Application of polyaniline impregnated mixed phase Fe2O3, MnFe2O4 and ZrO2 nanocomposite for rapid abatement of binary dyes from aqua matrix: response surface optimisation

ABSTRACT Wastewater loaded with toxic dyes causes serious environmental problems because of their high toxicity and needs to be treated before their disposal. In this research, mixed phase of Fe2O3, MnFe2O4 and ZrO2 nanocomposite was impregnated onto polyaniline to prepare a new nanocomposite (Fe-Mn-Zr/PANI-NC). The produced nanocomposite was utilised to adsorb methyl orange (MO) and eosin yellow (EY) dyes from binary dye solution. The response surface methodology (RSM) based optimisation showed that MO and EY dyes can be removed with ~98% efficiency under optimal experimental conditions. The adsorption process was quick in the wide solution pH range from 2.0 to 5.0 and the equilibrium was achieved within 5 min of reaction at solution pH 5.0. The adsorption process was mainly driven by strong electrostatic attraction and π-π interaction between dye molecules and Fe-Mn-Zr/PANI-NC. Kinetic study reveals the applicability of the pseudo-second-order and intra-particle diffusion model and the maximal adsorption capacity of Fe-Mn-Zr/PANI-NC was computed using the Langmuir isotherm model and found to be 217.39 mg/g for MO dye and 117.65 mg/g for EY dye. Therefore it can be concluded that Fe-Mn-Zr/PANI-NC can be a potential adsorbent for MO and EY dye remediation from wastewater due to high adsorption capacity, short equilibrium time, high reusability and adsorption favourability in a wide range of solution pH.


Introduction
The environmental integrity is in a great challenge due to frequent discharge of hazardous substances like dyes and pigments into the atmosphere [1].The textile, paper, plastics, leather, food and cosmetic industries are enormously increasing the environmental pollution through conscious and unconscious expulsion of their coloured effluents in the environment [2].The coloured effluents contain hazardous and toxic dyes which can immensely pollute the water bodies.The disposal of a very small quantity of dye causes severe environmental pollution to aquatic life which diminishes the efficacy of photosynthesis, as a result the water quality worsens, and aesthetic value of water deteriorates adversely [3].The major textile dyes are the azo dyes group with complex structure and non-biodegradable nature.These noxious pollutants can cause severe damage to human body by producing mutagenicity resulting to tumour and cancer [4].The active functional groups of dyes have the ability to interact with various active groups like SH − , -NH-, OH − , -CO-of fibre and get fixed.In an average, around 10-20% of colour dyes are discharged into the environment in course of synthesis and dyeing process as dyes are unable to fix with fibre due to saturation of active sites in fibre [5].Thus effluents containing contaminants with synthetic origin and obstinate nature are incalculably polluting the aquatic system [6].Methyl orange (MO) and eosin yellow (EY) are such anionic toxic dyes [7].
Nanoscale magnetic inorganic-organic hybrid composites have attracted considerable interest in the recent time research due to their wide range of potential applications [3,8,9].Organic materials with polymeric backbone of aniline, thiophene, pyrrole, o-aniside, α-napthylamine received significant attention as conductive polymer.Among these, polyaniline is one of the most promising materials owing to its ease of synthesis, unique electrochemical properties and facile doping possibility.Applicability of polyaniline in the field of wastewater treatment, energy storage, solar cell, light emitting diode, memory device, ion exchange materials and sensor materials are well established [10].But one of the major drawbacks of polyaniline is its poor mechanical strength which creates an impediment to become an efficient adsorbent.On the other hand inorganic metal oxide nanoparticles are interesting materials and draw much research attention due to their diversity in property [11].High mechanical strength, unusual particle size, anisotropic nature and high magnetic susceptibility introduce interesting and novel properties for such nanoparticles [12,13].But the problem associated with maximum metal oxide nanoparticles is that they exhibit their highest adsorption performance at highly acidic condition.Highly acidic effluent creates disposal problem and also increases the wastewater treatment cost.Impregnation of polyaniline on to metal oxide nano-particles improves their original paternal characteristics and also drives away the disadvantages of pure metal oxides or pure polymers.Deb et al. (2020) [14] reported that iron oxide/ polyaniline polymeric nanocomposite (Iron oxide-PANI-PNC) shows an enhanced adsorption capacity of 285.71 mg/g for Eriochrome black-T dye at almost neutral condition.Similarly, Liang et al. 2018 [15] had demonstrated the adsorption of alizarin red-S dye by NiFe 2 O 4 /PANI magnetic composite near to neutral pH condition.In addition to impregnation of polymeric materials on the surface of metal oxide nanoparticles, the novelty of this research also lies in assimilation of more than one metal oxide nanoparticles in lieu of single metal oxide nanoparticles.Incorporation of multi metal oxides not only improves the characteristics of paternal metal oxide but also ameliorate the synergistic effect of other available metal oxides.Bhowmik et al. (2018) [16] reported the fabrication of mixed phase Fe 2 O 3 /Mn 3 O 4 nanocomposite (MIMO) and its adsorptive application for dye loaded wastewater treatment.They reported that MIMO has shown enhanced performance for wastewater treatment as compared to the pure Fe 2 O 3 and pure Mn 3 O 4 nanoparticles.
In view of above, the overall aim and objective of this study were to investigate the adsorptive removal of toxic anionic dyes (MO and EY) onto a novel organometallic nanocomposite (Fe-Mn-Zr/PANI-NC) from aqueous media.In the present work, Fe-Mn-Zrtrimetal oxide polyanilinenanocomposite (Fe-Mn-Zr/PANI-NC) was synthesized by in situ oxidative sono-chemical polymerisation technique.The composite was characterised by X-ray diffraction (XRD), scanning electron microscopy (SEM),energy-dispersive X-ray (EDX), Fourier transforminfrared (FT-IR) spectroscopy, transmission electron microscopy (TEM) and Brunauer-Emmett-Teller (BET) analysis to study their crystalline nature, morphology, surface characteristics and specific surface area [17][18][19].Moreover to maximise the removal of MO and EY dye from binary dye matrix with least number of experiments, lesser amount of chemicals and energy,a powerful optimisation technique, response surface methodology (RSM) associated with central composite design (CCD) has been applied [20,21].Influence of various dye removal controlling parameters like solution pH, sonication timings, initial adsorbent dose and initial dye concentrations was systematically explored.The adsorption experimental data were also used to explore the adsorption mechanism and nature of adsorption through kinetic and isotherm modelling study.

