Aqueous phase catalytic conversion of agarose to 5-hydroxymethylfurfural by metal chlorides

Article abstract of DOI:10.1039/c3ra43293a

The production of 5-hydroxymethylfurfural (5-HMF) from agarose catalyzed by metal chlorides was studied in aqueous phase. A series of metal chlorides, including NaCl, CaCl₂, MgCl₂, ZnCl₂, CrCl ₃, CuCl₂ and FeCl₃, were comparatively investigated to catalyze agarose degradation for the production of 5-HMF at temperatures of 180°C, 200 °C and 220 °C, catalyst concentration of 0.5% (w/w), 1% (w/w) and 5% (w/w), time of 0-50 min, and substrate concentration of 2% (w/w). Results revealed that alkali and alkaline earth metal chlorides, including NaCl, CaCl₂ and MgCl₂, resulted in higher 5-HMF yields from agarose with negligible amount of byproducts, such as levulinic acid and lactic acid, derived from further degradation reactions. 1% (w/w) MgCl ₂ was the most efficient catalyst among the tested metal chlorides for 5-HMF production from agarose and resulted in both the highest yield of 40.7% and highest selectivity of 49.1% at 200 °C for 35 min. The cleavage of C-O-C bond in agarose with subsequent isomerization of galactose to its ketose was considered as a possible mechanism for formation of 5-HMF under MgCl ₂ catalyzed conditions.

Full text of DOI:10.1039/c3ra43293a

RSC Advances
View Article Online
View Journal | View Issue
PAPER
Aqueous phase catalytic conversion of agarose to
5-hydroxymethylfurfural by metal chlorides
Cite this: RSC Adv., 2013, 3, 24090
a
a
b
a
*
Lishi Yan, Dhrubojyoti D. Laskar, Suh-Jane Lee and Bin Yang
The production of 5-hydroxymethylfurfural (5-HMF) from agarose catalyzed by metal chlorides was studied
in aqueous phase. A series of metal chlorides, including NaCl, CaCl2, MgCl2, ZnCl2, CrCl3, CuCl2 and FeCl3,
were comparatively investigated to catalyze agarose degradation for the production of 5-HMF at
temperatures of 180 ꢀC, 200 ꢀC and 220 ꢀC, catalyst concentration of 0.5% (w/w), 1% (w/w) and 5%
(w/w), time of 0–50 min, and substrate concentration of 2% (w/w). Results revealed that alkali and
alkaline earth metal chlorides, including NaCl, CaCl2 and MgCl2, resulted in higher 5-HMF yields from
agarose with negligible amount of byproducts, such as levulinic acid and lactic acid, derived from
further degradation reactions. 1% (w/w) MgCl2 was the most efficient catalyst among the tested metal
chlorides for 5-HMF production from agarose and resulted in both the highest yield of 40.7% and
highest selectivity of 49.1% at 200 ꢀC for 35 min. The cleavage of C–O–C bond in agarose with
subsequent isomerization of galactose to its ketose was considered as a possible mechanism for
formation of 5-HMF under MgCl2 catalyzed conditions.
Received 28th June 2013
Accepted 4th October 2013
DOI: 10.1039/c3ra43293a
www.rsc.org/advances
amansii (a red algae species) is around 60–65%.5 Even though
agarose exists in a gel form caused by the intra- and inter-
1. Introduction
The decreasing reserves of fossil fuels and the growing green-
house gas emissions have caused an increase in bioenergy and
biochemical research in recent years.1 Biomass-derived carbo-
hydrates represent a promising carbon-based alternative as an
energy source and a sustainable feedstock. Substantial efforts
have recently been devoted towards converting biomass to
5-HMF, an important chemical building block used to make a
wide variety of chemicals (e.g. plastics, polymers, etc.) and fuels
that are currently derived from petroleum.2 However, the
current processes of 5-HMF production are primarily depen-
dent on edible monosaccharides such as fructose and
glucose.2,3 Using cellulosic biomass as the feedstock for 5-HMF
production has been also extensively studied over the years.
However, technical challenges in developing an efficient and
economic viable process for commercial application still
remain.2,4
Recently, polysaccharides from algal biomass were recom-
mended as an alternative feedstock for the large scale produc-
tion of biofuel and biochemicals because of their substantial
abundance, negligible food value, as well as use of non-arable
land.5,6 Agarose, mainly composed of galactose and 3,6-anhy-
drogalactose,7,8 is one of major polysaccharides stored in algal
biomass.5,9 For example, the agarose content in Gelidium
molecular hydrogen bonds at room temperature, these
hydrogen bonds can be dissociated when the temperature
exceeds 90 C,7 leaving agarose more susceptible to decompo-
ꢀ
sition. Furthermore, 3,6-anhydrogalactose with highly strained
additional cyclic unit in agarose presents relatively high reac-
tivity and could readily lead to the conversion of agarose into
degradation compounds.5
Various reaction media, including organic solvent/organic-
water biphasic systems, ionic liquids and the aqueous phase,
were used for 5-HMF production. The use of organic solvents or
ionic liquids as reaction medium has received signicant
attention because relatively high 5-HMF yield of 50–70% can be
realized.10,11 However, these reaction media present some limi-
tations for the large scale production of 5-HMF. Although the
5-HMF yield was increased in the organic solvent dimethyl
sulfoxide (DMSO), 5-HMF is highly soluble in DMSO which has
a high boiling point, thus prevents economical separation of
5-HMF.12 Although ionic liquids are reported to enhance the
5-HMF yields, ionic liquids are expensive and require a high
level of purity to maintain their physical properties. Thus,
industrial applications of ionic liquids are limited.13,14
Production of 5-HMF in aqueous phase (i.e. water) instead of
these expensive and difficult-to-separate solvents will signi-
cantly reduce its production cost. The physical properties of
ꢀ
water at elevated temperatures (e.g. 200 C) are different from
aBioproducts, Sciences and Engineering Laboratory, Department of Biological Systems
