Chen071231.qxd The Journal of Engineering Research Vol. 6, No. 2 (2009) 1-11 ___________________________________________ *Corresponding author’s e-mail: abuyaghi@ju.edu.jo Uptake of Cd(II) Using Natural Zeolite: Batch and Continuous Fixed-Bed Studies Luna M. Marashdeha and Ahmad M. Al-Haj-Ali*b a Department of Chemical Engineering, University of Jordan, Amman 11942, Jordan *b Department of Petroleum and Chemical Engineering, Sultan Qaboos University, P.O. Box 33, Al-Khod 123, Muscat, Sultanate of Oman Received 31 December 2007; accepted 7 October 2008 Abstract: Uptake of Cd(II) ions by natural phillipsite tuff was investigated both in shake-flask and fixed-bed columns. Equilibrium uptake, qe, was found to best fit Langmuir adsorption isotherm with a maximum value of 25.78 mg/g. Percent removal of Cd ions was close to 100% from initial metal ion concentrations in the range 50 - 75 mg/L at 5.0 g zeolite/L. Also, qe was found to vary exponentially with zeolite dose. Break points as high as 350 minutes were obtained from bed treatment at favorable conditions of a low solution flow rate and high bed depth. In batch experiments, equilibrium pH increased to < 8.0 excluding chemical precipitation as part of the removal while in fixed-beds the final pH exceeded 9.0. It is suggested that a sieve action of zeolite porous structure plays a role as an uptake mechanism in addition to the ion exchange. Keywords: Metal uptake, Ion exchange, Cadmium, Adsorption isotherms, Fixed-bed column, Natural Zeolite, Jordan Iôªà°ùŸŸG áàHÉãdG á≤Ñ£dG h áÑLƒdG »à≤jô£d á°SGQO /»©«Ñ£dG âj’ƒjGõdG ᣰSGƒH Ωƒ«eOɵdG ¢UÉ°üàeEG »∏Y êÉ◊G ~ªMG h I~°TGôŸG Éfƒd ::áá°°UUÓÓÿÿGG≠∏Ñj ¢UÉà°üeEÓd ≈°übC’G ~◊G ¿G ~Lh ~bh .áàHÉK äÉ≤ÑW h áLÉLQ ¥QGhO ‘ É«ÑjôŒ âjÉ°ùÑ«∏«a ´ƒf øe âj’ƒjR ᣰSGƒH Ωƒ«eOɵdG äÉfƒjCG ¢UÉ°üàeEG á°SGQO â“ øe %100 ‹GƒM ≠∏ÑJ ¥QGh~dG ‘ ádGREÓd ájƒÄŸG áÑ°ùædG ¿CG É° jCG ~Lh ~bh .Òª‚’ êPƒ‰ ™Ñàj ¿GõJ’G ¿G ɪc âj’ƒjõdG áYôL ™e á«°SCG ádG~H ¢übÉæàj h ºZ /º¨∏e 25^78 𫨰ûJ ±hôX ~æY ∂dP h âj’ƒjõdG IQ~b PÉØæà°SEG πÑb á≤«bO 350 √QG~≤e øeõd ¢UÉ°üeOE’G ôªà°SEG ~≤a áàHÉãdG á≤Ñ£dG ‘ ÉeCG .Îd/Ωƒ«eOÉc º¨∏e 75-50 õ«cÎdG äGP π«dÉÙG 8^0 øe πbG ¤G »æ«LhQ~«¡dG ¢SC’G ™ØJQG ¿GõJ’G á°SGQ~d áLÉLôdG ¥QGh~dG ÜQÉŒ ‘ .âj’ƒjõdG øe á≤«ªY á≤ÑWh Ωƒ«eOɵdG ∫ƒ∏Ù ¢ Øîæe ¿ÉjôL ∫~©e ‘ πãªàJ áªFÓe ¢UÉ°üeO’G á«dBG ¿CG ≈∏Y á°SGQ~dG èFÉàf ∫~J .9^0 ¤G »æ«LhQ~«¡dG ¢SC’G ™ØJQG ~≤a Iôªà°ùŸG áàHÉãdG á≤Ñ£dG ÜQÉŒ ‘ ÉeCG .Ωƒ«eOɵ∏d »FÉ«ª«c Ö«°SôJ çh~M »Øæj Gògh .»LQÉÿG í£°ùdG ≈∏Y ʃj’G ∫OÉÑàdG ¤G áaÉ°VEG ¿~©ŸG äGƒéa ∫ÓN ÉgQÉ°ûàfEG ~©H Ωƒ«eOɵdG äÉfƒjC’ »FÉjõ«ØdG ¢UÉ°üàeE’G ≈∏Y πª°ûJ áá««MMÉÉààØØŸŸGG ääGGOOôôØØŸŸGG .¿OQC’G ,»©«Ñ£dG âjÓjõdG ,áàHÉãdG á≤Ñ£dG êôH ,ÊGõJE’G RGõàeE’G äÉ«æëæe ,Ωƒ«eOɵdG ,ʃjC’G ∫OÉÑàdG ,äGõ∏ØdG ádGREG : Notation b Langmuir isotherm parameter related to adsorption energy (L/mg) Co Initial Cd ion concentration (mg/L) Ce Equilibrium Cd ion concentration (mg/L) D Zeolite Dose in equilibrium experiments (g /L solution). IUPAC International Union of Pure and Applied Chemistry K Equilibrium constant of Cd hydrolysis reaction KF Freundlich isotherm parameter related to adsorption capacity (L/g) n Freundlich isotherm parameter related to adsorption intensity (dimensionless) qe Equilibrium uptake (mg Cd/g zeolite) qm Experimental parameter indicating maximum uptake (mg/g) QL Langmuir isotherm parameter related to adsorption capacity (mg/g) r Dimensionless separation factor to describe the type of adsorption S Experimental coefficient (L/g) Z symbol representing active sites on the zeolite surface. 2 The Journal of Engineering Research Vol. 6, No. 2 (2009) 1-11 1. Introduction Heavy metal ions are major pollutants in the waste- waters of many industrial activities. Cadmium is intro- duced into the aquatic environment by discharge of liquid effluents with relatively low metal concentration from smelting, metal plating, cadmium-nickel batteries, phos- phate fertilizer, mining, paints, metal alloys and other chemical and engineering industries (Alloway, 1995; Stumm and Morgan, 1996). Such metal ions are soluble in water and could penetrate through soil layers thus pollut- ing soil, public water supplies and accumulating in agri- cultural crops. Cd is the most toxic element readily taken up by plants (Dabrowski et al. 2004). Therefore, the removal of Cd and other metal ions from waste effluents before their discharge is a health and environmental prior- ity required by international and local standards. The harmful effects of cadmium on human health are well doc- umented including a number of acute and chronic disor- ders such as renal damage, anemia and hypertension (Friberg et al. 1974). The discharge limits from industrial effluents vary according to fate or application of wastewater. For exam- ple, the maximum allowable concentration of Cd in the Jordanian Wastewater (both industrial and treated domes- tic sewage) Standards is 0.01 mg/L for reuse in irrigation and discharge to streams and lakes; 0.02 mg/L for under- ground water recharge and 0.07 mg/L for discharge to the sea. Drinking water guideline value for Cd recommended by World Health Organization (WHO) is 0.005 mg /L (WHO, 2004), which is adopted by the Jordanian standard for drinking water. Two engineering techniques for heavy metal removal are usually applied, namely chemical