Simulation of photocatalytic degradation of methylene blue in planar

microreactor with integrated ZnO nanowires

Fayazeh Rabanimehr, Mehrdad Farhadian* , Ali Reza Solaimany Nazar, Elham Sadat Behineh

Department of Chemical Engineering, Faculty of Engineering, University of Isfahan, Isfahan, Iran.

GRAPHICAL ABSTRACT

Experimental data
0.7 L- H kinetic model
0.6
0.5
0.4

cout/c0
0.3
0.2
0.1
0.0
0 45 90 135 180 225 270

Residence time, s

Photocatalytic degradation of methylene blue in planar micro-reactor

ARTICLE INFO ABSTRACT

Article history: In photocatalytic microreactors the catalyst layer is obtained by integration of

Received 6 February 2021 nanostructure films of semiconductors. One of these nanostructures that have a
Reviewed 17 April 2021 good photocatalytic activity is ZnO nanowires. The photocatalytic degradation of
Received in revised form 27 May 2021
methylene blue in a continuous flow microreactor with ZnO nanowires deposited
Accepted 29 May 2021
Available online 2 June 2021 film is simulated. A finite element model is developed using COMSOL Multiphysics
version 5.3 software to simulate the microreactor performance. The kinetic law of
the photocatalytic reaction is assumed to be Langmuir–Hinshelwood. The kinetic
Keywords: constants kLHa and K are determined 1.43 × 10-7 mol/m2s and 7.5 m3/mol,
Planar microreactor respectively. The percent of average absolute deviation of the model in predicting
ZnO nanowires the methylene blue outlet concentration obtained about 0.12% mol/m 3. The model
Simulation showed a very good agreement with the published experimental data. The effect of
Langmuir–Hinshelwood microreactor depth, methylene blue inlet concentration and flow rate on the
COMSOL Multiphysics
methylene blue degradation is also investigated. The simulation results showed
Article type: Research Article that the microreactor with shorter depth and lower values of inlet concentration and
flow rate has higher efficiency. Thiele modulus and Damköhler number are both
estimated lower than 1. It indicates that the photocatalytic reactions occur without
internal and bulk mass transfer limitations.
© The Author(s)
Publisher: Razi University

1. Introduction after pollutant removal is a difficult and time consuming process

(Dijkstra et al. 2002; Mills et al. 1998). Therefore, it is highly interesting
Nowadays, due to the increasing demand for clean water, water to immobilize the photocatalysts on a solid surface in the form of a film
purification is of particular importance (Wang et al. 2014). However, coating. Performing the photocatalytic reactions inside the
many contaminants in water and wastewater cannot be treated microreactors with integrated photocatalyst deposited film, has
efficiently by conventional methods (Ali et al. 2016; Jewell et al. 2016; advantages over than in large-size reactors. Short molecular diffusion
Wang et al. 2016). As a result, the photocatalytic water purification distance, high surface to volume ratio, fast mixing and high illumination
capable of decomposing and mineralizing a wide range of organic homogeneity are some of the many features of these types of reactors
pollutants has absorbed attentions (Wa ng et al. 2014). In conventional (Wang et al. 2014). Regarding these advantages, integration of
slurry photoreactors, photocatalysts nanoparticles are homogeneously nanoporous films of semiconductors like TiO2 (Krivec et al. 2013) and
suspended in an aqueous solution. However, this approach is not ZnO (Zhang et al. 2013) into photocatalytic microreactors has been
efficient due to the non-uniform distribution of light and low photon absorbed considerable attentions for organic pollutants removal from
transfer. Also, recycling the suspended photocatalyst nanoparticles water.
*Corresponding author Email: [email protected]

How to cite: F. Rabanimehr, M. Farhadian, A. R. Solaimany Nazar, E. S. Behineh, Simulation of photocatalytic degradation of methylene blue in planar
microreactor with integrated ZnO nanowires, Journal of Applied Research in Water and Wastewater, 8 (1), 2021, 36-40.
Rabanimehr et al. / Journal of Applied Research in Water and Wastewater 8 (1) 36-40

