Room temperature versatile conversion of biomass-derived compounds by means of supported TiO2 photocatalysts

Article abstract of DOI:10.1016/j.molcata.2012.09.018

The selective oxidation of glucose and degradation of phenol in liquid phase were studied in the presence of supported TiO₂ photocatalysts. Photocatalysts were synthesized by a modified ultrasound-assisted sol-gel method. The fact of supporting TiO₂ on zeolite type Y is giving more selective photocatalyst in glucose oxidation toward glucaric acid (GUA) and gluconic acid (GA) (total selectivity approx. 68%, after 10 min illumination time and 50%H₂O/50%acetonitrile solvent composition) than the unsupported TiO₂ and the commercially available photocatalyst Evonik P-25. Photocatalysts worked at room temperature, atmospheric pressure and very short reaction times (up to 15 min). Additionally, these photocatalysts were investigated in the mineralization of phenol as a cellulosic industries water contaminant. It was observed that fumed silica is a better option than zeolite as titania support in phenol aqueous degradation. Ultrasound-promoted sol-gel methodology is giving promising supported TiO₂ photocatalysts for water purification and solar chemicals production.

Full text of DOI:10.1016/j.molcata.2012.09.018

Journal of Molecular Catalysis A: Chemical 366 (2013) 156–162
Contents lists available at SciVerse ScienceDirect
Journal of Molecular Catalysis A: Chemical
journal homepage: www.elsevier.com/locate/molcata
Room temperature versatile conversion of biomass-derived compounds by
means of supported TiO2 photocatalysts
Juan C. Colmenares∗, Agnieszka Magdziarz
Institute of Physical Chemistry PAS, Kasprzaka 44/52, 01-224 Warsaw, Poland
a r t i c l e i n f o
a b s t r a c t
Article history:
The selective oxidation of glucose and degradation of phenol in liquid phase were studied in the presence
of supported TiO2 photocatalysts. Photocatalysts were synthesized by a modified ultrasound-assisted
sol–gel method. The fact of supporting TiO2 on zeolite type Y is giving more selective photocatalyst in
glucose oxidation toward glucaric acid (GUA) and gluconic acid (GA) (total selectivity approx. 68%, after
10 min illumination time and 50%H2O/50%acetonitrile solvent composition) than the unsupported TiO2
and the commercially available photocatalyst Evonik P-25. Photocatalysts worked at room temperature,
atmospheric pressure and very short reaction times (up to 15 min). Additionally, these photocatalysts
were investigated in the mineralization of phenol as a cellulosic industries water contaminant. It was
observed that fumed silica is a better option than zeolite as titania support in phenol aqueous degradation.
Ultrasound-promoted sol–gel methodology is giving promising supported TiO2 photocatalysts for
water purification and solar chemicals production.
Received 12 September 2012
Received in revised form
17 September 2012
Accepted 17 September 2012
Available online 25 September 2012
Keywords:
Selective photocatalysis
Glucose photo-oxidation
Phenol aqueous degradation
TiO2
Sol–gel synthesis
Sonication
© 2012 Elsevier B.V. All rights reserved.
1. Introduction
selective oxidation of cellulose-derived compounds can be driven
by photocatalytic reactions providing high-value chemicals [4,5],
Paper and pulp industry provides a significant amount of
residues that constitute a renewable source of lignocellulosic mate-
rial. This abundant, cheap and non-edible biomass replaced edible
biomass feedstock (e.g. sugars, starches and edible oils) after the
food versus fuel debate that emerged as a response to the sharp
increase in food prices several years ago. Such replacement permit-
ted sustainable production of fuels and chemicals without affecting
food supplies or forcing extensive changes in land use [1]. Ligno-
cellulose is a polymeric material composed of three primary units:
cellulose (a crystalline polymer of glucose units), hemicellulose (an
amorphous polymer of five different C5 and C6 sugars) and lignin (a
three-dimensional polymer of propyl-phenol that surrounds cellu-
lose and hemicellulose) [2].
In the cellulose-derived industries, phenols are among the
most common contaminants that can be found in water and,
due to their high toxicity, they are strictly regulated. Further-
more, their complete removal from water is relatively difficult
because of their high stability in solution. Hence, photocatalysis
has become a very promising method to eliminate such water
organic pollutants [3]. Alternatively, in the same type of industries,
among them the so-called platform molecules including a great
variety of carboxylic acids (succinic, fumaric and malic acids,
2,5-furandicarboxylic, 3-hydroxypropionic, aspartic, glucaric, glu-
tamic, itaconic, and levulinic acid) and many other building-block
molecules (3-hydroxybutyrolactone, glycerol, sorbitol, and xyli-
tol/arabitol) [6].
