Strategic synthesis of [Cu2], [Cu4] and [Cu5] complexes: inhibition and triggering of ligand arm hydrolysis and self-aggregation by chosen ancillary bridges

Article abstract of DOI:10.1039/c8dt03390k

The Schiff base ligand HL1 ({2,6-bis(allylimino)methyl}-4-methylphenol) having no coordinating donor arm has been examined for its reaction medium and ancillary bridge dependent reactivity for a hierarchical family of Cu^II complexes. The ligand showed a unique reactivity pattern toward Cu^II in solution. The bridging nature of in situ generated HO^? ions in the absence and presence of externally added carboxylates (RCOO^?; R = CF₃, C₆H₅ and CH₃) has been utilized to produce complexes {[Cu₂(μ-L2)₂(H₂O)]₂[Cu₂(μ-L2)₂(H₂O)₂](ClO₄)₆} (1) (HL2 = 3-{(allylimino)methyl}-2-hydroxy-5-methylbenzaldehyde), [Cu₄(μ₄-O)(μ-L1)₂(μ_1,3-O₂CCF₃)₄] (2), [Cu₄(μ₄-O)(μ-L1)₂(μ_1,3-O₂CC₆H₅)₄]·H₂O (3), and [Cu₅(μ₃-OH)₂(μ-L1)₂(μ_1,3-OAc)₂(OAc)₂(H₂O)₄][Cu₅(μ₃-OH)₂(μ-L1)₂(μ_1,3-OAc)₂(OAc)₃(H₂O)](ClO₄)₃·2C₂H₅OH (4). The absence of carboxylate anions did not yield HO^? ions in situ and triggered single ligand arm hydrolysis. The formation of tetra- and pentanuclear aggregates as well as ligand hydrolyzed dinuclear products has been rationalized to identify the possible roles of carboxylate anions in solution. Detailed characterization of the complexes in the solid state and in solution has been carried out using spectroscopic measurements, X-ray crystallography, variable temperature magnetic measurements and functional behavior. In MeOH solutions at 298 K, the complexes 1-4 showed catalytic oxidation of 3,5-di-tert-butyl catechol (3,5-DTBCH₂) saturated with O₂ from the air.

Full text of DOI:10.1039/c8dt03390k

Dalton
Transactions
View Article Online
View Journal | View Issue
PAPER
Strategic synthesis of [Cu ], [Cu ] and [Cu ]
2
4
5
complexes: inhibition and triggering of ligand arm
hydrolysis and self-aggregation by chosen
ancillary bridges†
Cite this: Dalton Trans., 2018, 47,
17160
a
b
c
b
Manisha Das, Angelos B. Canaj,
Valerio Bertolasi, Mark Murrie
and
a
Debashis Ray
*
The Schiff base ligand HL1 ({2,6-bis(allylimino)methyl}-4-methylphenol) having no coordinating donor
arm has been examined for its reaction medium and ancillary bridge dependent reactivity for a hierarchi-
II
II
cal family of Cu complexes. The ligand showed a unique reactivity pattern toward Cu in solution. The
bridging nature of in situ generated HO− ions in the absence and presence of externally added
carboxylates (RCOO−;
[Cu (µ-L2) (H O)] [Cu
benzaldehyde), [Cu (µ
(µ1,3-OAc)
4). The absence of carboxylate anions did not yield HO− ions in situ and triggered single ligand arm
R
=
CF
(H O)
-O)(µ-L1) (µ1,3-O
(OAc) (H O)
,
C
H
and CH
(1) (HL2
] (2), [Cu
(µ -OH) (µ-L1)
)
has been utilized to produce complexes
3-{(allylimino)methyl}-2-hydroxy-5-methyl-
(µ -O)(µ-L1) (µ1,3-O CC ]·H O (3), and
(µ1,3-OAc) (OAc) (H O)](ClO ·2C OH
3
6
5
3
{
2
2
2
2
2
(µ-L2)
2
2
2
](ClO
CCF
][Cu
4
)
6
}
=
4
4
2
2
3
)
4
4
4
2
2
6
H
5
)
4
2
[
(
Cu
5
(µ
3
-OH)
2
(µ-L1)
2
2
2
2
4
5
3
2
2
2
3
2
4
)
3
2 5
H
hydrolysis. The formation of tetra- and pentanuclear aggregates as well as ligand hydrolyzed dinuclear
products has been rationalized to identify the possible roles of carboxylate anions in solution. Detailed
characterization of the complexes in the solid state and in solution has been carried out using spectro-
scopic measurements, X-ray crystallography, variable temperature magnetic measurements and func-
tional behavior. In MeOH solutions at 298 K, the complexes 1–4 showed catalytic oxidation of 3,5-
Received 19th August 2018,
Accepted 12th November 2018
DOI: 10.1039/c8dt03390k
rsc.li/dalton
2 2
di-tert-butyl catechol (3,5-DTBCH ) saturated with O from the air.
hyde via hydrolysis. Quinone based Schiff bases are similar to
the aldehyde-based congeners in terms of their hydrolytic sus-
Introduction
Amine oxidases (CuAOs) containing a copper(II) ion and 2,4,5- ceptibility. The presence of a protein bound Cu(II) center plays
trihydroxyphenylalanine quinone (TPQ) are now known to be a crucial role in the hydrolysis of the product Schiff base.
responsible for the breakdown and regulation of biologically Using synthetic transition metal ion complexes, we and others
active amines in several organisms where the metal ion center have shown that the pH of the medium is crucial for the
1
does not play any redox role in catalysis. The catalytic process coordination-induced hydrolysis and release of free amines
2
generates an aldehyde from a wide range of primary amines in similar to the case of CuAOs. At pH < 7 the Schiff base sub-
the reductive half reaction, in which Cu(II)–TPQOX reacts with strate is prone to hydrolytic cleavage of the imine bond where
amines to form the quinoneimine ‘substrate Schiff base’. Next protonation of the imine nitrogen atom makes the adjacent
α-carbon proton abstraction leads to the quinolaldimine carbon electrophilic in nature for attack from a water molecule
‘product Schiff base’ responsible for the release of the alde- coordinated to the Cu(II) center. In this work we intend to
establish the hydrolytic susceptibility of a phenol-based alde-
hyde Schiff base in the presence of copper(II) ions, similar to
the hydrolysis of a quinone Schiff base in CuAOs.
a
Department of Chemistry, Indian Institute of Technology, Kharagpur 721302, India.
E-mail: dray@chem.iitkgp.ac.in; Fax: (+91) 3222-82252; Tel: (+91) 3222-283324
b
Thus, during the synthesis of multinuclear coordination
compounds, it is vital to recognize and interpret the self-aggre-
gation processes of different fragments in solution at room
WestCHEM, School of Chemistry, University of Glasgow, University Avenue,
Glasgow, G12 8QQ, UK
c
Dipartimento di ScienzeChimiche e Farmaceutiche, University of Ferrara,
4
4121 Ferrara, Italy
II
temperature for the creation of Cu -based compounds in a
†
Electronic supplementary information (ESI) available: X-ray crystallographic
phenoxido-bridging and imine supporting environment. The
role of ligand design and tuning of the reaction protocol with
the use of ancillary ligands in controlling the ligand stability
data in CIF format, Schemes S1 and S2, Fig. S1–S17, and Table S1. CCDC
857609, 1857607, 1857608 and 1857610 for 1–4. For ESI and crystallographic
data in CIF or other electronic format see DOI: 10.1039/c8dt03390k
1
17160 | Dalton Trans., 2018, 47, 17160–17176
This journal is © The Royal Society of Chemistry 2018

View Article Online
Dalton Transactions
Paper
are crucial to understand and control the aggregation process.
Understanding of the processes is dependent on the knowl-
edge of the solution stability and structure of the ligand
3
bound mononuclear and dinuclear fragments. The search for
a new ligand system is in progress for the synthesis and struc-
tural identification of [Cu
their participation in catalytic CH
2
] and [Cu
4
] complexes because of
oxidation. The latter
4
Scheme 1 Partial imine arm hydrolysis.
4
species can have a μ -oxido core and the ligand bound Cu O
4
4
unit is important both in industrial catalysis for substituted
5
phenol to quinone conversion and in the biological catalysis
6
cycle of nitrous oxide reduction.
2
In the biological realm the catechol oxidases (COs) are Cu -
based metalloenzymes of the oxido-reductase family. They
comprise a [Cu ] active site which catalyzes the two-electron
2
oxidation of a broad range of o-diphenols (catechols) to the
Chart 1 HL1 used in this work.
corresponding o-quinones, together with the reduction of O
2
7
from air to water. Thus a large number of copper(II)-based
model fragments and complexes have been studied as models
for CO in an attempt to establish the mechanism and discover
efficient mimics. Different dinuclear and mononuclear copper
complexes have been synthesized and examined for their
We have been interested in exploring the reactivity of (2,6-
bis(allylimino)methyl)-4-methylphenol (HL1; Chart 1) with
II
−
−
Cu ions in the presence of H O, ClO
and RCO2 .
2
4
Complexes of HL1 with transition metal ions have been
unknown thus far.
2
efficiency towards 3,5-DTBCH (3,5-di-tert-butylcatechol) oxi-
8
dation. The solution stability of Schiff base-bound Cu2L
species having several aqua coordination sites is dependent on
the pH of the reaction medium. The presence of these species
is important in synthesis for the isolation and crystallization Experimental section
of aggregates in the solid state. Solution pH and coordination
Reagents and materials
of ancillary bridges replacing the aqua H
2
O molecules direct
The chemicals used were obtained from the following sources:
trifluoroacetic acid from SRL; sodium acetate from SD Fine-
chem; sodium benzoate, copper hydroxide carbonate, triethyl-
amine from Merck, India and allyl amine from Alfa Aesar.
Copper(II) perchlorate hexa-hydrate was freshly prepared by
treating hydrated copper(II) carbonate (22.12 g, 0.1 mol) with a
the course of assembly reaction during the self-aggregation of
these Cu
more Cu
2
L fragments in solution. The association of two or
L units can provide various multinuclear architec-
2
9
10
11
tures, e.g., cubane, tetrahedron, stepped cubane, rhom-
1
2
13
14
boid, double cubane and cuboctahedron. The outcome of
reactions of a chosen Schiff base ligand system with different
copper(II) salts depends on the length and plasticity of side
arms with or without donor atoms, as well as on the particular
reaction conditions and the solvent used. However, the exact
role of each of these factors is still not well established to
1
: 1 aqueous solution of perchloric acid. Sodium trifluoroace-
tate was prepared by treating sodium hydroxide (4.00 g,
0
2
.1 mol) and trifluoroacetic acid (11.40 g, 0.1 mol) in water.
,6-Diformyl-4-methylphenol (2-hydroxy-5-methyl-benzene-1,3-
1
5
dicarbaldehyde) was prepared following a modified literature
predict and direct the preferential aggregation process. Our
present effort is dedicated to examine the reactivity pattern of
a new ligand system with copper(II) ions, in the environments
1
6
procedure providing a better yield. All the chemicals and sol-
vents used in this work were of reagent grade and used as
received without further purification.
−
−
2 4 2
of H O, ClO and RCO , for the modulation of the course of
the reaction. In different solvent media and in the presence of
Synthesis
the above-mentioned groups, the fragments CuL2 and Cu L1
2
(HL1
=
{2,6-bis(allylimino)methyl}-4-methylphenol, HL2
=
Ligand HL1 ((2,6-bis(allylimino)methyl)-4-methylphenol).
3
-{(allylimino)methyl}-2-hydroxy-5-methylbenzaldehyde) were 4-Methyl-2,6-diformylphenol was prepared following a modi-
1
6
stabilized. The presence of perchlorate anions triggers the fied literature procedure. To a MeOH solution of 4-methyl-
hydrolysis of one imine arm of the ligand (Scheme 1), 2,6-diformylphenol (0.820 g, 5 mmol), allyl amine (1.05 g,
whereas the presence of carboxylate anions has been shown 10 mmol) in MeOH (10 ml) was added with constant stirring
to attenuate the Schiff base hydrolysis and results in the and was refluxed for 2 h. Complete evaporation of the solvent
aggregation of initially formed fragments into μ
4
-oxido- in air over 12 h gave an orange oily substance, which was dried
10. The oily substance was character-
bridged [Cu ] and double µ -OH bridged [Cu ] complexes. We under vacuum over P O
4
3
5
4
were unable to isolate any type of dinuclear complex of intact ized by FT-IR and NMR spectroscopy and used directly for
−
L1 from the reaction medium due to the facile ‘spontaneous reactions with metal ion salts without further purification.
FTIR (cm− , KBr disk): 1636 (m). H NMR (400 MHz, CDCl
1
1
,
in situ self-aggregation’ reactions for [Cu
4
]
and [Cu
5
]
3
complexes.
δ ppm): 8.30–8.53 (1H, imine H), 7.26–7.65 (1H, Ar–H),
This journal is © The Royal Society of Chemistry 2018
Dalton Trans., 2018, 47, 17160–17176 | 17161

