Is surface fluorination of TiO2 effective for water purification? The degradation vs. mineralization of phenolic pollutants

Article abstract of DOI:10.1016/j.cattod.2016.03.010

The photocatalytic activity of surface fluorinated TiO₂ (F-TiO₂) for both the degradation and mineralization of bisphenol A (BPA) was compared with that of pure TiO₂. The degradation rate of BPA (i.e., the conversion rate of BPA to intermediates) was enhanced, but the mineralization rate (i.e., the conversion rate of BPA to CO₂) was reduced by surface fluorination. These behaviors are different from the general trend in photocatalysis, in which the photocatalyst with a higher activity for the degradation also shows a higher activity for the mineralization. The surface fluorination of TiO₂ enhanced the production of the hydroxyl radical ([rad]OH), which is primarily responsible for the degradation of BPA, by altering the [rad]OH generation pathway. However, the lower mineralization on F-TiO₂, which produced more [rad]OH, implies that the role of [rad]OH in the photocatalytic mineralization process is minor. The production of superoxide/hydroperoxyl radical (O₂[rad]^?/HO₂[rad]), which is suggested as an essential oxidant for the mineralization of phenolic pollutants, by F-TiO₂ was lower than that exhibited by pure TiO₂. The reduced photocurrent (I_ph) generation and the enhanced H₂O₂ production on F-TiO₂ indicate that fluorides on the TiO₂ surface reduce the interfacial electron transfer rate (i.e., the production of O₂[rad]^?/HO₂[rad]) and enhance the reduction of O₂[rad]^?/HO₂[rad] to H₂O₂. The degradation rate increased, but the mineralization efficiency decreased with increasing the surface coverage of fluorides, which depends on the pH and fluoride concentration in the solution. The reduced mineralization efficiency of other phenolic pollutants (4-chlorophenol, phenol, methylene blue, rhodamine B, and acid orange 7) was also observed on F-TiO₂. This result indicates that the negative effect of surface fluorination on the mineralization of phenolic pollutants is pervasive and is not restricted to BPA.

Full text of DOI:10.1016/j.cattod.2016.03.010

G Model
CATTOD-10104; No. of Pages7
ARTICLE IN PRESS
Catalysis Today xxx (2016) xxx–xxx
Contents lists available at ScienceDirect
Catalysis Today
journal homepage: www.elsevier.com/locate/cattod
Is surface fluorination of TiO2 effective for water purification? The
degradation vs. mineralization of phenolic pollutants
Jungho Ryua,1, Wooyul Kimb,1, Jaesung Kimc, Jinjung Juc, Jungwon Kimc,∗
a Mineral Resources Research Division, Korea Institute of Geoscience and Mineral Resources (KIGAM), Daejeon 34132, Republic of Korea
b Department of Chemical and Biological Engineering, Sookmyung Women’s University, Seoul 04310, Republic of Korea
c Department of Environmental Sciences and Biotechnology, Hallym University, Chuncheon, Gangwon-do 24252, Republic of Korea
a r t i c l e i n f o
a b s t r a c t
Article history:
The photocatalytic activity of surface fluorinated TiO2 (F-TiO2) for both the degradation and mineraliza-
tion of bisphenol A (BPA) was compared with that of pure TiO2. The degradation rate of BPA (i.e., the
conversion rate of BPA to intermediates) was enhanced, but the mineralization rate (i.e., the conversion
rate of BPA to CO2) was reduced by surface fluorination. These behaviors are different from the general
trend in photocatalysis, in which the photocatalyst with a higher activity for the degradation also shows
a higher activity for the mineralization. The surface fluorination of TiO2 enhanced the production of the
hydroxyl radical (•OH), which is primarily responsible for the degradation of BPA, by altering the •OH
generation pathway. However, the lower mineralization on F-TiO2, which produced more •OH, implies
that the role of •OH in the photocatalytic mineralization process is minor. The production of superox-
ide/hydroperoxyl radical (O2•−/HO2•), which is suggested as an essential oxidant for the mineralization
of phenolic pollutants, by F-TiO2 was lower than that exhibited by pure TiO2. The reduced photocurrent
(Iph) generation and the enhanced H2O2 production on F-TiO2 indicate that fluorides on the TiO2 surface
reduce the interfacial electron transfer rate (i.e., the production of O2•−/HO2•) and enhance the reduc-
Received 16 November 2015
Received in revised form 4 March 2016
Accepted 10 March 2016
Available online xxx
Keywords:
Photocatalysis
Surface fluorinated TiO2
Degradation
Mineralization
Reactive oxygen species
•
tion of O2•−/HO2 to H2O2. The degradation rate increased, but the mineralization efficiency decreased
with increasing the surface coverage of fluorides, which depends on the pH and fluoride concentration in
the solution. The reduced mineralization efficiency of other phenolic pollutants (4-chlorophenol, phenol,
methylene blue, rhodamine B, and acid orange 7) was also observed on F-TiO2. This result indicates that
the negative effect of surface fluorination on the mineralization of phenolic pollutants is pervasive and
is not restricted to BPA.
© 2016 Elsevier B.V. All rights reserved.
1. Introduction
superoxide/hydroperoxyl radical (O2•−/HO2•) and a hydroxyl radi-
cal (•OH), respectively (Reactions (2) and (3)). In addition, H2O2 can
•
be produced through the proton-coupled reduction of O2•−/HO2
Semiconductor photocatalysis has been regarded as a promis-
ing water treatment technology and extensively applied to the
purification of contaminated water. Among many photocatalysts
for water treatment, titanium dioxide (TiO2) is the most popular
material because of its high removal efficiency for various types
of pollutants, nontoxicity, stability over a wide range of pH, and
low cost of production [1–4]. Under UV (> bandgap energy) irradi-
ation, electrons in the valence band (VB) of TiO2 are excited to the
conduction band (CB), and this creates CB electrons (ecb−) and VB
holes (hvb+) (Reaction (1)). The reduction of oxygen by ecb− and the
+
by ecb− (Reaction (4)). Mostly, these four oxidizing species (i.e., hvb
,
O2•−/HO2•, •OH, and H2O2) are primarily involved in the oxidative
process.
