Chemical Engineering Journal 193–194 (2012) 256–266

Contents lists available at SciVerse ScienceDirect

Chemical Engineering Journal

journal homepage: www.elsevier .com/locate /ce j

Optimization of nickel biosorption by chemically modified brown macroalgae(Pelvetia canaliculata)

Amit Bhatnagar, Vítor J.P. Vilar ⇑, Catarina Ferreira, Cidália M.S. Botelho, Rui A.R. BoaventuraLSRE, Laboratory of Separation and Reaction Engineering, Associate Laboratory LSRE/LCM, Faculdade de Engenharia, Universidade do Porto, Rua Dr. Roberto Frias, 4200-465Porto, Portugal

a r t i c l e i n f o a b s t r a c t

Article history:Received 14 February 2012Received in revised form 9 April 2012Accepted 10 April 2012Available online 19 April 2012

Keywords:AlgaePelvetia canaliculataBiosorptionNickelIon-exchange

1385-8947/$ - see front matter � 2012 Elsevier B.V. Ahttp://dx.doi.org/10.1016/j.cej.2012.04.037

⇑ Corresponding author: Tel.: +351 918257824; faxE-mail address: vilar@fe.up.pt (V.J.P. Vilar).

In the present work, various forms of algae Pelvetia canaliculata were prepared by different chemicalmodifications, in order to get the best form of algae for the maximum uptake of nickel. Potentiometrictitration revealed that the carboxyl groups were more abundant (3.9 mmol/g) as compared to hydroxylgroups (2.0 mmol/g) on the biosorbent surface. Fourier transform infrared (FTIR) analysis of algae wasdone to identify the role of different functional groups present on algae surface during nickel biosorption.The protonated algae showed least sorption of nickel suggesting that after acid treatment, some of thebinding sites were destroyed. Among the various forms of prepared algae, Na-algae prepared directlyfrom raw algae (without protonation) showed highest uptake of nickel. The release of sodium ions duringthe uptake of nickel ions has shown that the current biosorption mechanism involves ion-exchange beinga stoichiometrical ratio of 2:1 between sodium and nickel ions.

� 2012 Elsevier B.V. All rights reserved.

1. Introduction

Heavy metals are essential for various biological activities in liv-ing organisms; nevertheless, they are considered as toxic at higherconcentrations. The presence of toxic metal ions in the environmenthas been recognized as deleterious to the ecosystem and humanhealth due to their non-degradability, biomagnification and toxic-ity. Besides the mining and metal related industries, other sourceswhich discharge metal-laden effluents include leather tanning, bat-tery, glassware, ceramics, electroplating, paints and photographicindustries [1]. Compared to conventional techniques such as pre-cipitation or synthetic ion exchange resins, biosorption process of-fers the economical and efficient approach for the remediation ofmetal bearing wastewaters because of eco-friendly characteristics,low cost, high uptake capacity, lack of toxicity constraints, lesssludge, possible regeneration of biosorbent and availability of bio-sorbents worldwide [2].

One of the important and widely used biosorbents is marine al-gae, which possess a high metal-binding capacity for various metals[3–5] with the cell wall playing an important role in binding [6,7].The high metal binding capacity in algae is due to the presence ofvarious functional groups such as carboxyl, amino, sulfonate andhydroxyl groups, which can act as binding sites for metals. Amongdifferent types of algae, brown algae has been found to be veryeffective biosorbents in removing heavy metals from water and

ll rights reserved.

: +351 225081674.

wastewater because of their high alginate content, higher uptakecapacities, similar to commercial ion-exchange resins and theirunlimited availability in the oceans [2,3,8–10]. Alginate, which iscomposed of mannuronic and guluronic acids, is a major polysac-charide in brown algae and offers carboxyl groups [11]. It consti-tutes between 10% and 40% of the brown algal dry weight [12].Brown algae also contain between 5% and 20% of the sulfated matrixpolysaccharide fucoidan [13] about 40% of which are sulfate esters.Alginate and fucoidan are known for their metal binding propertieswhereby ion exchange between metal ions occurs [14].

To enhance the removal efficiency of metal ions by different al-gae, various pretreatments have been reported in the literature.Pretreatment may be in terms of hardening the cell wall structurethrough a cross-linking reaction using epichlorohydrin [15] orincreasing the negative charge on the cell surface by NaOH treat-ment [16], or opening of the available sites for the adsorption byacid treatment [17], and enhancing the ion exchange by initial sat-uration of biomass with easily replaceable ions such as Ca or Na tofacilitate the metal sorption [18].

The importance of any given group for biosorption of a certainmetal by a certain biomass depends on various factors such as:the number of sites in the biosorbent material, the accessibility ofthe sites, the chemical state of the site (i.e., availability), and affinitybetween site and metal (i.e., binding strength) [19]. Bindingstrength is one of the important parameters in biosorption processin which ionic (electrostatic) binding and covalent binding are theimportant ones during metal binding. From different literaturesources, it can generally be concluded that the light metals (alkaline

A. Bhatnagar et al. / Chemical Engineering Journal 193–194 (2012) 256–266 257

and alkaline earth metals) bind less strongly than the heavy metalions [20]. Therefore, the former do not strongly interfere with thebinding of the latter.

Metal affinity to the biomass can be manipulated by pre-treat-ing the biomass with alkalis, acids, detergents and heat, which mayincrease the metal uptake. Further, the surface modification ap-proach has been suggested to be more cost-effective, as the modi-fication agents are normally less expensive than entrapmentmaterials, the sorptive capacity is enhanced, and the mass transferis not affected [21]. Understanding the binding mechanism canhelp to obtain stoichiometric information for the preparation ofcommercially applicable biosorbents and for the optimization ofprocess conditions. In order to determine the binding mechanism,one must identify the biopolymers and functional groups of thebiosorbent participating in the binding interaction.

In the present study, different chemical modification methodswere employed to raw algae (Pelvetia canaliculata) in order to pre-pare surface modified forms and their potential towards nickel[Ni(II)] biosorption was assessed considering the ion exchangemechanism. Nickel was selected as model pollutant as its saltsare used in many industrial applications. Ni(II) is essential for liv-ing organisms under permissible limits and it participates in vari-ous metabolic reactions such as ureolysis and acidogenesis [22].However, long term exposure to higher Ni(II) concentrations maylead to various health problems including skin dermatitis, gastro-intestinal distress, lung cancer, renal edema and pulmonary fibro-sis [23,24]. The World Health Organization’s (WHO’s) permissiblelimit for Ni(II) in drinking water is 0.5 mg/L [25]. The biosorptionkinetics on raw and modified algae was examined and the role offunctional groups in Ni(II) biosorption was discussed based onthe data obtained from potentiometric titration, Fourier transforminfrared (FTIR) analysis and esterification studies.

