This article was downloaded by: [Universiti Putra Malaysia]On: 14 February 2014, At: 20:26Publisher: Taylor & FrancisInforma Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House,37-41 Mortimer Street, London W1T 3JH, UK

Separation Science and TechnologyPublication details, including instructions for authors and subscription information:http://www.tandfonline.com/loi/lsst20

Adsorption of Nickel on Electric Arc Furnace Slag:Batch and Column StudiesMohammed Yusuf a b , Luqman Chuah b , Moonis Ali Khan c & Thomas S. Y. Choong ba Department of Chemical and Materials Engineering, Faculty of Engineering , University ofAuckland , New Zealandb Department of Chemical and Environmental Engineering, Faculty of Engineering ,Universiti Putra Malaysia , Selangor , DE , Malaysiac Advance Material Research Chair, Chemistry Department, College of Science , King SaudUniversity , Riyadh , Saudi ArabiaAccepted author version posted online: 26 Sep 2013.Published online: 17 Jan 2014.

To cite this article: Mohammed Yusuf , Luqman Chuah , Moonis Ali Khan & Thomas S. Y. Choong (2014) Adsorption ofNickel on Electric Arc Furnace Slag: Batch and Column Studies, Separation Science and Technology, 49:3, 388-397, DOI:10.1080/01496395.2013.843099

To link to this article: http://dx.doi.org/10.1080/01496395.2013.843099

PLEASE SCROLL DOWN FOR ARTICLE

Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) containedin the publications on our platform. However, Taylor & Francis, our agents, and our licensors make norepresentations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of theContent. Any opinions and views expressed in this publication are the opinions and views of the authors, andare not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied upon andshould be independently verified with primary sources of information. Taylor and Francis shall not be liable forany losses, actions, claims, proceedings, demands, costs, expenses, damages, and other liabilities whatsoeveror howsoever caused arising directly or indirectly in connection with, in relation to or arising out of the use ofthe Content.

This article may be used for research, teaching, and private study purposes. Any substantial or systematicreproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in anyform to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http://www.tandfonline.com/page/terms-and-conditions

Adsorption of Nickel on Electric Arc Furnace Slag: Batch andColumn Studies

Mohammed Yusuf,1,2 Luqman Chuah,2 Moonis Ali Khan,3 andThomas S. Y. Choong21Department of Chemical and Materials Engineering, Faculty of Engineering,University of Auckland, New Zealand2Department of Chemical and Environmental Engineering, Faculty of Engineering,Universiti Putra Malaysia, Selangor, DE, Malaysia3Advance Material Research Chair, Chemistry Department, College of Science,King Saud University, Riyadh, Saudi Arabia

The ability of electric arc furnace slag (EAFS), a by-product ofthe steel industry to adsorb nickel [Ni(II)] from an aqueous solution,was investigated by both batch and column operations. The charac-terization studies showed the mesoporous nature of EAFS withdominance of acidic sites. The adsorption was found to be dependenton the adsorbent dosage, contact time, the pH, and initial metal ionconcentration. Optimum Ni(II) uptake was 160.92mg/g at 1000mg/L initial concentration with equilibration time 216 h. Adsorption fol-lows the pseudo-second-order kinetics model. Linear and non-linearisotherm models revealed the applicability of the Langmuir modelconfirming monolayer adsorption. Both the column bed capacityand the exhaustion time increased with increase in bed height. Thesaturation time was found to increase from 42 to 46 h with a decreasein the flow rate from 15 to 5mL/min. The bed depth saturation timeand Thomas models were evaluated. The experimental breakthroughcurves agreed well with the predicted model.

Keywords batch and column process; chemisorption; electric arcfurnace slag; nickel; Thomas model

INTRODUCTION

Contamination of water resources by heavy metals isa global environmental issue. Excessive levels and long-termintake of heavy metals have been linked to a wide range ofhealth hazards, such as skin diseases, birth defects, and evencancer (1) in certain cases. Divalent heavy metals, such ascadmium (Cd), copper (Cu), and zinc (Zn) are commonlyfound in discharges from mining, electronics, motor vehicle,textile, electroplating, basic steel work, and metal-finishingindustries.

Nickel [Ni(II)], a divalent heavy metal, is essential intraces for human health. Excessive intake of Ni(II) may causerespiratory failure, birth defects, asthma and chronic bron-chitis, heart disorders and dizziness, dry cough, tightness ofthe chest, chest pain, headache, nausea, vomiting, cyanosis,and weakness (2). Hence, it is essential to remove or minimizeNi(II) to a permissible limit in industrial wastewater beforebeing discharged to natural water sources. The concentrationof Ni(II) in industrial wastewater has been reported to rangebetween 3.4 and 900mg=L (3). The maximum allowablelimits of Ni(II) in potable and surface water discharges are0.04mg=L (4) and 2.0mg=L (5), respectively.

Current methods for the removal and recovery of Ni(II)from industrial waste streams includes precipitation,ion-exchange, reverse osmosis, electrochemical treatment,solvent extraction, flocculation, membrane separation (6),and electrocoagulation (7). Several merits and demeritsare associated with each of the aforementioned techniques.Most of the methods involve high capita, not suitable forsmall-scale industries (8). Therefore, it is essential to comeup with a process which is economically appealing andeco-friendly. Considering these essentialities, adsorptionhas an upper hand for water decontamination among theaforementioned techniques. It is suitable even withadsorbate concentration as low as 1mg=L (9).

Slag is a solid waste material formed by flux reactionwith impurities during smelting and refining operation ofmetals. Electric arc furnace slag (EAFS), a valuable majorby-product of steel making industry. The reported annualglobal steel slag production is fifty million tons (10). Dueto its large scale production, dumping it off is graduallybecoming a major environmental issue. Studies showedextensive EAFS use in construction of roads and as a pave-ment construction material (11,12). The utilization ofEAFS as adsorbent for water decontamination before

Received 18 February 2013; accepted 7 September 2013.Address correspondence to Moonis Ali Khan, Advance

Material Research Chair, Chemistry Department, College ofScience, King Saud University, Riyadh, Saudi Arabia. E-mail:mokhan@ksu.edu.sa or moonisalikhan@gmail.com

Color versions of one or more of the figures in the article canbe found online at www.tandfonline.com/lsst.

