Ecological Engineering 83 (2015) 218–226
Contents lists available at ScienceDirect
Ecological Engineering journal homepage: www.elsevier.com/locate/ecoleng
Modiﬁed leaf biomass for Pb(II) removal from aqueous solution: Application of response surface methodology Suguna Madalaa,* , Mudumala Veera Narayana Reddya , Sreenivasulu Vudagandlab , Krishnaiah Abburia a b
Biopolymers and Thermo Physical Laboratories, Department of Chemistry, Sri Venkateswara University, Tirupati, 517 502, AP, India Institute of Atmospheric Environmental Safety and Pollution Control, Jinan University Huangpu Road West 601, Guangzhou 510632 China
A R T I C L E I N F O
A B S T R A C T
Article history: Received 12 November 2014 Received in revised form 3 June 2015 Accepted 20 June 2015 Available online xxx
A feasibility study was performed on Tephrosia Purpuria Leaf (TPL) biomass to observe its potential to remove Pb(II) ions from aqueous solution by biosorption. The sorbent was characterized by FTIR, SEM, and EDAX analyse. The inﬂuence of various operating parameters such as pH, biosorbent dose and contact time on biosorption process was optimized by using a quadratic model Box–Behnken design using surface methodology. The experimental results were ﬁtted with a multiple regression analysis. The results indicated that biosorption of Pb(II) on TPL was strongly dependent on pH. Using the ridge analysis the optimal conditions are: Equilibrium contact time for the removal of Pb(II) was 136 min, whereas the optimum amount of adsorbent dose was 0.4 g and pH was 5.4. Under these optimum conditions, the corresponding response value predicted for maximum biosorption yield was 90.6 mg/g. This was also conﬁrmed by validation experiments. The Pb(II) uptake by TPL biomass was quantitatively evaluated by using sorption isotherms such as Langmuir and Freundlich isotherm models, of which Langmuir isotherm gave better correlation and maximum adsorption capacity was found to be 100.0 mg/g. The biosorption process was found to be highly feasible, spontaneous and endothermic. ã 2015 Elsevier B.V. All rights reserved.
Keywords: Biosorption Lead(II) Tephrosia purpuria leaf (TPL) Response surface methodology Isotherm models Thermodynamics
1. Introduction Heavy metal pollution is becoming a serious environmental threat to our environment and the health of human being. Water pollution by toxic heavy metals is a worldwide concern for the last few decades. Among the various heavy metals, lead is one of the most common toxic pollutants released into natural water resources from various industrial activities such as paints and pigments, metal plating, mining and battery manufacturing etc. Unlike organic pollutants, Pb(II) is not biodegradable and tends to accumulate in living tissues, causing various diseases and disorder. According to the US Environmental Pollution Agency, it is a highly toxic cumulative element, causing a variety of negative effects on humans, even at low dosages. The toxicological effects of lead in humans include inhibition of haemoglobin formation, hypertension, learning disabilities, abortion, kidney damage, and mental retardation (Suguna et al., 2014; Tabaraki et al., 2014). The permissible limit of lead in drinking water as set by the World
* Corresponding author. E-mail address: [email protected]
(S. Madala). http://dx.doi.org/10.1016/j.ecoleng.2015.06.025 0925-8574/ ã 2015 Elsevier B.V. All rights reserved.
Health Organisation is 10 mg/L (WHO, 2011). Due to the aforementioned reasons in recent years there is great interest generated for removing these metal ions from aqueous solution due to their supreme toxicity to human health and ecological systems. The treatment of heavy metals bearing efﬂuents has been studied by several methods such as solvent extraction, reduction, precipitation, ion-exchange, electro-chemical reduction, evaporation, precipitation, reverse-osmosis and adsorption (Mc Donald et al., 1996; Kumar et al., 2008). Among these existing methods adsorption has becoming the most popular and effective technique due to its better performance, availability of various adsorbents, ease of handling, low cost, low sludge generation, short operation time, and no production of secondary compounds which might be toxic (Ramakul et al., 2012). Therefore, much attention has been paid to developing inexpensive sorbents with potential applications in metal removal from aqueous solutions. However, in recent years due to the application of various biomaterials for sorption purpose the term adsorption has become more popular as biosorption. Research in the area of water treatment suggests that biosorption is an alternative for decontamination of efﬂuents containing heavy metals. In recent years a number of cost effective
S. Madala et al. / Ecological Engineering 83 (2015) 218–226
Table 1 Levels and code of variables chosen for Box–Behnken design. Variables
pH Dose (g) Contact time (min)
X1 X2 X3
x1 x2 x3
2 0.1 30
4 0.3 90
6 0.5 180
and abundant biosorbents like modiﬁed chitosan (Suguna et al., 2013; Siva Kumar et al., 2010; Suguna et al., 2011), fungi (Suguna and Kumar, 2013), agricultural waste materials (Kadirvelu and Namasivayam, 2003; Suopajarvi et al., 2014; Naiya et al., 2009), plant leaves (Salim et al., 2008; Bhattacharyya et al., 2009; Qaiser et al., 2009) have been used for the removal of metals as well as organic matter from aqueous environment. Experimental design technique is a more powerful tool in the ﬁeld of engineering, which helps in better understanding and improving efﬁciency of the process. This technique well explains the interactive effect between the parameters. Response surface methodology (RSM) is a set of mathematical and statistical techniques that has been found to be a useful method for studying the effect of several factors inﬂuencing the response as well as optimizing the parameters in the adsorption process. This method was developed by Box and Wilson (1951) and since then it has been widely used as a technique for designing the experiments. That means it can reduce the number of variables in the process by taking into account only signiﬁcant factors. This leads to improving process performance, reducing the number of variables in the process by taking into account only the most signiﬁcant factors, and also in reducing operation costs and experimental time (Box and Draper, 2007; Ghorbani et al., 2008) Tephrosia Purpuria is a species of ﬂowering plant in the pea family, Fabaceae that has a pantropical distribution. It is a common wasteland weed. In many parts, it is under cultivation as a green manure crop. It is also used traditionally as folk medicine; it is used in the treatment of leprosy, ulcers, asthma, and tumours, as well as diseases of the liver, spleen, heart, and blood. The surfaces of TPL contain several polar functional groups like hydroxyl, amino, carbonyl etc., which are effective in chelating cations.
