Heterogeneous acidic TiO2 nanoparticles for efficient conversion of biomass derived carbohydrates

Article abstract of DOI:10.1039/c3gc40909k

Selective conversion of biomass derived carbohydrates into fine chemicals is of great significance for the replacement of petroleum feedstocks and the reduction of environmental impacts. Levulinic acid, 5-hydroxymethyl furfural (HMF) and their derivatives are recognized as important precursor candidates in a variety of different areas. In this study, the synthesis, characterization, and catalytic activity of acidic TiO₂ nanoparticles in the conversion of biomass derived carbohydrates were explored. This catalyst was found to be highly effective for selective conversion to value-added products. The nanoparticles exhibited superior activity and selectivity towards methyl levulinate from fructose in comparison to current commercial catalysts. The conversion of fructose to methyl levulinate was achieved with 80% yield and high selectivity (up to 80%). Additionally, conversions of disaccharides and polysaccharides were studied. Further, the production of versatile valuable products such as levulinic esters, HMF, and HMF-derived ethers was demonstrated using the TiO₂ nano-sized catalysts in different solvent systems.

Full text of DOI:10.1039/c3gc40909k

View Article Online
View Journal
Green Chemistry
Accepted Manuscript
This article can be cited before page numbers have been issued, to do this please use: C. Kuo, A. S. Poyraz, L. Jin, Y. Meng, L.
Pahalagedara, S. Chen, D. A. Kriz, C. Guild, A. Gudz and S. L. Suib, Green Chem., 2013, DOI: 10.1039/C3GC40909K.
This is an Accepted Manuscript, which has been through the RSC Publishing peer
review process and has been accepted for publication.
Accepted Manuscripts are published online shortly after acceptance, which is prior
Cutting-edge research for a greener sustainable future
www.rsc.org/greenchem
Volume 12 | Number 9 | September 2010 | Pages 1481–1676
to technical editing, formatting and proof reading. This free service from RSC
Publishing allows authors to make their results available to the community, in
citable form, before publication of the edited article. This Accepted Manuscript will
be replaced by the edited and formatted Advance Article as soon as this is available.
To cite this manuscript please use its permanent Digital Object Identifier (DOI®),
which is identical for all formats of publication.
More information about Accepted Manuscripts can be found in the
Information for Authors.
Please note that technical editing may introduce minor changes to the text and/or
graphics contained in the manuscript submitted by the author(s) which may alter
content, and that the standard Terms & Conditions and the ethical guidelines
that apply to the journal are still applicable. In no event shall the RSC be held
responsible for any errors or omissions in these Accepted Manuscript manuscripts or
any consequences arising from the use of any information contained in them.
ISSN 1463-9262
COMMUNICATION
CRITICAL REVIEW
Luque, Varma and Baruwati
Dumesic et al.
Magnetically seperable organocatalyst Catalytic conversion of biomass
for homocoupling of arylboronic acids to biofuels
1463-9262(2010)12:9;1-U
www.rsc.org/greenchem
Registered Charity Number 207890

Page 1 of 6
Green Chemistry
View Article Online
DOI: 10.1039/C3GC40909K
Heterogeneous Acidic TiO2 Nanoparticles for Efficient Conversion of
Biomass Derived Carbohydrates†
Chung-Hao Kuo, a Altug S. Poyraz, a Lei Jin, a Yongtao Meng, a Lakshitha Pahalagedara, a Sheng-Yu
Chen, a David A. Kriz, a Curtis Guild, a Anton Gudz,a and Steven L. Suib*a,b
5
Received (in XXX, XXX) Xth XXXXXXXXX 20XX, Accepted Xth XXXXXXXXX 20XX
DOI: 10.1039/b000000x
Selective conversion of biomass derived carbohydrates into fine chemicals is of great significance for the
replacement of petroleum feedstocks and the reduction of environmental impacts. Levulinic acid, 5ꢀ
hydroxymethyl furfural (HMF) and their derivatives are recognized as important precursor candidates in a
10 variety of different areas. In this study, the synthesis, characterization, and catalytic activity of acidic
TiO2 nanoparticles in the conversion of biomass derived carbohydrates were explored. This catalyst was
found to be highly effective for selective conversion to valueꢀadded products. The nanoparticles exhibited
superior activity and selectivity towards methyl levulinate from fructose in comparison to current
commercial catalysts. The conversion of fructose to methyl levulinate was achieved with 80 % yield and
15 high selectivity (up to 80 %). Additionally, conversions of disaccharides and polysaccharides were
studied. Further, the production of versatile valuable products such as levulinic esters, HMF, and HMFꢀ
derived ethers was demonstrated using the TiO2 nanoꢀsized catalysts in different solvent systems.
synthesis of platform molecules from biomassꢀderived feedstocks
such as cellulose17, saccharides18, and furfuryl alcohol19 using
various acids and solvents. Essayem et al. proposed a method for
55 the direct synthesis of methyl levulinate from cellulose using
solid acids in supercritical MeOH.17c Wang and coꢀworkers
developed an acid and metal free method for converting fructose
to HMF using an ionic liquid system.18d Mascal et al. investigated
the process of producing HMF derived ethers and levulinic acid
60 by 5ꢀ(chloromethyl)furfural (CMF).19d Furthermore, Tominaga et
al. explored the Lewis and Brønsted mixedꢀacid systems as an
efficient way to synthesize methyl levulinate.17d Despite all of the
aforeꢀmentioned progress, development of novel heterogeneous,
costꢀeffective and efficient catalysts for direct synthesis of
65 platform molecules from biomass resources is imperative for the
production of useful chemicals from biomass resources.
