Determination of iron and nickel by flame atomic absorption spectrophotometry after preconcentration on Saccharomyces cerevisiae immobilized sepiolite

Determination of iron and nickel by flame atomic absorption spectrophotometry after preconcentration on Saccharomyces cerevisiae immobilized sepiolite

Talanta 47 (1998) 689 – 696 Determination of iron and nickel by flame atomic absorption spectrophotometry after preconcentration on Saccharomyces cer...

105KB Sizes 0 Downloads 7 Views

Talanta 47 (1998) 689 – 696

Determination of iron and nickel by flame atomic absorption spectrophotometry after preconcentration on Saccharomyces cere6isiae immobilized sepiolite Hu¨seyin Bag˘ a, Mustafa Lale a, A. Rehber Tu¨rker b,* a

Kırıkkale U8 ni6ersitesi Fen Edebiyat Faku¨ltesi, 71450 Kırıkkale, Turkey b Gazi U8 ni6ersitesi Fen Edebiyat Faku¨ltesi, 06500 Ankara, Turkey

Received 2 December 1997; received in revised form 19 February 1998; accepted 27 February 1998

Abstract Iron and nickel have been preconcentrated on Saccharomyces cere6isiae immobilized sepiolite and determined by flame atomic absorption spectrophotometry (FAAS). Preconcentration studies were conducted by the column method. Effect of pH, amount of adsorbent, elution solution, flow rate and interfering ions on the recovery of the analytes have been investigated. Recoveries of Fe and Ni were 95 9 1 and 99.5 90.1%, respectively, at 95% confidence level. The breakthrough capacities of analytes were also investigated and found to be 0.042 mmol g − 1 for Fe and 0.055 mmol g − 1 for Ni. The proposed method was applied to the determination of iron and nickel in brass (NBS SRM 37e). The detection limit of iron and nickel were found as 0.065 and 0.087 mg ml − 1, respectively. The direct determination of trace metals by flame atomic absorption spectrometry (FAAS) is limited and difficult because of low concentration and/or matrix interferences. The proposed method is excellent for the determination of trace metal in matrixes, such as metal alloys. © 1998 Elsevier Science B.V. All rights reserved. Keywords: Trace metal determination; Preconcentration; Saccharomyces cere6isiae; Immobilization; Sepiolite; Flame atomic absorption spectrometry

1. Introduction Many methods, such as extraction, coprecipitation, electrodeposition and ion exchange, have been used for preconcentration of trace metals. In recent years preconcentration by microorganisms has been widely used [1 – 6]. The use of microorganisms as biosorbent for metals has become a * Corresponding author. Tel: + 90 312 2122900; fax: +90 312 2122279; e-mail: [email protected]

good alternative to the other preconcentration methods as regards higher recovery, economical advantages, simplicity and environmental protection. In general, microorganisms have the ability to selectively adsorb a specific element without preconcentrating the matrix [7–9]. Either living or nonliving microorganisms, such as yeast, fungi, bacteria and algae are capable of accumulating heavy metals and radionuclides from aqueous solution by a different chemical and biological mechanisms [10]. The products pro-

0039-9140/98/$ - see front matter © 1998 Elsevier Science B.V. All rights reserved. PII S0039-9140(98)00104-0

690

H. Bag˘ et al. / Talanta 47 (1998) 689–696

duced by or derived from microbial cell such as metabolites, polysaccharides, and cell wall constituents are of effective in metal accumulating. Yeasts are considered to be more effective in heavy metal accumulation because of their great tolerance towards metals and their high binding capacity to the cell [11]. Maquieira et al. [6] have explained the specificity of microorganisms. According to their explanation, the binding sites of microorganisms have a diversity property. For that reason, if one element has an affinity to coordinate with one functional group, i.e. COO − , and, if this element is present in a mixture of many other elements, each of them can form a stable complex with any of the other functional groups present in the cell wall of the microorganism. Therefore, by changing pH or elution conditions, selectivity can be obtained and this is well demonstrated in the interference effects where the elements in relatively high concentrations. There are many binding sites on the cell wall of microorganisms. The physical confinement or localization of microbial cells to a certain region of a support material is defined as ‘immobilization’. The use of immobilized cell systems has many advantages over the use of freely suspended cells. These include better capability of re-using the biomass, easy separation of cells from the reaction mixture, high biomass loadings and minimal clogging in continuous flow systems. In addition particle size can be controlled and high flow rates achieved with or without recirculation. Immobilized cell systems can be used in both batch and column experiments. Nakajima et al. [8] have used Saccharomyces cere6isiae and some other microorganisms in a comparative study for metal uptake by immobilizing on calcium alginate gel. Maquieira et al. [12] have used S. cere6isiae by immobilizing covalently on controlled pore glass (CPG) for trace metal preconcentration. Gencer et al. [13] have immobilized S. cere6isiae on wood chips by adsorption using a medium of glucose nutrients for ethanol fermentation in a tubular fermenter. They had described a mathematical model for the kinetics of ethanol fermentation. Daugulis et al. [14] have immobilized S. cere6isiae on ion exchange resins

