Preparation and characterization of CrIII, MnII, FeII, CoII and NiII complexes of a hexaazadithiophenolate macrocycle

Article abstract of DOI:10.1039/b513717a

The ligating properties of the 24-membered macrocyclic dinucleating hexaazadithiophenolate ligand (L^Me)^2- towards the transition metal ions Cr^II, Mn^II, Fe^II, Co ^II, Ni^II and Zn^II have been examined. It is demonstrated that this ligand forms an isostructural series of bioctahedral [(L^Me)M^II₂(OAc)]⁺ complexes with Mn^II (2), Fe^II (3), Co^II (4), Ni^II (5) and Zn^II (6). The reaction of (L^Me)^2- with two equivalents of CrCl₂ and NaOAc followed by air-oxidation produced the complex [(L^Me)Cr^IIIH₂(OAc)]²⁺ (1), which is the first example for a mononuclear complex of (L ^Me)^2-. Complexes 2-6 contain a central N₃M ^II(μ-SR)₂(μ-OAc)M^IIN₃ core with an exogenous acetate bridge. The Cr^III ion in 1 is bonded to three N and two S atoms of (L^Me)^2- and an O atom of a monodentate acetate coligand. In 2-6 there is a consistent decrease in the deviations of the bond angles from the ideal octahedral values such that the coordination polyhedra in the dinickel complex 5 are more regular than in the dimanganese compound 2. The temperature dependent magnetic susceptibility measurements reveal the magnetic exchange interactions in the [(L ^Me)M^II₂(OAc)]⁺ cations to be relatively weak. Intramolecular antiferromagnetic exchange interactions are present in the Mn^II₂, Fe^II₂ and Co^II₂ complexes where J = -5.1, -10.6 and ～-2.0 cm^-1 (H = -2JS₁S₂). In contrast, in the dinickel complex 5 a ferromagnetic exchange interaction is present with J = +6.4 cm^-1. An explanation for this difference is qualitatively discussed in terms of the bonding differences. The Royal Society of Chemistry 2006.

Full text of DOI:10.1039/b513717a

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/dalton | Dalton Transactions
Preparation and characterization of CrIII, MnII, FeII, CoII and NiII complexes
of a hexaazadithiophenolate macrocycle†
Yves Journaux,c Thorsten Glaser,b Gunther Steinfeld,a Vasile Lozana and Berthold Kersting*a
Received 27th September 2005, Accepted 25th November 2005
First published as an Advance Article on the web 22nd December 2005
DOI: 10.1039/b513717a
The ligating properties of the 24-membered macrocyclic dinucleating hexaazadithiophenolate ligand
2−
(LMe
)
towards the transition metal ions CrII, MnII, FeII, CoII, NiII and ZnII have been examined. It is
demonstrated that this ligand forms an isostructural series of bioctahedral [(LMe)MII2(OAc)]+ complexes
with MnII (2), FeII (3), CoII (4), NiII (5) and ZnII (6). The reaction of (LMe
)
with two equivalents of
2−
CrCl2 and NaOAc followed by air-oxidation produced the complex [(LMe)CrIIIH2(OAc)]2+ (1), which is
the first example for a mononuclear complex of (LMe)2−. Complexes 2–6 contain a central
N3MII(l-SR)2(l-OAc)MIIN3 core with an exogenous acetate bridge. The CrIII ion in 1 is bonded to three
N and two S atoms of (LMe
)
and an O atom of a monodentate acetate coligand. In 2–6 there is a
2−
consistent decrease in the deviations of the bond angles from the ideal octahedral values such that the
coordination polyhedra in the dinickel complex 5 are more regular than in the dimanganese compound
2. The temperature dependent magnetic susceptibility measurements reveal the magnetic exchange
interactions in the [(LMe)MII2(OAc)]+ cations to be relatively weak. Intramolecular antiferromagnetic
exchange interactions are present in the MnII2, FeII and CoII complexes where J = −5.1, −10.6 and
2
2
∼−2.0 cm−1 (H = −2JS1S2). In contrast, in the dinickel complex 5 a ferromagnetic exchange
interaction is present with J = +6.4 cm−1. An explanation for this difference is qualitatively discussed in
terms of the bonding differences.
Introduction
There is currently much interest in the coordination chemistry of
macrocyclic dinucleating polyaza-dithiophenolate ligands1,2 of the
Robson type.3,4 Motivations in this area are diverse and include the
preparation of model compounds for dinuclear metallo enzymes,5
the stabilization of unusual metal oxidation states,6 and attempts
to understand, achieve and control binuclear metal reactivity,7,8 to
name a few. We are interested in the coordination chemistry of the
hexaazadithiophenolate macrocycle H2LMe (Scheme 1) and have
recently reported on the ligating properties of this ligand system
towards various late 3d elements such as Co,9 Ni,10 Cu and Zn11
and also towards some heavy-metals such as Cd, Hg and Pb.12
As part of this program, we sought to extend our exploration
Scheme 1 Ligand and complexes.
to the syntheses of complexes incorporating early transition
metals. Herein we describe the synthesis and characterization of
chromium(III), manganese(II) and iron(II) complexes of H2LMe. To
the best of our knowledge these are the first examples of complexes
of macrobinucleating polyaza-dithiophenolate ligands with such
transition metals.13 The structures and magnetic properties of the
present complexes are compared with those of the CoII, NiII and
ZnII complexes that we have reported previously.9–11
aInstitut fu¨r Anorganische Chemie, Universita¨t Leipzig, Johannisallee
29, D-04103, Leipzig, Germany. E-mail: b.kersting@uni-leipzig.de;
Fax: +49/(0)341-97-36199
bFakulta¨t fu¨r Chemie der Universita¨t Bielefeld, Lehrstuhl fu¨r Anorganische
Chemie 1, Universita¨tsstr. 25, 33615, Bielefeld, Germany
cLaboratoire de Chimie Inorganique et Mate´rieux Mole´culaires, Universite´
Pierre et Marie Curie 6, case courier 42, F-75005, Paris, FranceCNRS
UMR 7071, F-75005, Paris, France
Results and discussion
Synthesis and spectroscopic characterization
† Electronic supplementary information (ESI) available: 1: Magnetic sus-
ceptibility data of [(LMe)MnII2(l-OAc)]BPh4 (2BPh4) as a function of the
temperature. 2: Magnetic susceptibility data of [(LMe)FeII2(l-OAc)]BPh4
(3BPh4) as a function of the temperature. 3: Magnetic susceptibility data
of [(LMe)CoII2(l-OAc)]BPh4 (4BPh4) as a function of the temperature.
4: Magnetic susceptibility data of [(LMe)NiII2(l-OAc)]BPh4 (5BPh4) as a
function of the temperature. See DOI: 10.1039/b513717a
The investigated compounds and their labels are collected in
Scheme 1. Of these, 4–6 have been reported previously.9–11
Attempts to prepare dinuclear chromium complexes were not
2−
successful (Scheme 2). Thus, treatment of (LMe
)
(prepared in
situ from H2LMe·6HCl and 8 equiv. of NEt3) with two equivalents
1738 | Dalton Trans., 2006, 1738–1748
This journal is
The Royal Society of Chemistry 2006
©

View Article Online
Both 2 and 3 display no absorptions in the 430–1600 nm range
with e >40 M−1 cm−1, indicative of high-spin d5 (Mn2+) and d6 (Fe2+)
configurations. The IR spectra of 2 and 3 are almost identical to
those of the acetato-bridged complexes 4–6 that we have reported
previously.9–11 The most prominent features are the absorptions at
1580 and 1430 cm−1 (see Table 1), which are associated with the
asymmetric [masym(OAc)] and symmetric acetate stretching modes
2−
[msym(OAc)].14 This indicates that the macrocycle (LMe
)
forms an
isostructural series of bioctahedral [(LMe)MII2(OAc)]+ cations with
the divalent ions MnII, FeII, CoII, NiII and ZnII. It is supported by
the crystal structures of 2 and 3 described below.
Electrochemistry
In view of the air-stability of the MnII and FeII complexes, it
2
2
was of interest to determine the redox properties of 2 and 3.
The electrochemical data listed in Table 1 have been measured
by cyclic voltammetry in CH3CN solution using NnBu4PF6 as the
supporting electrolyte. The data for 4 and 5 have been recorded
previously and are included for comparative purposes.9,10
Fig. 1 shows the CV of the MnII complex 2. The CV exhibits
2
two waves, one at E1 = +0.60 V (vs. SCE) with a peak to peak
1/2
separation of 180 mV and one at E2 = +1.19 V with a peak-
1/2
to-peak separation of 130 mV. The CV of the FeII complex 3 is
2
Scheme 2 Preparation and structures of complexes 1–3.
very similar, but the two redox waves are observed at less anodic
potentials at E1 = +0.32 V (vs. SCE) and at E2 = +1.00 V,
1/2
1/2
of anhydrous CrIICl2, followed by the addition of sodium acetate
and air-oxidation, gave a dark green solution, from which upon
addition of LiClO4 green crystals of [(LMe)CrIIIH2(OAc)](ClO4)2
(1(ClO4)2) could be isolated in 35% yield.
respectively. Thus, as was observed previously for the dinuclear Co
and Ni complexes, the divalent MnII and FeII oxidation states are
enormously stabilized over their trivalent ones.
