Heterogeneous catalytic reaction of microcrystalline cellulose in hydrothermal microwave-assisted decomposition: Effect of modified zeolite Beta

Article abstract of DOI:10.1039/c3gc42207k

Zeolite Beta, modified with some salts of alkali and alkaline earth metals (K, Zn, Sn), was tested in the hydrothermal heterogeneous catalytic decomposition of microcrystalline cellulose. The reactions were microwave-assisted, where the microwaves were issued by an in situ coaxial applicator. Zeolites were subjected to an ion-exchange process which determines the loss of crystallinity in the following order: Sn-Beta-IE > K-Beta > Zn-Beta > acid form H-Beta. The interaction between zeolites and microwaves was studied by irradiating zeolite powder under constant power and the heating response was in the following order: K-Beta > NH₄-Beta > Sn-Beta-IE ≈ Zn-Beta > H-Beta > alumina. These results show that the nature of the counterion strongly affects the absorption of microwaves. The catalytic activity of the different systems on the cellulose decomposition was studied, and resulted in the following order: H-Beta > K-Beta > Zn-Beta > Sn-Beta-IE > alumina, when the reaction medium contained 5 mM HCl. The most active catalyst was the acid zeolite Beta and the identified product distribution under the investigated conditions was (mol yield %): levulinic acid (22.3), glucose (12.1), lactic acid (4.1), formic acid (6.6), 5-(hydroxymethyl) furfural (14.6), acetic acid (15.2) and furfuraldehyde (3.1). The effect of temperature, time and the heterogeneous catalyst reuse (H-Beta) on the yields of different products was investigated. The use of MW radiation with a coaxial applicator instead of conventional heating gave clear advantages in the decrease of the reaction time (45 min) and in terms of yield enhancement (78.6% under the best conditions).

Full text of DOI:10.1039/c3gc42207k

Green Chemistry
View Article Online
View Journal
PAPER
Heterogeneous catalytic reaction of
microcrystalline cellulose in hydrothermal
microwave-assisted decomposition: effect of
modified zeolite Beta
Cite this: DOI: 10.1039/c3gc42207k
José González-Rivera,a Ignacio R. Galindo-Esquivel,a Massimo Onor,b
Emilia Bramanti,b Iginio Longoc and Carlo Ferrari*c
Zeolite Beta, modified with some salts of alkali and alkaline earth metals (K, Zn, Sn), was tested in the
hydrothermal heterogeneous catalytic decomposition of microcrystalline cellulose. The reactions were
microwave-assisted, where the microwaves were issued by an in situ coaxial applicator. Zeolites were
subjected to an ion-exchange process which determines the loss of crystallinity in the following order:
Sn-Beta-IE > K-Beta > Zn-Beta > acid form H-Beta. The interaction between zeolites and microwaves
was studied by irradiating zeolite powder under constant power and the heating response was in the
following order: K-Beta > NH4-Beta > Sn-Beta-IE ≈ Zn-Beta > H-Beta > alumina. These results show that
the nature of the counterion strongly affects the absorption of microwaves. The catalytic activity of the
different systems on the cellulose decomposition was studied, and resulted in the following order: H-Beta
> K-Beta > Zn-Beta > Sn-Beta-IE > alumina, when the reaction medium contained 5 mM HCl. The most
active catalyst was the acid zeolite Beta and the identified product distribution under the investigated con-
ditions was (mol yield %): levulinic acid (22.3), glucose (12.1), lactic acid (4.1), formic acid (6.6), 5-(hydroxy-
methyl) furfural (14.6), acetic acid (15.2) and furfuraldehyde (3.1). The effect of temperature, time and the
heterogeneous catalyst reuse (H-Beta) on the yields of different products was investigated. The use of
MW radiation with a coaxial applicator instead of conventional heating gave clear advantages in the
decrease of the reaction time (45 min) and in terms of yield enhancement (78.6% under the best conditions).
Received 25th October 2013,
Accepted 26th November 2013
DOI: 10.1039/c3gc42207k
www.rsc.org/greenchem
established for this purpose using enzymes,2,3 strong acids,4
and supercritical and subcritical water.5
Introduction
The green chemical process design considers the biomass as
Microwave (MW) technology is a powerful tool to optimize
the base for producing several value-added compounds for these processes because it significantly reduces the processing
different applications, such as energy production and polymer time, energy costs and equipment size with respect to conven-
processing.1 Cellulose is the most abundant fraction of tional methods.6 Therefore, in the last few years, special atten-
biomass, a partially crystalline, linear, unbranched, glucose- tion has been focused on biomass processing using MW
based homopolymer, with relatively high molecular weight, radiation.7–10 In addition, because of the synergy produced
made up of β-1,4-glycosidic bonds. Cellulose can be trans- using MW and heterogeneous catalysts in the MW-assisted
formed into several feedstock chemicals such as glucose, fur- processes, this seems to be a promising environmentally friendly
furals, levulinic acid and other compounds, mixtures or approach in cellulose chemical processing.11
combinations thereof. Different approaches have been
Recently, several Brønsted and Lewis solid acids have been
tested as heterogeneous catalysts in cellulose conversion using
water as
a
solvent in conventional heating11,12 or MW
irradiation,13 and ionic liquids with MW irradiation.14 Alu-
minium and siliceous-based inorganic materials, such as
alumina, zeolites and other similar frameworks, have shown
some advantages due to their robust performance related to
their high thermal stability (up to 600 °C), their strong acid
sites (Lewis and Brønsted catalytic sites), and their well-known
non-toxic and non-corrosive character.15
aChemical Engineering Department, University of Guanajuato, Noria Alta s/n 36050,
Guanajuato, Gto, Mexico
bItalian National Research Council, C.N.R., Istituto di Chimica dei Composti Organo
Metallici (ICCOM) – UOS Pisa, Via G. Moruzzi 1, 56124 Pisa, Italy
cNational Research Council of Italy, C.N.R., Istituto Nazionale di Ottica,
(INO) – UOS Pisa, Via G. Moruzzi 1, 56124 Pisa, Italy. E-mail: carlo.ferrari@ino.it;
Fax: +39-050-315-2247; Tel: +39-050-315-2245
This journal is © The Royal Society of Chemistry 2014
Green Chem.

