Biosorption of Fe3+ from aqueous solution by a bacterial dead Streptomyces rimosus biomass

Biosorption of Fe3+ from aqueous solution by a bacterial dead Streptomyces rimosus biomass

Process Biochemistry 39 (2004) 1643–1651 Biosorption of Fe3+ from aqueous solution by a bacterial dead Streptomyces rimosus biomass A. Selatnia∗ , A...

144KB Sizes 0 Downloads 15 Views

Process Biochemistry 39 (2004) 1643–1651

Biosorption of Fe3+ from aqueous solution by a bacterial dead Streptomyces rimosus biomass A. Selatnia∗ , A. Boukazoula, N. Kechid, M.Z. Bakhti, A. Chergui Département de Génie Chimique, Ecole Nationale Polytechnique d’Alger, 10 Avenue Pasteur, Belfort, EL-Harrach, Alger, Algeria Received 7 March 2003; accepted 13 July 2003

Abstract The iron biosorption capacity of a Streptomyces rimosus biomass treated with NaOH was studied in batch mode. After pretreatment of biomass at the ambient temperature, optimum conditions of biosorption were found to be: a biomass particle size between 50 and 160 ␮m, an average saturation contact time of 4 h, a biomass concentration of 3 g/l and a stirring speed of 250 rpm. The equilibrium data could be fitted by Langmuir isotherm equation. Under these optimal conditions, 122 mg Fe /gbiomass were fixed. © 2003 Elsevier Ltd. All rights reserved. Keywords: Biosorption; Iron; Streptomyces rimosus; Waste-water treatment; Batch processing

1. Introduction Many industries such as coatings, car, aeronautic and steel industries generate large quantities of waste water containing various concentrations of iron. These concentrations are usually too low to be treated by standard methods. Chemical precipitation lead to the production of toxic sludge. Solvent extraction techniques are not profitable for a stream containing less than 1 g/l of targeted heavy metals. On the other hand, ion exchange processes are too expensive due to the high cost of synthetic resins. Antibiotic fermentation produces large amounts of semi-solid wastes which are normally disposed by incineration. Such semi-solid wastes, after granulation and a heat treatment, have been used to recover and remove heavy metals from waste water streams [1–4]. These researchers tested several filamentous fungi (Mucor miehei, Penicillium chrysogenum, A. niger and Rhizopus arrhizus). In the literature, the capability of either living or nonliving organisms for fixing metal ions is widely described. Modak et al. [5] showed that nonliving A. niger biomass attached to wheat bran was selective for the extraction of copper and zinc. Guibal et al. [6] studied the biosorption of uranium by filamentous fungus Mucor and Gardea-Torresdey et al.



Corresponding author. Fax: +213-21-52-29-73. E-mail address: ammar [email protected] (A. Selatnia).

0032-9592/$ – see front matter © 2003 Elsevier Ltd. All rights reserved. doi:10.1016/S0032-9592(03)00305-4

[7] performed batch experiments with inactivated cells of M. rouxii for Cu(II) binding. Other studies were performed with different biomaterials as marine [8–12], bacteria [13–21], Chitosan [22]; humic substances [23] and sewage sludge [24,25]. All these studies were done to remove and recover heavy metals from dilute aqueous streams by biosorption. In the present work, dead Streptomyces rimosus biomass treated with NaOH (0.1 M) was used in granulated form to remove Fe3+ from laboratory-made aqueous solutions. The effects of the physical and chemical parameters on the performances of the biosorption were studied and the mechanism and the kinetics of the iron binding were determined.

