Clean Techn Environ Policy (2013) 15:955–965 DOI 10.1007/s10098-012-0565-3

ORIGINAL PAPER

Turning Laminaria digitata seaweed into a resource for sustainable and ecological removal of trivalent chromium ions from aqueous solutions Ingrid M. Dittert • Vı´tor J. P. Vilar • Eduardo A. B. da Silva • Selene M. A. Guelli U. de Souza • Antoˆnio Augusto U. de Souza Cida´lia M. S. Botelho • Rui A. R. Boaventura

•

Received: 1 October 2012 / Accepted: 14 December 2012 / Published online: 30 December 2012 Ó Springer-Verlag Berlin Heidelberg 2012

Abstract This study presents the application of a safe, cost effective, environmental friendly, and efficient technology for the removal of trivalent chromium ions from aqueous solutions, based on the valorisation of a renewable resource, Laminaria digitata seaweed. Insights into trivalent chromium speciation in solution and interaction with the active sites present in the surface of the brown algae were studied. Carboxyl and hydroxyl groups were identified as the major binding sites present in the surface of the biosorbent, in concentrations (Qmax) of 2.06 ± 0.01 and 1.4 ± 0.7 mmol g-1, and with proton binding parameters (pK) of 3.28 ± 0.01 and 11 ± 1, respectively. Trivalent chromium uptake at equilibrium conditions was well described at different acidic pH conditions and chromium concentrations, using a model which incorporates trivalent chromium hydrolysis reactions in the aqueous phase and its chemical interactions with the available active sites (carboxyl groups) present in the surface of biosorbent. The distribution profile of trivalent chromium species present in the solution as well as at the binding sites indicated that Cr3? and CrOH2? exhibit different affinities for the carboxyl groups present in the surface of the biomass according to the pH. A mass transfer kinetics model was applied to describe the kinetics at batch system, being possible I. M. Dittert  V. J. P. Vilar (&)  E. A. B. da Silva  C. M. S. Botelho  R. A. R. Boaventura LSRE - Laboratory of Separation and Reaction Engineering—Associate Laboratory LSRE/LCM, Faculdade de Engenharia, Universidade do Porto, Rua Dr. Roberto Frias, 4200-465 Porto, Portugal e-mail: [email protected] I. M. Dittert  S. M. A. G. U. de Souza  A. A. U. de Souza Laborato´rio de Transfereˆncia de Massa, Departamento de Engenharia Quı´mica e Engenharia de Alimentos, Universidade Federal de Santa Catarina, P.O. Box 476, CEP 88040-900 Floriano´polis, SC, Brazil

to obtain the distribution of CrOH2? and Cr3? species in solution and at the binding sites. Keywords Biosorption  Laminaria digitata  Trivalent Chromium  Modeling List of symbols ap Specific area of the particle (cm-1) B Representative of the functional group in the biomass BT or Qmax Total number of binding sites B per unit mass of biomass (mmol g-1) Ci Concentration of species i in the fluid phase (mmol L-1) CH Proton concentration in the solution (mmol L-1) Dh,i Coefficient of homogeneous diffusion inside the particle for each species i (cm2 s-1) Fobj-a Objective function i Experimental sample number k Reaction rate constant (s-1) kp,i Overall mass transfer coefficient of species i (cm s-1) KH0 Average of the affinity distribution of hydrogen ions Kint The intrinsic proton affinity constant at each i,H binding site i KH Dissociation constant of functional group (mol L-1) KM1 Binding constant of Cr3? to functional group (L mol-1) KM2 Binding constant of CrOH2? to functional group (L mol-1) Ks Thermodynamic equilibrium constant (mol L-1)

123

956

mH n q hqii qi QH Qexp H,i Qest H,i R V W t z sd,i ri hT,H

I. M. Dittert et al.

Width of the Sips distribution Number of samples Concentration of species i in the solid phase (mmol g-1) Average concentration of species i in the solid phase (mmol g-1) Equilibrium concentration of species i in the solid phase (mmol g-1) Weighted sum of the charge contributions of each active site (mmol g-1) Experimental charge of an acidic surface (mmol g-1) Estimated charge of an acidic surface (mmol g-1) Half of the thin plate thickness (cm) Volume of the liquid in the reactor (L) Algal mass (g) Time (s) Distance to the symmetry plane (cm) Time constant for diffusion of ionic species into the particle (s) Kinetic rate (mmol L-1 s-1) Total degree of protonation

Introduction Chromium is found in all phases of the environment, including air, water, and soil, and its many chemical forms are pollutants with serious implications for the environment and human health (Shanker and Venkateswarlu 2011). Chromium compounds are constantly released into the aquatic environment by natural processes (mainly volcanic activity and rock weathering) and by anthropogenic sources, which in the last 20 years have become of greatest importance in the worldwide emission balance of these substances (Di Natale et al. 2007). Chromium speciation in water is of special interest, because its toxicity to aquatic and terrestrial organisms including humans is dependent upon its oxidation state. The two main oxidation states, in natural waters, are trivalent and hexavalent, which have different chemical behavior. The trivalent state has broad stability, and hexavalent chromium occurs under strongly oxidizing conditions. The oxidation state of an element can have an important effect on its bioavailability and the toxicity (Kocaoba and Akcin 2002). Most chromium applications are in metallurgy and in the dye and paint industries. Trivalent chromium salts, especially chrome alum (KCr(SO4)2) and trivalent chromium sulfate [Cr2(SO4)312(H2O)], are used in leather tanning,

