Islem Chaari *,Atef Touil ,Mounir Medhioub

1 Useful Materials Valorization Laboratory,National Center for Research in Materials Sciences,Soliman 8027,Tunisia

2 Department of Geology,Faculty of Science,Sfax,Tunisia

Keywords:Clayey material Adsorption Phenolic compounds Olive mills wastewaters Desorption

ABSTRACT Olive mills wastewater (OMW)exhibit substantial contaminated properties due to their content of phenolic constituents and organic substances.The purpose of this work was the investigation of the efficiency of clay materials for the adsorption of phenolic compounds,which are contained in OMW.Furthermore,thermal activation of the clay took place in order to improve phenolic compounds uptake and afterwards,desorption process was studied.The adsorbent was characterized using XRD,XRF and BET surface area analyses.The adsorption efficiency of phenolic compounds by raw and calcined clay at 600°C was 77.61%and 84.21%,respectively at acidic pH.The values of Gibbs free energy indicated that adsorption process is spontaneous and beneficial at higher temperatures.Alkaline medium was propitious for phenolic compounds desorption.The obtained results showed that natural clay could be used as a low-cost adsorbent for OMW treatment.

1.Introduction

Tunisia is one of the countries in the olive oil producing world.It represents the first producer and exporter of olive oil in the Southern Mediterranean area,following the European Union [1].In spite of its important contribution to the national assets,the oleic industries use the three-phases milling method that was forbidden in Spain since 1972.This milling technique was characterized by the production of a solid waste,named olive pomace (OP)and a liquid waste named olive mill wastewater (OMW).This liquid represents a serious problem in olive oil producing areas due to its elevated toxicity and pollution risk for environmental and aquatic bodies,which is mainly related to its low pH [2],and polyphenol content [3].Some researchers affirmed that phenolic compounds are the major contributors to the toxicity of olive mill wastewater,which is considered to be harmful plants and aquatic organisms [4-7,8].On the other hand,other studies found that these phenolic compounds have strong antioxidant properties that can turn olive oil residues into an inexpensive source of natural antioxidants [9,10].So,polyphenols from OMW removal has become one of the focus of a large variety of researchers.Several performing techniques have already been used for OMW treatment and removal of its phenolic compounds based on flotation,sedimentation,advance oxidation,fenton and ozonation,adsorption and biological process,etc.[11-13,14].Among the proposed treatment methods,adsorption prove to be very efficient for the removal of organic pollutants from OMW [15].Adsorption processes offer several advantages in comparison to other listed techniques such as simple design,ease of operation,high efficiency for removing hazardous substances,low capital and operational costs[16].For example,Galiatsatouet al.[17],and Azzamet al.[18],used activated carbon as an adsorbent of organic matter present in OMW.Other investigations employed low-cost biosorbents such as banana peel [19],Pomegranate peel and orange juice [20],for removing phenolic compounds from OMW.Calcium carbonate and hydroxyapatite have been extensively employed after chemical treatment,succeeding efficient phenolic compounds removal[21].However,few studies have emphasized the use of natural geological materials such as clays,iron oxide,limestones,lignite,etc.as an adsorbents of phenolic compounds present in OMW.The application of natural clays as adsorbents of phenolic compounds have acquired extensive research attention for contaminants remediation owing to their physical and chemical properties.They exhibit high porosity,high cation exchange capacity,several types of active sites on the surface[22,23]and swelling properties providing significant capacity to adsorb water [24].Moreover,the relatively low cost,the easy avaibility and the inexpensive regeneration system make them more economically viable as adsorbents in comparison with other natural materials like iron oxide and lignite.A few studies have been published regarding the adsorption of the phenolic compounds onto natural clays.Al-Malahet al.[25],used activated clay and demonstrated that hydrophobic interactions was responsible for sorption of phenolic content.Similarly,Daheb and Yaddadene[26],used local clay as a low-cost adsorbent for removing phenolic compounds from OMW yielding an efficiency of 88%.Azzamet al.[27],treated OMW with a mixture of volcanic tuff,natural clay and charcoal and concluded that mixed adsorbents appeared to be a useful mineral in reducing the phenolic compounds amount (57.4%).

