Study of kinetics in the biosorption of lead onto native and chemically treated olive stone

G Model

JIEC-1659; No. of Pages 7 Journal of Industrial and Engineering Chemistry xxx (2013) xxx–xxx

Contents lists available at ScienceDirect

Journal of Industrial and Engineering Chemistry journal homepage: www.elsevier.com/locate/jiec

Study of kinetics in the biosorption of lead onto native and chemically treated olive stone G. Bla´zquez *, M. Calero 1, A. Ronda 1, G. Tenorio 1, M.A. Martı´n-Lara 2 Department of Chemical Engineering, University of Granada, 18071 Granada, Spain

A R T I C L E I N F O

Article history: Received 19 March 2013 Accepted 2 November 2013 Available online xxx Keywords: Biosorption Double exponential model Kinetics Lead Olive stone

A B S T R A C T

In this research, olive stone was used as precursor for the development of new biosorbents for lead ions. Chemical treatments were analyzed in terms of their effects on physical–chemical properties and kinetics of lead removal. A kinetic study of the biosorption of lead ions by olive stone was analyzed according to six different kinetic models (pseudo first, pseudo second, pseudo n-order, Elovich, solid diffusion and double exponential models). The biosorption kinetic data were successfully described with pseudo-nth order and double exponential models for all biosorbents. The double exponential model allowed estimating the values of external and internal mass transfer coefficients. The values of external mass transfer coefficient (ke) ranged from 42.62  106 to 508.3  106 m min1 and the internal mass transfer coefficient (ki) from 3.76  106 to 73.4  106 m min1. On the other hand, the analysis of experimental data showed that chemical treatments of the biomass led to increase biosorption capacity of the native biomass. ß 2013 The Korean Society of Industrial and Engineering Chemistry. Published by Elsevier B.V. All rights reserved.

1. Introduction Reclamation of metal-contaminated waters requires tertiary refinement treatments by activated carbons and synthetic resins in order to fulfill ongoing strict regulations. Both these materials can ensure the complete removal of pollutants, but with huge costs. In particular, activated carbon preparation is a highly energy consuming process, in which carbon-containing materials (wood, nut shells, fruit stones, peat, charcoal, brown coal, lignite, bituminous coal, mineral oil products, and other waste materials) are generally pyrolyzed (400–800 8C), and then activated (by gasification in oxidizing atmosphere at 800–1000 8C) to increase specific surface area and porosity [1]. In this scenario, researchers have suggested that biosorption is an emerging, competitive, effective and inexpensive technology which reduces the concentration of heavy metal ions to acceptable levels, especially for the treatment of low-concentration effluents [2,3]. The biosorption process involves a solid phase (sorbent or biosorbent; biological material) and a liquid phase (solvent, normally water) containing a dissolved species to be sorbed (sorbate, heavy metals). For this purpose, many kinds of vegetable wastes have been investigated. * Corresponding author. Tel.: +34 958 240770; fax: +34 958 248992. E-mail addresses: [email protected] (G. Bla´zquez), [email protected] (M. Calero), [email protected] (A. Ronda), [email protected] (G. Tenorio), [email protected] (M.A. Martı´n-Lara). 1 Tel.: +34 958 243311; fax: +34 958 248992. 2 Tel.: +34 958 240445; fax: +34 958 248992.

Table 1 shows the biosorption efficiency of these new biosorbents and activated carbon. In particular, olive stone may constitute promising low-cost biosorbent among biomaterials, since this material is produced in great quantities in the Mediterranean area, and is of no market value. The stone has very good properties as a fuel for heating, even for domestic installations. In addition to the use as fuel, stones are also used as abrasive material for cleaning walls, for example, in the manufacturing of furfural, and for the manufacturing of active carbon for the treatment of gases, water or other special applications. But also, olive stone was investigated both as biosorbent [14,24–30] and as raw material for activated carbon production [31–35]. Although these research activities have shown that olive stone is an effective biosorbent for the removal of Cd, Cr, Cu and Pb ions in aqueous solutions, a pretreatment with acids or alkali could improve its natural sorption capacity. Strong acids like H2SO4 or HNO3 can protonate unavailable functional groups contained in the structure of biosorbents. Furthermore, these acids can also transform functional groups mostly to carboxylic groups by oxidation. On the other hand, formation of carboxylate moieties from esters is carried out by NaOH treatment [36,37]. The scope of this work is to study the potential of untreated and chemically modified olive stone as biosorbents of Pb(II). New, low cost and ecofriendly biosorbents from olive stone have been developed and the kinetics of lead ions biosorption onto them has been elucidated.

1226-086X/$ – see front matter ß 2013 The Korean Society of Industrial and Engineering Chemistry. Published by Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.jiec.2013.11.003

Please cite this article in press as: G. Bla´zquez, et al., J. Ind. Eng. Chem. (2013), http://dx.doi.org/10.1016/j.jiec.2013.11.003

G Model

JIEC-1659; No. of Pages 7 G. Bla´zquez et al. / Journal of Industrial and Engineering Chemistry xxx (2013) xxx–xxx

2

Table 1 Biosorption efficiency of new biosorbents and activated carbon. Biosorption capacity (mg g1)

NaOH (Merck p.a.) solutions in a rotary shaker for 24 h, at 50 8C and at a biomass concentration of 10 g L1. Afterwards, the material was rinsed thoroughly with deionised water until neutral pH was attained. Following filtration, treated biomass was dried in an oven at 40 8C during 24 h.

