End-to-end azide-bridged manganese(III) chain compounds: Field-induced magnetic phase transitions and variation of T C to 38 K depending on the side groups of the schiff bases

Article abstract of DOI:10.1021/ic2013232

Three one-dimensional coordination polymers [Mn(L)(N₃)] _n [L = L1 (1), L2 (2), L3 (3); L1H₂ = N,N′-bis(5- chlorosalicylideneiminato)-1,3-diaminopentane, L2H₂ = N,N′-bis(5-bromosalicylideneiminato)-1,3-diaminopentane, L3H₂ = N,N′-bis(5-bromosalicylideneiminato)-1,3-diamino-2-dimethylpropane] bridged by end-to-end azides were prepared. The crystal systems differ according to the Schiff bases used. Each Mn atom adopts a typical Jahn-Teller distortion. The helicity of the chains occurs in a racemic manner only for 2. No noncovalent forces are relevant in 2, while π-π contacts are visible in 1 and 3. Magnetic measurements show the presence of apparent spin canting. Complexes 1 and 3 exhibit a field-induced metamagnetic transition from an antiferromagnetic state to a weak ferromagnetic phase, whereas 2 embraces a field-induced two-step magnetic phase transition. The critical temperature is observed at 38 K for 2, which is relatively higher than those for 1 (11 K) and 3 (10 K). The pronounced long-range order may contribute from intrachain exchange couplings and through-space dipolar interactions between adjacent chains.

Full text of DOI:10.1021/ic2013232

ARTICLE
pubs.acs.org/IC
End-to-End Azide-Bridged Manganese(III) Chain Compounds:
Field-Induced Magnetic Phase Transitions and Variation of TC to 38 K
Depending on the Side Groups of the Schiff Bases
Jung Hee Yoon,† Woo Ram Lee,† Dae Won Ryu,† Jin Wuk Lee,† Sung Won Yoon,‡ Byoung Jin Suh,‡
Hyoung Chan Kim,§ and Chang Seop Hong*,†
†Department of Chemistry, Research Institute for Natural Sciences, Korea University, Seoul 136-713, Korea
‡Department of Physics, The Catholic University of Korea, Buchon 420-743, Korea
§National Fusion Research Institute, Daejeon 305-333, Korea
S Supporting Information
b
ABSTRACT: Three one-dimensional coordination polymers [Mn(L)(N3)]n
[L=L1(1),L2(2), L3(3);L1H2 =N,N0-bis(5-chlorosalicylideneiminato)-1,
3-diaminopentane, L2H2 = N,N0-bis(5-bromosalicylideneiminato)-1,
3-diaminopentane, L3H2 = N,N0-bis(5-bromosalicylideneiminato)-1,
3-diamino-2-dimethylpropane] bridged by end-to-end azides were pre-
pared. The crystal systems differ according to the Schiff bases used. Each
Mn atom adopts a typical JahnꢀTeller distortion. The helicity of the
chains occurs in a racemic manner only for 2. No noncovalent forces are
relevant in 2, while πꢀπ contacts are visible in 1 and 3. Magnetic
measurements show the presence of apparent spin canting. Complexes 1
and 3 exhibit a field-induced metamagnetic transition from an antiferro-
magnetic state to a weak ferromagnetic phase, whereas 2 embraces a field-induced two-step magnetic phase transition. The critical
temperature is observed at 38 K for 2, which is relatively higher than those for 1 (11 K) and 3 (10 K). The pronounced long-range
order may contribute from intrachain exchange couplings and through-space dipolar interactions between adjacent chains.
’ INTRODUCTION
From a magnetic point of view, end-on azide generally provides
ferromagnetic interaction and end-to-end azide facilitates anti-
ferromagnetic coupling, although there are several exceptional
cases.14ꢀ16 Many structures with the azide ligand have been
examined using paramagnetic transition-metal ions (Mn2+, Mn3+,
Fe2+, Co2+, Ni2+, Cu2+, etc.),17ꢀ23 and the magnetic properties
relying on the metal contents were revealed. Among them, the
Co2+ ion shows strong magnetic anisotropy to give intriguing
magnetic characteristics of spin canting, metamagnetism, slow
magnetic relaxation, and long-range order.18,24ꢀ28 Similarly,
the MnIII ion, whose oxidation state is stabilized in the N2O2
environment of tetradentate Schiff bases, allows for significant
magnetic anisotropy in association with typical JahnꢀTeller
distortion.29 The use of the Schiff bases in an equatorial coor-
dination makes two axial sites available for incoming azide
ligands. Accordingly, the self-assembly of the MnIII Schiff bases
with azide led to the formation of dinuclear entities and fre-
quently constrained one-dimensional (1D) chains.30ꢀ38 The
SMM behavior was observed in some discrete molecules,30,39
and a few MnIII chain systems bridged by a single azide exhibited
spin canting and slow relaxation dynamics coupled with field-
induced magnetic transitions and/or antiferromagnetic order.36,37
Molecule-based magnetic materials have been actively studied
for their potential applications in magnetic devices, and con-
siderable progress has been made in the pursuit of constructing
anisotropic low-dimensional molecular nanomagnets with mag-
netization relaxations and high-dimensional coordination net-
works with long-range magnetic ordering. The former concerns
the development of single-molecule magnets (SMMs)1ꢀ5 and single-
chain magnets (SCMs)6ꢀ8 that require single-ion anisotropy and
negligible intermolecular or interchain magnetic interactions.
Under these conditions, the characteristic slow relaxation of
magnetization emerges in anisotropic molecular systems. For the
latter, it is generally believed that spin centers should interact
magnetically in a three-dimensional (3D) manner to bring about
long-range magnetic alignment.
To achieve the aforementioned molecular magnetic systems
with exotic properties, it is essential to select appropriate bridging
ligands that can hold paramagnetic centers closely together and
communicate effective magnetic exchange coupling between them.
