An efficient environmentally friendly CuFe2O4/SiO2catalyst for vanillyl mandelic acid oxidation in water under atmospheric pressure and a mechanism study

Article abstract of DOI:10.1039/d0nj04798h

With the aim of the green production of vanillin, a highly efficient environmentally friendly oxidation system was introduced to oxidize vanillyl mandelic acid (VMA) with a porous CuFe2O4/SiO2 component nano-catalyst in aqueous solution under atmospheric pressure. The N2 adsorption-desorption pattern indicated that CuFe2O4/SiO2 possessed a much higher specific surface area (49.98 m2 g-1) than that of CuFe2O4 (5.02 m2 g-1), which further indicated that the SiO2 substrate restrained the aggregation of CuFe2O4 nanoparticles. The conversion for VMA and selectivity for vanillin reached 98% and 96%, respectively, under atmospheric pressure. The excellent catalytic performance was attributed to the synergistic effect of the catalytic capacity of CuFe2O4 and the adsorption capacity for the reactant of SiO2. Simultaneously, the effect of different reaction conditions for catalyst activity and selectivity were investigated. Furthermore, the probable mechanism of VMA oxidation was investigated by in situ ATR-FTIR, H2-TPR, XPS and 1H NMR. More importantly, the decarboxylation was verified to proceed in basic conditions rather than in conventional acidic conditions. This journal is

Full text of DOI:10.1039/d0nj04798h

NJC
View Article Online
View Journal | View Issue
PAPER
An efficient environmentally friendly CuFe2O4/
SiO2 catalyst for vanillyl mandelic acid oxidation
in water under atmospheric pressure and a
mechanism study†
Cite this: New J. Chem., 2021,
45, 982
Haifang Mao,a Hongzhao Wang,a Tao Meng,a Chaoyang Wang,a Xiaojun Hu,*a
a
Zuobing Xiaob and Jibo Liu
*
With the aim of the green production of vanillin, a highly efficient environmentally friendly oxidation
system was introduced to oxidize vanillyl mandelic acid (VMA) with a porous CuFe2O4/SiO2 component
nano-catalyst in aqueous solution under atmospheric pressure. The N2 adsorption–desorption pattern
indicated that CuFe2O4/SiO2 possessed a much higher specific surface area (49.98 m2
g
À1) than that of
CuFe2O4 (5.02 m2 À1), which further indicated that the SiO2 substrate restrained the aggregation of
g
CuFe2O4 nanoparticles. The conversion for VMA and selectivity for vanillin reached 98% and 96%,
respectively, under atmospheric pressure. The excellent catalytic performance was attributed to the
synergistic effect of the catalytic capacity of CuFe2O4 and the adsorption capacity for the reactant of SiO2.
Simultaneously, the effect of different reaction conditions for catalyst activity and selectivity were investigated.
Furthermore, the probable mechanism of VMA oxidation was investigated by in situ ATR-FTIR, H2-TPR, XPS
Received 29th September 2020,
Accepted 4th December 2020
DOI: 10.1039/d0nj04798h
1
and H NMR. More importantly, the decarboxylation was verified to proceed in basic conditions rather than in
rsc.li/njc
conventional acidic conditions.
method and enzymatic method supplied an efficient method for
bio-vanillin production.10–13 The biosynthesis method satisfied
1 Introduction
As the major flavour component of vanilla beans, vanillin consumers’ desires for natural products and produced no
(3-methoxy-4-hydroxybenzaldehyde) is widely used in the food by-product pollution for the environment. However, these meth-
industry and perfumery due to its attractive aroma and favourable ods have many disadvantages such as strict conditions for the
antioxidative effects.1–3 Thanks to its antimicrobial properties, microorganism’s requirements, low yield, difficulties in produc-
vanillin is also used in the pharmaceutical industry as an impor- tion on a large-scale and so on. Simultaneously, using the
tant medical intermediate. Therefore, with the expansion of the biosynthesis method to produce vanillin results in a relatively
application fields, the annual worldwide demand for vanillin has high price. Conversely, chemically synthesized vanillin has the
increased sharply.4,5
advantages of a wide range of raw materials, a simple synthesis
At present, there are three sources for commercial vanillin: process and being easily industrialized, which caused commercial
natural extraction, microbiological synthesis and chemical vanillin to have a very cheap price compared with natural vanillin.
synthesis.6 Undoubtedly, the annual production of natural The nitrification method, guaiacol–glyoxylic acid method, lignin/
vanillin obviously cannot meet the annual global demand syringaldehyde catalytic oxidation method, and organic electro-
(about 1.5 million tons per year) due to the limited annual chemical synthesis method are the typical chemical synthesis
vanilla bean production and the vanillin content in the vanilla methods for vanillin production.14–18 However, the nitrification
bean.7–9 The microbial fermentation method, plant cell culture method and catalytic oxidation method face serious economic and
environmental challenges due to high pollution, long processes
a School of Chemical and Environmental Engineering, Shanghai Institute of
Technology, 100 Haiquan Road, 201418, Shanghai, China
E-mail: hu-xj@mail.tsinghua.edu.cn, jiboliu@sit.edu.cn
and so on. Therefore, the guaiacol–glyoxylic acid method for
vanillin production, which employs guaiacol and glyoxylic acid
as raw materials was the most competitive method among the
chemical methods due to the properties of low energy consump-
tion, simple operation, environmentally friendliness and so on.
