A new highly active La2O3-CuO-MgO catalyst for the synthesis of cumyl peroxide by catalytic oxidation

Article abstract of DOI:10.1039/d1ra00176k

In this study, different magnesium, copper, lanthanide single metal, and composite multimetal oxide catalysts were preparedviathe coprecipitation route for the aerobic oxidation of cumene into cumene hydroperoxide. All catalysts were characterized using several analytical techniques, including XRD, SEM, EDS, FT-IR, BET, CO₂-TPD, XPS, and TG-DTG. La₂O₃-CuO-MgO shows higher oxidation activity and yield than other catalysts. The results of XRD and SEM studies show that the copper and magnesium particles in the catalyst are smaller in size and have a distribution over a larger area due to the introduction of the lanthanum element. The CO₂-TPD results confirmed that the catalyst has more alkali density and alkali strength, which can excite active sites and prevent the decomposition of cumene hydroperoxide. XPS results show that due to the promotional effect of La₂O₃, there are more lattice and active oxygen species in the catalyst, which can effectively utilize the lattice defects under the strong interaction between metal oxides for rapid adsorption and activation, thus improving the oxidation performance. Besides, La₂O₃-CuO-MgO exhibits good stability and crystalline structure due to its high oxygen mobility inhibiting coking during the cycle stability test. Finally, the possible reaction pathway and promotional mechanism on La₂O₃-CuO-MgO in cumene oxidation are proposed. We expect this study to shed more light on the nature of the surface-active site(s) of La₂O₃-CuO-MgO catalyst for cumene oxidation and the development of heterogeneous catalysts with high activity in a wide range of applications.

Full text of DOI:10.1039/d1ra00176k

RSC Advances
View Article Online
View Journal | View Issue
PAPER
A new highly active La2O3–CuO–MgO catalyst for
the synthesis of cumyl peroxide by catalytic
oxidation†
Cite this: RSC Adv., 2021, 11, 12532
*
XiaoYan Cao, JiaXin Su and Zhenggui Gu
HanShuang Liu,
KaiJun Wang,
In this study, different magnesium, copper, lanthanide single metal, and composite multimetal oxide
catalysts were prepared via the coprecipitation route for the aerobic oxidation of cumene into cumene
hydroperoxide. All catalysts were characterized using several analytical techniques, including XRD, SEM,
EDS, FT-IR, BET, CO2-TPD, XPS, and TG-DTG. La2O3–CuO–MgO shows higher oxidation activity and
yield than other catalysts. The results of XRD and SEM studies show that the copper and magnesium
particles in the catalyst are smaller in size and have a distribution over a larger area due to the
introduction of the lanthanum element. The CO2-TPD results confirmed that the catalyst has more alkali
density and alkali strength, which can excite active sites and prevent the decomposition of cumene
hydroperoxide. XPS results show that due to the promotional effect of La2O3, there are more lattice and
active oxygen species in the catalyst, which can effectively utilize the lattice defects under the strong
interaction between metal oxides for rapid adsorption and activation, thus improving the oxidation
performance. Besides, La2O3–CuO–MgO exhibits good stability and crystalline structure due to its high
oxygen mobility inhibiting coking during the cycle stability test. Finally, the possible reaction pathway and
promotional mechanism on La2O3–CuO–MgO in cumene oxidation are proposed. We expect this study
to shed more light on the nature of the surface-active site(s) of La2O3–CuO–MgO catalyst for cumene
oxidation and the development of heterogeneous catalysts with high activity in a wide range of applications.
Received 9th January 2021
Accepted 23rd February 2021
DOI: 10.1039/d1ra00176k
rsc.li/rsc-advances
The liquid-phase oxidation of isopropyl-aromatic hydrocar-
bons proceeds according to a widely-known free radical chain
Introduction
mechanism. In most industrial production processes, cumene
oxidation is typically carried out in the liquid phase, using O2
(or air) as an oxidant in the absence of any catalyst (auto-
oxidation process) and with small amounts of CHP or other
substances added as initiators. Cumene conversion is usually
kept low to maximize the selectivity toward CHP and minimize
side products resulting from CHP decomposition, mainly 2-
phenyl-2-propanol (cumyl alcohol, PP) and aceto-phenone (AP).3
Despite these precautions, the productivity of the auto-
oxidation process is generally below the desired values. There-
fore, in order to solve this problem, it is very necessary to
develop highly active and reusable homogeneous and hetero-
geneous catalysts for cumene oxidation to increase the
productivity of CHP. Recent research on the liquid-phase
oxidation of cumene-involved metal–organic frameworks
(MOFs),8,9 Salenalen-type complexes,10 N-hydroxyphthalimide
(NHPI),11 ionic liquids,12 and metal oxides as important
research components. Many homogeneous catalysts can
improve CHP yield under mild conditions, but they have
become stumbling blocks in the process of catalyst preparation,
product purication and recovery. When used in the actual
production process, it is difficult to achieve the expected results
with above catalysts. Heterogeneous catalysts are widely used in
Currently, liquid-phase oxidation of hydrocarbons is having
a profound impact on the industry and economy and can be
regarded as the core of chemical synthesis. This reaction plays
an important role in turning petroleum hydrocarbon raw
materials into industrial chemicals, and in polymer and petro-
chemical industries.1 For example, the oxidation of cumene is
one of the important way to produce cumene hydroperoxide
(CHP), phenol, and acetone in industry.2,3 CHP is used in
industry as an initiator for chain autoxidation and polymeri-
zation reactions, an accelerator for rubber vulcanization, an
oxidant in ne chemicals, and in polymer synthesis.4,5 Phenol is
an important raw material for phenol resins, perfumes, medi-
cines, agricultural chemicals, and many other industries.6 At
present, more than 90% of phenol production in the world is
based on this route.7 The products or by-products of the whole
process are of great industrial value and the chemical reaction
involves extremely high atom utilization.
Jiangsu Provincial Key Laboratory of Materials Cycling and Pollution Control, Nanjing
Normal Univesity, Nanjing, Jiangsu 210023, China. E-mail: 07160@njnu.edu.cn
† Electronic supplementary information (ESI) available. See DOI:
10.1039/d1ra00176k
12532 | RSC Adv., 2021, 11, 12532–12542
© 2021 The Author(s). Published by the Royal Society of Chemistry

View Article Online
Paper
RSC Advances
the synthesis of chemical products and have the advantages of the free radical reaction of cumene and could provide some
simple preparation, easy recycling, and long catalytic life. It is directions and perspectives on heterogeneous catalytic oxida-
the main trend in the development of the chemical industry and tion of hydrocarbons.
one of the important way to realize green chemical processes.
