NJC

View Article Online

PAPER View Journal | View Issue

The effects of protonated heterocyclic

Published on 09 August 2018. Downloaded by Uniwersytet Wroclawski on 2/11/2019 3:30:38 PM.

cations on the structural and magnetic

Cite this: New J. Chem., 2018,
42, 15705 properties of tetrachlorocuprate(II) anions;
X-ray, magnetochemical and EPR studies†
Ülo Kersen,*a Andrzej Wojtczak,b Alina Bienko *c and Julia Jezierskac

Received 25th June 2018,
Accepted 8th August 2018
DOI: 10.1039/c8nj03155j
rsc.li/njc
The crystal and molecular structures and magnetic properties of two new complexes (HMepy)2[CuCl4]2H2O,
2, and (H2Me2ppz)[CuCl4], 3, formed by tetrachlorocuprate(II) anions and organic heterocyclic cations, bis(3-
amino-2-chloro-5-methylpyridinium) (HMepy) and trans-2,5-dimethylpiperazinium (H2Me2ppz), respectively,
have been determined by the analysis of their X-ray diffraction data, magnetic susceptibility behaviour and
EPR spectral parameters. These characteristics differentiate the [CuCl4]2 anions in 2 and 3 as well as that
previously reported in (H2bipip)[CuCl4]H2O, 1, (bipip = 4,40 -bipiperidine) (Tuikka et al., CrystEngComm, 2013,
15, 6177; Wikaira et al., J. Coord. Chem., 2016, 69(1), 57). The present study reveals a significant effect of the
organic cations on the geometry and magnetic properties of the [CuCl4]2 anions caused evidently by the
changes in the crystal packing structure. A characteristic relationship was observed between the average cis-
angles around the Cu(II) ions in [CuCl4]2 for 1–3, affected by the Cu(II) coordination geometry, and g-tensor
components derived from the EPR spectra, a measure of interaction of copper(II) unpaired electrons in
the [CuCl4]2 anions with an external magnetic field. The magneto-structural correlations revealed weak
antiferromagnetic exchange interactions between the [CuCl4]2 anions transmitted by different pathways.

Introduction EPR spectroscopy.4 A very specific feature discovered and studied

The [CuCl4]2 anion is one of the most frequently crystallo-
graphically characterized transition-metal complex ions. Corre-
lations between different geometric parameters of more than sixty
[CuCl4]2 anions indicated a significant dependence of their crystal
and molecular structures on the nature of the organic cation
(protonated aliphatic, cyclic and various substituted heterocyclic
amines). In general, an approximate planar geometry is observed
when shorter and/or a greater number of hydrogen-bonding inter-
actions are present as a result of more effective N–H  Cl hydrogen
bonds and reduction of the electrostatic repulsions between the
chlorides in [CuCl4]2 anions.3 It was also shown that the distortion
of the [CuCl4]2 anion geometry from planar to pseudo-tetrahedral
affects the values of the g-tensor components dependent on the
local symmetry of the paramagnetic center and are measured by
a
School of Chemical Engineering, Aalto University, P.O. Box 16100, 00076 Aalto,
Finland. E-mail: ulokersen@gmail.com
b
Department of Crystallochemistry and Biocrystallography, Faculty of Chemistry,
Nicolaus Copernicus University, 87-100 Torun, Poland
c
Faculty of Chemistry, University of Wroclaw, 14 F. Joliot-Curie, 50-383 Wroclaw,
Poland. E-mail: alina.bienko@chem.uni.wroc.pl
† Electronic supplementary information (ESI) available. CCDC 1455402 (2) and
1530155 (3). For ESI and crystallographic data in CIF or other electronic format
see DOI: 10.1039/c8nj03155j
for a wide variety of complexes containing [CuCl4]2 anions is
the transmission of magnetic exchange through non-bonding
Cu–Cl  Cl–Cu contacts between the adjacent anions. The magni-
tude of J corresponding to this exchange energy increases with (a)
shorter distance, d, between the chloride ions, (b) the Cu–Cl  Cl
and Cl  Cl–Cu angles, y1 and y2, respectively, approaching 1801
and (c) the Cu–Cl  Cl–Cu dihedral angle, t, approaching 01 or
1801.2,5 It was shown for [NH3(CH2)nNH3]CuCl4 that when n = 2, the
magnetic exchange through a Cu–Cl  Cl–Cu bridge between the
layers formed by the square planar [CuCl4]2 anions is only about
twice as small as through the Cu–Cl–Cu bridge between the anions
in the layers. For complexes with n-values increasing from 2 to 6,
the interlayer magnetic exchange integrals decreased rapidly from
J0 = 19.04 cm1 to J0 = 0.056 cm1 as the distance d increased
from 3.625 Å to 6.525 Å, whereas the J due to the intralayer
exchange decreased from 29.55 cm1 to 10.56 cm1.6 This depen-
dence is more distinct for the isomorphous salts of [CuBr4]2
anions for which the values of J0 are apparently greater.7 The same
general trend is observed for pseudo-tetrahedral complexes,
although there is more scatter in the data due to unpredictable
distortions of the geometry.5
The present work is a continuation of our search for differ-
ent structural effects responsible for the magnetic exchange
interaction between Cu(II)–Cu(II) ions through bridges as well as

This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2018 New J. Chem., 2018, 42, 15705--15713 | 15705
 View Article Online

Paper NJC

Cu–Cl  Cl–Cu contacts. The latter has never been systemati-
cally investigated in the same way as for the salts with [CuBr4]2
anions.5 Previously, we reported a correlation between the
magnetic and structural properties for 4,4 0 -bi-piperidinium
tetrachlorocuprate(II) monohydrate, (H2bipip)[CuCl4]H2O (1)1,2
which was treated as a reference for the comparative magneto-
structural studies of newly synthesized and crystallographically
characterized compounds, bis(3-amino-2-chloro-5-methylpyri-
dinium) tetrachloro-cuprate(II) dihydrate, (HMepy)2[CuCl4]2H2O
Published on 09 August 2018. Downloaded by Uniwersytet Wroclawski on 2/11/2019 3:30:38 PM.
(H2bipip)[CuCl4]H2O (1). The reaction of CuCl22H2O with
two equivalents of 4,4 0 -bipiperidine (bipip) and 6 M HCl yielded
complex 1, as reported previously.1,2
Anal. calc for C10H24Cl4CuN2O (393.65/mol): C 30.48%,
H 6.10%, N 7.11%, Cl 36.07%, Cu 16.13%; Found: C 30.42%,
H 5.92%, N 10.53%, Cl 36, 14%, Cu 16.21%.
(HMepy)2[CuCl4]2H2O (2). 3-Amino-2-chloro-5-methylpyridine
(0.250 g, 1.75 mmol) was dissolved in 3.0 mL of 6 M HCl.
A solution of CuCl22H2O (0.149 g, 0.875 mmol) in 2.0 mL of 6 M

