Towards sustainable hydrogenation of 5-(hydroxymethyl)furfural: A two-stage continuous process in aqueous media over RANEY catalysts

Article abstract of DOI:10.1039/c7ra03318d

The hydrogenation of 5-(hydroxymethyl)furfural (HMF) to 2,5-bis(hydroxymethyl)tetrahydrofuran (DHMTHF) in aqueous media under relatively mild reaction conditions has been investigated over heterogeneous RANEY Cu and Ni catalysts using a continuous-flow hydrogenation reactor. These RANEY catalysts were selected following a screening of several catalysts including precious metals supported on carbon for the hydrogenation of HMF. A single-stage versus a two-stage process for the hydrogenation of HMF into DHMTHF, i.e. via 2,5-dihydroxymethylfuran (DHMF) has been evaluated. The best result with an average selectivity of 98% for DHMTHF was obtained using a two-stage process; RANEY Cu was used as a catalyst for the highly selective hydrogenation of HMF to DHMF (92 mol%) in the first stage and this product was used without further purification for in a second-stage selective hydrogenation of DHMF into DHMTHF using RANEY Ni as a catalyst. The influence of the HMF concentration in the feeding solution (1-3 wt%), flow rate (0.05-0.25 mL min^-1) and total pressure (20-90 bar) were investigated for the first-stage hydrogenation of HMF into DHMF over RANEY Cu. HMF was found to exert an inhibiting effect on the conversion due to strong adsorption. The RANEY Ni catalyst used in the second stage gradually deactivated. A procedure for in situ regeneration of the partially deactivated RANEY Ni catalyst using acetic acid washing was investigated with limited success.

Full text of DOI:10.1039/c7ra03318d

RSC Advances
View Article Online
View Journal | View Issue
PAPER
Towards sustainable hydrogenation of 5-
hydroxymethyl)furfural: a two-stage continuous
process in aqueous media over RANEY® catalysts
(
Cite this: RSC Adv., 2017, 7, 31401
Sergio Lima,
´
David Chadwick
and Klaus Hellgardt
*
The hydrogenation of 5-(hydroxymethyl)furfural (HMF) to 2,5-bis(hydroxymethyl)tetrahydrofuran
DHMTHF) in aqueous media under relatively mild reaction conditions has been investigated over
(
heterogeneous RANEY® Cu and Ni catalysts using a continuous-flow hydrogenation reactor. These
RANEY® catalysts were selected following a screening of several catalysts including precious metals
supported on carbon for the hydrogenation of HMF. A single-stage versus a two-stage process for the
hydrogenation of HMF into DHMTHF, i.e. via 2,5-dihydroxymethylfuran (DHMF) has been evaluated. The
best result with an average selectivity of 98% for DHMTHF was obtained using a two-stage process;
RANEY® Cu was used as a catalyst for the highly selective hydrogenation of HMF to DHMF (92 mol%) in
the first stage and this product was used without further purification for in a second-stage selective
hydrogenation of DHMF into DHMTHF using RANEY® Ni as a catalyst. The influence of the HMF
ꢀ1
concentration in the feeding solution (1–3 wt%), flow rate (0.05–0.25 mL min ) and total pressure (20–
0 bar) were investigated for the first-stage hydrogenation of HMF into DHMF over RANEY® Cu. HMF
Received 21st March 2017
Accepted 12th June 2017
9
was found to exert an inhibiting effect on the conversion due to strong adsorption. The RANEY® Ni
catalyst used in the second stage gradually deactivated. A procedure for in situ regeneration of the
partially deactivated RANEY® Ni catalyst using acetic acid washing was investigated with limited success.
DOI: 10.1039/c7ra03318d
rsc.li/rsc-advances
attractive as alternative renewable polymer-building blocks are the
HMF-based di-ols 2,5-dihydroxymethylfuran (DHMF) and 2,5-
1
. Introduction
During the coming decades the world will be challenged to bis(hydroxymethyl)tetrahydrofuran (DHMTHF). DHMF is consid-
replace a major part of its use of oil based feedstocks for the ered an analogue of m/p-benzenedimethanol, which is currently
production of new commodity chemicals and fuels. A non- used for the production of phenolic and aromatic resins, homo-
petrochemical technology platform requires synthesis of next- polymers, polyesters and polyurethanes, and can be obtained by
generation as well as existing functional organic materials the selective hydrogenation of HMF, i.e. reductive hydrogenation of
10
from sustainable intermediates. In this context, particular the carbonyl group (see Scheme 1(1)). DHMTHF is considered an
attention is now given to the use of renewable biomass in the analogue of 1,3/1,4-cyclohexanedimethanol currently used for poly-
chemical industry, where lignocellulosic biomass and non- ester production and can nd several applications, for example as
edible resources have the potential to provide direct access to a solvent or an intermediate for pharmaceuticals production. It can
1–3
valuable chemicals. Ideally, this would employ continuous be obtained by total hydrogenation of HMF shown in Scheme 1. In
processes, inexpensive catalysts, and preferentially allow retro- principle, DHMTHF can be converted to 1,6-hexanediol (1,6HD) by

tting of existing industrial processes.