In past decades, many heterogeneous composite metal oxide
catalysts were investigated for the oxidation of cumene. In
Results and discussion
experimental studies on oxidation processes, many commonly-
Structural and morphology characterization
used transition metal catalysts, such as Cu(I)/Cu(II), Co(II)/(III),
Fe(II)/(III) and Mn(II)/(III), are typically investigated as oxides,13
complexes14 or loaded catalysts.15 Hsu et al.16 studied hetero-
geneous catalysis on transition metal oxides from the fourth
period for cumene oxidation, and found that these metal oxides
have both chain initiation and reaction acceleration capabil-
ities. Finally, 6.85% IPB conversion and 99% CHP selectivity
were obtained at 353 K. For all the above-mentioned catalytic
systems, copper compounds were excellent catalysts not only
with regard to the reaction activity but also with regard to the
CHP selectivity.17,18 Marella R. K. et al.19 found that the transi-
tion metal salt mixed with alkali metal oxides/alkaline earth
metal compounds showed an obvious improvement in the
catalytic activity. MgO is oen chosen as a support to enhance
the specic surface area of the catalyst.20 Based on infrared
studies reported in the literature, the hydrogen atom on the
isopropyl tertiary carbon of cumene chemically adsorbs on the
lattice oxygen with a negative basic center on the MgO surface,
thereby weakening the C–H bond on the isopropyl tertiary
carbon and facilitating the reaction. Cho et al.21 found that C–H
bonds were activated easily by Fe/MgO. V. V. Costa et al.20
investigated the liquid phase oxidation of various alcohols
catalyzed by Au/MgO with molecular oxygen as an oxidant.
However, none of the proposed catalysts were widely intro-
duced into the industry. Better suitability for industrial
production, boosting the overall activity and selectively
obtaining the desired products are signicant challenges for
cumene oxidation. There are few reports on the catalytic
performance of multi-metal oxides for cumene oxidation.
Therefore, based on the above-mentioned study on the oxida-
tion of cumene with binary heterogeneous solid bases, this
work was to continue exploration on the effects of rare earth
oxide doping on the catalyst structure and oxidation perfor-
mance. Some studies show that rare earth elements have
become important doping elements in various catalytic mate-
rials in recent years due to their special electronic structure.22–25
The catalytic initiative and stability of the mixed metal oxides
can be advanced by synergy with different metal ions.26,27 La is
a typical rare earth metal with good oxygen storage and release
capabilities, which can stabilize the structure of alkaline earth
metal oxides, improving the dispersion of active constituents24
and reduce the apparent activation energy of the catalyst.
Therefore, we report a high-activity La2O3–CuO–MgO ternary
composite solid base catalyst and evaluate its excellent activity
in the catalytic reaction of cumene. The material can be easily
prepared by the coprecipitation method, and the strong oxida-
tion capability of the prepared material is strongly proved by
sufficient characterization. Rich lattice oxygen and active
oxygen make this catalytic material have a long life. More
XRD measurements. The XRD patterns of C–M, L–M and
L–C–M catalyst samples are displayed in Fig. 1. The character-
istic reection of (200), (220), and (222) at 42.91ꢀ, 62.28ꢀ, and
78.59ꢀ, respectively, are assigned to the MgO phase. Two peaks
at 36.49ꢀ and 74.67ꢀ can be recognized for (211) and (220) crystal
planes, respectively, of CuO.28 The (100), (002), (101), (102), (110)
and (103) at 26.11, 29.13ꢀ, 29.96ꢀ, 39.53ꢀ, 46.08ꢀ and 52.13ꢀ
diffractions of the crystalline La2O3 phase are observed.29,30 The
diffraction peaks of La2CO5 at 22.84ꢀ, 29.55ꢀ, 30.82ꢀ, 44.51ꢀ and
46.56ꢀ can be indexed as (011), (013), (110), (020), and (022)
planes,31 respectively. For the multicomponent phase analysis,
diffraction peaks of (113) and (311) planes of La2CuO4 appear at
31.12ꢀ and 54.73ꢀ, and diffraction peaks of (311) and (331)
crystal planes of Cu2Mg appear at 39.28ꢀ and 57.17ꢀ,
respectively.
The peak intensity and particle size of the MgO and CuO
phases obviously decreased with the introduction of La. The
results indicate that La2O3 can effectively improve the disper-
sion of MgO and CuO phases and may form a new compound of
MgO and CuO in the medium or noncrystalline form. The
observed increase in the distribution of those phases can be
preliminarily assigned to the fact that the ion radius (0.106 nm)
of La3+ is much larger than that of Mg2+ and Cu2+ (0.072 nm and
0.073 nm, respectively). As a result, MgO and CuO microcrystals
are covered by La2O3 lms, which block the adhesion of parti-
cles, limit their grain growth during heat treatment, or cause
CuO to dissolve in the lattice of the solid solution of MgO, which
forms CuO–MgO. Furthermore, it was observed that some La2O3
was dissolved in CuO lattice, thanks to the solid–solid interplay
between La2O3 and CuO, which prompts the constitution of
importantly, this work discusses the mechanism of promoting Fig. 1 The XRD profiles of the C–M, L–M and L–C–M.
© 2021 The Author(s). Published by the Royal Society of Chemistry
RSC Adv., 2021, 11, 12532–12542 | 12533

View Article Online
RSC Advances
Paper
Fig. 2 The SEM pictures of the (a) C–M, (b) L–C–M.
lanthanum cuprate (La2CuO4).24 In the diffraction pattern of L–
The element distribution and composition of the catalyst
C–M, no diffraction peak of the La2O3 phase was found, indi- were further examined by EDS, as shown in Fig. 3g. Quantitative
cating the existence of extremely subtle La2O3 particles, whose elemental analysis shows that the weight percentages of C, O, Mg,
grain size became very small and lower than the diffraction Cu, and La elements in the L–C–M catalyst sample were 43.75%,
detection limit of the X-ray instrument. Nonetheless, the 32.14%, 4.28%, 0.38%, and 19.45%, respectively. The molar ratios of
absence of La2O3 as a single-phase may likewise be due to its Cu/Mg and La/Mg elements maintain good consistency with that
dissolution in the lattice of MgO and CuO, resulting in the designed at the time of preparation. Fig. 3a–f are the mapping
constitution of new complexes in a shapeless state or a badly diagrams of C, O, Mg, Cu, and La elements, respectively, showing
crystalline state.32
the distribution and density of each element on the sample surface.
