Photocatalytic conversion of acetate into molecular hydrogen and hydrocarbons over Pt/TiO2: pH dependent formation of Kolbe and Hofer-Moest products

Article abstract of DOI:10.1016/j.jcat.2017.02.033

The photocatalytic generation of H₂ and hydrocarbons from aqueous acetic acid has been studied employing TiO₂ P25-based photocatalyst particles. The effect of pH on the distribution of reaction products and their formation rates during the photochemical as well as the photoelectrochemical oxidation of acetic acid has been investigated. The photocatalytic formation rates for H₂ and hydrocarbons were found to be higher for Pt-loaded TiO₂ than for bare TiO₂. The major reaction products resulting from the photocatalytic decomposition of aqueous acetic acid, as determined quantitatively in the gas phase, were found to be H₂, CO₂, and CH₄. Furthermore, traces of C₂H₆, C₃H₈, CO, CH₃CHO, HCHO, CH₃OH, C₂H₅OH, and HCOOH were also detected. After 15 h of illumination, the average formation rates of H₂, CO₂, CH₄, and C₂H₆ evolved at pH 2 were found to be 22, 65, 35, and 2 μmol/h, respectively. The ratio of H₂ to hydrocarbons strongly depends on the pH values, i.e., at acidic pH the ratio of H₂ to CH₄ formation was found to be 0.6. On the contrary, at neutral and basic pH values negligible amounts of hydrocarbons were formed and H₂ was found to be the main product with formation rates of 12 and 5 μmol/h at pH 7 and 11, respectively. It is therefore assumed that the hydroxide ion has a significant effect on the reaction pathways. Due to the fact that methanol and ethanol are formed as reaction products, water or hydroxide ions are apparently required for the formation of the major oxidizing agent, that is the hydroxyl radical. Herein, a detailed mechanism for the photocatalytic decomposition of acetic acid at different pH values is presented.

Full text of DOI:10.1016/j.jcat.2017.02.033

Journal of Catalysis 349 (2017) 128–135
Contents lists available at ScienceDirect
Journal of Catalysis
journal homepage: www.elsevier.com/locate/jcat
Photocatalytic conversion of acetate into molecular hydrogen and
hydrocarbons over Pt/TiO : pH dependent formation of Kolbe and
2
Hofer-Moest products
Saher Hamid a, , Irina Ivanova , Tae Hwa Jeon , Ralf Dillert , Wonyong Choi , Detlef W. Bahnemann
⇑
a
b
a
b
a,c,
⇑
a
Institut für Technische Chemie, Leibniz Universität Hannover, Callinstr. 3, D-30167 Hannover, Germany
Division of Environmental Science and Engineering, Pohang University of Science and Technology (POSTECH), Pohang 37673, Republic of Korea
Laboratory for ‘‘Photoactive Nanocomposite Materials”, Saint-Petersburg State University, Ulyanovskaya str. 1, Peterhof, Saint-Petersburg 198504, Russia
b
c
a r t i c l e i n f o
a b s t r a c t
Article history:
The photocatalytic generation of H2 and hydrocarbons from aqueous acetic acid has been studied
employing TiO P25-based photocatalyst particles. The effect of pH on the distribution of reaction prod-
ucts and their formation rates during the photochemical as well as the photoelectrochemical oxidation of
acetic acid has been investigated. The photocatalytic formation rates for H and hydrocarbons were found
to be higher for Pt-loaded TiO than for bare TiO . The major reaction products resulting from the pho-
tocatalytic decomposition of aqueous acetic acid, as determined quantitatively in the gas phase, were
found to be H , CO , and CH . Furthermore, traces of C , C , CO, CH CHO, HCHO, CH OH, C OH,
and HCOOH were also detected. After 15 h of illumination, the average formation rates of H , CO , CH
and C evolved at pH 2 were found to be 22, 65, 35, and 2 mmol/h, respectively. The ratio of H to hydro-
carbons strongly depends on the pH values, i.e., at acidic pH the ratio of H to CH formation was found to
be 0.6. On the contrary, at neutral and basic pH values negligible amounts of hydrocarbons were formed
and H was found to be the main product with formation rates of 12 and 5 mmol/h at pH 7 and 11, respec-
Received 21 December 2016
Revised 27 February 2017
Accepted 28 February 2017
2
2
2
2
Keywords:
Photocatalysis
Acetate
Platinized titania
Surface charge
Solar fuel
2
2
4
2
H
6
3
H
8
3
3
2 5
H
2
2
4
,
2
H
6
2
2
4
2
tively. It is therefore assumed that the hydroxide ion has a significant effect on the reaction pathways.
Due to the fact that methanol and ethanol are formed as reaction products, water or hydroxide ions
are apparently required for the formation of the major oxidizing agent, that is the hydroxyl radical.
Herein, a detailed mechanism for the photocatalytic decomposition of acetic acid at different pH values
is presented.
Ó 2017 Elsevier Inc. All rights reserved.
1
. Introduction
organic acids [17–33] act as electron donors/sacrificial reagents,
2
while promoting the reduction reaction, i.e., the formation of H .
Energy needs and the corresponding environmental issues are
Most studies employing an organic acid as the electron donor have
used acetic acid which is a common organic compound present in
industrial wastes [3,34,35]. The transformation of acetic acid pre-
sent in wastewater into molecular hydrogen allows the combina-
tion of a necessary water treatment with the desired production
of a valuable fuel. The decomposition of acetic acid by means of
photocatalysis into useful fuels is a breakthrough as it does not
require expensive equipment. The use of acetic acid as an electron
an increasing concern in the present age. Since the last three dec-
ades, fuel production by the photocatalytic reforming of aqueous
organic compounds has been in the focus of attention for scientists
in order to meet the world’s energy requirements. The photocat-
alytic conversion of organic matter into molecular hydrogen (H
is considered as an ideal solution to address both energy and
related environmental problems. Among various methods for H
2
)
2
formation, photocatalytic reaction systems using inorganic semi-
conductors have widely been investigated [1–4]. In these reaction
systems, various organic compounds such as alcohols [5–16] and
donor in photocatalytic reaction results in the formation of H
2
and
valuable hydrocarbons (e.g., CH , C H ), which can be further uti-
4 2 6
lized as renewable solar fuels.
A series of reports have been published by Krauetler and Bard
18,19,36,37] discussing the photoelectrochemical and photocat-
[
⇑
Corresponding authors at: Institut für Technische Chemie, Leibniz Universität
alytic conversion of acetic acid into H
2
and hydrocarbons over pla-
Hannover, Callinstr. 3, D-30167 Hannover, Germany (D.W. Bahnemann).
E-mail addresses: hamid@iftc.uni-hannover.de (S. Hamid), bahnemann@iftc.
uni-hannover.de (D.W. Bahnemann).
tinized titania (Pt/TiO P25) in acidic media. These authors
2
http://dx.doi.org/10.1016/j.jcat.2017.02.033
0
021-9517/Ó 2017 Elsevier Inc. All rights reserved.

