Adsorption of Phenolic Compounds from Aqueous Solutions onto Chitosan-Coated Perlite Beads as Biosorbent

Adsorption of Phenolic Compounds from Aqueous Solutions ontoChitosan-Coated Perlite Beads as Biosorbent

N. S. Kumar,†,‡ M. Suguna,† M. V. Subbaiah,† A. S. Reddy,† N. P. Kumar,† and A. Krishnaiah*,†

Biopolymers and Thermophysical Laboratories, Department of Chemistry, Sri Venkateswara UniVersity,Tirupati - 517 502, A.P., India, and Department of Safety EnVironmental System Engineering,Dongguk UniVersity, Gyeongju 780-714, Republic of Korea

Chitosan-coated perlite (CCP) beads were prepared by dropwise addition of a liquid slurry containing chitosanand perlite to an alkaline bath. The resulting beads were characterized using Fourier transform infrared (FTIR)spectroscopy, scanning electron microscopy (SEM), and surface area analysis. The chitosan content of thebeads is 23% as determined by a pyrolysis method. Adsorption of phenolic compounds (phenol, 2-chlorophenol,and 4- chlorophenol) from aqueous solutions on chitosan-coated perlite beads was studied under batchequilibrium and column flow conditions. The binding capacity of the biosorbent was investigated as a functionof initial pH, contact time, initial concentration of adsorbate, and dosage of adsorbent. Adsorption kineticand isotherm studies, respectively, showed that the adsorption process followed a pseudo-first-order kineticmodel and the Langmuir isotherm. The maximum monolayer adsorption capacity of phenol, 2-CP, and 4-CPon to the chitosan-coated perlite beads was found to be 192, 263, and 322 mg g-1, respectively.

1. Introduction

Phenols are generally considered to be one of the importantorganic pollutants discharged into the environment causingunpleasant taste and odor. The major sources of phenol pollutionin the aquatic environment are waste waters from the paint,pesticide, coal conversion, polymeric resin, petroleum, andpetrochemicals industries. Degradation of these substancesproduces phenol and its derivatives in the environment. Thechlorination of natural waters for disinfection produces chlori-nated phenols. Phenols are considered as priority pollutants sincethey are harmful to organisms at low concentrations. Phenolcontents in the drinking water should not exceed 0.002 mg L-1

as per the Indian standard.1 In recent years, interest has beenfocused on the removal of phenols from aqueous solution. A varietyof techniques have been implemented to purify water contaminatedby phenols. Ozonolysis, photolysis, and photocatalytic decomposi-tion have been used with limited success.2 Traditionally, biologicaltreatment, activated carbon adsorption, reverse osmosis, ionexchange, and solvent extraction are the most widely usedtechniques for removing phenols and related organic substances.3-6

Adsorption of phenols onto solid supports such as activated carbonsallows for their removal from water without the addition ofchemicals. Activated carbon exhibits good adsorption ability formany organic pollutants but is expensive due to its difficultregeneration and high disposal cost.7

In recent years, polymeric adsorbents have been used increas-ingly as an alternative to activated carbon due to their economicfeasibility, adsorption-regeneration properties and mechanicalstrength. Chitosan is a polysaccharide prepared by the de-N-acetylation of chitin, which makes up shells and shrimps.8 Dueto the primary, secondary hydroxyl groups and highly reactiveamino groups of chitosan as well as the property of nontoxicityand biodegradability, it has been regarded as a useful materialto remove inorganic and organic substances from wastewater.9

However, several investigators have attempted to modify

chitosan to facilitate mass transfer and to expose the activebinding sites to enhance the adsorption capacity. Graftingspecific functional groups onto a native chitosan backboneallows its sorption properties to be enhanced.10 Many applica-tions are due to the secondary amino groups of chitosan whichshow polycationic, chelating, and film-forming properties alongwith high solubility in dilute acids. Chitosan has already beendescribed as a suitable natural polymer for the collection ofphenolic compounds, through chelation, due to the presence ofan amino and hydroxyl groups on the glucosamine unit.11 Inmost of the studies chitosan has been used in the form of flakes,powder, or hydrogel beads. Enzymatic removal of variousphenol compounds from a synthetic water sample was studiedby the use of mushroom tyrosinase and chitosan beads as afunction of pH, temperature, tyrosinase dose, and the hydrogenperoxide-to-substrate ratio.12 The adsorption of 4-nonylphenolethoxylates (NPEs) onto chitosan beads having cyclodextrin wasinvestigated by Aoki et al.13 Adsorption of phenol onto chitosan-coated bentonite was studied by Cheng et al.14 Biosorption ofphenol and o-chlorophenol from aqueous solutions ontochitosan-calcium alginate blended beads was reported by SivaKumar et al.15 Removal of chlorophenols from groundwater bychitosan sorption was studied by Zheng et al.16 Adsorption ofphenol, p-chlorophenol, and p-nitrophenol onto functionalchitosan was studied by Jian-Mei et al.17 Biosorption of phenoliccompounds from aqueous solutions onto chitosan-abrus preca-torius blended beads was studied by Siva Kumar et al.18 Themaximum adsorption capacities of different adsorbents obtainedfrom different sources are included in Table 1 along with thevalues obtained in the present study.

In this study a new composite chitosan biosorbent is preparedby coating chitosan, a glucosamine biopolymer, over perlite,an inorganic porous aluminosilicate and formed into beads.Perlite is a siliceous volcanic glassy rock with an amorphousstructure. It is expected that the more active sites of chitosanwill be available due to the coating thus enhancing theadsorption capacity. The percent of chitosan coated on perlitewas determined by pyrolysis technique. Surface area, porevolume, and pore diameter were obtained on the basis ofBrunauer, Emmet, and Teller (BET) measurements. The chito-

* To whom correspondence should be addressed. Tel.: +91-9393621986. E-mail address: [email protected].