1. Introduction
Since the use of fossil fuels as feedstocks for fine chemical
20 synthesis and energy resources is approaching unsustainable
levels, the development of innovative strategies and resources for
the sustainable production of fuels and chemicals from renewable
materials are stimulating interest.1ꢀ4 Among these resources,
biomass has attracted enormous attention due to its considerable
25 potential as a raw material for the production of green fine
chemicals, fuels and fuel additives.5 Therefore, it is highly
desirable to convert carbohydrates to platform molecules
selectively under mild conditions, which can subsequently be
used for the production of various chemicals. The typical
30 approach for biomass degradation involves acidꢀcatalyzed
hydrolysis6 followed by either chemical or biological conversion
to monosaccharides.7 For example, glucose produced from
cellulose through hydrolysis in an aqueous medium can be
transformed to useful molecules such as alkyl glucosides, 5ꢀ
35 hydroxymethyl furfural (HMF),8 levulinic acid, and gluconic
acid.9 Among these valuable molecules, levulinic acid and ester
derivatives are among the most promising building blocks in a
biomass refinery. Levulinic acid and its esters are classified as
some of the top 12 valuable chemicals converted from biomass
40 according to the US Department of Energy’s National Renewable
Energy Laboratory (NREL).10 Levulinic acid can also be
converted into other useful chemicals such as acrylic acid,
succinic acid, pyrrolidines, and diphenolic acid.11, 12
The proposed synthetic pathways20 for the direct conversion of
monosaccharides using acid catalysts are shown in Scheme 1.
The reaction starts with the isomerization of glucose to fructose
70 followed by the Brønstedꢀacidꢀcatalyzed dehydration to produce
HMF. Subsequent rehydration of HMF yields levulinic acid.
Further reaction of these acids (levulinic or formic acids) with
alcohols can produce acid esters. HMF also reacts with alcohols
to form HMF derived ethers under specific conditions. The
75 control of the acid type on the catalyst surface is crucial for its
selectivity, since different acid types determine the formation of
different products. Several solid acids have been widely utilized
to convert carbohydrates, such as zeolites,21 metal oxides,22
polymer based acids,20 and heteropoly acids.23 Although solid
80 acids are more easily separated, recycled, and environmentally
benign, there are still considerable opportunities for improving
the selectivity and mass transfer limitations between the insoluble
polymers and heterogeneous catalysts.
A traditional approach to the synthesis of levulinic acid from
45 biomassꢀderived carbohydrates requires the use of homogeneous
mineral acids such as sulfuric acid and metal chloride salts.13ꢀ16
The use of homogeneous catalysts for biomass conversion has
been extensively studied and is known to be highly effective.
However, there are several drawbacks to this approach, including
50 catalyst separation, reactor corrosion, and recyclability. Recently,
there have been several studies conducted regarding direct
Herein, we report the facile synthesis of acidic TiO2
85 nanoparticles that can be used as catalysts in biomass derived
carbohydrate conversion. In addition, several other solid acids
such as zeolites, polymer based acids, and sulfated metal oxides

Green Chemistry
Page 2 of 6
View Article Online
DOI: 10.1039/C3GC40909K
were chosen for comparison. Detailed process parameters
suspension was then refluxed overnight. Impurities and byꢀ
2ꢀ
including reaction time, temperature, catalyst loading, and the 30 products, such as excessive SO4 and NH4+ ions, were removed
reusability of the catalyst were investigated in terms of catalytic
performance. Moreover, different carbohydrate sources and
solvents were studied.
by filtration and washing with an adequate amount of distilled
deionized water until the filtrate pH was neutral. The obtained
white powder was then dried overnight at 120 °C.
5
35 2.2 Catalyst characterization
The catalyst was characterized by powder Xꢀray diffraction
(PXRD), field emission scanning electron microscopy (FESEM),
high resolution transmission electron microscopy (HRTEM),
thermogravimetric analysis (TGA), temperature programmed
40 desorption (TPD), Xꢀray photoelectron spectroscopy (XPS) and
nitrogen adsorption/desorption [BrunauerꢀEmmettꢀTeller (BET)
surface area]. The acid types of the catalysts were determined by
pyridineꢀadsorption and FTꢀIR experiments. The amount of acid
sites was determined by thermogravimetric titration using 2,6
45 lutidine and 3,5 lutidine.25 The detailed methods are presented in
the ESI.
