Removal of methyl violet from aqueous solution by perlite

en investi-increasee describedns and theponds to asperlite is

Journal of Colloid and Interface Science 267 (2003) 32–41www.elsevier.com/locate/jcis

Removal of methyl violet from aqueous solution by perlite

Mehmet Dogan and Mahir Alkan∗

Balıkesir University, Faculty of Science and Literature, Department of Chemistry, 10100 Balıkesir, Turkey

Received 5 November 2002; accepted 22 May 2003

Abstract

The use of perlite for the removal of methyl violet from aqueous solutions at different concentration, pH, and temperature has begated. Adsorption equilibrium is reached within 1 h. The capacity of perlite samples for the adsorption of methyl violet was found towith increasing pH and temperature and decrease with expansion and increasing acid-activation. The adsorption isotherms arby means of the Langmuir and Freundlich isotherms. The adsorption isotherm was measured experimentally at different conditioexperimental data were correlated reasonably well by the adsorption isotherm of Langmuir. The order of heat of adsorption corresphysical reaction. It is concluded that the methyl violet is physically adsorbed onto the perlite. The removal efficiency (P ) and dimensionlesseparation factor (R) have shown that perlite can be used for removal of methyl violet from aqueous solutions, but unexpandedmore effective. 2003 Elsevier Inc. All rights reserved.

Keywords: Adsorption isotherms; Methyl violet; Perlite; Dye

lorin-

az-arere-

logi-lor

ovalvi-uch

aveent

sseson,d caeenanelu-, red

t 20ng-arilylitesandon-ioused asast,

lite

eticen-hemic

rp-t onctstureher-h de-dia

1. Introduction

Many industries routinely use dyes or pigments to cotheir products. A number of these dyes or pigments areevitably left in the industrial waste, which could be a hard to the environmental [1]. Textile industry watersgenerally processed in biological treatment units formoval of biodegradable organic compounds. These biocal processes typically accomplish very little towards coremoval while handling these wastewaters [2]. The remof color from textile wastewaters is one of the major enronmental problems because of the difficulty of treating swater by conventional treatment methods [3].

In our day various physical–chemical techniques hbeen studied to assess their applicability for the treatmof this type of industrial discharges. Among these procemay be included coagulation, precipitation, flocculatiozonation, reverse osmosis, ion exchange, and activatebon adsorption [2,4]. Various types of materials have bused as adsorbents, such as activated carbon, mangoxide, silica gel, fly ash, wollastonite, lignite, peat, soil, amina, rutil, geothite, hematit, bentonit, sphalerit, anatasemud, mica, illite, kaolinite, and clays [4].

* Corresponding author.E-mail address: [email protected] (M. Alkan).

0021-9797/$ – see front matter 2003 Elsevier Inc. All rights reserved.doi:10.1016/S0021-9797(03)00579-4

r-

se

Perlite is a glassy volcanic rock that expands to aboutimes its original volume upon heating within its softenitemperature range of 760 to 1100◦C [5]. The uses of expanded perlite are many and varied and are based primupon its physical and chemical properties. As most perhave a high silica content, usually greater than 70%,are adsorptive, they are chemically inert in many envirments and hence are excellent filter aids and fillers in varprocesses and materials. Furthermore, perlite is also usa catalyst in chemical reaction [6]. Along the Aegean CoTurkey possesses about 70% (70× 109 tons) of the world’sknown perlite reserves [7]. The main consumption of peris in construction related fields.