Charles et al. (2011) investigated salicylic acid photodegradation

reaction in a microchannel reactor with immobilized TiO2 as It is considered that the photocatalytic degradation of MB takes place
photocatalyst. They observed that higher salicylic acid on the side covered with photocatalyst film in the microreactor.
photodegradation efficiency can be achieved in higher incident light It is assumed that the catalytic surface is smooth, unlike it is
intensity and lower pollutant concentration and flow rate. They also nanostructured.
determined the kinetic constant of salicylic acid photodegradation The catalytic surface reaction is considered to be equal to the boundary
reaction. Their observations showed that the reaction rate constant is flux at the catalytic surface.
only dependent on incident light intensity. Corbel et al. (2014) studied
mass transfer in microchannel photocatalytic reactor experimentally 2.3. Model
and numerically. They investigated the salicylic acid removal in the
presence of immobilized TiO2 nanoparticulates as photocatalyst. The A single phase, Newtonian and incompressible flow has been
results of the numerical simulation showed a very good representation considered in the microreactor. The Navier-Stokes and the continuity
of the experimental data. They also established a correlation for equations are used for conservation of momentum and mass
Sherwood number that is reliable for all microchannel geometries. respectively, to simulate the laminar flow through the microreactor.
Jayamohan et al. (2015) investigated the performance of a planar ∂u
⃗
microreactor with titanium nanotube array (TNA) and TiO2 ⃗ .∇ ⃗u=-∇⃗⃗⃗p+μ∇2⃗u
ρ ( ) +u (1)
∂t
nanoparticulate as photocatalysts for methylene blue (MB) degradation
and found that the microreactor with TNA photocatalyst has better ⃗ =0
ρ∇u (2)
performance compared to TiO2 nanoparticulate. They also simulated
the performance of the microreactor and deduced that MB degradation where, u is the flow velocity field, ρ is the fluid density, µ is the dynamic
efficiency is not affected by mass transfer limitation. He et al. (2010) viscosity of the fluid and p is the fluid pressure. 𝜌 = 1000 kg/m3 and 𝜇 =
demonstrated that ZnO nanowires (NWs) could be grown on the inner 0.001 Pas are used in calculations. The boundary conditions are
wall of capillaries and then coated by TiO2 nanoparticles. This imposed as follows: at the inlet of the microreactor, the inlet mass flow
composited photocatalyst showed a reasonable performance. rate is known so the inlet velocity is uniform in y direcrion and no-slip
Han et al. (2013) applied a planar microreactor operating in a close boundary condition is considered at the channel surfaces. Also, the
loop with integrated ZnO NWs over a glass substrate for relative fluid pressure at the outlet of the microreactor is zero.
photodegradation of MB. They found that the optimal photocatalytic A convection-diffusion equation is adopted to describe the species
conditions of microreactor can be obtained with annealed ZnO NWs mass transfer in the bulk of the microreactor:
subjected to UV irradiation. Because the structure of annealed ZnO ⃗
∂c
NWs has better crystallinity and its surface has lower defects compared =-∇.(-D∇c+c⃗u) (3)
∂t
to unannealed one. This study showed that the ZnO NWs integrated
where, c is the concentration of species and D is diffusion coefficient of
microreactor exhibits more photodegradation efficiency over using
the species. In Eq. 3, steady-state condition is considered. The
suspended ZnO NWs. As the fabrication of microreactors is costly,
concentration of MB at the inlet of the microreactor is known and the
difficult and time-consuming process, simulation of the microreactors,
MB concentration gradient at the outlet of the microreactor is equal to
can be useful for predicting their performances. However, in spite of
zero. At the reactor walls insulating condition is considered. The
many experimental works done on photodegradation of organic
degradation reaction of MB is done at the ZnO NWs catalytic surface of
pollutants in microreactors, simulation and modeling the performance
the microreactor. The photocatalyst absorbs a photon with energy equal
of these systems remains limited.
to or exceeded than its band gap energy and produces the electron and