Reference

Biosorbent

Metal

Agave bagasse

Pb

35.10

Appel residues Banana peels Coffe wastes

Pb Cu Pb

8.30 8.24 19.50

Grape stalks

Cu Ni

10.12 10.67

Villaescusa et al. [8]

Marine algae

Ni Cd Pb

56.58 114.38 228.6

Yalc¸in and Sezer [9]

Neem sawdust

Cr(VI)

25.51

Vinodhini and Das [10]

Olive stone

Cr(VI) Cr(III) Cu Pb

3.20 5.19 1.97 17.7

Martı´n-Lara et Bla´zquez et al. Blazquez et al. Martı´n-Lara et

Orange peel

Cd Zn Cr(III)

41.58 32.04 40.55

Pe´rez Marı´n et al. [15]

Copolymerization orange peel

Pb Cd Ni

Preteated orange peel with formaldehyde

Cr(III) Fe

7.59 17.36

Lugo-Lugo et al. [17]

Paper mill waste

Ni Cu Pb Cd

13.7 13.9 14.1 14.8

Suryan and Ahluwalia [18]

Rice bran

Cd Cu Pb Zn

476.1 293.3 162.6

Velazquez-Jimenez et al. [4] Chong et al. [5] Liu et al. [6] Boonamnuayvitaya et al. [7]

al. [11] [12] [13] al. [14]

Feng et al. [16]

1.10 0.52 4.55 0.13

Montanher et al. [19]

8.01

Almasi et al. [20]

195.38 139.03

Wilson et al. [21]

Walnut shell

Cr(VI)

Peanut shell carbon Commercial carbon (NORIT GRAN) Commercial carbon (MINOTAUR)

Pb

Activated carbon (DARACO 20–40)

Pb Cu Cr(III)

13.33 5.85 2.79

Sulaymon et al. [22]

Activated carbon from olive cake

Cd

21

Aljundi and Jarrah [23]

2.3. Preparation of lead solutions A stock solution of Pb(II) was prepared by dissolving an accurately weighed amount of lead nitrate (Pb(NO3)2) in distilled water. This reagent was analytical grade and was purchased from Panreac (Barcelona, Spain). All required initial Pb(II) concentrations were obtained by successive dilutions of stock solution with distilled water. 2.4. Biosorption experiments The experiments were performed using a batch system to determine the biosorption properties of the different biosorbents. First, in order to characterize the optimum biosorption conditions, experiments were performed varying the following experimental parameters: biosorbent dosage (1–10 g L1), initial Pb(II) concentration (50–250 mg L1) and pH range (2–5) (data here not reported). From these previous studies, the operation conditions were fixed (biosorbent dosage of 10 g L1, initial Pb(II) concentration of 150 mg L1 and pH of 5). Then, kinetic tests of Pb2+ sorption were performed by mixing 0.5 g of native or modified OS in 50 mL of the synthetic lead ion solutions (150 mg L1) in a batch reactor furnished with a thermostated jacket to control the temperature and stirred in a shaker for various contact times (0.5, 1, 2, 3, 4, 5, 10, 15, 20, 30, 40, 50, 60, 70, 80, 90 100, 110 and 120 min). The solution pH was adjusted to 5 using 0.1 M HCl or 0.1 M NaOH. Once the operation time had elapsed, the liquid phase was taken out of the reactor, centrifuged for 10 min, then the supernatant solution was filtered and analyzed for determined lead ion concentrations. The concentrations of lead ions were determined by a flame atomic absorption (AA) spectrophotometer (Perkin Elmer Model 3100). The biosorption capacity of the biosorbents (qt, mg g1) was calculated as follows:

248.69

qt ¼

ðC i  C f Þ V m

(1)

where V is the total solution volume (L), Ci and Cf are the initial and final lead concentration, respectively (mg L1), and m is the amount of OS on a dry basis (g). 2.5. Mathematical models

2. Materials and methods 2.1. Biosorbent material Olive stone (OS) was provided by an oil extraction plant ˜ ora del Castillo’’ located in Vilches, ‘‘Cooperativa Nuestra Sen province of Jaen (Spain). The stones were obtained from the separation process of the olive cake with an industrial pitting machine. Once in the laboratory, air-dried at room temperature to equilibrium moisture content, milled using a laboratory hammer mill (IKA MF-10) to a particle size smaller than 1 mm, homogenized and stored until used.

In order to determine the biosorption kinetics of lead by native and chemically treated OS, six kinetic models were applied to fit experimental data: the pseudo-first-order rate equation (PFORE, also called Lagergren’s equation), the pseudosecond-order rate equation (PSORE), the pseudo-nth-order rate equation (PNORE), the Elovich equation (EE), the resistance to intraparticle diffusion equation (RIDE) and the double exponential equation (DEE). Table 2 shows the kinetic models used in this work. 3. Results and discussion 3.1. Characterization of OS

2.2. Pretreatment of OS OS was chemically treated in order to transform this biomass into new forms by soaking and shaking it in a 2 M H2SO4, HNO3 or

Characterization of a biosorbent is an important analysis for understanding the behavior or the mechanism of lead removal on the surface of biosorbent. Table 3 shows the results obtained in the

Please cite this article in press as: G. Bla´zquez, et al., J. Ind. Eng. Chem. (2013), http://dx.doi.org/10.1016/j.jiec.2013.11.003

G Model

JIEC-1659; No. of Pages 7 G. Bla´zquez et al. / Journal of Industrial and Engineering Chemistry xxx (2013) xxx–xxx