Engineering, Washington State University, Richland, WA, 99354, USA. E-mail:
binyang@tricity.wsu.edu; Fax: +1 509 372 7690; Tel: +1 509 372 7640
bPacic Northwest National Laboratory, Richland, WA, 99354, USA
those under ambient conditions.15 For example, the solubility of
organic compounds in water can increase at high temperatures
(above 100 ꢀC) due to the lower dielectric constant of water.16,17
24090 | RSC Adv., 2013, 3, 24090–24098
This journal is ª The Royal Society of Chemistry 2013

View Article Online
Paper
RSC Advances
Furthermore, protons released from water at high tempera- selectivity due to the formation of undesired byproducts such as
tures18–20 could accelerate the formation of 5-HMF to some levulinic acid37 and lactic acid.38 Rasrendra et al.38 reported poor
extent.21 However, the degradation of saccharides (e.g. fructose 5-HMF selectivity (around 3–27%) when catalyzing glucose under
ꢀ
and glucose) in water-only conditions proceeds with poor 140 C for 6 h. The use of alkali and alkaline earth metal chlo-
selectivity to 5-HMF production, resulting in many byproducts. rides for 5-HMF production from mono-, oligo-, or poly-
Byproducts, such as glycolaldehyde, glyceraldehyde, and lactic saccharides were seldom reported. Nevertheless, Liu and
acid, were generated from the retro aldol reaction of hexose (e.g. Wyman42 investigated the degradation of xylotriose with 0.8%
glucose and fructose), while others including levulinic acid and (w/w) NaCl, CaCl2 and MgCl2 under hot water conditions at
ꢀ
formic acid were generated due to the further rehydration of 180 C and found that the rate constants of xylotriose degrada-
5-HMF.18–20 Formation of byproducts signicantly reduced the tion to furfural (k) using CaCl2 and MgCl2 were around 2.5 fold
5-HMF yield.18–20 Catalysts used for 5-HMF production in greater than that with water-only treatment. In addition, Sabesan
aqueous phase were mostly limited to mineral acids (e.g. H2SO4, and Spado43 obtained 66% furfural yield from biomass with 1%
H3PO4, and HCl),15,22 organic acids (e.g. oxalic acid, r-toluene- (w/w) CaCl2 (160 ꢀC, 30 min, water and THF mixture (1 : 1)
sulfonic acid),15 H-form zeolites,23 strong acid cation exchange system), which was 18% higher than that without CaCl2. The
resins,24,25 and solid metal phosphates.26,27 However, mineral potential of applying alkaline earth metal/alkali metal chlorides
acid and organic acid possess the catalytic ability of accelerating on decomposing polysaccharides (e.g. agarose) to 5-HMF needs
the rehydration of 5-HMF thereby limiting its yield;15,22 hetero- more studies.
geneous catalysts, such as H-form zeolites, strong acid cation
Further efforts are needed to realize enhanced selectivity and
exchange resins and solid metal phosphates, are oen not yield of 5-HMF in aqueous phase using efficient catalysts. The
suitable to catalyze water-insoluble polysaccharides (e.g. cellu- objective of this study was to investigate effects of different
lose) into degradation products (e.g. 5-HMF) due to mass groups of metal chlorides, including alkali metal chlorides,
transfer limitation.28-30
alkaline earth metal chlorides, and transition metal chlorides,
In recent years, metal chlorides were shown effective for on the degradation of agarose to 5-HMF, in order to improve
transforming saccharides into 5-HMF because they can assist the 5-HMF yield with less byproducts in aqueous phase using non-
isomerization of aldose to ketose (e.g. glucose to fructose),31,32 food based feedstocks. The mechanism of agarose to 5-HMF
which is a key step for 5-HMF production.3 In addition, as with desired metal chlorides was further discussed.
homogeneous catalysts,33 metal chlorides possess the potential of
being easily separated from reaction products such as 5-HMF
through semipermeable membranes. For example, Kotraro
et al.34 reported that the use of semipermeable composite
2. Materials and methods
2.1. Feedstocks, catalysts and standards
membranes for separating organic compounds (molecular
weight < 300) such as furan from aqueous inorganic salts (e.g.
NaCl) containing solutions. Use of CrCl3 for 5-HMF production
was reported by various authors. Qi et al.35 reported that 70%
glucose was converted into 5-HMF when 99% glucose was
degraded with 1.5% (w/w) CrCl3 in ionic liquid solvents. Simi-
larly, Li et al.36 also presented that 0.5% (w/w) CrCl3 led to 60%
5-HMF yield from cellulose substrate in ionic liquid under
microwave irradiation. However, with same catalyst (CrCl3) in
aqueous phase, the yield of 5-HMF was decreased to around
5%.37 Catalysis of cellulose by 0.01 M CrCl3 at 180 ꢀC for 120 min
led to 5% 5-HMF yield and 30% levulinic acid yield.37 In addition
to CrCl3, several other kinds of metal chlorides were studied for
5-HMF production in aqueous phase. 16.1% 5-HMF yield was
achieved when 63% (w/w) ZnCl2 was applied to 5-HMF produc-
tion from glucose in aqueous phase at 120 ꢀC.31 The limited
5-HMF yield from saccharides (e.g. cellulose and glucose) with
Agarose (BP160, Fisher Scientic) was purchased from Fisher
Scientic, Pittsburgh, PA. Galactan was dened as the unit of
agarose and the purity is 98%. The galactan content was
determined following the Laboratory Analytical Procedure (LAP)
of “Determination of Structural Carbohydrates and Lignin in
Biomass”.44 This procedure employed a two-step acid hydro-
lysis: (1) about 300 mg substrate was placed into a vial and
ꢀ
hydrolyzed in 72% (w/w) sulfuric acid at 30 C for 2 hours; (2)
the substrate was further hydrolyzed in 4% (w/w) sulfuric acid at
121 ꢀC for 1 hour. Sugars in the liquid were determined by
HPLC. NaCl, CaCl2, MgCl2, ZnCl2, CrCl3, CuCl2 and FeCl3 (Alfa
Aesar, Ward Hill, MA) were used as the catalysts for 5-HMF
production. All standard chemicals were purchased from
Sigma-Aldrich, St. Louis, Mo.
