ELSEVIER

Available online at www.sciencedirect.com

ScienceDirect

Karbala International Journal of Modern Science xx (2016) 1 — 11

http://www.journals.elsevier.com/karbala-international-journal-of-modern-science/

Optimisation using central composite design (CCD) and the desirability function for sorption of methylene blue from aqueous

solution onto Lemna major

Bikash Sadhukhan, Naba K. Mondal*, Soumya Chattoraj

Environmental Chemistry Laboratory, Department of Environmental Science, The University of Burdwan, Golapbag, Burdwan 713104, West

Bengal, India

Received 14 December 2015; revised 8 March 2016; accepted 9 March 2016

Abstract

Water pollution due to contamination of dye containing effluents is a great threat to water body. A study on the biosorption of methylene blue (MB) onto low-cost Lemna major biomass was conducted and the process parameters were optimized by response surface methodology (RSM). A two-level, four-factor central composite design (CCD) has been employed to determine the effect of various process parameters namely initial concentration (600—1000 mg L-1), bioadsorbent dose (0.20—1.50 g/100 mL), pH (5—12) and stirring rate (250—800 rpm) on MB uptake from aqueous solution. By using this design a total of 30 biosorption experimental data were fitted. The regression analysis showed good fit of the experimental data to the second-order polynomial model with coefficient of determination (R2) value of 0.9978 and model F-value of 953.48. The optimum conditions of initial concentration (1000 mg L-1), adsorbent dose (0.2 g), pH (5) and stirring rate (251.51 rpm) were recorded from desirability function. The adsorption isotherm data were best described by both Freundlich and Langmuir models with a maximum adsorption capacity of 488 mg MB g_1 L. major biomass at 30 °C which is higher than that available with adsorbents used by past researchers. Finally the pseudo second order kinetic model described the MB biosorption process with a good fitting (R2 = 0.999). The adsorbent was characterised by scanning electron micrograph (SEM) and Fourier transform infrared spectroscopy (FTIR). © 2016 The Authors. Production and hosting by Elsevier B.V. on behalf of University of Kerbala. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).

Keywords: Methylene blue; Biosorption; Lemna major biomass; Central composite design

Abbreviations: CCD, central composite design; RSM, response surface methodology; MB, methylene blue; ANOVA, analysis of variance; FTIR, Fourier transform infrared spectra; SEM, scanning electron microscope.

* Corresponding author. Tel.: +91 9434545694; fax: +91 (0) 3422634200.

E-mail address: nkmenvbu@gmail.com (N.K. Mondal).

Peer review under responsibility of University of Kerbala.

1. Introduction

Methylene blue (MB), a cationic dye, is used for colouring paper, temporary hair colourant, dyeing cottons, silk and wood. Therefore, it can easily be found in wastewater that can cause some harmful effects, such as heart beat increase, vomiting, shock, cyanosis, jaundice, and tissue necrosis in humans [1]. This necessitates the removal of MB from water by

http://dx.doi.org/10.1016/j.kijoms.2016.03.005

2405-609X/© 2016 The Authors. Production and hosting by Elsevier B.V. on behalf of University of Kerbala. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).

+ MODEL

B. Sadhukhan et al. / Karbala International Journal of Modern Science xx (2016) 1—11

appropriate methods. Amongst the numerous techniques of dye removal, the biosorption process is one of the effective processes that have been successfully employed for colour removal, recovery, and recycling of dyes from the wastewater [2].

Recently, cost effective, ecofriendly and easily available adsorbent particularly of biomaterial has received a huge interest. Various low-cost bioadsorbent materials are used in the literature such as Indian rosewood sawdust [3], modified peat-resin particle [4], hazelnut shell [5], modified polysaccharide [6], coconut husk [7], cellulose based waste [8], cotton stalk [9], sawdust composite [10], neem leaf powder [11], wheat shells [12], tamarind fruit shell [13], pre-treated rice husk [14], citrus fruit peel [15], peat [16], modified lignin from sugarcane bagasse [17], spent coffee grounds [18], fallen phoenix tree's leaves [19], wood apple shell (Feronia acidissima) [20], fly ash [21], activated carbon from waste biomass [22], agricultural waste [23], Parthenium hysterophorus [24], wood [25], rejected tea [26], Posidonia oceanica (L.) fibres [27], waste newspaper [28], untreated coffee husks [29], yellow passion fruit waste [30], growing Lemna minor [31], neem bark dust [32], biosolid [33], natural and modified clay [34], animal bone meal [35], Egyptian kaolin [36], eggshells and eggshell membrane [37], rice husk [38], spent activated clay [39], poplar leaf [40], etc. All these have been tested for removal of MB from aqueous systems, but no information is available in literature on the improved removal of MB by Lemna major bioadsorbent. It is inexpensive and easily available; this could make it a viable candidate as an economical adsorbent for removing unwanted hazardous components from contaminated water.

To optimize the process parameters for the sorption process the combined effect of initial concentration, bioadsorbent dose, pH and stirring rate a central composite design in response surface methodology (RSM) by Design Expert Version 7.0.3 [Stat-Ease] is used. RSM (response surface methodology), an empirical modelling technique, is used to find out the relationship between a set of experimental factors and observed results. It basically consists three major steps: performing statistically designed experiments, estimating the coefficients in a mathematical model and predicting the response and checking the adequacy of the model [41]. In this study a class of three level complete factorial designs (central composite design) was used to determine the optimization values of the operating variables.

