Highly efficient activated carbon loaded TiO2 for photo defluoridation of pentafluorobenzoic acid

Article abstract of DOI:10.1016/j.molcata.2009.10.029

The activated carbon loaded TiO₂-P25 catalysts were prepared and characterized by diffuse reflectance spectra (DRS), FT-IR, scanning electron micrograph (SEM), X-ray diffraction (XRD) and BET surface area analysis. The photocatalytic efficiency

Full text of DOI:10.1016/j.molcata.2009.10.029

Journal of Molecular Catalysis A: Chemical 317 (2010) 89–96
Contents lists available at ScienceDirect
Journal of Molecular Catalysis A: Chemical
journal homepage: www.elsevier.com/locate/molcata
Highly efficient activated carbon loaded TiO2 for photo defluoridation of
pentafluorobenzoic acid
L. Ravichandran, K. Selvam, M. Swaminathan∗
Department of Chemistry, Annamalai University, Annamalai Nagar 608002, India
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 17 July 2009
Received in revised form 23 October 2009
Accepted 27 October 2009
Available online 5 November 2009
The activated carbon loaded TiO2-P25 catalysts were prepared and characterized by diffuse reflectance
spectra (DRS), FT-IR, scanning electron micrograph (SEM), X-ray diffraction (XRD) and BET surface area
analysis. The photocatalytic efficiency of activated carbon loaded TiO2-P25 (AC-TiO2-P25) was evaluated
by UV assisted photodefluoridation of pentafluorobenzoic acid (PFBA) in aqueous medium. The catalyst
exhibited higher photo defluoridation efficiency of PFBA than that of bare TiO2-P25. The various experi-
mental parameters like concentration of PFBA, addition of oxidants, amount of catalyst and solution pH
for efficient defluoridation are reported. The higher efficiency of AC-TiO2-P25 is due to synergy effect of
activated carbon.
Keywords:
AC-TiO2
PFBA
Defluoridation
Synergistic effect
Photocatalysis
© 2009 Elsevier B.V. All rights reserved.
1. Introduction
pounds are found to have adverse impact on the health of animals
and human beings, the removal of these compounds from environ-
Though titanium dioxide had been widely used for toxic chem-
ical degradation, there are some practical difficulties such as either
filtration of fine TiO2 or fixation of catalyst particles and efficient
utilization of UV/solar light. For these reasons many researchers
have been working to increase the efficiencies of these processes
by modification of its surface. The design of TiO2 photocatalyst,
anchored or embedded onto support materials with large surface
areas that could condense diluted substances would be of great
significance, not only to avoid the filtration of small photocata-
lyst particles, but also to obtain higher efficiency [1,2]. Several
authors proposed a number of modifications of the photocatalysts
by loading them on supporting materials like alumina, zeolite, clay,
activated carbon, etc., and by doping with metal ions [3]. Acti-
vated carbon has a large specific surface area and a well developed
porous structure, resulting in an attractive force toward organic
molecules [4–8]. This increases the adsorption capacity of the
photocatalyst.
mental components becomes essential [11,13,14]. Photocatalytic
mineralisation of these compounds produces CO2, H2O and fluo-
ride ions. The fluoride ions can be used for the production of CaF2,
which can be used as a raw material for the manufacture of many
fluorinated compounds.
Earlier we reported the defluoridation of PFBA using TiO2-P25
[15], ZnO [16] and Fenton’s reagent [17]. Though TiO2-P25 is found
to be more efficient than ZnO in the defluoridation of PFBA, its
efficiency can be further increased by activated carbon loading.
Activated carbon support has been reported to give more promising
results, which are attributed to the synergistic effect between the
AC and the photocatalyst [18]. In the present paper, we report the
preparation and characterization of activated carbon loaded TiO2-
P25 catalyst and its photocatalytic activity on the defluoridation of
PFBA.
2. Experimental
Organofluoro compounds are now being widely used in phar-
maceutical, agrochemical, surfactant and polymer industries due
to their thermal stability and enhanced lipophilicity [9,10]. These
aromatic fluorinated compounds are quite stable and they have
no known natural decomposition [11,12]. Some of them have been
detected in environmental waters and in animals. Since these com-
2.1. Materials
Pentafluorobenzoic acid (PFBA) (99% purity) from Aldrich Chem-
ical Company was used without further purification. A gift sample
of TiO2-P25 was obtained from Degussa (Germany). It has the par-
ticle size of 30 nm and BET specific surface area of 55 m2/g. AnalaR
grade reagents, H2O2 (30%, w/w), (NH4)2S2O8, KBrO3 (E. Merck),
Na2CO3, NaCl and FeSO4·7H2O were used as received. The activated
carbon from SD Fine Chemicals was used as such. The double-
∗
Corresponding author. Tel.: +91 4144 220572; fax: +91 4144 238080.
