Synthesis, Structural Characterization (X-ray and EXAFS), and Magnetic Properties of Polynuclear Manganese(II) Complexes with Chlorobenzoato Bridges

Article abstract of DOI:10.1021/ic970935p

Three different types of polynuclear Mn¹¹ complexes with carboxylate bridges were obtained from the reaction of Mn(RCOO)₂ wiih 2.2′-bipyridine (bpy). Dinuclear complexes [Mn:(μ-RCOO)₂(bpy)₄](ClO₄)

Full text of DOI:10.1021/ic970935p

788
Inorg. Chem. 1998, 37, 788-798
Synthesis, Structural Characterization (X-ray and EXAFS), and Magnetic Properties of
Polynuclear Manganese(II) Complexes with Chlorobenzoato Bridges
Belen Albela,1a Montserrat Corbella,*,1a Joan Ribas,1a Isabel Castro,1b Jorunn Sletten,1c and
Helen Stoeckli-Evans1d
Departament de Quimica Inorga`nica, Universitat de Barcelona, Diagonal 647, 08028 Barcelona, Spain,
Departament de Quimica Inorga`nica, Universitat de Vale`ncia, Dr. Moliner 50, 46100 Burjassot
(Vale`ncia), Spain, Department of Chemistry, University of Bergen, Alle´gatan 41, N-5007 Bergen,
Norway, and Institut de Chimie, Universite´ de Neuchaˆtel, Avenue de Bellevaux 51,
CH-2000 Neuchaˆtel, Switzerland
ReceiVed July 25, 1997
Three different types of polynuclear MnII complexes with carboxylate bridges were obtained from the reaction of
Mn(RCOO)2 with 2,2′-bipyridine (bpy). Dinuclear complexes [Mn2(µ-RCOO)2(bpy)4](ClO4)2 with R ) 2-ClPh,
3-ClPh, 4-ClPh, Ph (1-4); trinuclear complexes [Mn3(µ-RCOO)6(bpy)2] with R ) 2-ClPh, 3-ClPh (5, 6), [Mn3(µ-
RCOO)6(2,2′-Me2-bpy)2] with R ) 2-ClPh, 3-ClPh, 4-ClPh (7-9), and 1D complexes [Mn(µ-RCOO)2(bpy)]n
with R ) 3-ClPh, 4-ClPh (10, 11). [Mn2(µ-PhCOO)2(bpy)4](ClO4)2 (4) and the [Mn(µ-3-ClPhCOO)2(bpy)]n‚nH2O
(10) have been characterized by X-ray diffraction. Complex 4 crystallizes in the triclinic system, space group P1h
with a ) 9.1886(10) Å, b ) 11.6135(9) Å, c ) 13.595(2) Å, R ) 66.225(13)°, â ) 84.073(12)°, γ ) 88.593(10)°,
Z ) 1. Complex 10 crystallizes in the monoclinic system, space group C2/c with a ) 26.376(5) Å, b ) 12.404(3)
Å, c ) 7.095(1) Å, â ) 96.10(3)°, Z ) 4. The other complexes were characterized by XANES and EXAFS
studies, by comparison with analogous complexes of known X-ray crystal structure. The Mn‚‚‚Mn distances in
the dinuclear complexes and in the infinite chains were similar (aV 4.5 Å) and longer than for the trinuclear
complexes (aV 3.6 Å); this is in agreement with the number of carboxylate bridges: two in the first case and
three for the trinuclear complexes. All the polynuclear complexes shown very weak antiferromagnetic coupling
(J between -0.7 and -3.22 cm-1). So, at low temperatures more than one spin state may be populated and
many possible transitions may be expected in the EPR spectra; each series shows a similar spectrum, which is
different from the others.
Introduction
Structurally characterized examples reported show either only
carboxylates or carboxylates and other bridging ligands. These
complexes model the bonding characteristics of biomolecules
with two neighboring MnII ions, since polypeptides often have
unbound -COO- groups which can be used in coordination to
metal ions. With only carboxylate bridges, there are reports of
species that possess only one,4,5 two,2,6-9 or three car-
boxylates6,10-12 bridging the two MnII centers. However, a
dimeric tetrabridged species with the copper(II) acetate structure
has not been reported. Recently, a MnII dimer bridged by two
The manganese(II) carboxylate system is one of the few areas
of MnII chemistry in which a significant amount of structural
data is available. Most MnII carboxylates and their hydrates
are stable in air but some of them, for example [Mn(ButCOO)2],
react readily to form red-brown trinuclear MnIII species.
Monomers are knownsfor example, [Mn(RCOO)(H2O)5]-
(RCOO); however, most structures involve bridging car-
boxylates, which can be coordinated to the metals in various
fashions,2 giving rise to polynuclear species. Indeed, as
underlined by Wieghardt,3 carboxylate groups show a pro-
nounced tendency to bind manganese in either a monodentate
or a bidentate manner; in the latter case, often with the formation
of a symmetrical µ-syn-syn carboxylate bridge.
(4) Chen, X. M.; Tang, Y. X.; Xu, Z. J.; Wak, T. C. W. J. Chem. Soc.,
Dalton Trans. 1995, 4001.
(5) Adams, H.; Bailey, N. A.; Debaecker, N.; Fenton, D. E.; Kanda, W.;
Latour, J. M.; Okawa, H.; Sakiyama, H. Angew. Chem., Int. Ed. Engl.
1995, 34, 2535
(6) Shake, A. R. Ph.D. Thesis, University of Indiana, Bloomington, IN,
1990.
(7) Oshio, 0. H.; Ino, E.; Mogi, I.; Ito,T. Inorg. Chem. 1993, 32, 5697.
(8) Che, C. M.; Tang, W. T.; Wong, K. Y.; Lai, T. F. J. Chem. Res.,
Synop. 1991, 30.
In contrast to dimanganese(III) complexes, effective self-
assembly routes to carboxylate-bridged dimanganese(II) com-
pounds are limited owing to the lability of MnII ions and their
tendency to undergo both substitution and redox reactions.
(9) Glowiak, T.; Kozlowski, H.; Erre, L. S.; Micera, G. Inorg. Chim. Acta
1995, 236, 149.
(10) Wieghardt, K.; Bossek, U.; Nuber, B.; Weiss, J.; Bonvoisin, J.;
Corbella, M.; Vitols, S. E.; Girerd, J. J. J. Am. Chem. Soc. 1988, 110,
7398.
(1) (a) Departament de Qu´ımica Inorga`nica. Universitat de Barcelona. (b)
Departamento de Qu´ımica Inorga´nica, Universidad de Valencia. (c)
Department of Chemistry, University of Bergen. (d) Institut de Chimie,
Universite´ de Neuchaˆtel.
(2) (a) Chiswell, B. In ComprehensiVe Coordination Chemistry; Perga-
mon: Oxford, 1987; Vol IV, p 43. (b) Rardin, R. L.; Tolman, W. B.;
Lippard, S. J. New J. Chem. 1991, 15, 417.
(11) Matsushima, H.; Ishiwa, E.; Nakashima, M.; Tokii, T. Chem. Lett.
1995, 129.
(12) Osawa, M.; Singh, U. P.; Tanaka, M.; Morooka,Y.; Kitajima, N. J.
Chem. Soc., Chem. Commun. 1993, 310.
(3) Wieghardt, K. Angew. Chem., Int. Ed. Engl. 1989, 28, 1153.
S0020-1669(97)00935-X CCC: $15.00 © 1998 American Chemical Society
Published on Web 02/04/1998

Mn(II) Complexes with Chlorobenzoato Bridges
Inorganic Chemistry, Vol. 37, No. 4, 1998 789
[Mn(RCOO)2]‚nH2O (R ) 2-ClPh; 3-ClPh; 4-ClPh, Ph) were synthe-
sized by the reaction of freshly prepared MnCO3 and R-COOH in hot
water: after several hours, the solution was filtered and the solvent
was reduced by rotatory evaporation, giving a pale pink precipitate of
the desired product.
carboxylate groups was described, in which the dicarboxylate
bridge belongs to a single large ligand.13 In other complexes
both metals are bridged not only by carboxylates, but also by
other typical ligands such as µ-aqua,14-16 µ-hydroxo,17
µ-phenoxo,18-23 and µ-alcoxo.24
Syntheses. [Mn2(µ-RCOO)2(bpy)4](ClO4)2, R ) 2-ClPh, 3-ClPh,
4-ClPh, Ph (1-4). [Mn(RCOO)2]‚nH2O (R ) 2-ClPh; 3-ClPh; 4-ClPh,
Ph) (1.0 mmol) and NaClO4‚H2O (0.14 g, 1.0 mmol) were mixed in
absolute ethanol (50-100 mL). The resulting mixture was filtered to
remove any impurity, and 2,2′-bipyridine (0.30 g; 2.0 mmol) in ethanol
(20 mL) was added. The solution turned yellow immediately and was
stored at room temperature for several hours. The yellow solid formed
was collected by filtration, washed in ethanol and ether, and dried in
air. The yield was high (85-90%). For complex 2, when the filtered
solution was left to stand in the open air for several days, a new product
was isolated, which did not present the perchlorate anion and cor-
responded to the chain (10). Complex 3 precipitated only in the
presence of excess NaClO4 (2-3 g); however, if the above ratio was
used the corresponding chain (11) was obtained. Suitable crystals were
isolated only for complex 4, by slow evaporation at room temperature
of an EtOH/H2O (1:1) solution of the initial solid. Anal. Calcd. for
complex 1, C54H40Cl4Mn2N8O12‚3H2O: C, 50.0; H, 3.54; N, 8.62.