Ultrasonic supported dye adsorption
Batch mode adsorption process was used to explore the ultrasound-assisted removal of dyes from their binary solution using Fe-Mn-Zr/PANI-NC as adsorbent.Batch adsorption studies were carried out in 100 mL beaker containing MO and EY dye binary solution.Appropriate amount of Fe-Mn-Zr/PANI-NC was loaded in the beaker and sonicated, until equilibrium was attained.At the end of the experiment, the treated dye solution was separated by centrifugation and analysed by UV-Vis spectrophotometer (Hach-DR5000) to calculate the available dye after treatment.Bath sonicator (Rivotek, India) of 1.5 L capacity was used to induce ultrasonic irradiation and 30 kHz was the operating frequency of the reactor.In order to study the sorption behaviour of Fe-Mn-Zr/PANI-NC, various process parameters like solution pH, sonication time, adsorbent dose and initial concentration of binary dye mixture were varied in the range of 2.0 to 9.0, 0.5 to 15.0 min, 0.1 to 0.8 g/L and 5 mg/L to 30 mg/L, respectively.All the experiments have been carried out at room temperature (30 ± 3°C).
The concentration of MO and EY dye from binary dye matrix was calculated by following equations [19]: In the Equations ( 1) and (2), C A and C B represent the MO and EY dye concentrations, respectively, after the reaction; K A1 and K B1 are the calibration coefficients for pure components 'A' (MO dye) and 'B' (EY dye) at the maximum wavelength of λ 1max = 464 nm.Similarly K A2 and K B2 are the calibration coefficients for pure components A (MO dye) and B (EY dye) at the maximum wavelength of λ 2max = 517 nm [19].