E-mail address: chemres50@gmail.com (M. Swaminathan).
1381-1169/$ – see front matter © 2009 Elsevier B.V. All rights reserved.
doi:10.1016/j.molcata.2009.10.029

90
L. Ravichandran et al. / Journal of Molecular Catalysis A: Chemical 317 (2010) 89–96
2.4. Defluoridation analysis
Fluoride ion concentration was determined using Orion expand-
able ion analyzer model (EA940). The analysis was done using
fluoride ion selective electrode in conjugation with Ag/AgCl ref-
erence electrode. The instrument was calibrated using standard
sodium fluoride solution of various concentrations between 10
and 250 ppm of fluoride ion. The sample 5 mL from photochemical
reactor was taken into 50 mL of polyethylene beaker at a regu-
lar interval of 30 min. Five milliliters water and 20 mL total ionic
strength adjustable buffer (TISAB) were added to the solution. This
solution was stirred for 5 min and fluoride ion concentration was
measured. The pH of the solutions was measured using Metrohm
744 pH meter.
2.5. Adsorption experiments
The adsorption experiments have been performed with the
PFBA solutions at different concentrations and at different pH. Fifty
milliliters aqueous PFBA solution and the catalyst were mixed for
3 h at 25 ◦C using mechanical shaker. In order to avoid the pho-
toreaction of catalyst, the samples were kept in dark for the entire
period of the experiment. After adsorption, the solutions were cen-
trifuged to separate the catalyst. The concentration of the substrate
in solution was determined from its absorbances at 270 nm.
Fig. 1. Schematic diagram of photoreactor.
distilled water was used to prepare experimental solutions. The
pH of the solutions was adjusted using H2SO4 or NaOH.
2.6. Equipments used
Scanning electron microscopic (SEM) analysis was performed on
platinum coated samples using a Jeol apparatus model JSM-5610
LV, equipped with INCA EDX probe for the energy dispersive X-ray
microanalysis (EDX). A Varian Cary 5E UV/VIS-NIR spectropho-
tometer equipped with an integrated sphere was used to record
the diffuse reflectance spectra (DRS). The baseline correction was
performed using a calibrated reference sample of barium sulfate.
The reflectance spectra of the AC-TiO2-P25 catalysts were analyzed
under ambient conditions in the wavelength range of 200–800 nm.
The specific surface area of the samples was determined through
nitrogen adsorption at 77 K on the basis of BET equation using a
Sorptomatic 1990 instrument.
Powder X-ray diffraction patterns of ZnO, AC and AC-ZnO
catalysts were obtained using a Philips PANalytical X’per PRO
diffractometer equipped with a Cu tube for generating a Cu K∝ radi-
ation (wavelength 1.5406 Å) at 40 kV, 25 mA. The particles were
spread on a glass slide specimen holder and the scattered inten-
sity was measured between 20◦ and 85◦ at a scanning rate of
2ꢀ = 1.2◦/min at 25 ◦C temperature. The digital output of the pro-
portional X-ray detector and the gonimeter angle measurements
were sent to an online microcomputer for storing and analysis.
Peak positions were compared with the standard files to identify
the crystalline phases.
2.2. Preparation of AC-TiO2-P25 catalysts
Catalysts with activated carbon were obtained by mixing TiO2-
P25 and activated carbon at different proportions in water and this
aqueous suspension was continuously stirred for 3 h. After this, the
mixture was filtered and dried. The catalysts have been denoted
as xAC-TiO2-P25, where ‘x’ is the weight percentage of activated
carbon (w/w) used in the preparation of the catalyst.
Catalysts with the lower AC contents (4%AC-TiO2-P25 and
8%AC-TiO2-P25) acquire a clear grey color. However, when the AC
concentration in the catalyst is increased above 8%, they become
darker. The catalyst with higher AC content (12AC-TiO2-P25) is
practically black.
2.3. Photoreactor
The photocatalytic defluoridation has been carried out with
a Heber immersion type photoreactor model HIPR-LP 6/8/116
whose schematic diagram is shown in Fig. 1. This model con-
sisted of a double walled immersion well made of a quartz reactor
of 175 cm3 capacity. A small inlet tube is extended down the
annular space to ensure flow of the coolant from bottom of
well upwards to outlet. In the center of cylindrical reactor, the
lamp used for light source was placed inside the quartz tube.
For the experiments, lamps with different wavelength emissions
were used. One was 16 W low pressure mercury with emission
mainly on the mercury resonance line at 253.7 nm. The other
was 8 W medium pressure mercury lamp with broadband emis-
sion predominantly at 365 nm. For both the lamps photon flux
of the light source was determined by ferrioxalate actinome-
try [19] and the values of 16 W low pressure Hg lamp and 8 W
For FT-IR experiments, FT-IR spectrophotometer Thermo Nick-
olet was used. The samples were incorporated in KBr pellets for the
measurements. Water reference spectrum was always subtracted
from every spectrum. UV spectral measurements were done using
Hitachi-U-2001 spectrophotometer. Total organic carbon was mea-
Table 1
Surface area and size of AC-TiO2-P25 catalysts.
