Selective oxidation of Mandelic acids catalyzed by copper (II) complexes

Article abstract of DOI:10.1016/j.molcata.2013.09.004

Two copper (II) complexes L¹Cu and L²Cu (H ₂L¹ = 4,9-dimethyl-2,11-dioxy-5,8-diazacycldode-2,4,8,10- tetraene; H₂L² = diethyl-3,8-dimethyl-4,7-diazadeca-2,8- diene-1,10-dioat) were prepared and displayed good catalytic efficiency for the oxidations of Mandelic acid (MA), 4-hydroxymandelic acid (4-HMA) and 4-methoxymandelic acid (4-MMA) by H₂O₂ in aqueous solution at 35 C. The apparent first-order rate constants k_obs of oxidations of MA, 4-HMA and 4-MMA were obtained in the range of pH 2.5-4.5. It was great interesting that the complexes LCu displayed excellent catalytic ability for the selective oxidation of Mandelic acids into aromatic aldehydes (S > 98%). The active species generated by reacting copper (II) complexes LCu with H ₂O₂ was characterized by UV-vis, electrospray ionization mass spectrometry (ESI-MS) and cyclic voltammetry (CV). The LCu^I-OOH species was suggested as the possible active species. The geometric configurations of LCu and the LCu^I-OOH species were calculated with Density Functional Theory (DFT) method. The LCu^I-OOH species is stable and possesses relatively high oxidation potential in aqueous solution at ambient temperature. This study also provides a new insight for the degradation of lignin.

Full text of DOI:10.1016/j.molcata.2013.09.004

Journal of Molecular Catalysis A: Chemical 379 (2013) 315–321
Contents lists available at ScienceDirect
Journal of Molecular Catalysis A: Chemical
journal homepage: www.elsevier.com/locate/molcata
Selective oxidation of Mandelic acids catalyzed by copper (II)
complexes
Wei-Feng Yu, Xiang-Guang Meng∗, Xiao Peng, Xiao-Hong Li, Ying Liu
Key Laboratory of Green Chemistry and Technology, Ministry of Education, College of Chemistry, Sichuan University, Chengdu 610064, PR China
a r t i c l e i n f o
a b s t r a c t
Two copper (II) complexes L1Cu and L2Cu (H2L1 = 4,9-dimethyl-2,11-dioxy-5,8-diazacycldode-2,4,8,10-
tetraene; H2L2 = diethyl-3,8-dimethyl-4,7-diazadeca-2,8-diene-1,10-dioat) were prepared and displayed
good catalytic efficiency for the oxidations of Mandelic acid (MA), 4-hydroxymandelic acid (4-HMA) and
4-methoxymandelic acid (4-MMA) by H2O2 in aqueous solution at 35 ◦C. The apparent first-order rate
constants kobs of oxidations of MA, 4-HMA and 4-MMA were obtained in the range of pH 2.5–4.5. It was
great interesting that the complexes LCu displayed excellent catalytic ability for the selective oxidation
of Mandelic acids into aromatic aldehydes (S > 98%). The active species generated by reacting copper (II)
complexes LCu with H2O2 was characterized by UV–vis, electrospray ionization mass spectrometry (ESI-
MS) and cyclic voltammetry (CV). The LCuI-•OOH species was suggested as the possible active species.
The geometric configurations of LCu and the LCuI-•OOH species were calculated with Density Functional
Theory (DFT) method. The LCuI-•OOH species is stable and possesses relatively high oxidation potential
in aqueous solution at ambient temperature. This study also provides a new insight for the degradation
of lignin.
Article history:
Received 20 March 2013
Received in revised form 24 August 2013
Accepted 4 September 2013
Available online 12 September 2013
Keywords:
Hydroperoxo free radical
Lignin
Mandelic acids
Selective oxidation
DFT
© 2013 Elsevier B.V. All rights reserved.
1. Introduction
species were generally studied in organic solvent at low tempera-
ture, owing to the instability and complexity in aqueous solution at
Metal-oxo species are very important active intermediates in
biological and biomimetic oxidation reactions [1–4]. In many bio-
logical metabolic processes, the generation and long-time stability
of the copper-hydroperoxo species are extremely important for
those enzyme catalytic oxidation reactions [5–8]. However, the
knowledge about the generation, properties and catalysis mech-
anism of the copper-hydroperoxo species is still not enough. In last
decade, some copper-oxo species derived from the oxidation states
of copper complex oxidized by O2 or H2O2, such as CuII-O2−, CuII-O•
and CuII-OOH species, were widely studied [9–12]. Recently Comba
et al. found a new mononuclear bispidine copper-hydroperoxo
species and investigated its some spectroscopic properties [13].
Choi et al. reported the synthesis and properties of copper-
hydroperoxo species bearing a Me6-tren ligand and found it was
inactive for electrophilic and nucleophilic oxidation reactions [14].
