Batch isotherm studies were carried out on a laboratory scale: (i) to investigate the effectiveness to remove lead of two wastes (olive stone (OS) and olive tree pruning (OTP)), untreated and chemically treated; and (ii) to examine the applicability of various adsorption isotherms to fit the experimental data. Results from tests were analyzed using seven equilibrium isotherm correlations (Langmuir, Freundlich, Dubinin–Radushkevich, Temkin, Redlich–Peterson, Sips, and Toth equations). The sum of the squares of the errors was determined for each isotherm and the Langmuir equation provided the best fit. Chemical treatments increased the biosorption properties of these materials. The maximum biosorption capacities were: 6.33, 49.13, 14.83, and 38.93 mg g−1 for untreated OS, HNO3-OS, H2SO4-OS, and NaOH-OS, respectively, and 26.72, 86.40, 72.78, and 123.80 mg g−1 for untreated OTP, HNO3-OTP, H2SO4-OTP, and NaOH-OTP, respectively. Finally, the loss of mass for each waste (13.9, 14.3, and 36.8% for HNO3-OS, H2SO4-OS, and NaOH-OS and 35.1, 27.5, and 46.7% for HNO3-OTP, H2SO4-OTP, and NaOH-OTP, respectively) was taken into account and an effectiveness coefficient was determined for each adsorbent material.

INTRODUCTION

The contamination of the environment from different activities has become an increasingly serious problem in recent years. Heavy metals are one class of toxic pollutants released into the surface water and groundwater as a result of various activities such as industry, mining, and agriculture. Lead is a hazardous heavy metal because it does not take part in biological processes and it tends to accumulate in ecosystems. As well, it may enter the body via inhalation, ingestion, and skin adsorption, having effects on children and adults, even at low concentrations (Gundogdua et al. 2009; Lourie et al. 2010; Tawfik et al. 2013). The high use of this metal in industry has caused a higher level in wastewaters. The United States Environmental Protection Agency (USEPA) requires lead in drinking water not to exceed 0.015 mg L−1. For that, all industrial effluents containing lead ions should be treated to reduce Pb(II) to an acceptable level (Bernardo et al. 2013; Calero et al. 2013b).

Biosorption is an emerging method with several advantages over conventional methods including ease of operation, high efficiency and no formation of any residual chemical sludge. However, the major advantage of biosorption is that it can be used in situ, and with proper design, it may not need any industrial process operations and can be integrated with many systems (Zhou et al. 2004; Martín-Lara et al. 2013).

To follow the path to sustainable development and resource renewability, an important point is searching for new uses of waste products. Lignocellulosic biomass is drawing attention as an economical and renewable source of energy. Spain is the world's leading producer of olive oil; just in Andalusia an average annual amount of 900,000 tons are produced, which generates a high amount of waste (mainly, olive pruning and olive stone). Olive stone (OS) is a waste from the separation process of the olive cake and olive tree pruning (OTP) is a waste from the olive pruning process. Considering the large amount of these residues generated yearly, a low-cost alternative to their disposal is necessary (AAE 2011; Feria et al. 2011). Andalusia produces per year an average of more than 2 million tons of OTP (assuming that 1 hectare of OTP generates 3 tons of olive oil) and more than 10,000 tons of OS (assuming that 11.5% of olive production is OS waste) (AAE 2011). Therefore, the obtained biomass from olive trees can be an abundant and renewable material to exploit, considering there are as yet no industrial applications for it.

Although there are several studies using these wastes as biosorbents (Martín-Lara et al. 2009; Ronda et al. 2013), there are few that study their behavior when they are chemically treated, that is, the comparison between them. Thus, the main novelty of this work is the study of the use of the two abundant wastes as biosorbent of lead and the effect of chemical treatments on them, as well as applying a high number of models to fit experimental data.

For all of the above, the proposed aims of this work are: the performance of batch isotherm studies on laboratory scale to investigate the effectiveness of untreated and chemically treated OS and OTP in the removal of lead and the analysis of the applicability of seven equilibrium isotherm correlations to fit experimental data.

MATERIALS AND METHODS

Biomass

Two biosorbents (OS and OTP) were obtained from olive farms in the province of Jaén (Spain). Solids were milled with an analytical mill (IKA MF-10) and the fraction <1.000 mm was chosen for the study.

Olive stone

The OS is a waste from the separation process of the olive cake that takes place during the months November–March.

Olive tree pruning

OTP is a waste from the olive pruning process. It is performed between the months of February and March.

Modification by chemical treatment of the raw biomass

The chemical modification of biomass was performed using three chemical solutions: nitric acid (HNO3), sulfuric acid (H2SO4), and sodium hydroxide (NaOH). A previous study was performed with several acid and basic solutions (HNO3, H2SO4, H3PO4, HCl, NaOH, and KOH). Chemical solutions with better results (HNO3, H2SO4, and NaOH) were chosen in this work to study more deeply the biosorption process and to compare their results. The solutions for treatment were prepared at different concentrations (0.1, 1, and 2 M). One liter of these solutions was used to treat 10 g of biomass in a flask at constant temperature (50 °C). Biomass and chemical solution were mixed for 24 h to establish complete contact. Then, the biomass was repeatedly washed with distilled water until the pH of rinsing water remained constant. The treated biosorbents were dried in an oven at 40 °C for 24 h and afterwards stored for later use.

