The study of photocatalytic degradation of a commercial azo reactive dye in a simple design reusable miniaturized reactor with interchangeable TiO2 nanofilm

Abstract A simple design and low cost miniaturized reactor integrated with interchangeable thin film TiO2 nanolayer was successfully fabricated for the photocatalytic degradation of azo dyes. The TiO2 nanofilms were prepared by sol-gel dip-coating method, while the miniaturized reactor was fabricated on poly methyl methacrylate (PMMA) substrates, using a laser cutting machine. The performance of the miniaturized reactor for the photocatalytic degradation process was investigated for the degradation of a commercial dye (Novacron Red C-2BL). About 98% degradation of the commercial dye was achieved after 100 min in a stopped flow system, and 15% in a continuous flow system. The effect of different operating variables such as pH, initial flow rate, light intensity, layers of the nanoparticles, and temperature on the photocatalytic degradation was studied and the optimum operating conditions were determined to be: inlet flow rate of 0.05 ml/s, pH of 7, UV power 82 W and using a multi-layer of TiO2 thin film in the miniaturized reactor. The reaction kinetics was described as pseudo first order kinetics and rationalized using the Langmuir–Hinshelwood model.


Introduction
Wastewater from textile industries is a major environmental problem, predominantly in developing countries.It was reported that about 1-20% of the total world production of dyes is discharged as textile effluents into surface water (Akpan & Hameed, 2009).These pollutants are toxic to microorganisms, aquatic life, and human beings, thus causing a great concern for both environment and public health.
Dye degradation from textile industrial wastewaters has therefore received increasing attention, and the demand toward more efficient and sustainable remediation has been upturned (Fosso-Kankeu, Waanders, & Geldenhuys, 2015; Katheresan, Kansedo, & Lau, 2018).Accordingly many methods of remediation have been preferred, among which photocatalysis was found to be one of the attractive techniques used effectively worldwide (Gangu, Maddila, & Jonnalagadda, 2019;Mohammed Redha et al., 2017;Szczepanik, 2017;Yusuf, Redha, Al Meshal, & Shehab, 2018).Traditional water treatment methods usually lead to the formation of secondary pollutants, while photocatalytic degradation has the ability to completely mineralize the target pollutant.Moreover, textile dyes are considered to be photocatalytically stable, thus making them more amenable for photocatalytic degradations.
Titanium dioxide (TiO 2 ) has been integrated as a 'close to' ideal semiconductor for photocatalysis, due to its low cost, high stability, and being environment friendly, which encouraged researchers to intensively employ it for photocatalytic wastewater treatments (Gupta & Tripathi, 2011;Tesana, Metha, Andersson, Ridings, & Golovko, 2018;Vu et al., 2013;Zhiyong, Laub, Bensimon, & Kiwi, 2008).The tendency to use TiO 2 nanoparticles has been massively increased in industries, especially when compared to fine particles.The smaller size which contributed to larger surface to volume ratio improves the catalytic activity of the particles and hence the production yield overall (Gupta & Tripathi, 2011).
The design of the photocatalytic system plays a significant role in the efficiency of the degradation process.For example, for practical applications, immobilized photocatalysts are more advantageous in comparison to suspended nanoparticles.Moreover, flow systems are preferred over batch systems for some industrial applications.Therefore, nanofilms could be considered as an effective form for such requirements.
In recent years, miniaturized reactors have been developed as a promising platform for different applications.The effectiveness of these reactors is featured with the achievement of laminar flow, providing longer residence time within the system (Das & Srivastava, 2016).Moreover, miniaturization can assure shorter molecular diffusion distance and larger surface to volume ratio, in comparison to conventional reactors.In terms of optical properties, with miniaturization uniform illumination and excellent light penetration could be achieved.
Flow systems have been implemented in the presented application due to the high demand for dye degradation in industries' wastes, especially in textile manufacturing fields, where flow processes are predominant.However, according to the authors' knowledge, degradation of azo dye using TiO 2 nanofilm in microfluidic systems is limited in the literature (de S a et al., 2018;Deepa, Meghna, & Kandasamy, 2014;Liu, Li, Wu, & Xia, 2018;Suhadolnik et al., 2019).
In light of these features, an in-house fabricated miniaturized reactor integrated with thin film TiO 2 nanoparticles for the degradation of azo dyes is presented in this study.This adds remarkable features over conventional reactors such as lower cost, faster reaction, developed yield, controlled process conditions, easier scalability, safer process, and more efficient energy and heat transfer.Such features are of high importance for large-scale industrial applications.TiO 2 nanofilms have been employed in this study for photocatalytic degradation of the azo dyes for several reasons.It is well documented that TiO 2 films are eco-friendly.They feature high oxidation efficiency and low solubility in water, which make them best candidates for industrial wastewater treatment (Stambolova, Shipochka, Blaskov, Loukanov, & Vassilev, 2012).Moreover, unlike suspended nanoparticles, nanofilms provide better stability, and eliminate the need for post-treatment separation systems.With such characteristics, TiO 2 nanofilms are considered as an excellent choice for continuous flow applications.The presented work highlights a simple design yet efficient miniaturized reactor that could well be considered as a suitable alternative for continuous flow industrial wastewater treatment.

