Dimerization, redox properties and antioxidant activity of two manganese(III) complexes of difluoro- and dichloro-substituted Schiff-base ligands

Article abstract of DOI:10.1016/j.jinorgbio.2016.11.019

Two mononuclear Mn^III complexes [Mn(3,5-F₂salpn)(H₂O)₂][B(C₆H₅)₄]·2H₂O (1·2H₂O) and [Mn(3,5-Cl₂salpn)(OAc)(H₂O)]·H₂O (2·H₂O), where H₂salpn?=?1,3-bis(salicylidenamino)propane, have been prepared and characterized. The crystal structure of 1·H₂O shows that this complex forms μ-aqua dimers with a short Mn???Mn distance of 4.93??. Under anaerobic conditions, the two complexes are stable in solution and possess trans-diaxial symmetry with the tetradentate Schiff base ligand symmetrically arranged in the equatorial plane. When left in air, these complexes slowly dimerize to yield high-valent [Mn^IV₂(3,5-X₂-salpn)₂(μ-O)₂] in which each X₂-salpn ligand wraps the two Mn ions. This process is favored in basic medium where the deprotonation of the bound water molecule is concomitant with air oxidation. The two complexes catalyze the dismutation of superoxide (superoxide dismutase (SOD) activity) and peroxide (catalase (CAT) activity) in basic medium. The phenyl-ring substituents play an important role on the CAT reaction but have little effect on SOD activity. Kinetics and spectroscopic results indicate that 1 and 2 catalyze H₂O₂ disproportionation through a cycle involving Mn^III₂ and Mn^IV₂ dimers, unlike related complexes with a more rigid and smaller chelate ring, which employ Mn^III/Mn^V[dbnd]O monomers.

Full text of DOI:10.1016/j.jinorgbio.2016.11.019

Dimerization, redox properties and antioxidant activity of two manganese(III)
complexes of difluoro- and dichloro-substituted Schiff-base ligands
Claudia Palopoli, Guillermo G o´ mez, Ana Foi, Fabio Doctorovich, Sonia
Mallet-Ladeira, Christelle Hureau, Sandra Signorella
PII:
DOI:
Reference:
S0162-0134(16)30415-9
doi:10.1016/j.jinorgbio.2016.11.019
JIB 10119
To appear in:
Journal of Inorganic Biochemistry
Received date:
Revised date:
Accepted date:
23 June 2016
8 November 2016
16 November 2016
Please cite this article as: Claudia Palopoli, Guillermo G ´o mez, Ana Foi, Fabio Doc-
torovich, Sonia Mallet-Ladeira, Christelle Hureau, Sandra Signorella, Dimerization, re-
dox properties and antioxidant activity of two manganese(III) complexes of difluoro-
and dichloro-substituted Schiff-base ligands, Journal of Inorganic Biochemistry (2016),
doi:10.1016/j.jinorgbio.2016.11.019
This is a PDF file of an unedited manuscript that has been accepted for publication.
As a service to our customers we are providing this early version of the manuscript.
The manuscript will undergo copyediting, typesetting, and review of the resulting proof
before it is published in its final form. 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.

ACCEPTED MANUSCRIPT
Dimerization, Redox Properties and Antioxidant Activity of Two
Manganese(III) Complexes of Difluoro- and Dichloro-substituted
Schiff-Base Ligands
a
a
b
b
Claudia Palopoli, Guillermo Gómez, Ana Foi, Fabio Doctorovich, Sonia
c
d
a,*
Mallet-Ladeira, Christelle Hureau, Sandra Signorella
a
IQUIR (Instituto de Química Rosario), Consejo Nacional de
Investigaciones Científicas y Tꢀcnicas (CONICET), Facultad de Ciencias
Bioquímicas y Farmacꢀuticas, Universidad Nacional de Rosario, Suipacha
5
31, S2002LRK Rosario, Argentina
b
Departamento de Química Inorgánica, Analítica y Química
Física/INQUIMAE-CONICET, Facultad de Ciencias Exactas y Naturales,
Universidad de Buenos Aires, Ciudad Universitaria, Pabellón 2, Buenos
Aires C1428EHA, Argentina
c
Institut de Chimie de Toulouse, FR 2599, 118 Route de Narbonne, F-
3
1062 Toulouse, France
d
CNRS, LCC (Laboratoire de Chimie de Coordination) and UPS, INPT,
LCC, Universitꢀ de Toulouse, 205 route de Narbonne, F-31077 Toulouse,
France
*Corresponding author: TEL: 54 341 4350214; Fax : 54 341 4370477
E-mail: signorella@iquir-conicet.gov.ar
1

ACCEPTED MANUSCRIPT
Abstract
III
Two mononuclear Mn complexes [Mn(3,5-F salpn)(H O) ][B(C H ) ]·2H O
2
2
2
6
5 4
2
(
1·2H O) and [Mn(3,5-Cl salpn)(OAc)(H O)]·H O (2·H O), where H salpn =
2 2 2 2 2 2
1
,3-bis(salicylidenamino)propane, have been prepared and characterized. The
crystal structure of 1·H O shows that this complex forms -aqua dimers with a
2
short Mn···Mn distance of 4.93 Å. Under anaerobic conditions, the two
complexes are stable in solution and possess trans-diaxial symmetry with the
tetradentate Schiff base ligand symmetrically arranged in the equatorial plane.
When left in air, these complexes slowly dimerize to yield high-valent
IV
2
[
Mn (3,5-X -salpn) (-O) ] in which each X -salpn ligand wraps the two Mn
2
2
2
2
ions. This process is favored in basic medium where the deprotonation of the
bound water molecule is concomitant with air oxidation. The two complexes
catalyze the dismutation of superoxide (superoxide dismutase (SOD) activity)
and peroxide (catalase (CAT) activity) in basic medium. The phenyl-ring
substituents play an important role on the CAT reaction but have little effect on
SOD activity. Kinetics and spectroscopic results indicate that 1 and 2 catalyze
III
2
IV
H O disproportionation through a cycle involving Mn
and Mn
dimers,
2
2
2
unlike related complexes with a more rigid and smaller chelate ring, which
III
V
employ Mn /Mn =O monomers.
Keywords: Mn catalysts; structure; SOD/CAT activity; dimerization; kinetics
2

