MenuISSN (0970-2083)
ISSN (0970-2083)
Deepak Kohli1, Sangeeta Garg1*, A.K Jana2, Ajay Bansal1 and Puniya Chopra1
1Department of Chemical Engineering, National Institute of Technology, Jalandhar 144 011, Punjab, India
2Department of Biotechnology, National Institute of Technology, Jalandhar 144 011, Punjab, India
Received date: 17 October, 2013; Accepted date: 12 December, 2013
Visit for more related articles at Journal of Industrial Pollution Control
Adsorption of organic pollutants, para-nitrophenol (pNP) and para-chlorophenol (pCIP) from aqueous solution using hexamethylene diisocynate (HMDI) cross-linked starch based adsorbent (CS3) was studied. Cross-linked starch (CS3) was characterized by fourier transform infrared spectroscopy (FTIR) and scanning electron microscopy (SEM). Thermodynamic studies showed that the adsorption of organic pollutants on to cross-linked starch (CS3) was spontaneous and exothermic and pseudo-second order kinetic model provided the best correlation of the experimental data. Breakthrough curves were plotted for the adsorption of pNP and pCIP using continuous flow column. Thomas model was applied for the adsorption of organic pollutants at different bed heights, flow rates and initial concentrations to predict the breakthrough curves. CS3 was capable of holding a maximum equilibrium capacity of 59.21 mg/g for pNP and 31.96 mg/g for pCIP with initial concentration of 100 mg/L, bed height 7.5 cm and flow rate of 4 mL/min. The experimental results fitted well with the Thomas model.
Organic pollutants, Cross-linked starch, Thermodynamics and kinetics, Fixed bed column, Thomas model
Phenolic compounds are pollutants of great concern because of the high toxicity and accumulation in the environment. Most of these phenolic compounds are recognised as organic contaminants in environmental systems (Bayramoglu and Arica, 2008). Phenols occur in waste water from a number of industries such as oil refineries, agrochemical, textile and pharmaceutical industries. The presence of phenolic compounds even at low concentration in water is restricted due to unpleasant taste and odour. Furthermore, phenols can cause adverse affect on digestive system, nervous system and respiratory system (Bayramoglu and Arica, 2008). Particularly, nitrophenols are listed as priority toxic pollutants and used to produce pharmaceutical drugs and pesticides (Marais and Nyokong, 2008). Chlorophenols are normally used in the synthesis of herbicides, insecticides and wood preservatives (Sze and Mckay, 2012).
Several methods, such as oxidation, electro-chemical treatment, pervaporation and adsorption have been used for the removal of phenols from waste water (Polaert et al., 2002; Korbhati and Tanyolac, 2003; Schutte, 2003; Gupta et al., 2003; Crini, 2005). However, adsorption is the most efficient, effective and extensively adopted method (Kalavathy et al., 2010). Recently, considerable attention has been directed towards low cost, naturally occurring adsorbents (Gupta et al., 1998; Kuleyin, 2007; Phan et al., 2000; Bhattacharya and Venkobachar, 1984; Delval et al., 2006). Among various adsorbents, starch has been considered as one of the best choice for the preparation of low cost adsorbents because of easy availability, low cost and can be used for the removal of pollutants from waste water (Crini, 2005). However, starch is hydrophilic in nature and limits the development of starch based adsorbents. Chemical modification of starch can be adopted to solve the problem and to produce water resistant materials (Koo et al., 2010). Recently, cross-linking of starch using various crosslinking agents such as citric acid, epichlorohydrin, phosphoryl chloride and hexamethylene diisocynate has been used widely to introduce desirable properties to starch (Reddy and Yang, 2010; Delval et al., 2005; Ratnayake and Jackson, 2008; Ozmen et al., 2008). Studies showed that the cross-linking of starch improved the surface area, swelling index, thermal properties and the sorption capacity of starch based materials (Wilpiszewska and Spychaj, 2007; Delval et al., 2002).
