Oxidation of carbon monoxide over Cu- and Ag-NaY catalysts with aqueous hydrogen peroxide

Article abstract of DOI:10.1016/j.materresbull.2007.01.004

A novel oxidation reaction of CO with aqueous H₂O₂ over Cu-NaY (2-15 wt%) and Ag-NaY (5-15 wt%) catalysts has been achieved at low temperatures (55-70 °C) using a flow mode system. The employed catalysts were prepared by the incipient wetness impregnation of NaY zeolite (Si/Al = 5.6, surface area = 910 m²/g) with an aqueous solution of known concentrations of copper acetate and silver nitrate. Solids were subjected to thermal treatment at 300-450 °C prior to catalytic measurements unless subjected to subsequent reduction with hydrogen at 350 °C. The physicochemical characterization of the catalysts was probed using X-ray diffraction (XRD), FT-IR and combined thermal analyses TGA-DrTGA. The XRD data indicated that, the Ag particles have an ordered location in the sodalite cavity and the center of a single six-ring. The FT-IR data also proved the presence of a new peak at 1385 cm^-1 that is assigned to Ag-coordinated with the framework. A slow induced oxidation of CO (induction period, t_ind) took place at the initial stage of the CO oxidation reaction after which the reaction obeyed first-order kinetics. The utilized metal ions are proposed to be reduced to lower oxidation states such as Cu⁺ and Ag⁰ during the first period of reaction, t_ind, where the reaction proceeded favorably on such sites. Such argument was evidenced by carrying out the oxidation reaction over H₂-reduced Cu10-NaY and Ag10-NaY catalysts. The reduction caused a decrease in the t_ind, giving an evidence that the lower oxidation states Cu⁺ and Ag⁰ are the active sites in the studied oxidation reaction. The enhancement in catalytic activity was interpreted in terms of the facile adsorption of CO on the low oxidation state species.

Full text of DOI:10.1016/j.materresbull.2007.01.004

Materials Research Bulletin 42 (2007) 2170–2183
www.elsevier.com/locate/matresbu
Oxidation of carbon monoxide over Cu- and Ag-NaY
catalysts with aqueous hydrogen peroxide
*
Zeinhom Mohamed El-Bahy
Chemistry Department, Faculty of Science, Al-Azhar University, Nasr City, 11884 Cairo, Egypt
Received 21 October 2006; received in revised form 8 December 2006; accepted 5 January 2007
Available online 11 January 2007
Abstract
A novel oxidation reaction of CO with aqueous H2O2 over Cu-NaY (2–15 wt%) and Ag-NaY (5–15 wt%) catalysts has been
achieved at low temperatures (55–70 8C) using a flow mode system. The employed catalysts were prepared by the incipient wetness
impregnation of NaY zeolite (Si/Al = 5.6, surface area = 910 m2/g) with an aqueous solution of known concentrations of copper
acetate and silver nitrate. Solids were subjected to thermal treatment at 300–450 8C prior to catalytic measurements unless
subjected to subsequent reduction with hydrogen at 350 8C. The physicochemical characterization of the catalysts was probed using
X-ray diffraction (XRD), FT-IR and combined thermal analyses TGA–DrTGA. The XRD data indicated that, the Ag particles have
an ordered location in the sodalite cavity and the center of a single six-ring. The FT-IR data also proved the presence of a new peak at
1385 cmꢀ1 that is assigned to Ag-coordinated with the framework.
A slow induced oxidation of CO (induction period, tind) took place at the initial stage of the CO oxidation reaction after which the
reaction obeyed first-order kinetics. The utilized metal ions are proposed to be reduced to lower oxidation states such as Cu+ and Ag0
during the first period of reaction, tind, where the reaction proceeded favorably on such sites. Such argument was evidenced by
carrying out the oxidation reaction over H2-reduced Cu10-NaYand Ag10-NaY catalysts. The reduction caused a decrease in the tind
,
giving an evidence that the lower oxidation states Cu+ and Ag0 are the active sites in the studied oxidation reaction. The
enhancement in catalytic activity was interpreted in terms of the facile adsorption of CO on the low oxidation state species.
# 2007 Elsevier Ltd. All rights reserved.
Keywords: A. Oxides; C. Infrared spectroscopy; C. X-ray diffraction; D. Catalytic properties
1. Introduction
The catalytic oxidation of CO with molecular oxygen has been extensively studied over numerous catalysts. This
simple reaction is being utilized in an increasing number of applications. For instance, the CO oxidation reaction is
often an integral component of pollution control devices designed to reduce industrial and automotive emissions [1].
Air purification devices [2–4] and CO gas sensors [5–7] commonly employ CO oxidation catalysts. Nevertheless, the
development of low temperature CO oxidation catalysts has received a considerable attention [8–10]. Haruta and co-
workers have prepared supported gold catalysts on Mn2O3, Fe2O3, Co3O4, etc., to determine their catalytic activities
toward the oxidation of H2 and/or CO [7,8].
* Tel.: +20 24129790; fax: +20 22629356.
E-mail address: zeinelbahy2020@yahoo.com.
0025-5408/$ – see front matter # 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.materresbull.2007.01.004

