Synthesis, crystal structures and magnetic properties of 1D polymeric [MnIII(salen)N3] and [MnIII(salen)Ag(CN)2] complexes

Article abstract of DOI:10.1039/b200384h

The syntheses, crystal structures and variable temperature (5-300 K) magnetic susceptibility measurements of two salen complexes of manganese(III) [where H₂salen = N,N′-bis(salicylidene)-1,2-diaminoethane] having azido and dicyano argentate(I) as bridging ligands are reported. Both complexes belong to one-dimensional systems in which the azido functions as μ-(1,3) and Ag(CN)₂^- as ... NC- Ag-CN ... bridging ligands. The azido-bridged compound is antiferromagnetic with an intrachain interaction constant J = -4.52(4) cm^-1 (Weng model) or J = 35.19(8) cm^-1 (Fisher model), and it is a candidate to observe the Haldane gap in a S = 2 system. The silver cyanide bridged complex shows single ion behaviour of the Mn(III) ion, perhaps in combination with a very weak antiferromagnetic interaction.

Full text of DOI:10.1039/b200384h

View Article Online / Journal Homepage / Table of Contents for this issue
Synthesis, crystal structures and magnetic properties of 1D polymeric
III III
[Mn (salen)N3] and [Mn (salen)Ag(CN)2] complexes
a
Anangamohan Panja, Nizamuddin Shaikh, Pavel Vojtısek, Song Gao and Pradyot Banerjee*
a
b
c
a
ˇ
´
a
Department of Inorganic Chemistry, Indian Association for the Cultivation of Science,
Kolkata 700 032, India
Department of Inorganic Chemistry, Universita Karlova, Albertov 2030, 128 40 Prague 2,
Czech Republic
College of Chemistry and Molecular Engineering, State Key Laboratory of Rare Earth Materials
Chemistry and Applications, Peking University, Beijing 100871, P.R. China
b
c
Received (in Montpellier, France) 8th January 2002, Accepted 2nd April 2002
First published as an Advance Article on the web 8th July 2002
The syntheses, crystal structures and variable temperature (5–300 K) magnetic susceptibility measurements of
two salen complexes of manganese(III) [where H2salen ¼ N,N0-bis(salicylidene)-1,2-diaminoethane] having
azido and dicyano argentate(I) as bridging ligands are reported. Both complexes belong to one-dimensional
ꢀ
systems in which the azido functions as m-(1,3) and Ag(CN)2 as ꢁ ꢁ ꢁNC– Ag–CNꢁ ꢁ ꢁ bridging ligands. The
azido-bridged compound is antiferromagnetic with an intrachain interaction constant J ¼ ꢀ4.52(4) cmꢀ1
(Weng model) or J ¼ ꢀ5.19(8) cmꢀ1 (Fisher model), and it is a candidate to observe the Haldane gap in a
S ¼ 2 system. The silver cyanide bridged complex shows single ion behaviour of the Mn(III) ion, perhaps in
combination with a very weak antiferromagnetic interaction.
The azide ion, N3ꢀ, can coordinate to metal ions by different
bridging modes1–3 viz., m-(1,3), mm-(1,1) along with a rarely
occurring one, m-(1,1,1). The cis and trans arrangement of
the bridged azide is also noteworthy. The polymeric chains
of Mn(III) often exhibit interesting magnetic properties and
could be used as precursors of molecular-based magnets. The
role of manganese-containing enzymes is well-known in biol-
ogy, and these Mn compounds could in some way mimic the
active site of these metalloenzymes. A literature survey reveals
that only a few structurally characterised complexes of Mn(III)
having N3ꢀ bridges have been reported. In 1998 two studies4,5
appeared almost concurrently on the synthesis, crystal struc-
ture and magnetic properties of a one-dimensional azide-
bridged manganese(III) complex, [Mn(salpn)N3] [H2sal-
pn ¼ N,N0-bis(salicylidene)-1,3-diaminopropane]. The Mn(III)
Schiff base complex [Mn(salen)N3] [H2salen ¼ N,N0-bis(salicy-
lidene)-1,2-diaminoethane] has also been reported but was not
characterised by X-ray crystallography6 due to a failure to
obtain suitable crystals for X-ray diffraction. However, based
on the spectral and magnetic studies, it was suggested to have
a linear chain structure identical to that in [Mn(acac)2N3]7
interactions leading to polymeric structures. There are several
reports in which Ag(CN)2ꢀ has been used to construct hetero-
bimetallic polymeric materials of various topologies with dif-
ferent transition metals,10 but to our knowledge there is no
III
I
report on any synthesis of Mn –Ag heterobimetallic poly-
mers. In this paper we wish to report the synthesis, single crys-
tal X-ray structures and variable temperature magnetic
properties of [Mn(salen)N3] and [Mn(salen)Ag(CN)2].
