Solution combustion synthesis of Ti0.75Ce0.15Cu0.05W0.05O2-δ for low temperature selective catalytic reduction of NO

Article abstract of DOI:10.1039/c4ra05862c

A Ti-Ce-Cu-W catalyst prepared by solution combustion synthesis (SCS) showed high low temperature activity for selective catalytic reduction (SCR) of NO with ammonia. Different fuels and different fuel/oxidant (F/O) ratios were introduced to investigate the optimum synthesis conditions for the Ti_0.75Ce_0.15Cu_0.05W_0.05O_2-δ catalyst. The results showed that samples using citric acid and glycine as a fuel showed good low temperature catalytic activity and the catalytic activity could be improved through increasing the F/O ratio. The sample with a F/O ratio of 1.5 obtained around 70% NO conversion at a temperature of 150°C.

Full text of DOI:10.1039/c4ra05862c

RSC Advances
View Article Online
View Journal | View Issue
PAPER
Solution combustion synthesis of
Ti0.75Ce Cu0.05W for low temperature
selective catalytic reduction of NO
O
0.15
0.05 2ꢀd
Cite this: RSC Adv., 2014, 4, 63909
Ting Chen, He Lin,* Qunhai Cao and Zhen Huang
A Ti–Ce–Cu–W catalyst prepared by solution combustion synthesis (SCS) showed high low temperature
activity for selective catalytic reduction (SCR) of NO with ammonia. Different fuels and different fuel/
oxidant (F/O) ratios were introduced to investigate the optimum synthesis conditions for the
Received 20th June 2014
Accepted 17th November 2014
0.05
Ti0.75Ce0.15Cu0.05W O2ꢀd catalyst. The results showed that samples using citric acid and glycine as a
fuel showed good low temperature catalytic activity and the catalytic activity could be improved through
DOI: 10.1039/c4ra05862c
increasing the F/O ratio. The sample with a F/O ratio of 1.5 obtained around 70% NO conversion at a
ꢁ
www.rsc.org/advances
temperature of 150 C.
the oven, in which solution combustion has taken place,
affected the performance of catalysts. In SCS, synthesis
1
. Introduction
11
Solution combustion synthesis (SCS) is an effective, low-cost conditions, such as fuel types and fuel/oxidant (F/O) may also
method for preparing various solid materials, such as spinel- impact the properties of products. Therefore, it is of interesting
1
type LiMn O for lithium ion batteries and bi-phasic calcium to investigate the effect of synthesis conditions on the perfor-
2
4
2
phosphate (BCP) for bone replacements. In SCS, organic mance of catalysts.
compounds containing reducible functional groups are
In recent years, vanadium SCR catalysts were enhanced their
employed as fuels and metal nitrates are used as oxidizers. It low temperature activity for NH
-SCR reaction by loaded on
–AlO and CeO
2 3
O . For the toxicity of the vanadium species, non-vanadium
3
12
was reported that the synthesis conditions, such as fuel types, various metal oxide supports, such as TiO
the amount of fuel and the pH values of the precursor solution, Al
–
2
2
3
3
–5
13–22
affect the properties of the products. The inherent nature of SCR catalysts were studied extensively,
for example, transi-
the fuel as well as its complex capability with ionic species is tion metal manganese is well known for its high low tempera-
3,4
16–19
crucial for the morphology of synthesized powders. Addi- ture activity.
However, the activity of manganese based
tionally, fuel/oxidant ratios may affect the particle size and the catalysts was inhibited obviously by SO
2
. Cerium (Ce) caught
catalyst showed excellent
NH -SCR performance in a rather wide temperature range of
3
5
dopant segregation. Therefore, controlling the synthesis lots of attention because Ce/TiO
2
conditions is critical in solution combustion synthesis.
Selective catalytic reduction (SCR) technique has been used 275–400 C. It was found that tungsten (W) doping could
to reduce nitrogen oxides (NO
) emission from power plants promoted the catalytic performance of Ce–Ti oxides at high
and diesel engines since 1990s. Recently, self-propagating temperature, for example, Ce0.2
high-temperature synthesis (SHS) method has been used to than 90% NO in the temperature window of 275 to 450 C.
synthesize SCR catalysts for the purpose of promoting the Copper-based Ce–Cu–Ti oxide catalysts presented high activity
22
ꢁ
20
x
6
–8
W
x
0.2TiO could reduce more
ꢁ
21
x
ꢁ
ꢁ
performance at low temperature. Roy et al. found that the in a rather wide temperature range from 200 C to 400 C,
catalyst Ti0.9Mn0.05Fe0.05
2ꢀd synthetized by SHS showed high which indicated that element Cu doping could promote low
catalytic activity and high N
et al. prepared Ti0.9Ce0.05
catalyst powder exhibited good SCR performance at low over above three catalysts.
temperature, due to its nanoparticle size, high specic surface
For the purpose of promoting the low temperature activity of
SCR catalysts, we intend to prepare Ti–Ce, Ti–Ce–W and Ti–Ce–
The so-called SHS method used above can be identied to a Cu–W catalysts with SCS. Effects of synthesis conditions on the
kind of SCS method. Guan also found that the temperature of structure and the performance of Ti0.75Ce0.15Cu0.05
O
ꢁ
9
selectivity in 150–450 C. Guan temperature performance of Ce–Ti based catalysts. It was also
2ꢀd with SHS, and found that the reported that SO have slight inuence on the catalytic activity
2
V
0.05
O
2
10
areas, appropriate pore volumes and average pore diameters.
