Synthesis, characterization and biological evaluation of new manganese metal carbonyl compounds that contain sulfur and selenium ligands as a promising new class of CORMs

Article abstract of DOI:10.1039/c9dt00616h

Three new manganese carbonyl compounds with heavy atom donors were synthesized and their potential use as photoCORMS was evaluated. Interestingly, all compounds had an elusive binding mode, in which the ligands adopted a κ²-X coordination (wher

Full text of DOI:10.1039/c9dt00616h

View Article Online
View Journal
Dalton
Transactions
Accepted Manuscript
This article can be cited before page numbers have been issued, to do this please use: A. L. Amorim, M.
M. Peterle, A. Guerreiro, D. F. Coimbra, R. S. Heying, G. F. Caramori, A. L. Braga, A. J. Bortoluzzi, A. Neves,
G. J. L. Bernardes and R. Peralta, Dalton Trans., 2019, DOI: 10.1039/C9DT00616H.
This is an Accepted Manuscript, which has been through the
Volume 45 Number
1 7 January 2016 Pages 1–398
Royal Society of Chemistry peer review process and has been
accepted for publication.
Dalton
Transactions
Accepted Manuscripts are published online shortly after
An international journal of inorganic chemistry
www.rsc.org/dalton
acceptance, before technical editing, formatting and proof reading.
Using this free service, authors can make their results available
to the community, in citable form, before we publish the edited
article. We will replace this Accepted Manuscript with the edited
and formatted Advance Article as soon as it is available.
You can find more information about Accepted Manuscripts in the
author guidelines.
Please note that technical editing may introduce minor changes
to the text and/or graphics, which may alter content. The journal’s
standard Terms & Conditions and the ethical guidelines, outlined
in our author and reviewer resource centre, still apply. In no
event shall the Royal Society of Chemistry be held responsible
for any errors or omissions in this Accepted Manuscript or any
consequences arising from the use of any information it contains.
ISSN 1477-9226
PAPER
Joseph T. Hupp, Omar K. Farha et al.
Efficient extraction of sulfate from water using
framework
a Zr-metal–organic
rsc.li/dalton

Please do not adjust margins
Page 1 of 10
Dalton Transactions
Dalton Transactions
View Article Online
DOI: 10.1039/C9DT00616H
ARTICLE
Synthesis, Characterization and Biological Evaluation of New
Manganese Metal Carbonyl Compounds That Contain Sulfur and
Selenium Ligands as a Promising New Class of CORMs
André L. Amorim,a,§ Marcos M. Peterle,a,§ Ana Guerreiro,b Daniel F. Coimbra,a Renata S. Heying,a
Giovani F. Caramori,a Antonio L. Braga,*a Adailton J. Bortoluzzi,a Ademir Neves,a Gonçalo J. L.
Bernardesb,c and Rosely A. Peralta*a
Received 00th January 2019,
Accepted 00th January 20xx
DOI: 10.1039/x0xx00000x
www.rsc.org/
Three new manganese carbonyl compounds with heavy atom donors were synthesized and their potential use as
photoCORMS was evaluated. Interestingly, all compounds had an elusive binding mode, in which the ligands adopted a 2-
X coordination (where X = S or Se), confirmed both by X-ray crystallography and IR spectroscopy. The stability of the title
compounds in the dark was determined by monitoring the changes in the UV spectra of the compounds in both
dichloromethane and acetonitrile. These studies show that in coordinating solvents there is an exchange of the bromide
bonded to the metal centre by a solvent molecule, which is evidenced by the changes in the UV and IR spectra and by DFT
analysis. EDA and natural bond order analyses were conducted to evaluate the influence of the heavy atom donors in the
first coordination sphere of the compounds. Photoexcitation at 380 nm demonstrated that all compounds showed release
of all three COs, as seen in the photoproducts detected by IR spectroscopy. Furthermore, CO release was observed when
the photoCORMs were incubated with living cells, and we observed a CO-dependent inhibition of cell viability.
pyTAm (2-(pyridyl)imino-triazaadamantane)25 and the dinuclear
rhenium-manganese carbonyl.26 These photoCORMs have
different applications concerning their biological stability,
Introduction
Since the discovery of the signalling properties of carbon
monoxide1,2 and its potential use as a therapeutic agent,3-5
several research groups have begun to study ways to deliver CO
in a safe manner and limit the hazards.6 One way to limit its
cytotoxicity is to use carbon monoxide-releasing molecules
(CORMs)7,8 for which the release of CO is prompted by the use
of an external trigger,9 such as photoexcitation. This is one of
the most commonly employed procedures that gave rise to so-
called photoCORMs.10 Even though some organic CO carriers
have been developed,11-18 metal carbonyl compounds (MCCs)8
are still common because of the diverse array of properties
associated with the metal centre,9 which lead to the
development of ruthenium, iron, rhenium, molybdenum and
manganese MCCs,19-23 the latter being of the most common.
Currently, a wide list of examples of manganese MCCs with
either different ligands or dinuclear species, which can help in
the design of new photoCORMs, can be found in the literature,
such as L1-k2N1,N2 (2,6-bis(benzimidazole-2′-yl)pyridine),24
photoswitch probes upon CO release and visible light activation.
In general, these compounds have MnI centres with either
bidentate or facial tridentate ligands. However, it has been
shown that the use of ligands with nonbonding groups, such as
tris(2-pyridylmethyl)amine (tpa), which adopts a 3 binding
mode, can facilitate the formation of inactivated CORMS
(iCORMs) after the CO-release procedure.27 After the
dissociation, either the released CO molecules are substituted
by a species present in the medium, typically a solvent
molecule, or the ligand suffers dissociation of the metal
framework, and the use of switch on probes, such as imdansyl
can help elucidate this mechanism and provide an insight into
the localization of the CO prodrug.28 Despite these recent
advances, most manganese photoCORMs have CO-release
activity only near the UV region, which limits their applicability.
In this context, the use of ligands that can shift the excitation
bands towards lower energy, such as bis(4-chloro-
phenylimino)acenaphthene and azopyridines,29,30 are highly
desirable.
a. Departamento de Química, Universidade Federal de Santa Catarina, Florianópolis,
Santa Catarina 88040-900, Brazil.
Although various examples of new and interesting ligands can
be found in the literature, the use of heavy atoms in the ligand
framework has not been reported. In this regard, the increase
in the spin-orbit coupling (SOC) with the use of heavier atoms,
such as sulfur and selenium, can modify the CO dissociation and
increase the intersystem crossing (ISC) between singlet and
triplet excited states.31
b. Instituto de Medicina Molecular, Faculdade de Medicina, Universidade de Lisboa,
Avenida Professor Egas Moniz, 1649-028, Lisboa (Portugal).
c. Department of Chemistry, University of Cambridge, Lensfield Road, Cambridge
CB2 1EW (UK)
*E-mail: rosely.peralta@ufsc.br; Tel: +55(48)3721 3627
Email: Braga.antonio@ufsc.br
†Electronic Supplementary Information (ESI) available free of charge: See
DOI: 10.1039/x0xx00000x.
Herein, we focus on the design of new ligands that contain
heavy atoms in the ligand framework of manganese
photoCORMs that could potentially modify the photoexcitation
‡Crystallographic information available free of charge at the Cambridge
Crystallographic Data Centre as supplementary publication numbers CCDC 1894674,
1894675 and 1894677.
§Both authors contributed equally to this work.
