Effects of iron precursor and loading on the catalytic performance of FeOx/CeO2 catalysts for NO reduction by CO

Article abstract of DOI:10.1016/j.mcat.2020.111123

During the catalyst synthesis process, the precursor chosen could have great impact on the physiochemical and surface properties of the catalysts. More importantly, the precursor could result in very different catalytic activities of the catalyst. In this work, two series of FeO_x/CeO₂ catalysts were prepared using iron nitrate/sulfate precursors with a wide range of Fe loading (1 wtpercent – 30 wtpercent). The catalytic activity, physical property, surface morphology and surface/molecular structure were investigated using GC, BET, XRD, Raman, FT-IR as well as SEM and EDS techniques. From the results, although the physical properties (specific surface area, total pore volume) of the catalyst were similar with different precursors, the surface composition and molecular structures were very different. The iron nitrate precursor fully transformed into FeO_x after the calcination process, whereas the iron sulfate precursor only partially transformed into FeO_x, leaving some FeSO₄ on the catalyst surface. The catalytic activity results suggested that a near monolayer coverage of FeO_x on the catalyst surface was crucial to increase the catalytic activity in the NO reduction by CO reaction, while the presence of FeSO₄ had advert impacts on the catalytic activity.

Full text of DOI:10.1016/j.mcat.2020.111123

MolecularCatalysis494(2020)111123
Contents lists available at ScienceDirect
Molecular Catalysis
journal homepage: www.elsevier.com/locate/mcat
Effects of iron precursor and loading on the catalytic performance of FeOx/
CeO2 catalysts for NO reduction by CO
Shuhao Zhang, Taejin Kim*
Materials Science and Chemical Engineering Department, Stony Brook University, Stony Brook, NY, 11794, USA
A R T I C L E I N F O
A B S T R A C T
Keywords:
NO reduction by CO
Iron oxide
Ceria
Precursor effects
During the catalyst synthesis process, the precursor chosen could have great impact on the physiochemical and
surface properties of the catalysts. More importantly, the precursor could result in very different catalytic ac-
tivities of the catalyst. In this work, two series of FeOx/CeO2 catalysts were prepared using iron nitrate/sulfate
precursors with a wide range of Fe loading (1 wt% – 30 wt%). The catalytic activity, physical property, surface
morphology and surface/molecular structure were investigated using GC, BET, XRD, Raman, FT-IR as well as
SEM and EDS techniques. From the results, although the physical properties (specific surface area, total pore
volume) of the catalyst were similar with different precursors, the surface composition and molecular structures
were very different. The iron nitrate precursor fully transformed into FeOx after the calcination process, whereas
the iron sulfate precursor only partially transformed into FeOx, leaving some FeSO4 on the catalyst surface. The
catalytic activity results suggested that a near monolayer coverage of FeOx on the catalyst surface was crucial to
increase the catalytic activity in the NO reduction by CO reaction, while the presence of FeSO4 had advert
impacts on the catalytic activity.
1. Introduction
activity comparing to CoOx/CeO2 catalysts, the dispersion of the surface
FeOx was better than CoOx. Zhang et al. [18] investigated CuO sup-
Removal of nitrogen oxides (NOx) produced from combustion pro-
cesses continues to be a major thrust area in heterogeneous catalysis. In
the past few decades, supported metal and metal oxide catalysts have
been utilized to catalytically reduce NOx gases with the help of reducing
agents such as hydrogen [1–4], and carbon monoxide (CO) [5–9]. The
catalytic reduction of NO by CO is a particularly attractive reaction
since NO and CO are both present in the automotive exhaust and could
be eliminated at the same time with this reaction. In general, supported
metal oxide catalysts relied on precious noble metals such as Pt and Rh
[10,11]. Due to the high cost and scarcity of precious metals, catalyst
systems containing earth-abundant metals and their oxides have gar-
nered increasing attention. Transition metal oxide catalysts are ad-
vantageous because of lower costs, higher thermal stability, and com-
petitive catalytic activities, especially for the NO reduction by CO
reaction [12–14].
ported by CeO2-Fe2O3 catalysts for NO reduction by CO and found that
the addition of Fe3+ to CeO2 had positive impacts on NO conversion, as
there was a strong interaction between Fe2O3 and CeO2.
Precursors used during the catalyst synthesis process are found to
have an impact on the crystallinity, surface area, morphology, and
catalytic activity. For example, Juárez-Marmolejo et al. [19] compared
Fe and Co precursors (nitrate/chloride) and their effects on the struc-
ture of carbon supported PdFe and PdCo catalysts. The authors found
that nitrate precursors facilitated the metal oxide formation more than
the chloride precursors, which led to a better catalytic performance for
the formic acid oxidation reaction. Xiong et al. [20] compared the
catalytic activities of carbon spheres supported iron catalysts in Fischer-
Tropsch synthesis using nitrate or acetate iron precursors. The authors
reported that the catalyst made from iron acetate precursor had lower
surface area than that of nitrate precursor (9.2 m2/g vs 21.4 m2/g) and
more uniform iron particles. It was also reported that iron acetate
precursor derived supported iron catalyst showed better performance
(in terms of metal time yield) than that of iron nitrate derived one.
Wang et al. [21] found that, instead of using iron nitrate precursor, the
use of ammonium ferric citrate in PtFe/C catalysts for oxygen reduction
reaction could facilitate a larger amount of PtFe alloy, which resulted in
Ceria – iron oxide catalyst systems have been studied in several
catalysis procedures including NO reduction by H2 and NH3 selective
catalytic reduction (NH3-SCR) and have shown great potential [15,16].
Recently, Peck et al. [17] studied FeOx/CeO2 and CoOx/CeO2 catalysts
for NO reduction by CO reaction in the presence of O2. Although the
authors found that FeOx/CeO2 catalysts generally showed lower areal
⁎ Corresponding author at: Materials Science and Chemical Engineering Department, Stony Brook University, Stony Brook, NY, 11794, USA.
E-mail address: taejin.kim@stonybrook.edu (T. Kim).
https://doi.org/10.1016/j.mcat.2020.111123
Received 6 May 2020; Received in revised form 10 July 2020; Accepted 13 July 2020
2468-8231/©2020ElsevierB.V.Allrightsreserved.

S. Zhang and T. Kim
MolecularCatalysis494(2020)111123
a higher activity (23 % increase). In other cases, different precursor did
not affect the catalytic activity much. Cherian et al. [22] found that
using an ammonium dichromate precursor for the synthesis of Cr2O3/
Al2O3 catalysts resulted in a lower surface area, compared to the cat-
alysts using the chromium nitrate precursors. However, the catalysts
prepared from both precursors showed similar catalytic activities in
propane ODH reaction.
Although several and series of CeO2 supported MOx (M = metal)
catalysts have been studied for the NO reduction by CO reaction, de-
tailed studies of precursor effect on the catalytic activity and physico-
chemical properties are still lacking. Herein we report the effect of ni-
trate and sulfate iron oxide precursors, as well as the Fe loading on the
overall catalytic activity of NO reduction by CO. Two series of FeOx/
CeO2 catalysts were prepared by incipient wetness impregnation (IWI)
on CeO2 with different precursors. The effects of iron precursors on the
structures and catalytic performances of FeOx/CeO2 catalysts were in-
vestigated by BET, XRD, Raman spectroscopy, FT-IR, SEM and EDS, as
well as gas phase NO reduction by CO reaction.
