Spectro-electrochemical and DFT studies of a planar Cu(II)-phenolate complex active in the aerobic oxidation of primary alcohols

Article abstract of DOI:10.1016/j.ica.2011.03.029

A square-planar compound [Cu(pyrimol)Cl] (pyrimol = 4-methyl-2-N-(2- pyridylmethylene)aminophenolate) abbreviated as CuL-Cl) is described as a biomimetic model of the enzyme galactose oxidase (GOase). This copper(II) compound is capable of stoichiometric aerobic oxidation of activated primary alcohols in acetonitrile/water to the corresponding aldehydes. It can be obtained either from Hpyrimol (HL) or its reduced/hydrogenated form Hpyramol (4-methyl-2-N-(2-pyridylmethyl)aminophenol; H₂L) readily converting to pyrimol (L^-) on coordination to the copper(II) ion. Crystalline CuL-Cl and its bromide derivative exhibit a perfect square-planar geometry with Cu-O(phenolate) bond lengths of 1.944(2) and 1.938(2) . The cyclic voltammogram of CuL-Cl exhibits an irreversible anodic wave at +0.50 and +0.57 V versus ferrocene/ferrocenium (Fc/Fc⁺) in dry dichloromethane and acetonitrile, respectively, corresponding to oxidation of the phenolate ligand to the corresponding phenoxyl radical. In the strongly donating acetonitrile the oxidation path involves reversible solvent coordination at the Cu(II) centre. The presence of the dominant Cu^II-L^? chromophore in the electrochemically and chemically oxidised species is evident from a new fairly intense electronic absorption at 400-480 nm ascribed to a several electronic transitions having a mixed π → π*(L^?) intraligand and Cu-Cl → L^? charge transfer character. The EPR signal of CuL-Cl disappears on oxidation due to strong intramolecular antiferromagnetic exchange coupling between the phenoxyl radical ligand (L ^?) and the copper(II) centre, giving rise to a singlet ground state (S = 0). The key step in the mechanism of the primary alcohol oxidation by CuL-Cl is probably the α-hydrogen abstraction from the equatorially bound alcoholate by the phenoxyl moiety in the oxidised pyrimol ligand, Cu-L ^?, through a five-membered cyclic transition state.

Full text of DOI:10.1016/j.ica.2011.03.029

Inorganica Chimica Acta 374 (2011) 406–414
Contents lists available at ScienceDirect
Inorganica Chimica Acta
journal homepage: www.elsevier.com/locate/ica
Spectro-electrochemical and DFT studies of a planar Cu(II)–phenolate complex
active in the aerobic oxidation of primary alcohols
Palanisamy Uma Maheswari a, František Hartl b, , Manuel Quesada a,1, Francesco Buda a, Martin Lutz c,
⇑
Anthony L. Spek c, Patrick Gamez a,1, Jan Reedijk a,d,
⇑
a Leiden Institute of Chemistry, Leiden University, P.O. Box 9502, 2300 RA Leiden, The Netherlands
b Department of Chemistry, University of Reading, Whiteknights, Reading RG6 6AD, UK
c Bijvoet Centre for Biomolecular Research, Crystal and Structural Chemistry, Utrecht University, Padualaan 8, 3584 CH Utrecht, The Netherlands
d Department of Chemistry, King Saud University, P.O. Box 2455, Riyadh 11451, Saudi Arabia
a r t i c l e i n f o
a b s t r a c t
Article history:
Available online 13 March 2011
A square-planar compound [Cu(pyrimol)Cl] (pyrimol = 4-methyl-2-N-(2-pyridylmethylene)aminopheno-
late) abbreviated as CuL–Cl) is described as a biomimetic model of the enzyme galactose oxidase (GOase).
This copper(II) compound is capable of stoichiometric aerobic oxidation of activated primary alcohols in
acetonitrile/water to the corresponding aldehydes. It can be obtained either from Hpyrimol (HL) or its
reduced/hydrogenated form Hpyramol (4-methyl-2-N-(2-pyridylmethyl)aminophenol; H2L) readily con-
verting to pyrimol (Lꢀ) on coordination to the copper(II) ion. Crystalline CuL–Cl and its bromide deriva-
tive exhibit a perfect square-planar geometry with Cu–O(phenolate) bond lengths of 1.944(2) and
1.938(2) Å. The cyclic voltammogram of CuL–Cl exhibits an irreversible anodic wave at +0.50 and
+0.57 V versus ferrocene/ferrocenium (Fc/Fc+) in dry dichloromethane and acetonitrile, respectively, cor-
responding to oxidation of the phenolate ligand to the corresponding phenoxyl radical. In the strongly
donating acetonitrile the oxidation path involves reversible solvent coordination at the Cu(II) centre.
The presence of the dominant CuII–LÅ chromophore in the electrochemically and chemically oxidised spe-
cies is evident from a new fairly intense electronic absorption at 400–480 nm ascribed to a several elec-
Dedicated to Prof. Dr. Wolfgang Kaim for his
significant contribution to coordination
chemistry and spectro-electrochemistry
Keywords:
Copper(II)
Galactose oxidase
Primary alcohol oxidation
Magnetic susceptibility
Spectro-electrochemistry
DFT
tronic transitions having a mixed
p ?
p⁄(LÅ) intraligand and Cu–Cl ? LÅ charge transfer character. The EPR
signal of CuL–Cl disappears on oxidation due to strong intramolecular antiferromagnetic exchange cou-
pling between the phenoxyl radical ligand (LÅ) and the copper(II) centre, giving rise to a singlet ground
state (S = 0). The key step in the mechanism of the primary alcohol oxidation by CuL–Cl is probably
the
oxidised pyrimol ligand, Cu–LÅ, through a five-membered cyclic transition state.
Crown Copyright Ó 2011 Published by Elsevier B.V. All rights reserved.
a-hydrogen abstraction from the equatorially bound alcoholate by the phenoxyl moiety in the
1. Introduction
unusual because, in contrast to most copper proteins carrying
out multielectron redox reactions at multinuclear Cu(I) M Cu(II)
active sites, it uses a single copper centre to carry out a two-
electron redox process. Spectroscopic studies of the GOase indicate
that the isolated inactive enzyme contains a CuII–tyrosinate frag-
ment that is converted upon treatment with oxidants into the ac-
tive form containing a CuII–tyrosinyl radical pair in the equatorial
position [11].
The search for efficient oxidation catalysts is increasingly being
guided and stimulated by considering the active site structures of
metalloenzymes. An example is the fungal enzyme galactose oxi-
dase (GOase) catalysing selectively aerobic oxidation of a wide
range of primary alcohols to the corresponding aldehydes, with a
concomitant formation of stoichiometric amounts of dihydrogen
peroxide by a coupled dioxygen reduction [1–10]. The enzyme is
Key evidences for the formulation of the active form include the
silent EPR response ascribed to antiferromagnetic exchange cou-
pling between the two S1/2 centres, the X-ray absorption edge data
indicative of the presence of Cu(II), the resonance Raman data
showing diagnostic phenoxyl radical vibrations, and an intense
⇑
Corresponding authors. Addresses: Department of Chemistry, University of
Reading, Whiteknights, Reading RG6 6AD, UK (F. Hartl), Leiden Institute of
Chemistry, Leiden University, P.O. Box 9502, 2300 RA Leiden, The Netherlands (J.
Reedijk). Tel.: +44 118 378 7695; fax: +44 118 378 6331 (F. Hartl).
E-mail address: f.hartl@reading.ac.uk (F. Hartl).
absorption band at kmax = 444 nm (
which has been postulated to involve a
e
max = 5190 Mꢀ1 cmꢀ1, GOase),
p⁄ electronic transition
p
?
of the phenoxyl radical [12–16]. The ability of the mononuclear ac-
tive site to carry out a two-electron process has been rationalised
1
Present address: ICREA, Universitat de Barcelona, Departament de Quimica
Inorganica, 08028 Barcelona, Spain.