This journal is © The Royal Society of Chemistry 2019
DaltonTrans., 2019, 00, 1-10 | 1
Please do not adjust margins

Please do not adjust margins
Dalton Transactions
Page 2 of 10
ARTICLE
Dalton Transactions
behaviour of these compounds. For this purpose, three starting material, and one of the sulfur or selenium moieties
View Article Online
DOI: 10.1039/C9DT00616H
chalcogen-containing ligands and their respective MCCs were remained uncoordinated.
synthesized (Figure 1), phS (1’), phSe (2’), bzlSe (3’), The crystallographic parameters are summarized in Table S1–S6
[MnBr(CO)3(phS-2 S)] (1), [MnBr(CO)3(phSe-2 Se)] (2) and and the relevant bond lengths and angles are displayed in Table
[MnBr(CO)3(bzlSe-2 Se)] (3) and their properties were 1, and show that the trans influence takes place, which
evaluated by common characterization techniques and influences the Mn–C bond lengths. The use of softer ligands
theoretical approaches.
shortens the Mn–C bond trans to the soft moiety, with the Mn–
C bonds trans to the bromide being slightly more compressed.
As a direct consequence, all three Mn–C bonds differ in some
way, deviating from the C3v symmetry, which is consistent with
the non-degenerate character of the CO antisymmetric
stretching seen in the IR spectra. As expected, the softer binding
moieties, sulfur and selenium, have a much higher M–X (where
X = S or Se) bond length relative to typical nitrogen ligands, due
to their higher covalent radii. The remainder of the bond lengths
and angles are within typical ranges considering previously
published manganese tricarbonyl compounds.34,35 The
molecular structures for 1–3 are represented in Figure 2.
Results and discussion
Ligands 1’–3’ were synthesized by the reaction of bis(2-
chloroehtyl)amine hydrochloride with the appropriate
selenolate or thiolate, which were generated in situ by the
cleavage of the selected dichalcogenide with sodium
borohydride. The products were purified by column
chromatography and obtained in good yields. The ligands were
characterized by IR, 1H, 13C and 77Se NMR spectroscopy (Figures
S1–S11).32
The manganese metal carbonyl compounds were synthesized
by heating at reflux a 4:1 diethyl ether/acetonitrile mixture of
ligands and Mn(CO)5Br precursor in dim light conditions
overnight, similar to already published procedures.33 Slow
diffusion of hexane into the solutions led to the formation of
yellow to orange crystals, which were collected and kept under
an argon atmosphere at –20C and used for characterization.
Solid-state characterization
The attenuated total reflectance (ATR) IR spectra for
compounds 1–3 showed three intense bands at around 2050
and 1850 cm–1, attributed to the CO stretching frequencies. This
is consistent with a facial orientation of the carbonyl groups and
a distorted C3v symmetry (Figures S12–S14). The higher energy
frequency, centred at 2020, 2015 and 2015 cm–1 for 1, 2 and 3,
respectively, is attributed to symmetric CO stretching whereas
the lower frequency energies centred at 1933/1912, 1919/1896
and 1927/1908 cm–1 are originated from a split in the
antisymmetric CO stretching band. In general, manganese
tricarbonyl compounds with tridentate symmetric ligands form
compounds with facial bonded ligands and a much less
degenerate unsymmetric stretching band. This splitting
indicates that the ligands of the compounds synthesized are
bonded in a bidentate fashion with no displacement of the
bromide from the first coordination sphere. The splitting of the
degenerate band, a direct comparison of the bands for
compounds 1 and 2, which are structurally similar, also reveals
that the use of heavier and softer ligands shifts the CO
stretching frequencies to lower energy. This indicates an
increase in the metal to ligand charge transfer (MLCT) that
increases the C–O bond length with heavier atoms. Analysis
with other manganese MCCs, such as [Mn(CO)3(tpa-k3 N)]Br27
(with CO: 2032, 1961 and 1905 cm–1) and CORM-ONN133 (with
CO: 2040, 1941, 1920 cm–1), reinforces this finding.
Figure 1 Structural representation of the ligands phS (1’), phSe (2’), bzlSe (3’) and their
respective manganese carbonyl compounds used in this study, from left to right:
[MnBr(CO)3(phS-2 S)] (1), [MnBr(CO)3(phSe-2 Se)] (2) and [MnBr(CO)3(bzlSe-2 Se)] (3).
Table 1 Selected bond lengths (Å) and angles (°) for 1–3.
1
2
3
Bonds
Mn1-C1
C1-O1
Mn1-C2
C2-O2
Mn1-C3
C3-O3
Mn1-N1
Mn1-S2
Mn1-Se1
Mn1-Se2
Mn1-Br1
1.810(3)
1.142(3)
1.804(3)
1.142(3)
1.795(3)
1.143(3)
1.803(4)
1.145(5)
1.801(4)
1.149(5)
1.790(4)
1.148(5)
2.131(3)
-
1.817(5)
1.140(6)
1.806(5)
1.141(6)
1.788(5)
1.138(6)
2.129(4)
-
2.1255(19)
2.3748(6)
-
-
-
2.4625(8)
-
2.5344(8)
2.4758(7)
2.5280(6)
2.5288(4)
Angles
C1-Mn1-C2
C1-Mn1-C3
C2-Mn1-C3
N1-Mn1-S2
N1-Mn1-Se1
N1-Mn1-Se2
S2-Mn1-Br1
Se1-Mn1-Br1
Se2-Mn1-Br1
88.43(12)
91.21(12)
90.67(11)
85.05(5)
88.39(19)
90.73(18)
90.70(18)
88.0(2)
92.2(2)
92.1(2)
-
-
-
-
-
85.11(11)
The above statement is corroborated by the crystal data
obtained for compounds 1–3. In all three cases, the ligand binds
in a bidentate mode to form a five-membered ring. The use of
mild synthetic conditions with diethyl ether as the solvent did
not favour the displacement of the bromide present in the
85.96(8)
-
-
83.918(18)
-
-
-
-
82.47(3)
-
82.49(2)
2 | DaltonTrans., 2019, 00, 1-10
This journal is © The Royal Society of Chemistry 2019
Please do not adjust margins

Page 3 of 10
Dalton Transactions
Please do not adjust margins
Dalton Transactions
ARTICLE
Table 2 Energy Decomposition Analysis (GKS-EDA) of M–CO bonds (kcal.mol–1) at
View Article Online
DOI: 10.1039/C9DT00616H
BP86-D3/def2-SVP level of theory.
Interaction ΔEint
ΔEelstat
ΔEexch
ΔErep
ΔEpol ΔEdisp ΔEcorr
1
Mn1-C1O1 -46.31 -146.59 -225.04 465.74 -94.69 -2.54 -43.19
Mn1-C2O2 -41.97 -148.37 -229.74 477.28 -97.59 -4.38 -39.15
Mn1-C3O3 -54.06 -148.92 -228.06 471.95 -98.41 -4.68 -45.93
2
Mn1-C1O1 -47.27 -146.80 -225.38 466.28 -94.63 -2.40 -44.34
Mn1-C2O2 -42.61 -151.31 -233.43 486.30 -101.16 -4.51 -38.50
Mn1-C3O3 -54.98 -148.96 -228.95 473.05 -98.27 -4.82 -47.04
3
Mn1-C1O1 -46.71 -143.01 -218.82 452.53 -90.58 -3.04 -43.79
Mn1-C2O2 -42.43 -152.79 -236.94 492.90 -103.84 -3.99 -37.77
Mn1-C3O3 -58.05 -149.55 -232.21 478.38 -101.21 -5.28 -48.18
The nature of the chalcogen employed has a remarkable effect
on the Mn1-C2O2 and Mn1-C3O3 bonding situations. For
instance, on going from 1 to 2, it can be noticed that the
electrostatic, ΔEelstat, exchange, ΔEexch, and polarization, ΔEpol,
contributions for Mn1-C2O2 become more stabilizing,
suggesting an increase of the electronic charge reorganization,
including charge transfer and polarization. A similar situation,
but with slight magnitude is also observed for Mn1-C3O3 (Table
2). By changing the ligand (phSe) to (bzlSe), that means going
from 2 to 3, it is noticed that Mn1-C3O3 becomes still stronger,
whereas the two other Mn–CO bonds are slightly less strong. In
fact, the GKS-EDA results reveal that both the nature of the
chalcogen as well as the ligand employed have non-negligible
and distinguishable electronic effects on the Mn–CO bonding
situations. The Wiberg bond indexes and natural population
analysis (NPA) obtained from natural bond order method, a
localized form of the molecular wavefunction also confirms the
differences on Mn-CO bond lengths as well as the electronic
charge reorganization (Table S10).
