Nickel-doped Mn/TiO2 as an efficient catalyst for the low-temperature SCR of NO with NH3: Catalytic evaluation and characterizations

Article abstract of DOI:10.1016/j.jcat.2012.01.003

The Mn/TiO₂ and a series of Mn-Ni/TiO₂ catalysts were prepared by adopting incipient wetness technique and investigated for the low-temperature SCR of NO with NH₃ in the presence of excess oxygen. Our XPS results illustrated that the MnO₂ is the dominant phase with respect to the Mn₂O₃ phase (Mn⁴⁺/Mn ³⁺ = 22.31, 96%), thus leading to a large number of Mn⁴⁺ species (Mn⁴⁺/Ti) over the titania support for the Mn-Ni(0.4)/TiO₂ catalyst. It is remarkable to note that the SCR performance of all the nickel-doped Mn/TiO₂ catalysts is accurately associated with the surface Mn⁴⁺ concentrations. The co-doping of nickel into the Mn/TiO₂ with 0.4 Ni/Mn atomic ratio promotes the formation of surface MnO₂ phase and inhibits the formation of surface Mn₂O₃ sites. Our TPR results revealed that the addition of nickel oxide to titania-supported manganese results in the stabilization of the former in the form of MnO₂ rather than Mn₂O ₃. Our TPR data results are in agreement with XPS results that the absence of the high-temperature (736 K) peak indicates that the dominant phase in the Mn-Ni/TiO₂ catalysts is MnO₂. The low-temperature reduction peak is shifted to much lower temperatures in nickel-doped Mn/TiO ₂ catalysts. This increase in reducibility and the extremely dominant MnO₂ phase seem to be the reason for the high SCR activity of the Mn-Ni/TiO₂ catalysts.

Full text of DOI:10.1016/j.jcat.2012.01.003

Journal of Catalysis 288 (2012) 74–83
Contents lists available at SciVerse ScienceDirect
Journal of Catalysis
journal homepage: www.elsevier.com/locate/jcat
Nickel-doped Mn/TiO2 as an efficient catalyst for the low-temperature SCR
of NO with NH3: Catalytic evaluation and characterizations
⇑
Boningari Thirupathi, Panagiotis G. Smirniotis
Chemical Engineering Program, School of Energy, Environmental, Biological and Medicinal Engineering, University of Cincinnati, Cincinnati, OH 45221-0012, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
The Mn/TiO2 and a series of Mn–Ni/TiO2 catalysts were prepared by adopting incipient wetness technique
and investigated for the low-temperature SCR of NO with NH3 in the presence of excess oxygen. Our XPS
Received 26 October 2011
Revised 4 January 2012
Accepted 9 January 2012
Available online 22 February 2012
results illustrated that the MnO2 is the dominant phase with respect to the Mn2O3 phase (Mn4+
/
Mn3+ = 22.31, 96%), thus leading to a large number of Mn4+ species (Mn4+/Ti) over the titania support
for the Mn–Ni(0.4)/TiO2 catalyst. It is remarkable to note that the SCR performance of all the nickel-doped
Mn/TiO2 catalysts is accurately associated with the surface Mn4+ concentrations. The co-doping of nickel
into the Mn/TiO2 with 0.4 Ni/Mn atomic ratio promotes the formation of surface MnO2 phase and inhibits
the formation of surface Mn2O3 sites. Our TPR results revealed that the addition of nickel oxide to titania-
supported manganese results in the stabilization of the former in the form of MnO2 rather than Mn2O3.
Our TPR data results are in agreement with XPS results that the absence of the high-temperature (736 K)
peak indicates that the dominant phase in the Mn–Ni/TiO2 catalysts is MnO2. The low-temperature
reduction peak is shifted to much lower temperatures in nickel-doped Mn/TiO2 catalysts. This increase
in reducibility and the extremely dominant MnO2 phase seem to be the reason for the high SCR activity
of the Mn–Ni/TiO2 catalysts.
Keywords:
Low-temperature NH3-SCR
Manganese dioxide (Mn4+
Nickel oxide (NiO)
XPS
)
Hydrothermal stability
Nitric oxide
Published by Elsevier Inc.
1. Introduction
at low temperatures. The manganese nitrate has been predomi-
nantly used in the preparation of the supported system in the
The current technology for post-combustion removal of NO is
the selective catalytic reduction (SCR) of NO by ammonia at med-
ium to high temperatures. Industrially adopted catalysts for the
practical applications of SCR reaction are V2O5–MoO3/TiO2 and
V2O5–WO3/TiO2 [1,2]. However, the existing commercial catalytic
system suffers from the drop of N2 selectivity at high temperatures
(573–673 K), the toxicity of vanadium pentoxide, high conversion
of SO2 to SO3, and need to reheat the stack gas [3–6]. The develop-
ment of low-temperature SCR catalysts working at 150–250 °C can
offer a great solution to avoid all the problems associated with the
existing commercial catalytic system.
The manganese oxide is well known for its high activity in the
SCR of NOx at low temperatures [7–10]. In recent years, manga-
nese-based catalysts attracted much attention due to their high
activity for various reactions such as direct decomposition of nitric
oxide [11], CO oxidation [12], CH4 oxidation [13], and total oxida-
tion of volatile organic compounds [14,15]. In all these reactions,
the manganese oxide undergoes oxidation–reduction cycles (redox
process). The Mn4+ species and their redox process might be the
reason for the high activity in the SCR reaction of NO with NH3
SCR reaction of NOx studies, because of its high solubility in water
and the ease of removal of the nitrate anion during calcination. The
calcination temperature and metal loading have been found to
determine the final oxidation state of the supported manganese
oxide. On a support, MnO2 is formed mainly below 700 K and
Mn2O3 is formed at 900 K [16–18]. The supported Mn3O4 phase
can be prepared by the slow reduction of MnO2 at the temperature
range of 665–700 K using hydrogen. Reduction at high tempera-
tures in H2 (900 K) is proposed to result in the formation of surface
MnO at higher loadings [19].
The interaction of Mn with promoters or co-dopant species also
influences its oxidation state [20]. For example, the co-doping of
Mn/TiO2 catalyst with copper cations formed mainly Mn2O3 phase
at 723 K calcination temperature [21]. The interaction of Cr species
with the Mn/TiO2 catalyst produced surface Mn3+ species at 923 K
calcination temperature [22]. The contact of Ce with Mn by 0.8 M
ratio creates surface Mn2O3, MnO, and Mn5O8 phases but not
MnO2 [23].
In the present work, we investigated whether phase change can
occur with co-dopant nickel cations and promote the manganese
phase transformation from Mn2O3 to MnO2, thus increase in cata-
lytic activity and stability of the catalyst. For this purpose, nickel
metal was induced in the Mn/TiO2 catalyst by varying the Ni/Mn
⇑
Corresponding author. Fax: +1 513 556 3473.
E-mail address: panagiotis.smirniotis@uc.edu (P.G. Smirniotis).
0021-9517/$ - see front matter Published by Elsevier Inc.
doi:10.1016/j.jcat.2012.01.003

B. Thirupathi, P.G. Smirniotis / Journal of Catalysis 288 (2012) 74–83
75
atomic ratio (Ni/Mn = 0, 0.2, 0.4, 0.6, and 0.8). However, the extre-
2.4. Temperature-programmed reduction (H2-TPR)
mely dominant surface MnO2 phase with respect to the Mn2O3
phase (Mn4+/Mn3+ = 22.31, 96%) was determined in our XPS analy-
sis of the Mn–Ni(0.4)/TiO2 catalyst. The high dispersion of surface
Mn4+ species on the TiO2 support is achieved by the Mn–Ni/TiO2
catalyst with Ni/Mn atomic ratio equal to 0.4. It is remarkable to
note that the SCR performance of all the prepared catalysts is accu-
rately correlated with the number of surface Mn4+ species (Mn4+/Ti
ratio). The concentration and acidity (distribution) of Lewis acid
sites on Mn/TiO2 catalyst have significantly improved by the
doping with Ni cations. The efficiency for NO removal, N2 selectiv-
ity, temperature window, and time on stream stability (durability)
enhanced greatly when Ni was substituted into the Mn/TiO2
catalyst.