pre- cipitation and ion exchange. These methods suffer from disadvantages such as non-selectivity, high cost and waste disposal problems. Natural zeolites are crystalline microporous aluminosil- icates with well-defined structures that consist of a frame- work formed by tetrahedra of SiO4 and AlO4 connected through oxygen atoms. The frame-work contains channels and cavities with molecular dimensions from 0.3 to 1 nm occupied by alkaline and alkaline earth metal cations as well as water molecules. These cations possess the ability to move freely and can be exchanged by more selective cations from aqueous solutions (Sand and Mumpton, 1978). In several parts of the world, natural zeolites have been used as ion exchangers for the removal of ammonium ions from municipal wastewater (Hedstrom, 2001); cesium and strontium ions from radioactive wastewater and heavy metals from industrial wastewater (Ouki et al. 1994). Fixed beds of zeolite may be used as packing materials in subsurface barriers to control ground water pollution (Cincotti et al. 2006). These applications are stimulated by the non-toxic nature of zeolites as well as their avail- ability in the many countries. Application of natural zeo- lites for the removal of heavy metals such as Cd(II), Pb(II), Ni(II), etc. has been reported in several studies considering both in equilibrium batch experiments (Berber-Mendoza, et al. 2006; Alvarez-Ayuso, 2003; Lee and Moon, 2001) as well as dynamic fixed-bed column experiments (Inglezakis and Grigoropoulou, 2004; Al- Haj-Ali and Al-Hunaidi, 2004) or both (Vaca-Mier, et al. 2001). Results of the above studies indicated that natural zeolites possess good ion exchange properties and thus can be used to substitute or assist ion exchange systems employing commercial synthetic resins in heavy metal removal from wastewater. This recommendation is based on relatively high removal efficiency as well as low cost of the mineral. In the northeastern part of Jordan, large deposits of nat- ural zeolite produced by alteration of basaltic glass are present. Currently this mineral is being mined and made commercially available for agricultural applications where it is used as a soil conditioner. Environmental appli- cations of this mineral, particularly as an adsorbent / ion exchange) has been the subject of investigation for more than 15 years. Published studies on this mineral have shown that it is of a promising potential in heavy metal removal from wastewater (Dwairi, 1992; Al-Haj-Ali and El-Bishtawi, 1997, Rashdan, 2002). This zeolite has become a commercial commodity in the local market available at low prices. The fact that natural zeolite can be regenerated after heavy metal removal makes it even more attractive as an ion exchanger. According to Rashdan (2002), reactivation of zeolite loaded with Pb, Cd, Ni, Cu and Zn ions with CaCl2, KCl or NaNO3 was successful up to 80-100% of the pre-adsorbed amounts of the indicated metals depending on the metal as well as the type and con- centration of eluant solution. The main objectives of this study are to evaluate the ability of natural Jordanian zeolite for Cd ions removal under certain conditions both in batch and continuous fixed-bed contact modes. 2. Experimental 2.1 Materials The zeolite used in this work was phillipsite-rich tuff mined from the Aritain mountain area in the northeastern part of Jordan and supplied by the Natural Resources Authority of Jordan. The zeolite was crushed using a Jaw crusher and sieved into three size cuts, namely, 355-500, 500-710 and 710-850 µm. Particles were washed with tap water until the dust and fines are removed. The miner- al was immersed in 25g.dm-3 solution of NaCl (analytical reagent, Analar Co.) in distilled, deionized water at room temperature for 24 h. The solution was replaced by a fresh one and the mineral continued to be immersed for anoth- er 24 h. The sodium homo-ionic zeolite was then filtered, washed gently with distilled, deionized water and oven- dried at 35 oC. The chemical composition of raw zeolite (as received) was obtained by x-ray fluorescence (XRF) analysis and is given in Table 1, which also presents relat- ed physical properties of the zeolite. The Langmuir and BET surface areas, total pore volume and average pore 3 The Journal of Engineering Research Vol. 6, No. 2 (2009) 1-11 diameter were determined for two zeolite sieve cuts (355- 500 and 500-710 µm) using a Quantachrome Autosorb automated gas sorption system. Data are calculated and reported using Autosorb 1 program for nitrogen at 77K. Cadmium nitrate was used as sources for Cd ions. Stock solutions of 1000 mg/L were prepared by dissolving the exact weight of Cd(NO3)2.4H2O in distilled-deionized water and used to provide the desired Cd ion concentra- tion by proper dilution. Glassware was cleaned by deter- gent and double rinsed thoroughly by tap water and dis- tilled-deionized water. 