In our previous work, we have reported the batch studies for the adsorption of organic pollutants using cross-linked starch based adsorbent (Kohli et al., 2012). In the present work, column studies were carried out to evaluate the performance of bio-adsorbent in continuous fixed bed column by varying the operating conditions such as flow rate, bed height and initial concentrations. Thomas model has been used to analyse the breakthrough curve for the adsorption of pNP and pCIP. Thermodynamic parameters and kinetic studies have also been investigated for bioadsorbent.
Experimental
Materials
Para-nitrophenol (pNP) and Para-chlorophenol (pCIP) were purchased from Loba Chemicals Limited, Mumbai (India). Corn starch was obtained from Sukhjit Starch and Chemicals Limited, Phagwara (India). Hexamethylene diisocynate (HMDI) and dimethyl formamide (DMF) were supplied from Merck Chemicals Limited, Mumbai (India).
Adsorbent preparation
Cross-linked starch based bio-adsorbent was prepared using HMDI as cross linking agent and DMF as solvent. The procedure was same as described by Yilmaz et al., 2007; Ozmen et al., 2008 with minor modifications. 20 g starch and 100 mL of DMF were taken into three necked round bottom flask, and after that calculated amount of HMDI was added. The reaction contents were stirred at 70°C for 4 h. After precipitating, the content of reaction mixture was filtered, washed with double distilled water and ethanol. The product was dried at 50°C for 12 h.
Fourier transform infrared spectroscopy (FT-IR)
FTIR spectra of native and cross-linked starch (CS3) were recorded by FTIR instrument (Perkin Elmer, Model RX-1) using potassium bromide (KBr) pellets (Sigma Aldrich). FTIR spectra were recorded at a resolution of 4 cm-1.
Scanning electron microscopy
The morphological images of the native and crosslinked starch samples were recorded by JEOL JSM- 6100, scanning electron microscope (Jeol Ltd, Tokoyo, Japan). Samples were gold plated and observed at magnification range of 200 X to 4000 X.
Adsorption methods for column studies
Column experiments were performed in a glass column (1cm internal diameter and 20 cm length). The column performance was studied at different initial concentration (50, 100 and 150 mg/L), bed height (2.5, 5 and 7.5 cm) and flow rates (4, 6 and 8 mL/min). A known quantity of adsorbent was placed in the column to yield the desired bed height of the adsorbent. pNP and pCIP solutions of known concentrations and pH were pumped downward through the column at a desired flow rate by a peristaltic pump. The effluent samples were collected at specified intervals and analysed for the residual concentrations using UV spectrophotometer at 400 nm for pNP and 500 nm for pCIP.
Analysis of column data
The loading behaviour of pNP and pCIP to be removed from aqueous solution in a fixed bed column was usually expressed in terms of (C/Co) where (C = effluent adsorbate concentration and Co = influent adsorbate concentration) as a function of time or volume giving a breakthrough (Guibal et al., 1995). The maximum column capacity (mg), for a given feed concentration and flow rate is equal to the area under the plot of the adsorbed adsorbate concentration Cad (Cad = Co-C) (mg/L) versus volume (L) and is calculated from equation (3).
The equilibrium uptake ( qeq (exp)), the weight of adsorbate adsorbed per unit dry weight of adsorbent (mg/g) in the column, is calculated as follows:
Where X is the total dry weight of adsorbent in column (g)
Total amount of adsorbate sent to column (Wtotal) is calculated as:
Where is the total volume of adsorbate passed through the column.
Total percentage removal (Y) of adsorbate is the ratio of the maximum capacity of the column (qtotal) to the total amount of adsorbate sent to column (Wtotal)
Modeling of column data
Thomas model
The Thomas model is used to calculate the adsorption rate constant and the solid phase concentration of the pollutant on the adsorbent from the continuous mode studies. The kinetic model suggested by Thomas is one of the widely used kinetic models for the evaluation of column performance and assumes Langmuir model of adsorption desorption, and obeys pseudo-second order reversible reaction kinetics (Thomas, 1944; Han et al., 2006).