Synthesis of Fe-Mn-Zr metal oxide/PANI nanocomposite (Fe-Mn-Zr/PANI-NC)
Firstly, the Fe-Mn-Zr metal oxide nanocomposite has been synthesised by simple chemical precipitation method as reported by Singh et al. (2018) [19].The schematic diagram of the synthesis process has been depicted in Figure S1.Secondly, aniline monomer was chemically polymerised onto Fe-Mn-Zr metal oxide nanoparticles.In a typical experiment, 6.0 ml of aniline was added to 94 ml of 1.0 M HCl solution making the total volume of 100 ml.Thereafter, the aniline contained HCl solution was taken in a 500 ml beaker and put into bath sonicator.With aniline solution, the pre-synthesised Fe-Mn-Zr nanomaterial was introduced to the beaker and proper dispersion was achieved by ultrasonication of the mixture for 5 min.Next, APS solution (12.25 g of APS powder was dissolved in 150 mL of double distilled water) was added slowly in aniline and Fe-Mn-Zr nanoparticles mix solution and the mixture was kept in ultrasonication for 15 min.Thereafter greenish colour precipitate was observed in the beaker confirming the formation of PANI.In order to complete the reaction properly, the beaker was kept in undisturbed condition for 48 hours.Thereafter the precipitate was collected, and washed with 0.5 N HCl solution to remove unreacted aniline and other impurities until colourless solution was obtained as filtrate.Then the precipitate was washed with distilled water, till the pH of the filtrate becomes near to neutral condition.Finally the obtained material was dried in oven and grinded properly to obtain Fe-Mn-Zr/PANI-NC for further study.

Response surface methodology based optimisation
Response surface methodology (RSM) with CCD was implemented to ascertain the effect of four independent variables on the adsorption of MO and EY dye by Fe-Mn-Zr/PANI-NC.The independent variables such as Fe-Mn-Zr/PANI-NC dose, sonication time, MO dye concentration and EY dye concentration were designated as X 1 , X 2 , X 3 and X 4 , respectively.The experimental data were designed and analysed by statistical software Design Expert (Version:8.0.6.1).A four-factor five-levels ( À α, −1, 0, +1, þ α) CCD was employed to determine the general equation which provides mathematical relationship between the independent variables and the output [22,23].The generalised second-order polynomial equation of quadratic order can be expressed as: In Equation ( 3), β 0 is the constant coefficient, β i is the first-order coefficient, β ii is the quadratic coefficient, β ij is the interaction coefficient, X i and X j are input variables, and k is the number of variables.

Characterisations of Fe-Mn-Zr/PANI-NC
The XRD pattern of Fe-Mn-Zr/PANI-NC has been shown in Figure 1(a), which describes various crystalline phase of the composite.The prominent peaks of the composite appears at 25.6°, 30.20°, 35.5°, 49.50° and 54.20° diffraction angle.The characteristic peaks occurred at 25.6° corresponds to (200) miller plane of polyaniline [24].The peak appears at 30.20° attributed to the tetragonal phase of ZrO 2 [25].Another diffraction peak at 35.5° can be indexed to (311) miller plane of MnFe 2 O 4 nanoparticles [26].The peaks at 2θ values of 49.50°, and 54.20° Correspond to the (024) and ( 116) crystal planes α-Fe 2 O 3 nanoparticles, respectively (JCPDS card number 33-0664) [27].It is apparent from FESEM image (Figure 1(b)) that the nanocomposite consists of nano rods having diameter in the range of 20-70 nm with some agglomeration of nanoscale particles due to its paramagnetic behaviour.The EDS spectrum of Fe-Mn-Zr/PANI-NC depicts the elemental ingredients of the composite as shown in Figure 1(c), which confirms that the composite mainly contains Fe, Mn and Zr elements along with the fundamental peaks of carbon and nitrogen from PANI matrix.
The FTIR spectrum of Fe-Mn-Zr/PANI-NC is illustrated in Figure 2(a).FTIR spectral data analysis confirms the nature of bonding mood that comprising the synthesise material [28].The spectral bands at 3473 cm -1 and 1633 cm -1 are mainly due to O-H stretching and O-H bending vibrations from water molecule present in the sample.The peak at 1094 cm −1 ascribed to bending vibration of benzene ring [29].The peaks observed at 1292 cm −1 and 1486 cm −1 assigned to C = C stretching vibration of the benzenoid ring, and C-N stretching vibration of PANI matrix, respectively [30,31].The broad peak in the range of 566 to 894 cm −1 can be ascribed to the metal oxide bending vibrations from Fe 2 O 3 , MnFe 2 O 4 and ZrO 2 nanoparticles.The N 2 adsorption-desorption study and the pore size distribution plot of Fe-Mn-Zr /PANI-NC are depicted in Figure 2(b,c), respectively.The surface area of the Fe-Mn-Zr/PANI-NC was found in order of 78.847 m 2 /g. Figure 2(b) shows that the shape of the isotherm resembles type IV isotherm which indicates Fe-Mn-Zr/PANI-NC may be mesoporous in nature [32].The pore volume vs. pore diameter plot (Figure 2(c)) of Fe-Mn-Zr/PANI-NC shows that a major peak concentrated around 10-14 nm, this shows the composite's mesoporous nature.The pore volume of Fe-Mn-Zr /PANI-NC was calculated as 0.067 cc/g.The surface morphology and size of the Fe-Mn-Zr /PANI-NC were explored by transmission electron microscope (TEM) (Figure 2(d)).This TEM technique is adjuvant to examine the particle size, shape, porosity and surface structure of adsorbent [33].TEM imaging reveals the combination of nano rods and nanoscale particles with some agglomeration in Fe-Mn-Zr/PANI-NC which is in agreement with FESEM image (Figure 1(b)).