The aim of this work is to investigate the efficiency of natural Tunisian clay,as low cost adsorbent for the removal of phenolic compounds from OMW.The influence of its composition and surface chemistry on its ability to adsorb phenolic compounds from OMW was evaluated and the effects of several processing parameters such as pH,contact time,adsorbent amount of clay,thermal activation of clay were studied.Moreover,kinetic and thermodynamic parameters for phenolic compounds were determined.

2.Materials and Methods

2.1.Adsorbent

The raw clay(KC)was collected in Lower Oligocene level of the Fortuna formation (Khledia area,North of Tunisia).The clay was used without further purification.Mineralogical analysis of the KC was determined by X-ray diffraction analysis using a PANalytical X’ Pert High Score Plus diffractometer with monochromated CuKα radiation.Specific surface area measurement of KC used a Quantachrome Autosorb Automated Gas Sorption System by performing N2-physisorption at 77 K.It was calculated using Brunauer-Emmett-Teller (BET) equation.For the determination of the point of zero charge of the raw clay,method adopted by Chaariet al.[28],was used.The granular activated carbon is a commercial product of the ‘‘Kanto chemical Co.INC.”Type.Cicareagent‘‘.The iron oxide was collected in the Tamra mine (Nefza region),which is very rich in hematite and goethite ores.The lignite was collected in the Cap Bon region.It is belong to Serravalian-Tortonian of Saouaf formation.

2.2.Olive mill wastewater

OMW sampling was withdrawn on October 2016-2017 from a mill located in Ghraiba area (Sfax,Tunisia).OMW was collected in plastic vessels,transported to the laboratory immediately after sampling,and was kept at 4 °C in a refrigerator.The characterization of OMW for pH,chemical oxygen demand (COD),phenolic compounds (PC),electrical conductivity (EC) was investigated.pH was determined with a pH meter (HANNA).A conductivimeter type proline was used to determine electrical conductivity (EC).Chemical oxygen demand was determined by a colorimetric method[29].Phenolic compounds were realized following the colorimetric method of Folin-Ciocalteau [30].The absorbance was 765 nm.

2.3.Adsorption studies

Batch adsorption experiments were carried out in 250 ml glass flasks by agitating the adsorbent in 20 ml of OMW solution of desired phenolic compounds concentration and pH at different temperatures in a rotary shaker (rotatest 74581) at 200 r·min-1at (25 ±2) °C.The pH (2-9) of the solutions was adjusted using NaOH (1 mol·L-1) or HCl (11 mol·L-1) by using Hanna pH-Meter.Other varying parameters were contact time(5-90 min),adsorbent amount(0.1-1.6 g),thermal activation(105-600°C)and temperature (25-55 °C).The solutions were separated from the adsorbent by filtration and centrifuged at 4000 r·min-1for 20 min.Phenolic compounds was determined using the spectrophotometric method.The adsorption efficiency,E(%),was calculated as equation (1):

whereC0andCeare the initial and equilibrium concentration of phenolic compounds (g·L-1).

2.4.Desorption of phenolic compounds

The used clay was separated from the OMW by centrifugation.The phenolics-loaded adsorbent were washed many times with distilled water to remove any unadsorbed compounds and dried in oven for about 5 h.Sample loaded with phenolic compounds,was shaken (4.0 g·L-1) with solutions at different pH (range 2-9).After 2 h of contact time,the phenolics concentration in the liquid phase was analyzed.The desorption efficiency,Y(%)of the phenolic compounds was calculated as follow:

3.Results and Discussion

3.1.Characterization of adsorbent

Mineralogical analysis indicate 44% of smectite,15% of illite,15% of kaolinite and 26% of quartz (Table 1).Chemical analysis of the adsorbent (Table 2) shows high amounts of iron oxides(7.42%) and SiO2(60.16%).The SiO2/Al2O3ratio is 3.54 and the sum (SiO2+Al2O3) represents 77% (mass) of the mineral.This chemical composition indicates the presence of several exchangeable cations (Mg+,K+and Na+) thus favoring the adsorption process.The specific surface area of KC was about 52.2 m2·g-1.The plot pHf=f (pHi) is shown in Fig.1.Results show that pHZPCof KC was 8.

Table 1 Mineralogical composition of clay sample

Table 2 Chemical composition of clay sample

Fig.1. pHZPC of KC.

