Synthesis, structure and magnetic properties of homotrinuclear Ni(II) complexes with asymmetric Schiff-base ligands

Article abstract of DOI:10.1016/j.ica.2014.07.020

By using 1,3-propanediamine, 2-hydroxyacetophenone and salicylaldehyde an asymmetric ONNO type Schiff base, N(hydroxyphenylidene)-N′(2- hydroxyacetophenylidene)-1,3-propanediamine (H₂metsalpn), has been prepared and isolated. This Schiff base has been reduced yielding N(2-hydroxybenzyl)-N′[1-(2-hydroxyphenyl)ethyl]-1,3-diaminopropane (H ₂metsalpn^H) These two ligands were used in preparing three trinuclear Ni(II) complexes with catena-[Ni₃] structural motif, where DMF and carboxylato (formato, acetato, benzoato) ligands occur. These complexes were characterized by EA, IR, TG, DTA and MS data. The X-ray diffraction confirms that the Ni(II) central atoms are in a distorted octahedral coordination environment: the terminal centers possess {NiN₂O ₄} octahedral coordination sphere whereas the very central atom possesses {NiO₆} chromophore. The coordinated DMF groups are liberated between 140-240 °C. The SQUID magnetometry confirms presence of weak exchange coupling of the antiferromagnetic nature, J/hc = -2 to -7 cm ^-1, with a moderate single-ion anisotropy reflected by the zero-field splitting D/hc = +4 to +7 cm^-1.

Full text of DOI:10.1016/j.ica.2014.07.020

Inorganica Chimica Acta 421 (2014) 531–537
Contents lists available at ScienceDirect
Inorganica Chimica Acta
journal homepage: www.elsevier.com/locate/ica
Synthesis, structure and magnetic properties of homotrinuclear Ni(II)
complexes with asymmetric Schiff-base ligands
b
c
d
d
b
Sevi Öz a, , Nurcan Acar , Ingrid Svoboda , Ján Titiš , Roman Boca , Orhan Atakol
⇑
ˇ
a Ahi Evran University, Faculty of Science and Arts, Department of Chemistry, 40100 Kırßsehir, Turkey
b Ankara University, Faculty of Science, Department of Chemistry, 06100 Ankara, Turkey
c Materialscience, Darmstadt University of Technology, 64287 Darmstadt, Germany
d Department of Chemistry, FPV, University of SS Cyril and Methodius, 91701 Trnava, Slovakia
a r t i c l e i n f o
a b s t r a c t
Article history:
By using 1,3-propanediamine, 2-hydroxyacetophenone and salicylaldehyde an asymmetric ONNO type
Schiff base, N(hydroxyphenylidene)-N0(2-hydroxyacetophenylidene)-1,3-propanediamine (H2metsalpn),
has been prepared and isolated. This Schiff base has been reduced yielding N(2-hydroxybenzyl)-N0[1-
(2-hydroxyphenyl)ethyl]-1,3-diaminopropane (H2metsalpnH) These two ligands were used in preparing
three trinuclear Ni(II) complexes with catena-[Ni3] structural motif, where DMF and carboxylato (for-
mato, acetato, benzoato) ligands occur. These complexes were characterized by EA, IR, TG, DTA and MS
data. The X-ray diffraction confirms that the Ni(II) central atoms are in a distorted octahedral coordina-
tion environment: the terminal centers possess {NiN2O4} octahedral coordination sphere whereas the
very central atom possesses {NiO6} chromophore. The coordinated DMF groups are liberated between
140–240 °C. The SQUID magnetometry confirms presence of weak exchange coupling of the antiferro-
magnetic nature, J/hc = ꢀ2 to ꢀ7 cmꢀ1, with a moderate single-ion anisotropy reflected by the zero-field
Received 21 April 2014
Received in revised form 24 June 2014
Accepted 2 July 2014
Available online 22 July 2014
Keywords:
Ni(II) complexes
Trinuclear complexes
Asymmetric Schiff base
Crystal structure
Magnetism
splitting D/hc = +4 to +7 cmꢀ1
.
Ó 2014 Elsevier B.V. All rights reserved.
1. Introduction
[(2-hydroxyphenyl)ethyl]-1,3-propanediamine (H2metsalpnH) has
been isolated. These ligands are sketched in Scheme 1.
The ONNO type symmetric Schiff base made of 1,3-propanedi-
amine and salicylaldehyde, N,N0(salicylidene)-1,3-propanedi-
amine, finds applications in the coordination chemistry for about
70 years [1]. From 1990 to today, a few trinuclear complexes of
the symmetric Schiff base were reported [2–11]. In the last decade
also the complexes of the ONNO type asymmetric Schiff bases were
prepared [12–19]. The asymmetric Schiff base complexes are usu-
ally obtained by a template synthesis; there are only a few studies
on the isolated asymmetric Schiff bases. In the amphyprotic sol-
vents, the Schiff bases can easily be reduced by NaBH4 to the
secondary amines [20–22].
In this study a preparation of the asymmetric Schiff base is
described by combining 1,3-propanediamine and two different
aldehydes (salicylaldehyde and 2-hydroxyacetophenone) giving
rise to N(2-hydroxyphenylidene)-N0(2-hydroxyacetophenylidene)-
1,3-propanediamine (H2metsalpn). In methanol solution this Schiff
base has been reduced by using NaBH4, and N(2-hydroxybenzyl)-N0
Using these ligands, the formate, acetate or benzoate anions in
presence of DMF (dimethylformamide), trinuclear Ni(II) complexes
have been prepared, namely [{Ni metsalpnH}2Ni(HCOO)2(DMF)2]
(1), [{Nimetsalpn}2Ni(AcO)2(DMF)2] (2), and [{Ni metsalpn}2Ni
(C6H5COO)2(DMF)2] (3). Single-crystal X-ray diffraction confirms
their trinuclear structure of the catena-[Ni3] type. In addition,
thermogravimetry and magnetometry have been used in order to
characterize thermal and magnetic properties of these species.