Many researchers have studied the applicability of low cost alternative materials for removal of methylene

blue from aqueous medium. However, the adsorption capacities of such materials are not up to the mark. Therefore, there is tremendous demand to explore a suitable biomaterial which can effectively remove methylene blue from aqueous solution. With this back drop present study was undertaken to evaluate the efficiency of L. major biomass as an effective adsorbent towards removal of methylene blue. Finally the experimental data were analyzed by fitting to a second order polynomial model, which was statistically validated by performing analysis of variance (ANOVA) and lack-of-fit test to evaluate the significance of the model. Desirability function is used to find optimum conditions where the maximum adsorption capacity was obtained for the removal of MB using L. major biomass to ensure the high uptake capacity at low adsorbent dosage and high MB concentration to reduce the time consumption.

2. Materials and methods

2.1. Preparation of adsorbent

L. major, a floating macrophyte, was collected from the wetland of Burdwan, West Bengal, India. It was washed 4—5 times with distilled water and then sundried for 10—12 days followed by drying in hot air oven at 343 ± 1 K for 48 h. The dried biomaterial was crushed and sieved through 250 mesh screen. The biomaterial was obtained washed and suctioned thoroughly with double distilled water to remove the impurities and then stored in closed glass container for further use as a bioadsorbent [42].

2.2. Reagents

A.R. grade chemicals were collected from M/S Merck India Pvt. Ltd., and used in the present study without further purification. Fig. 1 is the chemical structure of MB (basic blue 9), C16H18N3SCl with 1max = 665 nm. Double distilled water was used to prepare all reagents and standards. All glassware were cleaned by HNO3 and rinsed with double distilled water. For the adjustment of pH of MB solutions 0.1 mol L"1 NaOH and 0.1 mol L"1 HCl were used.

Fig. 1. Chemical structure of methylene blue (MB).

+ MODEL

B. Sadhukhan et al. / Karbala International Journal of Modern Science xx (2016) 1—11

2.3. Adsorbent characterization and instrumentation

The characterization of bioadsorbent was determined with the help of spectroscopic and quantitative analysis. The surface area of the adsorbent was determined by Quantachrome surface area analyzer (Model- NOVA 2200C, USA). Using flame photometer (Model No. SYSTRONICS 126) the concentrations of sodium and potassium were evaluated. Magnetic stirrer (TARSONS, Spinot digital model MC02, CAT No. 6040, S. No. 173) was utilised for stirring the solution. The Fourier transform infrared (FTIR) spectra of the bioadsorbent were studied using Fourier transform infrared spectropho-tometer (PERKIN-ELMER, FTIR, Model-RX1 Spectrometer, USA) working in the range of400—4000 cm-1. The scanning electron microscopy (SEM) analysis was used by scanning electron microscope (HITACHI, S-530 and ELKO Engineering, B.U. BURDWAN) at 25 kV for the study of surface morphology of the bioadsorbent used.

2.4. The physiochemical properties

The physico-chemical properties of the bio-adsorbent were reported in earlier investigation [42].

2.5. Batch adsorption process

The adsorption experiments were performed in triplicate, and mean values were used in the data analysis. The spectrophotometric determination of MB was done by the addition of bioadsorbent to the dye solution of known strength and stirred for 20 min each time on a magnetic stirrer and then centrifuged at 6000 rpm for 2 min. The supernatant liquid is directly taken for the absorbance at 665 nm using UV—VIS spectrophotometer (Systronics, Vis double beam Spec-tro 1203). The control experiments were performed without the addition of bioadsorbent under the same condition. The similar adsorption processes were also noted varying time duration, initial dye concentration at optimal pH, adsorbent dose and the stirring rate.

The influence of pH (5.0—12.0), initial MB concentration (600—1000 mg L-1), stirring rate (250—800 rpm), and bioadsorbent dose (0.20—1.50 g/ 100 mL) were evaluated during the study. Samples were collected from the flasks at predetermined time intervals for analyzing the residual MB concentration in the solution. The amount of MB adsorbed at equilibrium (Qe) was determined by using the following equation:

(Ci - Ce)V

where Ci and Ce are MB concentrations (mg L-1) before and after adsorption, respectively, V is the volume of the dye solution (L) and m is the weight of the adsorbent (g). Removal of MB in percentage was calculated from the following equation:

(C- - C )

Removal (%) = ^-^ x 100

The study of adsorption isotherm was analysed using Freundlich (Eq. (3)) and Langmuir isotherm equations (Eq. (4)) as follows:

log Qe = log kf + (-)log Ce (3)

Ce _ Ce 1

Qe Qm kLQm

where Qe (mg g-1) and Ce (mg L-1) are the solid phase concentration and the liquid phase concentration of adsorbate at equilibrium, respectively. Qm (mg g-1) is the maximum adsorption capacity and kL (L mg- J) is the energy of adsorption. The Freundlich isotherm constants kf and (1/n) can be evaluated by plotting of log (Qe) vs log (Ce). kf (mg g-1)(L/mg)1/n) and n indicates the adsorption capacity and adsorption intensity or the heterogeneity factor, respectively. The Langmuir constants Qm and kL can be estimated by plotting (Ce/Qe) vs Ce [10].

The kinetic study was determined using two most accepted models namely, Lagergren's Pseudo-first-order kinetic (Eq. (5)) and Pseudo-second-order kinetic models (Eq. (6)). The equations are as follows:

log(Qe - Qt) = log Qe - 2-3031

kzQ2 Q,

where Qt and Qe are the amount of MB adsorbed (mg g-1) at time t and at equilibrium and k1 (min-1) is the Lagergren rate constant of first order adsorption and k2 (gm g-1 min-1) is the second order adsorption rate constant, respectively [43].

