Mo(IV) and W(IV) cyanido complexes with Schiff bases. Synthesis, X-Ray single crystal structures, physicochemical properties and quantum chemical calculations

Article abstract of DOI:10.1016/j.poly.2013.07.030

In the reaction of salicylaldehyde derivatives and aminoethanol with Mo(IV) or W(IV) cyanido complexes, six new salts were isolated and characterized by physicochemical measurements. The single crystal X-ray analysis of four salts of the formula (PPh

Full text of DOI:10.1016/j.poly.2013.07.030

Polyhedron 68 (2014) 112–121
Contents lists available at ScienceDirect
Polyhedron
journal homepage: www.elsevier.com/locate/poly
Mo(IV) and W(IV) cyanido complexes with Schiff bases. Synthesis, X-ray
single crystal structures, physicochemical properties and quantum
chemical calculations
1
⇑
Janusz Szklarzewicz , Michael Skaisgirski , Patrycja Paciorek, Katarzyna Kurpiewska, Piotr Zabierowski,
´
Mariusz Radon
Faculty of Chemistry, Jagiellonian University, R. Ingardena 3, 30-060 Kraków, Poland
a r t i c l e i n f o
a b s t r a c t
Article history:
In the reaction of salicylaldehyde derivatives and aminoethanol with Mo(IV) or W(IV) cyanido complexes,
six new salts were isolated and characterized by physicochemical measurements. The single crystal X-ray
analysis of four salts of the formula (PPh4)2[M(CN)3O(LL)]ꢀnH2O, (where LL = Schiff bases formed
in situ in the reaction of aminoethanol and 5-bromo-, 5-chloro-, 5-methoxy-, 3,5-dichloro- or 5-bromo-
3-methoxy-substituted salicylaldehyde, M = Mo or W, n = 1, 1.5, 2 or 5 water molecules) reveals a dis-
torted octahedral geometry of the anions. All the complexes were characterized by elemental analysis,
IR and UV–Vis spectroscopy and by cyclic voltammetry measurements. The role of the salicylaldehyde
substituents on the structures and physicochemical properties is discussed. The results are compared
with quantum chemical calculations, indicating that, contrary to literature data, even strong hydrogen
bonds do not influence the anion structure.
Received 23 April 2013
Accepted 2 July 2013
Available online 24 October 2013
Keywords:
Salicylaldehyde
Cyanido complexes
Molybdenum
Tungsten
Crystal structure
Cyclic voltammetry
Ó 2013 Elsevier Ltd. All rights reserved.
1. Introduction
with organic ligands [2–20]. Many efforts were undertaken to syn-
thesize new complexes of this type, but still less than 50 were
The structures of anions (or cations) of d-electron metal com-
plexes are very often discussed in terms of ligand–metal interac-
tions (for example the trans effect), but they also serve as an
interesting source of information for the interactions of complex
cations (or anions) with solvent molecules. During the last few
years, much attention has been directed towards the syntheses of
Mo(IV) and W(IV) oxido-cyanido complexes as these complexes
are especially good for structure discussions due to their octahe-
dral structures with fixed ligand positions [1]. In complexes with
[M(CN)4OL]nꢁ anions (M = Mo or W, L denotes a monodentate li-
gand) the ligand L is always trans to the M@O bond. In complexes
with [Mo(CN)3O(LL)]nꢁ anions (LL denotes a bidentate ligand) one
end of LL ligand is also trans to the M@O bond, while the second
donor atom is in a cis position. The stability of the complex struc-
ture is very promising for an investigation of the correlations be-
tween the structure and physicochemical properties.
known, until now. The number of crystal structures resolved is
even smaller, which makes systematic comparisons not always
possible. Thus, it is difficult to draw correlations between the li-
gand and physicochemical properties for a given complex. It is
known, however, that complexes with the same anion and differ-
ent cations show anion distortion, discussed mainly in terms of
possible long range hydrogen interactions influencing the anion
structure [1].
In this work we synthesized several new complexes of Mo(IV)
and W(IV) with Schiff base ligands derived from salicylaldehyde
and 2-aminoethanol. To study the small effects of ligand and sol-
vent interactions we decided to use substituted salicylaldehydes.
The selected Schiff bases are identical to those used by us for cop-
per(II) complexes (with 5-chloro, 5-bromo and 5-bromo-3-meth-
oxy substituents), but the list of substituents was increased to
include 3,5-dichloro and 5-methoxy [21]. It has been found that
small changes on these ligands can dramatically influence the long
distance interactions for copper complexes, even by changing the
copper coordination number, which was interpreted in terms of
hydrogen bonding effects. However, for 4- and 5-d electron metals,
the hydrogen bonding interactions may not be so important as is
suggested for 3-d metals. We also carried out quantum chemical
calculations for isolated complex anions to examine if the observed
deformation of the Mo(IV) or W(IV) anion may be reproduced
The main problem is that for molybdenum(IV) and tungsten(IV)
cyanido complexes only a few complexes are known, especially
⇑
Corresponding author. Tel.: +48 12 6632231; fax: +48 12 6340515.
E-mail addresses: szklarze@chemia.uj.edu.pl, janusz.szklarzewicz@uj.edu.pl (J.
Szklarzewicz).
1
On leave from the Institut für Anorganische und Analytische Chemie, Albert-
Ludwigs-Universität Freiburg, Albertstr. 21, 79104 Freiburg, Germany.
0277-5387/$ - see front matter Ó 2013 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.poly.2013.07.030

J. Szklarzewicz et al. / Polyhedron 68 (2014) 112–121
113
without considering the hydrogen bonding network present in the
crystal structure.
junction potential; the final values are reported versus the stan-
dard hydrogen electrode (SHE).
