Titanium(IV) complexes of disulfide-linked schiff bases

Article abstract of DOI:10.1021/ic202709x

With the goal of preparing Ti(IV) complexes bearing a sulfur-based redox function of possible use in electrocatalytic oxidations of alcohols at electrode surfaces, a series of seven 2,2′-dithiodianiline Schiff-base derivatives, including two new variation

Full text of DOI:10.1021/ic202709x

Article
pubs.acs.org/IC
Titanium(IV) Complexes of Disulfide-Linked Schiff Bases
Alberto Donzelli, Imir Metushi, and Pierre G. Potvin*
Department of Chemistry, York University, 4700 Keele Street, Toronto, Ontario, Canada M3J 1P3
S
* Supporting Information
ABSTRACT: With the goal of preparing Ti(IV) complexes bearing a
sulfur-based redox function of possible use in electrocatalytic oxidations of
alcohols at electrode surfaces, a series of seven 2,2′-dithiodianiline Schiff-
base derivatives, including two new variations, were tested in reactions
i
t
with Ti(OR)4 (R = Pr, Bu). Instead of the expected dimetallic products
of general formula [LTi(OR)2]2, mononuclear species LTi(OR)2 were
obtained, confirmed by crystallographic determinations to have an
unprecedented, symmetrical, and macrocyclic arrangement with four-
point binding to the metal center and with the disulfide moieties
remaining uncoordinated. Cyclic voltammetry performed in CH2Cl2
displayed oxidations at potentials useful for fuel cells (+1.1−1.5 V vs
Ag/AgCl), but despite the uncoordinated disulfide moieties, the
complexes were reticent to engage in reduction processes.
oxidation. Disulfide-containing molecules such as DTDA have
been widely investigated because of their easy participation in
reversible redox processes3 and because of the important role
played by the RS-SR/RSH equilibrium in biological systems.4
DTDA has been considered for energy-storage devices,5 with its
oxidation occurring at a readily accessible +0.27 V vs Ag/AgCl,
but the large separation between oxidation and reduction
events indicated that the kinetics involved are relatively slow
and, hence, not suitable for practical storage applications.
DTDA-derived Schiff bases have also been used in coordination
chemistry, in particular as synthons of the anil forms of the 2-
aminothiophenol Schiff bases in complexes of Cu(II),6 Ni(II),7
Zn(II),8 Cd(II),9 Sn(II),10 V(III),11 and Fe(III).12 In direct
reactions, salicylaldehyde derivatives of DTDA could poten-
tially behave as doubly monobasic bis(O,N,S-tridentates) or
bis(O,N-bidentates) and could therefore potentially accom-
modate two metal centers. Only one example of this
coordination mode has been proposed for a di-Ru(II) species,
but it was not well characterized.13 In most cases, they behave
as pentadentates, with a metal−sulfur bond always
present.11−14 Prior to this work, the reactivity of salicylalde-
hyde-DTDA Schiff bases toward Ti(IV) was completely
unknown. With the possibility of isolating dimetallic products,
we investigated the reactions of seven examples (Chart 1) with
two Ti(OR)4 species (R = iPr, tBu) and report the
electrochemical behavior of the products in aprotic media.
INTRODUCTION
■
In pursuit of an efficient electrocatalyst of alcohol oxidation for
possible use in direct methanol fuel cells, we chose to base our
design on a dialkoxytitanium(IV) center carrying a redox-active
multidentate ligand. One alkoxide site would be used for
anchoring to an electrode surface, and the other would enable
rapid fuel loading and product unloading through alcohol−
alkoxide exchanges, while the redox-active ligand would store
up oxidation equivalents. We chose to examine a few families of
sulfur-containing ligands for redox activity, specifically for
oxidations that could be driven by O2 reduction at the cathode
of a fuel cell.
We first explored the coordination chemistry at Ti(IV) of
dithiocarbamates,1 which are known to reversibly oxidize to
thiurams. However, the dithiocarbamates prefentially formed
L3Ti(OR) complexes and the onsets of their oxidations were
too far positive to be driven by O2 reduction (i.e., above +1.24
V vs SHE) and up to 1.2 V more positive than the free
dithiocarbamates themselves. The more π-delocalized Schiff
bases derived from 2-aminothiophenol and various salicylalde-
hydes are dibasic tridentates that would oxidize to quinonoid
forms. They more usefully formed the desired L2Ti(OR)2
species or dimeric forms thereof. Relative to the free ligands,
the complexes’ oxidations were less positively shifted (by 0.5 V
or less) and were even easier in the presence of an alcohol.2
To further reduce the impact of the metal on the onset of
oxidation, we chose to segregate the redox function from the
metal-coordination function. This paper reports our exploration
of the coordination at Ti(IV) of salicylaldehyde-derived Schiff
bases of 2,2′-dithiodianiline (DTDA), with a view toward their
possible utility as two-electron stores in electrocatalytic
Received: December 16, 2011
Published: April 19, 2012
© 2012 American Chemical Society
5138
dx.doi.org/10.1021/ic202709x | Inorg. Chem. 2012, 51, 5138−5145

Inorganic Chemistry
Article
1H), 3.41 (q, 4H), 1.13 (t, 6H) ppm. 13C NMR (DMSO-d6): δ 162.6,
161.6, 151.8, 146.2, 134.5, 129.4, 127.6, 126.2, 125.2, 117.4, 108.4,
104.0, 96.6, 43.9, 12.4 ppm. Mp: 181−183 °C. Anal. Calcd for
C34H38N4O2S2: C, 68.19; H, 6.40; N, 9.36. Found: C, 66.73; H, 6.81;
N, 9.25.
Chart 1
H27.16 1H NMR (DMSO-d6): δ 15.04 (s, 1H), 9.75 (s, 1H), 8.69 (d,
1H), 8.01 (d, 1H), 7.86 (d, 1H), 7.80 (d, 1H), 7.61 (m, 2H), 7.40 (m,
2H), 7.27 (t, 1H), 7.15 (d, 1H) ppm. 13C NMR (DMSO-d6): δ 164.8,
158.7, 145.1, 136.1, 132.6, 129.4, 128.9, 128.6, 128.0, 127.6, 127.4,
127.2, 123.7, 120.8, 119.9, 119.0, 109.4 ppm.
