Reactivity of 9-anthracenecarboxylate in the presence of Mn(II) and Mn(III) ions: Biomimetic aerobic oxidative decarboxylation catalysed by a manganese(III) 2,2′-bipyridine complex

Article abstract of DOI:10.1016/j.ica.2020.119949

The stability of the 9-anthracenecarboxylate (AnthCO₂^–) in the presence of manganese ions and 2,2′-bipyridine (bpy) was explored. Two Mn(II) species were isolated: a one-dimensional system [Mn(bpy)(AnthCO₂)₂]_n (1) and a dinuclear [{Mn(bpy)(AnthCO₂)}₂(μ-OH₂)(μ-AnthCO₂)₂] compound (2), which was characterized by X-ray diffraction. Compound 2 can be obtained by hydrolysis of 1. These compounds present a specific electron paramagnetic resonance (EPR) signature and weak antiferromagnetic coupling, which is slightly stronger for the dinuclear complex (?4.8 cm^?1) than for the chain (?1.3 cm^?1). In the presence of Mn(III) ions AnthCO₂^– is not stable for long time. From the Mn(III) solution obtained by a comproportionation reaction between Mn(II) and Mn(VII) salts, first, a tetranuclear [Mn₄(μ-AnthCO₂)₆(μ-O)₂(bpy)₂(ClO₄)₂] compound was isolated, showing the antiferromagnetic properties typical of a butterfly [Mn₄O₂]⁸⁺ core (J₁ = ?41.0 cm^?1, J₂ = ?8.4 cm^?1, J₃ = ?10.6 cm^?1 and D_Mn = ?3.7 cm^?1). After that, the solution became yellow due to the reduction of the Mn(III) complex to a Mn(II) complex, and the decarboxylation and oxidation of the AnthCO₂^– ligand. This oxidation can lead to different compounds, depending on the reaction conditions (temperature and amount of O₂): an ester, derived from the AnthCO₂^– and the intermediate 10-hydroxyanthrone, and the 9,10-anthraquinone.

Full text of DOI:10.1016/j.ica.2020.119949

Journal Pre-proofs
Research paper
Reactivity of 9-anthracenecarboxylate in the presence of Mn(II) and Mn(III)
ions: Biomimetic aerobic oxidative decarboxylation catalysed by a manga‐
nese(III) 2,2'-bipyridine complex
François-Xavier Turquet, Beltzane Garcia-Cirera, Belén Albela, Laurent
Bonneviot, Mercè Font-Bardia, Montserrat Corbella
PII:
DOI:
Reference:
S0020-1693(20)31149-X
https://doi.org/10.1016/j.ica.2020.119949
ICA 119949
To appear in:
Inorganica Chimica Acta
Received Date:
Revised Date:
Accepted Date:
23 June 2020
4 August 2020
6 August 2020
Please cite this article as: F-X. Turquet, B. Garcia-Cirera, B. Albela, L. Bonneviot, M. Font-Bardia, M. Corbella,
Reactivity of 9-anthracenecarboxylate in the presence of Mn(II) and Mn(III) ions: Biomimetic aerobic oxidative
decarboxylation catalysed by a manganese(III) 2,2'-bipyridine complex, Inorganica Chimica Acta (2020), doi:
https://doi.org/10.1016/j.ica.2020.119949
This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover
page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version
will undergo additional copyediting, typesetting and review before it is published in its final form, but we are
providing this version to give early visibility of the article. Please note that, during the production process, errors
may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
© 2020 Elsevier B.V. All rights reserved.

Reactivity of 9-anthracenecarboxylate in the presence
of Mn(II) and Mn(III) ions: Biomimetic aerobic oxidative
decarboxylation catalysed by a manganese(III) 2,2'-
bipyridine complex
b]
François-Xavier Turquet,[a,
Beltzane Garcia-Cirera,[a] Belén Albela,*[b] Laurent
Bonneviot,[b] Mercè Font-Bardia,[c, d] and Montserrat Corbella*[a, e]
[a] Departament de Química Inorgànica
i
Orgànica, secció Inorgànica
Facultat de Química. Universitat de Barcelona. Martí i Franquès 1-11, 08028-
Barcelona. Spain. E-mail: montse.corbella@ub.edu
[b] Laboratoire de Chimie. Ecole Normale Supérieure de Lyon. Université de Lyon. 46
Allée d’Italie, 69364 Lyon Cedex 07, France. E-mail : belen.albela@ens-lyon.fr
[c] Unitat de Difracció de RX. Centres Científics i Tecnològics de la Universitat de
Barcelona (CCiTUB). Universitat de Barcelona. Solé i Sabarís 1-3. 08028-
Barcelona.
[d] Departament de Mineralogia, Petrologia i Geologia Aplicada. Facultat de Ciències
de la Terra. Universitat de Barcelona. Martí i Franquès s/n. 08028-Barcelona.
[e] Institut de Nanociència i Nanotecnologia (IN2UB). Universitat de Barcelona. Martí i
Franquès 1-11, 08028-Barcelona. Spain.
–
Abstract: The stability of the 9-anthracenecarboxylate (AnthCO2 ) in the presence of
manganese ions and 2,2’-bipyridine (bpy) was explored. Two Mn(II) species were
isolated: a one-dimensional system [Mn(bpy)(AnthCO2)2]n (1) and a dinuclear
[{Mn(bpy)(AnthCO2)}2(-OH2)(-AnthCO2)2] compound (2), which was characterized by
X-ray diffraction. Compound 2 can be obtained by hydrolysis of 1. These compounds
present a specific electron paramagnetic resonance (EPR) signature and weak
antiferromagnetic coupling, which is slightly stronger for the dinuclear complex (‒4.8
–
cm-1) than for the chain (‒1.3 cm-1). In the presence of Mn(III) ions AnthCO2 is not
stable for long time. From the Mn(III) solution obtained by a comproportionation
reaction between Mn(II) and Mn(VII) salts, first, a tetranuclear [Mn4(-AnthCO2)6-
O)2(bpy)2(ClO4)2] compound was isolated, showing the antiferromagnetic properties
typical of a butterfly [Mn4O2]8+ core (J1 = ‒41.0 cm-1, J2 = ‒8.4 cm‒1, J3 = ‒10.6 cm‒1 and
DMn = ‒3.7 cm‒1). After that, the solution became yellow due to the reduction of the
Mn(III) complex to a Mn(II) complex, and the decarboxylation and oxidation of the
–
AnthCO2 ligand. This oxidation can lead to different compounds, depending on the
reaction conditions (temperature and amount of O2): an ester, derived from the
–
AnthCO2 and the intermediate 10-hydroxyanthrone, and the 9,10-anthraquinone.
Keywords: 9-Anthracenecarboxylate• manganese • magnetic properties • EPR
1