In our previous works, we investigated the electrokinproperties of perlite [8], surface titrations of perlite suspsions [9], and adsorption of copper (II) onto perlite [4]. Tpresent work is aimed to study a convenient and economethod for methyl violet removal from water by adsotion on a low-cost and abundantly available adsorbenwhich no work could be found in the literature. The effeof expansion, solution pH, acid-activation, and temperaon methyl violet adsorption have been investigated. Furtmore, the results obtained have been applied to a batcsign for the removal of methyl violet from aqueous meusing perlite samples.

http://www.elsevier.com/locate/jcis

M. Dogan, M. Alkan / Journal of Colloid and Interface Science 267 (2003) 32–41 33

e obs ofe

a-per-lse-

les,ca-ter-py-ex-

.6)),rlite

ob-

ncet

nt

ab-

ingof

andthe

t tokerini-

entshionith-

eterure-as

en-

terngthce.se-bedforever-

io-The

rlitee in

2. Materials and methods

2.1. Materials

The unexpanded and expanded perlite samples wertained from the Cumaovası Perlite Processing PlantEtibank (Izmir, Turkey). The chemical composition of thperlite found in Turkey in the literature is given in Tble 1 [7]. Treatment of the unexpanded and expandedlite samples before use in experiments is described ewhere [8].

In order to obtain the acid-activated perlite sampH2SO4 solutions were used [10]. The cation exchangepacity (CEC) of the various perlite samples was demined by the ammonium acetate method, density bycnometry. The specific surface area of the samples ofpanded (EP), acid-activated expanded perlite (EHP(0unexpanded (UP), and acid-activated unexpanded pe(UHP(0.6)) were measured by BET N2 adsorption [10]. Theresults are summarized in Table 2. All chemicals weretained from Merck.

2.2. Preparation of methyl violet solutions

Methyl violet was dried at 70–80◦C for 4 h to removemoisture and then was dissolved in distilled water. Simethyl violet is difficult to dissolve in water, methyl viole

Table 1Chemical composition of perlite

Constituent Percentage prese

SiO2 71–75Al2O3 12.5–18Na2O 2.9–4.0K2O 4.0–5.0CaO 0.5–2.0Fe2O3 0.1–1.5MgO 0.03–0.5TiO2 0.03–0.2MnO2 0.0–0.1SO3 0.0–0.1FeO 0.0–0.1Ba 0.0–0.1PbO 0.0–0.5Cr 0.0–0.1

-

solution was allowed to stand for 1–2 days until thesorbance of the solutions remained unchanged [1].

2.3. Method

Adsorption experiments were carried out by shak0.5-g perlite samples with 50-ml aqueous solutionsmethyl violet of desired concentrations at various pHstemperatures for 1 h. Prior to adsorption experimentssolution was kept under N2 for 10 min. A preliminary exper-iment revealed that about 1 h is required for methyl violereach the equilibrium concentration. A thermostated shabath was used to keep the temperature constant. Thetial concentration of methyl violet solutes,C0, was variedin the range 5× 10−7–3 × 10−4 mol l−1 for unexpandedand expanded perlite samples. All adsorption experimwere performed at 30◦C and pH 11 except those in whicthe effects of temperature and pH of methyl violet solutwere investigated. The pH of the solution was adjusted wNaOH or HNO3 solution by using an Orion 920A pH meter equipped with a combined pH electrode. The pH mwas standardized with NBS buffers before every measment. At the end of the adsorption period, the solution wcentrifuged for 15 min at 3000 rpm and then the conctration of the residual methyl violet,Ce, was determinedwith the aid of a Cary |1E| UV-visible spectrophotome(Varian). The measurements were made at the waveleλ = 584 nm, which corresponds to maximum absorbanBlanks containing no methyl violet were used for eachries of experiments. The amounts of methyl violet adsorwere calculated from the concentrations in solutions beand after adsorption. Each experimental point was an aage of three independent adsorption tests [11].