(2) and trans-2,5-dimethylpiperaziniumtetrachlorocuprate(II),
(H2Me2ppz)[CuCl4] (3).
The aim of this study was to identify a relationship between
the crystal structure of the anions (affected by the organic
cations) and the magnetic properties of the Cu(II) ions in
[CuCl4]2 anions. These properties were analysed in terms of
magnetic exchange interactions between the Cu(II) ions
(manifested by the dependence of the magnetic susceptibility
on temperature) and of the interaction of the Cu(II) ions with an
external magnetic field detected by EPR and measured by the
g-tensor.
Experimental
Synthesis of complexes
3-Amino-2-chloro-5-methylpyridine, CuCl22H2O and trans-2,5-
dimethylpiperazine were purchased from Sigma-Aldrich and
used as received.
HCl was added to the above solution and warmed slightly. Green
crystals, formed at room temperature after slow evaporation of the
solution over one month, were collected by vacuum filtration and
allowed to air-dry. No attempt was made to optimize the yield.
Anal. calc. for C12H20Cl6N4O2 Cu(528.56/mol): C 27.24%,
H 3.78%, N 10.59%, Cl 40, 30%, Cu 12.01%; found: C 27.12%,
H 3.82%, N 10.53%, Cl 40, 34%, Cu 12.1%.
(H2Me2ppz)[CuCl4] (3). trans-2,5-Dimethylpiperazine (0.228 g,
2.0 mmol) was dissolved in 3.0 mL of 6 M HCl. A solution of
CuCl22H2O (0.170 g, 1.0 mmol) in 2.0 mL of 6 M HCl was
added to the above solution, followed by an additional 1 mL of
6 M HCl. The small amount of precipitate was removed by filtration
and the filtrate was left to evaporate slowly at room temperature.
After nearly one month, light yellow needle-shaped crystals, suitable
for X-ray analysis, were collected by vacuum filtration and allowed to
air-dry. No attempt was made to optimize the yield.
Anal. calc. for C6H16Cl4N2Cu (321.55/mol): C 22.39%, H 4.98%,
N 8.71%, Cl 44.16%, Cu 19.75%; Found: C 22.40%, H 4.92%,
N 8.75%, Cl 44.21%, Cu 19.70%.

Table 1 Crystal data and structure refinement for compounds 2 and 3

Identification code 2 3
Empirical formula C12H20Cl6CuN4O2 C6H16Cl4CuN2
Formula weight 528.56 321.55
Temperature; K 293(2) 293(2)
Wavelength; Å 0.71073 0.71073
Crystal system Triclinic Monoclinic
Space group P1% C2/c
Unit cell dimensions; Å, 1 a = 7.1563(6) a = 14.5594(6)
b = 8.0445(5) b = 14.5888(5)
c = 9.1508(8) c = 5.9318(3)
a = 94.831(6) a = 90
b = 90.917(7) b = 107.295(5)
g = 91.361(6) g = 90
Volume; Å3 524.70(7) 1202.97(9)
Z 1 4
Density (calculated); Mg m3 1.673 1.775
Absorption coefficient; mm1 1.819 2.663
F(000) 267 652
Crystal size; mm 0.516  0.417  0.111 0.551  0.273  0.161
Theta range for data collection 2.234 to 28.3701 2.792 to 28.4451
Index ranges 9 r h r 9, 10 r k r 10, 12 r l r 11 16 r h r 18, 19 r k r 17, 7 r l r 6
Reflections collected 3419 3912
Independent reflections 2329 [R(int) = 0.0167] 1382 [R(int) = 0.0214]
Completeness to theta 25.000 100.0% 25.242 99.8%
Absorption correction Analytical Analytical
Max. and min. transmission 0.817 and 0.464 0.730698 and 0.440975
Refinement method Full-matrix least-squares on F2 Full-matrix least-squares on F2
Data/restraints/parameters 2329/0/115 1382/0/62
Goodness-of-fit on F2 1.083 1.064
Final R indices [I 4 2sigma(I)] R1 = 0.0355, wR2 = 0.0902 R1 = 0.0224, wR2 = 0.0621
R indices (all data) R1 = 0.0483, wR2 = 0.0949 R1 = 0.0279, wR2 = 0.0633
Extinction coefficient n/a 0.0075(5)
Largest diff. peak and hole; e Å3 0.370 and 0.366 0.284 and 0.371

15706 | New J. Chem., 2018, 42, 15705--15713 This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2018
 View Article Online

NJC Paper

Physical measurements Table 2 Selected bond lengths (Å) and angles (1) for 2 and 3

X-ray crystallography. The green crystals of (HMepy)2[CuCl4] 2 3

2H2O (2) and the bright yellow crystals of (H2Me2ppz)[CuCl4] (3)
Cu1–Cl1 2.2600(7) Cu1–Cl3 2.2562(7)
were separated from a 6 M HCl solution. Diffraction data were Cu1–Cl2 2.2737(7) Cu1–Cl4 2.2744(6)
collected on an Oxford Sapphire CCD diffractometer using N1–C2 1.328(3) Cu1–Cl2 2.3623(5)
N1–C6 1.334(3) C2–C3 1.503(2)
MoKa radiation (l = 0.71073 Å) at 293 K by the o–2y method.
C2–C3 1.401(4) C2–C4 1.515(3)
The structures were solved by the Patterson method and refined C2–Cl21 1.700(2) C3–N1#3 1.483(2)
with the full-matrix least-squares method on F2 using the C3–N31 1.350(3) Cl3–Cu1–Cl4 180.0
C3–C4 1.391(3) Cl3–Cu1–Cl2 90.046(12)
SHELX-2014 program package.8 Analytical numeric absorption
Published on 09 August 2018. Downloaded by Uniwersytet Wroclawski on 2/11/2019 3:30:38 PM.

