- Email: [email protected]
Photocatalytic treatment of water containing imazalil using an immobilized TiO2 photoreactor
Accepted Manuscript Title: Photocatalytic treatment of water containing imazalil using an immobilized TiO2 photoreactor Author: Dunia E. Santiago M.R. Espino-Est´evez Gabriel V. Gonz´alez J. Ara˜na O. Gonz´alez-D´ıaz J.M. Do˜na-Rodr´ıguez PII: DOI: Reference:
S0926-860X(15)00179-9 http://dx.doi.org/doi:10.1016/j.apcata.2015.03.021 APCATA 15311
To appear in:
Applied Catalysis A: General
Received date: Revised date: Accepted date:
8-12-2014 28-2-2015 18-3-2015
Please cite this article as: D.E. Santiago, M.R. Espino-Est´evez, G.V. Gonz´alez, J. Ara˜na, O. Gonz´alez-D´iaz, J.M. Do˜na-Rodr´iguez, Photocatalytic treatment of water containing imazalil using an immobilized TiO2 photoreactor, Applied Catalysis A, General (2015), http://dx.doi.org/10.1016/j.apcata.2015.03.021 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
1 2
Photocatalytic treatment of water containing imazalil using an immobilized TiO2 photoreactor Dunia E. Santiago, M.R. Espino-Estévez, Gabriel V. González, J. Araña, O. GonzálezDíaz and J.M. Doña-Rodríguez.
Abstract
us
3 4 5 6
cr
ip t
Grupo de Fotocatálisis y Espectroscopía Aplicada al Medioambiente-FEAM (Unidad Asociada al ICMSE, Centro Mixto C.S.I.C.-USE), CIDIA-Dpto. de Química, Edificio Polivalente I del Parque Científico Tecnológico, Universidad de Las Palmas De Gran Canaria, Campus Universitario de Tafira, 35017, Las Palmas, Spain. Corresponding autor: [email protected]
In this study, the photoactivity of commercial and lab-made TiO2 when immobilized on different supports, namely borosilicate glass, alumina foam and refractory brick,
8
was examined for the removal of the fungicide imazalil from different water
9
matrices. Alumina foam provided the largest exposed photocatalyst surface but
an
7
degradation kinetics were not significantly improved by the use of this material.
11
TiO2 coatings were also subjected to thermal treatment at 450°C to improve
12
adhesion to the support and exhibited higher photocatalytic mineralization, at levels
13
comparable to the conventional suspended system.
14
However, successive photocatalyst reuse led to its deactivation. Different
15
regeneration methods were studied for the TiO2 films and it was concluded that
16
deionized water washes were the most effective regeneration procedure.
17
Finally, a typical industrial wastewater containing imazalil was successfully treated
18
using TiO2 supported on borosilicate glass under solar irradiation.
Ac ce
pt
ed
M
10
Keywords: solar photocatalysis, supported catalysts, dip-coating, imazalil removal. 1. Introduction
19 20 21
Imazalil is a fungicide that is widely applied in fruit and vegetable packing industries
22
to combat a variety of fungal diseases. Its range of application is broader than that of
23
other fungicides and it is active against strains that are resistant to other pesticides
24
[1]. As the disposal of water contaminated with toxic substances like imazalil into
25
sewage systems can have a severe environmental impact, various regulations have
Page 1 of 35
been introduced to control fungicide MCLs (maximum contaminant levels) in
27
wastewater. In this respect, Council Directive 98/83/EC on the quality of water
28
intended for human consumption states that total pesticide concentration cannot be
29
higher than 0.05 mg·L-1.
30
Due to its low biodegradability, conventional wastewater treatment plants are
31
unable to eliminate substances like imazalil. Adsorption methods and advanced
32
oxidation processes have been reported as alternatives for its removal from
33
deionized water solutions [2-6], but very few studies have focused on the
34
elimination of this fungicide from real wastewaters [7-9]. In this respect,
35
heterogeneous TiO2 photocatalysis has been reported as a possible technique for the
36
degradation and mineralization of imazalil in industrial wastewater [9].
37
Given the expense involved in separation of nano-sized TiO2 particles from treated
38
water and the difficulties faced in terms of photocatalyst reuse, there has been
39
growing interest in immobilization of the photocatalyst on inert solid supports [10].
40
However, supported TiO2 has been mainly reported for the treatment of emerging
41
pollutants or low pollutant concentrations [11-13]. This is because when the catalyst
42
is immobilized, there is an inherent decrease in the surface area available for
43
reaction and thus reaction rates are lower [14]. Different porous supports have been
44
studied to try to mitigate this problem [15-18], although the support may introduce
45
interfering species into the photocatalytic system [16].
46
Among the immobilization techniques available, dip-coating is a very simple
47
procedure which offers several advantages over others, including its low cost and
48
high film uniformity [19-20].
49
The purpose of this work was to study the performance of TiO2 immobilized on
50
different supports in the degradation and mineralization of 50 mg·L-1 imazalil using
51
deionized water (DW) and a simulated/synthetic wastewater (SW). Photocatalyst
52
stability was evaluated using the material under the same operating conditions over
53
several consecutive cycles. Finally, lab-scale studies were compared with solar
54
experiments.
Ac ce
pt
ed
M
an
us
cr
ip t
26
Page 2 of 35
2. Experimental
56 57 58
2.1 Reagents/Chemicals The commercial imazalil Fruitgard-IS-7.5 was used for this study. pH was adjusted
59
with diluted H2SO4 and NaOH aqueous solutions. Sodium chloride (NaCl), aluminium
60
sulphate (Al2(SO4)3·18H2O) and calcium hydroxide (Ca(OH)2) from Panreac were used
61
as PRS reagents for SW preparation. Ethanol (≥99.5%) from Panreac was employed
62
as TiO2 dispersant agent for the immobilization procedure.
cr
ip t
55
2.2 Preparation of the TiO2 films
The general procedure applied is described elsewhere [11]. In brief, immobilization
66
of the TiO2 powder was carried out by dip-coating procedure using a KSV-DC Dip-
67
Coater (KSV Instruments). The inert support was submerged in a TiO2-ethanol
68
suspension for 2 min and then withdrawn from the suspension for 4 min to dry the
69
surface and thereby ensure correct fixation of the catalyst. This procedure or cycle
70
was repeated 80 times. Suspensions of 2 g·L-1 and 4 g·L-1 were used for the
71
commercial and lab-made catalysts, respectively. In addition, 0.2 mL of a 0.1 M HNO3
72
aqueous solution was added to the lab-made photocatalyst suspensions to enhance
73
particle disaggregation. The mass of deposited TiO2 was measured using an
74
analytical balance (A&D HR-200 with 1 mg ±0.1 mg precision).
75
Three types of support were considered in this study, borosilicate glass, refractory
76
brick and alumina foam. Each support had a total covered surface area of
77
approximately 110 cm2.
78
All the coatings were thermally treated at 105°C for 2 h. These were named D105. A
79
series of thermally treated coatings were subjected to further heat treatment at
80
450°C for 2 h to increase adherence of the catalysts to the support [11]. These were
81
named D450.
82
Adherence of the coatings was evaluated by vigorous washing with deionized water.
83
Catalyst detachment was determined by turbidity measurements. For this purpose, a
84
turbidity vs. catalyst concentration calibration curve was used.
Ac ce
pt
ed
M
an
us
63 64 65
Page 3 of 35
85
2.3 Photocatalytic experiments Laboratory scale
88
Adsorption and degradation tests were conducted in a 300 mL photoreactor with
89
recirculation. The reactor consisted of two concentric cylindrical tubes, with the
90
photocatalyst fixed to the outer surface of the inner tube. The TiO2 fixation
91
procedure followed that described in Section 2.2.
92
Two sets of Solarium Philips HB175 lamps equipped with four 15W Philips CLEO
93
fluorescent tubes with emission spectrum from 300 to 400 nm (maximum around
94
365 nm) (9 mW) were employed as UV light source. Before irradiation was initiated,
95
the imazalil solution was pumped through the reactor for 15 minutes in the dark to
96
reach the adsorption-desorption quasi-equilibrium. The pump used for recirculation
97
purposes was a Resun SP-500 with a 90 L·h-1 flow rate.
98
Adsorption equilibrium experiments were performed at ambient temperature (22 ±
99
1°C).
M
an
us
cr
ip t
86 87
Blank experiments were carried out with the photocatalyst in suspension for
101
comparison purposes. For this, degradation tests were performed in the same
102
recirculation reactors, filled with 300 mL of the pollutant aqueous solution and 1 g·L-1
103
of photocatalyst. The samples were filtered using 0.45 µm syringe filters before
104
analysis.
105
Solar experiments
106
For solar experiments, the 300 mL photoreactor was exposed to natural sunlight.
107
Solar ultraviolet radiation was measured with a UV-A radiometer (Acadus 85-PLS).
108
This radiometer includes an LS-3200 integrator to provide the accumulated energy
109
E(t) received by the total irradiated surface area of the photoreactor (in W·h). The
110
relationship between the experiment duration time (t), the total volume of the
111
reactor (V), the average instantaneous irradiance flux (UV), the collector surface area
112
(A) and the accumulated energy E(t) is:
Ac ce
pt
ed
100
113 114
Page 4 of 35
115
A statistical treatment was performed on the data presented in this work. The
116
standard errors were calculated using 95% confidence limits.
117
2.4 FTIR studies For the FTIR (Fourier Transform Infrared) determinations, a FTIR Thermo Scientific
120
Nicolet iS10 spectrometer was used at intervals of 4000-1000 cm-1. The catalyst films
121
were placed between two CaF2 windows.
122
The ammonia/catalyst surface interaction was studied to determine the presence
123
and modification of Lewis or Brönsted acid centres. The experimental procedure
124
followed a similar method to that described in [21]. The system consisted of a vessel
125
containing a 25% wt. ammonia solution which was continuously air-bubbled at a
126
flow rate of 150 mL·min−1. The resulting air containing ammonia was introduced into
127
a 15 cm long, 4 mm diameter cylindrical glass reactor containing the catalyst for its
128
adsorption and the photocatalyst was then placed between the two CaF2 windows
129
for the FTIR measurements. 2.5 Analytical determinations
M
an
us
cr
ip t
118 119
Concentrations of imazalil at different reaction times were HPLC-measured using a
133
Supelco Discovery C18 column (25 cm x 4.6 mm ID, 5 µm particles) and an
134
acetonitrile-10mM KH2PO4 solution (45:55) with 100 mg·L-1 of sodium 1-
135
octanesulfonate as mobile phase (adjusted to pH 3 with phosphoric acid), using a UV
136
detector (λ = 225 nm). Quantification was performed using the least-squares fit
137
method. The detection and quantification limits for imazalil were 0.05 mg·L-1 and
138
0.15 mg·L-1, respectively. The adjusted R2 was 0.998.
139
Total organic carbon (TOC) was measured using a Shimadzu TOC-L analyser. Turbidity
140
was measured with a Velp Scientifica TB1 portable turbidimeter.BET surface area
141
measurements were carried out by N2 adsorption at 77K using a Micromeritics
142
Gemini instrument.
