Details make the difference: A family of tetranuclear CuIIMnIII complexes with cube-like and double open cube-like cores

Article abstract of DOI:10.1039/c7dt00957g

The "direct synthesis" approach, namely one-pot reaction of metal powders and ammonium salt with a methanol solution of a polydentate Schiff base (H₂L) formed in situ from salicylaldehyde and ethanolamine, has been successfully used for the pre

Full text of DOI:10.1039/c7dt00957g

Dalton
Transactions
View Article Online
View Journal
PAPER
Details make the difference: a family of
II
III
tetranuclear Cu Mn complexes with cube-like
Cite this: DOI: 10.1039/c7dt00957g
and double open cube-like cores†
a
b
a
Oleh Stetsiuk, Oksana V. Nesterova, * Vladimir N. Kokozay,
a
c
c
Kostiantyn V. Domasevitch,
Beata Vranovičová, Roman Boča,
Svitlana R. Petrusenko*
Iryna V. Omelchenko,
Oleg V. Shishkin,‡
d
d
b
Armando J. L. Pombeiro * and
a
The “direct synthesis” approach, namely one-pot reaction of metal powders and ammonium salt with a
methanol solution of a polydentate Schiff base (H L) formed in situ from salicylaldehyde and ethanol-
amine, has been successfully used for the preparation of the new heterometallic compounds [Cu Mn
O (3)
O (4). Crystallographic analysis revealed that 1–4 are based on the tet-
} where the metal centres are joined by the oxygen bridges of Schiff base
ligands forming a cube-like arrangement. The novel heterometallic compound [Cu Mn(L) (CH OH) [Mn
NCS) ]·2CH OH (5) has been obtained by the “building block” approach using the reaction of [Cu(HL)
with manganese acetate and NH
O) (µ -O) } metal core which can be viewed as a double open cube. In spite of a similar {Cu
set in the cores of 1–5, the complexes show rather different molecular structures and significantly differ
by the number and combinations of coordinated CH OH/H O solvent molecules. Variable-temperature
2–300 K) magnetic susceptibility along with variable-field magnetization measurements of 1–5 showed
2
3
(
L)
4
(CH
3
OH)
3
]I
3
4
(1), [Cu
3
Mn(L)
·0.6H
-O)
4
(CH
3
OH)
3
(H
2
O)]NCS·H
2
O (2), [Cu
3
Mn(L)
4
(CH
3
OH)(H
2
O)2.55]Br·0.45H
2
and [Cu
3
Mn(L)
(H O)3.4]BF
2
4
2
II
3
III
ranuclear core {Cu Mn (µ
3
4
3
4
3
3 2
]
(
4
3
2
]
II
III
4
NCS in methanol. The crystal structure of 5 revealed the {Cu Mn (µ-
3
2
3
2
3 4
MnO } atom
3
2
(
Received 16th March 2017,
Accepted 16th May 2017
a decrease of the effective magnetic moment value at low temperature, indicative of the antiferro-
magnetic coupling of medium size (−55 to −22 cm−1). For these systems resembling a compressed prism
DOI: 10.1039/c7dt00957g
the coupling constant in walls J
4 4
correlates with the averaged bonding angles in walls α: J vs. α develops
rsc.li/dalton
approximately according to a straight line.
tures to outstanding physico-chemical properties. It is known
that a synergetic effect caused by the presence of a few dissimi-
Introduction
Polynuclear transition metal complexes are attractive objects lar metals within one close-packed molecule can be respon-
in modern coordination chemistry since they can reveal a wide sible for a fascinating magnetic behaviour, such as single-
range of interesting features – from sophisticated crystal struc- molecule magnetism, a possibility to possess large spin
1
–9
ground states or large anisotropy barriers. Furthermore, this
effect can promote a significant influence on the catalytic
1
0–13
a
activity of the compounds in various processes.
Department of Inorganic Chemistry, Taras Shevchenko National University of Kyiv,
Volodymyrska str. 64/13, Kyiv 01601, Ukraine. E-mail: svitpetrusenko@gmail.com
The Cu/Mn pair is among the most interesting combi-
nations used for the construction of heterometallic complexes
due to the possibility to show various oxidation states and par-
ticularly flexible coordination spheres of the metals, what can
allow the formation of complicated and specific architectures.
According to the Cambridge Structural Database (CSD,
v. 5.37), 69 structures of Cu/Mn complexes where at least one
pair of metal atoms is joined by a single bridging atom (the
carbonyl, organometallic and heterotrimetallic Cu/Mn/M com-
pounds were excluded according to the search conditions)
b
Centro de Química Estrutural, Instituto Superior Técnico, Universidade de Lisboa,
Av. Rovisco Pais, 1049-001 Lisboa, Portugal.
E-mail: oksana.nesterova@tecnico.ulisboa.pt, pombeiro@tecnico.ulisboa.pt
c
SSI “Institute for Single Crystals” National Academy of Sciences of Ukraine,
6
0 Nauky Ave., Kharkiv 61072, Ukraine
d
Department of Chemistry, Faculty of Natural Sciences, University of SS Cyril and
Methodius, 91701 Trnava, Slovakia
1
4
†
Electronic supplementary information (ESI) available: Tables and figures with
calculated magnetic parameters and correlations for 1–5. CCDC 1516021,
516022, 1453203, 1453204 and 1516023. For ESI and crystallographic data in
CIF or other electronic format see DOI: 10.1039/c7dt00957g
Deceased
1
‡
have been reported. They reveal a variety of Cu Mnn combi-
m
This journal is © The Royal Society of Chemistry 2017
Dalton Trans.

View Article Online
Paper
Dalton Transactions
nations, from a quite simple, from the structural point of view, 2.0 mmol) were added to the hot yellow solution of the ligand
1
5
binuclear antiferromagnet [CuLMn(H O) ](ClO )
4 2
and a and magnetically stirred until the total dissolution of metal
2
3
1
6
single-molecule magnet [MnCuCl(5-Br-sap)
2
(MeOH)] to the powders was observed (4 h). Dark-green crystals of tetrakis
-salicylidene-2-ethanolamine)-tris-methanol-tri-copper(II)–
and the high manganese(III) triiodide suitable for X-ray analysis were
polynuclear cluster with high spin ground state (μ
a
3
1
7
[
Cu Mn O (tea) (HCO ) (H O) ]·36H O
1
7
28 40
12
2 6
2
4
2
1
8
nuclear aggregate [Mn18Cu
6
O
14(H
2
L)
6
Cl
2
(H
2
O)
6
]Cl
6
·H
2
O.
isolated after one day from the resulting dark-brown
Among them, only 8 tetranuclear Cu/Mn molecular complexes solution. Yield: 0.25 g, 61% (per Cu). Anal. calc. for
where the metal atoms are linked by one bridging atom were C H N O I Cu Mn: C, 34.06; H, 3.52; N, 4.07; Mn, 4.00; Cu,
3
9
48
4
11 3
3
found.
13.86%. Found: C, 33.70; H, 3.25; N, 3.89; Mn, 3.6; Cu, 13.6%.
In spite of the fact that a huge number of polynuclear com- IR (KBr, cm− ): 3290(br), 2915(w), 2868(w), 1633(s), 1538(s),
plexes have already been prepared, the prediction of the com- 1447(s), 1392(m), 1297(s), 1192(m), 1148(s), 1041(s), 936(m),
position and crystal structure of the final product is a very 896(m), 757(s), 650(s), 579(m), 526(w), 467(m), 435(m).
1
complicated task, especially if the spontaneous self-assembly The compound is soluble in DMSO, DMF and CH
method has been used for the synthesis. The large number of insoluble in water.
initial reagents under such conditions leads to a great variety
3
CN and
3 4 3 3 2 2
of their possible combinations, thereby the study of synthesis- Synthesis of [Cu Mn(L) (CH OH) (H O)]NCS·H O (2)
structure correlations in terms of the self-assembly approach
This complex was prepared in a way similar to that of 1, but
using barium oxide (0.15 g, 1.0 mmol) and NH NCS (0.15 g,
.0 mmol) instead of calcium oxide and NH I. Dark-green crys-
tals suitable for X-ray analysis were obtained after one day
from the resulting dark-brown solution. Yield: 0.13 g, 40%
towards polynuclear species is a considerable challenge.
Following our interest in the preparation and study of poly-
nuclear heterometallic complexes based on polydentate O,N-
4
2
4
1
9
donor ligands we have continued to apply the “direct syn-
thesis” method hereby expanding the classical spontaneous
self-assembly approach where the formation of a polynuclear
complex occurred upon one experimental stage. Among the
many different ligands that can be used in such synthetic
systems, the Schiff bases occupy a strong position since they
have the tendency to form single oxido- and hydroxido-bridges
between different metal centres which is of great importance
for magnetic materials design. In the cases of 1–4, we applied
the simplest way of the “direct synthesis” method using
copper and manganese powders as metal sources. The utiliz-
ation of a “building block” approach for comparative purposes
allowed us to obtain complex 5, revealing a rather rare type of
the molecular core, observed for the first time for heterometal-
lic transition metal complexes. Herein, we describe the syn-
thesis of five novel tetranuclear complexes containing the
(
per Cu). Anal. calc. for C H N O SCu Mn: C, 44.14; H,
40 52 5 13 3
4
.82; N, 6.43; S, 2.95; Mn, 5.05; Cu, 17.51%. Found: C, 43.90;
−1
H, 4.55; N, 6.15; S, 2.88; Mn, 5.3; Cu, 17.2%. IR (KBr, cm ):
3
1
5
372(br), 2915(w), 2868(w), 2056(s), 1631(s), 1532(s), 1443(s),
390(m), 1296(s), 1191(m), 1034(s), 971(w), 756(s), 646(s),
79(m), 525(w), 462(m), 437(m). The compound is soluble in
DMSO, DMF and CH CN and insoluble in water.
3
3 4 3 2 2
Synthesis of [Cu Mn(L) (CH OH)(H O)2.55]Br·0.45H O (3)
This complex was prepared in a way similar to that of 1, but
using barium oxide (0.15 g, 1.0 mmol) and NH Br (0.19 g,
4
2
.0 mmol) instead of calcium oxide and NH I. Dark-green crys-
4
tals suitable for X-ray analysis were formed after one day from
the resulting dark-brown solution. Yield: 0.19 g, 59% (per Cu).
II
III
Anal. calc. for C37
H N O12BrCu Mn: C, 41.76; H, 4.36; N,
46 4 3
{
3
4
Cu Mn O } core, the detailed analysis of their crystal struc-
5
.26; Mn, 5.16; Cu, 17.91%. Found: C, 41.23; H, 3.95; N, 4.90;
tures and the results of magnetic investigations.
−1
Mn, 5.1; Cu, 16.8%. IR (KBr, cm ): 3350(br), 2896(w), 1638(s),
1
7
541(s), 1444(s), 1388(m), 1300(s), 1198(m), 1035(s), 978(w),
59(s), 657(s), 580(m), 533(w), 467(m), 436(m). The compound
Experimental section
General
is soluble in DMSO, DMF and CH CN and insoluble in water.
3
Synthesis of [Cu
This complex was prepared in a way similar to that of 1, but
using barium oxide (0.15 g, 1.0 mmol) and NH BF (0.21 g,
.0 mmol) instead of calcium oxide and NH I. Dark-green crys-
3 4 2 4 2
Mn(L) (H O)3.4]BF ·0.6H O (4)
All chemicals were of reagent grade and used as received.
2
0
2
Complex [Cu(HL) ] was prepared as reported previously. All
experiments were carried out in air. Infrared spectra
4
4
1
(
4000–400 cm− ) were recorded on a BX-FT IR “PerkinElmer”
2
4
instrument in KBr pellets.
tals suitable for X-ray analysis were isolated after one day from
the resulting dark-brown solution. Yield: 0.16 g, 51% (per Cu).
Anal. calc. for C H N O BF Cu Mn: C, 40.90; H, 4.20; N,
3 4 3 3 3
Synthesis of [Cu Mn(L) (CH OH) ]I (1)
3
6
44
4
12
4
3
Salicylaldehyde (0.31 ml, 3.0 mmol) and ethanolamine 5.30; Mn, 5.05; Cu, 18.03%. Found: C, 40.25; H, 3.89; N, 4.90;
0.18 ml, 3.0 mmol) were dissolved in methanol (20 ml), Mn, 5.6; Cu, 18.1%. IR (KBr, cm− ): 3338(br), 2933(w), 2870(w),
1
(
forming a yellow solution of H L (salicylidene-2-ethanolamine) 1638(s), 1541(s), 1444(s), 1395(m), 1297(s), 1190(m), 1035(s),
2
which was magnetically stirred at 50–60 °C (10 min). Then, 981(w), 759(s), 649(s), 581(m), 533(w), 464(m), 435(m). The
manganese powder (0.05 g, 1.0 mmol), copper powder (0.06 g, compound is sparingly soluble in DMSO, DMF and CH
.0 mmol), calcium oxide (0.05 g, 1.0 mmol) and NH I (0.29 g, and insoluble in water.
3
CN
1
4
Dalton Trans.
This journal is © The Royal Society of Chemistry 2017

