New cobalt(ii) coordination designs and the influence of varying chelate characters, ligand charges and incorporated group I metal ions on enzyme-like oxidative coupling activity

Article abstract of DOI:10.1039/d0nj02347g

In transition-metal-mediated catalysis, designing new, well defined coordination architectures and subjecting them to catalysis testing under the same reaction conditions is a necessary tool for gaining an improved understanding of desirable active site geometries and characteristics. In this work, three ligands varying in chelation and charge characters (1= tridentate, dianionic N^N^N;2= bidentate, dianionic N^N;3= bidentate, neutral N^O) were synthesized and introduced as organic backbones around cobalt(ii) centres. Four multinuclear coordination assemblies incorporating various group I metal ions (i.e., (Na-Co1-i)₂, (Na-Co1-e)₂, (K-Co1-e)₂and (Cs-Co1-e)₂) and two mononuclear complexes (i.e.,Co2andCo3) were isolated, structurally characterized and studied as phenoxazinone synthase models. X-ray data and Continuous Shape Measure (CShM) calculations described unusual vacant trigonal bipyramidal geometries for (Na-Co1-i)₂, (Na-Co1-e)₂andCo3, trigonal bipyramids for (K-Co1-e)₂and (Cs-Co1-e)₂, and a tetrahedron forCo2. According to mass spectrometry data, the four multinuclear (M^I-Co1-solvate)₂complexes of ligand1easily dissociate in solution to yield the corresponding ions including a common mononuclearCo1unit. Testing as models in the oxidative coupling ofo-aminophenol revealed that an increasing chelate character of the organic ligand backbone produces an undesirable catalyst activity reduction. Furthermore, the similarity in coupling efficiencies observed among complexes (Na-Co1-i)₂, (Na-Co1-e)₂, (K-Co1-e)₂and (Cs-Co1-e)₂proves that the presence of varying group I metal ions contributed no catalytic effects to the coupling efficiencies and that theCo1species, which is generated by dissociation in solution, is the only active species responsible for the observed biomimetic behaviours. Therefore, the combination of anionic chelate ligands with easily detachable, neutral co-ligands (e.g., water or pyridine) around a cobalt(ii) centre could be recommended for the generation of a successful catalyst material, rather than the combination of a neutral chelate ligand with counter-anionic monodentate co-ligands like halides.

Full text of DOI:10.1039/d0nj02347g

NJC
View Article Online
View Journal
PAPER
New cobalt(II) coordination designs and the
influence of varying chelate characters, ligand
charges and incorporated group I metal ions on
enzyme-like oxidative coupling activity†
abc
Cite this:DOI: 10.1039/d0nj02347g
Hammed Olawale Oloyede,
Helmar Gorls,b Winfried Plass *b and Abiodun Omokehinde Eseola
Joseph Anthony Orighomisan Woods,a
bd
*
¨
In transition-metal-mediated catalysis, designing new, well defined coordination architectures and subjecting
them to catalysis testing under the same reaction conditions is a necessary tool for gaining an improved
understanding of desirable active site geometries and characteristics. In this work, three ligands varying in
chelation and charge characters (1 = tridentate, dianionic N^N^N; 2 = bidentate, dianionic N^N; 3 =
bidentate, neutral N^O) were synthesized and introduced as organic backbones around cobalt(II) centres.
Four multinuclear coordination assemblies incorporating various group I metal ions (i.e., (Na-Co1-i)2, (Na-
Co1-e)2, (K-Co1-e)2 and (Cs-Co1-e)2) and two mononuclear complexes (i.e., Co2 and Co3) were isolated,
structurally characterized and studied as phenoxazinone synthase models. X-ray data and Continuous Shape
Measure (CShM) calculations described unusual vacant trigonal bipyramidal geometries for (Na-Co1-i)2, (Na-
Co1-e)2 and Co3, trigonal bipyramids for (K-Co1-e)2 and (Cs-Co1-e)2, and a tetrahedron for Co2. According
to mass spectrometry data, the four multinuclear (MI-Co1-solvate)2 complexes of ligand 1 easily dissociate in
solution to yield the corresponding ions including a common mononuclear Co1 unit. Testing as models in
the oxidative coupling of o-aminophenol revealed that an increasing chelate character of the organic ligand
backbone produces an undesirable catalyst activity reduction. Furthermore, the similarity in coupling
efficiencies observed among complexes (Na-Co1-i)2, (Na-Co1-e)2, (K-Co1-e)2 and (Cs-Co1-e)2 proves that
the presence of varying group I metal ions contributed no catalytic effects to the coupling efficiencies and
that the Co1 species, which is generated by dissociation in solution, is the only active species responsible for
the observed biomimetic behaviours. Therefore, the combination of anionic chelate ligands with easily
detachable, neutral co-ligands (e.g., water or pyridine) around a cobalt(II) centre could be recommended for
the generation of a successful catalyst material, rather than the combination of a neutral chelate ligand with
counter-anionic monodentate co-ligands like halides.
Received 8th May 2020,
Accepted 29th July 2020
DOI: 10.1039/d0nj02347g
rsc.li/njc
design for structure–property correlation studies aided by a care-
ful assembly of ligands, co-ligands and transition metal centres
Introduction
Synthetic molecular science remains a major gateway to tapping into new coordination materials is attractive and has been a key
into unmined benefits in materials science. Systematic ligand tool for further understanding desirable catalyst active site com-
positions and geometries.1 However, the reported studies often
a Inorganic Chemistry Unit, Department of Chemistry, University of Ibadan, Ibadan,
subject few, rather molecularly similar, coordination candidates
to catalytic studies, which limits the extent of what can be
learned.2 Therefore, we were motivated to investigate new, system-
atically varied catalysts under the same reaction conditions.
