Selective conversion of 5-hydroxymethylfurfural to diketone derivatives over Beta zeolite-supported Pd catalysts in water

Article abstract of DOI:10.1016/j.jcat.2019.04.038

Conversion of 5-hydroxymethylfurfural (HMF) in water to the linear diketone derivatives 1-hydroxyhexane-2,5-dione (HHD) and 2,5-hexanedione (HXD) was investigated over a series of Beta zeolite-supported transition metal catalysts (Co, Ni, Cu, Ru, Pd). Their catalytic performance was tested in a batch stirred reactor (T = 110 °C, P_H2 = 20 bar) with Pd showing the highest activity and selectivity to HHD and HXD. The effects of Pd particle size, zeolite Si/Al ratio and reaction conditions (T = 80–155 °C, P_H2 = 5–60 bar) were also investigated. The incorporation of Pd into Beta zeolite by the deposition-coprecipitation method produced the most efficient catalyst, affording complete HMF conversion (T = 110 °C, P_H2 = 60 bar) predominantly to HHD (68% selectivity) and HXD (8% selectivity). The combination of a bifunctional acid/redox solid catalyst and water enhances the hydrolytic ring-opening and subsequent hydrogenation of the furan ring. Catalytic activity can be partially restored by a simple regeneration treatment. This work establishes a catalytic route to produce valuable diketone derivatives from renewable furanic platform sources in water.

Full text of DOI:10.1016/j.jcat.2019.04.038

Journal of Catalysis 375 (2019) 224–233
Contents lists available at ScienceDirect
Journal of Catalysis
journal homepage: www.elsevier.com/locate/jcat
Selective conversion of 5-hydroxymethylfurfural to diketone derivatives
over Beta zeolite-supported Pd catalysts in water
Rubén Ramos a, Alexios Grigoropoulos a, Ben L. Griffiths a,b, Alexandros P. Katsoulidis a, Marco Zanella a,
Troy D. Manning a, Frédéric Blanc b,c, John B. Claridge b, Matthew J. Rosseinsky a,
⇑
a Materials Innovation Factory, University of Liverpool, Liverpool L7 3NY, UK
b Department of Chemistry, University of Liverpool, Liverpool L69 7ZD, UK
c Stephenson Institute for Renewable Energy, University of Liverpool, Liverpool L69 7ZF, UK
a r t i c l e i n f o
a b s t r a c t
Article history:
Conversion of 5-hydroxymethylfurfural (HMF) in water to the linear diketone derivatives 1-
hydroxyhexane-2,5-dione (HHD) and 2,5-hexanedione (HXD) was investigated over a series of Beta
zeolite-supported transition metal catalysts (Co, Ni, Cu, Ru, Pd). Their catalytic performance was tested
in a batch stirred reactor (T = 110 °C, PH2 = 20 bar) with Pd showing the highest activity and selectivity
to HHD and HXD. The effects of Pd particle size, zeolite Si/Al ratio and reaction conditions (T = 80–
155 °C, PH2 = 5–60 bar) were also investigated. The incorporation of Pd into Beta zeolite by the
deposition-coprecipitation method produced the most efficient catalyst, affording complete HMF conver-
sion (T = 110 °C, PH2 = 60 bar) predominantly to HHD (68% selectivity) and HXD (8% selectivity). The com-
bination of a bifunctional acid/redox solid catalyst and water enhances the hydrolytic ring-opening and
subsequent hydrogenation of the furan ring. Catalytic activity can be partially restored by a simple regen-
eration treatment. This work establishes a catalytic route to produce valuable diketone derivatives from
renewable furanic platform sources in water.
Received 6 December 2018
Revised 30 March 2019
Accepted 23 April 2019
Keywords:
5-Hydroxymethylfurfural
HMF conversion
Beta zeolite
Supported Pd catalyst
Hydrogenation
Furan ring-opening
Ó 2019 Elsevier Inc. All rights reserved.
1. Introduction
In 2004, the U.S. Department of Energy (DOE) [3] released a
report, later revised by Bozell et al. [4], identifying the top value-
Non-edible lignocellulose is the most abundant, cheapest and
fastest growing sustainable biomass resource, composed of three
primary biopolymers: cellulose (a polymer of glucose), hemicellu-
lose (a polymer mainly of pentoses) and lignin (a highly cross-
linked polymer of substituted phenols) [1]. In order to produce
value-added bio-products which could displace petroleum feed-
stocks, lignocellulose must first be transformed into simpler and
more easily processed platform chemicals. This approach, similar
to that used in conventional petroleum refineries, would allow
the simultaneous production of biofuels and biochemicals in an
integrated facility, a biorefinery [2].
added platform chemicals in a future biorefinery. HMF was identi-
fied as one of the most appealing and promising building block
molecules. This furan derivative can be produced from agricultural
waste and forest residue such as polysaccharides (i.e. cellulose and
hemicellulose) by acid-catalysed hydrolysis to C6 monosaccha-
rides, followed by dehydration [5]. In contrast to most petrochem-
ical products, HMF is an oxygen-rich, functionalized compound. Its
conversion to value-added chemicals usually involves several
chemical transformations (e.g. hydrogenation, dehydration,
hydrogenolysis, oxidation, etc.) which are promoted by multifunc-
tional heterogeneous catalysts [6,7]. The development of related
catalytic heterogeneous processes has become highly topical to
produce valuable bioproducts such as: tetrahydrofuran 2,5-
diyldimethanol (THFDM) [8–10], 2,5-dimethylfuran (DMF) [11],
2,5-furandicarboxylic acid (FDCA) [12,13], C6 linear alcohols
[14,15], 3-hydroxymethylcyclopentanone (HCPN) and 3-
hydroxymethylcyclopentanol (HCPL) [16,17] (Fig. 1).
Abbreviations: HMF, 5-hydroxymethylfurfural; HHD, 1-hydroxyhexane-2,5-
dione; HXD, 2,5-hexanedione; FDM, furan-2,5-diyldimethanol; THFDM,
tetrahydrofuran-2,5-diyldimethanol;
DMF,
2,5-dimethylfuran;
HCPN,
3-
hydroxymethylcyclopentanone; HCPL, 3-hydroxymethylcyclopentanol; HHED, 1-
hydroxyhex-3-ene-2,5-dione.
⇑
One much less studied reaction is the conversion of HMF into
the linear diketone derivatives 1-hydroxyhexane-2,5-dione
Corresponding author at: Department of Chemistry, Materials Innovation
Factory, University of Liverpool, Liverpool L7 3NY, UK.
E-mail address: m.j.rosseinsky@liverpool.ac.uk (M.J. Rosseinsky).
https://doi.org/10.1016/j.jcat.2019.04.038
0021-9517/Ó 2019 Elsevier Inc. All rights reserved.

R. Ramos et al. / Journal of Catalysis 375 (2019) 224–233
225
Fig. 1. Chemical transformation of HMF into various biomass-derived compounds [8–17].
