Synthesis, structural characterization, magnetic behavior, and single crystal EPR spectra of three new one-dimensional manganese azido systems with FM, alternating FM-AF, and AF coupling

Article abstract of DOI:10.1021/ic990656x

Reaction of sodium azide with manganese(II) and pyridine derivatives such as 2-bzpy, 3-bzpy, and 3,5lut (2-benzoylpyridine, 3-benzoylpyridine, and 3,5-dimethylpyridine, respectively) led to the one-dimensional systems cis-[Mn(2-bzpy)(N₃)₂]_n (1), trans-[Mn(3-Bzpy)₂(N₃)₂]_n (2), and cis-[Mn(3,5lut)₂(N₃)₂]_n (3). Compound 1 crystallizes in the P2₁/n group: a = 14.413(4) ?, b = 16.157(4) ?, c = 18.478(5) ?, and Z = 12. Compound 2 crystallizes in the C2/c group: a = 14.179(5) ?, b = 9.698,(4) ?, c = 34.351(12) ?, and Z = 8. Compound 3 crystallizes in the P2₁/n group: a = 13.552(6) ?,b = 7.730(3) ?, c = 16.554(6) ?, and Z = 4. Structural determination shows a chain with double μ_1,1 azido bridges for 1, an alternating system with double μ_1,1 and μ_1,3 azido bridges for 2, and finally a chain with double μ_1,3 azido bridges for 3. Susceptibility data show ferromagnetic coupling for 1, antiferromagnetic coupling for 3, and alternating ferro-antiferromagnetic interactions for 2. EPR data measured on powdered samples and single crystals show the ideal Heisenberg narrowing angular dependence for the three kinds of coupled systems and an increasing dipolar broadening mechanism in the order 3 < 2 < 1 according to the decreasing Mn?Mn intrachain distances.

Full text of DOI:10.1021/ic990656x

5716
Inorg. Chem. 1999, 38, 5716-5723
Synthesis, Structural Characterization, Magnetic Behavior, and Single Crystal EPR Spectra
of Three New One-Dimensional Manganese Azido Systems with FM, Alternating FM-AF,
and AF Coupling
Morsy A. M. Abu-Youssef,† Albert Escuer,*,‡ Dante Gatteschi,§ Mohamed A. S. Goher,|
Franz A. Mautner, and Ramon Vicente‡
Chemistry Department, Faculty of Science, Alexandria University, Egypt, Departament de Qu´ımica
Inorga`nica, Universitat de Barcelona, Diagonal 647, 08028 Barcelona, Spain, Dipartimento di Chimica,
Universita´ degli Studi di Firenze, Via Maragliano 75-77, 50144 Firenze, Italy, Chemistry Department,
Faculty of Science, Kuwait University, P. O. Box 5969, Safat, 13060 Kuwait, and Institut fu¨r
Physikalische und Theoretische Chemie, Technische Universita¨t Graz, A-8010 Graz, Austria
ReceiVed June 7, 1999
Reaction of sodium azide with manganese(II) and pyridine derivatives such as 2-bzpy, 3-bzpy, and 3,5lut (2-
benzoylpyridine, 3-benzoylpyridine, and 3,5-dimethylpyridine, respectively) led to the one-dimensional systems
cis-[Mn(2-bzpy)(N3)2]n (1), trans-[Mn(3-Bzpy)2(N3)2]n (2), and cis-[Mn(3,5lut)2(N3)2]n (3). Compound 1 crystallizes
in the P21/n group: a ) 14.413(4) Å, b ) 16.157(4) Å, c ) 18.478(5) Å, and Z ) 12. Compound 2 crystallizes
in the C2/c group: a ) 14.179(5) Å, b ) 9.698(4) Å, c ) 34.351(12) Å, and Z ) 8. Compound 3 crystallizes
in the P21/n group: a ) 13.552(6) Å, b ) 7.730(3) Å, c ) 16.554(6) Å, and Z ) 4. Structural determination
shows a chain with double µ1,1 azido bridges for 1, an alternating system with double µ1,1 and µ1,3 azido bridges
for 2, and finally a chain with double µ1,3 azido bridges for 3. Susceptibility data show ferromagnetic coupling
for 1, antiferromagnetic coupling for 3, and alternating ferro-antiferromagnetic interactions for 2. EPR data
measured on powdered samples and single crystals show the ideal Heisenberg narrowing angular dependence for
the three kinds of coupled systems and an increasing dipolar broadening mechanism in the order 3 < 2 < 1
according to the decreasing Mn‚‚‚Mn intrachain distances.
Introduction
of the azido bridge was found to be µ1,3-N3 (end-to-end, EE),
but it is not rare in these kinds of systems to find µ1,1-N3 (end-
on, EO) coordination mode simultaneously in the same com-
pound, allowing alternated systems.2-4,6,8,9 No correlation has
so far been found between the properties of the L ligand and
the coordination mode of the azido bridge. The magnetic
behavior of these compounds follows the same general trends
observed in either copper or nickel(II) systems; i.e., EE bridges
lead to antiferromagnetic interaction,7 and EO bridges give
ferromagnetic coupling.11 Systematic synthesis and magnetic
study of pyridine derivatives have recently led to the charac-
terization of low-temperature magnetically ordered phases with
critical temperatures ranging between 16 and 40 K for several
molecular-based magnets of general formula [Mn(L)2(N3)2] (L
) py, 3- and 4-acpy, 3-Etnic, 4-CNpy).2,3
Following our research on compounds of this kind, we studied
the derivatives obtained by reaction of the ligands L ) 2- and
3-benzoylpyridine (2-bzpy and 3-bzpy) and 3,5-dimethylypy-
ridine (3,5lut) with manganese(II) and sodium azide. From these
reactions three new monodimensional systems cis-[Mn(2-bzpy)-
(N3)2]n (1), trans-[Mn(3-bzpy)2(N3)2]n (2), and cis-[Mn(3,5lut)2-
(N3)2]n (3) were structurally characterized. The three compounds
Recent research on the manganese(II)-(L)-azido-bridged
system, L ) N-aromatic ligands, is providing an exceptional
number of materials with new structural and magnetic features.
