Rollover cyclometalation with 2-(2′-pyridyl)quinoline

Article abstract of DOI:10.1021/ic400908f

Rollover cyclometalation of 2-(2′-pyridyl)quinoline, L, allowed the synthesis of the family of complexes [Pt(L-H)(X)(L′)] and [Pt(L)(X)(L′)][BF₄] (X = Me, Cl; L′ = neutral ligand), the former being the first examples of Pt(II) rollover complexes derived from the ligand L. The ligand L* is a C,N cyclometalated, N-protonated isomer of L, and can also be described as an abnormal-remote pyridylene. The corresponding [Pt(L-H)(Me)(L′)]/[Pt(L)(Me)(L′)]⁺ complexes constitute an uncommon Bronsted-Lowry acid-base conjugated couple. The species obtained were investigated in depth through NMR and UV-vis spectroscopy, cyclic voltammetry, and density functional theory (DFT) methods to correlate different chemico-physical properties with the nature of the cyclometalated ligand (e.g., L vs bipy or L* vs L) and of the neutral ligand (DMSO, CO, PPh₃). The crystal structures of [Pt(L-H)(Me)(PPh₃)], [Pt(L-H)(Me)(CO)] and [Pt(L)(Me)(CO)][BF ₄] were determined by X-ray powder diffraction methods, the latter being the first structure of a Pt(II)-based, protonated, rollover complex to be unraveled. The isomerization of [Pt(L)(Me)(PPh₃)]⁺ in solution proceeds through a retro-rollover process to give the corresponding adduct [Pt(L)(Me)(PPh₃)]⁺, where L acts as a classical N,N chelating ligand. Notably, the retro-rollover reaction is the first process, among the plethora of Pt-C bond protonolysis reactions reported in the literature, where a Pt-C(heteroaryl) bond is cleaved rather than a Pt-C(alkyl) one.

Full text of DOI:10.1021/ic400908f

Article
pubs.acs.org/IC
Rollover Cyclometalation with 2‑(2′-Pyridyl)quinoline
Antonio Zucca,*,† Diletta Cordeschi,† Luca Maidich,† Maria Itria Pilo,† Elisabetta Masolo,†
Sergio Stoccoro,† Maria Agostina Cinellu,† and Simona Galli‡
†Dipartimento di Chimica e Farmacia, Universita
̀
degli Studi di Sassari, via Vienna 2, 07100 Sassari, Italy
‡Dipartimento di Scienza e Alta Tecnologia, Universita
̀
dell′Insubria, via Valleggio 11, 22100 Como, Italy
S
* Supporting Information
ABSTRACT: Rollover cyclometalation of 2-(2′-pyridyl)quinoline, L, allowed
the synthesis of the family of complexes [Pt(L-H)(X)(L′)] and [Pt(L*)(X)-
(L′)][BF4] (X = Me, Cl; L′ = neutral ligand), the former being the first
examples of Pt(II) rollover complexes derived from the ligand L. The ligand L*
is a C,N cyclometalated, N-protonated isomer of L, and can also be described
as an abnormal-remote pyridylene. The corresponding [Pt(L-H)(Me)(L′)]/
[Pt(L*)(Me)(L′)]+ complexes constitute an uncommon Brønsted−Lowry
acid−base conjugated couple. The species obtained were investigated in depth
through NMR and UV−vis spectroscopy, cyclic voltammetry, and density
functional theory (DFT) methods to correlate different chemico-physical
properties with the nature of the cyclometalated ligand (e.g., L vs bipy or L* vs
L) and of the neutral ligand (DMSO, CO, PPh3). The crystal structures of
[Pt(L-H)(Me)(PPh3)], [Pt(L-H)(Me)(CO)] and [Pt(L*)(Me)(CO)][BF4] were determined by X-ray powder diffraction
methods, the latter being the first structure of a Pt(II)-based, protonated, rollover complex to be unraveled. The isomerization of
[Pt(L*)(Me)(PPh3)]+ in solution proceeds through a retro-rollover process to give the corresponding adduct [Pt(L)(Me)-
(PPh3)]+, where L acts as a classical N,N chelating ligand. Notably, the retro-rollover reaction is the first process, among the
plethora of Pt−C bond protonolysis reactions reported in the literature, where a Pt−C(heteroaryl) bond is cleaved rather than a
Pt−C(alkyl) one.
the preferred mechanism depends on the nature of the anionic
ligands coordinated to the metal ion.4
INTRODUCTION
■
The chemistry of cyclometalated compounds is of great current
interest because of the wide range of potential applications they
may have in many areas, such as organic synthesis,
homogeneous catalysis, photochemistry, and design of
advanced materials and biologically active agents.1
An emerging class of cyclometalated compounds consists of
the so-called “rollover complexes”, derived from bidentate
chelating ligands for which partial decomplexation and internal
rotation of the ligand allow an inactivated, remote C−H bond
to interact with the metal and give rise to a cyclometalated
species (Scheme 1). Such behavior, called “rollover” cyclo-
metalation, is still rather rare.2
Much effort has been devoted to shed light into the
mechanism of the rollover process. The fundamental
mechanistic difference between “classical” and “rollover”
cyclometalation lies in the nature of the starting adduct.
Rollover cyclometalation originates by definition from a
chelated adduct, and the crucial point is the internal rotation
of the ligand, which occurs before the C−H bond activation. In
the case of platinum(II), the first studies on the topic showed
the nucleophilic nature of the metal ion in the course of the C−
H bond activation.3 Successive studies in the gas phase
demonstrated a clear preference for an oxidative-addition/
reductive-elimination mechanism for platinum,4 whereas σ-
bond metathesis is favored for nickel. In the case of palladium,
The most studied ligand in this context is 2,2′-bipyridine
(bipy),5 but the family of rollover complexes comprises other
bidentate heterocyclic ligands, such as 2,2′:6′,2″-terpyridine,6 2-
(2′-thienyl)pyridines,7 pyrazolylmethanes,8 N-(2-pyridyl)-7-
azaindole,9 and even 2-phenylpyridines.10 The peculiarity of
rollover complexes derives from the presence of an
uncoordinated donor atom, usually a nitrogen or a sulfur,
which may influence the reactivity and properties of the whole
complex. The growing interest in this emerging field is
evidenced by the publication of a very recent review dedicated
to this topic.2 Worthy of note, rollover complexes have been
recently found to promote C−C bond formation in the gas
phase by activation of chloromethanes11 and by dehydrosulfu-
rization of thioethers.12 In addition, rollover cyclometalation
pathways have found catalytic applications in the hydroarylation
of alkenes and alkynes with 2,2′-bipyridines and 2,2′-biquino-
lines,13 and in the transformation of 3-alkynyl and 3-alkenyl-2-
arylpyridines into 4-azafluorene compounds.14 Moreover,
insertion reactions in the metal−carbon bond have been
successfully used for the synthesis of C(3) substituted
bipyridines,15 while double rollover metalation on the
Received: April 12, 2013
© XXXX American Chemical Society
A
dx.doi.org/10.1021/ic400908f | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Scheme 1. Rollover Cyclometalation
bipyridine ligand has afforded polymeric organometallic
species3a or bimetallic complexes where the metal centers are
connected by a highly delocalized planar system.16
The higher electronic delocalization of quinoline compared
to pyridine deserves a few comments: because of the additional
condensed ring, the quinoline nitrogen is expected to be more
basic, having the possibility to delocalize the positive charge,
after protonation, on a larger area. Actually, other effects
contribute to the basicity, such as solvation and steric
congestion. In particular, the peri-effect,28 generating repulsion
between the N−H hydrogen and the adjacent C−H one in the
N-protonated quinoline, lowers the stability of this species, so
that the basicity scale is quinoline < pyridine < isoquinoline
(pKb = 9.15, 8.83, 8.54, respectively; Chart 2).29
Rollover derivatives of 2,2′-bipyridine may be compared to
the analogous derivatives of 2-phenylpyridine (Chart 1), with
Chart 1
Chart 2
the striking difference that the formally anionic, deprotonated
rollover 2,2′-bipyridine is no more a spectator ligand, but may
be an active participant in the course of the chemical
transformations of the complex, through, for example,
coordination or protonation. In particular, protonation of the
uncoordinated nitrogen atom allows the synthesis of
uncommon cationic complexes which may be regarded either
as mesoionic species,17 or abnormal pyridylenes.18 These
compounds may convert into the corresponding Pt(N,N)
adducts through a “retro-rollover” reaction.19 The reversibility
of the rollover process, found in ruthenium,20 rhodium,21 and
platinum complexes,10,12 is extremely attractive, offering, in
principle, the possibility to design catalytic cycles based on
“rollover”/“retro-rollover” paths, employing the bipy/bipy-H
ligands as a hydrogen atom reservoir/acceptor system.
As for the more common cyclometalated species, also in the
case of rollover compounds the properties of the complexes
may be modulated on the basis of the nature of the
cyclometalated ligand. Rollover cyclometalation may indeed
create a highly delocalized system. Thus, following our
continuous efforts in the comprehension of the behavior of
cyclometalated22 and rollover16,23 derivatives of platinum(II),
we have decided to investigate a more delocalized ligand, such
as 2-(2′-pyridyl)-quinoline, L (Scheme 2). The latter, compared
On the whole, the electronic effects for 2-(2′-pyridyl)-
quinoline are not trivial and may furnish, together with the
augmented steric hindrance, additional interesting potentialities
with respect to 2,2′-bipyridine.
We report here the first example of an air- and moisture-
stable Pt(II) rollover complex containing 2-(2′-pyridyl)-
quinoline, a coherent picture of which is provided by means
of NMR (1H, 13C, 2D-COSY, NOE-1D experiments), IR and
electronic spectroscopy, as well as cyclic voltammetry. The
versatile chemistry of this species is further described in terms
of its behavior toward different acids and in substitution
reactions. Finally, an example of spontaneous retro-rollover
reaction19 is reported and described in detail. This reaction,
which may have considerable potential applications in
catalysis,2 proceeds through internal protonolysis and cleavage
of a Pt−C(heteroaryl) rather than a Pt−C(alkyl) bond. To the
best of our knowledge, this is the only case, among the plethora
of Pt−C bond protonolysis reactions reported in the literature,
where a Pt−C(heteroaryl) bond is cleaved rather than a Pt−
C(alkyl) one.30
It is worth remembering that protonolysis of M−CH3 bonds,
considered as the microscopic reverse reaction of the activation
of alkane C−H bonds, has been the subject of extensive
mechanistic investigations.31
Scheme 2. 2-(2′-Pyridyl)quinoline, L, with the Atom
Number Labeling Scheme Used Throughout the Paper
EXPERIMENTAL SECTION
■
Materials and Methods. Cis-[Pt(Me)2(DMSO)2] was synthe-
sized according to the literature.32,33 All the solvents were purified and
dried according to standard procedures.34 Elemental analyses were
performed with a Perkin-Elmer elemental analyzer 240B. Infrared
spectra were recorded with a FT-IR Jasco 480P in solution or using
1
Nujol mulls. H, 13C{1H}, and 31P{1H} NMR spectra were recorded
with a Varian VXR 300 or a Bruker Avance III 400 spectrometer.
to 2,2′-bipyridine, possesses an additional fused ring, potentially
imparting different electronic and steric properties, and has
been investigated until now only by classical coordination
means: examples available in the literature involve iron,24
rhodium,25 platinum,26 and gold.27
1
Chemical shifts are given in ppm relative to internal TMS for H and
13C{1H}, and to external 85% H3PO4 for 31P{1H}. J values are given in
hertz (Hz). NOE difference, 2D-COSY, 2D-NOESY and 13C apt
(attached proton test) experiments were performed by means of
standard pulse sequences.
B
dx.doi.org/10.1021/ic400908f | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
UV−vis spectra were recorded on a Hitachi U-2010 spectropho-
tometer. Cyclic voltammetric tests were performed using an Autolab
PGSTAT12 (Ecochemie) potentiostat/galvanostat interfaced with a
PC under GPES software, employing a single-compartment three-
electrode cell, at room temperature, under Ar atmosphere, at a
potential scan rate of 100 mV s−1. A 2 mm diameter Pt disk electrode
(CH Instruments) was used as working electrode, an aqueous Ag/
AgCl (Amel) with suitable salt bridge was adopted as the reference
electrode, and a graphite rod was the auxiliary electrode. All the
experiments were carried out in CH2Cl2 (Sigma-Aldrich, anhydrous,
≥99.8%) using 0.1 M tetraethylammonium hexafluorophosphate
(TEAPF6, Sigma-Aldrich, for electrochemical analysis, ≥99.0%) as
supporting electrolyte, with sample concentration about 2 × 10−3 M.
