Kinetics of Catalytic Hydrogenation of 5-Hydroxymethylfurfural to 2,5-bis-Hydroxymethylfuran in Aqueous Solution over Ru/C

Article abstract of DOI:10.1002/kin.20992

5-Hydroxymethylfurfural (5-HMF) is a cellulosic product of the hydrolysis of biomass, and it is widely considered for the production of several interesting chemicals and derivatives. In the present work, catalytic hydrogenation of 5-hydroxymethylfurfural to 2,5-bis-hydroxymethylfuran was investigated using 5% Ru/C in the aqueous phase. Kinetic data were experimentally obtained over a wide range of temperatures (313-343 K), H₂ partial pressure (0.69-2.07 MPa), initial HMF concentration (19.8-59.5 mM), and catalyst loading (0.3-0.7 kg/m³) in a three-phase slurry reactor. Disappearance of initial 5-HMF concentrations was modeled using the power law and Langmuir-Hinshelwood-Hougen-Watson models. A model based on the competitive adsorption of molecular H₂ and HMF was proposed. It is presumed that surface reaction between nondissociatively chemisorbed H₂ and 5-HMF was rate determining. This model provided the best fit for the kinetic data. From the Arrhenius equation, the activation energy for the surface reaction was found to be 104.9 kJ/mol.

Full text of DOI:10.1002/kin.20992

Kinetics of Catalytic
Hydrogenation of
5-Hydroxymethylfurfural to
2,5-bis-Hydroxymethylfuran
in Aqueous Solution over
Ru/C
ANANDKUMAR B. JAIN, PRAKASH D. VAIDYA
Department of Chemical Engineering, Institute of Chemical Technology, Matunga, Mumbai, 400 019, India
Received 16 October 2015; revised 26 February 2016; accepted 1 March 2016
DOI 10.1002/kin.20992
Published online 28 March 2016 in Wiley Online Library (wileyonlinelibrary.com).
ABSTRACT: 5-Hydroxymethylfurfural (5-HMF) is a cellulosic product of the hydrolysis of
biomass, and it is widely considered for the production of several interesting chemicals and
derivatives. In the present work, catalytic hydrogenation of 5-hydroxymethylfurfural to 2,5-bis-
hydroxymethylfuran was investigated using 5% Ru/C in the aqueous phase. Kinetic data were
experimentally obtained over a wide range of temperatures (313–343 K), H2 partial pressure
(0.69–2.07 MPa), initial HMF concentration (19.8–59.5 mM), and catalyst loading (0.3–0.7 kg/m3)
in a three-phase slurry reactor. Disappearance of initial 5-HMF concentrations was modeled
using the power law and Langmuir–Hinshelwood–Hougen–Watson models. A model based
on the competitive adsorption of molecular H2 and HMF was proposed. It is presumed that
surface reaction between nondissociatively chemisorbed H2 and 5-HMF was rate determining.
This model provided the best fit for the kinetic data. From the Arrhenius equation, the activa-
ꢀC
tion energy for the surface reaction was found to be 104.9 kJ/mol.
Inc. Int J Chem Kinet 48: 318–328, 2016
2016 Wiley Periodicals,
tionality, have provided several opportunities to
produce chemicals and polymer-building blocks
during the past century [1–3]. 5-HMF is a well-known
chemical, commonly referred to as the “sleeping
giant.” It is the dehydration product of fructose in the
presence of acid catalyst with levulinic acid as the side
product [4,5]. 5-HMF has two functionalities: One
is the hydroxyl group, and the other is the carbonyl
group attached to the furan ring, which provides high
INTRODUCTION
Cellulosic biomass compounds such as 5-
hydroxymethylfurfural (5-HMF), which show
high oxygen–carbon ratio with high chemical func-
Correspondence to: P. D. Vaidya; e-mail: pd.vaidya@
ictmumbai.edu.in.
ꢀC
2016 Wiley Periodicals, Inc.

CATALYTIC HYDROGENATION OF 5-HYDROXYMETHYLFURFURAL TO 2,5-BIS-HYDROXYMETHYLFURAN
319
O
O
O
5% Ru/C, H2
OH
HO
HO
313-343 K,WATER
5-hydroxymethylfurfural (HMF)
2,5-bis-(hydroxymethyl)furan (BHMF)
O
O
O
OH
HO
HO
MFA
BHMTHF
DMF
Figure 1 Reaction scheme of 5-HMF hydrogenation to BHMF.
reactivity for the production of valuable chemicals,
biofuels, and furan derivatives via oxidation or hydro-
genation processes [6]. For example, hydrogenation
of 5-HMF produces 2,5-bis-hydroxymethylfuran
on ligno-cellulosic biomass, at large, and covered a
lumped approach to functional group hydrogenation
and hydro-deoxygenation reaction kinetics [28–30].