† Sri Venkateswara University.‡ Dongguk University.

Ind. Eng. Chem. Res. 2010, 49, 9238–92479238

10.1021/ie901171b 2010 American Chemical SocietyPublished on Web 08/26/2010

san-coated perlite (CCP) beads were characterized before and afteradsorption of phenolic compounds by Fourier transform infrared(FTIR) spectroscopy, scanning electron microscopy (SEM), andsurface area analysis. In the present investigation, equilibrium anddynamic column adsorption characteristics of phenolic compoundson chitosan-coated perlite beads were studied. The column flowdata were used to generate break through curves. The loadedadsorbent with phenolic compounds was regenerated by solventelution method using 0.1 M NaOH as eluent.

2. Materials and Methods

2.1. Materials. Perlite is not a trade name but a generic termfor naturally occurring siliceous rock. The distinguishing feature,which sets perlite apart from other volcanic glasses, is that, whenheated to a suitable point in its softening range, it expands fromfour to twenty times of its original volume. This expansionprocess also creates one of perlite’s most distinguishingcharacteristicssits white colorswhile the crude rock may rangefrom transparent light gray to glassy black. The expanded formof perlite is obtained from Silbrico Corporations, IL, USA, andwas used as a substrate for the preparation of beads. Chitosan(molecular weight 1-3 lakhs), 99% pure oxalic acid dihydrate,and NaOH beads (95-100%) were purchased from FisherScientific Company. All the working solutions are obtained bydiluting the stock solution with double distilled water. Phenol(Ranbaxy, India, A.R. grade), 2-chlorophenol, and 4-chlorophe-nol (Spectrochem, India, A.R. grade) were used without furtherpurification. Stock solutions were prepared by dissolving 1.0 gof phenol, 2-chlorophenol, and 4-chlorophenol individually in1 L of double distilled water. These stock solutions were usedto prepare 100, 200, 300, and 400 mg L-1 solutions of phenol,2-chlorophenol, and 4-chlorophenol. Water used for preparationof solutions and cleaning adsorbents was generated in thelaboratory by double distilling the deionized water in a quartzdistillation unit.

2.2. Preparation of Chitosan-Coated Perlite Beads. Perlite,which is composed mainly of alumina and silica, is used as asubstrate for the preparation of beads.19 Perlite is first mixedwith 0.2 M oxalic acid, and the mixture is stirred for 12 h atroom temperature (30 °C) and filtered. The filtered perlite waswashed with deionized water and dried overnight at 70 °C andsieved through 100-mesh size. The acid-treated perlite wasstored in desiccator. About 30 g of medium molecular weightchitosan was slowly added to 1 L of 0.2 M oxalic acid solutionunder continuous stirring at 40-50 °C to facilitate the formationof a viscous gel. About 60 g of acid-treated perlite powder wasmixed with deionized water and slowly added to the diluted geland stirred for 12 h at 40-50 °C. The highly porous beads werethen prepared by dropwise addition of the perlite gel mixture intoa 0.7 M NaOH precipitation bath.20 The purpose of adding theacidic perlite-chitosan mixture to NaOH solution was to assistrapid neutralization of oxalic acid, so that the spherical shape couldbe retained. The beads were separated from the NaOH bath andwashed several times with deionized water to a neutral pH. Thebeads were dried in a freeze drier, oven, and by air.

2.3. Analysis of Phenolic Compounds. The concentrationof phenol, 2-chlorophenol, and 4- chlorophenol, in aqueousmedium, was determined by measuring absorbance at wave-lengths of 270, 274, and 280 nm, respectively, using a UV-spectrophotometer (model Shimadzu UV-2450). In order toreduce measurement errors in all the experiments, the UVabsorption intensity of each solution sample was measured intriplicates and the average value was used to calculate theequilibrium concentration based on standard calibration curve,whose correlation coefficient square (R2) was 0.999. Theexperimental error was observed to be within (2%.

2.4. Batch Studies. In order to study the effect of differentcontrolling parameters, such as solution pH, contact time,quantity of adsorbent, and the initial concentration of adsorbate,the experiments were conducted by varying one parameter at atime keeping all other parameters constant. The solution pHwas adjusted by adding 0.1 M HCl or 0.1 M NaOH solutions.Batch adsorption experiments were conducted in 125 mLErlenmeyer flasks with a specified amount (0.1 g) of adsorbentin contact with 100 mL of phenolic solutions of desiredconcentration at a desired pH. The contents of the flasks wereshaken at 175 rpm on a mechanical shaker at room temperature.It was confirmed through the preliminary experiments that 180min is sufficient to attain equilibrium between adsorbent andadsorbate. The samples were filtered through Whatman no. 5filter paper (2.5 µm size particle retention) to eliminate any fineparticules. The sorption capacity of the biosorbent was deter-mined by material balance of the initial and equilibriumconcentrations of the solution. Adsorption on the glassware wasfound to be negligible and was determined by running blankexperiments. The amount of solute adsorbed per unit mass ofadsorbent was calculated using the following equation:

where qe is the adsorption capacity (mg g-1) at equilibrium; C0

and Ce are the initial and equilibrium concentrations of solute(mg L-1), respectively; V is the volume of the aqueous solution(L); and W is the mass (g) of adsorbent.