ACCEPTED MANUSCRIPT
1
. Introduction
·–
Reactive oxygen species (ROS) such as H O and O , cause cell injury
2
2
2
and have been implicated in a number of neurodegenerative diseases [1,2].
Manganese catalases (MnCAT) and superoxide dismutases (MnSOD)
·–
catalyze disproportionation of intracellular H O and O efficiently and
2
2
2
constitute the main line of defense for controlling these toxic oxygen
metabolites in a number of organisms, through a mechanism involving
cyclic oxidation and reduction of the metal cofactor [3-5]. The active site
MnSOD contains one pentacoordinate Mn ion in a N O environment to
3
2
·–
decompose O [5] while MnCAT catalyses the disproportionation of H O
2
2
2
by using a Mn (-O CR)(-O/OH) structural unit [6]. Due to the potential
2
2
use as catalytic reactive ROS scavengers for preventing oxidative stress
injuries, numerous mono- and dinuclear Mn models exhibiting SOD- and
CAT-like activity have been reported [7,8]. Interestingly, a number of
mononuclear Mn complexes possess dual CAT/SOD activity [9-18]. These
catalytic ROS scavengers would be clinically superior to stoichiometric
ones [19] and should have better bioavailability than exogenously
administered antioxidant enzymes [9,11]. In particular, Mn-Schiff-base
complexes have proven to be efficient ROS scavengers and to protect cells
from oxidative damage in several animal models [20,21]. While a number
of studies have been reported for the dual SOD/CAT activity of
mononuclear Mn-Schiff-base complexes, the mechanism of their CAT-like
3

ACCEPTED MANUSCRIPT
reaction remains poorly understood. Complexes of the Mn-salen (salen =
1
,2-bis(salicylidenamino)ethane) family have been proposed to react
III
V
through a mechanism involving mononuclear Mn /Mn =O species [9,21-
V
2
4] and spectroscopic evidence for the formation of Mn =O species upon
O-O bond heterolysis of a peroxo species has been reported for Mn-
salophen (salophen = N,N’-bis(salicyliden)-1,2-phenylendiamine) [25,26]
in accordance with theoretical predictions for the Mn-salen model [27,28].
However, other authors related the catalytic activity of complexes of the
Mn-salen and/or Mn-salpn (salpn = 1,3-bis(salicylidenamino)propane)
families to their ability to form dimers [13,29,30]. This divergence could
imply that, depending on the phenol-ring substituent and geometry, these
Mn-Schiff-base complexes may react via a high-valent Mn-oxo form, or
through a mechanism involving dimeric Mn species as do the enzyme and
2
other mononuclear Mn complexes [31-35]. The kinetic study of the
antioxidant activity of new examples of this type of Mn complexes can
help to unravel some clues of their mode of action. With this aim, we report
here the synthesis, characterization, properties and kinetic studies of
SOD/CAT activity of [Mn(3,5-F salpn)(H O) ][B(C H ) ]·2H O (1·2H O)
2
2
2
6
5 4
2
2
and [Mn(3,5-Cl salpn)(OAc)(H O)]·H O (2·H O) (Scheme 1), and
2
2
2
2
compare their catalytic activity with that of other Mn-Schiff-base catalysts.
(Scheme 1)
4

ACCEPTED MANUSCRIPT
2
2
. Experimental Section
.1. Materials
All reagents or AR chemicals were used as purchased. Solvents were
purified by standard methods. The concentration of H O stock solution
2
2
was determined by iodometric titration.
2
.2. Synthesis of ligands
.2.1. Synthesis of 1,3-bis(3,5-difluorosalicylidenamino)propane (3,5-
2
F2salpn). A mixture of 3,5-difluorosalicylaldehyde (500.3 mg, 3.16 mmol)
and 1,3-diaminopropane (117.5 mg, 1.58 mmol) in ethanol (6.5 mL) was
stirred at reflux for 1 h. The reaction mixture was cooled to room
temperature and 3,5-F salpn precipitated as yellow needles, filtered off and
2
dried under vacuum. Yield: 399.9 mg (1.13 mmol, 72 %). Melting point:
1
1
18 - 119 ºC. H NMR (CCl D) δ: 8.34 (s, 2 H, N=CH-), 6.94 (ddd, 2 H,
3
4
3
4
3
H4, J = 3.1; J = 8.2 and 10.8 Hz), 6.79 (ddd, 2 H, H6, J = 3.1; J
HH
HF
HH
HF
5
=
8.2; J = 1.8 Hz), 3.77 (t, 4 H, N-CH -CH -, J = 6.75 Hz), 2.16 (q, 2 H,
HF
2
2
19
-
CH -CH -CH -, J = 6.75 Hz). F NMR (CCl D) δ (vs Cl CF): -133.24 (d,
2
2
2
3
3
1
9
F3, J = 10.8), -123.36 (t, F5, J = 8.2). F NMR (D -methanol) δ (vs
4
Cl CF): -135.31 (d, F3, J = 10.8), -127.19 (t, F5, J = 8.2). UV-visible (UV-
3
−1
−1
-4
vis) λmax nm (ε M cm ) in DMF: 323 (6660), 419 (1500); in 1.75 x 10
M NaOH (DMF): 327 (5050), 398 (7750). Significant IR bands (KBr, ν
−1
cm ): ν 3.434, ν 3088, 2953, 2877, ν 1643, νAr ring stretch 1604, 1488,
OH
CH
C=N
νAr-OH bend 1331, νAr-C-H out-of-plane 851, 804.
5

