Effect of liquid property on adsorption and catalytic reduction of nitrate over hydrotalcite-supported Pd-Cu catalyst

Article abstract of DOI:10.1016/j.molcata.2007.02.028

This study focused on investigating the effect of liquid property on adsorption and catalytic reduction of nitrate over hydrotalcite-supported Pd-Cu catalyst. Batch experiments were conducted under specific operating conditions. It was found that nitrate ions were adsorbed apparently by hydrotalcite-supported Pd-Cu catalyst at different reaction temperature (10, 25 and 35 °C). Adsorption isotherms for NO₃^- were described by Langmuir's equation. Higher reaction temperature accelerated nitrate adsorption and reduction, and simultaneously decreased the accumulation of NO₂^- and NH₄⁺. Nitrate adsorption and reduction were independent of the initial pH because of the high buffering capacity of hydrotalcite when the initial pH of solution was from 5.25 to 7.4. However, the dissolve of hydrotalcite-supported Pd-Cu catalyst at low initial pH (3.77) and higher concentration of OH^- ions at high initial pH (10.74) decreased nitrate adsorption and reduction. When using various nitrate salts as a source of nitrate ions, nitrate adsorption and reduction removal efficiency decreased in the order: Fe³⁺ > Mn²⁺ > Mg²⁺ > Ca²⁺ > Na⁺, which was attributed to different affinity toward OH^-. Co-existed cations in water with higher valence and lower crystal ionic radius and solubility constant of hydroxide showed stronger affinity toward OH^- and consequently increased the concentration of active sites accessible to NO₃^- and NO₂^-. The competitive adsorption of Cl^-, SO₄^2- and HCO₃^- with nitrate on the catalyst surface suppressed the adsorption and reduction of NO₃^-. Besides, few metals dissolved in neutral and alkaline solution. The activity of the hydrotalcite-supported Pd-Cu catalyst for nitrate reduction kept steady after repeated use.

Full text of DOI:10.1016/j.molcata.2007.02.028

Journal of Molecular Catalysis A: Chemical 272 (2007) 31–37
Effect of liquid property on adsorption and catalytic reduction
of nitrate over hydrotalcite-supported Pd-Cu catalyst
a,b
a,∗
a
Ying Wang , Jiuhui Qu , Huijuan Liu
a
State Key Laboratory of Environmental Aquatic Chemistry, Research Center for Eco-Environmental Sciences, Chinese Academy of Sciences,
8 Shuangqing Road, Haidian District, Beijing 100085, China
1
b
State Key Laboratory of Water Environment Simulation, School of Environment, Beijing Normal University, Beijing 100875, China
Received 10 January 2007; received in revised form 17 February 2007; accepted 19 February 2007
Available online 23 February 2007
Abstract
This study focused on investigating the effect of liquid property on adsorption and catalytic reduction of nitrate over hydrotalcite-supported
Pd-Cu catalyst. Batch experiments were conducted under specific operating conditions. It was found that nitrate ions were adsorbed apparently
◦
−
were described by
by hydrotalcite-supported Pd-Cu catalyst at different reaction temperature (10, 25 and 35 C). Adsorption isotherms for NO
Langmuir’s equation. Higher reaction temperature accelerated nitrate adsorption and reduction, and simultaneously decreased the accumulation of
3
−
+
NO
2
and NH . Nitrate adsorption and reduction were independent of the initial pH because of the high buffering capacity of hydrotalcite when
4
the initial pH of solution was from 5.25 to 7.4. However, the dissolve of hydrotalcite-supported Pd-Cu catalyst at low initial pH (3.77) and higher
−
concentration of OH ions at high initial pH (10.74) decreased nitrate adsorption and reduction. When using various nitrate salts as a source of nitrate
3
+
2+
2+
2+
+
ions, nitrate adsorption and reduction removal efficiency decreased in the order: Fe > Mn > Mg > Ca > Na , which was attributed to different
−
affinity toward OH . Co-existed cations in water with higher valence and lower crystal ionic radius and solubility constant of hydroxide showed
stronger affinity toward OH and consequently increased the concentration of active sites accessible to NO
−
−
−
3
and NO
2
. The competitive adsorption
−
2−
−
−
of Cl , SO
4
and HCO
3
with nitrate on the catalyst surface suppressed the adsorption and reduction of NO
3
. Besides, few metals dissolved in
neutral and alkaline solution. The activity of the hydrotalcite-supported Pd-Cu catalyst for nitrate reduction kept steady after repeated use.