Computational Details. All calculations were carried out using
the Firefly QC package,35 which is partially based on the GAMESS
(US)36 source code, using the hybrid PBE0 functional developed by
Perdew, Burke and Ernzerhof37,38 and implemented in the hybrid form
by Adamo and Barone.39 Ahlrichs and co-workers40 def2-SVP basis set,
as found in the EMSL basis set library,41 were used for all lighter atoms
(H, C, N, O, S, and P), while for platinum the same basis set was
integrated with an effective core potential (ECP) removing 60 core
electrons. Convergence criteria were tightened compared to the
default ones: an SCF cycle was considered converged if the density
change, in absolute value, between two subsequent iterations was less
than 10−6; the geometry was considered converged if the largest
component of the gradient, in hartree bohr−1, was less than 10−6.
Frequency analysis at the same level of theory (PBE0/def2-SVP)
and with the same convergence parameters was carried out for all
equilibrium geometries to check their nature on the potential energy
surface (PES). The calculations on all complexes showed no imaginary
frequencies in the output thus confirming that they are true minima.
Time-dependent density functional theory (TD-DFT),42,43 as
implemented in the Firefly program, was used to calculate the lowest
30 excited states having singlet multiplicity.
126.37; 127.42; 127.88; 129.20; 129.23 (JPt−C = 64.5 Hz); 138.36;
139.26 (JPt−C = 89.0 Hz); 140.80 (Pt−C, JPt−C = 1090 Hz); 146.32;
150.45; 161.75 (Cq, JPt−C = 52.4 Hz); 166.09 (Cq, JPt−C = 27.9 Hz).
When the reaction is followed at room temperature in acetone-d6, a
mixture of complexes 1 and 2a is observed in solution by means of 1H
NMR.
1
2
Complex 1, selected H NMR data: 1.23 (s sat, 3H, CH3, JPt−H
=
2
86.7 Hz); 1.16 ppm (s sat, 3H, CH3, JPt−H = 88.9 Hz); 8.90 (dd sat,
3
1H, H6′, JPt−H = 24.4 Hz).
Synthesis of [Pt(L-H)(Me)(PPh3)], 2b. To a solution of 2a (95.7 mg,
0.121 mmol) in CH2Cl2 (25 mL), 32.0 mg of PPh3 (0.121 mmol)
were added under vigorous stirring. The solution was stirred for 1 h,
then concentrated to a small volume and treated with n-hexane. The
precipitate formed was filtered off and washed with n-pentane to give
the analytical sample as a yellow solid. Yield 95%. m.p.: 201−205 °C.
Anal. (%) calcd for C33H27N2PPt: C = 58.49, H = 4.02, N = 4.13;
1
found C = 58.23, H = 3.95, N = 3.83. H NMR (ppm, CDCl3): 0.86
3
2
(d sat, 3H, CH3−Pt, JP−H = 7.5 Hz, JPt−H = 83.1 Hz); 6.73 (m, 1H,
H5′); 7.38−7.46 (m, 11H, aromatics); 7.58 (m, 1H, H7); 7.76−7.87
(m, 9H, aromatics); 8.02 (d, 1H, H8, JH−H = 8.0 Hz); 8.64 (d sat, 2H,
4
3
H4+H3′, JP−H = 6.0 Hz, JPt−H = 53.6 Hz). 31P NMR (ppm, CDCl3):
33.4 (s sat, PPh3, JPt−P = 2236 Hz).
Synthesis of [Pt(L-H)(Me)(CO)], 2c. CO was bubbled into a solution
of 2a (86.0 mg, 0.178 mmol) in CH2Cl2 (20 mL) at room temperature
for 2 h. The solution was concentrated to a small volume and treated
with n-pentane. The precipitate formed was filtered off, washed with n-
pentane, and vacuum pumped to give the analytical sample as a dark-
yellow solid. Yield 85%. Anal. (%) calcd for C16H12N2OPt C = 43.34,
H = 2.73, N = 6.32, found C = 43.37, H = 2.67, N = 5.94. m.p.: 180
1
2
°C. H NMR (ppm, CDCl3): 1.29 (s sat, 3H, CH3−Pt, JPt−H = 85.9
Hz); 7.37 (ddd, 1H, H5′, JH−H = 7.2, 5.4, 1.5 Hz); 7.50 (ddd, 1H, H6,
JH−H = 8.1 Hz); 7.61 (ddd, 1H, H7, JH−H = 8.2, 1.5 Hz); 7.79 (dd, 1H,
H5, JH−H = 8.2 Hz); 8.01 (m, 1H, H8); 8.03 (dt, 1H, H4′, JH−H = 7.7,
3
1.6 Hz); 8.46 (s sat, 1H, H4, JPt−H = 53.7 Hz); 8.61 (dd, 1H, H3, JH−H
3
Preparations. Synthesis of 2-(2′-Pyridyl)quinoline, L. Method A.
2-(2′-Pyridyl)quinoline was prepared according to literature meth-
ods44 from quinoline-N-oxide. The first step of the synthesis requires
the isolation of 2-cyano-quinoline, which was prepared by the reaction,
at 0 °C, of quinoline-N-oxide (4.00 g, 27.5 mmol), (CH3)2NCOCl
(3.17 g, d = 1.168 g/mL, 29.0 mmol) and (CH3)3SiCN (3.9 mL, 2.92
g, d = 0.744, 29.0 mmol) in CH2Cl2 (Yield 65%). 2-(2′-Pyridyl)-
quinoline was obtained from 2-cyano-quinoline through co-cyclo-
= 7.7 Hz); 8.71 (dd sat, 1H, H6′, JH−H = 5.4 Hz, JPt−H = 18 Hz). IR
(CH2Cl2, νmax/cm−1): 2057 s (CO); (Nujol, νmax/cm−1): 2053 s
(CO).
Synthesis of [Pt(L-H)(Cl)(DMSO)], 4a. To a solution of complex 2a
(100 mg, 0.203 mmol) in acetone (25 mL), 2.0 mL of HCl 0.1 M
(0.203 mmol) and 0.2 mL of DMSO (2.8 mmol) were added under
vigorous stirring. The mixture was stirred for 8 h, then complex 4a was
extracted with CH2Cl2 (3 × 10 mL), dried with Na2SO4, filtered, and
concentrated to a small volume. Addition of n-pentane produced a
precipitate that was filtered off, washed with n-pentane and vacuum
pumped to give the analytical sample as a yellow solid. Yield 65%.
Anal. (%) calcd for C16H15ClN2OSPt: C = 37.39, H = 2.94, N = 5.45.
Found C = 37.03, H = 2.71, N = 5.07. m.p.: 200 °C. 1H NMR (ppm,
trimerization of acetylene in the presence of Bonnemann catalyst
̈
(Co(cp)COD), in toluene at 120 °C, with a 30% yield.
Method B. 2-(2′-pyridyl)quinoline was prepared by cyclization
between o-aminobenzaldehyde and 2-acetylpyridine, according to
references 24 and 45. Yield 45%.
1H NMR (ppm, CDCl3): 7.36 (ddd, 1H, H5′, JH−H = 7.5, 4.8, 1.2
Hz); 7.55 (ddd, 1H, H6, JH−H = 8.4, 6.7 Hz); 7.75 (ddd, 1H, H7, JH−H
= 8.2, 6.7 Hz); 7.84−7.90 (m, 2H, H3′, 4′); 8.18 (d, 1H, H5, JH−H = 8.5
Hz); 8.28 (d, 1H, H4, JH−H = 8.6 Hz); 8.56 (d, 1H, H3, JH−H = 8.6 Hz);
8.65 (d, 1H, H8, JH−H = 8.2 Hz); 8.74 (ddd, 1H, H6′, JH−H = 5.6 Hz).
Synthesis of [Pt(L-H)(Me)(DMSO)], 2a. To a solution of cis-
[Pt(Me)2(DMSO)2] (100 mg, 0.262 mmol) in acetone (30 mL), 54
mg of 2-(2′-pyridyl)quinoline, L (0.262 mmol) were added under
vigorous stirring. The solution was heated to 50 °C for 4 h, then it was
evaporated to a small volume and treated with n-hexane. The
precipitate formed was filtered off, washed with n-hexane and vacuum
pumped to give the analytical sample as a yellow solid. Yield 90%.
Anal. (%) calc for C17H18N2OSPt: C = 41.38, H = 3.68, N = 5.68.
CDCl3): 3.71 (s sat, 6H, CH3 (DMSO), JPt−H = 24.0 Hz); 7.48 (m,
3
2H, H5′+H6); 7.63 (m, 1H, H7); 7.81 (d, 1H, H5, JH−H = 8.3 Hz); 8.00
(dd, 1H, H8, JH−H = 7.8 Hz); 8.05 (t, 1H, H4′, JH−H = 7.8 Hz); 8.55 (d,
1H, H3′, JH−H = 8.3 Hz); 9.02 (s sat, 1H, H4, 3JPt−H = 46.3 Hz); 9.71 (d
3
sat, 1H, H6′, JH−H = 4.8 Hz, JPt−H = 33.3 Hz). Assignments based on
2D- COSY and NOESY spectra.
Synthesis of [Pt(L-H)(Cl)(PPh3)], 4b. To a solution of 4a (50.0 mg,
0.097 mmol) in CH2Cl2 (25 mL), 30.7 mg of PPh3 (0.116 mmol)
were added under stirring. The solution was stirred for 4 h, then
concentrated to a small volume and treated with diethyl ether. The
precipitate formed was filtered off and washed with diethyl ether to
give the analytical sample as a yellow solid. Yield 94%. Anal. (%) calcd
for C32H24ClN2PPt: C = 55.06, H = 3.47, N = 4.01. Found C = 55.28,
1
1
Found C = 41.27, H = 3.43, N = 5.56. Mp 150 °C (dec). H NMR
H = 3.71, N = 3.88. m.p.: 195 °C. H NMR (ppm, CDCl3): 6.86 (d,
(ppm, CDCl3): 0.82 (s sat, 3H, CH3−Pt, 2JPt−H = 82.0 Hz); 3.29 (s sat,
1H, JH−H = 8.3 Hz); 7.16−7.21 (m, 2H); 7.35−7.54 (m, 12H); 7.80−
7.90 (m, 6H, aromatics); 8.05 (t, 1H, JH−H = 7.9 Hz); 9.98 (d sat, 1H,
H6′, 3JPt−H = 30 Hz). 31P NMR (ppm, CDCl3): 22.4 (s sat, PPh3, JPt−P
= 4288 Hz).
Synthesis of [Pt(L-H)(Cl)(CO)], 4c. To a solution of 4a (50.0 mg,
0.089 mmol) in CH2Cl2 (20 mL) CO was bubbled for 6 h at room
temperature. The crude mixture was then filtered, concentrated to a
small volume and treated with n-hexane. The precipitate formed was
filtered, washed with n-hexane and vacuum pumped to give the
3
6H, CH3(DMSO), JPt−H = 18.3 Hz); 7.43 (ddd, 1H, JH−H = 7.2, 5.5,
1.4 Hz, H5′); 7.47 (ddd, 1H, JH−H = 8.3, 6.8, 1.3 Hz, H6); 7.60 (td, 1H,
H7, JH−H = 8.3, 6.8, and 1.5 Hz); 7.79 (d, 1H, H5, JH−H = 8.3 Hz); 8.01
(td, 1H, H4′, JH−H = 7.8 Hz); 8.03 (d, 1H, H8, JH−H = 8.3 Hz); 8.41 (s
3
sat, 1H, H4, JPt−H = 60.4 Hz); 8.62 (d, 1H, H3′, JH−H = 7.9 Hz); 9.82
(ddd sat, 1H, H6′ JH−H = 5.6, 1.6, 0.8 Hz, 3JPt−H = 14.2 Hz). 13C NMR
(ppm, CDCl3): −13.07 (Pt-CH3, JPt−C = 766.1 Hz); 43.77 (DMSO,
JPt−C = 41.9 Hz); 122.46 (JPt−C = 23.2 Hz); 125.19 (JPt−C = 10.7 Hz);
C
dx.doi.org/10.1021/ic400908f | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
analytical sample as a yellow solid. Yield 65%. m.p.: 150 °C (dec).
7t, P trans to N(quinoline) (main species, ca. 56%) 1H: 0.72 (3H, d
3
2
Anal. (%) calc for C15H9ClN2OPt: C = 38.85, H = 1.96, N = 6.04.
sat, CH3, JP−H = 4.7 Hz, JPt−H = 74 Hz), 8.72 and 8.69 (AB system,
1
4
Found C = 38.50, H = 1.87, N = 5.83. H NMR (ppm, CDCl3): 7.54
2H, H3 + H43, JA−B = 8.5 Hz), 8.94 (m broad, 1H, JP−H ≈ JH−H ≈ ca.
(m, 2H, H5′+H6); 7.71 (m, 2H); 8.02 (d, 1H, H5, JH−H = 8.3 Hz); 8.12
(m, 2H, H3′+H4′); 8.53 (m, 1H, H8); 9.58 (d sat, 1H, H6′, JH−H = 5.4
4.5−5.0 Hz, JPt−H = 34 Hz, H6′). 31P NMR: 15.2 (JPt−P= 4449 Hz).
