Catalytic hydrothermal conversion of macroalgae-derived alginate: Effect of pH on production of furfural and valuable organic acids under subcritical water conditions

Article abstract of DOI:10.1016/j.molcata.2015.01.011

The hydrothermal decomposition of sodium alginate was analyzed in subcritical water as a function of pH in order to investigate the effects of catalysts on the production of value-added chemicals. A base-catalysed reaction at pH 13 promoted the decomposition of alginate, resulting in the production of lactic acid, fumaric acid and malic acid as major species. At pH 1, monomers (mannuronic acid and guluronic acid), furfural and glycolic acid were predominantly produced by the acid-catalysed hydrothermal decomposition of alginate. Increasing the reaction temperature enhanced both the acid- and base-catalysed reactions, albeit by varying degrees dependent upon the catalyst type. Our results demonstrate that optimizing the reaction temperature and pH is critical for the efficient conversion of seaweed-derived biomass into valuable products.

Full text of DOI:10.1016/j.molcata.2015.01.011

Journal of Molecular Catalysis A: Chemical 399 (2015) 106–113
Contents lists available at ScienceDirect
Journal of Molecular Catalysis A: Chemical
journal homepage: www.elsevier.com/locate/molcata
Catalytic hydrothermal conversion of macroalgae-derived alginate:
effect of pH on production of furfural and valuable organic acids
under subcritical water conditions
Wonjin Jeona, Chunghyeon Bana, Geonu Parka, Tae-Kyung Yub, Jeong-Yong Suhb,
Hee Chul Wooc, Do Heui Kima,∗
a School of Chemical and Biological Engineering, Institute of Chemical Processes, Seoul National University, Seoul 151-744, Republic of Korea
b Department of Agricultural Biotechnology, Seoul National University, Seoul 151-744, Republic of Korea
c Department of Chemical Engineering, Pukyong National University, Busan 608-739, Republic of Korea
a r t i c l e i n f o
a b s t r a c t
Article history:
The hydrothermal decomposition of sodium alginate was analyzed in subcritical water as a function
of pH in order to investigate the effects of catalysts on the production of value-added chemicals. A
base-catalysed reaction at pH 13 promoted the decomposition of alginate, resulting in the produc-
tion of lactic acid, fumaric acid and malic acid as major species. At pH 1, monomers (mannuronic
acid and guluronic acid), furfural and glycolic acid were predominantly produced by the acid-catalysed
hydrothermal decomposition of alginate. Increasing the reaction temperature enhanced both the acid-
and base-catalysed reactions, albeit by varying degrees dependent upon the catalyst type. Our results
demonstrate that optimizing the reaction temperature and pH is critical for the efficient conversion of
seaweed-derived biomass into valuable products.
Received 2 January 2015
Accepted 12 January 2015
Available online 13 January 2015
Keywords:
Biomass
Alginate
Catalytic hydrothermal treatment
Homogeneous catalysis
Subcritical water
© 2015 Elsevier B.V. All rights reserved.
1. Introduction
biomass in hot compressed water and has proven itself as a use-
ful alternative, as the reaction can be accomplished within a few
Marine algal biomass, composed of microalgae and macroalgae,
is emerging as a promising next-generation energy resource due
to its inherent advantages including a fast growth rate, ease of cul-
tivation, excellent CO2 consumption and absence of lignin [1–3].
Alginate, a major component of brown seaweeds, has been widely
utilized in various industrial and scientific fields. Alkali metal salts
of alginate, such as sodium alginate and calcium alginate, are useful
additives for foods that contribute to shape, texture and viscosity
[4]. The biocompatibility of alginate has been exploited for various
applications, including biological substrate construction and drug
delivery [5–8]. In an effort to produce novel biofuels from alginates,
the metabolic conversion of alginate has been proposed for the pro-
duction of ethanol using engineered microorganisms [9,10]. The
suggested enzymatic process provides high selectivity for desired
products, but involves long reaction times (ca. days), high costs
and complex genetic modifications of yeast strains, drawbacks that
have hampered its widespread adoption [11]. Hydrothermal treat-
ment with subcritical water produces valuable chemicals from
hours using a relatively simple procedure [12–16]. In addition, the
hydrothermal conversion process is cheap, sustainable and eco-
friendly due to its use of subcritical water instead of expensive and
toxic organic solvents [17].
Alginate has excellent potential for the generation of value-
added chemicals due to its polysaccharide structure of plant-like
origin. Alginate is composed of two monomeric subunits containing
the carboxylic groups mannuronic acid and guluronic acid [18,19].
The subunits are connected by ␤-1,4-glycosidic bonds analogous
to cellulose [20–23]. Similar to the hydrothermal conversion of
cellulose, the linkages in alginate are selectively broken through
a ␤-elimination reaction under hydrothermal reaction conditions,
as depicted in Fig. 1 [19,24]. Further hydrothermal decomposition
of alginate with the addition of hot compressed water leads to the
production of various organic compounds, via breaking of the C
or C O bonds within monomeric subunits [25–27].
C
Hydrothermal conversion of cellulose into valuable organic
compounds has been extensively investigated [28–33], but
hydrothermal treatment of alginate has only recently been applied
to produce useful organic compounds. Aida et al. described the
production of monomers and organic acids under subcritical and
supercritical water conditions, by investigating the effects of
∗
Corresponding author. +82 2880 1633.