But so far, there is no information on the mass transfer
analysis and kinetic modelling of 5-HMF hydrogena-
tion to BHMF over Ru/C.
(BHMF),
2,5-bis-hydroxymethyltetrahydrofuran
(BHMTHF), 5-methylfurfuryl alcohol (MFA), and
2,5-dimethylfuran (DMF). It can be used in phar-
maceutical intermediates, fine chemicals, dyes,
food, resins, polymers, fuel additives, and fibers
industries [7–12].
Catalytic hydrogenation in water provides a green
route for conversion of 5-HMF to BHMF (Fig. 1). In
this work, reaction kinetics of 5-HMF hydrogenation
in aqueous solution was investigated using the Ru/C
catalyst in a slurry reactor. Kinetic data were experi-
mentally obtained over a wide range of temperatures
(313–343 K), H2 partial pressure (0.69–2.07 MPa),
initial 5-HMF concentration (19.8–59.5 mM), and cat-
alyst loading (0.3–0.7 kg/m3). The initial rates for
the disappearance of 5-HMF were fitted to a kinetic
model, which presumed that the surface reaction be-
tween nondissociatively chemisorbed H2 and 5-HMF
is rate determining. From the temperature dependence
of the surface reaction rate constant, the activation en-
ergy was found to be very high (104.9 kJ/mol), which
indicated that the reaction conditions are favorable for
studying reaction kinetics.
Catalytic hydrogenation of 5-HMF over various
metals (Cu, Pd, Ni, Fe, Pt, Rh, and Ru) has been exten-
sively studied [13,14]. The choice of catalyst and reac-
tion variables governs product distribution. For exam-
ple, DMF is the major product when Ru/C [15–17] and
the bimetallic Cu-Ru/C [14,18] are used in nonaque-
ous media. Contrarily, BHMF is formed in aqueous
solution at low temperature (298 K) over Ru-clusters
immobilized in nanosized mesoporous zirconium sil-
ica [19]. Furthermore, BHMTHF is the preferred prod-
uct when Ru is supported on ceria, magnesia-zirconia,
and γ -alumina, whereas Pt and Pd are less selective
than Ru [20]. While Pd promotes the formation of 5-
methyl furfural [21], Pt and Ir-Re catalysts are selective
to BHMF [22,23].
Kinetics of hydrogenation of biomass-derived com-
pounds in aqueous solution over Ru/C was investigated
in our previous works, e.g., hydroxyacetone, hydroxy-
acetaldehyde, guaiacol, and 2-furanone [24], levoglu-
cosan [25], vanillin [26], and xylenol and maltol [27].
Often, a potential kinetic model was not enough to
represent the rate data owing to fractional reaction or-
ders and, consequently, hyperbolic kinetic models were
needed in these works. Some other studies focused
EXPERIMENTAL
Materials
5-Hydroxymethylfurfural (5-HMF; 98%) was pur-
chased from Sigma Aldrich (Mumbai, India). Hydro-
gen (H2) and nitrogen (N2) cylinders (99.9%) were
purchased from Inox Air Products (Mumbai, India).
A commercial 5% Ru/C catalyst was supplied by
International Journal of Chemical Kinetics DOI 10.1002/kin.20992

320
JAIN AND VAIDYA
Figure 2 MS data to confirm the formation of BHMF.
Arora-Matthey. To probe any possible influence of the
modifier Cu on the performance of Ru/C, a bimetallic
Cu-Ru/C (3:1) catalyst was prepared by using incipient
wetness impregnation method [18].
BHMF (see Fig. 2). From total organic carbon analysis
(ANATOC Instruments), it was found that the carbon
balance was 98%. The reproducibility of results, in
terms of analysis, was checked, and the error in exper-
imental measurements was <3%.
Methods and Analysis
Catalyst Characterization
The experimental setup was earlier discussed in pre-
vious works [24–27]. In this work, we investigated
kinetics in the 313–343 K range at different H2 partial
pressures (0.69–2.07 MPa) and catalyst concentrations
(0.3–0.7 kg/m3). 5-HMF was converted to BHMF in
the 313–343 K range. Few runs were conducted at
high temperature (T ࣙ 373 K) to just ascertain possi-
ble formation of by-products. When the temperature
was increased beyond 373 K, 5-HMF and BHMF were
converted to BHMTHF, MFA, and DMF. The fall in
the 5-HMF concentration was recorded using 5 mM
H2SO4 mobile phase by a refractive index (RI) de-
tector and HPLC technique (Agilent Technologies).