In this line, it is desirable to use ligands with short paths such as
oxide, cyanide, carboxylate, or azide and so on.9ꢀ12 Among them,
azide acts as a bridge in diverse binding modes, for instance, μ-1,1-
N3 (or single end-on), di-μ-1,1-N3 (double end-on), μ-1,3-N3
(single end-to-end), di-μ-1,3-N3 (double end-to-end), μ-1,1,1-N3,
μ-1,1,3-N3, μ-1,1,1,1-N3, μ-1,1,3,3-N3, and μ-1,1,1,3,3,3-N3.13
Received: June 20, 2011
Published: October 06, 2011
r
2011 American Chemical Society
10777
dx.doi.org/10.1021/ic2013232 Inorg. Chem. 2011, 50, 10777–10785
|

Inorganic Chemistry
ARTICLE
On the other hand, despite such 1D MnIII structures, the bulk
properties of the critical temperature and coercive field have been
demonstrated, as low as T < 11 K and as large as H = 2.3 kG,
respectively, as a result of the degree of interchain interaction and
the effect of magnetic anisotropy.30ꢀ38 In molecular magnets, it is
important to know the factors affecting the critical temperature
and coercive field for the sake of the fundamental understanding
of long-range order and practical application issues. It is thus
highly sought to establish a simple anisotropic system, and in this
regard, we have been interested in investigating such azide-linked
MnIII chains chelated with various tetradentate Schiff bases.
Herein we report the syntheses, crystal structures, and mag-
netic properties of three 1D chains, [Mn(L)(N3)]n [L = L1 (1),
L2 (2), L3 (3); L1H2 = N,N0-bis(5-chlorosalicylideneiminato)-
1,3-diaminopentane, L2H2 =N,N0-bis(5-bromosalicylideneiminato)-
1,3-diaminopentane, L3H2 =N,N0-bis(5-bromosalicylideneiminato)-
1,3-diamino-2-dimethylpropane], linked by end-to-end azides.
Antiferromagnetic interactions between MnIII spins within a
chain are operating through the azide bridges, and spin canting
clearly arises at low temperatures. Field-induced magnetic transi-
tions are realized in the 1D chain systems. It is noted that a long-
range order occurs at a high temperature of 38 K for 2 devoid of
any noncovalent interaction between chains, which may be due
to the extensive incorporation of dipolar interactions across the
lattice. A low critical temperature is seen at 11 K for 1 and at 10 K
for 3, where interchain πꢀπ contacts are involved. Compound 3
has a relatively high coercive field of 2.8 kG.
Table 1. Crystallographic Data for 1 - 3
1
2
3
formula
C
19H18Cl2-
MnN5O2
C38H36Br4-
Mn2N10O4
1126.29
100
C19H18Br2-
MnN5O2
563.14
298
Mr
474.22
293
T (K)
cryst syst
space group
a (Å)
orthorhombic
Pbca
triclinic
P1
monoclinic
P21/c
29.3009(17)
11.3200(6)
12.0634(7)
90
11.8425(10)
12.7124(10)
15.4367(16)
97.999(4)
112.472(3)
96.399(3)
2091.9(3)
2
12.6256(15)
12.8577(16)
12.4425(15)
90
b (Å)
c (Å)
α (deg)
β (deg)
γ (deg)
V (Å3)
Z
90
90
90
90
4001.3(4)
8
2019.9(4)
4
F
calc (g cmꢀ3
)
1.574
1.788
1.852
μ (mmꢀ1
)
0.953
4.477
4.636
F(000)
1936
1112
1112
total no. of reflns
GOF
60 355
1.029
33 430
15 490
0.885
0.994
R1a [I g 2σ(I)]
wR2b [I g 2σ(I)]
0.0541
0.0764
0.0533
0.1256
0.1558
2
0.0681
a R1 = ∑||Fo| ꢀ |Fc||/∑|Fc|. b wR2 = [∑w(Fo ꢀ Fc2)2/∑w(Fo ) ]
.
2 2 1/2
Absorption corrections were carried out using SADABS. The structures
were solved by direct methods and refined by full-matrix least-squares
analysis using anisotropic thermal parameters for non-H atoms with the
SHELXTL program. All H atoms for 1ꢀ3 were calculated at idealized
positions and refined with the riding models. For 3, we performed
ADDSYM, an algorithm for missing symmetry on WingX, to check the
space group, and no obvious space group change was suggested.
Crystallographic data for 1ꢀ3 are summarized in Table 1.
’ EXPERIMENTAL SECTION
To synthesize 1, an aqueous solution of 10 mL of NaN3 (7mg,0.10mmol)
was added to a methanol solution (30 mL) of [Mn(L1)(H2O)] ClO4
(48 mg, 0.10 mmol). After stirring for several minutes, the mixture was
filtered. After a few days, brown crystals were obtained. The crystals were
filtered off, washed with methanol, and dried in air. Yield: 42%. Anal.
Calcd for C19H18Cl2MnN5O2: C, 48.12; H, 3.83; N, 14.77. Found: C,
3
48.33; H, 3.79; N, 14.46. IR (KBr): νN 2056(sh), 2040(vs), 1983(w).
3
Complexes 2 and 3 were prepared by the same procedure as that of 1
using [Mn(L2)(H2O)] ClO4 for 2 or [Mn(L3)(H2O)] ClO4 for 3
’ RESULTS AND DISCUSSION
3
3
instead of [Mn(L1)(H2O)] ClO4. Yield of 2: 30%. Anal. Calcd for
3
Description of the Structures. The structures of 1ꢀ3 have
been characterized using single-crystal X-ray diffraction techniques.
Complexes 1 and 2. Complex 1 crystallizes in the orthorhom-
bic system (Pbca), whereas 2 belongs to the triclinic (P1) system.