Sample
BET specific surface area (m2/g)
Crystallite size (nm)
medium pressure Hg lamp are I254 nm = 2.54 × 10−5 Einstein L−1 s−1
,
AC
735.60
55.00
66.54
85.60
92.96
108.79
–
I365 nm = 2.08 × 10−6 Einstein L−1 s−1 respectively. The reaction ves-
sel had an arm at the top for gas purging. The temperature of the
experimental solution was maintained at 25 1 ◦C by circulating
water during the experiments. The above setup was placed in the
magnetic stirrer for complete mixing of the catalyst.
Bare TiO2-P25
4AC-TiO2-P25
8AC-TiO2-P25
10AC-TiO2-P25
12AC-TiO2-P25
21.5
27.3
30.9
74.8
74.9

L. Ravichandran et al. / Journal of Molecular Catalysis A: Chemical 317 (2010) 89–96
91
sured with Micro C Model 1997 of Analytica Gena, Germany. HPLC
analysis was performed using Shimadzu Prominence Analytical LC
model with C-18 Column and UV-detector.
3. Results and discussion
3.1. Characterization
3.1.1. Surface area measurements
The surface area of the catalyst is the most important factor
influencing the catalytic activity. Table 1 lists the BET surface area
of the TiO2-P25 and AC-TiO2-P25 catalysts.
The surface area of the catalyst increases with increase in AC
content. A similar trend has been reported in AC-TiO2 anatase [20].
Higher specific surface area may be of benefit to their high pho-
tocatalytic activity due to enhanced absorption of photons and
adsorption of pollutants.
3.1.2. Diffuse reflectance spectroscopic analysis
Fig. 2 shows the UV diffuse reflectance spectra of the pre-
pared AC-TiO2-P25 catalysts along with the bare TiO2-P25, catalyst.
TiO2-P25 shows a clear absorption edge at around 380 nm with
no absorption in visible light region. In AC-TiO2-P25 samples,
absorption of visible light is observed. The visible light absorption
increases up to 10AC-TiO2-P25 and then decreases slightly. It is
also noticeable that there is no correlation between the AC con-
tent and UV spectrum change. Enhancement of the same order has
been observed when the semiconductor has been immobilized on
a polymer [21]. It has been reported that the increment of surfa-
cial electric charge of oxides such as TiO2 and WO3 by the addition
of AC gives an absorption increment in the visible and IR region
and a color change from white to blue [22]. Similarly catalyst TiO2-
P25 also acquires a color change from white to grey at the lower
concentration of AC. This color change may be due to the incre-
ment of the electric charge in TiO2-P25 by the addition of AC. This
is produced by the acid base character increment of the surfacial
hydroxyl groups in the presence of AC. The fact that there is no cor-
relation between the AC content of catalyst and the UV spectrum
change reveals that a change of properties in some catalysts is not
only determined by their AC content [20].
Fig. 3. FT-IR spectra from the catalysts: (a) bare TiO2-P25, (b) 4AC-TiO2-P25, (c)
10AC-TiO2-P25, and (d) 12AC-TiO2-P25.
IR studies in the region between 4000 and 2500 cm−1 determine
possible changes in the distribution of hydroxyl groups that could
confirm the above mentioned acid–base character modification of
the catalysts. TiO2-P25 was treated with water as AC-TiO2 before
taking the FT-IR spectrum. In the spectrum of the bare TiO2-P25
(Fig. 3a) hydroxyl group vibration band is centered at 3412 cm−1. In
AC-TiO2-P25 catalysts a progressive ␯ vibration shift from 3412
OH
to 3424 cm−1 with the increase of AC content of the catalysts is
observed (Fig. 3b–d). These vibration shifts have been attributed
to an increment of the positive charge of these OH groups [23]
adsorbed on the surface of TiO2-P25, confirming the changes of the
acid–base character of the hydroxyl groups in these catalysts.
Furthermore, in the 800–1600 cm−1 region two new bands are
obtained. One band corresponding to C–O–C structure is observed
at 870–950 cm−1 for AC-TiO2-P25. The existence of C–O–C struc-
ture was reported in investigations of basic activated carbon and
also in pyrones which are found to have strong basic sites [24]. The
AC-TiO2-P25, samples show an additional band at the wavelength
of 1400–1600 cm−1 which can be attributed to C C skeleton vibra-
tion of aromatic ring. The observation of C–O–C and C C structure
reveals the presence of some basic character in AC-TiO2-P25.