C4–C5 1.383(4) Cl4–Cu1–Cl2 89,955(12)

corrections were made using CrysAlisPro1.171.38.439 based on C5–C6 1.379(4) Cl2–Cu1–Cl2#2 179.91(2)
the equations of Clark and Reid10 (Table 1). Extinction correc- C5–C51 1.496(3) C3#3–N1–C2 111.29(14)
Cl1#1–Cu1–Cl1 180.0 N1–C2–C3 109.58(14)
tion was applied for 3 (Table 1). All hydrogen atoms were
Cl1–Cu1–Cl2#1 89.38(3) N1–C2–C4 110.38(16)
located from the electron density maps and their positions Cl1–Cu1–Cl2 90.62(3) C3–C2–C4 110.80(18)
were constrained in the refinement with an appropriate model Cl2#1–Cu1–Cl2 180.00(4) N1#3–C3–C2 111.74(16)
C2–N1–C6 123.1(2)
as implemented in SHELXL-2014.8
N1–C2–C3 121.0(2)
The structural data for complexes 2 and 3 have been deposited N1–C2–Cl21 117.83(19)
with the Cambridge Crystallographic Data Centre, the deposition C3–C2–Cl21 121.1(2)
N31–C3–C4 122.7(2)
numbers CCDC 1455402 and 1530155, respectively.†
N31–C3–C2 121.8(2)
EPR spectra. X-band EPR spectra of the polycrystalline C4–C3–C2 115.5(2)
complexes were measured at 295 K using a Bruker ELEXSYS C5–C4–C3 122.7(2)
C6–C5–C4 117.9(2)
E500 spectrometer equipped with a NMR teslameter and a
C6–C5–C51 120.9(2)
frequency counter. The experimental spectra were simulated C4–C5–C51 121.1(2)
using the DoubletExact (S = 1/2) computer program.11 N1–C6–C5 119.7(2)
Magnetic measurements. Magnetic data were collected using
a Quantum Design MPMS-XL-5 SQUID magnetometer and cali-
brated with a palladium rod (Materials Research Corporation, are 2.2432(3) Å, 2.2475(3) Å, 2.2521(2) Å and 2.2568(3) Å. The
purity 99.9985%). The superconducting magnet operated at a tetrachlorocuprate(II) ion has a distorted tetrahedral geometry
field from 0 to 5 T. Magnetic susceptibility data were collected with a mean trans angle of 1371.1 Details of the diffraction
over the temperature range of 1.8–300 K and an applied mag- experiments and the crystal structure refinement for complexes
netic field of 0.5 T. Magnetization data were measured in a 2 and 3 are summarized in Table 1 with the corresponding bond
magnetic field from 0 to 50 kOe at 2 K. The data were corrected lengths and angles in Table 2.
for background signals and temperature independent paramag- The asymmetric unit of crystal structure 2, a half of the
netism, calculated from Pascal’s constants.12 chemical unit, (Fig. 2), consists of the 3-amino-2-chloro-5-methyl-
pyridinium cation (HMepy), a half of the tetrachlorocuprate(II)
dianion and one water molecule. The copper(II) ion is located at
Results and discussion the center of symmetry. Therefore, the rest of the chemical unit
Crystal structures is generated by the symmetry operator [x + 1, y + 1, z  1].
The crystal structure of complex 1 has been reported pre- The coordination sphere of Cu(II) has a square planar geo-
viously1,2 and arises from two CuCl4 tetrahedra, which are held metry. The average Cu–Cl distance equals 2.2500 Å, while the
together by hydrogen bonds with two water molecules and with Cl1–Cu1–Cl2 angle is 90.62(3)1. In 1 with H2bipip as the cation,
two 4,4 0 -bipiperidinium cations [Fig. 1]. The Cu–Cl bond lengths the [CuCl4]2 anion has a pseudo-tetrahedral geometry and the
average Cu–Cl distance of 2.2499 Å is shorter.1 The electron

Fig. 2 View of the chemical unit of 2, the symmetry generated atoms are
Fig. 1 The hydrogen bonding network and supported copper(II) chloride indicated by letter B (symmetry code: x + 1, y + 1, z  1). The thermal
chains in 1.1 ellipsoids are plotted at the 30% probability level.

This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2018 New J. Chem., 2018, 42, 15705--15713 | 15707
 View Article Online

Paper NJC

Table 3 Selected hydrogen bonds (Å and 1) bonds, with N  Cl distances of 3.350(3) and 3.374(2) Å, respec-
tively. An intra-cation N31–H2N3  Cl21 interaction is suggested
D–H  A d(D–H) d(H  A) d(D  A) +(DHA)
by a N  Cl distance of 3.007(3) Å, but is not considered to be a
Compound 2 H-bond due to a small N–H  Cl angle of 1041. In this way all
N31–H1  O1W 0.91 1.78 2.677(3) 172.1
N31–H1  Cl2#2 0.98 2.38 3.350(3) 169.7 N–H donors are involved in the H-bonds or polar interactions,
N31–H2  Cl2#3 0.94 2.71 3.374(2) 128.2 similar to the structures of the previously reported salts.2 Due to
N31–H2  Cl21 0.94 2.64 3.007(3) 104 the electrostatic effect of the protonated pyridine in HMepy, the
C4–H4A  Cl1#2 0.93 2.93 3.771(3) 151.1
C6–H6A  O1W#4 0.93 2.50 3.389(3) 161.1 polarization of the C–H groups is enhanced and these groups
O1W–H1W1  Cl1#5 0.80 2.50 3.237(2) 154.2 serve as donors in the C–H  Cl/O interactions. In particular,
Published on 09 August 2018. Downloaded by Uniwersytet Wroclawski on 2/11/2019 3:30:38 PM.

