Detoxification of Carcinogenic Dyes by Noble Metal (Ag, Au, Pt) Impregnated Titania Photocatalysts

Textile industries produce large quantities of waste water which is notoriously known to contain strong colour, suspended solids and COD. Although several treatment processes, such as filtration, coagulation, oxidation, ozonation, reverse osmosis, etc., have been practiced for many decades, they all have inherent limita-tions. They transfer the harmful pollutants from one phase to another and do not address detoxification. Semiconductor photocatalysis is a promising technique for photodegradation of hazardous chemicals found in waste waters. Among various semiconductors, TiO 2 has been studied widely. However, the main drawbacks associated with TiO 2 are (i) large band gap (Eg > 3.2 eV) (only UV active) and (ii) recombination of excitons. Dye sensitisation, coupling of semiconductors, and transitional metal doping are some of the methods reported to shift its optical response to visible region. Herein, nanoparticles of different noble metals such as Ag, Au and Pt were deposited on synthesised TiO 2 , characterized by XRD, TEM, FT-IR, BET, UV-Vis and AAS and were subjected to the degradation of some textile dyes namely Tartrazine (TAZ), Reactive Yellow-17 (RY-17) and Reactive Black-5 (RB-5) under both UV and visible irradiations. The reaction conditions such as catalyst concentration, dye concentration, pH, irradiation time, light intensity, and additives were optimized for complete decolourisation and discussed.


Introduction
Dyeing and finishing processes are the two important steps in textile industries. Large number of synthetic dyes are used extensively in the textile dyeing process and these industries produce large quantities of waste water containing carcinogenic dyes. When ground water mixes with such waste water it gets polluted and cause allergy, damage of eye, brain, liver and reproductive organs and deformities in babies during pregnant period. Thus, water pollution is a major concern in the developing nations. Though dyes are aesthetic pollutants by nature of their colour, they may interfere with light penetration in the receiving water bodies thereby disturbing the biological processes.
Among the various types of dyes, reactive dyes are extensively used for dyeing cotton fabrics and hence they add a lot to water pollution. The volume of textile effluent discharged from dyeing industries mainly using reactive dyes are approximately calculated to be 3,00,000-4,00,000 L ton −1 fabric materials in a year [1].
The increased public concern and the stringent international environmental standards have prompted the need to develop novel treatment methods for converting dye effluents to harmless compounds. Although some conventional treatment technologies such as filtration, coagulation, flocculation etc. have been tried in the past they are not viable and green technologies. Zero generation of sludge and complete mineralisation of dyes are the two important criteria for the technology to be economically attractive and environmentally benign. In this context the present study of dye treatment by using photocatalysts which mineralises the carcinogens into CO 2 , H 2 O and mineral salts assumes great significance.
The ideal photocatalyst should be stable, inexpensive, non-toxic and highly photoactive. Another primary criteria for the degradation of organic compounds is that the redox potential of the H 2 O/OH˙ couple lies within the band gap of the semiconductor [2]. Several semiconductors have band gap energies sufficient for catalysing wide range of chemical reactions. These include TiO 2 , WO 3 , SrTiO 3 , ZnO, ZnS, etc. Binary metal sulphide semiconductors such as CdS, CdSe and PbS are not sufficiently stable catalysts in aqueous media as they readily undergo photoanodic corrosion [3].These materials are also known to be toxic. The iron oxides are not suitable semiconductors as they readily undergo photocathodic corrosion [4]. ZnO is unstable in water as Zn(OH) 2 being formed on the particle surface. This results in catalyst deactivation [5]. However, the anatase form of TiO 2 seems to be the best. The photoactivity of TiO 2 is known for approximately 60 years and investigated extensively. Although TiO 2 absorbs only approximately 5% of the solar light reaching the surface of the earth, it is the best-investigated semiconductor in the field of chemical conversion and storage of solar energy. Accordingly, many efforts have been made to sensitise titania for visible light induced photocatalytic reactions. Such sensitization techniques include (i) Doping with altervalent ions, (ii) heterojunctioning with other semiconductors, (iii) dye sensitization (iv), deposition of noble metals over semiconductors etc.
Herein we report the deposition of noble metals such as Au, Pt and Pd over synthesised titania. Noble metal is important because of their own catalytic activity and they actually modify the photocatalytic properties of the semiconductor by changing the distribution of electrons. Further, it can enhance the yield of a particular product or the rate of the photocatalytic reaction. The addition of a metal to a semiconductor surface also changes the reaction products. On the other hand, the loading level is important in governing the net effect of metallation as heavy metal loading induces faster electron-hole recombination [6]. One of the main reasons for the numerous studies of titaniasupported metal catalysts is the so-called strong metal-support interaction (SMSI). The platinised titania exhibits a much higher activity in a series of photocatalytic reactions than pure anatase. Combined Pt-RuO 2 /TiO 2 catalysts are the most promising ones for photocatalytic water cleavage [7]. The Au, Ag and Pd impregnated titania catalysts were synthesised, characterised by various instrumental techniques and evaluated towards the decolourisation of one azo dye tartrazine (TAZ) and two reactive dyes Reactive Yellow and Reactive Black (RY-17 and RB-5) ( Table 1).
All other chemicals namely NaCl, C 2 H 5 OH, Na 2 CO 3 , H 2 O 2 , K 2 S 2 O 8 , NH 4 OH, H 2 SO 4, isopropanol and acetic acid were obtained from Merck (purity >98%) and were used as received. The water employed for the studies was doubly distilled.