Experimental
Materials and methods
High purity ethylenediamine, salicylaldehyde and sodium
azide were purchased from Aldrich and used as received. All
other reagents were of analytical grade. Elemental analysis
(C, H, N) was carried out using a Perkin–Elmer 240 elemental
analyser. Infrared spectra (400–4000 cmꢀ1) were recorded
from KBr pellets on a Nicolet Magna IR 750 series II FTIR
spectrophotometer. Magnetic measurements were carried out
with a MPMS SQUID magnetometer under 10 and 5 kOe
magnetic fields for 1 and 2, respectively. Diamagnetic correc-
tions were made using Pascal’s constants.
ꢀ
(H2acac ¼ acetylacetone). Here, N3 functions as a m-(1,3)
bridge leading to a polymeric chain.
Like azide, one of the most widely used and versatile anti-
ferromagnetic and ferromagnetic couplers is the cyanide
ligand. Cyanometallates have frequently been employed for
the design of various 1-D, 2-D or 3-D network architectures
and their magnetic properties have been studied in an effort
to prepare molecular-based magnetic materials. The most com-
mon cyanometallates for this purpose are M(CN)6nꢀ(n ¼ 2 or
3; M ¼ Fe, Co, Cr, V, Mn)8 and [M(CN)4]2ꢀ (M ¼ Ni, Pt, Cd,
Zn, Cu).9 Linear metallocyanates may be employed to design
various types of polymeric materials.Being interested in con-
structing heterobimetallic polymeric materials, we have
Synthesis
Caution! Since the azide complexes of metal ions are poten-
tially explosive, only small amounts of the materials should
be handled with care.
III
[Mn (salen)(l1,3-N3)] (1). This compound was conveniently
11
III
synthesised by reaction of [Mn (salen)OAc]ꢁH2O
and
sodium azide from a methanol–water mixture. Sodium azide
(0.065 g, 1 mmol) was dissolved in 40 cm3 of water in a 100
cm3 beaker. To this solution was slowly added 40 cm3 of a
ꢀ
employed Ag(CN)2 as a building block. In addition to serv-
ing as a cyanide bridge, it can also participate in AgIꢁ ꢁ ꢁAgI
III
methanolic solution of [Mn (salen)OAc]ꢁH2O (0.4 g, 1 mmol).
DOI: 10.1039/b200384h
New J. Chem., 2002, 26, 1025–1028
1025
This journal is # The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2002

View Article Online
Upon standing at room temperature for 2 h, brown-black crys-
tals separated out. These were filtered, washed first with the
mother liquor, followed by diethyl ether (30 cm3), and dried
in vacuo over calcium chloride. Yield 80%. Anal. calcd for
C16H14N5O2Mn: C, 52.89; H, 3.86; N, 19.28; found: C,
52.65; H, 3.92; N, 19.20.