Z.M. El-Bahy / Materials Research Bulletin 42 (2007) 2170–2183
2171
Many oxidation reactions in liquid phase using aqueous H2O2 as an oxidant have been of much attention in
presence of supported and non-supported catalysts [11–14]. H2O2 has been used in the oxidation of alcohols, ketons
[15–18] and amines [19–21] on titanium silicate (TS-1). The liquid phase ammoxidation of cyclohexanone over
Ti-zeolite catalyst in the presence of NH3 and H2O2 is an important process [22–24]. The conclusive oxidation
process of these reactions should lead to the production of CO gas. For instance, heterogeneous catalytic
epoxidation of olefins and parafins with oxygen usually gives poor selectivity because of the high reaction
temperature [14]. The low temperature oxidation of CO using H2O2 will be proficient in removing CO from gases in
hydrogen rich fuel produced from methanol in fuel cell. That will lead to avoid poisoning of Pt anode at low
temperatures (ca. 55–105 8C) and enhance the fuel cell performance. Under normal running conditions the product
hydrogen stream contains 0.5–1 vol.% CO. Thus, in order to obtain optimum performance, the total concentration
of CO in the gas stream should be reduced if possible to below 100 ppm. Only we have reported the oxidation of CO
with H2O2 in aqueous medium in the presence of Au- and Pt-NaY catalysts [25] to reduce the CO concentration in
liquid phase conditions.
The aim of this work is to prepare and characterize Cu-NaYand Ag-NaY catalysts to make use of the high surface
area of the NaY zeolite and the activity of Cu and Ag species in the decomposition of H2O2 and adsorption of CO
which has a great role in the CO oxidation reaction. Thus, we wish to report a new attempt to explore the oxidation of
CO with aqueous H2O2 in presence of Cu- and Ag-NaY catalysts and make advantage of their activity towards H2O2
decomposition and the facile adsorption of CO on such sites.
2. Experimental
Cu-NaY and Ag-NaY catalysts were prepared by the incipient wetness impregnation of NaY zeolite (Si/Al = 5.6,
surface area = 910 m2/g) with an aqueous solution of known concentrations of copper acetate and silver nitrate,
respectively, to obtain different Cu (2, 5, 10 and 15%) and Ag (5, 10, and 15%) metal loadings. The metal ion
concentrations were ensured by the conventional complexometric titration using EDTA complexone. The mixture was
refluxed at 80 8C with a continuous stirring for 3 h, washed thoroughly with distilled water, then subsequent air-dried
overnight at 120 8C, according to the method described elsewhere [23]. The calcination of the prepared samples
was performed in air at 400 8C for 5 h till constant weight (unless otherwise stated). The samples were referred to as
Mx-NaY, where M and x referred to the metal and wt% of M loading.
The reduction of samples was performed in a closed circulation system by introducing hydrogen (99.99% purity,
120 Torr) during installation of liquid nitrogen trap to prevent the re-oxidation of the reduced species by water evolved.
X-ray diffractograms were obtained using a Philips Analytical (type PW 1840) diffractometer. The patterns were
ꢀ1
˚
run with nickel-filtered Cu Ka radiation (l = 1.5405 A) at 30 kV and 20 mA with scanning speed of 28 in 2u min
The degree of relative crystallinity was calculated based on the five characteristic peak intensities at d-spacings of
.
˚
13.79, 5.60, 3.73, 3.27 and 2.89 A. The diffraction lines of reference NaY zeolite were used as the base (100%) for
comparing the percentage of zeolitic NaY structure in the overall metal-loaded samples [26].
The Fourier transform infrared (FT-IR) spectra were recorded on a Bruker (Vector 22), single beam spectrometer
with a resolution of 2 cmꢀ1. The samples were ground with KBr (1:100 ratio) as a tablet and mounted to the sample
holder in the cavity of the spectrometer.
Thermogravimetric analysis (TGA) and differential thermogravimetric analysis (DrTGA) were performed with a
Shimadzu-50 thermal analyzer. TGA and DrTGA curves were recorded with the following conditions: 9 mg; nitrogen
atmosphere of flow rate 30 ml minꢀ1; platinum crucible; ramping rate of 10 K minꢀ1 and a maximum temperature of
800 8C.
The oxidation of CO with aqueous H2O2 was carried out in a flow mode apparatus. In a typical reaction, 100 ml of
0.5 M H2O2 (BDH grade) was introduced into a double wall glass reactor. The gas mixture CO/N2 (1:1) was allowed to
pass through the reaction medium at a constant flow rate of 20 ml/min (unless otherwise stated) under constant stirring
conditions. The glass reactor was thermostated at the desired temperature by using temperature controller type (PEX-P
90). Then, 0.1 g of the catalyst was added to the reaction mixture supposing zero time. The kinetics of the reaction
were followed by introducing the gas mixture to a strong alkaline solution (NaOH) and measuring the change in its pH
value at interval times. The decrease in OHꢀ concentration with time, indicated by the decrease of pH value, was taken
as equivalent to the amount of CO2 evolved from the reaction. The natural logarithm of [OHꢀ] (ln C) was plotted
against time (t, min), applying the first-order rate equation: (kt = ln C0 ꢀ ln C) where, C0 and C are the initial OHꢀ