2. Experimental
2.2. Theoretical calculations
2.1. Materials and methods
Quantum chemical calculations were performed using Turbo-
mole 5.9 [23] and GAUSSIAN 09 [24] packages. The structures of the
isolated [Mo(CN)3O(LL)]2ꢁ anions (without counterions) were opti-
mized in Turbomole at the DFT:B3LYP/def2-TZVP level, both in the
gas phase and in a conductor-like screening model (COSMO) with
K3Na[Mo(CN)4O2]ꢀ6H2O and K3Na[W(CN)4O2]ꢀ6H2O were syn-
thesized according to the literature methods [22,16]. All other
chemicals were of analytical grade (Aldrich) and were used as sup-
plied. Microanalyses of carbon, hydrogen and nitrogen were per-
e
= 4. In the case of 3, the torsion angle C4–C5–O3–H3A (see Fig
formed using
a Vario Micro Cube elemental analyzer. Solid
2) was constrained to its value found in the crystal structure in or-
der to prevent spurious rotation of the C–CH2OH group leading to
the formation of intramolecular hydrogen bonding with one of the
CN ligands (which is not present in the crystal structure and thus
should be considered as an artifact of the present simple model,
not accounting for the environment of the [M(CN)3O(LL)]2ꢁ an-
ions). Mulliken and Hirshfeld atomic charges on the phenolic O
atom of each LL ligand were obtained from single-point
DFT:B3LYP/6-31G(d) calculations with GAUSSIAN 09, performed for
the isolated LLꢁ fragments at their geometries cut from the previ-
ously optimized [Mo(CN)3O(LL)]2ꢁ structures.
samples for IR spectroscopy were compressed as KBr pellets and
the IR spectra were recorded on a Bruker EQUINOX 55 FT-IR spec-
trophotometer. Electronic absorption spectra (qualitative) were
measured with a Shimadzu UV–Vis–NIR UV-3600 spectrophotom-
eter. Diffuse reflectance spectra were measured in BaSO4 pellets
with BaSO4 as a reference using a Shimadzu UV-3600 equipped
with an ISR-3100 attachment. Cyclic voltammetry measurements
were carried out in DMSO (dimethyl sulfoxide) with [Bu4N]PF6
(0.10 M) as the supporting electrolyte, using Pt working and coun-
ter and Ag/AgCl reference electrodes on an AUTOLAB/PGSTAT
128 N Potentiostat/Galvanostat. E1/2 values were calculated from
the average anodic and cathodic peak potentials, E1/2 = 0.5(Ea + Ec).
The redox potentials were calibrated versus ferrocene (0.440 V ver-
sus SHE), which was used as an internal potential standard for
measurements in organic solvents to avoid the influence of a liquid
2.3. Synthesis
The ligands used in the synthesis are presented in Scheme 1.
They were synthesized in situ by the reaction of the appropriate
substituted salicylaldehyde and aminoethanol. The formed Schiff
bases were denoted as Hclsb, Hbrsb, Hmetsb, Hdclsb and Hbrmetsb
for 5-chloro-, 5-bromo-, 5-methoxy-, 3,5-dichloro- and 5-bromo-
3-methoxy-salicylaldehyde, respectively.
2.3.1. Synthesis of [PPh4]2[Mo(CN)3O(brsb)]ꢀ3H2O (1)
2-Aminoethanol (39 ll, 0.650 mmol) and 5-bromosalicylalde-
hyde (130.6 mg, 0.650 mmol) in 15 cm3 ethanol were refluxed
for 10 min. The solution was cooled to room temperature and a
solution of K3Na[Mo(CN)4O2]ꢀ6H2O (378.8 mg, 0.788 mmol) in
25 cm3 water was added. The reflux was continued for 20 min.
Thereafter
tetraphenylphosphonium
chloride
(750.0 mg,
2.00 mmol) was added and the solution was left for crystallization.
After a few hours sparkling dark green crystals were obtained. The
compound was filtered, washed with water and dried in air. Yield
673.6 mg, 87%. MW 1165.89. Anal. Calc. for C60H55BrMoN4O6P2: C,
61.81; H, 4.75; N 4.81. Found: C, 62.16; H, 4.565; N, 4.96%.
Fig. 1. Structure of the [Mo(CN)3O(clsb)]2ꢁ ion in 2a with the atom labelling
scheme and 50% displacement ellipsoids.
Fig. 2. Structure of the [Mo(CN)3O(metsb)]2ꢁ ion in 3 with the atom labelling
scheme and 50% displacement ellipsoids.
Scheme 1.

114
J. Szklarzewicz et al. / Polyhedron 68 (2014) 112–121
2.3.2. Synthesis of [PPh4]2[Mo(CN)3O(clsb)]ꢀ2H2O (2)
compound was filtered, washed with water three times and dried
in air. Yield: 135.6 mg, 24%. MW 1119.8 Anal. Calc. for C60H50Cl2-
MoN4O4P2: C, 64.35; H, 4.50; N, 5.00. Found: C, 64.17; H, 4.45; N,
4.97%.
The synthetic procedure was identical as for 1, but as the alde-
hyde 5-chlorosalicylaldehyde (102.0 mg, 0.650 mmol) was used.
Some of the dark green crystals formed were kept in the mother li-
quor for crystal analysis (salt 2a) while the rest were filtered off,
washed with water and dried in air. Salts 2 and 2a (dark green)
were solvatomorphs. Yield 123.0 mg, 16%. MW 1103.53. Anal. Calc.
for C60H53ClMoN4O5P2: C, 65.31; H, 4.84; N, 5.08. Found: C, 65.57;
H, 4.838; N, 5.07%.