3.1.3. FT-IR study
In order to understand the surface nature, FT-IR spectra were
recorded for the TiO2-P25 and AC-TiO2-P25 samples (Fig. 3). FT-
3.1.4. SEM studies
The texture and morphology the catalyst are very important
parameters and might influence the photocatalytic activity. To
understand the differences between the TiO2-P25, activated car-
bon and AC-TiO2-P25, SEM images have been analyzed (Fig. 4). The
SEM image of AC in Fig. 4a depicts that the crystallinity is more in
activated carbon and the surface of the carbon is irregular and not
spherical as expected. The SEM image of TiO2-P25 (Fig. 4b) shows
that presence of a porous, sponge like network. It is found that in
10AC-TiO2-P25 (Fig. 4e) the AC particles are homogeneously cov-
ered by TiO2-P25 particles, which agrees with the previous results
[20]. But in 4AC-TiO2-P25 and 8AC-TiO2-P25, there is no complete
coverage of AC particles by the catalyst (Fig. 4c and d). At higher
AC content i.e., 12AC-TiO2-P25 (Fig. 4f), TiO2-P25 particles form
conglomerates and do not get homogeneously dispersed on AC
particles.
Fig. 2. Diffuse reflectance spectra from the catalysts: (a) bare TiO2-P25, (b) 4AC-
TiO2-P25, (c) 8AC-TiO2-P25, (d) 10AC-TiO2-P25, and (e) 12AC-TiO2-P25.

92
L. Ravichandran et al. / Journal of Molecular Catalysis A: Chemical 317 (2010) 89–96
Fig. 4. Scanning electron micrograph images. (a) AC, (b) bare TiO2-P25, (c) 4AC-TiO2-P25, (d) 8AC-TiO2-P25, (e) 10AC-TiO2-P25, and (f) 12AC-TiO2-P25 catalysts.
3.1.5. XRD analysis
Based on the above results, we have chosen 10AC-TiO2-P25 for
the photocatalytic defluoridation of PFBA.
Fig. 5 shows X-ray diffraction patterns for bare TiO2-P25, AC and
some AC-TiO2-P25 crystallites with a different content of AC. The
X-ray diffraction patterns of AC-TiO2-P25 almost coincide with that
of pure TiO2-P25 and show no diffraction pattern due to AC, thus
suggesting that the AC particles are uniformly distributed on the
TiO2-P25 surface and there is no change in the crystal structure by
AC.
3.2. Photo defluoridation
Initially the experiments were carried out to test the pho-
todefluoridation of PFBA using 254 nm light with AC-TiO2-P25
catalysts and the results are shown in Fig. 6. Complete defluori-
dation (180 mg L−1) was observed in 60 min with 10AC-TiO2-P25
in comparison to 90 min with bare TiO2-P25. Among the AC-TiO2-
P25 catalysts 10AC-TiO2-P25 was found to be most efficient. The
increased AC content of the catalyst from 4 to 10% increases the
fluoride removal. Further increase in AC content to 12%, decreases
the fluoride removal efficiency. The enhancement of removal rate
is mainly due to the increase in the amount of AC weight which
increases the number of PFBA molecules adsorbed. Above 10%
of AC in the catalyst, the decrease in removal rate is due to the
enhancement of light reflectance by the AC and decrease in light
penetration. Thus the optimum amount of AC in the catalyst for
efficient photodefluoridation of PFBA is 10%.
The defluoridation with 10AC-TiO2-P25 using 254 and 365 nm
light was studied to find out the effect of light wavelength and it was
compared with the defluoridation with bare TiO2-P25 (Fig. 7). The
defluoridation is more effective in AC-TiO2-P25 than bare TiO2-P25
with both 254 and 365 nm light. The 254 nm (UV-C) light is more
efficient than 365 nm (UV-A) light. The morphologies of TiO2-P25
may be changed with the AC content and the morphology is impor-
The crystal size was determined from the diffraction peak broad-
ening employing the following Eq. (1):
Kꢁ
D =
(1)
(ˇc − ˇs) cos ꢀ
where D is the crystal size of the catalyst, ꢁ is the wavelength of
X-ray, ˇc and ˇs are the FWHM of the catalyst and the standard
respectively, the value of the coefficient K is 0.89 and ꢀ is the diffrac-
tion angle. The sizes of particles are given in Table 1. The crystal size
of bare TiO2-P25 increases from 21.5 nm by the addition of AC.
The results have shown that in the catalysts with AC content
lower or equal to 10% in weight TiO2-P25 particles do not form
conglomerates, but get adsorbed on the AC particles surface giv-
ing a homogeneous distribution. This regular TiO2-P25 distribution
on AC particles forming layer may be the cause for the charge
transference between AC and TiO2-P25 resulting change in UV–vis
absorption spectrum which enhances the catalyst ability to absorb
more visible light.