E-mail address: dohkim@snu.ac.kr (D.H. Kim).
http://dx.doi.org/10.1016/j.molcata.2015.01.011
1381-1169/© 2015 Elsevier B.V. All rights reserved.

W. Jeon et al. / Journal of Molecular Catalysis A: Chemical 399 (2015) 106–113
107
2. Materials and experimental procedures
2.1. Materials
All organic acids were purchased from Sigma–Aldrich, except
fumaric acid which was purchased from Tokyo Chemical Industry.
Monomers of alginate were also prepared to elucidate the pathways
more clearly. Standard chemicals of those monomers, mannuronic
and guluronic acid, were purchased from Qingdao BZ Oligo Biotech,
China (purity > 98%). Alginic acid sodium salt (sodium alginate)
obtained from Sigma–Aldrich was used as a starting material in
the hydrothermal depolymerization reaction with no mechanical
pretreatment. Cellulose was purchased from Sigma–Aldirich and
mechanically treated with a milling machine for 20 h to reduce the
particle size and thus enhance the reactivity. Homogeneous cat-
alysts, hydrogen chloride and sodium hydroxide, were purchased
from Junsei Chemical and Sigma–Aldrich, respectively. For prepara-
tion of the acidic solvents, hydrogen chloride was added to distilled
water until the pH of the solvents reached 1 or 3. Likewise, sodium
hydroxide and distilled water were used to produce basic solvents
of pH 11 and 13.
2.2. Apparatus and experimental procedure
The hydrothermal conversion of sodium alginate was performed
using a tubular batch reactor consisting of SUS 316 tubing. A molten
salt bath including the mixture of alkali metal nitrates was utilized
as a heater. The procedure was comprised of five steps: sam-
ple loading, N2 gas purging, heating, cooling and analysis. As the
first step, sodium alginate (60 mg) and solvent (3 mL) were loaded
together into the reactor, which had an inner volume of 6 mL. The
solvent was prepared using pure distilled water, hydrogen chlo-
ride and sodium hydroxide to control pH values between 1 and 13.
Existing gases inside the reactor were removed by a vacuum pump
and the reactor was filled with N2 gas (99.999%). Next, the reactor
was immersed in the molten salt bath and taken out immediately
after the desired reaction time. The reaction time was recorded
from the initial time that a temperature sensor in the reactor first
detected the desired reaction temperature. After the heating step,
and within a few seconds, the reactor was quickly placed into a
cold-water bath. Prior to the analysis step, the final products were
pretreated using the following methods: dilution, neutralization,
centrifugation and filtration.
Fig. 1. Decomposition of alginate via the cleavage of 1, 4-glycosidic linkages in a
␤-elimination pathway.
reaction temperature (150–400 ◦C) on the conversion of alginate
into its monomers and monomer-derived organic acids, such as
monocarboxylic and dicarboxylic acids [26,27]. The impact of
different catalysts and pH on reactivity at high temperatures, how-
ever, remains unknown. We sought to systematically investigate
the effects of homogeneous catalysts on hydrothermal reactions
as a function of pH for the production of valuable organic acids.
The effects of homogeneous catalysts (such as mineral acids and
alkali salts) on hydrothermal degradation of alginate were first
investigated between the 1960s and 1980s [25,34–36], but the reac-
tion conditions were limited and quantification of the furfural and
organic acids was not satisfactory. Furthermore, the rapid decom-
position of alginate under conditions with homogeneous catalysts
prevented the analysis of the initial reaction states.
2.3. Product characterization
In this study, we investigated the effects of varying pH between
1 and 13 in order to better understand acid- and base-catalysed
reactions in hydrothermal decomposition of sodium alginate. In
addition, reaction temperatures between 150 and 250 ◦C were
used to understand the roles of catalysts and hot compressed
water at high temperatures, and to draw a comparison with previ-
ous findings conducted at temperatures below 150 ◦C [25,35–38].
A reaction time of less than 1 h was chosen to investigate the
reaction mechanism during the early stage of alginate decompo-
sition. We selected formic acid, acetic acid, glycolic acid, lactic
acid, fumaric acid, succinic acid, malic acid and furfural as target
products, because these C1–C5 organic compounds are considered
to be value-added chemicals by industrial chemists. Production
of the organic compounds was verified by HPLC equipped with
VWD and RID or GC/MS analysis. The production of gaseous prod-
ucts, such CO2, and solid residues were observed, but a detailed
analysis for those products was not performed in this study. The
current study characterizes a number of important aspects required
for our ongoing efforts to introduce heterogeneous catalysts into
the decomposition reaction of sodium alginates to produce value-
added organic compounds.
Product identification was performed with a GC–MS System
(Clarus 680/600T, PerkinElmer) equipped with an Agilent DB-5MS
column. Prior to GC–MS analysis, the samples were lyophilized and
silylated with a mixture of BSFTA with TMCS (99:1) and pyridine at
65 ◦C for 2 h. A representative result of GC–MS analysis for major
products was shown in Supplementary content Fig. 1S.
Products were quantified with an Agilent 1200 Series HPLC
equipped with two Shodex RSpak KC-811 columns in series.