Influence of solution pH onto dye uptake efficiency from binary dye matrix
Solution pH is a key factor which influences the various functional groups of adsorbent surface as well as the surface charge of the adsorbents during adsorption process [34].The degree of ionisation of pollutants, solubility and colour intensity of dye molecule are also controlled by solution pH.The effect of solution pH in the range of 2.0 to 9.0 has been studied by adding 0.0125 g of Fe-Mn-Zr /PANI-NC to 50 mL of binary dye solution (30ppmMO and EY dye each) at room temperature.The impact of solution pH on the removal process of MO and EY has been described in Figure 3(a), which delineates that for both MO and EY dye, maximal dye removal was seen at solution pH 5.0.The maximum removal of ~96% and ~78% was achieved for MO dye and EY dye, respectively.Hence, the solution pH was optimised at 5.0 for all subsequent experiments like determining the effect of other reaction parameters, kinetic modelling, isotherm modelling and RSM optimisation experiments.In this study, maximum dye removal efficiency was observed at solution pH 5.0, which is quite supportive concerning to the real time application point of view, as efficient treatment efficiency at highly acidic condition (pH 2.0 or 3.0) may not be suitable for field application.Polyaniline in the composite state is better dispersible in aqueous medium [35].The charge defect state in the backbone of the polyaniline enriched with enhanced electrostatic potential distribution.The surface functionalization with p-type conductivity and consequent electrostatic force of attraction made the surface better adsorption active for anionic dyes.Thus the adsorption process was mainly driven by strong electrostatic attraction and π-π interaction between dyes molecules and Fe-Mn-Zr/PANI-NC.The probable interaction mechanism between Fe-Mn-Zr/PANI-NC and anionic binary dyes has been depicted in Figure 3(b).Similar mechanism has been reported for adsorption of EBT dye onto polymer functionalised ZnO nanocomposite [10] and adsorption of alizarin red S dye onto NiFe 2 O 4 /polyaniline magnetic nanocomposite [24].

Effect of sonication time on removal percentage of dye
Sonication is a process where sound waves are used to agitate particles in solution.During sonication, the particles in solution vibrate because they experience cycles of pressure, then generate microscopic vacuum bubbles and then collapse into the solution, called cavitation.These vibrations can disrupt molecular interactions among the molecules of solvent and breaking the clumps of particles thereby leading towards enhanced interaction.Thus combining sonication with adsorption process can reduce the equilibrium time significantly and may enhance the pollutant uptake efficiency.The effect of sonication time in sono-assisted adsorption of MO and an EY dye from binary dye solution is shown in Figure 4(a).Excellent removal efficiencies were observed for MO dye (98.81%) and EY dye (98.02%) within 5.0 min of sonication using adsorbent dose of 0.2 g/L, and solution of pH of 5.0.The above results suggest that application of ultrasound reduces the time span to achieve equilibrium state for any adsorption process and similar findings were also reported by other researchers [36].The reduced span of sonication time with excellent dye removal efficiency (~99%) indicates that the process is very economic with a great deal of competence.