The Thomas equation can be expressed as follows:
Where KTh is the Thomas rate constant (mL/min mg); is the maximum solid phase concentration of solute (mg/g); is the amount of adsorbent in the column (g); is the flow rate (mL/min). The linearized form of Thomas model is as follows:
A plot of 
versus gives a straight line. The Thomas rate constant (KTh) was determined from the slop of the plot 
and the adsorption capacity of the column (qo) was calculated from the intercept of the plot 
.
Experimental conditions for the synthesis of polymers are shown in Table 1. The sorption capacity of crosslinked starch was found greater as compared to native starch. Whereas, sorption capacity of cross-linked starch (CS3) was more than CS1 and CS2, which may be due to rigid hexamethylene bridges formed in crosslinked starch (Kohli et al., 2012). Therefore, CS3 adsorbent is used for the present column study.
Table 1: Experimental condition for synthesis of adsorbents
Adsorbent characterization
FTIR spectra of native and cross-linked starch (CS3) are shown in Fig. 1 (a). In the spectra of native starch a broad band appeared at 3339 cm-1which was due to hydrogen bounded –OH groups. However, crosslinked starch (CS3) spectra also showed a band at 3331.47 cm-1 attributed to OH (and NH) groups. Other characteristic bands for cross-linked starch (CS3) appeared at 1578 cm-1 and 1620 cm-1 due to amide and carbonyl groups (Wilpiszewska and Spychaj, 2007).
Figure 1: FTIR spectra of native starch and CS3 (a); SEM micrographs of native starch and CS3 (b) and (c).
Surface morphological analysis of native and cross-linked starch (CS3) is shown in Fig. 1 (b-c). After cross-linking, smooth surface of starch converted to a rough surface which emphasized that starch granules were completely coated with HMDI layer and strong interactions between the HMDI and starch granules existed.
Thermodynamic parameters
Study of thermodynamic parameters Gibbs free energy of adsorption (ΔG°), enthalpy (ΔH°) and the entropy change (ΔS°) of an adsorption process explained the spontaneity of the process (Zakaria et al., 2009). Fig. 2 (a-b) shows the effect of temperature for the removal of organic pollutants on to cross-linked starch (CS3). It was found that the sorption capacity of pNP and pCIP decreased with increase in temperature. Thermodynamic parameters were evaluated at different temperatures 25°C, 30°C and 35°C and were calculated using the following equations:
Where Kc is the equilibrium constant, CAe is the amount of adsorbate on adsorbent per liter of the solution at equilibrium (mg/L), Ce is the equilibrium concentration of adsorbate in the solution (mg/L). T is the solution temperature (K) and R is the universal gas constant and equal to 8.314 J/mol K.
Figure 2: Effect of temperature for the removal of organic pollutants on CS3: pNP (a), pCIP (b); Variation of equilibrium constant (Kc) as a function of temperature: pNP (c), pCIP (d).
The value of ΔH° and ΔS° were calculated from the slope and intercept of the plot of 1/T versus ln Kc for pNP and pCIP and are shown in Fig. 2 (c-d). The values of Kc, ΔG,° ΔH° and ΔS° for organic pollutants are given in Table 2. The change in standard free energy ΔG° with negative values -3.51, -1.68 and -0.46 kJ/mol for pNP and -1.49, -0.80 and -0.125 kJ/mol for pCIP, at different experimental temperatures indicated that the adsorption of organic pollutants on CS3 was spontaneous and thermodynamically favourable.
Table 2: Thermodynamic values at various temperatures for pNP and pCIP
ΔH° had a value of -94.47 kJ/mol for pNP and - 42.17 kJ/mol for pCIP. The negative value of ΔH° showed that the adsorption of organic pollutants on CS3 was exothermic. The values of ΔS° for pNP and pCIP were -305.45 J/Kmol and -136.43 J/Kmol, reflected that the randomness decreased during the adsorption process.
Kinetic modeling
Adsorption kinetics showed strong dependence on the physical or chemical characteristics of the adsorbent materials and also used to determine the rate of adsorption process. Pseudo first and pseudo second order models were used to investigate the adsorption of organic pollutants on CS3. Pseudo-first order equation (Bilgili, 2006) is as follows:
Where qe and q are the amounts of adsorbate adsorbed (mg/g) at equilibrium and at any instant of time t (min), respectively, and k1 is the rate constant of pseudo-first order adsorption (min-1).
