A novel series of CoIII(salen-type) complexes containing a seven-membered metallacycle: Synthesis, structural characterization and factors affecting the metallacyclization rate

Article abstract of DOI:10.1021/om401038v

A series of electronically tuned trans-[Co^III(chel)(CH ₂Cl)]₂ complexes, where chel is a salen derivative (salen = 2,2′-ethylenebis(nitrilomethylidene)diphenol) containing either two or four methyl substituents in differen

Full text of DOI:10.1021/om401038v

Article
pubs.acs.org/Organometallics
A Novel Series of CoIII(salen-type) Complexes Containing a Seven-
Membered Metallacycle: Synthesis, Structural Characterization and
Factors Affecting the Metallacyclization Rate
Patrizia Siega,† Renata Dreos,*,† Giovanna Brancatelli,† Ennio Zangrando,† Claudio Tavagnacco,†
Visnja Vrdoljak,‡ and Tomica Hrenar‡
̌
†Department of Chemical and Pharmaceutical Sciences, University of Trieste, Via L. Giorgieri 1, 34127 Trieste, Italy
‡Department of Chemistry, Faculty of Science, University of Zagreb, Horvatovac 102a,10000 Zagreb, Croatia
S
* Supporting Information
ABSTRACT: A series of electronically tuned trans-[CoIII(chel)(CH2Cl)]2 com-
plexes, where chel is a salen derivative (salen = 2,2′-ethylenebis(nitrilomethylidene)-
diphenol) containing either two or four methyl substituents in different positions,
has been synthesized and characterized, both in solution and in the solid state. These
complexes undergo an intramolecular cyclization reaction in methanolic solution to
form the corresponding cis β organometallic derivative containing a seven-
membered metallacycle, by replacement of the Cl atom of the axial CH2Cl by the salen phenolate oxygen. The cyclization
rate increases on going from two to four methyl substituents in the chelate, in agreement with the electrochemical data that
evidence a general shift toward more negative values with an increase in the number of methyl substituents. The cyclization rate
is also affected by the substituent position, and both electrochemical and kinetic data evidence a remarkable influence of the
methyls on the −CN− groups of the chelate. The X-ray structures of cyclized complexes, [CoIII(chelCH2)(py)(H2O)]+, show
a dependence of the conformation of the seven-membered metallacycle on the different positions of substituents in the chelate.
In fact, in the complex having methyls on the −CN− groups, the conformation is characterized by having the methylene
carbon atom significantly displaced (ca. 1.26 Å) from the aromatic ring plane, whereas in the complex lacking methyl groups in
those positions, the atoms of the Ph−O−CH2 fragment are coplanar. The standard Gibbs energies obtained by quantum
chemical calculation reveal that the different conformations observed in the solid state are mainly the result of the energetically
unfavorable setup of the methyls on the −CN− groups and of the energetically favorable displacement of the CH2 group out
1
of plane of the aromatic ring. H NMR data suggest that the different conformations of the metallacycle are, at least partially,
retained in solution.
The interest in cis β metal Schiff base complexes arises from
the consideration that they proved to be efficient asymmetric
catalysts, being chiral and having two labile mutually cis
coordination sites.1d,2 Furthermore, as trans and cis β
configurations of metallosalen complexes are interchangeable
under the appropriate reaction conditions, the possibility that a
trans metallosalen complex reacts in the cis β configuration
should be considered.1c,d
Generally, the cis β configuration is induced either by a
bidentate ligand,3 which occupies two cis sites in the
coordination sphere, or by the bonding nature of the two
unidentate ligands, which strongly demand a cis coordination.4
A lengthening of the polymethylenic chain bridging the two
imine nitrogen atoms5 and an increase of the size of the central
metal atom6 may also favor a strained nonplanar configuration.
Little information is available about the effect of the presence of
substituents on the chelate on the conformation of metal
complexes. However, it was observed in a series of titanium−
Schiff base complexes that substantial steric effects within the
INTRODUCTION
■
The salen ligand is a Schiff base obtained by reaction of 2 equiv
of salicylaldehyde with 1 equiv of ethylenediamine, but the
more general term salen-type ligand currently refers to [N2O2]
tetradentate ligands prepared by condensation of variously
substituted salicylaldehydes with a diamine. salen-type ligands
are able to stabilize a large number of metals in different
oxidation states.1 The tetradentate salen-type ligands can adopt
three configurations in the formation of octahedral complexes:
trans, cis α, and cis β (Chart 1). Most of the complexes adopt a
trans configuration with two unidentate ligands occupying the
apical positions. There are relatively few examples in which the
quadridentate ligand assumes a cis configuration.
Chart 1
Received: October 25, 2013
Published: February 6, 2014
© 2014 American Chemical Society
909
dx.doi.org/10.1021/om401038v | Organometallics 2014, 33, 909−920

Organometallics
Article
ligand are required to achieve a folding of the ligand so that it
takes up a cis geometry.7 This result indicates that a
metallosalen complex can adopt a cis β configuration due to
the steric repulsion caused by introducing substituents on the
salen ligand.
both in solution and in the solid state, a number of complexes
containing either two or four methyl substituents in different
positions in the equatorial ligand (Chart 2) together with their
corresponding cyclized products, [CoIII(chelCH2)(py)(H2O)]-
ClO4. Herein we describe the effect of the substituents on the
cyclization rates of the organometallic [CoIII(chel)(CH2Cl)-
(L)] (L = H2O, py) complexes and the electrochemical
properties of the strictly related [CoII(chel)] and [CoIII(chel)-
(py)2]ClO4 derivatives. Furthermore, we discuss the metalla-
cycle conformational change induced by the methyl sub-
stituents in the [CoIII(chelCH2)(py)(H2O)]ClO4 products.
We have reported that trans-[CoIII(4,4′,7,7′tmsalen)-
(CH2Cl)(S)] (4,4′,7,7′tmsalen = 4,4′,7,7′-tetramethylsalen, S
= solvent; Chart 2) affords the cis β organometallic derivative
Chart 2
RESULTS
■
Syntheses. The chelates (chel) were obtained by reaction
of the appropriately substituted acetophenone or benzaldehyde
with ethylenediamine in methanol, according to the procedure
previously described for 4,4′,7,7′tmsalen (Scheme S1a,
Supporting Information).14 The corresponding CoII complexes
were prepared by reaction of the ligands with Co(CH3COO)2·
4H2O in alkaline methanol under nitrogen (Scheme S1b,
Supporting Information), whereas the [CoIII(chel)(py)2]ClO4
complexes were synthesized by carrying out the same reaction
in the presence of pyridine and with continuous stirring in air
(Scheme S1c, Supporting Information).14
The trans dimeric [CoIII(chel)(CH2Cl)]2 species were
obtained by reduction of the corresponding [CoIII(chel)(py)2]-
ClO4,with NaBH4/PdCl2 in alkaline methanolic solution
followed by the oxidative addition of CH2ClI (Scheme S1d,
Supporting Information). The products recovered from the
reaction mixture immediately after the completion of the
reaction are dimers in the solid state (the X-ray structure of
[CoIII(4,4′dmsalen)(CH2Cl)]2 (1) is reported below) and
monomers in solution, presumably with a solvent molecule in
the sixth coordination position.8,14 The cyclized cis β derivatives
[Co(chelCH2)(L)(S)]+ of the complexes containing a methyl
group at positions 7,7′, i.e. [CoIII(7,7′dmsalenCH2)(py)-
(H2O)]ClO4 and [CoIII(3,3′,7,7′tmsalenCH2)(py)(H2O)]-
ClO4, were obtained in the same way, by allowing the reaction
mixture to stand for 5 days after the alkylation reaction
(Scheme S2a, Supporting Information). This synthetic
procedure, when applied to complexes lacking the methyls at
positions 7,7′, led to mixtures of products which could not be
separated in spite of several attempts. Therefore,
[ C o I I I ( 3 , 3 ′ d m s a l e n C H 2 ) ( p y ) ( H 2 O ) ] C l O 4 a n d
[CoIII(4,4′dmsalenCH2)(py)(H2O)]ClO4 were prepared by
starting from the corresponding [CoIII(chel)(CH2Cl)]2 dis-
solved in methanol in the presence of an excess of pyridine
(Scheme S2b, Supporting Information). The reaction mixture
was heated and the progress of the reaction monitored by TLC.
The solution was evaporated to dryness in vacuo, and the
products were separated by flash chromatography on silica gel
using 3/1 CHCl3/C2H5OH as eluent. The mixture contained
the desired [CoIII(chelCH2)(py)(H2O)]Cl together with minor
amounts of unreacted [CoIII(chel)(CH2Cl)]2 and a third green
species identified as the pyridinium ylide [CoIII(chel)(CH2py)-
(Cl)], whose synthesis and characterization will be described
elsewhere. The corresponding [CoIII(3,3′dmsalenCH2)(py)-
(H2O)]ClO4 and [CoIII(4,4′dmsalenCH2)(py)(H2O)]ClO4
complexes were obtained by treatment of the chloride
derivatives with an excess of aqueous NaClO4 in methanolic
solution.
[CoIII(4,4′,7,7′-tmsalenCH2)(S)2]ClO4 through an intramolec-
ular cyclization reaction (Scheme 1).8 In this case, the
Scheme 1
conformational change of the ligand is driven by the chemical
reaction, which probably occurs through the internal
nucleophilic attack of a chelate oxygen on the axial chloroalkyl
group. Few examples of intramolecular reactions of XCH2 axial
groups (X = Cl, Br, I) with imino-oxime or amino-oxime
equatorial ligands have been described.9−11 In all cases, the
generation of an equatorial negatively charged nitrogen is
required and the reaction occurs only in strongly basic medium.
In contrast, the cyclization of [CoIII(4,4′,7,7′tmsalen)(CH2Cl)-
(S)] occurs also in neutral medium, likely due to the presence
of negatively charged oxygens in the equatorial ligand.8,12 The
cyclized product consists of a racemic mixture of Δ and Λ
enantiomers (Scheme 1; the numbering scheme of the
[CoIII(chelCH2)(py)(H2O)]+ cations refers to the Δ enan-
tiomer throughout the text).
Kinetic data obtained for the cyclization reaction of
[CoIII(4,4′,7,7′tmsalen)(CH2Cl)(S)] in the presence of L =
4-X-pyridine (X = CN, H, NH2), whose steric effect is assumed
to be similar, show that the cyclization rate constants for
[CoIII(4,4′,7,7′tmsalen)(CH2Cl)(L)] increase with the pKa of
L.13 Therefore, in this case the accelerating effect of the axial
ligand is mainly inductive in nature.
In order to gain a deeper insight into the factors that
influence the propensity of the trans-[CoIII(chel)(CH2Cl)(L)]
complexes to undergo the intramolecular cyclization in
methanolic solution, we have synthesized and characterized,
1H NMR Characterization. The assignments of the NMR
spectra of ligands and complexes were facilitated by a
910
dx.doi.org/10.1021/om401038v | Organometallics 2014, 33, 909−920

