Enhanced Eosin Mineralization in Presence of Au(III) Ions in Aqueous Solutions Containing TiO 2 as Suspension

: Photo-catalytic mineralization of eosin in aerated 0.1% (w/v) TiO 2 suspended aqueous systems with and without Au 3+ using 350 nm photo light was carried out. Eosin mineralization rate was significantly faster in 2 × 10 -4 M Au 3+ containing systems in contrast to sole TiO 2 systems, which is due to the participation of Au 3+ and it’s in situ generated various reduced intermediates including gold nanoparticles during mineralization. Furthermore, pulse radiolysis (a well known transient measurement technique) was adopted to analyze the reaction intermediates (eosin-OH adducts and/or eosin radical cation) produced in mineralization by generating in situ • OH and N 3 • species. The reaction rates for • OH and N 3 • reactions with eosin evaluated respectively 5.4 × 10 9 and 3.0 × 10 9 dm 3 mol -1 s -1 for the formation of radical cations were slower than the eosin-OH adduct formation rate (reaction rate = 1.4 × 10 10 dm 3 mol -1 s -1 ). Furthermore, it is proposed that the initially generated eosin- • OH/hole adduct is undergoing mineralization in the presence of air/oxygen.


INTRODUCTION
Color removal from industrial effluent is one of the most difficult requirements faced by the textile, dye manufacturing and paper industries. These industries are major consumers of water and, therefore cause water pollution; afterward it spreads to air and soil pollutions. Most of the dyes are harmful when brought in contact with living tissues. The discharge of such dyes into the river stream without proper treatment causes irretrievable damage to the crops and living beings, both aquatic and terrestrial. Some of them are not only toxic but also non biodegradable; hence they are not easily removed that is why there is an urgent need to develop effective methodology either by transforming them into harmless compounds or by mineralizing them completely. Mineralization of eosin, a model dye chosen under the study is generally used in cosmetics and biological stain for studying cell structures. Its' soluble salts are normally used as dyes. Eosin has also been utilized as a groundwater migration tracer by capillary electrophoresis/laserinduced fluorescence employing a multi wavelengths laser. Eosin is a carcinogen [1], it causes also cheilitis, dermatitis and stomatitis and emits toxic fumes when heated [2].
In the presence of an • OH/h + scavenger, it is possible to utilize TiO 2 containing aqueous systems as photo-reduction processes; otherwise, this system can be used as oxidation conditions by scavenging the photo-generated e -(a strong reductant), most commonly with oxygen.
The mineralization of eosin is known wherein researchers have used TiO 2 based photo-catalysts [7][8][9][10] and the degradation rates reported therein through its' absorption measurements, which were rather slow. In the present study, we have carried out the photocatalytic mineralization of eosin using photosensitive semiconductor TiO 2 powder as suspension in water in presence of air/O 2 . This work was explored further in presence of Au 3+ (a probable electron scavenger) to check better/different mineralization kinetics subsists if any, which was the motivation of the work. The enhanced mineralization in Au 3+ containing systems observed is reported in this article.
Water from Millipore gradient A10 polishing system, with conductivity ~18.2 MΩ cm -1 and organic carbon content < 5 ppb was used for all solutions preparations.

Photolysis and Evaluation
A quartz cell of 40 mL capacity was used for photoirradiation with 350 nm light using Rayonet photo reactor having photon flux 4×10 15