Found: C, 50.7; H, 3.6; N, 8.4. FT-IR data (KBr, main bands, cm-1):
1600 (vs), 1561 (s), 1474 (m), 1439 (s), 1401 (s), 1144 (m), 1118 (vs),
1088 (vs), 1016 (m), 765 (s), 754 (s), 648 (m), 624 (m). Analysis
calculated for complex 2: C54H40Cl4Mn2N8O12: C, 52.09; H, 3.22; N,
9.00; Cl, 11.41. Found: C, 52.2; H, 3.2; N, 8.9; Cl, 11.4. FT-IR data
(KBr, main bands, cm-1): 1609 (vs), 1563 (s), 1475 (m), 1439 (s),
1387 (s), 1120 (m), 1089 (vs), 1016 (m), 768 (s), 757 (m), 624 (m).
Anal. Calcd for complex 3, C54H40Cl4Mn2N8O12: C, 52.09; H, 3.22;
N, 9.00; Cl, 11.41. Found: C, 51.7; H, 3.2; N, 9.0; Cl, 11.4. FT-IR
data (KBr, main bands, cm-1): 1608 (vs), 1561 (vs), 1474 (m), 1438
(vs), 1400 (vs), 1147 (m), 1118 (s), 1088 (vs), 1015 (s), 776 (m), 764
(vs), 738 (m), 624 (m), 530 (m). Anal. Calcd for complex 4,
C54H42Cl2Mn2N8O12: C, 55.19; H, 3.61; N, 9.54; Cl, 5.96. Found: C,
55.1; H, 3.6; N, 9.4; Cl, 6.0. FT-IR data (KBr, main bands, cm-1):
1602 (vs), 1561 (vs), 1474 (m), 1438 (vs), 1392 (vs), 1118 (s), 1089
(vs), 1016 (s), 766 (s), 714 (m), 625 (m).
On the other hand, the number of trinuclear MnII complexes
with bridging carboxylato groups is limited, and all of them
present only these kinds of bridging ligands.25-30 Finally,
although manganese(II)-carboxylate systems are usually poly-
nuclear, few one-dimensional polymers have been described;
examples are reported elsewhere.31-39
In this paper we report the isolation and characterization of
three kinds of MnII compound of different nuclearity: dinuclear
complexes [Mn2(µ-RCOO)2(bpy)4](ClO4)2, trinuclear complexes
[Mn3(µ-RCOO)6(bpy)2], and one-dimensional compounds [Mn(µ-
RCOO)2(bpy)]n‚nH2O. The structural characterization was
achieved by X-ray diffraction and EXAFS techniques. The
magnetic behavior of these compounds has been studied, as has
their EPR spectra.
Experimental Section
Materials. All manipulations were performed under aerobic condi-
tions. Reagent grade solvents were used without further purification.
Organic reagents were used as received except 2-ClPhCOOH,
3-ClPhCOOH, and 4-ClPhCOOH, which were recrystallized from hot
water. Yield was calculated from the stoichiometric reaction. The
(13) Tanase, T.; Lippard, S. J. Inorg. Chem. 1995, 34, 4682.
(14) Caneschi, A.; Ferraro, F.; Gatteschi, D.; Melandri, M. C.; Rey, P.;
Sessoli, R. Angew. Chem., Int. Ed. Eng. 1989, 28, 1365.
(15) Yu, B. S.; Lippard, S. J.; Shweky, I.; Bino, A. Inorg. Chem. 1992,
31, 3502.
(16) Coucouvanis, D.; Reynolds, R. A.; Dunham, W. R. J. Am. Chem. Soc.
1995, 117, 7570.
(17) Bossek, U.; Wieghardt, K.; Nuber, B.; Weiss, J. Inorg. Chim. Acta
1989, 165, 123
(18) Gultneh, Y.; Farooq, A.; Liu, S.; Karlin, K. D.; Zubieta, J. Inorg. Chem.
1992, 31, 3607.
(19) Higuchi, C.; Sakiyava, H.; Okawa, H.; Isobe, R.; Fenton, D. E. J.
Chem. Soc., Dalton Trans. 1994, 1097.
(20) Mikuriya, M.; Fujii, T.; Kamisawa,S.; Kawasaki, Y.; Tokki, T.; Oshio,
H. Chem. Lett. 1990, 1181.
(21) Sakiyama, H.; Tamaki, H.; Kodera, M.; Matsumoto, N.; Okawa, H.
J. Chem. Soc., Dalton Trans. 1993, 591.
(22) Higuchi, C.; Sakiyawa, H.; Osawa, H.; Fenton, D. K. J. Chem. Soc.,
Dalton Trans. 1995, 4015.
(23) Wada, H.; Motoda, K.; Ohba, M.; Sakiyama, H.; Matsumoto, N.;
Okawa, H. Bull. Chem. Soc. Jpn. 1995, 68, 1105.
(24) Pessiki, P. J.; Khangulov, S. V.; Ho, D. M.; Dismukes, G. C. J. Am.
Chem. Soc. 1994, 116, 891.
(25) Hubner, K.; Roesky, H. W.; Noltemeyer, M.; Bohra, R. Chem. Ber.
1991, 124, 515.
(26) Menage, S.; Vitols, S. E.: Bergerat, P.; Codjovi, E.; Kahn, O.; Girerd,
J. J.; Guillot, M.; Solans, X.; Calvet, T. Inorg. Chem. 1991, 30, 2666.
(27) Vincent, B.; Christou, G. AdV. Inorg. Chem. Radiochem. 1989, 32,
197.
(28) Rardin, R. L.; Bino, A.; Poganiuch, P.; Tolman, W. B.; Liu, S.;
Lippard, S. J. Angew. Chem., Int. Ed. Eng. 1990, 29, 812.
(29) Rardin, R. L.; Poganiuch, P.; Bino, A.; Goldberg, D. P.; Tolman, W.
B.; Liu, S.; Lippard, S. J. J. Am. Chem. Soc. 1992, 114, 5240.
(30) Zhong, Z. J.; You, X.-Z.; Mak, T. C. W. Polyhedron 1994, 13, 2157.
(31) Cano, J.; Munno, D.; Sanz, J.; Ruiz, R.; Lloret, F.; Faus, J.; Julve, M.
J. Chem. Soc., Dalton Trans. 1994, 3465.
(32) Wagner, G. R.; Friedberg, S. A. Phys. Lett. 1964, 9, 11.
(33) Osaki, H.; Nakai, Y.; Watanabe, T. J. Phys. Soc. Jpn. 1963, 18, 919.
(34) Lis, T. Acta Crystallogr. A 1977, 24, 348.
(35) Chen, X. M.; Mak, T. C. W. J. Crystallogr. Spectrosc. Res. 1991, 21,
21.
(36) Chen, X. M.; Mak, T. C. W. Inorg. Chim. Acta 1991, 189, 3.
(37) Marioni, P.-A.; Marty, W.; Stoeckli-Evans, H.; Whitaker, C. Inorg.
Chim. Acta 1994, 219, 161.
(38) Solans, X.; Gali, S.; Font-Altaba, M.; Oliva, J.; Herrera, J. Afinidad
1989, 45, 247
[Mn3(µ-RCOO)6(bpy)2], R ) 2-ClPh (5), 3-ClPh (6). 2,2′-
Bipyridine (0.16 g, 1.0 mmol) in EtOH (20 mL) was added to a solution
of [Mn(RCOO)2]‚nH2O (0.40 g; 1.0 mmol) in absolute ethanol (30 mL).
The resulting mixture was left undisturbed at room temperature for
several hours. A yellow microcrystalline powder was formed, which
was filtered, washed in ethanol and ether, and air dried. The yield
was 70-85%. In the synthesis of complex 6, upon prolonged storage
of the filtrate at room temperature, a second yellow product was isolated,
whose analysis again corresponded to the chain (10), which is described
below. Anal. Calcd for complex 5, C62H40Cl6Mn3N4O12: C, 52.76;
H, 2.83; N, 3.97; Cl, 15.10. Found: C, 52.2; H, 3.0; N, 4.0; Cl, 15.0.
FT-IR data (KBr, main bands, cm-1): 1611 (vs), 1570 (s), 1450 (m),
1439 (s), 1387 (vs), 1058 (m), 1118 (vs), 762 (vs), 650 (m). Anal.
Calcd for complex 6, C62H40Cl6Mn3N4O12: C, 52.76; H, 2.83; N, 3.97;
Cl, 15.10. Found: C, 52.8; H, 2.9; N, 4.0; Cl, 15.1. FT-IR data (KBr,
main bands, cm-1): 1611 (vs), 1565 (vs), 1400 (vs), 1156 (m), 1010
(m), 768 (vs), 736 (s), 446 (m).