2. Diffusion and sorption models 2.1. Sorption isotherms The equilibrium of a solute separated between the liquid and solid phases is described by various models of sorption isotherms such as the Langmuir and Freundlich models. These models suggest a monolayer sorption, with lateral interactions between the sorbed molecules in the case of the Freundlich model: the energetic distribution of sites is heterogenous, due to the diversity of sorption sites or the diverse nature of the metal ions sorbed, free or hydrolysed

1644

A. Selatnia et al. / Process Biochemistry 39 (2004) 1643–1651

species. Langmuir model hypotheses suppose a monolayer sorption with an homogeneous distribution of sorption sites and sorption energies, without interactions between the sorbed molecules. The Langmuir model is described by the following equation: qe = qm

bCe 1 + bCe

(1)

where qe is the adsorbed metal ion quantity per gram of biomass at equilibrium (mg g−1 ); qm the maximum amount of metal ion per unit weight of biomass to form a complete monolayer on the surface bound at high Ce (mg g−1 ); b a constant related to the affinity of the binding sites (l mg−1 ). The Freundlich model equation is of the form qe =

1/n kCe

(2)

where k and n are the Freundlich’s constants characteristic of the system. 2.2. Diffusion models Sorption kinetics are mainly controlled by various steps including diffusion processes. Three steps can be enumerated and applied to iron removal [26–28]. Step 1: Iron transfer from the boundary film bordering to the surface of the particle. Step 2: Transfer of the iron from the surface to the intraparticular actives sites. Step 3: Uptake of metal ion on the active sites, via complexation, sorption and intraparticular precipitation phenomena. Step 1 describes film mass transfer resistance, Step 2 is related to the intraparticular diffusion model, and Step 3 is a rapid, non-limiting phase. Various models of diffusion were studied, including single steps of diffusion external or intraparticle or combined phenomena [29–33]. The objective of this study is not to propose a new diffusion model but to select the main limiting step in the overall uptake mechanism. 2.2.1. External mass transfer diffusion model This model, as an application of the Fick’s laws, expresses the evolution of the concentration of the solute in the solution C (mg/l), as a function of the difference in the concentrations of the metal ion in the solution, C, and at the particle surface, CS (mg/l) and according to Eq. (3) [34–36]. dC = −βS(C − CS ) dt

(3)

where β is the mass transfer coefficient (m s−1 ) and S the specific surface area (m−1 ). The coefficients are determined after making some assumptions such as a surface concentration CS negligible at

t = 0, a concentration in solution tending to the initial concentration C0 , and also negligible intraparticle diffusion. So the previous equation can be simplified to   d(C/C0 ) = −βS (4) dt t→0 The initial rate of sorption, −βS (s−1 ), is obtained by polynomial linearization of C/C0 , and subsequent derivation at t = 0. The specific surface area is approximated as the external surface area. Moreover, the particles are supposed spherical and S, calculated as the external surface compared to the solid/liquid ratio in the solution, gives S=

6m dp ρapp

(5)

where m is the sorbent mass concentration in the solution (kg m−3 ), dp the particle size diameter (m) and ρapp the apparent volume mass of the sorbent (kg m−3 ). 2.2.2. Intraparticular mass transfer diffusion model In this work, the models chosen refer to theories developed by [34,35,37]. According to the intraparticular diffusion model proposed by Weber and Morris [34], the initial rate of intraparticular diffusion is calculated by linearization of the curve q = f(t 0.5 ). q = Ki t 0.5

(6)

where q is the amount of adsorbed metal ion on the biomass at time t (mg g−1 ), t the time (s) and Ki the diffusion coefficient in the solid (mg g−1 s−1/2 ). Another kind of intraparticular diffusion model was proposed by Urano and Tachikawa [37]. In this model, the adsorption rate is considered as independent of the stirring speed, and external diffusion negligible relative to the low overall adsorption rate. The sorption kinetics are modelled according to the following equation.   2  q 4(π)2 Di t −log 10 1 − (7) = qe 2.3dp2 where Di is the diffusion coefficient in the solid (m2 s−1 ). 2.3. Kinetic modelling The first-order rate expression of Lagergreen [38], Ho and Mc Kay [39], and Aksu [40], based on solid capacity is generally expressed as follows:   qe − q k1 −log 10 t (8) = qe 2.3 where k1 is the rate constant of first-order biosorption (s−1 ).