3

Table 2 Kinetics models used to fit experimental data. Model

Equation

Reference k1 t

Pseudo-first-order equation (PFORE)

qt ¼ qe ð1  e

Pseudo-second-order equation (PSORE)

t qt ¼ ð1=hÞþðt=q

Pseudo-nth-order equation (PNORE)

qt ¼ qe  ½ðn  1Þkn t þ qe

Elovich equation (EE) Intraparticle diffusion equation (RIDE)

qt ¼ b1 Lnðvo  bÞ þ b1 Lnt qt ¼ kid  t 0:5 þ c

Double exponential equation (DEE)

qt ¼ qe  m1s expðK D1 tÞ  m2s expðK D2 tÞ

Lagergren [38]

Þ

eÞ

D

Ho an McKay [39,40] ð1nÞ 1=1n



¨ zer [41] O Zeldowitsch [42] and Chien and Clayton [43] Weber and Morris [44]

D

Chiron et al. [45] and Wilczak and Keinath [46]

Table 3 Physic-chemical characteristics of untreated and chemically treated OS. Parameters

Untreated OS

H2SO4-OS

HNO3-OS

NaOH-OS

Specific surface area (m2 g1) Total pore volume (cm3 g1) Average pore size (A˚) Elemental analysis C (%) H (%) N (%) S (%) O (%) Proximate analysis Moisture (%) Volatile material (%) Fixed carbon (%) Ash (%) Total titratable sites (mol g1) Acid titratable sites (mol g1) Basic titratable sites (mol g1) Point of zero charge (pHPZC) Loss of mass due to treatment (%)

0.16 1.84  103 453.02

0.51 1.88  103 146.64

2.45 3.84  103 62.75

0.25 4.63  104 72.76

52.34 7.11 0.03 <0.1 40.47

49.06 8.80 0.16 0.0 41.98

49.23 9.11 0.19 0.0 41.48

40.70 8.19 0.10 0.0 51.01

5.43 74.66 19.54 0.37 6.94  105 3.70  105 3.24  105 5.17 –

– 83.95 16.03 0.02 7.71  105 7.35  105 3.62  106 2.95 14.30

– 79.00 20.96 0.04 7.67  105 6.62  105 1.05  105 2.97 13.90

– 78.68 17.80 3.52 2.72  104 0 2.72  104 6.97 36.80

physical–chemical characterization of chemically treated OS compared to untreated OS. Biosorption of lead on olive stone depends of the structural and morphological features of the support. Respect to specific area of olive stone, the treatment with HNO3 multiplied by fifteen the specific area of untreated OS and duplicated its total pore volume. All treatments increased the specific surface area, in contrast, the total pore volume decreased with the NaOH treatment, and did not modify with H2SO4 treatment. Increase in surface area with base treatment has been attributed to extraction of plant components such as sugars, cementing materials and lignin which usually block pores on biosorbent materials and leading to further exposure of the biosorbent surface [47]. About average pore size, it decreased with treatments. Also, untreated and treated olive stone used in this study were analyzed by scanning electron microscopy (SEM) in order to examine its morphology. The SEM images (figures not shown) showed the granular structure of OS. After treatments, biomass had shown changes in surface morphology, mainly OS treated with nitric acid, presenting a more porous structure. They may contribute to high surface area of this biomass. Elemental analysis shown that native OS content was 52.34% C, 7.11% H, 0.03% N, 40.47% O and <0.1% S. Respect treated OS, the percentage of carbon decreased, while other ones increased with treatments. Mainly, results shown that for H2SO4, HNO3 and NaOH treated OS, the oxygenated groups density was improved with all treatments (41.98, 41.48 and 51.01% respectively). These results suggested that oxygenated groups could be introduced into native OS during treatments. The proportion of volatile matter was similar for all treated biosorbents (approximately 80%) and slightly lower for untreated OS (75%). The fixed carbon content for all biosorbents was approximately 15% or 20% on a dry basis. However, there is a

significant difference between type of OS in terms of ash content: that of NaOH-OS is higher (practically 4%) of the dry fuel. It can due to that basic treatment introduce metal in the sample (sodium) and it is forming part of ash. Finally, results of potentiometric titrations indicated that acid treated OS presents a higher number of acidic functional groups. On the other hand, when OS was treated by NaOH a great increase of total functional groups is produced but this increase is due to an increase on basic functional groups. This can be explained taking into account that alkali chemicals cause rupture of cell membrane, removal of surface impurities, unmasking of some of the cellular groups, swelling of the biomass probably due to the polymer chain breakage, release of polymers such as polysaccharides that have a high amount of acid groups. Respect to values of pH of point of zero charge, treatment by acids, showed lower values of pH of zero point of charge due to the increase of number to total acidic sites and the decrease of the total number of basic sites. Similar results were found in literature, for example, Li et al. [48] investigated orange peels as an adsorbent for cadmium adsorption and the effect of different citric acid concentrations on the adsorbent characters was studied. Upon treatment with more concentrated citric acid solutions, orange peels showed lower values of pH of zero point charge (pHzpc) due to the increase number of total acidic sites while the total number of basic sites decreased. 3.2. Effect of contact time on lead biosorption The effect of contact time on the biosorption process was studied in the time range from 0.5 to 120 min at 150 mg L1 initial Pb(II) concentration at pH 5 and temperature of 25 8C with a fixed biosorbent dose of 10 g L1. As shown in Fig. 1, for all biosorption experiments, the amount of lead ions biosorbed onto OS increased

Please cite this article in press as: G. Bla´zquez, et al., J. Ind. Eng. Chem. (2013), http://dx.doi.org/10.1016/j.jiec.2013.11.003

G Model

JIEC-1659; No. of Pages 7 G. Bla´zquez et al. / Journal of Industrial and Engineering Chemistry xxx (2013) xxx–xxx

