Tuning the catalase activity of dinuclear manganese complexes by utilizing different substituted tripodal ligands

Article abstract of DOI:10.1002/ejic.200300157

A series of five new dinuclear manganese (II/II) compounds with derivatives of 2-{[bis(pyridin-2-ylmethyl)amino]-methyl}phenol (HL₁) were synthesized and structurally characterized. All complexes crystallize in monoclinic space groups and exhib

Full text of DOI:10.1002/ejic.200300157

FULL PAPER
Tuning the Catalase Activity of Dinuclear Manganese Complexes by Utilizing
Different Substituted Tripodal Ligands
Nicole Reddig,[a] Daniel Pursche,[a] Michael Kloskowski,[a] Caroline Slinn,[a]
Sascha M. Baldeau,[a] and Annette Rompel*[a]
Keywords: Manganese / Tripodal ligands / Bioinorganic chemistry / Enzyme models
A series of five new dinuclear manganese(II
with derivatives of 2-{[bis(pyridin-2-ylmethyl)amino]-
/
II) compounds
redox potentials of the dinuclear transition metal complexes
and the catalase activity has been revealed. In analogy to
methyl}phenol (HL1) were synthesized and structurally char- manganese catalase, all complexes show saturation kinetics
acterized. All complexes crystallize in monoclinic space
at high substrate concentration. The investigated catalase ac-
tivity places the compounds in the upper range of functional
catalase mimics.
groups and exhibit Mn···Mn separations in the range of
˚
3.392(8)−3.493(2)
A
caused by bis(µ-phenoxo) bridging
modes. The derivatives of HL1 contain either electron-donat-
ing or -withdrawing substituents. A correlation between the
electronic character of the different ring substituents, the
( Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim,
Germany, 2004)
Introduction
catalase mimic.[13Ϫ15] At the moment the most efficient cat-
alase mimic is based on the ligand bpia [bis(picolyl)-
(N-methylimidazol-2-yl)amine], which belongs to the class
of tripodal ligands. The [Mn2(bpia)2] system consists of
four dinuclear complexes, that mimic structural, spectro-
scopic and functional features of manganese catalases.
[Mn2(bpia)2(µ-OAc)2](ClO4)2 and [Mn2(bpia)2(µ-O)2](ClO4)2-
(PF6)·2CH3CN are the most efficient catalysts using the
Mn2(/) and Mn2(/) redox couple reported to date for
the disproportion of hydrogen peroxide.[16]
In this paper we report the synthesis, characterization
and catalase activity of five new dinuclear Mn2(/) com-
plexes. They are based on tripodal ligands featuring an N3O
donor set, where the parent ligand is 2-({bis[(pyridin-2-yl)-
methyl]amino}methyl)phenol. The phenol moiety of the
ligands leads to a bis(µ-phenoxo) bridging motif, leaving
one vacant coordination site at each manganese center.
These are occupied by two mutually trans-positioned chlo-
ride ions. The catalytic activity regarding the dispro-
portionation of hydrogen peroxide depends on the ring sub-
stituents. It will be shown that electron-withdrawing groups
decrease the ability to disproportionate hydrogen peroxide,
whereas electron-donating groups increase this ability.
The catalase activities of the model compounds described
in this paper range between those of [Mn2(bpia)2] and
[Mn2(2-OHsalpn)2] and place the five dinuclear manga-
nese(/) complexes in second position for functional
manganese catalase mimics.
The manganese catalases are the most intensively studied
dinuclear manganese enzymes. By disproportionating hy-
drogen peroxide into water and dioxygen, manganese catal-
ases protect cells from deleterious effects caused by this
harmful by-product of respiration. In addition to heme-
type catalases,[1] there is another class of manganese catal-
ases that has been found in three bacterial organisms,
Lactobacillus plantarium,[2,3] Thermus thermophilus,[4,5] and
Thermoleophilum album.[6] The manganese catalase from
Thermus thermophilus is the best characterized one. During
the catalytic cycle the oxidation states of the dinuclear
manganese center change between Mn2(/) and Mn2(/
). There are also two inactive forms [Mn2(/) and
Mn2(/)], which are caused by one-electron transitions
during the disproportionation of hydrogen peroxide.[7,8]
So far only three functional model systems for the catal-
ase reaction have been described which employ Mn2(/)
and Mn2(/) in their catalytic cycles. The first system,
[Mn2(L1)(µ-OAc)]2ϩ, is based on a heptadentate ligand
L1 [N,N,NЈ,NЈ-tetrakis(2-methylenebenzimidazolyl)-1,3-
diaminopropan-2-ol] and was the first functional catalase
mimic.[9Ϫ12] The second model system is the [Mn2(2-
OHsalpn)2] [2-OHsalpn ϭ 1,3-bis(salicylideneamino)-2-pro-
panol] system. It has been fully crystallographically and spec-
troscopically characterized in four different oxidation states
and serves as both a structural and functional manganese
[a]
Institut für Anorganische und Analytische Chemie,
Results and Discussion
Westfälische Wilhelms-Universität,
Wilhelm-Klemm-Str. 8, 48149 Münster, Germany
Fax: (internat.) ϩ 49-251/833-8366
E-mail: rompela@uni-muenster.de
Reaction of manganese() chloride tetrahydrate with the
corresponding ligand under basic conditions yields the di-
Eur. J. Inorg. Chem. 2004, 879Ϫ887
DOI: 10.1002/ejic.200300157
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
879

N. Reddig, D. Pursche, M. Kloskowski, C. Slinn, S. M. Baldeau, A. Rompel
FULL PAPER
nuclear manganese complexes [Mn2(L1)2Cl2]·2CH2Cl2 (1),
[Mn2(L2)2Cl2]·C2H5OH (2), [Mn2(L3)2Cl2]·2CH2Cl2 (3),
[Mn(L4)2Cl2]·2CH2Cl2 (4) and [Mn2(L5)2Cl2] (5). All com-
pounds crystallize in the monoclinic crystal system. Com-
plexes 1Ϫ3 crystallize in the space group P21/c (no.14),
while 4 and 5 crystallize in I2/a (no.15). Each unit cell con-
sists of four complex molecules, except for compound 3,
where only two complex molecules can be found. Either
eight (1), four (2, 3, and 4), or no (5) solvent molecules
complete the unit cells. The manganese ions are in the oxi-
dation state ϩ. They are bis(µ-phenoxo)-bridged by the
ligands’ phenol groups resulting in Mn···Mn-distances in
˚
˚
the range from 3.392(8) A (5) to 3.493(2) A (4). The dis-
torted octahedral coordination environment of each
manganese center is formed by an N3O2Cl donor set, where,
except for the chloride ions, all donor atoms are provided
by the ligands. The crystal structure of the metal complexes
are depicted below. Selected bonds and angles for all com-
plexes are listed in Table 1.
Figure 1. Ellipsoid plot of 1 showing 50% probability thermal ellip-
soids; hydrogen atoms are omitted for clarity
range as for Mn(1). The average N(1)ϪMn(1)ϪN(X) (X ϭ
2, 3) and N(4)ϪMn(2)ϪN(Y) (Y ϭ 5, 6) angles of 72.7 and
72.3° differ significantly from 90° and show the distortion
from octahedral geometry due to constraints imposed by
the ligand.
