Synthesis and properties of a heterobimetallic iron-manganese complex and its comparison with homobimetallic analogues

Article abstract of DOI:10.1016/j.ica.2019.03.029

Heterobimetallic cofactors containing one manganese and one iron ion have recently been found within the di-metal carboxylate protein family. Herein we report the synthesis and characterization of three binuclear metal complexes with Fe-Fe, Mn-Mn, and Fe-Mn metal composition. All three complexes use the same ligand framework, the BPMP ligand (HBPMP = 2,6-bis[(bis (-2-pyridylmethyl)amine) methyl]-4-methylphenol)) with two additional acetate ligands bridging the two metals. In terms of stability towards metal exchange, the Fe-Mn is more stable than the Mn-Mn complex but less stable than the Fe-Fe complex. Cyclic voltammetry shows that the Fe-Mn complex behaves markedly different than the homobimetallic complexes. The Fe-Mn complex also shows higher reactivity with O₂ than both the Fe-Fe and the Mn-Mn counterparts.

Full text of DOI:10.1016/j.ica.2019.03.029

InorganicaChimicaActa490(2019)254–260
Contents lists available at ScienceDirect
Inorganica Chimica Acta
journal homepage: www.elsevier.com/locate/ica
Research paper
Synthesis and properties of a heterobimetallic iron-manganese complex and
its comparison with homobimetallic analogues
Michele Bedina, Hemlata Agarwalaa, Jennifer Marxb, Volker Schünemannb, Sascha Otta,
⁎
Anders Thappera,
a Department of Chemistry – Ångström Laboratory, Uppsala University, P. O. Box 523, S-751 20 Uppsala, Sweden
b Department of Physics, Technische Universität Kaiserslautern, Postfach 3049, 67653 Kaiserslautern, Germany
A R T I C L E I N F O
A B S T R A C T
Keywords:
Heterobimetallic cofactors containing one manganese and one iron ion have recently been found within the di-
metal carboxylate protein family. Herein we report the synthesis and characterization of three binuclear metal
complexes with Fe-Fe, Mn-Mn, and Fe-Mn metal composition. All three complexes use the same ligand frame-
work, the BPMP ligand (HBPMP = 2,6-bis[(bis (-2-pyridylmethyl)amine) methyl]-4-methylphenol)) with two
additional acetate ligands bridging the two metals. In terms of stability towards metal exchange, the Fe-Mn is
more stable than the Mn-Mn complex but less stable than the Fe-Fe complex. Cyclic voltammetry shows that the
Fe-Mn complex behaves markedly different than the homobimetallic complexes. The Fe-Mn complex also shows
higher reactivity with O2 than both the Fe-Fe and the Mn-Mn counterparts.
Iron
Manganese
Heterobimetallic
Oxygen reactivity
1. Introduction
they react when exposed to oxygen [6–7]. The first class is oxygen-
dependent, the second class works both in aerobic and anaerobic con-
Nature has incorporated transition metals in the active sites of many
enzymes that are involved in a diverse range of chemical reactions. For
some metalloenzymes, two different metals are present in the active
site. Examples of these heterobimetallic enzymes are the [NiFe]-hy-
drogenases that catalyse both the oxidation of hydrogen and the re-
duction of protons [1], the cytochrome-c oxygenases that reduce di-
oxygen [2], and the Cu-Zn superoxide dismutases that catalyse the
disproportionation of superoxide to dioxygen and hydrogen peroxide
[3–5].
ditions, while the third class is oxygen-sensitive. Class I is the most
related to this study and has been further subdivided into Class Ia
containing the classical motif with two iron ions in the active site, Class
Ib that can also function with two manganese ions, and Class Ic con-
taining one iron and one manganese ion [6,12–15].
Class I RNRs are composed of two non-identical proteins called R1
and R2 [6]. The R1 subunit has the binding site for the substrate and
effector molecules that alter the activity. The R2 subunit contains the
bimetallic centre that initiates the reaction with oxygen to create a
radical that is first stabilized as a tyrosyl radical (Y·), and subsequently
shuttled to a cysteine residue in the R1 subunit during catalysis
[16–18].
One class of enzymes that has recently been shown to have members
with heterobimetallic cofactors is the di-metal carboxylate protein fa-
mily. This family primarily has two Fe ions in the active site and the
best studied members are the ribonucleotide reductase (RNR) R2 pro-
teins [6–7], and soluble methane monooxygenases [8–9]. These en-
zymes have a carboxylate-bridged di-metal centre in their active sites
and are involved in dioxygen activation [10].
The Class Ic R2 proteins, e.g. CtR2c present in the Chlamydia tra-
chomatis [19], lacks the radical harbouring tyrosine [13–14]. The one-
electron oxidizing equivalent is instead proposed to be stored as a Fe
(III)-Mn(IV) species that replaces the Fe(III)-Fe(III)-Y· cofactor in stan-
dard R2s [20–24].
RNR is one of the most important enzymes for life and is present in
all living organisms, from bacteria to mammals (including humans).
RNR converts the four standard ribonucleotides to their deoxyr-
ibonucleotide counterparts and thereby provides the precursors needed
for both synthesis and repair of DNA [6,11].
A similar Fe-Mn cofactor was found in another human pathogen,
Mycobacterium tuberculosis [25]. This R2c like protein exhibits an un-
precedented tyrosine−valine cross-link in the vicinity of the iron ion
that results from a two-electron oxidation, probably catalysed by the
metal centre. Interestingly, it was also shown that not only the Fe(III)-
Three main classes of RNR have been identified according to how
⁎ Corresponding author.