0020-1693/$ - see front matter Crown Copyright Ó 2011 Published by Elsevier B.V. All rights reserved.
doi:10.1016/j.ica.2011.03.029

P. Uma Maheswari et al. / Inorganica Chimica Acta 374 (2011) 406–414
407
by a redox cycle between the active CuII–tyrosinyl unit and the re-
2. Experimental
duced Cu(I) state. The CuII–tyrosinyl radical moiety is assumed to
be directly responsible for the hydrogen atom abstraction from
the alcoholate substrate bound in the equatorial position.
2.1. Materials
The unusual nature of the cysteine-modified CuII–tyrosinyl rad-
ical species, its key role in the enzyme catalysis and the challenge
of fully understanding the mechanism of the alcohol oxidation
provide a sufficient thrust for undertaking thorough studies of
functional models of the active site of the GOase [17,18].
All chemicals for the syntheses were used as received without
further purification. The ligand (4-methyl-2-N-(2-pyridyl-
methyl)aminophenol (Hpyramol)) was obtained in a single-step
reaction as previously reported [42,43]. Dichloromethane and ace-
tonitrile (Acros) for spectro-electrochemistry were freshly distilled
from CaH2 under a dry nitrogen atmosphere. The supporting elec-
trolyte Bu4NPF6 (Aldrich) was recrystallized twice from absolute
ethanol and dried overnight under reduced pressure at 80 °C. Fer-
rocene, cobaltocenium hexafluoridophosphate and tris(4-bromo-
phenyl)aminium hexachloridoantimonate (all Aldrich) were used
as received.
Although a large number of CuII–phenolate-based structural
models of the active and inactive forms of the GOase have been
synthesized and spectroscopically characterised [19–27], only a
few functional models have been reported so far [6,8,28–30]. Pio-
neering studies by Kitajima et al. in the mid-1980s have dealt with
two tetrahedrally distorted phenolic Schiff-base copper(II) com-
plexes capable of oxidising primary alcohols to corresponding
aldehydes in the presence of an excess of alkoxide [31]. One decade
later, Stack and co-worker described the use of a functional mono-
nuclear Cu(II) model compound as a catalyst of aerobic oxidation of
activated benzyl and allylic alcohols with high TONs, again in the
presence of an excess of the base (alkoxide) as co-catalyst [32].
More recently, Wieghardt and co-workers reported a unique
mononuclear 2,20-thiobis(phenolato) triethylamine copper(II)
compound that oxidised both primary and secondary alcohols to
the corresponding aldehydes and ketones, respectively, and/or to
1,2-glycol derivatives [33–35]. The selective aerobic oxidation of
primary alcohols (but not methanol) was achieved with a related
Cu(II) compound containing two-electron-reducible iminosemi-
quinone ligands and NEt3. Mononuclear copper complexes with
sterically demanding Schiff-base ligands were applied by Stack
and co-workers in the stoichiometric conversion of benzyl alcohol
to benzaldehyde in the presence of a base and an oxidant [36]. Itoh
et al. performed the same oxidation using mononuclear copper
complexes bearing thioether-linked phenolic ligands. The impor-
2.1.1. Synthesis of 4-methyl-2-N-(2-pyridylmethylene)aminophenol
(Hpyrimol)
A solution of 10.7 g (0.1 mol) of picolinaldehyde in methanol
(75 mL) was added, under constant stirring, to a solution of
12.4 g (0.1 mol) of 2-amino-4-methylphenol in methanol (50 mL)
dried on molecular sieves (4 Å). The reaction mixture was refluxed
for 2 h at 70 °C. Next, methanol was evaporated under a reduced
pressure. The deep green residue was redissolved in CH2Cl2
(200 mL), filtered and layered with n-hexane (200 mL). The result-
ing bilayer was left at ꢀ20 °C to promote crystallization of the pure
compound. After 2 days, bright yellow crystals were obtained. The
crystals were collected and washed with diethyl ether, and subse-
quently dried under argon. Yield: 53%. 1H NMR (CDCl3): d 8.81 (s),
8.72 (d), 8.20 (d), 7.82 (t), 7.37 (m), 7.25 (d), 7.2 (s), 7.1 (s), 7.06 (d),
6.90 (d), 2.31 (s) ppm.
tance of
a reducible copper(II) site and the interconversion
2.1.2. Synthesis of [Cu(pyrimol)X]ꢁH2O (X = Cl, Br)
between the phenol and phenoxyl radical ligand redox forms in
the two-electron oxidation of alcohols were clearly demonstrated
by comparison with analogous but inactive Zn(II) complexes
[6,37–40].
As an alternative to the literature procedure [41] [Cu(pyri-
mol)Cl]ꢁH2O was prepared by layering a solution of Hpyrimol in
THF and an aqueous solution of CuCl2. A solution of CuCl2ꢁ2H2O
(127.3 mg, 0.747 mmol), in 15 mL H2O was added dropwise to a
solution of Hpyrimol (212.2 mg, 1 mmol) in 15 mL water. Dark
red crystals were formed in 34% yield. Anal. Calc. for C13H11ClCu-
N2OꢁH2O (Mw: 328.24): C, 47.57; H, 3.99; N, 8.53. Found: C,
47.54; H, 4.36; N, 8.50%. The analogous bromide compound was
prepared similarly and obtained as dark red needles. Yield: 19%.
Anal. Calc. for C13H11BrCuN2OꢁH2O (Mw: 372.70): C, 47.57; H,
3.99; N, 8.53. Found: C, 47.52; H, 3.96; N, 8.55%. [Cu(pyrimol)
Br]ꢁH2O is isomorphous to [Cu(pyrimol)Cl]ꢁH2O prepared from
the ligands Hpyramol or Hpyrimol (see below).
Herein, we report on a square-planar copper(II) compound
[Cu(pyrimol)Cl] (pyrimol = 4-methyl-2-N-(2-pyridylmethylene)-
aminophenolate, Lꢀ; Fig. 1b) capable of oxidising primary alcohols
to corresponding aldehydes. This functional model of the GOase
has been obtained from CuCl2 and Hpyrimol (HL; Fig. 1a). The
reduced and hydrogenated form of the ligand, Hpyramol (4-
methyl-2-N-(2-pyridylmethyl)aminophenol, H2L) can also be used,
readily converting to pyrimol (Lꢀ) upon coordination to the cop-
per(II) ion, as communicated elsewhere [41]. The corresponding
bromide compound, CuL–Br, has also been prepared and structur-
ally fully characterised.
2.2. X-ray crystallography
Reflections were measured on a Nonius Kappa CCD diffractom-
eter with
a
rotating anode (graphite monochromator,
k = 0.71073 Å) up to resolution of (sin h/k)max = 0.65 Åꢀ1. Intensities
were integrated with EvalCCD [44], using an accurate description
of the experimental setup for the prediction of the reflection con-
tours. The structures were refined with SHELXL-97 [45] against F2
of all reflections. Non-hydrogen atoms were refined with aniso-
tropic displacement parameters. All hydrogen atoms were located
in the difference Fourier map. The hydrogen atoms at C1 and in the
water molecule were refined freely with isotropic displacement
parameters; all other hydrogen atoms were refined with a riding
model. Geometry calculations and checking for higher symmetry
was performed with the PLATON program [46].
Fig. 1. Schematic molecular structures of (a) Hpyrimol (HL) and (b) [Cu(pyrimol)X]
(CuL–X, X = Cl and Br).

408
P. Uma Maheswari et al. / Inorganica Chimica Acta 374 (2011) 406–414
2.3. [Cu(pyrimol)Cl]ꢁH2O
oxidation of CuL–Cl with tris(4-bromophenyl)aminium hexa-
chloridoantimonate (Eo = +0.70 V versus Fc/Fc+) [50] was per-
formed in acetonitrile and dichloromethane under argon on UV–
Vis monitoring. Acetonitrile was evaporated under reduced pres-
sure to give a dark-green powder used for the magnetic measure-
ments. Corrections were applied for the diamagnetic responses of
the compound, as estimated from the Pascal constants [51].