Generally, the guaiacol–glyoxylic acid method for vanillin
production is divided into three processes.19 Firstly, the VMA is
b School of Perfume and Aroma Technology, Shanghai Institute of Technology,
100 Haiquan Road, 201418, Shanghai, China
† Electronic supplementary information (ESI) available: HPLC spectra, IR spectra,
1H NMR spectra and the stability of vanillin at different pH. See DOI: 10.1039/
d0nj04798h
982 | New J. Chem., 2021, 45, 982--992
This journal is ©The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2021

View Article Online
Paper
NJC
produced under alkaline conditions (pH = 10–11) from guaiacol ferrite (CuFe2O4/SiO2) component catalyst. Compared with the
and glyoxylic acid condensation. Then, as an intermediate 2.5 eq. (vs. VMA) chemical oxidizing agent requirement in
product in the reaction, VMA is used directly without separa- vanillin industrial production, the catalytic system just needed
tion from the reaction solution to produce 3-methoxy-4-hydroxy 0.1 eq. (vs. VMA) CuFe2O4/SiO2 catalyst to arrive at comparative
acetophenone under the oxidation of CoCl2, bismuth (Bi2+ and conversion and selectivity. In addition, to explore the potential
Bi3+), reducible metal salts, copper hydroxide, copper sulphate for industrial applications of the catalytic system, we also
and so on.20,21,41–43 Finally, 3-methoxy-4-hydroxy acetophenone investigated the effects of pH value, oxygen partial pressure,
is decarboxylated to give vanillin by adjusting the pH value to reaction temperature and reaction time for the yield and
3–4 using sulfuric acid. Undoubtedly, the oxidation of VMA is selectivity of vanillin. Simultaneously, a probable mechanism
the key process to influence the vanillin yield. In industrial for vanillin generation was proposed under in situ ATR-FTIR
production, in which VMA oxidation is catalysed by 2–3 eq. and 1H NMR research. It provides theoretical guidance for
CuO, there is the difficulty of catalyst recovery, regeneration vanillin catalytic production.
and the high-cost materials of construction.22 Meanwhile, a
yield of vanillin of only 88.3% is attributed to the low selectivity
for vanillin under the catalysis of CuO, although the conversion
of VMA reached nearly 100%. It is of great significance to
2. Experimental
2.1 Reagents and instruments
improve clean manufacturing by using environmentally
All materials were of AR grade. Guaiacol, glyoxylic acid (50%
friendly catalysts to replace the heavily consumed CuO. The
aqueous solution), tetraethyl orthosilicate (TEOS), absolute
by-product in manufacturing was one of the main reasons for
ethanol (EtOH, 99.5%), copper sulphate pentahydrate(CuSO4Á
environmental pollution especially for some fine chemicals
production. Thus, developing a highly efficient oxidation
method is pivotal for VMA oxidation. Many researchers have
5H2O, 99.5%), iron chloride hexahydrate(FeCl3Á6H2O, 99.5%),
citric acid monohydrate (C6H8O7ÁH2O, 99%), ammonium
hydroxide (NH3ÁH2O), sodium hydroxide (NaOH, 99%), hexadecyl
developed strategies for VMA oxidation such as Cu–Al hydrotalcite
catalyst, copper(II) complexes and so on.23,24
trimethyl ammonium bromide (CTMAB), sulfuric acid (H2SO4,
99%), tetraethyl silicate (99%), acetonitrile, ethyl acetate (99%),
However, many metal complex methods face difficulties in
and oxygen (99.99%) were purchased from Sinopharm Chemical
preparation and problems due to the inferior stability of the
Reagent Co., Ltd, China. All the mentioned chemicals were
complexes. So, metal oxides, especially spinel transition metal
directly used without additional purification.
oxides with the AB2O4 general type, such as CuFe2O4, MgAl2O4
and CoFe2O4 were widely used in super conductors, the magnetic
core, humidity sensors and catalysis fields due to their possessing
2.2 Experimental method
unique properties such as electronic, magnetic and optical
2.2.1 Preparation of the catalyst CuFe2O4/SiO2. The copre-
properties.25–27 In the field of catalytic oxidation, CuFe2O4 has cipitation method was implemented in the preparation of
been widely used as a high efficiency catalyst because of its CuFe2O4.33,34 In a typical experiment, the procedure was as
excellent catalytic activity, stable chemical properties and follows: CuSO4Á5H2O (14.98 g, 0.06 mol) and FeSO4Á6H2O
inexpensiveness.28–30 Due to the Jahn–Teller effect, the two crystal (32.44 g, 0.12 mol) were dissolved into 300 mL of water by
forms of CuFe2O4 (tetragonal and cubic), which are routed from stirring. C6H8O7ÁH2O (37.83 g, 0.18 mol) was added into the
the preparation and precursors have a pivotal role in the catalytic mixture to dissolve the copper salt and iron salt completely.
activities.31 Traditionally, CuFe2O4 nanoparticles catalysts can Ammonia water was added to adjust the pH value to 7. Then,
easily agglomerate during the calcination process. The agglome- the mixture was stirred at 100 1C for 3 h to obtain a colloid
ration of the catalyst caused much CuFe2O4 cladding in the liquid. The colloid liquid was filtered under vacuum to give a
particles, which further decreases the catalytic activity. The bluish grey solid. The solid was calcined in a muffle furnace at
agglomeration of the catalyst prevented the internal group contact 600 1C for 3 h. After cooling to room temperature, CuFe2O4
with the substrate of the reaction which further reduced the nanoparticles were collected with magnets. The freshly
catalytic efficiency.32 Employing a porous substrate to support prepared CuFe2O4 nanoparticles were added to a 500 mL flask,
the CuFe2O4 catalyst efficiently and thus further enhance the followed by the addition of water (50 mL), ethanol (300 mL),
dispersity of the catalyst is a useful strategy for increasing the CTMAB (10.0 g) and TEOS (35.0 g). After the addition of
catalytic activity. The nano-SiO2 substrate, as a typical porous ammonia water (100 mL) at 45 1C dropwise, the reaction mixture
material was introduced to increase the dispersity of CuFe2O4 was stirred for 6 h. The mixture was filtered after cooling to room
because of its huge specific surface area, good stability, and temperature. The solid was washed with water and dried in
inexpensive properties. Simultaneously, the porous SiO2 was also vacuum at 45 1C. Finally, the solid was calcined at 600 1C for 5 h
helpful in enhancing the concentration of the reacting substrate in a muffle furnace to obtain the CuFe2O4/SiO2 nanoparticles
due to the adsorption.
which was used for VMA oxidation directly.