Preparation of lead solutions

A stock solution of 2,000 mg L−1 Pb(II) was prepared to dissolve the desired amount of Pb(NO3)2 (lead(II) nitrate) in 500 mL of distilled water. Solutions of different concentrations were prepared by appropriate dilution of the above stock Pb(II) solution.

Characterization of untreated and chemically treated biosorbents

Untreated and chemically treated biosorbents were characterized in previous published works (Calero et al. 2013a; Martín-Lara et al. 2013). A summary of results obtained for both are shown in Supplementary Table S1 (available online at http://www.iwaponline.com/wst/072/153.pdf).

Experimental procedure

To study the biosorption equilibrium of Pb(II) with OS and OTP as sorbent solids (untreated and chemically treated) different experimental tests were performed. The operational conditions were chosen according to previous studies (Calero et al. 2013b) and were as follows: pH = 5, concentration of biosorbent = 10 g L−1, concentration of chemical agent = 2.0 M (for OS) and 1.0 M (for OTP), and temperature = 25 °C (constant during all the experiment). Experiments were carried out at different initial lead concentrations: 40, 50, 80, 150, 200, 250, 400, 800, 1,600, and 2,400 mg L−1. The total contact time was 120 min and samples were taken at initial and final times. The pH was initially adjusted to the desired value and it was kept constant with 0.1 N HCl and 0.1 N NaOH solutions. After 120 minutes the final lead concentrations were measured by an absorption spectrophotometer (Perkin Elmer, model A Analyst 200).

Theoretical background

Biosorption capacity is the one of the most important characteristics of a biosorbent. It is the amount of sorbate taken up by the biosorbent, per unit mass of the biosorbent. Biosorption capacity is of paramount importance to the capital cost because it dictates the amount of biosorbent required, which also fixes the volume of the adsorbed vessels. The biosorption capacity of Pb(II) at equilibrium qe (mg g−1) was calculated according to the following mass balance equation for the metal ion concentration: 
formula
1
where qe (mg g−1) is the amount biosorbed at equilibrium, Ci (mg L−1) and Ce (mg L−1) are the initial and the equilibrium metal concentrations, respectively, m (g) is the mass of biosorbent and V (L) is the volume of solution.

Isotherms were obtained by equilibrating metal ion solutions of different initial Pb(II) concentrations. The equilibrium can be described using adsorption isotherm models (Aksu & Tezer 2005). There are numerous isotherm models, which are usually based on different theoretical assumptions and that have a different number of parameters (degrees of freedom). Supplementary Table S2 (available online at http://www.iwaponline.com/wst/072/153.pdf) shows the main models from the literature with number of parameters and degrees of freedom.

In this work, four models of two parameters (Langmuir, Freundlich, Dubinin–Radushkevich, and Temkin) and three models of three parameters (Redlich–Peterson, Sips, and Toth) were selected. These models are described below.

Langmuir isotherm

The Langmuir isotherm assumes that biosorption takes place at specific homogeneous sites on the surface of the adsorbent, meaning once a lead molecule occupies a binding site, no further biosorption can occur at that site (Langmuir 1918). The Langmuir isotherm model can be presented by Equation (2): 
formula
2
where Ce is the equilibrium lead concentration in the solution (mg L−1), qe is the equilibrium lead uptake on the biosorbent (mg g−1), qm is the maximum biosorption capacity (mg g−1), and b is the Langmuir constant that is related to the affinity of binding sites and is related to the energy of sorption (L mg−1).

Freundlich isotherm

The Freundlich isotherm model describes a multilayer adsorption with the assumption of a heterogeneous surface in which the energy, a term in the Langmuir equation, varies as a function of the surface coverage (Freundlich 1906). The model can be presented as: 
formula
3
where KF (L g−1) and n (dimensionless) are characteristic constants that indicate the extent of the biosorption and the degree of non-linearity between solution concentration and biosorption, respectively.

Dubinin–Radushkevich isotherm

Another isotherm usually used in biosorption study is that proposed by Dubinin & Radushkevich (1947). Those authors indicated that solid characteristics are related to the porous structure of the solid. The isotherm can be represented by Equation (4): 
formula
4
where qm (mol g−1) is the maximum biosorption capacity in the equilibrium (it can be also represented as mg g−1), B is a model constant (mol2 kJ−2), and ɛ is the Polanyi potential and is represented by: 
formula
5
where R is the universal gas constant (kJ mol−1 K−1), T is the temperature (K). The constant B is the activity coefficient related to the mean free adsorption energy. 
formula
6
where E is the average value of sorption energy (kJ mol−1). This value gives information about the biosorption mechanism, either physical or chemical (Hany et al. 2014). If the E-value lies between 8 and 16 kJ mol−1, the biosorption process occurs chemically and if E < 8 kJ mol−1, the biosorption process takes place physically (Lodeiro et al. 2006; Sari & Tuzen 2008; Amjad et al. 2013).