Sample Complexation Procedure: Preparation of (1)Ti-
(OiPr)2. With Ar protection, ligand H21 (0.107 g, 0.23 mmol) was
suspended in 1 mL of anhydrous CHCl3, the suspension was stirred
for 1 min, and subsequently, Ti(OiPr)4 (0.07 mL, 0.23 mmol) was
added. The resultant dark red solution was stirred in a vortex mixer for
a few seconds and then in a sonicator for approximately 10 min at
room temperature. The solvent and reaction byproduct were removed
under reduced pressure, and the product was recovered as a red solid
in quantitative yield (0.142 g, 100%). All other complexes were
prepared on the same scale in quantitative yields. 1H NMR (CDCl3): δ
7.94 (s, 1H), 7.71 (d, 1H), 7.43 (d, 1H), 7.29 (t, 1H), 7.17 (t, 1H),
7.05 (d, 1H), 6.84 (t, 1H), 6.64 (m, 2H), 4.86 (h, 1H), 1.21 (d, 6H)
ppm. 13C NMR (CDCl3): δ 168.6, 164.1, 153.4, 136.7, 134.8, 134.2,
129.1 (2C), 128.0, 125.3, 120.7, 119.1, 116.9, 78.5, 25.3 ppm. Anal.
Calcd for C32H32N2O4S2Ti: C, 61.93; H, 5.20; N, 4.51. Found: C,
61.65; H, 5.34; N, 4.16.
EXPERIMENTAL SECTION
■
General Considerations. All reactions were carried out under Ar.
2,2′-Dithiodianiline (DTDA) was prepared by the literature
procedure.15 All other reagents were Sigma-Aldrich products.
Titanium tetraisopropoxide, Ti(OiPr)4, was distilled under Ar prior
to use. Titanium tetra-tert-butoxide, Ti(OtBu)4, was purchased in
Aldrich Sure-Seal bottles and used directly. Solvents were from
Caledon Laboratories (Georgetown, ON, Canada). Prior to use,
CHCl3 was dried and kept over MgSO4 and CH2Cl2 was distilled over
P2O5. HOiPr was kept over molecular sieves. NMR spectra were
acquired in CDCl3 or DMSO-d6 (Cambridge Isotope Laboratories) at
23 °C on a Bruker ARZ 300 MHz instrument. Signal assignments were
made with the help of COSY, HMQC, and HMBC spectra. Relative
integrations were obtained after careful baseline and phase corrections
on spectra acquired with a relaxation delay of 5 s. Elemental analyses
were performed with weighing under N2 by Guelph Chemical
Laboratories (Guelph, ON, Canada).
Ligand Synthesis. All ligands were prepared according to the
general literature procedure14f by reaction of DTDA with 2 equiv of
the corresponding salicylaldehyde in absolute EtOH. The reaction
mixture was heated to reflux for 3 h under a blanket of Ar. The yellow
solid precipitates were filtered, washed with cold absolute EtOH, and
dried under reduced pressure. The products were recovered in 82−
98% yields and used without further purification.
(1)Ti(OtBu)2. 1H NMR (CDCl3): δ 7.90 (s, 1H), 7.83 (d, 1H), 7.46
(d, 1H), 7.28 (bt, 1H), 7.20 (bt, 1H), 7.04 (d, 1H), 6.87 (bt, 1H), 6.60
(m, 2H), 1.29 (s, 9H) ppm. 13C NMR (CDCl3): δ 168.4, 165.0, 153.8,
136.6, 134.5, 134.1, 129.1, 128.7, 128.4, 125.1, 120.9, 118.8, 116.2,
83.9, 31.5 ppm. Anal. Calcd for C34H36N2O4S2Ti: C, 62.96; H, 5.59;
N, 4.32. Found: C, 63.14; H, 5.33; N, 3.98.
1
(2)Ti(OiPr)2. H NMR (CDCl3): δ 7.90 (s, 1H), 7.76 (d, 1H), 7.40
(d, 1H), 7.08 (bt, 1H), 6.90 (d, 1H), 6.81 (t, 1H), 6.69 (d, 1H), 6.54
(t, 1H), 4.94 (h, 1H), 3.94 (s, 3H), 1.20 (d, 6H) ppm. 13C NMR
(CDCl3): δ 168.5, 156.4, 153.4, 149.0, 136.4, 129.0 (2C), 127.8, 126.7,
125.3, 121.3, 119.3, 119.3, 78.7, 57.5, 25.3 ppm. Anal. Calcd for
C34H36N2O6S2Ti: C, 60.00; H, 5.33; N, 4.12. Found: C, 59.60; H,
5.10; N, 4.14.
(2)Ti(OtBu)2. 1H NMR (CDCl3): δ 7.90 (m, 2H), 7.45 (d, 1H), 7.14
(t, 1H), 6.96 (d, 1H), 6.85 (t, 1H), 6.74 (d, 1H), 6.53 (t, 1H), 4.07 (s,
3H), 1.31 (s, 9H) ppm. 13C NMR (CDCl3): δ 168.4, 157.1, 153.7,
148.8, 136.5, 129.0, 128.8, 128.2, 127.3, 125.2, 121.8, 121.0, 115.8,
84.0, 58.4, 31.6 ppm. Anal. Calcd for C36H40N2O6S2Ti: C, 61.01; H,
5.69; N, 3.95. Found: C, 60.77; H, 5.42; N, 4.23.
H21.14f 1H NMR (DMSO-d6): δ 12.57 (s, 1H), 9.03 (s, 1H), 7.73
(d, 1H), 7.58 (d, 1H), 7.50 (m, 2H), 7.37 (t, 1H), 7.30 (t, 1H), 7.04
(m, 2H) ppm. 13C NMR (DMSO-d6): δ 163.4, 160.0, 145.8, 133.8,
132.6, 130.2, 128.0, 127.8, 126.0, 119.3 (2C), 118.3, 116.6 ppm.
H22.11 1H NMR (DMSO-d6): δ 12.66 (s, 1H), 9.03 (s, 1H), 7.57 (d,
1H), 7.52 (d, 1H), 7.36 (t, 1H), 7.29 (m, 2H), 7.19 (d, 1H), 6.96 (t,
1H), 3.84 (s, 3H) ppm. 13C NMR (DMSO-d6): δ 163.6, 150.1, 147.8,
145.7, 130.2, 128.0, 127.8, 126.0, 123.8, 119.2, 118.8, 118.3, 115.9, 55.8
ppm.
(3)Ti(OtBu)2. 1H NMR (CDCl3): δ 7.77 (d, 1H), 7.74 (s, 1H), 7.44
(d, 1H), 7.19 (t, 1H), 6.89 (m, 2H), 6.19 (d, 1H), 6.02 (s, 1H), 3.82
(s, 3H), 1.29 (s, 9H) ppm. 13C NMR (CDCl3): δ 166.9, 165.3, 154.2,
136.7, 135.2, 129.2, 128.6 (2C), 124.8, 115.4, 105.3, 101.3, 83.8, 55.2,
31.6 ppm; Anal. Calcd for C36H40N2O6S2Ti: 61.01; H, 5.69; N, 3.95.