View Article Online
Paper
Dalton Transactions
5
.97–6.06 (1H, vinyl H), 5.13–5.23 (2H, vinyl H), 4.22–4.23 (2H, Yield: 0.664 g (54%). Micro analytical data are consistent with
1
3
allyl H), 2.27 (3H, methyl H). C NMR (100 MHz, CDCl3, the
δ ppm): 165.04 (imine C), 124.06–163.36 (Ar C), 116.62–127.54 CC
alkene C), 53.93 (secondary
0.32–20.35 (methyl C).
molecular
formula
of
[Cu (µ -O)(µ-L1) (µ1,3-
4
4
2
O
2
6
H
5
)
4
]·H O. Anal calcd for C58
2
H
56Cu
4
N
4
O
12 (1255.26
−
1
(
2
C attached to imine N), g mol ): C, 55.50; H, 4.50; N, 4.46. Found: C, 55.62; H, 4.54;
−
1
N, 4.44. Selected FT-IR bands: (KBr, cm , vs = very strong, br =
broad, s = strong, m = medium, w = weak): 3447 (br), 1624 (vs),
Complexes
A common synthetic practice was followed for the preparation L mol
of 1–4 using Cu(ClO ·6H O in the absence or presence of 259 (91 100). Molar conductance,
sodium salts of three different carboxylates. Orange MeCN or 27 Ω− cm mol
1
611 (vs), 1458 (m), 1370 (vs). UV-vis spectra [λmax, nm (ε,
−1
−1
cm )]: (MeOH solution) 635 (184), 366 (16 400),
(MeOH solution)
4
)
2
2
M
Λ :
1
2
−1
.
MeOH solutions of HL1 on reaction with Cu(ClO ) ·6H O in
[Cu (µ -OH) (µ-L1) (µ -OAc) (OAc) (H O) ][Cu (µ -OH) (µ-
4
2
2
5
3
2
2
1,3
2
2
2
4
5
3
2
−
−
MeOH in the absence or presence of CF
3
COO , C
COO resulted in all the four complexes.
Cu (µ-L2) (H O)] [Cu (µ-L2) (H O) ](ClO ) (1). A solution 2.5 mmol) was added to the as-prepared ‘solution of HL1’
6
H
5
COO
L1)
2
(µ1,3-OAc)
2
(OAc)
3
2 4 3 2 5
(H O)](ClO ) ·2C H OH (4). An EtOH
−
and CH
3
solution (10 mL) of copper(II) perchlorate hexahydrate (0.930 g,
[
2
2
2
2
2
2
2
2
4 6
of the as-prepared HL1 (approx. 0.24 g, 1 mmol) was dried (approx. 0.242 g, 1 mmol) in EtOH (20 mL) with stirring and
under vacuum and the gummy mass thus obtained was re-dis- the resulting green solution was stirred for 15 min. An EtOH
solved in MeCN (10 mL). To this copper(II) perchlorate hexa- solution (5 mL) of sodium acetate (0.348 g, 4.25 mmol) was
hydrate (0.37 g, 1 mmol) was added slowly and the resulting then added and the resulting reaction mixture was stirred for
green solution was magnetically stirred for ca. 15 min, fol- 2 h to give a dark green solution. The solution was filtered and
lowed by reflux for 2 h. The resulting deep-green solution was kept for slow evaporation in air. Green block-shaped crystals
filtered and kept undisturbed for slow evaporation in air. suitable for X-ray structure analysis were obtained after 1 week.
Block-shaped green crystals suitable for X-ray diffraction were Yield: 1.37 g (52%). Micro analytical data are consistent with
obtained after three weeks. Yield: 1.22 g (54%). Micro analyti- the above-mentioned formula for 4. Anal calcd for
1
cal data are consistent with the molecular formula of
Cu (µ-L2) (H O)] [Cu (µ-L2) (H O) ](ClO
C H Cu N O Cl (2263.36 g mol ): C, 38.21; H, 3.56; N, (KBr, cm , vs = very strong, br = broad, s = strong, m =
C
82
H
121Cl
3
Cu10
N
8
O
45 (2680.61 g mol− ): C, 36.74; H, 4.55; N,
[
2
2
2
2
2
2
2
2
4 6
) . Anal. calcd for 4.18. Found: C, 36.69; H, 4.53; N, 4.17. Selected FTIR bands:
−
1
−1
7
2
80
6
6
40
6
3
(
.71. Found: C, 38.14; H, 3.53; N, 3.73. Selected FTIR bands: medium, w = weak): 3447 (br), 1630 (m), 1564 (vs), 1412 (m),
KBr, cm− , vs = very strong, br = broad, s = strong, m = 1338 (m), 1092 (vs). UV-vis spectra [λmax, nm (ε, L mol
1
−1
−
1
medium, w = weak): 3448 (br), 1629 (vs), 1554 (s), 1090 (vs), cm )]: (MeOH solution) 670 (157), 367 (48 300), 258 (238 800).
−1
−1
−1
2
−1
6
M
22 (m). UV-vis spectra [λmax, nm (ε, L mol cm )]: (MeOH Molar conductance, Λ : (MeOH solution) 250 Ω cm mol .
solution) 708 (76), 366 (19 300), 243 (91 600). Molar conduc-
Caution! Metal ion complexes containing organic ligands
along with perchlorate counter anions are potentially explosive in
−1
2
−1
tance, Λ : (MeOH solution) 230 Ω cm mol
.
M
4 4 2 2 3 4
[Cu (µ -O)(µ-L1) (µ1,3-O CCF ) ] (2). To a stirred solution of nature. Therefore it is necessary to prepare only a small amount of
HL1 (approx. 0.242 g, 1 mmol) in MeOH (10 mL) solid copper(II) the samples at a time for safety reasons and treat them with
perchlorate hexahydrate (0.74 g, 2 mmol) was added under utmost care at all times.
stirring conditions which gave a light green solution. After
Physical measurements
1
0 min of stirring, a MeOH solution (10 mL) of sodium tri-
fluoroacetate (0.274 g, 2 mmol) was added, resulting in the Elemental analyses (C, H and N) of the compounds were per-
darkening of color and the stirring was continued for 2 h. The formed with a PerkinElmer model 240C elemental analyzer. A
solution was then filtered, and kept in air at room temperature Shimadzu UV 3100 UV–vis–NIR spectrophotometer and a
for slow evaporation. Green single crystals suitable for X-ray PerkinElmer RX1 spectrometer were used to record the solu-
analysis were obtained after 2 days. Yield: 0.710 g (59%). tion electronic absorption spectra and FTIR spectra respect-
Micro analytical data are consistent with the molecular ively. The purity of the powder compounds was determined by
formula of [Cu (µ -O)(µ-L1) (µ -O CCF ) ]. Anal. calcd for powder X-ray diffraction (PXRD) patterns using a Rigaku
4
4
2
1,3
2
3 4
−
1
C
38
H
34Cu
4
N
4
O
11
F
12 (1204.85 g mol ): C, 37.88; H, 2.84; N, MiniFlex II desktop X-ray diffractometer (30 kV, 15 mA) using
4
(
.65. Found: C, 37.97; H, 2.80; N, 4.67. Selected FTIR bands: Cu-K
α
radiation (λ = 1.5418 Å) within a 5–50° (2θ) angular
KBr, cm− , vs = very strong, br = broad, s = strong, m = range and a fixed-time counting of 4 s at 25 °C. A Bruker
medium): 1688 (vs), 1636 (m), 1436 (m), 1195 (vs). UV-vis Esquire 3000 Plus mass spectrometer was employed to collect
1
−1
−1
spectra [λmax, nm (ε, L mol cm )]: (MeOH solution) 666 the electrospray ionization (ESI) high resolution mass spectra
104), 372 (8100), 261 (45 400). Molar conductance, Λ : (MeOH of the three compounds. A CHI 1120A electrochemical
(
M
−
1
2
−1
solution) 34 Ω cm mol
Cu (µ -O)(µ-L1) (µ1,3-O
.
analyzer was used for cyclic voltammetric measurements.
[
4
4
2
2
CC
6
H
5
)
4
]·H
2
O (3). Complex 3 was Electrochemical measurements were performed at 25 °C in
prepared following a similar procedure as described above for MeOH in a one compartment cell with platinum as the
using sodium benzoate (0.288 g, 2 mmol) as the carboxylate working electrode, Pt wire as the counter electrode and Ag/
2
source in place of sodium trifluoroacetate. Block-shaped green AgCl as the reference electrode. Electron paramagnetic reso-
crystals suitable for X-ray analysis were obtained after 1 week. nance (EPR) spectra were recorded at 9.13 GHz (X-band) in
17162 | Dalton Trans., 2018, 47, 17160–17176
This journal is © The Royal Society of Chemistry 2018