Separation Science and Technology, 49: 388–397, 2014

Copyright # Taylor & Francis Group, LLC

ISSN: 0149-6395 print=1520-5754 online

DOI: 10.1080/01496395.2013.843099

388

Dow

nloa

ded

by [

Uni

vers

iti P

utra

Mal

aysi

a] a

t 20:

26 1

4 Fe

brua

ry 2

014

being used as a construction material could be an excellentsolid waste management alternative. Previous studiesreported EAFS use for removing Mn and Pb (13,14). Thisstudy testified EAFS adsorptive potential for Ni(II)removal from aqueous phase by employing both batchand column process. Various parameters such as the effectof initial metal concentration, contact time, particle size,dosage, pH, bed-height, and flow rate on the adsorptionhave been studied.

MATERIALS AND METHODS

Materials

Chemicals and reagents used during this work were ofanalytical reagent grade or as specified. The EAFS wasobtained from steel making plant in Southern Steel Berhad(Penang, Malaysia). Nickel chloride hexahydrate (NiCl2 �6H2O) was purchased from Analytical Univar Reagent,APS Ajax. Reagents such as sodium hydroxide (NaOH),hydrochloric acid (HCl), sodium carbonate (Na2CO3),sodium bicarbonate (NaHCO3), and potassium nitrate(KNO3) were purchased from Sigma-Aldrich, Malaysia.

Adsorbate and Adsorbent

Stock solution of Ni(II) (1000mg=L) was prepared indistilled water (DW) from NiCl2 � 6H2O. All the workingsolutions of varying concentrations were obtained bydiluting the stock solution with DW. The solution wasfurther diluted to required concentrations before use.

The EAFS was washed with DW in order to removesurface impurities and dried at 100�C for the period of24 h in an oven. The slag was crushed, grounded andscreened through sieves to various particle sizes (0.5, 1, 2,and 3mm) to collect samples for experimental studies.

Batch and Column Experiments

Batch experiments were performed at room temperature(25�C) in Erlenmeyer flasks (250mL) containing Ni(II)synthetic solution, 1 g of EAFS was added and agitatedusing a sweep shaker at a speed of 150 rpm, the suspensionafter equilibration was filtered using filter paper (WhatmanGrade No. 42). The Ni(II) concentration in the filtrate wasmeasured using atomic absorption spectrophotometer(AAS, TAS 990 Shimadzu, Kyoto, Japan). The experimentswere performed in triplicate by varying Ni(II) concentra-tions (50 to 1000mg=L), contact time (0 to 216 h), doses(0.1 to 1 g), pH (2 to 10), and particle size (0.5 to 3mm).

Column experiments were conducted in a glass column(internal diameter¼ 3 cm; length¼ 50 cm). The columnwas packed with the adsorbent between two supportinglayers of glass wool. The pH and temperature of the Ni(II)solution in the stream and column was maintainedconstant at 8.0 and 25�C, respectively, using a thermostaticbath. 1000mL of Ni(II) solution with 100mg=L initial

concentration (C0), was pumped upward using peristalticpump (PP 30, Miclins India, Limited, Chennai, India) witha flow rate of 5mL=min of metal solution through fixedbed column. The effluent samples were collected in 50mLfractions at regular time intervals and the amount of metal(C) was determined in each fraction by using AAS. Thebreakthrough curve was obtained by plotting the columneffluent concentration ðc=c0Þ versus the volume of theeffluent or time of treatment.

The percentage metal adsorption and adsorptioncapacity at equilibrium (qe, mg=g) were calculated by usingEqs. (1) and (2), respectively:

%Adsorption ¼ ðC � CeÞC

� 100 ð1Þ

qe ¼ ðC � CeÞ �V

mð2Þ

where C and Ce are the effluent metal ion concentrationand equilibrium concentration of the metal ion in the efflu-ent (mg=L), V is the volume of the metal ion solution (L),and m is the mass of the adsorbent (g).

The maximum column capacity (qtotal, mg=g) fora known inlet concentration and the flow rate is equal tothe area under the plot of the adsorbed Ni(II) concen-tration ½Cad ¼ C0 � Ce�, where Ce is the equilibriumconcentration of the metal ion in the effluent and C0 isthe influent metal ions concentration (mg=L), t is time(h). It is calculated as (15):

qtotal ¼QA

1000

Zt¼ttotal

t¼0

Cad � dt ð3Þ

where ttotal, Q, and A are the total flow rate time (min), theflow rate (mL=min), and the area under the breakthroughcurve, respectively.

The equilibrium uptake [qe(exp)], the weight of Ni(II)adsorbed per unit dry weight of EAFS (mg=g) in the col-umn, is calculated as follows:

qeðexpÞ ¼qtotalX

ð4Þ

where X is the total dry weight of EAFS in the column (g).The removal percentage of Ni(II) Y(%) can be obtained

from the following equation:

Y ð%Þ ¼ qtotalC0Ve

� 100% ð5Þ

where VeðVeðLÞ ¼ QteÞ is the volume of solution requiredto reach the exhaustion point.

