Combining the photocatalyst Pt/TiO2 and the nonphotocatalyst SnPd/Al2O3 for effective photocatalytic purification of groundwater polluted with nitrate

Article abstract of DOI:10.1021/cs5003564

We investigated photocatalytic reduction of NO₃^- in real groundwater in the presence of the photocatalyst Pt/TiO₂ and the nonphotocatalyst SnPd/Al₂O₃, which were dispersed in the groundwater, under irradiation at λ > 300 nm, with glucose as a hole scavenger. In this system, photocatalytic H₂ evolution (2H ⁺ + 2e^- → H₂) proceeded over Pt/TiO ₂, and nonphotocatalytic, that is, conventional catalytic, reduction of NO₃^- with H₂ (NO₃^- + 5/2H₂ → 1/2N₂ + 2H₂O + OH^-) occurred over SnPd/Al₂O₃. NO₃^- (1.0 mmol dm^-3) in the groundwater completely and selectively decomposed to N₂ (yield 83%) after 120 h with a 300 W Xe lamp (λ > 300 nm) over the Pt/TiO₂-SnPd/Al₂O₃ system in combination with photooxidative pretreatment of the groundwater over Pt/TiO ₂ to decompose organic compounds. The decomposition rate of NO ₃^- in the groundwater was still slower than that in an aqueous NO₃^- solution even after the pretreatment of the groundwater. The lower photocatalytic performance was due to poisoning of Pt/TiO₂ with sulfate and silicate ions and poisoning of SnPd/Al ₂O₃ with polymerized silicate ions. On the other hand, cations, including Na⁺, K⁺, Mg²⁺, and Ca ²⁺, in the groundwater did not affect the photocatalytic and catalytic performances of the system. Sulfate ions adsorbed on the Pt sites on Pt/TiO₂, where H₂ evolution occurs, and silicate ions deactivated the oxidation sites on TiO₂ by reacting with the surface hydroxyl groups, leading to a decline in the photocatalytic performance of Pt/TiO₂.

Full text of DOI:10.1021/cs5003564

Research Article
pubs.acs.org/acscatalysis
Combining the Photocatalyst Pt/TiO2 and the Nonphotocatalyst
SnPd/Al2O3 for Effective Photocatalytic Purification of Groundwater
Polluted with Nitrate
Jun Hirayama† and Yuichi Kamiya*,‡
‡
†Graduate School of Environmental Science, Research Faculty of Environmental Earth Science, Hokkaido University, Nishi 5, Kita
10, Kita-ku, Sapporo 060-0810, Japan
S
* Supporting Information
−
ABSTRACT: We investigated photocatalytic reduction of NO3 in real
groundwater in the presence of the photocatalyst Pt/TiO2 and the
nonphotocatalyst SnPd/Al2O3, which were dispersed in the groundwater,
under irradiation at λ > 300 nm, with glucose as a hole scavenger. In this
system, photocatalytic H2 evolution (2H+ + 2e− → H2) proceeded over Pt/
TiO2−, and nonphotocatalytic, that is, conventional catalytic, reduction of
−
NO3 with H2 (NO3 + 5/2H2 → 1/2N2 + 2H2O + OH−) occurred over
SnPd/Al2O3. NO3 (1.0 mmol dm−3) in the groundwater completely and
−
selectively decomposed to N2 (yield 83%) after 120 h with a 300 W Xe lamp
(λ > 300 nm) over the Pt/TiO2−SnPd/Al2O3 system in combination with
photooxidative pretreatment of the groundwater over Pt/TiO2 to decompose
−
organic compounds. The decomposition rate of NO3 in the groundwater
−
was still slower than that in an aqueous NO3 solution even after the
pretreatment of the groundwater. The lower photocatalytic performance was
due to poisoning of Pt/TiO2 with sulfate and silicate ions and poisoning of SnPd/Al2O3 with polymerized silicate ions. On the
other hand, cations, including Na+, K+, Mg2+, and Ca2+, in the groundwater did not affect the photocatalytic and catalytic
performances of the system. Sulfate ions adsorbed on the Pt sites on Pt/TiO2, where H2 evolution occurs, and silicate ions
deactivated the oxidation sites on TiO2 by reacting with the surface hydroxyl groups, leading to a decline in the photocatalytic
performance of Pt/TiO2.
KEYWORDS: photocatalysis, nitrate reduction, platium-modified titanium dioxide, groundwater purification, tin−palladium bimetal,
hydrogenation
−
reduction of NO3 in real groundwater. Westerhoff et al. first
INTRODUCTION
■
−
reported the photocatalytic reduction of NO3 in groundwater
−
Pollution of groundwater with nitrate (NO3 ) as a result of
intensive agricultural activities using nitrogen fertilizers, human
sewage, and industrial effluents is a serious global problem.1
in 2012.15 They use TiO2 to photocatalytically reduce NO3− in
−
groundwater under UV irradiation. Although NO3 decom-
poses in the presence of formic acid, the conversion decreases
from 51% in aqueous NO3 solution to 37% in groundwater.
To the best of our knowledge, the photocatalytic reduction of
NO3 to N2 with high selectivity in real groundwater has not
been reported so far.
Photocatalytic reactions over semiconductor photocatalysts,
such as TiO2 and WO3, proceed via three sequential
reactions:20,21 (1) photoexcited electrons (e−) and positive
holes (h+) are produced in the bulk of the semiconductor
photocatalyst by photoabsorption, (2) e− and h+ migrate to the
surface, (3) e− and h+ undergo reduction and oxidation,
respectively, with the substrates adsorbed on the surface, and
(4) unreacted e− and h+ recombine, which does not cause any
−
−
NO3 is reduced to nitrite (NO2 ) in the human body, which
causes methemoglobinemia in infants (i.e., blue-baby syn-
drome) and is a possible precursor of carcinogenic nitros-
amine.2 Thus, the World Heath Organization recommends that
−
−
−
the concentration of NO3 in drinking water should be below
0.8 mmol dm−3, which corresponds to 50 mg dm−3.3 Since
groundwater is an important fresh water resource that is
−
indispensable to human society, NO3 needs to be removed
from polluted groundwater.
In recent years, photocatalytic reduction of NO3 in water
−
has been extensively investigated as a promising technology to
remediate NO3 -polluted groundwater.
Guan’s group9 and
4−19
−
Kominami’s group10 have reported that Ag/TiO2 and CuPd/
TiO2, respectively, show high selectivity to N2 in the
−
−
photocatalytic reduction of NO3 in aqueous NO3 solutions
under UV irradiation in the presence of a hole scavenger.
However, there are only a few reports on the photocatalytic
Received: March 18, 2014
Revised: May 23, 2014
© XXXX American Chemical Society
2207
dx.doi.org/10.1021/cs5003564 | ACS Catal. 2014, 4, 2207−2215

ACS Catalysis
Research Article
chemical reactions to occur with the substrates. Observed rates
of the chemical reactions, i.e., rates of reactant consumption
and of product formation, depend on the rates of these three
reactions. The reaction rates of (1) and (3) positively influence
the observed rate, whereas the reaction rate of (4) negatively
influences it.
reaction rates. However, the biocatalysts and conventional
catalysts perform their reactions successively.
