Photodegradation of Rhodamine B over Ag modified ferroelectric BaTiO3 under simulated solar light: Pathways and mechanism

Article abstract of DOI:10.1039/c5ra00798d

The use of semiconductors with a 'built in' bias has now become of interest for a growing number of photoactive applications. Using a combination of spectroscopic techniques, gas chromatography in association with mass spectroscopy and NMR, we show that a sample of ferroelectric BaTiO₃ decorated with nanostructured Ag denatures a standard dye molecule (Rhodamine B) via a photocatalytic oxidation mechanism. The photosensitized oxidation was inhibited due to band bending induced by ferroelectric polarisation. In the Ag-BaTiO₃ system we find a slight hypsochromic wavelength shift during the initial stages of degradation (only 3 nm before 80% degradation percentage) and associate this shift with the cleavage of the chromophore structure which pre-empted deethylation. This shift in maximum absorption of the dye molecule did not occur until the later stages of molecule fragmentation. Our major identifiable breakdown intermediate was benzoic acid. A lack of other identifiable fragments during the breakdown of the dye is associated with retention of these fragments on the catalyst as full mineralisation of the dye liberates CO₂. This journal is

Full text of DOI:10.1039/c5ra00798d

View Article Online
View Journal
RSC Advances
This article can be cited before page numbers have been issued, to do this please use: Y. CUI, S. Goldup
and S. Dunn, RSC Adv., 2015, DOI: 10.1039/C5RA00798D.
This is an Accepted Manuscript, which has been through the
Royal Society of Chemistry peer review process and has been
accepted for publication.
Accepted Manuscripts are published online shortly after
acceptance, before technical editing, formatting and proof reading.
Using this free service, authors can make their results available
to the community, in citable form, before we publish the edited
article. This Accepted Manuscript will be replaced by the edited,
formatted and paginated article as soon as this is available.
You can find more information about Accepted Manuscripts in the
Information for Authors.
Please note that technical editing may introduce minor changes
to the text and/or graphics, which may alter content. The journal’s
standard Terms & Conditions and the Ethical guidelines still
apply. In no event shall the Royal Society of Chemistry be held
responsible for any errors or omissions in this Accepted Manuscript
or any consequences arising from the use of any information it
contains.
www.rsc.org/advances

View Article Online
DOI: 10.1039/C5RA00798D
Page 1 of 9
RSC Advances
Dynamic Article Links ►
Journal Name
Cite this: DOI: 10.1039/c0xx00000x
www.rsc.org/xxxxxx
ARTICLE TYPE
Photodegradation of Rhodamine B over Ag modified ferroelectric
BaTiO under simulated solar light: pathways and mechanism
3
a
b
a
Yongfei Cui, Stephen M. Goldup and Steve Dunn*
Received (in XXX, XXX) Xth XXXXXXXXX 20XX, Accepted Xth XXXXXXXXX 20XX
5
DOI: 10.1039/b000000x
45
into the semiconductor. This occurs when the conduction band
potential of the semiconductor is below the LUMO of the dye
molecule, or chemisorbed molecules such as oxygen molecules
followed by the formation of oxidative radicals.
The use of semiconductors with a ‘built in’ bias have now
become of interest for a growing number of photoactive
applications. Using a combination of spectroscopic techniques,
gas chromatography in association with mass spectroscopy
and NMR we show that a sample of ferroelectric BaTiO3
decorated with nanostructured Ag denatures a standard dye
molecule (Rhodamine B) via a photocatalytic oxidation
mechanism. The photosensitized oxidation was inhibited due
to band bending induced by ferroelectric polarisation. In the
A large number of semiconductor photocatalytic systems have
been developed and there are is a growing body of understanding
pertaining to these systems . In addition to the traditional
semiconductor systems, a new and interesting group of materials
that are proving of increasing and significant interest are
ferroelectric materials. This class of material, capable of
1
1
2
2
0
5
0
5
5
5
6
0
5
0
7
–11
Ag-BaTiO system we find a slight hypsochromic wavelength
3
developing an internal electric field due to the asymmetry in the
shift during the initial stages of degradation (only 3 nm before
12–15
crystal lattice show unique properties in photochemistry
. It
8
0% degradation percentage) and associate this shift with the
has been demonstrated that the spontaneous polarisation in
cleavage of the chromophore structure which pre-empted
deethylation. This shift in maximum absorption of the dye
molecule did not occur until the later stages of molecule
16
ferroelectrics lead to charge carrier separation and that this is
manifested as spatially selective redox chemistry. In addition, the
spontaneous polarisation induced surface charge can be screened
by adsorbing charged molecules from atmosphere. This leads to
fragmentation.
Our
major
identifiable
breakdown
1
7
intermediate was benzoic acid. A lack of other identifiable
fragments during the breakdown of the dye is associated with
retention of these fragments on the catalyst as full
mineralisation of the dye liberates CO2.
the formation of a strong Stern layer . This Stern layer can
influence the bond structure of an adsorbed or chemisorbed
18,19
molecule which may lead to a change in bond strength
. These
65 unique properties of a ferroelectric can overcome the common
drawbacks in other catalytic systems, through inhibiting charge
2
0
Introduction
carrier recombination .