Morphology and structures of catalysts. The morphology of Compared with Fig. 2a, it can be observed that Mg, Cu, and La
C–M and L–C–M catalyst samples was analyzed using a scan- elements are uniformly distributed on the sample surface, indi-
ning electron microscope, as shown in Fig. 2. It can be observed cating that the addition of La greatly improves the dispersion
that C–M consists of relatively uniform aggregates of rod-like performance of the C–M catalyst. These changes make it easier for
type29,33 of 2–3 mm of the width and 8–10 mm in length the metal oxide surface bind negatively charged lattice oxygen to the
(Fig. 2a). These aggregates are formed by MgO platelets and hydrogen atom on the isopropyl tertiary-carbon, weakening the
CuO. Aer the C–M modication with La2O3, the particle size bond energy between C–H bonds34 and increasing the catalytic
decreases from 3 to 0.1 and 10 to 1 mm of the width and length, activity of the catalyst. Besides, La2O3 itself has a good ability to store
respectively (Fig. 2b). Therefore, the addition of La2O3 reduces and release oxygen, which effectively enhances the catalytic life of
the particle size and improves the dispersion of the catalyst.
the catalyst, which is consistent with XRD and O 1s XPS analysis.
Fig. 3 X-ray energy spectrum analysis results of L–C–M sample: (a) scanning electron microscope image; (b) carbon element distribution
diagram; (c) oxygen element distribution diagram; (d) magnesium element distribution diagram; (e) copper element distribution; (f) lanthanum
element distribution diagram; (g) catalyst sample test energy spectrum.
12534 | RSC Adv., 2021, 11, 12532–12542
© 2021 The Author(s). Published by the Royal Society of Chemistry

View Article Online
Paper
RSC Advances
Table 1 Physical properties of catalysts
BET surface
area (m2 gꢁ1
Total pore volume
Average pore
diameter (nm)
Samples
)
(cm3 gꢁ1
)
C–M
L–M
L–C–M
48.98
20.84
4.15
0.28
0.18
0.06
22.87
30.15
43.58
is shown, the surface area and pore volume of catalysts
decreased greatly with the addition of La2O3, perhaps due to the
introduction of La2O3 particles into the pores of MgO or CuO,
which caused a part of the pores of the catalyst to be covered or
blocked, resulting in an order of magnitude reduction in the
total pore volume. All catalysts showed Langmuir adsorption
isotherms of type IV, which indicated that there was a large
proportion of mesopores in the catalysts. According to IUPAC
classication, the hysteresis in the isotherm for the catalyst is
H3 type,39 and in L–M and L–C–M, pores are stacked in a slit
shape. Its pore size distribution is more dispersed, which is
more conducive to molecular diffusion and promotes the
higher activity of the catalyst. Besides, a small amount of La2O3
in L–C–M diffuses to the CuO surface to form a CuO-based
La2O3–CuO solid solution. Due to the powerful interaction
between La and Cu, the bulk properties of the catalyst change,
which also results in the decrease of particulate surface area
and pore volume of catalyst.
Further, the BJH pore size distribution curve (inset), signies
that with the introduction of lanthanum element, the pore size
peak intensity of the catalyst is signicantly reduced, which
further indicates the blockage of the catalyst pores, in agree-
ment with XRD analysis outcomes.
Results of CO2-TPD analysis. As for the as-formed C–M, L–M
and L–C–M samples, the essential positions signify those
commonly associated with O2ꢁ ions on the surface.40 The
surface alkalinity of catalyst samples was detected using the
CO2-TPD technique.29 As exhibited in Fig. 6, all samples show
three CO2 desorption peaks, among which the feeble Brønsted
basic sites are concentrated in the 83–142 ꢀC region, and the
Fig. 4 The FT-IR transmission spectra of the C–M, L–M, L–C–M.
FT-IR studies. Fourier transform infrared (FT-IR) spectros-
copy is used essentially as an extra probe to demonstrate the
existence of OH groups and also additional organic and inor-
ganic chemical species. It not only proves the chemical changes
but may also reveal the interplay between the two metals. The
FTIR spectra in the range 4000–400 cmꢁ1 of the C–M, L–M and
L–C–M catalysts are shown in Fig. 4. For these three catalysts,
the wide absorption zone of 3350–3750 cmꢁ1 is related to the
superposition of the hydroxyl stretch zone n(OHstrs) produced by
water molecules between the metal hydroxyl group and the
hydrogen bond layer. This indicates the presence of hydroxyl
groups on the catalyst surface.35 The peak near 1640 cmꢁ1 is the
bending vibration of OH groups of physically adsorbed water.
Another transmission band is about 2920 cmꢁ1, corresponding
to the OH tensile vibration of carboxylic acid and CH asym-
metric tensile vibration. In the literature, the peaks at
1452 cmꢁ1 and 860 cmꢁ1 are ascribed toward the C–O extend
the pattern and the crooking pattern of structural carbonate,
individually. The bands observed in the low-frequency area may
be ascribed to lattice vibration patterns of M–O–M (M ¼ Mg, Cu,
La) and carbonate vibrations.36 Powerful and pointed bands
near 1507–1368 cmꢁ1 are assigned to extending and twisting
O–H vibrations of La(OH)3, with a general number of hydroxyl
groups at the surface, which plays the role of basic sites. The
absorption peak at 1100 cmꢁ1 corresponds to magnesium
propionate, and the band of CO3ꢁ in the skeletal oscillation in
propionates is situated around 1100–900 cmꢁ1. The intensity of
these bands decreases with the calcination of the catalyst, when
the metal propionate decomposes fully to metal oxide. The
frequency band below 800 cmꢁ1 is mainly due to metal–oxygen
vibrations.37 Another absorption band at about 678 cmꢁ1 stands
for the antisymmetric bending vibration of carbonate (n4).
