Structure and magnetism of Co(II) complexes with bidentate heterocyclic ligand Hsalbim derived from benzimidazole

Article abstract of DOI:10.1016/j.poly.2011.01.021

Two mononuclear Co(II) complexes based upon 2-(1H-benzimidazol-2-yl)phenol, abbreviation Hsalbim ligand, have been prepared and studied. The structure of Hsalbim and [Co(salbim)₂] have been confirmed by X-ray structure analysis. The second coba

Full text of DOI:10.1016/j.poly.2011.01.021

Polyhedron 30 (2011) 1163–1170
Contents lists available at ScienceDirect
Polyhedron
journal homepage: www.elsevier.com/locate/poly
Structure and magnetism of Co(II) complexes with bidentate heterocyclic ligand
Hsalbim derived from benzimidazole
a
b
b
c
a,b,
⇑
ˇ
ˇ
ˇ
M. Šebová , V. Jorík , J. Moncol , J. Kozíšek , R. Boca
a Department of Chemistry, FPV, University of SS Cyril and Methodius, Trnava, Slovakia
b Institute of Inorganic Chemistry, Technology and Materials, FCHPT, Slovak University of Technology, SK-812 37 Bratislava, Slovakia
c Institute of Physical Chemistry, FCHPT, Slovak University of Technology, 812 37 Bratislava, Slovakia
a r t i c l e i n f o
a b s t r a c t
Article history:
Two mononuclear Co(II) complexes based upon 2-(1H-benzimidazol-2-yl)phenol, abbreviation Hsalbim
ligand, have been prepared and studied. The structure of Hsalbim and [Co(salbim)2] have been confirmed
by X-ray structure analysis. The second cobalt(II) complex matches the formula [Co(salbim)2]ꢀ(Hsal-
bim)ꢀMeOH assuming a co-crystallization of one neutral ligand. The electronic spectra are consistent with
the tetrahedral pattern. Magnetic susceptibility measurements down to T = 2 K along with the magneti-
zation data until B = 7 T show that the Co(II) complexes are high-spin with a considerable zero-field split-
ting of the 4B1(D2d) term: D/hc = 67 and 55 cmꢁ1, respectively.
Received 16 September 2010
Accepted 17 January 2011
Available online 3 February 2011
Keywords:
Co(II) complexes
Magnetism
Heterocyclic ligands
Zero-field splitting
Ó 2011 Elsevier Ltd. All rights reserved.
1. Introduction
shown in Fig. 2. Therefore it could act as a versatile ligand whose
properties are tunable by changing pH of the reaction mixture.
Heterocyclic N-donor ligands are widely used in assembling
complexes which exhibit interesting magnetic properties. For in-
stance, the spin crossover systems, high-spin molecules and
molecular ferromagnets represent new materials that are promis-
ing in technical exploitation [1,2]. Herein we focused to three het-
erocyclic ligands that contain benzimidazole unit linked to the
other ring containing donor (N or O) atoms (Fig. 1). The Co(II) com-
plexes of salbim, in comparison with available data for analogous
poxbim and pybzim ligands, represent the center of interest of the
present communication.
Numerous complexes of this ligand exist as documented by entries
in the CCDC database. The ligand LH preferentially coordinates as a
bidentate-anionic unit, Lꢁ, forming tetracoordinate complexes (Sol
means a co-crystallized solvent): [CoIIL2] – KATHAP [5], [CoIIL2]ꢀSol
– RODWEN [6], [ZnIIL2] ꢁ HAVBOW [8], [ZnIIL2]ꢀSol ꢁ GEPDAH [9],
KANYEE [10] and UFONAF [11], [CuIIL2] – XIHNIM [12], [CuIIL2]ꢀSol
– MAVKAW [13], and [BeIIL2] – QAQFIY [14]. Hexacoordination oc-
curs in [FeIIIL2(O2N)]ꢀSol – TOQQOG [15] and [CrIIIL2phen]+ꢀSolꢁ
–
FURSOB [16]. The tris-Lꢁ coordination is exemplified by [CoIIIL3]ꢀSol
– OGAMUF [7]. Interesting compound is a hexacoordinate complex
[MnIII(LH)L(py)(MeOH)]ꢀMeOH – DEFREM [17], where both, the
neutral and the anionic forms of the ligand coexist. A bridging abil-
ity of the Lꢁ ligand is exemplified by [ClCuIIL]2ꢀSol – TEWJIP [18].
However, in the binuclear complex [(HO)FeIIIL2]2ꢀSol – TOQQUM
the Lꢁ unit acts as a terminal ligand [15]. A rather surprising is a
monodentate O-coordination of the ligand Lꢁ in [SnIVLPh3EtOH] –
VOPCUZ [19] and its analogs [19] VOPDAG, VOPDEK and VOPDIO
as opposed to [SnIVL2Ph2]ꢀSol – VOPDOU and VOPDOU01. A rare
example of the unicoordination exists in trans-[ReVLCl2OPPh3] ꢁ
IGEWUN [20], cis-[ReVLCl2OPPh3] – IGEXAU [20], and their analogs
[20] IGEXEY, IGEXIC and IGEXOI. On the contrary, bis-Lꢁ coordina-
tion occurs in [ReVL2O(OMe)]ꢀSol – IGEXUO [20].
The compound 2-(1H-benzimidazol-2-yl)phenol, abbreviated as
Hsalbim, was first prepared by reacting salicylamide with 1,2-
diaminobenzene [3]. It has been used in analytical chemistry as
selective agent in determining mercury. It can be prepared by con-
densation of salicylic acid with 1,2-diaminobenzene in presence of
P2O5, or using a catalyst SiCl4, and p-toluenesulfic acid [4].
We will focus to the cobalt(II) complexes formed of the ligand
LH = Hsalbim; only two tetracoordinate complexes, [CoIIL2] and
[CoIIL2]ꢀDMF, were structurally characterized so far [5,6] whereas
mer-[CoIIIL3]ꢀ3EtOHꢀH2O is a hexacoordinate complex [7].