L. Ravichandran et al. / Journal of Molecular Catalysis A: Chemical 317 (2010) 89–96
93
Fig. 7. Comparison of efficiencies of 10AC-TiO2-P25 in UV-254 and UV-365 nm
light [PFBA] = 400 ppm, 10AC-TiO2-P25 = 1 g/L, pH = 7, airflow rate = 8.1 mL s−1
,
Fig. 5. XRD pattern: (a) bare TiO2-P25, (b) 4AC-TiO2-P25, (c) 8AC-TiO2-P25, (d)
10AC-TiO2-P25, and (e) 12AC-TiO2-P25.
I254 nm = 2.54 × 10−5 Einstein L−1 s−1, I365 nm = 2.08 × 10−6 Einstein L−1 s−1
.
tant for the photocatalytic activity. To confirm this, the reaction was
carried out by the direct addition of activated carbon and TiO2-
P25 to the reactor with the same concentration of the prepared
10AC-TiO2-P25 sample. The defluoridation is more effective in pre-
pared AC-TiO2-P25 than with direct addition of AC and TiO2-P25
separately.
Since the defluoridation was effective with 10AC-TiO2-P25
using 254 nm light, the effects of pH and oxidants had been carried
out using this catalyst and 254 nm light to find out the optimum
conditions.
The pH of the solution is adjusted before irradiation and it is not
controlled during the course of the reaction. From the experimental
results the photocatalytic defluoridation efficiency of AC-TiO2-P25
is observed to be higher at neutral pH 7. To find out the reason for
this, adsorption experiments were carried out at different pH. The
percentages of PFBA remaining after adsorption at different pH are
shown in Fig. 9. Maximum adsorption was observed at pH 7. Hence
the higher defluoridation efficiency is due to the strong adsorption
of PFBA on the catalyst surface.
3.2.2. Effect of substrate concentration
The effect of initial PFBA concentration on the defluoridation
rate was investigated by varying the initial concentration from 100
to 600 mg L−1 at constant pH and catalyst loading. The results are
shown in Fig. 10. From the results, it is clear that there is no sig-
nificant change in defluoridation up to 200 ppm while decreasing
trend is observed at higher concentrations.
Complete defluoridation was observed in 40 min on irradia-
tion of 100 and 200 mg L−1 concentration with 10AC-TiO2-P25
in comparison to 90 min with 600 mg L−1. The increase of PFBA
concentration decreases the defluoridation rate. The rate of deflu-
3.2.1. Effect of solution pH
The photocatalytic defluoridation efficiency is affected by the
surface charge property of the photocatalyst AC-TiO2-P25, charge
of the molecule, adsorption of molecule onto photocatalyst sur-
face and hydroxyl radical concentration. As these properties are
pH dependent, it plays an important role in the degradation of pol-
lutants. The effect of pH from 3 to 11 on the defluoridation is shown
in Fig. 8.
Fig. 8. Effect of pH on photodefluoridation of PFBA with 10AC-TiO2-P25
[PFBA] = 400 ppm, 10AC-TiO2-P25 = 1 g/L, airflow rate = 8.1 mL s−1
, irradiation
Fig. 6. Primary defluoridation of PFBA [PFBA] = 400 ppm, catalyst concentra-
tion = 1 g/L, pH = 7, airflow rate = 8.1 mL s−1, I254 nm = 2.54 × 10−5 Einstein L−1 s−1
.
time = 60 min, I254 nm = 2.54 × 10−5 Einstein L−1 s−1
.

94
L. Ravichandran et al. / Journal of Molecular Catalysis A: Chemical 317 (2010) 89–96
Table 2
Effect of oxidants on photodefluoridation of PFBA.
Oxidants
Fluoride ion released (mg L−1
)
Without catalyst
With TiO2-P25
With AC-TiO2-P25
KIO4
H2O2
KBrO3
(NH4)2S2O8
KClO3
119.6
90.1
80.2
86
180
158.3
144
146.4
112.5
180
90.4
172.8
160.2
170.8
18.6
[PFBA] = 400 ppm, TiO2-P25/AC-TiO2-P25 = 175 mg, oxidants = 125 mg, H2O2
=
=
0.01 M, pH = 7, airflow rate = 8.1 mL s−1
,
irradiation time = 30 min, I254 nm
2.54 × 10−5 Einstein L−1 s−1
.
the following equation [27]:
kKC
r =
(2)
(3)
1 + KC
This equation can be modified as:
1
r
1
kKC
1
k
=
+
Fig. 9. Effect of adsorption of PFBA with AC-TiO2-P25 at different pH
[PFBA] = 100 ppm, AC-TiO2-P25 = 100 mg.
where C is the initial concentration of PFBA, k is the reaction rate
constant and K is the Langmuir adsorption constant. The k reflects
the limiting rate of the reaction at maximum coverage under the
given experimental conditions and K represents the equilibrium
constant for adsorption of PFBA on the illuminated catalyst.