We have previously reported that the photocatalytic
−
−
reduction of NO3 in an aqueous NO3 solution is effectively
and selectively promoted under light irradiation in the presence
of a photocatalyst (Pt/TiO2 or Pt/SrTiO3:Rh18) and a
17
nonphotocatalyst (SnPd/Al2O3). In this photocatalytic system,
H2 is formed by a photocatalytic reaction over Pt/TiO2 (eq 1),
Generally, semiconductor photocatalysts themselves, i.e.,
unmodified bare photocatalysts, show low photocatalytic
activities. However, the photocatalytic activity is dramatically
increased when a small amount of metal is added.14,22 There are
a few reasons for the increase in the photocatalytic activity. One
reason is that e− are trapped on the modified metal, leading to
elongation of the charge-separated state of e− and positive
holes, which decreases the rate of (4). Another reason is that
the modified metal becomes an adsorption site for the
reactants, resulting in an increase in the rate of (3). However,
addition of metal into the photocatalyst may increase the rate of
(4) because the excess metal acts as recombination sites for e−
and h+.7,9 Therefore, an optimal balance among the loading
amount, kind, crystalline structure, location, and particle size of
the metal must be obtained to afford high-performance
photocatalysts.
−
and the formed H2 thermally reduces NO3 in water over
SnPd/Al2O3 (eq 2):
2H+ + 2e− → H2
(1)
NO3− + 5/2H2 → 1/2N2 + 2H2O + OH−
(2)
The combined photocatalytic system shows much better
photocatalytic performance in terms of both activity and
selectivity than does TiO2 directly modified with Sn−Pd
particles. The photocatalyst (Pt/TiO2) and nonphotocatalyst
(SnPd/Al2O3) can be designed separately to maximize their
performance levels for the photocatalytic (eq 1) and non-
photocatalytic (eq 2) reactions, respectively.
In the present study, we investigated the photocatalytic
reduction of NO3− in real groundwater in the presence of both
the photocatalyst Pt/TiO2 and the nonphotocatalyst SnPd/
Al2O3 under light irradiation (λ > 300 nm). We chose glucose
as a hole scavenger for the photocatalytic reaction because it is
a readily available organic compound and has a high efficiency
of H2 production by photocatalysis. Moreover, the effects of
compounds in the groundwater on the photocatalytic and
nonphotocatalytic performance levels over Pt/TiO2 and SnPd/
Al2O3, respectively, were systematically investigated, and we
have proposed a guideline for the remediation of real
In some photocatalytic reactions, not only the loading
amount and particle size but also the elemental species of metal
strongly affect the photocatalytic performance. Photocatalytic
reduction of NO3− in water over a semiconductor photocatalyst
is a quite typical case that the elemental species of metal
strongly affects the photocatalytic performances. Semiconduc-
tor photocatalysts modified with silver and Cu−Pd particles
have high activity and high selectivity to N2 in the
9,10,14
−
photocatalytic reduction of NO3 ,
but Pt- and Rh-
modified ones show extremely low activities in the photo-
−
groundwater by photocatalytic reduction of NO3 .
5,14
−
catalytic reduction of NO3 ,
because Ag and Cu−Pd form
−
adsorption sites for NO3 , which become activated. Since the
EXPERIMENTAL METHODS
−
■
metal particles activate NO3 via adsorption, they must have a
Preparation of Catalysts. AEROXIDE TiO2 P25 (Evonik)
was used as a TiO2 photocatalyst. TiO2 was modified with 0.5
wt % Pt by using photodeposition. TiO2 (2 g) was dispersed in
distilled water (135 cm3), and then CH3OH (15 cm3, Wako
Pure Chem., Ind., Ltd.) and H2PtCl6·6H2O (1.25 cm3, 0.04 mol
dm−3, Wako Pure Chem., Ind., Ltd.) were added to the
suspension. The suspension, which was in a Pyrex glass cell, was
purged with a stream of N2 (15 cm3 min−1) for 30 min and
then irradiated using a 300 W Xe lamp (λ > 300 nm, USHIO
Inc., Optical Modulex) for 3 h with stirring. The suspension
was centrifuged to separate the catalyst powder, and the
resulting supernatant solution was replaced with distilled water
(200 cm3). The suspension was stirred for a few minutes and
centrifuged again. This process was repeated three times.
Finally, the catalyst powder was dried in air at 333 K overnight.
The obtained catalyst is denoted as Pt/TiO2.
high surface area for high photocatalytic activity. In addition,
the modified metal must help in the charge separation of the
e−−h+ pairs. If the two functions are maximized simultaneously
by optimizing the loading amount, elemental species, crystalline
structure, location, and particle size of the metals, extremely
high-performance semiconductor photocatalysts can be ob-
tained. However, it is usually difficult to achieve this with
semiconductor photocatalysts modified directly with metals,
because excess metal provides the recombination sites for e−
and h+.
Recently, reaction systems in which various catalytic
functions are distributed to separate catalysts present in the
same solution have been actively investigated to maximize the
performance of catalysts. Artificial Z-scheme-type photo-
catalytic water splitting, which affords H2 and O2, is an
example of function-distribution on separate catalysts. In the
artificial Z-scheme photocatalytic system, O2 and H2 evolution
sites are built on separate semiconductor photocatalysts (e.g.,
WO3 for O2 evolution and TaON for H2 evolution), and two
catalysts are added to the reaction solution with a redox shuttle
between the two catalysts.23−27 The combination of a
biocatalyst and a conventional catalyst (metal complexes and
precious metal-supported catalysts are examples for the latter)
has been reported to be a highly efficient system.28,29 In this
reaction system, the biocatalyst promotes difficult reactions
with high chemo-, regio-, and stereoselectivities, and conven-
tional catalysts perform relatively simple reactions with high
The nonphotocatalytic catalyst, 2.3 wt % Sn-4.2 wt % Pd/
Al2O3 (molar ratio of Sn/Pd was 0.5, denoted as SnPd/Al2O3),
was prepared by using an incipient wetness method. Al2O3
(AEROXIDE, Alu C, Evonik) was heated in air at 523 K for 4 h
before use. An aqueous solution of PdCl2 (7.38 cm3, 0.112 mol
dm−3, Wako Pure Chem., Ind., Ltd.) was dropped onto the
Al2O3 (2.0 g), and then the resulting wet solid was dried in air
at 373 K overnight, followed by calcination in air at 523 K for 3
h. An aqueous solution of SnCl2·2H2O (2.31 cm3, 0.172 mol
dm−3, Wako Pure Chem., Ind., Ltd.) was dropped onto the
resulting solid, and then the wet solid was dried in air at 373 K
overnight, followed by calcination in air at 523 K for 3 h. TiO2
2208
dx.doi.org/10.1021/cs5003564 | ACS Catal. 2014, 4, 2207−2215

ACS Catalysis
Research Article
Table 1. Concentrations of the Components Contained in the Groundwater and Reaction Solutions Used in this Study
concentration/mmol dm−3
i
Cl−
SO4
SiOxn−
NH4
Na+
K+
Mg2+
Ca2+
TOC
−
2−
+
reaction solution
initial pH
NO3
a
GW
8.1
8.1
7.8
5.3
5.5
5.4
5.5
9.8
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
0.5
0.5
0.5
0
1.1
1.1
1.1
0
1.1
0.7
0.7
0
<0.01
1.0
1.0
0
0.1
0.1
3.0
1.1
1.0
2.0
3.0
3.0
0.7
0.7
0
1.6
1.6
0
3.3
0.2
0.2
0.2
0
b
GW-Ox
<0.01
c
GW-Ox-Cation-removal
0
d
GW-Ox-Anion-removal
<0.01
1.0
0
0.7
0
1.6
0
e
KNO3-aq
0
0
0
0
0
0
0
f
KNO3−Cl−-aq
1.0
0
0
0
0
0
0
0
g
KNO3−SO42−-aq
1.0
0
0
0
0
0
0
h
KNO3−SiO32−-aq
0
1.0
0
0
0
0
a
b
c
Groundwater obtained from a well in Kitami, Hokkaido, Japan. GW treated with photocatalytic oxidation in the presence of Pt/TiO2. GW from
which cations were removed by passing GW through a cation-exchange resin, followed by treatment with photocatalytic oxidation to decompose
d
organic matter. GW from which anions were removed by passing GW through an anion-exchange resin, followed by treatment with photocatalytic
−
oxidation to decompose organic matter. After treatment, KNO3 was added to the water to adjust the concentration of NO3 to 1.0 mmol dm−3.
