Catalytic conversion of cellulose to levulinic acid by metal chlorides

Article abstract of DOI:10.3390/molecules15085258

The catalytic performance of various metal chlorides in the conversion of cellulose to levulinic acid in liquid water at high temperatures was investigated. The effects of reaction parameters on the yield of levulinic acid were also explored. The results showed that alkali and alkaline earth metal chlorides were not effective in conversion of cellulose, while transition metal chlorides, especially CrCl₃, FeCl₃ and CuCl₂ and a group IIIA metal chloride (AlCl₃), exhibited high catalytic activity. The catalytic performance was correlated with the acidity of the reaction system due to the addition of the metal chlorides, but more dependent on the type of metal chloride. Among those metal chlorides, chromium chloride was found to be exceptionally effective for the conversion of cellulose to levulinic acid, affording an optimum yield of 67 mol % after a reaction time of 180 min, at 200 °C, with a catalyst dosage of 0.02 M and substrate concentration of 50 wt %. Chromium metal, most of which was present in its oxide form in the solid sample and only a small part in solution as Cr³⁺ ion, can be easily separated from the resulting product mixture and recycled. Finally, a plausible reaction scheme for the chromium chloride catalyzed conversion of cellulose in water was proposed.

Full text of DOI:10.3390/molecules15085258

Molecules 2010, 15, 5258-5272; doi:10.3390/molecules15085258
OPEN ACCESS
molecules
ISSN 1420-3049
www.mdpi.com/journal/molecules
Article
Catalytic Conversion of Cellulose to Levulinic Acid by
Metal Chlorides
Lincai Peng, Lu Lin *, Junhua Zhang, Junping Zhuang *, Beixiao Zhang and Yan Gong
State Key Laboratory of Pulp and Paper Engineering, South China University of Technology,
Guangzhou 510640, Guangdong, China
* Authors to whom correspondence should be addressed; E-Mails: lclulin@scut.edu.cn (L.L.);
zhuangjp@scut.edu.cn (J.Z.).
Received: 18 June 2010; in revised form: 14 July 2010 / Accepted: 29 July 2010 /
Published: 2 August 2010
Abstract: The catalytic performance of various metal chlorides in the conversion of
cellulose to levulinic acid in liquid water at high temperatures was investigated. The effects
of reaction parameters on the yield of levulinic acid were also explored. The results
showed that alkali and alkaline earth metal chlorides were not effective in conversion of
cellulose, while transition metal chlorides, especially CrCl3, FeCl3 and CuCl2 and a group
IIIA metal chloride (AlCl3), exhibited high catalytic activity. The catalytic performance
was correlated with the acidity of the reaction system due to the addition of the metal
chlorides, but more dependent on the type of metal chloride. Among those metal chlorides,
chromium chloride was found to be exceptionally effective for the conversion of cellulose
to levulinic acid, affording an optimum yield of 67 mol % after a reaction time of 180 min,
at 200 °C, with a catalyst dosage of 0.02 M and substrate concentration of 50 wt %.
Chromium metal, most of which was present in its oxide form in the solid sample and only
a small part in solution as Cr3+ ion, can be easily separated from the resulting product
mixture and recycled. Finally, a plausible reaction scheme for the chromium chloride
catalyzed conversion of cellulose in water was proposed.
Key words: metal chlorides; cellulose; catalytic conversion; levulinic acid

Molecules 2010, 15
5259
1. Introduction
Cellulose is abundantly available, and is considered as a promising alternative to non-renewable
natural resources for the sustainable supply of fuel and chemicals in the future. Currently, extensive
research is being carried out worldwide to identify and study chemical or biological transformation
pathways to convert cellulose into biofuels and feedstock chemicals [1]. Among these, one attractive
approach is the single-step conversion of cellulose to levulinic acid (LA, Figure 1) by acid-catalyzed
hydrolysis. LA is a versatile building block containing a ketone carbonyl group and an acidic carboxyl
group, which can be used for preparation of various high-value organic chemicals, polymers, resin,
flavor substances, and fuel additives with numerous potential industrial applications [2,3].
Figure 1. Structure of levulinic acid (LA).
To date, researchers have reported the conversion of biomass into LA using a dilute mineral acid
such as HCl and H2SO4 as catalyst [4–7]. Although these hydrolysis reactions were effective, the use
of the mineral acid causes serious pollution and promotes equipment corrosion and being difficult to
recover from the reaction products for recycling [8]. Recently, solid acid catalysts have attracted
considerable interest as heterogeneous catalysts which can overcome the above mentioned
disadvantages of mineral acids in acid catalysis [9], but solid catalysts were found to display low
catalytic activity in the conversion of indissoluble cellulose to LA in water due to the poor accessibility
of the substrate. For instance, Wang et al. [10] used sulfated TiO2 as a solid acid catalyst to hydrolyze
cellulose into LA, however, the highest yield achieved under their optimal experimental conditions
was only 27.2 mol %.
Metal salts are expected to show higher catalytic activity than other types of acid catalysts, with the
additional possibility of being easy to be separate from the reaction products by supporting them on a
carrier [11]. Several studies have reported that some metal salts can catalytically hydrolyze
carbohydrates effectively into useful feedstock chemicals. Zhao et al. [12] studied the effect of a large
number of metal chlorides on the conversion of sugars into 5-hydroxymethylfurfural (HMF) in ionic
liquid solvents. Chromium (II) chloride was found to be uniquely effective, leading to the conversion
of glucose to HMF in nearly 70% yield. Li et al. [13] presented an efficient strategy for CrCl3-
mediated direct conversion of cellulose and glucose to HMF in ionic liquids under microwave
irradiation. Besides the above reactions in ionic liquids that mainly yield HMF, reactions of
carbohydrates with metal salts in aqueous solution and sub-critical water that chiefly produce LA
and/or lactic acid have also been explored. For example, Efremov et al. studied the conversion of
cellulose into LA in water using CoSO4, Fe2(SO4)3 and Al2(SO4)3 as catalysts [4]. The most active
among the studied catalysts was Al2(SO4)3, which produced a LA yield of about 18 wt % at 250 °C.
Rasrendra et al. reported the catalytic effect of a wide range of chloride and sulfate salts that were
shown to affect the chemo-selectivity considerably in the conversion of D-glucose in aqueous solutions