Column oven temperature, flow rate of mobile phase and concen-
tration of phosphoric acid aqueous solution were 40 ◦C, 1 cm3/min
and 5 mM, respectively. Both RI detector (Agilent G1362A) and
UV detector (Agilent G1314B) were used to cross-check the HPLC
analysis for more precise characterization. The wavelength of UV
detector was set to 210 nm for analyzing final products. The sepa-
ration and detection of each product were effectively conducted as
shown in Fig. 2S(a–c). Based on data obtained from HPLC analysis,
molar yields of products were calculated as:
nCi
6
ni
Yieldi(mol%) = 100 ×
×
nru

108
W. Jeon et al. / Journal of Molecular Catalysis A: Chemical 399 (2015) 106–113
where nCi = the number of carbon atoms in the organic acid i,
ni = the number of moles of the organic acid i as determined by
HPLC analysis, nru = the initial number of moles of repeating units
(C6H7O6Na) in the sodium alginate, equal to the mass of sodium
alginate divided by 198.
The molecular weight distribution of raw alginate and
hydrothermally treated alginate was analyzed by a gel permeation
chromatography (GPC) technique. The GPC system (Ultimate 3000,
Dionex) was configured with three types of Waters Ultrahydro-
gel columns: 120, 500 and 1000, in series. Sodium azide solution
(0.1 M) was used as a mobile phase at 1 mL/min at 40 ◦C. The GPC
calibration was conducted using Pullulan with a molecular weight
distribution from 342 to 80,500.
1H NMR analysis for raw alginate and final aqueous products
was performed with a 400 MHz spectrometer (JeolJNM-LA400/LFG,
JEOL). The 1H NMR spectra were obtained at 80 ◦C and 128
scans. Liquid products were lyophilized and then dissolved in D2O
(0.6 mL) containing 3-(trimethylsilyl) propionic-2,2,3,3-d4 acid
sodium salt (0.05 wt%) as an internal standard. At a minimum, tests
were conducted to ensure reproducibility three times, to obtain
reliable results in quantification.
We noticed the production of insoluble aggregates and gaseous
compounds during the sample preparation, but they were not fur-
ther characterized due to their negligible quantities.
3. Results and discussion
Fig. 2 shows the GPC chromatograms of raw alginates and
hydrothermally-treated alginates under various reaction condi-
tions. The GPC data were normalized to the intensities of the highest
peaks on each chromatogram to compare differences among the
chromatograms. The GPC analysis indicated that depolymerization
of sodium alginate was very sensitive to the reaction tempera-
ture and pH of the solvent. For example, no noticeable peaks were
observed for the reaction at pH 7 at 250 ◦C (Fig. 2(e)), but broad
and large peaks appeared for the reaction at pH 7 at 150 ◦C, which
gradually shifted toward longer retention times as the reaction time
increased (Fig. 2(b)). This means that the depolymerization of algi-
nate was clearly accelerated by the reaction temperature. As the
temperature of reaction medium, neutral water, increased from
150 to 250 ◦C, the concentrations of hydrogen and hydroxyl ions
dissociated from water molecules rose a few times, resulting in the
promotion of alginate decomposition via acid- and base-catalysed
hydrolysis by hydrogen and hydroxyl ions, respectively. In addition
to the reaction temperature, the effects of catalysts on the alginate
degradation were observed as shown in Fig. 2(a–c). At 150 ◦C, the
broad and large peaks for the reaction at pH 7 almost disappeared
as the pH varied to 1 or 13, which indicates that the initial cata-
lyst addition increased the concentrations of hydrogen or hydroxyl
ions, and consequently enhanced the hydrolysis of alginate. Inter-
estingly, any noticeable peaks were not observed within 25 min
of retention time when the alginate was degraded with a strong
base catalyst at 150 ◦C (Fig. 2(c)). This demonstrates that a strong
base catalyst significantly enhances the rate of depolymerization
and shows better catalytic performance than that of a strong acid
catalyst. The raw alginate was largely decomposed after catalyst-
free hydrothermal treatment at 250 ◦C, but small broad peaks were
observed within 25 minutes of the retention time, as shown in
Fig. 2(e), underlining the effect of catalysts on the depolymerization
of alginate.
Fig. 2. GPC chromatograms of raw alginate and products obtained by hydrothermal
depolymerization of raw alginate: (a) pH 1, 150 ◦C, (b) pH 7, 150 ◦C, (c) pH 13, 150 ◦C,
(d) pH 1, 250 ◦C, (e) pH 7, 250 ◦C, (f) pH 13, 250 ◦C.
molecules having similar molecular weights was accelerated at
higher temperatures due to vigorous hydrolysis triggered by higher
concentrations of hydrogens and hydroxyl ions. The increasing
thermal energy gives rise to a decrease of relative dielectric permit-
tivity and an increase in ion products in subcritical water systems
[17,39], which catalyses the hydrolysis of alginate more effectively.