The crystal structures of 1 and 2 and their extended two-
dimensional (2D) arrays are illustrated in Figures 1 and 2. The structure
of 1 is not identical with that of 2. This suggests that the change in
the halides (Cl for 1 and Br for 2) on the 5 position of the Schiff
bases is responsible for such discernible influence on the crystal
system and structure. Each Mn center consists of a distorted
octahedron surrounded by two N atoms from the bridging azides
and N2O2 donors from the corresponding tetradentate Schiff
bases (L1 and L2). For 1, the MnꢀNax (ax = axial) distances are
long, 2.317(4) Å for Mn1ꢀN1 and 2.252(4) Å for Mn1ꢀN3b
(b, x, 0.5 ꢀ y, 0.5 + z). Comparatively, the average MnꢀN(O)eq
(eq = equatorial) length in the equatorial N2O2 plane of L1
corresponds to 1.96(7) Å. The typical tetragonal elongation thus
points out the presence of a JahnꢀTeller distortion, as is found in
many octahedral high-spin MnIII entities.29ꢀ38 In the case of 2,
the JahnꢀTeller effect also occurs as judged by the long apical
MnꢀNax lengths [2.297(8) Å for Mn1ꢀN1, 2.249(9) Å for
Mn1ꢀN8d (d, ꢀ1 + x, y, z), 2.260(8) Å for Mn2ꢀN3, and
2.240(9) Å for Mn2ꢀN6] and the short average MnꢀN(O)eq
C38H36Br4Mn2N10O4: C, 40.52; H, 3.22; N, 12.44. Found: C, 40.42; H,
3.27; N, 12.23. IR (KBr): νN 2046(vs), 1990(sh). Yield of 3: 39%. Anal.
Calcd for C19H18Br2MnN5O2: C, 40.52; H, 3.22; N, 12.44. Found: C,
3
40.55; H, 3.24; N, 11.95. IR (KBr): νN 2052(vs), 1983(w). The purity
3
of the bulk samples was verified by powder X-ray diffraction (PXRD; see
Figure S1 in the Supporting Information).
Physical Measurements. Elemental analyses for C, H, and N
were performed at the Elemental Analysis Service Center of Sogang
University. IR spectra were obtained from KBr pellets with a Bomen
MB-104 spectrometer. PXRD data were recorded using Cu Kα
(λ = 1.5406 Å) on a Rigaku Ultima III diffractometer with a scan speed of
3° minꢀ1 and a step size of 0.01°. Magnetic susceptibilities for 1ꢀ3 in
eicosane were carried out using a Quantum Design SQUID susceptometer
(direct current, dc) and a PPMS magnetometer (alternating current, ac).
Corrections for the sample holder and eicosane were made, and dia-
magnetic corrections of all samples were estimated from Pascal’s Tables.
Crystallographic Structure Determination. X-ray data for
1ꢀ3werecollectedona BrukerSMARTAPEXII diffractometerequipped
with graphite-monochromated Mo Kα radiation (λ = 0.710 73 Å).
A preliminary orientation matrix and cell parameters were determined
from three sets of ω scans at different starting angles. Data frames were
obtained at scan intervals of 0.5° with an exposure time of 10 s frameꢀ1
The reflection data were corrected for Lorentz and polarization factors.
.
10778
dx.doi.org/10.1021/ic2013232 |Inorg. Chem. 2011, 50, 10777–10785

Inorganic Chemistry
ARTICLE
Figure 2. Top: Crystalstructure of2showingthe atom-labeling schemes.
Symmetry codes: c, 1 + x, y, z; d, ꢀ1 + x, y, z. Bottom: Extended 2D view
with the shortest interchain distance (dotted lines).
and S4 in the Supporting Information. The noticeable structural
alteration in 2, compared with 1, is that the shortest interchain
MnꢀMn distances are reduced to 6.566 and 6.659 Å, although
another MnꢀMn distance between the other chains slightly
increases to 11.633 Å. Interestingly, P- and M-helical chains with
a long pitch length of 11.843 Å in the lattice of 2 are efficiently
packed together in a heterochiral fashion, eventually resulting in a
racemate (Figure S5 in the Supporting Information). This
racemic arrangement was often encountered in systems com-
posed of not only archiral but also chiral ligands.40,41
Complex 3. The use of L3 with the dimethyl-substituted
propylene linker and 5-positioned Br groups affords the end-to-
end azide-bridged 1D chain system crystallized in the monoclinic
system with the space group P21/c (Figure 3). This material
exhibits an analogous octahedral environment around a MnIII
center in which the MnꢀNax lengths are 2.269(4) Å for
Mn1ꢀN1 and 2.289(4) Å for Mn1ꢀN3e (e, x, 1.5 ꢀ y, 0.5 + z)
and the average MnꢀN(O)eq length is 1.96(7) Å. The
MnꢀNaxꢀNax angles are 137.9(3)° for Mn1ꢀN1ꢀN2 and
128.3(3)° for Mn1ꢀN3eꢀN2e. The mean MnꢀNax length
(2.279 Å) lies between 1 and 2, while the average MnꢀNaxꢀNax
angle (133.1°) is larger than that of 1 and 2. The torsion angle
of MnꢀNax---NaxꢀMn along the chain corresponds to 128.6°,
and the intrachain distance through the azide bridge is 6.231 Å. In
the extended structure of 3 (Figures 3 and S6 and S7 in the
Supporting Information), weak πꢀπ stacking exists between the
phenoxide rings with a centroid distance of 3.742 Å and the
Figure 1. Top: Molecular view of 1 with the atom-labeling schemes.
Symmetry codes: a, x, 0.5 ꢀ y, ꢀ0.5 + z; b, x, 0.5 ꢀ y, 0.5 + z. Bottom:
Extended 2D viewshowing the shortest interchain distance (dotted lines).
distance of 1.97(8) Å. The MnꢀNaxꢀNax angles in the bridging
pathways are 125.6(3)° for Mn1ꢀN1ꢀN2 and 133.9(4)° for
Mn1ꢀN3bꢀN2b (1) and 119.1(6)° for Mn1ꢀN1ꢀN2,
129.9(8)° for Mn1ꢀN8dꢀN7d, 149.6(7)° for Mn2ꢀN3ꢀN2,
and 122.5(8)° for Mn2ꢀN6ꢀN7 (2). Notice that the average
MnꢀNax length (2.285 Å) of 1 is longer than that (2.268 Å) of 2,
while the mean MnꢀNaxꢀNax angle (129.8°) of 1 is similar to
that (130.3°) of 2. The torsion angles of MnꢀNax---NaxꢀMn are
116.0° for 1 and 131.1° and 159.3° for 2. The azide ligand
connects two adjacent MnIII ions in an end-to-end mode with
MnꢀMn distances of 6.069 Å for Mn1ꢀMn1a (1; a, x, 0.5 ꢀ y,
ꢀ0.5 + z) and 6.250 Å for Mn1ꢀMn2 and 6.009 Å for Mn2ꢀ
Mn1c (2; c, 1 + x, y, z) to generate a 1D linear-chain structure.