Fabrication of the miniaturized reactor
The miniaturized reactor was fabricated in-house at the University of Bahrain.A rectangular, thin-layer design of the miniaturized reactor with the required specification was drawn using AutodeskV R AutoCADV R 2015 program.The design consisted of two parts; the top part and the bottom part.The top part contained the inlet and outlet holes as well as a rectangular hinge designed to fit into the reaction area (to avoid solution leakage).The bottom part was designed to incorporate the reaction area where the catalyst sheet is to be inserted.The dimensions for each part are summarized in Table 1.The fabrication was made precisely such that upon joining the two parts, after the insertion of the catalyst, the effective depth of the reactor would be 500 mm.
The chip was fabricated on poly methyl methacrylate (PMMA) substrates, using a laser cutting machine (META 1.5 C; TLM Laser Ltd., Worcestershire, UK).PMMA was selected as a substrate for this application for many reasons; one of which is its excellent optical transparency.It is cheap, easy to handle, environmentally friendly and has high chemical stability (Liang, Liu, Zhang, Zhang, & Han, 2020).All these properties make PMMA suitable for industrial application.During fabrication, a focused laser beam was directed to cut the required dimensions.Then, 1 mm ID brass rod fittings were screwed at both inlets and outlets of the miniaturized reactor, after cleaning.Finally, a glass sheet of TiO 2 thin film nano-particles was inserted into the bottom part of the reactor, and the two parts were clamped together by screwing the four corners of the reactor.The complete miniaturized reactor is shown in Figure 1.

TiO 2 thin film synthesis
TiO 2 thin films were deposited onto pre-cleaned glass substrates (40 Â 25 Â 1 mm) by sol-gel dip-coating method.Tetraethyl-orthotitanate (95% purity; Merk) precursor was dissolved in a solution containing ethanol, water and nitric acid (69% purity; Sigma-Aldrich), then stirred for an hour.After aging  for 24 h, the obtained gel was deposited onto glass substrates by a dip-coater using a withdrawal speed adjusted at 1 mm/s.The sample was then heated for 10 min at 400 C for the removal of the solvent, residual organics and to be ready for the deposition of the second layer.These cycles of dipping and drying were repeated until the desired thickness was achieved.For better crystallization and chemical composition homogeneity, the obtained film was subjected to annealing in air for 1 h, using Carbolite BLF 1800 at 450 C. Finally, single layers of TiO 2 thin films, formed from 5 nanolayers (five dipping cycles), with an estimated thickness around 250 nm, were successfully obtained.The synthesized films were characterized by having a smooth surface, with a high transparency and uniform adherent on both sides of the substrate.

Apparatus and procedure
The photocatalytic degradation experiments were carried out on a thin film titanium dioxide sheet, integrated in a specially designed miniaturized reactor.In all experiments, UV light was illuminated on the nanoparticles using Mineralight R-52G Grid Lamp from UVP (USA).Novacron Red dye solutions were continuously pumped into the miniaturized reactor using a variable-flow pump (Cole-Parmer, UK; Model: EW-73160-40), and the degree of decolorization was determined at its maximum absorption (k ¼ 526 nm) using UV-Vis spectroscopy (path length 1 cm, apparatus: Shimadzu UV-1800).A heater (IKAV R -Werke, Germany) with a stirrer was used for temperature control experiments.All experiments were carried out at room temperature unless otherwise stated.The pH of solutions was measured using Thermo Scientific TM Orion TM Star A111 pH Benchtop Meter, from ThermoFisher Scientific, USA.