143
Diffuse reflectance spectra were recorded for all samples on a Varian Cary 5
144
spectrophotometer and the Kubelka–Munk function, F(R∞), was applied to obtain
145
the band-gaps.
Ac ce
pt
ed
130 131 132
Page 5 of 35
146
X-ray diffraction (XRD) patterns were obtained by using a Siemens D-500
147
difractometer (Cu Kα,λ = 1.5432
148
estimated from the line broadening of the corresponding X-ray diffraction peaks by
149
using the Scherrer equation.
150
Scanning electron microscopic (SEM) analyses were performed on a JSM-5400 Jeol
151
apparatus equipped with an X-ray dispersive energy (EDX) analyzer.
ip t
). Crystallite sizes in the different phases were
3. Results and discussion
153 154 155
3.1 Characterization of the films
The photocatalysts used in the experiments and their properties are listed in Table 1.
156
These characteristics remained unchanged after fixation of the photocatalysts to the
157
different supports. No thermal transformation from anatase to rutile phase was
158
observed at 450°C. Adequate adherence was confirmed for all supported TiO2.
159
Figure 1 shows the SEM images obtained for the P25-D450 films deposited on
160
borosilicate glass, alumina and refractory brick. Figures 1a and b show the
161
photocatalyst deposited on borosilicate glass. It can be observed that the resulting
162
film is uniform, with the support completely covered.
163
Figures 1c and d show the photocatalyst deposited on alumina. In this case the
164
distribution of TiO2 on the support is irregular and non-covered areas are detected.
165
The TiO2 film thickness is, in general, variable, although regular thicknesses of
166
around 5 µm were obtained in some areas with a more homogeneous distribution
167
(Figure 1d). This thickness is similar to that obtained for the TiO2 films supported on
168
borosilicate glass (6 µm) (Figure 1a).
169
The distribution of TiO2 deposited on refractory brick can be seen in Figure 1e. EDX
170
mapping is included in order to better discern the TiO2 deposits. Results indicate that
171
TiO2 distribution is random and that other components, such as Al, Fe, Si, K and Mg
172
are present on the brick surface. These elements are those employed in the
173
manufacturing of the brick. Figure 1f shows areas with the highest TiO2 coverage.
174
The thickness of the film in this case is around 10 µm, but numerous fissures can be
175
observed.
Ac ce
pt
ed
M
an
us
cr
152
Page 6 of 35
176
Figures 1 g and h show better detail of Evonik P25 alone and Evonik P25-D450.
177
Particle aggregation can be denoted in the D450 catalyst.
179 180 181 182
3.2 FTIR studies 3.2.1 Adsorption of water onto the photocatalysts FTIR spectra of the photocatalysts alone and when supported on borosilicate glass
ip t
178
(both D105 and D450) were analyzed. The results described below correspond to the
184
Evonik P25 catalyst.
185
The band observed in all the spectra at 1640 cm−1 is attributed to the water bending
186
mode (ν2). The broad band between 3650 and 3000 cm−1 is attributed to the
187
asymmetrical (ν3) and symmetrical (ν1) vibration modes of water, which represent
188
isolated molecules interacting via hydrogen bonds [22, 23]. The shape and relative
189
intensity of the bands attributed to vibrations ν3 and ν1 with respect to the
190
corresponding vibration ν2 differ depending on the thermal treatment the
191
photocatalyst was subjected to (see Figure 2). In this respect, the water layer on the
192
surface seems to decrease for the D105 and D450 photocatalysts, especially for the
193
latter. This indicates that thermal treatment may lead to particle agglomeration and,
194
consequently, a lower exposed surface available for adsorption of different species,
195
such as water. These results agree with that observed from SEM studies (Figures 1 g
196
and h).
us
an
M
ed
pt
Ac ce
197
cr
183
198
3.2.2 Surface acid centres
199
The modifications to the catalyst surface when deposited and calcined (see Section
200
3.2.1) may indicate changes to the adsorption and photocatalytic active centres. For
201
this reason, ammonia interaction with the catalyst surface was studied in order to
202
determine the presence and/or modification of Lewis or Brönsted acid centres for
203
the photocatalyst alone and when supported on borosilicate glass (D105 and D450).
204
Figure 3 shows the spectra obtained after ammonia interaction with the catalyst
205
surface in the dark and at different illumination times. Spectra are shown for Evonik
206
P25. After ammonia adsorption, bands attributed to Lewis acid centres (≈ 1200 cm-1),
207
Brönsted acid centres (≈ 1450 cm-1) and breaking centres (1340 and 1320 cm−1) have
Page 7 of 35
been detected [24]. In our study, the relative intensities of these bands differed
209
depending on the thermal treatment the catalyst had been subjected to. Slight shifts
210
of the bands attributed to Lewis and Brönsted acid sites are also observed. The
211
previously described agglomeration of the D450 photocatalyst could be responsible
212
for the modification of the acid sites.
213
Illumination of the photocatalysts after ammonia adsorption indicates that different
214
photoproducs are produced for the D450 catalyst if compared to those generated for
215
Evonik P25 alone and the D105 system.
216
For the last systems, a progressive reduction of the band attributed to Lewis acid
217
centres (≈ 1200 cm-1) is observed, and this is accompanied by the formation of
218
adsorbed nitrate and nitrite (bands at 1564 and 1192 cm-1, respectively). This can be
219
seen in Figures 3a and b.
220
On the contrary, new bands at 1400 and 1312 cm-1 (Figure 3c) appear for the D450
221
catalyst after only 5 minutes irradiation. These bands correspond to HNO3/H2O
222
mixtures, as described in [25]. These results indicate a higher activity and the
223
presence of different photoactive species for the D450 system.
224
Addionally, literature references indicate that high temperature thermal treatment
225
of TiO2 can lead to modification of its surface structure [26]. Such modifications
226
could generate new surface species, such as O2-, O22- or O-, which may affect the
227
activity of the photocatalysts.
228
Lewis acid centres have been correlated by other authors with a higher oxygen
229
adsorption on the catalyst surface and, consequently, higher O2·- production [27].
230
This would inhibit the electron-hole recombination rate. In this study, a higher
231
proportion of Lewis acid centres are detected for the D450 photocatalyst, which in
232
addition presents a higher activity towards ammonia oxidation, as was previously
233
described. It can be therefore affirmed that the presence of more Lewis acid centres
234
is associated with a higher activity.
Ac ce
pt
ed
M
an
us
cr
ip t
208
235 3.3 Preliminary studies: photolysis and adsorption of imazalil
236 237 238
Photolysis of imazalil has been previously reported at pH 7 [5, 21]. Adsorption of
239
imazalil was evaluated at different pH, namely natural pH (3.9), 5 and 7. It was
Page 8 of 35
observed that adsorption was enhanced at pH 7 for all photocatalysts (data not
241
shown). In agreement with previous studies [2, 4-5], adsorption of imazalil at low pH
242
values was greatly hindered by electrostatic repulsion between the photocatalysts
243
(pHPZC between 5.2-7.8, see Table 1) and the imazalil (pKa = 6.54) and can be
244
considered negligible.
245
It should be noted that adsorption was higher for the supported photocatalysts at
246
pH 7 after the thermal treatments. In this respect and as can be seen in Figure 4 for
247
studies at pH 7, the amount of imazalil adsorbed after the adsorption equilibrium
248
had been established was up to 3 times higher (expressed as mgIMZ·gcat-1) for the
249
photocatalysts which had been supported and treated at 450°C (D450) when
250
compared to the photocatalyst alone (in suspension). This is attributed to surface
251
modification as a result of the thermal treatment (as discussed in Section 3.2-FTIR
252
studies): the increase in surface agglomeration and the surface active sites
253
modification seemed to enhance imazalil adsorption at pH 7. This was most clearly
254
observed in the following order: Evonik P25 > Evonik P90 > EST-1023t.
255 256
3.4 Screening of photocatalysts Figure 5 shows the imazalil degradation profiles and the percentage of
257
mineralization after 120 min of irradiation, for a 25 mg·L-1 imazalil solution at pH 7,
258
using the different photocatalysts considered in this study in suspension (S) or
259
supported on borosilicate glass after thermal treatment at 105°C (D105) or 450°C
260
(D450).
261
Imazalil degradation was significantly slower for the supported systems than for the
262
suspended one. This is due to the lower photocatalyst surface area available in the
263
supported configurations. Despite this, IMZ conversion was above 96% in all cases
264
after 120 minutes irradiation. It should be noted that, with respect to the supported
265
systems, degradation rates were higher for the photocatalysts subjected to thermal
266
treatment at 450°C.
267
Table 1 shows the deposited mass of TiO2 for each of the supported systems. It can
268
be seen that the mass of TiO2 varies from one system to another. For this reason, the
269
apparent rate constant referred to catalyst mass was determined; taking into
270
account that imazalil degradation follows pseudofirst order kinetics. Results, shown
Ac ce
pt
ed
M
an
us
cr
ip t
240
Page 9 of 35
in Table 2, must be compared to those shown in Figure 5, where the photocatalyst
272
deposited mass is not considered. In this sense, although it is clear from Figure 5 that
273
imazalil degradation is slower for the deposited systems, from the results in Table 2
274
can be seen that, if the catalyst mass is considered, the apparent degradation rate
275
constant is very similar or even higher for the D450 configuration when compared to
276
the suspended (S) one. This agrees with FTIR studies, as was described in Section
277
3.2.2. However, it must be emphasized that, whatever the deposited mass, the
278
photocatalytic activity will strongly depends on the amount of photocatalyst with an
279
effective irradiated surface.
280
Mineralization was lower for the supported system (D105) compared to the
281
suspension/slurry configuration, maybe due to the lower available TiO2 exposed
282
surface area. The decrease in mineralization yield was highest for the lab-made
283
catalyst, maybe due to its lower surface area.
284
For all catalysts it was observed that thermal treatment at 450°C (D450 systems)
285
resulted in higher photoactivity compared to the D105 systems. This effect was
286
particularly noticeable in the mineralization results which were very similar for the
287
D450 and suspended systems.
288
It has been reported that TiO2 treated at 450°C contains more surface oxygen than
289
that treated at 105°C or left untreated [28]. The presence of surface oxygen may
290
result in a higher photoactivity because the photoexcited electrons would be
291
scavenged by surface oxygen to form superoxides on TiO2, thus preventing electron-
292
hole recombination. In addition, the higher photoactivity of the D450 system may be
293
associated with surface modifications of the photocatalyst that are related to the
294
creation of oxygen adsorption centres or a different nature of oxidizing species
295
present at the surface, in accordance with the FTIR studies shown in Section 3.2.2.
296
As for the activity of the different catalysts, it was observed that while imazalil
297
degradation took place at a faster rate with the EST-1023t catalyst, mineralization
298
was slightly higher for the commercial Evonik P25. It was therefore decided to
299
perform the remaining experiments using the Evonik P25 catalyst.
Ac ce
pt
ed
M
an
us
cr
ip t
271
Page 10 of 35
3.5 Efficiency of TiO2 on the different supports The use of different supports was studied in order to investigate the effect of
302
increasing the exposed catalyst surface area. The studies were conducted using
303
Evonik P25 and applying thermal treatment at 105°C or 450°C for its fixation.
304
Table 3 shows some characteristics of the different materials employed as supports.