e
f
g
h
i
Aqueous KNO3 solution. Aqueous KNO3 solution with KCl. Aqueous KNO3 solution with K2SO4. Aqueous KNO3 solution with K2SiO3. Total
organic carbon (ppm).
modified with 2.3 wt % Sn and 4.2 wt % Pd was prepared by
using a procedure similar to that for SnPd/Al2O3 and is
denoted as SnPd/TiO2.
solution was further treated with the photocatalytic oxidation,
like GW-Ox was. The anions were removed from GW (GW-
Ox-Anion) by using a procedure similar to that for GW-Ox-
Cation, but an anion exchange resin (ORGANO Co.,
AMBERLITE IRA410 OH AG) and CO2 were used instead
of the cation-exchange resin and KOH, respectively. To adjust
Just before the photocatalytic and nonphotocatalytic
reactions, SnPd/Al2O3 was reduced with NaBH4 (Wako Pure
Chem, Ind., Ltd.). NaBH4 (molar ratio of NaBH4/(Sn + Pd) =
10) was added to an aqueous suspension (30 cm3) of SnPd/
Al2O3, and the suspension was stirred at room temperature for
30 min. The catalyst powder was filtered and washed with
distilled water (∼200 cm3) to afford the reactants. SnPd/TiO2
was reduced by using a procedure similar to that for SnPd/
Al2O3.
−
the concentration of NO3 to 1.0 mmol dm−3, KNO3 (Wako
Pure Chem, Ind., Ltd.) was added to the solution.
KNO3 was dissolved in distilled water to afford KNO3-aq.
The concentration of NO3− in the solution was adjusted to 1.0
mmol dm−3. KNO3−Cl−-aq, KNO3−SO42−-aq, and KNO3−
SiO32−-aq were prepared by adding KCl (Wako Pure Chem.,
Ind., Ltd.), K2SO4 (Wako Pure Chem., Ind., Ltd.), and K2SiO3
(Wako Pure Chem., Ind., Ltd.), respectively, to KNO3-aq.
Component Analysis of Groundwater. Groundwater
−
polluted with NO3 was obtained from a well in Kitami,
−
Hokkaido, Japan (denoted as GW). Concentrations of the
anions and cations in the groundwater were determined by
using two ion-chromatographs (Tosoh Co., IC-2001). Ana-
lytical conditions are described in detail in our previous
report.18 The concentrations of Si species in the groundwater
were determined by using inductively coupled plasma atomic
emission spectroscopy (ICP-AES) on a Shimadzu ICPS-7000.
The Si species present in the groundwater were anions (silicate
ion, SiOxn−) because no Si was detected in the groundwater
that was treated with an anion-exchange resin. The amounts of
the organic compounds in the groundwater were measured by
using a total organic carbon (TOC) analyzer (Shimadzu Co.,
TOC-5000A). Components contained in the groundwater
(GW) are listed in Table 1.
Photocatalytic Reduction of NO3 in Water. Photo-
catalytic reduction of NO3− in water was carried out in a Pyrex
reaction vessel connected to a closed gas circulation system. Pt/
TiO2 and SnPd/Al2O3 were suspended in the reaction solution
(250 cm3, [NO3 ] = 1.0 mmol dm−3) containing glucose (1.0
−
mmol dm−3, Wako Pure Chem., Ind., Ltd.), and the suspension
was stirred using a magnetic stirrer. The reaction suspension
was thoroughly degassed and then exposed to He (101.3 kPa).
A 300 W Xe lamp covered with Pyrex glass (λ > 300 nm) was
used as the light source. The evolved gases were analyzed using
an online gas chromatograph (Agilent Technology Inc., 3000 A
Micro GC, He carrier) with columns of molecular sieves (for
H2, O2, and N2) and Porapak Q (for N2O). The concentrations
−
−
+
of NO3 , NO2 , and NH4 in the reaction solution were
determined by using two ion chromatographs.
Reaction Solutions. The reaction solutions and their
compositions are listed in Table 1. GW-Ox stands for GW
pretreated with photocatalytic oxidation in the presence of Pt/
TiO2 (500 mg dispersed in 300 cm3 of GW) to decompose the
organic compounds. The suspension in a Pyrex glass cell was
purged with a stream of air (100 cm3 min−1) and then
irradiated with a 300 W Xe lamp for 24 h. The catalyst was
separated from the suspension by filtration. TOC decreased
from 3.3 in GW to 0.2 ppm in GW-Ox by the treatment. The
Nonphotocatalytic (conventional catalytic) Reduction
of NO3− with H2 in Water. Nonphotocatalytic reduction, that
−
is, conventional catalytic reduction, of NO3 with H2 in water
was carried out in a batch reactor at 298 K in the dark. The
−
reaction solution (250 cm3, [NO3 ] = 1.0 mmol dm−3, 14 mg
N dm−3) containing SnPd/Al2O3 (10 mg) was purged with a
stream of He (30 cm3 min−1) for 30 min. Then the gas was
changed to a mixture of H2 (0.5 atm) and CO2 (0.5 atm) to
start the reaction. A small portion of the reaction solution was
periodically withdrawn and analyzed by using two ion
n−
concentrations of the other components besides SiOx were
n−
not affected by the treatment. The concentration of SiOx
decreased from 1.1 to 0.7 mmol dm−3.
chromatographs to determine the concentrations of NO3 ,
−
−
+
The cations were removed by passing GW through a glass
column with a cation-exchange resin (ORGANO Co.,
AMBERLYST 70), followed by addition of KOH to adjust
the pH of the solution to 7.8 (GW-Ox-Cation). The obtained
NO2 , and NH4 .
Photocatalytic H2 and O2 Evolutions. Photocatalytic H2
evolution from the reaction solution with glucose over Pt/TiO2
was carried out by using a procedure similar to that for the
2209
dx.doi.org/10.1021/cs5003564 | ACS Catal. 2014, 4, 2207−2215

ACS Catalysis
Research Article
a
−
Table 2. Photocatalytic Reduction of NO3 in Groundwater (GW) under Light Irradiation
selectivity [%]
−
+
entry
catalyst
Pt/TiO2
glucose
conversion [%]
NO2
NH4
N2
amnt of H2 detected in gas phase/μmol
1
2
3
4
5
6
7
absence
presence
presence
presence
presence
absence
presence
<1
<1
0
−
−
−
3
−
−
−
−
−
−
60
Pt/TiO2
227
d
SnPd/Al2O3
n.d.