˚
Table 1. Selected distances [A] and angles [°] of 1Ϫ5
1[a]
2[a]
3
4
5
Mn(1)···Mn(1a)[b]
Mn(1)ϪCl(1)
Mn(1)ϪO(1)
Mn(1)ϪO(1a)[b]
Mn(1)ϪN(1)
Mn(1)ϪN(2)
Mn(1)ϪN(3)
3.411(1)
2.451(2)
2.119(4)
2.203(4)
2.367(5)
2.348(5)
2.254(5)
3.440(1)
2.426(1)
2.138(2)
2.199(2)
2.360(3)
2.263(3)
2.288(3)
3.406(2)
2.472(3)
2.147(3)
2.173(3)
2.330(4)
2.399(4)
2.338(4)
3.493(2)
2.425(9)
2.156(2)
2.205(2)
2.305(2)
2.313(2)
2.361(2)
3.392(8)
2.454(1)
2.138(2)
2.142(2)
2.372(2)
2.273(2)
2.312(2)
Mn(1)ϪO(1)ϪMn(1a)[b] 105.2(2)
Cl(1)ϪMn(1)ϪO(1) 111.2(1)
Cl(1)ϪMn(1)ϪO(1a)[b] 105.2(1)
105.0(9)
104.1(1)
117.5(1)
97.0(1) 101.3(5)
106.4(7)
118.6(6)
104.8(7)
111.6(7)
101.5(7)
157.1(7)
98.2(9)
90.5(8)
75.0(9)
84.4(9)
114.7(5)
101.5(6)
152.9(7)
93.2(6)
88.5(8)
75.2(7)
84.4(7)
150.6(8)
87.8(7)
102.0(8)
90.8(7)
162.6(7)
73.2(8)
72.5(9)
102.9(8)
Cl(1)ϪMn(1)ϪN(1)
Cl(1)ϪMn(1)ϪN(2)
Cl(1)ϪMn(1)ϪN(3)
O(1)ϪMn(1)ϪO(1a)[b]
O(1)ϪMn(1)ϪN(1)
O(1)ϪMn(1)ϪN(2)
O(1)ϪMn(1)ϪN(3)
O(1a)ϪMn(1)ϪN(1)[b]
155.7(1)
88.1(1)
96.0(1)
76.0(1)
85.4(2)
154.9(1)
96.3(1)
90.6(2)
75.9(1)
85.0(1)
154.2(6)
89.1(7)
96.6(6)
73.6(7)
84.9(6)
92.9(2) 146.9(1)
88.5(1) 147.4(7)
151.1(2)
95.8(2)
92.2(1) 148.5(2) 86.8(7)
98.4(9)
85.5(1) 163.0(1)
87.9(2) 165.0(1)
99.5(1)
95.3(6)
85.0(7)
O(1a)ϪMn(1)ϪN(2)[b] 165.0(2)
O(1a)ϪMn(1)ϪN(3)[b]
N(1)ϪMn(1)ϪN(2)
N(1)ϪMn(1)ϪN(3)
N(2)ϪMn(1)ϪN(3)
87.4(2) 158.1(6)
73.0(2)
72.3(2)
72.1(1)
72.1(1)
71.9(1)
71.4(2)
72.8(7)
72.9(7)
97.8(2) 101.9(1)
103.1(2)
108.0(7)
Figure 2. Ellipsoid plot of 2 showing 50% probability thermal ellip-
soids; hydrogen atoms are omitted for clarity
[a]
Bond lengths and angles for Mn(2) are omitted for clarity.
[b] Mn(1a) ϭ Mn(2) and O(1a) ϭ O(2) for 1.
[Mn2(L2)2(Cl)2]·C2H5OH (2; Figure 2): In 1987 an anal-
ogous structure of complex 2 was published by Nishida
[Mn2(L1)2(Cl)2]·2CH2Cl2 (1; Figure 1): Mononuclear with NCS ligands instead of chloride ligands (P21/n).[24] In
Cu,[17] Fe,[18] Re,[19] and Zn[20] complexes of HL1 are addition, mono- and dinuclear copper complexes[23,25Ϫ27]
known. In addition, dinuclear Fe,[21,22] Cu,[23] and Zn[20] and mononuclear iron complexes[18,28] of HL2 have been
complexes have also been published.
described in the literature.
The MnϪN bond lengths around Mn(1) range from
There are two half complex molecules and one ethanol
˚
2.254(5) for Mn(1)ϪN(3) to 2.367(5) A for Mn(1)ϪN(1). molecule in the asymmetric unit. Both manganese() cen-
The distances for the Mn(1)ϪO bonds are 2.119(4) [O(1)] ters of each independent complex molecule are related by a
˚
and 2.203(4) A [O(2)]. The longest bond to be found is crystallographic inversion center. The coordination environ-
˚
Mn(1)ϪCl(1) [2.451(2) A]. Except for small deviations, the ment of Mn(1) is described as it is representative of both
manganeseϪdonor atom bonds for Mn(2) are in the same Mn(1) and Mn(2). The longest bond to be found is
880
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
Eur. J. Inorg. Chem. 2004, 879Ϫ887

Tuning the Catalase Activity of Dinuclear Manganese Complexes
FULL PAPER
˚
Mn(1)ϪCl(1) [2.426(1) A]. The chloride ion is bonded trans
to the aliphatic nitrogen atom N(1), with a bond length of
˚
2.360(3) A. This is considerably longer than the
˚
manganeseϪpyridine distances Mn(1)ϪN(2) [2.263(3) A]
˚
and Mn(1)ϪN(3) [2.288(3) A]. The average distance for the
˚
MnϪO bonds is 2.168 A.
The five-membered chelate rings containing Mn(1), N(1),
N(2/3) cause a distortion from an ideal octahedral ge-
ometry. Hence, for the average N(1)ϪMn(1)ϪN(X) (X ϭ 2,
3) angles a deviation of 17.8° from the ideal 90° angle is
observed. The distortion is shown best in the angles includ-
ing trans-bonded donor atoms. The largest difference from
linearity occurs in the angles O(1)ϪMn(1)ϪN(2)
[146.9(1)°].
Figure 3. Ellipsoid plot of 3 showing 50% probability thermal ellip-
soids; hydrogen atoms and a depiction of the chloride disorder are
omitted for clarity
[Mn2(L3)2(Cl)2]·2 CH2Cl2 (3; Figure 3): There is a crystal-
lographic inversion center found in the middle of the di-
nuclear complex, rendering the two halves of the molecule
crystallographically equivalent. The chloride ion is dis-
ordered. The occupation factors for both forms were re-
Figure 4. Ellipsoid plots of 4 (top) and 5 (bottom) showing 50%
probability thermal ellipsoids; hydrogen atoms and a depiction of
the chloride disorder are omitted for clarity
fined to 0.71 and 0.29. The manganeseϪchlorine bonds
˚
show an average value of 2.407 A. The aliphatic nitrogen positions have been refined to 0.84 and 0.16 for complex 4
atom N(1) is bonded trans to the chloride ion, with a bond and 0.95 and 0.05 for 5. In addition, compound 4 shows a
˚
length of 2.330(4) A. This is shorter than the disorder of the methyl substituent, where the occupation
manganeseϪpyridine distances. The bond Mn(1)ϪN(2) factors for both positions have been refined to 0.80 and
˚
[2.399(4) A] is the longest manganeseϪnitrogen bond in the 0.20.
complex, which can be attributed to the steric effects of the
methyl substituent in the ortho position of N(2). The both compounds and show values of 2.425(9) A (4) and
MnϪO bond lengths are in the same range. The average 2.454(1) A (5). The aliphatic nitrogen atom N(1) is bonded
N(1)ϪMn(1)ϪN(X) (X ϭ 2, 3) angle of 71.7° differs signifi- trans to the chloride ion with distances for 4 of 2.305(2) A
cantly from 90° and shows the distortion from octahedral and for 5 of 2.372(2) A to Mn(1). As expected, in 5 both
The Mn(1)ϪCl(1) bonds are the longest to be found in
˚
˚
˚
˚
geometry due to constraints imposed by the ligand.