A. Selatnia et al. / Process Biochemistry 39 (2004) 1643–1651 Table 1 The physical and chemical characteristics of the biomass

Particle size(␮m) Humidity (%) Density Specific area (m−1 ) Zeta potential (V)

1645

Table 2 IR absorption bands and corresponding possible groups

Untreated biomass

NaOH-treated biomass

Frequency (cm−1 )

Functional group

50–160 3.2217 0.43367 398.67 −0.062

50–160 4.3586 0.41479 398.67 −0.082

3431.58 2919.49 2852.92 1623.90 1398.58 1111.81 666.29

–OH, –NH –CH –CH –COO− , –C=O –COO− –C–O, –C–N –CH

The pseudo second-order equation is also based on the sorption capacity of the solid phase [40–42]. The pseudo second-order kinetic rate is expressed as     1 1 + k2 t (9) = qe − q qe where k2 is the rate constant of second-order biosorption (g mg−1 s−1 ).

3. Methods and materials S. rimosus biomass produced during oxytetracyclin antibiotic production was collected after fermentation. This biomass is washed with distilled water and dried at 50 ◦ C during 24 h. It is then crushed and sieved in order to select a fraction with particle diameters between 50 and 160 ␮m. This biomass is then treated with NaOH (0.1N) during 30 min and once again dried and sieved to obtain the particle size fraction between 50 and 160 ␮m. The physical and chemical properties of this biosorbant are shown in Table 1. The biosorption tests were carried out in closed system. A small quantity of (FeCl3 ·2H2 O) was added to biomass solution until adsorption equilibrium was reached. Equilibrium and kinetic curves are plotted by monitoring the Fe3+ residual concentration as a function of time. A magnetic stirrer was used to homogenize the mixture. For all the experiments, initial Fe3+ and biomass concentrations were fixed at 100 mg/l and 3 g/l, respectively. Fe3+ residual concentrations in solution were determined using a Unicam939 Atomic Absorption Spectrophotometer with a wavelength at 248.3 nm. All the experiments were carried out at unadjusted pH. Infrared Spectra were obtained with the help of a Perkin-Elmer FTIR1650.

4. Results and discussion 4.1. IR spectral analysis In order to find out which functions are responsible for the iron adsorption, IR analysis of the biomass was carried out. Figs. 1 and 2 show the IR spectra and the various functional groups corresponding to the absorption bands. The whole of

the frequencies of vibrations and their corresponding groups are gathered in Table 2. 4.2. Iron biosorption kinetics As can be seen from Fig. 3, iron biosorption kinetics is very fast—suggesting very active surface phenomena of the biomass. In fact, this biomass cell walls are made of great molecules (peptidoglycane) linked with techo¨ıd acid and polysaccharides. These molecules possess functional groups which can adsorb heavy metals. These groups are of the type (–NH), carboxylate anions (–COO− ), hydroxy (–OH) and others (–C–N), (–C–O), (–C–H), (–C=O), which present different affinities towards metallic ions. Fig. 3 shows adsorption kinetics of the biomass. Fig. 3 shows that for the NaOH-treated biomass, the adsorption of Fe3+ is total after 4 h. The chemical treatment of the biomass with NaOH (0.1N) shows that in the sodium form, the ion exchange sites are more easily able to exchange this cation (Na+ ) with ions Fe3+ than when the ion exchange sites are protonated. The metal biosorption depends strongly on the protonation or unprotonation of the functional groups on the cell wall (i.e. carboxylic, hydroxyl and amino groups) [6,13,43,44]. The ionic forms of the metal in solution and the electrical charge of the biomass depend on the solution pH. pH variations during the experiment are shown in Fig. 3. The following deductions can be made: • The adsorbed quantity tends to a value 33.32 mg Fe / gbiomass , corresponding to pH of 4.8. At the very beginning of the experiment, the pH falls from 9.45 to a value of 3. This can be explained by H+ liberation of some compounds of the biomass in solution suggesting that the biosorption mechanism is a ion exchange type between H+ ions and Fe3+ ions. • After a short time, solution pH stabilizes at the value of 4.8 and other biosorption mechanisms can occur such as complexation, electrostatic attraction and ion exchange between Na+ ions and Fe3+ ions. In our study, the biosorption mechanisms like ion exchange and complexation as well as electrostatic attractions