123

while several chromium compounds are used as catalysts (Vaiopoulou and Gikas 2012). Most of the physico-chemical processes for metal removal are not effective or economically feasible for achieving the required stringent standards for disposal to surface water bodies (Saravane et al. 2001). Low cost, alternative processes are thus being sought in order to serve as polishing stages in the overall treatment scheme, removing low to medium metal ion concentrations (Williams and Edyvean 1997). In recent years, applying biotechnology to control and remove metal pollution has been paid much attention and has gradually become a hot topic in the field of metal pollution control because of its potential application. There is a great variety of biomasses used for the biosorption of metals such as micro and macroalgae, yeasts, bacteria, and crustaceans. Adsorbents derived from dead organisms are economically attractive because they provide less expensive materials in comparison to conventional technologies. Algae are of special interest in the search for and the development of new biosorbent materials due to their high sorption capacity and their abundance in many parts of the world’s oceans (Wang and Chen 2009; Naja et al. 2010). Marine algae have been the focus of numerous biosorption studies and their excellent metal-binding capacity has been well documented (Davis et al. 2000; Hashim and Chu 2004; Sari et al. 2011). Seaweeds possess a high metal-binding capacity with the cell wall playing an important role in binding. This is due to the presence of various functional groups such as carboxyl, amino, sulfonic, and hydroxyl groups, which can act as binding sites for metals (Murphy et al. 2007). The biosorption of metals involves several mechanisms that differ qualitatively and quantitatively according to the species used and the origin of the biomass and its processing procedure. Metal sequestration during biosorption follows complex mechanisms, primarily ion exchange, chelation, adsorption by physical forces, and ion entrapment in the intra/interfibrillar capillaries and spaces of the structural polysaccharide network (Ofer et al. 2003). While several publications have reported the screening of the biosorptive properties of different types of biomass for chromium biosorption (Bishnoi et al. 2007; Batista et al. 2009; Vilar et al. 2007), few have focused on developing methods to predict metal binding under different conditions (e.g., metal concentrations, pH, ionic strength). Yun et al. (2001) proposed an equilibrium model for trivalent chromium adsorption using the brown seaweed Ecklonia sp. considering the binding sites present in the biomass responsible for biosorption and also the hydrolysis reactions that trivalent chromium undergoes in the aqueous phase. In this context, the scope of this paper was to investigate the trivalent chromium uptake by the Laminaria digitata

Turning Laminaria digitata seaweed into a resource

seaweed considering chromium speciation in the aqueous phase and its affinity constants to the binding sites present on the biomass surface. The diffusivity of the species (Cr3?, CrOH2?, and H?) inside the biomass particles, considering a non-rigid, low-porosity structure, similar to a gel matrix, and distribution profile of these species on the binding site as a function of pH were also determined.

Materials and methods

957

experiment was performed using the same amount of biomass (1.6 g) but using an initial chromium concentration of 2.69 mmol L-1 (140 mg L-1) without pH control during the sorption process (pHinitial 3.6). The suspension was mechanically stirred at 150 rpm (VWR VOS). The pH was maintained at the desired value, if necessary, using HNO3 or NaOH solutions. Samples were collected at predetermined time intervals and filtered through cellulose acetate membrane filters (Sartorius Stedim). The chromium concentration was determined by atomic absorption spectrometry (AAS, GBC 932 Plus).

Biomass preparation Equilibrium experiments Laminaria digitata seaweed biomass was collected at the Northern coast of Portugal, washed with tap water, sundried, and cut into pieces of 0.5–1 cm length. The protonated biomass was prepared by washing 10 g L-1 of biomass with 0.2 M HNO3 for two cycles of 3 h each. The biomass was then washed with distilled water until pH 4 and dried in oven at 45 °C for 24 h. Potentiometric titration The potentiometric titration was performed using an automatic titration system (Metrohm, 702 SM Titrino) and a shaker module (Metrohm, 728 stirrer). The pH electrode was calibrated with buffer solutions of pH 1.00, 4.01, 7.00, and 9.00. For each titration, 0.25 g of protonated algal biomass was added to 50 mL of a 0.1 mol L-1 NaCl solution, to adjust the ionic strength, and this suspension was then placed in a titration thermostatic cell at 25 °C. The titration was performed with additions of 0.02 mL of 0.1 mol L-1 NaOH solutions to the cell until pH 10, while the suspension was stirred under nitrogen atmosphere. When the drift between each addition was less than 0.5 mV min-1 or 30 min break time between each NaOH addition, it was considered that equilibrium had been reached. Equilibrium pH values and the volume of NaOH added were continuously recorded.

Equilibrium isotherms at pH 2.4 and 4.0 were performed using the following procedure: approximately 100 mg of dried protonated biomass was added to 50 mL of chromium solution (0.1–4.8 mmol L-1, 5–250 mg L-1) in 100 mL Erlenmeyer flasks. The system was agitated on an orbital shaker (VWR Advanced Digital System) at 150 rpm for 48 h. The pH was maintained at the desired value using HNO3 or NaOH solutions. To observe the effect of pH on the trivalent chromium maximum uptake capacity, 100 mg of dried protonated biomass was added to 50 mL of chromium solution (4.8 mM; *250 mg L-1) at eight different initial pH values (1.0, 1.5, 2.0, 2.5, 3.0, 3.5, 4.0, and 4.5) in 100 mL Erlenmeyer flasks. In this equilibrium experiments the pH was not controlled. The system was agitated on an orbital shaker (VWR Advanced Digital System) at 150 rpm for 48 h. After sorption, the final pH of all solutions was measured. All the samples were filtered through cellulose acetate membrane filters (Sartorious Stedim). The experiments were conducted in duplicate. The chromium concentration was determined by atomic absorption spectrometry (AAS, GBC 932 Plus).

Mathematical models Quantification of functional groups

Kinetic experiments Solutions of trivalent chromium were prepared in distilled water using Cr(NO3)3 9H2O (Merck, Darmstadt). Kinetics experiments were conducted in a 1 L reactor glass vessel, with a working volume of 800 mL, equipped with a cooling jacket to ensure a constant temperature during the experiments. The biomass (1.6 g) was brought into contact with approximately 4.47 and 4.71 mmol L-1 (*232 and *245 mg L-1) of trivalent chromium at controlled pH (2.5 and 4) and temperature (25 °C). Another two experiments were performed without controlling the pH during the sorption process (pHinitial 2.5 and 4). Another kinetic

Since the algal surface is polyfunctional, each functional group reacts with protons with different affinities. The distribution of affinities defines the chemical heterogeneity of the active sites present in the surface of the biomass. The total degree of protonation, hT,H, can be obtained by the Langmuir– Freundlich isotherm, considering a quasi-Gaussian distribution of the affinity constant suggested by Sips (Sips 1948): hT;H ¼

ðKH0 CH ÞmH ; 1 þ ðKH0 CH ÞmH

ð1Þ

where KH0 is the average of the affinity distribution of hydrogen ions, which determines the position of the affinity

