Kinetic and thermodynamic parameters of iron adsorption onto olive stones

Kinetic and thermodynamic parameters of iron adsorption onto olive stones

Industrial Crops and Products 49 (2013) 526–534 Contents lists available at SciVerse ScienceDirect Industrial Crops and Products journal homepage: w...

824KB Sizes 4 Downloads 85 Views

Industrial Crops and Products 49 (2013) 526–534

Contents lists available at SciVerse ScienceDirect

Industrial Crops and Products journal homepage: www.elsevier.com/locate/indcrop

Kinetic and thermodynamic parameters of iron adsorption onto olive stones Gassan Hodaifa a,∗ , J.M. Ochando-Pulido b , Saloua Ben Driss Alami b , S. Rodriguez-Vives b , A. Martinez-Ferez b a b

Molecular Biology and Biochemical Engineering Department, University of Pablo de Olavide, 41013 Seville, Spain Chemical Engineering Department, University of Granada, 18071 Granada, Spain

a r t i c l e

i n f o

Article history: Received 1 February 2013 Received in revised form 21 May 2013 Accepted 27 May 2013 Keywords: Iron Adsorption Olive stones Industrial wastewaters Kinetic study.

a b s t r a c t Olive stones biomass, by-product of olive oil industry, has been addressed in the present study as adsorbent for iron. Experimental results have shown that the pretreatments performed have not favored the iron adsorption capacity, demonstrating that direct reuse (as manufactured) or a simple washing with cold and hot water is sufficient. Results obtained indicate that the adsorption process is fast and spontaneous within the first 10–20 min. The experimental data supports both pseudo-first and pseudo-second order models. Kinetic parameters and equilibrium adsorption capacity were found to be increased upon stirring rates above 75 rpm. Also temperature effect was studied. Adsorption capacity values, qe , raise as temperature increases from 278 to 343 K, pin-pointing for an endothermic adsorption process. The adsorption isotherms were obtained from 5 different temperatures in the ranges of 278–343 K and 5–100 mg dm−3 iron (III) concentrations. These adsorption data were fitted with Langmuir isotherm. In addition, the mean values of thermodynamic parameters of activation energy (Ea = 8.04 kJ mol−1 ), standard free energy (G0 = −19.51 kJ mol−1 ), standard enthalpy (H0 = 8.86 kJ mol−1 ) and standard entropy (S0 = 91.4 J mol−1 K−1 ) of the adsorption mechanism were determined. What is more, the present process is environmentally friendly and may be able to reduce the iron load from different effluents, also providing an affordable technology for small and medium-scale industry. © 2013 Elsevier B.V. All rights reserved.

1. Introduction Iron is one of the most abundant metals of the Earth’s crust. It occurs naturally in water in soluble form as ferrous iron (bivalent iron in dissolved form Fe(II) or Fe(OH)+ ) or complexed form like ferric iron (trivalent iron Fe(III) which precipitates as Fe(OH)3 ) or even bacterial form, too. The presence of iron in water can also have an industrial origin such as metal plating, mining, iron and steel industry, metals corrosion, etc. There are many industrial situations where iron or impurities must be removed from solutions (Ghosh et al., 2008). Iron in drinking water and water supplies causes problems, such as giving reddish color and odor to water bodies (Cho, 2005). Iron removal is among the most problematic issues for water potabilization, and involves taste, visual effects and clogging, among others. Excess of iron may be present in groundwater or can occur due to corrosion of iron pipes or residual of iron based coagulants. In raw fresh water, iron concentration is usually in the range 0–50 mg dm−3 (Stegpniak et al., 2008).

∗ Corresponding author. Tel.: +34 954 978 206; fax: +34 954 349 813. E-mail address: [email protected] (G. Hodaifa). 0926-6690/$ – see front matter © 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.indcrop.2013.05.039

Advanced oxidation processes (AOPs) are known for their capability to mineralize a wide range of organic compounds in wastewaters. AOPs involve generation of highly reactive radical species, mainly hydroxyl radical (Legrini et al., 1993). Hydrogen peroxide has been used to reduce BOD, COD, offensive odor and foaminess in domestic or industrial wastewater for many years, and can be used as an autonomous treatment or as an improvement of existing physical or biological treatment processes, according to the situation. Also, it can be used alone (Millero et al., 1989) or with a catalyst such as iron salts (Fe3+ ) (Nieto et al., 2009). Olives are the most extensively cultivated fruit crop in the world. Olive cultivation is particularly widespread throughout the Mediterranean region and plays an important role in its rural economy, local heritage and environment protection. The largest producing countries are located in the Mediterranean and Middle East regions providing 98% of the total cultivated surface area, and 99% of the total olive fruit production (Niaounakis and Halvadakis, 2006). The total world production figures of olives and virgin olive oil, for the year 2010, were reported to be 20.8 and 3.27 million tons, respectively (FAOSTAT, 2012). In Spain, the main olive oil producer worldwide, there are more than 1700 olive oil factories, which gave rise to more than 1.54 million tons of virgin olive oil during the 2010 campaign (FAOSTAT, 2012). Olives and olive oil production grows

G. Hodaifa et al. / Industrial Crops and Products 49 (2013) 526–534

527

Table 1 Physicochemical properties of the crude olive stone used in this study.

