ABSTRACT
TiO2 nanoparticles have been synthesized by solvent-free hydrothermal process. TiO2 nanoparticles were annealed at 500°C for enhancing the characterization and the photocatalytic activity. The synthesized TiO2 was characterized by x-ray diffraction (XRD), field emission scanning electron microscopy (FE-SEM), atomic force microscopy (AFM), BET, and diffuse reflectance spectroscopy (UV-DR) techniques to study the morphology and structural configuration. The effects of different parameters such as the initial dye concentration, catalyst concentration, pH of the solution, light intensity, and reactive oxygen species (ROS) on relative photonic efficiencies and photocatalytic degradation kinetics of GRL were investigated, and the degradation of GRL follows pseudo-first order kinetics according to the Langmuir–Hinshelwood model. The ROS studies indicate that hydroxyl radicals and holes are the predominant reactive species within the same step, contributing up to 92.64%, hydroxyl radicals participate for about 55%, and holes share for about 37.64% in the photocatalytic degradation of GRL.

KEYWORDS
Maxilon blue dye; photocatalytic removal; photonic efficiency; reactive oxygen species; titanium dioxide; UVA-LED





Introduction

Undoubtedly, today, the environmental pollutants are considered as important problems in the human society. Water pollution requires main solutions because it is one of the most unfavorable environmental problems in the world. Textile industries produce a lot of wastewater, which contains a num- ber of undesirable compounds, including caustic dissolved solids or acidic and poisonous compounds. A lot of organic dyes (synthetic dyes) are hazardous and may do affect living organisms life causing different diseases (Vieira et al. 2009).
There are many ways for pollutant elimination such as adsorption on activated carbon, reverse osmosis, ultrafiltra- tion, etc.(Tang and An 1995; Simin 2014; Aljeboree, Alkaim, and Al-Dujaili 2015; Fil 2015). These methods generally cause the transfer of the organic pollutants from water to other media that naturally produces a new pollution. In recent dec- ades, a growing need for green technology has facilitated the development of alternative technologies that avoid the con- sumption of excessive chemicals and minimize the generation of toxic sludge after treatment(Zhou and Smith 2002). In this regard, advanced oxidation processes (AOPs) such as ozone, Fentons oxidation, ultrasound, electro-oxidation, and photo- catalytic degradation have been identified as one of the most promising options for this purpose through the formation of OH radicals (Wang et al. 2010; Bayar et al. 2014; F?l et al. 2014; Alkaim, Dillert, and Bahnemann 2015; Islam, Kurny, and Gulshan 2015; Kul et al. 2015).
Photocatalytic degradation is an important process for wastewater treatment. In particular, this method is notably useful for purification of wastewater containing organic

contaminants (Alkaim and Hussein 2012; Kandiel et al. 2013; Da Dalt et al. 2015; Sobana, Krishnakumar, and Swaminathan 2013). This technique has certain advantages over other processes, such as a small amount of by-products, complete mineralization, cost-effectiveness, and applicability at moderate temperatures (Sun et al. 2006; El-Mekkawi and Galal 2013). Among many proposed semiconductors for photocatalytic treatment, titanium dioxide is a suitable photocatalyst because of its acceptable band gap energy, easy availability, and low cost (Thiruvenkatachari, Vigneswaran, and Moon 2008). Such photocatalysts apply UV light to gen- erate electron–hole pairs at their conduction and valence bands (Gaya and Abdullah 2008). Then the electrons in the conduction band react with molecular oxygen in bulk solution to generate active oxidant species such as superoxide radical anions and hydrogen peroxide. On the other hand, the holes at valence band can oxidize surface hydroxyl groups to form OH• radicals or even organic pollutant molecules (Kansal et al. 2014). The mentioned oxidant species attack the organic pollutant targets, leading eventually to oxidation of them to CO2, H2O, etc.
Large amount of dyes are annually used in different kinds
of industries such as textile, cosmetics, food, pharmaceutical, and paper industries (Pitchaimuthu et al. 2014). The largest and the most important classes of commercial dyes have been found in azo dyes constitute.
The removal of azo dyes is an important process, because
many azo dyes are toxic to aquatic organisms(Chen 2009); therefore, in this work we choose a maxilon blue (GRL) as a model of azo-dye pollutant(Aljebori and Alshirifi 2012)