3.2.OMW characterization

According to the results shown in Table 3,the OMW has a pH of 4.73,which can be explained by the existence of organic acids.Phenolic compounds in OMW was 4.406 g·L-1.The quantity of dry matter (SM) of the OMW and its volatile matter (VM) was 16.23 and 2.10 g·L-1,respectively.The results showed that OMW has a very high organic matter concentration (chemical oxygen demand (COD) 36 g·L-1) and high levels in salts (Electric Conductivity EC=10.11 mS·cm-1) (Table 3).

The measurement of physico-chemical parameters of OMW after adsorption on the KC showed a decrease in pH value and in conductivity.However,two parameters suffered reductions,which are phenolic compounds and organic matter.Their amounts were 0.23 and 4.05 g·L-1,respectively which can be explained by the adsorption of a part of them onto KC.

3.3.Effect of operating conditions

The removal efficiency of PC onto clay sample increases from 66% to 77.61% with increasing adsorbent dosage from 0.1 to 1 g[Fig.2(a)].This effect may be related to the availability of more active surface sites for adsorption.

Table 3 OMW characterization before and after treatment with clay

Table 4 Kinetic parameters for the adsorption of phenolic compounds onto KC

pH of OMW is one of the most important parameters,which would affect both phenolic compounds molecules and surface binding sites of clay sample.According to Stasinakiset al.[2],and Papaoikonomouet al.[31],at OMW’s pH lower than the pKa(pKaof phenol=9.89),phenols remain undissociated.While,at pH higher than pKa,phenolic compounds dissociate into anionic forms.In addition,It has been reported that,at pH ＜pHZPC,the surface of KC gets positively charged,which increases the adsorption of the anions through electrostatic forces of attraction.However,at pH ＞pHZPC,the surface of adsorbent particles gets negatively charged,which favors the adsorption of cations.The ZPC for the KC was measured.It was estimated to be about 8.In this work,the adsorption efficiency decreased from 77.61% to 56.12% when the solution pH increased from 2 to 9 and the higher sorption rates were observed at pH 2 with removal efficiency of 77.61%[Fig.2(b)].So,at lower pH,the positively charged functional groups (SiOH+and AlOH+) were being available for the binding of the negatively charged phenolate ions.Wherefore,at higher pH values,it appears that the number of negatively charged sites of clay minerals increases,which must be less favorable to the adsorption of phenolate ions due to electrostatic repulsion.Banatet al.[32],and Djebbar [33],studying adsorption of phenols onto Maghnia clay and onto bentonite,also reported a decrease of adsorption at higher pH values.Contrarily,many researchers report good adsorption rates of phenolic compounds at basic pH [34].

Effect of contact time is a substantial parameter during adsorption.As it can be seen,the adsorption equilibrium of phenolic compounds was obtained after about 10 min with removal efficiency of approximately 77.61% [Fig.2(c)].No remarkable amelioration was observed after longer contact time.Adsorption rate of phenolic compounds onto KC was found to be relatively much faster than those reported for some others adsorbents,such as Pomegranate peel [20],raw and calcined clays [35].The temperature is also the most considerable parameter,which influence the adsorption of phenolic compounds onto clay.To investigate the effect of this parameter,four different temperatures (25 °C,35 °C,45 °C and 55 °C) were used.Increase of the temperature resulted in adsorption yield increase [Fig.2(d)],due to the fact that adsorption is an endothermic process.This increase should have attributed to the rising of adsorptive forces between the active sites of clay and the phenolic compounds.To discover the influence of thermal activation clay,three adsorption assays were released with raw,dried(105°C)and calcined(600°C)clays.The increase in phenolic compounds uptake-levels from 77.61% to 84.21% with the increase of thermal activation can be observed in Fig.2(e).This increase can be justified by the filling of the free sites due to the loss of adsorbed water appeared in the range of 25-200 °C [36],and the loss of constitution water of clay minerals appeared at almost 570 °C [37].

3.4.Desorption study

The high adsorption efficiency of phenolic compounds by clay was obtained in low pH values.Whereas,their high desorption efficiency was 52% (Fig.3) in an alkaline medium.The desorption of phenolic compounds could be justified by the electrostatic repulsion between phenolate ions and SiO-and AlO-on the edges surface of the clay.

