Selective Aerobic Oxidation of 5-Hydroxymethylfurfural to 2,5-Diformylfuran or 2-Formyl-5-furancarboxylic Acid in Water by using MgO?CeO2 Mixed Oxides as Catalysts

Article abstract of DOI:10.1002/cssc.201800334

Mixed oxides based on MgO?CeO₂ were used as efficient catalysts in the aerobic oxidation of 5-hydroxymethylfurfural (5-HMF) to afford, with very high selectivity, either 2,5-diformylfuran (DFF, 99 %) or 2-formyl-5-furancarboxylic acid (FFCA, 90 %), depending on the reaction conditions. 5-Hydroxymethyl-2-furancarboxylic acid (HMFCA, 57–90 %) was formed only at low concentration of 5-HMF (<0.03 m) or in presence of external bases. The conversion of 5-HMF ranged from a few percent to 99 %, according to the reaction conditions. The oxidation was performed in water, with O₂ as oxidant, without any additives. The surface characterization of the catalysts gave important information about their acid–base properties, which drive the selectivity of the reaction towards DFF. FFCA was formed from DFF at longer reaction times. Catalysts were studied by XPS and XRD before and after catalytic runs to identify the reason why they undergo reversible deactivation. XRD showed that MgO is hydrated to Mg(OH)₂, which, even if not leached out, changes the basic properties of the catalyst that becomes less active after some time. Calcination of the recovered catalyst allows recovery of its initial activity. The catalyst is thus recoverable (>99 %) and reusable. The use of mixed oxides allows tuning of the basicity of the catalysts, avoiding the need for external bases for efficient and selective conversion of 5-HMF and waste formation, resulting in an environmentally friendly, sustainable process.

Full text of DOI:10.1002/cssc.201800334

Accepted Article
Title: Selective aerobic oxidation of 5-(hydroxymethyl)furfural to 2,5-
diformylfuran or 2-formyl-5-furancarboxylic acid in water using
MgO·CeO2 mixed oxides as catalysts
Authors: Angela Dibenedetto, Maria Ventura, Francesco Lobefaro,
Elvira de Giglio, Monica Distaso, and Francesco Nocito
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: ChemSusChem 10.1002/cssc.201800334
Link to VoR: http://dx.doi.org/10.1002/cssc.201800334
A Journal of
www.chemsuschem.org

ChemSusChem
10.1002/cssc.201800334
FULL PAPER
Selective aerobic oxidation of 5-(hydroxymethyl)furfural to 2,5-
diformylfuran or 2-formyl-5-furancarboxylic acid in water using
MgO·CeO mixed oxides as catalysts
2
Maria Ventura, Francesco Lobefaro, Elvira de Giglio, Monica Distaso, Francesco Nocito, [a] and
[
b]
[c]
[a]
[d]
Angela Dibenedetto*[a,b]
Abstract: Mixed oxides based on MgO·CeO
2
were used, as efficient
subsequent dehydration of the latter in acidic conditions (Scheme
1). It has been identified as an important platform chemical with
high industrial potential as intermediate feedstock.3
2
catalysts, in the aerobic oxidation of 5-(hydroxymethyl)furfural (5-
HMF) to afford, with very high selectivity, either 2,5-diformylfuran
(DFF, 99%) or 2-formyl-5-furancarboxylic acid (FFCA, 90%)
depending on the reaction conditions. 5-Hydroxymethyl-2-
furancarboxylic acid (HMFCA, 57-90%) was formed only at low
concentration of 5-HMF (<0.03 M) or in presence of external bases.
The conversion of 5-HMF ranged from a few units to 99 %, according
to the reaction conditions. The oxidation was performed in water, with
2
O as oxidant, without any additives. The surface characterization of
Scheme 1. Reaction pathway for the synthesis of 5-HMF from glucose.
the catalysts gave important information about their acid-base
properties, which drive the selectivity of the reaction towards DFF.
FFCA was formed from DFF for longer reaction times. Catalysts were
studied by XPS and XRD before and after catalytic runs to identify the
reason why they undergo reversible deactivation. XRD has clearly
As shown in Scheme 2, 5-HMF can undergo oxidation to different
products. In particular, the oxidation of the aldehyde group leads
to 5-hydroxymethyl-2-furancarboxylic acid (HMFCA), while the
selective oxidation of hydroxyl moiety produces the furan
dialdehyde 2,5-diformylfuran (DFF). Oxidation of the latter two
compounds leads to 5-formyl-2-furancarboxylic acid (FFCA),
precursor of 2,5-furandicarboxylic acid (FDCA).
2
shown that MgO is hydrated to Mg(OH) which, even if not leached
out, changes the basic properties of the catalyst that becomes less
active after some time. Calcination of the recovered catalyst allows
recovering their initial activity. The catalyst is, thus, recoverable
(>99%) and reusable. The use of mixed oxides allows tune the
basicity of catalysts, avoiding external bases for efficient and selective
conversion of 5-HMF and waste formation, resulting in an
environmentally friendly, sustainable process.
Introduction
Over the last decades the depletion of fossil fuel resources and
1
the increased atmospheric CO
2
level, have forced to investigate
how to replace fossil carbon with renewable carbon, such as
biomass or even CO . 5-(Hydroxymethyl)furfural (5-HMF), a furan
2
ring bearing an aldehyde and a hydroxyl moiety, is obtained from
glucose by isomerization to fructose under basic catalysis and
Scheme 2. The pathway of 5-HMF oxidation to DFF, HMFCA, FFCA and FDCA.
DFF is a versatile chemical intermediate for the synthesis of
[
a]
Prof. A. Dibenedetto, Prof. E. de Giglio, Dr F. Nocito
Department of Chemistry,
4
5
6
pharmaceuticals, fungicides, furan-urea resins or heterocyclic
7
ligands. Several reports describe the formation of DFF in organic
University of Bari, Campus Universitario, 70126 Bari, Italy
E-mail: angela.dibenedetto@uniba.it
Prof. A. Dibenedetto, Dr M. Ventura
CIRCC, Via Celso Ulpiani, 27, 70126 Bari, Italy
MSc F. Lobefaro
8
solvents, despite the use of the latter contributes to the formation
[
[
b]
c]
9
of waste in the whole process. Even photocatalytic systems have
been used9b or metal-free processes9c,d or oxidant-free catalytic
systems.9e Expensive metals, such as ruthenium or gold, have
often been used as catalysts.10 Recent studies have shown that
DFF can be synthesized in water with good conversion and 87%
selectivity, when Mn0.70Cu0.05Al0.25OH was used as catalyst in mild
conditions in absence of any additives. The structural properties
2
IC R, Tecnopolis,
km 3 via Casamassima, 70018 Valenzano (BA), Italy
Dr M. Distaso
Institute of Particle Technology (LFG)
Cauerstrasse 4D-91058, Erlangen, Germany.
[d]
This article is protected by copyright. All rights reserved.