View Article Online
Paper
Green Chemistry
Glucose and some water soluble compounds are mainly Modified zeolite Beta and its characterization
obtained when zeolites are used as the catalysts, due to hydro-
lysis and dehydration reactions.16 Moreover, the product distri-
bution can be modulated if some salts of alkali and alkaline
earth metals are used (biorefining).17,18 High yield and high
selectivity for some target products, such as lactic acid, levuli-
nic acid or 5-hydroxymethylfurfural (5-HMF), can be achieved
using metal salts of zinc, chromium, aluminium, iron, copper
and so on, adopting two different approaches: conventional
heating17 and MW irradiation.18 However, the use of metal
salts involves other issues such as pollution, costs (reuse and
product purification) and safety considerations.
In order to tune the catalytic properties, zeolite Beta was modi-
fied with several metal salts using the conventional liquid ion
exchange process. A combination of the acid surface properties
(Lewis and Brønsted sites), the valency of extra-frame cation
and/or metal oxide particles deposited on the surface can turn
zeolites into special bifunctional catalysts.
Table 1 shows the results of elemental analysis using EDS.
The Si/Al ratio is increased by the activation process of NH4-
Beta to the acid form H-Beta. This variation can be attributed
to the loss of aluminium from the framework because the acti-
vation process was performed following a deep-bed high temp-
erature calcination.20
Despite the advantages of using electromagnetic radiation
and heterogeneous catalysts in carbohydrate chemistry, there
are some significant drawbacks. First, several heterogeneous
catalysts are not able to interact with MW because they are
MW-transparent. In addition, the microwave devices usually
used (microwave ovens) have a resonant behaviour and show
configurations of non-homogeneous and unpredictable elec-
tromagnetic fields. This uneven electromagnetic field can lead
to gradients in the reaction mixture temperature, which has a
direct effect on the reaction product distribution. Lastly, these
resonant devices cannot be easily scaled up to the industrial
level. The main limit is represented by the small depth of
penetration of MW power in reagents with high dielectric
losses. At the typical working frequency of 2.45 GHz, MW
power decreases to 1/e in less than two cm of water at 20 °C.
In a previous study we described the set-up and application
of a coaxial dipole antenna, a non-resonant microwave applica-
tor.19 An array of applicators can be utilized to activate uni-
formly a large volume of reagents with high MW power, each
applicator emitting radiation without interfering with the others.
The present study describes the catalytic hydrothermal
decomposition reaction of microcrystalline cellulose using the
acid form of zeolite Beta and its modification with some alka-
line metals (zinc, potassium and tin) as the heterogeneous
catalyst. The reactions were microwave-assisted, where the MW
were issued by a coaxial applicator. This “controlled” system
allowed us to investigate the interaction between the electro-
magnetic field and the different components of the reaction
system (solvent, reagents, catalyst, and products), which have
been scarcely investigated.
In the case of Sn-Beta-IE, HCl and tin tetrachloride promote
the isomorphic substitution of tetrahedral aluminium by Sn+4
into the zeolite framework.21 We obtained a high concen-
tration of tin, likely promoted by 0.3 M HCl (Table 1), although
no tetrahedral Sn+4 form was found as revealed by the absence
of the resonance shift at −450 ppm in the 119Sn magic angle
spinning NMR (results not shown).22 This result suggests that
tin is present in the form of SnO2 particles deposited on the
surface of the zeolite. A low concentration of chloride was also
found and it might be attributed to SnCl4·5H2O partially preci-
pitated at low pH.
Less incorporation of the cations K+ and Zn+2 into the Beta
framework were achieved when compared to the tin treatment.
These cations show basic behaviour in catalysis23 and were
easily exchanged into the framework.