2007 Elsevier B.V. All rights reserved.
©
Keywords: Liquid property; Pd-Cu/hydrotalcite; Catalytic; Adsorption; Nitrate reduction
1
. Introduction
the presence of residual organics and the possible increase in
chlorine demand in purified water [3] limit the application of
this process.
The nitrate ions removal from groundwater is becom-
ing an environmental problem of major concern due to its
high toxicity (e.g. causing cancer and nethemoglobinemia)
The catalytic reduction of nitrate is attracting considerable
interest due to many advantages, such as efficient chemical
reaction, flexibility, simple configuration of reactor, etc. In this
process, nitrate ions are reduced using hydrogen over bimetal-
lic catalyst. The reaction obeys a consecutive reaction scheme
in which nitrite appears as an intermediate, while nitrogen and
ammonia are the final products. Recently, developing a catalyst
possessed higher activity and selectivity for nitrate reduction
[9–11] and studying the influence of liquid properties [12–14],
such as temperature, pH, co-existed ions, etc., have received
many researchers’ attention.
[
1,2]. Conventional physicochemical methods, including ion-
exchange, reverse osmosis and electrodialysis [3,4], allow
effective removal of nitrate ions from water by concentrating
them in a secondary waste stream, but do not generally achieve
their complete disposal. The most environmental friendly meth-
ods for nitrate removal, based on selective reduction of nitrate
to nitrogen, are biological digestion [5,6] and catalytic reduc-
tion [7,8]. The biological denitrification process is currently the
most widely used, however, possible bacterial contamination,
In our previous study [15], hydrotalcite (HT)-supported
Pd-Cu catalysts for nitrate reduction using hydrogen were inves-
tigated preliminarily. Hydrotalcite like compounds, as one of
the most promising materials, were widely used as catalyst [16],
∗
Corresponding author. Tel.: +86 10 62849151; fax: +86 10 62923558.
E-mail address: jhqu@mail.rcees.ac.cn (J.H. Qu).
1
381-1169/$ – see front matter © 2007 Elsevier B.V. All rights reserved.
doi:10.1016/j.molcata.2007.02.028

3
2
Y. Wang et al. / Journal of Molecular Catalysis A: Chemical 272 (2007) 31–37
catalyst support [17], adsorbent [18], etc. The general chemical
Thermogravimetry (TG) and differential thermal analysis
(DTA) of the sample was carried out using a Labsys TG-DSC
II
III
x+ n−
x−
TM
formula is written as [M 1−xM x(OH)2] [A x/n] ·yH2O,
II
III
n−
where M and M are divalent and trivalent metal ions, A is
Simultaneous Analyser referenced against recalcined alumina
(SETRAM Co.), in flowing nitrogen (flow rate, 20 mL/min), at
a n-valent anion and x can have values approximately between
◦
0
.25 and 0.33. This material presents a brucite-like structure,
a heating rate of 10 C/min.
where metallic cations are located in the layer, while anions
and water are found in the interlayer space. Therefore, Pd-Cu
active metals could be introduced into the layers to obtain high
dispersion, which may increase the stability and activity of the
catalyst; toxic anions (such as nitrate) in contaminated water
may be concentrated in the interlayer space to be removed.
Our preliminary experiments have shown that hydrotalcite-
supported Pd-Cu catalyst possessed effective adsorptive and
catalytic capacity for nitrate in water [15]. The reaction scheme
for nitrate removal on HT3(Pd-Cu) catalyst, with H2 pro-
vided, was a consecutive and dynamic adsorption and catalytic
hydrogenation process. Then, it is a key to investigate the
influence of liquid property, including temperature, pH and
co-existed ions in water, on nitrate adsorption and reduction
over hydrotalcite-supported Pd-Cu catalyst, which will be favor-
able for the determining the appropriate reaction conditions for
nitrate reduction. From this aspect, systematical experiment was
necessary to conducted in this paper.
2.3. Adsorptive and catalytic tests
For all adsorptive tests in this work, the studies were car-
ried out in glass vessels with agitation provided by a shaker at
170 rpm and the temperature controlled by air bath. Nitrate solu-
tion (50 mL) was mixed with 0.1 g catalyst. Samples were taken
at different time intervals. The initial nitrate concentration was
equal to 100 mg/L. The adsorption equilibrium was obtained
within 180 min, which was demonstrated in our previous study
[15]. Consequently, the adsorption time was 180 min.
Catalytic capacity of different catalyst for nitrate reduction
was tested in a thermostated batch reactor equipped with H2 inlet
and outlet and a sample port. The catalyst (1 g) was suspended
in pure water (500 mL), which was saturated with the mixture
of argon (400 mL/min) and hydrogen (200 mL/min) from the
titanium plate situated in the bottom of the reactor for 60 min.
