Application of Taguchi L27 orthogonal array design to optimize Reactive Orange 12 Dye adsorption onto Magnetic Mn3O4 and MnFe2O4 Nanocomposite
Research Paper (postgraduate) 2016 37 Pages
In this study, Taguchi L27 experimental design was utilized to optimize adsorption of Reactive Orange 12 (RO 12) dye onto magnetic manganese oxide and manganese ferrite (MO-MnF) nanocomposite. The MO-MnF nanocomposite was characterized by X-ray diffraction (XRD), scanning electron microscopic (SEM), Fourier transform infrared spectroscopy (FT-IR), and vibrating sample magnetometer (VSM) measurements. The experimental design was constructed with five factors (solution pH, RO 12 dye concentration, MO-MnF dose, contact time, and reaction temperature) at three different levels. The experimental conditions were optimized to maximize the RO 12 dye removal efficiency where signal-to-noise (S/N) ratio was considered as criteria. Optimum values of 2.0, 1.0 g/L, 60 min, 50 mg/L and 60°C were obtained for solution pH, MO-MnF dose, contact time, concentration of RO 12 dye, and reaction temperature, respectively for maximum RO 12 dye removal efficiency of 99.32%. This adsorption process proficiently followed pseudo-second-order and intra-particle diffusion models and exhibited applicability of Langmuir monolayer adsorption with maximum adsorption capacity of 207.90 mg/g at room temperature.
Key words: Adsorption, Reactive Orange 12 dye, magnetic nanoparticles, Taguchi optimization
Release of industrial effluents containing toxic synthetic dyes into the aquatic environment has become a serious problem towards scientific community. Azo dyes are the most widely used synthetic dyes in industry and problems associated with most of the azo dyes are their stability, recalcitrant to microbial biodegradation, colorant and even potentially carcinogenic and toxic to human being [1,2]. The release of colored substance even at very low concentrations can cause serious damage because these dyes prevent sunlight penetration and then have a pejorative effect on photosynthesis process in aquatic ecosystem [3,4]. Reactive Orange 12 (RO 12) is an anionic dye and is very frequently used for coloring silk, wool, leather, jute and cotton and can be applied as biological staining, dermatology, veterinary medicine, and green ink manufacture . The anionic dyes are the brightest class of soluble dyes and produce severe environmental and health hazards [6,7]. RO 12 is toxic for humans and animals by causing permanent injuries to their eyes [5,8]. On account of above stated hazards of toxic dyes, the environmental scientists have shown remarkable interest to design and develop innovative approaches for their efficient and quantitative elimination by simple, economical and fast techniques. An environmentally benign technique with efficient removal efficiency in a short time with minimum requirements of chemicals is greatly recommended.
Several treatment techniques such as adsorption/biosorption [9-13], membrane techniques , coagulation  and flocculation , photo catalytic degradation [17,18] and chemical oxidation  are being applied for effective remediation of dye contaminated water. The adsorption process is one of the most effective treatment techniques for removal of wide range of pollutants owing to its high efficiency, simplicity of design, flexibility, less sludge generation, and availability of a large number of adsorbents [20,21]. Recently nano sized adsorbents have evoked a lot of interest for organic/inorganic pollutant removal [22-24] from water. A nanoscale material possesses some physical properties which make it a potential candidate as an adsorbent for remediation of toxic dyes, heavy metals, and other pollutants from polluted waters considering their unusual properties like high surface area, high number of reactive atoms, high mechanical and thermal strength and large number of vacant reactive surface sites in comparison to bulk particles . They can be functionalized easily with diverse chemical species to enhance their adsorption selectivity towards a particular pollutant. Higher surface area per mass provides more active sites for surface chemistry. However, post adsorption recovery of nano-adsorbents for reuse and regeneration involves centrifugation which implicates extra cost and time. Separation of magnetic nanoscale materials using external magnetic field for easy collection from wastewater offers a solution to this problem . Magnetic nanoparticles with super paramagnetic behaviors are receiving great importance due to their improved physical properties like high surface area and porosity, ease of functionalization and chemical stability. More importantly they provide several advantages due to their super paramagnetic behavior while designing separation and recovery units in complex heterogeneous systems . Moreover, magnetic separation technology could be cheaper than the conventional separation methods. That’s why application of magnetic nanoadsorbents has been extensively studied for environmental purification applications, recently [28,29].