2.6. Design of experiments

2.6.1. Central composite design (CCD)

CCD has been widely used statistical method based on the multivariate nonlinear model for the optimization of process variables of biosorption and also used to determine the regression model equations and operating conditions from the appropriate experiments. It is also useful in studying the interactions of the various parameters affecting the process [44]. The CCD was applied in this present study to determine the optimum process variables for biosorption of MB using L. major biomass. The CCD was used for fitting a second-order model which requires only a minimum number of experiments for modelling [45]. The CCD consists of

+ MODEL

4 B. Sadhukhan et al. / Karbala International Journal of Modern Science xx (2016) 1—11

2n factorial runs (coded to the usual ± notation) with 2n axial runs (±a, 0, 0, ..., 0), (0, ±a, 0, 0, ..., 0), ..., (0, 0, ..., ±a) and accentor runs (six replicates, 0,0,0,..., 0). The number of factors n increases the number of runs for a complete replicate of the design which is given in Eq. (7).

N = 2n + 2n + nc (7)

Basically the optimization process involves three major steps: (1) performing the statistically designed experiments, (2) estimating the coefficients in a mathematical model, and (3) predicting the response and checking the adequacy of the model [46—50]. An empirical model was developed to correlate the response to the bioadsorption process and is based on second order quadratic model for removal of MB using L. major biomass as given by Eq. (8) in order to analyse the effect of parameter interactions.

Y = b0 + E bx + X ir bjjXXj + XT biix2i + £ (8)

i=1 i=1 j=1 i=1

where Yis the response variable; bo is the intercept; bi, bj and bii are coefficients of the linear effect, double interactions; xi, Xj are the independent variables or factors and £ is error.

3. Results and discussion

3.1. Characterisation of bioadsorbent

The physico-chemical characteristics of the bio-adsorbent along with SEM and FTIR spectra of L. major have been done in the present study [51,52]. The results of FTIR and SEM study are presented in Figs. 2 and 3. From Fig. 2 it is clear that there are distinct peaks at 1046 cm-1, 1319 cm-1 and 1605 cm-1 which corresponds the functional groups such as C—O, N=O and C=O stretching, respectively. Similarly, Fig. 3

—I-'-1---1-'-1-•-1-'-1—

700 1400 2100 2800 3500 4200

Wavenumber(cm ')

Fig. 2. FTIR study of Lemna major after treatment with MB.

Fig. 3. SEM of Lemna major after treatment with MB.

demonstrated that the surface is rough with huge porosity. In the present study, the adsorption of MB onto the L. major was extensively done by batch process and the results of study are supported by the FTIR and SEM analysis (Figs. 2 and 3).

3.2. Central composite design studies

In the present study, four important parameters, initial concentration (A), bioadsorbent dose (B), pH (C) and stirring rate (D) (Table 1) were considered. Consequently, A, B, C and D were chosen as the independent variables while the removal of MB at equilibrium (Y) was selected as the response (dependent variable) of the study. The final central composite design obtained for percentage removal of MB with significant terms was quadratic as suggested by the software, and is given as:

Removal (Y) = +77.69 + 2.18*A + 3.06*B - 5.93*C

- 0.64*D - 4.18*A*B - 2.21*A*C

- 0.065*A*D + 6.11*B*C

+ 4.57*B*D - 1.43 *C*D + 0.94*A2

- 2.90*B2 - 2 . 38*C2 + 3 .17*D2

Eq. (9) reveals how the individual variables (quadratic) or double interaction affected MB removal from aqueous solution by L. major biomass as a

Table 1

Variables and levels considered for percentage of removal of MB.

Name (factor) Units Low High

Initial concentration (A) mg.L-1 600 1000

Bioadsorbent dose (B) g 0.20 1.50

pH(C) — 5 12

Stirring rate (D) rpm 250 800

+ MODEL

B. Sadhukhan et al. / Karbala International Journal of Modern Science xx (2016) 1—11 5

bioadsorbent. The negative coefficient values indicate that individual or double interactions factors negatively affect MB adsorption (i.e. adsorption percentage decreases), whereas positive coefficient values mean that factors increase MB adsorption in the tested range.

The adequacy (Table 2) of the models was justified (model summary statistics, Table 3) by the analysis of variance (ANOVA). The ANOVA of MB removal is

given in Table 4.The model F-value of 953.48 implies that the model is significant. The model F-value is the ratio of mean square for the individual term to the mean square for the residual. The Prob > F value is the probability of F-statistics value and is used to test the null hypothesis. The parameters having F-statistics probability value less than 0.05 are said to be significant [53].

Table 2

Adequacy of the model tested.

Source

Sum of squares

Mean square

F value

p-Value Prob > F

Mean vs total Linear vs mean 2FI vs linear Quadratic vs 2FI Cubic vs quadratic Residual Total

180,865.49 747.99 1319.24 523.43 1.15 1.76 183,459.06

1.00 4.00 6.00 4.00 11.00 4.00 30.00

180,865.49 187.00 219.87 130.86 0.10 0.44 6115.30

2.53 7.94 674.26 0.24

0.0655 0.0002 <0.0001 0.9740

Suggested Aliased

Table 3

Model summary statistics.

Source

Sum of squares

Mean square

F value

p-Value Prob> F

Linear 2FI

Quadratic Cubic Pure error

1843.821 524.5777 1.148546 0.00 1.76

22 16 12 1

83.81006 32.78611 0.095712 0.000225 0.587533

142.6473 55.80297 0.162905 0.000383

0.0008 0.0034 0.9910 0.9856

Suggested Aliased

Table 4

Analysis of variance (ANOVA), test of significance for MB uptake on Lemna major.