1646

A. Selatnia et al. / Process Biochemistry 39 (2004) 1643–1651

Fig. 1. Spectrum 1: Infrared spectrum of the NaOH-treated biomass.

seem to be the most occurring phenomena. Among the chemical groups possibly involved in such phenomena’s are the (–COO− ), (–OH), (–NH), (C–O–), (–C=O) groups, which are active sites for the Fe3+ sorption. 4.3. Biosorption rate constant In order to find out the Fe3+ biosorption kinetic parameters, many experiments were conducted under optimal conditions with NaOH-treated biomass. Fig. 4 shows a plot of the function log10 [(qe −q)/qe )] as a function of time for the NaOH-treated biomass. One can see that the adsorption kinetics is the first order. The corresponding rate constant is k1 = 4.1 × 10−4 s−1 . 4.4. External mass transfer coefficient By tracing the slope values at C = 0 of the plot (C/C0 ) = f(t), we can deduce the values of the external mass transfer coefficients β. The corresponding value of the external mass transfer is β = 1.95 × 10−6 m s−1 . This low value means that the resistance to the external mass transfer is quite important (Fig. 5).

4.5. Intraparticle diffusion coefficient The rate constant of intraparticle diffusion Ki has been determined by a plot q = f(t 0.5 ) by taking account only of the initial period according to Weber and Morris model [34]. The intraparticle diffusion coefficient Di was computed by plotting log10 [1−(q/qe )2 ] as a function of the time according to Urano and Tachikawa model [37]. Values of Ki and Di are given in Table 3. The low values of Ki = Di coefficients suggest that the intraparticle diffusion is negligible in comparison with the external mass transfer phenomena and the adsorption at the surface of the biomass (Figs. 6 and 7). 4.6. Effect of initial pH on Fe3+ biosorption Fig. 8 shows that the Fe3+ adsorbed quantity increases with increasing initial solution pH. That fact can be exTable 3 Values of Ki and Di for the NaOH-treated biomass Ki (mg g−1 s−1/2 ) Di (m2 s−1 )

0.775 1 × 10−13

A. Selatnia et al. / Process Biochemistry 39 (2004) 1643–1651

1647

Fig. 2. Spectrum 2: Infrared spectrum of the untreated biomass.

Time (min)

0 Log10[(qe-q)/qe)]

plained by the strong relation between the biosorption and the number of negative charges at the biomass surface which is itself related with the functional groups. The maximum of the Fe3+ adsorbed quantity at the surface of the biomass is 33.3 mg Fe /gbiomass . This maximum occurs at pH 10. There is a sharp increase of the adsorbed quantity from pH 4, which correspond to the dissociation of functional group or their deprotonation.

-0,5 0

50

100

150

200

250

-1 -1,5 -2 -2,5

Fig. 4. log10 [(qe −q)/qe ] vs. time (t) for the NaOH-treated biomass.

35 30

1,2

25 20

q pH

15 10

1 C/Co

q(mg Fe / g biomass)-pH

-3

0,8 0,6

5

0,4

0

0,2 0

100

200

300

Time (min)

0 0

100

200

300

Time (min) Fig. 3. Time evolution of the biosorption capacity and pH during the experiments. ω = 250 rpm, C0 (Fe3+ ) = 100 mg/l, Cbiomass = 3 g/l, PS = 50–160 ␮m.

Fig. 5. (C/C0 ) vs. time (t) according to the external mass transfer resistance model.

1648

A. Selatnia et al. / Process Biochemistry 39 (2004) 1643–1651

Log10[1-(q/qe)^2]

-0,5 0

50

100

150

200

35

qe( mg Fe /g biomass)

Time (min)

0

250

-1 -1,5 -2 -2,5

30 25 20 15 10 5 0 0

-3 Fig. 6. log10 [1−(q/qe )2 ] vs. time (t) according to the Urano and Tachikawa model.

30 25 20 15 10 5 0 0

5

10

15

800

Fig. 9. Effect of stirring speed on the biosorption capacity of the NaOH-treated biomass. C0 (Fe3+ ) = 100 mg/l, Cbiomass = 3 g/l, PS = 50–160 ␮m.