2. Experimental methods

2.1. Biomass preparation

The brown seaweed P. canaliculata was collected at the Northerncoast of Portugal. P. canaliculata is a common brown seaweed (Pha-eophyta) growing on the rocks of the upper shores of Europe fromIceland to Spain, including Norway, Ireland, Great Britain, theNetherlands, France and Portugal. The sun-dried seaweed waswashed with deionized (D.I.) water to remove sand and excesssalts followed by drying over night at 45 �C in an oven (BINDERED53). The biomass was then crushed using a mill (Retsch, ZM100), sieved (Retsch, AS 200) to obtain the size fraction of 1–2 mm, and stored until use. It was termed as raw algae. Usingthe raw algae, some biomass was subsequently protonated bysoaking into 0.2 M HNO3 under constant shaking (VWR Advanceddigital system) for different cycles of different time periods. Aftereach cycle, the solution was replaced with freshly prepared solu-tion. The biomass was then rinsed with D.I. water several times un-til pH reached �4.0 and, later, it was dried at 45 �C. The preparedalgae were termed as protonated (H+) algae.

The biomass was also converted to different ionic forms (Na-,K-, Ca- and Mg-) using the raw as well as the protonated formby soaking them into 0.5 M of respective chloride solutions (NaCl,KCl, CaCl2 and MgCl2) solution for 2 cycles of 24 h each under slowstirring. 1.0 M of respective metal hydroxide solutions (NaOH,KOH, Ca(OH)2 and/or Mg(OH)2) were used to control the solutionpH (�5.5–6.0) where protonated algae were used. After each cycle,the old solution was replaced with freshly prepared new solution.Afterwards, the Na-, K-, Ca- and/or Mg-loaded algae was rinsedwith D.I. water until achieving a low conductivity solution andpH about 4.0 (in case of H+-algae), and about 6.0–6.5 (in case of

raw algae). Finally, the algae were dried at 45 �C for 24 h andstored until use. The prepared algae were termed as metal-loadedalgae (Na-, K-, Ca- and/or Mg-loaded algae). Details of modifica-tion methods are listed in Table 1. Attempts were also made toprepare the protonated Na-algae by varying the experimental con-ditions and results are presented in Table 2. Sodium salts wereused as these are highly soluble in water as compared to Mgand Ca salts.

2.2. Blocking of carboxyl and sulfonate groups in algae

Esterification of carboxyl group was carried out by suspendingthe biomass (2 g) in methanol (130 mL) and concentrated hydro-chloric acid (1.2 mL) and equilibrated for 6 h at 25 �C [26]. Esterifi-cation of carboxyl acids present on the cell wall occurs according tofollowing reaction:

Afterwards, the biomass was washed with D.I. water nine timesin batch conditions (10 g/L; 20 min/wash) and dried at 45 �C for12 h. The modification of sulfonate groups was done by the methodreported elsewhere [27]. For the esterification of sulfonate groups,biomass (2 g) was suspended in methanol (130 mL) and concen-trated hydrochloric acid (1.2 mL) and equilibrated for four cyclesof 48 h continuous agitation, with replacement of the methanolicHCl between cycles [27]. Researchers [27] have also reported thatthe methanolic HCl treatment of the algal biomass resulted in sig-nificant decrease of the sulfonates concentration. They reportedthat after three successive treatments, there were no sulfonatesdetectable [27]. The sorption experiments with the modified bio-mass were carried out in batch conditions, as described in Section2.7 for the nickel biosorption experiments.

2.3. Biomass digestion

To determine the amount of metal ions present in raw and me-tal-loaded biomass, the samples were digested in a microwaveoven (Anton Paar, Multiwave 300) after adding 5.0 mL D.I. water,4.0 mL HNO3 (Merck) and 12.0 mL HCl (Merck) to 0.5 g of sample.The samples were cooked at 140 �C for 2 h and then the sampleswere left to cool. The metal concentrations in the digests weredetermined by atomic absorption spectrometry (AAS) (GBC 932Plus, Perkin Elmer) after filtration through cellulose acetate mem-brane filters (Ref. Albet-CA-045-25-BI). The obtained results areshown in Table 3.

2.4. Solutions preparation

Nickel(II) solutions were prepared by dissolving a weighedquantity of Ni(NO3)2�6H2O (Merck with purity > 98%) in D.I. water.The pH of each test solution was adjusted to the required valuewith diluted HCl and NaOH solutions. 0.5 M of NaCl, KCl, CaCl2

and MgCl2 solutions were prepared by dissolving respective chlo-ride salts (Merck with purity > 99.5%). 1.0 M NaOH, KOH, Ca(OH)2

and Mg(OH)2 solutions were prepared by dissolving the respectivehydroxide salts (Merck with purity > 99.5%).

2.5. Fourier transform infrared (FTIR) analysis

The chemical groups on the algae surface were detectedthrough FTIR spectroscopy (IRAffinity-1, Shimadzu, with EasiDiff

Table 1List of modification methods of Pelvetia canaliculata.

Experiment Reagent Ionic strength (M) Algae dose (g/L) No. of cycles Total contact time (h) Generic term used

1. Deionized water – 10 2 � 24 h 48 Raw algae2. HNO3 (68%) 0.2 10 2 � 3 h 6 Protonated (H+) algae -23. HNO3 (68%) 0.2 10 2 � 12 h 24 Protonated (H+) algae -34. HNO3 (68%) 0.2 10 2 � 24 h 48 Protonated (H+) algae-45. NaCl 0.5 1a (protonated) 2 � 24 h 48 Na-algae (after protonation)6. NaCl 0.5 5 (raw) 2 � 24 h 48 Na-algae (no protonation)7. KCl 0.5 5 (protonated) 2 � 24 h 48 K-algae (after protonation)8. KCl 0.5 5 (raw) 2 � 24 h 48 K-algae (no protonation)9. CaCl2 0.5 5 (protonated) 2 � 24 h 48 Ca-algae (after protonation)

10. CaCl2 0.5 5 (raw) 2 � 24 h 48 Ca-algae (no protonation)11. MgCl2 0.5 2a (protonated) 2 � 24 h 48 Mg-algae (after protonation)12. MgCl2 0.5 5 (raw) 2 � 24 h 48 Mg-algae (no protonation)

a Lower dosages were used here as if using higher dose, the resultant algae shape was destroyed.

Table 2Modification methods for the preparation of protonated Na-algae.

Algae sample ID NaCl (M) Protonated algae (g/L) Cycle Contact time (h) Total contact time (h) Remark

P-Na-1 0.4 10 1 24 24 Algae completely destroyedP-Na-2 0.5 5 1 24 24 Algae completely destroyedP-Na-3 0.5 5 1 24 24 Algae completely destroyedP-Na-4 0.5 4 1 24 24 Algae completely destroyedP-Na-5 0.5 1 1 24 24 Algae completely destroyedP-Na-6 0.5 1 2 5 10 Shape of algae not destroyedP-Na-7 0.5 2 2 24 48 Shape of algae not destroyedP-Na-8 0.5 4 2 4 8 Shape of algae not destroyedP-Na-9 0.5 5 2 4 8 Shape of algae not destroyedP-Na-10 0.5 1 2 24 48 Algae size reducedP-Na-11 0.5 5 4 2 8 Shape of algae not destroyedP-Na-12 0.5 5 3 2 6 Shape of algae not destroyedP-Na-13 0.5 1 3 24 72 Shape of algae not destroyed

Remark: 0.5–1.0 M NaOH was used to adjust the solution pH.