2. Experimental
2.1. Materials and instruments
All reagents and solvents were purchased from chemical
sources and used directly without purification. Elemental C, H
and N analysis was performed using Eurovector 1018 apparatus.
Ni analysis was conducted using atomic absorption spectrometer
(GBC Avanta PM) using the FAAS mode. The mass spectra were
obtained by a Shimadzu, 2010 plus with a DI (direct inlet) unit
with an electron impact ionizer. The interface temperature of DI
⇑
Corresponding author.
E-mail address: sevioz@hotmail.com (S. Öz).
http://dx.doi.org/10.1016/j.ica.2014.07.020
0020-1693/Ó 2014 Elsevier B.V. All rights reserved.

532
S. Öz et al. / Inorganica Chimica Acta 421 (2014) 531–537
crystalline product is
a mixture of H2metsalpn (67%) and
bis-N,N0(2-hydroxyacetophenylidene)-1,3-diaminopropane (33%)
[26]. For the purification a silica column was used: Merck silica
CH3
H
N
N
CH3
H
N
N
C
C
C
H
C
H
gel 60, 40–63 lm with 50 cm height and 2.0 cm diameter; EtA-
H
H
cO:n-C6H14 (1:1) was used as a solvent for separation. It was
observed that LACH2 moved rapidly within this solvent. In addi-
tion, the ratio of the Rf value of LACH2 and Rf value of H2metsalpn
was approximately 1:2 (both substances are yellow). The pure
compounds were isolated separately by evaporating the solvent
from the eluent. H2metsalpn was recrystallized from MeCN and
its purity was checked by HPLC. Yield: 36–40%, mp 92–93 °C.
IR (cmꢀ1): mC–H(Ar) 3023–3010, mC–H(Aliph) 2941–2854, mC@N
HO
OH
HO
OH
Scheme 1. Sketch of H2metsalpn (left) and H2metsalpnH (right) asymmetric Schiff
bases.
unit was varied between 40 and 230 °C and the ionization was
carried out with electrons of 70 eV energy.
1631–1624, mC
kmax = 242 nm,
@
e
1610, mC–O(Phenol) 1278–1159, dC–H(Ar) 752.
C(ring)
The NMR spectra of the ligands were recorded on a 400 MHz FT-
NMR (Varian Mercury) spectrometer using d6-DMSO as a solvent.
The IR spectra of the complexes and ligand were recorded using
Shimadzu Infinity model FTIR spectrometer equipped three reflec-
tions ATR attachment.
The thermogravimetric analysis was performed by a Shimadzu
DTG-60H apparatus; temperature varied between 30 and 600 °C
and scan rate was 10 °C/min under N2 atmosphere in Pt pans. Cal-
ibration of the instrument was done with metallic In and Zn. The
stoichiometry of the complexes was estimated using elemental
analysis results and with the help of the mass loss data of the coor-
dinated DMF ligands.
= 8164 dm3 molꢀ1 cmꢀ1 in DMSO. 1H NMR data
in d6-CH3SOCH3 (d, ppm): 16.55 (s), 13.47 (s), 8.58 (s), 7.62 (d),
7.45(d), 7.31 (p), 6.88 (t), 6.77 (t), 3.73 (t), 3.66 (t) 2.40 (s), 2.06
(p). 13C NMR data in d6-CH3SOCH3 (d, ppm): 167.70, 164.32,
161.13, 132.86, 132.74, 132.09, 129.21, 119.31, 119.11, 118.97,
118.95, 118.62, 117.13, 117.08, 116.91, 56.46, 46, 33, 14.80. MS
(m/z): 296 (molecular peak), 175, 162, 148 (base peak), 134, 107,
91, 77. Anal. Calc. for C18H20N2O2: C, 72.94; H, 6.80; N, 9.44. Found:
C, 73.07; H, 6.23; N, 9.81%.
2.4. Preparation of H2metsalpnH
The magnetic data was taken with
a SQUID apparatus
3.0 g of asymmetric Schiff base H2metsalpn was dissolved in
70.0 cm3 of MeOH by stirring and heating. This solution was heated
up to 50 °C and to this solution solid NaBH4 in small portions was
added until colorless under strong mixing [26–29]. After 10 min. of
stirring, 300 cm3 of ice water was added. The final mixture was left
to stand for 24 h. After filtration the white precipitate was
air-dried. The product H2metsalpnH was recrystallized from
MeOH:H2O (2:1,v/v). Yield: 54–67 %, mp: 107 °C. IR (cmꢀ1): mN–H
3307, mC–H(Ar) 3041–3014, mC–H(Aliph) 2965–2846, mC@C(ring) 1610–
1588, mC–O(Phenol) 1255–1107, dC–H(Ar) 752. 1H NMR data in d6-
CH3SOCH3 (d, ppm): 7.03 (m), 6.68 (m), 6.30(d), 3.84 (q), 3.78 (s),
2.52(t), 2.36 (m), 1.63 (s), 1.28 (d), 1.27 (d), 13C NMR data in d6-
DMSO (d, ppm): 157.49, 157.15, 157.09, 128.40, 127.91, 127.73,
127.52, 124.02, 118.32, 115.75, 115.31, 57.23, 50.63, 46.27,45.03,
28.98, 22.86. MS (m/z): 300 (molecular peak), 178, 164, 150 (base
peak), 135, 107, 91, 77. Anal. Calc. for C18H24N2O2: C, 71.97; H, 8.05;
N, 9.32. Found: C, 72.49; H, 7.58; N, 9.24%.