1
Other H NMR signals: 8.91 (d, 1H, JH−H = 8.5 Hz), 8.80 (d, 1H,
3
Hz, JPt−H = 33 Hz). IR (Nujol, νmax/cm−1): 2106 s (CO).
JH−H = 7.9 Hz), 8.47 (m, 1H), 8.42 (t, 1H, JH−H = 8.7 Hz), 8.22 (d,
1H, JH−H = 7.8 Hz), 8.09 (d, 1H, JH−H = 8.1 Hz), 7.84−7.21 (m,
PPh3), 7.01 (m, 1H), 6.83 (m, 1H).
Synthesis of [Pt(L*)(Me)(DMSO)][BF4], 5a-BF4. To a solution of 2a
(50 mg, 0.101 mmol) in CH2Cl2 (25 mL), 41.3 mg of [18-crown-6-
H3O][BF4] (0.111 mmol) were added under vigorous stirring. After 1
h the solution was filtered, concentrated to a small volume and treated
with diethyl ether. The precipitate formed was filtered off, washed with
diethyl ether, and vacuum pumped to give the analytical sample as a
green-yellow solid. Yield 80%. Anal. (%) calc for C17H19BF4N2OSPt: C
= 35.15, H = 3.29, N = 4.82. Found C = 35.02, H = 3.58, N = 4.63.
X-ray Powder Diffraction Structural Analysis. Polycrystalline
samples of compounds 2b, 2c and 5c-BF4, not containing single
crystals of suitable quality, were deposited in the hollow of an
aluminum sample-holder equipped with a quartz zero-background
plate.46 For all the species, diffraction data were collected by means of
overnight scans in the 2θ range of 5−105°, with steps of 0.02°, on a
Bruker AXS D8 Advance diffractometer, equipped with Ni-filtered
Cu−Kα radiation (λ = 1.5418 Å), with a Lynxeye linear position-
sensitive detector, and with the following optics: primary beam Soller
slits (2.3°), fixed divergence slit (0.5°), receiving slit (8 mm). The
generator was set at 40 kV and 40 mA. Standard peak search, followed
by indexing through the Singular Value Decomposition approach47
implemented in TOPAS-R,48 allowed the detection of the approximate
unit cell parameters of all the species. The space groups were assigned
on the basis of the systematic absences. Prior to structure solution, unit
cells and space groups were checked by means of Le Bail refinements.
Structure solutions were performed by the simulated annealing
technique, implemented in TOPAS, employing rigid, idealized models:
in the case of 2b and 5c-BF4, a rigid group was defined comprising the
whole Pt(II) complex (completed, for 5c-BF4, by a rigid group
describing the counterion). As for 2b, two rigid groups were defined,
comprising the PPh3 moiety and the residual portion of the complex,
respectively. Average values, retrieved by analyzing similar moieties
present in the Cambridge Structural Database, were assigned to the
bond parameters defining the stereochemistry at the metal ions, the
ligands, and the counterion. The rollover nature of the complexes was
established in all the cases by NMR. Thus, a rigid body comprising a
C,N-chelating ligand was adopted. Nonetheless, the possibility of
having a CH3 moiety trans to the carbon atom, not to the nitrogen
one, was investigated at the stage of structure solution. In all the cases,
the geometry suggested by NMR resulted in structural models with
lower figures of merit, which were thus adopted for the final structure
refinements.
A heavy preferred orientation along [1 0 1] affected the
diffractogram of 2c: at the structure solution stage, the mere
introduction of a correction in the March-Dollase formulation did
not allow to reach convergence to a sensible structural model. A
reasonable model was obtained only minimizing the preferred
orientation by acquiring XRPD data on a batch of 2c admixed with
a dispersing, amorphous material (flour). For the refinement stage, to
avoid the contribution of flour to the diffractogram, preferred
orientation was hampered by side-loading 2c onto the sample-holder.
The final refinements were carried out by the Rietveld method,
maintaining the rigid bodies introduced at the solution stage, but
allowing the ligand to depart from planarity. The background was
modeled by a Chebyshev polynomial function. One, isotropic thermal
parameter was assigned to the metal atoms (BM) and refined; lighter
atoms were given a Biso = BM + 2.0 Å2 thermal parameter. Final
Rietveld refinement plots are shown in Supporting Information, Figure
S1. Fractional atomic coordinates are supplied in the Supporting
Information as CIF files. X-ray crystallographic data in CIF format
have been deposited at the Cambridge Crystallographic Data Center as
supplementary publications no. CCDC 933079 (2b), CCDC 933080
(2c) and CCDC 933081 (5c-BF4). Copies of the data can be obtained
free of charge on application to the Director, CCDC, 12 Union Road,
Cambridge, CB2 1EZ, U.K. (Fax: +44−1223−335033; e-mail:
deposit@ccdc.cam.ac.uk or http://www.ccdc.cam.ac.uk).
1
m.p.: 140 °C (dec). H NMR (ppm, CDCl3): 0.85 (s sat, 3H, CH3−
2
3
Pt, JPt−H = 81.3 Hz); 3.33 (s sat, 6H, CH3 (DMSO), JPt−H = 20.3
Hz); 7.69 (m, 1H); 7.76 (m, 1H); 7.95−8.02 (m, 2H, aromatics); 8.31
(m, 1H); 8.53 (d, 1H, JH−H = 8.1 Hz); 8.95 (d, 1H, H8, JH−H = 7.9
Hz); 9.02 (s sat, 1H, H4, 3JPt−H = 63 Hz); 10.0 (d sat, 1H, H6′, JH−H
=
/
3
5.2 Hz, JPt−H = 13.2 Hz); 13.9 (broad, 1H, N−H). IR (Nujol, νmax
cm−1): 3280, 1058, s br (BF4 ).
−
Synthesis of [Pt(L*)(Me)(PPh3)][BF4], 5b-BF4. To a solution of 2b
(24.4 mg, 0.36 mmol) in CH2Cl2 (15 mL) 13.4 mg of [18-crown-6-
H3O][BF4] (0.036 mmol) were added. After 2 h, the mixture was
concentrated to a small volume and treated with Et2O to form a
precipitate. The solid was filtered off, washed with Et2O and vacuum-
pumped to give the analytical sample as a yellow solid. Yield 90%.
Anal. (%) calc for C33H28BF4N2PPt: C = 51.78; H = 3.69; N = 3.66;
found: C = 51.96; H = 3.57; N = 3.49. 1H NMR (ppm, CDCl3): 0.91
(s sat, 3H, 3JP−H = 7.2 Hz, 2JPt−H = 82 Hz); 7.04 (t, 1H, JH−H = 6.4 Hz,
H5′); 7.40−7.56 (m, 9H); 7.70−7.84 (m, 7H); 7.92−8.11 (m, 3H);
8.21 (td, JH−H = 9.0, 1.2 Hz, H4′); 8.53 (d, 1H, JH−H = 8.7 Hz, H3′);
4
3
9.35 (d sat, 1H, JP−H = 5.6 Hz, JPt−H = 55.2 Hz, H4); 9.91 (d, 1H,
JH−H = 8.1 Hz, H8); 13.79 (s, br, 1H, NH). 31P NMR (ppm, CDCl3):
32.0 (s sat, PPh3, JPt−P = 2507 Hz).
Synthesis of [Pt(L*)(Me)(CO)][BF4], 5c-BF4. To a solution of 2c
(17.7 mg, 0.040 mmol) in CH2Cl2 (15 mL) 14.7 mg of [18-crown-6-
H3O][BF4] (0.040 mmol) were added. After 2 h, the mixture was
concentrated to a small volume and treated with Et2O to form a
precipitate. The solid was filtered off, washed with Et2O and vacuum-
pumped to give the analytical sample as a yellow solid. Yield 85%.
Anal. (%) calc for C16H13BF4N2OPt: C = 36.18; H = 2.47; N = 5.27;
found: C = 36.04; H = 2.58; N = 5.11. 1H NMR (ppm, CDCl3): 1.36
(s sat, 3H, CH3, 2JPt−H = 84.0 Hz); 7.75 (m, 1H); 7.85 (m, 1H); 8.06
(m, 2H); 8.47 (m, 1H); 8.60 (d, 2H, JH−H = 7.9 Hz); 8.98 (d, 1H, H6′,
JH−H = 4.8 Hz); 9.10 (d, 1H, JH−H = 9.0 Hz); 9.17 (s, 1H, H4). IR
(Nujol, cm−1): 3340 m, 2079 vs, 1074 vs, br.
Synthesis of [Pt(L*)(Cl)(PPh3)][BF4], 6-BF4. To a solution of 4b
(25.0 mg, 0.036 mmol) in CH2Cl2 (15 mL) 13.4 mg of [18-crown-6-
H3O][BF4] (0.036 mmol) were added. After 2 h, the mixture was
concentrated to a small volume and treated with Et2O to form a
precipitate. The solid was filtered off, washed with Et2O and vacuum-
pumped to give the analytical sample as a yellow solid. Yield 90%.
Anal. (%) calc for C32H25BClF4N2PPt: C = 48.91; H = 3.21; N = 3.56.
Found: C = 48.60; H = 3.19; N = 3.66. 1H NMR (ppm, CDCl3): 7.05
(d, 1H, JH−H = 8.1 Hz); 7.43−7.66 (m, 10H, aromatics); 7.80−7.86
(m, 8H, aromatics); 7.90 (m, 1H); 8.39−8.48 (m, 2H); 9.02 (d, 1H,
3
H3′), 10.14 (m sat, 1H, H6′, JPt−H =25 Hz), 13.9 (s broad, 1H, NH).
31P NMR (ppm, CDCl3): 20.1 (s sat, PPh3, JPt−P = 4143 Hz).
Synthesis of [Pt(L)(Me)(PPh3)][BF4], 7-BF4. To a solution of 2b (6.8
mg, 0.01 mmol) in CD2Cl2 (1.0 mL) 3.6 mg of [18-crown-6-
H3O][BF4] (0.01 mmol) were added. The reaction was followed
1
through H and 31P NMR spectroscopy. After 2 days, the conversion
into the adducts [Pt(L)(Me)(PPh3)]+ was complete (almost 100%
conversion). Selected NMR data (ppm, CDCl3):
7c, P cis to N(quinoline) (minor species, ca. 44%) 1H: 0.96 (3H, d
sat, CH3, 3JP−H = 3.2 Hz, 2JPt−H = 70 Hz), 7.65 (d, 1H, H6, JH−H = 5.6
Hz), 8.85 and 8.76 (AB system, 2H, H3 + H4, JA‑B = 8.5 Hz). 31P
NMR: 17.1 (JPt−P = 4410 Hz).
Crystal Data for [Pt(L-H)(PPh3)(Me)], 2b. C33H27N2PPt, Mr =
677.64, triclinic, P1, a = 10.6066(4) Å, b = 12.1023(5) Å, c =
̅
12.8774(5) Å, α = 98.863(3)°, β = 94.042(2)°, γ = 55.167(2)°, V =
1340.3(1) Å3; Z = 2; ρcalc = 1.7 g cm−3; F(000) = 664; μ(CuKα) =
D
dx.doi.org/10.1021/ic400908f | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Chart 3
101.6 cm−1. Rp, Rwp, and RBragg, 0.045, 0.059, and 0.034, respectively,
Complex 1 is easily detected in solution by the presence of
two coordinated methyl groups (δ 1.23 ppm, 2JPt−H = 86.7 Hz;
δ 1.16 ppm, 2JPt−H = 88.9 Hz) with Pt−H coupling constants in
line with N−Pt−Me trans arrangements. The H6′ proton, as
expected, is slightly deshielded and coupled to 195Pt (δ 8.90
ppm, JPt−H = 24.4 Hz), this evidence confirming the
coordination of the pyridine ring.
Unfortunately, the simultaneous presence in solution of the
starting compounds, [Pt(Me)2(DMSO)2] and L, of the
intermediate adduct 1 and of the rollover complex 2a does
not allow the isolation of 1 as a pure product.
for 45 parameters.
Crystal Data for [Pt(L-H)(CO)(Me)], 2c. C16H12N2OPt, Mr
=
443.36, monoclinic, P21, a = 17.123(2) Å, b = 8.3068(6) Å, c =
9.7524(9) Å, β = 96.502(6)°, V = 1378.2(2) Å3; Z = 4 (Z′ = 2); ρcalc
=
,
2.1 g cm−3; F(000) = 416; μ(CuKα) = 182.9 cm−1. Rp, Rwp, and RBragg
3
0.078, 0.122, and 0.082, respectively, for 58 parameters.
Crystal Data for [Pt(L*)(Me)(CO)][BF4], 5c -BF4. C16H13BF4N2OPt,
Mr = 538.18, monoclinic, P21/n, a = 20.195(2) Å, b = 10.7569(7) Å, c
= 7.7571(5) Å, β = 100.703(4)°, V = 1655.8(2) Å3; Z = 4; ρcalc = 2.2 g
cm−3; F(000) = 1000; μ(CuKα) = 157.2 cm−1. Rp, Rwp, and RBragg
,
0.052, 0.073, and 0.035, respectively, for 56 parameters.