Temkin isotherm

The Temkin isotherm (Temkin & Pyhev 1940) takes into account the adsorbent–adsorbate interactions on the adsorbent surface. The Temkin isotherm assumes that sorption energy of all molecules in the layer decreases linearly with the coverage due to adsorbent–adsorbate interactions. Hamdaoui & Naffrechoux (2007) proposed Equation (7): 
formula
7
where θ is the fraction cover (qe/qm), ΔQ is the energy change sorption (kJ mol−1) and K0 is the Temkin equilibrium constant (L mg−1).

Redlich–Peterson isotherm

The Redlich–Peterson isotherm (Redlich & Peterson 1959) contains three parameters and it is an empirical equation that may be used to represent adsorption equilibrium over a wide concentration range. The mechanism of adsorption is a hybrid model featuring both Langmuir and Freundlich isotherms and it does not follow ideal monolayer adsorption (Prasad & Srivastava 2009). The Redlich–Peterson isotherm is given by Equation (8) (Ho et al. 2002): 
formula
8
where A and B are two constants (L g−1 and (L mg−1)g, respectively) and g is an exponent that describes the system heterogeneity (between 0 and 1). When g = 1, Equation (8) can be simplified to the Langmuir isotherm and when g = 0, Equation (8) can be simplified to the Henry law. From Equation (8) and by non-linear regression, model parameters can be obtained.

Sips isotherm

The Sips isotherm is a combination of the Langmuir and Freundlich models and it is mainly used to describe heterogeneous surfaces. When sorbate concentration is low, it is reduced to the Freundlich isotherm, while when sorbate concentration is high, it predicts a biosorption capacity in monolayer which is characteristic of the Langmuir isotherm (Günay et al. 2007). The Sips isotherm is given by Equation (9): 
formula
9
where all parameters have the same meaning as in the Langmuir and Freundlich isotherms.

Toth isotherm

The Toth isotherm (Toth 1962), derived from potential theory, has been used to describe the adsorption process in heterogeneous systems. This isotherm assumes an asymmetric quasi-Gaussian power distribution and it considers that most of the binding sites have sorption energy less than the mean value. The Toth isotherm is given by Equation (10): 
formula
10
where qm is the Toth maximum biosorption capacity (mg g−1), K1 is the Toth equilibrium constant (L mg−1), and n the Toth exponent (it is the same n as for the Freundlich equation). When n = 1, Equation (10) is reduced to the Langmuir equation.

RESULTS AND DISCUSSION

Characterization of biosorbents

A summary of physicochemical properties of both biosorbents and the main changes produced during chemical treatments is shown in Supplementary Table S1 (online at http://www.iwaponline.com/wst/072/153.pdf). The physicochemical characterization of biosorbent is vital to understand the metal binding mechanism onto biomass. Thus, most treatments improve properties of biosorbent in order to increase its biosorption capacity. Some changes are produced on the biosorbent surface (such as increasing the surface area or increasing the average pore volume and the pore diameter) and other changes are produced on functional groups of biosorbent (adding of new functional groups or changing percentages of lignin and holocellulose compounds). These changes have been comprehensively studied in previous works (Calero et al. 2013a; Martín-Lara et al. 2013), but a summary table helps to understand better the results of this paper.

Equilibrium biosorption

The study of equilibrium biosorption provides fundamental physicochemical data to evaluate the applicability of the biosorption process to real scale. This equilibrium biosorption is usually described by equations (isotherms), whose parameters are related to superficial properties of biosorbents, their biosorption capacity, and the affinity of sorbent and the sorbate at constant temperature and pH. The obtained isotherms for untreated and chemically treated OS and OTP at respective operational conditions are shown in Figures 1 and 2.

Figure 1

Experimental biosorption isotherms for untreated and chemically treated OS and fitted isotherm by Langmuir, Sips, and Toth models.

Figure 1

Experimental biosorption isotherms for untreated and chemically treated OS and fitted isotherm by Langmuir, Sips, and Toth models.

Figure 2

Experimental biosorption isotherms for untreated and chemically treated OTP and fitted isotherm by Langmuir, Sips, and Toth models.

Figure 2

Experimental biosorption isotherms for untreated and chemically treated OTP and fitted isotherm by Langmuir, Sips, and Toth models.

It is observed for both biosorbents that maximum biosorption capacity of lead increased significantly when biosorbents were chemically treated. This is related to modifications of the superficial properties of the biosorbents (facilitating the binding of ions with their surfaces) and to modifications in mechanisms of binding implicated in the biosorption process. Thus, the maximum biosorption capacity of untreated OS was 6.3 mg g−1 while for chemically treated OS with HNO3, H2SO4, and NaOH this value was 49.1 mg g−1, 14.8 mg g−1, and 38.9 mg g−1, respectively. The maximum biosorption capacity of untreated OTP and chemically treated OTP with HNO3, H2SO4, and NaOH was 26.7 mg g−1, 86.4 mg g−1, 72.78 mg g−1, and 123.80 mg g−1, respectively.

