A

sNUkfTm1©

K

1

ltcshotmmmtpptr

0d

Available online at www.sciencedirect.com

Colloids and Surfaces B: Biointerfaces 62 (2008) 97–104

Biosorption of Ni(II) from aqueous solutions by living andnon-living ureolytic mixed culture

Mustafa Isik ∗Aksaray University, Engineering Faculty, Environmental Engineering Department, 68100 Aksaray, Turkey

Received 25 July 2007; received in revised form 13 September 2007; accepted 19 September 2007Available online 25 September 2007

bstract

The present study explores the ability and the comparison of living and non-living ureolytic mixed culture (UMC) to remove Ni(II) from aqueousolution. Time dependency experiments for the Ni(II) uptake showed that adsorption equilibrium was reached almost 110 and 60 min after additioni(II) of 100 mg/L. The kinetic data were analyzed in term of pseudo-first-order and pseudo-second-order expressions. Ni(II) sorption of livingMC was appropriate with pseudo-first-order kinetic (k1 = 2.15 h−1, R2 = 0.93) while non-living UMC sorbed Ni(II) with respect to second-orderinetics (k2 = 1.64 g/mg h, R2 = 0.98). Also, comparison between the biosorption capacity of untreated living and non-living biomass was conductedor removal of Ni(II). The biosorption process was investigated in equilibrium batch tests for Langmiur, Freundlich and Temkin isotherm models.

he data pertaining to the sorption dependence upon Ni(II) ion concentration ranged from 5 to 320 mg/L could be fitted to a Freundlich isothermodel. The capacity constants K of Freundlich model for living and non-living UMC were 1.55 and 0.38 mg/g, respectively; the affinity constants

/n were 0.47 and 0.75, respectively. Based on the results, the UMC appear to be a potential biosorbent for removal of Ni(II) from wastewater. 2007 Elsevier B.V. All rights reserved.

ao

eiimobtobdnp

eywords: Biosorption; Living; Non-living; Ureolytic; Nickel; Kinetics

. Introduction

Enhanced industrial activity after the industrial revolution hased to the discharge of chemicals, which causes environmen-al and public health problems. The important group of toxichemicals is heavy metals, due to their high toxicity, pose aerious threat to biota and the environment. The presence ofeavy metals in the environment is of major concern becausef their extreme toxicity and tendency for bioaccumulation inhe food chain even in relatively low concentrations [1,2] Heavy

etals pollute the environment from various industries such asetal plating, electroplating, mining, ceramic, batteries, pig-ent manufacturing [3]. Although there are many methods for

he removal of metal ions from solutions, such as chemicalrecipitation, complexation, solvent extraction and membrane

rocesses, biosorption processes show many advantages overhese methods. It is selective, effective and cheap and is able toemove very low levels of heavy metals from solutions [4]. In

∗ Tel.: +90 382 2150953/132; fax: +90 382 2150592.E-mail address: misik@aksaray.edu.tr.

ane

mom

927-7765/$ – see front matter © 2007 Elsevier B.V. All rights reserved.oi:10.1016/j.colsurfb.2007.09.022

ddition, most of these processes are not eco-friendly becausef the production of sludge causing a solid disposal problem [5].

Recently, bioadsorbents have emerged as an eco-friendly,ffective and low cost material option. These bioadsorbentsnclude some agricultural wastes, fungi, algae and bacteria. Stud-es using bioadsorbents have shown that both living and dead

icrobial (non-living) cells are able to adsorb metal ions andffer potential inexpensive alternative to conventional adsor-ents [6]. However, living cells are subject to toxic effect ofhe heavy metals, resulting in cell death. Moreover, living cellsften require the addition of nutrients and hence increase theiochemical oxygen demand (BOD) and/or chemical oxygenemand (COD) in the effluent. For these reasons, the use ofon-living biomaterials or dead cells as metal-binding com-ounds has been gaining advantage because toxic ions do notffect them. In addition, dead cells require less care, mainte-ance and they are cheaper. Furthermore, dead biomass can beasily regenerated and reused [6].