S. Hamid et al. / Journal of Catalysis 349 (2017) 128–135
129
observed methane (CH
reaction products, while H
product in their studies. Yoneyama et al. investigated the transfor-
mation of acetic acid in suspensions containing Pt/TiO (rutile and
anatase) at different pH values ꢀ 7.5. There data show that CH is
the main reaction product in the presence of Pt-loaded anatase
in this pH regime but the H /CH ratio increased with increasing
pH [20]. Sakata et al. investigated the photocatalytic decomposi-
tion of aqueous acetic acid over Pt/TiO (rutile and anatase) at
pH values between 2.1 and 13 and observed significant amounts
of H , CO , hydrocarbons, and alcohols as reaction products in
acidic pH medium. On the other hand, the data analysis of these
authors revealed an increase in the formation rate for H and alco-
4
) and carbon dioxide (CO
2
) as the main
2.2. Photocatalytic measurements
2
was found to be the minor reaction
The photocatalytic measurements for the evolution of hydrogen
and hydrocarbons have been carried out in a 100 cm double jacket
3
2
4
quartz glass reactor having two flanges (one as inlet and the other
as outlet) being connected to a quadrupole mass spectrometer
(QMS, Hiden HPR-20) [10]. The system is a continuous gas flow
system, where argon was used as the carrier gas. A schematic illus-
tration of experimental setup is shown in Fig. S1. In a typical exper-
imental run, 0.05 g of the photocatalyst was suspended in 50 mL of
water containing aqueous acetic acid solution (0.5 M) by sonica-
tion. The suspension was then transferred into the photoreactor
and purged with Ar for about 20 min to remove molecular oxygen
present in the headspace of the reactor and in the suspension. The
2
4
2
2
2
2
hols when the pH was changed toward more basic values, while
hydrocarbons stopped being evolved [21]. Zheng et al. examined
the effect of various pH values on the photocatalytic transforma-
3
ꢃ1
flow rate of Ar gas was kept constant at 10 cm min throughout
the photocatalytic experiment. The actual inlet gas flow rate for
3
ꢃ1
tion of aqueous acetic acid into H
29] as well as CuO/SnO [30] as the photocatalysts. The authors
demonstrated a significant formation rate for H in acidic pH media
only. However, they observed a decreasing trend in the formation
rate of H as the pH value was increased toward the neutral regime
while nearly no H
formation was observed at pH ꢁ 7.
2
employing Pt/TiO
2
(anatase)
QMS was 1 cm min . Before starting the illumination, the system
was kept in the dark for 60 min to ensure a stable baseline. Subse-
quently, the reactor was illuminated employing an Osram XBO
1000 Watt Xenon Arc Lamp in a Müller LAX 1000 lamp housing
for 4–15 h. The temperature of the suspension was kept constant
at 25 °C during the entire time of the experimental run.
[
2
2
2
2
Due to these conflicting reports, we have decided to re-
investigate the influence of the pH on the photocatalytic decompo-
sition of aqueous acetic acid regarding the formation rates of the
main reaction products and the distribution of these products.
Although some reports have been published concerning the photo-
catalytic transformation of aqueous acetic acid, the exact details of
the reaction mechanism are still unclear. Here, the reaction mech-
anism for photocatalytic decomposition of aqueous acetic acid via
direct or indirect pathways is proposed. Moreover, quantum effi-
ciencies for the main reaction products are also presented.
2
.3. Photoelectrochemical measurements
Photoelectrochemical performances were measured by a three-
electrode electrochemical Pyrex cell using a potentiostat (Versastat
-400). As-coated TiO (or Pt/TiO ) electrode (working electrode),
Pt wire (counter electrode), and an Ag/AgCl electrode (reference
electrode) were immersed in NaClO solution (0.5 M; pH adjusted
to 2 or 9 by HClO or NaOH). The concentration of acetic acid was
3
2
2
4
4
adjusted to 0.5 M. A constant potential of 0 V (vs. Ag/AgCl) was
applied to the working electrode. Electrolyte was purged with Ar
gas for 15 min before any measurements, and during the measure-
ments purging was kept in headspace. The coated front side of the
2
. Experimental
working electrode was illuminated by a 150 W Xenon arc lamp
ꢃ2
(
ABET Tech.) with an air mass (AM) 1.5 G filter (100 mW cm ).
2.1. Materials
For slurry photocurrent measurements, a Pt wire working elec-
trode, a graphite rod counter electrode, and an Ag/AgCl reference
electrode were immersed in 0.5 M NaClO
Commercial titanium dioxide powder (Evonik Aeroxide TiO
2
4
solution containing
P25) and 1 wt% platinum-loaded P25 powders (prepared by an
impregnation method employing hexachloroplatinic acid (H PtCl
O) at H. C. Starck, Goslar, Germany) were used as received.
Briefly, 20 g TiO P25 powder was introduced into a conical flask
and 100 mL deionized water was added followed by the addition
of 4.45 mL H PtCl solution (8.9% Pt). After that, the conical flask
3+
1
g/L photocatalyst powder, 1 mM Fe , and 0.5 M acetic acid.
2
6
.
4
The pH was adjusted to 2 by HClO . Photocurrents were collected
to the Pt wire working electrode at a bias potential of 0.7 V (vs.
Ag/AgCl) under AM 1.5G illumination (100 mW cm ).
6H
2
2
ꢃ2
2
6
was sealed properly with a rubber stopper and placed in a ultra-
sonic bath for 6 h. The solution was then stirred and the solvent
was evaporated.
3. Results
The influence of the initial pH value on the formation rates of
the reaction products resulting from the light-induced conversion
of deaerated aqueous acetic acid has been investigated employing
The electrodes were prepared on electrically conductive FTO
substrates by the doctor blade method. 0.2 g bare TiO
Pt loaded TiO powder was mixed with 1 mL solution of carbowax
in a 1:1 ratio of polyethylene glycol and water). The mixture was
2
or 1 wt%
2
TiO
2
loaded with 1 wt% Pt as the photocatalyst. In order to confirm
surface and to study
(
the significance of the Pt-loading on the TiO
2
ground for 10 min until the solution was well dispersed and sticky.
the role of the sacrificial reagent in the photocatalytic reaction sys-
tem, some blank experiments were performed. These experiments
were carried out (a) under UV(A) irradiation in the presence of the
A few drops of the mixture were spread on the FTO substrate
2
(
area = 1.5 ꢂ 2.0 cm ) and coated via the doctor blading technique.
The as-coated electrodes were dried in air and annealed at 450 °C
for 30 min under Ar to enhance the adhesion between the particles
or the particles and the substrate and to remove any remaining
organic binder.
photocatalyst (TiO
diation in the presence of the co-catalyst and aqueous acetic acid,
and (c) in the presence of the TiO photocatalyst and aqueous
2
) suspended in water only, (b) under UV(A) irra-
2
acetic acid in the dark. The formation rates for molecular hydrogen
and other reaction products were found to be negligible in all of
these three blank experiments. On the other hand, UV(A) irradia-
4
Perchloric acid (HClO ) and NaOH were used for the initial
adjustment of the pH values. All materials were purchased from
Sigma-Aldrich and used as received without further purification.
Deionized water was used from a Sartorius Arium 611 apparatus
tion of suspensions of TiO
acid was found to result in the formation of H
2
and Pt-loaded TiO
2
in aqueous acetic
and hydrocarbons
2
(
resistivity = 18.2 M
X
cm) for the preparation of all aqueous
(Fig. 1), clearly indicating the photocatalytic transformation of
the organic compound. 0.5 M acetic acid and 1 wt% loading of the
solutions.

1
30
S. Hamid et al. / Journal of Catalysis 349 (2017) 128–135
1
00
100
5
2
0
.0
.5
.0
(a) TiO
Light on
(b) Pt/TiO
2
2
CO
2
8
6
4
2
0
0
0
0
0
80
60
40
20
0
CH
4
CO
CH
2
H2
Light off
4
0
200
400
H
2
Light on
Light off
0
100
200
300
400
0
200
400
600
800
1000 1200
Time / min
Time / min
Fig. 1. Photocatalytic formation of CO
2
, H
2
, and CH
4
from aqueous acetic acid employing (a) TiO
2
(P25) only and (b) 1 wt% Pt/TiO
2
(P25). Experimental conditions: photocatalyst
ꢃ1
concentration = 0.5 g L , initial acetic acid concentration = 0.5 M, initial pH 2, suspension volume = 50 mL, irradiation time = 4 h (a) and 15 h (b), irradiation intensity
ꢃ2
I250-I450 = 30 mW cm
.
co-catalysts were found to be the optimum conditions for all
experimental runs (Fig. S2).
quantum efficiencies are reported (Fig. 2). After 15 h of illumina-
tion, the average formation rates of CO , H , CH , and C evolved
2
2
4
2 6
H
at pH 2 were found to be 65, 22, 35, and 2 mmol/h, respectively.
