Salicylaldoxime in manganese(III) carboxylate chemistry: Synthesis, structural characterization and physical studies of hexanuclear and polymeric complexes

Article abstract of DOI:10.1016/j.poly.2008.08.021

The use of salicylaldehyde oxime (H₂salox) in manganese(III) carboxylate chemistry has yielded new members of the family of hexanuclear compounds presenting the [Mn₆(μ₃-O)₂(μ₂-OR) ₂]¹²

Full text of DOI:10.1016/j.poly.2008.08.021

Polyhedron 27 (2008) 3575–3586
Contents lists available at ScienceDirect
Polyhedron
journal homepage: www.elsevier.com/locate/poly
Salicylaldoxime in manganese(III) carboxylate chemistry: Synthesis, structural
characterization and physical studies of hexanuclear and polymeric complexes
a,
a
a
Catherine P. Raptopoulou a, , Athanassios K. Boudalis , Katerina N. Lazarou , Vassilis Psycharis ,
Nikos Panopoulos a, Michael Fardis a, George Diamantopoulos a, Jean-Pierre Tuchagues b, Alain Mari b,
George Papavassiliou a
*
*
a Institute of Materials Science, NCSR ‘‘Demokritos”, 15310 Aghia Paraskevi, Athens, Greece
b Laboratoire de Chimie de Coordination du CNRS, UPR 8241, 205 Route de Narbonne, 31077 Toulouse, France
a r t i c l e i n f o
a b s t r a c t
Article history:
The use of salicylaldehyde oxime (H2salox) in manganese(III) carboxylate chemistry has yielded new
members of the family of hexanuclear compounds presenting the [Mn6(
plexes [MnI6II
3-O)2(O2CPh)2(salox)6(L1)2(L2)2] (L1 = py, L2 = H2O (1); L1 = Me2CO, L2 = H2O (2);
L1 = L2 = MeOH (3)). Addition of NaOMe to the acetonitrile reaction mixture, afforded the 1D complex
[MnI3IINa(
3-O)(O2CPh)2(salox)3(MeCN)]n (4), whereas addition of NaClO4 to the acetone reaction mixture
afforded an analogous 1D complex [MnI3IINa(
3-O)(O2CPh)2(salox)3(Me2CO)]n (5). The structures of 1–3
present the [Mn6( 3-O)2(
2-OR)2]12+ core and can be described as two [Mn3( 3-O)]7+ triangular subunits
linked by two
2-oximato oxygen atoms of the salox2ꢀ ligands, which show the less common l3
2O: O0: O0
N coordination mode. The benzoato ligands are coordinated through the usual syn,syn-l2 O:
mode. The 1D polymeric structures of 4 and 5 consist of alternating [Mn3(
3-O)]7+ subunits and Na+ atoms
linked through two l3 N and one l4
j2O: O0: j2O:j2O0: N salox2ꢀ ligands as well as one syn,anti-l2
O:
O0 benzoato ligand. DC and AC magnetic susceptibility studies on 1 revealed the stabilization of an
l3-O)2(l
2-OR)2]12+ core, com-
Received 15 July 2008
Accepted 31 August 2008
Available online 8 October 2008
(l
l
Keywords:
l
Manganese(III) clusters
Hexanuclear complexes
Polymeric complexes
Magnetic studies
Salicylaldoxime
l
l
l
l
-
j
j
j
-j j
l
-
j
j
-
j
-
j
j
S = 4 ground state, and indications of single-molecule magnetism behavior, whereas the DC experimental
data from polycrystalline sample of 5 are indicative of antiferromagnetic interactions within the [Mn3] sub-
unit. Solid state 1H NMR data of 1 were used to probe the spin-lattice relaxation of the system.
Ó 2008 Elsevier Ltd. All rights reserved.
1. Introduction
by the proper use of oligonuclear precursors [5] (‘building blocks’)
and/or specially designed ligands [6].
Polynuclear homo- and/or hetero-metallic transition metal
complexes (clusters) stand in the heart of modern inorganic chem-
istry research mainly due to their biomimetic [1] and physical
properties [2] which are closely related to their structural charac-
teristics. In most of the cases, the synthesis of the clusters is gov-
erned by self-assembly principles [3]. The labile nature of the
coordination bonds, among other parameters, such as the crystal-
Our research efforts are mainly oriented towards the study of
the physical properties of clusters prepared either accidentally in
the realm of self-assembly [7] or under more ‘sophisticated’
rational synthetic approaches [8]. Our research interests include
also, the investigation of the effect of various key factors, like li-
gand substitution [9], solvent [10], crystallization method [11], as
well as reactivity of metal clusters [12]. The work presented herein,
constitutes part of our research project on oximato complexes
[8a,10,12,13], and in particular on complexes containing salicylal-
dehyde oxime (H2salox) as ligand. Our previous experience on the
lization process (which is governed by weak H-bonding, p–p stack-
ing and Van der Waals interactions), the nature of the ligands and
the coordinative flexibility of metal ions and ligands constitute
some of the features that make this chemistry so difficult to con-
trol. However, during the last 15 years a great progress has been
made on the rational synthesis of metal clusters either by the
chemical modification of accidentally preformed clusters [4] or
FeIII=H2salox=PhCO reaction system revealed that, depending on
ꢀ
2
the reaction solvent, trinuclear [10] or hexanuclear [8a] complexes
were isolated. Initial results concerning the MnIII=H2salox=RCO
ꢀ
2
(R = Me, Ph) reaction system have been reported by Perlepes and
his coworkers [14]. We present herein, the investigation of various
reaction parameters, such as the reaction solvent (MeCN, Me2CO,
MeOH), the presence of externally added base and/or counterions,
the use of oligonuclear manganese clusters as starting materials,
the stoichiometry of the reactants and the crystallization method,
* Corresponding authors. Tel.: +30 210 6503346; fax: +30 210 6519430
(C.P. Raptopoulou).
E-mail addresses: craptop@ims.demokritos.gr (C.P. Raptopoulou), tbou@ims.
demokritos.gr (A.K. Boudalis).
0277-5387/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.poly.2008.08.021

