The effect of surface modification of origin, base and acid/base treated activated carbons with ethyl xanthate on Zn(II) adsorption Ali Behnamfarda, Mohammad Mehdi Salarirada*, Francesco Vegliob a
Department of Mining and Metallurgical Engineering, Amirkabir University of Technology (Tehran Polytechnic), Tehran, Iran b Department of Chemistry, Chemical Engineering and Materials, University of L'Aquila, 67040 Monteluco di Roio, L'Aquila, Italy
Abstract The effect of surface modification of origin, base and acid/base treated activated carbons with potassium ethyl xanthate (PEX) on Zn(II) adsorption was investigated. Characterization of the activated carbon samples was performed by using scanning electron microscopy, N2 adsorption and desorption isotherms, pore size distribution, surface area determination, determination of point of zero charge and Boehm titration. The loading capacity of origin activated carbon (AC0) at initial Zn(II) concentration of 110mg/L after 24 hours aging in distilled water was 1.56mg/g and it increased by increasing the PEX concentration in the aging solution. But it never exceeded from the loading capacity of AC0 without aging which is 3.61mg/g. At Zn(II) initial concentration of 110mg/L, the loading capacity of AC0 and 4M nitric acid oxidized AC0 after 24 hours aging both of them in 0.1N NaOH solution were 7.18 and 15.03mg/g, respectively. They increased by increasing PEX concentration in the aging solution and ultimately reached to 8.05mg/g for base treated AC0 and to 16.43mg/g for acid/base treated AC0. The initial rate of Zn(II) adsorption decreased by increasing the PEX concentration in the aging solution of base and also acid/base treated AC0. The effect of Zn(II) initial concentration in the range of 59 to 165mg/L on its adsorption kinetics and loading capacity onto 4M nitric acid oxidized AC0 and then aging in 0.1N NaOH and 661mg/L PEX solution (ACABPEX) was investigated. The equilibrium and kinetic data were best represented by the Koble-Corrigan isotherm model and the pseudo second order kinetic model, respectively. It was understood that three mechanisms play the main role in the adsorption of Zn(II) onto ACABPEX. Keywords: Activated carbon; ethyl xanthate; surface modification; Zn(II) adsorption kinetics; loading capacity.
*
Corresponding author. Tel.: +98 2164542974; Fax: +98 2166405846.
E-mail address:
[email protected] (M.M. Salarirad)
1. Introduction Activated carbon (AC) is a well-known commercial material that is effectively used on an enormous scale in gas purification, wastewater treatment, drinking water purification, metal extraction, medicine and many other applications (Bansel and Goyal, 2005). AC is well suited to these tasks due to the very high surface area in the range of 800 to 1500m2/g, welldeveloped microporous structure and also the presence of various surface functional groups (Bansel and Goyal, 2005). While the effectiveness of AC to act as an adsorbent for adsorption of a variety of species from wastewater and drinking water is well noted, there has been an essential need for increasing the adsorption capacity of AC for heavy metals (Yin et al., 2007; Rivera-Utrilla et al., 2011). This is as a result of the increase in emission a higher amount of heavy metals with rapid development of industries like metal plating and finishing facilities (Fu and Wang, 2011). Also, the loading capacity of origin activated carbon for heavy metal isn't high with respect to the organic contaminants (Mugisidi et al., 2007). In order to enhance the adsorption capacity of AC, various factors can be tailored to increase its affinities toward heavy metals (Yin et al., 2007; Rivera-Utrilla et al., 2011). These factors include not only the specific surface area and pore size distribution, but also its surface chemistry (Yin et al., 2007; Rivera-Utrilla et al., 2011). The surface chemistry of AC can be changed by treating it with oxidizing agents either in gas phase or in aqueous solution or through impregnating foreign materials such as surfactants (Yin et al., 2007; Ahn et al., 2009; Rivera-Utrilla et al., 2011). Surfactants are hetero-polar organic molecules with a hydrophobic tail and a non-polar hydrophilic head, and have been applied to modify the surface properties of a solid surface (Rosen, 2004). According to the ability of surfactants to dissociate in water, they can be divided into distinct groups of nonionizing and ionizing
surfactants. Ionizing surfactants assumes the character of a cation or anion depending on the resulting charge (Rosen, 2004). Choi et al. (2009) studied the effect of pretreatment of AC with two cationic surfactant namely hexadecyltrimethylammonium bromide and cetylpridinium chloride on the adsorption of Cr(VI). They observed that the adsorption capacity of the modified ACs with the cationic surfactants is 3–5 times higher than that of the unmodified AC. They reported that this is as a result of increasing the number of positively charged adsorption sites on the surface of cationic surfactant modified AC compared to unmodified AC. Monser and Adhoum (2002) reported the results of fixed bed column studies on the adsorption of Cu(II), Zn(II), Cr(VI) ions onto the AC modified with anionic surfactant namely sodium diethyl dithiocarbamate (SDDC). It was found that the SDDC-modified AC had an effective removal capacity for Cu(II) (four times), Zn(II) (four times) and Cr(VI) (two times) greater than unmodified AC. They reported that the surfactant is adsorbed on the hydrophobic part of AC surface from nonpolar head while the ionic head will rest to solution for ion exchange with heavy metal cations. Ethyl Dithiocarbonate or ethyl xanthate is an anionic surfactant which widely used in the froth flotation of metal sulfide ores as a collecting agent and the hydrocarbon chain has two carbons (Fuerstenau et al., 1985; Bulatovic, 2007). In our previous investigations, we studied the kinetics and loading capacity of gold cyanide complexes (i.e., Au(CN)2-) onto AC which had been preload with different concentrations of ethyl xanthate (Salarirad and Behnamfard, 2010 and 2011). We observed that the adsorption kinetics of gold cyanide complex decreases significantly with increasing the preloading concentration of ethyl xanthate onto AC, while it has not significant effect on the loading capacity. The polar head of xanthate ions has a great affinity for heavy metal cations (Fuerstenau et al., 1985). This suggests that the modification of AC with ethyl xanthate may increase the adsorption of metal cations. The focus of this
research is to investigate the effect of surface modification of origin, acid and base treated AC with ethyl xanthate on the kinetics and loading capacity of Zn(II). After finding the most suitable condition, modeling of the loading capacity and kinetic adsorption of Zn(II) was carried out by using different models through linear and nonlinear regression methods. Furthermore, the mechanism of Zn(II) adsorption onto ethyl xanthate modified AC was assessed.
2. Materials and methods 2.1. Reagents Analytical reagent grade nitric acid, hydrochloric acid, sulfuric acid, sodium hydroxide, zinc sulfate heptahydrate (ZnSO4.7H2O), NH4Cl, ammonia solution, Eriochrome Black T and standard solutions of ethylenediaminetetraacetic acid (EDTA) were obtained from Merck Co., Germany. Analytical grade potassium ethyl xanthate (PEX) was attained from New Brunswick Co., USA. A commercial grade coconut shell-based granular AC, produced through a steam activation process was provided from Haycarb Co., Srilanka. A size fraction of −2.36+2mm of as received ACs was separated, rinsed with deionized water several times to remove any fine powder, dried in warm air to a constant mass and named as AC0. 2.2. Determination of acidity and basicity of AC samples The basicity and acidity of AC samples were determined according to the Boehm titration method (Barton et al., 1997; Boehm, 2002). The basicity of the carbon materials was estimated through mixing 2g of each sample with 100mL of 0.1M HCl solution in a closed flask and maintaining the contact under agitation by using an orbital shaker apparatus at 200rpm for 48h at ambient temperature. Then the suspension was decanted and the remaining HCl in solution was determined by titration with a 0.1M NaOH solution. The total acidity of AC samples was determined by a similar procedure where a 0.1M NaOH solution was put in contact with the AC and the titration solution was 0.1M HCl.