[Mn2(L4)2(Cl)2]·2CH2Cl2 (4; Figure 4)
pyridine nitrogen atoms N(2) and N(3) exhibit shorter
˚
and bonds to the metal center with values of 2.273(2) A and
˚
[Mn2(L5)2(Cl)2] (5; Figure 4): A crystallographic inversion 2.312(2) A, respectively. In complex 4 the Mn(1)ϪN(2, 3)
center in the middle of the complexes generates the corre- distances are longer compared to Mn(1)ϪN(1) (see Table
sponding symmetric equivalent half. Mn(1) is surrounded 1). This can be attributed to the steric effects of the dis-
by an N3O2Cl donor set while the position of the chloride ordered methyl substituent in the ortho position of N(2) and
ion is slightly disordered. The occupation factors for both N(3), respectively. The two oxygen atoms O(1) and O(1a)
Eur. J. Inorg. Chem. 2004, 879Ϫ887
www.eurjic.org
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
881

N. Reddig, D. Pursche, M. Kloskowski, C. Slinn, S. M. Baldeau, A. Rompel
FULL PAPER
are bonded trans to the aromatic nitrogen atoms N(2) and
The observed redox potentials for the oxidation
N(3). Distances of 2.156(2) A (4) and 2.138(2) A (5) for MnIIMnII/MnIIMnIII are in good agreement with data re-
˚
˚
˚
˚
Mn(1)ϪO(1) and 2.205(2) A (4) and 2.142(2) A (5) for ported in the literature for similar dinuclear manganese
Mn(1)ϪO(1a) have been determined. The bis(µ-phenoxo)- compounds.[29Ϫ32]
bridging mode results in Mn···Mn separations of 3.493(2)
A for compound 4 and 3.392(8) A for complex 5. The dis- peak separation of 59 mV has to be detected. For com-
tortion from ideal octahedral symmetry is best clarified by pounds 3 and 4 the differences in peak potentials for the
For a reversible single-electron redox transition an ideal
˚
˚
comparing
the
values
for
the
two
angles first redox transition show little difference from the ideal 59
N(1)ϪMn(1)ϪN(X) (X ϭ 2, 3) and O(1)ϪMn(1)ϪN(2) mV (∆E ϭ 70 mV). This shift can be attributed to solvent
with their ideal values. The latter differs significantly from effects.[33] Therefore these waves can clearly be attributed to
the ideal 180° with angles of 147.4(7)° (4) and 150.6(8)° (5), a single-electron transfer in both complexes from MnII/
while the average angles for N(1)ϪMn(1)ϪN(X) (X ϭ 2, 3) MnII to MnII/MnIII
.
deviate by around 17.2° from the ideal 90°.
In the case of quasi-reversible electron transitions the de-
There is a correlation between the Mn···Mn distances termination of n, the number of electrons involved in the
and the type of ring substituent. In particular, the influence redox process, is more complicated. The large peak-separ-
of the electron-withdrawing effect of the NO2 groups, that ation for the first redox transition in the cyclic voltammog-
are p-bonded to the phenol oxygen atom in 2 and 4, is con- ram of compounds 1, 2, and 5 (∆E ϭ 140Ϫ160 mV) indi-
spicuous. Because of its ϪM effect the electron density of cates that the electron transfer is slow. For a system under-
the phenolic oxygen atom is reduced through the aromatic going slow electron transfers it has been shown that the
π-system. This results in a weaker MnϪO bond, leading to number of electrons can be determined using the following
larger Mn···Mn distances than in complexes 1, 3, and 5. On equation:[31,34] αn ϭ 0.048/(|Ep Ϫ Ep/2|).
the other hand electron-donating groups like methyl and
In the present case a value of 0.5 can be considered for
methoxy moieties stabilize the MnϪO bonds resulting in the charge-transfer coefficient α due to the symmetrical
shorter Mn···Mn separations than in the parent complex 1. shape of the corresponding wave. Ep is the peak potential
of the considered wave and Ep/2 represents the potential
Electrochemistry
value at half the intensity of the peak current. From the
obtained data listed in Table 3 one can ascertain that n ϭ 1
for the first redox transition.
The redox potentials of complexes 1Ϫ5 are summarized
in Table 2. Figure 5 shows the cyclic voltammogram of
compound 2 in acetonitrile as the most representative.
Table 3. Potential peak values for the MnIIMnII/MnIIMnIII oxi-
dation in 1, 2, and 5
Table 2. Redox potentials of complexes 1Ϫ5
Compound
Epa [V]
Epa/2 [V]
Epc
∆E [V]
Complex
E1/2 (∆E) [V] vs. Ag/AgCl
II/III
Mn2II/II/Mn2
1
2
5
0.37
0.68
0.37
0.46
0.78
0.48
0.51
0.83
0.53
0.14
0.15
0.16
1
2
3
4
5
0.44 (0.14)
0.76 (0.15)
0.45 (0.07)
0.64 (0.07)
0.45 (0.16)
Thus, on the time scale of cyclic voltammetry (250 mV
Ϫ1) our investigations lead to the following conclusion. For
s
compounds 1Ϫ5 the first redox transition can clearly be
attributed to a single-electron transfer from MnIIMnII to
MnIIMnIII
.
It is noteworthy that both oxidation waves in the cyclic
voltammogram of 2 differ in intensity by a factor of 1.3.
For the remaining complexes this difference is significantly
smaller. To further investigate this second oxidation peak,
additional electrochemical studies were performed on the
pure ligand. In analogy to data reported by Blondin et al.,
oxidation of the phenolic part of the ligands occurs at
higher potentials.[31] Those irreversible, ill-defined cyclic
voltammetric traces overlay any supposed oxidation from
MnIIMnIII to MnIIIMnIII
.
However, there is a clear correlation between the revealed
redox potentials and the type of ring substituent. As shown
in Table 2, the increase of the redox potentials for the first
transition is directly commensurate with the intensity of the
Figure 5. Cyclic voltammogram of 2 in acetonitrile
882
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
Eur. J. Inorg. Chem. 2004, 879Ϫ887

Tuning the Catalase Activity of Dinuclear Manganese Complexes
FULL PAPER
electron-withdrawing effect of the ring substituents. This
was observed before by Walton et al.[35] Their investigations
on transition metal redox behavior were based on a series
of mononuclear manganese compounds. Using different
phenol-substituted triimine and triamine ligands an anal-
ogous influence on the redox potential was observed for
CH3O, CH3, and NO2 groups as para-substituents of the
phenol group.
Catalase Activity
Compounds 1Ϫ5 are stable under ambient conditions
and are not oxidized by atmospheric oxygen. All complexes
are able to disproportionate hydrogen peroxide in aqueous
solution. Like [Mn(bpia)(µ-OAc)]2(ClO4)2, they are ex-
amples of synthetic catalysts showing saturating kinetics at
high substrate concentrations (see Figure 6).[16]
Figure 7. LineweaverϪBurk plots for 1Ϫ5
(methyl substituted) and compound 5 (methoxy residue).