E-mail address: anders.thapper@kemi.uu.se (A. Thapper).
https://doi.org/10.1016/j.ica.2019.03.029
Received 18 December 2018; Received in revised form 15 March 2019; Accepted 18 March 2019
Availableonline19March2019
0020-1693/©2019ElsevierB.V.Allrightsreserved.

M. Bedin, et al.
InorganicaChimicaActa490(2019)254–260
Mn(IV) form is stable to incubations with H2O2, but the reduced forms
are efficiently activated by H2O2 treatment [25–27].
(82%) 1H NMR (CD3OD): δ 7.10 (s, 2H, ArH), 4.67 (s, 4H, CH2) 2.25 (s,
3H, CH3).
Synthetic binuclear model complexes containing iron and/or man-
ganese ions could be used to gain further insight into how hetero-
bimetallic cofactors are assembled and how they differ from the more
common homobimetallic counterparts. To date, only a few hetero-
bimetallic compounds which contain a Fe-Mn centre have been syn-
thesized and characterized but not as heterometallic active site mimics
[28–31].
2.1.3. Synthesis of di-(2-picolyl)amine (DPA)
The synthesis followed a modified literature procedure [34]. 2-
amino-methylpyridine (4.5 ml, 46.2 mmol) was put in a round bottom
flask with 20 ml of dry methanol. In another round bottom flask, the 2-
pyridinecarboxaldehyde (4.95 g, 46.2 mmol) was dissolved in 20 ml of
dry methanol and transferred in a dropping funnel. The pyridine car-
boxaldehyde was added to the 2-amino-methylpyridine slowly at 0 ˚C
under inert conditions. After allowing the solution to go to room tem-
perature, it was again cooled to 0 °C and NaBH4 (1.78 g, 47.1 mmol)
was added. The solution was allowed to go to room temperature again
and left stirring overnight. The solution was poured into ice (20 ml), the
pH was adjusted to 3–4 with conc. HCl, and extracted with CH2Cl2 until
the water phase became colourless (5 × 50 ml). The water phase was
neutralized with K2CO3 and extracted with CH2Cl2 3 times. The solvent
was dried with MgSO4 and removed under vacuum followed by a fur-
ther purification through column in alumina deactivated with 4% of
water, the eluent was a mixture of CH2Cl2/CH3OH/TEA in the ratio 75/
2/1. Yield: 4.9 g (53.3%) 1H NMR (CD3OD): δ 8.35 (s, 2H, PyH) 7.43 (s,
2H, PyH) 7.16 (s, 2H, PyH) 6.95 (s, 2H PyH) 4.56 (s, NH) 3.78 (4H,
CH2)
The main goal of this study is to examine the relative stability and
reactivity towards molecular oxygen of three binuclear complexes using
the same ligand framework, but with different metal composition – Fe-
Fe, Fe-Mn, and Mn-Mn, respectively. For this, the BPMP ligand
(HBPMP = 2,6-bis[bis(2-pyridylmethyl)aminomethyl]-4-methyl-
phenol) was chosen. BPMP binds two metal ions with a bridging phe-
nolate and the coordination sphere of the two metal ions can be com-
pleted with e.g. two additional bridging acetate groups. The three
complexes, [(BPMP)Fe2(OAc)2](PF6) (1), [(BPMP)Mn2(OAc)2](PF6)
(2), and [(BPMP)FeMn(OAc)2](PF6)(HPF6) (3) have been synthesized
and characterized and their stability and reactivity with molecular
oxygen has been investigated (Scheme 1).
2. Experimental section
2.1.4. Synthesis
of
(2,6-bis[(bis(-2-pyridylmethyl)amine)methyl]-4-
2.1. Synthesis
methylphenol) (HBPMP)
The synthesis followed a modified literature procedure [35]. A
mixture of DPA (4.21 g, 20.68 mmol) and triethylamine (4.18 g,
41.3 mmol) was dissolved in THF and added dropwise to a solution of
2,6-bis(chloromethyl)-4-methylphenol (2.12 g, 10.3 mmol) in THF at 0
°C. After the complete addition of the DPA mixture the solution was
warmed up to room temperature and left stirring for 4 days. The trie-
thylamine salt precipitate was removed by filtration and washed 3
times with THF, and the combined filtrates were concentrated under
vacuum. The residue was dissolved in 50 ml of water and extracted with
CH2Cl2 (3 × 50 ml). The extracts were combined, dried with MgSO4
and the solvent was removed under vacuum. The product was obtained
as a yellow oily residue. Yield: 5.24 g (95.6%). 1H NMR (CDCl3): δ 8.42
(4H, m), 7.68 (m, 4H), 7.53 (dd, J = 7.8, 3.2, 4H) 7.22 (m, 4H), 6.92 (d,
J = 3.2, 2H) 3.79 (d, J = 3.3, 8H) 3.71 (d, J = 3.2, 4H) 2.19 (d,
J = 3.3, 3H)
2.1.1. General synthetic procedures
All reagents and solvents were purchased from commercial sources.
The 2,6-bis-(hydroxymethyl)-4-methylphenol was purified by re-
crystallization in acetone before use. All the solvents used were distilled
under N2. Acetonitrile was distilled over CaH2, THF was distilled over
sodium-benzophenone. The dry solvents were collected and transferred
to dry flasks using a cannula before every reaction. Methanol was dis-
tilled over CaH2 and collected under argon in a Schlenk flask with 4 Å
molecular sieves. [Fe(CH3CN)6](PF6)2 was synthesized following a
procedure from the literature [32].