+
TiO2 + hꢀ(UVphotons) → ecb− + hvb
(1)
(2)
(3)
(4)
ecb− + O2 → O2 −(orecb− + O2 + H+ → HO2
•OH(orhvb+ + H2O → •OH + H+)
ecb− + O2 − + 2H+ → H2O2(orecb− + HO2 + H+ → H2O2)
•
•
)
hvb+ + OH−
→
•
•
+
oxidation of hydroxide (or a water molecule) by hvb generates a
The photocatalytic activity of TiO2 for the removal of aquatic pol-
lutants is strongly related to the pollutant-surface interaction, the
charge transfer/recombination efficiency, and the type of reactive
oxygen species (ROS), which is primarily generated by electron-
hole pairs. Because these behaviors are strongly dependent on the
∗
Corresponding author.
E-mail address: jwk@hallym.ac.kr (J. Kim).
Contribution as co-first author.
1
http://dx.doi.org/10.1016/j.cattod.2016.03.010
0920-5861/© 2016 Elsevier B.V. All rights reserved.
Please cite this article in press as: J. Ryu, et al., Is surface fluorination of TiO2 effective for water purification? The degradation vs.
mineralization of phenolic pollutants, Catal. Today (2016), http://dx.doi.org/10.1016/j.cattod.2016.03.010

G Model
CATTOD-10104; No. of Pages7
ARTICLE IN PRESS
2
J. Ryu et al. / Catalysis Today xxx (2016) xxx–xxx
surface properties of TiO2, a variety of surface modification meth-
ods have been developed to enhance the photocatalytic activity
of TiO2; they include carbon material coating [5–8], metal deposi-
tion [9,10], anion adsorption [11,12], and hybridization with other
semiconductors [13,14]. Among these modifications, the surface
fluorination of TiO2 (i.e., the adsorption of fluoride anions on the
TiO2 surface) has been a popular method because it is highly effi-
cient, simple, and inexpensive. Surface fluorinated TiO2 (F-TiO2)
has shown a photocatalytic activity that is higher than pure TiO2 for
the degradation of diverse aquatic pollutants, such as bromophe-
nol blue [15], reactive orange 4 [16], acid red 1 [17], methylene blue
[18,19], N-nitrosodimethylamine [20], arsenite [21], acid orange 7
[22], rhodamine B [23], and phenol [24]. However, in previous stud-
ies on F-TiO2, only the degradation rate of the parent compound
(i.e., the conversion rate of the parent compound to intermediates)
was compared.
In the case of organic pollutants, the reaction between the
parent compound and oxidizing species initiates the degrada-
tion of the parent compound, and the subsequent reactions of
intermediates with oxidizing species lead to their mineralization
(i.e., the conversion of intermediates to CO2). The intermediates
generated from the oxidation of the parent compound can be
more toxic than parent compound. The photocatalytic oxidation
of 2,4-dichlorophenoxy acetic acid generates the intermediate,
2,4-dichlorophenol, which is more toxic [25]. Even more toxic
intermediates, such as 2,6-dichlorodibenzo-p-dioxin and 2,4,6-
trichlorophenol, are generated from the photocatalytic oxidation of
2,4-dichlorophenol [26]. It has also been reported that the oxidation
intermediates of nicotine and diclofenac are more toxic than their
parent compound [27,28]. In this regard, not only is the degrada-
tion efficiency important but the mineralization efficiency should
also be measured to evaluate the practical applicability of a new
photocatalyst for water purification.
deionized water was ultrapure (18.3 Mꢁ·cm) and prepared using a
Human-Power I+ water purification system (Human corporation).
2.2. Photocatalytic and photoelectrochemical experiments
The TiO2 powder (50 mg) was dispersed in deionized water by
sonication for 1 min in an ultrasonic cleaning bath (Branson). An
aliquot of the phenolic pollutant stock solution was subsequently
added to the TiO2 suspension to yield the desired initial concen-
tration (200 ␮M). The initial pH (pHi) of the TiO2 suspension was
adjusted with concentrated HClO4 and NaOH solutions. The total
volume of the solution was 100 mL. For surface fluorination of TiO2,
NaF (usually 10.5 mg, 2.5 mM) was added to the TiO2 suspension
(50 mg/100 mL). Then, the pH of the TiO2 suspension was lowered
(usually to 3.0) to induce ligand exchange between fluoride anions
in the solution and hydroxyl groups on the TiO2 surface (Reaction
(5)) [29].
> Ti OH + F− ↔ > Ti F + OH−(pKF = 6.2)
(5)
The TiO2 suspension was stirred for 30 min in the dark to ensure
the adsorption equilibrium of phenolic pollutants and fluorides on
the surface of TiO2.
A 300 W Xe arc lamp (Oriel) was used as a light source. The
light beam was passed through a 5 cm IR water filter and a cut-
off filter (␭ > 320 nm) and focused onto a cylindrical Pyrex reactor
(volume = 120 mL) with a quartz window. The reactor was open
to ambient air to prevent the depletion of dissolved oxygen and
was magnetically stirred during the UV irradiation. All experiments
were performed at least in duplicate for a given condition to confirm
data reproducibility.
Photocurrent (Iph) measurements were performed using a con-
ventional three-electrode system. A Pt plate, a saturated calomel
electrode (SCE), and a graphite rod were used as the working, ref-
erence, and counter electrodes, respectively. The Fe3+/Fe2+ redox
couple was used as an electron mediator that transfers electrons
from the TiO2 CB to the working electrode, where +0.7 VSCE of
potential was applied. LiClO4 (0.1 M) was used as the electrolyte.
To remove dissolved oxygen, the aqueous suspension of TiO2 was
continuously purged with nitrogen gas (N2, 99.999%) for 1 h prior to
and during the UV irradiation. The Iph was recorded by a computer-
controlled potentiostat (EG&G 263A2).