The comparison of the ratios of the atoms constituting the reac-
3
.1. pH dependency of the reaction product distribution
tant CH
quantitatively determined reaction products (i.e., CO
and C ), which were found to be H/C = 1.9 and O/C = 1.3 at pH
, clearly indicates the formation of other reaction products. In fact,
thorough analysis of the data recorded by the QMS confirmed the
presence of CO, C , CH OH, C OH, HCHO, CH CHO, and
3
COOH (H/C = 2.0, O/C = 1.0) and the atom ratios of the
2
, H , CH ,
2
4
Plots of the formation rates vs. the irradiation time are pre-
2 6
H
sented in Fig. 1 for an experimental run performed at an initial
pH of 2. Prior to the start of the illumination, the concentrations
of all gaseous compounds were monitored for 60 min until a stabi-
lization of their respective signals was achieved. Thereafter, the
lamp was switched on thus initiating the photocatalytic conversion
of aqueous acetic acid into the reaction products including CO
and several alkanes C
takes almost 30–40 min (depending on the reaction conditions) to
reach the maximum observed formation rate. The main reaction
products resulting from the light-induced conversion of aqueous
acetic acid at pH 2 in the presence of the photocatalyst were found
2
H
3 8
3
2
H
5
3
HCOOH in the gas phase exiting the photoreactor (Fig. S3).
An increase in the initial pH value from pH 2 to pH 5 resulted in
a slight but significant decrease in the average formation rates
2 2
, H ,
x
H2x+2 with x ꢀ 3. For all reaction products it
(
and, consequently, of the average quantum efficiencies) of all
reaction products without affecting the type of reaction products
as shown in Figs. 2 and S3. This means, conversely, that a low pH
value favorably affects the rate of the light-induced conversion of
acetic acid in the presence of Pt/TiO
ratio between the rates of H and CH
indicating that the formation of CH is preferred.
No CO was found in the gas phase exiting the reactor at pH val-
2
. At acidic pH conditions, the
2 2 4
to be CO , H , and CH as shown in Fig. 1. It is also obvious from the
results shown in Fig. 1a and b that considerable amounts for all
reaction products are only obtained when Pt is deposited on
2
4
formation is <1 (Table 1)
4
2
2 2
TiO . Photocatalytic processes on bare TiO do show the formation
2
ꢃ
ues ꢁ 7 due to its conversion into carbonate (CO
3
) and bicarbon-
of the main reaction products; however, the produced amounts are
comparatively low. It is obvious from Fig. 1b that the formation
rates decrease as the illumination time is increased. Consequently,
the total time of illumination was also found to be a key factor for
the determination of the overall formation rate. Here, formation
rates averaged over the entire irradiation time as well as averaged
ꢃ
ate (HCO
3
) which remains in the aqueous suspension at these pH
values [38]. Fig. 2 also demonstrates that the increase in the pH
value results in an increase in both, the r(H )/r(CH ) and the r
)/r(C ) ratios (Table 1). Consequently, the formation of alka-
nes is suppressed at pH ꢁ 7 and H becomes the main gaseous
reaction product of the light-induced conversion of aqueous acetic
acid in the presence of 1 wt% Pt/TiO
2
4
(
H
2
2 6
H
2
2
.
pH 2
1
00
pH 5
pH 7
pH 9
pH 11
10
1
2 2
3.2. Photoelectrochemical performances of TiO and Pt/TiO
1
0
The photoelectrochemical oxidation of acetate on bare TiO
on Pt/TiO was investigated using catalyst-coated electrodes in an
aqueous electrolyte at pH 2 and pH 9, respectively. At any pH con-
dition, photocurrents of TiO and Pt/TiO electrodes were highly
2
and
1
2
0
.1
2
2
0.1
enhanced in the presence of acetic acid thus indicating the
0
.01
0.01
1
E-3
E-4
Table 1
Effect of the pH on the ratio of the formation rates of the main reaction products
formed from aqueous acetic acid in the presence of 1 wt% Pt/TiO .
2
1
CO
H
CH
4
C H
2 6
2
2
pH
r(H
2 2
)/r(CO )
r(H
2 4
)/r(CH )
2 2 6
r(H )/r(C H )
2
5
7
9
0.3
0.3
–
–
–
0.6
0.7
2.6
4.8
70
17
26
250
329
700
Fig. 2. Photocatalytic formation of the main reaction products, i.e., CO
2
, H
2
4
, CH and
C
2
H
6
from aqueous acetic acid employing 1 wt% Pt/TiO
2
(P25). Experimental
ꢃ1
conditions: photocatalyst concentration: 0.5 g L , initial acetic acid concentration:
0
.5 M, initial pH values: 2, 5, 7, 9 and 11 (adjusted by NaOH), suspension volume:
ꢃ2
11
5
0 mL, irradiation time: 15 h, irradiation intensity I250-I450 = 30 mW cm
.

S. Hamid et al. / Journal of Catalysis 349 (2017) 128–135
Table 2
131
photoelectrochemical oxidation of acetic acid (Fig. 3a and b). The
photocurrents for acetic acid oxidation on bare TiO and on Pt/
TiO electrodes were slightly higher at pH 2 than at pH 9
Fig. 3c). Under basic conditions (pH 9), the electrostatic repulsion
between the negatively charged TiO surface and the acetate anion
pK
ꢄ 4.8) [39] may retard the photoelectrochemical oxidation
process.
Possible reaction pathways for the photocatalytic transformation of organic com-
pounds and transient radical intermediates present on the photocatalyst surface
during the photocatalytic transformation of aqueous acetic acid.
2
2
(
ꢃ
+
Reactions of photogenerated charge carriers (e and h )
2
þ
PC þ h
m
! PCðh þ eꢃ
Þ
(1)
(2)
(
a
Pt
Å
ꢃ
þ
PCðe Þ þ H ! PC þ H
þ
Å
þ
PCðh Þ þ CH
3
COOH ! PC þ CH
3
COO þ H
(3)
(4)
(5)
(6)
þ
Å
þ
PCðh Þ þ C
2
H
5
COOH ! PC þ C
2
þ
H
5
COO þ H
þ
Å
PCðh Þ þ H
2
O ! PC þ OH þ H
4
. Discussion
þ
ꢃ
Å
PCðh Þ þ OH ! PC þ OH
Å
Å
Formation of CH
3
and C
þ CO
C H COO ! C H þ CO
2 5
H
4.1. Fundamental processes for the photocatalytic acetic acid
Å
Å
CH
3
COO ! CH
5
3
2
(7)
(8)
Å
Å
transformation
2
2
5
2
Formation of molecular hydrogen and alkanes (Kolbe-products)
It is well known that the initial step of the light-induced decom-
position of an aqueous organic compound in the presence of a
semiconductor photocatalyst (PC) material is the excitation of an
electron from its valence band into the conduction band resulting
in the formation of separated charge carriers, i.e., conduction band
electrons and valence band holes. These charge carriers can either
recombine or they react with molecules and ions being present at
the photocatalyst surface [4]. The formation of charge carriers (Eq.