Hydrogen temperature-programmed reduction (H2-TPR) of var-
ious catalyst samples was performed by means of an automated
catalyst characterization system (Micromeritics, model AutoChem
II 2910), which incorporates a thermal conductivity detector (TCD).
The experiments were carried out at a heating rate of 5 °C/min. The
reactive gas composition was 10% H2 in argon with a flow rate of
20 mL minꢁ1 (STP). The amount of H2 consumed by the catalyst
sample in a given temperature range (in mmol gꢁ1) was calculated
by the integration of the corresponding TCD signal intensities. Prior
to the analysis, approximately 50 mg of the catalysts were pre-
treated at 200 °C for 2 h in ultra-high-pure helium (30 mL minꢁ1
)
stream after preheating samples were tested by increasing the
temperature from 50 to 800 °C. The temperature was then kept
constant at 800 °C until the signal of hydrogen consumption re-
turned to the initial values. The TPR runs were carried out with a
linear heating rate (10 °C/min) in a flow of 4% H2 in argon with a
flow rate of 20 mL minꢁ1. The hydrogen consumption was mea-
sured quantitatively by a thermal conductivity detector. A mixture
of isopropanol and liquid nitrogen was used in the trapper to col-
lect the formed water during the TPR experiment.
2. Experimental
2.1. Catalysts preparation
Mn/TiO2 and a series of Mn–Ni/TiO2 (Ni/Mn atomic ratio = 0.2,
0.4, 0.6, 0.8) catalysts were prepared by adopting a wet impregna-
tion method. TiO2 anatase (Hombikat UV 100 from Sachtleben Che-
mie) was used as support material to prepare these catalysts.
Manganese nitrate (Mn(NO3)2ꢀxH2O, 99.99%, Aldrich) and nickel ni-
trate (Ni(NO3)2ꢀ6H2O, Aldrich) were used as the source of manga-
nese and nickel, respectively. The manganese loadings were
selected as 5 wt.%, and the nickel loadings were varied ranging
from 0 to 4 wt.%. The metal components of the catalysts are de-
noted as atomic ratios. All the ratios of the catalysts in this paper
are Ni/Mn = 0, 0.2, 0.4, 0.6, 0.8. For convenience, we have given
as Mn–Ni(x)/TiO2, where x indicates the atomic ratio of nickel/
manganese. To achieve this, the required quantities of precursors
were added to a 100-mL beaker containing 2.0 g of support in
100 mL deionized water. The excess water was then slowly evapo-
rated on a water bath with continuous stirring at 70 °C. For com-
parison purposes, TiO2 anatase (Hombikat) support alone was
mixed in deionized water and then water was evaporated with
continuous heating and stirring. The resulting materials were
oven-dried at 120 °C for 12 h and were ground and sieved (80–
120 mesh) to obtain homogeneous powder. Prior to the reaction
studies, the powder was calcined in a tubular oven at 400 °C for
2 h under continuous air flow (150 mL min–1).
2.5. Oxygen pulse chemisorptions
The site time yield (STY), oxygen uptake, and the active metal
dispersion on the surface of the titania support were determined
by oxygen chemisorption measurements. These measurements
were obtained in a pulse (Micromeretics Autochemi I 2910 system)
mode using He as carrier gas (30 STP cm3 minꢁ1). Prior to the anal-
ysis, approximately 0.050 g of catalyst samples were reduced in
flowing H2 (50 STP cm3 minꢁ1) at 250 °C for 2 h and then flushed
at the same temperature for 30 min in the He carrier flow [10].
Then, oxygen pulses (1 mL loop volume) were injected onto the
carrier gas until saturation of the sample was attained. The oxygen
uptake was quantified by a TCD connected to a 2910 Autochem I
(Micromeretics instrument).
2.6. NH3-TPD
NH3-TPD data were collected on a Micromeritics Autochem
2910 Automated Catalyst Characterization System. Each sample
(ca. 100 mg of finely ground powder) was initially pretreated
through heating at 200 °C in an ultra-high-pure He stream
(30 mL/min) with a 2 h hold. After that, the furnace temperature
was lowered to 100 °C, and the samples were then saturated with
anhydrous NH3 (4% in He) at a flow rate of 30 mL/min for 1 h. The
sample was flushed with 30 mL/min of He for 2 h to remove
weakly bound (physisorbed) NH3, after which the sample temper-
ature was reduced to 50 °C. Once a stable baseline by thermal con-
ductivity detector had been achieved, the temperature was then
ramped from 50 °C at a rate of 5 °C/min to 800 °C.
2.2. X-ray diffraction
The powder XRD patterns were recorded on a Phillips Xpert dif-
fractometer using nickel-filtered Cu Ka (wavelength 0.154056 nm)
radiation source and a scintillation counter detector. An aluminum
holder was used to support the catalyst samples. The intensity data
were collected over a 2h range of 10–80° with a step size of 0.025°
and a step time of 0.25 s. Crystalline phases were identified by
comparison with the reference data from International Center for
Diffraction Data (ICDD) files.
2.7. In situ FT-IR spectroscopy
FT-IR spectra were recorded using a Bio-Rad (FTS-40). The scans
were collected at a scan speed of 5 kHz, resolution of 2.0, and an
aperture opening of 2.0 cmꢁ1. Sixteen scans were averaged for each
normalized spectrum. Circular self-supporting thin wafers (8 mm
diameter) consisting of 12 mg of material were used for the study.
The wafers were placed in a high-temperature cell with CaF2 win-
dows and purged in situ in the IR cell with prepurified-grade he-
lium (30 mL minꢁ1, Wright Brothers) at 473 K for 2 h to remove
any adsorbed impurities. After this step, the samples were cooled
to 323 K and NH3 (3.9 vol.% in He) was introduced into the cell with
a flow of 30 mL minꢁ1 for 1 h at 323 K to ensure complete satura-
2.3. BET surface area and pore volume measurements
N2 adsorption–desorption isotherms were obtained at liquid
nitrogen temperature (77 K) using Micromeretics Gemini surface
area apparatus. Prior to N2 adsorption, 0.08–0.1 g of the samples
were degassed under helium atmosphere at 200 °C for 2 h in the
degassing port of the instrument. The adsorption isotherms of
nitrogen were collected by using approximately six values or rela-
tive pressure ranging from 0.05 to 0.99 and by taking 0.162 nm2 as
the molecular area of the nitrogen molecule.

76
B. Thirupathi, P.G. Smirniotis / Journal of Catalysis 288 (2012) 74–83
tion of the sample. Physisorbed ammonia was removed by flushing
the wafer with helium for sufficient time (3 h) at 373 K. Subse-
quently, the in situ FT-IR spectra were recorded by the evacuation
of ammonia at successive temperatures 323, 373, 423, 473, and
523 K.
3. Results and discussion
Initially, we evaluated the potential catalytic performance of all
the prepared catalysts (Mn/TiO2 and Mn–Ni/TiO2 with Ni/Mn
atomic ratio = 0.0, 0.2, 0.4, 0.6, 0.8) at 200 °C to optimize the Ni
content and to ensure the influence of Ni2+ co-cations on NO con-
version. From our results, one can observe that the partial loading
of Mn with Ni elements could indeed influence the SCR activity of
Mn/TiO2 catalyst. The catalytic performance results for the SCR of
NO with ammonia over titania-supported manganese as well as
manganese–nickel catalysts are tested at GHSV 50,000 hꢁ1 in the
presence of 2 vol.% of oxygen (Fig. 1). The nickel loading has a
strong influence on the conversion and N2 selectivity, since nickel
loading with the Ni/Mn ratio = 0.4 catalyst exhibits a maximum
conversion of 100% with huge N2 selectivity at 200 °C (Fig. 1). This
would indicate that an optimal dispersion of Mn–Ni species on the
support surface is attained with this amount of nickel in the Mn–
Ni/TiO2 catalyst. The Mn/TiO2 catalyst does not accomplish high
NO conversions at 200 °C. As the Ni/Mn atomic ratio increases up
to 0.4 on the TiO2-supported manganese surface, the NO conver-
sion increased monotonically and then decline for higher loadings
(Ni/Mn = 0.6, 0.8) of nickel. As can be noted from the Fig. 1, high
conversion of NO (100%), N2 selectivity achieved over the
Mn–Ni(0.4)/TiO2 catalyst. The Mn–Ni(0.4)/TiO2 afforded highly
remarkable catalytic activity and N2 selectivity relative to the other
catalysts tested with GHSV 50,000 hꢁ1 at 200 °C temperature.