Organometallics
Article
1
Table 1. H NMR Data for the Complexes [CoIII(chel)(CH2Cl)(S)] in CD3OD
complex
CH3-Ph
CH3-CN
CH2CH2
phenyl protons
CH2Cl
C7-H
[CoIII(3,3′dmsalen)
2.22 (s, 6H)
3.76 (s, 4H)
6.36 (d, 2H, C4-H), 6.88 (s, 2H, C2-H), 7.07
(d, 2H, C5-H)
5.34 (s, 2H)
7.92 (s, 2H)
(CH2Cl)(S)]
[CoIII(4,4′dmsalen)
2.18 (s, 6H)
2.20 (s, 6H)
3.79 (s, 4H)
6.97 (m, 6H)
5.35 (s, 2H)
5.28 (s, 2H)
5.26 (s, 2H)
7.94 (s, 2H)
(CH2Cl)(S)]
[CoIII(7,7′dmsalen)
2.56 (s, 6H)
2.52 (s, 6H)
3.87 (m, 4H) 6.52 (m, 2H), 7.08 (m, 4H), 7.55 (d, 2H)
a
(CH2Cl)(S)]
[CoIII(3,3′,7,7′tmsalen)
3.83 (m, 4H) 6.36 (d, 2H, C4-H), 6.90 (s, 2H, C2-H), 7.42
(d, 2H, C5-H)
(CH2Cl)(S)]
a
From ref 21.
comparison with the corresponding data previously obtained
for the fully characterized analogues.8,14 The following
discussion will be restricted to the most interesting features
of the spectra of organometallic derivatives.
salicylaldiminate fragment forming dihedral angles of 25.10(13)
and 2.9(2)° with the N2O2 coordination donor set. It is worth
noting that the bridging oxygen O1 causes an elongation of the
equatorial Co−O1 bond length (1.925(3) Å) with respect to
the Co−O2 length (1.890(3) Å). However, the bond lengths
do not show any anomalies and agree with those measured in
similar derivatives. The axial Co−C and Co−O1′ distances are
2.043(4) and 2.189(3) Å, the latter in agreement with the
The spectra of the [CoIII(chel)(CH2Cl)(S)] complexes
(Table 1) display a relatively low number of signals, owing to
the symmetry of the molecule, and are very similar to each
other. The singlet arising from the axial CH2Cl group (Figure
S1, Supporting Information) is observed at 5.34 ppm for
[CoI I I (3,3′dmsalen)(CH2 Cl)(S)], 5.35 ppm for
[CoI I I (4,4′dmsalen)(CH2 Cl)(S)], 5.28 ppm for
[CoIII(7,7′dmsalen)(CH2Cl)(S)], and 5.26 ppm for
[CoIII(3,3,′,7,7′tmsalen)(CH2Cl)(S)], suggesting that the
methyls at positions 7,7′ cause an upfield shift of these signals
by increasing the charge density on the axial group. The
CH2CH2 protons resonate in the range 3.8−3.9 ppm (Table 1).
In particular, they appear as a singlet for [CoIII(3,3′dmsalen)-
(CH2Cl)(S)] and [CoIII(4,4′dmsalen)(CH2Cl)(S)] and as a
multiplet for [CoIII(7,7′dmsalen)(CH2Cl)(S)] and
[CoIII(3,3′,7,7′tmsalen)(CH2Cl)(S)] (Figure S1).
14
values found in [Co(4,4′,7,7′tmsalen)(CH3)]2 and [Co-
14
(4,4′,7,7′tmsalen)(CF3CH2)]2 of 2.209(2) and 2.199(2) Å,
respectively, due to the trans influence of the methylene group.
The C(17)H2−Cl bond is oriented above the five-membered
chelating ring in the direction of the N−Co−N bond angle
bisector.
The structural determination of complexes
[CoIII(4,4′dmsalenCH2)(py)(H2O)]ClO4 (2) and
[CoIII(7,7′dmsalenCH2)(py)(H2O)]ClO4 (3) confirms the
cyclization reaction with the tetradentate salen ligand
coordinating the metal in a cis fashion.
In both cationic species the cobalt atom exhibits a slightly
distorted octahedral geometry, completing the coordination
sphere through an aqua and a pyridine molecule. The resulting
complexes (depicted in Figure 2) are chiral, in spite of the fact
that they crystallize as a racemic mixture. The coordination
bond distances Co−N(py), Co−O, and Co−C (Table 3) of
complexes [CoIII(4,4′dmsalenCH2)(py)(H2O)]+ and
[CoIII(7,7′dmsalenCH2)(py)(H2O)]+ are closely comparable
within their esd’s. On the other hand, the Co−N(salen)
distances in the latter evidence a slight difference (of 1.900(4)
and 1.942(4) Å), as previously observed in the β cis derivatives
[ C o I I I ( 4 , 4 ′ , 7 , 7 ′ t m s a l e n C H 2 ) ( p y ) ( H 2 O ) ] a n d
[CoIII(4,4′,7,7′tmsalenCH2)(N-methylimidazole)2],8 but in 2
the values are analogous to those measured in the trans salen
derivative [CoIII(4,4′dmsalen)(CH2Cl)]2. The Co−OH2 bond
lengths, 2.108(3) and 2.162(3) Å in 2 and 3, respectively, are
affected by the trans influence of the CH2 group, although they
are significantly shorter than the value measured in [Co-
(4,4′,7,7′tmsalenCH2)(H2O)(py)]+ of 2.2130(9) Å.8 Since the
aqua ligands in 2 and 3 are involved in H bonds of comparable
geometry with the oxygen atom O1 of a symmetry-related
complex and with a perchlorate oxygen, the differences may be
ascribed to an electronic effect of the number and position of
methyl groups on the salen ligand.
1
The H NMR spectra of the cis-β-[CoIII(chelCH2)(py)(S)]+
complexes (Table 2 and Figure S2 (Supporting Information))
show a larger number of signals owing to the lack of elements
of symmetry. The two methyls at positions 3,3′ or 4,4′
(Scheme 1) resonate as singlets in the range 2.1−2.3 ppm,
while those at 7,7′ appear as singlets in the range 2.5−2.7 ppm.
The ethylene bridge gives rise to three multiplets which
integrate for 1, 2, and 1 H, respectively, at 3.6, 3.9, and 5.0 ppm
f o r [ C o I I I ( 3 , 3 ′ d m s a l e n C H 2 ) ( p y ) ( S ) ] + a n d
[CoIII(4,4′dmsalenCH2)(py)(S)]+, and at 3.6, 4.2, and 4.7
+
ppm for [Co(7,7′dmsalenCH2)(py)(S)] and [Co(3,3′,7,7′
+
tmsalenCH2)(py)(S)]
(Table 2). The phenyl protons
resonate in the range 6.4−7.7 ppm, while the protons at 7,7′
give rise to two singlets at 7.9 and 8.8 ppm. Remarkably, the
two diastereotopic protons of CH2, i.e. those bonded to the
organometallic C, display a different chemical shift in the
complexes having methyls at positions 7,7′ with respect to the
complexes lacking them. Thus, two singlets are observed at 5.3
and 6.8 ppm for [CoIII(3,3′dmsalenCH2)(py)(S)]+ and
[CoIII(4,4′dmsalenCH2)(py)(S)]+, and at 6.2 and 6.9 ppm
for [Co(7,7′dmsalenCH2 )(py)(S)]+ and [Co-
(3,3′,7,7′tmsalenCH2)(py)(S)]+.
X-ray Structures. The molecular structure of
[CoIII(4,4′dmsalen)(CH2Cl)]2 (1) is depicted in Figure 1,
and coordination bond distances are reported in Table 3. The
crystal contains neutral dimeric units arranged about a
crystallographic inversion center where the tetradendate salen
adopts a planar trans geometry. The oxygen atom of one Co-
salen unit completes the coordination sphere of the metal ion
of the other unit and vice versa. The salen ligand assumes an
umbrella-shaped conformation with the mean plane of each
A feature worth noting in the two complexes is the different
conformation assumed by the seven-membered metallacycle
ring upon coordination (Figure 3). In fact, the C11−C16−
O2−C17 torsion angle of the chelating arm, −3.4(6) and
90.1(6)° in 2 and 3, respectively, indicates conformational
diastereomer complexes (neglecting the methyl groups on the
chelate). The conformation detected in 3 is same as that found
911
dx.doi.org/10.1021/om401038v | Organometallics 2014, 33, 909−920