123

958

I. M. Dittert et al.

int distributions along the axis log Kint i,H (Ki,H is the intrinsic proton affinity constant at each binding site i), CH is the proton concentration in the solution and mH is related to the width of the Sips distribution, which can have values between 0 and 1, which represent an infinite width and a null width, respectively. It should be noted that the parameter mH is a measure of the overall heterogeneity which includes the chemical heterogeneity. If the distribution of affinity displays more than one peak, then the charge of the acid surface, QH, is expressed as a weighted sum of the charge contributions of each active site X QH ¼ Qmax;j ð1  ðhT;H Þj Þ: ð2Þ

j

According to the FTIR spectra obtained in a previous work (Dittert et al. 2012), two main binding groups, carboxyl and hydroxyl, were identified in the surface of the macro-algae. Assuming the presence of two different types of ligands, carboxyl (j = 1) and hydroxyl (j = 2), the equation of the continuous model (Koopal et al. 2005; Milne et al. 1995) is obtained: QH ¼

Qmax;1 Qmax;2 0 C ÞmH;1 þ 1 þ ðK 0 C ÞmH;2 ; 1 þ ðK1;H H 2;H H

ð3Þ

where Qmax,1 and Qmax,2 are the maximum concentration of carboxyl and hydroxyl groups, respectively. The mathematical model was fitted to the experimental data obtained from equilibrium studies using the Excel Solver (Quasi-Newton algorithm). The objective function (Fobj-a) to minimize for optimal regression is n  2 X est Qexp  Q ð4Þ Fobja ¼ H;i H;i ; i¼1

Qexp H,i

where is the experimental charge of an acidic surface, Qest is the estimated charge of an acidic surface, calculated H,i by the model, i is the experimental sample number, and n is the number of samples. In order to determine the errors  0 0 associated with the parameters K1;H ; K2;H ; Qmax;1 ; Qmax;2 ; mH;1 ; mH;2 Þ; the matrix inverse approach (Chapra and Canale 1998) has been used.

Equilibrium model In order to study metal biosorption it is necessary to consider the speciation of metal ions at different pH values. Kocaoba and Akcin (2002) investigated the speciation of chromium in aqueous solution according to the solution pH and found that the predominant species at pH \3 is Cr3?, between 4 and 5 is Cr(OH)2? and at pH [6 occurs the precipitation as Cr(OH)3. Kratochvil et al. (1998) verified

123

that the species adsorbed by Sargassum biomass is not the Cr3? ion but the hydroxocomplex Cr(OH)2?. In fact, in aqueous solutions with pH values below 5, trivalent chromium can only exist in these forms, as expressed by the reaction: KS

2þ þ Cr3þ ðaqÞ þ H2 OðlÞ $ CrOHðaqÞ þ HðaqÞ ; KS ¼

½CrOH2þ ½Hþ  ½Cr3þ 

¼ 103:55 mol=L; ð5Þ 3?

where Ks is the first hydrolysis constant of Cr (Rai et al. 1987). The binding reactions of trivalent chromium with the biomass (B-) and their respective equilibrium constants can be expressed as follows:   K ½Cr1=3 B3 L  Cr3þ 3þ Cr3þ þ 3B $ 3Cr B ; K ¼ 1=3 ðsÞ Cr ðsÞ ðaqÞ ½Cr3þ ½B 3 mol ð6Þ KCrOH2þ

 2þ CrOH2þ ðaqÞ þ 2BðsÞ $ 2ðCrOH Þ1=2 BðsÞ ; KCrOH2þ   2 ½ðCrOH2þ Þ1=2 B L ¼ 2þ  2 mol ½CrOH ½B    ½HB L þ  KH HðaqÞ þ BðsÞ $ HBðsÞ ; KH ¼ þ  ½H ½B  mol

ð7Þ ð8Þ

Here, metal–ligand binding sites were considered as 3Cr1=3 B and 2ðCrOH2þ Þ1=2 B instead of CrB2 and CrOHB2 ; respectively, to emphasize that two or three bonds have to be broken in competitive binding or upon desorption of the metal, as formulated by Yun et al. (2001). When the biomass reacts with the metal in solution, the ligand sites begin to be occupied by Cr3?, CrOH2?, and protons or remain free. Therefore, the mass balance for the sites can be described as: i   h ½BT ¼ ½B  þ ½BH þ Cr1=3 B þ CrOH2þ 1=2 B ð9Þ Substituting Eqs. 5, 6, and 7 in Eq. 8, the concentration of free sites can be expressed as: ½B  ¼

½BT qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi 1 þ KH ½Hþ  þ 3 KCr3þ ½Cr3þ  þ 2 KCrOH2þ ½CrOH2þ  ð10Þ

The amount of chromium adsorbed can be calculated from the bound forms of two chromium species as shown in Eq. 10. i  1 h 1 qCr ¼ Cr1=3 B þ CrOH2þ 1=2 B ð11Þ 3 2 Moreover, combining Eqs. 6, 7, and 10 with Eq. 11 gives:

Turning Laminaria digitata seaweed into a resource

qCr

959

qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi  qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi  ffi  ffi 1 3 KCr3þ Cr3þ þ 12 2 KCrOH2þ CrOH2þ 3 qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi ¼  ffi qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi  ffi 1 þ KH ½Hþ  þ 3 KCr3þ Cr3þ þ 2 KCrOH2þ CrOH2þ ½BT