2172
Z.M. El-Bahy / Materials Research Bulletin 42 (2007) 2170–2183
concentration and its concentration at time t, respectively. The specific rate constant was used to express the activity of
catalysts, km:
rate consant ðkÞ
:
km
¼
catalyst weight
3. Results and discussion
3.1. X-ray diffraction
Powder diffractograms of different Cu- and Ag-NaY loading samples, along with parent NaY zeolite were recorded
as shown in Figs. 1 and 2, respectively. The diffraction lines characteristic of NaY at d-spacings of 13.79, 5.60, 4.71,
˚
4.32, 3.73, 3.27, 2.99, 2.89 and 2.84 A [PDF no. 38-0240], decreased with increasing metal loadings. Detailed analysis
of the X-ray diffraction of Cux-NaY and Agx-NaY showed evidence of CuO phase at d-spacings of 2.53, 2.33 and
˚
˚
1.87 A [PDF no. 48-1548]. Small lines at d-spacings of 2.51 and 2.30 A characteristic to Ag2O [PDF no. 75-1532]
were detected in the high Ag loading namely Ag10-NaY. This attenuation in the diffraction line intensities as
compared with those of NaY zeolite is caused by some loss of crystallinity evidently took place as shown in Table 1.
The crystallinity decreased especially in case of Ag10-NaY since it turned down to 58% and it was around 95% in case
of Cu10-NaY when compared with the parent NaY (taken at 100% crystallinity). This decrease can be a result of
the higher absorption coefficient of Cu and Ag species for the X-ray radiation [27] which increases with increasing
Cu-loading.
Fig. 1. XRD spectra of (a) parent NaY, (b) Cu5-NaY, (c) Cu10-NaY and (d) Cu15-NaY catalysts calcined at 400 8C.

Z.M. El-Bahy / Materials Research Bulletin 42 (2007) 2170–2183
2173
Fig. 2. XRD spectra of (a) parent NaY, (b) Ag5-NaY and (c) Ag10-NaY catalysts calcined at 400 8C.
Fig. 3 shows the change in the diffraction patterns of Cu10-NaY with the variation of calcination temperature
in the range of 350–400 8C. Elevating the calcination temperature leads to progressive increase in the intensity
˚
of CuO diffraction lines at d-spacings of 2.53, 2.33 and 1.87 A with a simultaneous decrease of the lines belonging
to NaY. This result undoubtedly shows a less effective interaction between surface CuO and NaY zeolite by
increasing temperature as due to increasing surface mobility of the latter phase and the agglomeration of CuO phase
on the NaY surface. The absence of a broad reflection in all samples between 2u value of 228 and 278, which would
indicate the presence of amorphous silica [28], demonstrates that the structure remains intact after the incorporation
of Cu and Ag inside the zeolite even with the decrease of crystallinity to a minimum value such as the case of Ag10-
NaY.
The unit cell parameters a0 for parent NaY and Cu- and Ag-Y samples has been determined by the help of
Rietveld method for optimizing the structural data and listed in Table 1. It is clear that the a0 value and the cell
volume has varied when compared with parent NaYas shown in Table 1. The cell volume of all samples is expected to
increase markedly when compared with the parent NaY suggesting the incorporation of Cu and Ag ions in the NaY
framework.
Table 1
Unit cell parameters, cell volume and crystallinity of the prepared samples
3
˚
Cell constant, a0 (A)
˚
Cell volume (A)
Catalyst
Crystallinity (%)
NaY
24.24
24.67
24.70
24.67
24.67
24.69
24.63
24.69
14,255
15,027
15,080
15,018
15,025
15,061
14,945
15,063
100
Cu5-NaY(400)
Cu10-NaY(350)
Cu10-NaY(400)
Cu10-NaY(450)
Cu15-NaY(400)
Ag5-NaY(400)
Ag10-NaY(400)
97.7
70.3
94.9
83.8
75.0
62.3
58.5