year by year, and so does olive oil mill wastewater (OMW). An average sized olive oil factory produces 10–15 m3 /day of OMW. Low pH, extremely high concentration of suspended and dissolved solids as well as heavy organic load are the regular characteristics of OMW. Among the latter, the high concentration of phenols and tannins commonly present in this effluent confers OMW phytotoxic and antimicrobial properties and low biodegradability (Garrido Hoyos et al., 2002; Rozzi et al., 1988). OMW must be treated before its disposal according to the present-day European environmental rules. This constitutes a high cost for the olive oil manufacturer. In addition, decentralization and small size of most olive oil production factories impede a centralized treatment for OMW and makes it necessary to find a feasible and flexible solution for the small plants (Nieto et al., 2009; Ochando et al., 2012). Nieto et al. (2009–2011) have shown the advantages of Fenton’s reagents in the treatment of OMW: (i) high efficiency, (ii) simplicity, (iii) lack of residues and (iv) ability to treat many different compounds. In addition, it can be used as a pretreatment stage before a biological step in order to increase the biodegradability of the recalcitrant compounds and thus lower the toxicity of these effluents (Bianco et al., 2011). In the industrial plant proposed by Nieto et al. (2010a) for the treatment of OMW from a continuous two-phase decanting process by advanced chemical oxidation (homogeneous Fenton-like reaction) using FeCl3 as catalyst, the total iron concentration at the outlet of the plant was reduced to 6.8 mg dm−3 by using a filter of raw olive stones. It is also worth highlighting that, just in Spain, olives and olive oil industry generate more than 370,000 tons of triturated olive stones as by-product per year, most of which are destined for combustion and for production of active carbon (Rodríguez et al., 2008). Following this line, different studies have shown that olive stones could be used as biomass sources to eliminate pollutants such as phenols (Stasinakis et al., 2008), dyes (Akar et al., 2009), or heavy metals like Cd(II), Pb(II), Ni(II) and Cu(II) (Blazquez et al., 2005; Fiol et al., 2006). In the present study, the elimination of iron (III) ions by olive stones was investigated as a function of different pretreatments, stirring rates, and temperatures. The sorption mechanism was investigated through various adsorption kinetic models including pseudo-first and pseudo-second order models. The activation energy, which is an indicator of the type and mechanism of the biosorption process, was also evaluated using the kinetic constants. Since the evaluation of the heat change of the biosorption mechanism is very important for the reactor design, the thermodynamics of the biosorption process was also investigated.

iron adsorption. Experiments were carried in 200 cm3 Erlenmeyer flasks. The different iron solutions were prepared from a standard 30% w/w aqueous iron (III) chloride solution from QP Panreac S.A., Spain. In this work, different experimental series were carried out. The first series was conducted to address the influence of the pretreatment of olive stones in the biosorption process. With this purpose, the following pretreatments were carried out: raw olive stones (without pretreatment), washing of olives stones with cold and hot water, olive stones extraction with various organic solvents (nhexane, ethyl acetate, and mixture of n-hexane and ethyl acetate in the ratio 1:1 v/v). In the second series, experiments at different stirring rates values of 0, 60, 75, and 117 rpm were conducted. In the third series, adsorption experiments were conducted at different temperatures, 278, 293, 303, 323, and 343 K. In all experimental series the common adsorption conditions were 20 mg dm−3 Fe(III), 37.5 g dm−3 olive stones, and pH equal to 2.9. The temperature of the biosorption process was fixed at 293 K in the first and second series, and the stirring rate in the first and third experimental series was maintained at 117 rpm. Also, for the determination of the adsorption isotherms at different temperatures, five additional series of experiments at different Fe(III) ions concentrations (5–100 mg dm−3 ) for each temperature in the range 278–343 K were performed. In these experiments the common adsorption conditions were olive stones concentration 37.5 g dm−3 , stirring rate 117 rpm and pH 2.9. The amount of adsorbed iron was spectrophotometrically determined by iron content difference before and after adsorption, and iron concentration after the adsorption process was determined in the supernatant obtained by centrifugation.

2. Materials and methods

2.3. Analytical methods

2.1. Samples and preparation of olive stones

In very first place, physicochemical characterization of the raw olive stones used in this study was fully completed (Table 1). Elemental composition was determined by means of an elemental analyzer (Fison’s-Carlo Erba, mod. 1108 CHNS) in the dry residue. The lignocellulosic products were determined according to lignin parameters (TAPPI T222 os-74), neutral detergent fiber (NDF) and acid detergent fiber (ADF) (Van Soest and Wine, 1967). The percentages of hemicellulose and cellulose in the solid residue were calculated according to the expressions:

Olive stones were a waste acquired from an olive oil extraction plant, S.A.T. Olea Andaluza, located in Baeza in the province of Jaén (Spain). Olive stones were obtained from the separation from pulp (initial particle size < 4.76 mm), then washed with cold and finally with boiled water for two hours in order to remove the remaining organic matter which could interfere in the results. They were dried at 333 ± 1 K in an oven.

Parameters

Amount

Elemental analysis

Carbon, % (w/w) Oxygen, % (w/w) Hydrogen, % (w/w) Nitrogen, % (w/w) Sulphur, % (w/w)

Lignocellulosic product

Lignin, g kg−1 Hemicellulose, g kg−1 Cellulose, g kg−1

Pore volume, cm3 g−1

VMacropores VMesopores

0.163 0.0340

Specific surface area

SBET (m2 g−1 )

0.600

%Hemicellulose = %NDF − %ADF

50.4 42.2 6.96 0.4 0.04 404.0 322.0 272.0

(I)

2.2. Iron adsorption experiments For the study of iron adsorption on olive stones biomass as adsorbent, the biomass of olive stones was introduced in iron solutions with concentrations higher than those usually registered in the real wastewater condition, so as to detect the exact amount of

%Cellulose = %ADF − %lignin

(II)

Mercury porosimetry (Thermo Electron Corporation, Pascal 440 and 140 Series) was employed to characterize the porosity of olive stones by applying various levels of pressure to samples (0.3 g

528

G. Hodaifa et al. / Industrial Crops and Products 49 (2013) 526–534

Table 2 Identification of the acidic and basic surface groups on olive stone used in this study.

Acidic surface groups, meq g−1 (a) Strong carboxylic groups (b) Weak carboxylic groups (c) Phenolic groups Basic surface groups, meq g−1

Crude olive-stone

Washed olive-stone

1.20 0.125 1.60 0.500

1.10 0.0250 1.47 0.500

approximately) immersed in mercury, at environmental temperature (Nieto et al., 2010b). Specific surface area, SBET , was determined by N2 adsorption at 77 K carried out with a Micromeritics 2010 analyzer after degas at 300 ◦ C for 48 h and by using BET equations (Brunauer et al., 1938). The identification of surface groups was carried out using the Boehm methods (Boehm, 1966, 1990, 2002). All iron ions were reduced to iron ions (II), which with thioglycolate medium and a derivative of triazine, formed a reddish-purple complex that was determined photometrically at 565 nm (Standard German methods ISO 8466-1 and German DIN 38402 A51). All adsorption experiments and analytical methods were made at least three times each. The calculation and statistical methods used are available in the OriginPro 8.0 program.