Simultaneously to the first step of the reaction (the
displacement of the DMSO ligands by L), the rollover reaction
takes place: the quinoline nitrogen (i.e., the most hindered one)
moves away from the metal, allowing rotation of the quinoline
moiety so that the initially remote C3−H bond may approach
the metal and, likely through an oxidative-addition/reductive-
elimination pathway, gives room-temperature C−H bond
activation and rollover cyclometalation. The process is
irreversible because of the elimination of methane.
This mechanism was already proposed for the rollover
metalation of 2,2′-bipyridines on the basis of NMR and kinetic
data,3 even though no evidence for a Pt(IV) species was
provided by NMR, and was subsequently confirmed by
mechanistic studies in the gas phase.51
RESULTS AND DISCUSSION
■
Synthesis and Characterization of the Rollover
Complex [Pt(L-H)(Me)(DMSO)], 2a. The ligand 2-(2′-
pyridyl)quinoline, L (Scheme 2), was synthesized following
two different routes: (a) from quinoline N-oxide via 2-CN-
pyridine and co-cyclotrimerization with acetylene in the
presence of Bonneman catalyst [CpCo(COD)];49 b) by
̈
cyclization between o-aminobenzaldehyde, obtained in situ by
reduction of o-nitrobenzaldehyde with iron, and 2-acetylpyr-
idine (see the Experimental Section).24,45
Despite its similarity with bipy, one of the most widely
studied ligands in coordination chemistry,50 2-(2′-pyridyl)-
quinoline has been the subject of only a few papers, where it
acts exclusively as a bidentate N,N donor.24,27 Compared to
2,2′-bipyridine, L possesses an additional fused ring, potentially
able to furnish supplementary electronic and steric properties.
In the case of platinum(II), only the adducts [Pt(L)X2] (X =
Cl, Br, I) are presently known, synthesized from the
corresponding bisbenzonitrile complexes [Pt(PhCN)2X2].26
By contrast, we report here that the reaction of the electron
rich platinum(II) derivative [Pt(Me)2(DMSO)2] follows a
different route, depending on the reaction conditions. As a
matter of fact, the reaction between [Pt(Me)2(DMSO)2] and L
in acetone at room temperature yields the adduct [Pt(L)-
(Me)2], 1, which rapidly converts into the corresponding
rollover complex [Pt(L-H)(Me)(DMSO)], 2a (Chart 3). To
the best of our knowledge, the latter is the first example of a
rollover complex containing the ligand L.
Complete conversion of the intermediate adduct into the
rollover species occurs, faster, in acetone at 50 °C, and complex
2a may be isolated in high yield in the solid state and
characterized.
The C−H bond activation is regiospecific: by contrast with
the symmetric 2,2′-bipyridine, in the present case two
geometric isomers are possible, but only one is formed, that
is, that derived from the activation of the C−H bond in the
quinoline ring, having an N−Pt−CH3 trans arrangement. No
activation in the pyridine ring is observed. This is in line with
previously reported data,3b which indicate that rollover
cyclometalation is extremely sensitive to steric factors. It should
be noted that the same reaction with 2,2′-bipyridine, which
gives the corresponding complex [Pt(bipy-H)(Me)(DMSO)],
3,5 occurs only under harsher conditions (toluene, reflux); the
different behavior of the two ligands could be ascribed to the
presence of the condensed ring on one of the pyridine rings of
L; the acceleration of the reaction may be thus attributed
mainly to steric factors, which destabilize the adduct 1, but
electronic factors, which can contribute to stabilize the
transition state, may be active as well and cannot be completely
ruled out.
The overall reaction from [Pt(Me)2(DMSO)2] to 2a was
1
followed by means of H NMR spectroscopy in deuterated
acetone at room temperature (see Scheme 2 for the atom
numbering). The spectra show the conversion of [Pt-
(Me)2(DMSO)2] into [Pt(L)(Me)2] (1) (molar ratio 5:1
after approximately 1 min) and of the latter into [Pt(L-
H)(Me)(DMSO)] (2a), according to the pathway [Pt-
(Me)2(DMSO)2] → [Pt(L)(Me)2] → [Pt(L-H)(Me)-
(DMSO)] (Chart 3). The two processes have comparable
rates, so that complex 2a is already present in solution before
complete conversion of [Pt(Me)2(DMSO)2] into 1. Indeed,
after 20 min, [Pt(Me)2(DMSO)2], [Pt(L)(Me)2], and [Pt(L-
H)(Me)(DMSO)] are present in solution in a 10:5:1 molar
ratio.
Complex 2a was thoroughly characterized by means of
analytical and spectroscopic methods. In the absence of a
structural characterization, a detailed NMR study (1H, 13C, 2D-
COSY, NOE-1D experiments) confirmed the proposed
molecular structure (Chart 4). The 1H NMR spectrum
confirms metalation, showing only nine aromatic protons. In
particular, the formation of the C(3)−Pt bond is demonstrated
E
dx.doi.org/10.1021/ic400908f | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
isomers is formed, that is, that with DMSO trans to the
nitrogen atom, instead of trans to the carbon atom, as in 2a.
This is confirmed by the 1H NMR spectrum of 4a which shows
Chart 4
3
a JPt−H value, for the coordinated DMSO, strongly enhanced
with respect to that found for 2a (24.0 Hz vs 18.3 Hz), in line
with the weaker trans influence of N with respect to C. A clear
downfield shift is also present for the H4 proton (δ = 9.02
ppm), attributable to the effect in the space of the DMSO
ligand. Also in this case, a 2D COSY spectrum enabled a
1
complete assignment of the H NMR signals. The assignment
was supported by a NOESY spectrum, which showed, inter alia,
a weak interaction between the DMSO protons, at 3.71 ppm,
and the H4 one, at 9.02 ppm, confirming the proposed
geometry.
In contrast, the behavior of 2a toward [18-crown-6-
H3O][BF4] is different: the Pt−C bonds remain intact and
only protonation of the uncoordinated nitrogen atom occurs, to
give the cationic derivative [Pt(L*)(Me)(DMSO)][BF4], 5a-
BF4, where L* is the κ2C,N neutral ligand formed by
deprotonation of the C(3)-H atom and protonation of the
quinoline nitrogen (Chart 5).
by the lack of the signal of the H3 proton, by the signal of the
3
H4 proton (a singlet with satellites at 8.41 ppm, JPt−H = 60.4
Hz) and by the signal, in the 13C NMR spectrum, of the
1
quaternary metalated C(3) (140.8 ppm), the JPt−C coupling
constant value (1090 Hz) being in agreement with a strong Pt−
C(sp2) bond.
The coordination of the pyridine nitrogen is confirmed by
the deshielding of the H6′ signal, at 9.82 ppm, typical of a
pyridine ring coordinated to Pt in cis to a DMSO moiety,5 and,
Chart 5
3
mostly, by the coupling 195Pt−H6′: the JPt−H value, 14.2 Hz, is
in line with a N−Pt−C trans arrangement.
The coordination sphere of the metal center is completed by
the methyl group and by a DMSO moiety, which show 1H and
13C NMR signals in agreement with a methyl coordinated in
2
trans to a pyridine nitrogen (1H: δ 0.82 ppm, JPt−H = 82 Hz;
1
13C: δ −13.1 ppm, JPt−C = 766 Hz), and a DMSO ligand in
trans to an sp2 carbon (1H: δ 3.29 ppm, 3JPt−H = 18.3 Hz; 13C: δ
2
43.8 ppm, JPt−C = 42 Hz).
The geometry of 2a was also ascertained by a series of NOE-
1D NMR spectra, which show, inter alia, an NOE contact
between the coordinated methyl group and the H4 proton.
Finally, a COSY experiment enabled us to fully assign the
signals in the 1H NMR spectrum (see the Experimental
Section).
1
The H NMR spectrum of 5a shows a broad signal at 13.9
ppm, accounting for one hydrogen atom, and ascribable to the
N−H proton, which disappears after addition of D2O. As a
further confirmation, the IR spectrum shows bands above 3000
cm−1, due to the N−H stretching.
Reactivity of 2a toward Acids. With the aim of
characterizing further the novel complex 2a, some aspects of
its reactivity were investigated, namely, its behavior in the
presence of acids and in substitution reactions (see next
Section). Notably, despite the presence of two metal−carbon
bonds, 2a is extremely stable in solution and in the solid state,
both in air and in the presence of moisture. As a matter of fact,
its stability is higher than that of the analogous complex derived
from 2,2′-bipyridine, [Pt(bipy-H)(Me)(DMSO)], 3, which
tends to decompose in solution. In the absence of any data
regarding decomposition pathways, we can tentatively ascribe
the extra stability of pyridyl-quinoline rollover complexes to the
peculiar electronic properties of the quinoline ring, able to
stabilize additional charges through delocalization in a better
way than pyridine.
With the intention to verify the chemical behavior of the Pt−
C bonds, we investigated the reactivity of 2a toward acids. Two
acids were tested: HCl and [18-crown-6-H3O][BF4], that is,
acids having a coordinating and a weakly coordinating
counteranion, respectively. As expected, the reaction with
aqueous HCl resulted in Pt−CH3 bond attack with subsequent
release of methane, giving the corresponding chloride, [Pt(L-
H)(Cl)(DMSO)], 4a. Only one of the two possible geometric
It is worth to note that the ligands L* and L are isomers. The
nature of the former deserves a few comments: it may be
described as a mesoionic ligand, but also as an abnormal remote
pyridylene.18 The nature of the metal−carbon bond in
compounds containing L* ligands has not been fully
ascertained, but the distinction between the two representa-
tions may be very subtle. Having notable potentialities in
catalysis,52 such uncommon species have received wide
attention in recent years. Thus, even if out of the scope of
the present contribution, the catalytic activity of 5-BF4
complexes is reasonably worth of thorough investigation.
Worthy of note is also the fact that species 2a and 5a
constitute an uncommon Brønsted−Lowry acid−base con-
jugated couple.
The scarce solubility of complex 5a-BF4 did not allow the
1
acquisition of a 13C NMR spectrum. The analysis of its H
NMR spectrum, compared to that of 2a, showed that the Pt−H
coupling constants involving the trans Pt−C(sp2) and Pt−S
3
bonds are greater in the former than in the latter (e.g., JPt−H
4
3
and JPt‑CH ‑DMSO, see the Experimental Section and reference
3
19), whereas the contrary is observed for the trans Pt−N and
3
2
Pt−C(sp3) bonds (e.g., JPt−H ′ and JPt‑CH in 5a and 2a).
6
3
F
dx.doi.org/10.1021/ic400908f | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Reactivity of 2a and 4a in Substitution Reactions.
Starting from 2a and 4a, a series of complexes with different
electronic and steric properties may be synthesized by
substitution of the labile DMSO ligand, as indicated in Chart
6. The substitution is easier in complex 2a because of the
stronger trans influence of carbon compared to nitrogen.
highly unstable because of the high trans influence of carbon
(trans-phobia).53 Despite this, complex 2c is highly stable,
differently from the analogous rollover complex [Pt(bipy-
H)(Me)(CO)],5 which tends to decompose, and from the
cyclometalated complex [Pt(phpy-H)(Me)(CO)] derived from
2-phenyl-pyridine, not isolable in the solid state.54
The geometry of complexes 2b and 2c in solution was
1
further confirmed by H-NOE 1D spectra, which show NOE
Chart 6
contacts between the Pt−Me and the H4 protons.
As already observed with 2a, the reaction of the methyl
complexes 2b and 2c with [18-crown-6-H3O]+ gave the
analogous mesoionic species [Pt(L*)(Me)(PPh3)]+, 5b, and
[Pt(L*)(Me)(CO)]+, 5c, (Chart 7) which were isolated in the
solid state as tetrafluoborate salts, 5b-BF4 and 5c-BF4.
As expected, the protonation of the uncoordinated nitrogen
atom produces, in the 1H NMR spectra, a generalized
downfield shift of the signals of the aromatic protons. This
effect is particularly evident for the H4 protons (δ 8.64, 2b;
9.35, 5b; 8.46, 2c; 9.17, 5c). The NH proton gives a broad
signal at about 13.8 ppm. A NOE difference spectrum of
complex 5b shows that irradiation at 13.8 ppm promotes
enhancement of the doublets at 8.91 and 8.53 ppm, due to the
H3′ and H8 protons (Chart 8). In addition, irradiation of the
Chart 8
Complexes 2b−c and 4b−c, [Pt(L-H)(X)(L′)] (X = Me, L′
= PPh3, 2b; X = Me, L′ = CO, 2c; X = Cl, L′ = PPh3, 4b; X =
Cl, L′ = CO, 4c) were isolated in the solid state with good
yields and characterized by means of microanalyses, IR and
NMR spectroscopy. In particular, the 31P NMR spectra of
species 2b and 4b clearly indicate a P-trans-C geometry for 2b
(1JPt−P= 2236 Hz) and a P-trans-N one for 4b (1JPt−P= 4288
Hz), as expected. In the 1H NMR spectrum of 2b the platinum-
bonded methyl group appears as a doublet with satellites (3JP−H
2
= 7.5 Hz, JPt−H = 83.1 Hz), providing evidence for a Pt−Me
singlet at 9.35 ppm (H4) shows NOE contacts at 8.05 ppm (d,
H5) and 0.90 ppm (d, Me), confirming the purported geometry
of the complex in solution.
bond in cis to a Pt−P one and in trans to a Pt−N one. The P-
trans-C geometry of 2b was further substantiated by the XRPD
structure determination (see next Section).