ACCEPTED MANUSCRIPT
2
.2.2. Synthesis of 1,3-bis(3,5-dichlorosalicylidenamino)propane (3,5-
Cl salpn). This ligand was prepared through a procedure slightly modified
2
from previously reported [36]. A mixture of 3,5-chlorosalicylaldehyde
(500.4 mg, 2.62 mmol) with 1,3-diaminopropane (97.4 mg, 1.3 mmol) in
ethanol (15 mL) was stirred at reflux for 2 h. The reaction mixture was
cooled to room temperature and 3,5-Cl salpn precipitated as a pure yellow
2
1
solid. Yield: 475 mg (1.13 mmol, 86 %). Melting point: 172 - 173 ºC. H
RMN (CDCl ) δ: 8.31 (s, 2 H, N=CH-), 7.42 (d, 2 H, H4, J = 2.64 Hz),
3
7
.16 (d, 2 H, H6, J = 2.64 Hz), 3.78 (t, 4 H, N-CH -CH , J = 6.75 Hz), 2.15
2
2
−1
(
q, 2 H, -CH -CH -CH -, J = 6.75 Hz). UV-visible (UV-vis) λ nm (ε M
2
2
2
max
−1
cm ) in MeOH: 234 (43150), 280 (17200), 334 (7740), 422 (11300); in
-
4
DMF 333 (6340), 427 (6380); in 5.0 x 10 M Bu NOH (DMF): 327
4
−1
(
4760), 421 (11500). Significant IR bands (KBr, ν cm ):  3443, CH
OH
3
071, 2948, 2859, C=N 1631, Ar ring stretch 1601, 1460, Ar-OH bend 1356, νAr-C-H
8
71, 733.
out-of-plane
2
.3. Synthesis of complexes
.3.1. Synthesis of [Mn(3,5-F salpn)(H O) ][B(C H ) ]·2H O (1·2H O). A
2
2
2
2
6
5 4
2
2
methanol solution (2 mL) of Mn(OAc) ·4H O (34.6 mg, 0.14 mmol) was
2
2
slowly added to a solution of 3,5-F salpn (50.2 mg, 0.14 mmol) in
2
methanol (3.5 mL) and the brownish mixture was left to stir for 24 h. Then,
a methanol solution of NaBPh (48 mg, 0.14 mmol in 0.5 mL) was added
4
6

ACCEPTED MANUSCRIPT
and a brown precipitate formed immediately. The solid was collected by
filtration, washed with hexane and dried under vacuum. Yield: 69 mg (0.09
mmol, 62%). Anal. calcd. for BC F H MnN O : C 61.7; H 5.0; Mn 6.9, N
4
1
4
40
2
6
3
.5%; found: C 61.5; H 4.9; Mn 6.8, N 3.8%. Molar conductivity
-
l
2
-l
+
(
methanol) = 55 Ω cm mol . ESI-MS: m/z = 407.02 [1 - BPh - 2H O] .
4
2
1
H NMR (D -methanol) δ: 18.9 (-CH -, w  1000 Hz), -14.3 (H4, w =
4
2
1/2
1/2
1
9
5
7
2
00 Hz). F NMR (D -methanol) δ (vs Cl CF): -60.22 (w = 737 Hz), -
4
3
1/2
−1
−1
4.76 (w = 311 Hz). UV-visible (UV-vis) λ nm (ε M cm ) in MeOH:
1
/2
max
65 (26240), 382 (8020), 551 (505); in DMF: 375 (9930), 500-600 (265,
-4
broad): in 1.75 x 10 M Bu NOH (DMF): 370 (11596), 500-600 (broad).
4
−1
Significant IR bands (KBr, ν cm ):  3472 (broad), CH 3050, 2999,
OH
2
926, C=N 1612,  1565, 1461, ν
857, 830, 738 (BPh ), 708
Ar
Ar-C-H out-of-plane
4
(
BPh ), 612 (BPh ). Single crystals of 1·H O suitable for X-ray diffraction
4
4
2
were obtained by crystallization from methanol upon standing for several
days.
2
.3.2. Synthesis of [Mn(3,5-Cl salpn)(OAc)(H O)]·H O (2·H O). A mixture
2
2
2
2
of 3,5-Cl salpn (251.2 mg, 0.60 mmol) and Mn(OAc) ·4H O (151.9 mg,
2
2
2
0
.62 mmol) in methanol (9.5 mL) was stirred at room temperature for 48 h
and a solid began to form slowly. After standing for 4 days, the resulting
dark green precipitate was collected by filtration, washed with hexane and
dried under vacuum. Yield: 221.3 mg (0.39 mmol, 65%). Anal. calcd. for
7

ACCEPTED MANUSCRIPT
C Cl H MnN O : C 40.2; H 3.4; Mn 9.7; N 4.9%; found: C 40.0; H 3.4;
1
9
4
19
2
6
-l
2
-l
1
Mn 9.9, N 4.5%. Molar conductivity (methanol) = 47 Ω cm mol . H
NMR (D -methanol) δ: 18.9 (-CH -, w  1000 Hz), -16.2 (H4, w = 370
4
2
1/2
1/2
1
-1
Hz). H NMR (D -DMF) δ: -8.2 (H4, w = 220 Hz). UV-vis λ . (ε M
7
1/2
máx
-
1
cm ) in MeOH: 226 (39160), 284 (19090), 386 (7825), 559 (504); in DMF:
-
4
3
74 (17980), 440 (sh); in 5.0 x 10 M Bu NOH (DMF): 367 (14720), 505
4
−1
(
sh). Significant IR bands (KBr, ν cm ): ν 3430 (broad),  3083, 2929,
OH
CH
2
860, C=N 1615,  1589, 1438, ν
1554/1389, νAr-C-H out-of-plane 870, 770.
Ar
CO2-
2
.4. Physical and analytical measurements
Electronic spectra were recorded on a JASCO V550 spectrophotometer
with thermostated cell compartments. IR spectra were recorded on a
Perkin-Elmer Spectrum One FT-IR spectrophotometer. ESI-mass spectra
were recorded on a Q-TRAP AB SCIEX instrument. The solutions for
electrospray were prepared from solutions of the complexes or reaction
-
5
mixtures and diluted with methanol to a ≈ 10 M concentration at a flow
-
1
rate of 5 mL min . Melting point was taken on a Fisher-Johns (Ionomex)
1
13
apparatus. H and C NMR spectra were recorded on a Bruker AC 300
NMR spectrometer at ambient probe temperature (ca. 26 °C). Chemical
19
1
shifts are referenced to Cl CF ( F NMR) and (CH ) Si ( H NMR) and
3
3 4
downfield shifts are given a positive sign. Paramagnetic NMR spectra were
acquired using super WEFT sequence, with acquisition time of 23 ms. For
8