When lanthanide was introduced, the absorption peaks at 870
and 860 cmꢁ1 are the characteristic peaks at La2O3 formed by
the decomposition of lanthanum propionate, and absorption
peaks at 500, 460 and 430 cmꢁ1 are assigned to CuO formed by
the decomposition of copper propionate.38
BET analysis. Physical features regarding several catalysts
are shown in Table 1, and the N2 adsorption–desorption
isotherms of the obtained new catalysts are shown in Fig. 5. As
Fig. 5 N2 adsorption–desorption isotherms and pore-size distribution
curve (inset) of C–M, L–M, L–C–M.
© 2021 The Author(s). Published by the Royal Society of Chemistry
RSC Adv., 2021, 11, 12532–12542 | 12535

View Article Online
RSC Advances
Paper
associated with Cu2+ type lies in the area of 933.28–935.66 eV,
while its disturbed satellite thread (s peak) lies between 941.29–
944.02 eV. There are obvious vibrational characteristics between
940–945 eV, which also excludes the possibility of Cu2O. The
results indicate that Cu on the MgO surface of the prepared
C–M and L–C–M catalysts exists in the form of Cu2+. Satellite
peaks indicate that if sudden approximation (unipolar selection
rule) and ligand eld theory are used, Cu2+ compounds are
caused by the unipolar charge transfer transition (ligand metal
3d). Since the 3d shell is lled, these satellite peaks cannot be
observed in Cu+ compounds or metallic Cu.44 In fact, transition
metal ions including unlled 3d orbits are widely reported to
display satellite peaks under the nuclear XP spectrum, which is
owing to electronic vibration and satellite structure, that is, the
number of peaks, intensity, and splitting reecting chemical
bond properties of transition metal ions.45 As for C–M and
Fig. 6 The CO2-TPD pictures of the C–M, L–M and L–C–M.
2+
L–C–M samples, the peak at 933.25–933.33 eV (CuAp peak) is
ꢀ
attributed to the extremely scattered Cu2+ species within the
moderate Lewis basic sites are concentrated in the 196–355 C
region. The powerful Lewis basic sites are concentrated near
550–730 ꢀC, which is connected to the surface OHꢁ groups and
coordinated unsaturated O2ꢁ ions. Nevertheless, the C–M
sample presents a weak desorption peak at 670 ꢀC, indicate that
only a few strongly basic (SB) sites.
In light of the quantitative analysis on the denseness of basic
sites on C–M, L–M and L–C–M samples (Table 2), the denseness
of the total basic sites and SB sites are as follows: C–M < L–M <
L–C–M. Previous CO2-TPD and FTIR studies41,42 have indicated
that the hydroxy groups on the surface of La2O3 have basic sites.
Combined with the results of XRD, due to the interaction
between La2O3 and CuO/MgO, CO2 is adsorbed on the strong
alkaline centers, that is, there are SB sites, which greatly
increase the total alkali content.
metal oxide (CuA2+), while the peak near 935.37–935.66 eV
2+
(CuBp peak) may be related to the covalent Cu2+–O bond
polarized by Mg2+ ions, thus reducing the efficient charge of
Cu2+ species. Noticeably, the doping of rare earth metal La is not
2+
only benecial for the appearance of the main peak CuAp
2+
species but also benecial for the formation of CuBp species.
This result can still be observed according to the data given in
Table 3. Therefore, compared with the C–M catalyst, L–C–M not
only has more highly dispersed Cu2+ species but also shows
a strengthened interaction between Cu–Mg species. These
results indicate the generation of lattice defects on the surface
catalyst, resulting in more amorphous CuO–MgO solid solu-
tions, which enhance the reaction activity of the catalyst. This is
consistent with XRD, BET, and SEM analysis results.
XPS results. As is known, the catalytic properties of catalysts
are closely associated with their surface chemistry states. The
surface component and electronic states of C–M, L–M and
L–C–M were analyzed by X-ray photoelectron spectroscopy
(XPS). The scanning spectra of the three samples show that
there are corresponding C, O, Mg, Cu, and La elements on the
surface, which is also very consistent with the elemental molar
ratio analyzed using EDX (Fig. 7a–c). Table 3 shows the distri-
bution characteristics of the corresponding element molar
ratios of Cu 2p and O 1s. According to the overall trend shown in
Fig. 7d, the main spectral peak (p peak) of Cu 2p in the catalyst
doped with La tends to shi to low energy.43 Ordinarily, the
binding energy (BE) of the primary Cu 2p3/2 peak (p peak)
Taken into consideration that the source of alkaline sites is
associated with the types of surface oxygen, XPS spectra of O 1s
of samples were analyzed. As shown in Fig. 7e, the peak at
529.41 eV is attributed to the lattice oxygen (Oa and O2ꢁ) and the
peak at 531.29 eV is attributed to the chemically adsorbed
surface-active oxygen (Ob, Oꢁ, and O22ꢁ). The peak with the
value of 532.23 eV may be related to hydroxyl groups or water
molecules (Og) adsorbed to the catalyst surface.46 It can be seen
from Table 3 that due to the doping of La, the concentration of
Ob is greatly increased, and Ob plays a vital role in the oxidation
reaction and is consistent with great inuence on the catalytic
activity concerning the catalyst. The molar quotient of Ob/(Oa +
Ob) on these catalysts follows the order, C–M < L–M < L–C–M.
This may be because La2O3 and CuO/MgO in the catalyst are
closely interlaced, which leads to defects, and molecular oxygen
can be quickly absorbed and activated by vacancy defects in the
catalyst. Therefore, active oxygen can be easily transferred to
supplement the surface-active oxygen consumed in the oxida-
tion process. Furthermore, it can be observed that the concen-
tration of Oa is also increased signicantly, which also indicates
that the introduction of La can indeed promote the release of
oxygen in the CuO/MgO lattice.47 According to the literature, the
presence of lattice oxygen is benecial to the selective oxidation
of C–H compounds to produce aldehydes, ketones, peroxides,
and other target products. The formation of strong alkaline
Table 2 The alkaline test results of the C–M, L–M and L–C–M
Total basic sites
SB sites
Quantity
Temperature
(ꢀC)
Quantity
Samples
(cm3 per g STP)
(cm3 per g STP)
C–M
L–M
L–C–M
21.80
55.14
60.55
670.25
633.92
672.43
2.32
30.14
46.60
12536 | RSC Adv., 2021, 11, 12532–12542
© 2021 The Author(s). Published by the Royal Society of Chemistry

View Article Online
Paper
RSC Advances
Fig. 7 XPS survey spectrum of (a) C–M, (b) L–M and (c) L–C–M. High resolution XPS survey at (d) Cu 2p region, and (e) O 1s region.
