Flexible Bench-Scale Recirculating Flow CPC Photoreactor for Solar Photocatalytic Degradation of Methylene Blue Using Removable TiO 2 Immobilized on PET Sheets

1Physical Chemistry Department, National Research Centre (NRC), 33 Elbohouth Street, P.O. Box 12622, Dokki, Giza, Egypt 2Polymeric Materials Research Department, Advanced Technology and New Materials Research Institute (ATNMRI), City of Scientific Research and Technological Applications (SRTA-City), New Borg El-Arab City, Alexandria 21934, Egypt 3Polymers and Pigments Department, National Research Centre (NRC), 33 Elbohouth Street, P.O. Box 12622, Dokki, Giza, Egypt 4Inorganic Chemistry Department, National Research Centre, 33 Elbohouth Street, P.O. Box 12622, Dokki, Giza, Egypt 5Nano-Photochemistry and Solar Chemistry Department, Laboratory of Chemistry, Faculty of Science, Ain Shams University, Abbassia, 11566 Cairo, Egypt


Introduction
Semiconductor photocatalysis is a fast growing area in terms of both research and commercial activities [1].In recent years, supported TiO 2 materials have been widely studied for the application on both indoor and outdoor air and water purification [2][3][4][5] disinfection and antibacteria [6,7], as well as self-cleaning surface [8,9].
Generally, photocatalytic oxidation follows the absorption of energy photons by semiconductors such as TiO 2 .Each photon should have energy (ℎ]) more than or equal to the bandgap energy of TiO 2 .The absorbed photon then promotes the excitation of electron (e − ) from valence band to conduction band.The generated electrons on conduction band and holes on valence band (h + ) would then directly or indirectly interact with adsorbed organic pollutants causing their destruction.UVA light energy (wavelength between 300 and 400 nm) is adequate to perform the photocatalytic oxidations on TiO 2 surfaces.Solar energy is a free sustainable source UVA light energy.Egypt has UVA irradiances that reach 40 W/m 2 or even higher on sunny days [10].
Compound Parabolic Concentrator (CPC) has been extensively interpreted as a good option for solar photochemical applications [11][12][13][14][15][16].CPC reactor consists of cylindrical 2 International Journal of Photoenergy absorber, on a combined reflecting profile.The reflector geometry mixes two parabolas and one involute whose curves relate to the diameter of the circle and to the concentration ratio.
Titanium dioxide can be used as slurry or fixed inside the CPC photocatalytic reactors.The major obstacle for the practical use of TiO 2 slurries in water treatment was the requirement for the expensive liquid-solid separation.This would consume more efforts, money, and time.In addition, the recovery loss of catalyst might be high [17].Therefore, catalyst immobilization is the key for successful implementation of heterogeneous photocatalytic oxidation of organic pollutants in water [18,19].
Some commonly used substrates such as glass, aluminum, and stainless steel are suitable for high temperature annealing treatment after the photocatalyst loading.However, the stiffness and weight of the materials impose limitations for largescale implementation.
Different methods have been developed on loading of TiO 2 on polymeric sheets from photocatalyst suspensions [20].All these approaches have to fulfill at least three main needs: the long term stability of the photocatalyst immobilization, its availability for heterogeneous photocatalysis, and the low cost of the deposition procedure, in view of a practical application.
Loading methods, such as sputtering [21,22], electrochemical deposition [23], spin coating [24], and electrophoretic coating [25], have been previously utilized for the preparation of titania supported thin films.However, these methods are often complicated, costly, and difficult for practical application [26,27].Large-scale application of photocatalytic oxidation for water purification can be achieved through direct coating of commercial photocatalysts such as Degussa P25 TiO 2 [28].When using direct coating, the primary issue of interest is the stability of the coating.Poor adhesions have been reported by direct coating method such as dip coating or spin coating [29,30], where even slight mechanical abrasion may remove the photocatalyst from the substrate surface.Thus, a binding agent is usually necessary for direct coating in order to form solid adhesion between catalyst and substrate.Either organic polymer or inorganic binding materials have been used for photocatalyst immobilization [31][32][33].However, the binder amount may affect the photocatalytic oxidation activity and stability of the coating [34].
Evaluation of the photocatalytic activity of surfaces modified photocatalysts is a crucial point.The dye method is one of the most commonly used methods for the evaluation of the photocatalytically active surfaces.Methylene blue (MB) has been widely addressed as a test pollutant in the evaluation of the photocatalytic activity of semiconductors.MB is extensively used as textile and leather and paper dye.It has been previously demonstrated that films of titania were able to mediate the complete photomineralization of MB [35].With MB as a highly colored organic material, all what is required for the evaluation process is following up its rate of photocatalytic bleaching via UV/vis spectrophotometery [36,37].Recently, the international standards organization (ISO) released a technical standard (ISO 10678:2010) for "determination of photocatalytic activity of surfaces in an aqueous medium by degradation of methylene blue" [1].
In this work, a bench-scale solar CPC photoreactor has been designed and fabricated for testing the photocatalytic oxidation of organic pollutants.TiO 2 Evonik Aeroxide P25 (previously known as Degussa P25) supported on PET sheets was used as a photocatalytically active surface inside the reactor.Immobilization of TiO 2 on PET was conducted by a simple and cost effective method.MB aqueous solutions were used to study the photocatalytic activity of TiO 2 /PET sheets under sunlight.COD analyses were conducted to test the complete mineralization of MB.Durability of TiO 2 /PET composite was also tested under sunlight irradiation.The physicochemical characterizations of the prepared TiO 2 /PET composite before and after degradation processes were also investigated. .In this study, strips of dimensions (0.2 × 0.7 m 2 ) were cut, weighed, and then immersed in double distilled water for 20 min.Concentrated TiO 2 Aeroxide P25 solution in double distilled water was prepared.The preweighed samples were immersed separately in this solution for 1 h.The wet TiO 2loaded PET substrates were pressed at 60 ∘ C for 20 min.The percentage of loaded TiO 2 was determined and the samples were washed with distilled water several times, and after each time the TiO 2 content was calculated as follows:

Materials and Methods
where  1 and  2 are the weights of PET substrate before and after TiO 2 loading.

Scanning Electron Microscope (SEM).
In SEM measurements, an electron beam was passed through the specimens followed by scattering them back as electrons and secondary electrons.Backscattered secondary electrons were used to form the image on the computer monitor.The acceleration of the electron beam was 10 kV.This was carried out on Quanta FEG250 Instrument.

Energy Dispersive X-Ray (EDX).
The spectra of TiO 2 coated PET fabrics before and after UV irradiation were obtained by EDX measurements.A field emission scanning electron microscope (QUANTA FEG 250) coupled with an energy dispersive X-ray spectrometer (EDX) unit was employed to evaluate the elemental composition of TiO 2 /PET sheets.Semiquantitative analyses in the inspection field were conducted using ZFA software where the energy of the emitted electrons for each element was counted in units of weight percent.

Pipeline Calculations.
Compound Parabolic Concentrator (CPC) prototype is made up of seven tubes connected in series, and water would flow from the lowest (the nearest to the ground) to the highest tube and then to a tank.It has been mounted on a platform tilted (30 ∘ N local latitude in Cairo).A centrifuge pump was used to circulate wastewater.Rods of temperature-resistant Teflon were used as support for titanium dioxide immobilized in polymer films.These rods were fitted in the middle of glass tubes used to flow the wastewater.As shown in Figure 1(b), the cross-sectional area of annulus  was calculated using ( 2 −  2 ), while the wetted perimeter was 2( + ).So, the equivalent diameter is The values of 2 (the inner diameter of glass tube) and 2 (the diameter of support Teflon rod) are specified depending on the required productivity and the limits of light penetration.The Reynolds number can be calculated using equivalent diameter   from   =   / in which  is the density and  is the dynamic viscosity of used wastewater [39,40].The flow is specified in laminar region, and the friction factor  (for plastic surface) can be calculated from = /  , where  is a constant, which depends on the ratio / [41].The glass surface is smooth.The head loss is obtained by using ℎ  = 2(/  )( 2 /) and Δ = 2(/  )( 2 ).Power to fluid is calculated from ṁℎ   in which mass flow rate ṁ can be obtained from ṁ = .Minor losses of connectors (T and U elbows) between tubes and inlet and outlet were included.The expansion and contraction effects were kept at minimum.The efficiency of the used pump was assumed to be 0.85.
Each tube with 0.25 m length (0.18 m exposed to the sunlight) was made from low iron borosilicate glass [((3 × 10 −6 )/) coefficient of thermal expansion and 1.51-1.54refractive index].Seven tubes were connected in series with total irradiated surface 1.26 m 2 and volume equal to 0.723×10 −3 m 3 .The inner diameter of each borosilicate tube is 0.025 m (with 0.015 m glass thickness) and the diameter of the Teflon rod is 0.01 m.The Reynolds number is specified as 700 with average velocity 0.031 m/s.Total volume of the system is 2.5×10 −3 m 3 .The used volumetric flow rate using a centrifugal pump was 0.072 m 3 /s and the irradiation time per cycle was about 4900 s.