Fujii et al. found that the reactivity of copper-hydroperoxo species
with N3O-type ligand was greatly influenced by their geometries,
the reactivity of the species with square planar geometry was
apparently larger than that of the species with the trigonal bipyra-
midal configuration [15]. In previous reports, metal-hydroperoxo
ambient temperature [16–19]. Sch•e−petkin et al. reported that sev-
eral free radical species •OOH, O2
and •OH could be formed in
the aqueous solution containing copper (II) complex and hydro-
gen peroxide [20]. Lin and Wu suggested that the hydroperoxo
free radical associates with the center copper ion to form copper-
peroxo or copper-hydroperoxo species [21]. Some researchers also
found that the formation of copper-hydroperoxo species could be
influenced by many factors including catalysts, pH of the solution,
solvent and temperature [22–24]. However until now, under what
condition the copper-hydroperoxo species can be generated pri-
marily is still unclear. In previous studies [25], we found that the
“copper complex associative radical” LCuI-•OOH may be the pre-
dominant oxidation species which is the oxidation state of copper
complex oxidized by H2O2 in buffer solution. In present work, this
kind of LCuI-•OOH species could be easily generated by the reac-
tion of copper complex with H2O2 in tartaric acid buffer solution
at ambient temperature. Some properties of this species were also
investigated in detail.
Mandelic acids are widely used as typical model substrate of
lignin degradation enzymes [26–28]. The efficient degradation of
lignin and its separation from the lignocellulose are challenging
problems encountered in biofuel production, lignin contains lots of
units of –OH and/or –OCH3 substituted aromatic secondary alco-
hols [29,30]. Recently we adopted a new route to degrade lignin by
∗
Corresponding author. Tel.: +86 28 85462979; fax: +86 28 85412907.
E-mail address: mengxgchem@163.com (X.-G. Meng).
1381-1169/$ – see front matter © 2013 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.molcata.2013.09.004

316
W.-F. Yu et al. / Journal of Molecular Catalysis A: Chemical 379 (2013) 315–321
O
selective oxidations of aromatic secondary alcohols and aromatic
ketones under mild condition [31,32]. In present work, we found
that the “copper complex associative radical” LCuI-•OOH species
could be readily generated and exhibited an excellent selectivity
for the oxidation of Mandelic acids by H2O2.
HO
O
H
OH
H2O2 / LCu
H+
R
R
2. Experimental
R: H, OH, OCH3
2.1. Materials
Scheme 1. Selective catalytic oxidation of Mandelic acids to aromatic aldehyde.
Mandelic acid (MA), 4-hydroxymandelic acid (4-HMA), 4-
methoxymandelic acid (4-MMA), 30% hydrogen peroxide,
benzaldehyde, 4-hydroxybenzaldehyde, anisaldehyde, ethyl
3-oxobutanoate, pentane-2,4-dione, ethane-1,2-diamine, sodium
hydroxide, tartaric acid were all analytical grade and purchased
from the commercial sources. All other chemicals and solvents
used were also analytical grade without further purification.
1.00
0.75
0.50
0.25
0.00
2.2. Synthesis of ligands and preparation of copper (II) complexes
The ligand H2L1 (4,9-dimethyl-2,11-dioxy-5,8-diazacy-cldode-
2,4,8,10-tetraene) C12H20N2O2 and H2L2 (diethyl-3,8-dimethyl-
4,7-diazadeca-2,8-diene-1,10-dioat) C14H24N2O4 were synthe-
sized according to literatures [33,34].
The copper (II) complex L1Cu and L2Cu were prepared according
to the reported methods [33,35,36]. The data of ligands and copper
complexes are listed in Table 1
0
20
60
80
100
40 t/ min
2.3. Instrumentations and methods
Fig. 1. The generation of anisaldehyde catalyzed by LCu at pH 3 and 35 ◦C; C0(4-
MMA) = 1 mmol L−1, C0(H2O2) = 0.03 mol L−1, C0(LCu) = 0.05 mmol L−1. (ꢀ) L1Cu, (ꢀ)
L2Cu, (ꢁ) without catalyst.
Ultraviolet–visible absorbance measurements were performed
with a Hitachi U-2910 spectrophotometer (Tokyo, Japan). Elec-
trochemical measurements were performed using a CHI660C
electrochemical analyzer. All cyclic voltammetry (CV) measure-
ments were carried out with a three-electrode-system with a small
glassy carbon as working electrode, a Pt wire counter electrode
and a KCl saturated calomel electrode reference electrode all being
placed into the solution. 1H NMR spectra were recorded on a Bruker
spectrometer, using CDCl3 as an internal standard. Elemental anal-
yses were implemented with a Carlo Erba 1106 instrument. The
FT-IR spectrum was measured by Fourier transform infrared spec-
trometer (FT-IR, NEXUS 6700, USA).
2.4. Computation
All Density Functional Theory (DFT) calculations were per-
formed with the Gaussian 03 programs. Geometry optimizations
were carried out at the >B3LYP/6-311++G(d, p), SDD level.