2.2 Equipment, Procedures and Sample Analysis 2.2.1 Batch Experiments Equilibrium uptake was determined by shake-flask batch experiments. A volume 150 mL of solution contain- ing 50 or 75 mg/L was placed in a 250 mL Erlenmeyer flask. The initial pH was adjusted to a specified initial value (4.0 or 5.5) while the final pH was not controlled. An accurately-weighed zeolite mass (in the range 0.12-1.2 g) was added to the solution thus making slurry concentra- tion in the range 1.0- 8.0 g/L. A series of such flasks was then agitated at a constant speed in a laboratory shaker at room temperature (20- 22°C). After 120 hours of contact to ensure that equilibri- um is attained, the zeolite particles were separated by fil- tration. A sample of the solid-free solution was analyzed for Cd remaining concentration using a using flame atom- ic absorption spectrophotometer (Pye-Unicam Model). Before sample analysis, the AAS was calibrated using cer- tified standard cadmium solutions. Meanwhile, the final pH of the solution was measured using Mettler Toledo pH meter calibrated with three standard buffer solutions of pH 4.01, 7.00 and 9.21. In another group of shake-flask experiments, blank runs involving zeolite in distilled-deionized water (no cadmi- um present) were conducted to investigate the effect of adsorbent on solution pH. In these experiments, the same solution volume, slurry concentration and initial pH were maintained and the same procedure above was followed. A third group of shake-flask experiments were conducted to investigate the effect of pH on chemical precipitation of cadmium in the absence of zeolite adsorbent. Solutions at initial concentrations of 24, 48, 70 and 91 mg Cd/L were shaked in 250-mL flasks at a specified pH value ranging from 4.0 to 10.0 then samples of precipitate-free solution were analyzed for remaining Cd concentration using AAS technique as indicated above. 2.2.2 Continuous Experiments These experiments were conducted using a laboratory scale glass column, 1.5cm inside diameter and 90 cm height. The column is gently packed with different mass- es of zeolite which provide different bed depths. The feed solution is pumped from a glass storage tank to an over- head constant-level tank made of plexi-glass and fed to the column through a glass flow meter capable of reading in the range 10-80 cm3.min-1. Flow control is achieved by valves fitted at the top and at the bottom of column. A fil- ter is located at both ends of the column and flexible con- nections permit column to be charged, sampled, emptied and washed. Samples at zero time (from solutions before contact with zeolite) as well samples from effluent solu- tions are withdrawn at predetermined times in the range of 3 to 450 min). The concentrations of Cd in these samples were measured using AAS as discussed above. Experimental column variables and their tested ranges are given in Table 2. 3. Results and Discussion 3.1 Equilibrium Uptake Equilibrium uptake (qe, mg Cd/g zeolite) was calculat- ed from the following mass balance equation: qe = (Co - Ce)/D (1) Weight Percent 34.20 10.57 7.02 11.35 7.28 2.00 1.65 1.23 0.33 Zeolite Composition * SiO2 Al2O3 FeO CaO MgO K2O Na2O TiO2 Other * Un-sieved 2.380 g/cm 3 1.035 g/cm 3 Yellowish Brown 124 m2/g 92 m2/g 4.02 x 10 -2 cm3/g 1.75 Å 120 m2/g 90 m2/g 3.90 x 10 -2 cm3/g 1.75 Å (Å =0.1ì m) For un-sieved zeolite True Density Bulk Density Color For size cut 355 -500 µm Langmuir Surface Area BET Surface area Total Pore Volume Avg. Pore Diameter For size cut 710 -850 µm Langmuir Surface Area BET Surface are a Total Pore Volume Avg. Pore Diameter Table 1. Physical and chemical properties of zeolite tuff 4 The Journal of Engineering Research Vol. 6, No. 2 (2009) 1-11 Experiments were carried out to determine the equilib- rium uptake using different zeolite doses. The results obtained are plotted as percent cadmium ion removal ver- sus D (Fig. 1a) and as qe versus D (Fig. 1b) for two initial Cd II ion concentrations (Co= 50 and 75 mg/L). Figure 1 (a) shows that almost 100% removal was possible using a minimum 5 g/L of zeolite at both initial concentrations. At lower D values, removal percent was higher for the lower Co value as expected with a difference of ~ 17% at 1 g zeolite/L. As a general trend and for all cases, it is clear that as the amount of zeolite per unit volume of solution is increased, metal uptake is significantly reduced. This observation is explained by the competition between metal ions for active (ion exchange) sites on the zeolite surface. Higher Co resulted in higher uptake per unit mass of zeolite (qe). Using non-linear regression technique, the data plotted in Fig. 1 (b) are found to best fit an exponen- tial decay function in the form: qe = qm. exp (- S.D) (2) Parameter qm (mg/g) is the upper limit of this decay function at minimum D indicating maximum uptake. The correlation coefficients for the two curves shown are 0.996 and 0.989 indicating good fit. Parameter S indicates how sharp is the change of metal uptake, qe, with zeolite dose, D, that is, the slope of this relationship. The quanti- ty qm increases as Co increases. 