The Beta framework of the modified zeolites after the ion
exchange process was further characterized using FTIR spec-
troscopy. The Beta framework shows two main bands associ-
ated with vibration of external linkages at 571 and 521 cm−1
(Fig. 1).24 The band at 571 cm−1 is assigned to the vibration of
five-membered rings and the band at 521 cm−1 to the presence
of double four-membered rings. Changes in the intensity of
these bands can be associated with changes in crystallinity.25
The relative crystallinity reported in Table 1 was calculated
on the basis of the integrated area of the bands at 571 and
521 cm−1 of FTIR spectra normalized with respect to the band
at 1095 cm−1 (100% crystallinity was assigned to the acid form
H-Beta and it was used as the reference).
Zeolite Beta treated with the different cation metals showed
loss of crystallinity as the intensity of both adsorption bands
The MW reactor configuration and the interaction of MW
with the catalytic system are discussed. The product distri-
bution observed in the hydrothermal decomposition of micro-
crystalline cellulose on the different catalytic systems is also
presented and discussed.
Table 1 Composition of zeolite catalyst obtained from EDS analysis
Relative crystallinityc
Si
Al
M load Si/Al
Zeolite
(w/w) (w/w) (% w)
ratio 571 cm−1 521 cm−1
Results and discussion
H-Betaa
K-Beta
53
48
50
3
3
3
1
—
4
6
32
30
30
75
100
97
99
100
87
90
The first part of this study reports the characterization of the
catalyst and its modified forms, including the interaction with
the electromagnetic field. The second part reports the pro-
ducts and yields of the catalytic MW-assisted reaction of micro-
crystalline cellulose decomposition.
Zn-Beta
Sn-Beta-IEb 36
22
91
90
a NH4-Beta (Si/Al = 25) data from the manufacturer. b 0.4 w/w of Cl was
found. c Calculated from the integrated peak area of FTIR spectra.
Green Chem.
This journal is © The Royal Society of Chemistry 2014

View Article Online
Green Chemistry
Paper
function of temperature, density and the microwave frequency,
and usually decreases with a temperature increase.9
Heterogeneous liquid phase reactions can benefit from MW
technology in terms of energy demands, faster reaction rates,
shorter residence times and space equipment saving.26 Fig. 3
shows the temperature profile of modified powder zeolites
under constant microwave power applied (100 W). Specific
sites such as silanol groups (–Si–OH) and the metal oxides
in the zeolite Beta framework have the ability to absorb
microwaves.26,27 The highest temperatures were achieved for
K-Beta ≫ NH4-Beta ≫ Sn-Beta-IE ≈ Zn-Beta ≫ H-Beta ≥ alumina.
Therefore the highest response to MW application was
obtained in the K-Beta system, while pure alumina showed the
poorest interaction of the tested systems.
Few data have been reported in the literature regarding the
interaction of microwaves with zeolites and the heterogeneous
catalytic processing of cellulose. When cellulose is processed
Fig. 1 FTIR spectra of modified zeolites: (a) K-Beta, (b) Sn-Beta-IE,
(c) Zn-Beta, (d) H-Beta.
under hydrothermal conditions,
a high temperature is
required and the dielectric properties of the medium usually
decrease. This is true if water is used as a solvent. In the reac-
tion dispersion of this study, a small amount of HCl was intro-
duced to promote the hydrolysis of microcrystalline cellulose.
Thus, the hydronium ion presence contributes to increasing
the permittivity of the medium, enhancing the heating rate.
decreases. Regarding the Sn modified zeolite, the loss of crys-
tallinity can be attributed to the dealumination promoted by
HCl and the presence of SnO2 particles in the structure (five
and double four membered rings).
SEM images of the different modified zeolites are shown in
Fig. 2. The acid form H-Beta presents small particle aggregates
(crystal size ≈ 0.5 μm) with wide shape morphology from ir-
regular to round ball aggregates. Some influence on the aggrega-
tion pattern and changes in the shape were observed due to
the variation in the Si/Al ratio and the cation exchanged. It is
well established that the structure of zeolite Beta is intrinsi-
cally deformable by changing the aluminium content and the
nature of the cation exchanged.20
Chromatographic analysis of the products of the MW-assisted
catalyzed reaction
The products obtained from the hydrothermal MW-assisted
catalysed-reaction of cellulose were analysed. Fig. 4–8 show the
results of the analysis performed by liquid chromatography
carried out on the reaction products. The effects on the reac-
tion products of processing temperature and time are shown
in Fig. 4 and 5 respectively.
Effects of MW on zeolites
The microwave interaction with a material is usually deter-
mined through the complex permittivity ε*. The dielectric loss
ε″ is a parameter representing the ability of the material to
absorb microwaves and dissipate them into heat. It is a
Four different temperatures were tested ranging from 180 to
210 °C. The reaction medium was irradiated with MW constant
Fig. 3 Temperature profile of modified powder zeolites subjected to
constant power applied of 100 W for 25 min: (a) alumina, (b) H-Beta,
(c) Sn-Beta-IE, (d) Zn-Beta, (e) NH4-Beta, (f) K-Beta.
Fig. 2 SEM microphotographs of modified zeolites: (a) H-Beta,
(b) K-Beta, (c) Zn-Beta, (d) Sn-Beta-IE.
This journal is © The Royal Society of Chemistry 2014
Green Chem.

