CHEMICAL ENGINEERING TRANSACTIONS  
 

VOL. 57, 2017 

A publication of 

 
The Italian Association 

of Chemical Engineering 
Online at www.aidic.it/cet 

Guest Editors: Sauro Pierucci, Jiří Jaromír Klemeš, Laura Piazza, Serafim Bakalis 
Copyright © 2017, AIDIC Servizi S.r.l. 

ISBN 978-88-95608- 48-8; ISSN 2283-9216 

Enhanced Performances of a Photocatalytic Reactor for 
Wastewater Treatment Using Controlled Modulation of LEDs 

Light 
Vincenzo Vaiano*a,Olga Saccoa, Diana Sanninoa, Giulia Di Capuab, Nicola Femiab 
a Department of Industrial Engineering, University of Salerno, Via Giovanni Paolo II 132, 84084 Fisciano (SA), Italy. 
b Department of Information and Electrical Engineering and Applied Mathematics, University of Salerno, Via Giovanni Paolo 
II 132, 84084 Fisciano (SA), Italy  
vvaiano@unisa.it 

One of the main limitations related to the industrial utilization of a photocatalytic process is the use of UV 
lamps that have several disadvantages such as the low quantum efficiency (QE). One of the approaches to 
increase the QE consists in the use of controlled periodic illumination (CPI). In this regard, Light Emitting 
Diodes (LEDs) are the best choice as light sources, since they can be electronically controlled by using LED 
dimming techniques, allowing variable turn-on and turn-off times on a microsecond time-scale.  
The aim of this work is to investigate the influence of controlled modulation of LEDs light on the performances 
of a photocatalytic reactor for wastewater treatment using TiO2 immobilized on glass spheres (TiO2/GS).  
The experiments were realized with a pyrex cylindrical batch photoreactor irradiated by three power UV-LEDs 
and placed surrounding the external body of the cylindrical photoreactor.  
The system adopted for the controlled modulation of LEDs light is composed by a photovoltaic panel (PV), a 
dc–dc converter dedicated to the maximum power point tracking of the PV panel and a dc–dc converter 
dedicated to drive the UV LEDs. A system controller is also included, whose goal is to ensure matching 
between the maximum available PV power, the LED power and the resulting low-frequency LEDs dimming 
modulation. The experimental results showed that the use of certain types of duty-cycle dimming modulation 
waveforms, including periodic and non-periodic waveforms and combinations of them, have been discovered 
to improve the photo-catalytic reactor performances. In particular, the use of a modulation of LEDs dimming 
like sinusoidal and pseudo-sinusoidal waveforms is more effective than the fixed dimming. The best results in 
terms of methylene blue degradation (about 25% in less than 1 h) have been achieved with a pseudo-
sinusoidal waveform. 

1. Introduction 

The main causes of surface and groundwater contamination are industrial effluents, excessive use of 
pesticides, fertilizers (agrochemicals) and domestic waste landfills. The traditional wastewater treatment is 
usually based on physical and biological processes. These processes have the limitation to transfer the 
organic pollutants from one phase to another phase leading to the need of further treatment. Advanced 
Oxidation Processes (AOPs) may become the most widely used water treatment technologies for organic 
pollutants (Vaiano et al., 2016) not treatable by conventional techniques (Sacco et al., 2015) due to their high 
chemical stability and/or low biodegradability (Pera-Titus et al., 2004). 
TiO2 is a well-known effective semiconductor photocatalyst for the purification of water and air due to its high 
photocatalytic activity, chemical as well as biological stability, relatively low-cost and especially non-toxicity 
(Lee and Park, 2013). However, photocatalysts are often used in slurry reactors (solid-liquid-gas system) and 
the treatment of industrial effluent with the TiO2 based photocatalytic reaction in dispersion medium needs 
separation of catalyst and the recovery of the ultrafine catalyst from the treated liquid. Moreover the use of 
photocatalysts in powder form causes damages to the recirculation pumps (expensive). This disadvantage is 
overcame realizing heterogeneous fixed bed reactors, in which the photocatalyst is supported in granular form 

                               
 
 