2.1.2. Synthesis of 2,6-bis(chloromethyl)-4-methylphenol
The synthesis followed a modified literature procedure [33]. To a
solution
of
2,6-bis-(hydroxymethyl)-4-methylphenol
(2.54 g,
15.1 mmol) in 50 ml of CH2Cl2, 100 ml of conc. HCl was added. The
mixture was left to stir for 24 h. The compound was extracted with
CH2Cl2; the organic phase was collected and dried with MgSO4 and
removed under reduced pressure to give yellow crystals. Yield: 2.54 g
2.1.5. Synthesis of [(BPMP)Fe2(OAc)2](PF6) (1)
The synthesis followed
a modified literature procedure [35].
HBPMP (1.10 g, 2.08 mmol) was put in a 100 ml round bottom flask and
dissolved in dry acetonitrile under nitrogen flow. Fe(CH3COO)2 (0.72 g,
4.16 mmol) and NH4PF6 (0.34 g, 2.10 mmol) were added. The solvent
was removed under vacuum and the residue was re-dissolved in CH2Cl2.
The solution was transferred to another flask with a cannula with a
filter avoiding transfer of the NH4(CH3COO). The solvent was removed
by vacuum, leaving a red-brown solid. Yield 1.80 g (95%).
2.1.6. Synthesis of [(BPMP)Mn2(OAc)2](PF6) (2)
The synthesis followed a modified procedure literature procedure
[36]. HBPMP (1.53 g, 2.89 mmol) was put in a 100 ml round bottom
flask and dissolved in dry acetonitrile under nitrogen flow. Mn
(CH3COO)2 (1.41 g, 5.77 mmol) and NH4PF6 (0.52 g, 3.19 mmol) were
added. The solvent was removed under vacuum and the residue was re-
dissolved in CH2Cl2. The solution was transferred to another flask with
a cannula avoiding transfer of the NH4(CH3COO), and the solvent was
removed the under vacuum, giving a yellow precipitate. Yield 1.90 g
(73%).
2.1.7. Synthesis of [(BPMP)FeMn(OAc)2](PF6)(HPF6) (3)
Under inert condition HBPMP (2.61 g, 4.92 mmol) was dissolved in
20 ml of dry and oxygen free acetonitrile. Mn(CH3COO)2 (0.84 g,
Scheme 1. Schematic structure of the cation of 1–3.
255

M. Bedin, et al.
InorganicaChimicaActa490(2019)254–260
4.89 mmol) and [Fe(CH3CN)6](PF6)2 (2.83 g, 4.89 mmol) was added in
that order. This procedure avoids the presence of species other than
what are needed for 3. The mixture was left stirring for 1 h and the
solvent was removed under vacuum, giving a dark blue precipitate.
Yield 4.22 g (4.02 mmol, 82%). The dark blue precipitate could be re-
crystallized from an acetonitrile/diethyl ether mixture. Elemental
analysis theoretical: C, 42.39%; H, 3.75%; N, 8.02%; found : C, 41.20%;
H 3.85%; N 8.21%. Metal ratio, Fe:Mn 1:1.2.
spectrum was collected. The procedure was repeated until the total
volume of oxygen injected was 1 ml (5 ml for 3).
3. Results and discussion
3.1. Synthesis
The ligand HBPMP was synthesized following a literature procedure
[34]. Crystalline 2,6-bis-(hydroxymethyl)-4-methylphenol was chlori-
nated with HCl in the 2,6 positions and subsequently di-(2-picolyl)
amine (DPA) was added to complete the ligand framework.
All metal complexes were stored in an inert atmosphere (glove box)
until use.
2.2. Instrumentation
The synthesis of the homobimetallic and heterobimetallic com-
plexes followed a modified version of the procedure published by
Borovik et al. [35]. For the synthesis of 1 the ligand was dissolved in
acetonitrile and two equivalents of Fe(CH3COO)2 were added. Ammo-
nium hexafluorophosphate was added and CH2Cl2 was used to pre-
cipitate 1 with PF6- as the counter ion. The same procedure was used to
synthesize 2 using Mn(CH3COO)2 as the metal source. The syntheses
were performed under inert conditions and the products stored in a
glove-box in an oxygen- and water-free environment.
1H NMR spectra was recorded with JEOL-400 MHz spectrometer at
293 K. Chemical shift are given in ppm and referenced internally to the
residual solvent signal. UV–Vis spectroscopy was performed on a Varian
Cary 50. EPR spectroscopy was performed using a Bruker ESR-500
spectrometer equipped with a dual mode DM9807 resonator an ESR900
cryostat and an Oxfords ITC503 temperature controller. EPR condi-
tions: Microwave frequency, 9.59 GHz, modulation frequency 100 kHz,
modulation amplitude 20 G. HR-MS was performed by Organisch-
Chemisches Institut der Westfälischen Willhelms-Universität, Munster
Germany. IR-spectroscopy was measured on a PerkinElmer Spectrum
One, FT-IR spectrometer with a PerkinElmer universal ATR sampling
accessory. Elemental analysis was performed by Analytische
Laboratorien Gmbh, Lindlar, Germany. Inductively Coupled Plasma
For the synthesis of the heterobimetallic complex 3, the ligand was
dissolved in dry acetonitrile and one equivalent of Mn(CH3COO)2 was
added as a source of manganese and acetate bridges followed by the
addition of Fe(CH3CN)6(PF6)2 as the source of iron and counter ion
[32]. The order in which the metal salts are added is crucial since
manganese is more labile in the ligand. This is shown in 3.6 by ex-
change studies with 1 and 2 separately, and is also expected from the
Irvin-William series [38]. Therefore, the addition of manganese before
iron minimises the formation of undesired homobimetallic side pro-
Atomic Emission Spectroscopy (ICP-AES) was measured on
PerkinElmer Avio 200ICP-OES.