(HHD) and 2,5-hexanedione (HXD). Although a large-scale syn-
thetic route to HHD is currently not available, the presence of a
hydroxymethyl functionality offers opportunities for the synthesis
of valuable chemicals, as recently highlighted [18–20]. HXD is
employed as a solvent and as an intermediate for the synthesis
of polymers, amines and surfactants [21,22]. HHD can be produced
from HMF in water under H2 pressure via the metal-catalysed
selective hydrogenation of the carbonyl group to furan-2,5-
diyldimethanol (FDM), followed by the acid-catalysed ring-
opening of the FDM unsaturated ring [23]. Notably, this pathway
does not involve the formation of a more stable saturated tetrahy-
drofuran ring, whose ring-opening requires harsh reaction condi-
tions (i.e. T > 140 °C; P > 60 bar) [24]. HXD can also be produced
from FDM via hydrogenolysis to DMF, followed by hydrolytic
ring-opening of the latter or by HHD via scission of the hydroxyl
group (Fig. 2) [22,25].
The formation of HHD from HMF was firstly reported in 1991 by
Schiavo et al. using Pd/C in an aqueous solution of oxalic acid
(pH = 2) at 70 bar H2 and 140 °C [23]. In 2009, Luijkx et al. reported
a similar process using HCl [26]. In 2014, Liu et al. developed two
binary catalytic systems using Pd/C either in CO2/H2O (forming
carbonic acid) [27], or in THF with co-added Amberlyst-15 [28],
affording in both cases 77% yield of HHD (see Table SI1). Recently
we reported that HHD is formed as a low yield intermediate (<7%)
in the conversion of HMF to HCPN and HCPL over M-Al2O3 catalysts
in H2O (M = Co, Ni, Cu). HHD was rapidly converted to HCPN via an
aldol condensation reaction, catalysed by basic sites, followed by
hydrogenation [17].
easily tuneable bifunctional catalysts. Among the most investi-
gated zeolites, Beta (with BEA topology) exhibits excellent proper-
ties to the aimed transformation due to its high hydrothermal
stability, large specific surface area (>600 m2 gÀ1), 3D large-pore
channel system (5.5–7.6 Å) and dual Lewis/Brønsted acidity [29].
Beta zeolite-based catalysts have been used for the conversion of
furfural into levulinic acid [30,31] and for the hydrodeoxygenation
of furoins into alkanes [32]. However, to the best of our knowledge,
there are no studies reporting the formation of linear diketone
derivatives using Beta zeolite-supported catalysts.
Herein, we present the catalytic production of the diketone
derivatives HHD and HXD from HMF by zeolite-supported transi-
tion metals in H2O. A series of transition metal-loaded (M) Beta
zeolites were prepared (M = Co, Ni, Cu, Ru, Pd), characterised and
tested in a batch stirred reactor under H2 pressure with Pd showing
the highest catalytic activity. Consequently, the effects of Pd parti-
cle size, zeolite Si/Al ratio and reaction conditions were investi-
gated, and catalyst stability and recyclability were evaluated.
This work establishes Beta zeolite-supported Pd catalysts as
promising candidates for the upgrading of HMF into valuable
biomass-derived linear diketone derivatives by demonstrating for
the first time the conversion of HMF to HHD and HXD in water
by a solid state, bifunctional (no acid co-added) catalyst.
2. Experimental
2.1. Catalyst preparation
The requirement of hydrogenating metal phases and acidic sites
for the production of HHD prompted us to investigate the deposi-
tion of various transition metals over zeolite supports to prepare
Metal-loaded zeolites were prepared by incipient wetness
impregnation (IWI) of commercially available Beta (Si/Al = 12.5)
and ZSM-5 (Si/Al = 11.5) zeolites purchased from Zeolyst Int. The
corresponding aqueous solutions of the metal precursors were
prepared using: PdCl2 (Sigma-Aldrich), Pd(NO3)2Á2H2O (Aldrich),
RuCl3 (Aldrich), NiCl2Á6H2O (Aldrich), CuCl2Á2H2O (Aldrich) and
CoCl2Á6H2O (Fluka). Prior to impregnation, the parent
NH4-zeolites were calcined at 550 °C (heating rate of 2 °C minÀ1
)
in static air for 5 h, producing the respective H-zeolites. The depo-
sition of the metal was carried out by adding dropwise the aqueous
solution of the precursor to the zeolite support at room tempera-
ture (3 wt% for the Pd and Ru samples; 10 wt% for the Cu, Ni and
Co samples). After impregnation, the catalysts were dried in a
rotary evaporator at 65 °C under vacuum for 1 h. Subsequently,
the dried samples were calcined in air at 500 °C for 5 h (heating
rate of 2 °C minÀ1). The reduction treatment was performed under
Fig. 2. Catalytic reaction pathway discussed in this work for the conversion of HMF
to the linear diketone derivatives HHD and HXD via the acid-catalysed hydrolytic
ring-opening of FDM.

226
R. Ramos et al. / Journal of Catalysis 375 (2019) 224–233
pure H2 flow (100 cm3 minÀ1) for 5 h at 200 °C (for the Pd and Ru
samples), 300 °C (Cu sample) and 500 °C (Ni and Co samples) with
a heating rate of 2 °C minÀ1. Finally, the catalysts were passivated
under a flow of 1% v/v O2/N2 (100 cm3 minÀ1) for 2 h at room
temperature.
Pd-loaded Beta zeolite was also prepared by deposition-
coprecipitation (DP-CP) using the urea-based method developed
by Geus et al. [33]. First, 2 g of the calcined Beta zeolite were placed
microcrucible and heated to 800 °C at 10 °C minÀ1 under a flow
of air (100 cm3 minÀ1). Elemental analysis (C and H content) of
the used catalysts was carried out on a Thermo EA1112 Flash CHNS
Analyser. TEM images were obtained with a JEOL 2100 transmis-
sion electron microscope operating at 200 kV. The samples were
dispersed in acetone, stirred in an ultrasonic bath and deposited
on a carbon-coated Cu grid. SEM imaging and energy-dispersive
X-ray (EDX) spectroscopy were run on a Hitachi S-4800 Field-
Emission scanning electron microscope.
in
a 250 ml round-bottom flask. Then, an aqueous solution
(100 ml) containing PdCl2 (0.005 M) and urea (1.2 M, Sigma) was
added dropwise with constant stirring (550 rpm) at room temper-
ature. The suspension (pH = 4–4.5) was heated to 95 °C to initiate
urea hydrolysis. After 3 h, the pH of the suspension remained
stable at pH ꢀ 7.5. The solution was cooled to room temperature
and the precipitate was collected by filtration, washed with deion-
ized water, dried at 110 °C overnight and subsequently calcined in
air at 500 °C for 5 h (heating rate of 2 °C minÀ1). The calcined sam-
ple (catalyst precursor) was then reduced under pure H2 flow
(100 cm3 minÀ1) at 200 °C for 5 h (heating rate of 2 °C minÀ1).
Finally, the reduced catalyst was passivated under a flow of 1% v/
v O2/N2 (100 cm3 minÀ1) for 2 h at room temperature. Hereafter,
the sample prepared by the DP-CP urea method will be referred
as Pd(u)/Beta.