2.3. SEM and WDX analysis High-resolution compositional maps of AC samples were obtained through imaging with backscattered electrons (BSE) by using a Philips XL30 scanning electron microscope. Information about morphology and surface topography of AC samples was obtained through imaging with secondary electrons (SE). WDX studies were carried out with a WDX 3PC, MICROSPEC Corp., USA. 2.4. Nitrogen adsorption and desorption measurements Textural characteristics including isotherm plot of adsorption and desorption of nitrogen gas, specific surface area and pore size distribution of the AC sample before and after surface modifications were determined from nitrogen adsorption at 77K using an automatic Micromeritics ASAP-2000 volumetric sorption analyzer. The AC samples were degassed at 200◦C in a vacuum condition prior to gas adsorption measurements. Nitrogen adsorption isotherms were measured over a relative pressure (p/p0) range from approximately 10−5 to 0.995. The surface areas of the AC samples were determined using Brunauer–Emmett–Teller (BET) and Langmuir methods (Bansel and Goyal, 2005). The Barrett–Joyner–Hanlenda (BJH) method (Barrett et al., 1951) was applied to calculate the pore size distributions of AC samples. 2.5. Point of zero charge (pHpzc) The pHpzc of the ACABPEX was determined by the solid addition method (Srivastava et al., 2008). To a series of 50ml flasks, 20mL of 0.01M KNO3 solution was transferred. The initial pH of the solutions was adjusted between 2 and 12 by adding 0.01N HNO3 or NaOH. The total volume of the solution in each flask was brought exactly to 25mL by adding the KNO3 solution and the initial pH of the solution was accurately recorded. 0.5g of ACABPEX was added to each flask, which were immediately capped. The suspensions were allowed to reach equilibrium for 48h with intermittent manual shaking and then the final pH of the supernatant
liquid was recorded. The difference between the initial and final pH values (∆pH) was plotted against the initial pH. The point of intersection of the resulting curve at which ∆pH=0 gave the pHpzc. This procedure was repeated for 0.1M KNO3 solution. 2.6. The effect of surface modification of AC0 with different concentrations of PEX on its Zn(II) adsorption ability 2g of AC0 was added to 500mL distilled water containing different concentrations of PEX in the range of 0 to 200mg/L and then agitated at ambient temperature by using a bottle roll apparatus at constant rotation speed of 100rpm. After 24 hours agitating, the AC0 were separated from solutions, rinsed with distilled water several times, and then they directly added to 500mL solution containing 110mg/L Zn(II). Sampling was performed by removing aliquots at predetermined time intervals and then analyzed for Zn(II) by the titration method (Vogel, 1989; Skoog et al., 2000). 10mL of the removed sample was buffered to pH=10 with ammonia/ammonium chloride solution and titrated directly with the standard EDTA solution in the presence of Eriochrome Black T as indicator. It must be mentioned that the accuracy and precision of the titration method was confirmed by the atomic absorption spectroscopy (UNICAM 939 AA spectrometer). The PEX concentration of the pretreatment solutions was determined by a UV-Visible spectrophotometer at the indicating wavelength of 301nm before adding AC0 and also at the end of experiments. It was observed that all of PEX in the range of 0 to 200mg/L adsorb onto AC0. 2.7. The effect of surface modification of base treated AC0 with different concentrations of PEX on its Zn(II) adsorption ability 2g of AC0 was added to 500mL 0.1N sodium hydroxide solution containing different concentrations of PEX in the range of 0 to 200mg/L and then agitated at ambient temperature by using a bottle roll apparatus at constant rotation speed of 100rpm. After 24 hours agitating,
the AC0 were separated from solutions, rinsed with distilled water several times until no pH change in the washed liquid could be detected. Afterwards, they were directly added to 500mL solution containing 112mg/L Zn(II). Sampling was performed by removing 10mL aliquots at predetermined time intervals and then analyzed for Zn(II) by the titration method. 2.8. The effect of surface modification of acid/base treated AC0 with different concentrations of PEX on its Zn(II) adsorption ability Initially 30g of AC0 was kept in contact with 250mL of 4M nitric acid solution in a closed flask and agitated for 24 hours by an orbital shaker at ambient temperature and 200rpm rotation speed. It was then separated from the solution, rinsed with distilled water several times until no pH change in the washed liquid could be detected and dried in warm air to a constant mass. Each 2g portion of nitric acid oxidized AC0 was transferred to 500mL 0.1N sodium hydroxide solutions containing different concentrations of PEX. The suspensions were agitated for 24h by a bottle roll apparatus at ambient temperature. They were then separated from the solution and rinsed with distilled water several times until no pH change in the washed liquid could be detected and then directly added to 500mL solution containing 112mg/L Zn(II). Sampling was performed by removing 10mL aliquots at predetermined time intervals and then analyzed for Zn(II) by the titration method. 2.9. Batch equilibrium experiments for the Zn(II) adsorption onto ACABPEX Each 2g portion of nitric acid oxidized AC0 was transferred to 500mL 0.1N sodium hydroxide solutions together with 661mg/L PEX. The suspensions were agitated for 24 hours by a bottle roll apparatus at ambient temperature. They were then separated from the solution and rinsed with distilled water several times until no pH change in the washed liquid could be detected and named as ACABPEX. They were directly added to 500mL solution containing different concentrations of Zn(II) in the range of 59 to 165mg/L and agitated by a bottle roll
apparatus at constant rotation speed of 100rpm and at ambient temperature for 24h to equilibrium was established. After equilibrium, the solution was analyzed for Zn(II) concentration in triplicate and averaged. The amount of Zn(II) adsorption at equilibrium, qe (mg/g), was calculated by: qe=(C0-Ce)V/W
(1)
where C0 and Ce (mg/L) are initial (t=0) and equilibrium Zn(II) concentrations, respectively. V is the volume of the solution (L) and W is the mass of dry carbon used (g). 2.10. Batch kinetic experiments for the Zn(II) adsorption onto ACABPEX The batch kinetic experiments were performed by a similar method as previously explained for the batch equilibrium experiments. Only, the sampling was performed at regular time intervals by removing 10mL aliquots and analyzed for Zn(II) concentration. The amount of Zn(II) adsorbed at time t, qt (mg/g), was calculated by: qt=(C0-Ct)V/W
(2)
where Ct (mg/L) is Zn(II) concentration at time t. 3. Results and discussion 3.1. Characterization of AC ACs generally have a strongly developed internal surface and are usually characterized by a complex tridimensional porous structure consisting of pores of different sizes and shapes. Pores of AC fall into three groups of the micropores with diameters less than 2 nm, mesopores or transitional pores with diameters between 2 and 50 nm, and macropores with diameters greater than 50 nm (Bansel and Goyal, 2005). Fig.1 shows the isotherms of nitrogen gas adsorption onto the AC0, 4M nitric acid oxidized AC0 and then 24 hours aging in 0.1M sodium hydroxide solution (ACAB) and also ACABPEX. A sharp increase of nitrogen adsorption is observed at relative pressures less than 0.1 for all three isotherms. This is due to the presence of micropores in the AC samples. When an adsorbent contains very