2174
Z.M. El-Bahy / Materials Research Bulletin 42 (2007) 2170–2183
Fig. 3. XRD spectra of (a) parent NaY and Cu10-NaY calcined at (b) 350, (c) 400 and (d) 450 8C.
It has been observed that empirical derived relationship exists between the relative I331, I311, and I220
peak intensities and cation location in faujasite type zeolites [29]. Cations are randomly distributed within the
lattice if I331 > I220 > I311, but if I331 > I311 > I220, the cations assume positions at sites Iꢀ, II. Using these
empirical criteria, the diffraction patterns of the prepared samples, except for Ag10-NaY (Figs. 1–3), suggest
that the metal ions are probably present in random positions within the NaY cavities. Whereas in case of
Ag10-NaY, Fig. 2, the order of the peak intensities was I331 > I220 > I311 and consequently, the ions are located
at site Iꢀ (sodalite cavity) and site II (the center of a single six-ring, S6R) or displaced from this point into a
supercage [30].
3.2. Mid-infrared spectra
The framework IR spectra of parent NaY zeolite and Cu-NaY with different Cu-loading samples are shown in
Fig. 4. The Cu-loaded samples exhibited bands of varying intensity and width at about 1167, 1025, 793, 724, 578 and
457 cmꢀ1 which are typical for NaY [31]. The bands at about 1025, 724 and 457 cmꢀ1 are caused by internal

Z.M. El-Bahy / Materials Research Bulletin 42 (2007) 2170–2183
2175
Fig. 4. FT-IR spectra of: (a) parent NaY, (b) Cu5-NaY, (c) Cu10-NaY and (d) Cu15-NaY catalysts calcined at 400 8C.
vibrations of (Si,Al)O4 tetrahedra of NaY, which tend to be insensitive to structure vibrations, whereas the bands at
about 1167, 793, 579 cmꢀ1 are due to vibrations related to external linkages between (Si,Al)O4 tetrahedra, and are
sensitive to the framework structure [32]. The structure-sensitive bands did not show significant changes in the
spectrum of Cu5-NaY sample as compared with those of NaY. On the other hand, the 578 cmꢀ1 absorption band
assigned to double ring polyhedra in the zeolite framework [33] decreased in intensity at Cu-loading of 15 wt%. This
indicates that a slight decomposition of the zeolite structure with increasing Cu concentration occurred. That is
confirmed also by the XRD data, since a decrease of the crystallinity was noticed up to 84% of the parent NaY. The
bands due to CuO, i.e. 610 and 500 cmꢀ1 [34] were obscured by the absorption bands of NaY, though the XRD analysis
showed free CuO crystallites grew at the external surface of zeolite.
From the recorded spectra of NaY, Ag5-NaY and Ag10-NaY, Fig. 5, it can be observed that, the spectra have the
same NaY characteristic sensitive and insensitive peaks. It is obvious also that, a new peak at 1385 cmꢀ1 is also
observed in Ag-NaY catalysts. This peak was not observed in either NaYor Cu-NaY spectra. The 1385 cmꢀ1 peak can
be assigned to coordinated framework silver [35]. These data are well correlated with the XRD data where, the
imbedding of Ag in the lattice in the sites Iꢀ and II and forms a coordinated structure leads to an increase of the cell
volume and a great decrease of the crystallinity of the Ag-loaded NaY catalysts.
Fig. 6 shows the IR spectra of Cu10-NaY precalcined in air at increasing temperatures, i.e. 350, 400 and 450 8C.
The bands characteristic of these samples resembled those of NaY zeolite, indicating no prominent decomposition of
the zeolite framework occurred during the heat treatment in the studied temperature range.