When L is a pyridine ligand, infinite arrays characterized by
different dimensionality can be obtained: 3-D for L ) pyridine
(py),1,2 2-D for L ) 3- and 4-acetylpyridine (3-acpy3 and
4-acpy4), methyl and ethyl isonicotinate (Meinc5 and Etnic2),
ethyl nicotinate (Enic6), or 4-cyanopyridine (4-CNpy3), and
finally 1-D for 2-hydroxopyridine (pyOH), 3-ethylpyridine
(3Etpy),7 or 3-ethyl-4-methylpyridine (3Et-4Mepy).8 Other 2-D
or 3-D networks have been described with 4,4′-bipyridine9 or
bipyrimidine10 ligands. The most common coordination mode
* To whom correspondence should be addressed. E-mail: aescuer@
kripto.qui.ub.es.
† Alexandria University.
‡ Universitat de Barcelona.
§ Universita` degli Studi di Firenze.
| Kuwait University.
Technische Universita¨t Graz.
(1) Goher, M. A. S.; Mautner, F. A. Croat. Chem. Acta 1990, 63, 559.
(2) Escuer, A.; Vicente, R.; Goher, M. A. S.; Mautner, F. A. Inorg. Chem.
1996, 35, 6386.
(3) Escuer, A.; Vicente, R.; Mautner, F. A.; Goher, M. A. S. Inorg. Chem.
1997, 36, 3440.
(4) Escuer, A.; Vicente, R.; Goher, M. A. S.; Mautner, F. A. Inorg. Chem.
1995, 34, 5707.
(5) Escuer, A.; Vicente, R.; Goher, M. A. S.; Mautner, F. A. J. Chem.
Soc., Dalton Trans. 1997, 4431.
(6) Goher, M. A. S.; Al-Salem, N. A.; Mautner, F. A. J. Coord. Chem.
1998, 44, 119.
(9) Shen, H. Y.; Liao, D. Z.; Jiang, Z. H.; Yan, S. P.; Sun, B. W.; Wang,
G. L.; Yao, X. K.; Wang, H. G. Chem. Lett. 1998, 469.
(10) (a) De Munno, G.; Julve, M.; Viau, G.; Lloret, F.; Faus, J.; Viterbo,
D. Angew. Chem., Int. Ed. Engl. 1996, 35, 1807. (b) Corte´s, R.;
Lezama, L.; Pizarro, J. L.; Arriortua, M. I.; Rojo, T. Angew. Chem.,
Int. Ed. Engl. 1996, 35, 1810.
(7) Escuer, A.; Vicente, R.; Goher, M. A. S.; Mautner, F. A. Inorg. Chem.
1998, 37, 782.
(8) Abu-Youssef, M.; Escuer, A.; Goher, M. A. S.; Mautner, F. A.;
Vicente, R. Eur. J. Inorg. Chem. 1999, 687.
(11) (a) Corte´s, R.; Pizarro, J. L.; Lezama, L.; Arriortua, M. I.; Rojo, T.
Inorg. Chem. 1994, 33, 2697. (b) During the refereeing process, a
compound similar to 1, was published: Manson, J. L.; Arif, A. M.;
Miller, J. S. Chem. Commun. 1999, 1479.
10.1021/ic990656x CCC: $18.00 © 1999 American Chemical Society
Published on Web 11/23/1999

Three New One-Dimensional Manganese Azido Systems
Inorganic Chemistry, Vol. 38, No. 25, 1999 5717
Table 1. Crystal Data and Structure Refinement for
[Mn(2-Bzpy)(N3)2]n (1), [Mn(3-Bzpy)2(N3)2]n (2), and
[Mn(3,5-lut)2(N3)2]n (3)
consist of well-isolated chains in which the manganese atoms
are bridged by two azido ligands, in the EO mode for 1,
alternating EO-EE modes for 2, and EE mode for 3, showing
the expected ferromagnetic (1), alternating ferro-antiferromag-
netic (2), and antiferromagnetic (3) coupling. It should be
underlined that compound 1 is the first example11b characterized
to date of one-dimensional manganese-azido systems with only
EO bridges. These systems are almost ideal for the investigation
of spin dynamics via EPR spectroscopy. In fact they provide
the possibility of finding all the features of a quasi-ideal
Heisenberg system and enable the response of unprecedented
manganese ferromagnetic and alternated ferro-antiferromag-
netic systems to be explored.
1
2
3
chem formula
formula weight
space group
a, Å
b, Å
c, Å
C12H9MnN7O C24H18MnN8O2 C14H18MnN8
322.20
P21/n
14.413(4)
16.157(4)
18.478(5)
90.0
505.40
C2/c
14.179(5)
9.698(4)
34.351(12)
90.0
353.30
P21/n
13.552(6)
7.730(3)
16.554(6)
90.0
101.69(3)
90.0
1698.2(12)
4
R, deg
â, deg
96.77(3)
90.0
4273(2)
12
97.81(3)
90.0
4680(3)
8
γ, deg
V, Å3
Z
In this paper we present the synthesis, structural characteriza-
tion, magnetic behavior, and selected EPR data (mainly room-
temperature measurements) of these fascinating new one-
dimensional systems.
T, K
293(2)
0.710 69
1.503
295(2)
0.710 69
1.435
0.603
0.0555
0.0928
295(2)
0.710 69
1.382
λ(Mo KR), Å
d
calc, Mg‚m-3
µ(Mo KR), mm-1 0.936
0.789
Ra
Rw
0.0669
0.1501
0.0617
0.1225
2 b
Experimental Section
a R(Fo) ) ∑Fo - Fc/∑Fo. b (Rw(Fo)2 ) {∑[w((Fo)2 - (Fc)2)2]/
∑[w((Fo)2)2]}1/2
[Mn(2-Bzpy)(N3)2]n (1) was synthesized by mixing an aqueous-
methanolic solution (1:1, 20 mL) of manganese(II) nitrate tetrahydrate
(1.00 g, ca. 4 mmol) and 1.83 g (ca. 10 mmol) of 2-benzoylpyridine
dissolved in 15 mL of methanol, followed by dropwise addition of a
concentrated aqueous solution of sodium azide (0.65 g, 10 mmol). The
resulting yellow-brown mixture was left to stand in a desiccator in a
dark place for several weeks to form long deep red crystals of 1 suitable
for X-ray determination; yield, ca. 75%. Anal. Calcd for MnC12H9N7O:
C, 44.74; H, 2.82; N, 30.43; Mn, 17.05. Found: C, 44.5; H, 2.7; N,
30.5; Mn, 17.0.