Figure 2 Molecular structures of compounds 1–3, ellipsoids with 40% probability. a)
[MnBr(CO)3(phS-2 S)]; b) [MnBr(CO)3(phSe-2 Se)], and c) [MnBr(CO)3(bzlSe-2 Se)].
Identification of the active species in solution
Bonding Analysis
Because the isolated compounds show the ligands binding in a
bidentate mode with a bromide in the first coordination sphere,
it is reasonable to expect that the absorption bands on the UV
spectra for these compounds will show solvent-dependent
behaviour, with the exchange of the bromide present in the first
coordination sphere for a coordinating solvent. Thus, the UV
spectra for 1–3 were analysed under two conditions: a
nonbonding and a bonding solvent, namely dichloromethane
and acetonitrile, respectively.
In dichloromethane, compounds 1, 2 and 3 present broad
absorption bands with max at 385 nm (2597 mol L–1 mol–1), 385
nm (2235 mol L–1 mol–1) and 388 nm (2014 mol L–1 mol–1),
respectively. Because these bands did not show any sign of
modification over a 24h incubation period in the dark (Figure 3
for 3 and S15–S16), the structures must retain the bidentate
binding mode [MnBr(CO)3(L-2 X)] (in which L = phS, phSe or
bzlSe and X = S or Se).
The Mn–CO bonding situations were analysed in the light of
both Kohn-Sham energy decomposition analysis (GKS-EDA) and
natural bond orbital (NBO) analyses. The GKS-EDA quantifies
the bonding strength and decomposes it on several meaningful
terms to reveal that in compounds 1–3 the carbonyl group
coordinated trans to bromide ion, C3O3, interacts more
strongly with the Mn1 centre than the two other carbonyl
groups, C1O1 and C2O2, which are coordinated trans to
nitrogen and chalcogen atoms (S or Se), respectively (Table 2).
The total binding energy values for Mn1-C3O3 vary from –54.06
to –58.05 kcal mol–1, whereas for Mn1-C1O1 and Mn1-C2O2
they vary from –41.97 to –47.27 kcal mol–1. In fact, the total
interaction energy values (ΔEint) shows that, in all cases, the
C2O2 group is the most labile one (–41.97– –42.61 kcal mol–1).
Such findings are totally in line with the solid-state
characterization, which reveals the three Mn-CO bonds are
different as a consequence of the trans effect. Such effect is also
confirmed by means of the GKS-EDA terms.
This journal is © The Royal Society of Chemistry 2019
DaltonTrans., 2019, 00, 1-10 | 3
Please do not adjust margins

Please do not adjust margins
Dalton Transactions
Page 4 of 10
ARTICLE
Dalton Transactions
Table 3 Experimental and calculated carbonyl stretching frequencies for the
View Article Online
manganese carbonyl compounds used in this study in both solvents (in which X = S
DOI: 10.1039/C9DT00616H
or Se) at BP86-D3/def2-SVP level of theory.
Ligand (L)
phS
phSe
[MnBr(CO)3L-2 X]
bzlSe
exp
calc
2014
1932
1904
exp
calc
2010
1928
1899
exp
Calc
2012
1929
1883
symmetric
2027
1938
1917
2022
1934
1906
2024
1934
1916
antisymmetric
[Mn(CH3CN)(CO)3L-2 X]+
exp
exp
Exp
symmetric
2023
1924
1900
2024
1932
1898
2023
1932
1911
antisymmetric
The changes in the CO stretching bands are shown in Figure 3
Figure 3 UV (top) and IR (bottom) spectral changes of 2 over 24 h period in the dark in for 2 (S17–S18 for 1 and 3) and the results displayed in Table 3.
dichloromethane (top) and acetonitrile (bottom).
Unfortunately, the isolation of the species [Mn(CH3CN)(CO)3(L-
2 X)]+ or [Mn(CO)3(L-3 X)]+ (in which L = phS, phSe or bzlSe and
It was also possible to analyse the stability of 1–3 in
X = S or Se), which could facilitate the identification of these
dichloromethane over the course of the incubation period by IR
species, could not be performed.
spectroscopy, which showed that the three CO stretching bands
remain unchanged, both in terms of their energy and their
intensity during analysis (Figure 3 for 2 and S17–S18).
Interestingly, a direct comparison between the solid ATR and
the diluted solution in KBr is not possible because the stretching
frequencies for the sample in KBr pellets are shifted towards
higher energy, 10 and 20 cm–1 for the symmetric and
antisymmetric stretching modes, respectively. The frequencies
for the L-2 binding modes are summarized in Table 3. This
difference observed for the solid ATR and the diluted solution
in KBr is due to the presence of intermolecular forces between
the sample and some of the water present in the KBr pellets.
With addition of acetonitrile the spectral profile slowly changes,
with the absorption bands shifting to higher energy (Figure 3 for
2 and S15–S16). Over the same 24h period in a 1:1
dichloromethane/acetonitrile the bands are hypsochromically
shifted by 20 nm, with max at 365, 365 and 368 nm,
respectively. These shifts could indicate that there is a
displacement of the bonded bromide by acetonitrile, which
would result in the formation of [Mn(CH3CN)(CO)3(L-2 X)]+
species (in which L = phS, phSe or bzlSe and X = S or Se). Further
displacement of the solvent molecule by the pendant moiety to
form Mn(CO)3(L-3)]+ was not detected over the course of the
experiment, since no red shift of the initial absorption bands
due to the coordination of the pendant moiety are detected.
To elucidate if there is a ligand exchange between the bromide
and a solvent molecule, the IR spectra were also obtained for
the species formed during the incubation periods. In this media
the three stretching bands shows minor shifts to lower energies,
and the nondegenerate character of the antisymmetric
stretching band is maintained for all three compounds. This
behaviour indicates that the distorted C3v character is
maintained, without exchange of the bromide by a solvent
molecule, which would shift the CO stretching to higher
frequencies, nor the coordination of the pendant moiety to the
metal centre, , which would increase the degeneracy of the
bands, similar behaviour is already reported in the literature.36
DFT and TD-DFT analysis of the active species
The [MnBr(CO)3(L-2 X)] and [Mn(CH3CN)(CO)3(L-2 X)]+ (in
which L = phS, phSe or bzlSe and X = S or Se) binding modes were
optimized to gain insight into the geometrical and electronic
structures of the species by using density functional theory
(DFT) in the gas phase. The absence of imaginary frequencies
corroborated the understanding that the optimized structures
are truly energy-minimum. These structures were once again
optimized to take into account the polarizability of the medium,
using the SMD solvation model, with the appropriate solvents.
