Hymavathi D; Prabhakar G

Research Article

Austin Chem Eng. 2017; 4(1): 1049.

Studies on the Removal of Pb(II) from Aqueous Solutions by Adsorption with E. Globulus Leaf Powder through Response Surface Methodology

Hymavathi D and Prabhakar G*

Department of Chemical Engineering, S.V. University College of Engineering, Tirupati, Andhra Pradesh, India

*Corresponding author: Garimella Prabhakar, Department of Chemical Engineering, S.V. University College of Engineering, Tirupati, Andhra Pradesh, India

Received: April 12, 2017; Accepted: May 17, 2017; Published: May 24, 2017

Abstract

An exhaustive batch experimental investigation to treat lead-laden waters by adsorption using leaf powder of Eucalyptus globulus is reported. Based on full factorial approach, optimum conditions to yield a removal of 96.58% are identified as initial ion concentration of 20mg/L, sorbent dosage 25g/L at a pH of 5.0 and a temperature of 303K. Freundlich model adequately represents the equilibrium and the maximum adsorption capacity is found to be 6.803mg/g. The process is endothermic and spontaneous. Adsorption follows second order kinetics. The adsorption process compares well with other similar studies.

Keywords: Pb(II) removal; Adsorption; E. globules; Equilibrium; Kinetics; RSM

Nomenclature

C_o: Initial Concentration of Pb(II) solution (mg/L); C_e: Equilibrium Concentratin of Pb(II) solution (mg/L); q_e: Amount Sorbed Per Unit Weight Of Biosorbent At Equilibrium (mg/g); q_et: Equilibrium Metal Uptake Capacity At Time (mg/g); q_max: Maximum Metal Uptake Capacity (mg/g); q_t: Metal Uptake Capacity At Time t (mg/g); K_diff: Intra-Particle Diffusion Rate Constant; c: Thickness Of Boundary Layer; m: Mass Of Adsorbent (g); w: Adsorbent Dosage (g/L); V: Volume of the Pb(II) solution (L); b=k_L: Affinity Constant Or Energy To Biosorption (L/g); n: number(dimensionless); K_f : Freundlich coefficient (mg/g); K₁ : First Order Equilibrium Rate Constant (1/min); K₂: Second Order Equilibrium Rate Constant (g/mg.min); A_T: Temkin Adsorption Intensity, (L/g); A_E: Elovich constant; B_E: Initial Adsorption Rate; E: Mean Free Energy Of Sorption Per Of The Molecule Of Sorbate; b_T: Heat Of Adsorption Constant

Abbreviations

R: Universal Gas Constant (8.314mol-1K-1); X₁: Initial Concentrations of Pb(II) (mg/L); X₂: pH(dimensionless); X₃: Adsorbent Dosage (g/L); X₄: Absolute Temperature (K); RSM: Response Surface Methodology; ANOVA: Analysis Of Variance; Max: Maximum Metal Uptake Capacity; Diff: Intra-Particle Diffusion Rate Constant

Introduction

Water contamination is mainly due to the accumulation of heavy metals, from different industries like metallurgical, battery, electroplating and metal finishing industries, chemical manufacturing and tanneries [1,2]. Among heavy metals viz., Co, Cr, Zn, Cd, Cu, Pb, Hg, As, Al and Ni, Pb(II) is one of the most toxic ion causing serious health issues related to liver damage, nervous system, kidneys, reproductive system, neurological activity and also causes high hypertension [3,4]. Maximum allowable concentration for Pb(II) ion limit value of 0.01mg/L in drinking water recommended by World Health Organization [5] and the permissible level of Pb(II) in wastewater is 0.05mg/L given by the Environmental Protection Agency (EPA) [6]. Conventional methods - chemical precipitation [7], reverse osmosis [8], ion exchange [9], coagulation [10], electro dialysis [11] and ultra filtration [12] were studied earlier. These methods were found to be not so promising due to incomplete metal removal, high reagent and high energy requirements.