[Mn(3-Bzpy)2(N3)2]n (2) was synthesized by mixing an aqueous-
methanolic solution (1:1, 30 mL) of manganese(II) chloride tetrahydrate
(0.59 g, 3 mmol) and 1.10 g (ca. 6 mmol) of 3-benzoylpyridine
dissolved in 20 mL of methanol, followed by dropwise addition of a
concentrated aqueous solution of sodium azide (0.65 g, 10 mmol). A
few drops of L-ascorbic acid in methanol were added to the resulted
yellow mixture in order to prevent oxidation of Mn(II). The clear
solution was left to stand in a dark place for several days to form light
yellow platelike crystals of 2 suitable for X-ray determination; yield,
ca. 60%. Anal. Calcd for MnC24H18N8O2: C, 57.04; H, 3.59; N, 22.17;
Mn, 10.87. Found: C, 56.9; H, 3.5; N, 22.3; Mn, 10.7.
.
centering of 36 reflections (8.9° < θ < 15.3°), 33 reflections (8.0° <
θ < 19.0°), and 55 reflections (7.5° < θ < 14.9°) and refined by least-
squares methods; 8230 reflections (7085 independent reflections, Rint
) 0.0386), 4921 reflections (4229 independent reflections; Rint
0.0299), and 4569 reflections (3685 independent reflections; Rint
)
)
0.0320) were collected in the ranges 2.75° < θ < 26.00°, 2.90° < θ <
26.00°, and 2.92° < θ < 27.99°. Intensity decays of 12, 2, and 3% for
control reflections (2 0 0; 0-3 3; 3 2-3), (1-1-6; 3 1-1; 2-2 1),
and (-2-1 2; 0 3 2), measured after every set of 100 reflections, were
observed during data collection. Corrections were applied for Lorentz-
polarization effects, for intensity decay, and for absorption using the
DIFABS12 computer program. The structures were solved by Patterson
and direct methods using the SHELXS-8613 computer program, and
refined by full-matrix least-squares methods on F2 (i.e. squared), using
the SHELXL-9314 program incorporated in the SHELXTL/PC V 5.0315
program library and the graphics program PLUTON.16 All non-
hydrogen atoms were refined anisotropically. The hydrogen atoms were
fixed geometrically with the HFIX utility.15 Final R factors for all
observed reflections are 0.0669, 0.0555, and 0.0617; the number of
refined parameters are 568, 320, and 212, respectively. Maximum and
minimum peaks in the final difference synthesis are as follows: 0.570
and -0.745 e Å-3, 0.209 and -0.218 e Å-3, and 0.299 and -0.440 e
Å-3. Significant bond parameters for 1-3 are given in Tables 2-4,
respectively.
[Mn(3,5-lutidine)2(N3)2]n (3) was prepared in the same way as
compound 1 but using the following quantities of the starting reagents:
1.00 g (4 mmol) of manganese(II) nitrate tetrahydrate, 1.07 g (10 mmol)
of 3,5-lutidine, and 0.65 g (10 mmol) of sodium azide. The resulting
solution was kept in the dark for about 2 weeks to yield large well-
formed greenish-yellow crystals of compound 3; yield, ca. 70%. Anal.
Calcd for MnC14H18N8: C, 47.60; H, 5.14; N, 31.72; Mn, 15.55. Found:
C, 47.7; H, 5.2; N, 31.5; Mn, 15.6.
Results and Discussion
Spectral and Magnetic Measurements. Infrared spectra (400-4000
cm-1) were recorded from KBr pellets on a Nicolet 520 FTIR
spectrophotometer. Magnetic measurements were carried out with a
DSM8 pendulum susceptometer, working in the temperature range
300-4 K. The applied external magnetic field was 1.5 T. Magnetic
measurements on the ferromagnetic compound 1 were performed with
a Quantum Desing instrument with a SQUID detector, working in the
temperature range 300-2 K under an external magnetic field of 100
G. Diamagnetic corrections were estimated from Pascal tables. EPR
spectra were recorded at X-band frequency with a Bruker ES200
spectrometer equipped with an Oxford liquid helium cryostat for
variable-temperature work.
Description of the Structure of cis-[Mn(2-bzpy)(N3)2]n (1).
An ORTEP plot of the basic unit of [Mn(2-bzpy)(N3)2]n is
shown in Figure 1a. The structure consists of neutral chains of
manganese atoms linked by EO azido bridges along the (0 0 1)
direction, Figure 1b. The manganese atoms are octahedrally
coordinated by a chelating 2-benzoylpyridine and four azido
ligands which act as end-on double bridges with the two
neighboring manganese ions. The azido ligands are linear and
show asymmetric N-N distances, close to 1.20/1.14 Å. Three
different manganese atoms, Mn(1), Mn(2), and Mn(3), may be
Crystal Structure Analysis of 1-3. The X-ray single-crystal data
for the three compounds were collected on a modified STOE four-
circle diffractometer. The crystal sizes of 1-3, respectively, are as
follows: 0.55 × 0.28 × 0.22; 0.44 × 0.40 × 0.30; and 0.47 × 0.25 ×
0.25 mm. The crystallographic data, conditions retained for the intensity
data collection, and some features of the structure refinements are listed
in Table 1. Graphite-monochromatized Mo KR radiation (λ ) 0.710 69
Å) with the ω-scan technique was used to collect the data sets. The
accurate unit-cell parameters of 1-3 were determined from automatic
(12) Walker, N.; Stuart, D. Acta Crystallogr. 1983, A39, 158.
(13) Sheldrick, G. M. SHELXS-86, Program for the Solution of Crystal
Structure; University of Gottingen: Gottingen, Germany, 1986.
(14) Sheldrick, G. M. SHELXL-93, Program for the Refinement of Crystal
Structure; University of Gottingen: Gottingen, Germany, 1993.
(15) SHELXTL 5.03 (PC Version), Program Library for the Solution and
Molecular Graphics; Siemens Analytical Instruments Division: Madi-
son, WI, 1995.
(16) Spek, A. L. PLUTON-92; University of Utrecht: Utrecht, The
Netherlands, 1992.

5718 Inorganic Chemistry, Vol. 38, No. 25, 1999
Table 2. Selected Bond Lengths (Å) and Angles (deg) for 1
Abu-Youssef et al.