The calculated frequencies at the BP86-D3/def2-SVP for the
[MnBr(CO)3(L-2 X)] were compared to the experimental IR
frequencies and are in good agreement with the experimental
values (Table 3) with deviations in the order of 10 cm–1. The
results are consistent with the proposed coordination.
Time-dependent DFT (TD-DFT) calculations were also made. The
first forty singlet excited states were calculated for both cases.
For [MnBr(CO)3(L-2 X)] )] (in which L = phS, phSe or bzlSe and X
= S or Se) the most prominent band is attributed either to a
HOMO-1 → LUMO+1 for phS and phSe ligands and a HOMO-2
→ LUMO+1 for the bzlSe ligand, in which for both cases the
orbitals comprise mostly manganese d orbitals, with some
contribution of the bromide p orbitals. The first unoccupied
orbitals are * orbitals delocalized on the chalcogen and
aromatic rings.37,38 Moreover, the increase in the polarization of
the donor did not further displace the MLCT bands to lower
energy, even though the LUMO+1 is 0.04 eV lower when going
from 1 to 2, probably due to the higher electrostatic character
of the M-X bond when going form S to Se, which is consistent
with the EDA results (Figure 4 and Table 4).
As expected, exchanging the solvent from dichloromethane to
acetonitrile and exchanging the bromide for acetonitrile,
increases the energy of the singlet excitation. Interestingly, for
[Mn(CH3CN)(CO)3(L-2 X)]+ compounds the excitations comprise
4 | DaltonTrans., 2019, 00, 1-10
This journal is © The Royal Society of Chemistry 2019
Please do not adjust margins

Page 5 of 10
Dalton Transactions
Please do not adjust margins
Dalton Transactions
ARTICLE
of inner orbitals that arise from the manganese d orbitals to the dichloromethane and acetonitrile solutions, by using an
View Article Online
* orbitals of the aromatic rings.
excitation wavelength of 380 ± 10 nm (3D80O),I:a10n.d10t3h9e/Cc9hDaTn0g0e6s16iHn
the UV, IR and NMR spectra were monitored. Lower energy
wavelengths were also employed however, as expected, show
CO release properties
Because the synthesized compounds have a medium- much slower CO release rates and for this reason not shown.
dependent structure, the CO release activity was tested in both In dichloromethane, in which the compounds retain the
coordinated bromide, all compounds showed a steady decrease
in the MLCT absorption bands at around 390 nm and the
appearance of a new band at around 480 nm, with the
formation of two isosbestic points during the processes, as seen
in Figure 5 for 2 and Figures S19 and S20 for 1 and 3. The decay
in the MLCT band is due to the CO release process that lowers
the concentration of the tricarbonyl species, forming
photoproducts with a lower CO content, which shifts the MLCT
bands to higher energies, as seen in the increase in absorbance
at around 300 nm for compound 2 (Figure 5 for 2 and S19 and
S20 for 1 and 3).
Interestingly, a second lower energy band is formed at around
480 nm during the photoexcitation procedure, which at the
time seemed unusual. However, mechanics studies by Kurz
group39,40 demonstrated that during the CO release of
manganese carbonyl compounds a bis-carbonyl species with a
solvent molecule that occupies the released CO is formed. In
addition, because [Mn(CO)3(tpa-3 N)]+ compound27 binds the
uncoordinated pyridine moiety after photoexcitation, it is
feasible to postulate that this second band is associated with a
[MnBr(CO)2(L-3 X)] species (in which L = phS, phSe or bzlSe and
X = S or Se) for which the uncoordinated chalcogen moiety binds
to the biscarbonyl species that is formed during the
photoexcitation procedure. This photoproduct should have
lower energy absorption values due to the presence of the
second bonded chalcogen.
To verify that the biscarbonyl intermediate is formed in
Figure 4. Calculated molecular orbitals for the first major transition in dichloromethane
dichloromethane, the changes in the IR spectra of the products
were also monitored. The analysis revealed a steady decrease
in the three major CO stretching bands during the
decomposition process. Concomitantly, formation of a new and
lower frequency band at around 1800 cm–1 was observed. This
CO stretching band is associated with antisymmetric stretching
of the bis-carbonyl intermediate that is continuously formed
during the decomposition process, as shown by Kurz’s
group.39,40 Calculating the area under the curve after 840 s of
exposure, 2.0, 2.2 and 2.2 equivalents of CO were released for
1-3 respectively, as seen in Figure 5 for 2 and S21 and S22 for 1
and 3.
Unfortunately, the attribution of the intermediates by NMR
spectroscopy was not possible due to rapid oxidation of the
metal centre during the CO release procedure, which resulted
in MnII species rapidly turning the solution paramagnetic, which
gave broader peaks and hindered the collection of reliable
results (Figure S23–S25).
(top) and acetonitrile (bottom) at B97X-D3/def2 level of theory.
Table 4 Energies, oscillator strengths (fosc) with main orbital contribution and type of
orbitals involved in the first excitations at B97X-D3/def2 level of theory.
State
 (nm)
fosc
Type of transition
[MnBr(CO)3(phS-2 S)]
1
2
422.3
403.5
0.003610293
0.013416896
 (Mn-CO); p (Br) →  (phenyl)
[MnBr(CO)3(phSe-2 Se)]
1
2
418.2
400.6
0.001774086
0.011913981
 (Mn-CO); p (Br) →  (phenyl)
[MnBr(CO)3(bzlSe-2 Se)]
1
2
419.4
403.0
0.001882577
0.009767997
 (Mn-CO); p (Br) →  (phenyl)
[Mn(CH3CN)(CO)3(phS-2 S)]+
1
2
421.7
401.6
0.000968382
0.015594213
 (Mn-CO);  (phenyl) → 
(phenyl)
[Mn(CH3CN)(CO)3(phSe-2 Se)]+
By applying a pseud-first-order kinetics model to the decay in
the UV spectra, it is possible to obtain the apparent CO release
1
2
413.6
394.8
0.000718936
0.017254963
 (Mn-CO);  (phenyl) → 
(phenyl)
[Mn(CH3CN)(CO)3(bzl-2 Se)] +
constants
for
1–3,
which
are,
respectively,
(1.82 ± 0.03) 10–3 s–1, (7.76 ± 0.15) 10–3 s–1 and (7.87 ± 0.15) 10–3
s–1.
1
2
400.7
383.9
0.000860679
0.007014534
 (Mn-CO);  (phenyl) → 
(phenyl)
This journal is © The Royal Society of Chemistry 2019
DaltonTrans., 2019, 00, 1-10 | 5
Please do not adjust margins

Please do not adjust margins
Dalton Transactions
Page 6 of 10
ARTICLE
Dalton Transactions
View Article Online
DOI: 10.1039/C9DT00616H
Figure 5 UV (top) and IR (bottom) spectral changes of 2 during excitation with a 380 ± 10
nm incident light in a dichloromethane solution.
Photoexcitation in acetonitrile with 380 resulted in the
formation of a brown to beige precipitate during the Figure 6. CO release in aqueous solution and cells. (a) Histogram representing the
emission intensity (RFUs - relative fluorescent units) measured at 60min after incubation
of 1 μM COP-1 (ex=475 nm) with 150μM of compounds 1–3 in PBS pH 7.4 at 37 °C. The
results correspond to Mean+ SEM of three independent experiments; (b) HeLa and
HepG2 cell viability determined after 24h incubation with 150μM of 1–3. Viability is
dissociation of CO by changing the colour of the solution from
colourless to light yellow to deep yellow. This solid is related to
the formation of MnII dimers after the dissociation of the first
carbonyl group.39,40 The precipitate formed during the
represented as percentage of control (Mean + SEM of three independent experiments).
irradiation inhibited the determination of the photoproducts in (c) Representative confocal microscopy images of CO release in live HeLa cells untreated
(control) and treated with 150 µM of 1–3 after 30min of incubation with COP-1. Left
panel- nucleic acid staining with Syto61 (red), middle panel- COP-1 turn-on response to
CO (green), right panel- merged channels.
the required concentrations. This is due to the high degree of
scattering of the irradiated light, which hindered the pseudo-
first-order kinetic constants and made the results for the
intermediates detected in KBr IR unreliable.