ChemSusChem
10.1002/cssc.201800334
FULL PAPER
of the catalysts have not been fully disclosed, and this leaves
unclear the justification of such high activity.11 To the best of our
knowledge, there are no other reports describing a comparable
selectivity towards DFF using water as solvent and homogeneous
or heterogeneous catalysts. In fact, metal systems such as Cu(II),
Fe(II), Cu(II) with SALEN ligand afforded 14.39% selectivity, using
H
2
O
2
as oxidant.12 13.3% selectivity was reached when Ru was
supported on poly(4-vinylpyridine)-functionalized Carbon
1
3
NanoTubes (CNT). Conversely, covalent triazine frameworks-
CTF14 enhance the selectivity up to 52%, while oxidation to FFCA
also takes place. Over activated carbon, a selectivity of 60.5%
1
0a
1
5
2 3 2
and 28.8% is reported. -Al O or CeO allow to reach selectivity
of 37.5%1 and 18%, respectively. Such large variety of data
gives an idea of how important are both the catalyst and the
support in this kind of oxidation reaction.16 Studies of catalytic
systems based of N-hydroxyphthalimide with ceria show that
increasing the amount of ceria produces different results. For
0b
16
Scheme 3. Reactivity of 5-HMF in aqueous media: a. Activation of the alcoholic
functionality in slightly basic or neutral media; b. Activation of aldehydic moiety
in strong basic media.
2
instance, the use of large amount of CeO causes a decrease of
selectivity.17 Au18 or Cr19 were also used but no more than 30% of
selectivity was reached. Even if recently the synthesis of FDCA
with excellent yield and selectivity has been reported,20 good
yields or selectivity of DFF, which is the intermediate in the FDCA
synthesis (Scheme 2), is still matter of research. Besides the
catalyst, also the solvent and the oxidant play a key role. In
Taking all such aspects into consideration, we have figured out
that a bi-functional metal oxide with the correct oxidizing power
and combination of basic and acidic sites would be an effective
and efficient catalyst. Noteworthy, high selectivity plays an
important role as it would both minimize the post reaction
separation costs (energetic and economic) and reduce waste
formation. However, based on our knowledge of the oxidation of
2
principle, the use of water as well as the use of O should
minimize the waste formation and make the process more
environmentally friendly. Noteworthy, when the target product is
DFF, reactions in water are difficult to control. In fact, pH also
plays a key role in the reactivity of 5-HMF and its derivatives. In
Scheme 3a the reactivity of 5-HMF and DFF in aqueous slightly-
basic or neutral media is shown, while Scheme 3b shows the
reactivity of 5-HMF in a strong basic medium. The first step in the
pathway of DFF formation from 5-HMF is the alcohol activation
-HMF to FFCA using tunable mixed oxides,22 we have
5
investigated the factors that can maximize the oxidation of 5-HMF
to DFF (or FFCA), suppressing the formation of HMFCA, that
implies the oxidation of aldehyde moiety instead of the alcoholic
moiety. We describe here, the selective (98%+) and quantitative
(
Scheme 3a). In slightly basic aqueous solutions (or in presence
(
2
99%) oxidation of 5-HMF to DFF in water with O as oxidant,
of solid basic Lewis catalysts in water) the alcoholic functionality
can be easily deprotonated to its conjugate base that favours the
oxidation, while at neutral or acidic pH, the equilibrium is shifted
towards 5-HMF, and the oxidation of the alcoholic moiety is not
favored. On the other hand, in strong basic media, the aldehyde
moiety of 5-HMF is hydrated, and its oxidation is favored,
achieving HMFCA (Scheme 3b). DFF shows the same behavior:
in a strong basic medium it can be hydrated and FFCA is formed,
while in an acidic medium such equilibrium is shifted towards DFF
using magnesium/cerium mixed oxides as catalysts. We recall
that physical mixtures of oxides show unpredictable and
serendipitous properties.
Results and Discussion
1. Characterization of the catalysts
(
Scheme 3a). Therefore, the most effective way to activate the
Table 2 shows the BET surface area and the weak and strong
basic and acidic sites (calculated from the area under the peaks
in the TPD experiments) of solids, expressed through the volume
hydroxyl group promoting the formation of DFF is the use of Lewis
bases because strong Broensted bases promote also the
aldehydic moiety hydration, enhancing the formation of FFCA
from DFF or the formation of HMFCA from 5-HMF, with loss of
selectivity.21
2 3
of CO and NH up-taken and released, respectively. The strength
of different basic/acidic sites can be divided into three
23
categories. Using CO as probe gas, peaks appearing in the
2
Targeting high conversion and selectivity in the absence of
soluble bases, using Earth crust abundant heterogeneous
catalytic systems (recoverable and reusable), in water with O as
2
oxidant is, thus, a challenging task, but a success would enable
the development of environmentally safe and sustainable
processes.
temperature region of 350-430 K belong to the interaction with
hydroxyl groups on the surface that are weakly basic in nature.
Basic sites of medium strength are found in the temperature range
of 430-670 K that are attributed to the presence of M=O moieties.
The strongest basic sites are observed above 670 K and are
attributed to isolated O2- species. To simplify the analysis, we
have merged together medium and strong sites that should be
responsible of the catalytic activity, excluding weak sites.
This article is protected by copyright. All rights reserved.