Azo-benzenes, which are more known for their applications
as dyes,3 optical materials,4 pharmaceuticals and bioactive
materials,5 are generally less deployed in metal coordination
architectures.6 Actinomycin-D has medicinal applications in
the treatment of various cancer cases and viral infections.7
Its biosynthesis is aided by the multi-copper phenoxazinone
Nigeria
b
¨
¨
Institut fur Anorganische und Analytische Chemie, Friedrich-Schiller-Universitat
Jena, Humboldtstraße 8, 07743 Jena, Germany. E-mail: sekr.plass@uni-jena.de;
Fax: +49 (0)3641 948132; Tel: +49 (0)3641 948130
c Department of Chemistry, School of Science, Adeyemi College of Education, Ondo,
Ondo State, Nigeria
d Materials Chemistry Group, Department of Chemical Sciences, Redeemer’s
University Ede, Osun State, Nigeria. E-mail: biodun.eseola@uni-jena.de,
bioduneseola@hotmail.com; Tel: +49 (0)15218016407
† Electronic supplementary information (ESI) available: CCDC 1986221 for (Na-Co1-i)2,
1986222 for (Na-Co1-e)2, 1986223 for (K-Co1-e)2, and 1986224 for (Cs-Co1-e)2. For ESI
and crystallographic data in CIF or other electronic format see DOI: 10.1039/d0nj02347g synthase via oxidative coupling of two units of o-aminophenol
New J. Chem.
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2020

View Article Online
Paper
NJC
suggests that such an assembly of ‘Co(II) + group I metal’
displayed doubtful cooperativity and this served as the motiva-
tion for further analyses. Furthermore, questions surrounding
ligand chelating capacity, ligand and metal centre charges, num-
ber and proximity of possible vacant coordination positions, etc.
are also aspects that deserve attention. For instance, information
could be obtained by comparing tridentate ligands with their
tetradentate analogues (compounds I and II, Scheme 1(b))14,15
as well as complexes bearing neutral chelate ligands along with
halides (compounds III and IV, Scheme 1(b))16 using the same
coupling procedures.
Therefore, in continuation of our efforts towards the deve-
lopment of synthetic phenoxazinone synthase models as well as
studying the effects of coordination environments and mole-
cular factors on favourable biomimetic activities, we herein
report the syntheses, structural characterization and biomi-
metic studies of some new cobalt(II) complexes stabilized by
azo-benzene ligand 1. Complexes of ligands 2 and 3 are also
incorporated for comparative purposes (Scheme 1(c)).
Experimental
General information
All starting materials for synthesis as well as substrates for the
catalytic experiments were obtained commercially as reagent
grade and used as supplied. The tridentate azo-benzene sulfo-
namide ligand 1 and the bis-imidazolephenol ligand 3 as well
as its complex Co3 (UV-Vis (CH3CN, l (nm), e (Mꢀ1 cmꢀ1)): 339
(e = 9340), 281 (e = 19 601)) were prepared as previously
reported.17 IR spectra were measured on a Bruker Equinox
FT-IR spectrometer equipped with a diamond ATR unit in the
range of 4000–600 cmꢀ1. Elemental analyses were either carried
out on a Leco CHNS-932 or on an El Vario III elemental
analyzer. Mass spectrometry (MS) analyses were conducted on
1
a Bruker MAT SSQ 710 spectrometer. H and 13C NMR spectra
were recorded on a Bruker AVANCE 400 MHz spectrometer
using deuterated solvents as internal standards.
Scheme 1 (a) Biosynthesis of actinomycin-D catalysed by phenoxazinone
synthase; (b) coupling rates of some reported cobalt complexes evaluated by
extrapolation at 2.5 ꢁ 10ꢀ3 M OAPH concentration (I,14 II,15 III,16 and IV16);
(c) structure of the o-sulfonamide and imidazole-phenol ligand architectures
utilized.
Synthesis of ligand 2
N,N0-Ditosylbenzene-1,2-diamine (2). Benzene-1,2-diamine
(4.0 g, 37.0 mmol) and 4-methylbenzene-1-sulfonyl chloride
(14.1 g, 74.0 mmol) were weighed and added into a 20 mL
derivatives (Scheme 1(a)).8 Thus, the importance of such a reac- microwave reactor vial equipped with a magnetic stirrer. Pyr-
tion and its industrial synthetic aspects are worthy of considera- idine (15 mL) was introduced to serve as the base catalyst as
tion and researchers have continued to design and investigate well as the solvent and the reactor vial was sealed, placed in the
several types of metal complexes as models.9 Among the investi- microwave synthesizer and exposed to microwave radiation for
gated metal species, cobalt complexes have less representation.10 four minutes at a setting of 120 1C. On cooling, the reaction
Efforts have continued to be made to obtain an understanding mixture was transferred into a quick fit round bottom flask and
of factors that influence biomimetic activities of reported the pyridine was removed under reduced pressure on a rotary
phenoxazinone synthase models.11 In particular, the coopera- evaporator (80 1C, 60 mbar). The residue dissolved in chloro-
tive benefit of metal centres in heterometallic complexes is still form was filtered over silica gel to exclude residual pyridine and
being reported.12 Scarborough, Borovik and others have attrib- pyridinium chloride using chloroform as the eluent. The pro-
uted improved coupling efficiencies to the presence of main duct fractions were combined, concentrated and treated with
group metal ions in the secondary coordination-sphere of some n-hexane to give a precipitate, which was dried to give ligand 2
cobalt(II) complexes.13 However, an on-going study in our group as a white solid (15.1 g, yield = 98%). MP 202 1C. Selected IR
New J. Chem.
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2020

View Article Online
NJC
Paper
peaks (ATR, cmꢀ1): n 3322m, 3218s, 1596m, 1498s, 1323vs, 551vs. obtained after one week. Yield (53 mg, 61%). MP = 311 1C.
1H NMR (400 MHz, d-DMSO): d 9.27 (s, 2H), 7.59 (d, J = 8.3 Hz, Selected IR data (ATR, cmꢀ1): v 2961m (alkyl), 1594m (CQC),
4H), 7.35 (d, J = 8.1 Hz, 4H), 6.98 (s, 4H), 2.36 (s, 6H). 13C NMR 1556s, 1377m, 1289m, 1228vs, 1117vs, 1031s, 944s, 800s, 736m,
(101 MHz, d6-DMSO): d 144.14, 136.49, 130.20, 130.12, 127.39, 653vs. MS (ESI) m/z 934 (1 + Co2+ + Cs, 100%, calc. = 934.20),
126.33, 123.82, 21.48. MS (EI) m/z 416 (M+, o100%): 416, 261, 182, 825 (1 + Co2+ + Na+, 60%), 761 (1 + H2O, 95%). Anal. calc. for
155. Anal. calc. for C20H20N2O4S2: C, 57.67; H, 4.84; N, 6.73; S,
15.40%. Found: C, 57.81; H, 4.80; N, 6.70; S, 15.56%.
C
92H130Co2Cs2N8O16S4ꢂH2O: C, 51.78; H, 6.23; N, 5.25; S, 6.01%.
