The residual dyes from textile and printing industries are considered a wide variety of organic pollutants introduced into the water resources. There are ten thousand types of textile dyes with an estimated annual production of seven hundred thousand metric tonnes commercially available worldwide. Thirty percent of these dyes are used more than one thousand tonnes per year, and ninety percent of the textile products are used at the level of one hundred tonnes per year approximately. About 10% - 25% of textile dyes are lost during the dyeing process whilst 2% - 20% of them are directly discharged as aqueous effluents in different environmental components   .
The discharge of dye-containing effluents into the wastewater system is undesirable since many of these are toxic, mutagenic and/or carcinogenic to living beings  . Furthermore, they affect biodegradation, light penetration and photosynthesis producing imbalance in the ecosystem. For instances, methylene blue, organic dye usually used in cotton and wool manufacturing, can cause serious health problems for human such as vomiting, hard breathing and mental disorder. The absence of effective treatments results in sustaining some components in the environment for a long period. For instance, the half-life of hydrolysed Reactive Blue-19 is ~46 years at pH = 7 and 25˚C  . Consequently, treatment methods such as filtration, chemical precipitation, electrolysis, sedimentation, oxidation, photo degradation and adsorption are essentially required.
Adsorption is found to be superior compared to the other methods in term of initial cost, simplicity of construction and insensitivity of toxics. Thus, a wide variety of natural and synthetic materials such as activated carbon, peat, various silica, activated clay, banana pith, natural manganese mineral, oil ash, goat hair, alum sludge, natural zeolite, mixtures of flash, soil and semiconducting oxides have been investigated as dye adsorbents. Among semiconducting oxides, Perovskite type oxides have been proven to be interest due to their efficiency in adsorbing and degrading recalcitrant organic compounds  .
Perovskite type oxides with the general formula ABO3 have been the focus of numerous studies. The major interest in such materials typically stems from their potential applications in the modern chemical industry such as catalysts, sensors and optical devices  . The Perovskite type oxides exist in a variety of compositions and structures each associated with unique physical properties. They contain two different cations, each surrounded by oxygen anion. The larger cation which can be alkali, alkaline earth, or rare earth has, ideally, dodecahedral symmetry, whilst the smaller cation which can be a transition metal ion, is six coordinate. Perovskite can accommodate different combinations of cations, so long as the crystal charge is neutral  .
This work investigates the removal of Aniline blue (AB) from aqueous solutions using the A-site doped perovskite Ce1−xBixCrO3 (x = 0, 0.5, 1). The impact of the replacing of Ce3+(4f1) with Bi3+(6s2) has been addressed to establish the role of the A-site cation in controlling the adsorption properties of the oxides. The incorporation of a cation with a stereo chemically active lone pair of electrons onto the A-site is known to have a significant effect on the structure and properties of perovskite. This has been explored in the double perovskite Bi2FeMoO6. It is postulated that the 6s2 electrons from the Bi3+ ions influences the structure and properties  .
Aniline Blue is typically organic soluble compound utilized in textile industry for dyeing of nylon, wool, silk and cotton. It belongs to triphenyl methane class of dyeing which a central carbon atom is bonded to two benzene rings and one p-quinoid group. Aniline Blue is acidic dye and has long residence time in water  . Thus, the presence of such pollutant in industrial discharge water is considered to have numerous impacts on the environment. The removal of industrial dyes from wastewater is a crucial process, from both economic and environmental points of view. It is thought that employing mixed metal oxides such as Ce1−xBixCrO3 in removal of organic dyes can be a good replacement with other high cost adsorbents due to its high accessibility and low cost.
