- Email: [email protected]
Lead removal in fixed-bed columns by zeolite and sepiolite
Chemosphere 60 (2005) 1487–1492 www.elsevier.com/locate/chemosphere
Short Communication
Lead removal in fixed-bed columns by zeolite and sepiolite Mustafa Turan a
a,*
, Ugur Mart b, Baris Yu¨ksel a, Mehmet S. C ¸ elik
b
Istanbul Technical University, Department of Environmental Engineering, Maslak 34469, Istanbul, Turkey b Istanbul Technical University, Faculty of Mining Engineering, Maslak 34469, Istanbul, Turkey Received 26 February 2004; received in revised form 2 February 2005; accepted 9 February 2005 Available online 19 April 2005
Abstract The removal efficiency of zeolite (clinoptilolite) and sepiolite from lead containing aqueous solutions was investigated. A series of experiments were conducted in batch-wise and fixed-bed columns. Synthetic wastewaters containing lead (50 mg l1) and acetic acid (0.001 N) along with untreated and regenerated clinoptilolites and sepiolites were used in the adsorption studies. Batch tests were mainly conducted to isolate the magnitude of lead precipitation from real adsorption. Adsorption isotherms for both abstraction and adsorption were constructed. The removal of lead is found to be a sum of adsorption induced by ion exchange and precipitation of lead hydroxide. The breakthrough curves were obtained under different conditions by plotting the normalized effluent lead concentration (C/C0) versus bed volume (BV). The ion exchange capacity of sepiolite and clinoptilolite for lead removal showed good performance up to approximately 100 and 120 BV where the C/C0 remained below 0.1, respectively. The lead removal capacity of clinoptilolite bed from wastewater containing only lead yielded 45% higher performance compared to that of acetic acid partly due to a decrease in the effluent pH and consequently in precipitation. Also, the presence of acetic acid in the sepiolite column decreased the bed volumes treated by about 40%. Removal efficiency of lead–acetic system both in untreated clinoptilolite and sepiolite columns was found higher than that in regenerated columns. 2005 Elsevier Ltd. All rights reserved. Keywords: Lead; Wastewater treatment; Clinoptilolite; Sepiolite; Ion exchange; Adsorption
1. Introduction Heavy metal ions accumulated in the receiving environment by natural and synthetic means are toxic. Industrial wastewaters are considered as the most important source of heavy metal pollution. Metal plating, automobile, and oil industries produce more heavy metal ions than other industries (Reed and Arunacha*
Corresponding author. Tel.: +90 212 285 6568; fax: +90 212 285 6587. E-mail addresses: [email protected] (M. Turan), [email protected] (M.S. C ¸ elik).
lam, 1994). Lead is often encountered in industrial wastewaters and one of the heavy metals that have been classified as priority pollutants by the US Environmental Protection Agency (USEPA). The national interim primary drinking water regulations gives the enforceable maximum contaminant level (MCL) of lead as 0.05 mg l1. Also the USEPA describes its proposed MCL as related to the optimal corrosion control. Although chemical precipitation is most economic for treatment of wastes with high lead concentrations, ion exchange and adsorption are widely used in the removal of lead at low concentrations (Reed and Arunachalam, 1994).
0045-6535/$ - see front matter 2005 Elsevier Ltd. All rights reserved. doi:10.1016/j.chemosphere.2005.02.036
1488
M. Turan et al. / Chemosphere 60 (2005) 1487–1492
Numerous investigators have studied lead removal from wastewater by adsorption using activated carbon in batch mode (Schindler et al., 1976; Wang, 1981; Corapcioglu, 1984; Netzer and Huges, 1984) and column reactor (Arulanantham et al., 1989; Reed and Arunachalam, 1994) using iron-containing industrial wastes as a low-cost adsorbent namely blast furnace sludge (Lopez et al., 1995), waste iron (III)/chromium (III) hydroxide (Namasivayam and Raganthan, 1995), and recycled iron material (Smith and Amini, 2000). Clinoptilolite is the most abundant natural zeolite. Removal by ion exchange particularly with zeolite is more effective when the metal species are cationic. Therefore, ion exchange methods using zeolite and sepiolite exhibit a significant potential for the removal of lead from wastewaters (Semmens and Martin, 1988; Bowman et al., 1995; Vaca-Mier et al., 2001). Mesoporous minerals, clinoptilolite and sepiolite, were also applied for ammonia removal (Sirkecioglu and Erdem-Senatalar, 1995; Celik et al., 2000, 2001; Turan et al., 2000) and color removal (Armagan et al., 2003a,b) from wastewaters. This paper presents lead removal from wastewaters by clinoptilolite and sepiolite as a low-cost adsorbent. A series of adsorption experiments was conducted both in batch-wise and in ion-exchange columns. Adsorption isotherms are obtained to isolate the contribution of precipitation to adsorption. The breakthrough curves were constructed by plotting the effluent lead concentration normalized with respect to the influent lead concentration (C/C0) versus bed volumes (BV).
