Y. Zhao et al. / Journal of Alloys and Compounds 829 (2020) 154508
3
3. Results and discussion
species and CrO3 species, respectively. The peak with the highest
binding energy is attributed to the Cr6þ species and the other peaks
correspond to the Cr3þ species [37]. The 3d spectrum of Ce is fitted
into eight peaks in several different regions (3d5/2: v series, 3d3/2:
u series) via Gaussian fitting, in which the peaks labeled v, v’’, v’’’
and u, u’’, u’’’ are assigned to the Ce4þ state and the v’ and u’ peaks
are attributed to the Ce3þ state [38]. Two main peaks can be
separated after fitting for the O 1s spectra: one peak is denoted Oa,
corresponding to the lattice oxygen (O2ꢁ) at a lower binding energy,
and the other is denoted Ob, corresponding to the chemisorbed
oxygen, such as Oꢁ2 or Oꢁ at a higher binding energy [39]. From the
Cr 2p spectrum, the peak intensities and ratios of Cr6þ to Cr3þ in-
crease with an increase in the Cr doping. Because Cr possesses a
higher electronegativity than Ce, a shift in the Cr3þ peak appears in
10CCE-U31 with the replacement of Ce by Cr. The Cr 2p and O 1s
spectra show that 10CCE-U31 exhibits the highest Cr6þ and Ob
content, as shown in Table 2. In the Ce-containing catalysis system,
the variation trends in the Ce3þ and Ob contents have the same
characteristics. In conclusion, based on the strong oxidizability of
Cr6þ and Ob, the doping of Cr into CeO2 clearly enhanced the
oxidizing ability of the catalysts. The influence of the Cr doping into
CeO2 on the electronic structure was further studied by DFT theo-
retical calculations, as shown in Fig. S3. The replacements of the Ce
atom in the lattice cell by one Cr atom and two Cr atoms were
calculated, which represented 5% and 10% Cr doping in moles,
respectively. The results show that the electron transport capacity
was enhanced after doping with one Cr atom. The same phenom-
enon was further confirmed after doping with two Cr atoms. The
calculation results were similar in the XPS analysis.
The influence of different urea dosages on the physicochemical
structure of the catalysts was also investigated, as shown in Fig. S2B
and Fig. S4. The catalysts with different urea dosages display similar
XRD patterns; they all shift to higher angles than that of pure CeO2.
Additionally, an obvious impurity phase appears over the catalysts
with a greater addition of urea. These series catalysts also exhibit a
similar N2 adsorption-desorption isotherm type, and 10CCE-U31
possesses the highest surface area and pore volume. The SEM im-
ages confirm the XRD results, displaying a similar morphology with
a double-shelled hollow structure under different urea dosages.
The surface element valence states over three typical catalysts
10CCE-Uy1 (y ¼ 1, 3, 5) were characterized via XPS, as shown in
Fig. S5. The peak fitting processes of Cr 2p, Ce 3d and O 1s over the
catalysts with different urea dosages are the same as those in the
above XPS analysis. However, the fitting results, such as the Cr6þ
and Ob ratios, are different. The optimal urea dosage for the prep-
aration of the double-shelled hollow CreCe mixed oxides is based
on the Cr6þ and Ob content. The specific Ce3þ, Cr6þ and Ob content
on the surface of the catalysts are listed in Table 2. The results show
that 10CCE-U31 exhibits the highest Ce3þ, Cr6þ and Ob content
among the three catalysts; thus, it possesses a relatively high
oxidizing ability.