In addition to develop novel adsorbents with high functionality towards toxic dyes, several mathematical models can be accomplished in such adsorption processes for determination of relevance and evaluation of statistical significance of various process variables and their simultaneous interaction . Multivariate designing tools can perform lateral estimations and expressions of process variables with minimum number of experimental data points, which reduces the experimental time tremendously. Moreover, multivariate optimization techniques are far better than one way classification in order to obtain exact concepts about the mutual interactions between experimental parameters . Taguchi optimization is such an optimization tool where the simultaneous interaction behaviors of the experimental variables can be studied efficiently for optimizing the experimental conditions for maximizing/minimizing any target output.
In view of above, we have reported a facile method for synthesis of manganese oxide and manganese ferrite (MO-MnF) magnetic nanocomposite for adsorption of RO 12 dye. In this adsorption study, the effect of each process variables were studied in details and the adsorption process was optimized by Taguchi L27 orthogonal array design for maximizing the RO 12 dye removal efficiency. The contribution of experimental parameters onto dye removal efficiency was determined from Taguchi design of experiment. Furthermore, several conventional kinetics, isotherms, and thermodynamic analysis were also performed to understand the adsorption mechanism.
2. EXPERIMENTAL DETAILS
2.1. Materials and methods
High purity MnCl2.4H2O, FeCl3, and NaOH pellets were purchased from Merck and deionized (DI) water (Millipore, 18 MΩ·cm) was used in entire experimental process. RO 12 dye powders were purchased from Leo Chemical Pvt. Ltd. (India) and used for preparing the aqueous solutions of different dye concentration for batch adsorption study. The RO 12 dye concentrations were determined using a Shimadzu UV-Vis-NIR spectrophotometer (model: UV–3101PC) at a wavelength of 415 nm (λmax). The RO 12 dye adsorption experiments were carried out in batch mode as follows: 50 mL of RO 12 dye solution was mixed with different dose of MO-MnF nanocomposite (0.125–1.50 g/L) and the mixture was dispersed thoroughly for a predefined contact time at room temperature. At every stage, the supernatant sample was immediately separated from the adsorbent by applying magnetic field and analyzed by UV-Vis spectrophotometry at maximum wavelength over working concentrations.
2.2. Synthesis and characterization of MO-MnF nanocomposite
In this synthesis process, 20.0 g of NaOH pellet was dissolved in 500 mL ultrapure deionized (DI) water and another two separate solutions of FeCl3 and MnCl2.4H2O were prepared by dissolving 10.0 g of FeCl3 powder and 6.0 g of MnCl2.4H2O powder in two beakers containing 200 mL DI water, respectively. Both the dissolved solutions were then mixed to the NaOH solution thoroughly for formation of brownish floc. The settled floc was separated by filtration and dried overnight in an oven at 85°C and dry powder samples were washed with DI water for several times. Then the samples were calcined at 400°C for six hours and obtained brownish black solid materials were grinded to powder form of MO-MnF nanocomposite. The prepared MO-MnF nanocomposite were characterized by X-ray diffraction (XRD) (Bruker, D-8 Advance), field emission scanning electron microscopic (FESEM, Hitachi, S-4800), Fourier Transform Infrared (FTIR) spectroscopy (Perkin Elmer), and vibrating sample magnetometer (VSM) measurement (Lakeshore, Model: 7410 series).