However, an important loss of mass is produced during treatment. Thus, for a more objective analysis, this is taken into account in the results of biosorption capacity. Table 1 shows the value of biosorption capacity of treated solid (qe), a relative value of biosorption capacity which included the loss of mass and refers to 1 g of untreated biosorbent (qer) and the relation between this value and the biosorption capacity of untreated biosorbent (qes).

Table 1

Biosorption capacity of each chemically treated biosorbent, its relative value, and the relation between this value and the biosorption capacity of untreated biosorbent

Biosorbent Loss of mass, % qe, mg g−1 qer, mg g−1 qer/qers 
OS: qes = 6.3 mg g−1 HNO3, 2 M 13.9 49.13 42.30 6.62 
H2SO4, 2 M 14.3 14.83 12.70 2.01 
NaOH, 2 M 36.8 38.93 24.60 4.00 
OTP: qes = 26.7 mg g−1 HNO3, 1 M 35.1 86.40 56.07 2.10 
H2SO4, 1 M 27.5 72.78 52.76 1.98 
NaOH, 1 M 46.7 123.80 65.98 2.47 
Biosorbent Loss of mass, % qe, mg g−1 qer, mg g−1 qer/qers 
OS: qes = 6.3 mg g−1 HNO3, 2 M 13.9 49.13 42.30 6.62 
H2SO4, 2 M 14.3 14.83 12.70 2.01 
NaOH, 2 M 36.8 38.93 24.60 4.00 
OTP: qes = 26.7 mg g−1 HNO3, 1 M 35.1 86.40 56.07 2.10 
H2SO4, 1 M 27.5 72.78 52.76 1.98 
NaOH, 1 M 46.7 123.80 65.98 2.47 

It is observed that the relation between relative biosorption capacity (qer) and biosorption capacity of untreated biosorbent (qes) was higher than 2 for all cases. The maximum value for each biosorbent was 6.62 for OS treated with 2 M HNO3 and 2.47 for OTP treated with 1 M NaOH. Results show that all chemical treatments improve the biosorption capacity of untreated biosorbent, including the most unfavorable treatment, which duplicates this value. From the viewpoint of practical application it is interesting that this value is as high as possible.

Fitting data models

For a better understanding of the Pb(II) biosorption mechanism, data were fitted to several equilibrium isotherm models. In this work, a total of seven models were selected: four of two parameters (Langmuir, Freundlich, Dubinin–Radushkevich, and Temkin) and three of three parameters (Redlich–Peterson, Sips, and Toth). Results are given in Tables 2 and 3 for OS and OTP, respectively.

Table 2

Biosorption isotherm constants for lead ion onto untreated and chemically treated OS