1. Introduction
Manganese chemistry with carboxylate ligands has been widely studied, due to the
facility of such ligands to lead to polynuclear compounds. These polynuclear systems
can be interesting both from the magnetic point of view, and for their similarities with
some redox enzymes, such as superoxide dismutase and catalase, that are involved in
protecting cells from the damaging effects of the reactive oxygen species ROS [1]. In
particular, the manganese catalases, present in bacteria, and responsible for the
disproportionation of the H2O2 into H2O and O2, have been largely studied [2–4]. The
active site of this metalloprotein consists of a manganese dinuclear site that shifts
between two oxidation states during hydrogen peroxide decomposition: Mn(III)2 and
Mn(II)2.
In the last years, we focused our attention on mimicking the oxidized and reduced form
of the active site of the manganese catalase, preparing Mn(III) and Mn(II) dinuclear
compounds of a general formula [{Mn(NN)L}2(-O)(-RC6H4CO2)2]X2 and [{Mn(NN)(R
C6H4CO2)}2(-OH2)(-RC6H4CO2)2] (NN = 2,2’-bipiridine, bpy or 1,10-phenanthroline,
–
–
phen; X = NO3 or ClO4 ; L = H2O or X), respectively. The Mn(III) dinuclear compounds
show antioxidant activity (catalase activity) [5,6] and are also interesting from the
magnetic point of view, because it is possible to obtain systems with a ground spin
state S = 0 or S = 4 [6–10]. It is interesting to notice that both the magnetic interaction
and the catalytic properties can be modulated by changing the carboxylate bridge from
acetate or benzoate-based derivatives with ad hoc electron donor or withdrawing
substituents [5,6]. Moreover, in some cases, Mn(III) tetranuclear complexes with
butterfly topology were obtained, instead of the dinuclear complexes. On the other
hand, Mn(II) carboxylate chemistry led to the formation of species of different
nuclearity: mononuclear [11], dinuclear [11–13], trinuclear [11,12,14–21] or one-
dimensional systems [12,15,16,22,23]. The magnetic interaction on these systems is
usually weak and antiferromagnetic, and the most characteristic feature is the EPR
spectrum.
The optimization of the antioxidant activity of the Mn-catalase model compounds and
the possibility of working in aqueous media can be obtained by the insertion of the
manganese complexes into the mesoporous silica matrix, which protects the catalyst in
the reaction medium. As a result, the catalytic activity is highly increased [24,25]. Going
further in this direction, we focused our attention on investigating what happens to the
complex inside the porous material during the redox reaction. In order to trace the
complex in the matrix, we changed the carboxylate bridging ligand by another one that
–
has a fluorophore group, the 9-anthracene carboxylate (AnthCO2 ). Several studies on
manganese compounds with different oxidation states using this fluorescent
carboxylate have already been reported. In some studies Mn(II) dinuclear compounds
were used [26,27], in others trinuclear [28] and tetranuclear mixed valence systems
were considered [29]. In addition, a Mn(III) trinuclear SMM [30], and even one Mn(III)
heterometallic Mn4Ca cluster [31] were also studied.
We report herein the synthesis of two related Mn(II) compounds, a chain
[Mn(bpy)(AnthCO2)2]n (1) and a dinuclear compound [{Mn(bpy)(AnthCO2)}2(-OH2)(-
AnthCO2)2] (2) with 2,2’-bipiridine as NN bidentate ligand (bpy). The characterisations
of these complexes, and in particular their magnetic study, are presented. Moreover,
the reaction of Mn(III) ions with the 9-anthracenecarboxylate is investigated. We also
2

report a very insoluble compound postulated as a Mn(III) tetranuclear compound (3), as
well as two organic compounds derived from the oxidation of the carboxylate ligand are
reported: the 9,10-anthraquinone (4) and an ester derived from the 9-
anthracenecarboxylate and the precursor of the 9,10-anthraquinone (5). This highlights
the role of Mn(III) ions to favour the decarboxylation and fixation of molecular oxygen
–
onto the AnthCO2 ligand.
2. Experimental
2.1. Physical characterization
Chemical analysis (C, H, N) were carried out by the “Centres Científics i Tecnològics”
of the Universitat de Barcelona. Infrared spectra were recorded on KBr pellets in the
4000–400 cm‒1 range with a Termo Nicolet Avatar 330 FT-IR spectrometer. Magnetic
susceptibility measurements between 2 and 300 K were carried out in a Quantum
Design MPMS XL5 magnetometer at the “Unitat de Mesures Magnètiques” (Universitat
de Barcelona), using a field of 200 G. Pascal’s constants were used to estimate the
diamagnetic corrections for each compound. Magnetization measurements were
carried out in the 1.8-6.8 K range and at six different magnetic fields (0.5, 1.0, 2.0, 3.0,
4.0 and 5.0 T). The fits were performed by minimizing the function R = ∑
[(MT)exp‒(MT)calc]2 /∑[(MT)exp]2 and the analogous expression for M. Solid-state EPR
spectra were recorded at X-band (9.4 GHz) frequency using a Bruker ESP-300E
spectrometer from room temperature to 12 K at the “Unitat de Mesures Magnètiques”
(Universitat de Barcelona).
2.2 Synthesis
All manipulations were carried out under aerobic conditions. Reagents and solvents
were obtained from commercial sources and used without further purification.
NBu4MnO4 was prepared as described in the literature [32]. Yields were calculated from
stoichiometric reactions. Caution! Perchlorate salts of compounds containing organic
ligands are potentially explosive. Only small quantities of these compounds should be
prepared and handled using suitable protective shields.
2.2.1. Synthesis of Mn(AnthCO2)2·3H2O
9-anthracene carboxylic acid (2.0 mmol, 0.44 g) was added to a solution of
Mn(AcO)2·4H2O (1.0 mmol, 0.25 g) in ethanol (50 mL) at room temperature. A white
precipitate instantaneously formed. The suspension was stirred for 10 minutes and
then, the Mn(AnthCO2)2 was filtered and washed with ether. Yield: 84%. Anal. Calcd for
C30H18MnO4·3H2O (M. W. 551.4) (%): C, 63.34; H, 4.39. Found (%): C, 63.3; H, 4.3.
Selected IR data (KBr, main bands, cm-1): 3660 (m), 3397 (s), 3265 (s), 3173 (s), 3056
(m), 1571 (s), 1487 (m), 1444 (s), 1392 (s), 1321 (s), 1279 (m), 1014 (m), 880 (m), 862
(m), 847 (m), 796 (m), 765 (m), 733 (s), 660 (m), 600 (m), 559 (m), 523 (m), 422 (m).
2.2.2 Synthesis of [Mn(bpy)(AnthCO2)2]n (1)
2,2’-bipyridine (1.0 mmol, 0.16 g) was to a yellow solution of Mn(AnthCO2)2 (1.0 mmol,
0.55 g) in acetonitrile (40 mL), at room temperature and quickly a pale precipitate
formed. The suspension was further stirred for 10 minutes, filtered, and the solid
washed with ether. Yield: 38%. Anal. Calcd for C40H26MnN2O4 (M. W. 653.6) (%): C,
3