With these interesting results in hand, we evaluated the SCR
activity of all the prepared catalysts to ensure about the tempera-
ture window and the influence of nickel content on the SCR activity
at the temperature range of 160–240 °C (Fig. 2). Under identical
operating conditions, the pure Mn/TiO2 catalyst demonstrated
poor catalytic performance at below 200 °C and it is evident that
the NO conversions drastically decline with time at above 200 °C.
This is probably due to the sintering of manganese species over
the surface of titania support at high temperatures. The undoped
Mn/TiO2 showed very narrow operation temperature window for
the NH3-SCR reaction, and the maximum NO conversions could
not reach 100%. Eventually, the addition of nickel species to the
Mn/TiO2 catalyst (Ni/Mn ratio = 0.4) broadened the temperature
window and demonstrated complete NO conversion at the temper-
ature range of 200–250 °C. The doping of nickel co-cations with 0.4
2.8. X-ray photoelectron spectroscopy (XPS)
The X-ray photoelectron spectroscopy (XPS) experiments were
carried out on a Pyris-VG thermo scientific X-ray photoelectron
spectrometer system equipped with a monochromatic Al K
a
(1486.7 eV) as radiation source at 300 W under UHV
a
(6.7 ꢂ 10ꢁ8 Pa). Sample charging during the measurement was
compensated by an electron flood gun. The electron takeoff angle
was 45° with respect to the sample surface. The spectra were re-
corded in the fixed analyzer transmission mode with pass energies
of 89.45 and 35.75 eV for recording survey and high-resolution
spectra, respectively. The powdered catalysts were mounted onto
the sample holder and evacuated overnight at room temperature
at a pressure on the order of 10ꢁ7 torr. Binding energies (BE) were
measured for C 1s, O 1s, Ti 2p, Mn 2p, and Ni 2p. A recorded Auger
spectrum for Mn was very weak. Sample charging effects were
eliminated by correcting the observed spectra with the C 1s bind-
ing energy (BE) value of 284.6 eV. An estimated error of 0.1 eV can
be considered for all the measurements. The overlapped Mn 2p
peaks were deconvoluted into several sub-bands by searching for
the optimal combination of Gaussian bands with the correlation
coefficients (r2) above 0.99. The Mn 2p peak was deconvoluted
using the Gaussian function.
2.9. Apparatus and catalytic experiments
The low-temperature reduction of NO by ammonia (NH3-SCR)
with excess oxygen activity measurements were carried out at
atmospheric pressure in a fixed-bed continuous flow quartz reac-
tor with an internal diameter of 6 mm. 0.1 g of catalyst (80–120
mesh) was placed in the reactor in between two glass wool plugs,
otherwise mentioned. All the gas flows were measured and cali-
brated using a digital flow meter (Humonics Hewlett Packard Opti-
flow 520). The reaction gas mixture normally consisted of 400 ppm
NO, 400 ppm NH3, 2 vol.% O2, and He in balance.
4NO þ 4NH3 þ O2 ¼ 4N2 þ 6H2O
ð1Þ
A total of 10 vol.% H2O vapor was used for the hydrothermal stabil-
ity test. The premixed gases oxygen (4% in He, Wright Brothers),
ammonia (3.99% in He, Wright Brothers), and nitric oxide (2.0% in
He, Matheson) were used as received. ISCO series D pump controller
was used to feed the 10 vol.% concentration of water vapor. The tub-
ing of the reactor system was heat-traced to prevent the formation
and deposition of ammonium sulfate/bisulfate and ammonium ni-
trate. The NO and NO2 concentrations were continually monitored
by a chemiluminescence NO/NOx detector (Eco Physics CLD 70S).
To avoid errors caused by the oxidation of ammonia in the con-
verter of the NO/NOx analyzer, an ammonia trap containing phos-
phoric acid solution was installed before the sample inlet to the
chemiluminescence detector [3]. The reactor was heated externally
via a tubular furnace regulated by a temperature controller (Omega
CN 2041), with a thermocouple inserted into the catalyst bed. Prior
to the catalytic experiments, the catalyst was activated in situ by
passing oxygen (4% in He, Wright Brothers) for 2 h at 200 °C tem-
perature. The reactants and products were analyzed online using a
Quadrapole mass spectrometer (MKS PPT-RGA) and a chemilumi-
nescence detector (Eco Physics CLD 70S). Reactant and product con-
tents in the reactor effluent were recorded only after steady state
was achieved at each temperature step. The N2 selectivity and NO
conversions were calculated as in our previous papers [7–10].
100
95
90
85
80
XNO% at 1hr
75
70
65
XNO% at 6hr
N2 Selectivity
0.0
0.2
0.4
0.6
0.8
Ni/Mn (atomic ratio)
Fig. 1. Effect on NO conversion and N2 selectivity in the SCR reaction of Mn/TiO2
and Mn–Ni/TiO2 catalysts with respect to the Ni/Mn atomic ratio; tempera-
ture = 200 °C; GHSV = 50,000 hꢁ1; feed: NO = 400 ppm, NH3 = 400 ppm, O2 = 2 vol.%,
He carrier gas, catalyst = 0.1 g, total flow = 140 mL minꢁ1. N2 selectivities at 6 h of
reaction.

B. Thirupathi, P.G. Smirniotis / Journal of Catalysis 288 (2012) 74–83
77
of continuous stream. It is apparent that the addition of nickel to
the Mn/TiO2 catalyst has enhanced the stability of the catalyst. In
particular, Mn–Ni(0.4)/TiO2 catalyst showed a remarkable perfor-
mance by exhibiting 100% NO conversion even after 240 h on
stream. The Mn–Ni(0.6)/TiO2 and Mn–Ni(0.8)/TiO2 catalysts were
rapidly deactivating with time. These results demonstrate that
the time on stream performance of the catalysts decreased in the
following order: Mn–Ni(0.4)/TiO2 ꢃ Mn–Ni(0.2)/TiO2 o Mn–
Ni(0.6)/TiO2 > Mn–Ni(0.8)/TiO2 > Mn/TiO2.
100
80
60
It is also important to evaluate the durability of SCR catalysts in
the presence of H2O vapors in the feed. Fig. 4 illustrates the toler-
ance of Mn–Ni(0.4)/TiO2 catalyst to steam poisoning for 240 h dur-
ing the SCR of NO with NH3 at low temperatures. Several reported
catalysts for the low-temperature SCR of NO with NH3 have the
problem of deactivation when water vapor was introduced into
the stream [24–26]. We carried out the SCR of NO reaction with
high gas hourly space velocity (GHSV) 50,000 hꢁ1 in the presence
of 10 vol.% H2O vapors in the feed (Fig. 4) to test the durability of
the catalyst.
Ni/Mn=0.0
40
20
Ni/Mn=0.2
Ni/Mn=0.4
Ni/Mn=0.6
Ni/Mn=0.8
160
180
200
220
240
Temperature (°C)
Initially, our best catalyst Mn–Ni(0.4)/TiO2 showed high NO
conversion in the absence of H2O (10 vol.%), a steady-state NO con-
version of 100% with N2 selectivity at 200 °C is obtained within the
first 6 h. In the present study, the Mn–Ni(0.4)/TiO2 catalyst showed
an extraordinary performance (XNO% = 100%) even in the presence
of 10 vol.% water vapor inlet concentrations. The bimetallic combi-
nation of Mn and Ni with 0.4 Ni/Mn atomic ratio produced very
stable and durable catalyst that did not deactivate for the tested
period. The N2 selectivity Mn–Ni(0.4)/TiO2 catalyst remained at
100% after 240 h on stream. These results clearly suggest that the
Mn–Ni(0.4)/TiO2 displays a satisfactory extent of stability and
durability properties. The system is successful because of its high
activity and resistance to steam poisoning. The addition of H2O
(10 vol.%) in the feed gas does not seem to create an eternal alter-
ation to the surface manganese species, most probably due to the
existence of high redox potential pairs of Mn and Ni in the struc-
ture, permitting an efficient active catalyst.