∼0.1 g of the catalyst samples were held at 300 °C for 4 h to remove
water and impurities under vacuum condition. During the analysis, the
catalyst samples were placed in liquid nitrogen (−196 °C).
Crystallite structures of the calcined CeO2 and FeOx/CeO2 catalysts
were explored by XRD technique. The XRD patterns were collected with
a
Phillips PW1729/APD3520 diffractometer with Cu radiation
(λ = 1.54184 Å) in the 2θ range of 20−80° with a step size of 0.02°.
Molecular structures of the FeOx/CeO2 catalysts were probed using
Raman spectroscopy (XploRa PLUS, Horiba). The Raman spectra were
collected under ambient temperature and pressure conditions with
785 nm laser. The Raman shift was calibrated with standard silicon
wafer sample. The spectral acquisition times were 10 scans accumu-
lated at 20 s/scan.
Fourier transform infrared (FTIR) spectra were obtained with
Perkin-Elmer Frontier IR spectrometer using an attenuated total re-
flectance (ATR) mode. The IR spectrum was collected with a 4 cm−1
resolution and an accumulation of 32 scans.
The surface topography and content were probed by SEM and EDS
techniques, respectively. The SEM images were taken with LEO 1550
SFEG SEM (Zeiss) with a EDS detector from EDAX, and software/elec-
tronics from iXRFsystems. The samples were coated with nominally
6 nm of Au to reduce charging. Additionally, the samples were imaged
using a Robinson backscatter detector (RBSD) and 20KeV electrons.
2. Experimental section
2.1. Catalyst synthesis
Two series of ceria supported iron oxide catalysts were synthesized
using incipient wetness impregnation method. Two precursors, iron
(III) nitrate and iron (II) sulfate, were impregnated on the same sup-
porting material (CeO2, HSA5 from Rhodia) during the synthesis pro-
cess. The precursor was first dissolved in de-ionized water completely,
then the prepared solution was added to the CeO2 powder slowly with
the mixture constantly being stirred. After impregnation, the samples
were dried under ambient conditions for 12 h. The samples were
transported to a tube furnace (Thermo Scientific Lindberg Blue Mini-
Mite Tube Furnace, Model TF55030A-1) and further dried in flowing air
(100 mL/min, dry grade, Airgas) at 120 °C for 12 h. The ramping rate
from room temperature to 120 °C was 2 °C/min. After fully removing
moisture from the catalyst, the catalyst was subsequently calcined in
flowing air (100 mL/min, dry grade, Airgas) at 400 °C for 6 h. The
ramping rate from 120 °C to 400 °C was 5 °C/min. Finally, the calcined
catalyst was sieved using a 425 μm size sieve (Fisherbrand). The pre-
pared catalysts were detonated as x% FeOx/CeO2 (N/S), with x being
the calculated wt% of Fe, and N/S indicating the precursor used during
the synthesis process (N-nitrate, S-sulfate).
3. Results
3.1. Gas-phase catalytic activity of FeOx/CeO2 catalysts for NO reduction
by CO
The gas phase catalytic activity results of the two series of FeOx/
CeO2 (N and S) are shown in Figs. 1 and 2, respectively. Overall, as the
reaction temperature increased from 25 °C to 500 °C, for both FeOx/
CeO2 (N) and FeOx/CeO2 (S) series of catalysts, the NO and CO con-
versions increased as well. However, with the same surface Fe wt% and
reaction temperatures, the FeOx/CeO2 (N) catalysts showed higher NO
or CO conversion values than their FeOx/CeO2 (S) counterparts, espe-
cially at medium reaction temperature ranges (200 °C – 400 °C). For
example, at 250 °C, the 5% FeOx/CeO2 (N) catalyst converted > 90 %
of NO whereas the 5% FeOx/CeO2 (S) catalyst displayed negligible NO
conversion. At 500 °C, on the other hand, all of the tested catalysts,
regardless of precursors, showed high NO and CO conversions
(both > 90 %), except for bulk FeOx (S), which displayed no catalytic
activity throughout the tested reaction temperature range
(25 °C – 500 °C). Also, for the FeOx/CeO2 (N) catalysts, the supported
catalysts performed better than both bulk FeOx (N) and bulk CeO2,
especially under lower reaction temperatures. But for FeOx/CeO2 (S)
catalysts, 1% and 5% FeOx/CeO2 (S) catalysts showed higher NO and
CO conversions than bulk CeO2. This result indicates that the surface
structure/composition of the two series of FeOx/CeO2 catalysts might
be different, resulting in different synergy effects as well as vastly dif-
ferent catalytic activity results under the same reaction conditions.
Moreover, comparing the catalysts made with iron nitrate precursor, it
could be noticed that, as the Fe loading increased from 1% to 5%, the
overall catalytic activity also improved. For instance, at 200 °C and
250 °C, the NO conversion increased from ∼30 % to ∼55 % and from
∼80 % to ∼95 %, respectively. However, as the surface Fe content
continue to increase from 5% to 30 %, the NO and CO conversion did
not follow the same trend but stayed the same (or worse) instead. This
could possibly be explained by the effects of surface Fe dispersion. It has
been reported that after the surface metal oxides (e.g., CoOx and FeOx)
reach monolayer coverage, the catalytic activity does not further in-
crease even with higher surface metal oxide loadings, as discussed by
Peck et al. and our previous work [17,23]. In the case of FeOx/CeO2 (S)
series of catalysts, NO and CO conversion decreased continuously with
increasing surface Fe loadings up to 10 wt%, while both 10 wt% and
30 wt% samples showed similar results. This could be caused by the
2.2. Gas phase catalytic activity measurements
The gas phase catalytic activity of the synthesized FeOx/CeO2 cat-
alysts in NO reduction by CO reaction was investigated using a gas
phase reaction setup. The catalyst (∼40 mg) was put in a fixed bed
quartz reactor (7.6 mm inner diameter, 2 mm wall thickness, 24.4 cm
length, Quartz Scientific). The reaction gas was made up by 5% NO
with He balance and 5% CO with He balance with a total flow rate of
40 mL/min. The flow rates were controlled manually with needle valves
and monitored by mass flow meters (FMA-1700 series, Omega) and the
gas hourly space velocity (GHSV) was 31,200 h−1. Reaction tempera-
ture range was 25 °C–500 °C and the temperature was controlled by
temperature control systems (ATS). Catalysts were treated at 400 °C in
He atmosphere for 1 h before the activity test. The gas phase reaction
products were detected and analyzed with GC equipped with a TCD
detector (Trace 1300 GC from Thermo Scientific).