Hence, based on the persistent researching of vanillin
production, a green catalytic system was designed to convert
2.2.2 Oxidation of VMA by CuFe2O4/SiO2
Preparation of VMA. Guaiacol (62.07 g, 0.5 mol) and glyoxylic
VMA to vanillin in aqueous solution under a low oxygen atmo- acid (62.19 g, 0.42 mol) were dissolved in deionized water (300 mL)
sphere by employing a high-efficiency silica supported copper under continuous stirring. The mixture was alkalized to pH = 11.5
This journal is ©The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2021
New J. Chem., 2021, 45, 982--992 | 983

View Article Online
NJC
Paper
1H NMR spectra were obtained on a Bruker AVANCE III spectro-
meter operating at 500 MHz. The Raman spectra were recorded
on Thermo DXR using a DPSS laser. In situ ATR-FTIR was
measured on a Mettler Toledo Reactir-15. The H2-TPR was
performed by passing 10 vol% H2/N2 over 50 mg of catalyst at
a heating rate of 5 1C minÀ1. Prior to the measurement, the
sample was pre-treated under argon atmosphere at 200 1C for
1 h, then the system was cooled to ambient temperature by
purging with pure argon gas.
Scheme 1 A schematic diagram of the synthesis of vanillin using glyoxylic
acid.
using aqueous NaOH (1 M) and then kept at 20 1C for 24 h. After
acidifying to pH = 4–5 with 1 M H2SO4, the excess guaiacol was
recovered by extracting with EtOAc (150 mL Â 2). The aqueous
solution was used for VMA oxidation directly without any further
purification.
3. Result discussion
3.1 Characterization of the prepared CuFe2O4/SiO2 composite
catalyst
Scheme 1 and Scheme S1 (ESI†) show the vanillin prepara-
tion and the by-product generation, respectively. Freshly prepared Field emission scanning electron microscopy (FESEM) was used
catalyst (CuFe2O4/SiO2 1 g) was added to an aqueous solution of to investigate the morphology of the CuFe2O4/SiO2 composite
VMA (100 mL, 10%). The mixture was heated to 90 1C under catalyst. As shown in Fig. 1a, the morphology of the CuFe2O4
constant stirring after the pH value was adjusted to 11 using particles was an irregular powder with a size of 100–200 nm,
aqueous NaOH (30%). Then, the mixture was bubbled with O2 which provides excellent conditions for the O2 and VMA adsorp-
and stirred for 5 h. The high-performance liquid chromatography tion. To further exhibit the dispersion of the CuFe2O4 nano-
(HPLC) spectra of the reaction mixture before and after oxidation particles on the SiO2 powders, transmission electron microscopy
are shown in Fig. S1 (ESI†). After cooling down, the mixture was (TEM) was employed to investigate the fine structure of the
acidified using H2SO4 10% to pH = 4 and extracted with toluene CuFe2O4/SiO2 component catalyst. Fig. 1b indicates the CuFe2O4
(60 mL) 3 times. The combined organic layers were dried over nanoparticles with the size of 20–40 nm deposited on the surface
MgSO4, filtered, concentrated and purified by silica-gel chromato- of SiO2 uniformly. Further details concerning CuFe2O4/SiO2 were
graphy to give the purified vanillin. The IR and 1H NMR spectra of observed by HR-TEM. As shown in the HR-TEM image, average
vanillin are shown in Fig. S2 and S3 (ESI†).
d-spacing values of 0.25 nm were observed clearly, which corre-
2.2.3 Materials characterization. Scanning electron micro- sponded to the distance of the (311) crystal planes of CuFe2O4.
scopy (SEM) was carried out on a Hitachi S-4800 transmission The equal dispersion of CuFe2O4 nanoparticles on a SiO2 porous
electron microscope. Transmission electron microscopy (TEM) substrate was beneficial to retain the activity of the CuFe2O4
was carried out on a Technai G2 20 TWIN. The X-ray powder nanoparticles for VMA oxidation by restraining its conglomera-
diffraction (XRD) pattern was performed on a d/max-2200pc tion. To better determine the CuFe2O4 loading on SiO2 obtained,
XRD diffractometer. The X-ray photoelectron spectroscopy the CuFe2O4/SiO2 was characterized by energy-dispersive spectro-
(XPS) experiments were tested on a Thermo Escalab 250Xi scopy (EDS). As shown in Fig. S4 (ESI†), the CuFe2O4 loading
system using Al-Ka radiation (hn = 1486.6 eV). The Brunauer– amount was 12.5%. To further examine the chemical elements
Emmett–Teller (BET) specific surface areas of typical products and states of the designed catalyst, XPS analysis was applied to
were performed at 77 K in a Micromeritics ASAP 2020 system. investigate the concentration and binding energy of the surface
Fig. 1 (a) A field emission scanning electron microscopy (FESEM) image of CuFe2O4/SiO2 and (b) a transmission electron microscopy (TEM) image of
CuFe2O4/SiO2.
984 | New J. Chem., 2021, 45, 982--992
This journal is ©The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2021

View Article Online
Paper
NJC
Fig. 2 (a) X-ray photoelectron spectroscopy (XPS) of CuFe2O4/SiO2 and (b–d) high-resolution XPS spectra of Cu 2p, Fe 2p and O 1s.