ADSORPTION OF NICKEL ON ELECTRIC ARC FURNACE SLAG 389

Dow

nloa

ded

by [

Uni

vers

iti P

utra

Mal

aysi

a] a

t 20:

26 1

4 Fe

brua

ry 2

014

Analytical Methods

Fourier transforms infrared (FT-IR) spectrometer(Model: FTIR-200, Jas co., Tokyo, Japan) was used todetect the surface functional groups of EAFS. The surfacearea and the average pore diameters were determined byusing the Braunauer-Emmeh-Teller (BET) surface analyzer.The acidic and basic surface sites of EAFS were analyzedby using the Boehm titration method (16). One hundredmilligrams of EAFS were placed in 100mL cappedvolumetric flasks containing 50mL of 0.1M Na2CO3,0.1M NaHCO3, 0.1M NaOH, and 0.1M HCl. Themixture was manually shaken for 3 days. A 0.45 mm Nylonfiber filter was use to filter the solution. Ten milliliter ofeach filtrate was pipeted. The excess acid was titrated with0.1M NaOH while, the excess base was titrated with 0.1MHCl. The acidic site concentrations were determinedfrom the assumption that NaOH neutralizes carboxylic,phenolic, and lactonic groups while Na2CO3, neutralizesthe carboxylic and lactonic groups, and NaHCO3 neutra-lizes the carboxylic groups only. The basic site concen-trations were determined from the amount of HCl thatreact with EAFS.

RESULTS AND DISCUSSION

Characterization of Adsorbent

The surface area, average pore diameter, and porevolume were determined by N2 adsorption=desorptionisotherm at 77K. It was observed that the position of N2

adsorption isotherm virtually correspond to that of desorp-tion, that is the hysteresis phenomenon identifying theirreversibility of sorption process can be ignored for allthe analyzed samples. Figure 1a illustrates that N2 adsorp-tion and desorption by an EAFS pores is virtually revers-ible. Figure 1b shows the pore size distribution diameterwhich is predominant in EAFS and falls in accordance tothe pore diameter classification of IUPAC. Surface areaand pore size distribution parameter are listed in Table 1.The pore can be termed as macropore (pore diame-ter>50 nm), mesopore (pore diameter between 2 and50 nm), and micropore (pore diameter<2 nm) (17). Thepresence of both micro and mesopores was observed inthe EAFS sample with dominance of mesopores.

Boehm’s acid-base titration studies showed presence of0.787mmol=g acidic groups (carboxylic–0.342mmol=g,phenolic–0.445mmol=g) and 0.011mmol=g basic groups.The results confirm the dominance of acidic functionalitiesover the EAFS surface.

The FT-IR spectra of EAFS (Fig. 2) displayed a numberof absorption peaks, indicating the complex nature of thematerial examined. An adsorption peak around 3920 cm�1

represented hydrated minerals such as Ca(OH)2 and a peakaround 3771 cm�1 indicating the S-O stretch such as gypsum.The peak around 1559 cm�1 correspond to C-O a symmetric

stretching, whereas a peak at 1472cm�1 may be due to thebonding in CO3

2� ion, which indicates the presence of somesort of carbonated mineral, possible due to the adsorption ofCO2 from the atmosphere. The peaks at 988cm�1 and874 cm�1 are Si�O and Al�O stretching while peaks at790 cm�1 to 709 cm�1 are due to C�O bonding (18). Shiftingin peaks was observed after Ni(II) adsorption; this may indi-cate that these functional groups are likely to participate inthe metal binding. A shift in the hydroxide group from3920 cm�1 to 3575 cm�1, the sulphonyl group from3771 cm�1 to 3637 cm�1, the alcohol group from 1559 cm�1

to 1557 cm�1, while a shift in the carbonate group from1472 cm�1 to 1498cm�1 was observed. The Al�O and Si�Ostretching was observed which tends to shift from 988cm�1

to 874cm�1 and 1220cm�1 to 969cm�1, respectively. Finally,the C�O bonding shift from (790 cm�1 and 709 cm�1) to(854 cm�1 and 812 cm�1). The shift of these functional

FIG. 1. N2 adsorption=desorption isotherm at 77K (a), and Pore size

distribution curve (b) of EAFS.

390 M. YUSUF ET AL.

Dow

nloa

ded

by [

Uni

vers

iti P

utra

Mal

aysi

a] a

t 20:

26 1

4 Fe

brua

ry 2

014

groups could correspond to the complexion and coordi-nation of these groups with metal ion (19,20).

Scanning Electron Microscopy (SEM) images revealedthat the surface is porous and irregular (Fig. S1a). Theimage of EAFS after Ni(II) adsorption showed a uniformlyoccupied surface confirming Ni(II) adsorption of EAFS(Fig. S1b). Energy-dispersive X-ray spectroscopy (EDX)analysis of EAFS before adsorption revealed the presenceof 48.87% Fe and 0.25% Sc, while, after adsorption, Ca(31.77%) and Ni (27.94%) were prevalent along with tracesof Cl (0.80%). The elemental analysis confirms binding ofNi(II) ions on the EAFS surface.

Batch Studies

Effect of pH

The importance of metal solution pH cannot be overem-phasized as it plays a critical role in the adsorption. This isbecause hydrogen ion competing with positively chargemetal ion on the active site of the adsorbent. The effectof pH on Ni(II) adsorption on EAFS was studied by vary-ing initial solution of pH. The Ni(II) uptake was found toincrease as pH increased from 2 to 9. Further, Ni(II)uptake decreases as the pH goes higher. Maximum Ni(II)adsorption (39.2mg=g) was observed at pH: 8 (Fig. S2a).

Furthermore, between pH: 6 and 8, there was a sharpincrease in Ni(II) adsorption, and this might be due tothe formation of nickel hydroxide which normally startsin this pH range (21). Below pH: 6, Ni(II) is in ionic form(21). A slight decrease in adsorption was observed abovepH: 8. This is due to the occurrence of simultaneous sorp-tion and precipitation. A similar result was reported else-where (2). Furthermore, the formation of CaOH fromCaO present in the slag might also be the possible reasonbehind decrease in adsorption. As the pH rises, the precipi-tation was higher; therefore, pH: 8 was selected for furtherstudies. The observed pHpzc of the adsorbent was at pH:6.7 (Fig. S2b). At pH> pHpzc, the surface of the adsorbentis negatively charged which enhances the adsorption ofpositively charged Ni(II) through the electrostatic forceof attraction (19). The positively charged surface reducedthe adsorption capacity at pH< pHpzc. This could be dueto a repulsive force between Ni(II) cations and the EAFSsurface.