Adsorptive treatment of lead waters was probed with different types of waste materials and a few of them are as follows. Nile rose plant [13], chaff [14], rice husk [15], coir fiber waste [16], banana stems [17], wheat bran[18], coffee grounds [19], tree ferns [20], palm kernel fibres [21], crop milling waste-black gram husk [22], pomegranate peels [23] , peanut skins [24], cone biomass of Pinus sylvestris [25], carbon derived from agricultural waste[26], almond shell [27], native and chemically treated olive stone [28], residue of all spice [29], ceder leaf ash [30], cashew nut shell [31], Peanut shell [32], native and chemically treated olive tree pruning [33], pine cone shell [34] and rapeseed biomass [35] were used in the studies and Eucalyptus globulus , leaf powder of cheap and widely available which is preferred as a natural antimicrobial agent, industrial solvent, and deodorant, is tried for the removal of lead from aqueous systems [36,37].

In present work, E. globules L., used as an adsorbent for removal of Pb(II) from aqueous solution and E. glubulus belongs to the family of myrtaceae. Eucalyptus bark was used for the removal of chromium [38] and mercury [39]. Till –to- date, non literature is available on E. globulus leaf powder for removal of Pb(II).

Materials and Methods

Preparation of stock solution

All chemical compounds used are of analytical grade (Merck).

A Stock solution of 500ppm Pb(II) is prepared by dissolving 0.4055mg of 98.5% pure Pb(NO₃)₂ in 500ml of distilled water. It is diluted to different levels, appropriate to the study. pH of the solution is adjusted using 0.1N NaOH and 0.1N H₂SO₄. Final Pb(II) ion concentration is obtained by Inductively Coupled Plasma Optical Emission Spectroscope (Perkin Elmer model Optima 8000). FTIR (ALPHA interferometer (ECO-ATR)), Bruker, Germany) in the range of 4000 – 500 cm-1 is employed to identify the functional groups that are involved in adsorption. Elemental composition is recorded by Scanning Electron Microscope, (SEM–EVO MA 15) with Electron Dispersive X- Ray Spectroscope of OXFORD INSTRUMENTS (Inca Penta FET x3).

Preparation and activation of biosorbent

E. globulus leaves are collected in the University Campus, water washed thoroughly to clear the surface impurities, and is then sun dried. They are grounded into a fine powder and 63μm – size particles are collected. The particles are further washed with water to remove coloring agents, dried at room temperature and are stored in air tight bottles for further studies.

Biosorption studies

Sufficient numbers of flasks, each containing 50ml solution of 20mg/L Pb(II) are taken and the solution pH adjusted. A known quantity of adsorbent is added and the flasks are agitated at constant speed on a shaker at room temperature, Flasks are withdrawn at suitable time intervals, the content filtered and Pb(II) estimation in the liquid sample is made. Similarly, the procedure is repeated with different quantities of adsorbent and other parameters to make the study complete. Percentage removal Pb(II) is calculated using the formula

$\frac{c_{o} - c_{e}}{c_{o}} * 100 (1)$

The equilibrium metal uptake capacity is estimated by using

$q_{e} = \frac{c_{o} - c_{e}}{c_{o}} * V (2)$

Response surface methodology (C.C.D) and optimization of Pb(II) removal

RSM is a group of mathematical and statistical techniques for modeling and analysis of problems in which the response of a model is influenced by several variables [40]. A 24 full-factorial design, 6 center points and 8 axial points leading to 30 experimental runs is performed to study the effect of the four contributing parameters using statistical software, Design expert 10.0.03 (Ease state, USA). Central Composite Design (CCD) consists of 2ⁿ factorial runs with 2*n axial runs and the minimum number of experiments that need be conducted is provided by equation 3.

No of experiments (N) = 2ⁿ+2n+6 (center points) = 2*4+2*4+6=30 (3)

Each variable is investigated at two levels and as the number of factors/operating variables increases, then the number of experimental runs for complete picture also increases.