atoms present in the CuFe2O4/SiO2 component catalyst. The C 1s binding energy. The peaks centered at 529.4 eV, 531.5 eV and
peak at 284.6 eV in the environment was used as a correction. 533.4 eV belong to the metal oxides (O–Cu–O and O–Fe–O),
In the survey spectrum in Fig. 2a, the elements of Si, Cu, Fe, and surface hydroxyl groups (OH) and SiO2, respectively.38
O can be observed clearly. The binding energy values of Fe 2p
The crystal structure of the catalyst was examined using
and Cu 2p for the sample were detected at 710.5 and 936.7 eV, X-ray diffraction pattern (XRD). As seen in Fig. 3a, the peaks
respectively. The obtained results obviously implied that both at 18.30, 29.94, 35.4, 36.94, 43.28, 53.4, 57.0 and 62.51 can
iron ions and copper ions are present in CuFe2O4/SiO2 powders. be ascribed to the reflection of (111), (220), (311), (222), (400),
Fig. 2b showed the XPS spectra of Cu 2p in freshly prepared (422), (511) and (440) CuFe2O4. In the XRD spectrum
CuFe2O4 and CuFe2O4/SiO2. The Cu 2p spectrum of fresh of CuFe2O4/SiO2, the position and relative intensity of the
CuFe2O4 and CuFe2O4/SiO2 overtly shows four peaks at 634.5 eV diffraction peaks were in accordance with those of CuFe2O4
for Cu 2p3/2, 654.5 eV for Cu 2p1/2, 942.1 eV for the Cu 2p3/2 no peaks of the other compounds were observed. In the
satellite feature and 962.1 eV for the Cu 2p1/2 satellite feature, CuFe2O4/SiO2 nanoparticle sample, there will be a broad peak
which indicates that the catalyst contained only Cu(II).35,36 Gen- at 221, which indicates that the crystallinity of SiO2 was
erally, as for the Fe 2p core level peak, the XPS spectra of the Fe reduced.39–41
2p regions can be fitted into three contributions (FeO, Fe2O3 and
As a porous material, SiO2 supplied an efficient adsorption
Fe3O4). XPS spectra of Fe 2p core level is resolved into four peaks site for the oxide reaction. Thus, SiO2 supported CuFe2O4 can
at 709.6, 710.9, and 712.7 eV for Fe 2p3/2, 724.4 eV for Fe 2p1/2 effectively increase the specific surface area of the catalyst and
and a satellite signal at 717.5 eV. For Fe 2p3/2, the peak at the contact area during the catalyst and the reactants, thus
709.6 eV is indexed to Fe(II), whereas the peaks at 710.9 and further improving the catalytic efficiency of the catalyst. As
712.7 eV are indexed to Fe(III).37 From the comparison of shown in Fig. 4, the surface area has dramatically increased
CuFe2O4 and CuFe2O4/SiO2 in the XPS spectrum, the valence after constructing CuFe2O4 nanoparticles on the surface of
state of the metal elements had no obvious changes, which SiO2. Additionally, with the CuFe2O4 loading on SiO2 the sur-
indicates that the CuFe2O4 is well maintained before and after face area was progressively increased from 5.017 to 49.98 m2 gÀ1
.
loading on SiO2. In addition, as shown in Fig. 2d, the shape of The large specific surface area of CuFe2O4/SiO2 can offer more
the wide and asymmetric peak of the O 1s spectrum indicates adsorption and reaction sites, consequently leading to enhanced
that there can be more than one chemical state according to the catalytic activity for VMA conversion.
This journal is ©The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2021
New J. Chem., 2021, 45, 982--992 | 985

View Article Online
NJC
Paper
Fig. 3 (a) X-ray diffraction (XRD) of CuFe2O4/SiO2 and CuFe2O4 and (b) X-ray diffraction (XRD) of CuFe2O4/SiO2 before and after using for 5 cycles.
Table 1 A comparison of the catalytic effect of CuFe2O4/SiO2 with
different catalytical systems
VMA conversion Vanillin selectivity Vanillin yield
Catalyst
(%)
(%)
(%)
CuFe2O4
CuFe2O4/SiO2
CuSO4
83.7
98.1
100
100
—
97
96.7
96.5
81.1
88.3
—
87
85
80.9
94.7
88.1
88.3
56
84.4
85
CuO22
42
Cu(OH)2
Bi(0)43
44
CoCl2
100
The reaction conditions as follows: reaction temperature: 90 1C,
O2 pressure: 0.1 MPa, pH value: 11.0, reaction time: 3 h, catalyst ratio:
10% vs. the amount of VMA.
further indicated the high efficiency of CuFe2O4/SiO2. The
selectivity for vanillin of the two catalysts was 96.7% and
96.5%, respectively, which indicated that the SiO2 substrate
had no obvious restrain for vanillin selectivity. As the optimal
process in the vanillin industry at present, CuO showed 100%
conversion VMA and 88.3% selectivity for vanillin.21 However, it
was noted that the CuO requirement was more than 2.5 eq.
(vs. VMA) to ensure VMA conversion. Such a CuO amount used
Fig. 4 The multilayer adsorption–desorption curve (BET) diagram of CuFe2O4/
SiO2 and CuFe2O4.
3.2 The CuFe2O4/SiO2 promoted highly efficient vanillin
production
Due to the synergistic effect of the CuFe2O4 nanoparticles and caused the difficulties of vanillin purification, the high cost of
SiO2 substrate, the CuFe2O4/SiO2 component catalyst showed oxidation agent regeneration, the waste of resources and so on.
excellent activity for VMA conversion. As exhibited in Fig. S1
The pH value of reaction mixture was investigated without
(ESI†), only vanillin was detected by HPLC after 3 hours of any changes made to the other conditions. The pH value of the
stirring, which indicated that the catalyst possesses excellent reaction system was about 4–5 after the recycling of the
selectivity for vanillin. Simultaneously, to evaluate the applica- redundant guaiacol. The influence of the pH value changing
tion prospects of the catalyst, various parameters such as the from 4 to 12.5 was investigated. As shown in Fig. 5a, after the
effect of SiO2 substrate, pH value, catalyst loading and so on for pH value was adjusted to 8, the yield of vanillin raised sharply.
vanillin production and VMA conversion were investigated. As a Following the constant rising of the pH value, the yield of
porous substrate, the SiO2 supplied a useful substrate for O2 vanillin increased accordingly. The vanillin yield reached 93.2%
and VMA adsorption, which further promoted the conversion of at a pH value of 11.0. However, the yield and conversion of
VMA. As shown in Table 1, the CuFe2O4 catalyst without vanillin declined as the pH value continued to increase. The
loading on the SiO2 substrate showed 83.7% VMA conversion, reason might be attributed to the poor stability of vanillin in
while the value in the CuFe2O4/SiO2 system was 98.1%, which the high alkali environment. Thus, the stability of vanillin
986 | New J. Chem., 2021, 45, 982--992
This journal is ©The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2021

View Article Online
Paper
NJC
Fig. 5 (a) The effect of pH on the yield and conversion of vanillin; (b) the effect of time on the yield and conversion of vanillin; (c) the effect of
temperature on the yield and conversion of vanillin; (d) the effect of pressure on the yield and conversion of vanillin; (e) the effect of CuFe2O4 loading
amount on the yield and conversion of vanillin; and (f) the effect of catalyst amount on the yield and selectivity of vanillin.
under different pH values was introduced to verify the assump- As shown in Fig. 5c, the conversion of vanillin reached 57.2%
tion. As shown in Fig. S5 (ESI†), the vanillin content was at 70 1C and the yield of vanillin was only 56.1% due to the low
decreased persistently following the pH value increase. Espe- conversion of VMA. When the temperature exceeded 90 1C, the
cially, the vanillin content showed a sharp decrease as the pH vanillin produced in the reaction system was converted into
value exceeded 12. Because of the activity of the aldehyde vanillic acid, which reduced the selectivity of vanillin. The
compound between alcohol and acid, the reaction time and results showed that the highest yield of vanillin was obtained
temperature will be important factors for the yield and selec- at 90 1C with the values of the yield and the conversion being
tivity of vanillin during industrial processes in the future. 92.9% and 97.8%, respectively.