2176
Z.M. El-Bahy / Materials Research Bulletin 42 (2007) 2170–2183
Fig. 5. FT-IR spectra of: (a) parent NaY, (b) Ag5-NaY and (c) Ag10-NaY catalysts calcined at 400 8C.
3.3. Thermal analysis
The TGA and DrTGA data of Cu10-NaY showed that the TG curve consists of three successive temperature
regions, Fig. 7. These regions are centered at 100, 268 and 407 8C. The temperature regions were accompanied by
mass loss of 14.80, 12.01 and 1.14%, respectively. The total weight loss due to heating up to 800 8C attained
27.95%.
It is well known that NaY zeolite almost contains 20% as physisorbed water molecules. However, The first region
represents the loss of water of crystallization. The second region reflects the transformation of Cu-acetate to CuO
during heating up the sample. The XRD data of this sample showed that CuO is formed at a temperature close to
400 8C. So the second region is attributed to the decomposition of Cu-acetate and the formation of CuO at 268 8C,
Fig. 7. The third region may be due to surface dehydroxylation of OH groups attached to Al in zeolite lattice. However,

Z.M. El-Bahy / Materials Research Bulletin 42 (2007) 2170–2183
2177
Fig. 6. FT-IR spectra of: (a) parent NaY and Cu10-NaY calcined at (b) 350 8C, (c) 400 8C and (d) 450 8C.
the dehydroxylation of pure Al2O3 took place at higher temperatures, e.g. 800–900 8C. It seems likely that the
presence of CuO species lowered the dehydroxylation temperature that extended over long range.
The TGA and DrTGA of Ag10-NaYare shown in Fig. 8. The figure displays four decomposition regions centered
at 47, 104, 166 and 566 8C. The first region is due to loosely held water (1.3%), the second is due to the water of
crystallization (8%) compared to ꢁ15% in case of Cu10-NaY. The loss of H2O of crystallization in Ag10-NaYis less
than that of Cu10-NaY which may be attributed to the higher dispersion of Ag+ in zeolite cage as proved before by the
XRD where, the order of the peak intensities I331 > I220 > I311, the large decrease of crystallinity and the large lattice
volume were identified. In addition, the presence of the peak at 1385 cmꢀ1 in the FT-IR spectra of Ag-NaY catalysts
means the occlusion of Ag ions in the zeolite lattice. Although the decomposition of AgNO3 occurs at relatively
higher temperature, the third peak at 166 8C may be due to the decomposition of AgNO3 to the oxide form according
to TGA calculations of weight loss. A fourth peak in the range of 284–746 and centered at 566 8C is attributed to the
transformation of the phase Ag2O to Ag [36] and the removal of the OH groups from NaY lattice which is lower than
regular because of the presence of Ag species.

2178
Z.M. El-Bahy / Materials Research Bulletin 42 (2007) 2170–2183
Fig. 7. The TG and DrTGA profiles of Cu10-NaY.
Fig. 8. The TG and DrTGA profiles of Ag10-NaY.
3.4. CO oxidation with aqueous H2O2 over Cu- and Ag-NaY catalysts
3.4.1. Effect of calcination temperature of Cu10-NaY catalyst
The catalytic performance of a given catalyst is generally governed by a number of variables in the experimental
conditions such as calcination temperature. The correlation between the catalytic activity (expressed as km) calculated
from the first-order plots and the calcination temperatures is shown in Fig. 9. It is clear that an induction period (tind
)
was observed during the first stage of the reaction, after which the reaction obeyed the first-order kinetic equation. It
has been reported that the plot of H2O2 decomposition on bulk CuO did not display tind [37,38]. Therefore, tind might
be associated with CO chemisorption on charged Cu centers. It is noticed that the catalytic activity of Cu-NaY
catalysts increases gradually by raising the calcination temperature from 300 to 400 8C, and then decreases again with
increasing temperature. The maximum catalytic activity was observed for the catalyst calcined at 400 8C. That is in
agreement with the data obtained from thermal analysis and XRD. These results are in a harmony with the data
obtained for the CO oxidation with molecular oxygen over CuO–Al2O3, though the reaction was carried out at a higher
temperature such as 125–225 8C [39] and at 350 8C [40]. From preceding data, the oxidation of CO using H2O2 in
liquid phase occurs at lower temperatures, which may give an advantage for the use of H2O2 oxidation method over the
conventional molecular oxygen oxidation method. The thermal analysis presented the complete transformation of