View Article Online
Paper
Green Chemistry
Fig. 6 Effect of different catalytic approaches on the hydrothermal
reaction of microcrystalline cellulose. Reaction conditions: 5% (w/w)
cellulose, 0.5 g of zeolite in 70 ml water, T = 210 °C, P = autogenous
(P0 = 1 bar, Pmax = 18 bar), 500 RPM. Time: 30 min to achieve the target
temperature (MW power = 240 W) and then kept for 15 min (MW at
steady state ∼100–120 W).
Fig. 4 The influence of reaction temperature on the yield of identified
products. Reaction conditions: 5% of microcrystalline cellulose, 0.5 g of
zeolite in 70 ml of water, 5 mM HCl, P = autogenous (P0 = 1 bar, Pmax
=
18 bar), 500 RPM. Time: 30 min to achieve the target temperature
(MW power = 240 W) and then kept for 30 min (MW at steady state
∼100–120 W).
Fig. 7 Effect of Sn, Zn, K ion exchanged zeolites, activated neutral
alumina and acid zeolite in the hydrothermal reaction of microcrystalline
cellulose. Reaction conditions: 5% (w/w) cellulose, 0.5 g of zeolite in
Fig. 5 The influence of the reaction time on the yield of different
products. Reaction conditions: 5% of microcrystalline cellulose, 0.5 g of
70 ml water, 5 mM HCl, T = 210 °C, P = autogenous (P0 = 1 bar, Pmax
=
zeolite in 70 ml of water, 5 mM HCl, P = autogenous (P0 = 1 bar, Pmax
18 bar), 500 RPM, T = 210 °C.
=
18 bar), 500 RPM. Time: 30 min to achieve the target temperature
(MW power = 240 W) and then kept for 15 min (MW at steady state
∼100–120 W).
power for 30 min until the target temperature was reached.
Then, isothermal conditions were kept for another 30 min. produced in good yield, whereas higher reaction temperatures
Fig. 4 shows the temperature influence on the total molar yield produce a decrease in molar yield. The latter can be explained
of the identified products, changing from 12.3% to 68.6% mol by consumption in subsequent side-reactions which results
when the temperature is increased from 180 to 210 °C. The in the synthesis of smaller molecules. Lactic acid was not
product distribution strongly depends on the temperature detected at temperatures below 190 °C.
value. At high reaction temperature, the yield of levulinic acid
To further explore the yield behaviour and the product dis-
benefits since it increases from 3.4% to 27.5%. Acetic acid tribution, the reaction time was varied from 5 to 30 min when
shows a similar trend. The molar yields for glucose and the reaction conditions were set at 210 °C and autogenous
5-HMF suggest that at low temperatures these compounds are pressure (see Fig. 5). After 15 min of isothermal reaction, the
Green Chem.
This journal is © The Royal Society of Chemistry 2014

View Article Online
Green Chemistry
Paper
The use of a small concentration of HCl (5 mM) is not
harmful to the zeolite framework, as it is generally accepted
that an amorphous phase may appear for acid concentrations
above 1 M, due to structure breaking. Moreover, low leaching
of extra framework aluminium and no leaching of aluminium
framework occurs at concentrations below 0.1 M under mild
treatment conditions.28
Zeolite Beta and Sn-Beta have shown high activity in the iso-
merization of glucose to fructose at low pH = 1.0.29 The pH of
the reaction precursors was between 3.0 and 4.0. Thus, an
improvement in the yield due to leaching of some aluminium
species is not expected in these reactions.
In all the catalytic systems examined in this study the yield
of 5-HMF was relatively low (≤14.9%), probably due to the
rehydration to levulinic acid and formic acid, other side reac-
tions or humins production.30
Fig. 8 Recycling catalytic reactions were performed following the
procedure described elsewhere:12 reaction conditions for R1 were the
same as listed for Fig. 6. A second cycle R2 was carried out when fresh
cellulose was added to the slurry dispersion and the same reaction
conditions for R1 were applied. Likewise, a third cycle R3 was performed
following the same procedure for R2. Solids from R3 were centrifuged,
dried and calcined to 550 °C. A final R4 test was run using this H-Beta
calcined powder and the starting conditions for R1.
The cellulose reaction seems to follow the generally
accepted pathway reported for acid processing of carbo-
hydrates (Scheme 1).30–32 First, the cellulose depolymerizes to
glucose in acid medium. Glucose isomerizes to fructose due to
the synergy between the different catalytic sites present in the
reaction system (H+, Lewis and Brønsted sites).33 From the iso-
merisation intermediate (1,2 enediol) some dehydration reac-
tions occur with the formation of 5-HMF.32 The rehydration of
5-HMF leads to levulinic acid and formic acid or humins for-
mation.34 Humins were macroscopically detected at the end of
the reaction.
yield of glucose and 5-HMF started to decrease and then after
30 min, both reached values close to zero. Conversely, the pro-
duction of small acid molecules like levulinic acid and acetic
acid follows a growing trend. The highest yield of lactic acid
was achieved after 15 min of reaction. Increments in the reac-
tion time seem to be harmful since the total yield of the identi-
fied products begins to decrease.