The compound LH = Hsalbim (CCDC name HAVBIQ [8] for the
polymorph-1 and HAVBIQ01 [7] for the polymorph-2), can exist
in various tautomeric forms (protonated and deprotonated) as
The condensation product of pyridinecarboxaldehyde N-oxide
with 1,2-diaminobenzene yields a ligand 2-(2-benzimidazolyl)-
pyridine N-oxide, hereafter poxbim, which has been isolated in a
free form and structurally characterized ꢁ XELDIB [21]. The biden-
tate ligand gave a series of complexes with Mn(II), Fe(II), Co(II),
⇑
Corresponding author. Tel.: +421 259325610.
ˇ
E-mail address: roman.boca@stuba.sk (R. Boca).
0277-5387/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.poly.2011.01.021

1164
M. Šebová et al. / Polyhedron 30 (2011) 1163–1170
H
N
The compound 2, [CoII(salbim)2], has been prepared as follows. A
H
N
H
N
solution of 0.18 g (0.49 mmol) of Co(ClO4)2ꢀ6H2O in 5 cm3 of water
and 7 cm3 of hot methanol was slowly added to the solution of
0.21 g (0.49 mmol) of Hsalbim in 35 cm3 of hot methanol under
an intense stirring. The dark-pink solution was boiled for 30 min.
The mixture was cooled to r.t. under stirring and then the dark-
pink product was separated. The product was dissolved in 50 cm3
of acetone, filtered, washed with small amount of water and etha-
nol and then dried on air. Yield 0.10 g (47%). Anal. Calc. for
N
N
N
N
N
HO
O
poxbim
Hsalbim = LH
pybzim
Fig. 1. Heterocyclic compounds used as ligands.
H
C
C
26H18N4O2Co (M = 477.39): C, 65.4; H, 3.80; N, 11.7. Found: C,
64.3; H, 3.88; N, 11.3%.
OH
HN+
CH
CH
+H+
The compound 3 was prepared from 0.08 g (0.21 mmol) Co(-
ClO4)2ꢀ6H2O dissolved in 10 cm3 of water. The solution was stirred
under heating, and warm solution of 0.1 g (0.47 mmol) of Hsalbim
dissolved in 10 cm3 of MeOH was added. The rose color solution
was stirred under heating for 30 min. The product was isolated
by filtration and left to dry on air. Yield 0.08 g (53%). The isolated
material in form of pearl-pink flakes does not allow the X-ray
N
H
HC
HC
C
H
CH
C
H
+
LH2
H
C
H
C
OH
N
O
HN+
CH
CH
CH
CH
−H+
N
H
N
H
HC
HC
C
H
HC
HC
C
H
structure determination. Anal. Calc. for
C40H32N6O4Co
CH
CH
CH
(M = 719.67, 3 = 2ꢀHsalbimꢀMeOH): C, 66.8; H, 4.48; N, 11.7. Found:
LH
C
H
C
H
LH
C, 66.5; H, 4.75; N, 11.2%.
H
C
H
The infrared spectra for 2 and 3 contain the basic features given
by the free ligand 1 (Supplementary material, Fig. S1). The spectra
for 2 and 3 are remarkably similar with some additional features in
3 due to the solvent molecules. An additional, rather sharp peak at
1026 cmꢁ1 is assigned to the CAO stretching vibration most prob-
ably from the MeOH which was used in synthesis. Formulation of
the complex 3 in the form of a solvated 2, 3 = 2ꢀSol, is consistent
with available structural data. The presence of a free ligand in
the complex 3 is confirmed by the TG/DTA analysis: the mass loss
until 150 °C is 3.4% (liberation of the methanol) and then proceeds
very gradually (with only a slight endothermic effect) until 400 °C
C
O
N
CH
CH
O
N
CH
CH
−H+
N
H
HC
HC
C
H
N
HC
HC
C
H
CH
C
H
C
H
L−
−
2
(L−
)
H
Fig. 2. Various forms of the ligand LH = Hsalbim.
Ni(II), Cu(II) and Zn(II) with 1:1, 1:2, 1:3, 1:4 and 1:5 metal-to-li-
gand ratios. The hexacoordination has been confirmed by X-ray
structure for [CoII(poxbim)2(H2O)2](ClO4)2 – XELDOH [21]. Unusual
molar ratio M:L = 1:4 has been rationalized by X-ray structure of
[Zn(poxbim)2(H2O)2](ClO4)2ꢀ(poxbim)2 where two ligands coordi-
nate, but two neutral solvent molecules lie outside the coordina-
tion sphere – XELDUN [21]. The central unit possesses the trans-
{MOw2N2O2} chromophore.
where
Dm = 17.5% (liberation of uncoordinated Hsalbim mole-
cules). Then the mass loss is more abrupt until the end of the heat-
ing (500 °C) where it is accompanied with a strong exothermic
effect (decomposition of the complex [Co(salbim)2]) – see Supple-
mentary material, Fig. S2.
2.2. Physical measurements
The family of complexes with the neutral pybzim ligand, 2,20-
(benzimidazol-2-yl)pyridine, is richer. However, only three com-
plexes with Co(II) have been structurally characterized so far and
they possess a polymeric structure: JEXNOQ [22], TINRUE [23],
and WEXQUM [24]. With the deprotonated ligand only two mono-
nuclear complexes of Co(III) were reported: [Co(pybzim)3]ꢀ2(ace-
tone)ꢀ1.33H2O – OGAMOZ [7] and [Co(pybzim)3]ꢀ2DMFꢀH2O –
KOKFUL [25].
Elemental analysis was carried out on FlashEA 1112 (Thermo-
Finnigan). IR spectra were measured in KBr pellets (Magna FTIR
750, Nicolet) in the 400–4000 cmꢁ1 region. Electron spectra were
measured in Nujol mull (Specord 200, Analytical Jena) in the range
9000–50000 cmꢁ1. TG/DTA analysis has been done with Shimadzu
apparatus (DTG60).
The powder diffraction patterns have been measured with X-
ray
powder
diffractometer
(Philips)
using
Co-anode
2. Experimental
(k = 1.78892 Å).