2.2. Experimental methods
ZnCl2 was mainly attributed to the formation of byproduct lactic Batch tubular reactors (1.27 cm OD ꢁ 15.24 cm length, 0.0889
acid during the process.38,39 Seri et al.40 reported that a 20% cm wall thickness, Hastelloy C-276) (Swagelok Northwest,
5-HMF yield was obtained from cellulose using 0.25% (w/w) LaCl3 Richland, WA) were used in this study. The total volume of each
in aqueous phase at 250 ꢀC within 2.5 min, and 5% levulinic acid reactor was 13.6 ml with 10 ml of working volume.
was simultaneously formed. De et al.41 applied AlCl3 to catalyze
0.2 g of substrate was loaded into the batch tubular reactor
conversion of insulin and starch to 5-HMF. Around 30% 5-HMF with 10 ml of water, sulfuric acid (e.g. 0.5% (w/w), 1% (w/w), 5%
yield was obtained from these polysaccharides at 120 ꢀC within 5 (w/w)) or metal chlorides (e.g. 0.5% (w/w), 1% (w/w) or 5%
min,41 and byproducts levulinic acid were considered as the cause (w/w)), respectively. Reactors were heated to reach the target
for limiting the 5-HMF yields.41 These studies of using transition temperature within 1 min in a 4 kW uidized sand bath (model
metal chlorides catalyzed reaction suggested low 5-HMF SBL-2D, Omega engineering, Inc., Stamford, CT) and treated at
This journal is ª The Royal Society of Chemistry 2013
RSC Adv., 2013, 3, 24090–24098 | 24091

View Article Online
RSC Advances
Paper
target temperatures. The reactions were quenched within 1 min
by immersing the batch tubular reactor in cold water bath aer
being subjected to the target reaction temperature for a speci-
ed time. Aer cooling the tubular reactor, the caps and plugs
were removed, samples were pushed out and separated into
liquid hydrolyzate and solid residue by vacuum ltration using
a 0.22 mm glass ber lter (Fisher Science, Pittsburgh, PA) for
analysis.
Galactose yield:
YGa ð%Þ ¼
5-HMF yield:
YHMF ð%Þ ¼
CGa ꢁ MWGan
CGan ꢁ MWGa
ꢁ 100%
(5)
CHMF ꢁ MWGan
CGan ꢁ MWHMF
ꢁ 100%
(6)
(7)
(8)
(9)
Levulinic acid yield:
2.3. Analytical methods
CLA ꢁ MWGan
YLA ð%Þ ¼
ꢁ 100%
CGan ꢁ MWLA
2.3.1. Liquid analysis
HPLC analysis. Galactose, 5-HMF, levulinic acid, formic acid,
glyceraldehyde, dihydroxyacetone and lactic acid in aqueous
solution were analyzed through Waters HPLC system (model
2695) equipped with a 410 refractive detector and a Waters 2695
auto-sampler using Waters Empower Build 1154 soware
(Waters Co., Milford, MA). Bio-Rad Aminex HPX-87H column
(Bio-Rad Laboratories, Hercules, CA) were operated under 65 ꢀC
for separation and quantication of compounds. Mobile phase
Formic acid yield:
CFA ꢁ MWGan
CGan ꢁ MWFA
YFAð%Þ ¼
ꢁ 100%
Glyceraldehyde yield:
CGly ꢁ MWGan
À
Á
YGly ð%Þ ¼
ꢁ 100%
ꢁ 100%
ꢁ 100%
CGan ꢁ 2 ꢁ MWGly
was 0.005 M H2SO4 with ow rate of 0.6 ml minꢂ1
.
Dihydroxyacetone yield:
GC-MS analysis. The collected samples were diluted with
acetone then analyzed with an Agilent gas chromatography
mass spectrometer (GC-MS; GC, Agilent 7890A; MS, Agilent
5975C) equipped with a DB-5MS column (30 m ꢁ 250 mm ꢁ
0.25 mm).45 The oven temperature was programmed from 40 ꢀC
to 300 ꢀC at a ramping rate of 10 ꢀC minꢂ1. Both the initial and
nal temperatures were held for 5 minutes. The ow rate of the
CDih ꢁ MWGan
YDih ð%Þ ¼
(10)
(11)
CGan ꢁ ð2 ꢁ MWDih
Þ
Lactic acid yield:
YLac ð%Þ ¼
CLac ꢁ MWGan
CGan ꢁ ð2 ꢁ MWLac
Þ
carrier gas (helium) was 1.3 ml minꢂ1
.
In these equations, CGan is the initial concentration of gal-
actan; CGa is the concentration of galactose; CHMF is the
concentration of 5-HMF; CLA is the concentration of levulinic
acid; CFA is the concentration of formic acid; CGly is the concen-
tration of glyceraldehyde; CDih is the concentration of dihy-
droxyacetone; CLac is the concentration of lactic acid. MW is the
molecular weight: MWGan ¼ 162; MWGa ¼ 180; MWHMF ¼ 126;
MWLA ¼ 116; MWFA ¼ 46; MWGly ¼ 90; MWDih ¼ 90; MWLac ¼ 90.
The agarose conversion was evaluated as below:
2.3.2. Solid analysis. The agarose in solid residue aer
separation was analyzed based on a standard analysis proce-
dure developed by National Renewable Energy Laboratory
(NREL).44 This procedure was a two-step acid hydrolysis: the
sample was rst treated with 72% (w/w) H2SO4 at 30 ꢀC for 1 h;
the reaction mixture was subsequently diluted to 4% (w/w)
H2SO4 and autoclaved at 121 ꢀC for 1 hour. The sugars in liquid
aer this two step procedure were then determined by HPLC.
Y0 ꢂ Yr
2.4. Calculation
X ¼
ꢁ 100%
(12)
Y0
The overall reactions include: galactan (galactan is dened as
the unit of agarose) to galactose; galactose to 5-HMF; 5-HMF to
levulinic acid and formic acid; galactose to glyceraldehyde,
dihydroxyacetone or lactic acid. These reactions can be
expressed as eqn (1)–(4), respectively:
X is the conversion of agarose, Y0 (g) is the initial weight of
agarose, Yr (g) is the agarose in solid residue.
The selectivity of degradation compounds from agarose was
calculated as below:
Y
S ¼ ꢁ 100%
(13)
C6H10O5 + H2O / C6H12O6
C6H12O6 / C6H6O3 + 3H2O
C6H6O3 + 2H2O / C5H8O3 + CH2O2
C6H12O6 / 2C3H6O3
(1)
(2)
(3)
(4)
X
S is the selectivity of product; Y is the yield of product; X is the
conversion of agarose.
3. Results and discussion
3.1. Catalytic effects of metal chlorides on the products
distribution and 5-HMF yield from agarose
The yields of degradation compounds were based on the
original amount of galactan. Concentrations (g lꢂ1) of galactose,
5-HMF, levulinic acid, formic acid, glyceraldehyde, dihydroxy- The effects of 0.5–5% (w/w) metal chlorides including alkali
acetone and lactic acid measured by HPLC were converted to metal chlorides (NaCl), alkaline earth metal chlorides (CaCl2
yields as a percent of the theoretical maximum, as follows:
and MgCl2), and transition metal chlorides (ZnCl2, CuCl2, FeCl3
24092 | RSC Adv., 2013, 3, 24090–24098
This journal is ª The Royal Society of Chemistry 2013

View Article Online
Paper
RSC Advances
and CrCl3), on the products distribution from 2% (w/w) agarose whereas the yields of glyceraldehyde and dihydroxyacetone
were investigated at varied temperatures ranging from 180– continuously increased, which further reduced the 5-HMF yield.