ChemSusChem
10.1002/cssc.201800334
FULL PAPER
Concerning the acidity, weak acidity is that belonging to the
materials is obtained. These results are in agreement with data
previously reported in the literature, whereby the solubility of
protons of the –OH groups on the surface, while strong acidity is
n+ 24
relevant to the Lewis acid centers M . The lowest strong acidity
Group 2 metal oxides in CeO
treatment at 1673-1873 K,25 whereas the solubility of CeO
2
2
has been reported upon thermal
was found for MgO followed by CeO
MgO·CeO exhibits the highest value (Table 1, Entry 3; 2.51 mL
2
, while the new mixed oxide
in MgO
has been found to be negligible.26 In the present work, milder
conditions are used by coupling HEM and calcination at 783-983 K.
2
-1
g ). It is worth to mention that CeO
for the weak acidity.
2
presents the highest value
Table 1. BET surface area and basicity/acidity strength of binary and ternary metal oxides
Total
Weak
CO
ads
Stron
g CO
ads
Total
Weak
NH
ads
Stron
NH
ads
BET
surf
VCO
ads
2
2
2
VNH
ads
3
3
g
3
cat
area
2
(
mL/g)
(mL/g)
(mL/g)
(mL/g)
(mL/g)
(mL/g)
(m /g)
1
2
MgO 2.21
0.24
1.01
1.97
0.93
3.05
0.09
1.26
0.85
41.35
66.12
CeO
2
2.34
3.06
1.32
1.80
MgO·
CeO
3
0.29
2.77
3.21
0.69
2.51
66.62
2
2
Note: a binary oxide is composed of two different elements (MO or M O); a ternary oxide is
composed of three different elements (MM’O) (IUPAC Nomenclature).
A comparison between single oxides and mixed oxide MgO·CeO
reveals a quite different basicity that increases in the order:
CeO <MgO<MgO·CeO Interestingly, the mixed oxide
MgO·CeO exhibits the highest strong adsorption of CO (Table
, Entry 3). CeO shows the highest weak CO adsorption. The
2
2
2
.
Figure 1. XRD of (A) CeO
indicate the reflections of CeO
as guidelines for the reflection positions.26
2
(B) MgO and (C) MgOCeO
2
. Dashed vertical lines
2
(red) and MgO (blue), respectively, and serve
2
2
1
2
2
BET surface area has the highest value in the case of the 1:1
mixed oxide and results to be quite close to the value proper of
CeO . Other mixed oxides with variable Mg/Ce molar ratio were
2
also synthesized (vide infra). Table 2 shows their BET surface
area and basic/acid properties: increasing the MgO content in the
mixed oxide leads to an expected increase of the basicity. The
Figure 2 shows the SEM micrographs of the as-synthesized CeO
2
,
MgO and MgOCeO particles. The surface of the CeO powder
2
2
appears to be rough and highly porous with cavities on the surface
between 5 and 8 micron in diameter (Fig. 2a). A close-up image
reveals that the bulk material comprises primary particles with
either spherical or ellipsoidal shape and Feret diameters between
total volume of CO
2
adsorbed increases gradually while
increasing the strong basic sites, at the same time the number of
1.4 and 1.8 (Fig. 2b).
weak adsorption sites for CO
increases.
2
decreases as the MgO content
Concerning the BET surface area, we have observed an increase
with the amount of Mg until the ratio Mg:Ce reaches the value of
1, then a decrease was observed.
Table 2. Basicity/acidity strength and BET surface area of xMgO·CeO
2
Total
Weak
CO
ads
Stron
Total
Weak
NH
ads
Stron
BET
surf
V CO
ads
2
2
g CO
ads
2
V NH
ads
3
3
g NH
ads
3
Cat
area
2
(
mL/g)
1.84
(mL/g)
(mL/g)
(mL/g)
(mL/g)
(mL/g) (m /g)
0
0
1
.1MgO·
1
2
3
4
0.72
0.58
0.42
0.38
1.12
1.83
2.64
3.30
3.45
2.99
3.21
1.93
0.57
2.93 17.30
CeO
2
.5MgO·
CeO
2.41
3.06
3.68
0.39
0.69
0.55
2.59 53.75
2.51 66.62
1.37 48.09
2
MgO·C
eO
2
2
.2MgO·
CeO
2
Figure 2. Scanning Electron Microscopy characterization of the particles: a)
CeO low magnification and b) high magnification; c) MgO; d) MgO·CeO
The XRD analysis of MgO·CeO
2
(Fig. 1, C) shows the reflections
2
2
of both CeO and MgO, indicating that a solid solution of the two
2
This article is protected by copyright. All rights reserved.

ChemSusChem
10.1002/cssc.201800334
FULL PAPER
Moving to MgO powder, the sample comprises particles with
spheroidal shape and mean diameter of 3 m (Figure 2c).
Every particle appears to be formed by the aggregation of nano-
sheets with thickness between 40 and 100 nm (Fig. 2c). The SEM
2. Study on the stability of DFF versus pH
1
DFF is very sensitive to Broensted basic and/or acidic media. H-
NMR experiments have clearly shown that acetic acid ([HAc]= 1
mol/L; pH < 3; Fig. 3 A), or NaOH ([NaOH]= 0.001 mol/L; pH>10;
Fig. 3 B), deeply affect the stability of DFF in water. Under acidic
conditions, conversion to 5-HMF was observed, while under basic
conditions several products were formed, among which FFCA
and 5-HMF. The tests above confirm that pH has a great influence
on the conversion of 5-HMF into DFF, as we have proposed in
Scheme 3, and summarize below:
i. Basic sites promote the hydroxyl activation of 5-HMF and get
DFF. Nevertheless, high concentration of strong basic centers
promote the aldehyde function activation and produce
HMFCA instead of DFF,30 Scheme 3b.
ii. Acid pH in the long term affects the stability of DFF in aqueous
solution and converts it back to 5-HMF.
iii. The use of external bases increases the system complexity
and the separation costs, causing a stronger environmental
impact.
2
analysis of the MgOCeO powder shows that the two materials
are closely associated with each other in clusters with diameters
between 1 and 5 micron (Fig. 2d). These evidences, taken
together with XRD results, suggest that the regular structure and
the control on shape and size of the starting materials (MgO and
CeO
2
2
) are able to provide a mixed solid oxide of MgOCeO with
a highly homogeneous structure and phase composition.
The basic/acidic strength of the catalyst can also be controlled by
modifying the number of defects in the framework of the material,
depending on the calcination temperature used in their
2
7
synthesis. The influence of the temperature of calcination of
MgO·CeO on the acid-basic sites and BET surface area is shown
in Table 3.
2
Table 3. BET surface area and basicity strength of MgO·CeO
different temperatures
2
calcined at
However, as an alternative to the use of external soluble
acids/bases, we have chosen to use heterogeneous catalysts
tunable for their acid/basic properties, in order to perform the
oxidation of 5-HMF to DFF with the best conversion and maximum
selectivity, making easy its isolation and, avoiding the generation
of undesirable waste. Although oxides can be hydrated to produce
M-OH moieties, the latter are immobilized on solid particles (vide
infra) and result to cause lesser pH modification of the medium
than soluble Broensted counterparts.
Total
Weak
CO
ads
Strong
CO
ads
Total
Weak
NH
ads
Strong
NH
ads
BET
surf
Calcination
temperature
VCO
ads
2
2
2
VNH
ads
3
3
3
area
(
K)
2
(
mL/g)
(mL/g)
(mL/g)
(mL/g) (mL/g)
(mL/g) (m /g)
1
2
3
723
823
923
3.62
3.06
2.58
0.26
0.28
0.10
3.34
2.76
2.43
6.11
3.21
2.53
0.72
0.69
0.52
5.37
65.4
66.6
28.1
2.51
2.01
The sample calcined at the lowest temperature (723 K) has a
larger amount of defect sites, leading to large number of sites of
higher basicity, Entry 1.
Raising the calcination temperature to 823 K, the crystal size of
28
2
CeO grows and the number of defects decreases. As a result,
the number of oxygen atoms associated with Ce exhibiting low
coordination number decreases, and the number of strong basic
sites decreases. Increasing the calcination temperature to 923 K,
the strong basic sites further decrease as shown by the value for
the absorption of CO
2
, which goes to 2.43 mL/g from 3.34 mL/g.
At the same time, the weak strength basicity is reduced due to the
decrease of the surface area of the MgO·CeO
at 923 K.
2
samples calcined
The lowest temperature of calcination also causes the highest
acidity of the solid as it is shown in Table 3 Entry 1, while at the
highest temperature of calcination a lower acidity is obtained,
even correlated to the drop in the BET surface area, Entry 3.
Therefore, the calcination temperature has a strong effect on both
the number and strength of acid and basic sites. In our work, we
b a
use the ratio strong basic to strong acid sites (n /n ) as a
comparison parameter, more than the absolute values, and it
works quite well. Noteworthy, moving from 723 to 823 and 923 K
.
the calcination temperature, the ratio n
b
/n
a
for MgO CeO
2
varies
from 0.62 to 1.10 and 1.21, indicating an important change in the
properties of the catalyst.
Figure 3. 1H-NMR of the mixture generated by: (A) the addition of HAc to
aqueous DFF. (After 12h); (B) the addition of NaOH to aqueous DFF. (After 12h).
This article is protected by copyright. All rights reserved.