Solid state 27Al NMR experiments were performed on a 9.4 T
Bruker DSX 400 MHz spectrometer using
a Bruker Triple
Resonance 4 mm HXY (in double resonance mode) probe under
Magic Angle Spinning (MAS) at a rotational rate of 10 kHz.
One-dimensional MAS NMR spectra were recorded using
rotor-synchronized (1 period) Hahn echo sequence with a radio
frequency pulse of 50 kHz ( /2 pulse of 1.7 s duration) and a
a
p
l
quantitative recycle delay of 1 s. Whilst the quantitative interpre-
tation of 27Al MAS NMR data has to be performed with caution
due to non-uniform excitation of sites with different magnitudes
of the quadrupolar coupling constants [34], the similar values
observed for tetrahedral and octahedral sites (i.e. 1–2 MHz) allow
for an estimation of their ratio [35,36].
Partial dealumination of the calcined Beta zeolite (Si/Al = 12.5)
was carried out by acid treatment using HNO3 aqueous solutions
of different concentration (0.1, 0.5, 2 and 5 M) at room tempera-
ture for 1 h (20 mL gÀ1 zeolite). After filtration and washing with
deionized water, the materials were dried overnight (110 °C) and
calcined in static air at 500 °C for 5 h (heating rate of 2 °C minÀ1).
Afterwards, the obtained dealuminated zeolites were impregnated
with Pd following the DP-CP urea method described above. Here-
after, the four dealuminated and impregnated samples will be
abbreviated as Pd(u)/Beta-dAlx (x = 1–4), where x = 4 refers to
the sample showing the highest degree of dealumination (higher
Si/Al ratio).
2.3. Catalytic experiments and product analysis
The performance of the catalysts was studied in high pressure
100 ml batch stirred reactors (Parr Instrument Co.) A glass liner
was loaded with 45 ml of an aqueous solution of HMF (0.04 M)
and 0.06 g of catalyst and placed into the stainless-steel reactor.
After sealing the vessel, the reactor was flushed three times with
N2 and heated to the required reaction temperature (80–155 °C).
Once the targeted temperature was reached, the vessel was pres-
surised with H2 to the respective value (5–60 bar of H2) and stir-
ring was set to 600 rpm. After the end of the reaction (typically
6 h), the identity and distribution of the products were determined
by the combination of 1H and 13C NMR spectroscopy (Bruker
AVANCE III HD spectrometer), GC-MS (Agilent 6890 N GC with a
5973 MSD detector) and GC (Agilent 7890A GC with an FID). GC
and GC-MS were equipped with a DB-WAXetr capillary column
2.2. Catalyst characterization
The prepared catalysts were characterised by powder X-Ray
diffraction (PXRD) on a Panalytical X’Pert Pro diffractometer with
(60 m, 0.25 mm i.d., 0.25 lm). Standard reference compounds
Co K 1 radiation (k = 1.7890 Å) in the 2h angle range 10À80° (scan-
a
used: HMF (Sigma), FDM (Manchester Organics), THFDM (Ambin-
ter) and HXD (Sigma-Aldrich). Details regarding calculations of
conversion, yield and selectivity are provided in the Supporting
Information (SI).
ning speed of 0.023° sÀ1). Metal content of the catalysts was deter-
mined by inductively coupled plasma
- optical emission
spectroscopy (ICP-OES) using an Agilent 5110 SVDV instrument.
The samples were digested in a strong acidic medium (10 ml HCl
and 20 ml HNO3) and then diluted with water (1:10 v/v). Textural
properties were evaluated through N2 adsorption-desorption iso-
therms at 77 K, using a Micromeritics TRISTAR II instrument. Prior
to the measurement, the samples were outgassed under vacuum at
120 °C for 20 h. The BET equation was used for specific surface area
calculation, whereas pore volume was determined by the BJH
method.
Acidity of the catalysts was determined by temperature pro-
grammed desorption of ammonia (NH3-TPD) in a Quantachrome
ChemBET 3000 unit. Firstly, the samples were outgassed under a
He stream (100 cm3 minÀ1) heating at 10 °C minÀ1 up to 350 °C.
Afterwards, the samples were cooled to 150 °C and saturated under
an ammonia stream (100 cm3 minÀ1) for 10 min. Subsequently, the
physically adsorbed ammonia was removed by flowing helium
(100 cm3 minÀ1) for 30 min at 150 °C. Finally, the chemically
adsorbed ammonia was desorbed by heating to 650 °C with a rate
of 10 °C minÀ1 under He flow (100 cm3 minÀ1). Ammonia concen-
tration was monitored continuously using a thermal conductivity
detector (TCD).
3. Results and discussion
3.1. Active metal screening for the conversion of HMF into HHD and
HXD
PXRD patterns (2h = 10–80°) of the Beta-supported metal cata-
lysts after reduction show the characteristic peaks of the corre-
sponding metallic phase (Fig. SI1). No crystalline phases of the
metal oxides precursors were observed, confirming their complete
reduction under the H2 treatment. Well-defined reflections associ-
ated with the zeolitic structures (Beta or ZSM-5) were identified
[37], verifying that crystallinity of the zeolitic support was
preserved after impregnation. The composition of the prepared
catalysts was determined by ICP-OES (Table 1), showing metal
contents close to the corresponding nominal values (Pd, Ru =
2.7–2.9 wt%; Ni, Cu, Co = 8.7–9.3 wt%).
The comparison of the catalytic performance of several zeolite-
supported metal catalysts, prepared by IWI, in the conversion of
HMF is presented in Table 1 (110 °C, 20 bar H2). Temperature
was set at 110 °C in order to minimize the extent of
Thermogravimetric analysis (TGA) was carried out on a Q600 TA
Instrument; ca. 5 mg of sample were loaded into an alumina

R. Ramos et al. / Journal of Catalysis 375 (2019) 224–233
227
Table 1
Product distribution from the conversion of HMF over zeolite-supported metal catalysts prepared by incipient wetness impregnation.(a)
M(b)
Conv.
(%)
Selectivity (%)
FDM
Cmb
(%)
(c)
Entry
Catalyst
(wt%)
THFDM
HCPN
HHD
HXD
1
2
3
4
5
6
7
8
Beta(d)
–
–
4
25
N/A
1
0
0
3
7
0
4
7
1
0
0
3
2
24
0
20
1
0
0
0
0
8
7
0
0
0
14
97
90
77
84
93
88
98
74
Beta(e)
10
80
41
21
24
15
74
Pd/Beta
Ru/Beta
Ni/Beta
Cu/Beta
Co/Beta
Pd/ZSM-5(f)
2.8
2.7
8.7
9
9.3
2.9
56
44
19
25
13
41
2
19
13
47
1
(a)
(b)
(c)
(d)
(e)
(f)
Metal chloride precursors. Reaction conditions: 0.23 gHMF, 45 ml H2O, 0.06 gcat, 110 °C, 20 bar H2, 600 rpm, 6 h.