The applicability of L–H equation for adsorption has been con-
firmed by the linearity of the plot of 1/r vs 1/C. This also indicates
that the defluoridation of PFBA occurred mainly on the surface of
catalyst. The values of k and K determined from the plot are found
to be 0.538 × 10−4 M−1 s−1 and 0.300 × 10−4 M−1, respectively for
AC-TiO2-P25.
oridation relates the amount of OH radical formation on catalyst
surface and the probability of OH radical reacting with PFBA
molecules. For all initial PFBA concentrations, the catalyst concen-
tration (175 mg) and UV power (2.54 × 10−5 Einstein L−1 s−1) are
same. Since the generation of hydroxyl radical remains constant,
the probability of PFBA molecule to react with hydroxyl radical
decreases.
Many authors [25] have used the Langmuir–Hinshelwood (L–H)
kinetic expression to analyze the heterogeneous photocatalytic
reaction. The experimental data have been rationalized in terms of
the modified form of L–H kinetic model to describe the solid–liquid
reaction successfully [26]. The rate of defluoridation of PFBA at the
surface is proportional to the surface coverage of PFBA on the cat-
alyst, assuming that the PFBA is strongly adsorbed on the catalyst
surface than the intermediate products [26]. The effect of initial
PFBA concentration on the initial rate of defluoridation is given in
3.2.3. Effect of inorganic oxidants
Table 2 gives the comparison of defluoridation by oxidants,
oxidants with catalysts TiO2-P25 and AC-TiO2-P25. It can be
observed from the table that defluoridation is enhanced by the
addition of oxidants except H2O2. The defluoridation efficiency
is in the following order: UV/AC-TiO2-P25/oxidants > UV/TiO2-
P25/oxidants > UV/oxidants.
The enhancement in UV/AC-TiO2-P25 and UV/TiO2-P25 by the
addition of oxidant is due increased charge separation caused by
oxidants. In case of H2O2 the decrease in defluoridation is observed.
The negative effect of H2O2 on some semiconductor photocataly-
sis of organic compounds had been reported by several authors
[28–30]. H2O2 may be adsorbed on AC-TiO2-P25, reducing the
activity of the catalyst.
3.3. Synergistic effect
Enhanced photocatalytic activity of semiconductors has been
achieved by the processes such as metal doping, surface modifica-
tion and mixing of two semiconductors, etc. But here we observed
the enhanced photocatalytic activity of TiO2-P25 when loaded with
activated carbon. Activated carbon has no photocatalytic activity
but it increases the photocatalytic activity of TiO2-P25 and this is
due to the increased adsorption of PFBA on AC-TiO2-P25. Increased
adsorption leads to higher concentration of PFBA around TiO2-P25.
This exhibits synergism, in which the adsorbed PFBA on AC is trans-
ferred to TiO2-P25. The complete degradation of PFBA adsorbed
on AC-TiO2-P25 was confirmed by FT-IR studies. The FT-IR spectra
of AC-TiO2-P25 photocatalyst after the adsorption of PFBA in dark
and after complete defluoridation on irradiation along with FT-IR
spectrum of catalyst alone are shown in Fig. 11a–c. New absorp-
tion bands such as 1650, 1613, 1527, 1490, 1402 and 1290 cm−1
appeared on the absorption spectra of the catalyst after PFBA
Fig. 10. Effect of initial PFBA concentration on photodefluoridation with
10AC-TiO2-P25.
10AC-TiO2-P25 = 1 g/L,
pH = 7,
airflow
rate = 8.1 mL s−1
,
I254 nm = 2.54 × 10−5 Einstein L−1 s−1
.

L. Ravichandran et al. / Journal of Molecular Catalysis A: Chemical 317 (2010) 89–96
95
Table 3
TOC and HPLC values of PFBA degradation.
Time (min)
Total organic carbon (ppm)
HPLC analysis (%)
Standard
157
98
54
99.4
30
60
90
68.63
37.45
6.96
10.3
Table 4
Percentages of PFBA defluoridation by various processes (irradia-
tion time = 60 min).
Process
% of defluroidation
UV/TiO2-P25
UV/ZnO
78.0
68.6
92.6
79.2
99.3
97.0
UV/Fe3/H2O2
UV/Fe2/H2O2
UV/AC-TiO2-P25
UV/Pt-TiO2-P25
In HPLC, the standard peak for PFBA with retention time of 6 min
decreased from 99.45% to 68.63 and 6.96% at the time of 30 and
90 min irradiation respectively. This again confirms about 93% of
PFBA degradation in 90 min. Analysis of HPLC after 30 min of irra-
diation reveals the formation of three peaks with retention times
4.38, 3.68 and 2.81. By matching with the standards, theses three
peaks are found to be due to the formation of pentafluorobenzene,
pentafluorophenol and tetrafluoroquinone as intermediates during
the irradiation. The formation of these intermediates in the deflu-
oridation of PFBA by UV-TiO2-P25 process was already confirmed
by GC–MS analysis [15].