2. Materials and methods Clinoptilolite and sepiolite samples used in the experiments were received from Incal zeolite (Gordes, Turkey) and Mayas sepiolite (Sivrihisar, Turkey) companies, respectively. The sample was classified into different size groups: 4 mm, 2.8–4 mm, 2–2.8 mm, 1–2 mm. The chemical analyses of the clinoptilolite and sepiolite samples are given in Table 1. Since the sample is rich in Ca2+ and K+ it was identified as Ca-clinoptilolite. The clinoptilolite sample contains 90.5–92.0% clinoptilolite, 4.2–5.0% smectite, 2.0–3.5% crystobalite and 1.0–1.3% mica. Go¨rdes clinoptilolite has the following properties: cation exchange capacity 1.9–2.2 meq g1, pore diameter ˚ , purity 92%, bed porosity 40%, density 2.15 g cm3, 4A apparent density 1.30 g cm3, and suspension pH of 7.5– 7.8 at 5% solids content. The X-ray diffraction and chemical analysis of sepiolite indicate that calcite and dolomite are the major impurities accompanying sepiolite. 2.1. Batch experiments Adsorption experiments were conducted in 40-ml vials using particles in the size range of 1–2 mm. One
Table 1 Chemical analyses of Go¨rdes clinoptilolite and Sivrihisar sepiolite Constituent
Clinoptilolite (wt%)
Sepiolite (wt%)
SiO2 CaO K2O SO3 Al2O3 MgO TiO2 P2O5 Fe2O3 Na2O LOI
70.00 2.50 2.30 0.01 14.00 1.15 0.05 0.02 0.75 0.20 9.02
51.93 0.12 0.33 – 1.52 24.20 0.08 – 0.70 0.12 21.00
gram of clinoptilolite (or sepiolite) sample was mixed with 20 ml lead solution of desired concentration at a solid-to-liquid ratio of 0.05 g ml1. The vials were shaken for 2 h on a shaker and centrifuged for 15 min. The supernatant was analyzed for the lead by atomic absorption spectrophotometer TJA Solutions model Solaar 969. All analyses were made at ambient temperature (22.5 ± 1 C). Precipitation tests were conducted in the absence of mineral bed by introducing Pb(NO3)2 at a particular constant pH value. Lead nitrate undergoes above a certain pH value depending on the initial lead concentration. The precipitated lead at each pH value is collected as a separate phase by centrifugation followed by supernatant analysis for lead using AAS. The depleted amount of lead is indicative of the lead ion removal through precipitation. 2.2. Column reactor experiments The laboratory scale experimental setup consists of a set of fixed ion exchange columns, wastewater and regeneration solution tanks, feed pumps, valves and treated water tank. The cylindrical Plexiglas column has a diameter of 3 cm and height of 100 cm. The particle size of clay mineral was 1–2 mm with a bed height of 50 cm and filling weight of 370 g. Aqueous solution made of Pb(NO3)2 was used. During the adsorption process, the samples were taken in 2 h periods and analyzed using AAS. Experiments were conducted with the synthetic wastewater containing only lead (50 mg l1) and lead (50 mg l1)–acetic acid (0.001 N) to simulate an acidic wastewater, respectively. A peristaltic pump (Master flex 100) was used to feed the column at a filter rate of 6 m h1. When the normalized effluent concentration (C/C0) that is the ratio of the lead concentration in treated water to the influent lead concentration approached 1.0, the process was terminated. The breakthrough
M. Turan et al. / Chemosphere 60 (2005) 1487–1492
curves were constructed by plotting the C/C0 versus bed volumes (BV) which is defined as follows: BV ¼
VF ; VR
ð1Þ
where VF is the total water volume passing through the column, VR is the fixed bed volume of natural mineral (350 cm3), C0 is the influent lead concentration, C is the effluent lead concentration. Bed volumes per hour (BV h1) were 12 and the empty bed contact time (EBCT) was found to be 5 min.
3. Results and discussion 3.1. Batch removal tests
Adsorption density, q e (mg g-1)
100
10
1
Sep/Adsorption Sep/Abstraction
0.1
Zeo/Adsorption Zeo/Precipitation Zeo/Abstraction
0.01 0.01
0.1
1
Lead hydroxide precipitation is plausible on thermodynamic grounds. The appearance of precipitation in the bulk is an indication that surface precipitate may be also forming on the clinoptilolite surface. Whether the precipitates form on the surface is debatable (Ananthapadmanabhan and Somasundaran, 1985). In order to determine the onset of precipitation and the corresponding solubility product, a series of precipitation tests has been carried out. It is envisaged that the lead hydroxide forms in the bulk above its solubility product of 1.1 · 1020 M3 (Lurye, 1965). Accordingly, the onset of precipitation occurs at pH = 5.5 for initial lead level of about 200 mg l1. Since precipitation is prevalent in the system, a convenient way to isolate precipitation from adsorption is to define a term abstraction, a sum of adsorption plus precipitation Abstraction ¼ Adsorption þ Precipitation.
In order to identify the mechanism of adsorption process, the adsorption of lead ions onto clinoptilolite and sepiolite is determined as a function of equilibrium (residual) lead concentration (Ce) and the corresponding adsorption isotherms are plotted in Fig. 1. The adsorption isotherm of clinoptilolite exhibits two regions of interest. In the first region, which is characterized by a sharp rise in abstraction, ion exchange between lead ions and those exchangeable ions on the clinoptilolite surface occurs; this extends up to approximately 1 mg l1 residual lead concentration (Fig. 1). Precipitation of lead hydroxide simultaneously occurs in this region as well. The second region marks the onset of plateau region and also the adsorption density (qe) remains nearly constant. Similar results were reported for cobalt elsewhere (Kara et al., 2003).
10
100
1000
10000
Equilibrium Concentration, Ce (mg l-1)
Fig. 1. Abstraction, adsorption and precipitation isotherms for lead–zeolite system and lead–sepiolite system (Zeo: zeolite, Sep: sepiolite).
1489
ð2Þ
Fig. 1 illustrates all the three isotherms, i.e. abstraction, precipitation and adsorption for lead–zeolite (clinoptilolite) system. Fig. 1 indicates, though small in magnitude (<10%), lead hydroxide precipitation is definite in the initial lead concentration range (Ci) of 5– 2000 mg l1; precipitation increases with an increase in pH. When the contribution of precipitation to abstraction is taken into account, the adsorption isotherm presented in Fig. 1 is obtained. Apparently, the abstraction and adsorption isotherms almost overlap due to insignificant contribution of precipitation on adsorption. These results indicate that clinoptilolite can be used as an adsorbent in the presence of precipitates without much loss in its adsorption capacity. Also, the adsorption of lead ions onto sepiolite is determined as a function of residual lead concentration and the corresponding adsorption isotherms are presented in Fig. 1. Precipitation of lead hydroxide simultaneously is determined to assess the magnitude of precipitation. It is envisaged that the lead hydroxide forms in the bulk above its solubility product of 1.1 · 1020 M3. Accordingly, similar to clinoptilolite, the onset of precipitation occurs at pH = 5.5 for initial lead level of about 200 mg l1. Though small in magnitude (<10%), lead hydroxide precipitation is definite in the initial lead concentration range of 5–500 mg l1; precipitation increases with an increase in pH. Apparently, the abstraction and adsorption isotherms almost coincide above about 60 mg l1 residual lead concentration due to insignificant contribution of precipitation to adsorption. Consequently, sepiolite can be also used as an adsorbent in the presence of precipitates without a significant loss in its adsorption capacity. The contribution of each lead species to adsorption is plotted against pH together with the percent lead abstraction in Fig. 2. It is shown that at pH 8, which is close to the natural pH of sepiolite, the distribution of adsorbed lead species is as follows: 9% Pb+2, 90%
1490
M. Turan et al. / Chemosphere 60 (2005) 1487–1492
100
% Adsorption
80
60
40 [Pb+2] [Pb(OH)+]
20
[Pb(OH)2aq] [Pb(OH)-3] % Abstraction
0 0
1
2
3
4
5
6
7
8
9 10 11 12 13 14
pH
Fig. 2. Distribution of various adsorbed lead species as a function of pH together with the percent lead abstraction in sepiolite/lead ion system.