220 ꢀC. Water-only and dilute sulfuric acid treatments were The distribution of galactose, 5-HMF, and various byproducts
used as references. The 5-HMF yields from agarose catalyzed by under water-only conditions showed a similar trend as when
these metal chlorides during the time course of 0–50 min were catalyzed by ZnCl2 and CrCl3 over the temperature range of
also investigated. In addition, the 5-HMF selectivity obtained by 180 ꢀC to 220 ꢀC. However, results indicated both slower
varied metal chlorides and the reference conditions (water-only formation and degradation of such compounds. Other transi-
and dilute sulfuric acid) was also studied.
tion metal chlorides used in this study, including CuCl2 and
3.1.1. Effects of temperature. Fig. 1 shows the effects of 1% FeCl3, resulted in low yields of galactose (2.2–19.4%) and
(w/w) metal chlorides on the distribution of products decom- signicant quantity of levulinic acid (27.3–41.2%), but negli-
ꢀ
ꢀ
ꢀ
posed from agarose at 180 C, 200 C and 220 C for 30 min. gible amount of 5-HMF at all tested temperatures. The
Transition metal chlorides ZnCl2 and CrCl3 resulted in various decomposition of agarose catalyzed by H2SO4 showed similar
compounds such as galactose, 5-HMF, lactic acid, glyceralde- pattern of products distribution compared with CuCl2 and
hyde and dihydroxyacetone, which was similar to water-only FeCl3: only galactose (0–8.5%) and levulinic acid (38.6–40.8%)
ꢀ
treatment. In addition to these compounds, CrCl3 led to the were observed at temperatures ranging from 180–220 C. The
formation of levulinic acid. Results suggested that transition formation of diverse byproducts (e.g. lactic acid, levulinic acid,
metal chlorides ZnCl2 and CrCl3 could catalyze the degradation etc.) catalyzed by transition metal chlorides contributed to the
process via both reactions of dehydration18 and retro aldol38 to relatively lower or negligible 5-HMF yield. Temperature (180–
these multiple compounds. Rasrendra et al.38 reported that 220 ꢀC) had negligible effects on the types of generated products
multiple products, including lactic acid and 5-HMF, were when using transition metal chlorides for the agarose degra-
observed from the decomposition of glucose in aqueous phase dation. However, temperatures inuenced the reaction rates of
at 140 ꢀC for 6 h using 6.8% (w/w) ZnCl2 and 6.1% (w/w) CrCl2. these degradation products, thereby impacting the yields of
Levulinic acid yields were reported at 6.1% (w/w) using CrCl2 5-HMF and overall byproducts.
only. Temperature played an important role in determining the
Addition of alkali and alkaline earth metal chlorides NaCl,
distribution of these degradation compounds from agarose CaCl2 and MgCl2 merely led to 5-HMF and galactose at all tested
catalyzed by ZnCl2, CrCl3, as well as that under water-only temperatures. The distribution of galactose and 5-HMF
conditions. When temperature was raised from 180 ꢀC to decomposed from agarose catalyzed by NaCl, CaCl2 and MgCl2
ꢀ
200 C, galactose yields decreased sharply from 22.6% to 2.7% over the three temperatures (180 ꢀC, 200 ꢀC and 220 ꢀC) fol-
ꢀ
with ZnCl2, and from 5.3% to 0.3% with CrCl3, while the yields lowed a similar trend. At 180 C, 29.3–32.5% 5-HMF and 37.5–
of 5-HMF catalyzed with ZnCl2 and CrCl3 increased from 15.9% 38.8% galactose yields were obtained with these alkali and
and 9.8% at 180 ꢀC to the highest yields of 20.9% and 12.9% at alkaline earth metal chlorides. As temperature increased to
ꢀ
200 ꢀC, respectively. The yields of byproducts also increased as 200 C, 5-HMF yield increased to 31.7–39.5%, while the galac-
ꢀ
ꢀ
temperature was elevated from 180 C to 200 C, thus limiting tose yield decreased to 19.1–23.1%. Yields of 5-HMF and
the production of 5-HMF. When the temperature was further galactose sharply decreased to 19.7–22.5% and 2.4–8.5%,
raised to 220 ꢀC, 5-HMF yields with ZnCl2 and CrCl3 decreased; respectively, when temperature was further raised to 220 ꢀC.
the yields of byproducts such as lactic acid declined as well; The reduced yield of 5-HMF could be attributed to the forma-
tion of humins, which were observed by Patil and Lund46 as the
decomposed product of 5-HMF under 135 ꢀC in aqueous phase.
The highest 5-HMF yields obtained with NaCl, CaCl2 and MgCl2
ꢀ
were 31.7%, 35.8% and 39.6% at 200 C, respectively.
3.1.2. Effects of catalyst loading. The distribution of
degradation products from agarose with metal chlorides was
also investigated under different catalyst loadings. The effects
of concentrations of the metal chlorides (0.5% w/w, 1% w/w and
5% w/w), including NaCl, CaCl2, MgCl2, ZnCl2, CrCl3, CuCl2 and
FeCl3, as well as H2SO4 treatment (0.5% w/w, 1% w/w and 5%
w/w) on degradation of agarose at 200 ꢀC for 30 min are shown
in Fig. 2. Only galactose and 5-HMF were obtained from agarose
using alkali and alkaline earth metal chlorides (NaCl, CaCl2 and
MgCl2) over the three catalyst loadings (0.5% w/w, 1% w/w and
5% w/w). Galactose yields decreased with the elevated loading of
these alkali and alkaline earth metal chlorides. 5-HMF yields
from agarose using these three catalysts increased to the high-
Fig. 1 Distribution of agarose decomposition products by various metal chlo-
est value, which were 31.7%, 35.8%, 39.6%, respectively, when
rides, water-only, H2SO4 treatments under different temperature (180 ꢀC, 200 ꢀ
C
loading was increased from 0.5% (w/w) to 1% (w/w). As the
catalyst loading was further increased to 5% (w/w), the 5-HMF
yield decreased sharply.
and 220 ꢀC). Conditions: 30 min, 1% (w/w) metal chlorides, water-only or 1%
(w/w) H2SO4. Lac: lactic acid; Gly: glyceraldehyde; Dhy: dihydroxyacetone; LA:
levulinic acid; 5-HMF: 5-hydroxymethylfurfural; Ga: galactose.