ð12Þ The concentration of two chromium species appears in Eq. 12. However, only the total trivalent chromium ([Cr3?] plus [CrOH2?]) can be experimentally measured. Thus, it is necessary to convert each species within a term to a total concentration ([Cr]T) using the relation of equilibrium speciation (Eq. 5). The resulting chromium adsorption equilibrium can be expressed as a function of the chromium and proton concentrations: ! rffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi rffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi 1 3 KCr3þ ½CrT 1 2 KCrOH2þ ½CrT þ2 ½BT 3 KS KS 1þ

qCr ¼

1 þ KH ½Hþ  þ

1þ

½Hþ 

sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi 3

KCr3þ ½CrT 1þ

KS þ

H

þ

½Hþ 

sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi 2

KCrOH2þ ½CrT 1þ

KS þ

H

ð13Þ



The mathematical model was fitted to the experimental data obtained from equilibrium studies using the Excel Solver (Quasi-Newton algorithm). The errors associated with the parameters were determined using the matrix inverse approach (Chapra and Canale 1998). Kinetic model For a quantitative description of the biosorption process dynamics, a mass transfer model was developed considering the following assumptions: all of the cell wall binding sites are readily available for metal uptake; there is negligible external film diffusion, corresponding to an adequate stirring rate; the sorption rate is controlled by a linear driving force (LDF) inside the particle; the biosorption process is isothermal and the amount of bound metal is in equilibrium with the metal concentration in the aqueous phase; and the biomass particles are unidimensional thin plates. Therefore, the overall sorption rate is controlled by intraparticle diffusion in the direction normal to the surface of the particles. The partial mass balance of the reactor is given by the following equation: V

dCi dhqi i þ Vðri Þ ¼W dt dt

ð14Þ

where V is the volume of the liquid in the reactor, Ci is the concentration of species i in the liquid phase, t is time, W is the algal mass, hqii is the concentration of species i in the solid phase, and ri is the kinetic rate of the hydrolysis reaction (Eq. 5). The kinetic rate for the disappearance of Cr3? from the solutions is:

rCr3þ

½CrðOHÞ2þ ½Hþ  ¼ k ½Cr   KS

!

3þ

ð15Þ

where k (h-1) is the reaction rate constant, KS (mol/L) is the thermodynamic chromium hydrolysis equilibrium constant, and [Cr3?] and [Cr(OH)2?] are the molar concentrations of chromium species (mol/L). In the simulations, a very large value for k (1000 h-1) was used since chromium speciation in solution is spontaneous. The rate of appearance of the other species is: rCrðOHÞ2þ ¼ rHþ ¼ ðrCr3þ Þ

ð16Þ þ

 If the pH is kept constant in the solution, d½H dt ¼ 0: The mass balance for an ionic species over the particle is represented by:

oqi o2 qi ¼ Dh;i 2 ot oz

ð17Þ

where Dh;i is the coefficient of homogeneous diffusion inside the particle for each species i (cm2 s-1) and z is the distance (expressed in cm) to the symmetry plane. If we consider the average metal concentration inside the particle instead of a concentration profile as in the model above described, we will get the following equation supposing a parabolic profile inside the particle (Glueckauf and Coates 1947):   dhqi i 1 ¼ kp;i  ap qi  hqi i ; ap ¼ ; kp;i  ap dt R Dh;i 3 ¼ 2¼ ; sd;i 3R

ð18Þ

where kp,i is the mass transfer coefficient for intraparticle diffusion of species i (cm s-1) ap is the specific area of the thin plate particles (cm-1), sd,i is the time constant for diffusion of ionic species into the particle (s), qi is the equilibrium concentration in the solid phase (mmol g-1), and R is half of the thin plate thickness (cm). The equilibrium relationships are as follows: qHþ ¼

qCr3þ

½BT KH ½H þ  qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi  ffi qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi  ffi 1 þ KH ½Hþ  þ 3 KCr3þ Cr3þ þ 2 KCrOH2þ CrOH2þ

ð19Þ qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi 3 1 KCr3þ ½Cr3þ  3 ½BT qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi ¼ 1 þ KH ½Hþ  þ 3 KCr3þ ½Cr3þ  þ 2 KCrOH2þ ½CrOH2þ  ð20Þ

qCrðOHÞ2þ

qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi 2 2þ 1 ½B T KCrOH2þ ½CrOH  2 q ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi ¼ ffi qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi   1 þ KH ½Hþ  þ 3 KCr3þ Cr3þ þ 2 KCrOH2þ ½CrOH2þ 

ð21Þ

123

960

I. M. Dittert et al.

Initial conditions: t = 0, [H?] = 10ðpH0 þ3Þ mmol L-1, hqii = 0 mmol g-1; [Cr3?]0 = [Cr]T,0/(1?KS/[H?]0) -1 2? mmol L ; [Cr(OH) ]0 = KS[Cr]T,0/([H?]0?KS) mmol L-1. Numerical solution The system of ordinary differential equations (ODE’s initial value problem) was integrated in time using the solver LSODA (Petzold 1983). This routine solves initial boundary value problems for stiff or non-stiff systems of first-order ODE’s. For non-stiff systems, it makes use of the Adams method with variable order (up to 12-th order) and step size, while for stiff systems it uses the Gear (or BDF) method with variable order (up to 5-th order) and step size. The parameters were obtained by fitting the model to the experimental results by minimizing the residual sum of squares using a successive quadratic programming (SQP) algorithm.

Results and discussion Potentiometric titration From the data obtained by potentiometric titration it was possible to qualitatively and quantitatively determine two main binding groups (carboxyl and hydroxyl groups)

present on the algal surface. The charge of the acid surface, QH, was determined considering the difference between the number NaOH moles consumed during titration with and without algal suspension divided by the mass of algae. Figure 1 shows that the continuous model is able to fit quite well the experimental data, where the proton affinity distribution to the active sites presents two peaks, regarding the carboxyl and hydroxyl groups. The width of the curves is directly related to the heterogeneity of each group. The curve for the hydroxyl groups is wider than that for the carboxyl groups, which means that are more heterogeneous. The protonation constants (pKH), the maximum concentration of carboxyl (Qmax,1) and hydroxyl (Qmax,2) groups and the measure of overall heterogeneity of the active centers for species (mH), are presented in Table 1. It can be seen that the carboxyl groups (2.06 mmol g-1) are present in greater quantity in the Laminaria seaweed than the hydroxyl groups (1.4 mmol g-1). In brown algae, carboxyl groups occur mainly in alginate, a polysaccharide composed of mannuronic and guluronic acids (Schiewer and Wong 1999), which has been shown to play a dominant role in the metal biosorption by brown seaweeds (Fourest 0 and Volesky 1995). The pK1;H value found lies in the range of 3.0-5.0, which are associated to the carboxylic functional group, according to Murphy et al. (2007). Hydroxyl groups have pKH values of [10 (Chojnacka et al. 2005), which is in agreement with the values obtained in this study. The presence of carboxyl and hydroxyl groups was identified by the FTIR technique as reported in a previous study (Dittert et al. 2012). Equilibrium