2.3. Taguchi design of experiments
Taguchi design of experiment was proposed by Genichi Taguchi , which simplified the statistical design efforts by using orthogonal array. Application of this method can considerably minimize the costs and number of experiments. Taguchi experimental design has been applied successfully to optimize various physicochemical processes since last decade [33-35] including optimization of adsorption processes [36-38] due to its ability to analyze the mutual involvement of multiple experimental variables at a time. The analysis in Taguchi approach is performed mainly based on orthogonal arrays and analysis of variance (ANOVA). Orthogonal arrays reduce the number of experiments and ANOVA calculates the influence of each process variables onto the output. Taguchi method uses the signal-to-noise (S/N) ratio which takes both average and variation into account. S/N ratio is used to calculate the quality characteristic deviating from the target value [39,40].
In this study, an orthogonal array (L27) is prepared using five process variables (solution pH, RO 12 dye concentration (mg/L), MO-MnF dose (g/L), contact time (min) and reaction temperature (◦C)) each with three levels (Table 1) to yield a stable design. The layout of the orthogonal array considering five operational variables, each taken at three levels is shown in Table 2. In Taguchi optimization, normally S/N ratio can be analyzed based on three different approaches i.e. ‘Larger is better’, ‘Target is best’, and ‘Smaller is better’ [40,41]. In this study we have adopted ‘Larger is better’ category as our objective is to maximize the RO 12 dye removal (%), which can be represented by the following equation.
Abbildung in dieser Leseprobe nicht enthalten (1)
where n is the number of experimental data points, and y is the RO 12 dye removal (%). Minitab 17 software was used to analyze S/N ratio and ANOVA for this study.
3. RESULTS AND DISCUSSION
3.1. Characterization of MO-MnF nanocomposite
The XRD pattern of the prepared adsorbent (Fig.1(a)) revealed that the samples are polycrystalline with mixed phase of manganese ferrite and manganese oxide. Two strong peaks of manganese oxide and five peaks of manganese ferrite can be observed in the in diffractogram. The strong XRD peaks indicate the well-crystalized structure of MO-MnF nanocomposite. The morphology of the MO-MnF nanocomposite by FE-SEM show mixed mode of nano-structure as rod and small grains (Fig.1(b)). The nanorods are of 600 nm length and nanoscale grains are of ~50 nm diameter. The FTIR study confirms the presence of metal oxygen vibrational modes arising from Fe–O stretching vibrations and Mn–O stretching vibrations of manganese ferrite and manganese oxide nanoparticles at 585 cm–1 and 472 cm–1 . Room temperature VSM data shown that the saturation magnetization of the powder sample is 30.12 emu/g. Thus the composite has been conveniently recovered from the aqueous solution by an external magnetic field and could have potential application as a magnetic nano-scale adsorbent to remove pollutants.
3.2. Taguchi optimization
The main objective of this Taguchi based optimization was to define the optimum experimental conditions for maximum RO 12 dye removal efficiency. The experimental conditions for which the calculated S/N ratio was higher were considered as optimal, as ‘Larger is better’ approach was adopted for calculating S/N ratio using Equation (1). Response Table for S/N ratio of L27 experimental design is shown in Table 3, where bold numeric values represent the optimum levels of the studied variables. Response table (Table 3) also shows the impact of considered factors onto the RO 12 dye removal (%) based on ‘Delta’ value. Solution pH (X1) is the most influencing factor followed by RO 12 dye concentration (X3), MO-MnF dose (X2), and temperature (X5). The parameter contact time (X4) is having the minimum impact on the output parameter RO 12 dye removal (%). The contribution of the experimental parameters were calculated as 43.80%, 14.29%, 25.00%, 7.35% and 9.56% for solution pH, MO-MnF dose, RO 12 dye concentration, contact time, and reaction temperature, respectively. The calculated S/N ratio for each level of all considered process variables is shown in Fig. 2. From the response table (Table 3) and Fig. 2, it is clear that the optimum experimental condition for adsorption of RO 12 dye onto MO-MnF nanoparticle is at a solution pH of 2.0, MO-MnF dose of 1.0 g/L, RO 12 dye concentration of 50 mg/L, contact time of 60 min, and reaction temperature of 60°C, and at optimum condition the Taguchi model predicted RO 12 dye removal efficiency comes to 99.32%. The experimental results were also predicted by Taguchi experimental design and the experimental and predicted values are shown in Table 2. The linear regression plot of experimental and Taguchi model predicted values are depicted in Fig. 3. The plot shows very good correlation (R2=0.957), which demonstrates the prediction ability of developed Taguchi orthogonal experimental design. Additionally, three replicates of optimum experimental conditions were performed at lab in order to check the potentiality of the predicted values by Taguchi model. It was found that the observed experimental values (98.43±1.02%) are in close agreement with the predicted value (99.32%), again confirming the applicability of this model for successful optimization of the adsorption process.