CONTACT Ayad F. Alkaim alkaim@iftc.uni-hannover.de Department of Chemistry, College of Sciences for Women, Babylon University, Hilla, Iraq. Color versions of one or more of the figures in the article can be found online at www.tandfonline.com/upst
© 2016 Taylor & Francis



and to investigate the photocatalytic degradation of synthe- sized TiO2 nanoparticles. The effect of operational parameters such as catalyst concentration, dye initial concentration, UVA light intensity, and pH of the solution on the relative photonic efficiency of synthesis TiO2 nanoparticles photocatalysis as well as the role of reactive oxygen species (ROS) on the photo-removal and relative photonic efficiency trends in the photocatalytic degradation of maxilon blue dye GRL has also been investigated.


Materials and methods

Materials

Titanium(IV) bis(ammoniumlactato) dihydroxide (TALH, 50% aqueous solution), aqueous ammonia solution (28.0– 30.0% NH3), sodium hydroxide (98%), nitric acid (70%), ethylenediaminetetraacetic acid disodium salt dehydrate (Sigma-Aldrich, Germany), 2-propanol anhydrous, (99.5% Sigma-Aldrich), potassium iodide anhydrous (99.5%, Sigma- Aldrich), nitric acid (70%, Sigma-Aldrich), N-phenylaniline (Merck Millipore), maxilon blue dye (GRL) were obtained from Al-Hilla of textile industries/Iraq. All the chemicals were used as received, without further purification.

done at room temperature (30°C). The pH of dye solutions was adjusted by adding 0.05 M HNO3 or NaOH.
The reaction was initiated when LED/UVA (365 nm-
Collimated LED for Olympus BX & IX, 700 mA Thorlab/ USA) interval light bulbs (measured by UVA radiometer, Dr Honle/Germany) were switched on. Samples were taken out at specific time intervals and centrifuged at 3500 rpm. The remaining concentration of GRL in solution was analyzed by UV–vis spectrophotometer (UV 1650 spectrometer, Shimadzu) at its maximum wavelength of 609 nm.
An estimate of the contribution of hydroxyl radicals for OH•, oxygen superoxide radical O—?, and positive holes (h?) during the heterogeneous photocatalytic degradation of GRL dye are determined by adding different concentrations
of ROS: isopropyl alcohol (IPA), diphenylamine (DPA), potassium iodide (KI), and ethylenediaminetetraacetic acid disodium salt dehydrate (Na2EDTA) were added in different concentrations (0.005, 0.01, 0,05, and 1 mM) to the reaction solution.
The relative photonic efficiency, photocatalytic degradation efficiency, and apparent first-order rate constant of photocata- lytic degradation of GRL dye were calculated using the follow- ing relationships:
R ? V

e ¼
o

? A ?1?

Methods
Pure anatase TiO2 nanoparticles with 97 m2 g—1 have been

Io ¼

I ? k
NA ? h ? c

?2?

synthesized by the thermal hydrolysis of titanium(IV) bis

PDE?%? ¼ 100 ? ?C0 — Ct ?=C0 ?3?

(ammoniumlactato) dihydroxide (TALH) as previously reported with annealing modifications (Alkaim et al. 2013).

Ct
ln ¼ —
0

kt ?4?

Typically, 10 mL of an aqueous TALH solution was mixed with an aqueous ammonia solution. The resulting solution volume was 100 mL and the concentration of aqueous ammonia was adjusted to be 1.0 M which was mixed with con- stant stirring at room temperature. Subsequently, the resultant solution was transferred into a Teflon-lined stainless steel autoclave with a volume of 250 mL. Then, the Teflon cup was sealed in an autoclave and placed into an electric furnace held at 160°C for 24 h.
After the growth, it was allowed to cool down naturally to
room temperature; the resulting TiO2 nanoparticles were collected, washed with water/ethanol different times with ultrasonic assistance, then dried at 80°C for 24 h, and finally annealed at 500°C for 4 h. The synthesized sample was charac- terized by XRD, the Brunauer—Emmett—Teller (BET) surface area, UV—Vis, diffused reflectance spectra (DRS), field- emission scanning electron microscopy (FE-SEM), and NTMDT Solver (P47-PRO) scanning probe microscope oper- ating in the contact atomic force microscopy (AFM) mode with a scan speed of 1 Hz.
The photocatalytic degradation experiments were carried out in a batch photoreactor, and reaction suspensions (200 mL) containing a suitable amount of TiO2 photocatalyst and GRL dye solution were kept in the dark for 60 min, in the presence of oxygen bubbling in order to reach adsorp- tion–desorption equilibrium. The reactor contained a stirring rod supported by a magnetic stirrer to confirm homogeneity of the mixture throughout the reactor; all the experiments were