ACCEPTED MANUSCRIPT
1
9
F NMR measurements, a fluorine frequency of 282.3 MHz was used.
EPR spectra were obtained on a Bruker ESP 300 E spectrometer with a
microwave frequency generated with a Bruker ER 04 (9–10 GHz).
Conductivity measurements were performed using a Horiba F-54 BW
conductivity meter, on 1.0 mM solutions of the complexes in methanol.
The electrochemical experiments were performed with a computer-
controlled Princeton Applied Research potentiostat, model VERSASTAT
II, with model 270/250 Research Electrochemistry Software. Studies were
carried out under Ar, in MeOH, DMF or DMSO solutions using 0.1 M
−3
Bu NPF as a supporting electrolyte and ≈ 10 M of the complex. The
4
6
working electrode was a Pt wire (MeOH, DMF) or a glassy carbon disk
DMSO) and the reference electrode was SCE with Pt as the auxiliary
(
electrode. All potentials are referred to the SCE electrode. Using the
described conditions, the ferrocene/ferrocenium redox couple was observed
at E = 409 mV in MeOH, E = 473 mV in DMF and E = 417 mV in
1/2
1/2
1/2
DMSO.
2
.5. Evaluation of CAT activity
Oxygen evolution studies were carried out polarographically using a Clark-
type oxygen electrode with an YSI oxygen-monitoring system (Model
5
300, Yellow Springs Instruments Co., Inc.). The initial rate method was
used to determine the rate constants (see ref. 37 for further details). Each
9

ACCEPTED MANUSCRIPT
rate constant reported here represents the mean value of multiple
determinations that fall within ±5%. Experiments were carried out at 25 °C.
Alternatively, rates were determined by volumetric measurement of the
evolved O from reaction mixtures. A round-bottom flask with a stopcock
2
equipped gas delivery side tube connected to a gas-measuring burette
(precision of 0.1 mL) was used. A closed vessel containing a solution of
catalyst was stirred at constant temperature on a water bath. Previously
thermostated H O was injected through a silicon stopper, and the evolved
2
2
O was volumetrically measured.
2
2
.6. SOD indirect assay
The SOD activity of the complexes was assayed by measuring the
inhibition of the photoreduction of nitro blue tetrazolium (NBT), by a
method slightly modified from that originally described by Beauchamps
−6
and Fridovich [38]. The solutions containing riboflavin (3.3 × 10 M),
−3
−5
methionine (9.5 × 10 M), NBT (4.2 × 10 M) and the complex of various
concentrations were prepared with phosphate buffer (pH 7.8). The mixtures
were illuminated by a fluorescent lamp with a constant light intensity at 25
°
C. The reduction of NBT was monitored at 560 nm with various
illumination periods (t). Rates in the absence and in the presence of
different concentrations of complex were determined and plotted vs.
complex concentration. Inhibition percentage was calculated according to:
1
0

ACCEPTED MANUSCRIPT
{(ΔAbs / t)without complex − (ΔAbs / t)with complex} × 100/(ΔAbs / t)without complex.
The IC value represents the concentration of the SOD mimic that induces
50
a 50% inhibition of the reduction of NBT. On the basis of competition with
•−
NBT, at 50% inhibition, the rates of the reactions of O with NBT and the
2
mimic are equal, k [complex] = kNBT [NBT], where kNBT (pH = 7.8) = 5.94
4
−1 −1
×
10 M s [39,40]. Control experiments were performed on mixtures of
NBT + complex, riboflavin + complex, and NBT + methionine + complex,
in phosphate buffer, to ensure that the complex does not react
independently with any of the components of the mixture.
2
.7. Crystal data collection and refinement
Single-crystal data were collected on an Oxford Diffraction Gemini E CCD
diffractometer equipped with a sealed tube with Mo K radiation (λ =
0
.71073 Å). Crystal structure data were corrected for absorption with
CrysAlisPro, Agilent Technologies, version 1.171.36.28 (1·H O) and
2
1
.171.34.49 (3·2DMSO), applying an empirical absorption correction using
spherical harmonics, implemented in SCALE3 ABSPACK scaling
algorithm. The structure were solved by direct methods with SIR-92 [41]
and refined by full-matrix least-squares on F² with SHELXL-2014 [42]. All
non-H atoms were refined with anisotropic displacement parameters. The
H atoms on carbon atoms were refined isotropically at calculated positions
using a riding model. For compound 1, hydrogen atoms bonded to O atoms,
1
1

ACCEPTED MANUSCRIPT
were located in a difference Fourier maps and refined using distance
restraints [O—H = 0.85 (1) Å] with U (H) = 1.5U (O). For compound 3,
iso
eq
the solvent molecule was disordered over two positions with the occupation
ratio 0.7:0.3. The atom ellipsoids were restrained to be similar using the
SIMU and DELU commands. Crystal data collection and refinement
parameters for compounds 1·H O and 3·2DMSO are summarized in Table
2
1
.
3
3
. Results and Discussion
.1. Solid state studies
Complexes [Mn(3,5-F salpn)(H O) ][B(C H ) ]·2H O (1·2H O) and
2
2
2
6
5 4
2
2
[
Mn(3,5-Cl salpn)(OAc)(H O)]·H O (2·H O) were prepared from 1:1
2
2
2
2
mixtures of the ligand with Mn(OAc) in methanol. While complex 2
2
precipitates from the reaction mixture, complex 1 required the addition of

the bulky BPh4 to separate from the solution. Color changes were
observed immediately after mixing. When Mn(ClO ) was used instead of
4
2
acetate, no color change was observed until a base was added to the
solution of complex, revealing the role of acetate to deprotonate the phenol
group and thus to induce its coordination to the metal center, an event
II
accompanied by a facilitated aerobic oxidation of the Mn [43-46].
Complexes 1 and 2 share some similar features in the IR spectra, exhibiting
−1
a strong band between 1620 and 1612 cm characteristic of the ν
C=N
1
2