[Mn3(µ-RCOO)6(Me2-bpy)2], R ) 2-ClPh (7), 3-ClPh (8), 4-ClPh
(9). [Mn(RCOO)2]‚nH2O (0.50 g, 1.2 mmol) and 4,4′-dimethyl-2,2′-
bypiridine (Me2-bpy) (0.18 g, 1.0 mmol), both dissolved in absolute
ethanol, were mixed with constant stirring. The total volume of EtOH
used was ca. 120 mL. The yellow solid formed was isolated by
filtration, washed in ethanol and ether, and dried in air. Yield: 85-
95%. Anal. Calcd for complex 7, C66H48Cl6Mn3N4O12: C, 54.02; H,
3.27; N, 3.82; Cl, 14.53. Found: C, 53.9; H, 3.4; N, 3.8; Cl, 14.3.
FT-IR data (KBr, main bands, cm-1): 1618 (vs), 1540 (s), 1450 (m),
1407 (vs), 1065 (m), 815 (m), 755 (s), 657 (m). Anal. Calcd for
complex 8, C66H48Cl6Mn3N4O12: C, 54.02; H, 3.27; N, 3.82; Cl, 14.53.
Found: C, 54.0; H, 3.3; N, 3.7; Cl, 14.1. FT-IR data (KBr, main bands,
cm-1): 1611 (vs), 1555 (vs), 1420 (m), 1394 (vs), 834 (m), 762 (s),
700 (m). Anal. Calcd for complex 9, C66H48Cl6Mn3N4O12: C, 54.02;
H, 3.27; N, 3.82; Cl, 14.53. Found: C, 53.6; H, 3.4; N, 3.8; Cl, 14.2.
(39) Borra´s-Almenar; J. J.; Burriel, R.; Coronado, E.; Gatteschi, D.; Go´mez-
Garc´ıa, C. J.; Zanchini, C. Inorg. Chem. 1991, 30, 947.

790 Inorganic Chemistry, Vol. 37, No. 4, 1998
Albela et al.
FT-IR data (KBr, main bands, cm-1): 1607 (vs), 1560 (s), 1415 (vs),
1097 (m), 1018 (m), 782 (s), 540 (m).
Table 1. Crystallographic Data for 4 and 10
4
10
[Mn(µ-3-ClPhCOO)2(bpy)]n‚nH2O (10). A solution of [Mn(3-
ClPhCOO)2]‚nH2O (0.40 g, 1.0 mmol) in absolute ethanol (50 mL)
was treated with 2,2′-bipyridine (0.16 g; 1.0 mmol) in EtOH (50 mL).
The resulting yellow solution was left undisturbed for 24 h. The yellow
microcrystalline solid formed corresponded to trinuclear complex 6.
This was filtered, and when the filtrate was left to stand for several
days in the open air, a new product formed, whose analysis indicated
the formula [Mn(µ-3-ClPhCOO)2(bpy)]n‚nH2O. The crystals obtained
were suitable for X-ray characterization. The yield was 30%. Anal.
Calcd for complex 10, C24H18Cl2MnN2O5: C, 53.33; H, 3.33; N, 5.18;
Cl, 13.14. Found: C, 53.3; H, 3.3; N, 5.1; Cl, 13.0. FT-IR data (KBr,
main bands, cm-1): 1644 (vs), 1598 (vs), 1565 (vs), 1442 (m), 1419
(m), 1394 (vs), 1071 (m), 1019 (m), 762 (vs), 735 (s), 650 (m).
empirical formula
fw
temp, K
space group
a, Å
b, Å
c, Å
C54H42Cl2Mn2N8O12
1175.74
293(2)
C24H18Cl2MnN2O5
540.26
108
C2/c (No 15)
26.376(5)
12.404(3)
7.095(1)
90.00
96.10(3)
90.00
2308.1(1.5)
4
0.710 73
1.555
8.202
0.041
P1h (No. 2)
9.1886(10)
11.6135(9)
13.595(2)
66.225(13)
84.073(12)
88.593(10)
1320.3(2)
1
0.710 73
1.479
6.51
0.067
0.120
R, deg
â, deg
γ, deg
V, Å3
Z
λ(Mo KR), Å
d
cal, g cm-3
[Mn(µ-4-ClPhCOO)2(bpy)]n‚2nH2O (11). 2,2′-bipyridine (0.16 g,
1.0 mmol) in EtOH (50 mL) was added to a solution of [Mn(4-
ClPhCOO)2]‚nH2O (0.40 g, 1.0 mmol) in ethanol (50 mL) with constant
stirring. The solution turned yellow immediately and was stored at
room temperature for 24 h. A yellow microcrystalline solid formed,
which was isolated by filtration, washed in absolute ethanol and ether
and dried in air. The yield was 75%. Anal. Calcd for complex 11,
C24H20Cl2MnN2O6, C, 51.61; H, 3.58; N, 5.01; Cl, 12.72. Found: C,
51.6; H, 3.7; N, 5.1; Cl, 12.7. FT-IR data (KBr, main bands, cm-1):
1630 (vs), 1585 (s), 1555 (s), 1445 (s), 1413 (vs), 1176 (m), 1091 (s),
1019 (s), 769 (vs), 742 (m), 532 (m).
µ(Mo KR), cm-1
Ra
wR2b (4); wRc (10)
0.050
a R (on F) ) ∑|Fo| - |Fc||/∑|Fo|. b wR2 (on F2) ) [∑w(|Fo| -
2
2
4
2
|Fc| )2/∑w|Fo| ]1/2
.
c wR (on F) ) [∑w(|Fo| - |Fc|)2/∑w|Fo| ]1/2
.
peaks in the final difference maps were 0.313 and -0.265 e Å-3
,
respectively. Figure 1 was drawn using the program PLATON.43
Complex 10. The crystals of complex 10 were light yellow, thin
plates. A crystal of 0.57 × 0.18 × 0.05 mm was chosen and X-ray
diffraction data were collected at 108 K with an Enraf-Nonius CAD-4
diffractometer using graphite-monochromated Mo KR radiation (λ )
0.710 73 Å). Unit cell parameters (Table 1) were determined from
least-squares refinement of the setting angles of 24 reflections with 2θ
angles in the range 31-49°. A total of 2019 unique reflections were
recorded within 2θ < 50°. Three reference reflections monitored
throughout the data collection showed an average intensity loss of 6%.
The data were corrected for Lorentz and polarization effects and for
linear decay. Experimental absorption corrections based on ψ-scans
of 7 reflections were carried out.44 The intensity statistics suggested a
centrosymmetric space group, thus C2/c was chosen rather than Cc.
The structure was solved by direct methods45 and successive Fourier
syntheses. All non-hydrogen atoms were refined anisotropically. The
extinction parameter was 2.659 × 10-7. The positions of all hydrogen
atoms were revealed in a difference Fourier map. The hydrogen bonded
to the water oxygen was refined isotropically, while hydrogen atoms
bonded to carbon were included at calculated, idealized positions, and
not refined. The final full-matrix least-squares refinements on F,
minimizing Σw(|Fo| - |Fc|)2, including 1748 reflections with I > 2σ,
adjusting 160 parameters, converged at R and Rw indices of 0.041 and
0.050. In the final difference map the residual maxima and minima
were 0.57 and -0.90 e Å-3. All calculations were carried out with
Spectral and Magnetic Measurements. Analysis of C, H, N, and
Cl were carried out by the “Serveis Cient´ıfico-Te`cnics” of the
Universitat de Barcelona. Infrared spectra (4000-400 cm-1) were
recorded from KBr pellets in a Nicolet Impact 400 FT-IR spectrometer.
Magnetic susceptibility measurements were performed on a MANICS-
DSM8 susceptometer equipped with an Oxford Instruments liquid
helium cryostat, working down to 4.2 K. Pascal’s constants were used
to estimate the diamagnetic corrections for each complex. EPR spectra
were recorded at X-band (9.4 GHz) frequencies with a Bruker ESP-
300E spectrometer from room temperature to 4 K. Both magnetic and
EPR measurements were carried out at the Servei de Magnetoqu´ımica
(Universitat de Barcelona).
X-ray Diffraction Data Collection and Refinement. Complex 4.
A yellow-red crystal of 4 (0.46 × 0.23 × 0.15 mm) was mounted on
a Stoe AED2 four-circle diffractometer, using graphite-monochromated
Mo KR radiation (λ ) 0.710 73 Å). Unit cell parameters were
determined by least-squares from the (ω values of 27 reflections in
the range 14 < θ < 18.5°. In all, 4635 independent reflections were
measured in the range 2.23 < θ < 25° at 293(2) K; 3289 reflections
[I > 2σ(I)] were considered observed. Three standard reflections were
measured every hour and showed a 3% intensity variation. The
extinction correction was 0.033(2). No correction for absorption was
made. The oxygen atoms of the perchlorate group were disordered
over two possible sites. Each group of four O atoms was refined with
an overall occupancy factor, final values 0.492 and 0.508. Crystal-
lographic data are given in Table 1. The crystal structure was solved
by direct methods and Fourier syntheses using the program SHELXS40
programs in the MolEN system.46 Neutral atomic scattering factors
were used,47 and anomalous scattering terms were included in Fcalc
.
48
Crystallographic data are given in Table 1. Complete lists of crystal
parameters, atomic coordinates, anisotropic thermal parameters, in-
tramolecular bond distances and angles, least-squares planes, torsion
angles and structure factors are given in the Supporting Material.
X-ray Absorption Data Recording and Processing. The XANES
(X-ray absorption near-edge structures) and EXAFS (extended X-ray
absorption fine structures) data were collected at LURE (Laboratoire
and refined by the full-matrix least-squares method on F2, using the
2
program SHELXL-93.41 The function minimized was Σw(|Fo|
-
2
2
|Fc| )2, where w ) |σ2(I) + (0.0227P)2 + (2.61P)|-1 and P ) (|Fo| +
2
2|Fc| )/3. Complex neutral atom scattering factors were taken from
(43) Spek, A. L. PLATON. Acta Crystallogr. 1990, A46, C34.