W O
0.05 2ꢀd
catalyst are investigated. The synthesis conditions are designed
separately and independently. Glycine, citric acid and urea are
used as a fuel respectively. The fuel/oxidant ratio (F/O) changes
from 0.5 to 1.0, and then to 1.5. Activities of the catalysts are
Key Laboratory for Power Machinery and Engineering of Ministry of Education,
Shanghai Jiao Tong University, 800 Dongchuan Road, Shanghai 200240, China.
E-mail: linhe@sjtu.edu.com; Fax: +86 21 34205949; Tel: +86 21 34206859
This journal is © The Royal Society of Chemistry 2014
RSC Adv., 2014, 4, 63909–63916 | 63909

View Article Online
RSC Advances
Paper
tested through temperature programmed reaction (TPR). The (C
6
H
8
O
7
) and urea (CON
2
H
4
) are 9, 18 and 6 respectively. Stoi-
catalysts are characterized by X-ray diffraction (XRD), trans- chiometric combustion reactions involving other fuels (citric
mission electron microscopy (TEM), high-resolution trans- acid and urea) are expressed in eqn (2) and (3).
mission electron microscopy (HRTEM), selected area electron
0
:75TiOðNO
3
Þ2 þ 0:15CeðNO
3
Þ3 þ 0:05CuðNO
3
Þ2
diffraction (SAED) and X-ray photoelectron spectroscopy (XPS).
0:05
þ
þ
þ
ðNH
4
Þ W12
O
41
1
0
1
2
0:25
1
1
2. Experimental
fC
2
H
5
NO
2
/Ti0:75Ce0:15Cu0:05
W
0:05
O
2ꢀd
9
2.1 Catalyst preparation
ꢀ
ꢁ
03:25
10:25
20:5
fH
2
O þ
f þ 1:025 N
2
þ
fCO
2
The SCS process of SCR catalysts based on glycine-nitrate is
3
5
8
9
9,10
proposed in literatures. Take catalyst Ti–Ce for an example.
Firstly, required amount of absolute alcohol and deionized
water was added orderly into the precursor of tetrabutyl titanate
0
:5
þ
NH
3
(1)
1
2
(
Ti[OCH
3
(CH
2
)
3
]
4
). Then, right amount of nitric acid (HNO
3
)
0:75TiOðNO Þ þ 0:15CeðNO Þ þ 0:05CuðNO Þ
3
2
3
3
3
2
solution was added to form titanyl nitrate (TiO(NO ) ).
Sequentially, cerium nitrate (Ce(NO ) $6H O) was added into
titanyl nitrate solution with stirring. Finally, a fuel (such as
glycine) was added to the solution. The prepared solution was
stirred for 1 h at 50 C, and then was exposed in the muffle
3
2
0:05
þ
þ
ðNH
Þ10W12
4 41
O
3
3
2
12
1
0:25
82:75
C
6
H
8
O
7
/Ti0:75Ce0:15Cu0:05
W
0:05
O
2ꢀd
þ
2
H O
1
8
36
ꢁ
10:25
0:5
þ 1:025N
2
þ
CO
2
þ 12 NH
3
ꢁ
furnace at the temperature of 400 C. Aer water being evapo-
3
rated, combustion took place and lasted for 5–6 minutes,
(2)
powder product was obtained. The combustion product was
ꢁ
calcined at 400 C for 1 h, then was cooled down and sieved
0:75TiOðNO
3
Þ2 þ 0:15CeðNO
3
Þ3 þ 0:05CuðNO Þ2
3
to 40–60 mesh. Under the same condition, element tungsten in
Ti–Ce–W catalyst was prepared by introducing ammonium
0:05
þ
þ
ðNH
4
Þ W12
O
41
1
0
1
10:25
2
tungstate ((NH
in oxalic acid (H
Ti–Ce–Cu–W, cupric nitrate (Cu(NO
4
)
10
W
12
O
41$xH
) solutions. When copper was used for
$3H O) was added likewise.
2
O), which was needed to dissolve
41:25
CON
2
H
4
/Ti0:75Ce0:15Cu0:05
W
0:05
O
2ꢀd
þ
2
H O
2
C
2
O
4
6
8:2
12
3 2
)
2
10:25
0:5
þ 3
N
2
þ
6
CO
2
þ 12 NH
3
In this study, ve Ti–Ce–Cu–W catalysts were prepared by
using three kinds of fuels and under different ratios of fuel to
oxidant (F/O). As shown in Table 1, all catalysts are shown as the
(3)
formula of Ti0.75Ce0.15Cu0.05W0.05
O
(X). Here the small-scale
2ꢀd
number denotes the mole ratio of respective component. X is
the ratio of F/O or the name of fuel. The balanced reaction for
2
.2 Catalytic activity tests
Ti0.75Ce0.15Cu0.05
0.05
W O2ꢀd synthesis with glycine is given in
Catalytic activity tests were carried out with temperature pro-
grammed reduction (TPR) in a xed bed quartz reactor (i.d.
9
in the reactor and was purged in N
the sample surface. The reactant gas which contained 600 ppm
NO, 600 ppm NH , and 5% O was led into the reactor with a
total ow rate of 950 mL min (N for balance). A high gas
hourly space velocity (GHSV) of 114 000 h was obtained. The
eqn (1), considering ammonium tungstate involved in a thermal
decomposition reaction, where f is the ratio of fuel to oxidant.