Metal content based on ICP-OES measurements.
Carbon mass balance.
Si/Al = 12.5.
FDM as the substrate.
Si/Al = 11.5.
oligomerisation reactions which are favoured by the presence of
acidic sites [24].
(MFI in ZSM-5 vs. BEA in Beta) which affect the shape selectivity
by either mass transfer or transition state effects [44,45]. Overall,
both zeolite-supported Pd catalysts gave the highest HMF conver-
sion and HHD selectivity but also showed the lowest Cmb due to the
formation of heavier undetectable oligomers [27,28].
A preliminary control reaction with non-impregnated Beta zeo-
lite (entry 1) showed negligible HMF conversion (4%) to FDM (1%
yield), verifying that a reduced metal phase is essential for the con-
version of HMF to the targeted diketone derivatives. An additional
control experiment using the more reactive FDM intermediate as
the substrate over Beta zeolite (entry 2) resulted in 10% FDM con-
version with a concomitant colour change of the reaction mixture
from pale to dark yellow. However, no products were detected by
GC and carbon mass balance (Cmb) was only 90%. The decrease in
Cmb suggests that the highly reactive unsaturated intermediates
formed by the hydrolytic ring-opening of FDM, such as 1-
hydroxyhex-3-ene-2,5-dione (HHED) [27,28], may lead to heavier
ill-defined products, such as humins [38,39]. This undesired
oligomerisation reaction always takes place in parallel with pro-
ductive FDM conversion. It should be noted that decarboxylation
of HHED to levulinic and formic acid [40] was not observed due
to the highly reducing conditions employed.
Among the screened active metal phases, the Pd/Beta catalyst
(entry 3) afforded the highest HMF conversion (80%) and selectiv-
ity to HHD (56%), whereas HXD was also detected as a minor pro-
duct (8% selectivity). However, Cmb was only 77%, consistent with
the formation of undetectable oligomers. The Ru/Beta catalyst
(entry 4) showed lower HMF conversion (41%) and HHD selectivity
(44%). The non-noble metal based catalysts (i.e. Ni, Cu and Co,
entries 5–7) showed even lower HMF conversion (15–24%) despite
having considerably higher metal loadings. Moreover, selectivity to
HHD was rather poor (13–25%), whereas HXD was not detected. It
should be noted that the higher Cmb observed for the Ru, Ni, Cu and
Co supported catalysts (84–98%) is a direct consequence of the
lower HMF conversion. The superior catalytic activity of Pd relative
to other transition metals has also been demonstrated for the
hydrogenation of HMF to THFDM in water, using Pd/C carbon
[41] or Pd@MIL-101(Al)-NH2 MOF [42].
In order to clarify the role of water in the reaction mechanism,
an isotopic labelling experiment was performed using D2O as the
solvent under the same reaction conditions (110 °C, 20 bar H2).
GC-MS revealed the formation of [D3]-HHD and [D4]-HXD as the
main products, as well as traces of [D4]-HCPN. FDM was also
detected but it was not deuterated (Fig. SI2). Specifically, higher
m/z values were observed in the mass spectra of the products when
D2O was employed as the solvent instead of H2O: m/z = 118 ([D4]-
HCPN), 133 ([D3]-HHD) and 118 ([D4]-HXD) compared to m/z = 114
(HCPN), 130 (HHD) and 114 (HXD). This in turn suggests that two
D2O molecules participate in the catalytic mechanism, specifically
in the ring-opening of FDM via consecutive hydration-dehydration
steps (Fig. 3), as originally proposed by Horvat et al. [40]. Impor-
tantly, none of the above compounds is formed in water-free reac-
tion mixtures [7]. Therefore, H2O not only serves as an
environmentally benign solvent but is also necessary for FDM
ring-opening [46].
3.2. Influence of Pd particle size
The effect of the Pd particles size in the Beta zeolite-supported
catalysts was also investigated. In addition to the catalyst prepared
by IWI and PdCl2 (Pd/Beta), two more catalysts were prepared by
either (i) IWI and Pd(NO3)2 as the precursor (Pd(n)/Beta) or (ii)
DP-CP with urea and PdCl2 (Pd(u)/Beta). Samples were then cal-
cined and reduced as before. The different preparation methods
led to different morphologies of the supported Pd nanoparticles
(NPs), as deduced by TEM imaging (Fig. 4 and SI3) and PXRD
(Fig. SI4).
The Pd/Beta catalyst (Fig. 4a) resulted in intermediate Pd NPs
(average diameter of 5.2 3.4 nm). Employment of Pd(NO3)2 as
the precursor (Fig. 4b) led to much larger Pd NPs with a signifi-
cantly less uniform particle size distribution (average diameter of
16.2 10.4 nm). The higher dispersion of Pd catalysts with PdCl2
as the precursor has been ascribed to the formation of complex
PdxOyClz species on alumina/aluminosilicate surfaces [47,48]. The
DP-CP method (Fig. 4c) resulted in the smallest Pd NPs and the
The effect of the zeolitic support was also explored by using
ZSM-5 (Si/Al = 11.5) instead of Beta (Si/Al = 12.5) and preparing
Pd/ZSM-5 (entry 8). The latter also showed high HMF conversion
(74%) and HHD selectivity (41%), albeit slightly lower than Pd/Beta.
Notably, Pd/ZSM-5 afforded the highest selectivity to HXD (14%)
which can be attributed to the higher concentration of Brønsted
acid sites in ZSM-5 [43] and the different structural frameworks

228
R. Ramos et al. / Journal of Catalysis 375 (2019) 224–233
Fig. 3. Proposed mechanism for formation of HHD via double hydration/dehydration of the FDM intermediate [40], based on experiments carried out in D2O.
35
D = 5.2 3.4 nm
a) Pd/Beta
30
25
20
15
10
5
0
20-40
1 2 3 4 5 6 7 8 9 10 30
Diameter (nm)
35
30
25
20
15
10
5
D = 16.2 10.4 nm
b) Pd(n)/Beta
0
0
5
0
< 5
> 40
5-10
10-1515-2 20-2525-3030-3 35-4
Diameter (nm)
c) Pd(u)/Beta
35
30
25
20
15
10
5
D = 3.5 1.5 nm
0
1
2
3
4
5
6
7
8
Diameter (nm)
Fig. 4. TEM images and histograms of Pd particle size distribution (after counting >200 particles) of (a) Pd/Beta, (b) Pd(n)/Beta and (c) Pd(u)/Beta catalysts.

R. Ramos et al. / Journal of Catalysis 375 (2019) 224–233
229
most uniform size distribution (average diameter of 3.5 1.5 nm).
The smaller and more uniform particle size observed for Pd(u)/Beta
can be associated with the slow and homogenous generation of
hydroxide ions through the hydrolysis of urea at 95 °C which hin-
ders the uneven precipitation of PdII species due to a sudden, local
increase of pH [49,50].