Fig. 1 Experimental data and model curves for biosorbent potentiometric titrations and  affinity distribution function for hydrogen ions P  int Qmax;i : triangle experimental data (IS = 0.1 M), F ¼ i fi log Ki;H solid line continuous model, dashed line Sips distribution

Biosorption mechanisms involving metals can be complex, depending on the type of biomass used. Different authors as Chojnacka et al. (2005) and Yang and Volesky (1999) reported that the metal uptake by algae is attributed mainly to ion exchange. At pH values below 5, as mentioned previously, trivalent chromium may be present or can be removed from solution as Cr3? or CrOH2?. Since chromium begins to precipitate at pH [5.0, the equilibrium studies were carried out at pH values between 1 and 4, as shown in Fig. 2a, for an initial chromium concentration of 4.8 mmol L-1 (*250 mg L-1). Even using a high concentration of metal, the uptake is highly dependent on the solution pH. Figure 2a shows that biosorption of chromium increases directly with the pH. This effect demonstrates the

Table 1 Parameters of the continuous distribution model for brown algae Laminaria digitata (IS = 0.1 M) Qmax,1 (mmol g-1)

Qmax,2 (mmol g-1)

0 pK1;H

0 pK2;H

mH,1

mH,2

R2

S2R (mmol g-1)2

2.06 ± 0.01

1.4 ± 0.7

3.28 ± 0.01

11 ± 1

0.77 ± 0.01

0.31 ± 0.05

0.999

2.7 9 10-4

123

Turning Laminaria digitata seaweed into a resource

Model

0.8

Experimental data

0.7

qCr (mmol/g)

0.6 0.5 0.4 0.3 0.2 0.1 0 0.5

1

1.5

2

2.5

3

3.5

4

4.5

pH

(b) 0.9 pH = 5.0 0.8

pH = 4.0

0.7 0.6

qCr (mmol/g)

competition between chromium ions and protons for actives sites, which is more important at low pH (Crist et al. 1988). Moreover, the acidic groups at low pH will be protonated according to their pKa values. For pH values higher than pKa (see Table 2), more acidic groups are available (deprotonated) for the chromium ions binding (Oliveira et al. 2011). An 80 % increase in the adsorption capacity was obtained by varying the pH from 1 to 4. Figure 2a, b show that the equilibrium model fits well the equilibrium experimental data at different pH values and equilibrium trivalent chromium concentrations. Considering the first hydrolysis constant (pKs) and the data obtained from the potentiometric titration (Qmax,1 and 0 pK1;H ), model equilibrium parameters for the species present in solution were estimated, as shown in Table 2. 0 , The values of BT and pKH correspond to Qmax,1 and pK1;H respectively, according to data obtained from the potentiometric titration. Figure 2b also shows the simulated equilibrium profile for pH 2.6 and 5.0, to show that for pH values higher than 5 and high metal concentrations, the increase in adsorption capacity is negligible, although for low pH values an increase of 0.2 in the pH value increases significantly the adsorption capacity, associated to the increase of the amount of available binding sites. However, as shown in Fig. 2c, the chromium uptake capacity increases substantially with the chromium concentration in 0 solution for pH values higher than pK1;H , since more free carboxylic groups are available for metal binding. At pH 4.0, the maximum Cr uptake was estimated to be 0.77 mmol g-1 (*40 mg g-1) for a total Cr concentration of 3.6 mM (*187 mg L-1), corresponding to 2.31 meq g-1 when only Cr3? species was considered. However, at pH 4.0, around 74 % of the Cr is present as CrOH2? (Dittert et al. 2012), corresponding to 1.74 meq g-1 (number of sites occupied by chromium). This value indicates that at this pH there are still sites available, since the amount of carboxylic groups is 2.06 meq g-1 (Table 2). According to Table 2, Cr3? has a slight higher affinity to the carboxylic groups than CrOH2?, which can be correlated to the charge of the ion. The same behavior was observed for trivalent chromium biosorption using protonated brown algae Ecklonia (Yun et al. 2001) and Pelvetia canaliculata (Vilar et al. 2012). The pH plays an important role in the distribution of chromium species in solution and binding sites present at the surface of the biomass. Figure 3 present the influence

(a) 0.9

pH=4.0 pH=2.4 Model

0.5 0.4

pH = 2.6

0.3

pH = 2.4 0.2 0.1 0

0

0.5

1

1.5

2

2.5

3

3.5

CCr (mmol/L)

(c)

0.9

[Cr]=0.1 mM 0.8

[Cr]=0.4 mM

0.7

[Cr]=1.0 mM [Cr]=2.0 mM

qCr (mmol/g)

Fig. 2 a Equilibrium sorption experimental and model predicted data c points at different pH values ([Cr(total)]initial = 4.8 mM), b Equilibrium sorption isotherms at two pH values and model predicted curves (for pH 2.6 and 5 only the model curve is presented), c Predicted concentration profiles as a function of pH for different equilibrium chromium concentrations

961

0.6

[Cr]=3.6 mM

0.5 0.4 0.3 0.2 0.1 0

1

1.5

2

2.5

3

3.5

4

4.5

5

pH

123

962

I. M. Dittert et al.

BT = Qmax,1, pKH ¼

4.1 ± 0.3

3.9 ± 0.4

pKS

3.55

R2

S2R (mmol g-1)2

0.871 9.2 9 10

-3

0 pK1;H

of pH (1–5), chromium concentration (0.1, 1.0, and 3.6 mM) and speciation in solution on the chromium speciation at the binding sites and respective uptake capacity. The Cr3? species begins to occupy the binding sites at pH 1.5. With increasing pH a greater amount of Cr3? species occupy the sites of the biomass and a maximum binding of this species is observed at pH 3 to 3.5, as the total chromium concentration decreases (Fig. 3a–c). The binding of species CrOH2? gradually increases as the pH is increased and, in general, begins at pH 2.0, but at lower levels than Cr3? until to pH around 3.5 when its binding exceeds that of Cr3?. According to the hydrolysis reaction, the concentrations of both species in the aqueous phase are the same at pH 3.55, and as the affinity of both species to the carboxylic groups are very similar, the uptake capacity of both chromium species are very similar in the pH range from 3.3 to 3.8, according to the equilibrium trivalent chromium in solution. However, for higher pH values, the CrOH2? species are preferentially linked to the binding sites. Figure 3a shows that only 10 % of the total binding sites is available for chromium equilibrium concentration of 3.6 mM (*187 mg L-1) at pH values higher than 3.3. The biosorption capacity increases with increasing chromium concentration in all situations and reaches a plateau at pH [4.0.