As evident from liquid chromatography–mass spec-
trometry (LC–MS) results, 5-HMF was converted to
To study the surface morphology of the catalyst, the
scanning-electron microscopy (SEM) technique was
used. Powder X-ray diffraction (XRD) patterns of the
fresh and spent catalyst were obtained using a Rigaku
Miniflex D 500 diffractometer and monochromic Cu
Kα radiation (XRD Commander). The instrument pa-
rameters were set at 25 mA, 40 kV, scan speed at
0.1 s−1, increment 0.01, and spinner rotation 5. The
mean catalyst particle size (21.7 μm) was measured
using a Coulter LS 230 particle size analyzer. Sur-
face metallic atom characteristics such as dispersion
(D = 0.028) and active particle diameter (94 nm)
were investigated by H2 chemisorption (Micromeritics
2920 unit). Procedures for BET analysis of Ru/C,
International Journal of Chemical Kinetics DOI 10.1002/kin.20992

CATALYTIC HYDROGENATION OF 5-HYDROXYMETHYLFURFURAL TO 2,5-BIS-HYDROXYMETHYLFURAN
321
determination of turnover frequency and confirmation
of the existence of a chemical control regime were
described in our previous work [27] and used here.
RESULTS AND DISCUSSION
As shown in Fig. 1, 5-HMF is hydrogenated over Ru/C
to BHMF in aqueous-phase. BHMF can be further re-
duced to BHMTHF, MFA, and DMF. Product distribu-
tion depends on the reaction conditions, solvent, and
catalyst selection. In the present work, we used mild
aqueous-phase reaction conditions for the fully selec-
tive hydrogenation of 5-HMF to BHMF. Further con-
version of BHMF to other products was not observed
in the lower temperature range 313–343 K. 5-HMF
was fully converted to BHMF (selectivity 100%). The
balance of 5-HMF and BHMF per the initial 100%
5-HMF was satisfactory (98%).
We observed that the selectivity of BHMF de-
creased when reaction conditions were intensified (T ࣙ
373 K). We also observed that the yield of DMF
increases when the Cu-Ru/C catalyst was used at
high temperature in 5-HMF hydrogenation. Hence, we
maintained lower temperature reaction conditions us-
ing 5% Ru/C in aqueous-phase hydrogenation of 5-
HMF to BHMF.
SEM images for fresh and spent Ru/C are shown
in Figs. 3a and 3b. It is evident that the honeycomb
monolithic structure of the catalyst (seen in Fig. 3a)
was retained even after reaction (Fig. 3b). SEM image
of Cu-Ru/C is shown in Fig. 4a, whereas the results of
EDX analysis are shown in Fig. 4b. The concentration
of Ru (0.99 ppm) and Cu (2.88 ppm) was measured by
the ICP-OES instrument. Both the fresh and spent Ru/C
samples showed a similar XRD pattern (see Fig. 5).
Diffraction peaks of Ruº species were observed
(2θ = 37° and 43°). The broad peak at 26.8° for carbon
suggests that the catalyst is amorphous in nature. In-
terestingly, no new intense peak was observed, which
may be due to the high dispersion of Ru on the support.
In this section, the effects of reaction variables on
5-HMF conversion and product distribution are dis-
cussed. We used different temperatures (T), molecu-
Figure 3 SEM image of (a) the fresh and (b) spent Ru/C
catalyst.
conversion after 60 min was 48.5%. Interestingly, the
concentration vs. time curve showed a distinct trend
at all temperatures. The reaction proceeded rapidly at
the beginning, so that unexpectedly high rate of con-
version was observed within the first 5 min; thereafter,
a gradual decrease in the rate and, finally, equilibrium-
type behavior around 60 min was seen. The reaction of
5-HMF hydrogenation was earlier reported in the lit-
erature by many research groups. Some of the earlier
studies include the work done by [1,13–18] and [21].
A comparison is given in Table II. Interestingly, Op
De Beeck et al. [1] reported complete conversion of 5-
HMF to BHMF using the Ru/C catalyst at T = 333 K,
lar H2 partial pressures PH , reactant concentrations
2
(CHMF), and catalyst loadings (ω) as shown in Table I.