View Article Online
Dalton Transactions
Paper
Synthesis of [Cu
Manganese acetate tetrahydrate (0.12 g, 0.5 mmol), [Cu(HL)
0.18 g, 0.5 mmol), and NH NCS (0.08 g, 1.0 mmol) were
dissolved in methanol (20 ml) and magnetically stirred at
0–60 °C (60 min). Dark-green crystals suitable for X-ray ana-
3
Mn(L)
4
(CH
3
OH)
3
]
2
[Mn(NCS)
4
]·2CH
3
OH (5)
with Uiso = 1.5Ueq(O, N). The CH-hydrogen atoms were added
geometrically [Uiso = 1.2U (C)] and refined as riding. For 2,
eq
2
]
the hydrogen atoms of the solvate water molecule were not
located and the positions of Cu and Mn ions were not dis-
tinguishable. Therefore, they were refined as mixed atoms with
estimated partial contribution factors of 0.75 (Cu) and 0.25
(
4
5
lysis were isolated after one day from the resulting dark-brown
solution. Yield: 0.06 g, 31% (per Cu). Anal. calc. for
(
Mn). For 1, partial isomorphic substitution was suggested by
mean square displacement amplitude (MSDA) analysis and
parameters of Hirshfeld rigid-bond tests. In this case, the
refinement of isotropic thermal parameters suggested the
presence of one Cu position, one preferable Mn position (0.6)
and two mixed positions with partial contributions of 0.8 (Cu)
and 0.2 (Mn). In 2, the NCS counter anion and solvate water
molecule are disordered over two unequally populated posi-
tions (0.65 and 0.35). These atoms were left isotropic and the
84 104 12 24 4 6 3
C H N O S Cu Mn : C, 43.11; H, 4.48; N, 7.18; S, 5.48;
Mn, 7.04; Cu, 16.29%. Found: C, 42.80; H, 4.00; N, 6.95; S,
.10; Mn, 6.9; Cu, 16.8%. IR (KBr, cm− ): 3500(br), 3050(w),
900(w), 2080(m), 1620(s), 1510(s), 1480(s), 1250(s), 1050(s),
80(m), 470(m). The compound is sparingly soluble in DMSO,
1
5
2
6
DMF and CH CN and insoluble in water.
3
Crystallography
−
NCS anion was refined with a set of geometry restraints.
Details of the data collection and processing, structure solu- Graphical visualization of the structures was made using the
2
3
tion and refinement are summarized in Table 1.
program Diamond 2.1e.
X-Ray diffraction studies of 3 and 4 were performed on an
The diffraction data for 1, 2 and 5 were collected with
graphite-monochromated Mo Kα radiation (λ = 0.71073 Å) “Xcalibur 3” diffractometer (graphite-monochromated MoKα
using a Stoe Image Plate Diffraction System (for 1 and 2), radiation (λ = 0.71073), CCD detector, ω-scans). Empirical cor-
2
1
numerical absorption correction using X-RED and X-SHAPE,
and a Bruker APEXII CCD area-detector diffractometer using spherical harmonics, implemented in the SCALE3
ω scans) (for 5). The data were corrected for Lorentz-polarization ABSPACK scaling algorithm of the CrysAlisPro program
rection for absorption was provided with a multi-scan method
(
2
4
effects and for the effects of absorption (multi-scans method). package. Structures were solved by direct methods and
2
The structures were solved by direct methods and refined by refined against F within anisotropic approximation for all
2
22
25
full-matrix least-squares on F using the SHELX-97 package.
non-hydrogen atoms using the OLEX2 program package with
2
6
All atoms of the cationic framework were refined anisotropi- SHELXS and SHELXL modules. All H atoms were placed in
cally. The OH-hydrogen atoms were located and then fixed idealized positions and constrained to ride on their parent
a
Table 1 Crystal data and structure refinement for 1–5
Complex
1
2
3
4
5
Empirical formula
Formula weight
Crystal system
Space group
C
39
H
48Cu
3
I
3 4
MnN O
11
C
40
H
52Cu
3
MnN
5 13
O S
C
37
H
46
N
4
O
3
12BrCu Mn
C
36
H
44
N
4
O
4
12BF Cu
3
Mn
C
84 6 3 12 24 4
H104Cu Mn N O S
1375.07
Monoclinic
P2 /n
1088.49
Monoclinic
P2 /n
1064.25
Monoclinic
P2 /n
1057.12
Monoclinic
P2 /n
2340.09
Triclinic
Pˉ1
1
1
1
1
a/Å
b/Å
c/Å
α/°
β/°
γ/°
13.3737(11)
15.6103(8)
23.0857(18)
90
92.535(9)
90
18.5500(12)
12.9260(8)
19.8784(14)
90
98.170(10)
90
17.9628(14)
13.0651(9)
19.7336(14)
90
96.001(7)
90
16.8093(9)
13.6009(7)
19.1235(10)
90
93.479(5)
90
13.5453(4)
15.1781(4)
26.9837(9)
90.892(2)
103.293(2)
114.678(2)
4866.7(3)
2
3
V/Å
Z
4814.8(6)
4
4718.0(5)
4
4605.8(6)
4
4364.0(4)
4
Calculated
density/g cm
Experiment
1.897
1.532
1.535
1.613
1.597
−3
213
213
105
298
173
temperature/K
−
1
μ(Mo-K
F(000)
α
)/mm
3.544
2676
1.705
2236
2.560
2156
1.804
2148
1.821
2394
Reflections
collected/unique
38 085/10 399
35 799/9684
23 351/10 477
51 481/14 779
55 094/18 104
R
int
0.0368
7203
0.0456
6144
0.1225
4162
0.0746
5567
0.1120
9405
Reflections with
2
2
F > 2σ(F )
Θ
min, Θmax/°
, F > 2σ(F )
2.20, 27.00
0.0309
4.14, 26.73
0.0466
2.866, 28.997
0.1023
2.930, 32.711
0.0778
1.72, 25.50
0.0573
2
2
R
1
wR (all data)
GoF
2
0.0718
0.859
0.1394
0.928
0.3183
0.978
0.2870
0.993
0.1118
0.904
a
For 3 and 4 the formula, formula weight, density, F(000) and μ parameters are calculated without masked solvent molecules.
This journal is © The Royal Society of Chemistry 2017
Dalton Trans.