Synthesis of photocatalysts
The catalytic materials used in the present study were synthesised as described below:

Synthesis of TiO 2
Sol gel process was adopted for the synthesis of TiO 2 . Titanium isopropoxide, isopropanol, water and acetic acid were used as starting materials. Solution A contained 17 mL titanium isopropoxide and 40 mL of isopropanol and Solution B contained 60 mL of isopropanol, 15 mL acetic acid and 5 mL water. Solutions A and B were mixed and stirred for 2 h. The formed TiO 2 sol was aged to get the gel of TiO 2 . The obtained gel was dried, ground with mortar and pestle and finally calcined at 500°C for 3 h [8].

Synthesis of M/TiO 2 (M = Ag, Au and Pt)
The M/TiO 2 (M = Ag, Au and Pt) catalysts were prepared by photoreduction method [9]. Calculated amounts of noble metal precursors (0.14 g of AgNO 3 for 1% Ag/TiO 2 , 0.15 g of AuCl 3 for 1% Au/TiO 2 , 0.14 g of [Pt(NH 3 ) 4 ]Cl 2 for 1% Pt/TiO 2 ) taken in minimum amount of water was added to 10 g of synthesised TiO 2 and was stirred at 70°C till the evaporation of the solvent. The colloidal solution was then irradiated with 125 W halogen lamp until complete reduction was observed visually by the formation of black precipitate for Ag/TiO 2 , Pt/TiO 2 and purple coloured product for Au/TiO 2 .

X-ray diffraction analysis
Synthesised TiO 2 and M/TiO 2 photocatalyst were subjected to powder X-ray diffraction to confirm the crystalline size and the presence of the metal ion. The powder X-ray diffractograms were recorded using a X-ray diffractometer (PANalytical X'Pert Pro) with Cu Kα source having wavelength of 1.54 Å operating at 20 mA and 50 kV. The samples were scanned from 0.5-80° (2θ) for the confirmation of phase and presence of metal particles. The diameter of the metal particles was calculated from the (111) reflection of metal atom using Debye Scherrer equation, where d, diameter of the metal particle; K, constant; λ, wavelength of the X-ray (1.54 Å); β, full width at half maximum; θ, angle of diffraction.

Transmission electron microscopy
Particle size of the noble metals deposited on TiO 2 was determined from electron micrographs taken by using transmission electron microscope (JEOL JEM-2010F) operating at 200 kV. A drop of alcoholic solution of the catalysts was placed on Cu grid and kept aside for about 45 s. Then transmission electron micrographs were recorded for the particles present on the grid. From the TEM images, average particle size of the catalysts was determined be taking minimum of 30 particles.

BET surface area analysis
BET surface area analyser (Nova-Quantachrome 4200 e) was used to measure the surface area of all the catalysts. The measurement was done by using N 2 as probe molecule at liquid nitrogen temperature (−190°C). Before the measurement, all the catalysts were degassed at 300°C at the pressure of 10 −5 Torr for 6 h. Linear portion of the BET plots was used for the determination of BET surface area.

Atomic absorption spectroscopic studies
The actual amount of metal content present in the synthesised catalysts (M/TiO 2 , M = Ag, Au and Pt) was determined using atomic absorption spectrophotometer (Perkin Elmer 2380) after digesting the photocatalysts in aqua regia.

UV-visible spectrophotometry
UV-visible spectra of all the samples were recorded using UV-Vis double beam spectrophotometer (Hitachi U-2000) in the range 200-800 nm. The course of reduction of noble metal ions on TiO 2 particles and extent of decolourisation were studied using this technique. Further, the band gap energies (E g ) of the synthesised catalysts were calculated according to the equation where h, Planck's constant; c, velocity of light (m/s); λ, wavelength (nm).

Preparation of synthetic dye samples
Stock solutions (10 −4 M for TAZ and 10 −5 M for both RY-17 and RB-5) of the dye samples were prepared for the present study.