Effect of Temperature
PH = 5 MPa in water. Scholz et al. [13] investi-
2
The dependency of concentrations of 5-HMF and
BHMF on time at 313, 328, and 343 K is represented in
Fig. 6. In all experiments, the reaction variables used
gated continuous transfer hydrogenolysis of 5-HMF
in isopropanol over Pd/Fe2O3 at T = 453 K, PH
=
2
2.5 MPa achieving 100% conversion; the yield of
BHMF decreasing with time. In addition to this,
Tamura et al. [23] reported 99% yield of BHMF in
were as follows: PH = 0.69 MPa, CHMF,0 = 39.7 mM,
2
and ω = 0.5 kg/m3. At T = 343 K, the highest
International Journal of Chemical Kinetics DOI 10.1002/kin.20992

322
JAIN AND VAIDYA
water at T = 303 K and PH = 0.8 MPa using Ir-
2
a
ReOx/SiO2 with complete conversion of 5-HMF. Chat-
terjee et al. [22] studied 5-HMF hydrogenation and re-
ported 100% conversion and 98.9% BHMF yield in 2 h
at T = 308 K, PH = 0.8 MPa, using Pt/MCM-41 in
2
the aqueous phase.
Effect of Catalyst Loading
The effect of catalyst loading on the initial turnover
frequency (TOF) values was investigated at 313, 328,
and 343 K in the 0.3–0.7 kg/m3 range. The values of
PH and CHMF,0 were 0.69 MPa and 39.7 mM, respec-
2
tively. The results are presented in Table III. With an in-
crease in catalyst loading, the initial rate increased from
4.6 × 10−4 to 10.4 × 10−4 kmol/(m3 min) at 328 K.
Thus, it can be concluded that the reaction rate exhibits
first-order dependence on the catalyst concentration.
001
b
1000
900
800
700
600
500
400
300
200
100
0
Effect of H2 Partial Pressure and
Concentration of 5-HMF
The effect of PH on the initial TOF values was studied
2
in the range of 0.69 to 2.07 MPa pressure at temper-
atures 313, 328, and 343 K. While CHMF,0 was kept
constant at 39.7 mM, a value of catalyst loading ω =
0.5 kg/m3 was used. The results are presented in Fig. 7.
Clearly, the dependence of initial TOF on PH was lin-
2
ear. This behavior suggests that H2 is weakly adsorbed
on the catalyst surface without dissociation. When the
H2 partial pressure was increased threefold from 0.69
to 2.07 MPa at T = 328 K, the 5-HMF concentration
decreased from 39.7 to 5.1 mM within a period of 1 h.
These results are shown in Table I.
0.00
1.00
2.00
3.00
4.00
5.00
keV
6.00
7.00
8.00
9.00
10.00
Figure 4 (a) SEM image of the fresh Cu-Ru/C catalyst and
(b) results of EDX analysis.
The effect of CHMF,0 on the initial TOF values was
studied at 313, 328, and 343 K in the 19.8–59.5 mM
range. The results are depicted in Fig. 8. From the plots
of ln r vs. ln CHMF,0 (see Fig. 9), the orders of reaction
at T = 313, 328, and 343 K were found to be equal to
1.15, 1.04, and 0.92, respectively.
Reaction Mechanism and Kinetic Modelling
Langmuir-Hinshelwood-Hougen-Watson (LHHW) ki-
netics was used for modeling initial rates of 5-HMF
disappearance. The hydrogenation reaction of HMF to
BHMF was represented as
C6H6O3 + H2 → C6H8O3
(1)
Usually, the surface reaction over Ru is the
rate-limiting step in the hydrogenation of biomass–
precursor compounds in aqueous solution [25–27]. As-
suming that the surface reaction between 5-HMF and
Figure 5 XRD images of fresh and spent Ru/C catalyst.
International Journal of Chemical Kinetics DOI 10.1002/kin.20992

CATALYTIC HYDROGENATION OF 5-HYDROXYMETHYLFURFURAL TO 2,5-BIS-HYDROXYMETHYLFURAN
323
Table I Effect of Experimental Conditions on Fractional Conversion (X) of 5-HMF and Turnover Frequency (TOF)
Values
Catalyst Loading
Fractional HMF
Conversion (X)
CHMF,0 (mM)
PH (MPa)
(ω) kg/m3
T (K)
TOF (1/min)
2
19.8
0.69
0.69
0.69
1.38
2.07
0.5
0.5
0.5
0.5
0.5
313
328
343
313
328
343
313
328
343
313
328
343
313
328
343
0.48
0.70
0.94
0.34
0.43
0.48
0.31
0.33
0.44
0.47
0.69
0.76
0.66
0.87
0.95
1.98
2.78
3.97
3.97
5.41
7.72
5.77
8.22
10.82
8.30
12.99
19.48
12.99
20.06
28.14
39.7
59.5
39.7
39.7
The surface reaction may be represented as:
k3,k−3
2HS + HMFS ←→ BHMF + 3 S
(4)
where BHMF is the product. For the case of molecular
adsorption of H2, Eq. (2) is replaced by
k1,k−1
H2 + S ←→ H2S
(5)
The surface reaction is now represented by
k3,k−3
H2S + HMFS ←→ BHMF + 2 S
(6)
The various steps for noncompetitive adsorption of
dissociatively chemisorbed H2 and HMF are given by
Figure 6 Dependency of concentrations of 5-HMF and
BHMF on time at 313, 328 and 343 K (CHMF,0 = 39.7 mM,
k1,k−1
H2 + 2 S1 ←→ 2 HS1
(7)
(8)
PH = 0.69 MPa and ω = 0.5 kg/m3).