and adsorption diffusion models were tested. The models which fitted the experimental data more faithfully were the pseudo-first and pseudo-second order models. The characteristic constants for the pseudo-first order equation of Lagergren (1898), based on solid capacity, were determined using the expression: dqt = k1 (qe − qt ) dt

(1)

Integrating this for the boundary conditions t = 0 to t =t∞ and qt = 0 to qt = qe , Eq. (1) may be rearranged to obtain: qt = qe (1 − e−tk1 )

(2)

where k1 is the rate constant of pseudo-first order sorption (g mg−1 min−1 ), qe is the amount of solute sorbed at equilibrium (mg g−1 ), qt is the amount of solute sorbed on the surface of the sorbent at any time t (mg g−1 ). The pseudo-second order model is based on solid phase (Ho and McKay, 1999), upon assumption that the rate-limiting step may be chemical sorption or chemical sorption involving valence forces through sharing or exchange of electrons between sorbent and sorbate. The surface site-sorbate reaction may be represented as follows: Active surface site + sorbate → surface site

3. Results and discussion

− sorbate surface complex

3.1. Characterization of olive stones Composition of the olive stones (OS) used in this study is given in Table 1. This characterization is an important analysis to understand the behavior of iron biosorption mechanism on OS surface. Elemental analysis shows that OS contains mainly carbon and oxygen and a little hydrogen percentage (≈7%). However, only small amounts of nitrogen and sulphur (<1%) were detected. Otherwise, OS is a lignocellulosic material, with hemicellulose, cellulose and lignin as main components. The higher lignin content of our samples can be justified by the fact that they presented some residues from skin and seeds (Rodríguez et al., 2008; Matos et al., 2010). Pore sizes are actually rated in accordance with the classification adopted by the International Union of Pure and Applied Chemistry (IUPAC), that is, micropores (diameter < 2 nm), mesopores (2 nm < diameter < 50 nm) and macropores (diameter > 50 nm). Table 1 shows that olive stones, a natural lignocellulosic biomass, are not a microporous material, and predominant pores are mainly macropores and a minor fraction of mesopores. The low total cumulative volume indicates that the surface presents a non-porous structure (confirmed by the low value of SBET = 0.6 m2 g−1 ), though upon determination of the porosity the surface showed certain roughness mainly manifested as macropores (Nieto et al., 2010b). Many properties of OS, in particular their wetting and adsorption behavior, are decisively influenced by chemisorbed oxygen. Oxygen in OS surface can be bound in the form of various functional groups which are similar to those known from organic chemistry (Boehm, 2002). Table 2 shows the identification of surface functional groups of OS. Carboxylic groups on OS surfaces enhance cation exchange properties, and OS always exhibit anion exchange capacity because basic surface oxides are always present when OS are exposed to the atmosphere. However, the concentration of basic surface sites is relatively small (0.5 meq g−1 , Table 2), whereas considerable cation exchange capacities can occur (Boehm, 1994). 3.2. Adsorption kinetic The adsorption mechanism was addressed through various adsorption kinetics models. Therefore, adsorption reaction models

It is assumed that the sorption capacity is proportional to the number of active sites occupied on the sorbent, and thus the kinetic rate law can be written as follows: dqt = K2 (qe − qt )2 dt

(3)

where K2 is the rate constant of pseudo-second order sorption (g mg−1 min−1 ). Integrating and applying boundary conditions t = 0 to t = t∞ and qt = 0 to qt = qe , Eq. (3) becomes: qt =

t 1 K2 q2 e

+

t qe

(4)

and h = K2 q2e

(6)

where h is the initial adsorption rate (g mg−1

min−1 ), and Eq. (4)

can be rearranged to obtain: qt =

t 1 h

+

t qe

(7)

3.2.1. Influence of OS pretreatment on adsorption kinetics It is well known that cellulosic waste materials can be obtained and employed as cheap adsorbents and their performance with regard to the removal of heavy metal ions can be affected by physical and chemical pretreatment (Selatnia et al., 2004; Ngah and Hanafiah, 2008). Fig. 1 shows experimental data of Fe (III) adsorption vs. time onto olive stones upon different pretreatments. It can be deduced that pretreatment affects the adsorption efficiency of Fe(III) onto OS, with the best results obtained using OS without any pretreatment, as straightly received from the olive oil extraction plant. The decrease of the adsorption capacity of iron (III) on the washed OS shows that in this case the chemical structure of OS is affected owed to the removal of some active surface groups, which participate or assist in iron binding. Washing of OS removes the remaining organic matter that could be stuck after being separated from the pulp, which could also occur by boiling in hot water. This second

0.997 0.982 0.995 0.998 0.988 0.992 1.13 0.585 2.79 3.66 2.57 3.08

0.456 0.132 2.36 3.05 1.42 1.85

2.49 × 10 2.30 × 10−3 3.85 × 10−4 8.42 × 10−5 4.99 × 10−4 2.62 × 10−4

R2 RSS

−4

Residual sum of squares.

3.2.2. Influence of stirring rate on adsorption kinetics The effect of stirring rate on OS adsorption was studied as well. From Fig. 2 it can be seen that the stirring rate, in the batch system, influences deeply the adsorbed amount of iron. Adsorption is possible in all cases, even without agitation. Increasing the

a

operation may cause hemicellulose autohydrolysis of olive stones, ˜ which starts at a temperature of 373 K (Fernández-Bolanos et al., 1999, 2001). The decrease of the adsorption capacity is even more noticeable when using olive stones extracted with hexane, ethyl acetate or mixture of both, being then removed even more functional surface groups involved in this mechanism. Experimental data were fitted to pseudo-first and pseudosecond order models by nonlinear regression. The obtained parameters for both kinetic models are given in Table 3. As seen from this table, the pseudo-first and pseudo-second order models fit with acceptable accuracy the experimental values. The parameters of the goodness of the fit for pseudo-first order model were R2 > 0.946 and residual sum of squares (RSS) < 7.03 × 10−3 , and the goodness of the fit for pseudo-second order model were R2 > 0.982 and RSS < 2.30 × 10−3 . The correlation coefficients calculated for both models and the other parameters are shown in Table 3. On the other hand, the calculated equilibrium sorption capacities for both models in all of cases, qe , are close to the experimental values, qe,exp , and these values are in the same order of those obtained by Nieto et al. (2010b) at the same operation conditions. Fig. 1 shows that adsorption of iron on untreated (unwashed) olive stone increases with time, and there are stability intervals with regard to the adsorbed amount of iron. This could be explained on the basis that iron was adsorbed onto the surface initially, and over time agitation causes a washing effect leading to the removal of a portion of the material layer attached to the surface of olive stones, thereby releasing a new surface which can further adsorb more iron ions. In Table 3, the mathematical models parameters for both stability intervals are reported, the first until 70 min (qe,exp = 0.335 mg g−1 ) and the second for 360 min (qe,exp = 0.381 mg g−1 ). It is finally worth indicating that the rest of the present work was performed with washed olive stones, since untreated OS (unwashed) cause turbidity in the samples, which then interferes in the photometrical determination of iron. Hence, samples analyzed with untreated OS were centrifuged prior to analysis to prevent the effect of turbidity interference in the results.