Table 3Metallic elements in raw and modified Pelvetia canaliculata obtained from digestion.

Sample ID Elements

0.83 mmol Ca/gRaw algae 0.48 mmol Mg/g

0.87 mmol Na/g0. 81 mmol K/g

Ca-loaded algae (No protonation) 0.75 mmol Ca/gCa-loaded algae (after protonation) 0.42 mmol Ca/gMg-loaded algae (No protonation) 1.28 mmol Mg/gMg-loaded algae (after protonation) 0.33 mmol Mg/gNa-loaded algae (No protonation) 1.30 mmol Na/gP-Na-6 0.15 mmol Na/gP-Na-7 0.21 mmol Na/gP-Na-8 0.10 mmol Na/gP-Na-9 0.09 mmol Na/gP-Na-10 0.17 mmol Na/gP-Na-11 0.09 mmol Na/gP-Na-12 0.08 mmol Na/gP-Na-13 0.28 mmol Na/gK-loaded algae (No protonation) 0.84 mmol K/gK-loaded algae (after protonation) 0.28 mmol K/g

258 A. Bhatnagar et al. / Chemical Engineering Journal 193–194 (2012) 256–266

diffuse reflectance accessory, Pike Technologies). Spectra were reg-istered from 4000 to 600 cm�1.

2.6. Potentiometric titration

Potentiometric titration was performed using an automatictitration system (Metrohm, 702 SM Titrino) and a stirrer module(Metrohm, 728 stirrer). The pH electrode was calibrated with buf-fer solutions of pH 1.0, 4.0, 7.0 and 10.0. For each titration, 0.25 g ofalgal biomass was added to 50 mL of 0.05 mol/L NaCl solution and,then, this suspension was placed in a thermostatic titration cell at

25 �C. Titration was carried out by stepwise addition of 0.02 mL ofa 0.1 M NaOH and/or 0.1 M HCl solutions to the cell while the sus-pension was stirred under a nitrogen atmosphere (in case of basictitration). After addition of acid/base, the drift rate was measuredover time, and readings were accepted when the drift was less than0.5 mV/min. For each data point, the maximum drift monitoringtime was 20 min.

2.7. Batch biosorption studies

Kinetic experiments were performed by two methods. In firstmethod, kinetics of the process was studied in 2 L capacity per-fectly mixed sorber, operating in batch mode at room temperature.0.5 g of biosorbent was added to 1 L of the nickel solution (50 mg/L). The solution pH was maintained �4.0 throughout these exper-iments. Aliquots of 10 mL were collected at different time intervalsafter commencing the experiment. In second method, biosorptionexperiments were carried out in 100 mL capped Erlenmeyer flaskscontaining 50 mL Ni(II) solution and 0.025 g biomass by shaking(170 rpm) at 25 �C until equilibrium was achieved. After adjustingthe desired pH, no further pH adjustment was done and final pHwas noticed after experiments. After achieving the equilibrium,solutions were separated from the biomass by filtration throughcellulose acetate membrane filters (Ref. Albet-CA-045-25-BI) andanalyzed for residual metal concentration (contaminant metalion (Ni) and exchangeable metal ion (Ca, Mg, Na and/or K) byatomic absorption spectrometry (AAS).

2.8. Analytical procedures

The concentrations of Ni, Na, K, Ca and Mg in the supernatantswere determined by Atomic Absorption Spectrometer (AAS) (GBC

Fig. 1. (a) FTIR spectra of different forms of algae (1): Raw algae, (2): Protonated algae, (3): Na-algae (after protonation), (4): Na-algae (no protonation), (5): Na-algae (noprotonation, after Ni(II) sorption, 300 mg/L, pH 4) and (b) FTIR spectra zoomed for wavelength lower than 1800 cm�1.

A. Bhatnagar et al. / Chemical Engineering Journal 193–194 (2012) 256–266 259

932 Plus, Perkin Elmer) with deuterium background correction anda spectral slit width of 0.5 nm for Ni, Ca, Mg and Na, and 0.2 nm forK. The working current/wavelength was adjusted to 4.0 mA/351.5 nm for Ni, 10 mA/422.7 nm for Ca, 5 mA/422.7 nm for Mg,5 mA/330.4 nm for Na and 5 mA/404.4 nm for K, giving a detectionlimit of 0.09 ppm for Ni, 0.0005 ppm for Ca, 0.0003 ppm for Mg,0.0002 ppm for Na and 0.003 ppm for K. In the analysis of Ca andMg, La2O3 solution (5.86% (w/v)) was added in the solution (1 mLLa2O3 in 10 mL solution), while in the analysis of Na and K, KCl(0.38% (w/v)) and NaCl (0.50% (w/v)) was added respectively tominimize the interferences. The instrument response was periodi-cally checked by respective standard solutions.

3. Mathematical models for functional groups quantification

Considering the surface of the algae P. canaliculata as polifunc-tional, each active site of one certain functional group will react

with hydrogen ions with different affinity and the total fractionof protonated sites, hT,H, can be given by the Langmuir–Freundlichisotherm. Assuming a Quasi-Gaussian distribution of the affinityconstant suggested by Sips [28], one can write:

hT;H ¼ðK 0HCHÞmH

1þ ðK 0HCHÞmHð1Þ

where K 0H is the average value of the affinity distribution for thehydrogen ions, which determines the position of the affinity distri-bution on the log Kint

i;H axis (Kinti;H – intrinsic equilibrium constant for

the hydrogen ions for each binding site i); CH is the concentrationof hydrogen ions in solution; and mH (0 < mH < 1) is the width of theirpeak in the Sips distribution (extreme values 0 and 1 represent a nullor infinite width, respectively). The charge of a biosorbent dependson the degree of protonation. If the affinity distribution exhibitsmore than one peak, then the charge of an acidic surface, QH, is ex-pressed as the weighted summation of the charge contributions of

Table 4Commonly observed stretching frequencies in seaweed FTIR spectra.

Wave number (cm�1) Assignment

3000–3600 Carboxylic/OH stretch and NAH stretcha

2920 Asymmetric stretch of aliphatic chains (ACH)b

2854 Symmetric stretch of aliphatic chains (ACH)b

1740 C@O stretch of COOHc

1630 Asymmetric C@Oc

1530 Amide IIa

1450 Symmetric C@Oc

1371 Asymmetric -SO3 stretchingd

1237 CAO stretch of COOHc

1160 Symmetric ASO3 stretchingd

1117 CAO (ether)a

1033 CAO (alcohol)a

817 S@O stretchingd

a [36].b [55].c [27].d [34].

Fig. 2. Experimental data and model curve for biosorbent potentiometric titration(IS = 0.05 M); IS – Ionic Strength.

