Synthesis and application of magnetized nanoparticles to remove lead from drinking water: Taguchi design of experiment

Contamination in drinking water from heavy metals like Pb2þ has severe effects on health. In this study, potato peel (PP) was used as the substrate and magnetic iron nanoparticles (MI) were deposited on PP using a co-precipitated method. Fourier transformation infrared spectroscopy (FTIR) and X-ray diffraction (XRD) analysis confirmed the deposition of MI on PP. The L16 (4^4) method of Taguchi design of experiment (DOE) was used for the optimization of adsorption condition, i.e., at 6 pH, 10 min of contact time, and a dose of 15 g/L can give more than 90% removal efficiency of Pb2þ using PP-MI. Contour maps, Taguchi response analysis, and analysis of variance (ANOVA) suggested that pH has a dominant contribution in the removal of Pb2þ. The adsorption process was favorable, spontaneous, and exothermic in nature and was followed by pseudo second order kinetics. A comparison of the sorption capacity of PP-MI for Pb2þ with literature values suggested that PP-MI has good potential for the removal of Pb2þ. doi: 10.2166/washdev.2020.097 om https://iwaponline.com/washdev/article-pdf/10/1/56/673685/washdev0100056.pdf 0 Muhammad Irfan Jalees Institute of Environmental Engineering and Research, University of Engineering and Technology, Lahore 54890, Pakistan E-mail: irfan611@gmail.com


INTRODUCTION
The cumulative, non-biodegraded, and persistent nature of heavy metals make them an environmental concern even when present in trace amounts (Gupta & Ali ).
Exposure to heavy metal like Pb 2þ , even in trace amounts, can cause adverse health effects (Gupta & Ali ). Many industries discharge industrial effluents containing Pb 2þ , which can pose a risk to humans. Therefore, the treatment of Pb 2þ prior to discharge is important, but the complex composition of effluents makes treatment difficult (Gupta & Ali ). Furthermore, the presence of dissolved organic compounds plays a key role in controlling the physiochemical properties of Pb 2þ ions which further enhance the difficulty level for Pb 2þ treatment (Gupta & Ali ). Out of many treatment methods, adsorption is considered as economic, efficient, and effective for the treatment of heavy metals (Günay et  The high absorption abilities of nanoparticles are further enhanced by the use of magnetic nanoparticles which increases the efficiency of solid-liquid separation problems (Plohl et al. ). These magnetic particles not only have a large surface area which increases the removal efficiency (%RE) but they are also easily prepared and can be separated from samples to reduce secondary solid waste generation (Plohl et al. ). Several attempts have been made to use various forms of nanoparticles, e.g., maghemite nanoparticles (Oukebdane et al. ), iron-humic acid, nanopolymers (Rao et al. ), etc. In this study, potato peel (PP) (agricultural waste) was used as substrate. Potato is abundantly used as potato chips. Based on the peeling method, 15-40% (% wt) of potato is wasted as potato peel.
This potato peel is a major contributor of solid organic waste in food industries (Gebrechristos & Chen ). Using potato peel as a substrate will help in reducing solid waste.
Magnetic iron (MI) particles were synthesized and coated on PP using the co-precipitated method. The PP-MI was also characterized using FTIR and XRD. The prepared PP-MI is then used as adsorbent material for optimization and isotherm studies for the removal of Pb 2þ from drinking water.

MATERIAL AND METHODS
The chemicals used during experiments were FeCl 3 .6H 2 O, FeSO 4 .7H 2 O, ammonia solution, H 2 SO 4 , and Pb(NO 3 ) 2 , were of analytical grade and purchased from Merck, Pakistan. The potato was purchased from the local market and its peel was used for experimentation. All glassware was Pyrex and washed with 2% HNO 3 and double rinsed with deionized water before being used in experimentation. The analysis of Pb 2þ was performed on an Atomic Absorption Spectrometer (AAS) (Analyst 800, Perkin Elmer) (American Public Health Association ) in triplicate after performing the limit of detection (LOD) and quality checks (QC) test. Samples were also analyzed from the chemistry department, UET and Institute of Chemistry Punjab University (PU) to minimize errors. Characterization of PP-MI was performed using FTIR and XRD from the physics department, UET Lahore.