fine micropores that have pore dimensions only a few molecular diameters, the potential field of force from the neighboring walls of the pores will overlap causing an increase in the interaction energy between the adsorbent surface and the gas molecules, which results in sharp adsorption of gas at low relative pressures (Bansel and Goyal, 2005). The adsorption process continues into the smallest of the mesopores by increasing the relative pressure up to the 0.45. At higher relative pressures, the adsorption of nitrogen gas occurs into mesopores of the AC samples. During this stage, along with the formation of multimolecular layers, condensation of the gaseous molecules takes place which called capillary condensation (Bansel and Goyal, 2005). This type of adsorption isotherm resembles to the Type IV of the five main types of adsorption isotherms which proposed by IUPAC. On lowering the pressure of the nitrogen in the equipment, the equilibrium positions, that is the desorption isotherm, do not follow the adsorption isotherm to create a hysteresis loop. This is due to the fact that the filling of the mesopores involves a different mechanism from their emptying. The adsorptiondesorption hysteresis is a direct consequence of capillary condensation in pores of the AC samples. The hysteresis loops in different adsorbents exhibit several forms which have been classified by lUPAC (Thommes, 2010). The hysteresis loops in Fig.1 resemble to the TypeH4. H4 hysteresis loops are generally observed with complex materials containing both micropores and mesopores (Thommes, 2010). As can be seen in Fig.1, the main difference between three nitrogen adsorption isotherm models is the amount of nitrogen adsorption at relative pressures less than 0.1 and it obeys the following order: AC0>ACAB>ACABPEX. This indicates that, surface modification of AC0 results in decreasing the amount of microporosity. Please insert Fig.1 about here Figs.2.a and b show the pore size distribution of the AC samples obtained from the N2 adsorption and desorption isotherm curves, respectively. The pore size distribution curves
demonstrate that the distribution of pore size of the AC samples is quite narrow in the range of micro and mesopores. The pattern of the pore size distribution curve of the AC samples in desorption mode is nearly the same with the curve in adsorption mode, only there is a peak around 40Ao which is due to the specific shape of pores in the AC samples. It can be seen that there is not any difference between three pore size distribution curves for the pores in the range of macropores, while a slight decrease of mesopores is observed for ACABPEX and ACAB than AC0. Also, the amount of micropores for ACAB and ACABPEX is lower than AC0. Please insert Fig.2 about here A summary of pore analysis report for AC0, ACAB and ACABPEX is presented in Table 1. As can be seen total surface area of AC0 in terms of the BET and Langmuir surface areas and also single point surface area at P/Po 0.2057 is larger than ACAB and it is larger than ACABPEX. It can be also seen that the main part of this decrease is due to the decrease of micropore area. This is agreement with the results observed in Figs.1 and 2. Please insert Table 1 about here Fig.3 shows the plot of ∆pH versus initial pH value for the ACABPEX at two different KNO3 concentrations of 0.01 and 0.1N. It can be seen that the intersection point of both curves with the longitudinal axis is at pH value of 9.75. This indicates that pHPZC for ACABPEX is at pH of 9.75. Also, pHPZC for AC0 was determined to be at pH of 9.8 through the same procedure. This indicates that the proposed surface modification method has not significant effect on the point of zero charge of AC. Please insert Fig.3 about here
3.2. The effect of surface modification of AC0 with different concentrations of PEX on its Zn(II) adsorption ability Fig.4 shows the effect of preloading of PEX onto ACO on the kinetics and loading capacity of Zn(II). It can be seen that the maximum adsorption capacity (i.e., 3.613mg/g) is for AC0 without aging for 24 hours in distilled water. Aging of AC0 in distilled water for 24 hours has a negative effect on both of the kinetics and loading capacity of Zn(II), so that the loading capacity of AC0 decreases from 3.16 to 1.56mg/g. As shown in Fig.4, the presence of PEX in the aging solution improves both of the kinetics and loading capacity of Zn(II) onto AC0, so that the adsorption capacity for AC0 samples after 24 hours aging in solutions containing 25, 50 and 200mg/L PEX are 1.65, 1.90 and 2.42mg/g, respectively. This also showed that the adsorption capacity of AC0 increases by increasing the PEX concentration in the aging solutions. Please insert Fig.4 about here Fig.5.a shows the SEM photograph of AC0 surface after aging in a solution containing 200mg/L PEX and then putting it in a solution containing 110mg/L Zn(II). Fig.5.b, c and d show the map of distribution and relative proportion of carbon, sulfur and zinc element over the scanned area, respectively. Fig.5.b shows that the carbon element is distributed over the scanned area. This is true because it is found that in elemental composition of an AC typically there is more than 90% carbon element. Fig.5.c shows a significant amount of sulfur element is detected over the scanned area, while AC0 surface has a negligible sulfur element before PEX adsorption. This is as a result of PEX adsorption on the surface of AC0 because each molecule of PEX has two atom of sulfur and we used from sulfur detection in WDX analysis for showing the PEX adsorption on the surface of AC0. PEX is a heteropolar surface-active organic molecule, which consists of two parts including an anionic polar group and an uncharged non-polar hydrocarbon chain containing two carbons. When the PEX
molecules adsorbed from their hydrocarbon chain head onto the surface of AC as a result of hydrophobic interactions, the charged polar head orients onto the solution and creates new adsorption sites, which results in increasing adsorption capacity of AC. Fig.5.d shows the distribution of Zn element over the scanned area. It can be seen that Zn(II) ions adsorbed on the surface of AC0, but their distribution are not proportional with the distribution of sulfur element. This indicates that only some of the PEX molecules adsorbed on the surface of AC0 from polar head and create new adsorption sites. The polar head of PEX molecules has a great affinity for Zn(II) ions and zinc ethyl xanthate (ZnX2) is formed according to the following equation: Zn2+ + 2C2H5OCSS- → Zn(C2H5OCSS)2
(3)
The solubility product (Ksp) of zinc ethyl xanthate is 4.9 ×10-9. Please insert Fig.5 about here Although the presence of PEX in the aging solution increases the adsorption capacity of AC for Zn(II) but the extent of this increase is low, so that the Zn(II) adsorption capacity for AC0 after 200mg/L PEX adsorption is lower than the AC0 without aging (i.e., 2.42 versus 3.16mg/g). In the next step we will try to improve the adsorption ability of AC0 to a greater extent by adding the sodium hydroxide to the aging solution. 3.3. The effect of surface modification of base treated AC0 with different concentrations of PEX on its Zn(II) adsorption ability Fig.6 shows the effect of surface modification of AC0 through aging in a solution containing 0.1N sodium hydroxide and different concentrations of PEX on the adsorption capacity of Zn(II) onto AC. It must be said that the aging of AC0 in a 0.1N sodium hydroxide solution for 24h significantly increases the Zn(II) adsorption capacity from 3.16 to 7.18mg/g. A comprehensive research about the surface modification of AC0 with sodium hydroxide has been made and the results published elsewhere.