Fig. 11. FT-IR spectra of (a) fresh AC-TiO2-P25, (b) AC-TiO2-P25 after PFBA adsorp-
tion and (c) AC-TiO2-P25 after complete defluoridation of PFBA.
adsorption can be assigned to the C–F stretching modes of PFBA
on to the catalyst surface (Fig. 11b). In general, fluorine containing
compounds exhibit strong absorptions in the region from 1400 to
1000 cm−1 due to C–F stretching modes. After irradiation of AC-
TiO2-P25 with the adsorbed PFBA, peaks in this range disappeared
giving a spectrum (Fig. 11c) similar to the spectrum of the catalyst
alone (Fig. 11a). This confirms that the adsorbed PFBA molecules on
AC are being transferred to the TiO2-P25, where they are degraded
under irradiation as shown in Scheme 1.
3.5. Comparison of photodefluoridation of PFBA
The percentages of photodefluoridation of 400 ppm PFBA with
this catalyst is compared to our earlier results reported for the
defluoridation of PFBA using the catalysts TiO2-P25 [15], ZnO [16],
Fenton [17] and Pt-TiO2-P25 [31] at their respective optimum con-
ditions. The results are given in Table 4. Among these catalysts,
AC-TiO2-P25 is found to be most efficient. The order of defluorida-
tion efficiency is as follows:
3.4. Mineralisation of PFBA
UV/AC-TiO2-P25 > UV/Pt-TiO2-P25 > UV/Fe3+/H2O2 > UV/Fe2+
/
In addition to measurement of defluoridation by fluoride ion
selective electrode, it is necessary to confirm the mineralisation by
other techniques also. Hence, the irradiated samples of PFBA were
analyzed by total organic carbon (TOC) measurement and by HPLC.
TOC values for the PFBA irradiated sample (400 ppm initial concen-
tration of PFBA) in UV/AC-TiO2-P25 catalyst at specified intervals of
time are given in Table 3. The value of 10.3 ppm at 90 min indicates
94% degradation, which is very close to complete mineralisation.
For observing the trend of degradation, HPLC analysis was done for
these samples and the percentages of PFBA are also given in Table 3.
H2O2 > UV/TiO2-P25 > UV/ZnO
The advantage of AC-TiO2-P25 process is that the maximum deflu-
oridation efficiency (99.3%) is observed at neutral pH 7 which is
beneficial and easy to implement in any wastewater treatment.
Also separation of catalyst becomes easier. The increased efficiency
of UV/AC-TiO2-P25 process can be attributed mainly to the syner-
gism of activated carbon. Moreover light absorption increases due
to increment of the surfacial electron charge of the hydroxyl groups
in the presence of AC.
Scheme 1. Synergistic effect of AC-TiO2-P25 on photodefluoridation of PFBA.

96
L. Ravichandran et al. / Journal of Molecular Catalysis A: Chemical 317 (2010) 89–96
4. Conclusions
[4] T. Tsumura, N. Kojitani, I. Izumi, N. Iwashita, M. Toyoda, M. Inagaki, J. Mater.
Chem. 12 (2003) 1391–1396.
[5] Y. Xu, W. Zheng, W. Liu, J. Photochem. Photobiol. A: Chem. 122 (1999) 57–60.
[6] K. Shimizu, T. Kaneko, T. Fujishima, T. Kodama, H. Yosida, Y. Kitayama, Appl.
Catal. A: Gen. 225 (2002) 185–191.
[7] J. Matos, J. Laine, J.M. Herrmann, J. Catal. 200 (2001) 10–20.
[8] J. Arana, J.M. Dona Rodriguez, C. Garriga i Cabo, O. Gonzalez-Diaz, J.A. Herrera-
Melian, J. Perez Pena, G. Colon, J.A. Navio, Appl. Catal. B: Environ. 44 (2003)
161–172.
[9] S.J. Tavener, J.H. Clark, J. Fluorine Chem. 123 (2003) 31–36.
[10] R. Natarajan, R. Azerad, B. Badet, E. Copin, J. Fluorine Chem. 126 (2005) 424–435.
[11] R. Renner, Environ. Sci. Technol. 35 (2001) 154A.
[12] J.P. Giesy, K. Kannan, Environ. Sci. Technol. 36 (2002) 146A–152A.
[13] K.J. Hansen, H.O. Johnson, J.S. Eldridge, J.L. Buhenhoff, L.A. Dick, Environ. Sci.