O1W–H2W1  Cl1#1 0.89 2.53 3.292(2) 143.5

the C6–H6A bond adjacent to the protonation site interacts
O1W–H2W1  Cl2 0.89 2.75 3.368(2) 127.2
C51–H51A  Cl21#6 0.96 2.83 3.386(3) 118 with O1(water) [x, y + 1, z] and C4–H4A interacts with Cl1
[x + 1, y + 2, z] with C  O and C  Cl distances of 3.389(3)
Compound 3 and 3.771(3) Å, respectively. In turn, the pyridinium-methyl
N1–H1  Cl2#6 0.89 2.42 3.2115(17) 148
N1–H1  Cl4#7 0.89 2.75 3.1952(17) 113 group of HMepy participates in the chain of C51–H51A  Cl21
N1–H2  Cl2 0.89 2.32 3.2000(17) 172 [x, y, 1 + z] interactions with a C  Cl distance of 3.386(3) Å.
C3–H3B  Cl4#8 0.97 2.77 3.412(2) 124
A water molecule is a donor in the O1(water)–H1(water1)  Cl1
Symmetry transformations used to generate equivalent atoms: #1 x + 1, [x  1, y, z] bond with a O  Cl distance of 3.237(2) Å and in a
y + 1, z  1, #2 x + 1, y + 2, z, #3 x, y + 1, z, #4 x, y + 1, z, bifurcated H-bond with the donor O1(water)–H2(water1) and
#5 x  1, y, z, #6 x, y, 1 + z, #7 x, y, z, #8 1/2 + x, 1/2 + y, z.
acceptors Cl1 [x + 1, y + 1, z  1] and Cl2, with the
respective O  Cl distances being 3.292(2) and 3.368(2) Å. As a
density maps for 2 demonstrate that protonation occurs at the consequence, the layers of the HMepy cations and the [CuCl4]2
pyridine N1 atom of 3-amino-2-chloro-5-methylpyridine. anions are bridged by the H-bonds involving water molecules,
The geometry of the resulting HMepy cation is typical of as presented in Fig. 3. In complex 3, the asymmetric part of the
such structures. The C2–Cl21 distance is 1.700(2) Å and the structure consists of half of the trans-2,5-dimethylpiperazinium
C3–N31 distance is 1.350(3) Å. A comparison of the N1–C2–Cl21 dication H2Me2ppz and half of the [CuCl4]2 anion (Fig. 4) with
angle of 117.83(19)1 with the observed C3–C2–Cl21 angle of Cu1, Cl3 and Cl4 positioned on the two-fold symmetry axis. As a
121.1(2)1 clearly reveals the effect of the bulk of the N31 amino- consequence, the rest of the chemical unit is generated by the
group on the valence geometry around C2. A p–p interaction is [x, y, z + 1/2] operator.
found between the N1–C6 pyridine ring and the [x, 2  y, z] We have previously reported two Cu(I) and Cu(II) complexes
symmetry equivalent, with the distance between the ring containing unsubstituted piperazinium dications and trigonal
centers of gravity equal to 3.6246(14) Å (Table 3 and Fig. 3). [CuCl3]2 or tetrahedral [CuCl4]2 anions, respectively.1 First, a
It should be mentioned that the protonation of the N1 of linear arrangement of the anions occurs, and second, a three
3-amino-2-chloro-5-methylpyridine results in the formation of dimensional network is formed as a result of hydrogen bonds,
3D charge assisted H-bonds involving the N–H and C–H groups where the columns of the [CuCl4]2 anions are interspersed with
as donors. the columns of the piperazinium molecules. In the [CuCl4]2
Both chloride atoms, Cl1 and Cl2, are acceptors in each of the anion of 3, the Cu1–Cl2 bond of 2.3623(5) Å is significantly longer
three H-bonds. In turn, the O atom is an acceptor in one NH  O than those for Cu1–Cl3 (2.2562(7) Å) and Cu1–Cl4 (2.2744(6) Å),
and one C–H  O interaction. The pyridinium N1–H1A in HMepy is indicating that their average value of 2.3138 Å is greater than that
a donor in the N  O(water) H-bond, with the N  O distance being for the complex with unsubstituted piperazinium and [CuCl4]2
2.677(3) Å. The N31 amino-group is a donor in the N31-H1N3  Cl2 anions1 with a distorted tetrahedral geometry (2.1955 Å). Since
[x + 1, y + 2, z] and N31–H2N3  Cl2 [x, y + 1, z] hydrogen the crystallographic two-fold axis generates the equivalent of

Fig. 3 Drawing of the coordination environment of 2. A 3D network of Fig. 4 View of the chemical unit of 3, the symmetry generated atoms
the layers of the HMepy cations and [CuCl4]2 anions is presented. Each of are indicated by letter A (symmetry code: x, y, z + 1/2). The thermal
the bridging water molecules forms three hydrogen bonds. ellipsoids are plotted at the 30% probability level.

15708 | New J. Chem., 2018, 42, 15705--15713 This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2018

NJC
Cl2 [x, y, z + 1/2] bound to the same Cu1, the pair of Cl2 and
its twofold related Cl2a (Fig. 4) complete the planar part of the
Cu(II) environment in 3. The Cl3–Cu1–Cl4 angle is 1801 and the
Cl2–Cu1–Cl3 and Cl2–Cu1–Cl4 angles are 90.045(12) and
89.955(12)1, respectively. Two significantly longer axial Cu1–Cl4
[x, y, z] and Cu1–Cl4 [x, y, 1  z] bonds, both being
2.9820(2) Å in length, complete the octahedral environment
around Cu1. Infinite dibridged polymeric chains are formed
along the c-axis by Cu1–Cl4 interactions (Fig. 5). The distance
Published on 09 August 2018. Downloaded by Uniwersytet Wroclawski on 2/11/2019 3:30:38 PM.
between the adjacent copper atoms is 3.9332 Å. Similar chain
structures were observed in the crystals of ternary Cu(II) com-
plexes with pairs of trans-coordinated chlorides and pyridine
derivatives.13,14 For Cu(2-amino-3,5-diX)dichlorocopper(II),
[X = Cl, Br]13, the observed geometry is planar and stabilized
by different hydrogen bonds, including those with the pyridine
halogen substituents. One Cl ion of each [CuCl4]2 anion plays
a role of an axial ligand and the infinite chains are formed
between the adjacent [CuCl4]2 anions. This coordination around
Cu(II) is very close to that for 3. To our knowledge, the crystal
structure of 3, composed of the chains of the dibridged [CuCl4]2
anions, is a very rare case and may be the only known example of
this motif. More common are the linear14,15 or two-dimensional
chains16 of monobridged [CuCl4]2 anions. Although in the bis(2-
amino-5-chloro-3-fluoropyridinium)tetrachlorocuprate(II) com-
plex the Cu–Cl  Cl–Cu contacts between the monobridged
planar [CuCl4]2 anions forming the linear chain are relatively
short of 3.965 Å, the Cu–Cu separation is greater than 8 Å, leading
to a very weak antiferromagnetic coupling with J = 0.74 cm1
and negligible magnetic intermolecular interactions.14 The
Fig. 5 (a) The interactions in the chain of the dibridged [CuCl4]2 anions
in 3. (b) The interactions between the H2Me2ppz dications and the [CuCl4]2
anions together with the formed 3D network of hydrogen bonds and direct
connections between the two [CuCl4]2 anions along the a-axis.
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2018
View Article Online
Paper
Cu1–Cu1 [x, y, 1  z] distance in the four-membered ring
in 3 is 3.9333(3) Å. The charge-assisted H-bonds involving all
N–H donors form an extensive network (Table 3). An infinite
H-bond chain along the c-axis consists of the piperazinium
moiety of the N1–H1N1 chain as a donor in the bifurcated
H-bonds to Cl2 [x, y, 1 + z] and Cl4 [x, y, z] with N  Cl
distances of 3.2115(17) and 3.1952(17) Å, respectively. In turn,
N1–H2N1 participates in an H-bond to Cl2, with a N  Cl
distance of 3.2000(17) Å. The C3–H3B H-bonds to Cl4 [1/2 + x,
1/2 + y, z] are linked with a C  Cl distance of 3.412(2) Å.
Similar to 2 and 1, a wider network of charge-assisted
hydrogen bonds is observed (Table 3 and Fig. 5), involving all
the N–H donors. The infinite hydrogen bond chain along the
c-axis is seen. The piperazinium N1–H1N1 group of H2Me2ppz
is the donor of hydrogen bond, bifurcated to Cl2 [x, y, 1 + z]
and Cl4 [x, y, z], with N  Cl bond distances of 3.2115(17)
and 3.1952(17) Å, respectively. The distance of the N1–H2N1–Cl2
bond is 3.2000(17) Å and the distance of the C3–H3B–Cl4 bond
[1/2 + x, 1/2 + y, z] is 3.412(2) Å.
It is noteworthy that the values of the average HN  Cl hydro-
gen bond distances, formed between [CuCl4]2 anions and nitro-
gen donors from the organic cations, change from 3.3409 Å1 to
3.2437 Å, and 3.2022 Å, in 1, and 2 and 3, respectively, corres-
ponding to a distorted tetrahedral geometry for 1 and a planar
geometry for 2 and 3. These experimental data confirm that the
shorter hydrogen HN  Cl bonds that result from removing a
charge from the Cl reduces electrostatic repulsions between the
[CuCl4]2 anions, favoring a planar geometry in these anions.3
EPR study
The experimental EPR spectra of polycrystalline complexes 1–3
shown in Fig. 6 correspond to the local symmetry of the Cu(II)
coordination sphere in the [CuCl4]2 close to axial and an
unpaired electron in the dx2y2 orbital. Simulation of these
spectra allowed us to determine the exact values of the g-tensor
parameters gz = g8, gx = gy = g>, a measure of the interaction
between the magnetic moment of the unpaired electron of the
Fig. 6 The EPR spectra for 1–3 at 295 K together with the spectra
simulated (1 sim–3 sim) using the parameters given in Table 4. The stars
show the resonance transitions of crystallites differently oriented towards
the magnetic field direction.
New J. Chem., 2018, 42, 15705--15713 | 15709
 View Article Online