2.3. Catalyst characterization
The N2 isotherm, multi-point BET surface area, BJH pore size dis-
tribution as well as pore properties (average pore size and total pore
volume) of the FeOx/CeO2 catalysts were measured and calculated by
NOVA 2200 series analyzer from Quantachrome. Before the test,
2

S. Zhang and T. Kim
MolecularCatalysis494(2020)111123
Fig. 1. Gas phase catalytic activity results of
FeOx/CeO2 (N) catalysts in NO reduction by CO
reaction under different reaction temperatures.
(a) NO conversion, (b) CO conversion, (c) N2
selectivity, and (d) N2O selectivity. Reaction
conditions: Total flow rate =40 mL/min, 5%
NO = 20 mL/min,
Catalyst amount =40 mg, Temperature
ranges = 25°C ∼ 500°C, GHSV =31,200 h−1
5%
CO = 20 mL/min,
.
formation and accumulation of inactive surface species on the catalyst.
Detailed surface characterization using Raman, XRD and FT-IR will be
discussed in later sections.
In addition to NO and CO conversion, the N2 and N2O selectivity
under each reaction temperature were also evaluated. As shown in
Fig. 1 (c–d) and Fig. 2(c–d), as the reaction temperature increased, the
N2 (N2O) selectivity increased (decreased) monotonically in sync with
the reaction temperatures. This could be explained with the well-ac-
cepted two-step reaction process for the NO reduction by CO reaction
over supported transition metal oxide catalysts [24–26]. In the first step
Fig. 2. Gas phase catalytic activity results of
FeOx/CeO2 (S) catalysts in NO reduction by CO
reaction under different reaction temperatures.
(a) NO conversion, (b) CO conversion, (c) N2
selectivity, and (d) N2O selectivity. Reaction
conditons: Total flow rate =40 mL/min, 5%
NO = 20 mL/min,
Catalyst amount =40 mg, Temperature
ranges = 25°C ∼ 500°C, GHSV =31,200 h−1
5%
CO = 20 mL/min,
.
3

S. Zhang and T. Kim
MolecularCatalysis494(2020)111123
of the reaction, especially when the reaction temperature is lower, NO
and CO are converted into mainly N2O and CO2. (Eq. 1) As reaction
temperature increased, the formed N2O started interacting with CO and
decompose into N2 and convert CO into more CO2 at the same time.
(Eq. 2)
patterns, indicating the presence of mesoporous structures. In the case
of FeOx/CeO2 (N and S) catalysts, the N2 isotherm patterns were very
similar to that of bulk CeO2, while 30 % FeOx/CeO2 sample showed a
little different type IV pattern. The 30 % FeOx/CeO2 samples displayed
H4 hysteresis loop patterns rather than the H1 hysteresis loop patterns,
which were observed in 1∼10 % FeOx/CeO2 samples. This result in-
dicated that for FeOx/CeO2 samples with lower Fe contents, the pore
size of the catalyst should be narrowly distributed [27]. (Figure S1) In
the case of 30 % FeOx/CeO2 sample, the H4 hysteresis loop is associated
with slit-like pores and a less well-defined pore size distribution profile.
The bulk FeOx (N) and bulk FeOx (S) both showed type IV isotherm as
well, with a H4 hysteresis loop similar to 30 % FeOx/CeO2 catalysts,
suggesting that the bulk FeOx (N/S) also had narrow slit-like pore
structures as well as a broad pore size distribution as shown in Figure
S1.
(1) (Step 1) 2NO + CO → N2O + CO2
(2) (Step 2) N2O + CO → N2 + CO2
(4) (Overall reaction) 2NO + 2CO → N2 + 2CO2
It can be noticed that during the first step of the reaction, the NO:CO
ratio is 2:1 instead of the 1:1 ratio if considering the overall reaction
(Eq. 3). This resulted in the different NO and CO conversion shown in
Fig. 1(a) and (b) under lower reaction temperatures. For example, the
5% FeOx/CeO2 (N) sample showed ∼55 % NO conversion at 200 °C,
and the corresponding CO conversion is ∼25 %. Whereas at higher
reaction temperatures (i.e. 350 °C), both NO and CO conversions are
∼100 %. Similar phenomenon could be observed in the FeOx/CeO2 (S)
cases as well. As shown in Fig. 2(a) and (b), at 250 °C, the 1% FeOx/
CeO2 (S) sample showed ∼55 % NO conversion and ∼28 % CO con-
version. At 350 °C, the corresponding NO and CO conversions are both
∼80 %.
3.3. Powder X-ray diffraction (XRD)
One of the major factors that could affect the catalytic activity is
molecular structure of the catalyst. The crystalline structures of the
FeOx/CeO2 catalysts in addition to bulk CeO2 and Fe2O3 as references
were investigated using powder X-ray diffraction (XRD) technique
(Fig. 5). Note that the XRD pattern of the aluminum sample holder was
also included in the figure, as some of peaks observed in the FeOx/CeO2
catalysts were from the sample holder, rather than the catalysts. For
both FeOx/CeO2 (N) and FeOx/CeO2 (S) catalysts, all the tested samples
displayed strong fluorite CeO2 patterns (FCC structure, PDF# 97−002-
8709) at 28.6°, 33.0°, 47.3°, and 56.2°, which are related to the (111),
(200), (220), and (311) reflections in fluorite CeO2, respectively. Fur-
thermore, the positions of CeO2 characteristic diffraction peaks on the
FeOx/CeO2 XRD patterns were not shifted (Table S1). This result sug-
gested that the fluorite cubic structure of CeO2 in the FeOx/CeO2 cat-
alysts was not changed. For the 30 % FeOx/CeO2 (N) sample, new peaks
were also observed at 33.1°, 35.5°, 54.1°, 62.5°, and 63.9°, which are
attributed to the (104), (110), (116), (214) and (300) reflections from
the α-Fe2O3 structure, respectively (PDF# 00−033-0664) [28–30]. As
for the 33.1° peak, it is difficult to distinguish solely from the XRD
patterns since both α-Fe2O3 and CeO2 have peak at ∼33.0°. It is
worthwhile to note that most FeOx characteristic peaks were not ob-
served up to 10 % FeOx/CeO2 (N) samples, possibly because the surface
FeOx species were relatively well dispersed on CeO2 rather than
forming crystalline α-Fe2O3 structures. The XRD pattern of FeOx/CeO2
(S) catalysts showed similar to that of the bulk CeO2, while 30 % FeOx/
CeO2 (S) sample’s diffraction peaks were weak and broad. (Fig. 5 (b))
Weak intensity peak at 35.5° was observed with 30 % FeOx/CeO2 (S)
sample, but the shape of the peak was not as sharp comparing to 30 %
FeOx/CeO2 (N). For bulk FeOx (S) sample, sharp and high intensity peak
was observed ∼27.5°, which corresponded to the FeSO4 structures, in
addition to the typical α-Fe2O3 peaks [31]. Note that the FeSO4 peak
3.2. Physical properties of FeOx/CeO2
To better understand the structure-activity relationship of FeOx/
CeO2 catalysts in NO reduction by CO reaction and explain the vastly
different trend displayed in the catalytic activity results using the two
series of FeOx/CeO2 catalysts, physical properties of the FeOx/CeO2
catalysts were investigated using N2 isotherm, BET and BJH methods.