2.5. Column Adsorption Studies. Dynamic flow adsorptionstudies were carried out in a column made of Pyrex glass of1.2 cm internal diameter and 30 cm length. The column is filledwith 1 g of chitosan-coated perlite beads by tapping so that thecolumn is filled without gaps. The column was fully jacketed

Table 1. Maximum Capacity, Q0 (mg g-1), for Adsorption ofPhenolic Compounds by Various Adsorbents

adsorbates, Q0 (mg g-1)

sorbent phenol 2-CP 4-CP refs

CS/CA blended beads 109 97 15functional chitosan 17CS 2.22 2.58CS-SA 8.50 20.49CS-CD 34.93 179.73EPI-CD 131.50 74.25CS/Ab blended beads 156 204 278 18olive stone- based activated

aarbon189 436 24

sugar cane bagasse fly ash 23.83 26ratan sawdust based

activated carbon149.25 27

granular activated carbon 165.80 36commercial activated

carbon49.72 37

activated bentonites 38M-bentonite 9.9Al-bentonite 8.7CTAB-bentonite 8.4T-bentonite 8.2CS850A 205.8 39fly ash 3.85 40activated carbon 380.2 422.1 41modified bentonite 176.6 42modified starch 43perlite 5.84 44chitosan-coated perlite

beads192 263 322 present

study

qe ) (C0 - Ce) × VW

(1)

Ind. Eng. Chem. Res., Vol. 49, No. 19, 2010 9239

to circulate water from a constant temperature water bath,enabling the experiment to be carried out at a constanttemperature. The adsorbent was washed thoroughly with waterand dried prior to use. The influent solution of known concentrationof aqueous solution of phenolic compounds was allowed to passthrough the bed at a constant flow rate (1 mL min-1) in a downflowmanner. The complete cycle of operation of each column experi-ment includes three steps: pH precondition of the adsorbate solution,solution flow, and adsorption of solute until column exhaustionoccurs. The effluent solution was collected at different time intervalsand concentrations of phenolic compounds were determined bymeasuring absorbance using Shimadzu UV-2450 spectrophotom-eter. The solutions were diluted appropriately prior to analysis. Allexperiments were carried out at room temperature. Breakthroughcurves were obtained by plotting volume of the solution passedthrough the column vs ratio of the column outlet concentration tothe initial concentration, C0/Ci.

2.6. Desorption Studies. Desorption (recovery) studies werevery important since the success of adsorption process dependson the regeneration of adsorbent. After the column wascompletely exhausted, the remaining aqueous solution in thecolumn was drained off by pumping air through the column.Desorption of phenolic compounds (phenol, 2-CP, and 4-CP)was tried with a number of eluents, and it was found that thedesorption occurred by sodium hydroxide easily. Desorption ofsolutes from the loaded adsorbent were carried out by a solventelution method using 0.1 M NaOH as an eluent maintained atconstant temperature at a fixed flow rate (1 mL min-1). Theeffluent samples at different time intervals were collected at thebottom of the column for analysis. When the concentration ofthe outlet solution was zero or close to zero, it was assumedthat the column is regenerated. After the regeneration, theadsorbent column was washed with distilled water to removeNaOH from the column before the influent adsorbate solutionwas reintroduced for the subsequent adsorption-desorptioncycles. The adsorption-desorption cycles were performed thricefor each phenolic solution using the same bed to check thesustainability of the bed for repeated use. Regeneration curveswere obtained by plotting volume of the solution passed throughthe column vs concentration of the column outlet solution.

3. Results and Discussion

3.1. Characterization of Chitosan-Coated Perlite (CCP)Beads. 3.1.1. Pyrolysis Studies. The amount of chitosan-coatedperlite is obtained by measuring the weight loss of biosorbentfrom pyrolysis. These experiments are conducted at hightemperature (800 °C) to determine the amount of chitosan coatedover perlite. Two ceramic crucibles, one containing acid washedpure perlite and the other containing chitosan-coated perlite(CCP) beads, are placed inside a furnace heated to 800 °C. Thechitosan burnt out at this temperature, and the chitosan contentis determined from weight difference. The results indicated that23% of chitosan is coated over perlite.21

3.2. Fourier Transform Infrared (FTIR) Studies. FTIRspectra of CCP in virgin form and loaded with phenoliccompounds, obtained using a Nicolet-740, Perkin-Elmer model283B (USA). The FTIR spectra of the chitosan-coated perlitebeads before and after adsorption are shown in Figure 1a-d,respectively. The FTIR spectrum of CCP beads before adsorp-tion (Figure 1a) shows a broad absorption peak at 3435 cm-1

corresponding to the overlapping of -OH and -NH peaks. Apeak at 2926 cm-1 represents the C-H group. A peak appearingat 1649 cm-1 is due to the bending mode of N-H in primaryamines. A significant difference can be seen in the FTIR spectra

of biosorbent before and after adsorption. As we observe inFigure 1b-d, some peaks are shifted and/or broadened indicat-ing that the functional groups present on the biosorbent isinvolved in interaction with the phenolic compounds. Theseresults confirm the participation of amino, carboxylic, andhydroxyl groups of CCP beads as potential active binding sitesfor adsorption of phenolic compounds.

3.3. Surface Area Analysis. Surface area, density, porevolume, pore diameter, and porosity of the composite biosorbentwere determined with a BET (Brunauer, Emmett and Teller)instrument (model no: Micromeritirics, USA). Surface area wasmeasured by assuming that the adsorbed nitrogen forms amonolayer and possess a molecular cross sectional area of 16.2Å2 /molecule. The isotherm plots were used to calculate thespecific surface area (N2/BET method) and average porediameter of CCP, while micropore volume was calculated fromthe volume of nitrogen adsorbed at P/Po ) 1.4. The sorbentmaterial shows an average surface area of 112.25 m2 g-1, porevolume of 0.47 cm3 g-1, porosity of 43.41%, pore diameter of0.97 nm, and density of 3.13 g cm-3.