Importantly, we could observe that the CO release is cytosolic
and seems to be localized in the perinuclear region, which is in
CO release in aqueous conditions and living cells
accordance to what was previously described for other CO-
Having proven the complete dissociation of carbon monoxide
releasing molecules.42 By considering that the aim of the use of
from the compounds in the organic solvents, it was important
CORMs is to selectively deliver CO to cells, as a therapeutic
to evaluate the ability of the compounds to release CO in
agent, we also addressed the cell viability of two different
physiological conditions. We could observe that the three
compounds were able to release CO in a buffered aqueous
solution (PBS, pH7.4), in response to COP-1, a fluorescent turn-
on probe that selectively reacts with CO through a
palladium-mediated carbonylation reaction.41 In this assay,
compound 2 was the most effective, because it was able to
release 2x more CO than compounds 1 and 3 (Figure 6(a) and
cancer cell lines, HeLa (cervical cancer) and HepG2 (liver cancer)
after treatment with compounds 1–3. Only 24h after treatment
with 150μM of 1–3 less than 30% of cells are viable (Figure 6b).
Although compound 3 is not the one that releases the most CO
it seems to be the most potent against both cell lines because it
is the one with lower IC50s (Figure S28). This effect may be due
to the ability of compound 3 to release CO faster than the other
S26). Moreover, we could also detect increasing fluorescence
compounds because 15min after incubation it shows a stronger
levels, in a time dependent manner in HeLa cells (derived from
intracellular fluorescence intensity relative to 1 and 2, a
cervical cancer) when incubated with the compounds for 15 min
tendency that is also visible after 30min of incubation (Figure 6b
(2 and 3) and for 30min (1–3) in the presence of COP-1 [Figure
and S27). In addition, at 45min after incubation we can already
6 (c) and S27] relative to the control, which indicates increasing
see that the cells look altered, with abnormal nuclei and
levels of CO release by compounds 1–3.
6 | DaltonTrans., 2019, 00, 1-10
This journal is © The Royal Society of Chemistry 2019
Please do not adjust margins

Page 7 of 10
Dalton Transactions
Please do not adjust margins
Dalton Transactions
ARTICLE
morphology, which is indicative of cell death (Figure S28). Thus, recorded with a Bruker micrOTOF-Q II APPI mass spectrometer
View Article Online
we believe that the effect of the compounds on cell death may equipped with an automatic syringe pumDpOfIo: r10s.a10m39p/lCe9iDnTje0c0t6i1o6nH.
occur within minutes after incubation, which is in accordance Infrared spectra were recorded with a Bruker Optics Alpha
with the IR results that indicate a complete CO dissociation after benchtop FT-IR spectrometer. The melting points were
15min exposure.
determined in a Microquimica MQRPF-301 digital model
equipment with heating plate. Column chromatography was
performed with silica gel (230-400 mesh). Thin layer
chromatography was performed with Merck Silica Gel GF254,
0.25 mm thickness. Samples were visualized either under
ultraviolet light or stained with iodine vapor and acidic vanillin.
Reactions under inert atmosphere were conducted in flame-
dried or oven-dried glassware equipped with tightly fitted
rubber septa and under a positive atmosphere of dry argon.
Reagents and solvents were handled by using standard syringe
techniques. Temperatures were maintained by use of a mineral
oil bath with a heating and stirring plate.
Conclusions
In this study, three new manganese metal-carbonyl compounds
were synthesized that contain heavy atoms in the first
coordination sphere, namely sulfur and selenium. Even though
the tridentate ligands where used, the MCCs synthesized had a
2-L X ( X = S or Se) binding mode, detected by both the crystal
structures and the asymmetry of the C-O stretching bands. The
bonding analysis showed that the sulfur and selenium heavy-
atom donors have unneglectable interactions with the
carbonyls groups that make the CO trans to these moieties
more stable in the ground state. Because the isolated
compounds had an elusive binding mode, studies were
conducted to determine the active species in solution, and the
results show that in acetonitrile solution only the substitution
of the bromide by an acetonitrile molecule occurs, without
formation of 3-L. Moreover, the photoactivity of compounds
1–3, showed that the CO release activity was solvent
dependent. In dichloromethane it was found that the pseudo-
first-order constants follow the trend k3 ≈ k2 > k1, which clearly
show that the use of heavier atoms in the first coordination
sphere can considerably increase the CO release rate, even
without the expected bathochromic shift of the MLCT bands,
which could prove useful in reducing the exposure time of the
patients to UV-A light. This increase in the CO release rate is
directly related to the dissociative populated excited states. In
this regard more measures are required to fully understand the
nature of these states, and how the increase in the SOC of the
ligand increases the CO release rate, contrary to what is known
of Mn and Re compounds, which in this case decreases the CO
release.
1
The IR, H, 13C and 77Se NMR spectra can be found in the
Supplementary Information.
Synthesis of ligands containing sulfur and selenium (1’–3’)
Ethanol (5.0 ml) and the appropriate diorganoyl dichalcogenide (3.0
mmol) were placed in a 25 mL round-bottomed flask under an argon
atmosphere. To the stirred reaction mixture, NaBH4 (6.0 mmol) was
added in small portions. After the complete evolution of gas, bis(2-
chloroethyl)amine hydrochloride (2.0 mmol) dissolved in ethanol
was added dropwise. The reaction mixture was then heated at 70 °C
for 18 h then allowed to cool to room temperature. The solvent was
removed by evaporation under reduced pressured. The crude
mixture was dissolved in ethyl acetate (30 mL) and extracted with
water (3 x 10 mL). The combined organic phase was washed with
brine, dried with anhydrous MgSO4, and concentrated under
reduced pressure. The crude material was purified by column
chromatography (hexane/ethyl acetate 7:3).
Bis(2-(phenylthio)ethyl)amine (1’). Yellow oil. Yield 288 mg;
289.46 g mol–1; 50%.1H NMR (400 MHz, CDCl3):  = 7.34 (d, J =
4.0 Hz, 3H), 7.25 (t, J = 8.0 Hz, 4H), 7.19–7.15 (m, 2H), 3.03 (t, J
= 8.0 Hz, 4H), 2.80 (t, J = 8.0 Hz, 4H), 1.88 (s, 1H) ppm. 13C NMR
(100 MHz, CDCl3):  = 135.7 (C), 129.8 (CH), 128.9 (CH), 126.3
(CH), 47.8 (CH2), 34.2 (CH2) ppm. Selected IR frequencies (KBr):
3300.3, 3057.6, 2920.9, 2829.1, 1945.9, 1640.0, 1582.8, 1478.8,
1438.0, 1121.9, 1024.0, 738.4, 691.5, 475.3 cm–1.32
Experimental
General
1H NMR spectra were obtained at 400 MHz with a Varian AS-
400 NMR spectrometer. Spectra were recorded in CDCl3
solutions. Chemical shifts are referenced to the solvent peak of
CDCl3 or tetramethylsilane as the external reference. Data are
reported as follows: chemical shift (), multiplicity, coupling
constant (J) and integration intensity. 13C NMR spectra were
obtained at 100 MHz with a Varian AS-400 NMR spectrometer.