View Article Online
Dalton Transactions
Paper
2
3
continuous wave mode with a Bruker ELEXSYS 580 X-band (SORTAV).
EPR spectrometer equipped with a standard accessory for (SIR97)
The structure was solved by direct methods
2
4
and the calculations were performed using
1
9
25
room temperature operation (298 K). Direct current (dc) mag- SHELXL-2014/6 and PARST implemented in the WINGX
2
6
netic measurements were performed on polycrystalline suite of programs. The locations of the heaviest atoms (Cu)
samples of compounds 2–4 using a Quantum Design SQUID were determined easily, and the O, N, and C atoms were sub-
magnetometer equipped with a 5 T magnet. The dc measure- sequently determined from the difference Fourier maps. These
ments were carried out under an applied field of 0.1 T. Data atoms are refined anisotropically. The H atoms were incorpor-
were corrected for the diamagnetism of the compounds and ated at calculated positions and refined with fixed geometry
for the diamagnetic contributions of the sample holder and riding thermal parameters with respect to their carrier
through measurements.
atoms. Crystallographic diagrams were presented using
2
7
DIAMOND software. A summary of the crystal data and rele-
vant refinement parameters is summarized in Table 1.
Crystallographic data (including structure factors) have been
Crystal data collection and refinement
Appropriate single crystals of 3 and 4 were chosen for data col- deposited with the Cambridge Crystallographic Data Centre as
lection on a Bruker SMART APEX-II CCD X-ray diffractometer supplementary publications CCDC 1857609, 1857607, 1857608
furnished with graphite-monochromated Mo Kα (λ
.71073 Å) radiation by the ω scan (width of 0.3° per frame)
method at 293 K with a scan rate of 5 s per frame. SAINT and
=
and 1857610.†
0
1
7
Kinetic experiments for catalytic oxidation of 3,5-DTBCH
2
XPREP softwares were used for data processing and space
group determination.
A
2
direct methods technique from 3,5-Di-tert-butylcatechol (3,5-DTBCH ) is widely used as a sub-
1
8
SHELXS-2014 was used for structure determination and then strate to study its catalytic oxidation and the oxidized product
refined by the full-matrix least squares technique using the 3,5-di-tert-butylquinone (3,5-DTBQ) can be monitored by Uv–
1
9
SHELXL (2014/7) program package within WINGX version vis spectroscopy (Shimadzu UV 3100 UV–Vis–NIR spectro-
2
0
1
.80.05. Data were corrected for Lorentz and polarization photometer) for appearance of the characteristic band at
effects; an empirical absorption correction was applied using 390 nm and reproducibility of the reactions was checked three
2
1
the SADABS. For complexes 1 and 2, the data were collected times. With the progress of the oxidation reaction i.e., oxi-
at 295 K using a Nonius Kappa CCD diffractometer. The dation of 3,5-DTBCH to 3,5-DTBQ, a characteristic and
2
2
2
dataset was integrated using the Denzo SMN package and intense absorption band was observed in the 300–400 nm
corrected for Lorentz, polarization and absorption effects range. MeOH solutions of the complexes (∼1 × 10− M) were
5
Table 1 Crystal data and structure refinement details for 1–4
Parameters
1
2
3
4
Formula
F.W. (g mol
Crystal system
Space group
Crystal color
Crystal size/mm
C
72
H
80Cu
6
N
6
O
40Cl
6
C
38
H
34Cu
4
N
4
O
11
F
12
C
58
H
56Cu
4
N
4
O
12
C
82 3 8 45
H121Cl Cu10N O
−
1
)
2263.36
Triclinic
Pˉ1
1204.85
Tetragonal
I4 /a
Green
0.43 × 0.34 × 0.24
12.5200(5)
12.5200(5)
35.8306(8)
90
90
90
5616.5(5)
4
1.425
1255.26
Monoclinic
P2 /c
Green
0.38 × 0.28 × 0.17
16.529(3)
22.599(4)
14.645(3)
90.00
99.695(5)
90.00
5392.3(18)
4
2680.61
Triclinic
Pˉ1
1
1
Green
Green
3
0.38 × 0.26 × 0.17
12.3517(4)
14.1130(5)
14.4440(5)
77.244(2)
75.5060(19)
71.192(2)
2280.45(14)
1
1.648
1.642
1150
295
0.30 × 0.21 × 0.12
14.3300(17)
15.8481(19)
24.442(3)
79.731(4)
80.665(4)
83.118(4)
5365.8(11)
2
1.659
2.105
2740
295
a/Å
b/Å
c/Å
α/°
β/°
γ/°
3
V/Å
Z
D
−
3
c
/g cm
1.543
1.624
2568
293
−
1
μ (mm
F(000)
T/K
)
1.583
2408
295
Total reflns
R(int)
Unique reflns
Observed reflns
Parameters
29 155
0.0394
8399
5928
16 627
0.0538
3058
2200
184
57 121
0.1299
8446
5764
63 654
0.0658
19 615
12 878
574
734
1335
R
1
; wR
2
2(I > 2σ(I))
0.0760, 0.2567
1.065
1.546, −0.719
1857609
0.0431, 0.1347
0.991
0.425, −0.251
1857607
0.0977, 0.2210
1.162
1.243, −0.796
1857608
0.0650, 0.1848
1.024
1.452, −0.756
1857610
GOF (F )
Largest diff peak and hole (e Å−
3
)
CCDC no.
2
2 1/2
2
2
o
R
1
o c o
= ∑(||F | − |F ||)/∑|F |. wR
2
= [∑w(|F
o
| − |F
c
|) /∑w(F
o
) ] . w = 0.75/(σ (F
o
) + 0.0010F
).
This journal is © The Royal Society of Chemistry 2018
Dalton Trans., 2018, 47, 17160–17176 | 17163

View Article Online
Paper
Dalton Transactions
separately reacted with incremental amounts of the substrate the reaction mixture to form I −
ions through I
(aq) + I → I
−
−.
3
2
3
3
,5-DTBCH2 (10 to 100 equivalents) at room temperature The H O liberated from the reaction mixture was detected by
2 2
−1
−1
under aerobic conditions. Time scans of ten minutes interval monitoring the band at 353 nm (ε = 26 000 L mol cm ),
for all the complex–substrate combinations were performed characteristic of I
−
3 2
. A blank experiment (without 3,5-DTBCH
and the data were treated following the initial rate method. or without catalyst) was also carried out to check the oxidizing
−
The slope of the absorbance versus time plot was used to find propensity of atmospheric O
the rate of the reaction. Following the Michaelis–Menten
2
toward I .
equation, a kinetic analysis was performed, and the resulting
kinetic parameters were extracted from the double-reciprocal
Results and discussion
Synthesis
Lineweaver–Burk plots for 1, 3 and 4. A modified rate equation
eqn (7)) was used to fit the data for 2 and extract the kinetic
(
Coordination of HL1 to copper(II) ions and the aggregation
flexibility of the initially formed species were explored in the
absence and presence of different carboxylate ions. Ligand
HL1 was synthesized in MeOH by Schiff base condensation of
parameters.
2 2
Detection of H O in catalytic reaction medium
Generation of hydrogen peroxide in solution during the cata- 2,6-diformyl-4-methyl phenol and allylamine in a 1 : 2 molar
lytic oxidation of DTBCH was spectrophotometrically detected ratio (Scheme S1†). The treatment of HL1 with Cu(ClO ) ·6H O
2
4 2
2
2
8
following a modified iodometric method. The reaction mix- in MeCN, under aerobic conditions, led to the formation of 1.
tures were obtained in a similar way to those used in the On the other hand, reactions of HL1 with Cu(ClO ·6H O in
kinetic experiments. The reaction mixture was acidified with the presence of different RCO Na salts (R = –CF , –C H , –CH )
4
)
2
2
2
3
6
5
3
2 4
H SO
(1 × 10− M) to pH 2 after 1 h to completely quench the in MeOH and EtOH, result in the formation of 2, 3 and 4,
3
reaction. The quinone formed in the reaction medium was respectively (Scheme 2). Several attempts have been made in
extracted three times with CH Cl2 and then 1 mL of 10% other solvents and varying reagent ratios to establish these
2
4 2 2
aqueous KI solution was added to the aqueous layer thus standardized procedures. The reaction of Cu(ClO ) ·6H O with
obtained. Ammonium molybdate (3%) solution was added in a HL1 in MeCN in a 1 : 1 molar ratio in the absence of any base
−
catalytic amount (3 drops) to accelerate triiodide (I ) for- under stirring followed by refluxing conditions, afforded a
3
2
8
mation in the presence of H
2
O
2
.
The reaction that occurs in green solution, from which 1 was isolated as a green block
−
+
the medium is H
2
O
2
+ 2I + 2H → 2H O + I , and I liberated shaped crystal in 54% yield. The use of Cu(ClO ·6H O
2
2
2
4
)
2
2
from the reaction combines with excess iodide ions present in without an added base made the reaction medium slightly
Scheme 2 Synthetic routes for 1–4.
17164 | Dalton Trans., 2018, 47, 17160–17176
This journal is © The Royal Society of Chemistry 2018