In this study, both the degradation and mineralization efficiency
of F-TiO2 were compared with those of pure TiO2. F-TiO2 showed
a higher photocatalytic activity for the degradation of bisphenol
A (BPA) than pure TiO2. However, the mineralization efficiency of
F-TiO2 was less than that of pure TiO2. The degradation and miner-
alization efficiencies were compared between pure TiO2 and F-TiO2
as a function of various experimental parameters, such as initial pH
(pHi), the fluoride concentration, and the type of phenolic pollu-
tants. Furthermore, detailed mechanistic investigations of the dual
effect of TiO2 surface fluorination on the degradation and mineral-
ization were performed by measuring the ROS production and the
photoelectrochemical property.
2.3. Chemical analyses
The concentrations of BPA, P, 4-CP, and BQ were measured
using a high performance liquid chromatography (HPLC, Agilent
1120) equipped with a UV–vis detector and a ZORBAX 300SB C-
18 column (4.6 mm × 150 mm). The eluent was a binary mixture of
phosphoric acid solution (0.1%) and acetonitrile. The volume ratio
of the phosphoric acid solution and acetonitrile was 7:3, 9:1, 8:2,
and 9:1 for the analysis of BPA, P, 4-CP, and BQ, respectively. The
eluent flow rate was 1.0 mL/min, and the temperature of column
was 30 ◦C. The detection wavelength was set to 229 nm, 270 nm,
228 nm, and 244 nm for BPA, P, 4-CP, and BQ analyses, respec-
tively. The concentrations of dye pollutants were determined using
a UV−vis spectrophotometer (Shimadzu UV-2600) by measuring
the absorbance at 663 nm, 553 nm, and 485 nm for MB, RhB, and
AO7, respectively. The mineralization of phenolic pollutants was
monitored by measuring the total organic carbon (TOC) concentra-
tion using a TOC analyzer (Shimadzu TOC-VCSH) equipped with a
nondispersive infrared sensor (NDIR) as a CO2 detector. The con-
centration of TOC was obtained by subtracting the concentration of
inorganic carbon (IC) from the concentration of total carbon (TC).
The presence of fluorides (up to 10 mM) had little effect on the TOC
measurement.
2. Experimental
2.1. Materials and chemicals
Materials and chemicals were used as received without fur-
ther purification. They include titanium dioxide (TiO2, Degussa
P25), sodium fluoride (NaF, Sigma-Aldrich), bisphenol A (BPA,
C15H16O2, Aldrich), 4-chlorophenol (4-CP, C6H5ClO, Aldrich), phe-
nol (P, C6H6O, Sigma-Aldrich), methylene blue (MB, C16H18N3SCl,
Aldrich), rhodamine B (RhB, C28H31ClN2O3, Sigma), acid orange
7 (AO7, C16H11N2NaO4S, Aldrich), coumarin (C9H6O2, Sigma),
benzoquinone (BQ, C6H4O2, Sigma-Aldrich), tert-butyl alcohol
(TBA, C4H10O, Junsei), N,N-diethyl-1,4-phenylenediamine (DPD,
(C2H5)2NC6H4NH2, Aldrich), peroxidase from horseradish (POD,
type VI-A, Aldrich), sulfuric acid (H2SO4, Kanto), phosphate buffer
reagent (NaH2PO4·H2O (Aldrich, 0.44 M) + Na2HPO4·7H2O (Aldrich,
0.06 M), pH = 6.0), iron(III) perchlorate hydrate (Fe(ClO4)3·xH2O,
Aldrich), and lithium perchlorate (LiClO4, Sigma-Aldrich). The
Please cite this article in press as: J. Ryu, et al., Is surface fluorination of TiO2 effective for water purification? The degradation vs.
mineralization of phenolic pollutants, Catal. Today (2016), http://dx.doi.org/10.1016/j.cattod.2016.03.010

G Model
CATTOD-10104; No. of Pages7
ARTICLE IN PRESS
J. Ryu et al. / Catalysis Today xxx (2016) xxx–xxx
3
pure TiO2 and F-TiO2. The degradation of BPA in the suspension of
F-TiO2 proceeded more rapidly than that in the suspension of pure
TiO2 (Fig. 1a). The pseudo-first-order degradation rate constant (k)
increased by three times with TiO2 surface fluorination (pure TiO2:
0.42 h−1 and F-TiO2: 1.25 h−1). In contrast to the degradation trend,
the mineralization (i.e., the removal of total organic carbon (TOC))
of BPA on F-TiO2 was slower than it was on pure TiO2 (Fig. 1b).
After 4 h of UV irradiation, 43% of TOC was reduced on pure TiO2,
whereas the TOC removal was only 4% on F-TiO2. From the HPLC
analysis, it was determined that the concentrations of the residual
intermediates were more significant for F-TiO2 than for pure TiO2,
at any point of reaction (see Fig. S1 in the Supplementary data).
This contrasting behavior (i.e., the lower mineralization efficiency
with the higher degradation efficiency) on F-TiO2 indicates that the
transformation of the parent compound (i.e., BPA) to intermediates
is enhanced, but the complete oxidation of the intermediates to CO2
is reduced by surface fluorination.
Most often, the photocatalyst with a higher activity for the
degradation also shows a higher activity for the mineralization. For
example, platinum (Pt) deposited on the TiO2 surface, which can
accelerate the electron transfer from the TiO2 CB to oxygen [34],
enhanced both the degradation and mineralization of BPA (see Fig.
S2 in the Supplementary data). In this regard, a mismatch between
degradation and mineralization on F-TiO2 is a unique phenomenon
in photocatalytic purification processes.
The higher mineralization efficiency of pure TiO2 compared
with F-TiO2 at pHi = 3.0 may be due to the higher adsorption of
intermediates on the surface of pure TiO2 because surface fluo-
rides on F-TiO2 can inhibit the adsorption of intermediates [35]. To
investigate the effect of intermediate adsorption on mineralization
efficiency of pure TiO2, the TOC was measured in a sample after 4 h
of UV irradiation and after increasing the pH to 11.0. At this pH, the
intermediates adsorbed on the surface of pure TiO2 at pHi = 3.0 can
be desorbed by electrostatic repulsion because the surface charge
of pure TiO2 changes from positive to negative (pKa2 = 8.7 [29]).
However, the increase of pH (i.e., the desorption of intermediates)
had little effect on the TOC value not only in the case of F-TiO2 but
also in the case of pure TiO2 (see Fig. S3 in the Supplementary data).