305
The data were provided by the support manufacturers.
306
The various materials support different amounts of TiO2 because of their different
307
pore size and structure. The highest deposited amount was obtained for the 50 ppi
308
alumina foam, followed by the fire brick and the glass tube, as shown in Table 4. This
309
is due to the higher porosity of the foam and fire brick which favours TiO2 adhesion.
310
The photocatalytic rate is believed to increase with increasing porosity of the
311
support [15].
312
In agreement with the literature [29-31], the photocatalytic activity of TiO2 films
313
supported on macroporous alumina foam was observed to increase with increasing
314
macroporosity of the substrate surface and so the highest catalytic activity was
315
found for TiO2 deposited on the 50 ppi foam.
316
Figure 6 shows the imazalil degradation profiles and mineralization percentages after
317
120 minutes illumination for the treatment of 25 mg·L-1 imazalil using TiO2 Evonik
318
P25 immobilized on different materials: borosilicate glass, fire brick and 50 ppi
319
alumina foam.
320
Degradation tests showed that mineralization was highest for the alumina foam
321
D105 system. However, this may be attributable to a slight detachment of the
322
photocatalyst, as was seen from the adhesion studies. Although the fire brick and
323
alumina foam provided higher TiO2 exposed area and higher deposited amounts,
324
mineralization was very similar for the D450 borosilicate glass and alumina 50 ppi
325
foam systems.
326
It should be noted that the fire brick results were not as good as expected. This may
327
be attributable to a high concentration of inorganic ions on this support surface
328
which may interact with TiO2 (see Section 3.1 for SEM images). Some studies have
329
correlated the decline in activity with the presence of cationic impurities in the TiO2
330
layer as a consequence of the thermal treatments required to improve adhesion of
331
the titania layer onto the support [16, 32].
Ac ce
pt
ed
M
an
us
cr
ip t
300 301
Page 11 of 35
3.6 Photocatalytic degradation of imazalil in synthetic agro-industrial water
334
Imazalil concentrations of up to 50 mg·L-1 have been reported in agro-industrial
335
wastewater effluents. For comparison purposes with a previous work [9], the
336
synthetic wastewater employed in this study contained 50 mg·L−1 imazalil, 100
337
mg·L−1 chloride (as NaCl), 300 mg·L−1 sulphate (as Al2(SO4)3·18H2O) and 20 mg·L−1
338
calcium (as Ca(OH)2). This simulates the composition generally found in the
339
wastewater samples analyzed from a collaborating banana packing company.
340
Photocatalysis is affected by the composition of the water. A high concentration of
341
inorganic ions can interfere, physically and chemically, with the photocatalysts [33].
342
Figure 7 shows the degradation profiles of imazalil, as well as the percentage of
343
mineralization after 120 minutes of irradiation for 50 mg·L−1 of imazalil in deionized
344
water and synthetic wastewater using different systems considered in this study.
345
In general terms, it was observed that the ionic strength of the SW matrix did not
346
hinder imazalil degradation or mineralization under the studied conditions. This is
347
because the working pH of 7 was established to be higher than the photocatalyst
348
pHPZC (see Table 1) and therefore anion adsorption was not favoured [9, 34-35].
349
From the results it can be concluded that both borosilicate glass and alumina foam
350
could be adequate supports for TiO2 in the treatment of waters contaminated with
351
imazalil.
cr
us
an
M
ed
pt
3.7 Reusability and recovery of TiO2 films
Ac ce
352 353
ip t
332 333
354
The results of catalyst activity over a prolonged period of use in synthetic
355
wastewater are shown in Figure 8. The treatment of 50 mg·L-1 imazalil in SW for 240
356
minutes under irradiation was repeated six times to evaluate the durability and
357
photoactivity of the immobilized photocatalyst under these operating conditions.
358
For the suspended system, no loss in activity was observed after six cycles. However,
359
a fall in photocatalytic activity was observed with increasing aging time of the TiO2
360
film for the supported systems. This phenomenon was accompanied by the
361
appearance of a pale yellow colour on the TiO2 film indicating the presence of
362
adsorbed intermediates on the titania active sites. These observations agree with
Page 12 of 35
363
those reported by other authors [36-38]. Deactivation of the photocatalyst after use
364
has also been attributed to the formation of oxygen vacancies after photocatalytic
365
reaction [36]. Oxygen vacancies are formed according to reaction (1), as reported in
366
[37].
367
h+ + 1/2O2-(lattice) → 1/4O2 + vacancy
368
The fall in mineralisation as the photocatalyst was reused was higher using the
369
alumina support than the borosilicate glass and, significantly, this was accompanied
370
by a much more noticeable yellowing. Accordingly, due to the higher mineralization
371
achieved during all cycles for the borosilicate D450 system, this was chosen as the
372
best support for the TiO2 photocatalyst.
373
In order to maintain high photocatalytic treatment effectiveness, the catalyst should
374
be cleaned or replaced. Several regeneration methods have been described for TiO2
375
films, including, amongst others, deionized water washes, NaOH washes, the
376
combination of H2O2 or air with UV irradiation or recalcination of the film [39-42].
377
Of these, recalcination of the film is not recommended because heat treatment may
378
cause further agglomeration of the TiO2 nanoparticles.
379
The regeneration effect on photocatalytic activity of three of these methods (water
380
washes, air with UV irradiation and recalcination) was tested in this study.
381
As can be seen in Figure 9, this technique did not result in favourable regeneration of
382
the TiO2 film. The air plus UV irradiation was also ineffective for film regeneration
383
film.
384
However, water washing was shown to be an effective technique to recover
385
deposited TiO2 activity. It has been reported to effectively remove the oxygen
386
vacancies that may have been responsible for TiO2 deactivation [36]. We have
387
observed that water washing of the TiO2 films at ambient temperature fully recovers
388
TiO2 activity if the washing procedure is performed after every use. Furthermore, the
389
films treated this way may be used daily for months without deactivation.
Ac ce
pt
ed
M
an
us
cr
ip t
(1)
Page 13 of 35
390
3.8 Solar experiments The optimal system determined at laboratory scale for the treatment of synthetic
393
agro-industrial wastewater containing imazalil was applied under solar irradiation in
394
order to validate the results.
395
Figure 10 shows the degradation and mineralization profiles of imazalil against
396
accumulated energy. It should be noted that the necessary time for the effluent
397
containing imazalil to be treated strongly depends on the climate conditions.
398
As can be observed from Figure 10, 75% mineralization was reached after around
399
110 kJ·L-1 of accumulated energy, which was equivalent to 8 hours at our location
400
considering perfect sunny days. IMZ was completely eliminated after 82 kJ·L-1
401
accumulated energy.
an
us
cr
ip t
391 392
4. Conclusions
TiO2 films, deposited on several supports by means of dip-coating of commercial and
405
lab-made photocatalysts, showed good adhesion to the substrate surface after
406
thermal treatment at 450°C. This treatment also resulted in enhanced photocatalytic
407
activity for all the photocatalysts tested. Of the photocatalysts studied, Evonik P25
408
and EST-1023t gave the best degradation results, and Evonik P25 and P90 returned
409
the best mineralization.
410
Coatings were studied next on different supports, namely borosilicate glass,
411
refractory brick and alumina foam, to increase the exposed surface area of the
412
photocatalyst and the amount of deposited mass. In view of its high mineralisation
413
results, the Evonik P25 photocatalyst was employed for this purpose. TiO2 supported
414
on alumina foam exhibited a higher imazalil removal activity, although mineralization
415
was similar for TiO2 films on alumina and borosilicate glass. However, the activity of
416
TiO2 on alumina was considerably diminished when the support was reused several
417
times for the treatment of synthetic agro-industrial wastewater (containing inorganic
418
ions) as water matrix.
419
Solar experiments were carried out and confirmed that TiO2 can be deposited on
420
borosilicate glass to efficiently remove up to 50 mg·L-1 imazalil from synthetic agro-
421
industrial wastewaters.
Ac ce
pt
ed
M
402 403 404
Page 14 of 35
422 423 424 425
Acknowledgements We thank the University of Las Palmas de Gran Canaria for its funding through the PhD Grant Program, the Spanish Ministry of Science and Innovation for its financial
427
support through the PhD Studentship BES-2010-036537, MINECO (Ministry of
428
Economy and Competitiveness, Government of Spain) for funding of the NANOBAC
429
project (IPT-2011-1113-310000) and the ERDF for co-funding with MINECO the
430
Infrastructure Project 2010-3E UNLP10-3E-726.
cr
ip t
426
us
5. References
[1] M. Duque, J.M. Torres, J.J. Oramas, J. Fernández, Pudrición de Corona en el
432
Plátano Canario, first ed. Coplaca, Tenerife, 2004.
433
[2] M.A. Martín-González, O. González-Díaz, P. Susial, J. Araña, J.A. Herrera-Melián,
434
J.M. Doña-Rodríguez, J. Pérez-Peña, Chem. Eng. J. 245 (2014) 348–358.
435
[3] A.K. Genena, D.B. Luiz, W. Gebhardt, R.F.P.M. Moreira, H.J. José, H.F. Schröder,
436
Ozone Sci. Eng. 33 (2011) 308–328.
437
[4] R. Hazime, C. Ferronato, L. Fine, A. Salvador, F. Jaber, J.M. Chovelon, Appl. Catal.,
438
B. 126 (2012) 90–99.
439
[5] D.E. Santiago, J.M. Doña-Rodríguez, J. Araña, C. Fernández-Rodríguez, O.
440
González-Díaz, J. Pérez-Peña, A.M.T. Silva, Appl. Catal., B. 138–139 (2013) 391–400.
441
[6] S. Malato, J. Blanco, D.C. Alarcón, M.I. Maldonado, P. Fernández-Ibáñez, W.
442
Gernjak, Catal. Today 122 (2007) 137–149.
443
[7] D.E. Santiago, E. Pulido Melián, C. Fernández-Rodríguez, J.A. Ortega Méndez, S.O.
444
Pérez-Báez, J.M. Doña-Rodríguez, Green Sustain. Chem. 1 (3) (2011) 39–46.
445
[8] I. Carra, S. Malato, M. Jiménez, M.I. Maldonado, J.A. Sánchez Pérez, Chem. Eng. J.
446
235 (2014) 132–140.
447
[9] D.E. Santiago, J. Araña, O. González-Díaz, M.E. Alemán-Domínguez, A.C. Acosta-
448
Dacal, C. Fernández-Rodríguez, J. Pérez-Peña, José M. Doña-Rodríguez, Appl. Catal.,
449
B 156-157 (2014) 284-292.
450
[10] R. L. Pozzo, Miguel A. Baltanfis, Alberto E. Cassano, Catal. Today 39 (1997) 219-
451
231.
Ac ce
pt
ed
M
an
431
Page 15 of 35
[11] M.R. Espino-Estévez, J.M. Doña-Rodríguez, C. Fernández-Rodríguez, O.
453
González-Díaz, J. Pérez-Peña, in: 7th European Meeting on Solar Chemistry and
454
Photocatalysis: Environmental Applications (SPEA 7) Oporto, Portugal, 2012, pp. 89.
455
[12] D. Li, H. Zheng, Q. Wang, X. Wang, W. Jiang, Z. Zhang, Y. Yang, Sep. Pur. Tech.