Fig. 2. SEM pictures of (a) un-treated TPL (b) pre-treated TPL (c) TPL loaded with Pb (II).
Fig. 1. FTIR spectra of TPL (a) pre-treated TPL (b) loaded with Pb(II) ions.
The objectives of the study are; 1. Investigation of the utility of cheap and freely available TPL for the removal of Pb(II) from aqueous solutions by studying various parameters, 2. Characterization of the prepared biosorbent by using FTIR, SEM and EDAX analysis, 3. Application of three level Box–Behnken experimental designs combining with RSM to optimize the various parameters and to obtain maximum response and 4. Determination of optimum operational conditions and to analyze equilibrium data by using different isotherm models.
S. Madala et al. / Ecological Engineering 83 (2015) 218–226
2. Materials and methods 2.1. Materials All the necessary chemicals are of analytical grade and obtained from Sigma–Aldrich, St. Louis, MO, U.S.A. Stock solution (1000 mg/ L) was prepared by dissolving Pb(NO3)24H2O in distilled water. The sample solutions were prepared at the required concentrations by serial dilution of stock solution for practical use. The pH of the solution was measured with a digital pH meter (Digisum D17007, India) using solid electrode calibrated with a standard buffer solution, pH of the solutions were adjusted by the addition of 0.1 M HCl or 0.1 M NaOH. 2.2. Preparation of biosorbent T. Purpuria leaves were collected from S.V. University campus, Tirupati, India. These leaves were washed thoroughly with distilled water several times to remove dust and dried under shade. The dried leaves were converted into ﬁne powder by grinding in a mechanical grinder. The powder was sieved to average size 50– 70 mm. 25 g of powder was soaked in 1000 ml of 0.1 M HNO3 for 24 h. Then the biosorbent was ﬁltered and washed with distilled water to remove acid content. The washing was continued upto pH of the ﬁltrate was close to 7. It was dried in an oven at 313 K for 15 h to completely remove moister. Finally it was stored in an air tight bottle to protect from humidity and used as a biosorbent for the removal of Pb(II). The objective of acid pre-treatment is to remove the surface impurities and expose the binding sites of biosorbent.
spectrophotometer (Shimadzu AA-6300, Japan) with deuterium background corrector. All measurements were carried out in an air/ acetylene ﬂame. A 10 cm long slot burner head, an air-acetylene ﬂame and hallow cathode lamp at wave length 217 nm were used. The operating parameters for working elements were set as recommended by the manufacturer. Scanning electron microscopy (Model Evo15, CarlZeiss, England) has been used to study the surface morphology of the adsorbent. The sample composition and element contents were analyzed by using energy dispersive analysis system of X-ray (EDAX) (EDAX, Ltd., USA). 2.4. Batch biosorption experiments Batch adsorption experiments were carried out in Erlenmeyer ﬂasks by adding 0.5 g of TPL biomass in 100 ml of aqueous Pb(II) solution at desired initial pH, metal ion concentration and temperature. The contents were shaken on a temperature controlled water bath shaker at 200 rpm. After completion of each batch, the solution was ﬁltered using Whatman No. 42 ﬁlter paper and the ﬁltrate was analyzed for remaining metal concentration in the sample using atomic absorption spectrophotometer. The amount of metal ion sorbed per unit mass of the biosorbent (mg g1) was evaluated by using the following equation, qe ¼
ðC i C e ÞV m
The biosorption percentage of Pb(II) ions was calculated using the equation,
2.3. Characterization of biosorbent Biosorption%ðYÞ ¼ Fourier transform infrared spectroscopy (FTIR) was used to identify the functional groups present on the biosorbent. The FTIR spectra were recorded using Thermo-Nicolet FTIR, Nicolet IR200 series, Germany. Concentration of Pb(II)) in aqueous solution, after biosorption could be analyzed by a ﬂame atomic absorption
ðC i C e Þ 100 Ce
where qe (mg g1) is the adsorption capacity at equilibrium, Ci and Ce denote respectively the initial and equilibrium concentrations of metal ion (mg L1), V (L) is volume of adsorbate and m (g) is the amount of adsorbent. 2.5. Box–Behnken design and Pb(II) biosorption optimization In the present work, optimum conditions for maximum Pb(II) biosorption by TPL was determined by the three levels, three factorial Box–Behnken experimental designs combining with response surface methodology. The three important parameters affecting Pb(II) biosorption, namely pH (x1), adsorbent dose (x2) and contact time (x3) were selected as independent variables, the factor levels were coded as 1(low), 0 (central point) and 1 (high). The removal biosorption% (Y) was considered as the dependent variable. The experimental range and levels of independent variables were shown in Table 1. In the optimization process, the response can be related to the independent variables by quadratic (second degree) polynomial equation. A quadratic model, which also includes the linear model, is given below: %Y ¼ bo þ b1 x1 þ b2 x2 þ b3 x3 þ b12 x1 x2 þ b13 x1 x3 þ b23 x2 x3 þ b11 x21 þ b22 x22 þ b33 x23
Fig. 3. Energy dispersive analysis (EDAX) pattern of (a) TPL (b) TPL loaded with Pb (II).