Solution containing nitrate was introduced and the time was
started. The initial nitrate concentration was equal to 100 mg/L.
The reaction time was 180 min.
2
. Experimental
2
.1. Catalyst preparation
2.4. Analysis methods
Hydrotalcite-supported Pd-Cu catalyst was prepared by co-
Samples were taken from the reactor at desired sampling
−
precipitation at low supersaturation method [19]. In this method,
two solutions, A and B, were added dropwise into a beaker con-
taining100 mLofdeionicwaterwhilevigorousstirring. Solution
A was Mg(NO3)2 (1.2 mol/L), Al(NO3)2 (0.4 mol/L), an appro-
priate amount of Pd(NO3)2·2H2O and Cu(NO3)2·3H2O mixed
aqueous solution. The palladium and copper content are 1 and
times and filtered through a 0.45 ␮m membrane. NO3 -N,
−
+
NO2 -N and NH4 -N were determined using a Hitachi-3010
model UV-spectrophotometer.
After the reaction, solution in the reactor was analyzed by
ICP-AES (Perkin-Elmer Co.) to quantify any dissolving of the
2+
3+
2+
2+
active metals (Mg , Al , Pd , Cu ).
0
.25 wt.%, respectively. Solution B contained 1.65 mol/L NaOH
and 0.5 mol/L Na2CO3. During the process of synthesis, the
pH of the suspensions was maintained at about 10. The result-
ing suspension was then maintained at 25 C, with stirring, for
3. Results and discussion
◦
3.1. Characterization of catalyst
4
h. The product was filtered, washed thoroughly with deionised
water until the pH of filtrate showed neutral and subsequently
The physical properties of the catalyst were very important
and influenced the catalytic capacity of the catalyst remark-
ably. Before the calcination, the sample had a low specific
◦
◦
dried overnight at 105 C and calcined at 550 C for 8 h, finally
◦
reduced at 200 C for 2 hr under flowing hydrogen/argon. The
2
final product was mentioned as HT3(Pd/Cu).
surface area of 6.2 m /g. The X-ray diffraction (XRD) anal-
ysis showed that a well-crystallized hydrotalcite-like phase
2
.2. Catalyst characterization
formed. After calcination, the specific surface area increased to
2
2
36.6 m /g. Meanwhile, hydrotalcite-like structure was changed
◦
The specific surface area (BET method) was determined
while MgO phase was formed according to the peaks at 43.2
and 62.7 . After the sample contacted with the aqueous solution
◦
using an ASAP2000 Surface Analyser (Micromeritics Co.,
USA) using N2 as the adsorbate.
X-ray powder diffraction (XRD) patterns of samples were
obtained with a Bruker diffractometer using Cu K␣ radiation
from 10 to 70 (in 2Θ).
Fourier transform infrared spectra were recorded with potas-
sium bromide-pressed disks, by accumulating 32 scans at
of nitrate, the specific surface area dropped to the value as low
2
as 9.05 m /g. Simultaneously, MgO phase collapsed and turned
into a hydrotalcite-like phase which was evidenced by the XRD
analysis. This result sample was just the active catalyst during
the denitrification process. In addition, due to the low active
metals content, no peak related with Pd-Cu could be observed.
In order to clarify the decomposition process during the cal-
cination, simultaneous thermogravimetry (TG) and differential
◦
◦
−1
−1
4
5
cm resolution between 400 and 4000 cm using a Nicolet
700 Fourier transform infrared spectrometer (FTIR).

Y. Wang et al. / Journal of Molecular Catalysis A: Chemical 272 (2007) 31–37
33
Fig. 1. Simultaneous TG/DSC thermogram of the HT3(Pd-Cu).
thermal analysis techniques were carried out. The result was
shown in Fig. 1. As TG analysis revealed, gradual weight loss
◦
◦
was observed from about 50 C to approximately 600 C, with
◦
two main endothermic effects at about 173 and 410 C. The
first large endothermic effect at about 173 C involved inter-
layer water loss [20]; the second endothermic effect at 410 C
◦
◦
accounted for hydrotalcite decomposition which was accompa-
nied by dehydroxylation and the decomposition of CO32 in the
brucite-like layers [20]. Carbon dioxide evolution induced the
formation of pores, which was the reason for the drastic increase
in the specific surface area. The total weight loss (ca. 40% of the
starting mass) is similar to that calculated from weights taken
before and after calcination.