Molecules 2010, 15
5260
at 140 °C [14]. The main water-soluble product was lactic acid for Al (III), while HMF was formed in
the highest yields by Zn (II). Lu et al. [15] determined glucose and HMF decomposition kinetics
catalyzed with some metal chlorides in liquid water at high temperatures by using glucose as
representative model for cellulose. The dominating product in the conversion of carbohydrates
catalyzed by metal salts in sub-critical water (T = 200–360 °C) appears to be lactic acid, according to
Bicker et al. [16] and Kong et al. [17]. However, there are only a few reports on the effects of some
types of metal chlorides catalyzing conversion of cellulose to LA in terms of the composition and yield
of products. Seri et al. [11] used lanthanum (III) chloride as catalyst during the conversion of cellulose
in water at 250 °C and identified the products, but the yield of LA was reported to be low. Therefore,
further studies are still necessary to understand the catalytic mechanisms of metal chlorides and find
more reactive catalysts for the conversion of cellulose to LA.
The objective of this study was to examine the catalytic performances of twelve common metal
chlorides, including alkali metals (Li, Na and K), alkaline earth metals (Mg and Ca), transition metals
(Cr, Mn, Fe, Co, Cu and Zn) and a group IIIA metal (Al), on the conversion of cellulose to LA in hot-
compressed water. Among those metal chlorides, CrCl3 was found to be uniquely effective. Based on
its use, we systematically sought to optimize the LA yield by altering the process conditions (agitation
speed, reaction time, reaction temperature, catalyst dosage and substrate concentration) and propose
the possible catalytic scheme of the metal chlorides such as CrCl3 for the conversion of cellulose to LA.
2. Results and Discussion
2.1. Catalytic Effects of Metal Chlorides on the Conversion of Cellulose
The cellulose was reacted at 180 °C for 120 min with 0.01 M of metal chlorides in water. After the
reaction, the target products (i.e., LA and formic acid) as well as the main intermediates (i.e., glucose,
fructose and HMF) in the liquid samples were analyzed. The glucose, HMF, LA and FA were detected
for all samples using various types of metal chlorides as catalyst. Fructose was in trace amounts or
could not be detected. This is due to the faster dehydration of fructose to HMF than that of glucose,
which already been shown in previous studies [18,19]. Other than the products detected, there were
probably several other compounds in the liquid samples. For instance, Rasrendra et al. showed that
glycolic acid, acetic acid and lactic acid were also observed for the conversion of glucose in water at
140 °C using transition metal and aluminum salts [14]. These substances were not checked here,
because the objective of this paper was to produce LA. The yields of main products including glucose,
HMF and LA with and without the addition of metal chlorides as catalyst during the conversion of
cellulose are compared in Figure 2. The results showed that alkali and alkaline earth metal chlorides
did not lead effectively to the conversion of cellulose, and there was no significant difference with the
blank (no catalyst). By contrast, several transition metal chlorides (CrCl3, FeCl3 and CuCl2) and the
group IIIA metal chloride (AlCl3) showed significantly higher activity than the blank reaction and the
yields of LA for these metal chlorides decreased in the order CrCl3 > AlCl3 > FeCl3 > CuCl2 under
identical operating conditions. However, the experiment with FeCl3 and CuCl2 as catalyst had high the
residual amounts of glucose in the products, which indicated that the FeCl3 and CuCl2 are unfavorable
for further converting glucose into LA. In order to further prove the catalytic effect of these metal