3.3. Analysis of variance (ANOVA) under Taguchi design
To understand the impact of each experimental variable onto RO 12 dye removal (%), the analysis of variance (ANOVA) was studied using Minitab 17 software and the results are shown in Table 4. In ANOVA, the p-value indicates the importance of a parameter in a model, and the F-values (Fisher test) exhibits the level of importance of the parameter in the model, irrespective of its positive or a negative effect. From F values and P values of Table 4, it is clear that solution pH, MO-MnF dose, and RO 12 dye concentration are exhibiting significant effect on the RO 12 removal efficiency due to high F values and very low P values (<0.001). However, out of five parameter, solution pH is the most critical factor (highest F value 861.81 and P value <0.001) and contact time is the least important factor (low F value 22.32 and highest P value 0.073).
3.4. Effect of various experimental parameters on adsorption of RO 12
Effects of various experimental parameters such as initial solution pH, adsorbent dose, initial RO 12 dye concentrations, and contact time were systematically explored in this study. Effect of initial solution pH in the range of 2.0–10.0 was studied by adding 1.0 g/L of adsorbent to 50 mL of RO 12 solution (300 mg/L) at room temperature and results depicted in Fig. 4(a), which shows that initial solution pH significantly affects the extent of adsorption and highly acidic condition is favorable for the process. Maximum adsorption capacity ~207.0 mg/g was obtained in equilibrium at pH 2.0 and hence all subsequent experiments in this study were conducted at pH 2.0.
The effect of initial RO 12 dye concentration was investigated in the range of RO 12 concentration of 50–250 mg/L at fixed adsorbent dose (1.00 g/L) and results depicted in Fig. 4(b). It is clear that the RO adsorption capacity increases with increase in initial dye concentration at a fixed adsorbent dose. The higher adsorption capacity at higher dye concentration at a fixed adsorbent dose is because of stronger interaction between dye molecule and adsorbent at higher ionic concentration of RO 12 . Moreover, from Fig. 4(b) it can be seen that RO 12 adsorption capacity increases with increase in contact time and reached equilibrium at 40 minutes and beyond that there is no improvement in adsorption capacity.
In order to explore the influence of adsorbent dose on the process, experiments were conducted by varying the adsorbent dose (0.125‒1.5 g/L), keeping the initial RO 12 concentration fixed at 225 mg/L and the results have been depicted in Fig. 4(c). The adsorption efficiency increased from about 45% to ~98.5% as the dose of the adsorbent increased from 0.125 g/L to 1.50 g/L. As the adsorbent dose has increased, the availability of active surface sites of adsorbent also increased which in turn increases the uptake percentage of RO 12 dye. However, adsorption capacity increased from ~150 mg/g to more than 850 mg/g, as the dose decreased from 1.5 g/L to 0.125 g/L at a fixed initial RO 12 concentration of 225 mg/L. Unsaturated active sites of adsorbent surface at higher dose and overlapping in available surface area may reduce the total effective surface area which in turn decreases the adsorption capacity at higher dose .