where Y is estimate response, b0 is model constant, b1,b2 and b3 are linear coefﬁcients, b12, b13 and b23 are interaction coefﬁcients among the three factors, b11, b22 and b33 are quadratic coefﬁcients and x1, x2 and x3 are independent variables A multiple regression analysis was done to obtain the coefﬁcients and the equation could be used to estimate the response. A total of 15 experiments were needed to estimate the biosorption of Pb(II) on TPL. The accuracy of the proposed model is then identiﬁed by using analysis of variance (ANOVA). The
S. Madala et al. / Ecological Engineering 83 (2015) 218–226
Table 2 Box–Behnken model design matrix, observed response and predicted values. Trail No.
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
1 1 1 1 1 1 1 1 0 0 0 0 0 0 0
1 1 1 1 0 0 0 0 1 1 1 1 0 0 0
0 0 0 0 1 1 1 1 1 1 1 1 0 0 0
2 6 2 6 2 6 2 6 4 4 4 4 4 4 4
0.1 0.1 0.5 0.5 0.3 0.3 0.3 0.3 0.1 0.5 0.1 0.5 0.3 0.3 0.3
105 105 105 105 30 30 180 180 30 30 180 180 105 105 105
20.30 63.50 29.37 95.36 25.54 65.4 28.74 90.81 54.54 60.23 55.87 76.80 64.54 64.54 64.54
23.86 64.19 28.68 91.79 25.16 68.46 26.25 90.62 50.65 61.44 55.51 79.84 64.54 64.54 64.54
3.56 0.69 0.69 3.56 0.38 3.05 2.48 0.19 3.89 1.21 0.36 3.04 0 0 0
Table 3 Analysis of variance regression model for Pb(II) biosorption by using TPL. Source
Sum of squares (SS)
Mean square (MSS)
Model pH D T (pH)2 (D)2 (T)2 (pH) (D) (pH) (T) (D) (T) Residual Lake-of-ﬁt Pure error
9 1 1 1 1 1 1 1 1 1 5 3 2
7158.05 5571.46 570.38 270.40 129.85 123.32 58.06 432.33 9.26 4.41 46.52 46.52 6.667E-005
795.34 5571.46 570.38 270.40 129.85 123.32 58.06 432.33 9.26 4.41 9.30 15.51 795.34
85.48 598.78 61.30 29.06 13.95 13.25 6.24 46.46 0.99 0.47
<0.0001 <0.0001 0.0005 0.0030 0.0135 0.0149 0.0546 0.0010 0.3644 0.5216
4.652E + 005
R2 = 0.9906; Adj R2 = 0.9736
3. Results and discussion 100
3.1. Characterization of biosorbent
80 70 60 50 40 30 20 10
Observed Value Fig. 4. Correlation between observed and predicted values.
property of ﬁt polynomial model is represented by the coefﬁcient of determination R2. The R2 values assure a measure of how variability in the observed response values can be clariﬁed by experimental factors and their interactions (Khajeh, 2011; Singh et al., 2010; Yetilmezsoy et al., 2009).
3.1.1. FTIR spectroscopy The FTIR spectra of pre-treated and Pb(II) loaded TPL are shown in Fig. 1. The pre-treated TPL (Fig. 1(a)) showed a broad band at 3280 cm1 indicating the presence of –OH and –NH groups. The peaks at 2917 and 2854 cm1 are due to the C H stretching vibrations. The peak at 1620 cm1 indicates the symmetrical stretching vibration of C¼O in carboxylic acids and ketones. The peaks appearing around 1390 and 1240 cm1 represent N H bending and C N stretching vibrations, respectively. Absorbance at 1020 cm1 indicates COH stretching vibration. FTIR spectra of lead loaded biomass (Fig. 2(b)) showed similar characteristics bands as pre-treated TPL, but the intensity of bands was either higher or lesser. The peaks on pre-treated TPL were shifted from 3280, 2917, 2854, 2357, 2099, 1620, 1390, 1240 and 1024 cm1 to 3300, 2914, 2850, 2347, 2049, 1629, 1401, 1255 and 1019 cm1 respectively after biosorption of Pb(II). This was due to interaction of various functional groups on TPL with Pb(II). From the FTIR spectral data, it was observed that the band at 3280 cm1 corresponding to OH functional group was shifted to 3300 cm1 after adsorption, indicating the involvement of hydroxyl group with adsorption of Pb(II). 3.1.2. Surface morphology Scanning electron microscopy (SEM) was carried for the pure, pre-treated and Pb(II) loaded TPL. SEM is used to verify the
S. Madala et al. / Ecological Engineering 83 (2015) 218–226
Table 4 Multiple regression results and signiﬁcance of the components for the quadratic model. Factor
Intercept pH D T (pH)2 (D)2 (T)2 (pH)(D) (pH)(T) (D)(T)
b0 b1 b2 b3 b11 b22 b33 b12 b13 b23
17.39 26.67 17.67 0.11 2.70 39.58 1.94 104 14.24 0.04 0.25
12.3 4.3 36.8 0.1 0.5 47.9 3.5 104 4.6 0.012 0.12
1.8 6.3* 0.4 0.4 5.6 0.8 1.2 3.1* 2.9* 1.9*
0.0001 0.0001 0.0013 0.0007 0.0027 0.4932 0.2965 0.0272 0.0345 0.1230
5571.46 570.38 270.40 129.85 123.32 58.06 432.33 9.26 4.41
morphological differences between the pure, pre-treated and Pb (II) loaded TPL. There are many studies which reported the utilization of the scanning electron microscopy analysis for showing the surface modiﬁcations in the developed biosorbents (Sener et al., 2014; Calero et al., 2013). As shown in Fig. 2, untreated TPL exhibited uniform and regular structure. After pre-treatment with acid, surface of TPL becomes more porous and have different surface morphology. SEM image of Pb(II) loaded biosorbent surface is signiﬁcantly different from that of unloaded biosorbent. Due to interaction with metal ions changed the surface morphology of TPL to rough texture. 3.1.3. EDAX analysis The sample composition and elemental contents were analyzed by energy dispersive analysis system (EDAX). The EDAX spectrum for TPL powder shown in Fig. 3(a), indicates the presence of C, O, S, Mg, K, and Si, but not Pb(II) ions on the surface of pure TPL. An EDAX spectrum of Pb(II) loaded TPL is shown in Fig. 3(b). The EDAX spectrum gives characteristic peak for Pb at 2.5 keV. The EDAX pattern of TPL did not exhibit characteristic signal of Pb(II), but Pb (II) loaded TPL showed a clear signal of the presence of Pb(II). This conﬁrms the binding of Pb(II) to TPL. 3.2. Experimental design 3.2.1. Determination of regression model and statistical analysis The effect of three parameters namely solution pH, pH (2–6), Dose (0.1–0.5) and contact time (30–180) on Pb(II) biosorption was studied with the help of the analyses of regression and variance using the RSREG procedure of the Statistical Analysis System (SAS, Version 9.03). Both canonical and ridge analyses were also carried
ss SS 100