Batch photocatalytic reactor
The photocatalytic reactor was made up of quartz having dimensions 30 × 3 cm (height × diameter) provided with water circulation arrangement to maintain the temperature. The top portion of the reactor has ports for sampling.
The irradiation was carried out using low pressure mercury arc lamp (wave length 254 nm) as UV source and tungsten lamp (365 nm) as visible light source built into a lamp housing with polished anodised aluminium reflectors placed 6.5 cm away from the lamps. The entire reactor system was cooled using an inbuilt fan set up.

Photocatalytic reaction
250 mL of the desired dye sample was taken in a photocatalytic reactor. Calculated amount of TiO 2 (P-25 Degussa) or synthesised TiO 2 or M/TiO 2 was added to the photocatalytic reactor and the reaction mixture was magnetically stirred before and during illumination. After specific time interval of irradiation, suitable aliquots of the sample was withdrawn and analysed to find out the extent of decolourisation and degradation.

UV-vis analysis of the decolourised products
The extent of photocatalytic decolourisation of dyes was studied by UV-visible spectrophotometer (Hitachi U-2000) from the decrease in the respective λ max values.
The colour intensity of the dyes was measured in terms of absorbance. Decolourisation was determined by using the calibration curve.
where C 0 , initial concentration of the dye; C, concentration of dye at time 't' .

Photodegradation studies
The photocatalytic degradation of dyes was evaluated in terms of total organic carbon (TOC). The total organic carbon of the pollutant was determined by using a TOC analyser (Shimadzu TOC V series).
Gold Nanoparticles -Reaching New Heights 6

Textile effluent study
To check the photocatalytic activity of M/TiO 2 catalyst for the photocatalytic degradation of actual textile effluent, the samples were collected from M/s Ramkay & Co, Erode, Tamil Nadu, India, filtered and diluted. Since the effluent contains number of unknown dyes it gave a broad peak in the wavelength range of 500-600 nm. Since the absorbance is additive, the effluent simply shows a broad peak in this range. To the diluted effluent (250 mL), 6 g of M/TiO 2 (M = Ag, Au and Pt) catalyst was added and the photocatalytic study was performed in the photocatalytic reactor.

COD experiment
Chemical oxygen demand (COD) of the samples collected at different intervals was determined by dichromate method in the presence of a catalyst Ag 2 SO 4 [10,11].

UV-visible spectrophotometry
The absorption spectra of synthesised TiO 2 (gel form) was recorded using UV-visible spectrophotometer and given in Figure 1. An absorption maxima observed around 320 nm is due to the charge transfer from the VB of 2p orbitals of the oxide ions to the CB of 3d t2g orbitals of the titania cations [12].
This figure shows that on visible irradiation, appearance of new bands centred on 370 nm for Ag, 520 nm for Au and 502 nm for Pt were observed. The reason for appearance of these peaks is due to the surface plasmon excitation of the respective metal colloids [13][14][15].
The band gap values of all the synthesised catalysts (TiO 2 and M/TiO 2 ) were calculated from the corresponding λ values obtained by extrapolating the rising portion of the spectrum to the × axis at zero absorbance and by using the following equation. The values were consistent with the values reported already [16].
The band gap values for the synthesised TiO 2 and metal loaded TiO 2 are given in Table 2. The band gap values were found to be lower for the metal loaded catalyst when compared to the synthesised TiO 2 catalyst.
Although lowering of band gap is not good for better catalytic activity under UV irradiation due to easy recombination, the presence of metals act as electrons traps and prevents the recombination process and also making the photocatalyst active in the visible range.

Atomic absorption spectrophotometric analysis
To determine the actual metal content of all the synthesised M/TiO 2 catalysts, they were subjected to AAS analysis after dissolving them. The metal content in each catalyst determined is given in Table 2. The results show that the experimentally determined metal content value is close to that of the theoretical value (1% w/w).

BET surface area measurements
The surface area of all the synthesised catalysts viz., TiO 2 (P-25 Degussa), TiO 2 and M/TiO 2 catalysts were determined and given in Table 2. Table 2 clearly shows that the synthesised catalysts show higher surface than commercial TiO 2 (P-25 Degussa). It is to be noted that impregnation of noble metals (Pt, Pd and Au) over TiO 2 did not alter the surface area values significantly. A very small reduction in the surface area observed may be due to the blocking of fine capillaries present on TiO 2 surface by metal thin islands. These islands prevent the entry of the probe molecule (nitrogen gas) into the pores during BET measurement [17].