Spectra were recorded in CDCl3 solutions. Chemical shifts are
referenced to the solvent peak of CDCl3. 77Se NMR spectra were
obtained at 38.14 MHz with a Bruker AC-200 NMR
spectrometer. Spectra were recorded in CDCl3 solutions.
Chemical shifts are referenced to diphenyl diselenide as the
external reference (463.15 ppm). Abbreviations to denote the
multiplicity of a particular signal are: s (singlet), d (doublet), t
(triplet) and m (multiplet). High-resolution mass spectra were
Bis(2-(phenylselanyl)ethyl)amine (2’). Off-white solid. Yield: 514
mg; 383.28 g mol–1; 71%. M.p. 110-112 °C (lit. 106 °C). 1H NMR
(400 MHz, CDCl3):  = 7.51–7.48 (m, 3H), 7.27–7.22 (m, 5H), 3.00
(t, J = 6.6 Hz, 4H), 2.85 (t, J = 6.6 Hz, 4H), 1.77 (s, 1H) ppm. 13C
NMR (100 MHz, CDCl3):  = 133.1 (CH), 129.7 (C), 129.2 (CH),
127.1 (CH), 48.6 (CH2), 28.6 (CH2) ppm. 77Se NMR (38.14 MHz,
CDCl3):  = 266.21 ppm. Selected IR frequencies (KBr): 3430.7,
3051.3, 2812.7, 1947.9, 1576.6, 1476.7, 1433.88, 1144.2,
1019.8, 732.2, 693.5 cm–1.32
Bis(2-(benzylselanyl)ethyl)amine (3’). Yellow oil. Yield: 658 mg;
1
411.33 g mol–1; 80%. H NMR (400 MHz, CDCl3):  = 7.31–7.18
(m, 9H), 3.77 (s, 4H), 2.73 (t, J = 6.6 Hz, 4H), 2.60 (t, J = 6.6 Hz,
4H), 1.66 (br. s, 1H) ppm. 13C NMR (100 MHz, CDCl3):  = 139.4
This journal is © The Royal Society of Chemistry 2019
DaltonTrans., 2019, 00, 1-10 | 7
Please do not adjust margins

Please do not adjust margins
Dalton Transactions
Page 8 of 10
ARTICLE
Dalton Transactions
(C), 128.9 (CH), 128.6 (CH), 126.8 (CH), 48.7 (CH2), 27.1 (CH2), were found from the Fourier difference map and treated as free
View Article Online
44
24.6 (CH2) ppm. 77Se NMR (38.14 MHz, CDCl3):  = 255.79 ppm. atoms. ORTEP plots were drawn with the PLATON program.
DOI: 10.1039/C9DT00616H
HRMS- (APPI): m/z calcd. for [C18H23NSe2]+ [M+nH]+ 414.0237; Selected crystallographic information are summarized in the
found 414.0236.
Supplementary Information file. Full crystallographic tables
(including structure factors) have been deposited with the
Cambridge Crystallographic Data Centre as supplementary
Synthesis of Mn compounds that contain sulfur and selenium 1–3
For the preparation of the carbonyl compounds, a solution of publication numbers CCDC 1894674, 1894675 and 1894677.
the appropriate organochalcogen ligand (0.25 mmol) was These data can be obtained free of charge from the Cambridge
solubilized in a mixture of dichloromethane (1 mL) and diethyl Crystallographic Data Centre at www.ccdc.cam.ac.uk.
ether (4 mL). The resulting solution was degassed by applying
Computational Methods
two freeze-pump-thaw cycles. The manganese precursor
[MnBr(CO)5] (0.25 mmol) was added to a frozen solution of the Geometries for all compounds were obtained by using DFT45,46
ligands, thaw degassed three times and then kept under with the BP8647 exchange correlation functional and Grimme’s
agitation until the mixture reached room temperature. The D3BJ dispersion correction48 along with the def2-TZVP basis
mixture was then heated to reflux under an inert atmosphere set49 as implemented in the Orca package, version 3.0.2.50
and dim light conditions overnight to give brown solutions. To Structures were confirmed as minima by the absence of
the cooled mixture, hexane (10 mL) was added and kept at 25°C imaginary eigenvalues in the generalized Hessian matrix.
for two days to give dark yellow crystals suitable for X-ray GKS-EDA51 as implemented in GAMESS-US 2012,52 was
analysis.
performed with the BP86 functional47 by using Grimme’s D3
correction48 and the def2-SVP basis set.53 The Boys and Bernardi
[MnBr(CO)3(phS-2 S)] (1). Dark yellow solid. Yield: 742 mg; approach54 was used to correct for basis set superposition
511.35 g mol–1; 72%. Selected IR frequencies (ATR): 2020 (s, errors. In all cases a CO ligand was considered as a neutral
CO), 1933 (s, CO), 1912 (s, CO), 1610-1433 (w, ar), 757 (m, ar), fragment whereas the remainder of the molecule was the
633 (m, ar) cm–1. Selected electronic absorption wavelengths second fragment without residual charges. The GKS-EDA
(CH2Cl2, max in nm) and their respective  values (L mol-1 cm-1): scheme decomposes the instantaneous interaction energy
385 (2757). Found and calculated elemental analysis (ΔEint) into the sum of six physical meaningful terms:
(C19H19NO3S2BrMn): C, 44.81 (44.89); H, 3.75 (3.77); N, 2.87 electrostatic (ΔEelstat), exchange (ΔEexch), repulsion (ΔErep),
(2.76).
polarization (ΔEpol), dispersion (ΔEdisp) and correlation (ΔEcorr).
[MnBr(CO)3(phSe-2 Se)] (2). Dark yellow solid. Yield: 742 mg; Natural bond orbital analysis55 was also performed in
605.14 g mol–1; 76%. Selected IR frequencies (ATR): 2015 (s, conjunction with GKS-EDA to support the bonding analysis of
CO), 1919 (s, CO), 1896 (s, CO), 1605-1435 (w, ar), 760 (m, ar), CO groups.
630 (m, ar) cm–1. Selected electronic absorption wavelengths For the frequencies and singlet excitations, geometries for all
(CH2Cl2, max in nm) and their respective  values (L mol–1 cm–1): compounds were obtained by using DFT45,46 with the B97X-
385 (2235). Found and calculated elemental analysis D356 exchange correlation functional along with the def2-SVP
(C19H19NO3Se2BrMn): C, 39.42 (39,37); H, 4.83 (4.72); N, 1.77 basis53 for carbon and hydrogen atoms, the def2-TZVP(-f) basis
(2.81).
set for carbon and nitrogen in the first coordination sphere, and
[MnBr(CO)3(bzlSe-2 Se)] (3). Dark yellow solid. Yield: 742 mg; oxygen atoms, and the def2-TZVP basis for the heavy atoms S,
633.20 g mol–1; 68%. Selected IR frequencies (ATR): 2015 (s, Se, Br and Mn,53 as implemented in the Orca package, version
CO), 1927 (s, CO), 1908 (s, CO), 1602-1430 (w, ar), 755 (m, ar), 4.0.1.57 To speed up the calculations, the RIJCOSX
627 (m, ar) cm–1. Selected electronic absorption wavelengths approximation was used.58 Structures were confirmed as
(CH2Cl2, max in nm) and their respective  values (L mol-1 cm-1): minima by the absence of imaginary eigenvalues in the Hessian
388 (2014). Found and calculated elemental analysis matrix. The resulting structures were then optimized by taking
(C21H23NO3Se2BrMn): C, 40.25 (40.02); H, 3.56 (3.68); N, 2.11 into account solvation effects by using the SMD solvation
(2.22).