Therefore, the effect of reaction time for vanillin production
Due to vanillin being a widely used food additive added to
was investigated with the other conditions unchanged. Experi- food directly, developing a suitable oxidant is vital to convert
mental results indicated that the conversion of VMA was VMA to vanillin. Oxygen is preferred because of its inherently
decreased following the reaction time and temperature increas- inexpensive, clean and nontoxic properties. Undoubtedly, a
ing (43 h, 495 1C). Simultaneously, the vanillin production higher O2 pressure has greater safety risks for the oxidation
increased persistently following the reaction time and tempera- reaction. Meanwhile, vanillin will be easily oxidized to vanillic
ture increase (o3 h, o95 1C). As shown in Fig. 5b, the yields for acid in the VMA oxidation process under a high oxygen pressure.
vanillin and conversion for VMA were 66.7% and 69.6%, The effect of oxygen partial pressure during vanillin production
respectively, after stirring at 90 1C for 1 h. As the reaction was investigated to determine the suitable O2 pressure with the
lasted for 3 h, the yield of vanillin reached 93.1% with the other conditions unchanged. As shown in Fig. 5d, the VMA
conversion of VMA at 98.3%. It indicated that the VMA conver- conversion and vanillin selectivity decreased rapidly following
sion increased continuously with high selectivity following the the increasing of O2 pressure. As the oxygen pressure increased
reaction time in the first 3 h. As the reaction lasted for 5 h, the to 0.2 MPa from 0.1 MPa, the yield of vanillin decreased to
vanillin yield decreased to only 53.6% with the VMA conversion 77.5% from 92.9% with the conversion increasing to 97.1%
remaining at 98.4%. The dominant by-product of the oxidation from 98.4%. The primary by-product was vanillic acid which
was vanillic acid which was produced by over oxidation of was produced by the high O2 pressure. Following reduction
vanillin. HPLC also indicated that vanillic acid increased to 43% under 0.5 MPa O2 pressure. As a result, the oxygen
obviously as the vanillin deceased. As the overall yield of partial pressure plays a decisive role in the vanillin generation.
vanillin decreasing was attributed to the selectivity of vanillin Exorbitant O2 pressure caused the generation of vanillic acid
reducing after 3 h, the reaction should be terminated after 3 h. which decreased the yield of vanillin. The loading amount of
The effect of reaction temperature on the vanillin production CuFe2O4 in the catalyst of CuFe2O4/SiO2 was one of the vital
was investigated with the other conditions kept the same. factors affecting the yield of vanillin. Therefore, the catalytic
This journal is ©The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2021
New J. Chem., 2021, 45, 982--992 | 987

View Article Online
NJC
Paper
effect of the catalyst under different CuFe2O4 loading amounts 3.4 Probable mechanism for vanillin production
was investigated with the other conditions unchanged.
As shown in Fig. 5e, VMA conversion increased from 47.1% to
96.4% following the CuFe2O4 loading amount added increasing
In order to explain the mechanism of VMA oxidation, the
reaction was investigated by employing in situ ATR-FTIR. The
IR spectra of vanillin and VMA in water were collected from a
from 2.5% to 12.5%. Simultaneously, the yield of vanillin
series of undersaturated solutions with known solute concen-
increased from 45.2% to 92.5%. Furthermore, the yield for
trations. As the ability of CuFe2O4/SiO2 to function as a VMA
vanillin had no further changes as the CuFe2O4 loading amount
oxidation catalyst was established, the IR spectrum was
over 12.5%. The ratio (vs. VMA) of CuFe2O4/SiO2 used signifi-
employed to detect any of the likely intermediates by examining
cantly affects the catalytic efficiency of the reaction. Thus, the
the evolution of the wavenumber in the range of 300–900 cmÀ1
influence of CuFe2O4/SiO2 ratio (vs. VMA) using for VMA con-
as a function of reaction time.45–47 To exhibit the reaction
version and vanillin production was investigated with the other
clearly, the differential absorption (DA) data was collected and
conditions unchanged (the reaction mixture was stirred at
processed as shown in Fig. 7a. Peaks pointing down indicate
90 1C for 3 h under the oxygen partial pressure of 0.1 MPa).
bands disappearing while those pointing up correspond to new
Fig. 5f exhibited the VMA conversion and vanillin selectivity
bands appearing due to the catalytic reaction. The increased
following the CuFe2O4/SiO2 ratio varying from 0% to 20%.
intensity of the DA value at 1633 cmÀ1 vested to the band of
CQO stretching vibration, which was attributed to the genera-
As the catalyst ratio increased to 10% from 0, the catalytic
efficiency of the reaction increased from 9.8% to 93.4% accord-
ingly. As the catalyst ratio further increased over 10%, the
catalytic efficiency showed no obvious changes, and the oxida-
tion of intermediate ketonic acid by oxidation of benzyl alcohol
in VMA.48 At the same time, the increased peaks of DA at 1305
and 1107 cmÀ1 belong to the C–C stretching vibration and
tion efficiency stabilized at about 93.5%. A recycling experiment
conjugation of aromatic ketones following the intermediate
of CuFe2O4/SiO2 for VMA oxidation was investigated to examine
ketonic acid generation. The peak at 786 cmÀ1 might be due
the stability of the catalyst, thus further evaluating the indus-
to the C–C stretching vibration. The decline of the peaks could
trial production prospects. Thanks to the synergistic effect of
be attributed to the fracture of the C–C bond caused by the
Cu(II)/Cu(I) and Fe(III)/Fe(II), the catalyst was equipped with
excellent catalytic stability for VMA oxidation. The VMA con-
version, vanillin yield and selectivity corresponding to the
decarboxylation of intermediate ketonic acid.49–51 The further
evolution of the DA spectra for VMA oxidation is seen more
clearly in Fig. 7b, in which the peaks at 1633 cmÀ1 (intermediate
CuFe2O4/SiO2 reusing times are summarized in Fig. 6a. Further-
ketonic acid), 1305 and 1107 cmÀ1 (aromatic ketones) and
more, the IR spectrum showed that the combined catalyst was
stable after cycling for five times. Simultaneously, as shown in
786 cmÀ1 (decarboxylation) were plotted versus reaction time.