where R is the rate of photocatalytic degradation (mg.L—1. min—1), e is the relative photonic efficiency, V is volume of irradiated solution (L), A is the irradiation area of photocata- lytic reaction, I is the incident light of irradiation source (mW cm—2), k is the wavelength of irradiation source (m), NA is the Avogadro’s number, h is the Planck constant, c
is the speed of light in space, C0 and Ct are the initial and photolyzed concentration (mg L—1), respectively, PDE (%) is
photocatalytic degradation efficiency, t is time of irradiation (min), and k is the apparent first-order rate constant (min—1).


Results and discussion

Characterization of TiO2 photocatalysts

The XRD pattern of annealed TiO2 is illustrated in Figure 1(a). The intensified diffraction peaks of annealed TiO2 at 2h ¼ 25.56, 38.02, 48.3, 54.14, 55.27, 62.69, 68.9, 70.52, and
75.22 were assigned to a pure anatase phase(Gnanaprakasam et al. 2015). No Rutil or Brookite peak was observed, clearly depicting the crystallization of synthesized TiO2. The crystal- line size of the annealed TiO2 was examined by using Scherrer’s equation, D ¼ 0.9k=bcosh, where b is the full width at half maximum intensity, k is the wavelength of Cu-k? radiation, and h is the angle obtained from the 2h value
corresponding to the (101) peak attributed to the anatase TiO2. According to Scherrer’s equation, the calculated average crystalline size was found to be ? 11 nm.





Figure 1. Typical (a) XRD pattern, (b) UV-vis diffuse reflectance (Tauc plot for absorption edge determination in the inset), (c) FE-SEM, and (d) AFM images for the calcined TiO2 nanoparticles.



The UV–vis diffuse reflectance spectra of synthesized TiO2 nanoparticles in the range of 200–800 nm are shown in Figure 1(b), assuming that TiO2 has an indirect optical transition (Sakthivel, Janczarek, and Kisch 2004); the calculated band gap energy for synthesized TiO2 is 3.223 eV.
The surface morphology and texture of catalyst are one of
the important parameters that might influence the photocata- lytic efficiency, and the annealed TiO2 nanoparticles were investigated by FE-SEM image as shown in Figure 1(c). The SEM image depicts that the particles are in the form of aggre- gates and the surface of the annealed TiO2 is irregular with the size range of 12–20 nm or even smaller.
Figure 1(d) shows typical AFM images of the TiO2 film. It can be seen that this film is quite uniform. The particle size distributions of TiO2 film in AFM image was about 55–82 nm, much greater than the crystallite sizes studied from the XRD. The aggregation of the primary crystallites caused an increase in the crystallite sizes of TiO2 particles measured in AFM results.


UV-visible spectra of GRL photocatalytic reaction

Figure 2 displays the changes in the UV–vis adsorption spectra of 15 mg L—1 GRL neutral solution exposed to the UVA for different time in the presence of 2 g L—1 TiO2.
The spectrum of GRL shows that the intensity of the peaks at 609 nm (kmax) decreases gradually during the UVA irradiation, resulting in decolorization of the solutions. The diminished absorption intensity of the kmax also expresses the loss of conjugation, e.g., especially the cleavage nears the azo bond from the —N ¼ N double bond as the most active site

for the oxidation attack (Sun et al. 2002). The nearly perfect disappearance of the band at 609 nm reveals that GRL is elimi- nated after about 60 min.
This is accompanied by a parallel decrease of the intensity of the peak in the UV region, 296 nm, attributed to the ben- zothioazol ring. No new absorption peaks appeared in either visible or UV regions.
Control experiments (figure not shown) proved that the decolorization of GRL was not feasible in the absence of either UVA light or photocatalyst TiO2. On the other hand, the use






Figure 2. UV–vis spectra changes of GRL dye at different times during adsorp- tion and photocatalysis process; experimental conditions: Catalyst loading 2 g L—1, pH ¼ 6.55, GRL concentration 15 mg • L—1, light intensity 33.5 mW cm—2, irradiation time 60 min. (inset shows the chemical structure of GRL dye).



of UVA light and photocatalyst resulted in significant GRL decolorization efficiency.