ChemSusChem
10.1002/cssc.201800334
FULL PAPER
Heterogeneous catalysts with an equilibrated distribution of
strong acid and basic sites result, thus, to be the most appropriate
to perform the conversion of 5-HMF into DFF
Products derived from the cleavage of the furan ring such as
formic acid (FA) and levulinic acid (LA) were also identified and
their concentration is function of the reaction conditions.
Interestingly enough, the initial concentration of 5-HMF influences
the nature of products. The best range for the selective production
of DFF is [5-HMF] = 0.15-0.5 M. At lower concentration (<0.1 M)
HMFCA is preferentially formed, while at high concentration (> 1
M) FFCA is the dominant product. Such concentration effect is still
under investigation in our laboratory.
3
. Catalysts screening in the oxidation of 5-HMF with O
2
in water.
As it is shown in Table 4, MgO (Entry 1) alone affords formic acid
(
(
FA) as major product, while, as we have already reported,22 CeO
Entry 2) shows good selectivity towards FFCA. The oxidation
calcined at 823 K (Entry
), gives DFF as almost the sole product with 98% of selectivity
accompanied by little by-products.
The conversion depends on the reaction time. To better
understand the behavior of the catalyst MgO·CeO , we have
2
reaction with the mixed oxide, MgO·CeO
4
2
.
Table 5. Oxidation of 5-HMF in presence of MgO CeO
2
in different conditions.
Selectivity (mol %)
PO
2
pH
T
Conv
2
cat
t(h) (MPa)
T
0
(K)
(%) DFF
HMFCA FFCA FDCA FA/LA
carried out studies where parameters such as catalyst
composition, PO2, reaction time and temperature were varied.
MgO·
0.9
3
6
1
2
3
4
5
6
373
373
403
373
373
373
15.1 99
97.8 98
0
0
0
<1
<2
5
CeO
2
MgO·
CeO
0.9
15
6
6
6
6
0
0
0
2
Table 4. Catalytic tests in the oxidation of 5-HMF using metal oxides as
catalysts
MgO·
CeO
0.9
9
99
0
0
90
10.1
0
0
2
Basic/
Acida
Time
h)
Conv.
(%)
Selectivity (mol %)
sites
MgO·
CeO
2
3
36.4/
14.8
cat
20.6 45.6
0
0
(
2
DFF
HMFCA
FFCA
FA
b a
n /n
MgO·
0.5
3
5.6
99
0
0
0
0
CeO
2
1
2
MgO
15
15
3
23.3
21.3
1.8
0.6
14.7
0
0
83.2
0
2.32
0.73
1.10
.
MgO
0.9
10
>10
17.9
(11.2)
41.8
57.5*
2.9
1.7
CeO
2
88.7
CeO2
MgO·
3
4
15.1
97.8
99
98
0
0
0
0
1
2
CeO
2
Reaction conditions: [5-HMF]i=0.14 M, 0.05 g of catalyst, 7 mL of water. *With
external bases.
.
MgO
CeO
15
1.10
2
5. Effect of the composition of the catalyst
Reaction conditions:[5-HMF]i=0.14M, 0.05g of catalyst, PO
K. Data from Table 1.
2
= 0.9 MPa. T = 373
a
The substitution of CeO with another oxidizing agent was tested
2
as it is shown in Table 6.
4. Influence of the reaction conditions on the conversion of 5-HMF
Table 6. Oxidation of 5-HMF in presence of different mixed oxides based on
MgO
Reactions were performed in different conditions of temperature,
time, PO and pH as shown in Table 5. Performing the reaction at
73 K (Entry 1) after 3 h DFF was formed as the sole product at
low conversion of 5-HMF (15 %) that increased at 15 h (97.8%)
keeping selectivity towards DFF at 98%. Under the conditions
above, the oxidation of the alcoholic moiety takes place with high
selectivity. When the temperature was increased to 403 K, Entry
2
Selectivity (mol %)
Time
(h)
T
(K)
Conv.
(%)
cat
3
DFF
0
HMFCA FFCA FDCA
FA/LA
0
MgO·
1
2
3
4
9
3
9
9
373
373
373
403
0
0
0
0
0
0
0
0
TiO
2
MgO·
CuO
24.04
15.7
98
0
16.3
0
82.2
74.9
64.5
0
3, FFCA was obtained as a single product after 9 h of reaction
without degradation, indicating the further oxidation of one of the
two formyl moieties of DFF into a carboxylic group (Scheme 2).
The oxygen pressure plays also a key role: when it was increased
from 0.9 to 2 MPa (Entry 4) poor selectivity was observed.
However, when the pressure was dropped to 0.5 MPa (Entry 5)
DFF was the only product, but a lower conversion was observed.
pH has a great importance in driving the selectivity of the reaction,
in fact when the reaction was performed under strong basic
conditions (pH >10) with the addition of external bases such as
NaOH, Entry 6, HMFCA (57.5%) was preferentially formed and
no DFF was detected.
MgO·
MnO
0
8.7/1.2
16.3/12.8
2
MgO·
MnO
2.6
2
i 2
Reaction conditions: [5-HMF]=0.14 M, 0.05 g of catalyst, 7 mL of water, PO =
0
.9 MPa.
CeO
2
has a crucial role in the reaction: when it was substituted
(Entry 1) the reaction did not proceed. CuO (Entry 2)
afforded FFCA, but not DFF. When CeO was substituted with a
stronger oxidant oxide such as MnO , (E° Ce /Ce =0.40 V; E°
with TiO
2
2
4
+
3+
2
This article is protected by copyright. All rights reserved.

