Copper catalysts based on fiberglass supports for hydrocarbon oxidation reactions with the participation of hydrogen peroxide

Article abstract of DOI:10.1134/S0023158408040071

Copper-containing catalysts were prepared by the adsorption of the ammonia complexes of Cu(II) on the surface of a silicate fiberglass material followed by the thermal and oxidative treatment of the samples. The states of copper after the adsorption of ammonia complexes and in the prepared samples were characterized using electronic diffuse reflectance spectroscopy. The catalytic activity of the samples in hydrogen peroxide decomposition and cyclohexane oxidation reactions was studied. It was found that molecular oxygen can be involved in the radical process of hydrogen peroxide oxidation. Based on spectroscopic data, it was hypothesized that partially reduced Cu(I)-Cu(O) compounds are active species in the catalysts of this type.

Full text of DOI:10.1134/S0023158408040071

ISSN 0023-1584, Kinetics and Catalysis, 2008, Vol. 49, No. 4, pp. 499–505. © MAIK “Nauka /Interperiodica” (Russia), 2008.
Original Russian Text © T.A. Balandina, T.Yu. Larina, N.I. Kuznetsova, B.S. Bal’zhinimaev, 2008, published in Kinetika i Kataliz, 2008, Vol. 49, No. 4, pp. 522–528.
Copper Catalysts Based on Fiberglass Supports for Hydrocarbon
Oxidation Reactions with the Participation of Hydrogen Peroxide
T. A. Balandina, T. Yu. Larina, N. I. Kuznetsova, and B. S. Bal’zhinimaev
Boreskov Institute of Catalysis, Siberian Branch, Russian Academy of Sciences, Novosibirsk, 630090 Russia
e-mail: balandina.tanya@gmail.com
Received April 27, 2007
Abstract—Copper-containing catalysts were prepared by the adsorption of the ammonia complexes of Cu(II)
on the surface of a silicate fiberglass material followed by the thermal and oxidative treatment of the samples.
The states of copper after the adsorption of ammonia complexes and in the prepared samples were characterized
using electronic diffuse reflectance spectroscopy. The catalytic activity of the samples in hydrogen peroxide
decomposition and cyclohexane oxidation reactions was studied. It was found that molecular oxygen can be
involved in the radical process of hydrogen peroxide oxidation. Based on spectroscopic data, it was hypothe-
sized that partially reduced Cu(I)–Cu(0) compounds are active species in the catalysts of this type.
DOI: 10.1134/S0023158408040071
INTRODUCTION
Copper complexes with organic ligands immobi-
lized on solid support surfaces belong to another cate-
gory of heterogeneous systems. Metal oxides [4, 5]; sil-
icon oxides [6–8]; and zeolites X, Y [9, 10], MCM-41
Hydrogen peroxide is the most promising reagent
for a number of oxidation processes in terms of cost and
environmental safety. The oxidative functionalization
of hydrocarbons is a problem to be solved frequently by
hydrogen peroxide oxidation. In this area, a great body
of data on catalysis by transition metal ions and com-
plexes, in particular, copper compounds, has been accu-
mulated.
Copper complexes with various mono-, bi-, and tri-
dentate organic ligands, such as pyridine, bipyridine,
tripyridine, tris(pyrazolyl)borate, tris(pyrazolyl)meth-
ane [1], and porphyrins, are used in the oxidation of
organic substrates. In the presence of these complexes,
the hydrogen peroxide oxidation of saturated and
unsaturated hydrocarbons, alcohols, and alkylbenzenes
occurs with the formation of corresponding alcohols,
ketones, and hydroperoxides [1].
[
8, 11], ZSM-5 [12, 13], and HMS [14] are most com-
monly used as supports. The examples of a sufficiently
high activity of heterogenized copper complexes are
well known. Thus, the conversion of phenol in an aque-
ous solution by hydrogen peroxide oxidation on heter-
ogeneous catalysts of this type was higher than 36%;
the total selectivity for catechol and hydroquinone was
higher than 95% [14]. Aromatic unsaturated hydrocar-
bons (toluene and benzene) were completely oxidized
to CO and H O in the presence of the most active cat-
2
2
alysts 5 wt % Cu/ZrO [4] and 1.5 wt % Cu/SiO [7].