2
k2,k−2
HMF + S2 ←→ HMFS2
H2 is rate controlling, four kinetic models (competitive
and noncompetitive) were considered.
Using the sequence of steps reported by us ear-
lier [25], the competitive adsorption of dissociatively
chemisorbed H2 and the reactant is represented as
For this case, the surface reaction is given by
k3,k−3
2HS1 + HMFS2 ←→ BHMF + 2 S1 + S2
(9)
k1,k−1
H2 + 2 S ←→ 2 HS
(2)
(3)
For the case of molecular adsorption of H2, Eq. (7)
is replaced by
k2,k−2
k1,k−1
HMF + S ←→ HMFS
H2 + S1 ←→ H2S1
(10)
International Journal of Chemical Kinetics DOI 10.1002/kin.20992

324
JAIN AND VAIDYA
Table II Comparison of this Study with Previous Works on Hydrogenation of 5-HMFa
Compound
Reaction Conditions
Products
Identified
X
(%)
Comment
Reference
This work
5-HMF
(39.7 mM)
328 K, 1 h, 0.69 MPa, 5%
Ru/C, Solvent: H2O
BHMF
43
Batch reactor, low
temperature, selective
hydrogenation of
carbonyl group
5-HMF
(39.7 mM)
423 K, 0.69 MPa, 5%
Ru/C, Solvent: H2O
BHMF,
BHMTHF,
MFA, DMF
82
Batch reactor, high
temperature reaction
conditions, formation of
different products
This work
5-HMF
(5.6 mmol)
5-HMF
333 K, 40 min, 5 MPa,
Ru/C, Solvent: H2O
453 K, 25 min, Pd/Fe2O3,
2.5 MPa, Solvent: IPA
BHMF
100 Batch reactor, high
pressure reaction
Op De Beeck
et al. [1]
Scholz et al. [13]
BHMF, DMF 100 Flow reactor, high
temperature and
(0.12 M)
pressure, organic solvent
100 Batch reactor, high
5-HMF
(2.5 wt%)
473 K, 2 h, 2 MPa, Ru/C
Solvent: THF
DMF
DMF
DMF
DMF
Hu et al. [16]
temperature, organic
solvent
100 Batch reactor, high
temperature, organic
solvent
5-HMF
(2.5 g 5-HMF)
493 K, 10 h, 0.68 MPa,
Cu-Ru/C (3:1); Solvent:
1-butanol
493 K, 10 h, H2(g),
Cu-Ru/C, solvent:
1-butanol
Roma´n- Leshkov
et al. [14]
Crude 5-HMF
(untreated corn
stover)
5-HMF
(6 wt%)
–
High temperature, organic
solvent
Binder and
Raines [18]
533 K, 1.5 h,
High temperature, organic
solvent
Zhang et al. [15]
H24.48×10−2 mol/g,
Ru/C, Solvent: 1-butanol
393 K, 1 h, 6.2 MPa, Pd/C
Solvent: ionic liquid
99.8
19
5-HMF
(126 mg)
MF, MFA,
BHMF, DMF,
HD, MTHFA
High temperature and
pressure required,
formation of different
products
Chidambaram
and Bell [21]
5-HMF
(1.2 wt%)
463 K, 6 h, Ru/C, 2 MPa
Solvent: IPA
DMF, BHMF, 100 High temperature and
Jae et al. [17]
MFA, MF
pressure required,
formation of different
products, catalytic
transfer hydrogenation
aAbbreviation: BHMF, 2,5-bis-hydroxymethylfuran; BHMTHF, 2,5-bis-hydroxymethyltetrahydrofuran; DMF, 2,5-dimethylfuran; HD, 2,5-
dexadione; 5-HMF, 5-hydroxymethylfurfural; MF, 5-methylfurfural; MFA, 5-methylfurfuryl alcohol; MTHFA, 5-methyltetrahydrofurfuryl
alcohol; X, fractional conversion.
and the surface reaction is now represented by
The average values of RSS, variance, and coeffi-
cient of determination (CoD) for all models are given
in Table IV at 313, 328, and 343 K. While the values
of CoD for models I and II were very close, kinetic
parameters estimated from model II did not fit to the
experimental data. However, model I reasonably rep-
resented kinetic data. The parity plot for model I is
shown in Fig. 10. Indeed, the experimental TOF values
are in good agreement with the predicted ones.
k3,k−3
H2S1 + HMFS2 ←→ BHMF + S1 + S2 (11)
TableIV presents the kineticmodels (I–IV) andtheir
rate-determining steps (i.e., Eqs. (6), (4), (9) and (11)).