a
-
2.2.1. Electrochemistry
ducts. Surprisingly elemental analysis shows the presence of two PF6
The measurements where done using an AUTOLAB potentiostat
with a three-electrode setup in a glove box, controlled using the GPES
electrochemical interface (eco chemie). The working electrode was a
glassy carbon disc (3 mm diameter), the counter electrode was a glassy
carbon rod and the reference electrode was an Ag wire in a solution of
10 mmol of AgNO3 in CH3CN. The working electrode was routinely
polished inside the glove box with alumina (porosity 0.05 µm) in
CH3CN slurry on a felt surface before the use. The scan was typically
−0.8 V to 1.2 V at 100 mV/s with 1 mM complexes in dry acetonitrile
and 0.1 M TBAClO4 as electrolyte unless otherwise indicated.
counter ions in the product. Since the complex was prepared under
anaerobic conditions, and since the Mössbauer spectroscopy described
in 3.4 shows the presence of 80% Fe(II) we do not believe that the
complex forms in a Fe(III)-Mn(II) oxidation state, even if the mass
spectrometry in 3.2 shows the presence of the oxidized complex. Also
the formation of the Fe(II)-Mn(III) oxidation state is very unlikely, and
the presence of a Mn(II) signal in the EPR spectrum shown in 3.3 is not
supporting this assignment. Instead we propose that the complex forms
in the Fe(II)-Mn(II) oxidation state and that is it isolated together with
-
and extra PF6 anion and presumably an extra proton as cation.
2.2.2. Mössbauer spectroscopy.
3.2. Mass spectrometry
Mössbauer spectra were recorded with a conventional Mössbauer
spectrometer operated in the constant acceleration mode in conjunction
with a 512-channel analyzer (WissEl GmbH). The isomer shift is given
with reference to α-iron at room temperature. A continuous flow
cryostat (Oxford Instruments) was used to cool the samples to 77 K.
Spectral analysis was performed with the public domain program Vinda
running on an Excel 2003® platform using Lorentzian line shapes with
the line width Γ [37].
The HR-MS of 3 showed a peak at m/z = 379.08 with z = 2 corre-
sponding to the [(BPMP)FeMn(OAc)2]2+ ion (Fig. S1) presumably in a
Fe(III)-Mn(II) oxidation state. The isotope pattern indicates the pre-
sence of an impurity (< 20%) of the Fe-Fe complex (1). No peak at m/
z = 758 corresponding to the [(BPMP)FeMn(OAc)2]+ ion could be
observed, likely due to oxidation of the complex when exposed to
oxygen. A similar behaviour was observed for 1 where only the Fe(III)-
Fe(II) form is observed (Fig. S2), but not for 2 where the Mn(II)-Mn(II)
form is the only one detected (Fig. S3).
2.3. Titration of 2 with Fe2+
Complex 2 (1 mM) was dissolved in 5 ml of acetonitrile (with 0.1 M
TBAClO4) and placed in an electrochemical cell. 150 µL of a solution of
[Fe(CH3CN)6](PF6)2 (15 mM) in acetonitrile was added to the cell and
the CV collected. The working electrode was polished, and the proce-
dure was repeated until 2 equiv. of [Fe(CH3CN)6](PF6)2 had been
added.
3.3. IR spectroscopy
The mid-range IR spectra of 1, 2, and 3 in solid state are very similar
(Fig. S4). Three bands originating from coordinated acetate and pyr-
idine ligands are present in all three spectra, at ∼1420, ∼1480
and ∼1590 cm−1 in 1 and 2, and at ∼1435, ∼1480 and ∼1605 cm−1
in 3 [39]. Since these bands are present in all three complexes we can
conclude that 3 must have a structure very similar to that of 1 and 2.
2.4. Oxygen activation
In the glove box, 0.15 mM solutions of 1–3 were prepared in dry and
oxygen free acetonitrile. 3.5 ml of each solution was put in gas-tight
cuvettes and removed from the glove box. Using a gas-tight syringe
oxygen gas was injected into the cuvettes (3–250 µL) and a UV–Vis
3.4. EPR spectroscopy
The EPR spectrum of 3 at 5 K in acetonitrile shows a feature at
g = 2.12 (Fig. 1) from a Mn(II) species. No signals corresponding to the
256

M. Bedin, et al.
InorganicaChimicaActa490(2019)254–260
typical for Fe(II)-high spin (HS) species [41]. The second component with
20% of abundance has δ = 0.43 mms−1 and ΔEQ = 0.45 mms−1, typical
of Fe(III)-HS species. The presence of Fe(III) again demonstrates that 3 is
easily oxidized. Also in the Mössbauer spectrum of 1 two iron species with
different oxidation states could be observed (Fig. 2b), ∼62% Fe(II)-HS
(δ = 1.19 mms−1 and ΔEQ = 2.80 mms−1
)
and ∼ 38% Fe(III)-HS
(δ = 0.46 mms−1 and ΔEQ = 0.70 mms−1) (Table S1). The parameters for
Fe(II)-HS and Fe(III)-HS are in good agreement with the previously re-
ported [(BPMP)Fe(III)Fe(II)(OAc)2](BF4)2·H2O (Fe(II): δ = 1.16 mms−1
,
ΔEQ = 2.42 mms−1; Fe(III): δ = 0.47 mms−1, ΔEQ = 0.51 mms−1).[42]
The parameters for the Fe(II) and Fe(III) components are not identical for
1 and 3 but similar enough that the Fe(III)-Fe(II) impurity of 1 observed in
the HR-MS might be indeed present as Fe(III)-HS in the Mössbauer spec-
trum of 3.