% wt of Fe ions removal

q( mg Fe / g biomass)

35

200 400 600 Stirring speed (Rpm)

120 100 80 60 40 20 0 0

20

2

4

6

8

10

12

biomass concentration (g/L)

t^0,5 ( min^ 0,5)

For low pH values, one may consider an adsorption competition between H+ protons and metallic ions on the active sites on the cell wall of the biomass, as it has been suggested by many others authors [6,13,43–47]. 4.7. Effect of stirring speed on Fe3+ biosorption

qe (mg Fe /g biomass)

The effect of stirring speed on biomass adsorption capacity was studied. Optimal value of adsorption capacity was obtained for a stirring speed ω = 250 rpm (Fig. 9). This stirring speed is used in all our experiments. A moderate speed gives the best homogeneity for the mixture suspension. At high stirring speed, vortex phenomena occurs and

35 30 25 20 15 10 5 0 0

5

10

15

initial pH

Fig. 8. Effect of initial pH on the Fe3+ adsorption capacity by the NaOH-treated biomass. ω = 250 rpm, C0 (Fe3+ ) = 100 mg/l, Cbiomass = 3 g/l, PS = 50–160 ␮m.

Fig. 10. Effect of the biomass concentration on the biosorption capacity of the NaOH-treated biomass. ω = 250 rpm, C0 (Fe3+ ) = 100 mg/l, PS = 50–160 ␮m.

the suspension is no longer homogenous which makes the adsorption of Fe3+ difficult. 4.8. Influence of biomass concentration Fig. 10 shows the metallic ions elimination in wt.% as a function of biomass concentration. This graph shows the growth of the elimination of the metal ions with an increase in the concentration of the biomass. A stage appears starting from a concentration of biomass equal to 3 g/l. This result makes it possible to conclude that starting from an optimal concentration in biomass of 3 g/l, the rate of elimination of qe( mg Fe / g biomass)

Fig. 7. q vs. t0.5 according to the Weber and Morris model.

140 120 100 80 60 40 20 0 0

200

400

600

800

Initial concentration of Fe (mg/L) Fig. 11. Influence of initial concentration of Fe3+ on the biosorption capacity of the NaOH-treated biomass. ω = 250 rpm, Cbiomass = 3 g/l, PS = 50–160 ␮m.

qe(mg Fe/g biomass)

A. Selatnia et al. / Process Biochemistry 39 (2004) 1643–1651

1649

Table 4 Sorption isotherm coefficients of Langmuir and Freundlich models

140 120 100 80 60 40 20 0

Langmuir qm (mg g−1 ) b (l mg−1 ) R2

0

100

200

300

400

Ce (mg/L)

Freundlich k (l g−1 ) 1/n R2

125 9.524 × 10−2 0.9904 27.76 0.2694 0.8346

Fig. 12. Adsorption data of Fe3+ onto the NaOH-treated biomass. ω = 250 rpm, Cbiomass = 3 g/l, PS: 50–160 ␮m.

the ions Fe3+ does not increase and it could be attributed to the biomass granulates which are agglomerated.

maximum quantity of adsorbed Fe3+ on biomass is around 122 mg Fe /gbiomass . 4.10. Adsorption isotherm analysis

4.9. Influence of the initial Fe3+ concentration Many studies have shown that for low Fe3+ concentrations, the quantity of adsorbed Fe3+ per unit mass of biosorbant is directly proportional to the ionic concentration in solution. Fig. 11 shows that the quantity of adsorbed Fe3+ per unit mass of biosorbant increases when the initial Fe3+ ions concentration increases. This study makes it possible to reach the level of saturation of the active sites of the biomass and to obtain the maximum for the ions Fe3+ capacity of adsorption of the biomass. In our study, initial Fe3+ concentration varies from 20 to 700 mg/l. As can be seen from Fig. 11, that the

In order to optimize the biosorption process parameters, we have modelised the equilibrium curve (Fig. 12). Two equations of isotherms of Langmuir and Freundlich were tested. The isotherm curve shows a limiting biosorption capacity attained at Fe3+ equilibrium concentration of about 334 mg/l. Figs. 13 and 14 are the transformed forms of these models which permit to calculate Langmuir’s constants (qm and b) and Freundlich constants (k and n). Table 4 shows the values of the computed constants. Values of coefficients of correlation R2 shows that the Langmuir’s models, fits fairly well our experimental data.