Å
Å
Pt
(9)
H þ H ! H
2
Å
Å
Å
Å
Å
CH
3
þ H ! CH
4
(10)
(11)
(12)
(13)
Å
CH3 þ CH ! C H
3
2
6
Å
C
C
2
H
H
5
þ H ! C
2
H
6
Å
2
5
þ CH
3
! C
3 8
H
Formation of C
2
H
5
COOH
Å
Å
CH
3
COOH þ OH ! CH
2
COOH þ H
2
O
(14)
(15)
Å
Å
CH
2
COOH þ CH
3
! C
2
H COOH
5
Formation of alcohols (Hofer-Moest products)
(
1)), their subsequent reactions with species present on the photo-
Å
Å
Å
OH þ CH
3
! CH
3
OH
OH
(16)
(17)
catalyst surface during irradiation (Eqs. (2)–(6)), as well as the
likely reactions of the assumed transient radical intermediates
formed during the photocatalytic transformation of acetic acid/
acetate (Eqs. (7)–(17)) are presented in Table 2.
Å
OH þ C
2
H
5
! C
2
H
5
Å
In the absence of molecular oxygen and in the simultaneous
presence of a suitable co-catalyst, conduction band electrons are
transferred to these co-catalysts where they are reducing
hydrogen atoms, H , (Eq. (2)). Generally, bare TiO
photocatalytic activity for H
electron donors [8,9]. Therefore, co-catalysts are usually deposited
2
shows negligible
formation even in the presence of
2
+
protons (H ) resulting initially in the formation of adsorbed
onto the TiO
2
surface acting as efficient electron traps that
1
1
0
0
0
0
.5
.2
.9
.6
.3
.0
25
light on off
on
off
on
off
pH 2
pH 9
(
a)
(b)
20
15
10
5
TiO
2
w/ acetic acid
Pt/TiO
2
w/ acetic acid
TiO
2
Pt/TiO
2
0
0
100
200
300
400
Pt/TiO
TiO
2
2
Time / s
3
2
2
1
1
0
0
.0
.5
.0
.5
.0
.5
.0
(
c)
TiO
2
Pt/TiO
2
TiO w/ acetic acid
2
Pt/TiO w/ acetic acid
2
0
200
400
600
800 1000 1200 1400
Time / s
Fig. 3. Photocurrent generation on TiO
2
and Pt/TiO
2
electrodes (a) at a potential of 0 V vs. Ag/AgCl at pH 2. Experimental conditions: [NaClO
4
] = 0.5 M, [acetic acid] = 0.5 M,
and Pt/TiO at pH 2
particles. Experimental conditions: [catalyst] = 1 g/L, [Fe ] = 1 mM, [acetic acid]
0.5 M, pH 2, Pt collector electrode at an applied potential of 0.7 V vs. Ag/AgCl, illumination: AM 1.5 G.
illumination: AM 1.5 G, pH was adjusted by HClO
4
or NaOH. (b) Steady-state photocurrent ratio R = (Iph with acetic acid)/(Iph without acetic acid) of TiO
2
2
and 9. (c) Fe3 -mediated photocurrent-time profiles in the suspension of TiO
+
and Pt/TiO
2 2
3+
=

1
32
S. Hamid et al. / Journal of Catalysis 349 (2017) 128–135
+
facilitate the reduction of H into H
2
to a great extent. Loading the
given some experimental evidence that this reaction occurs in
acidic suspensions containing acetic acid, Pt/TiO , and an efficient
electron donor such as Pd or Ag [20,21]. Despite the fact that
the added electron donor suppressed the formation of H , these
authors observed the evolution of CH and thus concluded that a
photocatalyst with a co-catalyst, such as Pt, results in an increase
in the rate of the electron transfer reaction. Under the experimen-
tal conditions employed in this study, the rate of H
2
2
+
+
2
formation was
with Pt
2
found to increase by a factor of 12 by loading the TiO
2
+
4
islands. It is, therefore, assumed that the reduction of H mainly
occurs on the Pt islands which requires that the electrons present
in the conduction band of the semiconductor are rapidly trans-
reaction according to Eq. (19) does indeed take place.
Since H-atom abstraction reactions from organic solutes by HÅ
are known to have usually higher rate constants in homogeneous
Å
ferred to the Pt islands present on the TiO
2
surface.
solutions than the corresponding reactions by CH
according to
3
, reactions
The valance band holes are able to react with both, the organic
ꢃ
solutes adsorbed at the TiO
2
surface and with H
2
O/ OH present at
Å
Å
H þ CH
4
! H
2
þ CH
3
ð20Þ
ð21Þ
ð22Þ
the photocatalyst surface (Eqs. (3)–(6)) [4]. The only organic solute
initially present in the system investigated here is acetic acid/acet-
ate which is known to be strongly adsorbed at the TiO
Å
Å
H þ C
2
H
6
! H
2
þ C
2 5
H
2
surface in
ꢃ
acidic suspension [40–42]. Hole oxidation of CH
3
3
COOH/CH COO
Å
Å
H þ CH
3
COOH ! H
2
þ CH
2
COOH
Å
yields the acetoxy radical, CH
poses into CO
3
COO , which immediately decom-
Å
2
and the methyl radical, CH
3
(Eqs. (3) and (7)).
as proposed by Sakata et al. [21] also have to be considered. Rate
Å
5
6
4
ꢃ1 ꢃ1
EPR studies have indeed revealed the presence of CH
light-induced transformation of CH
ence of TiO -based photocatalysts [42–47].
Meyerstein and co-authors have recently shown that CH
duced by pulse radiolysis of aqueous suspensions containing
dimethyl sulfoxide) rapidly react with both TiO and Pt resulting
3
during the
COO in the pres-
constants of <1 ꢂ 10 , 2.3 ꢂ 10 , and 9.8 ꢂ 10 L mol
s
have
ꢃ
3
COOH/CH
3
been reported for the three reactions given in Eqs. (20)–(22) in
homogeneous solutions, respectively [56].
2
Å
Å
Å
ꢃ
3
(pro-
2 2
The formation of CH COOH/ CH COO probably occurs pre-
ꢃ
Å
dominantly by the reaction between CH
(Eq. (14)); the reactions with CH
3
COOH/CH
(Eq. (19)) and H (Eq. (22)) are
3
Å
COO and OH
Å
2
3
in the methylation of the respective surfaces [48–52]. The rate con-
stants of these reactions have been found to approach the
at best insignificant side reactions.
The formation of propane, which has been detected as a minor
reaction product, might be explained by a sequence of radical
8
9
ꢃ1 ꢃ1
diffusion-controlled limit (10 ꢅ ꢅ ꢅ10 L mol
s
). On bare TiO
2
as
Å
well as on Pt two surface bound CH
3
recombine yielding C
2
H
6
.
forming and radical-radical recombination processes. The forma-
Å
with ÅCH
On Pt with preadsorbed H
2
the formation of methane was observed
tion is probably initiated by the reaction of CH
3
2
-
Å
ꢃ
by these authors. Higher amounts of hydrogen adsorbed at the Pt
surface increased the yield of methane and decreased the yield of
2
COOH/ CH COO resulting in the formation of propanoic acid (Eq.
(15)) which is subsequently oxidized by holes yielding CO
2
and
Å
ethane [52]. Therefore, we assume that a significant amount of
2 5
C H (Eqs. (4) and (8)). The latter radical is assumed to react with
Å
Å
Å
CH
3
(formed by reaction Eq. (7)) migrates to the Pt islands by sur-
both H and CH
3 2 6 3 8 2 6
yielding C H and C H (Eqs. (12) and (13)). C H
face diffusion. At the Pt islands two competing reactions, i.e., the
has been identified as the main reaction product of the photocat-
Å
dimerization of two H yielding H
2
(Eq. (9)) and the reaction
alytic transformation of propanoic acid in O
2
-free suspensions of
Å
Å
Å
Å
between H and CH
3
yielding CH
4
(Eq. (10)), occur during the pho-
Pt/TiO [21], indicating that the reaction of C
2
2
H
5
with H (Eq.