In the extended structure of 1 with the Cl-substituted Schiff
base L1 (Figures 1 and S2 in the Supporting Information), chains
are running along the c axis. The chains are separated with the
shortest MnꢀMn distance of 7.335 Å. The tiny πꢀπ interactions
with a centroid distance of 3.955 Å are established between the
phenoxide rings of L1, and the interchain MnꢀMn distance via
the contacts is equal to 11.204 Å (Figure S3 in the Supporting
Information). When Cl atoms are replaced with Br, the overall
crystal packing is subject to severe changes, as shown in Figures 2
10779
dx.doi.org/10.1021/ic2013232 |Inorg. Chem. 2011, 50, 10777–10785

Inorganic Chemistry
ARTICLE
Figure 4. Plots of χmT versus T for 1 and 2 at 1000 G. The solid lines in
the main panel show the best magnetic fits.
Figure 5. Plot of M versus H for 1 at 2 K. The inset is a blowup of the
low-field magnetization data.
Figure 3. Top:Crystal structure of3 showing the atom-labeling schemes.
Symmetry codes: e, x, 1.5 ꢀ y, 0.5 + z; f, x, 1.5 ꢀ y, ꢀ0.5 + z. Bottom:
Extended 2D view with the shortest interchain distance (dotted lines).
To probe the magnetic exchange coupling constants within
a chain, we employed an infinite-chain model derived by Fisher
(H = ꢀJ∑iSA SA ) in the high-temperature range to preclude
3
i
i+1
the interchain magnetic contributions and zero-field-splitting
effects.47 The fitting process gives the best-fit parameters of
g = 2.07(1) and J = ꢀ7.3(1) cmꢀ1 for 1 at T > 15 K and g =
2.05(1) and J = ꢀ11.5(1) cmꢀ1 for 2 at T > 40 K. The negative
J values indicate the existence of intrachain antiferromagntic
coupling between the Mn spins mediated by end-to-end azides,
and the coupling strength falls into the usual range of end-to-end
azide-bridged MnIII systems.29
shortest interchain MnꢀMn distance via the noncovalent force is
7.072 Å. The other interchain MnꢀMn distance along the c axis
is 12.626 Å.
Magnetic Properties of 1 and 2. The thermal variations in
the dc magnetic susceptibility data of 1 and 2 were collected at
H = 1 kG and T = 2ꢀ300 K (Figure 4). The χmT values at 300 K
are 2.87 and 2.52 cm3 K molꢀ1 for 1 and 2, respectively, which are
somewhat smaller than the spin-only one expected for a non-
coupled MnIII (SMn = 2) ion. The χmT product declines as the
temperature is lowered and a minimum arrives at 15 K for 1 and
at 40 K for 2. Below these temperatures, a cusp becomes visible,
followed by a further decay of χmT down to 2 K. The upturn in
χmT at low temperatures is due to the occurrence of spin canting
in the antiferromagnetic system arising from the involvement of
single-ion anisotropy of a MnIII ion and two types of MnIII ions
along the chain direction. These factors energetically favor the
perpendicular orientation for the coupled spins induced by the
antisymmetric DzyaloshinskyꢀMoriya (DꢀM) coupling, result-
ing in spin canting.42,43 Similar spin-canted phenomena were also
revealed in several cobalt(II), nickel(II), and manganese(III)
complexes bridged by diverse bridging ligands.44ꢀ46 The inter-
chain antiferromagnetic interactions and zero-field splitting of
the complexes may account for the final decrease in the χmT(T)
plots below the cusp temperatures.
To examine the magnetic nature of 1, we measured the field-
dependent magnetization data at several temperatures (Figures 5
and S8 in the Supporting Information). The magnetization
amounts to 0.53 Nβ in a field of 7 T, which is not saturated
and is away from the theoretical value of 4 Nβ assuming g = 2.
The coercive field (Hcr) at 2 K is 600 G. An extrapolation of the
high-field straight section of the magnetization to zero field
approaches 0.15 Nβ and the canting angle, based on sinꢀ1(Mw/Ms),
where Ms is the saturation magnetization and MW a magnetiza-
tion induced by a weak magnetic field, is 2.2° at 2 K. In the
derivative of M against H at 2 K (Figure S9 in the Supporting
Information), a positive peak is shown at HC = ∼10 kG. As the
temperature varies, HC tends to lessen (∼7 kG at 3 K, ∼6 kG at
4 and 5 K, and ∼5.5 kG at 7 and 9 K) and eventually vanish at
T = 11 K. To gain in-depth insight into the magnetic feature,
the field-cooled (FC) magnetizations were collected in the field
range from 20 G to 25 kG (Figure 6). An abrupt rise in the χm(T)
10780
dx.doi.org/10.1021/ic2013232 |Inorg. Chem. 2011, 50, 10777–10785

Inorganic Chemistry
ARTICLE
Figure 6. Plots of FC data versus T for 1 at the different magnetic fields.
Figure 8. Plot of M versus H for 2 at 2 K. The inset is a blowup of the
low-field magnetization data.
Figure 7. In-phase (χm0) and out-of-phase (χm00) components for 1
with increasing frequencies.