In addition to the reaction temperature, the average molecular
weights were drastically reduced when the alginate was depoly-
merized at 150 ◦C under acid or base catalysts, as evidenced by GPC
chromatograms. The rise of hydrogen and hydroxyl ion concen-
trations after the addition of catalysts increased the ion products
in the subcritical water, promoting hydrolysis, nucleophilic substi-
tution and elimination reactions. In particular, the products that
decomposed in the presence of a base catalyst exhibited much
lower average molecular weights (Mw, Mn) and polydispersity
index values than those with an acid catalyst, suggesting that
the base-catalysed elimination (known as a peeling reaction) is
more efficient for alginate decomposition. For example, when the
hydrothermal reaction occurred under the base catalyst, the val-
ues of Mw and Mn immediately converged to approximately 700
and 520, respectively. This means that the base-catalysed reac-
tion accelerated multiple and simultaneous cleavages of overall
glycosidic bonds in alginate polymer. However, under the acid cat-
alyst at 150 ◦C, Mw and Mn gradually decreased and PDI values
were higher than those values for base-catalysed reactions, which
indicates that the rate of alginate depolymerization was relatively
The average molecular weights (Mw, Mn) of products decreased
with increasing reaction temperatures from 150 to 250 ◦C, as
listed in Table 1. Polydispersity index (PDI) values of products
also decreased and converged toward 1 at higher temperatures,
strongly suggesting that the depolymerization of alginate into small

W. Jeon et al. / Journal of Molecular Catalysis A: Chemical 399 (2015) 106–113
109
Table 1
Weight average molecular weight (Mw), number average molecular weight (Mn)
and polydispersity index (PDI) values of raw alginate and hydrothermally-treated
alginate under different reaction conditions.
Sample
Mw (Da)
Mn (Da)
PDI
Raw alginate
277865
65185
4.26
pH 1–150 ◦
C
0 min
6 min
15 min
60 min
9453
4159
1396
1710
1269
1039
841
7.45
4
1.66
2.09
819
pH 7–150 ◦
C
0 min
268446
107320
2.5
6 min
15 min
60 min
129806
36506
3608
50913
15440
1870
2.55
2.36
1.93
pH 13–150 ◦
C
0 min
6 min
15 min
60 min
704
683
711
699
533
521
537
519
1.32
1.31
1.33
1.35
pH 1–250 ◦
pH 7–250 ◦
C
C
0 min
6 min
15 min
60 min
873
817
753
693
703
679
650
607
1.24
1.2
1.16
1.14
0 min
6 min
15 min
60 min
1367
1171
1251
1247
1009
905
994
962
1.35
1.29
1.26
1.30
pH 13–250 ◦
C
0 min
6 min
15 min
60 min
682
678
649
646
535
533
524
519
1.27
1.27
1.24
1.25
low and the acid-catalysed reaction induced random and partial
cleavages of glycosidic linkages.
The presence of GPC peaks after 27.5 min of retention time indi-
cates the breakdown of mannuronic and guluronic acid subunits
into smaller molecules due to cleavage of C C or C O linkages
within the monomer subunits. When sodium alginate decomposed
at pH 1 at 150 ◦C, the GPC peak from the monomer appeared
and then perished as reaction time increased (Fig. 2(a)). Coexis-
tence of GPC peaks from the monomer and other smaller products
implies that the acid-catalysed hydrolysis of sodium alginates
into monomers and the decomposition of monomers into small
organic compounds occur simultaneously. As shown in Fig. 2(c), the
monomers’ peak was not detected within 60 min at pH 13, show-
ing that the decomposition of monomers produced from sodium
alginate occurs extremely rapidly in the presence of a strong base
catalyst. However, in catalyst-free subcritical water (pH 7), the
monomers’ peak was observed after 60 min of reaction time. These
results were consistent with the 1H NMR analysis, as shown in
Fig. 3(a). The unique peaks for alginate between 4.40 and 5.17 ppm
were not observed for the reaction at pH 13 at 150 ◦C, whereas those
peaks clearly appeared when alginate was depolymerized at pH 1
and 7 at 150 ◦C. Instead of those unique peaks, alternative peaks
emerged between 1 and 4 ppm only for the sample treated at pH
13 for 0 min. Both NMR and GPC data provide evidence that sodium
alginate quickly converts to a complex organic mixture containing
hydroxyl and carboxyl groups smaller than monomers at 150 ◦C
under the strong base catalyst.
Fig. 3. 1H NMR spectra of raw alginate and hydrothermally treated alginate solu-
tions; H-1 of guluronic acid (ı = 4.95–5.17), H-5 of guluronic acid and H-1 of
mannuronic acid (ı = 4.55–4.82), H-5 of guluronic acid (ı = 4.40–4.50).
alginate at the higher temperature was also observed in 1H NMR
spectra as shown in Fig. 3(b). The characteristic peaks of alginate
disappeared completely and, at the same time, numerous other
peaks emerged, verifying the production of various organic com-
pounds from alginate decomposition. However, the catalysts were
still influential for the molecular distribution of depolymerized
alginate. Corresponding with the converged shapes of the chro-
matograms (Fig. 2(d–f)), the average molecular weights of products
obtained by the reaction at 250 ◦C under acid and base catalysts
were lower than those for products generated in neutral subcrit-
ical water, as listed in Table 1. This is strong evidence suggesting
that the catalyst-assisted hydrothermal treatment is still favorable
for depolymerizing sodium alginate at high reaction temperatures
(250 ◦C). It can be expected that the roles of catalysts in reducing the
molecular weight of sodium alginate may be relatively weakened
if reaction temperatures exceed 250 ◦C under supercritical water
conditions.