Cyclic voltammetry of 10ꢀ3 M CuL–Cl in a single-compartment
cell was carried out with an Autolab PGSTAT10 potentiostat (Eco
Chemie) controlled by the GPES4 software, or with an EG&G PAR
Model 283 potentiostat operated with the PAR Power CVÒ soft-
ware. The dichloromethane and acetonitrile solutions contained
10ꢀ1 M Bu4NPF6 as the supporting electrolyte. The three-electrode
system consisted of a carefully polished Pt disc working electrode,
a platinum auxiliary electrode, and an Ag/AgCl reference electrode.
The voltammetric response of the standard ferrocene/ferrocenium
(Fc/Fc+) couple was found in this system at +0.55 V versus Ag/AgCl.
Cobaltocenium hexafluoridophosphate was added as internal po-
tential standard (ꢀ1.33 V versus Fc/Fc+) when an Ag wire was used
as a pseudo-reference electrode.
C
13H11ClCuN2O(H2O), Mw = 328.24 g molꢀ1, dark red needle,
0.60 ꢂ 0.09 ꢂ 0.06 mm3, triclinic, P1 (No. 2), a = 6.6991(4),
ꢀ
b = 8.7519(4), c = 12.2726(7) Å,
a = 100.042(2), b = 100.131(2),
c
l
= 110.053(1)°,
V = 643.60(6) Å3,
Z = 2,
Dx = 1.694 g cmꢀ3
,
= 1.90 mmꢀ1. In total 11,897 reflections were measured at
T = 150 K; 2945 reflections were unique (Rint = 0.0178). An absorp-
tion correction based on multiple measured reflections was ap-
plied using the program SADABS [47] (0.34–0.89 correction range).
The structure was solved with the program DIRDIF-99 [48] using
automated Patterson Methods and found to be similar to the
room-temperature structure of the compound starting differently
prepared from Hpyramol [41]. The 185 parameters were refined
with no restraints. R1/wR2 [I > 2r(I)]: 0.0291/0.0799. R1/wR2 [all
reflections]: 0.0323/0.0819. S = 1.165. Residual electron density
ranges between ꢀ0.44 and 0.68 e/Å3.
2.4. [Cu(pyrimol)Br]ꢁH2O
C
13H11BrCuN2O(H2O), Mw = 372.70 g molꢀ1, dark red needle,
The air-tight UV–Vis and EPR spectro-electrochemical cells em-
ployed for the in situ oxidation of 2 mM CuL–Cl at variable temper-
atures have been described in detail elsewhere [52–55]. The
electrode potential was controlled during the in situ electrolyses
by a PA4 potentiostat (EKOM, Polná, Czech Republic). UV–Vis spec-
tra were recorded with a HP 8453 diode array spectrophotometer,
and the X-band EPR spectra with a Varian Century E-104A
spectrometer.
Density functional theory (DFT) calculations were performed
with the GAUSSIAN 03 package [56]. We used the hybrid B3LYP ex-
change and correlation functional [57,58] and a 6-311G(d,p) basis
set. All calculations are spin-unrestricted. The choice of this hybrid
functional is important to obtain accurate results both for the
structural and absorption properties of the CuL–Cl complex. The
DFT-optimised geometry of the compound is in excellent agree-
ment with the corresponding X-ray crystal structure data (see
Table S1 in Supplementary material). Time-dependent DFT calcula-
tions (TD-DFT) were performed with the same functional and basis
set to evaluate electronic absorption spectra. The geometry of all
complexes was optimised prior to the TD-DFT calculations. TD-
DFT calculations were also performed in the presence of a solvent
treated within the Polarizable Continuum Model (PCM) [59]. The
0.78 ꢂ 0.04 ꢂ 0.04 mm3, triclinic, P1 (No. 2), a = 6.7419(3),
ꢀ
b = 8.8574(6), c = 12.2550(6) Å,
a = 101.696(2), b = 97.836(2),
c
l
= 110.233(2)°,
V = 655.21(6) Å3,
Z = 2,
Dx = 1.889 g cmꢀ3
,
= 4.71 mmꢀ1. In total 13,266 reflections were measured at
T = 150 K; 2981 reflections were unique (Rint = 0.0837). The crystal
appeared to be non-merohedrally twinned with a twofold rotation
about the crystallographic a-axis as twin operation. This twin oper-
ation was taken into account during the integration of the intensi-
ties and the refinement as a HKLF5 refinement [49]. An analytical
absorption correction was applied (0.23–0.86 correction range).
The coordinates of the isomorphous [Cu(pyrimol)Cl]ꢁH2O were ta-
ken as starting model for the refinement. The 186 parameters were
refined with no restraints on all unique data. R1/wR2 [I > 2r(I)]:
0.0414/0.1041. R1/wR2 [all reflections]: 0.0546/0.1127. S = 1.152.
The twin fraction refined to 0.563(3). Residual electron density
ranges between ꢀ0.75 and 1.07 e/Å3.
2.5. Experimental procedures and methods
The oxidation of primary alcohols was carried out in air within a
100 mL three-necked round-bottom flask. Typically, the aerated
alcohol (2 mmol) was diluted with acetonitrile (15 mL) and the
base, tBuOK dissolved in H2O (5 mL), was added. The solution
was magnetically stirred for 5 min. CuCl2 (2 mmol) and Hpyramol
or Hpyrimol (2 mmol) were added stepwise at this stage, and the
reaction solution turned dark red. Crystalline CuL–Cl can also be
used instead of the in situ CuCl2–Hpyramol/Hpyrimol combination.
Samples for analyses were collected at regular intervals. The organ-
ic layer was extracted with diethyl ether/H2O (1:1, 10 mL), and the
extract was concentrated to 1 mL. The conversion (%) was then
determined by gas chromatography, using co-injected 1,2-dibro-
mobenzene as an internal standard. The GC analysis was carried
out using a Hewlett–Packard 5890 Series GC (CP-Sil-5 CB WCOT
Fused Silica capillary column, 50 m ꢂ 0.25 mm inside diameter
ionisation potential was estimated by a
DSCF calculation, i.e. by
taking the difference in the total energy of the complex before
and after its one-electron oxidation.
3. Results and discussion
3.1. Crystal structures
The compounds [Cu(pyrimol)X] (X = Cl, Br) can be obtained
both from the Hpyramol (H2L) and Hpyrimol (HL) ligands by a bi-
layer method, with the ligand dissolved in THF or ethyl acetate and
the Cu(II) salt in H2O. Crystals of the product appear typically with-
in 2 days at 293–273 K. The compounds reported herein, obtained
from the bilayer containing the Hpyrimol ligand and CuCl2/CuBr2
salts, are isomorphous to CuL–Cl prepared earlier [41] from CuCl2
and Hpyramol that readily dehydrogenates upon coordination to
give the pyrimol ligand, Lꢀ.
The coordination geometries of Cu(II) are perfectly square pla-
nar, with the NNO pyrimol and halide (Cl/Br) coordination. Views
of CuL–X (X = Cl, Br) are presented in Fig. 2 (for X = Br; bond
parameters are in the caption), Fig. S1 (for X = Cl) and Figs. S2
and S3 for the packing diagrams of both compounds. The structural
parameters regarding hydrogen bonds are collected in Table 1. The
(i.d.), 0.25 lm film thickness) with dihydrogen carrier gas. The
products of the reaction were determined by comparison with
the commercially available carbonyl compounds.