It can be seen from Fig.6 that the presence of PEX together with sodium hydroxide in the aging solution improves the loading capacity of AC0 for Zn(II). A sharp increase of loading capacity of AC0 for Zn(II) is observed after addition of 25mg/L PEX onto the aging solution. Afterwards, it gradually increases and reaches to 8.05 mg/g by increasing the PEX concentration in aging solution to 500 mg/L and then it nearly remains constant by the increasing the PEX concentration. As previously mentioned in previous section, increasing of the loading capacity of AC0 after aging for 24 hours in a solution containing 0.1N sodium hydroxide and PEX is due to the creation of new adsorption sites on AC surface as a result of PEX preloading. Please insert Fig.6 about here Fig.7.a shows the Zn(II) adsorption kinetics onto AC0 after 24 hours aging in 0.1N sodium hydroxide solution together with different concentrations of PEX in the range of 0-600mg/L. It can be seen that at all of the PEX concentrations, the initial rate of Zn(II) adsorption is fast and slows down afterwards. In order to observe the effect of PEX preloading on Zn(II) adsorption kinetics more clearly, Fig.7.b was drown. This figure shows the difference between the qt values for AC0 after aging in 0.1N sodium hydroxide solution together with different concentrations of PEX and AC0 after aging only in 0.1N sodium hydroxide solution. It can be seen that the preloading of PEX decreases the initial rate of Zn(II) adsorption and this decrease is much more by the increasing of PEX preloading. For example, the qt value at t=5.5h for AC0 with 24h aging in 0.1N sodium hydroxide solution is 5.7mg/g, while this value for AC0s with 24h aging in 0.1N sodium hydroxide solution together with 100 and 600mg/L PEX is 5.39 and 4.80mg/g, respectively. The difference between qt values is compensated at a given contact time and afterwards, the Zn(II) adsorption kinetics onto AC with PEX preloading surpasses than the AC without it. The time required for compensating
this difference is dependent on the PEX preloading concentration and it increases by the increasing the PEX preloading concentration. Please insert Fig.7 about here 3.4. The effect of surface modification of acid/base treated AC0 with different concentrations of PEX on its Zn(II) adsorption ability Fig.8 shows the effect of aging of nitric acid oxidized AC0 in 0.1N sodium hydroxide solution together with different concentrations of PEX in the range of 0 to1000mg/L. It must be said that oxidizing of activated carbon with nitric acid and then its modification with sodium hydroxide significantly improves the Zn(II) loading capacity. For example, at initial Zn(II) concentration of 112mg/L, the loading capacity of AC0 after modification with sodium hydroxide is 7.18mg/g, while this value for nitric acid oxidized AC0 after modification with sodium hydroxide is 15.03mg/g. This increase is mainly due to the increasing of acidic surface functional groups of AC0 as previously reported. As shown in Fig.8 the Zn(II) adsorption capacity of nitric acid oxidized AC0 increases by adding PEX into the aging solution. This increase initially is sharp so that by adding 100mg/L PEX into the aging solution, the loading capacity of nitric acid oxidized AC0 increases from 15.03 to 15.74mg/g. Afterwards, by increasing the PEX concentration from 100 to 600mg/L, the loading capacity increases from 15.74 to 16.32mg/g. Increasing the PEX concentration more than 600mg/L has no significant effect on the loading capacity, so that it is 16.43mg/g at PEX concentration of 1000mg/L. Increasing the loading capacity of nitric acid oxidized AC0 through aging in PEX containing solution is due to the creation of new adsorption sites for Zn(II) as a result of adsorption of PEX molecules through hydrophobic interactions with AC surface. Please insert Fig.8 about here
Fig.9.a shows the kinetics of Zn(II) adsorption onto nitric acid oxidized AC0 after 24 hours aging in 0.1N sodium hydroxide solution together with different concentrations of PEX in the range of 0 to 1000mg/L. It can be seen that adsorption kinetics of Zn(II) onto nitric acid oxidized AC0 after aging with sodium hydroxide and PEX is fast up to 10 hours contact time and then slows down to reach equilibrium. Fig.9.b was drown in order to better understand the effect of PEX addition on the aging solution on the Zn(II) adsorption kinetics. This figure shows the difference between qt values for nitric acid oxidized AC0 after aging in 0.1N sodium hydroxide solution together with and without PEX. It can be seen that the presence of PEX in the aging solution reduces the initial rate of Zn(II) adsorption and this decrease becomes greater at higher PEX concentrations. However, it is compensated at a given contact time based on the PEX concentration in the aging solution and after that it outruns. This can be explained as follows. Adsorption of some PEX molecules from polar head on the surface of AC due to the electrostatic interactions and/or laying of some PEX molecules on the AC surface as a result of hydrophobic interactions increase the surface hydrophobicity of AC. Increasing the surface hydrophobicity is the main factor in reducing the initial rate of Zn(II) adsorption onto the nitric acid oxidized AC0 after modification with sodium hydroxide and PEX with respect to the nitric acid oxidized AC0 after modification only with sodium hydroxide. However, some of the PEX molecules adsorbed onto the surface of AC from nonpolar hydrocarbon chain head through hydrophobic interactions and create new adsorption sites for Zn(II) ions which results in compensating the reduced adsorption kinetics and increasing the loading capacity. Please insert Fig.9 about here
3.5. The effect of Zn(II) concentration on its adsorption kinetics and loading capacity onto ACABPEX Fig.10 shows the effect of Zn(II) initial concentration on its loading capacity onto ACABPEX. It can be seen that the loading capacity of Zn(II) onto ACABPEX increases with increasing the initial concentration. The increase of loading capacity by increasing the initial Zn(II) concentration initially is significant, while it is negligible at higher initial concentrations. For example, the loading capacity of Zn(II) onto ACABPEX increases from 12.17 to 16.24mg/g by increasing the Zn(II) initial concentration from 59 to 108mg/L, while it increases only from 16.46 to 16.96 by increasing the initial concentration from 118 to 165mg/L. Please insert Fig.10 about here Fig.11 shows the kinetics of Zn(II) adsorption onto ACABPEX at different initial concentration in the range of 59 to 165mg/L. It can be seen that the rate of Zn(II) adsorption onto ACABPEX initially is fast to around 10h contact time and then slows down to reach equilibrium after about 24h. The kinetics of Zn(II) adsorption onto ACABPEX increases with increasing initial Zn(II) concentration. For example, the qt value at t=10h increases from 12.03 to 13.94mg/g by increasing the initial Zn(II) concentration from 59 to 165mg/L. Please insert Fig.11 about here Fig.12 shows the effect of initial Zn(II) concentration on its removal percentage onto ACABPX. The Zn(II) percent removal was calculated using the following equations: % Re moval = (1 −
Ct ) × 100 C0
(15)
It can be seen that the removal percentage decreases with increasing Zn(II) initial concentration. In other words, a higher amount of Zn(II) ions remain in the solution with increasing the initial Zn(II) concentration. For example, the removal percentage decreases from 85.31 to 43.55 by increasing the initial concentration from 69 to 165mg/L.
Please insert Fig.12 about here 3.6. Adsorption mechanism of Zn(II) adsorption Fig.13 shows the loading capacity of AC0, ACB, ACAB and ACABPEX for the adsorption of Zn(II) at different initial concentrations. It can be seen that the loading capacity of ACB is about twice than the AC0. This can be explained that after surface modification of AC0 with sodium hydroxide, the local pH of AC0 surface becomes suitable for precipitation of some of the Zn(II) ions as zinc hydroxide. Fig.13 also shows that the loading capacity of ACAB for Zn(II) ions is more than twice larger than ACB. This shows that oxidizing of AC0 with nitric acid and then its modification with sodium hydroxide results in increasing of the loading capacity significantly. This increase is as a result of increasing the acidic surface functional groups of AC0 after oxidizing with nitric acid. Because, the acidic surface functional groups of AC0 increases from 0.3 to 1.2mmol/g after oxidizing with nitric acids and they play a main role in adsorption of Zn(II) ions through a cation exchange mechanism. Fig.13 also shows that the loading capacity of ACABPEX is more than that of ACAB. In order to find the adsorption mechanism of Zn(II) onto ACABPEX some SEM and WDX analysis were carried out. Please insert Fig.13 about here Fig.14.a show the external surface of a ACABPEX particle after loading of Zn(II) in SE image mode. A smooth surface is observed in this image. Fig.14.b show the BSE image of the external surface of a ACABPEX particle after loading of Zn(II). In this image two distinct phases can be seen, so that a more bright BSE intensity is correlates with greater average atomic number. In order to identification of the elemental composition of ACABPEX surface after loading of Zn(II) ions, the map of distribution of carbon, sulfur and zinc elements is observed in Figs.14.c,d and e, respectively. It can be seen that the main element of the phase with more bright intensity in BSE image is zinc and the main element of the phase with dark