2
2
Several problems appear in the use of heterogeneous
catalysts for peroxide oxidation. The majority of heter-
ogeneous systems described in the literature, like
Under conditions of homogeneous liquid-phase homogeneous systems, contain organic components.
catalysis, the ligand environment of copper ions, on the The degradation of organic ligands occurs in an oxidiz-
one hand, retains them in solution and, on the other ing medium, and the support is also often affected. In
hand, forms the structure of active sites. The oxidation addition, many supports, particularly zeolites, exhibit
reactions of saturated [2, 3] and unsaturated aromatic high porosity. Therefore, a portion of an active compo-
hydrocarbons [3] analogous to homogeneous reactions nent becomes inaccessible to reactants or the desorp-
occur on the surface of solid copper complexes, which
are insoluble in the reaction mixture. Thus, the oxida-
tion of cyclohexane by hydrogen peroxide occurs inten-
sively in the presence of heterogeneous mononuclear
tion of oxidation products from the catalyst surface into
a solution is hindered [8–14].
In this study, copper catalysts were prepared by the
and tetranuclear porphyrin complexes of copper [2]. immobilization of inorganic copper complexes on a
The corresponding alcohol and ketone, cyclohexyl fiberglass support, which was characterized previously
hydroperoxide, and adipic acid are the oxidation prod- [15]. The important properties of the fiberglass used are
ucts. Other complexes, such as copper(II) trans-1,4- the following: (i) a low specific surface area and the
cyclohexanedicarboxylate, oxidize various alcohols to absence of pores; (ii) the presence of a great number of
corresponding ketones with selectivity higher than 99% surface silane and siloxane groups, which serve as
in a 20-fold excess of hydrogen peroxide [3].
immobilization sites for complex copper cations; and
4
99

5
00
BALANDINA et al.
(
iii) stability to degradation upon high-temperature
Physicochemical Methods
treatments to 900°C [15].
The concentration of copper was determined by
In this work, we were the first to study heteroge- X−ray fluorescence spectroscopy after the complete
neous copper-containing catalysts prepared by the dissolution of the sample.
adsorption of copper(II) acetate and copper(II) tetraam-
moniate on a fiberglass support in the liquid-phase oxi-
dation reaction of cyclohexane. Catalytic systems
based on fiberglass materials were used only in gas-
phase reactions [16, 17], and they were not used previ-
ously in liquid-phase oxidation catalysis.
The electronic diffuse reflectance spectra of copper-
containing fiberglass Cu/SiO samples and the UV
2
transmission spectra of the solutions of copper com-
plexes were measured with reference to BaSO over the
4
–1
range 11000–54000 cm on a UV-2501 PC spectro-
photometer (Shimadzu, Japan) with an ISR-240 A dif-
fuse reflectance attachment. The electronic diffuse
reflectance spectra were plotted in the wavenumber–
Kubelka–Munk function (F(R)) coordinates. The trans-
mission spectra of solutions of copper(II) complexes
were measured with reference to water using a quartz
cell with an optical path length of 2 mm.
The cyclohexane oxidation products were analyzed
by gas chromatography on a Kristall 2000m instrument
equipped with a DB-1701 capillary column (30 m in
length and 0.53 mm in i.d.) and a flame ionization
detector.
EXPERIMENTAL
Chemicals
The silica fiberglass support used in this study was
characterized in detail previously [15]. The material
consisted of threads slightly less than 1 mm in diameter,
which were weakly spun from thin fibers with d ~7–
1
0 µm; the threads easily disintegrated into individual
fibers. The fibers consisted of nonporous amorphous
2
SiO with S = 1 m /g.
2
sp
The following chemicals were used in this study:
cyclohexane of analytical grade, reagent-grade hydro-
gen peroxide (30%), aqueous ammonia (25%) of ana-
lytical grade, acetonitrile from Kriokhrom with a water and hydrogen peroxide decomposition were performed
content of 0.005% (was used without preliminary puri- in a thermostated glass reactor with a volume of
fication), distilled water, and 95% ethanol.