Source Sum of squares df Mean square F value p-Value Prob> F

Model 2590.66 14.00 185.05 953.48 <0.0001 Significant

A — conc 114.58 1.00 114.58 590.39 <0.0001

B — dose 164.49 1.00 164.49 847.54 <0.0001

C — pH 657.98 1.00 657.98 3390.30 <0.0001

D — rate 9.98 1.00 9.98 51.40 <0.0001

AB 279.56 1.00 279.56 1440.46 <0.0001

AC 77.88 1.00 77.88 401.29 <0.0001

AD 0.07 1.00 0.07 0.35 0.5639

BC 597.31 1.00 597.31 3077.72 <0.0001

BD 333.98 1.00 333.98 1720.85 <0.0001

CD 32.94 1.00 32.94 169.72 <0.0001

A2 24.35 1.00 24.35 125.47 <0.0001

B2 131.82 1.00 131.82 679.21 <0.0001

C2 88.48 1.00 88.48 455.90 <0.0001

D2 225.84 1.00 225.84 1163.65 <0.0001

Residual 2.91 15.00 0.19

Lack of fit 1.15 12.00 0.10 0.16 0.9910 Not significant

Pure error 1.76 3.00 0.59

Cor total 2593.58 29.00

Std. dev. 0.44 R-squared 0.9989

Mean 77.65 Adj R-squared 0.9978

C.V. % 0.57 Pred R-squared 0.9981

Adeq precision

151.554

+ MODEL

B. Sadhukhan et al. / Karbala International Journal of Modern Science xx (2016) 1—11

In this case A, B, C, D, AC, AD, CD, A2, B2, D2 are significant model terms. The "Lack of Fit F-value" of 0.16 implies the lack of fit is not significant relative to the pure error. There is 99.10% chance that "Lack of Fit F-value" is significant. This large value of F could occur due to noise. Therefore, it can be concluded that initial concentration (A), bioadsorbent dose (B), pH (C) and stirring rate (D) play an important role in case of MB adsorption. It implied that a good correlation between input and output variables could be drawn by the model developed. The value of actual and predicted removal percentage, leverage, internally studentized residuals, externally studentized residuals, DFFITS and Cook's distance of the data can be obtained from diagnostic case statistics (Table 3). The results portray that the leverage value was within 0— 1. The number of standard deviation separating actual and predicted values can be measured by internally studentized residuals. The limit of the internally studentized residuals is ±3 sigma [54]. The normality assumption was satisfied as the residual plot approximated along a straight line [55] (Fig. 4). The analysis of diagnostic case statistics of data shows that the model fits well to optimize the independent variables for the removal of MB.

A high value of the adjusted determination coefficient (RAdi = 0 •9978) was estimated. This result means that 99.78% of the total variation on MB adsorption data can be described by the selected model. The value of the signal-to-noise ratio (adequate precision ratio) was found to be 151.554, indicating the model has an adequate signal. Because the adjusted determination coefficient and adequate precision ratio exceeded 70% and 4, respectively, the quadratic model can be used to explore the design space and to find the optimal conditions of this process. A comparison of the effects of all factors at the optimal conditions of MB adsorption to the L. major biomass was performed by

using a perturbation plot (Fig. 5). The steep curvature of initial concentration (A), bioadsorbent dose (B), pH (C) and stirring rate (D) indicate the MB adsorption is highly affected by the operating parameters.

3.3. Optimisation using desirability function

In numerical optimization, we chose the desired goal for each factor and response. The possible goals were: to maximize, minimize, target, within range, none (for responses only) and set to an exact value (factors only). A minimum and a maximum level must be provided for each parameter included. A weight can be assigned to each goal to adjust the shape of its particular desirability function. The goals are combined into an overall desirability function. Desirability is an objective function that ranges from zero outside of the limits, to one at the goal. The program seeks to maximize this function. The goal seeking begins at a random starting point and proceeds up the steepest slope to a maximum. There may be two or more maximums because of curvature in the response surfaces and their combination in the desirability function. Starting from several points in the design space improves the chances for finding the 'best' local maximum [51—53]. A multiple response method was applied for optimization of any combination of four goals, namely initial concentration, bioadsorbent dose, pH, stirring rate and percentage removal of MB. Fig. 6 demonstrates the desirability values of the numerical optimization procedure in which the criterion was set, in range for pH (5—12), "minimum" for adsorbent dose (0.2 g) and "maximum" for initial concentration (1000 mg L_1) and within the range of stirring rate (250—800 rpm) to analyse economically viable optimal

Perturbation

Internally Studentized Residuals

Fig. 4. Normal probability plot of the studentised residuals.

Deviation from Reference Point (Coded Units)

Fig. 5. Perturbation plot of MB adsorption.

+ MODEL

B. Sadhukhan et al. / Karbala International Journal of Modern Science xx (2016) 1—11 7

600.00

1 ooo.oo

cono ■

1 ooooo

5.00 12.OO

pH = 5.00

250.0D 800.00

rate = 251 .51

Desirability = Q.999

smoval = 97.8379

Fig. 6. Desirability ramp of MB adsorption.

condition. The objective of this process was to find the maximum removal efficiency by utilizing less bio-adsorbent dosage. By seeking from 30 starting points (Diagonastic case statistics, Table 5) in the response changes the best local maximum was found at pH 5,

adsorbent dose at 0.2 g, initial concentration 1000 mg L-1 and stirring rate at 251.51 rpm. At this condition MB removal is close to 97.84% and desirability 0.999. These optimum values were checked experimentally which resulted 97.6% of MB removal by L. major

Table 5

Diagonastics case statistics.