Effect of Fe-Mn-Zr/PANI-NC dose on binary dye uptake efficiency
The consequence of Fe-Mn-Zr/PANI-NC dose on the adsorptive removal of MO and EY dye was investigated keeping initial concentration of both MO and EY fixed at 30 mg/L with solution pH of 5.0 at ambient temperature.The Fe-Mn-Zr/PANI-NC dose was varied from 0.1 g/L to 0.8 g/L and the obtained results have been depicted in Figure 4(b).It has been observed that when Fe-Mn-Zr/PANI-NC dose was varied from 0.1 g/L to 0.8 g/L with sonication time of 10 minutes, the dye removal efficiency increases with the increase in adsorbent dose.The removal efficiency for both the dyes MO (98.48%) and EY (97.02%) approaches towards saturation with a minimum Fe-Mn-Zr/PANI-NC dose of 0.5 g/L.However, further increase in Fe-Mn-Zr/PANI-NC dose did not show any considerable improvement in dye removal efficiency.With the increase of Fe-Mn-Zr/PANI-NC dose from 0.1 g/L to 0.5 g/L, the dye removal percentage also increases steadily.This may be due to increase in active adsorption sites of the adsorbent with the increase in Fe-Mn-Zr /PANI-NC dose.Similar results have been reported for Congo red dye adsorption by CaFe 2 O 4 nanoparticles [37] and amido black 10B dye adsorption by polyaniline/SiO 2 nanocomposite [31].However, adsorbent dose higher than the optimum level may result in overlapping of functional surface area which can reduce the effective surface area of adsorbent [37].This phenomenon can increase the diffusion path length for adsorbate and hinder the improvement in adsorption performance.Due to this phenomenon no significant improvement in MO and EY dye removal was observed beyond 0.5 g/L of Fe-Mn-Zr/PANI-NC dose.

ANOVA and other statistical analysis of RSM model
In RSM modelling, CCD was implemented to prepare the design of experiment with thirty experimental runs for examining the influence of four input parameters on two output parameters.The effective ranges of individual input parameters (Fe-Mn-Zr/PANI-NC dose, sonication time, MO dye and EY dye concentrations) were selected based on the previously conducted experimental results and their interactive effects were investigated on the removal of MO and EY dyes.The design of experiment comprising of thirty experiments with experimental and predicted MO and EY dye removal efficiencies are presented in Table 1.The polynomial equation between input parameters (Fe-Mn-Zr/PANI-NC dose, sonication time,MO dye and EY dye concentrations) and output parameters (MO and EY dye removal) can be expressed as: In Equations ( 4) and ( 5), the positive sign factors indicate the uprising of output, while the negative sign factors indicate inaction of the output.The experimental data for MO and EY dye removal were examined by analysis of variance (ANOVA) and the results are listed in Tables S1 and S2 of supplementary information.The developed regression model was analysed by ANOVA and the significance of the developed model was tested by 'p' values and 'F' values.The 'p' values of the models are less than 0.0001 (for both MO and EY dye) and also the corresponding 'F' values are high (60.08 for MO dye and 60.63 for EY dye), which indicate that the developed models are statistically significant [22].In addition, the calculated R 2 , adjusted R 2 and predicted R 2 values are also close to unity (more than 0.92) for both the developed models which indicate the high correlation between predicted and actual dye removal efficiency [23].

Three dimensional response surface plots and optimisation
Figure 5 shows the interactive effect of each pair of variables on dye removal percentage, keeping the other two input variables at a fixed level.The curvature nature of the plots confirms the type of interaction between the input variables on the output parameter.It is evident from Figure 5(a) that both sonication time and Fe-Mn-Zr/PANI-NC dose have positive correlation with the removal percentage of MO dye, so that the removal upsurges with escalating dose and time.has positive correlation with removal percentage of MO dye.Similarly, Figure 5(c) shows the positive and negative correlation of removal percentage of MO dye with sonication time and initial MO dye concentration, respectively.Similar interactive patterns were observed between input parameters and output parameters for EY dye adsorption also, which is shown in Figure 5(d-f).Figure S6 (a) and (b) shows the linear relationship between experimental dye removals with the RSM model predicted dye removals and very close agreements were observed for both MO dye (R 2 : 0.983) and EY dye (R 2 : 0.983) adsorption.This specifies normal distribution of error around the mean values which indicates that developed RSM models are capable to predict the experimental data accurately [27].Moreover, Figure S6(c) and (d) show that the deviation of the prediction is within ±3%, which endorse that models have approximated the experimental results with lesser deviation [30].
Thereafter, developed RSM model was used to optimise the experimental conditions to maximise the MO and EY dye removal efficiency.The optimisation results showed that more than 98% removal efficiency can be achieved for both MO and EY dye under optimal experimental conditions of Fe-Mn-Zr/PANI-NC dose: 0.4 g/L, initial MO dye concentration: 12 mg/L, initial EYdye concentration: 12 mg/L within 5 min of sonication time.