Parameters OS HNO3-OS H2SO4-OS NaOH-OS 
Two-parameter models 
 Langmuir model 
  qm, mg g−1 6.412 47.467 13.769 32.934 
  b, L mg−1 0.0390 0.0303 0.0126 1.9400 
  R2 0.943 0.943 0.903 0.902 
  ∑(qexpqcal)2 1.27 115.38 10.86 110.08 
 Freundlich model 
  KF, (mg g−1) (L mg−1)1/n 2.452 11.170 2.531 16.620 
  n 7.30 4.85 4.29 8.62 
  R2 0.772 0.919 0.929 0.828 
  ∑(qexpqcal)2 5.08 165.07 7.97 129.77 
 Dubinin–Radushkevich model 
  qm, mol g−1 (mg g−14.51 × 10−5 (8.53) 3.41 × 10−5 (70.78) 1.04 × 10−4 (21.45) 2.16 × 10−4 (44.80) 
  B, mol kJ−2 0.0017 0.0023 0.0029 0.0011 
  E, kJ mol−1 17.24 14.61 13.04 21.22 
  R2 0.839 0.965 0.951 0.885 
  ∑(qexpqcal)2 8.35 × 10−11 1.69 × 10−9 1.27 × 10−10 2.99 × 10−9 
 Temkin model 
  ΔQ, kJ mol−1 20.76 19.29 17.13 22.84 
  K0, L mg−1 4.722 1.573 0.374 9.761 
  R2 0.840 0.986 0.956 0.899 
  ∑(qexpqcal)2 3.56 29.33 5.00 112.87 
Three-parameter models 
 Redlich–Peterson model 
A, L g−1 2.110 4.970 0.525 90.200 
B, (L mg−1)g 0.784 0.256 0.112 3.650 
g 0.88 0.87 0.84 0.95 
R2 0.816 0.984 0.959 0.949 
 ∑(qexpqcal)2 4.10 33.44 4.63 57.38 
 Sips model 
qm, mg g−1 8.680 55.019 18.756 37.363 
b, L mg−1 0.231 0.104 0.063 0.799 
n 2.78 1.73 1.95 2.25 
R2 0.863 0.985 0.955 0.913 
 ∑(qexpqcal)2 3.04 30.33 5.09 97.72 
 Toth model 
qm, mg g−1 8.245 63.685 17.428 41.692 
K1, L mg−1 1.091 0.400 0.032 2.219 
n 2.93 2.81 1.95 2.00 
R2 0.891 0.988 0.955 0.953 
 ∑(qexpqcal)2 2.42 24.63 5.04 52.45 
Parameters OS HNO3-OS H2SO4-OS NaOH-OS 
Two-parameter models 
 Langmuir model 
  qm, mg g−1 6.412 47.467 13.769 32.934 
  b, L mg−1 0.0390 0.0303 0.0126 1.9400 
  R2 0.943 0.943 0.903 0.902 
  ∑(qexpqcal)2 1.27 115.38 10.86 110.08 
 Freundlich model 
  KF, (mg g−1) (L mg−1)1/n 2.452 11.170 2.531 16.620 
  n 7.30 4.85 4.29 8.62 
  R2 0.772 0.919 0.929 0.828 
  ∑(qexpqcal)2 5.08 165.07 7.97 129.77 
 Dubinin–Radushkevich model 
  qm, mol g−1 (mg g−14.51 × 10−5 (8.53) 3.41 × 10−5 (70.78) 1.04 × 10−4 (21.45) 2.16 × 10−4 (44.80) 
  B, mol kJ−2 0.0017 0.0023 0.0029 0.0011 
  E, kJ mol−1 17.24 14.61 13.04 21.22 
  R2 0.839 0.965 0.951 0.885 
  ∑(qexpqcal)2 8.35 × 10−11 1.69 × 10−9 1.27 × 10−10 2.99 × 10−9 
 Temkin model 
  ΔQ, kJ mol−1 20.76 19.29 17.13 22.84 
  K0, L mg−1 4.722 1.573 0.374 9.761 
  R2 0.840 0.986 0.956 0.899 
  ∑(qexpqcal)2 3.56 29.33 5.00 112.87 
Three-parameter models 
 Redlich–Peterson model 
A, L g−1 2.110 4.970 0.525 90.200 
B, (L mg−1)g 0.784 0.256 0.112 3.650 
g 0.88 0.87 0.84 0.95 
R2 0.816 0.984 0.959 0.949 
 ∑(qexpqcal)2 4.10 33.44 4.63 57.38 
 Sips model 
qm, mg g−1 8.680 55.019 18.756 37.363 
b, L mg−1 0.231 0.104 0.063 0.799 
n 2.78 1.73 1.95 2.25 
R2 0.863 0.985 0.955 0.913 
 ∑(qexpqcal)2 3.04 30.33 5.09 97.72 
 Toth model 
qm, mg g−1 8.245 63.685 17.428 41.692 
K1, L mg−1 1.091 0.400 0.032 2.219 
n 2.93 2.81 1.95 2.00 
R2 0.891 0.988 0.955 0.953 
 ∑(qexpqcal)2 2.42 24.63 5.04 52.45 
Table 3

Biosorption isotherm constants for lead ion onto untreated and chemically treated OTP