73.51; H, 4.01; N, 4.29. Found (%): C, 73.3; H, 4.1; N, 4.2. Selected IR data (KBr, main
bands, cm-1): 3440 (bd), 3045 (w), 1593 (s), 1566 (s), 1489 (m), 1473 (m), 1441 (s),
1394 (s), 1319 (s), 1279 (m), 1155 (m), 1016 (m), 887 (m), 866 (m), 847 (m), 760 (m),
735 (s), 665 (m), 652 (m), 638 (m), 559 (m), 525 (w), 415 (m).
2.2.3. Synthesis [{Mn(bpy)(AnthCO2)}2(-AnthCO2)2(-H2O)]·EtOH (2·EtOH)
2,2’-bipyridine (1.0 mmol, 0.16 g) was added to a suspension of Mn(AnthCO2)2 (1.0
mmol, 0.55 g) in ethanol (20 mL), at room temperature. The mixture was heated in a
microwave device from 25 to 85 ºC in 12 min. The final temperature was maintained for
2 min, and after that, the mixture was slowly cooled to 25 ºC. The mixture was filtered
to remove the remaining solid and the solution was kept undisturbed at room
temperature; after few days, some dark yellow crystals formed (yield 5%).
2.2.4. Synthesis [{Mn(bpy)(AnthCO2)}2(-AnthCO2)2(-H2O)](2)
Compound 2 could also be obtained from a solution of compound 1 (15.0 mg, 0.02
mmol) in CH3CN (50 mL) by addition of 0.5 mL of water. After few days, a yellow
microcrystalline sample of 2 was obtained. Yield: 32%. Anal. Calcd for C80H54Mn2N4O9
(M. W. 1325.2) (%): C, 72.51; H, 4.11; N, 4.23. Found (%): C, 72.3; H, 4.2; N,
4.1.Selected IR data (KBr, main bands, cm-1): 3444 (br), 3047 (m), 1600 (vs), 1562
(sh), 1439 (s), 1392 (m, 1319 (s), 1276 (w), 1015 (w), 764 (s), 737(s), 666 (w), 650 (w),
636 (w), 561 (w), 527 (w), 419 (w). The same compound was obtained using EtOH
instead of CH3CN.
2.2.5. Synthesis of [Mn4(-O)2(-AnthCO2)6(bpy)2(ClO4)2] (3)
A mixture of AnthCO2H (2.49 mmol, 0.55 g) and Mn(ClO4)2·6H2O (1.32 mmol, 0.48 g)
was dissolved in 100 mL of CH3CN. Then, 10 mL of an NBu4MnO4 (0.33 mmol, 0.12 g)
solution and 10 mL of a 2,2-bpyridine (0.83 mmol, 0.13 g) solution, both in CH3CN,
were added at the same time to the aforementioned solution, which immediately turned
dark brown. The mixture was stirred for 10 min and the black precipitate of 3 was
separated by filtration. Yield: 27%. Anal. Calcd for C110H70Mn4N4O22Cl2 (M. W. 2090.4)
(%): C, 63.20; H, 3.38; N, 2.68. Found (%): C, 63.8.4; H, 3.4; N, 2.6. Selected IR data
(KBr, main bands, cm-1): 3444(br), 3046 (w), 1589 (vs), 1562 (s), 1446 (m), 1423 (m),
1385 (s), 1319 (s), 1273 (m), 1123 (w), 1100 (w), 891 (w), 864 (w), 848 (w), 798 (w),
764 (m), 733 (s), 675 (m), 663 (m), 633 (m), 598 (m), 501 (m), 440 (m).
2.2.6. Synthesis of 9,10-anthraquinone (4) and 10-oxo-9,10-dihydroanthracen-9-
yl anthracene-9-carboxylate (5)
The dark mother solution of 3 changed colour after a week and became yellow-orange.
When this solution was kept undisturbed at room temperature in an open vessel, some
9,10-anthraquinone (4) crystals were obtained. However, when the solutions were kept
undisturbed in a closed container in the fridge, some crystals of the ester 10-oxo-9,10-
dihydroanthracen-9-yl anthracene-9-carboxylate (5) formed.
Selected IR data (KBr, main bands, cm-1) for compound 4: 3452 (br), 1726 (m), 1662
(s), 1600 (m), 1321 (m), 1305 (m), 1174 (s), 1151 (s), 996 (m), 929 (m), 882 (w), 756
(m), 726 (m), 702 (m), 679 (m).
Selected IR data (KBr, main bands, cm-1) for compound 5: 3452 (br), 1710 (s), 1672
(vs), 1600 (m), 1315 (s), 1295 (s), 1286 (m), 1196 (vs), 1168 (s), 1151 (s), 1000 (m),
932 (m), 918 (m), 893 (m), 756 (s), 731 (s), 684 (s).
4

2.3. X-ray crystallography.
The data collection for compounds 2, 4 and 5 was carried out at 100 K, on a D8
Venture system equipped with a multilayer monochromator and a Mo microfocus (λ =
0.71073 Å). The frames were integrated with the Bruker SAINT software package using
a narrow-frame algorithm. The structures were determined and refined using the
Bruker SHELXTL software package [33]. Data were corrected for absorption effects
using the multi-scan method (SADABS) [34]. Crystal data collection and refinement
are given in Table 1.
Table 1. Crystal data and structure refinement for compounds 2 and 5.
2·EtOH
5
Formula
Formula weight
Temperature / K
Wavelength / Ả
Crystal system
Space group
a / Ả
C82 H60 Mn2 N4 O10
1371.22
C29 H18 O3
414.43
100(2)
100(2)
0.71073
0.71073
Triclinic
P-1
Triclinic
P-1
16.1410(7)
17.5745(8)
24.6954(9)
108.768(2)
91.291(2)
90.413(2)
6630.4(5)
4
9.5385(7)
10.2860(7)
11.9881(8)
67.471(2)
83.343(2)
67.209(2)
1000.88(12)
2
b / Ả
c / Ả
 / deg
 / deg
 / deg
V / Ả3
Z
calcd / mg m-3
1.374
1.375

 / mm-1
0.448
0.088
F(000)
2840
432
crystal size / mm3
 range / deg
0.321x0.173x0.172
2.319 - 29.178
0.439x0.181x0.136
2.318 - 24.744
-22≤ h ≤22
-24≤ k ≤24
-33≤ l ≤33
-11≤ h ≤10
-12≤ k ≤12
-14≤ l ≤14
Index ranges
Reflections collected
368369
21676
5

35728
R(int) = 0.0592
3233
R(int) = 0.0410
Independent reflections
Data/restraints/
parameters
35728/14/1779
1.015
3233/0/289
1.041
2
Goodness-of-fit on F
R1 = 0.0364
wR2 = 0.0831
R1 = 0.0419
wR2 = 0.1047
Final R indices [I>2(I)]
R indices
(all data)
R1 = 0.0538
wR2 = 0.0914
R1 = 0.0602
wR2 = 0.1152
3. Results and Discussion
3.1. Synthesis
The behavior and reactivity of 9-anthracenecarboxlic acid in the presence of Mn(II) or
Mn(III) are different than those previously reported with other carboxylic acids. In the
literature, there are different kinds of Mn(II) compounds with carboxylate bridging
ligands and bidentate NN ligands (bpy, phen): dinuclear [{Mn(NN)(RCO2)}2(-OH2)(-
RCO2)2] [12,13], trinuclear [Mn3(NN)2(RCO2)6 [11,12,14–21] or one-dimensional
systems [Mn(NN)(RCO2)2]n [12,15,16,23]. Control over nuclearity depends on several
factors such as solvent, solubility or stability. In all cases, these compounds are
–
obtained by reaction between Mn(RCO2)2 and NN ligand, using different Mn:RCO2 :NN
molecular ratios and solvents. The use of other Mn(II) salts leads to the formation of
complexes without carboxylate ligand.
The first difference observed was in the preparation of the manganese(II) 9-
anthracenecarboxylate Mn(AnthCO2)2. In general, the manganese carboxylates
Mn(RCO2)2 are prepared by acid-base reaction between the MnCO3 and the carboxylic
acid RCO2H. However, with 9-anthracenecarboxylic acid the yield was very low.
Therefore, we changed the protocol, using Mn(MeCO2)2 as Mn(II) source, instead of
the carbonate salt. The lower pKa for for 9-AnthCO2H (3.68) compared to that of the
MeCO2H (4.75) and the volatility of the acetic acid promoted the substitution of the
carboxylate anion and Mn(AnthCO2)2 was obtained with a higher yield.
3.1.1 Synthesis of Mn(II) compounds
In order to explore the possibility of obtaining Mn(II) compounds of different nuclearity,
two Mn(AntnCO2)2:bpy ratios (3:2 and 1:1) and solvents (EtOH and MeCN) were
tested. The in situ preparation of the Mn(AnthCO2)2 upon reaction between
Mn(MeCO2)2 with either AnthCO2H or NaAnthCO2 was also tested. In all cases, the
one-dimensional system [Mn(bpy)(AnthCO2)2]n (1) and/or the dinuclear compound
[{Mn(bpy)(AnthCO2)}2(-AnthCO2)2(-H2O)] (2) were obtained. It was not easy to
control the formation of one or the other. Both compounds were poorly soluble in EtOH,
showing a slightly higher solubility in MeCN. In most cases, the obtained compound
corresponded to the one-dimensional system 1.
–
The different synthetic routes to obtain Mn(II) with 9-AnthCO2 ligand are shown in
Figure 1. It was only possible to obtain crystals suitable for X-ray diffraction analysis for
6