This journal is ª The Royal Society of Chemistry 2013
RSC Adv., 2013, 3, 24090–24098 | 24093

View Article Online
RSC Advances
Paper
which were 22.6% at 25 min and 12.9% at 30 min, respectively.
However, similar to CuCl2 and FeCl3, the 5-HMF yield catalyzed
by ZnCl2 and CrCl3 declined sharply as reaction time was pro-
longed. These observations suggested that transition metal
chlorides catalyzed 5-HMF production were sensitive to reaction
time, which required the reaction system with better control of
residence time (e.g. owthrough reactor).
5-HMF yields catalyzed by alkali and alkaline earth metal
chlorides NaCl, CaCl2, MgCl2, and water-only treatments
increased steadily over time, then declined slowly at around 35–
40 min. The highest yields of 5-HMF catalyzed by NaCl, CaCl2 or
with water-only treatment were 32.8%, 37.2% and 27.4% at 40
min, respectively. MgCl2 resulted in the highest yield of 5-HMF
40.7% at 35 min. Results implied that alkali and alkaline earth
metal chlorides (NaCl, CaCl2 and MgCl2) behaved with suitable
catalytic activity for the conversion of agarose to 5-HMF, while
simultaneously maintaining the relative stability of 5-HMF as
reaction prolonged.
Fig. 2 Distribution of agarose decomposition products by various metal chlo-
rides, water-only and H2SO4 treatments under different loading (0.5% (w/w), 1%
(w/w), 5% (w/w)). Conditions: 200 ꢀC for 30 min. Lac: lactic acid; Gly: glyceral-
dehyde; Dhy: dihydroxyacetone; LA: levulinic acid; 5-HMF: 5-hydroxy-
methylfurfural; Ga: galactose.
3.1.4. Comparison of the highest 5-HMF yield with varied
metal chlorides and the corresponding selectivity. 5-HMF yields
catalyzed by metal chlorides (NaCl, CaCl2, MgCl2, ZnCl2, CrCl3,
CuCl2 and FeCl3) were optimized under temperature 180–220 ꢀC,
Transition metal chlorides with loadings of 0.5% (w/w), 1% time 0–50 min and catalyst loading 0.5–5% (w/w), respectively.
(w/w) and 5% (w/w) led to various kinds of byproducts. Besides Table 1 summarized and compared the highest 5-HMF yields
galactose and 5-HMF, 0.5% (w/w) transition metal chlorides obtained with varied metal chlorides, water-only and H2SO4
ZnCl2 and CrCl3 generated byproducts such as lactic acid, treatment. In addition, the corresponding 5-HMF and byprod-
glyceraldehyde, and dihydroxyacetone. Moreover, levulinic acid ucts selectivity as well as the agarose conversion were also listed
was also observed with 0.5% (w/w) CrCl3. The increased catalyst and discussed.
loading presented insignicant impact on the types of
Results (Table 1) demonstrated that addition of alkali and
compounds generated from agarose, whereas had signicant alkaline earth metal chlorides (NaCl, CaCl2 and MgCl2) led to
inuence on their yields. Galactose yield was reduced sharply higher 5-HMF yields than transition metal chlorides (ZnCl2,
from 0.5–7.7% to negligible amounts as the loading of ZnCl2 CrCl3, CuCl2, FeCl3), water-only and H2SO4. Particularly, MgCl2
and CrCl3 was raised from 0.5% (w/w) to 5% (w/w). In contrast, resulted in the highest 5-HMF yield of 40.7%. The relatively
the yields of 5-HMF and other products catalyzed by ZnCl2 and higher 5-HMF yields with alkali and alkaline earth metal chlo-
CrCl3 increased when the catalyst loading was enhanced rides (NaCl, CaCl2 and MgCl2) compared with transition metal
from 0.5% (w/w) to 1% (w/w), and reached their highest yields. chlorides (ZnCl2, CrCl3, CuCl2, FeCl3), water-only and H2SO4
1% (w/w) loading of ZnCl2 and CrCl3 resulted in the highest could be attributed to fewer byproducts formed under tested
5-HMF yields of 19.0% and 12.9%, respectively. Further conditions. The corresponding 5-HMF selectivity with varied
increasing the catalyst loading from 1% (w/w) to 5% (w/w) led
the yields of 5-HMF and those byproducts to decline signi-
cantly. Other transition metal chlorides CuCl2 and FeCl3 as well
as H2SO4 merely resulted in less than 3.5% yield of galactose
and a considerable amount of levulinic acid (19.5–41.3% yield)
over the three tested catalyst loadings (i.e. 0.5%, 1% and
5% w/w), while 5-HMF was negligible. It was found that the
loading of metal chlorides and H2SO4 had a minor impact on
the variety of degradation compounds from agarose. However, the
catalyst loading had evident effects on the distribution of these
degradation compounds, thereby leading to different 5-HMF yield.
3.1.3. Effects of reaction time. The time courses of 5-HMF
yields from agarose by 1% (w/w) metal chlorides, water-only and
1% (w/w) H2SO4 at 200 ꢀC for 0–50 min are shown in Fig. 3.
Results showed that catalytic reaction with transition metal
chlorides CuCl2 and FeCl3 led to similar trends as 1% (w/w)
H2SO4, and the highest yield of around 5% 5-HMF was obtained
Fig. 3 Time course of 5-HMF yield decomposed from agarose by water-only, 1%
at 5 min, and then rapidly decreased. 5-HMF yields with tran-
(w/w) dilute sulfate acid and 1% (w/w) various metal chlorides treatments at
sition metal chlorides ZnCl2 and CrCl3 were relatively higher, 200 ꢀC.