3. Results and discussion
The typical reaction solution containing 0.05 mmol L−1 copper
(II) complex, 0.03 mol L−1 hydrogen peroxide and 0.001 mol L−1
Mandelic acids were kept at constant 35 ◦C in the tartaric acid buffer
solution. Test samples were extracted from the reactor periodically.
The concentrations of benzaldehyde (ε300 = 652 L mol−1 cm−1), 4-
hydroxybenzaldehyde (ε300 = 10 000 L mol−1 cm−1) and anisalde-
hyde (ε300 = 9500 L mol−1 cm−1) in the reaction solution were
determined by UV–vis analysis at wavelength 300 nm. MA, 4-HMA
and 4-MMA had no spectral absorbance contribution at 300 nm.
The mixture reaction solution, extracted with CH2Cl2 and dried
over Na2SO4, was analyzed by GC (GC9790, Fu li, China), GC–MS
(Agilent 5973 Network 6890 N) and HPLC (waters 1525) with a
UV–vis detector. GC measure conditions: KB-5 capillary column;
carrier gas, N2; gasification temperature, 300 ◦C; the column tem-
perature is programmed to keep at 60 ◦C for 2 min, then rise to
60–300 ◦C (25 ◦C/min); injection volume, 0.2 ␮L. GC–MS measure
3.1. Selective oxidation of Mandelic acids catalyzed by complexes
LCu
Lignin is one main component in lignocellulose. Up to now, its
efficient degradation is still a challenging problem in the utiliza-
tion of biomass [29,30]. MA, 4-HMA, 4-MMA, which are aromatic
secondary alcohols, were widely used as typical model substrates
of lignin [26–28]. In present work, we employed the copper com-
plexes L1Cu and L2Cu to catalyze the oxidation of Mandelic acids by
H2O2 (Scheme 1). Fig. 1 showed the formation process of anisalde-
hyde catalyzed by complexes LCu. From Fig. 1, it could be seen that
the oxidation reaction of 4-MMA with H2O2 was hardly occurred,
however the presence of catalyst LCu rapidly accelerated the reac-
tion. The mixture reaction solution extracted with CH2Cl2 and dried
by Na2SO4. The product of the reaction was analyzed by GC–MS
(see Fig. S2) and HPLC (Fig. 2). The analysis results indicated that
anisaldehyde was the only detectable product, the selectivity of
anisaldehyde was more than 98%. The oxidation reactions of other
two Mandelic acids (MA and 4-HMA) with H2O2 catalyzed by LCu
displayed the same characteristics as that of 4-MMA. The benzalde-
hyde or 4-hydroxybenzaldehyde is the sole detectable product.
The selectivity was more than 98% (the error was evaluated less
than 2%).
conditions: DB-5 capillary column; carrier gas, N2: 60 cm3 min−1
,
He: 2.0 cm3 min−1; gasification temperature, 300 ◦C; the column
temperature is programmed to keep at 40 ◦C for 4 min, then rise
to 40–260 ◦C (15 ◦C/min); injection volume, 0.1 ␮L. HPLC measure
conditions: C18 column; mobile phase, methanol/water (V/V = 1/1),
0.5 ml min−1; column temperature, 30 ◦C; injection volume, 20 ␮L;
UV detection wavelength, 280 nm.

W.-F. Yu et al. / Journal of Molecular Catalysis A: Chemical 379 (2013) 315–321
317
Table 1
Color, Mp, 1H NMR, IR data, ICP-AES data, elemental analyses of the ligands and copper complexes.
Compound
Color
Mp (◦C)
1H NMR, ı
IR data (KBr, mmax/cm−1
)
ICP-AES % Cu
–
Elemental analyses %
Found (calculated)
C H N
H2L1
Light yellow crystal
113 ◦
132 ◦
C
C
1.89 (s, 6H, CH3),
2.01 (s, 6H, CH3),
3.42 (d, 4H, CH2),
4.99 (s, 2H, CH),
10.99 (s, 2H, NH).
1.26 (t, 6H, CH3),
1.93 (s, 6H, CH3),
3.37 (m, 4H, CH2),
4.12 (q, 4H, CH2),
4.50 (s, 2H, CH),
8.66 (s, 2H, NH).
–
738 m, 851 m, 1288 vs, 1516 vs,
1577 vs, 1610 vs, 2948 m, 2996 m,
3083 m, 3159 m, 3450 br.
64.30 8.96 12.45
(64.26 8.99 12.49)
H2L2
Colorless crystal
724m, 784 s, 1169 s, 1289 vs, 1605
vs, 1648 vs, 2897 m, 2952 s, 2986 s,
3296 s.
–
59.09 8.48 9.90
(59.13 8.51 9.85)
L1Cu
L2Cu
Green
Purple
145 ◦
158 ◦
C
C
453 w, 780 m, 1021 w, 1275 w,
1405 s, 1530 s, 1578 vs, 3165 br,
3444 br.