Organometallics
Article
Figure 1. Molecular structure of the centrosymmetric neutral complex
1.
Table 3. Coordination Bond Distances (Å) for Complexes
1−3
complex 1
complexes 2 and 3
2
3
Co−N(1)
1.880(3)
1.882(4)
1.925(3)
1.890(3)
2.043(4)
2.189(3)
Co−N(1)
Co−N(2)
Co−N(3)
Co−O(1)
Co−C(17)
Co−O(1w)
1.891(3)
1.891(3)
1.982(3)
1.897(3)
1.900(4)
1.942(4)
1.982(4)
1.886(3)
1.960(5)
2.162(3)
Co−N(2)
Co−O(1)
Co−O(2)
Co−C(17)
1.964(4)
2.108(3)
a
Co−O(1)′
a
Atom O(1)′ in 1 at −x + 1, −y, −z + 1.
Figure 2. Molecular structures of the complex cations of 2 and 3.
Figure 3. View of complexes 2 and 3 showing the different
conformations of the metallacycle and the position of the coordinating
methylene group with respect to the mean plane through the aromatic
ring.
in the [CoIII(4,4′,7,7′tmsalenCH2)(py)(H2O)]+ derivative,8
where the cited torsion angle is 87.8°.
912
dx.doi.org/10.1021/om401038v | Organometallics 2014, 33, 909−920

Organometallics
Article
In [CoIII(4,4′,7,7′tmsalenCH2)(py)(H2O)]+ we observed8
that the delocalization in the methoxy moiety is strongly
reduced with respect to the salicylaldiminate group, mainly
evidenced by the differences detected in the NC−C groups.
Here these differences in bond lengths are not meaningful, the
values being comparable within their esd’s. However, in
complexes 2 and 3 significant variations are detected in bond
angles about the phenolate carbon atom C(16) in addition to a
widening of the C−O−C angle (from 112.2(4) to 121.5(3)°)
and a small elongation of the C(Ph)−O bond length observed
in the 7,7′ molecular cation (Table 4). In both cases, the
pyridine ring forms a π-stacking interaction with the salen
cyclized ring (centroid to centroid distance of ca. 3.64 Å).
where the axial CH2 group is placed out of plane of the
aromatic ring. These complexes were built by starting from the
geometry of 3. In the former set, where the axial CH2 group is
in plane with the aromatic ring, changes of one methyl group
position from 4′ to 7′ (4,7′ derivative) or of another methyl
from 4 to 7 (7,4′ derivative) are energetically unfavorable
(11.50 and 16.61 kJ mol−1, respectively) (first row of Table 5).
The change of placement for both methyl groups (7,7′
derivative) gives the even higher value of 22.66 kJ mol−1.
However, if the change of methyl group position from 4 to 7
(7,4′ derivative) is accompanied by the displacement of the
CH2 group out of plane, a stabilization of about 8 kJ mol−1
occurs (second and third row of Table 5) in comparison to the
complex with the axial CH2 group placed in plane with the
aromatic ring. Similar stabilization could be found for the 7,7′
derivative. Thus, the difference in standard Gibbs energies is the
result of the energetically unfavorable steric hindrance with
methyl groups in positions 7 and 7′ that causes the rotation of
both aromatic rings followed by the energetically favorable
displacement of the axial CH2 group out of plane and toward
the cobalt.
Electrochemical Studies. Cyclic voltammetry was carried
out on the [CoII(chel)] and [CoIII(chel)(py)2]ClO4 complexes,
whose redox processes are not complicated by the coupled
cleavage of the Co−C bond, which occurs for the
corresponding organometallic complexes.15 The CV measure-
ments were performed in DMSO + 0.1 M TEAP at 25.0 °C on
a glassy-carbon (GC) electrode; the CV signals are very similar
for all of the examined compounds and change only in peak
potential values (Epc and Epa).
Table 4. Geometrical Distortions (Distances in Å and Angles
in deg) Observed in the Ph−O−CH2 Fragment of 2 and 3
upon the out-of-Plane Displacement of CH2 in the Latter
2, R = H
3, R = Me
C(16)−O(2)
C(10)−C(11)
1.369(5)
1.455(6)
112.2(4)
128.0(4)
113.2(4)
118.8(4)
−3.4(6)
−34.2(7)
1.392(6)
1.492(7)
121.5(5)
119.3(5)
119.8(5)
120.8(5)
90.1(6)
C(17)−O(2)−C(16)
O(2)−C(16)−C(11)
O(2)−C(16)−C(15)
C(11)−C(16)−C(15)
C(17)−O(2)−C(16)−C(11)
N(2)−C(10)−C(11)−C(16)
49.3(8)
Quantum Chemical Calculations. Calculations of stand-
ard Gibbs energies reveal that complex 2 is 16.9 kJ mol−1 higher
in energy than complex 3 (Table 5). Apart from placement of
the methyl groups, the most notable feature in these two
complexes is the different position of the axial CH2 group
relative to the plane of the aromatic ring (Figure 3). In 2, the
carbon atom of the methylene group is coplanar with the
aromatic ring, whereas in 3 it is significantly displaced from it
(Figure 3). This conformational change appears to be induced
by the presence of the methyl groups in positions 7,7′. In order
to maintain the coordination to cobalt, the CH2 group is shifted
out of plane of the aromatic ring involved in the metallacycle.
This conformation is accompanied by the phenyl ring rotation
of ca. 15° (Figure 3).
To investigate the reasons that cause the difference in
standard Gibbs energies, we studied the geometries of six
additional complexes with different positions of methyl groups
and the axial CH2 group placed in two limiting conformations:
out of plane and in plane (Figure 3 and Table 5). Starting from
the geometry of complex 2, where the axial CH2 group is in
plane with the aromatic ring, the corresponding derivatives with
chel = 7,4′dmsalen, 4,7′dmsalen, 7,7′dmsalen were built and
optimized. Additionally, geometry optimizations were carried
out for 4,4′dmsalen, 7,4′dmsalen, and 4,7′dmsalen derivatives
The cyclic voltammograms of the [CoII(chel)] complexes in
the range from +0.5 to −2.0 V vs SCE at 50 mV/s scan rate
show two characteristic anodic/cathodic signal couples. The
more negative of these is attributed to the CoII/CoI process
(Figure 4), while the more positive couple is assigned to the
CoIII/CoII process. In any case, the Epa − Epc value is in
agreement with a quasi-reversible monoelectronic redox
process.16 The species CoI, CoII, and CoIII, are all stable on
the CV time scale under these experimental conditions, as was
proved by the fact that the peak currents ipa and ipc are
practically the same for all four peaks.
For [CoIII(chel)(py)2]ClO4 complexes, a scan in negative
direction from +0.5 to −2.0 V shows the presence of some
partially overlapped cathodic peaks between −0.1 and −0.5 V,
followed by a higher cathodic peak at more negative potentials.
The overlapped peaks at more positive potentials are attributed
to the CoIII/CoII reduction process, the [CoIII(chel)(py)2]+
complex being in equilibrium with other species in which either
one or both of the py ligands are replaced by the solvent. In
fact, for all of the complexes, the more positive of these
overlapped reduction peaks occurs at the same value observed
in the CV of the corresponding authentic [CoII(chel)] species
Table 5. Standard Gibbs Energies (ΔrG°, kJ mol−1) at 298.15 K and 1 atm for CoIII(dmsalen) complexes Calculated for the in-
a
Plane (ip) and the out-of-Plane (oop) Positions of the Axial CH2 Group Relative to the Plane of the Aromatic Ring
ΔrG°4,4′
ΔrG°7,4′
ΔrG°4,7′
ΔrG°4,7′ − ΔrG°7,4′
ΔrG°7,7′
ΔrG°7,7′ − ΔrG°7,4′
ΔrG°7,7′ − ΔrG°4,7′
b
ip
0.00
11.50
21.42
9.92
16.61
8.33
5.11
22.66
11.16
6.05
8.57
b
oop
9.57
9.57
−13.09
16.90
−4.52
ΔrGθ − ΔrGθ
−8.28
−5.76
oop
ip
a
M06/def2-TZVP level of theory, solvent effects incorporated using the SMD method. The positions of the methyls in the chelate are indicated by
b
subscripts. Geometry optimization starting from the crystal structure.
913
dx.doi.org/10.1021/om401038v | Organometallics 2014, 33, 909−920