3.6. Electrochemistry
Fig. 1. EPR spectra of 3 (top) and 2 (bottom, dashed) in CH3CN at 5 K. The
characteristic peak from 2 at g ∼ 2.6 is marked with an asterisk (*). Microwave
power, 20 µW.
The electrochemistry of 1 and 2 have previously been studied
[42–43] and both complexes show two redox waves in the cyclic vol-
tammogram (CV) measured in CH3CN, corresponding to the M(III)-M
(II)/M(II)-M(II) (E1a and E2a) and M(III)-M(III)/M(III)-M(II) (E1b and
Mn-Mn species 2, with an EPR-signal at g ∼ 2.6 could be observed
[36,40].
E
2b) couples (Fig. 3). The CV of 3 is more complex than those of 1 and 2
(Fig. 3, blue line). There are two redox events at −0.44 V and +0.28 V
(vs Fc+/Fc) that are at a similar potential to E1a and E1b in 1, but the
the anodic wave corresponding to E1a in 1 is not present in 3. Instead
there is an electrochemically irreversible oxidation wave at 0.08 V that
is presumably the anodic counterpart of the reduction at −0.44 V. The
substantial irreversibility of the Fe(III)-Mn(II)/Fe(II)-Mn(II) couple
3.5. Mössbauer spectroscopy
Mössbauer spectroscopy was used to investigate the two iron con-
taining complexes 1 and 3. The Mössbauer spectrum of 3, measured at
78 K shows the presence of two different iron components (Fig. 2a). The
most abundant species (80%) has an isomer shift δ of 1.13 mms−1 and a
quadrupole splitting ΔEQ of 3.16 mms−1 (Table S1). These parameters are
Fig. 3. Cyclic voltammetry of (top to bottom) 1, 3, 2, and a 1:1 mixture of 1 and
2 (dashed) in CH3CN with 0.1 M TBA(ClO4) (scan rate 100 mV/s). The half
wave potentials for Fe(III)-Fe(II)/Fe(II)-Fe(II) (E1a) and Fe(III)-Fe(III)/Fe(III)-Fe
(II) (E1b) from 1 and Mn(III)-Mn(II)/Mn(II)-Mn(II) (E2a) and Mn(III)-Mn(III)/
Mn(III)-Mn(II) (E2b) from 2 are indicated by dashed vertical lines.
Fig. 2. Mössbauer spectra of a) 3 and b) 1 taken at 78 K. The black solid line is
the sum of the subspectra originating from high-spin Fe(II) (middle) and high-
spin Fe(III) (top). The parameters used for analysis are displayed in Table S1.
257

M. Bedin, et al.
InorganicaChimicaActa490(2019)254–260
(ΔEp = 467 mV) in 3 compared to the reversible Fe(III)-Fe(II)/Fe(II)-Fe
(II) couple (ΔEp = 76 mV) in 1 can also be explained by the hump at
−0.007 V on the cathodic part of the CV of 3, which could be due to a
chemical change in the complex before it gets reduced to the Fe(II)-Mn
(II) form.
have been proposed to form a Fe(III)-super oxide species as a first in-
termediate. The reaction can continue through the formation of
mononuclear Fe(III)-peroxo or Fe(IV)-oxo species, or dimeric Fe(III)-µ
(peroxo)-Fe(III) species. Similar species have also been proposed and
observed for the activation of O2 by manganese complexes. The reac-
tion of O2 with diiron complexes in the Fe(II)-Fe(II) oxidation state can
lead to the formation of Fe(III)-Fe(II) complexes [45–46], but also Fe
(III)-µ(peroxo)-Fe(III) species have been observed at low temperatures
[47].
A quasi-reversible redox wave is found at 0.69 V (ΔEp = 138 mV),
this is in turn similar to the E2b (0.64 V) redox wave (ΔEp = 240 mV) in
2. A comparison with the CV of a 1:1 mixture of 1 and 2 (Fig. 3, dashed
line) that shows (quasi-)reversible redox waves at −0.4 V
(ΔEp = 85 mV), corresponding to E1a, and 0.1 V (ΔEp = 193 mV), cor-
responding to E2b, confirms that the CV of 3 is not contaminated to a
large extent by the presence of 1 or 2 in the solution.
The high-valent iron or manganese species that can oxidise sub-
strates through hydroxylation or oxygen atom transfer to the product.
One such reaction is the oxidation of benzylic protons on a substrate or
on the ligand of the metal complex itself. Recently a manganese com-
plex was reported to oxidise its own ligand in this fashion to give an
alkoxide and a ketone [48].
The CV of 3 indicates that the Fe-Mn complex is somewhat more
fluxional than the homobimetallic complexes 1 and 2 and undergoes
chemical transformation(s) (e. g. an alteration in the binding mode of
the bridging acetates.
The reactivity of 1–3 with molecular oxygen was studied by UV–Vis
spectroscopy. Solutions of the complexes (0.15 mM) were prepared in
gas tight cuvettes in the glove box and moved to the spectrometer
where oxygen gas was injected with a gas-tight syringe. Interestingly,
the three complexes react differently with O2. The Mn(II)-Mn(II) com-
plex 2 does not react with O2 under the conditions used (Fig. S6). The
Fe-Fe complex 1 reacts with the injected O2 and the peak at 435 nm
decreases in intensity over 1 h (Fig. 5). At the same time a band around
300 nm and a broad band around 600 nm increases. The resulting
spectrum is very similar to closely related Fe(III)-Fe(II) complexes using
the HBPMP ligand and different bridging carboxylates prepared in
presence of O2 [35,42]. The band around 600 nm has been assigned to
charge transfer from filled phenolato pπ orbitals to vacant half-filled d
(π*) orbitals of Fe(III) [42]. We therefore assign the species formed in
the reaction of 1 and O2 to be [(BPMP)Fe(III)Fe(II)(OAc)2]2+ The two
isosbestic points in the UV–vis spectra, show that 1 cleanly converts to
the Fe(III)-Fe(II) species.