Table 2 shows the conversion of HMF and the obtained product
distribution for the three Beta zeolite-supported Pd catalysts after
6 h under 20 bar of H2 at 110 °C. All the catalysts have similar Pd
contents varying between 2.6 and 2.8 wt% (based on ICP-OES). A
direct correlation between higher HMF conversion and smaller
Pd particle size was identified. Thus, the Pd(n)/Beta catalyst
(dM = 16.2 10.4 nm) showed the lowest HMF conversion (56%).
Moreover, the lower hydrogenation activity of Pd(n)/Beta resulted
in the lowest selectivity to HHD (39%) due to a higher degree of
oligomerisation of the unsaturated intermediates formed via
FDM ring-opening (Fig. 3). On the other hand, the Pd(u)/Beta cata-
lyst (dM = 3.5 1.5 nm) afforded almost complete HMF conversion
(96%) and the highest selectivity to HHD (56%).
The N2 adsorption-desorption isotherm of the Pd(u)/Beta cata-
lyst was similar to the parent Beta zeolite (Fig. SI5), presenting fea-
tures of Type I isotherms. The resulting textural properties showed
a slight decrease in specific surface area and pore volume after the
incorporation of Pd (621 m2 gÀ1 and 0.297 cm3 gÀ1 vs. 574 m2 gÀ1
and 0.286 cm3 gÀ1) due to partial blockage of the zeolite pores by
Pd NPs, characteristic of a highly dispersed metal phase [51]. Thus,
the enhanced hydrogenation activity with the decrease of Pd par-
ticle size can be attributed to the corresponding higher metal dis-
persion (higher metal active surface) [52] and to the higher
uniformity in Pd particle size which favours the adsorption and
hydrogenation of furanic compounds [41,42]. Additionally, well
dispersed and uniform Pd NPs increase catalytic lifetime by hinder-
ing leaching and sintering of particles [53]. In this sense, SEM-EDX
analysis of the Pd(u)/beta catalyst (Fig. SI6) confirmed the absence
of residual chlorine (from the PdCl2 precursor) which is known to
increase metal atom mobility and cause sintering.
applied to separate the oligomer fraction from the targeted com-
pounds [55,56].
3.3. Time evolution of HMF conversion
The time evolution of HMF conversion, product distribution and
Cmb over the Pd(u)/Beta catalyst are depicted in Fig. 5. HMF was
swiftly consumed, reaching 87% conversion in 120 min and 96%
in 360 min. FDM was detected at early reaction times, with a max-
imum yield of 4% at 20 min but was fully consumed in 240 min.
HHD and HXD yields rapidly increased during the first 120 min
(47% and 7%, respectively) but did not significantly change after-
wards, achieving final values of 54% (HHD) and 9% (HXD). Notably,
exposing a mixture of HHD and HXD to a fresh batch of Pd(u)/Beta
with or without H2 (T = 110 °C, 3 h) showed no interconversion
between HHD and HXD. A selectivity vs. conversion plot (Fig. SI8)
is consistent with HHD and HXD being formed via two separate
pathways (Fig. 2). Previous works have shown that hydrogenolysis
of FDM to DMF and hydrolysis of the latter can lead to HXD [21,25].
However, we did not detect any trace of DMF, indicating that HXD
is formed through a currently unidentified mechanism.
The observed time profile supports the proposed reaction
mechanism for the conversion of HMF into HHD (Fig. 3) which
begins with the hydrogenation of the HMF carbonyl group to form
FDM, followed by the hydrolytic furan ring-opening and hydro-
genation to form HHD [10,17,20,28]. The intermediate nature of
FDM was confirmed by a separate experiment using a lower
amount of catalyst (40 mg instead of 60 mg); FDM yield reached
a maximum of 23% before gradually decreasing as HHD was being
formed (Fig. SI9). The time evolution of the Cmb showed a pro-
nounced decrease within 120 min (from 100% to 73%) but
remained practically constant afterwards (71% in 360 min). This
is consistent with the formation of undetected heavier oligomers,
catalysed by the zeolite acid sites.
3.4. Influence of support Si/Al ratio
The product distribution obtained over the Pd(u)/Beta catalyst
(Table 2) revealed that at 96% HMF conversion no product
exceeded 1% selectivity apart from the targeted diketone deriva-
tives HHD and HXD. Notably, full conversion was achieved in
24 h and only two well-defined peaks corresponding to HXD and
HHD were observed in the respective GC chromatogram of the pro-
duct mixture (Fig. SI7). However, Cmb was found lower than 80% for
all runs. Taken together, these observations suggest that an unde-
tectable by GC fraction of products is produced via an acid-
catalysed oligomerisation of unsaturated intermediates, formed
via FDM ring-opening [24,27]. The formation of oligomers could
be potentially suppressed by co-addition of organic solvents [54].
Alternatively, techniques such as biphasic reactive extraction,
adsorbent-based separation or reactive distillation could be
Since the catalytic properties of the Pd(u)/Beta catalysts are
directly related to the acidity of the zeolite framework, it was
anticipated that the removal of Al atoms would affect catalyst
activity and selectivity [57]. In order to understand the influence
of the Si/Al ratio on the two competitive acid-catalysed reaction
pathways, i.e. FDM ring-opening and oligomerisation of unsatu-
rated intermediates, four Beta zeolite-supported Pd catalysts with
different Si/Al ratio were prepared (Pd(u)/Beta-dAlx, x = 1–4) via
dealumination (acid treatment) and subsequent Pd impregnation
(DP-CP with PdCl2). The parent Beta zeolite (Si/Al = 12.5) was par-
tially dealuminated by using gradually more concentrated HNO3
aqueous solutions of 0.1, 0.5, 2 and 5 M which led to increasingly
higher Si/Al atomic ratios in the range of 17.8–34.5 (Table 3).
Table 2
Product distribution from the conversion of HMF over Beta zeolite-supported Pd catalysts with different metal particle size.(a)
M(e) (wt%)
dM (nm)
Conv. (%)
Selectivity (%)
FDM
Cmb (%)
(f)
Catalyst
THFDM
HCPN
HHD
HXD
Pd/Beta(b)
2.8
2.8
2.6
5.2
16.2
3.5
80
56
96
1
2
0
3
2
1
3
2
1
56
39
56
8
7
9
77
74
71
Pd(n)/Beta(c)
Pd(u)/Beta(d)
(a)
(b)
(c)
(d)
(e)
(f)
Reaction conditions: 0.23 gHMF, 45 ml H2O, 0.06 gcat, 110 °C, 20 bar H2, 600 rpm, 6 h.
IWI and PdCl2.
IWI and Pd(NO3)2.
DP-CP with urea and PdCl2.
Metal content based on ICP-OES.
Mean metal particle diameter based on TEM images.

230
R. Ramos et al. / Journal of Catalysis 375 (2019) 224–233
110 °C vs. 53% HHD and 13% HXD at 140 °C). Further increasing
HXD
100
90
80
70
60
50
40
30
20
10
0
HHD
the temperature to 155 °C (entry 6) led to a significant decrease
of HHD selectivity, mainly due to oligomerisation (Cmb = 65%).