This result indicates that the adsorption of intermediates generated
from BPA degradation on both pure TiO2 and F-TiO2 is negligible.
Therefore, the higher mineralization efficiency of pure TiO2 is not
due to its higher adsorption capacity.
Fig. 1. (a) Photocatalytic degradation and (b) mineralization of BPA in suspensions
of pure TiO2 and F-TiO2. Experimental conditions: [TiO2] = 0.5 g L−1, [BPA] = 200 ␮M,
[NaF] = 2.5 mM, pHi = 3.0, and ␭ > 320 nm.
The production of •OH in the UV-irradiated suspensions of pure
TiO2 and F-TiO2 was compared by measuring the fluorescence
emission intensity of 7-hydroxycoumarin (a product of the reac-
tion between •OH and coumarin) in the presence of coumarin [30].
The fluorescence emission intensity of 7-hydroxycoumarin excited
by monochromatic light with a wavelength of 332 nm was mea-
sured at 460 nm using a spe•ctrofluorometer (Shimadzu RF-5301).
The production of O2•−/HO2 was indirectly evaluated by measur-
ing the degradation of BQ in the presence of an excess of TBA, as an
•OH scavenger [31,32]. The concentration of H2O2 in the UV/pure
TiO2 (or F-TiO2) system was measured by the colorimetric DPD
method [33]. 50 ␮L of DPD solution (0.1 g/10 mL (0.1 M H2SO4)),
50 ␮L of POD solution (0.01 g/10 mL), and 0.4 mL of phosphate
buffer reagent were added to a 3-fold diluted sample (3 mL). The
solution was vigorously mixed for 1 min and then its absorbance
was measured at 551 nm (␧ = 21000 M−1 cm−1 [33]) using a UV–vis
spectrophotometer (Shimadzu UV-2600).
3.2. Production of reactive oxygen species (ROS) on pure and
surface fluorinated TiO2
Under UV irradiation, TiO2 can produce four oxidizing species,
valence band holes (hvb+), hydroxyl radicals (•OH), superox-
ide/hydroperoxyl radicals (O2•−/HO2•), and hydrogen peroxide
(H2O2). The photocatalytic oxidation of BPA, which is hardly
adsorbed on the TiO2 surface, is primarily initiated by reactive oxy-
gen species (ROS) (i.e., •OH, O2•−/HO2•, and H2O2) but not by hvb
.
+
This is because ROS can diffuse out from the surface, however,
the adsorption of the pollutant on the surface is essential for the
hvb+-mediated oxidation [36,37].
To investigate which ROS is primarily involved in the degra-
dation of BPA, the production of ROS on F-TiO2 was measured
and compared with that on pure TiO2 (Fig. 2). The production of
•OH, which is proportional to the fluorescence emission inten-
sity of 7-hydroxycoumarin, was enhanced by surface fluorination
(Fig. 2a). This is consistent with the previous result obtained by
electron paramagnetic r•esonance (EPR) measurements [38]. The
production of O2•−/HO2 on F-TiO2 was compared with that on
pure TiO2 by monitoring the degradation of BQ (the bimolecu-
lar rate constant between BQ and O2•−/HO2• = 1.0 × 109 M−1 s−1
[39]) (Fig. 2b) [31,32]. To eliminate the possibility of BQ degra-
3. Results and discussion
3.1. Degradation and mineralization of bisphenol A (BPA) on pure
and surface fluorinated TiO2
The photocatalytic degradation and mineralization of bisphe-
nol A (BPA), as a model phenolic pollutant, on surface fluorinated
TiO2 (F-TiO2) were compared with those on pure TiO2 (Fig. 1). The
degradation of BPA (i.e., the removal of the parent pollutant) fol-
lowed pseudo-first-order kinetics with an R2 > 0.99, in both cases of
Please cite this article in press as: J. Ryu, et al., Is surface fluorination of TiO2 effective for water purification? The degradation vs.
mineralization of phenolic pollutants, Catal. Today (2016), http://dx.doi.org/10.1016/j.cattod.2016.03.010

G Model
CATTOD-10104; No. of Pages7
ARTICLE IN PRESS
4
J. Ryu et al. / Catalysis Today xxx (2016) xxx–xxx
Scheme 1. Schematic illustrations of (a) the photocatalytic degradation and (b)
mineralization of BPA in suspensions of pure TiO2 and F-TiO2 (B: BPA and I: inter-
mediates).
was limited, whereas it was significant on F-TiO2 (Fig. 2c). Over-
all, the surface fluorination of TiO2 enhances the production of •OH
and H2O2 but reduces the production of O2•−/HO2•. In this situa-
tion, the enhanced degradation of BPA on F-TiO2 implies that •OH
or H2O2 is primarily involved in the conversion of BPA to its inter-
mediates. However, the degradation of BPA was negligible in the
presence of H2O2 (1 mM) at pHi = 3.0 (see Fig. S4 in the Supplemen-
tary data). Therefore, the role of H2O2 can be ruled out, and •OH
can be proposed as the primary ROS for the degradation of BPA.
The enhanced production of •OH by surface fluorination (Fig. 2a)
can help to explain why the degradation rate of BPA on F-TiO2 was
higher than that on pure TiO2 (Fig. 1a).
The surface fluorination of TiO2 can enhance the production of
•OH by altering the •OH generation pathway [29,41,42]. On pure
+
TiO2, hvb primarily reacts with surface hydroxyl group (>Ti–OH)
to generate surface-bound •OH ( > Ti-•OH) (Reaction (6)), which is
likely to recombine with ecb− (Reaction (7)) [43]. On the other hand,
the surface fluorination (i.e., the adsorption of fluorides on the sur-
+
face) enhances the reaction between hvb and H2O (Reaction (8))
by reducing the density of >Ti-OH (Reaction (5)) and inhibiting the
reaction between hvb and >Ti-OH (Reaction (6)). •OH generated
+
from the oxidation of H2O (unbound •OH) can easily diffuse out
from the surface (Reaction (8)), which enhances the production
of •OH by preventing an external charge recombination process
between •OH and ecb (Reaction (7)) (Scheme 1a) [29,42,44].