Semiconductor photocatalysis has received significant attention
21
In 2013 we demonstrated that the presence of Ag on the surface
1
–3
due potential applications in environmental remediation and as
a possible alternative to provide future energy security i.e. to
produce solar fuels. The hazardous organic compounds in
industrial effluent can be decomposed through a series of
chemical redox reactions driven by photoexcited charge carriers
in an excited semiconductor. A number of previous publications
have described the reaction pathways that have resulted in some
of a ferroelectric phase of BaTiO3 could significantly enhance
photocatalytic degradation of dye molecules. We fully
7
7
8
0
5
0
3
3
4
0
5
0
characterised the Ag-BaTiO system and readers are encouraged
3
to review our earlier work for the details of synthesis and analysis
of material presented in this publication. Additionally we
demonstrated that there is an influence of the ferroelectric nature
of BaTiO by comparing systems of different crystal structures.
3
4
–6
widely accepted reaction schemes . These reactions can be
broadly divided into two mechanisms depending on the process
of charge carrier generation and are termed photocatalytic
When the BaTiO catalyst had a high ferroelectric tetragonal
3
phase there was a three times increase in decolourisation rate
compared to when the non-ferroelectric cubic phase was
5
2
1
oxidation (PCO) and photoassisted oxidation (PAO) .
predominant . We stated that this enhancement was due to the
internal dipole of the ferroelectric assisting separation of excited
charge carriers and the tight Stern layer on the surface of
Photocatalytic oxidation can occur when the semiconductor is
exposed to super-band gap irradiation. Electrons and holes are
formed directly in the conduction and valence band on interaction
with incident photons. Photosensitized oxidation occurs when
sub-band gap light is used. In this instance the organic dye is
excited resulting in the injection of carriers (normally an electron)
21
BaTiO3 .These findings demonstrate a promising future for
ferroelectric materials as photocatalysts.
This journal is © The Royal Society of Chemistry [year]
[journal], [year], [vol], 00–00 | 1

View Article Online
DOI: 10.1039/C5RA00798D
RSC Advances
Page 2 of 9
The current state of the art expresses a variety of reaction
intermediates associated with different reaction pathways for the
chemical degradation of dye molecules in a photocatalytic
system. There has been a focus on determining the reation
When illuminating the sample with a spectrum other than AM
1.5G optical filters termed, UV block (UQG Optics LTD, UV
blocking, 50 mm*50 mm, 1.1 mm) and visible-block (UQG
Optics LTD, Schott UG1, 50 mm*50 mm, 2 mm) were placed
5
0
5
intermediates from Rhodamine B, a common target dye, with the 60 between the irradiation source and dye catalyst suspension. The
2
2–31
results being widely reported
. Among these reports, a range
flux of incident light was maintained when filters were used by
adjusting the solar simulator output. The zeta potential of
photocatalysts was determined with Zetasizer Nano(Malvern
co.uk) by dispersing amount of catalyst powder in aqueous
2
2
23,26,31
of catalysts such as NaBiO , Bi WO
, in addition to TiO2
, have been investigated and show different
degradation pathways and chemical intermediates. In addition to
3
2
6
2
5,28–30
based catalysts
1
the range of catalysts generating different reaction intermediates, 65 solutions with different pH ranging from acidic to alkalinic.
changes in the surface state, crystalline structure and
3
2
Proton NMR spectrums of the dye solution were obtained using a
Bruker NMR spectrometer operating at 600 MHz at 300 K. The
samples for NMR analysis were prepared using the following
methodology: 200 ml of 10 ppm RhB solution was mixed with
0.6 g catalysts, followed by dark-equilibrium for 1 hour,
irradiation under solar light for the appropriate time, sampling,
and centrifugation (as detailed earlier). The remaining solvent
from the obtained solution (30 ml) was evacuated under reduced
pressure at 45°C using a rotary evaporator. The remaining residue
morphologies of the photocatalysts
have also been
3
0,33
demonstrated to influence the degradation process
. However,
to date there has been no contribution regarding the degradation
pathway for a target dye molecule when a ferroelectric catalyst is
used.
1
70
Here we investigate the degradation products formed during the
photochemical reaction with a modified ferroelectric catalyst
Ag:BaTiO ) addressing the interesting question – does the
(
3
7
5
was dissolved in 1ml D O ready for analysis.
2
2
3
3
0
5
0
5
spatial separation of photoinduced carriers influence the reaction
pathway? It should also be noted that the presence of
nanostructured silver on the surface of the BaTiO3 may also
influence the reaction pathway through the interaction with a
surface plasmon or photo assisted oxidation into the semi-
conductor. A number of studies on traditional semi-conductor
2
Intermediates formed during degradation were identified by GC-
MS (Agilent GC-MS, a 6890 gas chromatography interfaced with
5
973 Mass Selective Detector). To improve the signal to noise
values from the GC-MS, the initial concentration of RhB was
systems provide an insight into the reaction pathway when 80 increased from 10 ppm to 50 ppm, all other experimental
oxidation and reduction are not separated at the interface. Our
focus is on the reaction with RhB under simulated solar light over
our ferroelectric catalyst. The products of the reaction were
parameters remained the same. The pH of the solution was then
adjusted by 1M HCl to 3.0 and the aqueous solutions were
extracted bydichlromethane (DCM, Reagent Degrade) three
times. This solution was then dehydrated by anhydrous sodium
1
evaluated using UV-vis spectroscopy, H nuclear magnetic
resonance (1H NMR) and gas chromatography mass 85 sulphate and solvent removed by rotary evaporation. After the
spectrometry (GC-MS).To verify our methodology, we employ
residue was dissolved in 1ml DCM, 0.5 ml N,O-
Bis(trimethylsilyl)trifluoroacetamide (BSTFA, Sigma, ＞99 %)
was added and the resulting solution was kept at 50 °C for 30
mins to drive trimethylsilylation.
the widely used photocatalyst TiO (P25) as a reference following
2
the same experimental methods we have used or BaTiO to verify
3
our procedures and demonstrate context to the existing literature.