(MPMSXL7, Quantum Design) using the RSO mode of detection.
The temperature dependence of the magnetic susceptibility taken
at B = 0.1 T has been corrected for the underlying diamagnetism
and converted to the effective magnetic moment. The field depen-
dence of the magnetization has been measured at two tempera-
tures: T = 2.0 and T = 4.6 K.
2.2. X-ray structure analysis
Single crystals of 1 through 3 were mounted to Xcalibur (TM)
single crystal X-ray diffractometer (Oxford Diffraction) with a sap-
phire CCD detector using Mo K
a radiation (k = 0.71073 Å) operat-
ing in the /2h scan mode. The unit-cell dimensions were
x
determined and refined by using the angular settings of 25 auto-
matically centered reflections in 2.466 6 h 6 27.817 range for com-
plex 1, 2.74 6 h 6 26.37 range for complex 2, 2.615 6 h 6 27.705°
range for complex 3. The Data of complex 1 and 2 was collected
at T = 293 K and the data of complex 3 at 105 K. The empirical
absorption corrections were applied by the semi empirical method
via the CrysAlis CCD software [23]. Models were obtained from the
results of the cell refinement and the data reductions were carried
out using the solution software SHELXL97 [24]. The structure of all
complexes was solved by direct methods using the SHELXS97 soft-
ware implemented in the WinGX package [25].
2.5. Preparation of [{Ni metsalpnH}2Ni(HCOO)2(DMF)2] (1)
0.300 g (0.001 mol) of the reduced asymmetric Schiff base
H2metsalpnH was dissolved in 40.0 cm3 of DMF by heating. The
temperature was raised up to 95 to 110 °C. To this solution, added
were the solution of 0.237 g (0.0015 mol) NiCl2ꢁ6H2O salt in 10 cm3
MeOH : DMF (50/50, v/v) and the solution of 0.070 g (0.001 mol)
NaHCOO in MeOH : DMF (50/50, v/v), and the solution of 0.28 g
Et3N in 5 cm3 MeOH. The final mixture was allowed left to stand
for 4–5 days. The formed crystalline complex was separated by
vacuum filtration and air dried. Yield: 27%, IR (cmꢀ1): mN–H 3279,
3047–3020, mC–H(Aliph) 2966–2849, mC@O(DMF) 1648, mC@O(formate)
1635, mC@C(ring) 1597–1591, mC–O(Phenol) 1195, dC–H(Ar) 756. MS (m/
z): 356 (base peak). Anal. Calc. for C44H58N6Ni3O10: C, 52.36; H,
5.99; N, 8.32; Ni, 17.45. Found: C, 52.83; H, 5.66; N, 7.91; Ni,
18.11%.
2.3. Synthesis of H2metsalpn
The asymmetric Schiff base N-(2-hydroxyphenylidene)-N0-
(2-hydroxyacetophenylidene)-1,3 diaminopropane was prepared
in MeCN dehydrated by CaH2 [26]. 1,3-diaminopropane (3.7 g,
0.05 mol) was dissolved in 50 cm3 of MeCN and temperature was
decreased to ꢀ10 °C in a salt-ice bath. 2-hydroxyacetophenone
(6.80 g, 0.05 mol) was added under stirring for 5 min. This solution
was kept in ice-water mixture for 30 min and the temperature was
kept under 1 °C. Afterwards, salicylaldehyde (2-hydroxybenzalde-
hyde) (6.10 g, 0.05 mol) was added to the solution and again the
solution was kept in ice-water mixture for 30 min under stirring.
This solution was kept in the deep freeze at -20 °C for 12 h. A yel-
low crystalline substance was obtained, and filtered by vacuum.
The filtrate was rinsed with MeCN and Et2O and air-dried. This
2.6. Preparation of [{Ni metsalpn}2Ni(AcO)2(DMF)2] (2)
0.296 g (0.001 mol) of the asymmetric Schiff base, H2metsalpn
was dissolved in 50.0 cm3 of DMF by heating. This solution was
heated up to 110 °C and then the solution of 0.375 g (0.0015 mol)
of Ni(AcO)2ꢁ4H2O in 10 cm3 of MeOH : DMF (50/50, v/v) was added.

S. Öz et al. / Inorganica Chimica Acta 421 (2014) 531–537
533
The final mixture was left to stand for 4 days. The formed
crystalline complex was filtered by vacuum filtration, washed with
EtOH and dried in air. Yield: 23%, IR (cmꢀ1): 3053–3026, mC–H(Aliph)
2934–2847, mC@O(DMF) 1670, mC@O(Acetate) 1641, mC@N 1616,
mC@C(ring) 1583, mC–O(Phenol) 1257, dC–H(Ar) 750. MS (m/z): 356 (base
peak). Anal. Calc. for C46H56N6Ni3O10: C, 53.69; H, 5.48; N, 8.16;
Ni, 17.11. Found: C, 53.04; H, 5.21; N, 7.84; Ni, 17.95%.