Molecules 2010, 15
5261
chlorides including CrCl3, AlCl3, FeCl3 and CuCl2, we also conducted experiments with glucose
instead of cellulose under the same reaction conditions (Figure 3).
Figure 2. Catalytic effects of metal chlorides on the conversion of cellulose.
60
transition
m etal
chlorides
Ⅲ A
m etal
chlorides
H M F
G lucose
LA
50
40
30
20
10
0
alkaline
earth
m etal
alkali
m etal
chlorides chlorides
Reaction conditions: 2 wt % cellulose, 0.01 M metal chlorides, temperature 180 °C, time 120 min.
Figure 3. Catalytic effects of metal chlorides on the conversion of glucose.
100
80
60
40
20
0
100
80
60
40
20
0
Glucose
LA
HMF
Reaction conditions: 2 wt % glucose, 0.01 M metal chlorides, temperature 180 °C, time 120 min
It showed that these metal chlorides have similar activity for the conversion of glucose compared
with that of cellulose. FeCl3 and CuCl2 as catalyst during the conversion of glucose still left some
amount of residual glucose in the resulting products. On the other hand, glucose was totally consumed
and the yields of LA were very high when CrCl3 and AlCl3 were used as catalyst. A similar effect of
the type of salts on the glucose conversion was also shown in Rasrendra’s work [14], but the glucose
conversion there was somewhat low. Furthermore, either lactic acid or HMF was the major product,
and the amount of LA was relatively low. This is because the reaction temperature used in their
research was low at 140 °C. Lower temperature is disadvantageous to product LA [7], which was also
confirmed in the following experiments. However, different from the case of cellulose, AlCl3 showed
higher yield of LA and better selectivity for the conversion of glucose than CrCl3. Based on these
results it was concluded that AlCl3 is more favorable for promoting the isomerization of glucose to

Molecules 2010, 15
5262
fructose, whereas CrCl3 had higher activity than AlCl3 for the hydrolysis of cellulose. Cellulose was
rapidly depolymerized in the presence of CrCl3 so that it was more easily converted to glucose, and
then to LA.
2.2. Relationship between Reactivity and Acidity of Reaction System
Many metal chlorides are strong Lewis acids, and in water they may undergo hydrolysis to form
basic salts, resulting in the pH value of the solution becoming quite acidic. Figure 4 shows the
relationship between LA yield and initial pH value of the reaction system measured at room
temperature with 0.01 M metal chlorides. It is particularly interesting to note that the solutions of metal
chlorides including CrCl3, AlCl3, FeCl3 and CuCl2 with high activity in the conversion of cellulose to
LA were quite acidic, showing lower initial pH values in the reaction systems. The other added metal
chlorides showed very slight variation in the initial pH value of reaction system in comparison with the
blank (without the addition of metal chlorides), and did not work for the conversion of cellulose. The
water solution for a blank experiment exhibited weak acidity, which is due to dissolved carbon dioxide.
For this reason, a control experiment was also carried out at an intermediate pH value adjusted by
sodium hydroxide. It can be seen from Figure 4, the yield of LA for the control reaction was almost the
same as that of the blank. These finding suggested that a certain amount of acidity in the initial
reaction system was essential for the production of LA. However, this did not mean that the lower
initial pH value was closely related to the higher the yield of LA. For example, the initial pH value of
reaction system with FeCl3 was lower than that with CrCl3 and AlCl3, but the yield of LA was lower.
Accordingly, we believed catalytic performance depends mainly on the type of metal chloride used,
rather than the acidity of the solution. To verify the conclusions, the same initial pH value of the
reaction system with various metal chlorides was achieved by changing the amount of metal chlorides.
As shown in Figure 5, it can be seen that different types of metal chlorides as catalyst produced
significantly different yields of LA at the same initial pH value. The results further demonstrated the
types of metal chlorides play a major role in the cellulose hydrolysis and glucose dehydration.
Figure 4. Relationship between LA yield and initial pH value of reaction system with
0.01 M metal chlorides measured at room temperature.
50
4
Control
Blank
LiCl
3
40
NaCl
KCl
2
MgCl2
30
20
10
0
1
CaCl2
MnCl2
CoCl2
ZnCl2
CuCl2
FeCl3
CrCl3
AlCl3
0
5.7 5.8 5.9 6.0
pH value
2
3
4
5
6
7
8
pH value
Reaction conditions: 2 wt % cellulose, 0.01 M metal chlorides, temperature 180 °C, time 120 min.