Heterogeneous photocatalysis employs semiconductor materi-
als and among them titanium dioxide has been considered to be the
most attractive due to its low cost, non-toxicity, strong oxidizing
power, excellent photocatalytic activity and long-term photostabil-
ity [7]. However, its main drawback is that it can only be activated
under UV light irradiation (ꢀ < 388 nm) due to its large band gap
energy, which restricts its application under the whole solar light
spectrum [8]. To solve this problem, apart from other expensive
and complicated procedures, ultrasound-promoted methodologies
have been used to prepare visible-light active and well crystal-
lized mesoporous TiO2 photocatalysts [9–11]. Moreover, there are
several limitations to the use of “bare” TiO2 in photocatalytic
reactors, because of the fact that TiO2 aggregates rapidly in suspen-
sion, resulting in smaller effective surface area and lower catalytic
efficiency. The small size of TiO2 particles also complicates the
filtration of suspensions, making slurry photocatalytic reactors
impractical [12]. Therefore, it has been suggested that fine TiO2
particles should be anchored on supports for the ease of handling
as well as to improve adsorption and photocatalytic efficiency [3].
∗
Corresponding author. Tel.: +48 22 343 3215.
E-mail addresses: jcarloscolmenares@ichf.edu.pl, jcarlos@wp.pl
(J.C. Colmenares).
1381-1169/$ – see front matter © 2012 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.molcata.2012.09.018

J.C. Colmenares, A. Magdziarz / Journal of Molecular Catalysis A: Chemical 366 (2013) 156–162
157
In this context, there have been numerous attempts to immobilize
TiO2 catalyst on different supports as ceramic [13], silica [14,15],
activated carbon [16,17], clay [18,19], and zeolite [20,21]. From
this group, zeolites and silica distinguish themselves as supports
due to their photocatalytically attractive features as good pollutant
adsorption, diffusion properties and absence of light absorption
[3]. Additionally, zeolites have been reported to delocalise band
gap excited electrons of TiO2 and thereby minimize electron–hole
recombination as if to favor photoinduced electron-transfer reac-
tions [22].
In the present investigation, TiO2 supported on zeolite and sil-
ica materials has been synthesized for the first time by a modified
sol–gel method assisted by ultrasonic irradiation. According to
the chemical composition of lignocellulose, glucose as monomer
of cellulose and phenol as representative water pollutant of the
lignocellulosic industries have been chosen to serve as model
compounds in liquid phase photocatalytic reaction tests: selec-
tive photocatalytic oxidation of glucose and phenol photocatalytic
degradation.
reagents were of analytical grade and used without any further
purification.
All catalytic reactions were performed in a quartz (for glu-
cose)/borosilicate glass (for phenol) cylindrical double-walled
immersion well reactor with a total volume of 450 mL.
2.3. Case study #1: liquid-phase photocatalytic selective
oxidation of glucose
The reaction system was stirred magnetically at 700 rpm to get
a uniform suspension of the catalyst in the solution. A medium
pressure 125 W mercury lamp (ꢀmax = 365 nm) supplied by Pho-
tochemical Reactors Ltd. (Model RQ 3010) was placed inside the
quartz immersion well as light irradiation source. The reaction
temperature was established at 30 ◦C. Glucose solutions (2.8 mM)
were prepared in a mixture of Milli-Q water and acetonitrile (ACN)
(10:90, v/v) unless otherwise specified. Experiments were car-
ried out from 150 mL of mother solution and a concentration
of 1 g/L of the catalyst was used. All reactions were carried out
under ambient air (no oxygen bubbling conditions). Approx. 2 mL
of samples were taken from the photoreactor at pre-specified
periods of time and were filtrated (0.20 ␮m, 25 mm ␸ nylon fil-
ters) in order to remove TiO2-supported particles before HPLC
analyses.
2. Experimental
2.1. Preparation of photocatalysts
The quantitative analyses for glucose selective conversion
were measured, after calibration, by high-performance liquid
chromatograph (Waters HPLC Model 590 pump), equipped with
refractive index detector (Waters 2414 Refractive Index Detec-
tor). Separation was performed on a XBridgeTM Amide 3.5 ␮m
4.6 mm × 150 mm Column provided by Waters. Mobile phase was
MilliQ-water/acetonitrile (15:85, v/v) at a flow rate of 0.8 mL/min.
The injection volume was 10 ␮L.
All reaction products were identified by GC–MS (after sily-
lation using N,O-bis(trimethylsilyl)trifluoroacetamide with 1%
trimethylchlorosilane as derivatizating agent and performed at
60 ◦C for 1 h in a reactor CSB/COD-Reaktor ET 108, Aqualytic,
Germany). All products identified by GC–MS were confirmed by
LC/MS analysis. These analytical methods using both GC/MS and
LC/MS were able to detect all intermediates (also for the case study
#2 of phenol degradation).