2.3.6. Synthesis of [PPh4]2[Mo(CN)3O(brmetsb)]ꢀ2H2O (6)
2-Aminoethanol (60 ll, 1.00 mmol) and 5-bromo-3-methoxy-
salicylaldehyde (153 mg, 0.662 mmol) in 10 cm3 ethanol were re-
fluxed for 10 min. The solution was cooled to room temperature
and a solution of K3Na[Mo(CN)4O2]ꢀ6H2O (497 mg, 1.03 mmol) in
25 cm3 water was added. The reflux was continued for 15 min.
2.3.3. Synthesis of [PPh4]2[Mo(CN)3O(metsb)]ꢀ5H2O (3)
The synthetic procedure was identical as for 1, but as the alde-
hyde 5-methoxysalicylaldehyde (81 ll, 0.650 mmol) was used.
Yield 174.6 mg, 23%. MW 1153.03. Anal. Calc. for C61H62MoN4O9P2:
C, 63.54; H, 5.42; N, 4.86. Found: C, 63.37; H, 5.386; N, 4.85%.
Thereafter
a
solution of tetraphenylphosphonium bromide
(803 mg, 1.92 mmol) in 20 cm3 water was added and the solution
was left for crystallization. After a few hours, black sparkling crys-
tals were obtained. The compound was filtered, washed with water
three times and dried in air. Yield: 642 mg, 87%. MW 1119.8. Anal.
Calc. for C61H55BrMoN4O6P2: C, 62.20; H, 4.71; N, 4.76. Found: C,
62.30; H, 4.52; N, 4.65%.
2.3.4. Synthesis of [PPh4]2[W(CN)3O(brmetsb)]ꢀ1.25H2O (4)
2-Aminoethanol (60 ll, 1.00 mmol) and 5-bromo-3-methoxy-
salicylaldehyde (152.1 mg, 0.658 mmol) in 15 cm3 ethanol were
refluxed for 10 min. The solution was cooled to room temperature
and a solution of K3Na[W(CN)4O2]ꢀ6H2O (568.3 mg, 1.00 mmol) in
37.5 cm3 water was added. Additionally 2 M HCl (400
ll) was
added, too. The reflux was continued for 20 min. Thereafter a warm
solution of tetraphenylphosphonium chloride (562.3 mg,
1.5 mmol) in 30 cm3 water was added and the solution was left
for crystallization. After a few hours sparkling violet crystals were
obtained. The compound was filtered, washed with water and
dried in air. Yield: 131.9 mg, 16%. MW 1256.80. Anal. Calc. for C61-
H54BrWN4O5.25P2: C, 58.30; H, 4.33; N, 4.46. Found: C, 58.10; H,
4.288; N, 4.36%.
2.4. Crystallographic data collection and structure refinement
Crystals of 2a, 3 and 4 suitable for X-ray analysis were selected
from the materials prepared as described in the Experimental Sec-
tion. Intensity data were collected on a SuperNova diffractometer
(Agilent Technologies) at 110–113 K, equipped with an Atlas detec-
tor and a microfocus Mo Ka (k = 0.71073 Å) radiation source oper-
ating at 50 kV and 1 mA for 2a, 3 and 4. Data were processed using
CRYSALISPro. The crystal data and details of data collection and
structure refinement parameters are summarized in Table 1. The
positions of most of the atoms were determined by direct methods
2.3.5. Synthesis of [PPh4]2[Mo(CN)3O(dclsb)]ꢀH2O (5)
2-Aminoethanol (30 ll, 0.501 mmol) and 3,5-dichlorosalicylal-
dehyde (96.9 mg, 0.507 mmol) in 7.5 cm3 ethanol were refluxed
for 10 min. The solution was cooled to room temperature and a
solution of K3Na[Mo(CN)4O2]ꢀ6H2O (245 mg, 0.510 mmol) in
19 cm3 water as well as 2 M HCl (0.1 ml) were added. The reflux
was continued for 15 min. Thereafter a solution of tetraphenyl-
phosphonium chloride (284 mg, 0.758 mmol) in 15 cm3 water
was added and the solution was left for crystallization. After a
few hours, very small black sparkling crystals were obtained. The
(SIR-97) [25], whilst other non-hydrogen atoms were located on dif-
ference Fourier maps. Most of the non-hydrogen atoms were re-
fined anisotropically using weighted full-matrix least-squares on
F2. Atoms that exhibited disorder were refined isotropically. All
hydrogen atoms bonded to carbon were included in the structure
factor calculations at idealized positions. The structures were re-
fined by the SHELXL program [26]. Structural description graphics
were performed with the program Mercury 3.0.
Table 1
Crystal data and structure refinement parameters for 2a, 3 and 4.
2a
3
4
Empirical formula
Formula weight
Crystal size (mm)
Crystal system
Space group
a (Å)
C
60H59.5ClMoN4O8.25P2
C61H62MoN4O9P2
1153.03
C61H54BrWN4O5.25P2
1252.78
0.1 ꢂ 0.1 ꢂ 0.35
monoclinic
P21/n
9.83510(10)
23.8400(3)
23.6775(3)
90.00
98.4180(10)
90.00
5491.82(11)
4
1162.48
0.55 ꢂ 0.2 ꢂ 0.15
0.4 ꢂ 0.2 ꢂ 0.15
triclinic
triclinic
ꢀ
ꢀ
P1
P1
12.597(5)
14.203(5)
17.232(5)
73.398(5)
72.505(5)
85.310(5)
2817.8(17)
2
12.083(5)
13.815(5)
18.024(5)
72.898(5)
75.469(5)
85.204(5)
2783.5(17)
2
b (Å)
c (Å)
a
(°)
b (°)
c
(°)
V (Å3)
Z
T (K)
113(5)
1.374
0.396
110(10)
1.376
0.354
110(2)
1.515
2.943
Dc (Mg/m3)
l
(mmꢁ1
)
Reflections measured
40586
24537
27552
Reflections unique
12648
11953
11371
Reflections observed [I > 2
r
(I)]
10643
0.0354
0.0802
9990
9734
0.0272
0.0533
1.034
R1[F2 > 2
wR2(F2)a
S
r
(F2)]a
0.0371
0.0928
1.013
1.033
P
P
P
a
w = 1/[
r
2(Fo2) + (0.0297P)2 + 0.0712P] where P = (Fo2 + 2Fc2)/3, R1
=
(|F0| ꢁ |Fc|)/ (|F0|) and wR2 = { [w(F0 ꢁ Fc2)2}1/2
.