Elemental analyses (C, H, N) were carried out on a Perkin-Elmer
2400 series II analyser. X-band EPR measurements of a crystalline
sample of CuL–Cl were performed at 77 K, Bruker-EMXplus spec-
trometer (field calibrated with DPPH (g = 2.0036)). Magnetic sus-
ceptibility measurements at 5–300 K were carried out using a
Quantum Design MPMS-5S SQUID magnetometer, in a 1 kG ap-
plied field. Data were corrected for the experimentally determined
contribution of the sample holder. The selective one-electron

P. Uma Maheswari et al. / Inorganica Chimica Acta 374 (2011) 406–414
409
Fig. 3. Cyclic voltammogram of 10ꢀ3 M CuL–Cl in the anodic region. Conditions:
Freshly polished Pt working electrode (apparent surface area of 0.312 cm2),
v
= 100 mV sꢀ1, dichloromethane/10ꢀ1 M Bu4NPF6, room temperature.
values were confirmed against the standard cobaltocenium hexa-
fluoridophosphate used as an internal reference. The oxidised spe-
cies then reduces back on the reverse scan in dichloromethane at
Ep,c = ꢀ0.34 V versus Fc/Fc+ (0.14 mm2 Pt microdisc) or ꢀ0.45 V
versus Fc/Fc+ (0.312 cm2 Pt disc; the irreversible cathodic wave
R1 in Fig. 3) and in acetonitrile at ꢀ0.15 V versus Fc/Fc+
(0.14 mm2 Pt microdisc). The same cathodic wave was indeed also
observed in the cyclic voltammogram recorded after oxidation of
[CuL–Cl] with 1 equiv (4-BrC6H4)3N+SbCl6ꢀ. These electrode poten-
tial values for the irreversible redox reactions correspond to the
m
= 100 mV sꢀ1. The significant solvent dependence of
Fig. 2. Molecular structure of [Cu(pyrimol)Br]ꢁH2O (50% probability level). Selected
bond lengths (Å) and angles [°]: Cu1–O1 1.938(3), Cu1–Br1 2.3553(6), Cu1–N1
2.018(4), Cu1–N2 1.965(3); O1–Cu1–Br1 95.53(9), Br1–Cu1–N1 100.57(10), N1–
Cu1–N2 80.54(14), N2–Cu1–O1 83.36(13).
scan rate
the reduction potential R1 points to the coordination of acetonitrile
molecules to the oxidised form of CuL–Cl. To prove this assump-
tion, the voltammetric cycle of CuL–Cl in acetonitrile recorded at
m
= 500 mV sꢀ1 was interrupted for 10 s beyond the anodic wave
Table 1
O1 to accumulate the oxidised product at the electrode surface;
the scan direction was then reversed back beyond the cathodic
wave R1 of the oxidised product and switched again anodically
without interruption. Three complete redox cycles were recorded
in this way. Importantly, the back reduction of the oxidised prod-
uct at R1 forms a new neutral species that oxidises 460 mV less
positively than CuL–Cl; the anodic wave O1 of the parent complex
becomes indeed significantly smaller compared to the initial for-
ward scan. The new anodic wave was not observed in dichloro-
methane. The cyclovoltammetric results indicate axial
coordination of donating acetonitrile to planar [CuLÅ–Cl]+. In
dichloromethane, oxidised [CuLÅ–Cl]+ is assumed to coordinate
weakly a water molecule coming from crystals of parent [Cu(pyri-
mol)Cl]ꢁH2O (see Section 2) and from the electrolyte [60]. The ano-
dic wave O1 of CuL–Cl and the cathodic wave R1 of Fig. 3, hence do
not belong to the same (quasireversible) redox couple. The back
reduction of the five-coordinate oxidised species leads to the
recovery of parent CuL–Cl. For [CuLÅ–Cl(MeCN)]+ the corresponding
neutral transient CuL–Cl(MeCN) is readily detectable on the sub-
second voltammetric time scale. It is important to note that the po-
tential difference of 0.46 eV between the oxidations of CuL–Cl and
short-lived CuL–Cl(MeCN) agrees exactly with the difference in the
ionisation potentials of the two complexes (0.44 eV) obtained from
DFT calculations (see below).
Hydrogen-bonding parameters for [Cu(pyrimol)X]ꢁH2O (X = Cl, Br).
D–Hꢁ ꢁ ꢁA
D–H (Å)
Hꢁ ꢁ ꢁA (Å)
Dꢁ ꢁ ꢁA (Å)
D–Hꢁ ꢁ ꢁA (°)
[Cu(pyrimol)Cl]ꢁH2O
O2–H2Aꢁ ꢁ ꢁO1i
0.83(4)
0.75(4)
2.04(4)
2.10(4)
2.830(3)
2.832(3)
160(3)
165(4)
O2–H2Bꢁ ꢁ ꢁO1ii
[Cu(pyrimol)Br]ꢁH2O
O2–H2Aꢁ ꢁ ꢁO1i
0.75(7)
0.76(6)
2.13(7)
2.10(6)
2.857(5)
2.842(5)
162(7)
162(5)
O2–H2Bꢁ ꢁ ꢁO1ii
Symmetry operations: i—x, y ꢀ 1, z; ii—2 ꢀ x, 1 ꢀ y, 1 ꢀ z.
compound CuL–Cl has already been reported by some of us to
cleave DNA oxidatively without any specific reductant added
[41]. This process has revealed the ability of CuL–Cl to form reac-
tive radical species and to abstract hydrogen atoms from a sugar
substrate. The oxidation of primary alcohols to corresponding alde-
hydes under stoichiometric conditions is discussed in the light of
spectroscopic and electrochemical properties of CuL–Cl, with refer-
ences to the GOase catalytic cycle.
Two-dimensional hydrogen-bonding patterns were found be-
tween the non-coordinated H2O molecules present in the lattice
and the adjacent phenolate oxygen from the Cu-pyrimol
compound.
3.2. Cyclic voltammetry
3.3. UV–Vis spectro-electrochemistry
In the anodic region, the cyclic voltammogram of CuL–Cl shows
an irreversible wave (O1) at Ep,a = +0.50 V versus Fc/Fc+ in dichloro-
methane (Fig. 3) and +0.57 V versus Fc/Fc+ in acetonitrile. These
The UV–Vis spectrum of CuL–Cl in dichloromethane shows
prominent absorption bands at 561, 361, 300 and 238 nm. These
bands become replaced during the oxidation within an OTTLE cell

410
P. Uma Maheswari et al. / Inorganica Chimica Acta 374 (2011) 406–414
at the O1 electrode potential by new absorptions at 440, 343 and
271 nm, as shown in Fig. 4. Due to the significant involvement of
the pyrimol ligand (see below), it is not surprising that the low-
est-energy band is strongly affected by the largely pyrimol-based
oxidation, producing a phenoxyl radical species [61] that strongly
stable, but absorbing at 481 nm (Fig. 5b). Unfortunately, we did not
succeed to crystallize the oxidised complex for X-ray diffraction
analysis, despite several attempts. It cannot be excluded that
[CuLÅ–Cl(MeCN)]+ is sufficiently stable only in the excess of
acetonitrile.
absorbs at kmax = 435 nm (
e
max = 5750 Mꢀ1 cmꢀ1, Fig. 4).
Addition of N2-saturated water to the acetonitrile solution of
[CuLÅ–Cl(MeCN)]+ causes slow disappearance of the complex. As
noted above, parent CuL–Cl is stable in aqueous acetonitrile. When
monitoring the titration of the activated primary alcohol species
(deprotonated with NaOH in H2O; see Table 3) with CuL–Cl in ace-
tonitrile/H2O 3:1 (v/v) at room temperature by UV–Vis spectros-
copy, the decreasing visible absorption of parent CuL–Cl is an
indirect evidence for the conversion of the pyrimol ligand to the
oxidised phenoxyl radical species caused by the substrate
(alcoholate) binding to CuL–Cl. Compound CuL–Cl is inert at low
temperature (ꢀ20 °C) toward the alcohol oxidation and shows no
UV–Vis spectral changes. In order to partly elucidate the mecha-
nism of the alcohol oxidation, attempts were made to prove the
formation of H2O2 during the reaction, using the ‘Fox Assay’ [62].
However, no H2O2 was detected. The side product of the primary
alcohol to aldehyde oxidation is therefore assumed to be H2O or
a non-diffusible peroxido species, under stoichiometric conditions.
In acetonitrile, the parent compound absorbs in the visible
region prominently at 541 nm and the oxidised product
[CuLÅ–Cl(MeCN)]+ at 407 nm. The thin-layer cyclic voltammograms
recorded during the in situ oxidation (m
= 2 mV sꢀ1) were indeed
very close to the conventional scans (cf. Fig. 3), which supports
the assignment. The back reduction of the phenoxyl radical
product at the R1 wave led to ca 80% reappearance of the UV–Vis
spectrum of the starting CuL–Cl species.