To gain additional insight into the active surface manganese
species for the low-temperature SCR reaction, the H2 tempera-
ture-programmed reduction of the pure manganese oxide (from
Mn(NO3)2ꢀxH2O, 99.99% metals basis) was carried out and evalu-
ated for the SCR reaction. The reduction profile of unsupported
manganese oxide was characterized by three different reduction
peaks at 640, 728, and 790 K (Fig. 5). A low-temperature reduction
Fig. 2. Influence of Ni/Mn atomic ratio on NO conversion in the SCR reaction at a
temperature range (160–240 °C) over Mn–Ni/TiO2 catalysts; GHSV = 50,000 hꢁ1
;
feed: NO = 400 ppm, NH3 = 400 ppm, O2 = 2 vol.%, He carrier gas, catalyst = 0.1 g,
total flow = 140 mL minꢁ1. XNO% = conversion of NO at 6 h on stream.
Ni/Mn atomic ratio greatly improved the thermal stability of the
Mn/TiO2 catalyst since the Mn–Ni(0.4)/TiO2 demonstrated steady
high NO conversions at 300 °C (not shown). At the reaction tem-
perature 200 °C, further increase in the nickel content exceeding
that of 0.4 Ni/Mn atomic ratio decreased the SCR activity of the cat-
alyst. This is due to the interaction between the TiO2 and manga-
nese–nickel oxide which becomes feeble, and the number of
participating surface manganese oxide sites decrease with the in-
crease in nickel loading. This occurrence was also proved by our
XPS analysis as we will demonstrate below. The NH3-SCR activity,
N2 selectivity, stability, and operation temperature window of the
Mn/TiO2 catalyst improved greatly by the addition of optimal nick-
el content (Ni/Mn atomic ratio = 0.4).
Subsequently, the time on stream patterns of all the prepared
catalysts were evaluated at 200 °C, to check the impact of Ni spe-
cies in time stability of the Mn/TiO2 catalyst (Fig. 3). Among all
the catalysts, undoped Mn/TiO2 (Ni/Mn = 0.0) showed poor stabil-
ity since the catalytic activity monotonically dropped in six hours
Mn-Ni(0.4)/TiO2
100
100
Mn-Ni(0.2)/TiO2
Mn-Ni(0.4)/TiO2
80
NO conversion
N2 selectivity
90
10 vol.% H2O
60
Mn-Ni(0.6)/TiO2
ON
40
20
0
10 vol.% H2O
80
70
Mn-Ni(0.2)/TiO2
Mn/TiO2
OFF
1
2
3
4
5
6 238 239 240
40
80
120
160
200
240
Time (hours)
Time on stream (hours)
Fig. 3. SCR of NO with NH3 at 200 °C temperature over Mn/TiO2 and Mn–Ni/TiO2
catalysts; (j): GHSV = 50,000 hꢁ1
feed: NO = 400 ppm, NH3 = 400 ppm,
2 = 2 vol.%, He carrier gas, total flow = 140 mL minꢁ1, XNO% = conversion of NO;
catalyst = 0.1 g.
;
Fig. 4. Influence of inlet water concentrations (10 vol.%) on NO conversion in the
SCR reaction over Mn–Ni(0.4)/TiO2 catalyst at 200 °C; feed: NO. 400 ppm NH3/NO.
O
1.0, O2. 2 vol.%, He carrier gas, catalyst. 0.1 g, GHSV. 50,000 hꢁ1
.

78
B. Thirupathi, P.G. Smirniotis / Journal of Catalysis 288 (2012) 74–83
peak (T1) is observed at 640 K, which corresponds to the reduction
of MnO2 to Mn2O3. The subsequent reduction peak at 728 (T2) is
attributed to the reduction of Mn2O3 to Mn3O4. A high-tempera-
ture peak maxima at 790 K (T3) can be interpreted as a reduction
of Mn3O4 to MnO [27]. Usually, the manganese (II) oxide (MnO)
100
80
60
40
20
0
MnO2
Mn2O3
Mn3O4
ꢀ
is a green crystalline solid with FM3m space group [28], it is inter-
esting to note that the manganese (IV) oxide sample turned into a
green crystalline solid after the treatment for the sample with H2 at
1083 K (during H2-TPR experiment). It can be concluded from this
observation that the sample finally converts into MnO but not into
metallic manganese (Mn0). From our H2-TPR results, it is evident
that the unsupported manganese (IV) oxide sample undergoes
stepwise reduction of MnO2 to Mn2O3, Mn2O3 to Mn3O4, and
Mn3O4 to MnO. Subsequently, the catalytic reduction of NO over
the pure manganese oxides with respect to the temperature was
investigated to understand the active phase of manganese for the
SCR of NO at low temperatures (Fig. 6). To avoid the phase transfor-
mations, all the manganese oxides were prepared by in situ meth-
od. Prior to the catalytic experiments of pure MnxOy phases, the
manganese oxides (MnO2, Mn2O3, Mn3O4, MnO) were synthesized
in situ by passing hydrogen through a fixed-bed quartz reactor at
carefully selected temperatures.
400
420
440
460
480
500
520
Temperature (K)
Fig. 6. Influence of Mn oxidation state on NO conversion in the SCR reaction of NO
with respect to temperature: GHSV = 8000 hꢁ1
400 ppm, O2 = 2 vol.%, He carrier gas.
; feed: NO = 400 ppm, NH3 =
Prior to the catalytic performance tests of pure Mn oxides, a
reaction with undoped Hombikat TiO2 (calcined at 400 °C) was
performed at 200 °C to observe the catalytic activity of the TiO2
support and the efficiency of our SCR unit. As estimated, the NO
and NH3 conversions were zero when only TiO2 anatase (Hombi-
kat) was used as catalyst at relevant conditions (not shown in fig-
ure). The evaluation of the catalyst activity of pure Mn oxides for
the SCR of NO with ammonia at the temperature range 393–
513 K is tested in the presence of 2 vol.% of oxygen (Fig. 6). It
should be noted that a relatively low gas hourly space velocity
was used for this set of experiments in order to allow us to get sig-
nificant level of NO conversion of the pure Mn oxides. All these
reactions were performed in a quartz fixed-bed continuous flow
reactor (i.d. 6 mm, 51 cm) at atmospheric pressure. The typical
reactant gas composition was as follows: 400 ppm NO, 400 ppm
NH3, 2% O2, and balance helium. Prior to the each reaction, unsup-
ported manganese oxides of different oxidation state were pre-
pared in situ and evaluated for the SCR of NO with ammonia
between 393 and 513 K. These normalized results demonstrate
the nitric oxide reduction performance of the pure Mn oxides de-
creased in the following order: MnO2 ꢃ Mn2O3 n Mn3O4 ffi MnO.
The nitric oxide reduction results of MnO have not been incorpo-
rated in Fig. 6, since it had been progressively converted into
Mn3O4 during the SCR experiment. For this reason only, the MnO
also showed similar results as the Mn3O4 in the SCR of NO reaction.
All the above results clearly suggest that the oxidation state of the
surface Mn is decisive and the surface Mn4+ species are highly ac-
tive for the SCR of NO reaction with ammonia at low temperatures.
In order to obtain the information about the oxidation state of
manganese and nickel cations on the catalyst surface (which was
invisible under XRD) and to ensure the chemical compositions
(atomic concentrations) of the surface layer, XPS spectra of the
samples were recorded. The MnO2 and Mn2O3 were prepared using
aqueous solutions of manganese nitrates [29], and Mn3O4 was pre-
pared by the slow reduction of MnO2 at 473 K in H2 [30]. The pure
MnO was prepared by the slow reduction of MnO2 at 900 K in H2.
The above prepared samples were used for the reference manga-
nese phase compounds. However, our XPS spectra of the reference
samples of MnO2, Mn2O3, Mn3O4, and MnO exhibited the Mn 2p3/2
peak at around 642.3, 641.3, 641.4, and 641.5 eV, respectively (Ta-
ble 1). Two main peaks due to Mn 2p3/2 and Mn 2p1/2 of the Mn/
TiO2 and nickel-doped Mn/TiO2 samples are shown in Fig. 7. It
has been well established that the Mn4+ peak, Mn3+ peak, and
Mn nitrate peaks appear at 642.1 0.2 eV, 641.3 0.2 eV, and
644.2 0.4 eV, respectively [8,31,22]. For the identification of the
surface manganese oxide phases and the relative percentages of
Mn4+, Mn3+, Mnn+ species, the overlapped Mn 2p peaks were
deconvoluted into several peaks by searching for the optimal com-
bination of Gaussian bands with the correlation coefficients (r2)
above 0.99 (PeakFit, Version 4.0.6, AISN Software Inc.). The decon-
voluted peaks are signed as specific phases of manganese (MnO2,
Mn2O3, Mn nitrate) in each spectrum (Fig. 7). The relative percent-
ages of manganese, nickel, and titania species were calculated by
the area ratio of the corresponding characteristic peaks. The results
are listed in Table 1.