3.2 Equilibrium pH In Fig. 2, the equilibrium solution pH is plotted as a function of zeolite dose for the two initial concentration values considered. The solution pH has increased sharply from 4.0 for zeolite-free solution to 6.48 after addition of 0.8 g zeolite/L. Final pH continued to increase with increasing zeolite dose up to 7.72 at D = 8.0 g zeolite/L for Co = 50 mg/L. Similarly, pH increased between 6.37 and 7.34 when D increased from 0.8 to 8.0 g zeolite/L for Co = 75 mg/L. At any zeolite dose, equilibrium pH is found to be slightly lower for the higher initial metal ion concen- tration. At D = 8.0 g zeolite/L, the final pH is 7.34 for Co = 75 mg/L while it is 7.72 for 50 mg/L. Similar observa- tions have been reported by many authors investigating zeolite / metal systems. Al-Haj-Ali and El-Bishtawi (1997) investigated Pb(II) and Ni(II) sorption by a fine sieve cut from the same zeolite mineral and found that solution pH increased from an initial value of 4.0 to an equilibrium value of 5.7-7.9 for Pb(II) and to 6.9-7.3 for Ni(II) ions. Moderate changes in pH were also reported by Genc-Fuhrman, et al. (2007) for Cd(II) adsorption on nat- ural clinoptilolite, alumina and sand. The solution pH increased from 6.5 initially to 7.7, 7.4 and 7.2, respective- ly at a zeolite dose of 20 g/L at equilibrium. The results obtained in this study are also in agreement with those of by Alvarez-Ayuso et al. (2003) who investigated Cd ions uptake by clinoptilolite natural zeolite from Greece. They have found that starting with initial Cd ion concentrations of 50, 60 and 75 mg/L, the equilibrum pH values were 7.1, 7.0 and 6.9, respectively. 0.0 1.0 2.0 3.0 4.0 5.0 6.0 7.0 8.0 9.0 Zeolite Dose, D (g/L) 10 20 30 40 50 60 70 80 90 100 110 P e rce n t R e m o v a l Co = 75 mg Cd/L Co = 50 mg Cd/L Particle Size: 355-500 m Initial pH = 4.0 µµ (1a) 0.0 1.0 2.0 3.0 4.0 5.0 6.0 7.0 8.0 9.0 Zeolite Dose, D (g/L) 0 5 10 15 20 25 30 U p ta k e , q e (m g C d / g ) Co = 50 mg Cd/L Co = 75 mg Cd/L Particle Size: 355-500 m Initial pH = 4.0 µ (1b) Figure 1. Effect of zeolite dose on (a) Percent removal (b) Equilibrium uptake 0.0 1.0 2.0 3.0 4.0 5.0 6.0 7.0 8.0 9.0 Zeolite Dose, D (g/L) 4.0 4.5 5.0 5.5 6.0 6.5 7.0 7.5 8.0 8.5 9.0 9.5 10.0 S o lu tio n p H Particle Size: 355-500 m µµ Uptake: zeolite in Cd solution (C o=50 mg/L) at pH 4.0 Uptake: zeolite in Cd solution (C o=75 mg/L) at pH 4.0 Blank: zeolite in water at pH 4.0 Blank: zeolite in water at pH 5.5 Figure 2. Equilibrium pH for cadmium uptake and blank systems Pe rc en t R em ov al Zeolite Dose, D (g/L) U pt ak e, q e (m g C d/ g) Zeolite Dose, D (g/L) So lu tio n pH Zeolite Dose, D (g/L) 5 The Journal of Engineering Research Vol. 6, No. 2 (2009) 1-11 The final solution pH is the resultant of two effects; first- ly the initial pH of the cadmium nitrate solution, and sec- ondly the addition of zeolite particles to that solution. It is known that the pH of Cd(NO3)2.4H2O solution (and other similar metal salts) is below 7.0, the pH of pure water, since this is a weak salt obtained from a strong acid (HNO3) and a weak base [Cd(OH)2]. As an example, the pH of a solution whose Co = 70 mg Cd /L is 5.7; the lower the Cd initial concentration, the higher is the solution pH. The effect of the second parameter, that is, addition of zeolite was experimentally investigated for two different initial pH values (4.0 and 5.5) using a zeolite particle size cut of 355-500 µm in distilled-deionized water (blank runs). The results are shown also in Fig. 2 along with equi- librium pH from uptake experiments. In general, the final pH of zeolite-extract is always higher than that of the water. It is seen that as the zeolite dose added to water increased from 0.8 to 8.0 g/L, the final pH significantly increased from 4.0 to 8.9 and from 5.5 to 9.6. This is due to release of alkaline species from the active sites on the zeolite surfaces in contact with water which has been demonstrated by similar studies involving natural zeolite (Vaca-Mier, et al. 2001, Alvarez-Ayuso, et al. 2003; Camilo, et al. 2005; Kocaoba, et al. 2007). Raw natural zeolite possesses a significant content of alkaline metals (Ca, Mg, K and Na). Sodium is the major species contributing to pH rise in the case of the particu- lar natural zeolite used in the study since it has been added to the mineral preparing it as Na+ ion exchanger. Calcium carbonate, present as an impurity in the zeolite tuff is expected to take part in the pH rising process. It is note- worthy that the higher the initial pH of water, the higher the final equilibrium pH of zeolite-water slurry; the differ- ence narrows down at high D. The decrease in solution pH due to the removal of Cd ions from solution by zeolite will counter the rise in pH due to zeolite dissolution thus maintaining pH level below 8.0 compared to blank experimets resulting in high pH levels (9.1- 9.6). The fact that equilibrium pH increases during metal uptake (Pb, Ni, Cd, Cu, etc.) raises a concern about possi- ble chemical precipitation of metal ions as hydroxide thus No. Flow Rate, F (mL/min) Bed Depth, H (cm)* Initial Cd(II) Conc., C o (mg/L) Particle Size Cut, (µm) Initial Solution pH Break Time, tB (min) H/F (min /cm2) 1 22 14.4 50 355-500 4.0 95 0.65 2 15 14.4 50 355-500 4.0 215 0.96 3 10 14.4 50 355-500 4.0 350 1.44 4 10 9.6 50 355-500 4.0 190 0.96 5 10 4.8 50 355-500 4.0 75 0.48 6 15 14.4 100 355-500 5.5 105 0.96 7 15 14.4 75 355-500 5.5 130 0.96 8 15 14.4 50 355-500 5.5 230 0.96 9 15 14.4 75 355-500 4.0 130 0.96 10 15 14.4 75 710-850 4.0 110 0.96 11 15 14.4 50 500-710 4.0 130 0.96 12 15 14.4 50 500-710 5.5 140 0.96 Table 2. Break time in zeolite beds at