The 1JPt−P value is higher in 5b (2507 Hz) than in 2b (2236
Hz), as previously observed for the analogous bipyridine
complexes.5,19
The IR spectra of the carbonyl species 2c and 4c show, as
expected, very different CO stretching frequencies: a lower
frequency for the electron-rich complex 2c (νCO = 2053 cm−1)
and a higher frequency (2106 cm−1) for the electron poorer
one 4c. Complex 2c may have some points of interest having
three Pt−carbon bonds, each in a different hybridization state
(sp, sp2, sp3). Notably, as also confirmed by the XRPD
structural analysis (see next Section), the complex contains an
uncommon trans C−Pt−C arrangement, usually considered
The CO stretching frequency in 5c, 2079 cm−1, is higher
than in 2c, 2053 cm−1, as expected because of the minor
retrodonating properties of platinum(II) in the cationic
complex. These data may be compared to those of the
analogous bipyridine-containing couple [Pt(bipy-H)(Me)-
(CO)] and [Pt(bipy*)(Me)(CO)]+ (2044 and 2087 cm−1,
respectively55): the increase of the stretching frequencies, 26
Chart 7
G
dx.doi.org/10.1021/ic400908f | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Figure 1. Portion of the crystal structure of species 2b: (a) the Pt(II) complex; (b) the packing, from which the intermolecular π−π interactions
involving couples of symmetry-related L ligands can be appraised. Carbon, gray; hydrogen, light gray; nitrogen, blue; phosphorus, orange; platinum,
fuchsia.
Figure 2. Portion of the crystal structure of species 2c: (a) one of the two crystallographically independent complexes; (b) the packing, viewed along
[001]. Horizontal axis, a; vertical axis, b. The Pt···Pt non bonding interactions are highlighted with cyan fragmented lines. Hydrogen atoms have
been omitted in panel (b) for the sake of clarity. Carbon, gray; hydrogen, light gray; nitrogen, blue; oxygen, red; platinum, fuchsia.
cm−1 for L vs 43 cm−1 for bipyridine, is very different for the
two ligands and may be explained with the greater capacity of L
in delocalizing a positive charge, due to the quinoline fragment,
so that the Pt−C−O bonds are less affected by the additional
charge than in the bipyridine-based complexes, for which the
protonation directly results in a stronger change in the C−O
bond order.
powder diffraction (XRPD) methods. Relying on the extended
use of rigid bodies, structure determinations from XRPD do
not afford, for the geometrical parameters, the accuracy typical
of single-crystal X-ray diffraction analyses. Nevertheless, the
inter- and supra-molecular features can be confidently derived.
This relevant, otherwise inaccessible, piece of information will
thus be discussed in the following for 2b, 2c, and 5c-BF4.
As detailed in the Experimental Section, the rollover nature
of the Pt(II) complex was ascertained in all the cases by the
thorough NMR characterization: a rigid body model compris-
ing a C,N-chelating ligand was consequently adopted. None-
theless, the possibility of having the methyl group trans to the
Pt-coordinated carbon atom was investigated during the
structure solution stage. The structural models with the lowest
figures of merit always possessed the trans Me−Pt−N geometry
suggested by NMR. These models were thus selected for the
final structure refinements.
Reactivity of 4b. Protonation of complex 4b with [18-
crown-6·H3O]+ allows the isolation of the corresponding
mesoionic species [Pt(L*)(Cl)(PPh3)][BF4], 6-BF4. A broad
signal at 13.9 ppm in the 1H NMR spectrum provides evidence
of the protonation of the nitrogen atom. The analysis of the 1H
NMR spectra of 4b and 6 shows that all the signals of the
aromatic protons are deshielded after protonation, as expected.
In contrast to the 2b−5b couple, in the present case the
195Pt−31P coupling constant decreases after protonation (4143
Hz in 6 vs 4288 Hz in 4b), indicating that cis and trans
influences have a predominant effect on the coupling constant
in these species. In particular, coupling constant values for
bonds in trans to C3 increase after protonation, whereas those
for bonds in cis to C3 follow an opposite trend. Unfortunately, it
is not straightforward to establish whether these data really
indicate strengthening and weakening of the corresponding
Pt−ligand bonds.
Species 2b crystallizes in the triclinic space group P1. Its
̅
asymmetric unit comprises one [Pt(L-H)(Me)(PPh3)] com-
plex (Figure 1a), lying on a general position. The steric
hindrance is responsible for the actual orientation of the PPh3
moiety, no phenyl ring being obviously coplanar with the L
ligand. The reciprocal disposition of the complexes is dictated
by the formation of intermolecular π−π interactions between
couples of parallel L ligands, lying approximately on the (11−2)
plane, and facing each other at about 3.4 Å (Figure 1b). The
two π−π interacting complexes are mutually related by a
X-ray Powder Diffraction Structural Analysis of 2b, 2c,
and 5c-BF4. The crystal and molecular structures of species 2b,
2c, and 5c-BF4 were retrieved from state-of-the-art X-ray
H
dx.doi.org/10.1021/ic400908f | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Figure 3. Portion of the crystal structure of species 5c-BF4: (a) the cationic complex; (b) the packing, viewed along c. Horizontal axis, a; vertical axis,
−
b. The weak C−H···F interactions involving neighboring BF4 anions and cationic complexes are highlighted in panel (b) with cyan fragmented
lines. Carbon, gray; hydrogen, light gray; boron, light orange; fluorine, yellow; nitrogen, blue; oxygen, red; platinum, fuchsia.
crystallographic inversion center; thus, no Pt···Pt contacts can
be appraised together with the π−π ones.
discussion, we will call “experimental” the former, retrieved
from NMR spectroscopy, and “inverted” the latter; moreover a
prime will be used to distinguish the two species.
Species 2c crystallizes in the monoclinic space group P21.
The asymmetric unit contains two, crystallographically
independent, [Pt(L-H)(Me)(CO)] complexes (Figure 2a).
The two independent complexes lie approximately on the
(210) plane (and on the symmetry related one (2−10)), and
form piles of staggered moieties along [210]. Along the piles,
Pt···Pt contacts of about 3.4 Å can be envisaged between
couples of consecutive complexes (Figure 2b). Visualized along
the crystallographic axis c, the adjacent stacks describe a
herringbone motif.
Complexes [Pt(L-H)(CH3)(DMSO)], [Pt(L-H)(CH3)-
(PPh3)], [Pt(L-H)(CH3)(CO)], [Pt(L-H)(CH3)(DMSO)]+,
[Pt(L*)(CH3)(PPh3)]+, [Pt(L*)(CH3)(CO)]+ (2a−c and 5a−
c), and their corresponding “inverted” isomers (2a′-c′ and 5a′-
c′, Scheme 3) were optimized at the PBE0/def2-SVP level of
theory. In each case the isomer deduced from the NMR data is
the one that displays a lower energy; the calculated relative
enthalpy and Gibbs free energy differences (298.15 K and 1
atm) are collected in Table 1 where the isomer having the
Species 5c-BF4 crystallizes in the monoclinic space group
P21/n. Its asymmetric unit contains one cationic [Pt(L*)(Me)-
Table 1. DFT-Calculated Relative Enthalpy (ΔH) and Gibbs
a
(CO)]+ complex (Figure 3a) and one BF4 anion, both lying
−
Free Energy (ΔG) Differences between Experimental and
b
“Inverted” Isomers, Taking the Former As Zero
on general positions. The cationic complexes lie approximately
parallel to the (−302) plane and stack, staggered, along c with a
pace of c/2 (Figure 3b), so that no intermolecular π−π and
Pt···Pt interactions are at work. The anions are located between
the stacks (Figure 3b), their actual orientation being imposed
by the formation of weak C−H···F interactions involving three
nearby complexes (C···F 2.74−3.14 Å).
DFT-Calculated Molecular Structures of 2a−c and 5a−
c. In square planar complexes having a chelating ligand that
blocks two positions, as in 2a−c and 5a−c, two isomers are
possible, namely, one with the methyl residue trans to the
nitrogen atom, and the other where the methyl group is trans to
the metalated carbon atom (Scheme 3). In the next part of the
complex
ΔG
ΔH
[Pt(L-H)(CH3)(DMSO)], 2a
[Pt(L*)(CH3)(DMSO)]+, 5a
[Pt(L-H)(CH3)(CO)], 2c
[Pt(L*)(CH3)(CO)]+, 5c
[Pt(L-H)(CH3)(PPh3)], 2b
[Pt(L*)(CH3)(PPh3)]+, 5b
48.8
39.7
14.4
17.7
23.1
17.0
46.8
37.7
12.8
16.2
20.1
17.7
a
b
Both ZPE corrected, T = 298.15 K, P = 101325 Pa. All the values are
expressed in kJ mol−1.
lowest energy has been taken as reference. As expected, more
than 90% of the “stabilization energy” comes from an enthalpic
factor because the entropic contribution is very similar in both
isomers and, consequently, its role in stabilizing the
experimental isomer is less important. This observation fits
well in the framework of trans-phobia53 and antisymbiosis
effect.56
Scheme 3. Experimental (2a−c) and “Inverted” (2a′-c′)
a
Isomers Studied
As described above, upon reaction with [18-crown-6·H3O]-
[BF4], complexes 2a−c yield the cationic N-protonated
analogues 5a−c. As made evident from the calculations, both
protonation and change in the charge of the complex cause
only slight modifications in bond distances and angles (see
Table 2 and Table 3). Changes in bond distances involving
platinum can be explained in terms of push−pull interactions
between ligands trans to each other, that is, when a bond
elongates, that in trans shortens. In detail, the highest variations,
in absolute value, are evident for Pt−N1′, Pt−CH3, and Pt−L
a
The energies were evaluated also for the corresponding cationic
species 5a−c and 5a′−c′.
I
dx.doi.org/10.1021/ic400908f | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
density is somewhat pulled toward the protonated nitrogen,
which is more electron-deficient than in the neutral complex.
The bonds in the pyridinic ring rearrange without significant
or unexpected changes: all the differences are below 1 pm; in
detail, those bonds closer to the quinolinic system (i.e., N1′-C2′
and C2′-C3′) shorten, while the others are alternately shorter or
longer.
The trend of the unscaled stretching frequencies of the CO
bond in complexes 2c and 5c, recorded in Nujol, is correctly
reproduced by the harmonic analysis: 2053 vs 2183 cm−1 for
the neutral species, and 2079 vs 2225 cm−1 for the protonated
one (the first wavenumber value is the experimental one).
Finally, evaluation of the DFT-calculated proton affinity and
gas phase basicity of the species 2a−c (Table 4) and
Table 2. DFT-Calculated Bond Distances (pm) and Angles
(deg) around the Metal Center for Complexes 2a−c and the
a
Corresponding Protonated Species 5a−c
DMSO
CO
PPh3
2a
5a
2c
5c
2b
5b
Pt−L′
Pt−CH3
Pt−N1′
Pt−C3
N1′−Pt−C3
C3−Pt−CH3
CH3−Pt−L′
L′−Pt−N1′
N1′−Pt−CH3
C3−Pt−L′
233.0
205.5
215.6
201.5
80.3
233.1
204.6
217.3
201.4
79.5
190.3
205.9
216.5
205.3
79.7
190.9
205.3
217.9
205.3
79.0
235.0
205.8
217.5
203.8
79.6
235.8
204.9
219.1
203.7
78.8
91.0
91.5
90.8
91.4
89.9
90.5
90.4
90.4
89.7
88.7
86.0
85.7
98.2
98.5
99.8
100.9
170.4
179.9
104.6
169.3
175.6
105.0
169.2
176.0
171.3
178.5
171.1
178.0
170.5
179.5
Table 4. DFT-Calculated Proton Affinity (ΔH) and Gas
a
a
Phase Basicity (ΔG) for Complexes 2a−c and the
For the atom labels, consult Scheme 2.
b
Corresponding Bipyridine Analogues
Table 3. DFT-Calculated Bond Distances (pm) of the
Cyclometalated Ligand in Complexes 2a−c and 5a−c
complex
ΔG
ΔH
a
[Pt(L-H)(CH3)(DMSO)], 2a
[Pt(L-H)(CH3)(PPh3)], 2b
[Pt(L-H)(CH3)(CO)], 2c
[Pt(bipy-H)(CH3)(DMSO)]
[Pt(bipy-H)(CH3)(PPh3)]
1007.5
1027.4
980.3
1005.3
1026.8
978.6
DMSO
CO
PPh3
L′
2a
5a
2c
5c
2b
5b
995.1
993.1
N1′−C2′
C2′−C3′
C3′−C4′
C4′−C5′
C5′−C6′
C6′−N1′
C2′−C2
C2−N1
C2−C3
C3−C4
C4−C5
C5−C6
C6−C7
C7−C8
C8−C9
C9−C10
C10-N1
C10−C5
134.9
139.5
138.8
139.4
139.1
134.0
147.3
131.4
143.5
138.0
141.8
141.8
137.6
141.6
137.5
142.0
135.1
142.9
133.5
139.6
138.8
139.3
139.4
135.3
146.7
141.6
139.1
141.4
142.1
136.5
134.1
141.9
137.5
141.5
137.8
140.8
134.8
139.7
138.7
139.6
138.8
133.8
147.6
131.5
143.5
138.0
141.8
141.8
137.6
141.6
137.5
142.0
135.1
143.0
135.2
139.5
139.2
139.0
139.3
133.3
147.2
134.2
141.3
139.0
141.4
141.9
137.5
141.5
137.8
140.8
136.6
142.2
135.0
139.7
138.7
139.4
139.0
133.8
147.3
131.6
143.4
138.1
141.8
141.9
137.6
141.6
137.5
142.0
135.0
142.9
135.4
139.6
139.2
138.9
139.4
133.4
146.8
134.3
141.4
139.3
141.4
141.9
137.5
141.5
137.7
140.8
136.4
142.1
1018.6
1016.1
[Pt(bipy-H)(CH3)(CO)]
Both ZPE-corrected. All the values are expressed in kJ mol−1.