Microbial metal uptake generally involves the rapid,etabolism-independent uptake of metals to cell walls and

ther external surfaces (passive uptake), followed by a slow,etabolism-dependent transport across the cell membrane

9 B: Bi

(esoipi

Niadug[hps[

lcheTcctmsbsf

2

2

ep[e(CCSC1icftaad(

oa

2

dtHcmiws(cb

imfbNiwiNsrebwb

2

igms((wdw

3

3

t

8 M. Isik / Colloids and Surfaces

active uptake). In most studies, it was shown that passive uptake,specially the complexation by extracellular polymeric sub-tance is the dominant mechanism in metal removal [7,8]. On thether hand, for some metals, such as nickel, active uptake wasndicated to be important at low concentrations [9]. Extracellularolymeric substances are reported to have very important rolesn complexing and removing heavy metals from solution [1,2].

Industrial wastewater containing Ni(II) is common becausei(II) is used in a large number of industries, such as electroplat-

ng, batteries manufacturing, mine, metal finishing and forgingnd so on. The Ni(II) concentration in wastewater from minerainage, tableware plating, metal finishing and forging wasp to 130 mg/L [10]. Nickel and their compounds are carcino-enic and may constitute danger to human being and other lives11]. Ni(II) which is belong to the so-called “essential” metalsas been identified as a component in a number of enzymes,articipating in important metabolic reactions, such as ureoly-is, hydrogen metabolism, methane biogenesis and acidogenesis12].

Nowadays, ureolytic mixed culture has been produced in ouraboratories in order to investigate biocatalytic Ca removal fromalcium-rich wastewaters. The process bases on microbial ureaydrolysis, in which 1 mol of urea is hydrolyzed by the ureasenzyme to 2 mol of ammonium and 1 mol of carbon dioxide.hese products can subsequently react to form ammonium andarbonate ions, which, in the presence of soluble calcium ions,an react and precipitate as CaCO3 [13]. The present work inves-igates the potential use of the living and non-living ureolytic

ixed culture (UMC) as metal sorbent for nickel from aqueousolution. In addition, ureolytic mixed culture was chosen as aiosorbent because of the relative lack of information about itsorption ability and the potential use of waste sludge producedrom biocatalytic calcium removal process.

. Material and methods

.1. Feed solution and sludge production

The synthetic wastewater was prepared as taking consid-ration of simulating medium strength municipal, and lineraper manufacturing wastewater proposed by Holakoo et al.14] and Kim et al. [15], respectively. It was as following min-ral medium in mg/L: glucose-COD (750); urea (600); CaCl217); MgSO4·7H2O (1541); KH2PO4 (132); FeCl3·6H2O (19);uSO4·5H2O (0.118); MnSO4·H2O (0.123); ZnCl2 (0.229),oCl2·6H2O (0.404); Na2CO3 (477) and NaHCO3 (378).ulphuric acid was used to maintain a pH of 7.00 ± 0.10.omposition of synthetic wastewater resulted in COD/N/P of00/37/4 ratio. Urea was used in higher concentration than thats necessary for growth in the medium to serve microbiologicalarbonate precipitation (MCP) process. Culture was obtainedrom a fed-batch reactor, which is already settled in the labora-ory, receiving a synthetic wastewater described above and used

s the sorbent for the batch test. The sludge retention time (SRT)nd biomass concentration in this reactor were approximately 10ays and 2000 mg/L as mixed liquor volatile suspended solidsMLVSS), respectively. The sludge had an ability of conversion

tsto

ointerfaces 62 (2008) 97–104

f all urea to ammonia in a day. Dissolved oxygen was measuredbove 2 mg/L during sludge production.

.2. Experimental procedure

UMC taken from reactor, operated with sludge age of 10ays in the laboratory, was centrifuged, washed and resuspendedwice and more with distilled water for stock living sludge.ence, the cells were metabolically inactive, but referred living

ulture in this study. For the non-living culture, after above treat-ent, sludge inactivated using 1% formaldehyde [16], then dried

n an oven at 70 ◦C for 24 h and finally, for adsorption studies, aeighted amount pretreated cells were resuspended for stock

olution of 6 g/L. Sorption kinetics and sorption equilibriumisotherm) tests were run at 1 g/L of living and non-living UMConcentrations, which were transferred from stock solution ofiomass.