different experimental conditions *Corresponding to 10, 20 and 30 g of zeolite mass Table 3. Specification of cadmium due to aqueous dissolution and hydrolysis reactions 6 The Journal of Engineering Research Vol. 6, No. 2 (2009) 1-11 masking true uptake or ion exchange. Kocaoba, et al. (2007) declared that most of the heavy metal ions tend to precipitate at pH > 6 and thus making it difficult to quan- tify true adsorption. Cadmium may be an exception from this general statement as indicated by cadmium specifica- tion data in Table 3 which is compiled by referring to dif- ferent aqueous chemistry and metal removal publications. The main relevant conclusion from this Table is that chemical precipitation does nor occur at pH < 8.0. Other published studies give more insight about the possibility of precipitation during the course of sorption. Deliyanni and Matiz (2002) conducted their experiments on Cd sorption by synthetic akaganeite [β-FeO(OH)] up to pH 9.5 from a solution of 10 mg Cd/L considering that cadmi- um removal at this pH is due only to sorption. They based their conclusion on the thermodynamic aqueous specia- tion of Cd which indicates that chemical precipitation (as hydroxide) is not taking place below pH 9.5. Srivastava, et al. (2004) in their study on Cd adsorption on kaolinite considered Cd2+ as the only species dominant up to pH 10. Using solutions containing 15 mg Cd/L, they excluded chemical precipitation as a Cd removal mechanism in the studied range. However, Rangel-Mendez and Streat (2002) reported that Cd exists predominantly as Cd(H2O)62+ ions and precipitates as Cd(OH)2 just below pH 8. The results of shake-flask experiments conducted to determine the pH at which Cd precipitates from solutions at 24, 48, 70 and 91 mg/L in absence of zeolite are shown in Fig. 3 as percent cadmium reduction against pH. This Figure indicates that precipitation of Cd ions as Cd(OH)2 is rather slight (less than 10%) below pH 8.6 for both ini- tial concentration of 24 and 48 mg/L. For the two higher initial concentrations of 70 and 91 mg/L, precipitation is significant only beyond pH 8.3. These results are in agree- ment with the above-mentioned studies (Deliyanni and Matiz, 2002; Srivastava, et al. 2004; Berber-Mendoza, et al. 2006). Since the maximum pH reached in batch uptake experiments in this study is less than 8.0 (Fig. 2), therefore chemical precipitation of cadmium at pH < 8.0 is ruled out and Cd removal is confirmd as a totally sorption process. The uptake mechanism is discussed further in Section 3.5. 3.3 Modeling Uptake with Adsorption Isotherms The adsorptive capacity of a solid is concluded from equilibrium sets of data that need to be fitted to mathemat- ical models to facilitate design of adsorption systems and broaden their application. Equilibrium uptake data were fitted to popular models of adsorption isotherms, namely; Langmuir, Freundlich and Sips models. Langmuir isotherm has the following non-linear form: qe = QLb.Ce / (1 + b.Ce) (3) Based on parameter "b", the dimensionless separation factor, r, is defined as follows: r = 1/ (1+b.Co) (4) This factor is used to decide whether the adsorption process is favorable (0 < r < 1) or not. In the Cd(II) / phillipsite tuff system investigated, "r" has the values of 0.02 and 0.03 for initial concentrations of 75 and 50 mg/L, respectively, indicates a very favorite adsorption process. Freundlich isotherm has the following form: qe = KF. Ce 1/n (5) The predictions of the two models are shown in Fig. 4 compared to experimental data. Regression analysis has shown that Langmuir isotherm represents the data better than Freundlich isotherm (correlation coefficients, R2, are 0.995 and 0.977, respectively. This is an indication of mono-layer coverage of metal ions binded to specific sites rather than accumulation of ions on particle surfaces. Langmuir and Freundlich isotherms parameters obtained in this study by fitting experimental equilibrium data to these models are given in Table 4 compared to those reported in the literature for cadmium uptake by other adsorbents and ion exchangers. The Langmuir parameter, QL, represents the maximum sorption capacity which is 25.78 mg Cd/g for the system being studied. This is a rel- atively high sorption capacity compared with the perform- ance of other sorbents from the same nature, ie. soils and minerals. 3.4 Breakthrough Curves The effects of solution flow-rate, bed depth, initial con- centration of Cd ions in the feed, zeolite particle size and initial solution pH on the sorption behavior of Cd(II) ions are presented and discussed. The breakthrough time (tB) is taken as an indicator of zeolite bed efficiency and defined as the time required to reach the breakpoint assumed to be at 5% breakthrough (that is, the time at which C/Co = 0.05). The higher the break time, the higher is the metal uptake and the more efficient is the bed. The tB values obtained from various breakthrough curves describing Cd(II) uptake by zeolite bed are given in Table 2 along with the respective column parameters. Figure 3. Percent cadmium reduction versus solution pH Pe rc en t C ad m iu m R ed uc tio n Solution pH 7 The Journal of Engineering Research Vol. 6, No. 2 (2009) 1-11 