456
123 (2014) 130–138.
457
[13] N. Miranda-García, S. Suárez, B. Sánchez, J.M. Coronado, S. Malato, M. Ignacio
458
Maldonado, Appl. Catal., B 103 (2011) 294–301.
459
[14] J.A. Byrne, B.R. Eggins, N.M.D. Brown, B. McKinney, M. Rouse, Appl. Catal., B 17
460
(1998) 25-36.
461
[15] A. Rachel, B. Lavedrine, M. Subrahmanyam, P. Boule, Catal. Commun. 3 (2002)
462
165–171.
463
[16] A. Rachel, M. Subrahmanyam, P. Boule, Appl. Catal., B 37 (2002) 301–308.
464
[17] M. Radeka, S. Markov, E. Loncar, O. Rudic, S. Vucetic, J. Ranogajec, J. Eur.
465
Ceram. Soc. 34 (2014) 127–136.
466
[18] A. Gutiérrez-Alejandre, M. Trombetta, G. Busca, J. Ramirez, Microporous Mater.
467
12 (1997) 79-91.
468
[19] C. Guillard, B. Beaugiraud, C. Dutriez, J. Herrmann, Appl. Catal., B 39 (2002) 331–
469
342.
470
[20] C. Legrand-Buscema, C. Malibert, S. Bach, Thin Solid Films 418 (2002) 79–84.
471
[21] J. Araña, J.M. Doña-Rodrı́guez, O. González-Dı́az, E. Tello Rendón, J.A. Herrera
472
Melián, G. Colón, J.A. Navı́o, J. Pérez Peña, J. Mol. Catal. A: Chem. 215 (2004) 153-
473
160.
474
[22] H.A. Al-Abadleh, V.H. Grassian, Langmuir 19 (2) (2003) 341–347.
475
[23] L. Cammarata, S.G. Kazarian, P.A. Salterb, T. Welton. Phys. Chem. Chem. Phys. 3
476
(2001) 5192–5200.
477
[24] G. Martra. Appl. Catal., A, 200 (2000) 275-285.
478
[25] H. Yang, B.J. Finlayson-Pitts, J. Phys. Chem., A 105 (2001) 1890–1896.
479
[26] M.J. Elser, O. Diwald, J. Phys. Chem. C 116 (4) (2012) 2896–2903.
480
[27]M.I. Franch, J. Peral, X. Domènech, R.F. Howe, J.A. Ayllón, Appl. Catal., B 55
481
(2005) 105-113.
482
[28] J.C. Yu, J. Lin, D. Lo, S.K. Lam, Langmuir 16 (2000) 7304-7308.
Ac ce
pt
ed
M
an
us
cr
ip t
452
Page 16 of 35
[29] G. Plantard, V. Goetz, F. Correia, J.P. Cambon, Sol. Energy Mater. Sol. Cells 95
484
(2011) 2437-2442.
485
[30] G. Plesch, M. Gorbár, U.F. Vogt, K. Jesenák, M. Vargová, Mater. Lett. 63 (2009)
486
461-463.
487
[31] M. Vargová, G. Plesch, U.F. Vogt, M. Zahoran, M. Gorbár, K. Jesenák, Appl. Surf.
488
Sci. 257 (2011) 4678–4684.
489
[32] A. Fernández, G. Lassaletta, V.M. Jiménez, A. Justo, A.R. González-Elipe, J.M.
490
Herrmann, H. Tahiri, Y. Ait-Ichou, Appl. Catal., B 7 (1995) 49–63.
491
[33] A.G. Rincon, C. Pulgarin, Appl. Catal., B 51 (2004)283–302.
492
[34] W. Zhang, T. An, M. Cui, G. Sheng, J. Fu, J. Chem. Technol. Biotechnol. 80 (2005)
493
223-229.
494
[35] A. Aguedach, S. Brosillon, J. Morvan, E.l Kbir Lhadi, J. Hazard. Mater. 150 (2007)
495
250-256.
496
[36] X. Pan, N. Zhang, X. Fu, Y.J. Xu, Appl. Catal., A 453 (2013) 181-187.
497
[37] S. H. Szczepankiewicz, A. J. Colussi, M. R. Hoffmann, J. Phys. Chem., B 104 (2000)
498
9842-9850.
499
[38] D. Rubio, J.F. Casanueva, E. Nebot, J. Photochem. Photobiol., A 271 (2013) 16–
500
23.
501
[39] N. Miranda-García, S. Suárez, M. Ignacio Maldonado, S. Malato, B. Sánchez,
502
Catal. Today 230 (2014) 27-34.
503
[40] L. Nie, P. Zhou, J. Yu, M. Jaroniec, J. Mol. Catal. A: Chem. 390 (2014) 7-13.
504
[41] S. Liu, M. Lim, R. Amal, Chem. Eng. Sci. 105 (2014) 46–52.
505
[42] N. Madaan, S. Gatla, V.N. Kalevaru, J. Radnik, B. Lücke, A. Brückner, A. Martin, J.
506
Catal. 282 (2011) 103–111.
cr
us
an
M
ed
pt
Ac ce
507
ip t
483
Page 17 of 35
Figure 1. SEM images for TiO2 (Evonik P25) films deposited on: borosilicate glass (a and b), 50
508
ppi alumina foam (c and d), refractory brick (e and f) and detail of Evonik P25 untreated and
509
Evonik P25-D450.
510
Figure 2. FTIR spectra of Evonik P25 alone (P25), and after being supported on
511
borosilicate glass using thermal treatment at 105°C (P25-D105) and 450°C (P25-
512
D450).
513
Figure 3. FTIR spectra from the interaction of NH3 with Evonik P25 alone (a) and
514
after being supported on borosilicate glass using thermal treatment at 105°C (P25-
515
D105) b) and 450°C (P25-D450) c) under illumination at different times.
516
Figure 4. Adsorption of IMZ onto different TiO2 photocatalysts (initial concentration
517
25 mg·L-1 and initial pH 7). For the supported systems, the support was borosilicate
518
glass.
519
Figure 5. Degradation profiles (a) and % mineralization after 120 minutes (b) of 25
520
mg·L-1 IMZ using different TiO2 photocatalysts and configurations: suspended system
521
(S) and supported (D105 and D450) on borosilicate glass.
522
Figure 6. Degradation profiles (a) and % mineralization after 120 minutes (b) of 25
523
mg·L-1 IMZ using different TiO2 supports: borosilicate glass, fire brick and alumina
524
foam.
525
Figure 7. Degradation profiles (a) and % mineralization after 120 minutes (b) of 50
526
mg·L-1 IMZ using different TiO2 systems (suspended -S- or supported on alumina or
527
borosilicate glass - D450) and different water matrices: deionized water (DW) or
528
synthetic wastewater (SW).
529
Figure 8. IMZ conversion (%) (a) and % mineralization after 240 minutes (b) of 50
530
mg·L-1 IMZ using suspended (S), borosilicate D450 (B-D450) and alumina D450 (A-
531
D450) systems and synthetic wastewater (SW).
532
Figure 9. IMZ conversion (%) (a) and % mineralization after 240 minutes (b) of 50
533
mg·L-1 imazalil in SW using Evonik P25 supported on borosilicate glass (D450) on
534
the 1st cycle, 5th cycle and subsequent regeneration cycle.
535 536 537
Figure 10. IMZ and TOC evolution in the solar experiment using Evonik P25 supported on borosilicate glass (D450) for the treatment of 50 mg·L-1 IMZ in SW.
Ac ce
pt
ed
M
an
us
cr
ip t
507
Page 18 of 35
537 538
Table 1. Characteristics of the photocatalysts used in this work and deposited TiO2 mass
539
Table 2. Characteristics of the different supports.
540
Table 3. Deposited TiO2 mass on the different supports and adhesion studies.
for the different photocatalysts supported on borosilicate glass.
541
Ac ce
pt
ed
M
an
us
cr
ip t
542 543
Page 19 of 35
543 544
Table 2. Characteristics of the photocatalysts used in this work and deposited TiO2 mass for the different photocatalysts supported on borosilicate glass. Catalyst
Anatase/Rutile ratio (%) 80/20 86/14 70-80/30-20
Band Gap (eV) 3.18 3.29 2.96
Crystallite size (nm) Anatase Rutile 22.0 25.0 13 62.3 96.1
pHPZC 6.5 7.8 5.2
Deposited amount (mg) D105 D450 50 55 80.1 81.5 109.9 80.1
ip t
Evonik P25 Evonik P90 EST-1023t[5]
Specific surface area (m2·g-1) 52 100 13.5
Ac ce
pt
ed
M
an
us
cr
545 546
Page 20 of 35
546 547 548
Table 2. Apparent degradation rate constants for imazalil, (kIMZ/gcat) using different photocatalysts and systems. For the deposited systems, the support was borosilicate glass. kIMZ·g-1cat (min-1·g-1) S D105 D450 0.31 0.42 0.47 0.27 0.22 0.27 0.39 0.21 0.35
Photocatalyst
ip t
Evonik P25 Evonik P90 EST-1023t
Ac ce
pt
ed
M
an
us
cr
549 550
Page 21 of 35
Table 3. Characteristics of the different supports. Support
% Porosity
Apparent Density (g·cm-3)
Glass tube Fire brick Foam 50 ppi Foam 20 ppi Foam 10 ppi
16-20 ≥80 ≥80 ≥80
2.05 <0.5 <0.5 <0.5
Ac ce
pt
ed
M
an
us
cr
551 552
ip t
550
Page 22 of 35
552
Table 4. Deposited TiO2 mass on the different supports and adhesion studies. Deposited amount (mg)
Material Borosilicate glass Red brick Alumina foam (50 ppi)
D105 50 80 310.1
D450 55 80.1 304.6
Turbidity from adhesion tests (NTU) D105 D450 0.67 0.76 1.49 0.56 2.31 1.02
ip t
553
Ac ce
pt
ed
M
an
us
cr
554 555
Page 23 of 35
555 556
pt
ed
M
an
us
cr
ip t
The elimination of imazalil was proved for different water matrices. Suspended and supported TiO2 photocatalysis systems were compared. The effect of the water matrix was evaluated for the different systems. Different supports and thermal treatments were evaluated for the fixation of TiO2. Solar experiences confirmed the adequate treatment of synthetic wastewater.