Titanium (IV) isopropoxide (TTIP, >98%, Acros Organics) was
used as a precursor of titanium dioxide. Poly(ethylene) glycol
(PEG, M.W. 400, Acros Organics) was used as a template direction
reagent in photocatalysts preparation. d-glucuronic acid (>98%)
was obtained from Alfa Aesar. Zeolite (CBV780, Y type) was pur-
chased from Zeolyst International and silica (AEROSIL 200) was
obtained from Evonik.
The catalysts were synthesized by a modified sol–gel method
assisted by ultrasonic irradiation. Zeolite and silica supports were
thermally treated at 450 ◦C during 8 h in static air before the syn-
thesis procedure. A volume of 3.8 mL (12.5 mM, for a TiO2 nominal
content of 15 wt% on each support) titanium (IV) isopropoxide was
dissolved in 80 mL of 2-propanol at room temperature, 6 g of zeo-
lite and 3 g of poly(ethylene glycol) were added and the mixture
was treated with ultrasound (35 kHz, 560 W, Sonorex Digitec-RC,
Bandelin) for 1 h in 5 and 10 min cycles (two times/5 min and five
times/10 min, 2 min ultrasound off between cycles). Then, a mix-
ture of 1.27 mM of d-glucuronic acid and 1.4 g of PEG (total ratio
of PEG:Ti = 8:1) in water/2-propanol (25:47, v/v) was added using
a programmable syringe pump (model NE-1000-E New Era Pump
Systems Inc.) with the speed flow of 0.5 mL/min during ultrasonic
treatment (10 min cycles with 2 min ultrasound off between cycles,
the total time of this step was approx. 90 min). The final mix-
ture was continuously exposed to ultrasound for 30 min, left aging
at room temperature and magnetically stirred for 18 h, filtered
through a G5 funnel, dried at 110 ◦C for 3 h and calcined in static
air at 500 ◦C for 5 h. The resulting photocatalyst was designated as
TiO2/Ze.
Blank experiments were performed in the dark as well as with
illumination and no catalyst, without observable change in the ini-
tial concentration of glucose in both cases.
2.4. Case study #2: liquid-phase photocatalytic degradation of
phenol
All photocatalytic reactions with phenol were performed in the
set-up described above (for the case study #1), however with a few
modifications. The lamp was placed in a borosilicate glass immer-
sion well (300 nm, wavelength at which a 1 mm layer of the glass
absorbs at least 90% the light) in order to avoid autodegradation of
phenol (an important phenol disappearance caused by photolysis
under UV irradiation up to 300 nm was observed). Phenol solutions
(50 ppm) were prepared in Milli-Q water. The reactions were car-
ried out under ambient air. Phenol degradation was measured, after
external standard calibration, by HPLC (Waters HPLC Model 590
pump), equipped with a PDA detector. Separation was performed
on a XBridgeTM C18 5 ␮m 4.6 mm × 150 mm column provided by
Waters. The mobile phase was Milli-Q water/methanol (65:55, v/v)
with 0.1% of CF3COOH at a flow rate of 1 mL/min. The injection
volume was 10 ␮L.
According to the same procedure, silica supported TiO2 was
synthesized, however a different amount of silica was used,
3 g, preserving the proportions among the chemicals. The sam-
ple was labeled as TiO2/SiO2. For comparative purposes, the
results with the commercial photocatalyst Evonik P-25 and unsup-
ported TiO2(US) (synthesized by the same procedure) were also
presented.
2.2. Photocatalytic tests
The controlled experiments, without light or without photocat-
alyst, were performed to confirm that this reaction depends on the
presence of both, light and photocatalyst.
Glucose (ACS, pure p.a.) was purchased from POCH and phe-
nol (99+%) was obtained from Alfa Aesar. Solvents and other

158
J.C. Colmenares, A. Magdziarz / Journal of Molecular Catalysis A: Chemical 366 (2013) 156–162
Table 1
Summary of the structural, textural and optical features concerning characterization of photocatalytic systems.
Catalyst
UV–vis
XRD
N2-isotherms
b
b
b
Egap (eV)
Absorption threshold (nm)
Crystallitea size (nm)
Crystal phasea (%)
SBET (m2/g)
Vp (mL/g)a
rp (Å)
Zeolite Y
SiO2
Evonik-P25
TiO2/Ze
TiO2/SiO2
–
–
3.18
3.27
3.31
–
–
389
379
375
–
–
–
–
784
197
59
676
207
0.52
0.39
0.19
0.47
0.51
27
57
64
29
111
17.6 (A)/24.4 (R)
7.9 (A)
Not detectable
A (88)/R (12)
A (100)
Not detectable
a
A and R denote anatase and rutile, respectively.