The aim of this work is to simulate the performance of the integrated
hole in its conduction and valence bands, respectively. The
ZnO NWs planar microreactor studied by Han et al. (2013). For this
photogenerated holes oxidize water and produce hydroxyl radicals.
purpose, a computational 2-dimensional (2-D) model is setup using
Also, electrons produce the superoxide radicals by reduction of
COMSOL Multiphysics version 5.3 software to predict the concentration
dissolved oxygen. The generated free radicals degrade most organic
of MB at the outlet of the microreactor. The catalytic surface of the
water pollutants like MB at the surface of the photocatalyst. As
microreactor is assumed to be flat rather than the nanostructured. The
mentioned in assumptions, the catalytic surface reaction ra (mol/m2s) is
kinetic reaction rate constants are estimated by minimization of an
considered to be equal to the boundary flux at the catalytic surface:
objective function. The objective function is defined on the sum of
squares of residual concentrations. Dimensionless numbers and the ra =-D∇cs (4)
effect of microreactor geometry, inlet concentration and flow rate on MB
degradation efficiency are also investigated. where, cs is surface concentration of MB. The rate of removal of MB is
described by Langmuir–Hinshelwood (L-H) kinetic model which is an
2. Microreactor specifications adsorption-kinetic model and is applicable for photocatalytic removal of
organic pollutants from water (Corbel et al. 2014):
The simulated microreactor is planar with annealed ZnO NWs kLHa K cs
catalytic surface at the bottom, which is subjected to UV light (365 nm) ra = (5)
1+K cs
irradiation. The ZnO NWs have been grown on a glass substrate using
hydrothermal method and the microreactor has been made from where, K is adsorption constant in m 3/mol and kLHa is the surface
polydimethylsiloxane using soft lithography method. The inner diameter reaction rate constant in mol/m2s, which is calculated from kLH/κ, where
of inlet and outlet tubes of the microreactor is 0.89 mm. Fig. 1 shows kLH is the volumetric reaction rate constant in mol/m 3s and κ is the
the 3-D geometry of the investigated microreactor. Table 1 shows the specific surface area in 1/m. To predict the average MB concentration
microreactor geometric specifications. This microreactor is operated in at the outlet of the microreactor, the Langmuir–Hinshelwood kinetic
a closed loop and MB solution enters with a fixed flow rate of 400 µl/min. constants (kLHa and K) must be determined previously. For
According to the results, the microreactor volume is 75 mm3 and the determination of these parameters, an objective function is defined as:
exp. 2
residence time of MB solution in the microreactor is 11.25 s in each 1 ccalc. -ci
O.F.= ∑ni=1 ( i
exp. ) (6)
cycle. The microreactor totally operates in 24 cycles and the MB n ci
concentration at the outlet of the microreactor was measured every 4 exp.
cycles (Han et al. 2013). where, ccalc.
i and ci are calculated and experimental MB concentration
at the outlet of the microreactor at each residence time, respectively
2.1. Numerical modeling and n is the number of experimental data points. For determination of
kLHa and K, an initial guess was considered for each of them and the
The photodegradation of MB in a 2-D domain has simulated by objective function was calculated, then the value of parameters was
employing finite element method using COMSOL Multiphysics version modified several times in a way that the objective function reached its
5.3 software. The 2-D model domain of the microreactor is shown in minimum value. The final values of kLHa and K determined 1.43×10-7
Fig. 2. mol/m2.s and 7.5 m3/mol, respectively. For a complete triangular mesh,
the extremely fine mesh size was chosen.
2.2. Assumptions
3. Results and discussion
The assumptions made for the model are summarized as follows: 3.1. Validation of proposed model
Due to dilution of the solution, the physical properties of water are
considered. The steady state condition is considered. It is assumed that At the given microreactor geometry, the average outlet
the effects of the capillary force are negligible in the microreactor. concentration of MB (Cout) was predicted for each residence times. The