intensity is carbon. This show that one of the mechanism of Zn(II) adsorption onto ACABPEX is local precipitation of zinc hydroxide. Also, a significant amount of sulfur element is distributed over the scanned area evenly and no significant proportion is observed between distribution of sulfur and zinc elements. This shows a significant amount of PEX molecules are present on the surface of ACABPEX and some of them which adsorbed from nonpolar head can create adsorption sites for Zn(II) ions. Fig.14.f shows new cross sections of two broken ACABPEX particles after loading of Zn(II) ions. It can be seen that the phase with more bright intensity is observed only on the external surface of ACABPEX particles. In other words, zinc hydroxide precipitate has been formed on the external surface of ACABPEX. In order to more accurate observation of this hypothesis, SEM and WDX analysis of the marked area on the Fig.14.f were carried out and the results is illustrated in Fig.15. Please insert Fig.14 about here Fig.15.a shows the BSE image of the cross section of ACABPEX after loading of Zn(II). The more bright intensity in this figure which is correlated with zinc hydroxide precipitate is observed only on the external surface and also edge of ACABPEX particle. Fig.15.b. shows the map of distribution of carbon element over the scanned area. It can be seen that the distribution of carbon element inside the carbon particle is very high, but it decreases on external surface of carbon particle. Fig.15.c shows the map of distribution of sulfur element over the scanned area. It can be seen that the distribution of sulfur element decreases by increasing the distance from edge towards more interior parts of the carbon particle. This shows that the PEX molecules have been penetrated onto more interior parts of the carbon particle. Fig.15.d illustrates the map of distribution of zinc element over the scanned area. It can be seen that the distribution of zinc element on the external surface of the carbon particle
is high, especially at the regions with more bright intensity in Fig.15.a. Also, the distribution of zinc element decreases from edge towards more interior parts of the carbon particle. Please insert Fig.15 about here Based on the above discussion it can be concluded that three mechanisms play main role in the adsorption of Zn(II) onto ACABPEX : 1. Local precipitation of Zn(II) as zinc hydroxide on the surface of ACABPEX 2. Adsorption of Zn(II) ions by acidic surface functional groups of ACABPEX through cation exchange mechanism 3. Adsorption of Zn(II) ions by polar head of PEX molecules. 3.7. Modeling of equilibrium data for the adsorption of Zn(II) onto ACABPEX Modeling of equilibrium data is of great importance in finding an equation that can be used to compare different adsorbents under different operational conditions and to design and optimize an adsorption system. To examine the relationship between adsorbed amount and aqueous concentration of adsorbate at equilibrium, various sorption isotherm models are widely employed for fitting the data. In this study, two-parameter isotherm models including Froundlich, Langmuir and Temkin and three-parameter isotherm models including RedlichPeterson (RP) and Koble-Corrigan (KC) were assessed through linear and nonlinear regression methods for finding the most appropriate isotherm model for adsorption of Zn(II) onto ACABPEX. In both linear and non-linear regression methods, the average relative error (ARE) was applied to test the best-fitting isotherm model to the experimental data. The ARE values can be calculated by the following equation:
ARE =
100 N (q e exp − q e cal ) ∑ q exp N i =1 e i
(16)
where qeexp and qecal (mg/g) are experimental and calculated amount of Zn(II) adsorbed onto the ACABPEX at time t and N is the number of measurements made. The smaller ARE value indicates more accurate estimation of qe values. Table 2 shows the equations of different two- and three-parameter isotherm models and their linear form. It can be also seen that Langmuir isotherm model has four different linear forms. In linear regression method, the parameters of isotherm models can be determined from slope and intercept of linear plots, as can be seen in Table 2. Please insert Table 2 about here
Table 3 shows R2 and ARE values and also the parameters of different two- and threeparameter isotherm models. The R2 value for Langmuir Type(1) and Redlich-Peterson isotherm models has the closet value to unity among different isotherm models. This shows that the best linearity of the data is possible by using these two isotherm model. However, the ARE value for Langmuir Type(IV), RP and KC isotherm models is the lowest value among different isotherm models. This indicates that the best predication of experimental qe values is possible by using these three isotherm models. Please insert Table 3 about here
Table 4 shows the ARE value and also the parameters of different isotherm models for adsorption of Zn(II) onto ACABPEX which determined by using non-linear regression method. It can be seen that for all of the isotherm models the ARE value which obtained by non-linear regression method is lower than that of the linear regression method. This suggests that the non-linear regression method is a better way to obtain the isotherm parameters than the linear regression method. As can be seen in Table 4, the applicability of different isotherm models for predication of experimental qe values is in the order of KC ≥ Langmuir ≥ RP >> Froundlich > Temkin. Fig.16 schematically shows the applicability of different isotherm models for prediction of experimental qe values by using non-linear regression
method. This figure also shows the above mentioned order is valid for the applicability of different isotherm models for predication of experimental qe values. Please insert Table 4 about here Please insert Fig.16 about here 3.8. Modeling of the kinetic data for the adsorption of Zn(II) onto ACABPEX
Adsorption kinetic study provides valuable information on the reaction pathway and in the mechanism of adsorption reactions. Hence, it is of prime importance in designing of an adsorption system. As previously mentioned, in this research the kinetics of Zn(II) adsorption onto ACABPEX was studied at different initial concentrations in the range of 59 to 165mg/L. The kinetic data was then analyzed by using different kinetic models including Elovich, Fractional power (FP), pseudo first order (PFO) and pseudo second order (PSO) kinetics models through linear and non-linear regression methods. Table 5 shows the equations of different kinetic models. In linear regression method, the parameters of different kinetic models can be determined from slope and intercept of linear plots. Table 5 shows the linear form and parameters of different kinetic models. As can be seen in this table, PSO kinetic model has four different linear forms. Please insert Table 5 about here
Table 6 shows the R2 and ARE values and also parameters of different kinetic models at different initial concentrations in the range of 59-165mg/L which obtained by the linear regression method. It can be seen that the R2 value for PSO Type(I) is more near to unity among different kinetic models and their linear forms. Also, the ARE value for PSO Type(I) is the least value among the other models. This suggests that the kinetics of Zn(II) adsorption onto ACABPEX is best represented by the PSO kinetic model. Also, among four different linear forms of PSO kinetic model, Type(I) is the most linear form for predication of kinetics
of Zn(II) adsorption onto ACABPEX at different initial concentrations in the range of 59165mg/L. Please insert Table 6 about here
Table 7 shows the ARE value and the parameters of Elovich, FP, PFO and PSO kinetics models for the adsorption of Zn(II) onto ACABPEX at different initial concentrations in the range of 59-165mg/L which determined by the non-linear regression method. The results presented in this table also indicates that PSO kinetic model is the best kinetic model for representation of Zn(II) adsorption kinetics onto ACABPEX. It is obvious that the ARE value for each kinetic model and at each specific initial concentration which obtained by the non-linear regression method is less than that of the linear regression method. This indicates that the non-linear regression method is a better way than linear regression method for determination of parameters of kinetic models. Please insert Table 7 about here
Fig.17.a, b, c and d illustrate the applicability of different kinetic models for prediction of experimental qt values at Zn(II) initial concentration of 69, 98, 136 and 165mg/L, respectively. These figures clearly show that the PSO kinetic model is able to predicate the qt values for Zn(II) adsorption onto ACABPEX more accurate than the other models at different initial concentrations. Please insert Fig.17 about here Conclusion
This research showed that loading capacity of origin, base and acid/base treated activated carbons for Zn(II) increases by increasing the PEX concentration in the aging solution. This is due to the creation of new adsorption sites as a result of adsorption of PEX molecules through hydrophobic interactions with the surface of activated carbons. The presence of PEX molecules in the aging solution of base and acid/base treated AC0 results in the decreasing
initial rate of Zn(II) adsorption, since the surface hydrophobicity of activated carbons increases which results in the decreased mass transfer rates between the bulk solution and the adsorbent phase. The isotherm and kinetic studies on the Zn(II) onto ACABPEX were conducted at different initial concentrations in the range of 59 to 165mg/L. It was shown that both of the loading capacity and kinetics of Zn(II) adsorption onto ACABPEX increases by increasing the initial concentrations. Modeling of the equilibrium and kinetic data showed that the nonlinear regression method is a better way to obtain the model parameters than the linear regression method. The equilibrium data for the adsorption of Zn(II) on to the ACABPEX best represented by the KC isotherm model. Also, Langmuir and RP isotherm models are able to model the equilibrium data by a good precision. The PSO equation was best fitted to the kinetic data for the adsorption of Zn(II) onto ACABPEX at different initial concentrations. Three mechanism including ion exchange with acidic surface functional groups, local precipitation of zinc hydroxide and adsorption on the new created adsorption sites as a result of PEX adsorption through hydrophobic interaction play a main role in the adsorption of Zn(II) onto ACABPEX. References Aharoni, C., Tompkins, F.C., 1970. Kinetics of adsorption and desorption and the Elovich equation, In: Eley, D.D., Pines, H., Weisz, P.B. (Eds.), Advances in Catalysis and Related Subjects, vol. 21, Academic Press, New York, pp. 1–49. Ahn, C.K., Kim, Y.M., Woo, S.H., Park, J.M., 2009. Removal of cadmium using acid-treated activated carbon in the presence of nonionic and/or anionic surfactants, Hydrometallurgy 99, 209-213. Bansal, R.C., Goyal, M., 2005. Activated Carbon Adsorption. 1st ed. Taylor and Francis Group, New York, USA. Barrett, E. P., Joyner, L. G., Halenda, P. P., 1951. The determination of pore volume and area distributions in porous substances. I. Computations from nitrogen isotherms, J. Am. Chem. Soc. 73, 373–380.