Molecules 2010, 15
5263
2.3. Effects of Reaction Conditions on Yields of Products
Among the metal chlorides tested as catalyst, CrCl3 was found to be uniquely effective in the
conversion of cellulose to LA. Hence, it was chosen as the most suitable catalyst for further
exploration. To obtain the highest possible LA yield, it is necessary to optimize the reaction conditions
by varying the speed of agitation, reaction time, reaction temperature, catalyst dosage and substrate
concentration, and the results are discussed in the following sections.
Figure 5. Effect of various types of metal chlorides on the hydrolysis of cellulose at the
same initial pH value.
60
50
40
30
20
10
0
46.0%
37.3%
32.7%
2.9%
Reaction conditions: 2 wt % cellulose, pH value 3.8, temperature 180 °C, time 120 min.
2.3.1. Agitation Speed
The solid-liquid phase catalytic reaction system between the cellulose and the water solution of
metal chlorides may suffer from severe mass transfer limitations that affect the apparent reaction rate.
Increasing the agitation speed might increase the contact area of the two phases, and thus remove the
interfacial mass transfer resistance [20]. The reaction was conducted by changing the agitation speed
from 0–600 rpm and the results are given in Figure 6.
Figure 6. Effect of agitation speed on yields of products.
34
31
28
25
22
5
4
3
2
1
0
LA
Glucose
HMF
0
100
200
300
400
500
600
Agitation speed, rpm
Reaction conditions: 2 wt % cellulose, 0.01 M chromium chloride, temperature 180 °C, time 120 min.

Molecules 2010, 15
5264
As it can be observed from the figure, the yield of LA increased rapidly as agitation speed increased
from 0 rpm to 200 rpm, and then remained almost constant. However, there was no clear influence of
the agitation speed on the yields of glucose and HMF. These results indicated that the interfacial mass
transfer resistance between the cellulose surface and liquid phase was negligible when the agitation
speed was above 200 rpm. The conclusion also proved that the frontal experimental results with
agitation speed of 300 rpm were an intrinsic kinetic feature for the good conversion of cellulose to LA.
2.3.2. Reaction Time
Time courses of the main water-soluble products during CrCl3-catalyzed hydrolysis of cellulose
were as shown in Figure 7. It seems that with the prolonging of reaction time, the amount of glucose
reached a low constant level within a very short period of time. The yield of HMF, being slightly
higher during the initial stage of the reaction and then decreased, was kept very low throughout the
reaction. The amount of LA grew very fast for the first 140 min. Then it was observed slightly to rise
after 140 min. It was found that 180 min was sufficient for completion of the hydrolysis reaction. The
time course of the amount of formic acid (FA) was similar to that of LA. In addition, LA and FA were
always formed in a 1:1 molar ratio in line with the reaction stoichiometry given in Scheme 1. The
result was also proved in previous research [6]. This finding indicates that LA and FA are stable under
the reaction conditions employed and do not decompose to other products.
Figure 7. Time courses of the products during CrCl3-catalyzed hydrolysis of cellulose.
50
40
30
20
10
0
5
4
3
2
1
0
LA
FA
Glucose
HMF
0
20 40 60 80 100 120 140 160 180 200
Time, min
Reaction conditions: 2 wt % cellulose, 0.01 M chromium chloride, temperature 180 °C.
2.3.3. Reaction Temperature
The effect of reaction temperature on the cellulose conversion into LA was investigated, and the
experiments were conducted at 180, 190, 200, 210 and 220 °C, respectively. Figure 8 shows the effect
of the temperature on the yields of products. It can be observed that temperature has a significant effect
on LA yield. When the reaction temperature was increased from 180 to 200 °C, there was a significant
increase in the yield of LA and the amount of glucose and HMF were gradually decreased. When the
reaction temperature rose above 200 °C, the yield of LA began to fall, and intermediates such as
glucose and HMF weren’t detected. Higher temperatures thus could accelerate the rate of chemical

Molecules 2010, 15
5265
reactions, but unwanted side reactions also appeared at the same time. For example, LA is unstable
above 230 °C, which would be dehydrated to unsaturated lactones [7]. Additionally, more humins may
occur during the reaction. Therefore, elevation of temperature is unfavorable for the extent of reaction,
thus the reaction temperature was set to 200 °C in this experiment.
Figure 8. Effect of reaction temperature on yields of products.
60
55
50
45
40
35
4
3
2
1
0
LA
Glucose
HMF
180
190
200
210
220
Reaction temperature, oC
Reaction conditions: 2 wt % cellulose, 0.01 M chromium chloride, time 180 min.
2.3.4. Catalyst Dosage
The effect of catalyst dosage on the yields of products is shown in Figure 9. The concentration of
CrCl3 into the reaction mixture was varied from 0.002 M to 0.03 M. As can be seen, catalyst dosage
significantly affected the cellulose conversion. The effects on glucose and HMF were similar to those
in Figure 8. The yield of LA increased rapidly with increasing catalyst dosage to an optimum
condition, and then did not produce a corresponding increase with a further increase Addition of
excess catalyst has little effect. Similar results have also been observed for LA production catalyzed by
other catalysts such H2SO4 and HCl [2,5]. Accordingly, when catalyst dosage meets the requirements
for the reaction, no more would be needed. Taking the cost and the efficiency into considerations, the
most beneficial concentration of CrCl3 was chosen to be 0.02 M.
Figure 9. Effect of catalyst dosage on yields of products.
70
60
50
40
30
10
LA
Glucose
HMF
8
6
4
2
0
0.000 0.005 0.010 0.015 0.020 0.025 0.030 0.035
Chromium chloride concentration, M
Reaction conditions: 2 wt % cellulose, temperature 200 °C, time 180 min.