by adsorption in glucose nutrients. Black et al. [15] have immobilized S. cere6isiae and S. u6arum on stainless steel and polyester foam for the production of ethanol. In this study, S. cere6isiae has been immobilized on ‘sepiolite’ and used for the preconcentration of iron and nickel. Sepiolite is a hydrous magnesium mineral and has already been used for the adsorption of some gases [16], liquids [17] and as an adsorbent for preconcentration of trace elements alone [18], but not as an immobilization substrate.

2. Experimental

2.1. Apparatus A Philips PU 9285 model flame atomic absorption spectrometer equipped with deuterium lamp background correction, hallow cathode lamps (HCL) and air acetylene burner was used for the determination of the metals. All absorption measurements were performed under the following conditions: wavelengths, 248.3 nm for Fe and 232 nm for Ni; fuel flow rate, 1.1 l min − 1 for Fe and 0.9 l min − 1 for Ni; HCL lamp current, 11.2 mA; bandpass, 0.5 nm; and integration time, 4 s for both elements. All pH measurements were performed with a JENWAY 3010 model digital pH meter.

2.2. Reagents Doubly distilled deionized water and analytical grade reagent chemicals were used unless otherwise indicated. All metal stock solutions (1000 mg ml − 1) were prepared by dissolving the appropriate amounts of metals or compounds. The working solutions were prepared by dilution from the stock solutions.

2.3. Materials The sepiolite used as a substrate for the immobilization of S. cere6isiae in this study was collected from the trances dug in the Tu¨rktaciri sepiolite deposit and ground and sieved to 35–60 mesh. Full details for the characterization of sepi-

H. Bag˘ et al. / Talanta 47 (1998) 689–696

olite have been given in the study [18] carried out before.

2.4. Procedures A laboratory strain of S. cere6isiae was maintained on a medium comprising (g l − 1) malt extract (Difco) 3, yeast extract (Difco) 3, D-glucose (Carlo) 10, peptone (Difco) 5, and agar (Difco) 15. The yeast cultivated on the solid medium was stored in a refrigerator at 4°C before use, in order to extend their freshness and prevent contamination by the growth of other microorganisms. Liquid medium was prepared by mixing the substances mentioned above except agar. All the steps of every procedure were sterilized by autoclaving at 120°C for about 30 min. Firstly, starter culture was performed from the solid medium by loop-inoculating to 100 ml of liquid medium. Then, it is incubated for 24 h at 30°C on an orbital shaker (200 rpm). For experimental culture, 100 ml of liquid medium was prepared and inoculated with 5 ml of the starter culture and incubated on the orbital shaker for 24 h at 30°C. Then, the yeast grown in the experimental culture was separated from the growth media using centrifugation (5000 g for 5 min) to isolate the biomass. The biomass was treated according to the procedure described by Mahan et al. [19]. 10 ml of 0.1 mol l − 1 HCl was added to the isolated biomass. After 10 min, the mixture was centrifuged and the acid solution was discarded. This procedure was repeated three times and then followed by rinsing the acid-washed biomass in distilled water. These rinsed yeast were again centrifuged and the resulting biomass lyophilized to yield a dry yeast powder. The immobilization of S. cere6isiae was performed according to the procedure recommended by Mahan et al. [19]. 150 mg of dry yeast powder (S. cere6isiae) was mixed with 2 g of sepiolite. The mixture was wetted with 2 ml of doubly distilled deionized water and thoroughly mixed. After mixing, the paste was heated in an oven at 80°C for 24 h to dry the mixture. The wetting and drying step was repeated to maximize the contact between S. cere-

691

6isiae and sepiolite, thereby improving the immobilization efficiency. Then, the sepioliteyeast briquette was broken to get original size (35–60 mesh). S. cere6isiae immobilized sepiolite (0.2 g) was packed in a glass column (10 mm i.d and 200 mm length). Before use, 1 mol l − 1 HCl solution and doubly distilled deionized water were passed through the column in order to condition and clean it. Then, the column was conditioned to the studied pH. An aliquot of a solution (100 ml) of one or several elements containing 25 mg Ni, 50 mg Fe was taken and the pH was adjusted to the desired value with hydrochloric acid and ammonia. The resulting solution was passed through the column. By using a peristaltic pump, the flow rate was adjusted to the desired value. The retained metal ions were eluted from S. cere6isiae immobilized sepiolite with 10 ml of 1 mol l − 1 hydrochloric acid solution. The eluate was aspirated into an air-acetylene flame for trace determination by AAS. The S. cere6isiae immobilized sepiolite was used repeatedly after washing with 1 mol l − 1 HCl solution and distilled water, respectively. The recoveries of the elements were calculated from the ratio of the concentration found by FAAS to that calculated theoretically. All experiments done for the determination of the optimum conditions (pH, bed height, etc) were performed according to the general procedure described above.