Ac ce
557 558 559 560 561 562 563
Page 24 of 35
ed
pt
Ac ce
563
Page 25 of 35
us
an
M
cr
ip t
Ac ce p
te
d
M
an
us
cr
ip t
Figure 1
Page 26 of 35
Ac
ce
pt
ed
M
an
us
cr
i
Figure 2
Page 27 of 35
Ac
ce
pt
ed
M
an
us
cr
i
Figure 3
Page 28 of 35
Ac
ce
pt
ed
M
an
us
cr
i
Figure 4
Page 29 of 35
Ac
ce
pt
ed
M
an
us
cr
i
Figure 5
Page 30 of 35
Ac
ce
pt
ed
M
an
us
cr
i
Figure 6
Page 31 of 35
Ac
ce
pt
ed
M
an
us
cr
i
Figure 7
Page 32 of 35
Ac
ce
pt
ed
M
an
us
cr
i
Figure 8
Page 33 of 35
Ac
ce
pt
ed
M
an
us
cr
i
Figure 9
Page 34 of 35
Ac
ce
pt
ed
M
an
us
cr
i
Figure 10
Page 35 of 35
S0926-860X(15)00179-9 http://dx.doi.org/doi:10.1016/j.apcata.2015.03.021 APCATA 15311
To appear in:
Applied Catalysis A: General
Received date: Revised date: Accepted date:
8-12-2014 28-2-2015 18-3-2015
Please cite this article as: D.E. Santiago, M.R. Espino-Est´evez, G.V. Gonz´alez, J. Ara˜na, O. Gonz´alez-D´iaz, J.M. Do˜na-Rodr´iguez, Photocatalytic treatment of water containing imazalil using an immobilized TiO2 photoreactor, Applied Catalysis A, General (2015), http://dx.doi.org/10.1016/j.apcata.2015.03.021 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
1 2
Photocatalytic treatment of water containing imazalil using an immobilized TiO2 photoreactor Dunia E. Santiago, M.R. Espino-Estévez, Gabriel V. González, J. Araña, O. GonzálezDíaz and J.M. Doña-Rodríguez.
Abstract
us
3 4 5 6
cr
ip t
Grupo de Fotocatálisis y Espectroscopía Aplicada al Medioambiente-FEAM (Unidad Asociada al ICMSE, Centro Mixto C.S.I.C.-USE), CIDIA-Dpto. de Química, Edificio Polivalente I del Parque Científico Tecnológico, Universidad de Las Palmas De Gran Canaria, Campus Universitario de Tafira, 35017, Las Palmas, Spain. Corresponding autor: [email protected]
In this study, the photoactivity of commercial and lab-made TiO2 when immobilized on different supports, namely borosilicate glass, alumina foam and refractory brick,
8
was examined for the removal of the fungicide imazalil from different water
9
matrices. Alumina foam provided the largest exposed photocatalyst surface but
an
7
degradation kinetics were not significantly improved by the use of this material.
11
TiO2 coatings were also subjected to thermal treatment at 450°C to improve
12
adhesion to the support and exhibited higher photocatalytic mineralization, at levels
13
comparable to the conventional suspended system.
14
However, successive photocatalyst reuse led to its deactivation. Different
15
regeneration methods were studied for the TiO2 films and it was concluded that
16
deionized water washes were the most effective regeneration procedure.
17
Finally, a typical industrial wastewater containing imazalil was successfully treated
18
using TiO2 supported on borosilicate glass under solar irradiation.
Ac ce
pt
ed
M
10
Keywords: solar photocatalysis, supported catalysts, dip-coating, imazalil removal. 1. Introduction
19 20 21
Imazalil is a fungicide that is widely applied in fruit and vegetable packing industries
22
to combat a variety of fungal diseases. Its range of application is broader than that of
23
other fungicides and it is active against strains that are resistant to other pesticides
24
[1]. As the disposal of water contaminated with toxic substances like imazalil into
25
sewage systems can have a severe environmental impact, various regulations have
Page 1 of 35
been introduced to control fungicide MCLs (maximum contaminant levels) in
27
wastewater. In this respect, Council Directive 98/83/EC on the quality of water
28
intended for human consumption states that total pesticide concentration cannot be
29
higher than 0.05 mg·L-1.
30
Due to its low biodegradability, conventional wastewater treatment plants are
31
unable to eliminate substances like imazalil. Adsorption methods and advanced
32
oxidation processes have been reported as alternatives for its removal from
33
deionized water solutions [2-6], but very few studies have focused on the
34
elimination of this fungicide from real wastewaters [7-9]. In this respect,
35
heterogeneous TiO2 photocatalysis has been reported as a possible technique for the
36
degradation and mineralization of imazalil in industrial wastewater [9].
37
Given the expense involved in separation of nano-sized TiO2 particles from treated
38
water and the difficulties faced in terms of photocatalyst reuse, there has been
39
growing interest in immobilization of the photocatalyst on inert solid supports [10].
40
However, supported TiO2 has been mainly reported for the treatment of emerging
41
pollutants or low pollutant concentrations [11-13]. This is because when the catalyst
42
is immobilized, there is an inherent decrease in the surface area available for
43
reaction and thus reaction rates are lower [14]. Different porous supports have been
44
studied to try to mitigate this problem [15-18], although the support may introduce
45
interfering species into the photocatalytic system [16].
46
Among the immobilization techniques available, dip-coating is a very simple
47
procedure which offers several advantages over others, including its low cost and
48
high film uniformity [19-20].
49
The purpose of this work was to study the performance of TiO2 immobilized on
50
different supports in the degradation and mineralization of 50 mg·L-1 imazalil using
51
deionized water (DW) and a simulated/synthetic wastewater (SW). Photocatalyst
52
stability was evaluated using the material under the same operating conditions over
53
several consecutive cycles. Finally, lab-scale studies were compared with solar
54
experiments.
Ac ce
pt
ed
M
an
us
cr
ip t
26
Page 2 of 35
2. Experimental
56 57 58
2.1 Reagents/Chemicals The commercial imazalil Fruitgard-IS-7.5 was used for this study. pH was adjusted
59
with diluted H2SO4 and NaOH aqueous solutions. Sodium chloride (NaCl), aluminium
60
sulphate (Al2(SO4)3·18H2O) and calcium hydroxide (Ca(OH)2) from Panreac were used
61
as PRS reagents for SW preparation. Ethanol (≥99.5%) from Panreac was employed
62
as TiO2 dispersant agent for the immobilization procedure.
cr
ip t
55
2.2 Preparation of the TiO2 films
The general procedure applied is described elsewhere [11]. In brief, immobilization
66
of the TiO2 powder was carried out by dip-coating procedure using a KSV-DC Dip-
67
Coater (KSV Instruments). The inert support was submerged in a TiO2-ethanol
68
suspension for 2 min and then withdrawn from the suspension for 4 min to dry the
69
surface and thereby ensure correct fixation of the catalyst. This procedure or cycle
70
was repeated 80 times. Suspensions of 2 g·L-1 and 4 g·L-1 were used for the
71
commercial and lab-made catalysts, respectively. In addition, 0.2 mL of a 0.1 M HNO3
72
aqueous solution was added to the lab-made photocatalyst suspensions to enhance
73
particle disaggregation. The mass of deposited TiO2 was measured using an
74
analytical balance (A&D HR-200 with 1 mg ±0.1 mg precision).
75
Three types of support were considered in this study, borosilicate glass, refractory
76
brick and alumina foam. Each support had a total covered surface area of
77
approximately 110 cm2.
78
All the coatings were thermally treated at 105°C for 2 h. These were named D105. A
79
series of thermally treated coatings were subjected to further heat treatment at
80
450°C for 2 h to increase adherence of the catalysts to the support [11]. These were
81
named D450.
82
Adherence of the coatings was evaluated by vigorous washing with deionized water.
83
Catalyst detachment was determined by turbidity measurements. For this purpose, a
84
turbidity vs. catalyst concentration calibration curve was used.
Ac ce
pt
ed
M
an
us
63 64 65
Page 3 of 35
85
2.3 Photocatalytic experiments Laboratory scale
88
Adsorption and degradation tests were conducted in a 300 mL photoreactor with
89
recirculation. The reactor consisted of two concentric cylindrical tubes, with the
90
photocatalyst fixed to the outer surface of the inner tube. The TiO2 fixation
91
procedure followed that described in Section 2.2.
92
Two sets of Solarium Philips HB175 lamps equipped with four 15W Philips CLEO
93
fluorescent tubes with emission spectrum from 300 to 400 nm (maximum around
94
365 nm) (9 mW) were employed as UV light source. Before irradiation was initiated,
95
the imazalil solution was pumped through the reactor for 15 minutes in the dark to
96
reach the adsorption-desorption quasi-equilibrium. The pump used for recirculation
97
purposes was a Resun SP-500 with a 90 L·h-1 flow rate.
98
Adsorption equilibrium experiments were performed at ambient temperature (22 ±
99
1°C).
M
an
us
cr
ip t
86 87
Blank experiments were carried out with the photocatalyst in suspension for
101
comparison purposes. For this, degradation tests were performed in the same
102
recirculation reactors, filled with 300 mL of the pollutant aqueous solution and 1 g·L-1
103
of photocatalyst. The samples were filtered using 0.45 µm syringe filters before
104
analysis.
105
Solar experiments
106
For solar experiments, the 300 mL photoreactor was exposed to natural sunlight.
107
Solar ultraviolet radiation was measured with a UV-A radiometer (Acadus 85-PLS).
108
This radiometer includes an LS-3200 integrator to provide the accumulated energy
109
E(t) received by the total irradiated surface area of the photoreactor (in W·h). The
110
relationship between the experiment duration time (t), the total volume of the
111
reactor (V), the average instantaneous irradiance flux (UV), the collector surface area
112
(A) and the accumulated energy E(t) is:
Ac ce
pt
ed
100
113 114
Page 4 of 35
115
A statistical treatment was performed on the data presented in this work. The
116
standard errors were calculated using 95% confidence limits.
117
2.4 FTIR studies For the FTIR (Fourier Transform Infrared) determinations, a FTIR Thermo Scientific
120
Nicolet iS10 spectrometer was used at intervals of 4000-1000 cm-1. The catalyst films
121
were placed between two CaF2 windows.
122
The ammonia/catalyst surface interaction was studied to determine the presence
123
and modification of Lewis or Brönsted acid centres. The experimental procedure
124
followed a similar method to that described in [21]. The system consisted of a vessel
125
containing a 25% wt. ammonia solution which was continuously air-bubbled at a
126
flow rate of 150 mL·min−1. The resulting air containing ammonia was introduced into
127
a 15 cm long, 4 mm diameter cylindrical glass reactor containing the catalyst for its
128
adsorption and the photocatalyst was then placed between the two CaF2 windows
129
for the FTIR measurements. 2.5 Analytical determinations
M
an
us
cr
ip t
118 119
Concentrations of imazalil at different reaction times were HPLC-measured using a
133
Supelco Discovery C18 column (25 cm x 4.6 mm ID, 5 µm particles) and an
134
acetonitrile-10mM KH2PO4 solution (45:55) with 100 mg·L-1 of sodium 1-
135
octanesulfonate as mobile phase (adjusted to pH 3 with phosphoric acid), using a UV
136
detector (λ = 225 nm). Quantification was performed using the least-squares fit
137
method. The detection and quantification limits for imazalil were 0.05 mg·L-1 and
138
0.15 mg·L-1, respectively. The adjusted R2 was 0.998.
139
Total organic carbon (TOC) was measured using a Shimadzu TOC-L analyser. Turbidity
140
was measured with a Velp Scientifica TB1 portable turbidimeter.BET surface area
141
measurements were carried out by N2 adsorption at 77K using a Micromeritics
142
Gemini instrument.
143
Diffuse reflectance spectra were recorded for all samples on a Varian Cary 5
144
spectrophotometer and the Kubelka–Munk function, F(R∞), was applied to obtain
145
the band-gaps.