4

16

100

14 80

12 Untreated H2SO4

8

% Removal

qt, mg/g

10

HNO3 NaOH

6

60

40

4

Untreated H2SO4

20

2

HNO3 NaOH

0 0

20

40

60

80

100

120

140

t, min Fig. 1. Effect of contact time on lead biosorption capacity of untreated and chemically treated OS.

initially, and then become almost stable, denoting the attainment of equilibrium. The fast biosorption kinetics observed is typical for biosorption of metals involving no energy-mediated reactions, where metal removal from solution is due to purely physicochemical interactions between biomass and metal solution as physical adsorption or ion exchange at biosorbent surface [49] and the subsequent slower phase may involve other mechanisms such as complexation or micro-precipitation or just saturation of binding sites. Other studies published in literature also report similar results (two-stage sorption kinetics) for heavy metals biosorption such as the biosorption of copper and nickel sorption using rubber wood shavings [50] or treated alga (Undaria pinnatifida) [51] or cone biomass of Thuja orientalis [52] or the lead biosorption onto cattails (Typha angustifolia) leaves [53] or biosorption of lead by maize (Zea mays) stalk sponge [54]. Moreover, all treatments significantly increase the biosorption capacity of the biosorbent from an approximate value of 5 mg g1 for untreated OS to a value higher than 12 mg g1 for chemically treated OS. In particular, when OS is treated by NaOH biosorption capacity is tripled. Many researchers also reported significant enhancement in biosorption capacity after alkali treatment similar to results obtained in the present study. For example, Kumar and Bandyopadhyay [55] reported that rice husk treated with sodium hydroxide enhanced the biosorption capacity of cadmium. Tarley et al. [56] found that adsorption of Cd increase by almost double when rice husk was treated with NaOH. The reported adsorption capacities of Cd were 7 and 4 mg g1 for NaOH treated and unmodified rice husk, respectively. Also, treatment of spent grain with NaOH greatly enhanced the biosorption of Cd(II) and Pb(II) ions [57]. In general, three possible reasons for the increase in biosorption capacities of heavy metal ions when biomass is treated by NaOH were given by the authors: (i) Changes on wood surface-increase in surface area, average pore volume and pore diameter after alkaline treatment. (ii) Improvement in ion-exchange process especially with Na+ ions. (iii) Microprecipitation of metal hydroxides in the pores of biomass. Finally, as seen in Table 3, all the treatments caused a loss in the biomass ranging from 13.90 to 36.80%. The loss in the mass of biomass during chemical treatment may lead to some confusion during the quantitative assessment of the biosorption performance [52]. As economic perspective plays an important role in the

0 2

3

4

5

6

7

pH Fig. 2. Effect of pH on lead biosorption capacity of untreated and chemically treated OS.

selection of the biosorbent for industrial applications, it is important to assess the loss in biomass due to treatment along with the quantitative assessment of the biosorption performance. Enhancement of biosorption after treatment may be an offset to some extent, considering the loss of mass due to a treatment [34]. 3.3. Effect of pH on lead biosorption The pH value affects two aspects: speciation of metals in solution (metal ion solubility) and also the ionic state of the functional groups involved in the metal binding and presents in the biosorbent [25]. Fig. 2 shows the effect of pH as a function of the percent removal of the lead ions. Values of pH > 6 have not been studied since it precipitated as Pb(OH)2 at this pH, being the process of retention really a combination of biosorption and microprecipitation. From Fig. 2 the removal efficiency of Pb(II) increases initially from pH 3 to 5, and after pH 5 a plateau is obtained. As the pH of the system increases, the number of positively charged sites decreases and the number of negatively charged sites increases on the surface of biosorbents. Also, the acidity of the medium affects the competition of the hydrogen ions and the Pb(II) for the active sites on the biosorbent surface. At pH 3 the removal efficiency for untreated-OTP is low due to the presence of higher concentrations of H+ in the solution which compete with the Pb(II) ions for the active sites on the biosorbent surface and that the surface charge of the solids is positive. As the pH increases, the electrostatic repulsion decreases due to reduction of the positive charge density of protons on the sorption sites, thus resulting in an enhancement of Pb(II) biosorption. The maximum removal of Pb(II) for untreated OS was found to be 46% at pH 6 as is shown in Fig. 2. Similar results have been reported by Herna´inz et al. [25] that found a decrease in sorption of Cd(II), Cr(III) and Pb(II) using native olive stone when pH was reduced from 5 to 3 or Bla´zquez et al. [13] that reported a sharp increase in the copper removal by native olive stone from 22.0% to 63.4% when the pH value increased from 3 to 5. Respect to effect of pH on Pb(II) biosorption by chemically treated OS at studied concentration of Pb(II) ions, this effect is less marked, especially for OS treated by HNO3 and OS treated by NaOH, due to, at studied concentration (150 mg L1), practically all Pb(II) is removed from solution. Beyond pH 5 precipitate formation could be observed in Pb(II) solution and, therefore, all subsequent experiments were carried out at this pH level.

Please cite this article in press as: G. Bla´zquez, et al., J. Ind. Eng. Chem. (2013), http://dx.doi.org/10.1016/j.jiec.2013.11.003

G Model

JIEC-1659; No. of Pages 7 G. Bla´zquez et al. / Journal of Industrial and Engineering Chemistry xxx (2013) xxx–xxx Table 4 Comparison between biosorption rate constants, estimated qe, and coefficients of determination associated to the PFORE with untreated and chemically treated OS.

5

Table 6 Comparison between biosorption rate constants, estimated qe, and coefficients of determination associated to the PNORE with untreated and chemically treated OS.