2.7. Preparation of [{Ni metsalpn}2Ni(C6H5COO)2(DMF)2] (3)
0.296 g (0.001 mol) of the asymmetric Schiff base, H2metsalpn
was dissolved in 50.0 cm3 of DMF by heating. Temperature was
raised up to 115 °C. This solution was combined with the solution
of 0.237 g (0.0015 mol) of NiCl2ꢁ6H2O salt in 10 cm3 of MeOH:DMF
(50/50, v/v) and the solution of 0.432 g (0.003 mol) of NaC6H5COO
in 10 cm3 of MeOH : H2O (50/50, v/v); the mixture was mixed
for10 min and then left to stay for 2 days. The formed crystalline
complex was separated by vacuum filtration and iair dried. Yield:
46 %, IR (cmꢀ1): 3051–3017, mC–H(Aliph) 2939–2853, mC@O(DMF)
1647, mC@O(benzoate) 1633, mC@N 1626, mC@C(ring) 1595, mC–O(Phenol)
1195, dC–H(Ar) 740. MS (m/z): 356 (base peak). Anal. Calc. for
C56H60N6Ni3O10: C, 58.32; H, 5.24; N, 7.28; Ni, 15.27. Found C,
57.81; H, 4.97; N. 7.16; Ni, 14.51%.
3. Results and discussion
3.1. Crystal and molecular structure
The molecular structures obtained by the X-ray diffraction anal-
ysis are shown in Fig. 1 as Pluton drawing. The crystal data, data
collection and refinement parameters are given in Table 1. Selected
bond lengths and angles are listed in Table 2.
In all three complexes, the central and terminal Ni(II) ions are in
a distorted octahedral environment. The very central Ni(II) ions
possess the {NiO6} octahedral coordination sphere formed of the
phenolic oxygens of the two ligands and the oxygen atoms of the
carboxylato bridges. The terminal Ni(II) ions have the {NiN2O4}
coordination sphere formed of the phenolic oxygens of the two
ligands and nitrogen atoms, one carboxylato bridge oxygen and
the oxygen atom of the DMF molecule.
The coordination polyhedra are much distorted. In 1, three
angles along the trans coordinate are O4–Ni–O5 = 174.28(10),
O1–Ni–N2 = 172.81(10), and N1–Ni–O2 = 172.72(10) deg. In 2,
these angles amount to 175.6(4), 173.2(5), and 171.9(5) deg,
whereas in 3, they are 175.8(5), 171.1(6), and 170.0(5) deg, respec-
tively. There is no angle in these polyhedra that exceed 177 deg.
Bond lengths are dispersed more homogeneously. In both three
complexes, the chemical bonds between the very central Ni(II)
ion and the donor oxygen atoms are approximately 2.00 Å. In all
three complexes only the Ni–O5 bond length (to the DMF ligand)
exceeds 2.15 Å.
Fig. 1. The Pluton drawing of the complexes 1 through 3.
The ligand in the complex 1 is a kind of the asymmetric Schiff
base and possesses the phenol-amine structure. The ligands in
the complexes 2 and 3 have the phenol-imine structures. Owing
to these reason, the conformation of the six-membered chelate
ring that occurs in 1 is closer to the chair conformation. The chelate
rings in 2 and 3 are further from the chair conformation. In 1 the
six-membered chelate ring is formed of NiN1C8C9C10N2 atoms;
the angle between the C8N1N2C9 and C8C9C10 planes is ca 72
deg and the angle between the C8N1N2C9 and N1NiN2 planes is
37 deg. On the other hand, for the phenol-imine ligands in 2 these
angles are 65 and 23 deg, whereas in 3 they are 84 and 26 deg,
respectively. Notice, the imine nitrogen atom in 2 and 3 possesses
a double bond and it is harder for it to fit into the chair conforma-
tion. In 1 the amine N-atom is sp3 hybridized that is more
appropriate in forming the chair conformation. These data match
the literature sources on Schiff-base complexes [27,28]. In the
ligands of this study there is one methyl group that creates an
asymmetry and this causes that the chelate rings are more dis-
torted manifesting itself mostly in the torsion angles /. In 2 the
torsion angles (in deg) within the six-membered chelate ring are
/(N1NiN2C10) = 35, /(N2NiN1C8) = 19, /(C8C9C10N2) = 56, and
/(C10C9C8N1) = 74 deg. In 3 these values are found as 25, 27, 81
and 87. In the reduced ligand SALLACHH2 the nitrogen and carbon
atoms within the chelate rings are sp3 hybridized and the torsion
angles amount to 39, 35, 70 and 74 deg that are nearer to the ideal
chair conformation [28].

534
S. Öz et al. / Inorganica Chimica Acta 421 (2014) 531–537
Table 1
Data collection and crystal data for 1 through 3.