Kinetics modelling for MO and EY dye adsorption from binary dye matrix
Kinetic study helps to understand the underlying mechanism of any adsorption process [37].Hence, to investigate the adsorption mechanism of MO and EY dye onto Fe-Mn-Zr/PANI-NC three kinetic models were applied, namely pseudo-first-order kinetic model, pseudo-second-order kinetic model and intra-particle diffusion model.Linear fittings of adsorption experimental data for MO and EY dyes with abovementioned three kinetic models have been depicted in Figure S2 and Figure S3, respectively.The different kinetic parameters for three considered kinetic models determined from linear fitting of experimental data are shown in Table 2.
Table 2. General representation of kinetic models for MO dye adsorption (Fe-Mn-Zr/PANI-NC dose: 0.2 g/L; solution pH: 5.0; initial concentration of MO dye 15 ppm; reaction temperature: 30 ± 5°C) and EY dye adsorption (Fe-Mn-Zr/PANI-NC dose: 0.2 g/L; solution pH: 5.0; initial EY concentration 12 ppm; reaction temperature: 30 ± 5°C) onto Fe-Mn-Zr/PANI-NC.Considering the MO dye adsorption onto Fe-Mn-Zr/PANI-NC, the comparison between experimental adsorption capacity (Q e (exp)) and model predicted adsorption capacity (Q e ) indicates very high agreement for pseudo-second-order kinetic model.Moreover highest R 2 values are observed for pseudo-second-order kinetic model, suggests that this adsorption process mainly governed by the chemisorption.Similar phenomenon was observed for EY dye adsorption too.In case of intra-particle diffusion model, if the experimental data fitting line passes through the origin of the plot, then it indicates that the rate controlling step follows intra-particle diffusion model.But from Figure S2(c) and Figure S3(c), it is clear that no line is passing through origin, indicates that intra-particle diffusion is not the only rate controlling step.However, high R 2 values in linear fitting of experimental data with intra-particle diffusion model indicate the involvement of intra-particle diffusion in this adsorption process.

Adsorption equilibrium study of MO and EY dye from binary dye matrix
The adsorption equilibrium study is vital to unfold the interactive behaviour between adsorbates and adsorbent in any adsorption process [38].Three familiar adsorption isotherm models namely Langmuir, Freundlich and Temkin models were applied to analyse the adsorption equilibrium data [39].Linear fitting of MO and EY dye adsorption experimental data with above mentioned isotherm models have been depicted in Figure S4 and Figure S5, respectively.The isotherm parameters of three adsorption models obtained from MO and EY dye adsorption equilibrium onto Fe-Mn -Zr/PANI-NC are shown in Table 3.The linear regression coefficients depict that the Langmuir model (R 2 : 0.9933 and 0.9937) describe the processes better as compared to Freundlich (R 2 : 0.9918 and 0.9271) and Temkin (R 2 : 0.9847 and 0.9436) isotherm for MO and EY removal process.It confirms the suitability of Langmuir adsorption isotherm rather than Freundlich or Temkin adsorption isotherm for this adsorption process.Therefore the adsorption of MO and EY onto Fe-Mn-Zr/PANI-NC is more of monolayer adsorption rather than adsorption on a surface having heterogeneous  3).

Performance evaluation of Fe-Mn-Zr/PANI-NC
The performance of the Fe-Mn-Zr/PANI-NC for MO and EY adsorption was compared with other reported adsorbents in terms of adopted mixing methods, solution pH, contact time (min) and maximum adsorption capacity (mg/g).The comparison is shown Table 4 which illustrates that the adsorption equilibrium time for Fe-Mn-Zr /PANI-NC is only 5 min.However majority of reported adsorbents (multiwalled carbon nanotubes, modified coffee waste, haematite nanoparticles, γ-Fe 2 O 3 /SiO 2 /chitosan composite, Fe-Mn-Zr metal oxide nanocomposite, nano-sized chitosan blended polyvinyl alcohol) took much longer time (30 to 420 min) [19,[40][41][42][43][44][45] to reach the equilibrium.Thus quick attainment of equilibrium suggests the effectiveness of the proposed methodology in terms of treatment cost and energy consumption.
Considering solution pH, it can be seen that maximum reported adsorbents have shown highest dye removal at extremely acidic condition (solution pH: 2.0 to 3.0) which makes the process complicated for real field application.Moreover highly acidic effluent creates disposal problem, as it can cause serious damage to the aquatic ecosystem.On the contrary, Fe-Mn-Zr/PANI-NC could adsorb the MO and EY effectively at solution pH 5.0, which makes the process economical and suitable for field application.Some adsorbents like nano-sized chitosan blended polyvinyl alcohol [44] and CuO nanoparticles loaded on activated carbon [46] could remove the EY dye at solution pH: 6.0, but the their adsorption capacity (nano-sized chitosan blended polyvinyl alcohol:52.91 mg/g and CuO nanoparticles loaded on activated carbon:21.29 mg/g) for EY dye are not comparable with Fe-Mn-Zr/PANI-NC (117.65 mg/g).Finally, considering the maximum monolayer adsorption capacity, the prepared Fe-Mn-Zr/PANI-NC has superior or comparable adsorption capacity for MO (217.39 mg/g) and EY dye (117.65 mg/g) as compared to previously explored adsorbents (Table 4).