Pt/TiO2+SnPd/Al2O3
Pt/TiO2+SnPd/Al2O3
Pt/TiO2+SnPd/Al2O3
23
0
22
75
18
b
d
−
−
0
−
−
38
−
−
62
n.d.
d
<1
n.d.
c
SnPd/TiO2
5
21
a
−
Reaction conditions: catalyst weight, Pt/TiO2, 500 mg and SnPd/Al2O3, 100 mg; reactant NO3 , 1.0 mmol dm−3 with 1.0 mmol dm−3 glucose;
b
c
reaction volume, 250 cm3; light irradiation (λ > 300 nm) and reaction time, 12 h. Under dark conditions. 2.3 wt % Sn-4.2 wt % Pd/TiO2. The
amount of the catalyst loaded in the reactor was 500 mg. Not detected.
d
−
□
○
Figure 1. Time courses for photocatalytic reduction of NO3 in (Δ) GW, ( ) GW-Ox, and ( ) KNO3-aq in the copresence of Pt/TiO2 and SnPd/
Al2O3. (A) Conversion of NO3 , (B) selectivity to N2 + N2O, and (C) selectivities to NO2− (closed symbols) and NH4+ (open symbols). Reaction
−
conditions: catalyst weight, Pt/TiO2, 500 mg and SnPd/Al2O3, 100 mg; reactant NO3− (from KNO3), 1.0 mmol dm−3 with 1.0 mmol dm−3 glucose;
reaction volume, 250 cm3; light irradiation (λ > 300 nm).
−
photocatalytic reduction of NO3 in water, as described above,
but in the absence of SnPd/Al2O3. In addition, photocatalytic
O2 evolution from the reaction solution with AgNO3 (10 mmol
dm−3, Wako Pure Chem., Ind., Ltd.) was conducted by using a
procedure similar to that for the photocatalytic H2 evolution.
To prevent the formation of silver colloid, the pH of the
reaction solution was adjusted to 8.0 by adding HNO3 to the
solution with AgNO3 and K2SiO3.
the gas phase was larger than that for entry 1. SnPd/Al2O3
alone (entry 3) showed no photocatalytic or catalytic activity
for the decomposition of NO3− or for the evolution of H2 even
under light irradiation.
−
On the other hand, relatively high NO3 conversion was
obtained when both Pt/TiO2 and SnPd/Al2O3 (Pt/TiO2−
SnPd/Al2O3) were dispersed in the reaction solution in the
presence of glucose (entry 4). In addition, the formation of
−
undesirable NO2 was greatly suppressed (3% selectivity), and
Adsorption Measurement. The amount of adsorption of
SiO32− on Pt/TiO2 at 303 K was measured as follows. Pt/TiO2
desired N2 selectively formed (75% selectivity at 23%
−
(30 mg) was added to an aqueous SiO3 solution (10 cm3),
2−
conversion of NO3 ), though the selectivity may change with
further conversion. At the time, only N2 formed as a gaseous
nitrogen-containing product. When the reaction was conducted
which was prepared from K2SiO3, and the suspension was
heated at 303 K with stirring. After 24 h, Pt/TiO2 was filtered
off, and the concentration of SiO3 in the filtrate was
−
2−
in the dark, the conversion of NO3 was negligible even using
Pt/TiO2−SnPd/Al2O3 (entry 5). Thus, photocatalysis is
determined using ICP-AES. The initial concentrations of
−
SiO3 were varied from 0.1 to 8.0 mmol dm−3. Adsorption
2−
responsible for the decomposition of NO3 in entry 4. When
the reaction was conducted in the absence of glucose under
light irradiation (entry 6), NO3− did not decompose, and no H2
evolved in the gas phase. Although H2 evolved during the
photocatalytic reaction over Pt/TiO2 even in the absence of
glucose (entry 1), no photocatalytic reactions, including the
reached equilibrium within several hours.
RESULTS AND DISCUSSION
■
−
Photocatalytic Reduction of NO3 in Groundwater
−
(GW). Table 2 summarizes the data from the photocatalytic
decomposition of NO3 and H2 evolution, occurred when Pt/
−
reduction of NO3 in groundwater (GW) samples under light
TiO2 and SnPd/Al2O3 were copresent in GW (entry 6). This is
probably because the organic compounds are removed from
GW by adsorption on SnPd/Al2O3, which consequently
quenches the photocatalytic H2 evolution over Pt/TiO2.
TiO2 directly modified with SnPd particles exhibited
irradiation for 12 h. Pt/TiO2 alone in the absence of glucose
(entry 1) showed negligible activity for the decomposition of
−
NO3 , whereas H2 was produced. The H2 evolution under light
irradiation was due to a photocatalytic reaction over Pt/TiO2
(eq 1) consuming organic substances (3.3 ppm TOC), which
were originally present in GW (Table 1) and acted as a hole
scavenger because the amount of CO2 in the gas phase
increased with an increase in the reaction time (data not
−
photocatalytic activity for the decomposition of NO3 (entry
7). However, it was much less active than the Pt/TiO2−SnPd/
Al2O3 system was under the same reaction conditions. Since H2
−
evolution, as well as NO3 decomposition, occurred over
−
shown). Pt/TiO2 alone hardly decomposed NO3 even in the
presence of glucose (entry 2), but the amount of H2 evolved in
SnPd/TiO2 under light irradiation, photoexcited e− formed in
−
−
SnPd/TiO2 were consumed for the reduction of NO3 (NO3
2210
dx.doi.org/10.1021/cs5003564 | ACS Catal. 2014, 4, 2207−2215

ACS Catalysis
Research Article
+ 6H+ + 5e− → 1/2N2 + 3H2O) and for H2 evolution (2H+ +
2e− → H2). We calculated the number of e− consumed during
the reaction in entry 7 to be 116 μmol. On the other hand, that
calculated for the reaction in entry 4 (the Pt/TiO2−SnPd/
In order for the water obtained by the photocatalytic
decomposition of NO3 to be used as drinking water, glucose
and its partially oxidized compounds must ultimately be
removed from the water. Figure 2 shows time courses for
−
−
Al2O3 system), in which H2 evolved and NO3 decomposed,
was 369 μmol. A comparison of the amount of consumed e− in
entries 4 and 7 clearly indicates that the photocatalytic
performance of the Pt/TiO2−SnPd/Al2O3 system is much
higher than that of SnPd/TiO2. In addition, SnPd/TiO2 is less
−
active toward the nonphotocatalytic reduction of NO3− (NO3
+ 5/2H2 → 1/2N2 + 2H2O + OH−) than SnPd/Al2O3 is
because a similar amount of H2 formed over both SnPd/TiO2
and Pt/TiO2−SnPd/Al2O3. On the other hand, in the case of
the Pt/TiO2−SnPd//Al2O3 system, the photocatalytic function
on Pt/TiO2 and nonphotocatalytic function on SnPd/Al2O3
can be individually optimized. Therefore, we concluded that
this reaction system showed superior activity for the
−
decomposition of NO3 under light irradiation than SnPd/
TiO2 did.
Figure 1 shows time courses for the conversion of NO3− and
−
+
selectivities to N2 + N2O, NO2 , and NH4 for the
−
photocatalytic reduction of NO3 in GW, GW-Ox, and
Figure 2. Changes in the concentration of total organic compounds
(TOC) during the photocatalytic reduction of NO3− in GW-Ox in the
presence of Pt/TiO2 and SnPd/Al2O3. Reaction conditions (t ≤ 120
h): catalyst weight, Pt/TiO2, 500 mg and SnPd/Al2O3, 100 mg;
KNO3-aq over the Pt/TiO2−SnPd/Al2O3 system. The
−
conversion of NO3 increased as the reaction progressed up
to 12 h in GW but only slightly increased after that (Figure 1A:
Δ). That is, the catalytic activity in GW was less than that in
KNO3-aq. Since GW contained various compounds (Table 1),
these compounds must have a negative effect on the activities
over Pt/TiO2, SnPd/Al2O3, or both.