Therefore, as indicated in Scheme 2, a probable reaction pathway
was proposed based on the in situ ATR-FTIR result.52,53 Thanks
to the alkaline conditions and oxygen vacancies, VMA was
adsorbed on the surface of the catalyst to ensure the efficient
contact between VMA and the catalyst. Then, the Cu(II) ion in the
catalyst was converted to Cu(I) as soon as the electrons of VMA
transferred to the catalyst, accomplishing the pivotal inter-
mediate (a-keto carboxylate radical) formation. Vanillin was
the XRD spectrum of CuFe2O4/SiO2 after using it for five cycles
(Fig. 3b), all the typical diffraction peaks were observed on the
used catalyst, indicating the high stability of the catalyst.
Furthermore, trace amounts of CuO were also observed on
the XRD spectrum of the used CuFe2O4/SiO2, indicating the
partial decomposition of CuFe2O4/SiO2. However, the partial
decomposition of CuFe2O4/SiO2 had no obvious impact on the
catalytic activity compared with the fresh CuFe2O4/SiO2.
Fig. 6 (a) The stability of CuFe2O4/SiO2 under 5 times recycling; and (b) the IR spectrum of CuFe2O4/SiO2 after being recycled five times.
988 | New J. Chem., 2021, 45, 982--992
This journal is ©The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2021

View Article Online
Paper
NJC
Fig. 7 (a) The oxidation of VMA under in situ ATR-FTIR. (b) Absorbance versus time for selected wavenumber under in situ ATR-FTIR for VMA oxidation.
the lower temperature region direction, demonstrating the
synergy effect in CuFe2O4. The reduction sites of CuFe2O4/
SiO2 indicated that CuO was reduced first, hence the Cu ions
exhibited more active oxidizing properties when oxidizing VMA
by using CuFe2O4/SiO2.57,58 XPS was employed to investigate
the element valence change of the catalyst before and after
being used once. The XPS spectra indicated that the surface of
CuFe2O4/SiO2 was equipped with a high density content of
Cu(II) and Fe(III), which further enhanced the catalytic activity
by the convenient charge transfer between Cu(II)/Cu(I) and
Fe(III)/Fe(II). As shown in Fig. 8(a), two new peaks appeared at
933.2 eV and 936.8 eV indexed to Cu(I) and Cu(II) in the used
CuFe2O4/SiO2. The generation of Cu(I) indicated the catalytic
performance of the CuFe2O4/SiO2 catalyst.59–62 The generated
Cu(I) was re-oxidized to Cu(II) under the presence of Fe(III). As in
Fig. 8(b), similar with the fresh prepared CuFe2O4, the XPS
spectrum for Fe 2p after being used 5 times was also resolved
into five peaks at 709.6, 710.9, 712.7, 717.5 and 724 eV,
corresponding to Fe(II) and Fe(III). In addition, a change in
Scheme 2 A probable mechanism for oxidative decarboxylation of VMA
the Fe 2p spectrum in the fresh and spent catalysts was found,
and two peaks at 709.6 eV and 712.7 eV could be identified as
Fe(II) and Fe(III), respectively. The changes were attributed to the
conversion of Fe(III) to Fe(II) accompanied by the change of Cu(I)
to Cu(II). Simultaneously, little Cu(II) existed in the form of CuO
which was in accordance with the peaks in the XRD at 38.8 and
48.71 (as shown in Fig. 3b).63,64 Small amounts of CuO formed on
the surface of the catalyst after being used 5 times but the main
component showed no obvious changes. As shown in Fig. 8(c),
the XPS spectra of O 1s indicated that the valence state of O
before and after use showed no significant change.
Furthermore, it was interesting to note that the decarboxyla-
tion proceeded under alkaline conditions rather than under
acidic conditions. The traditional view for vanillin production
was the decarboxylation of the a-keto carboxylate radical under
acidic conditions. As soon as the reaction completed, the
1H NMR spectrum of the reaction mixture indicated that the
aldehyde group appeared at the chemical shift of 9.25 ppm
over CuFe2O4/SiO2.
generated following the decarboxylation of the pivotal inter-
mediate. Simultaneously, the catalyst was reborn under the
presence of O2.
To further investigate the oxidation process, H2-temperature
programmed reduction (H2-TPR) and XPS were employed to
reveal the reduction states of the CuFe2O4/SiO2 catalyst. As a
combined redox catalyst, CuFe2O4 was the catalytic centre
which was surrounded by SiO2 substrate. After calcination at
600 1C for 5 hours, the redox sites formed due to the oxygen
vacancies. The synergistic effects of CuFe2O4, CuO and Fe2O3
on the reduction behaviours were widely accepted.54–56 As
shown in Fig. S6 (ESI†), the pure CuO had a strong peak at
approximately 350 1C, corresponding to the reduction of Cu(II)
to Cu(0). Compared with CuO and Fe2O3, the reduction site in
the CuFe2O4/SiO2 samples could be clearly discovered to shift to
This journal is ©The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2021
New J. Chem., 2021, 45, 982--992 | 989

View Article Online
NJC
Paper
Fig. 9 The aldehyde group H in 1H NMR spectrum of the reaction mixture
compared with that of vanillin in different solutions and pH values.
4. Conclusions
In summary, a SiO2 supported CuFe2O4 composite catalyst was
designed and prepared for vanillin high efficiency generation
under low pressure in aqueous solution. The poly-porous
structure of the composite catalyst was beneficial for O2 VMA
adsorption which further causes the VMA oxidation at low O2
pressure (0.1 MPa). From the study on the condensation reac-
tion, the results show that the conversion of VMA and the yield
of vanillin were affected by the pH value, reaction time, catalyst
amount, partial pressure of oxygen, and reaction temperature.