The specific surface area and the total pore volume of the FeOx/CeO2
catalysts decreased with increasing Fe loading, regardless of the type of
precursors. (Fig. 3 and Table S1) The decrease in specific surface area
and total pore volume could be due to the accumulation and agglom-
eration of surface FeOx species, especially with high surface FeOx
content (> 10 wt%), as well as the blocking of the pores on the catalyst
surface. It was reported that the surface density of monolayer coverage
of surface FeOx in FeOx/CeO2 (N) catalyst system is around 7.5 Fe/nm2
[17]. In the current work, the surface density values of 10 wt% and
30 wt% FeOx/CeO2 (N) catalysts were 6.7 Fe/nm2 and 32.3 Fe/nm2,
and the surface density values of 10 wt% and 30 wt% FeOx/CeO2 (S)
catalysts were 7.7 Fe/nm2 and 53 Fe/nm2, as shown in Table S1.
Therefore, the steep decrease in specific surface area from 10 % FeOx/
CeO2 to 30 % FeOx/CeO2 catalyst is likely due to the formation of
crystalline structures of FeOx (or bulk like Fe2O3), which has very low
specific surface area of ∼25 m2/g in both nitrate and sulfate precursor
cases.
Fig. 4 shows the N2 isotherm curve of bulk CeO2, FeOx/CeO2 (N and
S), and bulk FeOx. The bulk CeO2 displayed typical type-IV N2 isotherm
Fig. 3. (a) Specific surface area of the FeOx/CeO2 catalysts, and (b) total pore volum of the FeOx/CeO2 catalysts.
4

S. Zhang and T. Kim
MolecularCatalysis494(2020)111123
Fig. 4. N2 adsorption/desorption isotherms of the (a) FeOx/CeO2 (N) catalysts, and (b) FeOx/CeO2 (S) catalysts (close symbols – adsorption, open symbols –
desorption).
was not observed on a series of FeOx/CeO2 (S) samples’ XRD patterns.
From the XRD results, it could be concluded that crystalline α-Fe2O3
structures were possibly formed in very high Fe loading (30 wt%) FeOx/
CeO2 catalysts over the studied samples. However, we cannot fully ig-
nore the presence of crystalline structure of FeOx with < 30 wt% sam-
ples due to the limitation in detection range of lab-scale powder XRD.
FeOx/CeO2 (N) catalyst. The results indicate that as the surface Fe
content kept increasing from 10 % to 30 %, the size of the surface Fe2O3
crystalline structure became larger with increasing ordered state.
Fig. 6(b) shows the bulk FeOx (S) and a series of FeOx/CeO2 (S) cata-
lysts’ Raman spectra. Similar to the FeOx/CeO2 (N) samples’ Raman
spectra, the FeOx/CeO2 (S) spectra contained the CeO2 (F2g mode) peak
at ∼458 cm−1 and peak intensity was decreased with increasing Fe
loadings. The 30 % FeOx/CeO2 (S) catalyst showed Fe2O3 peaks at
∼222 cm−1, ∼290 cm−1, ∼405 cm−1, and ∼612 cm−1, in addition to
the very weak intensity CeO2 peak, while Fe2O3 peaks were not present
up to 10 % FeOx/CeO2 (S) spectra. It was noticed that the Fe2O3 peaks
positions in 30 % FeOx/CeO2 (S) catalyst shifted to higher wavenumber
comparing to 30 % FeOx/CeO2 (N) catalyst. The peak shifting could be
due to the formation of multiple surface Fe species (e.g. Fe2O3 and
FeSO4) or other effects like photon confinement [38,39]. The 30 %
3.4. Raman and FT-IR spectroscopy
To further investigate the detailed surface structure and dispersion
of the FeOx, Raman spectroscopy was applied since it is more sensitive
to nanocrystalline metal oxide structures on the catalyst surface. The
Raman spectra of the series of FeOx/CeO2 (N) and FeOx/CeO2 (S) cat-
alysts were shown in Fig. 6(a) and (b), respectively. The bulk CeO2
sample displayed peaks at ∼250 cm−1 (weak), ∼458 cm−1 (strong)
and ∼597 cm−1 (weak), which are attributed to the second-order
transverse acoustic mode, F2g vibration of Ce-O bond, and D-band vi-
brations in CeO2, respectively [32–34]. It is worth pointing out that the
∼597 cm−1 peak from CeO2 was also believed to related to the defect
sites in CeO2 lattice [35]. For the FeOx/CeO2 (N) catalysts, starting from
10 % FeOx/CeO2 (N) catalyst, two new peaks at ∼218 cm−1 and
∼282 cm−1 were observed. (Fig. 6(a)) The ∼218 cm−1 peak is due to
the A1g vibration in Fe2O3 structures whereas the ∼282 cm−1 peak is
assigned to the Eg vibration in Fe2O3, indicating that the surface Fe2O3
nanocrystalline structures started to form when the surface Fe content
reached ∼10 wt%. [36,37]. In the case of 30 % FeOx/CeO2 (N) catalyst,
two more peaks were observed in the spectrum at ∼400 cm−1 and
∼600 cm−1, which are assigned to the Eg vibration in Fe2O3 structures
[36]. Note that the ∼218 cm−1 and ∼282 cm-1 peaks in the 30 %
FeOx/CeO2 (N) catalyst spectrum is much stronger than these of 10 %
FeOx/CeO2 (S) Raman spectrum also showed
a weak peak at
∼646 cm−1 and this could be due to the presence of FeSO4 [40]. In the
case of bulk FeOx (S) sample, two sharp peaks were observed at
∼183 cm−1 and ∼230 cm−1, which are also assigned to FeSO4 [40].
Based on the Raman results, it could be speculated that, for the FeOx/
CeO2 (N) catalysts, the surface Fe2O3 started forming nanocrystalline
structures at 10 wt% and monolayer coverage is < 10 wt%. On the
other hand, the FeOx/CeO2 (S) catalysts contain both Fe2O3 and FeSO4
structure at 30 wt% and monolayer coverage is > 10 wt%.
In addition to Raman spectroscopy, the IR spectra of the samples
were also collected to further investigate the molecular structures of the
catalysts. The FT-IR ATR results are shown in Figure S2. For the FeOx/
CeO2 (N) catalysts, no obvious difference in spectra was observed up to
10 % FeOx/CeO2 (N), while the 30 % FeOx/CeO2 (N) catalyst showed
signature bands of Fe2O3 at ∼450 cm−1 and ∼550 cm−1
,
Fig. 5. XRD patterns of the (a) FeOx/CeO2 (N) catalysts, and (b) FeOx/CeO2 (S) catalysts. XRD patterns of bulk CeO2, bulk FeOx (N and S), and alumium sample
holder are included as references.
5

S. Zhang and T. Kim
MolecularCatalysis494(2020)111123
Fig. 6. Raman spectra of the (a) FeOx/CeO2 (N) catalysts, and (b) FeOx/CeO2 (S) catalysts. Raman spectra of bulk CeO2 and bulk FeOx (N and S) are included as
references.
corresponding to the Fe-O stretching vibrations [41,42]. This result
agrees with the previous XRD and Raman results, where the evidence of
Fe2O3 crystalline structures existed in the 30 % FeOx/CeO2 (N) catalyst.