Biosorbent

qe (mg g1)

k1 (min1)

r2

SSE

Biosorbent

qe (mg g1)

kn (g mg1 min1)

n

r2

SSE

Untreated OS H2SO4-OS HNO3-OS NaOH-OS

4.113 12.568 14.804 15.256

0.216 0.115 0.757 0.312

0.713 0.969 0.717 0.956

2.224 3.708 3.080 3.795

Untreated OS H2SO4-OS HNO3-OS NaOH-OS

5.242 13.141 15.345 15.256

0.00874 0.0421 0.110 0.312

3.31 1.45 2.02 1.00

0.953 0.978 0.949 0.956

0.366 2.539 0.555 3.795

Table 5 Comparison between biosorption rate constants, estimated qe, and coefficients of determination associated to the PSORE with untreated and chemically treated OS. qe (mg g1) k2 (g mg1 min1) h (mg g1 min1) r2

Untreated OS 4.567 H2SO4-OS 14.109 HNO3-OS 15.331 NaOH-OS 16.247

0.0646 0.0108 0.114 0.0329

1.347 2.144 26.868 8.683

0.900 0.974 0.949 0.938

SSE

Biosorbent

a (mg g1 min1)

b (g mg1)

r2

SSE

0.801 3.126 0.555 5.269

Untreated OS H2SO4-OS HNO3-OS NaOH-OS

8.141 6.062 1965.6 592.94

1.498 0.393 0.695 0.634

0.983 0.908 0.718 0.505

0.134 10.846 3.065 42.290

3.4. Analysis of lead biosorption kinetics Biosorption kinetics depends on the sorbate–biosorbent interaction and on operating conditions. This is a critical parameter for evaluating the suitability of the system for practical applications to the treatment of heavy metals-bearing solutions. In order to determine the biosorption kinetics of lead by native and chemically treated OS, six kinetic models: the pseudo-first-order rate equation (PFORE, also called Lagergren’s equation), the pseudo-second-order rate equation (PSORE), the pseudo-nth-order rate equation (PNORE), the Elovich equation (EE), the resistance to intraparticle diffusion equation (RIDE) and the double exponential equation (DEE) were applied to fit experimental data. The applicability of these kinetic models was determined by measuring the coefficients of determination (R2) and the sum of squares error (SSE), using the Excel’s Solver optimization method. 3.4.1. Pseudo-first-order equation (PFORE) In Table 4 is observed that the PFORE reproduces well the experimental data only in some cases, particularly in OS treated by H2SO4 and NaOH. However, the values obtained of biosorption capacity, qe, are very similar to those found experimentally. On the other hand, values of the rate constant showed that the biosorption process is faster when OS treated by HNO3 is used as biosorbent. 3.4.2. Pseudo-second-order equation (PSORE) The PSORE (Table 5) reproduces the experimental results acceptably for untreated and chemically treated OS. The greater similarity between the calculated and experimental qe values and the high values of R2 showed the applicability of this kinetic model. Respect to the initial sorption rate, h, it was from 1.347 mg g1 min1 for untreated OS to 26.868 mg g1 min1 for HNO3-OS, which indicated that treatment by HNO3 produce a significant increase in the sorption speed (as before it has been commented). This fact was presented also with the rate constant, k2, that especially increased when OS is treated by HNO3. Also the maximum sorption capacity is markedly altered with changes in biosorbent. Ofomaja and Naidoo [47] found also differences between model parameters studying kinetics of copper biosorption by pine cone chemically activated by NaOH, KOH and Ca(OH)2. It is thought that instead of assuming order of the reaction as 1 or 2, the direct calculation of rate constant and order of the adsorption reaction is a more appropriate method [41]. Next, the paper presents the study of the mathematical n-order kinetic model, which contrarily to standard models based on the

laws of chemical kinetics, i.e. pseudo-first-order and pseudosecond-order kinetic model, enables categorization of the investigated process taking into account two parameters: the rate at which biosorption occurs, and order of reaction. 3.4.3. Pseudo-nth-order equation (PNORE) Table 6 shows the results of the fitting of experimental data to the pseudo-nth-order model. It is observed that the pseudo-nthorder equation reproduces well the experimental results for untreated OS and OS treated with all different agents. However, the values found for the reaction order are different, for the chemically treated OS the value of n is approximately 1 or 2, and for the untreated solid a higher value is obtained (n = 3.31). These results are consistent with those obtained by applying pseudo-first order and pseudo-second order models. 3.4.4. Elovich equation (EE) The results of Table 7 indicate that the EE only reproduce acceptably well kinetics of untreated OS but this model does not reproduce experimental results for any of the chemically treated biosorbents, so that the values of the constants of this model cannot be analyzed. Next, to identify the mass transfer steps that occur in the biosorption process, intraparticle diffusion equation was studied.

16 14 12 10

qt, mg/g

Biosorbent

Table 7 Comparison between biosorption rate constants, estimated qe, and coefficients of determination associated to the EE with untreated and chemically treated OS.

Untreated OS H2SO4-OS

8

HNO3-OS NaOH-OS

6 4 2 0 0

2

4

6

8

10

12

0,5

t , min Fig. 3. Weber–Morris plots of Pb(II) biosorption onto untreated and chemically treated OS.