2.1. Sample Preparation
To prepare the samples studied in this work the Sol-Gel method was utilized. Appropriate amounts of Ce(NO3)3∙6H2O (99.9%, BDH), Cr(NO3)3∙6H2O (99.9%, MERCK) and Bi(NO3)3 (99.9%, MERCK) were mixed and dissolved in 25 ml of 0.1 M nitric acid (65%, CODEX). The mixture was then added to a citric acid solution (99.7%, BDH) in the mole ratio 1:2 (oxide/acid) and heated on a hotplate at 200˚C until the gel turned ashes. The final products were ground using a mortar and pestle techniques and heated in an alumina crucible within a commercial furnace at 250˚C for 12 hrs and then reground and heated in stages at 400˚C for 12 hrs and 800˚C for 48 hrs.
The phase composition and purity of the samples was determined from X-ray diffraction. The X-ray diffractometer used was Philips PW 1800 X-Ray generator located at the Libyan Oil institution, Tripoli, Libya with a copper tube (Cu-Kα1 radiation), having a wavelength of 1.5406 Å. The operating voltage was 40 kV and the current was 30 mA. The samples were measured in flat plate mode at room temperature with a scan range of 10˚ < 2θ < 80˚ and a scan length of 10 mins were used. Diffraction patterns obtained from the Inorganic Crystal Structure Database (ICSD) were used for the comparison with obtained products.
The absorbance of solutions was determined using ultraviolet visible spectrophotometer (UV/V is, model Spect-21D) and (190 - 900 Perkin-Elmer) at maximum wavelength of absorbance (580 nλ). The concentrations of solutions were estimated from the concentration dependence of absorbance fit. The pH measurements were carried out on a WTW720 pH meter model CT16 2AA (LTD Dover Kent, UK) and equipped with a combined glass electrode.
2.3. Batch Mode
Batch mode removal studies were carried out by varying several parameters such as contact time, pH, temperature and mass of prepared oxide (adsorbent). Essentially, a 50 ml of dye solution with concentration of 50 ppm was taken in a 250 ml conical flask in which the initial pH was adjusted using HCl/NaOH. Optimized amount of adsorbent was added to the solution and stirred using magnetic stirrer for specific time.
3. Result and Discussions
3.1. Characterization of Oxides
The heating regime described above produced crystalline, green coloured samples. X-ray diffraction measurements showed the Ce0.5Bi0.5CrO3 oxide was isostructural with undoped CeCrO3 and have an orthorhombic (Pnma) structure whereas the BiCrO3 oxide has defected rhombohedra perovskite-type structure in space group R3C. Figure 1 illustrates the XRD patterns of the three oxides calcined at 800˚C. The Average Crystallite size Dp, Line broadening in radians β, Bragg angle θ, X-ray wavelength λ and lattice strain estimated from X-ray diffraction data are summarised in Table 1.
Figure 1. XRD patterns of CeCrO3, Ce0.5Bi0.5CrO3 and BiCrO3 calcined at 800˚C.
Table 1. Average crystallite size Dp, Line broadening in radians β, Bragg angle θ, X-ray wavelength λ, lattice strain estimated from X-ray diffraction data.
The crystallite size can be calculated using sheerer formula  :
Unexpectedly, Ce0.5Bi0.5CrO3 displayed the lowest crystallite size, whereas the CeCrO3 had the highest crystallite size in the series. The decrease in crystallite size of Ce0.5Bi0.5CrO3 and BiCrO3 is inconsistent with the relative ionic size of the Bi3+ (8 coordinate ionic radius, 1.17 Å) and Ce3+ (1.14 Å) cations  . The decrease in crystallite size of Bi doped oxides can be attributed to the high covalent bonding character of Bi ions, with small displacement of the A-site cations enhanced by the covalence   .
3.2. Batch Mode
3.2.1. Effect of Contact Time
The removal percentage of dyes over the adsorbents can be calculated as:
where R% is the removal percentage, Ci = 50 ppm is initial concentration of dye solutions, Ct is the concentration of dye at contact time estimated from the concentration dependence of absorbance fit. The effect of contact time on the AB removal was observed at the range of (0 - 150 min). Figure 2 shows the time dependence of AB removal at room temperature. There is no finite time was observed for the dye removal up to 150 min. The removal of dye increases as the contact time increased. The removal of AB on the surface of CeCrO3, Ce0.5Bi0.5CrO3 and BiCrO3 at 150 min were found to be 58.30%, 59.25% and 41.72% respectively.