3.7. Metal exchange
To investigate the relative coordinative stability of Fe and Mn to the
HBPMP ligand, titrations of the homobimetallic complexes 1 and 2 with
Mn and Fe salts respectively were performed under inert atmosphere in
a glove box and followed by cyclic voltammetry. When 1 was titrated
with Mn(ClO4)2 no changes were observed in the CV for the first
equivalent of Mn2+ (Fig. S5), indicating that the two iron centers stay
coordinated to the ligand, and that the complex is stable towards metal
exchange. With higher amounts of Mn2+ the intensity of the peak
currents of the redox waves in the CV decreased drastically but no new
redox waves appeared. In contrast, when 2 was titrated with up to 2
equiv. of Fe(CH3CN)6(PF6)2 drastic changes were observed in the CV
(Fig. 4). The peaks associated with the Mn(III)-Mn(II)/Mn(II)-Mn(II)
and Mn(III)-Mn(III)/Mn(III)-Mn(II) redox events gradually diminished
with the addition of Fe2+. The appearance of a cathodic feature at
−0.41 V that increases with increasing iron concentration indicates
that a new species is formed. In fact, the CVs of 2, after addition of 1
equiv. of Fe2+, and 3 show large similarities, noticeably the irreversible
reduction at ∼−0.4 V suggesting that a Fe-Mn complex has formed
Complex 3 reacts easily with the injected O2, the absorption over
the whole visible range increases and a band at 570 nm becomes much
more pronounced (Fig. 6). The reaction of O2 with 3 is faster than with
1 and is essentially completed already at the first injection of O2. Based
on the similarity with the UV–vis spectrum of 1 oxidized with O2 in
Fig. 5 we assign the product to be the one-electron oxidized 3, [(BPMP)
Fe(III)Mn(II)(OAc)2]2+
during the titration of 2 with Fe2+
.
We also used mass spectrometry to investigate the reaction with 1 or
3 with O2 but no new species other than [(BPMP)FeFe(OAc)2]2+ and
[(BPMP)FeMn(OAc)2]2+ could be observed, indicating that ligand
oxidation did not occur.
3.8. Oxygen reactivity
Iron and manganese complexes are known to activate molecular
oxygen in a large number of natural and synthetic systems [44]. In
mononuclear non-heme synthetic systems the reaction of Fe(II) with O2
The observed reactivity with oxygen shows that the hetero-
bimetallic complex reacts differently than a linear combination of the
Fig. 5. Complex 1 (0.15 mM) in CH3CN (solid line) and with added equivalents
of O2(g) (coloured traces). The solubility of O2 in pure acetonitrile at 20 °C is
2.4–2.6 mM [49–50].
Fig. 4. Cyclic voltammetry in CH3CN (0.1 M TBA(ClO4) as electrolyte) fol-
lowing the titration of 2 (1 mM) with Fe(CH3CN)6(PF6)2.
258

M. Bedin, et al.
InorganicaChimicaActa490(2019)254–260
References
[1] W. Lubitz, H. Ogata, O. Rüdiger, E. Reijerse, Chem. Rev. 114 (2014) 4081–4148,
https://doi.org/10.1021/cr4005814.
[2] H. Michel, J. Behr, A. Harrenga, A. Kannt, Annu. Rev. Biophys. Biomol. Struct. 27
(1998) 329–356, https://doi.org/10.1146/annurev.biophys.27.1.329.
[3] I.A. Abreu, D.E. Cabelli, Biochim. Biophys. Acta 1804 (2010) 263–274, https://doi.
org/10.1016/j.bbapap.2009.11.005.
[4] J.J. Perry, D.S. Shin, E.D. Getzoff, J.A. Tainer, Biochim. Biophys. Acta 1804 (2010)
245–262, https://doi.org/10.1016/j.bbapap.2009.11.004.
[5] J.M. McCord, I. Fridovich, Free Rad. Biol. Med. 5 (1988) 363–369, https://doi.org/
10.1016/0891-5849(88), 90109-8.
[6] P. Nordlund, P. Reichard, Annu. Rev. Biochem. 75 (2006) 681–706, https://doi.
org/10.1146/annurev.biochem.75.103004.142443.
[7] A. Jordan, P. Reichard, Annu. Rev. Biochem. 67 (1998) 71–98, https://doi.org/10.
1146/annurev.biochem.67.1.71.
[8] M. Merkx, D.A. Kopp, M.H. Sazinsky, J.L. Blazyk, J. Müller, S.J. Lippard, Angew.
Chem. 40 (2001) 2782–2807, https://doi.org/10.1002/1521-3773(20010803)
40:15<2782::AID-ANIE2782>3.0.CO;2-P.
[9] M.H. Sazinsky, S.J. Lippard, Acc. Chem. Res. 39 (2006) 558–566, https://doi.org/
10.1021/ar030204v.
[10] A.J. Jasniewski, L. Que Jr., Chem. Rev. 118 (2018) 2554–2592, https://doi.org/10.
1021/acs.chemrev.7b00457.
Fig. 6. Complex 3 (0.15 mM) in CH3CN (solid line) and with added equivalents
of O2(g) (coloured traces). The solubility of O2 in pure acetonitrile at 20 °C is
2.4–2.6 mM [49–50].