View Article Online
Paper
Dalton Transactions
3
atoms with Uiso = nUeq (n = 1.5 for OH and CH groups and n = Magnetic measurements
1
.2 for all other H atoms). H atoms of water molecules were
A SQUID magnetometer (MPMS-XL7, Quantum Design) has
been used for obtaining magnetic data using the RSO mode of
detection. The susceptibility taken at B = 0.1 T between T =
positioned to fit the hydrogen bond system. In 3, the Br1
anion, O4S water molecule, and C1S atom of the O1S–C1S
methanol molecule were disordered each one over two sites.
Their relative occupancy factors were refined in isotropic
approximation and fixed at the final steps of refinement at the
values of 0.68/0.32, 0.60/0.40, and 0.50/0.50, respectively. O1S–
C1SA and O1S–C1SB bonds were restrained to be approxi-
mately equal to within 0.01 Å. Anisotropic thermal parameters
of disordered C1S(A,B) and O4S(A,B) atoms were restrained to
2
–300 K has been corrected for the underlying diamagnetism
and converted to the effective magnetic moment. The magneti-
zation has been measured at two temperatures: T = 2.0 and
T = 4.6 K.
Results and discussion
be the same within 0.04 Å and approximately equal to within
2
0
.02 Å . In 4, three F atoms of the BF
4
anion were disordered Synthesis and spectroscopic analysis
around the B(1)–F(1) rotation axis over two sites with relative
occupancies of 0.55/0.45. All B–F bonds were restrained to be
approximately equal to within 0.01 Å. Anisotropic thermal
parameters of all atoms of the anion were restrained to be
approximately equal to within 0.02 Å . O4W molecule was dis-
ordered over two sites with relative occupancies of 0.60/0.40.
Complexes 1–4 were obtained from the reaction of manganese
powder, copper powder, ammonium salt (NH I for 1, NH NCS
4
4
for 2, NH Br for 3 and NH BF for 4) and calcium (1) or
4
4
4
barium oxide (2–4) with a methanol solution of the chelating
2
polydentate ligand source (H L), formed in situ, using a molar
2
ratio of Mn : Cu : NH X : H L = 1 : 1 : 2 : 3. The ratio in the
4
2
All oxygen atoms of water molecules in 4 were restrained to be
initial reaction mixture is fundamental, since only this stoi-
chiometry afforded the heterometallic complexes 1–4.
2
approximately isotropic to within 0.02 Å . All occupancy
factors were obtained by full-matrix refinement and fixed in
the final refinement cycles.
The Schiff base H L (Scheme 1) was obtained by conden-
2
sation of salicylaldehyde and ethanolamine. The reaction was
initiated and brought to completion by heating and stirring in
open air. Dark-brown solutions were obtained at the end of all
reactions. Dark-green crystals of 1–4 that showed analytical
data consistent with the Mn : Cu = 1 : 3 stoichiometry were
formed after standing at room temperature within one day.
The general reactions can be written as follows:
For 4, the low quality and small size of the crystals cause a
rather weak signal with small I/σ values of the reflections, up
to the total absence of strong reflections in the area of higher
angles (2θ > 35–40°). This makes the assignment of the metal
types a troublesome question, since the atomic scattering
factors of the Cu and Mn atoms are rather close. Thus the rela-
tive Cu/Mn content was determined from the elemental ana-
lysis, and the position of the Mn(1) cation was located on the
basis of the refinement indicators and coordination features of
the metal centres. It should be noted that the Mn(1) position
cannot be determined undoubtedly in this case and it can be
disordered among all positions of the metal centres. Assuming
the Mn/Cu disorder over all four metal positions, one can
4
Mn þ 12Cu þ 16H2L þ 12NH4I þ 12CH3OH þ 11O2
ð1Þ
! 4½Cu3MnðLÞ ðCH3OHÞ ꢀI3 þ 12NH3 þ 22H2O
4
3
4
Mn þ 12Cu þ 16H2L þ 4NH4NCS þ 12CH3OH þ 9O2
4½Cu3MnðLÞ ðCH3OHÞ ðH2OÞꢀNCSꢁH2O þ 4NH3 þ 10H2O
!
4
3
ð2Þ
1
obtain the R value of 0.0673 that is lower than the actual
4
Mn þ 12Cu þ 16H2L þ 4NH4Br þ 12CH3OH þ 9O2
value of 0.0778. Disordering over any two sites gives the R1
value in the range of 0.070–0.073. However, the metal coordi-
nation features and the absence of strong high-angle reflec-
tions do not allow us to ensure the Cu/Mn disorder, and thus
we have decided to maintain the non-mixed metal positions in 4.
In structures of 3 and 4, several isolated electron density
peaks were located during the refinement, which were believed
to be of a highly disordered solvent molecule(s). Satisfactory
!
4½Cu3MnðLÞ ðCH3OHÞðH2OÞ2:55ꢀBrꢁ0:45H2O þ 4NH3 þ 6H2O
4
ð3Þ
4
Mn þ 12Cu þ 16H2L þ 4NH4BF4 þ 9O2
ð4Þ
!
4½Cu3MnðLÞ ðH2OÞ3:4ꢀBF4ꢁ0:6H2O þ 4NH3 þ 2H2O
4
1
results (R = 0.108 in 3 and 0.081 in 4) were obtained modelling
the disordered C, N and O atoms, but very large displacement
parameters for them were observed. The SQUEEZE procedure
2
7
implemented in PLATON indicated solvent cavities (2 cavities
3
of 168 Å each containing approximately 49 electrons, in 3, and
3
4
cavities of 30 Å each containing approximately 4 electrons, in
4). In the final refinement, this contribution was removed from
the intensity data that produced better refinement results.
Crystallographic data for the structures reported can be
obtained by quoting the deposition numbers CCDC 1516021 (1),
1516022 (2), 1453203 (3), 1453204 (4) and 1516023 (5).
2
Scheme 1 The tridentate Schiff base H L.
Dalton Trans.
This journal is © The Royal Society of Chemistry 2017

View Article Online
Dalton Transactions
Paper
We suppose that the role of calcium (or barium) oxide is to
promote complete deprotonation of the amino alcohol Schiff
base ligand. Although the “direct synthesis” method foresees
1
9
deprotonation upon reaction of zerovalent metal, in our case
an additional deprotonating reagent (base) was required. None
of complexes 1–4 were reproduced in the absence of the
s-metal oxide. Compound 1 could also be obtained using
barium oxide instead of calcium oxide without any difference
in the crystallization process and composition. It should be
3 3 4
Fig. 1 The ball-and-stick representation of the {Cu Mn(µ -O) } mole-
noted that only the systems with methanol as the solvent cular core in 1–4. Color scheme: Cu, cyan; Mn, magenta; O, red.
resulted in the precipitation of the heterometallic compounds.
Such complexes, if formed, were not isolated from DMF,
DMSO and acetonitrile probably because of their high solubi- octahedral environments with O N donor sets formed by
5
lity in these solvents.
oxygen and nitrogen atoms from the Schiff base ligands and
Complex 5 was formed through the reaction of [Cu(HL)
2
]
oxygen atoms from the coordinated molecules of methanol
with manganese acetate and NH
4
NCS in methanol solution, (Fig. 2). The coordination geometry around the Cu(3) atom is
using the molar synthetic ratio of [Cu(HL) ] : Mn(OAc) : NH NCS = square pyramidal with the O N donor set. The equatorial
2
2
4
4
1
: 1 : 2. Dark-green microcrystals of 5 were isolated from the Cu–O(N) bond lengths range from 1.901(2) to 1.992(2) Å, while
dark-brown solution after one day standing at room tempera- the apical Cu–O distances vary from 2.379(3) to 2.580(2) Å
ture. The reaction proceeds in the following way:
(Table 2). The O–Cu–O(N)cis bond angles vary from 72.63(8)° to
1
1
09.52(9)°, while O–Cu–O(N)trans angles span from 155.88(8) to
77.40(11)°. A weak contact [3.8604(6) Å] between the Cu(3)
6
½CuðHLÞ ꢀ þ 3MnðOAcÞ þ 4NH4NCS þ 8CH3OH þ 1=2O2
2
2
!
½Cu3MnðLÞ ðCH3OHÞ ꢀ ½MnðNCSÞ ꢀꢁ2CH3OH þ 4H2L
4
3 2
4
atom of the metal core and the I(1) atom of the uncoordinated
I3 anion exists and, in spite of the fact that it is greater than
þ 4NH4OAc þ 2HOAc þ H2O
−
ð5Þ the sum of the van der Waals radii of the respective atoms,
2
9,30
3.38 Å,
it should not be ignored. The value of the angle
The IR spectra of 1–5 in the 4000–400 cm− range confirmed
the presence of the Schiff base ligands. The very strong bands at
633 (1), 1631 (2), 1638 (3, 4) and 1620 cm (5) were assigned
to the ν(CvN) stretching vibrations of the Schiff bases. The
broad bands in the region of 3600–3100 cm in the spectra of
all compounds included the ν(O–H) frequencies of the CH OH
or/and H O molecules. The presence of the uncoordinated thio-
cyanate ligand in 2 can be identified by the strong ν(C–N)
1
O(3)–Cu(3)⋯I(1) = 162.99(5)° also supports this assumption.
The Mn(III) atom adopts a distorted octahedral geometry with
the {MnO N} chromophore formed by the O and N atoms of
5
the Schiff bases and methanol with the equatorial Mn–O(N)
distances varying from 1.8675(2) to 1.958(3) Å and the apical
Mn–O bond distances equal to 2.320(3) and 2.411(2) Å
−1
1
−1
3
2
(
Table 2). The cis and trans O–Mn–O(N) bond angles range
from 75.26(8) to 101.81(9)° and from 159.06(9) to 176.71(11)°,
respectively. The intermetallic M⋯M non-bonded distances
within the metal core lie in the range of 3.115(1)–3.520(9) Å.
Strong intramolecular O–H⋯O hydrogen bonding in 1
involving the oxygen atoms from Schiff bases and the co-
−1
absorption peak at 2056 cm and the weak ν(C–S) vibration at
50 cm− . In the spectrum of 5 the strong band at 2080 cm
1
−1
7
−1
and the weak peak at 790 cm are attributed to the ν(C–N) and
ν(C–S) vibrations, respectively, that imply coordination of the
NCS-group through the nitrogen atom.
28
ordinated CH OH molecules strengthens the overall tetra-
3
Crystal structures
The single crystal X-ray analysis shows that the overall struc-
II
III
tural configurations of the heterometallic Cu Mn complexes
–4 are similar. All of them are based on the tetranuclear core
Cu Mn(µ -O) } (Fig. 1), where the metal centres are joined by
the µ -O bridges of Schiff base ligands forming a cube-like
1
{
3
3
4
3
arrangement. The Schiff base molecules, H L, in 1–4 are
2
doubly deprotonated and show tridentate (N,O,O) coordi-
nation. The tetranuclear core {Cu
3 3 4
Mn(µ -O) } belongs to the
{
M (µ -X) } (M = metal atom, X = bridging atom) molecular
4
3
4
structure type (MST), obtained by excluding all non-bridging
non-metal atoms of the structure. This MST, according to the
CSD, dominates over all other tetranuclear MSTs, taking
almost 30% of all known tetranuclear complexes.
Complex 1 contains three crystallographically independent
Fig. 2 The crystal structure of 1 with atom numbering. The hydrogen
copper(II) atoms, two of which, Cu(1) and Cu(2), have distorted atoms are omitted for clarity.
This journal is © The Royal Society of Chemistry 2017
Dalton Trans.