77.71 8.00 3.76 5.82 1.72 0.81 6.02 0.13 0.06
out. Statistical approaches with a Box–Behnken design was used for determination of the interaction between these factors. The number of experiments needed to investigate the optimization of biosorption process are 27 ((3)3), this was reduced to 15 by using a Box–Behnken experimental design. The results of this limited number of experiments provided a statistical model that was used to identify trends in high removal for the adsorption process. The data in Table 2, were used to ﬁt the polynomial second order equation to ﬁnd out the relationship between the variables and response, polynomial model representing the biosorption percentage (response) as a function of initial pH, dose and contact time. The equation below illustrates the relationship of the three variables and y. Y% ¼ 17:39 þ 26:68ðpHÞ 17:68ðDÞ 0:11ðTÞ 2:7ðpHÞ2 39:6ðDÞ2 þ 1:94 104 ðTÞ2 þ 14:2ðpHÞðDÞ þ 0:04ðpHÞðTÞ0:25ðDÞðTÞ
In Table 2, the percent removal of Pb(II) by TPL predicted by the ﬁnal quadratic model along with corresponding values observed were included. The agreement between the yield predicted by the model and the experimental data is very strong, with a small difference in the percent removal. Furthermore, the model capability was checked using the F-test and determination coefﬁcient R2. The analysis of variance ANOVA (Table 3) showed that this regression model was highly signiﬁcant with a very low probability value (P < 0.0001) and with F value of 58.3. The suitability of model was further conﬁrmed by the determination coefﬁcient (R2). The value of adjusted R2 (0.9736) indicates only 1.7% of the total variation and the value of determination coefﬁcient ((R2 = 0.9906) indicates good relation between the
Normal Plot of Residuals
Normal % Probability
Std Error of Design
0.600 0.400 0.200 0.000
95 90 80 70 50 30 20 10 5
Externally Studentized Residuals Fig. 5. Standard error design of the model adsorbent dose verses pH holding contact time at central level.s
Fig. 6. Normal plot of studentized residuals verses normal % probability.
S. Madala et al. / Ecological Engineering 83 (2015) 218–226
experimental and predicted values of the response. Fig. 4 indicates a satisfactory correlation between the observed and predicted values of lead removal efﬁciency. As seen in Fig. 4 the points around the diagonal line shows a good ﬁt of the model, since the deviation between experimental and predicted values was less, as similarly reported by Zhang and Zheng, (2009) and Yetilmezsoy et al. (2009). The regression coefﬁcients along with the corresponding P values are listed in Table 4. P value is used as a tool to check the signiﬁcance of each of the coefﬁcients. The greater the magnitude of the t value and smaller the P value, the more signiﬁcant are the parameters in the regression model (Yetilmezsoy et al., 2009; Yetilmezsoy and Saral, 2007), and P values showed that the ﬁrst order main effects of pH, dose and contact time were signiﬁcant. Based on the sum of squares obtained from the ANOVA, the contributions percentage (PC) for each term were calculated and listed in Table 4. As seen in Table 4, the initial pH of solution showed a very high level of signiﬁcance of 77.7%. Dose (D), contact time (T) and (pH)2 also have high signiﬁcance as compared to the remaining parameters as similarly done by other researchers (Meng et al., 2007; Khajeh, 2011). In Table 4, the ﬁnal part of ANOVA was completed in the same way to obtain total PC values for ﬁrst-order, quadratic and interaction terms. The results show that the contribution of ﬁrst order demonstrated the highest level of signiﬁcance than the PC values of quadratic and interaction terms. In Fig. 5, the minimum value of standard error design 0.12 around the centroid and maximum prediction variance 0.5 at the design points indicated that the present model is used to navigate the design space for the present study. The plot (Fig. 6) of studentized residuals against predicted probability values showed the homogeneously distributed data about either side of line indicating the suitability of the model in the present study. The canonical analysis of the response surface was performed to determine the shape of the ﬁtted response and estimated stationary point. According to the model, the predicted response at the stationary point (X1 = 0.24, X2 = 1.03 and X3 = 319.0 shown in Table 5) was 5.15 mg/L. The three Eigen values had different signs, indicating that the stationary point for this model was a saddle point as shown in Table 6, as similarly reported by Pasma et al., 2013. Therefore, the estimated surface did not have a unique optimum and ridge analysis was performed to determine the optimum. The results of ridge analysis in Table 7 show that all variables tested were positively related to the response and optimum level of them was found as 5.5 pH, 0.4 g of adsorbent dose and 136 min contact time with 90.6 % of Pb(II) removal by TPL. 3.2.2. Three dimensional (3D) response surfaces In order to gain a better understanding of the inﬂuence of the independent variables and their interactions on the dependent variables, 3D response surface plots for the measured responses were formed based on the model equation (Eq. (4)) in this work. Since the regression model has three independent variables, one variable was held at constant at the central level at each plot, thus a total of three response 3D plots were produced for response (Khajeh, 2011). Fig. 7 shows the 3D response surface as the function of two variables, one variable was held at constant at the central level for each plot. Table 5 Canonical analysis of response surface based on coded data. Factor
X1 X2 X3
1.88 6.65 5.65
0.24 1.03 319.00
Table 6 Eigen values and Eigen vectors. Eigenvalues
1.03 3.69 12.09
0.31 0.02 0.95
0.73 0.64 0.22
0.61 0.76 0.22
Stationary point is a saddle point.