J. Szklarzewicz et al. / Polyhedron 68 (2014) 112–121
115
Fig. 3. Structure of the [W(CN)3O(brmetsb)]2ꢁ ion in 4 with the atom labelling
scheme and 50% displacement ellipsoids.
3. Results and discussion
Fig. 4. The arrangement with distinguishable layers of anions and cations in 2a.
3.1. Crystal structures
The asymmetric part of the unit cell in the crystal structures of
2a, 3 and 4 contain the anion, two cations and water molecules.
The anions are shown in Figs. 1–3 (in those figures hydrogen atoms
are omitted for clarity). Selected bond lengths and angles for the
determined crystal structures of 2a, 3 and 4 are listed in Table 2.
The organic ligand coordinates in a bidentate mode to afford a
distorted octahedral geometry with three cyanido ligands and a
nitrogen atom of the Schiff base in the equatorial positions, and
with the oxido ligand and the phenolic oxygen occupying the axial
positions.
The anions are distorted: the average O1–M–O angle is 176.69°
and the average C1–M–O1 and N4–M–O1 angles are 93.68° and
97.44°, respectively. A similar type of deformation is observed for
all Mo(IV) and W(IV) complexes of the [M(CN)3O(LL)]nꢁ (n = 1 or
2) type known to date [1,18,7,19,20,27]. The highest distortion,
with the O–Mo–O angle equal to 164.30°, was found for the
Table 2
Fig. 5. The arrangement with distinguishable layers of anions and cations in 3.
Selected bond lengths [Å] and angles [°] for [PPh4]2[Mo(CN)3O(clsb)]ꢀ5H2O (2a),
[PPh4]2[Mo(CN)3O(metsb)]ꢀ5H2O (3) and [PPh4]2[W(CN)3O(brmetsb)]ꢀ1.5H2O (4) with
estimated standard deviations in parentheses. M = Mo or W.
2a
3
4
Bond lengths
M@O1
M–O2
M–N4
M–C1
M–C2
M–C3
C1–N1
C2–N2
C3–N3
1.694(2)
2.086(2)
2.184(2)
2.160(2)
2.193(3)
2.200(3)
1.153(2)
1.154(4)
1.147(5)
1.695(2)
2.069(2)
2.198(2)
2.162(3)
2.207(3)
2.201(3)
1.159(3)
1.157(3)
1.156(3)
1.727(2)
2.072(2)
2.132(2)
2.144(3)
2.187(3)
2.172(3)
1.156(3)
1.153(4)
1.154(4)
Bond angles
N4–M–O2
O1–M–O2
C1–M–N4
C2–M–C3
O1–M–N4
O1–M–C1
C2–M–N4
C3–M–N4
C1–M–O2
C2–M–O2
C3–M–O2
81.76(7)
176.69(7)
168.90(8)
168.44(9)
95.38(8)
95.70(8)
92.43(9)
85.86(10)
87.14(7)
85.59(7)
82.86(8)
81.61(7)
82.83(8)
175.67(7)
170.11(7)
169.95(8)
94.44(8)
95.41(8)
91.87(9)
87.18(8)
88.52(7)
85.88(7)
84.08(7)
177.49(7)
167.55(10)
167.93(11)
99.63(9)
92.64(10)
85.78(9)
89.34(9)
84.92(9)
83.07(9)
85.38(9)
Fig. 6. The arrangement with distinguishable layers of anions and cations in 4.

116
J. Szklarzewicz et al. / Polyhedron 68 (2014) 112–121
Fig. 9. Hydrogen bonding network in 4.
Table 3
The hydrogen bond parameters in salts 2a–4.
Complex
D–A
d(D–A) (Å)
DHA (°)
2a
N1–O22
N1–O220
N2–O21
N2–O210
N3–O23
O1–O26
O22–O25
O21–O25
O23–O24
O24–O25
O24–O3
2.914
2.970
3.001
2.926
3.034
2.790
2.780
2.771
2.757
2.712
2.759
177.59
173.99
174.33
164.06
176.12
–
173.03
161.71
164.96
166.51
167.41
Fig. 7. Hydrogen bonding network in 2a.
3
N1–O21
N1–O210
N2–O24
N2–O240
N3–O25
O21–O23
O23–O24
O22–O25
O22–O23
2.950
2.936
2.933
3.013
3.019
2.719
2.855
2.807
2.559
168.33
174.18
159.18
177.30
167.95
179.21
168.63
163.25
163.81
[PPh4]2[Mo(CN)3O(pzac)]ꢀ2H2O complex [27]. The average M–C
and C„N bond distances are 2.171 and 1.150 Å, respectively and
these are very close to those found in the structure of the
[Mo(CN)3O(pic)]2ꢁ anion, where the bond lengths are 2.161 and
1.148 Å, respectively [18]. The cyanido ligands are not bound with
the same strength, usually the M–C bond trans located to the Mo–N
is the shortest, whilst its C„N bond is the longest.
The crystal packing of complexes 2a–4 are shown in Figs. 4–6. It
can be seen that in the case of all the salts the layers of cations are
separated by the layers of anions. Thus, there are no significant dif-
ferences in the long distance packing.