The preparations of the dimanganese(II) and diiron(II) com-
plexes 2 and 3, respectively, are also illustrated in Scheme 2. Treat-
2−
ment of (LMe
)
with 2 equivalents of Mn(OAc)2·4H2O resulted
in a colourless solution, from which upon addition of LiClO4
colourless crystals of [(LMe)MnII2(OAc)]ClO4 (2·ClO4) precipitated
in ca. 90% yield. Likewise, the reaction of (LMe
)
with Fe(OAc)2
2−
(prepared in situ from FeCl2 and NaOAc) proceeded smoothly
and gave a pale-yellow solution of the analogous complex 3.
−
This complex was also readily isolated as its ClO4 salt. The
−
corresponding BPh4 salts were prepared in order to grow single-
crystals suitable for X-ray crystallography. All compounds gave
satisfactory elemental analyses and were further characterized
by IR and UV/Vis-spectroscopy, cyclic voltammetry and X-ray
structure analysis. The tetraphenylborate salts of compounds 2, 3
and 4 were also characterized by variable temperature magnetic
susceptibility measurements.
Fig. 1 Cyclic voltammograms of complexes 2–6 in MeCN solution.
Experimental conditions: Pt-disk working electrode, Ag wire reference
electrode, solutions were ca. 1.0 × 10−3 M in sample and 0.10 M in
supporting electrolyte (NnBu4PF6), scan rate = 100 mV s−1
.
Table 1 IR and CV data for 2–6a
Compound
IRa/cm−1
E11/2/V (DEp/V), E2 (DEp)b
Ref.
1/2
1
2
3
4
5
6
[(LMe)CrIIIH2(OAc)]2+
[(LMe)MnII2(OAc)]+
[(LMe)FeII2(OAc)]+
[(LMe)CoII2(OAc)]+
[(LMe)NiII2(OAc)]+
[(LMe)ZnII2(OAc)]+
1629, 1412
1583, 1422
1577, 1423
1587, 1434
1588, 1426
1585, 1428
n.d.c
This work
This work
This work
9
10
11
0.60 (0.18), 1.19 (0.13)
0.32 (0.09), 1.00 (0.13)
0.21 (0.12), 0.60 (0.15)
0.56 (0.15), 1.36 (0.14)
0.88 (irr.)
a Data refer to the perchlorate salts measured as KBr pellets. b Data recorded using the perchlorate salts in CH3CN solution. All potentials are referenced
to SCE. For experimental conditions see Experimental Section. Ex1/2 = (Eox + Eredp)/2 for reversible one-electron transfer processes; values in parentheses
p
represent peak-to-peak separations (DEp = (Eox,p − Ered,p)). c Not determined.
This journal is
The Royal Society of Chemistry 2006
Dalton Trans., 2006, 1738–1748 | 1739
©

View Article Online
Table 2 Crystallographic data for 1(ClO4)2, 2BPh4·3MeCN and 3BPh4·2MeCN·0.5EtOH
Compound
1(ClO4)2
2BPh4·3MeCN
3BPh4·2MeCN·0.5EtOH
Formula
Mr
C40H69Cl2CrN6O10S2
981.03
P21/c
14.490(3)
24.163(5)
13.777(3)
90.00
96.84(3)
90.00
C70H96BMn2N9O2S2
1280.37
P1
14.555(3)
16.707(3)
17.295(3)
113.54(3)
110.66(3)
94.92(3)
3479(1)
C69H96BFe2N9O2.5S2
1264.17
P1
14.568(3)
16.775(3)
17.512(4)
113.66(3)
110.39(3)
95.10(3)
3541(1)
¯
¯
Space group
˚
a/A
˚
b/A
˚
c/A
a/◦
b/◦
c /◦
3
˚
V/A
4789(2)
4
1.361
Z
2
2
Dc/g cm−3
1.222
1.186
Cryst. size/mm
l(Mo-Ka)/mm−1
2h Limits/◦
Measured refl.
Independent refl.
Observed refl.a
No. parameters
R1b (R1 all data)
wR2c (wR2 all data)
0.15 × 0.15 × 0.15
0.18 × 0.18 × 0.15
0.20 × 0.15 × 0.15
0.497
0.472
0.516
4.50–56.66
30037
11424
3881
545
3.56–56.58
31437
16227
11386
775
2.80–56.64
32034
16460
6953
751
0.0453 (0.1162)
0.1371 (0.1403)
0.544, −0.491
0.0665 (0.1607)
0.2049 (0.2192)
0.719, −0.727
0.0326 (0.0802)
0.0535 (0.0858)
0.400, −0.399
−3
˚
Max., min. peaks/e A
ꢀ
ꢀ
ꢀ
ꢀ
a Observation criterion: I > 2r(I). b R1 = ꢀFo| − |Fcꢀ/ |Fo|. c wR2 = { [w(Fo − Fc2)2]/ [w(Fo2)2]}
.
2
1/2
It should be noted that the dizinc complex 6 displays a single,
irreversible redox wave at E = +0.88 V vs. SCE. Since the two
zinc(II) ions are redox inactive, the anodic peak can only be
assigned to a ligand-centered redox reaction. It is very likely that
this process corresponds to the oxidation of the thiophenolate
sulfur atoms yielding a zinc-bound thiyl radical as indicated in
eqn (1). This is supported by recent results of Schro¨der and co-
workers who have demonstrated the redox non-innocence of the
bridging thiophenolate donors in the dinuclear nickel complex
[(L1)Ni2]2+ ((L1)2− represents a related dinucleating tetraimine-
dithiophenolate macrocycle).15 Schro¨der’s complex undergoes a
reversible oxidation at similar anodic potentials of E1/2 = +0.93 V
vs. SCE, assigned to the formation of a charge-delocalized
[(L1)Ni2]3+ trication with ca. 30% of the unpaired electron density
delocalized onto the bridging thiolate sulfur atoms.
chemical oxidation led to unidentified decomposition products.
Thus, while some of the above oxidations appear electrochemically
reversible, they are all chemically irreversible.16
Description of the crystal structures
The X-ray crystal structures of 1(ClO4)2, 2BPh4·3MeCN and
3BPh4·2MeCN·0.5EtOH were determined to establish the ge-
ometries about the metal ions as well as the bonding modes of
the macrocycle and the coligands. Experimental crystallographic
data are summarized in Table 2. Further data are available in the
supporting information.
[(LMe)CrIIH2(OAc)](ClO4)2 (1(ClO4)2). This salt crystallizes in
the monoclinic space group P21/c. The structure reveals the
presence of mononuclear [(LMe)CrIIIH2(OAc)]2+ dications and two
perchlorate anions. Fig. 2 shows the structure of the dication 1.
Selected bond lengths and angles are listed in the Figure caption.
The CrIII ion is in a distorted octahedral environment, being
coordinated by two sulfur and three nitrogen atoms of the macro-
cycle and a monodentate acetate ion. Two of the remaining three
•
[(LMe )ZnII2(OAc)]2+ + e− ↔ [(LMe)ZnII2(OAc)]+; E1
(1)
1/2
On the basis of these data, similar ligand-based oxidations are
likely for the present Mn2, Fe2 and Ni2 complexes. The zinc
complex 6 provides a lower limit of E1/2 = ca. 0.90 V for the
onset of ligand oxidation. Since the E1 values for 2–5 are all
1/2
˚
amine nitrogens are protonated. The average Cr–N (2.192(5) A),
well below this limit, the corresponding oxidations are tentatively
˚
˚
Cr–S (2.360(2) A) and Cr–O (1.958(4) A) bond lengths are
similar to those reported for [CrIII(en)2(SCH2CO2)]+,17 and other
octahedral CrIII complexes with mixed N,O,S coordination.18–20
The coordinated diethylene triamine unit binds facially with one
larger and two smaller N–Cr–N bond angles. This is normal for
this tridendate subunit21 and places the remaining donors in a cis-
orientation to each other. The structure is further stabilized by the
intramolecular hydrogen bonding interactions between O(2) and
H(6) and S(2) and H(5), respectively.
assigned to metal-centered oxidations of the MIIMII complexes to
their mixed-valent MIIMIII forms [eqn (2)]. Since the E2 values
1/2
are all above 0.90 V the subsequent oxidations are believed to be
largely ligand-centered in nature, as indicated in eqn (3).
[(LMe)MIIIMII(OAc)]2+ + e− ↔ [(LMe)MII2(OAc)]+: E1
(2)
(3)
1/2
•
[(LMe )MIIIMII(OAc)]3+ + e− ↔ [(LMe)MIIIMII(OAc)]2+: E2
1/2
It should be noted that the oxidized species, except the
kinetically inert CoIICoIII and CoIII complexes 42+ and 43+,9 are
[(LMe)MnII2(l-OAc)]BPh4 (2BPh4·3MeCN) and [(LMe)FeII2(l-
OAc)]BPh4 (3BPh4·2MeCN·0.5EtOH. The crystal structure
determinations of 2BPh4·3MeCN and 3BPh4·2MeCN·0.5EtOH
2
only stable on the time scale of a cyclic voltammetry experiment.
All attempts to prepare these compounds by electrochemical or
1740 | Dalton Trans., 2006, 1738–1748
This journal is
The Royal Society of Chemistry 2006
©

View Article Online
Fig. 4 displays the central coordination units of the isostructural
[(LMe)MII2(l-OAc)]+ cations. Selected bond lengths and angles are
listed in Tables 3 and 4.
Fig. 2 Structure of the cation 1 in crystals of [(LMe)CrIIIH2(OAc)](ClO4)2
(1(ClO4)2) with thermal ellipsoids drawn at the 50% probability level.