sites on the catalyst surface can be attributed to the existence of 660 ꢀC, during which, the total mass changes the least, and
lattice oxygen.48
gradually tends to thermal stability, which is mainly due to the
Thermo analysis. The TG curves and derivative TG (DTG) release of carbon dioxide from carbonate during roasting,
curves of the C–M, L–M and L–C–M materials are shown in reecting the existence of alkaline sites.49
Fig. 8. From the TG curve, it can be observed that the mass loss
However, the introduction of lanthanum metal moved the
of all materials occurs in three stages. The rst thermic event signal slightly towards a higher degradation temperature (DTG
compares to the loss of mass around 160–200 ꢀC, mainly owing curves) (Fig. 8a–c). This result can be attributed to the inuence
to the liberation of physically adsorbed water on the outer of the dehydroxylation temperature of La2O3 on the particle size
surface of the catalyst. The second thermal event is in the and crystallinity.50 Due to the change in crystal grains and size,
temperature range of 380 ꢀC to 450 ꢀC, corresponding to the the speed at which it reaches equilibrium is delayed because
DTG curve. It is observed that there is a large weight loss peak at different crystal grains and sizes have different gas interference
this stage, which is the removal of some chemically bound water diffusion rates. These results are consistent with those obtained
in the catalyst, the dehydroxylation of the interlayer hydroxyl using the structure size of the material changes, and the
and the decomposition of carbonate.29 The third is at 530– amorphous property increases with the introduction of
lanthanum. Therefore, all these factors may lead to an increase
in the decomposition temperature of oxides.
Catalytic performance evaluation. The catalytic activity test
results of C–M, L–M, L–C–M, and single metal oxide for IPB
total oxidation are shown in Fig. 9a. The catalytic activity of the
Table 3 Surface atomic concentration of different elements in C–M,
L–M and L–C–M catalysts using XPS results
composite metal oxide is better than that of single metal,
especially aer doping with rare earth metal lanthanum, its
conversion rate is greatly improved. According to BET and SEM
results, the doping of La makes the pore size distribution of
catalyst crystals more dispersed, strengthens the interaction
between Cu–Mg, produces more lattice defects, and promotes
Surfer element molar ratio/%
2+
2+
2+
2+
Samples CuAp
CuBp
CuAs
CuBs
Oa
Ob
Og
C–M
L–M
L–C–M
59.76
—
11.86
—
16.01
—
12.36
—
15.89 65.43 18.68
17.74 67.23 15.03
20.33 71.03
69.01
15.83
8.91
6.25
8.64
© 2021 The Author(s). Published by the Royal Society of Chemistry
RSC Adv., 2021, 11, 12532–12542 | 12537

View Article Online
RSC Advances
Paper
Fig. 8 TG and DTG curves of the (a) C–M, (b) L–M and (c) L–C–M.
the adsorption of active oxygen on hydrogen atoms on isopropyl impossible to continue to provide isopropyl radicals for branching
tertiary carbon more easily, thus showing better oxidation reactions.3 So in the exploration of reaction conditions, too high
performance of IPB. The element ratio of the catalyst and the a temperature or too fast oxygen ow rate will affect the state of
reaction conditions were also studied. As shown in Table 4, we provided free radicals, and then affect the yield of CHP. Aer single
investigated the elemental ratio of the catalyst and found that factor optimization (Fig. S1†), the conversion of IPB reaches 95.50%,
L
0.8–C0.03–M1 has the best catalytic activity. When the amount of the selectivity and yield of CHP reach 60.57% and 57.84%, respec-
Cu was increased, although the conversion of IPB increased tively. The oxygen ow rate was 500 ml minꢁ1 and 90 ꢀC for 6 h from
slightly, the selectivity to CHP decreased signicantly. Accord- 2 wt% L0.8–C0.03–M1.
ing to the literature, due to the strong redox performance of Cu,
Furthermore, based on the results of CO2-TPD and O 1s XPS
an increase in its content leads to poor dispersibility. The and the data in Table 4, the L–C–M sample not only shows more
interaction force between La and Cu is obviously greater than SB sites for stimulating the reaction activity but also has more
the rate of oxygen overowing to the carrier surface. It causes Ob. As we know, Ob performs a signicant part in the oxidation
the rapid decomposition of CHP, and the content of alcohol and reaction system, that can effectively utilize the lattice defects in
copper increases and the selectivity of CHP decreases. Therefore, it is catalytic molecules to adsorb and activate quickly and can
Fig. 9 IPB conversion (a) and cyclic stability test results of L–C–M samples (b).
12538 | RSC Adv., 2021, 11, 12532–12542
© 2021 The Author(s). Published by the Royal Society of Chemistry

View Article Online
Paper
RSC Advances
Table 4 SB sites, the molar ratio of Ob and catalytic oxidation activity of different composite oxide catalystsa
SB sites
S (%)
Quantity
(cm3 per g STP)
Ob/Ob + Oab (%)
CIPB (%)
CHPd
PP
AP
YCHPe (%)
c
Samples
Temperature (ꢀC)
C–M
L–M
*Co(BPZ)
*[MP4]Br
*Co-BTC
*NHPI/Cu(acac)2
670.25
633.92
—
—
—
—
—
672.43
—
—
2.32
30.14
—
—
—
—
—
46.60
—
—
0.65
0.67
—
—
—
—
—
0.71
—
—
53.26
67.50
34
16.7
49
56.59
58.90
84
87.7
69
34.69
39.56
15
—
—
8.43
0.35
1
—
—
30.14
39.76
28.56
14.65
33.81
29.92
57.46
57.84
53.02
48.49
44.83
68
44
45
10
L0.8–C0.01–M1
L0.8–C0.03–M1
L0.8–C0.05–M1
L0.8–C0.07–M1
L0.8–C0.09–M1
81.35
95.50
95.75
96.23
98.22
70.64
60.57
55.37
50.39
45.64
27.43
36.23
39.46
43.12
44.27
0.57
1.75
3.43
5.85
9.79
—
—
—
a
Reaction conditions: raw material (IPB) ¼ 20 ml, initiator (CHP) ¼ 0.2 ml, composite oxide catalyst ¼ 0.4 g, reaction temperature ¼ 90 ꢀC; time ¼
6 h; rate ¼ 500 ml minꢁ1
.