260 A. Bhatnagar et al. / Chemical Engineering Journal 193–194 (2012) 256–266

the different site types. In this case it is assumed the existence of twofunctional groups, carboxylic (j = 1) and hydroxyl (j = 2) [29,30]:

Q H ¼Q max;1 � ðK 01;HCHÞmH;1

1þ ðK 01;HCHÞmH;1þ Q max;2

1þ ðK 02;HCHÞmH;2ð2Þ

The first term is related with protonation and the second term isrelated to deprotonation.

4. Results and discussion

4.1. FTIR analysis

Infrared spectra of different biomass samples are shown inFig. 1a. In order to observe the clear peaks and changes in the algae,

Table 5Parameters of the continuous distribution model for brown algae P. canaliculata (IS = 0.05

Biosorbent Qmax,1 (mmol/g) Qmax,2 (mmol/g) pK1,H = logK1,H

P. canaliculata 3.9±0.4 2±1 2.6±0.1

spectra were zoomed (Fig. 1b) for wavelength lower than1800 cm�1. Some of the commonly observed stretching frequen-cies in seaweed FTIR spectra are given in Table 4. The spectrumof protonated biomass (H+-algae) displayed absorbance peaks at1740 cm�1 corresponding to the stretching band of the free car-bonyl double bond from the carboxyl functional group. The absor-bance bands at approximately 1650 cm�1 were found in all thealgae biomass. This peak corresponds to the stretching of the car-bonAoxygen bonds in the carboxyl groups of the algal cell wallcomponent [31]. A band at 1430 cm�1 was also prominent in rawalgae, which may be due to CAOH deformation vibration with con-tribution of OACAO symmetric stretching vibration of carboxylategroup [32]. After treating with Na or Ni solutions, shift of the car-bonyl stretching band to lower frequencies was observed in pro-tonated and Na-algae (after protonation) [33]. Another peakaround 1260 cm�1 in all forms of algae was found which corre-sponds to the complexation of the oxygen from the carbonylCAO bond. Absorbance peaks around 1160 cm�1 relates to sym-metric stretching of SO�3 bonds in sulfonic acids and can be seenin FTIR spectrum of protonated and Na-algae (after protonation)[34]. This group is mainly present in sulfonic acids of polysaccha-rides, such as fucoidan, in the biomass. The broad band at 3000–3600 cm�1 corresponds to OAH group from cellulose and NAHgroups from proteins in all the various forms of the algae [35].The peaks around 2920 cm�1 and 2855 cm�1 can be attributed toasymmetric and symmetric stretches of aliphatic chains [36,37].Since cellulose itself only contains hydroxyl groups that becomecharged only at pH > 10, it can be suggested that these hydroxylgroups are ineffective at lower pH. Consequently, cellulose has avery low metal uptake capacity and it hardly contributes to metalbinding in seaweeds. In brown algae cellulose may constitute 4–6%of the dry weight [18]. The bands near 1530 cm�1 are attributed toamide I and II from proteins. The peaks at 825 cm�1 may corre-spond to CASAO and S@O, respectively [27].

4.2. Potentiometric titration

The potentiometric titration data enables the qualitative andquantitative determination of the nature and number of activesites present at the algae surface. Experimental data were reported(Fig. 2) as negative (protonation)/positive (deprotonation) charge/gof biosorbent versus pH:

QH ¼ðCb � Ca þ ½Hþ� � ½OH��Þ � V

mð3Þ

where Cb (mM) and Ca (mM) are the base and acid concentrations inthe suspension for each titrant addition, V is the total volume (mL)of the suspension and m is the biosorbent amount (g). The experi-mental data were fitted to a model (Eq. (2)) that considers a contin-uous distribution of the groups on the biosorbent surface [10].Mainly carboxylic groups and hydroxyl groups were taken intoaccount, as these are the two main entities which constitute the(marine) algae. Fig. 2 presents the experimental results and theoret-ical curve for the potentiometric titration at 0.05 M ionic strength.Adjustable parameters for the continuous model of brown algaeP. canaliculata are shown in Table 5.

The obtained apparent pKa values are close to the lower limitof the range 3 < pH < 5 where the carboxylic groups generate a

M).

pK2,H = logK2,H mH,1 mH,2 SR2 (mmol/g)2 R2

11±4 0.9±0.1 0.3±0.3 6.00E�02 0.900

Fig. 3. (a) Uptake of Ni(II) and release of light metals during biosorption kinetics on raw algae (j: Ni; : Na; : K; : Ca; : Mg; : H+) and (b) Experimental results for ion-exchange of Ni(II) and light metals (j: Ni; : Sum of light metals released).

A. Bhatnagar et al. / Chemical Engineering Journal 193–194 (2012) 256–266 261

negative charge on the algae surface allowing electrostatics inter-actions with metal ions [37]. Polysaccharides of the brown algaealso contain hydroxyl groups, although they only become activeand available at pH > 10. Thus, their contribution to metal uptakeis secondary [9,38].

Metal biosorption is also influenced by the presence of algal pro-teins (pKa around 8) particularly between pH 6 and 9 [12]. On theother hand, at low pH, sulfonate groups (pKa around 1.0–2.5)actively contribute to metal binding [36]. In the present study,although those groups were not detected by titration, the presenceof sulfonated groups on the seaweed surface has already beendetected by FTIR analysis [37], and a fraction of the amount of

carboxylic groups determined by the model can be due to sulfonatedgroups.Fig. 2 shows the distribution function F ¼

Pifiðlog K int

i;HÞQ Hi

versus logK inti;H for P. canaliculata (where fi logK int

i;H represents Sips dis-tribution function for carboxylic and hydroxyl groups with totalcharge of Qmax,1 and Qmax,2, respectively). The carboxyl groups(Qmax,1) were found to be more abundant (3.9 mmol/g) as comparedto hydroxyl groups (Qmax,2) (2.0 mmol/g). Two peaks correspondingto carboxyl and hydroxyl groups can be observed in Fig. 2. Theparameter pK 0i;H (with i = 1, 2) is the abscissa value for the maxi-mum of the distribution function for each functional group. Theheterogeneity is represented by the width of the distribution(mH). As known, low mH values correspond to a wider distribution

Fig. 4. Biosorption kinetic studies of Ni(II) on protonated forms of algae (Ni(ini): 50 mg/L; pH: �4.0; algae dose: 0.5 g/L) (j: 2 cycles � 3 h; : 2 cycles � 12 h each; : 2cycles � 24 h each).

262 A. Bhatnagar et al. / Chemical Engineering Journal 193–194 (2012) 256–266

and, also, a higher heterogeneity of groups [10]. From Table 5 andFig. 2, it can be confirmed that the width of the distribution forcarboxylic groups, mH,1, is higher than those for hydroxyl groups,mH,2. Thus, it can be concluded that, besides the scarcity of hydro-xyl groups as compared to carboxylic groups, hydroxyl groups arealso more heterogeneous.

