005, 24, 2016.
(5) (a) Kossoy, E.; Iron, M. A.; Rybtchinski, B.; Yehoshoa, B. D.;
Shimon, L. J. W.; Konstantinovski, L.; Martin, J. M. L.; Milstein, D.
Chem.Eur. J. 2005, 11, 2319. (b) Kossoy, E.; Rybtchinski, B.; Diskin-
Posner, Y.; Shimon, L. J. W.; Leitus, G.; Milstein, D. Organometallics
2009, 28, 523.
(6) (a) Rubio, M.; Suar
Pizzano, A. Dalton Trans. 2007, 407. (b) Rubio, M.; Suar
D.; Galindo, A.; Alvarez, E.; Pizzano, A. Organometallics 2009, 28, 547.
(7) (a) Li, J.; Lutz, M.; Spek, A. L.; van Klink, G. P. M.; van Koten, G.;
Gebbink, R. J. M. Organometallics 2010, 29, 1379. (b) Niu, J. L.; Hao, X.
Q.; Gong, J. F.; Song, M. P. Dalton Trans. 2011, 40, 5135.
́
́
ez, A.; del Río, D.; Galindo, A.; Alvarez, E.;
́
ez, A.; del Río,
́
́
(8) (a) Morales-Morales, D.; Redon, R.; Yung, C.; Jensen, C. M. Chem.
Commun. 2000, 1619. (b) Salah, A. B.; Zargarian, D. Dalton Trans. 2011,
40, 8977.
J
dx.doi.org/10.1021/ic3006508 | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
(32) Ehlers, A. W.; Bohme, M.; Dapprich, S.; Gobbi, A.; Hollwarth, A.;
̈
̈
Jonas, V.; Kohler, K. F.; Stegmann, R.; Veldkamp, A.; Frenking, G. Chem.
̈
Phys. Lett. 1993, 208, 111.
(33) (a) Reed, E. A.; Weinstock, R. B.; Weinhold, F. J. Chem. Phys.
1985, 83, 735. (b) Reed, E. A.; Weinhold, F. J. Chem. Phys. 1985, 83,
1736. (c) Reed, E. A.; Curtiss, L. A.; Weinhold, F. Chem. Rev. 1988, 88,
899.
(34) Glendening, E. D.; Badenhoop, J. K.; Reed, A. E.; Carpenter, J. E.;
Bohmann, J. A.; Morales, C. M.; Weinhold, F. NGO 5.G; Theoretical
Chemistry Institute, University of Wisconsin: Madison, WI, 2001;
http://www.chem.wisc.edu/∼nbo5
(35) GABEDIT; http://sites.google.com/site/allouchear/Home/
gabedit.
K
dx.doi.org/10.1021/ic3006508 | Inorg. Chem. XXXX, XXX, XXX−XXX