Experimental Methods
90 Results and Discussion
Ag modified BaTiO3 (termed Ag-BTO) was prepared and
Adsorption of RhB molecules onto the surface of catalysts
2
1
characterised as described in our previous publication . In
summary we used BaTiO3 powder supplied by Sigma (99.9%
trace metal basis <2 μm), followed by heat treatment at 1200°C
for 10hours and photochemical silver deposition (UV irradiation
for 30 seconds and Ag nanoparticles around 5-10nm formed).
It is understood that the heterogeneous photodegradation process
4
,34
4
0
can be divided into five steps
:
(1) Transfer the organic reactants from the bulk liquid to the
95 surface region of catalysts;
(2) Adsorption of the reactants onto the surface of photo-excited
0
.15g of Ag-BTO was loaded into 50ml of 10ppm Rhodamine B
catalysts;
dye solution (RhB, Sigma, 99.9%). The dye solution and catalyst
suspension was kept in the dark with gentle stirring for 30 mins to
obtain adsorption-desorption equilibrium. Photochemical
degradation was performed under a solar simulator (Newport,
class ABB) fitted with an AM 1.5G filter at a distance of 17 cm
from the irradiation source giving a solar flux of 1sun (100 mW
cm ). During the degradation process, 2 ml aliquots of solution
were taken for sampling at fixed intervals during the degradation
process. The aliquot of solution and catalyst powder was then
separated by centrifugation at 4000 rpm for 30 minutes. The
optical absorption of the obtained dye solution was measured
using a PerkinElmer Lambda 950 UV−Vis spectrophotometer.
(3) Photocatalytic reactions of the adsorbed phases;
4
5
5
5
0
5
(4) Desorption of reaction products/intermediates from the
1
00 surface;
(5) Transfer of the products/intermediates to the bulk liquid.
The adsorption of a dye molecule onto the catalysts surface is an
important step that has implications for the rate of molecule
breakdown and, therefore, the overall reaction rate . During the
05 dark equilibrium phase dye molecules adsorb onto the surface of
the catalyst and develop equilibrium with dye molecules in
solution. The amount of dye adsorption on our catalyst system
was determined by comparing the absorption intensity of dye
-
2
35
1
2
| Journal Name, [year], [vol], 00–00
This journal is © The Royal Society of Chemistry [year]

View Article Online
DOI: 10.1039/C5RA00798D
Page 3 of 9
RSC Advances
solution before and after the equilibrium phase. Table 1 shows
direct photocatalysed. In comparison, for a standard TiO system,
2
the BET surface area and dye adsorption for Ag-BTO.
both visible light driven photosensitized oxidation and UV driven
photocatalytic oxidation have been shown to coexist during the
degradation process, shown in Figure S2, which is consistent with
45
Table 1 Dye adsorption on Ag-BTO sample
4
1
previously reported results .
BET surface area
Dye adsorption scaled
Catalysts
Ag-BTO
2
(
m /g)
0.61
by surface area (%)
7.48±0.19
It can be seen that the Ag-BTO shows a significant enhancement
in adsorption of RhB per unit surface area during the same
5
0
5
equilibrium period when compared to TiO (0.25±0.02 % of dye
2
per surface) absorbed (Table S1). This difference is due to the
polar nature of the ferroelectric catalyst which differentiates the
ferroelectric and non-ferroelectric material. The spontaneous
polarisation in a ferroelectric can induce macroscopic surface
charges, which are then screened by adsorbing the external
1
3
6–38
molecules and/ or internal charge carriers
. This results in a
higher loading of dye molecules at the interface with a tightly
3
9
bound Stern layer for a ferroelectric material. It has been
shown, for example, that BaTiO can adsorb more ethanol on 50 simulated solar light, visible light and UV light. After UV light
Figure 1 Photodegradation of RhB using Ag-BTO under
1
3
4
0
polar surfaces than on an unpolarized surface . This type of dye
adsorption enhancement has also been reported in the other
was blocked, the catalyst hardly showed any photocatalytic
activity.
2
0
ferroelectric LiNbO system .
3
It is known that the conduction band gap position of the
semiconductor is required to be lower than the lowest unoccupied
molecular orbital (LUMO) of the dye to make electrons transfer
successful for photoassisted oxidation. The structure of RhB is
accepted to give the LUMO and HOMO positions at1.1eV and -1
Photodegradation of RhB under different irradiation
conditions
5
5
2
0
While we have previously investigated the impact of ferroelectric
4
2
2
1
eV versus NHE . In terms of the band structures of a
semiconductor the specific band position of the semiconductor
crystal structures on Ag-BTO , in order to understand the impact
of ferroelectric dipoles on reaction pathways the interaction
between PCO and PAO must be determined. In order to achieve,
this photocatalytic properties of Ag-BTO were assessed through
degradation of RhB under full spectrum solar light, visible light
and UV. The latter two irradiation conditions were obtained by
passing the solar light through optical filters. The irradiation
spectra of the different test conditions are shown in Figure S1.