Molecules 2010, 15
5266
2.3.5. Substrate Concentration
The optimal concentration of substrate is very important for the efficient use of cellulose and the
final LA concentration. Figure 10 shows the influence of the substrate concentration with respect to the
LA concentration and LA yield. Higher LA yield was achieved at lower substrate concentration, but
the LA concentration was low, as shown in Figure 10. A higher LA concentration is extremely
favorable that can not only cut down energy consumption of the purification of LA, but also reduce the
amount of wastewater produced. When catalyst dosage was fixed, increasing substrate concentration
resulted in more cellulose available for conversion, which means higher LA concentration in the
reaction samples, but the yield of LA was found to decrease markedly from 10 wt % to 15 wt %. The
reason is probably due to product feedback inhibition, reactivity diminution or insufficient catalyst
dosage for the additional cellulose. Therefore, compromises have to be made between concentration
and yield of LA for economic reasons.
Figure 10. Effect of substrate concentration on concentration and yield of LA.
70
65
60
55
50
100
80
60
40
20
0
LA yield
LA concentration
0
5
10
15
20
Substrate concentration, wt%
Reaction conditions: 0.02 M chromium chloride, temperature 200 °C, time 180 min.
2.4. Metal Tracing and Recycling
The dosage of CrCl3 (0.02 M) during the production of LA in this paper was less compared with
that of H2SO4 (0.1–1 M) [2,5,6] and HCl (~1.2 M) [7] in the previous literature, but the disposal of the
catalysts will not only waste resources, but also cause environmental pollution, so it is extremely
important that the chromium metal be recycled. Here, the existing forms and the separation of
chromium metal from the reaction product were investigated. The amount of chromium ion in the
liquid product was determined using atomic absorption spectrometry. The results show that it was
significantly lower, with only 30% of that measured before the reaction. It is easy to separate by
supporting it on a carrier. This finding also indicates that the existing form of most chromium has
changed from the Cr3+ ion found in solution. During the reaction, some dark-brown insoluble
substances were observed. Previous studies have already shown that the substances known as humins
are formed from the side-reactions of the acid-catalyzed decompositions of cellulose, glucose and
HMF [6,21]. The presence of the solid products after the reaction was confirmed by XPS analysis, as
shown in Figure 11. Inspection of the XPS survey spectrum revealed that C and O are the predominant

Molecules 2010, 15
5267
species, occuring at 283.19 eV and 531.02 eV, respectively, and they are mainly derived from humins.
In addition, chromium was also detected, but chlorine did not appear. This suggests that most of the
chromium metal may be adsorbed in the solid products and no longer existed as chloride. High-
resolution scanning of the XPS spectrum of Cr 2p with its characteristic peaks at 585.62 eV of Cr1 and
575.93 eV of Cr2, respectively, are also presented in Figure 11.
Figure 11. XPS survey spectrum of solid product and high-resolution spectrum of Cr 2p.
595
590
585
Cr1
580
575
570
Cr2
O1s
C1s
Cr2p
1000
800
600
400
200
0
Binding Energy, eV
The chromium peaks might be caused by some its oxides, Cr1 of Cr2O3 and Cr2 of CrO2, which are
known to be existing forms of chromium in the solid product. The Cr1 to Cr2 peak intensity ratio was
found to be of 0.65. They can be formed during the reaction process because of high temperature and
pressure. For separating the chromium oxides from the humins, the solid product was calcined at 400
°C for 8 h in static air. It was observed that the color of solid product changed from dark-brown to
green. The humins were thus removed from the solid product. The XRD pattern of the resulting
product was shown in Figure 12.
Figure 12. XRD pattern of solid product calcined at 400 °C for 8 h.
200-500 °C
4 CrO2
2 Cr2O3
O2
+
36.2
33.6
54.9
50.3
24.5
65.2
41.5
63.5
58.4
76.9
72.9
44.3
10
20
30
40
50
60
70
80
2θ , ( °)
The resulting product was found to display some well distinguished peaks in the 2θ range from 10°
to 80°. These peaks were all assigned to Cr2O3. This is because chromium dioxide also breaks down