The slowest complex in terms of turnovers is the p-nitro-
substituted complex 2, exhibiting a threefold slower cata-
lytic conversion than 1. The electron-withdrawing effect of
the nitro group is presumably the decisive difference. This
effect causes a reduction of the electron density of the
manganese centers via the aromatic π-system. Since the dis-
proportionation of hydrogen peroxide is a redox process
this effect decreases the speed of substrate conversion.
In order to understand the observed kinetic data a com-
parison of kcat/KM as a criterion for catalytic efficiency and
the determined redox potentials is necessary. By comparing
the kcat/KM criteria of the five complexes a dependence of
the rate of the hydrogen peroxide disproportionation on the
ring substituents is also observed. Starting with complex 1,
which contains the unsubstituted parent ligand, the activity
increases if electron-donating groups like methyl or meth-
oxy substituents are bound to the pyridine or phenol arm
(2 and 5). Although the methoxy group is bigger than the
Figure 6. Initial rate of substrate consumption versus substrate
concentration at constant concentration of 1Ϫ5
Consequently, the data were fitted to the methyl group, complex 5 is more efficient than 2. This can
MichaelisϪMenten equation, and the turnover number kcat be attributed to the increased ability to donate electrons
,
the Michaelis constant KM and kcat/KM were determined into the aromatic π-system compared to the methyl group.
from the double reciprocal LineweaverϪBurk plot (Fig- It seems to be that the complexes’ catalase activity is more
ure 7). The values of kcat, KM and kcat/KM are listed in dependent on electronic than on steric effects. Therefore, by
Table 4.
using a methoxy substituent kcat/KM of 5 is larger than for
Taking KM as a measure of substrate affinity, the meth- both 1 and 2. On the other hand kcat/KM decreases if an
oxy-substituted compound 5 exhibits the highest affinity for NO2 group is bonded in the para position to the bridging
hydrogen peroxide (lowest value of KM) while the parent phenol oxygen atom. This is caused by its strong electron-
complex 1 shows the lowest value for substrate binding withdrawing effect, which reduces the electron density of
(highest value of KM). This is noteworthy if one considers the manganese centers via the aromatic π-system. The use
that the unsubstituted ligand should allow easier access to of both an electron-withdrawing and -donating group, as in
the metal center, resulting in a higher affinity for hydrogen 4, leads to a higher kcat/KM than for 2.
peroxide. Therefore electronic reasons must be the rate-de-
As shown by CV experiments (vide supra) the redox po-
termining effects for substrate binding. It is conceivable that tentials correlate with the insertion of electron-withdrawing
the methoxy lone pairs interact with the substrate leading or -donating substituents. The cyclic voltammograms of
to higher substrate binding.
compounds 1, 3 and 5 exhibit lower values for the observed
The turnover number kcat can be interpreted as the rate of redox transitions, while the corresponding E1/2 values for
substrate conversion into the corresponding products. Once the remaining complexes are shifted to higher potentials.
H2O2 is bonded to the metal center the rate of conversion This coherence between the redox potentials and the util-
is best for the parent complex 1, followed by complex 3 ized ring substituents underlines the determined catalase ac-
Eur. J. Inorg. Chem. 2004, 879Ϫ887
www.eurjic.org
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
883

N. Reddig, D. Pursche, M. Kloskowski, C. Slinn, S. M. Baldeau, A. Rompel
FULL PAPER
Table 4. Catalytic activity of manganese catalase and synthetic catalase mimics
kcat [sϪ1
]
KM [m]
kcat/KM [Ϫ1·sϪ1
]
Ref.
[36]
[7]
Thermus thermophilus catalase
Lactobacillus plantarum catalase
Thermoleophilum album catalase
[Mn2(bpia)2(µ-OAc)2](ClO4)2
2.6 ϫ 105
2.0 ϫ 105
2.6 ϫ 104
1070
140
165
283
115.6
87.7
2500
83
350
15
31.5
18
24.2
54.3
33.2
31.8
250
10.2Ϫ118.0
Ϫ
3.1 ϫ 106
0.6 ϫ 106
1.7 ϫ 106
34000
7800
6800
5400
3500
2750
1000
160Ϫ990
Ϫ
[6]
[16]
5
this work
this work
this work
this work
3
1
4
2
this work
[37,38]
[Mn2(salpn)2O2]
[13,14]
[39] [b]
[16]
[Mn2(2-OHsalpn)2]
4.2Ϫ21.9
13.2[a]
8.1[a]
[Mn2(tacn)(bipy)(µ-O)2(µ-OAc)(MeOH)](ClO4)2·MeOH
[Mn2(bpia)2(µ-O)2](ClO4)2(PF6)·2CH3CN
[Mn2(tacn)(µ-O)2(µ-OAc)(OAc)2]
[Mn2(L1)(µ-OAc)(H2)]2ϩ
[Mn2(salpentO)(µ-OAc)(µ-OMe)(MeOH)2]
[Mn(H2O)6](ClO4)2
Ϫ
Ϫ
0.3
36
Ϫ
Ϫ
700
18
[17] [b]
[9Ϫ12] [c]
[40]
5.5[a]
2.1[a]
0.66
[13]
0.0063[a]
Ϫ
Ϫ
[a] No saturation kinetics observed/reported. [b] tacn ϭ 1,4,7-trimethyl-1,4,7-triazacyclononane; bpy ϭ 2,2Ј-bipyridine. [c] L1 ϭ N,N,NЈ,NЈ-
tetrakis(2-methylenebenzimidazolyl)-1,3-diaminopropan-2-ol.
tivity. This is in good agreement with experiments reported drawing substituents cause the opposite effect, namely
previously concerning the binding efficiency of hydrogen higher E1/2 values and lower catalase activity. All com-
peroxide to the manganese center,[13] where it has been re- pounds belong to the class of synthetic model complexes
ported that the manganese compounds donating the most showing saturation kinetics at high substrate concen-
electrons have the highest affinity for the substrate.
trations. The observed catalase activity places the com-
Since complexes 1Ϫ5 are hexacoordinate in the solid plexes 1Ϫ5 within approximately three orders of magnitude
state it is most likely that at least one chloride ligand is of the three characterized manganese catalases in terms of
replaced in solution by a solvent molecule that can easily be both k2 and k2/KM criteria. Compared to the two most ef-
replaced by the substrate. This has been proved by MALDI ficient systems [Mn2(bpia)2] and [Mn2(2-OHsalpn)2], de-
experiments (see Exp. Sect.). This soft ionization method scribed by Triller et al.[16] and Gelasco et al.,[13] respectively,
ensures that each species in solution is detected without compounds 1Ϫ5 are the second best functional manganese
further conversion into fragments. As is shown in the Exp. catalase mimics.
Sect. either an [Mϩ Ϫ Cl] or [Mϩ Ϫ HCl] species was de-
tected in every spectrum. This is convincing evidence that
the dinuclear structure is preserved in solution without the
coordination of one chloride ion to the dinuclear metal
Experimental Section
core. Therefore one manganese ion provides a labile coordi-
nation site. This is sufficient to disproportionate H2O2 as it PerkinϪElmer Spectrum GX FT-IR spectrometer in the range of
Physical Measurements: IR spectra were recorded with
a
4000Ϫ400 cmϪ1. Samples were prepared as KBr pellets. Elemental
analyses were carried out with a PerkinϪElmer 2400 Series 2 ana-
lyzer. MALDI experiments were performed with a Bruker Dalton-
ics MALDI Reflex IV spectrometer. The compounds were dis-
solved in the same solvent system as used for the kinetic measure-
ments. Cyclic voltammetric experiments were performed with a
BAS CV-27 voltammograph equipped with a BAS C-1B cell stand
and a BAS RXY recorder. The concentration of the samples was
1 ϫ 10Ϫ3  in acetonitrile for 2, 1 ϫ 10Ϫ3  in acetonitrile/DMF
(15:1) for 4 and 5, 3 ϫ 10Ϫ3  in acetonitrile/DMF (15:1) for 1
and 3. Prior to use, solvents were purified by standard literature
methods.[41] Tetrabutylammonium hexafluorophosphate (recrys-
is postulated for the catalytic cycle of Mn catalase.[7]
The observed catalase activities place the complexes 1Ϫ5
within approximately three orders of magnitude of the three
characterized manganese catalases in terms of both k2 and
k2/KM criteria.