Figure 2. Time dependence of AB removal at room temperature. The volume, concentration and pH of the AB solution are 50 ml, 50 ppm and 4.3 respectively. The time dependent variation of the removal is fitted as labelled on the figure.
3.2.2. Effect of Adsorbent Mass
The amount of the dye adsorbed by one gram of the oxides (Q) was calculated as following:
where t = 150 min is the contact time, V = 50 ml is the volume of AB solution and W is the mass of oxides. As shown in Figure 3, Q decreases as the mass of adsorbents increased. The maximum capacity of adsorbent Qmax can be estimated from the intercept of the liner fit of 1/Qt at Y axis. Ce0.5Bi0.5CrO3 displayed the highest value of Qmax (779.67 (6) mg/g) whereas BiCrO3 exhibited the lowest value of Qmax (440.18 (4) mg/g). Qmax for CeCrO3 is 705.45 (1) mg/g. This result reflects the importance of the element composition and crystal structure of the oxides. The amount of AB adsorbed by CeCrO3 which has orthorhombic lattice and the largest crystallite size (87.14 nm) is much higher than that adsorbed by BiCrO3 which has rhombohedral lattice and crystallite size equal to 48.36 nm. The observed increase in the removal using Ce0.5Bi0.5CrO3 which has orthorhombic lattice and the smallest crystallite size (36.26 nm) suggested an enhancement in the adsorption properties occurs as result of the substitution of Bi3+ into the oxide plus the decrease in crystallite size. The decrease in crystallite size leads to an increase in the surface area of particles.
3.2.3. Effect of pH
To study the effect of pH, experiments were carried out at various pH values, ranging from 2 to 6 for constant dye concentration (50 ppm) and adsorbent mass (0.1 g). Figure 4 shows the removal of AB as a function of pH. The removal of AB increases as pH decreased. The highest removal of AB for CeCrO3,
Figure 3. Effect of mass adsorbent on the removal. The time, volume, concentration and pH of the AB solution are 150 min, 50 ml, 50 ppm and 4.3 respectively.
Figure 4. Effect of pH on the removal of AB. The time, volume and concentration of the AB solution are 150 min, 50 ml and 50 ppm respectively.
Ce0.5Bi0.5CrO3 and BiCrO3 are found to be 87.36%, 86.00%, 81.28% respectively at pH = 2. The oxides displayed almost an equal efficiency of removal (~53%) at pH = 6. BiCrO3 exhibited lower percentages of removal compare to the two other oxides. The interpretation of pH effects on the efficiency of the adsorption process is a very difficult task, because of its multiple roles. It is related to the acid base property of both the metal oxide and the organic dye. The adsorption of water molecules at metal sites is followed by the dissociation of OH− groups, leading to coverage with chemically equivalent metal hydroxyl groups (M-OH). Due to amphoteric behaviour of both the metal oxide and the organic dye, the equilibrium reactions below are considered. The electrostatic interactions between the positive catalyst surface and dye anions leading to strong adsorption of the last on the oxide support.
3.2.4. Effect of Temperature
Temperature has an important impact on the adsorption process. An increase in temperature helps the reaction to compete more efficiently with e-/H+ recombination. The removal of AB was investigated at 25˚C, 40˚C, 60˚C and 100˚C. The obtained results are illustrated below in Figure 5. The removal of AB decreased as temperature increased. For instance, the removal of AB using Ce0.5Bi0.5CrO3 decreased from 61.44% at 25˚C to 30.06% at 100˚C. The decrease in percentages of adsorption with rise in temperature may be due to desorption caused by an increase in the thermal energy. Higher temperature induces higher mobility of
Figure 5. Effect of temperature on the removal of AB. The time, volume and concentration of the AB solution are 150 min, 50 ml and 50 ppm respectively.
active sites of adsorbents and the surface area is decreased by increase the temperature.