Compound
1
2
3
Empirical formula
Formula weight (g molꢀ1
T (K)
C
44H58N6Ni3O10
1007.09
C46H56N6Ni3O10
1029.10
293(2)
C56H60N6Ni3O10
1153.23
105(2)
)
293(2)
Crystal size (mm)
Crystal system
Space group
a (Å)
b (Å)
c (Å)
0.30 ꢂ 0.22 x 0.12
monoclinic
P21/c
0.28 ꢂ 0.16 x 0.04
0.30 ꢂ 0.12 x 0.08
triclinic
triclinic
ꢀ
ꢀ
P1
P1
10.3856(4)
16.0360(7)
14.2395(6)
90.00
92.083(4)
90.00
10.9416(9)
12.3770(10)
18.623(2)
75.863(8)
85.947(8)
76.710(8)
2379.8(4)
2
10.9177(6)
11.3010(6)
11.6251(7)
76.722(5)
68.174(6)
89.648(6)
1290.91(13)
1
a
(°)
b (°)
c
(°)
V (Å3)
2369.93(17)
2
Z
Calc. density (g cmꢀ3
)
1.411
1.240
1.436
1.237
1.483
1.149
l
(mmꢀ1
)
F(000)
1056
1076
602
TminꢀTmax
0.7073–0.8654
2.466–27817
0.7233–0.9522
2.74–26.37
0.7243–0.9137
2.615–27.705
h Range (°)
Index ranges
Reflections collected
Reflections unique
ꢀ12 6 h 6 6, ꢀ17 6 k 6 20, ꢀ16 6 l 6 17 ꢀ12 6 h 6 13, ꢀ15 6 k 6 15, ꢀ23 6 l 6 22 ꢀ13 6 h 6 13, ꢀ13 6 k 6 14, ꢀ8 6 l 6 14
4783
9600
9259
3390
4316
4516
R1, wR2 (2
r
)
0.0478. 0.0901
0.0788, 0.0805
3390/336
1.071
0.1335, 0.4075
0.2417, 0.3516
4316/584
1.039
0.0594, 0.1482
0.0931, 0.1261
4516/678
0.994
R1, wR2 (all)
Data/parameters
GOOF of F2
Largest difference peak hole /e Åꢀ3 0.325/ꢀ0.258
1.481/ꢀ0.644
1.229/ꢀ1.139
CCDC number
993122
993123
993124
Table 2
Selected bond lengths in the coordination spheres of the complexes.a
1
2, unit-a
2, unit-b
3
Ni1–N1
Ni1–N2
Ni1–O1
Ni1–O2
Ni1–O4
Ni1–O5
Ni2–O1
Ni2–O2
Ni2–O3
2.086(3)
2.080(3)
2.040(2)
2.028(2)
2.041(2), c
2.168(2)
2.044(2)
2.046(2)
2.114(2), c
Ni1–N1
Ni1–N2
Ni1–O1
Ni1–O2
Ni1–O4
Ni1–O5
Ni2–O1
Ni2–O2
Ni2–O3
2.051(12)
2.015(11)
1.996(9)
Ni3–N4
Ni3–N5
Ni3–O6
Ni3–O7
Ni3–O8
Ni3–O10
Ni4–O6
Ni4–O7
Ni4–O9
2.042(13)
2.045(11)
2.003(10)
2.005(9)
2.069(11), c
2.214(13)
2.086(9)
Ni1–N1
Ni1–N2
Ni1–O1
Ni1–O2
Ni1–O5
Ni1–O6
Ni2–O1
Ni2–O3
Ni2–O2
Ni2–O4
Ni2–O7
Ni2–O8
Ni3–N4
Ni3–N5
Ni3–O3
Ni3–O4
Ni3–O9
Ni3–O10
2.093(13)
2.003(17)
2.042(9)
2.040(13)
2.164(11)
2.080(11), c
2.098(11)
2.076(11)
2.037(11)
2.080(9)
2.038(11), c
2.048(10), c
1.973(17)
1.992(14)
1.924(11)
2.011(12)
2.064(12), c
2.168(13)
1.985(9)
2.086(10), c
2.252(11)
2.109(9)
2.080(8)
2.068(10), c
2.102(8)
2.059(11), c
N1–H1A
N2–H2A
0.86(3)
0.85(3)
a
c – carboxylato oxygen atom.
3.2. Thermal properties
endothermic peak on the DTA curve. The theoretical mass of
DMF ligands in 1 is only 14.5% so that a fragmentation of the com-
plex probably occurs. This is the principal difference with respect
to 2 and 3 where a two-step process is evidenced: liberation of
DMF, and then an exothermic decomposition at ca 320 °C. The nat-
ure of these differences lies in stability intermediates and it is
mostly determined by donor properties of H2metsalpn and
H2metsalpnH ligands, respectively. Similar observations can be
found in literature [31].
The distortion caused by the methyl groups probably reflects
itself to the thermal stabilities of the complexes examined by ther-
mogravimetry (the recorded TG and DTA curves are deposited as
Supplementary Information). The TG curves show the mass loss
associated with liberation of DMF groups embodied in the com-
plexes 2 and 3. In 2 the theoretical mass loss of DMF molecules
is 14.1% as compared to the experimental value of 13.7 0.6%. In
3 the expected mass loss is 12.7% that matches the experimental
data of 12.4 0.3%. In both complexes DMF ligands are liberated
at 140 °C. Thermal decomposition of similar trinuclear complexes
is reported elsewhere [30]; however, those ligands were symmet-
ric Schiff bases and the decomposition of their complexes starts
with liberation of DMF.
3.3. Magnetic properties
The complex 1 behaves magnetically as an assembly of three
s = 1 units in an antiferromagnetic interaction and with some
zero-field splitting (Fig. 2). The effective magnetic moment
In 1 the DMF ligands are untouched up to 225 °C when
46.5 1.2% mass loss has been observed, associated with an
adopts at the room temperature a value of
leff = 5.45 lB that
decrease gradually down to 50 K (5.15 B); then it decreases more
l

S. Öz et al. / Inorganica Chimica Acta 421 (2014) 531–537
535
progressively and at the lowest temperature of the data taking
(1.9 K) it adopts a value of eff = 2.34 B. The room-temperature
value yields an estimate of leff lB ¼ gav½3sðs þ 1Þꢃ
three uncoupled s = 1 units with gav = 2.1.