Conclusion
The synthesis and structural characterization of five di-
nuclear manganese(/) compounds has been reported. The
utilized tripodal ligands share two pyridine and one phenol tallized from ethanol) was used as supporting electrolyte at a con-
centration of 0.1 . A three-electrode array consisting of a glassy
carbon working electrode, a platinum wire counter electrode and
an Ag/AgCl reference electrode was used.
moiety but differ in the type of substituents coordinated to
the aromatic rings. It seems that the electronic character of
the different ring substituents tunes the catalase activity,
which correlates with the observed redox potentials of the
dinuclear transition metal complexes. Groups donating
electrons into the aromatic π-system lead to higher catalytic
Kinetic Measurements: Catalase activities of all complexes were
measured at 25 °C on the basis of fluorescence quenching using an
oxygen sensor.[42] Hydrogen peroxide solution in water was placed
activity involving lower E1/2 values while electron-with- in a vessel equipped with a thermostat (5 mL, prepared in different
884  2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim www.eurjic.org
Eur. J. Inorg. Chem. 2004, 879Ϫ887

Tuning the Catalase Activity of Dinuclear Manganese Complexes
concentrations from 36.6% hydrogen peroxide solution in water,
FULL PAPER
chloroform (5 ϫ 20 mL). The combined organic layers were dried
concentration determined iodometrically and manganometrically). (MgSO4) and the solvent was removed under vacuum yielding the
Before each measurement, a calibration of the oxygen sensor was
carried out by saturating the solution with dioxygen (100% dioxy-
oily product. Yield: 5.73 g (23.6 mmol, 94%). 1H NMR (300 MHz,
CDCl3): δ ϭ 3.88 (s, 3 H, O-CH3), 3.92 (s, 2 H, CH2), 4.02 (s, 2
gen) and then with argon (0% dioxygen). The dioxygen concen- H, CH2), 6.59Ϫ7.69 (m, 6 H, Ar-H), 8.56 (d, J ϭ 4.8 Hz, 1 H,
tration in dioxygen-saturated water has been determined to be
1.388 m.[43] After a baseline had been established, 0.1 mL of a
0.01  solution of the complex in degassed methanol was added,
and the oxygen evolution was monitored. The average initial rate
Ar-H).
2-{[Bis(pyridin-2-ylmethyl)amino]methyl}-6-methoxyphenol (HL5):
2-Methoxy-6-{[(pyridin-2-ylmethyl)amino]methyl}phenol (4.88 g,
20.0 mmol) and 2-chloromethylpyridine hydrochloride (3.28 g,
20.0 mmol) were dissolved in 100 mL of methanol. A pH value of
9 was adjusted by adding solid potassium carbonate. The mixture
was stirred at room temperature for 3 d while the pH value was
controlled every 24 h. After the remaining K2CO3 was filtered off,
the mixture was acidified carefully with hydrochloric acid to pH ϭ
over
mol(H2O2)·sϪ1·[mol(catalyst)]Ϫ1 [12]
gression from the slope of concentration versus time plots.
three
independent
measurements,
expressed
as
,
was determined by linear re-
Syntheses: The ligands HL3 and HL4 were prepared according to
[21]
the previously reported syntheses of HL1
and HL2,[44] respec-
tively, using N-[(6-methylpyridin-2-yl)methyl]-N-(pyridin-2-yl- 1. Further purification was performed with the method described
methyl)amine instead of N,N-bis(pyridin-2-ylmethyl)amine.
above to yield an orange brown product. Yield: 3.99 g (11.9 mol,
50%). C20H21N3O2 (335.4): calcd. C 71.62, H 6.31, N 12.53; found
C 71.50, H 6.19, N 12.76. H NMR (300 MHz, CDCl3): δ ϭ 3.73
2-[{[(6-Methylpyridin-2-yl)methyl](pyridin-2-ylmethyl)amino}-
methyl]phenol (HL3): Yield: 5.43 g (17 mmol, 75%). C20H21N3O
(319.0): calcd. C 75.24, H 6.58, N 13.17; found C 75.13, H 6.34, N
1
(s, 3 H, OCH3), 3.89 (s, 2 H, CH2), 3.90 (s, 4 H, CH2), 6.71 (dd,
J ϭ 7.6 Hz, J ϭ 7.9 Hz, 1 H, Ar-H), 6.74 (d, J ϭ 7.9 Hz, 1 H, Ar-
H), 6.83 (d, J ϭ 7.6 Hz, 1 H, Ar-H), 7.13Ϫ7.66 (m, 6 H, Ar-H),
8.57 (d, J ϭ 4.8 Hz, 2 H, Ar-H) ppm. 13C NMR (75 MHz, CDCl3):
δ ϭ 55.9, 56.7, 59.0, 111.3, 118.5, 122.3, 122.4, 123.2, 123.3, 134.9,
137.0, 146.9, 148.7, 158.2 ppm. IR (KBr): ν˜ ϭ 3434 cmϪ1 (m, br),
3054 (m), 3010 (m), 2935 (m), 2917 (m), 2835 (m), 2805 (m), 2750
(w), 2609 (w), 1947 (w), 1888 (w), 1586 (s), 1570 (m), 1486 (s), 1472
(s), 1433 (s), 1389 (m), 1372 (m), 1299 (m), 1250 (s), 1235 (s), 1160
(m), 1134 (m), 1073 (s), 1049 (m), 1006 (m), 994 (m), 976 (m), 954
(m), 933 (m), 886 (w), 837 (m), 777 (s), 761 (m), 737 (m), 631 (m),
617 (w), 561 (w), 539 (w), 510 (w), 483 (w), 467 (w).
1
13.15. H NMR (300 MHz, CDCl3): δ ϭ 2.59 (s, 3 H, CH3), 3.80
(s, 2 H, CH2), 3.85 (s, 2 H, CH2), 3.88 (s, 2 H, CH2), 6.77Ϫ7.64
(m, 10 H, Ar-H), 8.57 (d, J ϭ 4.8 Hz, 1 H, Ar-H) ppm. 13C NMR
(75 MHz, CDCl3): δ ϭ 24.2, 56.9, 59.2, 59.3, 116.6, 118.9, 120.1,
121.7, 122.2, 122.9, 123.3, 129.0, 130.2, 136.7, 137.0, 149.0, 157.6,
157.7, 158.0, 158.5 ppm. IR (KBr): ν˜ ϭ 3440 cmϪ1 (m, br), 3049
(s), 2956 (s), 2922 (s), 2855 (s), 2801 (s), 2708 (m), 2608 (m), 2256
(w), 2007 (w), 1979 (w), 1935 (w), 1905 (w), 1876 (w), 1789 (w),
1688 (w), 1612 (m), 1583 (vs), 1512 (w), 1488 (vs), 1465 (s), 1437
(s), 1428 (s), 1375 (s), 1362 (s), 1316 (w), 1289 (m), 1246 (w), 1204
(m), 1181 (w), 1161 (m), 1141 (m), 1122 (m), 1091 (m), 1039 (m),
1003 (m), 990 (w), 973 (m), 955 (w), 937 (w), 899 (w), 892 (w), 870 Synthesis of the Dinuclear Complexes 1؊5: All complexes were pre-
(w), 789 (s), 760 (vs), 717 (m), 646 (w), 604 (m), 558 (m), 542 (w), pared according to the same method. Equivalent amounts of
479 (m), 444 (m).
manganese() chloride tetrahydrate (0.25 mmol, 49.5 mg) and cor-
responding ligand were dissolved in 10 mL of ethanol and triethyl-
amine (0.3 mmol, 42 µL) was added. After stirring the mixture for
12 h, the resulting precipitate was isolated and redissolved in 10 mL
of dichloromethane. Crystals suitable for X-ray diffraction analysis
were obtained by layering with hexane.