2.3 Typical condition for biomass conversion reaction
Carbohydrate conversion experiments were carried out in a 100
mL cylindrical stainless steel pressurized reactor made by PARR
50 Instrument Company, USA. In a typical reaction, 1 mmol of
monosaccharide (0.18 g of fructose/glucose), 0.5 mmol of
disaccharide (0.18 g of sucrose/ cellobiose) or biomass sources
(molecular weight of cellulose and starch are calculated as a
glucose unit) were used as substrates. Catalyst (0.01 to 0.20 g)
55 and solvent (20 mL) were charged into the reactor, followed by
purging with nitrogen, and pressurizing to 20 bar. The reactor
was then heated to designated temperatures for different reaction
times with stirring. After the reaction was completed, the reactor
was cooled in an ice bath. The reaction mixture was centrifuged
60 and filtered to remove the catalyst and insoluble solids before
analyses. For testing of the reusability of the catalyst, the catalyst
was collected by centrifugation after reaction, then washed with
methanol (20 mL) twice and dried at 100 °C before reuse. The
catalyst was regenerated by calcining under air at 400 °C for 1 h.
65 The detailed conversion and yield calculations are presented in
the ESI.
Scheme 1. Proposed reaction pathway for the conversion on
carbohydrates to platform molecules
10 2. Experimental Section
2.1 Catalyst preparation
3. Results and discussion
All chemicals were purchased from SigmaꢀAldrich and used
without further purification. Pꢀ25 were purchased from Degussa
15 Inc.. HꢀZSM5 (Si/Al = 50) and HꢀBeta (Si/Al = 30) were
purchased from Zeolyst Inc. and calcined at 550 °C before use.
Sulfated ZrO2 and TiO2 were made from the corresponding
hydroxides prepared by hydrolysis of a 0.4 M aqueous solution of
the metal precursor with NH4OH (25 wt% in water) at r.t. under
20 vigorous stirring, at a given final pH of ~ 10. The hydroxides
were then impregnated (via an incipient wetness technique) with
0.5 M sulfuric acid and dried at 150 °C.24 Finally, all the samples
were calcined in the air at temperatures between 450 ~ 650 °C
with a soak time of 3 h.
70 3.1 Catalytic conversion of fructose to methyl levulinate
The catalytic performance of the TiO2 nanoparticles was first
tested by fructose conversion in MeOH (Fig. 1). The reaction
started with dehydration of fructose to HMF followed by
rehydration of HMF to levulinic acid. Finally, levulinic acid was
75 further converted to methyl levulinate (ML) by esterification with
MeOH. In primary temperatureꢀdependent studies using fructose
as a substrate, TiO2 nanoparticles exhibited moderate activity (93
% conversion and 4 % ML yield) when the reaction temperature
was below 100 °C; however, when the temperature was increased
80 to 150 °C, the yield of ML was enhanced to 79 %. It was not until
225 °C that the yield of ML dropped to 68 %. This drop is
probably due to decomposition of ML at higher temperatures. In
25
The TiO2 nanoparticles were synthesized via modified
precipitation of TiOSO4.xH2SO4 (Alfa Aesar) with ammonia
solution at 85 °C. The pH value of the solution was controlled
under acidic conditions during precipitation. The resulting

Page 3 of 6
Green Chemistry
View Article Online
DOI: 10.1039/C3GC40909K
3.2 Comparison of different catalysts for the conversion of
40 fructose to ML
addition, HMF and levulinic acid were barely detected in any
experiment throughout this study, indicating rapid rehydration of
HMF and esterification to produce ML. Typically, the formation
of humins either from sugar decomposition or HMF oxidation is
believed to be the main factor that lowers the ML yield.26
Therefore, the origin of humins was further investigated by FTꢀIR
analyses (see ESI† Fig. S1). For these analyses, formed humins
were collected from two separate reactions utilizing fructose and
HMF as substrates, respectively. Both FTꢀIR spectra showed
10 similar features, such as several broad absorbance peaks located
in 900 ~ 1800 cmꢀ1 region as well as two distinct peaks located at
1596 and 1716 cmꢀ1. The latter two can be assigned to humins
formed from aldol addition/condensation polymerization of 2,5ꢀ
The catalytic performances of TiO2 nanoparticles were compared
to other TiO2 materials and solid acid catalysts such as
Amberlystꢀ15, sulfated ZrO2 and TiO2, niobic acid, and Hꢀtype
zeolites. Due to the fact that dehydration of fructose is associated
45 with the acidity of the catalyst,27 the acidities of the solid
catalysts were characterized by thermogravimetric titration using
2,6 and 3,5 lutidines.26b The catalytic results, surface areas, and
acidities are summarized in Table 1. For the acidꢀcatalyzed
dehydration of fructose, commercial TiO2 was inactive under the
50 same conditions. Pꢀ25 did not show any rehydration ability; only
HMF (20 %) was detected after reaction. Among the other tested
solid acids, Amberlystꢀ15 showed the highest activity and niobic
acid only produced a small amount of HMF (5 %). Sulfated ZrO2
and TiO2 showed the formation of the desired product ML in
55 moderate yields (62 % and 41 %, respectively). The catalytic
results are strongly correlated with surface acidity. Among the
studied materials, TiO2 nanoparticles exhibited the most acidic
sites as well as the largest surface areas. This might be the reason
why TiO2 nanoparticles were highly active for fructose
60 conversion. Moreover, the notable high activity of prepared
acidic TiO2 nanoparticles is attributed to its nanoꢀsized nature.