At 250 ◦C, the hydrothermal reactions under all pH condi-
tions were accelerated and the raw alginate was immediately
degraded as soon as the reaction initiated. As a result, shapes
of GPC chromatograms (Fig. 2(d–f)) quickly converged with the
average molecular weight values (Table 1). The properties of sub-
critical water at 250 ◦C changed significantly as mentioned above
and, as a result, strongly promoted the alginate decomposition
regardless of the addition of catalysts. The fast degradation of
After the depolymerization of sodium alginate through the
selective cleavage of 1,4-glycosidic bonds, the alginate was
expected to convert to monomers, followed by decomposition
of the monomers to organic products [25,27]. Fig. 4 shows yield
distributions of the monomers and organic compounds produced
by hydrothermal treatment at 150, 200 and 250 ◦C. Using a strong
acid catalyst (pH 1) at 150 ◦C, the monomers were instantly gen-
erated as the reaction started. This indicates that the selective

110
W. Jeon et al. / Journal of Molecular Catalysis A: Chemical 399 (2015) 106–113
cleavage of glycosidic linkages was promoted by acid-catalysed
hydrolysis. The production of monomers was also observed in
GPC data as shown in Fig. 2(a). The yield of monomers increased
and then decreased within a few minutes with the production
of furfural, glycolic acid and formic acid. This phenomenon was
explained by the further decomposition of monomers through
acid-catalysed hydrolysis and dehydration. Another research group
reported that mannuronic acid detaches from alginates faster than
guluronic acid [26]. Similarly, the quantity of mannuronic acid was
higher than that of guluronic acid in acidic and neutral subcrit-
ical water at 150 ◦C in this study. It indicates that the sequence
of monomer dissociation in subcritical water remains constant
regardless of the addition of an acid catalyst. On the other hand,
the rate of monomer production under a strong acid catalyst was
remarkably higher than the production rate in catalyst-free alginate
hydrothermal depolymerization.
Both monomers were not detected in the hydrothermal treat-
ment of sodium alginate in the presence of a strong base catalyst
(pH 13). Instead of monomers, carboxylic and dicarboxylic acids
were produced more abundantly, with the amount increasing as
reaction time increased. This suggests that a strong base cata-
lyst substantially promotes the decomposition of monomers via
the Lorby de Bruyn–Alberta van Ekenstein Transformation (LBET),
hydrolysis and dehydration reactions after the peeling reaction of
alginate [25]. As a result, the monomers liberated from sodium algi-
nate immediately converted into malic acid, fumaric acid, lactic
acid, formic acid, etc., as shown in Fig. 4(a). The production of lactic
and dicarboxylic acids was typical for a base-catalysed hydrother-
mal decomposition of alginate or other hexuronic acids [25,40]. In
contrast to the strong acid and base catalysts, at 150 ◦C, weak acid
(pH 3) and weak base (pH 11) catalysts did not show any remarkable
difference in performance compared to the catalyst-free reaction
(pH 7). Although the hydrothermal treatment of alginate in a pH
range from 3 to 11 produced monomers and organic acids, it
required at least 60 min to achieve that outcome.
Fig. 4(b) shows the yield distribution of organic compounds
produced at 200 ◦C. Monomers were almost consumed and the pro-
duction of organic compounds was enhanced in comparison to the
reaction at 150 ◦C. The production of formic acid, acetic acid and gly-
colic acid is common, and indicates that the reaction temperature
(200 ◦C) is sufficiently high to decompose monomers into smaller
organic acids regardless of the presence of an acid or base cata-
lyst. Interestingly, the yield distribution of organic products shows
clear differences between hydrothermal reactions under different
pH conditions. The strong acid catalyst (pH 1) assisted the produc-
tion of furfural, glycolic acid and small quantities of other organic
acids. Compared with hydrothermal treatment at pH 1, the quanti-
ties of succinic acid increased, whereas the quantities of furfural
and glycolic acid decreased in the pH range from 3 to 11. This
implies that the production of furfural and glycolic acid is a unique
feature of acid-catalysed hydrothermal alginate decomposition. On
the other hand, malic acid, fumaric acid and lactic acid were gen-
erated under a strong base catalyst (pH 13). This indicates that
mechanisms of monomer decomposition are completely different,
and are dependent upon the acidic or basic nature of the reaction
solvents.
As shown in Fig. 4(c), the overall yields of organic products for
all pH conditions were reduced as the reaction time increased,
in comparison with different reactions at 150 and 200 ◦C. Such
decrease indicates that thermally unstable compounds decom-
posed at 250 ◦C. For example, the yield of furfural decreased
drastically under the strong acid catalyst (pH 1). This decrease
was expected due to its low thermal stability in acidic aqueous
media. To ensure the decomposition of furfural in our exper-
imental system, the hydrothermal treatment of furfural in the
acidic subcritical water was performed with high-purity furfural
Fig. 4. Molar yields of monomers and organic acids produced by hydrothermal
decomposition. (a) 150 ◦C, (b) 200 ◦C, (c) 250 ◦C. Reaction times marked beneath
each column are as follows: 0, 3, 6, 9, 12, 15, 30 and 60 min.

W. Jeon et al. / Journal of Molecular Catalysis A: Chemical 399 (2015) 106–113
111
Table 2
Conversion and yields of organic acid products obtained from hydrothermal treatment of dicarboxylic acids in 250 ◦C of subcritical water at pH 13.
Starting Material (2 wt% in 3 mL of H2O)
Malic acid
3
Fumaric acid
3
Reaction time (min)
Conversion (%)
15
90.6
30
93.6
15
66.4
30
90.9
63
8.8
Compound
Formic acid
Acetic acid
Glycolic acid
Lactic acid
Succinic acid
Fumaric acid
Malic acid
Yield (mol%)
0
9.2
0.5
0.1
0.7
0
0
0
0
0
45.6
0.2
0.3
9.2
23
88.3
0.5
0.4
11.7
14.2
–
10.7
0.5
0
1.1
–
60.1
0.9
0.4
5.4
–
82.7
0.3
0.8
11.4
–
91.9
–
–
35.6
17.0
5.4
acid and fumaric acid, and depreciates the final product with the
large amount of acetic acid. Therefore, either short reaction times at
high temperatures or long reaction times at mild temperatures are
favorable for production of hydrothermally unstable compounds,
such as furfural, malic acid and fumaric acid.