Figure 9. Plots of FC data versus T for 2 at the different magnetic fields.
curve begins at ∼11 K in the low-field regime and then lowering
the temperature causes χm to be significantly reduced. The maxi-
mum peak at ∼10 K survives until H = 5 kG and disappears at
H g 10 kG. This behavior is typical of a field-induced metamag-
netic transition from an antiferromagnetic state to a spin-canted
phase (the weak ferromagnetic phase). Stabilization of the antifer-
romagnetic state seems to be in connection with the existence of
the πꢀπ-stacking interactions. The observed M(H) and M(T)
data demonstrate that the weak interchain interactions between
canted-spin chains are overcome by an external magnetic field of
H > HC, leading to a weak ferromagnetic phase. The HꢀT phase
diagram is plotted in Figure S10 in the Supporting Information.
The FC and zero-field-cooled (ZFC) magnetizations were
recorded at 50 G, and there is thermomagnetic irreversibility start-
ing from Tc = 11 K determined by the FC dM/dH curve (Figure
S11 in the Supporting Information). This divergence of FC and
ZFC may be related to superparamagnets, spin glasses, or mag-
netic domain movements. To elucidate the genuine character, ac
magnetic susceptibility data were collected at zero dc field, an ac
field of 5 G, and various oscillating frequencies (Figure 7). Because
no frequency dependence is detectable, the possibility of super-
paramagnets and spin glasses can be discarded. Consequently, the
irreversible response in the FC/ZFC curves supports the presence
of a magnetized phase, as ascertained by the hysteresis behavior in
The field dependence of the magnetization for 2 is displayed in
Figure 8. The magnetization data slowly increase with the field,
suggesting the operation of dominant antiferromagnetic interac-
tions. The magnetization value of 0.25 Nβ at 70 kG is much
smaller than that of 1. The coercive field (Hcr) of 650 G in 2 is
analogous to that in 1. The extrapolated remnant magnetization
at H = 0 G is 0.0035 Nβ, and the canting angle is calculated to be
0.05°. The inflection point in the M(H) plot shows that a field-
induced two-step transition occurs at critical fields of H1 = 8 kG
and H2 = 60 kG (Figure S12 in the Supporting Information). The
first critical field may be ascribed to a transition from a spin-
canted state to another spin-canted (spin-flop) phase, and the
second may be a spin reorientation to the field direction. It is
estimated that the anisotropy (HA) and exchange fields (HE) are
1.1 and 31 kG, respectively. The HA/HE ratio is 3.5 ꢁ 10ꢀ2
,
5
which is slightly larger than those of S = /2 Heisenberg anti-
ferromagnets with a spin-flop transition.48 Similar properties
were also observed in some anisotropic manganese(III) and
cobalt(II) complexes.49ꢀ51 To inspect the magnetic trait in
detail, the FC magnetizations were recorded at diverse magnetic
fields, as illustrated in Figure 9. No peaks are observed in the FC
plots in the low-field range, which is contrary to 1, where the
sharp maxima are placed at low temperatures and fields. This
behavior indicates the presence of a weak ferromagnetic state in
the FC process at those fields. An obvious transition evolves at
38 K in the low-field region, and the χm response is inclined to
decrease with increasing field and is almost suppressed at Hg20kG.
0
the M(H) data. From the ac plot, the in-phase (χm ) and out-of-
phase (χm00) signals are located at Tc = 11 K, implying the onset of
long-range magnetic ordering.
10781
dx.doi.org/10.1021/ic2013232 |Inorg. Chem. 2011, 50, 10777–10785

Inorganic Chemistry
ARTICLE
Figure 11. Temperature dependence of FC and ZFC for 2 at 50 G.
chains are magnetically correlated to some extent to realize a 2D
magnetic array. The combined 2D layers may be further coupled
through space, albeit weakly. Such an extended correlation may
create long-range order when anisotropic MnIII ions are incor-
porated in the chains. To check this scenario, we scrutinized the
critical temperature region (1 ꢀ T/Tc < 10ꢀ2ꢀ10ꢀ1).52 The
thermodynamic behavior of the spontaneous magnetization can
be expressed by the asymptotic power law M = M0(1 ꢀ T/Tc)β,
where M0 is the critical amplitude and β is the critical exponent.53
The β term is related to the spin and lattice dimensionality. In the
double-logarithmic plot of M versus the reduced temperature
(Figure S14 in the Supporting Information), a linear fit of
the data to the equation with Tc = 38K yields β = 0.21. The β value
remains between theoretical 2D Ising (β = 0.125) and 3D Ising
(β = 0.325) arrays or 3D Heisenberg (β = 0.364) magnetic
systems.52 Provided there are no noncovalent interchain contacts
in 2, the long-range order could not result from quantum exchange
interactions, which can be transmitted by noncovalent hydrogen
bonds or πꢀπ-stacking forces. The superexchange interaction, if
any, is also known to rapidly fade away with increasing interchain
spacing so that it seldom explains the observed high critical tem-
perature.54 The other possibility is a through-space dipolar inter-
action, as exemplified in copper(II) hybrid inorganicꢀorganic
layered compounds, where Tc was observed at ∼20 K with an
interlayer spacing as large as 40 Å.54ꢀ56 Accordingly, the critical
analysis (β = 0.21) of 2 supports that a 2D spin lattice is formed by
dipolar interactions through the shortest spacings (6.613 Å on
average) between the nearest neighboring chains (bottom of
Figure 2) and each quasi-2D magnetic layer is weakly interacted
through the other type of dipolar interaction via the longer spacing
(11.633 Å), eventually promoting 3D magnetic ordering.
Figure 10. (a) In-phase (χm0) and (b) out-of-phase (χm00) components
for 2 with increasing frequencies.
Interestingly, a hump is evident at 12 K, which disappears at H =
70 kG. This is tentatively assigned to antiferromagnetic ordering
0
throughout the 3D lattice because χm signals are present but
00
rough χm responses appear to be absent in the same tempera-
ture range (Figure 10). The HꢀT phase diagram is suggested in
Figure S13 in the Supporting Information.