Å
tocatalytic transformation of acetic acid in the absence of O
2
. The
(12)) is preferred compared to a reaction of C
alkyl radical.
H
2 5
with another
Å
competing formation of C
Eq. (11)) might occur on the TiO
The low formation rates of C
CH )/r(C ) ratios which can be calculated from the data given
2
H
6
by the dimerization of two CH
3
(
2
surface or at the Pt islands.
(Fig. 2) as well as the high r
In addition to H
methanol and ethanol as well as some of the known products of
the photocatalytic transformation of these alcohols in the absence
2
and the hydrocarbons C
x
H
2x+2 with x ꢀ 3,
2
H
6
(
4
2 6
H
in Table 1 (ꢆ1 at all pH values in accordance with published data
of O
detected as reaction products in the gas phase during the photocat-
alytic transformation of acetic acid. Earlier, CH OH, C OH,
CH CHO, (CH CO, and CH COOCH have been detected as by-
products of the photocatalytic transformation of CH COOH in the
2 3
, i.e., HCHO, HCOOH, CH CHO, and CO [5–7,13,14], have been
Å
[
20,21,42]) reveal that the dimerization of CH
3
is only a minor
Å
reaction, while the CH
4
formation by the reaction between CH
3
3
2 5
H
Å
Å
and H (Eq. (10)) appears to be the predominant reaction of CH
in acidic suspension.
3
3
3
)
2
3
3
3
ꢃ
Å
H-atom abstraction from CH
3
COOH/CH
3
COO by OH (formed
O/ OH, Eqs. (5) and (6)) results in the for-
liquid phase [21,21,24–27]. The presence of the alcohols can be
ꢃ
by hole oxidation of H
2
positively explained by reactions of the corresponding alkyl radi-
Å
Å
ꢃ
Å
mation of CH
2
COOH/ CH
2
COO (Eq. (14)). The formation of this
cals with OH at the TiO
2
surface (Eqs. (16) and (17)). But, it has
Å
radical during the light-induced transformation of acetate in the
presence of TiO has been proven by EPR [42,44–46].
It has been shown that H-atom abstraction from CH
OH in homogeneous aqueous solution occurs exclusively at
the -position with pH-dependent reaction rate constants
1 ꢂ 10 L mol
by hydrogen abstraction from CH
been mentioned above that the reaction of CH
3
with TiO
2
results
2
in the methylation of the oxide surface [48]. Therefore, the forma-
tion of methanol and other hydroxylated organic compounds
3
COOH by
Å
might also occur via the hydrolysis of the chemical bond fixing
Å
a
an alkyl radical CH
2
R (R = H, CH
3
) on the TiO
2
surface:
9
ꢃ1 ꢃ1
Å
ꢁ
s
[53]. Therefore, the formation of CH
COOH according to
3
COO
BTiAOACH
2
R þ H
2
O ! BTiAOH þ RCH
2
OH
ð23Þ
3
Å
Å
ꢃ
ꢃ
CH
3
COOH þ OH ! CH
3
COO þ H
2
O
ð18Þ
BTiAOACH R þ OH ! BTiAO þ RCH OH
ð24Þ
2
2
can most likely be excluded.
It has been reported by Kandiel et al. that the CH
results in the formation of HCHO (HCH(OH)
then further transform into H and CO as the main reaction prod-
3
OH oxidation
2
) and HCOOH which
2
2
ꢃ1 ꢃ1
Rate constants of 2 ꢂ 10 and 6 ꢂ 10 L mol
s
have been
in homo-
Å
reported for H-atom abstraction from acetic acid by CH
3
2
2
geneous aqueous solutions [54,55]. Therefore, the formation of CH
4
ucts [6,7]. We have also detected traces of HCHO and HCOOH as
by-products in the gas phase, thus evincing that the decomposition
ꢃ
by hydrogen abstraction from CH
3
COOH/CH
3
COO according to
Å
Å
of acetic acid results in the formation of CH
3
OH, which is then oxi-
CH
3
COOH þ CH
3
! CH
2
COOH þ CH
4
ð19Þ
+
+
ꢃ
dized into H and CO
contribute either to the formation of H
CH . C OH being formed in a sequence of reactions given in
2
. These H (after reduction by e ) can also
as proposed by several research groups [20,21,45] seems also to be
unlikely. However, Yoneyama et al. as well as Sakata et al. have
2
or to the formation of
4
2 5
H

S. Hamid et al. / Journal of Catalysis 349 (2017) 128–135
133
ꢃ
Eqs. (14), (15), (8), and (17) is subsequently oxidized mainly to H
and CO following reported reaction pathways. This clearly shows
that the oxidation of C
2
hole oxidation of H
2
O/ OH, Eqs. (5) and (6)) with the methyl group
Å
ꢃ
2
of the acetate anion yielding CH COO (Eq. (14)). The hydrogen
2
Å
H
2 5
OH via OH results in the formation of
evolution that has been observed to occur in alkaline suspension
(Fig. 2 and Table 1) might, therefore, be mostly the result of a reac-
tion sequence involving the anions of glycolic acid (hydroxyacetic
acid) and glyoxylic acid (oxoacetic acid):
CH
CH
3
CHO which is further oxidized giving rise to the formation of
and CO [5,13,14].
4
It is well understood that the distribution of the products
resulting from the photocatalytic transformation of CH
3
COOH/
Å
Å
ꢃ
ꢃ
ꢃ
OH þ CH COO ! HOCH COO
ð25Þ
3
CH COO (as well as their respective formation rates, see below)
2
2
strongly depends on the pH of the suspension. Considering the
acid-base properties of TiO
having a pHzpc ꢄ 6.25 (P 25 [57]) as
well as the dissociation constant of acetic acid (pK = 4.75 [39])
three different situations can be distinguished (Scheme 1):
Å
ꢃ
or in case that CH
methyl radical
2
COO reacts with the TiO
2
surface like the
2
a
ꢃ
ꢃ
ꢃ
ꢃ
BTiAOACH
2
COO þ OH ! BTiAO þ HOCH
2
COO
ð26Þ
ð27Þ
(
a 2
i) At pH values lower than the pK of acetic acid, the TiO sur-
ꢃ
HOCH
2
COO ! CHOCOOꢃ þ H
Glycolic acid that has been proposed to be an intermediate dur-
ing the photocatalytic transformation of acetic acid in the absence
2
face is positively charged while acetic acid is present in its
protonated form. Therefore, the organic solute only interacts
with the positively charged TiO
atively charged oxygen atoms of the carboxyl group.
ii) Under pH conditions where pK < pH < pHzpc holds the
2
surface via the slightly neg-
of O
Methanol is known to be photocatalytically transformed into CO
and H with a H formation rate being nearly unaffected by the
pH of the suspension [7].
2
[21] might behave similar to an alcohol, such as methanol.
(
a
2
organic acid is present in the suspension in its deprotonated
form. Consequently, the negatively charged acetate ion is
2
2
strongly attracted by the positively charged TiO
iii) At pH values higher than the pHzpc of TiO the negatively
charged carboxylate group of the acetate ion is strongly
interacting with the likewise negatively charged TiO sur-
2
surface.