37
Rabanimehr et al. / Journal of Applied Research in Water and Wastewater 8 (1) 36-40

MB inlet concentration (C0) of 0.03 mol/m3 (Han et al. 2013), the

diffusion coefficient of 1.6 × 10-10 m2/s (Jayamohan et al. 2015) and the
determined kinetic rate constants are considered. Fig. 3 shows the
predicted and experimental concentration of MB at the outlet of the
microreactor as a function of residence time. As shown in this figure,
there is a very good agreement between experimental data and model
results. The percent of average absolute deviation (AAD, %) of the
model in predicting the experimental data is determined 0.12 % mol/m3
using Eq. 7:
1 exp.
AAD= ∑ni=1|ccalc.
i -ci | (7)
n
As illustrated in Fig. 3, the deviation between the model results and the
experimental data points is low at all residence times. As expected, Cout
decreases with increasing the residence time and it corresponds to the
longer contact time of MB in the photocatalytic microreactor.
0.7 Experimental data
Fig. 5. Concentration distribution of MB in the 2-D domain of
L- H kinetic model microreactor, residence time=11.25 s, C0=0.03 mol/m3 and flow rate=
0.6
400 µl/min (For clarity reason dimensions are not real).
0.5
3.4. Characteristic times
0.4
For further scaling analysis on the investigated microreactor, the
cout/c0

0.3 characteristic times are estimated and summarized in Table 2.

Table 2. Characteristic times.
0.2 Parameter Symbol Definition Values
Residence time 𝑡𝑅 𝑉𝑅 ⁄ 𝑄 45-270 s
0.1 Diffusion time
𝑡𝐷 h2⁄ 𝐷 62.5 s
(in the fluid)
0.0 Effective diffusion
0 45 90 135 180 225 270 𝑡𝐷𝑒 h02 ⁄ 𝐷𝑒 5.5 s
time (in the ZnO NWs)
Reaction time 𝑡𝑘 1 ⁄ 𝑘𝑎𝑝𝑝 90.9 s
Residence time, s
Fig. 3. The model results and experimental data (Han et al. 2013) on The residence time, 𝑡𝑅=𝑉𝑅 ⁄𝑄, shows the advective transport of MB
MB concentration at the outlet of the microreactor as a function of molecules by the fluid flow parallel to the surface immobilized by ZnO
residence time, C0=0.03 mol/m3 and flow rate= 400 µl/min. NWs, where 𝑉𝑅 is the reaction volume and 𝑄 is the flow rate. The
diffusion time 𝑡𝐷 = ℎ2⁄𝐷, shows the time required for molecular diffusion
3.2. Velocity distribution of MB in x direction, where h is the depth of microreactor and 𝐷 is the
molecular diffusion coefficient. To intend the diffusion inside the porous
Fig. 4 shows the velocity profile in the microreactor. As shown in ZnO NWs catalyst layer, the effective diffusion time 𝑡𝐷𝑒 = ℎ02 ⁄𝐷𝑒 is
this figure, the flow regime is laminar. The velocity component employed, where 𝐷𝑒 = 𝐷𝜀⁄𝜏 is the effective diffusion coefficient, 𝜀 is the
perpendicular to the photocatalytic surface is assumed to be zero and porosity, and 𝜏 is the tortuosity of the photocatalyst film which
the velocity profile parallel to the photocatalytic surface, follows a considering 𝜀 = 0.45 and 𝜏 = 1.0. The characteristic reaction time can
Gaussian distribution. be defined as the reciprocal of the kinetic rate constant: 𝑡𝑘 =1⁄𝑘𝑎𝑝𝑝. As
mentioned in section 3, the values of kLHa and K are 1.43×10-7 mol/m2.s
and 7.5 m3/mol, respectively and kapp defined as kapp= kLHa×K×κ.
As shown in Table 2, the value of reaction time is more than both
the values of diffusion time and effective diffusion time, so it can be
concluded that this microreactor is reaction limited and the diffusion of
MB molecules in the bulk of fluid and solid is done faster than the
reaction. Also, to investigate the mass transfer resistance within the
catalyst film, Thiele modulus has been calculated. Thiele modulus 𝜙 =
(𝑡𝐷𝑒 ⁄ 𝑡𝑘)0.5 represents the ratio of internal diffusion to kinetic time (Aillet
et al. 2015, Yusuf et al. 2019) and is estimated to be 0.24 that is lower
than 1. So it can be assumed that the reaction occurs in the catalytic
film without internal mass transfer limitations.