ChemSusChem
10.1002/cssc.201800334
FULL PAPER
4
+
2+
Mn /Mn =1.23 V, Entry 3), the selectivity towards DFF was
dropped because 5-HMF was oxidized to FFCA. When this
reaction was carried out at higher temperature, Entry 4, DFF was
converted into FFCA and cleavage products such as FA and LA
were increased. Small amounts of FDCA were also formed.
Catalysts with strong oxidant properties drive, thus, to the further
oxidation of DFF. Mixed oxides with increasing amounts of strong
basic sites were prepared through the progressive addition of
2
MgO to CeO (Table 7) and each reaction carried out for 3 h.
The composition of the catalyst strongly influences both the
conversion yield and the selectivity. Entry 5 (Table 7) shows that
increasing the PO2 and reaction time promotes the conversion but
lowers the selectivity. DFF selectivity is plotted against the ratio
Mg/Ce and n
(
(
mixed oxides. Starting from 0.1MgO:1CeO
raised up to 2.2MgO:1CeO
b
/n
read on the right Y-axis) and the blue bars give the value of n
read on the left Y-axis); the X-axis gives the composition of the
, MgO content was
a
in Fig. 4. The yellow points give the selectivity
Figure 4. Correlation of selectivity towards DFF (right Y-axis), Mg content
b
/n
a
(
b a 2
X-Axis) and n /n of the MgO.CeO catalysts (left Y-axis). Data are taken
from Table 7. Catalytic runs were carried out for 3 h at 373 K in all cases.
The conversion depends on the catalyst used.
2
2
.
The basicity and acidity of the catalyst has thus a strong influence:
the best production of DFF is reached with the catalyst that shows
balanced properties in terms of strong basic/acid sites.
Table 7. Oxidation of 5-HMF in presence of different catalysts synthetized using
a different ratio Mg:Ce.
Selectivity (mol %)
Conv.
Cat
t (h)
T (K)
6. Effect of the calcination temperature
(
%)
DFF HMFCA FFCA FDCA FA/LA
We have discovered an interesting effect of the calcination
temperature over the selectivity (Table 8). We have used
MgO·CeO for such experiment as it is the most selective towards
2
DFF. When the solid calcined at 723 K was used, a high selectivity
towards FA was observed, Entry 1. With the solid calcined at 823
K (Entry 2) high selectivity towards DFF was obtained. The solid
calcined at 923 K produces mainly HMFCA (67%). As discussed
0
CeO
.1MgO·
2
14.8/
1
2
3
4
5
3
3
3
3
9
373
373
373
373
373
83.0
77.8
15.1
20.6
75.0
34.4 23.6
23.5
37.5
0
0
0
0
0
0
5.3
0
0
CeO
.5MgO·
2
62.6
97.0
80.0
80.6
0
0
0
0
1
eO
MgO·C
2
0
0
0
above, the n
b a
/n changes with the calcination temperature (Table
2
CeO
.2MgO·
2
0
3
and 4) influences the selectivity. The solid calcined at lower
temperature shows n
production of FA (Table 7, Entry 1). The solids calcined at 823
b
< n
a
(n
b
/n
a
=0.62), which causes the highest
2
CeO
.2MgO·
a
12.3
30
2
K shows n
Solids calcined at 923 K have n
b
≈ n
a
then the highest production of DFF was achieved.
> n , therefore a high production
b
a
Reaction conditions: [5-HMF]
i
=0.14 M, 0.05 g of catalyst, 7 mL of water, PO
2
=
a
of HMFCA was observed. It is worth to note that the latter exhibits
the same effect of added external bases.
0.9 MPa. PO
2
= 2 MPa.
High conversion was achieved with 0.1MgO:1CeO2, but the
selectivity towards DFF was poor (34% Fig. 5), because further
oxidation took place, affording FFCA at 23.5% (Table 7).
Table 8. Oxidation of 5-HMF in presence of MgO·CeO
2
calcined at different
temperatures in 3 h
Increasing the amount of MgO to 0.5MgO:1CeO
increase of the DFF selectivity (62%) was observed, with FFCA
still present (37.5% selectivity). Mixed oxide 1MgO:1CeO was
the most selective catalyst (98% DFF). However, increasing the
amount of MgO from 0.1 to 1 drives to an increase in the
selectivity towards DFF; when the amount of MgO was further
increased to 2.2 a lower selectivity was obtained but a higher
2
(Fig. 4), an
Calcination
Temperature
Conv.
(%)
Selectivity (mol%)
Basic/Ac
id Sites
nb/na
(
K)
2
DFF
3.77
99
HMFCA
0.75
0
FFCA
FA
1
2
3
723
823
923
10.6
15.1
16.3
10.37
0
85.66
<1
0.62
1.10
1.21
16.3
66.56
5.89
20.37
conversion of 5-HMF. As it is shown in Fig. 4, the most effective
.
catalyst for DFF production (MgO CeO
2
) is characterized by a
balanced number of strong basic and acid sites on the surface
/n =1.1).
Reaction conditions: [5-HMF]
i 2
=0.14 M, 0.05 g of catalyst, 7 mL of water, PO =
0.9 MPa
(
n
b
a
However, when the catalyst has a few strong basic and many acid
sites (0.1Mg1Ce) the selectivity towards DFF is dropped and
FFCA is formed in addition to HMFCA, fact that confirms the
hypothesis about the role of basic/acid sites.
In presence of strongly basic heterogeneous catalysts, thus, the
aldehyde moiety becomes strongly involved and either further
oxidation of DFF towards FFCA or the oxidation of the aldehydic
moiety of 5-HMF towards HMFCA occurs. The results shown
above confirm that the chemoselectivity of the reaction is mainly
This article is protected by copyright. All rights reserved.

ChemSusChem
10.1002/cssc.201800334
FULL PAPER
driven by a combination of strong basic-acidic sites in the
catalysts. The use of mixed oxides allows, thus, tune the ratio
that decreased after calcination). There were no differences on
the oxidation state of Mg before and after reaction, as expected.
XRD analysis was also performed on the solid before and after
n
b
a
/n in a smooth way, addressing the reaction towards one or the
other of the target products.
reaction. For MgO·CeO
2
, the pattern C in Fig. 1 is dominated by
CeO reflections with some contribution from MgO. The XRD
2
7. Tests on the recoverability and recyclability of the catalyst
pattern of the solid after the catalytic run (green line in Fig. 6)
shows that the reflections due to MgO at ca. 43° (200) and ca 62°
The catalyst recovery and reusability was tested by using either
(
3
220) are not present, whereas a new reflection appears at ca.
8° that corresponds to the most intense reflection of Mg(OH)
the second most intense reflection being present at ca. 58°, where
2
.2MgO·CeO (Fig. 5) or MgO·CeO . After each reaction run, the
2
2
2
,
catalyst was filtered off, washed with water (7 mL) three times,
and either used directly or calcined at 823 K and then reused in a
new run. Already at the second cycle of reaction using
a shoulder to the 222 reflection of CeO
observed.31
2
(Fig. 6) can be
2
2.2MgO·CeO reduction of DFF yield was evident (Fig. 5), while a
significant loss of the weight of the catalyst was verified. EDX
analyses on the recovered catalyst did show a loss of MgO and
the increase of the ratio Ce/Mg: before the catalytic run it was 0.45
after the first run was 0.76. It is interesting to note that after the
loss of Mg the catalysts was more robust for the consecutives
runs, the selectivity of DFF was kept constant (> 90%) for two
more runs.
2
Figure 6. XRD of MgO·CeO before and after a catalytic run. The blue dashed
vertical lines indicate the reflections of MgO that are missing in the solid after
the reaction. A new reflection appears at ca. 38°(*), suggesting the presence of
2 2
Mg(OH) . The red dashed vertical lines refer to CeO reflections.
2
Figure 5. Study on the reusability of 2.2MgO·CeO (9 h cycle)
This result, taken together with the elemental analysis, suggests
that the MgO is not leached during the reaction but converted into
2 2
Mg(OH) that is still present in the catalyst. Mg(OH) can be
The loss of Mg moved the composition of the catalyst towards the
:1 ratio with increase of stability and good selectivity towards
DFF.
formed by the reaction with solvent water, in agreement with what
1
reported by other authors.31,32 Calcination of the solid after the
.
catalytic cycle regenerates MgO CeO
2
by dehydration of Mg(OH)
2
By contrast, when MgO·CeO
2
was used as catalyst the
to MgO and re-establishes its activity. The powder after the
catalytic cycle was analyzed by SEM, finding that the morphology
and size of the particles were not considerably different from
those of the as-synthesized powders (Fig. 2d). Noteworthy, that
performance was constant even after repeated use (98%
selectivity towards DFF at 98% conversion). The catalyst was
calcined after each use. After three cycles of reaction, the
reduction of DFF yield was very limited (<9%) relative to the first
cycle. Also, no loss of weight of the catalysts was detected after
its use.
2
MgO or Mg(OH) is not lost during the catalytic run is
demonstrated by EDX that shows that Mg is not present in
solution or is present at trace level. This allows the catalyst
MgO.CeO
calcination. The results discussed above demonstrate that
MgO.CeO is an active, recoverable, recyclable and selective
catalyst in the aerobic oxidation of 5-HMF into DFF (98%
conversion, 99% selectivity) or FFCA (99% conversion, 90%
selectivity), depending on the reaction temperature, in water,
without generation of extra waste in the process, fully responding
to the principles of sustainable chemistry.
2
recover its efficiency and performance after
7
.1. Catalysts recycling: experiments and characterization by
XPS and XRD
2
Catalysts MgO·CeO
characterized by XPS and XRD to better clarify changes in their
structure/composition during a catalytic cycle.
2
and 2.2MgO·CeO
2
were further
2
XPS analysis of the 2.2MgO·CeO sample before and after
catalysis did show a decrease of the surface content of Ce(IV)
and MgO after the reaction. The reduction of Ce(IV) was due to
the formation of Ce(III) (13% of Ce(III) was found on the surface,
This article is protected by copyright. All rights reserved.