And f ¼ 1 refers to stoichiometric F/O ratio that metal nitrates
mm). Before each test run, 0.5 ml powder sample was placed
ꢁ
2
gas at 450 C for 1 h to clean
23
complete oxidation of fuel, while f > 1 means fuel rich.
Nevertheless, high fuel/oxidant ratios may increase combustion
intensity. In order to avoid danger, a controlled dosing of fuels
should be select.
The value of F/O can be calculated from the ratio of reducing
valencies in fuels to total oxidizing valencies in oxidizer salts
3
2
ꢀ
1
2
ꢀ1
ꢁ
ꢁ
sample was heated up from 50 to 450 C, with 50 C as incre-
mental. The outlet gas (NO, NO , N O and NH ) was measured
on line by a FTIR gas analyzer (Thermo NICOLET6700).
24,25
2
2
3
(
metal nitrates in eqn (1)).
TiO(NO , Ce(NO and Cu(NO
tively. The reducing valencies of glycine (C H NO ), citric acid
The oxidizing valencies of
) are 10, 15 and 10 respec-
3 2
3
)
2
3
)
3
2
5
2
2.3 Characterization of catalyst
The BET surface area and pore volume of catalyst samples were
measured using a surface area and pore size analyzer (Quan-
tachrome Nova 2000e). Prior to each measurement, the sample
was degassed at 300 C for 3 h. The special surface area was
calculated from a linear portion of the BET plot. The pore size
distribution was calculated from the desorption branch of the
Table 1 List of catalyst samples
Samples
Fuel
F/O
ꢁ
Ti0.75Ce0.15Cu0.05
Ti0.75Ce0.15Cu0.05
Ti0.75Ce0.15Cu0.05
Ti0.75Ce0.15Cu0.05
Ti0.75Ce0.15Cu0.05
W
W
W
W
W
0.05
0.05
0.05
0.05
0.05
O
O
O
O
O
2ꢀd (0.5)
2ꢀd (1.0)
2ꢀd (1.5)
2ꢀd (citric acid)
2ꢀd (urea)
Glycine
Glycine
Glycine
Citric acid
Urea
F/O ¼ 0.5
F/O ¼ 1
F/O ¼ 1.5
F/O ¼ 1
F/O ¼ 1
N
2
adsorption isotherm using Barrett–Joyner–Halenda (BJH)
formula.
63910 | RSC Adv., 2014, 4, 63909–63916
This journal is © The Royal Society of Chemistry 2014

View Article Online
Paper
Powder X-ray diffraction (XRD) measurement was performed
RSC Advances
on a Rigaku D/max-2200/PC system with Cu Ka (l ¼ 0.1528)
ꢁ
radiation. All patterns were obtained from 20 to 90 (2q) with a
ꢁ
ꢀ1
ꢁ
scanning rate of 5 (2q) min , as well as a step size of 0.02 at
ambient temperature.
Transmission electron microscopy (TEM), high-resolution
transmission electron microscopy (HRTEM) and selected area
electron diffraction (SAED) patterns were carried out on a JEOL
JEM-2010 analytical electron microscope, operating at an
accelerating voltage of 200 kV.
X-ray photoelectron spectroscopy (XPS) was used to analyze
the atomic surface concentration on each catalyst. The spectra
were acquired with a Kratos AXIS Ultra DLD spectrometer using
Al Ka radiation (1486.7 eV). Binding energies (BE) of Ti 2p, Ce Fig. 2 NO conversion and N
2
selectivity over Ti0.75Ce0.15Cu0.05-
W
0.05
O
2ꢀd prepared with different fuels. Reaction conditions: [NO] ¼
3d, Cu 2p, W 4f and O 1s were calibrated using C 1s (BE ¼
ꢀ1
[
NH
3
2
] ¼ 600 ppm, [O ] ¼ 5 vol%, N
2
as balance gas, GHSV ¼ 114 000 h
.
284.8 eV) as the standard.
3
. Results and discussions
synthesized with citric acid obtains nearly the same NO
conversion as the one prepared with glycine in the temperature
3.1 Activities of different catalysts
ꢁ
ꢁ
window of 100–330 C. NO conversion reaches 60% at 160 C
In order to investigate the low temperature activity of SCR
catalysts, transition metal Cu and W was doped on the Ti–Ce
catalyst. As shown in Fig. 1, Ti–Ce catalyst itself obtain more
than 80% NO conversion from 250–350 C. With the doping of
W, NO conversion is enhanced in the whole temperature
ꢁ
and almost 100% at 200 C, suggesting good low temperature
activity. And the catalyst synthesized with citric acid has a
ꢁ
ꢁ
higher activity above 350 C, due to its higher N2 selectivity
Fig. 2 inset). For the sample made using urea, the SCR
(
performance is not outstanding. Firstly, NO conversion
range over Ti–Ce–W catalyst. For Ti–Ce–Cu–W catalyst
ꢁ
increases gradually with the temperature. At 300 C, NO
(Ti0.75Ce0.15Cu0.05
0.05
W O
2ꢀd (1.0) in Table 1), the low tempera-
ꢁ
conversion reaches a peak point of 100%, later decreases
sharply with the temperature increasing. Hence, urea may not
ture NO conversion is further enhanced from 100–250 C. But
its activity begins to decrease over 300 C, with 88% NO conver-
sion at 350 C and 50% at 400 C. It can be seen that Ti–Ce–Cu–W
catalyst has the highest activity in the temperature range <250 C
ꢁ
ꢁ
ꢁ
the appropriate fuel for Ti0.75Ce0.15Cu0.05W0.05
2ꢀd
O synthesis
ꢁ
since the product shows a narrow temperature window.