3

5. Conclusions

Ce/qe (g/L)

2,5 2 1,5 1 0,5 0 0

100

200 Ce (mg/L)

300

400

Fig. 13. Application of the Langmuir equation to the adsorption data of Fe3+ onto the NaOH-treated biomass.

6 5

S. rimosus dead biomass treated with NaOH (0.1 M) is an efficient adsorbant of Fe3+ in dilute solutions. Up to 122 mg of iron can be fixed by each gramme of NaOH-treated biomass. The cell walls of this biomass contains anionic groups such as (–COO− , –C–O, –NH, –C=O, –OH), whose adsorbant ability towards Fe3+ ions is fairly high. Adsorption is moreover influenced by various parameters such as initial pH, initial Fe3+ concentration, biomass concentration and stirring speed. The results obtained during this study shows that this method of eliminating Fe3+ ions is very promising and confirm the technical and economic interest compared to the conventional processes such that the ion exchange on resins.

Ln qe

4 3

Appendix A. Nomenclature

2 1 0 -4

-2

0

2 Ln Ce

4

6

8

Fig. 14. Application of the Freundlich equation to the adsorption data of Fe3+ onto the NaOH-treated biomass.

List of symbols b Langmuir adsorption constant (l mg−1 ) C metal ion concentration in the solution (mg l−1 ) Cbiomass biomass concentration (g l−1 )

1650

Ce C0 (Fe3+ ) Cs dp Di k, n k1 k2 Ki m PS q qe qm

S t T

A. Selatnia et al. / Process Biochemistry 39 (2004) 1643–1651

residual metal ion concentration at equilibrium (mg l−1 ) Fe3+ initial concentration (mg l−1 ) metal ion concentration at the particle surface (mg l−1 ) particle size diameter (m) diffusion coefficient in the solid (m2 s−1 ) Freundlich’s adsorption constants rate constant of first-order biosorption (s−1 ) rate constant of second-order biosorption (g mg−1 s−1 ) diffusion coefficient in the solid (mg g−1 s−1/2 ) sorbent mass concentration in the solution (kg m−3 ) biomass particle size (m) amount of adsorbed metal ion on the biomass at time t (mg g−1 ) adsorbed metal ion quantity per gram of biomass at equilibrium (mg g−1 ) maximum amount of metal ion per unit weight of biomass to form a complete monolayer on the surface bound at high Ce (mg g−1 ) specific surface area (m−1 ) time for biosorption (s) temperature (◦ C)

Greek symbols β mass transfer coefficient (m s−1 ) ρapp apparent volume mass of the sorbent (kg m−3 ) ω stirring speed (rpm). References [1] Roux JC, Lhomme B, Nexton J, Lenon G, Robillon C. Biosorption of heavy metal from polluted waters by micellial dead biomasses of filamentous fungus Rhizopus arrhizus. In: Proceedings of the European Congress of Biotechnology. Copenhagen; 1990. p. 325–8. [2] Milande N, Roux JC, Fourest E. Improvement of heavy metal from industrial wastes by filamentous fungi. Récents Progrés en Génie des Procédés 1993;25:19–24. [3] Fourest E, Canal C, Roux JC. Improvement of heavy metals biosorption by micelial dead biomasses Rhizopus arrhizzopus, Mucor miehei, Penicillium chrysogenum: pH control and cationic activation. FEMS Microbiol Rev 1994;14(4):325–32. [4] Fourest E, Serre A, Roux JC. Contribution of carboxyl groups to heavy metal binding sites in fungal wall. Toxicol Environ Chem 1996;54(1–4):1–10. [5] Modak JM, Natarajan KA, Saha B. Biosorption of copper and zinc using waste Aspergillus niger biomass. Miner Metall Process 1996;13:52–7. [6] Guibal E, Roulph C, Lecloirec P. Uranium biosorption by filamentous fungus Mucor miehei, pH effect on mechanisms and performances of uptake. Water Res 1992;26(8):1139–45.