Barton, S.S., Evans, M.J., Halliop, E., Mac Donald, J.A.F.,1997. Acidic and basic sites on the surface of porous carbon. Carbon 35, 1361–6. Boehm, H.P., 2002. Surface oxides on carbon and their analysis: a critical assessment. Carbon 40, 145–9. Bulatovic, S.M., 2007. Handbook of Flotation Reagents. 1st ed. Elsevier, Amsterdam. Choi, H. D., Jung, W. S., Cho, J. M., Ryu, B. G., Yang, J. S., Baek, K., 2009. Adsorption of Cr(VI) onto cationic surfactant-modified activated carbon, J. Hazard. Mater. 166, 642-646. Freundlich, H.M.F., 1906. Uber die adsorption in losungen, Z. Phys. Chem. 57(A), 385–470. Fu, F., Wang, Q., 2011. Removal of heavy metal ions from wastewaters: A review, J. Environ. Manage. 92, 407-418. Fuerstenau, M.C., Miller, J.D., Kuhn, M.C., 1985. Chemistry of Flotation. Society of Mining Engineers, New York, USA. Ho, Y.S., McKay, G., 1998. Sorption of dye from aqueous solution by peat, Chem. Eng. J. 70, 115–124. Ho, Y.S., McKay, G., 1999. Pseudo-second order model for sorption processes, Process Biochem., 34, 451–465. Ho, Y.S., McKay, G., 2000. The kinetics of sorption of divalent metal ions onto sphagnum moss peat, Water Res. 34, 735–742. Juang, R.S., Chen, M.L., 1997. Application of the Elovich equation to the kinetics of metal sorption with solvent-impregnated resins, Ind. Eng. Chem. Res. 36, 813–820. Koble, R.A., Corrigan, T.E., 1952. Adsorption isotherms for pure hydrocarbons, Ind. Eng. Chem., 44, 383–387. Lagergren, S., 1898. About the theory of so-called adsorption of soluble substances, K. Sven. Vetenskapsakad. Handl., 24(4), 1-39. Langmuir, I., 1917. The constitution and fundamental properties of solids and liquids. II. Liquids, J. Am. Chem. Soc., 39, 1848–1906. Langmuir, I., 1918. The adsorption of gases on plane surfaces of glass, mica and platinum, J. Am. Chem. Soc. 40, 1361–1403. Monser, L., Adhoum, N., 2002. Modified activated carbon for the removal of copper, zinc, chromium and cyanide from wastewater, Sep. Purif. Technol., 26, 137–146. Mugisidi, D., Ranaldo, A., Soedarsono, J. W., Hikam, M., 2007. Modification of activated carbon using sodium acetate and its regeneration using sodium hydroxide for the adsorption of copper from aqueous solution, Carbon 45, 1081-1084. Redlich, O., Peterson, D.L., 1959. A useful adsorption isotherm, J. Phys. Chem., 63(6), 1024.
Rivera-Utrilla, J., Sanchez-Polo, M., Gomez-Serrano, V., Alvarez, P.M., Alvim-Ferraz, M.C.M., Dias, J.M., 2011. Activated carbon modifications to enhance its water treatment applications. An overview, J. Hazard. Mater. 187, 1–23. Rosen, M.J., 2004. Surfactants and Interfacial Phenomenon, John Wiley and Sons, New York, USA. Salarirad, M.M., Behnamfard, A., 2010. The effect of flotation reagents on cyanidation, loading capacity and sorption kinetics of gold onto activated carbon, Hydrometallurgy 105, 47–53. Salarirad, M.M., Behnamfard, A., 2011. Fouling effect of different flotation and dewatering reagents on activated carbon and sorption kinetics of gold, Hydrometallurgy 109, 23–28. Skoog, D. A., West, D. M., Holler, F. J., Crouch, S. R., 2000. Analytical Chemistry: An Introduction, 7th ed., Saunders College Publishing, USA, pp. 345-381. Sparks, D.L., 1986. Kinetics of Reaction in Pure and Mixed systems, In: soil physical chemistry, CRC Press, Boca Raton, pp. 12-18. Srivastava, V. C., Mall, I. D., Mishra, I. M., 2008. Adsorption of toxic metal ions onto activated carbon Study of sorption behaviour through characterization and kinetics, Chem. Eng. Process. 47, 1269–1280. Temkin, M.J., Pyzhev, V., 1940. Recent modifications to Langmuir isotherms, Acta Physchim. USSR, 12, 217. Thommes, M., 2010. Physical Adsorption Characterization of Nanoporous Materials, Chemie Ingenieur Technik, 82 (7), 1059-1073. Vogel, A., 1989. Vogel's Textbook of quantitative chemical analysis, 5th edition, Longman Scientific & Technical, UK, pp. 324-328. Yin, C.Y., Aroua, M.K., Daud, W.M.A.W., 2007. Review of modifications of activated carbon for enhancing contaminant uptakes from aqueous solutions, Sep. Purif. Technol. 52, 403–415.
Figure Captions Fig.1. Adsorption/desorption of N2 at 77K on the AC0, ACAB and ACABPEX. ............................................... 28 Fig.2. Pore volume distribution of the AC in (a) adsorption and (b) desorption of N2 at 77K. ............................ 29 Fig.3. Point of zero charge (pHPZC) of ACABPEX. .............................................................................................. 30 Fig.4. The effect of pretreatment of AC0 with different concentrations of PEX on its ability for the Zn(II) adsorption ........................................................................................................................................................ 31 Fig. 5. (a) SEM photograph of the external surface of a particle of AC0 after aging in 200 mg/L PEX solution and then Zn(II) loading in SE image mode; (b), (c) &(d) maps of the distribution and relative proportion of carbon, sulfur and zinc elements over the scanned area, respectively. ........................................................... 32 Fig.6. The effect of aging of AC0 in solution containing 0.1N sodium hydroxide and different concentrations of PEX on its Zn(II) adsorption capacity ............................................................................................................ 33 Fig.7 (a) The effect of aging of AC0 in a solution containing 0.1N sodium hydroxide and different concentrations of PEX on its Zn(II) adsorption kinetics; (b) The difference between Zn(II) adsorption kinetics onto AC0 after aging 24 hours in 0.1N sodium hydroxide solution and AC0 after aging 24 hours in o.1N sodium hydroxide together with different concentrations of PEX. ........................................................ 34 Fig.8. The effect of aging of nitric acid oxidized AC0 in solution containing 0.1N sodium hydroxide and different concentrations of PEX on its Zn(II) adsorption capacity ................................................................. 35 Fig.9 (a) The effect of aging of nitric acid oxidized AC0 in a solution containing 0.1N sodium hydroxide and different concentrations of PEX on its Zn(II) adsorption kinetics; (b) The difference between Zn(II) adsorption kinetics onto nitric acid oxidized AC0 after aging 24 hours in 0.1N sodium hydroxide solution and nitric acid oxidized AC0 after aging 24 hours in o.1N sodium hydroxide together with different concentrations of PEX. ................................................................................................................................... 36 Fig.10. The effect of Zn(II) initial concentration on its loading capacity onto ACABPEX and also AC0 ........... 37 Fig.11. The effect of Zn(II) initial concentration on its kinetics onto ACABPEX ............................................... 38 Fig.12. The effect of Zn(II) initial concentration on its removal percentage onto ACABPEX ............................ 39 Fig.13. Comparison between the loading capacity of Zn(II) onto AC0, ACB, ACAB and ACABPEX at different initial Zn(II) concentrations. ........................................................................................................................... 40 Fig.14. (a) & (b) SEM photographs of the external surface of a particle of ACABPEX after Zn(II) loading in SE and BSE image modes, respectively; (c), (d) &(e) maps of the distribution and relative proportion of carbon, sulfur and zinc elements over the scanned area, respectively; (f) . SEM photographs of two broken ACABPEX particles after Zn(II) loading in BSE image mode. ...................................................................... 41 Fig.15.(a) SEM image of cross section of an ACABPEX particle in BSE image mode; (b), (c) & (d) Maps of the distribution and relative proportion of carbon, sulfur and zinc elements over the scanned area, respectively. ........................................................................................................................................................................ 42 Fig.16. The calculated qe values for different isotherm models by using non-linear regression method vs. experimental qe values. ................................................................................................................................... 43 Fig.17. The applicability of different kinetic models for prediction of adsorption kinetics of Zn(II) onto ACABPEX at initial concentrations of (a) 69 and (b) 98 (c) 136 and (d) 165mg/L ....................................... 44
Table Captions Table 1. A summary report for the pore analysis of AC0, ACAB and ACABPEX.............................................. 45 Table 2. Isotherms and their linear forms ............................................................................................................. 46 Table 3. Isotherms parameters by linear regression method for the adsorption of Zn(II) onto ACABPEX. ........ 47 Table 4. Isotherms parameters by non-linear regression method for the adsorption of Zn(II) onto ACABPEX .. 48 Table 5. Kinetic models and their linear forms..................................................................................................... 49 Table 6. R2, ARE and model parameters of different kinetic models by linear regression method for the adsorption of Zn(II) onto the ACABPEX ....................................................................................................... 50 Table 7. ARE and model parameters of various kinetic models by Non-linear regression method for the adsorption of Zn(II) onto ACABPEX ............................................................................................................. 51
350
Volume Adsorbed, (cm3/g STP)
300
250
200
150
ACABPEX-Adsorption
ACABPEX-Desorption
ACAB-Adsorption
ACAB-Desorption
AC0-Adsorption
AC0-Desorption
100 0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
Relative Pressure, (P/P0) Fig.1. Adsorption/desorption of N2 at 77K on the AC0, ACAB and ACABPEX.