Standard Actual Predicted Residual Leverage Internally Externally Influence on Cook's Run

order value value studentized studentized fitted value distance order

residual residual DFFITS

1 80.67 80.62 0.05 0.63 0.18 0.17 0.22 0.00 28

2 97.92 97.89 0.03 0.63 0.10 0.09 0.12 0.00 5

3 73.80 73.75 0.05 0.59 0.17 0.16 0.19 0.00 12

4 74.36 74.31 0.05 0.59 0.19 0.19 0.22 0.00 27

5 63.74 63.82 -0.08 0.64 -0.29 -0.28 -0.37 0.01 15

6 72.20 72.27 -0.07 0.64 -0.25 -0.24 -0.31 0.01 13

7 81.34 81.39 -0.05 0.62 -0.18 -0.18 -0.23 0.00 9

8 73.06 73.12 -0.06 0.62 -0.21 -0.20 -0.26 0.00 26

9 73.24 73.20 0.04 0.62 0.15 0.15 0.19 0.00 2

10 90.26 90.21 0.05 0.62 0.18 0.18 0.23 0.00 10

11 84.67 84.61 0.06 0.58 0.23 0.22 0.26 0.00 11

12 84.96 84.90 0.06 0.58 0.22 0.21 0.25 0.00 21

13 50.59 50.68 -0.09 0.63 -0.35 -0.34 -0.45 0.01 4

14 58.77 58.87 -0.10 0.63 -0.38 -0.37 -0.48 0.02 16

15 86.45 86.53 -0.08 0.61 -0.29 -0.28 -0.35 0.01 24

16 77.91 78.00 -0.09 0.61 -0.32 -0.31 -0.39 0.01 23

17 77.09 77.07 0.02 0.61 0.08 0.08 0.10 0.00 3

18. 85.84 85.81 0.03 0.61 0.12 0.11 0.14 0.00 19

19 75.00 74.71 0.29 0.17 0.73 0.72 0.32 0.01 6

20 72.24 72.21 0.03 0.77 0.13 0.13 0.23 0.00 18

21 79.90 80.02 -0.12 0.76 -0.54 -0.52 -0.93 0.06 22

22 70.00 69.38 0.62 0.21 1.57 1.66 0.85 0.04 8

23 89.29 89.31 -0.02 0.53 -0.08 -0.07 -0.08 0.00 17

24 89.22 89.08 0.14 0.59 0.48 0.47 0.56 0.02 14

25 78.00 77.69 0.31 0.16 0.77 0.75 0.33 0.01 30

26 77.00 77.69 -0.69 0.16 -1.71 -1.84 -0.81 0.04 29

27 78.41 77.69 0.72 0.16 1.78 1.94 0.85 0.04 20

28 76.83 77.69 -0.86 0.16 -2.13 -2.47 -1.08 0.06 1

29 77.00 76.63 0.37 0.13 0.90 0.89 0.35 0.01 25

30 79.61 80.22 -0.61 0.13 -1.49 -1.56 -0.62 0.02 7

+ MODEL

B. Sadhukhan et al. / Karbala International Journal of Modern Science xx (2016) 1—11

biomass. The maximum adsorption capacity was calculated as 488 mg g-1 by using Eq. (1). The present adsorption capacity result is absolutely promising over other biomaterial such as Parthenium hystrophorous

weed which shows only 23.8 mg g 3.4. Adsorption isotherm

Freundlich and Langmuir isotherms were used to describe the equilibrium characteristics of adsorption of MB onto L. major biomass. The linear form of isotherms and their constants are given in Table 6. The experimental data obtained at equilibrium was fitted satisfactorily with Freundlich isotherm (figure not shown). The Freundlich isotherm reveals the multilayer adsorption [51]. Moreover, solution pH is an important factor, significantly affecting the adsorption of MB [57—59]. In the entire isotherm study, pH of the dye solution was fixed at 5.0.

3.5. Adsorption kinetic

The pseudo-first-order and pseudo-second-order kinetic models were tested to investigate the rate of adsorption of MB on L. major biomass. The linear form of adsorption kinetics [14,52] and their constants are presented in Table 7. From Table 7, it is confirmed that the adsorption of MB on L. major biomass at pH 5 followed the Pseudo-second order reaction (figure not shown). It is clear from Table 7 that the Pseudo-second-order kinetic model showed excellent linearity with high correlation coefficient (R2 > 0.99) at all the studied concentrations in comparison to the first-order kinetic model [13]. In addition the calculated Qe values also agree with the experimental data obtained from pseudo-second-order kinetic model [55]. Similar results for the adsorption kinetics of MB by magnetic iron oxide nanosorbent were also reported by Pacurariu et al. [60].

Table 6

Summary of parameters for isotherm models.

Isotherm model Equation Constants

Langmuir isotherm Freundlich isotherm Ce — Ce | 1 qe qm ' kL?m log qe = log kf + Q log Ce KL = 0.067 L mg-1, R2 = 0.991 n = 1.67, KF = 89.33 mg g-1(L/mg)(1/n), R2 = 0.999

Where qe (mg g-1) and ce (mgL-1) are the solid phase concentration and the liquid phase concentration of adsorbate at equilibrium respectively, qm (mg g-1) is the maximum adsorption capacity and kL (L mg-1) and KF (mg g-1) (L mg-1) and is the adsorption equilibrium constant. n is the heterogeneity factor.

Table 7

Summary of parameters for kinetic models.

Kinetic model Equation Constants

Pseudo-first-order Pseudo-second-order log(qe - qt) = log qe - os qt = + qe R2 = 0.975, K1 R2 = 0.999, K2 = 0.096 min-1 = 0.002 g mg-1 min-1

Where qt and qe are the amount of methylene blue adsorbed (mg g 1) at time t and at equilibrium and K1 (min 1) is the Lagergren rate constant of first order adsorption and k2 (g mg-1 min-1) is the second order adsorption rate constant.

Table 8

Reviewed results representing the adsorption capacity of different biomaterials for the adsorption of MB and their optimized pH values for maximum adsorption.