Ac ce
pt
ed
130 131 132
Page 5 of 35
146
X-ray diffraction (XRD) patterns were obtained by using a Siemens D-500
147
difractometer (Cu Kα,λ = 1.5432
148
estimated from the line broadening of the corresponding X-ray diffraction peaks by
149
using the Scherrer equation.
150
Scanning electron microscopic (SEM) analyses were performed on a JSM-5400 Jeol
151
apparatus equipped with an X-ray dispersive energy (EDX) analyzer.
ip t
). Crystallite sizes in the different phases were
3. Results and discussion
153 154 155
3.1 Characterization of the films
The photocatalysts used in the experiments and their properties are listed in Table 1.
156
These characteristics remained unchanged after fixation of the photocatalysts to the
157
different supports. No thermal transformation from anatase to rutile phase was
158
observed at 450°C. Adequate adherence was confirmed for all supported TiO2.
159
Figure 1 shows the SEM images obtained for the P25-D450 films deposited on
160
borosilicate glass, alumina and refractory brick. Figures 1a and b show the
161
photocatalyst deposited on borosilicate glass. It can be observed that the resulting
162
film is uniform, with the support completely covered.
163
Figures 1c and d show the photocatalyst deposited on alumina. In this case the
164
distribution of TiO2 on the support is irregular and non-covered areas are detected.
165
The TiO2 film thickness is, in general, variable, although regular thicknesses of
166
around 5 µm were obtained in some areas with a more homogeneous distribution
167
(Figure 1d). This thickness is similar to that obtained for the TiO2 films supported on
168
borosilicate glass (6 µm) (Figure 1a).
169
The distribution of TiO2 deposited on refractory brick can be seen in Figure 1e. EDX
170
mapping is included in order to better discern the TiO2 deposits. Results indicate that
171
TiO2 distribution is random and that other components, such as Al, Fe, Si, K and Mg
172
are present on the brick surface. These elements are those employed in the
173
manufacturing of the brick. Figure 1f shows areas with the highest TiO2 coverage.
174
The thickness of the film in this case is around 10 µm, but numerous fissures can be
175
observed.
Ac ce
pt
ed
M
an
us
cr
152
Page 6 of 35
176
Figures 1 g and h show better detail of Evonik P25 alone and Evonik P25-D450.
177
Particle aggregation can be denoted in the D450 catalyst.
179 180 181 182
3.2 FTIR studies 3.2.1 Adsorption of water onto the photocatalysts FTIR spectra of the photocatalysts alone and when supported on borosilicate glass
ip t
178
(both D105 and D450) were analyzed. The results described below correspond to the
184
Evonik P25 catalyst.
185
The band observed in all the spectra at 1640 cm−1 is attributed to the water bending
186
mode (ν2). The broad band between 3650 and 3000 cm−1 is attributed to the
187
asymmetrical (ν3) and symmetrical (ν1) vibration modes of water, which represent
188
isolated molecules interacting via hydrogen bonds [22, 23]. The shape and relative
189
intensity of the bands attributed to vibrations ν3 and ν1 with respect to the
190
corresponding vibration ν2 differ depending on the thermal treatment the
191
photocatalyst was subjected to (see Figure 2). In this respect, the water layer on the
192
surface seems to decrease for the D105 and D450 photocatalysts, especially for the
193
latter. This indicates that thermal treatment may lead to particle agglomeration and,
194
consequently, a lower exposed surface available for adsorption of different species,
195
such as water. These results agree with that observed from SEM studies (Figures 1 g
196
and h).
us
an
M
ed
pt
Ac ce
197
cr
183
198
3.2.2 Surface acid centres
199
The modifications to the catalyst surface when deposited and calcined (see Section
200
3.2.1) may indicate changes to the adsorption and photocatalytic active centres. For
201
this reason, ammonia interaction with the catalyst surface was studied in order to
202
determine the presence and/or modification of Lewis or Brönsted acid centres for
203
the photocatalyst alone and when supported on borosilicate glass (D105 and D450).
204
Figure 3 shows the spectra obtained after ammonia interaction with the catalyst
205
surface in the dark and at different illumination times. Spectra are shown for Evonik
206
P25. After ammonia adsorption, bands attributed to Lewis acid centres (≈ 1200 cm-1),
207
Brönsted acid centres (≈ 1450 cm-1) and breaking centres (1340 and 1320 cm−1) have
Page 7 of 35
been detected [24]. In our study, the relative intensities of these bands differed
209
depending on the thermal treatment the catalyst had been subjected to. Slight shifts
210
of the bands attributed to Lewis and Brönsted acid sites are also observed. The
211
previously described agglomeration of the D450 photocatalyst could be responsible
212
for the modification of the acid sites.
213
Illumination of the photocatalysts after ammonia adsorption indicates that different
214
photoproducs are produced for the D450 catalyst if compared to those generated for
215
Evonik P25 alone and the D105 system.
216
For the last systems, a progressive reduction of the band attributed to Lewis acid
217
centres (≈ 1200 cm-1) is observed, and this is accompanied by the formation of
218
adsorbed nitrate and nitrite (bands at 1564 and 1192 cm-1, respectively). This can be
219
seen in Figures 3a and b.
220
On the contrary, new bands at 1400 and 1312 cm-1 (Figure 3c) appear for the D450
221
catalyst after only 5 minutes irradiation. These bands correspond to HNO3/H2O
222
mixtures, as described in [25]. These results indicate a higher activity and the
223
presence of different photoactive species for the D450 system.
224
Addionally, literature references indicate that high temperature thermal treatment
225
of TiO2 can lead to modification of its surface structure [26]. Such modifications
226
could generate new surface species, such as O2-, O22- or O-, which may affect the
227
activity of the photocatalysts.
228
Lewis acid centres have been correlated by other authors with a higher oxygen
229
adsorption on the catalyst surface and, consequently, higher O2·- production [27].
230
This would inhibit the electron-hole recombination rate. In this study, a higher
231
proportion of Lewis acid centres are detected for the D450 photocatalyst, which in
232
addition presents a higher activity towards ammonia oxidation, as was previously
233
described. It can be therefore affirmed that the presence of more Lewis acid centres
234
is associated with a higher activity.
Ac ce
pt
ed
M
an
us
cr
ip t
208
235 3.3 Preliminary studies: photolysis and adsorption of imazalil
236 237 238
Photolysis of imazalil has been previously reported at pH 7 [5, 21]. Adsorption of
239
imazalil was evaluated at different pH, namely natural pH (3.9), 5 and 7. It was
Page 8 of 35
observed that adsorption was enhanced at pH 7 for all photocatalysts (data not
241
shown). In agreement with previous studies [2, 4-5], adsorption of imazalil at low pH
242
values was greatly hindered by electrostatic repulsion between the photocatalysts
243
(pHPZC between 5.2-7.8, see Table 1) and the imazalil (pKa = 6.54) and can be
244
considered negligible.
245
It should be noted that adsorption was higher for the supported photocatalysts at
246
pH 7 after the thermal treatments. In this respect and as can be seen in Figure 4 for
247
studies at pH 7, the amount of imazalil adsorbed after the adsorption equilibrium
248
had been established was up to 3 times higher (expressed as mgIMZ·gcat-1) for the
249
photocatalysts which had been supported and treated at 450°C (D450) when
250
compared to the photocatalyst alone (in suspension). This is attributed to surface
251
modification as a result of the thermal treatment (as discussed in Section 3.2-FTIR
252
studies): the increase in surface agglomeration and the surface active sites
253
modification seemed to enhance imazalil adsorption at pH 7. This was most clearly
254
observed in the following order: Evonik P25 > Evonik P90 > EST-1023t.
255 256
3.4 Screening of photocatalysts Figure 5 shows the imazalil degradation profiles and the percentage of
257
mineralization after 120 min of irradiation, for a 25 mg·L-1 imazalil solution at pH 7,
258
using the different photocatalysts considered in this study in suspension (S) or
259
supported on borosilicate glass after thermal treatment at 105°C (D105) or 450°C
260
(D450).
261
Imazalil degradation was significantly slower for the supported systems than for the
262
suspended one. This is due to the lower photocatalyst surface area available in the
263
supported configurations. Despite this, IMZ conversion was above 96% in all cases
264
after 120 minutes irradiation. It should be noted that, with respect to the supported
265
systems, degradation rates were higher for the photocatalysts subjected to thermal
266
treatment at 450°C.
267
Table 1 shows the deposited mass of TiO2 for each of the supported systems. It can
268
be seen that the mass of TiO2 varies from one system to another. For this reason, the
269
apparent rate constant referred to catalyst mass was determined; taking into
270
account that imazalil degradation follows pseudofirst order kinetics. Results, shown
Ac ce
pt
ed
M
an
us
cr
ip t
240
Page 9 of 35
in Table 2, must be compared to those shown in Figure 5, where the photocatalyst
272
deposited mass is not considered. In this sense, although it is clear from Figure 5 that
273
imazalil degradation is slower for the deposited systems, from the results in Table 2
274
can be seen that, if the catalyst mass is considered, the apparent degradation rate
275
constant is very similar or even higher for the D450 configuration when compared to
276
the suspended (S) one. This agrees with FTIR studies, as was described in Section
277
3.2.2. However, it must be emphasized that, whatever the deposited mass, the
278
photocatalytic activity will strongly depends on the amount of photocatalyst with an
279
effective irradiated surface.
280
Mineralization was lower for the supported system (D105) compared to the
281
suspension/slurry configuration, maybe due to the lower available TiO2 exposed
282
surface area. The decrease in mineralization yield was highest for the lab-made
283
catalyst, maybe due to its lower surface area.
284
For all catalysts it was observed that thermal treatment at 450°C (D450 systems)
285
resulted in higher photoactivity compared to the D105 systems. This effect was
286
particularly noticeable in the mineralization results which were very similar for the
287
D450 and suspended systems.
288
It has been reported that TiO2 treated at 450°C contains more surface oxygen than
289
that treated at 105°C or left untreated [28]. The presence of surface oxygen may
290
result in a higher photoactivity because the photoexcited electrons would be
291
scavenged by surface oxygen to form superoxides on TiO2, thus preventing electron-
292
hole recombination. In addition, the higher photoactivity of the D450 system may be
293
associated with surface modifications of the photocatalyst that are related to the
294
creation of oxygen adsorption centres or a different nature of oxidizing species
295
present at the surface, in accordance with the FTIR studies shown in Section 3.2.2.
296
As for the activity of the different catalysts, it was observed that while imazalil
297
degradation took place at a faster rate with the EST-1023t catalyst, mineralization
298
was slightly higher for the commercial Evonik P25. It was therefore decided to
299
perform the remaining experiments using the Evonik P25 catalyst.
Ac ce
pt
ed
M
an
us
cr
ip t
271
Page 10 of 35
3.5 Efficiency of TiO2 on the different supports The use of different supports was studied in order to investigate the effect of
302
increasing the exposed catalyst surface area. The studies were conducted using
303
Evonik P25 and applying thermal treatment at 105°C or 450°C for its fixation.
304
Table 3 shows some characteristics of the different materials employed as supports.
305
The data were provided by the support manufacturers.
306
The various materials support different amounts of TiO2 because of their different
307
pore size and structure. The highest deposited amount was obtained for the 50 ppi
308
alumina foam, followed by the fire brick and the glass tube, as shown in Table 4. This
309
is due to the higher porosity of the foam and fire brick which favours TiO2 adhesion.