Molecules 2010, 15
5268
into chromium trioxide and oxygen under high temperature condition. Cr2O3 is not only a raw material
for Cr production but also of possible use for the preparation of refractory ceramics and green
pigments.
2.5. Mechanism of Catalytic Degradation
The conversion of cellulose to LA is a complex multistep process. It is generally believed that the
polymer chains of cellulose are broken down into low molecular weight fragments and finally to
glucose. Then, the glucose is decomposed to HMF, which is further converted into LA and FA [6]. All
above-mentioned products were detected in this study. Some metal chlorides have shown obvious
catalytic effects for the conversion of cellulose and glucose in many studies, but it is still unclear how
metal chlorides affect the reaction process. In particular, the exceptional effectiveness of CrCl3 in the
conversion of cellulose not only reflected in our study for the LA production in water, but also in Li’s
work for the HMF production in ionic liquid [13]. It is likely that CrCl3 system have very strong
coordinating abilities with certain kinds of groups in the reactants. Li believed that CrCl3 in [C4mim]Cl
form [C4mim]n[CrCl3+n] complexes and coordination chemistry involving CrCl3 played a predominant
role for cellulose hydrolysis and glucose dehydration. Therefore, according to these findings and
results from our experiment, it could be induced that the catalytic reaction mechanism of CrCl3 in
water could be similar to that in ionic liquids. A plausible reaction mechanism for the CrCl3 catalyzed
conversion of cellulose to LA in water is thus proposed, as shown in Scheme 1.
Scheme 1. Proposed mechanism for the CrCl3 catalyzed conversion of cellulose to LA in water.
OH
OH
OH
CrCl ·xH O
3 2
OH
O
OH
O
O
O
O
HO
HO
HO
O
HO
O
+CrCl ·xH O
O
HO
O
HO
OH
3
2
O
OH
OH
OH
OH
OH
n
Cellulose
+
OH
OH
OH
+CrCl ·xH O
CrCl ·xH O
+H O
3
2
O
2
2
2
O
O
HO
HO
HO
HO
O
xH O·CrCl
2
3
-CrCl ·xH O
Cl
3
2
HO
OH
OH
Glucose
(α-glucopyranose anomer)
OH
OH
O
H
+
HO
+H O
2
-CrCl ·xH O
3
2
OH
O
CrCl ·xH O
H
3
2
HO
Cl
OH
OH
+CrCl ·xH O
HO
HO
-CrCl ·xH O
3
2
-
3
2
OH
OH
O
HO
HO
CrCl ·xH O
HO
2
2
OH
OH
OH OH
Glucose
(β-glucopyranose anomer)
O
-3H O
2
-CrCl ·xH O
3
2
-CrCl ·xH O
3
2
OH
O
HO
O
O
O
+2H O
-3H O
2
2
O
OH
HO
OH
H
H
OH
H
H C
3
H
OH
OH
Fructose
O
Formic acid
Levulinic acid
HMF

Molecules 2010, 15
5269
The CrCl3 could form hydrated complexes in aqueous solution. In this scheme, the cellulose
hydrolysis step and the coordination of the glycosidic oxygen of cellulose with chromium, which is
acting as the Lewis acid, would help the breakdown of the glycosidic linkages, with the coordinated
water molecules from the hydrated CrCl3 participating as a nucleophile to form glucose, similar to
what was proposed for the ZnCl2-mediated degradation of cellulose to glucose reported by
Amarasekara et al. [22]. In the glucose degradation process, the hydrated complex of CrCl3 facilitated
mutarotation of the α-anomer of glucose to the β-one through hydrogen bonds of chloride anions with
the hydroxyl groups. Then, the hemiacetal portion of β-glucopyranose forms a Cr (III) enolate anion
complex that leads to isomerization of glucose to fructose, which would be simultaneously dehydrated
to HMF. HMF is unstable in water and could be further converted into LA and FA under the reaction
condition.
3. Experimental
3.1. Materials
The twelve kinds of metal chlorides used in this study were lithium chloride (LiCl), sodium chloride
(NaCl), magnesium chloride (MgCl2), potassium chloride (KCl), calcium chloride (CaCl2), manganese
chloride (MnCl2), cobalt chloride (CoCl2), zinc chloride (ZnCl2), ferric chloride (FeCl3), copper
chloride (CuCl2), chromium chloride (CrCl3), and aluminium chloride (AlCl3) from Kermel (Tianjin,
China). Microcrystalline cellulose (DP~200) with an average particle size of 100 μm was obtained
from Sinopharm Chemical Reagent (Shanghai, China). The above reagents and chemicals were all of
analytical grade and used without further purification or treatment. Glucose, fructose,
5-hydroxymethylfurfural, levulinic acid, and formic acid used for calibration, with purity of over 99%,
were purchased from Aladdin Reagent (Shanghai, China). Deionized water was used for
all reactions.
3.2. Equipment and Procedure
The hydrolysis experiments were carried out in a cylindrical stainless steel pressurized reactor with
inner diameter 52 mm, depth 118 mm and 250 mL total volume. The reactor was heated in an
adjustable electric stove. The temperature of the reactor contents was monitored by a thermocouple
connected to the reactor. For each experiment, a weighed amount of microcrystalline cellulose or
glucose, water and the desired metal chlorides were mixed to form a suspension and put into the
reactor, which was then brought to the desired temperature by external heating. A typical reaction
procedure was as follows: cellulose (2 g) and water (100 mL) were introduced into the reactor along
with 0.001 mol chromium chloride. The reactor was heated at 180 °C for 120 min and shaken at
300 rpm. After the reaction, the reactor was taken from the stove and quenched in an ice cool water
bath to terminate the reaction.
3.3. Sample Analysis
The samples taken from the reactor were filtered and washed with a portion of deionized water to
recover the liquid and solid products. The amount of products in the liquid samples was quantitatively