Found: C, 60.77; H, 5.50; N, 4.02.
1
(4)Ti(OiPr)2. H NMR (CDCl3): δ 7.89 (s, 1H), 7.70 (d, 1H), 7.41
H23. 1H NMR (DMSO-d6): δ 13.07 (s, 1H), 8.93 (s, 1H), 7.60 (d,
1H), 7.54 (d, 1H), 7.48 (d, 1H), 7.34 (t, 1H), 7.26 (t, 1H), 6.61 (d,
1H), 6.55 (s, 1H), 3.83 (s, 3H) ppm. 13C NMR (DMSO-d6): δ 164.0,
162.8, 162.5, 145.8, 134.3, 129.9, 127.9, 127.3, 125.8, 118.1, 113.0,
107.1, 100.7, 55.5 ppm. Mp: 197−199 °C. Anal. Calcd for
C28H24N2O4S2: C, 65.09; H, 4.68; N, 5.42. Found: C, 63.87; H,
4.92; N, 5.41.
(d, 1H), 7.16 (t, 1H), 6.92 (d, 1H), 6.83 (t, 1H), 6.59 (d, 1H), 6.51 (s,
1H), 4.83 (h, 1H), 3.69 (s, 3H), 1.19 (d, 6H) ppm. 13C NMR
(CDCl3): δ 168.2, 159.1, 153.4, 150.5, 136.7, 129.1, 129.0, 128.0,
125.3, 123.3, 119.9 (2C), 115.5, 78.1, 55.7, 25.3 ppm. Anal. Calcd for
C34H36N2O6S2Ti: 60.00; H, 5.33; N, 4.12. Found: C, 60.24; H, 5.15;
N, 4.35.
(4)Ti(OtBu)2. 1H NMR (CDCl3): δ 7.83 (s, 1H), 7.78 (d, 1H), 7.44
(d, 1H), 7.17 (t, 1H), 6.88 (m, 2H), 6.50 (m, 2H), 3.71 (s, 3H), 1.32
(s, 9H) ppm. 13C NMR (CDCl3): δ 168.0, 160.3, 153.9, 150.0, 136.6,
129.0, 128.7, 128.5, 125.1, 123.2, 119.7 (2C), 115.4, 83.6, 55.8, 32.2
ppm. Anal. Calcd for C36H40N2O6S2Ti: 61.01; H, 5.69; N, 3.95.
Found: C, 60.84; H, 5.42; N, 3.70.
H24.14a 1H NMR (DMSO-d6): δ 11.99 (s, 1H), 8.99 (s, 1H), 7.58
(d, 1H), 7.48 (d, 1H), 7.36 (t, 1H), 7.30 (m, 2H), 7.10 (d, 1H), 6.96
(d, 1H), 3.76 (s, 3H) ppm. 13C NMR (DMSO-d6): δ 162.9, 154.2,
151.9, 146.0, 130.3, 128.0, 127.8, 125.9, 121.1, 119.2, 118.2, 117.5,
115.0, 55.4 ppm.
H25.16 1H NMR (DMSO-d6): δ 13.57 (s, 1H), 9.14 (s, 1H), 8.72 (s,
1H), 8.31 (d, 1H), 7.65 (d, 1H), 7.50 (d, 1H), 7.36 (m, 2H), 7.19 (d,
1H) ppm. NMR (DMSO-d6): δ 165.2, 161.2, 145.7, 139.6, 130.5,
128.6, 128.5, 128.3, 127.7, 127.2, 119.2, 118.6, 117.8 ppm.
H26. 1H NMR (DMSO-d6): δ 13.03 (s, 1H), 8.76 (s, 1H), 7.48 (d,
1H), 7.40 (m, 2H), 7.29 (t, 1H), 7.18 (t, 1H), 6.36 (d, 1H), 6.13 (s,
1
(5)Ti(OiPr)2. H NMR (CDCl3): δ 8.16 (d, 1H), 8.10 (s, 1H), 8.02
(s, 1H), 7.59 (d, 1H), 7.48 (d, 1H), 7.23 (t, 1H), 6.92 (t, 1H), 6.67 (d,
1H), 4.86 (h, 1H), 1.20 (2d, 6H) ppm. 13C NMR (CDCl3): δ 168.8,
168.0, 152.3, 138.0, 137.2, 131.2, 129.9, 129.3, 129.0, 127.2, 126.3,
119.8, 119.3, 81.0, 25.3, 25.2 ppm. Anal. Calcd for C32H30N4O8S2Ti:
54.09; H, 4.26; N, 7.88. Found: C, 54.24; H, 3.96; N, 8.07.
5139
dx.doi.org/10.1021/ic202709x | Inorg. Chem. 2012, 51, 5138−5145

Inorganic Chemistry
Article
(5)Ti(OtBu)2. 1H NMR (CDCl3): δ 8.16 (d, 1H), 8.09 (s, 1H), 7.98
(s, 1H), 7.68 (d, 1H), 7.49 (d, 1H), 7.23 (t, 1H), 6.92 (t, 1H), 6.63 (d,
1H), 1.29 (s, 9H) ppm. 13C NMR (CDCl3): δ 169.2, 167.8, 152.4,
137.7, 137.3, 131.2, 129.9, 129.1, 129.0, 127.5, 126.2, 119.7, 119.4,
86.4, 31.3 ppm. Anal. Calcd for C34H34N4O8S2Ti: 55.28; H, 4.64; N,
7.58. Found: C, 54.96; H, 4.48; N, 7.24.
H2113 and H24,14a detailed NMR spectroscopy was missing due
to a reported lack of solubility in deuterated solvents. Indeed,
we found them to be scarcely soluble in CDCl3 but soluble at
room temperature in DMSO-d6 after warming the samples.
Ligand H25 was soluble only in hot DMSO-d6, presumably
because of strong intermolecular H bonding or zwitterion
1
(6)Ti(OiPr)2. H NMR (CDCl3): δ 7.63 (d, 1H), 7.52 (s, 1H), 7.30
1
formation. In all cases, the H NMR spectra confirmed the
(d, 1H), 7.01 (t, 1H), 6.67 (m, 2H), 5.88 (d, 1H), 5.75 (s, 1H), 4.77
(h, 1H), 3.26 (2q, 4H), 1.09 (m, 12H) ppm. 13C NMR (CDCl3): δ
165.6, 165.0, 154.6, 153.1, 136.7, 135.6, 129.2, 128.8, 128.6, 124.1,
111.6, 102.8, 99.0, 76.9, 44.3, 25.5, 25.2, 12.7 ppm. Anal. Calcd for
C40H50N4O4S2Ti: 62.98; H, 6.61; N, 7.34. Found: C, 62.80; H, 6.86;
N, 7.31.
ligands’ 2-fold symmetry, with diagnostic singlets for the
aldimino hydrogens (labeled Hi in Chart 1) in the 8.7−9.8 ppm
range, at positions consistent with the expected electronic
effects of their substituents.