310
The photocatalytic rate is believed to increase with increasing porosity of the
311
support [15].
312
In agreement with the literature [29-31], the photocatalytic activity of TiO2 films
313
supported on macroporous alumina foam was observed to increase with increasing
314
macroporosity of the substrate surface and so the highest catalytic activity was
315
found for TiO2 deposited on the 50 ppi foam.
316
Figure 6 shows the imazalil degradation profiles and mineralization percentages after
317
120 minutes illumination for the treatment of 25 mg·L-1 imazalil using TiO2 Evonik
318
P25 immobilized on different materials: borosilicate glass, fire brick and 50 ppi
319
alumina foam.
320
Degradation tests showed that mineralization was highest for the alumina foam
321
D105 system. However, this may be attributable to a slight detachment of the
322
photocatalyst, as was seen from the adhesion studies. Although the fire brick and
323
alumina foam provided higher TiO2 exposed area and higher deposited amounts,
324
mineralization was very similar for the D450 borosilicate glass and alumina 50 ppi
325
foam systems.
326
It should be noted that the fire brick results were not as good as expected. This may
327
be attributable to a high concentration of inorganic ions on this support surface
328
which may interact with TiO2 (see Section 3.1 for SEM images). Some studies have
329
correlated the decline in activity with the presence of cationic impurities in the TiO2
330
layer as a consequence of the thermal treatments required to improve adhesion of
331
the titania layer onto the support [16, 32].
Ac ce
pt
ed
M
an
us
cr
ip t
300 301
Page 11 of 35
3.6 Photocatalytic degradation of imazalil in synthetic agro-industrial water
334
Imazalil concentrations of up to 50 mg·L-1 have been reported in agro-industrial
335
wastewater effluents. For comparison purposes with a previous work [9], the
336
synthetic wastewater employed in this study contained 50 mg·L−1 imazalil, 100
337
mg·L−1 chloride (as NaCl), 300 mg·L−1 sulphate (as Al2(SO4)3·18H2O) and 20 mg·L−1
338
calcium (as Ca(OH)2). This simulates the composition generally found in the
339
wastewater samples analyzed from a collaborating banana packing company.
340
Photocatalysis is affected by the composition of the water. A high concentration of
341
inorganic ions can interfere, physically and chemically, with the photocatalysts [33].
342
Figure 7 shows the degradation profiles of imazalil, as well as the percentage of
343
mineralization after 120 minutes of irradiation for 50 mg·L−1 of imazalil in deionized
344
water and synthetic wastewater using different systems considered in this study.
345
In general terms, it was observed that the ionic strength of the SW matrix did not
346
hinder imazalil degradation or mineralization under the studied conditions. This is
347
because the working pH of 7 was established to be higher than the photocatalyst
348
pHPZC (see Table 1) and therefore anion adsorption was not favoured [9, 34-35].
349
From the results it can be concluded that both borosilicate glass and alumina foam
350
could be adequate supports for TiO2 in the treatment of waters contaminated with
351
imazalil.
cr
us
an
M
ed
pt
3.7 Reusability and recovery of TiO2 films
Ac ce
352 353
ip t
332 333
354
The results of catalyst activity over a prolonged period of use in synthetic
355
wastewater are shown in Figure 8. The treatment of 50 mg·L-1 imazalil in SW for 240
356
minutes under irradiation was repeated six times to evaluate the durability and
357
photoactivity of the immobilized photocatalyst under these operating conditions.
358
For the suspended system, no loss in activity was observed after six cycles. However,
359
a fall in photocatalytic activity was observed with increasing aging time of the TiO2
360
film for the supported systems. This phenomenon was accompanied by the
361
appearance of a pale yellow colour on the TiO2 film indicating the presence of
362
adsorbed intermediates on the titania active sites. These observations agree with
Page 12 of 35
363
those reported by other authors [36-38]. Deactivation of the photocatalyst after use
364
has also been attributed to the formation of oxygen vacancies after photocatalytic
365
reaction [36]. Oxygen vacancies are formed according to reaction (1), as reported in
366
[37].
367
h+ + 1/2O2-(lattice) → 1/4O2 + vacancy
368
The fall in mineralisation as the photocatalyst was reused was higher using the
369
alumina support than the borosilicate glass and, significantly, this was accompanied
370
by a much more noticeable yellowing. Accordingly, due to the higher mineralization
371
achieved during all cycles for the borosilicate D450 system, this was chosen as the
372
best support for the TiO2 photocatalyst.
373
In order to maintain high photocatalytic treatment effectiveness, the catalyst should
374
be cleaned or replaced. Several regeneration methods have been described for TiO2
375
films, including, amongst others, deionized water washes, NaOH washes, the
376
combination of H2O2 or air with UV irradiation or recalcination of the film [39-42].
377
Of these, recalcination of the film is not recommended because heat treatment may
378
cause further agglomeration of the TiO2 nanoparticles.
379
The regeneration effect on photocatalytic activity of three of these methods (water
380
washes, air with UV irradiation and recalcination) was tested in this study.
381
As can be seen in Figure 9, this technique did not result in favourable regeneration of
382
the TiO2 film. The air plus UV irradiation was also ineffective for film regeneration
383
film.
384
However, water washing was shown to be an effective technique to recover
385
deposited TiO2 activity. It has been reported to effectively remove the oxygen
386
vacancies that may have been responsible for TiO2 deactivation [36]. We have
387
observed that water washing of the TiO2 films at ambient temperature fully recovers
388
TiO2 activity if the washing procedure is performed after every use. Furthermore, the
389
films treated this way may be used daily for months without deactivation.
Ac ce
pt
ed
M
an
us
cr
ip t
(1)
Page 13 of 35
390
3.8 Solar experiments The optimal system determined at laboratory scale for the treatment of synthetic
393
agro-industrial wastewater containing imazalil was applied under solar irradiation in
394
order to validate the results.
395
Figure 10 shows the degradation and mineralization profiles of imazalil against
396
accumulated energy. It should be noted that the necessary time for the effluent
397
containing imazalil to be treated strongly depends on the climate conditions.
398
As can be observed from Figure 10, 75% mineralization was reached after around
399
110 kJ·L-1 of accumulated energy, which was equivalent to 8 hours at our location
400
considering perfect sunny days. IMZ was completely eliminated after 82 kJ·L-1
401
accumulated energy.
an
us
cr
ip t
391 392
4. Conclusions
TiO2 films, deposited on several supports by means of dip-coating of commercial and
405
lab-made photocatalysts, showed good adhesion to the substrate surface after
406
thermal treatment at 450°C. This treatment also resulted in enhanced photocatalytic
407
activity for all the photocatalysts tested. Of the photocatalysts studied, Evonik P25
408
and EST-1023t gave the best degradation results, and Evonik P25 and P90 returned
409
the best mineralization.
410
Coatings were studied next on different supports, namely borosilicate glass,
411
refractory brick and alumina foam, to increase the exposed surface area of the
412
photocatalyst and the amount of deposited mass. In view of its high mineralisation
413
results, the Evonik P25 photocatalyst was employed for this purpose. TiO2 supported
414
on alumina foam exhibited a higher imazalil removal activity, although mineralization
415
was similar for TiO2 films on alumina and borosilicate glass. However, the activity of
416
TiO2 on alumina was considerably diminished when the support was reused several
417
times for the treatment of synthetic agro-industrial wastewater (containing inorganic
418
ions) as water matrix.
419
Solar experiments were carried out and confirmed that TiO2 can be deposited on
420
borosilicate glass to efficiently remove up to 50 mg·L-1 imazalil from synthetic agro-
421
industrial wastewaters.
Ac ce
pt
ed
M
402 403 404
Page 14 of 35
422 423 424 425
Acknowledgements We thank the University of Las Palmas de Gran Canaria for its funding through the PhD Grant Program, the Spanish Ministry of Science and Innovation for its financial
427
support through the PhD Studentship BES-2010-036537, MINECO (Ministry of
428
Economy and Competitiveness, Government of Spain) for funding of the NANOBAC
429
project (IPT-2011-1113-310000) and the ERDF for co-funding with MINECO the
430
Infrastructure Project 2010-3E UNLP10-3E-726.
cr
ip t
426
us
5. References
[1] M. Duque, J.M. Torres, J.J. Oramas, J. Fernández, Pudrición de Corona en el
432
Plátano Canario, first ed. Coplaca, Tenerife, 2004.
433
[2] M.A. Martín-González, O. González-Díaz, P. Susial, J. Araña, J.A. Herrera-Melián,
434
J.M. Doña-Rodríguez, J. Pérez-Peña, Chem. Eng. J. 245 (2014) 348–358.
435
[3] A.K. Genena, D.B. Luiz, W. Gebhardt, R.F.P.M. Moreira, H.J. José, H.F. Schröder,
436
Ozone Sci. Eng. 33 (2011) 308–328.
437
[4] R. Hazime, C. Ferronato, L. Fine, A. Salvador, F. Jaber, J.M. Chovelon, Appl. Catal.,
438
B. 126 (2012) 90–99.
439
[5] D.E. Santiago, J.M. Doña-Rodríguez, J. Araña, C. Fernández-Rodríguez, O.
440
González-Díaz, J. Pérez-Peña, A.M.T. Silva, Appl. Catal., B. 138–139 (2013) 391–400.
441
[6] S. Malato, J. Blanco, D.C. Alarcón, M.I. Maldonado, P. Fernández-Ibáñez, W.
442
Gernjak, Catal. Today 122 (2007) 137–149.
443
[7] D.E. Santiago, E. Pulido Melián, C. Fernández-Rodríguez, J.A. Ortega Méndez, S.O.
444
Pérez-Báez, J.M. Doña-Rodríguez, Green Sustain. Chem. 1 (3) (2011) 39–46.
445
[8] I. Carra, S. Malato, M. Jiménez, M.I. Maldonado, J.A. Sánchez Pérez, Chem. Eng. J.
446
235 (2014) 132–140.
447
[9] D.E. Santiago, J. Araña, O. González-Díaz, M.E. Alemán-Domínguez, A.C. Acosta-
448
Dacal, C. Fernández-Rodríguez, J. Pérez-Peña, José M. Doña-Rodríguez, Appl. Catal.,
449
B 156-157 (2014) 284-292.
450
[10] R. L. Pozzo, Miguel A. Baltanfis, Alberto E. Cassano, Catal. Today 39 (1997) 219-
451
231.
Ac ce
pt
ed
M
an
431
Page 15 of 35
[11] M.R. Espino-Estévez, J.M. Doña-Rodríguez, C. Fernández-Rodríguez, O.
453
González-Díaz, J. Pérez-Peña, in: 7th European Meeting on Solar Chemistry and
454
Photocatalysis: Environmental Applications (SPEA 7) Oporto, Portugal, 2012, pp. 89.
455
[12] D. Li, H. Zheng, Q. Wang, X. Wang, W. Jiang, Z. Zhang, Y. Yang, Sep. Pur. Tech.
456
123 (2014) 130–138.
457
[13] N. Miranda-García, S. Suárez, B. Sánchez, J.M. Coronado, S. Malato, M. Ignacio
458
Maldonado, Appl. Catal., B 103 (2011) 294–301.