Catalytic Tests
The catalytic reactions of cyclohexane oxidation
~30 ml, which was connected to a graduated glass gas
burette. As a rule, the catalyst (0.1 g), the solvent aceto-
nitrile (3.8 × 10 mol), and the cyclohexane substrate
–
2
–
3
Synthesis of Cu/SiO Samples
2
(1.85 × 10 mol) (some experiments on hydrogen per-
oxide decomposition was performed in the absence of
cyclohexane) were loaded in the reactor. The reactor
was thermostated (T = 60°ë) for several minutes and
connected to the gas burette, and atmospheric pressure
was maintained in the system. In order to perform the
cyclohexane oxidation reaction in an atmosphere of
oxygen or nitrogen, the reactor was purged with an
appropriate gas. Hydrogen peroxide (8.53 × 10 mol)
was introduced into the reactor through a special nozzle
without depressurization; thereafter, a magnetic stirrer
was turned on (500 rpm). Changes in the gas-phase vol-
ume in the course of the reaction were monitored using
the gas burette.
The Cu/SiO samples were prepared by the adsorp-
2
tion of copper from solutions of cationic ammonia
complexes of Cu(II) on a fiberglass support.
To prepare impregnating solutions, copper chloride
(
CuCl · 2H O) or copper acetate (Cu(CH COO) ) was
2 2 3 2
dissolved in water with the addition of 25% NH OH at
4
–
3
the molar ratio Cu(II)/NH OH = 1 : 10.
4
Adsorption was performed in a thermostated shaken
glass reactor with a volume of ~50 ml at a fiberglass-to-
solution-weight ratio of ~1 : 20 and at a fixed solution
temperature. The adsorption temperature and time were
varied over the ranges 30–70°ë and 10–120 min,
respectively. After the adsorption, the sample was
The catalytic activity of the samples in the reaction
removed from the reactor, repeatedly washed with dis- was evaluated as the total amount of products formed in
tilled water at room temperature, and dried on a filter in a specified time interval.
a flow of warm air.
For the reductive treatment of the samples, they
were placed in a glass reactor, which was initially
blown with helium at a rate of 80 ml/min for 5 min and
then with hydrogen at a rate of 70 ml/min. The sample
was heated to a specified temperature at a rate of
RESULTS AND DISCUSSION
UV Spectra of Impregnating Solutions
and the Cu/SiO Samples Prepared from Them
2
As can be seen in Fig. 1, the spectrum of a solution
1
0 K/min in a flow of hydrogen and kept for 1 h at this of CuCl · 2ç é in water changed upon the addition of
2 2
temperature. The main treatment temperature was var- ammonia. Even at the concentration ratio [CuCl ·
2
ied from 100 to 400°ë for different samples. The treat- 2ç é]/[NH OH] = 1 : 4, a tetracoordinated ammonia
2
4
ment was finished by cooling the sample to room tem- complex of copper was formed, as evidenced by
–1
perature in a flow of hydrogen and purging with helium. absorption at 16 500 cm (curve 5) [18]. Moreover, this
KINETICS AND CATALYSIS Vol. 49 No. 4 2008

COPPER CATALYSTS BASED ON FIBERGLASS SUPPORTS
solution exhibited absorption over the range 28000– D, arb. units
501
–
1
3
3000 cm due to polynuclear hydroxo complexes of 0.20
Cu(II). At a higher concentration of ammonium
hydroxide (the concentration ratio [NH OH]/[CuCl ·
4
2
3
2000
2
ç é] increased from 4 to 50), the absorption charac-
2
teristic of hydroxo complexes became less intense and 0.15
practically disappeared (curves 4, 3, and 2), whereas
the absorption band intensity of ammonia complexes
increased. Because the treatment of the support with
highly concentrated ammonia solutions could cause the 0.10
degradation of the material, a solution containing a ten-
fold amount of ammonia with respect to Cu(II) was
subsequently used for impregnation. As judged from
the UV spectrum (Fig. 1, curve 4) and published data 0.05
1
2
3
4
1
2100
1
6000
16500
1
5
2+
[18], Cu(II) ions occur in this solution as [Cu(NH ) ]
3 6
2
3
4
5
cations, whereas hydroxo complexes are almost absent.
In the case of copper acetate, the solutions exhibited
analogous spectra. Therefore, a solution with a tenfold
amount of ammonium hydroxide with respect to copper
acetate was also used in this case for impregnation.