−
reactant NO3 (from KNO3), 1.0 mmol dm−3 with 1.0 mmol dm−3
glucose; reaction volume, 250 cm3; light irradiation (λ > 300 nm). The
catalyst powder (Pt/TiO2−SnPd/Al2O3 mixture) was separated from
the reaction solution at 120 h and fresh Pt/TiO2 (500 mg) was added
to the solution, followed by light irradiation (λ > 300 nm) with a flow
of air to decompose the organic matter in the solution by
photocatalytic oxidation. Reaction conditions (t > 120 h): catalyst
weight, Pt/TiO2, 500 mg; reactant concentration, 1.0 mmol dm−3;
reaction volume, 250 cm3; air flow rate, 100 cm3 min−1; light
irradiation (λ > 300 nm).
In order to clarify the influence of the compounds in GW, we
−
conducted the photocatalytic reduction of NO3 with the Pt/
TiO2−SnPd/Al2O3 system in KNO3-aq, which did not contain
any cations other than K+, anions other than NO3 ions, and
−
−
organic compounds other than glucose. NO3 in KNO3-aq
decomposed more rapidly than it did in GW, and the
−
conversion of NO3 reached 100% at 36 h (Figure 1A: ○)
−
TOC during the photocatalytic reduction of NO3 in GW-Ox
without any catalyst deactivation. At 100% conversion, the
selectivity to N2 + N2O was 88% (Figure 1B: ○). From these
results, it was concluded that the compounds in GW decreased
the photocatalytic or catalytic performances (activity and
selectivity) of the Pt/TiO2−SnPd/Al2O3 system. During the
initial 24 h of the reaction in KNO3-aq, N2O formed in addition
to N2 but was completely transformed to N2 at 36 h (shown in
Figure S1 in the Supporting Information [SI]).
To investigate the effects of the organic compounds in GW
on the photocatalytic performance of the Pt/TiO2−SnPd/
Al2O3 system, we treated GW with photocatalytic oxidation in
the presence of Pt/TiO2 to decompose the organic compounds
(see Experimental Methods) and conducted the photocatalytic
over the Pt/TiO2−SnPd/Al2O3 system. The initial TOC was
73 ppm, and it decreased as the reaction progressed, reaching
59 ppm at 120 h, at which time NO3 was completely
−
decomposed. To completely decompose the organic com-
pounds in the reaction solution, the catalyst powder (Pt/TiO2−
SnPd/Al2O3 mixture) was separated from the reaction solution
at 120 h, and fresh Pt/TiO2 was added to the solution, followed
by light irradiation with a stream of air flowing into the reactor
(100 cm3 min−1). After an additional 24 h, the organic
compounds in the reaction solution were completely
decomposed. In other words, glucose added in GW and its
partially oxidized compounds can be easily decomposed so that
the groundwater can be used as drinking water.
−
reduction of NO3 in the treated solution (GW-Ox, Table 1)
Effects of the Compounds in the Groundwater (GW)
on Photocatalytic and Catalytic Performance. Figure 3a
shows the conversion of NO3− and selectivity for N2 + N2O in
over the Pt/TiO2−SnPd/Al2O3 system. The decomposition
−
rate of NO3 in GW-Ox was nearly twice that in GW, and
NO3− was completely decomposed at 120 h (Figure 1A: □). At
100% conversion (120 h), the selectivity to N2 + N2O was 83%
(Figure 1B: □), which was almost the same as that in KNO3-
aq. Thus, removing the organic compounds from GW makes it
−
the photocatalytic reduction of NO3 using the Pt/TiO2−
SnPd/Al2O3 system in the various reaction solutions under
light irradiation for a period of 12 h with an initial
concentration of NO3 of 1.0 mmol dm−3 in each reaction
−
−
possible to decompose NO3 selectively to N2 and N2O.
However, the reaction rate for NO3 -decomposition in GW-Ox
was slower than that in KNO3-aq (Figure 1A), indicating that
compounds other than the organic ones in GW negatively
affected the catalytic performance of the Pt/TiO2−SnPd/Al2O3
system. This will be discussed later.
solution. Although the pH of the solutions was different in each
case, as shown in Table 1, it barely affected the photocatalytic
performance (Figure S2A in the SI).
As was described above, compounds other than the organic
compounds still caused the NO3 -decomposition activity of the
−
−
Pt/TiO2−SnPd/Al2O3 system to decrease. Thus, cations and
2211
dx.doi.org/10.1021/cs5003564 | ACS Catal. 2014, 4, 2207−2215

ACS Catalysis
Research Article
Pt/TiO2 and nonphotocatalytic (conventional catalytic)
−
reduction of NO3 with H2 over SnPd/Al2O3, both reaction
rates affect the overall reaction rate of the NO3− decomposition
and the selectivities. Thus, to clarify the influence of the organic
and anionic compounds in GW on the performance of the Pt/
TiO2−SnPd/Al2O3 system, their effects on the photocatalytic
performance of Pt/TiO2 toward H2 evolution and the catalytic
performance of SnPd/Al2O3 for the reduction of NO3− with H2
must be individually investigated.
Figure 3b shows the photocatalytic H2 evolution rate over
Pt/TiO2 in the presence of glucose but the absence of SnPd/
Al2O3 in various reaction solutions. The H2 evolution rate
(μmol h−1) was estimated from the amount of H2 formed in 1 h
from the beginning of the reaction. Although the pH values of
the reaction solutions were different, as shown in Table 1, the
pH had little impact on the photocatalytic performance of Pt/
TiO2 (shown in Figure S2B in the SI). The H2 evolution rate
was 167 μmol h−1 in KNO3-aq, whereas it was only 59 μmol
h
−1 in GW. The H2 evolution rate was 93 μmol h−1 in GW-Ox,
indicating that H2 evolution over Pt/TiO2 was inhibited by the
organic compounds in GW. However, the H2 evolution rate in
GW-Ox was still slower than that in KNO3-aq. The effects of
the cations and anions in GW on the H2 evolution rate over Pt/
TiO2 were investigated in both GW-Ox-Cation-removal and
GW-Ox-Anion-removal. It was found that the cations in GW
had no impact on the H2 evolution rate, whereas the anions in
GW inhibited the photocatalytic H2 evolution over Pt/TiO2. In
KNO3−Cl−-aq, the H2 evolution rate was the almost same as
that in KNO3-aq. In contrast, when either SO42− or SiO32− was
present in KNO3-aq, the H2 evolution rate was significantly
slower than that in KNO3-aq. In particular, SiO32− dramatically
inhibited H2 evolution. From these results, it was concluded
Figure 3. Effects of different compounds in the groundwater on
photocatalytic and catalytic performances. (a) Photocatalytic reduction
of NO3− in water in the copresence of Pt/TiO2 and SnPd/Al2O3 under
−
■
□
light irradiation. ( ) Conversion of NO3 , and ( ) selectivity to N2 +
N2O. Reaction conditions: catalyst weight, Pt/TiO2, 500 mg and
−
SnPd/Al2O3, 100 mg; concentration of NO3 , 1.0 mmol dm−3;
concentration of glucose, 1.0 mmol dm−3; volume of reaction solution,
250 cm3; light irradiation (λ > 300 nm). (b) Photocatalytic hydrogen
evolution over Pt/TiO2 under light irradiation. Reaction conditions:
catalyst weight, Pt/TiO2, 500 mg; concentration of glucose, 1.0 mmol
dm−3; volume of reaction, 250 cm3; light irradiation (λ > 300 nm). (c)
−
Nonphotocatalytic reduction of NO3 with H2 in water over SnPd/
−
■
□
Al2O3 under the dark conditions. ( ) Conversion of NO3 , and ( )
selectivity to N2 + N2O. Reaction conditions: catalyst weight, SnPd/
n−
that the organic compounds, SO42−, and SiOx in GW
−
Al2O3, 10 mg; concentration of NO3 , 1.0 mmol dm−3; volume of
decreased the photocatalytic performance of Pt/TiO2, following
2−
the trend organic compounds > SiO3 > SO42−, which also
reaction solution, 250 cm3; gas composition, H2/CO2 = 1/1; gas flow
rate, 30 cm3 min−1.