The BET surface area, pore volume, and pore size of the
samples prepared with different fuels are shown in Table 2. The
and combines the high temperature SCR performance.
order of surface area of these samples is: Ti0.75Ce0.15Cu0.05
2ꢀd (citric acid) > Ti0.75Ce0.15Cu0.05 2ꢀd (glycine) >
Ti0.75Ce0.15Cu0.05 2ꢀd (urea). While, the average pore
diameter shows reverse order. Additionally, Ti0.75Ce0.15Cu0.05
2ꢀd (urea) obtains the largest average pore volume.
-
3
.2 Effects of fuel type on catalytic performance over
W
0.05
O
0.05
W O
Ti0.75Ce0.15Cu0.05
0.05 2ꢀd
W O
W O
0.05
Three fuels (glycine, citric acid and urea) were used to prepare
2ꢀd samples. We compare the perfor-
-
Ti0.75Ce0.15Cu0.05
0.05
W O
W O
0.05
mance of the catalysts prepared with different fuels under
stoichiometric F/O ratio of 1. As shown in Fig. 2, the catalyst
According to the molecule of fuels, one mole of fuel is
completely combusted, 7, 5 and 3.5 mol gases (i.e. CO , H O and
2
2
N ) are emitted from citric acid, glycine and urea respectively.
2
Gas emission in synthesis is benecial for enhancing the
surface area of catalyst.
XRD patterns of the powders are presented in Fig. 3. Crys-
talline phases are identied by being compared with ICDD les
(TiO
2 2
anatase phase: PDF # 21-1272; TiO rutile phase: PDF # 21-
1
276; cubic phase CeO : PDF # 34-0394; monoclinic phase CuO:
2
PDF # 45-0937; monoclinic phase WO : PDF # 83-0950). The
3
Table 2 Characterization of the catalysts prepared with different fuels
Sample prepared
condition
BET surface
area (m g
Pore volume
Average pore
diameter (nm)
2
ꢀ1
3
ꢀ1
)
(cm g
)
Fig. 1 NO conversion over Ti–Ce, Ti–Ce–W and Ti–Ce–Cu–W. Urea
44.33
142.61
107.78
0.133
0.065
0.091
12.22
4.13
4.90
Reaction conditions: [NO] ¼ [NH
3
] ¼ 600 ppm, [O
2
] ¼ 5 vol%, N
2
as Citric acid
ꢀ
1
balance gas, GHSV ¼ 114 000 h
.
Glycine
This journal is © The Royal Society of Chemistry 2014
RSC Adv., 2014, 4, 63909–63916 | 63911

View Article Online
RSC Advances
Paper
Fig. 3 Powder X-ray diffraction patterns of Ti0.75Ce0.15Cu0.05
0.05
W -
O2ꢀd with urea, glycine and citric acid.
XRD pattern of Ti0.75Ce0.15Cu0.05W0.05
O
(glycine) only shows
2ꢀd
anatase-phase TiO . In Ti0.75Ce0.15Cu0.05W0.05
O
2ꢀd
(urea), a
2
number of rutile-phase TiO
anatase TiO , and the diffraction peaks of CeO
are also observed. Compared to Ti0.75Ce0.15Cu0.05
urea) and Ti0.75Ce0.15Cu0.05 2ꢀd (glycine), Ti0.75Ce0.15-
2
appears besides main phase
2
2
, CuO and WO
3
W O
0.05 2ꢀd
(
W
0.05
O
Cu0.05
W
0.05
O
2ꢀd (citric acid) gives a broad and fuzzy peak in
Fig. 4 TEM images of Ti0.75Ce0.15Cu0.05
glycine (c and d); and citric acid (e and f).
W
0.05
O2ꢀd with urea (a and b);
ꢁ
theta angles of anatase TiO (2q ¼ 25.31 ) and rutile-phase TiO
2
2
ꢁ
(
2q ¼ 27.19 ), instead of high intensity narrower peaks. It
suggests that Ti Ce Cu
W
O
(citric acid) is in
0
.75
0.15
0.05 0.05 2ꢀd
amorphous state or even on non-crystalline material structure.
Through an analysis of TEM image (Fig. 4a), the particles in
on Ti0.75Ce0.15Cu0.05
0.05
W O2ꢀd (urea) shown by XRD, more these
species dispersed on the surface could not attribute to increase
the amount of active sites on the catalyst surface and improve
SCR performance. The surface atomic concentration of Ti on
Ti0.75Ce0.15Cu0.05
W
0.05
O
2ꢀd (urea) sample obtains large size
around 18 nm. The HRTEM image (Fig. 4b) presents lattice
fringes of anatase TiO (101), rutile TiO (110) and CeO (111)
2
2
2
Ti0.75Ce0.15Cu0.05W0.05
O
(glycine) is close to that on Ti0.75-
2ꢀd
planes with the interplanar spacing of about 0.348 nm, 0.323
nm and 0.315 nm respectively. And the polycrystalline ring
pattern consists of concentric diffraction spots. From Fig. 4c
Ce0.15Cu0.05W0.05
O
(citric acid), which is higher than that on
O2ꢀd (urea). It indicates that Ce, Cu and
2ꢀd
0.05
Ti0.75Ce0.15Cu0.05
W
W are most highly dispersed in TiO
2ꢀd (citric acid).