−
3
.2. Effect of liquid property on adsorption and catalytic
reduction of nitrate on HT3(Pd-Cu)
For the natural water, the property of the liquid phase was
obviously various in different area. Therefore, the effect of the
liquid property on the adsorption and catalytic hydrogenation
for nitrate over hydrotalcite-supported Pd-Cu catalyst should be
discussed in detail.
3
.2.1. The effect of temperature
The influence of temperature on nitrate uptake and catalytic
hydrogenation over HT3(Pd/Cu) was discussed, respectively.
−
Fig. 2(a) showed adsorption isotherm for NO3 at tempera-
◦
ture of 10, 25 and 35 C. The result showed that HT3(Pd/Cu)
catalyst possessed apparent adsorption capacity for NO3 , espe-
cially at 25 and 35 C, which could be explained by considering
−
◦
Fig. 2. The effect of temperature on the adsorption and reduction of nitrate
−
over HT3(Pd-Cu), without pH control: (a) adsorption isotherms of NO3 -N;
the capacity of hydrotalcite to concentrate anions in the inter-
layer space. Although the specific surface area of the regenerated
hydrotalcite was very low, nitrate ions could be concentrated by
ionic forces between the layers. Therefore, effective adsorption
for nitrate was obtained.
−
(
b) NO3 -N concentration with time during the catalytic hydrogenation of
−
+
−
aqueous nitrate solution; (c) NO2 and NH4 concentration with NO3 conver-
sion during the catalytic hydrogenation of aqueous nitrate solution (X
−
=
NO3
−
−
−
([NO3 ]t=0 − ([NO3 ]t)/([NO3 ]t=0).
Additionally, adsorption data were tested using Langmuir
model. It was discovered that, at different temperature, adsorp-
−
tion isotherm for NO3 fit well with Langmuir’s equation

3
4
Y. Wang et al. / Journal of Molecular Catalysis A: Chemical 272 (2007) 31–37
(
qe = qmbce/(1 + bce)), where qe (mg/g) was the amount of
−
NO3 -N adsorbed at equilibrium, qm (mg/g) the maximum
sorption capacity corresponding to complete monolayer cov-
erage, ce (mg/L) the equilibrium solute concentration and b
was the equilibrium constant (L/mg). The maximum adsorp-
tion capacity (qm) obtained by the linear regression was
− ◦ − ◦
.685 mg-NO3 -N/g at 10 C, 9.441 mg-NO3 -N/g at 25 C,
3
1
−
◦
5.491 mg-NO3 -N/g at 35 C, respectively, which suggested
that the nitrate uptake obviously increased with an increase
in the adsorption temperature. Usually, the uptake decreases
with temperature due to the exothermic nature of the simple
physical adsorption reaction. The present result indicated that
a chemical interaction involved the adsorption of nitrate on
HT3(Pd/Cu).
Fig. 2(b) showed the catalytic reduction of nitrate at differ-
◦
ent temperature (10, 25 and 35 C). It was found that nitrate
concentration decreased rapidly at the initial 2 min, especially
◦
Fig. 3. The effect of the initial pH on the adsorption and reduction of nitrate
over HT3(Pd-Cu), T = 25 C.
at 35 C nitrate removal efficiency reaching as high as 93.5%.
◦
−
Then, NO3 removal efficiency became slowly gradually. After
1
80 min, the total removal efficiency for nitrate was 64.37, 96.76
◦
ment of the final pH from adsorption process (about 10.8) to
reduction process (about 12).
and 98.78% at 10, 25 and 35 C, respectively. Thus, it was con-
cluded that higher reaction temperature may increase nitrate
removal.
Simultaneously, nitrite and ammonium ions were formed dur-
ing the catalytic hydrogenation for nitrate. The results were
shown in Fig. 2(c). With the decrease of the reaction temper-
ature, the concentration of NO2 and NH4 increased, that is,
the selectivity for N2 decreased. At 35 C, the concentration
of NO2 and NH4 formed at the same NO3 conversation
were the lowest. At 25 C, nitrite concentration increased first,
going through a maximum at 88% of nitrate conversion and
then decreased, which indicated that nitrite was an intermedi-
ate product of the reduction of nitrate. Ammonium increased
steadily because it was a final product. At 10 C, more NO3
was reduced to NH4 , and an increased accumulation occurred
in NO2 change. Consequently, the results obtained indicated
that higher temperatures favor the increase of nitrate reduction
efficiency and selectivity for N2.
Itwasconcludedthat, from5.25to7.4, bothnitrateadsorption
and catalytic reduction processes were almost pH independent.