(44) North, A. C. T.; Phillips, D. C.; Mathews, F. S. Acta Crystallogr.,
Sect. A 1968, 24, 351.
(45) Main, P.; Fiske, S. J.; Hull, S. E.; Lessinger, L.; Germain, G.; DeClerq,
J. P.; Woolfson, W. W. MULTAN80, A System Of Computer Programs
for the Automatic Solution of Crystal Structures from X-ray Diffraction
Data; University of York: England, 1980.
International Tables for Crystallography, Vol. C.42 The hydrogen
atoms were included in calculated positions as riding atoms using
SHELXL-9341 default parameters, and 390 parameters were refined
for 4611 reflections. Refinement converged at R(on F) 0.067, wR(on
F2) 0.120, for 3289 observed reflections. The maximun shift/esd and
the mean shift/esd were -0.002 and 0.000. Maximum and minimum
(46) MolEN, An InteractiVe Structure Solution Procedure; Enraf-Nonius:
Delft, The Netherlands, 1990.
(47) (a) Cromer, D. T. and Waber, J. T. International Tables for X-ray
Crystallography; The Kynoch Press: Birmingham, England, 1974;
Vol. IV; Vol. C, Table 2.2B. (b) Cromer, D. T.; Mann, J. B. Acta
Crystallogr., Sect. A 1968, 24, 321.
(48) Cromer, D. T. International Tables for X-ray Crystallography; The
Kynoch Press: Birmingham, England, 1974; Vol. IV, Table 2.3.1.
(40) Sheldrick, G. M. SHELXS. Acta Crystallogr. 1990, A46, 467.
(41) Sheldrick, G. M. SHELXL, Program for crystal structure refinement;
University of Go¨ttingen: Germany, 1993.
(42) International Tables for Crystallography; D. Reidel Publishing Co.:
Dordrecht, Holland, 1994; Vol. C.

Mn(II) Complexes with Chlorobenzoato Bridges
Inorganic Chemistry, Vol. 37, No. 4, 1998 791
d’Utilisation du Rayonnement Electromagnetique, Paris-Sud University,
France) on a storage ring with an energy of 1.85 GeV and a mean
intensity of 300-200 mA. The measurements were carried out at the
manganese K edge in the transmission mode on the EXAFS III
spectrometer equipped with a two-crystal monochromator (Si-311 and
Si-111 for XANES and EXAFS, respectively), the entrance slit being
0.5 mm for both types of spectrum. The monochromator was left
slightly out of tune to ensure harmonics rejection. Reduced-pressure,
air-filled ionization chambers were used to measure the flux intensity
before and after the sample. The spectra were recorded at 10 K in a
helium cryostat designed for X-ray absorption spectroscopy. The
XANES spectra were recorded over 150 eV step by step, every 0.25
eV with a 2 s accumulation time per point. The spectrum of a 5 µm
thick manganese foil was recorded just after or before an unknown
XANES spectrum to check the energy calibration, thus ensuring an
energy accuracy of 0.25 eV. The EXAFS spectra were recorded in
the same way over 1000-1200 eV, with 2 eV steps and 1 s
accumulation time per point. The experiments were calibrated by using
the 8991.1 eV peak at the top of the edge of a metallic foil of copper
and verifying that the first inflection point in all the spectra of the
manganese foils was 6539 eV. Each spectrum is the sum of several
independent recordings added after individual inspection (two for
XANES and four for EXAFS). Samples were well-pounded micro-
crystalline powders of a homogeneous thickness and calculated weight
that were compressed between two X-ray-transparent windows. The
thickness was computed to avoid saturation effects. The absorbance
jump at the edge was typically 1.
The XANES and standard EXAFS data were analyzed with the
“GALAAD” 49 and “EXAFS pour le MAC” 50 programs, following a
well-known procedure described elsewhere.51-53 The modeling of the
outer shells was not possible in the frame of the single scattering model,
and the multiple scattering/spherical wave EXAFS ab initio modeling
FEFF54,55 program was used. The amplitudes and phase shifts extracted
from FEFF calculations were then used to fit the outer shells. Multiple
scattering EXAFS contributions were included using the method
described in ref 56. The FEFF calculations were performed on a RISC
system/6000 running AIX 3.2.5 of the Physical Chemistry Department
of the University of Vale`ncia.
Figure 1. Drawing of the cation [Mn2(µ-PhCOO)2(bpy)4]2+ in complex
(4) showing the syn-anti conformation of the carboxylate bridges.
Hydrogen atoms have been omitted for clarity.
for R ) 4-ClPh, where the [Mn(µ-4-ClPhCOO)2(bpy)]n‚nH2O
(11) chain was obtained. The dinuclear complex (3) precipitated
only in the presence of excess NaClO4 in the solution. For R
) 3-ClPh, when the dinuclear complex (2) was collected, a new
product was formed later in the filtrate, which did not present
the perchlorate anion and corresponded to the chain (10).
For the trinuclear compounds, in the synthetic procedure given
by Shake6 for the benzoate complex, [Mn(PhCOO)2]‚nH2O and
2,2′-bipyridine (bpy) were used as starting materials; MnII/bpy
ratios of 1:2/3 and 1:1 were considered. However, in the
synthesis of the acetate analog reported by Me´nage et al.58
a
1:1 ratio was used. Both ratios were followed here with the
corresponding manganese(II) carboxylates, and only the tri-
nuclear compounds [Mn3(µ-RCOO)6(bpy)2] with R ) 2-ClPh,
3-ClPh (5, 6) were isolated. For R ) 4-ClPh the chain (11)
was again obtained. The analogous chain for R ) 3-ClPh (10)
was obtained after the trinuclear complex (16) was collected
by filtration. This synthesis was repeated using 4,4′-dimethyl-
2,2′-bipyridine (Me2-bpy) instead of bpy, and the trinuclear
complex was isolated in all cases, even for the 4-chlorobenzoate
(7-9). For this series, formation of the corresponding chain
was not observed.
Results and Discussion
Syntheses. In our previous paper57 we reported that treatment
of the MnIII complexes [MnIII2(µ-O)(µ-RCOO)2(bpy)2(H2O)2]-
(NO3)2, (R ) 2-ClPh, 3-ClPh, 4-ClPh) with hydrogen peroxide
followed by addition of NaClO4 leads to formation of the
dinuclear MnII complex [MnII (µ-RCOO)2(bpy)4](ClO4)2, but in
2
Description of Crystal Structure of [Mn2(µ-PhCOO)2-
(bpy)4](ClO4)2 (4). The structure of [Mn2(µ-PhCOO)2(bpy)4]-
(ClO4)2 (4)59 is depicted in Figure 1. Selected bond lengths
and angles are listed in Table 2. The molecule contains a
crystallographic inversion center, relating both halves of the
dinuclear unit. The perchlorate anions are disordered. Man-
ganese ions are bridged by two µ-benzoate groups in a syn-
anti fashion, that is, the lone pair on the carboxylate oxygens
are opposite each other. Six-coordination on each manganese
ion is completed by two chelating bpy groups; the geometry at
each metal center is distorted octahedral. The bond lengths from
the oxygen of the carboxylate (average dMn-O ) 2.128 Å) are
shorter than distances to the nitrogens (average dMn-N ) 2.273
Å). Further, the distances from the N trans to the oxygen of
the carboxylate are slightly larger (2.279, 2.292 Å) than those
from the N cis to the oxygen (2.258, 2.265 Å), probably caused
by the shorter Mn-O distances. The nonbonding interatomic
Mn‚‚‚Mn distance is 4.509 Å, similar to Mn‚‚‚Mn distances
found in analogous complexes in which carboxylate groups also
the reaction without ClO4- two different neutral complexes were
obtained. In order to improve both yield and quality of the
products and thus better characterize them, since these manga-
nese(II) products were often found as byproducts of the synthesis
of the dinuclear MnIII complexes, a direct procedure was studied.
In the synthesis of the dinuclear compounds, [Mn2(µ-RCOO)2-
-
(bpy)4](ClO4)2 (1-4), a 1:2:1 ratio of MnII/bpy/ClO4 was
considered. The desired product was isolated in all cases except
(49) Noinville, V.; Michalowicz, A. GALAAD; Socie´te´ Franc¸aise de
Chimie: Paris, 1991; pp 116-117.
(50) Michalowicz, M. EXAFS pour le MAC; Socie´te´ Franc¸aise de Chimie:
Paris, 1991; pp 102-103.
(51) Teo, B. K. EXAFS: basic Principles and Data Analysis; Springer-
Verlag: Berlin, 1986; Vol. 9.
(52) Ko¨nigsberger, D. C.; Prins, R. X-Ray Absorption Principles, Applica-
tions, Techniques of EXAFS, SEXAFS and XANES; Wiley: New York,
1988.
(53) Real, J. A.; Castro, I.; Bousseksou, A.; Verdaguer, M.; Burriel, R.;
Castro, M.; Linares, J.; Varret, F. Inorg. Chem. 1997, 36, 455-464.
(54) Rehr, J. J.; Zabinsky, S. I.; Albers, R. C. Phys. ReV. Lett. 1992, 69,
3397.