4.3. Ni(II) biosorption kinetics on raw algae

Biosorption kinetics was initially performed with raw algae(0.5 g/L) at an initial nickel concentration of 50 mg/L and pH 4.0in order to assess the Ni(II) removal, considering the ion exchangemechanism. The change in solution pH and release of light metalcations (Na, K, Ca, Mg) from the biomass was simultaneously mon-itored during biosorption experiments (Fig 3a). It was found thatthe Ni(II) bound by the biomass of P. canaliculata was proportionalto the sum of the released light metal cations [Na, K, Ca, and Mg] bythe biomass (Fig 3b). During the biosorption kinetics of Ni(II) on P.canaliculata, Na ions were released in the highest amounts amongall the light metal cations. From the digestion results (Table 3), itwas calculated that Ca is present in the highest amount (1.664 me-quiv Ca2+/g) followed by Mg (0.962 mequiv Mg2+/g), Na (0.865 me-quiv Na+/g) and K (0.810 mequiv K+/g) in raw algae. Whencompared with the kinetic results of raw algae for Ni biosorption,it was found that Na was released in highest amount (�77%), fol-lowed by Mg and K (�58%). The least release was observed in caseof Ca (only 8%) which is consistent with previous results where itwas reported that Ca has greater affinity for the binding sites[18]. The results suggest that Na ions have easily replaceable ten-dency to exchange with Ni ions in the solution. It can be assumedthat, in the case of P. canaliculata, Na cations played a dominantrole in the cation exchange in the biosorption process.

Ion-exchange has been considered as the critical phenomenonin biosorption by various researchers, as it describes the experi-mental observations made during metal uptake experiments. Fur-thermore, it is a natural extension to the premise that alginateplays a key role in biosorption by brown algae, since it has been

shown that ion-exchange takes place between metals when bind-ing to alginate [39]. It is well reported in the literature that metalcations are bound by algae by ion exchange, whereby acidic func-tional groups in the biomass exchange protons and/or cations ofalkaline earth metals [Ca,Mg] and/or alkali metals [Na,K] with me-tal ions from the solutions [40]. Light metals, present in marine andfreshwater environments, are naturally bound by functionalgroups on the surface of the algal cell wall. An enhanced releaseof light metals (Ca2+, Mg2+, K+, Na+) from the alga Ascophyllumnodosum was observed when reacted with a cobalt containingaqueous solution rather than cobalt-free solution [41,42]. Further-more, a 2:3 stoichiometric relationship was observed between Ca2+

release and Co2+ uptake when the algae was pre-treated with CaCl2

and HCl. Schiewer and Volesky [43] pointed out, however, that aratio closer to one would have been achieved if protons were in-cluded in the charge balance. It was concluded that ion-exchangewas the dominant mechanism.

The results of the present study suggest that, one of the mech-anisms involved in the biosorption of Ni(II) with P. canaliculatawould be the exchange between Ni(II) in the solution and lightmetals (especially Na present in the alginate of the algal cell wall).Other cations, aside from sodium, such as calcium, magnesium,and potassium could also be exchanged with metals in solutionduring the biosorption process. Different authors have also ob-served an increase in the concentrations of these ions during thebiosorption of cadmium, lead and copper with brown algae[44,45]. For instance, copper biosorption with Fucus vesiculosuswas estimated to be 77% due to ion exchange, with 13%, 9%, 24%and 31% accounting for interchange with Ca2+, Mg2+, Na+ and K+,respectively [45]. Chen and Yang described the influence of thesemetals in solutions during the biosorption of cadmium, lead andcopper with the brown algae Sargassum [21].

4.4. Ni(II) biosorption kinetics on modified algae

Preliminary experiments were conducted to see the influence ofpre-treatment of algae biomass by different protonation methods

Fig. 5. (a) Biosorption kinetic studies with Ni(II) on untreated (raw) and protonated forms of algae (Ni(ini): 50 mg/L; pH: �4.0; algae dose: 0.5 g/L), (j: Raw; : Protonated; :Ca (No protonation); : Mg (No protonation); : Na (No protonation); : K (No protonation); 4: Ca (after protonation); : Mg (after protonation); }: Na (afterprotonation); : K (after protonation) and (b) biosorption kinetic studies of Ni(II) on Na- algae prepared by different cycles (Ni(ini): 50 mg/L; pH: �4.0; algae dose: 0.5 g/L) (j:2 cycles � 24 h each; : 3 cycles � 24 h each; : 4 cycles � 24 h each).

A. Bhatnagar et al. / Chemical Engineering Journal 193–194 (2012) 256–266 263

on the uptake capacity for nickel (Fig. 4). As can be seen from thisfigure, the biosorption of Ni was higher on the H+-algae preparedby 2 cycles of 24 h each. Therefore, this form was further used toprepare the different ionic forms of biomass. The biosorption kinet-ics of nickel on different forms of prepared algae is presented inFig. 5a. As can be seen from this figure, the biosorption of Ni wasreduced by �65% in case of H+-algae as compared to the raw algae.The lower biosorption in H+-algae can be attributed to the destruc-tion of some of the binding sites responsible for metal biosorptionin algae after acid treatment. It has been reported in the literature

that an acid washing can provoke the dissolution of alginates orother constituents of the cell wall matrix of algae, and the subse-quent decrease of sorption potential [46]. On the other hand, rawalgae showed higher Ni(II) uptake as compared to H+-algae. As dis-cussed earlier, light metals e.g., calcium (Ca2+), magnesium (Mg2+),sodium (Na+), and potassium (K+) might attach to seaweeds (in theraw algae) through covalent or electrostatic attraction and could beinvolved in the heavy-metal biosorption process through ion ex-change, metal complex formation, coordination reactions, andother mechanisms [21]. Therefore, it can be concluded that acid

Table 6Binding strength characterization parameters.

H Na K Ca Mg Ni

z (charge) 1 1 1 2 2 2Rcryst

a (Å) – 1.02 1.51 1.00 0.72 0.69Rhyd

b (Å) 2.82 3.58 3.31 4.12 4.28 4.04Xm

c 2.20 0.93 0.82 1.00 1.20 1.80z2/rcryst

d1 (1/Å) – 0.98 0.66 4.00 5.56 5.80z2/rhyd

d2 (1/Å) 0.35 0.28 0.30 0.97 0.93 0.99

X2mðrcryst þ 0:85Þe (Å) – 1.51 1.51 1.85 2.26 4.99

DXmf 1.30 2.60 2.70 2.50 2.30 1.70

1� expð�DX2m=4g 0.34 0.82 0.84 0.79 0.73 0.51

xih = z2=rhyb=1� expð�DX2m=4Þ (1/Å) 1.03 0.34 0.36 1.23 1.27 1.92

a Shannon crystal radii [56].b Nightingale hydrated ion radii [57].c Pauling electronegativity [58].d Parameter for hydration (1) or ionic bonding (2) strength [59].e Parameter for covalent bond character (0.85 is an appropriate constant

assumed to reflect the radius of O and N donor atoms) [51].f Parameter for ionic bond character (Electronegativity of the metal relatively to

oxygen) [58].g Fraction of ionic bond character [60].h Parameter for total binding strength.