Magnetic data were taken with the SQUID magnetometer
(MPMS-XL7, Quantum Design) using the RSO mode of detection.
The susceptibility data were scanned in the temperature range
2–300 K at the applied field of B = 0.1 T. The magnetization has
been measured at T = 2.0 and 4.6 K. Raw data were corrected for
2.1. Synthesis
The compound 1, Hsalbim = 2-(1H-benzimidazol-2-yl)phenol,
was prepared from 12.7 g (0.092 mol) of salicylic acid and 10.0 g
(0.092 mol) freshly recrystallized 1,2-diaminobenzene in 100 cm3
(1.48 mol) of phosphoric acid (80%). The reaction mixture was
heated and stirred for 24 h at temperature of 160ꢁ180 °C. After
cooling, dark-blue reaction mixture was dissolved in 1000 cm3 of
distilled water to form dark blue solution. The suspension was sat-
urated by Na2CO3 until neutral pH and extracted in 400 cm3 of hot
methanol. Insoluble solid fraction was and than separated by filtra-
tion. The filtrate, after cooling, gave a crude crystalline product that
was isolated by filtration. The final product of lilac color was ob-
tained by recrystallization from methanol. Yield 4.2 g (22%). Anal.
Calc. for C13H10N2O (M = 210.24): C, 74.3; H, 4.79; N, 13.3. Found:
C, 74.3; H, 4.78; N, 13.3%.
the underlying diamagnetism using estimate of vdia
/
(10ꢁ12 m3 molꢁ1) = ꢁ5 M [g molꢁ1].
2.3. X-ray crystallography
Data collection and cell refinement of 1 and 2 were carried out
using a -axis diffractometer Gemini R CCD (Oxford Diffraction)
with graphite monochromated Mo K radiation. The diffraction
j
a
intensities were corrected for Lorentz and polarization factors.
The structures were solved by direct methods using SIR-97 [26] or
SHELXS-97 [27] and refined by the full-matrix least-squares proce-
dure with SHELXL-97 [27]. The analytical absorption correction

M. Šebová et al. / Polyhedron 30 (2011) 1163–1170
1165
create R21(7) rings [31] and yield a supramolecular chain (Fig. 3B).
A hydrogen bond between (iz)NH and the hydroxyl oxygen atom
N1AH1Aꢀ ꢀ ꢀO1 [2.00 Å], as well as much weaker intermolecular
interaction C9AH9Aꢀ ꢀ ꢀO1 [2.81 Å] are registered. A stronger intra-
molecular hydrogen bond exists between the (O)Hꢀ ꢀ ꢀN(iz):
O1AH1ꢀ ꢀ ꢀN2 [2.550(1) Å].
Table 1
Crystallographic data for 1 and 2.
Compound
1
2
Chemical formula
M
Cell setting, space group
T (K)
a (Å)
b (Å)
c (Å)
C
13H10N2O
C26H18CoN4O2
477.37
monoclinic, P21/c
293(2)
10.1590(10)
10.0280(9)
21.450(2)
94.597(9)
2178.2(4)
4
210.23
orthorhombic, Pna21
293(2)
18.2406(5)
4.7989(2)
11.9979(3)
90
1050.23(6)
4
Similar structure of 1 (HAVBIQ, orthorhombic) has been re-
ported for Hsalbim recrystallized from chloroform [8]. However, a
monoclinic polymorph (HAVBIQ01) has been isolated by recrystal-
lization of Hsalbim from DMSO [7]. The crystal structure of the
monoclinic form of Hsalbim also shows two types of intermolecular
hydrogen bonds (Fig. 3A) that result in a supramolecular chain; the
intermolecular interaction C10AH10Aꢀ ꢀ ꢀN2 is weaker and the
hydrogen bonds create R32(8) rings [31].
The complex 2 crystallizes in the monoclinic system and the
centrosymmetric space group P21/n, but molecules of [Co(salbim)2]
do not occupy special positions (Fig. 4). On the other hand, the
tetragonal polymorph of [Co(salbim)2] ꢁ KATHAP [5], which is a
racemic twin (space group P41212), contains the Co atom at the
twofold axis. In 2 the two bidentate salbim-N,O ligands form a dis-
torted tetrahedral chromophore {CoN2O2}; the angle between the
Co1/N1/O1 and Co1/N3/O2 planes is 89° (Fig. S4). The bond lengths
Co1AO1 (1.909(3) Å), Co1AO2 (1.913(3) Å), Co1AN1 (1.976(4) Å)
and Co1AN3 (1.958(4) Å) in 2 are similar to those found in the
tetragonal form of [Co(salbim)2] (KATHAP) [5] and also in the com-
plex [Co(salbim)2]ꢀ2DMF (RODWEN) [6]. Analogous geometry of
the chromophore exists in other {CoIIN2O2} complexes with deriv-
atives of salbim as ligands: GAJTER [32], GAJTIV [32], GAJTOB [32],
and WIPMAJ [33].
b (°)
V (Å3)
Z
Radiation type
Mo K
0.087
a
Mo Ka
0.820
l
(mmꢁ1
)
Crystal size (mm)
Diffractometer
Absorption correction
Tmin, Tmax
0.626 ꢂ 0.536 ꢂ 0.099
Gemini R CCD
analytical
0.954, 0.992
1.050
0.0284, 0.0749
2127/1/146
0.098, ꢁ0.136
793545
0.191 ꢂ 0.186 ꢂ 0.082
Gemini R CCD
analytical
0.727, 0.864
1.008
0.0734, 0.157
4437/0/308
0.479, ꢁ0.349
793546
S
R1[F2 > 2
Data/restrains/parameters
qmin (e Åꢁ3
CCDC code
r
(F2)], wR2(F2)
D
qmax
,
D
)
[28] was made by using CRYSALIS-RED [29]. Geometrical analyses
were performed with SHELXL-97. The structures were drawn with
MERCURY [30] and/or XP in SHELXTL [27]. Crystal data, conditions of
data collection, and refinement information are reported in
Table 1.