Table 4. Selected Bond Lengths (Å) and Angles (deg) for 3
Mn(1)-O(1)
Mn(1)-N(1)
Mn(1)-N(11)
Mn(1)-N(21)
Mn(1)-N(41)
Mn(1)-N(51)
Mn(3)-O(3)
Mn(3)-N(3)
Mn(3)-N(41)
Mn(3)-N(51)
Mn(3)-N(61)
Mn(3)-N(61b)
2.308(3)
2.305(4)
2.220(4)
2.197(4)
2.176(4)
2.205(4)
2.286(3)
2.265(4)
2.233(4)
2.198(4)
2.203(4)
2.204(4)
Mn(2)-O(2)
Mn(2)-N(2)
Mn(2)-N(11)
Mn(2)-N(21)
Mn(2)-N(31)
Mn(2)-N(31a)
2.278(3)
2.275(4)
2.176(4)
2.239(4)
2.200(4)
2.201(4)
Mn(1)-N(1)
Mn(1)-N(11)
Mn(1)-N(21)
2.251(3)
2.225(3)
2.220(3)
Mn(1)-N(2)
2.271(3)
2.252(3)
2.245(3)
Mn(1)-N(13a)
Mn(1)-N(23b)
N(1)-Mn(1)-N(2)
N(1)-Mn(1)-N(11)
N(1)-Mn(1)-N(13a)
N(1)-Mn(1)-N(21)
99.3(1) N(11)-Mn(1)-N(21)
87.3(1) N(11)-Mn(1)-N(23b)
89.1(1) N(13a)-Mn(1)-N(21)
173.4(1)
89.4(1)
90.1(1)
88.5(1) N(13a)-Mn(1)-N(23b) 85.0(1)
Mn(1)‚‚‚Mn(2)
Mn(1)‚‚‚Mn(3)
Mn(2)‚‚‚Mn(2a)
Mn(3)‚‚‚Mn(3b)
3.396(1)
3.386(1)
3.364(1)
3.415(1)
N(1)-Mn(1)-N(23b) 173.0(1) N(21)-Mn(1)-N(23b)
N(2)-Mn(1)-N(11) 88.8(1) N(12)-N(11)-Mn(1)
95.3(1)
121.9(3)
N(2)-Mn(1)-N(13a) 171.0(1) N(12)-N(13)-Mn(1b) 125.8(3)
N(2)-Mn(1)-N(21)
N(2)-Mn(1)-N(23b)
86.8(1) N(22)-N(21)-Mn(1)
86.8(1) N(22)-N(23)-Mn(1b) 124.8(3)
122.9(3)
O(1)-Mn(1)-N(1)
O(1)-Mn(1)-N(11)
O(1)-Mn(1)-O(21) 162.1(1) N(31)-Mn(2)-N(31a)
O(1)-Mn(1)-O(41)
O(1)-Mn(1)-O(51)
N(1)-Mn(1)-N(11)
N(1)-Mn(1)-N(21)
70.6(1) N(21)-Mn(2)-N(31)
93.7(1) N(21)-Mn(2)-N(31a) 173.9(1)
97.5(1)
N(11)-Mn(1)-N(13a) 94.9(1)
80.3(2)
70.5(1)
84.8(1)
91.6(1)
96.6(1)
163.7(1)
154.7(2)
95.6(1)
92.7(2)
94.0(2)
78.9(1)
95.9(1)
close to 90° due to the cis coordination, whereas they are roughly
coplanar with the corresponding next-nearest-neighbor rings
along the chain.
90.7(1) O(3)-Mn(3)-N(3)
86.7(1) O(3)-Mn(3)-N(41)
87.1(1) O(3)-Mn(3)-N(51)
92.3(1) O(3)-Mn(3)-N(61)
Description of the Structure of trans-[Mn(3-Bzpy)2(N3)2]n
(2). The structure of 2 is shown in Figure 2a. It consists of
octahedrally coordinated manganese atoms in which the coor-
dination sites are occupied by two 3-benzoylpyridine ligands
in trans arrangement and four azido ligands which form double
bridges between neighboring manganese ions. The azido groups
in the double bridges are alternately in the EE and EO modes,
giving an alternating chain, Figure 2a. The EO azido bridges
show asymmetric N-N distances of 1.182(4)/1.157(4) Å,
whereas the EE bridges are more symmetric, 1.178(4)/1.169-
(4) Å. The Mn2(µ1,3-N3)2 ring is quite regular: bond lengths are
Mn(1)-N(21) ) 2.235(3) Å and Mn(1)-N(23b) ) 2.244(3)
Å, and bond angles are Mn(1)-N(21)-N(22) ) 126.0(2)° and
Mn(1)-N(23b)-N(22)b ) 126.1(2)°. The Mn-(µ1,3-N3)2-Mn
ring shows the typical distortion from the planar to the chair
conformation, evaluated as a δ parameter7 ) 26.6(1)°. The
centrosymmetric Mn2N2 ring is strictly planar, as is usual for
double end-on bridges, and shows a bond angle Mn(1)-N(11)-
Mn(1)a ) 100.4(1)° and bond lengths Mn(1)-N(11) and Mn-
(1)-N(11)a of 2.229(3) and 2.219(3) Å, respectively. The Mn‚
‚‚Mn distances are asymmetric, the longer corresponding to the
EE ring, Mn(1)-Mn(1)b ) 5.229(2) Å, and the shorter to the
EO ring, Mn(1)-Mn(1)a ) 3.416(2) Å.