View Article Online
Dalton Transactions
Paper
acidic (pH 5–6), resulting in partial hydrolysis of one ligand water molecules in 1, a crystal lattice water molecule in 3, and
−
−
II
II
−
imine arm and conversion of L1 to L2 as Cu -bound forms. Cu -bridged OH groups, coordinated and lattice water mole-
II
During this hydrolysis process one Cu ion and one free allyla- cules in 4 is manifested by single broad and medium intensity
mine are also eliminated at the beginning. Thus the formation IR bands at 3478, 3447 and 3447 cm respectively. The charac-
of 1 can be summarized in eqn (1).
−
1
II
teristic Cu -bound ¯ν
stretching frequencies from the
CvN
−
1
ligand backbone appeared at 1636, 1624, and 1636 cm for 2,
CH3CN
3, and 4 respectively and at 1629 cm− for 1. The lattice ClO
1
−
6
HL1 þ 6CuðClO4Þ ꢀ 6H2O ꢁꢁꢁ! ½Cu2ðμ‐L2Þ ðOH2Þꢂ
Cu2ðμ‐L2Þ ðH2OÞ ꢂðClO4Þ þ 6½NH3CH2CHCH2ꢂClO4 þ 26H2O ions (in T site symmetry) in 1 showed a strong band at
2
2
2
4
½
2
2
6
d
1
−
1
6
104 cm
for ν
3
(T
(T
2
)(νClO) and a medium intense one at
2
)(δdOClO) modes. For 2 and 3 the pres-
ð1Þ
22 cm− for the ν
1
29
4
On the other hand, the addition of NaO
of Cu(ClO ·6H O and HL1 in MeOH, in a 4 : 2 : 1 molar ratio
gave a green solution under stirring, from which 2 can be iso-
lated as green block-shaped crystals in 59% yield. The
anions now serve dual roles by modifying the pH of
the reaction medium to prevent ligand arm hydrolysis and
function as a linker between two adjacent Cu centers. In
MeOH, the use of NaO CC H in place of NaO CCF in the
2 6 5 2 3
2
CCF
3
to a mixture
ence of bridging trifluoroacetates and benzoates was con-
firmed at 1688, 1436 and 1611, 1458 cm− , respectively for the
asymmetric ( ¯ν as(COO)
vibrations. The differences (Δ ¯ν = 252 cm and 153 cm ) in
these two values were characteristic of μ1,3-bridging modes of
four CF
In the case of CF COO bridging, the asymmetric stretching
vibration occurred at higher frequency due to the attachment
of electron-withdrawing fluorine atoms. For 4, the asym-
4
)
2
2
1
) and symmetric ( ¯ν s(COO)) stretching
−
1
−1
−
CF
3
CO
2
−
−
II
3
COO bridges in 2 and four C
6
H
5
COO bridges in 3.
−
3
same molar ratio resulted in 3 as green block-shaped crystals
in 54% yield. The generation of 2 and 3 from these two reac-
tions is summarized in eqn (2) and (3).
3
0
metric and symmetric stretching vibrations are detected at
1
−1
1
564 cm− ( ¯ν as(COO)) and 1412, 1338 cm ( ¯ν s(COO)) for μ1,3-brid-
−
1
CH3OH
ging type and monodentate coordination. Δ ¯ν = 152 cm thus
confirmed the former type and Δ ¯ν = 226 cm was for the
latter type. The presence of anionic ClO
symmetry) in 4 is indicated by a broad Cl–O stretching mode
at 1092 cm− characteristic of ν (T )(ν ) stretching and a
2
HL1 þ 4CuðClO4Þ ꢀ 6H2O þ 8CF3COONa ꢁꢁꢁ! ½Cu4ðμ ‐OÞ
ðμ‐L1Þ ðμ ‐O2CCF3Þ ꢂ þ 4CF3COOH þ 8NaClO4 þ 23H2O
2
4
−1
2
1;3
4
31
−
4 d
groups (in T site
ð2Þ
1
CH3OH
3
2
ClO
2
HL1 þ 4CuðClO4Þ ꢀ 6H2O þ 8C6H5COONa ꢁꢁꢁ!
Cu4ðμ ‐OÞðμ‐L1Þ ðμ ‐O2CC6H5Þ ꢂ ꢀ H2O þ 4C6H5COOHþ
−1
2
medium intensity vibration for Cl–O bending at 623 cm due
to a ν (T )(δdOClO) mode.
½
8
8
4
2
1;3
4
4
2
NaðClO4Þ þ 22H2O
2
Powder X-ray diffraction patterns
ð3Þ
The powder XRD patterns for the bulk materials of 1–4 were
recorded using a Bruker AXS X-ray diffractometer and com-
pared with the simulated ones derived from the single crystal
X-ray diffraction data. Fig. S2† shows good agreement of the
experimentally obtained powder patterns with that of the
simulated ones. The difference in intensity in some 2θ values
is due to the different orientations of the crystallites in the
powder samples. From the sharpness of the peaks and the cor-
respondence with the simulated patterns we conclude that the
powder samples are crystalline and phase pure with the
absence of extra peaks for impurities.
The use of the third type of carboxylate salt as NaOAc in
MeOH (molar ratio 4 : 2 : 1 for NaOAc : Cu(ClO ) ·6H O : HL1)
reaction medium did not yield any μ -oxido-bridged [Cu ] com-
4 4
plexes as obtained in eqn (2) and (3). Trial reactions using
different stoichiometries and solvents such as 4.25 : 2.5 : 1 in
EtOH on the other hand gave a green solution under stirring con-
ditions at room temperature, from which green crystals of 4 were
isolated. The formation of 4 can be rationalized as in eqn (4).
4
2
2
C2H5OH
4
HL1 þ 10CuðClO4Þ ꢀ 6H2O þ 17CH3COONa þ 2C2H5OH ꢁꢁꢁ!
2
½
½
Cu5ðμ ‐OHÞ ðμ‐L1Þ ðμ ‐O2CCH3Þ ðO2CCH3Þ ðOH2Þ ꢂ
3
2
2
1;3
2
2
4
Electronic spectra
Cu5ðμ ‐OHÞ ðμ‐L1Þ ðμ ‐O2CCH3Þ ðO2CCH3Þ ðOH2Þꢂ
3
2
2
1;3
2
3
ðClO4Þ ꢀ 2C2H5OH þ 8CH3COOH þ 17NaClO4 þ 49H2O
Electronic transitions for 1–4 in MeOH solution in the
3
200–800 nm range exhibit representative absorption bands for
ð4Þ
each aggregate. The ligand-field absorption bands (λ), with
The use of NEt , NaOMe or NaOH bases resulted in only
−1
−1
3
maxima at 708 nm (ε = 76 L mol cm ), 666 nm (ε = 104
intractable green gummy masses, which were not suitable for
further characterization and crystallization.
−1
−1
−1
−1
L mol cm ), 635 nm (ε = 184 L mol cm ) and 670 nm
(
−1
−1
ε = 157 L mol cm ), were found for 1, 2, 3 and 4 respect-
ively. The higher energy bands due to the ligand-to-metal
charge transfer (LMCT) transitions within the metal ion bound
Initial solid state characterization of the isolated reaction pro- ligand fragment were observed at 366 nm (ε = 19 300 L mol
Characterization by FTIR spectra
−
1
−
1
−1
−1
ducts was carried out using IR spectroscopy to pinpoint the cm ), >372 nm (ε = 8100 L mol cm ), 366 nm (ε = 16 400 L
−
1
−1
−1
−1
presence of the parent ligand system and carboxylates. It estab- mol cm ) and 367 nm (ε = 48 300 L mol cm ) respect-
–
II
–
−1
lished the binding of L1 to Cu centers in 2, 3 and 4 and L2
ively. Quite intense absorptions at 243 nm (ε = 91 600 L mol
II
II
−1
−1
−1
to Cu centers in 1 (Fig. S1†). The presence of Cu -bound cm ), 261 nm (ε
=
45 400
L
mol
cm ), 259 nm
This journal is © The Royal Society of Chemistry 2018
Dalton Trans., 2018, 47, 17160–17176 | 17165

View Article Online
Paper
Dalton Transactions
ε = 91 100 L mol cm−1) and 258 nm (ε = 238 800 L mol−1 of a MeCN solution, was used to find the molecular structure.
−
1
(
−
1
cm ) were dominated by the metal ion bound ligand-based Complex 1 crystallized in the triclinic P ˉ1 space group.
π → π* (LLCT) transitions (Fig. S3†).
Important bond distances and angles are summarized in
−
Table S1.† Two of the anionic hydrolyzed ligands (L2 ), provid-
II
EPR spectra
ing central phenoxido groups, were utilized to trap two Cu
centers in a centro-symmetric manner resulting in a Cu
motif. In two cases the coordination geometries around Cu
were different, square planar (Cu1 and Cu1*) and square pyra-
midal (Cu2 and Cu2*). For the central species both the Cu
2 2
O
Powdered samples of 1–4 were also characterized by recording
II
their X-band EPR spectra at 298 K. They show representative
II
32
axial signals at g ∼ 2.0 for Cu centers (Fig. S4†). Weak and
forbidden transitions at the half field of the g ∼ 2.0 signal
II
(
Cu3 and Cu3*) centers have square-pyramidal coordination
(
S
ΔM = ±2 transitions) were observed at 1605 G (g = 4.29), 1613
geometry. Within the basal planes the Cu–O and Cu–N dis-
tances from aldehyde O and imine N are close, e.g., Cu1–O4,
G (g = 4.25), 1607 G (g = 4.28) and 1607 G (g = 4.28) for com-
plexes 1–4 respectively indicating the presence of spin–spin
3
2
1.940(5) Å; Cu2–O2, 1.943(4) Å and Cu1–N1, 1.937(5) Å; Cu2–
∥ ∥
interactions. For complexes 1 and 2 definite g and A values
N2, 1.940(5) Å. Interestingly the bridging phenoxido functions
also display similar bond lengths e.g., Cu1–O1, 1.936(4) Å;
Cu1–O3, 1.944(4) Å; Cu2–O1, 1.959(4) Å; Cu2–O3, 1.945(4) Å
could not be determined due to poor resolution of split pat-
terns of the parallel component. The g⊥ signals were found at
2
.13 and 2.11 with hyperfine splitting in this region and the
−
4
−1
−4
−1
(
Fig. 1). The apical water (O1 W and O2 W) molecules showed
longer Cu–O distances at 2.300(5) and 2.291(5) Å. Within the
Cu motif the Cu⋯Cu separations are 3.0163(9) and 3.0307
12) Å and the Cu–O–Cu angles are in the 101.51(17)–101.71
⊥
corresponding A = 83.19 × 10 cm and 81.63 × 10 cm
respectively. On the other hand complexes 3 and 4 gave g =
∥
2 2
O
2
.27 and 2.25, and g
⊥
= 2.08 and 2.10 respectively. In this case
−
4
−1
(
(
the average hyperfine constants were A
∥
= 159.85 × 10 cm ,
3
3
−
4
−1
−4
−1
−4
17)° range and at 101.63(18)° for two types of [Cu
2
] units.
1
50.50 × 10 cm and A⊥ = 75.40 × 10 cm , 81.01 × 10
−
1
[
Cu (µ -O)(µ-L1) (µ1,3-O CCF ] (2). With Z = 4 compound 2
4
4
2
2
3 4
)
cm respectively.
crystallizes in the tetragonal I4 /a space group. Selected metric
1
Description of crystal structures
parameters are summarized in Table S1.† The tetranuclear
2
structure of 2 is shown in Fig. 2a. Two capping N O donor set
[
Cu (µ-L2) (H O)] [Cu (µ-L2) (H O) ](ClO ) (1).
A
green
2
2
2
2
2
2
2
2
4 6
−
II
bearing L1 parts bind two Cu centers to give {Cu (OH)} frag-
2
block-shaped single crystal of 1, obtained by slow evaporation
ments. Two such fragments in the next step assemble around
4 4
a central µ -O group to provide the tetrahedral {Cu O} core.
Bridging by four trifluoroacetate and two phenoxido groups
complete the Cu⋯Cu links on six faces of the cube (Fig. 3a).
The phenoxido-bridged faces record shorter Cu⋯Cu separ-
ations at 2.9979(6) Å and the clipping by trifluoroacetate
groups register longer intermetallic separations at 3.2231(1) Å.
II
The Cu center remains in a square-pyramidal O
4
N coordi-
3
4
nation environment (Addison τ = 0.12), where the shortest
5
distance (Cu–O, 1.9290(3) Å) was provided by the O-atom of
the μ
4
-oxido group and the apical coordination from the tri-
fluoroacetato O-atom gives a Cu–O distance of 2.342(3) Å
Fig. 1 Structural view of 1 with a partial atom numbering scheme. H
atoms are omitted for clarity. Colour code: Cu, green; N, blue; O, red; C, (Fig. 2b and c). Three possible carboxylate coordination modes
black.
for bridging two adjacent metal ion centers are known. These
Fig. 2 (a) Structural view of 2 with a partial atom numbering scheme. H atoms are omitted for clarity; (b) atom connectivity within the [Cu
c) coordination environment around each Cu center. Colour code: Cu, green; N, blue; O, red; C, black; F, yellow.
4
] core;
(
17166 | Dalton Trans., 2018, 47, 17160–17176
This journal is © The Royal Society of Chemistry 2018