Batch shake flask experiments were performed a rotaryncubator-shaker at 150 rpm using 1000 mL and 250 mL Erlen-

eyer flask containing 450 mL and 75 mL of 100 mg Ni(II)/Lor batch kinetic studies, and different Ni(II) concentrationsetween 5 and 320 mg/L for isotherm studies, respectively.ickel ions, in form of NiCl2·6H2O, were prepared by dissolv-

ng its corresponding chlorine salt in distilled water. The pHas adjusted to 6 by using 1 M NaOH and H2SO4 in all exper-

ments since pH 6 is reported to be the optimum near pH fori(II) biosorption [17–19]. The flasks were incubated in a rotary

haker at 20 ◦C for all experiments. A control flask free of bacte-ia culture with 100 mg/L Ni(II) ions was used to determine thextent of non-adsorptive Ni(II) removal from the solution. Allatch biosorption tests were performed with 5 h of contact time,hich guarantee to reach to the equilibrium between Ni(II) andiosorbents.

.3. Analytical methods

Samples were withdrawn from the mixed liquor medium afterncubation time, and were centrifuged at 2375 relative centrifu-al force (RCF) for 10 min to remove suspended solids from theedium. Clear supernatants were analyzed for Ni(II) ions with

tandard kits (Merck-Spectroquant) and a spectrophotometerThermoSpectronic AQUAMATE). Volatile suspended solidsVSS), suspended solids (SS) and sludge volume index (SVI)ere analyzed as specified in Standard Methods [20]. pH andissolved oxygen (DO) were measured by using apparatusesith the relevant probes (WTW, Germany).

. Results and discussion

.1. The effect of contact time and kinetic studies

As the adsorption process proceeds, the sorbed solute tendso desorb back into the solution. Eventually the rates of adsorp-

ion and desorption will attain an equilibrium state. When theystem reaches the sorption equilibrium, no further net adsorp-ion occurs. The time at which the adsorption equilibrium willccur was determined. Removal kinetics of Ni(II) by living and

M. Isik / Colloids and Surfaces B: Biointerfaces 62 (2008) 97–104 99

a) an

ntbi(tsacUcam

tN

q

q

wattS

scd6r

aol

wAcat[st

ttoctcsf

l

wa

Fig. 1. Effect of contact time on Ni(II) uptake by living (

on-living UMC were firstly studied in batch reactors with ini-ial metal concentration of 100 mg/L. More than 87% of Ni(II)iosorption was completed within the first 60 min by using liv-ng UMC; however, all of it apparently was uptaken in this time60 min) by using non-living UMC. The data obtained fromhe biosorption of Ni(II) ions on the ureolytic mixed culturehowed that a contact time of 110 and 60 min was sufficient tochieve equilibrium and the adsorption did not change signifi-antly with further increase in contact time, living and non-livingMC, respectively. Therefore, the uptake and unadsorbed Ni(II)

oncentrations at the end of the reached equilibrium are givens the equilibrium values, respectively (qe, mg/g and Ce,g/L).The amount of Ni(II) adsorbed by UMC was calculated from

he differences between Ni(II) quantity added to the biomass andi(II) content of the supernatant using the following equations:

= C0 − Ct

M(1)

e = C0 − Ce

M(2)

here q and qe are the solid phase Ni(II) ion concentrations atnytime (t) and at the equilibrium (mg/g), and C0, Ce and Ct arehe initial, equilibrium and anytime (t) metal concentrations inhe solution (mg/L), respectively. M is the quantity of UMC (gS/L).

Fig. 1 depicts variation of aqueous and solid phase (bio-orbed) Ni(II) concentrations by time. Solid phase Ni(II)

oncentration increased, the aqueous phase concentrationecreased with time and reached an equilibrium after 110 and0 min of operation period for living and non-living UMC,espectively. The equilibrium biosorption capacities were 16.81

fic

−

Fig. 2. Plots of ln(1 − q/qe) vs. time according to the pseudo-first-or

d non-living (b) UMC for 100 mg/L Ni and 1 g/L UMC.

nd 13.50 mg Ni(II)/g SS while equilibrium concentrations werebtained as 83.20 and 86.50 mg Ni(II)/L for living and non-iving UMC, respectively.

Ni(II) uptake by non-living UMC was taken place within 1 hhile living UMC sorbed Ni(II) about 2 h as shown in Fig. 1.s Ni(II) uptake is metabolism-independent passive binding to

ell walls (adsorption), and other external surfaces, is gener-lly considered as a rapid process [17]. These results agree withhe two-stage adsorption mechanism proposed in the literature8]. The first rapid uptake completed within the first hour corre-ponds to the passive uptake, while, the second slow stage seemso be the metabolism-dependent intracellular uptake.