Simultaneously, the relationship between the initial and equilib-
rium pH showed that, the final pH remained constant at about
10.8 for nitrate adsorption process and at 12 for nitrate reduction
process, respectively, in spite of obvious changes of the initial
pH. These results demonstrated that hydrotalcite-like compound
possessed pH buffering capacities at the initial pH of from 5.25
to 7.4, which was due to the release of hydroxyl during the regen-
eration of hydrotalcite structure. Therefore, at initial pH of from
−
+
◦
−
+
−
◦
5.25 to 7.4, nitrate adsorption (or nitrate reduction) was carried
◦
−
out at only one pH and, as a consequence, the nitrate removal
remained constant irrespective to the initial pH of the solution
due to the buffering effect of the HT3(Pd/Cu). Other authors
have also observed the buffering effect of hydrotalcite-like com-
+
−
2
−
pounds. You et al. [21] observed that the amount of SeO3
adsorption in Mg-Al and Mg-Fe hydrotalcites was almost con-
stant in a wide range of pH.
3
.2.2. The effect of pH
At too low initial pH of 3.77, a significant decline in nitrate
adsorption removal efficiency may be due to the dissolution
of HT3(Pd/Cu), which was confirmed by the examination of
dissolution ions by ICP-AES. At too high initial pH (10.74),
buffering effect of HT3(Pd/Cu) disappeared. High concentra-
tion of OH ions cause the competitive effect with NO3 , which
decreased nitrate removal efficiency for adsorption or reduction
and selectivity to N2.
The effect of the initial pH on nitrate adsorption and reduc-
tion over HT3(Pd/Cu) was shown in Fig. 3. Diluted HNO3 and
NaOH solutions were used for pH adjustments. Buffer solutions
were not used in order to avoid contamination of foreign anions.
The results indicated that, in the process of nitrate adsorption,
the nitrate removal remained practically constant from 5.25 to
−
−
7
.4, reaching over 47%. However, a significant decline occurred
when the initial pH was 3.77 and 10.74. In the process of nitrate
catalytic reduction, at the initial pH of from 5.25 to 10.74, nitrate
removal decreased slightly with the increasing of pH. There was
3.2.3. The effect of co-cations in water
+
no obviously different in the final NH4 concentration. However,
To study the influence of familiar anions in the aque-
ous solution on the adsorption and reduction of nitrate over
HT3(Pd/Cu), the following reagent-grade compounds were
employed as a source of nitrate ions: NaNO3, Mg(NO3)2,
Ca(NO3)2, Mn(NO3)2 and Fe(NO3)3. As shown in Fig. 4,
nitrate adsorption removal efficiency decreased in the order:
−
the final NO2 concentration increased with the increasing of
−
pH, especially for pH of 10.74, NO2 -N concentration reached
as high as 4.85 mg/L, which was about three times higher than
that at pH of 7.4. During the process of nitrate reduction to N2,
−
the electroneutrality of the system was maintained by the OH
3+
2+
2+
2+
+
anions produced, which was supported by the observed incre-
Fe > Mn > Mg > Ca > Na , which was on the contrary

Y. Wang et al. / Journal of Molecular Catalysis A: Chemical 272 (2007) 31–37
35
of the catalyst can be quickly withdrawed. Moreover, the lower
the solubility constant of hydroxide precipitation was, the easier
−
the combination of OH ions with cations to form the precip-
itation was. For the cations employed in this experiment, the
solubility constant of hydroxide precipitation according to the
order:
Fe(OH)3 < Mn(OH)2 < Mg(OH)2 < Ca(OH)2 < NaOH
−
38
−13
;
(
Fe(OH)3: Ksp = 4 × 10
;
11
Mn(OH)2: Ksp = 1.9 × 10
−
−6
Mg(OH)2: Ksp = 1.8 × 10 ; Ca(OH)2: Ksp = 5.5 × 10
at
298 K), which was just contrary to the order of the nitrate and
nitrite removal efficiency. That is, more hydroxide precipitation
−
made more OH in the solution reduced, and thus improved
nitrate removal efficiency and decreased the nitrite accumula-
tion. Additionally, the lower final pH in the presence of bivalent
−
and trivalent cations also demonstrated that more OH ions in
the solution were removed.
3
.2.4. The effect of co-anions in water
−
2−
−
Cl , SO4
and HCO3 are common co-existing anions
−
together with NO3 in water. The effect of these anions in
water on the adsorption and catalytic reduction of nitrate by
HT3(Pd/Cu) was studied in Fig. 5. Obviously, the presence of
Cl , SO4 and HCO3 anions suppressed the adsorption and
reduction of NO3 . The inhibitory effect was in the order of
HCO3 > SO4 > Cl . It was observed that there was only
a little decrease on the adsorption and reduction of NO3 at
the concentration range of 0–0.01 mol/L Cl ions. When the
concentration of Cl ions was higher than 0.01 mol/L, the
−
2−
−
−
−
2−
−
−
−
−
Fig. 4. The effect of the co-cations on the (a) adsorption and (b) reduction of
nitrate over HT3(Pd-Cu), T = 25 C, without pH control.