Parameters OTP HNO3-OTP H2SO4-OTP NaOH-OTP 
Two-parameter models 
 Langmuir model 
  qm, mg g−1 24.747 72.350 68.540 106.300 
  b, L mg−1 0.0265 0.0699 0.1700 0.0779 
  R2 0.952 0.908 0.960 0.926 
  ∑(qexpqcal)2 38.38 600.34 211.04 1031.69 
 Freundlich model 
  KF, (mg g−1) (L mg−1)1/n 4.639 14.598 21.410 21.800 
  n 4.14 4.21 5.81 4.09 
  R2 0.949 0.968 0.895 0.982 
  ∑(qexpqcal)2 41.89 207.76 551.18 244.20 
 Dubinin–Radushkevich model 
  qm, mol g−1 (mg g−11.96 × 10−4 (40.66) 5.63 × 10−4 (166.60) 4.67 × 10−4 (96.67) 8.59 × 10−4 (178.80) 
  B, mol kJ−2 0.0028 0.0025 0.0018 0.0024 
  E, kJ mol−1 13.34 14.16 16.77 14.33 
  R2 0.979 0.966 0.926 0.994 
  ∑(qexpqcal)2 3.95 × 10−10 5.09 × 10−9 9.09 × 10−9 1.97 × 10−9 
 Temkin model 
  ΔQ, kJ mol−1 17.31 21.09 22.20 24.26 
  K0, L mg−1 0.621 1.543 6.398 5.154 
  R2 0.990 0.963 0.919 0.964 
  ∑(qexpqcal)2 8.33 198.40 126.90 510.94 
Three-parameter models 
 Redlich–Peterson model 
  A, L g−1 1.34 18.57 13.22 44.75 
  B, (L mg−1)g 0.138 0.932 0.238 1.487 
  g 0.86 0.81 0.97 0.80 
  R2 0.990 0.975 0.962 0.998 
  ∑(qexpqcal)2 7.96 162.86 196.74 30.82 
 Sips model 
  qm, mg g−1 31.884 96.41 68.304 133.40 
  b, L mg−1 0.074 0.128 0.161 0.143 
  n 1.780 2.130 0.956 1.950 
  R2 0.988 0.953 0.960 0.986 
  ∑(qexpqcal)2 9.48 304.69 210.57 191.83 
 Toth model 
  qm, mg g−1 23.268 93.181 68.953 129.84 
  K1, L mg−1 0.147 0.252 0.185 0.210 
  n 2.67 2.39 1.09 2.03 
  R2 0.990 0.950 0.963 0.995 
  ∑(qexpqcal)2 8.05 194.93 196.38 76.05 
Parameters OTP HNO3-OTP H2SO4-OTP NaOH-OTP 
Two-parameter models 
 Langmuir model 
  qm, mg g−1 24.747 72.350 68.540 106.300 
  b, L mg−1 0.0265 0.0699 0.1700 0.0779 
  R2 0.952 0.908 0.960 0.926 
  ∑(qexpqcal)2 38.38 600.34 211.04 1031.69 
 Freundlich model 
  KF, (mg g−1) (L mg−1)1/n 4.639 14.598 21.410 21.800 
  n 4.14 4.21 5.81 4.09 
  R2 0.949 0.968 0.895 0.982 
  ∑(qexpqcal)2 41.89 207.76 551.18 244.20 
 Dubinin–Radushkevich model 
  qm, mol g−1 (mg g−11.96 × 10−4 (40.66) 5.63 × 10−4 (166.60) 4.67 × 10−4 (96.67) 8.59 × 10−4 (178.80) 
  B, mol kJ−2 0.0028 0.0025 0.0018 0.0024 
  E, kJ mol−1 13.34 14.16 16.77 14.33 
  R2 0.979 0.966 0.926 0.994 
  ∑(qexpqcal)2 3.95 × 10−10 5.09 × 10−9 9.09 × 10−9 1.97 × 10−9 
 Temkin model 
  ΔQ, kJ mol−1 17.31 21.09 22.20 24.26 
  K0, L mg−1 0.621 1.543 6.398 5.154 
  R2 0.990 0.963 0.919 0.964 
  ∑(qexpqcal)2 8.33 198.40 126.90 510.94 
Three-parameter models 
 Redlich–Peterson model 
  A, L g−1 1.34 18.57 13.22 44.75 
  B, (L mg−1)g 0.138 0.932 0.238 1.487 
  g 0.86 0.81 0.97 0.80 
  R2 0.990 0.975 0.962 0.998 
  ∑(qexpqcal)2 7.96 162.86 196.74 30.82 
 Sips model 
  qm, mg g−1 31.884 96.41 68.304 133.40 
  b, L mg−1 0.074 0.128 0.161 0.143 
  n 1.780 2.130 0.956 1.950 
  R2 0.988 0.953 0.960 0.986 
  ∑(qexpqcal)2 9.48 304.69 210.57 191.83 
 Toth model 
  qm, mg g−1 23.268 93.181 68.953 129.84 
  K1, L mg−1 0.147 0.252 0.185 0.210 
  n 2.67 2.39 1.09 2.03 
  R2 0.990 0.950 0.963 0.995 
  ∑(qexpqcal)2 8.05 194.93 196.38 76.05 

Comparison between models

In general, models fitted better the OTP results than OS ones (higher values of R2). Data were fitted by the Langmuir isotherm with values of R2 > 0.90 for all studied biosorbents. Values of maximum biosorption capacities (qm) were similar to experimental results, confirming the validity of this model. Best results for qm value were obtained for OS treated with 2 M HNO3 and OTP treated with 1 M NaOH. This is in accordance with results obtained from characterization studies (Calero et al. 2013a; Martín-Lara et al. 2013), where it is shown that chemical treatments that more greatly improved the biosorption capacity of each biosorbent were the same. However, the operational parameter b, which relates the affinity of biosorbent to the sorbate, did not undergo important changes with chemical treatment, indicating that chemical treatment did not affect the affinity that biosorbent presents by the ion (according to the definition of parameter b). The Dubinin–Radushkevich isotherm provided high values of R2. As well, values of mean free energy of biosorption (E) indicated that the mechanism of lead biosorption was mainly a chemical process, predominating the ionic interchange. These values were 17.24, 14.61, 13.04, and 21.22 kJ mol−1 for untreated OS, HNO3-OS, H2SO4-OS, and NaOH-OS respectively, and 13.34, 14.16, 16.77, and 14.33 kJ mol−1 for untreated OTP, HNO3-OTP, H2SO4-OTP, and NaOH-OTP, respectively. The Temkin model provided a good fit of results with values of energy change sorption (ΔQ) in the range between 17 and 24 kJ mol−1 for two biosorbents and the same magnitude order as that obtained by the Dubinin–Radushkevich model. Three-parameter models, in general, showed better values of R2 (all higher than 0.90 except for untreated OS). Thus, for the Redlich–Peterson model, data were highly fitted by this isotherm with values of R2 > 0.95 (except for untreated OS). Values of parameter g were close to 1 (higher than 0.80 for all studied biosorbents), indicating that the model had a trend to the Langmuir isotherm. The Sips model also represented adequately the experimental results, with obtained values of maximum biosorption capacity very similar to experimental values and obtained by the Langmuir model. However, the parameter n was lower than 1. Finally, the obtained results by the Toth model were very similar to those obtained by Sips, both in values of biosorption capacity as well as in values of parameter n. For untreated and treated OS, in all cases the Langmuir model was the best to fit the data (highest values of R2 and lowest values of ∑(qexpqcal)2. However, for OTP, the Redlich–Peterson and Toth models were the best fitted for experimental results. Moreover, for OTP the difference between models was lower than for OS, giving all of them a good fit. This is also observed in Table 4, which shows experimental biosorption capacity and that obtained by better models. Figures 1 and 2 present the experimental results and fitted ones by the same models.