PbOH+ and 1% Pb(OH)2. The abstraction of lead practically ceases around pH 8. These results clearly indicate that abstraction is mainly dominated by the precipitation of Pb+2 species below pH = 8. 3.2. Column studies of clinoptilolite with lead and lead–acetic acid The breakthrough curves reveal that the normalized effluent concentration (C/C0) remains below 0.1 up to approximately 120 BV followed roughly by a linear increase for both minerals (Fig. 3). At higher bed volumes, ion exchange capacity rapidly decreases and at 200 BV the C/C0 value exceeds the ratio of 0.7. The removal of lead, however, exhibits a higher performance in absence of acetic acid. Also, the lead removal efficiency in the clinoptilolite column increased with an efficient EBCT of 5 min, compared to the short contact time of about 1 Zeo-Lead (50 mg l-1)
0.9 0.8
Zeo-Lead (50 mg l-1), Acetic acid
0.7
Zeo-Lead (50 mg l-1), Acetic acid(R) Sep-Lead (50 mg l-1)
C/C o
0.6 0.5
Sep-Lead (50 mg l-1), Acetic acid
0.4
Sep-Lead (50 mg l-1), Acetic acid(R)
0.3 0.2 0.1 0 0
50
100
150
200
Bed volumes (BV= VF/VR)
Fig. 3. Breakthrough curves obtained for both zeolite (clinoptilolite) and sepiolite using wastewaters containing lead (50 mg l1) and acetic acid (0.001 N) (Zeo: zeolite, Sep: sepiolite).
10 s reported by Vaca-Mier et al. (2001). As seen in Fig. 3, breakthrough points are defined at the values of C/C0 of 0.08 and 0.1 (115 BV and 90 BV) for synthetic wastewater containing only lead (50 mg l1) and lead (50 mg l1) with acetic acid (0.001 N), respectively. Similarly, the value of C/C0 after 14 h of column operation (168 BV), gave 0.37 for the lead-only system and 0.68 for the lead–acetic acid system. Consequently, the presence of acetic acid decreased the bed volumes treated by about 45%. While the effluent pH for the wastewater containing lead/clinoptilolite decreased from 7.12 to 6.64 during the run, in the presence of acetic acid, the effluent pH exhibited a lower starting level and decreased from 5.42 to 4.18. The addition of acetic acid appears to decrease the effluent pH by about 30%; this is explained as follows: As emphasized in the batch adsorption tests, precipitation contributes to the uptake of lead by clinoptilolite. Lead ions undergo precipitation in the form of hydroxides at pH values above about 5 depending on initial lead concentration; the higher the pH value, the greater precipitation. Since the acetic acid containing system exhibits lower pH values and in turn less precipitation, the lower performance observed in the case of lead–acetic acid system can be ascribed to the less tendency of precipitation accompanying the overall less removal of lead ions. Interestingly, as opposed to precipitation, adsorption processes induced by electrostatic and ion exchange mechanisms become stronger as the pH becomes acidic. 3.3. Column studies of sepiolite with lead and lead–acetic acid The breakthrough curves showed a slow increase in the beginning followed by a rapid increase at higher C/ C0 values. At higher BV, ion exchange capacity rapidly decreases and at 200 BV the C/C0 value exceeds the ratio of 0.7. The removal of lead with untreated sepiolite exhibits a higher performance than that in the presence of acetic acid and regenerated sepiolite (Fig. 3). The normalized lead concentration C/C0 of 0.1 was attained at the BV values of 102 and 85 for only lead (50 mg l1) and lead–acetic acid (0.001 N), respectively. Similarly, the value of C/C0 after 14 h of column operation (168 BV) gave 0.5 and 0.84 for the same systems, respectively. Consequently, the presence of acetic acid decreased the bed volumes treated by about 40%. Effluent pH for the wastewater containing lead only, decreased from 9 at the beginning of the running period to 7 at the end, and that containing lead and acetic acid from 8.44 to 6.1. This is attributed to the acidic nature of acetic acid. As emphasized in the batch adsorption tests, precipitation contributes to the uptake of lead by sepiolite. Lead ions undergo precipitation in the form of hydroxides at pH values above 5 depending on initial lead concentra-
M. Turan et al. / Chemosphere 60 (2005) 1487–1492
3.4. Column regeneration When the normalized effluent lead concentration (C/C0) approached approximately 0.7 after 200 BV, the adsorption process was terminated and regeneration started. The column regeneration was carried out for two types of mineral using a solution consisting of 30 g l1 of NaCl and 1.5 g l1 of NaOH, at pH of 11.5. Regeneration solution with a flow rate of 0.25 l min1 was pumped to the bottom of bed, thus, the up flow in the bed accelerated the desorption of lead from particles. Desorption efficiencies can be determined calculating the mass of lead removed from wastewater during ion exchange process and lead desorbed from particles during regeneration run. The regeneration of clinoptilolite was found to achieve a significant unload of 1036 mg l1 Pb in the first 10 min and then reached 218 mg l1 Pb after 20 min and the regeneration completed in 55 min (Fig. 4). Thus, the desorption efficiency was found as 95% for the lead-only system while 40 BV of regenerant was required to desorp the lead adsorbed by 200 BV of solution. The results generally indicate that clinoptilolite has significant advantages over the other alternative adsorbents (Reed and Arunachalam, 1994; Smith and Amini, 2000). Similarly, the regeneration of sepiolite was found to achieve a significant unload of 1084 mg l1 lead in the first 10 min and then reached 181 mg l1 lead after 30 min and the regeneration was completed in 60 min (Yu¨ksel, 2001). From Fig. 3, Pb adsorption mass can be calculated as the following way: 50 mg Pb l1 192 BV 0.56C=C 0 50 mg Pb l1 ð192 108Þ BV=2Þ 0.35 l zeolite bed ¼ 3360 mg Pb 412 mg Pb ¼ 2948 mg Pb. On the other hand, from Fig. 4, desorption mass can be found as, 1036 mg Pb l1 15 min þ256 mg Pb l1 ð40 15Þ minÞ 0.25 l min1 =2 ¼ 1943 mg Pb þ 800 mg Pb ¼ 2743 mg Pb. Consequently, Pb desorption efficiency during regeneration process was obtained as, Desorption efficiency ¼ 2743 mg Pb=2948 mg Pb ¼ 93%. Thus, similar to clinoptilolite, desorption efficiency was found 93% for the lead-only system.
1200 1000
Sepiolite Clinoptilolite
-1
Lead concentration (mg l )
tion; the higher the pH value, the higher precipitation. Since the acetic acid containing system exhibits lower pH values and in turn less precipitation, the lower performance observed in the case of lead–acetic acid system can be ascribed to the less tendency of precipitation accompanying the overall less removal of lead ions.
1491
800 600 400 200 0 0
10
20
30 40 Time, t (min)
50
60
70
Fig. 4. Regeneration of lead from column reactor versus time for both clinoptilolite and sepiolite (lead-only system).