4
N1–O3
2.810
2.706
2.772
2.879
168.93
–
–
N2–O22
O22–O3
O1–O21
177.48
The anions and cations are bonded by a network of hydrogen
bonds, as presented in Figs. 7–9. Selected bond lengths are given
in Table 3. In 2a all the cyanido ligands and O1 are involved in a
three dimensional network presented in Fig. 7. The two cyanido li-
gand’s nitrogens (N1 and N2) form double bridges with water mol-
ecules, which form further water–water hydrogen bonds, yielding
a plane structure of the anions. Every second Mo@O oxygen as well
as every second organic ligand is located above or below the plane,
Fig. 8. Hydrogen bonding network in 3.

J. Szklarzewicz et al. / Polyhedron 68 (2014) 112–121
117
as shown in the bottom part of Fig. 7. A very reach network of
hydrogen bonds was found in 2a, therefore relatively large separa-
tions of the anions are observed in the structure. This kind of layer
arrangement is probably responsible for the easy dehydration of 2a
into 2, when three water molecules are released. The repeated syn-
thesis and analysis indicated only two water molecules in dried
samples of 2, while crystals kept in solution (2a) remain
pentahydrated.
As it can be seen in Table 2, the hydrogen bond formation has little
influence on the M@O, M–C and C„N bond lengths. For example in
3, the oxygen atom in Mo@O does not form hydrogen bonds. Fur-
thermore, this bond is the longest (1.739 Å) one, which is contrary
to expectation, but it is still very close to the value in 4 (1.727 Å).
Similarly, the N3 atom is either involved in single or no hydrogen
bonds, which has no influence on the C–N bond distances for salts
2–4.
In salt 3, the hydrogen bonding network is essentially similar to
that in 2a, the main difference is that the O1 oxygen does not form
hydrogen bonds with the water molecules, as shown in the bottom
part of Fig. 8. In 4 the hydrogen bonds arrange the anions in chains
and not in planes. The oxygen atom from the W@O bond forms a
hydrogen bond with one water molecule, but nitrogen N3 from
the cyanido ligand is not involved in such an interaction. The pre-
sented data for the hydrogen bonds in Table 3 reveal relatively
strong interactions, the D–H–A angle falling in the 159–178° range.
There has been a long discussion on the influence of hydrogen
bonds on the structure of [M(CN)3O(LL)]n- anions (M = Mo or W),
especially the M@O bond distance, so far without definite conclu-
sions [1,19]. Here, for the first time, we have obtained a series of
compounds with similar organic ligands and very different hydro-
gen bonds patterns involving the O1 atom (M@O group). While in 3
this oxygen atom remains free, in 2a and 4 it is involved in rela-
tively strong hydrogen bonds with water molecules. Despite this
profound difference, 2a and 3 have nearly identical M@O distances
(Table 4). Compound 4 has, clearly, a much longer M@O distance
than 2 and 3 for it contains W instead of Mo.
Table 4 contains the M@O distances calculated at the DFT level
in the gas-phase and in the COSMO model, as well as the experi-
mental data. The purpose of the COSMO approach is to roughly
simulate the effect of a polarizable environment in the condensed
phase, which is completely neglected in standard (gas-phase) cal-
culations. As can be seen in Table 4, the calculated M@O bond dis-
tances agree with experimental data very well, with discrepancies
below 0.005 Å (the COSMO distances, being 0.005–0.006 Å shorter
from the gas-phase ones, are closer to the experimental values). Gi-
ven that both sets of calculations refer to isolated [M(CN)3O(LL)]2ꢁ
anions, i.e. they completely neglect the intermolecular interactions
occurring in the crystal structures, the good agreement between
the calculations and the crystal structures goes in line with one
of the conclusions above: the M@O distance is hardly affected by
hydrogen bonding.
Table 4
Comparison of experimental and calculated M@O and M–O (phenolic) bond distances
and calculated charges on the O (phenolic) atom in the free ligands.
Bond distance (Å)
Salt
2
3
4
M@O
found
1.694
1.695
1.727
calculateda 1.689 (1.694) 1.692 (1.698) 1.723 (1.729)
M–O
found
2.086
2.069
2.072
(phenolic)
calculateda 2.170 (2.127) 2.142 (2.101) 2.127 (2.103)
O (phenolic) chargeb
ꢁ0.67
ꢁ0.69
ꢁ0.64
(ꢁ0.43)
(ꢁ0.45)
(ꢁ0.41)
a
Structures of isolated [M(CN)3O(LL)]2ꢁ anions optimized in the gas phase (in
parenthesis in the COSMO model) at the DFT:B3LYP/def2-TZVP level.
b
Mulliken (Hirshfeld) charge from single-point DFT:B3LYP/6-31G(d) calculations
on each LL taken from the crystal structure of the respective salt.
Fig. 10. IR spectra of a) 1 (black line), 2 (red line) and 3 (green line), b) 4 (red line), 5 (green line) and 6 (black line). Spectra measured in KBr in the 875–975 cmꢁ1 region.
(Color online)
Fig. 11. IR spectra (in the 2000–2200 cmꢁ1 region) of a) 1 (black line), 2 (red line) and 3 (green line) b) 4 (red line), 5 (green line) and 6 (black line). (Color online)

118
J. Szklarzewicz et al. / Polyhedron 68 (2014) 112–121
Fig. 12. The reflectance spectra of a) 1 (black line), 2 (red line) and 3 (green line) b) 4 (red line), 5 (green line) and 6 (blue line) after Kubelka–Munk transformation. BaSO4 as
internal standard. (Color online)
Fig. 13. UV–Vis spectra of 1 (a) 2 (b), 3(c) and 5 (d) vs. time in ethanol. T = 25.7 °C, d = 1 cm, time intervals 180 s, 15 spectra measured, arrows indicate the direction of the
changes.
Fig. 14. UV–Vis spectra of 6 (a) and 4 (b) vs. time in ethanol. T = 25.7 °C, d = 1 cm, time intervals 180 s, 15 spectra measured, arrows indicate the direction of the changes.