0
1
0000 15000 20000 25000 30000 35000 40000
–1
ν, cm
The samples containing from 0.05 to 0.08 wt % cop-
per were obtained by the treatment of the fiberglass
with a solution of the ammonia complex of copper.
Table 1 summarizes the treatment temperatures and the
support–solution contact times. As judged from the
data in Table 1, the adsorption equilibrium of the
ammonia complexes was rapidly established, so that
changes in the temperature and time of fiberglass treat-
ment with the ammonia solution of copper over the
specified ranges only slightly affected the metal content
of the final sample. Nevertheless, higher copper con-
tents of the samples after more intense treatments
Fig. 1. UV spectra of the ammonia complexes of copper(II)
in water: (1) CuCl · 2H O, (2–5) with the addition of
2
2
NH · H O to the molar ratios Cu/NH = (2) 1 : 50,
3
2
3
(
3) 1 : 30, (4) 1 : 10, and (5) 1 : 4.
due to the ammonia complex on the fiberglass surface
–1
shifted to 14900 cm , as compared to an analogous
–1
band for the solution (16000 cm ). This suggests the
chemical interaction of the ammonia complex with the
support surface, because of which the composition of
the coordination sphere of copper ions changed. The
band due to hydroxo complexes shifted to the long-
(
at higher temperatures and longer contact times) can
–1
wavelength region (to 30600 cm ); this band was
be noted. Thus, the subsequent experiments were per-
formed with catalyst sample no. 7 (Table 1).
much more intense than that in the spectrum of the
solution. As judged from the intensity ratio between the
The diffuse reflectance spectra of support samples bands due to the ammonia complex and hydroxo com-
after contact with solutions of the ammonia complex of plexes, the major portion of surface copper occurred as
copper [Cu(NH ) ]Cl were characterized. The band hydroxo complexes, unlike the starting solution con-
3
6
2
Table 1. Conditions of the adsorption of the ammonia complexes of copper on fiberglass and the copper contents of the re-
sulting samples
Treatment conditions
Sample
[Cu], wt %
temperature, °C
time, min
impregnating solution
[Cu(NH ) ]Cl
1
2
3
4
5
6
7
8
50
60
120
10
0.06
0.07
0.05
0.07
0.08
0.07
0.07
0.08
3
6
2
60
60
120
10
70
50
50
120
[Cu(NH ) ] · (CH COO)
2
3
6
3
–
2
Note: The impregnating aqueous solution contained [CuCl · 2H O] = [Cu(CH COO) ] = 1.52 × 10 mol/l and [NH · H O] =
2
2
3
2
3
2
1
.52 × 10–1 mol/l.
KINETICS AND CATALYSIS Vol. 49 No. 4 2008

5
02
BALANDINA et al.
4
F(R), arb. units
00
Amount of O × 10 , mol
2
4
0300
5
4
3
2
1
0
1
17000
1
6400
2
3
1800
4
3
0000
3
15700
1
2
3
4
1
1
0–1
2
1
4500
2
4
6
10 16 24 30 46 60 75 90 105 120
Time, min
1
Fig. 3. Kinetics of O release in the decomposition of H O
2
2
2
in the presence of (1) 0.05 or (2) 0.10 g of the catalyst.
1
0000
20000
30000
40000
50000
ν, cm
–
1
Hydrogen Peroxide Decomposition
Fig. 2. Diffuse reflectance spectra of fiberglass samples pre-
pared by impregnation from a solution of Cu(NH ) Cl and
reduced in a flow of hydrogen at (1) 100, (2) 200, (3) 300,
or (4) 400°ë.
The samples subjected to reductive treatment cata-
lyzed the decomposition of aqueous hydrogen peroxide
in an acetonitrile solution.
3
6
2
The initial portion of the kinetic curves is character-
ized by an induction period, which is related to the
development of a chain process. The rate of decompo-
sition subsequently increased with time (Fig. 3).
The rate of reaction and the shape of hydrogen per-
oxide decomposition curves depended on both catalyst
weight (Fig. 3) and the initial amount of hydrogen per-
oxide in solution. The dependence exhibited a compli-
cated character; however, the curves were similarly
shaped. As the catalyst weight and the amount of
taining only the ammonia complex. The intensities of
both of the bands were somewhat higher in the samples
treated under more severe conditions: at a higher tem-
perature and a long time of contact between the fiber-
glass and the solution.