(55) Rehr, J. J. Jpn. J. Appl. Phys. 1993, 32, 8.
(58) Me´nage, S.; Vitols, S. E.; Bergerat, P.; Codjovi, E.; Kahn, O.; Girerd,
J. J.; Guillot, M.; Solans, X.; Calvet, T. Inorg. Chem. 1991, 30, 2666.
(59) Stoeckli-Evans, H. Institut de Chimie, Universite´ de Neuchaˆtel,
Switzerland.
(56) Michalowicz, A.; Moscovici, J.; Ducourant, B.; Cracco, D.; Kahn, O.
Chem. Mat. 1995, 7, 1833-1842.
(57) Albela, B.; Corbella, M.; Ribas, J. Polyhedron 1996, 15, 91.

792 Inorganic Chemistry, Vol. 37, No. 4, 1998
Albela et al.
Table 2. Selected Interatomic Distances (Å) and Angles (deg) for
Complex [Mn2(µ-PhCOO)4(bpy)2](ClO4)2 (4) with Estimated
Standard Deviations (Esd’s) in Parentheses
Mn(1)-Mn(1) (i)a
N(2)-Mn(1)
N(4)-Mn(1)
O(2)-Mn(1)
C(21)-O(1)
4.509(2)
2.279(4)
2.265(4)
2.118(4)
1.257(6)
N(1)-Mn(1)
N(3)-Mn(1)
O(1)-Mn(1)
C(21)-O(2)
2.258(4)
2.292(4)
2.139(4)
1.242(6)
N(1)-Mn(1)-N(2)
N(1)-Mn(1)-N(3)
72.5(2) N(1)-Mn(1)-N(4) 168.7(2)
97.3(2) N(2)-Mn(1)-N(3)
93.1(2)
72.1(2)
N(4)-Mn(1)-N(2) 103.5(2) N(4)-Mn(1)-N(3)
O(1)-Mn(1)-N(1)
O(1)-Mn(1)-N(3)
O(2)-Mn(1)-N(1)
O(2)-Mn(1)-N(3) 167.2(2) O(2)-Mn(1)-N(4)
O(2)-Mn(1)-O(1) 103.6(2)
89.4(2) O(1)-Mn(1)-N(2) 161.39(14)
84.7(2) O(1)-Mn(1)-N(4)
92.6(2) O(2)-Mn(1)-N(2)
93.4(2)
82.2(2)
97.4(2)
a Symmetry transformation used to generate equivalent atoms: (i)
-x + 2, -y, -z + 1.
bridge in a syn-anti manner: [Mn2(µ-MeCOO)2(bpy)4](ClO4)2
(4.583 Å)2,6; [Mn2(µ-MeCOO)2(L)2](ClO4)2, L ) N,N′-dimethyl-
N,N′-bis(2-pyridylmethyl)ethane-1,2-diamine (4.298 Å)8; [Mn2(µ-
MeCOO)2(tpa)2](TCNQ)2‚2MeCN, tpa ) tris(2-pyridylmethyl)-
amine, TCNQ ) tetracyanoquinodimethane (4.145 Å).7
X-ray Absortion Studies of [Mn2(µ-3-ClPhCOO)2(bpy)4]-
(ClO4)2 (2). Attempts to obtain X-ray-quality crystals failed
for all other dinuclear complexes of this series with ClPhCOO-
as a ligand (1-3). Therefore, we decided to study the XANES
and EXAFS spectra of one of them (2) as a representative
example of this family of compounds in order to confirm that
they have the same structure as compound 4. In fact, the use
of X-ray absorption techniques in this structural area is not new
and they have been previously revelead as very useful tool in
coordination chemistry.53,60-63 The normalized XANES spec-
trum for complex 2 is shown in Figure 2a. The intensity and
energy of the K edge are typical of a MnII ion in an octahedral
surrounding,64 and the features at the top of the edge clearly
indicate a distorted compressed octahedral geometry.65 On the
other hand, the small shoulder observed at ca. 55 eV from the
pre-edge, which may be assigned to bielectronic transitions,66
is also typical of MnII compounds. The Fourier transform of
the experimental EXAFS spectrum is shown in Figure 3a. Each
atomic shell surrounding the metal ion is represented by a peak
on the Fourier transform. In this case, the Fourier transform
comprised four main peaks: the first one corresponded to the
four nitrogen atoms from two bpy and two oxygen atoms from
the carboxylates; the second and third peaks included essentially
the remaining carbon atoms of the amines and carboxylates;
and the last feature was tentatively assigned to the intramolecular
Mn‚‚‚Mn distance as the major contribution.
Figure 2. Normalized XANES spectra at the manganese K edge at
10 K for (a) [Mn2(µ-3-ClPhCOO)2(bpy)4](ClO4)2 (2), (b) [Mn(µ-4-
ClPhCOO)2(bpy)]n‚nH2O, (11), and (c) [Mn3(µ-3-ClPhCOO)6(bpy)2]
(6).
impossible in the frame of the single scattering approach of the
standard EXAFS formula, due to the multiple scattering. That
being so, a FEFF calculation using [Mn2(µ-PhCOO)2(bpy)4]-
(ClO4)2 (4) as a model compound was carried out. The complete
set of Cartesian coordinates was used for these calculations.
The FEFF model was compared with the experimental data
(Figure 3b) and showed good agreement, which confirmed that
the structure of the two compounds is similar. A quantitative
analysis of the first shell and the Mn‚‚‚Mn intramolecular
distance was then performed for complex 2 by using the
amplitudes and phase shifts extracted from the FEFF calculation.
The resulting average distances from the Mn were as follows:
two O atoms at 2.18 Å, four N atoms at 2.28 Å, and one Mn at
4.6 Å. Least-squares refinement resulted in the fit shown in
Figure S1 (Supporting Information), and the computed results
are listed in Table S1. These distances were similar to those
of complex 4 (average Mn-O ) 2.13 Å, average Mn-N )
2.27 Å, Mn‚‚‚Mn ) 4.509 Å), although the Mn-O and Mn-
Mn distances were somewhat larger, possibly due to the effect
of the chloro-substituent.
Because all complexes presented similar magnetic properties
(see below), the same structure, i.e. [Mn2(µ-RCOO)2(bpy)4]-
(ClO4)2, might be proposed for all the chloro-derivative
complexes. Besides, the analyses and IR data were in good
agreement with this formula.
A complete quantitative discussion of the local structure
around the metal ions including atoms in a sphere of 5 Å is
Description of Crystal Structure of [Mn(µ-3-ClPhCOO)2-
(bpy)]n‚nH2O (10).67 The structure consists of an infinite chain,
as shown in Figure 4. Bond lengths and angles are gathered in
Table 3. The manganese atom is located on a 2-fold axis and
presents a distorted octahedral coordination sphere, which
consists of the two N atoms of a bipyridine (bpy) ligand (dMn-N
) 2.277 Å) and four carboxylates, two trans to each other and
cis to the N of the bpy (dMn-O(1) ) 2.196 Å) and two cis to
each other and trans to the N of the amine (dMn-O(2) ) 2.104
Å). The Mn-O distance for the carboxylates trans to the amine
are shorter, due probably to the larger Mn-N bond, as was
observed for the dinuclear compound. The 3-chlorobenzoate
ligands bridge centrosymmetrically related manganese atoms.
(60) Gadet, V.; Mallah, T.; Castro, I.; Verdaguer, M. J. Am. Chem. Soc.
1992, 114, 9213.
(61) Lloret, F.; Ruiz, R.; Julve, M.; Faus, J.; Journaux, Y.; Castro, I. Chem.
Mat. 1992, 4, 1150.
(62) Lloret, F.; Ruiz, R.; Cervera, B.; Castro, I.; Julve, M.; Faus, J.; Real,
J. A.; Sapin˜a, F.; Journaux, Y.; Colin, J. C. J. Chem. Soc., Chem.
Commun. 1994, 2615.
(63) Ruiz, R.; Surville- Barland, C.; Journaux, Y.; Colin, J. C.; Castro, I.;
Cervera, B.; Julve, M.; Lloret, F.; Sapin˜a, F. Chem. Mater. 1997, 9,
201-209.
(64) Nakatani, K.; Carriat, J. Y.; Journaux, Y.; Kahn, O.; Lloret, F.; Renard,
J. P.; Pei, Y.; Sletten, J.; Verdaguer, M. J. Am. Chem. Soc. 1989,
111, 5739.
(65) Cartier, C.; Verdaguer, M.; Menage, S.; Girerd, J. J.; Tuchagues, J.
P.; Mabad, B. J. Phys. 1986, 47, 623-625.
(66) Cartier, C. Ph.D. Thesis, Centre Universitaire Paris-Sud, Orsay, France,
1988.
(67) This structure was solved by J. Sletten, Department of Chemistry,
University of Bergen, Norway.