It should be noted that Raman spectroscopy provided the presence of
Fe2O3 at 10 % FeOx/CeO2 (N) catalyst as shown in Fig. 6(a). For the
well and sulfur was identified in all the tested FeOx/CeO2 (S) catalysts.
For the FeOx/CeO2 (N) catalysts, the SEM images and EDS results were
shown in Figure S3. Ce, Fe, and Au elements were observed.
4. Discussion
FeOx/CeO2 (S) catalysts, Fe2O3 bands (∼450 cm−1 and ∼550 cm−1
)
and FeSO4 band (∼1120 cm−1) [43] were found in the 30 % FeOx/
CeO2 (S) catalyst and this result is well matched to the Raman spec-
troscopy result. Using the FTIR spectroscopy, the evidence of FeSO4 in a
series of FeOx/CeO2 (S) samples, however, was observed starting from
1% FeOx/CeO2 (S) catalyst. Considering the catalytic activity results
over the FeOx/CeO2 (S) catalysts, it was observed that as the Fe loading
increased, the catalytic activity decreased. Based on the FTIR results,
the decrease in catalytic activity might be due to the presence of FeSO4
because sulfur is well known as a catalyst poison and could have sig-
nificantly negative impacts on catalytic activity [44,45].
The precursors used during the catalyst synthesis process can have
great impacts on the surface/molecular structures of the catalyst, and
further resulting in improved or worsened catalytic activities. In this
study, multiple spectroscopic and microscopic techniques were applied
to understand the iron precursor effects on the structure-activity re-
lationship of the FeOx/CeO2 catalysts. In the FeOx/CeO2 (N) series of
catalysts, as the Fe loading increased from 1% to 30 %, a monotonic
decrease in specific surface area of the samples were observed (Fig. 3
and Table S1). The total pore volume of the catalysts also slightly de-
creased with the increasing Fe loading, possibly due to the blocking of
the existing pores on the CeO2 surface. Similar trend was observed for
the FeOx/CeO2 (S) catalysts. However, the specific surface area and
total pore volume decrease were steeper than their nitrate precursor
counterpart at higher Fe loading samples (> 10 %). For example, the 30
% FeOx/CeO2 (N) catalyst had specific surface area of 100 m2/g
whereas the 30 % FeOx/CeO2 (S) catalyst only had specific area of
61 m2/g. The larger drop in specific surface area could be from the
presence of both Fe2O3 and FeSO4. With increased Fe loading on the
catalyst surface, the dispersion and molecular structure of the surface
FeOx were expected to change from isolated species to monolayer
coverage of FeOx, then to the formation of large crystalline FeOx
structures. This speculation was then confirmed by spectroscopic re-
sults. XRD patterns of the FeOx/CeO2 (N) catalysts (Fig. 5) indicated
that Fe2O3 crystalline structures were formed in the 30 % FeOx/CeO2
(N) catalyst. Raman results (Fig. 6), having better resolution on smaller
structures, confirmed that the Fe2O3 nanocrystalline structures started
to emerge in the 10 % FeOx/CeO2 (N) catalyst. Based on previous lit-
erature [17,23], Raman spectroscopy could be used as a tool for surface
metal oxide monolayer coverage detection. Therefore, it could be
speculated that the 10 % FeOx/CeO2 (N) is above (or around) the
monolayer coverage region. For the FeOx/CeO2 (S) catalysts, the 30 %
FeOx/CeO2 (S) showed a peak of Fe2O3 in XRD results as well as the two
signature peaks of Fe2O3 in its Raman and IR spectra. Based on the XRD
and Raman spectroscopic results, it can be deduced that the monolayer
coverage range is > 10 % for FeOx/CeO2 (S) catalyst. This result pro-
vides that dispersibility of FeOx on FeOx/CeO2 (S) could be higher than
that of FeOx/CeO2 (N). It has been reported that the better dispersion of
surface metal oxide can improve the catalytic performances [46–48].
However, the nitrate and sulfate precursor effect on the molecular
structure and catalytic activity in this study were not fully followed the
general high-dispersion/high-activity relationship due to the sulfur
3.5. Scanning electron microscopy (SEM) and Energy Dispersive X-Ray
spectroscopy (EDS)
To further confirm the surface morphology and the presence of
sulfur on the surface of the FeOx/CeO2 (S) catalysts, SEM images were
taken, and EDS analyses were applied. As shown in Fig. 7, with in-
creasing the Fe loading from 1% to 30 %, surface texture was slowly
changed from smooth to rough. Especially, the 30 % FeOx/CeO2 (S)
catalyst’s SEM image shows agglomerated FeOx (and Fe3O4) species as
well as bumpy surface. On the other hand. the FeOx/CeO2 (S) catalysts
with lower Fe loading show relatively clean surface and surface FeOx
species looks well dispersed on the CeO2 surface. In the case of the
FeOx/CeO2 (N) catalysts, similar to the FeOx/CeO2 (S) catalysts, an
increased surface roughness could be observed in the SEM images as the
Fe loading increased (Figure S3). In the image of 30 % FeOx/CeO2 (N)
catalyst, larger crystalline structures could be noticed, while 10 %
FeOx/CeO2 (N) catalyst shows small but noticeable surface species. The
surface species could be mixed FeOx and Fe2O3 species based on the
XRD and Raman results. From the EDS results (below the SEM images),
Ce, Fe, and S elements could be identified. It should be noted that the
Au element was from the sample preparation process for SEM imaging,
where gold coating was applied to the sample surface to improve the
quality of the SEM images. Also, from Figure S4 and Figure S5, it can be
noticed that the EDS results of the FeOx/CeO2 catalysts are similar even
with different magnification powers, indicating an even distribution of
the surface FeOx species. From the previous Raman and IR results,
FeSO4 was found in the FeOx/CeO2 (S) catalysts. For instance, in IR
results (Figure S2), the FeSO4 bands were noticed starting from 1%
FeOx/CeO2 (S). According to the EDS results, similar trend was ob-
served. The existence of sulfur was confirmed in 1% FeOx/CeO2 (S) as
6

S. Zhang and T. Kim
MolecularCatalysis494(2020)111123
Fig. 7. SEM images and the corresponding EDS results of the (a) 1% FeOx/CeO2 (S) catalyst, (b) 5% FeOx/CeO2 (S) catalyst, (c) 10 % FeOx/CeO2 (S) catalyst, and (d)
30 % FeOx/CeO2 (S) catalyst.
residues. As shown in XRD (Fig. 5(b)), Raman (Fig. 6(b)), FTIR (Figure
S2(b)), and EDS (Fig. 7), FeSO4 was present in a series of FeOx/CeO2 (S)
catalysts. The presence of FeSO4 could be explained by the high thermal
decomposition temperature (> 600 °C) [49], while the calcination
temperature used in this study was 400 °C. It is worthwhile to note that
Fe2O3 species was also identified in the 30 % FeOx/CeO2 (S) catalyst
from XRD, Raman and IR results. This result could be speculated that
CeO2 facilitated the decomposition of FeSO4 into Fe2O3 during the
calcination process, even at 400 °C.