As reported in the literature for catalysts loaded with Ce, Ce
2
on Ti0.75Ce0.15Cu0.05-
and d, sample Ti0.75Ce0.15Cu0.05
spongy structure with middle particle size of 6 nm and shows
anatase TiO (101) planes. The sample of Ti0.75Ce0.15Cu0.05
2ꢀd (citric acid) is broken into pieces and observed with
0.05
W O2ꢀd (glycine) performs
W O
0.05
2
-
3+
4+
exists in the form of the intermediate between Ce and Ce
oxidation state, which could create surface oxygen vacancies
and be benecial for the SCR activity.
W O
0.05
particle size of 3 nm (Fig. 4e). In its HRTEM image (Fig. 4f), it is
hard to nd out any distinct crystal planes, indicating that the
sample is not crystallized well, which is consistent with the XRD
result. The presence of weak rings in the SAED pattern for the
citric acid sample is also an evidence for the loss of crystallinity.
This may be caused by different strength of the interaction
10,21
Fig. 5 shows the Ce 3d
0
0
XPS lines and their peak ttings. The bands labeled u and v
10
1
represent the 3d 4f initial electronic state corresponding to
3+
10
0
Ce , and the other six labeled peaks represent the 3d 4f state
4+ 26
3+
3+
4+
of Ce . The Ce /(Ce + Ce ) ratio of citric acid sample is
3
between fuels and metal nitrates. Citric acid is an acidic
substance, while urea is alkaline in nature and glycine has
amphoteric character. Hydrolysis of metal ion favors alkaline Table 3 XPS results of the catalysts prepared with different fuels
condition, so acid may hinder the hydrolytic action of metal ion
complexes. It makes citric acid stable bonding with metal ions,
Surface atomic concentration (%)
which could nally result in the difficulty of crystallization.
According to XPS results reported in Table 3, the surface condition
3+
Sample prepared
Ce /
1
2
3+
4+
Ti
Ce Cu W
O
(OA + OA )/O (Ce + Ce )
atomic concentration of Ce, Cu and W is most on Ti0.75Ce0.15
2ꢀd (urea), and following is Ti0.75Ce0.15Cu0.05
2ꢀd (glycine), and the least is Ti0.75Ce0.15Cu0.05
citric acid). For Ce, Cu, W species exist in the crystalline phase
-
-
Urea
Citric acid
Glycine
15.4 4.6 2.9 5.2 71.9 21.9
20.5 3.5 1.1 3.3 71.6 46.0
20.8 4.2 1.9 3.8 69.3 15.1
27.0
34.0
33.5
Cu0.05
0.05
0.05
W O
O
W
0.05 2ꢀd
W O
(
63912 | RSC Adv., 2014, 4, 63909–63916
This journal is © The Royal Society of Chemistry 2014

View Article Online
Paper
RSC Advances
follows the sequence of Ti0.75Ce0.15Cu0.05
Ti0.75Ce0.15Cu0.05W0.05 (urea) > Ti0.75Ce0.15Cu0.05W0.05-
2ꢀd
W
0.05
O
2ꢀd (citric acid)
>
O
O2 (glycine) (shown in Table 3).
ꢀd
Though the eld of nanomaterial is moving towards atomi-
cally precise activity in order to better correlate structure
28–30
performance,
it is still necessary to analyse size and shape-
dependent activity. Hence, by comparison, the fuel urea does
not cause an important change to the XPS result, which means
that the other properties of sample (such as particle size and
crystallite state) lead to its bad SCR performance. From the XPS
result, the biggest difference between citric acid and glycine
1
2
prepared samples is the (OA + OA ) ratio. But it is hard to come
1
2
to the conclusion that the higher (O
A
+ O
A
) ratio result in good
2ꢀd (citric
) ratio might enhance high
, and further inhibit SCR reaction.
high temperature activity over Ti0.75Ce0.15Cu0.05
W O
0.05
Fig. 5 XPS results of Ce 3d of Ti0.75Ce0.15Cu0.05
with different fuels.
W
0.05
O
2ꢀd prepared
1
2
acid), because the higher (O
A A
+ O
temperature oxidation of NH
3
Therefore, the fuel citric acid is most likely to improve the
activity of the product by affecting the morphology, such as
nanoparticle size and high specic surface areas.
slightly higher than that of glycine sample, but much higher
than that of urea sample (Table 3).
The XPS spectra of O 1s are presented in Fig. 6. Two kinds of
3
.3 Effects of F/O ratio on catalytic performance
surface oxygen species can be observed aer tting for Ti0.75
2ꢀd (urea), three for Ti0.75Ce0.15Cu0.05
2ꢀd (glycine), and four for Ti0.75Ce0.15Cu0.05
acid), respectively. In all cases, intense peaks are around
-
From the study on effects of fuel types, both catalysts synthe-
sized using citric acid and glycine performs good activity at low
Ce0.15Cu0.05
W
0.05
O
W
0.05
-
O
0.05
W O2ꢀd (citric
range temperature. But the structure of Ti0.75Ce0.15Cu0.05
2ꢀd (citric acid) is in amorphous state, which can be less
chemically stable. Therefore, to study the effects of F/O ratio on
the physicochemical characteristics of catalysts, Ti0.75Ce0.15-
0.05
W -
2
ꢀ
O
5
30.04–530.35 eV, attributed to the lattice oxygen O (denoted
27
as OL). Ionization characteristics of oxygen species integrated
ꢀ
2ꢀ
1
in the material as OH or O
(denoted as OA ) can be observed
2
Cu0.05
fuel.