Please cite this article in press as: G. Bla´zquez, et al., J. Ind. Eng. Chem. (2013), http://dx.doi.org/10.1016/j.jiec.2013.11.003

G Model

JIEC-1659; No. of Pages 7 G. Bla´zquez et al. / Journal of Industrial and Engineering Chemistry xxx (2013) xxx–xxx

6

Table 8 Comparison between biosorption rate constants, estimated qe, and coefficients of determination associated to the DEE with untreated and chemically treated OS. Biosorbent

D1 (mg L1)

D2 (mg L1)

KD1 (min1)

KD2 (min1)

r2

SSE

Untreated OS H2SO4-OS HNO3-OS NaOH-OS

20.45 103.51 77.28 121.10

19.90 13.19 43.34 44.71

0.353 0.106 1.944 0.382

0.0211 0.0182 0.188 0.202

0.989 0.984 0.972 0.941

0.0831 1.893 0.344 5.049

Table 9 External and internal mass transfer coefficients from DDE. Biosorbent

qe,exp (mg g1)

qe,cal (mg g1)

ke (106 m min1)

ki (106 m min1)

Untreated OS H2SO4-OS HNO3-OS NaOH-OS

4.95 13.12 15.22 15.21

4.04 11.67 12.06 16.58

508.3 42.62 61.46 206.2

4.81 3.76 35.70 73.4

3.4.5. Intraparticle diffusion equation (RIDE) The plot qt versus t0.5 (Fig. 3) shows multi-linearity for all biosorbents, and each portion represents a distinct mass transfer step. The first portion or the initial zone of higher slope could correspond to the binding of the metal in the external surface (external mass transfer or instantaneous biosorption step). The second portion or the substantially horizontal second zone would be related to the diffusion of metal inside the particle (intraparticle diffusion). Thus, one might consider that the biosorption of lead with these solids biosorbents occurs in more than one stage that can occur simultaneously. Similar results were reported by Dotto and Pinto [58] studying the biosorption of acid blue 9 and FD&C red no.40 onto Streptomyces platensis nanoparticles or Ramana et al. [59] in their study of the removal of Ni(II) from aqueous solutions using biomass prepared from Ceiba pentandra hulls powder modified with citric acid treatment or Mohanty et al. [60] that showed that the relationships for Eichhornia crassipes and chromium system for different initial concentrations at a particular sorbent dose were not linear over the entire time range, indicating that more than one process is affecting the adsorption. Since the kinetics of biosorption process seems to follow a twostep mechanism and in order to estimate mass transfer coefficients, the double exponential model has been now analyzed. 3.4.6. Double exponential equation (DEE) The results of Table 8 indicate that the DEE provided a better correlation for the biosorption of Pb(II) by untreated OS and chemically modified OS compared to the other proposed models. The calculated parameters for the rapid-step and slow-step, KD1 and KD2; respectively, are consistent with those obtained by Michard et al. [61] for the adsorption of uranyl ions by silica gel or Chiron et al. [45] for the adsorption of Cu(II) and Pb(II) onto a grafted silica. Nevertheless, the determination of KD1 and KD2 is not sufficient to describe and interpret the influence of both phenomena (external and internal diffusion). KD1, the rapid-phase sorption coefficient, takes into account both external and internal diffusions, while KD2, the slow-phase sorption coefficient, covers only intraparticle diffusion. Therefore, as the rapid step involves both

Table 10 Calculated initial rates from DEE. Biosorbent

r0 (mg g1 min1)

r01 (mg g1 min1)

r02 (mg g1 min1)

Untreated OS H2SO4-OS HNO3-OS NaOH-OS

0.764 1.119 15.838 5.533

0.722 1.097 15.023 4.629

0.042 0.022 0.815 0.904

phenomena, these parameters can only allow a comparison of the respective biosorption rate of Pb(II). From double exponential equation (Eq. (7)), and mathematically, the addition of D1/ms and D2/ms must be equal to qe to verify the initial condition for t = 0; qt = 0. Also, KD1 and KD2 coefficients are related with external (ke, m min1) and internal mass transfer coefficients (ki, m min1) by the following equations [61], K D1 ¼ ðke þ ki Þ  Se  w 

K D2 ¼ ki  Si  w 

C0 C 0  C eq

C0 C 0  C eq

(8)

(9)

where Se and Si are external and internal superficial areas, respectively, m2 g1, C0 and Ceq the initial and equilibrium lead concentrations, respectively, mg L1 and w the biosorbent dosage, g m3. The calculated and experimental values of qe are compared in Table 9. In this table also are reported the values of external and intraparticle mass transfer coefficients. When native OS was treated by NaOH, internal mass transport particularly increases (from 4.81  106 m min1 to 73.4  106 m min1). However, when native OS is chemically treated external mass transfer coefficients are decreased remarkably (from 508.3  106 m min1 to 42.62  106 m min1). They suggested that mass transport is influenced by the charge and potential conditions at the OS–solution interface. Also, the derivation of Eq. (7) allows the determination of the overall rate r as: r¼

dqt D1 D2 ¼  K D1  expðK D1 tÞ þ  K D2  expðK D2 tÞ dt ms ms

(10)

And, for initial conditions, r0 ¼

D1 D2  K D1 þ  K D2 ms ms

(11)

The overall initial rate can also be decomposed in two rates corresponding to the rapid and the slow steps (Table 10). In general, the initial rates of both steps of treated OS biosorbents are significantly higher than that of untreated OS and initial rates of first step are higher than the second one. These results showed that treatments improved affinity of untreated OS for lead ions. The differences between initial rates between both steps can be interpreted with respect to the biosorption mechanism, the rapid step involves external and internal diffusion and depends largely on the affinity of Pb(II) ions for the biosorbent and on the metal– ligand exchange rate, however, the slow step is controlled by the intraparticle diffusion, which is also dependent of the biosorbent.