The prediction of adsorption rate gives important informa-ion for designing batch adsorption systems. Information onhe kinetics of solute uptake is required for selecting optimumperating conditions for full-scale batch process. These modelsorrelate metal uptake rate, which are important in predictinghe reactor volume. Two different kinetic models were used fororrelation of biosorption data. The pseudo-first- (Eq. (3)) andecond-order (Eq. (4)) kinetic model [21,22] has the followingorms:

n

(1 − q

qe

)= k1t (3)

t

q= 1

k2q2e

+ t

qe(4)

here q and qe are the solid phase Ni(II) ion concentrations atny time (t) and at the equilibrium (mg/g), respectively, k1 the

rst-order rate constant (h−1) and k2 is the second-order rateonstant (g/mg h).

When the experimental data was plotted in form ofln(1 − q/qe) versus time, a straight line was obtained as shown

der adsorption kinetics for living (a) and non-living (b) UMC.

100 M. Isik / Colloids and Surfaces B: Biointerfaces 62 (2008) 97–104

-order

ir(ifpf

vasar(clpft

firhrac

3

ca

pi

tNdbftuuttoaclcwhco

tTvib

Fig. 3. Plots of t/q vs. time according to the pseudo-second

n Fig. 2a and b. From the slope of the lines, the first-orderate constants were found as k1 = 2.15 (R2 = 0.93) and 4.67 h−1

R2 = 0.90) for living and non-living UMC, respectively. For liv-ng UMC, the linear regression correlation coefficient values R2

ound as 0.93, which shows that this model can be applied toredict the adsorption kinetic model as compared with R2 (0.90)or non-living UMC.

When the data presented in Fig. 1 was plotted in form of t/qersus time, a straight line was obtained as shown in Fig. 3and b. From the slope and the intercept of the line in Fig. 3, theecond-order rate constants were found as k = 0.03 (R2 = 0.31)nd 1.64 (g/mg h) (R2 = 0.98) for living and non-living UMC,espectively. The value of correlation coefficient was very highR2 > 0.98) and the theoretical qe,cal value (13.72 mg/g) wasloser to the experimental qe,exp(13.50 mg/g) value for non-iving UMC. In the view of these results, it can be said that theseudo-second-order kinetic model provided a good correlationor the biosorption of Ni(II) onto non-living UMC in contrast tohe pseudo-first-order model.

Ni(II) sorption of living UMC was appropriate with pseudo-rst-order kinetic while non-living UMC sorbed Ni(II) withespect to second-order kinetics. For non-living UMC, Theigher R2 values confirm that the adsorption data are wellepresented by pseudo-second-order kinetics and supports thessumption behind the model that the adsorption is due tohemisorption [23].

.2. Isotherm studies

The adsorption isotherm models are characterized by certainonstants, the values of which express the surface propertiesnd affinity of the biosorbent and can also be used to com-

itto

Fig. 4. Effect of initial metal concentration on the adsorpt

adsorption kinetics for living (a) and non-living (b) UMC.

are biosorptive capacity of the biomass for different metalons.

Fig. 4 shows the effect of metal ion concentration onhe adsorption of Ni(II) by UMC. The data shows that thei(II) uptake increases and the percentage adsorption of Ni(II)ecreases with increase in metal ion concentration for twoiosorbent, except with initial Ni(II) concentration of 320 mg/Lor UMC. The plateau of Ni(II) uptake for UMC can speculatedhat all of binding sites of UMC was completely used and sat-rated at Ni(II) concentration of 320 mg/L. Increases in Ni(II)ptake is a result of increase in the driving forces, i.e. concentra-ion gradient. Though an increase in metal uptake was observed,he decrease in percentage adsorption may be attributed to lackf sufficient surface area to accommodate much more Ni(II)vailable in the solution. The percentage adsorption at higheroncentration levels shows a decreasing trend whereas the equi-ibrium uptake of Ni(II) displays an opposite trend. At loweroncentrations, all Ni(II) ions present in solution could interactith the binding sites and thus the percentage adsorption wasigher than those at higher Ni(II) ion concentrations At higheroncentrations, lower adsorption yield is due to the saturationf adsorption sites [24].