the dinuclear compound 2. These crystals were obtained by heating the ethanolic
reaction mixture above the boiling point of the solvent in a closed vessel, using
microwave irradiation.
The difference in the composition of 1 and 2 is just the water molecule present as
bridging ligand in the dinuclear compound 2. Therefore, in order to control and
understand the formation of these species, the one-dimensional system 1 was
dissolved in CH3CN or EtOH in the presence of water and, as expected, the solid
obtained was the dinuclear compound 2 (confirmed by EPR, see section 3.4). In fact,
1% of water sufficed to cleave the chain by hydrolysis of the carboxylate bridges. A
possible mechanism for this conversion is depicted in Figure S1 on ESI.
–
Figure 1. Synthetic routes to obtain Mn(II) complexes with 9-AnthCO2 ligand.
Compounds 1 and 2 were easily recognized from their specific IR signature in two
regions of the spectrum. In the 1600-1550 cm‒1 region, characteristic of the asymmetric
vibration of the carboxylate group ((CO2)a), the dinuclear compound 2 exhibits a single
strong band at 1600 cm-1, while the one-dimensional system 1 presents two ill-resolved
bands of similar intensity, at 1600 and 1575 cm‒1. It is interesting to mention that
compound 2 shows two bridging carboxylate ligands coordinated in a syn-syn mode,
and two monodentate carboxylate ligands with the dangling oxygen atom interacting by
hydrogen bond with the water bridging ligand. In spite of the different coordination
mode of the carboxylate groups, the C-O distances are similar (1.252 Ả, 1.255 Ả and
1.257 Ả, respectively) (see section 3.2); this fact can explain the presence of only one
band at 1600 cm‒1. For one-dimensional [Mn(RCOO2)2(NN)]n systems the carboxylate
bridges are in a syn-anti coordination mode, and both the C-Osyn and C-Oanti distances
are significatively different in most of the cases [12,15]. Unfortunately, it was not
possible to obtain suitable crystals for X-ray diffraction of compound 1, which is
postulated as a one-dimensional system. The other significant difference was observed
in the 600-700 cm-1 region, where three bands were detected at 638, 652 and 665 cm-1
for both compounds, with different relative intensities (Figure S2, on ESI). In fact, we
7

shall see below that the most relevant difference between these two compounds was
derived from the EPR spectrum (section 3.4).
3.1.2 Synthesis of tetranuclear Mn(III) compound
In our group, we synthesized a significant number of dinuclear ([{Mn(NN)L)2(-O)(-
RCO2)2]X2) and tetranuclear ([Mn4(-O)2(-RCO2)6(NN)2L2]X2) Mn(III) compounds with
carboxylate bridging ligands, by a comproportionation reaction between a Mn(II) salt
–
and MnO4 , in the presence of carboxylic acid and NN bidentate ligands (bpy or phen).
Attempts to obtain the analogous compounds with 9-anthracenecarboxylate presented
–
several difficulties. In order to obtain the dinuclear complex, the usual Mn:RCO2 :NN
ratio is 1:1:1, while for the tetranuclear complex, the usual ratio is 2:3:1. In general, for
both ratios the compound obtained is only one of them, which is the most insoluble
and/or stable species. However, It has been reported by R. Ruiz et al. that the
dinuclear complex with acetate and phen as NN blocking ligand,
[Mn2O(AcO)2(H2O)2(phen)2]2+, can be converted in solution to the tetranuclear entity
[Mn4O2(AcO)7(phen)2]+ [35]. In this work, Mn(ClO4)2 was used as Mn(II) source and bpy
–
as a capping amine. Both Mn:RCO2 :NN ratios were tested with analogous results: the
immediate precipitation of a dark brown powder (3).
The solid 3 was a pure compound that exhibits two characteristic IR bands assigned to
the asymmetric and symmetric vibrations modes of the carboxylate group ((CO2)a and
(CO2)s), separated by ~200 cm-1, which is consistent with a 1,3 coordination mode of
–
the AntCO2 ligand [36]. Moreover, there are the bands of the bpy ligand in the 1400-
1600 cm-1 range, and the characteristic bands of the perchlorate ion at 1100 and 625
cm-1 (Figure S3, on ESI). Therefore, all these ligands and ions are present in
compound 3. Moreover, CHN analysis fits better for a tetranuclear entity than for a
dinuclear one. Its low solubility in MeCN and the negligible ionic conductivity in solution
suggest that compound 3 is neutral. Therefore, we can expect that the perchlorate
anions act as ligands that are coordinated to the Mn(III) ions. The analysis of the
magnetic data, specifically the shape of the MT vs T graph, is characteristic of
tetranuclear complexes with a butterfly topology (section 3.3). With all this information
in mind, the proposed formula for this compound is [Mn4(-AnthCO2)6(-
O)2(bpy)2(ClO4)2]. Unfortunately, it was not possible to obtain crystals suitable for X-ray
diffraction. Only small crystals of poor quality were obtained.
One of the problems in obtaining good crystals was the low solubility of the Mn(III)
complex. The second problem was the side reaction between the Mn(III) and the 9-
anthracenecarboxylate. After removing compound 3 from the dark brown mother liquor,
we kept the solution undisturbed, in order to obtain crystals suitable for X-ray
diffraction. We tested different mixtures of solvents, slow evaporation and refrigeration.
In all the cases, no crystallization or precipitation of the compound 3 was observed.
Moreover, in a week, the solution turned orange and some yellow or yellow-orange
crystals were obtained, which correspond to different compounds. When the mother
solution was kept still at room temperature in an open vessel, the crystals formed
corresponded to 9,10-anthraquinone (4), while when the mother solution was kept cold
in the fridge in a closed recipient, an ester (5) crystallized, the 10-oxo-9,10-
dihydroanthracen-9-yl anthracene-9-carboxylate. When the reaction was carried out
under N2, the solution also became orange-yellow and the solid obtained contained
organic non identified compounds together with the Mn(II) dinuclear compound (2).
8

Figure 2 shows the evolution of the Mn(III) solution in different conditions, where the
most important effect is the amount of oxygen in the media.
–
Figure 2. Evolution of a solution containing Mn(III), bpy and AnthCO2 , in function of the
presence or absence of O2.
The organic compounds, 4 and 5, show significant differences in the IR spectrum in the
1400-1100 cm−1 and 800-600 cm−1 regions (Figure S4, on the ESI), which helped us to
identify which compound formed in each case. Moreover, these spectra were also
different compared to that of the Mn(II) dinuclear compound, which shows the a(CO2)
vibration around 1600 cm-1. Both 4 and 5 are obviously oxidation products of the
–
AnthCO2 ligand. Moreover, the low yield of the synthesis of the tetranuclear compound
–
3 could be due to the oxidation of some AnthCO2 as a side reaction.
In order to see if this oxidation was due to the powerful oxidant character of the starting
–
reactant MnO4 , a mixture of this anion with AnthCO2H in MeCN was maintained in
agitation for one day, and no change of the purple colour of the permanganate ion was
9

–
observed. Thus, the oxidation of the AnthCO2 could be attributed to the Mn(III) ion
–
present in solution, probably as a Mn(III) complex containing bpy and AnthCO2
ligands. It is interesting to note, that in the absence of bpy the oxidation of the
–
AnthCO2 was not observed. The formation of the Mn(II) dinuclear compound 2 in the
–
absence of oxygen clearly attests to an inner sphere oxidation of the AnthCO2 by the
–
Mn(III) ion. Indeed, the Mn(II) ion in the presence of the remaining AnthCO2 and bpy
ligands, led to the obtention of the low-soluble dinuclear compound 2.
The role of dinuclear Mn(III) complexes with [Mn2(-O)(-AcO)2]2+ core on the oxidative
carbon-carbon cleavage have been reported in the literature [37]. Moreover, several
Mn(III) compounds are reported to be efficient for oxidative decarboxylation of
carboxylic acids [38–40]. Taking into account the present results, we suggest that both
in the absence or in the presence of O2, the first step could be the decarboxylation of
–
the AnthCO2 ion with the formation of CO2, with the concomitant reduction of the
Mn(III) complex present in the solution to a Mn(II) species. Moreover, the anthracene
group can react with molecular oxygen in the presence of electron-transfer compounds,
yielding anthracene endoperoxide (A, Figure 3). The latter can be further converted
after electronic rearrangement into 10-hydroxyanthrone (B, Figure 3). This compound,
B, likely undergoes a second oxidation reaction, producing anthraquinone [41,42].
Accordingly, in our case, when dioxygen is present, we propose that after the
decarboxylation, the  system of the anthracene could interact with dioxygen, forming
the endoperoxide A, which, after the electronic rearrangement, leads to the formation
of the intermediate B. This intermediate should be the starting step to obtain
compounds 4 and 5 (Figure 3).
Figure 3. Proposed pathway for the oxidative decarboxylation of 9-
anthracenecarboxylate.
10