model,59 as implemented in the Orca package, by using the
appropriate solvent. The excited state wavefunctions and the
transition oscillator strengths within TDA approximation were
Single-Crystal X-ray Structure Determinations
Crystallographic analyses were carried out with a Bruker APEX II calculated using the TD-DFT calculations,60 in which the first
DUO diffractometer with graphite-monochromated Mo-Kα forty roots were considered on the same basis as before.
radiation (l = 0.71069 Å), at 200(2) K. Images were recorded by
Investigation in living cells
using the φ and ω scan method. Semi-empirical absorption
correction43 was also applied to all measured intensities with Cell culture. HepG2 (a human hepatoma cell line) and HeLa cells
maximum and minimum transmission factors shown in Table (derived from cervical cancer cells) were maintained in a
S1. The structures were solved by direct methods and refined humidified incubator at 37 °C under 5% CO2 and grown byh
by the full-matrix least-squares on F2.44 H atoms attached to C using 1x D-MEM ((Invitrogen, Life Technologies) supplemented
atoms were placed at their idealized positions, with C–H with 10% heat-inactivated fetal bovine serum (FBS) (Gibco, Life
distances and Ueq values taken from the default settings of the Technologies), 1x MEM NEAA (Gibco, Life Technologies), 1x
refinement program. Hydrogen atoms of the amine groups GlutaMAX (Gibco, Life Technologies), 200 units mL-1 penicillin
8 | DaltonTrans., 2019, 00, 1-10
This journal is © The Royal Society of Chemistry 2019
Please do not adjust margins

Page 9 of 10
Dalton Transactions
Please do not adjust margins
Dalton Transactions
ARTICLE
and 200 µg mL-1 streptomycin (Gibco, Life Technologies) and 10 the editing of this manuscript. The authors also acknowledge
View Article Online
DOI: 10.1039/C9DT00616H
mM HEPES (Gibco, Life Technologies).
CEBIME for the HRMS analysis.
Cell viability assay. To determine cell viability a CellTiter Blue
assay was performed. For this purpose 10.000 cells per well
were seeded in 96 well-plates and 24h later were treated with
1, 5, 10, 25, 50, 100 and 150μM of either phS, phSe and bzlSe.
After 24h or 48h after treatment, the cells incubated with
CellTiter blue for 1h 30min at 37°C after which the viability was
determined by measuring the Emission Intensity in RFUs with
an Infinite M200 plate reader.
CO release in aqueous solution. The quantification of the CO
release promoted by the compounds in aqueous solution was
determined by COP-1 fluorescence on different time points (0,
10, 20, 30, 40, 50, 60, 70, 80 and 90min) and different wave
lengths (490 to 610nm). Hence, 150 μM of the compounds in
PBS, were mixed with a 1 μM solution of COP-1 (synthesized
according to the literature)31 and COP-1 fluorescence was
measured using an Infinite M200 plate reader. A solution of
COP-1 with PBS was used as a negative control
Notes and references
1
H. P. Kim, S. W. Ryter and A. M. Choi, Annu. Rev. Pharmacol.
Toxicol., 2006, 46, 411-449.
J. M. Fukuto, S. J. Carrington, D. J. Tantillo, J. G. Harrison, L. J.
2
Ignarro, B. A. Freeman, A. Chen, and D. A. Wink, Chem. Res.
Toxicol., 2012, 25, 769-793.
3
R. Foresti, M. G. Bani-Hani, and R. Motterlini, Intensive Care
Med., 2008, 34, 649-658.
S. García-Gallego and G. J. L. Bernardes, Angew. Chem. Int.
Ed., 2014, 53, 9712-9721.
R. Motterlini, B. E. Mann and R. Foresti, Expert Opin. Invest.
Drugs, 2005, 14, 1305-1318.
R. Motterlini and L. E. Otterbein, Nat. Rev. Drug Discovery,
4
5
6
2010, 9, 728-743.
7
8
B. E. Mann, Organometallics, 2012, 31, 5728-5735.
C. C. Romão, W. A. Blättler, J. D. Seixas and G. J. L. Bernardes,
Chem. Soc. Rev., 2012, 41, 3571-3583.
9
U. Schatzschneider, Br. J. Pharmacol. 2015, 172, 1638-1650.
CO release in cells. A total of 3x104 HeLa cells per well were
seeded in a 8-chambered Ibidi cell plate. Thirty minutes before
measuring CO release, the cells were incubated with a 1:2000
dilution on Syto61 (Lyfe Technologies) for nuclear staining. After
the 30min incubation time, the cells were washed and
incubated with 1 μM of COP-1 and 150 μM of the compounds,
in medium with reduced phenol red, and were immediately
observed by using a Zeiss LSM 710 confocal Laser Point-
Scanning Microscope with a 40X oil objective lens and a
numerical aperture of 1.3. Throughout the experiment the cells
were maintained at 37 °C. COP-1 was excited by using an Argon
Laser with wavelength of 488 nm, whereas Syto61 was excited
by using a DPSS 561-10 laser (561nm). COP-1 was read at green
("em 500-550 nm) and Syto61 in red ("em 570-640 nm). Mean
of total fluorescence intensity of treated and untreated cells
were determined with the help of ImageJ software. Statistically
significant differences were tested with a Mann Whitney test.
Data are presented in the graphs as mean of total fluorescent
intensity ± SEM.
10 R. D. Rimmer, A. E. Pierri and P. C. Ford, Coord. Chem. Rev.,
2012, 256, 1509–1519.
11 X. Ji, C. Zhou, K. Ji, R. E. Aghoghovbia, V. Chittavong, B. Ke, B.
Wang, Angew. Chem. Int. Ed. Engl. 2016, 55, 15846-15851.
12 L. A. P. Antony, T. Slanina, P. Sebej, T. Solomek and P. Klán,
Org. Lett. 2013, 15, 4552-4555.
13 M. Popova, T. Soboleva, A. M. Arif, L. M. Berreau, RSC Adv.
2017, 7, 21997–22007.
14 S. N. Anderson, J. M. Richards, H. J. Esquer, A. D. Benninghoff,
A. M. Arif, L. M. Berreau, ChemistryOpen, 2015, 4, 590-594.
15 E. Palao, T. Slanina, L. Muchova, T. Solomek, L. Vitek, P. Klan,
J. Am. Chem. Soc., 2016 ,138, 126–133.
16 T. I. Ayudhya, P. J. Pellechia and N. N. Dingra, Dalton Trans.,
2018, 47, 538-543.
17 D. Wang, E. Viennois, K. Ji, K. Damera, A. Draganov, Y. Zheng,
C. Dai, D. Merlin and B. Wang, Chem. Commun., 2014, 50, 15890–
15893.
18 Y. Zheng, X. Ji, B. Yu, K. Ji, D. Gallo, E. Csizmadia, M. Zhu, M. R.
Choudhury, L. K. C. De La Cruz, V. Chittavong, Z. Pan, Z. Yuan, L. E.
Otterbein, B. Wang, Nat. Chem., 2018, 10, 787–794.
19 T. R. Johnson, B. E. Mann, I. P. Teasdale, H. Adams, R. Foresti,
C. J. Green and R. Motterlini, Dalton Trans, 2007, 15, 1500–1508.
20 R. Kretschemer, G. Guessner, H. Görls, S. H. Heinemann and
M. J. Westerhausen, Inorg. Biochem., 2011, 105, 6-9.
21 F. Zobi, O. Blacque, R. A. Jacobs, M. C. Schaub and A. Y.
Bogdanova, Dalton Trans., 2012, 41, 370-378.