The small amounts of hydrocarbons evolved during the photo-
catalytic acetate transformation suggest that the direct reaction
between holes and acetate (Eq. (3)) is only of minor importance
in alkaline suspension. But it cannot be excluded that CH
at pH > pHzpc reacts with OH in a fast reaction. The thus formed
(
2
2
Å
3
formed
face resulting in its repulsion. Hence, the acetate ions will
be adsorbed through the methyl group thus facilitating any
radical attack at this side of the molecule.
Å
2ꢃ
CH
HCO
3
OH might rapidly be further oxidized yielding H
in a sequence of reactions involving the formation of HCHO,
, and HCOOH as outlined by Kandiel et al. [6].
2
and CO
3
/
ꢃ
3
HCH(OH)
2
Adsorption of the organic solute is generally considered to be a
The products but not the product distribution found in this
investigation conform with those observed during the electrolysis
of acetate in aqueous solution. This electrosynthetic organic reac-
tion results in the formation of hydrocarbons (Kolbe reaction) or
alcohols (Hofer-Moest reaction) depending on the conditions
employed during the electrolysis [58–60].
prerequisite for the direct hole oxidation of the solute. The above
discussion concerning the interaction between the TiO
2
surface
ꢃ
3 3
and CH COOH/CH COO shows that the condition of adsorption
of the organic solute through the carboxylate group is fulfilled only
at pH values lower than the pHzpc of TiO
2
. Therefore, reaction prod-
Å
ucts formed via CH
3
being the intermediate radical of direct hole
In acetic suspensions the main reaction product of the photo-
catalytic transformation of acetic acid observed in the gas phase
was found to be methane accompanied by smaller amounts of
oxidation of the carboxylate group are expected to be formed in
significant amounts only in acidic suspensions. This is supported
by experimental data reported by Nosaka et al. and by Yoneyama
ethane and propane. The formation of these Kolbe products is
et al. The data obtained with Pt/TiO
2
in suspensions containing
Å
ꢃ1
resulting from the common intermediate CH
3
which is formed at
0
.1 mol L acetic acid enabled Nosaka and co-worker to calculate
the photocatalyst surface by the light-induced Kolbe decarboxyla-
tion of acetic acid. Our data analysis depicts that alkanes are
formed as the main products at pH 2 and 5 where the Kolbe decar-
boxylation is apparently the dominant reaction path. The simulta-
neous formation of alcohols reveals that the decomposition of
acetic acid not only includes the Kolbe reaction but also proceeds
via the Hofer-Moest reaction. Even though no alcohols were
detected in the gas phase at pH 9, the significant formation rate
the ratios between the photocatalytically generated amounts of
Å
Å
CH
3
and CH
2
COOH. The reported data show that 75% of the acetic
P25 is
acid is directly oxidized by holes when platinized TiO
2
employed as the photocatalyst [42]. Yoneyama et al. measured
Å
the highest CH
3
formation rate in suspensions containing acetic
acid and Pt/TiO
pH 7.5 resulted in a decrease of the CH
2
at pH 3.9. Shifting the pH of the suspension to
Å
3
formation rate by more
than 90% [20].
for H
2
as the main reaction product indicates the formation of
If the pH value is higher than the pHzpc, the product formation is
Å
(
most likely dissolved) CH OH which further transforms into H ,
3
2
presumably initiated indirectly by the reaction of OH (formed by
ꢃ
2ꢃ
3
HCO
3
, and CO
in alkaline suspension.
(
a)
(b)
(c)
O
HO
+
δ
C
CH
3
C
CH
3
⁻
-
δ
OH
O
OH
O
OH
OH
+
OH2+
+
2
+
O⁻
H O
H O
O⁻
2
2
pH < pKa
pK < pH < pH
pH > pHzpc
a
zpc
2
Scheme 1. Schematic illustration of the interaction between acetic acid/acetate and the TiO surface in different pH regimes.

1
34
S. Hamid et al. / Journal of Catalysis 349 (2017) 128–135
4
.2. Kinetics of the product formation
mation and organic pollutant degradation employing TiO
ried out at 2.4 ꢀ pH ꢀ 10, because pH values higher than 10 result
in the formation of large aggregates of TiO particles and affect the
2
are car-
The rates of the photocatalytic reactions of organic acids and
2
bases in the presence of semiconducting oxide have been reported
to be strongly dependent on the pH of the suspensions. Kormann
et al., who have used trichloroacetic acid and bis(2-chloroethyl)
amine as the probe molecules, explained this pH dependence of
the reaction rate with attractive and repulsive electrostatic interac-
performance of the photocatalyst [63]. The rather low H
2
formation
ꢃ1
rate of 5 mmol h obtained at pH 11 can therefore possibly also be
explained by a loss of photocatalytically active surface area due to
agglomeration of the photocatalyst particles.
2
tions between the ionic solutes and the charged TiO surface [57].
4
.3. Photoelectrochemical measurements
In the present work, a strong decrease in the formation rates of
methane and other hydrocarbons with increasing pH has been
observed. A similar pH dependent decrease in the methane forma-
tion rate has also been reported by Sakata et al. [21]. As discussed
in Section 4.1 all hydrocarbons formed during the photocatalytic
Fig. 3 shows that the Pt/TiO
tocurrent than the bare TiO electrode regardless of the presence
of acetic acid. This does not mean that Pt/TiO is less active for
acetate photooxidation, but can be explained by the fact that the
Pt loaded on TiO acts as electron trapping site and intercepts
the electrons during their transport from the TiO nanoparticles
to the FTO base electrode [64,65]. Although the apparent photocur-
rent generated on Pt/TiO was lower than that on bare TiO , the
steady-state photocurrent ratio (R = (Iph with acetic acid)/(Iph with-
out acetic acid)) is higher with Pt/TiO as compared to bare TiO
This result is consistent with the enhanced photocatalytic oxida-
tion of acetate on Pt/TiO
The effect of acetic acid on the photoinduced electron transfer
process occurring on TiO and Pt/TiO was also investigated by
2
electrode exhibits a lower pho-
2
2
2
transformation of acetic acid in the absence of O have the methyl
2
radical as the common reaction intermediate (Eqs. (10), (11), and
Å
2
(
13) in Table 2). A prerequisite for the significant CH
3
formation
(
Eqs. (3) and (7)) is the adsorption of acetic acid/acetate at the
surface through the carboxylate group thus enabling the
2
2
TiO
2
direct oxidation of the adsorbate by the photogenerated valance
band holes. This condition is, however, only provided in acidic sus-
pensions with pH < pHzpc. The importance of adsorption for the
efficient photocatalytic transformation of an organic solute is also
confirmed by the results of the photoelectrochemical measure-
ments (Section 3.2) with photocurrents for acetic acid oxidation
2
2
.
2
.
2
2
3+
2+
measuring the Fe /Fe redox-mediated photocurrent collected
by a Pt electrode immersed in the photocatalyst suspension [66].
The photocurrent generation was evidently enhanced with Pt/
2
on Pt/TiO electrodes being slightly higher at pH 2 than those at
pH 9.
The CH
than the H
decreased from 23 mmol h
4
formation rate was found to be significantly higher
formation rate at pH 2 and 5. The H formation rate
when the pH value
forma-
2 2
TiO in comparison with bare TiO thus indicating that the pres-
2
2
ence of Pt facilitates the interfacial electron transfer on the sus-
pended particles. The photocurrent was further enhanced in the
presence of acetic acid, which is consistent with the photocatalytic
ꢃ1
ꢃ1
to 18 mmol
was shifted from 2 to 5. In the pH regime 5 ꢀ pH ꢀ 9, the H
2
tion rates were found to be nearly unaffected by changes of the pH.