Reflector Design.
The reflectors of CPC were designed using the general equation of acceptance angles [38]: Part AB in Figure 1(a) (2a) is the distance of ,   is the half angle of acceptance, and the CPC concentration ratio is given by: The reflectors are made of stainless steel 304 with 8 × 10 −4 m thickness.Concentration ratio that equals one and 90 degrees' acceptance half angle (  ) were used.Dimension of the collector is 0.17 m width and 0.2 m length.

The Fabricated Bench-Scale Recirculating Flow CPC.
The photoreactor has been designed in small bench-scale dimensions (platform = 0.7 × 0.2 × 0.4 m 3 ), so it facilitates conducting lab-scale experiments (see Figure 2).Almost all the parts are flexible.The Teflon rods are removable so several kinds of supported photocatalysts can be used.Moreover, experiments containing slurry photocatalysts can also be conducted.T connectors were used to obtain a flexible labscale system (use only one tube or more and withdrawing sample at any tube is possible).
International Journal of Photoenergy 2.5.Measurement of Solar UV Irradiance.The UV light intensity was measured by YK-35UV radiometer, Taiwan.The radiometer had the same inclination as that of the platform where experiments were conducted.

Photocatalytic Degradation Experiments.
All the experiments were carried out under direct sun irradiation at the roof of the National Research Centre, Cairo, Egypt (latitude 30 ∘ N, longitude 31 ∘ E) [10].The initial concentration of MB solution was 1.5 × 10 −5 M. The total volume (  ) of the treated MB solution was 2.5 L. This volume is divided into two parts: the total irradiated volume (in glass tubes) (  ) is 0.723 L and the dead reactor volume (i.e., connectors, tank, pipes, etc.).Prior to the degradation experiments, the photoreactor was covered with black plastic to permit the dark adsorption equilibrium of MB onto TiO 2 /PET.After 30 min the plastic cover is removed and the reactor was exposed to the sunlight.The total exposure time for each experiment was approximately 7 hours, with exposure started at 9:00 and finishing at 16:00.During experiments, the MB solution was pumped at flow rate of 1.2 L/min through the glass tubes and finally into the recirculating tank.4 mL of MB solution was collected for analysis from the reactor tank at timed intervals (1 h) during the degradation experiment.Durability of the TiO 2 /PET sheets was also tested for seven days.The photocatalytic degradation experiment was repeated as mentioned previously using the same TiO 2 /PET sheets.Fresh solutions of MB were used at each repetition.The collected data at each day were interpreted using the following equation [42]: where   is the experimental time for each sample, UV is the average solar ultraviolet radiation measured during Δ  , and  30 w is a normalized illumination time.In this case, time refers to a constant solar UV power of 30 W/m 2 (typical solar UV power on a perfectly sunny day around noon).
The initial amounts of adsorbed MB molecules on TiO 2 /PET sheets at each repetition were calculated as follows: where  0B and  0A are the concentrations of MB in solution before and after the dark adsorption, respectively.The values of  0B and  0A of MB solutions are spectrophotometrically measured at  max = 665 nm. is the area of PET strip and  is the number of strips (seven strips).

UV/Visible
Measurements.The change in MB color during the degradation experiments was followed up using UV-visible spectrophotometer, Lambda 40, PerkinElmer, Germany.

COD Measurements.
Hanna COD Meter and Multiparameter Photometer Model HI 83099 was used for the Chemical Oxygen Demand (COD) measurements.The percentages of the COD removal of MB were calculated by applying the following equation: where  0 is the initial COD value of MB and   is the COD value at each time interval of sunlight irradiation.
2.9.Temperature Measurements.The temperature of the treated water was measured during the degradation experiments using Scichem Tech infrared thermometer, gun type professional, SCT658, USA.The maximum temperature achieved inside the reactor during the experiments is 50 ∘ C.