Complex Synthesis. Direct reactions of 1:1 ratios of the
ligands with Ti(OiPr)4 in anhydrous CHCl3 gave instantaneous
color changes, turning the thick, yellow suspensions of the
ligands into clear, dark orange-red solutions. All cases except
that of H27 led to complexes of apparent formulas LTi(OiPr)2
as the only NMR-detectable products. With ligands L2− acting
as bis(bidentates), either surrounding a single metal or bridging
pairs of metals, the S−S moieties were evidently not
participating in the coordination. The analogous reactions
with Ti(OtBu)4 were generally slower, requiring longer reaction
times, but led to the same results, except that neither H26 nor
H27 gave a clear result. Even with H24, the reaction was not
clean, and the desired product was accompanied by one or
more side products. Nevertheless, NMR assignments for
(4)Ti(OtBu)2 were possible by comparisons with the spectra
from (4)Ti(OiPr)2 and through 2D experiments. In any case,
elemental analysis supported the LTi(OR)2 formulation, as was
the case with the other ligands. The combination of H23 with
Ti(OiPr)4 produced an impure product, though the main NMR
signals could readily be assigned to (3)Ti(OiPr)4 by analogy
with the OtBu analogue (see the Supporting Information).
Mass spectral analysis of the complexes (EI, MALDI) has
proven fruitless, showing no molecular ions and too much
fragmentation.
When a 1:2 ratio of ligand to Ti was instead used, 1 equiv of
unreacted Ti(OR)4 always remained, indicating the impossi-
bility of isolating a dimetallic complex and confirming the
preference for 1:1 stoichiometries. The formation of the
complexes produced a general change in the NMR signal
patterns in the aromatic region. The ligand symmetry was
preserved, as the number of ligand signals remained the same.
There was a consistent upfield shift of the Hi signal (by 1.1−1.2
ppm from free ligand positions in DMSO-d6) to fall in the 7.5−
8.0 ppm range (Figure 1) in all cases. This differs from what we
had found for complexes of 2-aminothiophenol-derived Schiff
bases,2 where the thiazoline forms of the starting ligands
underwent ring opening to engage in tridentate coordination
and the corresponding Hi signals understandably shifted upon
X-ray Crystallography. Red crystals of (6)Ti(OiPr)2 were grown
overnight after layering petroleum ether over a CDCl3 solution. Yellow
plates of [(4)TiO]3 were similarly grown over several months from a
solution of (4)Ti(OtBu)2. Crystal structure data collection, structural
analysis, and refinement were carried out by Dr. Alan Lough at the
University of Toronto. Diffraction intensities were collected on a
Bruker-Nonius Kappa CCD instrument using a fine-focus sealed-tube
Mo Kα source and graphite monochromator. Unique reflections were
corrected for absorption (Denzo-SMN) and used in all calculations.
Heavy-atom positions were determined by direct methods (SHELXS-
97) and refined anisotropically. The hydrogen atoms were assigned
idealized positions according to a riding model and refined
isotropically. Structure refinement used full-matrix least squares on
F2 (SHELXL-97). The C(CH3)2 groups of (6)Ti(OiPr)2 were
disordered (52%:48%) over two sets of correlated positions, both
refined anisotropically. During the refinement of [(4)TiO]3, electron
density peaks corresponding to approximately 47 electrons were
located within the 318 Å3 of void volume within the unit cell
(PLATON), believed to be due to highly disordered solvent molecules
(possibly chloroform) on a 3 axis. Attempts to model these were not
̅
successful. Following the literature treatments of disordered solvent
molecules,17 this contribution to the electron density was removed
from the observed data in the final cycles of refinement. The reported
density, F(000) value, molecular weight, and formula are given without
taking into account the results using the SQUEEZE option of
PLATON.18
Electrochemistry. All electrochemical analyses were performed in
a one-compartment, three-electrode cell with a Pt-disk working
electrode, a Ag/AgCl pseudo reference electrode and a graphite
counter electrode, under an Ar atmosphere, connected to an Obbligato
Objectives Faraday MP potentiostat. The sample solutions were 0.05
M in freshly distilled CH2Cl2 (10 mL) containing 0.1 M nBu4NPF6 as
supporting electrolyte. All scans were performed at a 100 mV s−1
sweep rate. They were limited to either oxidation or reduction regimes
or covered both in succession and were then initiated in both anodic
and cathodic directions. Selected examples (2,2′-dithioaniline, H21,
H22, and H24) were also tested in protic solvent (20% water in
CH3CN), keeping all concentrations unchanged. The potentials were
internally calibrated with ferrocene and are reported with respect to
Ag/AgCl (saturated KCl).
RESULTS AND DISCUSSION
■
Ligand Synthesis. The new examples of DTDA-derived
Schiff bases, H23 and H26, were prepared in high yields by
condensations of DTDA with 2 equiv of the appropriate,
commercially available salicylaldehydes, and their H and 13C
1
NMR spectra were consistent with their formulations.
Combustion analyses showed lower than expected carbon
contents, but these ligands nevertheless produced complexes
that analyzed as pure (vide infra). All of the other ligands were
previously reported but had been characterized mainly by IR
spectroscopy, elemental analysis,11,14f,16 and X-ray crystallogra-
phy.11,14e,19 For comparisons with their complexes, 1H and 13
NMR spectra were acquired and fully assigned for the known
ligands as well (Supporting Information). With the exception of
C
1
Figure 1. Aromatic regions of the H NMR spectra in CDCl3 of (A)
ligand H21 and (B) (1)Ti(OiPr)2.
5140
dx.doi.org/10.1021/ic202709x | Inorg. Chem. 2012, 51, 5138−5145

Inorganic Chemistry
Article
complexation, but to positions further downfield into the 8.6−
8.8 ppm range. The Hi magnetic environment here is consistent
with bidentate ligation and the nonparticipation of the S atoms
in coordination in the present series, since the aminothiophenol
moiety needs to be more or less coplanar with the salicylidene
portion to enable S coordination in the previous series, but
must become noncoplanar for bidentate coordination in the
present cases.
produced slowly evolving mixtures of multinuclear or even
polymeric products, where the requisite exchanges were
particularly slow, mixtures of coordination isomers, products
involving S coordination, or a combination of these.