TiO 2 thin film characterization
Figure 2 displays the X-ray diffraction (XRD) pattern of the deposited and annealed TiO 2 film.Welldefined peaks with relatively low intensity and large broadening were observed, indicating the nanocrystalline nature of the deposited film.All peaks were indexed within a tetragonal phase of anatase-type structure in agreement with JCPDS card No. 12-1272.The halo observed in the range 2h ¼ 20-35 can be attributed to the amorphous glass substrate.
The crystallite size (D) was estimated using the well-known Scherrer equation: Where k is the X-ray wavelength, h is the Bragg diffraction angle and b represents the full width at half maximum (FWHM).The estimated value of the crystallite size was found to be around 13 nm, indicating that the as-grow film is composed of "grains" at the nanoscale regime.
The surface morphology of TiO 2 film, as observed by AFM, exhibits homogeneous arrangement of grains with typical size of 25 nm and maximum size of 40 nm (Figure 3a), while the 3 D image revealed sharp cones with a mean height of 4.8 nm (Figure 3b).The surface roughness was found almost constant (flat region), the RMS (root mean square) was about 0.90 nm.
The UV-Vis spectrum (Figure 3c) revealed very pronounced interferences in the transparency region 350 À 1500 nm with sharp fall of transmittance at the band edge, which indicated the good homogeneity of the deposited film thickness.
The spectral absorbance A(k) is calculated from the experimental data of R(k) is reflectance and T(k) is transmittance by using the following equation (Bensouici et al., 2015): The absorbance (A) is related to the absorption coefficient (a) is related to the film thickness (d) by the simple equation: Then, the optical indirect bandgap E g can be determined from the well-known energy-exponential relation (Bensouici et al., 2015), as follows: where A Op is a constant for each film.The plot of (AE) 0.5 versus E shown in the inset of Figure 3(c) and using the Tauc method, gave the indirect band gap (E g ) value, which was found to be 3.33 eV for TiO 2 film.This value is much higher compared to TiO 2 quantum dots below 3 nm (QDs) functionalized with surface hydroxyl group, having a narrow E g of 2.85 eV (Deng et al., 2018), but closer to 3.2 eV for TiO 2 (Anatase) nanoparticles with a mean size below 25 nm (D'Amato et al., 2018), and to TiO 2 (Anatase and Rutile) synthesized by green mediated combustion route using Aloe Vera peel, which was reported to have broad variation of E g in the range 3.50 À 3.20 eV for irregular porous cone-shaped nanoparticles (20 À 45 nm) (Harlapur et al., 2017).
The change in energy band gap is attributed to both size and shape effects, known as quantum confinement, when the size of particles (grains) are at nanoscale.In turns, this would have a direct influence on the physical properties and performance, such as the photocatalytic activity during the degradation of dyes.

Miniaturized reactor validation
The miniaturized reactor with the integrated nanofilm was utilized to perform all the photocatalytic experiments described in this study.The photocatalytic performance of the miniaturized reactor for the degradation of the dye was evaluated using the following equation: Where, D is the percentage degradation, C 0 , is the initial concentration of the solution, C t is the final concentration measured at time t.The corresponding concentration values were obtained from the calibration graph (with R 2 ¼0.9989), using the following relationship: Where A t is the measure absorbance corresponding to the given concentration at time t.