Solution pH is one of the most important parameters effecting the heavy metal removal from aqueous solution and thus effect of pH was studied in the range of 2–6. Fig. 7 shows the interactive effect of two variables pH (2–6) and adsorbent dose (0.1–0.5) on Pb (II) removal percentage and the response was increased with increase in adsorbent dosage. The results show that the quantitative removal (95%) was obtained in the pH between 5 and 6. The removal percentage of Pb(II) was increased with increase of pH from 2.0 to 6.0. The pH could not maintain above 6.0, as the Pb(II) got precipitated at higher pH. A lower percentage removal was observed under highly acidic conditions, as most of the active sites of TPL were occupied by H+ and H3O+. Hence biosorbent surface becomes more positively charged resulting in electrostatic repulsion between adsorbate and adsorbent. As pH increases, deprotonation starts and the metal ion complexed with TPL, hence removal percentage increases. Point of zero charge determination of the biosorbent is also important in elucidating the biosorption mechanism. The pHpzc of TPL biomass is 4.4 and the sorption of positively charged species will be favoured only at pH > pHpzc. For the pH at PZC, the surface charge of biosorbents is neutral and the electrostatic forces between metal ion and surface of adsorbents are balanced. This balance is disturbed when pH is deviated from pHPZC. At pH < pHPZC, surface charge is positive which results in an electrostatic repulsion with metal ion and causes low metal sorption. At pH > pHPZC, the surface charge of the biosorbent becomes negative and Pb(II) ions in solution are attracted to their surface. Maximum sorption is likely to occur at pH values greater than pHPZC when adsorbents have a net negative charge. Hence maximum adsorption of Pb(II) was obtained at pH 5.4. In the case of interactive effect between pH and contact time a similar trend was observed. Fig. 7 shows interactive effect between adsorbent dose and contact time on removal percentage of Pb(II). From the ﬁgure we found that with increase of adsorbent dose there is signiﬁcant increase in percentage removal. This is due to the increase of number of active sites and overall surface area of the adsorbent, thus removal percentage was increased with increase of adsorbent dose. But contact time does not exhibit any signiﬁcant changes.
Table 7 The RSREG procedure. Estimated ridge of maximum response for variable Y. Coded radius
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
65.79 68.62 71.33 73.93 76.46 78.91 81.31 83.67 86.00 88.31 90.60
2.19 2.19 2.18 2.15 2.13 2.09 2.08 2.07 2.09 2.14 2.24
Uncoded factor values pH
4.00 4.18 4.36 4.52 4.68 4.83 4.97 5.09 5.22 5.33 5.45
0.30 0.31 0.32 0.32 0.33 0.34 0.36 0.37 0.38 0.39 0.41
1050.0 106.71 108.82 111.32 114.18 117.35 120.77 124.41 128.22 132.17 136.24
S. Madala et al. / Ecological Engineering 83 (2015) 218–226
1 1 1 ¼ þ qe qm K L C e qm
60 40 20
60 40 20
180 150 120 90 60
1 1 þ K L C0
1 log qe ¼ log K F þ logC e n 100 80
60 40 20
Contact time (min)
where C0 (mg L1) is the initial amount of adsorbate and KL is the Langmuir sorption constant (L mg1) described above. RL indicates the nature of adsorption. Adsorption is favourable when the value of RL is between 0 and 1.The RL parameter is considered as a more reliable indicator of the adsorption. The value of RL indicates the type of the isotherm to be either unfavourable (RL > 1), linear (RL = 1), favourable (0 < RL < 1) or irreversible (RL = 0). The RL values for adsorption of Pb(II) ions on to TPL are shown in Table 9 at the temperature range from 298 K to 318 K. The values show that the adsorption is favourable for the metal ions by TPL. The Freundlich model assumes a heterogeneous surface with involvement of sites of different energies in adsorption process. The Freundlich model is given as:
where qe (mg g1) is the amount of metal ion adsorbed per unit mass of adsorbent, Ce (mg L1) is the equilibrium concentration of metal ions, qm is the monolayer biosorption capacity of the adsorbent (mg g1) and KL (mg L1) is the Langmuir equilibrium constant, respectively. The values of qm and KL were determined from the plot of 1/qe vs 1/Ce (Fig. 8). The values were shown in Table 8. The essential characteristics of Langmuir isotherm can be expressed by a dimensionless constant called separation factor (or equilibrium parameter), RL, which is deﬁned by Weber and Chakkravorti as (Weber and Chakkravorti, 1974):
Fig. 7. Response surface plots obtained from Box–Behnken design for removal efﬁciency of lead.