488 m, 769 m, 1069 m, 1290 s,
1425 s, 1468 vs, 1508 vs, 1583 vs,
3168 br, 3450 br.
22.24
18.45
50.53 6.40 9.74
(50.42 6.35 9.80)
–
48.68 6.32 8.06
(48.61 6.41 8.10)
IR data, br: broad, s: strong, vs: very strong, m: medium, w: weak.
2.00
O
H
1.50
O
HO
OCH
OH
1.00
0.50
OCH
0.00
0.00 2.00 4.00 6.00 8.00 10.00 12.00 14.00 16.00 18.00 20.00 22.00 24.00 26.00 28.00 30.00 32.00 34.00 36.00 38.00 40.00 42.00
Minutes
Fig. 2. HPLC profile of the reaction solution after 2 h.
In order to investigate the reaction kinetics, we employed exces-
sive concentration of H2O2 over that of 4-MMA to carry out the
reaction. The apparent first-order formation rate constants kobs
were calculated based on the plots of ln[(A∞ − A0/A∞ − At)] vs
reaction time t (Fig. 3), where A0 and At were the absorbance of
anisaldehyde were listed in Table 2.The initial benzaldehyde, 4-
hydroxybenzaldehyde and anisaldehyde formation rates kobs were
affected by pH values of solution. From Table 2, it can be seen
that the initial aromatic aldehyde formation rate constants kobs
increased with decreasing pH. The pH effect may attribute to both
protonation of substrate Mandelic acids and generation of active
species under acidic aqueous solution, as illustrated in Scheme 2.
We also investigated the oxidation reaction in neutral and weak
alkaline solutions, when pH > 8 several active species including
superoxide anion free radical, hydroperoxo free radical, hydroxide
radical could be generated simultaneously in the aqueous solution
and the reaction products became complex.
reaction product anisaldehyde at time 0 and t respectively. A was
∞
the absorbance of anisaldehyde at the same concentration with
that of C0(4-MMA). From Fig. 3 it can be seen that the plots of
ln[(A∞ − A0)/(A∞ − At)] vs time t were all straight lines and approx-
imately passed the origin (the correlative coefficients r > 0.995),
which indicated this was a first-order reaction with respect to the
concentrations of 4-MMA for the initial reaction time 15 min. In
this work, in order to avoid the influences of the decomposition of
H2O2 on reaction, the experimental data for the initial 15 min reac-
tion time were employed to calculate the anisaldehyde formation
rate constants. The calculated apparent first-order rate constants
kobs of formation of benzaldehyde, 4-hydroxybenzaldehyde and
3.2. The generation and solution properties of active species
The active copper-oxo species played an important role in cat-
alytic oxidation reaction. In order to obtain the information of the
Table 2
Apparent first-order rate constants kobs of formation of aromatic aldehyde at 35 ◦C.a
pH
L1Cu
L2Cu
MA
4-HMA
4-MMA
MA
4-HMA
4-MMA
104kobs (s−1
)
104kobs (s−1
)
2.5
3.0
3.5
4.0
4.5
2.91
2.16
1.63
1.45
1.12
5.72
4.07
3.27
2.56
2.08
11.7
8.35
6.69
5.48
4.57
0.85
0.66
0.42
0.29
0.23
2.18
1.75
1.08
0.78
0.65
3.86
2.89
2.03
1.36
1.02
a
C0(MA) = 1 mmol L−1, C0(4-HMA) = 1 mmol L−1, C0(4-MMA) = 1 mmol L−1, C0(H2O2) = 0.03 mol L−1, C0(LCu) = 0.05 mmol L−1
.

318
W.-F. Yu et al. / Journal of Molecular Catalysis A: Chemical 379 (2013) 315–321
2.0
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0.0
0.4
2.0
1.5
1.0
0.5
0.0
pH=2.5
pH=3
pH=3.5
pH=4
pH=4.5
pH=2.5
pH=3
pH=3.5
pH=4
pH=4.5
a
b
1.5
1.0
0.5
0.0
0.3
0.2
0.1
0.0
0
20
40
t/min
60
80
wav7e0l0ength/nm
400
500
600
800
900
1000
1100
0
200
400
600
800 1000
0
100 200
400 500 600
t3/00s
t / s
Fig. 5. The UV–vis spectral changes showing the formation of the L2CuI-•OOH
species vs time in tartaric acid solution at 25 ◦C, pH = 3, C0(H2O2) = 0.02 mol L−1
C0(L2Cu) = 5 mmol L−1. Inset: time course of the absorbance change of the L2CuI-
•OOH species at 480 nm for the reaction with 4-MMA, C0(4-MMA) = 0.1 mol L−1
,
Fig. 3. Plots of ln[(A∞ − A0)/(A∞ − At)] vs reaction time t for the oxidations of 4-
MMA by H2O2 using L1Cu (a) and L2Cu (b) as catalyst in tartaric acid solution at
.