790
728
640
As shown in Fig. 7 and Table 1, the manganese species (Mn4+
,
Mn3+) can be characterized by Mn 2p3/2 and Mn 2p1/2 peaks located
between 639.0 eV and 658 eV. It has been found in our studies and
established in the literature that the 2p3/2 binding energy of Mn4+
(MnO2) is 642.1 1 eV, while the corresponding Mn3+ (Mn2O3) has
a binding energy at 641.3 1 eV [8,31,22,32,33]. In all cases, the as-
prepared catalyst surfaces were Mn enriched, as depicted by the
consistently higher value of Mn/Ti surface relative to Ni/Ti. More
specifically, in Mn–Ni(0.4)/TiO2, Mn4+ species are most enriched
on the TiO2 surface with higher surface MnO2 phase relative to that
of other catalysts.
400
500
600
700
800
900
1000
Temperature (K)
Fig. 5. The reduction profile (H2-TPR) of unsupported manganese oxide.

B. Thirupathi, P.G. Smirniotis / Journal of Catalysis 288 (2012) 74–83
79
Table 1
Binding energy, surface atomic ratio of Mn4+/Mn3+, Ni/Mn, and Mn4+/Ti for the prepared catalysts determined from deconvoluted XPS spectra.
a
a
a
a
Catalyst
B.E. (eV)
M0
(Mn4+/Mn3+
)
(Mn4+/Mnn+
)
Ni/Mn
Mn4+/Ti
Ti 2p3/2
Ti 2p1/2
O 1s
Mn 2p3/2
Mn 2p1/2
Ni 2p3/2
Ni 2p1/2
MnO2
–
–
–
–
–
–
–
–
–
–
–
–
642.1
641.3
641.4
641.5
642.1
641.3
642.1
641.3
642.1
641.3
642.1
641.3
642.1
641.3
653.6
653.0
652.9
653.1
653.6
653.0
653.6
653.0
653.6
653.0
653.6
653.0
653.6
653.0
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
Mn2O3
Mn3O4
MnO
5 Mn/TiO2
458.6
464.3
529.9
02.02
0.58
0.020
Mn–Ni(0.2)/TiO2
Mn–Ni(0.4)/TiO2
Mn–Ni(0.6)/TiO2
Mn–Ni(0.8)/TiO2
458.6
459.1
458.8
459.0
464.1
464.6
464.5
464.6
530.1
530.3
530.0
530.3
2p3/2
855.8
2p3/2
856.3
2p3/2
856.1
2p3/2
856.2
2p1/2
869.5
2p1/2
874.6
2p1/2
873.8
2p1/2
874.1
04.88
22.31
01.82
01.44
1.18
1.33
0.63
0.51
0.17
0.43
0.67
0.80
0.069
0.073
0.053
0.049
a
Relative amounts are according to the metal atomic ratio.
Table 1 shows the atom percentage of catalysts determined by
composition [35]. In addition, the presence of other transition me-
XPS. The relative atomic percentage value of Mn4+/Mn3+ character-
ized by XPS was significantly high for the Mn–Ni(0.4)/TiO2 catalyst
(Mn4+/Mn3+ = 22.31, 96%), whereas Mn2O3 phase is in contest with
MnO2 in other catalysts (Mn4+/Mn3+ = 1.34–12.67) (Table 1, Fig. 7).
The formation of surface MnO2 phase is improved a lot by the co-
doping of nickel into the Mn/TiO2 with 0.4 Ni/Mn atomic ratio.
tal cations could also favor the existence surface manganese (Mn4+
species over the support [36].
)
The highly dispersed manganese and nickel species over the
surface of titania anatase is ascertained by the XRD. Interestingly,
our XRD studies revealed that the promoter oxide (NiO) does not
favor any titania phase transformation. Powder X-ray diffraction
patterns of the various nickel-doped titania-supported manganese
catalysts calcined at 673 K for 2 h are shown in Fig. 9. For pure TiO2
sample, the strong characteristic peaks of titania anatase can be
observed. All the samples with various Ni/Mn atomic ratio showed
broad diffraction lines typically at d = 3.54, 1.90, and 2.40 Å, which
corresponds to anatase phase (JCPDS #71-1169 and 21-1272). For
all the prepared catalysts, X-ray reflections at d = 2.41, 1.63, and
2.11 Å are not found, which are corresponding to MnO2 phase
(JCPDS #04-0779). These results are suggesting that the manga-
nese oxide is in a highly dispersed state or the crystallites formed
are less than 5 nm, and also insertion of manganese ions into the
titania lattice [37]. Our XPS studies on surface atomic ratio of
Mn/Ti would explain that the manganese is intercalated in the
TiO2 (Hombikat) lattice structure at lower loadings due to the high
surface and the smaller particle size of the support. Due to this rea-
son, thought that the Mn is highly dispersed at a certain level
called two-dimensional monolayer coverage. This highly dispersed
manganese or poor crystalline structure of manganese oxide has
been suggested to improve the existence of oxygen vacancies and
perform a best catalytic performance [7]. It was proposed that a
small amount of co-doping metal prevents manganese oxide
reaching a crystalline structure [38,39].
The BET surface area, average pore diameter, and pore volume
of the titania (anatase), Mn/TiO2, Mn–Ni/TiO2 (Ni/Mn = 0.2, 0.4,
0.6, 0.8) catalysts are summarized in Table 2. The specific surface
area of the pure titania catalyst dropped to 161 m2 gꢁ1. It can be
explained that the crystallinity and particle size of the support
material increased after calcination at 400 °C. By increasing the
Ni/Mn atomic ratio from zero to 0.4, we observe a monotonic in-
crease in the surface area and pore volume. However, the BET sur-
face area and pore volume decrease gradually with the increase in
nickel content (Ni/Mn = 0.6, 0.8), which may be caused by the
blocking effect of the support pores by the loading of nickel oxide.
However, the co-doping of nickel cations onto the Mn/TiO2 has
apparently enhanced the specific surface area of the catalyst.
Among all the catalysts, Mn–Ni(0.4)/TiO2 catalyst attained the high
surface area (200 m2/g) and high pore volume (0.42 cm3/g). It is
due to high dispersion of manganese and nickel oxides over the
titania support. In order to investigate the synergistic promoting
Among all the catalysts, high surface Mn4+ concentrations (Mn4+
/
Ti) were observed for the Mn–Ni(0.4)/TiO2 catalyst. This is due to
the higher surface coverage by manganese (IV) oxide and/or its
higher dispersion on the TiO2 support. It is interesting to note that
the SCR performance of all the prepared catalysts (Fig. 1) is accu-
rately correlated with the surface Mn4+ concentrations (Fig. 8).
The MnxOy catalytic activity has been ascribed to the potential of
Mn to form several oxides and to provide oxygen selectively from
its crystalline lattice. Since its labile oxidation state, Mn is acting as
an active component in the redox process.