Hydrogen bonds in the crystal structure of monoclinic and
tetragonal forms of [Co(salbim)2] are drawn in Fig. 5. The complex
molecules of 2 are linked by intermolecular hydrogen bonds be-
tween (iz)NH and phenolate oxygen atoms of adjacent complex
molecules in 2: N2AH2Nꢀ ꢀ ꢀO2i and N4AH4Nꢀ ꢀ ꢀO1ii [symmetry
codes: (i) ꢁx + 3/2, y + 1/2, ꢁz + 3/2; (ii) ꢁx + 1, ꢁy, ꢁz + 1] with
Nꢀ ꢀ ꢀO separations of 2.811(5) and 2.830(5) Å, respectively; a two-
dimensional hydrogen-bonding network is formed (Fig. S5). Two
3. Results and discussion
3.1. Structural data
The compound Hsalbim (1) gave suitable single crystals from
methanol. It is a planar molecule where N1AC1 [1.359(2) Å], and
N2AC1 [1.322(2) Å] bond lengths indicate that the whole molecule
is conjugate (Fig. S3, Supplementary material). The crystal struc-
ture of 1 shows two types of intermolecular hydrogen bonds which
molecules in
2
are linked to R22(12) ring [31] through
Fig. 3. Different hydrogen bond networks in two polymorphs of Hsalbim: (a) monoclinic HAVBIQ01 [7], (b) this work, similar to orthorhombic HAVBIQ [8].

1166
M. Šebová et al. / Polyhedron 30 (2011) 1163–1170
Fig. 4. The molecular structure of two polymorphs of [Co(salbim)2]: (a) one part of a racemic twin of KATHAP in space group P41212 [5], (b) complex 2 in P21/n.
Fig. 5. The molecules of [Co(salbim)2] connected through NAHꢀ ꢀ ꢀO bonds: (a) in KATHAP; (b) in the complex 2.
N4AH4Nꢀ ꢀ ꢀO1ii hydrogen bonds. There are
p–p
stacking interac-
identical with that piece of the single crystal. Therefore the powder
diffraction technique is a powerful tool in proving the identity and
purity of the powder material.
The powder diffraction patterns have been taken for com-
pounds 1 through 3. The patterns have been reconstructed from
the available X-ray structure data by MERCURY [30].
According to Fig. 6 the powder of LH = Hsalbim is identical with
the crystalline polymorph-2 (HAVBIQ01). This differs from the XPD
of the polymorph-1 (HAVBIQ) that matches the present X-ray
structure redetermination (1).
tions [34] between and the [C13AC19] benzene rings of salbim li-
gands and adjacent symmetry-related [C21AC26] benzene ring of
salbim ligands at (x + 1/2, ꢁy + 1/2, +z + 1/2); centroidꢀ ꢀ ꢀcentroid
distance is 3.73 Å and distances between two benzene planes are
in the range 3.30–3.80 Å. On the other hand, crystal structure of
KATHAP [5] exhibits a three-dimensional hydrogen-bonding net-
work (Fig. S6). It can be concluded that 2 and KATHAP represent
supramolecular isomers [35–37] with different hydrogen-bonding
networks.
The X-ray structure for 2 has been solved in the space group
P41212 as a racemic twin (KATHAP [5]). In this case all unit cell an-
gles were 90°. The theoretical powder diffraction pattern is pre-
sented in Fig. 7. The present structure redetermination refers to
the space group P21/n where the angle b = 94.597(9)° in the mono-
clinic crystal system. This manifests itself in a different powder dif-
fraction pattern where the first line is markedly split. The recorded
XPD pattern for 2 matches a theoretical prediction based upon the
full structural determination.
3.2. Powder diffraction patterns
In general, the cobalt complexes display a variety of instabilities
and interconversions. First, the Co(II) complexes are sometimes
oxidized to Co(III) and the Co(III) ones are kinetically inert. Second,
Co(II) complexes are susceptible for isomerism among which the
salt isomerism is quite common. Even if the X-ray structure had
been done, there is no guarantee that the powder material was

M. Šebová et al. / Polyhedron 30 (2011) 1163–1170
1167
by an intense charge transfer band. The d–d transitions are well
visible at 10 400 (a broad band), 18 000, and 19 200 (narrow)
cmꢁ1; there is shoulder at 17 000 cmꢁ1. Since the structure of the
complex 2 has been determined as tetrahedral, and the powder dif-
fraction pattern matches the theoretical prediction, the interpreta-
tion of the spectrum should be done in the Td group of symmetry in
the first approximation.
A - Expt. LH
B - Theor. HAVBIQ01
C - Theor. HAVBIQ
D - Theor. 1
D
C
The estimate for the crystal field strength for tetrahedral sys-
tems is Dq(Td) = (4/9)Dq(Oh) = (4/9)(1/6)F4(R). Therefore the first
transition D1 (4A2 ! 4T2) lies in the NIR region and it has not been
detected.
4
The first observed band refers to the 4A2 ! T1ðFÞ transition and
its energy is about D2 = 18Dq = 10400 cmꢁ1. However, the configu-
4
ration interaction between 4T1(F)ꢀ ꢀ ꢀ T1(P) terms is effective and
causes a reduction of the above estimate. With Dq = 578 the esti-
4
mate for the last allowed d–d transition 4A2 ! T1ðPÞ is
B
A
D3 ¼ 12Dq þ 15B; using the unreduced Racah parameter of Co(II),
B0 = 980 cmꢁ1, we get D3 = 21800 cmꢁ1. This value need be low-
ered owing to the nephelauxetic effect: the Racah electron repul-
sion parameter B is reduced relative to its free-ion value B0.
The assignment of the bands in the 17000–22000 cmꢁ1 region is
intricate since the 4T1g(P) mother term is split into {4A2, 4E} daugh-
ter terms in the more realistic D2d (or C2v) group of symmetry.
Moreover, the close-lying 2E(G) term might borrow the intensity
from the spin-allowed transitions. The spin–orbit coupling is also
in the play and it further modifies the term scheme.
6
8
10 12 14 16 18 20 22 24 26 28 30 32 34 36 38
2θ/°
Fig. 6. Powder diffraction patterns for the free ligand Hsalbim.