3.5. Effect of regeneration on column performance The ion exchange capacities of the regenerated clinoptilolite and sepiolite for 1–2 mm size fraction using synthetic wastewater containing lead (50 mg l1) and acetic acid (0.001 N) are given as a function of BV in Fig. 3. The regenerated clinoptilolite appears to have a lower efficiency than the untreated clinoptilolite. Breakthrough curves have a significant increase at the C/C0 of 0.1 for both untreated and regenerated clinoptilolites at 90 and 72 BV, respectively. After 14 h of running time (168 BV), the C/C0 rises to 0.68 and 0.90 for untreated and regenerated clinoptilolites, respectively. Therefore, regeneration hinders the performance of ion exchange column by approximately 25%; this is contrary to the finding of clinoptilolite/ammonia system, which exhibited enhanced performance even after three regeneration steps (Celik et al., 2001). On the other hand, the C/C0 reaches to 0.76 and 0.90 at 168 BV for untreated and regenerated sepiolites, respectively. Similarly, lead–acetic system in untreated sepiolite column showed a higher performance than that in regenerated column. On the other hand, the C/C0 becomes nearly equal for both clinoptilolite and sepiolite-lead system after about 14 h of operation time.
4. Conclusions Bottle adsorption experiments together with our earlier results reveal that the removal of lead ions both by clinoptilolite and sepiolite involves the contribution of both adsorption by ion exchange and precipitation. The contribution of precipitation is found to be pH dependent and significant compared to that of adsorption. The ion exchange capacity of clinoptilolite for lead removal from wastewater showed good performance up to approximately 120 BV where the normalized
1492
M. Turan et al. / Chemosphere 60 (2005) 1487–1492
effluent concentration C/C0 is maintained below 0.1 in the column. The lead removal capacity of clinoptilolite bed from wastewater containing only lead exhibited 45% higher performance compared to that in lead and of acetic acid. The optimum regeneration of clinoptilolite was obtained with 30 g l1 NaCl solution at pH 11.5. Lead–acetic system both in untreated clinoptilolite and sepiolite columns showed higher performance than that in regenerated ones.
Acknowledgments The financial support of Turkish State Planning Commission (DPT-107) is greatly acknowledged. The help of Dr. M. Kara in Fig. 2 is appreciated.
References Ananthapadmanabhan, K.P., Somasundaran, P., 1985. Surface precipitation of inorganics and surfactants and its role in adsorption and flotation. Colloid Surface 13, 151–167. Armagan, B., Ozdemir, O., Turan, M., Celik, M.S., 2003a. Adsorption of negatively-charged azo dyes onto surfactant—modified sepiolite. J. Environ. Eng. ASCE 129, 709– 715. Armagan, B., Ozdemir, O., Turan, M., Celik, M.S., 2003b. The removal of reactive azo dyes by natural and modified zeolites. J. Chem. Technol. Biotechnol. 78, 725–732. Arulanantham, A., Balasubramanian, N., Ramakrishna, T.V., 1989. Coconut shell carbon for treatment of cadmium and lead containing wastewater. Met. Finish. 87, 51–55. Bowman, R.S., Haggerty, G.M., Huddleston, R.G., Neel, D., Flynn, M., 1995. Sorption of nonpolar organic compounds, inorganic cations, and inorganic oxyanions by surfactantmodified zeolites. In: Proc. International ACS Symposium, p. 54. Celik, M.S., Ozdemir, B., Turan, M., Benligonu¨l, B., 2000. Utilization of sepiolite in the removal of ammonia from contaminated drinking waters. In: Proc. World Filtration Congress 8 (WFC8), the Brighton Centre, Brighton, UK, pp. 377–380. Celik, M.S., Ozdemir, B., Turan, M., Koyuncu, I., Atesok, G., Sarikaya, H.Z., 2001. Removal of ammonia by natural clay materials using fixed and fluidised bed column reactors. Water Sci. Technol.: Water Supply. 1, 81–88.
Corapcioglu, M.O., 1984. Adsorption characteristics of copper(II), lead, nickel(II), and zinc(II) onto activated carbon surface in dilute aqueous solution: the effect of complex formation. Ph.D. Thesis, University of Delaware, Newark, Del. Kara, M., Yu¨zer, H., Sabah, E., C ¸ elik, M.S., 2003. Adsorption of cobalt from aqueous solution onto sepiolite. Water Res. 37, 224–232. Lopez, F.A., Perez, C., Sainz, E., Alonso, M., 1995. Adsorption of Pb2+ on blast furnace sludge. J. Chem. Technol. Biotechnol. 62, 200–206. Lurye, Y.Y., 1965. Khimiya Spravochnik po Analitickeskoy Khimii, Moscow. Namasivayam, C., Raganthan, K., 1995. Removal of lead (II) by adsorption onto waste iron(III)/chromium(III) hydroxide from aqueous solution and radiator manufacturing industry wastewater. Ind. Eng. Chem. Res. 34, 869–873. Netzer, A., Huges, D.E., 1984. Adsorption of copper, lead, and cobalt by activated carbon. Water Res. 18, 927–933. Reed, B.E., Arunachalam, S., 1994. Use of granular activated carbon columns for lead removal. J. Environ. Eng. ASCE 120, 416–436. Schindler, P.W., Faust, B., Dick, R., Wolf, P.U., 1976. Ligand properties of surface silanol groups—1. Surface complex formation with Fe3+, Cu2+, Cd2+ and Pb2+. J. Colloid Interface Sci. 55, 469. Semmens, M.J., Martin, W.P., 1988. The influence of pretreatment on the capacity and selectivity of clinoptilolite for metal ions. Water Res. 22, 537. Sirkecioglu, A., Erdem-Senatalar, A., 1995. Removal of ammonium ions from wastewaters by Bigadic¸ clinoptilolite. Tr. J. Eng. Environ. Sci. 19, 399–405. Smith, E.H., Amini, A., 2000. Lead removal in fixed beds by recycled iron material. J. Environ. Eng. ASCE 126, 58– 65. Turan, M., Aslan, F., Benligo¨nu¨l, B., Celik, M.S., 2000. Ammonia removal from a contaminated drinking water reservoir by clinoptilolite. In: Proc. World Filtration Congress 8 (WFC8), the Brighton Centre, Brighton, UK, pp. 373–376. Vaca-Mier, M., Callegas, R.L., Gehr, R., Jimenez, C.B., Alvarez, P.J., 2001. Heavy metal removal with Mexican clinoptilolite: multi-component ionic exchange. Water Res. 35, 373–378. Wang, Y.T., 1981. The removal of Pb(II) from dilute aqueous solution. M.S. Thesis, University of Delaware, Newark, Del. Yu¨ksel, B., 2001. Lead removal by natural clay materials using fixed bed column reactor. M.S. Thesis, Istanbul Technical University, Istanbul, Turkey (in Turkish).