N(1)-Mn(1)-N(41) 100.5(1) O(3)-Mn(3)-N(61b)
N(1)-Mn(1)-N(51) 157.3(1) N(3)-Mn(3)-N(41)
N(11)-Mn(1)-N(21) 79.5(1) N(3)-Mn(3)-N(51)
N(11)-Mn(1)-N(41) 172.1(1) N(3)-Mn(3)-N(61)
N(11)-Mn(1)-N(51) 93.7(1) N(3)-Mn(3)-N(61b)
N(21)-Mn(1)-N(41) 98.2(1) N(41)-Mn(3)-N(51)
N(21)-Mn(1)-N(51) 110.1(2) N(41)-Mn(3)-N(61)
N(41)-Mn(1)-N(51) 80.0(1) N(41)-Mn(3)-N(61b) 111.0(1)
O(2)-Mn(2)-N(2)
70.4(1) N(51)-Mn(3)-N(61)
169.8(2)
95.2(1)
78.4(2)
101.1(2)
99.9(2)
100.3(2)
100.5(2)
99.7(2)
O(2)-Mn(2)-N(11) 163.1(1) N(51)-Mn(3)-N(61b)
O(2)-Mn(2)-N(21)
O(2)-Mn(2)-N(31)
92.2(1) N(61)-Mn(3)-N(61b)
85.2(1) Mn(1)-N(11)-Mn(2)
O(2)-Mn(2)-N(31a) 93.2(1) Mn(1)-N(21)-Mn(2)
N(2)-Mn(2)-N(11)
N(2)-Mn(2)-N(21)
94.8(1) Mn(1)-N(41)-Mn(3)
90.5(1) Mn(1)-N(51)-Mn(3)
N(2)-Mn(2)-N(31) 154.6(1) Mn(2)-N(31)-Mn(2a)
N(2)-Mn(2)-N(31a) 93.9(1) Mn(3)-N(61)-Mn(3b) 101.6(2)
N(11)-Mn(2)-N(21) 79.5(1)
N(11)-Mn(2)-N(31) 110.3(2) Mn(1)‚‚‚Mn(2)‚‚‚Mn(2a) 132.52(4)
N(11)-Mn(2)-N(31a) 95.9(1) Mn(1)‚‚‚Mn(3)‚‚‚Mn(3b) 127.88(4)
Mn(2)‚‚‚Mn(1)‚‚‚Mn(3) 132.38(3)
Table 3. Selected Bond Lengths (Å) and Angles (deg) for 2
Mn(1)-N(1)
Mn(1)-N(11)
Mn(1)-N(21)
2.308(3)
2.229(3)
2.235(3)
Mn(1)-N(2)
Mn(1)-N(11a)
Mn(1)-N(23b)
2.264(3)
2.219(3)
2.244(3)
The N(1)-Mn(1)-N(2) bond angle of the axial pyridine
ligands is 172.6(1)°. This deviation from 180° has the effect of
moving away the axial ligands on the EO ring. In contrast, the
shortest contact between the benzoyl groups, smaller than 4 Å,
was found between the ligands on the EE rings.
N(1)-Mn(1)-N(2)
N(1)-Mn(1)-N(11)
N(1)-Mn(1)-N(11a) 92.1(1) N(11)-Mn(1)-N(23b) 173.9(1)
172.6(1) N(11)-Mn(1)-N(11a)
94.7(1) N(11)-Mn(1)-N(21)
79.6(1)
91.0(1)
N(1)-Mn(1)-N(21)
N(1)-Mn(1)-N(23b) 85.8(1) N(11a)-Mn(1)-N(23b) 94.3(1)
N(2)-Mn(1)-N(11) 92.6(1) N(21)-Mn(1)-N(23b) 95.1(1)
N(2)-Mn(1)-N(11a) 90.6(1) Mn(1)-N(11)-Mn(1a) 100.4(1)
126.0(2)
86.0(1) N(11a)-Mn(1)-N(21) 170.3(1)
The packing of the chains shows an unusual feature: the
chains are placed along the 1 1 0 and 1 -1 0 directions in
pseudoplanes of parallel chains, Figure 2b. The parallel chains
are well-isolated by the pyridine ligands, allowing minimum
Mn‚‚‚Mn interchain distances of 7.753(3) Å, whereas the mini-
mum Mn‚‚‚Mn distance between the pseudoplanes placed at 0,
1/2 and 1 in the z direction is greater than 17 Å, as a consequence
of the trans coordination of the 3-bzpy rings along the z axis.
Description of the Structure of cis-[Mn(3,5lut)2(N3)2]n (3).
An ORTEP plot of cis-[Mn(3,5lut)2(N3)2]n is shown in Figure
3a. The structure consists of neutral chains of manganese atoms
linked by azido bridges, placed along the (0 1 0) direction,
Figure 3b. In this case the octahedral coordination around the
manganese atoms consists of two 3,5-dimethylpyridine ligands
coordinated by means of the pyridine nitrogen atom and four
azido bridges which act as EE double bridges with the two
neighboring manganese ions. The two sets of EE bridges are
not equivalent, giving an alternating system. In this case, the
N-N bond distances are practically symmetrical, around 1.70
Å. The bond angles Mn(1)-N(11)-N(12) and Mn(1)-N(13A)-
N(2)-Mn(1)-N(21)
92.4(1) N(22)-N(21)-Mn(1)
N(2)-Mn(1)-N(23b) 87.1(1) N(22)-N(23)-Mn(1b) 126.1(2)
found along the chain, which, combined with the inversion
centers placed between Mn(2)-Mn(2A) and Mn(3)-Mn(3A)
and the absence of an inversion center on Mn(1), give rise to
the Mn(3)-Mn(1)-Mn(2)-Mn(2A)-Mn(1A)-Mn(3A)-Mn-
(3B) sequence with four different bridges between them. Bond
angles in the two centrosymmetric Mn(2)-N(31)-Mn(2A) and
Mn(3)-N(61)-Mn(3B) units are 99.7(2) and 101.6(2)°, re-
spectively, whereas the bond angles for the remaining noncen-
trosymmetric units are Mn(3)-N(41)-Mn(1) and Mn(3)-
N(51)-Mn(1) 100.3(2) and 100.5(2)°, and Mn(1)-N(11)-
Mn(2) and Mn(1)-N(21)-Mn(2), 101.1(2) and 99.9(2)°. The
four different Mn‚‚‚Mn distances range between 3.364(1) and
3.415(1) Å. The chains are well-isolated due to the large 2-bzpy
ligand, the minimum Mn‚‚‚Mn interchain distance being 9.931-
(1) Å. Dihedral angles between neighboring Mn2N2 rings are

Three New One-Dimensional Manganese Azido Systems
Inorganic Chemistry, Vol. 38, No. 25, 1999 5719
Figure 1. (a, left) ORTEP drawing for cis-[Mn(2-bzpy)(N3)2]n (1). Ellipsoids at the 50% probability level. (b, right) Packing of the well-isolated
chains along the c axis.
Figure 2. (a, left) ORTEP drawing for trans-[Mn(3-bzpy)2(N3)2]n (2). Ellipsoids at the 50% probability level. (b, right) View of the packing along
the (110) direction showing the different orientations of the chains placed at z ) 1 and 0 from those at z ) 1/2 and the large interchain distance (>17
Å) along the c axis.
N(12A), 121.9(2) and 125.8(2)°, Mn(1)-N(21)-N(22) and Mn-
(1)-N(23B)-N(22B), 122.9(3) and 124.8(3)°, lie in the normal
range5 found for related EE compounds. Deviation from
planarity to chair conformation for the Mn(N3)2Mn eight-
membered rings, measured as the δ torsion angle between the
plane defined by the six N coplanar atoms of the bridges and
the corresponding N-Mn-N planes, takes the δA 32.7(1) and
δB 31.6(1)° values with the corresponding manganese atoms
deviating 0.818(1) and 0.788(1) Å from A and B planes. Mn‚
‚‚Mn intrachain distances are 5.159(1) and 5.140(1) Å for A
and B rings, respectively, whereas the minimum interchain
distance is 9.277(1) Å due to the large size of the lutidine ligands
which efficiently isolate the chains. It is worth pointing out that
the manganese skeleton consists of a zigzag chain due to the
cis coordination of the bridges with a Mn(1A)‚‚‚Mn(1)‚‚‚Mn-
(1B) angle of 97.28(3)°, Figure 3a.