3.4. Scanning Electron Microscopic (SEM) Studies. Scan-ning electron micrographs of CCP beads recorded, using asoftware controlled digital scanning electron microscope-JEOLJSM 5410 (Eucentric Gonimeter state type) Japan, aregiven in Figure 2. The SEM micrograph of the pure perlitepowder, outer surface, and cross section of CCP beads are shownin Figure 2a-c, respectively. The figure also illustrates thesurface texture and porosity of CCP beads with holes and smallopenings on the surface, thereby increasing the contact area,which facilitates the pore diffusion during adsorption. Theporous nature is clearly evident from this micrograph. The innersurface appears to have similar type of texture and morphologyas the outer surface. The surface morphology of the pure perliteappears to change significantly following coating with chitosan.

3.5. Effect of pH. The pH of aqueous medium is animportant factor that may influence the uptake of the adsorbate.The pH of the solution affects the degree of ionization andspeciation of various pollutants which subsequently leads to achange in reaction kinetics and equilibrium characteristics ofthe adsorption process. In order to optimize the pH for maximumremoval efficiency, experiments were conducted in the pH rangefrom 4.0 to 10.0 using 0.1 g of CCP beads with 100 mL of 100mg L-1 adsorbate solutions at room temperature. In the alkalinerange, the pH was varied using aqueous NaOH, whereas in theacidic range, pH was varied using HCl. Results are shown inFigure 3. The adsorption capacities increase with increase inpH up to pH 7 and decreases from there. The interaction forcesbetween phenol or chlorophenols and biomass are rather weakin the acidic solutions. The highest adsorption of phenoliccompounds occurred at pH 7.0 for all species. The decrease inthe sorption capacity of the biosorbent toward the chlorophenolsas the pH increases above 7 can be explained as being the resultof increased electrostatic repulsion between the sorbate andsorbent, since both are negatively charged over this pH range.The phenolic compounds considered in this study, viz. phenol,2-CP, and 4-CP, have pKa values of 9.9, 8.3, and 9.2,respectively. When the pH of a solution goes beyond the pKa,phenols chiefly exist as negative phenolate ions, whereas theyexist as neutral molecules below the pKa. Due to the electronrich nature of the oxygen atom in phenolate ions, the hydrogenbonding efficiency decreases. Therefore, phenols effectivelyadsorbed on to the adsorbent as molecules but not phenolateions. Thus, it may be concluded that the molecular interactionsinvolved in the adsorption process are through hydrogen bonding

9240 Ind. Eng. Chem. Res., Vol. 49, No. 19, 2010

and van-der Waals forces. A similar behavior for phenols, ingeneral, has been reported by Gupta et al.,22 Zogoroski et al.,23

Termoul et al.,24 Mangrulkara et al.,25 Srivastava et al.,26 andHameed et al.27

3.6. Effect of Agitation Time. The effect of agitation timeon the extent of adsorption of phenolic compounds at differentconcentrations is shown in Figures 4, 5, and 6 for phenol, 2-CP,and 4-CP, respectively. The extent of adsorption increases withtime and attained equilibrium for all the concentrations of phenol,2-CP, and 4-CP studied (100, 200, 300, and 400 mg L-1) at 180min. After this equilibrium period, the amount of solute adsorbeddid not change significantly with time, indicating that this time issufficient to attain equilibrium for the maximum removal ofphenolic compounds from aqueous solutions by CCP beads. Theadsorption capacity for chlorophenols is higher than that of phenol,possibly due to the higher solubility of phenol in water.

3.7. Batch Adsorption Kinetic Modeling. Kinetic modelsare used to determine the rate of adsorption process. Data onremoval of phenol, 2-CP, and 4-CP by CCP beads as a functionof time at pH 7.0 at various initial concentrations (100-400mg L-1) are presented graphically in Figures 7-9. Both pseudo-first-order and pseudo-second-order kinetic models were usedto correlate the adsorption data28,29

Where, qt is the amount adsorbed at time t, and k1 and k2 arethe kinetic parameters to be determined. The slope and intercept

Figure 1. FTIR spectra of phenolic compounds: (a) before adsorption of chitosan-coated perlite beads; (b) after phenol adsorption; (c) after 2-chlorophenoladsorption; (d) after 4-chlorophenol adsorption.

log(qe - qt) ) log qe -k1t

2.303(2)

tqt

) 1

k2qe2+ ( 1

qe)t (3)

Ind. Eng. Chem. Res., Vol. 49, No. 19, 2010 9241

of plot of log(qe - qt) versus t were used to determine the first-order rate constant k1. The slope and intercept of the plot of t/qt

versus t were used to calculate the second-order rate constant

k2. It is more likely to predict behavior over the whole range ofadsorption and is in agreement with the chemisorption mech-anism being the rate controlling step. The kinetic parametersare included (Table 2). Higher correlation coefficients of thepseudo-second-order model and agreement between experimen-tal and calculated qe values indicate that the adsorption ofphenol, 2-CP, and 4-CP on to CCP beads follows second-orderkinetics.

3.8. Intraparticle Diffusion Model. The intraparticle dif-fusion model is used to investigate the diffusion controlledadsorption system. A process is diffusion controlled if its rateis dependent upon the rate at which components diffuse towardeach another.30 The intraparticle diffusion equation is expressedin the following form.

Figure 2. SEM of (a) pure perlite powder, (b) outer surface of chitosan-coated perlite bead, and (c) cross section of chitosan-coated perlite bead.

Figure 3. Effect of pH on the adsorption of ([) phenol, (9) 2-chlorophenol,and (2) 4-chlorophenol onto chitosan-coated perlite beads.

Figure 4. Effect of agitation time on adsorption of phenol on chitosan-coated perlite beads at different initial concentrations.

Figure 5. Effect of agitation time on adsorption of 2-chlorophenol onchitosan-coated perlite beads at different initial concentrations.