3.5.Kinetics adsorption

The kinetic data of phenolic compounds retained were fitted to these models:The pseudo-first-order [38],and pseudo-second order [39],:

Fig.2. Effects of adsorbent amount(a),initial solution pH(b),contact time(c),temperature(d)and thermal activation of clay(e)on the adsorption of phenolic compounds onto KC.

Fig.3. Desorption of phenolic compounds by KC.

whereqt(mg·g-1) andqe(mg·g-1) are the amounts of substrate in OMW adsorbed at any time,t(min) and at equilibrium,respectively;k1is the rate constant of pseudo first order (L·min-1),k2is pseudo second order rate constant (g·g-1·min-1).

Comparison ofR2values indicated that adsorption of phenolic compounds onto KC followed second-order kinetics.Furthermore,the calculatedqe(16.28 mg·g-1) value obtained from Eq.(5) was in agreement with experimentalqe(16.54 mg·g-1)value(Table 4).These two models are empirical;they are not able to determine the adsorption mechanism,which dominate the process.For this reason,intra-particle diffusion was used to examined the limiting controlling step of adsorption [40],

Fig.4. Kinetic plot for the adsorption of phenolic compounds onto KC:Intraparticle diffusion.

whereCis the constant varied with the boundary layer thickness andkidis the intra-particle diffusion rate constant (g·g-1·min-1/2).The quantity of phenolic compounds adsorbed at any time (q)was plottedversusthe square root of time (t1/2) (Fig.4) [41].There is an initial curve followed by a straight line,which indicates that two types of mechanism are operating in the removal of phenolic compounds.The initial curve can be explained by the occupation of most of available sites on the clay surface by PC,while the linear part corresponds to interlayer diffusion.The linear plot does not pass through origin.Thus,the intraparticle diffusion is not the limiting controlling step.The external mass transfer plays an important role in the adsorption of PC.Similar results are reported by Ververiet al[20].

3.6.Thermodynamic studies

Based on van’t Hoff equation (Eq.(6)),enthalpy (ΔH°) and entropy (ΔS°) change were calculated.

Table 5 Thermodynamic parameters for adsorption of phenolic compounds onto KC

Fig.5. Comparative study between clay and others materials.

By plotting lnqe/Ceagainst 1/T,ΔH°and ΔS°were calculated from the slope and the intercept,respectively,of the straight line.The Gibbs free energy change,ΔG°,was given by the relationship.

Thermodynamic parameters were summerized in Table 5.The positive values of ΔH°confirm the endothermicity of the adsorption.Based on ΔG°values,adsorption process was spontaneous.

3.7.Comparative study between KC and others materials

Fig.5 presents the results of activated carbon,iron oxide,lignite and clay to treat OMW.It was noticed that phenolic compounds uptake-level by iron oxide was little higher(87.29%)than raw clay(77.61%)at lower pH.However,lignite produced the lowest reduction in phenolic compounds levels in OMW (71%).The increase of pH values could be attributed partially to the low reduction in the levels of phenolic compounds.Their adsorption efficiencies at pH 9 were 58.42%,56.12% and 9.23% for iron oxide,clay and lignite,respectively.It is very clear from the above results that activated carbon,which it is very expensive materials,could be replaced by other natural materials found in Tunisia.

3.8.Mechanism of adsorption

Fig.6. Adsorption mechanism.

The adsorption mechanism was well described in Fig.6.It is clearly that at pH ＜pHZPC=8,the negatively charged phenolate ions bind through electrostatic attraction to the positively charged functional groups on the clay surface.At pH ＞pHZPC,it seems that the number of negatively charged sites increases,which must be less favorable to the adsorption of phenolate ions due to electrostatic repulsion [42,43].

4.Conclusions

This study was about using adsorption as a technique to decrease the phenolic compounds concentrations in OMW.KC material is considered as an excellent adsorbent.The adsorption occurred by an interaction between the silanol/aluminol groups of the clay with phenolate ions at acidic pH.This interaction increases with increasing the thermal activation of clay.Taking into consideration the removal efficiencies,comparative study between activated carbon and others natural materials available in Tunisia was depicted:Activated carbon ＞ iron oxide ＞clay ＞lignite.

Desorption studies show that negatively charged surface site on the KC favours the desorption of phenolic compounds due to electrostatic repulsion.

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.