Having established that T = 110 °C is the optimal temperature
for the production of the targeted compounds, the effect of H2
pressure was explored. Running the reaction under 40 bar of H2
afforded again 96% HMF conversion, albeit with a slightly higher
Cmb (77%, entry 9). Increasing the H2 pressure to 60 bar resulted
in 98% HMF conversion with 66% HHD selectivity in 6 h and
100% conversion with 68% HHD selectivity in 24 h (entries 10–
11). Cmb increased to 82%, consistent with a faster hydrogenation
rate of the unsaturated intermediates (vs. oligomerisation) to form
HHD.
HCPN
THFDM
HMF
Carbon mass balance
FDM
3.6. Stability and reusability studies
0
30 60 90 120 150 180 210 240 270 300 330 360
Time (min)
The stability of the Pd(u)/Beta catalyst was examined by testing
the catalytic activity of the supernatant after physically separating
the catalyst from the reaction mixture. HMF conversion did not
increase any further (ꢀ50% conversion at 110 °C and 20 bar H2)
and product distribution did not change once the catalyst was fil-
tered off after 40 min (Fig. SI12). Likewise, Pd concentration in the
supernatant was less than 0.2 ppm (<0.5% of the total Pd content),
according to ICP-OES. Both results verify that Pd does not leach into
the solution phase and confirm the heterogeneous nature of the
catalytic system.
The reusability of the Pd(u)/Beta catalyst was investigated by
evaluating its catalytic activity upon consecutive runs (110 °C,
20 bar H2). The used catalyst was recovered after each run by filtra-
tion at room temperature, washed with deionized water and dried
at 25 °C overnight. Fig. 6a depicts the TGA curves of the fresh and
the used Pd(u)/Beta catalysts, showing a noticeable increase in the
total weight loss for the used catalyst (28% vs. 7%). Furthermore,
the elemental microanalysis of the used catalyst (Fig. 6a inset)
revealed a significant carbon content (9.32% weight), consistent
with deposition of organic compounds on the catalyst’s surface
during turnover. This result compensates to a certain extent (5–
6%) for the lower Cmb observed. The PXRD pattern of the used Pd
(u)/Beta catalyst (Fig. 6b) showed the expected reflections of the
Pd metallic phase, indicating that Pd remains reduced after turn-
over. However, TEM images of the used catalyst (Fig. 6c and
SI13) revealed an increase in Pd particle size (dM = 7.6 2.3 nm,
Fig. 6d) compared to the fresh catalyst (dM = 3.5 1.5 nm,
Fig. 4c), indicative of aggregation and formation of larger Pd parti-
cles [60].
Fig. 5. Time evolution of HMF conversion to HHD over the Pd(u)/Beta catalysts.
Reaction conditions: 0.23 gHMF, 45 ml H2O, 0.06 gcat, 110 °C, 20 bar H2, 600 rpm.
Crystallinity and microporosity of the zeolitic support were pre-
served after the dealumination treatment based on the respective
PXRD patterns and N2 isotherm profiles (Fig. SI10). Removal of Al
resulted in a progressive decrease of acid sites according to
NH3-TPD measurements (Fig. SI11 and Table SI2), due to removal
of: (i) extra-framework aluminium, associated to Lewis acidity
and (ii) tetrahedrally coordinated aluminium, associated to
Brønsted acidity [58,59]. Important differences were observed in
the catalytic performance of the Pd(u)/Beta-dAlx catalysts (Table 3).
A non-negligible increase of Cmb was observed over the dealumi-
nated supports (from 71% for Pd(u)/Beta to 79–82% for Pd(u)/
Beta-dAlx). However, a gradual decrease in HMF conversion and
selectivity to HHD was also observed as the Si/Al ratio was
increased due to the lower extent of the hydrolytic FDM ring-
opening, catalysed by Brønsted acid sites.
3.5. Influence of temperature and H2 pressure on HHD production
The effect of reaction temperature (80–155 °C) and H2 pressure
(5–60 bar) on the production of diketone derivatives from HMF
over the Pd(u)/Beta catalyst was also investigated (Table 4). Com-
parison with the original run (entry 1) revealed that lowering the
temperature below 110 °C resulted in lower HMF conversion
(entries 2–3). Raising the temperature to 125 °C or 140 °C (entries
4–5) restored HMF conversion (ꢁ95%). Notably, HCPN was also
detected as a minor product (4–5% selectivity) as the temperature
increased due to promotion of FDM ring-rearrangement, resulting
in a slight improvement of Cmb (from 71% at 110 °C to 78% at
140 °C). However, the combined selectivity of targeted HHD and
HXD was practically not affected (65 1%), although a marginal
shift towards HXD was observed (56% HHD and 9% HXD at
Recycling tests were conducted after recovering the Pd(u)/Beta
catalyst (Fig. 7a). HMF conversion gradually decreased from 96%
(1st run) to 10% (4th run) due to deposition of organic compounds
and aggregation of Pd NPs (Fig. 6 and Fig. SI13). In order to restore
the catalytic activity, the used Pd(u)/Beta catalyst (after 4 consec-
utive runs) was subjected to a regular regeneration treatment: cal-
cination (air/500 °C/5h) and reduction (H2/200 °C/5 h). Catalytic
Table 3
Catalytic conversion of HMF to HHD and HXD over dealuminated Beta zeolite-supported Pd catalysts.(a)
Si/Al(b)
dM (nm)
Conv. (%)
Selectivity (%)
FDM
Cmb (%)
(c)
Catalyst
THFDM
HCPN
HHD
HXD
Pd(u)/Beta
12.5
17.8
20.6
27.0
34.5
11
11
12
14
15
96
70
64
52
49
0
1
2
2
4
1
1
2
2
2
1
1
1
0
2
56
53
47
42
39
9
71
79
81
80
82
Pd(u)/Beta-dAl1
Pd(u)/Beta-dAl2
Pd(u)/Beta-dAl3
Pd(u)/Beta-dAl4
11
16
12
12
(a)
(b)
(c)
Reaction conditions: 0.23 gHMF, 45 ml H2O, 0.06 gcat, 110 °C, 20 bar H2, 600 rpm, 6 h.
Based on SEM-EDX measurements.
Mean metal particle diameter based on PXRD patterns and Scherrer equation.

R. Ramos et al. / Journal of Catalysis 375 (2019) 224–233
231
Table 4
Product distribution from the conversion of HMF over the Pd(u)/Beta catalyst.(a)
Entry
T (°C)
PH2 (bar)
t (h)
Conv. (%)
Selectivity (%)
FDM
Cmb (%)
THFDM
HCPN
HHD
HXD
1
2
3
4
5
6
7
8
110
80
95
20
20
20
20
20
20
5
10
40
60
60
6
6
6
6
6
6
6
6
6
6
24
96
67
88
96
95
99
69
90
96
98
100
0
3
1
0
0
1
0
0
1
1
0
1
1
1
1
1
2
0
1
1
1
2
1
0
0
4
5
7
1
1
2
1
2
56
39
57
54
53
37
48
59
61
66
68
9
3
6
10
13
14
9
8
7
71
68
71
75
78
65
73
75
77
82
82
125
140
155
110
110
110
110
110
9
10
11
7
8
(a)
Reaction conditions: 0.23 gHMF, 45 ml H2O, 0.06 gcat, 600 rpm.