Reaction kinetics
It is well known that the most commonly endorsed photocatalytic degradation mechanism in the presence of titanium dioxide catalyst involves the oxidation of the organic dye by the generated hydroxyl radical (Deepa et al., 2014;Naeem & Ouyang, 2013;Rauf & Ashraf, 2009).The suggested mechanism is shown in the following schematic (Figure 4) and can be summarized through reactions RXN 1 to 5. In brief, the titanium dioxide nanoparticles will be photo-excited upon irradiation by UV light, producing electron-hole pairs RXN 1 (Linden & Mohseni, 2014).The produced electrons will subsequently react with the dissolved oxygen molecules to produce superoxide radical anions (O À 2 Þ according to RXN 2, whereas the produced holes will react with surface bound water molecules to produce hydroxyl radicals (OH Þ as in RXN 3. Hydroxyl radical generated in reactions RXN 3 and 4 will oxidize the organic dye causing its degradation according to reaction 5.
Where, e À cb and h þ vb are the electron-hole pairs, respectively.
The photocatalytic degradation rate of azo dyes could be described by pseudo first order kinetics and rationalized using the Langmuir-Hinshelwood model (Barka, Qourzal, Assabbane, Nounah, & Ait-Ichou, 2010;Li, Li, Li, & Yin, 2006;Mir, Khan, Dar, & Muneer, 2014).At low concentrations, this equation can be simplified to: Integrating Equation (1) with the limit of C¼C 0 at t ¼ 0, where C 0 being the initial concentration in the bulk solution, will yield to the well-known relation: Where, k 0 (min À1 ) is the apparent pseudo first order rate constant.
In this study, a plot of ln C 0 C versus t for all the experiments had yielded a straight line with R 2 above 0.9 in all cases; representing a good correlation of the Langmuir model.An example of this correlation is shown in Figure 5(a) and (b) for both continuous and stopped flow conditions respectively.The UV-Vis spectrum at different concentrations are shown in Figure 5(c).The details of optimization are discussed in the following sections.Using the miniaturized reactor, a degradation as high as 98% was achieved in only 100 min at stopped flow and 12% in continuous flow using low power intensity.In comparison to literature, this work represents a remarkable degradation efficiency at low luminescence intensity.Almost complete degradation was achieved in this work at 100 min compared to similar studies which obtained the same degradation efficiency but under 5 times higher luminescence industry (Habibi, Hassanzadeh, & Mahdavi, 2005).Bansal, Singh, and Sud (2010) obtained similar degradation after 15 h using commercial available titanium oxide catalyst in a batch process, while Liu et al. (2018) achieved the same degradation efficiency in 2 h in a continuous process.With TiO 2 nanofilm, Stambolova et al. (Stambolova et al., 2012) reported approximately 80% degradation in 180 min in a batch process where the nanofilm was sprayed with the dye solution.Recent studies by Suhadolnik et al. (2019) reported only 10% degradation in their continuous design.This efficiency was further improved to 100% using more sophisticated design that utilized a complicated hybrid system of photo-electro catalytic degradation.In comparison to the latter, our study achieved similar degradation yet with an efficient and simple design.A summary of these studies in comparison to the current paper is shown in Table 2.

Effect of increasing the number layers of the nanoparticles
Figure 6 shows the effect of titanium dioxide nanoparticles on the photocatalytic degradation of Novacron Red dye.Improved degradation was achieved in the presence of the TiO 2 nanoparticles over with just illuminating the sample with UV light.The photocatalytic degradation reaction showed pseudo first order reaction kinetics as confirmed by the linear fit between the ln(C 0 /C) and irradiation time for both cases.Similar results were obtained by other researchers as well (Kaur & Singh, 2007a).Moreover, the apparent pseudo first order rate constants obtained from the slopes of the two conditions were compared and the results showed that the degradation rate constant was doubled in the presence of the nanoparticles in comparison to only UV light as shown in inset of Figure 6.
Investigating the effect of mono and double layered nanoparticles on the degradation of the Novacron Red dye revealed improvement in the percentage dye removal with the double layer in comparison to the monolayer (Figure 7).This could be attributed to the increase in the effective surface area of the nanoparticles.As a result, more H 2 O and O 2 molecules could be captured, forming more active species for the degradation process (Deepa et al., 2014;Sayılkan et al., 2007).The rate constants shown in the inset of Figure 7 confirm this increase in the photocatalytic performance.

Effect of flow rate
Studying the effect of solution flow rate is an essential parameter in continuous processes because it determines the residence time at which the dye is in contact with the nanoparticles.As a matter of fact, the higher the flow rate the less the residence time and thus, the shorter the contact between the dye molecule and the generated active species (Deepa et al., 2014).As a consequence, the dye degradation rate will decrease.Similarly, in this experiment, the degradation of the dye was observed to decrease as the solution flow rate increased from 0.05 to 0.2 ml/s confirming the above discussion.The obtained results are shown in Figure 8.Moreover, from the inset of Figure 8, it is revealed that the reaction rate constant is considerably higher at lower flow rates than that obtained at higher flow rates.Here the apparent rate constant was approximately three times more at 0.05 mL/s than that at 0.2 mL/s, and twice more at 0.1 mL/s.