View Article Online
Paper
Dalton Transactions
Table 2 Selected geometrical parameters (distances/Å and angles/°)
for 1
Cu1–O1
Cu1–O2
Cu1–O3
Cu1–O7
Cu1–O9
Cu1–N1
Cu2–O3
Cu2–O4
Cu2–O5
Cu2–O7
Cu2–O10
Cu2–N2
1.992(2)
1.909(2)
1.958(2)
2.580(2)
2.501(3)
1.942(3)
1.972(2)
1.904(2)
2.576(2)
1.968(2)
2.379(3)
1.938(3)
Cu3–O1
Cu3–O3
Cu3–O5
Cu3–O6
Cu3–N3
Mn1–O1
Mn1–O5
Mn1–O7
Mn1–O8
Mn1–O11
Mn1–N4
1.963(2)
2.474(2)
1.968(2)
1.901(2)
1.929(3)
2.411(2)
1.942(2)
1.944(2)
1.865(2)
2.320(3)
1.958(3)
Fig. 3 The crystal structure of 2 with atom numbering. The hydrogen
atoms are omitted for clarity.
O1–Cu1–O2
O1–Cu1–O3
O1–Cu1–O7
O1–Cu1–O9
O1–Cu1–N1
O2–Cu1–O3
O2–Cu1–O7
O2–Cu1–O9
O2–Cu1–N1
O3–Cu1–O7
O3–Cu1–O9
O3–Cu1–N1
O7–Cu1–O9
O7–Cu1–N1
O9–Cu1–N1
O3–Cu2–O4
O3–Cu2–O5
O3–Cu2–O7
O3–Cu2–O10
O3–Cu2–N2
O4–Cu2–O5
O4–Cu2–O7
O4–Cu2–O10
O4–Cu2–N2
O5–Cu2–O7
O5–Cu2–O10
O5–Cu2–N2
O7–Cu2–O10
O7–Cu2–N2
O10–Cu2–N2
169.24(10)
87.58(9)
74.83(8)
91.88(9)
83.86(10)
94.68(9)
95.76(9)
98.69(10)
94.14(11)
72.81(8)
88.41(8)
171.17(10)
157.15(8)
106.83(9)
89.72(10)
170.35(10)
76.22(8)
88.11(9)
91.16(8)
83.52(9)
95.08(9)
93.32(9)
98.45(10)
94.21(10)
72.63(8)
155.88(8)
100.93(9)
86.69(9)
170.52(10)
97.88(10)
O1–Cu3–O3
O1–Cu3–O5
O1–Cu3–O6
O1–Cu3–N3
O3–Cu3–O5
O3–Cu3–O6
O3–Cu3–N3
O5–Cu3–O6
O5–Cu3–N3
O6–Cu3–N3
O1–Mn1–O5
O1–Mn1–O7
O1–Mn1–O8
O1–Mn1–O11
O1–Mn1–N4
O5–Mn1–O7
O5–Mn1–O8
O5–Mn1–O11
O5–Mn1–N4
O7–Mn1–O8
O7–Mn1–O11
O7–Mn1–N4
O8–Mn1–O11
O8–Mn1–N4
O11–Mn1–N4
75.07(8)
86.11(9)
95.60(10)
168.50(11)
78.82(8)
99.70(9)
109.52(9)
177.40(11)
84.53(10)
93.99(11)
75.26(8)
79.84(8)
100.06(10)
159.06(9)
101.81(9)
89.50(9)
93.66(11)
89.31(11)
172.86(10)
176.71(11)
86.06(10)
83.55(10)
94.89(12)
93.27(11)
91.81(11)
Fig. 4 The crystal structure of 3 with atom numbering. The hydrogen
atoms are omitted for clarity.
nuclear structure of the complex [the O⋯O distances vary from
.672(8) to 2.691(12) Å, while the O–H⋯O bond angles range
2
from 166.71(22) to 175.17(19)°].
In the crystal structures of compounds 2, 3 and 4 all metal
5
atoms reveal an octahedral O N coordination environment
with strong axial distortion (Fig. 3, 4 and 5). The equatorial
O N donor set of each metal centre is formed by oxygen and
3
nitrogen atoms from the Schiff base ligand with the M–O(N) Fig. 5 The structure of 4 with atom numbering. The hydrogen atoms
bond lengths ranging from 1.862(5) to 1.986(3) Å. One of the are omitted for clarity.
axial oxygen atoms belongs to the Schiff base too while the
second one belongs to the coordinated methanol or water
molecule, with M–O bonds ranging from 2.355(9) to 2.741(11) second part as the non-coordinated one. This results in a
Å (Tables 3–5). In the case of 3 and 4 a high degree of thermal mixed O N octahedral/O N square pyramidal coordination
5
4
and/or structural disorder of coordinated water molecules was environment of Cu2 in 3 and Cu3 in 4, so that 55% of complex
observed. Particularly, atoms O4S in 3 and O4W in 4 are both cations in 3 and 40% in 4 reveal octahedral coordination of all
disordered in such a way that the Cu–O bond length floats metal centres, similar to 2, while 45% of cations in 3 and 60%
from 2.536(18) to 3.56(3) Å and from 2.45(2) to 3.055(12) Å, in 4 reveal coordination type similar to 1, where one of the
respectively. Therefore, the first part of each of these water metal centres has a square pyramidal coordination environ-
molecules can be treated as coordinated to Cu centres and the ment. The O–M–O(N)cis bond angles in 2–4 range from 72.4(2)°
Dalton Trans.
This journal is © The Royal Society of Chemistry 2017

View Article Online
Dalton Transactions
Paper
Table 3 Selected geometrical parameters (distances/Å and angles/°)
for 2
Table 4 Selected geometrical parameters (distances/Å and angles/°)
for 3
M1–O1
M1–O2
M1–O3
M1–O8
M1–O9
M1–N1
M2–O3
M2–O4
M2–O5
M2–O8
M2–O10
M2–N2
1.986(3)
1.916(3)
1.957(2)
2.564(2)
2.466(4)
1.934(3)
1.957(3)
1.898(3)
2.544(3)
1.964(3)
2.373(3)
1.943(3)
M3–O1
M3–O3
M3–O5
M3–O6
M3–O12
M3–N3
M4–O1
M4–O5
M4–O7
M4–O8
M4–O11
M4–N4
1.956(3)
2.489(3)
1.964(3)
1.904(3)
2.476(4)
1.929(4)
2.480(3)
1.949(3)
1.895(3)
1.949(3)
2.383(5)
1.943(4)
Cu1–O1
Cu1–O2
Cu1–O6
Cu1–O8
Cu1–O2S
Cu1–N1
Cu2–O2
Cu2–O3
Cu2–O4
Cu2–O8
Cu2–O4SA
Cu2–N2
1.894(7)
1.985(7)
2.508(7)
1.983(7)
2.472(7)
1.901(9)
1.953(7)
1.892(6)
1.951(7)
2.438(7)
2.536(18)
1.932(8)
Cu3–O4
Cu3–O6
Cu3–O7
Cu3–O8
Cu3–O3S
Cu3–N4
Mn1–O2
Mn1–O4
Mn1–O5
Mn1–O6
Mn1–O1S
Mn1–N3
2.553(7)
1.961(6)
1.883(7)
1.970(6)
2.382(10)
1.927(8)
2.451(7)
1.948(6)
1.883(7)
1.940(6)
2.355(9)
1.960(8)
O1–M1–O2
O1–M1–O3
O1–M1–O8
O1–M1–O9
O1–M1–N1
O2–M1–O3
O2–M1–O8
O2–M1–O9
O2–M1–N1
O3–M1–O8
O3–M1–O9
O3–M1–N1
O8–M1–O9
O8–M1–N1
O9–M1–N1
O3–M2–O4
O3–M2–O5
O3–M2–O8
O3–M2–O10
O3–M2–N2
O4–M2–O5
O4–M2–O8
O4–M2–O10
O4–M2–N2
O5–M2–O8
O5–M2–O10
O5–M2–N2
O8–M2–O10
O8–M2–N2
O10–M2–N2
170.58(13)
87.47(11)
75.85(10)
91.70(13)
83.80(13)
94.64(11)
95.96(10)
97.58(13)
94.33(13)
72.93(9)
86.73(12)
171.00(13)
156.38(11)
106.89(12)
91.30(15)
173.58(13)
78.09(9)
88.30(11)
88.84(12)
83.52(12)
96.81(11)
93.92(13)
97.20(13)
94.02(14)
72.54(10)
157.89(11)
103.91(11)
89.49(12)
171.64(13)
92.09(14)
O1–M3–O3
O1–M3–O5
O1–M3–O6
O1–M3–O12
O1–M3–N3
O3–M3–O5
O3–M3–O6
O3–M3–O12
O3–M3–N3
O5–M3–O6
O5–M3–O12
O5–M3–N3
O6–M3–O12
O6–M3–N3
O12–M3–N3
O1–M4–O5
O1–M4–O7
O1–M4–O8
O1–M4–O11
O1–M4–N4
O5–M4–O7
O5–M4–O8
O5–M4–O11
O5–M4–N4
O7–M4–O8
O7–M4–O11
O7–M4–N4
O8–M4–O11
O8–M4–N4
O11–M4–N4
74.59(9)
O1–Cu1–O2
O1–Cu1–O6
O1–Cu1–O8
O1–Cu1–O2S
170.9(3)
95.9(3)
94.3(3)
98.2(3)
93.9(3)
75.9(2)
87.2(3)
90.8(3)
84.9(3)
73.8(2)
158.3(2)
106.9(3)
88.7(3)
171.6(3)
88.5(3)
94.9(3)
86.4(3)
76.2(3)
83.9(5)
168.1(3)
176.9(3)
98.8(3)
94.1(5)
94.0(3)
78.8(3)
88.9(5)
85.1(3)
157.1(5)
110.2(3)
87.6(5)
O4–Cu3–O6
O4–Cu3–O7
O4–Cu3–O8
O4–Cu3–O3S
O4–Cu3–N4
O6–Cu3–O7
O6–Cu3–O8
O6–Cu3–O3S
O6–Cu3–N4
O7–Cu3–O8
O7–Cu3–O3S
O7–Cu3–N4
O8–Cu3–O3S
O8–Cu3–N4
O3S–Cu3–N4
O2–Mn1–O4
O2–Mn1–O5
O2–Mn1–O6
O2–Mn1–O1S
O2–Mn1–N3
O4–Mn1–O5
O4–Mn1–O6
O4–Mn1–O1S
O4–Mn1–N3
O5–Mn1–O6
O5–Mn1–O1S
O5–Mn1–N3
O6–Mn1–O1S
O6–Mn1–N3
O1S–Mn1–N3
72.4(2)
95.8(3)
75.6(3)
157.4(3)
103.9(3)
94.3(3)
88.0(3)
88.1(3)
170.7(3)
170.0(3)
96.9(3)
94.6(3)
93.0(3)
82.8(3)
93.6(3)
73.9(3)
98.1(3)
78.1(3)
156.7(3)
105.6(3)
94.3(3)
88.3(3)
86.7(3)
171.8(3)
174.6(3)
96.1(3)
93.9(3)
88.8(3)
83.6(3)
91.8(3)
87.59(11)
94.49(13)
86.73(14)
169.98(13) O1–Cu1–N1
79.36(10)
95.64(11)
157.66(13) O2–Cu1–O2S
109.25(12) O2–Cu1–N1
173.89(13) O6–Cu1–O8
87.86(13)
84.13(13)
97.98(14)
94.32(15)
87.31(15)
74.46(10)
97.81(11)
78.55(11)
156.06(13) O2–Cu2–O4SA
109.33(14) O2–Cu2–N2
95.30(12)
88.01(11)
85.46(15)
170.42(14) O3–Cu2–N2
174.27(14) O4–Cu2–O8
96.89(15)
92.90(14)
88.02(14)
84.24(13)
88.65(18)
O2–Cu1–O6
O2–Cu1–O8
O6–Cu1–O2S
O6–Cu1–N1
O8–Cu1–O2S
O8–Cu1–N1
O2S–Cu1–N1
O2–Cu2–O3
O2–Cu2–O4
O2–Cu2–O8
O3–Cu2–O4
O3–Cu2–O8
O3–Cu2–O4SA
O4–Cu2–O4SA
O4–Cu2–N2
O8–Cu2–O4SA
O8–Cu2–N2
O4SA–Cu2–N2
to 112.96(16)°, while the O–M–O(N)trans angles span from According to the CSD, only two complexes of this MST
3
1
1
55.29(15)° to 176.9(3)°. Systematic differences between bonds were found: [(μ
and angles involving Cu and Mn centres are negligible. and [Li (tmeda) Cp*TaS
The M⋯M non-bonded distances within the metal core for ethylenediamine, Cp* = C Me ). Hence, 5 represents the first
2
-SC
6
H
4
NO
2
-4)
2
(μ
3 6 4 2 2 3 4
-SC H NO -4) (CuPPh ) ]
3
2
3
Cl]
2
(µ-tmeda) (tmeda = tetramethyl-
3
2
5
5
complexes 2–4 vary in the 3.1040(18)–3.5242(7) Å range. example of a heterometallic transition metal complex with the
Similar to 1, the tetranuclear cations in 2–4 are fastened by {M (µ-X) (µ -X) } molecular structure type. The O atoms from
strong intramolecular O–H⋯O hydrogen bonds involving the the deprotonated Schiff bases bridge the Cu and Mn centres in
oxygen atoms from Schiff bases that act as acceptors and the both double open cube-like cations, whereby two µ [O2, O6]
coordinated solvent molecules (CH OH or H O) as donors and two µ bridge atoms [O4, O8] exist in one cation and two
Table 6). In each of these compounds, uncoordinated anions µ [O10, O12] and two µ bridge atoms [O14, O16] in another
4
2
3
2
3
3
2
2
(
3
2
occupy cavities between closely packed bulky cations and are one (Fig. 7).
linked to them via non-directional electrostatic forces as well
All copper atoms in 5 have five-coordinated distorted
as intermolecular H-bonding.
square-pyramidal coordination environments with O N donor
4
Crystallographic analysis reveals that complex 5 consists of sets which in the case of Cu1, Cu3, Cu4 and Cu5 are formed
2
+
two heterometallic [Cu
slightly different geometries, the [Mn(NCS)4]
two uncoordinated molecules of methanol. Each cation is are formed exclusively by three deprotonated Schiff bases. The
based on the {Cu Mn(µ-O) (µ -O) } metal core which can equatorial Cu–O(N) distances vary from 1.889(4) to 1.996(4) Å,
3
Mn(L)
4
(CH
3
OH)
3
]
cations having by donor atoms of two Schiff base ligands and one coordinated
2
−
anion and molecule of methanol, while in the case of Cu2 and Cu6 they
3
2
3
2
be viewed as the double open cube {M (µ-X) (µ -X) } (Fig. 6). while the apical Cu–O bond lengths lie in the 2.376(4)–2.504(5)
4
2
3
2
This journal is © The Royal Society of Chemistry 2017
Dalton Trans.