bridged azido. The corresponding vibration of the C–N bond
of the salen ligand is at 1626 cmꢀ1. For 2 the asymmetric
stretching vibration for C N is observed at 2140 cmꢀ1, which
=
=
is close to that reported for several other metal cyanide
bridges.10
3
III
[Mn (salen)Ag(CN)2] (2). To an aqueous solution (20 cm )
of potassium dicyanoargentate(I), [KAg(CN)2] (0.2 g, 1 mmol),
was slowly added a methanolic solution (40 cm3) of [Mn(sale-
n)OAc]ꢁH2O (0.4 g, 1 mmol) in a 100 cm3 beaker. Upon stand-
ing for 3 days at room temperature, black crystals separated
out. These were filtered, washed successively with mother
liquor and diethyl ether, and dried in vacuo. Yield 80%. Anal.
calcd for C18H14N4O2MnAg: C, 44.90; H, 2.91; N, 11.64;
found: C, 44.82; H, 3.02; N, 11.69.
Description of the structures of complexes 1 and 2
Complex 1 features an azido-bridged Mn(III) chain as shown in
Fig. 1. The Mn(III) ion assumes a slightly distorted octahedral
geometry in which the metal ion binds to one salen in the equa-
torial mode and two azide ions in axial positions. Each azide
functions as a trans-m-(1,3) bridge, resulting in a one-dimen-
sional polymer. Based on the classification by Escuer et al.,15
it can be categorised as a type I azido chain. In the equatorial
plane the Mn–O(1), Mn–O(2), Mn–N(1) and Mn–N(2) bond
˚
lengths are 1.875(15), 1.888(16), 1.989(17) and 1.978(18) A,
respectively, which are comparable to those observed in other
Mn(III) salen complexes. The O(1) and N(2) atoms are 0.041
Crystal data collection and refinement
Suitable single crystals of the complexes 1 and 2 were mounted
on an Enraf–Nonious CAD4 diffractometer with graphite-
˚
and 0.043 A above the mean plane whereas the O(2) and
˚
˚
N(1) are 0.041 and 0.043 A below the mean plane. The central
monochromated Mo-Ka radiation (l ¼ 0.71070 A) at 293 K.
˚
Mn(III) ion itself is, however, 0.019 A above the mean equator-
Information regarding the crystallographic data collection
and refinement of the structures is complied in Table 1. Unit
cell parameters and orientation matrixes were determined by
a least-squares fit. Intensity data were corrected for Lorentz
polarisation but not for absorption. All calculations for data
reduction, structure solution and refinement were done by
standard procedures (SHELXS97),12 (SHELXL97).13 For 2,
our programme for data reduction merges reflections if the
absorption correction is not made. These data are input in
SHELX, which cannot calculate a value for Rint in this case.
All non-hydrogen atoms were refined anisotropically by full-
matrix least-squares techniques on F2. All the hydrogen atoms
were located in Fourier-difference maps and refined isotropi-
cally. The final crystal structures were drawn using the ORTEP
programme.14
ial plane. The coordination of the azide bridge is nearly sym-
metrical: Mn(1)–N(11)–N(12) ¼ 118.42(16)ꢂ and Mn(1a)–
N(13)–N(12) ¼ 114.64(16)ꢂ. Owing to Jahn–Teller distortion,
the axial Mn–N(11) and Mn–N(13) distances of 2.280 and
˚
2.334 A, respectively, are considerably longer. The intrachain
˚
Mnꢁ ꢁ ꢁMn distance is 5.56 A. However, the interchain
˚
Mnꢁ ꢁ ꢁMn separation is considerably larger at 9.99 A. Selected
bond distances and angles are listed in Table 2.
The coordination environment of the Mn(III) ion in complex
2 with atom numbering scheme is shown in Fig. 2. The coordi-
nation environment of each manganese atom is best described
by a distorted octahedral geometry. Two nitrogen atoms and
two oxygen atoms of the Schiff base ligand [having Mn–
O(1), Mn–O(2), Mn–N(1) and Mn–N(2) distances of
˚
CCDC reference numbers 188032–188033. See http://
www.rsc.org/suppdata/nj/b2/b200384h/ for crystallographic
files in CIF or other electronic format.