3.4.1. Effect of Solution Flow Rate Breakthrough curves obtained using three solution flow-rates keeping other parameters constant (zeolite mass = 30 g equivalent to 14.4 cm of bed depth) are shown in Fig. 5. Break times of 350 and 215 min obtained at the low F values of 10 and 15 mL/min are considered high and denotes an efficient bed. However, the bed per- formance would be lower at higher solution flow rates. This is inferred from the significantly shorter break time at 22 mL/min. 3.4.2. Effect of Bed Depth In Fig. 6, the bed depth (H) was varied keeping all other parameters constant. As expected, the highest value of tB was obtained using the highest bed depth (14.4 cm) corre- sponding to the highest zeolite mass available. Decreasing H while keeping F constant at 10 mL/min) resulted in shorter break time. It is concluded that at the conditions tested, a bed depth less than 9.6 cm (zeolite mass less than 20 g) will not give a satisfactory performance. The effect of bed depth, H, and solution flow rate, F, Reference Sorbent Material and Country Langmuir Parameters QL (mg/g) b (L/mg) Freundlich Parameters KF (L/g) n Garcia-Sanchez et al. (1999) Sepiolite, Spain 8.26 1.67 6.17 3.80 Singh et al. (2001) Phosphatic Clay; Florida, US 24.5 0.111 --- --- Alvarez-Ayuso et al. (2003) Clinoptilolite, Greece 4.60 0.162 --- --- Berber-Mendoza et al. (2006) Clinoptilolite, Mexico 29.11 0.022 20.57 2.22 Cincotti et al. (2006) Clinoptilolite; Sardinia, Italy 10.40 1.08 --- --- Sprynskyy et al. (2006) HCl-treated Clinoptilolite, Ukraine 4.88 0.078 0.636 2.20 Genc-Fuhrman et al. (2007) Clinoptilolite, UK --- --- 0.21 1.07 Genc-Fuhrman et. al. (2007) Sand, Denmark --- --- 0.42 3.57 Genc-Fuhrman et al. (2007) Alumina, Denmark --- --- 6.35 0.86 Genc-Fuhrman et al. (2007) Fly ash, Denmark --- --- 5.01 0 .64 This work Phillipsite tuff, Jordan 25.78 0.636 10.29 3.78 Table 4. Cd(II) adsorption parameters of Jordanian zeolite compared to other zeolites and mineral adsorbents 0 5 10 15 20 25 30 35 40 45 50 55 Ce (mg Cd/L) 0 5 10 15 20 25 30 35 q e (m g C d / g ) Experimental Data Freundlich Isotherm Langmuir Isotherm Figure 4. Modeling experimental data with a adsorption isotherms Figure 5. Breakthrough curves at variable solution flow rate C / C o Time (min)Ce (mg Cd/L) q e (m g C d/ g) 8 The Journal of Engineering Research Vol. 6, No. 2 (2009) 1-11 can be combined in one parameter, the H/F ratio which has the units of min/cm2. The break time, tB is plotted versus H/F for the data given in Table (2) obtained at con- stant Co (50 mg/L), constant initial pH of solution (4.0) and using one size cut of particles (355-500 µm). The point of original (0, 0) is also included in the data. Linear regression supported a straight line (Fig. 7) with 0.957 correlation coefficient, 246.7 as the slope and -30.4 as the intercept. The basic idea behind the H/F ratio is the complement- ing roles of sorbent bed depth and flow rate of feed solu- tion. Furthermore, it is an interaction of two effects, name- ly, the availability of active surface sites from the sorbent side and the adequacy of residence time from the solution side. Both factors are crucial for a successful sorption process which usually involves mass transfer of ions from the aqueous phase to solid surface and the subsequent binding to that surface. Using high bed depth, which pro- vides easily accessible ion exchange sites, compensates for lack of sufficient residence time of ions in contact with the bed. To simplify the idea, it can be said that for a given removal efficiency, using high bed depth with high flow rate is the same as using a low bed depth with a low flow rate. 3.4.3. Effect of Initial Cd ion Concentration Figure 8 shows the effect of Cd(II) ion concentration in feed solution, Co, on break time is shown in maintain- ing constant F = 15 mL/min and H = 14.4 cm. The slopes of breakthrough curves decreased as Co decreased indicat- ing better sorption behavior at lower Co, where break time reached 230 min compared to 150 and 105 min at the higher Co values of 75 and 100 mg/L, respectively. This observation is in agreement with the findings of Takac and Calik (1998) for ion exchange. 3.4.4. Effect of Zeolite Particle Size The effect of zeolite particle size on breakthrough curves is shown in Fig. 9. The effect is not large where break time decreased only by 15% when coarse particles (710-850 µm) are used compared to the case of the finer particles (355-500 µm). A look at Table (1) reveals that these two size cuts have similar physical properties in terms of specific surface area (91- 92 m2/g) and average pore diameter (1.75Å). Therefore, it is not surprising that the two zeolite fractions exhibit comparable sorption per- formances. Figure 6. Breakthrough curves at variable bed depth Figure 7. Effect of bed-depth to solution flow-rate ratio, H/F 0 50 100 150 200 250 300 350 400 450 Time (min) 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 C / C o 100 mg/L 75 mg/L 50 mg/L Initial Cd Ion Concentration Fixed Conditions : Bed Depth: 14.4 cm Flow Rate: 15 mL/min Particle Size: 355- 500 m Initial pH: 5.5 µµ Figure 8. Breakthrough curves at variable initial Cd ion concentration 0 50 100 150 200 250 300 350 400 450 500 Time (min) 