In the course of adsorption, support threads partially
separated into thin fibers; therefore, the threads were
specially divided into fibers in the subsequent experi-
ments. This resulted in a more uniform wetting of the hydrogen peroxide were increased, the induction
sample with the impregnating solution and had a weak period shortened; in this case, the rates of hydrogen
effect on the spectra of the samples after adsorption.
peroxide decomposition and reaction product forma-
tion increased with time.
After the reductive treatment with hydrogen from
the gas phase, intense absorption in the region 20000–
–
1
2
5000 cm , which corresponds to mixed-valence
Cyclohexane Oxidation in the Presence
of Fiberglass Samples
Cu(I)–Cu(0) compounds [19–21], appeared or
increased in the spectra of the samples. In the next
series of samples, the adsorption of the copper complex
was performed at 50°C for 120 min followed by treat-
ment in a flow of hydrogen. Figure 2 shows the diffuse
reflectance spectra of the samples thus prepared. After
The oxidation of cyclohexane occurred under the
reaction conditions of hydrogen peroxide decomposi-
tion. Cyclohexanol, cyclohexanone, and minor
amounts of cyclohexyl hydroperoxide were the main
oxidation products. Note that the fiberglass samples
the treatment of the samples at the highest temperatures that were dried in air after the adsorption of
(
1
300 and 400°C), absorption bands at 17000 and [ëu(NH ) ]Cl were found to be almost inactive. Cata-
3
6
2
–
1
6400 cm (curves 3 and 4), which were attributed to lytic activity in cyclohexane oxidation, as well as in
copper metal [19], appeared in the spectrum in addition hydrogen peroxide decomposition, developed only
to absorption characteristic of Cu(I)–Cu(0) com- after the reductive treatment of the samples. Figure 4
pounds. As the reduction temperature was increased shows the amounts of cyclohexane oxidation products
from 300 to 400°C, the absorption characteristic of the formed upon the addition of copper-containing fiber-
glass prepared by adsorption from an ammonia solution
of the copper complex [Cu(NH ) ]Cl and reduced in
mixed-valence compounds decreased, whereas the
absorption band due to the metal became more intense.
3 6
2
hydrogen at various temperatures from 100 to 400°ë.
An analogous change in the spectrum resulted from As the reduction temperature was increased, the cata-
the reductive treatments of catalysts prepared from cop- lytic activity increased and reached a maximum after
per acetate.
reduction at 300°ë. As the catalyst reduction tempera-
KINETICS AND CATALYSIS Vol. 49 No. 4 2008

COPPER CATALYSTS BASED ON FIBERGLASS SUPPORTS
503
Products, µmol
Products, µmol
3
3
2
2
1
1
5
0
5
0
5
0
5
3
3
2
2
1
1
5
0
5
0
4
3
5
0
5
0
2
1
uncalcined H , 100°C H , 200°C H , 300°C H , 400°C
0
30
60
90
120
Time, min
2
2
2
2
Fig. 4. Amounts of cyclohexane oxidation products (cyclo-
hexanol + cyclohexanone) formed in the presence of cop-
per-containing fiberglass prepared by adsorption from an
Fig. 5. Effect of the gas atmosphere on the rate of cyclohex-
ane oxidation: (1) O , (2) N + H O , (3) air + H O , and
ammonia solution of Cu(NH ) Cl and reduced at various
3
6
2
2
2
2
2
2 2
temperatures. Time: 120 min. Atmosphere: é .
(4) O + H O .
2
2 2 2
ture was further increased to 400°ë, a smaller amount latter two cases, the reactor with the loaded reactants
of cyclohexane oxidation products was obtained. At the was purged with O or N and connected to a burette
2
2
same time, as the calcination temperature was filled with the corresponding gas.
increased, the diffuse reflectance spectra exhibited
more intense absorbance in the region 20000–
In the interaction of cyclohexane with oxygen in the
absence of hydrogen peroxide, the formation of oxida-
tion products was not detected (curve 1). Nevertheless,
oxygen positively affected the rate of cyclohexane oxi-
–1
2
5000 cm (Fig. 2), which corresponds to mixed-
valence Cu(I)–Cu(0) compounds.