Effect of Particle Size

Particles of different sizes have different surface area andadsorption capacity (20).The effects of EAFS particle sizeon Ni(II) adsorption was investigated at fixed Ni(II) con-centration �100mg=L. The operating conditions were keptconstant at pH: 8, agitation speed 150 rpm, and contacttime 9 days. The result showed that Ni(II) removal tendsto decrease from 82 to 23% with increase in particle sizefrom 0.5 to 3mm, respectively (Fig. S3). The smaller par-ticle has a shorter diffusion path, thus allowing the adsorb-ate to penetrate deeply and quickly into the adsorbentparticles resulting in a higher rate of adsorption. Also, thereis a possibility of intra-particle diffusion from the outermost

TABLE 1Surface analysis data of EAFS

Parameters Values

BET surface area, m2=g 1.513Pore volume, cc=g 0.004Pore radius, A 24.597

FIG. 2. FT-IR spectra of EAFS before and after Ni(II) adsorption.

ADSORPTION OF NICKEL ON ELECTRIC ARC FURNACE SLAG 391

Dow

nloa

ded

by [

Uni

vers

iti P

utra

Mal

aysi

a] a

t 20:

26 1

4 Fe

brua

ry 2

014

surface into the pores of the material. As the particle sizeincreases the diffusion resistance to mass transfer becomesgreater. This is due to various factors, such as contact time,diffusion path, and blockage in sections of the particlesresulting in lowering in adsorption (20).

Effect of Dosage

The effect of EAFS dose on Ni(II) adsorption was stud-ied at an optimum pH: 8 and 100mg=L initial Ni(II) con-centration. The increase in EAFS dosage from 0.1 to 1 gresulted in an increase in adsorption of Ni(II) from 3.5 to82.1%, respectively (Fig. S4). An increase in adsorptionmight be due to the increase in the adsorbent surface andtherefore, more active functional groups resulting in theavailability of more binding sites for Ni(II) adsorption (21).

Effects of Contact Time and Concentration

With increase in Ni(II) concentration, the adsorptioncapacity increases from 11.2 to 160.9mg=g, respectively(Fig. 3). This is because it provides a driving force in orderto overcome mass transfer resistance between the adsorb-ent and the adsorption medium (22). Initially for 26 h, fastincrease in adsorption at various initial Ni(II) concentra-tions as visible by a steep slope of contact time plot wasobserved on EAFS. This is due to a large number of avail-able active sites for metal adsorption, and gradually showsdue to the low diffusion rate in the adsorbent internalporous structures (23) attaining equilibrium in 216 h.

Adsorption Isotherm Studies

The mathematical model that describes the distributionof the adsorbate between the solid (adsorbent) and liquidphase is known as an adsorption isotherm (24). The iso-therm data analysis by fitting them to different models is

an important step in finding out a suitable model thatcan be used for design purposes. During this study bothlinear and non-linear Langmuir, Freundlich, andRedlich-Peterson (RP) isotherms models have beenapplied.

The Langmuir isotherm assumes monolayer adsorptiononto a surface containing a finite number of adsorptionsites of uniform strategies of adsorption with no trans-migration of adsorbate in the plane of the surface (25).In a non-linear form the Langmuir isotherm is given as:

qe ¼KLCe

1þ aLCeð6Þ

The linear form of the Langmuir isotherm is given as:

Ce

qe¼ 1

kLþ aLCe

kLð7Þ

where Ce is the equilibrium concentration of the metal inthe effluent, qe is the equilibrium sorption capacity (mg=g), and aL and kL are the Langmuir constants.

The Freundlich isotherm assumes heterogenous surfaceenergies, in which the energy term in the Langmuir equa-tion varies as a function of the surface coverage (26). Theisotherm in a non-linear form is given as:

qe ¼ kFC1=ne ð8Þ

The linear form of the Freundlich isotherm is given as:

log qe ¼ log kF þ 1

nlogCe ð9Þ

where kf and n are the Freundlich constants. kf (mg=g)(L=mg)1=n is related to the adsorption capacity of theadosorbent and n is the heterogeneity factor.

The Redlich-Peterson (RP) isotherm is a hybrid iso-therm featuring both Langmuir and Freundlich isotherms(27). The liner form of Redlich-Peterson is described below:

ln KRCe

qe� 1

� �¼ lnaR þ blnCe ð10Þ

The isotherm in non-linear form is given as:

qe ¼KRCe

1þ aRCbe

ð11Þ

where KR (mg=g) and aR (mg=gb) are the R-P constants andb is the R-P isotherm exponent, lies between 0 and 1,respectively.

If b¼ 1, R-P isotherm will covert to Langmuir isotherm.FIG. 3. Effect of contact time on Ni(II) adsorption on EAFS at various

initial concentrations.

392 M. YUSUF ET AL.

Dow

nloa

ded

by [

Uni

vers

iti P

utra

Mal

aysi

a] a

t 20:

26 1

4 Fe

brua

ry 2

014

aRCbe � 1, the R-P isotherm will reduce to Freundlich

equation.aRC

be � 1, the R-P isotherm reduced to Henry’s law.

Non-linear isotherm parameters for the aforementionedmodels were evaluated by using Microsoft Excel SOLVERsoftware. The residual or sum of squares error (SSE) andaverage relative error (ARE) function were used to mea-sure the goodness of fit. The SSE and ARE can be definedas:

Sumof Error Squares ðSSEÞ ¼

ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiPðqe;exp � qe;calÞ2

N

!vuutð12Þ

AverageRelative Error ðAREÞ ¼ 100

n

Xni¼1

qe;exp � qe;calcqe;exp

� �

ð13Þ

Table 2 summarizes constants, regression coefficient(R2), and error functions values for the applied isothermmodels. Linear and non-linear (Fig. 4) models for theLangmuir model confirmed better applicability of themodel as inferred by the highest R2 values for linear andnon-linear model and lowest error functions (SSE and

ARE) values. The results are in good agreement witha previous study reporting Mn(II) adsorption on EAFS (13).