Z.M. El-Bahy / Materials Research Bulletin 42 (2007) 2170–2183
2179
Fig. 9. The correlation between the km values and the calcination temperatures.
copper acetate to CuO around 300 8C. In addition, the XRD showed agglomeration of the CuO phase on the surface of
the catalyst which leads to the decrease of the number of active sites.
3.4.2. Effect of catalyst reduction with hydrogen
To demonstrate the effect of catalyst reduction on the catalyst activity, Cu10-NaYand Ag10-NaY were reduced at
350 8C using 120 Torr of H2 prior to the reaction. Fig. 10 shows the first-order plots of the reaction performed using
reduced and unreduced Cu10- and Ag10-NaYat the studied reaction conditions. It is obvious that tind is affected by the
mode of catalyst treatment, where it becomes 15 min in the reduced Cu10-NaY instead of 25 min in the unreduced
sample and it decreased to 50 min in the reduced Ag10-NaY instead of 60 min in the unreduced one. It is obvious that
the km was very close for both reduced and unreduced catalysts, indicating that the nature and amount of active sites in
the steady state are the same. This data is different from the data of CO oxidation over reduced Pt-NaY since the
catalytic activity of the reduced form was 50 times more than that of the unreduced form [25]. These differences may
be due to the fact that Pt ions are easy to be reduced more than Cu or Ag. Another observation may be taken is that
Cu-NaY catalyst is more active than that of Ag-NaY catalyst which may be due to the imbedding of Ag ions in the
coordination of the lattice structure but CuO is present mainly on the surface of the NaY zeolite and hence can be
readily in contact with the reactant species.
Fig. 10. First-order plots of CO oxidation with (0.5 M) H2O2 at 60 8C over reduced and unreduced (a) Cu10-NaY and (b) Ag10-NaY.

2180
Z.M. El-Bahy / Materials Research Bulletin 42 (2007) 2170–2183
Scheme 1. Proposed scheme for CO oxidation with H2O2 over Cu-NaY and Ag-NaY catalysts.
It has been shown that no stable carbonyls can be formed when CO is adsorbed on Cu(II) sites at room temperature
[41,42]. Consequently, the catalyst is reduced with the action of CO to lower oxidation state in the first stage of
reaction that is known as the induction period, tind, i.e. the first-order kinetics was subsequent to the formation of active
sites of Cu+ and Ag0 from Cu2+ and Ag+, respectively, Scheme 1. However, such lower oxidation state species are
stabilized by back p-donation [43–45] and may be considered as the active sites for this CO oxidation due to the facile
chemisorption of CO on lower oxidation states [46].
3.4.3. Effect of reaction temperature
The first-order plots of CO oxidation over employed catalysts (not shown) gave an impression that, the reaction
temperature slightly affected the activity of the catalyst during the first stage of the reaction, i.e. the temperature did not
affect the tind in the first stage of reaction but it increased the reaction rate in the steady state. The activation energy for
the CO oxidation reaction over the utilized catalysts was calculated from Arrhenius plots used in the temperature range
of 55–70 8C, Fig. 11 (ln km versus 1/T). The km values increased respectively with increasing reaction temperatures,
i.e. 1.02, 1.79, 2.51 and 4.79 minꢀ1 in case of Cu10-NaY and 0.249, 0.43, 0.67 and 1.21 minꢀ1 in case of Ag10-NaY.
The evaluated activation energy from Arrhenius plots was 46 kJ molꢀ1 for Cu10-NaYand 48 kJ molꢀ1 for Ag10-NaY.
The activation energy of CO oxidation reaction over Cu10-NaY is slightly less than that of CO oxidation reaction
over Ag10-NaY catalyst indicating that the CO oxidation reaction pathway is the same in both catalysts.
Fig. 11. Arrhenius plots of CO oxidation over: (upper) Cu10-NaY and (lower) Ag10-NaY catalyst.