964.8
963.2
a
b
comparison with the corresponding bipyridine analogues,
respectively [Pt(bipy-H)(CH3)(DMSO)], [Pt(bipy-H)(CH3)-
(PPh3)], and [Pt(bipy-H)(CH3)(CO)], permits to note the
following: (a) both with L and bipy the easiness of protonation
depends on the neutral ligand coordinated trans to the
metalated carbon atom and increases in the order PPh3 >
DMSO > CO; (b) complexes with cyclometalated bipy always
show lower values than the corresponding with L. These
observations are in good agreement and show a good
correlation with the well-known donor properties of the
neutral ligands involved in this work; moreover it seems that
the basicity order expected for pyridine and quinoline is
respected.
a
For the atom labels, consult Scheme 2.
Electronic Spectroscopy and Electrochemical Behav-
ior. Platinum derivatives 2a−c and 5a−c were characterized by
UV−vis spectroscopy in CH2Cl2. By comparison with the free
ligand, the higher energy bands are attributable to ligand
centered π−π* transitions and the lower ones are tentatively
assigned to metal-to-ligand charge transfer transitions (380−
430 nm).57 N-protonated cationic derivatives show MLCT
absorptions at higher wavelengths than the corresponding
neutral species, suggesting a more effective extension of the
charge delocalization on the complex. It seems reasonable to
ascribe this red-shift of the MLCT band to the effects that the
protonation has on the electron density, as highlighted also by
the DFT calculations (see Table 5). The same effect can be
observed comparing complexes 2a and 3, which differ only for
the cyclometalated ligand (L and bipy, respectively) thus we are
prone to think that the higher extent in the electronic
delocalization originating from the quinoline system instead
of a pure pyridine ring is the reason of the observed red-shift.
The analysis of the UV−vis bands position permits also to
roughly estimate the σ-donor effect of the cyclometalated
ligand.57 Complexes 2a and 3 show λmax respectively at 403 and
380 nm suggesting that bipy-H has a stronger σ donating effect
bonds (ca. 1 pm). Quite surprisingly, the Pt−C3′ bond length is
almost unchanged, showing a 0.1 pm difference which is too
small to be considered significant.
The angles insisting on the metal center do not show
dramatic changes in their absolute value, but it can be noted
that corresponding angles have the same trend, that is, N1′−
Pt−C3 and L−Pt−CH3 close, while the other two become
wider (Table 2). Finally, the planarity of the molecule is not
affected by the protonation of the noncoordinating quinolinic
nitrogen.
The effect of protonation on the cyclometalated ligand was
also investigated (Table 3). As is expected, the changes in the
molecular structure are more pronounced in the surroundings
of the protonation site and, particularly, in the N1−C2, C2−C3,
N1−C10 and C9−C10 bond lengths, with elongations of about
3.0, 2.0, 1.4, and 1.2 pm, respectively. Interestingly, a kind of
“contraction” of the metallacycle bonds is evident on the
quinolinic moiety, that is, C2′−C2, C2−C3, and, to a lesser
extent, C3−Pt. This is in line with the fact that the electron
J
dx.doi.org/10.1021/ic400908f | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Table 5. Electrochemical, Optical, and Computational Data
Eg,opt
b
Eg,theor
Eox
HOMOexp
(eV)
HOMOtheor
LUMOtheor
a
c
d
d
c
c
compound
λ(nm)
(eV)
(eV)
(V)
Ered(V)
LUMOexp (eV)
(eV)
(eV)
e
f
[Pt(bipy-H)(DMSO)(Me)] 276, 307, 380
2.99
4.71
0.72
−5.13
−2.14
−6.28
−1.57
(3)
f
[Pt(bipy*)(DMSO)(Me)]
255, 314, 364
306, 366, 403
2.95
2.94
3.75
4.37
−0.58
−0.85
−7.42
−5.04
−4.47
−2.89
−9.73
−6.15
−5.98
−1.78
e
f
[Pt(L-H)(DMSO)(Me)]
0.64
(2a)
f
e
[Pt(L*)(DMSO)(Me)]+
(5a)
325, 460
2.18
3.43
−6.74
−4.56
−9.54
−6.11
e
f
[Pt(L-H)(PPh3)(Me)] (2b) 262,302, 381
[Pt(L*)(PPh3)(Me)]+ (5b) 264, 329, 429
2.88
2.46
3.59
2.96
3.62
4.23
1.35
1.21
1.58
−5.70
−7.00
−5.61
−7.69
−2.71
−4.54
−2.02
−4.73
−5.89
−8.87
−6.48
−10.17
−6.66 (cis)
−6.64 (trans)
−1.66
−5.81
−2.12
−6.46
−1.71 (cis)
−1.81 (trans)
f
e
3.06
−0.50
−0.44
e
f
[Pt(L-H)(CO)(Me)] (2c)
256, 289
4.36
f
e
[Pt(L*)(CO)(Me)]+ (5c)
275, 284, 350
253, 320, 335
3.71
e
f
L
4.95 (cis)
−5.59
−1.97
4.83 (trans)
a
b
Measured in CH2Cl2 solution. Concentration for UV−vis measurements was about 2 × 10−5 M. The energy-gap (Eg,opt) value was determined
c
d
from the λonset in the UV−vis spectrum. Theoretical values obtained from DFT calculations at PBE0/def2-SVP level. Potential values reported vs
e
Ag/AgCl in CH2Cl2-TEAPF6 0.1 M solvent system. Concentration 2 × 10−3 M. Calculated from the anodic (HOMO) or cathodic (LUMO) onset
f
potential value. The LUMO energy value was calculated from the equation LUMO(eV) = HOMO + Eg,opt, and the HOMO value from
HOMO(eV) = LUMO − Eg,opt. (reference 59).
than L-H; the same trend is visible in the corresponding N-
protonated species 5a, [Pt(L*)(Me)(DMSO)]+, and [Pt-
(bipy*)(Me)(DMSO)]+ (460 vs 364 nm).
electron density on the quinoline ring is lower than in 2c, thus
the oxidation process in the latter is easier than in 2b. In the
case of the dimethylsulfoxide (DMSO) derivative (2a), the
known lability of the Pt−DMSO bond makes it difficult to
compare its electrochemical response with that of the analogues
2b and 2c.
The cathodic portion of the voltammograms does not
evidence the appearance of reductive processes in the potential
range exploited. With regards to the LUMO values, the DFT
calculations seem to confirm that the associated cathodic
process would be at very negative onset potential values,
reasonably not detectable in the CH2Cl2-TEAPF6 solvent
system.
As for the cationic species (5a−c), no oxidative process is
evidenced in the potential range exploited. Evaluation of the
HOMO energy values by DFT calculations suggest that the
anodic response due to the cationic species, mainly located on
the metal center, would be at potential values higher than the
anodic limit of the CH2Cl2-TEAPF6 0.1 M solvent system.
Conversely, the cathodic behavior shows a reduction process
(Table 5) attributable to the reduction of the protonated
portion. The DFT data indicate that the LUMO is always
located on the pyridylquinoline ligand, and the dipole moment
is always oriented toward the cyclometalated ligand, L,
accounting for the relatively small difference between the
reduction potential values of 5b and 5c.
The analysis of the energy-gap (Eg) values derived from the
UV−vis spectra evidence, as predictable, a lower Eg value for
the cationic complexes (5a−c) than for the corresponding
neutral ones (2a−c). Eg values obtained from UV−vis spectra
were compared with estimated data by DFT method (see Table
5), confirming a decrease in Eg values going from neutral to
cationic species. Differences in experimental with respect to
calculated Eg data is between 18 and 36%: although such values
seem quite high, they are in line with those reported for some
Pt(II) derivatives with phenylpyridine and benzoquinoline
ligands.58 We reasonably impute the difference to the
functional/basis set used in the computations and the absence
of the solvent which can play a not negligible role.
Finally, the analysis of the molecular orbitals suggests that
the higher Eg displayed by the bipy-H complexes is due to a
concurrent stabilization of the highest occupied molecular
orbital (HOMO) and destabilization of the lowest unoccupied
molecular orbital (LUMO).
The electrochemical behavior of the neutral (2a−c) and of
the corresponding cationic (5a−c) species was investigated in
the CH2Cl2-TEAPF6 0.1 M solvent system by cyclic
voltammetry.
Neutral complexes show an anodic irreversible process
between 0.6 and 1.4 V, while the oxidation of the free ligand
occurs at about 1.6 V (Table 5). DFT calculations indicate that
in the free ligand the HOMO is located on the whole of the
molecule, while in the complexes it encompasses the quinoline
fragment, the metal center, and, especially in the case of 2c,
even the L′ ligand. Theoretical data support the hypothesis that
the anodic processes involve primarily the pyridylquinoline
ligand (L), and to a minor extent the Pt(II) center and,
eventually, the L′ ligand. Moreover, DFT data show a dipole
moment oriented toward the L′ ligand in 2a and 2b, and
toward the pyridine ring in 2c. The comparison between the
direction of the dipole moment, together with the fact that the
oxidation process is mainly located on L, might explain the
higher oxidation potential of the PPh3 derivative (2b)
compared to the CO one (2c): as a matter of fact, in 2b the
Accordance between experimental and theoretical data is in
the range of analogous comparisons reported in literature.58
Retro-Rollover Process Involving 5b. The mesoionic
species 5a and 5c are rather stable in solution and do not
decompose in the presence of moisture or oxygen. However,
the NMR spectrum of complex 5b, acquired in CDCl3,
1
undergoes a slow change; as a matter of fact, both H and 31P
NMR show the conversion of 5b into two new species (7),
formed in a 1.3:1 molar ratio. The new complexes appear to be
very similar, each having, inter alia, one methyl group and one
phosphorus atom bound to platinum, with similar 195Pt−1H
and 195Pt−31P coupling constant values (see the Experimental
Section), suggesting P−Pt−N and Me−Pt−N trans coordina-
tion for both complexes.
K
dx.doi.org/10.1021/ic400908f | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Scheme 4. Retro-Rollover Process
These similarities suggest that the two species are closely
related, being the two geometric isomers of [Pt(L)(Me)-
(PPh3)]+ (Scheme 4), originated through a “retro-rollover”
process, as recently reported by some of us for the analogous
bipyridine complexes.19 In the starting complex 5b, the C3
atom is metalated, so the appearance of the signals due to the
H3 protons in the spectra of both species strongly corroborates
our assumption. Furthermore, in the aromatic region, a signal
with satellites is clearly visible which is typical of an H6 proton
trans to phosphorus,19 due to a N−Pt−P trans arrangement,
whereas in the other isomer the signal due to H6 is strongly
upfield shifted (δ = 7.65 ppm) by the shielding cones of the
phosphine aromatic rings. In the light of this, the two species
may be indicated as 7t (phosphorus trans to quinoline) and 7c
(phosphorus cis to quinoline). As expected, complex 7t, having
the bulky PPh3 far from the quinoline moiety, is the main
species. In line with this assignment, NOESY-1D spectra
showed contacts between the methyl group at 0.72 ppm (Pt−
Me in 7t) and a doublet at 8.40 ppm belonging to the adjacent
H8 proton. For the other Pt−Me hydrogens, that is, those of
the 7c isomer, a clear contact with the H6′ proton, at 8.94 ppm,
is observed. A H−H COSY spectrum further helped in the
assignments; in particular, two AB systems, due to the H3 and
H4 protons, are clearly visible in the spectra.