An analogous behavior in cyclohexane oxidation dation in the presence of peroxide. Figure 5 shows that,
was observed in the samples prepared from copper ace- in the presence of oxygen (curve 4), oxidation occurred
tate. An unreduced sample did not exhibit activity. As in much more rapidly than in nitrogen or air (curves 2 and
the previous series of samples, activity appeared only 3, respectively).
after reductive treatment; the activity of the samples
Oxygen was consumed in the course of the reaction
from Cu(CH COO) was lower than that of the samples with hydrogen peroxide. It is likely that oxygen reacted
3
2
prepared from a solution of [Cu(NH ) ]Cl by a factor with radicals formed by the decomposition of hydrogen
3
6
2
of ~2. This can be due to the following reasons: (i) a peroxide to result in chain branching. In accordance
–
positive effect of Cl anions, which can be minor con- with the reactions given below, this increased the rate of
stituents of the samples prepared from a chloride-con- cyclohexane oxidation (R is cyclohexane).
taining solution of [Cu(NH ) ]Cl , on catalysis and (ii)
3
6
2
RH + OH•
•
R + H O,
(I)
(II)
2
different concentrations of mixed-valence compounds,
as evidenced by different absorption intensities in the
•
ROO•,
R + O
2
–
1
region 20000–25000 cm .
•
•
ROO + RH
R + ROOH,
(III)
(IV)
(V)
Note that the fiberglass samples exhibited suffi-
ciently high catalytic activity in the reaction of cyclo-
hexane oxidation even at a low copper content. Com-
paring the results of catalytic tests with the spectro-
scopic characteristics of the catalysts, we found that the
catalytic activity of the samples correlated with the
intensity of absorption in the region that is characteris-
tic of Cu(I)–Cu(0) compounds. These compounds
formed in the partial reduction of the initial ammonia
complexes of Cu(II).
•
•
ROOH
RO + OH ,
ROH + R•.
•
RO + RH
In air, the oxidation reaction occurred more slowly
than in oxygen, whereas the buildup of products in
nitrogen occurred only because of hydrogen peroxide
decomposition. The cyclohexanol/cyclohexanone
product ratio remained almost the same regardless of
the composition of the gas mixture (air, oxygen, or an
inert gas—nitrogen). This product ratio suggests the
The composition of the gas phase in the reactor and simultaneous formation of cyclohexanol and cyclohex-
the temperature of the reaction solution, as well as the anone from a common precursor, cyclohexane, whereas
amounts of the catalyst and hydrogen peroxide added, the oxidation of cyclohexanol to cyclohexanone did not
affect the rate of buildup of the main oxidation prod- occur.
ucts: cyclohexanol and cyclohexanone. The effects of
these factors are considered below.
The temperature of the reaction atmosphere also
considerably affected the rate of cyclohexane oxidation
The oxidation reaction was performed in various gas (Fig. 6). We found that cyclohexane oxidation occurred
atmospheres: air, oxygen, and nitrogen (Fig. 5). In the very slowly over a range from room temperature to
KINETICS AND CATALYSIS Vol. 49 No. 4 2008

5
04
BALANDINA et al.
F(R), arb. units
8
7
6
5
4
3
2
1
00
00
00
00
00
00
00
00
0
4
3
2
1
0
4
2
1
100
17100
3
0
60
90
120
2
1
Time, min
1
3000
3
10–1
1
, 2
3
0
60
90
120
Time, min
10–2
10000
Fig. 6. The temperature dependence of the rate of cyclohex-
ane oxidation in an atmosphere of oxygen: (1) 40, (2) 50,
(3) 60, and (4) 65°C.
20000
30000
40000
50000
ν, cm
–
1
Fig. 7. Diffuse reflectance spectra of fiberglass samples (1)
before and (2) after cyclohexane oxidation. T = 65°C; oxy-
gen atmosphere.
5
0°ë (curves 1, 2). The rapid buildup of the oxidation
products cyclohexanol and cyclohexanone came into
play at 60°ë (curves 3, 4).