0.354 0.384 0.302 0.228 0.214 0.196

Fig. 1. Influence of different pretreatment of olive stones on the Fe(III) ions adsorption. Olive stone pretreatment: crude (unwashed) (,) washed (), extracted with: hexane (䊉), with ethyl acetate (♦), with hexane and ethyl acetate (). (−) Solid line corresponding to adjusted experimental data by pseudo-second order model. Adsorption conditions: Fe(III): 20 mg dm−3 , olive stones: 37.5 g dm−3 , stirring rate: 117 rpm, pH 2.9 and temperature 293 K.

1.000 0.946 0.999 0.997 0.972 0.996

400

3.65 × 10 7.03 × 10−3 4.72 × 10−5 1.53 × 10−4 1.14 × 10−3 1.23 × 10−4

350

0.174 0.124 0.206 0.222 0.225 0.180

300

0.336 0.367 0.299 0.225 0.209 0.192

250

0.335 0.381 0.299 0.226 0.215 0.192

200 t, min

Unwashed (70 min) Unwashed (360 min) Washed Extracted with hexane Extracted with ethyl acetate Extracted with hexane and ethyl acetate

150

h (g mg−1 min−1 )

100

K2 (g mg−1 min−1 )

50

−5

0

Pseudo-second order model

0.00

qe (mg g−1 )

0.05

R2

0.10

RSSa

0.15

k1 (g mg−1 min−1 )

0.20

qe (mg g−1 )

0.25

Pseudo-first order model

qt, mg g

-1

0.30

qe,exp (mg g−1 )

0.35

529

Olive stone pretreatment

0.40

Table 3 Influence of different pretreatment of olive stones on the Fe(III) ions adsorption and pseudo-first and pseudo-second order model parameters for sorption process. Adsorption conditions: Fe(III): 20 mg dm−3 , olive stones: 37.5 g dm−3 , stirring rate: 117 rpm, pH 2.9 and temperature 293 K.

G. Hodaifa et al. / Industrial Crops and Products 49 (2013) 526–534

530

G. Hodaifa et al. / Industrial Crops and Products 49 (2013) 526–534

0.35

a)

0.30

-1

0.20

qt, mg g

qt, mg g

-1

0.25

0.15 0.10 0.05 0.00 0

50

100

150

200

250

300

350

0.55 0.50 0.45 0.40 0.35 0.30 0.25 0.20 0.15 0.10 0.05 0.00

400

0

50

100

150

t, min

200

250

300

350

400

250

300

350

400

t, min 0.55

b)

Fig. 2. Influence of different stirring rates on the Fe(III) adsorption onto olive stones. Stirring rates: () 0 rpm, () 60 rpm, (䊉) 75 rpm, and (♦) 117 rpm. (−) Solid line corresponding to adjusted experimental data by pseudo-first order model. Adsorption conditions: Fe(III): 20 mg dm−3 , olive stones: 37.5 g dm−3 , pH 2.9 and temperature 293 K.

0.50

stirring rate enhanced the adsorption capacity of iron (III) onto olive stones (Fig. 2). About 60 rpm was sufficient to initiate mixing. Operating at a stirring rate equal to 117 rpm allowed reaching values close to the maximum equilibrium adsorption within about 10 min, whilst in the other experiments the equilibrium was achieved later (Fig. 2). Furthermore, as it can be observed in Fig. 2, for operating time greater than 210 min no significant differences were detected between 0 and 75 rpm. However, when the stirring rate was increased from 75 to 117 rpm, the adsorption capacity (qt ) was promoted. This increase highlights certainly the limitation of the mass transfer through the solution and in the boundary layer during the adsorption process when a stirring rate of 75 rpm or lower was selected. Experiments at higher stirring rates were not conducted, since major stirring rate may incur in considerable cost in scale-up, and moreover care must be taken to avoid fragmentation of olive stones. Table 4 shows the model parameters of the pseudo-first and pseudo-second order mechanisms determined from nonlinear regression fit of the experimental data. For pseudo-first order model the values of k1 and qe increased with the stirring rates from 0.0672 to 0.206 g mg−1 min−1 and from 0.160 to 0.299 mg g−1 , respectively. In the case of the pseudo-second order model, the values of K2 , h, and qe became incremented with the stirring rate from 0.752 to 2.79 g mg−1 min−1 , 0.0227 to 0.255 g mg−1 min−1 , and from 0.174 to 0.302 mg g−1 , respectively. The values of k1 and K2 confirm the effect of the boundary layer, where these values are similar when stirring rates lower than 75 rpm agitation are fixed. Also, it is important to indicate that in both models the adsorption capacity values calculated at equilibrium are similar to those obtained experimentally (Fig. 2). On the other hand, the second equilibrium zone observed when operating without agitation (Fig. 2) can be explained by the appearance of another

qt, mg g

-1

0.45 0.40 0.35 0.30 0.25 0.20 0.0 0

50

100

150

200 t, min

Fig. 3. Effect of temperature on the Fe(III) adsorption onto olive stone. Temperatures: () 278 K, () 293 K, () 303 K, () 323 K, and () 343 K. (−) Solid line corresponding to adjusted experimental data by pseudo-second order model for (a) 360 min and (b) 90 min of time operation. Adsorption conditions: Fe(III): 20 mg dm−3 , olive stones: 37.5 g dm−3 , stirring rate: 117 rpm, and pH 2.9.