Specific surface area (SBET), cumulative pore volume (Vp) and pore mean radius (rp).
b
2.5. Characterization of photocatalysts
phenomenon to the aggregation of TiO2 particles on the surface of
the zeolite. Furthermore, it may suggest that in the case of catalyst
with low TiO2 loading, as in the studied case, the small amount of
well-dispersed TiO2 on the zeolite inhibits the mutual approach of
fine TiO2 particles, hence leading to the formation of crystals of finer
size [12]. Also Zainudin et al. [26] report that the addition of bar-
rier materials such as organic ligands, crystalline zeolites, porous
amorphous polymers among others, can control the growth of the
TiO2 crystallites.
The textural properties of photocatalysts were determined from
N2 adsorption–desorption isotherms at liquid nitrogen tempera-
ture by using a Micromeritics ASAP 2020 instrument. Surface areas
and pore sizes were calculated by the Brunauer–Emmett–Teller
(BET) and the Barret–Joyner–Halenda (BJH) methods, respectively.
Prior to measurements, all samples were degassed at 110 ◦C to
0.1 Pa.
X-ray powder analysis was carried out with
a
XRD
Peaks of anatase are not detectable on the XRD spectra of
TiO2/SiO2 catalyst. This observation can be explained by a high dis-
persion of TiO2 particles on the SiO2 surface or due to the very
small TiO2 clusters located within the silica pores [24]. However,
the presence of TiO2 in this sample can be deduced from the UV–vis
spectrum that is presented in Fig. 1. As one can observe, a character-
istic absorption peak ascribed to TiO2 anatase phase is clearly seen
for the TiO2/SiO2 photocatalyst. Also, a blue-shift (to shorter wave-
lengths) is observed, and it represents a slight increase (Table 1
and Fig. 1) in the band gap energy value of supported TiO2 in com-
parison with the bare TiO2 (commercial Evonik P-25) from 3.18 to
3.3 eV., approximately. This can be explained by the quantum con-
finement effect of titania clusters within the supports texture and
indicates that the size of titania nanoparticles in the zeolite and
silica supports is smaller than that of P-25 [27]. A marked absorp-
tion of visible light in the range of 400–700 nm for TiO2/Ze (Fig. 1)
was detected. Osorio et al. [28] prepared a visible-light responsive
TiO2 by low-frequency ultrasounds (LFUS) irradiation. Based on
electron spin resonance (ESR) measurements, they found evidence
for the formation of oxygen vacancies (Vo) on this material, which
were also found responsible for the visible-light absorption [28].
The Vo surface defects might result from high-speed interparticle
Siemens D5000 diffractometer using Ni-filtered CuK˛ radia-
˚
tion (ꢀ = 1.5406 A) at 40 kV and 30 mA. The diffraction angle 2ꢁ was
scanned at a rate of 2◦ min−1. The average crystallite size of anatase
was determined according to the Scherrer equation (Eq. (1)) using
the full width at half maximum of the peak corresponding to 1 0 1
reflections and taking into account the instrument broadening.
Kꢀ
D =
(1)
ˇ cos ꢁ
where D is the average crystallite size of the catalyst (nm), ꢀ is
the wavelength of the Cu K˛ X-ray radiation (ꢀ = 1.5406 A), K is a
˚
coefficient usually taken as 0.94, ˇ is the full width at half maximum
(FWHM) intensity of the peak observed at 2ꢁ (radian), and ꢁ is the
diffraction angle.
Ultraviolet–visible Diffuse Reflectance spectroscopy was per-
formed using a UV-2501PC Shimadzu spectrophotometer. Band-
gaps values were calculated based on the Kubelka–Munk functions,
f(R), which are proportional to the absorption of radiation, by plot-
ting [f(R)hꢂ)]1/2 against hꢂ. The function f(R) was calculated using
Eq. (2) [23]:
(1 − R)2
f (R) =
.
(2)
2R
3. Results and discussion
3.1. Photocatalysts characterization
It is well-known that the amount of loaded TiO2 in zeolite is
significantly affected by structural parameters of zeolite such as
SiO2/Al2O3 ratio, cation type, and topology [24]. In this study, Y
type zeolite, with the ratio SiO2/Al2O3 = 80 and hydrogen (acidic)
form has been chosen. The Y type zeolite exhibits the FAU (fauja-
site) structure is composed of tetrahedral SiO4 and AlO4 building
blocks, which results in the formation of 3-dimensional pore struc-
ture material.
X-ray powder diffraction was used to determine the crystal
phase of the catalyst particles. The XRD patterns of catalysts are
shown in Table 1. The characteristic peak of anatase [1 0 1] phase
at 2ꢁ = 25.4◦ has been observed for TiO2/Ze catalyst [25]. TiO2 par-
ticle size calculated by means of Scherrer’s equation was 7.9 nm
for the nominal TiO2 loading of 15 wt% (Table 1). Wang et al. [12]
reported that the size of TiO2 particles increases with increas-
ing TiO2 loading (28.9 nm. for 50 wt% of TiO2). He ascribed this
Fig. 1. Diffuse reflectance UV–visible spectra of catalyst samples: TiO2/Ze, TiO2/SiO2
and Evonik P-25.