Found: C, 51.22; H, 5.72; N, 5.43; S, 6.09%. UV-Vis (CH3CN,
l (nm), e (Mꢀ1 cmꢀ1)): 479 (e = 44 667), 331 (e = 64 657).
Co2. To a mixture of ligand 2 (100 mg, 0.2 mmol), cobalt(II)
Syntheses of the cobalt(II) complexes
(Na-Co1-i)2. Ligand 1 (70 mg, 0.094 mmol) and cobalt(II) acetate tetrahydrate (60 mg, 0.2 mmol) and tetrahydrofuran
acetate tetrahydrate (23 mg, 0.094 mmol) were reacted together (2 mL) in a glass vial was added 2 drops of pyridine. The vial
in 3 mL of ethanol. Pyridine (0.1 mL) was added followed by the was capped and allowed to stand overnight. Co2 was obtained
addition of NaOH solution (0.1 mL of 0.5 M in MeOH). The as self-assembled reddish crystals, which were filtered, washed
solution was then carefully layered with isopropanol and with a few drops of tetrahydrofuran and air-dried (50 mg, yield
allowed to stand under slow evaporation. Crystals of (Na-Co1-i)2 = 39%). MP 337 1C. Selected IR peaks (ATR, cmꢀ1): n 3050m,
suitable for X-ray measurement were obtained after three weeks. 1606m, 1579m, 1448s, 968vs, 547vs. MS (ESI) m/z 664 (Co2ꢂ
Yield (55 mg, 62%). MP = 247 1C. Selected IR data (ATR, cmꢀ1): MeOH, 20%, calc. = 664.60), 538 (2–py–Me, 25%), 439 (2,
v 3050w (Ar–H), 2958s (alkyl), 1589s (CQC), 1561s, 1442s, 1238vs, 100%). Anal. calc. for C30H28CoN4O4S2: C, 57.05; H, 4.47;
1123vs, 1031vs, 949s, 758vs, 658vs. MS (ESI) m/z 824 (1 + Co2+
Na+, 100%, calc. = 824.96), 760 (1 + H2O, 80%). Anal. calc. for 10.34%. UV-Vis (CH3CN, l (nm), e (Mꢀ1 cmꢀ1)): 569 (e = 623),
47H65CoN4NaO7S2: C, 59.79; H, 6.94; N, 5.93; S, 6.79%. Found: 414 (e = 898), 285 (e = 50 274), 245 (e = 18 268).
+
N, 8.87; S, 10.15%. Found: C, 57.16; H, 4.47; N, 8.79; S,
C
C, 59.66; H, 6.88; N, 5.98; S, 6.68%. UV-Vis (CH3CN, l (nm),
e (Mꢀ1 cmꢀ1)): 486 (e = 39 420), 332 (e = 56 972).
Catalytic oxidation of o-aminophenol to 2-aminophenoxazin-3-
one
(Na-Co1-e)2. Ligand 1 (70 mg, 0.094 mmol) and cobalt(II)
acetate tetrahydrate (23 mg, 0.094 mmol) were reacted in 3 mL The catalytic oxidation of o-aminophenol was carried out in
of EtOH together with 0.1 mL of NaOH solution (0.5 M in the presence of the prepared complexes (2.5 ꢁ 10ꢀ5 M) and
MeOH). The solution was allowed to stand under slow evapora- o-aminophenol (OAPH) (2.5 ꢁ 10ꢀ3 M) in acetonitrile under
tion. After one week, crystals of (Na-Co1-e)2 suitable for X-ray aerobic conditions at a thermostatted temperature of 25 1C.
measurement were obtained. Yield (53 mg, 58%). MP = 299 1C. The reaction was examined spectroscopically by following the
Selected IR data (ATR, cmꢀ1): v 2950w (alkyl), 1591w (CQC), increase in the absorbance of the phenoxazinone chromophore
1564m, 1284w, 1238s, 1125vs, 1033s, 950s, 836s, 753vs, 658s. as a function of time at 426 nm, which is the characteristic
MS (EI) m/z 906 (L + Co + OAcꢀ + Na+, 7%, calc. = 906.99), 824 band of 2-aminophenoxazin-3-one in acetonitrile.18 Absorbance
(1 + Co + Na+, 12%), 759, 616, 555, 497, 349, 269, 204, 189, 161. spectral scans of these solutions were recorded at regular time
MS (ESI) m/z 824 (1 + Co2+ + Na+, 100%, calc. = 824.24), 760 intervals of 30 minutes in the wavelength range of 600–300 nm
(1 + H2O, 100%). Anal. calc. for C46H63CoN4NaO7S2ꢂ(AcOH): C, for all the complexes. The initial rate method was used to
58.23; H, 6.82; N, 5.66; S, 6.48%. Found: C, 58.12; H, 6.42; determine the rate of reaction by linear regression from the
N, 6.01; S, 6.59%. UV-Vis (CH3CN, l (nm), e (Mꢀ1 cmꢀ1)): 487 slope of the absorbance against time plot.18
(e = 45 517), 333 (e = 56 989).
Structure determination
(K-Co1-e)2. Ligand 1 (70 mg, 0.094 mmol), cobalt(II) acetate
tetrahydrate (23 mg, 0.094 mmol) and KCl (7 mg, 0.094 mmol) The intensity data for the compounds were collected on a Nonius
were reacted together in 2 mL of EtOH. To this solution was Kappa CCD diffractometer using graphite-monochromated Mo-Ka
added 0.1 mL of triethylamine. The solution was allowed to radiation. Data were corrected for Lorentz and polarization
evaporate slowly and crystals of (K-Co1-e)2 suitable for X-ray effects; absorption was taken into account on a semi-empirical
measurement were obtained after 5 days. Yield (64 mg, 71%). basis using multiple-scans.19,20 The structures were solved using
MP = 305 1C. Selected IR data (ATR, cmꢀ1): v 2956m (alkyl), direct methods (SHELXS) and refined using full-matrix least
1592m (CQC), 1556m, 1405s, 1289s, 1228vs, 1159s, 1121vs, squares techniques against Fo2 (SHELXL-97).20 The hydrogen
1034vs, 946vs, 802s, 755vs, 648vs. MS (EI) m/z 840 (1 + Co + K, atoms bonded to the propan-2-ol hydroxy-group O7 of (Na-Co1-i)2
35%, calc. = 840.26), 776, 742 (1 – 2H, 40%), 632, 571, 515, 210. and to the water molecules O5 and O7 of (K-Co1-e)2 and (Cs-Co1-e)2,
MS (ESI) m/z 840 (1 + Co2+ + K, calc. = 840.26), 824 (1 + Co2+
+
respectively, were located by difference Fourier synthesis and refined
Na+, 100%), 760 (1 + H2O, 48%). Anal. calc. for C44H57Co- isotropically. All other hydrogen atoms were included at calculated
KN4O6S2: C, 58.71; H, 6.38; N, 6.22; S, 7.12%. Found: positions with fixed thermal parameters. The crystals of (Na-Co1-e)2,
C, 59.17; H, 6.40; N, 6.30; S, 6.94%. UV-Vis (CH3CN, l (nm), and (K-Co1-e)2 contain large voids, filled with disordered solvent
e (Mꢀ1 cmꢀ1)): 487 (e = 43 928), 331 (e = 64 417).