264 A. Bhatnagar et al. / Chemical Engineering Journal 193–194 (2012) 256–266

washing was not found to be a successful step to enhance the bio-sorption of nickel.

The Ni(II) biosorption was also examined on the ionic forms(Ca-, Mg-, Na- and K-) prepared from raw (washing with D.I. wateronly) and after protonation (acid treatment) algae and the resultsare shown in Fig. 5a. The results showed that Ni(II) biosorptionwas found lower in case of all the ionic forms of algae which wereprepared from the H+-algae as compared to the raw algae. This canbe attributed to the loss of active binding sites after protonation(acid treatment) as explained earlier. On the other hand, Ni(II)biosorption was increased on all the ionic forms of algae (exceptCa-form) prepared from the raw algae, which can be attributedto the increase of specific metal content in respective algae. Aftertreatment with metal chloride salt, the specific metal contentincreased in the algae which helped to enhance the Ni(II)

Fig. 6. Experimental results for ion-exchang

biosorption. This trend of biosorption is in consistent with diges-tion results where an increase in metal content was observed aftertreatment with specific metal salt treatment.

The improvement in Ni(II) biosorption after metal chloridetreatment in raw algae follows the order: Na > K > Mg > Ca. Inter-estingly, Ni(II) biosorption on Ca-loaded algae was found on lowerside as compared to other ionic forms of algae (Na-, K- and Mg-)prepared from raw algae. This could be consequence of its higherelectrostatic accumulation and its greater affinity for the bindingsites [18]. Similar results were reported by Chubar et al. [47],who showed that Cu2+ biosorption by cork biomass was more effi-cient after NaCl pretreatment. The explanation they offered is thatthe interaction of the biomass binding sites with the divalent cal-cium ions is stronger than that of the monovalent sodium cations,thus hindering the biosorption.

These observations can be viewed in a wider context: Ni is clas-sified as predominantly ‘‘borderline: intermediate metal’’ ion,while Na, K, Ca and Mg are classified as predominantly ‘‘hard’’ ions[48]. According to the literature, ‘‘hard’’ ions tends to form strongbonds with highly electronegative donors, which are difficult topolarize (hard donors – O or F), while soft cations form more stablecomplexes with soft donors (N, S, P, As) [49]. Soft ions participatein more covalent bonding where the free energy is mostly enthal-pic, and hard ions participate in more ionic bonding [50]. The sur-face groups of the biosorbents used in this work have beencharacterized as, predominantly, carboxylic and hydroxyl groups.As the experiments were conducted at pH values below 7, carbox-ylic groups (donor – O) are the only free active sites to bind metalions. Some characteristics of the studied metal ions are listed inTable 6. It can be seen that, according to the (z2/rhyd) criterion,the strength of the ionic bonds is Ni > Ca > Mg > K > Na. Using theDXm, or the 1� expð�DX2

m=4Þ criteria, one can conclude that thebinding of K is more electrostatic. Ni binding is highly covalentwhen compared with the other metals. Nieboer and McBryde[51] introduced the parameter X2

mðrcryst þ 0:85Þ as a measure forthe strength of covalent bonding, where 0.85 stands for thecontribution of N or O donors to the bond distance. ‘‘Hard’’ ionsare characterized by X2

mðrcryst þ 0:85Þ <� 4.2, and ‘‘soft’’ ions byX2

mðrcryst þ 0:85 > 7Þ. The X2mðrcryst þ 0:85Þ criterion confirms that

e of nickel with sodium during kinetics.

Fig. 7. Ni(II) biosorption kinetics on raw and esterified algae (Ni(ini): 50 mg/L; algae dose: 0.5 g/L), (j: Na-algae (no esterification) pH 4.0; : Na-algae (esterified) pH 4.0; :Na-algae (esterified) pH 2.0).

A. Bhatnagar et al. / Chemical Engineering Journal 193–194 (2012) 256–266 265

the relative contribution of covalent bonding follows the order,Ni > Mg > Ca > K = Na. From Table 6, it is clear that the parameterfor total binding strength (n) is highest for Ni followed by Mg, Ca,K and Na. This explains why Ni easily exchanged with Na ions.

As can be seen from Fig. 5a, an increase of 10–20% on Ni(II) bio-sorption was observed on ionic forms of algae (Na-, K- and Mg-). AsNa-form (no protonation, 2 cycles of 24 h each) showed the bestresults, efforts were made to saturate the binding sites with Na,by conducting 3 and 4 cycles also of 24 h each (Fig. 5b), but the bestresults were obtained with the algae prepared after 2 cycles, there-fore, this form was chosen for further studies. The other advantageof using Na-loaded biomass is the stability of the solution pH dur-ing sorption experiments, in contrast with protonated biomass, inwhich exchange of hydrogen ions occurs with the metal ions,thereby decreasing the solution pH and also decreasing the equi-librium metal uptake. The higher exchange of Ni2+ by sodium ionsmight be attributed to the fact that Ni2+ is above hydrogen in theelectrochemical activity series in the direction of increasingstrength of reduction with standard reduction potentials of�0.25 V [52,53]. Also, Ni2+ is beneath sodium in the series withstandard reduction potential of�2.76 V. Thus, sodium is effectivelyreplaceable by Ni2+ ions than the H atom of ACOOH of carboxylicgroup.

A preliminary digestion of the Na-form algae revealed a concen-tration of around 1.3 mmol Na+ per gram of the biosorbent. The to-tal of 3.9 mmol/g of carboxylic groups at the surface of thebiosorbent has been obtained by potentiometric titration. In thisway, one can conclude that, the number of cycles used for sodiumalgae pre-treatment was not sufficient to saturate all the bindingsites. The release of sodium ions during the uptake of nickel ionshas shown that the current biosorption mechanism involves theion exchange mechanism being a stoichiometrical ratio of 2:1 be-tween sodium and nickel ions (Fig. 6).

4.5. Ni biosorption kinetics on esterified forms of Na-algae

To ascertain the role of functional groups of biosorbent in thebinding of Ni(II), the carboxylic groups were esterified and later

the sulfonate groups were blocked and the binding of Ni(II) wasstudied in comparison with the control biosorbent at pH 4 and 2.The data (Fig. 7) shows a dramatic decrease (�40%) in the removalof Ni(II) by the esterified biosorbent as compared to the un-ester-ified one indicating a major role of ACOO groups in the binding ofNi(II) ions at pH 4.0. A good correlation between the degree ofblocking of COOA groups by esterification in Sargassum fluitansand the corresponding decrease in metal uptake has been previ-ously reported [27]. Similar results were also obtained for the car-boxyl groups in the biomass of the freshwater algae Chlorellapyrenoidosa and Cyanidium caldarium [26]. These results indicatethat carboxyl groups are important in the uptake of metal ions atpH 4.0, but little metal binding occurred at pH 2.0 supporting thesupposition that groups other than carboxylic functionalities areinvolved in metal ion uptake at pH 2.0 (e.g., sulfonate groups)[54]. The biosorption experiments with esterified algae also re-vealed that among the total available carboxyl groups, 3.9 mmol/g (calculated from potentiometric titration), the number of sulfo-nate groups are about 25% (1.0 mmol/g). The binding of Ni(II)was also studied with esterified algae in which sulfonate groupswere blocked, but no biosorption of Ni(II) was observed in thiscase, revealing the importance of sulfonate groups in metalbiosorption.