The uptake capacity of the viable biomass was higher thanhat of the non-living biomass at lower concentrations of Ni(II).wo different processes are involved in metal ion uptake byiable and non-living biomass. The first uptake process isndependent of cell metabolic activity, and is referred to asiosorption or passive uptake. It involves the binding of metal

ons to the cell surface. The second uptake process involves,he uptake of metal ions into the cell across the cell membrane,his process is referred to as intracellular uptake, active uptaker bioaccumulation [11]. The metal uptake by active mode has

ion of Ni(II) for living (a) and non-living (b) UMC.

M. Isik / Colloids and Surfaces B: Biointerfaces 62 (2008) 97–104 101

s for

bMndtbtc

vvtwabdlawuagphtboo

obie

fimb

q

wbtL

Tlifm

stT

q

wctF

Fig. 5. Langmuir adsorption isotherm

een observed for metals such as Cu, Cd, Ni, Zn, Mn, Sr, Co,g and Ca [25]. The first process occurs for both viable and

on-viable biomass, the second process, which is metabolism-ependent, happens only in viable biomass. For viable biomass,he metal uptake is also facilitated by the production of metal-inding proteins [25]. Therefore, at low initial concentrations,he higher sorption capacity of the living UMC observed in Fig. 4ould be due to the active uptake of the biomass.

Another possible explanation why sorption was higher for theiable biomass is the difference of surface area of viable and non-iable biomass [26]. The biomass was killed by basically dryinghe biomass in the oven and treatment with chloroform. Wateras the most abundant single compound in the cell and it makes

lmost 70% of the total weight of the cell. Thus by drying theiomass into the form of pellets the surface area of the cell wouldecrease, i.e. less area is exposed to the metal ions, therefore,ess metal uptake. The ability of bacterial cells to bind metals isssociated with the components of the cell itself in organic form,hich is about 75% of solid fraction in cell. It has been well doc-mented that several biomolecules, proteins, polysaccharidesnd extracellular polymers contain different surface functionalroups such as carboxylate, carbonyl, hydroxyl, amino, phos-horyl and sulphide groups [27]. The different functional groupsave a high affinity towards heavy metals that they can complexhe metal ions [28]. The heat [29] and chloroform treatment ofiomass could cause possibly a loss of some functional groupsn the cell surface, which would decrease the uptake capacityf the biomass.

The equilibrium established between adsorbed component

n the biosorbent and unadsorbed component in solution cane represented by adsorption isotherms. The most widely usedsotherm equation for modeling equilibrium is the Langmuirquation, which is valid for monolayer sorption on to a surface a

l

Fb

Fig. 6. Freundlich adsorption isotherms fo

living (a) and non-living (b) UMC.

nite number of identical sites [30]. The Langmuir [31] sorptionodel was chosen for the estimation of maximum Ni(II) sorption

y the biosorbent. The Langmuir isotherm can be expressed as

e = qmbCe

1 + bCe(5)

here qm indicates the monolayer adsorption capacity of adsor-ent (mg/g) and the Langmuir constant b (L/mg) is related tohe energy of adsorption. For fitting the experimental data, theangmuir model was linearized as

1

qe= 1

qm+ 1

bqmCe(6)

he plots of 1/qe versus 1/Ce are presented in Fig. 5a and b,iving and non-living UMC, respectively. From the slope andntercept of the line in figures, the Langmuir constants wereound as qm = 12.58 and 7.41 mg/g and b = 0.080 and 0.059 L/g for living and non-living UMC.The Freundlich expression is an empirical equation based on

orption on a heterogeneous surface suggesting (as expected)hat binding sites are not equivalent and/or independent. [30].he Freundlich [32] model is represented by the equation:

e = KC1/ne (7)

here K (mg/g) is the Freundlich constant related to adsorptionapacity of adsorbent and n is the Freundlich exponent relatedo adsorption intensity. For fitting the experimental data, thereundlich model was linearized as follows:

n qe = ln K + 1

nln Ce (8)

ig. 6 depicts a plot of ln qe versus ln Ce for Ni(II) ioniosorption. From the slope and intercept of the line, the fol-

r living (a) and non-living (b) UMC.