Hydrogen atoms are omitted for reasons of clarity. Selected bond
◦
˚
lengths (A) and angles ( ): Cr–O(1) 1.958(4), Cr–N(1) 2.231(4), Cr–N(2)
2.165(5), Cr–N(3) 2.181(4), Cr–S(1) 2.362(2), Cr–S(2) 2.357(2), O(1)–C(39)
1.292(7), O(2)–C(39) 1.207(7), H(6) · · · O(2) 1.979, H(5) · · · S(2) 2.559;
N(1)–Cr–N(2) 82.1(2), N(1)–Cr–N(3) 102.2(2), N(2)–Cr–N(3) 82.6(2),
S(1)–Cr–S(2) 91.62(6), O(1)–Cr–N(1) 165.8(2), N(2)–Cr–S(2) 170.6(1),
N(3)–Cr–S(1) 168.0(1).
Fig. 4 Structure of the bioctahedral N3MnII(l-SR)2(l-OAc)MnIIN3 core
of 2. Thermal ellipsoids are drawn at the 50% probability level.
As can be seen from Fig. 3, the supporting ligand adopts a
bowl-shaped “calixarene-like” conformation, analogous to that
found in the acetato-bridged complexes 4–6.9–11 The metal ions
are in a strongly distorted octahedral N3S2O environment, being
coordinated by two sulfur and three facially oriented nitrogen
atoms of the macrocycle and a symmetrically bridging acetate ion
to give a N3MII(SR)2(OAc)MIIN3 core. Similar dimanganese(II)
complexes with phenolate bridges instead of the thiophenolates
have been reported by Nagata and co-workers.22 The distortions
from the ideal octahedral geometry are manifested in the cis and
trans L–M–L bond angles, which deviate by as much as 17◦ from
their ideal values (see Table 4).
confirmed the presence of well-separated bioctahedral [(LMe)-
MII2(l-OAc)]+ cations, tetraphenylborate anions and acetonitrile
and ethanol molecules of solvent of crystallization. Shown in Fig. 3
is the structure of the cation 2. The cation 3 is isostructural with 2.
The long average metal–ligand bond lengths in 2 are nor-
mal for octahedral manganese(II) with high-spin electronic
configuration.23 For example, the average Mn–N distance of
˚
˚
2.368(2) A in 2 corresponds nicely to the average value of 2.337 A
in [(bpea)MnII2(l-OAc)3]+,24,25 and 2.35(1) A in [(Me3tacn)Mn 2(l-
OAc)3],26 which both contain high-spin MnII. Likewise, the Mn–
S distance of 2.609(1) A is very close to the average Mn –S
bond distance of 2.566(2) A in a related octahedral Mn complex
with an N3S2O donor set.27 In the case of 3, the average Fe–N
II
˚
II
˚
Fig.
3
Structure of the [(LMe)MnII2(OAc)]+ cation in crystals of
II
˚
2BPh4·3MeCN. Thermal ellipsoids are drawn at the 50% probability level.
Hydrogen atoms are omitted for reasons of clarity.
˚
Table 3 Selected bond lengths (A) in complexes 2–6
2 (M = Mn)
3 (M = Fe)
4 (M = Co)
5 (M = Ni)
6 (M = Zn)
M(1)–O(1)
M(1)–N(1)
M(1)–N(2)
M(1)–N(3)
M(1)–S(1)
M(1)–S(2)
M(2)–O(2)
M(2)–N(4)
M(2)–N(5)
M(2)–N(6)
M(2)–S(1)
M(2)–S(2)
M–Na
2.091(1)
2.407(1)
2.355(2)
2.337(2)
2.6017(7)
2.6166(8)
2.083(1)
2.401(2)
2.333(2)
2.380(2)
2.630(1)
2.588(1)
2.368(2)
2.087(1)
2.609(1)
3.456(1)
2.029(2)
2.406(3)
2.256(3)
2.311(3)
2.526(1)
2.512(1)
2.021(2)
2.377(3)
2.248(3)
2.366(3)
2.546(2)
2.500(2)
2.327(3)
2.025(2)
2.521(1)
3.421(1)
2.011(2)
2.345(3)
2.188(3)
2.277(3)
2.525(1)
2.476(1)
2.015(2)
2.285(3)
2.195(3)
2.335(3)
2.472(1)
2.537(2)
2.271(3)
2.013(2)
2.503(1)
3.448(1)
1.998(2)
2.281(2)
2.152(2)
2.251(2)
2.4929(9)
2.448(1)
2.008(2)
2.244(2)
2.158(2)
2.295(2)
2.493(1)
2.4500(9)
2.230(2)
2.003(2)
2.471(1)
3.483(1)
2.026(2)
2.379(2)
2.202(2)
2.316(2)
2.5508(9)
2.515(1)
2.027(2)
2.314(2)
2.219(2)
2.377(2)
2.572(1)
2.508(1)
2.301(2)
2.027(2)
2.536(1)
3.427(1)
M–Oa
M–Sa
M · · · M
a Average values.
This journal is
The Royal Society of Chemistry 2006
Dalton Trans., 2006, 1738–1748 | 1741
©

View Article Online
Table 4 Selected bond angles (◦) in complexes 2–6
2 (M = Mn)
3 (M = Fe)
4 (M = Co)
5 (M = Ni)
6 (M = Zn)
N(1)–M(1)–N(2)
N(1)–M(1)–N(3)
N(2)–M(1)–N(3)
N(4)–M(2)–N(5)
N(4)–M(2)–N(6)
N(5)–M(2)–N(6)
N–M–Na
76.50(6)
102.69(5)
77.48(6)
77.12(6)
103.64(7)
77.63(5)
85.84(6)
12.93(6)
77.32(10)
100.87(10)
79.33(11)
78.78(11)
102.16(11)
78.54(11)
86.17(10)
11.51(10)
79.40(10)
99.21(10)
81.34(12)
80.68(10)
99.51(9)
79.87(10)
86.67(10)
9.57(10)
81.23(8)
98.90(8)
82.47(9)
82.10(8)
98.83(7)
81.06(7)
87.43(8)
8.48(8)
78.92(8)
98.84(7)
80.98(9)
80.23(7)
99.00(7)
79.05(7)
86.17(8)
9.77(8)
devb
N(1)–M(1)–S(1)
N(2)–M(1)–S(1)
N(2)–M(1)–S(2)
N(3)–M(1)–S(2)
N(4)–M(2)–S(2)
N(5)–M(2)–S(2)
N(5)–M(2)–S(1)
N(6)–M(2)–S(1)
N–M–Sa
88.62(5)
102.14(5)
97.42(5)
87.87(5)
87.67(5)
99.48(5)
97.55(5)
87.56(5)
93.54(5)
5.61(5)
88.72(8)
101.93(9)
99.39(9)
89.14(8)
88.19(9)
100.49(9)
98.99(9)
88.36(9)
94.40(9)
5.79(9)
90.17(8)
100.32(9)
98.89(8)
89.51(8)
90.16(7)
98.27(8)
99.28(8)
89.31(7)
94.49(8)
4.78(8)
91.0(6)
89.74(6)
99.90(7)
98.41(6)
89.06(6)
90.19(5)
97.84(6)
98.88(6)
88.59(5)
94.08(6)
4.72(6)
98.52(7)
98.09(7)
90.60(6)
91.43(6)
97.38(6)
98.43(6)
90.25(6)
94.46(6)
4.46(6)
devb
O(1)–M(1)–N(1)
O(1)–M(1)–N(3)
O(2)–M(2)–N(4)
O(2)–M(2)–N(6)
O–M–Na
86.75(6)
88.79(6)
88.67(6)
87.04(5)
87.81(6)
2.19(6)
85.90(10)
87.75(11)
88.38(11)
86.16(11)
87.05(11)
2.95(10)
86.42(10)
87.24(10)
87.71(9)
86.54(9)
86.98(10)
3.02(10)
87.69(8)
87.77(8)
87.79(8)
88.11(7)
87.84(8)
2.16(8)
86.11(7)
86.76(8)
87.31(7)
86.46(7)
86.66(7)
3.34(7)
devb
O(1)–M(1)–S(1)
O(1)–M(1)–S(2)
O(2)–M(2)–S(1)
O(2)–M(2)–S(2)
O–M–Sa
95.47(5)
102.63(5)
100.40(5)
99.38(4)
99.47(5)
9.47(5)
94.14(8)
100.14(8)
97.14(8)
97.63(8)
97.26(8)
7.26(8)
93.61(7)
97.29(7)
94.65(7)
97.56(7)
95.78(7)
5.78(7)
93.30(5)
94.78(6)
93.55(6)
95.44(6)
94.27(6)
4.27(6)
94.81(5)
98.52(6)
95.73(6)
98.95(5)
97.00(6)
7.00(6)
devb
S(1)–M(1)–S(2)
S(1)–M(2)–S(2)
S–M–Sa
80.41(3)
80.41(5)
80.41(4)
9.59(4)
81.20(4)
81.05(6)
81.13(5)
8.87(5)
81.10(4)
80.92(4)
81.01(4)
8.99(4)
79.49(4)
79.45(3)
79.47(4)
10.53(4)
82.31(4)
82.04(3)
82.18(4)
7.82(4)
devb
O(1)–M(1)–N(2)
N(1)–M(1)–S(2)
N(3)–M(1)–S(1)
O(2)–M(2)–N(5)
N(4)–M(2)–S(1)
N(6)–M(2)–S(2)
X–M–Ya
155.22(5)
166.13(4)
168.15(4)
155.82(5)
166.02(4)
167.19(4)
163.08(5)
16.92(4)
156.35(11)
168.56(8)
170.34(7)
157.35(11)
168.46(8)
169.10(8)
165.03(8)
14.97(8)
160.07(10)
170.70(7)
170.61(8)
160.35(9)
171.00(7)
169.66(7)
167.06(8)
12.94(8)
163.89(7)
170.28(5)
170.09(5)
163.84(7)
170.86(5)
169.28(5)
168.04(6)
11.96(6)
158.88(8)
171.09(5)
171.37(5)
159.08(7)
172.00(5)
169.58(5)
167.00(6)
13.00(6)
devb
M(1)–S(1)–M(2)
M(1)–S(2)–M(2)
M–S–Ma
82.67(4)
83.20(4)
82.94(4)
7.06(4)
84.79(5)
86.06(5)
85.42(5)
4.58(5)
85.87(4)
88.37(4)
87.12(4)
2.88(4)
88.63(4)
90.66(3)
89.65(4)
0.35(4)
84.00(4)
86.05(3)
84.53(4)
5.47(4)
devb
M(1)–O(1)–C(39)
M(2)–O(2)–C(39)
C(4) · · · C(20)
Ph/Phc
134.40(12)
132.65(12)
10.183
95.4
127.7
133.6(2)
133.0(2)
10.094
94.4
135.1(2)
133.3(2)
9.418
81.0
132.7
134.8(2)
133.3(2)
9.306
80.8
134.9
133.9(2)
132.5(2)
9.404
80.0
131.2
M1L4/M2L4
130.8
d
a Average values. b Average deviation from the ideal 90 or 180◦ L–M–L bond angles of a perfect octahedron. c Angle between the normals of the planes
of the two aryl rings of H2LMe d Angle between the normals of the M(1)S(1)S(2)N(1)N(3) and M(2)S(1)S(2)N(4)N(6) planes.