4
3,44
60
can be calculated according to the empirical equation
:
2
5
e
ECB  X  E  0.5E
g
(1)
Where ECB is the conduction band edge,
X
is the
e
electronegativity of the semiconductor, E is the energy of free
electrons on the hydrogen scale (about 4.5eV), and Eg is the band
3
0
The degradation profile for Ag-BTO under the different 65 gap energy of the semiconductor. The electronegativity of the
illumination sources is shown in Figure 1.
semiconductor can be calculated from the geometric mean of the
electronegativity of the constitute atoms. We have measured the
direct band gap of Ag-BTO to be around 3.14eV by extrapolating
the linear portion of (h) vs h using a Tauc plot that is
shown in Figure 2. Our measured value for the system in use here
In the case of our Ag-BTO system the results do not support the
hypothesis that both sub and super band gap illumination are
driving the reaction. As shown in Figure 1, after the UV
component of the full solar spectrum was blocked, there was
negligible degradation of the dye throughout the course of the
experiment period while the photodegradation profiles of full
spectrum solar illumination and visible-light filter block almost
overlap. This illustrates that in the Ag-BTO system, visible light
hardly contributes to the degradation and only UV excited charge
carriers are participating in the degradation reaction. The
photoassisted oxidation process is hindered with respect to the
2
45
7
0
3
5
is consistent with other publications, where the band gap of
46
BaTiO powder was reported to be direct at 3.18 eV
.
3
Table 2 Absolute electronegativity, band gap, conduction and
valence band edge of Ag-BTO
4
0
75
8
0
This journal is © The Royal Society of Chemistry [year]
Journal Name, [year], [vol], 00–00 | 3

View Article Online
DOI: 10.1039/C5RA00798D
RSC Advances
Page 4 of 9
0
.97 eV) and LUMO of RhB (-1 eV) and upward band bending of
Absolute
Electro -
the ferroelectric c- domain. Our results support the hypothesis
that this is enough of a barrier to prevent electron injection from
the excited dye to the conduction band of BTO. This reduced
injection of electrons will be manifested by a reduction in the rate
of dye degradation. Once the electron transport cannot overcome
the barrier formed at the catalyst surface and solution the
photoassisted oxidation process will be suppressed. Accordingly
visible light cannot contribute to the degradation process, which
is consistent with the phenomenon observed in Fig.1. This is
Eg
Ec Edge
eV)
Ev Edge
(eV)
2
5
Material
Ag-BTO
(
(
eV)
negativity﹡
5.13
3.14
-0.94
2.2
30
﹡
The absolute electronegativity of constitute atoms are adapted
different from a TiO system, where the photocatalytic oxidation
and photoassisted oxidation reactions can coexist (see Fig.S2).
4
7
2
from Ralph’s work .
35
5
Figure
3 Calculated energy level diagram and proposed
degradation reactions for the Ag-BTO system under simulated
solar light. Photoassisted oxidation was hindered due to the
barriers arising from band bending.
Figure 2 Tauc plot for Ag-BTO using the relationship between
the square of αhμ versus photon energy for
a direct
4
4
5
0
5
0
semiconductor system. By extrapolating of the linear portion we
determine the band gap to be 3.14 eV.
Figure 4(a) shows the UV-vis absorption spectra of RhB dye
solution at different stages of decolourisation represented by the
percentage of absorption at a fixed wavelength. The spectra
1
0
Table 2 shows the adapted and calculated values for the band
edge positions using Equation 1. The energy level diagrams of
2
2,26,29,30
shows a previously observed
shift of wavelength during
the Ag-BTO catalytic systems from Table
2 are shown
degradation of RhB. The change of maximum absorption
wavelength as a function of decolourisation for the dye is shown
in Figure 4(b). There were two clear stages to the trend for the
shift of wavelength. Up to 80% decolourisation there is only a
slight shift in the wavelength of absorption (to a maximum of
schematically in Figure 3. In Figure 3 we demonstrate the
proposed reaction scheme. When Ag-BTO was dispersed in the
cationic dye RhB solution, the dye molecules will be attracted to
the negative domains (c-) on the surface of the ferroelectric
powder. The c+ surfaces will hold a net positive charge and so
will not attract and absorb the cationic dye. As Ag is a conductor
it will screen the charge in the ferroelectric but the whole c+
surface will have a net positive charge. The c- face is where holes
accumulate leading to upward band bending, as shown in
Figure.3. Due to the proximity of the conduction band edge (-
1
5
2
nm) while in the second stage of reaction the maximum
wavelength of absorption exhibited by the dye shifts significantly
by 55nm.
2
0
60
5
5
4
| Journal Name, [year], [vol], 00–00
This journal is © The Royal Society of Chemistry [year]

View Article Online
DOI: 10.1039/C5RA00798D
Page 5 of 9
RSC Advances
Dynamic Article Links ►
Journal Name
Cite this: DOI: 10.1039/c0xx00000x
www.rsc.org/xxxxxx
ARTICLE TYPE
-
3
4
4
5
5
0
5
0
charge carries from the catalysts and other species (O , OH , etc.).
2
This is typical for systems undergoing such a photocatalysed
reaction where the production of OH and other active species is
-
2
2
the initial step in photochemical reaction . In a photoassisted
oxidation, the reactive species are produced near the adsorption
site of the RhB molecules, where the excited dye molecules inject
electrons into the semiconductor. Therefore, it is be easier for the
reactive species to selectively attack the adsorbed group
selectively. In Ag-BTO system, where only UV-induced
photocatalytic oxidation exists, it is less likely that reactive
radicals will selectively attack of adsorbed group (alkylamine
5
group(-NEt ) in Ag-BTO, see Figure. S3) which will lead to non-
2
obvious deethlylation and a slight wavelength shift. This
argument is consistent with the observations from previous work,
where under UV irradiation the deethylation process was not so
2
1
obvious as when under visible irradiation . The deethylation of
4
4
the dye was regarded as initiated by visible light .