In pure dichloromethane, acetonitrile and acetonitrile/H2O 3:1
(v/v) without the electrolyte, the maximum of the visible absorp-
tion of CuL–Cl shifts from 566 to 540 and 488 nm, respectively;
the dipole moment is likely to change for a phenolate-to-imino-
pyridine intramolecular charge transfer in this region (see below).
One-electron oxidation of CuL–Cl in dichloromethane at room
temperature with (4-BrC6H4)3N+SbCl6ꢀ gives in steps a fully stable
product absorbing at 438 nm (Fig. 5a), most probably identical to
that observed during the in situ spectro-electrochemical experi-
ment. In dry acetonitrile, the one-electron-oxidised product is also
3.4. EPR spectroscopy
The EPR spectrum of CuL–Cl in dichloromethane shows at room
temperature
a well-resolved isotropic signal with a slightly
decreasing line width in a stronger magnetic field, which appears
as typical for d9 Cu(II) (Fig. 6, Aav = 82.5(5) G, gav = 2.107(1); a sim-
ulated spectrum for these parameters in given in Fig. S4). More
precisely, the Mulliken analysis has revealed 63% of the spin den-
sity on the Cu centre, the rest residing on the four coordinated
donor atoms (Cl, N1, N2, O) (cf. Fig. 9). A broadened axial spectrum
was observed for a frozen solution of CuL–Cl at 77 K, with no clear
anisotropy and hyperfine splitting resolved (g = 2.11(1); spectrum
not shown).
As a result of the stoichioꢀmetric one-electron oxidation of CuL–
Cl with (4-BrC6H4)3N+SbCl6 the deep red colour of the parent
complex in dichloromethane turned dark green (see Fig. 5a). The
EPR spectrum of the solution containing oxidised [CuLÅ–Cl]+ (likely
stabilized by coordination of residual water) is completely silent,
pointing to the formation of the radical phenoxyl ligand antiferro-
magnetically coupled to the Cu(II) centre [63]. In situ anodic
spectro-electrochemistry with CuL–Cl in dichloromethane at
ꢀ40 °C led to the same result. The EPR signal of CuL–Cl, which
disappeared at the anodic wave O1, was fully recovered at the
Fig. 4. UV–Vis spectral changes accompanying the oxidation of CuL–Cl in DCM/
Bu4NPF6 to a phenoxyl radical species at the anodic potential O1. Conditions: Pt
minigrid anode, OTTLE cell, room temperature.
2.0
2.0
a
A
b
A
1.5
1.0
0.5
0.0
1.5
1.0
0.5
0.0
400
500
600
700
800
400
500
600
700
800
Wavelength, nm
Wavelength, nm
Fig. 5. UV–Vis spectral changes accompanying the stepwise one-electron oxidation of CuL–Cl to a phenoxyl radical species in dichloromethane (a) and acetonitrile (b), using
(4-BrC6H4)3N+SbCl6
.
ꢀ

P. Uma Maheswari et al. / Inorganica Chimica Acta 374 (2011) 406–414
411
cathodic wave R1 (Fig. 3). This result excludes any extensive
irreversible change in the chemical composition induced by the
oxidation of the parent compound, e.g., dissociation of the chloride
ligand. This behaviour is also in agreement with the magnetic sus-
ceptibility data presented in the following section.
compound is in a very good agreement with the available X-ray
data (see Table S1; Cartesian coordinates of the optimised complex
are given in Supplementary material).
In Table S2 (Supplementary information), we report orbital
energies and localisation of frontier molecular orbitals (MOs) of
CuL–Cl and other close-lying orbitals relevant for the assignment
of major electronic transitions in this complex (Table 2). The
3.5. Magnetic behaviour
ground state of CuL–Cl is a doublet with one excess
Fig. 9 shows the spin density obtained as the difference between
the and b electron densities. The HOMO and HOMO b (79
and 78b, respectively) of CuL–Cl are very close in energy and con-
sist of a orbital largely localised on the pyrimol ligand, in partic-
a electron.
Magnetic susceptibility measurements were carried out with a
a
a
a
crystalline sample of CuL–Cl at variable temperature. The
lue is 0.393 cm3 molꢀ1 K at room temperature (Fig. 7, circles),
which is close to the value for single free electron
MT = 0.375 cm3 molꢀ1 K, g = 2.0023). The value remains almost
vMT va-
p
a
ular on the phenolate oxygen and aromatic ring, and to some
(v
extent also on the imino nitrogen (see Fig. 9 and Table S2). The sin-
unchanged till very low temperatures (T ꢃ 50 K), where a small de-
crease is observed. These data are indicative of very weak antifer-
romagnetic interactions between the paramagnetic Cu(II) centres
in the lattice. On the other hand, the product obtained by the chem-
gly occupied r⁄ HOMO-1
a (78a) is about 1.0 eV lower in energy
than the HOMO and is largely localised on the Cu–Cl moiety, with
strong contributions from all three donor heteroatoms of the pyri-
mol ligand (Fig. 9 and Table S2).
ical one-electron oxidation of CuL–Cl with (4-BrC6H4)3N+SbCl6 in
ꢀ
As shown in Fig. 5, the UV–Vis spectra of CuL–Cl feature a dom-
inant lowest-energy absorption band at 566 nm in dichlorometh-
ane and 540 nm in acetonitrile. In order to rationalise the
solvatochromic shifts in these spectra and the nature of the corre-
sponding electronic transitions, we performed TD-DFT calculations
for CuL–Cl in vacuum and in the presence of a solvent treated
acetonitrile is diamagnetic (throughout the same temperature
range (5–300 K)), in agreement with the absent EPR signal (see
above). No remainders of any non-oxidised sample were observed.
Further analysis of [CuLÅ–Cl(MeCN)]+ was carried out by density
functional theory (DFT) calculations.
3.6. DFT and TDDFT study of CuL–Cl and one-electron-oxidised [CuLÅ–
Cl]+
Table 2
TD-DFT computed excitation energies
transitions (E.T.) in CuL–Cl.
E
and oscillator strengths
E.T.b
f
for electronic
Weightc (%)
DFT calculations were carried out to characterise the CuL–Cl
compound and its EPR silent oxidised state [CuLÅ–Cl]+, respectively
[CuLÅ–Cl(MeCN)]+. The geometrical structure of the CuL–Cl
E (eV) (k, nm)
fa
CuL–Cl
2.207 (562)
0.170
79
a
? 80
a
37
37
10
69
39
41
78b ? 79b
70b ? 80b
70b ? 80b
2.279 (544)
3.431 (361)
0.041
0.286
75
a ? 80a
75b ? 79b
a
Only the excitation energies with an oscillator strength larger than 0.01 are
reported.
b
Electronic transitions from occupied to virtual MOs. Note that in [CuL–Cl] there
are 79 electrons and 78b electrons.
a
c
For each excitation energy, the corresponding most relevant molecular orbital
transitions are indicated with the relative weight.
270
290
310
330
Field (mT)
350
370
Fig. 6. X-band EPR spectrum of [CuL–Cl] in dichloromethane at room temperature;
g = 2.107, Aiso = 8.25 mT. The corresponding simulated spectrum is shown in Fig. S4.
Fig. 8. TDDFT simulated UV–Vis absorption spectra of parent CuL–Cl in vacuum
(red, full line), in dichloromethane (green, dashed line), in acetonitrile (blue, dotted
line) and the one-electron oxidised species [CuLÅ–Cl]+ in dichloromethane (purple,
densely dotted line). (For interpretation of the references to colours in this figure
legend, the reader is referred to the web version of this paper.)
Fig. 7. The plots
vMT and 1/vM vs. T obtained for the compound [CuL–Cl].

412
P. Uma Maheswari et al. / Inorganica Chimica Acta 374 (2011) 406–414
Fig. 9. Molecular orbitals of CuL–Cl relevant for the calculated electronic transitions beyond the absorption peak at 562 nm, and the spin density delocalisation in the
complex.
within the PCM method. To simulate the absorption spectra we
have considered a superposition of Gaussian peaks centred at the
calculated excitation energies, with a height equal to the oscillator
strength and a width of 0.15 eV. In Table 2, we report the calcu-
lated excitation energies and the corresponding oscillator
strengths in vacuum.