Table 4

Experimental value of biosorption capacity and that obtained for models which reproduced better results

 qm, mg g−1 
Biosorbent Experimental Langmuir Sips Toth 
OS Untreated 6.33 6.41 8.68 8.26 
HNO3 49.13 47.47 55.02 63.69 
H2SO4 14.83 13.77 18.76 17.43 
NaOH 38.93 32.93 37.36 41.69 
OTP pruning Untreated 26.72 24.75 31.68 36.27 
HNO3 86.40 72.35 96.41 93.18 
H2SO4 72.78 68.54 68.30 68.95 
NaOH 123.80 106.30 133.40 129.84 
 qm, mg g−1 
Biosorbent Experimental Langmuir Sips Toth 
OS Untreated 6.33 6.41 8.68 8.26 
HNO3 49.13 47.47 55.02 63.69 
H2SO4 14.83 13.77 18.76 17.43 
NaOH 38.93 32.93 37.36 41.69 
OTP pruning Untreated 26.72 24.75 31.68 36.27 
HNO3 86.40 72.35 96.41 93.18 
H2SO4 72.78 68.54 68.30 68.95 
NaOH 123.80 106.30 133.40 129.84 

Comparison between biosorbents

Comparing values of biosorption capacity for each biosorbent (Tables 2 and 3), it is observed that values for treated OTP were higher than for treated OS. Moreover, comparing data obtained from the best models (Table 4), it is observed that results for OTP were better than for OS, with values of biosorption capacity around four times higher. Thus, the best results of qm for OS and OTP were 49.13 mg g−1 and 123.80 mg g−1, respectively, and they were obtained with HNO3 treatment for OS and NaOH treatment for OTP.

However, taking into account values for untreated biosorbent (four times higher for OTP than for OS) and loss of mass (also higher for OTP during chemical treatment), it is concluded that chemical treatment for OS was more effective than for OTP, as values of qer/qes were higher for the first biosorbent. Regarding different chemical treatments for each biosorbent, it is determined that for OS the best results were obtained with treatment with 2 M HNO3, giving a maximum value of biosorption capacity (49.13 mg g−1) and a minimum value of loss of mass (13.9%). For OTP the best result was obtained with treatment with 1 M NaOH, reaching a maximum value of biosorption capacity (123.80 mg g−1); however the loss of mass was also high (46.7%). Maximum values of the relation (qer/qes) were 6.62 and 2.47 for HNO3-OS and NaOH-OTP, respectively. It would be interesting to perform a more exhaustive economical analysis to determine from those values whether chemical treatment can be considered viable to use in real scale.

CONCLUSIONS

Two wastes of olive-oil production (OS and OTP) were chemically modified with three chemical solutions (HNO3, H2SO4, and NaOH) and changes in physicochemical characteristics and biosorption properties were studied. Results showed that chemical treatments improved greatly the biosorption capacity of lead. Experimental isotherm tests were performed and the experimental maximum biosorption capacities obtained were 6.33, 49.13, 14.83, and 38.93 mg g−1 for untreated, HNO3-OS, H2SO4-OS, and NaOH-OS, respectively, and 26.72, 86.40, 72.78, and 123.80 mg g−1 for untreated, HNO3-OTP, H2SO4-OTP, and NaOH-OTP, respectively. Moreover, an effectiveness coefficient (qer/qes) was determined for each adsorbent material in order to take into consideration the loss of mass during chemical treatment. Results showed that the treatments that improved lead biosorption capacity more were treatments with HNO3 for OS (qer/qes = 6.62) and with NaOH for OTP (qer/qes = 2.47). Data obtained were subject to equilibrium modeling using different isotherm models: Langmuir, Freundlich, Dubinin–Radushkevich, Temkin, Redlich–Peterson, Sips, and Toth. The better models were Langmuir, Toth, and Sips for OS, and Langmuir, Redlich–Peterson, and Toth for OTP.