However, lactic acid was generated under all of the pH
conditions at 250 ◦C without the decrease of yield, since the decom-
position of lactic acid is feasible in near critical water (320–400 ◦C)
[43]. The production of lactic acid was typical for the base-catalysed
hydrothermal depolymerization of alginate [25], and means that
the high reaction temperature induced a base-catalysed reaction
by changing the properties of the subcritical water and promot-
ing nucleophilic substitutions and elimination reactions. The high
quantities of lactic acid at pH 13 at all of the reaction temperatures
indicate that a strong base catalyst is very useful for the produc-
tion of lactic acid via hydrothermal treatment of alginate in a wide
range of temperature.
As summarised in Fig. 6, various organic compounds could be
produced by the hydrothermal treatment of alginate over acid and
base catalysts in subcritical water. The acid-catalysed reaction pro-
moted the production of furfural and glycolic acid, whereas the
base-catalysed reaction induced the production of lactic acid and
dicarboxylic acids, such as malic acid and fumaric acid. Catalyst-
free reaction at pH 7 showed poor performance in both alginate
depolymerization and organic acid production. However, the per-
formance was remarkably enhanced and all of the target products
were detected at reaction temperatures higher than 200 ◦C. This
indicates that the high reaction temperature induces both acid- and
base-catalysed hydrothermal reactions. Increasing reaction tem-
perature influenced reaction pathways of hydrothermal conversion
of alginate and changed the product distributions with the aid of
catalysts. For example, lactic acid was produced under the strong
acid catalyst at reaction temperatures higher than 200 ◦C, although
it is a typical product from base-catalysed hydrothermal decom-
position of carbohydrate biomass including alginate [25,28,44].
Decomposition of furfural in acidic subcritical water is considered
as a major factor for the production of lactic acid at pH 1, as shown
in Fig. 4(c) and Fig. 5. This means that lactic acid can be produced
by acid- or base-catalysed hydrothermal decomposition of alginate
at relatively high temperatures via totally different reaction path-
ways. Namely, in addition to catalysts, the reaction temperature
has considerable influence on reaction pathways in hydrothermal
conversion of alginate into organic compounds.
Fig. 5. Conversion of furfural and yields of products after hydrothermal treatment
of furfural at pH 1 and 250 ◦C.
(>99.0%). As expected, the conversion of furfural rapidly proceeded,
resulting in the production of small amounts of organic acids
and dark solid residues as shown in Fig. 5. The solid residues
are considered as humins formed by polymerisation of furfural,
known as “furfural resinification”, in the acidic hydrothermal con-
ditions. These insoluble particles are observed when hexose or
pentose are hydrothermally converted to furan compounds, such
as 5-hydroxymethyl furfural or furfural [41,42]. These insoluble
particles were also generated in small quantities when alginate
was hydrothermally treated in acidic reaction media. The produc-
tion of humins is a major obstacle to produce furfural from the
hydrothermal process, and further investigation for inhibiting the
humin formation is required to enhance the yield of furfural.
Yields of malic acid and fumaric acid were also reduced as reac-
tion temperatures increased from 200 to 250 ◦C at pH 13, as shown
in Fig. 4(c). This indicates that these particular acids are unstable
in that subcritical water condition. Those two dicarboxylic acids
were likely to convert into succinic acid and acetic acid at pH 13
at 250 ◦C, since it was observed that the yields of succinic acid and
acetic acid increased as the reaction time increased. To clarify this,
separate hydrothermal treatments of fumaric acid and malic acid
was conducted at 250 ◦C in subcritical water at pH 13, as listed in
Table 2. It was observed that malic acid and fumaric acid were con-
sumed as the yields of succinic acid and acetic acid increased. It is
believed that malic acid was rapidly transformed to fumaric acid by
dehydration reaction and then fumaric acid produced converted to
succinic acid and acetic acid. Although fumaric acid was hydrated to
form malic acid, the dehydration of malic acid was likely to be more
dominant in the reaction condition. To sum up, this high reaction
temperature under the strong base catalyst decomposes both malic
We compared the hydrothermal decomposition of cellulose
with that of alginate by using acid and base catalysts in our
efforts to produce valuable organic compounds. The cellulose was
mechanically milled for over 20 h to enhance the susceptibility of
crystalline cellulose to hydrothermal conversion, for effective con-
version in subcritical water [45]. As shown in Table 3, a monomer of
cellulose (glucose) also completely converted during hydrothermal

112
W. Jeon et al. / Journal of Molecular Catalysis A: Chemical 399 (2015) 106–113
Fig. 6. Reaction pathways of alginate decomposition by catalysts and reaction temperatures.
Table 3
catalyst promotes the production of mannuronic acid and guluronic
acid (monomers of alginate) at 150 ◦C, whereas the presence of
a strong base catalyst immediately converts the monomers to
carboxylic acids and dicarboxylic acids after the dissociation of
monomers from alginate. As the reaction temperature increases,
monomers completely convert to various organic compounds with
different acidic and basic distributions for the reaction solvents.