Mn(II) Complexes with Chlorobenzoato Bridges
Inorganic Chemistry, Vol. 37, No. 4, 1998 793
Figure 3. Fourier transform of the experimental EXAFS spectrum (left) and comparison of the experimental (dashed line) and calculated (solid
line) k space EXAFS spectra (right) for [Mn2(µ-3-ClPhCOO)2(bpy)4](ClO4)2 (2) using the FEFF method on the model compound [Mn2(µ-PhCOO)2-
(bpy)4](ClO4)2 (4) at the manganese K edge.
stabilizing the zigzag chain structure. A similar zigzag arrange-
ment has recently been described for [Mn(µ-Cl)2(bpy)]n,68 where
the chloride ions occupy the position of the carboxylates in the
chain studied here. Furthermore, the hydrogen bond with the
O(1) of the carboxylate may cause the Mn-O(1) distance to
be slightly larger than that with the O(2), which is not hydrogen-
bonded to the water molecule, and the octahedral environment
of the metal becomes distorted.
X-ray Absortion Studies of [Mn(µ-4-ClPhCOO)2-
(bpy)]n‚nH2O (11). In the absence of the crystal structure for
the complex with 4-ClPhCOO- as carboxylate (11), XANES
and EXAFS spectroscopies were used to characterize it structur-
ally. The XANES spectrum for 11 is shown in Figure 2b. The
intensity and energy of the K edge is similar to the dinuclear
manganese(II) complex 2 and the shoulder at ca. 55 eV from
the pre-edge also appears. The top of the edge is structured,
Figure 4. Plot of the one-dimensional complex [Mn(µ-3-ClPhCOO)2-
and the shape suggests that the local environment of MnII metal
ion is octahedral elongated.65 As in compound 2, four main
peaks are observed in the experimental Fourier transform (Figure
5a): the first one corresponds to the first coordination sphere
of the metal, the second and third peaks include mainly the
remaining carbon atoms of the ligands, and the fourth one might
be tentatively assigned to distances to the closest Mn as the
greater contribution. Only the distance with the first Mn was
tentatively assigned, since a second Mn may be located quite
far from the absorber atom to have a considerable contribution
in the spectrum. A FEFF calculation for complex [Mn(µ-3-
ClPhCOO)2(bpy)]n‚nH2O (10) was performed, using the com-
plete set of Cartesian coordinates. The superimposability of
the spectrum calculated for this compound and the experimental
one of 11 indicates the good quality of the model considered
(Figure 5b). The amplitudes and phase shifts deduced from
the FEFF calculation were used to fit the experimental data of
complex 11 (4-ClPh). The resulting average distances were the
following: Mn-O ) 2.14 Å, Mn-O ) 2.20 Å, Mn-N ) 2.30
Å, and Mn‚‚‚Mn ) 4.6 Å, which are similar to those of [Mn(µ-
3-ClPhCOO)2(bpy)]n‚nH2O (10) (Table 3). Least-squares re-
finement resulted in the fitting shown in Figure S1 (Supporting
Information), and the computed results are listed in Table S1.
X-ray Absortion Studies of [Mn3(µ-3-ClPhCOO)6(L)2], L
) bpy, Me2-bpy (6, 8). In the absence of crystal structure of
any trinuclear compound (5-9), a formula [Mn3(µ-RCOO)6-
(L)2] was proposed on the basis of the analyses, IR data, and
the study of the magnetic properties (see below). This structure
was confirmed by the analysis of the XANES and EXAFS
spectra at the Mn K edge. Two of these complexes were
(bpy)]n‚nH2O, (10). The benzenic rings of the µ-3-ClPhCOO- groups
are omitted for clarity.
Table 3. Main bond Distances (Å) and Angles (deg) for
[Mn(µ-3-ClPhCOO)2(bpy)]n‚nH2O (10) with Estimated Standard
Deviations (Esd’s) in Parenthesesa
Mn-MnA
Mn-O(2)
O(1)-C(7)
4.515(1)
2.104(2)
1.260(3)
Mn-O(1)
Mn-N
O(2)-C(7)
2.196(2)
2.277(2)
1.261(3)
O(1)-Mn-O(1)B 168.88(6) O(2)A-Mn-O(2)C 111.77(7)
N-Mn-NB
71.27(7) O(1)-Mn-O(2)A
87.05(6) O(1)-Mn-O(2)C
83.92(6) O(2)A-Mn-N
89.05(7)
100.52(6)
85.76(6)
158.16(7)
O(1)-Mn-N
O(1)-Mn-NB
O(2)A-Mn-NB
a Symmetry codes: A ) 1 - x, -y, -z; B ) 1 - x, y, 1/2 - z; C )
1
x, -y, z + /2.
In this way, infinite zigzag chains consisting of doubly car-
boxylate-bridged manganese atoms are formed (Figure 4). The
Mn‚‚‚Mn distance within the chain is 4.515 Å. This value is
comparable with the dMn-Mn of 4.509 Å in the former complex,
[Mn2(µ-PhCOO)2(bpy)4](ClO4)2 (4), as expected since in the
chain the two carboxylate bridges also present a syn-anti
conformation. Indeed, the structure of the dinuclear complex
is observed within the chain, which might be considered as a
polymer of “Mn(µ-RCOO)2Mn” units bridged by two µ-car-
boxylates. So, such similarity between both structures might
explain the finding that, when we attempted to synthesize the
dinuclear complex [Mn2(µ-RCOO)2(bpy)4](ClO4)2, the [Mn(µ-
RCOO)2(bpy)]n‚nH2O chain was also isolated in some cases.
Besides, the water of crystallization, which is situated on a 2-fold
axis, is hydrogen-bonded to two carboxylate O(1) atoms of
alternate manganese ions within the same chain (dO(1)-H ) 1.957
Å, dO(1)-O(3) ) 2.864 Å, O(3)-H(3)‚‚‚O(1) ) 153°), thus
(68) Lubben, M.; Meetsma, A.; Feringa, B. L. Inorg. Chim. Acta 1995,
230, 169.

794 Inorganic Chemistry, Vol. 37, No. 4, 1998
Albela et al.
Figure 5. Fourier transform of the experimental EXAFS spectrum (left) and comparison of the experimental (dashed line) and calculated (solid
line) k space EXAFS spectra (right) for [Mn(µ-4-ClPhCOO)2(bpy)]n‚nH2O (11) using the FEFF method on the model compound [Mn(µ-3-ClPhCOO)2-
(bpy)]n‚nH2O (10) at the manganese K edge.
Chart 1
main peaks were observed in the Fourier transforms and the
last feature might be tentatively assigned to distances Mnc-
Mnt as the larger contribution.
A FEFF calculation of the EXAFS spectrum was carried out
using the crystallographic coordinates of the benzoate analogue
[Mn3(µ-PhCOO)6(bpy)2]6 as a theoretical model. This simula-
tion was difficult, since the two terminal Mn and the central
ion showed different contributions and average values were
considered. Therefore, the EXAFS study of this system may
be more complicated than that of the former dinuclear and
monodimensional complexes, and only average distances may
be derived. Figure 6b displays a comparison between the
experimental data of complex 6 and the theoretical model. The
fit of the filtered first and third main peaks of the Fourier
transform, considering the parameters obtained from the FEFF
calculations for this model, gave Mn-O distances between 2.09
and 2.28 Å, Mn-N distances of 2.23-2.25 Å, and a Mn‚‚‚Mn
distance of 3.6 Å in both cases. Least-squares refinement
resulted in the fit shown in Figure S1 (Supporting Information),
and the computed results are listed in Table S1. The values
obtained are similar to the average of the distances around the
Mn atom for the model complex [Mn3(µ-PhCOO)6(bpy)2]6
(Mnterminal-O ) 2.075, 2.098, 2.274, 2.286 Å; Mncentral-O )
2.128, 2.170, 2.239 Å; Mn-N ) 2.261, 2.268 Å; Mnt-Mnc )
3.588 Å), thereby confirming the proposed structure, i.e.,
[Mn3(µ-RCOO)6(bpy)2].
selected as representative compounds for such studies: one of
the bpy series, complex 6, and the corresponding complex of
the Me2-bpy series with the same carboxylate (8). The analog
with benzoate as a bridge [Mn3(µ-PhCOO)6(bpy)2] described
elsewhere6 was considered as a model. This structure consists
of a linear array, in which each pair of Mn atoms is bridged by
three carboxylates, showing two types of bonding mode. Two
of the benzoates bridge symmetrically in a syn-syn bidentate
fashion, whereas the third is described as “asymmetrical”, with
one carboxylate oxygen bridging the two metals and the second
bound to the terminal manganese atom. The central ion thus
exhibits an octahedral coordination sphere made of six oxygen
atoms from six different benzoate groups, while the terminal
ions present a distorted octahedral environment of four oxygen
atoms and two nitrogen atoms that come from a bpy molecule
(see Chart 1). The bridging mode found in this compound is
quite rare among carboxylate bridged complexes.
The XANES spectrum of complex 6 at the Mn K edge is
shown in Figure 2c. The spectrum is typical of manganese(II)
in an octahedral environment, as pointed out for the former
complexes. The top of the edge was featureless, which may
indicate that (i) the distortion of the octahedral environment is
weak69 or (ii) the spectrum is the result of the average
contributions of two opposite distorted manganese ions, one
compressed octahedral manganese and the other elongated. A
comparison between the XANES spectra of 6 and 8 (Figure
S2) reveals that the two edges are almost superimposable,
suggesting strong similarities in the geometry at the metal ion.
The similarity between the experimental EXAFS spectra and
the corresponding Fourier transforms for complexes 6 and 8
(Figure 6 and Figure S3) also suggests that the local structure
of the MnII ion might be the same. In this case, only three
Magnetic Susceptibility Study. Magnetic susceptibility data
were recorded for all complexes (1-12) from room temperature
to 4 K. For the dinuclear complexes (1-4) the value of øM
was 0.04 cm3 mol-1 at room temperature and increased with
temperature, reaching a maximum of øMT ≈ 0.35-0.38 cm3
mol-1 at 4-5 K, and then decreased, tending to zero at 0 K.