View Article Online
Dalton Transactions
Paper
Fig. 3 (a) Trifluoroacetato connectivity in 2; (b) distorted syn–syn coordination of each trifluoroacetate group for complementary basal–apical
coordination.
are syn–syn, syn–anti and anti–anti bridging modes. The dis-
torted syn–syn coordination from four bridging carboxylates in crystallizes in the monoclinic P2
the μ1,3-mode was responsible for placing one syn O-atom of the asymmetric unit has one [Cu
[Cu (µ -O)(µ-L1) (µ -O CC H ) ]·H O (3). Compound
3
4
4
2
1,3
2
6
5 4
2
1
/c space group with Z = 4 and
] unit and a lattice water
4
the carboxylate anion at the basal site of one copper center molecule. The structure is shown in Fig. 4a and important
and another distorted syn O-atom of the same carboxylate bond lengths and angles are listed in Table S1 in the ESI.†
anion at the apical site of another copper center for comp-
Unlike the previous case, the nature of bridging of four ben-
lementary basal–apical clipping (Fig. 3b). The basal Cu–O zoate groups was different. Now two of the cube faces were
bond (1.962(2) Å) is shorter compared to the apical one (2.342 spanned by two benzoates leading to intermediate Cu⋯Cu
(
3) Å). The amount of twisting of carboxylate anions for the separations of 3.193(2) and 3.082(2) Å. The other two faces
distorted syn–syn coordination can be estimated from the C– where no clipping groups were present the Cu⋯Cu distances
C–O–Cu/C–C–O–Cu torsion angles (179.59/141.85°). were the longest at 3.204 and 3.286(2) Å (Fig. 5a). Similar to
−
Careful checking of the crystal packing arrangements the previous case Cu⋯Cu separations in L1 bridged faces are
II
revealed no hydrogen bonding network or CH⋯π interactions the smallest (3.021(2) and 3.0105(18) Å). Herein all four Cu
to be present in 2.
5
centers showed varying Addison τ values (Cu1 = 0.03, Cu2 =
Fig. 4 (a) Structural view of 3 with a partial atom numbering scheme and H atoms are omitted for clarity; (b) atom connectivity within the [Cu
tetrahedron; (c) representative coordination geometry around each Cu centre. Colour code: Cu, green; N, blue; O, red; C, black.
4
]
II
Fig. 5 (a) Benzoato connectivity in 3; (b) complementary basal–apical clipping by benzoate groups.
This journal is © The Royal Society of Chemistry 2018
Dalton Trans., 2018, 47, 17160–17176 | 17167

View Article Online
Paper
Dalton Transactions
0
.13, Cu3 = 0.22 and Cu4 = 0.08). The distorted syn–syn coordi-
nation modes of two benzoates to Cu1–Cu3 and Cu2–Cu4
brought the syn O-atom to the basal site and another syn
II
O-atom to the apical site of Cu centers in a regular and comp-
lementary fashion (Fig. 5b). For all the metallic sites the basal
Cu–O bonds were shorter at 1.930(7) Å–1.964(7) Å compared to
the apical ones at 2.214(9) Å–2.397(7) Å (Fig. 4b and c).
[
Cu (µ -OH) (µ-L1) (µ -OAc) (OAc) (H O) ][Cu (µ -OH) (µ-
5
3
2
2
1,3
2
2
2
4
5
3
2
L1)
2
(µ1,3-OAc)
2
(OAc)
3
(H
2
O)](ClO
4
)
3
·2C
2
5
H OH (4). Compound 4
forms green crystals in the triclinic P ˉ1 space group and the
structural view is shown in Fig. 6. The unit cell of this com-
pound is very much similar to the copper-based shattuckite
Fig. 7 View of the cationic pentametallic core of the first part of 4.
5
mineral and consists of two [Cu ] units, three perchlorate
anions, and two lattice EtOH molecules. Important bond
lengths and angles are listed in Table S1.† The pentanuclear
Cu -based aggregate was formed by trapping of a central Cu
ing the acetato coordination network and depending upon the
preference for coordination geometries and availability of
II
II
center by two ligand bound {Cu (OH)} fragments which were
2
coordination sites, water molecules were coordinated to some
instrumental to give 2 and 3 by a direct assembling process. In
II
Cu centers. Two [Cu ] coordination units have two types of
5
these two cases, condensation of two bridging hydroxido
acetato coordination networks within 4 (Fig. 6). In the first
−
2−
groups results in a central µ -O group (2HO → H O + O ),
II
4
2
part (dicationic unit), including the central one, four Cu
necessary for the formation of the {Cu
4
O} core. In the case of 4
centers are in square pyramidal (τ values 0.11, 0.08, 0.03 and
5
two of these hydroxido bridges have the control on the gene-
4
0.05) geometry and Cu3 in square planar (τ = 0.17) geometry
ration of two [Cu ] entities as is known in the case of copper-
5
2
having unusual coordination of one H O (Fig. 7). Within the
based minerals shattuckite, Cu
Cu (OH) (AsO and with chlorido in place of the hydroxido
group in Cu Cl (SeO ) .
5
(OH)
2
(SiO
3
)
4
,
cornubite,
two L1− bridged parts the Cu⋯Cu separations are 3.0442(12)
5
4
4 2
)
and 3.0396(13) Å.
In the second part the number of acetato groups and their
connectivity pattern were different. Now three centers were in
3
5,36
Two slightly different [Cu ] coordi-
5
2
3 4
5
nation units, having two or one positive charges, were present
within one unit cell. In these two cases the bridging network
4
square planar (τ = 0.16, 0.17 and 0.16) geometry, while the
II
between the five Cu ions is ensured by two water derived
5
remaining two were in square pyramidal (τ = 0.01 and 0.10)
−
μ-HO ions, two μ1,3 acetato bridges, terminal acetates and two
geometry (Fig. 8). Within the two L1− bridged parts the
–
μ-phenoxido fragments from deprotonated L1 ligands.
Cu⋯Cu separation is 3.0401(13) Å for square pyramidal–
square planar centers and 3.0007(13) Å for two in square
planar geometry.
II
In both the cases the central Cu centers remained in a dis-
torted square-pyramidal (τ
5
= 0.05) coordination geometry
having five O-donor atoms from the coordination of two hydro-
xido O centers in basal sites and, two acetato groups in μ1,3
mode and one as monodentate. The O atoms of the bridging
acetato group were responsible for long Cu–O apical bonds as
5
Both the [Cu ] parts were more open and distorted com-
pared to other known arrangements based on vertex-shared
10
open-dicubanes.
expected from a Jahn–Teller distortion (Fig. 7). After establish- Magnetic measurements
The magnetic properties of complexes 2, 3 and 4 were investi-
gated by direct current (dc) magnetic susceptibility measure-
ments on powdered samples under an applied field of 0.1 tesla
and are plotted as χ T vs. T in Fig. 9. For 1, 2 and 3 the mag-
M
Fig. 6 POV-ray view of the two parts in 4 with an atom numbering
scheme. Hydrogen atoms are omitted for clarity. Color code: Cu, green;
N, blue; O, red; C, black.
Fig. 8 View of the cationic pentametallic core of the second part of 4.
17168 | Dalton Trans., 2018, 47, 17160–17176
This journal is © The Royal Society of Chemistry 2018

View Article Online
Dalton Transactions
Paper
4
more efficient interaction compared to the single μ -oxido
II
11
39
group bridging the two [Cu ] units. Using the program PHI
2
we were able to fit the magnetic susceptibility data for com-
plexes 2 and 3 by adopting a simple 1-J model (see Fig. 10 left
and eqn (5)), which avoids any over parameterization in the fit.
4
X
!
ˆ
ˆ
ˆ
ˆ
ˆ
~
H ¼ ꢁ2JðS1 ꢀ S2 þ S3 ꢀ S4Þ þ gμ B
si
ð5Þ
B
i¼1
The best fit afforded J = −204 (±1) cm− and J = −174 (±1)
cm− , for 2 and 3 respectively, with the g-value fixed at 2.15.
Small S = 1/2 spin impurities were also considered for 2 (3.5%)
and 3 (2.9%), consistent with those previously reported in
1
1
1
1,38,40,41
similar complexes.
The fit results in a singlet ground
state (S = 0) with the first excited state (S = 1) located at
400 cm− and ∼350 cm for 2 and 3, respectively. The strong
1
−1
∼
antiferromagnetic coupling in 2 and 3 is not uncommon when
II
Cu ions are bridged through phenoxido groups and the Cu–
Fig. 9
M
χ T vs. T plot for complexes 2–4 under an applied dc field of 0.1
Ophenoxido–Cu angle is larger than 97.6°, with the strength of
T. The solid lines represent a fit of the data (see the text for details).
the antiferromagnetic interaction increasing with the
4
2,43
angle.
angles are 100.5° and 102° for complex 2 and 98° and 104° for
, it is not surprising that the exchange interaction, J, through
Given that the Cu–Ophenoxido–Cu and Cu–Ooxido–Cu
netic susceptibility measurements were difficult. In particular
for 1 the magnetic susceptibility signal was extremely low at
room temperature, indicative of strong antiferromagnetic
3
the μ -oxido/phenoxido bridge is large and antiferromagnetic.
4
For complex 4, the data were fitted considering one domi-
II
interactions between the two Cu centers suggesting a very
II
nant exchange interaction between the phenoxido-bridged Cu
well-isolated singlet spin ground state (S = 0) and no further
measurements were undertaken. For 2 and 3 the signal
becomes extremely low below 110 K and 50 K, respectively.
ions (Fig. 10) adopting the approach previously reported for
similar complexes, which avoids any overparameterization in
3
7,44
39
the fit.
Using the program PHI
and adopting the
II
Complex 4, [Cu ], presents better behavior and a complete
5
Hamiltonian shown in eqn (6)
measurement, from 290 K to 5 K, was possible.
The room temperature χ
nificantly lower compared to the theoretical values, indicating
M
T values of all complexes are sig-
X
5
!
ˆ
ˆ
ˆ
ˆ
ˆ
~
H ¼ ꢁ2JðS1 ꢀ S2 þ S3 ꢀ S4Þ þ gμ B
si
ð6Þ
B
i¼1
that strong dominant antiferromagnetic interactions exist
II
between the Cu ions; the room temperature χ
M
T values of we were able to simultaneously fit the magnetic susceptibility
3
−1
3
−1
0
.81 cm K mol and 0.91 cm K mol for 2 and 3 respect- and magnetization data (Fig. 11). The best fit yielded J = −205
3
−1
−1
ively are below the expected value of 1.73 cm K mol for four (±2) cm with the g-value fixed at 2.15. The fit results in an
non-interacting S = 1/2 ions with g = 2.15. Upon cooling, χ
S = 1/2 ground state with the first excited state located at
M
−1
T
3
−1
decreases reaching a minimum value of 0.17 cm K mol at ∼400 cm . The magnitude and the nature of the exchange
10 K for 2 and ∼0.1 cm K mol at 50 K for 3. Below 110 K interaction is in agreement with previously reported Cu com-
3
−1
II
1
and 50 K, for 2 and 3 respectively, further measurements were plexes with similar geometric parameters; the Cu–Ophenoxido–
extremely difficult due to the very small signal and subsequent Cu angles are ∼101° (above the critical angle of 97.6°).42,43
poor signal-to-noise ratio, suggesting a well-isolated singlet
spin ground state (S = 0) for both the complexes (vide infra).
3
−1
For complex 4 the experimental χ T value of 1.10 cm K mol
M
3
at 290 K is lower than the theoretical value of 2.17 cm
mol for five non-interacting S = 1/2 ions with g = 2.15. Upon
cooling, χ T decreases reaching a minimum value of 0.38 cm
K
−
1
3
M
K mol− at 2 K.
1
II
It has been reported previously that in Cu complexes the
magnetic interactions mediated through bridges coordinated
on the basal plane are stronger than those in the apical posi-
3
7
tions. With this in mind, complexes 2 and 3 can be con-
II
sidered as two phenoxido-oxido-bridged [Cu ] units interact-
2
3
8
ing weakly through a central μ
4
-oxido bridge. In each Cu⋯Cu
Fig. 10 Exchange interaction (J) scheme for 2 and 3 (left) and 4 (right).
Color code: cyan dashed line = J, O = red, Cu = green, N = dark blue, C
dinuclear unit the metals are bridged through a double μ
4
-
oxido/phenoxido bridge, which has been described as a much = grey.
This journal is © The Royal Society of Chemistry 2018
Dalton Trans., 2018, 47, 17160–17176 | 17169