Paper NJC

Table 4 Relationship between the EPR parameters and the geometry of different Cu(II) complexes

EPR parameters
Compounds gx = gy = g>, gz = g8 gav Dg = g8  g> Averaged cis-angles of [CuCl4]2 Ref.
(HMepy)2[CuCl4]2H2O (2) 2.045, 2.238 2.109 0.185 90.00 This work
K2[Pd(Cu)Cl4] 2.049, 2.236 2.110 0.187 90.00 17
(H2Me2ppz) [CuCl4] (3) 2.060, 2.302 2.140 0.243 90.00 This work
[H52,6Acgu2] [CuCl4]ClH2O gx = 2.092, gy = 2.159, gz = 2.239 2.163 97.26 18
(H2bipip)[CuCl4]H2O (1) 2.067, 2.372 2.160 0.305 97.67 This work
(Me3PhN)2[CuCl4] 2.067, 2.374 2.170 0.307 98.20 4
(EtPh3P)2[CuCl4] 2.076, 2.386 2.179 0.310 97.80 4
Published on 09 August 2018. Downloaded by Uniwersytet Wroclawski on 2/11/2019 3:30:38 PM.

gav = 1/3(2g> + g8), Dg = g8  g>.

Cu(II) ion in the [CuCl4]2 anions resulting from different describing the isotropic magnetic exchange interaction, anti-
organic cations. A relationship exists between the g-tensor ferromagnetic for J o 0 and exchange interaction zJ 0 (where z is
anisotropy of the new compounds and those reported earlier, the number of adjacent binuclear or paramagnetic species
which is expressed by g8, or gav = 1/3(g8 + 2g>) or Dg = g8  g>, around a given binuclear unit), according to the molecular field
and the average cis-angles of different [CuCl4]2 anions (Table 4). approximation (MFC).22 The experimental magnetic data were
This result is in agreement with that observed previously.4 An fitted according to Bleaney–Bowers (eqn (2))23 modified eqn (3),
increase in the cis-angle causes a greater anisotropy of the
H = JSASB  zJ 0 hSziSz (1)
g-tensor. This characteristic tendency is observed for com-
  1
pounds 1 and 2. The smallest g-tensor anisotropy for complex Nb2 g2 1 J
wM ¼ 1 þ exp (2)
2 (g8 = 2.238, gav = 2.109 and Dg = 0.185) corresponds to the 3kT 3 kT
planar geometry of the anion with an average cis-angle of 901
similar to that determined previously for the compound also wM
wcorr
M ¼ (3)
containing planar [CuCl4]2 anions17, which is only the second zJ 0 w
1  2 M2
known example with a square planar geometry, next to our com- Ng b
plex 2, examined by EPR.17 The distorted tetrahedral [CuCl4]2 The temperature independent paramagnetism (TIP) was included
anion in 1 with an average cis-angle of 97.671 is characterized by in the fitting procedure.
g8 = 2.238, gav = 2.167 and Dg = 0.305, very close to those The fitting of the magnetic data based on the HDVV spin
reported for other compounds containing the anions of similar Hamiltonian was also carried out by using the PHI program.24
geometry.17–19 The observed increase of g8 and gav upon change In addition, the presence of different magnetic pathways in the
of the geometry from planar to tetrahedral is in agreement with crystal structures of the complexes prompted us to analyze the
a decrease in the crystal field energy pronounced in the red temperature dependence of the magnetic susceptibility of 2
shift of maximum in the band caused by the d–d (DEdd)3 using the Heisenberg ladder model.25,26 In 3, the [CuCl4]2
transition, which consequently increases the g-tensor compo- anions are arranged in linear chains due to the long Cu–Cl4
nents as their deviation from ge = 2.0023 (for free electron) is axial bonds (Fig. 5). A simplified Heisenberg–Ising model, corres-
inversely proportional to DEdd. ponding to the Hamiltonian (4) and the fitting of the magnetic
The slightly higher values of g8 = 2.302, gav = 2.140 and data according to eqn (5) was used.27
Dg = 0.2430 for 3 with a cis-angle of 901 may be associated with
X
n1
the interaction of the Cu(II) ion in the planar [CuCl4]2 anion H ¼ J SAi  SAiþ1 (4)
with the Cl4 ion at the axial positions from the neighboring i¼1
parallel [CuCl4]2 anion at a distance of 2.9820(2) Å. An addi-
tional binding to Cu(II) along the z-axis results in a weaker Ng2 b2 0:25 þ 0:0749755x þ 0:075235x2
wM ¼ with
in-plane Cu–Cl interaction than that in the strictly planar Cu(II) kT 1:0 þ 0:9931 þ 0:172135x2 þ 0:757825x2
coordination.20,21 This effect is supported by the geometrical
x ¼ jj=kTj (5)
parameters of 3 showing that its in-plane average Cu–Cl distance
(2.3138 Å) is apparently greater than that in 2 (2.2668 Å). It seemed necessary to confirm the magnetic properties of
complex 1, which are needed for comparison with those of
Magnetic properties complexes 2 and 3. The synthesis of 1 and the measurement
The exchange interaction between the nearest two copper(II) ions of temperature dependence of the magnetic susceptibility were
(SA = SB = 1/2) in the crystal lattice was approached first to the repeated by us to perform the best match of the experimental
model of binuclear units realized through Cu–Cl  OW  Cl–Cu data by obtaining the parameter g that is most similar to
bridges in 1 and 2 and Cu–Cl–Cu–Cl bridges in 3 together gav = 2.167 derived for the first time by the analysis of the
with different additional Cu–Cl  Cl–Cu contacts in all com- EPR spectra. The calculated values of J and J 0 appeared to be
pounds. The calculations were based on the spin Hamiltonian about 1 cm1 lower and 2 cm1 greater, respectively, than those
(Heisenberg–Dirac–Van Vleck) in the zero field given by eqn (1) obtained previously using g = 2.02 (Table 5 and Fig. S1, ESI†).