type of adsorption that may be chemical (slower than physical adsorption). Similar results for other metallic ions were obtained by Martinez-Garcia et al. (2006), who reported a stirring rate equal to 120 rpm as optimal for the biosorption of cadmium solution onto olive oil waste (mixture of pulp and olive stone) from a two-phase decanter system. 3.2.3. Effect of temperature on adsorption kinetics and thermodynamic parameters Fig. 3 shows the equilibrium removal of Fe(III) ions as a function of temperature, for experiments conducted at constant concentrations of Fe(III) equal to 20 mg dm−3 . The adsorption of Fe(III) onto the surface of olive stones took place quickly disregarding the temperature (278–343 K), in a way that around 40–90% of the metal iron ions in solution were removed within the first 10–20 min. After

Table 4 Influence of stirring rate on the Fe(III) adsorption and determination of pseudo-first and pseudo-second order model parameters for sorption process. Adsorption conditions: Fe(III): 20 mg dm−3 , olive stones: 37.5 g dm−3 , pH 2.9 and temperature 293 K. Stirring rate (rpm)

qe,exp (mg g−1 )

Pseudo-first order model −1

qe (mg g 0 60 75 117 a

0.162 0.206 0.215 0.298

Residual sum of squares.

0.160 0.204 0.215 0.299

)

k1 (g mg 0.0672 0.0674 0.0990 0.206

−1

min

Pseudo-second order model −1

)

a

2

−1

RSS

R

qe (mg g

4.19 × 10−4 1.95 × 10−3 6.35 × 10−4 4.72 × 10−5

0.980 0.954 0.986 0.999

0.174 0.216 0.223 0.302

)

K2 (g mg−1 min−1 )

h (g mg−1 min−1 )

RSS

R2

0.752 0.602 1.03 2.79

0.0227 0.0280 0.0510 0.255

4.29 × 10−4 9.50 × 10−4 1.31 × 10−3 3.85 × 10−4

0.979 0.978 0.971 0.995

G. Hodaifa et al. / Industrial Crops and Products 49 (2013) 526–534

531

Table 5 Influence of temperature on the Fe(III) adsorption and determination of pseudo-first and pseudo-second order model parameters for sorption process. Adsorption conditions: Fe(III): 20 mg dm−3 , stirring rate 117 rpm, olive stones: 37.5 g dm−3 , pH 2.9, and adsorption time 90 min. Temperature (K)

qe , exp (mg g−1 )

Pseudo-first order model qe (mg g−1 )

278 293 303 323 343

0.211 0.299 0.343 0.425 0.480

0.183 0.206 0.294 0.285 0.326

Pseudo-second order model RSSa

qe (mg g−1 )

R2 −4

1.34 × 10 1.88 × 10−5 1.06 × 10−4 1.90 × 10−4 1.68 × 10−5

0.996 1.00 0.999 0.999 1.00

0.220 0.308 0.348 0.432 0.485

K2 (g mg−1 min−1 ) 2.15 2.13 4.13 3.02 4.71

h (g mg−1 min−1 )

R2

RSS −4

1.68 × 10 2.89 × 10−5 4.88 × 10−5 1.02 × 10−4 4.66 × 10−5

0.104 0.202 0.501 0.564 1.11

0.995 0.997 0.999 0.999 1.000

Residual sum of squares.

20 min of adsorption, the removal rates of this ion slowed significantly, probably due to saturation on the biomass surface. On the other hand, in this figure it can be seen that the adsorption capacity values, qe , become raised as the temperature increased from 278 to 343 K. Enhancement of adsorption capacity of OS at higher temperatures may be attributed to the activation of the adsorbent surface and increase in the mobility of metal ions. Also, this fact demonstrated an endothermic biosorption process (Blazquez et al., 2005). In Fig. 3b, two equilibrium zones (states of stability) are confirmed in the adsorption curves at temperatures above 303 K, highlighting that the second zone starts at 90 min, which can be explained whether by the appearance of another type of adsorption that may be chemical (slower than physical adsorption) or owed to modification of the biomass surface due to its exposure to high temperatures, increasing the density of accessible sites for adsorption of iron. In light of this fact, the experimental data were adjusted to the kinetic models at two operating times of 90 and 360 min (Fig. 3). The kinetic parameters, obtained for the two best-fitting models depicting experimental data at adsorption time intervals of 90 and 360 min, show little significant variation between themselves for each case except for some temperatures (Table 5 shows only the parameters of the two models at 90 min of adsorption time). In general, the values of k1 , K2 , and h increased with temperature, as expected. Both models fit the experimental data in quite an acceptable manner (R2 > 0.99, and RSS < 1.90 × 10−4 , Table 5). The qt values calculated from the pseudo-first and pseudosecond order models adjustment increase from 0.211 to 0.480 mg g−1 and 0.220 to 0.485 mg g−1 when operating temperature was incremented from 278 to 343 K in an interval of 90 min (Table 5), respectively. The same trend was observed by studying the kinetics at 360 min at a temperature range between 278 and 303 K (data not shown for brevity). It is noted however that for a temperature of 323 K, although the adsorption capacity continues to increase, the values of k1 , K2 , and h decreased, which may indicate a variation in the adsorption process, for example instead of applying in a monolayer it takes place in two or more layers, while there is a decrease of the adsorption rate. The rate of adsorption is affected by the temperature dependence of the sorption constant rate (rate constant of pseudo-first or pseudo-second order model sorption, k1 and K2 ) and the temperature is given by the Arrhenius equation:

 E 1 a

k1 or K2 = A exp −

11 10 9 8 2.9

3.0

3.1

3.2

3.3 3

1/T x 10 , K

3.4

3.5

3.6

3.7

-1

Fig. 4. Arrhenius graphic and calculation of activation energy of the Fe(III) adsorption. Values of Napierian logarithm of k1 () and k2 () for the pseudo-first and pseudo-second order model, respectively. Adsorption conditions: Fe(III): 20 mg dm−3 , olive stones: 37.5 g dm−3 , stirring rate: 117 rpm, and pH 2.9.