 

 
   

                                                 
DOI: 10.3303/CET1757093

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Please cite this article as: Vaiano V., Sacco O., Sannino D., Di Capua G., Femia N., 2017, Enhanced performances of a photocatalytic reactor 
for wastewater treatment using controlled modulation of leds light, Chemical Engineering Transactions, 57, 553-558   
DOI: 10.3303/CET1757093 

553



(Vaiano et al., 2014a) or thin film on transparent support (Vaiano et al., 2015b). However, one of the main 
limitations related to the industrial utilization of a photocatalytic reactor is the use of UV lamps that have 
several disadvantages such as low quantum efficiency (QE) (Wang and Ku, 2006). One of the approaches to 
increase the QE consists in the use of controlled periodic illumination (CPI). In this regard, Light Emitting 
Diodes (LEDs) are the best choice as light sources, since they can be electronically controlled by using LED 
dimming techniques, allowing variable turn-on and turn-off times on a microsecond time-scale. LEDs sizes are 
much smaller that UV lamps, allowing a spatial arrangement that guarantees a highly uniform illumination, with 
a reduced waste of light energy. Moreover, thanks to their controllable-energy generation, LEDs can realize 
operating temperatures lower than those of traditional UV lamps, with a consequent reduction of cooling and 
electrical energy costs (Mottier, 2010). The additional use of a renewable energy source, such as a 
photovoltaic (PV) panel feeding the LEDs in a light-to-light converter (Femia et al., 2013), and of an electric 
storage unit to stabilize the renewable energy production (Femia and Zamboni, 2012), opens perspectives of 
further innovation in LED-lighting applications.  
The aim of this work is to investigate the influence of controlled modulation of LEDs light on the performances 
of a photocatalytic reactor for wastewater treatment using TiO2 immobilized on glass spheres.  

2. Experimental: preparation of photocatalyst and photocatalytic activity tests 

2.1 Sol–gel synthesis 

The sol-gel synthesis was carried out following the method reported by Vaiano et al. (Vaiano et al., 2015b). In 
particular Triton X-100 (nonionic surfactant, Sigma–Aldrich) has been used as binder (Vaiano et al., 2015b). 
Triton X-100 was dissolved in isopropyl alcohol (i-PrOH, 99.8 wt %, Sigma–Aldrich) and the pH of solution was 
adjusted with sulfuric acid (H2SO4, 95-98 wt %, Carlo Erba) until to reach a value of about 3. Then, titanium 
(IV) isopropoxide (TTIP, 97%, Sigma–Aldrich), used as titania precursor, was added to the mixture. The molar 
ratio of the ingredients was Triton X-100:i-PrOH:TTIP:sulfuric acid = 1:45:1:1. The obtained solution was 
stirred for 24 h at room temperature. After this time, the solution was slightly yellow, and appeared 
homogeneous and stable. The final solution was used for the immobilization of TiO2 on glass spheres. 

2.2 Immobilization of TiO2 on glass spheres 

Pyrex spheres (dp=5.3 mm, from Microglass Heim) were used as glass substrate for the immobilization of 
TiO2, which has been performed through dip-coating technique.  
Before dip-coating, the whole surface of glass spheres was rinsed with MilliQ grade water and calcined at 
450°C for 30 minutes. TiO2 coating was realized by immersing the glass spheres in the solution prepared as 
reported in the section 2.1. The glass spheres were maintained in the solution for 30 minutes and then 
calcined for 30 minutes at 450°C. The dip-coating and calcination processes were repeated four times (Vaiano 
et al., 2015b). The final structured catalyst is named TiO2/GS. The TiO2 amount coated on glass spheres has 
been measured using precision balance (Mettler Toledo). 

2.3 Photocatalyst characterization 

TiO2/GS catalyst was characterized by several techniques. Specific surface area (BET) was evaluated from 
dynamic N2 adsorption measurement at -196°C, performed by a Costech Sorptometer 1040 after pretreatment 
at 150°C for 30 min in He flow. The Raman spectra were recorded with a Dispersive MicroRaman system 
(Invia, Renishaw), equipped with 514 nm diode-laser, in the range 100-900 cm-1 Raman shift. Total TiO2 
content on glass spheres was determined by X-ray fluorescence spectrometry (XRF) in a thermoFischer ARL 
QUANT’X EDXRF spectrometer equipped with a rhodium standard tube as the source of radiation and with Si-
Li drifted crystal detector. 