The FC and ZFC curves of 2 are depicted in Figure 11. The
saturated behavior at low temperatures denotes the existence
of weak ferromagnetism in FC and ZFC. They are split below
Tc = 38 K, which is determined by the inflection point in the
M(T) data. This can be attributed to the formation of magnetic
domains. Under the FC process, the domain walls move freely
and the size of the domains grows. In the ZFC process, the
domains are randomly oriented and frozen. The application of a
magnetic field is not sufficient to unlock the domain wall
motions, giving the observed MZFC < MFC. In the ac magnetic
data (Figure 10), the critical temperature is confirmed at Tc =
When it comes to the critical feature of 2, it is worth men-
tioning that Tc surprisingly enhances by 27 K with a change of Cl
in L1 (1) to Br in L2 (2). In 2D layered magnets, in-plane
exchange coupling (Jin‑plane) imposes a substantial impact on Tc,
while variation of the interplanar spacing (jinterplane) slightly
affects Tc. The relationship of Tc with J and j can be expressed
as Tc = 4πJin‑plane/ln(Jin‑plane/jinterplane).54 On the basis of this
expression, the noteworthy Tc in 2 may be primarily associated
with the enhanced magnetic correlation in the quasi-2D lattices,
the stronger intrachain magnetic coupling (J) of 2 than of 1, as
well as the closer interchain distance in the ab plane of 2 than in
the bc plane of 1.
00
38 K, where a sharp peak in χm0 and χm00 coincides. The apex in χm
pertains to the beginning of a ferromagnetically ordered state.
The peak position is independent of the oscillating frequency,
suggesting that there are no glassy or superparamagnetic beha-
vior. It is noted that, as far as the 1D chain structure is concerned,
the Tc of 38 K in 2 is relatively high and is even greater than that
of 2D cobalt(II) azide multilayer systems with Tc = 23 K at best.27
Theoretically, it is impossible to obtain a magnetized phase in a
1D chain. However, when interchain magnetic interactions turn
out to be sizable in a real system, a 1D compound can give rise to
long-range magnetic order. From the structural data of 2, it is
apparent that the chains in the bc plane are most closely positioned
among all of the complexes (Figure 2). It is conjectured that the
Magnetic Properties of 3. The thermomagnetic data of 3
were collected at 1 kG and at temperatures in the range from 2
10782
dx.doi.org/10.1021/ic2013232 |Inorg. Chem. 2011, 50, 10777–10785

Inorganic Chemistry
ARTICLE
Figure 12. Plot of χmT versus T for 3 at 1000 G. The solid line presents
the best magnetic fit.
Figure 14. Plots of FC data versus T for 3 at various magnetic fields.
Figure 13. Plots of M versus H for 3 at several fields. The inset is a
blowup of the low-field magnetization data.
to 300 K, as shown in Figure 12. The χmTvalue of 2.72 cm3 Kmolꢀ1
at 300 K, which is similar to those of 1 and 2, decreases steadily
with cooling of the sample and reaches a shoulder at T ≈ 9 K.
Below that temperature, χmT decreases again. The overall
reduction in χmT is a signature of the antiferromagnetic interac-
tions. The occurrence of a shoulder can be ascribed to the canted
phenomenon of neighboring spins. The intrachain magnetic
exchange interaction can be assayed using the Fisher chain
model.47 A best fit affords g = 2.05(1) and J = ꢀ8.0(1) cmꢀ1
at T > 15 K, and the J magnitude is similar to that of 1 but smaller
than that of 2. The order of magnetic strength appears to be
paralleled by the average MnꢀNax bond length.
Figure 15. (a) In-phase (χm0) and (b) out-of-phase (χm00) components
for 3 with increasing frequencies.
Isothermal magnetization of 3 was measured in the field range
of ꢀ70 kG e H e 70 kG at several temperatures (Figure 13).
The magnetization increases slowly and reaches 0.34 Nβ at
70 kG, which is much smaller than saturation. This feature shows
the involvement of the antiferromagnetic interactions and mag-
netic anisotropy of the material. The linear part of the high-field
region is taken to compute a canting angle of 0.49°. In the
hysteresis loop, Hcr at 2 K is 2.8 kG, which is the highest value of
the azide-bridged MnIII systems.29 Hcr decreases to 120 G at 4 K
and 10 G at 7 K. In the azide-bridged cobalt(II) layered com-
pounds, the coercive field tends to decrease with increasing inter-
layer spacing because of weakening of the anisotropy field.27
In addition, intralayer structures together with other factors
(particle shape and size) should also be taken into account
because in some cases the interlayer separation only could not
account for the coercivity.27 In the case of 3, the shortest inter-
chain distance (7.072 Å) may magnetically engender a quasi-2D
lattice and the other separation (12.626 Å) may act as an
interplanar spacing. With a larger interplanar distance of 3 than
those of 1 and 2, it is impossible to address the larger Hcr of 3
based on the situation applied to the cobalt(II) azide layered
systems. Therefore, all of the structural and relevant factors
should be combined to justify the hysteresis trait.
The critical fields (HC) are obtained from the peaks of the
dM/dH curves (Figure S15 in the Supporting Information).
10783
dx.doi.org/10.1021/ic2013232 |Inorg. Chem. 2011, 50, 10777–10785

Inorganic Chemistry
ARTICLE
HC at 2 K is 8 kG, at which a field-induced metamagnetic
transition transpires. When the temperature is elevated, HC is
prone to dwindle to ca. 1.6 kG at T = 4 and 7 K and finally pass
away at T = 15 K. The magnetic conversion is ascertained by the
FC plots at several magnetic fields (Figure 14). The maximum at
Tmax ≈ 7.5 K in the FC plot is retained up to H = 100 G and
disappears above H = 2 kG. The peak feature designates that an
antiferromagnetic state is stabilized in that field. The antiferro-
magnetic phase may be established because of the πꢀπ contacts.
funded by the Ministry of Education, Science and Technology
(NRF20110018396).
’ REFERENCES
(1) Das, A.; Gieb, K.; Krupskaya, Y.; Demeshko, S.; Dechert, S.;
Klingeler, R.; Kataev, V.; Buchner, B.; Muller, P.; Meyer, F. J. Am. Chem.