Effect of UV power
The light intensity plays an important role in the photocatalytic degradation reaction as it has a direct effect on the generation of the active hydroxyl radicals (Deepa et al., 2014).In semiconductors, the production of electron and hole pairs is governed by the intensity of the illuminated light.This is an essential step in the formation of the active radicals  for the photocatalytic degradation process.Thus, the higher the light intensity the more the transfer of the electrons from the valence band to the conduction band, consequently the more generation of the hydroxyl or oxygen radical.In this experiment, the effect of light intensity was investigated at 32, 40 and 82 W and the results are shown in Figure 9.
Confirming the above discussion, the degradation efficiency increased by increasing the illuminated light intensity.Moreover, the reaction rate constant was observed to increase linearly with the increase in the light intensity for the investigated range.Due to the instrumental limitation, higher intensities could not be investigated and thus, 82 W was selected as optimum luminous intensity.

Effect of pH
The effect of solution pH on the photocatalytic degradation efficiency was intensively studied by many  researchers (Alahiane, Qourzal, El Ouardi, Abaamrane, & Assabbane, 2014;Deepa et al., 2014;Kaur & Singh, 2007a, 2007b).Studies have shown that the degradation kinetics of some organic dyes are highly influenced by the pH of the solution since it can change the surface charge of the photocatalyst.Generally, the surface of the titanium oxide is known to be covered with hydroxyl groups and the protonation or deprotonation of these groups is assumed to originate the surface change of the photo catalyst as represented by the following reaction (Rauf & Ashraf, 2009): Which of these two reaction is predominant depends strongly on the solution pH.In our experiment, the dependency of the dye degradation efficiency was investigated at three pH values of 4, 7 and 9 (Figure 10).Slight improvement in the dye degradation at around pH 7 was observed in contrast to acidic and basic conditions.Comparing the rate constant values at the three conditions showed 20-30% improvement in the rate constant values at neutral media in comparison to acidic and basic media respectively.Thus, pH 7 is more advantageous than alkaline or acidic pH.Moreover, the lowest degradation was observed at alkaline conditions.This could be attributed to the negatively changed surface of the nanoparticles as described in reaction (RXN 6), which will repel the hydroxide anions thereby preventing the formation of the hydroxyl radical and reducing the degradation efficiency (Rauf & Ashraf, 2009;Sriwong, Wongnawa, & Patarapaiboolchai, 2012).Similarly, there will be electrostatic repulsion forces between the surface and the dye anions, which will impede the adsorption of the dye molecules on the nanoparticles thereby reducing the degradation efficiency.
On the other hand, the slight improvement in the degradation rate at acidic conditions in comparison to alkaline conditions observed in Figure 10 could be explained by reaction (RXN 7), where the positively charged surface of the nanoparticles causes the electrostatic attraction of the dye anions, leading to their absorption on the surface.However, this improvement was not very significant.It was reported that at very acidic conditions, the presence of excess hydrogen cations can reduce the photocatalytic degradation reaction (Rauf & Ashraf, 2009).
In this experiment, the highest degradation efficiency observed at pH 7 which is close to the point of zero net charge of TiO 2 (pH 6.8) (Samsudin et al., 2015) suggests that the dye anions are not being radically adsorbed to the nanoparticles, thus, the main degradation mechanism here is controlled by the hydroxyl radical in the solution.

Effect of temperature
The effect of temperature was investigated in the range of 25-60 C at a flow rate of 0.05 ml/s and constant pH of 7. From the experimental results, it was observed that the temperature dependency of the dye degradation was not significant for the investigated range (Figure 11).The rate of the reaction remained almost constant.Similar observation was recorded by many researchers for certain azo dyes (Habibi et al., 2005).It was reported that the photocatalysis process of semiconductors is not very temperature sensitive and the dependency of the rate constant on temperature is not very significant (Chen & Ray, 1998).Moreover, in the    investigatedrange of temperatures, 25-60 C, there is usually weak dependency of the rate of degradation on temperature as was observed by many researchers in the field (Zhang, Wang, & Li, 2005).