3. Results and discussion

3.1. Adsorption isotherms

Figure 1 shows the adsorption isotherms of methyl vlet on the unexpanded and expanded perlite samples.adsorbed amount of methyl violet for unexpanded peis greater than that for expanded perlite. The decreas

ential

Table 2Some physicochemical properties of perlite samples used in the study

Sample Nomenclature CEC Density Specific surface area Zeta pot(meg 100 g−1) (g ml−1) (m2 g−1) (mV)

Expanded, purified in water EP 33.30 2.24 2.30 −46.8Expanded, 0.2-M acid-activated EHP(0.2) 38.20 2.10 – −47.1Expanded, 0.4-M acid-activated EHP(0.4) 43.38 2.04 – −46.3Expanded, 0.6-M acid-activated EHP(0.6) 54.24 1.93 2.33 −44.0Unexpanded, purified in water UP 25.97 2.30 1.22 −23.5Unexpanded, 0.2 M acid-activated UHP(0.2) 32.79 2.32 – −21.8Unexpanded, 0.4-M acid-activated UHP(0.4) 35.00 2.38 – −22.0Unexpanded, 0.6-M acid-activated UHP(0.6) 36.56 2.46 1.99 −21.1

34 M. Dogan, M. Alkan / Journal of Colloid and Interface Science 267 (2003) 32–41

Fig. 1. The effect of thermal treatment on the adsorption of methyl violet on perlite.

lt ofe inostctratha

rmalun-oxylitest tondeEC)nex-

hylngase-rved

asàleze totionr-n inf dy

on

ac-re

ng-ydye

pH,sam-ing

m

con-

the amount of adsorption by expansion may be a resuevents occurring during the calcination: (i) the decreasthe amount of hydroxyl groups and (ii) the removal of mof the micropores due to heating the sample. Infrared speof the unexpanded and expanded perlite samples showthe number of hydroxyl groups is decreased by the thetreatment in the production of expanded perlite fromexpanded perlite. The decrease in the amount of hydrgroups of the adsorbent, which are mainly effective sfor adsorption, during the expansion of perlite is thoughcause a decrease in adsorption capacity, although expaperlite has greater values of cation exchange capacity (Czeta potential (ZP), and specific surface area than upanded perlite [12].

The effect of acid activation on the adsorption of metviolet onto perlite samples is given in Fig. 2, indicatithat the adsorbed amounts of methyl violet slightly decrewith the concentration of H2SO4 used for the acid activation for both of the perlite samples. This decrease obsemay be due to the partial destruction of perlite structureshown by Gonzàlez-Pradas et al. [13] and López-Gonzand Gonzàlez-Garcià [14] for bentonite and may be duthe decrease in OH groups in perlite during the activaprocess, as was shown by Dogan using IR-spectra. Furthemore, the increase in CEC with acid activation, as seeTable 2, may cause a decrease in adsorption capacity o

t

d,

e

as well, which is an indicator of charged replacable sitesthe surface [12].

To study the influence of pH on the adsorption capity of perlite samples for methyl violet, experiments weperformed using various initial solution pH values, chaing from 3 to 11 (Fig. 3). The removal of methyl violet bperlite samples has been increased when the pH of thesolution was increased. In the discussion of the effect ofit is necessary to discuss the pKa-values of the S–OH groupof perlite samples. Surface charge will develop via thephoteric ionization of the surface hydroxyl groups accordto the reactions

(1)SOH+2

K inta1� SOH+ H+

s ,

(2)SOHK int

a2� SO− + H+s ,

where the subscripts denotes the surface and the equilibriuconstant for Eqs. (1) and (2):

(3)K inta1

= [SOH][H+s ]

[SOH+2 ] ,

(4)K inta2

= [SO−][H+s ]

[SOH] .