The theoretically predicted UV–Vis spectrum of the CuL–Cl
compound in vacuum shows two peaks at 562 and 361 nm, in good
agreement with the experimental spectrum (see Figs. 4 and 8 for
comparison). The lowest-energy excitation at 562 nm arises pre-
at around 440 nm, consistently with the experiment. The main
electronic transitions responsible for this absorption have a mixed
p
(see Fig. S5 in Supplementary information). A residual band of low-
er intensity is also observed at 550 nm, differently from the exper-
imental spectrum. This might be an artefact of the TD-DFT
description and another argument for further theoretical
investigations.
?
p⁄(LÅ) intraligand and Cu–Cl ? LÅ charge transfer character
The UV–Vis spectrum of [CuLÅ–Cl]+ in the triplet state was also
evaluated in acetonitrile. At the same time we considered the axial
coordination of a solvent molecule in [CuLÅ–Cl(MeCN)]+, as indi-
cated by the spectro-electrochemical data. The calculated spectra
in these two cases are very similar to the spectrum in Fig. 8 (purple
line), showing no significant shift in the position or change in the
intensity of the absorption band at 440 nm.
The calculated ionisation potentials for CuL–Cl, CuL–Cl(MeCN),
and [CuL–MeCN]+ complexes reach the values of 7.49, 7.05, and
10.33 eV, respectively. Thus, the formation of the pentacoordinate
complex [CuLÅ–Cl(MeCN)]+ is further supported by the DFT results
since the anodic potential difference of ca. 0.45 eV between the
oxidations of CuL–Cl and short-lived CuL–Cl(MeCN) (see above)
corresponds well with the difference in the ionisation potential be-
tween the two complexes (0.44 eV) due to the axial acetonitrile
coordination. The formation of the oxidised species [CuLÅ–MeCN]2+
can be also discarded on the basis of the much higher computed
ionisation energy.
dominantly from the HOMO(
tion, which can be described as a
a
,b) to LUMO(
a,b) electronic transi-
p–
p⁄(pyrimol) intraligand or
phenolate-to-iminopyridine intramolecular charge transfer transi-
tion (see Table 2 and Fig. 9). The lowest-energy absorption band for
the parent complex CuL–Cl is predicted to shift to 544 and 535 nm
in the presence of the dichloromethane and acetonitrile solvents,
respectively. The shift to a higher energy in acetonitrile compared
to dichloromethane is consistent with the experimental data
shown in Fig. 5, indicating that TD-DFT with the implicit solvent
model is able to capture the observed solvatochromic shift (see
Fig. 8).
Upon oxidation of the CuL–Cl compound, one electron is ex-
pected to be removed from the HOMO, giving rise to a biradicaloid
state with two unpaired electrons (in the HOMO and HOMO-1),
which can be either ferromagnetically (triplet state) of antiferro-
magnetically (singlet state) coupled. According to the EPR analysis,
the oxidised state is EPR silent and characteristic of an antiferro-
magnetic coupling (see above). The triplet state would also con-
trast the diamagnetic nature of the oxidised species. However,
when performing the unrestricted DFT/B3LYP calculation on the
singlet state of the oxidised complex, the unpaired electron from
3.7. Oxidation of primary alcohols with CuL–Cl in aerated aqueous
acetonitrile
The oxidative activity of CuL–Cl was probed in water/acetoni-
trile at room temperature, using benzyl alcohol as a model sub-
the HOMO-1
a is removed, resulting in the formation of a cationic
singlet state, [CuL–Cl]+ (S = 0), with no biradical character. The fail-
ure of unrestricted DFT in this case is not surprising due to the
intrinsically single determinant nature of the DFT wave function:
the biradical singlet state of [CuLÅ–Cl]+ needs a multi-configuration
approach to be described properly, and further theoretical calcula-
tions on this issue are planned in the near future. Here, we look in-
stead at the triplet state of [CuLÅ–Cl]+, which, although not
corresponding with the experimentally observed state, carries
the correct biradical character (with the two unpaired electrons
having parallel spin) and can provide some insight into the nature
of the new visible electronic transition. In Fig. 8 we show the cal-
culated electronic absorption spectrum of [CuLÅ–Cl]+ in dichloro-
methane in the triplet state. A new absorption band is emerging
t
strate. The addition of BuOK has been essential for the alcohol
activation. Other bases like NaOH and KOH can also be employed.
The base has to be added after the alcohol in the exact stoichiom-
etric amount and not in an excess that retards the reaction. The
complex CuL–Cl enters the reaction mixture after the base to avoid
a decrease of the catalyst efficiency. A similar behaviour for the
GOase was noted by Hamilton et al. [64]. The activity of the GOase
increased 20–30 times in the presence of oxidants such as hexac-
yanidoferrate(III), hexachloridoiridate(IV) or a Mn(III) edta com-
plex. Conversely, the alcohol oxidation was completely inhibited
in the presence of hexacyanidoferrate [64]. The reaction also did
not occur under argon, while benzaldehyde was smoothly pro-
duced in air. The reaction rate further increased after air had been

P. Uma Maheswari et al. / Inorganica Chimica Acta 374 (2011) 406–414
413
Table 3
CuL–Cl mediated oxidation of selected primary alcohols under stoichiometric
conditions.a
Entry
Alcohol
Conv.b (%)
Conv.c (%)
Conv.d (%)
1
2
3
4
Benzyl alcohol
Crotyl alcohol
Geraniol
4-Hydroxy-3-methoxy-
benzyl alcohol
1-Octanol
15
18
32
7
45
56
78
23
92
90
95
65
5
21
56
87
a
In the presence of 2 mmol of primary alcohol activated by 2 mmol of tBuOK and
2 mmol of CuL–Cl in CH3CN/H2O at room temperature. Reaction conditions are
given in Section 2.
b
Products analysed after 1 h.
Products analysed after 3 h.
Products analysed after 8 h.
Fig. 10. Schematic representation of the active site of the GOase enzyme [11].
c
d
a reliable theoretical description of its electronic structure is cur-
rently not available (see above).
replaced by neat dioxygen. Using these conditions, with CuL–Cl we
could oxidised a series of primary alcohols stoichiometrically to
the corresponding aldehydes (Table 3). No reactivity of CuL–Cl
towards secondary alcohols was observed.
The Cu(II) centre in CuL–Cl is coordinated by a phenolate
moiety (part of the tridendate pyrimol ligand) which closely
resembles the Tyr272 moiety in the GOase. The pyrimol ligand
is coordinated to Cu(II) in a nearly perfect square-planar envi-
ronment and the bond distances in the plane match closely to
those found at the GOase active site. In particular, the Cu–O1
distance in the equatorial plane is 1.94 Å, both in CuL–Cl and
the GOase [67].
In the GOase, Tyr272, His496, His581 and the acetate ion
from the crystallization buffer also form an almost perfect
square. A fifth ligand, i.e. Tyr495, occupies the axial position at
a longer distance than Tyr272 (Fig. 10). The four-coordinate
Cu(II)–pyrimol compound features a strong rigidity brought in
by the tridentate pyrimol ligand. Pyrimol forms two five-
membered chelate rings, Cu1–N2–C21–C22–O1 and Cu1–N2–
C1–C11–N1, thereby stabilizing the square-based geometry and
preventing large structural changes induced by the oxidation of
the phenolate moiety (as suggested by the spectro-electrochem-
ical and computational studies). The rigid square-planar coordi-
nation is most likely crucial for the activity, being ensured by
the protein backbone for the natural enzyme. In the
[CuLÅ–Cl(MeCN)]+ coordination sphere, the chloride ligand may
act as a facile leaving group in the presence of alcoholate. The
likely solvent-assisted replacement of the chloride in the equato-
rial plane by the deprotonated alcoholate species is favourable
for a five-membered cyclic transition state (Fig. 11) and for the
When different copper(II) salts and Hpyrimol or Hpyramol are
used in situ as catalyst precursors (see Section 2), the rate increases
in the following order: CuBr2 > CuCl2 > Cu(ClO4)2 > Cu(BF4)2 > Cu(-
NO3)2. It is noteworthy that the oxidation catalyst formed in situ
from CuCl2 and Hpyramol/Hpyrimol gives results comparable to
those obtained with the crystallized sample of CuL–Cl. When car-
rying out the oxidations at different temperatures, the reactivity
increases in the order: 60 °C > 40 °C > 20 °C. At 90 °C, however,
the reaction is slower than at 20 °C, indicating some instability of
the catalyst above 60 °C. Good to excellent conversions were
achieved with the various primary alcohols used which were
deprotonated with tBuOK initially for their activation (Table 3).