Soc. 2011, 133, 3433–3443.
(2) Hewitt, I. J.; Tang, J.; Madhu, N. T.; Anson, C. E.; Lan, Y.; Luzon,
J.; Etienne, M.; Sessoli, R.; Powell, A. K. Angew. Chem., Int. Ed. 2010,
49, 6352–6356.
The critical field is estimated at H ≈ 1.6 kG in the vicinity of Tmax
,
and above that field, a weak ferromagnetic phase results. The ac
(3) Jiang, S. D.; Wang, B. W.; Su, G.; Wang, Z. M.; Gao, S. Angew.
Chem., Int. Ed. 2010, 49, 7448–7451.
data disclose that the broad peak in χm0 is centered at TN ≈ 10 K
00
(4) Rinck, J.; Novitchi, G.; Van den Heuvel, W.; Ungur, L.; Lan,
Y. H.; Wernsdorfer, W.; Anson, C. E.; Chibotaru, L. F.; Powell, A. K.
Angew. Chem., Int. Ed. 2010, 49, 7583–7587.
(5) Gatteschi, D.; Sessoli, R. Angew. Chem., Int. Ed. 2003, 42, 268–297.
(6) Bogani, L.; Vindigni, A.; Sessoli, R.; Gatteschi, D. J. Mater. Chem.
2008, 18, 4750–4758.
without any χm responses, suggesting the existence of long-
range antiferromagnetic order with a hidden spin canting (Figure 15).
The HꢀT phase diagram is shown in Figure S16 in the Supporting
Information.
’ CONCLUSIONS
(7) Harris, T. D.; Bennett, M. V.; Clerac, R.; Long, J. R. J. Am. Chem.
Soc. 2010, 132, 3980–3988.
(8) Sun, H.-L.; Wang, Z.-M.; Gao, S. Coord. Chem. Rev. 2010, 254,
1081–1100.
(9) Davies, J. E.; Gatehouse, B. M.; Murray, K. S. J. Chem. Soc., Dalton
Trans. 1973, 2523–1973.
(10) Kennedy, B. J.; Murray, K. S. Inorg. Chem. 1985, 24, 1552–1557.
(11) Sun, H.-L.;Wang, Z.-M.;Gao, S. Inorg. Chem. 2005, 44, 2169–2176.
(12) Price, D. J.; Batten, S. R.; Moubaraki, B.; Murray, K. S. Poly-
hedron 2003, 22, 2161–2167.
(13) Adhikary, C.; Koner, S. Coord. Chem. Rev. 2010, 254, 2933–2958.
(14) Hong, C. S.; Do, Y. Angew. Chem., Int. Ed. 1999, 38, 193–195.
(15) Hong, C. S.; Koo, J. E.; Son, S. K.; Lee, Y. S.; Kim, Y. S.; Do, Y.
Chem.—Eur. J. 2001, 7, 4243–4252.
(16) Yoo, H. S.; Kim, J. I.; Yang, N.; Koh, E. K.; Park, J. G.; Hong,
C. S. Inorg. Chem. 2007, 46, 9054–9056.
(17) Nayak, S.; Evangelisti, M.; Powell, A. K.; Reedijk, J. Chem.—Eur.
J. 2010, 16, 12865–12872.
(18) Li, R.-Y.; Wang, Z.-M.; Gao, S. CrystEngComm 2009, 11, 2096.
(19) Wang, X.-Y.; Wang, Z.-M.;Gao, S. Chem. Commun. 2008, 281–294.
(20) Liu, X. T.; Wang, X. Y.; Zhang, W. X.; Cui, P.; Gao, S. Adv.
Mater. 2006, 18, 2852–2856.
(21) Wang, X.-Y.; Wang, L.; Wang, Z.-M.; Su, G.; Gao, S. Chem.
Mater. 2005, 17, 6369–6380.
(22) Reddy, K. R.; Rajasekharan, M. V.; Tuchagues, J. P. Inorg. Chem.
1998, 37, 5978–5982.
(23) Monfort, M.; Ribas, J.; Solans, X.; Font-Bardí, M. Inorg. Chem.
1996, 35, 7633–7638.
(24) Yin, P.; Gao, S.; Zheng, L.-M.; Wang, Z.; Xin, X.-Q. Chem.
Commun. 2003, 1076–1077.
(25) Wang, X.-Y.; Wang, L.; Wang, Z.-M.; Gao, S. J. Am. Chem. Soc.
2006, 128, 674–675.
(26) Jia, Q.-X.; Tian, H.; Zhang, J.-Y.; Gao, E.-Q. Chem.—Eur. J.
2011, 17, 1040–1051.
(27) Gao, E.-Q.; Liu, P.-P.; Wang, Y.-Q.; Yue, Q.; Wang, Q.-L. Chem.
—Eur. J. 2009, 15, 1217–1226.
(28) Liu, T.-F.; Fu, D.; Gao, S.; Zhang, Y.-Z.; Sun, H.-L.; Su, G.; Liu,
Y.-J. J. Am. Chem. Soc. 2003, 125, 13976–13977.
(29) Miyasaka, H.; Saitoh, A.; Abe, S. Coord. Chem. Rev. 2007,
251, 2622–2664.
(30) Ge, C.-H.; Cui, A.-L.; Ni, Z.-H.; Jiang, Y.-B.; Zhang, L.-F.; Ribas,
J.; Kou, H.-Z. Inorg. Chem. 2006, 45, 4883–4885.
(31) Ko, H. H.; Lim, J. H.; Kim, H. C.; Hong, C. S. Inorg. Chem.
2006, 45, 8847–8849.