◦
inhibitory effect increased apparently. Generally, the concentra-
−
tion of Cl ions in the groundwater or surface water was lower
−
than 0.01 mol/L. So Cl ions in natural water exhibited little
to the order of final pH decrease. Similarly, in the process of
nitrate catalytic reduction, the order of nitrate catalytic reduc-
tion removal efficiency was consistent with the order of nitrate
adsorption removal efficiency. The final pH in the presence of
effect on nitrate adsorption and reduction over HT3(Pd/Cu) cat-
−
2−
alyst. However, the effect of HCO3 and SO4 on the removal
−
of NO3 by adsorption and reduction was significant at the
concentration range of 0–0.1 mol/L. With the increasing concen-
+
−
2−
−
Na was apparently higher than others, which was the same
tration of HCO3 and SO4 , the amounts of NO3 removal
by adsorption and reduction decreased quickly. For example,
as the result during the process of adsorption. Simultaneity,
−
+
−
2−
the final NO2 and NH4 concentrations were obviously dif-
in the presence of HCO3 or SO4 at 0.001 mol/L, the nitrate
removal by HT3(Pd/Cu) during the catalytic hydrogenation pro-
cess dropped from 97 to about 86%; at a concentration of
0.05 mol/L HCO3 or SO4 , the nitrate removal declined to
about 70%; while at a concentration of 0.01 mol/L, the removal
of nitrate decreased to 46%. Additionally, during the catalytic
−
ferent for various nitrate salts. The final NO2 concentration
decreasedwiththeincreaseofnitratecatalyticreductionremoval
efficiency, which meant favorable effect on nitrate catalytic
reduction advanced nitrite reduction alike. Additionally, the
−
2−
+
+
NH4 accumulation was the lowest in the presence of Na .
It has been widely known that the nitrate reduction efficiency
and the selectivity to N2 are severely limited by diffusion [22].
−
2−
−
−
hydrogenation process, the presence of Cl , SO4 or HCO3
increased the nitrite and ammonia accumulation. That is, Cl ,
−
2−
−
During the process of nitrate catalytic reduction, OH ions were
SO4 or HCO3 as co-existed anions inhibited the reduction
of nitrite and decreased the selectivity to N2. The inhibitory
produced continuously and baffled the access of nitrate and
nitrite to the surface of the catalyst. With this respect, one can
−
2−
−
order of HCO3 > SO4 > Cl was similar to that for nitrate
adsorption and reduction. It was found that the inhibitive effect
−
conclude that faster removal of OH ions from the surface of
−
catalyst would lead to faster regeneration of active sites. In other
words, a large number of Pd-Cu sites can be accessed by nitrate
and nitrite in the subsequent reaction cycles. Accordingly,
different effect of cations on nitrate and nitrite hydrogenation
of HCO3 ions on the adsorption and reduction of nitrate was
higher than that of SO4 and Cl , which might be due to the
identical structures of NO3 and HCO3 ions. They are both
planar, and the angles between the N O and C O bonds are
2−
−
−
−
−
◦
might be ascribed to various affinities toward OH , which was
equal to 120 .
−
2−
−
related with physical properties of cations, such as valence,
crystal ionic radius and solubility constant of hydroxide. With
the increase of valence and the decrease of ionic radius, attrac-
To clarify the inhibitory effect of Cl , SO4 or HCO3
on nitrate adsorption and reduction, we took the influence of
SO4 ions, for example. During nitrate reduction process,
infrared spectra of the catalyst at different reaction time was
2−
−
−
tion forces towards OH improved. Thus, OH on the surface

3
6
Y. Wang et al. / Journal of Molecular Catalysis A: Chemical 272 (2007) 31–37
Fig. 6. Infrared spectra of HT3(Pd-Cu) catalyst withdrawn at different reaction
◦
2−
times, T = 25 C, without pH control, the concentration of SO4 is 0.01 mol/L.
ually. Therefore, by IR study, the effect of SO42 on nitrate
reduction over HT3(Pd/Cu) catalyst could be attributed to com-
petitive adsorption with nitrate on the catalyst surface, which
leads to the active sites decreased apparently, and consequently
nitrate adsorption and reduction decreased and nitrite was accu-
mulated.