J.C. Colmenares, A. Magdziarz / Journal of Molecular Catalysis A: Chemical 366 (2013) 156–162
159
collisions and shock waves generated during the ultrasound-
assisted sol–gel methodology used for supported TiO2 photocat-
alyst synthesis. Further investigations are needed to confirm the
presence of Vo defects.
The pore volume and specific surface area of supported catalysts
were determined using nitrogen adsorption–desorption isotherms
and are presented in Table 1. Loading of TiO2 on the zeolite surface
results in the reduction of pore volume and surface area of zeolites
(from 0.52 to 0.47 mL/g and from 784 to 676 m2/g, respectively).
This can indicate that TiO2 is dispersed on the zeolite surfaces and
the pores of zeolite are partially blocked by TiO2. As the pore size
of zeolite was smaller (<2.0 nm) than the size of the TiO2 particles
(approx. 7.9 nm) supported on the zeolite, the particles might not
be deposited inside the pores of zeolite [13]. On the other hand,
the surface area after loading of TiO2 on the silica surface does not
change significantly but the pore radius and volume changes are
more evident. These changes may suggest an aggregation, in some
cases under sonication [28], of TiO2 particles on silica surface and
the development of voids in this aggregation, which might create
an additional pore volume [12].
Fig. 2. Glucose disappearance for TiO2/Ze and TiO2/SiO2 photocatalysts (reaction
conditions: 150 mL of mother solution, Cglucose = 2.8 mM, 150 mg of photocatalyst,
temperature 30 ◦C, pressure 1 bar, solvent composition H2O:ACN: (a) 10:90, v/v (b)
50:50, v/v; (c) 100:0, v/v).
3.2. Photocatalytic test reactions
better distribution of active titania (anatase phase) sites on the
silica surface.
3.2.1. Case study #1: liquid-phase photocatalytic selective
oxidation of glucose
The optimization of water/acetonitrile (ACN) reaction medium
has also been investigated and the results are presented in Table 2.
The presence of acetonitrile has been reported to improve selec-
tive photo-oxidation reactions [29]. It is clearly seen that the higher
amount of water gives lower conversion (except for P-25 where the
situation is the opposite). Herein, better photodegradation proper-
ties of silica-supported titania material are confirmed, as if in each
H2O/ACN volume ratio glucose conversion on silica-supported tita-
nia is the highest (Fig. 2). The influence of H2O/ACN ratio on the
selectivity has also been checked. No selectivity to carboxylic acids
was observed for 100% H2O as solvent (Tab. 2). Interestingly, the
selectivity to GUA and GA for 50%H2O/50%ACN solvent composition
was improved (arabitol was not detected) under these conditions
(64.8% and 68.1% for TiO2/SiO2 and TiO2/Ze, respectively, Table 2).
In this H2O/ACN volume ratio both photocatalysts are slightly more
selective to gluconic acid (results not shown).
Controlling the photocatalytic degradation of target substrates
in a selective way is a challenging issue since the reactivity of the
hydroxyl radical is difficult to control [30]. Interesting properties
of zeolites are negative charges (in our case balanced by counter
protons present in zeolites pores) carried by the aluminosilicate
framework, which results from the substitution of Al3+ into the
silica structure [31]. The electrostatic repulsion of anionic prod-
ucts (carboxylic acids) by the negatively charged zeolite framework
facilitated the selective photocatalytic oxidation of glucose.
While the point of zero charge pzc (pzc describes the condition
when electrical charge density on a surface is zero) of Evonik P-25 is
approx. 6.5, those of TiO2/Ze could be significantly shifted to lower
pH (approx. 2) [24] because of the negative charges of the zeolite
framework. As a result, the surface of TiO2/Ze carries dominant neg-
ative charges at pH >2 (pH of the glucose solution before reaction is
6.3 0.2, pKa (GA) = 3.35 and pKa (GUA) = 2.99) under these condi-
tions carboxylic acids products should be repulsed preventing their
total photo-oxidation. As it can be seen (Tab. 2), TiO2/Ze (better
than TiO2/SiO2, TiO2(US) and much better than P-25) exhibits the
charge-selective adsorption characteristics, which might affect the
photocatalytic selectivity. Not only does the repulsion of carboxylic
acids from TiO2/Ze surface (also observed for TiO2/SiO2 and unsup-
ported TiO2(US)) avoid their degradation but also their diffusion
into the protective environment created by ACN (pKa = 25, as a weak
base could stabilize the carboxylic acids by solvation and suppress
their further oxidation), which demonstrates the important role
of ACN especially for the solvent composition of 50%H2O/50%ACN.