molecules. The sizes of the voids are 701 and 373 Å3 per unit cell,
(Cs-Co1-e)2. The complex was prepared in a similar way to respectively. Their contribution to the structure factors was secured
the preparation of (K-Co1-e) using CsCl (16 mg, 0.094 mmol) in via back-Fourier transformation using the SQUEEZE routine of the
place of KCl. Crystals suitable for X-ray measurement were program PLATON,21 resulting in 242 and 126 electrons per unit cell,
New J. Chem.
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2020

View Article Online
Paper
NJC
Table 1 Selected bond lengths (Å) and bond angles (1) around the metal
centres of the prepared complexes
coordinates as a dianionic, bidentate N^N donor, ligand 3
functions as a neutral, bidentate N^O chelator. Analytical char-
acterization data for the complexes agree with their molecular
compositions and all the studied complexes have been crystal-
lographically confirmed (see the ESI†). Isopropanol was pre-
ferentially coordinated to (Na-Co1-i)2, which was prepared in
the presence of both ethanol and isopropanol. It is also
noteworthy that the electron impact (EI) and electron spray
ionization (ESI) mass spectrometry data regularly indicate the
stability of the core, monomer Co1 units of complexes derived
from ligand 1 and this highlights the strong chelating effect
of the tridentate dianionic ligand 1. It also indicates the readiness
of the heteronuclear complexes (Na-Co1-i)2, (Na-Co1-e)2, (K-Co1-e)2
and (Cs-Co1-e)2 to dissociate in solution into their constituent ions
and the core cobalt(II) units (i.e., Co1).
Length (Å)/angles (1)
Bonds
(Na-Co1-i)2
(Na-Co1-e)2
(K-Co1-e)2
(Cs-Co1-e)2
Co1–N1
Co1–N2/N3
Co1–N4
1.976 (2)
2.053 (2)
1.993 (2)
1.962 (2)
7.833
2.002 (3)
2.062 (4)
1.982 (3)
1.952 (4)
7.739
2.006 (3)
2.081 (3)
2.052 (3)
2.127 (3)
8.030
2.013 (4)
2.081 (4)
2.055 (4)
2.037 (4)
8.195
Co1–O5/O6
Co–Co
MI–MI
3.214
3.361
4.605
5.441
N1–Co1–N4
133.4 (1)
136.8 (1)
130.9 (1)
133.1 (2)
respectively. All non-hydrogen and non-disordered atoms were
refined anisotropically.20 The XP program was used for structural
representation.22 The crystallographic data as well as structure
solution and refinement details are summarized in Table 1. The
structures of Co2 and Co3 are displayed in the ESI,† Fig. S1 and
S2, respectively.
X-ray characterization and continuous shape measurement
(CShM) analyses
Suitable crystals for X-ray measurement were obtained for all
the complexes under slow evaporation from or slow solvent
diffusion into the solutions of the complexes in ethanol or
ethanol/isopropanol. Their structural diagrams are shown in
Results and discussion
Syntheses and characterization of the cobalt complexes
The cobalt(II) complexes (Scheme 2) were synthesized at room Fig. 1 and 2 while the corresponding structural refinement data
temperature by reacting cobalt(II) acetate tetrahydrate with any of and processing parameters are presented in Table S1 of the
the ligands 1–3 in certain solvent systems. All reactions with ESI.† Complexes (Na-Co1-e)2, (K-Co1-e)2 and (Cs-Co1-e)2 crystallize
%
%
%
ligand 1 were carried out in the presence of a varying group I in the triclinic space groups P1, P1 and P1, respectively, while the
metal base or halide (i.e., sodium hydroxide, potassium chloride compound (Na-Co1-i)2 crystallizes in the monoclinic space
or caesium chloride). The complexes were obtained in moderate group P21/n. These structures reveal the stable, tridentate
to high yields. While ligand 1 contains the cobalt(II) metal centres N^N^N chelation of ligand 1 to the cobalt(II) centres in five-
as a dianionic, tridentate N^N^N ligand backbone and ligand 2 and six-membered chelate rings.
Scheme 2 Compositions of the isolated cobalt(II) complexes.
New J. Chem.
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2020

View Article Online
NJC
Paper
It is noteworthy that the two cobalt centres of complexes four-coordinate complexes Co2 and Co3 were in agreement
(Na-Co1-i)2 and (Na-Co1-e)2 (i.e., the Co1 cores) were found to with a distorted tetrahedron and a vacant trigonal bipyramid,
be in unusual four-coordinate geometries (Fig. 1), where the respectively (Table 2(c)). The little difference between distor-
acetate O-donor anion occupied the fourth coordination posi- tions from ideal tetrahedron (1.897%) and vacant trigonal
tion, while (K-Co1-e)2 and (Cs-Co1-e)2 have trigonal bipyramid bipyramid (1.376%) for Co3 is noteworthy.