[11] A.B. Tomter, G. Zoppellaro, N.H. Andersen, H.-P. Hersleth, M. Hammerstad,
Å.K. Røhr, G.K. Sandvik, K.R. Strand, G.E. Nilsson, C.B. Bell, A.-L. Barra, E. Blasco,
L. Le Pape, E.I. Solomon, K.K. Andersson, Coord. Chem. Rev. 257 (2013) 3–26,
https://doi.org/10.1016/j.ccr.2012.05.021.
two homobimetallic complexes. The presence of two different metals
gives rise to higher reactivity than those of the corresponding homo-
bimetallic complexes.
[12] A.M. Magherusan, A. Zhou, E.R. Farquhar, M. García-Melchor, B. Twamley, L. Que
Jr., A.R. McDonald, Angew. Chem. 57 (2018) 918–922, https://doi.org/10.1002/
anie.201709806.
[13] M. Högbom, P. Stenmark, N. Voevodskaya, G. McClarty, A. Gräslund, P. Nordlund,
Science 305 (2004) 245–248, https://doi.org/10.1126/science.1098419.
[14] M. Högbom, Metallomics 3 (2011) 110–120, https://doi.org/10.1039/
C0MT00095G.
4. Conclusions
[15] J.A. Cotruvo, J. Stubbe, Biochemistry 50 (2011) 1672–1681, https://doi.org/10.
1021/bi101881d.
We have presented a study of three different dinuclear complexes
containing iron and/or manganese. In terms of stability of the com-
plexes towards metal scrambling in the M(II)-M(II) oxidation state, the
Fe-Fe complex 1 is the most stable and is not altered by the presence of
manganese ions, while the Mn-Mn complex 2 is converted to the Fe-Mn
complex (3) by the addition of one equivalent of Fe(II) salts. This places
the heterometallic complex as the intermediate in terms of stability. In
cyclic voltammetry the Fe-Mn complex 3 shows some similarities to
both the Fe-Fe and Mn-Mn complexes but has a more complex elec-
trochemical behavior than the homobimetallic reference complexes.
The first oxidation wave of 3 is not electrochemically reversible and is
probably accompanied with a chemical change in the coordination of
the bridging acetate ligands, something that is not observed for the
homobimetallic complexes. Also, the reactivity towards oxygen, shows
that the heterometallic complex gains new reactivity compared to the
homometallic complexes. The Mn-Mn complex is virtually unreactive
towards oxygen, while both the Fe-Fe complex and the Fe-Mn complex
react with oxygen, with the heterobimetallic Fe-Me complex doing so
much more rapidly. Taken together, this study shows that the exchange
of one metal ion in homobimetallic complexes in otherwise identical
ligand environments can alter the chemical reactivity of the complexes.
Further studies of symmetric and asymmetric homo- and hetero-
bimetallic complexes are underway in our lab and will hopefully con-
tribute further to the understanding of the chemistry of hetero-
bimetallic iron-manganese complexes and the necessity why nature
chooses such iron-manganese containing co-factors in certain enzymes.
[16] P. Reichard, Biochem. Biophys. Res. Commun. 396 (2010) 19–23, https://doi.org/
10.1016/j.bbrc.2010.02.108.
[17] E.M. Sproviero, Dalton Trans. 46 (2017) 4724–4736, https://doi.org/10.1039/
c6dt03893j.
[18] Y. Kwak, W. Jiang, L.M.K. Dassama, K. Park, C.B. Bell, L.V. Liu, S.D. Wong, M. Saito,
Y. Kobayashi, S. Kitao, M. Seto, Y. Yoda, E.E. Alp, J. Zhao, M.J. Bollinger, C. Krebs,
E.I. Solomon, J. Am. Chem. Soc. 135 (2013) 17573–17584, https://doi.org/10.
1021/ja409510d.
[19] C. Roshick, E.R. Iliffe-Lee, G. McClarty, J. Biol. Chem. 275 (2000) 38111–38119,
https://doi.org/10.1074/jbc.m006367200.
[20] N. Voevodskaya, F. Lendzian, A. Ehrenberg, A. Gräslund, FEBS Lett. 581 (2007)
3351–3355, https://doi.org/10.1016/j.febslet.2007.06.023.
[21] W. Jiang, D. Yun, L. Saleh, E.W. Barr, G. Xing, L.M. Hoffart, M.-A.A. Maslak,
C. Krebs, J.M. Bollinger, Science 316 (2007) 1188–1191, https://doi.org/10.1126/
science.1141179.
[22] W. Jiang, L.M. Hoffart, C. Krebs, M.J. Bollinger, Biochemistry 46 (2008)
8709–8716.
[23] J.M. Bollinger Jr., W. Jiang, M.T. Green, C. Krebs, Curr. Opin. Struct. Biol. 18
(2008) 650–657, https://doi.org/10.1016/j.sbi.2008.11.007.
[24] M. Carboni, J.-M. Latour, Coord. Chem. Rev. 255 (2011) 186–202, https://doi.org/
10.1016/j.ccr.2010.08.003.
[25] C.S. Andersson, M. Högbom, Proc. Nat. Acad. Sci. USA 106 (2009) 5633–5638,
https://doi.org/10.1073/pnas.0812971106.
[26] J.J. Griese, K. Roos, N. Cox, H.S. Shafaat, R.M. Branca, J. Lehtio, A. Gräslund,
W. Lubitz, P.E. Siegbahn, M. Högbom, Proc. Nat. Acad. Sci. USA 110 (2013)
17189–17194, https://doi.org/10.1073/pnas.1304368110.
[27] J.J. Griese, R. Kositzki, P. Schrapers, R.M. Branca, A. Nordström, J. Lehtio,
M. Haumann, M. Högbom, J. Biol. Chem. 290 (2015) 25254–25272, https://doi.
org/10.1074/jbc.M115.675223.