View Article Online
Paper
Dalton Transactions
lected in Table 2. In the case of complex 2, inhibition by sub-
strate molecules was observed and the reaction rate did not
reach a plateau at high substrate concentration as observed in
the other three cases (Fig. 12). This behavior may be assumed
due to the formation of inactive mononuclear species in the
presence of excess 3,5-DTBCH . In the other three cases, the
2
catalytic potency may be explained due to the difference in the
availability and nature of catalytically active ligand bound
{
Cu
2
} fragments. For 2, the decomposition of hydroxido-
} species in solution limits the possibility of two
bridged {Cu
2
electron oxidation of 3,5-DTBCH to 3,5-DTBQ. In the case of 2
2
a simple Michaelis–Menten model with the establishment of a
pre-equilibrium cannot be used to describe the kinetic
process. A modified rate equation was used to fit the rate data
c
as per the mechanism represented in Scheme S2,† where K is
4
6
Fig. 11 M vs. H plot for complex 4 in the 1–5 T field range. The solid the formation constant of the inactive mononuclear species.
lines represent the fit of the data (see the text for details).
Kinetic treatment considering initial rates and a large excess
of 3,5-DTBC resulted in the following equation.
Catechol oxidation activity
kcat½DTBCꢂ½Catꢂ
r ¼
ꢀ
ꢁ
ð7Þ
2
KM 1 þ KC½DTBCꢂ þ ½DTBCꢂ
In this part of the work, we examined all four complexes for
the catechol oxidase model study. Our intention was to reveal
the efficacy of the reported compounds of varying nuclearities
for any catalytic oxidation of the most widely used model sub-
7
The parameters obtained by applying eqn (7) to the data for
are listed in Table 2 along with 1, 3 and 4.
2
2
strate 3,5-di-tert-butylcatechol (3,5-DTBCH ) at 298 K in MeOH
High resolution mass spectroscopic evidence for fragments in
solution
solutions saturated with O . The generation of the oxidized
2
quinone product (3,5-DTBQ) was followed by monitoring the
increase in absorbance in the UV-vis region of the character- To obtain a reasonable idea about the nature of the various
istic band of quinone as a function of time. A lower redox fragments present in solution, possible catalyst–substrate
potential for this substrate and the presence of bulky substitu- intermediates during oxidation reactions and insight into the
ents for a slower ring opening reaction during any kind of mechanism of catalytic cycles, MeOH solutions of complexes
4
5
follow-up oxidation prompted us to choose this substrate. To 1–4 were examined by HRMS analysis. All four complexes
measure the catechol oxidation potential of complexes 1–4, ∼1 exhibited a common peak at an m/z value of 243.2511 corres-
10 mol L solutions were treated with a ∼1 × 10−3 mol L
−
5
−1
−1
ponding to the protonated ligand [HL1 − H] (C H N O;
+
×
1
5
19 2
solution of 3,5-DTBCH in air for the collection of time-depen- calcd 243.1497) (Fig. S6†). In the case of 1, the base peak
2
dent UV-vis spectra up to 50 min (Fig. S5†). The treatment of obtained at an m/z value of 204.1013 is due to the protonated
+
Cu(ClO
4
)
2
·6H
2
O with 3,5-DTBCH
2
did not exhibit any change and hydrolyzed ligand fragment [HL2 − H] (C H NO ; calcd
1
2
14
2
in the absorption intensity, even after 6 h of addition under 204.1025). The same peak, of lesser intensity, was also present
similar reaction conditions.
in the solutions of 2–4.
Initial rate methods were applied to examine the kinetic be-
A medium intensity peak at an m/z value of 445.1441 for 1
+
havior of oxidation of 3,5-DTBCH by complexes 1–4, by moni- was considered for [(HL2)2
−
K] (C H KN O ; calcd
2
24 26
2 4
toring the characteristic absorption band for substituted 445.1530) and other peaks at 266.0243, 468.1140, 490.0925
quinone at 390 nm in MeOH as a function of time. A first- and 530.0331 (Fig. S6†) were assigned to the fragments [Cu(L2)
+
+
+
+
order dependence of the rates on substrate concentrations was
−
H] (C H CuNO ; calcd 266.042), [Cu(L2)2
−
H]
Na]
1
2
13
2
observed on treating the complexes 1–4 with varying concen- (C H CuN O ;
calcd
468.1110),
490.0930)
[Cu(L2)2
−
2
4
25
2 4
trations of 3,5-DTBCH
2
. However complexes 1, 3 and 4 show (C H CuN NaO ;
calcd
and [Cu (L2) ]
2
4
24
2
4
2 2
saturation kinetics at higher 3,5-DTBCH concentrations. Thus (C H Cu N O ; calcd 530.0323). For a mixture of 1 and 3,5-
2
24 24
2 2 4
a Michaelis–Menten model is used for fitting the rates DTBCH2 in a 1 : 100 molar ratio in MeOH (Fig. S7†) after
obtained by the initial rate method) for 1, 3 and 4 (Fig. 12) 10 min of mixing the peaks at m/z = 243.1381 and 463.2868 for
(
+
+
and the kinetic parameters were extracted from the [3,5-DTBQ − Na] and [(3,5-DTBQ) − Na] , respectively, clearly
2
Lineweaver–Burk plots (Fig. 12 insets).
The binding constant (K ), the maximum velocity (Vmax), intensity peak at 529.2902 is obtained for the catalyst–semiqui-
and the rate constants for the dissociation of the complex–sub- nonate adduct
strate intermediate (i.e., the turnover number, kcat) are (C H Cu N NaO ; calcd 529.1872) establishing the
showed the formation of free quinone in solution. The low
M
2
+
[Cu (L2) (DTSQ) (CH OH)
2
−
HNa]
2
2
2
3
5
4
73
2
2
10
+
obtained from the Lineweaver–Burk plot of 1/V vs. 1/[S]. The [Cu (L2) ] fragment as the catalytically active species in
2
2
kinetic parameters Vmax, K , and K for 1, 3 and 4 are col- solution.
M
cat
17170 | Dalton Trans., 2018, 47, 17160–17176
This journal is © The Royal Society of Chemistry 2018

View Article Online
Dalton Transactions
Paper
Fig. 12 Dependence of the reaction rates on the substrate concentration for the oxidation of 3,5-DTBCH
The Lineweaver–Burk plots (inset).
2
catalyzed by complexes 1–4 in MeOH.
Table 2 Kinetic parameters for the oxidation of 3,5-DTBCH
MeOH
2
by 1–4 in
A solution of 3 in MeOH gave a medium intensity peak at
+
m/z = 413.2657 for [Cu
2 6 5 3 3
(L1)(OH)(C H COO) (CH OH) − 2Na]
(
C H Cu N Na O ; calcd 413.5429) (Fig. S11†), whereas a
3
7
37
2
2
2 9
V
max
K
M
k
cat
k
cat/K
M
K
C
7
mixture of 3 and 3,5-DTBCH
resulted in peaks at m/z = 243.1381 and 463.2868 for [3,5-
DTBQ − Na] and [(3,5-DTBQ) − Na] , respectively (Fig. S12†),
confirming the formation of a free quinone unit. A character-
istic catalyst–partially oxidized substrate aggregate [Cu
(
2
in a 1 : 100 molar ratio in MeOH
Complex (10 M S−1
−
6
)
(10 M) (h−1)
−4
(10 M h−1
3
−1
)
(10 M−2)
+
+
1
2
3
4
0.1001
—
0.3394
1.2603
0.5569
1.2879
1.4883
2.1377
36.045 632
274.896 2134.4
122.174 820.8
453.732 2122.4
—
1.3911
—
2
—
2
(L1)
2
+
OH)(C H COO) (DTSQ) (H O) − 2H] (C H Cu N Na O ;
6
5
2
2
2
37 37
2
2
2 9
calcd 544.1918) was found as a low intensity peak at m/z =
544.3038.
An MeOH solution of 4 gave a peak at m/z = 545.0192 for
The mass spectrum of 2 shows a peak at m/z = 409.0576 for
the dinuclear fragment [Cu (L1)(OH)(CF COO) (CH OH) −
the diagnostic {Cu
2
} fragment [Cu
2 3 2 2
(L1)(OH)(CH COO) (H O) −
2
3
3
3
+
2
+
(
2H)Na] (C20
H
30Cu
2
N
2
NaO ; calcd 545.0386) (Fig. S13†).
7
HNaK] (C H Cu F KN NaO ; calcd 408.9688) (Fig. S8†). A
2
2
23
2
9
2
9
When mixed with 3,5-DTBCH2 in MeOH in a 1 : 100 molar
ratio the species with a coordinated semiquinonate form of
mixture of 2 and 3,5-DTBCH
2
in a 1 : 20 molar ratio in MeOH
after 10 min of mixing gave a peak at 587.1411 for the catalyst–
semiquinonate adduct [Cu (L1)(OH)(CF COO) (DTSQ)
the substrate as [Cu
2
(L1)(OH)(CH
3 2 2 3 2
COO) (DTSQ) (CH OH) −
−
2
3
2
2
2
+
2
+
2
K] (C H Cu K N O ; calcd 543.1461) was obtained as a
(
(
8
2Na)(2K)]
Fig. S9†). But for a 1 : 100 molar ratio a peak at a m/z value of
19.4127 could be assigned to the mononuclear species
(C47
H
58Cu
2
F
6
K
2
N
2
Na
2
O
10
;
calcd
587.0923)
49 72
2 2 2 12
low intensity peak at m/z = 543.3044 (Fig. S14†).
+
Cyclic voltammetric traces in solution
2 3 2 2 7
[Cu(L1)(3,5-DTBCH) (CH OH)(H O) − Na] (C44H65CuN NaO ;
calcd 819.3985) (Fig. S10†) indicating the collapse of the The redox sensitivities of the complexes required for the cata-
dinuclear fragment at high 3,5-DTBCH concentrations.
lytic oxidation of 3,5-DTBCH were examined by cyclic voltam-
2
2
This journal is © The Royal Society of Chemistry 2018
Dalton Trans., 2018, 47, 17160–17176 | 17171