.
˚
˚
˚
˚
˚
(2.327(3) A), Fe–S (2.521(1) A) and Fe–O (2.025(2) A) distances
are long and consistent with high-spin electronic configurations of
the two ferrous ions (S = 2). Low-spin ferrous complexes reveal
much shorter bond lengths.28–30 As expected, the Mn–ligand bond
distances in 2 are significantly longer than the corresponding Fe–
ligand bond-lengths in 3; the difference between the arithmetic
averages of all bond lengths (2.402 A in 2, 2.342 A in 3) amounts
˚
to 0.060 A. This is only slightly larger than the difference of the
ionic radii of MnII and FeII.31
As can be seen in Table 3, the metal–ligand bond lengths in
all [(LMe)MII2(OAc)]+ cations 2–6 clearly depend on the ionic radii
II
II
II
II
˚
˚
˚
of the metal ions (Mn : 0.97 A, Fe : 0.92 A, Co : 0.885 A, Ni :
1742 | Dalton Trans., 2006, 1738–1748
This journal is The Royal Society of Chemistry 2006
©

View Article Online
II
˚
˚
moment decreases from 7.55 lB at 295 K to 0.34 lB at 2 K. This is
characteristic of an intramolecular antiferromagnetic exchange
interaction between the two high-spin MnII ions in 2.32 The
expected maximum in the v vs. T curve occurs at 41 K.
0.83 A and Zn : 0.88 A). This dependence upon metal ion radius
appears in several other features of the structures. For instance,
there is a consistent decrease in the deviations of the bond angles
from the ideal 90 and 180◦ values upon going from 2 to 5. Thus, the
coordination polyhedra in the dinickel complex 5 are more regular
than in the dimanganese compound 2. The dependencies on the
metal ion radii are also coupled with the bonding constraints
of the macrocycle. Thus, there is a clear correlation between
the folding angle of the two N2M(l-S)2 planes and the folding
angle between the two phenyl rings. As the former angle steadily
increases from 127.7 to 134.9◦, the latter steadily decreases from
95.4 to 80.8◦. As a consequence, the M · · · M distances remain
almost equidistant in the five complexes, despite of the large
To determine the magnitude of the exchange interaction, the
v vs. T data were fitted by least squares using eqn (4), where
vdim and vmon refer to the molar susceptibilities of complex 2BPh4
and a fraction q of a paramagnetic mononuclear manganese(II)
impurity, respectively.33 The intramolecular exchange constant J
is defined for the isotropic Heisenberg–Dirac–van-Vleck (HDVV)
exchange Hamiltonian (H = −2JS1S2) (with S1 = S2 = 5/2) for
dinuclear complexes.34 The parameters were optimized to fit v as
a function of temperature and are listed in Table 5. An excellent
fit was obtained with J = −5.05 cm−1, g = 1.95 and q = 0.16%,35
with an agreement factor R of 5.94 × 10−3.36 Similar values of
J have been reported for the complex [(L2)MnII2(OMe)(OAc)]+
(HL2 = 2,6-bis[N-(2-pyridylethyl)iminomethyl]-4-methylphenol,
J = −3.8 cm−1)37 and for rat liver arginase (J = −2.0 cm−1)38
both of which contain related {MnII2(OR)2(OAc)} cores. The small
difference between the refined g value and g = 2.00 expected for
MnII is inside experimental uncertainties. Thus replacement of the
l-OR by l-SR groups does not change significantly the strength of
the antiferromagnetic interaction. To date, a magneto-structural
correlation for dithiolato-bridged MnII clusters has not appeared
in the literature.39
˚
contraction of the M–S bond lengths by ca. 0.13 A across the
series. With respect to the magnetic properties described below it
is also worth mentioning that the M(1)–O(1)–C(39) and the M(2)–
O(2)–C(39) bond angles remain almost constant at 133.5◦ whereas
the M–S–M angles involving the bridging sulfur atoms increase
steadily from 82.94(4)◦ in 2 to an almost 90◦ angle in 5.
Magnetic susceptibility measurements
We have carried out magnetic susceptibility measurements to
see whether magnetic exchange interactions are present in our
compounds. The temperature dependent susceptibility data for
powdered samples of complexes 2BPh4–5BPh4 have been mea-
sured between 2.0 and 300 K by using a SQUID magnetometer in
an applied external magnetic field of 0.1 T (for 2) or 0.2 T (for 3–5).
Fig. 5 displays the temperature dependencies of v and leff (per
dinuclear complex) for complex 2BPh4. The effective magnetic
v = vdim(1 − q) + 2vmon
q
(4)
For the diiron(II) complex 3BPh4 the effective magnetic moment
per dinuclear complex decreases from 6.44 lB at 295 K to 0.37 lB
at 2 K (Fig. 6). A plot of v vs. T exhibits a maximum at 65 K
Fig. 5 Temperature dependence of v (open squares) and leff (open circles)
for 2BPh4. The full lines represent the best theoretical fits to eqn (4). The
fit parameters are summarized in Table 5. Experimental and calculated
values are provided in the ESI.†
Fig. 6 Temperature dependence of v and leff (per dinuclear complex) for
3BPh4. The full lines represent the best theoretical fits to eqn (4). The fit
parameters are summarized in Table 5. Experimental and calculated values
are provided in the ESI.†
Table 5 Magnetic properties of complexes 2–5a
b
d
complex
J/cm−1
,
gc
|D|/cm−1
,
2
3
4
5
[(LMe)MnII2(l-OAc)]BPh4
[(LMe)FeII2(l-OAc)]BPh4
[(LMe)CoII2(l-OAc)]BPh4
[(LMe)NiII2(l-OAc)]BPh4
−5.05
1.95
2.09
2.85 (fixed)
2.25
n.d.
n.d.
73 10
37.2
−10.63
−(1.9 0.6)
+6.4
a Parameters resultant from least-squares fit to the vMT data under the spin Hamiltonians in eqn (4) or (5) (see text for details). b J = Coupling constant.
c g = g-Value. d D = Zero-field-splitting parameter.
This journal is
The Royal Society of Chemistry 2006
Dalton Trans., 2006, 1738–1748 | 1743
©

View Article Online
indicative of a significant antiferromagnetic exchange interaction
and the increase of the susceptibility values at low temperatures
implies that there is a small amount of a paramagnetic impurity.
The susceptibility data were fitted by the eqn (4), used for
complex 2BPh4 with vdim and vmon modified for two spin S = 4/2
ions.40 A reasonable fit to the spin-Hamiltonian H = −2JS1S2 was
only possible when the low temperature data (T = 2–7 K) were
not included in the fitting procedure. Under these conditions the
fitting parameters were found to be J = −10.63 cm−1, g = 2.09
and q = 0.88% (R = 6.5 × 10−3).41
orbit coupling, the spin–orbit coupling of excited states into the
ground state can be treated by a second-order perturbation. The
resulting orbital-angular momentum contribution may then be
treated phenomenologically by the zero-field splitting in the spin-
Hamiltonian approach.45 The actual energy splittings imposed
by symmetry reduction and by spin–orbit coupling are difficult
to evaluate. Thus, some controversy exists in the literature for
the applicability of the spin-Hamiltonian approach for cobalt(II)
complexes with lower then cubic coordination environments.23,46
However, a recent review summarized a large number of applica-
tions of the spin-Hamiltonian approach on cobalt(II) complexes.47
In this respect, we tried to simulate the temperature dependence
of the magnetic data of complex 4BPh4 using the appropriate
spin-Hamiltonian (5) including the isotropic HDVV exchange,
the single-ion zero-field splitting and the single-ion Zeeman
interaction by using a full-matrix diagonalization approach.48
There has been much interest in the coordination chemistry of
carboxylato-bridged diiron(II) complexes as such units occur in the
active sites of several proteins. The exchange interactions between
the two ferrous ions is generally much smaller (|J| ∼<10 cm−1)
than that for binuclear ferric systems as the metal–ligand bonds
are less covalent.42 Our observation of a weak antiferromagnetic
exchange interaction in 3BPh4 is in good agreement with the
reported trend.