0.05
W O2ꢀd samples were synthesized using glycine as a
at around 530.91 eV. The peaks at 531.56–531.75 eV are
assigned to the ionization of oxygen species that could allow
compensation for some deciencies in the subsurface of tran-
The performance of the catalyst samples with F/O ratio of
0.5, 1 and 1.5 is shown in Fig. 7. It is noted that the F/O ratio has
an obvious inuence on low temperature SCR activity. The
sample Ti0.75Ce0.15Cu0.05
performance than Ti0.75Ce0.15Cu0.05
Ce0.15Cu0.05W0.05
conversion ratio combined with N
above 300 C. As shown in Fig. 7, the NO conversion ratio of
Ti0.75Ce0.15Cu0.05
2
ꢀ
sition metal oxides (denoted as O
A
), such as O species. The
peaks at higher binding energy (532.36.1–533.19 eV) are char-
3
W O2ꢀd (1.5) shows better SCR
0.05
A
acteristic of weakly adsorbed species (denoted as O ). Surface
2
ꢀ
ꢀ
1
2
W
0.05
ꢁ
O
2ꢀd (0.5) and Ti0.75
-
chemisorbed oxygen, such as O2 or O species (OA and OA ),
is thought to be more reactive in oxidation reactions due to its
O
(1.0) under 200 C, but has lowest NO
2ꢀd
10,21
1
2
2
selectivity (Fig. 7 inset)
higher mobility.
The relative ratio of (OA + OA ), calculated
+ O )/O (O is only observed in case of citric acid
A A A
sample, while in other two samples, O
ꢁ
1
2
1
by (O
ꢁ
1
W
0.05
O
2ꢀd (1.5) reaches 70% at 150 C, but the
A
is considered as zero),
Fig. 7 NO conversion and N
2
selectivity over Ti0.75Ce0.15Cu0.05-
W
0.05
O
2ꢀd prepared with different F/O ratios. Reaction conditions:
Fig. 6 XPS results of O 1s of Ti0.75Ce0.15Cu0.05
with different fuels.
W
0.05
O
2ꢀd prepared [NO] ¼ [NH
3
] ¼ 600 ppm, [O
2
] ¼ 5 vol%, N as balance gas, GHSV ¼
2
ꢀ
1
114 000 h
.
This journal is © The Royal Society of Chemistry 2014
RSC Adv., 2014, 4, 63909–63916 | 63913

View Article Online
RSC Advances
Paper
corresponding ratios of Ti0.75Ce0.15Cu0.05
0.05
W O2ꢀd (0.5) and
Ti0.75Ce0.15Cu0.05W0.05
O
(1.0) are no more than 40%. High F/
2ꢀd
O ratio means sufficient fuels for reduction of oxidizer salts,
23,31
which may optimize the morphology of catalysts,
and then
improve the catalytic performance. To verify it, BET surface
areas, pore volumes, and pore sizes of the samples were checked
and are shown in Table 4.
It can be seen that BET surface areas increase with the
increase of F/O ratio, for instance, the BET surface area of
Ti0.75Ce0.15Cu0.05W0.05
Ti0.75Ce0.15Cu0.05W0.05
O
O
(1.5) is two times larger than that of
(0.5). It was reported that the BET
2ꢀd
2ꢀd
surface area of the product increases with the F/O ratio, because
a larger amount of gases release from metal glycinates decom-
position, consequently enhances powder porosity and surface
5,23,32
Fig. 8 Powder X-ray diffraction patterns of Ti0.75Ce0.15Cu0.05W0.05-
area.
The average pore diameters in these samples were
O2ꢀd with F/O ratios of 0.5, 1 and 1.5.
2
.98–5.03 nm. Sample Ti0.75Ce0.15Cu0.05
0.05
W O
2ꢀd (0.5) obtained
3
ꢀ1
the largest pore volume (0.091 cm g ).
XRD peaks of three samples (Fig. 8) closely conform to
anatase-phase TiO2 (PDF-ICDD 21-1272), without any rutile-
XPS results are reported in Table 5, Fig. 10 and 11. In Table 5,
the obvious differences of three samples are from the surface
atomic concentration of Ce, Ti and O on Ti0.75Ce0.15Cu0.05
2ꢀd (1.5), which have most surface Ce and O atoms but
least surface Ti atoms. As discussed, the XPS lines of Ce 3d can
phase TiO . In the patterns of three samples, there are no
2
-
x x x
diffraction peaks of CeO , CuO , WO or any other mixed oxides.
W O
0.05
It indicates that Ce, Cu and W elements are in a highly
dispersed state and in amorphous form on titania, or the
formed crystallites are too small to be detected by XRD.
0
0
be tted to 8 peaks, in which the peaks labeled u and v are
3
+
4+
attributed to Ce and other peaks are related to Ce (Fig. 10).