Magnetic Susceptibility and Powder EPR Spectra. The
øMT product vs T for 1 measured with a low external field (100
G) shows a continuous increase from the room-temperature
value (4.39 cm3‚K‚mol-1) down to 2 K, where it reaches the
value of 32.8 cm3‚K‚mol-1, Figure 4. This plot indicates weak
ferromagnetic coupling, with no evidence of interchain interac-
tions, which is consistent with the large interchain distance.
According to the structural data, compound 1 shows a compli-

5720 Inorganic Chemistry, Vol. 38, No. 25, 1999
Abu-Youssef et al.
Figure 3. (a, left) ORTEP drawing for cis-[Mn(35lut)2(N3)2]n (3). Ellipsoids at the 50% probability level. (b, right) Packing of the well-isolated
chains along the b axis.
Figure 4. Plot of the øMT product vs T for cis-[Mn(2-bzpy)(N3)2]n
(1). Solid line shows the best fit of the data (see text).
Figure 5. Plot of the molar susceptibility and øMT product vs T for
trans-[Mn(3-bzpy)2(N3)2]n (2) (open symbols) and cis-[Mn(35lut)2(N3)2]n
(3) (filled symbols). Solid lines show the best fit of the data (see text).
cated alternance of coupling constants Mn3′-(J1)-Mn1′-(J2)-
Mn2′-(J3)-Mn2-(J2)-Mn1-(J1)-Mn3-(J4)-Mn3′′. No
model is available for fitting the temperature dependence of øT
for such complex systems, but, given the similar Mn-N-Mn
bond angles for the different bridges along the chain, the
coupling was evaluated as an average value using the conven-
tional equations17 derived by Fisher for the magnetic suscep-
tibility of an infinite chain of classical spins based on the
Hamiltonian H ) -J∑SiSi+1 for local spin values S ) 5/2 with
Hamiltonian H ) -J1∑S2iS2i+1 - J2∑S2iS2i+2, and the best-fit
parameters were J1 ) -12.31(3) cm-1, J2 ) +3.5(3) cm-1
,
and g ) 1.99(1). This result compares well with results reported
for similar alternating compounds with ferro-antiferromagnetic
interactions such as [Mn(bpy)(N3)2]n for which18 J1 was -11.8
cm-1 and J2 was +9.6 cm-1, or the more recent [Mn(3-Et,4-
Mepy)2(N3)2]n compound for which8 J1 was -13.7 cm-1 and
J2 was +2.4 cm-1
.
Compound 3. The plot of the magnetic data in the 300-4
K range is shown in Figure 5. The general behavior corresponds
to an antiferromagnetically coupled system, with a room-
temperature øMT value of 3.21 cm3‚K‚mol-1, which decreases
continuously and tends to zero when the temperature decreases.
øM shows a broad maximum at 70 K. Fit of the susceptibility
data was performed with the Fisher equation,17 Hamiltonian H
) -J∑SiSi+1, and the best-fit parameters were J ) -10.59(4)
cm-1 and g ) 2.007(3). In this case, the results are also com-
parable to those obtained for related compounds7 such as [Mn-
(pyOH)2(N3)2]n (J ) -7.0 cm-1) or [Mn(3Etpy)2(N3)2]n (J )
-13.8/-11.7 cm-1) with comparable chair distortions in the
bridge region.
one J parameter. The best-fit parameters are J ) +0.80(2) cm-1
,
g ) 2.05(1). The magnetization measured at 2.5 K reaches 4.3
µB at 5 T, showing that under these conditions the magnetization
is not yet saturated.
Compound 2. The øT vs T plot in the 300-4 K range is
shown in Figure 5. The overall behavior indicates a dominant
antiferromagnetic interaction, with a room-temperature øMT
value of 3.70 cm3‚K‚mol-1, which decreases continuously and
tends to zero when the temperature decreases. øM shows a broad
maximum centered at 40 K. The magnetic data were analyzed
by means of the reported expression18 derived from the
(17) Fisher, M. E. Am. J. Phys. 1964, 32, 343.
(18) Corte´s, R.; Drillon, M.; Solans, X.; Lezama, L.; Rojo, T. Inorg. Chem.
1997, 36, 677.
The room-temperature polycrystalline powder X-band EPR
spectrum for 1 shows a broad band with a peak-to-peak width

Three New One-Dimensional Manganese Azido Systems
Inorganic Chemistry, Vol. 38, No. 25, 1999 5721
∆Hpp ) 750 G centered at g ) 2.05, confirming well the g
value obtained from susceptibility measurements. However, the
spectrum of 3 shows a very sharp line ∆Hpp ) 40 G centered
at g ) 2.00 and a weak half-field signal at g ) 4.00, whereas
2 shows a line with ∆Hpp ) 220 G centered at g ) 2.02. The
line width of the spectra of one-dimensional compounds is the
result of two conflicting contributions: intrachain superexchange
interactions which tend to narrow the lines and dipolar interac-
tions which tend to broaden them.19 The exchange coupling
constants in 2 and 3 are of the same order, while that in 1 is an
order of magnitude smaller. The dipolar interactions are mainly
determined by the Mn-Mn distance: the shorter the distance
the stronger the interaction. In qualitative terms exchange
narrowing becomes dominant for 3, in which the Mn‚‚‚Mn
distances are relatively long, 5.159 and 5.140 Å, and the J value
is relatively high (-10.59 cm-1). In contrast 1, which shows
very short Mn‚‚‚Mn distances of 3.364 and 3.414 Å, and a low
J value (less than 1 cm-1), has the broadest line. The intermedi-
ate line width value obtained for the alternating chain 2 is
consistent with the alternating short-long Mn‚‚‚Mn distances
(5.229 and 3.416 Å), and J values comparable to 1 and 3 offer
the mixing of the two opposite effects. These results are also
comparable to those obtained for the alternating EE/EO systems
[Mn(bpy)(N3)2]n 18 or [Mn(3Et,4Mepy)(N3)2]n 8 which show line
widths of 500 and 320 G, respectively, and the EE bridged
systems [Mn(L)2(N3)2]n, L ) pyOH or 3-Etpy, with line widths
of 35-42 G.7
Figure 6. Angular dependence of the peak-to-peak line width ∆Hpp
of cis-[Mn(2-bzpy)(N3)2]n (1) at room temperature and X-band. Dot-
centered circles show the dependence of ∆Hpp between the parallel (θ
) 0°) and the perpendicular (θ ) 90°) orientations of the external field
with the chain axis c. The solid line shows the expected angular
dependence [3 cos2 θ - 1]4/3. Filled squares are ∆Hpp for a constant θ
) 90° rotation.