All models were simplified to the initial rate equa-
tions. A model discrimination technique was used for
selecting the most appropriate model. The model pa-
rameters were estimated by using the optimization pro-
gram POLYMATH 6.0 and the Simplex–Levenberg–
Marquardt algorithm. Models 3 and 4 did not provide
a good fit of the data and hence were rejected.
Furthermore, model I was validated by a straight-
forward graphical procedure. If model I holds, a plot
of (CHMF,0/r0)1/2 vs. CHMF,0 should be linear at con-
stant values of CH . Certainly, this plot confirmed that
2
model I was appropriate (see Fig. 11). An intrinsic
International Journal of Chemical Kinetics DOI 10.1002/kin.20992

CATALYTIC HYDROGENATION OF 5-HYDROXYMETHYLFURFURAL TO 2,5-BIS-HYDROXYMETHYLFURAN
325
Table III Effect of Catalyst Loading (ω) on Initial 5-HMF Disappearance Rates (r0) and TOF Value (PH = 0.69 MPa,
2
CHMF,0 = 39.7 mM)
r0 × 104 (kmol/(m3 min))
Initial TOF (1/min)
328 K
ω (kg/m3)
313 K
328 K
343 K
313 K
343 K
0.3
0.5
0.7
3.9
6.46
8.85
4.6
7.5
10.4
5.3
8.7
12.1
4.69
4.66
4.74
5.53
5.41
5.36
6.37
6.28
6.24
Figure 7 Effect of H2 partial pressure (PH ) on the initial
Figure 9 Plots of ln r vs. ln CHMF,0 at T = 313, 328, and
343 K (PH2 = 0.69 MPa and ω = 0.5 kg/m3).
2
TOF values at 313, 328, and 343 K (CHMF,0 = 39.7 mM and
ω = 0.5 kg/m3).
parameter (C2) method earlier reported by Brahme
and Doraiswamy [31] for glucose hydrogenation was
used for further discrimination. The proposed models
were expressed in terms of either fractional conver-
sion or partial pressure (or concentration). The method
involves two different parameters C1 and C2. This pro-
cess describes C2 in terms of the initial concentration
of the reactant. To find out the suitable model from
the equation for C2, the raw data obtained at differ-
ent initial concentrations of the reactant was used to
find the rate at different reaction conditions. Model I
P(1−X )
(r0 =
) can be expressed in terms of C1 and
[C1−C2X ]2
C2 as
1 + αPKH + CHMF,0KHMF
2
C1 =
(12)
(13)
ꢀ
ꢁ
1/2
α CHMF,0KHMFKH2k3
ꢀ
ꢁ
Figure 8 The dependence of the initial TOF on the ini-
1/2
KHMF CHMF,0
tial HMF concentration at various temperatures (PH = 0.69
2
C2 =
ꢀ
ꢁ
MPa and ω = 0.5 kg/m3).
1/2
α KHMFKH k3
2
International Journal of Chemical Kinetics DOI 10.1002/kin.20992

326
JAIN AND VAIDYA
Table IV Plausible Kinetic Models (I–IV) and Model Discrimination Parameters for 5-HMF Hydrogenation
Model
I
Rate-Controlling Step and Initial-Rate Expression
Temperature (K)
Variance
RSS
CoD
Eq. (6)
313
328
343
3.62 × 10−8
9.50 × 10−8
2.86 × 10−7
5.38 × 10−5
8.72 × 10−5
1.51 × 10−4
0.997
0.996
0.994
k3 KH2 KHMF CH2 CHMF
(1+KH2 CH2 +KHMF CHMF
r =
2
)
II
Eq. (4)
313
328
5.93 × 10−8
6.89 × 10−5
0.978
0.972
k3 KH2 KHMFCH2 CHMF
r =
2.00 × 10−7
1.26×10−4
√
3
(1+ KH2 CH2 +KHMF CHMF
)
343
4.82 × 10−7
1.96 × 10−4
0.967
III
IV
Eq. (9)
313
328
3.38 × 10−8
5.20 × 10−5
0.977
0.961
k3 KH2 KHMFCH2 CHMF
(1+ KH2 CH2 )2(1+KHMF CHMF
r =
9.04 × 10−8
8.50 × 10−5
√
)
343
2.97 × 10−7
1.54 × 10−4
0.969
Eq. (11)
313
328
343
3.99 × 10−8
1.29 × 10−7
4.82 × 10−7
5.65 × 10−5
1.02 × 10−4
1.96 × 10−4
0.955
0.941
0.967
k3 KH2 KHMFCH2 CHMF
(1+KH2 CH2 )(1+KHMF CHMF
r =
)
Figure 11 Plots of (CHMF,0/r0)1/2 vs. CHMF,0 at T = 313,
328, and 343 K.