Material Characterization
Membrane filters (Ultipor ® N 66 ® Nylon 6.6 membrane 0.45 µm from PALL Life Sciences) were used for filtration under vacuum to separate TiO 2 from aqueous TiO 2 suspended systems when required.
Characterizations of the photo-irradiated solutions and filtrates were done with a spectrophotometer (JASCO V-650) and Dynamic light scattering (DLS), whereas the residues on membrane filters were characterized with different analytical techniques such as X-ray Diffractometer (XRD), Raman spectroscopy, Scanning Electron Microscopy (SEM) and Transmission Electron Microscopy (TEM).
Under DLS (Malvern Instruments Ltd, UK) studies, the gold nanoparticles (AuNP) over TiO 2 generated in photolysis were found highly poly-disperse in nature with average particle size of 100 nm. X-ray diffraction (XRD) measurement was carried out with Philips diffractometer (X'Pert PRO, PANalytical, Netherland) using Ni filtered Cu Kα radiation of 1.54 A o (~8 keV energy). XRD patterns of unknown samples were generally confirmed from the comparison of available data in literature and/or in library available in the software. Raman spectra were recorded at room temperature within the spectral range of 200-1800 cm -1 using the 633 nm line from a HeNe laser with a chargecoupled device (CCD) (Synapse, Horiba Jobin Yvon) based monochromator (LabRAM HR800, Horiba Jobin Yvon, France). The data presented herein are free from the background interference. Furthermore, morphological characterization/images of the materials were obtained with JEOL JSM-T330 Scanning Electron Microscope (SEM) and Carl Zeiss LIBRA 120 kV (for TEM).

RESULTS AND DISCUSSIONS
Eosin (C 20 H 6 Br 4 O 5 Na 2 ), a red fluorescent dye possesses chemical structure as shown in Scheme 1. The mineralization of eosin under the study along with the morphological characterization of in situ generated materials is discussed below: Scheme 1: Chemical structure of Eosin.

Photolysis of Eosin
The aerated 5×10 -5 M eosin in aqueous solution was photo-irradiated at 350 nm light using abovementioned photo reactor up to 120 min. 5 mL of photoirradiated solutions were taken out at different time intervals and the UV-Vis absorption spectra recorded are shown in Figure 1a. The spectrum recorded for eosin in aqueous solution prior to photolysis exhibits four peaks with absorption maxima at 255, 300, 340 and 516 nm (Figure 1a spectrum 'a'). The peak at 516 nm possesses high absorption intensity, which is the prime reason for taking low concentration of eosin for this particular experiment. With respect to photoirradiation time, the absorbance at 516 nm was reduced as shown in the inset figure. This change is rather low indicating the photo-degradation/photodecoloration of eosin was less under this condition. It is important to note that in this particular study the mineralization (transformation of eosin into CO 2 and water) should be negligible (discussed in the later section). Moreover, the decrease in absorbance at 516 nm is possibly due to the breaking of some of the weak chromophoric bonds of eosin molecule. This leads to ~ 20% of original color fading in 120 min photolysis. Furthermore, at other peaks positions (255, 300 and 340 nm), the changes in absorbance due to photolysis were quite low, hence are not considered for analysis.

Photolysis of Eosin in Presence of TiO 2
Similarly, photo-mineralization experiments were carried out in aerated 2×10 -4 M eosin solutions in water containing 0.1% TiO 2 as suspension using 350 nm photolight up to 120 min with constant stirring conditions. The preference of 0.1% w/v TiO 2 in the present study was confirmed earlier for requisite TiO 2 amount for maximum photo-catalytic product yield [11]. The 5 mL photo-irradiated samples were taken out at different time intervals and filtered using membrane filter and recorded absorption spectra of the filtrate solutions. The spectra recorded (not shown) exhibit abnormal behavior due to adsorption of eosin onto TiO 2 , therefore we have tried to separate TiO 2 by centrifugal method (another separation technique) and the absorption spectra of the superannuated solutions were also recorded. In this method too it was difficult to measure eosin accurately; as a part of eosin adsorbed on TiO 2 , moreover, the report on eosin adsorption in such system is available elsewhere [12]. Eventually, KMnO 4 demand method (discussed above) was adopted to estimate TOC content in eosin containing samples under different experimental conditions. The mineralization of eosin in TiO 2 containing systems interprets the decrease in TOC percentage (curve c) with respect to photolysis time as observed is shown in Figure 1b. The mineralization trend observed was relatively faster as compared to the mineralization of eosin without TiO 2 (inset of Figure 1a, which was generated using 5×10 -5 M eosin) and without photolysis of the same system (curve b Figure 1b). The curve 'a' in Figure 1b (Figure 2a), which is due to the absorption of eosin (<2 min) for the initial experimental solutions and later (>5 min photolysis) due to AuNP formation revealing the active interference of AuNP in eosin analysis at λ = 516 nm. To overcome the eosin analysis problem through absorption measurement, KMnO 4 demand method (discussed above) was adopted to determine the TOC contents. The results obtained are shown in Figure 1b