X-ray diffraction analysis
To obtain information regarding the phase formation and crystallite size, X-ray diffraction measurements were performed for the synthesised TiO 2 and M/TiO 2 (M = Ag, Au and Pt) photocatalysts and the XRD patterns are shown in Figure 2.
The synthesised TiO 2 has both anatase and rutile phases but not the brookite phase. Anatase and rutile phases are confirmed by the appearance of major peaks at 2θ = 25.4 and 48° respectively. The corresponding d (111) reflections of the noble metal atoms were found at 2θ = 38.1, 38.8 and 40° for Ag, Au and Pt, which confirms the impregnation of metal particles on TiO 2 lattice. From the X-ray diffraction patterns the average particle size of the synthesised TiO 2 and M/TiO 2 was calculated using Debye-Scherrer equation.
The average particle diameter of the synthesised TiO 2 and M/TiO 2 was found to be in the nanometre range as shown in Table 2.

Transmission electron microscopy
TEM images and particle size histograms of synthesised TiO 2, Ag, Au and Pt deposited TiO 2 catalysts are shown in Figure 3. Almost uniform dispersion was obtained  with all the catalysts. The average size of the particles was calculated by averaging particle sizes of 20-30 particles. The plot of particle size distribution is also shown.
The average particle size of TiO 2 , Ag/TiO 2 , Au/TiO 2 and Pt/TiO 2 catalyst was found to be 27 nm, 11 nm, 10 nm and 19 nm respectively.

Effect of initial concentration of dye
The effect of initial concentration of dye on the rate of decolourisation was studied by taking 250 mL of dye solutions and varying the concentrations between 8 × l0 −5 and 1.3 × l0 −4 M for Tartrazine (TAZ), 8 × l0 −6 and 1.6 × l0 −5 M for RY-17 and 6 × l0 −6 and 1.6 × l0 −5 for RB-5 with constant catalyst weight (1.5 g). The irradiation was carried out for 6 h by using 125 W low pressure mercury arc lamp (wave length 254 nm) and 85 W tungsten lamp (wave length 365 nm) as UV and visible light sources, respectively. Figure 4 shows that percentage decolourisation decreases as the initial concentration of the dye increases under both UV and visible light illumination.
Similar results in the photocatalytic degradation of phenol were reported in the literature [18,19]. The decolourisation rate relates to the probability of formation of hydroxyl radicals (OH˙) on the catalyst surface and the probability of hydroxyl radicals reacting with dye molecules. Hence the rate constant depends on the probabilities of the formation of these two. Hence k' can be expressed as k ̕ = k 0 x P OH x P dye molecules (6) where k' is the overall rate constant and k 0 is the reaction rate constant. P OH˙ and P dye molecules refer to the probabilities of generation of OH˙ radicals and OH˙ radicals reacting with dye molecules. We all know that the reaction rate constant k 0 is independent of initial dye concentration but P OH˙ and P dye molecules will be affected by the dye concentration. Literature suggests that photocatalytic degradation of aromatic compounds mainly occurs by hydroxyl radicals [20]. The rate determining step of the reaction may be the formation of OH˙ radicals that are formed through the reaction of holes with adsorbed OH − and H 2 O [21,22]. As the dye concentration increases, the available hole sites may be occupied by dye ions which are generated from the sodium salt of the dye molecules, Since there are only a few active sites available for the generation of OH˙ radicals the generation of OH˙ will be reduced.
It is concluded that as the initial dye concentration increases, the catalyst surface needs to generate more amount of OH˙ radicals and other oxidising species. But, illumination time and required amount of the catalyst are constant, so that OH˙ radicals and other oxidising species formed on the TiO 2 surface are also constant. So the relative number of OH˙ attacking the dye molecules decreases with increasing dye concentration [23]. Moreover, as the initial concentration of the dye increases, the path length of photons entering the solution decreases and at low dye concentration the reverse effect is observed [24,25]. Also, at higher concentration, degradation decreases at sufficiently long distances from the light source or reaction zone due to retardation in the penetration of light. Hence, under both UV and visible light sources the rate of degradation decreases considerably with increase in dye concentration above the optimal concentration.

Effect of catalyst weight
A series of experiments were carried out to assess the optimum weight of the catalyst by varying the amount of TiO 2 (P-25 Degussa) from 0 to 3 g for the decolourisation of 250 mL of dye solution (TAZ = 1 × 10 −4 M, RY-17 and RB-5 = 1 × 10 −5 M) at the irradiation time of 6 h at neutral pH. The effects of catalyst weight on the percentage decolourisation of all the dyes are shown in Figure 5.
In the absence of the catalyst no decolourisation occurred. From the figure it is also evident that the rate of photodecolourisation increased linearly with the weight of catalyst up to 1.5 g irrespective of light sources. On increasing the catalyst weight further, the percentage decolourisation decreased. This decrease in decolourisation on increasing the catalyst weight may be due to the formation of turbidity (Shadowing effect). Similar observation was also reported [26]. Optimum catalyst  loading is always essential and the higher amount of TiO 2 may not be useful both in view of aggregation as well as reduced irradiation field due to shadowing effect [27].
As the weight of the catalyst increased, the quantity of photons adsorbed increased and consequently the decolourisation rate increased [28], Hence, optimisation of the catalyst loading for a given dye concentration is an important parameter to avoid excess catalyst and to ensure total absorption of light either UV or visible light for efficient photodecolourisation.