1.886(3), 1.874(3), 1.985(4), 1.990(4) A, respectively] define
the equatorial plane around the Mn atom. The axial sites are
occupied by two nitrogen atoms of the bridging [N–C–Ag–
C–N] moiety and show bond distances Mn–N(98) ¼ 2.321(4)
˚
and Mn–N(99) ¼ 2.286(4) A, respectively. The central Mn(III)
˚
ion is slightly above the equatorial plane (0.006 A). The O(1)
Results and discussion
˚
and N(2) atoms are 0.037 and 0.039 A above the mean plane
The IR spectrum of 1 shows a very strong band at around 2027
cmꢀ1, assigned to the asymmetric stretching vibrations of the
˚
whereas the O(2) and N(1) atoms are 0.037 and 0.039 A,
Table 1 Data collection and structure refinement for complexes 1
and 2
Empirical formula
Formula weight
Crystal system
Space group
C16H14N5O2Mn (1)
363.26
C18H14N4O2AgMn (2)
481.14
Orthorhombic
Pca21 (no. 29)
10.9390(3)
12.9470(2)
11.1140(4)
90
Monoclinic
P21/c (no. 14)
8.148(5)
6.509(5)
33.158(5)
96.346(5)
1747.8(17)
4
˚
a/A
˚
b/A
˚
c/A
b/ꢂ
U/A
3
˚
1574.05(8)
4
0.858
Z
m(Mo-Ka) /mmꢀ1
T/K
1.863
293
293
20 584
Total data
Unique data
Rint
2802
2802
3601
0.043
Data [I > 2s(I)]
R [I > 2s(I)]
wRa [I > 2s(I)]
3193
0.0295
0.0727
2292
0.0348
0.0888
w ¼ 1/[s2(Fo2) + (0.0403 P)2 + 0.1134 P] where P ¼ (Fo2 + 2 Fc2)/
3 for 1; w ¼ 1/[s2(Fo2) + (0.0336 P)2 + 2.15 P] where P ¼ (Fo2 + 2
Fc2)/3 for 2.
a
Fig. 1 ORTEP view of 1, showing the atom labeling scheme. Hydro-
gen atoms are omitted for clarity and thermal ellipsoids are repre-
sented at the 30% probability level.
1026
New J. Chem., 2002, 26, 1025–1028

View Article Online
ꢂ
˚
Table 2 Selected bond distances (A) and angles ( ) for complexes 1
and 2a
Magnetic properties
The thermal variation of the molar susceptibility wM shows a
broad maximum around ca. 43 K for 1, which is indicative
of antiferromagnetic coupling within a chain-type structure
(Fig. 3). Using the expression for an antiferromagnetic one-
dimensional S ¼ 2 chain given by Weng and modified by Hil-
ler,16 a reasonably good fit is possible from room temperature
down to 5 K. The J values were obtained by minimising the
Complex 1
Mn–O(1)
Mn–N(1)
Mn–N(11)
1.8754(15)
1.9885(17)
2.280(2)
94.32(7)
173.73(7)
91.50(7)
91.85(7)
89.80(7)
93.38(7)
87.06(7)
176.73(8)
Mn–O(2)
Mn–N(2)
Mn–N(13a)
1.8878(16)
1.9780(18)
2.334(2)
91.73(7)
91.77(7)
172.99(7)
89.99(7)
82.20(7)
86.49(7)
89.68(7)
O(1)–Mn–O(2)
O(1)–Mn–N(2)
O(1)–Mn–N(13a)
O(2)–Mn–N(2)
O(2)–Mn–N(13a)
N(1)–Mn–N(11)
N(2)–Mn–N(11)
N(11)–Mn–N(13a)
Complex 2
O(1)–Mn–N(1)
O(1)–Mn–N(11)
O(2)–Mn–N(1)
O(2)–Mn–N(11)
N(1)–Mn–N(2)
N(1)–Mn–N(13a)
N(2)–Mn–N(13a)
P
P
calc
function R ¼ (wM ꢀ wMobs)2/ (wMobs)2. The best-fit