Ultraꢀsonicated TiO2 in MeOH is able to remain in suspension for
more than one week. Such behaviour minimizes mass transfer
limitations under reaction conditions.
5
dioxoꢀ6ꢀhydroxyhexanal (DHH), which is
a highly active
15 intermediate formed from HMF. In addition, humins collected
from the HMF reaction exhibited two additional peaks at 758 and
795 cmꢀ1, which can be attributed to polymerized HMF.26b Thus,
fructose and HMF polymerization are not the main sources for
the formation of humins. Instead, the intermediate HMF is most
20 likely rapidly converted to ML contributing to the higher yields
of ML at short reaction times.
To investigate the effects of catalyst loading on activity,
different amounts of catalyst were used (see ESI† Fig. S2). The
studies showed that TiO2 nanoparticles exhibit high activity even
25 with lower catalyst loading. For example, only 5.5 wt% catalyst
loading (with respect to fructose) was required for a 67 % yield of
ML with 95 % conversion. The product mixture in methanol
solution was analyzed by NMR without further purification and
was identified as ML with minimal impurities (see ESI† Fig. S3).
30 In contrast to homogeneous acid catalyzed reactions, the
purification process in this system only requires filtration and
solvent removal, which affords a straightforward and economical
approach.
65
Classic acidꢀrich Hꢀtype zeolites like HZSMꢀ5 and Hꢀbeta
were also tested under the same reaction conditions.
Thermogravimetric titration results showed that the acidities of
HZSMꢀ5 and Hꢀbeta catalysts were higher than that of TiO2
nanoparticles. However, HZSMꢀ5 and Hꢀbeta exhibited lower
70 selectivities for fructose conversion than the TiO2 nanoparticles.
Both zeolites converted fructose not only to ML and methyl
lactate but also unidentified side products. This might be due to
the mixture of acid types in zeolites. The Hꢀtype zeolites contain
both Brønsted and Lewis acid sites on the surface.28 As shown in
75 Scheme 1, Brønsted acids allow reactions to undergo dehydration
then rehydration to produce ML; Lewis acids, on the other hand,
catalyze a retroꢀaldol reaction to generate methyl lactate. Also,
the relatively stronger acidities of Hꢀtype zeolites were not
suitable for fructose conversion under optimized conditions for
80 TiO2 nanoparticles. Consequently, the selectivity toward ML was
significantly lowered to 27~15 % for Hꢀtype zeolites. The TiO2
nanoparticle catalyst provide appropriate acidity to catalyze both
the dehydration and further rehydration of fructose.
85 3.3 Investigations on conversion of different biomass sources
Investigation of the conversion of different biomass sources
demonstrated the practical applicability of the synthesized TiO2
catalyst (Table 2). In glucose conversion, the reaction pathway
involves isomerization of glucose to fructose. Fructose is then
90 converted to ML in MeOH. The timeꢀdependent studies (see ESI†
Fig. S4) showed the coꢀexistence of glucose and fructose in the
first 5 hours, indicating a slow glucose isomerization. The yield
of ML stabilized at a maximum of 61% after 9 hours. The rate of
glucose isomerization controls the selectivity to ML. The use of
95 disaccharides such as sucrose (a disaccharide of glucose and
35 Fig.1 Influence of the reaction temperature on fructose conversion and
ML yield. (Reaction conditions: 20 mL 0.05 M fructose in MeOH, 0.1 g
TiO2 nanoparticles, 1 h.)

Green Chemistry
Page 4 of 6
View Article Online
DOI: 10.1039/C3GC40909K
fructose) and cellobiose (a disaccharide of glucose) requires
initial hydrolysis of disaccharide into the monosaccharides.
Sucrose and cellobiose were also selectively converted to ML
with high yields (65 % and 58 %, respectively), indicating that
sequential reactions
5
Table 1. Summary of catalytic results and characterization for different solid acid catalystsa
Yield / Selectivity (%)
N of Acid sites by N of Acid sites by
BET
Conversion (%)b
2,6 Lutidine
3,5 Lutidine
ꢁmol/ge
218
HMFb
MLc
MLAc
Catalyst
(m2/g)
ꢁmol/ge
303
19
TiO2 nanoparticles
Commercial TiO2
Pꢀ25 TiO2
Sulfated ZrO2
Sulfated TiO2
Amberlystꢀ15
Nb2O5
>99
23
87
89
82
0/(0)
3/(13)
20/(23)
25/(28)
33/(40)
0/(0)
79/(80)
0/(0)
0/(0)
62/(70)
41/(50)
66/(60)
0/(0)
ꢀd
ꢀ
ꢀ
ꢀ
ꢀ
238
18g
56g
106
98
20
48
42
207
144
457f
5
154
127
>99
76
ꢀ
ꢀ
35
20
457f
5/(7)
125
HꢀZSM5
(Si/Al =50)
HꢀBeta
98
72
0/(0)
0/(0)
26/(27)
11/(15)
21/(21)
23/(32)
426g
272
607
674
628
680g
(Si/Al =30)
a Reaction conditions: MeOH (20 mL), fructose (1 mmol), catalyst (0.1 g), 175 °C, 1 h under N2. The average yields are reported. b Conversions and yields
c
of HMF were obtained by HPLC analysis. Yields and selectivities of methyl levulinate (ML) and methyl lactate (MLA) were obtained by GCꢀFID
d
f
analysis. Not observed product. e The number of acid amount measurements were done by thermogravimetric titration method. Due to the thermal
10 instability of Amberlystꢀ15, the acidity data was acquired from literature.31 g acquired from literature. 31
(hydrolysis of disaccharides, isomerization of glucose to fructose
and finally conversion to ML) were successfully catalyzed. The
main differences between the conversions of sucrose and
cellobiose are the reaction times; sucrose only requires 3 hours
rehydrated products such as levulinic acid were detected in
aprotic solvents. Among all tested aprotic solvents,
tetrahydrofuran (THF) gave the highest HMF yield (54 %).