X-ray Crystallography. Diffraction-quality crystals were
obtained with (6)Ti(OiPr)2 (Figure 2) and [(4)TiO]3 (Figure
i
Pairs of Pr 13CH3 signals were detected for three of the
LTi(OiPr)2 complexes, indicating diastereotopicity and a chiral
metal center given by a cis disposition of two salicylidene units
in, moreover, a C2-symmetric fashion that preserves the ligands’
internal symmetry. Given the C2 symmetry and the spectral
similarities between these three cases and the others, with no
reason to anticipate different behaviors, this arrangement is very
probably common to all five examples observed. Furthermore,
the close spectral similarities displayed by the 5 LTi(OtBu)2
complexes and their OiPr analogues suggests that all 10
complexes have identical ligand arrangements. NMR cannot a
priori distinguish between 1:1 and higher-order, closed n:n
assemblies that two independent bidentate sites make possible.
However, entropic considerations favor a 1:1 assembly. This
requires the ligands to loop back onto the same metal, and the
preferred conformation about the S−S bond in disulfides, with
an approximate 90° dihedral angle,20 makes this possible, as
confirmed by X-ray diffraction (vide infra). Although the earlier
literature reports two analogous but dinuclear complexes with
Cu(II)14a and Mn(II),14d these were merely dimeric, looped-
back 1:1 complexes, linked through phenoxy bridges.
We used several differently substituted salicylaldehydes to
gauge substituent effects. The substituents will influence not
only the ligand oxidizability, as revealed by electrochemistry
(see below), but also the ease with which ligated alkoxides will
become oxidized. The effects of the salicylaldehyde substituents
on the acidities of Ti−O−C−H groups can be gauged by the
TiOC1H positions in the OiPr complexes and the TiO13C
chemical shifts in the OiPr and OtBu complexes, relative to
their positions in the unsubstituted complexes of 12−. The
signals for all three kinds of signals spanned limited ranges, but
they were nevertheless consistent with the anticipated inductive
and resonance effects. Not surprisingly, the 4-NEt2 group in 62−
exerted the strongest upfield shifts, and the 5-NO2 group in 52−
exerted strong downfield shifts, but the 3-OMe group in 22−
was more interesting in that this π-donating group also caused
the strongest deshielding on the TiOC1H signal and also
deshielded the TiO13C signals, an effect that probably has a
steric origin.
In those cases where no clean LTi(OR)2 product was
obtained (L = 62− with R = tBu and L = 72− with R = iPr, tBu),
NMR showed only broad signals that did not resolve over time,
yet we see no evident source of product destabilization within
these ligands. In reactions with monobasic bidentates, such as
Hacac, the second substitution step is usually faster than the
first, owing to a relief of the steric congestion about Ti(IV), and
such ligands tend to quickly form hexacoordinate L2Ti(OR)2
complexes even at 1:1 metal−ligand ratios. In the present
heterogeneous reactions between insoluble free ligands and
soluble Ti(OR)4, however, where the metal was present in
relative excess over dissolved ligand throughout, the process
may have been biased toward multinuclear intermediates that
would need to, then, in a second phase, disproportionate to 1:1
products through exchange reactions involving liberated
alcohol. The three unclear cases encountered here may have
Figure 2. ORTEP plot of the crystal structure of (6)Ti(OiPr)2, with H
atoms and disordered alkoxy positions omitted for clarity.
3). Tables 1 and 2 report the crystallographic data and selected
bond metrics. The latter crystal was obtained from a solution of
(4)Ti(OtBu)2 and evidently arose by preferential crystallization
after adventitious hydrolysis, occurring over the months of
crystal growth, and trimerization of a complex analogous to
(6)Ti(OiPr)2. Unfortunately, we were unsuccessful in deliber-
ately preparing [(4)TiO]3. The unit cell of (6)Ti(OiPr)2 was
composed of a pair of molecules of one enantiomeric form next
to a pair of the opposite enantiomers. One of the diethylamino
groups is coplanar with the attached benzene rings, with both
ethyl groups falling to the same side. The other is twisted out of
coplanarity, with an ethyl group falling on each side of the
nitrogen plane. The difference is no doubt packing-induced, but
as a result, the ligand halves and the two OiPr groups are not
quite crystallographically equivalent. Nevertheless, this struc-
ture completely supported the NMR analysis. Both alkoxy
groups of (6)Ti(OiPr)2 were found to occupy two disordered
positions to nearly equal extents. The Ti−O bond lengths were
virtually the same and were in line with those found in similar
Schiff base and dithiocarbamate complexes of Ti(IV)
alkoxides,1,2 and as expected, they were shorter than the
5141
dx.doi.org/10.1021/ic202709x | Inorg. Chem. 2012, 51, 5138−5145

Inorganic Chemistry
Article
Table 1. Crystallographic Data for (6)Ti(OiPr)2 and
a
[(4)TiO]3
(6)Ti(OiPr)2
[(4)TiO]3
formula
Mr
C40H50N4O4S2Ti
762.86
C84H66N6O15S6Ti3
1735.49
space group
a (Å)
P21/c
R3
̅
16.3908(7)
11.9907(4)
20.1174(8)
90
21.2378(3)
21.2378(3)
31.7593(7)
90
b (Å)
c (Å)
α (deg)
β (deg)
γ (deg)
V (Å3)
Z
91.946(2)
90
90
120
3951.5(3)
4
12 405.7(4)
6
Dcalcd (g cm−3
)
1.282
1.394
λ(Mo Kα) (mm−1
)
0.367
0.502
b
R(Fo)
0.0635
0.0543
b
c
d
2
Rw(Fo )
0.1547
0.1541
a
In both cases, T = 150(1) K and λ = 0.710 73 Å. Esd’s are expressed
b
as uncertainties in the least significant digits in brackets. For
c
reflections where I > 2σ(I). The weights w used in the calculation of
2
2
Rw(Fo ) are given by w = 1/[σ2(Fo ) + (0.0831P)2 + 6.523P], where P
d
= (Fo + 2Fc2)/3. The weights are given by w = 1/[σ2(Fo ) +
2
2
(0.0751P)2], where P = (Fo + 2Fc2)/3.