(a)
Used catalyst
Fresh catalyst
(b)
100
* Pd
95
90
85
80
75
70
Elemental microanalysis
65
60
55
50
Element
Weight%
9.32
*
C
H
1.57
*
0
100 200 300 400 500 600 700 800
Temperature (°C )
10
20
30
40
50
60
70
80
2
35
(d)
(c)
30
25
20
15
10
5
D = 7.6 2.3 nm
0
0 1 2 3 4 5 6 7 8 9 101112
Diameter (nm)
Fig. 6. (a) TGA curves (air, 100 cm3 minÀ1) and C and H elemental microanalysis (inset), (b) PXRD pattern, (c) TEM image and (d) histogram of Pd particle size distribution
(after counting >200 particles) of the used Pd(u)/Beta catalyst.
activity was partially restored (5th run), affording 90% HMF con-
version to HHD (40% yield) and HXD (4% yield). A non-negligible
amount of FDM (7% yield) and HCPN (12% yield) was also detected
in the product mixture.
FDM ring-opening. Moreover, PXRD verified an increase in Pd par-
ticle size after regeneration (Fig. SI15), consistent with a lower
hydrogenation activity. Therefore, the observed differences in
activity and selectivity during recycling can be ascribed to aggrega-
tion of Pd NPs and partial loss of Brønsted acidity [61–65].
In order to investigate the observed change in selectivity, a sep-
arate set of experiments was conducted during which the catalyst
was regenerated at the end of each run (Fig. 7b). HMF conversion
gradually decreased from 96% (1st run) to 56% (3rd run) with a
concomitant increase in FDM yield (6% in 3rd run), as also observed
for the dealuminated samples (Table 3). Measurement of the 27Al
MAS NMR spectra of the fresh and the regenerated Pd(u)/Beta cat-
alyst (Fig. SI14) revealed a decrease in the fraction of tetrahedrally
coordinated Al after turnover and regeneration. This in turn sug-
gests a lower number of Brønsted acid sites [61] which promote
4. Conclusions
Diketone derivatives such as HHD and HXD were produced
from HMF over a bifunctional Beta zeolite-supported Pd catalyst
in water under relatively mild reaction conditions. The DP-CP
method afforded the most active catalyst, compared to IWI, due
to smaller and more uniformly dispersed Pd particles among the
zeolitic support. Complete conversion of HMF was achieved at

232
R. Ramos et al. / Journal of Catalysis 375 (2019) 224–233
Fig. 7. HMF conversion (lined, left column) and product yield (right column) over the Pd(u)/Beta catalyst (6 h, 110 °C, 20 bar H2): (a) consecutive runs before (runs 1–4) and
after catalyst regeneration (run 5); (b) consecutive runs after catalyst regeneration at the end of each run. Cmb is less than 100% due to formation of heavier oligomers.
[15] B. Xiao, M. Zheng, X. Li, J. Pang, R. Sun, H. Wang, X. Pang, A. Wang, S. Wang, T.
Zhang, Green Chem. 18 (2016) 2175.
[16] J. Ohyama, R. Kanao, Y. Ohira, A. Satsuma, Green Chem. 18 (2016) 676.
110 °C and 60 bar of H2 with 68% selectivity to HHD. Leaching of Pd
was not observed and catalytic activity could be partially restored
after a simple regeneration step. Selectivity to HHD is mainly lim-
ited by the formation of heavier ill-defined oligomers. The key dis-
tinguishing feature of this study is the synergic effect of the zeolite
acid sites and the highly active hydrogenating Pd metallic phase
which promotes the hydrolytic ring-opening and subsequent
hydrogenation of the FDM intermediate without necessitating co-
addition of an acid or use of an organic solvent.
[17] R. Ramos, A. Grigoropoulos, N. Perret, M. Zanella, A.P. Katsoulidis, T.D.
Manning, J.B. Claridge, M.J. Rosseinsky, Green Chem. 19 (2017) 1701.
[18] Z. Xu, P. Yan, H. Li, K. Liu, X. Liu, S. Jia, Z.C. Zhang, ACS Catal. 6 (2016) 3784.
[19] B. Wozniak, Y. Li, S. Hinze, S. Tin, J.G. de Vries, Eur. J. Org. Chem. 17 (2018)
2009.
[20] Z. Xu, P. Yan, W. Xu, X. Liu, Z. Xia, B. Chung, S. Jia, Z.C. Zhang, ACS Catal. 5
(2014) 788.
[21] D. Ren, Z. Song, L. Li, Y. Liu, F. Jin, Z. Huo, Green Chem. 18 (2016) 3075.
[22] J.J. Roylance, K.S. Choi, Green Chem. 18 (2016) 2956.
[23] V. Schiavo, G. Descotes, J. Mentech, Bull. Soc. Chim. Fr. 128 (1991) 704.
[24] Y. Nakagawa, M. Tamura, K. Tomishige, ACS Catal. 3 (2013) 2655.
[25] J. Tuteja, H. Choudhary, S. Nishimura, K. Ebitani, ChemSusChem 7 (2014) 96.
[26] G.C.A. Luijkx, N.P.M. Huck, F. van Rantwijk, L. Maat, H. van Bekkum,
Heterocycles 77 (2009) 1037.
Acknowledgements
This work was supported by the Engineering and Physical
Sciences Research Council (EPSRC), UK (EP/K014749).
[27] F. Liu, M. Audemar, K.D.O. Vigier, J.M. Clacens, F. De Campo, F. Jérôme,
ChemSusChem 7 (2014) 2089.
[28] F. Liu, M. Audemar, K. De Oliveira Vigier, J.-M. Clacens, F. De Campo, F. Jérôme,
Green Chem. 16 (2014) 4110.
[29] A. Corma, J. Catal. 216 (2013) 298.
Appendix A. Supplementary material
[30] B. Hernández, J. Iglesias, G. Morales, M. Paniagua, C. López-Aguado, J.L.G.
Fierro, P. Wolf, I. Hermans, J.A. Melero, Green Chem. 18 (2016) 5777.
[31] M.M. Antunes, S. Lima, P. Neves, A.L. Magalhães, E. Fazio, A. Fernandes, N.
Fortunato, C.M. Silva, M. Ribeiro, M. Pillinger, A. Urakawa, A.A. Valente, J. Catal.
329 (2015) 522.
Supplementary data to this article can be found online at
https://doi.org/10.1016/j.jcat.2019.04.038.
[32] B.L. Wegenhart, L. Yang, S.C. Kwan, R. Harris, H.I. Kenttämaa, M.M. Abu-Omar,
ChemSusChem 7 (2014) 2742.
References
[33] L.A.M. Hermans, J.W. Geus, Preparation of Catalysts II, Elsevier, Amsterdam,
1979, p. 113.