[7] Gardea-Torresdey JL, Cano-Aguilera I, Webb R, Tiemann K, Gutierrez-Corona F. Copper adsorption by inactivated cells of Mucor rouxii: effect of esterification of carboxyl groups. J Hazard Mater 1996a;48:171–80. [8] Kuyucak N, Volesky B. Accumulation of cobalt by marine alga. Biotechnol Bioeng 1989;33:809–22. [9] Hao Y, Zhao Y, Ramelow GJ. Uptake of metal ions by non living biomass derived from marine organisms: effect of pH and chemical treatments. J Environ Sci Health 1994;29:2235–54. [10] Leusch A, Volesky B. Influence of film diffusion on cadmium biosorption by marine biomass. J Biotechnol 1995;43:1–10. [11] Khoshmanesh A, Lawson F, Prince IG. Cadmium uptake by unicellular green microalgae. Chem Eng J 1996;62:81–8. [12] Chong KH, Volesky B. Metal biosorption equilibria in ternary system. Biotechnol Bioeng 1996;49:629–38. [13] Sautel G, Roulph C, Leclerc P. Cadmium biofixation by Pseudomonas putida bacteria. Récents Progrés en Génie des Procédés 1991;5:203– 8. [14] Urrutia M, Beveridge TJ. Remobilization of heavy metals retained as oxyhydroxides or silicates by Bacillus subtilis cells. Appl Environ Microbiol 1993;59(12):4323–9. [15] Mago R, Srivostava S. Uptake by zinc in Pseudomonas sp. Strain UD626. Appl Environ Microbiol 1994;60:2367–70. [16] Butter TJ, Evision LM, Hancock TYC, Holland, FC. Removal and recovery of cadmium from dilute aqueous streams by biosorption, elution and electrolysis. In: Proceedings of the 9th Forum for Applications of Biotechnology. Gent; 1995. p. 2581–3. [17] Engl A, Tholmas C, Kune B, Biosorption of heavy metals by paracoccus denitrificans. In: Proceedings of the 9th Forum for Applications of Biotechnology, Gent; 1995. p. 2589–92. [18] Sag Y, Kutsal T. Biosorption of heavy metals by Zoogloea ramigera: use of adsorption isotherms and a comparison of biosorption characteristics. Chem Eng J Biochem Eng J 1995;60:181–8. [19] Mishra S, Chaudhury P, Roy G. Kinetics of Zn2+ adsorption by Pénicillium sp. Hydrometallurgy 1996;40:11–23. [20] Singleton I, Simmons P. Factors affecting silver biosorption by an industrial strain of Saccharomyces cerevisiae. J Chem Technol Biotechnol 1996;65:21–8. [21] Thomas R, Macaskie LE. Biodegradation of tributyl phosphate by naturally occurring microbial isolates and coupling to removal of uranium from aqueous solution. Environ Sci Technol 1996;30:2371– 5. [22] Jaanson-Cherrier M, Guibal E, Roussy J, Belanghe B, Lecloirec P. Vanadium(IV) sorption by chitosan: kinetics and equilibrium. Water Res 1996;30(3):465–75. [23] Gardea-Torresdey JL, Tang L, Salvador JM. Copper adsorption by esterified and unesterified fractions of sphagnum peat moss and its different humic acid. J Hazard Mater 1996;48:191–206. [24] Remacle J, Muguruza I, Fransolet M. Cadmium removal by strain of denitrificans isolated from metal-polluted pond. Water Res 1992;7:923–9. [25] Solari P, Zouboulis AI, Matis KA, Stalidis GA. Removal of toxic metals by biosorption onto nonliving sewage sludge. Sep Sci 1996;31:1075–92. [26] Fritz W, Merk W, Schlünder EU. Competitive adsorption of two dissolved organics onto activated carbon-II. Adsorption kinetics in batch reactors. Chem Eng Sci 1981;36:731–41. [27] Hand DW, Crittenden JC, Thacker WE. User-Oriented batch reactor solutions to the homogenous surface diffusion model. J Environ Eng Div 1983;109:82–101. [28] Mc Kay G. The adsorption of basic dye onto silica from aqueous solution–solid diffusion model. Chem Eng Sci 1984;39:129–38. [29] Van Vliet BM, Weber Jr WJ, Hozumi H. Modelling and prediction of specific compounds adsorption by activated carbon and synthetic adsorbents. Water Res 1980;14:1719–28. [30] Mattews AP, Weber Jr WJ. Modelling and parameter evaluation for adsorption in slurry reactors. Chem Eng Commun 1984;25:157–71.