-(Hydroxymethyl)furfural (HMF) can be obtained from carbo- opens a route to the industrial production of polyamines such as
ring opening and hydrogenolysis or hydrodeoxygenation, which
5
3,10,11
hydrate biomass, i.e. lignocellulosic biomass, in particular from the nylon-6.
dehydration of C-6 carbohydrates (for example D-glucose and D-
Several studies have been reported concerning the catalytic
fructose – constituent lignocellulosic biomass monomers), using hydrogenation of HMF to DHMF using metal-based heteroge-
4–6
homogeneous and heterogeneous acid catalysts.
HMF is neous catalysts especially over various metals such as Au, Cu,
considered one of the most promising platform chemicals in the Ni, Pt, Pd, Ir, Ru – for example bimetallic nickel and iron sup-
biorenery, providing access to a wide range of alternative polymer- ported on carbon nanotubes catalyst or copper supported in g-
7–9
10–22
building blocks. Among the furanic intermediates that are highly alumina.
The choice of catalysts and reaction conditions
10–22
governs the product distribution.
Various organic solvents
including 1,4-dioxane, ethanol, 1-propanol or organic solvent–
water biphasic mixtures such as n-butanol/water, THF/water or
toluene/water have been used in those reported studies.
Department of Chemical Engineering, Imperial College London, South Kensington
Campus, London, SW7 2AZ, UK. E-mail: k.hellgardt@imperial.ac.uk; Tel: +44 (0)20
10–22
7
59 45577
This journal is © The Royal Society of Chemistry 2017
RSC Adv., 2017, 7, 31401–31407 | 31401

View Article Online
RSC Advances
Paper
Scheme 1 Schematic reductive hydrogenation and hydrogenolysis/hydrodeoxygenation of HMF. HMF: 5-hydroxymethylfurfural, DHMF: 2,5-
dihydroxymethylfuran, 5-MF: 5-methylfurfural, DMF: dimethylfurfural, DMTHF: 2,5-dimethyltetrahydrofuran, DHMTHF: 2,5-bis(hydroxymethyl)
tetrahydrofuran, 1,2,6-HT: 1,2,6-hexanetriol.
DHMF selectivities and HMF conversions higher than 96% continuous production of DHMTHF from fructose, in which
using 1 wt% Pt on MCM-41 (Pt/MCM-41), gold sub-nano clus- fructose is rst dehydrated to HMF, then HMF is hydrogenated
40
ters supported on alumina (Au/Al O ) or ReO modied SiO - to DHMTHF. A certain amount of DHMTHF is recovered as
2
3
x
2
supported metal catalysts with iridium (Ir–ReO /SiO ) as cata- product, while the remainder is recycled to serve as the co-
x
2
ꢁ
40
lysts have been reported at temperatures between 30 and 110 C solvent in the dehydration reaction. However, no catalytic
and pressures of 6.9 to 65 bar aer 2–6 hours using water as results were reported by the authors using a ow hydrogenation
1
8–22
40
monophasic solvent system.
A considerable number of studies have been reported for the
reactor.
In this paper we report the selective catalytic synthesis of
Due to the relatively DHMTHF from HMF using a continuous hydrogenation reactor
11,22,23,23–30
synthesis of DHMTHF from HMF.
stable furan ring, typically harsher conditions are required (i.e. system. A single-stage process to DHMTHF and a two-stage
higher pressure and temperatures) compared to the reduction process via the selective synthesis of the intermediate DHMF
23
of the aldehyde group of HMF to form DHMF. Common have been investigated. These studies have been performed in
ꢁ
materials investigated as heterogeneous catalysts include non- aqueous media at the relatively mild conditions of 90 C and 90
noble and noble metals, especially bifunctional catalysts bar H
derived from hydrotalcite-like compounds, bimetallic catalysts catalysts has been compared with common supported precious
supported on silica as Ni–Pd/SiO and Pd–Ir/SiO and RANEY®- metal catalysts. The catalysts were in the form of commercial
type metal catalysts (Co, Cu, Ni) and others using different CatCart® supplied by ThalesNano Inc. RANEY® catalysts have
2
. The performance of heterogeneous RANEY® Cu and Ni
2
2
11,17,22–26,28
monophasic organic solvents.
the advantage of being inexpensive compared to precious metal
RANEY®-type metals (Cu, Co and Ni) have been investigated catalysts. We show that the two-stage process using RANEY® Cu
as catalyst for total hydrogenation of HMF using 1,4-dioxane as in the rst stage and RANEY® Ni in the second stage is the most
25
solvent by Kong et al. Among the RANEY®-type metals, efficient system for selective synthesis of DHMTHF from HMF
RANEY® Ni was revealed to be the most selective with a 96% at the mild conditions used. A pathway for in situ regeneration
ꢁ
DHMTHF selectivity at full conversion, at 100 C and 15 bar of of RANEY® Ni catalyst is also proposed in this paper.
25
2
H aer 15 h of reaction. RANEY® Ni has also been success-
fully used to catalyse the hydrogenation of HMF into DHMTHF
in other solvents such as methanol or ethanol with yields higher
2
. Experimental section
11,26
than 95%.
The best result was reported by Connolly et al.