3576
C.P. Raptopoulou et al. / Polyhedron 27 (2008) 3575–3586
to the identity of the MnIII=H2salox=RCO
(R = Ph) reaction
C, 49.27; H, 3.38; N, 6.96. Found: C, 49.02; H, 3.36; N, 6.93%. FT-
IR (KBr pellets, cmꢀ1): 3450(br), 3054(w), 3013(w), 2927(w),
1598(vs), 1529(vs), 1472(m), 1440(vs), 1394(vs), 1358(w),
1324(m), 1283(vs), 1200(m), 1154(m), 1126(m), 1068(w),
1043(s), 1027(vs), 1009(m), 961(w), 918(s), 851(w), 826(m),
754(s), 721(s), 678(vs), 652(m), 625(w), 530(w), 472(m).
ꢀ
2
system, i.e. the synthesis, structural and physical properties studies
of the hexanuclear complexes [Mn6III
(l
3-O)2(O2CPh)2(salox)6
(L1)2(L2)2] ꢁ xMeCN ꢁ yMeOH ꢁ zH2O ꢁ wMe2CO (L1 = py, L2 = H2O,
x = 6, y = 0, z = 0, w = 0 (1); L1 = Me2CO, L2 = H2O, x = 0, y = 0, z = 0,
w = 2 (2); L1 = L2 = MeOH, x = 0, y = 1, z = 0.5, w = 0 (3)) and the 1D
complexes [MnI3IINa(
l
3-O)(O2CPh)2(salox)3(S)]n ꢁ nS (S = MeCN (4);
S = Me2CO (5)).
2.2.2. [MnI6II
(2 ꢁ 2Me2CO)
2.2.2.1. Method A. Solid H2salox (0.034 g, 0.25 mmol) was added to
a refluxed red brown solution of [Mn3( 3-O)(O2CPh)6(py)2(H2O)]
(
l3-O)2(O2CPh)2(salox)6(Me2CO)2(H2O)2] ꢁ 2Me2CO
2. Experimental
l
(0.092 g, 0.085 mmol) in Me2CO (30 mL) and the resulting brown
solution was further refluxed for 15 min. Vapor diffusion of Et2O
to the brown reaction mixture afforded X-ray quality brown nee-
dle-like crystals of 2, which were filtered off, and dried in vacuo
(yield: 0.040 g, 60% based on the metal). The resulting powder
was analyzed as solvent free. C62H56Mn6N6O22 requires: C, 47.53;
H, 3.60; N, 5.36. Found: C, 47.29; H, 3.57; N, 5.33%. FT-IR (KBr pel-
lets, cmꢀ1): 3410(br), 3056(w), 3015(w), 1701(m), 1634(sh),
1597(vs), 1536(vs), 1472(s), 1440(vs), 1396(vs), 1327(m),
1282(vs), 1202(s), 1152(m), 1125(m), 1044(vs), 1028(vs), 956(w),
917(vs), 856(w), 825(m), 755(s), 722(s), 676(vs), 648(s), 550(w),
531(m), 465(s).
2.1. General and spectroscopic measurements
All manipulations were performed under aerobic conditions
using materials as received (Aldrich Co). All chemicals and solvents
were of reagent grade. [MnI2IIMnII(
l3-O)(O2CPh)6(py)2(H2O)] [15]
and (NBun4)[MnI4IIO2(O2CPh)9(H2O)] [16] were synthesized as previ-
ously described. Elemental analysis for carbon, hydrogen, and nitro-
gen was performed on a Perkin Elmer 2400/II automatic analyzer.
Infrared spectra were recorded as KBr pellets in the range 4000–
400 cmꢀ1 on a Bruker Equinox 55/S FT-IR spectrophotometer. Vari-
able-temperature magnetic susceptibility measurements were car-
ried out on polycrystalline samples of 1 and 5 using a Quantum
Design MPMS SQUID susceptometer. DC data were collected in the
2–300 K temperature range under magnetic fields of 0.1, 0.5, 1, 2.5
and 5 T for 1 and under a field of 0.5 T for 5. AC data for 1 were col-
lected between 2 and 7.5 K, with an oscillating field of 1 G amplitude
oscillating at frequencies of 0.1, 1, 10, 33, 99.9, 332, 997, 1400 Hz.
Diamagnetic corrections for the complexes were estimated from
Pascal’s constants. The magnetic susceptibility of 5 has been com-
puted by exact calculation of the energy levels associated with the
spin Hamiltonian through diagonalization of the full matrix with a
general program for axial symmetry [17]. Least-squares fittings were
accomplished with an adapted version of the function-minimization
2.2.2.2. Method B. Solid H2salox (0.034 g, 0.25 mmol) and solid
NaOMe (0.027 g, 0.50 mmol) were added to a refluxed red brown
solution of [Mn3(l3-O)(O2CPh)6(py)2(H2O)] (0.092 g, 0.085 mmol)
in Me2CO (30 mL) and the resulting brown solution was further re-
fluxed for 20 min. Small portions of a white precipitate (probably
insoluble NaOMe) were filtered off, and the brown filtrate was lay-
ered with Et2O/n-hex to afford brown crystals, which were filtered
off and dried in vacuo (yield: 0.043 g, 65% based on the metal). The
brown crystals were identified as compound 2 ꢁ 2Me2CO by unit
cell determination (a = 32.948(1), b = 32.948(1), c = 14.381(1) Å,
V = 15611.6(2) Å3, see Table 1).
2
P
ðxexpꢀxcalcÞ
program MINUIT [18]. The error factor R is defined as R ¼
,
Nx2exp
where N is the number of experimental points. 1H pulsed NMR
experiments on polycrystalline sample of 1 were performed at
2.4 T using a coherent spectrometer operating at 2.4 T (100 MHz
for 1H NMR). An Oxford 1200 CF continuous flow cryostat was em-
ployed for measurements in the range 5–300 K. The T1 spin-lattice
relaxation time was obtained using the standard saturation tech-
nique [19].
2.2.3. [MnI6II
(
l
3-O)2(O2CPh)2(salox)6(MeOH)4] ꢁ MeOH ꢁ 0.5H2O
(3 ꢁ MeOH ꢁ 0.5H2O)
Solid H2salox (0.034 g, 0.25 mmol) and solid MeONa (0.027 g,
0.50 mmol) were added to a red brown solution of [Mn3(l3
-
O)(O2CPh)6(py)2(H2O)] (0.092 g, 0.085 mmol) in MeOH (30 mL)
and the resulting green-brown solution was refluxed for half hour,
upon which time the solution turned to brown. Small portions of a
dark brown precipitate were filtered off, and the brown filtrate was
left for slow evaporation. After a period of one month, brown pris-
matic crystals of 3 were formed, which were filtered off, washed
with cold MeOH and dried in vacuo (yield: 0.039 g, 60% based on
the metal). The resulting powder was analyzed as solvent free.
C60H56Mn6N6O22 requires: C, 46.71; H, 3.66; N, 5.45. Found: C,
46.47; H, 3.63; N, 5.42%. FT-IR (KBr pellets, cmꢀ1): 3420(br),
3052(w), 3011(w), 2926(w), 1597(vs), 1527(vs), 1471(m),
1440(vs), 1393(vs), 1358(w), 1322(m), 1282(vs), 1200(m),
1154(m), 1126(m), 1067(w), 1042(s), 1027(vs), 1009(m), 960(w),
918(s), 851(w), 826(m), 755(s), 722(s), 679(vs), 652(m), 625(w),
531(w), 474(m).
2.2. Compound preparations
2.2.1. [MnI6II
2.2.1.1. Method A. Solid H2salox (0.014 g, 0.10 mmol) was added to
a red brown solution of [Mn3( 3-O)(O2CPh)6(py)2(H2O)] (0.108 g,
(
l3-O)2(O2CPh)2(salox)6(py)2(H2O)2] ꢁ 6MeCN (1 ꢁ 6MeCN)
l
0.10 mmol) in MeCN (30 mL) and the resulting brown solution
was refluxed for 1 h. Small portions of an orange-brown precipitate
were filtered off, and the brown filtrate was left for slow evapora-
tion. After 5 days, a mixture of green-brown prismatic crystals of 1
and orange cubic crystals were formed. The latter identified as the
mixed-valence
hexanuclear
complex
[MnII4MnIII
(l4-O)2-
2
(O2CPh)10(py)2(H2O)(MeCN)] ꢁ 2MeCN by unit cell determination
2.2.4. [MnI3IINa(
l
3-O)(O2CPh)2(salox)3(MeCN)]n ꢁ nMeCN (4 ꢁ nMeCN)
[20].
2.2.4.1. Method A. Solid H2salox (0.034 g, 0.25 mmol) and solid
2.2.1.2. Method B. Solid H2salox (0.069 g, 0.50 mmol) was added to
NaOMe (0.027 g, 0.50 mmol) were added to a refluxed red-brown
a red brown solution of [Mn3(l3-O)(O2CPh)6(py)2(H2O)] (0.184 g,
solution of [Mn3(l3-O)(O2CPh)6(py)2(H2O)] (0.092 g, 0.085 mmol)
0.17 mmol) in MeCN (50 mL) and the resulting brown solution
was refluxed for 1 h. The brown solution was left for slow evapora-
tion and after 20 days green-brown prismatic crystals of 1 were
formed, which were filtered off, washed with cold MeCN and dried
in vacuo (yield: 0.093 g, 67% based on the metal). The resulting
powder was analyzed as solvent free. C66H54Mn6N8O20 requires:
in MeCN (30 mL), and the resulting brown solution was further re-
fluxed for half hour. Small portions of a white precipitate (probably
insoluble NaOMe) were filtered off, and the brown filtrate was left
undisturbed in closed vials. After one week X-ray quality brown
needle-like crystals were formed, which were filtered off and dried
in vacuo (yield: 0.053 g, 70% based on the metal). The resulting

C.P. Raptopoulou et al. / Polyhedron 27 (2008) 3575–3586
3577
Table 1
Crystallographic data for complexes 1 ꢁ 6MeCN, 2 ꢁ 2Me2CO, 3 ꢁ MeOH ꢁ 0.5H2O, 4 ꢁ nMeCN and 5 ꢁ nMe2CO
1 ꢁ 6MeCN
2 ꢁ 2Me2CO
3 ꢁ MeOH ꢁ 0.5H2O
4 ꢁ nMeCN
5 ꢁ nMe2CO
Formula
Fw
Space group
T (°C)
k (Å)
C78H72Mn6N14O20
1855.14
P1
C68H68Mn6N6O24
C61H61Mn6N6O23.5
1583.80
P1
C39H31Mn3N5NaO11
C41H37Mn3N3NaO13
1682.92
I41/a
25
933.50
P21/c
25
967.56
P21/c
25
ꢀ
ꢀ
25
ꢀ170
Mo Ka
(0.71073 Å)
Cu Ka
(1.54187 Å)
Mo K
a
(0.71069 Å)
Cu Ka
(1.54187 Å)
Mo Ka (0.71069 Å)
a (Å)
b (Å)
c (Å)
12.588(7)
12.841(7)
13.438(7)
77.16(2)
77.89(2)
87.81(2)
2070.6(2)
1
32.8892(6)
32.8892(6)
14.3012(3)
12.797(1)
14.146(2)
19.830(2)
86.310(3)
69.983(3)
81.465(3)
3335.2(7)
2
14.5372(3)
17.6493(3)
16.1490(3)
15.2662(9)
17.8828(10)
16.1926(10)
a
(°)
b (°)
108.2725(7)
93.456(2)
c
(°)
V Å3
15469.6(5)
8
1.445
8.382 (Cu K
0.0475c
0.1236c
3934.5(1)
4
1.576
4412.6(5)
4
1.456
0.920 (Mo K
0.0972f
0.2270f
Z
qcalc (g cmꢀ3
)
1.488
1.577
l
R1
wR2
(mmꢀ1
)
0.965 (Mo K
0.0660b
0.1769b
a)
a)
1.184 (Mo K
a)
8.405 (Cu K
a)
a)
0.0773d
0.0524e
a
0.1938d
0.0969e
a
P
P
P
P
2(F2o)+(aP)2 + bP] and P = (max(F2o,0) + 2Fc2)/3; R1
=
(|Fo| ꢀ |Fc|)/ (|Fo|) and wR2 = { [w(F2o ꢀ Fc2)2]/ [w(Fo2)2]}1/2
.
a
b
c
d
e
f
w = 1/[
r
For 5740 reflections with I > 2
For 4251 reflections with I > 2
For 6754 reflections with I > 2
For 5301 reflections with I > 2
For 3739 reflections with I > 2
r(I).
r(I).
r(I).
r(I).
r(I).
powder was analyzed as solvent free. C37H28Mn3N4NaO11 requires:
C, 49.80; H, 3.16; N, 6.28. Found: C, 49.55; H, 3.14; N, 6.24%. FT-IR
(KBr pellets, cmꢀ1): 3428(br), 3056(w), 3012(w), 2927(w), 1654(s),
1596(vs), 1534(vs), 1473(s), 1440(vs), 1394(vs), 1367(s), 1322(m),
1277(vs), 1203(s), 1152(m), 1121(m), 1105(w), 1067(w), 1040(sh),
1022(vs), 957(w), 916(vs), 853(w), 825(m), 758(s), 720(s), 707(s),
672(vs), 653(sh), 551(w), 531(m), 478(m), 462(s).
on a Crystal Logic Dual Goniometer diffractometer using graph-
ite-monochromated Mo K radiation. Unit cell dimensions were
a
determined and refined by using the angular settings of 25 auto-
matically centred reflections in the range 11° < 2h < 23°. Three
standard reflections monitored every 97 reflections showed less
than 3% intensity fluctuation and no decay. Lorentz, polarization
corrections were applied using Crystal Logic software. Brown crys-
tals of 2 (0.08 ꢂ 0.12 ꢂ 0.60 mm), 4 (0.08 ꢂ 0.25 ꢂ 0.25 mm), and 5
(0.08 ꢂ 0.55 ꢂ 0.65 mm) were mounted in capillaries with drops of
mother liquor. A brown crystal of 3 (0.09 ꢂ 0.09 ꢂ 0.19 mm) was
taken directly from the mother liquor and immediately cooled to
ꢀ170 °C. Diffraction measurements for 2–5 were made on a Rigaku
R-AXIS SPIDER Image Plate diffractometer using graphite mono-
2.2.4.2. Method B. Solid H2salox (0.014 g, 0.10 mmol) and solid Na-
ClO4 ꢁ H2O (0.028 g, 0.20 mmol) were added to a red-brown solu-
tion of (NBun4)[Mn4O2(O2CPh)9(H2O)] (0.040 g, 0.025 mmol) in
MeCN (30 mL). The resulting brown solution was refluxed for
45 min. Layer of Et2O to the brown reaction mixture afforded
X-ray quality brown crystals which were filtered off and dried in
vacuo (yield: 0.014 g, 63% based on the metal). The brown crystals
were identified as compound 4 ꢁ nMeCN by unit cell determination
(a = 14.554(2), b = 17.670(3), c = 16.168(3) Å, b = 108.258(6)°,
V = 3948.6(2) Å3, see Table 1).
chromated Mo K
a (3, 5) and Cu Ka radiation (2, 4). Data collection
(x
-scans) and processing (cell refinement, data reduction and
empirical absorption correction) were performed using the CRYSTAL-
CLEAR program package [21]. Important crystal data and parameters
for data collection are reported in Table 1. The structures were
solved by direct methods using SHELXS-97 [22] and refined by full
matrix least-squares techniques on F2 with SHELXL-97 [23]. Further
experimental crystallographic details for 1: 2hmax = 50°; reflections
collected/unique/used, 7648/7276 [Rint = 0.0217]/7276; 611
2.2.5. [MnI3IINa(
(5 ꢁ nMe2CO)
l
3-O)(O2CPh)2(salox)3(Me2CO)]n ꢁ nMe2CO)
Solid H2salox (0.014 g, 0.10 mmol) and solid NaClO4 ꢁ H2O
(0.028 g, 0.20 mmol) were added to a red-brown solution of
(NBun4)[Mn4O2(O2CPh)9(H2O)] (0.040 g, 0.025 mmol) in Me2CO
(30 mL). The resulting brown solution was refluxed for 45 min.
Layering or vapor diffusion of Et2O or Et2O/n-hex to the brown
reaction mixture afforded X-ray quality brown crystals which were
filtered off and dried in vacuo (yield: 0.024 g, 74% based on the me-
tal). The resulting powder was analyzed as solvent free.
C41H37Mn3N3NaO13 requires: C, 50.90; H, 3.86; N, 4.34. Found: C,
50.64; H, 3.84; N, 4.31. FT-IR (KBr pellets, cmꢀ1): 3422(br),
3057(w), 3011(w), 2930(w), 1705(m), 1652(s), 1595(vs),
1537(vs), 1472(s), 1440(vs), 1395(vs), 1365(s), 1323(m),
1278(vs), 1204(s), 1154(m), 1123(m), 1106(w), 1069(w),
1040(sh), 1024(vs), 959(w), 917(vs), 854(w), 826(m), 759(s),
721(s), 706(s), 673(vs), 653(sh), 550(w), 531(m), 478(m), 462(s).
parameters refined;
(D/r)max = 0.004; (Dq)max/(Dq)min = 1.192/
ꢀ01.409 e/Å3; R/Rw (for all data), 0.0830/0.1955. Hydrogen atoms
were located by difference maps and were refined isotropically, ex-
cept those of the coordinated pyridine and the solvate molecules
which were introduced at calculated positions as riding on bonded
atoms. All non-H atoms were refined anisotropically. The crystals
of 2 were very thin needles. Repeating efforts to increase their size
proved unsuccessful. Data were collected to 2hmax = 144° but only
those to 2hmax = 125° were used for the refinement, since in greater
h values more than 50% of the collected data were unobserved. Fur-
ther experimental crystallographic details for 2: 2hmax = 125°;
reflections collected/unique/used, 31770/5864 [Rint = 0.0491]/
5864; 557 parameters refined;
(
D
/
r
)
max = 0.008;
(
D
q
)
/
max
(D
q)
min = 0.564/ꢀ0.341 e/Å3; R/Rw (for all data), 0.0758/0.1423.
Hydrogen atoms were located by difference maps and were refined
isotropically, except those of the methyl groups which were intro-
duced at calculated positions as riding on bonded atoms. All non-H
atoms were refined anisotropically. The crystals of 3 were very
small in size. Repeating efforts to increase their size proved unsuc-
2.3. Single-crystal X-ray crystallography
A green-brown prismatic crystal of 1 (0.18 ꢂ 0.36 ꢂ 0.70 mm)
was mounted in capillary. Diffraction measurements were made