Adsorption Kinetics

To examine the Ni(II) adsorption mechanism on EAFS,pseudo-first-order and pseudo-second-order models atdifferent initial Ni(II) concentrations (50 to 1000mg=L)have been applied.

The pseudo-first-order rate kinetics model equation,originally formulated by Lagergren (28), is given as:

logðqe � qÞ ¼ log qe �K1

2:303� t ð14Þ

where qe and q are the equilibrium sorption capacity andsorption capacities of Ni(II) on EAFS at equilibrium andat time t (mg=g), respectively, and K1 is the pseudo-first-order rate constant (1=min).

The pseudo-second-order kinetics model by Ho andMcKay (29) is given as follows:

t

q¼ 1

K2q2eþ 1

qe� t ð15Þ

where K2 is the pseudo-second-order rate constant (g=mg-min).

Results showed comparatively higher pseudo-second-order model R2 values at various Ni(II) concentrations(Table 3). The experimental (qe,exp) and calculated (qe,cal)adsorption capacities at various concentrations are nearerconfirming applicability of pseudo-second-order kineticsmodel.

TABLE 2Isotherm parameters for Ni(II) adsorption on EAFS

Isotherms

Parameters

Linear Non-linear

LangmuiraL 0.0091 0.0086KL 1.8570 1.7598R2 0.9961 0.9999SEE – 4.56ARE – 4.05FreundlichKF 4.4781 6.15361=n 0.6304 0.5703R2 0.9519 0.9456SEE – 23.17ARE – 14.12Redlich-PetersonKPR 0.0057 0.0081qPR 300.04 216.767B 1.057 1.009R2 0.9957 0.9999SEE – 14.27ARE – 4.83

FIG. 4. Non-linear isotherm models.

ADSORPTION OF NICKEL ON ELECTRIC ARC FURNACE SLAG 393

Dow

nloa

ded

by [

Uni

vers

iti P

utra

Mal

aysi

a] a

t 20:

26 1

4 Fe

brua

ry 2

014

Column Studies

The flow of the adsorbate solution causes the adsorptionzone to move. The effluent concentration will start to riseat a particular time, denoted as the breakthrough point.The breakthrough time (ta) is defined as the time neededto approach a specific through curve concentration whilethe saturation time is defined as the time taken for c

c0to reach

1, usually becomes longer as the bed depth increases. Theloading behavior of Ni(II) to be removed from a solutionin a fixed bed is normally expressed in terms of Ce=C0 asa function of time, in a given breakthrough curve anda Ni(II) solution with pH: 8.0 for column experiments.

Effect of Bed Height

The accessibility of the metals in a fixed-bed columnlargely dependent on the quantity of adsorbent inside thecolumn. The column experiments were performed byvarying the amount of EAFS [102 g (10 cm bed height),203 g (20 cm bed height), and 302 g (30 cm bed height)] ata fixed flow rate-7mL=min and an initial Ni(II) concen-tration-100mg=L. The void fraction of the column¼ 0.419;0.419; density of the particle¼ 2.47 g=cm3. Generally, theamount of metal adsorbed increases with increasing bedheight. Figure 5a represents the breakthrough graph forNi(II) adsorption onto EAFS at different bed heights ina fixed-bed column.

It is clearly seen that the gradient of S-shaped curvesdropped with bed height, indicating that the breakthroughcurve became steeper as the bed height decreased. Thesaturation time for the adsorption of Ni(II) increasedfrom 37 to 40 h as the bed height increased from 10 to30 cm, respectively. An increase in the amount of adsorbent

usually resulted in higher adsorption capacities andsaturation time. The maximum bed capacities were60.311, 29.955, and 22.098mg=g for bed heights of 30,20, and 10 cm, respectively. The Ni(II) uptake capacity ofthe EAFS was found to increase with increasing bed heightas more bonding sites are available for adsorption (20).An increase in bed height resulted in longer contact timesof Ni(II) with the adsorbent, thereby increasing the contacttime required for the solution to pass through the column.As a result, the adsorption capacity increased.

Bed-depth saturation time (BDST) is a simple modelbased on a linear relationship between the bed height (Z)and the saturation time (t), as expressed in Eq. (16):

t ¼ N0

C0FZ � 1

KaC0ln

C0

C� 1

� �ð16Þ

where C is the breakthrough metal concentration (mg=L),N0 is the adsorption capacity of the bed (mg=L), F is thelinear velocity (mL=min), and Ka is the rate constant (L=mg-min). The column saturation time (t) was selected asthe time required for the effluent Ni(II) to reach a concen-tration of 100mg=L. A plot of t against Z at a flow rate-7mL=min is shown in Fig. S5. The linear representationobtained (R2¼ 0.9959) validated the BDST model for thepresent system (30).

The adsorption capacity of the bed per unit of bed vol-ume (N0) and Ka were calculated from the slope and theintercept of the linear plot, respectively, with the initialconcentration C0 and the linear velocity (F) remaining con-stant during the experiment. Ka and N0 were determined as0.3651L=mg-min and 1336.9mg=L, respectively (Table 4).