2.4 Photocatalytic activity tests 

Methylene blue (MB) was used as model dye. The experiments were carried out with initial concentration of 
MB equal to 5 mg L-1, at ambient temperature and pressure.  
The experiments were realized with a pyrex cylindrical batch photoreactor (ID = 2.5 cm; height=18 cm) 
(Vaiano et al., 2014b) equipped with an air distributor device (Qair = 150 cm

3 min-1 under standard temperature 
and pressure conditions) to assure the presence of oxygen in the liquid medium. The amount TiO2/GS catalyst 
used during the photocatalytic test was equal to 60 g in a total solution volume of 80 mL.  
The continuous mixing of the aqueous sample was realized by external recirculation of the aqueous solution 
through the use of a peristaltic pump. Thermocouple was inserted inside the reactor to monitor the 
temperature during irradiation. The photoreactor was irradiated with three power UV LEDs (5 W of nominal 
maximum power, 700 mA of nominal current and emission spectrum centred at 400 nm). The UV LEDs are 
placed surrounding the external body of the cylindrical photoreactor. The schematic picture of experimental 

554



set-up is reported in a previous paper (Vaiano et al., 2014b). The system was left in dark until reaching MB 
adsorption equilibrium, and then photocatalytic reaction was initiated under UV light. Liquid samples were 
analyzed continuously by spectrophotometric measurement. In particular, a special assembly with a flow 
quartz cuvette and an external pump for the recirculation of liquid was used, permitting to determine the 
change of MB measured at λ = 663 nm.  
The electronic system adopted for the experimental tests is composed by a PV panel, a dc–dc boost converter 
dedicated to the maximum power point tracking of the PV panel and a dc–dc buck converter dedicated to drive 
the LEDs array. A system digital controller is also included, whose goal is to ensure matching between the 
maximum available PV power, the LED power and the resulting low-frequency LEDs dimming modulation.  
The average LEDs current (Iavg) depends on the on-state LED current (If) and on the dimming duty-cycle ddim, 
defined as the ratio between the time during which the buck converter is in operation feeding the LEDs array 
(ton) and the total duration of the dimming period (Tdim), according to Eq(1) and Eq(2):  

favg
IdI 

dim
   (1) 

dim

dim

on
t

d
T

    (2) 

The average LEDs current was modified according PWM approach, which consists in modulating the duration 
of current pulses at fixed on-state LED current. 
In this work, PWM technique have been investigated and tested. Such modulation approaches have been 
performed both at fixed (100 %) and at variable dimming duty-cycle. In particular, two kind of variable dimming 
duty-cycle have been herein considered, based on a sinusoidal and on a pseudosinusoidal modulation of the 
duty. 
In the case of sinusoidal dimming, the duty-cycle is modulated according to Eq(3): 

 )(1
2

)( minmax
mindim

tsen
DD

Dtd 


     (3) 

where Dmin and Dmax are the minimum and maximum dimming duty-cycle values. In the case of pseudo-
sinusoidal dimming, the duty-cycle is modulated according to Eq(4): 

 ))((1
2

)( minmax
mindim

ttsen
DD

Dtd 


     (4) 

where the angular frequency ω(t) is not a constant value, but varies in a certain range, as shown in Figure 1 
(c). The sinusoidal and pseudosinusoidal modulations of the dimming duty cycle used in this work are reported 
in Figure 1. 

 

Figure 1: Behaviour of the dimming duty-cycle as a function of run time: (a) sinusoidal modulation with 

constant amplitude and constant frequency; (b) pseudo-sinusoidal modulation with constant frequency and 

variable amplitude; (c) pseudo-sinusoidal modulation with constant amplitude and variable frequency. 