0.9
1
(a) 0.3
ACABPEX ACAB AC0
log differential pore valume, (cc/g)
0.25
0.2
0.15
0.1
0.05
0 10
(b)
100 Pore Diameter, (A)
0.5
ACABPEX ACAB AC0
0.45 0.4 log differential pore valume, (cc/g)
1000
0.35 0.3 0.25 0.2 0.15 0.1 0.05 0 10
100 Pore Diameter, (A)
Fig.2. Pore volume distribution of the AC in (a) adsorption and (b) desorption of N2 at 77K.
1000
1
∆pH= Initial pH-Final pH
0 0
1
2
3
4
5
6
7
8
9
10
11
12
-1 -2 -3 -4 -5
0.01M KNO3
-6
0.1M KNO3
-7 Initial pH Fig. 3. Point of zero charge (pHPZC) of ACABPEX.
13
4 3.5 3
qt (mg/g)
2.5 2 1.5 AC0-N0 aging AC0-aging-No PEX AC0-aging-25mg/L PEX
1 0.5
AC0-aging-50 mg/L PEX AC0-aging-200 mg/L PEX
0 0
5
10
15
20
25
Time (h) Fig.4. The effect of pretreatment of AC0 with different concentrations of PEX on its ability for the Zn(II) adsorption
(a)
(b)
(c)
(d)
Fig. 5. (a) SEM photograph of the external surface of a particle of AC0 after aging in 200 mg/L PEX solution and then Zn(II) loading in SE image mode; (b), (c) &(d) maps of the distribution and relative proportion of carbon, sulfur and zinc elements over the scanned area, respectively.
8.2 8
qe (mg/g)
7.8 7.6 7.4 7.2 7 0
100
200
300
400
500
600
PEX concentration, mg/L Fig.6. The effect of aging of AC0 in solution containing 0.1N sodium hydroxide and different concentrations of PEX on its Zn(II) adsorption capacity
9
1.5
(a)
8
(b) 1 qt With PEX- qt No PEX (mg/g)
7 qt (mg/g)
6 5 4 No PEX PEX 25 mg/L PEX 50 mg/L PEX 100 mg/L PEX 150 mg/L PEX 200 mg/L PEX 300 mg/L PEX 400 mg/L PEX 500 mg/L PEX 600 mg/L
3 2 1
5
10
15
20
0 0
-0.5
-1
0 0
0.5
25
10
20 PEX25-No PEX50-No PEX100-No PEX150-No PEX200-No PEX300-No PEX400-No PEX500-No PEX600-No
-1.5 Time (h) Time (h) Fig.7 (a) The effect of aging of AC0 in a solution containing 0.1N sodium hydroxide and different concentrations of PEX on its Zn(II) adsorption kinetics; (b) The difference between Zn(II) adsorption kinetics onto AC0 after aging 24 hours in 0.1N sodium hydroxide solution and AC0 after aging 24 hours in o.1N sodium hydroxide together with different concentrations of PEX.
16.6 16.4
qt (mg/g)
16.2 16 15.8 15.6 15.4 15.2 15 14.8 0
200
400
600
800
1000
PEX concentration (mg/L) Fig.8. The effect of aging of nitric acid oxidized AC0 in solution containing 0.1N sodium hydroxide and different concentrations of PEX on its Zn(II) adsorption capacity
18
2
(a) 16
1.5
qtWith PEX-qt No PEX (mg/g)
14
qt (mg/g)
12 10 No PEX PEX 25 mg/L PEX 50 mg/L PEX 100 mg/L PEX 150 mg/L PEX 200 mg/L PEX 300 mg/L PEX 400 mg/L PEX 600 mg/L PEX 800 mg/L PEX 1000 mg/L
8 6 4 2
5
10
15
Time (h)
20
1 0.5 0 0
10
20
-0.5 -1
0 0
(b)
PEX25-No PEX50-No PEX100-No PEX150-No PEX200-No PEX400-No PEX300-No PEX600-No PEX800-No PEX1000-No
25
-1.5 Time (h)
Fig.9 (a) The effect of aging of nitric acid oxidized AC0 in a solution containing 0.1N sodium hydroxide and different concentrations of PEX on its Zn(II) adsorption kinetics; (b) The difference between Zn(II) adsorption kinetics onto nitric acid oxidized AC0 after aging 24 hours in 0.1N sodium hydroxide solution and nitric acid oxidized AC0 after aging 24 hours in o.1N sodium hydroxide together with different concentrations of PEX.
17 16.5 16 qe (mg/g)
15.5 15 14.5 14 13.5 13 12.5 12 50
60
70
80
90
100
110
120
130
140
150
160
170
Initial Zn(II) concentration (mg/L) Fig.10. The effect of Zn(II) initial concentration on its loading capacity onto ACABPEX and also AC0
20 18
qt (mg/g)
16 14 12 10 8
Initial Zn(II) Conc. 59 mg/L Initial Zn(II) Conc. 78 mg/L Initial Zn(II) Conc. 98 mg/L Initial Zn(II) Conc. 118 mg/L Initial Zn(II) Conc. 136 mg/L Initial Zn(II) Conc. 155 mg/L
6 4 2
Initial Zn(II) Conc. 69 mg/L Initial Zn(II) Conc. 88 mg/L Initial Zn(II) Conc. 108 mg/L Initial Zn(II) Conc. 128 mg/L Initial Zn(II) Conc. 146 mg/L Initial Zn(II) Conc. 165 mg/L
0 0
5
10
15
20
Time (h) Fig.11. The effect of Zn(II) initial concentration on its kinetics onto ACABPEX
25
100 90 80
Zn(II) % Removal
70 60 50 40 30 20 10
Initial Zn(II) Conc. 69 mg/L
Initial Zn(II) Conc. 78 mg/L
Initial Zn(II) Conc. 88 mg/L
Initial Zn(II) Conc. 98 mg/L
Initial Zn(II) Conc. 108 mg/L
Initial Zn(II) Conc. 118 mg/L
Initial Zn(II) Conc. 128 mg/L
Initial Zn(II) Conc. 136 mg/L
Initial Zn(II) Conc. 146 mg/L
Initial Zn(II) Conc. 155 mg/L
Initial Zn(II) Conc. 165 mg/L
0 0
5
10
15 20 Time (h) Fig.12. The effect of Zn(II) initial concentration on its removal percentage onto ACABPEX
25
18 ACABPEX
16
ACAB 14
ACB AC0
12 qe (mg/g)
10 8 6 4 2 0 0
20
40
60
80
100
120
140
160
Initial Zn(II) concentration (mg/L) Fig.13. Comparison between the loading capacity of Zn(II) onto AC0, ACB, ACAB and ACABPEX at different initial Zn(II) concentrations.