corresponds to the photocatalytic performance of the Pt/
−
TiO2−SnPd/Al2O3 system toward NO3 decomposition
(Figure 3a). The mechanism for the inhibition of the
anions were removed from GW by passing it through ion-
exchange resins, followed by photocatalytic oxidation over Pt/
TiO2. The solutions obtained from GW by passing through a
cation- and an anion-exchange resins are denoted as GW-Ox-
Cation-removal and GW-Ox-Anion-removal, respectively. The
2−
2−
photocatalytic reaction over Pt/TiO2 by SO4 and SiO3
will be discussed in detail later.
Next, we investigated the influence of the compounds in GW
on the catalytic performance of SnPd/Al2O3. Figure 3c shows
−
−
the decomposition rate of NO3 and selectivity for N2 + N2O
conversion of NO3 and the selectivity to N2 + N2O in GW-
−
in the nonphotocatalytic reduction of NO3 with H2 over
Ox-Cation-removal were almost the same as those in GW-Ox,
indicating that the cations in GW did not affect the
photocatalytic and nonphotocatalytic performance levels of
the Pt/TiO2−SnPd/Al2O3 system. However, in GW-Ox-Anion-
SnPd/Al2O3 in the absence of Pt/TiO2 in the dark. The
decomposition rate of NO3 in KNO3-aq was 5.2 mmol h−1
−
gcat−1, and the selectivity for N2 + N2O was 97%. Since the
amount of N2O was negligible due to very low selectivity (less
than 1%), the formation of N2O did little harm to the
environment. In the case of GW, the decomposition rate of
−
removal, the conversion of NO3 and the selectivity to N2 +
N2O were recovered, being almost the same as those in KNO3-
aq. The results clearly indicate that the anions in GW render
the Pt/TiO2−SnPd/Al2O3 system ineffective.
−
NO3 was much slower than that in KNO3-aq, with a value of
0.5 mmol h−1 gcat−1, and the selectivity for N2 + N2O was only
Next, we investigated the influence of each anion (Cl−,
SO42−, and SiO32−), which is contained in GW, on the
photocatalytic performance of the Pt/TiO2−SnPd/Al2O3
−
63%. The decomposition rate of NO3 in GW-Ox was still
about a quarter of that in KNO3-aq, and the selectivity for N2 +
N2O was 88%, indicating that the organic compounds in GW
−
system toward the reduction of NO3 by using aqueous
−
−
NO3 solutions that contained each anion separately. Cl− in
significantly affected the nonphotocatalytic reduction of NO3
KNO3−Cl−-aq did not affect the conversion of NO3 and
−
with H2 over SnPd/Al2O3. Since the activity and selectivity in
GW-Ox-Cation-removal were similar to those in GW-Ox, the
cations in GW did not affect the catalytic performance of SnPd/
Al2O3. On the other hand, in GW-Ox-Anion-removal, the
selectivity to N2 + N2O (Figure 3a). On the other hand, when
2−
SO4 or SiO32− was present in the solution, the conversion of
−
NO3 decreased. In particular, the decrease in the conversion
−
was most notable in KNO3−SiO32−-aq.
decomposition rate of NO3 and selectivity to N2 + N2O were
−
Since the photocatalytic reduction of NO3 over Pt/TiO2−
the same as those in KNO3-aq. Thus, the anions in GW
inhibited the catalytic performance of SnPd/Al2O3. However,
SnPd/Al2O3 proceeded with photocatalytic H2 evolution over
2212
dx.doi.org/10.1021/cs5003564 | ACS Catal. 2014, 4, 2207−2215

ACS Catalysis
Research Article
a
Table 3. Catalytic Reduction of NO3− in the Dark in an Aqueous KNO3 Solution over SnPd/Al2O3 in the Presence of K2SiO3,
KCl, and K2SO4
selectivity/[%]
b
c
activity/mmol h−1 gcat
NO2
NH4
N2 + N2O
absorbance @ 440 nm
−1
−
+
entry
anion
−
1
2
3
4
5
6
NO3
5.2
1.7
5.2
5.2
3.9
5.1
0
1
3
7
4
6
5
8
97
92
96
93
95
91
−
0.38
NO3−, SiO32−, Cl−, SO4
2−
NO3−, Cl−, SO4
0
−
2−
NO3−, SiO32−, SO4
1
0.46
0.43
0.47
2−
NO3−, SiO32−, Cl−
<1
NO3−, SiO3
1
2−
a
−
Reaction conditions: catalyst weight, SnPd/Al2O3, 10 mg; reactant NO3 (from KNO3), 1.0 mmol dm−3; reaction volume, 250 cm3; gas
composition, H2/CO2 = 1/1; gas flow rate, 30 cm3 min−1. Initial concentrations of K2SiO3, KCl, and K2SO4 were 1.1, 0.5, and 1.1 mmol dm−3,
respectively. Selectivity to N2 + N2O was calculated by subtracting the selectivities of NO2 and NH4 from 100%. Determined by using a
molybdenum yellow method.30
b
c
−
+
when the anions, including Cl−, SO42−, and SiOxn−, were
individually present in KNO3-aq, the decomposition rate of
NO3− by the catalytic reduction with H2 over SnPd/Al2O3 and
the selectivity for N2 + N2O were almost the same as those in
KNO3-aq.
absorbances (Table 3). This correlation clearly indicates that
the polymerized silicate ions deactivate SnPd/Al2O3 in the
nonphotocatalytic reduction of NO3 with H2.
−
From the results mentioned above, we concluded that
K2SO4, K2SiO3, and organic compounds in GW decreased the
photocatalytic performance of Pt/TiO2 and that the organic
compounds and polymerized silicate ions in GW lowered the
nonphotocatalytic performance of SnPd/Al2O3. Thus, the
Thus, we thought that the anions (Cl−, SO42−, and SiOxn−
)
only had a significant effect on the catalytic performance of
SnPd/Al2O3 when a combination of these anions was present.
Therefore, we determined the catalytic performance of SnPd/
−
decomposition rate of NO3 and selectivity to N2 + N2O by
−
Al2O3 for the reduction of NO3 with H2 in solutions
the Pt/TiO2−SnPd/Al2O3 system also decreased for the
containing different combinations of Cl−, SO42−, and SiOxn−
.
photocatalytic reduction of NO3 in GW and GW-Ox.