Although fructose solubility is higher in solvents with relatively
15 while cellobiose requires 9 hours to complete the reaction. This 45 high boiling points [such as dimethyl sulfoxide (DMSO),
difference may be due to the hydrolysis of sucrose to glucose and
fructose units, which is feasible for conversion of ML. Besides
these monosaccharides (fructose, glucose) and disaccharides
(sucrose, cellobiose), polysaccharides (cellulose and starch) were
dimethylformamide (DMF), and dimethylacetamide (DMA)] the
use of these solvents produced more humins during the reaction
(as indicated by a deep brown color observed after the reaction).
The formation of humins suppressed the HMF yield in these
20 also tested under the same conditions as fructose conversion. 50 solvents.
Both cellulose and starch reached ML yields around 40% after 20
hours. Conversion of cellulose to ML involves depolymerization
of cellulose to glucose and the subsequent transformation of
glucose to ML; therefore, a longer reaction time was needed for
Regarding the utilization of levulinic esters as diesel blend
components, lower molecular weight levulinic esters have shown
some disadvantages. ML is miscible with water, which is difficult
to manage. Another disadvantage is that ML might separate from
25 this reaction. The depolymerization of cellulose/starch to glucose 55 biodiesel at low temperatures. However, higher molecular weight
may be hindered in organic solvents by solubility constraints.
levulinic esters have higher octane numbers, which are thermally
more stable. Mixing diesel with higher boiling point levulinic
esters can also reduce volatile organic compound (VOC)
emissions, and is more environmentally friendly.29 Therefore, nꢀ
60 ethanol, propanol, and butanol were used as solvents in order to
produce levulinic esters with higher molecular weights. The
results showed that fructose converted to different nꢀalkyl
levulinates can reach 71~ 78% yield in three hours. Zhang et al.
tried to produce alkyl levulinates in high yields.29b However, the
65 expensive furfuryl alcohol was needed as a precursor while our
process uses economical raw precursors such as fructose to
produce versatile levulinic esters using a TiO2 nanoparticle
catalyst.
The formation of levulinic esters from fructose is a consecutive
70 reaction which probably proceeds by either rehydration of HMFꢀ
ether to form levulinic ester or by rehydration of HMF to
levulinic acid, followed by esterification to form levulinic ester.
Interestingly, the use of higher steric hindrance alcohols such as
2ꢀpropanol, 2ꢀbutanol and tꢀbutanol led to the HMFꢀderived
75 ethers as main products as opposed to levulinic esters. The
reaction was not favorable for further rehydration to form
levulinic esters, so the generation of other products such as
levulinic acid or esters was suppressed. The yields of HMFꢀ
Table 2. Methyl levulinate yields from different biomass carbohydrates
catalyzed by TiO2 nanoparticlesa
Yield / Selectivity (%)
Substrate
Conversion (%)b
MLc
80/(80)
Fructose
Glucose
Sucrose
Cellobiose
Cellulose
Starch
>99
>99
82, >99
68, >99
72
61/(61)e
51/(62), 65/(65)d
17/(25), 58/(58)e
42/(58)f
67
40/(60)f
a
30 Reaction conditions: MeOH (20 mL), fructose (1 mmol), catalyst (0.1
g), 175 °C, 1 h under N2. All reactions were repeated three times. The
b
average yields are reported. Conversions were obtained by HPLC
c
analysis. Yields and selectivities of ML were obtained by GCꢀFID
d
e
f
analysis. Reaction time: 3 hours. Reaction time: 9 hours. Reaction
35 time: 20 hours .