These results are in good agreement with the results published by
Zheng et al., who observed that an acidic suspension is more favor-
2 2
activities of TiO and Pt/TiO for acetic acid oxidation. This implies
that the electron separation is enhanced in the presence of Pt with
the simultaneous oxidation of acetic acid on the suspended cata-
lysts [66]. The photoelectrochemical measurements confirmed fur-
thermore that the adsorption of acetate molecules is stronger on
the TiO surface at acidic pH values as compared with the more
2
alkaline regions. Moreover, the use of a co-catalyst promotes the
interfacial charge transfer, hence resulting in a higher photocat-
able for photocatalytic H
when acetic acid/acetate is employed as the electron donor
29,30]. But these results are dissimilar from the results published
by Sakata et al., who observed significant increase in the photocat-
alytic H formation rate as the pH was shifted to the basic medium
21]. Apparently, that the reaction pathway presented here cannot
2
formation than an alkaline suspension
[
2
[
alytic activity for the simultaneous H
degradation.
2
production and the acetate
be generalized.
An alternative explanation for the observed decrease in the
reaction rates can be related to the energetic position of the
valance band and the conduction band of the semiconductor. We
5. Conclusions
2
have determined the flatband potential for TiO P 25 (which is
approximately the potential of the conduction band) to be Ufb = -
ꢃ0.58 V vs. NHE at pH 7. The flatband potential and, consequently,
the potentials of the valance band and the conduction band are
dependent on the pH of the suspension according to
During the photocatalytic transformation of acetic acid/acetate
in the presence of Pt-loaded TiO (Evonik Aeroxide TiO P25) car-
2
2
bon dioxide, molecular hydrogen, and methane were obtained as
the main reaction products in the gas phase exiting the photoreac-
tor, accompanied by other reaction products including other
hydrocarbons, alcohols, and aldehydes. It was shown that the pho-
tocatalytic acetic acid transformation proceeds via its light-
induced Kolbe decarboxylation resulting in the formation of the
Kolbe products, i.e., hydrocarbons, and the Hofer-Moest products,
i.e., alcohols.
fbðpHÞ ¼ UfbðpH0Þ ꢃ 59 ꢂ 10ꢃ3 ꢂ pH V
U
v
s: NHE
ð28Þ
Hence, the valance band of TiO
values upon an increase in pH thus lowering the oxidation power
of the holes. Apparently, the CH formation almost stops at pH val-
ues higher than 7. This can be explained by the fact that the ener-
getic position of the TiO
valance band (EVB = ECB – Ebg ꢄ Ufb – Ebg
with the band gap energy Ebg = 3.2 eV for anatase TiO ) at pH 7 is
2
shifts toward more negative
4
The rate of the photocatalytic formation of molecular hydrogen
2
from acetic acid and acetate in the presence of Pt-loaded TiO
found to be affected by the pH of the suspension with the highest
formation rate observed at pH 2. Increasing the pH resulted in a
small but significant decrease in the H formation rate. Hydrocar-
bons, C
2
was
2
approximately 2.6 V vs. NHE, which is close to the published value
for the critical anodic potential required for acetic acid decarboxy-
lation at a Pt electrode in aqueous solution (ꢄ+2.4 V vs. NHE)
H
2
2
x
H
2x+2 with x ꢀ 3, as well as alcohols were detected as other
[
21,37,61,62].
gaseous products of the photocatalytic transformation of CH
COOH/CH
3
-
ꢃ
Finally, it should be mentioned that the reaction rates might
3
COO . The formation rates of these products were found
also be affected by the suspension pH due to agglomeration of
the photocatalyst particles resulting in a decrease in the surface
area. Usually, photocatalytic experiments for simultaneous H for-
2
to be affected much more strongly by the pH of the suspension
than the H formation rate with significant amounts of these prod-
ucts being only evolved from acidic suspensions.
2

S. Hamid et al. / Journal of Catalysis 349 (2017) 128–135
[19] B. Kraeutler, A.J. Bard, J. Am. Chem. Soc. 100 (1978) 5985–5992.
135
These results are indicative for the appearance of two parallel
[
20] H. Yoneyama, Y. Takao, H. Tamura, A.J. Bard, J. Phys. Chem. 87 (1983) 1417–
reaction pathways. The direct pathway is initiated by hole oxida-
1
422.
ꢃ
tion of CH
3
COOH/CH
3
COO adsorbed on the TiO
2
surface resulting
[
21] T. Sakata, T. Kawai, K. Hashimoto, J. Phys. Chem. 88 (1984) 2344–2350.
in the intermediate formation of methyl radicals. Subsequent reac-
tions of these radicals yield mainly methane and ethane (Kolbe
[22] J. Araña, O.G. Díaz, Rodrıguez, J.M. Doña, J.A.H. Meliána, C. Garriga i Caboa, J.P.
´
Peña, M.C. Hidalgo, J.A. Navío-Santos, J. Moc. Catal. A: Chem. 197 (2003) 157–
1
71.
23] S. Mozia, Int. J. Photoenergy (2009), ID 469069.
[24] S. Mozia, A. Heciak, A.W. Morawski, J. Photochem. Photobiol. A: Chem. 216
2010) 275–282.
products). The indirect pathway is initiated by hole oxidation of
[
ꢃ
adsorbed H
2
O/ OH resulting in the formation of hydroxyl radicals.
(
These radicals subsequently react mainly to the methyl group of
the acetate molecule yielding hydroxylated products such as gly-
colic acid and alcohols (Hofer-Moest products). The observed pH
dependence of the formation rates of the gaseous products sug-
gests that the indirect reaction pathway becomes predominant in
alkaline suspensions.
[
[
25] S. Mozia, A. Heciak, A.W. Morawski, Catal. Today 161 (2011) 189–195.
26] S. Mozia, A. Heciak, A.W. Morawski, Appl. Catal. B: Environ. 104 (2011) 21–29.
[27] S. Mozia, A. Kulagowska, A.W. Morawski, Molecules 19 (2014) 19633–19647.
[
[
[
[
[
[
28] A. Heciak, A.W. Morawski, B. Grzmil, S. Mozia, Appl. Catal. B: Environ. 140–141
2013) 108–114.
29] X.-J. Zheng, L.-F. Wei, Z.-H. Zhang, Q.-J. Jiang, Y.-J. Wei, B. Xie, M.-B. Wei, Int. J.
Hydrogen Energy 34 (2009) 9033–9041.
(
30] X.-J. Zheng, Y.-J. Wei, L.-F. Wei, B. Xie, M.-B. Wei, Int. J. Hydrogen Energy 35
(
2010) 11709–11718.
Conflict of interest
31] S. Asal, M. Saif, H. Hafez, S. Mozia, A. Heciak, D. Moszy n´ ski, M. Abdel-Mottaleb,
Int. J. Hydrogen Energy 36 (2011) 6529–6537.
32] T. Puangpetch, S. Chavadej, T. Sreethawong, Energy Convers. Manage. 52
The authors confirm that there are no known conflict of inter-
ests associated with this publication.
(
2011) 2256–2261.
33] L.-F. Wei, X.-J. Zheng, Z.-H. Zhang, Y.-J. Wei, B. Xie, M.-B. Wei, X.-L. Sun, Int. J.
Energy Res. 36 (2012) 75–86.
[
[
[
34] L. Kos, I. Michalska, R. Zylla, J. Perkowski, Environ. Prot. Eng. 38 (2012) 29–39.
35] J.Y. Park, I.H. Lee, Korean J. Chem. Eng. 26 (2009) 387–391.
36] B. Kraeutler, A.J. Bard, J. Am. Chem. Soc. 99 (1977) 7729–7731.