2
relatively hard Ti(IV) centers is perhaps not surprising but
contrasts with most of the DTDA-derived Schiff base
complexes previously reported.14c−f,21 On the assumption that
the coordination sphere of (4)Ti(OtBu)2 was preserved in
Table 2. Selected Bond Lengths (in Å) and Angles (in deg),
with Uncertainties in the Least Significant Digits in
a
Parentheses
(6)Ti(OiPr)2
Ti1−O1
Ti1−O2
Ti1−O3
Ti1−O4
Ti1−N1
Ti1−N2
1.911(2)
1.911(2)
1.820(3)
1.811(3)
2.260(3)
2.282(3)
O3−C35/38
O4−C35A/C38A
S1−C13
S1−S2
S2−C14
1.435(5)
1.434(5)
1.777(4)
2.059(2)
1.788(4)
Ti1−O2−C26
Ti1−O3−C35
Ti1−O3−C35A
Ti1−O4−C38
Ti1−O4−C38A
140.5(2)
O1−Ti1−O2
O2−Ti1−O3
O3−Ti1−N1
S1−S2−C14
S2−S1−C13
166.8(1)
95.8(1)
133.1(6)
140.0(5)
131.5(5)
142.4(6)
167.1(1)
105.7(1)
105.8(1)
[(4)TiO]3
Ti1−O1
Ti1−O2
Ti1−O3
Ti1−O3*
1.915(2)
1.909(2)
1.817(2)
1.807(2)
2.279(3)
161.5(1)
103.1(1)
96.3(1)
95.0(1)
96.9(1)
94.7(1)
84.7(1)
81.8(1)
86.5(1)
Ti1−N2
S2−C6
S2−S3
2.300(3)
1.780(4)
2.064(2)
1.776(4)
Figure 3. ORTEP plots of the crystal structure of [(4)TiO]3, with H
atoms omitted for clarity: (top) view of the propeller-shaped trimer
down the C3 axis; (bottom) side view of one metallic center.
S3−C15
Ti1−N1
O1−Ti1−O2
O3−Ti1−O3*
O2−Ti1−O3
O2−Ti1−O3*
O1−Ti1−O3
O1−Ti1−O3*
N1−Ti1−N2
N1−Ti1−O1
N1−Ti1−O2
N1−Ti1−O3
N1−Ti1−O3*
N2−Ti1−O1
N2−Ti1−O2
N2−Ti1−O3
N2−Ti1−O3*
C6−S2−S3
84.8(1)
171.7(1)
84.2(1)
average phenolic O−Ti bond length. The Ti−O−CCH3 angles
fell in the 132−142° range and, along with the short Ti−O
bonds, are evidence of modest π donation. The unit cell of
[(4)TiO]3 also includes both enantiomers. Each of those
molecules contains crystallographically equivalent metal centers
but inequivalent ligand halves and inequivalent Ti−oxo bonds.
Both structures confirm the four-point binding deduced by
NMR, with no involvement of sulfur and with the looped ligand
conformation. The absence of coordination by neutral sulfur at
80.6(1)
169.2(1)
87.6(1)
104.6(1)
105.9(1)
C15−S3−S2
a
The labels marked with an asterisk refer to symmetry equivalents.
5142
dx.doi.org/10.1021/ic202709x | Inorg. Chem. 2012, 51, 5138−5145

Inorganic Chemistry
Article
forming [(4)TiO]3, the same was therefore true with
(4)Ti(OtBu)2. One previously reported example, a V(III)
complex of a symmetrical disulfide,11 similarly lacked S
coordination. On the other hand, the two nitrogen atoms in
the V(III) case were inequivalent, whereas the structures of
both of our crystalline Ti(IV) complexes are consistent with the
from partial hydrolyses of complexes with salicylidenimine,23
cyclopentadiene,26 or tripodal tertiary amine ligands.27 Within
the annular core of [(4)TiO]3, the almost identical Ti−O3 and
Ti−O3* bond lengths and the Ti1−O3−Ti2−O3* angle of 5.39°
account for the unusually planar six-membered ring, while all
the other reported Ti3(μ-O)3 compounds showed a chairlike
configuration and inequivalence between the three units.
Because the Ti−O bond lengths are consistent, on average,
across all the structures reported, the electronic densities and
the steric demands of the different ligands are not apparently
affecting the Ti3(μ-O)3 core.
1
single sets of H and 13C NMR signals detected, in which the
internal ligand symmetry was preserved in the complexes. The
exception was a Ti(IV) complex of ligand H48 which we
reported earlier,2 a 2:2 assembly of a bis(dibasic tridentate) that
arose by postcomplexation modifications.
Electrochemistry. All of the isolated complexes were tested
by standard cyclic voltammetry (CV) in comparisons with the
corresponding free ligands. All scans were obtained at the same
concentrations and scan rates under identical conditions, but
the free ligands were examined in two different solvents. The
redox processes were all essentially irreversible, but barring
small differences in kinetics that one would expect within a
family of complexes, the steady-state peak potentials (Table 3)
Table 3. Peak Anodic (Epa) and Cathodic (Epc) Potentials (in
Monobasic bidentate ligands such as acac− normally form 2:1
ligand−Ti complexes, adopting cis arrangements with chiral
metal centers.22 The same arrangement has been reported for
N-phenylsalicylidenimine23 with, moreover, cis nitrogen atoms.
This arrangement is retained in the trinuclear product of its
partial hydrolysis, analogous to the case for [(4)TiO]3, and is
replicated with the present ligands. Planar (equatorial or
meridional) arrangements are more usual with salen-type
tetradentate ONNO ligands24 (except with bidentate auxiliary
ligands and bridging ligands),25 but the present ligands are
twisted out of coplanarity, resulting in cis coordination of the
alkoxide ligands. The looped conformations of the ligands in
(6)Ti(OiPr)2 and [(4)TiO]3 evidently allowed for entropically
favored 1:1 complexation in preference over bridged species
but necessitated somewhat strained S−S linkages. In compar-
ison to the average S−S bond length reported for similar
V vs Ag/AgCl) for H2L and LTi(OR)2 Complexes in CH2Cl2
a
n
(0.1 M in Bu4NPF6)
Epa
Ligands
Epc
a
DTDA
H21
H22
H23
H24
H25
H26
H27
(+0.77)
(nd)
+1.79, +2.03
+1.37, +1.62
+1.40, +1.61
nd
−1.57, −1.87
nd
−1.71, −2.11
nd
nd
−1.25, −1.72
−1.92, −2.48
−1.21, −1.65
+1.22, +1.90
+1.61
LTi(OiPr)2 Complexes
H21
H22
H23
H24
H25
H26
nd
nd
+1.52
−1.68
ca. −1.83
nd
+1.54, +1.94
ligands in their free states (2.02
0.03 Å),11,18,27 the S−S
+1.31
bonds in the present complexes were somewhat elongated
(2.059 Å in (6)Ti(OiPr)2 and 2.064 Å in [(4)TiO]3). While
these fall in line with the values measured in those metal
complexes that do include sulfur coordination, one exception
being the Fe(III) complex,12 they were nevertheless noticeably
shorter than in the comparable V(III) species,11 which also
lacked S coordination but may have been strained, and in
[(8)Ti]2 (2.086 Å on average),2 in which the S−S linkages were
perhaps more relaxed but which were also engaged in H
bonding. The C−S−S−C torsion angles (106.9° in (6)Ti-
(OiPr)2 and 109.5° in [(4)TiO]3) are among the largest
reported. In [(8)Ti]2,2 these angles were significantly smaller
(77.6° on average) and more acute than is ideal but are more in
line with those found in the literature. The torsional distortions
from coplanarity between the two aromatic rings in each
bidentate half of (6)Ti(OiPr)2 (averaging about 67°) were
stronger than those previously found in the analogous ONS
ligands,2 and that distortion appears to be required for
coordination of the bidentate halves to the same metal. In
[(4)TiO]3, the average angle was about 79°. These torsions
move the aldimino hydrogens out of the deshielding regions of
the thiophenol rings and account for the upfield shifts of the Hi
NMR signals upon complexation.