24094 | RSC Adv., 2013, 3, 24090–24098
This journal is ª The Royal Society of Chemistry 2013

View Article Online
Paper
RSC Advances
Table 1 The highest 5-HMF yield and product selectivity from agarose degradation by various metal chlorides, H2SO4 and water-only treatmentb
Selectivity
5-HMF
yield (%)
Agarose
conv. (%)
Gly +
Dhy
Catalyst
Conditionsa
5-HMF
Ga
LA
Lac
MgCl2
CaCl2
NaCl
None
ZnCl2
CrCl3
CuCl2
FeCl3
H2SO4
200 ꢀC/1% (w/w)/35 min
200 ꢀC/1% (w/w)/40 min
200 ꢀC/1% (w/w)/40 min
220 ꢀC/1% (w/w)/20 min
200 ꢀC/1% (w/w)/25 min
200 ꢀC/1% (w/w)/30 min
200 ꢀC/0.5% (w/w)/10 min
200 ꢀC/0.5% (w/w)/10 min
200 ꢀC/0.5% (w/w)/5 min
40.7
37.2
32.8
29.5
22.6
13.0
8.2
82.9
85.3
81.1
79.3
92.1
100
91.6
93.8
90.3
49.1
43.6
40.5
37.2
24.5
13.0
9.0
20.5
21.1
25.2
20.5
4.2
0
0
0
0
0
0
0
0
0
0
6.2
15.8
7.5
0
0
0
12.1
9.4
0
0
0
0.5
13.7
20.4
26.9
22.7
31.9
29.1
28.2
7.7
7.1
8.2
7.9
0
0
a
Conditions to reach the highest 5-HMF yields. b Gal: galactose; LA: levulinic acid; Gly: glyceraldehyde; Dhy: dihydroxyacetone; Lac: lactic acid.
metal chlorides listed in Table 1 suggested that alkali and agarose, when using metal chlorides to hydrolyze cellulose in
alkaline earth metal chlorides (NaCl, CaCl2 and MgCl2) with aqueous phase under 180 ꢀC over 120 min, alkali and alkaline
fewer byproducts resulted in higher 5-HMF selectivity than earth metal (e.g. NaCl, CaCl2 and MgCl2) showed negligible
transition metal chlorides (ZnCl2, CrCl3, CuCl2, FeCl3), water- catalytic ability on cellulose conversion while transition metal
only and H2SO4. 5-HMF selectivity obtained using MgCl2 chlorides (e.g. CrCl3, FeCl3 and CuCl2) converted cellulose
resulted in the highest value of 49.1%. Such selectivity was efficiently.37
comparable to the 5-HMF selectivity (40–53%) obtained from
glucose in acidic biphasic systems, as reported by Dumesic and
coworkers.47 In comparison, the considerable selectivity for
3.2. Kinetic analysis of 5-HMF formation from agarose with
MgCl2
byproducts (e.g. lactic acid) formed via retro–aldol reaction
could serve as a possible explanation of the relatively low 5-HMF
selectivity from water-only or ZnCl2 treatment. For example,
lactic acid selectivity for ZnCl2 catalyzed reactions was 15.8%,
while the corresponding 5-HMF selectivity was merely 24.5%.
Using CuCl2, FeCl3 and H2SO4 led to 20.4–26.9% selectivity
toward levulinic acid through further rehydration of 5-HMF,
while their selectivity toward 5-HMF didn't exceed 9.0%. CrCl3
resulted in all byproducts listed in Table 1 with the selectivity
ranging from 7.5–13.7% for each compound, consequently
leading to 13.0% 5-HMF selectivity.
However, the galactose selectivity for alkali and alkaline earth
metal chlorides (NaCl, CaCl2 and MgCl2) catalyzed reaction was
maintained on a relatively high level (20.5–25.2%) compared with
some transition metal chlorides (e.g. CrCl3) (0.5%), thus reducing
the 5-HMF selectivity. Furthermore, the limited 5-HMF selectivity
could be also a consequence of the formation of aforementioned
(Section 3.1.1) insoluble humins.46
MgCl2 appeared to result in the highest 5-HMF yield and
selectivity among various tested metal chlorides under experi-
mental conditions. A kinetic analysis of 5-HMF production from
agarose was perfoꢀrmed with 1% (w/w) MgCl2 at a temperature
range of 180–220 C within 50 min.
A pseudo-homogeneous irreversible rst order reaction
model48 was used to describe the degradation of agarose to
5-HMF with low substrate concentration (2% (w/w)). This
model was also used to describe cellulose (0.8–2.4% (w/w))
degradation to 5-HMF and levulinic acid in aqueous phase
with dilute acid.22
Results indicated that when agarose was decomposed with
MgCl2, 5-HMF and galactose were major products without
formation of other byproducts, such as levulinic acid, lactic
acid, etc. Therefore, the reaction pathway of agarose decom-
position in MgCl2 solution can be simply summarized as
follows:
k1
ƒƒƒƒ!
k2
ƒƒƒƒ!
k3
ƒƒƒƒ!
Agarose
Galactose
5-HMF
Degradation products ðe:g: HuminsÞ
The agarose conversion with alkali and alkaline earth
Based on the reaction pathway, the kinetic model equations
metal chlorides (NaCl, CaCl2 and MgCl2), transition metal were developed to describe the agarose degradation process as
chlorides (ZnCl2, CrCl3, CuCl2 and FeCl3) were obtained 81.1– follows:
100% (Table 1) under conditions for achieving the highest
dA/dt ¼ ꢂk1A
dG/dt ¼ k1A ꢂ k2G
dM/dt ¼ k2G ꢂ k3M
(14)
(15)
(16)
5-HMF yield, which demonstrated the agarose's susceptibility
to decomposition with both alkali and alkaline earth metal
chlorides (NaCl, CaCl2 and MgCl2) and transition metal
chlorides (ZnCl2, CrCl3, CuCl2 and FeCl3). Different from
This journal is ª The Royal Society of Chemistry 2013
RSC Adv., 2013, 3, 24090–24098 | 24095

View Article Online
RSC Advances
Paper
where A (glucan equivalents, g), G (glucan equivalents, g) and M
(glucan equivalents, g) represent the weight of agarose, galactose
and 5-HMF, respectively. k1 (minꢂ1), k2 (minꢂ1) and k3 (minꢂ1
)
are rate constants of agarose hydrolyzed into galactose, galactose
dehydrated into 5-HMF, and 5-HMF condensed to humins,
respectively. Table 2 shows the value of the rate constants
increased with reaction temperature. The rate constant k3 was
lower than both k1 and k2 over the tested temperatures (180–
220 ꢀC), which was especially apparent at 200 ꢀC (i.e. k3 ¼ 0.0399
minꢂ1, k1 ¼ 0.0937 minꢂ1, k2 ¼ 0.0729 minꢂ1).