3.3. Adsorption isotherm modelling Isotherms are the equilibrium relations between the concentration of adsorbate and the surface of the adsorbent material. Further sorption isotherm studies are essential to design an adsorption system. From the isotherms the maximum adsorption capacity can be obtained. These data provide information on the capacity of the adsorbent or the amount required to remove a unit mass of pollutant under the system conditions. Langmuir and Freundlich isotherms are the most common isotherms describing solid–liquid adsorption system. The Langmuir model assumes that adsorption takes place at speciﬁc homogeneous sites on the surface of the adsorbent and also, when a site is occupied by an adsorbate molecule, no further adsorption can take place at this site. The Langmuir equation is expressed as:
where KF ((mg g1) (L mg1)1/n) is related to the biosorption capacity and 1/n is an empirical parameter relating the biosorption intensity. Values of n and KF for Pb(II) were calculated at different temperatures (298, 308 and 318 K) from the slope and intercept of log Ce vs log qe and are presented in Table 8. The R2 values of Freundlich isotherm are greater than 0.980, but lower than 0.991, indicate that this model is unable to describe adequately the relationship between the amounts of Pb(II) adsorbed by the biomass and its equilibrium concentration in the solution. The values of KF are found to be increasing with increase in temperature suggesting that adsorption process is endothermic in nature. The values of n between 1 and 10 represent a favourable sorption. Langmuir isotherm gave R2 values close to unity, indicating that the adsorption of Pb(II) onto TPL is best described by Langmuir model. The biosorption capacity of different biosorbent show the different sorption capacities and is mainly due to the type, surface structure, morphology and the type of functional groups present in the biosorbent, and the maximum biosorption capacity of the biosorbent obtained from Langmuir isotherm shown in Table 11. TPL showed comparable biosorption capacity towards Pb(II) with previous literature results. 3.4. Thermodynamic analysis The thermodynamic parameters such as enthalpy change (DH0), the entropy change (DS0) and the Gibbs free energy change (DG0) for the sorption process were calculated from the variation of Langmuir constant (KL) with temperature (T) using well known relations,
DG0 ¼ RTlnK L
S. Madala et al. / Ecological Engineering 83 (2015) 218–226 Table 10 Values of thermodynamic parameters for the adsorption of Pb(II) on TPL.
298 K 308 K 318 K
298 308 318
0.558 1.889 3.156
DG0 ¼ DH0 T DS0
lnK L ¼
1/Ce Fig. 8. Langmuir isotherm model plot for the biosorption of Pb(II) onto TPL.
Table 8 Langmuir and Freundlich isotherm constants for adsorption of Pb(II) on TPL at different temperatures.
298 308 318
76.9 90.9 100.0
0.016 0.023 0.033
0.994 0.999 0.996
1.74 2.93 4.77
1.29 1.34 1.55
0.991 0.991 0.980
Table 9 RL values for Pb(II) ions on TPL. Temperature (K)
298 308 318
where R is the universal gas constant (8.314 10 kJ mol K1) and KL is Langmuir constant. Enthalpy change due to adsorption of metal ions by TPL over the temperature range studied can be determined from the linear plots of ln KL against 1/T using the least squares analysis The values of DG0, DH0 and DS0 for sorption of Pb (II) by TPL at different temperature (298–318 K), given in Table 10, show that DG0 is small and negative but decreases with increasing temperature. The negative values of DG0 demonstrate the process to be spontaneous and positive values of DH0 indicating that the process requiring some energy input from the outside. Hence the process of removal of Pb(II) on TPL is endothermic in nature. The magnitude of the DH0 value gives an indication on the type of adsorption, which can be either physical or chemical. The enthalpy or heat of biosorption, ranging from 2.1–20.9 kJ/mol corresponds a physical sorption as it ranges from 20.9 to 418.4 kJ/mol in case of a chemical sorption (Sari and Tuzen., 2008; Deng et al., 2007). The biosorption heat of Pb(II) onto TPL was of the same magnitude as the heat of chemisorption. The DH0 value obtained (Table 10) showed that chemisorption plays a dominant role in the adsorption process. The positive value of DS0 suggested the increase of randomness at the solid/solution interface during the biosorption of metal ions on TPL. 3.5. Adsorption mechanism
Initial metal ion concentration (mg/L) 10
0.999 0.813 0.752
0.999 0.465 0.377
0.998 0.303 0.232