5. Conclusions

In the present study, biosorption of nickel by different forms ofchemically modified algae (P. canaliculata) was studied. The bio-sorption of Ni(II) was reduced by �65% in case of H+-algae whichmight be due to the destruction of some of the binding sitesresponsible for metal biosorption in algae after acid treatment.Furthermore, Ni(II) biosorption was also found to be lower in caseof all the ionic forms of algae which were prepared from the H+-al-gae as compared to the raw algae. On the other hand, Ni(II) bio-sorption was increased by 10–20% on all the ionic forms of algae(except Ca-form), which can be attributed to the increase of spe-cific metal content in respective algae. Na-algae prepared fromraw algae showed the highest uptake capacity for nickel at pH

266 A. Bhatnagar et al. / Chemical Engineering Journal 193–194 (2012) 256–266

4.0. The biosorption mechanism was attributed to ion exchangebetween sodium and nickel ions with stoichiometry 2:1 (Na:Ni).A marked decrease (�40%) in the removal of Ni(II) by esterifiedbiosorbent as compared to the un-esterified one indicated a majorrole of ACOO groups in the binding of Ni(II) ions at pH 4.0. Littlemetal binding also occurred at pH 2.0 supporting the suppositionthat groups other than carboxylate are involved in metal ion up-take at pH 2.0 (e.g., sulfonate groups).

Acknowledgments

This work was supported by project PEst-C/EQB/LA0020/2011,financed by FEDER through COMPETE – Programa OperacionalFactores de Competitividade and by FCT – Fundação para a Ciênciae a Tecnologia. Amit Bhatnagar acknowledges his post-doctoralscholarship (DFRH-SFRH/BPD/62889/2009) supported by thePortuguese Foundation for Science and Technology (FCT).

References

[1] S.E. Brown, W.C. Welton, Heavy Metal Pollution, Nova Science Pub Inc., 2008.[2] S.K. Mehta, J.P. Gaur, Use of algae for removing heavy metal ions from

wastewater: progress and prospects, Crit. Rev. Biotechnol. 25 (2005) 113–152.[3] E. Romera, F. González, A. Ballester, M.L. Blázquez, J.A. Muñoz, Biosorption with

algae: a statistical review, Crit. Rev. Biotechnol. 26 (2006) 223–235.[4] G.J. Ramelow, D. Fralick, Y.F. Zhao, Factors affecting the uptake of aqueous

metal-ions by dried seaweed biomass, Microbios 72 (1992) 81–93.[5] Z.R. Holan, B. Volesky, Biosorption of lead and nickel by biomass of marine

algae, Biotechnol. Bioeng. 43 (1994) 1001–1009.[6] R.H. Crist, K. Oberholser, J. McGarrity, D.R. Crist, J.K. Johnson, J.M. Brittsan,

Interaction of metals and protons with algae. 3. Marine algae, with emphasison lead and aluminum, Environ. Sci. Technol. 26 (1992) 496–502.

[7] N. Kuyucak, B. Volesky, The mechanism of cobalt biosorption, Biotechnol.Bioeng. 33 (1989) 823–831.

[8] S. Ahmady-Asbchin, Y. Andrès, C. Gérente, P.L. Cloirec, Biosorption of Cu(II)from aqueous solution by Fucus serratus: surface characterization and sorptionmechanisms, Bioresource Technol. 99 (2008) 6150–6155.

[9] T.A. Davis, B. Volesky, A. Mucci, A review of the biochemistry of heavy metalbiosorption by brown algae, Water Res. 37 (2003) 4311–4330.

[10] J.F.d.S.S. Costa, V.J.P. Vilar, C.M.S. Botelho, E.A.B. da Silva, R.A.R. Boaventura,Application of the Nernst–Planck approach to lead ion exchange in Ca-loadedPelvetia canaliculata, Water Res. 44 (2010) 3946–3958.

[11] R.E. Lee, Phycology, Cambridge University Press, Cambridge, UK, 1980.[12] E. Percival, R.H. McDowell, Chemistry and Enzymology of Marine Algal

Polysaccharides, Academic Press, London, UK, 1967.[13] V.J. Chapman, Seaweeds and Their Uses, Chapman and Hall, London, UK, 1980.[14] A. Haug, O. Smidsrod, Selectivity of some anionic polymers for divalent metal

ions, Acta Chem. Scand. 24 (1970) 843–854.[15] Y.H. Kim, J. Yeon Park, Y.J. Yoo, J.W. Kwak, Removal of lead using xanthated

marine brown alga, Undaria pinnatifida, Process Biochem. 34 (1999) 647–652.[16] E. Güris�ik, M.Y. Arica, S. Bektas�, Ö. Genç, Comparison of the heavy metal

biosorption capacity of active, heat-inactivated and NaOH-treatedPhanerochaete chrysosporium biosorbents, Eng. Life Sci. 4 (2004) 86–89.

[17] J. Yang, B. Volesky, Modeling uranium-proton ion exchange in biosorption,Environ. Sci. Technol. 33 (1999) 4079–4085.

[18] B. Volesky, Sorption and Biosorption, BV Sorbex, Inc., Quebec, 2003.[19] R.H.S.F. Vieira, B. Volesky, Biosorption: a solution to pollution?, Int Microbiol. 3

(2000) 17–24.[20] G.M. Naja, V. Murphy, B. Volesky, M.C. Flickinger, Biosorption, metals, in:

Encyclopedia of Industrial Biotechnology, John Wiley & Sons, Inc., 2009.[21] J.P. Chen, L. Yang, Chemical modification of Sargassum sp. for prevention of

organic leaching and enhancement of uptake during metal biosorption, Ind.Eng. Chem. Res. 44 (2005) 9931–9942.

[22] M.A. Khan, M. Ngabura, T.S.Y. Choong, H. Masood, L.A. Chuah, Biosorption anddesorption of nickel on oil cake: batch and column studies, BioresourceTechnol. 103 (2012) 35–42.

[23] V.K. Gupta, A. Rastogi, A. Nayak, Biosorption of nickel onto treated alga(Oedogonium hatei): application of isotherm and kinetic models, J. ColloidInterface Sci. 342 (2010) 533–539.

[24] Y.G. Bermúdez, I.L.R. Rico, O.G. Bermúdez, E. Guibal, Nickel biosorption usingGracilaria caudata and Sargassum muticum, Chem. Eng. J. 166 (2011) 122–131.

[25] WHO, Guidelines for Drinking Water Quality, Incorporating First addendum toThird ed., vol. 1 Recommendations, Geneva, 2006.