Figure 10 Parity plot for model I.
From Eq. (12) and (13), for the model I, C2 was
calculated from experimental data using the following
relation:
ꢀ
ꢁ
1/3
KHMF CHMF,0
C2 =
(16)
(α KHMF KH2 k3)1/3
ꢂ
ꢄ
ꢃ
ꢂ
ꢃ
1/2
1/2
From Eqs. (15) and (16), for the model II, C2 was
calculated from experimental data using the following
relation:
P
r0
1
P (1 − X )
1
C2 =
−
(14)
X
r
X
ꢅ
P(1−X)
ꢂ
ꢃ
ꢂ
ꢃ
Model II r0 =
can be expressed in terms
1/3
1/3
[C1−C2X]3
P
r0
1
P (1 − X )
1
C2 =
−
(17)
of C1 and C2 as follows:
X
r
X
1/2
1 + (αPKH2
)
+ CHMF,0 KHMF
The intrinsic parameter C2 is a function of CHMF,0
.
C1 =
(15)
ꢀ
ꢁ
1/3
At constant values of P, models I and II require plots of
α CHMF,0 KHMF KH2 k3
International Journal of Chemical Kinetics DOI 10.1002/kin.20992

CATALYTIC HYDROGENATION OF 5-HYDROXYMETHYLFURFURAL TO 2,5-BIS-HYDROXYMETHYLFURAN
327
kinetics of 5-HMF to BHMF was studied in a three-
phase slurry reactor using the Ru/C catalyst. Using a
limited range of temperature, 313–343 K, H2 partial
pressure 0.69–2.07 MPa, initial 5-HMF concentration,
19.8–59.5 mM, and catalyst loading, 0.3–0.7 kg/m3,
kinetic data were experimentally obtained. The appar-
ent reaction orders with respect to 5-HMF and H2 were
close to one. Several LHHW models were proposed to
describe reaction kinetics. Using model discrimination
techniques, it was found that the kinetic data could be
best fitted to a LHHW-based model with surface re-
action between competitively adsorbed molecular H2
and 5-HMF as the rate-controlling step. Finally, it was
found that the rise in temperature beyond 373 K re-
sulted in the conversion of 5-HMF and BHMF into
BHMTHF, MFA, and DMF. Furthermore, the yield of
DMF increased when the Cu-Ru/C catalyst was used
at high temperature. These inferences will aid in the
designing and operation of hydrogenation reactors for
producing useful chemicals from biomass. Also, this
work will stimulate further research on hydrogenation
kinetics of biomass–precursor compounds in aqueous
solution.
a
b
NOMENCLATURE
C2
CH
Intrinsic parameter in Eqs. (12) and (13)
Concentration of H2 in the liquid phase,
kmol/m3
2
CHMF
Concentration of HMF in the liquid phase,
kmol/m3
CHMF,0 Initial concentration of HMF in the liquid
phase, kmol/m3
D
KHMF
Metal dispersion
Figure 12 Plots for validation of models I (a) and II (b).
Adsorption equilibrium constant for HMF,
m3/kmol
C2 vs. C1/2 and C2 vs. CH2/M3F,0 as straight lines pass-
KH
Adsorption equilibrium constant for H2,
m3/kmol
Reaction rate constant, kmol/(kgcat min)
Total pressure, atm
Partial pressure of H2, MPa
Coefficient of determination
Rate of reaction, kmol/(kgcat min)
Initial rate of reaction, kmol/(kgcat min)
Active site on catalyst surface
Temperature, K
2
HMF,0
ing through the origin. Such graphs for hydrogenation
of 5-HMF are shown in Figs. 12a and 12b. Clearly,
the plot for model I (i.e., Fig. 11a) is linear with zero
intercept. On the other hand, Fig. 11b does not present
a linear relation. Thus, model II is ruled out. From the
Arrhenius equation, the activation energy for surface
reaction of 5-HMF was calculated as 104.9 kJ/mol.
Finally, it may be noted that, among the different
mechanistic models considered, model I fits the data
well.
k3
P
PH
2
R2
r
r0
S
T
t
Time, h
X
Fractional conversion
CONCLUSIONS
GREEK SYMBOLS
Catalytic hydrogenation in water provides a green
route for conversion of 5-HMF to useful chemi-
cals. In the present work, selective hydrogenation and
α
Henry’s law constant, kmol/(m3 kPa)
Catalyst loading, kg/m3
ω
International Journal of Chemical Kinetics DOI 10.1002/kin.20992

328
JAIN AND VAIDYA
Anandkumar B. Jain is grateful to University Grants Com-
mission, New Delhi, India, for financial aid.