Pseudo-second order model equation (Kuleyin, 2007) is as follows:
Where qt is the amount of adsorbate adsorbed (mg/g) at time t, K2 is the rate constant (g/mg min). Pseudo-first order kinetic plot is shown in Fig. 3 (a-b). Based on Eq. (16), the values of R2, K1 and qe were 0.966, 0.0115 (min-1) and 0.998 (mg/g) for pNP and 0.967, 0.0253 (min-1) and 0.982 (mg/g) for pCIP, respectively.
Figure 3: Adsorption kinetics: Pseudo first order kinetic plot: pNP (a), pCIP (b); Pseudo second order kinetics plot: pNP (c), pCIP (d).
Fig. 3(c-d) shows the pseudo-second order kinetic plot for organic pollutants on cross-linked starch (CS3). The pseudo-second order model had higher correlation coefficient value, R2 > 0.99 for both pNP and pCIP. The value of K2 and qe calculated from Eq. (17) were 0.0788 g/mg min and 4.424 mg/g for pNP and 0.1334 g/mg min and 3.1847 mg/g for pCIP. It was found that the calculated qe values were closer to the experimental data as compared to values of pseudo-first order model, while the correlation coefficients were very high for pseudo-second order kinetic model. Therefore, the adsorption of pNP and pCIP were found more favorable by the pseudo-second order model.
Column studies
Effect of initial concentration on breakthrough curves
The adsorption potential of cross-linked starch (CS3) was examined at various initial concentrations (50, 100 and 150 mg/L) of organic pollutants. Fig. 4 (a-b) shows the effect of increase in initial concentration of pNP and pCIP on the breakthrough curves with a bed height of 2.5 cm and a flow rate of 4 mL/min. It was found that with the increase in initial concentration from 50 to 150 mg/L, the slope of breakthrough curve became steeper, which was due to increase in driving force and decrease in the adsorption zone length (Goel et al., 2005). It was observed that at low initial concentration, the treated volume was higher because lower concentration gradient had a slower transport due to decreased diffusion rate. On the other hand, the availability of the adsorbate for the adsorption site was also increased at higher concentration, which leads to the higher uptake of adsorbate (Goel et al., 2005). KTh, qo and correlation coefficients were calculated by Thomas equation and are shown in Table 3. The removal efficiency of organic pollutants decreased as the initial concentration increased. Equilibrium capacities (qeq(exp)) increased from 34.12 to 59.21 mg/g for pNP and 21.89 to 31.96 mg/g for pCIP as the initial concentration increased from 50 mg/L to 150 mg/L. As the initial concentration increased the KTh values decreased. The linear correlation coefficients (R2) ranging from 0.871 to 0.953 for pNP and 0.959 to 0.974 for pCIP. It was observed that the difference between values estimated by Thomas model and those measured experimentally were insignificant. So, results showed that the Thomas model gave well-fit experimental data under these conditions.
Figure 4: Effect of initial concentration on breakthrough curves for adsorption of organic pollutants on CS3 (flow rate = 4 mL/min, pH 5 and bed height 2.5 cm): pNP (a), pCIP (b).
Table 3: Parameters predicted from Thomas model of organic pollutants adsorption at different initial concentrations (Flow rate 4mL/min, pH 5 and bed height 2.5 cm)
Effect of bed height
The adsorption of pNP and pCIP in the packed bed is dependent on the quantity of adsorbent inside the packed column. The effect of bed height on breakthrough curves with a flow rate of 4 mL/min and at a concentration of 100 mg/L of pNP and pCIP is shown in Fig. 5 (a-b). Different bed heights of 2.5 cm, 5 cm and 7.5 cm were obtained by adding 1g, 2 g and 3 g of the adsorbent respectively. It was observed that with the increase in bed height from 2.5 cm to 7.5 cm, volume of solution treated increased at various breakthroughs, which was due increase in number of adsorption sites and surface area (Vijayaraghavan et al., 2004).