Conversely, the amount of MnxOy needed to cover the support
surface as a monomolecular layer can be estimated from structural
calculations [34]. The monolayer surface coverage is defined as the
maximum amount of MnxOy in contact with the metal oxide sup-
port. From the M–O bond lengths of the crystalline Mn2O3, mono-
layer surface coverage is estimated to be 0.05 wt.% Mn per m2 of
the support [26]. Very interestingly, our best catalyst Mn–
Ni(0.4)/TiO2 composition is equivalent to near or just below the
monolayer capacity of surface manganese on the TiO2 support. This
is probably due to the involvement of Ni species with MnxOy dis-
persion over the titania anatase support. Our previous XRD studies
clearly indicate that the formation of bulk manganese oxide com-
mences from the 8–9 wt.% of manganese over the titania support
[10]. The high dispersion of Mn4+ species monolayer on the TiO2
support is achieved by a total 7 wt.% of Mn–Ni/TiO2 with Ni/Mn
atomic ratio equal to 0.4. Moreover, the Mn–Ni loadings are nearly
equivalent to the theoretical monolayer capacity. As can be seen
from the Fig. 8, the existence of the number of surface Mn4+ species
over the TiO2 support (Mn4+/TiO2) is directly related to the poten-
tial activity of the particular catalyst. In other words, a high con-
centration of surface Mn4+ species equivalent to the monolayer
capacity on the support surface is highly essential for obtaining
better yields. These studies are in good agreement with our activity
studies of pure manganese oxides. Kapteijn et al. [27] also investi-
gated for the active phase of manganese toward the SCR of NO at
low temperatures (Fig. 6) and found that the MnO2 is the preemi-
nent phase for the SCR reaction. The MnO2 showed an excellent
catalytic performance, this is probably due to the poor crystalline
structure of manganese oxide material and defective structural

80
B. Thirupathi, P.G. Smirniotis / Journal of Catalysis 288 (2012) 74–83
(c)
(f)
Mn 2p3/2
Mn 2p3/2
(a)
(b)
(c)
(a)
Mn 2p1/2
Mn 2p1/2
(c)
(b)
Mn 2p3/2
(e)
(b)
Mn 2p3/2
(a)
(b)
(c)
(a)
(b)
Mn 2p1/2
(c)
Mn 2p1/2
Mn 2p3/2
Mn 2p3/2
(a)
(d)
(a)
(b)
(c)
(a)
(b)
(c)
Mn 2p1/2
Mn 2p1/2
635
640
645
650
655
660
635
640
645
650
655
660
Binding Energy (eV)
Binding Energy (eV)
Fig. 7. Deconvoluted Mn 2p (XPS) spectra of (a) MnOx/TiO2 (b) Mn–Ni(0.2)/TiO2, (c) Mn–Ni(0.4)/TiO2, (d) Mn–Ni(0.6)/TiO2, (e) Mn–Ni(0.8)/TiO2, (f) spent Mn–Ni(0.4)/TiO2
catalysts.
effect of Ni species in Mn–Ni/TiO2 catalyst, we normalized the NO
conversion over the catalysts with different Ni/Mn atomic ratios
using BET surface area (Fig. 10). With the increase in Ni/Mn atomic
ratio equal to 0.4, the normalized NO conversion showed a mono-
tonic enhancement. Further increase in Ni/Mn atomic ratio de-
creased the normalized NO conversion, indicating that the Mn/
TiO2 catalyst is promoted by the optimized amount of Ni species,
and Mn species are the main active components. Though the nor-
malized NH3-SCR activity of Mn–Ni(0.2)/TiO2 catalyst is equivalent
to the Mn–Ni(0.4)/TiO2 at 200 °C, from the viewpoint of reaction
temperature broadening application, we chose Mn–Ni(0.4)/TiO2
as the model catalyst for further study due to its high apparent
NH3-SCR activity at low temperatures. However, the normalized
NO conversion of Mn–Ni(0.2)/TiO2 decreased at 240 °C.
The Mn/TiO2 sample prepared by incipient wetness technique
shows TPR patterns characterized by three narrow peaks at 576
(T1), 665 (T2), and 736 K (T3), whose intensity appears to be inde-
pendent of the manganese loading. These peaks are interpreted
as a stepwise reduction of MnO2 to Mn2O3, Mn2O3 to Mn3O4, and
Mn3O4 to MnO, respectively (Table 2, Fig. 11). In agreement with
thermodynamic predictions, further reduction of MnO to metallic
manganese does not occur in between 273 and 1073 K [40].
H2-TPR results for nickel-promoted Mn/TiO2 catalysts showed the
shift of peak positions to lower temperatures, suggesting that the
reduction potential of manganese oxide species is increased
compared to those of monometallic catalyst. A two-step reduction
profiles (only two peaks) were observed for the Mn–Ni/TiO2
catalysts, whereas three-step reduction profile was observed for

B. Thirupathi, P.G. Smirniotis / Journal of Catalysis 288 (2012) 74–83
81
0.08
0.07
0.06
0.05
0.04
0.03
0.02
25
20
15
10
5
1.0
0.9
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0
200 °C
240 °C
0.0
0.2
0.4
0.6
0.8
Ni/Mn atomic ratio
Fig. 8. Surface atomic concentration of Mn4+ and the surface atomic ratio of Mn4+
/
0.2
0.4
0.8
0.0
0.6
Mn3+ with respect to the Ni/Mn atomic ratio in the catalyst acquired from the XPS
analysis: SCR activity of the particular catalyst is comparable with the surface
Ni/Mn ratio
atomic concentration of Mn4+
.
Fig. 10. Low-temperature NH3-SCR activity of the Mn/TiO2 and Mn–Ni/TiO2
catalysts normalized by BET surface area. Reaction conditions: NO = 400 ppm,
NH3 = 400 ppm, O2 = 2 vol.%, balance He and GHSV = 50,000 hꢁ1
.
Α
Α
Α
Α
Α
(e)
(d)
Mn-Ni(0.8)/TiO2
Mn-Ni(0.6)/TiO2
(c)
(b)
(a)
Mn-Ni(0.4)/TiO2
Mn-Ni(0.2)/TiO2
Mn/TiO2
20
30
40
50
60
70
2θ (°)
400
500
600
700
800
900
1000
Fig. 9. Powder X-ray diffraction patterns of various titania-supported catalysts;
according to Ni/Mn atomic ratio (a) Mn/TiO2 (b) Mn–Ni(0.2)/TiO2 (c) Mn–Ni(0.4)/
TiO2 (d) Mn–Ni(0.6)/TiO2 and (e) Mn–Ni(0.8)/TiO2, respectively.
Temperature (K)
Fig. 11. Disappearance of high-temperature peak and low-temperature shift of
reduction peaks in H2-TPR patterns of titania-supported manganese–nickel oxide
catalysts.
Mn/TiO2. The absence of the high-temperature (736 K) peak repre-
sents that the dominant phase is MnO2 [41]. Thus, the co-doping of
manganese with nickel oxides results in stabilization of the former
in the form of MnO2, which is reduced at lower temperatures, com-
pared to Mn2O3. This raise in reduction potential of manganese and
dominant phase of MnO2 seems to be the reason for the increased
activity of Mn–Ni/TiO2 catalysts. H2-TPR results are in agreement
with the XPS results, where the formation of surface MnO2 phase
is increased by the doping of nickel oxide in Mn/TiO2 catalysts.
To verify the dispersion of active metal species over the support,
the most frequently used and widely accepted method is selective
chemisorption of suitable gases like hydrogen and carbon monox-
ide. Parekh and Weller [42,43] have developed a simple oxygen
chemisorption method for determining the dispersion of equiva-
lent metal area over the Al2O3 support. Consequently, the choice
of temperature for the oxygen chemisorption is also crucial. The
Table 2
BET surface area and pore volume and H2-TPR measurements.