TABLE 3Kinetics parameters for Ni(II) adsorption on EAFS

Kinetics equations

Pseudo-first-order Pseudo-second-order

Co

(mg=L)qe,exp(mg=g)

qe,cal(mg=g)

k1(1=min) R2

qe,cal(mg=g)

k2(g=mg-min) R2

50 11.2 10.5 0.00023 0.951 11.9 0.000068 0.996100 24.9 24.7 0.00023 0.893 25.6 0.000033 0.974200 47.4 48.0 0.00023 0.997 47.6 0.000051 0.997300 70.5 62.4 0.00023 0.971 71.4 0.000041 0.999400 88.7 81.7 0.00023 0.976 90.9 0.000026 0.999500 109.6 97.1 0.00023 0.938 111.1 0.000021 0.999600 131.3 112.7 0.00023 0.916 125.0 0.000024 0.999700 135.5 113.0 0.00046 0.835 142.9 0.000019 0.999800 147.8 115.2 0.00069 0.749 147.1 0.000012 0.971900 154.3 130.5 0.00092 0.912 151.5 0.000011 0.9741000 160.9 147.2 0.00138 0.928 161.3 0.000013 0.974

394 M. YUSUF ET AL.

Dow

nloa

ded

by [

Uni

vers

iti P

utra

Mal

aysi

a] a

t 20:

26 1

4 Fe

brua

ry 2

014

As shown in Table 5, with an increase in bed depth, the Ve

increased due to the more contact time, and removalefficiency of (Y) of Ni(II) increase from 21.718% to23.752% as well. The increase in Ni(II) adsorption withthe increasing bed depth in the fixed bed column mightbe due to the increase adsorbent doses in larger bed, which

provided a greater service area (or adsorption sites). Thebreakthrough time (tb) obtained were 6, 9, and 9.5,respectively.

Effect of Flow Rate

The effect of the flow rate on the adsorption of Ni(II) onEAFS was examined by varying the flow rate from 5 to15mL=min, while keeping the initial Ni(II) concentration(100mg=L) and bed height (30 cm) constant. These con-ditions allowed operation of the bed depth at its maximumcapacity with a minimum flow rate (31). In contrast to thebed height results, the column performed well at the lowestflow rate. The breakthrough curves for Ni(II) adsorptionon EAFS at different flow rates in a fixed-bed columnsystem are shown in Fig. 5b. The uptake decreases with timeas these active sites become occupied. The breakthroughtimes were found to shorten (i.e., higher breakthroughslopes) with the flow rate. An increase in service timefrom 42 to 46h led to a decrease in the flow rate as themetal solution interacted more strongly with the adsorbentas a result of the lower saturation time.

Breakthrough and exhaustion times were observed inthe profile at a high flow rate of 15mL=min. The flow rate

TABLE 4BDST model constants for Ni(II) adsorption on EAFS

C=C0 Slope Intercept Flow rate (mL=min) Ka (L=mg-min) No (mg=L) R2

0.991 13.5 1420 7.00 0.3651 1336.9 0.9959

FIG. 5. (a) Breakthrough curves for the adsorption of Ni(II) on EAFS at

different bed heights (flow rate 7mL/min) and (b) Breakthrough curves for

the adsorption of Ni(II) on EAFS at different flow rate with 30 cm bed depth.

TABLE 5Parameters in fixed-bed column for Ni(II) adsorption onto

EAFS

C0

(mg=L)Z

(cm)Q

(mL=min)tb(h)

te(h)

qe(mg=g)

Ve(L)

Y(%)

100 10 7 6 26 22.098 10.92 21.718100 20 7 9 28 29.955 11.76 21.916100 30 7 9.5 30.5 60.311 12.81 23.752100 30 5 13 29 69.984 8.7 24.293100 30 10 10 23.5 59.343 14.1 12.710100 30 15 8.5 22 52.943 19.8 8.075

TABLE 6Thomas model constants for Ni(II) adsorption onto EAFS

Flow rate(mL=min)

KTh�10�3

(mL=min-mg)qo

(mg=g) R2

15 3.240 52.943 0.976710 3.183 59.343 0.98045 2.808 69.984 0.9509

ADSORPTION OF NICKEL ON ELECTRIC ARC FURNACE SLAG 395

Dow

nloa

ded

by [

Uni

vers

iti P

utra

Mal

aysi

a] a

t 20:

26 1

4 Fe

brua

ry 2

014

was also found to strongly influence uptake capacity ofNi(II) on EAFS (69.984, 59.343, and 52.943mg=g at 5,10, and 15mL=min, respectively). The Thomas modelwas applied to data given as:

lnC0

C� 1

� �¼ kThq0M

Q� kThC0

Ft ð17Þ

where C is the effluent concentration (mg=L) at time t, C0 isthe influent concentration (mg=L), KTh is the Thomas rateconstant (mL=min-mg), q0 is the maximum slag adsorptioncapacity (mg=g), M is the mass of the adsorbent (g), Q isthe volumetric flow rate (mL=min). KTh and q0 can beobtained from the slope of the plot of ln c0

c � 1� �

againstt as shown in Fig. S6. The KTh values are listed in Table 6.

Table 6 showed a decrease in KTh values with a decreasein flow rate. Optimum adsorption capacity (69.983mg=g)was observed at 5mL=min flow rate. The R2 values showedthat the model provides a good fit to experimental dataover the entire range of flow rates examined. Also, fromTable 5, at a higher flow rate the Ni(II) did not haveenough time to diffuse into the pores of the Electric ArcFurnace Slag (EAFS) and they passed the column fastbefore equilibrium occurred. Hence, an early breakthroughcurve occurred, resulting in low bed adsorption capacityand low removal efficiency. When the linear flow ratedecreased, the contact time between Ni(II) and EAFS inthe column was longer, which favored a better adsorptioncapacity of EAFS and relative high removal efficiency(Table 5).

Comparison of the EAFS Ability to Remove Ni(II) Ionwith Other Adsorbent

Different types of adsorbents have been tested for theirability to remove Ni(II) ion. A comparative assessment ofthe Ni(II) ion adsorption capacity of various types ofadsorbents is provided in Table 7. From the table it shows

EAFS has good adsorption capacity when compared withother adsorbent and this indicates that it could beconsidered a promising material for the removal of thismetal ion from aqueous solutions.