Molecules 2010, 15
5270
analyzed by ion chromatography (Dionex ICS-3000, USA) after appropriate dilution with pure water.
A CarboPac PA1 (2 mm × 250 mm) analytical column with electrochemical detection was employed
to detect the glucose and HMF. A solution of sodium hydroxide (100 mM) was used as the eluent with
a volumetric flow rate of 0.325 mL/min at 30 °C. An IonPac AS11-HC (4 mm × 250 mm) analytical
column with conductivity detection was employed to detect the LA and FA. A solution of sodium
hydroxide (10 mM) was used as the eluent with a volumetric flow rate of 1.0 mL/min at 30 °C. The
sample injection volume was 20 μL. The amount of each compound in the liquid products was
determined using calibration curves obtained by analyzing standard solutions with known amounts.
The yield of products on a molar base was calculated using the following equation [23]:
Product yield (mol %) = 100M/N
where N is total amount (mol) of glucose monomer in cellulose and M denoted the amount (mol) of
products after reaction. The sample was further analyzed when CrCl3 was used as a catalyst. The
amount of chromium ion in the liquid product was determined using atomic absorption spectrometry
(Hitachi Z-2000, Japan). The elemental composition of the solid product was analyzed with X-ray
photoelectron spectroscopy (XPS). XPS measurement was performed on a Kratos Ultra system with
1 eV per step for survey spectrum over a binding energy range of 0–1,100 eV. High-resolution
spectrum of Cr 2p was an average of five scans acquired at pass energy of 20 eV and resolution of
0.1 eV per step. X-ray diffraction (XRD) analysis of the solid product calcined at 400 °C for 8 h was
carried out using Bruker D8 X-ray diffractometer with Cu Ka radiation operated at 40 kV and 40 mA.
Data was collected from 2θ between 10° and 80° in a step of 0.02°/s.
4. Conclusions
In this study it was shown that different types of metal chlorides have markedly different catalyst
effects on the conversion of cellulose. Some transition metal chlorides (such as CrCl3, FeCl3 and
CuCl2) and a group IIIA metal chloride (AlCl3) showed higher catalytic activity for the reaction. The
catalytic performance was correlated with the acidity of the reaction system due to addition of the
metal chlorides, but more dependent on the types of metal chlorides. The reaction time, reaction
temperature, catalyst dosage and substrate concentration during the degradation of cellulose play
significant roles on the production of LA. The highest yield of LA was found to be 67 mol % when
CrCl3 was used as catalyst. An advantage of using CrCl3 as catalyst is that less is used. Additionally,
chromium metal, most of which was present in its oxide form in the solid products and only a small
part in solution as Cr3+ ion, can be easily separated from the reaction products and recycled.
Acknowledgements
The authors are grateful to the financial support from Natural Science Foundation of China
(50776035, U0733001), Foundation of Scientific Research for Universities (20070561038) from
Ministry of Education of China, National High Technology Project (863 project) (2007AA05Z408)
and National Key R&D Research Program (2007BAD34B01) as well as National Key Basic Research
Program (2010CB732201) from the Ministry of Science and Technology of China.