−
Table 3 summarizes the results for the catalytic reduction of
Mechanism for Deactivation of Pt/TiO2 by K2SO4 and
K2SiO3. As described above, SO4 and SiOx in GW
2−
n−
NO3− with H2 over SnPd/Al2O3 without irradiation in reaction
2−
solutions containing Cl−, SO42−, and SiO3 in the same
decreased the photocatalytic activity of Pt/TiO2 (Figure 3b).
concentrations (0.5, 1.1, and 1.1 mmol dm−3, respectively) as
those in GW. In the reaction in entry 2, the decomposition rate
Possible reasons for the deactivation of Pt/TiO2 by SO4 and
2−
SiOx are poisoning of the reduction sites (Pt_e−), oxidation
n−
−
2−
of NO3 was significantly slower in comparison with that in
sites (TiO2_h+), or both on Pt/TiO2 with SO4 and SiOxn−
KNO3-aq (entry 1). On the other hand, the decomposition
rates of NO3− in the solutions containing SO42− and Cl− (entry
3) and SO42− and SiO32− (entry 4) were comparable to that in
(Figure S3 in the SI). In order to clarify which sites were
poisoned by SO42− and SiOxn−, we investigated the influence of
the concentrations of both compounds on the photocatalytic
performance for H2 evolution (2H+ + 2 (Pt_e−) → H2 + 2Pt)
from an aqueous glucose solution over Pt/TiO2 and for the O2
evolution (2H2O + 4 (TiO2_h+) → O2 + 4H+ + 4TiO2) from
an aqueous AgNO3 solution over bare TiO2. The former
reaction can be used to estimate the level of poisoning of Pt_e−
sites due to these compounds, and the latter one can be used to
estimate the level of poisoning of TiO2_h+ sites.
2−
KNO3-aq (entry 1). In contrast, in a solution containing SiO3
−
and Cl− (entry 5), the decomposition rate of NO3 decreased
−1
to 3.9 mmol h
(5.2 mmol h
g
cat
−1, which was slower than that in KNO3-aq
cat
−1
g
−1). A possible reason for the deactivation of
SnPd/Al2O3 with SiO32− in solutions that also contained Cl− or
Cl−-SO4 is that SiO3 polymerizes in the presence of Cl−
and SO42−, after which the formed silicate oligomers are
adsorbed on SnPd/Al2O3.
2−
2−
The H2 evolution rate over Pt/TiO2 under photocatalytic
conditions was 298 μmol h−1 in the absence of K2SO4 (Figure
S4 in the SI). When 0.1 mmol dm−3 of K2SO4 was added to the
solution, the H2 evolution rate decreased to 245 μmol h−1.
Further addition of K2SO4 did not change the H2 evolution
rate. Since a small amount of K2SO4 had a substantial effect on
2−
Thus, the amount of SiO3 monomer in each reaction
solution was determined by using a molybdenum yellow
method,30 where Mo6+ is added to the solution and reacts with
monomeric silicate ions to form a heteropolyanion
([SiMo12O40]4−). The amount of [SiMo12O40]4− in the solution
is determined from the absorbance at 440 nm in a UV−vis
spectrum of the solution. Therefore, if the intensity of the
absorbance at 440 nm decreases, then the amount of silicate
oligomers has increased. Table 3 lists the absorbances at 440
nm in the reaction solutions after 1 h of the catalytic reduction
of NO3− with H2 over SnPd/Al2O3.31 The absorbance in entry
2−
the H2 evolution rate over Pt/TiO2, we concluded that SO4
was strongly adsorbed on the H2 evolution sites on Pt/TiO2,
i.e., Pt atoms exposed on the surface of the Pt particles,
resulting in the deactivation of Pt/TiO2. In contrast to the H2
2−
evolution rate, the presence of SO4 in the solution did not
cause a decrease in the O2 evolution rate over the bare TiO2,
indicating that SO42− did not poison the TiO2_h+ sites on TiO2
(Figure S4 in the SI).
Figure 4 shows the effects of the concentration of K2SiO3 on
the H2 evolution rate from the photocatalytic reaction over Pt/
TiO2 in an aqueous glucose solution and on the O2 evolution
rate from the photocatalytic reaction over bare TiO2 in an
aqueous AgNO3 solution. The H2 evolution rate did not change
with the concentration of K2SiO3 up to 0.5 mmol dm−3,
4 (A440 = 0.46) was almost the same as that in entry 6 (A440
=
0.47). On the other hand, the absorbances in entries 2 and 5
(A440 = 0.38 and 0.43, respectively), in which the catalytic
performance was smaller, were lower than that in entry 6. In
particular, the absorbance in entry 2 was much smaller than that
in entry 5. The results indicate that a larger amount of silicate
ions was polymerized in the reaction solution in entry 2 than it
was in entry 6. The catalytic activity correlated with the
2213
dx.doi.org/10.1021/cs5003564 | ACS Catal. 2014, 4, 2207−2215

ACS Catalysis
Research Article
2−
2−
Figure 5. Adsorption of SiO3 on Pt/TiO2. Adsorption conditions:
○
Figure 4. Effects of the concentration of SiO3 on the rates of ( ) H2
2−
weight of Pt/TiO2, 30 mg; initial concentration of SiO3 (from
evolution for the photocatalytic reactions in an aqueous glucose
solution over Pt/TiO2 and ( ) O2 evolution for the photocatalytic
K2SiO3), 0.1−8.0 mmol dm−3; temperature, 303 K; and volume of
●
solution, 10 cm3.
reaction in an AgNO3 solution over TiO2. Reaction conditions for H2
evolution: catalyst weight, Pt/TiO2, 500 mg; reactant glucose, 100
mmol dm−3; reaction volume, 250 cm3; and light irradiation (λ > 300
nm). Reaction conditions for O2 evolution: catalyst weight, TiO2, 500
mg; initial pH, 9.8; reactant AgNO3, 10 mmol dm−3; reaction volume,
250 cm3; and light irradiation (λ > 300 nm).
nonphotocatalyst SnPd/Al2O3, which were dispersed in the
groundwater, irradiated at λ > 300 nm in the presence of
glucose as a hole scavenger. However, pretreatment of the
groundwater by photooxidation with Pt/TiO2 was necessary to
remove the organic compounds from the groundwater because
they poisoned both Pt/TiO2 and SnPd/Al2O3. The decom-
2−
although the total amount of SiO3 (125 μmol) in the 0.5
mmol dm−3 K2SiO3 solution was much larger than the number
of exposed Pt sites on the surface (3 μmol), which was
estimated from the adsorption amount of CO at 323 K.32
−
position rate of NO3 and the selectivity to N2 + N2O in the
photocatalytic reduction of NO3 in the groundwater were
−
2−
Therefore, the adsorption of SiO3 on the surface of the Pt
lower than those in an aqueous solution of KNO3. The lower
photocatalytic performance of the Pt/TiO2−SnPd/Al2O3
system in the groundwater was due to sulfate and silicate
ions and the organic compounds in the groundwater. Sulfate
and silicate ions deactivated Pt_e− and TiO2_h+ sites,
respectively, on the Pt/TiO2 surface by adsorption, causing a
decrease in the photocatalytic performance toward H2
evolution. In addition, polymerized silicate ions in the
groundwater decreased the catalytic performance of SnPd/
particles was very weak. On the other hand, the O2 evolution
rate significantly decreased, even in the presence of a small
amount of K2SiO3 (1.0 mmol dm−3). These results clearly
indicated that TiO2_ h+ sites on TiO2 were deactivated with
K2SiO3. Although the rate-limiting step for the H2 evolution
reaction from the aqueous glucose solution over Pt/TiO2 was
the H2 evolution reaction from the Pt_e− sites and SiO3
2−
deactivated TiO2_h+ sites on TiO2, the H2 evolution rate
decreased when an excess amount of K2SiO3 (>1.0 mmol
dm−3) was added. This is probably because the rate-limiting
step for the H2 evolution reaction from the aqueous glucose
solution changed from the consumption of Pt_e− to form H2 to
that of TiO2_h+ due to the poisoning of most of the TiO2_h+
−
Al2O3 for the hydrogenation of NO3 with H2.