To understand the adsorption mechanism of Pb(II) adsorption onto TPL, from the FTIR spectral data, it was observed that the band at 3280 cm1 corresponding to OH functional group was shifted to 3300 cm1 after adsorption indicating that the involvement of hydroxyl group with adsorption of Pb(II). Whereas no signiﬁcant changes were observed in case of other functional groups .This conﬁrms that chemisorption plays a role in Pb(II) biosorption onto
Table 11 Comparison of maximum adsorption capacity (mg/g) of TPL for Pb(II) on different biosorbents from the literature. Biosorbent
Adsorption capacity (mg/g)
Ficus religiosa leaves Olive tree pruning (OTP) Un treated OTP H2SO4-OTP HNO3-OTP Bael leaves (Aegle marmelos) NaOH and HCl treated Melia azedarach L. leaves Cinnamomum camphora leaves Lichen (Parmelina tiliaceae) Macro fungus (Amanita rubescens) Macro fungus (Lactarius scrobiculatus) Sporopollenin biomass Cactus (Opuntia ﬁcus indica) cladodes Tephrosia purpurea leaves
Qaiser et al., 2009 Calero et al., 2013
65.62 85.09 104 35.06 and 28.5 respectively 73.15 75.8 38.4 56.2 6.1 3.5 100.0
Chakravarty et al., 2010 Asima et al., 2015 Chen et al., 2010 Uluozlu et al., 2008 Sari & Tuzen, 2009 Anayurt et al. 2009 Sener et al., 2014 Barka et al., 2013 Present study
S. Madala et al. / Ecological Engineering 83 (2015) 218–226
TPL. The DH0 value obtained (Table 10) showed that chemisorption plays a dominant role in the adsorption process. 4. Conclusion Tephrosia purpurea leaves powder was used for the removal of Pb(II) from aqueous medium over a wide range of concentrations. The following conclusions are arrived at based on the results of the present study: Pre-treatment with acid causes the increase in the roughness of the surface. Response Surface methodology (RSM) based on three factor, three level Box–Behnken design was used to optimize the various inﬂuencing parameters on Pb(II) biosorption process. The optimum conditions for maximum removal percentage of Pb (II) by TPL were found to be at pH 5.4 with 0.4 g of adsorbent and 136 min contact time. Equilibrium data were ﬁtted to Langmuir and Freundlich isotherm models and it is found that the equilibrium data were best described by the Langmuir isotherm model. The maximum monolayer biosorption capacity was found to be 100 mg g1 at 318 K. The thermodynamic results show the feasibility, spontaneous and endothermic nature of biosorption of Pb(II) on TPL. Based on the results, it can be concluded that the TPL is an effective biosorbent for the removal of Pb(II) from aqueous medium. Acknowledgement One of the authors (MS) is thankful to DST, New Delhi, India for the ﬁnancial support of this research project, SR/WOS-A/CS/76/ 2011. References Anayurt, R.A., Sari, A., Tuzen, M., 2009. Equilibrium, thermodynamic and kinetic studies on biosorption of Pb(II) and Cd(II) from aqueous solution by macro fungus (lactarius scrobiculatus) biomass. Chem. Eng. J. 151, 255–261. Asima, K., Zumaira, S., Misbah, 2015. Removal of heavy metal ions by chemically treated Melia azedarach L. leaves. J. Environ. Chem. Eng. 3, 944–952. Barka, N., Abdennouri, M., El Makhfouk, M., Qourzal, S., 2013. Biosorption characteristics of cadmium and lead onto eco-friendly dried cactus (Opuntia ﬁcus indica) cladodes. J. Environ. Chem. Eng. 1, 144–149. Bhattacharyya, G., Sarma, J., Sarma, A., 2009. Azadirachta indica leaf powder as a biosorbent for Ni(II) in aqueous medium. J. Hazard. Mater. 165, 271–278. Box, G.E.P., Wilson, K.B., 1951. On the experimental attainment of optimum conditions. J. R. Stat. Soc. Ser. B 13, 1–45. Box, G.E.P., Draper, N.R., 2007. Response Surface, Mixtures and Ridge Analyses, 2nd ed. Wiley, USA. Calero, M., Perez, A., Blazquez, G., Ronda, A., Martin-Lara, M.A., 2013. Characterization of chemically modiﬁed biosorbents from olive tree pruning for the biosorption of lead. Ecolog. Eng. 58, 344–354. Chakravarty, S., Ashok Mohanty, Nag Sudha, T., Upadhyay, A.K., Konar, J., Sircar, J.K., Madhukar, A., Gupta, K.K., 2010. Removal of Pb(II) ions from aqueous solution by adsorption using bael leaves (Aegle marmelos). J. Hazard. Mater. 173 (1–3), 502– 509. Chen, H., Zhao, J., Dai, G., Wu, J., Yan, H., 2010. Adsorption characteristics of Pb(II) from aqueous solution onto a natural biosorbent, fallen Cinnamomum camphora leaves. Desalination 262 (1-3), 174–182. Deng, L., Su, Y., Su, H., Wang, X., Zhu, X., 2007. Sorption and desorption of lead(II) from wastewater by green algae Cladophora fascicularis. J. Hazard. Mater. 143, 220–225. Ghorbani, F., Younesi, H., Ghasempouri, S.M., Zinatizadeh, A.A., Amini, M., Daneshi, A., 2008. Application of response surface methodology for optimization of cadmium biosorption in an aqueous solution by Saccharomyces cerevisiae. Chem. Eng. J. 145, 267–275.