Moreover, it seems that different LL ligands in compounds 1–3,
5 and 6 point to very similar Mo@O distances. While experimental
values are available only for 2 and 3, the DFT ones can be computed
for all of them, and they turn out to be very similar (gas-phase val-
ues for the remaining species: 1.688 Å for 1, 1.686 Å for 5 and
1.688 Å for 6). In contrast, the M-O (phenolic) bond distances show
larger variations with respect to the organic ligand and also a
worse agreement between the calculated and experimental values.

J. Szklarzewicz et al. / Polyhedron 68 (2014) 112–121
119
The M–O (phenolic) bond is a softer degree of freedom than the
M@O distance, and thus it is presumably more affected by short-
comings of the computational method and the model used here
(isolated anion). Nonetheless, the differences in the M–O (pheno-
lic) bond distances between compounds 2–4 are nicely reproduced
by the calculations, the DFT distances being systematically longer
than the experimental ones (cf. Table 4). Interestingly, the M–O
distance seems to be correlated with the atomic charge on the phe-
nolic O atom: with increasing negative charge, the bonds become
shorter.
2103 cmꢁ1, for 4 at 2084 and 2104 cmꢁ1, for 5 at 2095 and
2110 cmꢁ1 and for 6 at 2089 and 2103 cmꢁ1. For the Mo(IV) com-
plexes, the lowest energy band can be arranged in the series: 5-
methoxy- < 5-bromo-3-methoxy- < 5-bromo- < 5-chloro- = 3,5-di-
chloro-, while the highest energy one can be divided in two groups:
mC„N ꢃ 2103 cmꢁ1 for the salts with methoxy- and bromo-meth-
oxysalicylaldehyde and mC„N ꢃ 2110 cmꢁ1 for the other salts.
3.3. UV–Vis spectra
The sample reflectance electronic spectra in the UV–Vis range
for salts 1–3 are presented in Fig. 12. In general the spectra mea-
sured in the solid state for all the Mo(IV) complexes are similar
(1–6). The lowest energy bands (at ca 650 nm) appear to be of d–
d origin. In the reflectance spectra these forbidden transitions are
well visible and their intensity is similar to the allowed transitions.
However, for the Mo(IV) and W(IV) cyanido complexes with biden-
tate organic ligands the d–d transitions are overlapped with the
charge-transfer ones which are also well visible in the spectra mea-
sured in solution.
3.2. IR spectra
The IR spectra of salts 1–6 are presented in Figs. 10 and 11 and
are in agreement with the X-ray crystal structures for salts 2a–4.
The most interesting features for the Mo(IV) and W(IV) complexes
are the
m
M@O (in the 900–1000 cmꢁ1 range) and
m
C„N bands (in the
2050–2150 cmꢁ1 range). The
m
M@O bands for salts 1–6 are observed
at 928, 925, 924, 922, 934 and 934 cmꢁ1 respectively. Thus the
studied substituents of the salicylaldehyde can be arranged for
Mo(IV) complexes in order of increasing band energy as: 5-meth-
oxy- < 5-chloro- < 5-bromo- < 3,5-dichloro-=5-bromo3-methoxy-.
The spectra in EtOH for all the complexes are presented in
Figs. 13 and 14. The observed instability of the complexes in EtOH
solution results in synthetic problems (low yield) and problems
with recrystallisation of the complexes for structure measure-
The
m ,
C„N bands for 1 are observed at 2094, 2098sh and 2110 cmꢁ1
for 2 at 2095, 2098sh and 2110 cmꢁ1, for 3 at 2077sh, 2087 and
Fig. 15. The electronic spectra of 6 (a) and 4 (b) in different solvents. Qualitative spectra normalized for absorbance = 1 for the most intense CT band. T = 25 °C, d = 1 cm,
qualitative spectra.
Fig. 16. The cyclic voltammetry of complexes 1–6 in DMSO. The overlaid curves above each main curve represent different potential scan rates – 20, 50, 100, 200, 500,
1000 mV/s with raising peak current, respectively. In the upper left corner a scale of current is presented. T = 25 °C, 20 mg of sample per 5 ml of solvent.

120
J. Szklarzewicz et al. / Polyhedron 68 (2014) 112–121
ments. To isolate salt 4 it was necessary to stabilize the aldehyde
with excess of aminoethanol over aldehyde.
tems. What is interesting is that we do not observe well defined
cathodic waves at ca ꢁ1.3 V attributed to the reduction of the li-
gands, as in the case of previously reported copper complexes [21].
The Mo(IV)/Mo(V) redox potential values determined herein lay
in the range between the slightly lower value reported for the sal-
icylaldehydehydrazone complex (ꢁ629 mV versus Fc/Fc+ [13]) and
the slightly higher value reported for the picolinic acid complex
(ꢁ301 mV versus Fc/Fc+ [18]). The difference between the redox
potentials of the corresponding picolinc acid complexes of Mo(IV)
and W(IV) is lower (117 mV lower value for the potential of the
W(IV) complex) than that of the corresponding salicylidene-2-eth-
anolamine complexes 4 and 6 (287 mV).
These data indicate that the salicyaldehydehydrazone com-
plexes of Mo(IV) are better reductants than the corresponding sal-
icylidene-2-ethanolamine and picolinic acid complexes. What is
more, the fourth oxidation state of tungsten is less stabilized in
complexes with salicylidene-2-ethanolamine than with picolinic
acid.