In the photocatalytic oxidation of glucose three organic com-
pounds have been obtained in the liquid phase, namely: glucaric
acid (GUA), gluconic acid (GA) and arabitol (AOH). Carboxylic acids
are very important building blocks for pharmaceutical, food, per-
fume and fuel industries. The arabitol component in the organic
phase is also an important platform molecule [4,6] and it was
detected only when 10%H2O/90%ACN solvent composition was
used. Besides these liquid phase products, CO2 and traces of hydro-
carbons (methane, ethane, acetaldehyde) in gas phase have been
detected. Scheme 1 presents a chart of the idea of the production
of GUA and GA through a photocatalytic selective oxidation of glu-
cose in liquid phase. As it can be observed in Scheme 1, all steps
(except for step nr. 1) are very well known technologies of distil-
lation (steps 3, 4, 7 and 8), filtration (step 2), drying (step 2) and
reactive extraction of pure carboxylic acids (step 5), which guar-
antee good flexibility and effectiveness of producing high value
carboxylicacids in a low-cost and environmentallyfriendly process.
Table 2 compares the selectivity results obtained for the studied
photocatalysts for the first 10 min of photoreaction (this illumina-
tion time was chosen as it showed the best selectivity performance
and difference among catalysts). As it is seen, zeolite-supported
titania is the most selective in this time interval (total selectivity
for GUA + GA is 68.1% with 15.5% glucose conversion). Moreover, the
most popular photocatalyst P-25 showed several times lower car-
boxylic acids selectivity and higher glucose conversions (Table 2)
than the catalysts prepared by the ultrasound-assisted sol–gel
methodology. Unsupported TiO2(US) was giving lower glucose con-
version and comparable selectivity to GUA and GA; at this point it is
important to highlight the benefits of supporting TiO2 (easy filtra-
tion) and the fact that for our catalysts, TiO2 represents a nominal
value of 15 wt% of the whole catalyst used in the reaction (150 mg
of unsupported TiO2 was applied), which makes supported photo-
catalysts more active and selective per gram of titania used. In
the case of silica-supported titania the selectivity decreases more
visibly in each time interval (results not shown), but the glucose
conversion rate increases more rapidly (Fig. 2) in each solvent
composition. Hence, the attribute of revealing better glucose degra-
dation properties can be ascribed to titania supported on SiO2,
while zeolite-supported titania showed better selectivity. Faster
disappearance (Fig. 2) of glucose in the presence of silica-supported
titania photocatalysts can be assigned to a larger amount or

160
J.C. Colmenares, A. Magdziarz / Journal of Molecular Catalysis A: Chemical 366 (2013) 156–162
Scheme 1. Chart of the concept of photocatalytic selective valorization of sugars (glucose) in liquid phase (H2O/ACN).
Shiraishi et al. [32] and Morishita et al. [33] demonstrated the
positive influence of acetonitrile on formation of epoxides in the
selective photo-oxidation of olefins. One more thing to highlight is
the fact that the more acidic character of zeolite support in com-
parison to SiO2, unsupported TiO2(US) and Evonik P-25 could help
in obtaining higher selectivity to carboxylic acids as it was found
by An et al. [34]. They observed that the acidity of their tested cat-
alysts (polyoxometalates-supported gold nanoparticles) not only
favored the conversion of cellobiose in the presence of oxygen
but also resulted in higher selectivity to gluconic acid by facilitat-
ing desorption and inhibiting its further degradation in aqueous
phase.
TiO2/Ze visible light absorption (Fig. 1) could be ascribed to oxy-
gen vacancies (Vo) which might additionally induce the creation
of surface photoactive centers in our hybrid material for a better
photoselectivity in the formation of carboxylic acids from glucose
oxidation (approx. 19% of the spectrum of the lamp used in pho-
tocatalytic tests was between 400 and 440 nm light wavelength).
Further research will be focused on the demonstration whether or
not Vo defects are present on TiO2/Ze surface and also on the effect
of such defects on the selective photocatalytic oxidation of glucose
through techniques such as electron spin resonance spectroscopy
(ESR).
To sum up, all previous observations allowed us to con-
clude that the properties of TiO2, the nature of titania support
and the reaction conditions (solvent composition, illumina-
tion time) play a key role in determining the photocatalytic
behavior, particularly in determining the selectivity for GA and
GUA.
3.2.2. Case study #2: liquid-phase photocatalytic degradation of
phenol
Phenol disappearance, as a model reaction [35], was employed
to investigate photocatalytic degradation properties of synthesized
samples under UV irradiation. To verify the photocatalytic abili-
ties of the samples, the experiments were conducted during 4 h of
illumination. No appreciable phenol degradation was found in the
absence of UV irradiation or catalyst. The concentration of phenol
by photolysis decreased only to 11% after 4 h of illumination.