Figure 6. Effect of agitation time on adsorption of 4-chlorophenol onchitosan-coated perlite beads at different initial concentrations.

qt ) Kidt0.5 + C (4)

9242 Ind. Eng. Chem. Res., Vol. 49, No. 19, 2010

where Kid (mg g-1 min-1/2) is the rate constant of intraparticlediffusion. C is the value of the intercept of qt versus t0.5, whichgives an idea about the boundary layer thickness, i.e. the largerthe intercept, the greater the boundary layer effect. The plotsare not linear over the whole time range, indicating that morethan one step affecting the adsorption of phenolic compounds.

For example, the first step might be due to the boundary layerdiffusion at the initial stage of the adsorption and the intraparti-cle diffusion which gives the other two linear parts. Theintraparticle diffusion starts with a rapid transport of adsorbatemolecules in to macropores and wider mesopores and then,penetrating the smaller meso- and micropores at a much slowerpace. Therefore, the second portion of linear curve attributes tothe gradual adsorption, where intraparticle diffusion is a ratelimiting. The third portion refers to the final equilibrium stagesignified by a formation of plateau, indicating a weak activityof the intraparticle diffusion due to low adsorbate concentrationleft in the solution.31 On the other hand, most of the adsorptionsites have been occupied after lapse of time thus limited freesites for the adsorbate molecules to attach on. If the intraparticlediffusion is the only rate-controlling step, then the plot passesthrough the origin; otherwise, the boundary layer diffusionaffects the adsorption to some degree.32

4. Effect of Adsorbent Dose

One of the parameters that strongly affect the biosorptioncapacity is the amount of the biosorbent. The effect of adsorbentdose on the uptake of phenol, 2-CP, and 4-CP on CCP beadswas studied and is shown in Figures 7, 8, and 9, respectively.It can be seen from the figures, that percentage removal ofphenol, 2-CP, and 4-CP increases with the increase in adsorbentdose while the adsorption capacity at equilibrium, qe (mg g-1),decreases. The latter result can be explained as a consequenceof partial aggregation, which occurs at high biomass concentra-tion giving rise a decrease of active sites.26 It is apparent thatthe percent removal of phenol, 2-CP, and 4-CP increases rapidlywith increase in the dose of CCP beads due to the availabilityof greater amount of active sites of adsorbent. It can also beseen from these figures that the uptake of solute markedlyincreased up to adsorbent dose of 0.5 g and thereafter nosignificant increase was observed. Adsorption is maximum with0.5 g of CCP beads and the maximum percent removal is about91% for phenol, about 95% for 2-CP, and about 98% for 4-CP.

5. Adsorption Isotherm Models

The experimental data on adsorption were analyzed usingLangmuir, Freundlich, and Dubinin-Radushkevich (DR) iso-therm models. The Langmuir isotherm model assumes uniformenergies of adsorption onto the surface with no transmigrationof adsorbate in the plane of the surface. The linear form of theLangmuir isotherm is given by the following equation.22

where qe is the amount adsorbed (mg g-1), Ce is the equilibriumconcentration of the adsorbate (mg L-1), and Q0 and b are theLangmuir constants related to maximum adsorption capacityand energy of adsorption, respectively. The values of theparameters were obtained from the plots of 1/qe versus 1/Ce.Langmuir parameter, b, can be used to predict the affinitybetween the sorbate and adsorbent using the dimensionlessseparation factor, RL, and defined by6

If the value of RL is equal to zero or one, the adsorption iseither linear or irreversible, and if the value is in between zeroand one, adsorption is favorable to chemisorption. The value

Figure 7. Effect of adsorbent dose (percent removal of phenol andadsorption capacity (mg g-1)) for adsorption of phenol on chitosan-coatedperlite beads.

Figure 8. Effect of adsorbent dose (percent removal of 2-chlorophenol andadsorption capacity (mg g-1)) for adsorption of phenol on chitosan-coatedperlite beads.

Figure 9. Effect of adsorbent dose (percent removal of 4-chlorophenol andadsorption capacity (mg g-1)) for adsorption of phenol on chitosan-coatedperlite beads.

1qe

) 1Q0

+ 1bQ0Ce

(5)

RL ) 11 + bC0

(6)

Ind. Eng. Chem. Res., Vol. 49, No. 19, 2010 9243

of RL is less than 1 and great than 0, suggesting the favorableuptake of phenolic compounds by chitosan-coated perlitebeads.33

The Freundlich isotherm is given as

where KF ((mg g-1)(L mg-1)1/n) and 1/n are indicators of theadsorption capacity and the adsorption intensity, respectively.The values of KF and 1/n were calculated by plotting ln(qe)against ln(Ce). The values of Langmuir constants Q0 and b andFreundlich parameters KF and n along with the correlationcoefficients (R2) are presented in Table 3. Among these twomodels, the Langmuir isotherm gives a better representation ofadsorption of phenol, 2-CP, and 4-CP on chitosan-coated perlitebeads compared to the Freundlich model.

Another model for the analysis of isotherms of a high degreeof rectangularity is the Dubinin-Radushkevich isotherm. Theequilibrium data were also correlated with the DR model todetermine if adsorption occurred by physical or chemicalprocesses. In the case of liquid phase adsorption, several studieshave shown that the adsorption energy can be estimatedaccording to the Dubinin-Radushkevich equation. Assumingthat the adsorption in micropores is limited to a monolayer andthe DR equation is applicable, the adsorption capacity per unitsurface area of the adsorbent at equilibrium, qe, can be writtenas34

where ε can be correlated by

where qs is the ultimate capacity per unit area of the adsorbentand the constant B gives the mean free energy E of adsorptionper molecule of the adsorbate when it is transferred to the surface

of the solid from infinity in the solution and can be computedby using the relationship:

where R is the gas constant (8.314 J mol-1 K-1) and T is theabsolute temperature. A plot of ln(qe) versus ε2 enables theconstants qs and E to be determined (Table 3). It is knownthat Freundlich and Langmuir isotherms could not reveal theadsorption mechanism. The purpose of applying equilibriumdata to the DR model is mainly to clarify the adsorption typeand evaluate the nature of interaction between sorbate andsolid. On the basis of the theory of the DR model, the sorptionspace in the vicinity of the solid surface is characterized bya series of equipotential surfaces having same sorptionpotential. The sorption mean free energy is the energyrequired to transfer 1 mol of the sorbate from infinity insolution to the surface of the solid. The magnitude of sorptionmean free energy E is widely used for estimating the type ofadsorption.35 The mean adsorption energy (E) was found tobe 10.20 kJ mol-1 for phenol, 11.62 kJ mol-1 for 2-CP, and13.36 kJ mol-1 for 4-CP onto chitosan-coated perlite beads.These values indicate that the adsorption processes isassociated with chemical ion-exchange mechanism. Theparameters of the three isotherms were computed and listedin Table 3. The Langmuir isotherm fitted quite well with theexperimental data (correlation coefficient R2 ) 0.99), com-pared to the other two isotherm models.

6. Column Adsorption Data

To be useful in separation and removal processes, adsorbedspecies should be easily desorbed under mild conditions andadsorbents should be reused many times in order to decreasematerial costs. The results of dynamic flow experiments wereused to obtain the breakthrough curves for adsorption ofphenol, 2-CP, and 4-CP from aqueous solutions by plotting

Table 2. Adsorption Rate Constants of Phenolic Compounds on Chitosan-Coated Perlite Beads

first-order kinetic model second-order kinetic model Weber-Morris

(mg L-1) q(e,exp) (mg g-1) q(e,cal) (mg g-1) k1 (min-1) R2 q(e,cal) (mg g-1) k2 (g mg-1 min-1) R2 Kid C R2

phenol

100 74.6 53.8 0.0085 0.993 75.1 3.06 × 10-4 0.992 3.961 11.49 0.994200 94.3 62.5 0.0080 0.997 91.7 3.05 × 10-4 0.996 4.604 19.87 0.993300 117 84.7 0.0085 0.995 119 1.90 × 10-4 0.993 6.258 17.47 0.994400 139.2 100 0.0082 0.997 140 1.59 × 10-4 0.994 7.479 20.06 0.995

2-chlorophenol

100 85.3 56.5 0.0085 0.990 84.03 3.29 × 10-4 0.998 4.275 17.26 0.984200 109.4 81.7 0.0085 0.991 112.3 1.77 × 10-4 0.997 6.130 11.35 0.983300 144.3 103.8 0.0087 0.993 149.2 1.49 × 10-4 0.998 7.974 20.15 0.984400 192.1 137.4 0.0089 0.991 196.1 1.22 × 10-4 0.982 9.939 35.52 0.992

4-chlorophenol

100 92.3 63.9 0.0096 0.980 93.4 3.11 × 10-4 0.998 4.780 20.17 0.989200 134.6 96.4 0.0087 0.995 136.9 1.70 × 10-4 0.996 7.304 20.55 0.990300 167.9 123.8 0.0078 0.988 169.4 1.21 × 10-4 0.999 9.266 17.98 0.982400 202.7 153.3 0.0078 0.993 200 1.09 × 10-4 0.997 11.37 18.56 0.990

Table 3. Isotherm Parameters of Langmuir, Freundlich, and DR Isotherms for Adsorption of Phenol, 2-Chlorophenol, and 4-Chlorophenol onChitosan-Coated Perlite Beads

Langmuir constants Freundlich constants Dubinin-Raduskvich isotherm

adsorbates Q0 b R2 KF n R2 B qs E R2

phenol 192 0.004 0.997 1.068 1.097 0.987 0.0042 3.394 10.20 0.83592-chlorophenol 263 0.003 0.993 1.437 1.037 0.997 0.0037 3.276 11.62 0.78954-chlorophenol 322 0.005 0.999 1.825 1.078 0.993 0.0028 3.512 13.36 0.8123

qe ) KFCe1/n (7)

qe ) qs exp(-Bε2) (8)

ε ) RT ln[1 + 1Ce] (9)

E ) 1

√2B(10)

9244 Ind. Eng. Chem. Res., Vol. 49, No. 19, 2010

the volume of effluent vs Ce/Ci. The breakthrough curves areshown in Figures 10-12. Breakthrough capacities, theamount adsorbed until the effluent concentration of theadsorbate is equal to the influent solution concentration, arecomputed from the breakthrough curves. An examination ofthe curves indicates that no leakage of solute is observed up

to a volume of about 100 mL of influent sorbate solution inall cases in the first cycle.

When the bed is exhausted or the effluent coming out of thecolumn reaches the allowable maximum discharge level, theregeneration of the adsorption bed to recover the adsorbedmaterial and/or to regenerate the adsorbent becomes quiteessential. The regeneration could be accomplished by a varietyof techniques such as thermal desorption, steam washing, solventextraction, etc. Each method has inherent advantages andlimitations. In this study, several solvents were tried to regener-ate the adsorption bed. A 0.1 M NaOH solution is found to beeffective in desorbing and recovering adsorbates quantitativelyfrom the adsorption bed. The fixed bed columns of CCP beadssaturated with phenol or chlorophenol is regenerated by passing0.1 M NaOH solution as an eluent at a fixed flow rate of 1 mLmin-1. To evaluate the sorbate recovery efficiency, the percentof phenol, 2-CP, and 4-CP recovered is calculated from thebreakthrough and recovery curves. The desorption profile isgraphically represented in Figures 13-15. From the plots, it isobserved that the rate of desorption increases sharply reachinga maximum with 4 mL of 0.1 M NaOH solution and completeregeneration occurred at about 30 mL. The regenerated columnis further used for the removal of phenol. The results indicatethat the column gets saturated early and adsorption capacitydecreases. As a result, the percent desorption also decreases fromthe first to the third cycle.