−
3.3. The stability and reusability of HT3(Pd-Cu)
From a practical viewpoint, to evaluate the stability and
reusability of HT3(Pd-Cu) must be required. After the reduc-
tion of nitrate, the dissolution of metals in solution at different
initial pH was analyzed by ICP-AES (Table 1). In the neutral (pH
7.4) and alkaline (pH 10.74) solution, no palladium ions were
detected in the solution. The concentrations of copper, magne-
sium and aluminium ions were lower than 0.0025, 0.477 and
0.58 mg/L, which were all lower than the standard for drink-
ing water in China. However, in acid (pH 3.77) solution, the
dissolution of metals increased apparently. The concentrations
of palladium and copper reached as high as 5.8 and 1.24 mg/L,
which accounted for 29 and 24.8% of the amount loaded on
the catalyst, respectively. The concentrations of magnesium
and aluminium ions were higher than 200 mg/L. These results
demonstrated that HT3(Pd-Cu) catalyst dissolved in acid solu-
tion, but keep stable in neutral and alkaline solution. Generally,
most waters were about neutral or alkalescent. Therefore, the
reduction of nitrate on HT3(Pd-Cu) catalyst may be used in
most case.
Fig. 7 showed average activity of HT3(Pd-Cu) for nitrate
reduction during five cycles. The experiment was conducted as
follows. After the reaction, the catalyst was taken out by the
centrifugation, and then washed to remove excess ions on the
surface of the catalyst. Whereafter, the recycled catalyst was
−
2−
−
Fig. 5. The effect of the co-anions (Cl , SO4 and HCO3 ) on the adsorption
◦
and reduction of nitrate over HT3(Pd-Cu), T = 25 C, without pH control: (a)
nitrate removal efficiency with different co-anions concentration during adsorp-
−
+
−
tion and reduction process; (b) NO2 and NH4 concentration with NO3
conversion at different co-anions concentration during nitrate reduction process
−
− ([NO3−]t)/([NO3−
(
X
−
= ([NO3
]
]
).
NO3
t=0
t=0
studied in Fig. 6. We can see that, when SO42 ions occurred in
−
nitrate solution or not, it presented two infrared bands at 1647
Table 1
−1
and 1385 cm assigned to the bending O H vibration of the
water molecules and to the asymmetric stretching N O vibra-
tion of the nitrate ions, respectively. However, in nitrate solution
−
The dissolution of HT3(Pd-Cu) at different initial pH (C0(NO3 ) = 100 mg/L,
◦
2
g/L catalyst, at 25 C)
pH
Pd (mg/L)
Cu (mg/L)
Mg (mg/L)
Al (mg/L)
with SO42 ions, the existence of an apparent infrared band
−
−
1
3.77
5.8
–
–
1.24
0.0025
–
>200
>200
0.39
0.58
at 1109 cm can be found, but not be observed in the solu-
2
−
7.40
0.74
0.477
0.0051
tion without SO4 ions. Therefore, it was concluded that the
1
–
−
1
2−
band at 1109 cm was assigned to SO4 . Moreover, during
−1
: lower than detective level by ICP-AES.
the course of the reaction, this band (1109 cm ) increased grad-

Y. Wang et al. / Journal of Molecular Catalysis A: Chemical 272 (2007) 31–37
37
and lower crystal ionic radius and solubility constant of hydrox-
−
ide showed stronger affinity toward OH and consequently
−
increased the concentration of active sites accessible to NO3
and NO2 . Thus, when using various nitrate salts as a source of
−
nitrate ions, nitrate adsorption and reduction removal efficiency
3
+
2+
2+
2+
+
decreased in the order: Fe > Mn > Mg > Ca > Na . The
−
2−
−
presence of Cl , SO4
and HCO3 anions suppressed the
−
adsorption and reduction of NO3 on HT3(Pd-Cu), which could
be attributed to competitive adsorption with nitrate on the cat-
alyst surface. Additionally, few metals dissolved in neutral and
alkaline solution. The activity of HT3(Pd-Cu) for nitrate reduc-
tion kept steady after repeated use, which favored the long-term
use.
Acknowledgements
−
Fig. 7. Reusability of HT3(Pd-Cu) for nitrate reduction for an initial NO3
◦
concentration of 100 mg/L, 2 g/L catalyst, at 25 C, without pH control.
This work was supported by the Funds for Creative Research
Groups of China (Grant No. 50621804) and National High Tech-
nology Research and Development Program of China (Grant No.