Variable-temperature EPR spectra show common features for
1-3: the line width decreases slightly when T decreases, and
at low temperatures, between 100 and 4 K, a significant
broadening of the signal was observed, reaching values close
to 2000 G for 1, 750 G for 2, and 1300 G for 3 as a consequence
of the increased correlations along the chain. In addition,
compounds 1 and 3 show new signals giving a complicated
pattern, probably due to the anisotropic shift of the signals
expected at low T.
Single-Crystal EPR Experiments. EPR properties of man-
ganese quasi-ideal Heisenberg 1-D antiferromagnets have been
studied for several systems, mainly for [(CH3)4N)][MnCl3]
(TMMC).19 On the basis of the structural data and of the
variation of the line width observed in the powder EPR spectra
for 1-3, single-crystal measurements were performed, to
characterize in more detail the spin dynamics in systems with
different kinds of magnetic interactions and to estimate the
intrachain to the interchain exchange interaction.
(a) cis-[Mn(2-bzpy)(N3)2]n (1). For the single-crystal experi-
ments a well-formed prismatic crystal in which the chain
direction corresponds to the elongation axis of the prism was
used. The spectra were recorded by rotating around the
perpendiculars in the chain direction. A third rotation around
the chain direction was also performed. The first rotations were
practically identical to each other, and the angular dependence
of the peak-to-peak width ∆Hpp for one of them is shown in
Figure 6. Maximum broadening was found in the parallel
direction, ∆Hpp ) 2095 G; a well-defined minimum in the line
width was found at the magic angle θ ) 54.7°, ∆Hpp ) 500 G,
and a local maximum along the direction perpendicular to the
chain, ∆Hpp ) 760 G. This is the expected behavior for one-
dimensional magnets in which spin diffusion determines a large
secular enhancement.19-21 The dependence of the line width is
consistent with the ∆Hpp ) a + b(3‚cos2 θ - 1)4/3, as
Figure 7. Angular dependence of the g value cis-[Mn(2-bzpy)(N3)2]n
(1) at room temperature and X-band.
expected.19 The best fit values are a ) 309 G, b ) 639 G. A
further confirmation of the one-dimensional behavior comes
from the analysis of the line shape which is Lorentzian at the
magic angle. Some deviation from this is observed at θ ) 90°,
and it is diffusive at θ ) 0°. The rotation around the chain axis
shows only a minor variation in the line width in the 180°
rotation (100 G approximately), Figure 6. The important
information which emerges from the EPR spectra is linked to
the fact that the line shape, following the characteristic one-
dimensional behavior, provides an upper limit22 to the interchain
exchange interactions, J′, which would restore the Lorentzian
line shape as J′/J e 10-3
.
The g parameter was also checked along the rotation, from
the observation of the anomalous g value of 2.05 found from
susceptibility or powder EPR spectra. The g dependence vs θ
is plotted in Figure 7, which shows a maximum of g ) 2.10 at
θ ) 90°, g ) 2.01 for θ ) 0° and a minimum value of 1.91 for
θ ) 25°. Similar behavior was observed in several compounds,
such as in [Cu(hfac)2(NITMe)],23 a ferromagnetic Cu(II)-radical
(20) Kalt, H.; Siegel, E.; Pauli, N.; Mosebach, H.; Wiese, J.; Edgar, A. J.
Phys. C 1983, 16, 6427.
(21) Richards, P. M.; Quinn, R. K.; Morosin, B. J. Chem. Phys. 1973, 59,
(19) Bencini, A.; Gatteschi, D. EPR of Exchange Coupled Systems;
Springer-Verlag: Berlin, Heidelberg 1990, Chapter 10, and references
cited therein.
4474.
(22) Hennessy, M. J.; McElwee, C. D.; Richards, P. M. Phys. ReV. B 1973,
7, 930.

5722 Inorganic Chemistry, Vol. 38, No. 25, 1999
Abu-Youssef et al.
Figure 9. Scheme of the two different chain directions along (1 1 0)
and (1 -1 0) for trans-[Mn(3-bzpy)2(N3)2]n (2). Magic angle 1-4
indicate the expected directions at which the magic angle for one set
of chains should be found and the two θ common angles for all chains.
Figure 8. Angular dependence of the resonance field Ho of cis-[Mn-
(2-bzpy)(N3)2]n (1) at room temperature and X-band with the orientation
of the microwave field for a constant θ ) 90° orientation. φ ) 0 and
φ ) 90° refer to the orientation of the external field with the chain
axis. The dotted line shows the expected dependence of Ho with φ
according to the cos 2φ law.
system, and attributed to short-range interactions. G-shifts may
also be due to demagnetization effects, but they should become
important only at low temperature. In our case, at high
temperature these explanations are not acceptable and seem to
agree with the angular dependence of g pointed out by Benner
et al. for TMMC.24 These authors suggested that, at low
symmetry and in the presence of broad lines, transverse dynamic
susceptibility øxx(ω) cannot be efficiently described by the
circular polarized susceptibility ø++(ω), which is generally used
in theoretical treatments. They showed that in some cases the
nondiagonal contributions, ø+-(ω) and ø-+(ω), must be taken
into account. When this is done, dynamic shifts may be observed
in the lines. In fact small shifts were observed in TMMC and
CsNiF3.24
The proposal of Benner et al. also suggests the dependence
of the resonance field (for a constant θ angle), as a function of
the microwave field angle φ with the chain, following the cos
2φ law.24 In our case this g-shift was confirmed by the different
resonance fields for θ ) 90° and φ ) 90 or φ ) 0° observed
during the three rotations of the crystal. In addition, a third
spectrum at φ ) 45° was measured. As is shown in Figure 8,
compound 1 follows the expected φ dependence.