The formation of surface species, [SOH+2 and SOH], is the

principal mechanism by which protons are released (or


Fig. 2. The effect of acid-activation on the adsorption of methyl violet on perlite.

olu-

a-x-s 2.hat

on-velylues

inre-2)),rlite

sumed) by many oxide surface in aqueous electrolyte stions. For oxides in general, as pK int

a1and pK int

a2increase,

the acidity of both surface species, S–OH+2 and S–OH, de-

creases. In this sense, pK inta values are considered as a me

sure of surface acidity. pK inta2

values for expanded and unepanded perlite samples were, respectively, determined aand 3.0 with NaCl as an electrolyte [4]. It can be said t

7

the surface hydroxyl groups on perlite are acidic. When csidered together with the fact that the surface is negaticharged through the entire range of the studied pH va(i.e., pH 3–10), this result shows that the reaction (2) isfavor of the right-hand side at low concentrations. Thefore, as the pH of the dye solution becomes higher (Eq. (the association of dye cations with negatively charged pe


Fig. 3. The effect of pH of the solution on the adsorption of methyl violet on perlite.

tatic

eacngead-s of-

ownof

thylypi-on.ingdyece a

surface can more easily take place through an electrosinteraction as follows (Eq. (5)):

(5)≡SO− + Dye+ � S–O− −+ Dye.

A study of the temperature dependence of adsorption rtions gives valuable information about the enthalpy chaduring adsorption. The effect of temperature on thesorption isotherm was studied by carrying out a serieisotherms at 30, 40, 50, and 60◦C for both of the perlite sam

-

ples (unexpanded perlite and expanded perlite) and shin Fig. 4. Results indicate that the adsorption capacityunexpanded and expanded perlite for adsorption of meviolet increases with increasing temperature, which is tcal for the adsorption of most organics from their solutiThe effect of temperature is fairly common and increasthe temperature must increase the mobility of the largecation. Furthermore, increasing temperature may produ


Fig. 4. The effect of temperature on the adsorption of methyl violet on perlite.

n-

theoun

a is

re-pur-o ofFre-

tionbe-ng-

swelling effect within the internal structure of the perlite eabling large dyes to penetrate further.

3.2. Isotherm analysis

The purpose of the adsorption isotherms is to relateadsorbate concentration in the bulk and the adsorbed amat the interface [15]. The analysis of the isotherm dat

t

important to develop an equation which accurately repsents the results and which could be used for designposes [16]. Several isotherm equations are available. Twthem have been selected in this study: Langmuir andundlich isotherms.

The Langmuir isotherm assumes that all the adsorpsites are equivalent [17] and that there is no interactiontween adsorbed species [16]. The linear form of the La


acefre

ntnt

rit-

nge oftion,If

(6)quationnhatellthet th

rm

0–

withle”ng-less

in-

ylthatt thesing

-’tper-

ea-

denesderthationandad-

over

dsor-

n ofc di-

at-nd

oveds

dedivescan),

romed

muir isotherm for adsorption onto a single site solid surf(a homogeneous surface having one type of site) hasquently been applied as

(6)Ce

Qe

= 1

QmK+ Ce

Qm

,

where Qe is equilibrium dye concentration on adsorbe(mol g−1), Qm is monolayer capacity of the adsorbe(mol g−1), K is adsorption constant (l mol−1), and Ce isequilibrium dye concentration in solution (mol l−1). Accord-ing to the Eq. (6), a plot ofCe/Qe versusCe should be astraight line with slope 1/Qm and intercept 1/QmK whenadsorption follows the Langmuir equation.

The Freundlich equation in logarithmic form can be wten as

(7)logQe = logKF + 1

nlogCe,

whereKF andn are empirical Freundlich constants, beiindicative of the extent of the adsorption and the degrenonlinearity between solution concentration and adsorprespectively. The value ofn is usually between 2 to 10.Eq. (7) applies, a plot of logQe against logCe will give astraight line, of slope 1/n and intercept logKF [18].

Adsorption isotherms were obtained in terms of Eqs.and (7) using experimental adsorption results in these etions. The Langmuir equation represents the adsorpprocess very well; ther-values were almost all higher tha0.99, indicating a very good mathematical fit. The fact tthe Langmuir isotherm fits the experimental data very wmay be due to homogenous distribution of active sites onperlite surface, since the Langmuir equation assumes thasurface is homogenous [17,19]. Furthermore,r-values forfitting of the experimental data to the Freundlich isothewere in the range 0.61–0.97.