ACCEPTED MANUSCRIPT
−1
stretching mode, which is shifted ≈ 20 cm lower respect to the free
Schiff-base ligand, indicating the coordination to the manganese through
the nitrogen atoms of the imine group. The peak at 1331 (3,5-F salpn)
2
−1
/
1356 (3,5-Cl salpn) cm , assigned to O–H in plane bending vibration of
2
the phenol groups in the free ligands, is absent in the spectra of the
complexes, indicating that deprotonated phenol groups are bound to
manganese. Besides, the two compounds show strong absorptions centered
-
1
at 3472/3430 cm corresponding to the 
stretching of the water
OH
molecules. Complex 1 exhibits out-of-plane C-H bending at 738, 708 and
Ar
-
1
-1
6
12 cm , and strong C-H stretching at 3050 cm , characteristic of the
Ar

non-coordinated BPh4 counter-ion. Powdered samples (1·2H O) and
2
crystals of 1·H O display identical IR spectra. Complex 2 shows two
2
−
1
additional bands at 1554 and 1392 cm assigned to asymmetrical and
−
symmetrical carboxylate CO2 vibrational modes of bound acetate,
respectively.
Crystals of 1·H O suitable for X-ray diffraction studies separated from a
2
methanol solution of 1·2H O. The molecular structure is shown in Fig. 1.
2
The compound crystallizes in the P2 /c space group with the asymmetric
1
III
unit containing a discrete Mn
complex cation, non-coordinated

BPh anion and one solvate water molecule. The fully deprotonated
4
tetradentate Schiff base ligand is tightly bound to the Mn ion via the Nimine
and Ophenolato atoms, with two capping water molecules completing the
1
3

ACCEPTED MANUSCRIPT
coordination sphere of the metal centre. The Mn atom is in an axially
elongated octahedral environment, with the Mn–O(3), Mn–O(5) bond
distances (av. 2.225 Å) distinctly longer than the equatorial Mn–(O1), Mn–
O(4), Mn–N(1) and Mn–N(2) ones (av. 1.968 Å). These values are
4
III
consistent with Jahn–Teller elongation of a d Mn ion in an octahedral
III
ligand field, and are comparable to those reported for other Mn Schiff-
base complexes [30,44,47,48]. The values of the trans-angles in the
coordination environment of the Mn ion are between 173.83 and 175.10°,
indicating a weak distortion from pure octahedral geometry.
(Figure 1)
Complex cations are associated in pairs through H-bonds and aryl
interactions as shown by O–H···O short contacts between capping water
and the two Ophenolato atoms of the neighboring Schiff-base, with distances
O(5)H···O(1) = 2.85 Å and O(5)H···O(4) = 2.92 Å (Figure 2). This results
in a short Mn···Mn distance of 4.93 Å typical of -aqua dimers
[13,43,49,50]. There is an additional intermolecular H-bond between the
other coordinated H O molecule and the solvate water with distance
2
O(3)H···O(0aa) 2.72 Å. H- interaction between the bound water molecule

and the phenyl ring of BPh , with a centroid-to-O(3) distance of 3.40 Å,
4
1
4

ACCEPTED MANUSCRIPT
provides a plausible reason for the short O(3)H···O(0aa) length. Aromatic
rings of the two neighboring complex cations interact through - stacking,
with a centroid-to-centroid distance of 3.70 Å between overlapping rings.
The pairs of complex cations are isolated from each other by the

BPh anions. The crystal packing probably results from Van der Waals
4
interactions.
(Figure 2)
3
.2. Solution studies
The specific conductivity of 1.0 mM solutions of the complexes was
−1
2
−1
measured in methanol. The molar conductivity for 1 is 55  cm mol .
This value is the expected for a complex that contains the low mobility

BPh anion and behaves as 1:1 electrolyte in solution [51]. In line with
4
this, the ESI-mass spectrum of a freshly prepared solution of 1 in DMF

shows only one peak (negative) at m/z = 319 corresponding to BPh anion,
4
and two main peaks (positive) at m/z = 407.4 (100%) and 480.3 (99%)
+
+
belonging to MnL and MnL(DMF) (Figure 3(a)), respectively. No traces
1
of free ligand or dimeric species were found. The diamagnetic H NMR
spectrum of 1 in D -methanol shows peaks belonging to the aromatic
4

protons of BPh without traces of the Schiff-base ligand (Figure S1),
4
1
5

ACCEPTED MANUSCRIPT
indicating the complex is stable towards metal dissociation even in a protic
solvent.
(Figure 3)
For 2, the molar conductivity of the equilibrated methanol solution is 46
−1
2
−1

cm mol , thus revealing partial dissociation of coordinated acetate
with time, in the protic solvent.
1
The paramagnetic H NMR spectra of 1 and 2 in D -methanol (Figure 4)
4
reveal a simple pattern for the two complexes outside the diamagnetic
region (ca. 0–10 ppm). In both cases, one broad resonance is observed at 19
ppm corresponding to the central –CH - of the aliphatic chain. A second
2
resonance appears up-field at -14.3 (1) and -16.2 ppm (2), which can be
assigned to the H4 aromatic ring proton on the basis of comparison with
III
reported spectra for other phenolato-based Mn complexes [52,53].
Because of their closeness to the Mn centre, protons adjacent to the donor
groups of the Schiff base ligand (aromatic H6, N=CH- and -CH -N=C) are
2
III
not observed, and this is consistent with previous finding for related Mn
1
complexes [54]. Paramagnetic H NMR is a useful probe of the ligand
conformation of Mn-Schiff-base complexes in solution. The observation of
1
only one resonance for the aromatic ring protons in the H NMR spectra of
1
–2 indicates that these complexes possess trans-diaxial symmetry with the
tetradentate Schiff base ligand symmetrically arranged in the equatorial
plane [52,55-57]. A more complex pattern is expected when the ligand
1
6