Co1 cores, where water and acetate O-donors occupy the fourth
Analysis of the geometric distortion of the coordination
and fifth coordination positions (Fig. 2). It is also noteworthy environment of complex (Na-Co1-i)2 by minimal distortion
that the O-donors from the sulfonamide –SO2, acetate or water inter-conversion pathway analysis25 between square planar
serve as bridging donors, which aid the dimer assembly of two (D4h) and vacant trigonal bipyramid (C3v) indicated that the
Co1 units into the (MI-Co1-solvate)2 architectures (Fig. 1 and 2). deviation from the ideal polyhedra may be attributed to rota-
The bond lengths and angles around the coordination polyhe- tion of the coordination planes of the complex from C3v (vacant
dra in the structures of the complexes are within the expected trigonal bipyramid) to D4h (square planar) (Table 3).26 However,
values and the diazene –NQN– bond lengths range from 1.264 a possible contribution from elongation of the trigonal bipyr-
to 1.279 Å (Table 1).23 Based on the Co–Co and MI–MI distances, amidal coordination polyhedra cannot be ruled out. It could be
the general space between the paired Co1 cores in the four concluded that the isolated complexes varied broadly in coor-
complexes of ligand 1 can be seen to increase as the ionic radii dination geometries, ligand chelation characters and ligand
of the group I ions increase (Table 1).
charges.
Results from continuous shape measure (CShM)24 calcula-
tions carried out on input files generated from the crystal
Phenoxazinone synthase mimicking studies
atomic positions define the four-coordinate Co1 environments Estimation of the phenoxazinone synthase-like capacity of the
of complexes (Na-Co1-i)2 and (Na-Co1-e)2 as vacant trigonal complexes was typically done in acetonitrile under aerobic
bipyramids (Table 2(a)) while defining the five-coordinate Co1 conditions and at 25 1C by running experiments on the oxida-
environments of (K-Co1-e)2 and (Cs-Co1-e)2 as trigonal bipyr- tive coupling of 2-aminophenol (OAPH, 2.5 ꢁ 10ꢀ3 M) in the
amids (Table 2(b)). The significant distortion of the unusual, presence of the complexes (2.5 ꢁ 10ꢀ5 M). The formation of the
vacant trigonal bipyramidal geometries of (Na-Co1-i)2 and (Na- 2-aminophenoxazin-3-one product was spectroscopically followed at
Co1-e)2 is noteworthy. The corresponding calculations for the about its absorption maximum of 426 nm. Illustrations of spectral
Fig. 1 Structures (top) and the corresponding metal coordination environments (bottom) of (Na-Co1-i)2 and (Na-Co1-e)2 with thermal ellipsoids drawn
at the 30% probability level. In order to improve clarity, some bonds are presented as dashes while some protons, solvent molecules and atom labels have
been omitted.
New J. Chem.
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2020

View Article Online
Paper
NJC
Fig. 2 Structures of (K-Co1-e)2 and (Cs-Co1-e)2 with thermal ellipsoids drawn at the 30% probability level. In order to improve clarity, some bonds are
presented as dashes while some protons, solvent molecules and atom labels have been omitted.
Table 2 Calculated CShM parameters for the complexes
426 nm while spectral changes in the presence of the complexes
confirmed that all the complexes yielded active catalysts for the
(a)
(Na-Co1-i)2
(Na-Co1-e)2
oxidative coupling of OAPH. The initial rate values were used to
compare catalytic performance. It is important to note that the
strong visible absorption of the diazene compounds is responsible
for the originally high absorbance observed at the inception of
catalyses conducted in the presence of the (MI-Co1-solvate)2
complexes (e.g., Fig. 3(a) and (b)). This high ligand absorbance
is absent in the spectra of reactions conducted with Co2 and Co3
(e.g., Fig. 3(c)).
SP-4 (D4h
)
18.882
5.976
4.874
3.481
16.959
6.771
4.465
4.407
T-4 (Td)
SS-4 (C2v
)
v-TBPY-4 (C3v
)
(b)
(K-Co1-e)2
(Na-Co1-e)2
PP-5 (D5h
vOC-5 (C4v
TBPY-5 (D3h
SPY-5 (C4v
)
30.150
5.595
1.653
4.982
3.635
29.204
5.295
1.799
4.813
3.866
)
)
)
Comparison of the phenoxazinone synthase-like activity of the
complexes
JTBPY-5 (D3h
)
(c)
Co2
Co3
Since the comparison of the catalytic influence of the various
cobalt complexes is principal to this study and since a linear
relationship exists between absorbance and the amount of a
solute, it was considered sufficient to estimate and compare the
initial rates of absorbance changes for the catalytic reactions.
This was achieved by linear regression fitting at the origin of
the absorbance change versus time plot.18 Using a similar
reaction setting (1 mol% catalyst, 25 1C, acetonitrile media),
the plots of absorbance change versus time for the catalytic and
blank runs were recorded and are represented in Fig. 4 while
the corresponding initial rates of absorbance increase are
reported in Table 4. It could be observed that the studied
complexes possess varying degrees of phenoxazinone synthase
mimicking activity when compared with the blank experiment,
where only autoxidation in the absence of a metal complex took
place. The temperature effect also indicated increasing activ-
ities as the temperature of the oxidative coupling increased
(Fig. 5(a)).
SP-4 (D4h
)
25.319
3.040
8.349
4.835
29.655
1.897
8.232
1.376
T-4 (Td)
SS-4 (C2v
)
v-TBPY-4 (C3v
)
SP-4 (D4h) = square planar; T-4 (Td) = tetrahedron; SS-4 (C2v) = seesaw;
vTBPY-4 (C3v) = vacant trigonal bipyramid; PP-5 (D5h) = pentagon; vOC-5
(C4v) = vacant octahedron; TBPY-5 (D3h) = trigonal bipyramid; SPY-5 (C4v) =
spherical square pyramid; JTBPY-5 (D3h) = Johnson trigonal bipyramid J12.
Table 3 Minimal distortion path analysis from SP-4 (0%) to vTBPY-4
(100%) for (Na-CoL-i)2
(Na-CoL-i)2
S (D4h
S (C3v
)
)
18.9
3.5
0.9
DevPath
Gencoord
71.2
SP-4 (D4h) = square planar; vTBPY-4 (C3v) = vacant trigonal bipyramid;
DevPath = path deviation; Gencoord = generalized coordinate; deviation
threshold to calculate generalized coordinate = 10%.
An important feature that can be observed in Fig. 4 and Table 4
is the similarity of phenoxazinone synthase mimicking behaviour
stacks of UV-Vis scans, which were collected at 30 min intervals of the four complexes (Na-Co1-i)2, (Na-Co1-e)2, (K-Co1-e)2 and (Cs-
for several hours after addition of OAPH, are presented in Co1-e)2, which suggests that the differing group I metal ions
Fig. 3(a)–(c) for (K-Co1-e)2, (Cs-Co1-e)2 and Co2, respectively. played no tangible role in the oxidative coupling of OAPH. The
Blank experiments under the same reaction conditions did not similarity in catalyst performance of these complexes points to
show a tangible increase in the phenoxazinone band intensity at the generation of Co1 in solution as the active species, which is
New J. Chem.