A. Selatnia et al. / Process Biochemistry 39 (2004) 1643–1651 [31] Mac Kay G, Bino MJ. Adsorption of polluants from wastewater onto activated carbon based on external mass transfer and pore diffusion. Water Res 1988;22:279–86. [32] Glover MRL, Young BD, Bryson AW. Modelling the binary adsorption of gold and zinc cyanides onto a strong-base anion exchange resin. Int J Miner Process 1990;30:217–28. [33] Mac Kay G, Allen SJ. In: Fuchman CH, editor. Proceedings of the Seminar on Peat as a Raw Material. Bord NaMona Peat Research Center, Newbridge, Co Kildare; 1990. p. 124–46. [34] Weber WJ, Morris JC. Advances in water pollution research: removal of biologically-resistant polluants from waste waters by adsorption. In: Proceedings of the International Conference on Water Pollution Symposium, vol. 2. Oxford: Pergamon Press; 1962; p. 231–66. [35] Mac Kay G, Poots VJP. Kinetics and diffusion processes in colour removal from effluent using wood as an adsorbant. J Chem Technol Biotechnol 1980;30:279–92. [36] Mc Kay G, Blair, HS, Findon, A. Sorption of metal ions by chitosan. In: Immobilisation of ions by Bio-sorption, Eccles H, Hunt S, editors. Ellis Horwood, Chichester, UK; 1986, p. 59–69. [37] Urano K, Tachikawa H. Process development for removal and recovery of phosphorus from wastewater by new adsorbent-2. Adsorption rates and breathrough curves. Ind Eng Chem Res 1991;30:1897–9. [38] Lagergreen S. Handlingar 1898;24:1.

1651

[39] Ho YS, Mac Kay G. Water Res 1999;33:578. [40] Aksu Z. Equilibrium and kinetic modelling of cadmium(II) biosorption by C. vulgaris in batch system: effect of temperature. Sep Purif Technol 2001;21:285–94. [41] Ho YS, Mac Kay G. Pseudo-second order model for sorption process. Process Biochem 1983;34:451. [42] Yakup Arica M, Kaçar Y, Genç Ö. Entrapment of white-rot fungus Trametes versicolor in Ca-alginate beads: preparation and biosorption kinetic analysis for cadmium removal from an aqueous solution. Bioresource Technol 2001;80:121–9. [43] Fourest E, Volesky B. Alginate properties and heavy metal biosorption by marine algae. Biochem Biotechnol 1997;67:215–26. [44] Fourest E, Roux JC. Heavy metal biosorption by fungal mycelial by-products. Mechanisms and influence of pH. Appl Microb Biotechnol 1992;37:399. [45] Benedetti MF, Milne CJ, Kinninburg DG, Van Riemsddijik K, Koopal LK. Metal binding to humic substances. Application of the non-ideal competitive adsorption model. Environ Sci Technol 1995;29:446–50. [46] Fourest E, Volesky B. Contribution of sulphonate groups and alginate to heavy metal biosorption by dry biomass of Sargassum fluitants. Environ Sci Technol 1996;30:277–302. [47] Huang CP, Westman D, Huang C, Morehart AL. Water Sci Technol 1988;20:369.