ACCEPTED MANUSCRIPT
adopts a cis- or cis- folded conformation because the different extent of
charge transfer for protons of the phenolato trans to different groups results
in the magnetic nonequivalence of protons of the two phenolato rings
[
52,58]. The chemical shift of H4 correlates with the Hammett parameter of
the aromatic p-substituent ( ):  is larger for the complex with the more
p
H4
electron-withdrawing group, as shown in the inset of Figure 4 where the
complex with the NO derivative is included. This trend in the chemical
2
shifts as a function of the aromatic substituent (NO > Cl > F) indicates that
2
these ligands are predominantly -donor and correlates with the strength of
the phenolato-Mn bond: thus, the stronger the electron-withdrawal from the
Ophenolato, the stronger the Mn-ligand bond and the larger the contact shift on
H4.
(Figure 4)
19
The paramagnetic F NMR spectrum of complex 1 in D -methanol affords
4
two peaks confirming the symmetrical disposition of the ligand around the
Mn ion (Figure S2). The peaks at -127 (F5) and -135 (F3) ppm (vs Cl CF)
3
in the free ligand, were assigned to F5 and F3, respectively, based on the
coupling pattern (Figures S3 and S4). In the complex, these peaks broaden
and shift to -75 (F5) and -60 (F3) ppm.
1
7

ACCEPTED MANUSCRIPT
1
The paramagnetic H NMR spectra of freshly prepared solution of 2 in D -
7
DMF (Figure S5) also exhibit a simple pattern outside the diamagnetic
region with an up-field resonance at -8 ppm corresponding to H4, which is
not modified after addition of base. This indicates that 2 retains the trans
geometry in DMF and basic DMF [53].
The mass spectrum of a freshly prepared DMF solution of 1 shows peaks at
+
+
m/z 407.4 (MnL ) and 480.3 (MnL(DMF) ). With time, new peaks appear
IV
2
at m/z 847.5 and 869.6, corresponding to the Mn
dimeric species
+
+
Mn L O H and Mn L O Na , respectively. After 48 h, the relative
2
2
2
2
2
2
intensities of peaks belonging to monomeric and dimeric species are: 100%
+
+
+
+
MnL , 60% MnL(DMF) , 40% Mn L O H and 64% Mn L O Na (Figure
2
2
2
2
2
2
3
(b)). The same behavior was observed in DMSO. These results reveal this
complex is prone to form the high valence dimer in air, although the rate is
slow. This was confirmed through spectrophotometric monitoring of a
solution of 1 in DMF and basic DMF. In this solvent the complex shows a
band at 375 nm corresponding to a L  M charge transfer transition from
p orbitals of the phenolato oxygen to the partially filled d orbitals of the
III
-1
-1
Mn ion [44,59,60], and a broad band centered at 550 nm ( 265 M cm )
III
that can be assigned to a d–d transition of Mn in a pseudotetragonal
environment [13,44]. Addition of base (up to 10 equiv.) shifts the ligand to
metal charge transfer (LMCT) band to 370 nm. When left in air, the
spectrum does not alter immediately, but after several hours a new broad
1
8

ACCEPTED MANUSCRIPT
band grows up at 460 - 550 nm, which can be attributed to the formation of
the high-valence oxo-bridged diMn dimer [61-63]. In neat DMF, the
intensity of this band raises slower than in basic DMF. The spectrum of a
solution of 1 + 10 equiv. of base registered after 1 day is shown in figure 5.
When 20 or more equiv. of base are added, the band at 460 – 550 nm clears
up and the intensity of the LMCT band at 370 nm increases (Figure 5(b)),

probably because of the formation of the [Mn(F -salpn)(OH) ] complex.
2
2
Deoxygenated DMF solutions of 1, do not evolve with time.
(Figure 5)
IV
The formation of the Mn dimer was also verified for 2 in DMSO. Several
days after preparing the solution and left in air, crystals of [Mn (3,5-
2
Cl salpn) (O) ]·2DMSO (3·2DMSO) separated from the solution, the
2
2
2
structure of which was confirmed by X-ray diffraction. The molecular
structure of 3, shown in Figure 6(a), consists of a binuclear Mn complex
bridged by two oxo groups and two Schiff-base ligands acting as N,O-
donors to each manganese centre, resulting in a Mn···Mn separation of
2
.7239(15) Å. Each Mn atom is in a hexacoordinate environment
[
MnN O ], with two O from the two 3,5-Cl salpn ligands and two
2
4
phenolato
2
bridging oxo ligands occupying the equatorial plane of a distorted
octahedron. The axial sites of the Mn atoms are occupied by two Nimino
,
1
9

ACCEPTED MANUSCRIPT
each one from a different Schiff-base ligand. Bond distances Mn(1)-O(2)
i
1
.809(3) Å, Mn(1)-O(2 ) 1.812(3) Å, Mn-Ophenolato(av) 1.927 Å and Mn-
IV
Nimine(av) 2.017 Å are consistent with a di--oxo-bridged Mn complex
[
36,44,50,64-65] and are similar to the DMF adduct [Mn (Cl -salpn) (-
2
2
2
O) ]·4.5DMF previously characterized by X-ray diffraction [66].
2
(Figure 6)
Solutions of 2 in DMF, MeCN and DMSO were left in air and the IR
spectra of the separated solids compared. It must be noted that the relative
rate of precipitation is well different in the three solvents -MeCN >>
DMSO > DMF-, so in DMF and DMSO solids were collected after several
days. The IR spectra of the three solids –which are essentially
superimposable to the spectrum of a crystalline sample of 3·2DMSO-
-
1
exhibit a sharp absorption band at 647 cm attributable to the highly
symmetrical {Mn O } core with the O occupying the positions trans
2
2
phenolato
to the two Mn-Ooxo bonds [36,67]. Therefore, the same dimer with the
bridging Schiff-base ligand (3) forms in the three solvents (although at
different rates), the IR spectrum of which is well different from that of the
starting complex 2, particularly in the regions of OAc and low frequencies
(Figure 6(b)).
2
0