1
0
To provide a reference and demonstrate the robustness of our test
regimes, TiO was used as a model photocatalyst and used to
2
compare products of the dye at the same decolourisation stage,
55
shown in Figure S4. Overall we find that TiO system presents a
2
similar degradation process, that is the cleavage of chromophore
structure predominated over deethlylation. This observation is in
2
5,30,49
accordance with the other reported results
, where
degradation of RhB with TiO under visible light was regarded to
2
60
be mainly destruction of chromophore structure. Our findings
indicate that the presence of the metal nanostructures and the
ferroelectric nature of the catalyst have not significantly
influenced the process of cleavage of the chromophore and
indicates that our experimental procedure is robust.
1
5
1
65
H NMR spectroscopy
Figure 4 (a) UV-Vis absorption spectra of RhB dye solutions with
different degradation percentage in Ag-BTO system and (b) The
maximum absorption wavelength shift of RhB as a function of
degradation percentage in Ag-BTO system
To observe the structural changes of RhB during the degradation
2
0
processes proton NMR analysis was performed. Six samples at
1
the same percentage decolourisation, were evaluated using
H
NMR. Figure.5 shows the NMR spectra for the degradation of
RhB using Ag-BTO catalyst. The NMR signals from different
protons in the RhB structure have been assigned peaks d, e, f, g,
h, i and j represent aromatic hydrogen atoms Hd, He, Hf, Hg, Hh,
Hi and Hj (Fig. S5). The hydrogen atoms in the ethyl group of
RhB structure contributed to the NMR peaks at δ 1.15-1.25 ppm
70
The shift in absorption maximum wavelength has been associated
with the deethylation process. The four ethyl group in RhB are
removed one by one and a series of N-de-ethylated intermediates
2
5
2
9
are formed . This deethylation process accompanies and
competes with cleavage of the conjugated chromophore structure 75 and δ 3.6 - 3.6 ppm. In addition, there are a series of small peaks
during the overall degradation process. Our Ag-BTO system
shows a slight hypsochromic wavelength shift during the initial
stage of degradation, indicating that cleavage of chromophore
structure predominated.
between peak a and peak b in the initial RhB sample. After
careful examination of the NMR spectra of pure RhB and after
catalyst loading, these peaks have been assigned to an impurity
from of RhB and/or the NMR sample preparation process.
3
0
In Ag-BTO system, as described above, only UV induced 80 As the photodegradation progresses the intensity of the
photocatalytic oxidation contributes to the whole process, and the
characteristic peaks associated with aromatic hydrogen atoms
oxidative radicals are generated from the reactions between
(peaks d-j) and ethyl group(peaks a and b) decrease while new
This journal is © The Royal Society of Chemistry [year]
[journal], [year], [vol], 00–00 | 5

View Article Online
DOI: 10.1039/C5RA00798D
RSC Advances
Page 6 of 9
peaks appear at δ 3.0-3.1ppm and around peak a (δ1.15-1.25ppm)
as highlighted in Figure 5.The disappearance of characteristic
peaks indicates the breakdown of chromophore structures and
simultaneous removal of ethyl groups during the degradation
process. The new peaks in the spectrum are associated with the
intermediate products formed during degradation. In addition to
the intensity decreasing for the characteristic peaks, a peak shift
can be observed. As highlighted in Fig.5, peaks d, e and f moved
to a higher chemical shift while peaks g, h, I and j remain in
place. If we consider the hydrogen atoms which these peaks
represent in the RhB structure, this chemical shift could be
associated with either the deethylation process or break down of
the xanthene ring.
Figure S6 shows the NMR spectra for the degradation of RhB for
the TiO2 system. Being the same as Ag-BTO system, the
characteristic peaks for aromatic hydrogen atoms and ethyl group
disappeared. Meanwhile, new peaks appeared at δ 3.0-3.1 ppm
and at δ 1.15-1.25 ppm were observed. The latter peaks were
1
5
5
considered to be signals from –CH and analogous group. These
3
20
features were consistent with the other RhB degradation works
2
9,32
1
0
using TiO as catalysts
.
2
2
5
3
0
1
3
5
Figure 5 H NMR spectra of samples with different degradation percentage with Ag-BTO
reasonable to assume that they are intermediate products in the
Identification of intermediates by GC-MS
degradation.
4
4
5
0
5
0
Figure 6 shows the gas chromatography of samples with
Comparing between Figure 6(a) and Figure 6(b), with the
irradiation for 10 and 90 minutes in Ag-BTO system. After 60 irradiation time increasing to 90 minutes, some new peaks
discounting peaks associated with the impurities in the system
see Fig.S7), new peaks (marked by ) were observed for the
observed in Figure 6(a) have diminished while the intensity of
benzoic acid peak dropped. The change in the pattern of the GC-
MS relates the species that are in solution during the degradation
of the RhB. The results shown here indicate that as the
(
Ag-BTO system. These have been associated with intermediate
products produced during the degradation process. The peaks at
the retention time 5.82 minutes (Fig. 6(a)) and the peak with a 65 photochemical process proceeds the reaction intermediates are
retention time of 5.928 minutes (Fig. 6(b)) have been assigned to
benzoic acid according to the mass fragmentation pattern. The
small differences in retention time arise from individually sample
injections.
reacting to form a new mixture during degradation of the RhB. A
number of publications have investigated the distribution of
2
2–31
intermediates and degradation pathways for Rhodamine B
using different catalysts. These previous works describe a number
of discrepancies in the intermediates found during the
degradation of RhB. However, phthalic acid and benzoic acid
have been widely observed and have become the accepted, or the
most probable, intermediates during the degradation of RhB.