[28] R.M. Buchanan, M.S. Mashuta, J.F. Richardson, R.J. Webb, K.J. Oberhausen,
M.A. Nanny, D.N. Hendrickson, Inorg. Chem. 29 (1990) 1299–1301, https://doi.
org/10.1021/ic00332a001.
[29] P. Karsten, A. Neves, A.J. Bortoluzzi, M. Lanznaster, V. Drago, Inorg. Chem. 41
(2002) 4624–4626, https://doi.org/10.1021/ic025674k.
[30] T.R. Holman, Z. Wang, M.P. Hendrich, L. Que Jr., Inorg. Chem. 34 (1995) 134–139,
https://doi.org/10.1021/ic00105a024.
Acknowledgements
[31] S.K. Dutta, R. Werner, U. Flörke, S. Mohanta, K.K. Nanda, W. Haase, K. Nag, Inorg.
Chem. 35 (1996) 2292–2300, https://doi.org/10.1021/ic9508334.
[32] J.A.S. Robert A Heintz, Paul S. Szalay, Amy Weigerber, Kim R. Dunbar, in: Useful
Reagents and Ligands, (Ed. D. Cocouvanis), John Wiley & Sons, Inc., 2002, pp.
75–121.
Ashleigh Castner and Timofey Liseev, Uppsala University are ac-
knowledged for help with the ICP-AES measurements. Financial support
for this work from the Swedish Research Council through a Röntgen
Ångström Cluster is gratefully acknowledged.
[33] P. Chirakul, P.D. Hampton, Z. Bencze, J. Org. Chem. 65 (2000) 8297–8300, https://
doi.org/10.1021/jo001094y.
[34] Y.-L. Wong, C.-Y. Mak, H.S. Kwan, H.K. Lee, Inorg. Chim. Acta 363 (2010)
1246–1253, https://doi.org/10.1016/j.ica.2009.12.039.
Appendix A. Supplementary data
[35] A.S. Borovik, V. Papaefthymiou, L.F. Taylor, O.P. Anderson, L. Que Jr., J. Am.
Chem. Soc. 111 (1989) 6183–6195, https://doi.org/10.1021/ja00198a032.
[36] L. Sun, M.K. Raymond, A. Magnuson, D. LeGourriérec, M. Tamm, M. Abrahamsson,
P. Huang Kenéz, J. Mårtensson, G. Stenhagen, L. Hammarström, S. Styring,
Supplementary data to this article can be found online at https://
doi.org/10.1016/j.ica.2019.03.029.
259

M. Bedin, et al.
InorganicaChimicaActa490(2019)254–260
B. Åkermark, J. Inorg. Biochem. 78 (2000) 15–22, https://doi.org/10.1016/s0162-
[44] S. Sahu, D.P. Goldberg, J. Am. Chem. Soc. 138 (2016) 11410–11428, https://doi.
0134(99)00200-7.
org/10.1021/jacs.6b05251.
[37] H.P. Gunnlaugsson, Hyperfine Interact. 237 (2016) 79, https://doi.org/10.1007/
s10751-016-1271-z.
[45] J.D. Cohen, S. Payne, K.S. Hagen, J. Sanders-Loehr, J. Am. Chem. Soc. 119 (1997)
2960–2961, https://doi.org/10.1021/ja9638871.
[38] H. Irving, R. Williams, J. Chem. Soc. (1953) 3192–3210.
[39] S. Blanchard, G. Blondin, E. Riviere, M. Nierlich, J.J. Girerd, Inorg. Chem. 42
(2003) 4568–4578, https://doi.org/10.1021/ic0301066.
[46] S.C. Payne, K.S. Hagen, J. Am. Chem. Soc. 122 (2000) 6399–6410, https://doi.org/
10.1021/ja991885l.
[47] M. Yamashita, H. Furutachi, T. Tosha, S. Fujinami, W. Saito, Y. Maeda,
K. Takahashi, K. Tanaka, T. Kitagawa, M. Suzuki, J. Am. Chem. Soc. 129 (2007)
2–3, https://doi.org/10.1021/ja063987z.
[40] P. Huang, A. Magnuson, R. Lomoth, M. Abrahamsson, M. Tamm, L. Sun, B. van
Rotterdam, J. Park, L. Hammarström, B. Åkermark, S. Styring, J. Inorg. Biochem. 91
(2002) 159–172, https://doi.org/10.1016/s0162-0134(02), 00394-x.
[41] V. Schünemann, H. Winkler, Rep. Prog. Phys. 63 (2000) 263.
[42] M. Suzuki, A. Uehara, H. Oshio, K. Endo, M. Yanaga, S. Kida, K. Saito, Bull. Chem.
Soc. Japan 60 (1987) 3547–3555, https://doi.org/10.1246/bcsj.60.3547.
[43] G. Eilers, C. Zettersten, L. Nyholm, L. Hammarström, R. Lomoth, Dalton Trans.
(2005) 1033–1041, https://doi.org/10.1039/b415148h.
[48] C. Deville, S.K. Padamati, J. Sundberg, V. McKee, W.R. Browne, C.J. McKenzie,
Angew. Chem. 55 (2016) 545–549, https://doi.org/10.1002/anie.201508372.
[49] M. Quaranta, M. Murkovic, I. Klimant, Analyst 138 (2013) 6243–6245, https://doi.
org/10.1039/c3an36782g.
[50] C. Franco, J. Olmsted III, Talanta 37 (1990) 905–909, https://doi.org/10.1016/
0039-9140(90)80251-a.
260