The constant rates obtained in the first equilibrium zone at 90 min represent a linear equation, where the intercept and the slope of the line give the frequency factor and activation energy of the adsorption process (Fig. 4). The activation energy provides information about the dependence of the adsorption process with temperature: the greater the activation energy value, the more temperature affects the adsorption process. To evaluate the nature of the interaction between Fe(III) ions and olive stone sites, the value of activation energy obtained (7.05 and 9.02 kJ mol−1 , Table 6) allows to know the nature of the adsorption process. In this sense, the Ea values fall within the 5–20 kJ mol−1 range generally considered for processes in which physisorption predominates. This is revealed not only by the low activation energy value but also by the speed of the process (Glasstone et al., 1941). Blazquez et al. (2005) reported a value of 8.44 kJ mol−1 for removal of cadmium ions with olive stones. The study of the temperature effect on iron removal by OS enabled us to determine the thermodynamic parameters (G0 , H0 and S0 ) of these reactions. The standard Gibbs free energy change, G0 , when the adsorption reaches equilibrium state can be written as follows: G0 = −RT ln K

where A is the frequency factor (min−1 ), Ea the activation energy (kJ mol−1 ), T the temperature (K) and R the gas constant (8.314 J mol−1 K−1 ). This equation can be expressed with the Napierian logarithm:

E 1 a R

12

(10)

(8)

R T

ln(k1 or K2 ) = ln (A) −

13

ln(k1 or k2)

a

0.213 0.298 0.344 0.429 0.481

k1 (g mg−1 min−1 )

T

(9)

Table 6 Activation energy calculated for Fe(III) adsorption on olive stones at the experimental conditions.

Pseudo-first order model Pseudo-second order model

A (min−1 )

Ea (kJ mol−1 )

2.24 × 10 5.91 × 106

7.05 9.02

5

532

G. Hodaifa et al. / Industrial Crops and Products 49 (2013) 526–534

Table 7 Parameters of Langmuir isotherms and the values of the thermodynamic parameters of Fe(III) adsorption onto olive stone at various temperatures. t* (min)

Temperature (K)

Langmuir constants −1

qmax (mg g

)

K (L mol

−1

)

G◦ (kJ mol−1 )

H kJ mol−1

S (J mol−1 K−1 )

2

R

90

278 293 303 323 343

0.463 1.20 1.33 1.32 1.91

1537.5 1886.5 1820.1 2312.1 3425.1

0.962 0.963 0.964 0.902 0.912

−17.0 −18.4 −18.9 −20.8 −23.2

9.12

92.0

120

278 293 303 323 343

0.457 1.67 1.56 1.73 2.12

1677.6 1340.3 1706.1 1847.4 3386.0

0.957 0.967 0.985 0.923 0.985

−17.2 −17.5 −18.7 −20.2 −23.2

8.60

90.8

*

Adsorption time.

where T is the temperature (K), R is the gas constant (8.314 J mol−1 K−1 ) and K denotes the equilibrium constant. Therefore, G0 for an adsorption reaction can be estimated if Langmuir constant, K, for the adsorption mechanism is known. Hence, in order to determine the adsorption isotherms at different temperatures (from 278 to 343 K), the initial iron concentration for each temperature was varied in the interval ranging from 5.0 to 100 mg iron dm−3 . In all experiments performed, the adsorption proceeds at a high rate (≈5 min) and the adsorption equilibrium is attained in 10–20 min, being achieved around 80% of the total iron (III) adsorption by the olive stones in the first 5 min, and from then on existing certain variation in this equilibrium during the remainder of the experiment. Only in two cases it was observed the existence of two consecutive equilibrium intervals, the first equilibrium interval was detected from 10 to 90 min, and the second interval was determined from 120 min until the end of the experiment at 360 min (Fig. 3). This fact was observed only when working with high iron concentrations above 70 mg dm−3 and high temperatures over 323 K. Similar results were observed by Nieto et al. (2010b) who detected the appearance of a second adsorption layer. This second layer begins to form before reaching the maximum capacity of the model, and corroborates that the adsorption of iron on olive stones occurs both in a monolayer and in a multilayer form. Fig. 5 shows the adsorption isotherms of iron on olive stones at different temperatures (only for the first equilibrium intervals

1.6

qe =

Kqmax CFe 1 + KCFe

(11)

where qe is the load of Fe (mg) adsorbed per gram of support (OS) on equilibrium, K (L mol−1 ) is the Langmuir constant, qmax the maximum load of Fe (mg) adsorbed by one g of OS, and CFe (mg dm−3 ) is the iron concentration (III) at equilibrium. Table 7 shows the kinetic parameters (qmax and K) of the Langmuir equation determined from nonlinear regression fit of the experimental data for the two equilibrium intervals at 90 and 120 min. These values are in the same order of those determined by Fiol et al. (2006) for the adsorption of metallic ions at 293 K (Cd(II): qmax = 2.76 mg g−1 and K = 640 L mol−1 , Pb(II): qmax = 9.26 mg g−1 and K = 2450 L mol−1 , Ni(II): qmax = 2.13 mg g−1 and K = 3150 L mol−1 , Cu(II): qmax = 2.03 mg g−1 and K = 12,030 L mol−1 ) onto olive stones. Having determined the values of the Langmuir constant, the values of the standard Gibbs free energy change, G0 , were calculated for each temperature (Table 7). The negative value of G0 confirms the feasibility of the process and the spontaneous nature of the sorption. It is of note that G0 values up to −20 kJ mol−1 are consistent with electrostatic interaction between sorption sites and the metal ion (physical adsorption) while G0 values more negative than −40 kJ mol−1 involve charge sharing or transfer from the biomass surface to the metal ion to form a coordinate bond (Horsfall et al., 2006). Also, the change in apparent enthalpy, H0 , and entropy S0 of the adsorption mechanism can be calculated using the following equation: G0 = H 0 − T S 0

1.4 qe (mg iron/g olive-stone)

of 90 min). The experimental data (Fig. 5) were adjusted to the Langmuir model, Eq. (11):

(12)

Dividing this equation by the terms, R T, and equating to Eq. (10) the following equation can be written:

1.2 1.0

ln K =

0.8 0.6 0.4 0.2 0.0 0

20

40

60

80

100

-3

CFe, mg dm

Fig. 5. Adsorption isotherms of iron to olive stone determined at different temperatures: () 278 K, () 293 K, () 303 K, () 323 K, and () 343 K. Adsorption conditions: Fe(III): 5–100 mg dm−3 , olive stones: 37.5 g dm−3 , stirring rate: 117 rpm, pH 2.9, and operation time 90 min.