Please cite this article in press as: G. Bla´zquez, et al., J. Ind. Eng. Chem. (2013), http://dx.doi.org/10.1016/j.jiec.2013.11.003

G Model

JIEC-1659; No. of Pages 7 G. Bla´zquez et al. / Journal of Industrial and Engineering Chemistry xxx (2013) xxx–xxx

On the other hand, the double-exponential function can also describe a process where the biosorbent offers two different types of biosorption sites [45]. On the first site rapid metal biosorption occurs within a few minutes whereas on the second site type, metal is biosorbed more slowly. Since the surface of OS is heterogeneous it can be considered as a biosorbent with two different types of functional groups. Similarly, a multiple-site model was used to explain the biosorption of Pb(II) onto the twophase olive mill solid by Bla´zquez et al. [62] and Martı´n-Lara et al. [63]. From all the results obtained, it can be observed that the best fits to the experimental data are obtained with the doubleexponential and pseudo-nth-order models for all biosorbents. The double exponential is an analytical solution that can adequately express a two-step mechanism. 4. Conclusions The present study investigates the Pb(II) biosorption potential of untreated and chemically treated OS. Results indicated the effectiveness of the chemical treatments on biosorption capacity. The OS is a material that is regionally available at low cost and can be used as an alternative biosorbent for the removal of Pb(II) from aqueous solutions. The analysis of the physic-chemical characteristics of the OS demonstrated that the chemical treatment changed the surface of the biomass and affected its adsorption capacity. The best models to describe the kinetics of biosorption process, among the tested models, were the pseudo-nth-order and double exponential kinetic models which satisfactorily described the experimental data for the untreated, acid and alkaline treated OS, with high coefficients of determination. On the other hand, the double exponential model allowed estimating the values of external and internal mass transfer coefficients. The values of external mass transfer coefficient (ke) ranged from 42.62  106 to 508.3  106 m min1 and the internal mass transfer coefficient (ki) from 3.76  106 to 73.4  106 m min1. Acknowledgement The authors are grateful to the Spanish Ministry of Science and Innovation for financial support received (Project CTM200910294). References [1] F. Pagnanelli, S. Mainelli, L.N. Toro, Water Res. 42 (2008) 2953. [2] B. Volesky, in: R. Amils, A. Ballester (Eds.), Biohydrometallurgy and Environment Toward the Mining of the 21st century, Part B, Elsevier, Amsterdam, 1999, p. 161. [3] F. Pagnanelli, A. Esposito, F. Veglio`, Water Res. 36 (2002) 4095. [4] L.H. Velazquez-Jimenez, A. Pavlick, J.R. Rangel-Mendez, Ind. Crops Prod. 43 (2013) 200. [5] S.H. Chong, H. Jung, H. Chung, M.Y. Lee, J. Yang, Process Biochem. 33 (1998) 205. [6] C. Liu, H.H. Ngo, W. Guo, K.-L.V. Tung, Bioresour. Technol. 119 (2012) 349. [7] V. Boonamnuayvitaya, C. Chaiya, W. Tanthapanichakoon, S. Jarudilokkul, Sep. Purif. Technol. 35 (2004) 11. [8] I. Villaescusa, N. Fiol, M. Martı´nez, N. Miralles, J. Poch, J. Serarols, Water Res. 38 (2004) 992. [9] S. Yalc¸ın, S. Sezer, R. Apak, Environ. Sci. Pollut. Res. 19 (2012) 3118. [10] V. Vinodhini, N. Das, Desalination 264 (2010) 9.