parameters are: J ¼ ꢀ4.52(4) cmꢀ1
,
g ¼ 2.006(8) and
R ¼ 4.5 ꢃ 10ꢀ4
Except at very low temperatures where some anisotropy
may be significant, manganese(III) is expected to be Heisen-
berg-like in its magnetic properties6 We, therefore, also made
an attempt to explain the magnetism in terms of the Heisen-
berg 1D chain using the Fisher17 expression:
Ag–C(99)
Mn–O(1)
2.067(5)
1.886(3)
Ag–C(98b)
Mn–O(2)
2.070(5)
1.874(3)
ꢀ
ꢁ
Nb2g2
3kT
1 þ U
1 ꢀ U
w ¼
SðS þ 1Þ
Mn–N(1)
Mn–N(98)
1.985(4)
2.321(4)
Mn–N(2)
Mn–N(99)
1.990(4)
2.286(4)
ꢀ2JSðS þ 1Þ ꢀ2JSðS þ 1Þ
C(98b)–Ag–C(99)
O(1)–Mn–N(1)
O(1)–Mn–N(98)
O(2)–Mn–N(1)
O(2)–Mn–N(98)
N(1)–Mn–N(2)
N(1)–Mn–N(99)
N(2)–Mn–N(99)
155.00(17)
91.59(13)
95.18(12)
173.73(14)
93.80(14)
82.77(14)
87.23(13)
83.45(13)
O(1)–Mn–O(2)
O(1)–Mn–N(2)
O(1)–Mn–N(99)
O(2)–Mn–N(2)
O(2)–Mn–N(99)
N(1)–Mn–N(98)
N(2)–Mn–N(98)
N(98)–Mn–N(99)
94.14(12)
174.01(13)
94.28(12)
91.58(13)
94.86(14)
83.14(13)
86.19(13)
166.66(13)
U ¼ coth
ꢀ
kT
kT
Fitting of the observed data gave best-fit parameters as:
.
J ¼ ꢀ5.19(8) cmꢀ1, g ¼ 2.05(1), and R ¼ 9.4 ꢃ 10ꢀ4
Owing to the small axial overlap, the manganese(III) azide-
bridged linear chain having long axial bonds has only one
dominant exchange pathway, and it results in a weak antifer-
romagnetic interaction. The only magnetic orbital that can
make significant contribution to the coupling is the one derived
from the Mn(III) dz2 orbital. The axial ligands will provide
mainly a s-type superexchange pathway. A superexchange
mechanism involving the p orbitals of the bridging group
may well be conceived. The intrachain Mnꢁ ꢁ ꢁMn distance
a
Symmetry codes: (a) 2 ꢀ x, y, 12 + z for 1; (b) 1 + x, ꢀ1 + y, z for 2.
1
respectively, below the plane. The bridging Mn–N(98)–N(98)
and Mn–N(99)–N(99) angles are 170.4(3)ꢂ and 144.4(3)ꢂ,
respectively. The [N–C–Ag–C–N] moiety is almost linear and
forms endless chains involving the Mn (salen) moiety. They
are parallel with diagonal directions a, b and the distance
˚
˚
(5.56 A) in 1 is comparable to that observed (5.63 A) in [Mn(a-
cac)2N3]18. The magnitudes of the exchange interaction in
these two complexes are also close. The J values obtained
for 1 are comparable to those reported4,5 for [Mn(salpn)N3].
The two values (J ¼ ꢀ5.19 and ꢀ4.52 cmꢀ1) obtained by us
using two different models are virtually identical to those
reported by Kennedy and Murray6 (J ¼ ꢀ5.42 and ꢀ4.45
cmꢀ1) through powder susceptibility and magnetisation mea-
surements.
III
˚
Agꢁ ꢁ ꢁAg is about 6.5 A. No direct Agꢁ ꢁ ꢁAg interaction can
be conceived. Main bond distances and angles are listed in
Table 2.