102 M. Isik / Colloids and Surfaces B: Biointerfaces 62 (2008) 97–104

Fig. 7. Temkin adsorption isotherm for l

Table 1Isotherm constants for Ni(II) sorption on living and non-living UMC

Isotherm type Living UMC Non-living UMC

Langmuir isothermqm (mg/g) 12.58 7.41b (L/g) 0.080 0.059R2 0.88 0.69

Freundlich isothermK (mg/g) 1.55 0.38n 2.12 1.341/n 0.47 0.75R2 0.91 0.91

Temkin isotherm

l0

iei

q

wtt

e

q

TtTb

Tt(te

1ilcfhus

TC

B

NLNPCPSASBNL

n

AT (L/g) 0.48 0.15bT 657 460R2 0.86 0.75

owing Freundlich isotherm constants were found: K = 1.55 and.38 mg/g and n = 2.12 and 1.34 for living and non-living UMC.

Temkin isotherm assumes that the fall in the heat of sorptions linear rather than logarithmic, as implied in the Freundlichquation. The Temkin isotherm [33] has generally, been appliedn the following form:

e = RTln(AT Ce) (9)

bT

here AT (L/mg) and bT are Temkin isotherm constants. R ishe universal gas constant (8.314 J mol/K) and T is the absoluteemperature in K (293.15 K).

s

pb

able 2omparison of adsorption capacities of living and non-living UMC with literature

iosorbent Operation conditions

Initial pH T I

on-living mixed activated sludge 4.5 25 1iving activated sludge 5 30 non-living C. vulgaris, a green alga, 5 25 1. versicolor 5 25 na-anaerobic biomass 5.5 20 nseudomonas syringae n.a. 22 0treptomyces coelicolor 8 25 1rtrobacter sp. 5–5.5 30 ntreptomyces noursei 5.9 30 (acillus thuringiensis 6 35 1on-living UMC 6 20 1iving UMC 6 20 1

.a.: Not available.

iving (a) and non-living (b) UMC.

For fitting the experimental data, the Temkin model was lin-arized as follows:

e = RT

bTln AT + RT

bTln Ce (10)

he plots of qe versus In Ce are presented in Fig. 7a and b, respec-ively. From the slope and intercept of the line in figures, theemkin constants were found as AT = 0.48 and 0.15 L/mg, andT = 657 and 430 for living and non-living UMC, respectively.

The calculated results of the Langmuir, Freundlich, andemkin isotherm constants are given in Table 1. It is found

hat the biosorption of Ni(II) on the UMC were correlated wellR2 = 0.91 for living and non-living UMC, respectively) withhe Freundlich equation as compared to Langmiur, and Temkinquation under the concentration range studied.

From inspection of Table 1, The K value was found as.55 mg/g for living UMC and 0.38 mg/g for non-living UMC. Its observed by comparison of these results that Ni(II) is sorbed oniving UMC more than non-living UMC at the this experimentalonditions. It is also noted that the value of 1/n are 0.47 and 0.75or living and non-living UMC, respectively. The value of 1/nas been shown to be indicative of the sorption mechanism. Val-es of 1/n close to one are indicative of a constant partitioningorption mechanism, where sorbate can easily penetrate into the

orbent [34] for non-living culture as compared to living UMC.

The qmax values of the living and non-living UMC were com-ared with the metal adsorption capacities reported for otheriosorbent in Table 2. Living and non-living UMC has medium

q (mg/g) Reference

nitial C Biomass (g/L)

00.2 0.5 71.6 [30].a. 1 40.6 [6]00 1 42.3 [35].a. 0.6 37.0 [34].a. n.a. 26.0 [26]–12 0.28 6.0 [36]50 1 11.1 [37].a 1.4 13.0 [38]0.6–60) 3.5 0.8 [18]00 1 21.5 [17]00 1 13.50 Present study00 1 16.81 Present study

es B:

ac

4

b

•

•

•

•

•

A

((fC

R

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

M. Isik / Colloids and Surfac

ffinity for Ni(II) ion when compared with other bacteria andultures.