–
For one side, B can react with remaining intact AnthCO2 , leading to the formation of
the ester (5). On the other side, in the presence of some Mn(III) complex in the
solution, B could be oxidized to the anthraquinone 4, with the consequent formation of
a soluble Mn(II) species. The Mn(II) species generated could act as an electron-donor
–
to the decarboxylated AnthCO2 , assisting in the fixation of a new dioxygen molecule.
So, in presence of oxygen, a Mn(III)/Mn(II) couple may catalyze the fixation of the
–
molecular oxygen and the oxidation of the AnthCO2 , leading to the formation of the
ester (5) and the quinone (4). Hence, the Mn(III) complex in solution acts like a
decarboxylase-dioxygenase, as it proceeds by successive decarboxylation, oxidation
and insertion of two oxygen atoms from the molecular oxygen to the substrate. This is
consistent with formation of 5 via a slow oxidation process at low temperature without
excess of oxygen, and the formation of the anthraquinone 4 by deeper oxidation in
excess oxygen.
3.2. Crystal structures
The crystal structure of compound 2·EtOH is shown in Figures 4 and S5 (ESI). The unit
cell contains two dinuclear complexes and two ethanol molecules. The Mn(II) ions are
–
bridged by two AnthCO2 ligands coordinated in a syn-syn conformation, and one aqua
ligand. The hexacoordination of each manganese ion is completed by a bpy ligand and
‒
one monodentate AnthCO2 . The monodentate carboxylate ligand shares hydrogen
bonds with the aqua bridge ligand, while the bpy rings interact via -stacking between
the two subunits. Selected interatomic distances and angles are shown in Table 2.
Figure 4. Crystal structure of the two subunits of compound 2·EtOH
11

Table 2. Selected bond lengths (Å) and angles (º) for compound 2·CH3CH2OH
Mn(1A)-O(3A)
2.1152(10)
2.1242(10)
2.1448(10)
2.2408(11)
2.2706(10)
2.2821(12)
2.0983(10)
2.1129(10)
2.1399(10)
2.2419(11)
2.2557(10)
2.2945(12)
3.6200(7)
106.22(4)
107.64(4)
-90.24(5)
Mn(1B)-O(5B)
Mn(1B)-O(1B)
Mn(1B)-O(3B)
Mn(1B)-N(2B)
Mn(1B)-O(9B)
Mn(1B)-N(1B)
Mn(2B)-O(4B)
Mn(2B)-O(6B)
Mn(2B)-O(7B)
Mn(2B)-N(4B)
Mn(2B)-O(9B)
Mn(2B)-N(3B)
Mn(1B)···Mn(2B)
2.1190(10)
2.1447(9)
2.1566(10)
2.2340(11)
2.2457(9)
2.2801(12)
2.1168(10)
2.1188(10)
2.1757(10)
2.2438(11)
2.2555(10)
2.2736(11)
3.6333(7)
Mn(1A)-O(5A)
Mn(1A)-O(1A)
Mn(1A)-N(1A)
Mn(1A)-O(9A)
Mn(1A)-N(2A)
Mn(2A)-O(6A)
Mn(2A)-O(4A)
Mn(2A)-O(7A)
Mn(2A)-N(3A)
Mn(2A)-O(9A)
Mn(2A)-N(4A)
Mn(1A)···Mn(2A)
Mn(1A)-O(9A)-Mn(2A)
Mn(1B)-O(9A)-Mn(2B)
O1A-Mn1A-Mn2A-O7A
O1B-Mn1B-Mn2B-O7B
80.43(5)
The octahedra around the Mn(II) ions are compressed in the monodentate ligand
direction. The degree of distortion can be evaluated as for Mn(III) dinuclear complexes.
Approximate values of the octahedron axis length can be found by addition of Mn-
ligand distances. Arbitrarily, we assigned the x axis in the water ligand direction and the
z axis in the monodentate ligand direction. x = d(Mn-Ow) + d(Mn-N), y = d(Mn-Oc) +
d(Mn-N) and z = d(Mn-Oc) + d(Mn-Oc*), where Oc represents the oxygen of the
carboxylate bridge, and Oc* corresponds to the oxygen atom of the monodentate
carboxylate. The values of the axes and the distortion parameters, elongation () and
rhombicity () calculated as reported previously for Mn(III) dinuclear compounds (eq
1)[7] are shown on Table 3.
 = 푧 ― 푥푦 and  = 푦 ― 푥
eq. 1
푥푦
푥
The negative value of  is indicative of a compression on the z direction. The
disposition of the compression axis (z) is close to the orthogonality. The torsion angle
(Oc* -Mn-Mn-Oc*) is ~ 85º.
12

Table 3. x, y and z axis lengths,* elongation () and rhombicity () parameters for the
two subunits of compound 2.
x (Ả)
4.50
4.49
y (Ả)
4.41
4.41
z (Ả)
4.25
4.28
 
-4.64
-3.90
 
-2.17
-1.67
Sub. A
Sub B
* Average values of the two Mn(II) ions on each subunit
The crystal structure of 9,10-anthraquinone (4) is well known and several data are
available on the CCDC. Therefore, only the preliminary resolution was done for this
compound (Figure S5 on ESI). The crystal structure of the new compound 5 is shown
in Figure 5. The anthracene and anthracene-derivative group are not coplanar; there is
an angle of 79.77º between the planes. It is worth highlighting that in this compound,
the C-O distances in the carboxylate group are well differentiated (d(C15-O2) = 1.206
Å and d(C15-O1) = 1.346 Å), while in the dinuclear compound 1, all the C-O distances
are in the 1.24‒1.26 Å range, without a significant difference between the monodentate
and the bridging carboxylate ligands. This fact could be due to the hydrogen bond
between the non-coordinated oxygen atom of the monodentate carboxylate and the
aqua ligand. The value of the distance d(C8-O3) is 1.223 Ả, similar to that found for the
quinone (4) (1.221 Ả).
Figure 5. Crystal structure of compound 5.
3.3. Magnetic properties
Magnetic susceptibility measurements for compounds 1, 2 and 3 were carried out from
room temperature to 2 K. For the Mn(II) dinuclear compound 2, the MT values
decrease as the temperature falls, indicative of an antiferromagnetic coupling between
the Mn(II) ions (Figure 6). Moreover, the M vs T plot shows a maximum at 20 K. The
data were fitted with the PHI program [43] and the best fit (reported considering the H =
‒J S1·S2 spin Hamiltonian) corresponds to J = ‒4.8 cm-1 and g = 2.1, with R = 6.5 10-5
(MT ) and 2.1 10-5 cm-1 (M).
13