15710 | New J. Chem., 2018, 42, 15705--15713 This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2018
 View Article Online

NJC Paper

Table 5 Magnetic data parameters fitted for 1

Model J (cm1) zJ 0 (cm1) g g(EPR)

1. Heisenberg–Dirac–Van Vleck/Bleaney–Bowers with MFC 7.49 3.06 2.170 2.167
8.51 1.35 2.00
2. Two-leg Heisenberg spin-ladder 8.32 3.32 2.165 2.167
9.54 1.97 2.00
3. Heisenberg–Dirac–VanVleck–PHI program 8.48 4.98 2.175 2.167
0 2 1
The J and zJ values in K are converted to cm .
Published on 09 August 2018. Downloaded by Uniwersytet Wroclawski on 2/11/2019 3:30:38 PM.

It was shown that 1 consists of the dimers formed by the

Cu–Cl  Ow  Cl–Cu bridge with a Cu  Cu distance of 5.3372 Å.
Furthermore there are intradimer Cu–Cl  Cl–Cu contacts of
which the Cl1  Cl1A along the c-axis is the shortest with
d = 3.695 Å and the angles very close to 1801. Among these two
pathways the two-chloride contacts were considered responsible
for the strongest exchange interactions with J = 7.49 cm1. The
interdimer chloride–chloride contacts between Cu1 and Cu1B
and Cu1 and Cu1D, measured by J 0 are at comparable distances
to the intradimer ones but with much smaller angles.2
The magnetic susceptibility data for complexes 2 and 3 were
collected over the temperature range of 1.8–300 K under a
constant magnetic field of 0.5 T. The temperature dependence
of the wM, wMT and 1/wMT products for 2 and 3 are shown in Fig. 8 Thermal dependence of wMT (J) and wM (Q) for 3. The inset shows
Fig. 7 and 8, respectively. At room temperature, the values of the thermal dependence of the inverse magnetic susceptibility. The solid
wMT of 0.43 cm3 mol1 K (1.85 mB) for 2 and 0.45 cm3 mol1 K black line is calculated by using the HDVV spin Hamiltonian and the
Bleaney–Bowers equation, the red line by the Heisenberg–Ising chain
(1.90 mB) for 3 are slightly higher than those expected for the
model, the green line by the HDVV spin Hamiltonian and the PHI
[CuCl4]2 anions in 2 and 3 without any exchange interactions; program.24
meff = 1.83 mB for S = 1/2 and gav = 2.11 taken from EPR for 2 and
gav = 2.14 from EPR (meff = 1.85 mB) for 3. Upon lowering the
temperature, wMT decreases very slowly, followed by a rapid and 6.15 K for 3, confirming an antiferromagnetic exchange
decrease up to 0.18 cm3 mol1 K (meff = 1.20 mB) for 2 and coupling. Inspection of the selected crystal packing character-
0.34 cm3 mol1 K (meff = 1.66 mB) for 3 at 1.8 K, which indicates an istics of complex 2 suggests that the weak magnetic exchange
antiferromagnetic exchange coupling. The measured magnetic coupling may be transmitted through the Cu–Cl  Ow  Cl–Cu
susceptibility data in the temperature range of 30–300 K follows bridge with alternating Cu–Cu separations of 7.1563 and
the Curie–Weiss law with a Weiss constant of y = 4.81 K for 2 12.6827 Å, respectively (Fig. S2, ESI†). The exchange interaction
may be realized also through the nonbonding Cu1–Cl1  Cl1A–
Cu1A contact with distance d = 4.2442 Å and an optimal t angle
of 1801, comparable to the Cu1B–Cl3B  Cl2C–Cu1C contact
with both t and Y angles closer to 1801 but with d being longer
(4.6372 Å), (Table 6, red line in Fig. S2, ESI†). In addition the

Table 6 Selected bond lengths (Å) and angles (1) in 2 and 3 of the crystal
packing

Chloride contact d (Å) Y1 (1) Y2 (1) t (1)

Complex 2
Cu1–Cl1  Cl2A–Cu1A 5.8950 126.2 68.9 107.2
Cu1–Cl1  Cl1A–Cu1A 4.2442 106.3 106.3 180.0
Cu1A–Cl3A  Cl2B–Cu1B 10.333 105.9 131.1 76.4
Cu1–Cl1  Cl2D–Cu1D 7.3420 108.9 72.4 95.6
Cu1–Cl3  Cl4D–Cu1D 7.3420 108.9 72.4 95.6
Fig. 7 Thermal dependence of wMT (J) and wM (Q) for 2. The inset shows
Cu1B–Cl3B  Cl2C–Cu1C 4.6372 170.0 170.0 180.0
the thermal dependence of the inverse magnetic susceptibility. The solid
Complex 3
red line is calculated by using the HDVV spin Hamiltonian and the Cu1–Cl3  Cl2C–Cu1C 5.2610 117.8 150.9 17.6
Bleaney–Bowers equation, the blue line by the two-leg Heisenberg Cu1–Cl3  Cl2B–Cu1B 7.3365 109.6 127.8 73.8
spin-ladder model, and the green line by the HDVV spin Hamiltonian Cu1C–Cl3C  Cl3D–Cu1D 8.0571 158.4 158.4 180.0
and the PHI program.24 Cu1B–Cl3B  Cl3C–Cu1C 10.180 153.4 153.4 179.6