Factors affecting photocatalytic degradation of GRL dye

Effect of catalyst loading
A set of experiments were carried out to attain the optimum catalyst loading by varying the amount of TiO2 catalyst from
0.25 to 4 g L—1 (Figure 3). Results show that the apparent rate
constant and relative photonic efficiency of GRL degradation increased as TiO2 content increased to 3 g L—1 from 0.25 g L—1. However, further adding TiO2 content in the reacting system resulted in reduced photocatalytic degradation of GRL dye.
This is attributed to the fact that the increase in the amount of TiO2 nanoparticles caused an increase in the number of active sites on the photocatalyst surface, which in turn increases the number of superoxide and hydroxyl radicals. This could be explained by the fact that the collision frequency between oxidants and maxilon blue GRL would be affected by an increase in TiO2 dosage (Miao et al. 2014). Furthermore, the apparent rate constant and relative photonic efficiency reduced due to the retardation of the light by the suspension when the concentration of the catalyst increases above the optimum value (Muruganandham and Swaminathan 2006).

Influence of pH

TiO2 nanoparticles was studied at different pH values ranging from 3 to 11. Prior to TiO2 addition, the photostability of GRL at different pH values was tested in the absence of TiO2. GRL dye becomes completely dissociated at high pH values due to the complete ionization of hydroxyl and sulfonate groups.
The results show that TiO2 nanoparticles exhibit better apparent rate constant and relative photonic efficiency at pH 5 (Figure 4), since pH of solutions is dependent on the surface charge properties of the semiconductors. The highly photo- stability of the undissociated species under acidic conditions, at pH < 5, caused a decrease in the apparent rate constant and relative photonic efficiency.


Effect of initial dye concentration
The effect of various initial dye concentrations at different interval times on the degradation of GRL on TiO2 surface has been investigated. Since the generation of ROS has remained constant, the probability of dye molecule to react with reactive species decreases. The photocatalytic degradation efficiency decreases at higher initial dye concentrations, while at low dye concentration, the photo absorbed by the catalyst will increase thereby reverse effect will observe.
However, the formation of OH•, O—?, or h? on the catalyst
surface remains constant for a given catalyst amount, light intensity, and irradiation time. Therefore, at higher concentra-
•

The performance of solution pH effects is a very difficult func-

tions, the available OH

radicals are inapplicable for pollutant

tion on the efficiency of dye photocatalytic degradation pro- cess because of its multiple roles (Akpan and Hameed 2009) such as the related to the ionization state of the surface accord- ing to the following reactions:

degradation. Consequently, the rate of photocatalytic degra-
dation increased when the pollutant concentration increases (Bahnemann, Muneer, and Haque 2007).
It is well known that the rate of photocatalytic degradation
strongly depends both on the concentration of organic com-

TiOH ? H? ¼ TiOH?

?5?

pound used as the test molecule (GRL) and on the intensity

TiOH ? OH— ¼ TiO— ? H2O ?6?

Also the role of hydroxyl radicals can be formed by the reaction between hydroxide ions and positive holes. In addition, it should also be cleared that under acidic condition the surface area available for dye adsorption and photon

of the incident light (Tschirch, Dillert, and Bahnemann
2008). This expression is usually explained by the empirical kinetic model given in Equation (7), i.e., called Langmuir– Hinshelwood rate law (L-H): (Kormann, Bahnemann, and Hoffmann 1991; Hoffmann et al. 1995)

absorption would be reduced because TiO2 particles tend to

rGRL ¼

kr :K

ads

:½GRL?ads

:Ib 7?

agglomerate (Fox and Dulay 1993). The apparent rate constant

1 ? Kads:½GRL?0

hv ?

and relative photonic efficiency of GRL in the presence of




Figure 3. Variation of photonic efficiency and apparent rate constant values as a function of amount of catalyst. Experimental conditions: pH ¼ 6.55, GRL concentration 25 mg•L—1, light intensity 33.5 mW cm—2.

rGRL ¼ kap:½GRL? ?8?