Effect of pH
The effect of pH on decolourisation of dyes irradiated under UV and visible irradiation are given in Figure 6.
The percentage decolourisation of all the dyes was found to be maximum at the pH of 7 for UV and visible irradiations. On increasing the pH from acidic to neutral under UV and visible irradiations, the percentage decolourisation increased significantly. However, when the pH was increased beyond these values to the basic ranges the percentage decolourisation decreased drastically. The main reaction is presented by the hydroxyl radical attack on the dye anion. Hydroxyl radicals  which are considered as the predominant species at neutral or alkaline pH values are generated by oxidising more hydroxide ions [29]. At low pH values (pH < 5) the photodecolourisation of dyes is retarded under both UV and sunlight sources by the high concentration of the proton. Lack of availability of hydroxyl radicals in the pH range less than 5 leads to the decrease in decolourisation of dyes. In highly alkaline conditions (>pH 9) the % decolourisation of dyes decreased drastically due to the electrostatic columbic repulsion between the anionic dye surface (negatively charged) and the hydroxyl anions. Due to this repulsion, the dyes do not interact closely with the anions [30][31][32][33]. Thus it is deduced that the efficient condition for the maximum degradation of all the mentioned dyes is at neutral pH (pH = 7).

Photocatalytic degradation studies
The extent of degradation of dyes was followed by total organic carbon analyser (TOC). The experimental results revealed that as the irradiation time increased, some dye molecules may degrade into components of lower molecular fragments and they mineralise. TOC analysis was carried out for all the dye samples collected at different intervals of time and the results are shown in the Figure 7.
Since the decrease in the TOC content is the direct measure of degradation, TOC studies have been carried out to check whether the photocatalyst converts the harmful dye into harmless products. The studies revealed that, about 45% of TOC of dye samples was reduced under UV and 35% under visible irradiations. Similar type of observations was also made by [34,35]. The degradation of dye involved the cleavage of ▬N〓N▬ and sequential evolution of N 2 in the early stage of degradation [36]. The fate of nitrogen containing compounds in the photodegradation was also explained by [37].

Recycling of TiO 2
In order to test the recyclability of TiO 2 catalyst, the experiment was repeated with 250 mL of dye solutions (TAZ =1 × 10 −4 M, RY-17 and RB-5 = 1 × 10 −5 M) with the used catalyst. The photocatalytic experiments were carried out with 1.5 g of TiO 2 (P-25 Degussa) for 5 h. At the end of reaction, the dye samples were filtered and the catalyst was filtered off. There was not much difference between the weight of the recovered catalyst with its initial weight (<1%). Repeated experiments were carried out for another 5 h by taking 250 mL of fresh dye solution. The pseudo first order rate constant values were calculated for each cycle as mentioned before and the values are given in Table 3.  The rate constant values were found to be almost the same for all the dyes under both UV and visible irradiations even after two cycles. Nevertheless at the end of II recycle, 91% decolourisation occurred in the illumination period of 5 h. This shows that the TiO 2 (P-25 Degussa) is a very good catalyst which does not undergo significant deactivation even after 2 cycles. A very slight decrease in the rate constant values obtained in the photocatalysis experiments may be due to the adsorption of dye on the active sites of the catalysts which could not be removed during filtering.

Effect of inhibitors
Large amount of sodium chloride is used in the dyeing process in the textile industries and hence it usually comes out in the dye effluent along with sectional water of textile mills. Sodium carbonate also plays an important role in fixing the dye on fabrics and in the fastness of colour and hence it is mainly used in the dye bath to adjust the pH of the bath. Alcohols are used for padding process during dyeing the fabrics. Alcohols such as ethanol are commonly used to quench hydroxyl radicals. Hence, it is important to study the influence of chloride ions, carbonate ions and ethanol on the photomineralisation efficiency of the catalyst. Irradiation experiments were carried out by taking 250 mL of different dyes of various concentration (TAZ =1 × 10 −4 M, RY-17 and RB-5 = 1 × 10 −5 M), 1.5 g of TiO 2 (P-25 Degussa) under neutral pH and with different concentrations of (0-2 g) for both sodium chloride and sodium carbonate and 0 to 1 mL of ethanol. The irradiation was carried out by using 125 W low pressure mercury arc lamp and 85 W tungsten lamp as UV and visible light sources, respectively. The results obtained under both UV-visible irradiations are shown in Figure 8.  Table 3.