Fortunately, a modeling of the crystal-field terms (and multi-
plets), as described in details elsewhere [38], brings answer; using
the crystal-field poles F4 = 8000 cmꢁ1 and b = B/B0 = 0.90 the calcu-
lated energy spectrum is: 4A2 (ground state), 4T2 – 5925, 4T1(F) –
10 120, 2E(G) – 17 370, 4T1(P) – 17 720, 2A1(G) – 20 830, 2T1(G) –
20 890, and 2T2(G) – 22 420 cmꢁ1. Indeed, the terms 2E(G) and
4T1(P) are close in energy. To this end, the band registered at
18 000 cmꢁ1 is assigned to the last allowedꢁ1d–d transition
4A2 ! 4T1ðPÞ whereas the shoulder at 17 000 cm is assigned to
A - Expt.
B - Theor. 2
C - Theor. KATHAP
D - Exptl. 3
2
D
C
4
the spin-forbidden transition 4A2 ! EðGÞ.
For 3 the d–d transitions are well visible at 10 000–10 700 (a
broad band), 15 500 (narrow), 17 100 (narrow) cmꢁ1 and
19 000 cmꢁ1 (broad). A modeling with F4 = 8000 cmꢁ1 and b = B/
B0 = 0.80 gave the energy spectrum: 4A2 (ground term), 4T2
–
5925, 4T1(F) – 10 030, 2E(G) – 15 520, 4T1(P) – 15 850, 2A1(G) –
19 180, 2T1(G) – 19 510, and 2T2(G) – 20 600 cmꢁ1. Again the spin
forbidden transition to 2E(G) is close to the last spin allowed tran-
sition to 4T1(P). Accordingly, the registered d–d transitions are as-
B
A
signed
as
follows:
D2(4A2 ! 4T1ðFÞ) = 10000 cmꢁ1
and
D3(4A2 ! T1ðPÞ) = 17100 cmꢁ1. The feature at 15500 cmꢁ1 is the
4
4A2 ! 2EðGÞ transition that borrows intensity from the close lying
allowed transition D3
.
Definitely, the electronic spectrum of 3 is not consistent with
octahedral patterns. For an octahedral high-spin Co(II)
6
8
10 12 14 16 18 20 22 24 26 28 30 32
4
complex the spin-allowed d–d transitions are T1gðOhÞ !
2θ/°
4
4T2g
;
4T1g
!
4A2g; and 4T1g ! T1gðPÞ. Then the d–d transitions re-
Fig. 7. Powder diffraction patterns for the Co(II) complexes of salbim.
fer to D1 = 8Dq, D2 = 18Dq and D3 = 12Dq + 15B. Taking
D1 = 8Dq = 10 000 cmꢁ1
, the second band should lie around
The XPD data for 3 unambiguously shows that this sample is
different from 2. However, any embodiment of the solvent mole-
cules will change the lattice parameters and consequently the
whole XPD pattern. Therefore it cannot be excluded that the com-
pound 3 is 2ꢀSol. This hypothesis is supported by inspecting the
CCDC-database according to which the MII centers form with the
salbim ligand exclusively [MIIL2]ꢀSol complexes.
D2 = 18Dq = 22 500 cmꢁ1 which contradicts to the observation.
The modeling of the electronic spectra has been improved by
considering a realistic geometry of the chromophore with the
C2v symmetry: the angle
a(NACoAN) = 129°, and c(OACoAO) =
114°. The crystal field poles were F4(N) = 8500, and
F4(O) = 7000 cmꢁ1. The nephelauxetic ratio was b = 0.9 for 2 and
0.8 for 3; orbital reduction factors were
j = 0.9. Within the C2v
group of symmetry the energy levels are split into a set of Kra-
mers doublets whereas the spin–orbit coupling only shifts the en-
ergy values. According to Fig. 9 the reconstruction of the energy
transitions is fairly good. The shoulder at 17 000 cmꢁ1 for 2 does
not contain a single transition to 2E(G), that is expected to be
3.3. Electronic spectra
The electronic spectrum of 2 (Fig. 8) in the range of 9000–
23 000 cmꢁ1 exhibits a number d–d transitions which are followed

1168
M. Šebová et al. / Polyhedron 30 (2011) 1163–1170
1.0
0.8
0.6
0.4
0.2
4T1g(P)
4T1(P)
4P
4A2+4E
15
2G
0.4
0.2
Δ3 > 15B + 12Dq
Δ3 ~ 15B + 12Dq
Δ2 ~ 18Dq
2E(G)
CI
CI
4B+3C
4F
4A2g
Δ2 < 18Dq
4T1(F)
4A2+4E
14000
16000
18000
20000
8Dq
4T2
10Dq
4T2g
Δ1
4B2+4E
4T1(P)
Δ1
2E(G)
10Dq
4A2
0.20
0.18
0.16
4T1(F)
8Dq
4T1g
4B1
9000
10000
11000
9000
12000
15000
18000
21000
-1
24000
flattening
compression
Wavenumber/cm
Td
Oh
D4h
D2d
R3
a) Electronic spectrum
b) Crystal field terms (not to scale)
Fig. 8. Electronic spectrum of 2 (dashed) and 3 (solid) measured in Nujol mull.
0.4
0.2
0.0
0.4
0.2
0.0
8000 10000 12000 14000 16000 18000 20000 22000
Wavenumber/cm-1
8000 10000 12000 14000 16000 18000 20000 22000
Wavenumber/cm-1
Complex 2
Complex 3
Fig. 9. Modeling of the electronic spectra: horizontal bars – crystal-field terms (lower row) and crystal-field multiplets (upper row).
sharp, but also to a set of transitions to daughter terms of 4T1(P)
on symmetry lowering.
The magnetization data taken at T = 2.0 K show a saturation at
B = 7.0 T but the magnetization per formula unit is only
The differences in the electronic spectrum for 2 and 3 are prob-
ably due to a different degree of flattening of the parent tetrahedra.