Figure 10. Column breakthrough curves for adsorption of phenol onchitosan-coated perlite beads.

Figure 11. Column breakthrough curves for adsorption of 2-chlorophenolon chitosan-coated perlite beads.

Figure 12. Column breakthrough curves for adsorption of 4-chlorophenolon chitosan-coated perlite beads.

Figure 13. Regeneration curves of chitosan-coated perlite beads loaded withphenol.

Figure 14. Regeneration curves of chitosan-coated perlite beads loaded with2-chlorophenol.

Ind. Eng. Chem. Res., Vol. 49, No. 19, 2010 9245

7. Conclusions

Chitosan is effectively coated on an inert substrate, perlite,and is made in the form of spherical beads. Pyrolysis resultsindicated that 23% of chitosan was coated on perlite. Scanningelectron micrographs showed the beads to be porous in nature.The effect of pH on the extent of adsorption was investigated.The adsorption process was found to follow the second-orderkinetic model. The Langmuir isotherm model represents theexperimental data adequately compared to Freundlich and DRisotherm models. The equilibrium adsorption data show thatchitosan-coated beads adsorb a significant amount of phenoliccompounds compared to chitosan or other chemically modifiedchitosan as reported in the literature. Sodium hydroxide (0.1M) was found to be effective in regenerating the column loadedwith phenolic compounds. The present work elucidates that theCCP beads are potential biosorbents for their application in theremoval of phenol and its derivatives. The results indicate thatCCP beads show higher adsorption capacity for chlorophenolsthan phenol. This could be attributed to the difference insolubility and hydrophobicity of phenol, 2-CP, and 4-CP inwater.

Literature Cited

(1) Ahmaruzzaman, Md. Adsorption of phenolic compounds on low-cost adsorbents: A review. AdV. Colloid Interface Sci. 2008, 143, 48.

(2) Hu, M. Q.; Xu, Y. M.; Zhao, J. C. Efficient photosensitizeddegradation of 4-chlorophenol over immobilized aluminium tetrasulfoph-thalocyanine in the presence of hydrogen peroxide. Langmuir 2004, 20,6302.

(3) Tepe, O.; Dursun, A. Y. Combined effects of external mass transferand biodegradation rates on removal of phenol by immobilized Ralstoniaeutropha in a packed bed reactor. J. Hazard. Mater. 2008, 151, 9.

(4) Dabrowski, A.; Podkoscielny, P.; Hubicki, M.; Barczak, M. Adsorp-tion of phenolic compounds by activated carbonsa critical review.Chemosphere 2005, 58, 1049.

(5) Patterson, J. F. Industrial Waste water Treatment Technology, 2nded.; Butterworths: London, 1985.

(6) Dursun, G.; Cecek, H.; Dursun, A. Y. Adsorption of phenol fromaqueous solution by using carbonised beet pulp. J. Hazard. Mater. 2005,125, 175.

(7) Maugans, C. B.; Akgerman, A. Catalytic wet oxidation of phenol ina trickle bed reactor over a pt/Tio2 catalyst. Water Res. 2003, 37, 319.

(8) Chao, A. C.; Shyu, S. S.; Lin, Y. C.; Mi, F. L. Enzymatic graftingof carboxyl groups on to chitosan-to confer on chitosan the property of acationic dye adsorbent. Biores. Technol. 2004, 91, 157.

(9) Crini, G. Non-Conventional low-cost adsorbents for dye removal: areview. Biores. Technol. 2006, 97, 1061.

(10) Juang, R. S.; Ju, C. Y. Kinetics of sorption of Cu (II)- ethylene-diaminetetraacetic acid chelated anions on cross-linked, polyaminatedchitosan beads. Ind. Eng. Chem. Res. 1998, 32, 386.

(11) Rhee, J. S.; Jung, M. W.; Paeng, K. J. Evaluation of chitin andchitosan as a sorbent for the preconcentration of phenol and chlorophenolsin water. Anal. Sci. 1998, 14, 1089.

(12) Yamada, K.; Akiba, Y.; Shibuya, T.; Kashiwada, A.; Matsuda, K.;Hirata, M. Water purification through bioconversion of phenol compoundsby tyrosinase and chemical adsorption by chitosan beads. Biotechnol. Prog.2005, 21, 823.

(13) Aoki, N.; Kinoshita, K.; Mikuni, K.; Nakanishi, K.; Hattori, K.Adsorption of 4-nonylphenol ethoxylates onto insoluble chitosan beadsbearing cyclodextrin moieties. J. Incl. Phenom. Macrocycl. Chem. 2007,57, 237.

(14) Cheng, T. S.; Xu, W. Q.; Yan, J. Study on the adsorption capacityof phenol by chitosan coated bentonite. Chemical 2006, 20, 1.

(15) Siva Kumar, N.; Kalyani, S.; Boddu, V. M.; Krishnaiah, A.Biosorption of phenol and o-chlorophenol from aqueous solutions on tochitosan calcium alginate blended beads. J. Hazard. Mater. 2008, 162, 482.

(16) Zheng, S.; Yang, Z.; Jo, D. H.; Park, Y. H. Removal of chlorophe-nols from ground water by chitosan sorption. Water Res. 2004, 38, 2315.

(17) Li, J.-M.; Meng, X.-G.; Hu, C.-W.; Du, J. Adsorption of phenol,p-chlorophenol and p-nitrophenol on to functional chitosan. Biores. Technol.2009, 100, 1168.

(18) Siva Kumar, N.; Venkata Subbaiah, M.; Subba Reddy, A.;Krishnaiah, A. Biosorption of phenolic compounds from aqueous solutionson to chitosan-abrus precatorius blended beads. J. Chem. Technol. Bio-technol. 2009, 84, 972.