Fig. 7 displays the temperature dependencies of v and leff (per
dinuclear complex) for the dinuclear cobalt(II) complex 4BPh4.
The effective magnetic moment, leff, has a value of 6.72 lB at
295 K which is significantly higher than the theoretical value of
5.48 lB for two uncoupled Si = 3/2 spins (gi = 2.00). This is
indicative of a strong orbital-angular momentum contribution
to the magnetic moment as it is typical for octahedral high-spin
cobalt(II) ions with a 4T1g ground state (in strict Oh symmetry).43,44
Decreasing the temperature leads to a steadily decrease of leff
to a value of 2.21 lB at 2 K. This behavior is characteristic for
antiferromagnetic exchange interactions between the two CoII high
spin centers with an St = 0 spin ground state and/or the effect
of the orbital-angular momentum contribution. Below 10 K, vM
exhibits a steep increase indicative to the presence of a significant
amount of a paramagnetic impurity.
2
ꢁ
2
zi
1
ˆ
H = −2JS1S2 +
[Di(S − / Si(Si + 1)) + lBgSiB]
(5)
3
i=1
Fitting the data without a zero-field splitting D resulted in a
poor reproduction of the experimental data and J = −4.3 cm−1.
Taking into account zero-field splitting improved the fitting results
tremendously. However, it turned out that a significant amount of
a paramagnetic impurity had to be taken into account. While
the sample used for the magnetic measurement was crystalline
and its elemental analysis proved its purity, we assume that the
dinuclear Co(II) complex might be partially oxidized to the mixed
valence Co(II)Co(III) dimer which has been described recently.9
This dimer contains only one paramagnetic cobalt(II) ion with
Si = 3/2 while the cobalt(III) is low spin d6 with Si = 0. In order
to reduce the number of parameters, we constrained the g-values
for the cobalt(II) species. In order to evaluate the (J, D) parameter
space, we calculated a two-dimensional projection of the relative
error surface for varying J and D (Fig. 8). A local minimum
exists with D = 0 cm−1 and J = −4.3 cm−1 which corresponds
to the bad fit without taking into account zero-field splitting (vide
supra). Instead of one global minimum the error surface exhibits
two minima corresponding to the same relative error. These two
minima belong to different signs in D. This behaviour relates to
the known phenomenon that the sign of D cannot be determined
by a powder measurement using only one magnetic field strength.
Interestingly, these minima correspond to different absolute values
of D and different J values. The two minima correspond to the
combinations D = −82 cm−1 with J = −1.3 cm−1 and D =
+63 cm−1 with J = −2.5 cm−1. The fitting result for the later
Fig. 7 Temperature dependence of the effective magnetic moment, leff,
of 4BPh4 at 0.2 T. The solid line is a fit to the experimental data using the
spin-Hamiltonian (5) with J = −2.51 cm−1, D = 63 cm−1, g = 2.85 and
14.8% of a paramagnetic impurity (S = 3/2).
The application of the spin-Hamiltonian approach is not
appropriate for metal ions with a T ground state due to the
first-order orbital-angular momentum and in-state spin–orbit
coupling.43 However, the local symmetry (approximate Cs) splits
4
the T1g into orbitally non-degenerate states (4Aꢁ + 2 × 4Aꢁꢁ).
Fig. 8 Two-dimensional contour projection of the relative error surface
for varying J and D. The surface was calculated with g = 2.85 and 14.8%
paramagnetic impurity (S = 3/2) held constant.
This splitting affects a quenching of the first-order orbital-angular
momentum. If this splitting is large as compared to the spin–
1744 | Dalton Trans., 2006, 1738–1748
This journal is
The Royal Society of Chemistry 2006
©

View Article Online
combination is shown in Fig. 7. It must be noted, that the two
fit results are indistinguishable. Thus, we are not able to predict
the correct combination. However, this analysis establishes a weak
antiferromagnetic coupling and a significant zero-field splitting as
it is typical for high-spin cobalt(II) ions.47
Fig. 9 shows the temperature dependence of leff (per dinuclear
unit) for the dinickel complex 5BPh4.49 For this complex leff
gradually increases from 4.51 lB at 295 K to a maximum of 4.91 lB
at 22 K and then decreases rapidly to 4.38 lB at 2 K. This behaviour
indicates an intramolecular ferromagnetic exchange interaction
between the two NiII ions in 5. The abrupt decrease in vMT below
22 K is presumably due to zero-field splitting of NiII. Thus, in
contrast to the above complexes, 5BPh4 exhibits a high-spin (S =
2) ground state.
ferromagnetic exchange interaction in 5 in which the Ni–(l-SR)–
Ni angles are very close to 90◦ is in good agreement with the
reported trends, despite of the different core structures.
Conclusion
The main findings of the present work can be summarized as
follows: (a) the 24-membered hexaazadithiophenolate macro-
cycle H2LMe forms an isostructural series of bioctahedral
[(LMe)MII2(OAc)]+ cations with early and late transition metal
ions (MnII, FeII, CoII, NiII and ZnII); (b) attempts to prepare
dinuclear CrIII complexes were unsuccessful, but resulted in the
formation of our first mononuclear complex of (LMe)2−. This
offers the opportunity to prepare heterodinuclear species. (c) The
binuclear complexes readily dissolve in organic solvents without
decomposition and air-oxidation. (d) Several structural features
of the [(LMe)MII2(OAc)]+ cations clearly depend on the ionic radii
of the metal ions, as for instance the consistent decrease in the
deviations of the bond angles, the folding angle of the two N2M(l-
S)2 planes and the folding angle between the two phenyl rings. (e)
The magnetic exchange interactions in all the [(LMe)MII2(OAc)]+
cations are relatively weak, indicative of a high degree of ionic
character in the metal–ligand bonds.
Experimental
General
Unless otherwise noted the preparations of the metal complexes
were carried out under an argon atmosphere by using stan-
dard Schlenk techniques. The compounds H2LMe·6HCl,55 FeCl2,56
[(LMe)CoII2(l-OAc)]BPh4 (4BPh4),9 and [(LMe)NiII2(l-OAc)]BPh4
(5BPh4)10 were prepared as described in the literature. Anhydrous
CrIICl2 was purchased from Aldrich. Melting points were deter-
mined in capillaries and are uncorrected.
Fig. 9 Temperature dependence of v and leff (per dinuclear complex) for
5BPh4. The full lines represent the best theoretical fit to eqn (5). The fit
parameters are summarized in Table 5. Experimental and calculated values
are provided in the ESI.†
A least-squares fit of the susceptibility data under the spin
Hamiltonian (5) yielded J = +6.4 cm−1, g = 2.25 and |D| =
37.2 cm−1.50 Again, the sign of D could not be determined from the
data. Nevertheless, the value of J is unambiguous and represents
an accurate measure of the magnetic coupling in this complex.51
The observed values are also in excellent agreement with those
reported for other bis(l-thiophenolato)-l-carboxylato bridged
nickel(II) complexes.49 Recently, Thompson et al. have described
a magneto-structural correlation for a series of bis(l-phenolato)-
bridged nickel complexes.52 There is a linear relationship between
the exchange parameter J and the Ni–O–Ni bond angle. The bridg-
ing angles range from 105.7 to 99.5◦ and J varies between −67.1
and −17.0 cm−1. The crossover-point occurs at an angle of 97◦.
For an angle greater than 97◦ the interaction is antiferromagnetic,
while angles less than 97◦ lead to a ferromagnetic interaction.
A similar correlation exists for face-sharing bioctahedral nickel
complexes. In this case, a ferromagnetic exchange interaction is
predicted if the Ni–X–Ni bridging angle is at 90◦, while for smaller
angles, the orthogonality of the magnetic orbitals is cancelled and
antiferromagnetic pathways become available to produce a change
of the sign of J. An antiferromagnetic exchange interaction, for
example, is observed in tris(l-chloro)- or tris(l-thiophenolato)-
bridged complexes, where the average bridging angle is below
80◦.53 The other extreme is represented by tris(l-phenolato)-
CAUTION! Perchlorate salts of transition metal complexes are
hazardous and may explode. Only small quantities should be
prepared and great care taken.
[(LMe)CrIIIH2(OAc)](ClO4)2 (1(ClO4)2). To an argon-purged
solution of H2LMe·6HCl (890 mg, 1.00 mmol) in MeOH (50 mL)
was added solid CrIICl2 (245 mg, 2.00 mmol). A solution of NEt3
(808 mg, 8.00 mmol) in MeOH (2 mL) was then added, followed
by sodium acetate (164 mg, 2.00 mmol). After stirring for 12 h,
the flask was exposed to air to give a dark green solution, to which
LiClO4·3H2O (802 mg, 5.00 mmol) was added. Upon standing for
several weeks, crystals of the title compound formed. Yield: 343 mg
(35%). IR (KBr pellet): m/cm−1 3031(w), 2955(s), 2905(s), 2866(s),
1629(s) masym(OAc), 1462(s), 1412(sh) msym(OAc), 1395(m), 1365(m),
1305(w), 1278(vw), 1264(vw), 1231(m), 1094(vs) (ClO4−), 966(m),
929(m), 884(m), 845(vw), 816(m), 787(m), 750(m), 740(w), 625(s).