Catalyst
XRD phases
SBET (m2/g)
Average pore
diameter (nm)
Pore volume
(cm3/g)
T (K)
H2 consumption (l )
mol gꢁ1
T-1
T-2
T-3
TiO2 Hombikat (commercial)
TiO2 Hombikatb
A
A
A
A
A
A
A
309
161
161
189
200
176
171
4.5
9.3
9.3
8.5
8.6
8.1
7.7
0.37
0.42
0.37
0.40
0.42
0.35
0.33
–
–
–
–
–
–
736
–
–
–
–
Mn/TiO2
Mn–Ni(0.2)/TiO2
Mn–Ni(0.4)/TiO2
Mn–Ni(0.6)/TiO2
576
500
510
471
496
665
689
684
673
685
282.9
335.6
339.9
352.1
398.6
a
a
a
–
–
a
Mn–Ni(0.8)/TiO2
a
Relative amounts are according to the atomic ratio of Ni/Mn; A-anatase;
Calcined at 400 °C.
b

82
B. Thirupathi, P.G. Smirniotis / Journal of Catalysis 288 (2012) 74–83
Table 3
Oxygen uptake, active metal dispersion, and STYs of Mn/TiO2 and Mn–Ni(0.2–0.8)/TiO2 catalysts.
c
SBET (m2/g)
Dispersionb
NO reaction rate in
l
mol gꢁ1 sꢁ1 (XNO%)d
240 °C
STY (sꢁ1) at 50,000 GHSV
a
ꢁ1
S. no.
Catalyst
O2 uptake (
l
mol g
)
cat
200 °C
200 °C
240 °C
1
2
3
4
5
Mn/TiO2
161
189
200
176
171
215.92
238.88
245.36
219.84
208.06
0.48
0.52
0.53
0.48
0.45
08.77 (73.0)
11.78 (98.0)
12.01 (100)
09.73 (81.0)
09.25 (77.0)
02.21 (20.0)
09.41 (85.0)
10.63 (96.00)
08.31 (75.0)
07.97 (72.0)
0.49
0.61
0.61
0.55
0.55
0.12
0.49
0.55
0.42
0.45
Mn–Ni(0.2)/TiO2
Mn–Ni(0.4)/TiO2
Mn–Ni(0.6)/TiO2
Mn–Ni(0.8)/TiO2
a
b
c
Pretreatment and O2 pulse chemisorption temperature are 250 °C.
Fraction of manganese atoms at the surface.
Site time yield (STY): number of NO molecules converted per manganese atom site per second.
XNO% = NO conversion at corresponding temperature.
d
pre-reduction temperature must avoid bulk and over reduction of
the metal species and sintering of the support material [44]. As
shown in our TPR patterns of all catalysts, the first reduction tem-
perature peak appeared around 250 °C related to the reduction of
surface interacted support and active bridge (Ti–O–Mn) compo-
nent, which is very important for the activity of catalyst. For this
reason, the pre-reduction temperature of catalysts was selected
as 250 °C in favor of the oxygen pulse chemisorption experiments
in order to start with the surface reduction of active metal species.
In our previous studies, very low SCR activity (5% NO conversion)
was observed for the 20 wt.% Ni/TiO2 with 50,000 hꢁ1 GHSV at
175 °C suggesting that the nickel is not an active metal for the
low-temperature reaction [10]. Hence, in these oxygen chemisorp-
tion studies, we have concentrated on the dispersion of active me-
tal species (manganese) over the surface of titania support but not
concerning nickel dispersion. In addition, as shown in our TPR pat-
terns, NiO is not supposed to reduce at this pre-reducing tempera-
ture. This (single site activity) method is of considerable intrinsic
attention as an example of complex reactions operating on single
sites (manganese sites). Recently, Zhdanov proposed a kinetic
model of standard selective catalytic reduction in NO by NH3 on
single sites [45].
Mn-Ni(0.4)/TiO2
324
755
619
743
355
506
Mn/TiO2
100
200
300
400
500
600
700
800
Temperature (°C)
Fig. 12. NH3-TPD patterns of Mn/TiO2 and Mn–Ni(0.4)/TiO2 catalysts.
The oxygen uptake, active metal dispersion, NO reaction rate,
and site time yield (STY) values obtained on various Mn–Ni/TiO2
(Ni/Mn = 0.0–0.8) samples are presented in Table 3. To get insight
into the specific activity of the classified Mn species and their rel-
ative fraction is combined with the STY values demonstrated in Ta-
ble 3. As expected, the oxygen chemisorption results indicate that
the uptake of O2 per unit weight of catalyst increases with Ni/Mn
ratio in the catalyst because of the increased active metal disper-
sion. The addition of Ni is promoting the dispersion of manganese
over the surface of titania support. The oxygen chemisorption and
metal dispersion values are increased monotonically up to 0.4 Ni/
Mn ratio Mn–Ni/TiO2 catalyst. It becomes virtually high at a load-
ing of Ni/Mn ratio equals to 0.4 in the Mn–Ni/TiO2 catalyst. Further
increase in Ni content decreases the oxygen sorption capacity in
the catalyst (Mn–Ni(0.6–0.8)/TiO2). This is probably due to the cov-
erage of surface manganese species by approaching high amount of
nickel species. This observation is also in agreement with our XPS
results, where the ratio of surface active species (Mn4+/Ti) de-
creased behind the 0.4 Ni/Mn ratio.
1228
1156
1098
Mn-Ni(0.4)/TiO2
523 K
473 K
1599
1540
423 K
373 K
323 K
2000
1800
1600
1400
-1
1200
Wavenumber (cm )
The acidic sites distribution of the Mn/TiO2 and Mn–Ni(0.4)/TiO2
catalysts was determined using ammonia TPD technique. The sim-
ilar TPD patterns were observed for the corresponding catalysts
(Fig. 12). It is established in the literature that the acid site distribu-
tion can broaden by the addition of nickel oxide [46,47]. In the pres-
ent study, broad acid site distribution is observed for Mn–Ni(0.4)/
TiO2 catalyst. It is found in our TPD studies that the doping of nickel
can remarkably improve the concentration and acidity of Lewis acid
sites on Mn/TiO2 catalyst. Usually, the low-temperature peak is
attributed to ammonia coordinated to Lewis acid sites; percentage
of this peak area is increased by the addition of nickel with 0.4 Ni/
Fig. 13. In situ FT-IR spectra of adsorbed ammonia species over the Mn–Ni(0.4)/TiO2
catalyst at 323 K and successive outgassing at various temperatures (323–523 K).
Mn atomic ratio. These results are in good agreement with the re-
ported literature, where the doping of nickel can significantly im-
prove the concentration and acidity of Lewis acid sites on CeO2–
ZrO2 catalysts [48]. To support this statement, the adsorption and
desorption characteristics of NH3 over the Mn–Ni(0.4)/TiO2 catalyst
have been investigated to ensure the influence of nickel oxide. In
situ FTIR experiments were performed in order to gain a better

B. Thirupathi, P.G. Smirniotis / Journal of Catalysis 288 (2012) 74–83
83
[4] H. Tounsi, S. Djemal, C. Petitto, G. Delahay, Appl. Catal. B: Environ. 107 (2011)
158.
[5] J.P. Dunn, P.R. Koppula, H.G. Stenger, I.E. Wachs, Appl. Catal. B: Environ. 19
(1998) 103.
[6] Vanadium Pentoxide, MSDS No. V2220, B. Mallinckrodt, N.J. Phillipsburg, July
1, 2009, <http://www.jtbaker.com/msds/englishhtml/v2220.htm> (accessed
20.01.10).
[7] P.G. Smirniotis, D.A. Peña, B.S. Uphade, Angew. Chem. Int. Ed. 40 (2001) 2479.
[8] D.A. Peña, B.S. Uphade, P.G. Smirniotis, J. Catal. 221 (2004) 421.
[9] D.A. Peña, B.S. Uphade, E.P. Reddy, P.G. Smirniotis, J. Phys. Chem. B 108 (2004)
9927.
[10] E.P. Reddy, E. Neeraja, M. Sergey, P. Boolchand, P.G. Smirniotis, Appl. Catal. B:
Environ. 76 (2007) 123.
[11] T. Ishihara, M. Ando, K. Sada, K. Takiishi, K. Yamada, H. Nishiguchi, Y. Takita, J.
Catal. 220 (2003) 104.
[12] A.K.H. Nohman, D. Duprez, C. Kappenstein, S.A.A. Mansour, M.I. Zaki, in: B.
Delmon, P.A. Jacobs, G. Poncelet (Eds.), Preparation of Catalysts V, Elsevier,
Amsterdam, 1991, p. 617.
[13] R. Craciun, B. Nentwick, K. Hadjiivanov, H. Knözinger, Appl. Catal. A 243 (2003)
67.
[14] C. Lahousse, A. Bernier, P. Grange, B. Delmon, P. Papaefthimiou, T. Ioannides, X.
Verykiosy, J. Catal. 178 (1998) 214.
[15] C. Reed, Y. Xi, S.T. Oyama, J. Catal. 235 (2005) 378.
[16] Brian R. Strohmeier, David M. Hercules, J. Phys. Chem. 88 (1984) 4922.
[17] P.W. Selwood, T.E. Moore, M. Ellis, K. Wethingt, J. Am. Chem. Soc. 71 (1949)
693.