Figure 6. MT vs T and M vs T (inset) plots for compound 2. The solid line corresponds
to the best fit of the experimental data.
For the one-dimensional system 1, the shape of the MT vs T and M vs T plots are
similar to that of the dinuclear compound 2 (Figure 7), but in this case, the maximum of
M appears at lower temperature (7 K). The experimental data were fitted with the
Fisher equation [44] for a one-dimensional system of spins S = 5/2. (H = ‒ijJij Si·Sj).
The best-fit parameters were J = ‒1.3 cm‒1 and g = 2.0, with R = 6.1 10‒4 (MT) and 9.0
10‒4 (M).
Figure 7. MT vs T and M vs T (inset) for the one-dimensional system 1. The data are
reported per Mn(II) ion.
Comparison between the magnetic coupling of the dinuclear compound (2) and the
chain (1) shows a higher antiferromagnetic interaction for 2, in agreement with the
presence of an extra aqua bridge ligand. There are some other dinuclear Mn(II)
compounds described in the literature with the same [Mn2(-H2O)(-RCOO)2]2+ core
and the structural and magnetic data are collected in Table 4. The magnetic coupling
14

for this kind of compound is weak, with J values ranging from ‒2.0 to ‒5.8 cm‒1. The
difference between the largest and the smallest Mn-Ow-Mn angles is 13º and in
general, the highest Mn···Mn distance is found for compounds with larger Mn-Ow-Mn
angles. However, no significant magneto-structural correlation was found. The
difference between the magnetic coupling constant is small. Indeed, the 5 d orbitals of
the Mn(II) ions are involved in the interaction. Consequently, the net magnetic behavior
depends on small structural changes propitiated by the terminal ligands, together with
the carboxylate bridges: conformation (syn-anti or syn-syn), steric and/or electronic
factors of the substituents.
The one-dimensional systems with only carboxylate bridges [Mn(NN)(-RCOO)2]n are
scarce and in most cases show a weaker antiferromagnetic coupling than the dinuclear
compounds (Table 5).
Table 4. Selected structural and magnetic data for Mn(II) dinuclear compounds with
[Mn2(-H2O)(-RCO2)2]2+ core.
 b/ deg
114.6
110.0
110.2
114.4
108.3
101.9
108.8
110.4
105.7
106.9
Compound
da/ Å
3.739
3.621
3.595
3.777
3.618
3.484
3.614
3.709
3.584
3.627
J c /cm‒1
–3.3
–5.8
–5.5
–2.5
–2.0
–3.6
–3.6
–2.6
–4.9
–4.8
ref
[Mn2(F5C2CO2)4(H2O)3L2]
[Mn2(MeCO2)4(H2O)(tmeda)2]
[Mn2(H2O)(Me2bpy)2(piv)4]
[Mn2(MeCO2)4(H2O)2(Im)4]
[Mn2(MeCO2)5(H2O)(py)4]─
[Mn2(4-ClC6H4CO2)4(H2O)(phen)2]
[Mn2(butca)(H2O)5]
[45]
[13]
[13]
[46]
[47]
[12]
[46]
[48]
[27]
*
[Mn2(H2O)(tolf-O,O’)2(tolf-O)2(py)4]
[Mn2(AnthCO2)4(phen)4(H2O)
[Mn2(bpy)2(AnthCO2)4(H2O)] (2)
a
b
c
d(Mn···Mn); ((Mn-Ow-Mn); H = ‒JS1·S2; * this work. Abbreviations: L = 2-ethyl-
4,4’,5,5’-tetramethyl-3-oxo-4,5-dihidro-1H-imidazolyl-1-oxyl; tmeda N,N,N,N-
=
tetramethylethylenediamine; piv = pivalate; Me2bpy = 4,4’-dimethyl-2,2’-bipyridine; Im
= imidazole; py = pyridine; butca = 1,2,3,4-butanetetracarboxylate; tolf = tolfenamate;
phen = 1,10-phenanthroline.
15

Table 5. Magnetic coupling constant for Mn(II) one-dimensional systems
[Mn(NN)(RCO2)2]n.
Compound
J a / cm‒1
–1.8
Ref
[15]
[12]
[15]
[16]
[23]
*
[Mn(bpy)(3-ClC6H4CO2)2]n
[Mn(phen)(3-ClC6H4CO2)2]n
[Mn(bpy)(4-ClC6H4CO2)2]n
[Mn(phen)(ClCH2CO2)2]n
[Mn(phen)(FcC6H4CO2)2]n
–1.7
–0.7
–0.9
–6.5
[Mn(bpy)(AnthCO2)2]n (1)
–1.3
a
H = ‒ijJij Si·Sj; Abbreviations: phen = 1,10-phenanthroline; Fc = ferrocene. * This
work
For the Mn(III) tetranuclear compound 3 MT vs T plot is shown in Figure 8. The MT
value at room temperature is 8.5 cm3 mol-1 K, which is quite below the expected value
for four Mn(III) ions (12.0 cm3 mol-1 K). MT decreases with temperature almost linearly
until ~50 K and below this temperature it decreases most drastically until ~2.7 cm3 mol-
1
K at 4 K. The shape of this graph is characteristic of tetranuclear compounds with a
butterfly structure [49–54].
Figure 8. MT vs T plot for compound 3. The solid line is the best fit of the experimental
data.
To acquire more information about compound 3, magnetization (M) data were collected
in the 1.8-6.8 K range, applying magnetic fields between 0.5 and 5.0 T. M/N vs HT‒1
plot is shown in Figure 9. The non-superposition of the various isofield lines is
indicative of significant zero-field splitting (ZFS). The maximum M/N value is around
4.5 N
16

Figure 9. M/N vs HT‒1 plot for compound 3. The solid lines are the best fits of the
experimental data.
The MT vs T and M/N vs HT‒1 data were fitted simultaneously with the PHI program
[43]. Three different Mn···Mn exchange pathways may be considered (Figure 10): the
interaction through two oxo bridges (J1), and the magnetic interaction between the
terminal-central Mn(III) ions (J2 and J3). In some of the compounds reported in the
literature, the experimental data can be fitted by considering an average of these
magnetic coupling constants [50–53].
Figure 10. Scheme of the possible Mn···Mn exchange magnetic pathways in the
tetranuclear compound 3.
The Heisenberg spin Hamiltonian considered was  = ‒J1(S1·S2) ‒ J2(S1·S3+ S2·S4) ‒
J3(S1·S4+S2·S3), where S1 = S2 = S3 = S4. In this kind of compound, the Jahn-Teller
axes of the terminal Mn(III) ions are mostly orthogonal. Thus, a rotation of the
reference frame, of -85º for Mn3 and +85º for Mn4, through the Euler angle  was
considered. The best fit led to J1 = ‒41.0 cm-1, J2 = ‒8.4 cm‒1, J3 = ‒10.6 cm‒1 and DMn
= ‒3.7 cm‒1, with R = 2.5 10‒4 ( T) and R = 8.4 10‒4 (M/N). Attempts to fit the
M
experimental data with an average of J2 and J3 were unsuccessful.
For all the Mn(III) tetranuclear compounds reported in the literature of the butterfly
family, J1 is the strongest antiferromagnetic interaction, ranging between ‒28 and ‒55
cm‒1, while J2 and J3 are weaker, ranging from 0 to ‒16 cm‒1 (Table 6). In most cases
the fit of the experimental data was achieved considering J2 = J3. However, the
‒
magnetic behavior of two compounds with 1,10-phenanthroline and 4-tBuC6H4CO2 or
17