Technol. 36 (2002) 1681–1685.
[14] K. Kannan, J.W. Choi, N. Iseki, K. Senthilkumar, D.H. Kim, S. Masunaga, G.P. Giesy,
Chemosphere 49 (2002) 225–231.
[15] L. Ravichandran, K. Selvam, M. Muruganandham, M. Swaminathan, J. Fluorine
Chem. 127 (2006) 1204–1210.
[16] L. Ravichandran, K. Selvam, M. Swaminathan, Sep. Purif. Technol. 56 (2007)
192–198.
[17] L. Ravichandran, K. Selvam, M. Swaminathan, J. Photochem. Photobiol. A: Chem.
188 (2007) 392–398.
The increase in the photocatalytic activity of TiO2-P25 by the
addition of activated carbon was observed. Activated carbon was
loaded into TiO2-P25 in 4, 8, 10 and 12% by weight. These cat-
alysts were characterized by various surface analytical methods
such as BET surface area measurement, diffuse reflectance spec-
troscopy, FT-IR spectroscopy, scanning electron microscopy and
X-ray diffraction analysis. Among AC-TiO2-P25 catalysts, 10AC-
TiO2-P25 was found to be most efficient.
The experiments revealed that the defluoridation is more effec-
tive in 10AC-TiO2-P25 than bare TiO2-P25 and 254 nm is more
effective than 365 nm. Complete defluoridation was observed in
60 min with 10AC-TiO2-P25 in comparison to 90 min with bare
TiO2-P25. The higher photocatalytic defluoridation efficiency of
10AC-TiO2-P25 at neutral pH is due to synergistic effect of AC.
As experienced with other catalysts, increase of PFBA concentra-
tion decreases the defluoridation rate. The addition of inorganic
oxidants to catalyst enhances the defluoridation efficiency and
is in the following order: UV/AC-TiO2-P25/oxidants > UV/TiO2-
P25/oxidants > UV/oxidants.
[18] E. Carpio, P. Zuniga, S. Ponce, J. Solis, J. Rodriguez, W. Estrada, J. Mol. Catal. A
228 (2005) 293–298.
[19] C.G. Hatchard, C.A. Parker, Proc. R. Soc. London A 235 (1956) 518.
[20] J. Arana, J.M. Dona-Rodriguez, E. Tello-Rendon, C. Garriga-i-Cabo, O. Gonzalez-
Diaz, J.A. Herrera-Melian, J. Perez-Pena, G. Colon, J.A. Navio, Appl. Catal. B 44
(2003) 153–160.
Acknowledgements
[21] V. Brezova, M. Jankovicova, M. Soldan, A. Blazkova, M. Rehakova, I. Surina, M.
Ceppan, B. Havilinova, J. Photochem. Photobiol. A 89 (1994) 69–75.
[22] I. Bedja, S. Hotchandani, P.V. Kamat, J. Phys. Chem. 97 (1993) 11064–11070.
[23] P.A. Connor, K.D. Dobson, A.J. McQuillan, Langmuir 15 (1999) 2402–2408.
[24] E. Papirer, S. Li, J.B. Donnet, Carbon 25 (1987) 243–247.
[25] L. Wenhua, L. Hong, C. Sao’an, Z. Jianqing, C. Chunan, J. Photochem. Photobiol.
A 131 (2000) 125–132.
One of the authors, K. Selvam is thankful to CSIR, New Delhi,
for the award of Senior Research Fellowship. We thank Catalysis
Laboratory, IIT Madras, Chennai for BET measurements.
[26] H. Al-Ekabi, N. Serpone, J. Phys. Chem. 92 (1988) 5726–5731.
[27] R.W. Matthews, J. Phys. Chem. 91 (1987) 3328–3333.
[28] J. Harbour, J. Tromp, M.L. Hair, Can. J. Chem. 63 (1985) 204–208.
[29] C. Kormann, D.W. Bahnemann, M.R. Hoffmann, Environ. Sci. Technol. 22 (1988)
798–806.
[30] E. Evgenidou, K. Fytianos, I. Poulios, Appl. Catal. B 59 (2005) 81–89.
[31] L. Ravichandran, K. Selvam, B. Krishnakumar, M. Swaminathan, J. Hazard. Mater
167 (2009) 763–769.
References
[1] S. Sakthivel, S.U. Geissen, D.W. Bahnemann, V. Murugesan, A. Vogelpohl, J.
Photochem. Photobiol. A: Chem. 148 (2002) 283–293.
[2] A. Dipaola, E. Garicia-Lopez, S. Ikeda, G. Marci, B. Ohtani, L. Palmisano, Catal.
Today 75 (2002) 87–93.
[3] R.L. Pozzo, M.A. Baltanas, A.E. Cassano, Catal. Today 39 (1997) 219–231.