Kadirvelu, K., Namasivayam, C., 2003. Activated carbon from coconut coir pith as metal adsorbent: adsorption of Cd(II) from aqueous solution. Adv. Environ. Res. 7, 471–478. Khajeh, M., 2011. Response surface modelling of lead pre-concentration from food samples by miniaturised homogenous liquid–liquid solvent extraction. Box– Behnken design. Food Chem. 129, 1832–1838. Kumar, J.R., Lee, H.I., Lee, J.Y., Kim, J.S., Sohn, J.S., 2008. Comparison of liquid–liquid extraction studies on platinum(IV) from acidic solutions using bis (2,4,4trimethylpentyl) monothiophosphinic acid. Sep. Purif. Technol. 63, 184–190. Mc Donald, M., Mila, I., Scalbert, A., 1996. Precipitation of metal ions by plant polyphenols: optimal conditions and origin of precipitation. J. Agric. Food Chem. 44, 599–606. Meng, H., Hu, X., Neville, A., 2007. A systematic erosion-corrosion study of two stainless steels in marine conditions via experimental design. Wear 263, 355–362. Naiya, T.K., Bhattacharya, A.K., Mandal, S., Das, S.K., 2009. The sorption of lead(II) ions on rice husk ash. J. Hazard. Mater. 163, 1254–1264. Pasma, S.A., Daik, R., Maskat, M.Y., Hassan, O., 2013. Application of Box–Behnken design in optimization glucose production from oil palm empty fruit bunch cellulose. J. Polym. Sci. 1–8. Qaiser, S., Saleemi, A.R., Umar, M., 2009. Biosorption of lead from aqueous solution by Ficus religiosa leaves: batch and column study. J. Hazard. Mater. 166, 998– 1005. Ramakul, P., Yanachawakul, Y., Leepipatpiboon, N., Sunsandee, N., 2012. Biosorption of palladium(II) and platinum(IV) from aqueous solution using tannin from Indian almond (Terminalia catappa L.) leaf biomass: kinetic and equilibrium studies. Chem. Eng. J. 193–194, 102–111. Salim, R., Al-Subu, M., Dawod, E., 2008. Efﬁciency of removal of cadmium from aqueous solutions by plant leaves and the effects of interaction of combinations of leaves on their removal efﬁciency. J. Environ. Mange. 87, 521–532. Sari, A., Tuzen, M., 2008. Biosorption of total chromium from aqueous solution by red algae (Ceramium virgatum): equilibrium, kinetic and thermodynamic studies. J. Hazard. Mater. 160, 349–355. Sari, A., Tuzen, M., 2009. Kinetic and equilibrium studies of biosorption of Pb(II) and Cd(II) from aqueous solution by macrofungus (Amanita rubescens) biomass. J. Hazard. Mater. 64, 1004–1011. Sener, M., Harikishore Kumar Reddy, D., Kayan, B., 2014. Biosorption properties of pretreated sporopollenin biomass forlead(II) and copper(II): application of response surface methodology. Ecol. Eng. 68, 200–208. Singh, R., Chadetrik, R., Kumar, R., Bishnoi, K., Bhatia, D., Kumar, A., Bishnoi, N.R., Singh, N., 2010. Biosorption optimization of lead(II), cadmium(II) and copper(II) using response surface methodology and applicability in isotherms and thermodynamics modelling. J. Hazard. Mater. 174, 623–634. Siva Kumar, N., Suguna, M., Venkata Subbaiah, M., Subba Reddy, A., Phani Kumar, N., Krishnaiah, A., 2010. Adsorption of phenolic compounds from aqueous solutions on to chitosan-coated perlite beads as biosorbent. Ind. Eng. Chem. Res. 49, 9238–9247 (American Chemical Society). Suguna, M., Siva Kumar, N., Subba Reddy, A., Boddu, V.M., Krishnaiah, A., 2011. Biosorption of lead(II) and from aqueous solutions on glutaraldehyde crosslinked chitosan beads. Can. J. Chem. Eng. 9999, 1–11. Suguna, M., Siva Kumar, N., Sreenivasulu, V., Boddu, V.M., Krishnaiah, A., 2013. Equilibrium, kinetics and thermodynamics of cadmium(II) biosorption on to composite chitosan biosorbent. Arabian J. Chem.. Suguna, M., Kumar, N.S., 2013. Equilibrium, kinetics and thermodynamic studies on biosorption of lead(II) and cadmium(II) from aqueous solutions by using polypores biomass. Ind. J. Chem. Technol. 20, 57–69. Suguna, M., Siva Kumar, N., Sreenivasulu, V., Krishnaiah, A., 2014. Removal of Pb(II) from aqueous solutions by using chitosan coated zero valent iron nanoparticles. Sep. Sci. Technol. 49, 1613–1622. Suopajarvi, T., Liimatainen, H., Karjalainen, M., Upola, H., Niinimaki, J., 2014. Lead adsorption with sulfonated wheat pulp nanocelluloses. J. Water Process Eng. (in press). Tabaraki, R., Nateghi, A., Ahmady-Asbchin, S., 2014. Biosorption of lead(II) ions on Sargassum ilicifolium: application of response surface methodology. Int. Biodeterior. Biodegrad. 93, 145–152. Uluozlu, O.D., Sari, A., Tuzen, M., Soylak, M., 2008. Biosorption of Pb(II) and Cr(III) from aqueous solution by lichen (Parmelina tiliaceae) biomass. Biores. Technol. 99 (8), 2972–2980. WHO, 2011. Who Guidelines For Drinking-water Quality, 4th ed. WHO Press, Geneva. http://www.who.int. Weber, T.W., Chakkravorti, R.K., 1974. Pore and solid diffusion models for ﬁxed-bed adsorbers. Aiche J. 20, 228–238. Yetilmezsoy, K., Demirel, S., Vanderbei, R.J., 2009. Response surface modelling of Pb (II) removal from aqueous solution by Pistacia vera L.: Box–Behnken experimental design. J. Hazard. Mater. 171, 551–562. Yetilmezsoy, K., Saral, A., 2007. Stochastic modelling approaches based on neural network and linear-nonlinear regression techniques for the determination of single droplet collection efﬁciency of counter current spray towers. Environ. Model. Assess. 12, 13–26. Zhang, Z., Zheng, H., 2009. Optimization for decolorization of azo dye acid green 20 by ultrasound and H2O2 using response surface methodology. J. Hazard. Mater. 172, 1388–1393.