Short Communication
Lead removal in fixed-bed columns by zeolite and sepiolite Mustafa Turan a
a,*
, Ugur Mart b, Baris Yu¨ksel a, Mehmet S. C ¸ elik
b
Istanbul Technical University, Department of Environmental Engineering, Maslak 34469, Istanbul, Turkey b Istanbul Technical University, Faculty of Mining Engineering, Maslak 34469, Istanbul, Turkey Received 26 February 2004; received in revised form 2 February 2005; accepted 9 February 2005 Available online 19 April 2005
Abstract The removal efficiency of zeolite (clinoptilolite) and sepiolite from lead containing aqueous solutions was investigated. A series of experiments were conducted in batch-wise and fixed-bed columns. Synthetic wastewaters containing lead (50 mg l1) and acetic acid (0.001 N) along with untreated and regenerated clinoptilolites and sepiolites were used in the adsorption studies. Batch tests were mainly conducted to isolate the magnitude of lead precipitation from real adsorption. Adsorption isotherms for both abstraction and adsorption were constructed. The removal of lead is found to be a sum of adsorption induced by ion exchange and precipitation of lead hydroxide. The breakthrough curves were obtained under different conditions by plotting the normalized effluent lead concentration (C/C0) versus bed volume (BV). The ion exchange capacity of sepiolite and clinoptilolite for lead removal showed good performance up to approximately 100 and 120 BV where the C/C0 remained below 0.1, respectively. The lead removal capacity of clinoptilolite bed from wastewater containing only lead yielded 45% higher performance compared to that of acetic acid partly due to a decrease in the effluent pH and consequently in precipitation. Also, the presence of acetic acid in the sepiolite column decreased the bed volumes treated by about 40%. Removal efficiency of lead–acetic system both in untreated clinoptilolite and sepiolite columns was found higher than that in regenerated columns. 2005 Elsevier Ltd. All rights reserved. Keywords: Lead; Wastewater treatment; Clinoptilolite; Sepiolite; Ion exchange; Adsorption
1. Introduction Heavy metal ions accumulated in the receiving environment by natural and synthetic means are toxic. Industrial wastewaters are considered as the most important source of heavy metal pollution. Metal plating, automobile, and oil industries produce more heavy metal ions than other industries (Reed and Arunacha*
Corresponding author. Tel.: +90 212 285 6568; fax: +90 212 285 6587. E-mail addresses: [email protected] (M. Turan), [email protected] (M.S. C ¸ elik).
lam, 1994). Lead is often encountered in industrial wastewaters and one of the heavy metals that have been classified as priority pollutants by the US Environmental Protection Agency (USEPA). The national interim primary drinking water regulations gives the enforceable maximum contaminant level (MCL) of lead as 0.05 mg l1. Also the USEPA describes its proposed MCL as related to the optimal corrosion control. Although chemical precipitation is most economic for treatment of wastes with high lead concentrations, ion exchange and adsorption are widely used in the removal of lead at low concentrations (Reed and Arunachalam, 1994).
0045-6535/$ - see front matter 2005 Elsevier Ltd. All rights reserved. doi:10.1016/j.chemosphere.2005.02.036
1488
M. Turan et al. / Chemosphere 60 (2005) 1487–1492
Numerous investigators have studied lead removal from wastewater by adsorption using activated carbon in batch mode (Schindler et al., 1976; Wang, 1981; Corapcioglu, 1984; Netzer and Huges, 1984) and column reactor (Arulanantham et al., 1989; Reed and Arunachalam, 1994) using iron-containing industrial wastes as a low-cost adsorbent namely blast furnace sludge (Lopez et al., 1995), waste iron (III)/chromium (III) hydroxide (Namasivayam and Raganthan, 1995), and recycled iron material (Smith and Amini, 2000). Clinoptilolite is the most abundant natural zeolite. Removal by ion exchange particularly with zeolite is more effective when the metal species are cationic. Therefore, ion exchange methods using zeolite and sepiolite exhibit a significant potential for the removal of lead from wastewaters (Semmens and Martin, 1988; Bowman et al., 1995; Vaca-Mier et al., 2001). Mesoporous minerals, clinoptilolite and sepiolite, were also applied for ammonia removal (Sirkecioglu and Erdem-Senatalar, 1995; Celik et al., 2000, 2001; Turan et al., 2000) and color removal (Armagan et al., 2003a,b) from wastewaters. This paper presents lead removal from wastewaters by clinoptilolite and sepiolite as a low-cost adsorbent. A series of adsorption experiments was conducted both in batch-wise and in ion-exchange columns. Adsorption isotherms are obtained to isolate the contribution of precipitation to adsorption. The breakthrough curves were constructed by plotting the effluent lead concentration normalized with respect to the influent lead concentration (C/C0) versus bed volumes (BV).