Fig. 7 shows the distribution of the main products obtained
from the hydrothermal MW-assisted reaction of cellulose cata-
lyzed using ion exchanged zeolites. The catalytic activity was
Chromatographic analysis of the products of reaction in acid
medium and using ion exchanged zeolites
Once the optimum conditions of temperature and residence
time were achieved, different catalytic schemes were investi-
gated (see Fig. 6). For these experiments the temperature and
time were fixed at 210 °C and 15 min of isothermal conditions.
These conditions were chosen to avoid complete transform-
ation of glucose and 5-HMF, and have the simultaneous pres-
ence of all the chemical species which can be synthesized.
When H-Beta was used as a catalyst, a total yield of 31.6% was
achieved, with glucose as the main product (18.5%).
The addition of 5 mM HCl in the reaction medium
increased the solubility of cellulose and promoted its hydro-
lysis. Fig. 6 shows the enhancement in the total yield (78.6%)
obtained combining the heterogeneous (H-Beta) and the
homogeneous catalyst (HCl). Under these conditions levulinic
acid was the main product and its yield increased from 2.5%
to 22.3%.
The addition of 0.5 g alumina caused a decrease in the
system activity leading to 68.5% yield, whereas the selectivity
was modified to favour the production of lactic acid, at the
expense of levulinic acid.
Scheme 1 Microcrystalline cellulose decomposition pathway catalyzed
by metal zeolites.
This journal is © The Royal Society of Chemistry 2014
Green Chem.

View Article Online
Paper
Green Chemistry
Table 2 Comparison between MW and conventional heating
Experimental conditions
% Yield (mol)
Glucose
Heating method
Heating time (min)
Temp. (°C)
Levulinic acid
5-HMF
Formic acid
Conventional
MW
360
60
180
180
4.0
3.2
1.6
3.4
—
2.8
1.0
0.5
found in the following order: acid form H-Beta > K-Beta > Zn-
Beta > Sn-Beta-IE > alumina, with the following total yields: recovered by centrifugation and calcined again at 550 °C.
78.6% > 52.7% > 39.9% > 35.5% > 27.5% mol. Accordingly, the reused H-Beta of reaction R3 was recovered,
To completely restore catalytic abilities, the zeolite must be
The acid form of zeolite Beta was the most active catalyst. calcined and used under the same reaction conditions
The highest yield of glucose (29.0%) was achieved when Zn- (reaction R4 in Fig. 8), getting high levels of the total yield
Beta was used. This catalyst also promoted the lowest for- (84%). Thus, the feasibility of the catalysts reuse in the
mation of the other by-products, which implies a high rate heterogeneous microcrystalline cellulose decomposition was
associated with the hydrolysis reaction of microcrystalline demonstrated.
cellulose and the inhibition of subsequent reactions.
Comparison with conventional heating
Sn-Beta-IE shows the lowest activity of all the exchanged
zeolites, probably due to the dealuminization process induced
during the catalyst synthesis (Table 1). The fact that the dealu-
minization process decreases the amount of acid sites, as men-
tioned before, supports the idea that the systems under study
follow the acid processing of the carbohydrate pathway.
The acid form H-Beta promoted the highest formation of
levulinic acid (and formic acid) and an appreciable amount of
lactic acid. The latter is probably the product of a secondary
reaction that does not involve the 5-HMF and the well known
capability of zeolites to perform isomerisation/dehydration
reactions (see Scheme 1).35
Trace amounts of acrylic acid (41.6 and 69.4 μM) were also
detected when H-Beta and K-Beta were used. Acrylic acid is a
very desirable product due to its versatility for the production
of water soluble polymers. For comparison purposes, activated
alumina was tested under the same conditions. We found that
it presented the lowest activity and did not favour the
formation of any specific product.
In this study we also investigated the hydrothermal reaction of
microcellulose with 0.5 g of H-Beta and 5 mM HCl using con-
ventional heating for longer reaction times (360 min), achiev-
ing only yields below 7% mol while only 60 min are required
to achieve a total yield of more than 12% using microwaves at
low temperature. Table 2 summarizes the comparison of the
results.
The data in the table demonstrate the advantages of the
MW-assisted approach from every point of view: treatment
time, energy consumption and efficiency of the reaction.
The reactor used in this work is promising also because it
allows an effortless scale-up.36 Moreover, due to the fast
heating of the electromagnetic field, the homogeneously
applied electromagnetic field and the magnetic stirring, the
pressure and temperature inside the vessel can be carefully
controlled.
Experimental section
Materials
Reuse of the catalyst
Microcrystalline cellulose (Sigmacell type 50, particle size
50 μm, S5504) purchased from Sigma-Aldrich (Milan, Italy)
was employed as a substrate and used without further treat-
ment. Ammonium form commercial zeolite Beta (NH4-Beta,
CP814E) was supplied by Zeolyst International (USA) (SiO2/
Al2O3 = 25, surface area = 680 m2 g−1). Activated neutral
alumina (150 mesh, 58 Å, 267 740) was purchased from Sigma-
Aldrich (Milan, Italy). HCl (37%, m/m) was purchased from
Carlo Erba (Rodano, Milan, Italy). Deionized water obtained
with a Milli-Q system (Millipore, Bedford, MA, USA) was used
as the solvent and the reactant in all reactions.