The removal efficiency,P , is given as [17]

(8)P = C0 − Ce

C0× 100.

The removal efficiency rose from 98.0–80.0% at 30◦C up to99.8–70.0% at 60◦C for unexpanded perlite and from 98.55.0% at 30◦C up to 99.8–60.0% at 60◦C for expandedperlite.

The shape of the isotherm may also be considereda view to predicting if an adsorption system is “favorabor “unfavorable.” The essential characteristics of a Lamuir isotherm can be expressed in terms of a dimensionseparation factor or equilibrium parameterR [20], which isdefined by

(9)R = 1

1+ KCe

.

According to the value of R the isotherm shape may beterpreted as follows:

-

-

e

Value ofR Type of adsorption

R > 1.0 unfavorableR = 1.0 linear0< R < 1.0 favorableR = 0 irreversible

The fact that all theR-values for the adsorption of methviolet on the perlite are in the range 0.004–0.999 showsthe adsorption process is favorable. This indicates thaadsorption process becomes more favorable with increatemperature [15].

The enthalpy of adsorption,�Hads, as a function of coverage fraction (θ = Qe/Qm) can be estimated from vanHoff isochore using the adsorption data at various tematures for methyl violet [18,20]. The subscriptθ meansthat the equilibrium constant at each temperature is msured at constant coverage. The values of�Hadsat θ = 0.5were calculated as 13.4 kcal mol−1 for expanded perlite an16.5 kcal mol−1 for unexpanded perlite from the data givin Fig. 5. The heat of physical adsorption, which involvonly relatively weak intermolecular forces such as vanWaals and electrostatic interactions, is low compared toof chemisorption, which involves essentially the formatof a chemical bond between the sorbate and moleculethe surface of the adsorbent. The upper limit for physicalsorption may be higher than 20 kcal mol−1 for adsorptionon adsorbents. The heat of chemisorption ranges from100 kcal mol−1 to less than 20 kcal mol−1 [21]. The resultsabove show that the interaction between surface and abate molecules is a physical interaction.

3.3. Designing batch adsorption from isotherm data

Adsorption isotherms can be used to predict the desigsingle-stage batch adsorption systems [17]. A schematiagram is shown in Fig. 6 where the effluent containsV l ofwater and an initial methyl violet concentrationC0, which isto be reduced toC1 in the adsorption process. In the trement stageW g perlite (dye-free) is added to solution athe dye concentration on the solid changes fromQ0 = 0 (ini-tially) to Q1. The mass balance that equates the dye remfrom the liquid effluent to that accumulated by the solid i

(10)V (C0 − C1) = W(Q1 − Q0) = WQ1.

In the case of the adsorption of methyl violet on unexpanand expanded perlite samples the Langmuir isotherm gthe best fit to experimental data. Consequently equationbe best substituted forQ1 in the rearranged form of Eq. (10giving adsorbent/solution ratios for this particular system

(11)W

V= C0 − C1

Qe

≡ C0 − Ce(

QmKCe

1+KCe

) .

Figures 7a and 7b show a series of plots derived fEq. (11) for the adsorption of methyl violet on unexpandand expanded perlite. An initial dye concentration of 5.0 ×


Fig. 5. Plot of− lnCe versus 1/T for methyl violet adsorption on perlite.

Fig. 6. Single-stage batch adsorber.

esuceing

of

well

e-

ith

ith

mef-

t of

10−5 mol l−1 at 30◦C and pH 11 is assumed and figurshow the amount of effluent which can be treated to redthe methyl violet content by 50, 60, 70, 80, and 90% usvarious masses of adsorbent.

3.4. Conclusions

The following points may be mentioned as the resultsthis study.

1. The experimental data were correlated reasonablyby the Langmuir adsorption isotherm.

2. The adsorbed amount of methyl violet slightly dcreased with increasing concentration of H2SO4 usedfor the acid activation for both of perlite samples.