0.8

1.8

0.7

1.6

(CrOH)1/2B

1.4

0.6

1.2

0.5

1

0.4

0.8

0.3

Cr1/3B

0.6

0.2

0.4

B0.1

0.2 0

0

1

1.5

2

2.5

3

3.5

4

4.5

5

pH

(b) 2.2

0.9 BH

2

0.8

q

1.8

0.7

1.6 0.6

1.4

(CrOH)1/2B

1.2

0.5

1

0.4

0.8

0.3

Cr1/3B

0.6

0.2

B-

0.4

0.1

0.2 0

0

1

1.5

2

2.5

3

3.5

4

4.5

5

pH

(c) 2.2

0.9 BH

2

0.8

Mathematical models that describe the transient behavior of biosorption in batch processes under different experimental conditions are very useful in optimizing processes and for large-scale studies. Figures 4, 5, and 6 show that the mass transfer model described in ‘‘Kinetic model’’ section fits well the experimental kinetic data at different initial pH values and initial chromium concentration. In these figures the predicted kinetic profiles at the liquid and solid phase for Cr3? and CrOH2? species are also shown. The experimental data presented in Figs. 4 and 5 were obtained at controlled pH conditions. However, the kinetic data presented in Fig. 6 were obtained without pH control. At pH 2.5 (Fig. 4a) it can be observed that the amount of chromium bound to the actives is higher in the form of Cr3?, since 92 % of the total chromium in solution is in the form of Cr3?. In the other hand, at pH 4.0, 73 % of the total chromium in solution is in the form

123

Concentration (mmol/g)

1.8

Kinetics

Cr uptake (mmol/g)

3.28 ± 0.01

pKCrðOHÞ2þ

Cr uptake (mmol/g)

pKCr3þ

0.7

1.6 0.6

1.4

q

0.5

1.2 1

0.4

B-

0.8

0.3

(CrOH)1/2B

0.6

Cr uptake (mmol/g)

2.06 ± 0.01 a

pKH a

0.9

q

BH

2

Concentration (mmol/g)

BaT (mmol g-1)

(a) 2.2

Concentration (mmol/g)

Table 2 Estimated model equilibrium parameters for trivalent chromium biosorption onto brown algae Laminaria digitata

0.2

0.4

Cr1/3B

0.1

0.2 0

0

1

1.5

2

2.5

3

3.5

4

4.5

5

pH Fig. 3 Speciation of binding sites as a function of the solution pH: a [Cr(total)] = 3.6 mM, b [Cr(total)] = 1.0 mM, and c [Cr(total)] = 0.1 mM

of CrOH2?, and as illustrated in Fig. 5a, higher fraction of total chromium bound to the actives sites is in the form of CrOH2?.

Turning Laminaria digitata seaweed into a resource

963

(a) 1.0

0.5

(a) 1.0

0.8

0.4

0.8

0.8 0.7

0.4

0.2

0.2

0.1

0.0

0.0

0.5

0.6

0.4 0.4

0.3

qCr (mmol/g)

0.3

CCr /CCr,0

0.6

qCr (mmol/g)

CCr /CCr,0

0.6

0.2 0.2 0.1

30

40

50

0.0

60

(b)1.0

0.30

0.9 0.8

qCr3+

CCr /CCr,0 CCr 3+/CCr,0

qCrOH2+

CCr(OH)2+/CCr,0

qCr, Total

0.4

30

40

50

2.45 0.15

2.40 0.10

0.8 Exp. data (CCr /CCr,0) CCr /CCr,0 CCr3+/CCr,0

0.8

CCr(OH)2+/CCr,0

2.35

0.7 0.6 0.5

0.7 qCr3+

0.6 0.5

0.3

qCr, Total

0.4

0.2

3.8

pH

0.05

0.3

0.1

0

10

20

30

40

0.1

0.2

0.00

3.7

50

0

Time (hours)

Figure 4b shows the simulated profiles in the absence of pH control. It can be observed that within the first 10 h, the pH decreases from 2.50 to 2.36, associated to ion exchange mechanism between chromium ions in solution and hydrogen ions on the solid phase. Without pH control, there is a decrease in the total chromium biosorption capacity due to the release of hydrogen ions to the solution exchanged with the chromium ions, leading to a decrease in the pH value, increasing the competition to the binding sites and at same time decreasing the number of binding sites available. The simulated pH profile presented in Fig. 5b shows a slight increase in pH at the beginning of the experiment which is related to the need for protons in the hydrolysis reaction. The subsequent decrease is attributed to the ion exchange between hydrogen and chromium ions until stabilizing at a value of 3.75 after 15 h. In this case, the decrease in pH did not affected the chromium uptake capacity because in the range of those pH values all the

20

30

40

0.0

50

Time (hours)

Fig. 5 Kinetics of chromium biosorption onto algae Laminaria (pHi = 4.0; T = 25 °C). [Cr(total)] = 4.47 mmol L-1. square experimental data with pH control, dashed line Cr(OH)2?, dotted dashed line Cr3?. Simulated profiles with (a) and without (b) pH control

qCr3+

1.0

3.6

0.6

3.5

0.5

3.4

0.4

qCrOH

2+

Exp. data (CCr /CCr,0 )

qCr, Total

0.9

CCr /CCr,0

0.8

CCr 3+/CCr,0 CCr(OH)2+/CCr,0

0.7

CCr/CCr,0

Fig. 4 Kinetics of chromium biosorption onto algae Laminaria (pHi = 2.5; T = 25 °C). [Cr(total)] = 4.71 mmol L-1. square experimental data with pH control, dashed line Cr(OH)2?, dotted dashed line Cr3?. Simulated profiles with (a) and without (b) pH control

10

3.3

0.6

pH

2.30

0.0

0.4

3.9

qCrOH2+

0.3 0.2

4.1

0.9

4.0

CCr/CCr,0

Exp. data (CCr /CCr,0)

pH

CCr/CCr,0

pH

qCr (mmol/g)

0.20

0.5

20

(b) 1.0

0.25

2.50

0.6

10

Time (h) 2.55

0.7

0.0 0

Time (h)

qCr (mmol/g)

20

0.3

0.5 3.2

0.2

3.1

0.1

3.0

0.0

qCr (mmol/g)

10

pH

0

0.4 pH

0.3

Exp. data (pH)

0.2 0

10

20

30

40

50

60

70

Time (hours)

Fig. 6 Kinetics of chromium biosorption onto algae Laminaria (pHi = 3.6; T = 25 °C). [Cr(total)] = 2.69 mmol L-1. sqaure experimental data without pH control, dashed line Cr(OH)2?, dotted dashed line Cr3?