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2020

View Article Online
NJC
Paper
Fig. 3 Stacks of UV-vis spectral scans showing the increase in the phenoxazinone chromophore band at 426 nm after the addition of OAPH (2.5 ꢁ 10ꢀ3 M)
to a cobalt complex solution (2.5 ꢁ 10ꢀ5 M). (a) (K-Co1-e)2; (b) (Cs-Co1-e)2; (c) Co2.
essentially common to the four complexes. The ESI mass spectro- sites around the cobalt(II) centre. A similar lack of support from
metry data obtained during characterization of the complexes also group I elements and the limiting influence of coordinative
agree with dissociation of the multinuclear (MI-Co1-solvate)2 saturation have also been observed in our recent study27 and
complexes into Co1 and M1+ species.
the chelation influence could also be inferred by comparing data
Furthermore, the higher catalytic performance of Co2 (i.e., in the literature (e.g., Scheme 1(b), I versus II or III versus IV).14–16
199 ꢁ 10ꢀ4
h
ꢀ1) relative to the multinuclear (MI-Co1-solvate)2
Finally, in light of the fact that the structure of Co3 appears
complexes (i.e., 62 ꢁ 10ꢀ4–86 ꢁ 10ꢀ4 hꢀ1) could be attributed to to possess a free, uncrowded coordination environment around
the increase in number of vacant coordination sites offered by the cobalt(II) centre, it is surprising that the dichloride complex
Co2 (Fig. 4). That is, while the bidentate chelation of the yielded only a barely higher phenoxazinone synthase-like activ-
dianionic ligand 2 represents lower coordinative saturation, ity than the blank experiment (Fig. 4 and Table 4). However,
the tridentate chelation from ligand 1 implies a relatively since the bidentate chelation would stabilize the neutral ligand
higher limitation of substrate access to active site. This conclu- 3 on the cobalt(II) centre and the anionic charge of the chlorides
sion is considered plausible since both ligands 1 and 2 have would also sustain their coordination, it could be suggested
similar dianionic charge characters and N-donor strengths, but that the stability of the tetrahedron prevents interaction
differ in the number of donor atoms that occupy coordination between Co3 and the substrates. This character of complex
Co3 strengthens the thinking that co-ligands are dissociated in
the (MI-Co1-solvate)2 and Co2 complexes.
In attempts to force chloride dissociation from Co3, the same
oxidative coupling experiments were repeated in the presence
of 1 mol equivalent of potassium or caesium carbonate and the
Table 4 Catalytic performance of the complexes in phenoxazinone
synthase-like activity
Co(II) complex
Rate (ꢁ 10ꢀ4 hꢀ1
)
(Na-Co1-i)2
(Na-Co1-e)2
(K-Co1-e)2
(Cs-Co1-e)2
Co2
Co3
No complex
70.0 ꢃ 1.0
73.6 ꢃ 1.2
85.8 ꢃ 1.4
62.7 ꢃ 1.3
199.1 ꢃ 3.7
10.4 ꢃ 0.0
7.1 ꢃ 0.5
Fig. 4 Plot of absorbance vs. time for the phenoxazinone synthase-like activity
of the complexes at 25 1C (OAPH = 2.5 ꢁ 10ꢀ3 M, complex = 2.5 ꢁ 10ꢀ5 M).
New J. Chem.
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2020

View Article Online
Paper
NJC
Fig. 5 (a) Plot of absorbance against time for mimicking activity of (K-Co1-e)2 at different temperatures and (b) coupling in the presence of potassium or
caesium carbonates indicating that removal of chloride in Co3 enables improved active site generation ([OAPH] = 2.5 ꢁ 10ꢀ3 M, [complex] = 2.5 ꢁ 10ꢀ5 M).
resultant yield of 2-aminophenoxazin-3-one was dramatically Electrospray ionization (ESI) mass spectrometry studies
increased within 2 hours relative to outcomes in the absence of
the carbonates (Fig. 5(b)). Although the presence of a base medium
solution in order to possibly probe the mechanism of the biomi-
alone stimulated the oxidative coupling of o-aminophenol, reactivity
in the presence of ‘Co3 + K2Co3’ or ‘Co3 + Cs2Co3’, which was
significantly higher and above that of the base-only experiments,
For the purpose of studying the nature of species formed in
metic activity, electrospray ionization spectral measurements (ESI-
MS positive and negative) were carried out on solutions of a 1 : 50
mixture of (Na-Co1-i)2 or (K-Co1-e)2 and OAPH in acetonitrile after
provides evidence that chloride dissociation enabled the observed
dramatic increase in activity (Fig. 5(b)). It could be concluded that
coordination to anionic chelate ligands like 1 and 2 supported by
neutral monodentate co-ligands like water or pyridine is desirable
for the ease of generating active species as opposed to coordination
with a neutral chelating ligand like 3 supported by anionic mono-
dentate co-ligands.
stirring for 30 minutes. Interestingly, the mass spectrometry
results for these reaction mixtures containing the biomimetic
models (Na-Co1-i)2 and (K-Co1-e)2 show similar m/z peaks for both
positive and negative ESI mode measurements. This supports
the earlier conclusion that the (MI-Co1-solvate)2 complexes
carry out their biomimetic activities through generation of
the same core Co1 active species and via the same mechanistic
Scheme 3 Proposed mechanism for the catalytic oxidation of o-aminophenol to 2-aminophenoxazin-3-one under aerobic conditions by (K-Co1-e)2.
New J. Chem.
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2020

View Article Online
NJC
Paper
pathway of substrate–complex intermediate formation that Co1 species generated from dissociation of the multinuclear
precludes the group I metal ions. The prominent peaks at (MI-Co1-solvate)2 complexes is responsible for the observed
m/z = 824.2 and 865.8 represent the sodium and solvent adducts outcomes. It was also concluded that the presence of halide
of the ligand–metal aggregate of the form [L + Co + Na]+ and donors blocked access of substrates to metal binding even
[L + Co + Na + MeCN]+, respectively. Furthermore, the peak
when the metal centre is not crowded by organic chelate
representing the formation of the substrate–complex intermediate
can be found at m/z = 955.1, which is assigned to [L + Co + BQMI +
2Na]+ (BQMI = o-benzoquinone monoamine). The peak at m/z =
212.0 represents the 2-aminophenoxazin-3-one product.
ligands, as for complex Co3. High oxidative coupling efficien-
cies obtained after using potassium or caesium carbonate to
remove the chlorides lend support to this conclusion.