70
In addition to the benzoic acid peak, there are a new peaks at
4
.172 minutes, 5.044 minutes, 6.068 minutes, 7.862 minutes in
Figure 6(a) and 4.179 minute, 6.087 minute in Fig. 6(b). However,
we have not been able to assign a chemical structure to these
peaks according to the mass fragmentation pattern. As they do
not exist in the blank experiments and are not impurities so it is
5
5
75
6
| Journal Name, [year], [vol], 00–00
This journal is © The Royal Society of Chemistry [year]

View Article Online
DOI: 10.1039/C5RA00798D
Page 7 of 9
RSC Advances
Table 3 Gas chromatography Peaks detected in Ag-BTO system
Catalytic system
Retention time of Peaks obtained
Ag-BTO
5
4
.172 minutes, 5.044 minutes, 5.82
T=10 min
T=90 min
minute s (benzoic acid), 6.068 minutes,
7
4
.862 minutes
.179 minutes, 5.928 minutes (benzoic
1
0
acid),
.087 minutes
6
45
The intermediates in TiO system were also analysed (see Figure
2
S9-10) following the same methodolog. Ethylene glycol (at
retention time 3.771minutes) and benzoic acid (at retention time
5
.8 minutes) have been identified as part of the intermediates in
1
5
50
addition to a few other peaks unassigned in the sample at
irradiation time 10 minutes. The confirmed species have been
observed and reported as intermediates during degradation of
3
2,50
22,26
RhB in TiO2
and other catalytic systems
, supporting the
feasibility of the experimental process.
5
5
The impact of ferroelectric polarisation on surface chemistry has
been extensively investigated, and it ranges from the varied
5
1,52
desorption activation energy
coefficient . Furthermore, the dipoles of ferroelectrics were
demonstrated to bend the carbon dioxide molecule , and affect
to the reactive sticking
4
0,53
2
0
Figure 6 Gas chromatography of RhB with Ag-BTO-anneal after
irradiation for (a) 10 minutes and (b) 90 minutes under solar
simulator, with mass spectrum of the peak at 5.82 minutes (a),
indicating the presence of benzoic acid. The peak at 5.928
minutes in (b) presents the same mass spectrum and can be
assigned to benzoic acid as well.
1
9
.
54
60
the adsorption configuration of OH radicals . The polarisation-
dependent adsorption has been observed in different species, e.g.
5
3,55,56
40,51,53
52
water
, alcohols
, and acetic acid . In addition to the
influence on adsorption, the interaction between the dipoles of
adsorbed molecules and surface dipole in ferroelectrics can also
2
5
5
7
In order to test this and support our observation, we performed 65 affect the activation energy of the surface catalysed reactions .
experiments where we ran phthalic acid and benzoic acid as the
target molecule through the GC-MS. From this we were able to
determine the expected output for the two molecules using our
GC-MS system. The retention time of phthalic acid and benzoic
Based on previous work, it was expect that there will be some
intriguing phenomenon on the ferroelectric surface when
interacting with dye molecules, and that would lead to a varied
degradation pathway. However, according to our results, it seems
3
3
4
0
5
0
acid were found to be 8.644 minutes and 5.864 minutes (Figure 70 there is no significant change in the intermediates distribution and
S8). After examining the GC-MS results for the degradation of
RhB over Ag-BTO, no peaks could be found that matched the
peak position and mass fragmentation pattern of phthalic acid.
This excluded the possibility of phthalic acid being produced as
an intermediate when RhB is degraded over Ag-BTO. The peaks
which have been assigned to benzoic acid in Figure 6 are
consistent with standard benzoic acid in terms of retention time
and mass spectra. This further demonstrates the existence of
benzoic acid as intermediate. Overall, the intermediates observed
in Ag-BTO system are summarized in Table 3.
degradation pathway with respect to the nonpolar TiO system is
2
seen for the ferroelectric system. There could be a number of
reasons to account for this.
First, our ferroelectric system is ferroelectric powder, of which
the polarisation orientation is not systematically controlled. Thus
the potential effect of polarisation will not be evident as poled
thin film or single crystal which is widely used in the
75
5
8
investigation of polarisation effect . In addition, it was proposed
that the geometry of transition metals on ferroelectric substrate
This journal is © The Royal Society of Chemistry [year]
Journal Name, [year], [vol], 00–00 | 7

View Article Online
DOI: 10.1039/C5RA00798D
RSC Advances
Page 8 of 9
55
(4)
Chong, M. N.; Jin, B.; Chow, C. W. K.; Saint, C. Water Res.
2010, 44, 2997–3027.
plays an important role in enlarging or inhibiting the polarisation
5
9
effect . When the metal clusters are too thick for their surface to
be affect by the underlying polarisation, it will behave similarly
to the other nonpolar systems. But in our system, the size of Ag
nanoparticles is only 5-10nm, which is not likely to block the
effect from polarisation, although the exact critical thickness is
not known. Thus the first reason is more likely. Nevertheless, our
initial attempt to probe into the degradation mechanism,
intermediates and pathways in degradation of RhB dye molecules
using ferroelectric BaTiO3 provides a deeper understanding in
ferroelectric catalytic system and add directions to the
development of a new efficient photocatalyst.
(
(
(
5)
6)
7)
Konstantinou, I. K.; Albanis, T. A. Appl. Catal. B Environ.