Preparation of the adsorbent
Potato peel (PP) was dried in sunlight for 8 hr daily for 6 days. The average temperature was 37-41 C during the drying period. After drying, PP was transferred to an oven at 45 C. The dried PP was then cut into small pieces, ground into powder form and stored in airtight jars to prevent damp. The mesh size of powdered PP was 60-200 micron. Powdered PP (1 g) was added into 200 mL of distilled water and heated at 80 C. After that, 10 mL of NH 4 OH was added to maintain a pH of 10. Ferrous sulfate hepta-hydrate (4.2 g) and ferric chloride hexa-hydrate (6.1 g) were mixed in 100 mL de-ionized water, separately. These salts were then slowly mixed with PP solution at 80 C.
After the complete addition of two iron salts, the combined

Taguchi method for design of experiments (DOE)
Design of experiments (DOE) for the Taguchi method is used for the optimization of conditions. Four parameters, i.e., pH (2-8), dose of adsorbent (PP-MI) (5-20 g/L), contact time of adsorption (10-40 min) and initial concentration of Pb 2þ (70-90 ppm) was selected. The method L16 (4^4) was selected for Taguchi DOE. It has four parameters and four levels. Table 1 contains the various inputs (16) for the optimization of adsorption conditions. Known solutions of various Pb 2þ concentrations (10-90 ppm) were prepared using the standard solutions provided by Perkin Elmer, US (Oukebdane et al. ). Analysis of iron which may be leached out during the adsorption process was also carried out using an atomic absorption spectrometer.

Characterization of PP-MI
Chemical analysis of PP was performed to measure moisture content, carbohydrates, proteins, fats, ash, and heavy metals. The results of the composition are given in Table 1.

Optimization of parameters
Sixteen experimental sets were applied for the removal of Pb 2þ from the drinking water.  ( Figure 3(b)). Furthermore, pH 6 and 15 g/L adsorbent dose along with 100 ppm concentration of Pb 2þ also gave more than 90% removal efficiency of Pb 2þ (Figure 3(d)).  from drinking water using PP-MI particles.

Isotherm studies
Various isotherms, i.e., Langmuir isotherm, Freundlich isotherm, Temkin isotherm, Dubinin-Radushkevich (D-R) isotherm, and Flory-Huggins isotherm were used to study the mechanism of adsorption. These isotherms of adsorption are very important to discover the behavior of adsorbate on specific adsorbents. To find the maximum     The Langmuir adsorption model was adopted for the homogenous monolayer adsorption process (Theivarasu & Mylsamy ). The model equation is given in Table 5.
The linear plot of C e vs C e /q e is given in Figure 5. The values (Table 6) of q m and b were calculated using the slope and intercept of the plot (Figure 5(a)). The value of R L was less than zero, which suggested the adsorption was favorable (Theivarasu & Mylsamy ).
The Freundlich isotherm model was adopted to calculate adsorption intensity on the adsorbent surface. The plot of logC e vs logQ e was used for this purpose ( Figure 5(b)).
The values of n and K f were calculated from the slope and intercept of the graph, respectively. The value of n was smaller than zero, which suggested a favorable adsorption process.
The Dubinin-Radushkevich isotherm model was used for porosity and energy of adsorption measurements. The plot between ε 2 vs LnQ e was used ( Figure 5(c)). The values of K ad and q m were calculated using the slope and intercept of the graph, respectively ( Table 6). The value of E (mean free energy of adsorption) was 0.32 kJ/mol. The Temkin isotherm model was used to predict the uniform distribution of binding energy over the population of surface binding adsorption (Theivarasu & Mylsamy ).
The plot of LnC e vs Q e was used ( Figure 5(d)). The values   1.
The Flory-Huggins isotherm model was used for the degree of surface coverage of adsorbate on the adsorbent.
The values of K FH and n can be calculated by slope and intercept, respectively ( Table 6). The value of ΔG indicated a spontaneous exothermic condition.

Kinetics
Kinetics of removal of Pb 2þ was studied at optimized condition (mentioned in the previous section) at a concentration of 60 ppm of Pb 2þ . The pseudo first order (Equation (6)) and pseudo second order (Equation (7)) kinetics were studied using the following equations (Jalees et al. ): The plot of t vs log(Q e À Q t ) and t vs T/Q t was used for pseudo first (Figure 6(a)) and second order (Figure 6(b)), respectively. Various kinetic parameters measured from these plots are given in Table 7. The regression values suggested that the adsorption followed pseudo second order kinetics as R 2 value was very close to 1.

CONCLUSION
The use of agro-nanoparticles for the removal of Pb 2þ from drinking water shows good results. Characteristic peaks of FTIR and XRD confirmed the deposition of MI on PP.
The Taguchi design of experiment (16 experiments) was performed, which indicated that at pH 6, the contact time of 10 min and an adsorbent dose of 15 g/L can give more than 90% removal efficiency of Pb 2þ using PP-MI. Contour