G Model
CATTOD-9913; No. of Pages9
ARTICLE IN PRESS
W. Jeon et al. / Catalysis Today xxx (2015) xxx–xxx
3
2.2. Reaction procedure
compare the catalytic activity of metal ion catalysts. The determi-
nation of the reaction condition is important since the production
of furfural from alginate over catalysts is greatly influenced by
reaction temperature and time [10]. For example, the catalytic
hydrothermal conversion of alginate to furfural is promoted by
increasing the reaction temperature and time, however, the yield
of furfural suddenly decreases in the reaction condition due to
side reactions of furfural produced. Therefore, the reaction tem-
perature (200 ◦C) and time (30 min) were selected as an optimal
reaction condition in the evaluation of catalytic activity of various
metal ions, in order to minimize the loss of furfural produced by the
hydrothermal reaction of alginic acid. As shown in Fig. 2, Cu(II) ions
shows the highest furfural yield (13.19 mol%) higher than two times
the furfural yield obtained in a blank test (5.14 mol%). Based on the
furfural yield of the blank test, Cu(II), Fe(III) and Pb(II) ions pro-
moted the conversion of alginic acid to furfural. In the hydrothermal
reaction under Zn(II), Co(II) and Ni(II) ions, there was little or no
catalytic effect. Other metal cations except the six metal cations
mentioned above inhibited the production of furfural rather than
catalyzing it. In particular, the yield of furfural decreased from 5.14
to 2.5 mol% when Y(III) ions participated in the reaction. Gener-
ally, xylose obtained by hydrolysis of hemicellulose is sequentially
lulose [23,24]. In contrast, glucose, a monomer of cellulose, tends to
convert to hydroxymethylfufrual (HMF) rather than furfural, since
glucose has an extra alcohol functional group compared to xylose
[25]. The conversion of HMF to furfural was feasible via the loss
of formaldehyde in HMF, but it is not favorable [26]. In the same
manner, the carboxylic functional group of mannuronic acid and
guluronic acid should be eliminated to produce furfural from alginic
acid. In other words, both decarboxylation and dehydration are
important steps for the conversion of alginic acid to furfural, which
would be influenced by metal cations.
A stainless steel batch reactor (50 mL) lined with Teflon was used
for hydrothermal treatment of alginic acid. A stirrer was located
inside the reactor for an effective contact between insoluble alginic
acid and metal cations in water (600 rpm). Alginic acid (0.6 g) and
metal cation solution (30 mL) were added to the reactor. The sealed
reactor was purged with nitrogen gas and mounted in a heater. The
heating time to different target temperatures varied as shown in
Fig. 1. The heating time, approximately 30 min, was excluded in
counting reaction time. For example, 61 min of reaction time was
actually needed to complete the hydrothermal reaction of alginic
acid at 180 ◦C for 30 min, since it took 31 min for heating the reactor
to the reaction temperature. After dwelling at the target temper-
atures, the reactor was immediately quenched with a cold-water.
The final products obtained were filtered and centrifuge in order to
separate liquid products from solid-liquid mixtures before analysis.
2.3. Product analysis
The furfural and organic acids in liquid products were quantified
with an Agilent 1200 Series HPLC equipped with two Shodex RSpak
KC-811 columns in series. RI detector (Agilent G1362A) and UV
detector (Agilent G1314B) were used together for crosschecking the
data. The wavelength of the UV detector was set to 210 nm in order
to observe furfural and organic acids simultaneously. Phosphoric
acid aqueous solution (5 mM), as a mobile phase, was run through
the column (40 ◦C) at a flow rate of 1.0 mL min−1
Based on data obtained from HPLC analysis, molar yields of prod-
ucts were calculated as:
.
nCi
ni
Yieldi mol% = 100 ×
×
(
)
6
nru
where nCi = the number of carbon atoms in the organic product i,
ni = the number of moles of the organic product i as determined by
HPLC analysis, nru = the initial number of moles of repeating units
(C6H8O6) in alginic acid, equal to the mass of alginic acid divided
by 176.
The molecular weight distribution of products was analyzed by
gel permeation chromatography (GPC). The GPC system (Ultimate
3000, Dionex) was composed of three types of columns (Waters
Ultrahydrogel column: 120, 500 and 1000) in series. Sodium azide
solution (0.1 M), as a mobile phase, flowed through the colum
(40 ◦C) at a flow rate of 1.0 mL min−1. Pullulan with a molecular
weight distribution from 342 to 80,500 was used to calibrate the
GPC system.
Furfural and intermediates of the hydrothermal reaction were
identified with a LC–MS system (Surveyor, Thermo Finnigan) in
combination with a mass spectrometer (LCQ Deca XP Plus, Thermo
Finnigan) equipped with an electrospray ionization module and
working in positive or negative mode with a capillary temper-
ature of 275 ◦C. Three types of mobile phases (0.1% of formic
acid dissolved in distilled water, acetonitrile or methanol) were
run through a column (SynergiTM 4 m Polar-RP 80 Å, LC Column
150 × 2 mm, Phenomenex) at a flow rate of 0.25 mL min−1. The UV
wavelength was set from 210 to 280 nm.
In order to explain the effect of metal cations on the production
of furfural from alginic acid, the physical or chemical properties of
the metal ions were correlated with the yields of furfural. Fig. 3(a)
shows the correlation between the furfural yield and ionic radius
of metal ions. For lanthanide metal ions, the yield of furfural lin-
early increases with the ionic radius of the metal ions. However, the
lanthanide metal cations shows poor catalytic performance in the
production of furfural, showing lower furfural yields than that of
the blank test. The yields of furfural for post-transition metal ions
are also proportional to the size of metal ions. The largest metal
ions, Pb(II) (119 pm), exhibits the highest furfural yield (7.12 mol%)
among the post-transition metal ions. On the other hand, there was
no clear relation between the furfural yield and the ionic radius of
the hydrothermal reaction of alginic acid catalyzed by those two
groups of metal ions are lower than by transition metal ions.
The yields of furfural under post-transition or transition metal
ions are almost proportional to the electronegativity as shown in
Fig. 3(b). The strong electronegativity of metal cations suggests
that the metal ions can play a role of Lewis acid catalysts in the
hydrothermal conversion of alginic acid, since the electronegative
metal ions attract electrons towards itself, promoting the elec-
tron transfer in the reaction. Lewis acidity is proportional to the
electronegativity of metal cations [27], which may lead to the acid-
catalyzed dehydration in the conversion of alginic acid to furfural.
On the other hand, the yield of furfural increases with decreasing
electronegativity under the lanthanide metal ions, implying that
these metal ions likely participate in the hydrothermal reaction
through different pathways, compared to the transition metal ions.
In addition to furfural, a few organic acids were produced in the
hydrothermal reaction of alginic acid. As listed in Table 1, glycolic
acid, lactic acid and formic acid were mainly produced with furfural
in the conversion of alginic acid at 200 ◦C for 30 min. In the blank
3. Results and discussion
3.1. Effect of different metal cations on the conversion of alginic
acid to furfural
The various metal cations were used as catalysts in the
hydrothermal conversion of alginic acid to furfural at 200 ◦C for
30 min. It should be noticed that the reaction condition was deter-
mined based on the results of our previous research in order to
Please cite this article in press as: W. Jeon, et al., Hydrothermal conversion of alginic acid to furfural catalyzed by Cu(II) ion, Catal. Today