The presence of chloride ion in the dye samples had a negative effect on decolourisation. The percentage decolourisation decreased drastically with increase in the concentration of sodium chloride. This decrease in the percentage decolourisation of dyes in the presence of chloride ion is due to the hole-scavenging properties of chloride ion (equations 7 and 8).
Effect of catalyst recyclability on the degradation of textile dyes. This is a distinctive competitive inhibition reaction where the reaction of dye molecules with the holes has to compete with the chloride ions, which are not readily oxidisable. This inhibitive effect was also observed by [38]. However, the photocatalytic activity of the catalyst can be fully restored by washing the catalyst with pure water.
The decrease in the degradation of dyes in the presence of carbonate and bicarbonate ions is due to the hydroxyl scavenging property of carbonate ions according to the following Eqs. (11) and (12).
The presence of carbonate and bicarbonate ions reduces the concentration of hydroxyl radicals significantly and hence the percentage decolourisation of the dyes decreases [39].
When the volume of ethanol was increased the photodecolourisation efficiency gradually decreased. This shows that the dye degradation occurs via positive holes (h + vb ) that are not involved in recombination with electrons during illumination process. Further it is evident that hydroxyl radicals are not the only species responsible for degradation and minor degradation also occurs through holes generated during illumination [40].

Effect of electron acceptors
The practical problem in using TiO 2 as a photocatalyst is the undesired electron/ hole recombination. In the absence of proper electron acceptors or donors the electron/hole recombination is extremely efficient and represents the major energy wasting step which limits the achievable quantum yield. One strategy to inhibit DOI: http://dx.doi.org /10.5772/intechopen.80467 electron/hole recombination is to add other electron acceptors such as hydrogen peroxide, persulphate ions to the reaction mixture.
Hydrogen peroxide is extensively used in commercial bleaching of textiles, especially cellulose fibres and wool. In order to study the influence of hydrogen peroxide in the decolourisation of dyes, experiments were performed by adding different quantities of hydrogen peroxide(30% v/v) (0-10 mL) and 0 to 2 g of potassium persulphate to 250 mL of different dyes of various concentration (TAZ = 2 × 10 −4 M, RY-17 and RB-5 = 2 × 10 −5 M), 1.5 g of TiO 2 (P-25 Degussa) under neutral pH. The irradiation was carried out by using 125 W low pressure mercury arc lamp and 85 W tungsten lamp as UV and visible light sources respectively.
The presence of hydrogen peroxide will prevent the (e − /h + ) pair recombination and enhances the formation of hydroxyl radicals (Eq. 13).
Since hydrogen peroxide is better than molecular oxygen as electron acceptor it leads to 100% decolourisation in comparison to the results obtained in the absence of H 2 O 2 . But, in the absence of TiO 2 , and with only H 2 O 2 the percentage decolourisation was found to be less under both the light sources as shown in Figure 9.
The percentage decolourisation increased on increasing the volume of H 2 O 2 from 0 to 10 mL. However, percentage decolourisation decreased on increasing the volume of H 2 O 2 beyond 10 mL. Since, H 2 O 2 is a powerful hydroxyl scavenger, the percentage decolourisation decreased at the higher concentration as shown in Eqs. (14) and (15).
The above decolourisation results carried out by both UV and visible sources show the necessity of choosing the proper dosage of hydrogen peroxide for maximum decolourisation. Thus, it is clear from the Figure 9 that addition of H 2 O 2 up to 10 mL concentration showed a beneficial effect on the photocatalytic decolourisation of the dyes [41].
The rate of photo-assisted decolourisation of dyes is significantly improved by the presence of persulphate ions.
The percentage decolourisation increased with increase in the amount of K 2 S 2 O 8 and 100% degradation was attained at highest concentration of K 2 S 2 O 8 for both UV and visible irradiations. The added persulphate ion acts as electron traps resulting in the formation of reactive radical intermediate SO 4˙− as shown in Eq. (16).
The sulphate radical anion is a strong oxidant and removes electron from neutral molecules like water and generates hydroxyl radical.
The persulphate ions oxidise the dyes as well as act as electron scavenger. Due to their scavenging property they inhibit e − /h + pair recombination at the semiconductor surface [42]. S 2 O 8 2− ions produce sulphate radical anion (SO 4˙− ) which in turn produce highly oxidative hydroxyl radicals. These hydroxyl radicals degrade the dyes at a faster rate.