nd
−1.28, −2.09
nd
+1.13, +1.25
LTi(OtBu)2 Complexes
H21
H22
H23
H24
nd
−1.72, −2.38
−1.74
−1.84
+1.33
+1.31
+1.26
nd
H25
+1.39
−1.26
a
a
nd = no clear wave detected. Values measured in 1:4 H2O−CH3CN
are given in parentheses.
are nevertheless useful points of comparison. Figure 4 presents
representative steady-state traces in the positive direction, all
involving the new ligand H23, but scans were also initiated in
the negative direction with the same result. Unfortunately, few
species showed activity in both oxidation and reduction regimes
and some showed none at all, allowing for few useful
comparisons.
The oxidizable LTi(OR)2 complexes showed one or two
distinct and irreversible processes when scanning in the positive
direction, as exemplified by Figure 4b,c. These had been
previously identified14a as being due to oxidations of the
phenolic moiety (process I) and of the sulfur atoms (process
II), and those processes seem to coalesce in the case of the free
Several complexes showing a Ti3(μ-O)3 core have been
previously reported and, as in the case of [(4)TiO]3, resulted
5143
dx.doi.org/10.1021/ic202709x | Inorg. Chem. 2012, 51, 5138−5145

Inorganic Chemistry
Article
instead due to reductions of the nitro group, as had been the
case with the nitro-substituted salicylaldehyde-2-aminothiophe-
nol Schiff base.2
CONCLUSIONS
■
Seven DTDA-derived Schiff-base ligands, including two new
variations, formed ten mononuclear Ti(IV) complexes of novel
architecture, while three other combinations were ill-defined
and perhaps oligomeric. As revealed in two hitherto-
unprecedented crystal structures, the mononuclear species
feature a C2-symmetrical cis ligation of the two monobasic
bidentate portions that take advantage of the preference for
right-angle conformations about the noncoordinated S−S
linkages. These mononuclear complexes then uniquely present
a redox function independent of the coordination function.
However, the products were less redox active in aprotic solvent
than anticipated, showing difficult reductions with no
detectable reoxidation waves. The diethylamino-substituted
complex showed a potentially useful oxidation at +1.13 V vs
Ag/AgCl. We will reinvestigate their electrochemical behavior
after chemisorption on electrode surfaces.
Figure 4. Full cyclic voltammometric scans at 100 mV s−1 of 0.04 M
solutions in CH2Cl2 containing 0.1 M nBu4NPF6 of (a) H23, (b)
(3)Ti(OiPr)2, and (c) (3)Ti(OtBu)2.
ASSOCIATED CONTENT
* Supporting Information
■
S
ligand (Figure 4a). In the case of H26 and its complex, the
lower-potential oxidation (Table 3) was no doubt due to an
easy oxidation of the anilino nitrogen.
CIF files giving crystallographic data and text and tables giving
NMR signal assignments and coupling constants for all species
and 1H and 13C NMR spectra for H23, (3)Ti(OiPr)2, and H26.
This material is available free of charge via the Internet at
http://pubs.acs.org.
Our earlier work with oxidizable ligands had revealed a
pronounced retardation of the ligand oxidations once attached
to Ti(IV).1,2 Setting aside the anilino oxidations of H26 and its
complex, the remaining available comparisons with LTi(OiPr)2
in this series show only a small retardation of the oxidation,
while the two comparable LTi(OtBu)2 complexes oxidized at
slightly lower potentials than the free ligands. It is therefore
unlikely that process I involves the Schiff base itself. Instead, the
ordering is perhaps reversed from that reported earlier,14a with
process I associated with sulfur oxidation and process II with a
retarded Schiff-base oxidation. The difference in oxidation
behavior between LTi(OtBu)2 and LTi(OiPr)2 species has been
noted before,1,2 owing to the greater electron-donating effect of
OtBu groups, but the only two comparable cases here show
larger differences. A steric, rather than electronic, effect may be
responsible for the easier oxidation at sulfur.
Several of the free ligands were also tested in 20:80 H2O−
CH3CN, whence the first oxidations were detected at potentials
up to 0.8 V less positive than in CH2Cl2, which is consistent
with the formation of charged species upon oxidation.
Unfortunately, water sensitivity precluded similar measure-
ments with the complexes. When they were scanned in the
negative direction, H23 (Figure 4a) and most of the other free
ligands (Table 3) showed a first irreversible reduction (process
IV), associated with the cleavage of the disulfide bond, followed
by a more reversible reduction (processes V/V′), reportedly
that of the aldimino functionality.14a Process IV′ (Figure 4a),
not detected during the first positive scan, was evidently
associated with the re-formation of the disulfide bond.
However, half of the complexes did not show such waves.
The few available comparisons reveal more negative waves with
the complexes than with the free ligands, with no reoxidation
waves, but we cannot ascertain whether these are due to
electrostatically retarded S−S bond reductions, to aldimine
reductions, or to TiIV/III couples. In the case of H25 and its
complexes, however, the lower-potential waves were probably
AUTHOR INFORMATION
Corresponding Author
*E-mail: pgpotvin@yorku.ca.
■
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
We thank the Natural Sciences and Engineering Research
Council (Canada) for funding and Imperial Oil for a University
Research Grant.
REFERENCES
■
(1) Donzelli, A.; Potvin, P. G. Inorg. Chem. 2009, 48, 3171−8.
(2) Donzelli, A.; Potvin, P. G. Eur. J. Inorg. Chem. 2012, 741−750.
(3) Cremlyn, R. J. An Introduction to Organosulphur Chemistry; Wiley:
New York, 1996; p 60.
(4) Bertini, I.; Gray, H. B.; Stiefel, E. I.; Valentine, J. S. Biological
Inorganic Chemistry; University Science Books: Mill Valley, CA, 2007;
pp 508−17.