ChemSusChem
10.1002/cssc.201800334
FULL PAPER
Conclusions
for 3 h at 723 K giving a solid which color depends on the compounds
mixed. The calcined solids were transferred into a flask and stored under
2
N atmosphere to prevent uncontrolled surface deterioration prior to
catalysis.
2
The non-precious mixed oxide MgO·CeO prepared by the simple
HEM method was used as effective catalyst in the selective
oxidation of 5-HMF (0.14 to 0.5 M) to DFF using water as solvent,
Table 1. Mass (g) used for the synthesis of the mixed oxides
2
O as oxidant, without any type of additives. In this work we have
also demonstrated that the reaction parameters can be
modulated to oxidize the aldehyde moiety of 5-HMF instead. The
key role of basic/acidic sites has been proved. To reach the
maximum selectivity towards DFF, an almost equal number of
strong basic and acid sites is required. Relationship between
initial concentration of 5-HMF and selectivity or conversion has
also been demonstrated. Working at low concentration of
substrate (0.02 M) and increasing the basicity of the catalyst,
HMFCA is preferentially obtained. The composition of the catalyst
plays a key role: to reach more than 98% of selectivity towards
Entry
Mixed Oxide
MgO·TiO
g Mg(NO
3 2
) ·7H
2
O
g other nitrate/oxide
0.079 TiO2
1
2
3
2
0.25
0.25
0.25
MgO·CuO
MgO·MnO
0.19 Cu(NO
0.087 MnO
3 2 2
) ·3H O
2
2
3. Catalytic tests
DFF, 1MgO:1CeO
paper that MgO·CeO
catalytic runs after a calcination step. Transient deactivation is
due to hydration of MgO to Mg(OH) , as shown by XRD, more
than to modification of the composition by leaching. The process
that we have discussed in this work is totally eco-friendly and no
secondary residues have been obtained in optimized conditions.
This makes the catalyst a good candidate for an eventual up-scale.
2
is needed. We have clearly shown in this
The conversion of 5-HMF at a fixed temperature was studied in a 50 mL
stainless-steel reactor equipped with a withdrawal valve and an electrical
heating jacket. An appropriate amount of 5-HMF, depending on the desired
concentration, was dissolved in 7 mL of distilled water in a glass reactor
containing a magnetic stirrer, and 0.05 g of the catalyst under study were
added. The glass-reactor was then transferred into the autoclave that was
2
is recoverable and reusable in next
2
2
closed and purged three times with O .
Experimental Section
Experimental Details
1. Materials
Cerium ammonium nitrate ≥98% (by titration); cerium (IV) oxide
nanopowder, <25 nm particle size 99.95% trace rare earth metals basis;
magnesium oxide 99.99% trace metal basis; magnesium (II) nitrate
hexahydrate 99%; 2,5-furandicarboxaldehyde ≥97%; 5-formyl-2-furoic
acid 99%; 5-hydroxymethyl-2-furancarboxylic acid 99%, were ACS grade
reagents purchased from Sigma Aldrich. 5-(hydroxymethyl)furfural was
prepared as we have reported in ref. 2a,b.
2. Catalysts preparation
a) Synthesis of MgO·CeO
2
:
Cerium (IV) ammonium nitrate and magnesium (II) nitrate hexahydrate (the
relative mass depends on the desired ratio Mg/Ce in the final mixed oxide:
in the case of MgO·CeO , 0.43 g of Ce(IV) ammonium nitrate were reacted
2
with 0.25 g of Mg(II) nitrate hexahydrate) were mixed in a High Energy
Milling-HEM apparatus and pulverized at 790 rpm during 1h with pause of
1
min every 15 min and inversion of the rotation sense. The pale yellow
mixture was calcined for 3 h at 723-923 K giving a yellow solid. The solid
was transferred into a flask and stored under N atmosphere to prevent
uncontrolled surface deterioration prior to catalysis.
2
Figure 7. a. 1H-NMR, b. 13C-NMR of isolated DFF in CDCl
(Bruker, 600 MHz)
3
at 300 K.
n m
b) Synthesis of MgO·M O : (M=Ti, Cu, Mn)
An equivalent number of mmol of magnesium (II) nitrate hexahydrate and
the nitrate or oxide of the desired oxidant (depending on the commercial
It was charged with the appropriate pressure of oxygen and heated to the
reaction temperature as specified in Results and Discussion. At fixed
intervals of time, stirring was stopped so to allow the solid to settle, a 0.1
mL sample was withdrawn and analysed by HPLC following the conversion
of 5-HMF. When the concentration of the latter dropped to a constant value
availability of the solids: for example for the synthesis of MgO·TiO
2
, 0.25 g
of magnesium (II) nitrate hexahydrate were reacted with 0.079 g of TiO
2
,
more details are shown in Table 9), were mixed in a High Energy Milling-
HEM apparatus and treated as reported in (a). The mixture was calcined
(or even zero), the reaction was stopped. The solution and the solid were
This article is protected by copyright. All rights reserved.