−
Fig. 2. (a) Photocatalytic production of 7-hydroxycoumarin in the presence of
coumarin, (b) degradation of BQ in the presence of TBA, and (c) production of H2O2
> Ti OH + hvb+ → > Ti •OH
(6)
(7)
(8)
in suspensions of pure TiO2 and F-TiO2. Experimental conditions: [TiO2] = 0.5 g L−1
,
[coumarin] = 1 mM (for a), [BQ] = 50 ␮M (for b), [TBA] = 5 M (for b), [NaF] = 2.5 mM,
pHi = 3.0, and ␭ > 320 nm.
> Ti •OH + ecb− → > Ti OH
> Ti F + hvb+ + H2O → > Ti F + (unbound)•OH + H+
dation by •OH (the bimolecular rate constant between BQ and
•OH = 1.2 × 109 M−1 s−1 [40]), an excess of TBA (5 M) was added as
an •OH scavenger (the bimolecular rate constant between TBA and
•OH = 1.7 × 105 M−1 s−1 [40]). The degradation of BQ (with TBA) in
the suspension of F-TiO2 was slower than that in the suspension of
pure TiO2. This result indicates that the surface fluorination reduces
the production of O2•−/HO2•. The production of H2O2 on pure TiO2
In many studies, it has been proposed that •OH is primarily
involved in the photocatalytic mineralization of organic pollutants
[45–51]. However, the mineralization efficiency of F-TiO2, which
produces more •OH, was less than that of pure TiO2. This result
implies that other oxidizing species (i.e., hvb+, O2•−/HO2•, and
H2O2) rather than •OH play an important role in the photocatalytic
mineralization process. The hindered adsorption of intermediates,
Please cite this article in press as: J. Ryu, et al., Is surface fluorination of TiO2 effective for water purification? The degradation vs.
mineralization of phenolic pollutants, Catal. Today (2016), http://dx.doi.org/10.1016/j.cattod.2016.03.010

G Model
CATTOD-10104; No. of Pages7
ARTICLE IN PRESS
J. Ryu et al. / Catalysis Today xxx (2016) xxx–xxx
5
eralization efficiency is reduced. It should be noted that for F-TiO2,
the production of O2•−/HO2• as a source of H2O2 was lower (Fig. 2b),
but the production of H2O2 was higher than with pure TiO2 (Fig.•2c).
These results may indicate that the conversion of O2•−/HO2 to
H2O2 through a PCET reaction (Reaction (4)) on F-TiO2 is faster
than that on pure TiO2. Although the concentration of protons in
the aqueous bulk phase of pure TiO2 and F-TiO2 suspensions is
the same, the concentration of protons near the surfaces of pure
TiO2 and F-TiO2 can be dissimilar due to the different surface-
proton interactions. A negatively charged surface can attract the
positively charged protons to the surface, which increases the con-
centration of protons near the surface [58]. At pH = 3.0, most surface
hydroxyl groups on pure TiO2 are protonated, which makes the sur-
face charge positive (Reaction (9)). Whereas the surface charge of
F-TiO2 is negative at pH = 3.0 due to the presence of anionic fluo-
rides on the surface (Reaction (10)) [29].
Fig. 3. Time profiles of the generation of photocurrent (Iph) in suspensions of
pure TiO2 and F-TiO2. Experimental conditions: [TiO2] = 0.5 g L−1, [LiClO4] = 0.1 M,
[Fe3+] = 0.5 mM, [NaF] = 2.5 mM, pHi = 2.0, ␭ > 320 nm, and the applied potential to
> Ti OH](1/3)− + H+ ↔ > Ti OH2](2/3)+(pKa1 = 3.9)
> Ti-OH](1/3)− + F− ↔ > Ti-F](1/3)− + OH−(pKF = 6.2)
(9)
(10)
the working electrode = +0.7 VSCE
.
Under these conditions, the concentration of protons near the
surface of F-TiO2 should be higher than that near the surface
of pure TiO2 at pH = 3.0 because there is an electrostatic attrac-
tion between H+ and >Ti-F](1/3)− and an electrostatic repulsion
between H+ and >Ti-OH](2/3)+. The local concentration of protons
which reduces the hole-mediated oxidation, by surface fluorina-
tion may be related to the lower mineralization efficiency on
F-TiO2. However, the adsorption of intermediates generated from
BPA degradation was negligible not only on F-TiO2 but also on
pure TiO2 (see Fig. S3 in the Supplementary data). A few stud-
ies have suggested that the reaction between an organic radical
(i.e., intermediates generated from the reaction between the parent
•
near the surface of F-TiO2 accelerates the conversion of O2•−/HO2
to H2O2 by providing a suitable environment for a PCET reaction
•
(Reaction (4)). A more rapid reduction of O2•−/HO2 to H2O2 on
•
F-TiO2 inhibits the reaction between intermediates and O2•−/HO2
•
compound and oxidizing species) and O2•−/HO2 is an essen-
and eventually reduces the mineralization efficiency (Scheme 1b).
However, surface fluorides on F-TiO2 can inhibit the adsorption of
H2O2 ( > Ti-OH + H2O2 → > Ti-OOH + H2O) and the subsequent pho-
todegradation of surface peroxides [59,60]. Therefore, the higher
residual concentration of H2O2 on F-TiO2 may be mainly due to the
lower degradation rate (but not the higher formation rate) of H2O2.
If so, a reduced interfacial electron transfer rate by surface fluori-
nation is primarily responsible for the lower O2•−/HO2• production
and mineralization efficiency on F-TiO2.
tial step for the mineralization of phenolic pollutants through
the cleavage of the aromatic structure [26,52–55]. The presence
of polyoxometalate (as an alternative electron acceptor) reduced
the photocatalytic activity of TiO2 for the mineralization of 2,4-
dichlorophenol by inhibiting the electron transfer from the TiO2 CB
to oxygen (i.e., the production of O2•−/HO2•) [26]. In the absence of
oxygen but in the presence of platinum deposits (as an alternative
electron acceptor) on the TiO2 s•urface (i.e., under the conditions
that the generation of O2•−/HO2 is not possible), 4-chlorophenol
was degraded, but its mineralization was not achieved at all [56].