The magnetization per formula unit M1 ¼ Mmol=NAlB ¼ gavS,
however, is only M1 = 4.19 at B = 7 T and T = 2.0 K; this is a finger-
print of a sizable single-ion anisotropy reflected into the axial zero-
field splitting (ZFS) parameter D. Even better evidence for a sizable
ZFS is seen when the magnetization is plotted vs the reduced field
B/T: the divergence is evident.
existence of a non-symmetric chromophore of the central Ni(II)
atom with trans-pairs of Ni–O distances: axial (2.037, 2.048), equa-
torial (2.037, 2.079) and (2.076, 2.098) Å. This may be the source of
an enlarged magnetic anisotropy measured by the axial zero-field
splitting parameter D. As a consequence, the magnetization func-
tion is far from its saturation value (M1 ꢄ 6). There is some inter-
molecular contact between the aromatic rings of neighboring
trinuclear units (Cꢁ ꢁ ꢁC = 3.58 Å, see Fig. S2); this probably mani-
fests itself in an enhancement of the molecular-field correction (zj).
The model Hamiltonian [32] applied for the data analysis
assumes the anisotropic g-factor; however, in order to avoid over-
parametrization, it was ignored and only the isotropic value has
been considered. For the same reason the asymmetric exchange
parameter DNi–Ni was also neglected and the Ni(II) centers were
considered equivalent either.
l
l
1=2
=
= 5.14 for
The susceptibility and magnetization data sets have been fitted
simultaneously by applying an error functional F ¼ RðvÞ ꢂ RðMÞ
that accounts uniformly for the susceptibility and magnetization
data. The calculated values have been reconstructed by considering
the spin Hamiltonian [32]
The presence of the non-zero D-value is a characteristic feature
of the Ni(II) complexes that are not strictly octahedral. In mononu-
clear Ni(II) complexes this parameter varies between ꢀ7 cmꢀ1 (for
compressed tetragonal bipyramid) to +7 cmꢀ1 (for an elongated
bipyramid) [33]. This was the basis for the formulation of the first
magnetostructural D-correlation in metal complexes [34]. In dinu-
clear and trinuclear complexes also the exchange interaction is in
the play. Therefore, in the case of J < 0 a reliable determination of
the D-parameter is problematic when only the susceptibility data
is treated. When also the magnetization is involved into the data
analysis, the determination of the D-value is reliable. However, in
the case of |D|/hc < 2 cmꢀ1 the reconstructed magnetic data is
almost insensitive to the sign of the D-parameter. In such a case
the only way is to utilize the electron spin resonance (high-field/
high-frequency ESR, syn. EPR, EMR) [35].
3
3
X
X
ꢀ ꢀ2
Ha ¼ ꢀJðS1 ꢁ S2 þ S2 ꢁ S3Þh
ꢀ ꢀ1
ꢀ ꢀ2
DðSAz ꢀ SA=3Þh
2
^
2
^
^
~
~
~
~
^
þ
lBBgA h SAa
þ
A¼1
A¼1
ð1Þ
for a = z, x. The susceptibility data were corrected for the molecular-
field correction (zj) and eventually for the temperature-indepen-
dent magnetism
(
vTIM
)
as follows vcorr ¼ vmol=½1 ꢀ ðzj=
NAl0l2BÞvmolꢃ þ
v
TIM. The latter term accounts to the uncompen-
sated underlying diamagnetism and the temperature-independent
paramagnetism, along with the signal of the sample holder.
The optimization routine converged to the following set of mag-
netic parameters for 1: J/hc = ꢀ2.10 cmꢀ1, gav = 2.031, D/hc = +5.71
cmꢀ1, (zj)/hc = ꢀ0.079 cmꢀ1
,
vTIM ꢄ 7 ꢂ 10ꢀ9 m3 molꢀ1; the dis-
crepancy factors R(v) = 0.033 and R(M) = 0.025. The sign of the
It might be noticed that for an analogous catena-[Ni3] complex
D-parameter is consistent with an elongated tetragonal bipyramid
which matches the structural data for the central Ni(II) atom: the
axial distance is Ni–O = 2.114, whereas the equatorial ones are
2.044 and 2.040 Å, respectively.
bridged by NOꢀ2 unit (4) the following set of magnetic parameters
was reported [10]: J/hc = ꢀ12.5 cmꢀ1
, gpz = 2.114, gpx = 2.150,
gc = 2.244, |Dp/hc| = +5.7 cmꢀ1 (the peripheral centre is labeled as
p, the very central as c).
The magnetic data for the complex 2 are essentially analogous
to those of the complex 1 (Fig. 3). The fitting procedure gave the
According to Table 3, the D-parameter in the complexes
under study is positive which is implied by the distortion of the
octahedron towards to an elongated tetragonal bipyramid. In 1
the very central Ni2 atom lies in the centre of the elongated
tetragonal bipyramid, as four equatorial Ni–O distances average
to Re = 2.045 Å and two axial distances (to the formato
ligands) are longer, Ra = 2.114 Å. Then the distortion parameter is
d = Ra–Re = 0.069 Å and the magnetostructural D-correlation
following set of magnetic parameters: J/hc = ꢀ2.67 cmꢀ1, gav
=
-
2.023, D/hc = +4.06 cmꢀ1, (zj)/hc = ꢀ0.091 cmꢀ1
,
vTIM ꢄ 7 ꢂ 10ꢀ9
m3 molꢀ1; R(
v) = 0.036 and R(M) = 0.036.