5-(Hydroxymethyl)furfural was purchased from Sigma-Aldrich
using methanol as solvent, achieving 99% selectivity to
and used as received without further purication. For the
catalytic reactions a xed-bed bench-top continuous-ow
hydrogenation reactor, H-CUBE® model HC 2, developed by
ThalesNano Inc. (Budapest, Hungary) was used.
pre-packed heterogeneous catalyst cartridges (CatCat®) of 4
DHMTHF and over 95% HMF conversion under relatively mild
ꢁ
reaction conditions, 60 C under 4.8 bar of H
operation.
2
aer 4 h of batch
26
35,36
Commercial
All the above research has been conducted in batch reactors.
To our knowledge only three papers have addressed the
synthesis of DHMF or DHMTHF using heterogeneous catalytic
continuous hydrogenation of HMF, although indirectly, despite
the potential advantages of catalytic continuous-ow hydroge-
nation for large-scale industrial production compared with
mm inner diameter and 30 mm length (tubular) lled with ca.
140 mg of catalyst (particle size of 50–400 mm) were used. The
general protocol was as follows: the CatCart® was placed
vertically in the H-Cube system cartridge holder and rinsed with
degassed, deionised water followed by activation by pretreat-
ment with water saturated with hydrogen through the CatCart®
3
1–39
batch reactor processes.
bifunctional oxides with Brønsted or Lewis acidy such as g-
Al , ZrO , TiO , Al /SBA-15, ZrO /SBA-15, TiO /SBA-15, H-
BEA, Sn-BEA ZrO , and supported metals such as Fe -sup-
ported Pd, and Ni, Pd, and Cu supported on g-alumina.
Tucker et al. proposed a tandem catalytic approach to
The catalysts studied include
ꢀ1
by an HPLC-pump (ow rate of 0.25 mL min ) for 30 min at the
2
O
3
2
2
2
O
3
2
2
selected temperature and pressure, usually 90 bar. The catalytic
reaction was performed by feeding a continuous ow of the
substrate solution saturated with hydrogen. The performance of
the selected CatCart® was evaluated by analysing the liquid
2
2 3
O
31–39
31402 | RSC Adv., 2017, 7, 31401–31407
This journal is © The Royal Society of Chemistry 2017

View Article Online
Paper
RSC Advances
samples collected on an hourly basis aer the start of the catalysts. These included precious metals supported on carbon
experimental run feeding the substrate solution. Steady state and SiO , heterogeneous RANEY® catalysts (Ni, Cu), and Ni/
2
was reached aer 1 h of time on stream. The average conver- SiO –Al O , 5% Rh/Al O as CatCart® catalysts. The reactions
2
2
3
2 3
sions and product selectivities were calculated, for example over were performed using aqueous solutions of 1 wt% HMF as
ꢀ
1
the period of 1–8 h, by averaging the point conversions and feeding solution (feed ow rate of 0.05 mL min ) under mild
ꢁ
selectivities obtained from the liquid samples collected at each reaction condition (90 C, 90 bar H
2
) and a liquid hourly space
ꢀ
1
hour interval from 1 to 8 h time-on-stream (7 points). The velocity (LHSV) of 7.95 h , calculated considering the internal
ꢀ7
3
product solutions were analysed by a GC (Hewlett-Packard HP cartridge volume ¼ 3.77 ꢂ 10 m , equivalent to a mean liquid
6
890 series) equipped with a 30 m StabilWAX(R)-DA column (30 residence time for this feed ow rate of 4.16 min based on the
m length, 0.32 mm internal diameter, and 0.5 mm lm thick- average dead volume for 30 mm lled CatCart ¼ 0.208 mL (value
ness) and a ame ionization detector (FID), injection volume 0.5 provided by ThalesNano). The conversion of HMF to DHMF,
ꢁ
ꢁ
mL, inlet temperature 250 C, detector temperature 250 C, and DHMTHF or by-products using Darco® CatCart® (carbon black)
ꢁ
a split ratio 1 : 5. The initial column temperature was 50 C (5 as blank at these conditions was negligible aer 8 h on stream.
min) with a temperature rise of 10 C min and nal temper-
ature was 240 C. Nitrogen was used as the carrier gas. Each catalysts are given in Table 1. The average conversion of HMF was
ꢁ
ꢀ1
The average conversion and product selectivity for the studied
ꢁ
peak of the GC chromatogram was properly integrated and the in range of 70–100% for most of the catalysts, with the exception of
actual concentration of each product was obtained from a pre- Ni/SiO –Al O , 5% Ru/C (Table 1, entry 1 and 4). In general, the
2 2 3
calibrated plot of peak area against concentrations. Phenol catalysts investigated exhibited poor selectivity for the single-stage
was used as internal standard. Conversion, selectivities and synthesis of DHMTHF. Average selectivity from 1–8 h on stream
yields were calculated with the following eqn (eqn (1)–(3)):
was less than 20% for DHMTHF, showing that these catalysts are
not effective for a continuous single-stage synthesis of DHMTHF
from HMF in aqueous solution under the reaction conditions
investigated. The mass balance are not closed for most of catalysts
investigated, except for Ni/SiO –Al O and Rh/Al O , which can be
2 2 3 2 3
consequence of either formation of insoluble by-products and/or
because not all by-products formed were detected by the GC-FID
mole HMF inlet ꢀ mole HMF outlet
Conversion ð%Þ ¼
Selectivity ð%Þ ¼
ꢂ 100
1)
mole HMF inlet
(
mole product
ꢂ 100
mole HMF inlet ꢀ mole HMF outlet
(
2) analysis. RANEY® Ni catalyst gave a 13% average selectivity to
DHMTHF which corresponded to an average productivity of 0.2
ꢀ
1
ꢀ1
mole product
mmolDHMTHF.gcatalyst
h
aer 8 h of operation. Nevertheless,
Yield ð%Þ ¼
ꢂ 100
(3)
mole HMF inlet
RANEY® Cu as catalyst (Table 1, entry 3) showed a stable average
selectivity to DHMF of 84% at and an average HMF conversion of
9
4%. This result suggested that a two-stage continuous process
Inductively coupled plasma atomic emission spectroscopy
ICP-AES) analyses were performed using a ICP-OES spectrom-
might prove attractive.