It may be pointed out that a sharp increase in susceptibility
at temperatures below 15 K was observed in [Mn(salpn)N3].4,5
No such behaviour is observed in the title [Mn(salen)N3] com-
pound, ruling out long-range order or impurity. To the best of
our knowledge, just one experimental example of a Haldane
gap system with S ¼ 2 has been documented, in 1996,19
although several numerical studies have been performed since
Haldane’s conjecture in 1983.20 All theoretical work predicts
Fig. 2 ORTEP view of 2, showing the atom labeling. Hydrogen
atoms are omitted for clarity and thermal ellipsoids are represented
at the 30% probability level.
Fig. 3 Variable-temperature magnetic susceptibility data for 1.
Experimental points are represented by (X). (—) data fitted by Fisher
model. (L) data fitted by Weng model.
New J. Chem., 2002, 26, 1025–1028
1027

View Article Online
in comparison to the silver cyanide group. It is quite likely that
the Ag(I) ion inhibits the magnetic interaction between the
manganese(III) centres in 2, leading to a very weak antiferro-
˚
magnetic interaction. The long Mnꢁ ꢁ ꢁMn distance (10.43 A)
in 2 would also be quite unfavorable for a more direct
pathway.
Acknowledgements
This work was partly supported by the Council of Scientific
and Industrial Research, India, and the National Science Fund
for Distinguished Young Scholars in China (20125104).
References
Fig. 4 Field dependence of magnetisation of 1 at 1.83 K; the line is
calculated using the Brillion function for an isolated Mn(III) ion.
1
2
3
(a) J. Ribas, M. Monfort, B. K. Ghosh, R. Cortes, X. Solans and
M. Font-Bardia, Inorg. Chem., 1996, 35, 864; (b) F. Tuczek and
W. Bensch, Inorg. Chem., 1995, 34, 1482.
(a) A. F. Mautner and M. A. S. Goher, Polyhedron, 1996, 15,
1133; (b) D. De Munno, T. Poerio, G. Viau, M. Julve and F.
Lloret, Angew. Chem., Int. Ed. Engl., 1997, 36, 1459.
M. W. Wemple, D. M. Adams, K. S. Hagen, K. Folting, D. N.
Hendrickson and G. Christou, J. Chem. Soc.,Chem. Commun.,
1995, 1591.
4
5
H. Li, Z. J. Zhong, C.-Y. Duan, X.-Z. Yu, T. C. W. Mak and B.
Wu, Inorg. Chim. Acta, 1998, 271, 99.
K. R. Reddy, M. V. Rajasekharan and J.-P. Tuchagues, Inorg.
Chem., 1998, 37, 5978.
6
7
8
B. J. Kennedy and K. S. Murray, Inorg. Chem., 1985, 24, 1552.
A. K. Gregson and N. T. Moxon, Inorg. Chem., 1982, 21, 586.
(a) M. Ohba, N. Usuki, N. Fukita and H. Okawa, Angew. Chem.,
Int. Ed., 1999, 38, 1795; (b) H. Z. Kou, S. Gao, B. Q. Ma and
D. Z. Liao, Chem. Commun., 2000, 713; (c) A. Marvilliers, S.
Parsons, E. Riviere, J.-P. Audiere and T. Mallah, Chem. Com-
mun., 1999, 2217; (d ) J. A. Smith, J.-R. Galan-Mascaros, R.
Clerac and K. R. Dunbar, Chem. Commun., 2000, 1077; (e)
M. S. El Fallah, E. Rentschler, A. Caneschi, R. Sessoli and D.
Gatteschi, Angew. Chem., Int. Ed. Engl., 1996, 35, 1947.
(a) L. R. Falvello and M. Tomas, Chem. Commun., 1999, 273; (b)
F. Olbrich, J. Kopf and E. Weiss, J. Organomet. Chem ., 1993,
456, 293.
Fig. 5 Variable-temperature magnetic susceptibility data for 2.
Experimental points are represented by (K). (—) data fitted by the
Curie–Weiss Law.
9
10 (a) T. Soma and T. Iwamoto, Inorg. Chem., 1996, 35, 1849; (b) T.