View Article Online
Paper
Dalton Transactions
Table 5 Selected geometrical parameters (distances/Å and angles/°)
for 4
Table 6 Hydrogen bonding distances (Å) and angles (°) for complexes
1–5
Cu1–O2
Cu1–O6
Cu1–O7
Cu1–O8
Cu1–OW1
Cu1–N4
Cu2–O1
Cu2–O2
Cu2–O4
Cu2–O8
Cu2–OW3
Cu2–N1
2.451(4)
1.967(4)
1.894(4)
1.965(4)
2.606(6)
1.940(4)
1.862(5)
1.942(4)
1.977(4)
2.501(4)
2.465(7)
1.948(5)
Cu3–O2
Cu3–O4
Cu3–O5
Cu3–O6
Cu3–OW4B
Cu3–N3
Mn1–O3
Mn1–O4
Mn1–O6
Mn1–O8
Mn1–OW2A
Mn1–N2
1.965(3)
2.534(5)
1.880(4)
1.949(4)
2.45(2)
1.926(5)
1.882(4)
1.955(4)
2.475(4)
1.947(4)
2.741(11)
1.940(5)
D–H⋯A
D–H
H⋯A
D⋯A
D–H⋯A
1
O9–H9O⋯O6
O10–H10O⋯O2
O11–H11O⋯O4
2
O9–H1W⋯O6
O10–H2W⋯O2
O11–H3W⋯O4
O12–H5W⋯O7
3
0.85
0.85
0.85
1.84
1.85
1.84
2.687(4)
2.691(12)
2.672(8)
175
171
167
0.85
0.85
0.85
0.85
1.84
1.86
1.89
1.92
2.679(6)
2.688(8)
2.716(16)
2.734(6)
173
165
162
161
O1S–H1S⋯O7
O2S–H2SA⋯O3
O3S–H3SA⋯O1
0.86
0.87
0.85
1.03
1.80
1.87
1.93
1.92
2.635(11)
2.696(10)
2.706(11)
2.75(2)
162.6
157.4
151.1
135.6
O2–Cu1–O6
O2–Cu1–O7
O2–Cu1–O8
O2–Cu1–OW1
O2–Cu1–N4
O6–Cu1–O7
O6–Cu1–O8
O6–Cu1–OW1
O6–Cu1–N4
O7–Cu1–O8
O7–Cu1–OW1
O7–Cu1–N4
O8–Cu1–OW1
O8–Cu1–N4
OW1–Cu1–N4
O1–Cu2–O2
O1–Cu2–O4
O1–Cu2–O8
O1–Cu2–OW3
O1–Cu2–N1
O2–Cu2–O4
O2–Cu2–O8
O2–Cu2–OW3
O2–Cu2–N1
O4–Cu2–O8
O4–Cu2–OW3
O4–Cu2–N1
O8–Cu2–OW3
O8–Cu2–N1
OW3–Cu2–N1
73.80(13)
93.81(16)
79.99(15)
157.86(13)
112.96(16)
94.97(17)
87.74(16)
85.99(16)
167.78(19)
172.28(17)
96.91(17)
94.70(18)
90.47(17)
83.63(17)
85.44(18)
174.55(19)
95.0(2)
96.93(17)
98.26(19)
94.3(2)
87.44(17)
79.10(15)
86.66(17)
83.47(18)
72.84(14)
88.5(2)
O2–Cu3–O4
O2–Cu3–O5
O2–Cu3–O6
O2–Cu3–OW4B
O2–Cu3–N3
O4–Cu3–O5
O4–Cu3–O6
O4–Cu3–OW4B
O4–Cu3–N3
O5–Cu3–O6
O5–Cu3–OW4B
O5–Cu3–N3
O6–Cu3–OW4B
O6–Cu3–N3
OW4B–Cu3–N3
O3–Mn1–O4
O3–Mn1–O6
O3–Mn1–O8
O3–Mn1–OW2A
O3–Mn1–N2
O4–Mn1–O6
O4–Mn1–O8
O4–Mn1–OW2A
O4–Mn1–N2
O6–Mn1–O8
O6–Mn1–OW2A
O6–Mn1–N2
O8–Mn1–OW2A
O8–Mn1–N2
OW2A–Mn1–N2
72.80(14)
95.45(16)
171.46(19) O4SA–H4SB⋯O5
89.7(5)
4
169.37(19) OW1–HW1A⋯O3
0.86
0.85
0.87
0.85
1.95
1.96
1.85
2.08
2.790(4)
2.761(5)
2.665(4)
2.897(19)
166.0
157.4
155.7
159.7
92.87(15)
79.71(15)
160.7(4)
OW2A–HW2A⋯O1
OW3–HW3A⋯O5
OW4B–HW4D⋯O7
5
109.25(19)
171.46(19) O17–H17O⋯O5
0.85
0.85
0.85
0.85
0.85
0.85
0.85
0.85
1.84
1.89
1.80
1.83
1.93
1.89
1.80
1.88
2.664(7)
2.678(7)
2.602(7)
2.668(7)
2.733(6)
2.674(7)
2.653(8)
2.644(7)
163
154
157
168
152
154
178
148
97.0(4)
94.87(18)
91.4(4)
83.73(18)
86.5(5)
173.7(2)
94.06(16)
95.57(17)
101.5(2)
93.9(2)
81.16(16)
87.26(16)
84.43(18)
83.7(2)
75.02(14)
155.29(15)
108.5(2)
84.37(18)
169.6(2)
89.6(2)
O19–H19O⋯O3
O18–H18O⋯O20
O20–H20O⋯O1
O21–H21O⋯O13
O22–H22O⋯O9
O24–H24O⋯O11
O23–H23O⋯O24
170.4(2)
156.87(15)
108.37(18)
87.8(2)
Fig. 6 The ball-and-stick representation of the {Cu
3 2 3 2
Mn(µ-O) (µ -O) }
molecular core in 5. Color scheme: Cu, cyan; Mn, magenta; O, red.
Å range. The cis angles around the Cu atoms range from
7
1
3.39(15) to 124.94(17)° and the trans bond angles span from
58.44(19) to 179.13(16)° (Table 7). In assessing the molecular distances varying from 1.857(4) to 2.331(4) Å. The cis and trans
core type for all 1–5 complexes the upper limits of the apical O–Mn–O(N) bond angles range from 76.60(14) to 102.76(16)°
Cu–O bond lengths were set using the sum of van der Waals and from 165.12(16) to 173.75(16)°, respectively. The Mn3
2
−
radii of the respective atoms, 2.92 Å (van der Waals radius of atom from the [Mn(NCS)4] anion has distorted tetrahedral
Cu is 1.4 Å and of O is 1.52 Å (ref. 29 and 30)). All apical Cu–O coordination geometry with Mn–N bond lengths in the 2.020(6)–
distances in 1–4 are lower than this value. In the case of 5, the 2.061(7) Å range and N–Mn–N bond angles ranging from
distances over or close to it, namely Cu1–O4 = 2.888(4), Cu3– 102.8(3)° to 118.4(3)°. The intermetallic M⋯M non-bonded
O8 = 2.907(4), Cu4–O16 = 2.819(5) and Cu5–O14 = 3.044(5) Å, distances within both metal cores are in the range of 3.087(2)–
were excluded during the core type definition. The correctness 3.608(2) Å. Similar to 1–4, the tetranuclear cations in 5 are
of such differentiation and assignment of the molecular core reinforced by strong intramolecular O–H⋯O hydrogen bonds
type in 1–4 and 5 to the cube and double open cube realiz- involving the oxygen atoms from Schiff bases and coordinated
ations, respectively, are supported by the data obtained from methanol molecules. Besides, each tetranuclear cation in 5 is
the magnetic studies of 1–5 and are in agreement with estab- H-bonded to the uncoordinated molecule of methanol that
lished magnetostructural correlations (see below).
additionally stabilizes the overall double open cube-like metal
Both Mn1 and Mn2 atoms of the double open cube-like cores (Table 6).
cations adopt a distorted octahedral geometry formed by the O
In spite of the similarity of the crystal structures of 1–4
and N atoms of the ligands and methanol with the Mn–O(N) they drastically differ by the number and combinations of
Dalton Trans.
This journal is © The Royal Society of Chemistry 2017