UV/Vis (CH3CN, 2.0 × 10−3 M): kmax/nm (e/M−1 cm−1) 586 (222).
Elemental analysis (%) calc. for C40H69Cl2CrN6O10S2·4H2O (M =
981.04 + 72.06): C 45.62, H 7.37, N 7.98; found C 45.70, H 7.12,
N 8.38.
[(LMe)MnII2(l-OAc)]ClO4 (2ClO4). To
a
suspension of
bridged complexes, which feature wider angles at 90
hence parallel spin alignment occurs.54 Our observation of a
8◦ and
H2LMe·6HCl (890 mg, 1.00 mmol) in MeOH (50 mL) was added
a solution of Mn(OAc)2·4H2O (490 mg, 2.00 mmol) in MeOH
This journal is
The Royal Society of Chemistry 2006
Dalton Trans., 2006, 1738–1748 | 1745
©

View Article Online
(30 mL). A solution of NEt3 (808 mg, 8.00 mmol) in MeOH
(2 mL) was then added to give a pale-yellow solution. After stirring
for 2d at room temperature a solution of LiClO4·3H2O (802 mg,
5.00 mmol) in MeOH (2 mL) was added. The colourless solid
([(LMe)Mn2(l-OAc)]ClO4, (M = 937.46 g mol−1)) was isolated by
filtration, washed with cold EtOH (10 mL) and Et2O (5 mL) and
dried in vacuo. The crude product was purified by recrystallization
from a mixed MeCN–EtOH (1 : 1) solvent system. Yield: 840 mg
(90%). Mp 360 ◦C (decomposes without melting). IR (KBr pellet):
m/cm−1 3044(w), 3007(w), 2962(s), 2900(m), 2866(s), 2807(w),
1583(s) masym(OAc), 1478(w), 1459(s), 1438(w), 1422(sh) msym(OAc),
1395(m), 1365(m), 1352(w), 1314(m), 1292(w), 1264(s), 1230(m),
1202(m), 1169(w), 1154(m), 1097(vs) (ClO4−), 1077(sh), 1054(m),
1041(s), 1007(w), 982(w), 928(m), 913(m), 882(m), 820(s), 803(m),
746(m), 687(w), 653(m), 625(s), 560(m). UV/Vis (CH3CN, 2.0 ×
10−3 M): kmax/nm (e/M−1 cm−1) 435 (3.8), 482 (1.8), 615 (1.2).
CV (CH3CN, 295 K, 0.1 M NnBu4PF6, m = 100 mV s−1): E/V
vs. SCE (DEp/V) = +0.60 (0.18) (MnIIIMnII/MnIIMnII), +1.19
(0.126) (MnIIIMnIII/MnIIIMnII). The tetraphenylborate salt was
prepared by adding an excess of NaBPh4 (342 mg, 1.00 mmol) to a
stirred solution of 2·ClO4 (94 mg, 0.10 mmol) in MeOH (50 mL).
The colourless compound was isolated by filtration, recrystallized
once from a mixed EtOH–MeCN (1 : 1) solution and dried in
vacuum. Colourless crystals. Yield: 107 mg (92%). Mp 326–327 ◦C
(decomposes without melting). Elemental analysis (%) calc. for
C64H87BMn2N6O2S2 (M = 1157.23): C 66.42, H 7.58, N 7.26, S
5.54; found C 66.09, H 7.22, N 7.79, S 4.48. IR (KBr pellet): m/cm−1
3054(m), 3031(m), 3009(w), 2996(w), 2979(sh), 2963(s), 2924(w),
2901(m), 2864(s), 2837(w), 2805(w), 1579(s) masym(OAc−), 1477(w),
1457(s), 1437(s), 1424(sh) msym(OAc−), 1393(m), 1364(w), 1349(w),
1310(m), 1291(w), 1264(m), 1229(m), 1200(m), 1183(w), 1168(w),
1153(m), 1131(w), 1113(w), 1077(s), 1053(s), 1042(s), 1007(m),
998(w), 981(w), 928(m), 909(m), 883(m), 842(w), 817(s), 802(m),
745(m), 733(s), 704(vs), 653(m), 612(s), 559(m). UV/Vis (CH3CN,
2.0 × 10−3 M): kmax/nm (e/M−1 cm−1) 433 (4.2), 481 (2.0), 616 (1.1).
pale yellow compound was isolated by filtration, recrystallized
once from a mixed EtOH/MeCN (1:1) solution and dried in
vacuum. Pale yellow blocks. Yield: 95 mg (82%). Mp 317–318 ◦C
(decomposes without melting). Elemental analysis (%) calc. for
C64H87BFe2N6O2S2 (M = 1159.05): C 66.32, H 7.57, N 7.25, S 5.53;
found C 66.08, H 7.70, N 7.41, S 4.83. IR (KBr pellet): m/cm−1
3053(m), 3029(m), 2993(w), 2980(sh), 2963(s), 2898(m), 2864(s),
2803(w), 1577(s) masym(OAc), 1478(w), 1451(s), 1423(w) msym(OAc−),
1393(m), 1374(w), 1362(m), 1349(w), 1304(m), 1289(w), 1264(m),
1231(m), 1201(m), 1168(w), 1152(m), 1076(s), 1054(m), 1041(s),
1001(m), 981(w), 928(m), 912(m), 884(m), 842(w), 822(s), 816(w),
806(m), 747(m), 733(s), 705(vs), 655(m), 625(m), 612(s), 558(m).
UV/Vis (CH3CN, 2.0 × 10−3 M): kmax/nm (e/M−1 cm−1) 441 (39),
481 sh (21), 490 (22), 518 sh (10), 530 sh (10), 935 (14), 1236 (36).
[(LMe)Zn2(l-Ac)]ClO4 (6·ClO4). This compound was prepared
as described in the literature.10 The complete assignment of the 1H
NMR signals has not been reported previo◦usly and is included
1
here: H NMR (200 MHz, [D3]CH3CN, 25 C, TMS): d 7.06 (s,
4 H, ArH), 4.40 (d, 2J = 11.6 Hz, 4 H, ArCH2), 3.50 (dt, J = 13.9
and 3.4 Hz, 4 H, NCHHCH2N), 3.22 (dt, J = 13.7 and 3.5 Hz, 4 H,
NCHHCH2N), 2.86 (s, 6 H, NCH3), 2.77 (dd, J = 13.8 and 3.2 Hz,
4 H, NCH2CHHN), 2.63 (d, 2J = 11.6 Hz, 4 H, ArCH2), 2.42 (s,
12H, NCH3), 2.38 (dd, only one doublet is observed, the other
is obscured by the previous NCH3 signal, J = 14.0 (estimated)
and 3.4 Hz, 4 H, NCH2CHHN), 1.22 (s, 18 H, tBu), 0.79 (s,
3 H, CH3). 13C NMR (50 MHz, [D]CHCl3, 25 ◦C, TMS): d 174.5,
164.5 (BPh4), 145.7 (C), 142.5 (C), 136.8 (CH), 134.3 (C), 128.3
(BPh4), 126.1 (m, BPh4), 122.3 (m, BPh4), 64.2 (CH2), 59.2 (CH2),
57.9 (CH2), 49.8 (NCH3), 46.7 (NCH3), 34.4 (C), 31.8 (CH3), 22.8
(CH3). UV/Vis (CH3CN, 2.0 × 10−3 M): kmax/nm (e/M−1 cm−1)
342sh (223), 440 (170).
Crystal structure determinations
Single crystals of 2BPh4·3MeCN and 3 BPh4·2MeCN·0.5EtOH
suitable for X-ray structure analysis were grown by recrystalliza-
tion from a mixed MeCN–EtOH (1 : 1) solvent system. Crystals
of 1(ClO4)2 were taken directly from the mother liquor. The
crystals were mounted on glass fibers using perfluoropolyether oil.
Intensity data were collected at 210(2) K, using a Bruker SMART
CCD diffractometer. Graphite-monochromated Mo-Karadiation
[(LMe)FeII2(l-OAc)]ClO4 (3·ClO4). To
a
suspension of
H2LMe·6HCl (890 mg, 1.00 mmol) in MeOH (50 mL) was added
a solution of FeCl2 (254 mg, 2.00 mmol) in MeOH (30 mL).
A solution of NEt3 (808 mg, 8.00 mmol) in MeOH (2 mL)
was then added to give a pale-yellow solution. After stirring for
24 h at room temperature a solution of sodium acetate (98 mg,
1.20 mmol) in MeOH (40 mL) was added. After stirring for 2 h at
ambient temperature, solid LiClO4·3H2O (802 mg, 5.00 mmol)
was added to give fine pale-yellow needles of 3·ClO4. Yield:
800 mg (85%). Mp 260–261 ◦C (decomposes without melting).
IR (KBr pellet): m/cm−1 3044(w), 3007(w), 2959(s), 2898(m),
2864(s), 2803(m), 1579(s) masym(OAc), 1478(w), 1459(s), 1438(w),
1422(sh) msym(OAc), 1394(m), 1364(m), 1350(w), 1304(m), 1292(w),
1263(s), 1231(m), 1202(m), 1168(w), 1154(m), 1100(vs) (ClO4−),
1077(sh), 1055(m), 1041(s), 1003(m), 982(w), 928(m), 912(m),
882(m), 822(s), 806(m), 750(m), 689(w), 654(m), 625(s), 560(m).