. Conclusions

Based on experimental results the following conclusions cane drawn:

Ni(II) biosorptions in excess of 16 and 13% were obtained for100 mg/L Ni(II) at the equilibrium times of 110 and 60 minfor living and non-living UMC, respectively.The pseudo-first-order and pseudo-second-order kinetic mod-els were used to describe the kinetic data for initial Ni(II)(100 mg/L) and biomass concentrations (1 g/L) and the rateconstants were evaluated. The used experimental data werefitted by the second-order kinetic model, which indicates thatchemisorption is the rate-limiting step for non-living UMCwhile Ni(II) sorption of living UMC was appropriate withpseudo-first-order kinetic.In the comparison of living and non-living UMC, non-livingbiomass showed relatively rapid sorption characteristics ofNi(II), but living UMC indicated higher sorption capacity.Biosorption isotherms were modeled with the Langmuir Fre-undlich, and Temkin isotherms. Based on the correlationcoefficients, Freundlich models better described the nickelbiosorption isotherms compared to Langmuir and Temkinmodels.It was found that the living and non-living UMC are a potentialadsorbent for removal of Ni(II) from aqueous solutions.

cknowledgments

The Turkish Scientific and Technical Research CouncilTUBITAK) and Turkish Government Planning InstitutionDPT) funded this study. The author would like to thank themor the financial support given to the project with grant numbersAYDAG 105Y262 and 2006K120880-1.

eferences

[1] P. Malakul, K.R. Srinivasan, H.Y. Wang, Metal adsorption and desorptioncharacteristics of surfactant-modified clay complexes, Ind. Eng. Chem.Res. 37 (1998) 4296–4301.

[2] D. Mohan, K.P. Singh, Single and multi-component adsorption of cadmiumand zinc using activated carbon derived from bagasse—an agriculturalwaste, Water Res. 36 (2002) 2304–2318.

[3] M. Iqbal, R.G.J. Edyvean, Biosorption of lead, copper and zinc ions on loofasponge immobilized biomass of Phanerochaete chrysosporium, Miner.Eng. 17 (2004) 217–223.

[4] A. Hammaini, A. Ballester, M.L. Blazquez, F. Gonzales, J. Munoz, Effectof the presence of lead on the biosorption of copper, cadmium and zinc byactivated sludge, Hydrometallurgy 67 (2002) 109–116.

[5] R. Gupta, P. Ahuja, S. Khan, R.K. Saxena, H. Mohapatra, Microbial biosor-bents: meeting challenges of heavy metal pollution in aqueous solutions,Current Sci. 78 (2000) 967–973.

[6] Z. Al-Qodah, Biosorption of heavy metal ions from aqueous solutions byactivated sludge, Desalination 196 (2006) 164–176.

[7] P.S. Lawson, R.M. Sterritt, J.N. Lester, Adsorption and complexation mech-anisms of heavy metal uptake in activated sludge, J. Chem. Technol.Biotechnol. 34B (1984) 253–262.

[

[

Biointerfaces 62 (2008) 97–104 103

[8] M.J. Brown, J.N. Lester, Role of bacterial extracellular polymers in metaluptake in pure bacterial culture and activated sludge—II effects of meancell retention time, Water Res. 6 (1982) 1549–1560.

[9] T. Viraraghavan, G.A.K. Rao, Adsorption of cadmium and chromium fromwastewater by flyash, J. Environ. Sci. Heal. A 26 (1991) 721–753.

10] J.W. Patterson, Industrial Wastewater Treatment Technology, second ed.,Butterorth Publisher, Stoneham, MA, USA, 1985.

11] B. Volesky, Biosorption of Heavy Metals, CRC Press Inc., Boca Raton,Florida, 1990.

12] S.B. Mulrooney, R.P. Hausinger, Nickel uptake and utilization by microor-ganisms, FEMS Microbiol. Rev. 27 (2003) 239–261.

13] F. Hammes, A. Seka, S. Knijf, W. Verstraete, A novel approach to calciumremoval from calcium-rich industrial wastewater, Water Res. 37 (2003)699–704.

14] L. Holakoo, G. Nakhla, A.S. Bassi, E.K.F. Yanful, Long term performanceof MBR for biological nitrogen removal from synthetic municipal wastew-ater, Chemosphere 66 (2007) 849–857.

15] Y.H. Kim, S.H. Yeom, J.Y. Ryu, B.K. Song, Development of a novelUASB/CO2-stripper system for the removal of calcium ion in paperwastewater, Process Biochem. 39 (2004) 1393–1399.