Y=f (X₁, X₂, X₃ …Xn) (4)

In the present case, four factors - initial metal ion concentration (X₁), initial solution pH (X₂), adsorbent dosage (X₃) and temperature of solution (X₄) are selected as independent variables in equation 4 and by fixing the contact time and size of the adsorbent, the percentage removal of Pb(II) is obtained. Percentage adsorption (%Y) is considered as the dependent variable and the experimental design made with range and levels (-2, -1, 0, 1, 2) of independent variables. In the optimization process, the response can be related to the independent variables by quadratic (second degree) equation and the model equation is given in equation 5.

$% (y) = β_{0} + β_{1} X_{1} + β_{2} X_{2} + β_{3} X_{3} + β_{4} X_{4} + β_{12} X_{1} X_{2} + β_{13} X_{1} X_{3} +$ (5)

$β_{14} X_{1} X_{4} + β_{23} X_{2} X_{3} + β_{24} X_{2} X_{4} + β_{34} X_{3} X_{4} + β_{11} X_{1}^{2} + β_{22} X_{2}^{2} + β_{33} X_{3}^{2} + β_{44} X_{4}^{2}$ where Y is estimate response of the system, β₀ is constant coefficient, β₁, β₂, β₃ and β₄ are linear coefficients, β₁₂, β₁₃, β₁₄, β₂₃, β₂₄ and β₃₄ are interaction coefficients among the four factors, β₁₁, β₂₂, β₃₃ and β₄₄ are quadratic coefficients, X₁, X₂, X₃ and X₄ are independent variables. A multiple regression analysis is then performed to obtain the values of the coefficients. A total of 30 experiments are needed to estimate the biosorption of Pb(II) on to E. globulus L. The responses and corresponding parameters are modeled and optimized using analysis of variance (ANOVA) and by the correlation coefficient (R2). The R2 value shows a measure of how variability in the observed response values can be simplified by experimental factors and their interactions [41].

Results and Discussion

Optimization using response surface methodology (RSM)

It is pointed out that six factors are critical in adsorption. The smaller the particle, the higher is the surface available for transfer and the higher the transfer. However, post adsorption separation puts a limit on the particle size. In the present study a particle of 63μm is used. The time progress of adsorption is studied first and the process has taken 60 minutes to reach equilibrium, Thus, keeping 63μm particle size and 60 minutes of contact time fixed, the effect of other vital parameters, namely, Pb(II) ion concentration, initial solution pH, sorbent dosage and temperature is studied. A total of 30 experimental runs are conducted and the results of the experimentation are in (Table 1). Levels of different process variables in coded and uncoded form for adsorption of Pb(II) using E. globulus L. leaf powder in (Table 2). Thus, 96.58% removal of Pb(II) could be achieved, when 20mg/L Pb(II) is treated with 25g/L E. globulus L. at a pH of 5.0 and temperature of 303K. The interactive contributions of the variables are further studied with model expression given as equation 3.

$% Y = 96 - {0.028}^{*} X_{1} - {0.78}^{*} X_{2} + {0.22}^{*} X_{3} + {0.12}^{*} X_{1} X_{2} - {0.17}^{*} X_{1} X_{3} + {0.14}^{*} X_{1} X_{4} + {0.66}^{*} X_{2} X_{3}$

$- {0.13}^{*} X_{2} X_{4} + {0.22}^{*} X_{3} X_{4} - {0.40}^{*} X_{1}^{2} - 0.51 * X_{2}^{2} + {0.042}^{*} X_{3}^{2} - 0.34 * X_{4}^{2} (6)$

Table 1: Result from CCD for Pb(II) Adsorption by E. globulus L.