(
eter Perkin-Elmer Optima 2000 DV ICP.
3.2 Two-stages catalytic hydrogenation of HMF
3
. Results and discussion
3
.2.1 First-stage reaction: catalytic hydrogenation of HMF
into DHMF over RANEY® Cu. The hydrogenation of HMF to
DHMF over the RANEY® Cu catalyst was explored in more detail
with regard to use in a possible rst stage. The catalyst proved to
exhibit exceptional activity and selectivity toward DHMF over
4 h on stream at 90 C and 90 bar pressure, with an average
conversion of 94% and 92% selectivity, Fig. 1. The average
selectivity obtained over 24 h (92%) is slightly higher than that
observed over 8 h on stream (84%), mainly due to the higher
consistency (i.e. less uctuation) of calculated selectivities over
a longer period of time on stream. The DHMF moles formed per
mole of Cu (RANEY® Cu: 91 wt% Cu), i.e. the turnover number
The reaction network for the catalytic hydrogenation of HMF is
given in Scheme 1. The reactions of interest in the present study
are HMF to DHMF (reaction 1) and DHMF to DHMTHF (reac-
tion 2). The hydrogenolysis/hydrodeoxygenation of HMF to 5-
methylfurfural (5-MF), and the further conversion of DMF,
DHMF and DHMTHF to dimethylfurfural (DMF), 2,5-dime-
thyltetrahydrofuran (DMTHF) among others are herein referred
to as giving rise to by-products.
The present catalytic hydrogenation results were obtained in
aqueous media using heterogeneous pre-packed catalyst
cartridges in a continuous-ow hydrogenation bench-top
reactor (H-Cube) as noted above, where molecular hydrogen
was generated in situ by the electrolysis of water and continu-
ously mixed into the owing aqueous solution of feed, for
example HMF.
ꢁ
2
(
TON), was low, 2.4 aer 24 h of time on stream due to the low
concentrations used. Operating for longer periods of time would
increase product/catalyst ratio, although this depends on catalyst
stability over extended periods of operation. By-products from
over-hydrogenation or hydrogenolysis/hydrodeoxygenation of
DHMF (e.g. DMF) were not observed using RANEY® Cu as
catalyst. ICP-OES analyses of the recovered solutions aer
3.1 Single-stage catalytic hydrogenation of HMF
The single-stage hydrogenation of HMF to DHMTHF was 24 h of on-stream operation showed that no measurable Cu
investigated using a range of commercially available cartridge or Al leaching occurred.
This journal is © The Royal Society of Chemistry 2017
RSC Adv., 2017, 7, 31401–31407 | 31403

View Article Online
RSC Advances
Paper
ꢁ
Table 1 Performance of commercial catalysts (CatCart®) in the catalytic hydrogenation reaction of HMF and selected bio-products at 90 C and
a
9
0 bar
b,c
Bio-products selectivity (%)
b
d
Entry
Catalyst
Ni/SiO –Al
RANEY® Ni
RANEY® Cu
5% Ru/C
Conv. (%)
DMF
5-MF
DMTHF
DHMF
DHMTHF
Total yield (%)
1
2
3
4
5
6
7
8
9
2
2
O
3
52
100
94
100
96
69
99
100
100
74
—
—
—
—
—
—
1 (1)
—
—
—
2 (1)
3 (3)
9 (8)
—
4 (4)
5 (3)
10 (10)
2 (2)
—
—
—
—
4 (4)
7 (7)
—
3 (3)
4 (4)
21 (21)
7 (5)
93 (48)
60 (60)
84 (79)
27 (27)
30 (30)
78 (54)
44 (44)
17 (17)
—
5 (3)
13 (13)
—
18 (18)
6 (6)
—
52
76
87
49
45
57
58
39
38
74
4.5% Ru 0.5% Pd/C
1% Pt/SiO
10% Pt/C
2
1 (1)
5% Pd/SiO
10% Pd/C
2
16 (16)
17 (17)
7 (5)
1
0
5% Rh/Al
2
O
3
5 (4)
81 (60)
a
ꢀ1
ꢁ
b
Reaction conditions: [HMF]
0
¼ 1 wt% in water, 0.05 mL min , 90 C, 90 bar h. The average HMF conversion, average selectivities, and average
yields were calculated by averaging the moles of HMF or bio-products obtained from the GC-FID analysis of the liquid samples collected at each
c
hour from 1 h to 8 h of time on-stream (7 points). The steady state was reached aer 1 h of time on stream. In brackets is provided the yield.
d
Total yield was calculated as the sum of the yields for the identied by-products.
conversions reported in Fig. 2(a) are not limited by thermody-
namic equilibrium based on the temperature and pressure
dependence of the thermodynamic equilibrium constant for the
45,46
hydrogenation of HMF to DHMF as previously reported.