Organometallics
Article
[CoIII(3,3′,7,7′tmsalen)(CH2Cl)(S)] (Figure 5), with absorb-
ance vs time traces showing a good first-order exponential
Figure 4. Cyclic voltammetry of [CoIII(3,3′,7,7′tmsalen)(py)2]ClO4
(black line) and of [CoII(3,3′,7,7′tmsalen)] (red line). Experiments
were performed in DMSO + 0.1 M TEAP as supporting electrolyte on
a glassy-carbon electrode at a 50 mV/s scan rate at 25 °C. The
concentration of the complexes was about 1 mM.
Figure 5. Spectral changes with time of an approximately 2 × 10−4
M
solution of [CoIII(3,3′,7,7′tmsalen)(CH2Cl)(S)] in 60% methanol−
40% water (v/v, I = 0.1 M NaClO4) at 25 °C. Spectra were recorded at
time intervals of 30 min.
(Figure 4). The well-defined peaks at more negative potentials
are attributed to the CoII/CoI process.
behavior (Figure S3, Supporting Information). For
[CoIII(3,3′dmsalen)(CH2Cl)(S)] and [CoIII(4,4′dmsalen)-
(CH2Cl)S], a further slow process becomes evident before
the conclusion of the cyclization reaction at elevated temper-
atures (Figure S4, Supporting Information). It had been
previously observed that the product of cyclization of
[CoIII(4,4′,7,7′tmsalen)(CH2Cl)(S)], which presumably con-
tains two solvent molecules as unidentate ligands, is not stable
in methanolic solution and further reactions occur after the
completion of cyclization.8
The second process is attributed to the decomposition of the
cyclized species in the absence of a nitrogen base. The present
results suggest that the cyclization/decomposition processes
tend to overlap for complexes lacking the methyls in positions
7,7′. Despite this complication, it was possible to perform a
kinetic study of the first process by carrying out the reactions at
wavelengths for which the changes in absorbance related to the
second reaction were very small.
The presence of electron-donating methyl groups in the
equatorial chelate shifts the redox potentials of both the CoIII/
CoII and CoII/CoI electron transfer processes toward more
negative values. The effect is clearly evident in the CoII/CoI
processes of both [CoIII(chel)(py)2]+ and [CoII(chel)]
compounds and in the CoIII/CoII reduction signals of
[CoII(chel)]. A similar trend can be detected for the CoIII/
CoII reduction peaks of [CoIII(chel)(py)2]+, although these
signals are quite broad (see above). A comparison of the E1/2
values (Table 6) with that obtained for the CoII/CoI reduction
of unsubstituted [CoII(salen)] (−1.218 V)15 shows that the
insertion of methyls in positions 3,3′ or 4,4′ shifts the potentials
of 0.045 and 0.022 V, respectively, toward more negative values.
Substitution at positions 7,7′ has a stronger effect (0.120 V).
For the tetrasubstituted [CoII(3,3′,7,7′tmsalen)] complex
(Table 6) the shift of the redox potentials toward more
negative values is further enhanced. The already reported E1/2
v a l u e f o r t h e C o I I I / C o I I r e d o x p r o c e s s o f
[CoII(4,4′,7,7′tmsalen)] (−0.083 V)15 does not agree with
this trend: the result was confirmed by a repeated experiment
and could not be rationalized.
Time-resolved UV−vis spectra of the [CoIII(chel)(CH2Cl)-
(S)] complexes under the same experimental conditions but in
the presence of an excess of py show two processes for all of the
complexes. The processes are well separated for
[CoIII(7,7′dmsalen)(CH2Cl)(S)] and [CoIII(3,3′,7,7′tmsalen)-
(CH2Cl)(S)] at any concentration and temperature but again
tend to overlap for [CoIII(3,3′dmsalen)(CH2Cl)(S)] and
[CoIII(4,4′dmsalen)(CH2Cl)(S)] at high pyridine concentra-
tions and elevated temperatures (Figure S5, Supporting
Kinetics and Mechanism. When time-resolved UV−vis
spectra of the [CoIII(chel)(CH2Cl)(S)] complexes were
recorded in 60% methanol−40% water (v/v, I = 0.1 M
NaClO4) in the temperature range 20−40 °C, a well-defined
process was observed for [CoIII(7,7′dmsalen)(CH2Cl)(S)] and
Table 6. Redox Potentials (V vs SCE) in DMSO + TEAP 0.1 M Solution for [CoII(chel)] on a GC Electrode at 25 °C
a
a
complex
Epc(II/I)
Epa(I/II)
E1/2
Epc(III/II)
Epa(II/III)
E1/2
[CoII(3,3′dmsalen)]
[CoII(4,4′dmsalen)]
[CoII(7,7′dmsalen)]
[CoII(3,3′7,7′tmsalen)]
−1.312
−1.270
−1.386
−1.525
−1.213
−1.210
−1.290
−1.379
−1.263
−1.240
−1.338
−1.452
−0.162
−0.140
−0.180
−0.228
−0.031
−0.045
−0.030
−0.060
−0.097
−0.093
−0.105
−0.144
a
E1/2 = (Epc + Epa)/2.
914
dx.doi.org/10.1021/om401038v | Organometallics 2014, 33, 909−920

Organometallics
Article
Information). The problem was circumvented as described
above.
Scheme 2
As previously found for [CoIII(4,4,′7,7′tmsalen)(CH2Cl)-
(S)],13 the plots of kobs vs [L] show a nonzero intercept and a
significant curvature at high pyridine concentrations (Figure 6).
Table 7 gives kinetic rate constants and activation parameters
for the cyclization reaction.
This scheme involves the substitution of the solvent by py in
a fast pre-equilibrium step followed by the cyclization of both
[Co(chel)(CH2Cl)(S)] and [Co(chel)(CH2Cl)(py)] at differ-
ent rates. The corresponding rate law is given in eq 1, where K
k1 + k2K[py]
1 + K[py]
kobs
=
(1)
represents the equilibrium constant for the substitution of the
solvent by py in the axial position and k1 and k2 are the
cyclization rate constants of [Co(chel)(CH2Cl)(S)] and
[Co(chel)(CH2Cl)(py)], respectively. The value of k1 has
been independently determined by carrying out the cyclization
reaction in the absence of pyridine (Table 7). Preliminary
estimates of k2 and K have been obtained from the linear plots
Figure 6. Plots of the pseudo-first-order rate constant, kobs, versus [py]
for the cyclization of [CoIII(3,3′,7,7′tmsalen)(CH2Cl)(S)] in 60%
methanol−40% water (v/v, I = 0.1 M NaClO4) at various
temperatures. The solid lines show the fit based on eq 1 of the text
with the parameters reported in Table 7.
The results have been interpreted as indicated in Scheme 2.
Table 7. Kinetic Rate Constants and Activation Parameters for the Cyclization Reaction in 60% Methanol−40% Water (v/v, I =
0.1 M NaClO4)
complex
[CoIII(3,3′dmsalen)(CH2Cl)(S)]
T, K
105k1, s−1
1.73 0.01
104k2, s−1
K, M−1
293.1
298.1
303.1
308.1
313.1
1.7 0.1
3.3 0.4
6.0 0.2
85
33
32
26
27
4
3
1
1
1
3.78 0.02
7.30 0.01
12.5 0.1
10.33 0.5
20.7 0.2
16.7 0.7
ΔH1* = 92 4 kJ mol−1
ΔS1* = −23 13 J mol−1K−1
3.59 0.01
ΔH2* = 85 2 kJ mol−1
ΔS2* = −27 7 J mol−1 K−1
2.7 0.4
[CoIII(4,4′dmsalen)(CH2Cl)(S)]
[CoIII(7,7′dmsalen)(CH2Cl)(S)]
[CoIII(3,3′,7,7′tmsalen)(CH2Cl)(S)]
293.1
298.1
303.1
308.1
313.1
35
4
6.96 0.01
4.3 0.1
31.5 0.8
12.1 0.1
8.4 0.8
22
22
13
3
1
2
20.4 0.2
14.5
1
41.8 0.4
24.1
2
ΔH1* = 67 7 kJ mol−1
ΔS1* = −99 22 J mol−1K−1
4.77 0.01
ΔH2* = 64 5 kJ mol−1
ΔS2* = −96 17 J mol−1 K−1
2.5 0.2
293.1
298.1
303.1
308.1
313.1
62
25
35
4
5
2
a
a
a
8.39 0.01
5.6 0.8
15.1 0.1
8.2 0.5
27.9 0.1
15.2 0.2
26.7 0.2
14.0 0.4
45.6 0.1
27.8 0.8
ΔH1* = 85 1 kJ mol−1
ΔS1* = −39 5 J mol−1K−1
6.86 0.01
ΔH2* = 86 4 kJ mol−1
ΔS2* = −21 14 J mol−1 K−1
4.4 0.3
293.1
298.1
303.1
308.1
313.1
46
59.2 0.7
38
12.9 0.4
27
2
16.4 0.1
8.1 0.1
22.3 0.1
13.5 1.2
3
50.4 0.1
27.5 0.7
63.8 0.1
36
2
1
ΔH1* = 83 9 kJ mol−1
ΔH2* = 78 6 kJ mol−1
ΔS1* = −41 29 J mol−1K−1
ΔS2* = −42 21 J mol−1 K−1
915
dx.doi.org/10.1021/om401038v | Organometallics 2014, 33, 909−920