UV/Vis (CH3CN, 2.0 × 10−3 M): kmax/nm (e/M−1 cm−1) 440 (39),
481 sh (21), 491 (22), 520 sh (10), 529 sh (10), 937 (14), 1239 (36).
CV (CH3CN, 295 K, 0.1 M NnBu4PF6, m = 100 mV s−1): E/V vs.
SCE (DEp/V) = +0.32 (0.086) (FeIIIFeII/FeIIFeII), +1.00 (0.126)
(FeIIIFeIII/FeIIIFeII). The tetraphenylborate salt was prepared by
adding an excess of NaBPh4 (342 mg, 1.00 mmol) to a stirred
solution of 3·ClO4 (94 mg, 0.10 mmol) in MeOH (50 mL). The
˚
(k = 0.71073 A) was used throughout. The data were processed
with SAINT and corrected for absorption using SADABS57
(transmission factors: 1.00–0.82 for 2, 1.00–0.84 for 2, 1.00–0.80
for 3). The structures were solved by direct methods using the
program SHELXS-8658 and refined by full-matrix least-squares
techniques against F2 using ShelXL-97.59 The ShelXTL version
5.10 program package was used for the structure solutions and
refinements.60 PLATON was used to search for higher symmetry.61
Ortep3 was used for the artwork of the structures.62 Unless
otherwise noted the anisotropic thermal parameters of all non-
hydrogen atoms were refined. Hydrogen atoms were included in
calculated positions with their bonding distances and thermal
parameters 1.2 times (1.5 times for CH3 groups) the thermal
parameter of the atoms to which they were attached. Selected
details of the data collection and refinement are given in Table 2.
In the crystal structure of [(LMe)CrIIIH2(OAc)](ClO4)2 (1(ClO4)2)
one perchlorate anion was found to be disordered over two
positions. Two split positions were successfully refined using the
1746 | Dalton Trans., 2006, 1738–1748
This journal is
The Royal Society of Chemistry 2006
©

View Article Online
SADI instruction (equal Cl–O and O · · · O distances) implemented
in the ShelXL program to give the following site occupancies:
Cl(2)O(7a)–O(10a)/Cl(2)O(7b)–O(10b) = 0.63(2)/0.37(2). The
hydrogen atoms H(6) and H(5) were located from the final
difference map and were refined isotropically with Ueq(H) 1.2 times
that of the nitrogen atom to which they are attached.
CCDC reference numbers 281645 (1(ClO4)2), 281646
(2BPh4·3MeCN) and 281647 (3BPh4·2MeCN·0.5EtOH).
For crystallographic data in CIF or other electronic format see
DOI: 10.1039/b513717a
7 A. L. Gavrilova, C. Jin Qin, R. D. Sommer, A. L. Rheingold and B.
Bosnich, J. Am. Chem. Soc., 2002, 124, 1714–1722.
8 (a) S. Brooker, Coord. Chem. Rev., 2001, 222, 33–56; (b) S. Brooker,
P. D. Croucher, T. C. Davidson, G. S. Dunbar, C. U. Beck and S.
Subramanian, Eur. J. Inorg. Chem., 2000, 169–179.
9 B. Kersting and G. Steinfeld, Inorg. Chem., 2002, 41, 1140–1150.
10 B. Kersting, Angew. Chem., 2001, 113, 4109–4112; B. Kersting, Angew.
Chem., Int. Ed., 2001, 40, 3987–3990.
11 J. Hausmann, S. Ka¨ss, S. Klod, E. Kleinpeter and B. Kersting,
Eur. J. Inorg. Chem., 2004, 4402–4411.
12 V. Lozan and B. Kersting, Eur. J. Inorg. Chem., 2005, 504–512.
13 (a) T. Beissel, T. Glaser, F. Kesting, K. Wieghardt and B. Nuber, Inorg.
Chem., 1996, 35, 3936–3947; (b) T. Glaser, F. Kesting, T. Beissel, E. Bill,
T. Weyhermu¨ller, W. Meyer-Klaucke and K. Wieghardt, Inorg. Chem.,
1999, 38, 722–732.
14 K. Nakamoto, Infrared and Raman Spectra of Inorganic and Coordina-
tion Compounds, VCH-Wiley, New York, 5th edn, 1997.
15 N. D. J. Branscombe, A. J. Atkins, A. Marin-Becerra, E. J. L. McInnes,
F. E. Mabbs, J. McMaster and M. Schro¨der, Chem. Commun., 2003,
1098–1099.
Physical measurements
Elemental analyses were performed on a Vario EL analyzer
(Elementaranalysensysteme GmbH). IR spectra were taken on
a Bruker VECTOR 22 FT-IR-spectrophotometer as KBr pellets.
Cyclic voltammetry measurements were carried out at 25 ◦C with
an EG & G Princeton Applied Research potentiostat/galvanostat
model 263 A. The cell contained a Pt working electrode, a Pt
wire auxiliary electrode and an Ag wire as reference electrode.
Concentrations of solutions were 0.10 M in supporting electrolyte
NnBu4PF6 and ca. 1.0 × 10−3 M in sample. Cobaltocenium hex-
afluorophosphate (Cp2CoPF6) was used as an internal standard.
All potentials were converted to the SCE reference using tabulated
values.63 Temperature-dependent magnetic susceptibility measure-
ments on powdered solid samples were carried out on a SQUID
magnetometer (MPMS Quantum Design) over the temperature
range 2.0–300 K. The magnetic field applied was 0.1 (for 2) or 0.2
Tesla (for 3–5). The observed susceptibility data were corrected
for the underlying diamagnetism by using Pascal’s constants.
16 Experiments are underway to prepare heterodinuclear complexes of
the type [(LMe)CoIIZnII(OAc)]n+. We believe that the two-electron
•
oxidized [(LMe )CoIIIZnII]3+ form, in which a thiyl-radical is stabilized by
coordination to a kinetically inert CoIII ion, is stable enough to allow its
isolation and characterization. The results of these investigations will
be reported in due course.
17 R. C. Elder, L. R. Florian, R. E. Lake and A. M. Yacynych, Inorg.
Chem., 1973, 12, 2690–2699.
18 (a) S. Merlino and F. Sartori, Acta Crystallogr., Sect. B, 1972, 28,
972–976; (b) C. L. Raston and A. H. White, Aust. J. Chem., 1977, 30,
2091–2094.
19 J. I. Bruce, L. R. Gahan, T. W. Hambley and R. Stranger, J. Chem.
Soc., Chem. Commun., 1993, 702–704.
20 K. Shiren and K. Tanaka, Inorg. Chem., 2002, 41, 5912–5919.
21 D. A. House, Comprehensive Coordination Chemistry, ed. G. Wilkinson,
R. D. Gillard and J. A. McCleverty, 1996, vol. 2, pp. 23–72.
22 T. Nagata and J. Mizukami, J. Chem. Soc., Dalton Trans., 1995, 2825–
2830.
23 U. Bossek, D. Nu¨hlen, E. Bill, T. Glaser, C. Krebs, T. Weyhermu¨ller, K.
Wieghardt, M. Lengen and A. X. Trautwein, Inorg. Chem., 1997, 36,
2834–2843.
24 I. Romero, L. Dubois, M.-N. Collomb, A. Deronzier, J.-M. Latour and
J. Pecaut, Inorg. Chem., 2002, 41, 1795–1806.
Acknowledgements
We are particularly grateful to Prof. Dr H. Vahrenkamp for provid-
ing facilities for NMR and X-ray crystallographic measurements.
Financial support of this work from the Deutsche Forschungs-
gemeinschaft (Priority programme “Molecular Magnetism”, KE
585/4-1,2) is gratefully acknowledged.
25 Abbreviations:
bpea
=
N,N-bis(2-pyridylmethyl)ethylamine,
Me3tacn = N,Nꢁ,Nꢁꢁ-trimethyl-1,4,7-triazacyclononane).
26 K. Wieghardt, U. Bossek, B. Nuber, J. Weiss, J. Bonvoisin, M. Corbella,
S. E. Vitols and J.-J. Girerd, J. Am. Chem. Soc., 1988, 110, 7398–
7411.
27 K. Wieghardt, U. Bossek, B. Nuber and J. Weiss, Inorg. Chim. Acta,
1987, 126, 39–43.
28 V. A. Grillo, L. R. Gahan, G. R. Hanson, R. Stranger, T. W. Hambley,
K. S. Murray, B. Moubaraki and J. D. Cashion, J. Chem. Soc., Dalton
Trans., 1998, 2341–2348.
29 M. Li, D. Bonnet, E. Bill, F. Neese, T. Weyhermu¨ller, N. Blum, D.
Sellmann and K. Wieghardt, Inorg. Chem., 2002, 41, 3444–3456.
30 U. Heinzel, A. Henke and R. Mattes, J. Chem. Soc., Dalton Trans.,
1997, 501–508.
31 (a) R. D. Shannon and C. T. Prewitt, Acta Crystallogr., Sect. B, 1969,
25, 925–946; (b) R. D. Shannon, Acta Crystallogr., Sect. A, 1976, 32,
751–767.
References
1 (a) A. J. Atkins, A. J. Blake and M. Schro¨der, J. Chem. Soc., Chem.
Commun., 1993, 1662–1665; (b) N. D. J. Branscombe, A. J. Blake, A.