2-[{[(6-Methylpyridin-2-yl)methyl](pyridin-2-ylmethyl)amino}-
methyl]-4-nitrophenol (HL4): Yield: 2.4 g (17 mmol, 57%).
C20H20N4O3 (365.1): calcd. C 65.74, H 5.48, N 15.34; found C
65.60, H 5.39, N 15.26. H NMR (300 MHz, CDCl3): δ ϭ 2.60 (s,
3 H, CH3), 3.85 (s, 2 H, CH2), 3.90 (s, 2 H, CH2), 3.92 (s, 2 H,
1
CH2), 6.95Ϫ8.12 (m, 9 H, Ar-H), 8.58 (d, J ϭ 4.8 Hz, 1 H, Ar-H) [Mn2(L1)2(Cl)2]·2CH2Cl2 (1): Yield:113.78 mg (0.12 mmol, 47%).
ppm. 13C NMR (75 MHz, CDCl3): δ ϭ 24.0, 56.0, 58.8, 59.0, 117.2, C38H36Cl2Mn2N6O2 (789.52): calcd. C 57.81, H 4.60, N 10.64;
120.0, 122.0, 122.4, 123.2, 123.6, 125.6, 126.6, 137.0, 137.2, 137.9,
found C 56.96, H 4.46, N 10.49. MS (MALDI): m/z ϭ 754 [Mϩ
Ϫ
148.8, 156.8, 157.9, 161.2, 162.6 ppm. IR (KBr): ν˜ ϭ 3434 cmϪ1 Cl]. IR (KBr): ν˜ ϭ 3436 cmϪ1 (m, br), 3050 (m), 3021 (m), 3002
(m, br), 3074 (m), 3049 (m), 3002 (m), 2930 (m), 2820 (m), 2717 (m), 2982 (w), 2954 (m), 2913 (m), 2878 (m), 2830 (m), 2776 (w),
(m), 2491 (w), 1736 (w), 1619 (m), 1600 (m), 1582 (s), 1511 (s),
2604 (w), 2497 (w), 1885 (w), 1602 (vs), 1571 (s), 1476 (vs), 1452
1488 (s), 1460 (m), 1431 (m), 1377 (m), 1333 (vs), 1292 (s), 1262 (vs), 1396 (w), 1368 (w), 1349 (s), 1326 (w), 1308 (w), 1269 (vs),
(m), 1240 (s), 1178 (m), 1149 (m), 1133 (m), 1110 (m), 1088 (m), 1190 (w), 1150 (m), 1129 (m), 1095 (m), 1052 (m), 1037 (m), 1014
1049 (m), 1009 (m), 976 (m), 931 (w), 904 (w), 878 (m), 839 (m), (m), 999 (m), 978 (m), 960 (m), 926 (m), 881 (s), 837 (m), 759 (vs),
823 (m), 750 (s), 733 (m), 641 (w), 610 (w), 566 (w), 511 (w), 494
(w), 450 (m).
737 (s), 660 (w), 638 (m), 576 (m), 517 (w), 493 (w), 477 (m).
[Mn2(L2)2(Cl)2]·EtOH (2): Yield: 155.41 mg (0.17 mmol, 67%).
The ligand 2-{[bis(pyridin-2-ylmethyl)amino]methyl}-6-methoxy-
phenol (HL5) was synthesized as follows.
C38H34Cl2Mn2N8O6 (879.51): calcd. C 51.89, H 3.90, N 12.74;
found C 51.46, H 3.86, N 12.24. MS (MALDI): m/z ϭ 843 [Mϩ
Ϫ
HCl]. IR (KBr): ν˜ ϭ 3427 cmϪ1 (s, br), 3056 (w), 2956 (w), 2848
(w), 1719 (m, br), 1600 (s), 1573 (m), 1477 (s), 1437 (m), 1395 (w),
1329 (w), 1286 (vs), 1180 (m), 1153 (m), 1126 (m), 1090 (s), 1051
(m), 1014 (m), 977 (w), 961 (w), 933 (m), 894 (w), 838 (w), 763 (m),
737 (s), 671 (m), 648 (s), 516 (w), 461 (m, br).
2-Methoxy-6-{[(pyridin-2-ylmethyl)amino]methyl}phenol: Equiva-
lent amounts of 2-pyridylmethylamine (2.70 g, 25 mmol) and 2-
hydroxy-3-methoxybenzaldehyde (3.80 g, 25 mmol) were dissolved
in 100 mL of methanol and stirred for 2 h. After reduction with
NaBH4 (0.47 g, 12.5 mmol), the solution was stirred for a further
12 h. The mixture was acidified with hydrochloric acid to pH ϭ 1
and, after concentrating under vacuum, extracted with chloroform
[Mn2(L3)2(Cl)2]·CH2Cl2 (3): Yield: 106.27 mg (0.12 mmol, 47%).