Z.M. El-Bahy / Materials Research Bulletin 42 (2007) 2170–2183
2181
Fig. 12. The correlation between the GHSV (hꢀ1) and the reaction rate constant (km).
3.4.4. Effect of flow rate of reactants
The effect of feed gas flow rate on CO oxidation with aqueous H2O2 over Cu10-NaY catalyst was studied at 60 8C
under similar experimental conditions as described previously. The km values deduced from the first-order plots of CO
oxidation at different gas hourly space velocity (GHSV) ca. 12,000, 15,000 and 18,000 hꢀ1 are given in Fig. 12. A
progressive increase of km values was observed as GHSV decreases, where they amounted 0.82, 1.23 and 1.66 minꢀ1
at GHSV of 18,000, 15,000 and 12,000 hꢀ1, respectively. Such increase of the km values with decreasing the GHSV
values is interpreted in terms of the increase of the contact time of reactants on the catalyst surface.
Fig. 13. The correlation between the km values and (a) Cu- and (b) Ag-loadings in the oxidation of CO with (0.5 M) H2O2 at 60 8C.

2182
Z.M. El-Bahy / Materials Research Bulletin 42 (2007) 2170–2183
3.4.5. Effect of metal loading
The Effect of Cu- and Ag-loading on the catalytic activity of CO oxidation was studied and the relation between
km values and metal loading is shown in Fig. 13. In case of Cu-NaY catalysts, a pronounced increase in the curve
when Cu-loading increased from 2 to 5%, then passing through a slight increase with further increase in Cu-loading,
i.e. 10–15 wt%. These results indicate that CuO is well dispersed in the interior surface of NaY zeolite up to 5 wt% as
indicated by XRD technique, though the population of Cu centers was not high enough. This reflects the extension of
tind in the curves of these samples. Aggregates of CuO particles have been formed by increasing Cu-loading beyond
5% on the zeolite surface thereby a slight increase in km values was observed. That is supported by XRD data which
˚
demonstrates the increase of CuO lines at d-spacings of 2.53, 2.33 and 1.87 A. In case of Ag-NaY samples, the activity
did not change much with increasing the silver loading which may be due to the presence of Ag species inside the site
Iꢀ and II as mentioned and proved by XRD and FT-IR data.
4. Conclusions
The catalytic oxidation of CO over Cu- and Ag-NaY catalysts in aqueous H2O2 has been successfully done at low
temperatures in the range of 55–70 8C. The reaction temperature was much lower than that used for the oxidation of
CO by molecular oxygen. The reaction showed a period (tind) prior to the first-order kinetics. Reduction of Cu2+ and
Ag+ is necessary to enhance and complete the oxidation of CO to CO2 while CO is partially reducing the metal ions in
the beginning of reaction. The CO is then easily adsorbed on the reduced form and then being oxidized with the
adsorbed oxygen on the surface of the utilized catalyst. That was evidenced by the decrease of tind in the reduced
catalyst in both catalysts. Thus the activity of the catalyst is related to the transformation of higher oxidation states to
lower oxidation states. Coordinated Ag species are formed inside the supercage of NaY zeolite. On the other hand,
CuO forms aggregates on the surface of NaY zeolite which leads to the slight increase of the catalytic activity with
increasing Cu-loading.
References
[1] L. Zhao, X. Wang, Q. He, H. Wang, G. Sheng, L.Y. Chan, J. Fu, D.R. Blakec, Atmos. Environ. 38 (2004) 6177.
[2] J. Tamaki, C. Xu, N. Miura, N. Yamazoe, Chem. Lett. 441 (1990).
[3] T. Kobayashi, M. Haruta, H. Sano, M. Nakane, Sens. Actuators Soc Auto. Eng. 13 (1988) 339.
[4] T. Kobayashi, M. Haruta, H. Sano, Chem. Express 4 (1989) 217.
[5] P.G. Harrison, M.J. Wittet, Nature 337 (1988) 332.
[6] D.S. Stark, J. Harris, J. Phys. E: Sci. Instrum. 21 (1988) 715.
[7] M. Haruta, T. Kobayashi, N. Yamada, Chem. Lett. 405 (1987).
[8] M. Haruta, N. Yamada, T. Kobayashi, S. Lijima, J. Catal. 115 (1989) 301.
[9] A.K. Tripathi, V.S. Kamble, N.M. Gupta, J. Catal. 187 (1999) 332.
[10] D.R. Schryer, B.T. Upchurch, J.D. Van Norman, K.G. Brown, J. Catal. 122 (1990) 193.
[11] L. Formi, C. Oliva, A.V. Vishniakov, I.E. Ezerets, I.E. Mukovozov, F.P. Vatti, V.N. Zubkovskaja, J. Catal. 145 (1994) 194.
[12] H. Fu, S. Kaliaguine, J. Catal. 148 (1994) 540.
[13] K. Hafedh, F. Francois, J. Catal. 171 (1997) 420.
[14] X. Zuwei, Z. Ning, S. Yu, L. Kunlan, Science 292 (2001) 1139.
[15] M.F. Jose, I.G. Jose, A.M. Jose, V. Eugenio, J. Catal. 189 (2000) 40.
[16] T. Tatsumi, M. Nakamura, S. Negishi, H. Tominogo, J. Chem. Soc., Chem. Commun. 746 (1990).
[17] D.R.C. Huybrechts, L. De Bruycker, P.A. Jacobs, Nature 345 (1990) 240.
[18] A. Corma, P. Esteve, A. Martinez, S. Valencia, J. Catal. 152 (1995) 18.
[19] R. Joseph, T. Ravindranathan, A. Sudalai, Tetrahedron Lett. 36 (1995) 1903.
[20] S. Suresh, R. Joseph, B. Jayachandran, A.V. Pol, M.P. Vinod, A. Sudalai, H.R. Sonawane, T. Ravindranathan, Tetrahedron 51 (1995) 11305.
[21] L. Gang, G. Xinwen, W. Xiangsheng, Z. Qi, B. Xinhe, H. Xiuwen, L. Liwu, Appl. Catal. A 185 (1999) 11.
[22] A. Thangaray, S. Sivasanker, P. Ratnasmy, J. Catal. 131 (1991) 394.
[23] J.R. Sudahakar, S. Sivaskar, P. Ratnasmy, J. Mol. Catal. 69 (1991) 383.
[24] W. Peng, K. Takayuki, Y. Tatsuaki, J. Catal. 168 (1997) 400.
[25] T.M. Salama, Z.M. El-Bahy, F.I. Zidan, J. Mol. Catal. A 264 (2006) 128.
[26] T.M. Salama, I. Othman, M.S. Sirag, G.A. El-Shobaky, Micropor. Mesopor. Mater. 95 (2006) 312.
[27] M.M. Mohamed, N.A. Eissa, Mater. Res. Bull. 38 (2003) 1993.
[28] M.M. Mohamed, T.M. Salama, J. Colloid Interf. Sci. 249 (2002) 104.
[29] W.H. Quayle, J.H. Lunsford, Inorg. Chem. 21 (1982) 2226.
[30] M.N. Bae, M.K. Song, Y. Kim, Bull. Kor. Chem. Soc. 22 (2001) 1081.