G0 = RT



H 0 R



1 S 0 − T R

(13)

The Vant’Hoff plot of ln K as a function of 1/T (at 90 min of operation, Fig. 6) gave a straight line (R2 = 0.905). Here it is noteworthy that a bad value for R2 = 0.629 was attained when adjusting ln K versus 1/T for the experimental data obtained at operation time of 120 min. The calculated slope and intercept from the plot were used to determine H0 and S0 , respectively (Table 7). In both operating times (90 and 120 min), the values of H0 are positive, indicating that the sorption reaction is endothermic. The positive values of S0 show the increasing randomness at the solid/liquid interface during the adsorption process and the entropy is a driving force (Nieto et al., 2010b; Aziz et al., 2009). Al-Anber and Al-Anber (2008) obtained values for G0 , H0 and S0 in the same order for iron (III) adsorption by olive cake (mixture between pulp and olive

G. Hodaifa et al. / Industrial Crops and Products 49 (2013) 526–534

9.0 8.5 8.0

ln(K)

7.5 7.0 6.5 6.0 5.5 2.9

3.0

3.1

3.2

3.3 3

1/T x 10 , K

3.4

3.5

3.6

3.7

-1

Fig. 6. Vant’Hoff plot for Fe(III) adsorbed by olive stone at 278, 298, 303, 323, and 343 K. Adsorption conditions: Fe(III): 20 mg dm−3 , olive stones: 37.5 g dm−3 , stirring rate: 117 rpm, pH 2.9, and operation time 90 min.

533

(iv) Kinetic pseudo-first and pseudo-second order models allow accurate fit of the experimental data. (v) The comparison between olive stones biomass and other bioadsorbent indicates the efficiency and its advantage as adsorbent medium. (vi) The use of olive stones straightly without pretreatment for metals adsorption is feasible, and can also be used as metallic cations (iron) adsorbent in industrial applications. (vii) The removal efficiency of Fe(III) ions increase with incrementing temperature. (viii) The activation energy value fall within the range considered for processes when physisorption predominates, which also accounts for the speed at which the operation occurs. (ix) The values of thermodynamic functions show that the adsorption of Fe(III) ions onto olive stones is spontaneous and of endothermic nature. (x) The adsorption mechanism is predominantly physisorption with some chemisorption also occurring. (xi) The adsorption process is environmentally friendly and is able to reduce the iron load from different effluents, and it may further provide an affordable technology for small and medium-scale industry.

Acknowledgements We are grateful to the Ministry of Science and Technology for financial support through Project PPQ2003-07873 “Treatment of olive oil wastewater for its reuse in agriculture irrigation”, and the project CTQ 2007-66178 “Depuration process of the olive mill wastewaters by Fenton like treatment and later purification by biosorption”.

References

Fig. 7. System of three packed filters (sand, olive stone, and sand) operated in the Wastewaters Treatment Plant of the company of S.A.T. Olea Andaluza located in Baeza (Spain).

stone) in a batch system for a temperature range between 301 and 318 K. In this sense, Nieto et al. (2009, 2010a) have already demonstrated the effectiveness of the use of crude olive stones to eliminate residual iron ions from the effluents of olive oil industry in real conditions, installed at the end of a wastewater treatment plant located in Baeza (Spain) operating using Fenton’s reagent, consisting of a system composed of three packed filters (sand, olive stones and gravel), depicted in Fig. 7. 4. Conclusion The results obtained in this study allow us to state the following: (i) Olive stones biomass is an agro-industrial by-product, abundant, and cheap remainder from olive oil industry. (ii) Olive stones have considerable potential for the removal of metal ions from effluents not only from olive oil industry but also from other contaminated wastewaters over a wide range of concentrations. (iii) The adsorption rate of Fe(III) ions by olive stones was rapid, stable and occurred in less than 20 min.

Akar, T., Tosun, I., Kaynak, Z., Ozkara, E., Yeni, O., Sahin, E.N., Akar, S.T., 2009. An attractive agro-industrial by-product in environmental cleanup: bye biosorption potential of untreated olive pomace. J. Hazard. Mater. 166, 1217–1225. Al-Anber, Z.A., Al-Anber, M.A.S., 2008. Thermodynamics and kinetic studies of iron(III) adsorption by olive cake in a batch system. J. Mex. Chem. Soc. 52 (2), 108–115. Aziz, A., Elandaloussi, E., Belhalfaoui, B., Ouali, M.S., De Ménorval, L.C., 2009. Efficiency of succinylated-olive stone biosorbent on the removal of cadmium ions from aqueous solutions. Colloids Surf. B: Biointerfaces 73, 192–198. Bianco, B., Michelis, I.D., Vegliò, F., 2011. Fenton treatment of complex industrial wastewater: optimization of process conditions by surface response method. J. Hazard. Mater. 186, 1733–1738. ˜ Blazquez, G., Hernainz, F., Calero, M., Ruiz-Nunez, L.F., 2005. Removal of cadmium ions with olive stones: the effect of some parameters. Process Biochem. 40, 2649–2654. Boehm, H.P., 1966. Chemical identification of surface groups. Adv. Catal. 16, 179–274. Boehm, H.P., 1990. Surface oxides on carbon. High temperatures-high pressures. In: International Symposium on the Development of Carbon Fibres and Their Applications, Seoul, South Korea, vol. 22 (3), pp. 275–288. Boehm, H.P., 1994. Some aspects of the surface chemistry of carbon blacks and other carbons. Carbon 35 (2), 759–769. Boehm, H.P., 2002. Surface oxides on carbon and their analysis: a critical assessment. Carbon 40, 145–149. Brunauer, S., Emmett, P.H., Teller, E., 1938. Adsorption of gases in multimolecular layers. J. Am. Chem. Soc. 60, 309–319. Cho, B.Y., 2005. Iron removal using an aerated granular filter. Process Biochem. 40, 3314–3320. DIN 38402 A51, 1986. German standard methods for the examination of water, waste water and sludge; general information (group A); calibration of analytical methods, evaluation of analytical results and linear calibration functions used to determine the performance characteristics of analytical methods (A 51). FAOSTAT, 2012. Website of Food and Agriculture Organization of the United Nations., http://faostat.fao.org/ ˜ Fernández-Bolanos, J., Felizón, B., Heredia, A., Guillén, R., Jiménez, A., 1999. Characterization of the lignin obtained by alkaline delignification and of the cellulose residue from steam-exploded olive stones. Bioresour. Technol. 68 (2), 121–132.