Adsorbent

Adsorption capacity

Reference with year

Activated carbon from newspaper Indian rosewood sawdust Neem leaf powder Wheat shells Peat

Activated carbon from waste biomass Hazelnut shell Sawdust composite Growing Lemna minor Poplar leaf

Modified Strychnos potatorum seeds Neem bark dust (NBD) Lemna major

11.43 5.00

390.00 mg g-1 11.80 mg g-1

16.56 8.82

10.93 mg g-1

49.00 mg g-1

488.00 mg g-1

[28] Okada et al., 2003 [3] Garg et al., 2004

[11] Bhattacharyya et al., 2005

[12] Bulut et al., 2006 [16] Fernandes et al., 2007 [22] Karagoz et al., 2008 [5] Dogen et al., 2009 [10] Ansari et al., 2010 [35] Reema et al., 2011 [40] Xiuli et al., 2012

[55] Senthamarai et al., 2013 [32] Sadhukhan et al., 2014 In this study.

+ MODEL

B. Sadhukhan et al. / Karbala International Journal of Modern Science xx (2016) 1—11

3.6. Comparison of L. major biomass with other sorbents

A comparative study of the maximum MB uptake capacity of L. major biomass has been carried out with other reported sorbents. The maximum MB sorption capacity of other reported sorbents under different pH and are presented in Table 8. From Table 8, it is observed that the maximum sorption capacity of L. major biomass for MB is comparable and moderately higher than that of many corresponding sorbent materials. The easy availability and cost effectiveness of L. major biomass are some additional advantages, which make it better bioadsorbent for treatment of MB.

4. Conclusion

In this study, L. major biomass was tested and evaluated as a possible bioadsorbent for removal of MB, a cationic dye from its aqueous solution using batch sorption technique. The biosorption studies were carried out as a function of initial concentration (A), bioadsorbent dose (B), pH (C) and stirringrate (D). Percentage removal of the dye molecule decreased with increase in initial dye concentration while it increased with increase in bio-adsorbent dose, pH, and stirring rate up to a certain level. So removal of MB by L. major biomass is very much effective within this concentration range. The hierarchical quadratic model navigates adequately the response surface space based on the adjusted determination coefficient (RAdj = 0 •9978) and the adequate precision ratio which measure the signal to noise ratio is 151.554. At the optimum conditions, the predicted removal efficiency achieved is close to 100% of MB removal from aqueous solutions, when using L. major biomass. Finally, the reported results in this research demonstrate the feasibility of Box—Benken model to optimize the experiments for MB removal by adsorption using L. major biomass as a low-cost bioadsorbent.

Conflict of interest

The authors have declared no conflict of interest. Acknowledgement

The authors are thankful to all faculty members and non-teaching staff of the Department of Environmental Science, University of Burdwan, West Bengal, India for providing infrastructural facilities and active moral support towards completion of this work.

References

[1] A. Bhatnagar, A.K. Minocha, Assessment of the biosorption characteristics of lychee (Litchi chinensis) peel waste for the removal of Acid Blue 25 dye from water, Environ. Technol. 31 (1) (2010) 97—105.

[2] S. Banerjee, M.G. Dastidar, Use of jute processing wastes for treatment of wastewater contaminated with dye and other or-ganics, Bioresour. Technol. 96 (2005) 1919.

[3] V.K. Garg, M. Amita, R. Kumar, R. Gupta, Basic dye (methylene blue) removal from simulated wastewater by adsorption using Indian rosewood sawdust, Dyes Pigments 63 (2004) 243—250.

[4] Q. Sun, L. Yang, The adsorption of basic dyes from aqueous solution on modified peat-resin particle, Water Res. 37 (2003) 1535—1544.

[5] M. Dogen, H. Abak, M. Alkan, Adsorption of methylene blue onto hazelnut shell: kinetics and mechanism and adsorptive energy, J. Hazard. Mater. 164 (2009) 172—181.

[6] A.T. Paulino, M.R. Guilherme, A.V. Reis, G.M. Campese, E.C. Muniz, J. Nozaki, Removal of methylene blue dye from an aqueous media using superabsorbent hydrogel supported on modified polysaccharide, J. Colloid Interface Sci. 301 (1) (2006) 55—62.

[7] I.A.W. Tan, A.L. Ahmad, B.H. Hameed, Adsorption of basic dye on high-surface-area activated carbon prepared from coconut husk: equilibrium, kinetic and thermodynamic studies, J. Hazard. Mater. 154 (2008a) 337—346.

[8] G. Annadurai, R.S. Juang, D.J. Lee, Use of cellulose-based wastes for adsorption of dyes from aqueous solutions, J. Hazard. Mater. B92 (2002) 263—274.

[9] D. Hui, L. Jianjiang, L. Guoxue, Z. Genlin, W. Xugen, Adsorption of methylene blue on adsorbent materials produced from cotton stalk, Chem. Eng. J. 172 (2011) 326—334.

[10] R. Ansari, Z. Mosayebzadeh, Removal of basic dye methylene blue from aqueous solution using sawdust and sawdust coated with polypyrrole, J. Iran. Chem. Soc. 7 (2) (2010) 339—350.

[11] K.G. Bhattacharyya, A. Sharma, Kinetics and thermodynamics of methylene blue adsorption onneem (Azadirachtaindica) leaf powder, Dyes Pigments 65 (2005) 51—59.

[12] Y. Bulut, H. Aydin, A kinetics and thermodynamics study of methylene blue adsorption on wheat shells, Desalination 194 (1—3) (2006) 259—267.

[13] S. Chowdhury, P. Das, Pseudo-second-order kinetic model for biosorption of methylene blue onto tamarind fruit shell: comparison of linear and non-linear methods, Biorem. J. 14 (4)

(2010) 196—207.