View Article Online
Dalton Transactions
Paper
Table 7 Selected geometrical parameters (distances/Å and angles/°)
for 5
Cu1–O1
Cu1–O2
Cu1–O8
Cu1–O17
Cu1–N1
Cu2–O2
Cu2–O3
Cu2–O4
Cu2–O6
Cu2–N2
Cu3–O2
Cu3–O5
Cu3–O6
Cu3–O19
Cu3–N3
Cu4–O9
Cu4–O10
Cu4–O12
Cu4–O21
1.929(4)
1.996(4)
1.965(3)
2.278(3)
1.950(4)
2.379(3)
1.889(4)
1.954(4)
1.985(3)
1.923(4)
1.945(3)
1.906(3)
1.978(3)
2.504(4)
1.926(4)
1.904(4)
1.975(4)
1.958(4)
2.435(4)
1.938(5)
1.916(4)
1.964(4)
1.953(4)
Cu5–O22
Cu5–N6
Cu6–O10
Cu6–O12
Cu6–O13
Cu6–O14
Cu6–N7
Mn1–O4
Mn1–O6
Mn1–O7
Mn1–O8
Mn1–O18
Mn1–N4
Mn2–O10
Mn2–O14
Mn2–O15
Mn2–O16
Mn2–O23
Mn2–N8
Mn3–N9
Mn3–N10
Mn3–N11
Mn3–N12
2.376(4)
1.937(5)
1.967(4)
2.466(4)
1.896(4)
1.958(3)
1.918(5)
1.916(3)
2.330(4)
1.878(3)
1.899(3)
2.223(4)
1.967(4)
2.315(4)
1.937(4)
1.857(4)
1.917(4)
2.215(4)
1.977(5)
2.060(7)
2.060(7)
2.020(6)
2.040(7)
Fig. 7 The crystal structure of 5 with atom numbering. The hydrogen Cu4–N5
atoms are omitted for clarity.
Cu5–O11
Cu5–O12
Cu5–O16
O1–Cu1–O2
O1–Cu1–O8
O1–Cu1–O17
O1–Cu1–N1
174.81(15)
94.88(15)
91.25(15)
91.42(18)
89.62(14)
91.39(14)
83.76(17)
89.46(14)
169.72(17)
98.54(16)
96.26(13)
85.27(13)
74.81(12)
113.99(15)
178.43(15)
95.91(15)
94.68(17)
84.83(14)
84.36(17)
165.38(16)
95.69(14)
85.76(13)
92.70(14)
169.72(16)
169.95(16)
94.59(16)
94.49(16)
95.28(14)
83.96(16)
87.90(16)
167.09(16)
96.81(16)
97.73(15)
94.36(18)
85.84(15)
94.89(14)
83.23(17)
90.09(15)
168.84(18)
88.58(16)
173.89(17)
94.31(16)
94.65(16)
92.15(19)
88.19(15)
90.68(14)
84.94(18)
94.95(15)
O16–Cu5–N6
O22–Cu5–N6
O10–Cu6–O12
O10–Cu6–O13
O10–Cu6–O14
O10–Cu6–N7
O12–Cu6–O13
O12–Cu6–O14
O12–Cu6–N7
O13–Cu6–O14
O13–Cu6–N7
O14–Cu6–N7
O4–Mn1–O6
O4–Mn1–O7
O4–Mn1–O8
O4–Mn1–O18
O4–Mn1–N4
O6–Mn1–O7
O6–Mn1–O8
O6–Mn1–O18
O6–Mn1–N4
O7–Mn1–O8
O7–Mn1–O18
O7–Mn1–N4
172.12(19)
89.00(18)
73.39(15)
95.77(16)
84.97(15)
158.44(19)
92.95(17)
87.69(16)
124.94(17)
179.13(16)
94.51(18)
84.64(18)
76.76(13)
97.04(15)
88.32(15)
91.67(15)
171.17(16)
90.69(14)
87.39(14)
168.39(13)
100.75(15)
173.74(16)
89.77(15)
91.42(16)
93.31(15)
83.09(16)
90.84(17)
76.57(14)
89.84(15)
88.41(14)
165.20(15)
102.76(17)
95.35(16)
90.60(15)
88.63(16)
173.56(19)
173.24(17)
92.14(17)
91.05(19)
91.22(16)
82.98(18)
91.88(18)
102.8(3)
coordinated CH
methanol molecules are coordinated to the tetranuclear metal
core forming the {Cu
The coordination of a fourth solvent molecule has not
occurred, possibly due to the existence of a weak contact
3 2
OH and H O molecules. In the case of 1, only
3
Mn(µ
3
-O)
4
}(CH
3
OH)
3
fragment (Fig. 8). O2–Cu1–O8
O2–Cu1–O17
O2–Cu1–N1
O8–Cu1–O17
between the five-coordinated copper atom of the core and the O8–Cu1–N1
−
O17–Cu1–N1
O2–Cu2–O3
O2–Cu2–O4
uncoordinated I
four solvent molecules (one to each metal atom) to the core
with mixed-solvent CH
3 3
the disorder, the formation of two mixed-solvent {Cu Mn(µ -
3
anion. Complex 2 displays coordination of
3
OH
3
/H
2
O combination, while, due to O2–Cu2–O6
O2–Cu2–N2
O3–Cu2–O4
O3–Cu2–O6
O) }(H O) (CH OH) and {Cu Mn(µ -O) }(H O) (CH OH) frag-
4
2
3
3
3
3
4
2
2
3
ments is observed in 3. In the case of compound 4, similarly O3–Cu2–N2
O4–Cu2–O6
O4–Cu2–N2
O6–Cu2–N2
O2–Cu3–O5
O2–Cu3–O6
O2–Cu3–O19
O2–Cu3–N3
to 1, only one type of solvent molecule, namely water, is co-
ordinated to the metal core. Taking into account the disorder
of one of the coordinated water molecules, the two {Cu
O) }(H O) and {Cu Mn(µ -O) }(H O) fragments can be
selected (Fig. 8) in 4. Such a diversity of the observed combi-
3
Mn(µ
3
-
4
2
4
3
3
4
2
3
nations of coordinated solvent molecules in 1–4 could be a O5–Cu3–O6
O8–Mn1–O18
O8–Mn1–N4
O5–Cu3–O19
O5–Cu3–N3
O6–Cu3–O19
result of a whole set of different factors. The pronounced influ-
ence of the synthetic conditions (temperature, kinetics, etc.)
O18–Mn1–N4
O10–Mn2–O14
O10–Mn2–O15
O10–Mn2–O16
O10–Mn2–O23
O10–Mn2–N8
O14–Mn2–O15
O14–Mn2–O16
O14–Mn2–O23
O14–Mn2–N8
O15–Mn2–O16
O15–Mn2–O23
O15–Mn2–N8
O16–Mn2–O23
O16–Mn2–N8
O23–Mn2–N8
N9–Mn3–N10
N9–Mn3–N11
N9–Mn3–N12
N10–Mn3–N11
N10–Mn3–N12
N11–Mn3–N12
and the chemical composition of the synthetic mixture can be O6–Cu3–N3
O19–Cu3–N3
O9–Cu4–O10
O9–Cu4–O12
understood if one considers these mixtures as dynamic combi-
natorial libraries (DCL). The DCL concept,
3
3,34
introduced by
Jean-Marie Lehn, describes a library of components, which O9–Cu4–O21
O9–Cu4–N5
O10–Cu4–O12
O10–Cu4–O21
can interact in a reversible manner (such as coordination
bondings). In this way, the structure and composition of a
reaction product is a complex function of all possible equili- O10–Cu4–N5
O12–Cu4–O21
O12–Cu4–N5
O21–Cu4–N5
bria in such systems and, moreover, may be influenced by the
crystallization process, as the latter is also a reversible one.
Hence, even the lattice energy may be crucial in determining O11–Cu5–O12
O11–Cu5–O16
O11–Cu5–O22
O11–Cu5–N6
which structure (coordination compound) will dominate and
crystallize from the solution phase.
105.9(3)
109.8(3)
118.4(3)
106.6(3)
In the crystal structure of 5 both tetranuclear metal cores O12–Cu5–O16
O12–Cu5–O22
O12–Cu5–N6
O16–Cu5–O22
3 3 4 3 3
take part in the formation of similar {Cu Mn(µ -O) }(CH OH)
fragments (Fig. 9) showing the coordination of three methanol
112.8(3)
This journal is © The Royal Society of Chemistry 2017
Dalton Trans.

View Article Online
Paper
Dalton Transactions
Fig. 8 Ball-and-stick representations showing the coordination of the
solvents to molecular cores in 1–4. Color scheme: Cu, cyan; Mn,
magenta; O, red; C, light grey.
Fig. 10 The MSTs and distribution of known tetranuclear Cu/Mn com-
plexes (according to the CSD) and compounds 1–5.
4 3 4
ous MST {M (µ -X) }, where in all cases the bridging atoms are
oxygens [ref. 37 and 38 and complexes 1–4]. In the M X
4
6
4
Fig. 9 Ball-and-stick representations showing the coordination of the
solvents to molecular cores in 5. Color scheme: Cu, cyan; Mn, magenta;
O, red; C, light grey.
3
9,40
group the star-like MST {M
4
(µ-X)
6
}
4
and MST {M (µ-X)
38,41
3 2
(µ -X) },
which can be viewed as combination of uncompleted
cubes, were found. Obviously, the synthetic procedure used for
the complex preparation can be one of the main determining
molecules to two Cu(II) and one Mn(III) metal centre in each factors for the MST formation. While complexes from the
tetranuclear core. The absence of such solvent coordination to
group with MST {M (µ-X) } were obtained using metal
the remaining copper (Cu2 and Cu6) atoms of the cores can be salts or metal complexes as sources of the metals, the com-
explained as follows: in the case of Cu2 – by the manner of the pounds from this group with MST {M (µ -X) } were synthesized
molecular cation packing that leads to the appearance of steric using the direct synthesis approach only. For the M group,
hindrance and formation of insufficient space for the solvent the complexes with star-like MST {M (µ-X) } were obtained
coordination and, in the case of Cu6 – by the existence of a using the “complex as ligand” strategy only, while compounds
very weak copper–sulfur contact (Cu6⋯S3 = 4.701(2) Å).
possessing MST {M (µ-X) (µ -X) were prepared by the
M
4
X
4
4
4
4
3
4
X
4 6
4
6
}
2
4
4
3
Finally, when analyzing the crystal structures of 1–5, it can “complex as ligand” strategy and by direct synthesis. Thus, in
be concluded that all of them have similar metal atom arrange- some cases, the preparation methodology is just a way for
ments described by the general tetranuclear M
in which the molecular {Cu Mn(µ -O) } and {Cu
µ -O) } cores for 1–4 and 5, respectively, can be assigned. In cular structural type.
4
X
4
group, obtaining the desired compound, while in other cases it is the
Mn(µ-O)
2
only possible way to prepare a complex with a certain mole-
3
3
4
3
(
3
2
the cases of 1–4, the cube-like shape of the heterometallic
cores is supported by the coordination of four or three solvent
Magnetic properties
molecules with mixed or single-type solvent combination, Three Cu(II) centres and one Mn(III) centre yield a high-temp-
while in 5, a similar coordination of three methanol molecules erature limit for the effective magnetic moment μ
eff
=
2
2
1/2
to both double open cube-like cores is observed. Therefore, the [3gCu SCu(SCu + 1) + gMn SMn(SMn + 1) μB which amounts to
various synthetic mixtures, as well as the different synthetic 5.7µB when all g = 2.0 (SCu = 1/2, SMn = 2). On cooling from
approaches, used for complex 1–5 preparation lead to the for- room temperature the effective magnetic moment for 1 gradu-
mation of the products showing quite similar but at the same ally decreases from µeff = 5.54µ to the value of 1.98µB at T =
B
time rather different molecular structures.
2.0 K (Fig. 11). This feature indicates that an antiferromagnetic
Analysis of the MSTs of known crystal structures of the tet- coupling among the magnetic centres applies. Down to 2.0 K
ranuclear Cu/Mn molecular complexes (according to the CSD), the molar magnetic susceptibility only increases showing that
where all metal atoms are linked by a single bridging atom, is S > 0 is the ground state. At a very low temperature, also some
shown in Fig. 10. Depending on the number of bridging (mononuclear) fragments of the tetranuclear bulk complex
atoms, two groups with general formulae M
4 4 4 6
X and M X can might be present; a modeling of such paramagnetic impurities
be distinguished. The M X group includes three MSTs with is rather problematic though they could influence the low-
4
4
the following realizations: diamond-like O-bridged {M
4
(µ- temperature susceptibility data.
double open
The molar magnetization per formula unit possesses the
O-bridged cube {M (µ-X) (µ -X) } [complex 5] and more numer- saturation limit M1 = Mmol/NAμB = 3gCuSCu + gMnSMn which
3
5
36
X)
4
},
4 4
consecutive N-bridged {M (µ-X) },
4
2
3
2
Dalton Trans.
This journal is © The Royal Society of Chemistry 2017