16] T. Bahadir, G. Bakan, L. Altas, H. Buyukgungor, The investigation oflead removal by biosorption: An application at storage battery industrywastewaters, Enzyme Microb. Technol. 41 (2007) 98–102.

17] A. Ozturk, Removal of Nickel from aqueous solution by the bacteriumBacillus thuringiensis, J. Hazard. Mater. 147 (2007) 518–523.

18] B. Mattuschka, G. Straube, Biosorption of metals by a waste biomass, J.Chem. Technol. Biot. 58 (1993) 57–63.

19] Z.R. Holan, B. Volesky, Biosorption of lead and nickel by biomass of marinealgae, Biotechnol. Bioeng. 43 (1994) 1001–1009.

20] APHA–AWWA–WEF, Standard Methods of Examination of Water andWastewater, twentieth ed., American Public Health Association, ControlFed. 57, Washington DC, USA, 2005.

21] Y.S. Ho, G. McKay, Kinetic models for the sorption of dye from aqueoussolution by wood, Trans. Ins. Chem. Eng. 76B (1998) 183–191.

22] Y.S. Ho, G. McKay, Batch lead (II) removal from aqueous solution by peat:Equilibrium and kinetics, Trans. Ins. Chem. Eng. 77B (1999) 165–173.

23] Y.S. Ho, Review of second-order models for adsorption systems, J. Hazard.Mater. B136 (2006) 681–689.

24] Y.P. Kumar, P. King, V.S.R.K. Prasad, Zinc biosorption on Tectona grandisL.f. leaves biomass: Equilibrium and kinetic studies, Chem. Eng. J. 124(2006) 63–70.

25] J. Wase, C. Forster, Biosorbents for Metal Ions, Taylor and Francis Inc.,Bristol, PA, 1997.

26] A.H. Hawari, C.N. Mulligan, Biosorption of lead(II), cadmium(II), cop-per(II) and nickel(II) by anaerobic granular biomass, Biores. Technol. 97(2006) 692–700.

27] S. Senthilkumaar, S. Bharathi, D. Nithyanandhi, V. Subburam, Biosorp-tion of toxic heavy metals from aqueous solutions, Bioresour. Technol. 75(2000) 163–165.

28] A. Delgado, A.M. Anselmo, J.M. Novais, Heavy metal biosorption bydried powdered mycelium of Fusarium flocciferum, Water Environ. Res.70 (1998) 370–375.

29] G. Yan, T. Viraraghavan, Effect of pretreatment on the bioadsorption ofheavy metals on Mucor rouxii, Water SA 26 (2000) 119–123.

30] Z. Aksu, D. Akpinar, Modelling of simultaneous biosorption of phenoland nickel (II) onto dried aerobic activated sludge, Sep. Purif. Technol. 21(2000) 87–99.

31] I. Langmuir, The adsorption gasses on plane surface of glass, mica andplatinum, J. Am. Chem. Soc. 40 (1916) 1361–1368.

32] H.M.F. Freundlich, Over the adsorption in solution, J. Phys. Chem. 57(1906) 385–470.

33] C. Aharoni, M. Ungarish, Kinetics of activated chemisorption. Part 2.Theoretical models, J. Chem. Soc., Faraday Trans. 73 (1977) 456–464.

34] F.B. Dilek, A. Erbay, U. Yetis, Ni(II) biosorption by Polyporous versicolor,Process Biochem. 37 (2002) 723–726.

35] G. Donmez, Z. Aksu, A. Ozturk, T. Kutsal, A comparative study on heavymetal biosorption characteristics of some algae, Process Biochem. 34(1999) 885–892.

1 B: Biointerfaces 62 (2008) 97–104

[

[

04 M. Isik / Colloids and Surfaces

36] J.P.S. Cabral, Selective binding of metal ions to Pseudomonas syringaecells, Microbios 71 (1992) 47–53.

37] A. Ozturk, T. Artan, A. Ayar, Biosorption of nickel(II) and copper(II) ionsfrom aqueous solution by Streptomyces coelicolor A3(2), Colloid Surf. B34 (2004) 105–111.

[

38] F. Veglio, F. Beolchini, A. Gasbarro, Biosorption of toxic metals: an equi-librium study using free cells of Arthrobacter sp, Process Biochem. 32(1997) 99–105.