C40H40Cl2Mn2N6O2 (817.58): calcd. C 58.76, H 4.93, N 10.27;
(5 ϫ 20 mL) to separate side products and impurities. In order to found C 58.40, H 4.23, N 9.67. MS (MALDI): m/z ϭ 780 [Mϩ
neutralize the hydrochloride, the pH was adjusted to 8 with an
aqueous solution of NaOH and the free amine was extracted with
Ϫ
HCl]. IR (KBr): ν˜ ϭ 3399 cmϪ1 (s, br), 3054 (m), 3024 (m), 3001
(m), 2963 (m), 2923 (m), 2833 (m), 1907 (w), 1677 (s), 1602 (vs),
Eur. J. Inorg. Chem. 2004, 879Ϫ887
www.eurjic.org
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
885

N. Reddig, D. Pursche, M. Kloskowski, C. Slinn, S. M. Baldeau, A. Rompel
Table 5. Details of structure determination, refinement and experimental parameters of compounds 1Ϫ5
FULL PAPER
1
2
3
4
5
Empirical formula
Formula mass [g/mol]
Temperature [K]
C40H40Cl6Mn2N6O2 C40H40Cl2Mn2N8O7 C41H42Cl4Mn2N6O2 C41H40Cl4Mn2N8O6 C40H42Cl2Mn2N6O4
959.36
213(2)
925.58
213(2)
902.33
213(2)
992.33
100(2)
851.56
100(2)
˚
Wavelength [A]
0.71073
monoclinic
P21/c (No. 14)
11.104(2)
15.394(3)
25.306(5)
92.80(3)
4320.4(15)
4
0.71073
monoclinic
P21/c (No. 14)
20.214(4)
11.620(2)
17.324(3)
96.94(3)
4039.4(13)
4
0.71073
monoclinic
P21/n (No. 14)
11.870(2)
13.171(3)
15.417(3)
111.72(3)
2239.2(8)
2
0.71073
monoclinic
I2/a (No. 15)
22.997(5)
8.762(2)
23.602(5)
101.37(3)
4662.5(18)
4
1.54178
monoclinic
I2/a (No. 15)
21.8425(5)
8.7350(2)
21.9876(7)
113.81(1)
3838.2(1)
4
Crystal system
Space group
˚
a [A]
˚
b [A]
˚
c [A]
β [°]
3
˚
V [A ]
Z
Density (calcd.) [g/cm3]
1.475
1.522
1.459
1.526
1.470
µ [mmϪ1
]
0.998
0.818
0.965
0.941
7.040
Crystal size [mm]
θ range [°]
0.31 ϫ 0.20 ϫ 0.12 0.48 ϫ 0.40 ϫ 0.24 0.40 ϫ 0.40 ϫ 0.32 0.13 ϫ 0.11 ϫ 0.09 0.12 ϫ 0.10 ϫ 0.08
8.5 Յ 2θ Յ 52.0
Ϫ13 Յ h Յ 13
Ϫ18 Յ k Յ 17
Ϫ30 Յ l Յ 31
30989
8.8 Յ 2θ Յ 52.0
Ϫ24 Յ h Յ 24
Ϫ14 Յ k Յ 14
Ϫ21 Յ l Յ 21
29776
9.1 Յ 2θ Յ 52.0
Ϫ14 Յ h Յ 14
Ϫ16 Յ k Յ 16
Ϫ18 Յ l Յ 18
15763
3.52 Յ 2θ Յ 56.58 8.80 Յ 2θ Յ 142.36
Index ranges
Ϫ30 Յ h Յ 30
Ϫ11 Յ k Յ 11
Ϫ31 Յ l Յ 31
23512
Ϫ26 Յ h Յ 26
Ϫ10 Յ k Յ 10
Ϫ26 Յ l Յ 25
10378
Reflections collected
Independent reflections
8406
[R(int) ϭ 0.1966]
8406
7886
[R(int) ϭ 0.1376]
7886
4356
[R(int) ϭ 0.1474]
4356
5795
[R(int) ϭ 0.0337]
5795
3558
[R(int) ϭ 0.0342]
3558
Data
Parameters
505
1.033
518
1.025
292
1.084
339
1.007
335
0.998
Goodness-of-fit on F2
Final R indices [I Ͼ 2σ(I)] R1 ϭ 0.0701
wR2 ϭ 0.1258
R1 ϭ 0.0572
wR2 ϭ 0.1363
R1 ϭ 0.0792
wR2 ϭ 0.1510
0.585/Ϫ0.667
R1 ϭ 0.0703
wR2 ϭ 0.1843
R1 ϭ 0.0916
wR2 ϭ 0.2266
1.076/Ϫ1.018
R1 ϭ 0.0417
wR2 ϭ 0.1105
R1 ϭ 0.0586
wR2 ϭ 0.1134
0.570/Ϫ0.662
R1 ϭ 0.0380
wR2 ϭ 0.0980
R1 ϭ 0.0526
wR2 ϭ 0.1036
0.587/Ϫ0.220
R indices (all data)
R1 ϭ 0.1624
wR2 ϭ 0.1609
0.581/Ϫ0.382
Largest diff. peak/hole
Ϫ
3
˚
[e /A ]
1575 (w), 1478 (vs), 1455 (vs), 1396 (w), 1378 (w), 1349 (w), 1330 a Bruker AXS SMART 6000 with Cu-Kα radiation (λ ϭ 1.54178
˚
(w), 1297 (w), 1271 (s), 1163(s), 1156 (m), 1120 (m), 1092 (s), 1038
A) at 100 K. All structures were solved with direct methods and
(m), 1009 (s), 979 (w), 965 (w), 888 (s), 844 (w), 787 (vs), 757 (w), refined by full-matrix least-squares procedures on F2o values
738 (w), 726 (w), 656 (w), 641 (m), 576 (w), 524 (w), 495 (w), 474 (SHELXTL-Plus-97[45,46]). Anisotropic displacement parameters
(w).
were used to refine the positions of all non-hydrogen atoms. Hydro-
gen atoms were placed at calculated positions according to a riding
model with group isotropic temperature factors. The residual value
wR2 in Table 5 is defined as [w(F2o Ϫ Fc2)2/w(F2o)2]1/2. Selected dis-
tances and angles are listed in Tables 1Ϫ4. CCDC-205842 (1),
-205843 (2), -205844 (3), -205845 (4), and -205846 (5) contain the
supplementary crystallographic data for this paper. These data can
be obtained free of charge at www.ccdc.cam.ac.uk/conts/retriev-
ing.html [or from the Cambridge Crystallographic Data Centre, 12
Union Road, Cambridge CB2 1EZ, UK; Fax: (internat.) ϩ 44-
1223/336-033; E-mail: deposit@ccdc.cam.ac.uk].
[Mn2(L4)2(Cl)2]·CH2Cl2 (4): Yield: 141.63 mg (0.14 mmol, 57%).
C40H38Cl2Mn2N8O6 (907.57): calcd. C 52.93, H 4.22, N 12.35;
found C 53.59, H 4.05, N 11.31. MS (MALDI): m/z ϭ 870 [Mϩ
Ϫ
HCl]. IR (KBr): ν˜ ϭ 3436 cmϪ1 (s, br), 3054 (w), 2955 (m), 2922
(m), 2675 (w), 2492 (w), 1974 (w), 1595 (s), 1575 (s), 1477 (s), 1437
(w), 1327 (w), 1287 (s), 1180 (m), 1127 (w), 1091 (s), 1051 (w), 1011
(m), 966 (w), 935 (m), 895 (w), 880 (w), 834 (m), 782 (m), 764 (m),
736 (w), 671 (m), 646 (m), 558 (w), 514 (w), 488 (w), 455 (w).
[Mn2(L5)2(Cl)2] (5): Yield: 131.51 mg (0.16 mmol, 62%).
C40H40Cl2MnN6O4 (848): calcd. C 56.55, H 4.75, N 9.89; found C
56.23, H 4.89, N 9.24. MS (MALDI): m/z ϭ 813 [Mϩ Ϫ Cl]. IR
(KBr): ν˜ ϭ 3429 cmϪ1 (s, br), 3055 (w), 2921 (m), 2849 (m), 1603
(s), 1571 (m), 1478 (vs), 1456 (m), 1441 (m), 1388 (w), 1346 (w),
1308 (m), 1273 (m), 1237 (m), 1154 (w), 1126 (w), 1079 (m), 1051
(m), 1014 (m), 1000 (m), 970 (w), 897 (w), 849 (m), 767 (m), 736
(m), 638 (w), 618 (w), 494 (w), 442 (w), 410 (w).
Acknowledgments
Financial support by the DFG (Deutsche Forschungsgemeinschaft,
SPP 1118) and the FCI (Fonds der Chemischen Industrie) is grate-
fully acknowledged.
[1]
´
M. J. Mate, G. Murshudov, J. Bravo, W. R. Melik-Adamyan,
P. C. Loewen, I. Fita, in Handbook of Metalloproteins (Eds.:
A. Messerschmidt, R. Huber, T. Poulos, K. Wieghardt), John
Wiley & Sons, New York 2001, p. 468.
Y. Kono, I. Fridovich, J. Biol. Chem. 1983, 258, 6015.
W. F. Beyer Jr., I. Fridovich, Biochem. 1985, 24, 6460.
V. V. Barynin, A. A. Vagin, W. R. Melik-Adamyan, A. I. Greb-
enko, S. V. Khangulov, A. N. Popov, M. E. Andrianova, B. K.