The inuence of the ow rate on the hydrogenation of HMF
at a concentration of 1 wt% was investigated in the range of
ꢀ1
ꢀ1
ꢁ
0
.05–0.25 mL min (LHSV: 7.95–39.78 h ) at 90 C and 90 bar
by using the RANEY® Cu as catalyst (Fig. 2(b)). As the ow rate
ꢀ1
increases in the range of 0.05–0.25 mL min the conversion of
2
HMF decreases, as expected. An excellent linear correlation (R
¼
0.99) was observed, being consistent with a zero-order kinetic
regime in the concentration of HMF. In a similar way, the
hydrogenation reaction of furfural to furfuryl alcohol in the
liquid-phase, was reported to be negative rst-order to zero-
order in furfural concentration (0.025–0.1 M), and close to

rst-order with respect to hydrogen pressure, in the tempera-
Fig. 1 Conversion of HMF into DHMF over RANEY® Cu catalyst.
ꢁ
46
ture range of 50–90 C. However, closer to rst-order depen-
dence was observed with respect to both HMF and hydrogen,
reported for the hydrogenation of HMF to DHMF in aqueous
media in the concentration range 0.05–0.3 M and temperature
ꢀ1
Reaction conditions: flow rate of 0.05 mL min , 1 wt% HMF in water,
ꢁ
9
0 bar, 90 C.
ꢁ
21
range 40–70 C, over Ru/C.
The effect of total pressure of the system on the HMF
The inuence of the HMF concentration in the feeding
solution was also investigated. Fig. 2(c) shows the results for the
hydrogenation of HMF to DHMF with feed concentrations in
conversion to DHMF was investigated in range of 20–100 bar
ꢁ
over the RANEY® Cu catalyst at 90 C, Fig. 2(a). Each point in
the plot corresponds to an average value obtained aer 8 h of
ꢁ
the range of 1–3 wt% (i.e. from 8.2 to 24.7 mM) at 90 C and 90
2
on-stream operation. An excellent linear correlation (R ¼ 0.99)
bar total pressure over RANEY® Cu. As the concentration of
HMF increases from 1 to 3 wt%, the conversion of HMF
decreases, as expected due to strong adsorption of HMF, and
matches the observed ow rate dependence. Interestingly, in
the range of HMF concentrations investigated, the selectivity to
DHMF (average ca. 92%) did not change signicantly with the
increase of the HMF concentration in the feeding aqueous
solutions.
for the reaction activity was observed up to 70 bar (ow rate:
ꢀ
1
ꢁ
0
.05 mL min , 90 C, 1 wt% HMF), showing that the hydro-
genation of HMF is strongly dependent on the hydrogen
concentration up to this pressure (Fig. 2(a)), due to the limiting
stoichiometric hydrogen availability up to this pressure (mole
fraction of hydrogen per moles of HMF in aqueous phase ¼ 0.74
41,42
at 70 bar and 0.96 at 90 bar).
Similar results were reported by
Tukacs et al. for the hydrogenation of levulinic acid to g-valer-
olactone using Ru/C® and butyl-bis-(m-sulfonated-phenyl)
3.2.2 Second-stage reaction: catalytic hydrogenation of
DHMF to DHMTHF over RANEY® Ni. RANEY® Ni as catalyst for
the hydrogenation of DHMF to DHMTHF as a second stage in
43
phosphine ligand as co-catalyst in aqueous media. The
31404 | RSC Adv., 2017, 7, 31401–31407
This journal is © The Royal Society of Chemistry 2017

View Article Online
Paper
RSC Advances
and 90 bar of total pressure) were used aer have been mixed,
without further purication as feeding solution for the second-
stage catalytic experiments over RANEY® Ni.
The RANEY® Ni catalyst exhibited exceptional selectivity
toward DHMTHF, with an average selectivity of 98% and a high
ꢀ
1
ꢀ1
productivity ¼ 19 mmolDHMTHF.gcatalyst
operation, Fig. 3. Measurement of the amount of leaching gave
.9 ppm of Ni and 0.023 ppm of Al in the product liquid cor-
h
aer 15 h of
4
responding to the total mass of Ni and Al leached aer 15 h of
operation of 0.18 wt% and <0.01 wt% respectively, based on the
initial specied composition of RANEY® Ni alloy (85 wt%
Ni min., 12 wt% Al max.) and 140 mg of catalyst lled in the
cartridge. Other two well-known hydrogenating catalysts, 10%
Pd/C and 5% Rh/C were also investigated as catalysts for the
selective hydrogenation of DHMF to DHMTHF in aqueous
phase under the same operation reaction conditions (temper-
ature, pressure and feeding ow rate). An average DHMTHF
selectivity less than 5% was observed for 10% Pd/C aer 8 h of
time on stream and 5% Rh/C catalyst completely deactivated
aer 2 h (result not shown). When RANEY® Ni was used as
catalyst for the single-stage conversion of HMF into DHMTHF
(Table 1, entry 2) an average selectivity of only 13% was obtained
aer 8 h of time on stream.