Organometallics
Article
of 1/(kobs − k1) versus 1/[py]. The final values were derived
from a nonlinear fit of kobs versus [py], and the activation
parameters calculated according to the Eyring equation (eq 2)
are summarized in Table 7.
circumvent the problem, owing to the scarce solubility of the
starting products at low temperatures and the fast formation of
ylides at high temperatures (see above).
In the proposed reaction mechanism, the cyclization is a
formally unimolecular reaction,19 which should have ΔS* near
zero or positive, if the activation process involves bond
breaking. The considerably negative values of the activation
entropy may stem from the loss of conformational freedom on
going from the ground state to an almost cyclic transition state.
A second contribution may arise from the change in solvation,
as the cyclization involves the formation of ions from a neutral
complex. If the activated complex may be described almost as
an ion pair or a polar complex approaching an ion pair, a large
negative value of the activation entropy due to the freezing of
the solvent around the incipient ions should be expected.
The X-ray structures of the cyclized complexes indicate that
the conformation of the seven-membered metallacycle shows a
dependence on the positions of the methyl substituents in the
chelate. In fact, the conformation assumed by the Ph−O−
CH2−Co fragment within the seven-membered metallacycle
ring in complexes having methyl substituents on the −CN−
groups is different from that of the complexes lacking methyls
at these positions. In the [CoIII(4,4′dmsalenCH2)(py)(H2O)]+
complex, the carbon atom of the methylene group lies in the
plane of the aromatic ring, whereas in the
[CoIII(7,7′dmsalenCH2)(py)(H2O)]+ complex this carbon
atom is significantly displaced from the phenol ring plane, as
previously found in the [CoIII(4,4′,7,7′dmsalenCH2)(py)-
(H2O)]+ complex.8
Quantum chemical calculations reveal that the standard
Gibbs energy for the optimized structure of the 7,7′-methyl-
substituted derivative is 16.9 kJ mol−1 greater than that of the
4,4′ species (Table 5). This appears to agree with the largely
preferred coplanar conformation (also in the presence of an
ortho substituent) of methoxyphenyl groups observed in crystal
structure data.20 In addition, the ΔG° values of variously methyl
substituted salen species indicate that the difference in
conformation is a balanced effect between the energetically
unfavorable state represented by methyl groups in positions 7
and 7′ and the energetically favorable out-of-plane conforma-
tional displacement of the CH2 group from the aromatic ring.
The computations corroborate the conformational features
resulting from the X-ray structural analyses, indicating that the
conformation of 3 is likely induced by steric hindrance of the
methyl groups in positions 7,7′. In order to allow the
coordination to the metal, the out-of-plane coordinating
methylene in 3 implicates significant geometrical distortions
in the geometry of the Ph−O−CH2−Co fragment in
comparison to complex 2.
*
RT
*
ΔSi
R
ki
T
ΔHi
κ
h
ln
= ln
−
+
(2)
DISCUSSION
■
The trans-[CoIII(chel)(CH2Cl)]2 complexes, which are dimers
in the solid state and presumably monomers in solution with a
solvent molecule in the sixth coordination position,8,14 form cis
β organometallic derivatives containing a seven-membered
metallacycle, by replacement of the Cl atom of the axial CH2Cl
ligand by the salen phenolate oxygen. The syntheses of the
complexes [CoIII(7,7′dmsalenCH2)(py)(H2O)]ClO4 and
[CoIII(3,3′,7,7′tmsalenCH2)(py)(H2O)]ClO4 were carried out
b y t h e p r o c e d u r e a l r e a d y d e s c r i b e d f o r
[CoIII(4,4′,7,7′tmsalenCH2)(py)(H2O)]ClO4.8 This method
was revealed to be useless for complexes whose chelates lack
methyls in positions 7,7′, leading to intractable mixtures of
products. Therefore, [CoIII(3,3′dmsalenCH2)(py)(H2O)]ClO4
and [CoIII(4,4′dmsalenCH2)(py)(H2O)]ClO4 were synthe-
sized from the corresponding trans chloromethyl derivatives
in the presence of a large excess of pyridine, which was
demonstrated to accelerate the cyclization rate.13 Under these
experimental conditions, the formation of the rather
uncommon nonstabilized pyridinium ylide species [CoIII(chel)-
(CH2py)(Cl)]17 is observed in addition to the desired
products.
The proposed metallacyclization mechanism involves the
internal nucleophilic attack of the phenolate oxygen on the axial
chloromethyl group with the detachment of a chloride ion
(Scheme 2). From the data of Table 7 it appears that the
cyclization rate of [Co(chel)(CH2Cl)(py)] is about 1 order of
magnitude greater than that of [Co(chel)(CH2Cl)(S)].
It has been previously shown for [CoIII(4,4′,7,7′tmsalen)-
(CH2Cl)(S)] that the k2 values increase remarkably with the
pKa of the nitrogen base.13 The data of Table 7 show that the
cyclization rate depends on the number of the methyl
substituents in the chelate: the values of k1 and k2 for
3,3′,7,7′tmsalen and 4,4′,7,7′tmsalen13 derivatives are almost
double those of the dmsalen complexes. This result is in
agreement with the electrochemical data that evidence a general
shift of the E1/2 potentials toward more negative values with an
increase in the number of methyl sustituents. The cyclization
rate is also affected by the substituent position: in the dmsalen
derivatives it increases in the order 3,3′ < 4,4′ < 7,7′. The
reactivity order does not match exactly the trend of redox
potentials, but both the electrochemical and the kinetic data
evidence a remarkable influence of methyls at the 7,7′ positions.
The above results suggest that the substituent effect on the
intramolecular cyclization rate is inductive in origin and that the
rate enhancement is due to the increased electron density on
the equatorial chelate, which makes the oxygen atoms more
nucleophilic.
It is worth noting that the 1H NMR spectra of the complexes
lacking methyls on the −CN− groups show a considerable
upfield shift of one of the two diastereotopic protons bonded to
the axial methylene group. This effect can be ascribed to the
different conformations of this group in 2 and 3, which cause
different orientations of the protons in the magnetic anisotropy
cones of aromatic rings of the complexes. Thus, these data
suggest that the different conformations of the metallacycle are,
at least partially, retained in solution by 2 and 3, and
presumably a similar behavior can be ascribed to
[ C o I I I ( 3 , 3 ′ d m s a l e n C H 2 ) ( p y ) ( H 2 O ) ] + a n d
[CoIII(3,3′,7,7′tmsalenCH2)(py)(H2O)]+, whose X-ray struc-
tures are not available.
The activation entropies do not show a clear trend with the
position and the number of methyl substituents (Table 7); the
lack of correlation may arise from the inaccuracy of obtaining
ΔS* by a long extrapolation of Eyring plots to 1/T = 0.18
Unfortunately, we could not use a wider temperature range to
916
dx.doi.org/10.1021/om401038v | Organometallics 2014, 33, 909−920