Preparation and Characterization of TiO 2 /PET Sheets.
Coating of PET with TiO 2 was conducted by dip coating method followed by exposing to mild heat and pressure.Wet PET polymeric sheets were first immersed in concentrated TiO 2 solution.The TiO 2 /PET sheets were then heated at 60 ∘ C under mild pressure.The mild temperature and pressure result in the dehydration of water molecules formed by the -OH groups at both PET and TiO 2 surfaces leading to successful binding between the adsorbed TiO 2 and PET substrate.
After coating PET sheet with TiO 2 for the first time, the TiO 2 content was found to be 30.5%.Washing with water reduced the TiO 2 content to 26.4% and 25% for one and two times, respectively.From the third to fifth washing, the TiO 2 content decreased slightly from 24.8% to 24.1%.Further washings result in no further leaching.
The FT-IR spectra of TiO 2 and pristine PET and TiO 2 and coated PET are given in Figures 3(a)-3(c).As shown in Figure 3(a), the primary O-H stretching of the hydroxyl functional group appeared at 3427 cm −1 .Besides, the band around 1630 cm −1 is attributed to the bending vibration H-OH groups.For pristine PET (Figure 2(b)), the main characteristic peaks are as follows: (a) strong peak at 1731 is due to carbonyl group, (b) peak at 3630 cm −1 is assigned to end hydroxyl group, and (c) the peak at 3431 cm −1 is attributed to overtone vibration of carbonyl group.After coating with TiO 2 (Figure 3(c)), the peak assigned to hydroxyl group was shifted to 3639 cm −1 due to bonding with PET.
In addition, SEM measurements also reveal the successful preparation of TiO 2 /PET sheets.SEM images of TiO 2 , pristine PET, and TiO 2 coated PET are depicted in Figures 4(a)-4(c).As shown in Figure 4(b), pristine PET appeared as smooth yarns.However, after coating (Figure 4(c)), TiO 2 nanoparticles appeared to be homogeneously distributed on PET yarns.

Solar Photocatalytic Degradation of MB.
Under the sunlight irradiation, remarkable changes in the color of MB solution were noticed.Complete decolorization was achieved within 5 hours of light illumination.The change in color was followed up spectrophotometrically. Figure 5(a) shows the change in the absorption spectrum of MB solution with time of illumination.The corresponding measured solar UV doses during the whole experiment are indicated in Figure 5(b).As shown in Figure 5(a), all peaks in the absorption spectrum of MB decrease as the time of illumination increases suggesting the complete mineralization of MB molecules.It is also worthy to mention that the treated MB solutions were clear during the whole degradation experiment.No suspended TiO 2 particles were observed and consequently there is no need for the separation step.This indicates the efficient performance of the prepared TiO 2 /PET sheets.
After complete decolorization, the recirculating colorless MB solution was left for 2 hours for further sunlight exposure.COD of all collected samples were then measured.The results indicated the complete mineralization of the MB molecules (Figure 6).
The complete photomineralization of MB using films of titania has been demonstrated as given in the following equation [35]: It is clear from Figures 5 and 6 that the mineralization process occurs on a longer time than the oxidative photobleaching of MB.It has been previously also found that the measurement of the rate of photobleaching of MB is expected not to be equivalent to the rate of mineralization of the dye, which is usually a slower process [1].

Durability of the TiO 2 /PET Sheets.
Reuse of TiO 2 /PET sheets for the photocatalytic degradation of MB inside the CPC reactor was investigated.The solar photodegradation experiments using the same TiO 2 /PET sheets were repeated using fresh MB solutions.TiO 2 /PET showed a high durability for 7 repetitions at 7 daylights.Comparison between the different photocatalytic experiments during the 7 days was made using the normalized illumination time ( 30 w ) as shown previously in (4).Complete decolorization of fresh MB solutions occurred at approximately  30 w = 300 min at the first 5 days (Figure 7).The rates of MB photobleaching during International Journal of Photoenergy the first to the fifth runs nearly coincided (not shown here).However, as shown in Figure 7, at runs 6 and 7, the rate of photobleaching became slower.No more decolorization was observed for MB solutions after 7 repetitions.On the other side, the color of TiO 2 /PET sheets appeared white after first 3 repetitions.This means that, at early use of TiO 2 /PET sheets, both adsorbed and dissolved MB molecules were involved in the photooxidative reactions.However, as the number of repetitions increases, the sheets' color turns toward deep blue.The initial amounts of adsorbed MB molecules on TiO 2 /PET sheets at each repetition were calculated according to (5) and listed in Table 1.These experimental and visual observations suggest that the extra adsorption of MB dye on TiO 2 /PET sheets results in blocking the photocatalytically active sites of TiO 2 particles.
The physicochemical characteristics of TiO 2 /PET sheets were investigated before and after the solar photodegradation experiments.Besides, some agglomeration for TiO 2 particles was observed by SEM measurements on PET surfaces after 7 repetitions (Figure 4(d)).On the other hand, EDX spectroscopy reveals the good binding of TiO 2 to PET surfaces.As shown in Figures 8(a) and 8(b), the two characteristic signals of Ti atom at 4.6 and 5 eV were observed in the EDX spectra of TiO 2 /PET sheets.EDX data indicated that the percentage amounts of TiO 2 on PET surfaces were 42.17 and 36.54%,respectively.This means that only 6% of TiO 2 was lost after seven cycles giving rise to both the high performance and durability of the prepared composite.