The magnetic properties of the complex 3 are displayed in
Fig. 4. The set of magnetic parameters reads: J/hc = ꢀ7.14 cmꢀ1
,
gav = 2.250, D/hc = +6.89 cmꢀ1, (zj)/hc = ꢀ0.095 cmꢀ1
,
vTIM ꢄ 3.4
ꢂ 10ꢀ9 m3 molꢀ1
;
R(v) = 0.0087 and R(M) = 0.014. Notice the
6
6
5
4
3
2
1
0
6
5
4
3
2
1
0
B = 0.1 T
5
T = 4.6 K
T = 2.0 K
T = 2.0 K
4
6
μ
μ
μ
4
3
μ
2
2
T = 4.6 K
χ
1
0
0
0
10
150
T/K
20
200
30
250
40
300
0
1
2
3
4
0
50
100
0
1
2
3
4
5 6 7
(B/T)/TK-1
B/T
Fig. 2. Magnetic functions per formula unit for the complex 1. Left – temperature dependence of the effective magnetic moment, centre – field dependence of the
magnetization, right – magnetization vs reduced field. Lines – fitted.

536
S. Öz et al. / Inorganica Chimica Acta 421 (2014) 531–537
6
5
4
3
2
1
0
6
5
4
3
2
1
0
6
B = 0.1 T
5
T = 4.6 K
6
4
2
0
T = 2.0 K
4
T = 2.0 K
μ
μ
μ
3
μ
2
T = 4.6 K
χ
1
0
10
150
T/K
20
200
30
250
40
300
0
0
1
2
3
4
0
50
100
0
1
2
3
4
5 6 7
(B/T)/TK-1
B/T
Fig. 3. Magnetic functions per formula unit for the complex 2. Left – temperature dependence of the effective magnetic moment, right- field dependence of the
magnetization. Lines – fitted.
6
5
4
3
2
1
0
6
5
4
3
2
1
0
6
5
4
3
2
1
0
B = 0.1 T
6
4
2
0
μ
μ
μ
μ
T = 2.0 K T = 4.6 K
T = 2.0 K
T = 4.6 K
χ
0
10
150
T/K
20
200
30
250
40
300
0
1
2
3
4
0
50
100
0
1
2
3
4
5 6 7
(B/T)/TK-1
B/T
Fig. 4. Magnetic functions per formula unit for the complex 3. Left – temperature dependence of the effective magnetic moment, right- field dependence of the
magnetization. Lines – fitted.
Table 3
Magnetic parameters for catena-[Ni3] complexes.
Complex
Ni–O(Ph)–Ni /deg
J/hc (cmꢀ1
)
g
D/hc (cmꢀ1
)
Ref.
1
2
96.54, 96.86
96.25, 96.86
95.57, 96.79
95.14, 97.10
96.54, 99.45
94.67, 95.38
92.57, 92.86
92.50
ꢀ2.10
ꢀ2.67
2.031
2.023
5.71
4.06
this work
this work
3
ꢀ7.14
2.250
6.89
this work
a
4
ꢀ12.5
2.46
1.89
1.73
2.114, 2.150, 2.244
+5.7
[10]
[36]
[36]
[36]
[Ni2(HL1)4(H2O)]
[Ni2(HL2)4(H2O)]
[Ni2(HL3)4(H2O)]
2.091
2.12
2.10
ꢀ2.11
ꢀ3.56
ꢀ3.84
93.26, 92.81
a
Cobalt analogue [5].
predicts D/hc = +4.1 cmꢀ1. A similar analysis for the peripheral
Ni2 centers is complicated by the fact that the Ni1–N1 and
Ni1–N2 bonds are cis-positioned. In 2 there are two different
molecular units and 3 is not centrosymmetric so that the predic-
tion of the D-parameter from metal–ligand distances remains
unmodeled.
Table 3 brings also an information about the relationship
between the bonding angle Ni–O(Ph)–O and the exchange coupling
constant. For bonding angles 95–99 deg the coupling is of a weak
antiferromagnetic nature. In analogous complexes possessing a
bit lower bonding angle the exchange coupling switches to the
weakly ferromagnetic one.

S. Öz et al. / Inorganica Chimica Acta 421 (2014) 531–537
537
[6] Y. Song, P. Gomez, O. Roubeau, I. Mutikainen, U. Turpeinen, J. Reedijk, Inorg.
Chim. Acta 358 (2005) 109.
4. Conclusions
[7] J. Reglinski, M.K. Taylor, A.R. Kennedy, Inorg. Chem. Commun. 9 (2006) 736.
[8] D.H. Shi, Z.L. You, C. Xu, Q. Zhang, H.L. Zhu, Inorg. Chem. Commun. 10 (2007)
397.
[9] Q.L. Wang, C. Yang, L. Qi, D.Z. Liao, G.M. Yang, H.X. Ren, J. Mol. Struct. 892
(2008) 88.
ONNO type Schiff base has been prepared and isolated by using
2-hydroxybenzaldehyde, 2-hydroxyacetophenone and ethylenedi-
amine. This species has been reduced yielding ONNO type phe-
nol-amine ligand. Asymmetric Schiff bases and the reduced
ligand were used in preparing three catena-[Ni3] complexes which
were characterized by X-ray diffraction. The X-ray diffraction con-
firms that the Ni(II) are in a distorted octahedral coordination envi-
ronment: the terminal centers possess distorted octahedral
coordination sphere {NiN2O4} whereas the very central atom pos-
sesses {NiO6} chromophore. In Schiff-base complexes the coordi-
nated DMF ligands leave the bulk powder at about 140 °C. After
leaving the DMF molecules, trinuclear Ni(II) complexes decompose
in an exothermic way. The SQUID magnetization data is compati-
ble with exchange coupling of an antiferromagnetic nature operat-
ing along with a moderate zero-field splitting.