The active benzyl alcohol, crotyl alcohol and geraniol (entries
1–3) were almost completely converted to the corresponding
aldehydes. The lower conversion reached with 4-hydroxy-3-meth-
oxy-benzyl alcohol (entry 4) is most likely due to the deactivating
character of the p-hydroxy substituent; nevertheless, 65% of
vanillin resulted after a reaction time of 8 h. The selective oxida-
tion of 1-octanol to 1-octanal is remarkable. Indeed, non-activated
linear alkyl alcohols are known to be difficult to oxidise [65]. With
CuL–Cl, a conversion of 87% was obtained (entry 5), which was
comparable to the conversions achieved for the activated alcohols
in entries 1–3. Moreover, no traces of the overoxidised octanoic
acid were detected by gas chromatography.
facile
a-hydrogen abstraction as the key step in the alcohol oxi-
dation, in contrast to alcoholate attachment in the axial position
[68]. Importantly, the alcoholate oxidation is faster when CuBr2
is used instead of CuCl2 (see above), which provides an indirect
evidence that the halide ligand indeed acts as a good leaving
group for the alcoholate substrate.
3.8. CuL–Cl as a functional model of the GOase
The active site of the galactose oxidase (Fig. 10) is rather unu-
sual, as the coordinated phenolate ligands (Tyr272 and Tyr495)
together with a copper ion are involved in two-electron transfer
catalysis. The modelling of this active site requires assembling a
variety of functional groups in a specific geometric arrangement.
The neutral complex CuL–Cl and its oxidised form can be consid-
ered as models of the native enzyme GOase with regard to the
coordination environment of Cu(II). Similar to the GOase, the
oxidation of the pyrimol ligand in CuL–Cl leads to a biradical
complex in a singlet ground state (S = 0). The comparison of
the antiferromagnetic coupling and stabilization of the singlet
ground state (S = 0) in the oxidised GOase and [CuLÅ–Cl(solvent)]+
would require qualitative analysis of the relative orientations
and overlap of the magnetic orbitals of the Cu(II) centre and
the phenoxyl radical LÅ according to the Goodenough-Kanamori
rules for exchange coupling [66]. Unfortunately, the crystal
structure of [CuLÅ–Cl(solvent)]+ could not be determined and
the optimised geometry of the singlet oxidised state based on
N2O donor pyrimol ligand
N
.
O
H
Cu(II)
Coordinating
N
primary
alcoholate
O
α
C
R
CH3-CN
H
Fig. 11. The five-membered cyclic transition state involving [CuLÅ–Cl(MeCN)]+,
proposed for the -hydrogen abstraction during primary alcohol oxidation.
a

414
P. Uma Maheswari et al. / Inorganica Chimica Acta 374 (2011) 406–414
[13] N. Ito, S.E.V. Phillips, C. Stevens, Z.B. Ogel, M.J. McPherson, J.N. Keen, K.D.S.
4. Conclusions
Yadav, P.F. Knowles, Nature 350 (1991) 87.
[14] J. Stubbe, W.A. van der Donk, Chem. Rev. 98 (1998) 705.
[15] M.M. Whittaker, D.P. Ballou, J.W. Whittaker, Biochemistry 37 (1998) 8426.
[16] M.M. Whittaker, J.W. Whittaker, J. Biol. Chem. 263 (1988) 6074.
[17] S.G. Minasian, M.M. Whittaker, J.W. Whittaker, Biochemistry 43 (2004) 13683.
[18] M.M. Whittaker, J.W. Whittaker, Biochemistry 40 (2001) 7140.
[19] E. Bill, J. Muller, T. Weyhermuller, K. Wieghardt, Inorg. Chem. 38 (1999) 5795.
[20] F. Michel, F. Thomas, S. Hamman, E. Saint-Aman, C. Bucher, J.L. Pierre, Chem.
Eur. J. 10 (2004) 4115.
[21] R.C. Pratt, T.D.P. Stack, Inorg. Chem. 44 (2005) 2367.
[22] J.G. Radziszewski, M. Gil, A. Gorski, J. Spanget-Larsen, J. Waluk, B.J. Mroz, J.
Chem. Phys. 115 (2001) 9733.
[23] I. Sylvestre, J. Wolowska, C.A. Kilner, E.J.L. McInnes, M.A. Halcrow, Dalton
Trans. (2005) 3241.
The copper(II)–pyrimol compound (CuL–Cl) is capable of oxidis-
ing a series of primary alcohols to the corresponding aldehydes,
under stoichiometric condition. The presence of dioxygen and
tBuOK is essential for the oxidation process. The rigid square-
planar coordination, owing to the double-chelating conjugated
pyrimol ligand, appears to stabilize the singlet ground state
(S = 0), on the phenoxyl radical formation. This geometry is also
expected to assist the a-hydrogen abstraction from the alcoholate
substrate through a five-membered cyclic transition state. The re-
sults of the spectro-electrochemical studies in dichloromethane
and acetonitrile are indicative of a preserved chemical composition
and square-planar geometry of the CuL–Cl moiety after the
oxidation of the phenolate group, with a likely axial coordination
of the donor solvent (acetonitrile, water). DFT and time-dependent
DFT calculations provide a valuable insight into the molecular
orbitals involved in the intraligand (L) and charge-transfer visible
electronic transitions in CuL–Cl and its oxidised state.
The CuL–Cl complex is an effective bioinspired example of the
aerobic oxidation of primary alcohols, bearing close resemblance
to the native GOase enzyme. Functionally, some differences exist
due to the rigid planar coordination of the single pyrimol ligand
at Cu(II) and the flexible coordination of the separate amino acids
in the innate enzyme. The likely catalytic activity of CuL–Cl
presents a challenge for future studies.
[24] M. Taki, H. Hattori, T. Osako, S. Nagatomo, M. Shiro, T. Kitagawa, S. Itoh, Inorg.
Chim. Acta 357 (2004) 3369.
[25] R. Uma, R. Viswanathan, M. Palaniandavar, M. Lakshminarayanan, J. Chem.
Soc., Dalton Trans. (1994) 1219.
[26] M. Vaidyanathan, M. Palaniandavar, R.S. Gopalan, Indian J. Chem., Sect. A:
Inorg. Bioinorg. Phys. Theor. Anal. Chem. 42 (2003) 2210.
[27] C. Zondervan, R. Hage, B.L. Feringa, Chem. Commun. (1997) 419.
[28] P. Milko, J. Roithova, N. Tsierkezos, D. Schroder, J. Am. Chem. Soc. 130 (2008)
7186.
[29] D. Rokhsana, D.M. Dooley, R.K. Szilagyi, J. Biol. Inorg. Chem. 13 (2008) 371.
[30] J. Manzur, H. Mora, A. Vega, D. Venegas-Yazigi, M.A. Novak, J.R. Sabino, V.
Paredes-Garcia, E. Spodine, Inorg. Chem. 48 (2009) 8845.
[31] N. Kitajima, K. Whang, Y. Morooka, A. Uchida, Y. Sasada, J. Chem. Soc., Chem.
Commun. (1986) 1504.
[32] Y.D. Wang, T.D.P. Stack, J. Am. Chem. Soc. 118 (1996) 13097.
[33] P. Chaudhuri, M. Hess, U. Florke, K. Wieghardt, Angew. Chem., Int. Ed. 37
(1998) 2217.
[34] P. Chaudhuri, M. Hess, J. Muller, K. Hildenbrand, E. Bill, T. Weyhermuller, K.