The retro-rollover isomerization reaction likely proceeds
through protonation of platinum(II) following an oxidative-
addition/reductive-elimination reaction pathway. The reductive
elimination step is followed by rotation around the C−C bond
between the aromatic rings and by chelation. A second possible
reaction pathway, that is, a concerted mechanism involving
electrophilic attack of the platinum−carbon bond, cannot be
completely ruled out, but in the case of electron-rich
platinum(II) organometallic complexes it is unlikely.60
According to the first mechanism, electron-richer complexes
react faster than electron-poorer species.
the effect of the phosphane properties in the rollover complexes
[Pt(bipy*)(Me)(PR3)]+, it emerged that better donors (e.g.,
P(p-tolyl)3 vs PPh3) highly accelerate the process. Moreover,
we have found that also the nature of the cyclometalated ligand
strongly affects the rate of reaction. The condensed aromatic
ring on the cyclometalated pyridine is likely to strongly favor
the retro-rollover process. However, although electronic effects
seem to be dominant, also steric ones may play a non negligible
role. In particular, repulsion between the N−H and the H8
protons (in mutual peri position, Chart 9), added to the N−H
vs H3′ repulsion, may also influence the process.
Chart 9
Protonolysis of M−CH3 bonds is considered as the
microscopic reverse reaction of the activation of alkane C−H
bonds, and hence it has been the subject of extensive
mechanistic investigations.31 To the best of our knowledge,
all the cases reported in the literature refer to Pt−C(alkyl)
bond cleavage, rather than to Pt−C(aryl) rupture. Thus, the
retro-rollover process is the first case of preferred Pt−C(sp2)
bond protonolysis in place of Pt−C(sp3). This occurrence may
have future potential applications in catalysis, in processes
where a heterocyclic bidentate donor may act as a hydrogen
reservoir/acceptor.
Complexes 7 are not static in solution: a 1D-EXSY
experiment in CD2Cl2 showed that irradiation of the methyl
signal at 0.92 ppm (7t, main species), results in magnetization
transfer to the other methyl group at 0.67 ppm, demonstrating
a dynamic process in solution which rapidly interconverts the
two isomers on the NMR time scale (Chart 10).
The retro-rollover reaction rate is solvent-dependent: the
process is slower in CDCl3 (approximately 30% conversion in a
10−2 M solution after 2 days) and faster in CD2Cl2 (100%
conversion after 2 days). Addition of excess crown ether acid
does not affect the reaction rate, probably because protonation
of Pt does not occur on the cationic species 5 but on the
neutral complexes 2, originated from 5 by deprotonation.
Dilution of the solution results in a slower reaction rate, so it is
likely that the process is not intramolecular, but may involve
water. Saturation with H2O almost inhibits the reaction, but
this may be due to the presence of a water phase, so that H3O+
is partially removed from the chloroform solution.
Chart 10
The analogous conversion of the bipyridine rollover19
complex [Pt(bipy*)(Me)(PPh3)]+ is significantly slower,
reaching completion in CD2Cl2 only after 30 days. In this
case, the reaction rate was shown to be strongly dependent on
the electron density on the metal center. In particular, studying
L
dx.doi.org/10.1021/ic400908f | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
An analogous behavior was observed by Romeo and co-
workers in the case of the symmetric chelating ligand 2,9-
dimethyl-1,10-phenanthroline (N,N) in a series of cationic
[Pt(N,N)(Me)(PR3)]+ complexes.61 By contrast, analogous
complexes of the less-hindered ligands 1,10-phenanthroline and
2,2′-bipyridine, appear static in solution on the NMR time
scale. The severe distortion of the square planar geometry
around the platinum center, due to the methyl groups on the
bidentate ligand 2,9-dimethyl-1,10-phenanthroline, was re-
ported to be responsible for the fluxionality of the complexes.
Detailed kinetic studies showed that the mechanism of the
process may switch from associative to dissociative pathways
according to several factors, such as solvent or counterion
properties, the presence of external nucleophiles, the electronic
and steric properties of the phosphane ligand, and so forth. In
addition to the flipping of the bidentate N,N ligand, also
rotation of the phosphane occurs. Indeed, it was shown that
tuning the bulkiness of the PR3 ligand, the two motions display
identical rates. Because of this synchronized fluxional motion,
these complexes act as a molecular gear.61d,e
To the best of our knowledge, complexes 7c and 7t are the
first platinum(II) species with nonsymmetric N,N chelating
ligands which show a dynamic flipping of the chelated ligand.
The possibility to design molecular gears whose dynamics can
be controlled by the stereoelectronic properties of the ligands is
of outstanding interest. Studies are in progress to evaluate, also
in the present case, the possible behavior as a molecular gear
with different phosphane ligands.
whereas those in cis position decrease. This is a clear example of
trans vs cis influence.
The retro-rollover process, which is particularly fast with the
ligand L, may be of future interest because of its potential
applications in catalytic cycles, for example, in the dehydrogen-
ation of alcohols,1a with L serving as hydrogen atom reservoir,
or in functionalization reactions.13
It is also a unique example of Pt−C(sp3) vs Pt−C(sp2)
protonolysis, an interesting result in view of the close
relationship between M−C protonolysis reactions and metal-
mediated C−H bond activation, that is, the reverse process.
Notably, for the first time a retro-rollover process is observed
for a “non-symmetric” rollover complex. Furthermore, the final
species, the cationic adduct [Pt(L)(Me)(PPh3)]+, has been
demonstrated to be fluxional in solution.
Finally, for the first time, the X-ray crystal structure of a
protonated rollover species, 5c-BF4, has been determined and
reported.
ASSOCIATED CONTENT
* Supporting Information
■
S
Final Rietveld refinement plots for species 2b, 2c, and 5c-BF4.
DFT calculated equilibrium geometries, energy values, dipole
moments, molecular orbitals and TD-DFT calculated tran-
sitions. This material is available free of charge via the Internet
at http://pubs.acs.org.
AUTHOR INFORMATION
Corresponding Author
*E-mail: zucca@uniss.it.
■
Finally, it is worth noting that [Pt(N,N)(Me)(neutral
ligand)]+ cationic complexes are also of interest in the C−H
bond activation of alkanes,62 even if their synthesis is usually
not so straightforward, involving the use of silver salts and
several steps. On the contrary, our methodology opens a new
and simpler way to obtain this class of complexes, at least when
the rollover reaction is accessible.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
Financial support from Universita
■
̀
di Sassari and Ministero
dell’Universita
̀
e della Ricerca Scientifica is gratefully acknowl-
CONCLUSIONS
■
edged. We thank Johnson Matthey for a generous loan of
platinum salts, and Norberto Masciocchi for definitely fruitful
discussions. L.M. gratefully acknowledges a PhD fund, financed
by POR/FSE 2007-2013, from Regione Autonoma della
Sardegna and the resources given by the Cybersar Project
managed by the “Consorzio COSMOLAB”.
The 2-pyridyl-quinoline ligand, despite of its similarity to 2,2′-
bipyridine, has a more complex behavior because of differences
both in the electronic and in the steric properties. Because of its
specific features, both rollover and retro-rollover processes are
highly favored and accelerated.
Protonation of the uncoordinated nitrogen atom in the
rollover species 2a−c allowed the synthesis of the series of
uncommon species 5a−c. Complexes 2 and 5 are unusual
Brønsted−Lowry acid−base couples, which may be considered
from several points of view. Complexes 5a−c may be described
as mesoionic species, abnormal-remote carbenes, or pyridy-
lenes, and the ligand L* is an isomer, or a tautomer, of the free
ligand L. Furthermore, the L/L* couple is an example of a
“Ligand with multiple personalities”,63 a class of ligands with
additional N−H bonds which is receiving growing interest.
These species have been reported to respond to variations in
the solution environment, such as pH changes, modifying the
properties of the transition metal center and raising interesting
perspectives in C−H bond activation64 and in the design of
molecular devices.65
The nature of the mesoionic species 5a−c and 6 has not
been fully investigated. An interesting feature is the trend of the
NMR data, which show that in this family of complexes the
coupling constant values around the platinum center are
governed by position more than by charge: the coupling
constant values in trans to C(3) increase after protonation,
REFERENCES
■
(1) (a) For example, see Albrecht, M. Chem. Rev. 2010, 110 (2),
576−623. (b) Omae, I. Coord. Chem. Rev. 2004, 248 (11−12), 995−
1023.
(2) Butschke, B.; Schwarz, H. Chem. Sci. 2012, 3, 308−326.
(3) (a) Skapski, A. C.; Sutcliffe, V. F.; Young, G. B. J. Chem. Soc.,
Chem. Commun. 1985, 609−611. (b) Minghetti, G.; Stoccoro, S.;
Cinellu, M. A.; Soro, B.; Zucca, A. Organometallics 2003, 22 (23),
4770−4777.
(4) Butschke, B.; Schwarz, H. Organometallics 2010, 29 (22), 6002−
6011.
(5) Zucca, A.; Petretto, G. L.; Stoccoro, S.; Cinellu, M. A.;
Manassero, M.; Manassero, C.; Minghetti, G. Organometallics 2009,
28 (7), 2150−2159.
(6) Doppiu, A.; Minghetti, G.; Cinellu, M. A.; Stoccoro, S.; Zucca, A.;
Manassero, M. Organometallics 2001, 20 (6), 1148−1152.
(7) (a) Giordano, T. J.; Rasmussen, P. G. Inorg. Chem. 1975, 14 (7),
1628−1634. (b) Nonoyama, M. Bull. Chem. Soc. Jpn. 1979, 52, 3749−
3750. (c) Nonoyama, M.; Kajita, S. Transition Met. Chem. 1981, 6 (3),
163−165. (d) Chassot, L.; Von Zelewsky, A. Inorg. Chem. 1987, 26
(17), 2814−2818. (e) Chang, S. C.; Hauge, R. H.; Billups, W. E.;
M
dx.doi.org/10.1021/ic400908f | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Margrave, J. L.; Kafafi, Z. H. Inorg. Chem. 1988, 27 (1), 205−206.
(f) Gianini, M.; Von Zelewsky, A.; Stoeckli-Evans, H. Inorg. Chem.
1997, 36 (26), 6094−6098. (g) Fuchita, Y.; Ieda, H.; Wada, S.;
Kameda, S.; Mikuriya, M. J. Chem. Soc., Dalton Trans. 1999, 4431−
4435. (h) Lee, K.-E.; Jeon, H.-T.; Han, S.-Y.; Ham, J.; Kim, Y.-J.; Lee,
S. W. Dalton Trans. 2009, 6578−6592. (i) Callear, S. K.; Spencer, J.;
Patel, H.; Deadman, J.; Hursthouse, M. B. J. Chem. Crystallogr. 2011,
41 (4), 523−527.
(c) Romeo, R.; Plutino, M. R.; Elding, L. I. Inorg. Chem. 1997, 36 (25),
5909−5916. (d) Zhong, H. A.; Labinger, J. A.; Bercaw, J. E. J. Am.
Chem. Soc. 2002, 124 (7), 1378−1399. (e) Alibrandi, G.; Minniti, D.;
Romeo, R.; Uguagliati, P.; Calligaro, L.; Belluco, U. Inorg. Chim. Acta
1986, 112 (2), L15−L16.
(31) (a) Peters, R. G.; White, S.; Roddick, D. M. Organometallics
1998, 17 (20), 4493−4499. (b) Zhang, F.; Prokopchuk, E. M.;
Broczkowski, M. E.; Jennings, M. C.; Puddephatt, R. J. Organometallics
2006, 25 (7), 1583−1591. (c) Hickman, A. J.; Villalobos, J. M.;
Sanford, M. S. Organometallics 2009, 28 (18), 5316−5322.
(d) Mazzone, G.; Russo, N.; Sicilia, E. Inorg. Chem. 2011, 50 (20),
10091−10101. (e) Bercaw, J. E.; Chen, G. S.; Labinger, J. A.; Lin, B.-L.
Organometallics 2010, 29 (19), 4354−4359. (f) Guido, E.; D’Amico,
G.; Russo, N.; Sicilia, E.; Rizzato, S.; Albinati, A.; Romeo, A.; Plutino,
M. R.; Romeo, R. Inorg. Chem. 2011, 50 (6), 2224−2239. (g) Kalberer,
E. W.; Houlis, J. F.; Roddick, D. M. Organometallics 2004, 23 (17),
4112−4115. (h) Stahl, S. S.; Labinger, J. A.; Bercaw, J. E. J. Am. Chem.
Soc. 1996, 118 (25), 5961−5976. (i) Heiberg, H.; Johansson, L.;
Gropen, O.; Ryan, O. B.; Swang, O.; Tilset, M. J. Am. Chem. Soc. 2000,
122 (44), 10831−10845. (j) Romeo, R.; D’Amico, G. Organometallics
2006, 25 (14), 3435−3446. (k) Wik, B. J.; Lersch, M.; Tilset, M. J. Am.