Organometallics
Article
Table 8. Elemental Analyses and Synthetic Yields for Chelates and [CoII(chel)] and [CoIII(chel)(py)2]ClO4 Complexes
calcd (found) (%)
compound
formula
C
H
N
yield (%)
3,3′dmsalen
4,4′dmsalen
C18H20N2O2
72.9 (72.8)
72.9 (72.3)
74.0 (73.4)
61.2 (60.6)
61.2 (60.5)
61.2 (59.8)
63.0 (61.4)
55.1(54.5)
55.1(54.5)
56.4(56.0)
6.80 (7.15)
6.80 (6.75)
7.46 (8.08)
5.13 (5.53)
5.13 (5.86)
5.13 (6.27)
5.81 (7.43)
4.62(4.92)
4.62(4.71)
5.05(5.25)
9.45 (9.59)
9.45 (9.60)
8.59 (8.73)
7.93 (7.80)
7.93 (7.88)
7.93 (7.77)
7.35 (7.18)
9.17(9.05)
9.17(8.95)
8.77(8.56)
89.1
92.5
90.2
76.7
67.1
73.2
78.2
98.0
97.1
97.3
C18H20N2O2
3,3′,7,7′tmsalen
C20H24N2O2
[CoII(3,3′dmsalen)]
C18H18CoN2O2
C18H18CoN2O2
C18H18CoN2O2
C20H22CoN2O2
C28H28CoN4O6Cl
C28H28CoN4O6Cl
C30H32CoN4O6Cl
[CoII(4,4′dmsalen)]
[CoII(7,7′dmsalen)]
[CoII(3,3′,7,7′tmsalen)]
[CoIII(3,3′dmsalen)(py)2]ClO4]
[CoIII(4,4′dmsalen)(py)2]ClO4]
[CoIII(3,3′,7,7′tmsalen)(py)2]ClO4]
mmol) dissolved in methanol (200 mL). The reaction mixture was
warmed to 50 °C for 2 h and then allowed to stand overnight. The
yellow solid was collected by filtration and dried.
EXPERIMENTAL SECTION
General Methods. The synthesis and characterization of the Co
complexes with chel = 4,4′,7,7′tmsalen have already been reported by
■
General Procedure for Synthesis of [CoII(chel)] Complexes. A
suspension of 2 mmol of the ligand in 120 mL of CH3OH was
deaerated with nitrogen several times. A slight excess of Co-
(CH3COO)2·4H2O (0.5 g, 2.0 mmol) and one pellet of NaOH
were added under a nitrogen atmosphere, and the stirring was
continued for 1 h at 35 °C. The orange solid was collected by filtration
under nitrogen, washed with methanol, and dried.
Table 9. Yields and ESI-MS Data (MeOH) for
[CoIII(chel)(CH2Cl)]2 and [CoIII(chelCH2)(py)(H2O)]+
Complexes
ESI-MS for [CoIII(chel)(CH2Cl)]+
(m/z)
General Procedure for Synthesis of [CoIII(chel)(py)2]ClO4
Complexes. To a solution of 0.5 mmol of the ligand suspended in
methanol (200 mL) were added a slight excess of Co(CH3COO)2·
4H2O (1.4 g, 5.6 mmol) and then 10 mL of pyridine. Air was bubbled
in the solution with stirring for 3 h. An aqueous solution of NaClO4
was added and the brown product collected by filtration and washed
with water.
yield
(%)
complex
calcd
found
[CoIII(3,3′dmsalen)
27.0
402.8 402.0 (+H+, 3%),
(CH2Cl)]2
425.0 (+Na+, 8%)
[CoIII(4,4′dmsalen)
53.9
74.2
402.8 402.1 (+H+, 17%)
430.8 453.0 (+Na+, 6%)
(CH2Cl)]2
[CoIII(3,3′,7,7′tmsalen)
Caution! Perchlorate salts of metal complexes with organic ligands
are potentially explosive.
(CH2Cl)]
2
ESI-MS for [CoIII(chelCH2)
(py)(H2O)]+ (m/z)
General Procedure for Synthesis of [CoIII(chel)(CH2Cl)]2
Complexes. As the complexes are light sensitive, all manipulations
were performed in the dark. A methanol solution of [CoIII(chel)-
(py)2]ClO4 (1.4 mmol in 80 mL of CH3OH) containing one pellet of
NaOH was thoroughly dearated by several vacuum/nitrogen cycles. A
solution of NaBH4 (0.060 g, 1.59 mmol) in 1 mL of water was added
to the suspension, followed by addition of a few drops of an aqueous
10% PdCl2 solution. A very fast reduction occurred, evidenced by the
instantaneous change of color from light brown to red. A slight excess
of CH2ClI (0.25 mL, 3.43 mmol) was then added, and the stirring
under nitrogen was continued for 1.30 h. After filtration, 100 mL of
water was added to the solution drop by drop and the precipitate was
collected by filtration, washed with water, and dried in air. X-ray-
quality crystals of [CoIII(4,4′dmsalen)(CH2Cl)]2 (1) were obtained by
slow evaporation from a saturated methanolic solution at 4 °C.
[CoIII(7,7′-dmsalenCH2)(py)(H2O)]ClO4 and [CoIII(3,3′,7,7′-
tmsalenCH2)(py)(H2O)]ClO4. A methanol solution of the appropriate
[CoIII(chel)(py)2]ClO4 (0.82 mmol in 80 mL of CH3OH) containing
one pellet of NaOH was thoroughly deaerated by several vacuum/
nitrogen cycles. A solution of NaBH4 (0.060 g, 1.59 mmol) in 1 mL of
water was added to the suspension, followed by addition of a few
drops of an aqueous 10% PdCl2 solution. A very fast reduction
occurred, evidenced by the instantaneous change of color from light
brown to red. A slight excess of CH2ClI (0.25 mL, 3.43 mmol) was
then added, and the stirring under nitrogen was continued for 1.30 h.
The solution was poured in a flask and allowed to stand for 5 days in
yield
(%)
complex
calcd
found
[CoIII(3,3′dmsalenCH2)(py)
35.4
73.6
51.5
52.7
62.0
464.1
464.1
464.1
464.1
492.2
367.1 (−py, 100%)
(H2O)]Cl
[CoIII(3,3′dmsalenCH2)(py)
367.1 (−py, 100%)
367.1 (−py, 100%)
367.1 (−py, 100%)
395.1 (−py, 100%)
(H2O)]ClO4
[CoIII(4,4′dmsalenCH2)(py)
(H2O)]Cl
[CoIII(4,4′dmsalenCH2)(py)
(H2O)]ClO4
[CoIII(3,3′,7,7′tmsalenCH2)(py)
(H2O)]ClO4
us,8 as well as [CoI I I (7,7′dmsalen)(py)2 ]ClO4 and
[CoIII(7,7′dmsalen)(CH2Cl)]2.21 All other reagents were analytical
grade and were used without further purification. UV−vis spectra were
monitored with an UVIKON 941 PLUS spectrophotometer. 1D and
2D NMR spectra were recorded on a JEOL EX-400 instrument (1H at
400 MHz and 13C at 100.4 MHz) with the solvent as internal
reference. Electrospray mass spectra were recorded in positive mode
by using an API 1 mass spectrometer (Perkin-Elmer).
General Procedure for Synthesis of Ligands. A solution of
ethylenediamine (0.6 g, 10 mmol) in methanol (50 mL) was mixed
with the appropriately substituted acetophenone or benzaldehyde (20
1
Table 10. H NMR Data (CDCl3) for Chelates
chel
CH3-Ph
CH3-CN
CH2CH2
phenyl protons
C7-H
OH
3,3′dmsalen
4,4′dmsalen
3,3′,7,7′tmsalen
2.31 (s, 6H)
2.25 (s, 6H)
2.29 (s, 6H)
3.89 (s, 4H)
3.92 (s, 4H)
3.94 (s, 4H)
6.66 (d, 2H), 6.74 (s, 2H), 7.10 (d, 2H)
6.84 (d, 2H), 7.00 (s, 2H), 7.09 (d, 2H)
6.58 (d, 2H), 6.71 (s, 2H), 7.38 (d, 6H)
8.30
8.29
13.21 (bs, 2H)
12.97 (s, 2H)
15.96 (s, 2H)
2.34 (s, 6H)
917
dx.doi.org/10.1021/om401038v | Organometallics 2014, 33, 909−920