(19) Hasan, S.; Krishnaiah, A.; Gosh, T. K.; Boddu, V. M.; Smith, E. D.Chromium(VI) adsorption capacity of chitosan coated on an inert substrateperlite. Sep. Sci. Technol. 2003, 38, 3375.

(20) Kawamura, Y.; Yoshida, H.; Asai, S.; Kurahashi, I.; Tanibe, H.Effects of chitosan concentration and precipitation bath concentration onthe material properties of porous cross linked chitosan beads. Sep. Sci.Technol. 1997, 32, 1959.

(21) Kalyani, S.; Krishnaiah, A.; Boddu, V. M. Adsorption of divalentcobalt from aqueous solution onto chitosan-coated perlite beads as biosor-bent. Sep. Sci. Technol. 2007, 32, 2767.

(22) Gupta, V. K.; Ali, I.; Saini, V. K. Removal of chlorophenols fromwastewater using red mud: An aluminum industry waste. EnViron. Sci.Technol. 2004, 38, 4012.

(23) Zogoroski, J. S.; Faust, S. D.; Haas, J. H., Jr. The kinetics ofadsorption of phenols by granular activated carbon. J. Colloid InterfaceSci. 1976, 55, 329.

(24) Termoul, M.; Bestani, B.; Benderdouche, N.; Belhakem, M.;Naffrechoux, E. Removal of Phenol and 4-Chlorophenol from AqueousSolutions by Olive Stone-based Activated Carbon. Adsorp. Sci. Technol.2006, 24, 375.

(25) Mangrulkara, P. A.; Kamble, S. P.; Meshram, J.; Rayalu, S. S.Adsorption of phenol and o-chlorophenol by mesoporous MCM-41. J.Hazard. Mater. 2008, 160, 414.

(26) Srivastava, V. C.; Swamy, M. M.; Mall, I. D.; Prasad, B.; Mishra,I. M. Adsorptive removal of phenol by bagasse fly ash and activated carbon:equilibrium kinetics and thermodynamics. Colloids Surf. Surf. A: Physi-cochem. Eng. Aspects. 2006, 272, 89.

(27) Hameed, B. H.; Rahman, A. A. Removal of phenol from aqueoussolutions by adsorption onto activated carbon prepared from biomassmaterial. J. Hazard. Mater. 2008, 160, 576.

(28) Ho, Y. S.; McKay, G. The sorption of Lad (II) on peat. Water Res.1999, 33, 578.

(29) Ho, Y. S. Review of second-order models for adsorption systems.J. Hazard. Mater. 2006, 136, 681.

(30) Dogan, M.; Alkan, M. Adsorption kinetics of methyl violet ontoperlite. Chemosphere 2003, 50, 517.

(31) Malik, P. K. Use of activated carbons prepared from sawdusts andrice-husk for adsorption of acid dyes: a case study of acid yellow 36. DyesPigments 2003, 56, 239.

(32) Ozcan, A. S.; Ozcan, A. Adsorption of acid dyes from aqueoussolutions onto acid-activated bentonite. J. Colloid Interface Sci. 2004, 276,39.

(33) Rozada, F.; Otero, M.; Moran, A.; Garcia, A. I. Adsorption of heavymetals onto sewage sludge-derived materials. Biores. Technol. 2008, 99,6332.

(34) Hasany, S. M.; Saeed, M. M. A kinetic and thermodynamic studyof silver sorption onto manganese dioxide from acid solutions. Sep. Sci.Technol. 1992, 27, 1789.

(35) Mahramanlioglu, M.; Kizilcikli, I.; Bicer, O. Adsorption of fluoridefrom aqueous solution by acid treated spent bleaching earth. J. FluorineChem. 2002, 115, 41.

(36) Kumar, A.; Kumar, S.; Kumar, S.; Gupta, D. V. Adsorption ofphenol and 4-nitrophenol on granular activated carbon in basal salt medium:equilibrium and kinetics. J. Hazard. Mater. 2007, 147, 155.

Figure 15. Regeneration curves of chitosan-coated perlite beads loaded with4-chlorophenol.

9246 Ind. Eng. Chem. Res., Vol. 49, No. 19, 2010

(37) Ozkaya, B. Adsorption and desorption of phenol on activatedcarbon and a comparison of isotherm models. J. Hazard. Mater. 2006,129, 158.

(38) Al-Asheh, S.; Banat, F.; Abu-Aitah, L. Adsorption of phenol usingdifferent types of activated bentonites. Sep. Purif. Technol. 2003, 33, 1.

(39) Mohd Din, A. T.; Hameed, B. H.; Ahmad, A. L. Batch adsorptionof phenol onto physiochemical-activated coconut shell. J. Hazard. Mater.2009, 161, 1522.

(40) Singh, B. K.; Rawat, N. S. Comparative sorption equilibrium studiesof toxic phenols on fly ash and impregnated fly ash. J. Chem. Technol.Biotechnol. 1994, 61, 307.

(41) Aksu, Z.; Yerner, J. A Comparative adsorption/biosorption studyof mono-chlorinated phenols onto various sorbents. Waste Manag. 2001,21, 695.

(42) Akcay, M.; Akcay, G. The removal of phenolic compounds fromaqueous solutions by organophilic bentonite. J. Hazard. Mater. 2004, 113,189.

(43) Delval, F.; Crini, G.; Vebrel, J. Removal of organic pollutants fromaqueous solutions by adsorbents prepared from an agroalimentary by-product. Biores. Technol. 2006, 97, 2173.

(44) Koumanova, B.; Peeva-Antova, P. Adsorption of p-chlorophenolfrom aqueous solutions on bentonite and perlite. J. Hazard. Mater. 2002,90, 229.

ReceiVed for reView July 23, 2009ReVised manuscript receiVed July 27, 2010

Accepted August 13, 2010

IE901171B

Ind. Eng. Chem. Res., Vol. 49, No. 19, 2010 9247