If this hypothesis is true, then a different zero-field splitting should
be detected in magnetic measurements. Notice, the zero-field split-
ting for a perfect tetrahedron is exactly D = 0.
M1 = Mmol/(NAlB) = 2.5. This subnormal value again reflects a con-
siderable zero-field splitting.
The magnetic data have been fitted by assuming the anisotropic
spin Hamiltonian (SH) in the form
ꢁ2
ꢁ1
2
^
^
^
Ha ¼ D½Sz ꢁ SðS þ 1Þ=3ꢄꢀh þ ga
l
BBSaꢀh
ð1Þ
3.4. Magnetic data
(a = z, x) where D is the zero-field splitting parameter and the sec-
ond contribution is the spin Zeeman term. The anisotropy has been
accounted by two different g-factors (gz and gx) whereas the D-
parameter eventually can be constrained by the SH formula
The molar magnetic susceptibility for 2, corrected for the under-
lying diamagnetism, has been converted to the effective magnetic
moment that is displayed in Fig. 10. At the room temperature the
value of
leff = 4.77 lB matches the S = 3/2 spin system with some
D ¼ kðgz ꢁ gxÞ=2
ð2Þ
orbital contribution to the g-factor (g ꢃ 2.5). On cooling the effec-
tive magnetic moment slightly decreases along a straight line
which is a fingerprint of some temperature-independent paramag-
netism. Below 100 K the decrease is more rapid and reflects the
zero-field splitting (the splitting of the ground 4B1(D2d) term into
two Kramers doublets).
For Co(II) system the value of the spin–orbit splitting parameter
is k ¼ ꢁnCo=2S which amounts to k/hc = ꢁ172 cmꢁ1
.
An advanced fitting procedure accounts simultaneously for the
two data-sets:
v ¼ fðT; B0 ¼ 0:1TÞ and M ¼ fðB; T0Þ with T0 = 2.0
and 4.6 K, respectively. The powder average has been done by

M. Šebová et al. / Polyhedron 30 (2011) 1163–1170
1169
4
6
5
4
3
2
1
0
B = 0.1 T
3
2
1
T = 2.0 K
T = 4.6 K
5.0
4.5
4.0
3.5
3.0
μ
μ
0
5
10 15 20 25 30 35 40
0
0
1
2
3
4
5
6
7
0
50
100
150 200 250 300
B/T
T/K
Fig. 10. Magnetic functions for 2 = [Co(salbim)2]. Left – temperature dependence of the effective magnetic moment, right – field dependence of the magnetization. Circles –
experimental data. Lines – fitted by gz = 2.00, gx = 2.60, D/hc = 66.7 cmꢁ1
.
averaging 210 points of polar coordinates ð#i; uiÞ distributed over
the top hemisphere. It converged to the following set of magnetic
parameters: gz = 2.00, gx = 2.60, D/hc = 66.7 cmꢁ1 [the discrepancy
The experimental magnetic data for 3 are displayed in Fig. 11. It
can be seen that the magnetic functions are essentially similar to
the complex 2; the best-fit values for magnetic parameters are
factors R(v) = 0.005 and R(M) = 0.098]. The susceptibility is
gz = 2.00, gx = 2.80, D/hc = 54.9 cmꢁ1
,
v
TIP = 26.6 ꢂ 10ꢁ9 m3 molꢁ1
reproduced excellently, however the magnetization less satisfac-
tory. The involvement of the rhombic zero-field splitting parame-
ter E does not improve the quality of the fit. With the SH
formula, the estimate of the axial zero-field splitting parameter is
D/hc = 52 cmꢁ1 that matches the value obtained by optimization.
There is not so much freedom to improve the model unless one
considers a multiterm reference. This means a consideration of full
space of 120 levels as they result from all terms of the electron con-
figuration d7. An alternative is to deal with the spin admixed states
– a theory well developed for (quasi) octahedral Fe(III) systems
where the spin–orbit coupling operator mixes states of different
[R(v) = 0.009, R(M) = 0.048]. These data again show a considerable
magnetic anisotropy represented by the g-factor difference and the
zero-field splitting parameter D.
We tested a hypothesis that the complex 3 contains more (less)
than one uncoordinated Hsalbim molecule. For instance, probing
[Co(salbim)2]ꢀ2salbimH for 3 would increase the molar mass, giving
rise a higher molar magnetization, and consequently to an unreal-
istically high g-factor.
The zero-field splitting of the order of magnitude 55–66 cmꢁ1 is
high when compared to the other central atoms [40]. However, for
hexacoordinate Co(II) complexes D as high as 100 cmꢁ1 has been
reported [41]. For tetracoordinate Co(II) complexes D = 10 cmꢁ1
has been found in Hg[Co(NCS)4], based upon susceptibility mea-
surements [42]. Based upon HF/HF-ESR [43], D = ꢁ14 cmꢁ1 in
[Co(PPh3)2Cl2]. In Cs3CoCl5 the value of D = ꢁ4.3 cmꢁ1 has been de-
tected by HF/HF-ESR whereas in Cs3CoBr5 D = ꢁ5.3 cmꢁ1 [44].
The sign of the D-parameter arises from the assignment of the
lowest crystal-field multiplet. When the Kramers doublet
6
so 4
^
spin multiplicity h A1jH j A1i [38,39]. In the present case the situ-
4
so 2
^
ation is more complex since in addition to h A2jH j A2i some other
mixings contribute. The principal effect is that in addition to two
Kramers doublets arising from the 4A2(F) term another one origi-
nating in the 2A2(G) term is in the play, owing to which the overall
magnetoactivity is altered (Td reference group is used for labeling
the crystal-field terms).
6
4
B = 0.1 T
5
3
T = 2.0 K
4
5.5
μ
μ
N
5.0
4.5
4.0
3.5
3
2
1
0
2
T = 4.6 K
μ
M
1
0
5
10 15 20 25 30 35 40
0
0
50
100
150
200
250
300
0
1
2
3
4
5
6
7
T
/K
B/T
Fig. 11. Magnetic functions for 3 = [Co(salbim)2]ꢀsalbimHꢀMeOH. Left – temperature dependence of the effective magnetic moment, right – field dependence of the
magnetization. Circles – experimental data. Lines – fitted by gz = 2.00, gx = 2.80, D/hc = 54.9 cmꢁ1
,
v
TIP = 26.6 ꢂ 10ꢁ9 m3 molꢁ1
.