3578
C.P. Raptopoulou et al. / Polyhedron 27 (2008) 3575–3586
cessful. Data were collected to 2hmax = 55° but only those to
2hmax = 46.5° were used for the refinement, since in greater h val-
ues more than 50% of the collected data were unobserved. Further
experimental crystallographic details for 3: reflections collected/
unique/used, 19418/9471 [Rint = 0.0684]/9471; 884 parameters re-
were introduced at calculated positions as riding on bonded atoms.
All non-H atoms were refined anisotropically.
3. Results and discussion
fined; (
D
/r)
max = 0.007; (
D
q
)max/(D
q)
min = 0.847/ꢀ0.673 e/Å3; R/
3.1. Syntheses
Rw (for all data), 0.1034/0.2107. Hydrogen atoms were introduced
at calculated positions as riding on bonded atoms; no H-atoms for
the water solvate molecule were included in the refinement. All
non-H atoms were refined anisotropically. The solvate water mol-
ecule was refined anisotropically with occupation factor fixed to
0.50. Further experimental crystallographic details for 4:
2hmax = 144.1°; reflections collected/unique/used, 27919/7418
Previous investigation of the use of H2salox in manganese car-
boxylate chemistry by Perlepes and his co-workers has led to the
hexanuclear
compounds
[MnI6II
(l3-O)2(O2CR)2(salox)6(EtOH)4]
(R = Me, Ph) [14] isolated from the 1:1 Mn(O2CR)2 ꢁ 2H2O/H2salox
molar ratio reaction in EtOH. Both complexes exhibit single-mole-
cule magnetism (SMM) behaviour [24], and have been isolated under
identical reaction conditions suggesting that the nature of the
carboxylate ligand does not affect the identity of the products. Re-
cently Brechin and co-workers have expanded this family of clusters
by changing the carboxylate groups and/or by introducing steric
bulk at the oximate carbon atom [25]; these modifications led to
dramatic improvement of the SMM properties [25b,c]. In order to
further investigate the parameter space of this reaction system, we
have decided to modify other key factors and in particular: (i) the
Mn/H2salox stoichiometry, (ii) the reaction media, (iii) the source
of the manganese ions, (iv) the addition of base and/or counterions,
and (v) the crystallization method, in order to study their influence
on the metal ions’ topology, oxidation state and nuclearity. Thus,
[Rint = 0.0662]/7418; 634 parameters refined; (D/r)max = 0.005;
(D
q)
max/(D
q)
min = 0.418/ꢀ0.705 e/Å3; R/Rw (for all data), 0.0814/
0.1110. Hydrogen atoms were located by difference maps and were
refined isotropically, except those of the methyl groups which
were introduced at calculated positions as riding on bonded atoms.
All non-H atoms were refined anisotropically. The crystals of 5
showed poor diffraction ability (despite their sufficient size) and
signs of twinning. Data were collected to 2hmax = 54° but only those
to 2hmax = 45° were used for the refinement, since in greater h val-
ues more than 50% of the collected data were unobserved. The
twinning problem was handled by omitting reflections showing
large F2 compared to the expected F2 values because an exact
o
c
reaction of one equivalent of [MnI2IIMnII(
l3-O)(O2CPh)6(py)2(H2O)]
with one equivalent of H2salox in MeCN and slow evaporation of
twin law could not be found. As it was proved after the solution
of the structure, compound 5 is analogous to 4 thus we felt that
the complete structural characterization would not add further sig-
nificant information to our work. Therefore, we have not at-
tempted to improve the quality of the crystals. Further
experimental crystallographic details for 5: 2hmax = 45°; reflections
collected/unique/used, 40444/5493 [Rint = 0.1240]/5493; 554
the reaction mixture afforded green-brown prismatic crystals of
[MnI6II(
l
3-O)2(O2CPh)2(salox)6(py)2(H2O)2] (1) and orange cubic crys-
tals of the known mixed-valence hexanuclear complex [MnI4IM-
nI2II
4-O)2(O2CPh)10(py)2(H2O)(MeCN)] [20] in mixture. With the
(
l
identity of 1 established by X-ray crystallography the stoichiometric
Mn:H2salox reaction has been performed which afforded pure 1 for
further physical measurements (Scheme 1). Further investigation of
parameters refined;
(D/r)max = 0.017; (Dq)max/(Dq)min = 2.162/
ꢀ0.608 e/Å3; R/Rw (for all data), 0.1204/0.2448. Hydrogen atoms
Scheme 1. The synthetic routes employed for the preparation of 1–5 (see text for details).