The effect of hydrogen pressure on the DHMF conversion
into DHMTHF was also investigated in the range of 20–90 bar
ꢁ
using RANEY® Ni at 90 C, Fig. 4. The hydrogenation is not
44,45
limited by thermodynamic equilibrium at these conditions.
As can be observed, the highest catalytic performance (conver-
sion and selectivity) for the conversion of DHMF to DHMTHF
(Scheme 1(2)) was reached at the same operation pressure as the

rst-stage reaction, i.e. 90 bar, which is convenient from
a process stand point.
While the selectivity did not change signicantly over the
operating time investigated, shown in Fig. 3, the DHMF conversion
gradually decreased over time. Aer 32 h of operation (result not
ꢀ1
Fig. 2 Effect of: (a) total pressure (flow rate: 0.050 mL min , 1 wt%
HMF), (b) feed flow rate (pressure: 90 bar, 1 wt% HMF in water) and (c)
ꢀ1
concentration of substrate (flow rate: 0.05 mL min , pressure: 90 bar)
on the HMF hydrogenation to DHMF in the presence of RANEY® Cu at
ꢁ
9
0 C in aqueous media. Each point correspond to an average after 8 h
of time on stream.
the hydrogenation of HMF to DHMTHF has been investigated.
The product aqueous solutions collected over 24 h of operation
at the outlet of the reactor obtained from the experimental runs Reaction conditions: flow rate of 0.05 mL min , 90 bar, 90 C. Feed
Fig. 3 Conversion of DHMF into DHMTHF over RANEY® Ni catalyst.
ꢀ1
ꢁ
solution: aqueous solution of DHMF (molar ratio DHMF/HMF ¼ 0.92)
obtained from the catalytic experiment using RANEY® Cu (first-stage
reaction) under the similar operation conditions.
performed using RANEY® Cu as catalyst (and 1 wt% HMF
aqueous solution, feeding ow rate of 0.05 mL min at 90 C
ꢀ1
ꢁ
This journal is © The Royal Society of Chemistry 2017
RSC Adv., 2017, 7, 31401–31407 | 31405

View Article Online
RSC Advances
Paper
ꢁ
47
higher than 120 C aer a very severe washing procedure. In
order to recover the RANEY® Ni activity we used a procedure based
on early literature reported by Hauschild et al. to perform the
regeneration in situ, i.e. without unpacking the catalyst from the
48
cartridge. The spent RANEY® Ni cartridge was ushed in situ
with a degassed, alcoholic solution of acetic acid (acetic acid in
methanol/water) saturated with hydrogen (generated in situ by
electrolysis of water) at a ow rate of 1 mL min at 40 C and 50
bar for 2 h, followed by washing the CatCart® with deionised water
saturated with hydrogen for another 3 h under the same
conditions.
ꢀ1
ꢁ
The activity of catalyst treated according to the above
procedure was not comparable to the activity of the fresh cata-
lyst, even aer repeating the severe washing procedure two
times, as shown in Fig. 5. These results show that regeneration
of RANEY® Ni was achieved with only limited success. Harsher
conditions than those that can be achieved in the H-Cube
Fig. 4 Effect of total pressure on the hydrogenation of DHMF into
DHMTHF over RANEY® Ni (reaction conditions: flow rate of 0.050 mL
ꢀ1
ꢁ
min , 1 wt% HMF in water, 90 C). Feed solution: aqueous solution of hydrogenation reactor, for example higher temperature under
DHMF (molar ratio DHMF/HMF ¼ 0.92) from the catalytic reaction
using RANEY® Cu as catalyst under similar operation conditions. Each
point correspond to an average after 8 h of time on stream.
hydrogen, might be needed to effectively regenerate the Ni
surface aer being washed.
47
4
. Conclusions
shown), the conversion decreased to ca. 71% of the initial value.