ChemSusChem
10.1002/cssc.201800334
FULL PAPER
separated by centrifugation and both extracted three times with 7 mL of
ethyl acetate (EtOAc), collecting the EtOAc phases. Such procedure was
necessary because DFF is only slightly soluble in water and can precipitate
in the reaction medium or even deposit on the catalyst. The EtOAc phase
was washed with water to remove the residual 5-HMF eventually dissolved
Acknowledgements
The work was funded by the MIUR Industrial Research Project
CTN01_00063_49393 “REBIOCHEM” in the frame of the
Operative National Program-Research and Competitiveness
2007-2013. VALBIOR is gratefully thanked for gracious use of
equipment.
2 4
into it, dried with Na SO , filtered and analysed by GC. Evaporation in
vacuum of EtOAc gave the pure DFF (isolated yield: 98%) which was
analysed by NMR (Figure 7). Spectra show that DFF is very pure, with only
very minor traces of by-products. It is worth to emphasize that increasing
the amount of catalyst, speeds-up the reaction, but leaves unchanged the
selectivity.
Keywords: Oxidation of 5-HMF • selective non-precious
catalysts • aerobic oxidation in water• 2,5-diformylfuran (DFF) •
2-formyl-5-furancarboxylic acid (FFCA).
4. Analytical methods
5
-HMF and derivatives were analysed by using a JASCO HPLC equipped
[
1]
2]
a) S. Shafiee, E. Topal, Energy Policy 2009, 37, 181-189; b) M. Höök, X.
Tang, Energy Policy 2013, 52, 797-809.
with a Refractive Index (RI) detector and a Phenomenex Rezex RHM
Monosaccharide H (8%) 300x7.8mm at 343 K. A 0.005 N water solution
+
[
a) A. Dibenedetto, M. Aresta, L. di Bitonto, C. Pastore, ChemSusChem
of sulphuric acid was used as the mobile phase. The flow rate was
between 0.5-0.9 mL/min. The concentration of residual 5-HMF and
reaction products were determined using a RI detector. DFF concentration
was analysed in ethyl acetate solution, using a Thermo Scientific GC with
initial temperature of 323 K for 2 min. The rate of the first ramp was 7 º/min
until 393 K for 8 min. The second ramp rate was 15 º/min until 403 K for
2016, 9, 118-125; b) A. Dibenedetto, M. Aresta, C. Pastore, L. di Bitonto,
A. Angelini, E. Quaranta, RSC Adv. 2015, 5, 26941- 26948; c) W. Deng,
Q. Zhang, Y. Wang, Sci. China-Chem. 2015, 58, 29-46; d) X.L. Tong, Y.
Ma, Y.D. Li, Appl Catal A: Gen 2010, 385, 1-13.
[
3]
4]
a) M. Besson, P. Gallezot,C. Pinel, Chem. Rev. 2014, 114, 1827-1870;
b) R.-J. van Putten, J.C. van der Waal, E. de Jong, C.B. Rasrendra, H.J.
Heeres, J.G. de Vries, Chem. Rev. 2013, 113, 1499-1597; c) B. Xiao, M.
Zheng, X. Li, J. Pang, R. Sun, H. Wang, X. Pang, A. Wang, X. Wang, T.
Zhang, Green Chem. 2016, 18, 2175-2184.
10 min.
Pulse ChemiSorb 2750 Micromeritics instrument was used for the surface
characterization of the catalysts. Analyses of the acidic/basic sites were
[
a) K. T. Hopkins, W.D. Wilson, B.C. Bender, D.R. McCurdy, J.E. Hall,
R.R. Tidwell, A. Kumar, M. Bajic, D.W. Boykin, J. Med. Chem. 1998, 41,
carried out using NH
catalyst. The samples were pre-treated under N
K. The Pulse Chemisorb was performed with NH
carrier gas (30 mL min-1). Brunauer Emmett Teller (BET) surface area was
determined using N /He as carrier gas at 273 K followed by heating up to
23 K. Temperature Programmed Desorption (TPD) were performed
3
or CO
2
, respectively, as probe-gas using 100 mg of
(30 mL min-1) flow at 673
or CO gas using He as
2
3872-3878; b) D.W. Boykin, A. Kumar, J. Spychala, M. Zhou, R.J.
3
2
Lombardy, W.D. Wilson, C.C. Dykstra, S.K. Jones, J.E. Hall, R.R. Tidwell,
C. Loughton, C.M. Nunn, S. Neidle, J. Med. Chem. 1995, 38, 912-916.
M. Del Poeta, W.A. Schell, C.C. Dykstra, S.K. Jones, R.R. Tidwell, A.
Kumar, D. W. Boykin, J. R. Perfect, Antimicrob. Agents Chemother. 1998,
2
[
5]
6]
9
under He flow at 30 mL min-1.
42, 2503-2510.
[
a) A.S. Amarasekara, D. Green, L.D. Williams, Eur. Polym. J. 2009, 45,
595-598; b) A. Gandini, N.M. Belgacem, Polym. Int. 1998, 47, 267-276.
D.T. Richter, T. D. Lash, Tetrahedrom Lett. 1999, 40, 6735-6738.
a) B. Karimi, H.M. Mirzaei, E. Farhangi, Chem.Cat.Chem 2014, 6, 758-
762; b) Y. Zhu, M. Shen, Y. Xia, M. Lu, Catal. Comm. 2015, 64, 37-43;
c) R. Fang, R. Luque, Y. Li, Green Chem. 2017, 19, 647-655.
All reported values are the average of three measurements, with a
standard deviation < 3%.
[7]
8]
[
X-Ray Diffraction (XRD) analysis was performed using a D8 Advance
instrument (Bruker AXS GmbH, Germany) with Cu Kα radiation (λ = 1.5406
Å). The patterns were collected in the 2 range of 20-80° (step size 0.014°,
[9]
a) D. Prat, A. Wells, J. Hayler, H. Sneddon, C.R. McElroy, S. Abou-
Shehada, P.J. Dunn, Green Chem. 2016, 18, 5082-5085; b) H. Zhang,
Q. Wu, C. Guo, Y. Wu, T.Wu, ACS Sust Chem and Eng, 2017, 5, 3517-
3523; c) P.V. Rathod, S.D. Nale, V.H. Jadhav, ACS Sust Chem and Eng,
2017, 5(1), 701-707; d) J., Li, G. Lv, B. Lu, Y. Wang, T. Deng, X. Hou, Y.
Yang, Energy Technology, 2017, 5, 1429-1434; e) G. Li, Z. Sun, Y. Yan,
Y. Zhang, Y. Tang, ChemSusChem, 2017, 10, 494-498.
2
seconds). The powders were manually compressed inside low volume
sample holders comprising low background sample cups with a vicinal
911) Si crystal of 25 mm diameter (Bruker AXS GmbH). All the presented
patterns were background subtracted. X-ray photoelectron spectroscopy
XPS) was conducted on a Thermo VG scientific ESCA MultiLab-2000
(
(
spectrometer with a monochromatized Al-Kα source (1486.6 eV) at
constant analyser pass energy of 25 eV.
[10] a) Nie, J.; Xie, J.; Liu, H. J. Catal. 2013, 301, 83-91; (b) Antonyraj, C. A.;
Jeong, J.; Kim, B.; Shin, S.; Kim, S.; Lee, K.-Y.; Cho, J. K. J. Ind. Eng.
Chem. 2013, 19, 1056-1059; (c) Stahlberg, T.; Eyjolfsdottir, E.; Gorbanev,
Y. Y.; Sadaba, I.; Riisager, A. Catal. Lett. 2012, 142, 1089-1097; (d)
Takagaki, A.; Takahashi, M.; Nishimura, S.; Ebitani, K. ACS Catal. 2011,
1, 1562-1565.
ULTRATM 55 instrument (Carl Zeiss AG, Germany) was used for Scanning
Electron Microscopy (SEM) analyses. The particles were deposited
directly on a conductive sticky carbon pad (Plano GmbH). Image analysis
was performed by using the freely available software.33 The size of
particles with ellipsoidal morphology was estimated through the Feret´s
diameter, i.e. as ratio between the projection of the long and short axis in
electron micrographs.
[11] F. Neaţu, N. Petrea, R. Petre, V. Somoghi, M. Florea, V.I. Parvulescu,
Catal. Today 2016, 278, 66-73.
[
12] D.X. Martinez-Varga, J. Rivera de la Rosa, L. Sandoval-Rangel, J.L.
Guzman-Mar, M.A. Garza-Navarro, C.J. Lucio-Ortiz, D.A. De Haro-
DelRio, Applied Cat A: General, 2017, 547, 132-145.
Energy Dispersion X-ray Spectrometer (EDX) Shimadzu EDZ-720 was
used for elemental analyses of the catalysts.
[13] J. Chen, J. Zhong, Y. Guo, L. Chen, RSC Adv. 2015, 5, 5933-5940.
[14] J. Artz, S. Mallmann, R. Palkovits, ChemSusChem 2015, 8, 672-679.
[15] J. Nie, J. Xie, H. Liu, Cuihua Xuebao 2013, 34, 871- 875.
[16] Y.Y. Gorbanev, S. Kegnaes, A. Riisager, Top. Catal. 2011, 54, 1318-
1324.
[17] Y. Yan, X. Tong, K. Wang, X. Bai, Catal. Commun. 2014, 43, 112-115.
This article is protected by copyright. All rights reserved.