15. Zhang, J.; Lin, L.; Liu, S. Energy Fuels 2012, 26, 4560–
4567.
16. Hu, L.; Tang, X.; Xu, J.; Wu, Z.; Lin, L.; Liu, S. Ind Eng
Chem Res 2014, 53, 3056–3064.
BIBLIOGRAPHY
17. Jae, J.; Zheng, W.; Lobo, R. F.; Vlachos, D. G. Chem
Sus Chem 2013, 6, 1158–1162.
1. Op De Beeck, B., Dusselier, M.; Geboers, J.; Holsbeek,
J.; Morre´, E.; Oswald, S.; Giebeler, L.; Sels, B. F. Energy
Environ Sci 2015, 8, 230–240.
2. Teong, S. P.; Yi, G.; Zhang, Y. Green Chem 2014, 16,
2015–2026.
3. Van Putten, R. J.; Van Dar Wass, J. C.; Jong, E. D.;
Rasrendra, C. B.; Heeres, H. J.; de Vries, J. G. Chem
Rev 2013, 113, 1499–1597.
18. Binder, J. B.; Raines, R. T. J Am Chem Soc 2009, 131,
1979–1985.
19. Chen, J.; Lu, F.; Zhang, J.; Yu, W.; Wang, F.; Gao, J.;
Xu, J. Chem Cat Chem 2013, 5, 2822–2826.
20. Alamillo, R.; Tucker, M.; Chia, M.; Paga´n-Torres, Y.;
Dumesic, J. A. Green Chem 2012, 14, 1413–1419.
21. Chidambaram, M.; Bell, A. T. Green Chem 2010, 12,
1253–1262.
4. Tong, X.; Ma, Y.; Li, Y. Appl Catal A: Gen 2010, 385,
1–13.
22. Chatterjee, M.; Ishizaka, T.; Kawanami, H. Green Chem
2014, 16, 4734–4739.
5. Alonso, D. M.; Bond, J. Q.; Dumesic, J. A. Green Chem
2010, 12, 1493–1513.
6. Kwon, Y.; de Jong, E.; Raoufmoghaddam, S.; Koper,
M. T. M. Chem Sus Chem 2013, 6, 1659–1667.
7. Hu, X.; Westerhof, R. J. M.; Wu, L.; Dong, D.; Li, C. Z.
Green Chem 2015, 17, 219–224.
23. Tamura, M.; Tokonami, K.; Nakagawa, Y.; Tomishige,
K. Chem Commun 2013, 49, 7034–7036.
24. Bindwal, A. B.; Bari, A. H.; Vaidya, P. D. Chem Eng J
2012, 207–208, 725–733.
25. Bindwal, A. B.; Vaidya, P. D. Ind Eng Chem Res 2013,
52, 17781–17789.
8. Gallezot, P. Chem Soc Rev 2012, 41, 1538–1558.
9. Climent, M. J.; Corma, A.; Iborra, S. Green Chem 2011,
13, 520–540.
10. Buntara, T.; Noel, S.; Phua, P. H.; Melia´n-Cabrera, I.; de
Vries, J. G.; Heeres, H. J. Top Catal 2012, 55, 612–619.
11. Climent, M. J.; Corma, A.; Iborra, S. Green Chem 2014,
16, 516–547.
12. Liu, D.; Chen, Y. X. Green Chem 2014, 16, 964–981.
13. Scholz, D.; Aellig, C.;Hermans, I. ChemSusChem2014,
7, 268–275.
26. Bindwal, A. B.; Vaidya, P. D. Energy Fuels 2014, 28,
3357–3362.
27. Jain, A. B.; Vaidya, P. D. Energy Fuels 2015, 29, 361–
368.
28. Grilc, M.; Lizokar, B. Levec, J. Catal Today 2015, 256,
302–314.
29. Grilc, M.; Lizokar, B. Levec, J. Appl Catal B: Environ
2014, 150–151, 275–287.
30. Grilc, M.; Lizokar, B. Levec, J. Biomass Bioenergy
2014, 63, 300–312.
14. Roma´n-Leshkov, Y.; Barrett, C. J.; Liu, Z. Y.; Dumesic,
J. A. Science 2007, 447, 982–985.
31. Brahme, P. H.; Doraiswamy, L. K. Ind Eng Chem, Pro-
cess Des Dev 1976, 15, 130–137.
International Journal of Chemical Kinetics DOI 10.1002/kin.20992