C.P. Raptopoulou et al. / Polyhedron 27 (2008) 3575–3586
3579
the reaction media, by reaction of one equivalent of [MnI2IIMnII(l3
-
O)(O2CPh)6(py)2(H2O)] with three equivalents of H2salox in Me2CO
afforded compound [MnI6IIO2(O2CPh)2(salox)6(Me2CO)2(H2O)2] (2)
either by slow evaporation or layering of Et2O or vapour diffusion
of Et2O or Et2O/n-hexane (Scheme 1).
Compounds 1 and 2 are quite similar to the previously reported
hexanuclear
complexes
[Mn6(l3-O)2(O2CR)2(salox)6(EtOH)4]
(R = Me, Ph), suggesting that these species are spontaneously
formed, perhaps due to their thermal and/or thermodynamic sta-
bility, and/or their insolubility in the reaction solvent. Thus,
changes in the source of the manganese ions and the reaction med-
ia have not affected the identity of the product isolated, whereas
the change of the reactants stoichiometry (excess of manganese
ions in the case of 1) most likely promotes the simultaneous isola-
tion of the known mixed-valence compound [MnI4IMn2III
(l4-
O)2(O2CPh)10(py)2(H2O)(MeCN)]. Of course the presence of other
more soluble species in the reaction mixture cannot be excluded.
In order to examine the influence of the base and/or counterions
we have employed reactions analogous to those leading to 1 and 2
in the presence of NaOMe and/or NaClO4.H2O. Thus, reaction of one
equivalent of [MnI2IIMnII(
l3-O)(O2CPh)6(py)2(H2O)] with three
Fig. 1. Partially labelled ORTEP plot of
probability level. For clarity hydrogen atoms have been omitted. Primed atoms
are generated by symmetry (0 = ꢀx, 2 ꢀ y, 1 ꢀ z).
1
with ellipsoids drawn at the 30%
equivalents of H2salox in the presence of six equivalents of NaOMe
in MeOH yielded compound [MnI6IIO2(O2CPh)2(salox)6(MeOH)4] (3),
which is completely analogous to 1 and 2. On the other hand, the
analogous reaction in MeCN yielded only the 1D complex [Mn3III
Na(l3-O)(O2CPh)2(salox)3(MeCN)]n (4) whose structure is com-
pletely different than 1–3, whereas the same reaction mixture in
geometry around Mn(1) is distorted octahedral where the equato-
rial positions are occupied by one salox2ꢀ ligand (through Ophenolic
and Noximato atoms), the Ooximato atom of a second salox2ꢀ ligand
and the oxide group, while the Jahn–Teller axis of the octahedron
is completed by a pyridine and a coordinated water molecule.
The coordination geometry around Mn(2) is square pyramidal
comprised by one salox2ꢀ ligand (through Ophenolic and Noximato
atoms), the Ooximato atom of a second salox2ꢀ ligand and the oxide
group occupying the equatorial plane, while the apical position is
occupied by an oxygen atom of the l2 benzoato ligand. Finally,
the coordination geometry about Mn(3) is distorted octahedral
comprised by one salox2ꢀ ligand (through Ophenolic and Noximato
atoms), the oxide group and the Ooximato atom of a second salox2ꢀ
directed in the equatorial plane of the octahedron, while the Jahn–
Teller positions are completed by a benzoato oxygen and an
Me2CO afforded only 2 as confirmed by X-ray crystallography
(Scheme 1). Finally the reaction of one equivalent of (NBun
)
4
[MnI4IIO2(O2CPh)9(H2O)] with four equivalents of H2salox in the pres-
ence of eight equivalents of NaClO4 ꢁ H2O in MeCN also afforded
compound 4, whereas the analogous reaction in Me2CO afforded
the 1D complex [MnI3IINa(
l3-O)(O2CPh)2(salox)3(Me2CO)]n (5) which
is analogous to 4 (Scheme 1).
In conclusion, the reaction media, the use of oligonuclear man-
ganese complexes as starting materials and the crystallization
methods have not affected the identity of the hexanuclear com-
plexes 1–3, whereas analogous modifications on the salicylaldoxi-
mato iron(III) carboxylate chemistry revealed that the reaction
media is determinative for the identity of the product. Trinuclear
Ooximato atom of a third salox2ꢀ ligand. Thus, in each [Mn3(
subunit two of the salox2ꢀ ligands adopt the common l2
O0:
N coordination mode, while the third one presents the rare
2O: O0:
N coordination mode (Scheme 2), that has been also
observed in [Mn6(
3-O)2(O2CR)2(salox)6(R0OH)4] [14,25a], [Fe6-
l3-O)]
[Fe3(
L2 = H2O depending on the reaction solvent) [10] or hexanuclear
[Fe6( 3-O)2(O2CPh)10(salox)2(L)2] (L = H2O, MeCONH2) [8a] com-
l3-O)(O2CPh)5(salox)(L1)(L2)] (L1 = L2 = MeOH, L1 = EtOH and
-j
O:
j
j
l
l3-
j
j
j
plexes have been isolated from alcoholic or MeCN solutions,
respectively; the former being transformed to the latter by simple
recrystallization from MeCN solutions. Compounds 1–3 and
l
(
(
l
l
3-O)2(O2CPh)10(salox)2(H2O)2] [8a], and in [M6III
(
l
3-O)2(salox)6-
2-O2CR)2(OH2)2 (RCN)2] (M = VIII
,
CrIII, MnIII
,
FeIII
,
RCO2
¼
ꢀ
3-O)2(O2CPh)10(salox)2(L)2] contain the same [M6III
(
l3-
Ph3CCO2ꢀ; Me3CCO2
;
Ph2CðOHÞCO2ꢀ; PhCO2ꢀ; C2H5CO2
)
[26],
ꢀ
ꢀ
[Fe6(
O)2(
l
l
2-OR)2]12+ core, although the rest of the ligation pattern is dif-
but also in trinuclear and tetranuclear complexes [27].
ferent. On the other hand, the presence of sodium ions in the reac-
tion mixtures in MeCN and Me2CO, either as NaOMe or
NaClO4 ꢁ H2O, revealed dramatic changes to the identity of the prod-
ucts isolated yielding the 1D complexes 4 and 5.
The three MnIII ions in each [Mn3( 3-O)]7+ subunit form an al-
l
most isosceles triangle (see Table 2). The Mn(2)ꢁ ꢁ ꢁMn(3) separation
(3.158(2) Å) is shorter than the other two in the triangular subunit
(3.248(5), 3.273(5) Å) due to the presence of the extra benzoate
bridge. The Mnꢁ ꢁ ꢁMn distance in the central [Mn2(
3.335(5). The central 3-O)] subunit is dis-
3-O2ꢀ ion in each [Mn3(
placed at 0.17 Å out of the [Mn3] mean plane, an observation that
has been also seen in the analogous compound [Mn6(l3
O)2(O2CR)2(salox)6(R0OH)4] [14,25a] as well as in the trinuclear
complexes [Mn3( 3-O)(O2CR)3(mpko)3](ClO4) (R = Me, Et, Ph;
mpkoH = methyl-2-pyridyl ketone oxime, 0.30 Å) [28]. The Mn–
Ooxo bond distances are in the range 1.863–1.881 Å. The Mn–
Ophenoxo and the Mn–Noximato bond lengths are ꢃ1.86 and ꢃ2.0 Å,
respectively. The Mn–Ooximato chelating bond distances are
ꢃ1.94 Å while the corresponding bridging bond distances are sig-
nificantly longer (2.420 Å). All the salox2ꢀ ligands are planar and
make an angle of 5.1°, 13.3° and 28.7° (for the salox2ꢀ ligands de-
fined by O(1), O(11), and O(21), respectively) with the [Mn3] mean
l2-O)2] core is
3.2. Description of structures
l
l
ꢀ
Compound 1 crystallizes in the triclinic space group P1 with one
molecule per unit cell. The molecular structure of 1 is given in
Fig. 1 and selected bond distances and angles are listed in
Table 2. The structure contains the [Mn6(
whose topology consists of six manganese(III) ions arranged as
two centrosymmetrically related off-set stacked [Mn3(
3-O)]7+
-
l
l
3-O)2(
l
2-OR)2]12+ core,
l
subunits bridged by two oximato oxygen atoms (O(12) and
O(120)). In each [Mn3(
l
3-O)]7+ subunit, there are two Mnꢁ ꢁ ꢁMn
pairs [Mn(1)/Mn(3) and Mn(1)/Mn2)] bridged by a l2 oxide and
a
l2
-
j
O:
j
O0:
j
N salox2ꢀ ligand, while the third pair [Mn(2)/
2 benzoato ligand. The coordination
Mn(3)] is further bridged by a
l

3580
C.P. Raptopoulou et al. / Polyhedron 27 (2008) 3575–3586
Table 2
Selected bond distances (Å) and angles (°) for complex 1 ꢁ 6MeCN
Distances
Mn(1)–O(1)
Mn(1)–O(41)
Mn(1)–O(22)
Mn(1)–N(1)
Mn(1)–O(w)
Mn(1)–N(41)
Mn(1)ꢁ ꢁ ꢁMn(3)
1.859(3)
1.882(3)
1.922(3)
1.998(4)
2.310(4)
2.391(4)
3.248(5)
Mn(2)–O(11)
Mn(2)–O(41)
Mn(2)–O(2)
Mn(2)–N(11)
Mn(2)–O(31)
Mn(1)ꢁ ꢁ ꢁMn(2)
Mn(2)ꢁ ꢁ ꢁMn(3)
1.864(3)
1.862(3)
1.900(3)
2.011(4)
2.123(4)
3.273(5)
3.158(2)
Mn(3)–O(21)
Mn(3)–O(41)
Mn(3)–O(12)
Mn(3)–N(21)
Mn(3)–O(32)
Mn(3)–O(120)
Mn(3)ꢁ ꢁ ꢁMn(30)
1.866(3)
1.868(3)
1.942(3)
1.998(4)
2.137(4)
2.419(3)
3.335(5)
Angles
O(1)–Mn(1)–O(41)
O(1)–Mn(1)–O(22)
O(41)–Mn(1)–O(22)
O(1)–Mn(1)–N(1)
O(41)–Mn(1)–N(1)
O(22)–Mn(1)–N(1)
O(1)–Mn(1)–Ow
O(41)–Mn(1)–Ow
O(22)–Mn(1)–Ow
N(1)–Mn(1)–Ow
O(1)–Mn(1)–N(41)
O(41)–Mn(1)–N(41)
O(22)–Mn(1)–N(41)
N(1)–Mn(1)–N(41)
Ow–Mn(1)–N(41)
174.5(2)
88.8(2)
92.3(1)
91.9(2)
87.6(1)
174.8(2)
88.2(2)
86.3(1)
97.8(2)
87.5(2)
89.3(2)
96.1(1)
90.9(2)
83.9(2)
170.9(2)
O(11)–Mn(2)–O(41)
O(11)–Mn(2)–O(2)
O(41)–Mn(2)–O(2)
O(11)–Mn(2)–N(11)
O(41)–Mn(2)–N(11)
O(2)–Mn(2)–N(11)
O(11)–Mn(2)–O(31)
O(41)–Mn(2)–O(31)
O(2)–Mn(2)–O(31)
N(11)–Mn(2)–O(31)
Mn(1)–O(41)–Mn(2)
Mn(1)–O(41)–Mn(3)
Mn(2)–O(41)–Mn(3)
Mn(3)–O(12)–Mn30)
169.4(2)
87.2(1)
91.0(2)
89.0(2)
89.4(1)
161.4(2)
98.1(2)
92.5(1)
104.2(2)
94.3(2)
121.9(2)
120.1(2)
115.7(2)
99.2(1)
O(21)–Mn(3)–O(41)
O(21)–Mn(3)–O(12)
O(41)–Mn(3)–O(12)
O(21)–Mn(3)–N(21)
O(41)–Mn(3)–N(21)
O(12)–Mn(3)–N(21)
O(21)–Mn(3)–O(32)
O(41)–Mn(3)–O(32)
O(12)–Mn(3)–O(32)
N(21)–Mn(3)–O(32)
O(21)–Mn(3)–O(120)
O(41)–Mn(3)–O(120)
O(12)–Mn(3)–O(120)
O(32)–Mn(3)–O(120)
N(21)–Mn(3)–O(120)
175.6(2)
88.4(1)
89.8(1)
90.9(2)
89.9(1)
165.9(2)
94.1(2)
90.1(1)
96.6(2)
97.5(2)
88.9(1)
86.8(1)
80.8(1)
176.0(1)
85.1(1)
0
Symmetry transformations used to generate equivalent atoms: = ꢀx, 2 ꢀ y, 1 ꢀ z.
metric unit. The molecular structures of 2 and 3 (Figs. 2 and 3,
respectively) are completely analogous to 1 and they will not be
discussed further. Selected bond distances and angles are given
as Supplementary material (Tables S1 and S2, respectively); a com-
parison of their structural characteristics with those of 1 is given in
Table 4.
Compound 4 crystallizes in the monoclinic space group P21/c
with one molecule per asymmetric unit. The molecular structure
of 4 consists of anionic trinuclear oxo-centered units, [Mn3III
(l3-
O)(O2CPh)2(salox)3]ꢀ, sodium cations and MeCN molecules. An OR-
TEP plot of the structure of 4 is given in Fig. 4, selected bond dis-
tances and angles are listed in Table 3. The trinuclear anion
contains a central [Mn3(
l
3-O)]7+ core, in which two Mnꢁ ꢁ ꢁMn pairs
[Mn(1)/Mn(2) and Mn(2)/Mn(3)] are bridged by a l2 oxide and a
l2-j j j
O0: N salox2ꢀ ligand, while the third pair [Mn(1)/Mn(3)] is
O:
further bridged by a l2 benzoato ligand. The coordination geometry
Scheme 2. The coordination modes of salox2ꢀ ligands in the structures of 1–5 (see
text for details).
plane. The benzoate ligands are coordinated through the common
syn,syn l2-jO:j
O0 mode. The Mn–Ocarboxylate bond distances are
ꢃ2.13 Å. There is a moderate twisting of the Mn–O–N–Mn moieties
within each Mn3 subunit; this is evidenced by the Mn–N–O–Mn
torsion angles which are 5.9°, 22.3° and 19.8° for the Mn(1)/
Mn(2), Mn(2)/Mn(3) and Mn(3)/Mn(1) pairs (and their symme-
try-equivalents), respectively.
Compound 2 crystallizes in the tetragonal space group I41/a,
with eight molecules per unit cell, and compound 3 crystallizes
in the triclinic space group P1 with two half molecules in the asym-
Fig. 2. Partially labelled ORTEP plot of
probability level. For clarity hydrogen atoms have been omitted. Primed atoms
are generated by symmetry (0 = 1 ꢀ x, 1 ꢀ y, ꢀz).
2 with ellipsoids drawn at the 30%
ꢀ