Z.M. El-Bahy / Materials Research Bulletin 42 (2007) 2170–2183
[31] J. Scherzer, J.L. Bass, J. Catal. 28 (1973) 101.
2183
[32] R. Cid, F.J. Gil Llambfas, J.L.G. Fierro, A. Lopez Agudo, J. Villasenor, J. Catal. 89 (1984) 478.
[33] E.M. Flanigen, H. Khatami, H.A. Szymanski, Adv. Chem. Ser. 101 (1971) 201.
[34] J.R. Ferraro, Low-frequency Vibrations of Inorganic and Coordination Compounds, Plenum Press, NY, 1971.
[35] V. Sundaramurthy, N. Lingappan, J. Mol. Catal. A 60 (2000) 367.
[36] I.N.W. Geoffrey, A.B. Graham, B.M. James, Phys. Chem. Chem. Phys. 3 (2001) 3838.
[37] Z. Zhong, P. Hong, Z. Xin, S. Dai, Chem. Res. Chin. Univ. 8 (3) (1992) 327.
[38] N.A. Yousef, T. Farid, M.M. Selim, Afinedad 50 (47) (1993) 443.
[39] M.D. Nasr-Allah, Colloids Surf. A 207 (2002) 197.
[40] M.D. Nasr-Allah, Thermochim. Acta 401 (2003) 175.
[41] M. Kanttcheva, K. Hadjiivanov, A. Budneva, A. Davydov, Appl. Surf. Sci. 55 (1992) 49.
[42] B. Busca, J. Mol. Catal. 43 (1987) 225.
[43] K. Hadjiivanov, M. Kantcheva, D. Klissurski, J. Chem. Soc., Faraday Trans. 92 (1996) 4595.
[44] Y. Itho, S. Nishiyama, S. Tsuruya, M.J. Masai, Phys. Chem. 98 (1994) 960.
[45] G.J. Millar, C.H. Rochester, K.C. Waugh, J. Chem. Soc., Faraday Trans. 88 (1992) 1477.
[46] B. Jenshi, B. Wang, L. Sheng-Chin, H. Ta-Jen, Appl. Catal. A 232 (2002) 107.