[26] J.L. Gardea-Torresdey, M.K. Becker-Hapak, J.M. Hosea, D.W. Darnall, Effect ofchemical modification of algal carboxyl groups on metal ion binding, Environ.Sci. Technol. 24 (1990) 1372–1378.

[27] E. Fourest, B. Volesky, Contribution of sulfonate groups and alginate to heavymetal biosorption by the dry biomass of Sargassum fluitans, Environ. Sci.Technol. 30 (1996) 277–282.

[28] R. Sips, On the structure of a catalyst surface, J. Chem. Phys. 16 (1948) 490–495.

[29] L.K. Koopal, T. Saito, J.P. Pinheiro, W.H.v. Riemsdijk, Ion binding to naturalorganic matter: general considerations and the NICA–Donnan model, ColloidsSurf. A: Physicochem. Eng. Aspects 265 (2005) 40–54.

[30] C.J. Milne, D.G. Kinniburgh, J.C.M. de Wit, W.H. van Riemsdijk, L.K. Koopal,Analysis of proton binding by a peat humic acid using a simple electrostaticmodel, Geochim. Cosmochim. Acta 59 (1995) 1101–1112.

[31] S.J. Kleinübing, R.S. Vieira, M.M. Beppu, E. Guibal, M.G.C.d. Silva,Characterization and evaluation of copper and nickel biosorption on acidicalgae Sargassum filipendula, Mater. Res. 13 (2010) 541–550.

[32] M. Mathlouthi, J.L. Koenig, Vibrational spectra of carbohydrates, in: R.S.Tipson, H. Derek (Eds.), Advances in Carbohydrate Chemistry andBiochemistry, Academic Press, 1987, pp. 7–89.

[33] M.M. Figueira, B. Volesky, V.S.T. Ciminelli, F.A. Roddick, Biosorption of metalsin brown seaweed biomass, Water Res. 34 (2000) 196–204.

[34] M.M. Figueira, B. Volesky, H.J. Mathieu, Instrumental analysis study of ironspecies biosorption by Sargassum biomass, Environ. Sci. Technol. 33 (1999)1840–1846.

[35] L. Deng, Y. Su, H. Su, X. Wang, X. Zhu, Sorption and desorption of lead (II) fromwastewater by green algae Cladophora fascicularis, J. Hazard. Mater. 143 (2007)220–225.

[36] P.X. Sheng, Y.-P. Ting, J.P. Chen, L. Hong, Sorption of lead, copper, cadmium,zinc, and nickel by marine algal biomass: characterization of biosorptivecapacity and investigation of mechanisms, J. Colloid Interface Sci. 275 (2004)131–141.

[37] V. Murphy, H. Hughes, P. McLoughlin, Cu(II) binding by dried biomass of red,green and brown macroalgae, Water Res. 41 (2007) 731–740.

[38] S. Schiewer, B. Volesky, Biosorption processes for heavy metal removal,Environ. Microbiol. Metal Interact. 14 (2000) 329–362.

[39] S. Myklestad, Ion-exchange properties of brown algae. II. Rate mechanism forcalcium–hydrogen ion exchange for particles from Ascophyllum nodosum, J.Appl. Chem. 18 (1968) 222–227.

[40] D. Kratochvil, P. Pimentel, B. Volesky, Removal of trivalent and hexavalentchromium by seaweed biosorbent, Environ. Sci. Technol. 32 (1998) 2693–2698.

[41] N. Kuyucak, B. Volesky, Accumulation of cobalt by marine alga, Biotechnol.Bioeng. 33 (1989) 809–814.

[42] N. Kuyucak, B. Volesky, Desorption of cobalt-laden algal biosorbent,Biotechnol. Bioeng. 33 (1989) 815–822.

[43] S. Schiewer, B. Volesky, Biosorption by marine algae, in: J.J. Valdes (Ed.),Remediation, Kluwer Academic Publishers, Dordrecht, The Netherlands, 2000,pp. 139–169.

[44] O. Raize, Y. Argaman, S. Yannai, Mechanisms of biosorption of different heavymetals by brown marine macroalgae, Biotechnol. Bioeng. 87 (2004) 451–458.

[45] E.L. Cochrane, S. Lu, S.W. Gibb, I. Villaescusa, A comparison of low-costbiosorbents and commercial sorbents for the removal of copper from aqueousmedia, J. Hazard. Mater. 137 (2006) 198–206.

[46] A.A. Hamdy, Biosorption of heavy metals by marine algae, Curr. Microbiol. 41(2000) 232–238.

[47] N. Chubar, J.R. Carvalho, M.J.N. Correia, Heavy metals biosorption on corkbiomass: effect of the pre-treatment, Colloids Surf. A: Physicochem. Eng.Aspects 238 (2004) 51–58.

[48] J.H. Duffus, ‘‘Heavy Metals’’ – a meaningless term?, Pure Appl Chem. 74 (2002)793–807.

[49] R.G. Pearson, Hard and soft acids and bases, J. Am. Chem. Soc. 85 (1963) 3533–3539.

[50] J. Buffle, Complexation Reactions in Aquatic Systems: An Analytical Approach,Ellis Horwood Ltd., Chichester, 1988.

[51] E. Nieboer, W.A.E. McBryde, Free-energy relationships in coordinationchemistry. III. A comprehensive index to complex stability, Can. J. Chem. 51(1973) 2512–2524.

[52] L. Wang, R.M. Shelton, P.R. Cooper, M. Lawson, J.T. Triffitt, J.E. Barralet,Evaluation of sodium alginate for bone marrow cell tissue engineering,Biomaterials 24 (2003) 3475–3481.

[53] H.M. Nizam El-Din, M.F. Abou Taleb, A.W.M. El-Naggar, Metal sorption andswelling characters of acrylic acid and sodium alginate based hydrogelssynthesized by gamma irradiation, Nuclear Inst. Method Phys. Res. Sec. B:Beam Int. Mater. Atoms 266 (2008) 2607–2613.

[54] L.R. Drake, S. Lin, G.D. Rayson, P.J. Jackson, Chemical modification and metalbinding studies of Datura innoxia, Environ. Sci. Technol. 30 (1995) 110–114.

[55] M.N. Pons, S.L. Bonte, O. Potier, Spectral analysis and fingerprinting forbiomedia characterisation, J. Biotechnol. 113 (2004) 211–230.

[56] H.T. Evans, Ionic radii in crystals, in: D.R. Lide (Ed.), CRC Handbook ofChemistry and Physics, CRC Press, Boca Raton, FL, 1993, pp. 12.18–12.19.

[57] Y. Markus, A.S. Kertes, Ion Exchange and Solvent Extraction of Metal, JohnWiley & Sons, London, UK, 1969.

[58] J.A. Dean, Lange’s Handbook of Chemistry, McGraw-Hill, New York, 1985.[59] C.S.G. Phillips, R.J.P. Williams, Inorganic Chemistry, Oxford University Press,

New York, Principle and Non-Metals., 1965.[60] L. Pauling, Nature of the Chemical Bond, Cornell University Press, Ithaca, NY,

1967.