Chem. Soc. 2002, 124 (41), 12116−12117.
(8) (a) Canty, A. J.; Honeyman, R. T. J. Organomet. Chem. 1990, 387
(2), 247−263. (b) Canty, A. J.; Minchin, N. J. J. Organomet. Chem.
1982, 226 (1), C14−C16. (c) Canty, A. J.; Minchin, N. J.; Patrick, J.
M.; White, A. H. J. Chem. Soc., Dalton Trans. 1983, 1253−1259.
(9) Zhao, S.-B.; Wang, R.-Y.; Wang, S. J. Am. Chem. Soc. 2007, 129
(11), 3092−3093.
(10) Crosby, S. H.; Clarkson, G. J.; Rourke, J. P. Organometallics
2011, 30 (13), 3603−3609.
(11) Butschke, B.; Schwarz, H. Int. J. Mass Spectrom. 2011, 306 (2−
3), 108−113.
(12) Butschke, B.; Schlangen, M.; Schroder, D.; Schwarz, H. Int. J.
̈
Mass Spectrom. 2009, 283 (1−3), 3−8.
(13) Kwak, J.; Ohk, Y.; Jung, Y.; Chang, S. J. Am. Chem. Soc. 2012,
134 (42), 17778−17788.
(14) Shibata, T.; Takayasu, S.; Yuzawa, S.; Otani, T. Org. Lett. 2012,
14 (19), 5106−5109.
(32) Eaborn, C.; Kundu, K.; Pidcock, A. J. Chem. Soc., Dalton Trans.
1981, 933−938.
(15) Petretto, G. L.; Zucca, A.; Stoccoro, S.; Cinellu, M. A.;
Minghetti, G. J. Organomet. Chem. 2010, 695 (2), 256−259.
(16) (a) Zucca, A.; Petretto, G. L.; Stoccoro, S.; Cinellu, M. A.;
Minghetti, G.; Manassero, M.; Manassero, C.; Male, L.; Albinati, A.
Organometallics 2006, 25 (9), 2253−2265. (b) Petretto, G. L.; Rourke,
J. P.; Maidich, L.; Stoccoro, S.; Cinellu, M. A.; Minghetti, G.; Clarkson,
G. J.; Zucca, A. Organometallics 2012, 31 (8), 2971−2977.
(17) (a) Crabtree, R. H. Coord. Chem. Rev. 2013, 257, 755−766.
(b) Iglesias, M.; Albrecht, M. Dalton Trans. 2010, 39, 5213−5215.
(18) For example, see Schuster, O.; Yang, L.; Raubenheimer, H. G.;
Albrecht, M. Chem. Rev. 2009, 109 (8), 3445−3478.
(33) Romeo, R.; Monsu Scolaro, L. Inorg. Synth. 1998, 32, 153.
̀
(34) Vogel’s Textbook of Practical Organic Chemistry, 5th ed.;
Longman Scientific and Technical Ed.: Harlow, U.K., 1989.
(35) Granovsky, A. Firefly, version 7.1.G; www.http://classic.chem.
msu.su/gran/firefly/index.html.
(36) Schmidt, M. W.; Baldridge, K. K.; Boatz, J. A.; Elbert, S. T.;
Gordon, M. S.; Jensen, J. H.; Koseki, S.; Matsunaga, N.; Nguyen, K. A.;
Su, S.; Windus, T. L.; Dupuis, M.; Montgomery, J. A. J. Comput. Chem.
1993, 14 (11), 1347−1363.
(37) Perdew, J. P.; Burke, K.; Ernzerhof, M. Phys. Rev. Lett. 1996, 77
(18), 3865−3868.
(38) Perdew, J. P.; Burke, K.; Ernzerhof, M. Phys. Rev. Lett. 1996, 78
(7), 1396.
(39) Adamo, C.; Barone, V. J. Chem. Phys. 1999, 110 (13), 6158−
6169.
(40) Weigend, F.; Ahlrichs, R. Phys. Chem. Chem. Phys. 2005, 7 (18),
3297−3305.
(41) Feller, D. J. Comput. Chem. 1996, 17 (13), 1571−1586.
(42) Runge, E.; Gross, E. K. U. Phys. Rev. Lett. 1984, 52 (12), 997−
1000.
(19) Maidich, L.; Zuri, G.; Stoccoro, S.; Cinellu, M. A.; Masia, M.;
Zucca, A. Organometallics 2013, 32 (2), 438−448.
(20) (a) Constable, E. C.; Dunne, S. J.; Rees, D. G. F.; Schmitt, C. X.
Chem. Commun. 1996, 1169−1170. (b) Moorlag, C.; Clot, O.; Wolf,
M. O.; Patrick, B. O. Chem. Commun. 2002, 3028−3029. (c) Moorlag,
C.; Wolf, M. O.; Bohne, C.; Patrick, B. O. J. Am. Chem. Soc. 2005, 127
(17), 6382−6393.
(21) Zuber, M.; Pruchnik, F. P. Polyhedron 2006, 25 (14), 2773−
2777.
(22) For example, see (a) Stoccoro, S.; Chelucci, G.; Cinellu, M. A.;
Zucca, A.; Minghetti, G. J. Organomet. Chem. 1993, 450 (1−2), C15−
C16. (b) Stoccoro, S.; Cinellu, M. A.; Zucca, A.; Minghetti, G.;
Demartin, F. Inorg. Chim. Acta 1994, 215 (1−2), 17−26. (c) Zucca, A.;
Stoccoro, S.; Cinellu, M. A.; Minghetti, G.; Manassero, M.; Sansoni,
M. Eur. J. Inorg. Chem. 2002, 12, 3336−3346.
(43) Gross, E. K. U.; Kohn, W. Phys. Rev. Lett. 1985, 55 (26), 2850−
2852.
(44) (a) Azzena, U.; Chelucci, G.; Delogu, G.; Gladiali, S.; Marchetti,
M.; Soccolini, F.; Botteghi, C. Gazz. Chim. Ital. 1986, 116, 307.
(b) Botteghi, C.; Chelucci, G.; Chessa, G.; Delogu, G.; Gladiali, S.;
Soccolini, F. J. Organomet. Chem. 1986, 304 (1−2), 217−225.
(45) Smith, L. I.; Opie, J. W. Org. Synth., Coll. Vol. 1955, 3, 56; Org.
Synth. 1948, 28, 11.
(46) Remarkably, at least in the case of 2b and 2c, grinding the
sample in an agate mortar provoked partial degradation of the material.
(47) Coelho, A. A. J. Appl. Crystallogr. 2003, 36, 86−95.
(48) TOPAS, Version 3.0; Bruker AXS: Karlsruhe, Germany, 2005.
(23) For example, see (a) Minghetti, G.; Doppiu, A.; Zucca, A.;
Stoccoro, S.; Cinellu, M. A.; Manassero, M.; Sansoni, M. Chem.
Heterocycl. Compd. 1999, 8 (386), 1127−1137. (b) Doppiu, A.;
Minghetti, G.; Cinellu, M. A.; Stoccoro, S.; Zucca, A.; Manassero, M.
Organometallics 2001, 20 (6), 1148−1152.
(24) Harris, C. M.; Kokot, S.; Patil, H. R. H.; Sinn, E.; Wong, H. Aust.
J. Chem. 1972, 25 (8), 1631−1643.
(49) Bonnemann, H. Angew. Chem., Int. Ed. Engl. 1978, 17 (7), 505−
̈
(25) Qaseer, H. A. Croat. Chem. Acta. 2005, 78 (1), 79−84.
(26) Zaghal, M. H.; Qaseer, H. A. Inorg. Chim. Acta 1989, 163 (2),
193−200.
515.
(50) Kaes, C.; Katz, A.; Hosseini, M. W. Chem. Rev. 2000, 100 (10),
3553−3590.
(27) O’Connor, C. J.; Sinn, E. Inorg. Chem. 1978, 17 (8), 2067−
2071.
(28) Hoefelmeyer, J. D.; Schulte, M.; Tschinkl, M.; Gabbai, F. P.
(51) Butschke, B.; Schlangen, M.; Schroder, D.; Schwarz, H. Chem.
̈
Eur. J. 2008, 14 (35), 11050−11060.
(52) See, as an example, (a) Albrecht, M.; Stoeckli-Evans, H. Chem.
Commun. 2005, 4705−4707. (b) Poulain, A.; Neels, A.; Albrecht, M.
Eur. J. Inorg. Chem. 2009, 13, 1871−1881.
Coord. Chem. Rev. 2002, 235, 93−103.
(29) Hosmane, R. S.; Liebman, J. F. Struct. Chem. 2009, 20, 693−697.
(30) (a) Bonnington, K. J.; Zhang, F.; Moustafa, M. M. A. R.;
Cooper, B. F. T.; Jennings, M. C.; Puddephatt, R. J. Organometallics
2011, 31 (1), 306−317. (b) Johansson, L.; Tilset, M.; Labinger, J. A.;
Bercaw, J. E. J. Am. Chem. Soc. 2000, 122 (44), 10846−10855.
(53) (a) Vicente, J.; Arcas, A.; Bautista, D.; Jones, P. J.
Organometallics 1997, 16 (10), 2127−2138. (b) Crespo, M.; Granell,
J.; Solans, X.; Font-Bardia, M. J. Organomet. Chem. 2003, 681 (1−2),
N
dx.doi.org/10.1021/ic400908f | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
143−149. (c) Vicente, J.; Abad, J.-A.; Frankland, A. D.; Ramírez de
Arellano, M. C. Chem.Eur. J. 1999, 5 (10), 3066−3075.
(54) Owen, J. S.; Labinger, J. A.; Bercaw, J. E. J. Am. Chem. Soc. 2004,
126 (26), 8247−8255.
(55) Zucca A. et al., unpublished results.
(56) Pearson, R. G. Inorg. Chem. 1973, 12 (3), 712−713.
(57) Janzen, D. A.; VanDerveer, D. G.; Mehne, L. F.; da Silva Filho,
́
D. A.; Bredas, J.-L.; Grant, G. J. Dalton Trans. 2008, 14, 1872−1882.
(58) Ghedini, M.; Pugliese, T.; La Deda, M.; Godbert, N.; Aiello, I.;
Amati, M.; Belviso, S.; Lelj, F.; Accorsi, G.; Barigelletti, F. Dalton
Trans. 2008, 32, 4303−4318.
(59) Hallett, A. J.; Kariuki, B. M.; Pope, S. J. A. Dalton Trans. 2011,
40, 9474−9481.
(60) Bercaw, J. E.; Chen, G. S.; Labinger, J. A.; Lin, B.-L.
Organometallics 2010, 29 (19), 4354−4359.
(61) (a) Romeo, R.; Fenech, L.; Monsu Scolaro, L.; Albinati, A.;
̀
Macchioni, A.; Zuccaccia, C. Inorg. Chem. 2001, 40 (14), 3293−3302.
(b) Romeo, R.; Fenech, L.; Carnabuci, S.; Plutino, M. R.; Romeo, A.
Inorg. Chem. 2002, 41 (11), 2839−2847. (c) Romeo, R.; Carnabuci, S.;
Plutino, M. R.; Romeo, A.; Rizzato, S.; Albinati, A. Inorg. Chem. 2005,
44 (5), 1248−1262. (d) Romeo, R.; Carnabuci, S.; Fenech, L.; Plutino,
M. R.; Albinati, A. Angew. Chem., Int. Ed. 2006, 45, 4494−4498.
(e) Romeo, R.; D’Amico, G.; Guido, E.; Albinati, A.; Rizzato, S. Inorg.
Chem. 2007, 46 (25), 10681−10692.
(62) For example, see (a) Lersch, M.; Tilset, M. Chem. Rev. 2005,
105, 2471−2526. (b) Johansson, L.; Tilset, M.; Labinger, J. A.; Bercaw,
J. E. J. Am. Chem. Soc. 2000, 122, 10846−10855. (c) Johansson, L.;
Tilset, M. J. Am. Chem. Soc. 2001, 123, 739−740. (d) Johansson, L.;
Ryan, O. B.; Rømming, C.; Tilset, M. J. Am. Chem. Soc. 2001, 123,
6579−6590. (e) Zhong, H. A.; Labinger, J. A.; Bercaw, J. E. J. Am.
Chem. Soc. 2002, 124, 1378−1399.
(63) Crabtree, R. H. Science 2010, 293, 455−456.
(64) Hashiguchi, B. G.; Young, K. J. H.; Yousufuddin, M.; Goddard,
W. A.; Periana, R. A. J. Am. Chem. Soc. 2010, 132 (36), 12542−12545.
(65) Koo, C.-K.; Lam, B.; Leung, S.-K.; Lam, M. H.-W.; Wong, W.-Y.
J. Am. Chem. Soc. 2006, 128 (51), 16434−16435.
O
dx.doi.org/10.1021/ic400908f | Inorg. Chem. XXXX, XXX, XXX−XXX