Marin-Becerra, W.-S. Li, S. Parsons, L. Ruiz-Ramirez and M. Schro¨der,
Chem. Commun., 1996, 2573–2574.
2 (a) S. Brooker and P. D. Croucher, J. Chem. Soc., Chem. Commun.,
1995, 1493–1494; (b) S. Brooker and P. D. Croucher, J. Chem. Soc.,
Chem. Commun., 1995, 2075–2076; (c) S. Brooker, P. D. Croucher and
F. M. Roxburgh, J. Chem. Soc., Dalton Trans., 1996, 3031–3037; (d) S.
Brooker and P. D. Croucher, Chem. Commun., 1997, 459–460.
3 N. H. Pilkington and R. Robson, Aust. J. Chem., 1970, 23, 2225–2236.
4 (a) P. A. Vigato, S. Tamburini and D. Fenton, Coord. Chem. Rev., 1990,
106, 25–170; (b) D. Fenton, Chem. Soc. Rev., 1999, 28, 159–168; (c) M.
Yamami, H. Furutachi, T. Yokoyama and H. Okawa, Inorg. Chem.,
1998, 37, 6832–6838; (d) H. Okawa, H. Furutachi and D. E. Fenton,
Coord. Chem. Rev., 1998, 174, 51–75; (e) M. Shinoura, S. Kita, M.
Ohba, H. Okawa, H. Furutachi and M. Suzuki, Inorg. Chem., 2000, 39,
4520–4526.
5 A. J. Atkins, D. Black, A. J. Blake, A. Marin-Becerra, S. Parsons, L.
Ruiz-Ramirez and M. Schro¨der, Chem. Commun., 1996, 457–464.
6 (a) S. Brooker, P. D. Croucher, T. C. Davidson, G. S. Dunbar, A. J.
McQuillan and G. B. Jameson, Chem. Commun., 1998, 2131–2132;
(b) D. J. E. Spencer, A. C. Marr and M. Schro¨der, Coord. Chem. Rev.,
2001, 219–221, 1055–1074.
32 The theoretical values of vmT and leff for two non-interacting Mn2+ (S =
5/2), Fe2+ (S = 2), Co2+ (S = 3/2) and Ni2+ (S = 1) ions are calculated to
be 8.753 cm3 K mol−1 (8.367 lB), 6.93 cm3 K mol−1 (6.00 lB), 3.75 cm3 K
mol−1 (5.48 lB) and 2.00 cm3 K mol−1 (4.00 lB), respectively, for g = 2.
33 For S1 = S2 = 5/2, vdim = (Ng2lB2/(kT)) [2 exp(2J/kT) + 10 exp(6J/
kT) + 28 exp(12J/kT) +60 exp(20J/kT) + 110 exp(30J/kT)]/[1 +
3 exp(2J/kT) + 5 exp(6J/kT) + 7 exp(12J/kT) + 9 exp(20J/kT) +
11 exp(30J/kT)], see ref. 34; vmon = (Ng2lB2/(3kT))(5/2(5/2 + 1)).
34 C. J. O’Connor, Prog. Inorg. Chem., 1982, 29, 203–283.
35 A least-squares fit of eqn (4) to the high-temperature data (T > 60 K)
yielded values of J = −5.12 cm−1 and g = 1.96 when the presence of the
paramagnetic impurity was neglected q = 0 (fixed)), R = 2.9 × 10−3
.
ꢀ
ꢀ
36 R = |(vdim,exp − vdim,calc)|/ | vdim,exp|.
37 M. Mikuriya, Y. Kawasaki, T. Tokii, S. Yanai and A. Kawamori, Inorg.
Chim. Acta, 1989, 156, 21–22.
This journal is
The Royal Society of Chemistry 2006
Dalton Trans., 2006, 1738–1748 | 1747
©

View Article Online
38 Z. F. Kanyo, L. R. Scolnick, D. E. Ash and D. W. Christianson, Nature,
1996, 383, 554–557.
50 In our earlier work (see ref. 49), we have refined the susceptibility data
with the g-value fixed at 2.21 (J and D were found to be +7.9 and
−37.7 cm−1). In the present work the refinements were carried out with
unconstrained variables. This explains the small discrepancies between
the earlier and the present values.
39 S. V. Khangulov, P. J. Pessiki, V. V. Barynin, D. E. Ash and G. C.
Dismukes, Biochemistry, 1995, 34, 2015–2025.
40 For S1 = S2 = 4/2, vdim = (Ng2lB2/(kT))[2 exp(2J/kT)+ 10 exp(6J/
kT) + 28 exp(12J/kT) + 60 exp(20J/kT)]/[1 + 3 exp(2J/kT) +
51 D. M. Duggan, E. K. Barefield and D. N. Hendrickson, Inorg. Chem.,
5 exp(6J/kT) + 7 exp(12J/kT) + 9 exp(20J/kT)], see ref. 34. vmon
(Ng2lB2/(3kT))(4/2(4/2 + 1)).
=
1973, 12, 985–991.
52 (a) K. K. Nanda, L. K. Thompson, J. N. Bridson and K. Nag, J. Chem.
Soc., Chem. Commun., 1994, 1337–1338; (b) K. K. Nanda, R. Das,
L. K. Thompson, K. Venatsubramanian, P. Paul and K. Nag, Inorg.
Chem., 1994, 33, 1188–1193.
41 By using the spin-Hamiltonian eq 5 (with D, g and J being the variables)
a fit is possible over the full temperature range. In this case the least-
squares fit yielded values of J = −10.52 cm−1, g = 2.09 and D =
−0.007 cm−1 (R = 9.2 × 10−3).
53 T. Beissel, F. Birkelbach, E. Bill, T. Glaser, F. Kesting, C. Krebs, T.
Weyhermu¨ller, K. Wieghardt, C. Butzlaff and A. X. Trautwein, J. Am.
Chem. Soc., 1996, 118, 12376–12390.
54 Y. Elerman, M. Kabak, I. Svoboda, H. Fuess, K. Griesar and W. Haase,
Z. Naturforsch., B, 1996, 51, 1132–1136.
42 E. I. Solomon, T. C. Brunold, M. I. Davis, J. N. Kemsley, S.-K. Lee, N.
Lehnert, F. Neese, A. J. Skulan, Y.-S. Lang and J. Zhou, Chem. Rev.,
2000, 100, 235–349.
43 (a) C. J. Ballhausen, Introduction to Ligand Field Theory, McGraw-Hill,
New York, 1962; (b) A. T. Casey and S. Mitra, in Theory and Application
of Molecular Paramagnetism, ed. E. A. Boudreaux and L. N. Mulay,
Wiley-Interscience, New York, 1976, pp. 135–255; (c) B. N. Figgis,
Introduction to Ligand Fields, Interscience Publishers, New York, 1964.
44 L. H. Gade, Koordinationschemie, Wiley-VCH, Weinheim, 1998, p. 313.
45 F. Neese and E. I. Solomon, in Magnetism: Molecules to Materials, ed.
J. S. Miller and M. Drillon, Wiley-VCH, Weinheim, 2003, Germany,
vol. IV, pp. 345–466.
55 (a) M. H. Klingele, G. Steinfeld and B. Kersting, Z. Naturforsch.,
B, 2001, 56, 901–907; (b) G. Siedle and B. Kersting, Z. Anorg. Allg.
Chem., 2003, 629, 2083–2090; (c) B. Kersting and G. Steinfeld, Chem.
Commun., 2001, 1376–1377.
56 P. Kovacic and M. O. Brace, J. Am. Chem. Soc., 1954, 76, 5491–
5494.
57 SADABS, An Empirical Absorption Correction Program Part ot the
SAINTPlus NT version 5.10 package, BRUKER AXS, Madison, WI,
1998.
58 G. M. Sheldrick, Acta Crystallogr., Sect. A, 1990, 46, 467–473.
59 G. M. Sheldrick, SHELXL-97, Computer Program for Crystal Structure
Refinement, University of Go¨ttingen, Go¨ttingen, Germany, 1997.
60 ShelXTL version 5.10, Bruker AXS, Madison, WI, 1998.
61 A. L. Spek, PLATON-A Multipurpose Crystallographic Tool, Utrecht
University, Utrecht, The Netherlands, 2000.
46 K. Fink, C. Wang and V. Staemmler, Inorg. Chem., 1999, 38, 3847–3856.
47 R. Boca, Coord. Chem. Rev., 2004, 248, 757–815.
48 The routine JULIUS was used for spin Hamiltonian simulations of the
data (C. Krebs, E. Bill, F. Birkelbach and V. Staemmler, unpublished
results). For the calculation of the magnetic susceptibiliy of the
paramagnetic impurity JULIUS uses the equation vM = vM,dim(1 −
q) + vM,monq.
49 The magnetic properties of this complex have been reported previously:
J. Hausmann, M. H. Klingele, V. Lozan, G. Steinfeld, D. Siebert, Y.
Journaux, J. J. Girerd and B. Kersting, Chem. Eur. J., 2004, 10, 1716–
1728.
62 L. J. Farrugia, J. Appl. Crystallogr., 1997, 30, 565.
63 Under our experimental conditions E(Cp2Co+/Cp2Co) = −1.345 V
vs. E(Cp2Fe+/Cp2Fe); E(Cp2Co+/Cp2Co) = −0.937 V vs. SCE: N. G.
Connelly and W. E. Geiger, Chem. Rev., 1996, 96, 877–910.
1748 | Dalton Trans., 2006, 1738–1748
This journal is
The Royal Society of Chemistry 2006
©