For a proper band assignment, the spectra in a series of organic
solvents were also measured and are presented in Fig. 15 for
Mo(IV) and W(IV) representatives, with the same LL ligand (salts
6 and 4). The lowest energy bands (at ca. 600 nm) can be attributed
to d–d transitions overlapped with MLCT (metal-to-ligand charge
transfer) transitions. Such an assignment is in agreement with
the literature data for [M(CN)3O(LL)]nꢁ type ions and with the rel-
atively low intensity of this band (for the Mo complexes) [1]. These
bands are especially well visible in the reflectance spectra (see
Fig. 12). For the Mo(IV) complexes (see for example the spectrum
of salt 6 in Fig. 15) this band is almost solvent-independent, sup-
porting its assignment as d–d. For the W salt (complex 4) this band
is much more intense (
while for 6
e
615 nm = 20.3 1.4 ꢂ 102 molꢁ1 dm3 cmꢁ1
,
e
643 nm = 705 10 molꢁ1 dm3 cmꢁ1 in DMSO) and is sol-
vent sensitive, indicating its stronger CT character. This is a typical
behavior for similar W(IV) complexes [1].
The most intense band at ca. 500 nm has MLCT character and is
strongly solvatochromic (Fig. 15). The molar extinction coefficients
4. Conclusions
indicate allowed transitions (for example, salt 4
e
485 nm = 55 2 ꢂ
102 molꢁ1 dm3 cmꢁ1, for 6 e615 nm = 45 7 ꢂ 102 molꢁ1 dm3 cmꢁ1
in DMSO). The solvatochromism of [M(CN)3O(LL)]nꢁ ions (M = Mo
or W) was investigated intensively earlier and support the MLCT
character of the bands [28]. In the UV part of the spectra, bands
attributed to intraligand transitions overlapped with anion ones
are visible. The band closest to 400 nm is connected with the pres-
ence of the C@N bond and its position is typical for complexes with
Schiff base ligands.
The synthesized complexes of Mo(IV) and W(IV) with Schiff
bases formed from aminoethanol and 5-bromo-, 5-chloro-,
5-methoxy-, 3,5-dichloro- and 5-bromo-3-methoxy-substitued sal-
icylaldehyde, isolated as their tetraphenylphosphonium salts, be-
long to a very narrow class of compounds where changes in the
aromatic ring substituents can influence the properties of the com-
plexes. Quantum chemical calculations for the isolated complex an-
ions show remarkable correlations between the calculated bond
distances and those found by X-ray single crystal measurements,
suggesting that the formation of even strong hydrogen bonds do
not influence the structures of the complex anions considerably.
Similarly, the calculated charge on phenolic oxygen is almost iden-
tical for all the complexes studied. On the other hand, the redox
potentials of the complexes change strongly with the ligand substi-
tuent, by almost 0.1 V (between salt 6 and 3). For substituents in the
5 position (complexes 1, 2 and 3) the E1/2 value increases in order:
methoxy < bromo < chloro. In the same order, the increase of the
d–d transition energy (Fig. 12) can also be observed. In the solid
state, the substituents influence the IR spectra as well as the struc-
tures. This mainly results in a change of short contact interactions
and also in the hydrogen bonding network formation. On the basis
of all the presented data, it seems that for the 4th and 5th row ele-
ments, the main effect on the complex anion structure is metal
and ligand type and not the network of hydrogen bonds. This is,
however, very interesting, as substituents can be used to tune some
physicochemical properties (like, for example, redox potentials)
without changing the structure of the anion significantly.
Solutions of complexes 1–6 are not very stable with time, as
presented in Figs. 13 and 14. The most stable solutions are those
in MeCN, CHCl3, 1,2-dichloroethane and acetone. In other solvents
a decrease of the bands is observed, connected with the organic li-
gand release. The process of ligand release in [M(CN)3O(LL)]nꢁ type
ions was studied earlier in detail for LL = picolinic acid or salicylal-
dehyde hydrazide and was used in a reaction mechanism study
and in the synthesis of coordination isomers of Mo(IV) complexes
[29].
3.4. Cyclic voltammetry
The cyclic voltammograms for compounds 1–6 measured in
DMSO are presented in Fig. 16. The redox potentials are included
in Table 5. As can be seen in the Fig. 16, each of the synthesized
complexes can be reversibly oxidized from [Mo(CN)3O(LL)]2ꢁ to
[Mo(CN)3O(LL)]ꢁ. The redox potentials of the Mo(IV)/Mo(V) couple
in the studied molybdenum complexes are influenced by the elec-
tron withdrawing effects of 5-substituents to a larger extent than
3-substituents. The potential-current transients for each complex
for different potential scan rates indicate that the redox processes
are diffusion controlled. Moreover, the anodic and cathodic poten-
tials do not change significantly with the potential scan rates,
which is associated with the reversibility of the studied redox sys-
Acknowledgment
M.R. thanks Foundation for Polish Science (FNP) for supporting
his research work with the scholarship from the START program.
The research was carried out with the equipment purchased
thanks to the financial support of the European Regional Develop-
ment Fund in the framework of the Polish Innovation Economy
Operational Program (contract no. POIG.02.01.00-12-023/08). This
work was supported in part by the Jagiellonian University.
Table 5
The cyclic voltammetry data for complexes 1–6 in DMSO.
Potential sweep rate 100 mV/s, T = 298 K. The redox
potentials are reported vs. Fc/Fc+.
Complex
E1/2 (mV)
DE (mV)
Appendix A. Supplementary data
1
2
3
4
5
6
ꢁ510
ꢁ444
ꢁ532
ꢁ726
ꢁ453
ꢁ439
82
92
88
78
78
73
CCDC 934895, 934896 and 934897 contain the supplementary
crystallographic data for complexes 4, 3 and 2a. These data can
be obtained free of charge via http://www.ccdc.cam.ac.uk/conts/
retrieving.html, or from the Cambridge Crystallographic Data

J. Szklarzewicz et al. / Polyhedron 68 (2014) 112–121
121
[17] J.G. Leipoldt, S.S. Basson, A. Roodt, I.M. Potgieter, Transition Met. Chem. 11
(1986) 323.