35 ◦C. C0(4-MMA) = 1 mmol L−1, C0(H2O2) = 0.03 mol L−1, C0(LCu) = 0.05 mmol L−1
.
more than 2 h (Fig. 4). It was also found that an optic feature with
ꢀmax = 780 nm appeared and simultaneously an optic feature with
ꢀmax = 980 nm decreased in the infrared region, isosbestic points
were observed at 605 nm and 870 nm (see Fig. S3). These indicated a
new stable species formed in aqueous solution at 25 ◦C. It was noted
that the generated species could slowly decompose when it was
placed in the aqueous solution beyond 3 h at ambient temperature.
Similarly, after H2O2 was added to the L2Cu complex solution,
an UV–vis spectral feature with ꢀmax = 480 nm was observed and
simultaneously an optical feature with ꢀmax = 800 nm increased
(Fig. 5). From Figs. 4 and 5, we found that although the structure of
L2Cu was similar to that of L1Cu, the characteristic UV spectrum of
species L2CuI-•OOH was different from that of L1Cu, this implied
that ligand L could cause sensitive influence to the copper-oxo
species. From Fig. 5, it could be seen that the new species could
be rapidly reduced by 4-MMA, the reaction followed a pseudo-
first-order kinetic behavior in the case of excessive 4-MMA and the
pseudo-first-order rate constant of 2.3 × 10−3 s−1 was estimated
under the experimental conditions as in Fig. 5.
According to the investigations of UV–vis absorbance, a new
active species could be found under this condition. In order to
identify the new species further, the reaction aqueous solution
was detected by electrospray ionization mass spectrometry (ESI-
MS) at ambient temperature. The spectrum of positive ion (see Fig.
S4) showed a ion peak at m/z of 319.25 attributed to [H2L1CuI-
•OOH]+ (Calc. m/z of 320.08), a peak at m/z of 374.31 attributed to
[H2L1CuI-•OOH(3H2O)]+ (Calc. m/z of 374.11) and a peak at m/z of
390.29 attributed to [H2L1CuI-•OOH(2H2O)(H2O2)]+ (Calc. m/z of
390.11). It could be observed that the reaction of L1Cu with H2O2
results in the formation of [H2L1CuI-•OOH]+ species. The spectrum
of negative ion (Fig. S6) showed an anion peak at m/z of 334.93
attributed to [L1CuI-OOH(H2O)]− (Calc. m/z of 336.07) and a peak
at m/z of 486.88 attributed to [L1CuI-OOH(2H2O2)(N(C2H5)3)]−
(Calc. m/z of 487.20). Similarly, according to the ESI-MS spec-
trum of positive ion for the reaction solution of L2Cu with H2O2,
a prominent peak corresponding to [H2L2CuI-•OOH]+ appeared
at m/z of 377.13 (Calc. m/z of 380.10) (see Fig. S5). It could be
found that the other two peaks at m/z of 399.15 and m/z of
499.29 attributed to [H2L2CuI-•OOH(H2O)]+ (Calc. m/z of 398.11)
and [H2L2CuI-•OOH(H2O)(3H2O2)]+ (Calc. m/z of 500.13), respec-
tively. This indicated the generation of [H2L2CuI-•OOH]+ derived
from the interaction of L2Cu with H2O2. It was very possible that
there were different state forms due to the proton transfer between
2H+
[H2LCu]2+
LCu
H2O2
H2O2
2H+
[LCu-OOH]-
[H2LCuI-OOH]+
Scheme 2. The possible proton transfer balances and reaction paths.
1.5
1.5
1.0
0.5
0.0
1.0
0.5
0.0
0
30
60
t/min
90
120
wa3v2e0length/nm
240
280
440
480
Fig. 4. The UV–vis spectral changes showing the formation of the L1CuI-•OOH
species vs time t in tartaric acid solution at 25 ◦C, pH = 3, C0(H2O2) = 0.2 mmol L−1
,
C0(L1Cu) = 0.05 mmol L−1. Inset: the generation of the L1CuI-•OOH species and the
absorbance change at 252 nm vs time t.
LCuI-•OOH species, the generation and properties of the species
were investigated in detail.
The reaction processes of L1Cu with 4 equiv. H2O2 were mon-
itored by UV–vis in tartaric acid solution at 25 ◦C. The spectral
changes were shown in Fig. 4. It can be observed that an optic fea-
ture with ꢀmax = 252 nm appeared after adding H2O2 to the L1Cu
complex solution. This indicated that a new species formed. Fur-
ther the formation kinetics showed that the new species could be
readily generated within 30 min and then could stably exist for

W.-F. Yu et al. / Journal of Molecular Catalysis A: Chemical 379 (2013) 315–321
319
Fig. 6. Cyclic voltammogram of 1 mmol L−1 LCu and addition of different amounts of H2O2 at a scan rate of 100 mV s−1 in unstirred deaerated tartaric acid solution at 25 ◦C;
pH = 3; A: L1Cu, a: 0 mmol L−1 H2O2, b: 3 mmol L−1 H2O2, c: 6 mmol L−1 H2O2; B: L2Cu, e: 0 mmol L−1 H2O2, f: 5 mmol L−1 H2O2, g: 10 mmol L−1 H2O2.