leveled off at > 5%, while 30 % FeOx/CeO2 (N) show little lower cat-
alytic performance between 250 °C and 300 °C. The stop in activity
increase at ≥10 % FeOx/CeO2 (N) catalysts could be due to the ex-
cessive amount of surface FeOx, which has weaker interaction with the
surface CeO2 and could not improve the catalytic activity. In the case of
the series of FeOx/CeO2 (S) catalysts, catalytic activity and N2 se-
lectivity displayed the totally different trend compared to the FeOx/
CeO2 (N) catalysts. Fig. 8 showed a more direct comparison between
the T50 for NO conversion and temperature at 50 % N2 selectivity for
the FeOx/CeO2 (N and S) catalysts. As shown in Fig. 8, there was a big
increase in NO T50 between 1%, 5% and 10 % FeOx/CeO2 (S) catalysts,
possibly due to the increased FeSO4 amount with increasing Fe loading.
Considering the gas phase catalytic activity results (Fig. 1), in the
case of FeOx/CeO2 (N) catalysts, both NO and CO conversion and N2
selectivity were improved with increasing Fe loading up to 5% and then
7

S. Zhang and T. Kim
MolecularCatalysis494(2020)111123
Fig. 8. (a)The T50 temperature for NO conversion and (b) Temperature at 50 % N2 selectivity over the FeOx/CeO2 (N and S) catalysts in NO reduction by CO reaction.
From the comparison between the catalytic activity of FeOx/CeO2 (N)
and FeOx/CeO2 (S), it is reasonable to conclude that the presence of
surface FeSO4 could have adverse effects on the catalytic activity in NO
reduction by CO reaction, while physical property of both catalysts
were relatively similar.
Acknowledgments
S. Zhang and T. Kim would like to thank the Advanced Energy
Research and Technology Center (AERTC) for the facilities and the fi-
nancial support from Sensor CAT (Award number: 83242) at the Stony
Brook University. The authors would also like to thank Prof. Rina
Tannenbaum for providing Raman spectrometer, Prof. Joel Hurowitz
and Donald Hendrix for providing BET analyzer, as well as Dr. Jim
Quinn for the assistance in XRD, SEM and EDS experiments.
5. Conclusions
In the current work, two series of CeO2 supported FeOx catalysts
(from 1 wt% to 30 wt%) with two different iron precursors (iron nitrate
and iron sulfate) were synthesized, characterized and tested for the NO
reduction by CO reaction. The two series of FeOx/CeO2 catalysts
showed similar trend in physical property changes – as the Fe loading
increased, the catalysts’ specific surface area as well as total pore vo-
lume both decreased accordingly. Combining the XRD, Raman, FT-IR,
and EDS results, it could be concluded that, for the FeOx/CeO2 (N)
catalysts, the surface Fe species only consisted of FeOx and Fe2O3, and
the monolayer coverage for Fe2O3 is < 10 wt%. For the FeOx/CeO2 (S)
catalysts, both Fe2O3 and FeSO4 were identified in the surface Fe spe-
cies and the monolayer coverage could be ≥ 10 wt%. The precursor
effects on the catalytic activity were significant. Firstly, with the same
Fe loading, the FeOx/CeO2 (N) catalysts performed better in terms of
NO and CO conversion, as well as N2 selectivity; Secondly, as the Fe
loading increase, the FeOx/CeO2 (N) catalysts showed an increase in
catalytic activity up to ∼5%, then the catalytic activity leveled off even
with higher Fe loading. On the other hand, the series of FeOx/CeO2 (S)
catalysts showed different trend where the catalytic activity decreased
as the Fe loading increased, which was speculated to be caused by the
catalyst poisoning by FeSO4. Overall, this work could serve as a guide to
choose a proper precursor to achieve higher catalytic activity.
Appendix A. Supplementary data
Supplementary material related to this article can be found, in the
online version, at doi:https://doi.org/10.1016/j.mcat.2020.111123.
References
[1] Q. Du, X. Cheng, M.H. Tahir, D. Su, Z. Wang, S. Chen, Investigation on NO reduction
by CO and H2 over metal oxide catalysts Cu2M9CeOx, Int. J. Hydrogen Energy 45
(2020) 16469–16481, https://doi.org/10.1016/j.ijhydene.2020.04.088.
[2] M. Machida, S. Ikeda, D. Kurogi, T. Kijima, Low temperature catalytic NOx - H2
reactions over Pt/TiO2-ZrO2 in an excess oxygen, Appl. Catal. B Environ. 35 (2001)
107–116, https://doi.org/10.1016/S0926-3373(01)00243-0.
[3] G. Qi, R.T. Yang, L.T. Thompson, Catalytic reduction of nitric oxide with hydrogen
and carbon monoxide in the presence of excess oxygen by Pd supported on pillared
clays, Appl. Catal. A Gen. 259 (2004) 261–267, https://doi.org/10.1016/j.apcata.
2003.09.040.
[4] R. Burch, M.D. Coleman, An investigation of promoter effects in the reduction of NO
by H2 under lean-burn conditions, J. Catal. 208 (2002) 435–447, https://doi.org/
10.1006/jcat.2002.3596.
[5] Y. Bai, X. Bian, W. Wu, Catalytic properties of CuO/CeO2-Al2O3 catalysts for low
concentration NO reduction with CO, Appl. Surf. Sci. 463 (2019) 435–444, https://
doi.org/10.1016/j.apsusc.2018.08.229.
[6] Y. Wang, A. Zhu, Y. Zhang, C.T. Au, X. Yang, C. Shi, Catalytic reduction of NO by
CO over NiO/CeO2catalyst in stoichiometric NO/CO and NO/CO/O2 reaction, Appl.
Catal. B Environ. 81 (2008) 141–149, https://doi.org/10.1016/j.apcatb.2007.12.
005.
Author contributions
[7] M. Haneda, H. Hamada, Promotional role of H2O in the selective catalytic reduction
of NO with CO over Ir/WO3/SiO2 catalyst, J. Catal. 273 (2010) 39–49, https://doi.
org/10.1016/j.jcat.2010.04.021.
[8] L. Ilieva, G. Pantaleo, N. Velinov, T. Tabakova, P. Petrova, I. Ivanov, G. Avdeev,
D. Paneva, A.M. Venezia, NO reduction by CO over gold catalysts supported on Fe-
loaded ceria, Appl. Catal. B Environ. 174–175 (2015) 176–184, https://doi.org/10.
1016/j.apcatb.2015.03.004.
The manuscript was written through contributions of all authors. All
authors have given approval to the final version of the manuscript.
CRediT authorship contribution statement
[9] H. ou Zhu, J.R. Kim, S.K. Ihm, Characteristics of Pt/WO3/CeO2/ZrO2 catalysts for
catalytic reduction of NO by CO, Appl. Catal. B Environ. 86 (2009) 87–92, https://
doi.org/10.1016/j.apcatb.2008.07.024.