3.4 Investigations on conversion in different solvents
Utilization of TiO2 nanoparticles for converting fructose in
various solvents was investigated, and the results are given in
40 Table 3. Fructose was dehydrated to HMF, but no further

Page 5 of 6
Green Chemistry
View Article Online
DOI: 10.1039/C3GC40909K
derived ethers were 63 ~ 68% in 2ꢀpropanol, 2ꢀbutanol and tꢀ 40 use. The results are shown in Figure 2. The yield of ML from 80
butanol. HMFꢀderived ethers such as 5ꢀethoxymethylfurfural
have already been tested, and exhibited comparable energy
density relative to standard gasoline.30 The advantage of forming
HMFꢀderived ethers is that they are much more stable and easier
% dropped to 70 % after the 3rd use but stabilized at around 70 %
in further uses for TiO2 nanoparticles. On the other hand, the
activity of sulfated ZrO2 and TiO2 catalysts drastically dropped;
both of the sulfated catalysts were almost completely deactivated
5
to work with than HMF. Therefore, by adjusting nꢀalkyl alcohols 45 after the 5th use. Strong acidity for the sulfated catalysts is
to higher sterically hindered alcohols, another selection of
valuable platform chemicals (HMFꢀderived ethers) was obtained.
primarily due to the chelation of sulfate ions on the surface of
metal oxides.31 Therefore, the thermal stability of TiO2
nanoparticles was also tested. In Table S2 (see ESI† Table S2),
the results indicated the sulfur concentration of the TiO2
50 nanoparticles for different treatments. Almost no sulfur was
detected for the catalyst calcined at 800 °C which showed no
activity toward ML production. The deactivation might occur due
to sulfate ion decomposition. Deactivation of the sulfated metal
oxide catalysts due to sulfate ion leaching or the thermal
55 instability of the catalysts has been discussed in several other
studies.31 The postꢀtreatment of catalysts to remove humins might
also be responsible for sulfate ion removal on the metal oxide
surface due to thermal decomposition. Ultimately, synthesized
acidic TiO2 nanoparticles showed superior behavior in preserving
60 activity and thermal stability.
Table 3. Effect of solvents on fructose conversions and products yields
10 catalyzed by TiO2 nanoparticlesa
Conversion
Solvent Time (h)
(%)b
Yield (%)
HMF Levulinic
HMFb
LAb
etherc
esterc
1
3
3
3
3
3
1
3
1
3
1
3
1
3
1
3
1
3
>99
>99
66
>99
>99
>99
97
>99
98
>99
96
>99
>99
>99
96
>99
98
42
54
19
31
30
ꢀd
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
THF
CH3CN
DMF
DMSO
DMA
35
ꢀ
ꢀ
ꢀ
n.d.
n.d.
n.d.
n.d.
< 1%
< 1%
n.d.
n.d.
< 1%
< 1%
9
18
n.d.
14
n.d.
52
68
12
n.d.
54
67
45
63
58
71
66
78
8
13
67
75
4
n.d.
n.d.
n.d.
n.d.
2
EtOH
nꢀPrOH
2ꢀPrOH
nꢀBuOH
4
n.d.
n.d.
2
6
8
2ꢀBuOH
10
3
7
tꢀBuOH
>99
7
9
a
Reaction conditions: solvent (20 mL), fructose (1 mmol), catalyst (0.1
b
g), 150 °C, under N2. Conversion and yield of HMF and levulinic acid
were obtained by HPLC analysis. Yields of levulinic esters and HMFꢀ
derived ethers were obtained by GCꢀFID analysis.
15 product.
C
d
Not observed
3.5 Investigations of the catalysts reusability
Reusability is also a key factor in heterogeneous catalysis.
Results indicate that the yields of ML gradually decreased to 40
20 % after the 5th use (Fig. 2). The color of the catalyst changed
from white to deep brown, indicating an accumulation of humins
on the surface of catalyst over time, possibly causing
deactivation. Therefore, post treatment of the catalyst was
performed by calcination at 400 °C for 1 h. The white color of the
25 catalyst was restored, and the yields reverted to 70 % with only
around a 10 % loss of yield after the 6th use. The recovered
catalysts were further characterized. The XRD patterns (see ESI†
Fig. S5) and SEM images (see ESI† Fig. S7) of recovered TiO2
nanoparticles showed no significant changes after calcination.
30 These characterization studies showed that the morphology of
TiO2 nanoparticles remained the same after reaction and
regeneration.
Sulfated metal oxides are well known strong solid acid
catalysts and are widely tested for catalytic reactions.31 The main
35 drawback of sulfated metal oxides is the unstable activity due to
sulfate leaching problems. Therefore, we compared the stability
of TiO2 nanoparticles to other sulfated ZrO2 and TiO2 prepared
by impregnation methods. To reduce interferences, humins
adsorbed on the catalysts were removed by calcination after each
Fig.2 Reusability of TiO2 nanoparticles and sulfated metal oxides. The
o
catalysts were calcined at 400 C for 1 h before reuse except the one for
TiO2 nanoparticles without calcination (▼) and calcined at the 6th use (◆
65 ). (Reaction conditions: 20 mL 0.05 M fructose in MeOH, 0.1 g catalyst,
175 °C, 1 h.)