Furfural and glycolic acid was mainly produced by acid-catalysed
reactions, whereas lactic acid, fumaric acid and malic acid were
detected in base-catalysed reactions. Higher reaction tempera-
ture and longer reaction time were favorable for hydrothermal
decomposition of alginate, but it was observed that a few useful
compounds were transformed to less valuable products, such as
furfural to humins or C4 dicarboxylic acids to acetic acid. To sum
up, the acid and base catalysts are further influential in deter-
mining the distribution of final products, whereas the reaction
temperature is a more important factor for reducing the molec-
ular weight of alginate in hydrothermal conditions. Our research
demonstrates that an efficient selective production of value-added
organic compounds can be achieved by carefully adjusting the
reaction conditions such as pH, temperature and time. This will
be helpful in understanding the effects of solid catalysts on the
hydrothermal depolymerization of alginate, studies of which are
now in progress. It is expected that more selective production of
valuable organic chemicals will be possible if engineered solid cat-
alysts are applied using this depolymerization reaction.
Yields of organic compounds produced by hydrothermal treatment of sodium algi-
nate and cellulose at 150 ◦C for 30 min in the presence of a strong acid and base
catalyst.
Product
Yield (mol%)
pH 1
pH 13
Alginate
Cellulose
Alginate
Cellulose
Monomer
Furfural
4.3
1.7
0.0
0.0
2.5
0.0
0.0
0.5
0.0
6.6
3.3
0.0
0.0
0.0
0.0
0.0
0.6
0.9
0.0
0.0
3.1
0.4
0.6
2.8
3.5
4.9
1.1
0.0
0.0
0.0
0.0
0.2
0.0
4.8
0.0
2.8
Malic acid
Succinic acid
Glycolic acid
Fumaric acid
Lactic acid
Acetic acid
Formic acid
Total
9.0
13.2
16.4
7.8
treatment at pH 13, similar to the decomposition of mannuronic
acid and guluronic acid with a strong base catalyst. In addition,
monocarboxylic acids, such as lactic acid and formic acid, were
obtained from the hydrothermal conversion of both cellulose and
alginate. However, cellulose did not convert into dicarboxylic acids,
such as malic acid, succinic acid or fumaric acid. At pH 1, glucose
and furfural were produced significantly rather than organic acids.
These differences likely arose from an extra carboxylic functional
group in an alginate monomer. The acid-catalysed degradation of
cellulose showed better performance for production of furfural,
compared with alginate. Most final products serve as useful plat-
form chemicals in scientific and industrial fields [46]. Among those
products, furfural, lactic acid and succinic acid were selected as
important biomass-derived chemicals [47]. Prior to that, malic acid
and fumaric acid were also ranked as top value-added chemicals
from biomass in 2004, released by the US Department of Energy
(DOE) [48]. In that respect, alginate is regarded as a promising and
competitive biomass feedstock for production of furfural and value-
added organic acids, especially without treating it with mechanical
process like cellulose.
Acknowledgement
This work was supported by Ministry of Oceans and Fisheries,
Republic of Korea.
Appendix A. Supplementary data
Supplementary data associated with this article can be found,
in the online version, at http://dx.doi.org/10.1016/j.molcata.2015.
01.011.
References
4. Conclusions
[1] F.M. Kerton, Y. Liu, K.W. Omari, K. Hawboldt, Green Chem. 15 (2013) 860–871.
[2] K. Gao, K. McKinley, J. Appl. Phycol. 6 (1994) 45–60.
[3] P.J.L.B. Williams, L.M.L. Laurens, Energy Environ. Sci. 3 (2010) 554–590.
[4] A. Imerson, Food stabilisers, in: Thickeners and Gelling Agents,
Wiley-Blackwell, 2009.
Alginate represents a promising alternative feedstock to replace
cellulose for the production of furfural and organic acids. We found
that homogeneous acid and base catalysts accelerate the depoly-
merization of sodium alginate in subcritical water. A strong acid
[5] M.T. Cook, G. Tzortzis, V.V. Khutoryanskiy, D. Charalampopoulos, J. Mater.
Chem. B 1 (2013) 52–60.

W. Jeon et al. / Journal of Molecular Catalysis A: Chemical 399 (2015) 106–113
113
[6] A.C.S. Alcantara, P. Aranda, M. Darder, E. Ruiz-Hitzky, J. Mater. Chem. 20
[24] P. Gacesa, A. Squire, P.J. Winterburn, Carbohydr. Res. 118 (1983) 1–8.
[25] K. Niemela, E. Sjostrom, Carbohydr. Res. 144 (1985) 241–249.
(2010) 9495–9504.
[7] C. Riordan, G. Love, N. Barron, P. Nigam, R. Marchant, L. McHale, A.P. McHale,
Bioresour. Technol. 55 (1996) 171–173.
[26] T.M. Aida, T. Yamagata, M. Watanabe, R.L. Smith, Carbohydr. Polym. 80 (2010)
296–302.
[8] M. Gimeno-Fabra, M. Peroglio, D. Eglin, M. Alini, C.C. Perry, J. Mater. Chem. 21
(2011) 4086–4089.
[27] T.M. Aida, T. Yamagata, C. Abe, H. Kawanami, M. Watanabe, R.L. Smith, J.
Supercrit. Fluids 65 (2012) 39–44.
[9] H. Takeda, F. Yoneyama, S. Kawai, W. Hashimoto, K. Murata, Energy Environ.