3. The adsorbed amounts of methyl violet increased wincreasing pH for both of perlite samples.

4. The adsorbed amount of methyl violet increased wincrease in temperature for both of perlite samples.

5. The dimensionless separation factor(R) showed thatperlite can be used for removal of methyl violet froaqueous solutions, but unexpanded perlite is morefective. Its adsorption capacity is greater than thaexpanded perlite.


Fig. 7. Volume of effluent (V ) treated against adsorbent mass (W ) for different percentages of methyl violet removal.

er-ol

abletem

ni-

o-

ch-

.De-

rk,

7)

6. The values of�Hadsfor unexpanded and expanded plite samples were calculated as 16.5 and 13.4 kcal m−1,respectively.

7. As a result, it can be said that perlite has considerpotential as an adsorbent of dyes in a commercial sysbecause it is cheap.

Acknowledgment

The work was financially supported by the Balıkesir Uversity Research Fund (Project 2000/1).

References

[1] M. Dai, J. Colloid Interface Sci. 164 (1994) 223–228.[2] I. Arvanitooyannis, I. Eleftheriadis, E. Tsatsaroni, Chem

sphere 18 (9/10) (1989) 1707–1711.[3] S.K. Khare, K.K. Panday, R.M. Srivastava, V.N. Singh, J. Chem. Te

nol. Biotechnol. 38 (1987) 99–104.[4] M. Alkan, M. Dogan, J. Colloid Interface Sci. 243 (2001) 280–291[5] P.W. Harben, R.L. Bates, Industrial Minerals: Geology and World

posits, Metal Bulletin Inc., London, 1990.[6] C.W. Chesterman, Industrial Minerals and Rocks, AIME, New Yo

1975.[7] S.S. Uluatam, J. Am. Water Works Assoc. 83 (6) (1991) 70–71.[8] M. Dogan, M. Alkan, Ü. Çakir, J. Colloid Interface Sci. 192 (199

114–118.[9] M. Alkan, M. Dogan, J. Colloid Interface Sci. 207 (1998) 90–96.


0)

m-

cìa,111.954)

m.

44.3)

w

80)

d

[10] O. Inel, Turkish J. Chem. 19 (4) (1995) 323–330.[11] M. Dogan, M. Alkan, Y. Onganer, Water Air Soil Pollut. 120 (200

229–248.[12] M. Dogan, MSc Thesis, University of Balıkesir, Department of Che

istry, Balıkesir, Turkey, 1997 (in Turkish).[13] E. Gonzàlez-Pradas, M. Villafranca-Sànchez, A. Valverde-Gar

M. Socias-Viciana, J. Chem. Technol. Biotechnol. 42 (1988) 105–[14] J.D. López-Gonzàlez, S. Gonzàlez-Garcìa, An. Fis. Quim. B 50 (1

465–470.[15] H.M. Asfour, O.A. Fadali, M.M. Nassar, M.S. El-Geundi, J. Che

Technol. Biotechnol. 35 (1) (1985) 21–27.

[16] J. Eastoe, J.S. Dalton, Adv. Colloid Interface Sci. 85 (2000) 103–1[17] G. McKay, M.S. Otterburn, A.J. Aga, Water Air Soil Pollut. 24 (

(1985) 307–322.[18] K.J. Laidler, J.H. Meiser, Physical Chemistry, Houghton Mifflin, Ne

York, 1999.[19] A. Gürses, S. Bayrakçeken, M.¸S. Gülaboglu, Colloids Surf. 64 (1)

(1992) 7–13.[20] G. McKay, V.J.P. Poots, J. Chem. Technol. Biotechnol. 30 (6) (19

279–292.[21] K.E. Noll, V. Gounaris, W.S. Hou, Adsorption Technology for Air an

Water Pollution Control, Lewis Publishers, 1992, pp. 21–22.

Documents

Removal of methyl violet from aqueous solution by perlite