Conclusions
A home-made designed and fabricated bench-scale solar CPC recirculating photocatalytic reactor has been utilized for the photocatalytic oxidation of organic pollutants.Evonik TiO 2 Aeroxide P25 immobilized on PET was prepared via an effective cost method.The photocatalytic performance of TiO 2 /PET sheets was tested using MB as a model pollutant.COD analyses reveal the complete mineralization of MB.Durability of TiO 2 /PET composite was also tested under sunlight irradiation.TiO 2 /PET showed good durability.Results indicate only 6% reduction in the amount of TiO 2 after seven cycles.No remarkable change was observed for the physicochemical characteristics of TiO 2 /PET after the irradiation processes.
cpc : The CPC concentration ratio, dimensionless COD: Chemical oxygen demand, mg/L   : Equivalentdiameter,m  0 : The initial COD value of MB, mg/L  0A : Concentration of MB in solution after the dark adsorption, mg  0B : Concentration of MB in solution before the dark adsorption, mg   : The COD value at each time interval of sunlight irradiation, mg/L EDX: Energy dispersive X-ray : The friction factor, dimensionless FT-IR: Fourier transform infrared : The local acceleration due to gravity, m/s 2 ℎ  : Theheadlossduetofriction,m : Kelvin temperature, degree : Thelengthofthepipe,m ṁ: The mass flow rate, kg/s : NumberofPETstrips MB: Methylene blue PET: Polyethylene terephthalate   : Reynoldsnumber,whichisaquantityusedto characterize different flow regimes (laminar or turbulent), dimensionless   : The experimental time for each withdrawn sample, min  30 w : Normalized illumination time at UV power of 30 W/m 2 : AreaofPETstrip,m 2 SEM: Scanning electron microscope UV: Ultraviolet UV: The average solar ultraviolet radiation, W/m 2 : The mean velocity of the fluid, m/s   : The irradiated volume of the MB solution, L   : The total volume of the treated MB solution, L  1 : Weights of PET substrate before loading with TiO 2 , g  2 : Weights of PET substrate after loading with TiO 2 , g : The density of the fluid, kg/m 3 : The dynamic viscosity of the fluid kg/(m⋅s) : A mathematical constant, the ratio of a circle's circumference to its diameter, commonly approximated as 3.14159 : The distance of CPC arc, m   : The half angle of acceptance, degree  max : Wavelength at the maximum absorption value of MB, nm.

Figure 1 :
Figure 1: (a) CPC acceptance angles, (b) Schematic cross section representation of used annulus to flow the wastewater [38].

Figure 2 :
Figure 2: Fabricated bench-scale CPC photoreactor was used in this work.It contains seven low iron borosilicate tubes connected in series (a).Each tube was fixed on a reflector fabricated from stainless steel grade 304 (b).T connectors were used to obtain a flexible lab-scale system (use only one tube or more and withdrawing sample at any tube is possible).The simulated wastewater is fed from the lower tube to the upper one against the gravity using centrifuge pump connected to 4-L stirrer tank (not shown).The output treated wastewater was circulated to the tank (closed system).TiO 2 immobilized on PET sheets were wrapped on temperature-resistant removable Teflon rods hanged in the middle of each glass tube.

Figure 6 :Figure 7 :
Figure 6: The change in the COD removal percentage at different intervals of irradiation time.
Composite.PET/TiO 2 composite was prepared by dip coating of PET textile substrate with TiO 2

Table 1 :
The initial amount of adsorbed MB molecules on TiO 2 /PET sheets at different repetitions.
the composite.However, the vibrational peak of carbonyl group in PET at 869 cm −1 is slightly shifted to 871 cm −1 .This shift corresponds to the surface complexation of MB and PET through functional groups (-CO) via hydrogen bonding.