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 C / C o 710-850 m 355-500 m Zeolite Particle Size µµ µµ Fixe d Conditions : Bed De pth: 14.4 cm Flow Rate : 15 mL/m in Co = 75 mg/L Initial pH: 4.0 Figure 9. Breakthrough curves at different zeolite particle sizes C / C o Time (min) C / C o Time (min) Bed Depth / Solution Flow Rate, H/F (min/cm2) B re ak T im e, T B (m in ) C / C o Time (min) 9 The Journal of Engineering Research Vol. 6, No. 2 (2009) 1-11 3.4.5. Effect of Initial Solution pH and Change in pH Figure 10 involves two breakthrough curves obtained using solutions at two initial pH values keeping other parameters constant. Little effect is observed (about 10% more) in break time as pH increased from 4.0 to 5.5. This is a well-known observation in metal uptake studies which is explained by the fact that the concentration of hydrogen ions (H+), which competes with Cd ions for attachment on zeolite sites, decreases with pH increase thus enhancing Cd ions removal. Hence, the bed reaches break point ear- lier at pH 5.5 than in the case of pH 4.0. The changes of solution pH during cadmium ion uptake in a typical experiment are shown in Fig. 11. Starting with feed solutions initially adjusted to pH 4.0 or 5.5; the solu- tion pH during sorption has risen sharply in the first 15 minutes to 8.4- 9.0 then declined slowly and leveled off at 6.2- 6.5. Based on the discussion in Section 3.2, chemical precipitation in the form of Cd(OH)2 will occur at when pH 8.0-8.5 is exceeded thus masking the sorption process. The reason for this sharp rise in pH during the initial sorp- tion stage is explained in the light of the kinetics of ion exchange occurring in the bed. Both the sodium release and cadmium uptake occur initially at high rates in the upper fresh zone of the bed where the driving forces for these processes are high. At later stages down the bed, the driving forces of these processes will decrease and solu- tion pH is levelled off. It is recommended, therefore, that pH inside zeolite beds should also be maintained below 8.0 to avoid deposition of precipitate on the surface of par- ticles. Such deposition will make it difficult and impracti- cal to regenerate the zeolite bed for further use. 3.5 Mechanism of Cd Uptake Based on the unique structure of zeolite discussed ear- lier which is characterized by being both crystalline and porous, the removal of metal ions can be described as ion exchange, sieve action or both depending on the relative ionic sizes and pore diameter. The average pore diameter of the investigated size cuts of zeolite is 1.75 Å (Table 1) which indicates a macroporous structure. According to the IUPAC standard, pores larger than 50 nm (0.5 Å) are con- sidered as macropores. The radii of the crystal, Cd2+, and hydrated, Cd(OH)+, ions are 0.97 and 5.0 Å (Pauling, 1970). The following sequence compares ionic sizes to average zeolite pore size: Ionic radius of Cd2+ < Zeolite average pore size < Ionic radius of Cd(OH)+ This comparison reveals that it is possible for Cd2+ ions, the major cadmium species present at the experimen- tal conditions (pH < 8.0) to diffuse inside the pores of zeo- lite in addition to their sorption on the external surface. However, the hydrated ion, Cd(OH)+, is too large to pen- etrate into the pores and thus its removal is restricted to ion exchange at the external surface. This means that two mechanisms of uptake are plausible: (a) ion exchange process which may occur at the external surface or at the active sites inside the large pores which may be represent- ed by the following chemical equation: Cd2+(aq) + 2 Na+ (Z) Cd2+ (Z) + 2 Na+(aq) (6) (b) sieve action followed by physical adsorption on the internal surface of zeolite pores. There are similar obser- vations of these phenomena reported in the literature (Lee and Moon, 2001; Sprynskyy, et al. 2006). 4. Conclusions Jordanian phillipsite tuff is a cost-effective material for the removal of cadmium ions from aqueous solutions. The study has shown that zeolite dose (slurry concentration) and initial cadmium ion concentration are key factors influencing equilibrium uptake.Slurry pH increased slightly with sorption progress. Equilibrium uptake is best modeled by Langmuir model implying monolayer cover- 0 50 100 150 200 250 300 350 400 450 Time (min) 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 C / C o 4.0 5.5 Initial Solution pH: Fix e d Conditions : Be d De pth: 14.4 cm Flow Ra te : 15 m L/m in Co = 50 m g/L Pa rticle Size : 500-710 mµ Figure 10. Breakthrough curves at different initial solution pH values 0 50 100 150 200 250 300 350 400 450 Time (min) 4.0 4.5 5.0 5.5 6.0 6.5 7.0 7.5 8.0 8.5 9.0 9.5 10.0 E fflu e n t p H Fixed Conditions : Bed Depth: 14.4 cm Flow Rate: 15 mL/min Co = 50 mg/L Particle Size: 500- 710 mµ Initial Solution pH: 4.0 5.5 Figure 11. Changes in effluent solution pH with time in zeolite fixed-bed C / C o Time (min) E ff ic ie nt p H Time (min) 10 The Journal of Engineering Research Vol. 6, No. 2 (2009) 1-11 age at specific charged sites. Breakthrough curves obtained demonstrated the applicability of the tested zeo- lite mineral in continuous fixed bed column operation. The combination of bed height and solution flow rate are the