View Article Online
Dalton Transactions
Paper
paths which involve pairs of sharp and open angles so that a
competition between ferromagnetic and antiferromagnetic
interactions is expected: Mn–O–Cu2 (88°, 107°), Mn–O–Cu1
(
1
91°, 106°), Cu1–O–Cu2 (87°, 108°), and Cu1–O–Cu3 (89°,
04°). Thus J could be positive or negative, but less negative
4
than J
= J′
The diagonalization of the above Hamiltonian produces a
2
2
and J′ . During the fitting procedure it was found that
J
2
2
can be safely fixed.
set of eigenvalues entering the partition function. Then the
magnetic susceptibility and the magnetization are evaluated
4
3
by using the formulae of the statistical thermodynamics. The
susceptibility data were corrected for the molecular-field cor-
rection (zj ) and the temperature-independent magnetism χTIM
via χcorr = χmol/[1 − (zj )χmol] + χTIM. The latter term accounts
for the uncompensated underlying diamagnetism and the
temperature-independent paramagnetism, along with the dia-
magnetic signal of the sample holder.
An advanced fitting procedure using a genetic algorithm
has been applied. The final set of magnetic parameters for 1
through 4 is listed in Table 8. In accordance with expectations,
4
9
g
Mn < 2 for the d system and gCu > 2 for the d complex hold
true. (There is a second set of magnetic parameters with | J | <
J4| fitting well the magnetic functions listed in the ESI† – fit-
2
|
b.) The magnetic properties of complex 2 are analogous to
those of complex 1 (Fig. 11) and an almost identical set of
magnetic parameters was obtained.
The magnetic functions for complex 3 are a bit different: (i)
the susceptibility at T = 2 K is much lower; (ii) the magnetiza-
tion at B = 7 T is much lower (Fig. 12). This indicates an
increase of the antiferromagnetic exchange. The last complex
Fig. 11 Magnetic functions for 1 and 2. Left – Effective magnetic
moment, right – magnetization per formula unit, inset – molar magnetic
susceptibility (SI units). Solid lines – fitted: blue – fit-a, red – fit-b.
yields M
adopts a value of only M
1
= 7 for Smax = 7/2. The measured magnetization
= 1.2 at T = 2.0 K and B = 7 T. This
4
behaves analogously to 3.
Compound 5 consists of two [Cu Mn ] cubes of the open
1
III
3
reduction is a fingerprint of a considerable antiferromagnetic
exchange, eventually combined with single-ion zero-field split-
ting at the Mn(III) centre.
The susceptibility and magnetization data were fitted simul-
taneously by applying an error functional F = E(χ) × E(M) that
accounts uniformly for the relative errors in susceptibility and
magnetization. The calculated values have been reconstructed
by considering the following spin Hamiltonian
II
2−
type and one [Mn (NCS)
4
]
anion. The high-temperature limit
2
of the effective magnetic moment is μeff = [6gCu
S
Cu(SCu + 1) +
gMn SMn(SMn + 1) + gMn′ SMn′(SMn′ + 1)] μB which amounts to
0.34µ when all g = 2.0 (SCu = 1/2, SMn = 2, SMn′ = 5/2). At
) owing
to higher g-factors (Fig. 13). At the same time the magnetiza-
tion per formula unit should saturate to M = 6gCu
Mn′ = 19 but its value at T = 2.0 K and B = 7 T is
2
2
1/2
2
1
B
room-temperature, however, µeff is much higher (12.7µ
B
1
Cu
S +
Mn
2g SMn + gMn′S
ex
a
ꢂ2
only M1 = 7.4. This again confirms a dominating antiferro-
magnetic coupling along with sizable zero-field splitting.
ˆ
H
~
~
~
~
¼ ꢂ ½ J2ðSMnꢁSCu1Þ þ J′2ðSCu2ꢁSCu3Þꢀℏ
~
~
~
~
þ J4½ðSCu1ꢁSCu2Þ þ ðSCu1ꢁSCu3Þ
The data fitting for 5 has been based upon the assumption
ꢂ2
~
~
~
~
III
þ ðSCu2ꢁSMnÞ þ ðSCu3ꢁSMnÞꢀℏ
þ DMnðSMn;z ꢂ SMn =3Þℏ
that the formula unit contains 2/3 of the [Cu Mn ] cube and
3
2
2
ꢂ2
II
ˆ
ˆ
1/3 of the [Mn ] complex (S = 5/2, g = 2). Thus the magnetic
ꢂ1
data were divided by a factor of 3, then fitted as before with a
ˆ
ˆ
ˆ
ˆ
þ μ BaðgMnSMn;a þ gCuSCu1;a þ gCuSCu2;a þ gCuSCu3;aÞℏ
B
II
paramagnetic impurity of Mn (xPI = 1/3) and the fitted data
ð6Þ
were backtransformed to the original scale by multiplying the
(
a = x, y, z) where the first term refers to the isotropic calculated susceptibility and magnetization by a factor of
exchange, the second – the zero-field splitting, and the third – 3. The final set of magnetic parameters is also listed in Table 8
the Zeeman term. In the isotropic exchange part several and it is seen that the coupling constants for such an open-
different coupling constants occur. The constants J
2
and J′
2
structure system are much lower relative to compounds 1–4.
reflect open angles between Mn–O–Cu1 (100°, 105°) and Cu2–
O–Cu3 (101°, 104°) centres and they are expected to be nega-
Magnetostructural J–α correlations
4
2
tive (an antiferromagnetic exchange interaction).
The The magnetic parameters of the complexes under study are
remaining four interactions precede along the superexchange listed in Table 9 along with structural parameters that charac-
This journal is © The Royal Society of Chemistry 2017
Dalton Trans.

View Article Online
Paper
Dalton Transactions
a
Table 8 Calculated magnetic parameters for 1–5
b
Complex
g
Mn
g
Cu
2
J /hc
J
4
/hc
D
Mn/hc
(zj )/hc
χ
TIM
R(χ)/%
R(M)/%
1
2
3
4
5
1.859
1.869
1.988
2.000
1.851
2.313
2.349
2.009
2.000
2.178
−54.8
−54.8
−45.0
−52.5
−22.7
−27.0
−27.0
−37.5
−39.1
−11.6
−13.6
−14.4
−2.8
−0.075
−0.013
−0.079
∼0
+0.2
4.0
3.5
2.2
1.9
2.5
3.4
4.3
0.7
1.0
3.3
+14.4
+12.2
+15.0
+40.0
−2.5
c
−23.4
0.268
a
Coupling constants confirming the definition Hˆ ij ¼ ꢂJðSiꢁSjÞ in cm−1. J
b
~
~
2
– in two bases, J
4
– within four walls. Temperature independent mag-
netism χTIM in units of 10− m mol (SI). g(Mn ) = 1.950 and 1.956.
9
3
−1
c
II
terize the cube-like core. Data for two analogous complexes
have been retrieved from the CSD and added for
3
7,38
comparison.
These data form a basis for magnetostructural correlations.
The problem is that there are six M⋯M contacts, twelve M–O
distances, twelve M–O–M angles, but only two J-constants at
the disposal. This causes that some selection and/or averaging
of geometrical parameters is necessary. An attempt to
correlate the bonding angle β in the basal planes (defined by
the longest Cu⋯Cu contacts) with the coupling constant
2 4 4
occurring twice (J ) appeared recently for tetrahedro-{Cu O }
4
4
complexes.
An enlarged set of 41 complexes has been treated by
methods of chemometry giving rise to six clusters according to
4
5
their similarity by means of Cluster Analysis (CA). Magnetic
and structural data for 1–4 were added to this set and renum-
bered as complexes 42–45. Complex 5 possesses two structural
units but the common set of magnetic parameters giving rise
to members 46 and 47. Then CA was repeated as displayed in
Fig. 14. Four heterometallic cubes span the cluster 1 (no. 42,
4
3, 44, 45) and cluster 3 (no. 46 and 47). In these cases the
core of the complex refers to a compressed prism (R < R ).
The clusters 4, 5, and 6 resemble an elongated prism with
R > R .)
4
2
(
4
2
For such a case the appropriate magnetostructural corre-
lation is represented by the J vs. α plot as displayed in Fig. 15.
Here the new members (complexes 1–5, no. 42–47) are distin-
Fig. 12 Magnetic functions for 3 and 4. Left – Effective magnetic
moment, right – magnetization per formula unit, inset – molar magnetic
susceptibility (SI units). Solid lines – fitted: blue – fit-a, red – fit-b (in
fact coincide).
4
guished by triangle symbols. It can be concluded that the
coupling constant within the walls of the compressed prism J
4
approximately correlates with the averaged Cu–O–Cu angle in
walls α according to a straight line with a negative slope.
However, a better correlation would be a curved line of the
hyperbolic nature.
Conclusions
The present study shows the successful utilization of the
“direct synthesis” approach for the preparation of highly
nuclear heterometallic complexes with polydentate Schiff base
ligands formed in situ. Open-air reactions of copper and
manganese powders with a solution containing the conden-
sation product of salicylaldehyde and ethanolamine yielded
Fig. 13 Magnetic functions for 5. Left – Effective magnetic moment,
right – magnetization per formula unit, inset – molar magnetic suscep-
tibility (SI units). Solid lines – fitted: blue – fit-a, red – fit-b.
II
III
four new Cu Mn
3 4 3 3 3
complexes [Cu Mn(L) (CH OH) ]I (1),
[Cu Mn(L) (CH OH) (H O)]NCS·H O (2), [Cu Mn(L) (CH OH)
3
4
3
3
2
2
3
4
3
Dalton Trans.
This journal is © The Royal Society of Chemistry 2017

View Article Online
Dalton Transactions
Paper
This journal is © The Royal Society of Chemistry 2017
Dalton Trans.