[18] A.K.H. Nohman, D. Duprez, C. Kappenstein, S.A.A. Mansour, M.I. Zaki,
Preparation of catalysts, in: G. Poncelet, P.A. Jacobs, P. Grange, B. Delmon
(Eds.), Elsevier, Amsterdam, 1991, p. 617.
[19] M.L. Jacono, M. Schiavello, Preparation of catalysts, in: B. Delmon, P.A. Jacobs,
G. Poncelet (Eds.), Elsevier, Amsterdam, 1976, p. 474.
understanding of molecular behavior of ammonia and to acquire
the information about surface species. The adsorption–desorption
of ammonia provided information about the nature of acid sites
present, and its presence as a function of temperature (323–
523 K) revealed the potential strength of the acid sites. In our pre-
vious studies, the 1167, 1301, 1467, 1599, and 1680 cmꢁ1 peaks
are observed for Mn/TiO2 [9] and they correspond to ammonia
coordinatively adsorbed on Lewis sites. The in situ FTIR spectra of
the adsorbed species arising from the interaction of ammonia with
the surface of Mn–Ni(0.4)/TiO2 catalyst followed by evacuation at
successively increasing temperatures are shown in Fig. 13. The
bands at 1098, 1156, and 1228 cmꢁ1 peaks were attributed to
coordinatively adsorbed NH3 on Lewis sites [48]. Interestingly, an
absorption peak at 1156 cmꢁ1 (ds NH3 coordinated to Lewis acid
sites) decreases down to disappear after evacuation at 473–523 K
range. Bands at 1540 and 1599 cmꢁ1 are attributed to the das NH3
coordinated to Lewis acid sites [48]. In agreement with the TPD re-
sults, we observed similar distinctive in situ FTIR spectra for the
Mn–Ni(0.4)/TiO2 and Mn/TiO2. The similar features of coordinative
adsorbed species (1098, 1156, and 1228 cmꢁ1, ds NH3 bound to Le-
wis acid sites) were observed in both cases. These results suggest
that the Mn–Ni(0.4)/TiO2 surface sites are having Lewis acidity,
with no apparent Brönsted acidity.
[20] A. Wollner, F. Lange, H. Schmetz, H. Knozinger, Appl. Catal. A 94 (1993) 181.
[21] V.H. Vu, J. Belkouch, A. Ould-Dris, B. Taouk, AIChE J. 54 (2008) 6, doi:10.1002/
aic.
4. Conclusions
[22] Z. Chen, Q. Yang, H. Li, X. Li, L. Wang, S.C. Tsang, J. Catal. 276 (2010) 56.
[23] W.-J. Hong, S. Iwamoto, S. Hosokawa, K. Wada, H. Kanai, M. Inoue, J. Catal. 277
(2011) 208.
[24] W.S. Kijlstra, J.C.M.L. Daamen, J.M. van de Graaf, B. van der Linden, E.K. Poels, A.
Bliek, Appl. Catal. B: Environ. 7 (1996) 337.
[25] T. Valdes´-Solıs´, G. Marban´ , A.B. Fuertes, Appl. Catal. B: Environ. 46 (2003) 261.
[26] X.T. Tang, J.M. Hao, W.G. Xu, J.H. Li, Catal. Commun. 8 (2007) 329.
[27] F. Kapteijn, L. Singoredjo, A. Andreini, Appl. Catal. B 3 (1994) 173.
[28] A.H. Reidies, Manganese Compounds, Ullmann’s Encyclopedia of Chemical
Technology, Wiley-VCH, Weinheim, 2007, doi:10.1002/14356007.a16_123,
<http://en.wikipedia.org/wiki/Manganese(II)_oxide#cite_note-Greenwood-1>.
[29] F.C. Buciuman, F. Patcas, T. Hahn, Chem. Eng. Proc. 38 (1999) 563.
[30] F. Kapteijn, L. Singoredjo, M. Vandriel, A. Andreini, J.A. Moulijn, G. Ramis, G.
Busca, J. Catal. 150 (1994) 105.
A series of novel nickel-promoted titania-supported manganese
catalysts were prepared by adopting wet impregnation method
and developed for the low-temperature selective catalytic reduc-
tion (SCR) of NO with ammonia in the presence of oxygen. 100%
of NO conversion with 100% N2 selectivity was obtained over the
Mn–Ni(0.4)/TiO2 catalyst, at 200 °C, under flow conditions of
GHSV = 50,000 hꢁ1. It was found from the XRD patterns that the
Mn and Ni are highly dispersed over the titania support. Shift in
manganese reduction peak (576–510 K) and disappearance of
high-temperature peak at 736 K in the TPR profiles suggested that
the MnO2 is the dominant phase in Mn–Ni/TiO2 catalysts. Among
all the catalysts, deconvoluted XPS spectrum of Mn–Ni(0.4)/TiO2
revealed the high existing amount of Mn4+ ions over the surface
of titania (Mn4+/Ti = 0.073). The Mn–Ni(0.4)/TiO2 catalyst proved
its remarkable performance by exhibiting a superior hydrothermal
stability with 10 vol.% inlet water concentrations with 240 h on
stream. In conclusion, Mn–Ni(0.4)/TiO2 catalyst offers high NO
conversion, high N2 selectivity, broadening of temperature win-
dow, time on stream patterns, and hydrothermal stability.
[31] F. Kapteijn, A.D. van Langeveld, J.A. Moulijn, A. Andreini, M.A. Vuurman, A.M.
Turek, J.-M. Jehng, I.E. Wachs, J. Catal. 150 (1994) 94.
[32] A. Machocki, T. Ioannides, B. Stasinska, W. Gac, G. Avgouropoulos, D.
Deli2007maris, W. Grzegorczyk, S. Pasieczna, J. Catal. 227 (2004) 282.
[33] S. Ponce, M.A. Pena, J.L.G. Fierro, Appl. Catal. B: Environ. 24 (2000) 193.
[34] J.M.G. Amores, T. Armaroli, G. Ramis, F. Finaochio, G. Busca, Appl. Catal. B:
Environ. 22 (1999) 259.
[35] M.R. Morales, B.P. Barbero, L.E. Cadús, Appl. Catal. B: Environ. 67 (2006) 229.
[36] M.R. Morales, B.P. Barbero, L.E. Cadús, Appl. Catal. B: Environ. 74 (2007) 1.
[37] G. Pechi, P. Reyes, T. Lopez, R. Gomez, A. Moreno, J.L.G. Fierro, J. Chem. Technol.
Biotechnol. 77 (2002) 944.
[38] M.R. Morales, B.P. Barbero, L.E. Cadús, Fuel 87 (2008) 1177.
[39] M.R. Morales, B.P. Barbero, T. Lopez, A. Moreno, L.E. Cadús, Fuel 88 (2009)
2122.
[40] I.R. Leith, M.G. Howden, Appl. Catal. 37 (1988) 75.
[41] S.M. Saqer, D.I. Kondarides, X.E. Verykios, Appl. Catal. B: Environ. 103 (2011)
275.
Acknowledgment
The authors wish to acknowledge financial support from the
National Science Foundation (Grant No. NSF-0828226).
[42] B.S. Parekh, S.W. Weller, J. Catal. 47 (1977) 100.
[43] S.W. Weller, Acc. Chem. Res. 16 (1983) 101.
[44] B.M. Reddy, B. Manohar, E.P. Reddy, Langmuir 9 (1993) 1781.
[45] V.P. Zhdanov, React. Kinet. Mech. Cat. 103 (2011) 11.
[46] Z. Si, D. Weng, X. Wu, J. Yang, B. Wang, Catal. Commun. 11 (2010) 1045.
[47] P.M. Sreekanth, P.G. Smirniotis, Catal. Lett. 122 (2008) 37.
[48] J.M.G. Amores, V.S. Escribano, G. Ramis, G. Busca, Appl. Catal. B: Environ. 13
(1997) 45.
References
[1] F. Janssen, F.V. Kerkhof, H. Bosch, J.R.H. Ross, J. Phys. Chem. 91 (1987) 5921.
[2] J.A. Dumesic, N.-Y. Topsoe, H. Topsoe, T. Slabiak, J. Catal. 163 (1996) 409.
[3] G. Busca, L. Lietti, G. Ramis, F. Berti, Appl. Catal. B: Environ. 18 (1998) 1.