Characterization of In Situ Generated Materials
As mentioned above, the photo-catalytic mineralization of eosin in presence of Au 3+ or pregenerated AuNP follows similar degradation kinetics, which insists the detailed study of TiO 2 catalyst and its' modified materials while doing photolysis. The filtrates, superannuated solutions and residues characterized by different techniques under the study are discussed below.
The changes in colors of the solutions and the residues on membrane filters from orange to dark purple (Figure 2b) revealed the degradation of eosin as well as the formation of AuNP. The size of AuNP produced during the study was ~ 100 nm as determined in DLS experiments with superannuated AuNP samples (wherein TiO 2 was removed through centrifuge).  ) were identical to that of the anatase TiO 2 [13,14]. It is also noticed from Figure 3 that, the XRD patterns of TiO 2 remained unaffected in eosin-TiO 2 systems containing Au 3+ . However, for 1 and 20 min photoirradiated eosin-TiO 2 -Au 3+ samples, the considerable change in peaks intensities at 2θ 38 o and 62 o (see Figure 3) is due to AuNP formation [14,15]. It is important to note that with formation of AuNP the spectral peaks tends to be broader and the broadness increases with increase in AuNP concentrations [16,17]. The peak intensity at 2θ = 25 o and 38 o decreased considerably (spectrum c Figure 3) when the samples were photo-irradiated for 1 min. This was decreased drastically (spectrum d Figure 3) when samples were photo-irradiated further to 20 min, hinting the incorporation of AuNP onto TiO 2 . Similarly the Raman scattering spectra recorded for all these above samples (residues on membrane filters) are shown in Figure 4. The background membrane filter (Raman spectrum not shown) did not show any active peak within 200-1000 cm -1 spectral region.
Moreover, in presence of TiO 2 (Figure 4 spectrum 'a') three distinct peaks were observed below 1000 cm -1 region (640, 517 and 397.6 cm -1 ), which are quite similar to previous report [18]. The presence of eosin and Au 3+ ions did not lead to any significant change in spectral properties of TiO 2 (spectrum 'b') within 1000 cm -1 region except there is decrease in peaks intensities. Moreover, in presence of two different concentrations of AuNP (produced due to photoirradiation for 1 & 20 min), the peaks de-shaping (broadening, and blue shifting to 630, 510 and 394 cm -1 ) within 1000 cm -1 were observed, which is due to the active interaction between TiO 2 and AuNP (spectra c & d) that changed the vibration & structural properties of TiO 2 [18]. The high content of AuNP in 20 min photoirradiated sample was exhibited significant peak shifting & broadening in contrast to 1 min photolysis sample. SEM images (Figure 5a) exhibit larger size (~100 nm) and morphological change (coagulated) in TiO 2 AuNP materials (Figure 5a image 2) than in the plain TiO 2 (Figure 5a image 1). Furthermore, in Figure 5b