‒
4-MeOC6H4CO2 as bridging ligand, are described with different J2 and J3. The spin
ground state depends on the J2/J3 ratio. Indeed, the energy of the spin states is a
function of the J2/J3 ratio, and for values around 0,7 (for J1 = ‒45. 6 cm‒1 and J3 = ‒15
cm‒1) the spin states ST = 0, 1, 2 and 3 are practically degenerated [49].
The results obtained for compound 3 were in the expected range, and it was necessary
to differentiate J2 of J3 to obtain a good fit of the experimental data. Seven practically
degenerated energy levels (energy gap of 10‒12 cm‒1) were postulated for this system,
with the following total spin values: ST = 3 (two), 2 (two), 1 (two), and 0 (one). The first
excited state, ST = 0 was found at 1.7 cm‒1 from the ground state. For this compound
the J2/J3 ratio was 0.79, in agreement with the minor J3 value.
Table 6. Magnetic coupling constants for compounds with a [Mn4O2(RCO2)6]2+ core.
J1 / cm‒1
‒47.0
‒49.2
‒55.0
‒33.0
‒51.4
‒29.8
‒45.5
‒43.0
‒28.4
‒29.0
‒28.4
‒31.0
‒41.0
J2 / cm‒1
‒15.6
‒10.6
‒0.8
J3 / cm‒1
‒15.6
‒10.6
‒0.8
ref
[Mn4O2(MeCO2)7(bpy)2]+
[50]
[51]
[52]
[53]
[53]
[54]
[49]
[49]
[55]
[55]
[55]
[55]
*
[Mn4O2(MeCO2)7(pic)2]‒
[Mn4O2(C6H4CO2)7(dpm)2]
[Mn4O2(NO3)(EtCO2)6(bpy)2]+
[Mn4O2(EtCO2)7(bpya)2]+
‒3.4
‒3.4
‒6.6
‒6.6
[Mn4O2(py)2(MeCO2)6(dbm)2]
[Mn4O2(4-MeOC6H4CO2)7(phen)2]+
[Mn4O2(4-tBuC6H4CO2)6(phen)2]2+
[Mn4O2(ThCO2)9(EtOH)] ‒
‒10.0
‒15.1
‒14.7
‒9.2
‒10.0
‒4.4
‒8.2
‒9.2
[Mn4O2(C6H4CO2)9(H2O)]‒
‒9.7
‒9.7
[Mn4O2(4-MeC6H4CO2)9(H2O)]‒
[Mn4O2(3,5-(Me)2C6H3CO2)9(H2O)]‒
[Mn4(-AnthCO2)6(-O)2(bpy)2(ClO4)2].
‒9.4
‒9.4
‒9.8
‒9.8
‒8.4
‒10.6
Abbreviations: bpy = 2,2’-bipyridine; bpya = bis(2-pyridylamine; py = pyridine; dbmH
= dibenzoylmethane; phen = 1,10-phenanthroline; Th = thiophene; *this work.
3.4. EPR spectra
The EPR spectra of compounds 1 and 2 were recorded on powdered samples at
different temperatures. At room temperature the spectra of the dinuclear compound
and the 1D system showed a broad band around g ~2 with a peak-to-peak linewidth
(Hp-p) of 558 G for the dinuclear compound (2) and of 382 G for the chain (1). At low
temperature, and particularly below 50 K, the spectrum of 2 was obviously structured at
low and high field. This is characteristic of Mn(II) dinuclear compounds with weak
antiferromagnetic coupling (Figure 11). By contrast, any change in the shape of the
18

band appeared for 1 at low temperature (Figure S6 on ESI), providing us a convenient
tool to differentiate both compounds.
Figure 11. X-band EPR spectra on a powdered sample of compound 2 at different
temperatures.
We previously mentioned that compound 1 could be converted into 2 by hydrolysis in
the presence of some water. EPR spectroscopy was used to monitor this evolution.
Figure 12 shows the EPR spectra on powdered samples of the one-dimensional
system (1), the dinuclear compound (2) and the solid obtained after treatment of 1 with
CH3CN/H2O or EtOH/H2O, indicated as 2*. The spectrum of the solid obtained after
hydrolysis of 1 (2*) is very similar to that of the dinuclear compound 2. Indeed, the
appearance of new features at low and high fields is characteristic of Mn(II) dinuclear
complexes. Based on this result we propose that the presence of humidity, or some
amount of water, led to the hydrolysis of 1 and the formation of 2.
19

Figure 12. EPR spectra at 100 K on powdered samples for 1, 2, and 2* (compound
obtained after treatment of 1 with CH3CN/H2O or EtOH/H2O).
Analyzing of the EPR spectrum of dinuclear compounds is a challenging task, since it
combines the effect of both the ZFS parameter of the Mn(II) ions DMn, and the magnetic
coupling constant J [11]. For compound 1, the magnetic interaction between the
manganese centers was slightly stronger than what was found in this previous study,
leading to different contributions of the spin states S =1 and S = 2 and a different EPR
line shape. The simulation calculated here using the PHI program [43] is meant to fit
with the spectrum recorded at 12 K. The orthogonal orientation of the distortion axes of
the Mn(II) ions was taken into account, by considering the Euler angle  of ‒45º for
Mn1 and 45º for Mn2 and including the magnetic interaction between the Mn(II) ions
found from the fit of the magnetic data (‒4.8 cm‒1) and the ZFS parameters of the DMn
and EMn. The g value was fixed to 2.0. The best simulation was obtained with DMn
0.28 cm‒1 and EMn = ‒0.03 cm‒1 (Figure 13).
=
Figure 13. X-band EPR spectrum of compound 1 on a powdered sample at 12 K (red
line), and simulation (dotted black line).
4. Conclusion
Two Mn(II) compounds were synthesized using 2,2’-bipyridine and 9-
anthracenecarboxylate as ligands: a dinuclear compound with three bridges located
between the two metal ions (-(H2O)(-AnthCO2)2) and a chain with only two bridges
((-AnthCO2)2) connecting each Mn(II) neighbor. The one-dimensional compound can
be converted into the dinuclear by hydrolysis in presence of some water. EPR
spectroscopy in solid state was the best tool to differentiate these two Mn(II)
compounds.
A tetranuclear Mn(III) compound, [Mn4(-AnthCO2)6(-O)2(bpy)2(ClO4)2], was prepared
by comproportionation between the Mn(ClO4)2 complex and a permanganate salt. Its
butterfly topology was determined from the magnetic signature. Its spontaneous
decomposition during crystallization is consistent with a catalytic decarboxylase
dioxygenase activity by the Mn(III) species. Depending on the amount of dioxygen
20

present in the solution, it yields either an ester derived from the 9-
anthacenecarboxylate, or the anthraquinone.
Acknowledgments
This work was supported by the Spanish Ministerio de Economia y Competitividad
through the CTQ 2015-63614-P and PGC2018-094031-B-I00 projects. F.X. Turquet
gratefully thanks the Auvergne - Rhône Alpes Region and the IDEXLYON of the
Université de Lyon for their support in the collaboration between the Ecole Normale
Supérieure de Lyon and the Universitat de Barcelona.
Appendix A. Supplementary data
Crystallographic data for compounds 2 and 5 have been deposited with the Cambridge
Crystallographic Data Center, CCDC No 1879985 (for 2) and 1879986 (for 5). These
data can be obtained free of charge from CCDC.
Electronic supplementary information contains a proposed mechanism for the
conversion of 1 to 2, IR spectra comparatives, a picture of 4, and the EPR spectra of 1
at different temperatures.
References
[1]
[2]
[3]
[4]
A. Bhattacharyya, R. Chattopadhyay, S. Mitra, S.E. Crowe, Physiol. Rev. 94
(2014) 329.
V. V. Barynin, M.M. Whittaker, S. V. Antonyuk, V.S. Lamzin, P.M. Harrison, P.J.
Artymiuk, J.W. Whittaker, Structure 9 (2001) 725.
M.M. Whittaker, V. V. Barynin, S. V. Antonyuk, J.W. Whittaker, Biochemistry 38
(1999) 9126.
S. V. Khangulov, V. V. Barynin, S. V. Antonyuk-Barynina, BBA - Bioenerg. 1020
(1990) 25.
[5]
[6]
V. Gómez, M. Corbella, Eur. J. Inorg. Chem. 2012 (2012) 3147.
M. Corbella, G. Fernández, P. González, M. Maestro, M. Font-Bardia, H.
Stoeckli-Evans, Eur. J. Inorg. Chem. 2012 (2012) 2203.
[7]
L. Escriche-Tur, M. Font-Bardia, B. Albela, M. Corbella, Dalt. Trans. 45 (2016)
11753.
[8]
[9]
V. Gómez, M. Corbella, G. Aullón, Inorg. Chem. 49 (2010) 1471.
V. Gómez, M. Corbella, O. Roubeau, S.J. Teat, Dalt. Trans. 40 (2011) 11968.
[10] G. Fernández, M. Corbella, G. Aullón, M.A. Maestro, J. Mahía, Eur. J. Inorg.
Chem. 3 (2007) 1285.
[11] L. Escriche-Tur, M. Font-Bardia, B. Albela, M. Corbella, Dalt. Trans. 46 (2017)
2699.
[12] V. Gomez, M. Corbella, Eur. J. Inorg. Chem. (2009) 4471.
[13] S.B. Yu, S.J. Lippard, I. Shweky, A. Bino, Inorg. Chem. 31 (1992) 3502.
[14] V. Tangoulis, D.A. Malamatari, K. Soulti, V. Stergiou, C.P. Raptopoulou, A.
Terzis, T.A. Kabanos, D.P. Kessissoglou, Inorg. Chem. 35 (1996) 4974.
[15] B. Albela, M. Corbella, J. Ribas, I. Castro, J. Sletten, H. Stoeckli-Evans, Inorg.
21