Soma, H. Yuge and T. Iwamoto, Angew. Chem., Int. Ed. Engl.,
1994, 33, 1665; (c) C. Kappenstein, A. Onli, M. Guerin, J. Cernak
and J. Chomic, Inorg. Chim. Acta., 1988, 147, 189; (d ) J. Cernak,
F. Gerard and J. Chomic, Acta Crystallogr., Sect. C, 1993, 49,
1294
11 A. van den Bergen, K. S. Murray, M. J. O’Connor and B. O.
West, Aust. J. Chem., 1969, 22, 39.
12 SHELXS 97: G. M. Sheldrick, Acta Crystallogr., Sect. A, 1990,
46, 467.
13 G. M. Sheldrick, SHELXL 97, Program for Crystal Structure
Refinement, University of Go¨ttingen, Germany, 1997.
14 C. K. Johnson and M. N. Burnett, ORNL-6895, Oak Ridge
National Laboratory, Oak Ridge TN, USA, 1996.
15 A. Escuer, R. Vicente, M. S. E. Fallah, J. Ribas and X. J. Solans,
J. Chem. Soc., Dalton Trans., 1993, 2973.
that the gap between the singlet ground state and the first
excited triplet state Eg for an S ¼ 2 chain is correlated with
the intrachain antiferromagnetic interaction J as Eg ¼ (0.05–
0.09)J. The J for the title S ¼ 2 chain, a candidate for obser-
ving a Haldane gap, is ca. 7 K; therefore, Eg may be estimated
at ca. 0.5 K. The field dependence of the magnetisation at 1.83
K (Fig. 4) shows a very low value, even at 70 kOe, compared
with the calculated value using a Brillion function for an iso-
lated Mn(III) ion with S ¼ 2, g ¼ 2.0, also suggesting a singlet
ground state; however, the critical field (Hc) to a gapless state
was not observed up to 70 kOe. Apparently, more experimen-
tal evidence is needed, especially low-temperature and high-
field investigations.
16 (a) C. Y. Weng, Ph. D. Thesis, Carnegie Institute of Technology,
USA, 1968; (b) W. Hiller, J. Strahle, A. Datz, M. Hanack, W. E.
Hatfield, L. W. ter Haar and P. Guetlich, J. Am. Chem. Soc.,
1984, 106, 329.
17 G. R. Wagner and S. A. Friendberg, Phys. Lett., 1964, 9, 11.
18 B. R. Stults, R. S. Marianelli and V. W. Day, Inorg. Chem., 1975,
14, 722.
19 G. E. Granroth, M. W. Meisel, M. Chaparala, T. Jolicoeur, B. H.
Ward and D. R. Talham, Phys. Rev. Lett., 1996, 77, 1616.
20 (a) F. D. M. Haldane, Phys. Rev. Lett., 1983, 50, 1153; (b) S.
Yamamoto, Phys. Rev. Lett., 1995, 75, 3348; (c) X. Wang, S.
Qin and L. Yu, Phys. Rev. B, 1999, 60, 14 529; (d ) S. Todo and
K. Kato, Phys. Rev. Lett., 2001, 87, 47 203.
Complex 2, on the other hand, shows single ion behaviour
mainly, perhaps combined with a very weak antiferromagnetic
coupling between Mn(III) ions (Fig. 5). The magnetic suscept-
ibility obeys the Curie–Weiss law [wM ¼ C/(T ꢀ y)] with
C ¼ 2.994(3) cm3 molꢀ1 K and y ¼ ꢀ1.4(1) K. The Curie con-
stant C corresponds to one Mn(III) ion with S ¼ 2 and
g ¼ 1.998. The slight decrease of wMT below 30 K and the
negative Weiss constant y suggest a very weak, if it exists at
all, antiferromagnetic interaction between Mn(III) ions.
The stronger coupling observed for 1 than in 2 may arise
from a more symmetrical orbital system in the azide bridge
1028
New J. Chem., 2002, 26, 1025–1028