  
    Run
    X₁(Co)
    X₂(pH)
    X₃ (w)
    X₄(T)
    %    Adsorption of Pb(II)
  
  
    
    
    
    
    
    Experimental
    Predicted
  
  
    1
    -2
    0
    0
    0
    94.42
    94.4391
  
  
    2
    1
    -1
    1
    1
    95.34
    95.3214
  
  
    3
    -1
    -1
    -1
    1
    96.01
    95.9773
  
  
    4
    0
    0
    -2
    0
    95.7
    95.733
  
  
    5
    0
    0
    0
    -2
    94.42
    94.4155
  
  
    6
    -1
    1
    -1
    1
    92.42
    92.4071
  
  
    7
    -1
    1
    1
    -1
    94.77
    94.7846
  
  
    8
    0
    0
    0
    0
    96
    96
  
  
    9
    -1
    -1
    1
    1
    95.881
    95.8978
  
  
    10
    -1
    1
    -1
    -1
    93.158
    93.1381
  
  
    11
    0
    0
    2
    0
    96.58
    96.6018
  
  
    12
    1
    1
    -1
    -1
    93.634
    93.601
  
  
    13
    1
    1
    -1
    1
    93.43
    93.4143
  
  
    14
    0
    0
    0
    0
    96
    96
  
  
    15
    1
    1
    1
    -1
    94.555
    94.5493
  
  
    16
    0
    -2
    0
    0
    95.53
    95.5361
  
  
    17
    0
    0
    0
    0
    96
    96
  
  
    18
    0
    2
    0
    0
    92.36
    92.4086
  
  
    19
    -1
    -1
    -1
    -1
    96.19
    96.2035
  
  
    20
    1
    -1
    -1
    1
    96.13
    96.0991
  
  
    21
    1
    -1
    -1
    -1
    95.77
    95.7811
  
  
    22
    0
    0
    0
    0
    96
    96
  
  
    23
    0
    0
    0
    2
    94.84
    94.8993
  
  
    24
    2
    0
    0
    0
    94.29
    94.3256
  
  
    25
    -1
    -1
    1
    -1
    95.25
    95.2273
  
  
    26
    -1
    1
    1
    1
    95
    94.9504
  
  
    27
    1
    -1
    1
    -1
    94.11
    94.1066
  
  
    28
    0
    0
    0
    0
    96
    96
  
  
    29
    1
    1
    1
    1
    95.289
    95.2593
  
  
    30
    0
    0
    0
    0
    96
    96



Table 1: Result from CCD for Pb(II) Adsorption by E. globulus L.

Table 2: Levels of different process variables in coded and uncoded form for Adsorption of Pb (II) using E. globulus L. leaf powder.




  
    Variable
    Name
    Range    and level
  
  
    -2 -1 0 1 2
  
  
    X₁
    Initial    concentration(C_o)mg/L
    10 15 20 25 30
  
  
    X₂
    pH    of the solution
    3 4 5 6 7
  
  
    X₃
    Biosorbent    dosage(w), g/L
    5 10 15 20 25
  
  
    X₄
    Temperature(T),K
    283 293 303 313 323



Table 2: Levels of different process variables in coded and uncoded form for
Adsorption of Pb (II) using E. globulus L. leaf powder.

The equation 6 suggests that a metal ion concentration (Co) = 20.904 mg/L, pH = 5. 258, sorbent dosage (w) = 15.372g/L, and temperature (T) = 302.006K yield the highest removal 97.087% and there is good agreement between experimental values and those, calculated with the model equation. A comparison of optimal values (predicted) and the percentage removal (actual), and the normal plot for residual is in (Figure 1(i)). Interactive contributions of the four variables - Pb(II) ion concentration, initial solution pH, adsorbent dosage and temperature, are presented in surface contour plots in (Figure 1(ii)) for better understanding. The ANOVA and regression coefficient values are presented as (Table 3).

Table 3: ANOVA and estimated regression coefficients for the Pb (II) biosorption onto E. globulus L.