4. Conclusions
In summary, we report a oneꢀstep synthesis of acidic TiO2
70 (anatase) nanoparticles that can produce high yields of ML in
MeOH from different biomass derived carbohydrates under
moderate conditions. The ML yields obtained from fructose were
as high as 80 % with 80 % selectivity. Moreover, HMF, HMFꢀ
derived ethers and higher molecular weight levulinic esters can
75 also be obtained by changing solvent feeds. Remarkable catalytic
activity of the nanoparticles is attributed to in-situ sulfation on the
surface and their superior dispersion in reaction media. These

Green Chemistry
Page 6 of 6
View Article Online
DOI: 10.1039/C3GC40909K
65 18. a) S. Saravanamurugan, A. Riisager, Catal. Commun., 2012, 17, 71ꢀ
75; b) L. C. Peng, L. Lin, H. Li, Q. L. Yang, Appl. Energy, 2011, 88,
4590ꢀ4596; c) S. Saravanamurugan, O. N. Van Buu, A. Riisager,
ChemSusChem, 2011, 4, 723ꢀ726; d) Y. N. Li, J. Q. Wang, L. N. He,
Z. Z. Yang, A. H. Liu, B. Yu, C. R. Luan, Green Chem., 2012, 14,
novel catalytic nanoparticles allow for the conversion of biomass
derived carbohydrates efficiently at low temperatures ranging
from 150 to 200 °C. The catalysts also show high recyclability
with a minor loss of performance. These newly discovered acidic
TiO2 nanoparticles for biomass conversion open up a new avenue
of cost effective biomass refinery processes toward the
production of affordable bioꢀchemicals and biofuels.
70
2752ꢀ2758.
5
19. a) G. M. G. Maldonado, R. S. Assary, J. A. Dumesic, L. A. Curtiss,
Energy Environ. Sci., 2012, 5, 8990ꢀ8997; b) Z. H. Zhang, K. Dong,
Z. B. Zhao, ChemSusChem 2011, 4, 112ꢀ118; c) J.ꢀP. Lange, W. D.
van de Graaf, R. J. Haan, ChemSusChem 2009, 2, 437ꢀ441; d) M.
Mascal, E. B. Nikitin, ChemSusChem 2009, 2, 859ꢀ861.
20. A. Takagaki, M. Ohara, S. Nishimura and K. Ebitani, Chem.
Commun., 2009, 6276ꢀ6278.
75
Acknowledgements
21. a) M. S. Holm, S. Saravanamurugan and E. Taarning, Science, 2010,
328, 602ꢀ605; b) J. Jow, G. L. Rorrer, M. C. Hawley and D. T. A.
Lamport, Biomass, 1987, 14, 185; c) K. Lourvanij and G. L. Rorrer,
Appl. Catal. A, 1994, 109, 147.
22. J. L. a. L. Wang, Adv. Mater. Res., 2010, 96, 183.
23. a) Z. Sun, M. Cheng, H. Li, T. Shi, M. Yuan, X. Wang and Z. Jiang,
RSC Advances, 2012, 2, 9058; b) R. Palkovits, K. Tajvidi, A. M.
Ruppert and J. Procelewska, Chem. Commun., 2011, 47, 576ꢀ578.
24. A. Corma, A. Martínez and C. Martínez, Appl. Catal. A, 1996, 144,
249ꢀ268.
10 This work was supported by the National Science Foundation
CBET Division, under GOALI program 0827800. We would like
to thank Dr. Francis Galasso for many helpful discussions.
80
85
90
Notes and references
a Department of Chemistry, University of Connecticut, U-3060, 55 North
15 Eagleville Rd., Storrs, Connecticut 06269 (USA). Fax: +1 860 486 2981;
Tel: +1 860 486 2797; E-mail: steven.suib@uconn.edu
b Institute of Materials Science, University of Connecticut, U-3060, 55
North Eagleville Rd., Storrs, Connecticut 06269 (USA)
25. S. L. Soled, G. B. McVicker, L. L. Murrell, L. G. Sherman, N. C.
Dispenziere, S. L. Hsu, D. Waldman, J. of Catal., 1988, 111, 286ꢀ
295.
20 † Electronic
Supplementary
Information
(ESI)
available:
26. a) X. Hu and C.ꢀZ. Li, Green Chem., 2011, 13, 1676; b) S. K. R.
Patil, J. Heltzel, C. R. F. Lund, Energy Fuels, 2012, 26, 5281ꢀ5293.
27. M. Watanabe, Y. Aizawa, T. Iida, R. Nishimura, H. Inomata, Appl.
Catal. A, 2005, 295, 150ꢀ156.
95 28. a) G. L. Woolery, G. H. Kuehl, H. C. Timken, A. W. Chester and J.
C. Vartuli, Zeolites, 1997, 19, 288; b) Y. Meng, H. C. Genuino, C.ꢀH.
Kuo, H. Huang, S.ꢀY. Chen, L. Zhang, A. Rossi, S. L. Suib, J. Am.
Chem. Soc., 2013, 135, 8594ꢀ8605.
Characterizations details of TiO2 nanoparticles, including XRD
diffractograms, SEM and TEM images, FTꢀIR spectra, NH3ꢀTPD
analysis, and TGA analysis. Experimental detalis, effect of catalyst
loading on conversion of fructose, effect of time on conversion of
25 gulcose, and NMR spectrum of reaction products. See
DOI: 10.1039/b000000x/
29. a) L. Lomba, B. Giner, I. Bandrés, C. Lafuente and M. R. Pino, Green
1. C. H. Zhou, X. Xia, C. X. Lin, D. S. Tong and J. Beltramini, Chem.
Soc. Rev., 2011, 40, 5588ꢀ5617.