•
These results corroborate the requirement of O2•−/HO2 for the
3.3. Effects of TiO2 surface fluorination under various conditions
mineralization of phenolic pollutants.
The lower mineralization on F-TiO2 indicates that the surface
fluorination reduces the production of O2•−/HO2•, which is primar-
ily responsible for the mineralization of phenolic pollutants. There
Because the adsorption of fluoride on the TiO2 surface takes
place by ligand exchange between the fluoride in the solution and
the surface hydroxyl group on TiO2, its extent should depend on
both the fluoride concentration ([F−]) and the pH. Fig. 4 shows the
effects of TiO2 surface fluorination on the degradation and miner-
alization of BPA as a function of fluoride concentration and pHi. At
the same pHi (i.e., at pHi = 3.0), the photocatalytic activity towards
degradation (i.e., the pseudo-first-order degradation rate constant
(k)) increased, but the activity towards mineralization (i.e., the TOC
removal) decreased with increasing fluoride concentration; both
quantities were saturated at [F−] = 2.5 mM (Fig. 4a). These trends
are consistent with the surface coverage of fluorides (the ratio of
the number of surface sites adsorbed by fluoride to the total num-
ber of surface sites, >Ti-F (%)), which increases with the fluoride
concentration up to [F−] = 2.5 mM (Fig. 4a). The increase in surface
coverage of fluorides should enhance the production of •OH but
reduce the production of O2•−/HO2•. Therefore, the positive effect
of surface fluorination on degradation and its negative effect on
mineralization were more pronounced with increasing the surface
coverage of fluorides (i.e., with increasing fluoride concentration
up to [F−] = 2.5 mM).
•
are two possible reasons for the lower production of O2•−/HO2
(i.e., the lower mineralization efficiency) on F-TiO2: (1) a reduced
interfacial•electron transfer rate and (2) an enhanced conversion of
O2•−/HO2 to H2O2 by surface fluorination.
If the interfacial electron transfer from the TiO2 CB to oxygen
•
is inhibited by surface fluorides, the production of O2•−/HO2 on
F-TiO2 should be lower than that on pure TiO2. To investigate the
effect of surface fluorides on the interfacial electron transfer, the
generation of a photocurrent (Iph) in UV irradiated suspensions
of TiO2 was monitored, in the presence and absence of fluorides
(Fig. 3). The Iph for F-TiO2 was lower than that observed for pure
TiO2, which indicates that surface fluorination inhibits the inter-
facial electron transfer process. Because the fluorine atom has a
strong electronegativity, it can retain electrons in the CB by hin-
dering the electron transfer from the CB to the electron acceptor in
the solution (i.e., oxygen) [57]. A holding of electrons in the CB by
surface fluorides limits the production of O2•−/HO2• and eventually
the mineralization of phenolic pollutants.
•
In addition, if the conversion of O2•−/HO2 to H2O2 through a
The effects of initial pH (pHi) on both the degradation and
mineralization were investigated at the same fluoride concentra-
tion (i.e., at [F−] = 2.5 mM) (Fig. 4b). Both the degradation rate and
the TOC removal efficiency remained almost constant indepen-
proton-coupled electron transfer (PCET) reaction (Reaction (4)) is
accelerated by surface fluorides, the production (more precisely,
•
the residual concentration) of O2•−/HO2 is reduced, and the min-
Please cite this article in press as: J. Ryu, et al., Is surface fluorination of TiO2 effective for water purification? The degradation vs.
mineralization of phenolic pollutants, Catal. Today (2016), http://dx.doi.org/10.1016/j.cattod.2016.03.010

G Model
CATTOD-10104; No. of Pages7
ARTICLE IN PRESS
J. Ryu et al. / Catalysis Today xxx (2016) xxx–xxx
Table 1
6
Pseudo-first-order degradation rate constant (k) and TOC removal of various phe-
nolic pollutants on pure TiO2 and F-TiO2.a
Type
k (h−1
)
TOC removal (%)
pure TiO2
F-TiO2
pure TiO2
F-TiO2
4-Chlorophenol
Phenol
Methylene blue
Rhodamine B
Acid orange 7
0.93( 0.03)
0.85( 0.08)
1.01( 0.05)
0.47( 0.02)
0.53( 0.02)
2.61( 0.13)
2.53( 0.15)
1.52( 0.07)
1.44( 0.03)
1.05( 0.07)
83.7( 3.2)
71.3( 2.7)
61.2( 0.8)
53.4( 3.3)
52.2( 2.5)
45.0( 4.8)
43.2( 1.4)
44.7( 4.2)
32.5( 2.6)
29.6( 1.8)
a
Experimental conditions: [TiO2] = 0.5 g L−1
, [phenolic pollutant] = 200 ␮M,
[NaF] = 2.5 mM, pHi = 3.0, and ␭ > 320 nm for 4 h.
phenolic pollutants is not restricted to BPA, which reduces the
practical applicability of F-TiO2 for water purification.
4. Conclusions
We investigated the effects of TiO2 surface fluorination on both
the degradation and mineralization of BPA. Most often, a photo-
catalyst with a high activity for the degradation (i.e., the oxidation
of the parent compound to intermediates) shows a high activity
for the mineralization (i.e., the complete oxidation of the parent
compound to CO2). Therefore, in many cases, only the degradation
rate is measured and is used as an index for the performance of
a photocatalyst. However, the mineralization efficiency of F-TiO2
for BPA was less than that of pure TiO2, despite its higher activity
for the degradation of BPA. The dual effect of surface fluorina-
tion on the degradation (positive) and mineralization (negative)
is attributed to the unique role of fluoride adsorbed on the TiO2
surface, which enhances the production of •OH (a primary oxidant
for the conversion of•BPA to intermediates) but reduces the pro-
duction of O2•−/HO2 (an essential oxidant for the conversion of
intermediates to CO2). The lower mineralization efficiency of F-
TiO2 compared with pure TiO2 was observed in various phenolic
pollutants, such as 4-chlorophenol, phenol, methylene blue, rho-
damine B, and acid orange 7. From a practical point of view, the
mineralization efficiency of a photocatalyst is more important than
the degradation efficiency because the water containing a high con-
centration of intermediates cannot be used and discharged to the
surface water. Therefore, F-TiO2 is not a practical photocatalyst,
when high purity is required in a water purification system.