123

964

I. M. Dittert et al.

Table 3 Estimated parameters for the mass transfer model [Cr]T (mmol L-1)

pH

4.71

2.5b 2.5

T (°C)

c

25

Cr3?

CrOH2?

H?

kp,i 9 ap (h-1)

Dah,i (cm2 s-1)

kp,i 9 ap (h-1)

Dah,i (cm2 s-1)

kp,i 9 ap (h-1)

2.7

5.6 9 10-8

0.5

1.0 9 10-8

–

2.1

-8

1.7

-8 -9

4.4 9 10

3.5 9 10

Dah,i (cm2 s-1) 6.7 9 10-3

–

5.0

1.0 9 10

-7

2.1 9 10-2

4.47

b

4.0 4.0c

2.7 2.0

5.6 9 10 4.2 9 10-8

0.2 0.2

4.2 9 10 4.2 9 10-9

– 5.0

– 1.0 9 10-7

1.7 9 10-2 2.8 9 10-2

2.69

3.6c

1.8

3.8 9 10-8

0.05

1.0 9 10-9

5.0

1.0 9 10-7

3.0 9 10-2

a

-8

S2R (mmol L-1)2

R = 0.05 mm, V = 0.8 L, W = 1.6 g

b

pH controlled

c

pH not controlled

carboxylic groups are deprotonated and available for the chromium ions binding. Figure 6 presents the kinetic experimental data obtained without pH control using a lower initial chromium concentration. The pH varied during the experiment and the simulated profile provided a very good fit for the first few hours, fluctuating by less than 0.1 pH unit for the remained of the experiment. Table 3 presents the intraparticle homogeneous diffusion coefficient (Dh) of the species Cr3?, CrOH2?, and H? onto the protonated algae for different pH values and chromium concentrations (thickness of the thin plate particles as 0.1 mm). Dh is about 3.8–5.6 9 10-8 cm2 s-1 for Cr3?, 0.1–3.5 9 10-8 cm2 s-1 for CrOH2?, and 1.0 9 10-8 cm2 s-1 for H? indicating that Cr3? diffuses faster than CrOH2?, which is in agreement with the nightingale hydrated ion radii (rhyb) (Marcus and Kertes 1969) and molecular diffusion coefficients given by Nernst–Haskell expression (Dean 1979; Reid et al. 1987). Vilar et al. (2012) reported average values of 4.6 9 10-7 cm2 s-1 to Cr3? and 1.8 9 10-8 cm2 s-1 to CrOH2? at 25 °C and pH 4.0, which are in the same magnitude order of those presented in this study.

Conclusions The equilibrium model based on the trivalent chromium speciation in aqueous solutions and chemical interaction with the carboxyl groups present in the surface of the biomass was successfully applied to the isothermal equilibrium, according to variations in the pH and total chromium concentration. The Cr3? species revealed a higher affinity to the carboxylic groups than CrOH2? at the same species concentration, although for pH values higher than 3.5 hydroxyl chromium species preferentially bind to the sites present on the biomass surface, due to its higher concentration in solution. From the kinetic results it was

123

shown that Cr3? species has a higher diffusivity compared to CrOH2? species, within the pH range studied. Acknowledgments The authors are grateful to the project of International Cooperation, Edital-CGCI-010/2009, Projeto CAPES/ FCT no. 279/2010, financed by CAPES-Brazil and FCT-Portugal. This work was also partially supported by project PEst-C/EQB/ LA0020/2011, financed by FEDER through COMPETE—Programa Operacional Factores de Competitividade and by FCT—Fundac¸a˜o para a Cieˆncia e a Tecnologia. Ingrid M. Dittert also acknowledges her Doctoral fellowship provided by CAPES. V.J.P. Vilar acknowledges financial support from Programme Cieˆncia 2008 (FCT).

References Batista APS, Roma˜o LPC, Arguelho MLPM, Garcia CAB, Alves JPH, Passos EA, Rosa AH (2009) Biosorption of Cr(III) using in natura and chemically treated tropical peats. J Hazard Mater 163(2–3): 517–523. doi:10.1016/j.jhazmat.2008.06.129 Bishnoi NR, Kumar R, Kumar S, Rani S (2007) Biosorption of Cr(III) from aqueous solution using algal biomass spirogyra spp. J Hazard Mater 145(1–2):142–147. doi:10.1016/j.jhazmat.2006.10.093 Chapra SC, Canale RP (1998) Numerical methods for engineers, 3rd edn. McGraw-Hill, New York Chojnacka K, Chojnacki A, Go´recka H (2005) Biosorption of Cr3?, Cd2? and Cu2? ions by blue–green algae Spirulina sp.: kinetics, equilibrium and the mechanism of the process. Chemosphere 59 (1):75–84. doi:10.1016/j.chemosphere.2004.10.005 Crist RH, Oberholser K, Schwartz D, Marzoff J, Ryder D, Crist DR (1988) Interactions of metals and protons with algae. Environ Sci Technol 22(7):755–760. doi:10.1021/es00172a002 Davis TA, Volesky B, Vieira RHSF (2000) Sargassum seaweed as biosorbent for heavy metals. Water Res 34(17):4270–4278. doi: 10.1016/s0043-1354(00)00177-9 Dean JA (ed) (1979) Lange’s handbook of chemistry, 12th edn. McGraw-Hill Book Company, New York Di Natale F, Lancia A, Molino A, Musmarra D (2007) Removal of chromium ions form aqueous solutions by adsorption on activated carbon and char. J Hazard Mater 145(3):381–390. doi: 10.1016/j.jhazmat.2006.11.028 Dittert IM, Vilar VJP, da Silva EAB, de Souza SMAGU, de Souza AAU, Botelho CMS, Boaventura RAR (2012) Adding value to marine macro-algae Laminaria digitata through its use in the separation and recovery of trivalent chromium ions from aqueous solution. Chem Eng J 193–194:348–357. doi:10.1016/j.cej.2012.04.048