Acknowledgements
Financial Support from the Bundesministerium für Bildung und
Forschung BMBF (Project ‘‘DuaSol” Nr. 03SF0482C) and Global
Research Laboratory program (2014K1A1A2041044), Korea gov-
ernment (MSIP) through NRF is gratefully acknowledged. We thank
H.C. Starck Company for cooperation and for providing materials
for the present study.
[37] B. Kraeutler, A.J. Bard, Nouv. J. Chim. 3 (1979) 31–38.
[
[
38] J. Zosel, W. Oelßner, M. Decker, G. Gerlach, U. Guth, Meas. Sci. Technol. 22
2011) 72001.
(
39] A.M. Toth, M.D. Liptak, D.L. Phillips, G.C. Shields, J. Chem. Phys. 114 (2001)
4595.
[40] H.P. Boehm, Discuss. Faraday Soc. 52 (1971) 264–275.
[41] S.J. Hug, B. Sulzberger, Langmuir 10 (1994) 3587–3597.
[42] Y. Nosaka, M. Kishimoto, J. Nishino, J. Phys. Chem. B 102 (1998) 10279–10283.
[43] B. Kraeutler, C.D. Jaeger, A.J. Bard, J. Am. Chem. Soc. 100 (1978) 4903–4905.
[44] M. Kaise, H. Kondoh, C. Nishihara, H. Nozoye, H. Shindo, S. Nimura, O. Kikuchi,
J. Chem. Soc., Chem. Commun. (1993) 395–396.
Appendix A. Supplementary data
[
45] M. Kaise, H. Nagai, K. Tokuhashi, S. Kondo, S. Nimura, O. Kikuchi, Langmuir 10
(1994) 1345–1347.
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.jcat.2017.02.033.
[
[
46] Y. Nosaka, K. Koenuma, K. Ushida, A. Kira, Langmuir 12 (1996) 736–738.
47] I.R. Macdonald, S. Rhydderch, E. Holt, N. Grant, J.M. Storey, R.F. Howe, Catal.
Today 182 (2012) 39–45.
References
[48] O. Golberg-Oster, R. Bar-Ziv, G. Yardeni, I. Zilbermann, D. Meyerstein, Chem.
Eur. J. 17 (2011) 9226–9231.
[
1] R.M. Navarro, M.C. Sánchez-Sánchez, M.C. Alvarez-Galvan, F.d. Valle, J.L.G.
Fierro, Energy Environ. Sci. 2 (2009) 35–54.
2] K. Shimura, H. Yoshida, Energy Environ. Sci. 4 (2011) 2467.
3] A.V. Puga, Coord. Chem. Rev. 315 (2016) 1–66.
4] J. Schneider, M. Matsuoka, M. Takeuchi, J. Zhang, Y. Horiuchi, M. Anpo, D.W.
Bahnemann, Chem. Rev. 114 (2014) 9919–9986.
5] G.R. Bamwenda, S. Tsubota, T. Nakamura, M. Haruta, J. Photochem. Photobiol.
A: Chem. 89 (1995) 177–189.
6] T.A. Kandiel, R. Dillert, D.W. Bahnemann, Photochem. Photobiol. Sci. 8 (2009)
[49] R. Bar-Ziv, I. Zilbermann, O. Oster-Golberg, T. Zidki, G. Yardeni, H. Cohen, D.
Meyerstein, Chem. Eur. J. 18 (2012) 4699–4705.
[50] R. Bar-Ziv, I. Zilbermann, T. Zidki, G. Yardeni, V. Shevchenko, D. Meyerstein,
Chem. Eur. J. 18 (2012) 6733–6736.
[51] R. Bar-Ziv, I. Zilbermann, M. Shandalov, V. Shevchenko, D. Meyerstein, Chem.
Eur. J. 21 (2015) 19000–19009.
[52] R. Bar-Ziv, T. Zidki, I. Zilbermann, G. Yardeni, D. Meyerstein, ChemCatChem 8
(2016) 2761–2764.
[53] P. Neta, M. Simic, E. Hayon, J. Phys. Chem. 73 (1969) 4207–4213.
[54] B.C. Gilbert, R.O.C. Norman, G. Placucci, R.C. Sealy, J. Chem. Soc., Perkin Trans. 2
(1975) 885–891.
[55] M.J. Davies, B.C. Gilbert, C.B. Thomas, J. Young, J. Chem. Soc., Perkin Trans. 2
(1985) 1199–1204.
[56] G.V. Buxton, C.L. Greenstock, W.P. Helman, A.B. Ross, J. Phys. Chem. Ref. Data
17 (1988) 513–886.
[57] C. Kormann, D.W. Bahnemann, M.R. Hoffmann, Environ. Sci. Technol. 25 (1991)
494–500.
[
[
[
[
[
[
[
[
6
83–690.
7] T.A. Kandiel, R. Dillert, L. Robben, D.W. Bahnemann, Catal. Today 161 (2011)
96–201.
8] A.Y. Ahmed, T.A. Kandiel, I. Ivanova, D. Bahnemann, Appl. Surf. Sci. 319 (2014)
4–49.
9] L.M. Ahmed, I. Ivanova, F.H. Hussein, D.W. Bahnemann, Int. J. Photoenergy
014 (2014) 1–9.
1
4
2
[
[
[
10] T.A. Kandiel, I. Ivanova, D.W. Bahnemann, Energy Environ. Sci. 7 (2014) 1420.
11] A. Patsoura, D.I. Kondarides, X.E. Verykios, Catal. Today 124 (2007) 94–102.
12] H. Bahruji, M. Bowker, P.R. Davies, F. Pedrono, Appl. Catal. B: Environ. 107
[58] G. Atherton, M. Fleischmann, F. Goodridge, Trans. Faraday Soc. 63 (1967)
1468–1477.
(
2011) 205–209.
13] M. Antoniadou, V. Vaiano, D. Sannino, P. Lianos, Chem. Eng. J. 224 (2013) 144–
48.
14] A.V. Puga, A. Forneli, H. García, A. Corma, Adv. Funct. Mater. 24 (2014) 241–
48.
[59] S. Glasstone, A. Hickling, Trans. Electrochem. Soc. 75 (1939) 333–352.
[60] A.J. Fry, Synthetic Organic Electrochemistry, second ed., Wiley, New York,
1989.
[61] S.N. Shukla, O.J. Walker, Trans. Faraday Soc. 27 (1931) 722–730.
[62] T. Dickinson, W.F.K. Wynne-Jones, Trans. Faraday Soc. 58 (1962) 382–387.
[63] S. Malato, P. Fernández-Ibáñez, M.I. Maldonado, J. Blanco, W. Gernjak, Catal.
Today 147 (2009) 1–59.
[64] N. Lakshminarasimhan, A.D. Bokare, W. Choi, J. Phys. Chem. C 116 (2012)
17531–17539.
[65] J. Lee, W. Choi, J. Phys. Chem. B 109 (2005) 7399–7406.
[66] H. Park, W. Choi, J. Phys. Chem. B 108 (2004) 4086–4093.
[
[
[
[
[
[
1
2
15] A. Beltram, I. Romero-Ocaña, J. Josè Delgado Jaen, T. Montini, P. Fornasiero,
Appl. Catal. A: Gen. 518 (2016) 167–175.
16] E.P. Melián, C.R. López, D.E. Santiago, R. Quesada-Cabrera, J.O. Méndez, J.D.
Rodríguez, O.G. Díaz, Appl. Catal. A: Gen. 518 (2016) 189–197.
17] A.F. Alkaim, T.A. Kandiel, R. Dillert, D.W. Bahnemann, Environ. Technol. 37
(
2016) 2687–2693.
18] B. Kraeutler, A.J. Bard, J. Am. Chem. Soc. 100 (1978) 2239–2240.