two parameters to adjust for enhanced bed efficiency at a given cadmium ion concentration in the feed. Solution chemistry at pH < 8.0 as well as pore diameter of zeolite supports cadmium uptake mechanism by both ions exchange and sieve action. References Al-Haj-Ali, A. and El-Bishtawi, R., 1997, "Removal of Metal Ions Using Jordanian Zeolite Tuff", J. of Chem. Technol. Biotechnol., Vol. 69, pp. 27-34. Al-Haj-Ali, A. and Al-Hunaidi, T., 2004, "Breakthrough Curves and Column Design Parameters for Sorption of Lead Ions by Natural Zeolite", Environmental Technology, Vol. 25, pp. 1009-1019. Alloway, B.J., 1995, "Heavy Metals in Soil", Wiley, New York. Alvarez-Ayuso, E.; Garcia-Sanchez, A. and Querol, X., 2003, "Purification of Metal Electroplating Wastewaters Using Zeolites", Water Research Vol. 37, pp. 4855-4862. Berber-Mendoza, M.S., Leyva-Ramos, R., Alonso- Davilla, P., Mendoza-Barron, J. and Diaz-Flores, P.E., 2006, "Effect of pH and Temperature on the Ion- Exchange Isotherm of Cd(II) and Pb(II) on Clinoptilolite", J. of Chem. Technol. Biotechnol., Vol. 81, pp. 966-973. Burgess, J., 1988, "Ions in Solution: Basic Principles of Chemical Interactions", Ellis Horwood, London, UK. Camilo, C.; Carmen, G. and Paula, M., 2005, "Sorption Characteristics of Heavy Metal Ions by a Natural Zeolite", J. of Chem. Technol. Biotechnol. Vol. 80(4), pp. 477-481 Cincotti, A., Mameli, A., Locci, A. M., Orru, R. and Cao, G., 2006, "Heavy metal Uptake by Sardinian Natural Zeolites: Experiment and Modeling", Industrial & Eng. Chem. Research Vol. 45, pp. 1074-1084. Dabrowski, A., Hubicki, Z., Podkoscielny, P. and Robens, E., 2004, "Selective Removal of the Heavy Metal Ions from Waters and Industrial Wastewaters by Ion- Exchange Method", Chemosphere, Vol. 56, pp. 91- 106. Dwairi, I. M., 1992, "Evaluation of Jordanian Zeolite Tuffs for Decontamination and Immobilization of Cs", DIRASAT, Vol. 19B, pp. 61-73. Friberg, L.M., Piscaor, G.F. and Nordberg, T., 1974, "Cadmium in the Environment", Second edition, CRC Press, Cleveland, US. Garcia-Sanchez, A., Alvarez-Ayuso, E. and Jimenez de Blas, O., 1999, "Sorption of Heavy Metals from Industrial Wastewater by Low-Cost Mineral Silicates", Clay Minerals, Vol. 34, pp. 469-477. Genc-Fuhrman, H., Mikkelson, P. and Ledin, A., 2007, "Simultaneous Removal of As, Cd, Cr, Cu, Ni and Zn from Stormwater: Experimental Comparison of 11Different Sorbents", Water Research, Vol. 41, pp. 591-602. Hedstrom, A., 2001, "Ion Exchange of Ammonium in Zeolites: A Literature Review", J. of Environ. Eng., Vol. 127(8), pp. 673-681. Inglezakis, V.J. and Grigoropoulou, H., 2004,"Effects of Operating Conditions on the Removal of Heavy Metals by Zeolite in Fixed Bed Reactors", J. of Hazardous Materials, Vol. 112(1-2), pp. 37-43. Kocaoba, S.; Orhan, Y. and Akyuz, T., 2007, "Kinetics and Equilibrium Studies of Heavy Metal Ions Removal by Use of Natural Zeolite", Desalination, Vol. 214, pp. 1- 10. Lee, D.H. and Moon, H., 2001, "Adsorption Equilibrium of Heavy Metals on Natural Zeolites", Korean J. of Chem. Eng., Vol. 18(2), pp. 247-256. Ouki, K.S., Cheeseman, C.R. and Perry, R., 1994, "Natural Zeolite Utilization in Pollution Control: A Review of Applications to Metals' Effluents", J. of Chem. Technol. Biotechnol., Vol. 59(2), pp. 121-126. Pauling, L., 1970,"General Chemistry", W.H. Freeman & Co., New York. Rangel-Mendez, J.R. and Streat, M., 2002, "Adsorption of Cadmium by Activated Carbon Cloth: Influence of Surface Oxidation and Solution pH", Water Researech, Vol. 36, pp. 1244-1252. Rashdan, Z F., 2000, "Investigation of Natural Zeolitic Tuffs on Their Ability for Sewage Cleaning Purposes", PhD. Thesis, Dept. of Chemistry, Oldenburg University, Oldenburg, Germany, pp. 102- 107. Sand, L.B. and Mumpton, F.A., 1978, "Natural Zeolites: Occurance, Properties, Use", Pergamon Press, Oxford, UK. Singh, S.P., Ma, L.Q. and Harris, W.G., 2001," Heavy Metal Interactions with Phosphatic Caly: Sorption and Desorption Behavior", J. of Environmental Quality, Vol. 30, pp. 1968-1975. Sips, R., 1948, "On the Structure of a Catalyst Surface", J. of Chemical Physics, Vol.16, p.490. Sprynskyy, M., Buszewski, B., Terzyk, A. and Namiesnik, J., 2006, "Study of the Selection Mechanism of Heavy Metal (Pb2+, Cu2+, Ni2+ and Cd2+) Adsorption on Clinoptilolite", J. of Colloid & Interface Sci., Vol. 304, pp. 21-28. Srivastava, P., Singh, B. and Angrove, M.J., 2004, "Competitive Adsorption of Cadmium (II) onto Kaolinite as Affected by pH", 3rd Australian New Zealand Soils Conference, University of Sydney, Australia, 5-9 December, 2004. Stumm, W. and Morgan, J.J., 1996, "Aquatic Chemistry: Chemical Equilibria and Rates in Natural Waters", Third edition, Wiley, New York. Takac, S., Calik, G., Aytar, M. and T. Ozdamar, T., 1998, "Separation Kinetics of 1-Phenyl-Amine by Ion- Exchange Process", Biochem. Eng. J., Vol. 2, pp. 101-112. 11 The Journal of Engineering Research Vol. 6, No. 2 (2009) 1-11 Vaca-Mier, M., Callejas, R.L., Gehr, R., Cisneros, E.E.J. and Alvarez, P.J., 2001, "Heavy Metal Removal with Mexican Clinoptilolite: Multi-Component Ionic Exchange", Water Research, Vol. 35(2), pp. 373-378. WHO (World Health Organization), 2004, "Guidelines for Drinking Water Quality", Third Edition, WHO, Geneva.