View Article Online
Paper
Dalton Transactions
metry measurements. This gave an idea about the involvement Probable mechanism for catechol oxidation
of redox active species during catalysis. The same MeOH
medium was chosen for measurement under a N atmosphere
2
with a TBAP supporting electrolyte, platinum working elec-
trode, platinum wire counter electrode and Ag/AgCl reference
electrode (Fig. S15†). Bare HL1 showed a reduction peak at
In order to establish a reasonable course of the metal ion
complex mediated oxidation reaction, we need to know
whether the O2 in air reduces to H O or H O during the
2
2 2
2 2
process. The estimation of H O formed in the medium was
performed by a modified iodometric titration procedure accel-
−0.898 V which was shifted upon complex formation in the
−
erated by molybdate ions. The oxidation of I to I followed by
2
−
0.698 to −0.888 V range vs. Ag/AgCl for 1–4 and was most
−
the generation of I
3
was apparent from the UV-vis spectral
II
likely for the reduction of the bare and Cu bound CvN bond.
An oxidative response at 0.733 V vs. Ag/AgCl appeared for free
HL1 and in the 0.708 to 0.860 V range for the metal bound
ligand in 1–4. The nature of the voltammograms was in favor
of only ligand-centered redox processes without showing any
change in the oxidation state of the involved metal ion centers
study of the solution obtained after work-up indicating the
8
generation of H O as identified by others. A significant
2
2
2 2
amount of H O is found to be accumulated in the reaction
medium in all the studied cases (Fig. S17†).
From HRMS, CV and EPR analysis a mechanistic pathway
can be tentatively proposed for the catalytic oxidation of 3,5-
DTBCH2 to 3,5-DTBQ by 1–4. In MeOH solution the active
species like {Cu (L2) }, {Cu (L1)(OH)(CF COO) (CH OH)},
II
I
from Cu to Cu during the oxidation of 3,5-DTBCH
2
to 3,5-
DTBQ.
2
2
2
3
3
3
{
Cu
2
(L1)(OH)(C
6
H
5
COO)
3
(CH
3
OH)}
and
{Cu
2
(L1)(OH)
EPR sensitive species in solution
(
{
{
CH COO) (H O)} were present to bind 3,5-DTBCH giving
Cu (L2) (DTSQ) (CH OH) }, {Cu (L1)(OH)(CF COO) (DTSQ) },
Cu
3 2 2 2
X-band EPR spectra of MeOH solutions of 1–4 at 298 K showed
very weak isotropic signals at g ∼ 2.0 and a second one at g ∼
2
2
2
3
2
2
3
2
2
2
(L1)(OH)(C
6
H
5
COO)
2
(DTSQ)
2
(H
2
O)} and {Cu
2
(L1)(OH)
4
2
.0 indicating the presence of ligand bound {Cu } entities in
(
CH COO) (DTSQ) (CH OH) } respectively for 1, 2, 3 and 4. In
3 2 2 3 2
solution (Fig. S16† left and Fig. 13 left). In particular, the
signals at g ∼ 4.0 were attributable to {Cu } species (cf. the low-
field ΔM = ±2 features) as indicated in HRMS analysis. The
S
isotropic g parameters (giso) could not be ascertained unam-
biguously due to the weak and noisy signals around g ∼ 2.0.
The g values associated with the ΔM
found to be 4.25, 4.26, 4.25 and 4.28 respectively for 1–4. The
EPR spectra were recorded after 10 min of mixing of 1–4 and
all these cases the generation of semiquinone radical species
and metal ion bound imine-centered radical fragments was
detected from EPR signals. No reduction peak corresponding
to the conversion of Cu to Cu is observed for 1–4 in MeOH
solution and each ligand imine bond accepts an electron to
remain in the reduced state. The lone electron is later
accepted by a dioxygen molecule completing the catalytic cycle
2 2
releasing 3,5-DTBQ and H O (Scheme 3). Thus the used pro-
radical system is expected to exploit ligand redox activity cen-
tered on the imine groups for catalytic activity. Complex 2 in
2
the presence of a high concentration of DTBCH breaks down
to catechol bound mononuclear Cu species {Cu(L1)(3,5-
DTBCH) (CH OH)(H O)} as indicated by HRMS, which was cat-
alytically inactive inhibiting the catalytic oxidation.
2
II
I
S
= ±2 transitions were
47
3
2
,5-DTBCH in a 1 : 100 molar ratio in MeOH at 298 K
(Fig. S16† right and 13 right). The spectra for 1, 3 and 4
showed sharp signals at g = 2.01, 2.00 and 2.00 respectively
confirming the generation of organic free radicals in solu-
II
4
7
tion. Repeat scans in the 3420–3475 G range, keeping the
number of data points the same as before, showed splitting of
these signals into two with additional lower intensity
shoulders to the right in all the three cases. This observation
of doubly split signals justifies the presence of an unpaired
2
3
2
Comparison of catalytic activity
4
8
electron on C4 of DTSQ bearing a single hydrogen atom. The With respect to the determined catalytic turn over number,
signals appearing as shoulders arise from a CvN centered cat, the reactivity order of the complexes for the oxidation of
k
radical in L1 and L2. An unpaired electron on the imine 3,5-DTBCH follows the trend 4 > 2 > 3 > 1. From HRMS ana-
2
carbon would have given a six line spectrum. Clear resolution lysis it was found that in solution the parent ligand bound
was obscured in the presence of a strong DTSQ radical signal. {Cu
An interesting observation was made in the EPR spectrum of 2 through crystallization. The highest kcat value in the series for
with 100 equivalents of 3,5-DTBCH . It was different from that 4 is evident from the fact that it provides four {Cu } fragments
of the complex alone giving a giso = 2.15 signal of much higher in solution because its unit cell consists of two [Cu ] units.
2
} fragments predominate over other species obtained
2
2
5
intensity (Fig. 13 right). This was due to the formation of cate- The presence of double phenoxido-bridges in 1 rendered it
II
chol bound mononuclear Cu species as evidenced from mass least active possibly due to the availability of less room for sub-
4
9
analysis.
strate binding. Monodentate coordination of two different
As the analysis of the reaction rate data for 2 gave the carboxylate groups in 2 and 3 resulted in two types of {Cu2}
highest rate for 20 equivalents of the substrate, EPR measure- fragments showing difference in catalytic activity. The
ments were performed on a mixture of 2 and 3,5-DTBCH
: 20 molar ratio in MeOH (Fig. 13 bottom). The spectrum catechol when the reaction velocity is equal to one half of the
showed a strong and sharp signal at g = 2.01 indicating the maximal velocity for the reaction and also it is a measure of
presence of organic radical species and a doubly split signal binding affinity of the catechol molecules to the {Cu } frag-
with lower intensity shoulders similar to 1, 3 and 4. ments. When the KM value is less it indicates a greater binding
2 M
in a Michaelis–Menten constant K shows the concentration of the
1
2
17172 | Dalton Trans., 2018, 47, 17160–17176
This journal is © The Royal Society of Chemistry 2018

View Article Online
Dalton Transactions
Paper
2 2
Fig. 13 EPR spectra of 3 and 2 (left), mixtures of 3 and 2 with 3,5-DTBCH in a 1 : 100 molar ratio (right) and a mixture of 2 with 3,5-DTBCH in a
1
: 20 molar ratio (bottom) in MeOH at 298 K.
affinity and vice versa. In this work the K
M
value for 2 is less ments. The coordination of trifluoroacetate groups facilitates
than that for 3 indicating superior binding affinity of frag- the transfer of electrons from catechol and semiquinone mole-
II
ments generated from 2. The observed difference in the kcat cules to bound imine functions via Cu centers in 2, compared
values can be explained by considering the rate determining to 3 with attached benzoate groups, resulting in a higher kcat
step as the transfer of electrons from the substrate to the value. A higher kcat value indicates that more substrate mole-
II
imine functions of the ligand through the Cu centers. cules were turned over per second. The catalytic efficiency can
Trifluoroacetate groups, having electron withdrawing fluorine also be compared from the kcat/K
M
values and a catalyst with
atoms, were responsible for the withdrawal of electron density high kcat and low K values is more efficient in this regard.
M
II
−
from Cu centers compared to benzoate group bound frag- The effect of the electron-withdrawing nature of the CF COO
3
This journal is © The Royal Society of Chemistry 2018
Dalton Trans., 2018, 47, 17160–17176 | 17173

View Article Online
Paper
Dalton Transactions
2
Scheme 3 Proposed course of catalytic oxidation of 3,5-DTBCH .
Scheme 4 Proposed routes of aggregation in solution followed by crystallization.
17174 | Dalton Trans., 2018, 47, 17160–17176
This journal is © The Royal Society of Chemistry 2018