[14] S. Chowdhury, P. Das, Linear and non-linear regression analyses for binary sorption kinetics of methylene blue and safranin onto pre-treated rice husk, Bioremediat. J. 15 (2)

(2011) 99—108.

[15] S. Dutta, A. Bhattacharyya, A. Ganguly, S. Gupta, S. Basu, Application of response surface methodology for preparation of low-cost adsorbent from citrus fruit peel and for removal of methylene blue, Desalination 275 (2011) 26—36.

[16] A.N. Fernandes, C.A.P. Almeida, C.T.B. Menezes, N.A. Debacher, M.M.D. Sierra, Removal of methylene blue from aqueous solution by peat, J. Hazard. Mater. 449 (2007) 412—419.

[17] N.C. Filho, E.C. Venancio, M.F. Barriquello, A.A.W. Hechenleitner, E.A.G. Pineda, Methylene blue adsorption onto modified lignin from sugarcane bagasse, Eclet. Quim. 32 (4) (2007) 63—70.

+ MODEL

B. Sadhukhan et al. / Karbala International Journal of Modern Science xx (2016) 1—11

[18] A.S. Franca, S.O. Leandro, E.F. Mauro, Kinetics and equilibrium studies of methylene blue adsorption by spent coffee grounds, Desalination 249 (2009) 267-272.

[19] R. Han, W. Zou, Yu, S. Cheng, Y. Wang, J. Shi, Biosorption of methylene blue from aqueous solution by fallen phoenix tree's leaves, J. Hazard. Mater. 141 (2007) 156-162.

[20] S. Jain, R. Jayaram, Removal of basic dyes from aqueous solution by low-cost adsorbent: wood apple shell (Fer-oniaacidissima), Desalination 250 (3) (2010) 921-927.

[21] S. Wang, Y. Boyjoo, A. Choueib, A comparative study of dye removal using fly ash treated by different methods, Chemo-sphere 60 (2005) 1401-1407.

[22] S. Karagoz, T. Tay, S. Ucar, M. Erdem, Activated carbons from waste biomass by sulphuric acid activation and their use on methylene blue adsorption, Bioresour. Technol. 99 (14) (2008) 6214-6222.

[23] P.S. Kumar, R.V. Abhinaya, K.G. Lashmi, V. Arthi, R. Pavithra, V. Sathyaselvabala, Adsorption of methylene blue dye from aqueous solution by agricultural waste: equilibrium, thermodynamics, kinetics, mechanism and process design, Colloid J. 73 (2011) 651-661.

[24] H. Lata, V.K. Garg, R.K. Gupta, Removal of a basic dye from aqueous solution by adsorption using Parthenium hyster-ophorus: an agricultural waste, Dyes Pigments 74 (2007) 653-658.

[25] G. McKay, V.J.P. Poots, Kinetics and diffusion processes in colour removal from effluent using wood as an adsorbent, J. Chem. Technol. Biotechnol. 30 (1980) 279-292.

[26] N. Nausuha, B.H. Hameed, A.T. Din, Rejected tea as a potential low-cost adsorbent for the removal of methylene blue, J. Hazard. Mater. 175 (2010) 126-132.

[27] M.C. Ncibi, B. Mahjoub, M. Seffen, Kinetic and equilibrium studies of methylene bluebiosorption by Posidoniaoceanica (L.) fibres, J. Hazard. Mater. 139 (2007) 280-285.

[28] K. Okada, N. Yamamoto, Y. Kameshima, A. Yasumori, Adsorption properties of activated carbon from waste newspaper prepared by chemical and physical activation, J. Colloid Interface Sci. 262 (2003) 194-199.

[29] L.S. Oliveira, A.S. Franca, T.M. Alves, S.D.F. Rocha, Evaluation of untreated coffee husks as potential biosorbents for treatment of dye contaminated waters, J. Hazard. Mater. 155 (2008) 507-512.

[30] F.A. Pavan, E.C. Lima, S.L.P. Dias, A.C. Mazzocato, Methylene blue biosorption from aqueous solutions by yellow passion fruit waste, J. Hazard. Mater. 150 (2008) 703-712.

[31] A.R.M. Reema, A.P. Saravanan, M.D. Kumar, S. Renganathan, Accumulation of methylene blue dye by growing Lemna minor, Sep. Sci. Technol. 46 (6) (2011) 1052-1058.

[32] B. Sadhukhan, N.K. Mondal, S. Chattoraj, Biosorptive removal of cationic dye from aqueous system: a response surface methodological approach, Clean Tech. Environ. Policy 16 (6) (2014) 1015-1025.

[33] M. Sariglu, U.A. Aatay, Removal of methylene blue by using biosolid, Global Nest J. 8 (6) (2006) 113-120.

[34] G.K. Sarma, S. Sengupta, K.G. Bhattacharyya, Methylene blue adsorption on natural and modified clays, Sep. Sci. Technol. 46 (2006) 1602-1614.

[35] R. Slimani, A. Anouzla, Y. Abrouki, Y. Ramly, S.E.l. Antri, R. Mamouni, S. Lazar, M.E.l. Haddad, Removal of a cationic dye, methylene blue, from aqueous media by the use of animal bone meal as a new low cost adsorbent, J. Mater. Environ. Sci. 2 (1) (2011) 77-87.

[36] S.J. Tarek, H.A.G. Hany, S.I. Hanan, H.A. Islam, El-Maksoud, Removal of methylene blue by two zeolites prepared from naturally occurring Egyptian kaolin as cost effective technique, Solid State Sci. 13 (2011) 1844—1851.

[37] W.T. Tsai, J.M. Yang, C.W. Lai, Characterization and adsorption properties of eggshells and eggshell membrane, Bioresour. Technol. 97 (2006) 488—493.