CONCLUSIONS

The adsorptive potential of EAFS for Ni(II) from aque-ous medium was investigated by both the batch and thecolumn process. Optimum adsorption of Ni(II) on EAFSwas observed at pH: 8. Kinetics studies showed applica-bility of pseudo-second-order kinetics model confirmingchemisorption process. The adsorption of Ni(II) on EAFSwas monolayer adsorption as inferred by the applicabilityof Langmuir model. Column studies at various bed heightsrevealed Ka� 0.3651L=mg-min and No� 1336.9mg=L.The adsorption capacity was found to be strongly depen-dent on the influent flow rate, concentration, and columnbed height. The breakthrough time decreased with increasein time. The Thomas model was applied to predict thebreakthrough curves showed good fit to experimental data.Hence, it could be concluded that EAFS, a readilyavailable low-cost residue from a steel plant, could bea suitable alternative for the removal of Ni(II) from aque-ous solution.

FUNDING

The authors thank the Faculty of Engineering at theUniversiti Putra Malaysia for their financial support.

SUPPLEMENTAL MATERIAL

Supplemental data for this article can be accessed on thepublisher’s website.

REFERENCES

1. Ewecharon, A.; Thiravetyan, P.; Wendel, E. (2009) Nickel adsorption

by solution polycrylate grafted activated carbon. J. Hazard. Mater.,

171: 335–339.

TABLE 7Comparison of the adsorption capacity at equilibrium of different adsorbents for Ni(II) removal

Adsorbent q (mg=g) Experimental conditions Reference

Zeolite X 46 pH¼ 5.2, T¼ 293–313K, Dose¼ 2–3.5 gm, Initial conc¼ 100mg=L 32Grape stalk 14 pH¼ 5.2, Dose¼ 1 g, Initial conc¼ 100mg=L, T¼ 20–60�C 33Brown algae 56 pH¼ 4–7, T¼ 293K, Conc¼ 6–201mg=L 34Silylated clay 5.3 pH¼ 6.0, Initial conc¼ 100mg=L, T¼ 373 k 35Glycidyl methacrylate 48 pH¼ 6.0–7.0, dose¼ 0.2, T¼ 23�C, Con¼ 20–4000mgdm�3 3Irish peat moss 14.5 pH¼ 2–8, T¼ 298K, Initial conc¼ 100mg=L 36Crab shell 25 pH¼ 4.5, Initial conc¼ 100mg=L, Dose¼ 21.5 g 37Lotus stalk 31.45 pH¼ 6.5, dose¼ 0.1 g, Initial conc¼ 100mg=L, T¼ 298K 38Calcined Bofe betonite clay 1.91 pH¼ 5.3, T¼ 293K, Initial conc¼ 50mg=L, Dose¼ 1 g 39EAFS 160.92 pH¼ 8, Initial conc¼ 1000mg=L, T¼ 25�C, Dose¼ 1 g This study

396 M. YUSUF ET AL.

Dow

nloa

ded

by [

Uni

vers

iti P

utra

Mal

aysi

a] a

t 20:

26 1

4 Fe

brua

ry 2

014

2. Khan, M.A.; Ngabura, M.; Choong, T.S.Y.; Masood, H.; Chuah,

L.A. (2012) Biosorption and desorption of Nickel on oil cake: Batch

and column studies. Bioresour. Technol., 103: 35–42.

3. David, W.O. (2006) A modified cellulose adsorbent for the removal

of nickel (II) from aqueous solution. J. Chem. Tech. & Biotech., 81:

1820–1828.

4. Sheng, P.X.; Ting, Y.P.; Chen, J.P.; Hong, L. (2004) Sorption of

lead, copper, cadmium, zinc, and nickel by marine algal biomass:

Characterization of biosorptive capacity and investigation of

mechanisms. J. Colloid Interf. Sci., 275: 131–141.

5. MINAS pollution control acts, rules, notification issued there under

Central Pollution Control Board, Ministry of Environment and

Forests, Government of India, New Delhi, September 2001.

6. Kwon, J.S.; Yun, S.T.; Lee, J.H.; Kim, S.O.; Jo, H.Y. (2010) Removal

of divalent heavy metals (Cd, Cu, Pb, and Zn) and arsenic (III) from

aqueous solutions using scoria: Kinetics and equilibria of sorption.

J. Hazard. Mater., 174: 307–313.

7. Espinoza-Quinones, F.R.; Modenes, A.N.; Theodoro, P.S.; Palacio,

S.M.; Trigueros, D.E.G.; Borba, C.E.; Abugderah, M.M.; Kroumov,

A.D. (2012) Optimization of the iron electro-coagulation process of

Cr, Ni, Cu, and Zn galvanization by-products by using response

surface methodology. Sep. Sci. & Tech., 47 (5): 688–699.

8. Khan, M.A.; Lee, S.-H.; Kang, S.; Paeng, K.-J.; Lee, G.; Oh, S.-E.;

Jeon, B.-H. (2011) Adsorption studies for the removal of methyl

tert-butyl ether on various commercially available GACS from an

aqueous medium. Sep. Sci. & Tech., 46 (7): 1121–1130.

9. Khan, M.A.; Ahn, Y.-T.; Kumar, M.; Lee, W.; Min, B.; Kim, G.;

Cho, D.-W.; Park, W.-B.; Jeon, B.-H. (2011) Adsorption studies for

the removal of nitrate using modified lignite granular activated car-

bon. Sep. Sci. & Tech., 46 (16): 2575–2584.

10. Altun, I.; Yilmaz, I. (2002) Study on steel furnace slags with high MgO

as additive in Portland cement. Cem. and Concr. Res., 32: 1247–1249.

11. Stock, A.F.; Colin, M.I.; Taylor, I.F. (1996) Skidding Characteristics

of Pavement Surface Incorporating Steel Slag Aggregates, Transport

Research Record 1545, National Research Council, Washington,

DC, pp 35–40.

12. Emery, J.J. (1982) Slag Utilization in Pavement Construction.

Extending Aggregate Resources ASTM, STP 774 American Society

for Testing and Materials, pp 95–118.

13. Beh, C.; Chuah, L.; Choong, T.S.Y.; Kamarudzaman, M.Z.B.;

Abdan, K. (2010) Adsorption study of electric arc furnace slag for

the removal of manganese from solution, Am. J. Appl. Sci., 7: 442–446.