Regeneration study of Fe-Mn-Zr/PANI-NC
After the successful adsorption of MO and EY onto the Fe-Mn-Zr/PANI-NC, the regeneration study was carried out using 1.0 N NaOH solution, 1.0 M HCl solution and double distilled water.Firstly, the MO and EY dye loaded Fe-Mn-Zr/PANI-NC was dipped in 1.0 N NaOH solution to desorb the dye molecules.Next to reactivate the desorbed adsorbent, Fe-Mn-Zr/PANI-NC was kept in 1.0 M HCl solution for 8 hours [47].After that the composite was washed for several times with double distilled water and used for next cycle.This study was performed up to five consecutive cycles and the obtained results are delineated in Figure 6.No significant reduction in MO and EY dye removal efficiency was observed up to three regeneration cycles with dye uptake efficacy of 91.67% for MO dye and 89.87% for EY dye, respectively.However at the end of the fifth cycle the obtained dye removal efficiency was ~ 80%.The above results suggest the reusability of the Fe-Mn-Zr/PANI-NC for dye loaded wastewater treatment.

Conclusions
Ultrasound-assisted quick and enhanced adsorption of MO and EY dye from binary dye solution onto crystalline Fe-Mn-Zr/PANI-NC was reported in this study.The adsorption process was efficient in the wide solution pH range from 2.0 to 5.0, but considering the real field application the solution pH was optimised at 5.0.Due to involvement of ultrasound irradiation more than 98% dye removal efficiency was achieved for both MO and EY dye within 5 min of sonication.Kinetic studies for both MO and EY adsorption confirm the applicability of pseudo-second-order kinetic model and the contribution of intra-particle diffusion was also observed.Isotherm study reveals the suitability of Langmuir monolayer adsorptive model and the maximum uptake capacities of 217.39 mg/g and 117.65 mg/g were found for MO and EY dye, respectively.RSM study explores that highest removal efficiency (99% for MO dye and 98% for EY dye) was achieved with optimum values of experimental parameters, i.e.Fe-Mn-Zr/PANI-NC dose of 0.4 g/L, initial MO concentration of 12 mg/L, initial EY concentration of 12 mg/L and sonication time of 5.0 minutes.Regeneration study reveals that no significant reduction in MO and EY dye removal efficiency up to three regeneration cycles and after third cycle ~90% dye uptake efficacy was observed for MO dye and EY dye.However at the end of the fifth cycle the obtained dye removal efficiency was ~ 80%.Therefore due to high adsorption capacity, short equilibrium time, and good reusability, the Fe-Mn-Zr/PANI-NC can be a promising adsorbent for sonoassisted adsorptive remediation of dye loaded wastewater.
Figure 5(b) shows that initial MO dye concentration has negative correlation with the removal percentage of MO dye but Fe-MnZr/PANI-NC dose

Table 1 .
Experimental parameters with their levels and design of experiment for RSM study; experimental MO and EY dye removal efficiency versus RSM model predicted dye removal efficiencies.

Table 3 .
Isotherm parameters calculated from Langmuir, Freundlich and Temkin isotherms fitting The monolayer adsorption capacity of 217.39 mg/g and 117.65 mg/g was calculated from Langmuir adsorption isotherm for MO and EY dye, respectively.The values of Langmuir isotherm parameter 'R L ' (0.0 to 1.0) and Freundlich isotherm parameter 'n' (more than 2.0) indicate the favourability of this adsorption process (Table

Table 4 .
Performance evaluation of Fe-Mn-Zr/PANI-NC with various other adsorbents for MO and EY dye adsorption.