3.5. Parametric study

3.5.1. Effect of microreactor depth on MB degradation

As shown in Table 1, the depth of the microreactor is very low

Fig. 4. Velocity distribution in the 2-D domain of microreactor, (h=0.1 mm) and its specific surface area has a higher value (κ=10
residence time=11.25 s, C0=0.03 mol/m3 and flow rate= 400 µl/min. 1/mm-1) compared to conventional reactors. To investigate the effect of
(For clarity reason dimensions are not real). microreactor depth on MB degradation, the performance of the same
microreactor with different values of depth is simulated using COMSOL
3.3. Concentration distribution Multiphysics software. The degradation of MB is defined as:
C0 -Cout
Fig. 5 shows the concentration distribution of MB in the 2-D domain X= (8)
of the microreactor at residence time equal to 11.25 s (first cycle). The C0
molecular diffusion of MB from the bulk towards the catalytic surface of where, X is the mean conversion ratio. By increasing the depth of
ZnO NWs can be seen in this figure. As the photocatalytic degradation microreactor, the residence time of MB solution in the reactor will be
reaction happens at the catalytic surface, the MB concentration reduces increase. The effect of microreactor depth on the MB degradation in
at the bottom wall of the microreactor. The average outlet concentration constant flow rate is illustrated in Fig. 6. As shown in the figure,
of MB at this low residence time is about 0.027 mol/m3. In this residence increasing the depth of the microreactor leads to decreasing the MB
time, a concentration gradient is formed in the x-direction, hence the degradation. This can be explained by a lower mass transfer limitation
reactant molecules diffuse to the catalytic film, but the residence time is in microreactor with lower depth. To confirm this aspect, the average
not enough to allow the conversion of all the reactant molecules. mass transfer coefficients in microreactors with the depth of 0.1 and 0.2
mm are calculated using simulation results.

38
Rabanimehr et al. / Journal of Applied Research in Water and Wastewater 8 (1) 36-40

0.36 1
MB degradation (X)

MB degradation (X)
0.8
0.34

0.6

0.32
0.4

0.3 0.2
0.1 0.2 0.3 0.4 0.5 0.6 45 90 135 180 225 270 315
Depth of microreactor, mm Residence time, s

Fig. 6. Effect of microreactor depth on MB degradation, C0=0.03 Fig. 7. Effect of MB inlet concentration on MB degradation at different
mol/m3 and flow rate= 400 µl/min. residence times, ♦ symbol shows C0=0.01 mol/m3, ■ symbol shows
C0=0.03 mol/m3, ▲ symbol shows C0=0.05 mol/m3 and × symbol
The MB mass flux toward the photocatalytic surface is equal to the
shows C0=0.07 mol/m3.
rate of reaction. Therefore, the mass transfer coefficient in a horizontal
cross section of the microreactor can be calculated using Eq. (9): 0.6
ra =km (c̅MB -cMBs ) (9)
0.5
where, km is mass transfer coefficient in a horizontal cross section of the
MB degradation (X)
microreactor, c̅ MB is the average MB concentration in a horizontal cross
section of the microreactor and cMBs is the MB concentration at the 0.4
catalytic surface of the microreactor. The average mass transfer
coefficient km is considered to be the arithmetical average of mass 0.3
transfer coefficients at different horizontal cross sections of the
microreactor.
For microreactor with depth of 0.1 mm and flow rate of 400 µl/min, 0.2
km is determined 1.36×10-5 m/s and when the depth of microreactor is
0.2 mm at the same flow rate it is determined 7.5×10-6 m/s. Therefore, 0.1
it can be deduced that increasing the depth of microreactor leads to
decreasing mass transfer coefficient and decreases MB degradation.
To investigate the effect of mass transfer limitation on the kinetic of MB
0
removal, Damköhler number is calculated. Damköhler number 1 2 3 4
determines that the kinetic of reaction is mass transfer limited or not
and is the ratio of the heterogeneous reaction rate at the microreactor Number of cycles
wall to radial diffusion of reactant molecules from bulk towards the wall. Fig. 8. Effect of flow rate on MB degradation, C0=0.03 mol/m3, ♦
For photocatalytic reactions that obey the Langmuir–Hinshelwood symbol shows flow rate= 200 µl/min, ■ symbol shows flow rate= 400
kinetics, Damköhler number, α is defined as: µl/min, ▲ symbol shows flow rate= 800 µl/min.
k
α= kmLHakm (10) 4. Conclusions
+
K C0