The surfer element molar ratio of Ob in O 1s XPS test. c Conversion of IPB based on the DAD-HPLC results ¼ moles of IPB
b
reacted/[initial moles of IPB used] ꢂ 100. d Selectivity of product calculated by total moles of CHP formed/total moles of IPB converted. e Yield of
CHP ¼ conversion ꢂ selectivity.
transfer and replenish automatically, thus accelerating the the possible reaction mechanism of the oxidation of IPB on the
circulation of the reaction. It is precise because of its high L–C–M catalyst is presented in Fig. 11. The peroxidation of
oxygen mobility that effectively inhibits the coking of the cata- cumene can be described as a typical free radical chain reac-
lyst. Notably, rare earth lanthanum itself has good oxygen tion.13,51 In most cases, the process can be regarded as a pre-
storage and release capacity, thus effectively enhances the determined automatic oxidation process for reactions (1)–
catalytic life of the catalyst. In this study, the recovered L–C–M (5).52,53 Its catalytic peroxidation process on L–C–M involves the
sample was washed with ethanol (specic operations are added combination of free radical reactions on the surface of and
in the ESI†), and then the IPB catalytic oxidation experiment inside of particles.
was repeated many times. As shown in Fig. 9b, the catalytic
activity did not change signicantly, indicating that it has good
stability and catalytic life.
The samples were characterized and compared using XRD
aer several stability tests of the recovery cycle. As displayed in
Fig. 10, the diffraction patterns on fresh samples and used
samples have no obvious changes, which indicates that each
phase and particle size about the samples are not affected aer
being used many times.
Initiation:
ROOH / ROc + cOH
(1)
(2)
ROc + RH / ROH + Rc
Possible promotional mechanism analysis. Summarizing
the results obtained from this study and works from literature,
Fig. 10 XRD patterns profiles of fresh and used samples.
Fig. 11 Catalytic mechanism of cumene free radical chain oxidation.
RSC Adv., 2021, 11, 12532–12542 | 12539
© 2021 The Author(s). Published by the Royal Society of Chemistry

View Article Online
RSC Advances
Paper
Propagation:
Magnesium nitrate hexahydrate, copper nitrate trihydrate,
lanthanum nitrate hexahydrate, and anhydrous sodium
carbonate were provided by Sinopharm Chemical Reagent Co.,
Ltd., oxygen was procured from Nanjing Tianze Gas Co., Ltd.
Rc + O2 / ROOc
(3)
(4)
ROOc + HR / ROOH + Rc
Preparation of CuO–MgO
Termination:
C–M was prepared by dropping a 1 mol lꢁ1 of Na2CO3 aqueous
solution into 100 ml stirred aqueous solution of Cu(NO3)2-
$3H2O and Mg(NO3)2$6H2O weighed under different molar
ratios as n(Cu) : n(Mg) under constant churning. The precipitate
was washed to a pH of 7. Filtered and dried in an oven at 373 ꢀC
for 12 hours. The dried sample was next calcined in air at 873 K
for 6 h to obtain C–M.
2ROOc / non-radical products
(5)
The initiation (eqn (1) and (2)), propagation (eqn (3) and (4)),
and termination (eqn (5)) are typical steps that occur in free
radical chain reactions.
The particle surface involves rapid adsorption of CHP
(initiator formed and added during the entire reaction).
Because of the strong interaction between La2O3–CuO and Cu2+
ions, well-dispersed CuO species are distributed around its site
(–La3+–Ov–Cu2+–O–La3+–) and form more oxygen vacancies (Ov)
to become the adsorption and activation sites of CHP, which are
then decomposed into ROc and HOc.54 This process is also well
reected in the internal reaction, especially for the part that
induces the activation of IPB. Cu2+ itself has the effect of initi-
ating a chain reaction to generate ROc radicals, thereby accel-
erating the effective progress of the reaction.
Preparation of La2O3–MgO
L–M was prepared by dropping a 1 mol lꢁ1 Na2CO3 aqueous
solution into 100 ml of a blended aqueous solution of
La(NO3)3$6H2O and Mg(NO3)2$6H2O weighed in line with
different molar ratios of n(La) : n(Mg) under continual
agitating. The precipitate was washed until the pH of the
washing was 7, ltered, and the cake was dried in the oven at
373 ꢀC for 12 hours. The dried sample was later calcined in air at
973 K for 6 h to acquire L–M.
The alkaline earth metal magnesium oxide has a special
effect on the activation of IPB. The negatively charged lattice
oxygen provided by the alkali center on the surface of the
alkaline earth metal catalyst is chemically adsorbing the
hydrogen atom from the isopropyl tertiary carbon, thus weak-
ening the C–H bond on the isopropyl tertiary carbon. According
to the results of O 1s XPS, more lattice and active oxygen species
are generated in the L–C–M catalyst, making more O2ꢁ interact
with the hydrogen bond of IPB tertiary carbon. As a result,
a large number of surface oxygen vacancies are used for the
adsorption–activation of oxygen molecules, which improves the
mobility of oxygen and accelerates the overall conversion rate.55
In the reaction process, the dissociation enthalpy of the O–H
bond is 104 kcal molꢁ1 in benzol RO–H and 88 kcal molꢁ1 in
CHP ROO–H.56 The rate of extracting hydrogen atoms (S2) from
IPB molecules by ROc free radicals is much faster than that of
hydrogen abstraction reaction (S1) between ROOc and IPB.3 The
formation of CHP and Rc (S2) becomes the control step of the
whole radical chain reaction. Improving transformation capa-
bility is an efficient way to effectively stimulate the tertiary
carbon–hydrogen bond for IPB. Therefore, improvements in
oxygen adsorption–activation and L–C–M interaction both
contribute to the superior catalytic performance in IPB total
oxidation.
Preparation of La2O3–CuO–MgO
L–C–M was prepared by dropping 1 mol lꢁ1 of Na2CO3 aqueous
solution into 100 ml of a mixed aqueous solution of Cu
(NO3)2$3H2O, La(NO3)3$6H2O and Mg(NO3)2$6H2O weighed
according to different molar ratios of n(La) : n(Cu) : n(Mg) under
continuous stirring. The precipitate was washed until the pH of the
supernatant was 7, then ltered and the cake was dried in the oven
at 373 ꢀC for 12 hours. The dried sample was calcined in air at
a temperature of 873 K for 6 h to obtain L–C–M.