555



3. Results and Discussion 

3.1 Characterization of TiO2/GS photocatalyst 

Specific surface area (SSA) values for TiO2/GS and uncoated glass spheres are reported in Table 1. SSA of 
TiO2/GS was is 0.96 m

2 g-1, while a lower value (0.08 m2 g-1) was found for uncoated glass spheres.  
The increase of surface area of TiO2/GS could be ascribed to the formation of TiO2 coating on the surface of 
glass spheres which results in a porous and rough surface, as previously observed for N-doped TiO2 
immobilized on glass spheres (Vaiano et al., 2015a).  
The total amount of TiO2 effectively deposited was determined by XRF (Table 1) and it was equal to 0.6 wt %. 
The real TiO2 content well agrees to the TiO2 content evaluated with the precision balance, indicating a good 
yield of the dip-coating process under the used preparation conditions. 

Table 1: List of the samples and their characteristics 

Sample SSA,  
m2 g-1 

TiO2 loading 
wt% 
(balance) 

TiO2 
measured 
loading 
wt% (XRF) 

TiO2  
crystallite size  
nm (Raman) 

Glass spheres 0.08 - - - 
TiO2/GS 0.96 0.6 0.57 6 
 
The Raman spectra of the uncoated glass spheres in comparison with TiO2/GS structured catalyst is shown in 
Figure 2. Uncoated glass spheres showed no Raman signals while TiO2/GS sample displayed bands at 144, 
396, 514 637 cm-1 and a weak shoulder at 195 cm-1 due to the anatase form of TiO2 (Sannino et al., 2013) 
meaning that the coating method is able to induce the formation of anatase TiO2 deposited on the surface of 
glass spheres. The TiO2 particle size immobilized on glass spheres has been estimated by Raman spectra 
following the method finding that the value is about 6 nm, accordingly to previous results (Vaiano et al., 
2015b). 

 

Figure 2: Raman spectra of glass spheres and TiO2/GS. 

3.2 Photocatalytic activity results 

Preliminary experiments were carried out in order to verify that MB was degraded by the heterogeneous 
photocatalytic process. In the absence of TiO2/GS photocatalyst, no significant decrease in MB concentration 
was observed during the UV irradiation. 
Figure 3 reports the comparison between the photocatalytic activity of TiO2/GS at fixed (100 %) and at 
variable LEDs dimming duty-cycle (according to the profiles reported in Figure 1). 

556



 

Figure 3: Behaviour of the MB relative concentration as a function of UV irradiation time obtained at fixed 

dimming duty cycle (100%), with sinusoidal modulation with constant amplitude and constant frequency (a), 

with pseudo-sinusoidal modulation with constant frequency and variable amplitude (b) and with pseudo-

sinusoidal modulation with constant amplitude and variable frequency (c).  

The effect of the light intensity modulation on the photocatalytic activity is clearly evident.  
The sinusoidal modulation (a) did not promote the photocatalytic performances with respect to the irradiation 
realized at fixed dimming duty cycle. When the pseudo-sinusoidal modulation changes both in frequency and 
in amplitude, an increase of photocatalytic activity was achieved. The best results in terms of methylene blue 
degradation (about 25% in less than 1 h) have been achieved with a pseudo-sinusoidal modulation with 
constant amplitude and variable frequency (curve c of Figure 3). 
Such PWM techniques provide an optimal trade-off between the photocatalytic performances and the energy 
consumption caused by the lighting system. In fact, variable duty-cycle in PWM dimming techniques may help 
in improving the effectiveness of the photocatalytic process and the energy efficiency of the lighting system, 
even if the average LED current (and relevant power consumption) and the resulting light intensity are lower 
than those ones obtained with fixed duty-cycle based PWM dimming. 