Comparison of photocatalytic activities of synthesised TiO 2 with TiO 2 (P-25 Degussa)
The photocatalytic decolourisation of the dyes TAZ, RY-17 and RB-5 were carried out with the synthesised TiO 2 catalyst under the conditions optimised for the best decolourisation with TiO 2 (P-25 Degussa). The optimised conditions for the complete decolourisation of various dyes are given in Table 4 and the time taken for complete decolourisation of the dyes over the synthesised titania is given in Table 5.
Among the three dyes, RY-17 reached complete decolourisation at a shorter irradiation time under both the light sources. RB-5 took longer irradiation time for complete decolourisation. This may be due to its chemical nature (i.e.) RB-5 being diazo dye. Diazo dyes are less degradable than mono azo dye during ozonation [43]. Similar trend of decolourisation was observed with visible light irradiation.  However, the times at which complete decolourisation took place were different and higher for visible light when compared with UV.
The higher efficiency of the synthesised catalyst when compared to P-25 Degussa may be due to the larger surface area of the synthesised TiO 2 (90.5 m 2 /g) when compared with TiO 2 (P-25 Degussa) (50 m 2 /g). Similar type of results was also observed by [44]. Because of the increase in the surface area of synthesised TiO 2 there may be increase in the adsorption of the three dyes.
From the above discussions, it is clear that both degradation and decolourisation of the dyes are highly effective with synthesised TiO 2 compared to TiO 2 (P-25 Degussa) under both the light sources. To further the photocatalytic activity further in the visible region, the modification in the synthesised catalyst has to be carried out by noble metal deposition. Noble metals such as Ag, Au and Pt have been deposited on the synthesised catalyst and they were evaluated for their photocatalytic activity.

Photodecolourisation studies of metal impregnated titania photocatalysts
To evaluate the photocatalytic activity of noble metal deposited titania catalyst, the decolourisation of dyes (TAZ, RY-17 and RB-5) was carried out under visible and UV light irradiations. The reaction was carried out with 250 mL of dyes of various concentration (TAZ =1 × 10 −4 M, RY-17 and RB-5 = 1 × 10 −5 M) and 1.5 g of M/TiO 2 catalyst at neutral pH. The irradiation was carried out by using 125 W low pressure mercury arc lamp (wave length 254 nm) and 85 W tungsten lamp (wave length 365 nm) as UV and visible light sources respectively. The results obtained in the decolourisation of various dyes with M/TiO 2 under UV and visible irradiations are shown in Figures 10 and 11.
From the above photocatalytic decolourisation studies the following observations were made: Photocatalytic efficiency of the different catalysts for the decolourisation of different dyes were found to be in the following order: Au / TiO 2 > Ag / TiO 2 ~Pt / TiO 2 > synthesised TiO 2 > TiO 2 ( P − 25 Degussa ) The time at which the complete decolourisation occurred for the different catalysts is given in Table 6.
• As far as the decolourisation of the dyes are concerned, all the catalysts found to decolourise RY-17 to the maximum extent followed by TAZ and RB-5. The percentage decolourisation of different dyes obtained with the different catalysts were determined at the time at which the Au/TiO 2 catalyst showed 100% decolourisation of the respective dyes (TAZ, RY-17 and RB-5) using different catalysts are given in Table 7.   Among all the catalysts, M/TiO 2 catalysts show the higher photocatalytic activity towards the decolourisation of all the three dyes. The higher catalytic activity of M/TiO 2 catalysts may be due to the smaller band gap values when compared to TiO 2 (P-25 Degussa) and the ability of the noble metals trap the electrons, thus preventing the recombination. The rate constant values obtained with different M/TiO 2 catalysts in the photocatalytic decolourisation of all the three dyes were found to be highest for Au/TiO 2 followed by Ag/TiO 2 and Pt/TiO 2 . The higher photocatalytic activity of the Au/TiO 2 catalyst when compared to the other metals is due to the lowered band gap values [45,46], which leads to the effective absorption of visible light by the catalyst resulting in higher decolourisation.