Wieghardt, J. Am. Chem. Soc. 121 (1999) 9599.
[35] J.S. Panek, C.E. Masse, Angew. Chem., Int. Ed. 38 (1999) 1093.
[36] Y.D. Wang, J.L. DuBois, B. Hedman, K.O. Hodgson, T.D.P. Stack, Science 279
(1998) 537.
Acknowledgements
[37] R. Behrendt, C. Renner, M. Schenk, F.Q. Wang, J. Wachtveitl, D. Oesterhelt, L.
Moroder, Angew. Chem., Int. Ed. 38 (1999) 2771.
[38] S. Itoh, S. Takayama, R. Arakawa, A. Furuta, M. Komatsu, A. Ishida, S. Takamuku,
S. Fukuzumi, Inorg. Chem. 36 (1997) 1407.
[39] A. Philibert, F. Thomas, C. Philouze, S. Hamman, E. Saint-Aman, J.L. Pierre,
Chem. Eur. J. 9 (2003) 3803.
[40] E. Zueva, P.H. Walton, J.E. McGrady, Dalton Trans. (2006) 159.
[41] P.U. Maheswari, S. Roy, H. den Dulk, S. Barends, G. van Wezel, B. Kozlevcar, P.
Gamez, J. Reedijk, J. Am. Chem. Soc. 128 (2006) 710.
[42] L.D. Pachon, A. Golobic, B. Kozlevcar, P. Gamez, H. Kooijman, A.L. Spek, J.
Reedijk, Inorg. Chim. Acta 357 (2004) 3697.
[43] Y.L. Wong, D.K.P. Ng, H.K. Lee, Inorg. Chem. 41 (2002) 5276.
[44] A.J.M. Duisenberg, L.M.J. Kroon-Batenburg, A.M.M. Schreurs, J. Appl.
Crystallogr. 36 (2003) 220.
This research was financially supported by the Dutch Economy,
Ecology and Technology (EET) programme, a joint programme of
the Ministry of Economic Affairs, the Ministry of Education, Culture
and Science, and the Ministry of Housing, Spatial Planning and the
Environment. M.L. and A.L.S. acknowledge financial support of the
Netherlands Organization for Scientific Research (NWO-CW). Dr.
Paul Alsters (DSM, Pharmaceuticals) and Dr. Ronald Hage (Unilever)
are gratefully acknowledged for fruitful discussions. Dr. Roberta
Pievo and Dr. Ilaria Gamba are thanked for recording and reproduc-
ing the EPR spectrum. The use of supercomputer facilities was
sponsored by the Stichting Nationale Computerfaciliteiten (NCF),
with a support from NWO-CW.
[45] G.M. Sheldrick, Acta Crystallogr., Sect. A 64 (2008) 112.
[46] A.L. Spek, J. Appl. Crystallogr. 36 (2003) 7.
[47] G.M. Sheldrick, University of Göttingen, Germany, 1999.
[48] P.T. Beurskens, G. Admiraal, G. Beurskens, W.P. Bosman, S. Garcia-Granda, R.O.
Gould, J.M.M. Smits, C. Smykalla, The DIRDIF Program System, Technical Report of
the Crystallography Laboratory, University of Nijmegen, The Netherlands, 1999.
[49] R. Herbst-Irmer, G.M. Sheldrick, Acta Crystallogr., Sect. B 54 (1998) 443.
[50] R.C. Pratt, T.D.P. Stack, J. Am. Chem. Soc. 125 (2003) 8716.
[51] E.A. Boudreaux, L.N. Mulay, Theory and Applications of Molecular Magnetism,
John Wiley & Sons, New York, 1976.
Appendix A. Supplementary material
CCDC 629865 and 629866 contain the supplementary crystallo-
graphic data for this paper. These data can be obtained free of
charge from The Cambridge Crystallographic Data Centre via
www.ccdc.cam.ac.uk/data_request/cif. Supplementary data associ-
ated with this article can be found, in the online version, at
doi:10.1016/j.ica.2011.03.029.
[52] F. Hartl, R.P. Groenestein, T. Mahabiersing, Collect. Czech. Chem. Commun. 66
(2001) 52.
ˇ
ˇ
[53] M. Krejcík, M. Danek, F. Hartl, J. Electroanal. Chem. 317 (1991) 179.
[54] T. Mahabiersing, H. Luyten, R.C. Nieuwendam, F.E. Hartl, Collect. Czech. Chem.
Commun. 68 (2003) 1687.
[55] F. Hartl, H. Luyten, H.A. Nieuwenhuis, G.C. Schoemaker, Appl. Spectrosc. 48
(1994) 1522.
[56] J.A. Pople, GAUSSIAN 03, Revision C.02, Gaussian, Inc., Wallingford, CT, 2004.
[57] P.J. Stephens, F.J. Devlin, C.F. Chabalowski, M.J. Frisch, J. Phys. Chem. 98 (1994)
11623.
[58] A.D. Becke, J. Chem. Phys. 98 (1993) 5848.
[59] J. Tomasi, B. Mennucci, R. Cammi, Chem. Rev. 105 (2005) 2999.
[60] F. Hartl, Inorg. Chim. Acta 232 (1995) 99.
[61] J.A. Halfen, V.G. Young, W.B. Tolman, Angew. Chem., Int. Ed. 35 (1996) 1687.
[62] R. van Gorkum, E. Bouwman, J. Reedijk, 43 (2004) 2456.
[63] L. Benisvy, E. Bill, A.J. Blake, D. Collison, E.S. Davies, C.D. Garner, G. McArdle,
E.J.L. McInnes, J. McMaster, S.H.K. Ross, C. Wilson, Dalton Trans. (2006) 258.
[64] G.A. Hamilton, P.K. Adolf, J. Dejersey, G.C. Dubois, G.R. Dyrkacz, R.D. Libby, J.
Am. Chem. Soc. 100 (1978) 1899.
[65] D. Geisslmeier, W.G. Jary, H. Falk, Monatsh. Chem. 136 (2005) 1591.
[66] J. Muller, T. Weyhermuller, E. Bill, P. Hildebrandt, L. Ould-Moussa, T. Glaser, K.
Wieghardt, Angew. Chem., Int. Ed. 37 (1998) 616.
References
[1] C.D. Borman, C.G. Saysell, A. Sokolowski, M.B. Twitchett, C. Wright, A.G. Sykes,
Coord. Chem. Rev. 192 (1999) 771.
[2] P. Chaudhuri, K. Wieghardt, Prog. Inorg. Chem. 50 (2001) 151.
[3] P. Chaudhuri, K. Wieghardt, T. Weyhermuller, T.K. Paine, S. Mukherjee, C.
Mukherjee, Biol. Chem. 386 (2005) 1023.
[4] S. Itoh, M. Taki, S. Fukuzumi, Coord. Chem. Rev. 198 (2000) 3.
[5] B.A. Jazdzewski, W.B. Tolman, Coord. Chem. Rev. 200 (2000) 633.
[6] F. Thomas, Eur. J. Inorg. Chem. (2007) 2379.
[7] S.K. Alamsetti, S. Mannam, P. Mutupandi, G. Sekar, Chem. Eur. J. 15 (2009) 1086.
[8] A. John, M.M. Shaikh, P. Ghosh, Dalton Trans. (2008) 2815.
[9] L. Que, W.B. Tolman, Nature 455 (2008) 333.
[10] C. Mukherjee, U. Pieper, E. Bothe, V. Bachler, E. Bill, T. Weyhermuller, P.
Chaudhuri, Inorg. Chem. 47 (2008) 8943.
[11] J.W. Whittaker, Chem. Rev. 103 (2003) 2347.
[12] B.P. Branchaud, M.P. Montaguesmith, D.J. Kosman, F.R. McLaren, J. Am. Chem.
Soc. 115 (1993) 798.
[67] N. Ito, S.E. Phillips, K.D. Yadav, P.F. Knowles, J. Mol. Biol. 238 (1994) 794.
[68] A. Kohen, J.P. Klinman, Chem. Biol. 6 (1999) R191.