  
    Source
    Sum of
      Squares
    df
    Mean
      Square
    F
      Value
    p-value
      Prob > F
    
  
  
    Model
    38.15
    14
    2.72
    2256.97
    < 0.0001
    significant
  
  
    X₁-Initial Concentration
    0.019 
    1 
    0.019 
    16 
    0.0012 
    
  
  
    X₂-pH
    14.67 
    1 
    14.67 
    12152.03 
    < 0.0001 
    
  
  
    X₃-Biosorbent dosage
    1.13 
    1 
    1.13 
    937.84 
    < 0.0001 
    
  
  
    X₄-Temperature
    0.35 
    1 
    0.35 
    290.83 
    < 0.0001 
    
  
  
    X₁ *X₂
    0.78 
    1 
    0.78 
    649.07 
    < 0.0001 
    
  
  
    X₁ *X₃
    0.49 
    1 
    0.49 
    403.82 
    < 0.0001 
    
  
  
    X₁ *X₄
    0.3 
    1 
    0.3 
    245.34 
    < 0.0001 
    
  
  
    X₂*X₃
    6.88 
    1 
    6.88 
    5697.4 
    < 0.0001 
    
  
  
    X₂*X₄
    0.25 
    1 
    0.25 
    211.02 
    < 0.0001 
    
  
  
    X₃*X₄
    0.8 
    1 
    0.8 
    666.05 
    < 0.0001 
    
  
  
    X₁²
    4.49 
    1 
    4.49 
    3715.37 
    < 0.0001 
    
  
  
    X₂²
    7.05 
    1 
    7.05 
    5837.42 
    < 0.0001 
    
  
  
    X₃²
    0.048 
    1 
    0.048 
    39.78 
    < 0.0001 
    
  
  
    X₄²
    3.09 
    1 
    3.09 
    2559.5 
    < 0.0001 
    
  
  
    R²(Adj) = 0.9991 and R²(Pred) = 0.9973



Table 3: ANOVA and estimated regression coefficients for the Pb (II) biosorption
onto E. globulus L.

Figure 1(i): A comparison of predicted values and the percentage removal (Actual values) and the normal plot for Residual for Pb(II) onto E. globulus. L.

    
    
    Figure 1(i):  A comparison of predicted values and the percentage removal
(Actual values) and the normal plot for Residual for Pb(II) onto E. globulus. L.

The Model F-value of 2256.97 in (Table 3) implies that the model is significant. There is only a 0.01% chance that an F-value this large could occur due to noise. All the terms with F< Probability <0.05 are significant. In this case all model terms X₁, X₂, X₃, X₄, X₁*X₂, X₁*X₃, X₁*X₄, X₂*X₃, X₂*X₄, X₃ *X₄, X₁ 2, X₂ 2, X₃ 2 X₄ 2 are significant. A value greater than 0.05 indicate the model terms are not significant.

Isotherm studies

Four different standard, the two-parameter adsorption isotherm models are examined for suitability to represent equilibrium, concerning Pb(II) removal.

Langmuir isotherm model [42]

$\frac{c_{e}}{q_{e}} = \frac{1}{q_{m^{b}}} + \frac{c_{e}}{q_{m}} (7)$

Feasibility of the Langmuir isotherm depends on dimensionless constant is expressed by separation factor, RL.

$R_{L} = \frac{1}{1 + b c_{e}} (8)$

In the present study the value RL is 0.913, indicates a favorable condition.

Freundlich isotherm model [43]

$q_{e} = K_{f} C_{_{e}}^{\frac{1}{n}} (9)$

Temkin isotherm model [44]

$q_{e} = \frac{R T}{b_{T}} \ln A_{T} + \frac{R T}{b_{T}} \ln c_{e} (10)$

Dubinin–Radushkevich isotherm model

Sorption energy is calculated by Dubinin–Radushkevich [45] isotherm model to predict the nature of adsorption process, i.e., physical or1 chemical. The linear from of the model is described as.

$\ln q_{e} = \ln q_{D} - K ε^{2} (11)$

The Polanyi potential [46] which is equal to,

$ε = R T \ln (1 + \frac{1}{c_{e}}) (12)$

The plot of lnq_e against e² gave a straight line from which the values of K and q_D, where q_D is the Dubinin-Radushkevich isotherm constant related to the degree of sorbate sorption by the sorbent surface. From the K value, the mean sorption energy (E) is evaluated as

$E = \frac{1}{\sqrt{2 K}} (13)$

The E (kJ/mol) value gives the information regarding type of adsorption, physical or chemical. If E < 8kJ/mol, the adsorption process is physical in nature and in the 8–16 kJ/mol range and it is chemical in nature [47]. In the present case, the adsorption is in chemical in nature. The parametric values of these models are given in (Table 4) and represent the system well at equilibrium, with Freundlich model being marginally superior to the other models.