C.P. Raptopoulou et al. / Polyhedron 27 (2008) 3575–3586
3581
structural characteristics of the planar [Mn3(
l
3-O)]7+ core. The Mn–
Ooxo bond distances are in the range 1.845–1.867 Å. The average
Mn–Ophenoxo, Mn–Ooximato and Mn–Noximato bond lengths are 1.863,
1.914 and 2.008 Å, respectively.
All the salox2ꢀ ligands are planar and make an angle of 18.6°,
16.8° and 25.8° (for the salox2ꢀ ligands defined by O(1), O(11),
and O(21), respectively) with the [Mn3] mean plane. The three sal-
ox2ꢀ ligands adopt different coordination modes, bridging pairs of
MnIII ions of the anionic complex with the sodium cations. Two of
the salox2ꢀ ligands (defined by O(1) and O(21)) adopt the l3
-
j2O: O0:
j jN coordination mode bridging Mn(3)/Mn(1) and Mn(3)/
Mn(2) pairs to Na through the l2 phenoxo oxygen O(1) and
through the l2 oximato oxygen O(22), respectively. The third sal-
ox2ꢀ ligand (defined by O(11)) adopts the l4
-
j
2O:j2O0:
jN coordi-
nation mode bridging Mn(1)/Mn(2) pair to two adjacent Na atoms
through the l2 phenoxo oxygen O(11) and through the l2 oximato
oxygen O(12). The coordinational flexibility of the salox2ꢀ ligands
is reflected in the Mn–N–O–Mn torsion angles in the Mn3 subunit,
suggesting a moderate twisting of the Mn–O–N–Mn moieties in
the case of l3- j jN ligands (22.1° and 16.7° for the Mn(1)/
j2O: O0:
Mn(3) and Mn(3)/Mn(2) pairs, respectively), whereas the same
moiety is planar (0.8° for the Mn(2)/Mn(1) pair) in the case of
the l4-j jN ligand.
2O:j2O0:
The two benzoato ligands also adopt different coordination
modes. One of them bridges Mn(1) and Mn(3) in the trinuclear an-
ion through the common syn,syn l2
whereas the second one bridges the trinuclear anion to the sodium
cation through the syn,anti l2 O:
O0 coordination mode. The Mn–
Fig. 3. Partially labelled ORTEP plot of one of the crystallographically independent
hexanuclear molecules in the structure of 3 with ellipsoids drawn at the 30%
probability level. For clarity hydrogen atoms have been omitted. Primed atoms are
generated by symmetry (0 = 1 ꢀ x, ꢀy, ꢀz).
-jO:j
O0 coordination mode,
-j j
Ocarboxylate bond distances range from 2.045 to 2.170 Å. The coordi-
nation geometry around the Na+ ions is distorted octahedral com-
prised of two phenoxo and two oximato oxygen atoms belonging
to four salox2ꢀ ligands, one carboxylato oxygen atom and one
MeCN molecule. Each Na+ ion is triply and doubly bridged to its
adjacent trinuclear anions forming 1D polymeric arrays which ex-
tend parallel to the crystallographic c-axis. The Na–O bond inter-
atomic distances range from 2.406 to 2.673 Å.
Compound 5 crystallizes in the monoclinic space group P21/c
with one molecule per asymmetric unit, and is completely analo-
gous to 4; the only difference being the solvate molecule coordi-
nated to the Na+ ions. An ORTEP plot of a small fragment of the
chains of 5 is given in Fig. 5, selected bond distances and angles
are given as Supplementary material (Table S3), and a comparison
of its structural characteristics with those of 4 is given in Table 4.
Compounds 1–5 join a small family of structurally characterized
Mn complexes with the mono-(Hsalox-) and/or the dianion (sal-
ox2ꢀ) of salicylaldoxime as ligand. Other members are the ‘‘butter-
fly” complex [Mn4IIIO2(salox)2L3L02](ClO4) [30], where
L is the
diphenylglycolate(ꢀ1) ligand and L0 is 1,4,7-trimethyl-1,4,7-triaza-
cyclononane, and the structurally similar [MnI6IIO2(O2CR)2(sa-
lox)6(R0OH)4] (R = H, Me, Ph; R0 = Me, Et) mentioned above
[14,25a]. Compounds 1–3 are also members of a small family of Mn6III
Fig. 4. Partially labelled ORTEP plot of a small fragment of the chains of 4 with
ellipsoids drawn at the 30% probability level. For clarity hydrogen atoms have been
0
omitted. Symmetry transformations used to generate equivalent atoms: = x,
00
0.5 ꢀ y, ꢀ0.5 + z; = x, 0.5 ꢀ y, 0.5 + z.
complexes whose cores are based on two linked trinuclear [Mn3III
(l3-
O)]7+ subunits [14,25,31]. Compounds 4 and 5 represent the first tri-
nuclear oxo-centered MnIII anionic complexes with salox2ꢀ. Re-
cently, a neutral trinuclear oxo-centered MnIII complex has been
around each MnIII ion is square pyramidal where the basal positions
are occupied by one salox2ꢀ ligand (through Ophenolic and Noximato
atoms), the Ooximato atom of a second salox2ꢀ ligand and the oxide
group, and the apical position is occupied by a carboxylato oxygen
atom [O(31), O(41) and O(32) for Mn(1), Mn(2) and Mn(3),
respectively].
reported which contains Me-salox2ꢀ
Me)(Me-salox)3(py)4] [32].
, compound [Mn3O(O2C-
3.3. Magnetic susceptibility studies
The three MnIII ions in the [Mn3(
triangle with edges 3.138(1)–3.305(1) Å (Table 3). The central l3
O2ꢀ ion lies in the plane defined by the three MnIII ions, as in the
l
3-O)]7+ subunit form a scalene
-
3.3.1. DC susceptibility
The
vMT product of
1
(Fig. 6) at 300 K (1 T) is
case of the ‘basic carboxylates’ of the general formula [M3III
(
l3-
15.62 cm3 molꢀ1 K, lower than the theoretically expected value
for six non-interacting S = 2 spins (18.01 cm3 molꢀ1 K, g = 2), indi-
cating the interplay of antiferromagnetic interactions. Upon cool-
O)(O2CR)6(L)3]+ (M = Mn, Fe, Cr; L = neutral monodentate ligands)
[29], suggesting that the replacement of the carboxylato and mono-
dentate ligands by the tridentate salox2ꢀ ligand has not affected the
ing, this progressively decreases to
a
minimum of