Sci. 4 (2011) 2575–2581.
[28] D. Esposito, M. Antonietti, ChemSusChem 6 (2013) 989–992.
[29] S.J. Dee, A.T. Bell, ChemSusChem 4 (2011) 1166–1173.
[30] X. Qi, H. Guo, L. Li, R.L. Smith, ChemSusChem 5 (2012) 2215–2220.
[31] F. Tao, H. Song, L. Chou, J. Mol. Catal. A: Chem. 357 (2012) 11–18.
[32] F. Jiang, Q. Zhu, D. Ma, X. Liu, X. Han, J. Mol. Catal. A: Chem. 334 (2011) 8–12.
[33] K.K. Yadav, S. Ahmad, S.M.S. Chauhan, J. Mol. Catal. A: Chem. 394 (2014)
170–176.
[34] B.L. Arne Haug, Acta Chem. Scand. 17 (1963) 1466–1468.
[35] B.L. Arne Haug, O. Smidsrød, Acta Chem. Scand. 20 (1966) 8.
[36] B.L. Arne Haug, O. Smidsrød, Acta Chem. Scand. 21 (1967) 12.
[37] B.L. Arne Haug, O. Smidsrød, Acta Chem. Scand. 21 (1967) 14.
[38] Z. Yang, J.-P. Li, H.-S. Guan, Carbohydr. Polym. 58 (2004) 115–121.
[39] N. Akiya, P.E. Savage, Chem. Rev. 102 (2002) 2725–2750.
[40] K. Niemelä, E. Sjöström, Carbohydr. Res. 144 (1985) 93–99.
[41] M.L. Granados, A.C. Alba-Rubio, I. Sadaba, R. Mariscal, I. Mateos-Aparicio, A.
Heras, Green Chem. 13 (2011) 3203–3212.
[42] B. Danon, G. Marcotullio, W. de Jong, Green Chem. 16 (2014) 39–54.
[43] C.T. Lira, P.J. McCrackin, Ind. Eng. Chem. Res. 32 (1993) 2608–2613.
[44] C. Sánchez, I. Egüés, A. García, R. Llano-Ponte, J. Labidi, Chem. Eng. J. 181–182
(2012) 655–660.
[45] D.M. Alonso, J.Q. Bond, J.A. Dumesic, Green Chem. 12 (2010) 1493–1513.
[46] P. Gallezot, Chem. Soc. Rev. 41 (2012) 1538–1558.
[47] J.J. Bozell, G.R. Petersen, Green Chem. 12 (2010) 539–554.
[48] T. Werpy, G. Petersen, Top value added chemicals from biomass Results of
Screening for Potential Candidates from Sugars and Synthesis Gas, vol. 1, U. S.
Department Energy, 2004, pp. 1–76.
[10] A.J. Wargacki, E. Leonard, M.N. Win, D.D. Regitsky, C.N.S. Santos, P.B. Kim, S.R.
Cooper, R.M. Raisner, A. Herman, A.B. Sivitz, A. Lakshmanaswamy, Y.
Kashiyama, D. Baker, Y. Yoshikuni, Science 335 (2012) 308–313.
[11] M. Enquist-Newman, A.M.E. Faust, D.D. Bravo, C.N.S. Santos, R.M. Raisner, A.
Hanel, P. Sarvabhowman, C. Le, D.D. Regitsky, S.R. Cooper, L. Peereboom, A.
Clark, Y. Martinez, J. Goldsmith, M.Y. Cho, P.D. Donohoue, L. Luo, B.
Lamberson, P. Tamrakar, E.J. Kim, J.L. Villari, A. Gill, S.A. Tripathi, P.
Karamchedu, C.J. Paredes, V. Rajgarhia, H.K. Kotlar, R.B. Bailey, D.J. Miller, N.L.
Ohler, C. Swimmer, Y. Yoshikuni, Nature 505 (2014) 239–243.
[12] S. Yin, Y. Pan, Z. Tan, Int. J. Green Energy 8 (2011) 234–247.
[13] F.M. Jin, J. Yun, G.M. Li, A. Kishita, K. Tohji, H. Enomoto, Green Chem. 10 (2008)
612–615.
[14] K. Anastasakis, A.B. Ross, Bioresour. Technol. 102 (2011) 4876–4883.
[15] M. Sasaki, B. Kabyemela, R. Malaluan, S. Hirose, N. Takeda, T. Adschiri, K. Arai,
J. Supercrit. Fluids 13 (1998) 261–268.
[16] M. Bicker, S. Endres, L. Ott, H. Vogel, J. Mol. Catal. A: Chem. 239 (2005)
151–157.
[17] M. Möller, P. Nilges, F. Harnisch, U. Schröder, ChemSusChem 4 (2011)
566–579.
[18] Arne Haug, Acta Chem. Scand. 16 (1962) 1908–1918.
[19] H. Grasdalen, B. Larsen, O. Smidsrød, Carbohydr. Res. 68 (1979) 23–31.
[20] Y.-B. Huang, Y. Fu, Green Chem. 15 (2013) 1095–1111.
[21] R. Rinaldi, F. Schüth, ChemSusChem 2 (2009) 1096–1107.
[22] P.L. Dhepe, A. Fukuoka, ChemSusChem 1 (2008) 969–975.
[23] D. Klemm, B. Heublein, H.-P. Fink, A. Bohn, Angew. Chem. 117 (2005)
3422–3458.