Photodegradation studies
The photodegradation studies was carried out with 250 mL of dyes of various concentration (TAZ = 1 × 10 −4 M, RY-17 and RB-5 = 1 × 10 −5 M) and 1.5 g of M/TiO 2 catalyst at neutral pH. The irradiation was carried out by using 125 W low pressure mercury arc lamp (wave length 254 nm) and 85 W tungsten lamp (wave length 365 nm) as UV and visible light sources respectively. The degradation was monitored by a TOC analyser and the results are shown in Figure 12.
All the above photodegradation experiments show that like photodecolourisation that photodegradation efficiency of all the synthesised catalyst follow similar trend as mentioned below: • Au/TiO 2 > Ag/TiO 2 ≈ Pt/TiO 2 > Synthesised TiO 2 > TiO 2 (P-25 Degussa) • As far as the dyes are concerned, all the catalysts found to degrade RY-17 to the maximum extent followed by TAZ and RB-5.  The degradation percentage of the dyes (TAZ, RY-17 and RB-5) using different catalysts at the time period of 8 h are given in Table 8.
The photocatalytic studies carried out under UV and visible irradiations leads to the following observations: • The photocatalytic activity of the synthesised catalyst in the decolourisation and degradation of all the dyes was found to be better than TiO 2 (P-25 Degussa) under both UV and visible irradiations.
• Higher percentage of decolourisation and degradation occurred for both TiO 2 (P-25 Degussa) and the synthesised catalyst under UV irradiation when compared to visible irradiation.
• The photocatalytic activities of M/TiO 2 catalysts were found to be higher under both UV and visible irradiations than the synthesised TiO 2 .
• The photocatalytic efficiency of the M/TiO 2 was remarkable even under visible light irradiation when compared to the UV irradiation.
The higher activity of M/TiO 2 catalysts under visible region is explained below: • The activity of noble metals may be attributed to the electronic interaction between the impregnated noble metal and TiO 2 . Due to this interaction, Schottky barriers are formed leading to electron hole pair [45].
• The metal deposits traps the electrons and quickly transfers the electrons to oxygen present in the solution to form the highly oxidative O 2˙− radical [47].
• The trapping rate of electrons by noble metals is faster rate when compared to the electron transfer from the semiconductor to O 2 [48].

Photocatalytic activity of M/TiO 2 with the actual effluent sample
The photocatalytic activity of the synthesised M/TiO 2 catalysts (M = Ag, Au and Pt) was tested with the real effluent samples collected from a textile industry (M/s Ramkay, Erode, Tamil Nadu, India).
The industrial effluent (1 L) sample was first filtered to remove any insoluble matter (carbonates, hydroxides, etc.). As the initial pH was in weakly acidic, it was increased to 7 by adding dil. NH 4 OH and the pH adjusted effluent was then subjected to photocatalytic treatment by adding 6 g of the M/TiO 2 (M = Ag, Au and Pt) catalyst. Before the lamp was switched on, the effluent was equilibrated for few minutes with continuous stirring. The effluent was then irradiated using visible light source. Aliquots of the sample were withdrawn at regular time intervals and the extent of decolourisation was monitored using UV-Vis spectrophotometer (Hitachi U 2000). The degradation of the dyes was also determined by using total   organic carbon analyser (TOC). The decolourisation and degradation results obtained are shown in Figure 13.
The effluent studies show that, all the M/TiO 2 catalysts decolourise and degrade the effluent samples significantly. However, these catalysts decolourise the effluent to the extent of only 65-75% and degrade to the extent of 25-30% that too after longer time of irradiation (12 h). The lesser percentages of decolourisation and degradation may be due to the presence of infinite number of dyes, salts etc. in the effluent.

Conclusions
The optimised reactions parameters for the degradation of TAT, RY-17 and RB-5 were found to be: The initial dye concentration (TAZ = 1 × 10 −4 M and for both RY-17 and RB-5 = 1 × 10 −5 M), catalyst weight (1.5 g/L), pH (7) and the time taken for complete decolourisation (TAZ: 5 ½ h (UV) 6 ½ h (visible), RY-17: 5 h (UV) 6 h (visible), RB-5: 6 h (UV) 7 h (visible). Decolourisation and degradation were comparatively faster under UV than under visible. Inhibitors such as sodium chloride, sodium carbonate and ethanol decreased the degradation rate and electron acceptors such as H 2 O 2 and potassium per sulphate showed beneficial effect. As far as the photocatalytic efficiency of the catalysts are concerned, the following order was observed under both UV and visible irradiations, Au / TiO 2 > Ag / TiO 2 ~Pt / TiO 2 > Synthesised TiO 2 > TiO 2 ( P − 25 Degussa ) Among the noble metals, Au/TiO 2 catalyst was found to be the most active catalyst. This may be attributed to the small band gap value of Au/TiO 2 leading to better visible light absorption.
All the M/TiO 2 catalysts were found to be more active towards the decolourisation of all the dyes even under visible irradiation. The enhanced photocatalytic activity of the M/TiO 2 under visible light irradiation may be due to the presence of impregnated metals, which act as electron traps and by this way they prevent electron hole recombination. The synthesised M/TiO 2 catalysts were found to be very active towards the decolourisation and degradation of textile dye effluent collected from an industry as well.