1170
M. Šebová et al. / Polyhedron 30 (2011) 1163–1170
[5] H.-Y. Bu, Y.-J. Liu, Q.-F. Liu, J.-F. Jia, Acta Crystallogr. E61 (2005) m1986
(KATHAP).
[6] F. He, Y. Xi, J. Li, F. Zhang, Acta Crystallogr. E64 (2008) m1311 (RODWEN).
[7] X. Quezada-Buendia, A. Esparza-Ruiz, A. Pena-Hueso, N. Barba-Behrens, R.
Contreras, A. Flores-Parra, S. Bernes, S.E. Castillo-Blum, Inorg. Chim. Acta 361
(2008) 2759 (OGAMUF, HAVBIQ01, OGAMOZ).
MS ¼ ꢅ1=2 is ground state and that MS ¼ ꢅ3=2 refers to the ex-
cited state, then
= 2D > 0 holds true. For instance, the [CoCl4]2ꢁ
ion as found in Cs3CoCl5 possesses D2d symmetry and is slightly
elongated, while the same anion in Cs2CoCl4 has nearly C2v symme-
try and is slightly compressed; the former has D < 0, and the latter
has D > 0.
D
[8] Y.-P. Tong, S.-L. Zheng, X.-M. Chen, Eur. J. Inorg. Chem. (2005) 3734 (HAVBIQ,
HAVBOW).
[9] Y.-H. Zhao, Z.-M. Su, Y. Wang, X.-R. Hao, K.-Z. Shao, Acta Crystallogr. E62 (2006)
m2361 (GEPDAH).
Theoretical modeling with the generalized crystal-field theory
gave D = 10 cmꢁ1 for 2 when the spin-Hamiltonian approximation
[10] Y.-P. Tong, Acta Crystallogr. E61 (2005) m1771 (KANYEE).
[11] Y.-P. Tong, Y.-W. Lin, J. Chem. Crystallogr. 38 (2008) 613 (UFONAF).
[12] Y.-P. Tong, S.-L. Zheng, J. Mol. Struct. 841 (2007) 34 (XIHNIM).
[13] Y. Xi, J. Li, F. Zhang, Acta Crystallogr. E61 (2005) m1953 (MAVKAW).
[14] Y.-P. Tong, S.-L. Zheng, X.-M. Chen, Inorg. Chem. 44 (2005) 4270 (QAQFIY).
[15] A.K. Boudalis, J.M. Clemente-Juan, F. Dahan, V. Psycharis, C.P. Raptopoulou, B.
Donnadieu, Y. Sanakis, J.-P. Tuchagues, Inorg. Chem. 47 (2008) 11314
(TOQQOG, TOQQUM).
[16] Y.-P. Tong, Y.-W. Lin, Inorg. Chim. Acta 362 (2009) 2167 (FURSOB).
[17] W. Shi, Y. Liu, B. Liu, Y. Song, Y. Xu, H. Wang, Y. Sha, G. Xu, S. Styring, P. Huang,
J. Coord. Chem. 59 (2006) 119 (DEFREM).
[18] Q.-X. Li, X.-L. Yang, H.-B. Xu, Chin. J. Struct. Chem. 26 (2006) 149 (TEWJIP).
[19] A. Esparza-Ruiz, A. Pena-Hueso, I. Ramos-Garcia, A. Vasquez-Badillo, A. Flores-
Parra, R. Contreras, J. Organomet. Chem. 694 (2009) 269 (VOPCUZ, VOPDAG).
[20] B. Machura, M. Wolff, J. Kusz, R. Kruszynski, Polyhedron 28 (2009) 2949
(IGEWUN, IGEXAU, IGEXEY, IGEXIC, IGEXOI, IGEXUO).
is used. The splitting of the crystal-field multiplets is
D
= 17.8 cmꢁ1
= 1.
for the orbital reduction factor
j
= 0.9, and 20 cmꢁ1 when
j
To this end, the present communication represents a step for-
ward in getting magnetostructural D-correlations for Co(II) com-
plexes. These were extensively studied for a series of Ni(II)
complexes elsewhere [45–50].
4. Conclusions
The reported X-ray structure data for the complex 2 – [Co(sal-
bim)2] is not a simple redetermination of the already published
structure (KATHAP) since a different crystal system, different space
group and different hydrogen bond network has been found. The
measured X-ray powder diffractogram confirms that the powder
material used later in magnetic measurements is the complex 2
and not KATHAP. The electronic spectra exhibit transitions
that match the tetrahedral pattern. The magnetic data show a
considerable zero-field splitting of the 4B1(D2d) ground term
(D/hc = 67 cmꢁ1) which causes a marked deviation of the magne-
tization from its saturation value as well as a drop of the effective
magnetic moment at low temperature.
ˇ
[21] M. Vrbová, P. Baran, R. Boca, H. Fuess, I. Svoboda, W. Linert, U. Schubert, P.
Wiede, Polyhedron 19 (2000) 2195 (XELDOH, XELDUN).
[22] L.-J. Zhou, Y.-Y.-U. Wang, C.-H. Zhou, C.-J. Wang, Q.-Z. Shi, S.-M. Peng, Cryst.
Growth Des. 7 (2007) 300 (JEXNOQ).
[23] C.-K. Xia, W. Wu, Q.-Y. Chen, J.-M. Xie, Acta Crystallogr. E63 (2007) m2726
(TINRUE).
[24] C.-K. Xia, C.-Z. Lu, D.-Q. Yuan, Q.-Z. Zhang, X.-Y. Wu, J.-J. Zhang, D.-M. Wu, J.
Mol. Struct. 831 (2007) 195 (WEXQUM).
[25] S.-M. Peng, H.-F. Chen, Bull. Inst. Chem. Acad. Sin. 37 (1990) 49 (KOKFUL).