2. Materials and methods Clinoptilolite and sepiolite samples used in the experiments were received from Incal zeolite (Gordes, Turkey) and Mayas sepiolite (Sivrihisar, Turkey) companies, respectively. The sample was classified into different size groups: 4 mm, 2.8–4 mm, 2–2.8 mm, 1–2 mm. The chemical analyses of the clinoptilolite and sepiolite samples are given in Table 1. Since the sample is rich in Ca2+ and K+ it was identified as Ca-clinoptilolite. The clinoptilolite sample contains 90.5–92.0% clinoptilolite, 4.2–5.0% smectite, 2.0–3.5% crystobalite and 1.0–1.3% mica. Go¨rdes clinoptilolite has the following properties: cation exchange capacity 1.9–2.2 meq g1, pore diameter ˚ , purity 92%, bed porosity 40%, density 2.15 g cm3, 4A apparent density 1.30 g cm3, and suspension pH of 7.5– 7.8 at 5% solids content. The X-ray diffraction and chemical analysis of sepiolite indicate that calcite and dolomite are the major impurities accompanying sepiolite. 2.1. Batch experiments Adsorption experiments were conducted in 40-ml vials using particles in the size range of 1–2 mm. One
Table 1 Chemical analyses of Go¨rdes clinoptilolite and Sivrihisar sepiolite Constituent
Clinoptilolite (wt%)
Sepiolite (wt%)
SiO2 CaO K2O SO3 Al2O3 MgO TiO2 P2O5 Fe2O3 Na2O LOI
70.00 2.50 2.30 0.01 14.00 1.15 0.05 0.02 0.75 0.20 9.02
51.93 0.12 0.33 – 1.52 24.20 0.08 – 0.70 0.12 21.00
gram of clinoptilolite (or sepiolite) sample was mixed with 20 ml lead solution of desired concentration at a solid-to-liquid ratio of 0.05 g ml1. The vials were shaken for 2 h on a shaker and centrifuged for 15 min. The supernatant was analyzed for the lead by atomic absorption spectrophotometer TJA Solutions model Solaar 969. All analyses were made at ambient temperature (22.5 ± 1 C). Precipitation tests were conducted in the absence of mineral bed by introducing Pb(NO3)2 at a particular constant pH value. Lead nitrate undergoes above a certain pH value depending on the initial lead concentration. The precipitated lead at each pH value is collected as a separate phase by centrifugation followed by supernatant analysis for lead using AAS. The depleted amount of lead is indicative of the lead ion removal through precipitation. 2.2. Column reactor experiments The laboratory scale experimental setup consists of a set of fixed ion exchange columns, wastewater and regeneration solution tanks, feed pumps, valves and treated water tank. The cylindrical Plexiglas column has a diameter of 3 cm and height of 100 cm. The particle size of clay mineral was 1–2 mm with a bed height of 50 cm and filling weight of 370 g. Aqueous solution made of Pb(NO3)2 was used. During the adsorption process, the samples were taken in 2 h periods and analyzed using AAS. Experiments were conducted with the synthetic wastewater containing only lead (50 mg l1) and lead (50 mg l1)–acetic acid (0.001 N) to simulate an acidic wastewater, respectively. A peristaltic pump (Master flex 100) was used to feed the column at a filter rate of 6 m h1. When the normalized effluent concentration (C/C0) that is the ratio of the lead concentration in treated water to the influent lead concentration approached 1.0, the process was terminated. The breakthrough
M. Turan et al. / Chemosphere 60 (2005) 1487–1492
curves were constructed by plotting the C/C0 versus bed volumes (BV) which is defined as follows: BV ¼
VF ; VR
ð1Þ
where VF is the total water volume passing through the column, VR is the fixed bed volume of natural mineral (350 cm3), C0 is the influent lead concentration, C is the effluent lead concentration. Bed volumes per hour (BV h1) were 12 and the empty bed contact time (EBCT) was found to be 5 min.
3. Results and discussion 3.1. Batch removal tests
Adsorption density, q e (mg g-1)
100
10
1
Sep/Adsorption Sep/Abstraction
0.1
Zeo/Adsorption Zeo/Precipitation Zeo/Abstraction
0.01 0.01
0.1
1
Lead hydroxide precipitation is plausible on thermodynamic grounds. The appearance of precipitation in the bulk is an indication that surface precipitate may be also forming on the clinoptilolite surface. Whether the precipitates form on the surface is debatable (Ananthapadmanabhan and Somasundaran, 1985). In order to determine the onset of precipitation and the corresponding solubility product, a series of precipitation tests has been carried out. It is envisaged that the lead hydroxide forms in the bulk above its solubility product of 1.1 · 1020 M3 (Lurye, 1965). Accordingly, the onset of precipitation occurs at pH = 5.5 for initial lead level of about 200 mg l1. Since precipitation is prevalent in the system, a convenient way to isolate precipitation from adsorption is to define a term abstraction, a sum of adsorption plus precipitation Abstraction ¼ Adsorption þ Precipitation.
In order to identify the mechanism of adsorption process, the adsorption of lead ions onto clinoptilolite and sepiolite is determined as a function of equilibrium (residual) lead concentration (Ce) and the corresponding adsorption isotherms are plotted in Fig. 1. The adsorption isotherm of clinoptilolite exhibits two regions of interest. In the first region, which is characterized by a sharp rise in abstraction, ion exchange between lead ions and those exchangeable ions on the clinoptilolite surface occurs; this extends up to approximately 1 mg l1 residual lead concentration (Fig. 1). Precipitation of lead hydroxide simultaneously occurs in this region as well. The second region marks the onset of plateau region and also the adsorption density (qe) remains nearly constant. Similar results were reported for cobalt elsewhere (Kara et al., 2003).
10
100
1000
10000
Equilibrium Concentration, Ce (mg l-1)
Fig. 1. Abstraction, adsorption and precipitation isotherms for lead–zeolite system and lead–sepiolite system (Zeo: zeolite, Sep: sepiolite).
1489
ð2Þ
Fig. 1 illustrates all the three isotherms, i.e. abstraction, precipitation and adsorption for lead–zeolite (clinoptilolite) system. Fig. 1 indicates, though small in magnitude (<10%), lead hydroxide precipitation is definite in the initial lead concentration range (Ci) of 5– 2000 mg l1; precipitation increases with an increase in pH. When the contribution of precipitation to abstraction is taken into account, the adsorption isotherm presented in Fig. 1 is obtained. Apparently, the abstraction and adsorption isotherms almost overlap due to insignificant contribution of precipitation on adsorption. These results indicate that clinoptilolite can be used as an adsorbent in the presence of precipitates without much loss in its adsorption capacity. Also, the adsorption of lead ions onto sepiolite is determined as a function of residual lead concentration and the corresponding adsorption isotherms are presented in Fig. 1. Precipitation of lead hydroxide simultaneously is determined to assess the magnitude of precipitation. It is envisaged that the lead hydroxide forms in the bulk above its solubility product of 1.1 · 1020 M3. Accordingly, similar to clinoptilolite, the onset of precipitation occurs at pH = 5.5 for initial lead level of about 200 mg l1. Though small in magnitude (<10%), lead hydroxide precipitation is definite in the initial lead concentration range of 5–500 mg l1; precipitation increases with an increase in pH. Apparently, the abstraction and adsorption isotherms almost coincide above about 60 mg l1 residual lead concentration due to insignificant contribution of precipitation to adsorption. Consequently, sepiolite can be also used as an adsorbent in the presence of precipitates without a significant loss in its adsorption capacity. The contribution of each lead species to adsorption is plotted against pH together with the percent lead abstraction in Fig. 2. It is shown that at pH 8, which is close to the natural pH of sepiolite, the distribution of adsorbed lead species is as follows: 9% Pb+2, 90%
1490
M. Turan et al. / Chemosphere 60 (2005) 1487–1492
100
% Adsorption
80
60
40 [Pb+2] [Pb(OH)+]
20
[Pb(OH)2aq] [Pb(OH)-3] % Abstraction
0 0
1
2
3
4
5
6
7
8
9 10 11 12 13 14
pH
Fig. 2. Distribution of various adsorbed lead species as a function of pH together with the percent lead abstraction in sepiolite/lead ion system.