Fig. 4 Product profiles in hydrolysates from treatments with 1% (w/w) MgCl2 at
Results implied that the formation of 5-HMF was much more
feasibly than its degradation under such conditions, thus
resulted in 5-HMF yield maintaining at a relatively high level
(e.g. 40.7%). The Arrhenius equation was also established:
200 ꢀC for 35 min analyzed by GC-MS.
the formation of 5-HMF is caused by the pathway of C2 and C5
ring closure.
k ¼ Aexp(ꢂEa/RT)
(17)
The reaction mechanism for MgCl2 catalyzing agarose to
5-HMF in aqueous phase is proposed in Fig. 5. It can be
anticipated that Mg2+ ions (MgCl2) would be able to coordi-
nate with the oxygen atom of C–O–C bond and subsequently
catalyze the cleavage of the C–O–C bond acting as Lewis acid.
Coordination of Mg2+ ions (MgCl2) to the oxygen atom of C–O–
C in sugars (e.g. sucrose) in aqueous phase has also been
where k (minꢂ1) is the rate constant, A (minꢂ1) is the pre-
exponential factor, Ea (kJ molꢂ1) is the activation energy, R is
universal gas constant (8.3143 ꢁ 10ꢂ3 kJ molꢂ1 Kꢂ1), T (K) is
temperature.
The activation energy values of the degradation of agarose,
galactose and 5-HMF calculated based on eqn (17) were 67.5 kJ
molꢂ1, 104.6 kJ molꢂ1 and 105.6 kJ molꢂ1, respectively. The
lower activation energy for agarose degradation compared with
its monomer and 5-HMF indicated the requirement of relatively
lower temperature for agarose degradation in 1% (w/w) MgCl2,
which consequently could be favorable for the stability of the
released 5-HMF.
13
supported with IR and C NMR spectra.49 Once agarose is
degraded into galactose, magnesium ions from MgCl2 coor-
dinate with oxygen atom of the hemiacetal portion of galac-
tose and the closest hydroxyl group to form an enediol
intermediate. This leads to the isomerization of galactose to
its ketose form, which can be easily dehydrated to 5-HMF.
Although other alkali and alkali earth metal ions, including
Na+ ions and Ca2+ ions, appeared to interact with oxygen atom
in sugar (e.g. oxygen atom of C–O–C bond),49 thereby
enhancing the agarose to 5-HMF reactions as compared to
water-only operations, Mg2+ ions may possess better activity
due to its higher charge density (charge-to-size ratio) than Na+
ions and Ca2+ ions.49
3.3. Postulated mechanism of agarose to 5-HMF catalyzed by
MgCl2
In order to gain more insights into the mechanism of 5-HMF
formation from agarose with MgCl2 treatment, the formation of
ꢀ
degradation products derived from MgCl2 treatments (200 C,
1% (w/w), 35 min) was further veried by GC-MS analysis.
The GC-MS spectra showed that 5-HMF was the dominant
compound from agarose catalyzed by MgCl2, and no levulinic
acid from further rehydration of 5-HMF or compounds from
retro aldol of galactose were detected (Fig. 4). Trace amount of
furyl hydroxymethyl ketone (C6H6O3), an isomer of 5-HMF was
observed. Formation of this isomer can be explained as a C3
carbonyl and C6 hydroxy group intermolecular closure, while
Table 2 Rate constant of each degradation step of agarose with 1% (w/w)
a
MgCl2
Temperature (ꢀC)
Parameters
180
200
220
k1 (minꢂ1
k2 (minꢂ1
k3 (minꢂ1
)
)
)
0.0426
0.0327
0.0145
0.977
0.947
0.976
0.0937
0.0729
0.0399
0.959
0.972
0.960
0.1822
0.3133
0.1416
0.961
0.969
0.944
2
RAga
2
RGa
2
R5-HMF
Fig. 5 Proposed reaction mechanism for Mg2+ catalyzing the decomposition of
a
Aga: agarose; Ga: galactose.
agarose to 5-HMF.
24096 | RSC Adv., 2013, 3, 24090–24098
This journal is ª The Royal Society of Chemistry 2013

View Article Online
Paper
Results implied that Mg2+ ions under the tested conditions
RSC Advances
5 B. Kim, J. Jeong, S. Shin, D. Lee, S. Kim, H. J. Yoon and
J. K. Cho, ChemSusChem, 2010, 3, 1273–1275.
6 J. Ferrell and V. Sarisky-Reed, National Algal Biofuels
Technology Roadmap, U.S. Department of Energy, Maryland,
2010.
7 K. T. Nijenhuis, Adv. Polym. Sci., 1997, 130, 194–202.
8 R. Armisen, Hydrobiologia, 1991, 221, 157–166.
9 X. T. Fu and S. M. Kim, Mar. Drugs, 2010, 8, 200–218.
10 Y. Roman-Leshkov and J. A. Dumesic, Top. Catal., 2009, 52,
possess suitable properties for 5-HMF production: Mg2+ ions
can promote the cleavage of C–O–C bond of agarose and
enhance the subsequent isomerization of galactose to its
ketose. Additionally, our results also suggested that Mg2+ ions
have little ability to induce the retro aldol reaction of galactose
or further rehydration of 5-HMF to levulinic acid, etc., thereby
maintaining a relatively higher 5-HMF yield with few byprod-
ucts (as depicted in Fig. 4).
´
´
297–303.
11 H. Zhao, J. E. Holladay, H. Brown and Z. C. Zhang, Science,
2007, 316, 1597–1599.
4. Conclusion
Metal chlorides had different effects on catalyzing agarose 12 R. M. Musau and R. M. Munavu, Biomass, 1987, 13, 67–74.
degradation in aqueous phase. The addition of alkali and 13 A. A. Shamsuri and D. K. Abdullah, Makara Seri Sains., 2010,
alkaline earth metal chlorides was found to result in the
formation of only galactose and 5-HMF while the addition of 14 J.-P. Mikkola, P. Virtanen, H. Karhu, T. Salmi and
transition metal chlorides led to the formation of various D. Y. Murzin, Green Chem., 2006, 8, 197–205.
degradation products, including levulinic acid, lactic acid, etc. 15 F. S. Asghari and H. Yoshida, Ind. Eng. Chem. Res., 2006, 45,
Although temperature and catalyst loading had signicant 2163–2173.
14, 101–106.
effects on reaction rates and the products distribution, the 16 E. U. Franck, Pure Appl. Chem., 1987, 59, 25–34.
variety of agarose degradation products catalyzed by these metal 17 D. J. Miller and S. B. Hawthorne, Anal. Chem., 1998, 70, 1618–
chlorides was dependent on the intrinsic properties of the
catalysts used. Alkali and alkaline earth metal chlorides NaCl, 18 T. M. Aida, Y. Sato, M. Watanabe, K. Tajima, T. Nonaka,
CaCl2 and MgCl2 resulted in higher 5-HMF yields and selectivity H. Hattori and K. Arai, J. Supercrit. Fluids, 2007, 40, 381–388.
than transition metal chlorides ZnCl2, CuCl2, FeCl3 and CrCl3, 19 G. Bonn and O. Bobleter, J. Radioanal. Chem., 1983, 79, 171–
water-only or dilute sulfuric acid treatments. The addition of
177.