The use of zeolites in different reactions has been demon-
strated and due to its good thermal properties their reuse is
very feasible.12 Fig. 8 shows the results of the reuse of the
solid catalyst in the MW-assisted reaction of microcrystalline
cellulose. The successive reactions were performed without
addition of fresh HCl. Good yields were achieved after 3 cycles.
The first approach was carried out adding fresh cellulose
to the slurry dispersion obtained from a reaction carried
out under standard conditions (5% (w/w) cellulose, 0.5 g of
zeolite in 70 ml water, T = 210 °C, P = autogenous, 500 RPM,
30 min to achieve the target temperature, then kept
for 15 min).12 Two processes are shown in Fig. 8 (R1 to R3).
Catalysts preparation
In these processes the system started to saturate due to the The heterogeneous catalysts were prepared using NH4-Beta as
production of unknown products or humins and the total the starting material. The acid form H-Beta was obtained after
yield tends to decrease. Interestingly, for lactic acid, in whose controlled heating of NH4-Beta for four hours at 550 °C under
synthesis the 5-HMF is not involved, the yield increased.
an air atmosphere. Potassium Beta (K-Beta) and zinc Beta
Green Chem.
This journal is © The Royal Society of Chemistry 2014

View Article Online
Green Chemistry
Paper
(Zn-Beta) were prepared using the corresponding 0.15 M aqueous
solutions of potassium sulfate (Merck, 99%) and zinc acetate
(Merck, 99.5%). Tin Beta (Sn-Beta) was prepared using a solu-
tion of 0.3 M HCl and 0.15 M tin tetrachloride pentahydrate
(Sigma-Aldrich, 98%).
The liquid ion exchange process was carried out using 10 g
of NH4-Beta in 100 ml of the corresponding salt solution. The
dispersions were stirred at 600 RPM for 15 h at 85 °C. After
this process the zeolites were filtered, washed several times (re-
dispersed in water, centrifuged and decanted, until near
neutral pH) and dried at 80 °C for 24 h and finally calcined for
four hours at 550 °C in a deep-bed under an air atmosphere.
Catalyst characterization
The morphology and particle size were analyzed using a Joel
5600 LVS Scanning Electron Microscope (SEM). The Si/Al ratio
and the metal loaded into the zeolite framework were deter-
mined using an energy-dispersive X-ray spectroscope (EDX)
mounted on the Jeol SEM detector.
Fig. 9 Schematic diagram of the microwave reactor: (1) stainless steel
vessel, (2) Teflon vessel, (3) quartz tube, (4) O-ring, (5) magnetic stir bar,
(6) thermocouple, (7) choke, (8) antenna, (9) coaxial cable, (10) pressure
sensor, (11) in-line filter, (12) data acquisition, (13) gas pressure inlet,
(14) pressure release, and (15) sampling valve.
The crystallinity of the zeolite Beta framework after the ion
exchange process was determined using FTIR spectropho-
tometry, pelletting the samples with KBr. Spectra were
recorded in the range 4000–400 cm−1 using a Spectrum 100
FTIR spectrophotometer (PerkinElmer Inc., USA). The FTIR
active section of the applicator, as a function of the permitti-
vity of the reagents. The exposed applicator tip is provided with
a threaded metal cap whose position is a function of the
dielectric permittivity of the reagents. It was therefore possible
to reduce to zero the reflected power from the reactor to the
microwave source.
The reactor is provided with pressure and temperature
sensors. A versatile PID control approach is used since the
system may operate in two ways: isothermally when the power
is regulated or adiabatically when the power is kept constant.
The latter causes a constant temperature increase, within
limits, until the power is switched off.
spectra were baseline-corrected and normalized at 1095 cm−1
.
The interaction between the solid catalysts and microwaves
was evaluated by the determination of the temperature profile
of a weighted quantity of zeolite powder under constant
applied MW power (100 W) for 25 min. For these experiments,
the solid catalyst was loaded in a quartz tube and held in a micro-
strip line. The temperature was measured using an optical
fiber thermometer placed in the middle of the catalyst powder.
Heterogeneous catalytic reaction: equipment and procedure
A typical reaction was performed loading into the reactor
microcrystalline cellulose (5% w/w) in 70 mL water and
different substrate/catalyst ratios (HCl, H-Beta, K-Beta, Zn-
Beta, Sn-Beta-IE or alumina or combinations thereof). The
reactor was closed and the pressure adjusted. The mixture was
stirred at 500 RPM, irradiated by MW at the desired power or
temperature and time. After the reaction time was established,
the power supplied was switched off and the reactor was
quenched in a cool water bath to avoid further reaction.