Fig. 4. Effects of surface fluorination on the degradation and mineralization of
BPA as a function of (a) fluoride concentration and (b) pHi. Experimental condi-
tions: [TiO2] = 0.5 g L−1, [BPA] = 200 ␮M, [NaF] = 2.5 mM (for b), pHi = 3.0 (for a), and
␭ > 320 nm for 4 h. The fluoride surface coverage (>Ti-F (%)) was calculated using pK
values from Ref. [29].
dent of the pHi for pure TiO2. On the other hand, the degradation
rate decreased, but the mineralization efficiency increased, with
increasing the pHi for F-TiO2. That is, the effects of surface flu-
orination on both the degradation and mineralization kinetics
gradually decreased with increasing the pHi and then completely
disappeared at pHi = 11.0. These results are also related to the
pH dependency of fluoride surface coverage (Fig. 4b). However, it
should be noted that the positive and negative effects of surface
fluorination on the degradation and mineralization, respectively,
were observed even at pHi = 9.5, where the adsorption of fluorides
is negligible. This result should be ascribed to a decrease of pH by
protons released from the degradation of BPA. The initial pH (9.5)
was gradually reduced with UV irradiation time to pH conditions,
where the adsorption of fluorides is favored (i.e., to 6.5 after 1 h, 5.7
after 2 h, and 5.0 after 4 h). The in situ pH decrease accompanied by
BPA degradation facilitates the adsorption of fluorides on the TiO2
surface and induces the dual effect of surface fluorination on the
degradation (positive) and mineralization (negative).
Acknowledgments
This research was supported by Basic Science Research Pro-
gram (NRF-2013R1A1A1007312) and Space Core Technology
Development Program (NRF-2014M1A3A3A02034875) through
the National Research Foundation of Korea (NRF) funded by the
Ministry of Science, ICT & Future Planning. This work was also sup-
ported by Hallym University Research Fund (HRF-201507-011) and
Institute for Information & communications Technology Promo-
tion (IITP) grant (B0132-15-1001, Development of Next Imaging
System) funded by the Korea government (MSIP).
The TOC removal of various phenolic pollutants (4-
chlorophenol, phenol, methylene blue, rhodamine B, and acid
orange 7), in which the enhanced degradation by surface fluori-
nation has already been reported [18,19,22–24], is presented for
pure TiO2 and F-TiO2 (Table 1). For both pure TiO2 and F-TiO2,
the removal of TOC varied depending on the type of phenolic
pollutants. However, a general trend was clearly observed. The
mineralizing capability of F-TiO2 was lower than that of pure TiO2
for all phenolic pollutants tested. This result indicates that the
negative effect of surface fluorination on the mineralization of
Appendix A. Supplementary data
Supplementary data associated with this article can be found,
in the online version, at http://dx.doi.org/10.1016/j.cattod.2016.03.
010.
References
[1] J. Lee, J. Kim, W. Choi, TiO2 photocatalysis for the redox conversion of aquatic
pollutants, in: P.G. Tratnyek, T.J. Grundl, S.B. Haderlein (Eds.), Aquatic Redox
Please cite this article in press as: J. Ryu, et al., Is surface fluorination of TiO2 effective for water purification? The degradation vs.
mineralization of phenolic pollutants, Catal. Today (2016), http://dx.doi.org/10.1016/j.cattod.2016.03.010

G Model
CATTOD-10104; No. of Pages7
ARTICLE IN PRESS
J. Ryu et al. / Catalysis Today xxx (2016) xxx–xxx
7
Chemistry, ACS Symposium Series, American Chemical Society, Washington,
DC, 2011, pp. 199–222.
[30] K.-i. Ishibashi, A. Fujishima, T. Watanabe, K. Hashimoto, Hashimoto.
Electrochem. Commun. 2 (2000) 207–210.
[2] H. Park, Y. Park, W. Kim, W. Choi, J. Photochem. Photobiol. C 15 (2013) 1–20.
[3] J. Schneider, M. Matsuoka, M. Takeuchi, J. Zhang, Y. Horiuchi, M. Anpo, D.W.
Bahnemann, Chem. Rev. 114 (2014) 9919–9986.
[31] M.-Q. Yang, Y. Zhang, N. Zhang, Z.-R. Tang, Y.-J. Xu, Sci. Rep. 3 (2013) 3314.
[32] C. Cai, H. Zhang, X. Zhong, L. Hou, Water Res. 66 (2014) 473–485.
[33] H. Bader, V. Sturzenegger, J. Hoigneı´, Water Res. 22 (1988) 1109–1115.
[4] O. Rojviroon, T. Rojviroon, S. Sirivithayapakorn, Environ. Eng. Res. 19 (2014)
255–259.
[5] G.-h. Moon, D.-h. Kim, H.-i. Kim, A.D. Bokare, W. Choi, Environ. Sci. Technol.
Lett. 1 (2014) 185–190.
[34] A. Yamakata, T.-a. Ishibashi, H. Onishi, J. Phys. Chem. B 105 (2001) 7258–7262.
[35] S. Kim, W. Choi, J. Phys. Chem. B 109 (2005) 5143–5149.
[36] D. Monllor-Satoca, R. Gómez, M. González-Hidalgo, P. Salvador, Catal. Today
129 (2007) 247–255.
[6] P. Gao, Z. Liu, M. Tai, D.D. Sun, W. Ng, Appl. Catal. B 138–139 (2013) 17–25.
[7] G. Jiang, Z. Lin, C. Chen, L. Zhu, Q. Chang, N. Wang, W. Wei, H. Tang, Carbon 49
(2011) 2693–2701.
[8] C. Andriantsiferana, E.F. Mohamed, H. Delmas, Environ. Eng. Res. 20 (2015)
181–189.
[9] J.A. Anderson, Catal. Today 181 (2012) 171–176.