ˇ
[10] O. Atakol, R. Boca, I. Ercan, F. Ercan, H. Fuess, W. Haase, R. Herchel, Chem. Phys.
Lett. 423 (2006) 192.
[11] B. Cristóvao, B. Miroslaw, C. Estarelles, Polyhedron 34 (2012) 121.
¯
[12] A.S. Munde, A.N. Jagdale, S.M. Jadhav, T.K. Chondhekar, J. Serbian Chem. Soc. 75
(2010) 349.
[13] A.A. Nejo, G.A. Kolawole, A.O. Nejo, J. Coord. Chem. 63 (2010) 4398.
[14] H. Keypour, M. Ahmadi, M. Rezaeivala, A. Chehregani, R. Golbedaghi, A.G.
Blackman, Polyhedron 30 (2011) 1865.
[15] M. Asadi, Z. Asadi, J. Coord. Chem. 61 (2008) 640.
[16] B. Tang, X.P. Sun, G.L. Liu, H. Li, J. Mol. Struct. 984 (2010) 111.
[17] M. Shebl, S.M.E. Khalil, S.A. Ahmed, H.A.A. Medien, J. Mol. Struct. 980 (2010)
39.
[18] G.A. Kolawole, A.A. Osowole, J. Coord. Chem. 62 (2009) 1437.
[19] M. Ghosh, M. Fleck, B. Mahanti, A. Ghosh, G. Pilet, D. Bandyopathyay, J. Coord.
Chem. 65 (2012) 3884.
[20] A. Mustapha, C. Busche, J. Reglinski, A.R. Kennedy, Polyhedron 30 (2011) 1530.
[21] H. Aneetha, K. Pannerselvam, T.F. Liao, T.H. Lu, C.S. Chung, JCS Dalton Trans.
(1999) 2689.
Acknowledgments
[22] A. Akay, C. Arıcı, O. Atakol, H. Fuess, I. Svoboda, J. Coord. Chem. 59 (2006) 337.
[23] Oxford Diffraction, CrysAlis CCD and CrysAlis RED. Version 1. 170. 14, Oxford
Diffraction, Oxfordshire, England, 2002.
Grant Agencies (Slovakia: VEGA 1/0522/14, VEGA 1/0233/12,
APVV-0014-11) are acknowledged for the financial support. In
addition, financial support of this work by the University of Ankara
Scientific Research Fund under contract no. 12B4240003 and Ahi
Evran University Scientific Research Fund under contract no.
4001.12.014 are gratefully acknowledged.
[24] G.M. Sheldrick, SHELXS97 and SHEXL97, Program for Crystal Structure Solution
and Refinement, University of Gottingen, Germany, 1997.
[25] L.J. Farrugia, J. Appl. Crystallogr. 32 (1999) 837.
_
[26] S. Öz, Ü. Ergun, M. Yakut, I. Svoboda, A. Atakol, E. Kübra Inal, N. Yılmaz, O.
Atakol, accepted article, Russ. J. Coord. Chem. 40 (2014).
[27] A. Biswas, M.G.B. Drew, J. Ribas, C. Diaz, A. Ghosh, Inorg. Chim. Acta 379 (2011)
28.
[28] M. Aksu, S. Durmußs, M. Sarı, K.C. Emregül, I. Svoboda, H. Fuess, O. Atakol, J.
Appendix A. Supplementary material
Therm. Anal. Calorim. 90 (2007) 541.
[29] B.M. Ateßs, F. Ercan, M.L. Aksu, O. Atakol, Z. Kristallogr. 223 (2008) 530.
[30] S. Durmußs, Ü. Ergun, J.C. Jaud, K.C. Emregül, H. Fuess, O. Atakol, J. Therm. Anal.
Calorim. 86 (2006) 337.
[31] O. Çakırer, E. Eker, U. Ergun, E. Gökçınar, L.T. Yıldırım, H. Dal, O. Atakol, J.
Therm. Anal. Calorim. 101 (2010) 1167.
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.ica.2014.07.020.
ˇ
[32] R. Boca, A Handbook of Magnetochemical Formulae, Elsevier, Amsterdam,
References
2012.
ˇ
[33] R. Boca, J. Titiš, Coordination Chemistry Research Progress, Nova Science
[1] A.E. Martell, M. Calvin, Die Chemie der Metallchelat Verbindungen, Verlag
Chemie GMBH, 1958, p. 199.
[2] C. Fukuhara, K. Tsuneyoshi, N. Matsumoto, S. Kida, M. Mikuriya, M. Mori, J.
Chem. Soc., Dalton Trans. (1990) 3473.
[3] A. Gerli, K.S. Hagen, L.G. Marzilli, Inorg. Chem. 30 (1991) 4673.
[4] S. Uhlenbrock, R. Wegner, B. Krebs, J. Chem. Soc., Dalton Trans. (1996) 3731.
[5] F. Ercan, O. Atakol, C. Arici, I. Svoboda, H. Fuess, Acta Crystallogr., Sect. C 58
(2002) m193.
Publishers, New York, 2008.
[34] J. Titiš, R. Boca, Inorg. Chem. 49 (2010) 3971.
[35] R. Herchel, R. Boca, J. Krzystek, A. Ozarowski, M. Duran, J. van Slageren, J. Am.
Chem. Soc. 129 (2007) 10306.
[36] R. Modak, Y. Sikdar, S. Mandal, S. Chatterjee, A. Bienko, J. Mrozinski, S.
Goswami, Inorg. Chim. Acta 416 (2014) 122.
ˇ
ˇ