ACKNOWLEDGEMENTS

This work was funded in part by the Spanish Government (Project CTM2009-10294). The work of the first author was partially funded by the University of Granada, Spain (Research grant A.6-2012).

REFERENCES

REFERENCES
Agencia Andaluza de la Energía
2011
Aksu
Z.
Tezer
S.
2005
Biosorption of reactive dyes on the green alga Chlorella vulgaris
.
Process Biochem.
40
(
3–4
),
1347
1361
.
Amjad
H.
Alan
P.
Adi
J.
Abdullah
M.
Malek
M.
2013
Improving the adsorption efficiency of phenolic compounds into olive wood biosorbents by prewashing with organic solvents: Equilibrium, kinetic and thermodynamic aspects
.
J. Environ. Manage.
118
(
30
),
1
10
.
Bernardo
M.
Mendes
S.
Nuno
L.
Goncalves
M.
Mendes
B.
Pinto
F.
Lopes
H.
Fonseca
I.
2013
Removal of lead (Pb(II)) from aqueous medium by using chars from co-pyrolysis
.
J. Colloid Interf. Sci.
409
,
158
165
.
Calero
M.
Pérez
A.
Blázquez
G.
Ronda
A.
Martín-Lara
M. A.
2013a
Characterization of chemically modified biosorbents from olive tree pruning for the biosorption of lead
.
Ecol. Eng.
58
,
344
354
.
Calero
M.
Ronda
A.
Martín-Lara
M. A.
Pérez
A.
Blázquez
G.
2013b
Chemical activation of olive tree pruning to remove lead(II) in batch system: Factorial design for process optimization
.
Biomass Bioenergy
58
,
322
332
.
Dubinin
M. M.
Radushkevich
L. V.
1947
Equation of the characteristic curve of activated charcoal
.
Proc. Acad. Sci. Phys. Chem. Sect.
55
,
331
333
.
Feria
M. J.
López
F.
García
J. C.
Pérez
A.
Zamudio
M. A. M.
Alfaro
A.
2011
Valorization of Leucaena leucocephala for energy and chemicals from autohydrolysis
.
Biomass Bioenergy
35
,
2224
2233
.
Freundlich
H. M. F.
1906
Uber die adsorption in losungen (About the adsorption in solutions)
.
Zeitschrift fur Phys Chem
57
(
A
),
385
470
.
Gundogdua
A.
Ozdesa
D.
Durana
C.
Buluta
V. N.
Soylakb
M.
Senturka
H. B.
2009
Biosorption of Pb(II) ions from aqueous solution by pine bark (Pinus brutia Ten.)
.
J. Trace Microprobe Technol.
153
,
62
69
.
Hany
H.
Azza
M.
Nabila
S.
Mohamed
A.
2014
Lead biosorption from aqueous solution by raw and chemically modified green fresh water algae Scenedesmus obliquus
.
Desalination Water Treat.
52
(
40–42
),
7906
7914
.
Ho
Y. S.
Porter
J. F.
Mckay
G.
2002
Divalent metal ions onto peat: copper, nickel and lead single component systems
.
Water Air Soil Pollut.
141
(
1–4
),
1
33
.
Martín-Lara
M. A.
Hernaínz
F.
Calero
M.
Blázquez
G.
Tenorio
G.
2009
Surface chemistry evaluation of some solid wastes from olive-oil industry used for lead removal from aqueous solutions
.
Biochem. Eng. J.
44
(
2–3
),
151
159
.
Martín-Lara
M. A.
Blázquez
G.
Ronda
A.
Pérez
A.
Calero
M.
2013
Development and characterization of biosorbents to remove heavy metals from aqueous solutions by chemical treatment of olive stone
.
Indust. Eng. Chem. Res.
52
(
31
),
10809
10819
.
Redlich
O.
Peterson
D. L.
1959
A useful adsorption isotherm
.
J. Phys. Chem.
63
(
6
),
1024
1026
.
Ronda
A.
Martín-Lara
M. A.
Blázquez
G.
Moreno-Bachs
N.
Calero
M.
2013
Copper biosorption in the presence of lead onto olive stone and pine bark in batch and continuous systems
.
Environ. Prog. Sustain. Energy J.
13
,
192
204
.
Temkin
M. I.
Pyhev
V.
1940
Kinetics of ammonio synthesis on promoted iron catalysts
.
Acta Physicochem.
12
(
3
),
327
356
.
Toth
J.
1962
Acta Chimica Academiae Scientiarum Hungaricae
.
Akademiai Kiado
,
Budapest
,
Hungary
.
Zhou
D.
Zhang
L.
Zhou
J.
Guo
S.
2004
Development of a fixed-bed column with cellulose/chitin beads to remove heavy metals ions
.
J. Appl. Polym. Sci.
94
,
10809
10819
.