Therefore, it could be concluded that coordination to anionic
chelate ligands like 1 and 2 supported by easily detachable,
neutral co-ligands like water or pyridine favours the easy genera-
tion of active species in contrast to complexes bearing neutral
chelates like 3 supported by anionic monodentate co-ligands.
In the ESI negative mode spectra, the most noticeable of
the peaks is the one at m/z = 910.2, which is assignable to
the substrate–complex aggregate of composition [(L + Co) +
OAPH]+. Furthermore, it is noteworthy that the reactive inter-
mediate, o-benzoquinone monoamine (BQMI), is observed at
m/z = 107.9 while the peak at m/z = 236.1 also represents the
sodium adduct of 2-aminophenoxazin-3-one. Therefore, it can
be concluded that the catalytic cycle for the aerobic oxidation of
OAPH to 2-aminophenoxazin-3-one catalysed by these dimeric
complexes proceeded by the formation of a stable substrate–
complex intermediate through the coordination of OAPH to the
cobalt active centre.
Conflicts of interest
There are no conflicts to declare.
Acknowledgements
HOO is grateful to the Government of the Federal Republic of
Nigeria for financial support through TETfund research grants
(TETfund AST&D year 2012 intervention) and to Adeyemi Col-
lege of Education, Ondo, Nigeria, for granting study leave. AOE
is thankful to the Alexander von Humboldt Foundation for
granting a post-doctoral scholarship. The financial provision by
the Deutsche Forschungsgemeinschaft (DFG) is thankfully
acknowledged (PL155/11, PL155/12 and PL155/13).
Having obtained the substrate–complex intermediate in the
rate determining step, the OAP radical would be subsequently
formed under aerobic condition. This would then lead to the
formation of BQMI from the OAP radical via disproportionation
of the OAP radical itself. The BQMI produced then reacted with
another OAPH to form 2-aminophenoxazin-3-one through oxi-
dative dehydrogenation of the intermediate I, as proposed in
Scheme 3 using the complex (K-Co1-e)2 for illustration.
References
Conclusion
1 (a) F. Li, Y. Bu, G.-F. Han, H.-J. Noh, S.-J. Kim, I. Ahmad,
Y. Lu, P. Zhang, H. Y. Jeong, Z. Fu, Q. Zhong and J.-B. Baek,
Nat. Commun., 2019, 10, 2623–2629; (b) G.-G. Luo, H.-L.
Zhang, Y.-W. Tao, Q.-Y. Wu, D. Tian and Q. Zhang, Inorg.
Two dianionic disulfonamides, 1 (N^N^N) and 2 (N^N), as well
as the neutral imidazole-phenol 3 (N^O) have been synthesized
and used to prepare the structurally diverse cobalt(II) complexes
(Na-Co1-i)2, (Na-Co1-e)2, (K-Co1-e)2, (Cs-Co1-e)2, Co2 and Co3.
Ligand 1 forms multinuclear complexes incorporating Na+, K+
or Cs+. Based on X-ray data and Continuous Shape Measure
(CShM) calculations, the cobalt(II) coordination geometries
were an unusual vacant trigonal bipyramid for (Na-Co1-i)2, (Na-
Co1-e)2 and Co3, trigonal bipyramid for complexes (K-Co1-e)2 and
(Cs-Co1-e)2, and tetrahedron for Co2. Mass spectrometry and
catalytic results revealed that the four multinuclear (MI-Co1-
solvate)2 complexes of ligand 1 predominantly dissociate in
solution to yield the same Co1 cores as the active species.
Relative to blank experiments, all the complexes possess
varying extents of phenoxazinone synthase-like activity, which
generally increased as temperature increased. The significantly
higher catalytic performance effected by the bidentate ligand 2
¨
Chem. Front., 2019, 6, 343–354; (c) C. Schottle, E. Guan,
A. Okrut, N. A. Grosso-Giordano, A. Palermo, A. Solovyov,
B. C. Gates and A. Katz, J. Am. Chem. Soc., 2019, 141,
4010–4015; (d) E. Peris and R. H. Crabtree, Chem. Soc.
Rev., 2018, 47, 1959–1968; (e) S. K. Dey and A. Mukherjee,
Coord. Chem. Rev., 2016, 310, 80–115.
2 (a) A. Panja and P. Guionneau, Dalton Trans., 2013, 42,
5068–5075; (b) A. Panja, Dalton Trans., 2014, 43, 7760–7770.
3 O. A. Blackburn, B. J. Coe, J. Fielden, M. Helliwell, J. J. W.
McDouall and M. G. Hutchings, Inorg. Chem., 2010, 49,
9136–9150.
4 B. Tylkowski, A. Trojanowska, V. Marturano, M. Nowak,
L. Marciniak, M. Giamberini, V. Ambrogi and P. Cerruti,
Coord. Chem. Rev., 2017, 351, 205–217.
of Co2 (199 ꢁ 10ꢀ4
h
ꢀ1) relative to the tridentate ligand 1 of the
5 (a) K. G. Samper, S. C. Marker, P. Bayon, S. N. MacMillan,
´
Co1 species (62 ꢁ 10ꢀ4–86 ꢁ 10ꢀ4 hꢀ1) revealed that coordinative
saturation is detrimental. Furthermore, the notable similarity in
catalytic performance by complexes (Na-Co1-i)2, (Na-Co1-e)2,
(K-Co1-e)2 and (Cs-Co1-e)2 indicated no catalytic contribution
from the presence of differing group I metal ions and that the
I. Keresztes, O. Palacios and J. J. Wilson, J. Inorg. Biochem.,
`
2017, 174, 102–110; (b) F. Hamon, F. Djedaini-Pilard,
F. Barbot and C. Len, Tetrahedron, 2009, 65, 10105–10123.
˜
´
´
6 S. Banerjee, P. Brandao, A. Bauza, A. Frontera, M. Barcelo-Oliver,
A. Panja and A. Saha, New J. Chem., 2017, 41, 11607–11618.
New J. Chem.