The morphology of Ti0.75Ce0.15Cu0.05
analyzed in Section 3.2 (based on Fig. 4c and d). Fig. 9 shows the
morphology of Ti0.75Ce0.15Cu0.05W0.05 produced with 0.5
0.05
W O2ꢀd (1.0) has been
3
+
3+
4+
The Ce /(Ce + Ce ) ratio of Ti0.75Ce0.15Cu0.05
is slight higher than Ti0.75Ce0.15Cu0.05
Ti0.75Ce0.15Cu0.05 2ꢀd (0.5) (as shown in Table 5). From
Fig. 11, the O 1s lines of these samples can be tted into three
peaks. The ratio of more reactive oxygen species (O
Ti0.75Ce0.15Cu0.05W0.05 (1.5) sample is highest, and
following is Ti0.75Ce0.15Cu0.05W0.05 (1.0), and the least is
W
0.05
O
2ꢀd (1.0)
W
0.05
O2ꢀd (1.5) and
O
2ꢀd
W
0.05
O
and 1.5 ratios of glycine. Spongy structure composed by very
small particles is observed in the two samples (Fig. 9a and c).
2
A
) on
The particle size (6–21 nm) is not uniform in the Ti0.75Ce0.15
-
O
2ꢀd
Cu0.05
0.05
W O
2ꢀd (0.5) sample. It is caused by nonhomogeneous
O
2ꢀd
distribution of energy during the combustion process. The
increase of F/O ratio results in smaller size particles with more
porous morphology, as shown in Fig. 9c. The typical nano-
Ti0.75Ce0.15Cu0.05
0.05
W O2ꢀd (0.5) (Table 5). Hence, it suggests
4
+
that more amount of fuel may conduce to transform Ce to
3
+
Ce irons, leading to the increase of surface chemisorbed
oxygen. Once the amount of fuel is exceed the one calculated by
particles well disperse in Ti0.75Ce0.15Cu0.05W0.05
O
2ꢀd
(1.5)
sample with particle size in the range of 6–7 nm. Combustion
synthesis between oxidizer salts and fuels is a self-sustaining
exothermic redox process. In the cases of larger F/O ratio, the
combustion rate of oxidizer salts is higher, which provides more
heat necessary to decompose the oxidizer salts into smaller
3
+
stoichiometric F/O ratio, which may not keep increasing of Ce
irons ratio, but be benecial for the richness of surface Ce and O
4
particles. An additional benet of larger F/O is to generate
larger amount of gas, leading to more drastic collisions between
the gases and the product particles, which leaves the product
more porous and spongy. As anatase-phase TiO (110) planes
2
show in HRTEM images and polycrystalline rings present in
SAED patterns (Fig. 9b and d), the two samples are well crys-
tallized, which is consistent with XRD results.
Table 4 Characterization of the catalysts prepared with different F/O
ratios
Sample prepared
condition
BET surface
area (m g
Pore volume
Average pore
diameter (nm)
2
ꢀ1
3
ꢀ1
)
(cm g
)
F/O ¼ 0.5
F/O ¼ 1
75.31
107.78
184.80
0.066
0.091
0.047
5.03
4.90
2.98
Fig. 9 TEM images of Ti0.75Ce0.15Cu0.05W0.05
0.5 (a and b) and F/O ratios of 1.5 (c and d).
O
2ꢀd
with F/O ratios of
F/O ¼ 1.5
63914 | RSC Adv., 2014, 4, 63909–63916
This journal is © The Royal Society of Chemistry 2014

View Article Online
Paper
RSC Advances
Table 5 XPS results of the catalysts prepared with different F/O ratios and are characterized to observe the structure and morphology.
Both fuel/oxidant ratios and fuel types play a role in character-
ization of the product, leading to an effect on SCR performance.
The catalysts synthesized using citric acid and glycine
performs good activity at low range temperature. But the crys-
Surface atomic concentration (%)
Sample prepared
condition
2
A
3+
3+
4+
Ti
Ce Cu
W
O
O
/O Ce /(Ce + Ce
)
F/O ¼ 0.5
F/O ¼ 1
20.6 4.6 1.4 3.5 69.9 13.5
20.8 4.2 1.9 3.8 69.3 15.1
12.0 6.9 1.4 3.8 75.9 18.4
27.0
33.5
31.5
tallization of Ti0.75Ce0.15Cu0.05
The inuence of urea is stronger, that the product phase is
evidently changed and the SCR activity is drastically worsen.
0.05
W O2ꢀd (citric acid) is not well.
F/O ¼ 1.5
The catalyst Ti0.75Ce0.15Cu0.05
0.05
W O2ꢀd (1.5) using glycine as a
fuel with a higher F/O exhibits higher low temperature activity,
3
+
due to its high specic surface areas and richness of Ce irons
and surface chemisorbed oxygen. Therefore, glycine as a fuel
with a fuel/oxidant of 1.5 should be the best condition for the
synthesis of catalyst Ti0.75Ce0.15Cu0.05
0.05 2ꢀd
W O .
Acknowledgements
This work was supported by the National Natural Science
Foundation of China (51176118).
References
1
2
3
4
5
6
C. Y. Zhu, A. Nobuta, G. Saito, I. Nakatsugawa and
T. Akiyama, Adv. Powder Technol., 2014, 25, 342–347.
M. A. Aghayan and M. A. Rodr ´ı guez, Mater. Sci. Eng., C, 2012,
Fig. 10 XPS results of Ce 3d of Ti0.75Ce0.15Cu0.05
with different F/O ratios.
0.05
W O2ꢀd prepared
32, 2464–2468.
S. Samantaray, B. G. Mishra, D. K. Pradhan and G. Hota,
Ceram. Int., 2011, 37, 3101–3108.