[18] J. Szklarzewicz, A. Makuła, D. Matoga, J. Fawcett, Inorg. Chim. Acta 358 (2005)
1749.
Centre, 12 Union Road, Cambridge CB2 1EZ, UK; fax: (+44) 1223-
336- 033; or e-mail: deposit@ccdc.cam.ac.uk.
[19] J. Szklarzewicz, P. Paciorek, P. Zabierowski, K. Kurpiewska, M. Mikuriya, D.
Yoshioka, Polyhedron 37 (2012) 35.
References
[20] J. Szklarzewicz, K. Stadnicka, Inorg. Chim. Acta 393 (2012) 131.
[21] P. Zabierowski, J. Szklarzewicz, K. Kurpiewska, K. Lewin´ ski, W. Nitek,
Polyhedron 49 (2013) 74.
[22] A. Samotus, M. Dudek, A. Kanas, J. Inorg. Nucl. Chem. 37 (1975) 943.
[23] R. Ahlrichs, H. Horn, A. Schaefer, O. Treutler, M. Haeser, M. Baer, S. Boecker, P.
Deglmann, F. Furche, Turbomole, version 5.9.
[1] A. Samotus, J. Szklarzewicz, D. Matoga, Bull. Pol. Acad. Sci., Chem. 50 (2002)
145.
[2] H. Arzoumanian, M. Pierrot, F. Ridouane, J. Sanchez, Transition Met. Chem. 16
(1991) 422.
[3] H. Arzoumanian, A. Bouraoui, V. Lazzeri, M. Rajzmann, H. Teruel, New J. Chem.
16 (1992) 965.
[24] M.J. Frisch, G.W. Trucks, H.B. Schlegel, G.E. Scuseria, M.A. Robb, J.R. Cheeseman,
G. Scalmani, V. Barone, B. Mennucci, G.A. Petersson, H, Nakatsuji, M. Caricato,
X. Li, H.P. Hratchian, A.F. Izmaylov, J. Bloino, G. Zheng, J.L. Sonnenberg, M.
Hada, M. Ehara, K. Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y.
Honda, O. Kitao, H. Nakai, T. Vreven, J.A. Montgomery, J.E. Peralta, F. Ogliaro, M.
Bearpark, J.J. Heyd, E. Brothers, K.N. Kudin, V.N. Staroverov, R. Kobayashi, J.
Normand, K. Raghavachari, A. Rendell, J.C. Burant, S.S. Iyengar, J. Tomasi, M.
Cossi, N. Rega, J.M. Millam, M. Klene, J.E. Knox, J.B. Cross, V. Bakken, C. Adamo,
J. Jaramillo, R. Gomperts, R.E. Stratmann, O. Yazyev, A.J. Austin, R. Cammi, C.
Pomelli, J.W. Ochterski, R.L. Martin, K. Morokuma, V.G. Zakrzewski, G.A. Voth,
P. Salvador, J.J. Dannenberg, S. Dapprich, A.D. Daniels, O. Farkas, J.B. Foresman,
J.V. Ortiz, J. Cioslowski, D.J. Fox, Gaussian 09, Revision A.1.
[25] A. Altomare, G. Cascarano, G. Giacovazzo, A. Guagliardi, M.C. Burla, G. Polidori,
M. Camalli, J. Appl. Crystallogr. 27 (1994) 435.
[4] D. Matoga, J. Szklarzewicz, A. Samotus, J. Burgess, J. Fawcett, D.R. Russell,
Polyhedron 19 (2000) 1503.
[5] A. Roodt, J.G. Leipoldt, S.S. Basson, I.M. Potgieter, Transition Met. Chem. 13
(1988) 336.
´
[6] D. Matoga, J. Szklarzewicz, A. Samotus, K. Lewinski, J. Chem. Soc., Dalton Trans.
(2002) 3587.
[7] D. Matoga, J. Szklarzewicz, K. Lewin´ ski, Polyhedron 27 (2008) 2643.
[8] A. Roodt, J.G. Leipoldt, S.S. Basson, I.M. Potgieter, Transition Met. Chem. 15
(1990) 439.
[9] S.S. Basson, J.G. Leipoldt, I.M. Potgieter, Inorg. Chim. Acta 87 (1984) 71.
[10] S.S. Basson, J.G. Leipoldt, I.M. Potgieter, Inorg. Chim. Acta 90 (1984) 57.
[11] J. Szklarzewicz, A. Samotus, N.W. Alcock, M. Moll, J. Chem. Soc., Dalton Trans.
(1990) 2959.
[12] B. Burda, J. Burgess, S.A. Parsons, A. Samotus, J. Szklarzewicz, Transition Met.
Chem. 20 (1995) 291.
[13] J. Szklarzewicz, Transition Met. Chem. 26 (2001) 189.
[14] B. Nowicka, A. Samotus, J. Szklarzewicz, F.W. Heinemann, H. Kisch, J. Chem.
Soc., Dalton Trans. (1998) 4009.
[26] G.M. Sheldrick, Acta Crystallogr., Sect. A 64 (2008) 112.
[27] A. Ryniewicz, M. Tomecka, J. Szklarzewicz, D. Matoga, W. Nitek, Polyhedron 45
(2012) 229.
[28] B. Nowicka, J. Burgess, S.A. Parsons, A. Samotus, J. Szklarzewicz, Transition Met.
Chem. 23 (1998) 317.
[15] J. Szklarzewicz, A. Samotus, J. Burgess, J. Fawcett, D.R. Russell, Transition Met.
Chem. 26 (2001) 1.
[29] J. Szklarzewicz, A. Gruszewska, Transition Met. Chem. 30 (2005) 27.
[16] A. Roodt, S.S. Basson, J.G. Leipoldt, Polyhedron 13 (1994) 599.