Under the optimum conditions, the selectivity for vanillin and
the yield of vanillin can reached 98% and 93%. The catalytic
oxidation process used CuFe2O4/SiO2 catalyst to avoid the
use of large amounts of CuO, and the catalyst can be easily
recovered during magnetism of CuFe2O4/SiO2. More impor-
tantly, the probable mechanism of the CuFe2O4/SiO2 catalysed
VMA oxidation was introduced by in situ ATR-FTIR, H2-TPR
and NMR. It provided new insights for the green oxidation of
vanillyl mandelic acid.
Fig. 8 High-resolution Cu 2p (a), Fe 2p (b), O 1s (c) XPS peaks of
CuFe2O4/SiO2 before and after.
Conflicts of interest
There are no conflicts to declare.
1
(as shown in Fig. 9 and Fig. S7, ESI†). The H NMR spectrum
of the mixture compared with that of vanillin in different
solutions and pH values further indicated the H at 9.25 ppm,
which is attributed to the aldehyde group generated from the
decarboxylation under basic conditions. As a result, vanillin
Acknowledgements
was finally generated after decarboxylation of the a-keto carboxylate The authors are grateful to the National Natural Science Foun-
radical in basic conditions rather than acidic conditions. dation of China (NSFC, Project No. 21906105), the Shanghai
Furthermore, the loading of CuFe2O4 on the SiO2 substrate Gaofeng & Gaoyuan Project for University Academic Program
provided efficient adsorption sites for oxygen molecules and Development and the Scientific Research Foundation of Shanghai
favourable conditions for decarboxylation due to the large Institute of Technology (39120K186028-YJ2018-7) for financial
surface area.32
support.
990 | New J. Chem., 2021, 45, 982--992
This journal is ©The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2021

View Article Online
Paper
NJC
28 R. Parella and N. S. A. Babu, Catal. Commun., 2012, 29,
118–121.
29 P. Koley, S. C. Shit, B. Joseph, S. Pollastri, Y. M. Sabri,
E. L. H. Mayes, L. Nakka, J. Tardio and J. Mondal, ACS Appl.
Mater. Interfaces, 2020, 12, 21682–21700.
30 Q. Z. Wang, J. J. He, Y. B. Shi, S. L. Zhang, T. J. Niu,
H. D. She, Y. P. Bi and Z. Q. Lei, Appl. Catal., B, 2017, 214,
158–167.
Notes and references
1 J. H. Hu, Y. F. Hu, J. Y. Mao, J. Yao, Z. R. Chen and H. R. Li,
Green Chem., 2012, 14, 2894–2898.
2 W. Fu, L. M. Yue, X. G. Duan, J. Li and G. Z. Lu, Green Chem.,
2016, 18, 6136–6142.
3 M. Fache, B. Boutevin and S. Caillol, ACS Sustainable Chem.
Eng., 2016, 4, 35–46.
4 A. R. R. P. Almeida, V. L. S. Freitas, J. I. S. Campos,
M. D. M. C. Ribeiro da Silva and M. J. S. Monte, J. Chem.
Thermodyn., 2019, 128, 45–54.
5 R. S. Fernandes, M. Dinc, I. M. Raimundi Jr and
B. Mizaikoff, Anal. Methods, 2017, 9, 2883–2889.
6 G. Banerjee and P. Chattopadhyay, J. Sci. Food Agric., 2019,
99, 499–506.
7 N. J. Gallage and B. L. Møller, Mol. Plant, 2015, 8, 40–57.
8 A. J. J. Straathof, Chem. Rev., 2014, 114, 1871–1908.
9 T. X. Yang, L. Q. Zhao, J. Wang, L. S. Guo, H. M. Liu,
H. Cheng and Z. Yang, ACS Sustainable Chem. Eng., 2017, 5,
5713–5722.
31 O. S. G. V. M. Barbu and P. B. M. Stefanescu, J. Therm. Anal.
Calorim., 2013, 113, 1245–1253.
32 F. H. Shi, H. R. Shan, D. Li, X. Yin, J. Y. Yu and B. Ding,
J. Colloid Interface Sci., 2019, 538, 620–629.
33 J. Feng, L. Su, Y. H. Ma, C. L. Ren, Q. Guo and X. G. Chen,
Chem. Eng. J., 2013, 221, 16–24.
34 Z. P. Sun, L. Liu, D. Z. Jia and W. Y. Pan, Sens. Actuators, B,
2007, 125, 144–148.
35 Y. l. Guo, L. l. Zhang, X. Y. Liu, B. Li, D. Tang, W. S. Liu and
W. W. Qin, J. Mater. Chem. A, 2016, 4, 4044–4055.
36 N. Gutta, V. K. Velisoju, J. Tardio, J. Patel, L. Satyanarayana,
A. V. S. Sarma and V. Akula, Energy Fuels, 2019, 33,
12656–12665.
37 S. Chen, J. Deng, C. Ye, C. Xu, L. Huai, J. Li and X. Li,
Sci. Total Environ., 2020, 742, DOI: 10.1016/j.scitotenv.2020.
140587.
38 M. A. Taleb, R. Kumar, A. A. Al-Rashdi, M. K. Seliem and
M. A. Barakat, Arabian J. Chem., 2020, 13(10), 7533–7543,
DOI: 10.1016/j.arabjc.2020.08.028.
39 M. Y. Zhu, D. H. Meng, C. J. Wang and G. W. Diao, ACS Appl.
Mater. Interfaces, 2013, 5, 6030–6037.
´
10 F. R. Teran, I. P. Amador and A. L. Munguia, J. Agric. Food
Chem., 2001, 49, 5207–5209.
11 D. D. Gioia, L. Sciubba, M. Ruzzi, L. Settiet and F. Favaal,
J. Chem. Technol. Biotechnol., 2009, 84, 1441–1448.
12 R. H. H. Heuvel, M. W. Fraaije, C. Laane and W. J. H. Berkel,
J. Agric. Food Chem., 2001, 49, 2954–2958.
13 R. T. Winter, H. L. Van Beek and M. W. Fraaije, J. Chem.
Educ., 2011, 89, 258–261.
14 S. L. Bhanawase and G. D. Yadav, ACS Sustainable Chem.
Eng., 2016, 4, 1974–1984.
15 T. J. Sorensen, Chem. Eng. Technol., 2014, 37, 1959–1963.
16 M. P. Pandey and C. S. Kim, Chem. Eng. Technol., 2011, 34,
29–41.