(5) Cotarelo, M. A.; Huerta, F.; Quijada, C.; Per
́
ez, J. M.; del Valle,
M. A.; Vaz
́
quez, J. L. J. Electrochem. Soc. 2006, 153, A2071−A2076.
(6) Bastida, R.; Bermejo, M. R.; Louro, M. S; J. Romero, J.; Sousa, A.;
Fenton, D. E. Inorg. Chim. Acta 1988, 145, 167−169.
(7) Castro, J.; Romero, J.; Garcia-Vaz
́
quez, J. A.; Duran, M. L.;
Castineiras, A.; Sousa, A.; Fenton, D. E. J. Chem. Soc., Dalton Trans.
̃
1990, 3255−3258.
́
(8) Labisbal, E.; Garcia-Vazquez, J. A.; Gomez, C.; Marcias, A.;
Romero, J.; Sousa, A.; Englert, U.; Fenton, D. E. Inorg. Chim. Acta
1993, 203, 67−72.
́
(9) Labisbal, E.; Romero, J.; Garcia-Vazquez, J. A.; Gomez, C.; Sousa,
A.; Pritchard, R.; McAuliffe, C. A. Polyhedron 1994, 13, 1735−1740.
(10) Labisbal, E.; De Blas, A.; Garcia-Vazquez, J. A.; Romero, J.;
́
Duran, M. L.; Sousa, A.; Bailey, N. A.; Fenton, D. E.; Leeson, P. B.
Polyhedron 1992, 11, 227−233.
5144
dx.doi.org/10.1021/ic202709x | Inorg. Chem. 2012, 51, 5138−5145

Inorganic Chemistry
Article
(11) Wang, D.; Behrens, A.; Farahbakhsh, M.; Gatjens, J.; Rehder, D.
Chem. Eur. J. 2003, 9, 1805−1813.
(12) Pyrz, J. W.; Pan, X.; Britton, D.; Que, L., Jr. Inorg. Chem. 1991,
30, 3461−3464.
(13) Bhowon, M. G.; Jhaumeer-Laulloo, S.; Dowlut, M.; Curpen, S.;
Jumnoodoo, V. Transition Met. Chem. 2005, 30, 35−39.
(14) (a) Gili, P.; Martin-Reyes, M. G.; Martin-Zarza, P.; Machado, I.
L. F.; Guedes da Silva, M. F. C.; Lemos, M. A. N. D. A.; Pombeiro, A.
J. L. Inorg. Chim. Acta 1996, 244, 25−36. (b) Uma, R.; Palaniandavar,
M. Transition Met. Chem. 1993, 18, 629−634. (c) Bertrand, J. A.;
Breece, J. L. Inorg. Chim. Acta 1974, 8, 267−272. (d) Kessissoglou, D.
P.; Butler, W. M.; Pecoraro, V. L. Inorg. Chem. 1987, 26, 495−503.
(e) Roy, N.; Sproules, S.; Weyhermuller, T.; Wieghardt, K. Inorg.
Chem. 2009, 48, 3783−3791. (f) Manzur, C.; Bustos, C.; Schrebler, R.;
Carrillo, D.; Knobler, C. B.; Gouzerh, P.; Jeannin, Y. Polyhedron 1989,
8, 2321−2330.
(15) Sathe, M.; Ghorpade, R.; Kaushik, M. P. Chem. Lett. 2006, 35,
1048−1049.
(16) Prabhu, P. M.; Mehta, B. H. Asian J. Chem. 1995, 7, 551−555.
Mehta, B. H.; Prabhu, P. M. Orient. J. Chem. 1997, 13, 297−299.
(17) Stahler, R.; Nather, C.; Bensch, W. Acta Crystallogr. 2001, C57,
̈
̈
26−27. Cox, J. P.; Kumarasammy, Y.; Nahar, L.; Sarkar, D. S.; Shoeb,
M. Acta Crystallogr. 2003, E59, o975−o977. Mohamed, A. A.; Krause
Bauer, A. J.; Bruce, E. A.; Bruce, M. R. M. Acta Crystallogr. 2003, C59,
m84−m86. Athimoolam, S.; Kumar, J.; Ramakrishnan, V.; Rajaram, R.
K. Acta Crystallogr. 2005, E61, m2014−m2017.
(18) Spek, A. L. J. Appl. Crystallogr. 2003, 36, 7−13.
(19) Ruiz-Perez, J.; Gonzalez-Platas, J.; Rodriguez-Romero, F. V.;
Gili, P.; Martin Zarza, P.; Martin Reyes, M. G. Acta Crystallogr. 1995,
C51, 1016−1018.
(20) Lee, J. D. Naturwissenschaften 1972, 59, 36.
(21) Elmali, A.; Elermana, Y.; Svoboda, I. Acta Crystallogr. 2001, C57,
375−376.
(22) Potvin, P. G.; Fieldhouse, B. Can. J. Chem. 1995, 73, 401−413.
(23) Sharma, N.; Sharma, V.; Bohra, R.; Raju, V. S.; Lorenz, I.-P.;
Krinninger, C; Mayer, P. Inorg. Chim. Acta 2007, 360, 3002−3012.
(24) For a recent example, see: Tzubery, A.; Tshuva, E. Y. Inorg.
Chem. 2012, 51, 1796−1804.
(25) Tsuchimoto, M. Bull. Chem. Soc. Jpn. 2001, 74, 2101−2105.
Carroll, K. M.; Schwartz, J.; Ho, D. M. Inorg. Chem. 1994, 33, 2707−
2708.
(26) Blanco, S. G.; Gomez-Sal, M. P.; Carreras, M. P.; Mena, M.;
Royo, P.; Serrano, R. J. Chem. Soc., Chem. Commun. 1986, 1972−1973.
Babcock, L. M.; Klemperer, W. G. Inorg. Chem. 1989, 28, 2003−2007.
Troyanov, S. I. J. Organomet. Chem. 1991, 402, 201−207. Carofiglio,
T.; Floriani, T.; Sgamellotti, A.; Rosi, M.; Chiesi-Villa, A.; Rizzoli, C. J.
Chem. Soc., Dalton Trans. 1992, 1081−1087. Amor, I.; Cuenca, I.;
Galakhov, M; Gomez-Sal, M. P.; Manzanero, M; Royo, P. J.
Organomet. Chem. 1997, 535, 155−168.
(27) Okamatsu, T.; Irie, R.; Katsuki, T. J. Organomet. Chem. 2007,
692, 645−653.
5145
dx.doi.org/10.1021/ic202709x | Inorg. Chem. 2012, 51, 5138−5145