[38] V. Vadivelan, K.V. Kumar, Equilibrium, kinetics, mechanism, and process design for the sorption of methylene blue onto rice husk, J. Colloid Interface Sci. 286 (2005) 90—100.

[39] C.H. Weng, Y.F. Pan, Adsorption of cationic dye (methylene blue) onto spent activated clay, J. Hazard. Mater. 144 (1) (2007) 335—362.

[40] H. Xiuli, N. Xiaona, M. Xiaojian, Adsorption characteristics of methylene blue on poplar leaf in batch mode: equilibrium, kinetics and thermodynamics, Korean J. Chem. Eng. 29 (4) (2012) 494—502.

[41] R. Bhaumik, N.K. Mondal, S. Chattoraj, J.K. Datta, Application of response surface methodology for optimization of fluoride removal mechanism by newely developed biomaterial, Am. J. Anal. Chem. 4 (2013) 404—419.

[42] S. Chattoraj, N.K. Mondal, B. Das, P. Roy, B. Sadhukhan, Carbaryl removal from aqueous solution by Lemna major biomass using response surface methodology and artificial neural network, J. Environ. Chem. Eng. 2 (2014) 1920—1928.

[43] M.M. AbbdEl-Latif, A.M. Ibrahim, M.F. El-Kady, Adsorption equilibrium, kinetics and thermodynamics of methylene blue from aqueous solutions using biopolymer oak sawdust composite, J. Am. Sci. 6 (6) (2010) 267—283.

[44] S. Chattoraj, N.K. Mondal, B. Sadhukhan, Predictability by Box—Behnken model for carbaryl adsorption by soils of Indian origin, J. Environ. Sci. Health, B 48 (2013) 626—636.

[45] R. Kumar, R. Singh, N. Kumar, K. Bishnoi, N.R. Bishnoi, Response surface methodology approach for optimization of biosorption process for removal of Cr(VI), Ni (II) and Zn (II) ions by immobilized bacterial biomass sp. Bacillus brevis, Chem. Eng. J. 146 (2009) 401—407.

[46] N.K. Mondal, S. Chattoraj, B. Sadhukhan, B. Das, Evaluation of carbaryl sorption in alluvial soil, Songklanakarin J. Sci. Technol. 35 (6) (2013) 727—738.

[47] L. Peng, X. Xu, H. Mu, C.E. Hoy, J. Adler-Nissen, Production of phospholipids by lipase-catalyzed acidolysis: optimization using response surface methodology, Enzyme Microb. Technol. 31 (2002) 523.

[48] B.K. Körbahti, M.A. Rauf, Application of response surface analysis to the photolytic degradation of basic red 2 dye, Chem. Eng. J. 138 (1) (2008) 166—171.

[49] D.C. Montgomery, Design and Analysis of Experiments, fourth ed., John Wiley & Sons, 1996.

[50] R.H. Myers, D.C. Montgomery, Response Surface Methodology: Process and Product, 2002.

[51] S. Chattoraj, N.K. Mondal, B. Das, P. Roy, B. Sadhukhan, Biosorption of carbaryl from aqueous solution onto Pistia stratiotes biomass, Appl. Water Sci. 4 (2014) 79—88.

[52] S. Senthilkumaar, P. Kalaamani, C.V. Subburaam, Liquid phase adsorption of Crystal violet onto activated carbons derived from male flowers of coconut tree, J. Hazard. Mater. 136 (3) (2006) 800—808.

[53] P. Saha, S. Chowdhury, S. Gupta, I. Kumar, R. Kumar, Assessment on the removal of malachite green using tamarind fruit shell as biosorbent, Clean Soil Air Water 38 (2010) 437—445.

+ MODEL

B. Sadhukhan et al. / Karbala International Journal of Modern Science xx (2016) 1—11

[54] S. Arivoli, M. Hema, S. Parthasarathy, N. Manju, Adsorption dynamics of methylene blue by acid activated carbon, J. Chem. Pharm. Res. 2 (5) (2010) 626-641.

[55] C. Senthamarai, K.P. Senthil, M. Priyadharshini, P. Vijayalakshmi, K.V. Vinoth, P. Baskaralingam, et al., Adsorption behavior of methylene blue dye onto surface modified Strychnos potatorum seeds, Environ. Prog. Sustain. Energy 32 (2013) 624-632.

[56] M. Mulugeta, B. Lelisa, Removal of methylene blue (Mb) dye from aqueous solution by bioadsorption onto untreated Par-thenium hystrophorous weed, Mod. Chem. Appl. 2 (2014) 146, http://dx.doi.org/10.4172/2329-6798.1000146.

[57] T. Madrakian, A. Afkhami, M. Ahmadi, H. Bagheri, Removal of some cationic dyes from aqueous solutions using magnetic

modified multi-walled carbon nanotubes, J. Hazard. Mater. 196 (2011) 109-114.

[58] S.K. Giri, N.N. Das, G.C. Pradhan, Magnetite powder and kaolinite derived from waste iron ore tailings for environmental applications, Powder Technol. 214 (2011) 513-518.

[59] R. Rakhshaee, Rule of Fe0 nano-particles and biopolymer structures in kinds of the connected pairs to remove Acid Yellow 17 from aqueous solution: simultaneous removal of dye in two paths and by four mechanisms, J. Hazard. Mater. 197 (2011) 144-152.

[60] C. Pacurariu, O. Paska, R. Lanos, S.G. Muntean, Effective removal of methylene blue from aqueous solution using a new magnetic iron oxide nanosorbent prepared by combustion synthesis, Clean. Technol. Environ. Policy (2015), http:// dx.doi.org/10.1007/s10098-015-1041-7.