ACCEPTED MANUSCRIPT
The electronic spectrum of complex 2 in DMF shows an intense phenolato
to Mn charge transfer band at 374 nm, a shoulder at 440 nm and a low
intensity broad absorption centered at 505 nm. The LMCT band decreases
with time while three broad bands at 450, 500 and 680 nm grow up (Figure
7
). A similar behavior is observed when up to 10 equiv. of Bu NOH are
4
added to the DMF solution of 2. Higher excess of base depletes the bands
at longer wavelengths and enhances absorbance at 367 nm, as observed for
1
, a fact also verified in DMSO. The bands at 450, 500 and 680 nm are the
signature of the formation of 3 as shown in the inset of Figure 7, where the
DMF spectrum of a solution of 2 aged 24 h is compared to the spectrum of
3
in MeCN.
(Figure 7)
These results confirm that the formation of 3 in DMF is a slow process that
requires several days for completion.
3
.3. Electrochemical Studies
The electrochemical properties of complexes 1-2 were investigated by
cyclic voltammetry in DMSO, DMF and MeOH (Figure 8, S6-S8). The two
complexes exhibits one quasi-reversible wave with E in the range -27 to
1
/2
1
01 mV vs. SCE (depending on the solvent, Table 2), with ΔE = 120-160
p
2
1

ACCEPTED MANUSCRIPT
III
II
mV and Ia/Ic  0.9, corresponding to the Mn /Mn redox couple of the
complexes. These E values can be correlated with the donor numbers
1/2
(
DN) of the three solvents; DN: 29.8 (DMSO) > 26.6 (DMF) > 19
MeOH). Thus, the higher oxidation state of Mn is more stabilized in the
(
III
II
solvents with higher DN values. As expected, the Mn /Mn redox
potential increases with the inductive effect of the electron-withdrawing
substituent on the phenolato ring; thus, the Cl-substituted complex (2) is
III
II
easier to reduce than the F-substituted one (1). The E for the Mn /Mn
1
/2
couple of 2 occurs at a potential 23 and 63 mV higher than for [Mn(3,5-
Cl -salpn)]ClO and [Mn(3,5-Cl -salen)]ClO , respectively. Salen
2
4
2
4
complexes are reduced at more negative potential than the salpn analogues
which possess a more flexible propylene spacer between the imino groups
III
that destabilizes the Mn state with respect to reduction (Table 2, entries 3,
7
).

In the anodic scan of 1, the BPh4 oxidation appears as an
irreversible peak at E ≈ 1062 in DMF and 1040 mV in DMSO.
p
(Figure 8)
For complex 2 in DMF, an additional nonreversible oxidation peak is
observed at E = 772 mV in the anodic scan (Figure 8(b)), attributable to
p
III
IV
IV
the Mn /Mn couple. This stabilization of the Mn oxidation level is
2
2

ACCEPTED MANUSCRIPT
likely the result of the presence of the basic acetate coordinated to the
metal. The stabilizing effect of the axial anionic ligand had already been
observed for other Mn-salpn(X) and Mn-salen(X) complexes (Table 2,
entries 5, 6, 9; X = axial substituent), which are oxidized bellow 1 V.
Addition of acid to the DMF solution of 2 causes disappearance of the
III
II
wave at 772 mV (Figure S9(b)), shift of the Mn /Mn couple to a slightly
lower potential (E = 71 mV) and emergence of the reduction peak of
1/2
dissociated acetatic acid in the cathodic scan. Subsequent addition of
III
IV
Bu NOH restores the Mn /Mn wave (Figure S9(c)). The oxidation of
4
unbound acetate at 772 mV was discarded based on the lack of such an
oxidation peak in DMF solution of NaOAc and the high oxidation potential
value reported for acetate [71].
Voltammograms registered after addition of substoichiometric
amounts of Bu NOH to a freshly prepared DMF solution of 1, show the
4
III
II
wave corresponding to the Mn /Mn couple at 37 mV and an oxidation
peak at 707 mV (Figure S10), which might result from the conversion of
III
the starting complex into the hydroxo-Mn complex [72]. The hydroxo
group stabilizes the high oxidation state of the metal favoring the oxidation
IV
+
to [Mn (OH)(3,5-F salpn)] at a potential much lower than for the cationic
2
III
II
complex [44] but does not affect the Mn /Mn couple significantly (entries
/6, 8/9, 10/11 in Table 2). Voltammograms of freshly prepared solutions
of complexes 1 and 2 in DMF and basic DMF, exclude the formation of
4
2
3

ACCEPTED MANUSCRIPT
oxo-bridged diMn dimers at short times, the signature of which is
significantly different [44,73-75]
3
.4. SOD activity
The complexes’ stability was checked by UV–vis spectroscopy in the
aqueous phosphate buffer employed for the SOD test. Electronic spectra of
complex 1 registered at different time–lengths (up to 2 h) after preparation
of the solution were identical (Figure S11(a)), revealing the compound is
stable in the buffer. The electronic spectra of 2 in phosphate buffer change
with time, probably as a consequence of acetate/hydroxide substitution in
the aqueous basic medium. However, spectral changes are slow. So, if the
SOD activity is measured with the freshly prepared solution, most acetate
is still bound to Mn as revealed by the minor changes observed in the
electronic spectra registered during the first 45 minutes after preparing the
solution (Figure S11(b)). After 24 h, the spectral pattern of 2 is similar to
that of 1 and can be attributed to [Mn(3,5-Cl salpn)OH] formed through
2
acetate replacement by hydroxide.
The SOD activity of complexes 1 – 2 was determined in aqueous buffer of
pH 7.8 using the NBT assay, which is based on kinetic competition
•−
between NBT and the complex for reacting with O . In this assay SOD
2
activity is inversely related to the amount of formazan, the purple product
•−
formed by the reaction of NBT with O , observed at 560 nm. The two
2
2
4

ACCEPTED MANUSCRIPT
complexes inhibit the reduction of NBT with increasing inhibition as the
concentration of complexes increases, as shown in Figure 9. The IC50
values, graphically evaluated, are 2.72 and 3.22 μM and the rate constants,
k = kNBT [NBT] / IC , are shown in Table 3. These k values are independent
50
of the detector concentration and appropriate for comparison with other
catalysts.
(Figure 9)
Based on k, the activity of 1 – 2 is in the range of other Mn complexes of
tetradentate Schiff-base ligands (Table 3), meaning that ring substitutions
and spacer length modifications have little effect on the SOD activity of
this kind of Mn complexes. SOD activity seems to be more related to the
solubility of the complex. In this way, water soluble Mn complexes of
sulphonato-derived ligands (Table 3) react faster than 1 and 2, even when
the last two complexes possess metal-centered redox couples closest to the
•−
midpoint between the oxidation and reduction of O (120 mV vs. SCE).
2
2
5