Figure 5: Effect of bed height on breakthrough curves for adsorption of organic pollutants on CS3 (flow rate = 4 mL/min, pH 5 and Co= 100 mg/L): pNP (a), pCIP (b).
The values of qo and KTh at different bed heights are presented in Table 4. The values of equilibrium capacities (qeq(exp)) increased from 48.80 to 57.2 mg/g for pNP and 26.5 to 30.0 mg/g for pCIP as the bed height increased from 2.5 cm to 5 cm. In addition, the values of KTh decreased with increase in bed height. Higher values of (R2) showed that the Thomas model could describe the breakthrough data under these conditions. The maximum adsorption capacities (qo(cal)) of organic pollutants predicted by the model were found effected by bed height. The values of qo(cal) estimated by Thomas model were near to the experimental values.
Table 4: Parameters predicted from Thomas model of organic pollutants adsorption at different bed heights (Flow rate 4mL/min, pH 5 and Co 100 mg/L)
Effect of solution flow rate
The effect of solution flow rate on the adsorption of organic pollutants by cross-linked starch (CS3) was investigated by varying the flow rate from 4 mL/min to 8 mL/min. The bed height and inlet organic concentration were fixed at 2.5 cm and 100 mg/L. The breakthrough curves for different flow rates are shown in Fig. 6 (a-b). Results indicated that, as the flow rate increased, the breakthrough became steeper and reached the breakthrough quickly. This was due to the shorter contact time between the adsorbent and the adsorbate. The percentage removal Y (%) decreased from 57.41 % to 53.3 % for pNP and 55.78 % to 50.22 % for pCIP (Table 5). This was due to the fact that, when the residence time of the organic pollutants was not enough for adsorption equilibrium at that flow rate, the solute solution left the column before equilibrium occurred. Therefore, the contact time of adsorbate with adsorbent was very short at higher flow rate, causing a reduction in removal efficiency (Ghorai and Pant, 2005). However, equilibrium adsorption capacities (qeq(exp)) of pNP was better than pCIP due to strong interaction between the pollutant and the adsorbent. Whereas, pCIP had the lowest interaction with the adsorbent because of steric restrictions in the network (Delval et al., 2006).
Figure 6: Effect of flow rate on breakthrough curves for adsorption of organic pollutants on CS3 (Co = 100 mg/L, pH 5 and bed height 2.5 cm): pNP (a), pCIP (b).
Table 5: Parameters predicted from Thomas model of organic pollutants adsorption at different flow rates (Co 100 mg/L, pH 5 and bed height 2.5 cm)
The values of qo and KTh at different flow rates are presented in Table 5. It was found that the removal efficiency of pNP and pCIP decreased as the flow rate increased. With the increase in flow rate from 4 mL/ min to 8 mL/min the equilibrium capacities (qeq(exp)) increased from 37.31 to 48.80 mg/g for pNP and 20.09 to 26.5 mg/g for pCIP. Higher value of R2 indicated that the Thomas model was well fitted by the column data. The values of qo(cal) estimated by Thomas model were not significantly different with those measured experimentally. So, it was concluded that the Thomas model gave well-fit experimental data under this study conditions.
The bio-adsorbent prepared by cross-linking of starch with HMDI was found to be an effective adsorbent for the removal of organic pollutants (pNP and pCIP) from waste water. The negative values of ΔG° indicated the feasibility and spontaneity of adsorption process. Negative values of ΔH° and ΔS° suggested that the adsorption of pNP and pCIP on CS3 was exothermic and showed decreased randomness. The kinetics of organic pollutants on CS3 was examined with pseudo-first and second order models. The results indicated that pseudo-second order model provided the best correlation of the adsorption data. The adsorption capacity (qeq(exp)) increased with increase in influent concentration, bed height and decreased with increase in flow rate. The Thomas model analysis of linear regression analysis was well fitted with the experimental data.
Copyright © 2024 Research and Reviews, All Rights Reserved