Molecules 2010, 15
5271
References
1. Naik, S.N.; Goud, V.V.; Rout, P.K.; Dalai, A.K. Production of first and second generation
biofuels: A comprehensive review. Renew. Sustain. Energy Rev. 2010, 14, 578–597.
2. Girisuta, B.; Danon, B.; Manurung, R.; Janssen, L.P.; Heeres, H.J. Experimental and kinetic
modelling studies on the acid-catalysed hydrolysis of the water hyacinth plant to levulinic acid.
Bioresour. Technol. 2008, 99, 8367–8375.
3. Bozell, J.J.; Moens, L.; Elliott, D.C.; Wang, Y.; Neuenscwander, G.G.; Fitzpatrick, S.W.; Bilski,
R.J.; Jarnefeld, J.L. Production of levulinic acid and use as a platform chemical for derived
products. Resour. Conserv. Recy. 2000, 28, 227–239.
4. Efremov, A.A.; Pervyshina, G.G.; Kuznetsov, B.N. Production of levulinic acid from wood raw
material in the presence of sulfuric acid and its salts. Chem. Nat. Compd. 1998, 34, 182–185.
5. Chang, C.; Cen, P.L.; Ma, X.J. Levulinic acid production from wheat straw. Bioresour. Technol.
2007, 98, 1448–1453.
6. Girisuta, B.; Janssen, L.P.B.M.; Heeres, H.J. Kinetic study on the acid-catalyzed hydrolysis of
cellulose to levulinic acid. Ind. Eng. Chem. Res. 2007, 46, 1696–1708.
7. Yan, L.F.; Yang, N.K.; Pang, H.; Liao, B. Production of levulinic acid from bagasse and paddy
straw by liquefaction in the presence of hydrochloride acid. Clean-Soil, Air, Water 2008, 36,
158–163.
8. Lange, J.P.; van de Graaf, W.D.; Haan, R.J. Conversion of furfuryl alcohol into ethyl levulinate
using solid acid catalysts. Chem. Sus. Chem. 2009, 2, 437–441.
9. Fan, G.D.; Shen, M.; Zhang, Z.; Jia, F. Preparation, characterization and catalytic properties of
2-
S2O8 /ZrO2-CeO2 solid superacid catalyst. J. Rare Earth. 2009, 27, 437–442.
10. Wang, P.; Zhan, S.H.; YU, H.B. Production of levulinic acid from cellulose catalyzed by
environmental-friendly catalyst. Adv. Mater. Res. 2010, 96, 183–187.
11. Seri, K.I.; Sakski, T.; Shibata, M.; Inoue, Y., Ishida, H. Lanthanum (III)-catalyzed degradation of
cellulose at 250 °C. Bioresour. Technol. 2002, 81, 257–260.
12. Zhao, H.B.; Holladay, J.E.; Brown, H.; Zhang, Z.C. Metal chlorides in ionic liquid solvents
convert sugars to 5-hydroxymethylfurfural. Science 2007, 316, 1597–1600.
13. Li, C.Z.; Zhang, Z.H.; Zhao, Z.B.K. Direct conversion of glucose and cellulose to
5-hydroxymethylfurfural in ionic liquid under microwave irradiation. Tetrahedron Lett. 2009, 50,
5403–5405.
14. Rasrendra, C.B.; Makertihartha, I.G.B.N.; Adisasmito, S.; Heeres, H.J. Green chemicals from
D-glucose: Systematic studies on catalytic effects of inorganic salts on the chemo-selectivity and
yield in aqueous solutions. Topic Catalysis 2010, 53, 1241–1247.
15. Lu, C.B.; Lü, X.Y. Effect of metal chlorides on glucose decomposition kinetics in high
temperature liquid water. CIESC J. 2009, 60, 3035–3041.
16. Bicker, M.; Endres, S.; Ott, L.; Vogel, H. Catalytical conversion of carbohydrates in subcritical
water: A new chemical process for lactic acid production. J. Mol. Catal. A: Chem. 2005, 239,
151–157.
17. Kong, L.Z.; Li, G.M.; Wang, H.; He, W.Z.; Ling, F. Hydrothermal catalytic conversion of
biomass for lactic acid production. J. Chem. Technol. Biotechnol. 2008, 83, 383–388.

Molecules 2010, 15
5272
18. Kuster, B.F.M.; van der Baan, H.S. Dehydration of D-fructose (formation of 5-hydroxymethy
l-2-furaldehyde and levulinic acid).2.Influence of initial and catalyst concentrations on
dehydration of D-fructose. Carbohydr. Res. 1977, 54, 165–176.
19. Girisuta, B.; Janssen, L.P.B.M.; Heeres, H.J. Green chemicals: A kinetic study on the conversion
of glucose to Levulinic acid. Chem. Eng. Res. Des. 2006, 84, 339–349.
20. Devulapelli, V.G.; Weng, H.S. Synthesis of cinnamyl acetate by solid-liquid phase transfer
catalysis: Kinetic study with a batch reactor. Catal. Commun. 2009, 10, 1638–1642.
21. Mckibbins, S.; Harris, J.F.; Saeman, J.H.; Neill, W.K. Kinetics of the acid catalyzed conversion of
glucose to 5-hydroxymethyl-2-furaldehyde and levulinic acid. Forest Prod. J. 1962, 12, 17–23.
22. Amarasekara, A.S.; Ebede, C.C. Zinc chloride mediated degradation of cellulose at 200 °C and
identification of the products. Bioresour. Technol. 2009, 100, 5301–5304.
23. Suganuma, S.; Nakajima, K.; Kitano, M.; Yamaguchi, D.; Kato, H.; Hayashi, S.; Hara, M.
Hydrolysis of cellulose by amorphous carbon bearing SO3H, COOH, and OH groups. J. Am.
Chem. Soc. 2008, 130, 12787–12793.
Sample Availability: Samples of the compounds are available from the authors.
© 2010 by the authors; licensee MDPI, Basel, Switzerland. This article is an Open Access article
distributed under the terms and conditions of the Creative Commons Attribution license
(http://creativecommons.org/licenses/by/3.0/).