This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2018 New J. Chem., 2018, 42, 15705--15713 | 15711

Paper
Table 7 Magnetic data parameters fitted for 2
Model
1. Heisenberg–Dirac–Van Vleck/Bleaney–Bowers with MFC
2. Two-leg Heisenberg spin-ladder
3. Heisenberg–Dirac–VanVleck–PHI program
magnetic interactions are strengthened by p  p stacking
Published on 09 August 2018. Downloaded by Uniwersytet Wroclawski on 2/11/2019 3:30:38 PM.
between the pyridine rings.
The best agreement with the experimental magnetic data for
2 was obtained using the Bleaney–Bowers eqn (2) and (3) with
J = 1.59 cm1, zJ 0 = 0.40 cm1, g = 2.11 and TIP = 195 
P P
106 cm3 mol1, R = [(wT)exp  (wT)calc]2/ [(wT)exp]2 = 6.81 
105 (red lines in Fig. 7). The g-values of 2.12 determined by the
analysis of the magnetic susceptibility data are close to those
obtained by the simulation of the EPR spectra (Table 7). The
experimental data were analyzed additionally using the PHI pro-
gram (green line in Fig. 7) based on single J isotropic HDVV. The
magnetic susceptibility dependence was determined also by the
Heisenberg ladder model (blue line, Fig. 7) with the parameters
supporting the antiferromagnetic exchange interactions along the
legs and rungs, with the constants J|| and J>, respectively, corres-
ponding to transmission through the Cu–Cl1  Cl1A–Cu1A (legs)
and Cu1B–Cl3B  Cl2C–Cu1C (rungs) shown in Fig. S3 (ESI†). The
fitting results in J8 = 1.52 cm1, J> = 0.62 cm1, g = 2.12 and
P
TIP = 196  106 cm3 mol1, and R = [(wT)exp  (wT)calc]2/
2 5
[(wT)exp] = 2.9  10 . The simultaneously obtained value of zJ0 =
0.58 cm1 indicates that the exchange chloride–chloride contacts
Fig. 9 Magnetization as a function of magnetic field at 2.00 K for com-
plexes 1 (K), 2 (K) and 3 (K). The solid line is the magnetization calculated
for the isolated Cu(II) complexes with J = 8.48 cm1, g = 2.175 for 1,
J = 1.93 cm1, and g = 2.12 for 2 and J = 2.80 cm1, and g = 2.155 for 3.
The dashed line is the Brillouin function for S = 1/2 and g = 2.0 for 1–3.
Table 8 Magnetic data parameters fitted for 3
Model
1. Heisenberg–Dirac–Van Vleck/Bleaney–Bowers with MFC
2. Heisenberg–Dirac–Van Vleck/PHI program with MFC
4. Heisenberg–Ising for chain structure
15712 | New J. Chem., 2018, 42, 15705--15713 This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2018
View Article Online
NJC
J (cm1) zJ 0 (cm1) g g(EPR)
1.59 0.40 2.11 2.109
1.52 0.58 2.12 2.109
0.62
1.93 1.18 2.12 2.109
between the legs of the neighboring ladders are comparable to
those through the chloride contacts (rungs) with the optimal
angles of 1801 and water bridges additionally transmitting mag-
netic interaction. Other chloride contacts with much longer Cl  Cl
distances and suboptimal angles are shown in Table 6.
For complex 2, magnetization at 2 K reaches saturation at a
field of 50 000 Oe with a value of 0.79 mB but smaller than the
value of 1.0 mB for the paramagnetic Cu(II) ion with the S = 1/2
ground state (Fig. 9). The measured and fitted magnetization
curves do not follow the curve of the Brillouin function.
Since the field dependent magnetization cannot be fitted to
the spin ladder model, we decided to ignore the weak magnetic
interactions along the shoulder of the ladder (JII), considering
the system as a set of isolated Cu(II) dimers, while leaving the
rest of parameters unchanged.
The best fit of the magnetic data for 3, based on the
modified Bleaney–Bowers eqn (2) and (3), was achieved with
J = 1.85 cm1, zJ 0 = 0.56 cm1, g = 2.133, TIP = 119 
106 cm3 mol1 and R = 2.74  104 (black and green lines in
Fig. 8). This model only partly reveals the structural properties
of 3. The magnetic data were also analyzed according to the
isotropic HDVV spin Hamiltonian model using the PHI program
(green line in Fig. 7 and 8). The model expected to be suitable
for the infinite chains, according to the Heisenberg–Ising
Hamiltonian eqn (5), allowed the best agreement with the experi-
mental magnetic data for gav(Cu) = 2.152 and J = 1.68 cm1,
zJ 0 = 0.25 cm1, g = 2.173 and TIP = 89  106 cm3 mol1. The
discrepancy factor is 1.37  104. The parameters are close to
those obtained by using the other models (Table 8).
As shown before, a similar chain structure was observed for
Cu(2-amino-3,5-di-X)dichloro cuprate(II), [X = Cl, Br]13 with
almost the same long axial Cu–Cl bonds and Cu–Cu distances
but the exchange interactions are greater (J = 3.10 cm1 and
3.37 cm1, using Cl or Br, respectively) most likely due to the
additional contacts between the halogen atoms, substituted on
the pyridines and the Cl ions, bound to copper(II).
A weak interdimer antiferromagnetic interaction with zJ0 values
from 0.3 cm1 to 0.13 cm1 in 3 can be attributed to the
presence of the superexchange pathway, mediated through the
Cu–Cl  Cl–Cu contact with a rather long Cl  Cl distance (Table 6).
The magnetic superexchange pathway in 2 is formed in a similar
J (cm1) zJ 0 (cm1) g g(EPR)
1.85 0.56 2.133 2.140
2.80 0.28 2.155 2.140
1.68 0.25 2.173 2.140