30 2. D. M. Alonso, J. Q. Bond and J. A. Dumesic, Green Chem., 2010, 12,
1493.
100
Chem., 2011, 13, 2062; b) Z. Zhang, K. Dong and Z. K. Zhao,
ChemSusChem, 2011, 4, 112ꢀ118; c) H. Joshi, B. R. Moser, J. Toler,
W. F. Smith and T. Walker, Biomass and Bioenergy, 2011, 35, 3262ꢀ
3266; d) E. Christensen, A. Williams, S. Paul, S. Burton and R. L.
McCormick, Energy Fuels, 2011, 25, 5422ꢀ5428.
3. P. N. R. Vennestrom, C. M. Osmundsen, C. H. Christensen and E.
Taarning, Angew. Chem. Int. Ed. Engl., 2011, 50, 10502ꢀ10509.
4. D. J. Hayes, Catalysis Today, 2009, 145, 138ꢀ151.
35 5. A. J. Ragauskas, C. K. Williams, B. H. Davison, G. Britovsek, J.
Cairney, C. A. Eckert, W. J. Frederick, Jr., J. P. Hallett, D. J. Leak,
C. L. Liotta, J. R. Mielenz, R. Murphy, R. Templer and T.
Tschaplinski, Science, 2006, 311, 484ꢀ489.
6. R. Rinaldi and F. Schuth, ChemSusChem, 2009, 2, 1096ꢀ1107.
40 7. R. Kumar, S. Singh and O. V. Singh, J. Ind. Microbiol. Biotechnol.,
2008, 35, 377ꢀ391.
8. R. J. van Putten, J. C. van der Waal, E. de Jong, C. B. Rasrendra, H.
J. Heeres and J. G. de Vries, Chem. Rev., 2013, 113, 1499ꢀ1597.
9. C. Baatz and U. Pruse, J. Catal., 2007, 249, 34ꢀ40.
45 10. T. Werpy and G. Petersen, Top Value Added Chemicals from
Biomass: Volume 1 - Results of Screening for PotentialCandidates
from Sugars and Synthesis Gas, National Renewable Energy
Laboratory, Golden, CO, 2004.
105 30. a) G. A. Kraus, T. Guney, Green Chem., 2012, 14, 1593ꢀ1596; b) L.
E. Manzer, Top Catal., 2010, 53, 1193ꢀ1196.
31. a) S. C. Brown and J. S. J. Hargreaves, Green Chem., 1999, 1, 17; b)
J. A. Melero, J. Iglesias and G. Morales, Green Chem., 2009, 11,
1285; c) K. Suwannakarn, E. Lotero, J. Goodwinjr and C. Lu, J. of
110
Catal., 2008, 255, 279ꢀ286; d) G. D. Yadav and A. D. Murkute, J. of
Catal., 2004, 224, 218ꢀ223; d) Y.ꢀH. Chin, W. E. Alvarez and D. E.
Resasco, Catalysis Today, 2000, 62, 159; e) A. Corma, A. Martinez
and C. Martinez, Appl. Catal. A, 1996, 144, 249; f) S. Talwalkar, P.
Kumbhar, S. Mahajani, Catal. Commun., 2006, 7, 717ꢀ720.
115 32. a) J. R. Sohn and D. C. Shin, Appl. Catal. B, 2008, 77, 386ꢀ394; b) H.
Nakabayashi, Bull. Chem. Soc. Jpn., 1992, 65, 914.
33. K. Okumura, T. Tomiyama, S. Shirakawa, S. Ishida, T. Sanada, M.
Arao and M. Niwa, J. Mater. Chem., 2011, 21, 229.
11. M. Kitano and F. Tanimoto, Chem. Econ. Eng. Rev., 1975, 7, 25.
50 12. B. V. Timokhin, V. A. Baransky and G. D. Eliseeva, Russ. Chem.
Rev., 1999, 68, 73.
120
13. S. W. Fitzpatrick, U.S. Pat. 5608105, 1997.
14. K. Seri, T. Sakaki, M. Shibata, Y. Inoue and H. Ishida, Bioresour.
Technol., 2002, 81, 257.
55 15. L.ꢀP. Peng, L. Lin, J. Zhang, J. Zhuang, B. Zhang and Y. Gong,
Molecules, 2010, 15, 5258.
16. J. Hegner, K. C. Pereira, B. DeBoef and B. L. Lucht, Tetrahedron
Lett., 2010, 51, 2356.
17. a) X. Y. Wu, J. Fu, X. Y. Lu, Carbohydr. Res., 2012, 358, 37ꢀ39; b)
60
R. Le Van Mao, Q. Zhao, G. Dima, D. Petraccone, Catal. Lett., 2011,
141, 271ꢀ276; c) F. Rataboul, N. Essayem, Ind. Eng. Chem. Res. 2011,
50, 799ꢀ805; d) K.ꢀI. Tominaga, A. Mori, Y. Fukushima, S. Shimada,
K. Sato, Green Chem., 2011, 13, 810ꢀ812; e) Y. Hishikawa, M.
Yamaguchi, S. Kubo, T. Yamada, J. Wood Sci., 2013, 59, 179ꢀ182.