For α less than 0.1, reaction is not mass transfer limited (Charls et al. A finite element model was developed using COMSOL Multiphysics
2011). The calculated α for microreactor with the depth of 0.1 and 0.2 5.3 software to simulate photodegradation of MB in a planar
mm are 0.117 and 0.64, respectively. It can be deduced that microreactor with integrated ZnO NWs as photocatalyst. The
microreactor with the depth of 0.1 mm is not mass transfer limited and Langmuir–Hinshelwood kinetic constants kLHa and K determined as
with increasing the depth to 0.2 mm the mass transfer limitation comes 1.43×10-7 mol/m2.s and 7.5 m3/mol, respectively. The model predicted
in to existence. Hence, in the microreactor with the depth of 0.1 mm, the average concentration of MB at the outlet of the microreactor and
the intrinsic kinetics can be extracted without computing the mass showed a very good representation of published experimental data. The
transfer effects. percent of average absolute deviation of the model in predicting the
experimental data was about 0.12% mol/m3 and the deviation between
3.5.2. Effect of MB inlet concentration on MB degradation the model and the experimental data points was low at all residence
times. The simulation results showed that increasing the residence time
The effect of MB concentration at the inlet of the microreactor on causes decreasing the MB concentration at the outlet of the
the degradation is investigated at different residence times. As microreactor that corresponds to the longer contact time of MB with
illustrated in Fig. 7, in lower amounts of initial concentration, the photocatalyst. The photocatalytic degradation of MB was simulated in
microreactor has better efficiency. Because in lower amounts of inlet different microreactor geometric configurations and it was found that at
concentration, there are more active sites of photocatalyst in contact constant residence time, microreactor with shallower depth has better
with MB molecules. As shown in the figure, with increasing the photocatalytic performance because increasing the depth of the
residence time, due to the occupation of active sites by MB molecules, microreactor, leads to higher mass transfer limitation. The simulation
the changes in MB degradation decreases and the curves become results showed that the microreactor has the highest efficiency in the
plateau. lowest amounts of inlet MB concentration and flow rate. Thiele modulus
and Damköhler numbers were estimated and showed that the
3.5.3. Effect of flow rate on MB degradation investigated microreactor performance is not mass transfer limited.

Fig. 8 illustrates the effect of flow rate on the MB degradation in four References
cycles. As shown in the figure, the microreactor with the lowest flow rate
has the highest value of MB degradation. This corresponds to longer Aillet T., Loubie`re K., Prat L., Impact of the Diffusion limitation in
residence time of MB solution at lower flow rates. microphotoreactors, American Institute of Chemical Engineers
Journal, 61 (2015) 1284-1299.