The catalytic reactions were carried out in a 250 ml homemade
stainless steel pressurized reactor, equipped with a coaxial
antenna as a microwave applicator (Fig. 9). The reactor consists
of a cylindrical chamber of stainless steel with two circular
caps closed with several bolts. The cap above presents a deep
recess accomplished by a tube of quartz. The seal between the
cap of stainless steel and the quartz tube is ensured by an
O-ring. The recess is used to introduce the coaxial antenna pro-
viding the MW power. The reactor has two other connections
for the introduction of the temperature probe (type K thermo-
couple) and for connection to a cylinder of nitrogen with a
Catalytic reaction product analysis
relative pressure meter (DS Europe, Mod. LP 652-100). The The reaction products in aqueous solution were analyzed
sample holder of 100 ml was made with Teflon and shaped to using an HPLC gradient pump (P4000, ThermoQuest)
reduce the transmission of heat to the steel walls of the equipped with a Rheodyne 7125 injector (Rheodyne, Cotati,
reactor. The magnetic stirrer is placed under the base of the CA, USA), a 50 μL injection loop, coupled with a mechanical
reactor.
degassing system (SC1000, ThermoFinningan) and a UV6000
The source of microwave is a magnetron oscillator equipped diode array and pulse amperometric (Au) detectors. Levulinic
with forward and reflected power indicators (SAIREM, Mod. acid, 5-HMF, lactic acid, furfuraldehyde, acetic acid, formic
GMP 03 K/SM, up to 300 W of continuous MW irradiation acid and acrylic acid were separated using a reversed-phase
power at a frequency of 2.45 GHz). A satisfying MW impedance HPLC column Hydra RP C18 (Phenomenex) 250 × 4.6 mm
matching between the reactor and the MW source was (silica particle size 4 mm), with an isocratic elution in
obtained with suitable selection of the configuration of the 100% 0.05 M PBS, pH = 2.50, and 16% methanol flowing at
This journal is © The Royal Society of Chemistry 2014
Green Chem.

View Article Online
Paper
Green Chemistry
0.8 mL min−1 and final rinsing of 100% methanol to clean the proved to be very useful especially to perform hydrothermal
column. Chromatograms were recorded at 210 nm. Calibration reactions.
curves of analytical standards were used for the quantitation
of the reaction products. Glucose was quantified using the
amperometric detector and post-column derivatization with
0.1 M NaOH flowing at 1 mL min−1. All the solutions were fil-
tered and diluted 10 times in water before analysis.
Acknowledgements
The authors gratefully acknowledge the Consejo Nacional de
Ciencia y Tecnología (CONACYT) and Instituto de Financia-
miento e Información para la Educación del Estado de Guana-
juato (EDUCAFIN) for providing financial support. The authors
would like to thank C. Lanza and F. Pardini (INO-CNR) for
their valuable technical support.
The molar yields of glucose and the products determined
by HPLC were calculated using the following equation:37
ꢀ
ꢁ
ꢀ
ꢁ
#ci
6
V ꢀ Ci
Yield mol% ¼ 100
ꢀ
ð1Þ
ncellulose
where #ci = number of carbon atoms in the product i, Ci = molar
concentration of the product i (HPLC), v = volume of aqueous
initial g of cellulose
dispersion precursors and ncellulose
¼
.
162
Notes and references
1 P. T. Anastas, L. G. Heine and T. C. Williamson, in Green
Chemical Syntheses and Processes, ed. ACS, Washington, DC,
2000, ch. 1.
2 P. J. Weimer, J. M. Lopez-Guisa and A. D. French, Appl.
Environ. Microbiol., 1990, 56(8), 2421.
3 M. Finetti, M. Daz, G. Ellenrieder and M. Marx-Figini,
World J. Microbiol. Biotechnol., 1993, 9, 251.
4 W. A. Farone and J. E. Cuzens, US Pat, 5 562 777,
1996.
5 M. Sasaki, B. Kabyemela, R. Malaluan, S. Hirose,
N. Takeda, T. Adschiri and K. Arai, J. Supercrit. Fluids, 1998,
13, 261.
6 P. Lidström, J. Tierney, B. Wathey and J. Westman, Tetra-
hedron, 2001, 57, 9225.
7 R. D. Perlack, L. L. Wright, A. Turhollow, R. L. Graham,
B. Stokes and D. C. Erbach, Report No. DOE/GO-102995-
2135 ORNL/TM-2005/66, Oak Ridge National Laboratory,
Oak Ridge, TN, 2005.
8 G. W. Huber, S. Iborra and A. Corma, Chem. Rev., 2006,
106, 4044.
9 S. Tsubaki and J. I. Azuma, in Advances in Induction and
Microwave Heating of Mineral and Organic Materials, ed.
S. Grundas, InTech, 2011, ISBN: 978-953-307-522-8.
Conclusions
The use of a microwave reactor and a heterogeneous catalyst
allowed us to obtain short reaction times and yields much
higher than reactions carried out using conventional heating.
Zeolites interact with MW, decreasing the energy consumption
and the time required for heating the reaction systems. The
heterogeneous catalyst can be easily recovered by filtration and
reused with good yield.
The addition of minimal amounts of HCl to the reagents
enhances the catalytic activity observed on the microcrystalline
cellulose hydrothermal decomposition reaction, probably due
to an enhanced permittivity of the reaction medium.