◦
dried at 60 C and weighed. Finally, the capacity of the recy-
cled catalyst for nitrate reduction was examined again. It was
observed that the average activity decreased obviously in the
first cycle and then decreased slowly with the increase of cycles,
which indicated that HT3(Pd-Cu) catalyst exhibited excellent
stability and may be used for a long term.
2006AA06Z302).
References
[
[
1] A. Pintar, Catal. Today 77 (2003) 451–465.
2] M. Super, H.V. Heese, D. MacKenzie, W.S. Dempster, J. Plessis, J.J. Fer-
reira, Water Res. 15 (1981) 1265–1270.
3
.4. Conclusions
[
3] A. Santaf e´ -Moros, J.M. Goz a´ lvez-Zafrilla, J. Lora-Garc ´ı a, Desalination
185 (2005) 281–287.
[
4] J. Wi s´ niewski, A. R o´ z˙ a n´ ska, Desalination 191 (2006) 210–218.
In this paper, the effect of liquid property on adsorption
[
5] H.S. Moon, S.W. Chang, K. Nam, J. Choe, J.Y. Kim, Environ. Pollut. 144
and catalytic reduction of nitrate over hydrotalcite-supported
Pd-Cu catalyst were investigated systematically. The results
demonstrated that temperature, pH and co-existed ions in water
showed apparent influence on nitrate removal. At different
(
2006) 802–807.
6] V. Laurin, N. Labb e´ , P. Juteau, S. Parent, R. Villemur, Water Res. 40 (2006)
836–1840.
[7] K.D. Vorlop, T. Tacke, Chem. Ing. Tech. 61 (1989) 836–845.
8] U.M. Meytal, M. Sheintuch, Catal. Today 102 (2005) 121–127.
[9] A.E. Palomares, J.G. Prato, F. Marquez, A. Corma, Appl. Catal. B: Environ.
1 (2003) 3–13.
[
1
◦
[
reaction temperature (10, 25 and 35 C), HT3(Pd/Cu) catalyst
−
possessed apparent adsorption capacity for NO3 , which could
4
be explained by considering the capacity of hydrotalcite to con-
centrate anions in the interlayer space. Adsorption isotherms for
[
[
10] A. Garron, K. L a´ z a´ r, F. Epron, Appl. Catal. B: Environ. 65 (2006) 240–248.
11] A. Garron, K. L a´ z a´ r, F. Epron, Appl. Catal. B: Environ. 59 (2005) 57–69.
[12] G. Centi, S. Perathoner, Appl. Catal. B: Environ. 41 (2003) 15–29.
−
NO3 were good agreement with Langmuir’s equation. More-
[
13] L. Lemaignen, C. Tong, V. Begon, R. Burch, D. Chadwick, Catal. Today
5 (2002) 43–48.
over, Higher reaction temperature advanced nitrate adsorption
and reduction, and simultaneously decreased the accumulation
7
[
[
14] A. Pintar, J. Batista, J. Levec, Chem. Eng. Sci. 56 (2000) 1551–1559.
15] Y. Wang, J.H. Qu, H.J. Liu, The Second Japan–China Workshop on Envi-
ronmental Catalysis and Eco-materials, Fukuoka, Japan, 2005.
−
+
of NO2 and NH4 . When the initial pH of solution was from
.25 to 7.4, the initial pH exhibited little influence on the nitrate
5
adsorption and catalytic reduction, which was due to high pH
buffering capacities of HT3(Pd-Cu) catalyst. However, at too
low initial pH of 3.77, the dissolution of HT3(Pd/Cu) resulted
in a significant decline in nitrate adsorption removal. At too high
initial pH of 10.74, buffering effect of HT3(Pd/Cu) disappeared.
[16] T.M. Jyothi, T. Raja, B.S. Rao, Mol. Catal. A: Chem. 168 (2001) 187–191.
[
[
[
[
17] F. Basile, G. Fornasari, M. Gazza, Appl. Clay Sci. 16 (2000) 185–200.
18] S. Yata, Clays Clay Miner. 31 (1983) 305–311.
19] F. Cavani, F. Trifiro, A. Vaccari, Catal. Today 11 (1991) 173–301.
20] D. Tichit, M.H. Lhouty, A. Guida, B.H. Chiche, F. Figueras, A. Auroux,
D. Bartalini, E. Garrone, J. Catal. 151 (1995) 50–59.
−
High concentration of OH ions decreased nitrate adsorption
and reduction. Co-existed cations in water with higher valence
[21] Y. You, G.F. Vance, H. Zhao, Appl. Clay Sci. 20 (2001) 13–25.
22] U. Prusse, K.D. Vorlop, J. Mol. Catal. A: Chem. 173 (2001) 313–328.
[