Catalysts characterization
X-ray powder diffraction (XRD) patterns were collected on a Rigaku
XRD-6000 diffraction instrument equipped with Cu Ka radiation (l ¼
ꢀ
1.5406 A) (Shimadzu, Kyoto, Japan). The microstructures of the
materials were characterized using a scanning electron microscope
(SEM, Hitachi-S4800, Tokyo, Japan). Energy-dispersive (EDS) study
was performed using an EX 250 spectrophotometer (Horiba, Kyoto,
Japan). Fourier transform infrared spectroscopy (FT-IR) was per-
formed on TNEXUS-670 (Thermo Nicolet Corporation, USA). The
surface area (BET) measurements by nitrogen adsorption at 77 K were
determined by ASAP 2020. Before the measurement, samples were
heated in a vacuum at 353 K. CO2-TPD was conducted using an
AutoChem II 2920 apparatus to survey the total alkalinity and alkali
strength. X-ray photoelectron spectroscopy (XPS) examination was
performed on an ESCALAB 250Xi instrument (Thermo, USA). TG-DTA
was conducted from 100 to 800 ꢀC at a heating rate of 10 ꢀC minꢁ1
and 100 ml minꢁ1 of airow on a thermobalance (PerkinElmer, USA).
Experimental section
Materials
The catalytic oxidation of cumene was implemented on three
catalysts viz., CuO–MgO (C–M), La2O3–MgO (L–M), and La2O3–
CuO–MgO (L–C–M). All the chemicals used to prepare the
Oxidation on the catalyst
catalyst were of AR grade. Cumene and CHP were purchased As a representative reaction, liquid-phase oxidation of cumene
from Shanghai Aladdin Biochemical Technology Co., Ltd. was performed in a circular-bottom ask equipped with
12540 | RSC Adv., 2021, 11, 12532–12542
© 2021 The Author(s). Published by the Royal Society of Chemistry

View Article Online
Paper
RSC Advances
a condenser. Raw materials cumene, solid base catalyst (2%),
and initiator CHP (1%) were added in succession into the ask.
The reaction was conducted with continuous stirring at
a certain time and temperature. Aer the reaction, the products
were analyzed using Agilent High-Performance Liquid Chro-
matography (Model 1260). The product identication was per-
formed by diluting the standard sample with methanol and
injecting it into the high-performance liquid chromatography.
The catalyst was reclaimed through centrifugation and washed
thoroughly with ethanol for reuse.
6 N. M. Saleh, S. Asman and B. S. Yih, Anal. Methods, 2021, 13,
607–619.
7 B. A. Mantooth, E. Sykes, P. Han, A. M. Moore,
Z. J. Donhauser and P. S. Weiss, J. Phys. Chem. C, 2007,
111, 6167–6182.
8 A. Nowacka, R. Vismara, G. Mercuri, M. Moroni and
A. Rossin, Inorg. Chem., 2020, 59, 8161–8172.
´
9 A. E. Nowacka, P. Briantais, C. Prestipino and i. Llabres, ACS
Sustainable Chem. Eng., 2019, 7, 7708–7715.
10 B. Ghanbari and S. Tafazolian, Res. Chem. Intermed., 2012,
38, 871–883.
11 G. Dobras and B. Orlinska, Appl. Catal., A, 2018, 561, 59–67.
12 X. Shuai, Z. Jie, B. Chen and Z. Lei, Chem. Eng. Process., 2018,
130, 88–92.
13 S. A. Xu, C. P. Huang, J. Zhang and B. H. Chen, Korean J.
Chem. Eng., 2009, 26, 1568–1573.
14 W. J. Yang, C. C. Guo, N. Y. Tao and J. Cao, Kinet. Catal.,
2010, 51, 194–199.
15 H. Wang, W. Fan, Y. He, J. Wang, J. N. Kondo and
T. Tatsumi, J. Catal., 2013, 299, 10–19.
Conclusions
The following are the major conclusions of the work: (1) the
results show that the IPB oxidation reaction catalyzed on the
L–C–M catalyst showed higher activity and yield. Which
provides a potential scheme for the related research due to its
low cost, convenient post-treatment, and being environment-
friendly. (2) The higher the dispersion of copper and magne-
sium particles, the stronger is the synergism between
lanthanum and copper and magnesium, which seems to be the
reason for the higher activity of the L–C–M catalyst, according to
the characterization results of XRD, SEM, CO2-TPR, and XPS. Its
role is to promote the transfer of electrons between ions,
induces the formation of more oxygen vacancies, thereby
improving the volume of chemically adsorbed oxygen on the
catalyst surface and improve the oxidation performance. (3) The
cyclic stability test of L–C–M shows that its high oxygen mobility
inhibits the coking of the catalyst, so it maintains a good crystal
structure and stability, as evidenced from O 1s XPS analysis.
16 Y. F. Hsu and C. P. Cheng, J. Mol. Catal. A: Chem., 1997, 120,
109–116.
17 S. Matsui and T. Fujita, Catal. Today, 2001, 71, 145–152.
18 O. Furuhashi, T. Kinugawa, T. Hirayama, T. Koizumi,
C. Yamada and S. Ohtani, Chem. Phys., 2003, 295, 185–193.
19 R. K. Marella, C. Neeli, S. Kamaraju and D. R. Burri, Catal.
Sci. Technol., 2012, 2, 1833–1838.
20 V. V. Costa, M. Estrada, Y. Demidova, I. Prosvirin,
V. Kriventsov, R. F. Cotta, S. Fuentes, A. Simakov and
E. V. Gusevskaya, J. Catal., 2012, 292, 148–156.
21 S. H. Cho, M. S. Cheong, K. D. Jung, C. S. Kim and S. H. Han,
Appl. Catal., A, 2004, 267, 241–244.
Conflicts of interest
22 V. H. Rane, S. T. Chaudhari and V. R. Choudhary, J. Nat. Gas
Chem., 2010, 19, 25–30.
23 V. R. Choudhary, V. H. Rane and S. T. Chaudhari, Indian J.
Chem. Technol., 2004, 11, 569–574.
There are no conicts to declare.
24 G. El-Shobaky, N. R. E. Radwan, M. S. El-Shall, A. M. Turky
and H. Hassan, Colloids Surf., A, 2009, 345, 147–154.