ChemSusChem
10.1002/cssc.201800334
FULL PAPER
[
18] a) X. Wan, C. Zhou, J. Chen, W. Deng, Q. Zhang, Y. Yang, Y. Wang,
ACS Catal. 2014, 4, 2175-2185; b) S. Albonetti, A. Lolli, V. Morandi, A.
Migliori, C. Lucarelli, F. Cavani, Appl. Catal., B 2015, 163, 520-530; c) A.
Villa, M. Schiavoni, S. Campisi, G.M. Veith, L. Prati, ChemSusChem
2
013, 6, 609-612.
19] F. Tao, Y. Cui, P. Yang, Y. Gong, Russ. J. Phys. Chem. A 2014, 88,
091-1096.
[
[
1
20] a) S. Li, K. Su, Z. Li, B.Cheng, Green Chem. 2016, 19, 914-918; b) Z.
Zhang, J. Zhen, B. Liu, K. Lv, K. Deng, Green Chem. 2015, 17, 1308-
1317; c) L. Zheng, J. Zhao, Z. Du, B; Zong, H. Liu, Science China
Chemistry, 2017, 60, 950-957; d) S. Xu, P. Zhou, Z. Zhang, C. Yang, B.
Zhang, K. Deng, S. Bottie, H. Zhu, J Am Chem Soc, 2017, 139, 14775-
14782.
[21] X.L. Tong, Y. Ma, Y.D. Li, Appl. Catal. A-Gen. 2010, 385, 1-13.
[22] a) M. Ventura, M. Aresta, A. Dibenedetto, ChemSusChem 2016, 9, 1096-
1100; b) M. Aresta, A. Dibenedetto, M. Ventura, Patent
102016000014339, 2016; c) M. Ventura, A. Dibenedetto, M. Aresta,
Inorg. Chim. Acta, 2018, 470, 11-21.
[23] G.A.H. Mekhemer, S.A. Halawy, M.A. Mohamed, M.I. Zaki, J. Phys.
Chem. B 2004, 108, 13379-13386.
[24] P. Carniti, A. Gervasini, F. Bossola, V.D. Santo, Appl. Catal. B: Environm.
2016, 193, 93-102.
[
25] S.V. Chavan, A.K. Tyagi, Thermochim. Acta 2002, 390, 79-82.
26] A. Kumar, S. Thota, S. Sivakumar, S. Priya, J. Kumar, J. Sol-Gel Sci.
Technol 2013, 68, 46-53.
[
[
27] Q. Dai, X. Wang, G.Lu, Appl. Catal. B: Environm. 2008, 81, 192-202.
28] V.V. Pushkarev, V.I. Kovalchuk, J.L. d’Itri, J. Phys. Chem. B 2004, 108,
[
5341-5348.
[
29] H. Ait Rass, N. Essayem, M. Besson, Green Chem. 2013, 15, 2240-2251.
30] S.K. Triantafyllidis, A.A. Lapas, M. Stöcker, The role of catalysis for the
sustainable production of bio-fuels and bio-chemicals; Elsevier, 2013.
31] M. Uda, H. Okuyama, T.S. Suzuki, Y. Sakka, Sci. Tech. Adv. Mat. 2012,
[
[
[
13, 025009.
32] F. Al-Hazmi, A. Umar, G.N. Dar, A.A. Al-Ghamdi, S.A. Al-Sayari, A. Al-
Hajry, S.H. Kim, R.M. Al-Tuwirqi, F. Alnowaiserb, F. El-Tantawy, J Alloys
Compd 2012, 519, 4-8.
[
33] W.S. Rasband, J. Image, U.S. National Institutes of Health, Bethesda,
Maryland, USA, http://imagej.nih.gov/ij/, 1997–2014.
This article is protected by copyright. All rights reserved.

ChemSusChem
10.1002/cssc.201800334
FULL PAPER
FULL PAPER
Maria Ventura, Francesco Lobefaro,
Elvira de Giglio, Monica Distaso,
Francesco Nocito, and Angela
Dibenedetto
MgO·CeO , O , H O
2
2
2
Page No. – Page No.
Selective aerobic oxidation of 5-
(hydroxymethyl)furfural to 2,5-
diformylfuran or 2-formyl-5-
By using a tunable mixed oxide such as MgO·CeO
and quantitative oxidation of 5-HMF to DFF (or FFCA) in water in the presence of O
2
we have obtained a selective
2
as oxidant. Interestingly, the oxidation of the aldehyde moiety, which brings to the
formation of HMFCA, is suppressed.
furancarboxylic acid in water using
MgO·CeO mixed oxides as catalysts
2
[
a]
Prof. A. Dibenedetto, Prof. E. de Giglio, Dr F. Nocito
Department of Chemistry,
University of Bari, Campus Universitario, 70126 Bari, Italy
E-mail: angela.dibenedetto@uniba.it
This article is protected by copyright. All rights reserved.[b]
Prof. A. Dibenedetto, Dr M. Ventura
CIRCC, Via Celso Ulpiani, 27, 70126 Bari, Italy