[10] M. Li, X. Wu, Y. Cao, S. Liu, D. Weng, R. Ran, NO reduction by CO over Rh/Al2O3
and Rh/AlPO4 catalysts: metal-support interaction and thermal aging, J. Colloid
Interface Sci. 408 (2013) 157–163, https://doi.org/10.1016/j.jcis.2013.07.023.
[11] P. Xiao, R.C. Davis, X. Ouyang, J. Li, A. Thomas, S.L. Scott, J. Zhu, Mechanism of
NO reduction by CO over Pt/SBA-15, Catal. Commun. 50 (2014) 69–72, https://doi.
org/10.1016/j.catcom.2014.02.027.
Shuhao Zhang: Data curation, Formal analysis, Investigation,
Writing - original draft, Writing - review & editing. Taejin Kim:
Supervision, Writing - original draft, Writing - review & editing, Project
administration, Funding acquisition.
Declaration of Competing Interest
[12] P. Broqvist, I. Panas, H. Persson, A DFT study on CO oxidation over Co3O4, J. Catal.
210 (2002) 198–206, https://doi.org/10.1006/jcat.2002.3678.
[13] Y. Hu, L. Dong, M. Shen, D. Liu, J. Wang, W. Ding, Y. Chen, Influence of supports on
the activities of copper oxide species in the low-temperature NO + CO reaction,
Appl. Catal. B Environ. 31 (2001) 61–69, https://doi.org/10.1016/S0926-3373(00)
The authors declare that they have no known competing financial
interests or personal relationships that could have appeared to influ-
ence the work reported in this paper.
8

S. Zhang and T. Kim
MolecularCatalysis494(2020)111123
00269-1.
pp. 35–37.
[14] L. Ma, M.F. Luo, S.Y. Chen, Redox behavior and catalytic properties of CuO/
Ce0.8Zr0.2O2 catalysts, Appl. Catal. A Gen. 242 (2003) 151–159, https://doi.org/
10.1016/S0926-860X(02)00509-4.
[32] S. Agarwal, X. Zhu, E.J.M. Hensen, L. Lefferts, B.L. Mojet, Defect chemistry of ceria
nanorods, J. Phys. Chem. C. 118 (2014) 4131–4142, https://doi.org/10.1021/
jp409989y.
[15] D. Prieto-Centurion, T.R. Eaton, C.A. Roberts, P.T. Fanson, J.M. Notestein, Catalytic
reduction of NO with H2 over redox-cycling Fe on CeO2, Appl. Catal. B Environ.
168–169 (2015) 68–76, https://doi.org/10.1016/j.apcatb.2014.11.055.
[16] Y. Ma, X. Tang, F. Gao, H. Yi, S. Zhao, Y. Shi, C. Wang, Selective catalytic reduction
of NOx with NH3 over iron-cerium mixed oxide catalyst prepared by different
methods, J. Chem, Technol. Biotechnol. 95 (2020) 232–245, https://doi.org/10.
1002/jctb.6226.
[17] T.C. Peck, G.K. Reddy, M. Jones, C.A. Roberts, Monolayer detection of supported Fe
and Co oxides on Ceria to establish structure-activity relationships for reduction of
NO by CO, J. Phys. Chem. C. 121 (2017) 8435–8443, https://doi.org/10.1021/acs.
jpcc.7b00398.
[18] X. Zhang, X. Cheng, C. Ma, Z. Wang, Effects of the Fe/Ce ratio on the activity of
CuO/CeO2-Fe2O3 catalysts for NO reduction by CO, Catal. Sci. Technol. 8 (2018)
3336–3345, https://doi.org/10.1039/c8cy00709h.
[19] L. Juárez-Marmolejo, S. Pérez-Rodríguez, M.G.M. de Oca-Yemha, M. Palomar-
Pardavé, M. Romero-Romo, A. Ezeta-Mejía, P. Morales-Gil, M.V. Martínez-Huerta,
M.J. Lázaro, Carbon supported PdM (M= Fe, Co) electrocatalysts for formic acid
oxidation. Influence of the Fe and Co precursors, Int. J. Hydrogen Energy. 44 (2019)
1640–1649, https://doi.org/10.1016/j.ijhydene.2018.11.112.
[20] H. Xiong, M. Moyo, M.A.M. Motchelaho, L.L. Jewell, N.J. Coville, Fischer-Tropsch
synthesis over model iron catalysts supported on carbon spheres: the effect of iron
precursor, support pretreatment, catalyst preparation method and promoters, Appl.
Catal. A Gen. 388 (2010) 168–178, https://doi.org/10.1016/j.apcata.2010.08.039.
[21] Q. Wang, H. Tao, Z. Li, G. Wang, Effect of iron precursor on the activity and stability
of PtFe/C catalyst for oxygen reduction reaction, J. Alloys. Compd. 814 (2020)
152212, https://doi.org/10.1016/j.jallcom.2019.152212.
[22] M. Cherian, M.S. Rao, W.T. Yang, J.M. Jehng, A.M. Hirt, G. Deo, Oxidative dehy-
drogenation of propane over Cr2O3/Al2O3 and Cr2O3 catalysts: effects of loading,
precursor and surface area, Appl. Catal. A Gen. 233 (2002) 21–33, https://doi.org/
10.1016/S0926-860X(02)00132-1.
[23] S. Zhang, Y. Li, J. Huang, J. Lee, D.H. Kim, A.I. Frenkel, T. Kim, Effects of molecular
and electronic structures in CoOx/CeO2 catalysts on NO reduction by CO, J. Phys.
Chem. C. 123 (2019) 7166–7177, https://doi.org/10.1021/acs.jpcc.8b12442.
[24] C. Deng, J. Qian, C. Yu, Y. Yi, P. Zhang, W. Li, L. Dong, B. Li, M. Fan, Influences of
doping and thermal stability on the catalytic performance of CuO/Ce20M1Ox (M =
Zr, Cr, Mn, Fe, Co, Sn) catalysts for NO reduction by CO, RSC Adv. 6 (2016)
113630–113647, https://doi.org/10.1039/c6ra21740k.
[25] L. Xue, C. Zhang, H. He, Y. Teraoka, Catalytic decomposition of N2O over CeO2
promoted Co3O4 spinel catalyst, Appl. Catal. B Environ. 75 (2007) 167–174,
https://doi.org/10.1016/j.apcatb.2007.04.013.
[26] L.F. Liotta, G. Pantaleo, G. Di Carlo, G. Marcì, G. Deganello, Structural and mor-
phological investigation of a cobalt catalyst supported on alumina-baria: effects of
redox treatments on the activity in the NO reduction by CO, Appl. Catal. B Environ.
52 (2004) 1–10, https://doi.org/10.1016/j.apcatb.2004.03.003.
[27] K.S.W. Sing, D.H. Everett, R.A.W. Haul, L. Moscou, R.A. Pierotti, J. Rouquerol,
T. Siemieniewska, Reporting physisorption data for Gas/Solid systems with special
reference to the determination of surface area and porosity, Pure Appl. Chem. 57
(1985) 603–619, https://doi.org/10.1351/pac198557040603.