Stability of catalyst performance is of extreme importance in
industrial processing. From previous studies the main reasons for
deactivation of RANEY®-type catalysts are considered to be: (1) loss
of active Ni surface by sintering; (2) leaching of Ni and promoter
metal into the acidic and chelating reaction mixture. However, the
negligible Ni and Al leaching measured aer 15 h of operation
suggests that the observed deactivation of RANEY® Ni may not be
a result of loss of metal and/or promoter species; (3) poisoning of
the active Ni surface by organic species principally produced by
DHMTHF was obtained efficiently from the hydrogenation of HMF
in aqueous media using a continuous two-stage catalytic process
via DHMF under relatively mild reaction conditions, 90 C at 90
ꢁ
bar. An average selectivity of 98% was achieved using different non-
noble catalysts in each stage: RANEY® Cu to hydrogenate HMF to
DHMF and RANEY® Ni to hydrogenate DHMF to DHMTHF under
the same reaction conditions (productivity ¼ 19 mmolDHMTHF.-
ꢀ1
ꢀ1
g
catalyst
h
aer 15 h of operation). This approach was shown to
be preferable to a single-stage process using RANEY® Ni catalyst
where only 13% average selectivity to DHMTHF was obtained
47
side reactions. Hoffer et al. showed that RANEY®-type Ni cata-
lysts used in hydrogenation of aqueous solutions of D-glucose can
be effectively regenerated by a hydrogen treatment at temperatures
ꢀ
1
ꢀ1
(average productivity ¼ 0.2 mmolDHMTHF.gcatalyst
h
aer 8 h of
operation). The rst-stage hydrogenation of HMF to DHMF over
the RANEY® Cu catalyst proved to give stable performance, but the
RANEY® Ni in the second stage gradually deactivated while
maintaining high selectivity to DHMTHF. Negligible Ni and Al
leaching occurred over 15 h of operation, suggesting that the
observed decrease of activity over time-on-stream might be a result
of either poisoning of the active Ni surface by organic species or
due to irreversible oxidation of the active Ni surface species.
Regeneration of the RANEY® Ni catalyst by washing the spent
catalyst with an alcoholic aqueous solution of acetic acid in the
presence of hydrogen had only a partial success. The conditions for
the regeneration are limited by the H-Cube which would not
necessarily apply to an industrial process.
The two-stage approach described here can be an important
contribution to the development of sustainable catalytic routes
to DHMTHF production, where continuous hydrogenation is
required for large-scale production of this high value interme-
diate for polymer manufacture.
Fig. 5 Conversion of DHMF into DHMTHF over RANEY® Ni in
aqueous media. Comparison of DHMF conversion after the CatCart®
have been regenerated/washed (after regeneration) and after the same
CatCart® have been regenerated for second time (after regeneration
for the second time).
Acknowledgements
This work was nancially supported by EPSRC(UK) under grant
EP/K014749/1.
31406 | RSC Adv., 2017, 7, 31401–31407
This journal is © The Royal Society of Chemistry 2017

View Article Online
Paper
RSC Advances
A. Sharma and K. Sutherland, Org. Process Res. Dev., 2010,
References
14, 459–465.
1
2
3
4
5
6
7
G. W. Huber, S. Iborra and A. Corma, Chem. Rev., 2006, 27 J. Chen, R. Liu, Y. Guo, L. Chen and H. Gao, ACS Catal., 2015,
06(9), 4044–4098.
5, 722–733.
J. N. Chheda, G. W. Huber and J. A. Dumesic, Angew. Chem., 28 S. Yao, X. Wang, Y. Jiang, F. Wu, X. Chen and X. Mu, ACS
Int. Ed., 2007, 46(38), 7164–7184.
Sustainable Chem. Eng., 2014, 2, 173–180.
G. Berndes and J. Hansson, Energy Policy, 2007, 35(12), 5965– 29 Y. Nakagawa and K. Tomishige, Catal. Commun., 2010, 12,
979.
154–156.
P. K. Rout, A. D. Nannaware, O. Prakash, A. Kalra and 30 Y. Nakagawa, K. Takada, M. Tamura and K. Tomishige, ACS
R. Rajasekharan, Chem. Eng. Sci., 2016, 142, 318–346.
Catal., 2014, 4, 2718–2726.
A. Chinnappan, C. Baskar and H. Kim, RSC Adv., 2016, 6, 31 J. Luo, J. Yu, R. J. Gorte, E. Mahmoud, D. G. Vlachos and
3991–64002.
M. A. Smith, Catal. Sci. Technol., 2014, 4, 3074–3081.
I. Agirrezabal-Telleria, I. Gandarias and P. L. Arias, Catal. 32 D. Scholz, C. Aellig and I. Hermans, ChemSusChem, 2014, 7,
Today, 2014, 234, 42–58.
268–275.
L. Hu, G. Zhao, W. Hao, X. Tang, Y. Sun, L. Lin and S. Liu, 33 D. P. Duarte, R. Mart ´ı nez and L. J. Hoyos, Ind. Eng. Chem.
1
5
6
RSC Adv., 2012, 2, 11184–11206.
Res., 2016, 55, 54–63.
8
9
K. Yan and A. Chen, Fuel, 2014, 115, 101–108.
Y. Rom ´a n-Leshkov, J. N. Chheda and J. A. Dumesic, Science,
34 B. Xiao, M. Zheng, X. Li, J. Pang, R. Sun, H. Wang, X. Pang,
A. Wang, X. Wang and T. Zhang, Green Chem., 2016, 18,
2175–2184.
2006, 321, 1933–1938.
1
1
0 I. Delidovich, P. J. C. Hausoul, L. Deng, R. Pf u¨ tzenreuter, 35 N. G. Anderson, Org. Process Res. Dev., 2012, 16, 852–869.