Organometallics
Article
the dark. The suspension was then filtered, 15 mL of water was added,
and the solvent was removed under reduced pressure to give a red
solid, which was collected by filtration and dried on P2O5. X-ray-
quality crystals of [CoIII(7,7′dmsalenCH2)(py)(H2O)]ClO4 (3) were
obtained by slow evaporation of the solvent.
[CoIII(4,4′-dmsalenCH2)(py)(H2O)]Cl and [CoIII(3,3′-
dmsalenCH2)(py)(H2O)]Cl. A 0.0947 g amount (0.118 mmol) of
the corresponding [CoIII(chel)(CH2Cl)]2 was dissolved in 100 mL of
methanol, and 5 mL of pyridine was added. The solution was warmed
to 40 °C in the dark, and the progress of the reaction was monitored
by TLC. After ca. 2 h the heating was stopped and the solvent was
evaporated in vacuo. The mixture was separated by flash
chromatography on a silica gel column using 3/1 CHCl3/C2H5OH
as eluent. The first brown and the third green fractions were discarded.
The red central fraction afforded the desired product after evaporation
of the solvent in vacuo.
[CoIII(4,4′-dmsalenCH2)(py)(H2O)]ClO4 and [CoIII(3,3′-
dmsalenCH2)(py)(H2O)]ClO4. A solution of NaClO4·H2O (2.00 g,
1.42 × 10−2 mol) in 5 mL of MeOH was added to 90 mg (2.2 × 10−4
mol) of the corresponding [CoIII(chelCH2)(py)(H2O)]Cl dissolved in
about 15 mL of MeOH. After addition of 3 mL of H2O the solution
was set aside for partial evaporation and the precipitate was collected
by filtration. X-ray-quality crystals of [CoIII(4,4′dmsalenCH2)(py)-
(H2O)]ClO4 (2) were obtained by slow evaporation of the solvent.
Characterization of Compounds. Characterization data (ele-
mental analysis, ESI MS mass spectra, and 1H NMR spectra) are
collected in Tables 1, 2, and 8−11.
Crystallographic Measurements. Data for complexes
[CoIII(4,4′dmsalen)(CH2Cl)]2 (1) and [CoIII(7,7′dmsalenCH2)(py)-
(H2O)]ClO4 (3) were collected at room temperature on a Bruker
rotating anode diffractometer equipped with a Kappa-CCD detector
and graphite-monochromated Cu Kα radiation (λ = 1.54178 Å).
Intensity data for [CoIII(4,4′dmsalenCH2)(py)(H2O)]ClO4 (2) were
collected at the X-ray diffraction beamline (λ = 1.0000 Å, 100 K) of
the Elettra synchrotron (Trieste, Italy). Data reductions were made by
using the programs Denzo and Scalepack.22 All of the structures were
solved by direct methods23 and refined with full-matrix least squares
based on F2 with all observed reflections using the SHELXL program
on F2 with all observed reflections.23 Hydrogen atoms were positioned
geometrically and allowed to ride on their respective parent atom, with
the exception of water molecules in [CoIII(4,4′dmsalenCH2)(py)-
(H2O)]ClO4 and [CoIII(7,7′dmsalenCH2)(py)(H2O)]ClO4 located
on the difference Fourier map with O−H distances restrained at 0.85
Å. [CoIII(7,7′dmsalenCH2)(py)(H2O)]ClO4 revealed the presence of
two chloroform molecules, one of which was refined with an
occupancy of 0.5. The calculations were made using the WinGX
System, vesion 1.80.05.24 Crystal data and details of refinements are
given in Table S1 (Supporting Information), whereas selected bond
distances and angles are given in Table S2 (Supporting Information).
Hydrogen bond parameters are given in Table S3 (Supporting
Information).
Computational Methods. All quantum chemical calculations
were performed using the Gaussian 09 program package (Table S4,
Supporting Information).25 Geometry optimizations for ground states
were performed using the hybrid functional M0626 and def2-
TZVP27−29 basis set.30,31 For all optimized structures harmonic
frequencies were calculated to ensure that obtained geometries
correspond to the local minimum on the potential energy surface.
Solvation effects were incorporated in the calculations using the
reformulation of polarizable continuum model (PCM)32,33 known as
the integral equation formalism (IEFPCM) of Tomasi and co-
workers34−38 with radii and nonelectrostatic terms for Truhlar and co-
workers’ SMD solvation model.39 The standard Gibbs energies were
calculated at T = 298.15 K and p = 1 atm.
Cyclic Voltammetry Experiments. The electrochemical meas-
urements were carried out in freshly distilled DMSO (Fluka) that was
kept under molecular sieves in dark bottles. The supporting electrolyte
was tetraethylammonium perchlorate (TEAP, 0.1 M; Fluka), which
was used after recrystallization from hot water and vacuum drying at
30 °C. Voltammetric experiments were performed using an Amel 552
918
dx.doi.org/10.1021/om401038v | Organometallics 2014, 33, 909−920