Eosin-• OH/ N 3 • Reaction Intermediates
To understand the eosin oxidative degradation mechanism through hole (considered as • OH [19,20]) initiative reactions, the pulse radiolysis studies were revisited on the reactions of eosin with • OH/N 3 • radicals at pH 6.8. The experimental set up for pulse radiolysis study in this institute has been reported elsewhere [21] in which 7 MeV electron beam of 200 ns pulse duration (dose rate 20 Gy per pulse determined as described elsewhere [22]) was used for sample irradiation. In water/aqueous medium, the primary species generated due to interaction of high-energy radiation (electron beam in the present case) in pico second time and diffused homogeneously throughout the medium within 0. allowing the radiolytically generated • OH in the system to interact with solute available (eosin in this case). In presence of NaN 3 under identical conditions, N 3 • was generated through N 3 -+ • OH à N 3 • + OH - [23]. The selection of • OH and N 3 • under the study was made because of their different reaction nature, former undergoes both addition and electron transfer reactions while the later reacts mainly through electron transfer type mechanism. The reactions of • OH and N 3 • with eosin studied separately are discussed below.
The spectrum 'a' in Figure 6 shows the transient absorption spectrum obtained in eosin-• OH reaction at Similarly, the spectrum 'b' in Figure 6 depicts the transient absorption spectrum recorded during eosin-N 3 • reaction. This spectrum exhibits only one peak with absorption maximum at 450 nm along with a strong bleaching at 510 nm. The reaction rate for N 3 • reaction with eosin determined from the formation time profiles of transient intermediates absorption at 450 nm was found to be 3.0x10 9 M -1 s -1 , which is close to • OH reaction with eosin evaluated using 450 nm time profiles.
On comparison of spectra a & b in Figure 6 it was observed that in • OH reaction there were two different intermediates: one is responsible for the absorption around 400 nm and other possesses absorption in 600 nm region. In N 3 • reaction the absence of 600 nm absorbing species reveals that the formation of 600 nm absorbing species was exclusively due to • OH addition and/or its subsequent reaction intermediates originated from eosin-• OH adduct species. The percentage of two different species generated in • OH reactions with eosin was evaluated to be 25% radical cation (450 nm absorbing species) and 75% • OH adduct species (600 nm absorbing species). This was evaluated based on the comparison of the absorbance values obtained from equal quantity of radicals ( • OH/ N 3 • ) generated in two different systems (G-values for • OH/N 3 • = 5.5 [5] at identical conditions. The possible reactions are: The following interpretations are therefore made: in presence of TiO 2 the mineralization of eosin was more as compared to only eosin systems where photo colorfading took place to a lesser extent as breaking of some weak chromophoric bonds occurred. The mineralization in aerated TiO 2 systems was more because the oxygen presence in the system holds dual role: i) as an electron scavenger: a part of photogenerated e was scavenged by O 2 (preventing recombination reaction (backward reaction 1) resulting more hole available for oxidation of eosin. ii) Secondly, the generated eosin-hole/ • OH reaction intermediate species react with available oxygen undergoing subsequent mineralization through oxy/hydroxy intermediates. It is noteworthy to include at this juncture that the radical cations in general are less/negligible reactive to oxygen [26]. Hence, the eosin-• OH adduct and/or its' associated intermediates are responsible for eosin mineralization.
In aerated Au 3+  Nevertheless, in metal ions containing systems, after formation of metal nanoparticles (as soon as AuNP color appears) the systems behave both as metal doped as well as metal ions containing systems, wherein both metal ions and its nanoparticles play significant role in eosin mineralization. This happens only when the photolysis was kept on, because after photolysis (photolight off condition) the colors of nanoparticles remained steady, indicating the negligible contribution of eosin mineralization due to thermal reaction of the nanoparticles with eosin. It is noteworthy to include that the active role of various intermediates starting from reduced metal ions/metal atom to oligomers/small aggregates of sub-nano-or nano-sized gold metal particles was significant in eosin mineralization. The possible photo-catalytic reactions taking place during mineralization of eosin are shown in Scheme 2.

CONCLUSION
Enhanced mineralization of eosin has been demonstrated under the study in presence of both Au(III) ions and its' in situ generated AuNP in oxygen containing systems. The strong interaction of AuNP with TiO 2 photo-catalyst was observed in Raman shifting during material characterization. The reaction intermediates both eosin-• OH/hole adduct as well as radical cation (differentiated in pulse radiolysis studies) were generated initially in mineralization process, wherein eosin-• OH/hole adduct is probably undergoing mineralization in presence of oxygen on photoirradiation. In presence of Au 3+  ) as compared to sole TiO 2 systems (0.004 min -1 ), which is due to the contribution of Au 3+ and the in situ generated AuNP in organics degradation process. This finding not only contributes in updating the eosin mineralization method, but this methodology also evolves as a convenient technique for their utilization in pollution control.