39
Rabanimehr et al. / Journal of Applied Research in Water and Wastewater 8 (1) 36-40

Ali I., Al-Othman Z.A., Alwarthan A., Synthesis of composite iron nano Lee K.M., Lai C.W., Koh S.N., Juan J.C., Recent developments of zinc
adsorbent and removal of ibuprofen drug residue from water, Journal oxide based photocatalyst in water treatment technology: A review,
of Molecular Liquids 219 (2016) 858-864. Water Research 88 (2016) 428-448.
Charles G., Roques-Carmes T., Becheikh N., Falk L., Commenge J.M., Lei L., Wang N., Zhang X.M., Tai Q., Tsai D.P., Chan H.L.W.,
Corbel S., Determination of kinetic constants of a photocatalytic Optofluidic planar reactors for photocatalytic water treatment using
reaction in microchannel reactors in the presence of mass-transfer solar energy, Biomicrofluidics 4 (2010) 1-12.
limitation and axial dispersion, Journal of Photochemistry and
Liao W., Wang N., Wang T., Xu J., Han X., Liu Zh., Zhang X., Yu W.,
Photobiology A: Chemistry 223 (2011) 202-211.
Biomimetic microchannels of planar reactors for optimized
Corbel S., Becheikh N., Roques-Carmes T., Zahraa O., Mass transfer photocatalytic efficiency of water purification, Biomicrofluidics 10
measurements and modeling in a microchannel photocatalytic (2016) 1-8.
reactor, Chemical Engineering Research and Design 92 (2014) 657-
662. Meng Z., and Zhang XQin., A high efficiency microfluidic-based
photocatalytic microreactor using electrospun nanofibrous TiO 2 as a
Dijkstra M.F.J., Panneman H.J., Winkelman J.G.M., Kelly J.J., photocatalyst, Nanoscale 5 (2013) 4687-4690.
Beenackers A., Modeling the photocatalytic degradation of formic
acid in a reactor with immobilized catalyst, Chemical Engineering Mills A., and Wang J.S., Photomineralisation of 4-chlorophenol
Science 57 (2002) 4895-4907. sensitised by TiO2 thin films, Journal of Photochemistry &
Photobiology. A Chemistry118 (1998) 53-63.
Gorges R., Meyer S., Kreisel G., Photocatalysis in microreactors,
Journal of Photochemistry and Photobiology A: Chemistry 167 (2004) Ould-Mame S.M., Zahraa O., Bouchy M., Photocatalytic degradation of
95-99. salicylic acid on fixed TiO2- kinetic studies, International Journal of
Photoenergy 2 (2000) 59-66.
Han Zh., Li J., He W., Li S., Li Zh., Chu J., Chen Y., A microfluidic device
with integrated ZnO nanowires for photodegradation studies of Ramos B., Ookawara S., Matsushita Y., Yoshikawa S., Low-cost
methylene blue under different conditions, Microelectronic polymeric photocatalytic microreactors: Catalyst deposition and
Engineering 111 (2013) 199-203. performance for phenol degradation, Journal of Environmental
Chemical Engineering 2 (2014) 1487-1494.
He Z., Li Y., Zhang Q., Wang H., Capillary microchannel based
microreactors with highly durable ZnO/TiO2 nanorod arrays for rapid Wang Y., Ma J., Zhu J., Ye N., Zhang X., Huang, H., Multi-walled carbon
high efficiency and continues flow photocatalysis, Applied Catalysis nanotubes with selected properties for dynamic filtration of
B: Environmental 93 (2010) 376-382. pharmaceuticals and personal care products, Water Research 92
(2016) 104-112.
Helen L., Microfluidic reactors for photocatalytic water purification, Lab
on a Chip, 14 (2014) 1074-1082. Wang N., Zhang X., Chen B., Song W., Chan N.Y., Chan H.L.,
Microfluidic photo-electrocatalytic reactors for water purification with
Jayamohana H., Smith Y.R., Hansen L.C., Mohanty S.K., Gale B.K., an integrated visible-light source, Lab on a Chip, 12 (2012) 3983-
Misra M., Anodized titania nanotube array microfluidic device for 3990.
photocatalytic application: Experiment and simulation, Applied
Catalysis B: Environmental 174-175 (2015) 167-175. Yusuf A., Oladipo H., Yildiz Ozer L., Garlisi C., Loddo V., Abu-Zahra M.
R.M., Palmisano G., Modelling of a recirculating photocatalytic
Jewell K.S., Falas P., Wick A., Joss A., Ternes T.A., Transformation of microreactor implementing mesoporous N-TiO2 modified with
diclofenac in hybrid biofilm activated sludge processes, Water graphene, Chemical Engineering Journal 391 (2020) 123574-1-33.
Research 105 (2016) 559-567.
Zhang Q., Wang H., Li Y., A high efficiency microreactor with Pt/ZnO
Krivec M., Zagar K., Suhadolnik L., eh M., Drazi G., A highly efficient nanorod arrays on the inner wall for photodegradation of phenol,
TiO2-based microreactor for photocatalytic applications, Applied Journal of Hazardous Materials 15 (2013) 318-324.
Material Interfaces 5 (2013) 9088-9094.
Kuo T., Lin Ch., Kuo C., Huang M.H., Growth of ultralong ZnO
nanowires on silicon substrates by vapor transport and their use as
recyclable photocatalysts, Chemistry of Material 19 (2007) 5143-
5147.

40