2.5. Dissolution of standard reference material As an application sample, standard reference brass material (NBS SRM 37e) was used. 0.5 g of the standard reference material were weighed and dissolved in 10 ml of doubly distilled deionized water and 10 ml of concentrated nitric acid. The solution was heated until the volume was one fourth of the beginning solution. Then, it was made up to 100 ml with doubly distilled deionized water and adjusted to pH 6 with hydrochloric acid and ammonia. Iron and nickel were determined after applying the general preconcentration procedure described above.

692

H. Bag˘ et al. / Talanta 47 (1998) 689–696

3. Results and discussion

3.1. Effect of pH The retention of metal ions on the column containing S. cere6isiae immobilized sepiolite was studied as a function of pH. For that purpose, the pH values of element solutions were adjusted to a range of 2– 10 with HCl or NH3. As shown in Fig. 1, the optimum pH of the sample solution is 6 for nickel and 8 for iron. From the same figure, uptake of the metal ions was quantitative in part of the pH range 6 – 8. For the binding of metal ions, at least, two types of adsorption sites or two functional groups could be present in the yeast surface [21]. This causes the difference in metal binding to the surface with the pH of the medium. The decrease in binding at lower pH values was attributed to the protonation of weakly basic coordination groups on the yeast surface [22]. In a study [12], S. cere6isiae was immobilized on controlled pore glass (CPG) and the maximum uptake of iron was obtained at pH 6. The difference between the pH values may be due to the change of metal uptake mechanism according to the substrate used for the immobilization and killing process of yeast. In our previous study [18] in which sepiolite was used alone as an adsorbent the maximum retention of iron was at pH 2 with

a recovery of 80%. But in the present study, the retention of iron was found extremely high and the maximum recovery was obtained at pH 8. This shows that the metal ions were adsorbed mainly by immobilized S. cere6isiae instead of sepiolite.

3.2. Effect of the amount of adsorbent (bed height) The retention of the elements studied was examined in relation to the amount of S. cere6isiae immobilized sepiolite, which was varied from 0.05 to 0.4 g. It was found that above 0.15 g of adsorbent the recovery of Ni was gradually increased, but about 0.2 g of adsorbent that of Fe reached plateau. Therefore, 0.2 g of adsorbent was found to be optimum of all preconcentration purposes.

3.3. Effect of the type and 6olume of elution solutions The elution studies were performed with 0.5 and 1 mol l − 1 hydrochloric and nitric acid solutions. The eluate volume was 5, 10 and 15 ml. As can be seen in Table 1, for all elements studied 10 ml of 1 mol l − 1 hydrochloric acid solution was found to be satisfactory.

3.4. Effect of flow rates of sample solutions The retention of elements on an adsorbent depends upon the flow rate of the metal solution. Therefore, the effect of the flow rate of sample solution was examined under optimum conditions (pH, eluent type, etc.) by using a peristaltic pump. The solution was passed through the column with the flow rates adjusted in a range of 1–7.5 ml min − 1. As shown in Fig. 2, the optimum flow rates was found as 2.5 ml min − 1 for both metal ions. The flow rate of elution solution used was 1 ml min − 1.

3.5. Effect of the 6olume of sample solution Fig. 1. The effect of pH on recovery of iron and nickel by S. cere6isiae immobilized sepiolite; A, iron; and B, nickel.

The effect of changes in the volume of sample solution passed through the column on the reten-

H. Bag˘ et al. / Talanta 47 (1998) 689–696

693

Table 1 Effect of the type and volume of elution solutions on recovery of iron and nickel by S. cere6isiae immobilized sepiolite Type of elution solution

HCl

HCl

HNO3

HNO3

Volume (ml)

5 10 15 5 10 15 5 10 15 5 10 15

tion of Fe and Ni was investigated. 100, 250, 500, 750 and 1000 ml of sample solutions containing 0.25, 0.10, 0.05, 0.033 and 0.025 mg ml − 1 of Ni, 0.5, 0.2, 0.1, 0.066 and 0.05 mg ml − 1 of Fe respectively, were passed through the column. It was found that iron and nickel up to 500 ml of sample solution could be recovered quantitatively. At higher sample volumes, the recoveries decreased gradually with increasing volume of sample. In this study, the elution volume was 10 ml and the preconcentration factors were 50 for both iron and nickel.