14. Dimitrova, S.V. (2002) Use of granular slag column for lead removal.

Water Res., 36: 4001–4008.

15. Meng, M.; Feng, Y.; Zhang, M.; Liu, Y.; Ji, Y.; Wang, J.; Wua, Y.;

Yan, Y. (2013) Highly efficient adsorption of salicylic acid

from aqueous solution by Wollastonite-based imprinted adsorbent:

A fixed-bed column study. Chem. Eng. J., 225:331–339.

16. Goertzen, S.L.; Theriault, K.D.; Oickle, A.M.; Tarasuk, A.C.;

Andreas, H.A. (2010) Standardization of the Boehm titration. Part

I. CO2 expulsion and endpoint determination. Carbon, 48: 1252–1261.

17. Eddy, M. (1991) Wastewater Engineering: Treatment, Disposal, and

Reuse; McGraw-Hill: New York.

18. Malkoc, E.; Nuhuoglu, Y. (2006) Removal of Ni(II) from aqueous

solution using waste tea factory: Adsorption on a fixed-bed column.

J. Hazard. Mater., 40: 326–336.

19. Yanli, M.; Hongwei, H.; Yongsheng, Y.; Hongxiang, O. (2012)

A novel extracellular biopolymer produced from pseudonas fluoresces

C-2: Preparation, characterization, and adsorption behaviors toward

nickel (II) ions in aqueous solution. Inter. J. Mater. Struc. Intergr.,

6: 74–93.

20. Malkoc, E. (2006) Ni (II) removal from aqueous solutions using cone

biomass of Thuja orientalis. J. Hazard. Mater., 137: 899–908.

21. Kumar, P.S.; Kirthika, K. (2009) Equilibrium and kinetic study of

adsorption of nickel from aqueous solution onto bael tree leaf

powder. Eng. Sci. Technol., 40: 351–363.

22. Ozer, A.; Ozer, D. (2004) The adsorption of copper (II) ions on to

dehydrated wheat bran (DWB): Determination of the equilibrium

and thermodynamic parameters. Process Biochem., 39: 2183–2191.

23. Ozacar, M.; Sengil, A. (2005) Adsorption of complex dyes from

aqueous solution by pine sawdust. Bioresour. Technol., 7: 791–795.

24. Ofomaja, A.E. (2010) Intraparticle diffusion process for lead (II)

biosorption onto mansonia wood Sawdust. Bioresour. Technol., 101:

5868–5876.

25. Langmuir, I. (1916) The adsorption of gases on plane surface of glass,

mica and platinum. J. Amer. Chem. Soc., 40: 1361–1403.

26. Freundlich, H. (1985) Uber die. Adsorption in lunsungen. J. Phys.

Chem., 57: 347–470.

27. Redlich, O.; Peterson, D.L. (1959) A useful adsorption isotherm.

J. Phys. Chem., 63: 1024–1026

28. Lagergren, S. (1898) About the theory of so-called adsorption of

soluble substances. Handlingar Band, 24: 1– 39.

29. Ho, Y.S.; Ng, J.C.Y.; McKay, G. (2000) Kinetics of pollutant

sorption by biosorbents: A review. Sep. Purif. Methods, 29: 189–232.

30. Sivakumar, P.; Palanisamy, P.N. (2009) Packed bed column studies

for the removal of Acid blue and Basic red 29 using non-conventional

adsorbent. Ind. J. Chem. Eng., 16: 301–307.

31. Fox, C.R.; Kennedy, D.C. Conceptual Design of Adsorption Technology:

A Step-by-step Approach to Process Evaluation and Application;

Slejko F. L. (Ed.), pp 91–165. Marcel Dekker: New York.

32. Surinder, S.; Lokesh, K.V.; Sambi, S.S.; Sharma, S.K. (2008)

Adsorption behavior of Ni(II) from water onto Zeolite X: Kinetic

and Equilibrium studies. Proceeding of the World Congress on

Engineering and Computer Science, pp 978–988.

33. Cesar, V.; Jaime, A.; Arevato, E. (2010) Modeling of the Ni (II)

removal from aqueous solution onto grape stack wastes in fixed-bed

column. Journal of Hazardous Material, 174: 144–150.

34. Pahlavanzadeh, H.; Keshtkar, A.R.; Safdari, J.; Abadi, Z. (2010)

Biosorption of nickel (II) from aqueous solution by brown

algae: Equilibrium, dynamic and thermodynamic studies. Journal of

Hazardous Materials, 175 (1–3): 304–310.

35. Wagner, A.C.; Vignado, C.; Juliane, F. (2008) Ni (II) removal from

aqueous effects by silylated clays. Journal of Hazardous Material,

153: 1240–1247.

36. Sen Gupta, B.; Curran, M.; Hasan, S.; Ghosh, T.K. (2009)

Adsorption characteristics of Cu and Ni on Irish peat moss. Journal

of Environmental Management, 90 (2): 954–960.

37. Vijayaraghavan, K.; Jegan, J.; Palanivelu, K.; Velan, M. (2004)

Removal of nickel (II) ions from aqueous solution using crab

shell particles in a packed bed up-flow column. Journal of Hazardous

Materials, 113 (1–3): 223–230.

38. Littui, H.; Yuan yuan, S.; Tao, Y.; Li, L. (2011) Adsorption behavior

of Ni (II) on Lotus stalk derived active carbon by phosphoric acid

activation. Desalination, 268: 12–19.

39. Vieira, M.; Neto, A.; Gimenes, M.; da Silva, M. (2010) Sorption

kinetics and equilibrium for the removal of nickel ions from aqueous

phase on calcined Bofe bentonite clay. Journal Hazardous Material,

177 (1–3): 362–371.

ADSORPTION OF NICKEL ON ELECTRIC ARC FURNACE SLAG 397

Dow

nloa

ded

by [

Uni

vers

iti P

utra

Mal

aysi

a] a

t 20:

26 1

4 Fe

brua

ry 2

014