PbOH+ and 1% Pb(OH)2. The abstraction of lead practically ceases around pH 8. These results clearly indicate that abstraction is mainly dominated by the precipitation of Pb+2 species below pH = 8. 3.2. Column studies of clinoptilolite with lead and lead–acetic acid The breakthrough curves reveal that the normalized effluent concentration (C/C0) remains below 0.1 up to approximately 120 BV followed roughly by a linear increase for both minerals (Fig. 3). At higher bed volumes, ion exchange capacity rapidly decreases and at 200 BV the C/C0 value exceeds the ratio of 0.7. The removal of lead, however, exhibits a higher performance in absence of acetic acid. Also, the lead removal efficiency in the clinoptilolite column increased with an efficient EBCT of 5 min, compared to the short contact time of about 1 Zeo-Lead (50 mg l-1)
0.9 0.8
Zeo-Lead (50 mg l-1), Acetic acid
0.7
Zeo-Lead (50 mg l-1), Acetic acid(R) Sep-Lead (50 mg l-1)
C/C o
0.6 0.5
Sep-Lead (50 mg l-1), Acetic acid
0.4
Sep-Lead (50 mg l-1), Acetic acid(R)
0.3 0.2 0.1 0 0
50
100
150
200
Bed volumes (BV= VF/VR)
Fig. 3. Breakthrough curves obtained for both zeolite (clinoptilolite) and sepiolite using wastewaters containing lead (50 mg l1) and acetic acid (0.001 N) (Zeo: zeolite, Sep: sepiolite).
10 s reported by Vaca-Mier et al. (2001). As seen in Fig. 3, breakthrough points are defined at the values of C/C0 of 0.08 and 0.1 (115 BV and 90 BV) for synthetic wastewater containing only lead (50 mg l1) and lead (50 mg l1) with acetic acid (0.001 N), respectively. Similarly, the value of C/C0 after 14 h of column operation (168 BV), gave 0.37 for the lead-only system and 0.68 for the lead–acetic acid system. Consequently, the presence of acetic acid decreased the bed volumes treated by about 45%. While the effluent pH for the wastewater containing lead/clinoptilolite decreased from 7.12 to 6.64 during the run, in the presence of acetic acid, the effluent pH exhibited a lower starting level and decreased from 5.42 to 4.18. The addition of acetic acid appears to decrease the effluent pH by about 30%; this is explained as follows: As emphasized in the batch adsorption tests, precipitation contributes to the uptake of lead by clinoptilolite. Lead ions undergo precipitation in the form of hydroxides at pH values above about 5 depending on initial lead concentration; the higher the pH value, the greater precipitation. Since the acetic acid containing system exhibits lower pH values and in turn less precipitation, the lower performance observed in the case of lead–acetic acid system can be ascribed to the less tendency of precipitation accompanying the overall less removal of lead ions. Interestingly, as opposed to precipitation, adsorption processes induced by electrostatic and ion exchange mechanisms become stronger as the pH becomes acidic. 3.3. Column studies of sepiolite with lead and lead–acetic acid The breakthrough curves showed a slow increase in the beginning followed by a rapid increase at higher C/ C0 values. At higher BV, ion exchange capacity rapidly decreases and at 200 BV the C/C0 value exceeds the ratio of 0.7. The removal of lead with untreated sepiolite exhibits a higher performance than that in the presence of acetic acid and regenerated sepiolite (Fig. 3). The normalized lead concentration C/C0 of 0.1 was attained at the BV values of 102 and 85 for only lead (50 mg l1) and lead–acetic acid (0.001 N), respectively. Similarly, the value of C/C0 after 14 h of column operation (168 BV) gave 0.5 and 0.84 for the same systems, respectively. Consequently, the presence of acetic acid decreased the bed volumes treated by about 40%. Effluent pH for the wastewater containing lead only, decreased from 9 at the beginning of the running period to 7 at the end, and that containing lead and acetic acid from 8.44 to 6.1. This is attributed to the acidic nature of acetic acid. As emphasized in the batch adsorption tests, precipitation contributes to the uptake of lead by sepiolite. Lead ions undergo precipitation in the form of hydroxides at pH values above 5 depending on initial lead concentra-
M. Turan et al. / Chemosphere 60 (2005) 1487–1492
3.4. Column regeneration When the normalized effluent lead concentration (C/C0) approached approximately 0.7 after 200 BV, the adsorption process was terminated and regeneration started. The column regeneration was carried out for two types of mineral using a solution consisting of 30 g l1 of NaCl and 1.5 g l1 of NaOH, at pH of 11.5. Regeneration solution with a flow rate of 0.25 l min1 was pumped to the bottom of bed, thus, the up flow in the bed accelerated the desorption of lead from particles. Desorption efficiencies can be determined calculating the mass of lead removed from wastewater during ion exchange process and lead desorbed from particles during regeneration run. The regeneration of clinoptilolite was found to achieve a significant unload of 1036 mg l1 Pb in the first 10 min and then reached 218 mg l1 Pb after 20 min and the regeneration completed in 55 min (Fig. 4). Thus, the desorption efficiency was found as 95% for the lead-only system while 40 BV of regenerant was required to desorp the lead adsorbed by 200 BV of solution. The results generally indicate that clinoptilolite has significant advantages over the other alternative adsorbents (Reed and Arunachalam, 1994; Smith and Amini, 2000). Similarly, the regeneration of sepiolite was found to achieve a significant unload of 1084 mg l1 lead in the first 10 min and then reached 181 mg l1 lead after 30 min and the regeneration was completed in 60 min (Yu¨ksel, 2001). From Fig. 3, Pb adsorption mass can be calculated as the following way: 50 mg Pb l1 192 BV 0.56C=C 0 50 mg Pb l1 ð192 108Þ BV=2Þ 0.35 l zeolite bed ¼ 3360 mg Pb 412 mg Pb ¼ 2948 mg Pb. On the other hand, from Fig. 4, desorption mass can be found as, 1036 mg Pb l1 15 min þ256 mg Pb l1 ð40 15Þ minÞ 0.25 l min1 =2 ¼ 1943 mg Pb þ 800 mg Pb ¼ 2743 mg Pb. Consequently, Pb desorption efficiency during regeneration process was obtained as, Desorption efficiency ¼ 2743 mg Pb=2948 mg Pb ¼ 93%. Thus, similar to clinoptilolite, desorption efficiency was found 93% for the lead-only system.
1200 1000
Sepiolite Clinoptilolite
-1
Lead concentration (mg l )
tion; the higher the pH value, the higher precipitation. Since the acetic acid containing system exhibits lower pH values and in turn less precipitation, the lower performance observed in the case of lead–acetic acid system can be ascribed to the less tendency of precipitation accompanying the overall less removal of lead ions.
1491
800 600 400 200 0 0
10
20
30 40 Time, t (min)
50
60
70
Fig. 4. Regeneration of lead from column reactor versus time for both clinoptilolite and sepiolite (lead-only system).