Chem. 37 (1998) 788.
[16] G. Fernández, M. Corbella, J. Mahía, M.A. Maestro, Eur. J. Inorg. Chem. 2002
(2002) 2502.
[17] S.G. Baca, Y. Sevryugina, R. Clérac, I. Malaestean, N. Gerbeleu, M.A.
Petrukhina, Inorg. Chem. Commun. 8 (2005) 474.
[18] C.J. Milios, T.C. Stamatatos, P. Kyritsis, A. Terzis, C.P. Raptopoulou, R.
Vicente, A. Escuer, S.P. Perlepes, Eur. J. Inorg. Chem. 2004 (2004) 2885.
[19] Z. Chen, Y. Ma, F. Liang, Z. Zhou, J. Organomet. Chem. 693 (2008) 646.
[20] A. Escuer, B. Cordero, X. Solans, M. Font-Bardia, T. Calvet, Eur. J. Inorg.
Chem. 2008 (2008) 5082.
[21] S. Menage, S.E. Vitols, P. Bergerat, E. Codjovi, O. Kahn, J.J. Girerd, M. Guillot,
X. Solans, T. Calvet, Inorg. Chem. 30 (1991) 2666.
[22] S.G. Baca, I.L. Malaestean, T.D. Keene, H. Adams, M.D. Ward, J. Hauser, A.
Neels, S. Decurtins, Inorg. Chem. 47 (2008) 11108.
[23] H. Hou, L. Li, Y. Zhu, Y. Fan, Y. Qiao, Inorg. Chem. 43 (2004) 4767.
[24] L. Escriche-Tur, M. Corbella, M. Font-Bardia, I. Castro, L. Bonneviot, B. Albela,
Inorg. Chem. 54 (2015) 10111.
[25] B. Garcia-Cirera, M. Corbella, L. Bonneviot, B. Albela, Microporous Mesoporous
Mater. 261 (2018) 150.
[26] R.Q. Zou, C. Sen Liu, X.S. Shi, X.H. Bu, J. Ribas, CrystEngComm 7 (2005) 722.
[27] C. Sen Liu, E.C. Sañudo, L.F. Yan, Z. Chang, J.J. Wang, T.L. Hu, Transit. Met.
Chem. 34 (2009) 51.
[28] N. Taleb, V.J. Richards, S.P. Argent, J. Van Slageren, W. Lewis, A.J. Blake,
N.R. Champness, Dalt. Trans. 40 (2011) 5891.
[29] C.C. Beedle, C.J. Stephenson, K.J. Heroux, W. Wernsdorfer, D.N. Hendrickson,
Inorg. Chem. 47 (2008) 10798.
[30] D.I. Alexandropoulos, A.M. Mowson, M. Pilkington, V. Bekiari, G. Christou, T.C.
Stamatatos, Dalt. Trans. 43 (2014) 1965.
[31] A.A. Alaimo, D. Takahashi, L. Cunha-Silva, G. Christou, T.C. Stamatatos, Inorg.
Chem. 54 (2015) 2137.
[32] T. Sala, M. V. Sargent, J. Chem. Soc. Chem. Commun. (1978) 253.
[33] G.M. Sheldrick, SHELXL-2018/3 (Sheldrick 2014-Progr. Solut. Cryst. Struct.
(2018).
[34] 2008. SADABS, Version 2008/1; Sheldrick, Bruker AXS Inc., (n.d.).
[35] R. Ruiz, C. Sangregorio, A. Caneschi, P. Rossi, A.B. Gaspar, J.A. Real, M.C.
Muñoz, Inorg. Chem. Commun. 3 (2000) 361.
[36] G. Deacon, Coord. Chem. Rev. 33 (1980) 227.
[37] R. Ruiz, A. Caneschi, D. Gatteschi, A.B. Gaspar, J. A. Real, I. Fernandez, M.
Carmen Muñoz, Inorg. Chem. Commun. 2 (1999) 521.
[38] V. Mirkhani, S. Tangestaninejad, M. Moghadam, M. Moghbel, Bioorganic Med.
Chem. 12 (2004) 903.
[39] V. Mirkhani, S. Tangestaninejad, M. Moghadam, Z. Karimian, Bioorganic Med.
Chem. Lett. 13 (2003) 3433.
22

[40] M. Nasr-Esfahani, M. Montazerozohori, P. Akhlaghi, Bull. Korean Chem. Soc. 30
(2009) 1583.
[41] H. Kotani, K. Ohkubo, S. Fukuzumi, J. Am. Chem. Soc. 126 (2004) 15999.
[42] M. Klaper, P. Wessig, T. Linker, Chem. Commun. 52 (2016) 1210.
[43] N.F. Chilton, R.P. Anderson, L.D. Turner, A. Soncini, K.S. Murray, J. Comput.
Chem. 34 (2013) 1164.
[44] M.E. Fisher, Am. J. Phys. 32 (1964) 343.
[45] A. Caneschi, F. Ferraro, D. Gatteschi, M.C. Melandri, P. Rey, R. Sessoli, Angew.
Chemie Int. Ed. English 28 (1989) 1365.
[46] B.-H. Ye, T. Mak, I.D. Williams, X. Li, Chem. Commun. 18 (1997) 1813.
[47] D. Coucouvanis, R.A. Reynolds, W.R. Dunham, J. Am. Chem. Soc. 117 (1995)
7570.
[48] M. Zampakou, N. Rizeq, V. Tangoulis, A.N. Papadopoulos, F. Perdih, I. Turel, G.
Psomas, Inorg. Chem. 53 (2014) 2040.
[49] L. Escriche-Tur, B. Albela, M. Font-Bardia, M. Corbella, New J. Chem. 41 (2017)
2934.
[50] J.B. Vincent, C. Christmas, G. Christou, D.N. Hendrickson, Q. Li, H.R. Chang,
J.C. Huffman, P.D.W. Boyd, J. Am. Chem. Soc. 111 (1989) 2086.
[51] E. Libby, E.A. Schmitt, G. Christou, J.K. McCusker, D.N. Hendrickson, K.
Folting, Inorg. Chem. 30 (1991) 3486.
[52] C. Cañada-Vilalta, J.C. Huffman, G. Christou, Polyhedron 20 (2001) 1785.
[53] G. Aromí, S. Bhaduri, P. Artús, J.C. Huffman, D.N. Hendrickson, G. Christou,
Polyhedron 21 (2002) 1779.
[54] S. Wang, K. Folting, W.E. Streib, E.A. Schmitt, J.K. McCusker, D.N.
Hendrickson, G. Christou, Angew. Chemie Int. Ed. English 30 (1991) 305.
[55] R.F.M. Elshaarawy, Y. Lan, C. Janiak, Inorganica Chim. Acta 401 (2013) 85.
Declaration of Competing Interest
The authors declare that they have no known competing financial interests or personal
relationships that could have appeared to influence the work reported in this paper.
Declaration of interests
☒ The authors declare that they have no known competing financial interests or
personal relationships that could have appeared to influence the work reported in this
paper.
23

☐The authors declare the following financial interests/personal relationships which may
be considered as potential competing interests:
Highlights
–
Role of Mn ions on the decarboxylation and fixation of O2 onto the AnthCO2 anion.
Some drops of water, crucial for obtaining a dinuclear compound instead of 1D system.
EPR shape is very different for Mn(II) dinuclear complexes and 1D systems.
MnIII O2 core shows a characteristic magnetic signature.
4
24

25