Turning Laminaria digitata seaweed into a resource Fourest E, Volesky B (1995) Contribution of sulfonate groups and alginate to heavy metal biosorption by the dry biomass of sargassum fluitans. Environ Sci Technol 30(1):277–282. doi:10. 1021/es950315s Glueckauf E, Coates JI (1947) 241. Theory of chromatography. Part IV. The influence of incomplete equilibrium on the front boundary of chromatograms and on the effectiveness of separation. J Chem Soc (Resumed) Hashim MA, Chu KH (2004) Biosorption of cadmium by brown, green, and red seaweeds. Chem Eng J 97(2–3):249–255. doi: 10.1016/s1385-8947(03)00216-x Kocaoba S, Akcin G (2002) Removal and recovery of chromium and chromium speciation with MINTEQA2. Talanta 57(1):23–30. doi:10.1016/s0039-9140(01)00677-4 Koopal LK, Saito T, Pinheiro JP, Riemsdijk WHv (2005) Ion binding to natural organic matter: general considerations and the NICA– Donnan model. Colloids Surf A 265(1–3):40–54. doi:10.1016/j. colsurfa.2004.11.050 Kratochvil D, Pimentel P, Volesky B (1998) Removal of trivalent and hexavalent chromium by seaweed biosorbent. Environ Sci Technol 32(18):2693–2698. doi:10.1021/es971073u Marcus Y, Kertes AS (1969) Ion exchange and solvent extraction of metal complexes. Wiley Interscience, London Milne CJ, Kinniburgh DG, de Wit JCM, van Riemsdijk WH, Koopal LK (1995) Analysis of proton binding by a peat humic acid using a simple electrostatic model. Geochim Cosmochim Acta 59(6): 1101–1112. doi:10.1016/0016-7037(95)00027-w Murphy V, Hughes H, McLoughlin P (2007) Cu(II) binding by dried biomass of red, green and brown macroalgae. Water Res 41(4):731–740. doi:10.1016/j.watres.2006.11.032 Naja G, Murphy V, Volesky B (2010) Biosorption, metals. Encyclopedia of industrial biotechnology: bioprocess, bioseparation, and cell technology 1–29 Ofer R, Yerachmiel A, Shmuel Y (2003) Marine macroalgae as biosorbents for cadmium and nickel in water. Water Environ Res 75:246–253 Oliveira RC, Palmieri MC, Garcia Jr O (2011) Biosorption of metals: state of the art, general features, and potential applications for environmental and technological processes. Progress in biomass and bioenergy production. InTech, Araraquara Petzold L (1983) Automatic selection of methods for solving stiff and nonstiff systems of ordinary differential-equations. Siam J Sci Stat Comput 4(1):136–148 Rai D, Sass BM, Moore DA (1987) Chromium(III) hydrolysis constants and solubility of chromium(III) hydroxide. Inorg Chem 26(3):345–349. doi:10.1021/ic00250a002

965 Reid RC, Prausnitz JM, Poling BE (1987) The properties of gases & liquids, 4th edn. McGraw-Hill Book Company, New York Saravane R, Sundararajan T, Reddy SS (2001) Chemically modified low cost treatment for heavy metal effluent management. Environ Manag Health 12:215–224 ¨ D, Tu¨zen M (2011) Equilibrium, thermodynamic Sari A, Uluozlu¨ O and kinetic investigations on biosorption of arsenic from aqueous solution by algae (Maugeotia genuflexa) biomass. Chem Eng J 167(1):155–161. doi:10.1016/j.cej.2010.12.014 Schiewer S, Wong MH (1999) Metal binding stoichiometry and isotherm choice in biosorption. Environ Sci Technol 33(21): 3821–3828. doi:10.1021/es981288j Shanker AK, Venkateswarlu B (2011) Chromium: environmental pollution, health effects and mode of action. In: Editor-in-Chief: Jerome ON (ed) Encyclopedia of environmental health. Elsevier, Burlington, pp 650–659 Sips R (1948) On the structure of a catalyst surface. J Chem Phys 16:490–495 Vaiopoulou E, Gikas P (2012) Effects of chromium on activated sludge and on the performance of wastewater treatment plants: a review. Water Res 46(3):549–570. doi:10.1016/j.watres.2011.11. 024 Vilar VJP, Botelho CMS, Boaventura RAR (2007) Chromium and zinc uptake by algae Gelidium and agar extraction algal waste: kinetics and equilibrium. J Hazard Mater 149(3):643–649. doi: 10.1016/j.jhazmat.2007.04.023 Vilar VJP, Valle JAB, Bhatnagar A, Santos JC, Guelli U, de Souza SMA, de Souza AAU, Botelho CMS, Boaventura RAR (2012) Insights into trivalent chromium biosorption onto protonated brown algae Pelvetia canaliculata: distribution of chromium ionic species on the binding sites. Chem Eng J 200–202:140–148. doi:10.1016/ j.cej.2012.06.023 Wang J, Chen C (2009) Biosorbents for heavy metals removal and their future. Biotechnol Adv 27(2):195–226. doi:10.1016/jbiotechadv. 2008.11.002 Williams CJ, Edyvean RGJ (1997) Optimization of metal adsorption by seaweeds and seaweed derivatives. Process Saf Environ Prot 75(1):19–26. doi:10.1205/095758297528733 Yang J, Volesky B (1999) Modeling uranium-proton ion exchange in biosorption. Environ Sci Technol 33(22):4079–4085. doi:10.1021/ es990435q Yun Y-S, Park D, Park JM, Volesky B (2001) Biosorption of trivalent chromium on the brown seaweed biomass. Environ Sci Technol 35(21):4353–4358. doi:10.1021/es010866k

123