LCu and [H2LCu]2+ or between [LCu-OOH]− and [H2LCuI-•OOH]+ in
aqueous solution as shown in Scheme 2.
aqueous solutions of LCu. The results were shown in Fig. 6. Incre-
mental addition of H2O2 to the solution of L1Cu resulted in a new
increasing reduction wave at EpIIIc = −0.65 V/SCE (Fig. 6A). Similarly,
Fig. 6B showed the current responses of L2Cu in the absence and
presence of H2O2. From Fig. 6B, a new increasing reduction wave
at EpIIIc = −0.67 V/SCE with the addition of H2O2 can be observed,
Further evidence for the formation of the LCuI-•OOH species
was obtained with electrochemical method. We attempted to iden-
tify voltammetric responses corresponding to the active species by
titration experiments by adding different amounts of H2O2 to the
Fig. 7. Optimized geometric configurations of L1Cu, L2Cu, [H2L1CuI-•OOH]+ and [H2L2CuI-•OOH]+ (nitrogen atoms are illustrated in blue, oxygen atoms in red and copper
atom in orange). (For interpretation of the references to color in this figure legend, the reader is referred to the web version of the article.)

320
W.-F. Yu et al. / Journal of Molecular Catalysis A: Chemical 379 (2013) 315–321
the increase of the reduction current intensity indicated that a new
oxidation species formed in the presence of H2O2 in a similar man-
ner to other reports early [28,37–40]. It could be observed that the
two small reduction waves Ic and IIc, which may be correspond to
the state forms of Cu2+ and Cu+ respectively [41–43], also caused
some tiny changes (see Table S1) with the addition of H2O2. An
obvious phenomenon was that for the two complexes L1Cu and
L2Cu, both the position and the electric current intensity of the
oxidation waves at EpIa = 0.311–0.316 V/SCE hardly changed after
adding H2O2, which was different from those metal complexes
reported by literatures [37–39]. It was also found that the new
4. Conclusions
The reactive species LCuI-•OOH could be generated by reac-
ting copper complexes LCu with H2O2 in tartaric acid solution. The
reactive species was relatively stable and possesses relatively high
oxidation potential in aqueous solution at ambient temperature.
The DFT computation results indicated that the geometry of LCuI-
•OOH was five-coordinate square-pyramidal configuration and this
may lead to the extreme sensitivity of species LCu-•OOH as well as
reaction to environment. The copper (II) complexes LCu displayed
excellent catalytic ability for the selective oxidation of Mandelic
acids (MA, 4-HMA, 4-MMA) into aromatic aldehydes by H2O2 in
tartaric acid solution.
active species possesses relatively high reduction potential (EpIIIc
)
in aqueous solution at ambient temperature (Table S1). The redox
potential of the LCuI-•OOH species is larger than those of most
reported metal oxo species [2,23], the species is a more stable than
reported radical complexes [44,45] in aqueous solution.
For metal complexes, the properties of ligands generally
determine the degree of electron transfer. Owing to the Ligand
coordinated with Cu2+ was electronic conjugate system, the elec-
tron easily transfers within the ligand. Further, it is possible that
for the LCuI-•OOH species, the electron could easily transfer among
ligand L, Cu2+(or Cu+) and •OOH. Therefore, in the process of LCuI-
•OOH species associating and reacting with substrate, the electron
and proton H+ could easily transfer between the LCuI-•OOH species
and substrate.
Supplementary data
Some experimental data and computational data.
Acknowledgments
This work has been supported by the National Natural Science
Foundation of China (Nos. 21073126 and 21273156).
Appendix A. Supplementary data
Supplementary data associated with this article can be
found, in the online version, at http://dx.doi.org/10.1016/
j.molcata.2013.09.004.
3.3. Density Functional Theory computation
The geometric configurations of copper (II) complexes LCu,
[H2LCuI-•OOH]+ (Fig. 7), [H2LCu]2+ and [LCu-OOH]− (Fig. S7) were
calculated by using DFT method. The computation results indi-
cate that the geometries of complexes LCu and [H2LCu]2+ are
square plane configurations with Cu2+ at the center. The geomet-
ric configurations of [LCu-OOH]− (Fig. S7) and [H2LCuI-•OOH]+ are
five-coordinate square pyramidal structures: four atoms includ-
ing two N atoms, one carbonyl O atom and one hydroperoxo O
atom encircle a central Cu2+ ion to construct an approximate square
plane configuration and one carbonyl O atom is in an axial position.