Acid form H-Beta and K-, Sn-, Zn-modified zeolite Beta are
capable of effectively catalyzing the reaction in the presence of
MW, due to their capacity to absorb MW. Moreover, it has
been demonstrated that the nature of the exchanged cation is
a key factor that affects the absorption of MW in the zeolites.
Thus, the different exchanged cations induce changes in the
product selectivity, Zn-Beta favours glucose formation, while
H-Beta favours the production of levulinic acid. Furthermore,
an increase in the amount of zeolite added to the reaction 10 M. Nüchter, U. Müller, B. Ondruschka, A. Tied and
system favours the formation of by-products through the con-
version of glucose.
W. Lautenschläger, Chem. Eng. Technol., 2003, 26, 1207.
11 A. Richel, P. Laurent, B. Wathelet, J. P. Wathelet and
M. Paquot, Catal. Today, 2011, 167, 141.
Zeolite-Beta coupled with MW seems to be a promising
approach in the development of new green processes for cellu- 12 A. Onda, T. Ochi and K. Yanagisawa, Green Chem., 2008, 10,
lose processing. 1033.
The use of the coaxial applicator developed to introduce the 13 Y. Wu, Z. Fu, D. Yin, Q. Xu, F. Liu, C. Lu and L. Mao, Green
MW into the reaction system allowed for a precise control of Chem., 2010, 12, 696.
the temperature and pressure of the system, avoiding hot 14 Z. Zhang and Z. Zhao, Carbohydr. Res., 2009, 344,
spots, and permitting to activate with MW power a chemical 2069.
reaction inside a metal vessel. The MW reactor can be easily 15 A. Corma, J. Catal., 2003, 216, 298.
scaled up and can be adapted for continuous operation, using 16 E. Taarning, C. M. Osmundsen, X. Yang, B. Voss,
an array of applicators along a tube, also made of metallic
material, without giving rise to electromagnetic resonance,
S. Andersen and C. H. Christensen, Energy Environ. Sci.,
2011, 4, 793.
without loss of radiation and therefore with great energy 17 L. Peng, L. Lin, J. Zhang, J. Zhuang, B. Zhang and Y. Gong,
efficiency. The high pressure MW-assisted chemical reactor
Molecules, 2010, 15, 5258.
Green Chem.
This journal is © The Royal Society of Chemistry 2014

View Article Online
Green Chemistry
Paper
18 J. M. Robinson, U.S. Patent Application No. US2010/0044210 28 D. M. Roberge, H. Hausmann and W. F. Hölderich, Phys.
A1, 2010. Chem. Chem. Phys., 2002, 4, 3128.
19 C. Ferrari, I. Longo, E. Tombari and E. Bramanti, J. Photo- 29 M. Moliner, Y. Román-Leshkov and M. E. Davis, Proc. Natl.
chem. Photobiol., A, 2009, 204, 115. Acad. Sci. U. S. A., 2010, 14, 6164.
20 P. J. Kunkeler, B. J. Zuurdeeg, J. C. van der Waal, J. A. van 30 S. K. R. Patil and C. R. F. Lund, Energy Fuels, 2011, 25,
Bokhoven, D. C. Koningsber and H. van Bekkum, J. Catal.,
1998, 180, 234.
21 D. E. Vaughan and S. B. Rice, US Pat., 4 933 161, 1990.
22 M. Renz, T. Blasco, A. Corma, V. Fornés, R. Jensen and
L. Nemeth, Chem.–Eur. J., 2002, 8, 4708.
4745.
31 X. Hu and C. Z. Li, Green Chem., 2011, 13, 1676.
32 B. Girisuta, L. P. B. M. Janssen and H. J. Heeres, Chem. Eng.
Res. Des., 2006, 84(A5), 339.
33 M. Moliner, Y. Roman-Leshkov and M. E. Davis, Proc. Natl.
Acad. Sci. U. S. A., 2010, 107, 6164.
23 P. Adryan, R. Ismail and F. Roessner, Kinet. Catal., 2008, 49, 587.
24 H. G. Karge, in Verified Syntheses of Zeolitic Materials, ed 34 J. Horvat, B. Klaic, B. Metelko and V. Sunjic, Tetrahedron
H. Robson, Elsevier, 2001. Lett., 1985, 26, 2111.
25 D. Prasetyoko, Z. Ramli, S. Endud, H. Hamdan and 35 R.-J. van Putten, J. C. van der Waal, E. de Jong,
B. Sulikowski, Waste Manage., 2006, 26, 1173.
26 E. Altman, G. D. Stefanidis, T. van Gerven and
A. I. Stankiewicz, Ind. Eng. Chem. Res., 2010, 49, 10287.
C. B. Rasrendra, H. J. Heeres and J. G. de Vries, Chem. Rev.,
2013, 113, 1499.
36 I. Longo, US Pat, 11/722,723, 2009.
27 M. D. Turner, R. L. Laurence, W. C. Conner and 37 F. Chambon, F. Rataboul, C. Pinel, A. Cabiac, E. Guillon
K. S. Yngvesson, AICHE J., 2000, 46, 758.
and E. Nadine, Appl. Catal., B, 2011, 105, 171.
This journal is © The Royal Society of Chemistry 2014
Green Chem.