534

G. Hodaifa et al. / Industrial Crops and Products 49 (2013) 526–534

˜ Fernández-Bolanos, J., Felizón, B., Heredia, A., Rodrıˇıguez, R., Guillén, R., Jiménez, A., 2001. Steam-explosion of olive stones: hemicellulose solubilization and enhancement of enzymatic hydrolysis of cellulose. Bioresour. Technol. 79 (1), 53–61. Fiol, N., Villaescusa, I., Martinez, M., Miralles, N., Poch, J., Serarols, J., 2006. Sorption of Pb(II), Ni(II), Cu(II), and Cd(II) from aqueous solution by olive stone waste. Sep. Purif. Technol. 50, 132–140. Garrido Hoyos, S.E., Nieto, L.M., Rubio, F.C., Cormenzana, A.R., 2002. Kinetics of aerobic treatment of olive mill wastewater (OMW) with Aspergillus terreus. Process Biochem. 37 (10), 1169–1176. Ghosh, D., Solanki, H., Purkait, M.K., 2008. Removal of Fe(II) from tap water by electrocoagulation technique. J. Hazard. Mater. 155, 135–143. Glasstone, S., Laidler, K.J., Eyring, H., 1941. The Theory of Rate Processes. McGrawHill, New York. Ho, Y.S., McKay, G., 1999. Pseudo-second order model for sorption process. Process Biochem. 34 (5), 451–465. Horsfall Jr.M., Abia, A.A., Spiff, A.I., 2006. Kinetic studies on the adsorption of Cd2+ , Cu2+ and Zn2+ ions from aqueous solutions by cassava (Manihot sculenta Cranz) tuber bark waste. Bioresour. Technol. 97, 283–291. ISO 8466-1, 1990. Water quality-Calibration and evaluation of analytical methods and estimation of performance characteristics-Part 1: statistical evaluation of the linear calibration function, Genève, Switzerland. Lagergren, S., 1898. Zur theorie der sogenannten adsorption gelöster stoffe, Kungliga Svenska Vetenskapsakademiens. Handlingar 24 (4), 1–39. Legrini, O., Oliveros, E., Braun, A.M., 1993. Photochemical processes for water treatment. Chem. Rev. 93, 671–698. Martinez-Garcia, G., Bachmann, R.T.H., Williams, C.J., Burgoyne, A., Edyvean, R.G.J., 2006. Olive oil waste as a biosorbent for heavy metals. Int. Biodeterior. Biodegrad. 58 (3–4), 231–238. Matos, M., Filomena, B.M., Gandini, A., 2010. Olive stone as a renewable source of biopolyols. Ind. Crop. Prod. 32, 7–12. Millero, F.J., Le Ferriere, A., Fernandez, M., Hubinger, S., Hershey, J.P., 1989. Oxidation of hydrogen sulfide with hydrogen peroxide in natural waters. Environ. Sci. Technol. 23, 209–213. Niaounakis, M., Halvadakis, C.P., 2006. Olive Processing Waste Management: Literature Review and Patent Survey. Elsevier, Amsterdam.

Ngah, W.S., Hanafiah, M.A.K.M., 2008. Biosorption of copper ions from dilute aqueous solutions on base treatedrubber (Hevea brasiliensis) leaves powder: kinetics, isotherm, and biosorption mechanisms. J. Environ. Sci. 20, 1168–1176. Nieto, L.M., Hodaifa, G., Vives, S.R., Casares, J.A., Driss, S.B., Grueso, R., 2009. Treatment of olive-mill wastewater from a two-phase process by chemical oxidation on an industrial scale. Water Sci. Technol. 59 (10), 2017–2027. Nieto, L.M., Hodaifa, G., Vives, S.R., Casares, J.A., 2010a. Industrial plant for olive mill wastewater from two-phase treatment by chemicals oxidation. J. Environ. Eng. 136 (11), 1309–1313. Nieto, L.M., Driss, S.B., Hodaifa, G., Faur, C., Vives, S.R., Gimenez, J.A., Ochando, J., 2010b. Adsorption of iron on crude olive Stones. Ind. Crop. Prod. 32, 467–471. Nieto, L.M., Hodaifa, G., Vives, S.R., Casares, J.A., Ochando, J., 2011. Degradation of organic matter in olive-oil mill wastewater through homogeneous Fenton-like reaction. Chem. Eng. J. 173, 503–510. Ochando, J., Vives, S.R., Hodaifa, G., Martinez-Ferez, A., 2012. Fouling inhibition upon Fenton-like oxidation pretreatment for olive mill wastewater reclamation by membrane process. Chem. Eng. Process. 62, 89–98. ˜ Rodríguez, G., Lama, A., Rodríguez, R., Jiménez, A., Guillen, R., Bolanos, J.F., 2008. Olive Stone an attractive source of bioactive and valuable compounds. Bioresour. Technol. 99, 5261–5269. Rozzi, A., Limoni, N., Menegatti, S., Boari, G., Liberti, L., Passino, R., 1988. Influence of Na and Ca alkalinity on UASB treatment of olive mill effluents. Part 1. Preliminary results. Process Chem. 23, 86–90. Selatnia, A., Bakhti, M.Z., Madani, A., Kertous, L., Mansouri, Y., 2004. Biosorption of cadmium ion from aqueous solution by a NaOH-treated bacterial dead Sterptomyces rimosus biomass. Hydrometallurgy 75, 11–24. Stasinakis, A.S., Elia, I., Petalas, A.V., Halvadakis, C.P., 2008. Removal of total phenols from olive-mill wastewater using an agricultural by-product, olive pomace. J. Hazard. Mater. 160, 408–413. ´ E., 2008. The research on the possibilStegpniak, L., Kegpa, U., Stanczyk-Mazanek, ity of ultrasound field application in iron removal of water. Desalination 223, 180–186. Van Soest, P.J., Wine, R.H., 1967. Use of detergents in the analysis of fibrous feeds, IV. Determination of plant cell-wall constituents. J. Assoc. Off. Anal. Chem. 50, 50–55.