(b) trans-[Mn(3-Bzpy)2(N3)2]n (2). Single-crystal spectra of
2 were recorded by rotating 180° with the static magnetic field
in the well-formed (0 0 -1) face. There are two crystallographi-
cally equivalent, but magnetically nonequivalent, chains in the
unit cell, parallel to the (1 1 0) and (1 -1 0) directions, forming
an angle of 68.74° between them, Figure 9. The angular
dependence of the line width shows two maxima close to the
directions of the chains, and four minima, which nicely agree
with the magic angles of the chains. In fact we assume that the
spectra in each crystal setting are the sum of two contributions,
corresponding to the nonequivalent chains. The two signals have
different line widths, and in general the experimental spectrum
will have a peak-to-peak width which is close to that of the
Figure 10. Angular dependence of the peak-to-peak line width ∆Hpp
of trans-[Mn(3-bzpy)2(N3)2]n (2) at room temperature and X-band. The
numbering 1-4 and the two θ angles refer to Figure 9. Dotted lines
show the angular dependence [3 cos2 θ - 1]4/3 for the two chains with
a θ gap of 68.74°. The solid line is a guide for the eye.
narrower signal. In Figure 10 we show the calculated line widths,
assuming a (3 cos2 θ - 1)4/3 behavior. Line shape analysis
performed for the equivalent orientations θ ) 34.37, 54.7, and
90° shows characteristic one-dimensional behavior, and the
expected line widths of 694, 293, and 350 G, respectively. It is
also worth noting that 2 also exhibits a g-shift as a function of
the θ angle, but in this case the effect was less intense than the
effect observed for compound 1. The minimum g value of 1.97
was found for θ ) 34.37°, and the maximum g of 2.03 appears
for the perpendicular spectrum; so we tentatively give the same
explanation for the g shift in 1.
Therefore, although in this case the analysis of the data is
made more difficult by the presence of nonequivalent chains in
the lattice which do not permit the θ ) 0° spectrum to be
obtained, the EPR data confirm the one-dimensional nature of
the interaction.
(c) cis-[Mn(3,5lut)2(N3)2]n (3). To perform the single-crystal
experiments, an indexed crystal of 3 was mounted on the (-1
0 -1) well-developed face. In all the crystal orientations the
resonance was observed at g ) 2.00. As above, three rotations
were measured, two of them from parallel to orthogonal to the
chain in two orthogonal planes and the third was a rotation with
respect to the external magnetic field for a constant θ ) 90°
value. The results are apparently surprising: in the first rotation
(from 010 to 101), Figure 11, the line width is highest when
orthogonal to the chain, and the minimum is at the magic angle
(23) Caneschi, A.; Gatteschi, D.; Zanchini, C.; Rey, P. J. Chem. Soc.,
Faraday Trans. 1987, 83, 3603.
(24) Benner, H.; Brodehl, M.; Seitz, H.; Wiese, J. J. Phys. C 1983, 16,
6011.

Three New One-Dimensional Manganese Azido Systems
Inorganic Chemistry, Vol. 38, No. 25, 1999 5723
Figure 11. Angular dependence of the peak-to-peak line width ∆Hpp
of cis-[Mn(35lut)2(N3)2]n (3) at room temperature and X-band. The two
orthogonal rotations show the normal angular dependence (filled circles)
and reversed parallel-perpendicular behavior (solid squares).
Figure 12. Angular dependence of the intensity of the half-field signal
(g ) 4.00) of cis-[Mn(35lut)2(N3)2]n (3) at room temperature and
X-band. The signal disappears at θ ) 0 and shows maximum intensity
for θ ) 45° according to the expected sin2 θ cos2 θ law (dotted line).
from the perpendicular to the chain. In fact, maximum broaden-
ing was found for θ ) 90° (95 G), the magic angle at θ ) 35°
(33 G) and the local maximum in the parallel direction θ ) 0°
(70 G). In the second rotation (orthogonal to the first rotation),
the positions of the maximum and minimum line widths are
normal, but the broadening effect is very weak. In the third
rotation (constant θ ) 90°), a complicated dependence of the
line width was observed. The inverse broadening effects for
the rotation from 010 to 101 are also reflected in the corre-
sponding line shape analysis, which shows a more diffusive
component for the θ ) 90° spectrum and a more Lorentzian
component for θ ) 0° and the magic angle.
This surprising feature may be related to the structural data,
since the manganese atoms along the chain form a zigzag
arrangement with an unusual angle Mn‚‚‚Mn‚‚‚Mn of 97°. For
this reason the dipolar effects are minimized along the chain
axis, and then weak normal or anomalous reversed behavior
may be found for selected θ ) 90° orientations.
Due to the narrowing of the main line in the EPR spectrum,
a well-defined half-field line is observed at g ) 4.00. Rotation
between θ ) 0° and 90° shows the same line width angular
dependence as the main line at g ) 2.00. According to the
expected properties, the intensity of the g ) 4.00 signal
decreases when θ approaches the parallel direction and disap-
pears for θ ) 0°. The observation of the half-field line is
excellent evidence19 of the one-dimensional nature of 2. In
Figure 12 the angular dependence of the half-field signal with
the expected25 sin2 θ cos2 θ law is shown.
Conclusions
Three one-dimensional manganese-azido systems were
prepared and structurally characterized. Each compound is
characterized by different chain structure and magnetic behav-
ior: double EO bridges by ferromagnetic coupling (1), alternat-
ing EE and EO bridges by alternating ferro-antiferromagnetic
coupling (2), and finally EE bridges by antiferromagnetic
interactions (3). Single-crystal EPR experiments showed that
these new types of one-dimensional magnetic materials can be
interpreted on the same grounds as were originally developed
for the regular Heisenberg antiferromagnet TMMC. Therefore,
we have shown that the spin dynamics in one-dimensional
ferromagnets and alternating ferro-antiferromagnets can easily
be related to the dynamics in antiferromagnets.
Acknowledgment. A.E. and R.V. thank the CICYT (Grant
PB96/163) for supporting this research. F.A.M. thanks Prof. C.
Kratky and Dr. F. Belaj (University of Graz) for use of
experimental equipment. F.A.M. and M.A.M.A.-Y. are grateful
for partial support from an OENB grant (Project 6630). D.G.
thanks the EU for the TMR Grant 3MD.b.
Supporting Information Available: Three X-ray cryatallographic
files, in CIF format. This material is available free of charge via the
Internet at http://pubs.acs.org.
IC990656X
(25) Benner, H.; Wiese, J. Physica B 1979, 96, 216.