7

[11] M.A. Martı´n-Lara, F. Herna´inz, G. Bla´zquez, G. Tenorio, M. Calero, J. Environ. Eng. ASCE 136 (2010) 1389. [12] G. Bla´zquez, M. Calero, F. Herna´inz, G. Tenorio, M.A. Martı´n-Lara, Environ. Prog. Sustain. 30 (2011) 576. [13] G. Bla´zquez, M.A. Martı´n-Lara, E. Dionisio-Ruiz, G. Tenorio, M. Calero, J. Ind. Eng. Chem. 17 (2011) 824. [14] M.A. Martı´n-Lara, G. Bla´zquez, A. Ronda, I.L. Rodrı´guez, M. Calero, J. Ind. Eng. Chem. 18 (2012) 1006. [15] A.B. Pe´rez Marı´n, J.F. Ortun˜o, M.I. Aguilar, V.F. Meseguer, J. Sa´ez, M. Llore´ns, Biochem. Eng. J. 53 (2010) 2. [16] N. Feng, X. Guo, S. Liang, Y. Zhu, J. Liu, J. Hazard. Mater. 185 (2011) 49. [17] V. Lugo-Lugo, C. Barrera-Dı´az, F. Uren˜a-Nu´n˜ez, B. Bilyeu, I. Linares-Herna´ndez, J. Environ. Manage. 112 (2012) 120. [18] S. Suryan, S.S. Ahluwalia, Int. J. Environ. Sci. 2 (2012) 1331. [19] S.F. Montanher, E.A. Oliveira, M.C. Rollemberg, J. Hazard. Mater. 117 (2005) 207. [20] A. Almasi, M. Omidi, M. Khodadadian, R. Khamutian, M. Gholivand, Toxicol. Environ. Chem. 94 (2012) 660. [21] K. Wilson, H. Yang, C.W. Seo, W.E. Marshall, Bioresour. Technol. 97 (2006) 2266. [22] A.H. Sulaymon, T.J. Mohammed, J. Al-Najar, Can. J. Chem. Eng. Technol. 3 (2012) 86. [23] I.H. Aljundi, N. Jarrah, J. Anal. Appl. Pyrolysis 81 (2008) 33. [24] N. Fiol, I. Villaescusa, M. Martı´nez, N. Miralles, J. Poch, J. Serarols, Sep. Purif. Technol. 50 (2006) 132. [25] F. Herna´inz, M. Calero, G. Bla´zquez, M.A. Martı´n-Lara, G. Tenorio, Environ. Prog. 27 (2008) 469. [26] A. Aziz, E.H. Elandaloussi, B. Belhalfaoui, M.S. Oualia, L.C. De Me´norval, Colloids Surf. B 73 (2009) 192. [27] F. Herna´inz, M. Calero, G. Bla´zquez, G. Tenorio, M.A. Martı´n-Lara, J. Environ. Eng. ASCE 135 (2009) 1348. [28] G. Bla´zquez, M. Calero, F. Herna´inz, G. Tenorio, M.A. Martı´n-Lara, Chem. Eng. J. 160 (2010) 615. [29] M. Calero de Hoces, G. Bla´zquez Garcı´a, A. Ronda Ga´lvez, M.A. Martı´n-Lara, Ind. Eng. Chem. Res. 49 (2010) 12587. [30] M.A. Martı´n-Lara, I.L. Rodrı´guez, G. Bla´zquez, M. Calero, Desalination 278 (2011) 132. [31] A.H. El-Sheikh, A.P. Newman, H.K. Al-Daffaee, S. Phull, N. Cresswell, J. Anal. Appl. Pyrolysis 71 (2004) 151. [32] M.L. Martı´nez, M.M. Torres, C.A. Guzma´n, D.M. Maestri, Ind. Crops Prod. 23 (2006) 23. [33] R. Ubago-Pe´rez, F. Carrasco-Marı´n, D. Faire´n-Jime´nez, C. Moreno-Castilla, Microporous Mesoporous Mater. 92 (2006) 64. [34] I. Kula, M. Ug˘urlu, H. Karaog˘lu, A. C¸elik, Bioresour. Technol. 99 (2008) 492. [35] C. Ku¨tahyali, M. Eral, J. Nucl. Mater. 396 (2010) 251. [36] K.K. Krishnani, S. Ayyappan, Rev. Environ. Contam. Toxicol. 188 (2006) 59. [37] W.S.S. Ngah, M.A.K.M. Hanafiah, Bioresour. Technol. 99 (2008) 3935. [38] S. Lagergren, Kungliga Svenska Ventenskapsaka demiens Handlingar 24 (1898) 1. [39] Y.S. Ho, G. McKay, Chem. Eng. J. 70 (1998) 115. [40] Y.S. Ho, G. Mckay, Water Res. 34 (2000) 735. ¨ zer, J. Hazard. Mater. 141 (2007) 753. [41] A. O [42] J. Zeldowitsch, Acta Physicochim. URSS 1 (1934) 449. [43] S.H. Chien, W.R. Clayton, Soil Sci. Soc. Am. J. 44 (1980) 265. [44] W. Weber, J. Morris, J. Sanit. Eng. Div. ASCE 89 (1963) 31. [45] N. Chiron, R. Guilet, E. Deydier, Water Res. 37 (2003) 3079. [46] A. Wilczak, T.M. Keinath, Water Environ. Res. 65 (1993) 238. [47] A.E. Ofomaja, E.B. Naidoo, Chem. Eng. J. 175 (2011) 260. [48] X. Li, Y. Tang, Z. Xuan, Y. Liu, F. Luo, Sep. Purif. Technol. 55 (2006) 69. [49] Z. Aksu, Sep. Purif. Technol. 21 (2001) 285. [50] N. Nordin, Z.A. Zakaria, W.A. Ahmad, Water Air Soil Pollut. 223 (2012) 1649. [51] Z. Chen, W. Ma, M. Han, J. Hazard. Mater. 155 (2008) 327. [52] Y. Nuhoglu, E. Oguz, Process Biochem. 38 (2003) 1627. [53] Y.L. Sharain-Liew, C.G. Joseph, S.-E. How, J. Serb. Chem. Soc. 76 (2011) 1037. [54] G. Garcı´a-Rosales, A. Colı´n-Cruz, J. Environ. Manage. 91 (2010) 2079. [55] U. Kumar, M. Bandyopadhyay, Bioresour. Technol. 97 (2006) 104. [56] C.R.T. Tarley, S.L.C. Ferreira, M.A.Z. Arruda, Microchem. J. 77 (2004) 163. [57] K.S. Low, C.K. Lee, S.C. Liew, Process Biochem. 36 (2000) 59. [58] G.L. Dotto, L.A.A. Pinto, Biochem. Eng. J. 68 (2012) 85. [59] D.K.V. Ramana, D.H.K. Reddy, B.N. Kumar, Y. Harinath, K. Seshaiah, Can. J. Chem. Eng. 90 (2012) 111. [60] K. Mohanty, M. Jha, B.C. Meikap, M.N. Biswas, Chem. Eng. J. 117 (2006) 71. [61] P. Michard, E. Guibal, T. Vincent, P. Le Cloirec, Microporous Mater. 5 (1996) 309. [62] G. Bla´zquez, F. Herna´inz, M. Calero, M.A. Martı´n-Lara, G. Tenorio, Chem. Eng. J. 148 (2009) 473. [63] M.A. Martı´n-Lara, F. Herna´inz, M. Calero, G. Bla´zquez, G. Tenorio, Biochem. Eng. J. 44 (2009) 151.

Please cite this article in press as: G. Bla´zquez, et al., J. Ind. Eng. Chem. (2013), http://dx.doi.org/10.1016/j.jiec.2013.11.003