X-ray Crystallographic Study: Single-crystal X-ray studies of com-
plexes 1Ϫ3 were performed with a Stoe IPDS diffractometer with
graphite-monochromated Mo-Kα radiation (λ ϭ 0.71073 A) at
˚
[2]
[3]
[4]
213 K; studies on compound 4 were done with a Bruker AXS
SMART APEX with graphite-monochromated Mo-Kα radiation
˚
(λ ϭ 0.71073 A) at 100 K; studies on compound 5 were done with
886
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
Eur. J. Inorg. Chem. 2004, 879Ϫ887

Tuning the Catalase Activity of Dinuclear Manganese Complexes
FULL PAPER
M. Vaidyanathan, R. Viswanathan, M. Palaniandavar, T. Bala-
subramanian, P. Prabhaharan, T. P. Muthiah, Inorg. Chem.
1998, 37, 6418.
R. Viswanathan, M. Palaniandavar, T. Balasubramanian, T. P.
Muthiah, Inorg. Chem. 1998, 37, 2943.
H.-R. Chang, S. K. Larsen, P. D. W. Boyd, C. G. Pierpont, D.
N. Hendrickson, J. Am. Chem. Soc. 1988, 110, 4565.
M. Marappan, V. Narayanan, M. Kandaswamy, J. Chem. Soc.,
Dalton Trans. 1998, 3405.
[27]
Vainshtein, Dokl. Akad. Nauk SSSR 1986, 288, 877; Chem.
Abstr. 1986, 105, 74919q.
V. V. Barynin, P. D. Hempstead, A. A. Vagin, S. V. Antonyuk,
[5]
[28]
[29]
[30]
[31]
[32]
[33]
W. R. Melik-Adamyan, V. S. Lamzin, P. M. Harrison, P. Ar-
tymiuk, J. Inorg. Biochem. 1997, 67, 196.
[6]
G. S. Allgood, J. J. Perry, J. Bacteriol. 1986, 168, 563.
[7]
J. E. Penner-Hahn, in Manganese Redox Enzymes (Ed.: V. L.
Pecoraro), VCH Publishers, Inc., New York, 1992, p. 29.
[8]
J. E. Penner-Hahn, Struct. Bonding (Berlin, Ger.) 1998, 90, 1.
[9]
´ `
A. E. M. Boelrijk, G. C. Dismukes, Inorg. Chem. 2000, 39,
3020.
P. Mathur, M. Crowder, G. C. Dismukes, J. Am. Chem. Soc.
C. Hureau, E. Anxolabehere-Mallart, M. Nierlich, F. Gonnet,
E. Riviere, G. Blondin, Eur. J. Inorg. Chem. 2002, 2710.
D. P. Kessissoglou, W. H. Butler, V. L. Pecoraro, Inorg. Chem.
`
[10]
1987, 109, 5227.
P. J. Pessiki, S. V. Khangulov, D. M. Ho, G. C. Dismukes, J.
1987, 26, 495.
[11]
J. Heinze, Angew. Chem. 1984, 96, 823; J. Heinze, Angew. Chem.
Int. Ed. Engl. 1984, 84, 831.
Am. Chem. Soc. 1994, 116, 891.
[12]
[34]
[35]
P. J. Pessiki, G. C. Dismukes, J. Am. Chem. Soc. 1994, 116, 898.
R. S. Nicholson, I. Shain, Anal. Chem. 1964, 36, 706.
E. A. Lewis, J. R. Lindsay Smith, P. H. Walton, S. J. Archibald,
S. P. Foxon, G. M. P. Giblin, J. Chem. Soc., Dalton Trans.
2001, 1159.
[13]
A. Gelasco, S. Bensiek, V. L. Pecoraro, Inorg. Chem. 1998,
37, 3301.
A. Gelasco, V. L. Pecoraro, J. Am. Chem. Soc. 1993, 115, 7928.
A. Gelasco, M. L. Kirk, J. W. Kampf, V. L. Pecoraro, Inorg.
[14]
[15]
[36]
[37]
M. Shank, V. Barynin, G. C. Dismukes, Biochemistry 1994,
Chem. 1997, 36, 1829.
M. U. Triller, W.-Y. Hsieh, V. L. Pecoraro, A. Rompel, B.
33, 15433.
[16]
A. C. Rosenzweig, C. A. Frederick, S. J. Lippard, P. Nordlund,
Nature 1993, 366, 537.
E. J. Larson, V. L. Pecoraro, J. Am. Chem. Soc. 1991, 113, 7809.
U. Bossek, M. Saher, T. Weyhermüller, K. J. Wieghardt, J.
Chem. Soc., Chem. Commun. 1992, 1780.
C. Palopoli, B. Chansou, J.-P. Tuchagues, S. Signorella, Inorg.
Chem. 2000, 39, 1458.
Krebs, Inorg. Chem. 2002, 41, 5544.
S. Ito, S. Nishino, H. Itoh, S. Ohba, Y. Nishida, Polyhedron
[17]
[38]
[39]
1998, 17, 1637.
[18]
S. Ito, M. Suzuki, T. Kobayashi, H. Itoh, A. Harada, S. Ohba,
[40]
[41]
[42]
[43]
[44]
Y. Nishida, J. Chem. Soc., Dalton Trans. 1996, 2579.
[19]
H. Sugimoto, Y. Sasaki, Chem. Lett. 1997, 541.
[20]
A. Trösch, H. Vahrenkamp, Eur. J. Inorg. Chem. 1998, 827.
S. Yan, L. Que Jr., L. F. Taylor, O. P. Anderson, J. Am. Chem.
J. F. Coetzee, in Recommended Methods for Purification of Sol-
vents, Pergamon, Oxford, 1982.
[21]
Soc. 1988, 110, 5222.
S. Yan, X. Pan, L. F. Taylor, J. H. Zhang, C. J. O’Connor, D.
R. Könecke, A. Comte, H. Jürgens, O. Kohls, H. Lam, T.
Scheper, Chem. Ing. Tech. 1998, 70, 1611.
Handbook of Chemistry and Physics (Ed.: R. C. Weast), 54th
ed., CRC Press, Cleveland, 1973؊1974, B 116.
Y. Nishida, H. Shimo, S. Kida, J. Chem. Soc., Chem. Commun.
1984, 1611.
G. M. Sheldrick, SHELXS-97, Universität Göttingen 1997.
G. M. Sheldrick, SHELXL-97, Universität Göttingen 1997.
Received March 25, 2003
[22]
Britton, O. P. Anderson, L. Que Jr., Inorg. Chim. Acta 1996,
243, 1.
[23]
H. Adams, N. A. Bailey, I. K. Campbell, D. E. Fenton, Q.-Y.
He, J. Chem. Soc., Dalton Trans. 1996, 2233.
[24]
[45]
[46]
Y. Nishida, Chem. Lett. 1987, 16, 2151.
[25]
U. Rajendran, R. Viswanathan, M. Palaniandavar, M. Laksh-
minarayanan, J. Chem. Soc., Dalton Trans. 1992, 3563.
R. Uma, R. Viswanathan, M. Palaniandavar, M. Lakshmina-
[26]
Early View Article
rayanan, J. Chem. Soc., Dalton Trans. 1994, 1219.
Published Online January 19, 2004
Eur. J. Inorg. Chem. 2004, 879Ϫ887
www.eurjic.org
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
887