25 V. R. Choudhary, V. H. Rane and S. T. Chaudhari, Fuel, 2000,
79, 1487–1491.
26 S. Irusta, L. M. Cornaglia and E. A. Lombardo, J. Catal., 2002,
210, 7–16.
Acknowledgements
The authors acknowledge nancial support from Jiangsu
Daoming Petroleum & Chemical Corporation (S11090B41904),
the relevant test and analysis results of this work are supported
by the Analysis and Test Center of Nanjing Normal University.
27 H. Y. Wang and E. Ruckenstein, Appl. Catal., A, 2000, 204,
143–152.
28 H. Dizaj ghorbani, H. Azadi, M. Esmaeilzadeh, A. Moemen
Bellah and R. Malekfar, Opt. Mater., 2019, 91, 433–438.
29 D. S. Desai and G. D. Yadav, Ind. Eng. Chem. Res., 2019, 58,
16096–16105.
References
1 A. K. Suresh, M. M. Sharma and T. Sridhar, Ind. Eng. Chem.
Res., 2000, 39, 3958–3997.
2 G. M. Westhoff, B. F. M. Kuster, M. C. Heslinga, H. Pluim and
M. Verhage, Lactose and Derivatives, American Cancer 30 G. Li, C. Peng, C. Zhang, Z. Xu, M. Shang, D. Yang, X. Kang,
Society, 2014.
W. Wang, C. Li, Z. Cheng and J. Lin, Inorg. Chem., 2010, 49,
10522–10535.
3 M. Weber, J. Daldrup and M. Weber, Noncatalyzed Radical
Chain Oxidation: Cumene Hydroperoxide, John Wiley & Sons, 31 D. Singh, B. Reddy, A. Ganesh and S. Mahajani, Ind. Eng.
Ltd, 2016.
Chem. Res., 2014, 53, 18786–18795.
4 S. Liao, F. Peng, H. Yu and H. Wang, Appl. Catal., A, 2014, 32 N. R. E. Radwan, Appl. Catal., A, 2006, 299, 103–121.
478, 1–8.
33 M. Zhang, L. Wang, H. Ji, B. Wu and X. Zeng, J. Nat. Gas
Chem., 2007, 16, 393–398.
5 S. Liao, Y. Chi, H. Yu, H. Wang and F. Peng, ChemCatChem,
2014, 6, 555–560.
© 2021 The Author(s). Published by the Royal Society of Chemistry
RSC Adv., 2021, 11, 12532–12542 | 12541

View Article Online
RSC Advances
Paper
34 R. Wilson, R. Thalji, R. Bergman and J. Ellman, Org. Lett., 46 V. D. Makwana, Y. C. Son, A. R. Howell and S. L. Suib, J.
2006, 8, 1745–1747. Catal., 2002, 210, 46–52.
35 Q. Y. Mu and Y. D. Wang, J. Alloys Compd., 2011, 509, 396– 47 G. Liu, L. Sun, J. H. Liu, F. Wang and C. J. Guild, Mol. Catal.,
401. 2017, 440, 148–157.
36 Y. W. Guo, Z. L. Zhu, Y. L. Qiu and J. F. Zhao, J. Hazard. 48 W. Luo, L. Sun, Y. Yang, Y. Q. Chen, Z. Zhou, J. H. Liu and
Mater., 2012, 239, 279–288. F. Wang, Catal. Sci. Technol., 2018, 8, 6468–6477.
37 P. Gao, F. Li, H. Zhan, N. Zhao, F. Xiao, W. Wei, L. Zhong, 49 S. Zadaviciute, A. Bankauskaite, K. Baltakys and A. Eisinas, J.
H. Wang and Y. Sun, J. Catal., 2013, 298, 51–60. Therm. Anal. Calorim., 2018, 131, 521–527.
38 N. Murugesan, A. M. Remona, S. K. Kumar and S. Suresh, 50 R. M. Cornell and U. Schwertmann, The Iron Oxides:
Mater. Lett., 2018, 222, 100–104.
Structure, Properties, Reactions, Occurences and Uses, 2nd
edn, 2004.
51 C. O. L. Crites, G. L. Hallett-Tapley, M. Frenette,
M. Gonzalez-Bejar, J. C. Netto-Ferreira and J. C. Scaiano,
ACS Catal., 2013, 3, 2062–2071.
52 K. E. Kharlampidi, K. A. Tereshchenko, T. S. Nurmurodov,
D. A. Shiyan and N. V. Ulitin, Chem. Eng. J., 2020, 392,
123811.
39 M. Thommes, B. Smarsly, M. Groenewolt, P. Ravikovitch and
A. Neimark, Langmuir, 2006, 22, 756–764.
40 J. Valente, E. Lima, J. Toledo, M. Cortes-Jacome, L. Lartundo-
Rojas, R. Montiel and J. Prince, J. Phys. Chem. C, 2010, 114,
2089–2099.
41 S. C. Shen, X. Y. Chen and S. Kawi, Langmuir, 2004, 20, 9130–
9137.
42 G. Colon, J. A. Navio, R. Monaci and I. Ferino, Phys. Chem. 53 A. Bhattacharya, Chem. Eng. J., 2008, 137, 308–319.
Chem. Phys., 2000, 2, 4453–4459.
43 S. Pradhan, A. S. Reddy, R. N. Devi and S. Chilukuri, Catal.
Today, 2009, 141, 72–76.
44 G. Avgouropoulos and T. Ioannides, Appl. Catal., A, 2003,
244, 155–167.
54 F. F. Wang, S. Y. Jia, D. L. Li, B. Yu, L. W. Zhang, Y. Liu,
X. Han, R. Zhang and S. H. Wu, Mater. Lett., 2016, 173, 72–75.
55 Y. Zeng, K. G. Haw, Z. Wang, Y. Wang, S. Zhang,
P. Hongmanorom, Q. Zhong and S. Kawi, J. Hazard. Mater.,
2021, 404, 124088.
45 D. C. Frost, A. Ishitani and C. A. McDowell, Mol. Phys., 1972, 56 F. Minisci, F. Recupero, A. Cecchetto, C. Gambarotti,
24, 861–877.
C. Punta, R. Paganelli, G. F. Pedulli and F. Fontana, Org.
Process Res. Dev., 2004, 8, 163–168.
12542 | RSC Adv., 2021, 11, 12532–12542
© 2021 The Author(s). Published by the Royal Society of Chemistry