The introduction of hydroperoxo free radical •OOH reorganizes the
coordination of Cu2+ ion with coordinating other atoms. To our sur-
prise, the hydroperoxo O atom occupies one position of carbonyl
O atom and then the carbonyl O atom is arranged at the position
References
[1] B.F. Gherman, D.E. Heppner, W.B. Tolman, C.J. Cramer, J. Biol. Inorg. Chem. 11
(2006) 197–205.
[2] S.O. Kim, C.V. Sastri, M.S. Seo, J. Kim, W. Nam, J. Am. Chem. Soc. 127 (2005)
4178–4179.
[3] A. Serafimidou, A. Stamatis, M. Louloudi, Catal. Commun. 9 (2008) 35–39.
[4] R.A. Geiger, D.F. Leto, S. Chattopadhyay, P. Dorlet, E. Anxolabéhère-Mallart, T.A.
Jackson, Inorg. Chem. 50 (2011) 10190–10203.
[5] S. Itoh, Curr. Opin. Chem. Biol. 10 (2006) 115–122.
[6] M.R. Maurya, A.K. Chandrakar, S. Chand, J. Mol. Catal. A: Chem. 270 (2007)
225–235.
[7] J.A. Bumpus, M. Tien, D. Wright, S.D. Aust, Science 228 (1985) 1434–1436.
[8] S.T. Prigge, A.S. Kolhekar, B.A. Eipper, R.E. Mains, L.M. Amzel, Science 278 (1997)
1300–1305.
[9] M. Kodera, T. Kita, I. Miura, N. Nakayama, T. Kawata, K. Kano, S. Hirota, J. Am.
Chem. Soc. 123 (2001) 7715–7716.
[10] A. Crespo, M.A. Marti, A.E. Roitberg, L.M. Amzel, D.A. Estrin, J. Am. Chem. Soc.
128 (2006) 12817–12828.
[11] K. Yoshizawa, N. Kihara, T. Kamachi, Y. Shiota, Inorg. Chem. 45 (2006)
3034–3041.
[12] A. Decker, E.I. Solomon, Curr. Opin. Chem. Biol. 9 (2005) 152–163.
[13] P. Comba, C. Haaf, S. Helmle, K.D. Karlin, S. Pandian, A. Waleska, Inorg. Chem.
51 (2012) 2841–2851.
[14] Y.J. Choi, K.B. Cho, M. Kubo, T. Ogura, K.D. Karlin, J. Cho, W. Nam, Dalton Trans.
40 (2011) 2234–2241.
[15] T. Fujii, A. Naito, S. Yamaguchi, A. Wada, Y. Funahashi, K. Jitsukawa, S. Nagatomo,
T. Kitagawa, H. Masuda, Chem. Commun. (2003) 2700–2701.
[16] S.H. Kim, H. Park, M.S. Seo, M. Kubo, T. Ogura, J. Klajn, D.T. Gryko, J.S. Valentine,
W. Nam, J. Am. Chem. Soc. 132 (2010) 14030–14032.
[17] A. Wada, S. Ogo, S. Nagatomo, T. Kitagawa, Y. Watanabe, K. Jitsukawa, H.
Masuda, Inorg. Chem. 41 (2002) 616–618.
[18] R. Mas-Balleste, L. Que Jr., J. Am. Chem. Soc. 129 (2007) 15964–15972.
[19] J. Cho, S. Jeon, S.A. Wilson, L.V. Liu, E.A. Kang, J.J. Braymer, M.H. Lim, B. Hedman,
K.O. Hodgson, J.S. Valentine, E.I. Solomon, W. Nam, Nature 478 (2011) 502–505.
[20] I. Schepetkin, A. Potapov, A. Khlebnikov, E. Korotkova, A. Lukina, G. Malovichko,
L. Kirpotina, M.T. Quinn, J. Biol. Inorg. Chem. 11 (2006) 499–513.
[21] T.Y. Lin, C.H. Wu, J. Catal. 232 (2005) 117–126.
[22] S. Wang, L. Wang, M. Dakovic, Z. Popovic, H. Wu, Y. Liu, ACS Catal. 2 (2012)
230–237.
[23] S. Fukuzumi, L. Tahsini, Y.M. Lee, K. Ohkubo, W. Nam, K.D. Karlin, J. Am. Chem.
Soc. 134 (2012) 7025–7035.