Optimum conditions
From the single variable optimization experiments, the optimum conditions for the investigated ranges of this study were found to be pH of 7, flow rate of 0.05 mL/min, UV power of 82 W and room temperature.A comparison between the optimized and unoptimized conditions for the degradation of the dye was investigated and the pseudo first order kinetic plot is shown in Figure 12.It is clear from the figure that there is a comparable difference in the reactor's performance between the optimized and un-optimized conditions.The apparent pseudo first order rate constants increased by a factor of 8 when optimized conditions were employed.This concludes that the operating conditions have significant effect on the degradation reaction.

Conclusions
This study investigated the feasibility of a specially designed miniaturized reactor for the photocatalytic degradation of reactive dye.A thin film of titanium dioxide nanoparticles was integrated into the miniaturized reactor producing an effective reaction depth of 500 mm.It was evident that Novacron Red dye was effectively degraded by TiO 2 nanoparticles upon its illumination by UV light.The designed miniaturized reactor was successful in achieving 40% degradation in 40 min and 98% degradation in 100 min in stopped flow configuration.Moreover, the double layer nanoparticles proved to be more effective than the single layer.The photocatalytic degradation kinetics was proved to follow pseudo first order and was found to be affected by the operating conditions: flow rates, illumination power, and pH.More degradation was obtained at lower flow rates, higher UV illumination power and neutral media.However, the effect of temperature was found insignificant.

Figure 2 .
Figure 2. X-ray diffraction patterns of TiO 2 film deposited onto glass substrate.

Figure 4 .
Figure 4. Schematic diagram of the photocatalytic dye degradation mechanism on titanium oxide nanoparticles. e

Figure 5 .
Figure 5. Pseudo first order kinetic plot of the Novacron Red dye solution.Conditions: (a) continuous flow system at 0.05 mL/ min, (b) stopped flow, 15 mg/l initial dye concentration and UV power 27.6 W (c) UV-Vis spectra of the dye at different dye concentrations.

Figure 6 .
Figure 6.Photocatalytic degradation of the Novacron Red dye with and without TiO 2 nanoparticles (Inset: Pseudo first order kinetic plots of the dye solution for both experiments).Conditions: stopped flow system, 15 mg/l initial dye concentration, UV power 27.6 W.

Figure 7 .
Figure 7.The effect of the number of the deposited TiO 2 nanoparticles layers on the degradation of Novacron Red dye solution.Conditions: continuous flow system, UV power 27.6 W. (Inset: Pseudo first order kinetic plots of the dye solution for both experiments).

Figure 8 .
Figure 8.The effect of flow rate on the photocatalytic degradation of the Novacron Red dye solution.Conditions: continuous flow system, UV power 27.6 W (Inset: Corresponding plots of the effect of flow rate on the apparent pseudo first order rate constant).

Figure 9 .
Figure 9.The effect of the applied light intensity on the photocatalytic degradation of the Novacron Red dye solution.Conditions: continuous flow system, with flow rate of 0.05 ml/s and pH of 7 and using a multi-layer of TiO 2 thin film.

Figure 10 .
Figure 10.The effect of pH on the photocatalytic degradation of the Novacron Red dye solution.Conditions: continuous flow system, with flow rate of 0.05 ml/s, UV power 27.6 W. (Inset: Pseudo first order kinetic plots of the dye solution for both experiments).

Figure 11 .
Figure11.The effect of temperature on the photocatalytic degradation of the Novacron Red dye solution.Conditions: flow rate of 0.05 ml/s, pH at 7 and using multi-layers of TiO 2 thin film.

Figure 12 .
Figure12.Pseudo first order kinetic plot of the Novacron Red dye solution of 15 mg/l initial dye concentration in continuous flow system at optimized condition: 0.05 mL/min, pH 7, UV power 82 W, and un-optimized conditions: 0.2 mL/min, pH 7 and UV power 27.6 W.

Table 1 .
Summary of the miniaturized reactor's dimensions.

Table 2 .
Summary of similar studies in the field compared to the current work.