[34] M.E. Smith, E.R.H. van Eck, Prog. Nucl. Magn. Reson. Spectrosc. 34 (1999) 159.
[35] J. Pérez-Pariente, J. Sanz, V. Fornés, A. Corma, J. Catal. 124 (1990) 217.
[36] A. Abraham, S.-H. Lee, C.-H. Shin, S.B. Hong, R. Prins, J.A. van Bokhoven, Phys.
Chem. Chem. Phys. 6 (2004) 3031.
[37] J.B. Higgins, R.B. LaPierre, J.L. Schlenker, A.C. Rohrman, J.D. Wood, G.T. Kerr, W.
J. Rohrbaugh, Zeolites 8 (1988) 446.
[38] G. Tsilomelekis, M.J. Orella, Z. Lin, Z. Cheng, W. Zheng, V. Nikolakis, D.G.
Vlachos, Green Chem. 18 (2016) 1983.
[1] Y.C. Lin, G.W. Huber, Energy Environ. Sci. 2 (2009) 68.
[2] F. Cherubini, Energy Convers. Manag. 51 (2010) 1412.
[3] T. Werpy, G. Petersen, A. Aden, J. Bozell, J. Holladay, J. White, A. Manheim, Top
Value Added Chemicals from Biomass: Results of Screening for Potential
Candidates from Sugars and Synthesis Gas U.S. DoE Report, NREL and Pacific
Northwest National Laboratory, 2004.
[4] J.J. Bozell, G.R. Petersen, Green Chem. 12 (2010) 539.
[5] J.N. Chheda, Y. Román-Leshkov, J.A. Dumesic, Green Chem. 9 (2007) 342.
[6] R.J. van Putten, J.C. van der Waal, E.D. De Jong, C.B. Rasrendra, H.J. Heeres, J.G.
de Vries, Chem. Rev. 113 (2013) 1499.
[39] A. Gandini, M.N. Belgacem, Prog. Polym. Sci. 22 (1997) 1203.
[40] J. Horvat, B. Klaic´, B. Metelko, V. Sunjic´, Tetr. Lett. 26 (1985) 2111.
[41] H. Cai, C. Li, A. Wang, T. Zhang, Catal. Today 234 (2014) 59.
[42] J. Chen, R. Liu, Y. Guo, L. Chen, H. Gao, ACS Catal. 5 (2014) 722.
[43] F. Lónyi, J. Valyon, J. Thermochim Acta 373 (2001) 53.
[44] B. Smit, T.L. Maesen, Chem. Rev. 108 (2008) 4125.
[45] J. Jae, G.A. Tompsett, A.J. Foster, K.D. Hammond, S.M. Auerbach, R.F. Lobo, G.W.
Huber, J. Catal. 279 (2011) 257.
[46] R. Ma, X.-P. Wu, T. Tong, Z.-J. Shao, Y. Wang, X. Liu, Q. Xia, X.-Q. Gong, ACS
Catal. 7 (2017) 333.
[47] M. Haneda, Y. Kintaichi, I. Nakamura, T. Fujitani, H. Hamada, J. Catal. 218
(2003) 405.
[7] S. Chen, R. Wojcieszak, F. Dumeignil, E. Marceau, S. Royer, Chem. Rev. 118
(2018) 11023.
[8] R. Alamillo, M. Tucker, M. Chia, Y. Pagán-Torres, J.A. Dumesic, Green Chem. 14
(2012) 1413.
[9] Y. Nakagawa, K. Takada, M. Tamura, K. Tomishige, ACS Catal. 4 (2014) 2718.
[10] N. Perret, A. Grigoropoulos, M. Zanella, T.D. Manning, J.B. Claridge, M.J.
Rosseinsky, ChemSusChem 9 (2016) 521.
[11] Y. Román-Leshkov, C.J. Barrett, Z.Y. Liu, J.A. Dumesic, Nature 447 (2007) 982.
[12] O. Casanova, S. Iborra, A. Corma, ChemSusChem 2 (2009) 1138.
[13] S.E. Davis, B.N. Zope, R.J. Davis, Green Chem. 14 (2012) 143.
[14] S. Yao, X. Wang, Y. Jiang, F. Wu, X. Chen, X. Mu, ACS Sust. Chem. Eng. 2 (2013)
173.

R. Ramos et al. / Journal of Catalysis 375 (2019) 224–233
233
[48] F.B. Noronha, M.A.S. Baldanza, M. Schmal, J. Catal. 188 (1999) 270.
[49] W. Song, C. Zhao, J.A. Lercher, Chem. Eur. J. 19 (2013) 9833.
[50] P. Burattin, M. Che, C. Louis, J. Phys. Chem. B 102 (1998) 2722.
[51] R. Ramos, A. García, J.A. Botas, D.P. Serrano, Ind. Eng. Chem. Res. 55 (2016)
12723.
[57] R. Srivastava, N. Iwasa, S.I. Fujita, M. Arai, Catal. Lett. 130 (2009) 655.
[58] R. Baran, Y. Millot, T. Onfroy, J.M. Krafft, S. Dzwigaj, Microp. Mesop. Mat. 163
(2012) 122.
[59] D.M. Roberge, H. Hausmann, W.F. Hölderich, Phys. Chem. Chem. Phys. 4 (2002)
3128.
[52] P. Mäki-Arvela, J. Hájek, T. Salmi, D.Y. Murzin, Appl. Catal. A Gen. 292 (2005) 1.
[53] S.T. Gentry, S.F. Kendra, M.W. Bezpalko, J. Phys. Chem. C 115 (2011) 12736.
[54] V.V. Ordomsky, J. Van der Schaaf, J.C. Schouten, T.A. Nijhuis, J. Catal. 287 (2012)
68.
[55] H. Nguyen, R.F. DeJaco, N. Mittal, J.I. Siepmann, M. Tsapatsis, M.A. Snyder, W.
Fan, B. Saha, D.G. Vlachos, Annu. Rev. Chem. Biomol. Eng. 8 (2017) 115.
[56] P.S. Metkar, E.J. Till, D.R. Corbin, C.J. Pereira, K.W. Hutchenson, S.K. Sengupta,
Green Chem. 17 (2015) 1453.
[60] Y.F. Han, D. Kumar, C. Sivadinarayana, D.W. Goodman, J. Catal. 224 (2004) 60.
[61] Y. Xiong, W. Chen, J. Ma, Z. Chen, A. Zeng, RSC Adv. 5 (2015) 103695.
[62] S. Vitolo, B. Bresci, M. Seggiani, M.G. Gallo, Fuel 80 (2001) 17.
[63] A. Marcilla, M.I. Beltrán, R. Navarro, J. Anal. Appl. Pyrol. 74 (2005) 361.
[64] B. Puértolas, A. Veses, M.S. Callén, S. Mitchell, T. García, J. Pérez-Ramírez,
ChemSusChem 8 (2015) 3283.
[65] S. Wan, C. Waters, A. Stevens, A. Gumidyala, R. Jentoft, L. Lobban, D. Resasco, R.
Mallinson, S. Crossley, ChemSusChem 8 (2015) 552.