E. Thomas, D. H. Lee, M. Y. Yoon, S. H. Ehrman and
H. J. Hwang, Ceram. Int., 2011, 37, 2269–2274.
F. Deganello, G. Marc `ı and G. Deganello, J. Eur. Ceram. Soc.,
2009, 29, 439–450.
L. Lietti, I. Nova, G. Ramis, L. Dall'Acqua, G. Busca,
E. Giamello, P. Forzatti and F. Bregani, J. Catal., 1999, 187,
419–435.
7
8
9
T. Morimune, H. Yamaguchi and Y. Yasukawa, Exp. Therm.
Fluid Sci., 1998, 18, 220–230.
N. Y. Topsoe, H. Topsoe and J. A. Dumesic, J. Catal., 1995,
151, 226–240.
S. Roy, B. Viswanath, M. S. Hegde and G. Madras, J. Phys.
Chem. C, 2008, 112, 6002–6012.
1
1
1
1
1
1
0 B. Guan, H. Lin, L. Zhu and Z. Huang, J. Phys. Chem. C, 2011,
115, 12850–12863.
1 B. Guan, H. Lin, L. Zhu, B. Tian and Z. Huang, Chem. Eng. J.,
Fig. 11 XPS results of O 1s of Ti0.75Ce0.15Cu0.05
with different F/O ratios.
0.05
W O2ꢀd prepared
2
012, 181–182, 307–322.
2 B. Thirupathi, R. Koirala and P. G. Smirniotis, Appl. Catal., B,
013, 140–141, 289–298.
3
+
atoms. Hence, the surface atomic concentration of Ce irons
and surface chemisorbed oxygen are enhanced a lot in fact,
which nally contribute to improve the low temperature NH3-
SCR activity.
2
3 W. Shan, F. Liu, H. He, X. Shi and C. Zhang, Chem. Commun.,
2011, 47, 8046–8048.
4 J. Xue, X. Wang, G. Qi, H. Wang, M. Shen and W. Li, J. Catal.,
2013, 297, 56–64.
5 J. Park, H. J. Park, H. H. Baik, I. Nam, C. Shin, J. Lee,
B. K. Cho and S. H. Oh, J. Catal., 2006, 240, 47–57.
4
. Conclusions
A Ti–Ce–Cu–W catalyst obtains high low temperature activity. A 16 B. Thirupathi and P. G. Smirniotis, J. Catal., 2012, 288, 74–
series of catalysts Ti0.75Ce0.15Cu0.05W0.05 are synthesized 83.
with solution combustion synthesis (SCS) technique under 17 P. R. Ettireddy, N. Ettireddy, B. Thirupathi, R. Pardemann
O
2ꢀd
different conditions. Catalysts are tested to check the activity
and P. G. Smirniotis, J. Catal., 2012, 292, 53–63.
This journal is © The Royal Society of Chemistry 2014
RSC Adv., 2014, 4, 63909–63916 | 63915

View Article Online
RSC Advances
Paper
1
8 B. Thirupathi and P. G. Smirniotis, Appl. Catal., B, 2011, 110, 25 R. K. Lenka, T. Mahata, P. K. Sinha and A. K. Tyagi, J. Alloys
95–206. Compd., 2008, 466, 326–329.
9 J. M. Kim, H. J. Kwon, L. Heo, I. Nam, B. K. Cho, 26 B. M. Reddy, A. Khan, Y. Yamada, T. Kobayashi, S. Loridant
1
1
J. W. Choung, M. Cha and G. K. Yeo, Appl. Catal., B, 2012,
26, 9–21.
0 W. Xu, Y. Yu, C. Zhang and H. He, Catal. Commun., 2008, 9,
453–1457.
and J. C. Volta, Langmuir, 2003, 19, 3025–3030.
27 J. C. Dupin, D. Gonbeau, P. Vinatier and A. Levasseur, Phys.
Chem. Chem. Phys., 2000, 2, 1319–1324.
1
2
2
2
2
2
1
28 G. Li and R. Jin, Acc. Chem. Res., 2013, 46(8), 1749–1758.
1 W. Shan, F. Liu, H. He, X. Shi and C. Zhang, Appl. Catal., B, 29 H. Li, L. Li and Y. Li, Nanotechnol. Rev., 2013, 2(5), 515–528.
2012, 115–116, 100–106.
30 J. Liu, K. S. Krishna, Y. B. Losovyj, S. Chattopsdhyay,
N. Lozova, J. T. Miller, J. J. Spivey and C. S. S. R. Kumar,
Chem.–Eur. J., 2013, 19(31), 10201–10208.
2 X. Gao, X. S. Du, L. W. Cui, Y. C. Fu, Z. Y. Luo and K. F. Cen,
Catal. Commun., 2010, 12, 255–258.
3 A. S. Mukasyan, P. Epstein and P. Dinka, Proc. Combust. Inst., 31 H. Nasiri, E. Bahrami Motlagh, J. Vahdati Khaki and
007, 31, 1789–1795. S. M. Zebarjad, Mater. Res. Bull., 2012, 47, 3676–3680.
4 S. R. Jain, K. C. Adiga and V. R. Verneker, Combust. Flame, 32 Y. Li, L. Xue, L. Fan and Y. Yan, J. Alloys Compd., 2009, 478,
981, 40, 71–79. 493–497.
2
1
63916 | RSC Adv., 2014, 4, 63909–63916
This journal is © The Royal Society of Chemistry 2014