Table 2
Effect of the solvent composition (water and acetonitrile ACN) on the conversion of glucose and selectivity to carboxylic acids for: TiO2/SiO2, TiO2/Zeolite, TiO2(US) and Evonik
P-25 photocatalysts (reaction conditions: Cglucose = 2.8 mM, 150 mL of mother solution, 150 mg of photocatalyst, temperature 30 ◦C, pressure 1 bar, 10 min illumination time).
Photocatalyst
Solvent composition
10%H2O/90%ACN
Conversion (%)
50%H2O/50%ACN
Conversion (%)
100%H2O
ꢃSelectivity (%)
ꢃSelectivity (%)
Conversion (%)
ꢃSelectivity (%)
TiO2/SiO2
TiO2/Zeolite
TiO2(US)
58.1
49.2
28.8
60.4
28.1
29.5
27.3
12.1
14.0
15.5
11.0
67.3
64.8
68.1
71.3
8.5
9.5
13.7
41.2
78.8
0.0
0.0
0.0
0.0
Evonik P-25

J.C. Colmenares, A. Magdziarz / Journal of Molecular Catalysis A: Chemical 366 (2013) 156–162
161
4. Conclusions
Among the photocatalytic systems tested, the best perform-
ances in terms of organic products selectivity (GA + GUA) in glucose
photocatalytic oxidation were especially remarkable in the case
of the catalysts synthesized by the ultrasound-modified sol–gel
method. TiO2 supported on zeolite (TiO2/Ze) was the most selective
toward glucaric/gluconic acid products (68.1% selectivity for a sol-
vent composition of 50%H2O/50%ACN and 10 min of illumination).
It was found that reactions conditions, especially solvent composi-
tion and short illumination times, also have considerable effect on
the activity/selectivity of tested photocatalysts.
As far as phenol degradation is concerned, TiO2/SiO2 is the
best option as photocatalyst (ca. 35% of phenol disappearance for
240 min of illumination) because it is promoting mineralization of
phenol to CO2 and H2O with insignificant production (traces) of
harmful and dangerous water contaminants (e.g. hydroquinone,
catechol).
Studies on the selective photocatalytic conversion of glucose
may provide important clues for the rational design of efficient
photocatalysts for cellulose transformations and also for more
effective photocatalysts in the fight against water pollutants.
Fig. 3. Photocatalytic degradation of phenol in the presence of investigated cata-
lysts: TiO2/Ze, TiO2/SiO2, and Evonik P-25 (reaction conditions: 150 mL of mother
solution, 150 mg of photocatalyst, Cphenol = 50 ppm prepared in Mili-Q water, tem-
perature 30 ◦C, reaction pressure 1 bar).
Although upon approx. 100 min of illumination (Fig. 3) TiO2/Ze
was more active than TiO2/SiO2, after 240 min of illumination the
same phenol degradation rate is achieved for both photocatalysts
(ca. 35% of degradation). It can be seen in Fig. 3 that the most
active photocatalyst is Evonik P-25 with 69.3% of phenol degra-
dation after 240 min of irradiation. Unfortunately, in the aqueous
phase P-25 left, apart from phenol, the highest quantity of phenol-
derived harmful compounds of negative effect on water quality
(e.g. hydroquinone, catechol) among the investigated photocata-
lysts (TiO2/SiO2 has shown negligible amount of these compounds,
Fig. 4). HPLC–MS analyses (Fig. 4), upon 240 min of illumination
time, showed intermediate by-products of partial photo-oxidation
obtained in photocatalytic degradation of phenol evidencing a
more selective mineralization to CO2 and H2O in the presence of
the photocatalysts prepared by the modified ultrasound-assisted
sol–gel procedure (TiO2/SiO2 > TiO2/Ze > >P-25, sequence of direct
selectivity to complete mineralization).
Acknowledgments
This research was supported by
a Marie Curie Interna-
tional Reintegration Grant within the 7th European Community
Framework Programme. This scientific work was financed from
the 2012–2014 science financial resources, granted for the
international co-financed project implementation (Project Nr.
473/7.PR/2012, Ministry of Science and Higher Education of
Poland). The authors express gratitude to Dr K. Noworyta for his
help with the UV–vis Diffuse Reflectance measurements.
References
[1] J.C. Serrano-Ruiz, R. Luque, A. Sepulveda-Escribano, Chem. Soc. Rev. 40 (2011)
5266–5281.
[2] G. Centi, P. Lanzafame, S. Perathoner, Catal. Today 167 (2011) 14–30.
[3] M.V. Shankar, S. Anandan, N. Venkatachalam, B. Arabindoo, V. Murugesan,
Chemosphere 63 (2006) 1014–1021.
[4] J.C. Colmenares, R. Luque, J.M. Campelo, F. Colmenares, Z. Karpinski, A.A.