In the presence of copper-containing fiberglass sam-
ples, an insignificant growth of cyclohexane oxidation
products was observed in the first minutes of reaction;
this is consistent with the analogous shape of hydrogen
peroxide decomposition curves. The rate of the oxida-
tion reaction increased with time with increasing cata-
lyst sample weight and the amount of hydrogen perox-
ide in solution.
CONCLUSIONS
The silica fiberglass support is promising for liquid-
phase reaction catalysts because of its low surface area,
the absence of a pore structure, and chemical resistance
over a wide range of pH. The adsorption of the ammo-
nia complexes of Cu(II) from a weakly alkaline solu-
tion results from coordination bonds with functional
groups on the support surface. Mixed-valence Cu(I)–
Cu(0) compounds are formed on the surface by the sub-
sequent treatments. Although the copper content of the
resulting catalysts is low, they exhibit good catalytic
activity in the reactions of hydrogen peroxide decom-
position and cyclohexane oxidation. Table 2 compares
the activity of the resulting catalysts (Cu/fiberglass)
with the activities of various copper compounds based
on other supports in the oxidation reactions of organic
substrates. The amount of oxidation products was
After completion of the reaction of cyclohexane
with hydrogen peroxide, the catalyst was removed and
used repeatedly. In the repeated test, the sample exhib-
ited a negligibly low activity, as compared with that of
the fresh catalyst. The amount of detected products was
3
5 µmol, which is smaller than that in the first cycle by
a factor of 20. As judged from the spectra of a sample
recorded before and after the catalytic reaction (Fig. 7),
deactivation was related to the oxidation of reduced
Cu(I)–Cu(0) compounds to Cu(II). A band at
9
0 mol cyclohexanol + cyclohexanone per mole of cop-
per as a catalyst constituent per hour. It is of importance
that, in an oxygen-containing atmosphere, molecular
oxygen was involved in the chain radical oxidation pro-
–
1
1
3000 cm corresponds to a Cu(II) compound [21].
–
1
For comparison, a band due to CuO lies at 15400 cm
[
19]. We found that the reductive treatment of the sam- cess, probably through preactivation, which also
ple allowed us to reactivate it to an insignificant extent. occurred on copper compounds. The participation of
Table 2. Comparison of the activity of the prepared catalysts with published data
Reaction rate, mol product ×
Catalyst
Oxidized substrate
–1 –1
Reference
[3]
(mol active component)
h
Copper(II) trans-1,4-cyclo- Linear, cyclic, and aromatic alcohols
0.40–0.21
hexanedicarboxylate
[
Cu(sal-ambmz)(Cl)]Y
Phenol
Styrene
Benzyl alcohol
Cyclohexanol
Cyclohexanol
565.7
127.4
75.6
220.3
90
[9]
Cu/MCM-41
Cu(sal-oaba)(H O)]Y
[22]
[23]
This work
[
2
Cu/fiberglass
KINETICS AND CATALYSIS Vol. 49 No. 4 2008

COPPER CATALYSTS BASED ON FIBERGLASS SUPPORTS
505
molecular oxygen in cyclohexane oxidation allowed us
to decrease the consumption of hydrogen peroxide:
9. Maurya, M.R., Chandrakar, A.K., and Chand, S., J. Mol.
Catal. A: Chem., 2007, vol. 263, p. 227.
0
.08 mol of oxidation products was formed per mole of 10. Jacob, R.C., Varkey, S.P., and Ratnasamy, P.,
H O , whereas a higher peroxide consumption is usu-
Microporous Mesoporous Mater., 1998, vol. 22, p. 465.
2
2
ally characteristic of chain radical oxidation reactions.
11. Wu, Q., Hu, X., Yue, P.L., Zhao, X.S., and Lu, G.Q.,
Appl. Catal., B, 2001, vol. 32, p. 151.
Now, it is clear that copper particles are sufficiently
stable to washing off, whereas sample deactivation was
mainly related to the reversible oxidation of copper.
The repeated use of the catalyst should be further stud-
ied in more detail.
1
1
1
1
2. Shevade, S.S., Raja, R., and Kotasthane, A.N., Appl.
Catal., A, 1999, vol. 178, no. 2, p. 243.
3. Villa, A.L. and Caro, C.A., Montes De Correa C, J. Mol.
Catal. A: Chem., 2005, vol. 228, nos. 1–2, p. 233.