Concentration (mol l−1)

0.5

1.0

0.5

1.0

Recovery (%) Fe

Ni

81 89 90 87 95 95 79 87 88 87 90 91

85 89 89 90 99 99 83 89 92 89 91 94

corresponding to three times the standard deviation of the blank signal) were 0.065 mg ml − 1 for iron and 0.087 mg ml − 1 for nickel.

3.8. Capacity studies The capacity study used was adapted from that recommended by Maquieira et al. [20]. 50 ml aliquots of a series of concentrations (5–80 mg ml − 1) was adjusted to the appropriate pH, then, preconcentrated and eluted. The amount of metal

3.6. Precision of the method For the precision of the method, the optimum conditions mentioned above were used. For this purpose, successive retention and elution cycles (with 0.25 mg ml − 1 of Ni and 0.50 mg ml − 1 of Fe) were performed and nickel and iron were determined in the solution by FAAS. As can be seen in Table 2, the recoveries of nickel and iron are quantitative and the precision of the method is very good.

3.7. Calibration graph and detection limit The calibration graphs were linear up to 5 mg ml − 1 for nickel and up to 8 mg ml − 1 for iron. The detection limit (evaluated as the concentration

Fig. 2. The effect of flow rate of sample solutions on recovery of iron and nickel by S. cere6isiae immobilized sepiolite; A, iron; and B, nickel.

H. Bag˘ et al. / Talanta 47 (1998) 689–696

694 Table 2 Precision of the method

Table 3 Capacity for the metal ions and recovery

Element

Recoverya % R9 ts/ N

N

Metal ion

Capacity (mmol g−1)

Recovery (%)

Fe Ni

959 1 99.590.1

7 7

Fe3+ Ni2+

0.042 0.055

95 99

a

Average of N determinations with 95% confidence level.

adsorbed at each concentration level was determined from the following equation: C= c6/w where C is the capacity (in mg g − 1), c is the concentration (in mg ml − 1) of metal eluted, 6 is the volume (in ml) of solution used and w is the mass (in g) of the immobilized material. Evaluation of breakthrough capacity was made from a breakthrough curve by plotting the total metal concentration (mg ml − 1) versus the milimoles of metal adsorbed per gram. The breakthrough capacity was evaluated from a breakthrough curve plot, which is shown in Fig. 3. As shown in Table 3, it was found that the breakthrough capacity of nickel was higher than that of iron. However, the breakthrough capacity of iron found in this study is higher than that obtained in a study [12] in which S. cere6isiae was immobilized on CPG and the breakthrough ca-

Fig. 3. The breakthrough curve of iron and nickel on S. cere6isiae immobilized sepiolite; A, iron; and B, nickel.

pacity of iron was found 0.020 mmol g − 1. As it is explained before, this difference in the breakthrough capacity of iron may be due to the change in the structure of the yeast during the killing, immobilization process and the type of immobilization substrate (sepiolite).

Table 4 The effect of interfering ions on recovery Interfering ions

Concentration (mg ml−1)

Recovery (%)

Fea

Nib

Cu2+

— 0.5 1.0 2.5 5.0 10.0

95 95 92 92 90 90

99 98 98 97 94 94

Zn2+

— 0.5 1.0 2.5 5.0 10.0

95 96 96 94 93 92

99 97 97 95 94 94

Fe3+

— 0.5 1.0 2.5 5.0 10.0

Ni2+

— 0.5 1.0 2.5 5.0 10.0

a b

0.5 mg ml−1. 0.25 mg ml−1.

99 100 100 102 103 104 95 89 91 97 99 102

H. Bag˘ et al. / Talanta 47 (1998) 689–696

695

3.9. Effect of interfering ions In order to investigate the effect of the interference of elements to each other on biosorption, the recoveries of elements was examined when they existed together in a same medium. The concentrations of interfering metal ions were adjusted in a range of 0.5 – 10 mg ml − 1. The results were given in Table 4. As can be seen, the effect of divers ions can be negligible. This shows that nickel and iron can be determined quantitatively in metal alloys although the matrix concentration is high.

3.10. The effect of column reuse The stability and potential recyclability of the column were assessed by monitoring the change in the recoveries of iron and nickel ions through several adsorption-elution cycles. Each five runs were performed in the same day and the next five runs were made one day later. The columns were stored in doubly distilled deionized water. As shown in Fig. 4, a small decrease in the recoveries ( 5 and 8% for iron and nickel, respectively) occurred. The columns seems to be relatively stable up to 20 runs.