View Article Online
Dalton Transactions
Paper
group is further exemplified by the fact that 2 has the highest 2, 3 and 4 are dominated by strong antiferromagnetic
catalytic efficiency (kcat/K ) in the group. Of the four com- exchange interactions, consistent with the presence of the Cu–
M
plexes studied only 2 showed substrate inhibition, most prob- O–Cu phenoxido-bridges. From their bio-relevant behavior, all
ably due to the failure of the trifluoroacetate group to stabilize the complexes act as efficient catalysts for the aerobic oxi-
the {Cu } species in the presence of a large excess of 3,5- dation of 3,5-DTBCH to 3,5-DTBQ in MeOH and their reactiv-
2
2
DTBCH
2
. Oxidation of 3,5-DTBCH
2
by 1, 3 and 4 took place in ity order is 1 < 3 < 2 < 4. For complex 2, the inhibition of reac-
a single phase as confirmed from the first-order exponential tion by substrate molecules was observed at high 3,5-DTBCH
2
growth of 3,5-DTBQ over time.
concentration due to the formation of an inactive mono-
nuclear copper complex–catechol adduct as confirmed by
HRMS analysis and EPR study. ESI-MS also proved that in solu-
Paths for coordination aggregation
tion the {Cu } fragment is the active catalyst for oxidation of
Two important and competitive paths have been identified for
the first time in this work: metal ion coordination induced
ligand arm hydrolysis and hydroxide ion coordination driven
self-aggregation of the initially formed fragments. Varying
reaction conditions and the presence of different carboxylate
2
2
3,5-DTBCH to 3,5-DTBQ. Further work in the same line is cur-
rently in progress in our laboratory to grow larger cages sup-
ported by the in situ generated, metal salt supplied and exter-
nally added anions.
anions lead to the stabilization of two types of {Cu
A and B (Scheme 4).
2
} fragments
Conflicts of interest
In the absence of any carboxylate anion the reaction of HL1
II
with Cu ions triggered a single imine arm hydrolysis to
The authors declare no competing financial interest.
produce {Cu(L2)} fragment A, whose dimerization in solution
resulted in 1. The reaction in the presence of carboxylate ions
provided their termination coordination and in situ generated
Acknowledgements
2
(µ-L1)
−
II
HO ions were utilized to bridge two Cu centers in [Cu
µ-OH)(O CR) ] fragment B. Trifluoroacetate, benzoate and
(
2
2
M. D. is grateful to the IIT Kharagpur for her research
fellowship. M. D. is thankful to Dr K. L. Pruseth for his kind
help in collecting PXRD data. We are also thankful to DST,
New Delhi, for providing the Single Crystal X-ray
Diffractometer facility in the Department of Chemistry, IIT
Kharagpur under the FIST program. The EPR facility by CRF,
IIT Kharagpur is duly acknowledged. M. D. is thankful to Mr
acetate anions, showing varying intermetallic clipping abil-
ities, were responsible for two different types of homometallic
aggregations as [Cu ] complexes 2 and 3, and [Cu ] complex 4.
4
5
Conclusions
Sudipto
Khamrui
for
his
help
with
EPR
While examining the coordination reactivity pattern of HL1
measurements. A. B. C. thanks the University of Glasgow for
financial support.
II
with Cu in the absence or presence of three ancillary bridges
a new family of compounds have been recognized showing
characteristic physical and biophysical behaviors. Ancillary
ligands were utilized to inhibit the ligand-arm hydrolysis and
trap the reactive fragments. Formation of the aggregated pro-
ducts avoiding ligand-arm hydrolysis is reliant on the choice
of solvent system and carboxylate bridges for the self-assembly
References
1 E. M. Shepard and D. M. Dooley, Acc. Chem. Res., 2015, 48,
1218–1226.
of the preformed {Cu
2
} units. In these reactions, coordination
2 C. M. Chang, V. J. Klema, B. J. Johnson, M. Mure,
J. P. Klinman and C. M. Wilmot, Biochemistry, 2010, 49,
2540–2550.
3 I. A. Koval, K. van der Schilden, A. M. Schuitema,
P. Gamez, C. Belle, J.-L. Pierre, M. L. Lueken, B. Krebs,
O. Roubeau and J. Reedijk, Inorg. Chem., 2005, 44, 4372–
4382.
4 M. R. Halvagar, P. V. Solntsev, H. Lim, B. Hedman,
K. O. Hodgson, E. I. Solomon, C. J. Cramer and
W. B. Tolman, J. Am. Chem. Soc., 2014, 136, 7269.
5 H. Sun, K. Harms and J. Sundermeyer, J. Am. Chem. Soc.,
2004, 126, 9550–9955.
6 M. Prudencio, A. S. Pereira, P. Tavares, S. Besson,
I. Cabrito, K. Brown, B. Samyn, B. Devreese, J. Van
Beeumen, F. Rusnak, G. Fauque, J. J. G. Moura, M. Tegoni,
C. Cambillau and I. Moura, Biochemistry, 2000, 39, 3899–
3907.
and bridging of more than one type of smaller groups drive
the resulting reaction paths and products. The presence of a
carboxylate group was crucial to modulate the course of aggre-
gation. In the absence of these carboxylate anions, the avail-
able perchlorate ions from metal ion salts were responsible for
hydrolyzing the imine arm of the ligand to yield 1. The pres-
ence of carboxylate anions on the other hand prevents this
path and takes part in coordination aggregation. Externally
−
−
added CF
3
CO
2
and PhCO
2
ions thus showed preference for
the inter-fragment reaction 2{Cu L1(OH)} → {Cu (μ -O)L1 } +
2
4
4
2
−
H
2
O in 2 and 3. The use of CH
3
COO ions on the other hand
II
lead to the trapping of central Cu ions by two such {Cu
2
L1
OH)} fragments to provide 4 in a different sequence of aggre-
(
−
3
gation. This emphasizes the role of CH COO ions for the self-
assembly of two L1 bound {Cu } fragments, responsible for the
2
trapping of the central Cu center. The magnetic properties of
II
This journal is © The Royal Society of Chemistry 2018
Dalton Trans., 2018, 47, 17160–17176 | 17175

View Article Online
Paper
Dalton Transactions
7
8
9
S. K. Dey and A. Mukherjee, Coord. Chem. Rev., 2016, 310, 28 A. Neves, L. M. Rossi, A. J. Borotoluzzi, B. Szpoganicz,
8
0–115.
C. Wiezbicki, E. Schwingel, W. Haase and S. Ostrovsky,
Inorg. Chem., 2002, 41, 1788–1794.
29 K. Nakamoto, Infrared and Raman Spectra of Inorganic and
coordination compounds, Wiley, New York, 4th edn, 1986.
30 A. Ozarowski, I. B. Szymanska, T. Muziol and J. Jezierska,
J. Am. Chem. Soc., 2009, 131, 10279–10292.
K. Chattopadhyay, B. K. Shaw, S. K. Saha and D. Ray, Dalton
Trans., 2016, 45, 6928–6938.
M. Pait, E. Colacio and D. Ray, Polyhedron, 2015, 88, 90–
1
00.
1
1
1
1
1
0 A. Escuer, G. Vlahopoulou, S. P. Perlepes and
F. A. Mautner, Inorg. Chem., 2011, 50, 2468–2478.
31 G. B. Deacon and R. J. Phillips, Coord. Chem. Rev., 1980, 33,
227–250.
32 G. Tabbì, A. Giuffrida and R. P. Bonomo, J. Inorg. Biochem.,
2013, 128, 137–145.
33 A. Biswas, L. K. Das, M. G. B. Drew, C. Diaz and A. Ghosh,
Inorg. Chem., 2012, 51, 10111–10121.
34 A. W. Addison, T. N. Rao, J. Reedijk, J. V. Rijn and
G. C. Verschoor, J. Chem. Soc., Dalton Trans., 1984, 1349–1356.
1 M. Sarkar, R. Clerac, C. Mathoniere, N. G. R. Hearns,
V. Bertolasi and D. Ray, Inorg. Chem., 2010, 49, 6575–6585.
2 E. Bosch and C. L. Barnes, J. Coord. Chem., 2003, 56, 329–
3
36.
3 H. Liu, H. Wang, H. Wu and D. Niu, J. Coord. Chem., 2005,
8, 1345–1349.
4 A. K. Ghosh, M. Pait, R. Clérac, C. Mathonière, V. Bertolasi,
5
A. Bauza, A. Frontera, K. Pramanik and D. Ray, Dalton 35 Z. Johan and J. Hak, Chem. Erde, 1958, 20, 49–50.
Trans., 2014, 43, 4076–4085.
36 D. Zhang, H. Berger, R. K. Kremer, D. Wulferding,
P. Lemmens and M. Johnsson, Inorg. Chem., 2010, 49,
9683–9688.
37 A. K. Ghosh, R. Clérac, C. Mathoniére and D. Ray,
Polyhedron, 2013, 54, 196–200.
1
5 H. Kumari, A. V. Mossine, S. R. Kline, C. L. Dennis,
D. A. Fowler, S. J. Teat, C. L. Barnes, C. A. Deakyne and
J. L. Atwood, Angew. Chem., Int. Ed., 2012, 51, 1452–1454.
6 R. R. Gagne, C. L. Spiro, T. J. Smith, C. A. Hamann,
1
W. R. Thies and A. D. Shiemke, J. Am. Chem. Soc., 1981, 38 M. Sarkar, R. Clerac, C. Mathoniere, N. G. R. Hearns,
03, 4073–4081. V. Bertolasi and D. Ray, Inorg. Chem., 2011, 50, 3922–3933.
7 SAINT, SMART and XPREP, Siemens Analytical X-ray 39 N. F. Chilton, R. P. Anderson, L. D. Turner, A. Soncini and
Instruments Inc., Madison, WI, 1995. K. S. Murray, J. Comput. Chem., 2013, 34, 1164–1175.
8 G. M. Sheldrick, SHELXS-2014, University of Göttingen, 40 X. Wang, K.-Q. Zhao, M. R. J. Elsegood, T. J. Prior, X. Liu,
1
1
1
1
Göttingen, Germany, 2014.
L. Wu, S. Sanz, E. K. Brechin and C. Redshaw, RSC Adv.,
2015, 5, 57414–57424.
9 G. M. Sheldrick, Crystal structure refinement with SHELXL,
Acta Crystallogr., Sect. A: Found. Crystallogr., 2008, 64, 112– 41 M. S. Jana, S. Dey, J. L. Priego, R. Jiménez-Aparicio,
22. T. K. Mondal and P. Roy, Polyhedron, 2013, 59, 101–106.
0 L. J. Farrugia, WinGX-Version 2014.1, J. Appl. Crystallogr., 42 L. K. Thompson, S. K. Mandal, S. S. Tandon, J. N. Bridson
012, 45, 849–854. and M. K. Park, Inorg. Chem., 1996, 35, 3117–3125.
1 G. M. Sheldrick, SADABS Software for Empirical Absorption 43 F. Tuna, L. Patron, Y. Journaux, M. Andruh, W. Plass and
Correction, University of Göttingen, Institute für J.-C. Trombe, J. Chem. Soc., Dalton Trans., 1999, 539–545.
Anorganische Chemieder Universität, Göttingen, Germany, 44 M. Das, G. A. Craig, D. Escudero, M. Murrie, A. Frontera
999–2003. and D. Ray, New J. Chem., 2018, 42, 14349–14364.
2 Z. Otwinowski and W. Minor, Processing of X-ray 45 A. Rompel, H. Fischer, D. Meiwes, K. Buldt-
1
2
2
2
1
2
Diffraction Data Collected in Oscillation Mode, in Methods
in Enzymology: Macromolecular Crystallography, part A, ed.
Karentzopoulos, A. Magrini, C. Eicken, C. Gerdemann and
B. Krebs, FEBS Lett., 1999, 445, 103–110.
C. W. Carter Jr and R. M. Sweet, Academic Press, New York, 46 E. Monzani, G. Battaini, A. Perotti, L. Casella, M. Gullotti,
1
997, vol. 276, pp. 307–326.
3 R. H. Blessing, Acta Crystallogr., Sect. A: Found. Crystallogr.,
995, 51, 33–37.
L. Santagostini, G. Nardin, L. Randaccio, S. Geremia,
P. Zanello and G. Opromolla, Inorg. Chem., 1999, 38, 5359–
5369.
2
2
1
4 A. Altomare, M. C. Burla, M. Camalli, G. L. Cascarano, 47 T. Ghosh, J. Adhikary, P. Chakraborty, P. K. Sukul,
C. Giacovazzo, A. Guagliardi, A. G. G. Moliterni, G. Polidori
and R. J. Spagna, J. Appl. Crystallogr., 1999, 32, 115–119.
5 M. Nardelli, J. Appl. Crystallogr., 1995, 28, 659.
6 L. J. Farrugia, J. Appl. Crystallogr., 1999, 32, 837–838.
7 DIAMOND, Visual Crystal Structure Information System, 49 J. Mukherjee and R. Mukherjee, Inorg. Chim. Acta, 2002,
version 3.1, Crystal Impact, Bonn, Germany, 2004. 337, 429–438.
M. S. Jana, T. K. Mondal, E. Zangrando and D. Das, Dalton
Trans., 2014, 43, 841–852.
48 S. M. Yi, K. V. Narasimhulu, R. I. Samoilova, R. B. Gennis
and S. A. Dikanov, J. Biol. Chem., 2010, 285, 18241–18251.
2
2
2
17176 | Dalton Trans., 2018, 47, 17160–17176
This journal is © The Royal Society of Chemistry 2018