[26] SIR-97: A. Altomare, M.C. Burla, M. Camalli, G.L. Cascarano, C. Giacovazzo, A.
Guagliardi, A.G.G. Moliterni, G. Polidori, R. Spagna, J. Appl. Crystallogr. 32
(1999) 115.
[27] SHELXS-97 and SHELXL-97: G.M. Sheldrick, Acta Crystallogr. A64 (2008) 112.
[28] R.C. Clark, J.S. Reid, Acta Crystallogr. A51 (1995) 887.
[29] Oxford Diffraction, CRYSALIS-CCD and CRYSALIS-RED, Oxford Diffraction Ltd.,
Abingdon, England, 2010.
[30] MERCURY: C.F. Macrae, P.R. Edgington, P. McCabe, E. Pidcock, G.P. Shields, R.
Taylor, M. Towler, J. van de Streek, J. Appl. Crystallogr. 39 (2006) 453.
[31] J. Bernstein, R.E. Davis, L. Shimoni, N.-L. Chang, Angew. Chem., Int. Ed. Engl. 34
(1995) 1555.
[32] L. Benisvy, E. Bill, A.J. Blake, D. Collison, E.S. Davies, C.D. Garner, C.I. Guindy,
E.J.L. McInnes, G. McArdle, J. McMaster, C. Wilson, J. Wolowska, Dalton Trans.
(2004) 3647 (GAJTOB, GAJTER, GAJTIV).
[33] J.D. Crane, E. Sinn, B. Tann, Polyhedron 18 (1999) 1527 (WIPMAJ).
[34] C. Janiak, J. Chem. Soc., Dalton Trans. (2000) 3885.
[35] B. Moulton, M.J. Zaworotko, Chem. Rev. 101 (2001) 629.
[36] J.W. Steed, J.L. Atwood, Supramolecular Chemistry, second ed., Wiley,
Chichester, 2009.
The composition of the complex 3 = [Co(salbim)2]ꢀsalbimHꢀ
MeOH is confirmed by elemental analysis, IR spectra indicating a
CAO bond from the methanol solvent, a gradual mass loss on
TG/DTA experiments showing first a liberation of MeOH and then
liberation of salbimH, electron spectra that match a tetrahedral
pattern, and magnetic data since [Co(salbim)2] will bring too low
molar mass and [Co(salbim)2]ꢀ2salbimH too high. The magnetic
data again show a considerable zero-field splitting of the 4B1(D2d
)
ground term (D/hc = 55 cmꢁ1).
Acknowledgments
[37] M. Schröder, N.R. Champness, Supramolecular isomerism, in: J.L. Atwood, J.W.
Steed (Eds.), Encyclopedia of Supramolecular Chemistry, vol. 2, Taylor and
Francis, London, 2004, pp. 1420–1423.
Grant agencies (VEGA 1/0213/08, 1/1005/09, APVV 0202-10,
VVCE 0004-07) are acknowledged for the financial support.
ˇ
[38] R. Boca, Struct. Bonding 117 (2006) 1.
[39] M. Weissbluth, Struct. Bonding 2 (1967) 1.
Appendix A. Supplementary data
ˇ
[40] R. Boca, Coord. Chem. Rev. 248 (2004) 757.
ˇ
ˇ
ˇ
[41] B. Papánková, R. Boca, L. Dlhán, I. Nemec, J. Titiš, I. Svoboda, H. Fuess, Inorg.
Chim. Acta 363 (2010) 147.
CCDC 793545 and 793546 contain the supplementary crystallo-
graphic data for 1 and 2. These data can be obtained free of charge
via http://www.ccdc.cam.ac.uk/conts/retrieving.html, or from the
Cambridge Crystallographic Data Centre, 12 Union Road,
Cambridge CB2 1EZ, UK; fax: (+44) 1223-336-033; or e-mail:
deposit@ccdc.cam.ac.uk. Supplementary data associated with
this article can be found, in the online version, at doi:10.1016/
j.poly.2011.01.021.
[42] D. Nelson, L.W. ter Haar, Inorg. Chem. 32 (1993) 182.
[43] J. Krzystek, S.A. Zvyagin, A. Ozarowski, A.T. Fiedler, T.C. Brunold, J. Telser, J. Am.
Chem. Soc. 126 (2004) 2148.
[44] R.P. van Stapele, H.G. Beljers, P.F. Bongers, H. Zijlstra, J. Chem. Phys. 44 (1966)
3719.
ˇ
ˇ
ˇ
[45] P. Baran, M. Boca, R. Boca, A. Krutošíková, J. Miklovic, J. Pelikán, J. Titiš,
Polyhedron 24 (2005) 1510.
ˇ
[46] A. Mašlejová, R. Ivaniková, I. Svoboda, B. Papánková, L. Dlhán, D. Mikloš, H.
ˇ
Fuess, R. Boca, Polyhedron 25 (2006) 1823.
ˇ
ˇ
[47] R. Ivaniková, R. Boca, L. Dlhán, H. Fuess, A. Mašlejová, V. Mrázová, I. Svoboda, J.
Titiš, Polyhedron 25 (2006) 3261.
References
ˇ
ˇ
ˇ
ˇ
ˇ
[48] J. Titiš, R. Boca, L. Dlhán, T. Durceková, H. Fuess, R. Ivaniková, V. Mrázová, B.
Papánková, I. Svoboda, Polyhedron 26 (2007) 1523.
ˇ
[49] R. Boca, J. Titiš, In Coordination Chemistry Research Progress, Nova Science
[1] O. Kahn, J. Kröber, C. Jay, Adv. Mater. 4 (1992) 718.
[2] D. Gatteschi, Adv. Mater. 6 (1994) 635.
Publishers, New York, 2008. pp. 247–304.
ˇ
[50] J. Titiš, R. Boca, Inorg. Chem. 49 (2010) 3971.
[3] D.A. Horton, G.T. Bourne, M.L. Smythe, Chem. Rev. 103 (2003) 893.
[4] L.J. Walter, H. Freiser, Anal. Chem. 25 (1953) 127.