Table 4: Isotherm equations and constants at optimum conditions.




  
    Isotherm    Models
    Equation
    Parameters
    R²
  
  
    Langmuir
    c_e/q_e = 0.147c_e+0.885
    q_max=6.803mg/g
      k_L=b=0.166L/g
    0.9486
  
  
    Freundlich
    lnq_e = 0.5724lnc_e + 0.047
    K_f=1.046mg/g
      n=1.75
    0.9918
  
  
    Temkin
    q_e=    1.340 lnc_e+1.033
    b_T=1879.95    kJ/mol
      A_T=2.162    K L/g
    0.9492
  
  
    D-R    model
    lnq_e=-3E-07e²+1.2743
    K=3e^-7
      q_m=3.576mg/g
      E=13.519    kJ/mol
    0.8446



Table 4: Isotherm equations and constants at optimum conditions.

Thermodynamics of adsorption

Adsorption process is temperature dependent. A plot of 1/T against the ratio of equilibrium concentrations of Pb(II) ion in liquid and solid phases is plotted on a semi logarithmic coordinate system and the plot is as shown in (Figure 2). Changes in enthalpy, Gibb’s free energy and entropy are calculated to explore the nature of the adsorption, using the following equations 12 and 13.

$Δ G^{o} = - R T \ln K_{c}^{o} (14)$

$\ln (\frac{q_{e}}{c_{e}}) = K_{c}^{o} = - \frac{Δ H}{R} (\frac{1}{T}) + \frac{Δ S}{R} (15)$

Kinetic model	Equation	Parameters	R²
Pseudo1^st order	log(q_et–q_t) = - 0.027t-0.665	K₁=0.0306min^-1	0.959
Pseudo 2^nd order	(t/ q_t) = 0.511 + 0.434	K₂=0.602 g/mg.min	1
Elovich model	q_t=0.069 lnt+1.6555	A_E=14.492 B_E=1.8*10⁹	0.9577
Intra-particle diffusion	q_t=0.0261t^1/2+1.8222	K_dif= 0.0261 C=1.7411	0.9009

Functional group	*E. globulus* L.
Functional group	Before adsorption	After adsorption	Shift
Esters impurities		3826-3716
N-H and O-H stretch	3327	3336	9
(3200-3500)	3327	3336	9
strong O-H stretch	3249	3244	-5
C-H stretch	2911	2920	-9
alkyne triple bonds and nitrile triple bonds	2354	2353	-1
C=O stretch in aldehyde group	----	1718
C=C stretch	1628	1633	5
N-O stretch in Nitro group	------	1524
-C-H bending	1441	1441	0
C-N stretch	1317	1320	3
(amide III) modes of the residual N-acetyl groups	1317	1320	3
C-N stretch	1221	1218	-3
C-N stretch	1153	1153	0
skeletal vibration of the C-O stretch	1046	1055	11
C-Cl stretching vibrations	771	-----
C-Br stretch in alkylhalide	582	----

Adsorbent	Concentration	Time	pH	Metal uptake, mg/g	References
Adsorbent	(mg/l)	min	pH	Metal uptake, mg/g	References
Ficus religiosa leaves	10-1000	15	4	37.45	[53]
Bael leaves (Aegle marmelos)	8.7-180.2	30	5.1	104	[54]
Cinnamomum camphora leaves	50-400	60	5	73.15,	[55]
Black cumin	20-May	60	5.1	8.08	[56]
Maize bran	100-150	100	6.5	142.86	[57]
Tephrosia purpurea	25–200	130	5.4	100	[58]
Phaseolus vulgaris L	40–80	20	5	42.68	[59]
*Sargassum ilicifolium*	20-200	120	3.7	195	[60]
*E.globulus* L	20-150	60	5	6.803	Present study