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2020

View Article Online
Paper
NJC
7 (a) M. Liu, Y. Jia, Y. Xie, C. Zhang, J. Ma, C. Sun and J. Ju, 15 M. Mahato, K. van Hecke and H. P. Nayek, Appl. Organomet.
Mar. Drugs, 2019, 17, 1–13; (b) R. L. Rill and K. H. Hecker, Chem., 2018, 32, e4336–e4344.
Biochemistry, 1996, 35, 3525–3533; (c) C. Chen, F. Song, 16 A. Jana, P. Brandao, H. Jana, A. D. Jana, G. Mondal, P. Bera,
˜
Q. Wang, W. M. Abdel-Mageed, H. Guo, C. Fu, W. Hou,
H. Dai, X. Liu, N. Yang, F. Xie, K. Yu, R. Chen and L. Zhang,
A. Santra, A. K. Mahapatra and P. Bera, J. Coord. Chem.,
2019, 72, 2636–2653.
¨
Appl. Microbiol. Biotechnol., 2012, 95, 919–927; (d) T. Das, 17 (a) H. O. Oloyede, J. A. Orighomisan Woods, H. Gorls, W. Plass
R. R. Nair, R. Green, S. Padhee, M. Howell, J. Banerjee,
S. S. Mohapatra and S. Mohapatra, Anticancer Res., 2017, 37,
1655–1663.
and A. O. Eseola, J. Mol. Struct., 2020, 1199, 127030;
¨
(b) A. O. Eseola, H. Gorls, M. Bangesh and W. Plass, New
J. Chem., 2018, 42, 7884–7900.
¨
8 (a) A. W. Smith, A. Camara-Artigas, M. Wang, J. P. Allen and 18 H. O. Oloyede, J. A. O. Woods, H. Gorls, W. Plass and
W. A. Francisco, Biochemistry, 2006, 45, 4378–4387; A. O. Eseola, New J. Chem., 2019, 43, 18322–18330.
(b) C. E. Valdez, Q. A. Smith, M. R. Nechay and 19 (a) B. V. Nonius, COLLECT, Data Collection Software, Nether-
A. N. Alexandrova, Acc. Chem. Res., 2014, 47, 3110–3117.
9 (a) S. K. Dey and A. Mukherjee, J. Mol. Catal. A: Chem., 2015,
407, 93–101; (b) P. Mahapatra, M. G. B. Drew and A. Ghosh,
Cryst. Growth Des., 2017, 17, 6809–6820; (c) N. Sarkar,
M. Das and S. Chattopadhyay, Inorg. Chim. Acta, 2017,
457, 19–28.
lands, 1998; (b) Z. Otwinowski and W. Minor, in Macro-
molecular Crystallography Part A. Processing of X-Ray Diffraction
Data Collected in Oscillation Mode, ed. C. W. Carter and
R. M. Sweet, Academic Press, 1997; (c) Bruker-AXS Inc.,
SADABS, Madison, WI, U.S.A., 2002.
20 G. M. Sheldrick, Acta Crystallogr., Sect. A: Found. Crystallogr.,
2008, 64, 112–122.
´
10 (a) K. Ghosh, S. Roy, A. Ghosh, A. Banerjee, A. Bauza,
A. Frontera and S. Chattopadhyay, Polyhedron, 2016, 112, 21 A. L. Spek, Acta Crystallogr., Sect. C: Struct. Chem., 2015, 71,
6–17; (b) A. Hazari, A. Das, P. Mahapatra and A. Ghosh, 9–18.
Polyhedron, 2017, 134, 99–106; (c) M. Mahato, D. Mondal 22 Interactive Molecular Graphics Program, Siemens Analytical
and H. P. Nayek, ChemistrySelect, 2016, 1, 6777–6782.
X-ray Instruments Inc., Siemens (1994), XP., Madison, Win-
scosin, USA, 1994.
˜
11 (a) A. Panja, N. C. Jana and P. Brandao, New J. Chem., 2017,
41, 9784–9795; (b) P. Mahapatra, S. Ghosh, S. Giri, V. Rane, 23 (a) Z. H. Chohan and H. A. Shad, Appl. Organomet. Chem.,
R. Kadam, M. G. B. Drew and A. Ghosh, Inorg. Chem., 2017,
56, 5105–5121.
12 (a) Y. J. Park, J. W. Ziller and A. S. Borovik, J. Am. Chem. Soc.,
2011, 133, 9258–9261; (b) Y. Sano, A. C. Weitz, J. W. Ziller,
2011, 25, 591–600; (b) V. G. Vlasenko, D. A. Garnovskii,
G. G. Aleksandrov, N. I. Makarova, S. I. Levchenkov, A. L.
Trigub, Y. V. Zubavichus, A. I. Uraev, Y. V. Koshchienko and
A. S. Burlov, Polyhedron, 2019, 157, 6–17.
M. P. Hendrich and A. S. Borovik, Inorg. Chem., 2013, 52, 24 (a) J. Cirera, E. Ruiz and S. Alvarez, Organometallics, 2005,
10229–10231.
24, 1556–1562; (b) M. Pinsky and D. Avnir, Inorg. Chem.,
1998, 37, 5575–5582.
13 (a) C. M. Wallen, M. Wielizcko, J. Bacsa and C. C.
Scarborough, Inorg. Chem. Front., 2016, 3, 142–149; 25 D. Casanova, J. Cirera, M. Llunell, P. Alemany, D. Avnir and
(b) C. M. Wallen, J. Bacsa and C. C. Scarborough, Inorg. S. Alvarez, J. Am. Chem. Soc., 2004, 126, 1755–1763.
Chem., 2018, 57, 4841–4848; (c) D. C. Lacy, Y. J. Park, 26 (a) S. Alvarez, P. Alemany, D. Casanova, J. Cirera, M. Llunell
J. W. Ziller, J. Yano and A. S. Borovik, J. Am. Chem. Soc.,
2012, 134, 17526–17535.
and D. Avnir, Coord. Chem. Rev., 2005, 249, 1693–1708;
(b) S. Alvarez, Chem. Rev., 2015, 115, 13447–13483.
¨
14 K. Ghosh and S. Chattopadhyay, Polyhedron, 2019, 170, 27 H. O. Oloyede, J. A. O. Woods, H. Gorls, W. Plass and
495–507.
A. O. Eseola, Inorg. Chim. Acta, 2020, 509, 11964–11972.
New J. Chem.
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2020