Conflicts of interest
22 A. R. Marques, L. Kromer, D. J. Gallo, N. Penacho, S. S.
Rodrigues, J. D. Seixas, G. J. L. Bernardes, P. M. Reis, S. L.
Otterbein, R. A. Ruggieri, A. S. G. Gonçalves, A. M. L. Gonçalves,
M. N. D. Matos, I. Bento, L. E. Otterbein, W. A. Blättler and C. C.
Romão, Organometallics, 2012, 31, 5810-5822.
23 U. Schatzschneider, Inorg. Chim. Acta, 2011, 374, 19-23.
24 A. M. Mansour, A. Friedrich, Inorg. Chem. Front., 2017, 4,
1517-1524.
25 J. Jimenez, I. Chakraborty, S. J. Carrington, P. K. Mascharak,
Dalton Trans., 2016, 45, 13204-13213.
26 Z. Li, A. E. Pierri, P. Huang, G. Wu, A. V. Iretskii, P. C. Ford,
Inorg. Chem., 2017, 56, 6094-6104.
27 C. Nagel, S. McLean, R. K. Poole, H. Braunschweig, T. Kramer
and U. Schatzschneider, Dalton Trans., 2014, 43, 9986-9997.
28 J. Jimenez, I. Chakraborty, A. Dominguez, J. Martinez-
Gonzalez, W. M. C. Sameera and P. K. Mascharak, Inorg. Chem.,
2018, 57, 1766-1773.
29 S. J. Carrington, I. Chakraborty and P. K. Mascharak, Dalton
Trans., 2015, 44, 13828-13834.
There are no conflicts to declare.
Acknowledgements
Funded under the Brazilian governmental agencies CNPq (PVE -
400210/2014-2), CNPq, and CERSusChem GSK/FAPESP (grant
2014/50249-8) grants for A. Neves, A. L. Braga, A. L. Amorim and
R. A. Peralta, CAPES (Finance Code 001). We also thank the
FAPESC, EU Horizon 2020 programme, Marie Skłodowska-Curie
ITN GA no. 675007 and TWINN-2017 ACORN, GA no. 807281,
the Royal Society (UF110046 and URF\R\180019 to G.J.L.B.),
FCT Portugal (iFCT IF/00624/2015 to G.J.L.B. and PhD
studentship SFRH/BD/115932/2016 to A.G.), and an ERC StG
(GA no. 676832) for funding. The authors thank Prof. Chris
Chang for providing COP-1, and Dr Vikki Cantrill for her help with
This journal is © The Royal Society of Chemistry 2019
DaltonTrans., 2019, 00, 1-10 | 9
Please do not adjust margins

Please do not adjust margins
Dalton Transactions
Page 10 of 10
ARTICLE
Dalton Transactions
30 E. Kottelat, A. Ruggi, F. Zobi, Dalton Trans., 2016, 45, 6920– 45 P. Hohenberg and W. Kohn, Phys. Rev., 1964, 136, B864-B871.
View Article Online
6927.
31 H. C. Braga, C. A. M. Salla, I. H. Bechtold, A. J. Bortoluzzi, B. de 47 J. P. Perdew, Phys. Rev. B, 1986, 33, 8822-8824.
Souza and H. Gallardo, Dyes Pigments, 2017, 117, 149-156. 48 S. Grimme, J. Antony, S. Ehrlich and H. Krieg, J. Chem. Phys.,
46 W. Kohn and L. J. Sham, Phys. Rev. 1965, 140, A1133-A1138.
DOI: 10.1039/C9DT00616H
32 S. Kumar, G. K. Rao, A. Kumar, M. P. Singh, A. K. Singh, Dalton 2010, 132, 154104.
Trans., 2013, 42, 16939–16948.
49 K. Eichkorn, F. Weigend, O. Treutler and R. Ahlrichs, Theor.
33 R. Mede, J. Traber, M. Klein, H. Görls, G. Gessner, P. Chem. Acc., 1997, 97, 119-124.
Hoffmann, M. Schmitt, J. Popp, S. H. Heinemann, U. Neugebauer 50 F. Neese, WIREs Comput. Mol. Sci. 2012, 2, 73-78.
and M. Westerhausen, Dalton Trans., 2017, 46, 1684-1693.
51 P. Su, Z. Jiang, Z. Chen and W. Wu, J. Phys. Chem. A, 2014, 118,
34 M. A. Gonzalez, S. J. Carrington, N. L. Fry, J. L. Martinez and P. 2531-2542.
K. Mascharak, Inorg. Chem., 2012, 51, 11930-11940.
52 M. W. Schmidt, K. K. Baldridge, J. A. Boatz, S. T. Elbert, M. S.
35 F. Mohr, J. Niesel, U. Schatzschneider and C. W. Z. Lehmann, Gordon, J. H. Jensen, S. Koseki, N. Matsunaga, K. A. Nguyen, S. Su,
Anorg. Allg. Chem., 2011, 638, 543-546.
T. L. Windus, M. Dupuis and J. A. Montgomery, J. Comput. Chem.,
36 M. A. Gonzalez, S. J. Carrington, N. L. Fry, J. L. Martinez and P. 1993, 14, 1347-1363.
K. Mascharak, Inorg. Chem., 2012, 51, 11930-11940.
53 K. Eichkorn, O. Treutler, H. Öhm, M. Häser and R. Ahlrichs,
37 I. Chakraborty, S. J. Carrington and P. K. Mascharak, Acc. Chem. Phys. Lett., 1995, 240, 283-290.
Chem. Res., 2014, 47, 2603-2611.
54 S. F. Boys and F. Bernardi, Mol. Phys. 2002, 100, 65-73.
38 R. N. Pickens, B. J. Neyhouse, D. T. Reed, S. T. Ashton and J. K. 55 E. D. Glendening, C. R. Landis and F. Weinhold, WIREs Comput.
White, Inorg. Chem. 2018, 57, 11616-11625.
Mol. Sci., 2012, 2, 1-42.
39 H-M. Berends and P. Kurz, Inorg. Chim. Acta., 2011, 380, 141- 56 J. -D. Chai, M. Head-Gordon, J. Chem. Phys., 2008, 128,
147. 084106.
40 U. Sachs, G. Schaper, D. Winkler, D. Kratzert and P. Kurz, 57 F. Neese, WIREs Comput. Mol. Sci. 2018, 8, e1327.
Dalton Trans., 2016, 45, 17464-17473. 58 T. Petrenko, S. Kossmann and F. Neese, J. Chem. Phys., 2011,
41 B. W. Michel, A. R. Lippert and C. J. Chang, J. Am. Chem. Soc., 134, 054116.
2012, 134, 15668-15671.
59 A. V. Manerich, C. J. Cramer and D. G. Truhlar, J. Phys. Chem.
42 M. Chaves-Ferreira, I. S. Albuquerque, D. Matak-Vinkovic, A. B, 2009, 113, 6378.
C. Coelho, S. M. Carvalho, L. M. Saraiva, C. C. Romão and G. J. L. 60 F. Neese and G. Olbrich, Chem. Phys. Lett., 2002, 362, 170-
Bernardes, Angew. Chem. Int. Ed., 2015, 54, 1172 -1175.
43 APEX2, SAINT and SADABS. version 2011.8-0; Bruker AXS Inc.,
Madison, Wisconsin, USA.
178.
44 G. M. Sheldrick, Acta Cryst., 2015, C71, 3-8.
10 | DaltonTrans., 2019, 00, 1-10
This journal is © The Royal Society of Chemistry 2019
Please do not adjust margins