˚
of an axial coordination. The calculated distance (1.888–1.898 A)
of Cu atom with hydroperoxo O atom is shorter than that of other
four coordination atoms (Table S2), which indicate that the associ-
ation of hydroperoxo free radical •OOH with copper complex LCu
is steady. The axial ligand also plays an important role in the activ-
ity for catalysis reaction [46,47]. The DFT calculation displays that
the property of copper complex associated hydroperoxo free radi-
cal species may be easily changed through the replacement of the
axial ligand of carbonyl O atom by other group containing O, N, P or
S coordination atom. In this work it was found that when ethanol
was added to the LCu/H2O2 mixture solution, the UV–vis spectrum
recovered its former shape of LCu solution, and the product analysis
showed that the ethanol had not be oxidized. This implied that the
ethanol may disturb the coordination form of copper complexes,
changed the geometric configurations of active species by replacing
the axial ligand of carbonyl O atom and resulted in the disappear-
ance of species L1CuI-•OOH. This could also explain the sensitivity
of species LCu-•OOH as well as reaction to environment. Further we
found that the generation of species L1CuI-•OOH was sensitive to
solution medium, and the tartaric acid played an important role to
the generation of this species. Compared with H2O2, The O-O bond
length in complex [H2LCuI-•OOH]+ is slightly shorter (Table S2),
which implied that the copper complex could induce the electron
transfer of O atoms and activate the free radical •OOH.
[24] E.I. Solomon, P. Chen, M. Metz, S.K. Lee, A.E. Palmer, Angew Chem. Int. Ed. 40
(2001) 4570–4590.
[25] J.M. Li, X.G. Meng, C.W. Hu, J. Du, X.C. Zeng, J. Mol. Catal. A: Chem. 299 (2009)
102–107.
[26] V. Resch, W.M.F. Fabian, W. Kroutil, Adv. Synth. Catal. 352 (2010) 993–997.

W.-F. Yu et al. / Journal of Molecular Catalysis A: Chemical 379 (2013) 315–321
321
[27] M.E. Brown, T. Barros, M.C.Y. Chang, Chem. Biol. 7 (2012) 2074–2081.
[28] E.E. Ferapontova, J. Castillo, L. Gorton, BBA 1760 (2006) 1343–1354.
[29] N. Sathitsuksanoh, A. George, Y.H.P. Zhang, J. Chem. Technol. Biotechnol. 88
(2013) 169–180.
[30] T.Q. Yuan, F. Xu, R.C. Sun, J. Chem. Technol. Biotechnol. 88 (2013) 346–352.
[31] W.P. Zeng, X.H. Li, J. Du, J.M. Li, P. Zhang, C.W. Hu, X.G. Meng, Acta Chim. Sinica
68 (2010) 27–32.
[32] B. Liu, X.G. Meng, W.Y. Li, L.C. Zhou, C.W. Hu, J. Phys. Chem. A 116 (2012)
2920–2926.
[33] A.E. Martell, R.L. Belford, M. Calvin, J. Inorg. Nucl. Chem. 5 (1958) 170–181.
[34] C.K.Z. Andrade, A.F.S. Barreto, W.A. Silva, Arkivoc 7 (2008) 226–232.
[35] L. Mouni, M. Akkurt, S.O. Yıldırım, T.B. Hadda, Z.H. Chohan, J. Chem. Crystallogr.
40 (2010) 169–172.
[39] Q. Yu, Z.S. Shi, X.Y. Liu, S.L. Luo, W.Z. Wei, J. Electroanal. Chem. 655 (2011) 92–95.
[40] V.L.N. Dias, E.N. da Fernandes, L.M.S. Silva, E.P. Marques, J.J. Zhang, A.L. Brandes
Marques, J. Power Sources 142 (2005) 10–17.
[41] M. Meyer, L. Fremond, E. Espinosa, R. Guilard, Z. Ou, K.M. Kadish, Inorg. Chem.
43 (2004) 5572–5587.
[42] A.M. Lazarin, C.A. Borgo, Y. Gushikem, J. Membr. Sci. 221 (2003) 175–184.
[43] K.K. Sarker, S.S. Halder, D. Banerjee, T.K. Mondal, A.R. Paital, P.K. Nanda, P.
Raghavaiah, C. Sinha, Inorg. Chim. Acta 363 (2010) 2955–2964.
[44] K. Nakahara, S. Iwasa, J. Iriyama, Y. Morioka, M. Suguro, M. Satoh, E.J. Cairns,
Electrochim. Acta 52 (2006) 921–927.
[45] F.S. de Paula, A.G. Cioletti, J.F. da Silva Filho, A.E.G. Santana, A.F. dos San-
tos, M.O.F. Goulart, M. Vallaro, R. Fruttero, J. Electroanal. Chem. 544 (2003)
25–34.
[36] J.P. Costes, J.P. Laurent, Inorg. Chem. 27 (1988) 3235–3237.
[37] C.V. Sastri, K. Oh, Y.J. Lee, M.S. Seo, W. Shin, W. Nam, Angew. Chem. Int. Ed. 45
(2006) 3992–3995.
[46] L. Duan, F. Bozoglian, S. Mandal, B. Stewart, T. Privalov, A. Llobet, L. Sun, Nat.
Chem. 4 (2012) 418–423.
[47] J.L. Fillol, Z. Codola, I. Garcia-Bosch, L. Gomez, J.J. Pla, M. Costas, Nat. Chem. 3
(2011) 807–813.
[38] M. Senel, E. Cevik, M.F. Abasıyanık, Sens. Actuator B: Chem. 145 (2010) 444–450.