4. Fu, Z., Chen, J., Yin, D., Yin, D., Zhang, L., and
Zhang, Y., Catal. Lett., 2000, vol. 66, p. 105.
ACKNOWLEDGMENTS
5. Simonova, L.G., Barelko, V.V., Lapina, O.B., Pauksh-
tis, E.A., Terskikh, V.V., Zaikovskii, V.I., and
Bal’zhinimaev, B.S., Kinet. Katal., 2001, vol. 42,
no. 5, p. 762 [Kinet. Catal. (Engl. Transl.), vol. 42,
no. 5, p. 693].
We are grateful to Dr. Sci. (Chem.) E.A. Paukshtis
for helpful discussions of data on the electronic state of
copper in fiberglass catalysts.
1
1
6. Simonova, L.G., Barelko, V.V., Toktarev, A.V.,
Cherashev, A.F., Chumachenko, V.A., and Bal’zhini-
maev, B.S., Kinet. Katal., 2002, vol. 43, no. 1, p. 227
REFERENCES
1
2
. Costas, M. and Llobet, A., J. Mol. Catal. A: Chem.,
999, vol. 142, no. 2, p. 113.
. Silva, A.C., Fernandez, T.L., Carvalho, N.M.F.,
Herbst, M.H., Bordinhão, J., Horn, A., Jr., Wardell, J.L.,
Oestreicher, E.G., and Antunes, O.A.C., Appl. Catal.,
A, 2007, vol. 317, no. 2, p. 154.
[
Kinet. Catal. (Engl. Transl.), vol. 43, no. 1, p. 61].
1
7. Bal’zhinimaev, B.S., Barelko, V.V., Suknev, A.P., Pauk-
shtis, E.A., Simonova, L.G., Goncharov, V.B., Kiril-
lov, V.L., and Toktarev, A.V., Kinet. Katal., 2002,
vol. 43, no. 4, p. 586 [Kinet. Catal. (Engl. Transl.),
vol. 43, no. 4, p. 542].
3
. Kato, C.N., Hasegawa, M., Sato, T., Yoshizava, A.,
1
1
2
8. Lever, A.B.P., Inorganic Electronic Spectroscopy,
Inoue, T., and Mori, W., J. Catal., 2005, vol. 230, p. 226.
Amsterdam: Elsevier, 1984, 2nd ed.
4
. Tidahy, H.L., Siffert, S., Wyrwalski, F., Lamonier, J.-F.,
and Aboukais, A., Catal. Today, 2007, vol. 119, nos. 1–4,
p. 317.
. Salavati-Niasari, M., Hassani-Kabutarkhani, M., and
Davar, F., Catal. Commun., 2006, vol. 7, no. 12, p. 955.
9. Gang, L., van Grondelle, J., Anderson, B.G., and van
Santen, R.A., J. Catal., 1999, vol. 186, p. 100.
0. Moretti, G., Ferraris, G., Fierro, G., Lo Jacono, M., Mor-
purgo, S., and Faticanti, M., J. Catal., 2005, vol. 232,
p. 476.
5
6
. Kuzmin, A.O., Elizarova, G.L., Matvienko, L.G., Savi- 21. Montanari, B., Vaccari, A., Gazzano, M., Kaner, P.,
nova, E.R., and Parmon, V.N., Mendeleev Commun.,
998, p. 210.
. Lambert, S., Cellier, C., Gaigneaux, E.M., Pirard, J.-P.,
and Heinrichs, B., Catal. Commun., 2007, vol. 8, 22. Wong, S.-T., Lee, C.-H., Lin, T.-S., and Mou, C.-Y.,
p. 1244.
J. Catal., 2004, vol. 228, p. 1.
Papp, H., Pasel, J., Dziembaj, R., Makowski, W., and
Lojewski, T., Appl. Catal., B, 1997, vol. 13, nos. 3–4,
p. 205.
1
7
8
. Ray, S., Mapolie, S.F., and Darkwa, J., J. Mol. Catal. A: 23. Maurya, M.R., Chandrakar, A.K., and Chand, S., J. Mol.
Chem., 2007, vol. 267, nos. 1–2, p. 143.
Catal. A: Chem., 2007, vol. 274, p. 192.
KINETICS AND CATALYSIS Vol. 49 No. 4 2008