[10] P. Panagiotopoulou, E.E. Karamerou, D.I. Kondarides, Catal. Today 209 (2013)
91–98.
[11] S. Kim, M. Kim, S.-H. Hwang, S.K. Lim, Appl. Catal. B 123-124 (2012) 391–397.
[12] H. Sheng, Q. Li, W. Ma, H. Ji, C. Chen, J. Zhao, Appl. Catal. B 138–139 (2013)
212–218.
[37] J. Kim, G.-h. Moon, S. Kim, J. Kim, J. Photochem. Photobiol. A 311 (2015) 35–40.
[38] M. Mrowetz, E. Selli, Phys. Chem. Chem. Phys. 7 (2005) 1100–1102.
[39] B.H.J. Bielski, D.E. Cabelli, R.L. Arudi, A.B. Ross, J. Phys. Chem. Ref. Data 14
(1985) 1041–1100.
[40] G.V. Buxton, C.L. Greenstock, W.P. Helman, A.B. Ross, J. Phys. Chem. Ref. Data
17 (1988) 513–886.
[41] H. Park, W. Choi, J. Phys. Chem. B 108 (2004) 4086–4093.
[42] C. Minero, G. Mariella, V. Maurino, D. Vione, E. Pelizzetti, Langmuir 16 (2000)
8964–8972.
[43] S. Jiang, Y. Li, X. Zhang, Y. Li, Catal. Today 259 (2016) 360–367.
[44] J. Kim, D. Monllor-Satoca, W. Choi, Energy Environ. Sci. 5 (2012) 7647–7656.
[45] W. Kim, T. Tachikawa, G.-h. Moon, T. Majima, W. Choi, Angew. Chem. Int. Ed.
53 (2014) 14036–14041.
[13] P. Dhatshanamurthi, B. Subash, M. Shanthi, Mater. Sci. Semicond. Process. 35
(2015) 22–29.
[14] C. Wang, C. Shao, X. Zhang, Y. Liu, Inorg. Chem. 48 (2009) 7261–7268.
[15] L.N. Dlamini, R.W. Krause, G.U. Kulkarni, S.H. Durbach, Appl. Water Sci. 1
(2011) 19–24.
[46] T. Fotiou, T. Triantis, T. Kaloudis, A. Hiskia, Chemosphere 119 (2015) S89–S94.
[47] N.M. Mahmoodi, M. Arami, N.Y. Limaee, N.S. Tabrizi, J. Colloid Interface Sci.
295 (2006) 159–164.
[16] A. Vijayabalan, K. Selvam, R. Velmurugan, M. Swaminathan, J. Hazard. Mater.
172 (2009) 914–921.
[17] M.V. Dozzi, E. Selli, Catal. Today 206 (2013) 26–31.
[18] T.K. Le, D. Flahaut, D. Foix, S. Blanc, H.K.H. Nguyen, T.K.X. Huynh, H. Martinez,
J. Solid State Chem. 187 (2012) 300–308.
[19] J. Kim, W. Choi, H. Park, Res. Chem. Intermed. 36 (2010) 127–140.
[20] J. Lee, W. Choi, J. Yoon, Environ. Sci. Technol. 39 (2005) 6800–6807.
[21] J. Ryu, W. Choi, Environ. Sci. Technol. 38 (2004) 2928–2933.
[22] J.H. Pan, Z. Cai, Y. Yu, X.S. Zhao, J. Mater. Chem. 21 (2011) 11430–11438.
[23] Y. Chen, F. Chen, J. Zhang, Appl. Surf. Sci. 255 (2009) 6290–6296.
[24] N. Wang, Z. Chen, L. Zhu, X. Jiang, B. Lv, H. Tang, J. Photochem. Photobiol. A
191 (2007) 193–200.
[25] J. Kim, W. Choi, Appl. Catal. B 106 (2011) 39–45.
[26] C. Chen, P. Lei, H. Ji, W. Ma, J. Zhao, H. Hidaka, N. Serpone, Environ. Sci.
Technol. 38 (2004) 329–337.
[48] S.K. Kansal, M. Singh, D. Sud, J. Hazard. Mater. 141 (2007) 581–590.
[49] N. Daneshvar, D. Salari, A.R. Khataee, J. Photochem. Photobiol. A 162 (2004)
317–322.
[50] A.R. Khataee, M.B. Kasiri, J. Mol. Catal. A: Chem. 328 (2010) 8–26.
[51] G. Mills, M.R. Hoffmann, Environ. Sci. Technol. 27 (1993) 1681–1689.
[52] X. Li, J.W. Cubbage, T.A. Tetzlaff, W.S. Jenks, J. Org. Chem. 64 (1999)
8509–8524.
[53] H. Kawaguchi, Chemosphere 22 (1991) 1003–1009.
[54] X. Li, J.W. Cubbage, W.S. Jenks, J. Org. Chem. 64 (1999) 8525–8536.
[55] L. Amalric, C. Guillard, P. Pichat, Res. Chem. Intermed. 21 (1995) 33–46.
[56] J. Kim, J. Lee, W. Choi, Chem. Commun. (2008) 756–758.
[57] D. Monllor-Satoca, R. Gómez, J. Phys. Chem. C 112 (2008) 139–147.
[58] H. Park, W. Choi, Langmuir 22 (2006) 2906–2911.
[59] M. Mrowetz, E. Selli, New J. Chem. 30 (2006) 108–114.
[60] M.V. Dozzi, G.L. Chiarello, E. Selli, J. Adv. Oxid. Technol. 13 (2010) 305–312.
[27] S. Rodriguez, A. Santos, A. Romero, Desalination 280 (2011) 108–113.
[28] X. Zhao, Y. Hou, H. Liu, Z. Qiang, J. Qu, Electrochim. Acta 54 (2009) 4172–4179.
[29] C. Minero, G. Mariella, V. Maurino, E. Pelizzetti, Langmuir 16 (2000)
2632–2641.
Please cite this article in press as: J. Ryu, et al., Is surface fluorination of TiO2 effective for water purification? The degradation vs.
mineralization of phenolic pollutants, Catal. Today (2016), http://dx.doi.org/10.1016/j.cattod.2016.03.010