M. Rose and R. Palkovits, Chem. Rev., 2016, 116, 1540–1599. 36 R. Jones, L. G ¨o d ¨o rh ´a zy, N. Varga, D. Szalay, L. Urge and
1 T. Buntara, S. Noel, P. H. Phua, I. Meli ´a n-Cabrera, J. G. de
F. Darvas, J. Comb. Chem., 2006, 8, 110–116.
Vries and H. J. Heeres, Angew. Chem., Int. Ed., 2011, 50, 37 E. A. Artiukha, A. L. Nuzhdin, G. A. Bukhtiyarova,
¨
7
083–7087.
Y. S. Zaytsev, P. E. Plyusnin, Y. V. Shubin and
V. I. Bukhtiyarov, Catal. Sci. Technol., 2015, 5, 4741–4745.
38 M. C. Bryan, D. Wernick, C. D. Hein, J. V. Petersen,
J. W. Eschelbach and E. M. Doherty, Beilstein J. Org. Chem.,
2011, 7, 1141–1149.
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2 Y. Zhu, X. Kong, H. Zheng, G. Ding, Y. Zhu and Y.-W. Li,
Catal. Sci. Technol., 2015, 5, 4208–4217.
3 G. C. A. Luijkx, N. P. M. Huck, F. van Rantwijk, L. Maat and
H. van Bekkum, Heterocycles, 2009, 77, 1037–1044.
4 A. J. Kumalaputri, G. Bottari, P. M. Erne, H. J. Heeres and 39 P. J. Cossar, L. Hizartzidis, M. I. Simone, A. McCluskey and
K. Barta, ChemSusChem, 2014, 7, 2266–2275.
C. P. Gordon, Org. Biomol. Chem., 2015, 13, 7119–7130.
5 L. Yu, L. He, J. Chen, J. Zheng, L. Ye, H. Lin and Y. Yuan, 40 M. Tucker, H. R. Alamillo, A. J. Crisci, G. M. Gonzalez,
ChemCatChem, 2015, 7, 1701–1707.
6 Y. Liu, M. A. Mellmer, D. M. Alonso and J. A. Dumesic,
ChemSusChem, 2015, 8, 3983–3986.
7 R. Alamillo, M. Tucker, M. Chia, Y. Pag ´a n-Torres and
J. A. Dumesic, Green Chem., 2012, 14, 1413–1419.
S. L. Scott and J. A. Dumesic, ACS Sustainable Chem. Eng.,
2013, 1, 554–560.
41 E. L. Stephan, N. S. Hateld, R. S. Peoples and H. A. H. Pray,
USAEC Report BMI-1067, Battelle Memorial Institute,
Columbus, Ohio, 1956.
8 M. Chatterjee, T. Ishizaka and H. Kawanami, Green Chem., 42 H. A. Pray, C. E. Schweickert and B. H. Minnich, Ind. Eng.
014, 16, 4734–4739.
Chem., 1952, 44(5), 1146–1151.
9 M. Tamura, K. Tokonami, Y. Nakagawa and K. Tomishige, 43 J. M. Tukacs, R. V. Jones, F. Darvas, G. Dib ´o , G. Lezs ´a k and
Chem. Commun., 2013, 49, 7034–7036.
L. T. Mika, RSC Adv., 2013, 3, 16283–16287.
0 J. Ohyama, A. Esaki, Y. Yamamoto, S. Arai and A. Satsuma, 44 R. S. Assary, P. C. Redfern, J. R. Hammond, J. Greeley and
RSC Adv., 2013, 2, 1033–1036.
L. A. Curtiss, Chem. Phys. Lett., 2010, 497, 123–128.
1 A. B. Jain and P. D. Vaidya, Int. J. Chem. Kinet., 2016, 48, 318– 45 S. P. Verevkin, V. N. Emel'yanenko, E. N. Stepurko,
2
3
28.
2 A. J. Sanborn and P. D. Bloom, US Pat. number 7393963 B2,
008.
R. V. Ralys and D. H. Zaitsau, Ind. Eng. Chem. Res., 2009,
48, 10087–10093.
46 H. Rojas, J. J. Mart ´ı nez and P. Reyes, Dyna, 2010, 163, 151–
159.
47 B. W. Hoffer, E. Crezee, F. Devred, P. R. M. Mooijman,
W. G. Sloof, P. J. Kooyman, A. D. van Langeveld,
F. Kapteijn and J. A. Moulijn, Appl. Catal., A, 2003, 253,
437–452.
48 U. Hauschild and H. Nicolaus, US Pat. number 3165478 A,
1965.
2
3 K. Weissermel and H.-J. Arpe, Industrielle Organische Chemie,
VCH, Weinheim, 4th edn, 1994.
4 X. Kong, R. Zheng, Y. Zhu, G. Ding, Y. Zhu and Y.-W. Li,
Green Chem., 2015, 17, 2504–2515.
5 X. Kong, Y. Zhu, H. Zheng, F. Dong, Y. Zhu and Y.-W. Li, RSC
Adv., 2014, 4, 60467–60472.
6 T. J. Connolly, J. L. Considine, Z. Ding, B. Fordatz,
M. N. Jennings, M. F. MacEwan, K. M. McCoy, D. W. Place,
This journal is © The Royal Society of Chemistry 2017
RSC Adv., 2017, 7, 31401–31407 | 31407