NJC
way. It is noteworthy that taking into account the long bridge
Cl  OW  Cl between Cu(II) and the long distances of the Cl  Cl
contacts, the exchange interactions should be weaker than those for
3 with the double-chloride bridge and shorter Cl–Cl contacts. This
suggests that the magnetic exchange in 2 is efficiently strengthened
by p–p stacking between the substituted pyridinium rings.
Conclusions
Published on 09 August 2018. Downloaded by Uniwersytet Wroclawski on 2/11/2019 3:30:38 PM.
A significant effect of the organic cations: trans-2,5-dimethyl-
piperazinium, 3-amino-2-chloro-5-methylpyridinium, and 4,4 0 -
bipiperidinium on the structural and magnetic properties of
the [CuCl4]2 anions was revealed in their cationic–anionic
complexes, 2 and 3 (new) and 1 (studied earlier1,2), respectively.
The EPR spectral studies indicated a characteristic decrease of
the anisotropy of the axial g-tensor components (from Dg = g8 
g>= 0.305 for 1 to 0.243 for 3 and 0.185 for 2) with the change of
the geometry of the [CuCl4]2 anions from pseudo-tetrahedral
for 1 to elongated octahedral for 3, where the anion by itself is
square planar (see Fig. 6) and square planar for 2, respectively.
The magnetic exchange interactions between the [CuCl4]2
anions in the complexes evidently resulted from the crystal
packing structure dependent on the organic cations and water
of crystallization (W), which modified the distances in (a) the
Cu–Cl  Cl–Cu contacts between the [CuCl4]2 anions, (b) the
Cu–Cl  W  Cl–Cu bridges in 1 and 2, (c) the dichloride bridge in
the dinuclear units formed by the [CuCl4]2 anions and the adjacent
Cl4 atoms in the chain structure of 3, (d) p–p stacking in 2. In 2 and
3, all these magnetic exchange pathways occurred on relatively long
distances, leading to a weak and comparable antiferromagnetic
exchange; their intra- and intermolecular interactions are similar,
corresponding to J = 1.6 cm1, J0 = 0.5 cm1 and J = 1.8 cm1,
J0 = 0.4 cm1, respectively, where H = JSASB  zJ0 hSziSz. For 1, the
closer Cu–Cl  Cl–Cu contacts cause the stronger exchange inter-
actions in agreement with J = 8.48 cm1 and J0 = 3.32 cm1.
Conflicts of interest
There are no conflicts to declare.
Acknowledgements
We would like to thank the Wroclaw Centre of Biotechnology
for support with the EPR and magnetic measurements within
the program of the National Research Centre (KNOW) for years
2014–2018, and also Dr Rein Kirss of the Department of
Chemistry, Northeastern University, Boston, MA (USA) for his
help in reading the manuscript and for useful comments.
Notes and references
1 M. Tuikka, Ü. Kersen and M. Haukka, CrystEngComm, 2013,
15, 6177.
2 L. Wikaira, R. J. Butcher, Ü. Kersen and M. M. Turnbull,
J. Coord. Chem., 2016, 69(1), 57.
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2018
View Article Online
Paper
3 K. E. Halvorson, C. Patterson and R. D. Willett, Acta Crystal-
logr., Sect. B: Struct. Sci., 1990, B46, 508.
4 K. Winter, A. Thiel, T. Zabel, A. Klamroth, A. Pōppl,
U. Kelling, A. Schilde, P. Taubert and P. Strauch, New
J. Chem., 2014, 38, 1019.
5 M. M. Turnbull, Ch. P. Landee and B. M. Wells, Coord.
Chem. Rev., 2005, 249, 2567–2576.
6 L. O. Snively, G. F. Tuthill and J. E. Drumheller, Phys. Rev. B:
Condens. Matter Mater. Phys., 1981, 24, 5349–5355.
7 K. Halvorson and R. D. Willett, Acta Crystallogr., Sect. C:
Cryst. Struct. Commun., 1988, C44, 2071.
8 G. M. Sheldrick, Acta Crystallogr., Sect. A: Found. Crystallogr.,
2008, A64, 112.
9 CrysAlisPro 1. 171.38.43, Rigaku Oxford Diffraction, Yarnton,
England, 2015.
10 C. Clark and J. S. Reid, Acta Crystallogr., Sect. A: Found.
Crystallogr., 1995, 51, 887–897.
11 G. A. Bain and J. F. Berry, J. Chem. Educ., 2008, 5, 532–536.
12 The program DoubletExact written by Dr Andrew Ozarowski
from NHMFL, Florida University, Tallahassee, USA.
13 K. C. Shortsleeves, L. N. Dawe, Ch. P. Landee and M. M.
Turnbull, Inorg. Chim. Acta, 2009, 362, 1859–1866.
14 B. L. Solomon, C. P. Landee, M. M. Turnbull and J. L.
Wikaira, J. Coord. Chem., 2014, 67, 3953–3971.
15 S. N. Herringer, M. M. Turnbull, C. P. Landee and J. L.
Wikaira, J. Coord. Chem., 2009, 62, 863–875.
16 P. R. Hammar, D. C. Dender, D. H. Reich, A. S. Albrecht and
C. P. Landee, J. Appl. Phys., 1997, 81, 4615–4617.
17 C. Chow, K. Chang and R. D. Willett, J. Chem. Phys., 1973,
59, 2629.
18 J. Valdes-Martinez, J. H. Alstrum-Acevedo, R. A. Toscano,
S. Hernandez-Ortega, G. Espinosa-Perez, D. X. West and
B. Helfrich, Polyhedron, 2002, 21, 409–416.
19 A. Winter, K. Thiel, A. Zabel, T. Klamroth, A. Pōppl,
A. Kelling, U. Schilde, A. Taubert and P. Strauch, New
J. Chem., 2014, 38, 1019.
20 G. Tabbı̀, A. Giuffrida and R. P. Bonomo, J. Inorg. Biochem.,
2013, 128, 137–145.
21 B. A. Goodman and J. B. Raynor, in Adv. Inorg. Chem.
Radiochem., ed. H. J. Emeléus and A. G. Sharpe, Academic
Press, New York, 1970, vol. 13, pp. 135–362.
22 O. Kahn, Molecular Magnetism, VCH, Weinheim, 1993.
23 B. Bleaney and K. D. Bowers, Proc. R. Soc. A, 1952,
214, 451.
24 N. F. Chilton, R. P. Anderson, L. D. Turner, A. Soncini and
K. S. Murray, J. Comput. Chem., 2013, 34, 1164–1175.
25 G. Chaboussant, P. A. Crowell, L. P. Lévy, O. Piovesana,
A. Madouri and D. Mailly, Phys. Rev. B: Condens. Matter
Mater. Phys., 1997, 55, 3046.
26 D. C. Johnston, M. Troyer, S. Miyahara, D. Lidsky, K. Ueda,
M. Azuma, Z. Hiroi, M. Takano, M. Isobe, Y. Ueda, M. A.
Korotin, V. I. Anisimov, A. V. Mahajan and J. L. Miller,
eqn (14), arXiv:cond-mat/0001147, 2000.
27 D. B. Brown, J. A. Donner, J. W. Hall, S. R. Wilson,
R. B. Wilson, D. J. Hodgson and W. E. Hatfield, Inorg.
Chem., 1979, 18, 2635–2641.
New J. Chem., 2018, 42, 15705--15713 | 15713