3.5. Effect of regeneration on column performance The ion exchange capacities of the regenerated clinoptilolite and sepiolite for 1–2 mm size fraction using synthetic wastewater containing lead (50 mg l1) and acetic acid (0.001 N) are given as a function of BV in Fig. 3. The regenerated clinoptilolite appears to have a lower efficiency than the untreated clinoptilolite. Breakthrough curves have a significant increase at the C/C0 of 0.1 for both untreated and regenerated clinoptilolites at 90 and 72 BV, respectively. After 14 h of running time (168 BV), the C/C0 rises to 0.68 and 0.90 for untreated and regenerated clinoptilolites, respectively. Therefore, regeneration hinders the performance of ion exchange column by approximately 25%; this is contrary to the finding of clinoptilolite/ammonia system, which exhibited enhanced performance even after three regeneration steps (Celik et al., 2001). On the other hand, the C/C0 reaches to 0.76 and 0.90 at 168 BV for untreated and regenerated sepiolites, respectively. Similarly, lead–acetic system in untreated sepiolite column showed a higher performance than that in regenerated column. On the other hand, the C/C0 becomes nearly equal for both clinoptilolite and sepiolite-lead system after about 14 h of operation time.
4. Conclusions Bottle adsorption experiments together with our earlier results reveal that the removal of lead ions both by clinoptilolite and sepiolite involves the contribution of both adsorption by ion exchange and precipitation. The contribution of precipitation is found to be pH dependent and significant compared to that of adsorption. The ion exchange capacity of clinoptilolite for lead removal from wastewater showed good performance up to approximately 120 BV where the normalized
1492
M. Turan et al. / Chemosphere 60 (2005) 1487–1492
effluent concentration C/C0 is maintained below 0.1 in the column. The lead removal capacity of clinoptilolite bed from wastewater containing only lead exhibited 45% higher performance compared to that in lead and of acetic acid. The optimum regeneration of clinoptilolite was obtained with 30 g l1 NaCl solution at pH 11.5. Lead–acetic system both in untreated clinoptilolite and sepiolite columns showed higher performance than that in regenerated ones.
Acknowledgments The financial support of Turkish State Planning Commission (DPT-107) is greatly acknowledged. The help of Dr. M. Kara in Fig. 2 is appreciated.
References Ananthapadmanabhan, K.P., Somasundaran, P., 1985. Surface precipitation of inorganics and surfactants and its role in adsorption and flotation. Colloid Surface 13, 151–167. Armagan, B., Ozdemir, O., Turan, M., Celik, M.S., 2003a. Adsorption of negatively-charged azo dyes onto surfactant—modified sepiolite. J. Environ. Eng. ASCE 129, 709– 715. Armagan, B., Ozdemir, O., Turan, M., Celik, M.S., 2003b. The removal of reactive azo dyes by natural and modified zeolites. J. Chem. Technol. Biotechnol. 78, 725–732. Arulanantham, A., Balasubramanian, N., Ramakrishna, T.V., 1989. Coconut shell carbon for treatment of cadmium and lead containing wastewater. Met. Finish. 87, 51–55. Bowman, R.S., Haggerty, G.M., Huddleston, R.G., Neel, D., Flynn, M., 1995. Sorption of nonpolar organic compounds, inorganic cations, and inorganic oxyanions by surfactantmodified zeolites. In: Proc. International ACS Symposium, p. 54. Celik, M.S., Ozdemir, B., Turan, M., Benligonu¨l, B., 2000. Utilization of sepiolite in the removal of ammonia from contaminated drinking waters. In: Proc. World Filtration Congress 8 (WFC8), the Brighton Centre, Brighton, UK, pp. 377–380. Celik, M.S., Ozdemir, B., Turan, M., Koyuncu, I., Atesok, G., Sarikaya, H.Z., 2001. Removal of ammonia by natural clay materials using fixed and fluidised bed column reactors. Water Sci. Technol.: Water Supply. 1, 81–88.
Corapcioglu, M.O., 1984. Adsorption characteristics of copper(II), lead, nickel(II), and zinc(II) onto activated carbon surface in dilute aqueous solution: the effect of complex formation. Ph.D. Thesis, University of Delaware, Newark, Del. Kara, M., Yu¨zer, H., Sabah, E., C ¸ elik, M.S., 2003. Adsorption of cobalt from aqueous solution onto sepiolite. Water Res. 37, 224–232. Lopez, F.A., Perez, C., Sainz, E., Alonso, M., 1995. Adsorption of Pb2+ on blast furnace sludge. J. Chem. Technol. Biotechnol. 62, 200–206. Lurye, Y.Y., 1965. Khimiya Spravochnik po Analitickeskoy Khimii, Moscow. Namasivayam, C., Raganthan, K., 1995. Removal of lead (II) by adsorption onto waste iron(III)/chromium(III) hydroxide from aqueous solution and radiator manufacturing industry wastewater. Ind. Eng. Chem. Res. 34, 869–873. Netzer, A., Huges, D.E., 1984. Adsorption of copper, lead, and cobalt by activated carbon. Water Res. 18, 927–933. Reed, B.E., Arunachalam, S., 1994. Use of granular activated carbon columns for lead removal. J. Environ. Eng. ASCE 120, 416–436. Schindler, P.W., Faust, B., Dick, R., Wolf, P.U., 1976. Ligand properties of surface silanol groups—1. Surface complex formation with Fe3+, Cu2+, Cd2+ and Pb2+. J. Colloid Interface Sci. 55, 469. Semmens, M.J., Martin, W.P., 1988. The influence of pretreatment on the capacity and selectivity of clinoptilolite for metal ions. Water Res. 22, 537. Sirkecioglu, A., Erdem-Senatalar, A., 1995. Removal of ammonium ions from wastewaters by Bigadic¸ clinoptilolite. Tr. J. Eng. Environ. Sci. 19, 399–405. Smith, E.H., Amini, A., 2000. Lead removal in fixed beds by recycled iron material. J. Environ. Eng. ASCE 126, 58– 65. Turan, M., Aslan, F., Benligo¨nu¨l, B., Celik, M.S., 2000. Ammonia removal from a contaminated drinking water reservoir by clinoptilolite. In: Proc. World Filtration Congress 8 (WFC8), the Brighton Centre, Brighton, UK, pp. 373–376. Vaca-Mier, M., Callegas, R.L., Gehr, R., Jimenez, C.B., Alvarez, P.J., 2001. Heavy metal removal with Mexican clinoptilolite: multi-component ionic exchange. Water Res. 35, 373–378. Wang, Y.T., 1981. The removal of Pb(II) from dilute aqueous solution. M.S. Thesis, University of Delaware, Newark, Del. Yu¨ksel, B., 2001. Lead removal by natural clay materials using fixed bed column reactor. M.S. Thesis, Istanbul Technical University, Istanbul, Turkey (in Turkish).