ACCEPTED MANUSCRIPT
3
.5. CAT activity
The ability of complexes 1-2 to catalyze H O disproportion was tested in
2
2
basic DMF. In these studies, solutions of catalyst were prepared in
degasified solvent and used immediately to ensure only monomeric species
are present when H O is added. Without addition of base, the reaction
2
2
rates are extremely slow. Addition of H O to a solution of the catalyst +
2
2
Bu NOH in DMF causes evolution of dioxygen coupled to color change
4
from brown to orange. In the presence of 10 equiv. of Bu NOH, complexes
4
1
-2 were able to disproportionate up to 58 and 115 equiv. of H O ,
2
2
respectively. These TON values, although low, are higher than reported for
other Mn-Schiff-base complexes in protic solvent (Table 4).
The initial rate of H O disproportionation (r ) by 1 + Bu NOH in DMF
2
2
i
4
was measured as a function of the complex and substrate concentrations at
5 °C. In this medium, at constant [H O ] , r shows quadratic dependence
2
2
2 0
i
-
1
-1
on [1] (Figure 10), with second-order rate constant k ” = 24.2(4) mM s
0
1
(
[H O ] = 10 mM). At constant [1] , the initial rate of H O dismutation
2
2 0
0
2
2
exhibits saturation kinetics with [H O ] (Fig. 10), and the experimental
2
2 0
data could be modeled by the Michaelis–Menten equation from which the
-
1
maximum rate Vmax = 0.034(2) mM s and the Michaelis constant K =
M
1
.5(1) mM, were determined.
(Figure 10)
2
6

ACCEPTED MANUSCRIPT
Room-temperature UV-vis absorption spectra were taken during the
progress of the reaction of complex 1 with 150 equiv. of H O in basic
2
2
DMF (Figure 11). Immediately after H O addition, the 370 nm band of
2
2
complex 1 decreases and a shoulder appears at 405 nm (brown line in
Figure 11). The intensity of this shoulder rapidly diminishes concurrently
with the apparition of a broad band in the range 450-550 nm (orange line in
Figure 11). As the reaction progresses the solution becomes turbid,
probably resulting from the formation of a poorly soluble inactive or
degraded form of the complex. The absorption in the 450 – 550 nm region
holds during the reaction, and based on the above discussion, might
correspond to a di--oxo-bridged diMn dimer in accordance with the peak
at m/z 847.5 in the ESI-mass spectrum taken on a reaction mixture (Figure
1
1, Inset). Notably, the reaction mixtures are EPR silent. No mononuclear
IV
III
IV
Mn , which would be expected to have a feature near g = 4, nor Mn Mn
mixed-valence species characterized by 16-line spectrum centered at g = 2,
are observed, during or at the end of the reaction.
(Figure 11)
For complex 2, the initial rate of H O disproportionation in basic DMF
2
2
shows second-order dependence on [2] (Figure 12(A)), with k ” = 1.84(5)
0
2
2
7

ACCEPTED MANUSCRIPT
-
1
-1
mM s at saturating substrate concentrations. The electronic spectra
registered on a reaction mixture of 50 equiv. of H O over catalyst (Figure
2
2
1
2(B)) show the decrease of the band at 367 nm, corresponding to the
starting complex, the appearance of a new band at 423 nm and a shoulder at
445 nm. After reacting with 100 equiv. of H O , the band at 367 nm
~
2
2
disappears completely giving rise to the band at 423 nm. The reaction
mixture remains clear until the end of O evolution, without traces of solid.
2
Upon addition of NH OH, the band at 423 nm converts into two bands at
2
3
67 and 320 nm (Inset in Figure 12(B)). Further addition of NH OH causes
2
the vanishing of absorptions at 367 and 423 nm and growth of the band at
20 nm. At the end of the H O disproportionation reaction, the catalyst is
3
2
2
EPR silent, but the species generated by complete reduction with NH OH
2
II
shows the six-line signal at g = 2 typical of non-coupled Mn .
(Figure 12)
Second-order dependence of the initial reaction rate on [catalyst] implies
the involvement of two molecules of the complex at (or before) the slow
step of the catalytic cycle. Based on kinetics and spectroscopic results, a
plausible mechanism for H O disproportionation by complexes 1 and 2 is
2
2
proposed in Scheme 2. The reaction starts upon binding of peroxide to the
III
III
hydroxo (or acetato) Mn complex, LMn X (X = OH or OAc), through
2
8

ACCEPTED MANUSCRIPT
substitution of the hydroxo/acetato group followed by reaction with a
III
second LMn X (X = OH or OAc) molecule to yield the oxidized active
form of the complex that initiates the catalytic cycle: a high-valent
IV
[
LMn (O)] species. This is consistent with the observation of an initial
2
time-lag at the onset of the O evolution. The oxidized catalyst reacts with
2
one H O molecule in the oxidative half-reaction to yield the reduced active
2
2
III
form of catalyst [LMn OH] coupled to O formation, without change in
2
2
III
the total formal charge of the catalyst. The so-formed Mn complex reacts
2
with another H O molecule in the reductive half-reaction to yield water
2
2
and restore the oxidized active form of the catalyst, thus closing the cycle.
III
IV
The catalytic cycle is proposed to run between Mn and Mn oxidation
2
2
states which correspond to Mn in configurations that are EPR-silent in
perpendicular mode and is consistent with the ability of 1 and 2 to access
higher oxidation state in basic medium. The formation of dimeric forms of
the catalyst is in agreement with the second order kinetics, the observation
of a peak corresponding to the diMn species in the mass spectra and the
lack of a peak belonging to the monomer. The low turnover numbers
suggest the conversion of the complex into inactive species through a
mechanism competitive with catalysis. In the case of complex 1, these
species precipitate from the reaction mixture, while for 2 remain in solution
III
2
and could correspond to a Mn inactive form of the complex which can be
II
fully reduced to Mn with NH OH.
2
2
9