459
[14] J.A. Byrne, B.R. Eggins, N.M.D. Brown, B. McKinney, M. Rouse, Appl. Catal., B 17
460
(1998) 25-36.
461
[15] A. Rachel, B. Lavedrine, M. Subrahmanyam, P. Boule, Catal. Commun. 3 (2002)
462
165–171.
463
[16] A. Rachel, M. Subrahmanyam, P. Boule, Appl. Catal., B 37 (2002) 301–308.
464
[17] M. Radeka, S. Markov, E. Loncar, O. Rudic, S. Vucetic, J. Ranogajec, J. Eur.
465
Ceram. Soc. 34 (2014) 127–136.
466
[18] A. Gutiérrez-Alejandre, M. Trombetta, G. Busca, J. Ramirez, Microporous Mater.
467
12 (1997) 79-91.
468
[19] C. Guillard, B. Beaugiraud, C. Dutriez, J. Herrmann, Appl. Catal., B 39 (2002) 331–
469
342.
470
[20] C. Legrand-Buscema, C. Malibert, S. Bach, Thin Solid Films 418 (2002) 79–84.
471
[21] J. Araña, J.M. Doña-Rodrı́guez, O. González-Dı́az, E. Tello Rendón, J.A. Herrera
472
Melián, G. Colón, J.A. Navı́o, J. Pérez Peña, J. Mol. Catal. A: Chem. 215 (2004) 153-
473
160.
474
[22] H.A. Al-Abadleh, V.H. Grassian, Langmuir 19 (2) (2003) 341–347.
475
[23] L. Cammarata, S.G. Kazarian, P.A. Salterb, T. Welton. Phys. Chem. Chem. Phys. 3
476
(2001) 5192–5200.
477
[24] G. Martra. Appl. Catal., A, 200 (2000) 275-285.
478
[25] H. Yang, B.J. Finlayson-Pitts, J. Phys. Chem., A 105 (2001) 1890–1896.
479
[26] M.J. Elser, O. Diwald, J. Phys. Chem. C 116 (4) (2012) 2896–2903.
480
[27]M.I. Franch, J. Peral, X. Domènech, R.F. Howe, J.A. Ayllón, Appl. Catal., B 55
481
(2005) 105-113.
482
[28] J.C. Yu, J. Lin, D. Lo, S.K. Lam, Langmuir 16 (2000) 7304-7308.
Ac ce
pt
ed
M
an
us
cr
ip t
452
Page 16 of 35
[29] G. Plantard, V. Goetz, F. Correia, J.P. Cambon, Sol. Energy Mater. Sol. Cells 95
484
(2011) 2437-2442.
485
[30] G. Plesch, M. Gorbár, U.F. Vogt, K. Jesenák, M. Vargová, Mater. Lett. 63 (2009)
486
461-463.
487
[31] M. Vargová, G. Plesch, U.F. Vogt, M. Zahoran, M. Gorbár, K. Jesenák, Appl. Surf.
488
Sci. 257 (2011) 4678–4684.
489
[32] A. Fernández, G. Lassaletta, V.M. Jiménez, A. Justo, A.R. González-Elipe, J.M.
490
Herrmann, H. Tahiri, Y. Ait-Ichou, Appl. Catal., B 7 (1995) 49–63.
491
[33] A.G. Rincon, C. Pulgarin, Appl. Catal., B 51 (2004)283–302.
492
[34] W. Zhang, T. An, M. Cui, G. Sheng, J. Fu, J. Chem. Technol. Biotechnol. 80 (2005)
493
223-229.
494
[35] A. Aguedach, S. Brosillon, J. Morvan, E.l Kbir Lhadi, J. Hazard. Mater. 150 (2007)
495
250-256.
496
[36] X. Pan, N. Zhang, X. Fu, Y.J. Xu, Appl. Catal., A 453 (2013) 181-187.
497
[37] S. H. Szczepankiewicz, A. J. Colussi, M. R. Hoffmann, J. Phys. Chem., B 104 (2000)
498
9842-9850.
499
[38] D. Rubio, J.F. Casanueva, E. Nebot, J. Photochem. Photobiol., A 271 (2013) 16–
500
23.
501
[39] N. Miranda-García, S. Suárez, M. Ignacio Maldonado, S. Malato, B. Sánchez,
502
Catal. Today 230 (2014) 27-34.
503
[40] L. Nie, P. Zhou, J. Yu, M. Jaroniec, J. Mol. Catal. A: Chem. 390 (2014) 7-13.
504
[41] S. Liu, M. Lim, R. Amal, Chem. Eng. Sci. 105 (2014) 46–52.
505
[42] N. Madaan, S. Gatla, V.N. Kalevaru, J. Radnik, B. Lücke, A. Brückner, A. Martin, J.
506
Catal. 282 (2011) 103–111.
cr
us
an
M
ed
pt
Ac ce
507
ip t
483
Page 17 of 35
Figure 1. SEM images for TiO2 (Evonik P25) films deposited on: borosilicate glass (a and b), 50
508
ppi alumina foam (c and d), refractory brick (e and f) and detail of Evonik P25 untreated and
509
Evonik P25-D450.
510
Figure 2. FTIR spectra of Evonik P25 alone (P25), and after being supported on
511
borosilicate glass using thermal treatment at 105°C (P25-D105) and 450°C (P25-
512
D450).
513
Figure 3. FTIR spectra from the interaction of NH3 with Evonik P25 alone (a) and
514
after being supported on borosilicate glass using thermal treatment at 105°C (P25-
515
D105) b) and 450°C (P25-D450) c) under illumination at different times.
516
Figure 4. Adsorption of IMZ onto different TiO2 photocatalysts (initial concentration
517
25 mg·L-1 and initial pH 7). For the supported systems, the support was borosilicate
518
glass.
519
Figure 5. Degradation profiles (a) and % mineralization after 120 minutes (b) of 25
520
mg·L-1 IMZ using different TiO2 photocatalysts and configurations: suspended system
521
(S) and supported (D105 and D450) on borosilicate glass.
522
Figure 6. Degradation profiles (a) and % mineralization after 120 minutes (b) of 25
523
mg·L-1 IMZ using different TiO2 supports: borosilicate glass, fire brick and alumina
524
foam.
525
Figure 7. Degradation profiles (a) and % mineralization after 120 minutes (b) of 50
526
mg·L-1 IMZ using different TiO2 systems (suspended -S- or supported on alumina or
527
borosilicate glass - D450) and different water matrices: deionized water (DW) or
528
synthetic wastewater (SW).
529
Figure 8. IMZ conversion (%) (a) and % mineralization after 240 minutes (b) of 50
530
mg·L-1 IMZ using suspended (S), borosilicate D450 (B-D450) and alumina D450 (A-
531
D450) systems and synthetic wastewater (SW).
532
Figure 9. IMZ conversion (%) (a) and % mineralization after 240 minutes (b) of 50
533
mg·L-1 imazalil in SW using Evonik P25 supported on borosilicate glass (D450) on
534
the 1st cycle, 5th cycle and subsequent regeneration cycle.
535 536 537
Figure 10. IMZ and TOC evolution in the solar experiment using Evonik P25 supported on borosilicate glass (D450) for the treatment of 50 mg·L-1 IMZ in SW.
Ac ce
pt
ed
M
an
us
cr
ip t
507
Page 18 of 35
537 538
Table 1. Characteristics of the photocatalysts used in this work and deposited TiO2 mass
539
Table 2. Characteristics of the different supports.
540
Table 3. Deposited TiO2 mass on the different supports and adhesion studies.
for the different photocatalysts supported on borosilicate glass.
541
Ac ce
pt
ed
M
an
us
cr
ip t
542 543
Page 19 of 35
543 544
Table 2. Characteristics of the photocatalysts used in this work and deposited TiO2 mass for the different photocatalysts supported on borosilicate glass. Catalyst
Anatase/Rutile ratio (%) 80/20 86/14 70-80/30-20
Band Gap (eV) 3.18 3.29 2.96
Crystallite size (nm) Anatase Rutile 22.0 25.0 13 62.3 96.1
pHPZC 6.5 7.8 5.2
Deposited amount (mg) D105 D450 50 55 80.1 81.5 109.9 80.1
ip t
Evonik P25 Evonik P90 EST-1023t[5]
Specific surface area (m2·g-1) 52 100 13.5
Ac ce
pt
ed
M
an
us
cr
545 546
Page 20 of 35
546 547 548
Table 2. Apparent degradation rate constants for imazalil, (kIMZ/gcat) using different photocatalysts and systems. For the deposited systems, the support was borosilicate glass. kIMZ·g-1cat (min-1·g-1) S D105 D450 0.31 0.42 0.47 0.27 0.22 0.27 0.39 0.21 0.35
Photocatalyst
ip t
Evonik P25 Evonik P90 EST-1023t
Ac ce
pt
ed
M
an
us
cr
549 550
Page 21 of 35
Table 3. Characteristics of the different supports. Support
% Porosity
Apparent Density (g·cm-3)
Glass tube Fire brick Foam 50 ppi Foam 20 ppi Foam 10 ppi
16-20 ≥80 ≥80 ≥80
2.05 <0.5 <0.5 <0.5
Ac ce
pt
ed
M
an
us
cr
551 552
ip t
550
Page 22 of 35
552
Table 4. Deposited TiO2 mass on the different supports and adhesion studies. Deposited amount (mg)
Material Borosilicate glass Red brick Alumina foam (50 ppi)
D105 50 80 310.1
D450 55 80.1 304.6
Turbidity from adhesion tests (NTU) D105 D450 0.67 0.76 1.49 0.56 2.31 1.02
ip t
553
Ac ce
pt
ed
M
an
us
cr
554 555
Page 23 of 35
555 556
pt
ed
M
an
us
cr
ip t
The elimination of imazalil was proved for different water matrices. Suspended and supported TiO2 photocatalysis systems were compared. The effect of the water matrix was evaluated for the different systems. Different supports and thermal treatments were evaluated for the fixation of TiO2. Solar experiences confirmed the adequate treatment of synthetic wastewater.
Ac ce
557 558 559 560 561 562 563
Page 24 of 35
ed
pt
Ac ce
563
Page 25 of 35
us
an
M
cr
ip t
Ac ce p
te
d
M
an
us
cr
ip t
Figure 1
Page 26 of 35
Ac
ce
pt
ed
M
an
us
cr
i
Figure 2
Page 27 of 35
Ac
ce
pt
ed
M
an
us
cr
i
Figure 3
Page 28 of 35
Ac
ce
pt
ed
M
an
us
cr
i
Figure 4
Page 29 of 35
Ac
ce
pt
ed
M
an
us
cr
i
Figure 5
Page 30 of 35
Ac
ce
pt
ed
M
an
us
cr
i
Figure 6
Page 31 of 35
Ac
ce
pt
ed
M
an
us
cr
i
Figure 7
Page 32 of 35
Ac
ce
pt
ed
M
an
us
cr
i
Figure 8
Page 33 of 35
Ac
ce
pt
ed
M
an
us
cr
i
Figure 9
Page 34 of 35
Ac
ce
pt
ed
M
an
us
cr
i
Figure 10
Page 35 of 35