Organometallics
Article
potentiostat/galvanostat connected with an Amel 568 function
generator and equipped with a three-electrode thermostated jacked
(b) Carroll, K. M.; Schwartz, J.; Ho, D. M. Inorg. Chem. 1994, 33,
2707−2708. (c) Calligaris, M.; Manzini, G.; Nardin, G.; Randaccio, L.
J. Chem. Soc., Dalton Trans. 1972, 543−547. (d) Kushi, Y.; Tada, T.;
Fujii, Y.; Yoneda, H. Bull. Chem. Soc. Jpn. 1982, 55, 1834−1839.
(4) (a) Yamanouchi, K.; Yamada, S. Inorg. Chim. Acta 1974, 9, 161−
164. (b) Cilla, A. C.; Goghi, L.; Manfredotti, A. G.; Guastini, G. Cryst.
Struct. Commun. 1974, 3, 551−554. (c) Thornback, J. R.; Wilkinson,
G. J. Chem. Soc., Dalton Trans. 1978, 110−115. (d) Motevalli, M.;
Parkin, B.; Ramnauth, R.; Sullivan, A. Dalton Trans. 2000, 2661−2662.
(e) Xu, Z.-J.; Fang, R.; Zhao, C.; Huang, J.-S.; Li, G.-Y.; Zhu, N.; Che,
M.-C. J. Am. Chem. Soc. 2009, 121, 4405−4417.
(5) (a) Tisato, F.; Refosco, F.; Mazzi, U.; Bandoli, G.; Dolmella, A.;
Nicolini, M. Inorg. Chim. Acta 1991, 189, 97−103. (b) Gatehouse, B.
M.; Reichert, B. E.; West, B. Acta Crystallogr., Sect. B 1976, 32, 30−34.
(c) Chattopadhyay, S.; Drew, M. G. B.; Ghosh, A. Eur. J. Inorg. Chem.
2008, 1693−1701.
(6) (a) Belokon, Y. N.; Caveda-Cepas, S.; Green, B.; Ikonnikov, N.
S.; Khrustalev, V. N.; Larichev, V. S.; Moscalenko, M. A.; North, M.;
Orizu, C.; Tararov, V. I.; Tasinazzo, M.; Timofeeva, G. I.; Yashkina, L.
V. J. Am. Chem. Soc. 1999, 121, 3968−3973. (b) Belokon, Y. N.;
Blacker, A. J.; Carta, P.; Clutterbuck, L. A.; North, M. Tetrahedron
2004, 60, 10433−10447.
(7) Corden, J. P.; Errington, W.; Moore, P.; Wallbridge, M. G. H.
Chem. Commun. 1999, 323−324.
(8) Dreos, R.; Nardin, G.; Randaccio, L.; Siega, P.; Tauzher, G.;
Vrdoljak, V. Inorg. Chem. 2003, 42, 6805−6811.
cell at 25
0.1 °C, under a stream of purified N2 to avoid O2
reduction signals. Many electrodes as Au, Pt, and glassy-carbon (GC)
discs were tried as working electrodes, but a 2 mm GC disc (EG&G)
was finally chosen because it showed the best signal/noise ratio. Prior
to each experiment the electrode was micropolished, using successively
1 and 0.05 μm γ-alumina (Buehler) on a polishing cloth, and was then
sonicated three times for 5 min in DMSO. Finally, an electrochemical
cleaning was performed, which consisted of repeated potential scans in
DMSO between +0.7 and −2.0 V at a 50 mV/s scan rate. All of the
potentials were measured and quoted with respect to a Hg|Hg2Cl2,
NaCl saturated electrode (SCE) contained in a glass tube filled with
the supporting electrolyte solution, separated from the solution by a
Vycor frit and located close to the tip of the working electrode to
minimize the ohmic drop. The counter electrode was a Pt ring or rod
directly dipped in the solution. The complex concentration was about
1 mM.
Kinetic Studies. The cyclization kinetics were followed spec-
trophotometrically in the mixed solvent 60% methanol−40% H2O (v/
v, I = 0.1 M NaClO4). Generally, (2−4) × 10−4 M solutions of
complex were used. As the complexes did not dissolve completely, the
suspension was filtered before the addition of pyridine. The ligand
concentration was in large excess (pseudo-first-order conditions).
Observed rate constants were obtained from the linear plots of ln(A −
A∞), where A is the absorbance at the time t and A∞ is the final
absorbance, versus time. Kinetic data were analyzed with ORIGIN,
version 6.
(9) (a) Polson, S. M.; Hansen, L.; Marzilli, L. G. J. Am. Chem. Soc.
1996, 118, 4804−4808. (b) Marzilli, L. G.; Polson, S. M.; Hansen, L.;
Moore, S. J.; Marzilli, P. Inorg. Chem. 1997, 36, 3854−3860.
(10) (a) Dreos, R.; Felluga, A.; Nardin, G.; Randaccio, L.; Siega, P.;
Tauzher, G. Inorg. Chem. 2001, 40, 5541−5546. (b) Dreos, R.; Felluga,
A.; Nardin, G.; Randaccio, L.; Tauzher, G. Organometallics 2003, 22,
2486−2491.
ASSOCIATED CONTENT
* Supporting Information
■
S
CIF files giving X-ray data for compounds 1−3, Schemes S1
1
and S2, Figures S1−5, giving H NMR spectra in CD3OD of
the complexes [CoIII(chel)(CH2Cl)(S)], H NMR spectra in
1
(11) Dreos, R.; Randaccio, L.; Siega, P.; Vrdoljak, V. Croat. Chem.
Acta 2009, 82, 455−461.
CD3OD of complexes [CoIII(chelCH2)(py)(S)]+, a plot of first-
order kinetic data for the cyclization reaction of
[CoIII(3,3′,7,7′tmsalen)(CH2Cl)(S)], a plot of absorbance vs
time for the cyclization reaction of [CoIII(4,4′dmsalen)-
(CH2Cl)(S)], and a plot of absorbance vs time for the
cyclization reaction of [CoIII(4,4′dmsalen)(CH2Cl)(S)] in the
presence of 0.25 M py, and Tables S1−S4, giving crystallo-
graphic data and details of refinement for complexes 1−3,
complete bond distances and angles, H bond parameters for
complexes 2 and 3, and optimized geometries and electronic
energies (Eel) for CoIII(dmsalen) complexes. This material is
available free of charge via the Internet at http://pubs.acs.org.
(12) Dreos, R.; Mechi, L.; Randaccio, L.; Siega, P.; Zangrando, E.;
Ben Hassen, R. J. Organomet. Chem. 2006, 691, 3305−3309.
(13) Dreos, R.; Siega, P. Organometallics 2006, 25, 5180−5183.
(14) Dreos, R.; Nardin, G.; Randaccio, L.; Siega, P.; Tauzher, G.;
Vrdoljak, V. Inorg. Chim. Acta 2003, 349, 239−248.
(15) Siega, P.; Vrdoljak, V.; Tavagnacco, C.; Dreos, R. Inorg. Chim.
Acta 2012, 387, 93−99 and references therein.
(16) Bard, A. J.; Faulkner, L. R. Electrochemical Methods,
Fundamentals and Applications, 2nd ed.; Wiley: New York, 2001.
(17) (a) Saito, T.; Urabe, H.; Sasaki, Y. Transition Met. Chem. 1980,
5, 35−9. (b) Saito, T. Bull. Chem. Soc. Jpn. 1978, 51, 169−73.
(18) Wilkins, R. G. Kinetics and Mechanism of Reactions of Transition
Metal Complexes; VCH: Weinheim, Germany, 1991; p 89.
(19) de La Mare, P. B. D.; Swedlun, B. E. In The Chemistry of the
Carbon-Halogen Bond; Patai, S., Ed.; Wiley: Bristol, U.K., 1973;
Chapter 7.
AUTHOR INFORMATION
Corresponding Author
*E-mail for R.D.: dreos@units.it.
■
(20) Hummel, W.; Huml, K.; Burgi, H.-B. Helv. Chim. Acta 1988, 71,
1291−1302.
Notes
(21) Mechi, L.; Siega, P.; Dreos, R.; Zangrando, E.; Randaccio, L.
Eur. J. Inorg. Chem. 2009, 2629−2638.
The authors declare no competing financial interest.
(22) Otwinowski, Z.; Minor, W. Processing of X-ray Diffraction Data
Collected in Oscillation Mode. In Methods in Enzymology; Carter, C.
W., Jr., Sweet, R. M., Eds.; Academic Press: New York, 1997; Vol. 276,
pp 307−326.
REFERENCES
■
(1) For selected reviews on this subject see: (a) Canali, L.;
Sherrington, D. C. Chem. Soc. Rev. 1999, 28, 85−93. (b) Cozzi, P. G.
Chem. Soc. Rev. 2004, 33, 410−421. (c) Katsuki, T. Coord. Chem. Rev.
1995, 140, 189−214. (d) Katsuki, T. Chem. Soc. Rev. 2004, 33, 437−
444. (e) Wezenberg, S. J.; Kleij, A. W. Angew. Chem., Int. Ed. 2008, 47,
2354−2364. (f) Kleij, A. W. Chem. Eur. J. 2008, 14, 10520−10529.
(g) Kleij, A. W. Eur. J. Inorg. Chem. 2009, 193−205. (h) Jacobsen, E.
N. Acc. Chem. Res. 2000, 33, 421−431.
(2) Matsumoto, K.; Saito, B.; Katsuki, T. Chem. Commun. 2007,
3619−3627.
(3) (a) Cyriac, A.; Jeon, J. Y.; Varghese, J. K.; Park, J. H.; Choi, S. Y.;
Chung, Y. K.; Lee, B. Y. Dalton Trans. 2012, 41, 1444−1447.
(23) Sheldrick, G. M. SHELX97 Programs for Crystal Structure
Analysis (Release 97-2); University of Gottingen: Gottingen, Germany,
̈
̈
1998.
(24) Farrugia, J. J. Appl. Crystallogr. 1999, 32, 837−838.
(25) Frisch, M. J. G.; Trucks, W.; Schlegel, H. B.; Scuseria, G. E.;
Robb, M. A.; Cheeseman, J. R.; Scalmani, G.; Barone, V.; Mennucci,
B.; Petersson, G. A.; Nakatsuji, H.; Caricato, M.; Li, X.; Hratchian, P.;
Izmaylov, A. F.; Bloino, J.; Zheng, G.; Sonnenberg, J. L.; Hada, M.;
Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa, J.; Ishida, M.; Nakajima,
T.; Honda, Y.; Kitao, O.; Nakai, H.; Vreven, T.; Montgomery, J. A., Jr.;
919
dx.doi.org/10.1021/om401038v | Organometallics 2014, 33, 909−920

Organometallics
Article
Peralta, J. E.; Ogliaro, F.; Bearpark, M.; Heyd, J. J.; Brothers, E.; Kudin,
K. N.; Staroverov, V. N.; Kobayashi, R.; Normand, J.; Raghavachari, K.;
Rendell, A.; Burant, J. C.; Iyengar, S. S.; Tomasi, J.; Cossi, M.; Rega,
N.; Millam, J. M.; Klene, M.; Knox, J. E.; Cross, J. B.; Bakken, V.;
Adamo, C.; Jaramillo, J.; Gomperts, R.; Stratmann, R. E.; Yazyev, O.;
Austin, A. J.; Cammi, R.; Pomelli, C.; Ochterski, J. W.; Martin, R. L.;
Morokuma, K.; Zakrzewski, V. G.; Voth, G. A.; Salvador, P.;
Dannenberg, J. J.; Dapprich, S.; Daniels, A. D.; Farkas, O.;
Foresman, J. B.; Ortiz, J. V.; Cioslowski, J.; Fox, D. J. Gaussian 09,
Revision A.02; Gaussian, Inc., Wallingford, CT, 2009.
(26) Zhao, Y.; Truhlar, D. G. Theor. Chem. Acc. 2008, 120, 215−241.
(27) Schaefer, A.; Horn, H.; Ahlrichs, R. J. Chem. Phys. 1992, 97,
2571−2577.
(28) Schaefer, A.; Huber, C.; Ahlrichs, R. J. Chem. Phys. 1994, 100,
5829−5835.
(29) Weigend, F.; Ahlrichs, R. Phys. Chem. Chem. Phys. 2005, 7,
3297−3305.
(30) Feller, D. J. Comput. Chem. 1996, 17, 1571−1586.
(31) Schuchardt, K. L.; Didier, B. T.; Elsethagen, T.; Sun, L.;
Gurumoorthi, V.; Chase, J.; Li, J.; Windus, T. L. J. Chem. Inf. Model.
2007, 47, 1045−1052.
(32) Miertus,
129.
̌
S.; Scrocco, E.; Tomasi, J. Chem. Phys. 1981, 55, 117−
(33) Tomasi, J.; Persico, M. Chem. Rev. 1994, 94, 2027−2094.
(34) Cances
(35) Cances
3032−3041.
(36) Mennucci, B.; Cances
10506−10517.
̀
, E.; Mennucci, B. J. Math. Chem. 1998, 23, 309−326.
̀
, E.; Mennucci, B.; Tomasi, J. J. Chem. Phys. 1997, 107,
̀
, E.; Tomasi, J. J. Phys. Chem. B 1997, 101,
(37) Mennucci, B.; Cammi, R.; Tomasi, J. J. Chem. Phys. 1998, 109,
2798−2807.
(38) Scalmani, G.; Frisch, M. J. J. Chem. Phys. 2010, 132, 114110.
(39) Marenich, A. V.; Cramer, C. J.; Truhlar, D. G. J. Phys. Chem. B
2009, 113, 6378−6396.
920
dx.doi.org/10.1021/om401038v | Organometallics 2014, 33, 909−920