3582
C.P. Raptopoulou et al. / Polyhedron 27 (2008) 3575–3586
Table 3
Selected bond distances (Å) and angles (°) for complex 4 ꢁ nMeCN
Distances
Mn(1)–O(1)
Mn(1)–O(51)
Mn(1)–O(12)
Mn(1)–N(1)
Mn(1)–O(31)
Na–O(220)
1.865(3)
1.867(2)
1.915(3)
1.992(3)
2.170(3)
2.406(3)
2.414(3)
3.206(1)
3.187(2)
Mn(2)–O(51)
Mn(2)–O(11)
Mn(2)–O(22)
Mn(2)–N(11)
Mn(2)–O(41)
Na–N(31)
1.845(2)
1.869(3)
1.920(2)
2.028(3)
2.045(3)
2.419(5)
2.451(3)
3.138(1)
3.443(2)
Mn(3)–O(21)
Mn(3)–O(51)
Mn(3)–O(2)
Mn(3)–N(21)
Mn(3)–O(32)
Na–O(42)
1.856(3)
1.865(2)
1.906(3)
2.003(3)
2.085(3)
2.490(3)
2.673(3)
3.305(1)
Na–O(1)
Na–O(110)
Na–O(12)
Mn(2)ꢁ ꢁ ꢁMn(3)
Mn(1)ꢁ ꢁ ꢁMn(2)
Mn(1)ꢁ ꢁ ꢁMn(3)
Mn(1)ꢁ ꢁ ꢁNa
Mn(2)ꢁ ꢁ ꢁNa00
Angles
O(1)–Mn(1)–O(51)
O(1)–Mn(1)–O(12)
O(51)–Mn(1)–O(12)
O(1)–Mn(1)–N(1)
O(51)–Mn(1)–N(1)
O(12)–Mn(1)–N(1)
O(1)–Mn(1)–O(31)
O(51)–Mn(1)–O(31)
O(12)–Mn(1)–O(31)
N(1)–Mn(1)–O(31)
O(220)–Na–O(1)
169.4(1)
86.9(1)
92.7(1)
90.8(1)
89.0(1)
176.2(1)
98.6(1)
92.1(1)
93.4(1)
89.9(1)
86.3(1)
108.0(1)
149.0(1)
64.1(1)
102.1(1)
125.9(1)
86.3(1)
104.9(1)
O(51)–Mn(2)–O(11)
O(51)–Mn(2)–O(22)
O(11)–Mn(2)–O(22)
O(51)–Mn(2)–N(11)
O(11)–Mn(2)–N(11)
O(22)–Mn(2)–N(11)
O(51)–Mn(2)–O(41)
O(11)–Mn(2)–O(41)
O(22)–Mn(2)–O(41)
N(11)–Mn(2)–O(41)
N(31)–Na–O(110)
O(220)–Na–O(42)
O(1)–Na–O(42)
167.5(1)
89.2(1)
85.7(1)
89.2(1)
90.7(1)
155.2(1)
95.9(1)
96.7(1)
111.0()
93.9(1)
108.9(1)
165.2(1)
79.1(1)
83.8(1)
121.5(1)
119.5(1)
95.4(1)
O(21)–Mn(3)–O(51)
O(21)–Mn(3)–O(2)
O(51)–Mn(3)–O(2)
O(21)–Mn(3)–N(21)
O(51)–Mn(3)–N(21)
O(2)–Mn(3)–N(21)
O(21)–Mn(3)–O(32)
O(51)–Mn(3)–O(32)
O(2)–Mn(3)–O(32)
N(21)–Mn(3)–O(32)
O(220)–Na–O(12)
O(1)–Na–O(12)
N(31)–Na–O(12)
O(110)–Na–O(12)
O(42)–Na–O(12)
Mn(3)–O(51)–Mn(1)
Mn(2)–O(11)–Na00
174.8(1)
90.7(1)
90.9(1)
90.7(1)
85.8(1)
155.1(1)
91.3(1)
93.4(1)
98.1(1)
106.8(1)
99.2(1)
61.2(1)
88.8(1)
158.3(1)
71.4(1)
114.5(1)
104.9(1)
O(220)–Na–N(31)
O(1)–Na–N(31)
O(220)–Na–O(110)
O(1)–Na–O(110)
Mn(2)–O(51)–Mn(3)
Mn(1)–O(12)–Na
Mn(2)–O(22)–Na00
N(31)–Na–O(42)
O(110)–Na–O(42)
Mn(2)–O(51)–Mn(1)
Mn(1)–O(1)–Na
0
00
Symmetry transformations used to generate equivalent atoms: = x, 0.5 ꢀ y, ꢀ0.5 + z; = x, 0.5 ꢀ y, 0.5 + z.
7.63 cm3 molꢀ1 K at 30 K, and then increases to a maximum of
9.22 cm3 molꢀ1 K at 10 K, before falling to 4.71 cm3 molꢀ1 K at
2 K. This overall behaviour shows the interplay of both ferro- and
antiferromagnetic interactions. The drop below 10 K is associated
with zero-field splitting effects of the MnIII ions. The maximum
around 10 K is field-dependent, being suppressed and shifted to
higher temperatures with increase of the applied magnetic field.
This is attributed to the additional Zeeman splitting of the ground
multiplet.
The overall behavior is very similar to that encountered in com-
plexes [Mn6(l3-O)2(O2CR)2(salox)6(EtOH)4] (R = Me, Ph) [8], which
is in accord with the structural similarities of the complexes. M
versus HTꢀ1 plots (Fig. 6) confirm that in 1 an S = 4 ground state
is also stabilized. Fits of the low-temperature data, and assuming
a well-isolated ground state, yield values of |D| ꢃ 0.5 cmꢀ1 and
g = 2.00 for the ground state. The magnitude of D for the ground
state of complex 1, is lower than the values previously reported
(|D| = 1.22 [14] and 1.39 cmꢀ1 [25a]) for the respective Mn6 com-
plexes with S = 4 ground states. Although these fits qualitatively
confirm the ground state as an S = 4 one, they fail to accurately
reproduce the data, probably due to the existence of low-lying ex-
cited states. This is to be expected, given the relatively weak ex-
change couplings between MnIII ions, which lead to a closely
Fig. 5. Partially labelled ORTEP plot of a small fragment of the chains of 5 with
ellipsoids drawn at the 30% probability level. For clarity hydrogen atoms have been
0
omitted. Symmetry transformations used to generate equivalent atoms: = x,
00
0.5 ꢀ y, 0.5 + z; = x, 0.5 ꢀ y, ꢀ0.5 + z.
Table 4
Structural characteristics of compounds 1–5
Mnꢁ ꢁ ꢁMn in [Mn3O] core (Å)
Mnꢁ ꢁ ꢁMn in [Mn2O2] core (Å)
Displacement of O2ꢀ/[Mn3] plane (Å)
Mn–N–O–Mn torsion angle (average) (°)
1
3.159(2)–3.274(5)
3.163(2)–3.268(2)
3.150(1)–3.278(2)
3.164(2)–3.262(2)
3.138(1)–3.305(1)
3.142(2)–3.301(2)
3.335(5)
3.314(2)
3.287(2)
3.304(2)
0.17
0.17
0.18
0.20
0.00
0.00
6.1, 19.7, 22.5 (16.1)
1.1, 21.6, 24.5 (15.7)
6.6, 22.7, 32.9 (20.7)
13.1, 13.3, 29.8 (18.7)
0.8, 16.7, 22.1 (13.2)
2.5, 14.0, 22.6 (13.0)
2
3a
4
5
a
For the two crystallographically independent molecules in the asymmetric unit.

C.P. Raptopoulou et al. / Polyhedron 27 (2008) 3575–3586
3583
former was accounted for by a mean-field correction term zj, while
the latter by an axial zfs parameter D, common for all spins
(D = D1 = D2 = D3). Initial fitting attempts over 2–300 K yielded very
poor results. It was then decided that the intramolecular interac-
tions might be too complex to be successfully accounted for by a
mean-field term, while the single-ion zfs parameters might be sub-
stantially different for the three sites. In addition, it was also
acknowledged that since both parameters zj and D are mainly
manifested in the low temperature range, inclusion of both in
our model might lead to overparametrization of our problem. Thus,
a simpler model was used, considering only the intra-unit ex-
change couplings, while the data over the 300–100 K range were
fitted. The corresponding Hamiltonian is given in Eq. (1):
^
^ ^
^ ^
^ ^
H ¼ ꢀ2½J1ðS1S2 þ S1S3Þ þ J1S2S3ꢄ
ð1Þ
Best-fit parameters for this model were J1 ꢃ J2 = ꢀ11.8 cmꢀ1
,
g = 2.01 with R = 4.1 ꢂ 10ꢀ5, corresponding to a S = 2 ground state.
3.3.2. AC susceptibility
Fig. 6.
vMT vs. T data for 1 under various magnetic fields. The M
vs. HTꢀ1
Given the pronounced similarities of
1 with [Mn6(l3-
dependence along with a fit assuming an isolated S = 4 ground state are shown in
O)2(O2CR)2(salox)6(EtOH)4] (R = Me, Ph), we decided to test
whether it also behaves as an SMM. Thus, in order to probe its
relaxation properties, AC susceptibility experiments were carried
out at low temperatures (Fig. 8). These showed characteristic
out-of-phase signals attributed to slow magnetic relaxation phe-
nomena. The signals show clear maxima between 2.13 K (10 Hz)
and 3.50 K (1400 Hz). For lower frequencies the maxima occur be-
the inset.
spaced energy spectrum. Thus, the values of the above fits should
not be considered as quantitatively accurate.
The S = 4 ground state is in accord [25] with the moderate twist-
ing of the Mn–N–O–Mn moieties within each Mn3 subunit, the
low 1.9 K, which is the lower limit of our instrument. Fits of these
average torsion angle being
av = 16°. Brechin and co-workers re-
eff =kB
data to an Arrhenius law,
s
¼
s0 ꢁ eU
T ; yielded parameters
cently showed [25b,c] that when this twisting in this family of
MnI6II SMMs becomes severe (as a consequence of the use of bulky
substituents on the oximate carbon), e.g.
av = 36.5° or 39.1°, the
resulting ground state is S = 12 and record Ueff values up to 86.4 K
can be achieved.
The
v
MT product of
5
(Fig. 7) at 300 K (0.5 T) is
6.64 cm3 molꢀ1 K, lower than the theoretically expected value for
three non-interacting S = 2 spins (9.00 cm3 molꢀ1 K, g = 2), indicat-
ing the interplay of antiferromagnetic interactions. Upon cooling,
this rapidly decreases, attaining a value of 0.51 cm3 molꢀ1 K at 2 K.
In order to interpret the experimental data, an isosceles model
of three interacting S = 2 spins was assumed, based on the repeat-
ing unit of the complex. Due to its polymeric structure, intra-unit
interactions are expected, while significant single-ion zero-field
splitting parameters D, are also common for MnIII complexes. The
Fig. 8. v0M (top) and v00MT vs. T data for 1 between 1.9 and 6 K for oscillating
magnetic fields of 0.1, 1, 10, 33, 100, 333, 1000 and 1400 Hz. A fit of the out-of-
phase data to an Arrhenius law is shown in the inset.
Fig. 7.
vMT vs. T data for 5 and best fit according to the model described in the text.