2
004, 49, 1–14.
5
Akpan, U. G.; Hameed, B. H. J. Hazard. Mater. 2009, 170,
520–529.
60
Moniz, S. J. A.; Shevlin, S. A.; Martin, D. J.; Guo, Z.-X.; Tang,
J. Energy Environ. Sci. 2015, -.
1
0
(8)
Tong, H.; Ouyang, S.; Bi, Y.; Umezawa, N.; Oshikiri, M.; Ye, J.
Adv. Mater. 2012, 24, 229–251.
Conclusion
65
(9)
Maeda, K. J. Photochem. Photobiol. C Photochem. Rev. 2011,
12, 237–268.
We have used UV-Vis spectroscopy, NMR and GC-MS to
investigate the degradation process of Rhodamine B under
1
2
2
3
5
0
5
0
simulated solar light with silver modified BaTiO3 as
a
(10)
Xu, H.; Ouyang, S.; Liu, L.; Reunchan, P.; Umezawa, N.; Ye, J.
J. Mater. Chem. A 2014, 2, 12642–12661.
photocatalyst. Our results indicate that a tight Stern layer is
formed consisting of the dye cation and counter anions around the
ferroelectric catalyst. The polarisation of the dipole associated
with the ferroelectric forms band bending and this hinders the
photosensitized oxidation process between adsorbed dye and the
catalyst, such that we see no evidence of direct electron injection
from the dye into the Ag-BTO. The hypsochromic wavelength
shift of the maximum absorption peak shown by UV-vis
demonstrates that the cleavage of the chromophore structure was
the primary initial process of molecule breakdown. The use of
NMR analysis further supported the hypothesis that during
degradation there was cleavage of the chromophore structure
followed by deethylation of the dye molecule. Analysis of the
breakdown products shows that we have been able to identify
benzoic acid as the main intermediate based on GC-MS analysis.
(
11)
12)
Di Paola, A.; García-López, E.; Marcì, G.; Palmisano, L. J.
Hazard. Mater. 2012, 211–212, 3–29.
7
0
(
Giocondi, J. L.; Rohrer, G. S. J. Phys. Chem. B 2001, 105,
8
275–8277.
(
13)
14)
Giocondi, J. L.; Rohrer, G. S. Chem. Mater. 2001, 13, 241–242.
Tiwari, D.; Dunn, S. J. Mater. Sci. 2009, 44, 5063–5079.
Li, L.; Salvador, P. A.; Rohrer, G. S. Nanoscale 2013.
(
75
(15)
(
16)
17)
Roy, S. S.; Gleeson, H.; Shaw, C. P.; Whatmore, R. W.; Huang,
Z.; Zhang, Q.; Dunn, S. Integr. Ferroelectr. 2000, 29, 189–213.
Notes and references
(
Tiwari, D.; Dunn, S.; Zhang, Q. Mater. Res. Bull. 2009, 44,
1219–1224.
a
Materials Research Institute, School of Engineering and Materials Science, Queen
Mary University of London, E1 4NS; E-mail: s.c.dunn@qmul.ac.uk
b
3
4
4
5
0
5
Department of Chemistry, University of Southampton, Southampton, SO17 1BJ
80 (18)
Cabrera, A. L.; Vargas, F.; Zarate, R. A. J. Phys. Chem. Solids
1
994, 55, 1303–1307.
†
Electronic Supplementary Information (ESI) available: [Dye adsoprtion
on referenced sample P25, irradiation spectra under different wavelength
conditions, zeta potential of Ag-BTO, photodegradation profiles, reaction
mechanisms, uv-vis spectra and H NMR spectra in P25 system, GC-MS
spectra and identified intermediates in P25 system and the comparsion
with standard compounds ]. See DOI: 10.1039/b000000x/
(
19)
20)
Cabrera, A. L.; Vargas, F.; Albers, J. J. Surf. Sci. 1995, 336,
80–286.
2
1
(
Stock, M.; Dunn, S. J. Phys. Chem. C 2012, 116, 20854–20859.
Acknowledgement
85
(21)
Cui, Y.; Briscoe, J.; Dunn, S. Chem. Mater. 2013, 25, 4215–
4
223.
The author, Yongfei Cui, would like to acknowledge Chinese Scholarship
Council for supporting this work. Dr. Joes Briscoe is acknowledged for
helpful discussion in this work.
(22)
(23)
(24)
(25)
Yu, K.; Yang, S.; He, H.; Sun, C.; Gu, C.; Ju, Y. J. Phys. Chem.
A 2009, 113, 10024–10032.
Fu, H.; Pan, C.; Yao, W.; Zhu, Y. J. Phys. Chem. B 2005, 109,
22432–22439.
5
0
(1)
Hoffmann, M. R.; Martin, S. T.; Choi, W.; Bahnemann, D. W.
Chem. Rev. 1995, 95, 69–96.
90
HE, Z.; YANG, S.; JU, Y.; SUN, C. J. Environ. Sci. 2009, 21,
(
2)
3)
Robertson, P. K. J. J. Clean. Prod. 1996, 4, 203–212.
2
68–272.
(
Bhatkhande, D. S.; Pangarkar, V. G.; Beenackers, A. A. C. M. J.
Chem. Technol. Biotechnol. 2002, 77, 102–116.
Chen, F.; Zhao, J.; Hidaka, H. Int. J. Photoenergy 2003, 05,
09–217.