[28] S. Xu, A.H. Habib, S.H. Gee, Y.K. Hong, M.E. McHenry, Spin orientation, structure,
morphology, and magnetic properties of hematite nanoparticles, J. Appl. Phys. 117
(2015) 17A315, , https://doi.org/10.1063/1.4914059.
[29] M. Zainuri, Hematite from natural iron stones as microwave absorbing material on
X-band frequency ranges, IOP Conf. Ser. Mater. Sci. Eng. IOP Publishing, 2017, p.
12008, , https://doi.org/10.1088/1757-899X/196/1/012008.
[30] N. Li, Q. Zhang, R. Bibi, Q. Shen, R. Ngulube, Y. Liu, J. Zhou, Catalytic hydro-
genation of acetic acid to acetaldehyde: synergistic effect of bifunctional Co/Ce‐Fe
oxide solid solution catalysts, Chinese J. Chem. 37 (2019) 709–719, https://doi.
org/10.1002/cjoc.201900055.
[31] P.C. Silva, N.P. Ferraz, E.A. Perpetuo, Y.J.O. Asencios, Treatment of OIL produced
water using advanced oxidative processes: heterogeneous-photocatalysis and photo-
fenton, Environ. Div. 2018 - Core Program. Area 2018 AIChE Annu. Meet 4 (2018),
[33] T. Yu, Y. Zhu, X. Xu, Z. Shen, P. Chen, C.T. Lim, J.T.L. Thong, C.H. Sow, Controlled
growth and field-emission properties of cobalt oxide nanowalls, Adv. Mater. 17
(2005) 1595–1599, https://doi.org/10.1002/adma.200500322.
[34] Z. Wu, M. Li, J. Howe, H.M. Meyer, S.H. Overbury, Probing defect sites on CeO2
nanocrystals with well-defined surface planes by raman spectroscopy and O2 ad-
sorption, Langmuir 26 (2010) 16595–16606, https://doi.org/10.1021/la101723w.
[35] B.M. Reddy, A. Khan, Y. Yamada, T. Kobayashi, S. Loridant, J.C. Volta, Structural
characterization of CeO2-TiO2 and V2O5/CeO2-TiO2 catalysts by Raman and XPS
techniques, J. Phys. Chem. B 107 (2003) 5162–5167, https://doi.org/10.1021/
jp0344601.
[36] A. Zoppi, C. Lofrumento, E.M. Castellucci, P. Sciau, Al-for-Fe substitution in he-
matite: the effect of low Al concentrations in the Raman spectrum of Fe2O3, J.
Raman Spectrosc. 39 (2008) 40–46, https://doi.org/10.1002/jrs.1811.
[37] H. Mansour, H. Letifi, R. Bargougui, S. De Almeida-Didry, B. Negulescu, C. Autret-
Lambert, A. Gadri, S. Ammar, Structural, optical, magnetic and electrical properties
of hematite (α-Fe2O3) nanoparticles synthesized by two methods: polyol and pre-
cipitation, Appl. Phys. A Mater. Sci. Process. 123 (2017) 1–10, https://doi.org/10.
1007/s00339-017-1408-1.
[38] D. Bersani, P.P. Lottici, Confinement effects on the LO‐phonons in CdSexS1–x doped
glasses, Phys. Status Solidi 174 (1992) 575–582, https://doi.org/10.1002/pssb.
2221740227.
[39] C. Baratto, P.P. Lottici, D. Bersani, G. Antonioli, G. Gnappi, A. Montenero, Sol-gel
preparation of α-Fe2O3 thin films: structural characterization by XAFS and raman,
J. Solgel Sci. Technol. 13 (1998) 667–671, https://doi.org/10.1023/
A:1008694519106.
[40] H.C. Chi, S.K. Sharma, D.W. Muenow, The hydrates and deuterates of ferrous sulfate
(FeSO4): A Raman spectroscopic study, J. Raman Spectrosc. 38 (2007) 87–99,
https://doi.org/10.1002/jrs.1623.
[41] S. Yariv, E. Mendelovici, The effect of degree of crystallinity on the infrared
Spectrum of hematite, Appl. Spectrosc. 33 (1979) 410–411, https://doi.org/10.
1366/0003702794925651.
[42] S. Zolghadr, S. Kimiagar, A.M. Davarpanah, Magnetic property of α -Fe2O3-GO
nanocomposite, IEEE Trans. Magn. 53 (2017), https://doi.org/10.1109/TMAG.
2017.2733503.
[43] J. Majzlan, C.N. Alpers, C.B. Koch, R.B. McCleskey, S.C.B. Myneni, J.M. Neil,
Vibrational, X-ray absorption, and Mössbauer spectra of sulfate minerals from the
weathered massive sulfide deposit at Iron Mountain, California, Chem. Geol. 284
(2011) 296–305, https://doi.org/10.1016/j.chemgeo.2011.03.008.
[44] M. Casapu, O. Kröcher, M. Elsener, Screening of doped MnOx-CeO2 catalysts for
low-temperature NO-SCR, Appl. Catal. B Environ. 88 (2009) 413–419, https://doi.
org/10.1016/j.apcatb.2008.10.014.
[45] C. Xie, Y. Chen, Y. Li, X. Wang, C. Song, Sulfur poisoning of CeO2-Al2O3-supported
mono- and bi-metallic Ni and Rh catalysts in steam reforming of liquid hydro-
carbons at low and high temperatures, Appl. Catal. A Gen. 390 (2010) 210–218,
https://doi.org/10.1016/j.apcata.2010.10.012.
[46] L. Zhang, X. Yao, Y. Lu, C. Sun, C. Tang, F. Gao, L. Dong, Effect of precursors on the
structure and activity of CuO-CoOx/γ-Al2O3 catalysts for NO reduction by CO, J.
Colloid Interface Sci. 509 (2018) 334–345, https://doi.org/10.1016/j.jcis.2017.09.
031.
[47] L. Zhang, L. Dong, W. Yu, L. Liu, Y. Deng, B. Liu, H. Wan, F. Gao, K. Sun, L. Dong,
Effect of cobalt precursors on the dispersion, reduction, and CO oxidation of CoOx/
γ-Al2O3 catalysts calcined in N2, J. Colloid Interface Sci. 355 (2011) 464–471,
https://doi.org/10.1016/j.jcis.2010.11.076.
[48] J. Li, J. Chen, R. Ke, C. Luo, J. Hao, Effects of precursors on the surface Mn species
and the activities for NO reduction over MnOx/TiO2 catalysts, Catal. Commun. 8
(2007) 1896–1900, https://doi.org/10.1016/j.catcom.2007.03.007.
[49] M.A. Shuvaeva, G.S. Litvak, V.A. Varnek, G.A. Bukhtiyarova, Preparation of sup-
ported iron-containing catalysts from a FeSO4 solution: the effect of the support,
React. Kinet. Catal. Lett. 50 (2009) 874–877, https://doi.org/10.1134/
S0023158409060123.
9