3584
C.P. Raptopoulou et al. / Polyhedron 27 (2008) 3575–3586
Fig. 9. Simplified view of 1, showing the coordination spheres of the MnIII ions. The
Jahn–Teller elongated bonds (shown in gray) are roughly parallel to each other.
s
0 = 5.5 ꢂ 10ꢀ8 s and Ueff = 27.4 K = 19.0 cmꢀ1. For an S = 4 ground
state and considering that
D
E = S2|D| > Ueff, it turns that the lowest
limit for |D| is 1.2 cmꢀ1, or in other words D < ꢀ1.2 cmꢀ1. These
results are in close agreement with the results obtained for
[Mn6(
[14,25a].
The relatively high value of D predicted by the relaxation exper-
l
3-O)2(O2CR)2(salox)6(R0OH)4] (R = H, Me, Ph; R0 = Me, Et)
Fig. 10. 1H NMR 1/T1 spin-lattice relaxation rate of complex 1 as a function of
temperature (solid line represents the fitted data according to Eq. (2)).
iments deserves some comment. Generally speaking, the anisotropy
of the ground state is related to the single-ion anisotropies of the
coupled spins. The projection of each single-ion anisotropy on the
molecular anisotropy axis (i.e. the z-axis) contributes to the total
anisotropy. High-spin (S = 2) MnIII ions, are highly anisotropic ions,
with their easy axes coinciding with their Jahn–Teller axes. In the
case of 1 we observe that the Jahn–Teller elongation axes are
roughly parallel (Fig. 9). This should have an important effect on
For the analysis of the NMR data, we adopt a simple model,
originally proposed by Hardeman et al. [34] based on a two level
scheme for the fluctuations of the orientation of the magnetic
moment.
In this model the NMR relaxation is caused by the local field
fluctuations originating from the magnetic excitations from the
ground state to the excited state, separated by the energy differ-
the total D value. We should note that for a D value of ꢀ1.2 cmꢀ1
,
ence
ground state for an average time s0 and in the excited state
for a lifetime 1, during which an effective transverse local field
D. Assuming that the Mn spin system remains in the
a cluster like Mn12 (S = 10) should present a spin-reversal energy
barrier of ca. 120 cmꢀ1. This relatively high D value should be
responsible for the fact that clear out-of-phase signals are observed
despite the relatively low spin (S = 4) of the ground state.
s
h\ appears at the nuclear site, the relaxation rate is expressed as
[34]
1
T1
s1s0
s
2 ðs1
þ
s0
Þ
ð
xN
s
Þ þ 1
3.4. Solid state NMR studies
¼ ðcNh Þ
ð2Þ
?
2
2
Nuclear magnetic resonance is a powerful tool in order to probe
the internal fields that are present in paramagnetic and magnetic
solids. In particular the nuclear T1 spin-lattice relaxation is sensi-
tive to the electron spin dynamics of the system since the nuclei
under investigation probe the fluctuating local magnetic fields in-
duced at the nuclear sites through the hyperfine interaction.
The 1H T1 spin-lattice relaxation time of complex 1 was mea-
sured at an applied magnetic field H = 2.4 T, between room and li-
quid-helium temperatures. The nuclear relaxation rate 1/T1 as a
function of temperature is shown in Fig. 10.
At high temperatures 1/T1 is nearly temperature independent
while below 70 K the rate increases exponentially with an appear-
ance of a peak at around 15 K. This behaviour has been frequently
encountered in previous 1H NMR studies of molecular clusters,
rings and single-molecule magnets and it is observed when the
temperature is of the order of the magnetic exchange constant
J/kB between the magnetic ions [33].
where
1
s
1
1
¼
þ
s0 s1
;
cN is the proton gyromagnetic ratio and xN is the Larmor angular
frequency.
This model has been frequently employed for the nuclear
magnetic relaxation in paramagnets [29d,33a,35].
Considering an activation-type process for the excitation, we set
ꢀ
ꢁ
1
s0
D
kBT
¼ B0 exp
ꢀ
ð3Þ
where
D represents the energy-level spacing between the ground
state and the first excited state. For the lifetime s1 we consider a
temperature independent term:
1=s1 ¼ C1
ð4Þ
In these molecular clusters, due to finite size of the spin system,
the low-lying energy states are well separated among themselves
and at low temperatures, the spin dynamics is governed by mag-
netic excitations between the ground state and the next excited
Setting N = 100 MHz (H = 4.7 T), and using Eqs. (3) and (4), the
experimental NMR points are fitted to Eq. (2), shown as solid line
m
in Fig. 10. The fitted parameters are
D/kB = 20 K for the energy
state separated by an energy difference
D
. This energy difference
gap and B0 = 5ꢂ109 sꢀ1 and C = 5ꢂ108 sꢀ1, which are slow fluctua-
tions of the order of the Larmor frequency. We observe that the line
describes the overall experimental features rather well.
D
is largely determined by the exchange interaction J/kB and the
peak of the nuclear relaxation rate 1/T1 occurs when the tempera-
ture is of the order of the energy gap. Thus the appearance of the
NMR peak provide us with a direct estimate of the energy differ-
ence between the ground state and the first excited state and can
be compared with results of magnetic measurements.
Therefore the NMR analysis of the low-frequency dynamics of
the Mn electron spin system clearly demonstrates the existence
of a sizeable energy gap between the ground state and the first
excited multiple. Also we have given a measure of the magnitude

C.P. Raptopoulou et al. / Polyhedron 27 (2008) 3575–3586
3585
(d) X. Yang, R.A. Jones, W.-K. Wong, Dalton Trans. (2008) 1676;
(e) I.S. Lee, Y.K. Chung, Inorg. Chem. Commun. 10 (2007) 593.
[3] R.E.P. Winpenny, J. Chem. Soc., Dalton Trans. (2002) 1.
of the local-field fluctuations present in the electronic spin
system.
[4] (a) As the most characteristic example of this approach we mention the
replacement of the bridging carboxylato ligands in the [Mn12
family of
]
4. Conclusions
complexes. See for example: P. Artus, C. Boskovic, J. Yoo, W.E. Streib, L.-C.
Brunel, D.N. Hendrickson, G. Christou, Inorg. Chem. 40 (2001) 4199;
(b) N.E. Chakov, W. Wernsdorfer, K.A. Abboud, D.N. Hendrickson, G. Christou,
J. Chem. Soc., Dalton Trans. (2003) 2243;
Three hexanuclear compounds presenting the [Mn6(
l3-O)2(l2-
3-O)2(O2CPh)2(salox)6(L1)2(L2)2]
OR)2]12+ core, complexes [MnI6II
(l
(c) J.M. Lim, Y. Do, J. Kim, Eur. J. Inorg. Chem. (2006) 711;
(d) E. Coronado, A. Forment-Aliaga, A. Gaita-Ariño, C. Giménez-Saiz, F.M.
Romero, W. Wernsdorfer, Angew. Chem., Int. Ed. 43 (2004) 6152.
[5] (a) See for example: V. Marvaud, C. Decroix, A. Scuiller, C. Guyard-Duhayon,
J. Vaissermann, F. Gonnet, M. Verdaguer, Chem. Eur. J. 9 (2003) 1677;
(b) Z.-G. Gu, W. Liu, Q.-F. Yang, X.-H. Zhou, J.-L. Zuo, X.-Z. You, Inorg. Chem. 46
(2007) 3236;
(L1 = py, L2 = H2O (1); L1 = Me2CO, L2 = H2O (2); L1 = L2 = MeOH (3))
have been prepared. The influence of the reactants stoichiometry,
the reaction media, the use of oligonuclear manganese clusters as
starting materials and crystallization, on the isolation of the conge-
ner compounds 1–3 is negligible. On the other hand, the presence
of Na+ ions (either as NaOMe or NaClO4 ꢁ H2O) in the corresponding
MeCN and Me2CO solutions, dramatically influenced the identity of
(c) A.J. Tasiopoulos, A. Vinslava, W. Wernsdorfer, A. Abboud, G. Christou,
Angew. Chem., Int. Ed. 43 (2004) 2117;
(d) M. Soler, E. Rumberger, K. Folting, D.N. Hendrickson, G. Christou,
Polyhedron 20 (2001) 1365;
the products, yielding the 1D complexes [MnI3IINa(
l3-O)(O2CPh)2(sa-
(e) M. Soler, W. Wernsdorfer, K. Folting, M. Pink, G. Christou, J. Am. Chem. Soc.
126 (2004) 2156;
(f) X. Ottenwaelder, J. Cano, Y. Journaux, E. Rivière, C. Brennan, M. Nierlich, R.
Ruiz-García, Angew. Chem., Int. Ed. 43 (2004) 850.
lox)3(S)]n (S = MeCN (4), Me2CO (5)). The magnetic properties of the
representative compounds 1 (hexanuclear) and 5 (polymeric) have
been studied. The magnetic susceptibility measurements from poly-
crystalline samples of 1 revealed the interplay of both ferro- and
antiferromagnetic interactions stabilizing an S = 4 ground state as
confirmed by M versus HTꢀ1 plots. The AC susceptibility measure-
ments at 1.9–6 K showed characteristic out-of-phase signals due to
slow magnetic relaxation phenomena. Fits of the AC data to the
[6] (a) See for example: J.-P. Costes, F. Dahan, W. Wernsdorfer, Inorg. Chem. 45
(2006) 5;
(b) J.-P. Costes, F. Nicodème, Chem. Eur. J. 8 (2002) 3442;
(c) J.-P. Costes, F. Dahan, F. Nicodème, Inorg. Chem. 42 (2003) 6556;
(d) L.K. Thompson, T.L. Kelly, L.N. Dawe, H. Grove, M.T. Lemaire, J.A.K. Howard,
E.C. Spencer, C.J. Matthews, S.T. Onions, Inorg. Chem. 43 (2004) 7605;
(e) S.K. Dey, L.K. Thompson, L.N. Dawe, Chem. Commun. (2006) 4967;
(f) L.N. Dawe, L.K. Thompson, Angew. Chem., Int. Ed. 46 (2007) 7440.
[7] (a) See for example: C.P. Raptopoulou, V. Tangoulis, E. Devlin, Angew. Chem.,
Int. Ed. 41 (2002) 2386;
Arrhenius law, yielded parameters
s
0 = 5.5 ꢂ 10ꢀ8
s
and
Ueff = 27.4 K = 19.0 cmꢀ1. The magnetic behavior of 1 is analogous
to the SMM behavior displayed by the prototype complexes of this
(b) A.K. Boudalis, C.P. Raptopoulou, B. Abarca, R. Ballesteros, M. Chadlaoui, J.-P.
Tuchagues, A. Terzis, Angew. Chem., Int. Ed. 45 (2006) 432;
(c) C.P. Raptopoulou, V. Tangoulis, V. Psycharis, Inorg. Chem. 39 (2000)
4452.
series, compounds [Mn6(l
3-O)2(O2CR)2(salox)6(R0OH)4] (R = Me, Ph;
R0 = Et). Investigation of the electron spin dynamics of 1 carried
out by solid state 1H NMR measurements, clearly revealed an energy
[8] (a) See for example: C.P. Raptopoulou, A.K. Boudalis, Y. Sanakis, V. Psycharis,
J.M. Clemente-Juan, M. Fardis, G. Diamantopoulos, G. Papavassiliou, Inorg.
Chem. 45 (2006) 2317;
gap of
D/kB = 20 K between the ground and the first excited state.
Magnetic susceptibility measurements from polycrystalline samples
of 5 revealed the presence of antiferromagnetic interactions in the
[MnI3II] unit. The data were fitted in the 100–300 K range by consid-
(b) A.K. Boudalis, B. Donnadieu, V. Nastopoulos, J.M. Clemente-Juan, A. Mari,
Y. Sanakis, J.-P. Tuchagues, S.P. Perlepes, Angew. Chem., Int. Ed. 43 (2004)
2266;
(c) A.K. Boudalis, Y. Sanakis, J.M. Clemente-Juan, B. Donnadieu, V.
Nastopoulos, A. Mari, Y. Coppel, J.-P. Tuchagues, S.P. Perlepes, Chem. Eur.
J. 14 (2008) 2514.
ering only intra-unit exchange couplings and yielded J = ꢀ11.8 cmꢀ1
corresponding to an S = 2 ground state.
,
[9] For a comprehensive review of our work see A.K. Boudalis, Y. Sanakis, C.P.
Raptopoulou, V. Psycharis, A. Terzis, Polyhedron 25 (2006) 1391 and references
7–10 cited therein.
Appendix A. Supplementary data
[10] C.P. Raptopoulou, A.K. Boudalis, Y. Sanakis, V. Psycharis, Polyhedron 24 (2005)
711.
[11] (a) A.K. Boudalis, V. Nastopoulos, A. Terzis, C.P. Raptopoulou, S.P. Perlepes,
Z. Naturforsch. 56b (2001) 122;
(b) C. Gkioni, A.K. Boudalis, Y. Sanakis, L. Leondiadis, V. Psycharis, C.P.
Raptopoulou, Polyhedron 27 (2008) 2315.
[12] Th.C. Stamatatos, A.K. Boudalis, Y. Sanakis, C.P. Raptopoulou, Inorg. Chem. 45
(2006) 7372.
[13] C.P. Raptopoulou, V. Psycharis, Inorg. Chem. Commun. 11 (2008) 1194.
[14] C.J. Milios, C.P. Raptopoulou, A. Terzis, F. Lloret, R. Vicente, S.P. Perlepes,
A. Escuer, Angew. Chem., Int. Ed. 43 (2004) 210.
[15] J.B. Vincent, H.-R. Chang, K. Folting, J.C. Huffman, G. Christou, D.N.
Hendrickson, J. Am. Chem. Soc. 109 (1987) 5703.
[16] M.W. Wemple, H.-L. Tsai, S. Wang, J.P. Claude, W.E. Streib, J.C. Huffman, D.N.
Hendrickson, G. Christou, Inorg. Chem. 35 (1996) 6437.
[17] J.-M. Clemente-Juan, C. Mackiewicz, M. Verelst, F. Dahan, A. Bousseksou,
Y. Sanakis, J.-P. Tuchagues, Inorg. Chem. 41 (2002) 1478.
[18] F. James, M. Roos, Comput. Phys. Commun. 10 (1975) 343.
[19] E. Fukushima, S.B.W. Roeder, Experimental Pulse NMR, Addison Wesley,
Reading, MA, 1981.
CCDC 694623, 694624, 694625, 694626 and 694627 contain the
supplementary crystallographic data for 1 ꢁ 6MeCN, 2 ꢁ 2Me2CO,
3 ꢁ MeOH ꢁ 0.5H2O, 4 ꢁ nMeCN and 5 ꢁ nMe2CO. These data can be
obtained free of charge via http://www.ccdc.cam.ac.uk/conts/
retrieving.html, or from the Cambridge Crystallographic Data Cen-
tre, 12 Union Road, Cambridge CB2 1EZ, UK; fax: (+44) 1223-336-
033; or e-mail: deposit@ccdc.cam.ac.uk. Tables of selected bond
distances and angles for 2 ꢁ 2Me2CO, 3 ꢁ MeOH ꢁ 0.5H2O and
5 ꢁ nMe2CO are also available. Supplementary data associated with
this article can be found, in the online version, at doi:10.1016/
j.poly.2008.08.021.
References
[1] (a) For some recent examples on manganese iron complexes see: P. Baran, R.
ˇ
Boca, I. Chakraborty, J. Giapintzakis, R. Herchel, Q. Huang, J.E. McGrady, R.G.
[20] S. Parsons, M. Murie, R. Winpenny, P. Wood, Private Communication to the
Raptis, Y. Sanakis, A. Simopoulos, Inorg. Chem. 47 (2008) 645;
(b) J. Ballmann, S. Dechert, E. Bill, U. Ryde, F. Meyer, Inorg. Chem. 47 (2008)
1586;
(c) A. Mishra, W. Wernsdorfer, K.A. Abboud, G. Christou, Chem. Commun.
(2005) 54;
CCDC, Code XAJVEK.
[21] Rigaku/MSCCRYSTALCLEAR, Rigaku/MSC Inc., The Woodlands, Texas, USA, 2005.
[22] G.M. Sheldrick, SHELXS-97: Structure Solving Program, University of Göttingen,
Germany, 1997.
[23] G.M. Sheldrick, SHELXL-97: Crystal Structure Refinement, University of
Göttingen, Germany, 1997.
[24] (a) For reviews see: G. Aromi, E.K. Brechin, Struct. Bond. 122 (2006) 1;
(b) D. Gatteschi, R. Sessoli, Angew. Chem., Int. Ed. 42 (2003) 268;
(c) G. Christou, D. Gatteschi, D.N. Hendrickson, R. Sessoli, MRSBull. 25(2000)66.
[25] (a) C.J. Milios, A. Vinslava, A.G. Whittaker, S. Parsons, W. Wernsdorfer,
G. Christou, S.P. Perlepes, E.K. Brechin, Inorg. Chem. 45 (2006) 5272;
(b) C.J. Milios, A. Vinslava, P.A. Wood, S. Parsons, W. Wernsdorfer, G. Christou,
S.P. Perlepes, E.K. Brechin, J. Am. Chem. Soc. 129 (2007) 8;
(c) C.J. Milios, A. Vinslava, W. Wernsdorfer, S. Moggach, S. Parsons, S.P.
Perlepes, G. Christou, E.K. Brechin, J. Am. Chem. Soc. 129 (2007) 2754;
(d) A. Mishra, Y. Pushkar, J. Yano, V.K. Yachandra, W. Wernsdorfer, K.A.
Abboud, G. Christou, Inorg. Chem. 47 (2008) 1940;
(e) R. Tagore, R.H. Crabtree, G.W. Brudvig, Inorg. Chem. 47 (2008) 1815.
[2] (a) For some recent examples see: A. Mishra, A.J. Tasiopoulos, W. Wernsdorfer,
E.E. Moushi, B. Moulton, M.J. Zaworotko, K.A. Abboud, G. Christou, Inorg. Chem.
47 (2008) 4832;
(b) V. Chandrasekhar, B. Murugesa Pandian, R. Boomishankar, A. Steiner, J.J.
Vittal, A. Houri, R. Clérac, Inorg. Chem. 47 (2008) 4918;
(c) D. Li, R. Clé rac, O. Roubeau, E. Harté, C. Mathonière, R. Le Bris, S.M. Holmes,
J. Am. Chem. Soc. 130 (2008) 252;