Romero, Materials 2 (2009) 2228–2258.
[5] J.C. Colmenares, A. Magdziarz, A. Bielejewska, Bioresour. Technol. 102 (2011)
11254–11257.
[6] (a) T. Werpy, G. Pedersen, Top Value Chemicals from Biomass, 1, US Department
of Energy report, 2004;
(b) J.J. Bozell, G.R. Petersen, Green Chem. 12 (2010) 539–554.
[7] M.A. Henderson, Surf. Sci. Rep. 66 (2011) 185–297.
[8] X. Chen, S.S. Mao, Chem. Rev. 107 (2007) 2891–2959.
[9] J.C. Yu, L. Zhang, J. Yu, Chem. Mater. 14 (2002) 4647–4653.
[10] J.C. Colmenares, M.A. Aramendía, A. Marinas, J.M. Marinas, F.J. Urbano, Appl.
Catal., A 306 (2006) 120–127.
[11] S.Y. Kim, T.S. Chang, C.H. Shin, Catal. Lett. 118 (2007) 224–230.
[12] Ch.-C. Wang, C.-K. Lee, M.-D. Lyu, L.-Ch. Juang, Dyes Pigments 76 (2008)
817–824.
[13] F. Sunada, A. Heller, Environ. Sci. Technol. 32 (1998) 282–286.
[14] Y. Chen, K. Wang, L. Lou, J. Photochem. Photobiol. A 163 (2004) 281–287.
[15] Y. Xu, W. Zheng, W. Liu, J. Photochem. Photobiol. A 122 (1999) 57–60.
[16] J. Hermann, J. Matos, J. Disdier, C. Guillard, J. Laine, S. Malato, J. Blanco, Catal.
Today 54 (1999) 255–265.
[17] H. Yoneyama, T. Torimoto, Catal. Today 58 (2000) 133–140.
[18] C. Ooka, H. Yoshida, K. Suzuki, T. Hattori, Microporous Mesoporous Mater. 67
(2004) 143–150.
[19] Z. Sun, Y. Chen, Q. Ke, Y. Yang, J. Yuan, J. Photochem. Photobiol. A 149 (2002)
169–174.
[20] M.G. Kang, H.S. Park, K.J. Kim, J. Photochem. Photobiol. A 149 (2002) 175–181.
[21] A. Bhattacharyya, S. Kawi, M.B. Ray, Catal. Today 98 (2004) 431–439.
[22] S. Sankararaman, K.B. Yoon, T. Yabe, J.K. Kochi, J. Am. Chem. Soc. 113 (1991)
1419–1421.
[23] S. Sakthivel, H. Kisch, Angew. Chem. Int. Ed. 42 (2003) 4908–4911.
[24] G. Zhang, W. Choi, S.H. Kim, S.B. Hong, J. Hazard. Mater. 188 (2011) 198–205.
[25] Y.H. Hsien, C.F. Chang, Y.H. Chen, S. Cheng, Appl. Catal., B 31 (2001) 241–249.
[26] N.F. Zainudin, A.Z. Abdullah, A.R. Mohamed, J. Hazard. Mater. 174 (2010)
299–306.
Fig. 4. HPLC–MS chromatographs for 240 min of photocatalytic degradation of phe-
nol.

162
J.C. Colmenares, A. Magdziarz / Journal of Molecular Catalysis A: Chemical 366 (2013) 156–162
[27] M.V. Rama-Krishna, R.A. Friesner, J. Chem. Phys. 95 (1991) 8309–8322.
[28] P.A. Osorio-Vargas, C. Pulgarin, A. Sienkiewicz, L.R. Pizzio, M.N. Blanco, R.A.
Torres-Palma, C. Pétrier, J.A. Rengifo-Herrera, Ultrason. Sonochem. 19 (2012)
383–386.
[29] Y. Shiraishi, T. Hirai, J. Photochem. Photobiol. C 9 (2008) 157–170.
[30] S.H. Zhan, D.R. Chen, X.L. Jiao, Y. Song, Chem. Commun. (2007) 2043–2045.
[31] M.N. Chretien, Pure Appl. Chem. 79 (2007) 1–20.
[32] Y. Shiraishi, M. Morishita, T. Hirai, Chem. Commun. (2005) 5977–5979.
[33] M. Morishita, Y. Shiraishi, T. Hirai, J. Phys. Chem. B 110 (2006) 17898–17905.
[34] D. An, A. Ye, W. Deng, Q. Zhang, Y. Wang, Chem. Eur. J. 18 (2012) 2938–2947.
[35] V. Gomez, A.M. Balu, J.C. Serrano-Ruiz, S. Irusta, D.D. Dionysiou, R. Luque, J.
Santamaria, Appl. Catal., A 441–442 (2012) 47–53.