4. Conclusions 

The influence of controlled modulation of LEDs light on the performances of a photocatalytic reactor for 
wastewater treatment using TiO2 immobilized on glass spheres has been studied. The experiments were 
realized with a pyrex cylindrical batch photoreactor irradiated by three power UV-LEDs and placed 
surrounding the external body of the cylindrical photoreactor. The system adopted for the controlled 
modulation of LEDs light is composed by a photovoltaic (PV) panel, a dc–dc converter dedicated to the 
maximum power point tracking of the PV panel and a dc–dc converter dedicated to drive the UV LEDs.  
Photo-catalytic reactor performances has been improved by means of the time domain modulation of the 
square-wave LEDs light dimming, realized by means of the duty-cycle modulation waveforms. 
The use of certain types of duty-cycle dimming modulation waveforms, including periodic and non periodic 
waveforms and combinations of them, have been discovered to improve the photocatalytic reactor 
performances. In particular, the duty-cycle has been run-time modulated during the experimental tests 
allowing to achieve superior photocatalytic performances. In particular, the best results in terms of methylene 
blue degradation (about 25% in less than 1 h) have been achieved with a sinusoidal modulation with constant 
amplitude and variable frequency. 
 
 

557



References  

Femia, N., Fortunato, M. & Vitelli, M. 2013. Light-to-Light: PV-Fed LED Lighting Systems. IEEE Transactions 
on Power Electronics, 28, 4063-4073. 

Femia, N. & Zamboni, W. Photovoltaic-fed LED lighting system with SOC-based dimmable LED load.  IECON 
2012 - 38th Annual Conference on IEEE Industrial Electronics Society, 25-28 Oct. 2012 2012. 1132-1137. 

Lee, S.-Y. & Park, S.-J. 2013. TiO2 photocatalyst for water treatment applications. Journal of Industrial and 
Engineering Chemistry, 19, 1761-1769. 

Mottier, P. 2010. Frontmatter. LEDs for Lighting Applications. ISTE. 
Pera-Titus, M., Garcı́a-Molina, V., Baños, M. A., Giménez, J. & Esplugas, S. 2004. Degradation of 

chlorophenols by means of advanced oxidation processes: a general review. Applied Catalysis B: 
Environmental, 47, 219-256. 

Sacco, O., Vaiano, V., Han, C., Sannino, D., Dionysiou, D. D. & Ciambelli, P. 2015. Long afterglow green 
phosphors functionalized with Fe-N doped TiO2 for the photocatalytic removal of emerging contaminants. 
Chemical Engineering Transactions. 

Sannino, D., Vaiano, V., Sarno, G. & Ciambelli, P. 2013. Smart tiles for the preservation of indoor air quality. 
Chemical Engineering Transactions, 32, 355-360. 

Vaiano, V., Sacco, O., Pisano, D., Sannino, D. & Ciambelli, P. 2015a. From the design to the development of 
a continuous fixed bed photoreactor for photocatalytic degradation of organic pollutants in wastewater. 
Chem. Eng. Sci., 137, 152-160. 

Vaiano, V., Sacco, O., Sannino, D. & Ciambelli, P. 2015b. Nanostructured N-doped TiO2 coated on glass 
spheres for the photocatalytic removal of organic dyes under UV or visible light irradiation. Applied 
Catalysis B: Environmental, 170-171, 153-161. 

Vaiano, V., Sacco, O., Sannino, D., Ciambelli, P., Longo, S., Venditto, V. & Guerra, G. 2014a. N-doped 
TiO2/s-PS aerogels for photocatalytic degradation of organic dyes in wastewater under visible light 
irradiation. Journal of Chemical Technology and Biotechnology, 89, 1175-1181. 

Vaiano, V., Sacco, O., Sannino, D., Stoller, M., Ciambelli, P. & Chianese, A. 2016. Photocatalytic removal of 
phenol by ferromagnetic N-TiO2/SiO2/Fe3O4 nanoparticles in presence of visible light irradiation. Chemical 
Engineering Transactions. 

Vaiano, V., Sacco, O., Stoller, M., Chianese, A., Ciambelli, P. & Sannino, D. 2014b. Influence of the 
photoreactor configuration and of different light sources in the photocatalytic treatment of highly polluted 
wastewater. Int. J. Chem. React. Eng., 12, 1-13, 13. 

Wang, W.-Y. & Ku, Y. 2006. Photocatalytic degradation of Reactive Red 22 in aqueous solution by UV-LED 
radiation. Water Research, 40, 2249-2258. 

 

558