The øMT product began at ≈8-9 cm3 mol-1 K at ca. 300 K,
5
which was close to the value for two uncoupled S ) /2 ions
(8.75 cm3 mol-1 K). The øMT values decreased slowly with
temperature to 7.5 cm3 mol-1 K at 50 K, followed by a further
larger decrease, reaching a value of 1-2 cm3 mol-1 K at 4 or
2 K for complex 3. This behavior is typical for weak
antiferromagnetic coupling between two MnII centers.
The spin Hamiltonian H ) -J12S1S2 was used to fit the
magnetic results assuming that the two MnII ions present the
same g value. The parameters J and g obtained from least-
squares fitting of the øM Vs temperature data to the theoretical
equation,70 are listed in Table 4. All J values are similarly low.
The g parameter presented a value close to 2.0 in all cases,
(70) (a) Van Vleck, J. H. The Theory of Electric and Magnetic Susceptibili-
ties; Oxford University Press: London, 1932. (b) O’Connor, C. J. Prog.
Inorg. Chem. 1982, 29, 238
(69) Roe, A. L.; Schneider, D. J.; Mayer, R. J.; Pyrz, J. W.; Widom, J.;
Que, L. J. J. Am. Chem. Soc. 1984, 106, 1676.

Mn(II) Complexes with Chlorobenzoato Bridges
Inorganic Chemistry, Vol. 37, No. 4, 1998 795
Figure 6. Fourier transform of the experimental EXAFS spectrum (left) and comparison of the experimental (dashed line) and calculated (solid
line) k space EXAFS spectra (right) for [Mn3(µ-3-ClPhCOO)6(bpy)2] (6) using the FEFF method on the model compound [Mn3(µ-PhCOO)6(bpy)2]
at the manganese K edge.
Table 4. Parameters Obtained from the Fit of the Susceptibility
Data of Complexes 1-9
complex
J (cm-1
)
g
complex
J (cm-1
)
g
1
2
3
4
5
-1.73
-1.79
-1.70
-1.76
-2.73
2.03
2.05
2.00
2.00
2.01
6
7
8
9
-3.04
-3.22
-3.15
-2.07
2.02
2.02
2.03
2.04
which agrees with the presence of manganese(II) ions, which
do not exhibit spin-orbit coupling. From the magnetic point
of view, the coupling constant J, which reflects the manganese-
manganese interactions, is determined in a first approach by
the bridges between the two metal ions. It is generally
assumed14,71 that the carboxylates are ineffective in transmitting
the exchange interaction. Accordingly, complexes with only
carboxylate bridging ligands may exhibit small negative ex-
change coupling constants, as observed here. This is consistent
with the great distance between the two manganese ions found
in these cases. J values reported in the literature for manga-
nese(II) complexes with only carboxylates as bridging ligands
are always weak, between -0.2 and -5 cm-1 approxi-
mately.4,5,7,11,12,31
Figure 7. øMT Vs T plot for [Mn3(µ-RCOO)6(L)2] (R ) 2-ClPh, 3-ClPh,
4-ClPh; L ) bpy, Me2-bpy) (5-9). The solid line is the best fit to the
experimental data.
For the trinuclear complexes (5-9) the value of øM was ≈0.05
cm3 mol-1 at room temperature and increased continuously with
temperature. Since a maximum in the øM versus temperature
plot was not observed, øMT provided the best fit to a theoretical
equation. For all complexes, øMT at room temperature (≈12
cm3 mol-1 K at 290 K) was smaller than the expected value
Scheme 1
5
for three MnII atoms uncoupled with Si ) /2 and gi ) 2 each
(øMT ) 13.125 cm3 mol-1 K). This indicates antiferromagnetic
coupling, which was confirmed by the decrease in øMT when T
decreased, reaching 4.3 cm3 mol-1 K at 4.2 K, which corre-
sponds (within experimental uncertainty) to the expected value
In fact, all authors fix J13 ) 0 to fit the experimental
susceptibility data,26,29,73 owing to the long distance between
the two terminal MnII ions and the linearity of the known
structures (with acetate and benzoate).
5
øMT ) 4.375 cm3 mol-1 K for a spin state S ) /2 with g ) 2
(Figure 7). Similar behavior has been reported for the acetate26
and benzoate6 analogues.
From the Van Vleck formula,70a the susceptibility expression
for three linear MnII (d5) ions is obtained,6 and this equation
was incorporated into a nonlinear, least-squares computer
program, which was then used to fit the experimental øMT versus
temperature data as a function of an exchange coupling
parameter J and an isotropic g value. The solid lines in Figure
7 represent the best fit obtained for each complex. The
corresponding J and g values are listed in Table 4. The J values
obtained range between -2.1 and -3.2 cm-1, which are of the
same order as the values found for similar complexes: [Mn3(µ-
MeCOO)6(bpy)2]26 (J ) -4.4 cm-1) and [Mn3(µ-MeCOO)6-
Taking into account the structure of these compounds, the
following Kambe vector-coupling scheme (Scheme 1)72 was
considered:
The spin Hamiltonian that describes the low-lying electronic
states is given by eq 1, where it is assumed that J12 ) J23 ) J.
HS ) -J(S1S2 + S2S3) -J13S1S3
(1)
(71) Hartman. J. A. R.; Rardin, R. L.; Chaudhuri, P.; Pohl, K.; Wieghardt,
K.; Nuber, B.; Weiss, J.; Papaefthymiou, G. C. J. Am. Chem. Soc.
1987, 109, 7387.
(73) Baldwin, M. J.; Kampf, J. W.; Kirk, M. L.; Pecoraro, V. L. Inorg.
Chem. 1995, 34, 5252.
(72) Kambe, K. J. Phys. Soc. Jpn. 1950, 5, 48.

796 Inorganic Chemistry, Vol. 37, No. 4, 1998
Albela et al.
(L)2], L ) BIPhMe29 (J ) -5.6 cm-1). This indicates that weak
antiferromagnetic coupling is mediated by the bridging car-
boxylate in these complexes, which is as expected, since the
Mn-O distances are relatively long in such compounds.26,29 The
coupling is even weaker than the exchange observed for
hydroxide- and alkoxide-bridged species,29 as seen for the
dinuclear MnII compounds. As pointed out by Baldwin et al.,73
it does not appear that the different carboxylate bridging modes
have a significant effect on the magnetic coupling in these
5
systems. The J values obtained lead to an (S13,ST) ) (5, /2)
spin ground state, which is energetically separated from the (4,
7
5
7
3/2) and (5, /2) lowest excited states by /2J and /2J, respec-
tively.26,66 As the J values are quite small, the first excited states
may lie close to the ground state, and at room temperature many
states may be populated.
For complexes 10 and 11 the value of øM was ≈0.025 cm3
mol-1 at room temperature, and increased continuously with
temperature (Figure 8) giving a maximum close to 10 K for 10
and at ca. 4 K for 11. øMT values decreased monotonically
from ≈4.5 cm3 mol-1 K at 290 K (a typical value for an S )
5/2 ion, assuming g ) 2.0). The shape of the two curves clearly
indicates that complex 10 showed greater antiferromagnetic
coupling than 11.
As reported in the literature for similar one-dimensional
manganese(II) complexes,74 the magnetic susceptibility data for
complexes 10 and 11 may be analyzed according to two
isotropic Heisenberg models for linear chains: (a) the scaling
model of Wagner and Friedberg32 with S ) 5/2 and (b) the Weng
model75 with coefficients generated by Hiller et al.76 for S )
5/2.
Figure 8. øM Vs temperature plot for (A) [Mn(µ-3-ClPhCOO)2-
(bpy)]n‚nH2O (10) and (B) [Mn(µ-4-ClPhCOO)2(bpy)]n‚nH2O (11). The
dotted and solid lines are the best fit to the experimental data using
the Wagner-Friedberg and Weng-Hiller models, respectively.
Table 5. Parameters Obtained from the Fit of the Susceptibility
Data of Complexes 10 and 11
Wagner-Friedberg Model
Weng-Hiller Model
complex
J (cm-1
)
g
J (cm-1
)
g
10
11
-1.72
-0.74
2.05
2.1
-1.58
-0.72
2.05
2.1
The isotropic Heisenberg Hamiltonian considered in both
cases was
EPR Spectra. The three families of manganese(II) com-
pounds studied show weak Mn‚‚‚Mn interactions, so even at
low temperatures a number of spin states may be populated.
This may lead to many possible transitions and thus a
complicated EPR spectrum. At room temperature all the
complexes showed a signal centered at g ∼ 2, which was quite
broad in general. However, at 4 K the spectrum of each type
of compound was quite different, which allowed us to identify
each kind of complex.
H ) -J ΣSiSj
According to the Wagner and Friedberg model (a), the molar
magnetic susceptibility is given by
Ng2â2S(S + 1) (1 + U)
øM )
×
(1 - U)
3kT
The spectrum at 4 K for the dinuclear complexes (1-4)
presented a broad signal centered at g ∼ 2, which was flanked
by one feature at ∼2800 G and another at ∼4200 G. Other
less intense signals were observed at lower field (Figure 9A).