We have prepared three 1D Mn(III) coordination polymers
with the Schiff bases (L1 for 1, L2 for 2, and L3 for 3) bridged by
end-to-end azides. The crystal systems and structural parameters
are altered by relying on the Schiff bases used. The variable-
temperature magnetic measurements reveal that intrachain anti-
ferromagnetic couplings operate between the magnetic centers
via azide bridges, and spin-canting behavior is obvious at low tem-
peratures. Interestingly, the field-induced metamagnetic transi-
tion from an antiferromagnetic state to a weak ferromagnetic
phase takes place in 1 and 3, whereas the field-induced two-step
magnetic phase transition occurs in 2. The exchange coupling
constant (J) of 2 is larger than those of 1 and 3, which is likely
relevant to the MnꢀNax bond lengths. Notably, a significantly
high Tc of 38 K is observed in 2, which is compared with critical
temperatures of 11 K for 1 and 10 K for 3. In addition, 3 exhibits
the highest coercive field (2.8 kG) among azide-bridged MnIII
systems examined so far. The long-range order can be under-
stood in terms of the combined contributions of the intrachain
exchange couplings (J) and through-space dipolar interaction
between the neighboring chains. The finding of such a high
ordering temperature (38 K) in the spin-canted 1D chain system
(2) without hydrogen-bonding and/or πꢀπ contacts between
the chain structures is surprising, which may provide a strategy to
developing high-Tc molecular magnets by manipulating the
dipolar interactions.
’ ASSOCIATED CONTENT
S
Supporting Information. X-ray crystallographic files in
b
CIF format and additional structural and magnetic data for the
complexes. This material is available free of charge via the
Internet at http://pubs.acs.org.
’ AUTHOR INFORMATION
Corresponding Author
*E-mail: cshong@korea.ac.kr.
’ ACKNOWLEDGMENT
(32) Liu, C. M.; Zhang, D. Q.; Zhu, D. B. Inorg. Chem. 2009, 48,
4980–4987.
(33) Saha, S.; Mal, D.; Koner, S.; Bhattacherjee, A.; G€utlich, P.;
Mondal, S.; Mukherjee, M.; Okamoto, K.-I. Polyhedron 2004, 23,
1811–1817.
This work was supported by grants from the National Research
Foundation of Korea funded by the Korean Government (Grant
2011-0003264) and by the Priority Research Centers Program
through the National Research Foundation of Korea (NRF)
10784
dx.doi.org/10.1021/ic2013232 |Inorg. Chem. 2011, 50, 10777–10785

Inorganic Chemistry
ARTICLE
(34) Stamatatos, T. C.; Christou, G. Inorg. Chem. 2009, 48, 3308–3322.
(35) Viciano-Chumillas, M.; Tanase, S.; Mutikainen, I.; Turpeinen,
U.; de Jongh, L. J.; Reedijk, J. Inorg. Chem. 2008, 47, 5919–5929.
(36) Yoon, J. H.; Lee, J. W.; Ryu, D. W.; Yoon, S. W.; Suh, B. J.; Kim,
H. C.; Hong, C. S. Chem.—Eur. J. 2011, 17, 3028–3034.
(37) Yoon, J. H.; Ryu, D. W.; Kim, H. C.; Yoon, S. W.; Suh, B. J.;
Hong, C. S. Chem.—Eur. J. 2009, 15, 3661–3665.
(38) Yuan, M.; Zhao, F.; Zhang, W.; Wang, Z.-M.; Gao, S. Inorg.
Chem. 2007, 46, 11235–11242.
(39) Lu, Z.; Yuan, M.; Pan, F.; Gao, S.; Zhang, D.; Zhu, D. Inorg.
Chem. 2006, 45, 3538–3548.
(40) Tabellion, F. M.; Seidel, S. R.; Arif, A. M.; Stang, P. J. Angew.
Chem., Int. Ed. 2001, 40, 1529–1532.
(41) Yoo, H. S.; Ko, H. H.; Ryu, D. W.; Lee, J. W.; Yoon, J. H.; Lee,
W. R.; Kim, H. C.; Koh, E. K.; Hong, C. S. Inorg. Chem. 2009, 48,
5617–5619.
(42) Dzyaloshinsky, I. J. Phys. Chem. Solids 1958, 4, 241–255.
(43) Moriya, T. Phys. Rev. 1960, 120, 91–98.
(44) Abu-Youssef, M. A. M.; Mautner, F. A.; Vicente, R. Inorg. Chem.
2007, 46, 4654–4659.
(45) Bellitto, C.; Bauer, E. M.; Ibrahim, S. A.; Mahmoud, M. R.;
Righini, G. Chem.—Eur. J. 2003, 9, 1324–1331.
(46) Cao, D.-K.; Li, Y.-Z.; Zheng, L.-M. Inorg. Chem. 2007, 46,
7571–7578.
(47) Fisher, M. E. Am. J. Phys. 1964, 32, 343–346.
(48) Manson, J. L.; Kmety, C. R.; Palacio, F.; Epstein, A. J.; Miller,
J. S. Chem. Mater. 2001, 13, 1068–1073.
(49) Jia, H.-P.; Li, W.; Ju, Z.-F.; Zhang, J. Chem. Commun. 2008,
371–373.
(50) Wang, X.-Y.; Wei, H.-Y.; Wang, Z.-M.; Chen, Z.-D.; Gao, S.
Inorg. Chem. 2005, 44, 572–583.
(51) Yin, P.; Gao, S.; Zheng, L.-M.; Wang, Z.; Xin, X.-Q. Chem.
Commun. 2003, 1076–1077.
(52) Coronado, E.; Delhaes, P.; Gatteschi, D.; Miller, J. S. Magnetic
Phenomena in Molecular Materials; Kluwer Academic Publishers: Dor-
drecht, The Netherlands, 1996.
(53) Coronado, E.; Gomez-Garcia, C. J.; Nuez, A.; Romero, F. M.;
Waerenborgh, J. C. Chem. Mater. 2006, 18, 2670–2681.
(54) Drillon, M.; Panissod, P. J. Magn. Magn. Mater. 1998, 188,
93–99.
(55) Fujita, W.; Awaga, K. Inorg. Chem. 1996, 35, 1915–1917.
(56) Rabu, P.; Rouba, S.; Laget, V.; Hornick, C.; Drillon, M. Chem.
Commun. 1996, 1107–1108.
10785
dx.doi.org/10.1021/ic2013232 |Inorg. Chem. 2011, 50, 10777–10785