Figure 4. Variation of photonic efficiency and apparent rate constant values as a function of solution pH. Experimental conditions: catalyst loading 3 g L—1, GRL concentration 25 mg • L—1, and light intensity 33.5 mW cm—2.




kap ¼ kr :h:Ib Kads:½GRL?ads

?9?

h ¼ 1 ? K


ads

:½GRL?ads

?10?

where a constant of different intermediate rate steps, Ihv is an incident light intensity of photocatalytic degradation, b is the empirical constant, h is the fraction of the surface covered by the reactant, and the Kads is the adsorption equilibrium con- stant for GRL.
The applicability of L–H equation for degradation has been confirmed by the nonlinear plot (correlation coefficient of 0.944), obtained for rate of photodegradation (R) vs. initial dye concentration (Figure 5). This indicates that the degra- dation of GRL occurred mainly on the surface of TiO2.

Effect of light intensity
The extent of light absorption by the semiconductor catalyst at a given wavelength was determined by light intensity. The rate of initiation of photocatalysis and electron–hole formation in the photochemical reaction is strongly dependent on the light intensity(Cassano and Alfano 2000).












Figure 6. Variation of photonic efficiency and apparent rate constant values as a function of light intensity. Experimental conditions: Catalyst loading 3 g L—1, pH
¼ 6.55, GRL concentration 25 mg • L—1.


check their effects on the relative photonic efficiencies of GRL dye.
The reaction pathway of GRL degradation through gener- ation of radicals from photogenerated electron-hole pairs
CB; h? ) is shown as follows (Subramonian and Wu 2014):

Figure 6 shows plots of the apparent rate constant as well as
the relative photonic efficiency as a function of the applied light intensity. Figure 6 shows a nonlinear relation between

(e— VB


TiO2 ? GRL ! GRL=TiO2 ?11?
. ?

the apparent rate constant and the employed light intensity

GRL=TiO2 ? hv ! GRL=TiO2 e— ?

?12?

was observed, while the relative photonic efficiency would be

TiO2?h? 2

CB ? hVB

VB? ? H O ! OH? ? H? ?13?

independent light intensity at high light intensity more than 25 mW cm—2. Because at higher light intensity the relative

TiO2?h? ? ? OH—

! OH?

?14?

CB? ? O2 ! O2 ?15?

photonic efficiency will decrease, this is attributed to the pri-
mary initiator of oxidation of electron-donating substrates and the surface-bound hydroxyl radical would be the principal hole trap when the intensity is increased by about 25 mW cm—2(Garcia and Takashima 2003).

TiO2?e— ?—
OH ? GRL ! Intermediate Product ! Degradation ?16?
2 ? GRL ! Intermediate Product ! Degradation ?17?
It is well known that two main species have the major contributions: hydroxyl radicals and electrons and holes (e— ; h? ) have an affected on photocatalytic degradation pro-

CB VB

Role of reactive oxygen species (ROS)

In order to distinguish the contribution of the surface reaction with OH•, O—?, or h? species, different ROS were employed to

cess (Saien and Soleymania 2007). The recombination lifetime
of the photogenerated electrons and holes and the interfacial electron-transfer rate were used to determine the overall quan- tum efficiency of photocatalysis.
To strengthen the quantum efficiency, the most general way is to try to retard the recombination of photogenerated holes and electrons. Therefore, filling the valence band holes by the electrons of some kind of reductant may confirm the photocatalytic efficiency.
In the present study, several scavengers were used, such as EDTA was adopted as the scavengers for h? (Lv, Zhu, and Zhu 2013), isopropanol for OH• (Martin, Lee, and Hoffmann

1995), N-phenyl aniline for O—?

(Zhang et al. 2014), and











Figure 5. Langmuir–Hinshelwood model: (inset variation of PDE % after
—1

potassium iodide (KI) for both h? and OH•. (Zhang et al. 2008; Van Doorslaer et al. 2012). As can be seen from Figure 7, the photocatalytic degradation efficiency of GRL is 62.93% without scavengers after irradiation for 60 min.
When isopropanol was added, the degradation efficiency changed and reduced to 18.9%, indicating that OH• can be important in the photocatalytic process; also the presence of N-phenyl aniline remarkably slightly reduces the photocataly- tic activity of GRL under the same condition to 38.56%, which
suggests that O—? is the secondary active species in the reac-

75 min). Experimental conditions: Catalyst loading 2 g L sity 33.5 mW cm—2.

, pH ¼ 6.55, light inten-

2
tion. On the other hand, the presence of N-phenyl aniline or