ASSOCIATED CONTENT
■
S
* Supporting Information
−
Time courses for the photocatalytic reduction of NO3−,
2−
2−
sites with SiO3 when excess SiO3 was present in the
reaction solution.
influence of pH on the photocatalytic reduction of NO3 ,
and dependences of concentrations of glucose and AgNO3 on
the rates of H2 and O2 evolutions, respectively, as mentioned in
the text. This material is available free of charge via the Internet
at http://pubs.acs.org.
2−
Figure 5 shows an adsorption profile for SiO3 on Pt/TiO2
at 303 K. The dashed line in Figure 5 represents the number of
surface hydroxyl groups on Pt/TiO2 (295 μmol g−1), which was
calculated from the surface area of the Pt/TiO2 and the density
of surface hydroxyl groups (3.3 OH nm−2).32 The adsorption
AUTHOR INFORMATION
■
2−
amount of SiO3
increased with an increase in the
Corresponding Author
concentration of SiO3 and saturated at ∼350 μmol g−1,
2−
*E-mail: kamiya@ees.hokudai.ac.jp. Tel/Fax:+81-11-706-2217.
which was approximately equal to the number of surface
Notes
hydroxyl groups (295 μmol g−1). This result implies that SiO3
2−
The authors declare no competing financial interest.
stoichiometrically reacted with the surface hydroxyl groups in a
1:1 ratio (Ti−O−Si−OH) and supports the hypothesis that the
REFERENCES
2−
TiO2_h+ sites were poisoned by SiO3
.
■
(1) Wakida, F. T.; Lerner, D. N. Water Res. 2005, 39, 3−16.
(2) Cantor, K. P. Cancer, Causes Control, Pap. Symp. 1997, 8, 292−
308.
(3) WHO. Guidelines for drinking-water quality. First addendum to
volume 1, Recommendations, 3rd edition; World Health Organization:
Geneva, Switzerland, 2006.
CONCLUSIONS
■
Nitrate in real groundwater was decomposed mainly into N2
with an 83% selectivity by using a photocatalytic reaction
system composed of the photocatalyst Pt/TiO2 and the
2214
dx.doi.org/10.1021/cs5003564 | ACS Catal. 2014, 4, 2207−2215

ACS Catalysis
Research Article
(4) Kudo, A.; Domen, K.; Maruya, K.; Onishi, T. Chem. Lett. 1987,
1019−1022.
(5) Ranjit, K. T.; Viswanathan, B. J. Photochem. Photobiol., A 1997,
108, 73−78.
(6) Bems, B.; Jentoft, F. C.; Schlogl, R. Appl. Catal., B 1999, 20, 155−
̈
163.
(7) Hamanoi, O.; Kudo, A. Chem. Lett. 2002, 838−839.
(8) Gao, W.; Jin, R.; Chen, J.; Guan, X.; Zeng, H.; Zhang, F.; Guan,
N. Catal. Today 2004, 90, 331−336.
(9) Zhang, F.; Jin, R.; Chen, J.; Shao, C.; Gao, W.; Li, L.; Guan, N. J.
Catal. 2005, 232, 424−431.
(10) Kominanmi, H.; Nakaseko, T.; Shimada, Y.; Furusho, A.; Inoue,
H.; Murakami, S.; Kera, Y.; Ohtani, B. Chem. Commun. 2005, 2933−
2935.
(11) Sa, J.; Aguera, C. A.; Gross, S.; Anderson, J. A. Appl. Catal., B
̀
̈
2009, 85, 192−200.
(12) Wehbe, N.; Jaafar, M.; Guillard, C.; Herrmann, J. M.; Miachon,
S.; Puzenat, E.; Guilhaume, N. Appl. Catal., A 2009, 368, 1−8.
(13) Li, L.; Xu, Z.; Liu, F.; Shao, Y.; Wang, J.; Wan, H.; Zheng, S. J.
Photochem. Photobiol., A 2010, 212, 113−121.
(14) Gekko, H.; Hashimoto, K.; Kominami, H. Phys. Chem. Chem.
Phys. 2012, 14, 7965−7970.
(15) Doudrick, K.; Monzon, O.; Mangonon, A.; Hristovski, K.;
Westerhoff, P. J. Environ. Eng. 2012, 138, 852−861.
(16) Doudrick, K.; Yang, T.; Hristovski, K.; Westerhoff, P. Appl.
Catal., B 2013, 136−137, 40−47.
(17) Hirayama, J.; Kondo, H.; Miura, Y.; Abe, R.; Kamiya, Y. Catal.
Commun. 2012, 20, 99−102.
(18) Hirayama, J.; Abe, R.; Kamiya, Y. Appl. Catal., B 2014, 144,
721−729.
(19) Shand, M.; Anderson, J. A. Catal. Sci. Technol. 2013, 3, 879−
899.
(20) Maeda, K. ACS Catal. 2013, 3, 1486−1503.
(21) Ohtani, B. J. Photochem. Photobiol., C 2010, 11, 157−178.
(22) Amano, F.; Yasumoto, T.; Shibayama, T.; Uchida, S.; Ohtani, B.
Appl. Catal., B 2009, 89, 583−589.
(23) Sayama, K.; Mukasa, K.; Abe, R.; Abe, Y.; Arakawa, H. Chem.
Commun. 2001, 2416−2417.
(24) Sayama, K.; Mukasa, K.; Abe, R.; Abe, Y.; Arakawa, H. J.
Photochem. Photobiol., A 2002, 148, 71−77.
(25) Abe, R.; Sayama, K.; Sugihara, H. J. Phys. Chem. B 2005, 109,
16052−16061.
(26) Abe, R.; Takata, T.; Sugihara, H.; Domen, K. Chem. Commun.
2005, 3829−3831.
(27) Kato, H.; Hori, M.; Konta, R.; Shimodaira, Y.; Kudo, A. Chem.
Lett. 2004, 33, 1348−1349.
(28) Schwartz, T. J.; Goodman, S. M.; Osmundsen, C. M.; Taarning,
E.; Mozuch, M. D.; Gaskell, J.; Cullen, D.; Kersten, P. J.; Dumesic, J. A.
ACS Catal. 2013, 3, 2689−2693.
(29) Denard, C. A.; Hartwig, J. F.; Zhao, H. ACS Catal. 2013, 3,
2856−2864.
(30) Potapov, V. V.; Serdan, A. A.; Kashpura, V. N.; Gorbach, V. A.;
Tyurina, N. A.; Zubakha, S. V. Glass Phys. Chem. 2007, 33, 44−49.
(31) Two cubic centimeters of a 10% (w/v) aqueous (NH4)6Mo7O24
solution and 1 cm3 of hydrochloric acid (6 mol dm−3) were added to
50 cm3 of the reaction solution (entries 2 and 4−6 in Table 3). The
solution was allowed to stand for 5 min at 293 K, and then the
absorbance of the solution at 440 nm was measured on a
spectrophotometer to estimate the amount of [SiMo12O40]4− in the
solution.
(32) Erdem, B.; Hunsicker, R. A.; Simmons, G. W.; Sudol, E. D.;
Dimonie, V. L.; El-Aasser, M. S. Langmuir 2001, 17, 2664−2669.
2215
dx.doi.org/10.1021/cs5003564 | ACS Catal. 2014, 4, 2207−2215