3586
C.P. Raptopoulou et al. / Polyhedron 27 (2008) 3575–3586
(d) C.J. Milios, A. Vinslava, W. Wernsdorfer, A. Prescimone, P.A. Wood, S. Parsons,
(d) A. Vlachos, V. Psycharis, C.P. Raptopoulou, N. Lalioti, Y. Sanakis, G.
Diamantopoulos, M. Fardis, M. Karayanni, G. Papavassiliou, A. Terzis, Inorg.
Chim. Acta 357 (2004) 3162;
S.P. Perlepes, G. Christou, E.K. Brechin, J. Am. Chem. Soc. 129 (2007) 6547;
(e) C.J. Milios, R. Inglis, A. Vinslava, R. Bagai, W. Wernsdorfer, S. Parsons, S.P.
Perlepes, G. Christou, E.K. Brechin, J. Am. Chem. Soc. 129 (2007) 12505.
(e) V. Psycharis, C.P. Raptopoulou, A.K. Boudalis, Y. Sanakis, M. Fardis, G.
Diamantopoulos, G. Papavassiliou, Eur. J. Inorg. Chem. (2006) 3710.
[30] P. Chaudhuri, E. Rentschler, F. Birkelbach, C. Krebs, E. Bill, T. Weyhermüller, U.
Flörke, Eur. J. Inorg. Chem. (2003) 541.
[31] (a) N.E. Chakov, L.N. Zakharov, A.L. Rheingold, K.A. Abboud, G. Christou, Inorg.
Chem. 44 (2005) 4555;
[26] P. Chaudhuri, Coord. Chem. Rev. 243 (2003) 143.
[27] (a) J.M. Thorpe, R.L. Beddoes, D. Collison, C.D. Garner, M. Helliwell, J.M.
Holmes, P.A. Tasker, Angew. Chem., Int. Ed. 38 (1999) 1119;
(b) P. Chaudhuri, M. Hess, E. Rentschler, T. Weyhermüller, U. Flörke, New J.
Chem. (1998) 553;
(c) V. Zerbib, F. Robert, P. Gouzerh, J. Chem. Soc., Chem. Commun. (1994)
2179.
(b) X. Xia, M. Verelst, J.-C. Daran, J.-P. Tuchagues, J. Chem. Soc., Chem.
Commun. (1995) 2155.
[28] (a) Th.C. Stamatatos, D. Foguet-Albiol, C.C. Stoumpos, C.P. Raptopoulou, A.
Terzis, W. Wernsdorfer, S.P. Perlepers, G. Christou, J. Am. Chem. Soc. 127
(2005) 15380;
[32] C.J. Milios, P.A. Wood, S. Parsons, D. Foguet-Albiol, C. Lampropoulos, G.
Christou, S.P. Perlepes, E.K. Brechin, Inorg. Chim. Acta 360 (2007) 3932.
[33] (a) M. Fardis, G. Diamantopoulos, M. Karayianni, G. Papavassiliou, V.
Tangoulis, A. Konsta, Phys. Rev. B 65 (2001) 014412;
(b) Th.C. Stamatatos, D. Foguet-Albiol, S.-C. Lee, C.C. Stoumpos, C.P.
Raptopoulou, A. Terzis, W. Wernsdorfer, S.O. Hill, S.P. Perlepers, G. Christou,
J. Am. Chem. Soc. 129 (2007) 9484.
(b) D. Procissi, B.J. Suh, A. Lascialfari, F. Borsa, A. Caneschi, A. Cornia, J. Appl.
Phys. 91 (2002) 7173;
[29] (a) R. Wu, M. Poyraz, F.E. Sowrey, C.E. Anson, S. Wocadlo, A.K. Powell, U.A.
Jayasooriya, R.D. Cannon, T. Nakamoto, M. Katada, H. Sano, Inorg. Chem. 37
(1998) 1913;
(c) M. Belesi, A. Lascialfari, D. Procissi, Z.H. Jang, F. Borsa, Phys. Rev. B 72
(2005) 014440.
[34] G.E.G. Hardeman, N.J. Poulis, W.v.D. Lugt, Physica (Utrecht) 22 (1956) 48.
[35] (a) T. Kohmoto, T. Goto, S. Maegawa, N. Fujiwara, Y. Fukuda, M. Kunitomo, M.
Mekata, Phys. Rev. B 49 (1994) 6028;
(b) J. An, Z.-D. Chen, J. Bian, X.-L. Jin, S.-X. Wang, G.-X. Xu, Inorg. Chim. Acta
287 (1999) 82;
(c) J. Li, S. Yang, F. Zhang, Z. Tang, S. Ma, Q. Shi, Q. Wu, Z. Huang, Inorg. Chim.
Acta 294 (1999) 109;
(b) T. Goto, T. Koshiba, T. Kubo, K. Agawa, Phys. Rev.
104408.
B 67 (2003)