Fig. 4. The effect of column reuse on recovery of iron and nickel by S. cere6isiae immobilized sepiolite; A, iron; and B, nickel.

iron were determined with high accuracy and precision although the matrix concentration (e.g. copper and zinc) was very high in standard reference material.

3.11. Application

4. Conclusion

Proposed preconcentration method was applied to the determination of Ni and Fe in standard reference material (NBS SRM 37e). Reference sample material was preconcentrated and analyzed as explained in Section 2. As shown in Table 5, it is clear that the calculated values for both metal agreed very well with the certified values. By using yeast immobilized sepiolite, nickel and

The method proposed by the use of S. cere6isiae immobilized sepiolite for the preconcentration of iron and nickel is simple, sensitive and accurate. Iron and nickel were quantitatively recovered from the column with a high precision. In conclusion, the proposed method is excellent as regards simplicity, sensitivity, selectivity, precision, accuracy and column stability.

Table 5 Analysis of standard reference material Element

Fe Ni a b

Concentration (% m/m) Calculated valuea

Certified valueb

(3.7 90.5)×10−3 0.50 9 0.01

0.004 0.53

Relative error (%)

LOD (mg ml−1)

−7.5 −5.7

0.065 0.087

Mean and S.D. of eight determinations. The composition of the brass (NBS SRM 37e) was Cu 69.61, Zn 27.85, Pb 1.00 and Sn 1.00% (m m−1).

696

H. Bag˘ et al. / Talanta 47 (1998) 689–696

Acknowledgements The support Gazi University Research Fund are gratefully acknowledged.

References [1] V. Majidi, J.A. Holcombe, J. Anal. At. Spectrom. 4 (1989) 439. [2] H.A.M. Elmahadi, G.A. Greenway, J. Anal. At. Spectrom. 8 (1993) 1011. [3] H.A.M. Elmahadi, G.A. Greenway, J. Anal. At. Spectrom. 6 (1991) 643. [4] C.A. Mahan, V. Majidi, J.A. Holcombe, Anal. Chem. 61 (1989) 624. [5] L.C. Robles, C.G. Ollalla, A.J. Aller, J. Anal. At. Spectrom. 8 (1993) 1015. [6] A. Maquiera, H. Elmahadi, R. Puchades, J. Anal. At. Spectrom. 11 (1996) 99. [7] T. Sakaguchi, T. Tsuji, A. Nakajima, T. Horikoshi, Appl. Microbiol. Biotechnol. 8 (1979) 207. [8] A. Nakajima, T. Sakaguchi, Appl. Microbiol. Biotechnol. 24 (1986) 59. [9] V. Majidi, J.A. Holcombe, Spectrochim. Acta 43B (12)

.

(1988) 1423. [10] G.M. Gadd, C. White, L. de Rome, Biohydrometallurgy Sci. Technol. Lett., Kew Survey, UK (1988) 421. [11] G.M. Gadd, C. White, J. Gen. Microbiol. 131 (1985) 1875. [12] A. Maquieira, H.A.M. Elmahadi, R. Puchades, Anal. Chem. 66 (1994) 1462. [13] M.A. Gencer, R. Mutharasan, Biotechnol. Bioeng. 25 (1983) 2243. [14] A. J. Daugulis, N.M. Brown, W.R. Cluett, D.B. Dunlop, Biotechnol. Lett. 3 (1981) 651. [15] G.M. Black, C. Webb, T.M. Mathews, B. Atkinson, Biotechnol. Bioeng. 26 (1986) 134. [16] Y. Sarıkaya, S. Aybar, Comm. Fac. Sci. University d’Ankara 24 (1978) 33. [17] B. Erdog˘an, H.I: . U8 nal, I: . Kayabalı, H. O8 cal, Seventh National Clay Symposium, Ankara, 1985. [18] A.R. Tu¨rker, H. Bag˘, B. Erdog˘an, Fresenius J. Anal. Chem. 357 (1997) 351. [19] C.A. Mahan, J.A. Holcombe, Anal. Chem. 64 (1992) 1933. [20] A. Maquieira, H.A.M. Elmahadi, R. Puchades, Anal. Chem. 66 (1994) 3632. [21] L.C. Robles, C.G. Olalla, A.J. Aller, J. Anal. At. Spectrom. 8 (1993) 1015. [22] B. Greene, M.T. Henz, M. Hosea, D.W. Darnall, Biotechnol. Bioeng. 28 (1986) 764.