Compared with the conventional syntheses of hollow-type CeO2
structures, which require complicated, time-consuming, expensive
templates [17], a one-pot technique for synthesizing CeO2 hollow
microspheres (adopting urea as a cheap template) is much more
convenient. The synthesis process of CreCe mixed oxides with a
double-shelled hollow morphology is illustrated in Fig. 1A. The
addition of urea has a surprising influence on the hollow structure
[28]. The synthesized 10CCE-U31 hollow microspheres display a
double-shelled structure in which four cores are located within the
second shell. The effects of urea on the physical structure of the
sample are shown in Fig. 1B, and C. Both 10CCE-U31 and 10CCE
exhibit four main peaks corresponding to the CeO2 cubic structure
(JCPDS No. 34e0394) [33]. Two catalysts display similar patterns,
indicating that the addition of urea did not influence the formation
of crystal structures. Additionally, compared with pure CeO2, the
XRD peaks of both catalysts shift to higher angles. Because the
3þ
radius of Cr3þ (r
¼ 0.0615 nm) is smaller than that of Ce4þ
Cr
4þ
(r ¼ 0.097 nm), we conclude that the Cr was doped into the CeO2
Ce
lattice. As listed in Table 1, 10CCE-U31 exhibits a higher BET surface
area than 10CCE. In addition, the N2 adsorption-desorption iso-
therms of both catalysts are different; 10CCE displays a type III
isotherm with a type H3 hysteresis loop, whereas 10CCE-U31 dis-
plays a type IV isotherm with a type H4 hysteresis loop [34,35], as
shown in Fig. S1. The SEM images reveal the differences in the BET
surface areas of the two catalysts; the BET surface area of 10CCE is
mainly generated by the aggregates of small nanospheres, whereas
the BET surface area of 10CCE-U31 results from its hollow structure,
thus the average pore size of 10CCE is higher than that of 10CCE-
U31, as shown in Table 1. The above results indicate that the addi-
tion of urea mainly influenced the morphologies of CreCe mixed
oxides rather than the crystal structure.
The varying Cr content in CeO2 was investigated to study its
impacts on the structure of the mixed metal oxides. As shown in
Fig. 2, with an increasing content of the Cr-containing precursor,
obvious angle shifts appear for the 5CCE-U31, 7CCE-U31 and
10CCE-U31 samples in their XRD patterns. Moreover, there is no
CrOx peak among the catalyst peaks, which indicates that all Cr
atoms were doped into the CeO2 lattice. The N2 adsorption-
desorption isotherms of the catalysts with different Cr doping
amounts are shown in Fig. S2A. The catalysts all exhibit type IV
isotherms with type H4 hysteresis loops, indicating the existence of
a narrow stenopaic pore structure [35]. The BET results are illus-
trated in Table 1, displaying that the surface area, pore volume and
pore size increased as the Cr doping increased. These findings
indicate that the addition of Cr atoms in conjunction with Ce atoms
was beneficial for the generation of pores during the hydrothermal
process, confirming the results of a previous study [36]. Addition-
ally, the increase in Cr doping could enlarge the pore size. SEM
images of the xCCE-U31 series catalysts are shown in Fig. 3A. The
varying Cr doping did not obviously influence the morphologies of
the catalysts. However, all catalysts display the obvious double-
shelled shape in the TEM images (Fig. 3B). The number of parti-
cles attached to the outmost shell increased as the Cr doping
increased, which is attributed to the expansile pore structure
caused by Cr doping. Combined with the BET results, the generated
pores on the shell could facilitate the transport of reactants and
increase the number of the exposed active sites, thereby promoting
the catalytic performance.
SEM images of 10CCE-U31 at different resolutions are shown in
Fig. S6. Double-shelled hollow microspheres are clearly observed.
The particle sizes are concentrated over 1.5
particle size is 2.25 m.
mm, and the average
m
A typical TEM image of 10CCE-U31 is displayed in Fig. 5A, which
clearly confirms its double-shelled hollow structure. As exhibited in
the HRTEM image (Fig. 5B), 10CCE-U31 nanocrystal subunits in the
shell with an interplanar spacing of 0.321 nm matches the (1 1 1)
crystallographic plane of cubic CeO2. The selected area electron
diffraction (SAED) image (Fig. 5C) of 10CCE-U31 displays four main
continuous rings corresponding to the (111), (2 0 0), (2 2 0) and (3
1 1) planes of CeO2 in accordance with the XRD analysis. The uni-
form distribution of each element in every shell and the cores of the
CreCe mixed oxide hollow microspheres investigated by energy-
The surface element valence state over the catalysts with
varying Cr doping was analyzed by XPS characterization, as shown
in Fig. 4. The survey spectra confirm the survival of Ce, Cr, and O in
the catalysts. In the 2p3/2 region of the Cr 2p spectra, all catalysts
are fitted into Cr3þ and Cr6þ states, which are assigned to the Cr2O3