1% (w/w) MgCl2 was the most favorable additive among the 20 M. J. Antal, W. S. L. Mok and G. N. Richards, Carbohydr. Res.,
tested metal chlorides, resulting in 40.7% of the 5-HMF yield 1990, 199, 91–109.
1621.
and 49.1% 5-HMF selectivity at 200 ꢀC within 35 min. The 21 B. F. M. Kuster, Starch/Staerke, 1990, 42, 314–321.
mechanism of the MgCl2 catalyzed reaction of agarose to 5-HMF 22 J. Shen and C. E. Wyman, AIChE J., 2012, 58, 236–246.
was proposed to proceed through assisting the cleavage of C–O– 23 C. Moreau, R. Durand, F. Alies, M. Cotillon, T. Frutz and
C bond of agarose and accelerating the subsequent isomeriza-
tion of galactose to its ketose. The specic ability of MgCl2 in 24 Y. Roman-Leshkov, J. N. Chheda and J. A. Dumesic, Science,
aqueous phase is thought to be the driving force for favoring the
2006, 312, 1933–1937.
5-HMF formation, while diminishing the generation of unde- 25 L. Rigal, A. Gaset and J.-P. Gorrichon, Ind. Eng. Chem. Prod.
M. Theoleyre, Ind. Crops Prod., 2000, 11, 237–242.
sired byproducts such as levulinic acid, lactic acid, etc.
Res. Dev., 1981, 20, 719–721.
26 F. Benvenuti, C. Carlini, P. Patrono, A. M. R. Galletti,
G. Sbrana, M. A. Massucci and P. Galli, Appl. Catal., A,
2000, 193, 147–153.
Acknowledgements
We are grateful for the support from the Department of Bio- 27 C. Carlini, M. Giuttari, A. M. R. Galletti, G. Sbrana,
logical Systems Engineering and the Bioproducts, Sciences and
Engineering Laboratory at Washington State University. The
T. Armaroli and G. Busca, Appl. Catal., A, 1999, 183, 295–
302.
material in this work was supported by DARPA YFA Award # 28 Y. B. Huang and Y. Fu, Green Chem., 2013, 15, 1095–1111.
N66001-11-1-4141/P00001. The authors also thank Dr. Huamin 29 F. Chambon, F. Rataboul, C. Pinel, A. Cabiac, E. Guillon and
Wang from PNNL for his internal review of this paper and
helpful suggestions.
N. Essayem, Appl. Catal., B, 2011, 105, 171–181.
30 C. Carlinia, G. Mario, M. R. G. Anna, S. Glauco, A. Tiziana
and B. Guido, Appl. Catal., A, 1999, 183, 295–302.
31 T. Deng, X. Cui, Y. Qi, T. Wang, X. Hou and Y. Zhu, Chem.
Commun., 2012, 48, 5494–5496.
References
1 J. N. Chheda, G. W. Huber and J. A. Dumesic, Angew. Chem., 32 Y. Yang, C. Hu and M. M. Abu-Omar, Green Chem., 2012, 14,
Int. Ed., 2007, 46, 7164–7183.
509–513.
2 A. A. Rosatella, S. P. Simeonov, R. F. M. Frade and 33 S. Dutta, S. De, M. I. Alam, M. M. Abu-Omar and B. Saha,
C. A. M. Afonso, Green Chem., 2011, 13, 754–793. J. Catal., 2012, 288, 8–15.
3 J. B. Binder, A. V. Cefali, J. J. Blank and R. T. Raines, Energy 34 R. Kotraro, C. Linder, M. Nemas and M. Perry, US Pat.,
Environ. Sci., 2010, 3, 765–771. US4767645, 1988.
4 X. Qi, M. Watanabe, T. M. Aida and R. L. Smith, Cellulose, 35 X. Qi, M. Watanabe, T. M. Aida and R. L. Smith, Jr,
2011, 18, 1327–1333.
ChemSusChem, 2010, 3, 1071–1077.
This journal is ª The Royal Society of Chemistry 2013
RSC Adv., 2013, 3, 24090–24098 | 24097

View Article Online
RSC Advances
Paper
36 C. Li, Z. Zhang and Z. K. Zhao, Tetrahedron Lett., 2009, 50, 43 S. Sabesan and C. J. Spado, US Pat., US20130109869, 2013.
5403–5405.
44 A. Sluiter, B. Hames, R. Ruiz, C. Scarlata, J. Sluiter,
D. Templeton and D. Crocker, Technical Report. NREL/TP-
510-42618, 2008.
37 L. Peng, L. Lin, J. Zhang, J. Zhuang, B. Zhang and Y. Gong,
Molecules, 2010, 15, 5258–5272.
38 C. B. Rasrendra, I. G. B. N. Makertihartha, S. Adisasmito and 45 S. O. Bahaffi and S. S. Al-lihaibi, Commum.Fac.Sci. Univ.Ank.
H. J. Heeres, Top. Catal., 2010, 53, 1241–1247. Series B, 2005, 51, 1–12.
39 L. Kong, G. Li, H. Wang, W. He and F. Ling, J. Chem. Technol. 46 S. K. R. Patil and C. R. F. Lund, Energy Fuels, 2011, 25, 4745–
Biotechnol., 2008, 83, 383–388. 4755.
40 K. I. Seri, T. Sakski, M. Shibata, Y. Inoue and H. Ishida, 47 J. N. Chheda, Y. Romin-Leshkov and J. A. Dumesic, Green
Bioresour. Technol., 2002, 81, 257–260. Chem., 2007, 9, 342–350.
41 S. De, S. Dutta and B. Saha, Green Chem., 2011, 13, 2859–2868. 48 J. F. Saeman, Ind. Eng. Chem. Res., 1945, 37, 43–52.
42 C. Liu and C. E. Wyman, Carbohydr. Res., 2006, 341, 2550– 49 P. Rondeau, S. Sers, D. Jhurry and F. Cadet, Appl. Spectrosc.,
2556.
2003, 57, 466–472.
24098 | RSC Adv., 2013, 3, 24090–24098
This journal is ª The Royal Society of Chemistry 2013