View Article Online
Paper
Dalton Transactions
2
6
1
014 and UID/QUI/00100/2013 and fellowship SFRH/BPD/
3710/2009) and Slovak grant agencies (VEGA 1/0534/16,
/0919/17 and APVV-14-0078). Special thanks to a crystallo-
graphic referee for the valuable comments allowing us to
improve the paper.
Notes and references
4 4
Fig. 14 Cluster analysis for 41 {Cu O } cubes and 6 additional
{Cu
3
MnO
4
} cubes. Ward’s method, city-block metrics.
1 S. K. Langley, N. F. Chilton, L. Ungur, B. Moubaraki,
L. F. Chibotaru and K. S. Murray, Inorg. Chem., 2012, 51,
1
1873–11881.
L. R. Piquer and E. C. Sanudo, Dalton Trans., 2015, 44,
771–8780.
2
3
8
S. W. Zhang, H. Li, E. Y. Duan, Z. S. Han, L. L. Li,
J. K. Tang, W. Shi and P. Cheng, Inorg. Chem., 2016, 55,
1202–1207.
4
D. D. Zhang, F. Cao, P. T. Ma, C. Zhang, Y. Song,
Z. J. Liang, X. J. Hu, J. P. Wang and J. Y. Niu, Chem. – Eur.
J., 2015, 21, 17683–17690.
5
6
7
K. Liu, W. Shi and P. Cheng, Coord. Chem. Rev., 2015, 289,
74–122.
S. K. Langley, N. F. Chilton, B. Moubaraki and K. S. Murray,
Inorg. Chem. Front., 2015, 2, 867–875.
M. Maity, M. C. Majee, S. Kundu, S. K. Samanta,
E. C. Sanudo, S. Ghosh and M. Chaudhury, Inorg. Chem.,
Fig. 15 Magnetostructural correlation for cubes resembling a com-
pressed prism with {Cu } and {Cu MnO } cores; confidence and pre-
4
O
4
3
4
diction intervals are shown at 95% probability level, correlation coeffi-
cient r = −0.47.
2
015, 54, 9715–9726.
8
J. L. Liu, J. Y. Wu, Y. C. Chen, V. Mereacre, A. K. Powell,
L. Ungur, L. F. Chibotaru, X. M. Chen and M. L. Tong,
Angew. Chem., Int. Ed., 2014, 53, 12966–12970.
(
(
H O) ]Br·0.45H O (3) and [Cu Mn(L) (H O) ]BF ·0.6H O
4). The use of the “building block” approach for comparative
9 V. Mereacre, A. M. Ako, R. Clerac, W. Wernsdorfer,
I. J. Hewitt, C. E. Anson and A. K. Powell, Chem. – Eur. J.,
2008, 14, 3577–3584.
2
2.55
2
3
4
2
3.4
4
2
purposes allowed us to synthesize the novel compound
[
Cu Mn(L) (CH OH) ] [Mn(NCS) ]·2CH OH (5). Both the used 10 P. Buchwalter, J. Rose and P. Braunstein, Chem. Rev., 2015,
3 4 3 3 2 4 3
preparation methodologies afforded complexes with tetranuc-
lear crystal structures but with different molecular core types – 11 J. A. Mata, F. E. Hahn and E. Peris, Chem. Sci., 2014, 5,
115, 28–126.
cube-like {Cu Mn(µ -O) } in 1–4 and the rather rare double
1723–1732.
3
3
4
open cube {Cu
3
Mn(µ-O)
2
(µ
3
-O)
2
} in 5, which is the first hetero- 12 B. G. Cooper, J. W. Napoline and C. M. Thomas, Catal.
metallic example with such a metal arrangement built by
oxygen bridges only. Moreover, the different numbers and 13 D. S. Nesterov, E. N. Chygorin, V. N. Kokozay, V. V. Bon,
Rev., 2012, 54, 1–40.
combinations of coordinated to the metal core methanol and
water molecules in 1–5 allowed us to disclose a wide range of
R. Boca, Y. N. Kozlov, L. S. Shul’pina, J. Jezierska,
A. Ozarowski, A. J. L. Pombeiro and G. B. Shul’pin, Inorg.
Chem., 2012, 51, 9110–9122.
mixed-solvent or single-type solvent heterometallic {Cu Mn
3
(
O)
along with variable-field magnetization measurements of 1–5
showed an antiferromagnetic coupling of medium magnitude 15 A. Biswas, L. Mandal, S. Mondal, C. R. Lucas and
4
}(H
2
O)
x
(CH
3
OH)
y
fragments. Magnetic susceptibility data 14 C. R. Groom and F. H. Allen, Angew. Chem., Int. Ed., 2014,
53, 662–671.
(
−55 to −22 cm− ). For these systems resembling a compressed
1
S. Mohanta, CrystEngComm, 2013, 15, 5888–5897.
prism the coupling constant in walls J
4
correlates with the aver- 16 H. Oshio, M. Nihei, A. Yoshida, H. Nojiri, M. Nakano,
aged bonding angles in walls α: the function J vs. α develops
approximately according to a straight line with a negative slope.
A. Yamaguchi, Y. Karaki and H. Ishimoto, Chem. – Eur. J.,
2005, 11, 843–848.
4
1
7 W. G. Wang, A. J. Zhou, W. X. Zhang, M. L. Tong,
X. M. Chen, M. Nakano, C. C. Beedle and
D. N. Hendrickson, J. Am. Chem. Soc., 2007, 129, 1014–1015.
8 V. A. Milway, F. Tuna, A. R. Farrell, L. E. Sharp, S. Parsons
and M. Murrie, Angew. Chem., Int. Ed., 2013, 52,
1949–1952.
Acknowledgements
1
This work was supported by Fundação para a Ciência e
Tecnologia (FCT), Portugal (projects PTDC/QEQ-QIN/3967/
Dalton Trans.
This journal is © The Royal Society of Chemistry 2017

View Article Online
Dalton Transactions
Paper
1
2
9 D. S. Nesterov, O. V. Nesterova, V. N. Kokozay and 32 K. Tatsumi, Y. Inoue, H. Kawaguchi, M. Kohsaka,
A. J. L. Pombeiro, Eur. J. Inorg. Chem., 2014, 4496–4517.
A. Nakamura, R. E. Cramer, W. Vandoorne, G. J. Taogoshi
0 A. C. D. Midoes, P. E. Aranha, M. P. dos Santos, E. Tozzo,
and P. N. Richmann, Organometallics, 1993, 12, 352–364.
S. Romera, R. H. D. Santos and E. R. Dockal, Polyhedron, 33 J. M. Lehn, Chem. – Eur. J., 1999, 5, 2455–2463.
008, 27, 59–64. 34 J. M. Lehn, Angew. Chem., Int. Ed., 2013, 52, 2836–2850.
1 Stoe & Cie, X-SHAPE. Revision 1.06, Stoe & Cie GmbH, 35 Y. S. Moroz, L. Szyrwiel, S. Demeshko, H. Kozlowski,
2
2
Darmstadt, Germany, 1999; Stoe & Cie, X-RED. Version 1.22,
F. Meyer and I. O. Fritsky, Inorg. Chem., 2010, 49, 4750–
Stoe & Cie GmbH, Darmstadt, Germany, 2001.
4752.
2
2 G. M. Sheldrick, SHELX97, A system of computer programs 36 C. H. Ge, A. L. Cui and H. Z. Kou, Inorg. Chem. Commun.,
for X-ray structure determination, University of Göttingen, 2009, 12, 926–928.
Göttingen, Germany, 1997; G. M. Sheldrick, Acta 37 M. Nihei, A. Yoshida, S. Koizumi and H. Oshio, Polyhedron,
Crystallogr., Sect. A: Fundam. Crystallogr., 1990, 46, 467. 2007, 26, 1997–2007.
3 K. Brandenburg, Diamond 2.1e, Crystal Impact GbR, Bonn, 38 Q. Wu, Q. Shi, Y. G. Li and E. B. Wang, J. Coord. Chem.,
999. 2008, 61, 3080–3091.
4 Agilent, CrysAlisPro, Agilent Technologies, Yarnton, England, 39 S. Biswas, S. Naiya, C. J. Gomez-Garcia and A. Ghosh,
011. Dalton Trans., 2012, 41, 462–473.
5 O. V. Dolomanov, L. J. Bourhis, R. J. Gildea, J. A. K. Howard 40 S. Mondal, S. Mandal, L. Carrella, A. Jana, M. Fleck,
2
2
2
2
2
2
1
2
and H. Puschmann, J. Appl. Crystallogr., 2009, 42, 339–341.
6 G. M. Sheldrick, Acta Crystallogr., Sect. A: Fundam.
Crystallogr., 2008, 64, 112–122.
A. Kohn, E. Rentschler and S. Mohanta, Inorg. Chem., 2015,
54, 117–131.
41 Y. Sunatsuki, H. Shimada, T. Matsuo, M. Nakamura, F. Kai,
N. Matsumoto and N. Re, Inorg. Chem., 1998, 37, 5566–
5574.
7 A. L. Spek, Acta Crystallogr., Sect. C: Cryst. Struct. Commun.,
2
015, 71, 9–18.
8 K. Nakamoto, Infrared and Raman Spectra of Inorganic and 42 O. Kahn, Molecular Magnetism, VCH, New York, 1993.
Coordination Compounds, Part B, Applications in 43 R. Boča, A Handbook of Magnetochemical Formulae, Elsevier,
Coordination, Organometallic and Bioinorganic Chemistry,
Wiley, Chichester, 5th edn, 1997.
9 A. Bondi, J. Phys. Chem., 1964, 68, 441–451.
Amsterdam, 2012.
44 Z. Puterová-Tokarová, V. Mrázová, J. Kozisek, J. Valentová,
B. Vranovičova and R. Boča, Polyhedron, 2014, 70,
52–58.
2
3
3
0 A. Bondi, J. Phys. Chem., 1966, 70, 3006–3007.
1 C. H. Li, S. C. F. Kui, I. H. T. Sham, S. S. Y. Chui and 45 M. Makohusová, V. Mrázová, W. Haase and R. Boča,
C. M. Che, Eur. J. Inorg. Chem., 2008, 2421–2428. Polyhedron, 2014, 81, 572–582.
This journal is © The Royal Society of Chemistry 2017
Dalton Trans.