The energy of activation (Ea), was calculated from the Arrhenius plot of lnR vs 1000/T (Figure 6). Arrhenius plot shows that the activation energies of the removal of AB using CeCrO3, Ce0.5Bi0.5CrO3 and BiCrO3 are negative and equal to −22.58 ± 1, −8.53 ± 1 and −22.21 ± 1 kJ∙mol−1 respectively. This result is contrary to normal expectations, and is a consequence of the competition between two effects: the increase of intrinsic kinetics with temperature, and the decrease of adsorption strength and the concentration of active intermediates with temperature  . The adsorption of AB dye can be described as barrierless reaction, in which the reaction proceeding relies on the capture of the molecules in a potential well. Increasing the temperature leads to a reduced probability of the colliding molecules capturing one another (with more glancing collisions not leading to reaction as the higher momentum carries the colliding particles out of the potential well), expressed as a reaction cross section that decreases with increasing temperature. Such a situation no longer leads itself to direct interpretations as the height of a potential spot  .
The removal of Aniline blue from aqueous solution by the mixed metal perovskite Ce1−xBixCrO3 has been studied. In general, the amount of AB adsorbed by Ce0.5Bi0.5CrO3 is higher than those adsorbed by CeCrO3 and BiCrO3. The study showed no finite time for the AB removal up to 150 min. The AB removal increases as the mass of the oxides increases. The adsorption of AB was
Figure 6. Arrhenius plot of ln(Removal %) vs 1000/T(K−).
temperature and pH dependent. The three oxides showed maximum removal efficiency of AB at 25˚C and pH = 2.0. The adsorption capacity of AB from water was 779.67, 705.45 and 440.18 mg/g using Ce0.5Bi0.5CrO3, CeCrO3 and BiCrO3 respectively. The study indicated that doping with Bismuth ions into CeCrO3 can successfully enhance the adsorption properties of the oxide.
Authors thank the Libyan Oil institution for the XRD measurements. Many thanks for Prof. Abd El-Salam. M. Elmehob for the paper proofreading.
 Baban, A., Yediler, A. and Ciliz, N.K. (2010) Integrated Water Management and CP Implementation for Wool and Textile Blend Processes. CLEAN—Soil, Air, Water, 38, 84-90.
 Robinson, T., McMullan, G., Marchant, R. and Nigam, P. (2001) Remediation of Dyes in Textile Effluent: A Critical Review on Current Treatment Technologies with a Proposed Alternative. Bioresource Technology, 77, 247-255.
 Carmen, Z. (2012) Textile Organic Dyes—Characteristics, Polluting Effects and Separation/Elimination Procedures from Industrial Effluents—A Critical Overview. InTech, Janeza Trdine Rijeka, Croatia.
 Lan, Y., Feng, X., Zhang, X., Shen, Y. and Wang, D. (2016) Effects of Bi Doping on Structural and Magnetic Properties of Double Perovskite Oxides Sr2FeMoO6. Physics Letters A, 380, 2962-2967.
 Langford, J.I. and Wilson, A.J.C. (1978) Scherrer after Sixty Years: A Survey and Some New Results in the Determination of Crystallite Size. Journal of Applied Crystallography, 11, 102-113.
 Takagi, S., Subedi, A., Cooper, V.R. and Singh, D.J. (2010) Effect of A-Site Size Difference on Polar Behavior in MBiScNbO6 (M=Na, K, and Rb): Density Functional Calculations. Physical Review B, 82, 134108.
 Corker, D.L., Glazer, A.M., Kaminsky, W., Whatmore, R.W., Dec, J. and Roleder, K. (1998) Investigation into the Crystal Structure of the Perovskite Lead Hafnate, PbHfO3. Acta Crystallographica Section B, 54, 18-28.
 Mozurkewich, M. and Benson, S.W. (1984) Negative Activation Energies and Curved Arrhenius Plots. 1. Theory of Reactions over Potential Wells. Journal of Physics and Chemistry, 25, 6429.