The spin of the ground state for these compounds is S ) 0.
Therefore, at 4 K they should be EPR-silent. However, in our
case, as the J value is so small (J ∼ -1.7 cm-1), the lowest
excited states may also be populated besides the S ) 0 ground
state. As described by Adams et al.5 for weakly coupled
dinuclear complexes, the features observed probably correspond
to the superposition of the spectra of various excited spin states
(S ) 1-5) of the manganese(II) pair, modulated by a Boltzmann
distribution, since more states may contribute due to the small
J values.
where U ) coth K - 1/K and K ) JS(S + 1)/kT.
According to the second model (b),
(A + BX2)
Ng2â2
kT
øM )
×
(1 + CX + DX3)
where A ) 2.9167, B ) 208.04, C ) 15.543, D ) 2707.2, and
X ) |J|/2kT.
Experimental magnetic susceptibility data were analyzed
using both models. Dotted and solid lines in Figure 8 represents
the best fits to the Wagner-Friedberg and Weng-Hiller models
respectively. Similar J and g values were obtained in both fits,
and an average value was considered as a final parameter in
each case (Table 5). However, the Weng-Hiller model gave
better agreement. In fact, the main feature was the low values
of J, in accordance with the presence of only bridging
carboxylates.
For the trinuclear complexes (5-9) the intensity of the signal
at g ∼ 2 observed at room temperature decreased with the
temperature. At 4 K the spectrum became quite complicated:
it showed intense signals at low field and other less intense
signals, at high field (Figure 9B). These systems present an ST
5
) /2 ground state. At 4 K this spin state may be populated,
(74) Betz, P.; Bino, A.; Du, J.-L.; Lo, I. S.-M.; Thompson, R. C. Inorg.
Chim. Acta 1990, 170, 45.
(75) Weng, C. H. Ph.D. Dissertation, Carnegie-Mellon University, Pitts-
burgh, PA, 1968.
(76) Hiller, W.; Strahle, J.; Dtaz, A.; Hanack, M.; Hatfield, W. E., terHaar,
L. W.; Gutlich, P. J. Am. Chem. Soc. 1984, 106, 329.
which can exhibit zero-field splitting (ZFS). If the ZFS effects
were negligible, i.e., |J| > |A| (A is the hyperfine splitting
parameter), a narrow signal may arise around g ) 2.26,77
However, if the ZFS is considerable, this signal should be split

Mn(II) Complexes with Chlorobenzoato Bridges
Inorganic Chemistry, Vol. 37, No. 4, 1998 797
are narrow, and the line widths often become comparable to
those of the individual spins embedded in the magnetic lattice:
this regime is referred to as exchange narrowing.75 In this
theory it is necessary to take into account the relaxation time:
the absorption line is centered at νo and is broadened by ν1(t),
the time-dependent perturbation, which is dipolar in nature; in
other words, the resonance at νo is modulated by ν1(t). On the
other hand, it is possible to demonstrate that the exchange
narrowing is greater in 3D > 2D > 1D systems. In isolated
one-dimensional systems, the lines are broader than in three-
dimensional systems. The possible effect of weak interchain
coupling (J′) may then be strong: J′ yields more “three-
dimensional” behavior (exchange narrowing).
There are additional broadening mechanisms: hyperfine
coupling, anisotropic and antisymmetric exchange, g anisotropy,
and crystal field effects. Fortunately, for MnII low-dimensional
systems, only the hyperfine coupling and the crystal field effects
are considerable. Finally, for S > 1/2 (such as the MnII ions, S
5
) /2) another possible broadening mechanism is caused by
single-ion ZFS effects.
As pointed out above, the EPR spectra of 10 and 11 showed
different behavior with temperature: for 10 a considerable
variation of the signal width was observed when temperature
decreased, while for 11, only a slight broadening occurred, since
at room temperature the g ∼ 2 feature was already rather broad.
At this stage, it is impossible to explain the origin of this
difference. The fact that complex 10 exhibited antiferromag-
netic coupling greater than 11 might be related to the narrower
signal that the former presented at room temperature, since the
higher J (exchange interaction) might contribute to the exchange
narrowing (Hex . Hdip). When the temperature decreases and
approaches the critical point at which three-dimensional order
may be established, the dipolar interactions may become larger
(Hex , Hdip) and the line width may then increase quite rapidly,
giving a broad feature, as observed for both compounds.
Figure 9. Comparative X-band EPR spectra of a microcrystalline
sample at 4 K of complexes (A) [Mn2(µ-3-ClPhCOO)2(bpy)4](ClO4)2
(2), (B) [Mn3(µ-3-ClPhCOO)6(bpy)2] (6), and (C) [Mn(µ-3-ClPhCOO)2-
(bpy)]n‚n H2O (10).
into several features, and thus signals at low field may be
observed, as found here. No quantitative explanation of these
low-temperature features has been reported so far. Nevertheless,
these spectra at 4 K are similar to that reported by Me´nage et
al.26 for the acetate analogue at 10 K. On the other hand,
complex 8 showed a narrow band at g ∼ 2 at room temperature,
in contrast to the broad features presented by the other
complexes of the series. The narrowing of this signal might
be due to intermolecular interactions between the molecules,
as will be discussed for the manganese(II) chains (see below).
The EPR spectrum for the chains (10, 11) presented a single
band centered at g ∼ 2 at room temperature, which was narrow
for 10 (∼100 G) and quite broad for 11 (∼500 G). Complex
11 also showed a broad shoulder at 2200-2300 gauss. A broad
feature at g ∼ 2 (∼700 G) was observed at 4 K in both cases
(Figure 9C).
EPR spectra of one-dimensional systems are difficult to
interpret, even from a qualitative point of view. In fact, as
pointed out by Gatteschi and Bencini,78 the effect of the dipolar
interaction is great in these cases, leading to a broadening of
the main line of the spectrum. The addition of one more spin
to a cluster yields an increasing number of transitions, only some
of which are centred at νo, the rest flanking it at different
distances. The result is a homogeneous broadening of the line.
On the other hand, when isotropic exchange is also present the
exchange interaction yields a number of levels, which increases
with the number of interacting spins. As long as the various
multiplets are not split in zero field, the resonance may be
centered at νo, and no shift or broadening may occur. In other
words, the Hexchange Hamiltonian may not affect the spectrum,
while the Hdipolar Hamiltonian may yield broadening. If Hdip
Conclusions
The three chlorobenzoato anions (ortho, meta, and para) are
not “innocent” when forming new polynuclear MnII complexes.
Working with 2,2′-bipyridine (bpy) as blocking ligand and a
stoichiometric amount of sodium perchlorate for 2-Cl and 3-Cl
and a large excess for 4-Cl, dinuclear cationic complexes (with
ClO4- as counter-anion) may be prepared; neutral trinuclear
complexes, [Mn3(µ-RCOO)6(bpy)2], only for 2-Cl and 3-Cl and,
finally, unexpected one-dimensional systems, [Mn(µ-RCOO)2-
(bpy)]n‚nH2O, for 3-Cl and 4-Cl. Their structures have been
studied by X-ray and EXAFS determination. All show only
bridging carboxylate ligands and, thus, long distances between
the Mn centers.
The magnetic measurements showed, irrespective of their
nuclearity, antiferromagnetic Mn‚‚‚Mn interaction, owing to the
presence of five unpaired electrons in the d orbitals, which forms
a possible pathway for the interchange coupling. This anti-
ferromagnetism is very small due to the presence of only
carboxylate bridges.
The EPR spectra of the different polynuclear compounds are
difficult to interpret because, even when the spin ground state
is 0, they have low-lying excited states that are EPR-active,
which causes the formation of wide bands at low field.
. Hex, the broadening effect dominates; in contrast, if Hdip
,
Hex, then a completely new situation arises, in which the signals
Acknowledgment. This work was supported by The Spanish
Goverment (DGICYT), Grants PB93/0772 (to J.R.) and PB94-
1002 (to I.C.), and The Swiss National Science Foundation,
Grant 20.45315.95 (to H.S.). B.A. received an FPI Grant from
(77) Luneau, D.; Savariault, J. M.; Tuchagues, J. P. Inorg. Chem. 1988,
27, 3912.
(78) Bencini, A.; Gatteschi, D. EPR of Exchange Coupled Systems;
Springer-Verlag: Berlin, 1990.

798 Inorganic Chemistry, Vol. 37, No. 4, 1998
Albela et al.
spectra of 6 and 8 (Figure S2), Fourier transform of the experimental
EXAFS spectrum and comparison of the experimental and calculated
k-space EXAFS spectra for 8 using the FEFF method on the model
compound [Mn3(µ-PhCOO)6(bpy)2] (Figure S3), and a table containing
other parameters resulting from quantitative analyses of EXAFS data
(22 pages). Ordering information is given on any current masthead
page.
the Spanish Government. We also express our gratitude to Dr.
F. Villain for the help in the use of the EXAFS 3 spectrometer
and cryogen device and to Dr. N. Clos for recording the
magnetic and EPR data.
Supporting Information Available: Tables of fractional atomic
coordinates, bond distances and angles, anisotropic thermal parameters
for 4 and 10; figures showing the fitting of the EXAFS experimental
data for compounds 2, 11, 6, and 8 using the parameters extracted from
the FEFF calculation (Figure S1), the comparison between the XANES
IC970935P