2
8
| Journal Name, [year], [vol], 00–00
This journal is © The Royal Society of Chemistry [year]

View Article Online
DOI: 10.1039/C5RA00798D
Page 9 of 9
RSC Advances
(
26)
He, Z.; Sun, C.; Yang, S.; Ding, Y.; He, H.; Wang, Z. J. Hazard.
Mater. 2009, 162, 1477–1486.
(48)
(49)
Ma, Y.; Yao, J. J. Photochem. Photobiol. A Chem. 1998, 116,
167–170.
(27)
Lei, P.; Chen, C.; Yang, J.; Ma, W.; Zhao, J.; Zang, L. Environ.
Sci. Technol. 2005, 39, 8466–8474.
Fan, Y.; Chen, G.; Li, D.; Luo, Y.; Lock, N.; Jensen, A. P.;
Mamakhel, A.; Mi, J.; Iversen, S. B.; Meng, Q.; Iversen, B. B.
Int. J. Photoenergy 2012, 2012, 1–7.
45
5
(28)
LI, J.; MA, W.; LEI, P.; ZHAO, J. J. Environ. Sci. 2007, 19,
8
92–896.
(50)
(51)
Natarajan, T.; Natarajan, K.; Bajaj, H.; Tayade, R. J.
Nanoparticle Res. 2013, 15, 1–18.
(29)
(30)
(31)
(32)
Wu, T.; Liu, G.; Zhao, J.; Hidaka, H.; Serpone, N. J. Phys.
Chem. B 1998, 102, 5845–5851.
Yun, Y.; Kampschulte, L.; Li, M.; Liao, D.; Altman, E. I. J.
Phys. Chem. C 2007, 111, 13951–13956.
Wang, Q.; Chen, C.; Zhao, D.; Ma, W.; Zhao, J. Langmuir 2008,
24, 7338–7345.
1
0
50 (52)
(53)
Yun, Y.; Altman, E. I. J. Am. Chem. Soc. 2007, 129, 15684–
1
5689.
Fu, H.; Zhang, S.; Xu, T.; Zhu, Y.; Chen, J. Environ. Sci.
Technol. 2008, 42, 2085–2091.
Li, D.; Zhao, M. H.; Garra, J.; Kolpak, A. M.; Rappe, A. M.;
Bonnell, D. A.; Vohs, J. M. Nat. Mater. 2008, 7, 473–477.
Li, J.; Ma, W.; Chen, C.; Zhao, J.; Zhu, H.; Gao, X. J. Mol.
Catal. A Chem. 2007, 261, 131–138.
(54)
Hölscher, R.; Sanna, S.; Schmidt, W. G. Phys. status solidi
2012, 9, 1361–1365.
55
1
5
(33)
Liu, G.; Li, X.; Zhao, J.; Hidaka, H.; Serpone, N. Environ. Sci.
Technol. 2000, 34, 3982–3990.
(55)
(56)
Geneste, G.; Dkhil, B. Phys. Rev. B 2009, 79, 235420.
(
34)
35)
Herrmann, J.-M. Catal. Today 1999, 53, 115–129.
Sanna, S.; Hölscher, R.; Schmidt, W. G. Phys. Rev. B 2012, 86,
2
05407.
(
Vinodgopal, K.; Kamat, P. V. J. Phys. Chem. 1992, 96, 5053–
5
059.
(57)
Zhao, M. H.; Bonnell, D. A.; Vohs, J. M. Surf. Sci. 2009, 603,
284–290.
60
2
0
(36)
Yang, W. C.; Rodriguez, B. J.; Gruverman, A.; Nemanich, R. J.
J. Phys. Condens. Matter 2005, 17, S1415–S1426.
(58)
(59)
Dunn, S. J. Appl. Phys. 2003, 94, 5964–5968.
(
37)
38)
Dunn, S.; Jones, P. M.; Gallardo, D. E. J. Am. Chem. Soc. 2007,
Garrity, K.; Kolpak, A. M.; Ismail-Beigi, S.; Altman, E. I. Adv.
Mater. 2010, 22, 2969–2973.
1
29, 8724–8728.
(
Dunn, S.; Tiwari, D.; Jones, P. M.; Gallardo, D. E. J. Mater.
Chem. 2007, 17, 4460–4463.
2
5
65
(
39)
40)
Jones, P. M.; Dunn, S. J. Phys. D. Appl. Phys. 2009, 42, 065408.
Zhao, M. H.; Bonnell, D. A.; Vohs, J. M. Surf. Sci. 2008, 602,
(
2
849–2855.
(
(
(
41)
42)
43)
Zhao, J.; Wu, T.; Wu, K.; Oikawa, K.; Hidaka, H.; Serpone, N.
Environ. Sci. Technol. 1998, 32, 2394–2400.
3
3
4
0
5
0
Hashimoto, K.; Hiramoto, M.; Sakata, T. Chem. Phys. Lett.
1
988, 148, 215–220.
Butler, M. A.; Ginley, D. S. J. Electrochem. Soc. 1978, 125,
28–232.
2
(44)
Kim, Y. Il; Atherton, S. J.; Brigham, E. S.; Mallouk, T. E. J.
Phys. Chem. 1993, 97, 11802–11810.
(
45)
46)
Butler, M. A. J. Appl. Phys. 1977, 48, 1914.
(
Fan, H.; Li, H.; Liu, B.; Lu, Y.; Xie, T.; Wang, D. ACS Appl.
Mater. Interfaces 2012, 4, 4853–4857.
(47)
Pearson, R. G. Inorg. Chem. 1988, 27, 734–740.
This journal is © The Royal Society of Chemistry [year]
Journal Name, [year], [vol], 00–00 | 9