40 K. M. R. Karim, M. Tarek, H. R. Ong, H. Abdullah, A. Yousuf,
C. K. Cheng and M. M. R. Khan, Ind. Eng. Chem. Res., 2019,
58, 563–572.
41 F. Wang, K. Wang, Y. Muhammad, Y. Wei, L. Shao and
X. Wang, ACS Sustainable Chem. Eng., 2019, 7, 14716–14726.
42 X. R. Shen, X. M. Yang, Y. Zhu, X. H. Liu and H. L. Wen,
J. Labelled Compd. Radiopharm., 2014, 57, 658–659.
17 A. D. Paola, M. Bellardita, B. Megna, F. Parrino and
L. Palmisano, Catal. Today, 2015, 252, 195–200.
18 A. Franco, J. F. de Souza, P. F. P. Nascimiento,
´
M. M. Pedroza, L. S. Carvalho, E. Rodriguez-Castellon and
˜
´
43 I. Favier, F. Giulieri, E. Dunach, D. Hebrault and J. Desmurs,
R. Luque, ACS Sustainable Chem. Eng., 2019, 7, 7519–7526.
19 D. F. Niu, H. C. Li and X. S. Zhang, Tetrahedron, 2013, 69,
8174–8177.
Eur. J. Org. Chem., 2002, 1984–1988.
˜
´
´
44 I. Favier, E. Dunach, D. Hebrault and D. Hebrault, New
J. Chem., 2004, 28, 62–66.
˜
20 I. Favier and E. Dunach, Tetrahedron, 2003, 59, 1823–1830.
45 G. Paul, C. Bisio, I. Braschi, M. Cossi, G. Gatti, E. Gianotti
and L. Marchese, Chem. Soc. Rev., 2018, 47, 5684–5739.
46 S. Supriya, S. Sushmitha and K. Srinivasan, Cryst. Growth
Des., 2019, 19, 6315–6323.
47 V. Kannan, M. Jayaprakasan, R. Bairava Ganesh and
P. Ramasamy, Phys. Status Solidi A, 2006, 203, 2488–2495.
48 N. Fujii, K. Fujimoto, T. Michinobu, M. Akada, J. P. Hill,
S. Shiratori, K. Ariga and K. Shigehara, Macromolecules,
2010, 43, 3947–3955.
49 S. Schmidt, M. Koldevitz, J. M. Noy and P. J. Roth, Polym.
Chem., 2015, 6, 4–54.
50 X. Sun, S. Liu, A. Khan, C. Zhao, C. Yan and T. Mu, New
J. Chem., 2014, 38, 3449–3456.
21 S. L. Bhanawase and G. D. Yadav, Ind. Eng. Chem. Res., 2017,
56, 12899–12908.
22 W. Fu, X. G. Duan, L. M. Yue, S. W. Dai, L. Xu, J. Li, G. Z. Lu
and D. S. Mao, React. Kinet., Mech. Catal., 2015, 116,
191–204.
23 W. F. Yu, X. G. Meng, X. Peng, X. H. Li and Y. Liu, J. Mol.
Catal. A: Chem., 2013, 379, 315–321.
24 J. A. Jiang, C. Chen, J. G. Huang, H. W. Liu, S. Cao and
Y. F. Ji, Green Chem., 2014, 16, 1248–1254.
25 Z. Jiang, W. H. Zhang, W. F. Shangguan, X. J. Wu and
Y. Teraoka, J. Phys. Chem. C, 2011, 115, 13035–13040.
26 Y. P. Liu, F. L. Formal, F. Boudoire, L. Yao, K. Sivula and
N. Guijarro, J. Mater. Chem. A, 2019, 7, 1669–1677.
27 Z. M. Pan, G. G. Zhang and X. C. Wang, Angew. Chem., Int.
Ed., 2019, 58, 7102–7106.
51 S. C. Shit, P. Koley, B. Joseph, C. Marini, L. Nakka, J. Tardio
and J. Mondal, ACS Appl. Mater. Interfaces, 2019, 11,
24140–24153.
This journal is ©The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2021
New J. Chem., 2021, 45, 982--992 | 991

View Article Online
NJC
Paper
52 D. Choi and D. J. Jang, New J. Chem., 2017, 41, 2964–2972. 58 R. Bonyasi, M. Gholinejad, F. Saadati and C. Najera, New
53 C. Sarkar, P. Koley, I. Shown, J. Lee, Y.-F. Liao, K. An, J. Chem., 2018, 42, 3078–3086.
J. Tardio, L. Nakka, K.-H. Chen and J. Mondal, ACS Sustainable 59 F. Cao, Z. Pan and X. Ji, New J. Chem., 2019, 43, 11342–11347.
Chem. Eng., 2019, 7, 10349–10362.
54 G. Perin, J. Fabro, M. Guiotto, Q. Xin, M. M. Natile, P. Cool,
60 B. Bridier and J. Perez-Ramıirez, J. Am. Chem. Soc., 2010,
132, 4321–4327.
P. Canu and A. Glisenti, Appl. Catal., B, 2017, 209, 214–227. 61 J. B. Wang, D. H. Tsai and T. J. Huang, J. Catal., 2002, 208,
55 X. Shi, B. Chu, F. Wang, X. Wei, L. Teng, M. Fan, B. Li, 370–380.
L. Dong and L. Dong, ACS Appl. Mater. Interfaces, 2018, 10, 62 E. Amini, M. Rezaei and M. Sadeghinia, Chin. J. Catal., 2013,
40509–40522. 34, 1762–1767.
56 K. Xiao, X. Qi, Z. Bao, X. Wang, L. Zhong, K. Fang, M. Lin 63 X. Zeng, C. Gong, H. Guo, H. Xu, J. Zhang and J. Xie, New
and Y. Sun, Catal. Sci. Technol., 2013, 3, 1591–1602. J. Chem., 2018, 42, 17346–17350.
57 Z. Xiao, S. Jin, X. Wang, W. Li, J. Wang and C. Liang, 64 S. Swami, A. Agarwala and R. Shrivastava, New J. Chem.,
J. Mater. Chem., 2012, 22, 16598–16605.
2016, 40, 9788–9794.
992 | New J. Chem., 2021, 45, 982--992
This journal is ©The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2021