(a)
(b)
(c)
(d)
(e)
(f)
Fig. 14. (a) & (b) SEM photographs of the external surface of a particle of ACABPEX after Zn(II) loading in SE and BSE image modes, respectively; (c), (d) &(e) maps of the distribution and relative proportion of carbon, sulfur and zinc elements over the scanned area, respectively; (f) . SEM photographs of two broken ACABPEX particles after Zn(II) loading in BSE image mode.
(a)
(b)
(c)
(d)
Fig. 15.(a) SEM image of cross section of an ACABPEX particle in BSE image mode; (b), (c) & (d) Maps of the distribution and relative proportion of carbon, sulfur and zinc elements over the scanned area, respectively.
17
Calculated qe (mg/g)
16
15
Experimental values Freundlich
14
Temkin Langmuir RP
13
KC 12 12
13
14 15 Experimental qe (mg/g)
16
17
Fig.16. The calculated qe values for different isotherm models by using non-linear regression method vs. experimental qe values.
20
20 (a)
(b)
18
16
16
14
14
12
12
qt (mg/g)
qt (mg/g)
18
10 8
Experimental Values
6
Elovich
10 8
Experimental Values
6
Elovich
4
FP
PSO PFO
2
FP
4 2 0
PSO PFO
0 0
10
20
0
10
Time (h) 20
20 (d)
18
16
16
14
14
12
12
qt (mg/g)
qt (mg/g)
Time (h)
(c)
18
20
10 8
Experimental Values Elovich FP PSO PFO
6 4 2
10 8
Experimental Values Elovich FP PSO PFO
6 4 2
0
0 0
5
10
15
20
25
0
5
Time (h)
10
15
20
25
Time (h)
Fig.17. The applicability of different kinetic models for prediction of adsorption kinetics of Zn(II) onto ACABPEX at initial concentrations of (a) 69 and (b) 98 (c) 136 and (d) 165mg/L
Table 1. A summary report for the pore analysis of AC0, ACAB and ACABPEX Type of activated carbon
AC0
ACAB
ACABPEX
Area
BET surface area Langmuir surface area Single point surface area at P/P0 0.2057 BJH cumulative adsorption surface area of pores between 17 and 3000 A diameter BJH cumulative desorption surface area of pores between 17 and 3000 A diameter Micropore area
1025.0320 1361.3854 1053.9333 88.6050
908.8878 1202.3763 937.8924 71.6299
804.0222 1066.7009 829.3566 75.5723
106.3752
84.7216
86.3966
798.1767
727.4421
633.7935
Single point total pore volume of pores less than 1233.8228 A diameter at P/P0 0.9841 BJH cumulative adsorption pore volume of pores between 17 and 3000 A diameter BJH cumulative desorption pore volume of pores between 17 and 3000 A diameter Micropore volume
0.512316
0.450685
0.404873
0.083709
0.067092
0.070541
0.090805
0.071380
0.073301
0.388723
0.355372
0.309728
Average pore diameter (4V/A by Langmuir) BJH adsorption average pore diameter (4V/A) BJH desorption average pore diameter (4V/A)
15.0528 37.7896 34.1452
14.9931 37.4660 33.7012
15.1823 37.3368 33.9368
2
(m /g)
Volume 3
(cm /g)
Pore Size (A)
Table 2. Isotherms and their linear forms Isotherms Freundlich
Equation **
Langmuir**
qe=KF(Ce)
Linear form 1/n
-1
lnqe=lnKF + n lnCe
lnqe vs. lnCe
KF=exp(intercept),
(Freundlich,
1/n=slope
1906)
(Ce/qe) vs. Ce
Type(II)
1 1 1 = + qe KLq m Ce qm
1/qe vs. 1/Ce
qe=(qmKLCe)/(1+KLCe)
qe=qm ln(KTCe)
Ref.
Ce C 1 = + e qe K Lq m qm
Type(IV)
Temkin
Parameters
Type(I)
Type(III)
**
Plot
qe = qm − (
1 qe ) K L Ce
qe = KLqm − K Lqe Ce
qe=qmlnKT + qmlnCe
-1
qm=(slope) , KL=slope/intercept
qe vs. qe/Ce
qm=(intercept)-1,
(Langmuir,
KL=intercept/slope
1917 &
qm= intercept,
1918)
KL= - (slope)-1 qe/Ce vs. qe
qm= -(intercept/slope), KL=-slope
qe vs. lnCe
qm=slope, KT=exp(intercept/slope)
(Temkin
g=slope,
(Redlich &
ln[(ARPCe/qe) -1]
BRP=exp(intercept),
Peterson,
vs. lnCe
ARP*
1959)
AKC=(slope)-1,
(Koble &
BKC=intercept/slope
Corrigan,
and Pyzhev, 1940)
RP***
KC***
qe=(ARPCe)/(1+BRPCeg)
qe=(AKCCeP)/(1+BKCCeP)
ln[(ARPCe/qe) -1]=glnCe+lnBRP
(1/qe)=(1/AKCCeP)+(BKC/AKC)
(1/qe) vs. (1/CeP)
*
P *
optimized using a trial and error method ** two-parameter isotherm *** three-parameter isotherm
1952)
Table 3. Isotherms parameters by linear regression method for the adsorption of Zn(II) onto ACABPEX. Isotherms
R2
ARE
parameters
Freundlich
0.904
2.394
KF=10.206 (mg g-1)(L mg-1)1/n; 1/n=0.117
Type(I)
0.999
1.346
qm=17.544 (mg g-1); KL= 0.285 (L mg-1)
Type(II)
0.949
1.380
qm=17.544 (mg g-1); KL=0.305 (L mg-1)
Type(III)
0.935
1.422
qm= 17.420 (mg g-1); KL= 0.312 (L mg-1)
Type(IV)
0.935
1.330
qm= 17.545 (mg g-1); KL=0.292 (L mg-1)
Temkin
0.927
2.133
qm=1.746 (mg g-1); KT=215.230 (L mg-1)
RP
0.998
1.335
g=1.001; BRP=0.297899 (L mg-1)g;
Langmuir
ARP= 5.24502 (mg g-1) (L mg-1) KC
0.949
1.343
AKC=5.494505 (mg g-1) (L mg-1)P; BKC=0.313187 (L mg-1)P; P= 0.9845
Table 4. Isotherms parameters by non-linear regression method for the adsorption of Zn(II) onto ACABPEX Isotherms
ARE
parameters
Freundlich
1.914
KF=11.130 (mg g-1)(L mg-1)1/n; 1/n=0.095168
Langmuir
1.318
qm=17.542 (mg g-1); KL= 0.299 (L mg-1)
Temkin
2.120
qm=1.735 (mg g-1); KT=215.237 (L mg-1)
RP
1.332
g=1.0009; BRP=0.297881 (L mg-1)g; ARP=5.245018 (mg g-1) (L mg-1)
KC
1.271
AKC=3.86547 (mg g-1) (L mg-1)P; BKC= 0.22167 (L mg-1)P; P= 1.105431
Table 5. Kinetic models and their linear forms Kinetic
Equations
Linear form
Plot
Parameters
Ref.
qt=β-1ln(αβt)
qt=β-1ln(αβ) + β-1lnt
qt vs. lnt
β=slope-1,
(Aharoni&Tompkins,1970;
α=slope ×
Juang&Chen, 1997)
models Elovich
exp(intercept/slope) FP
qt = ktv
lnqt=lnk+vlnt
lnqt vs. lnt
k=exp(intercept), v=slope
(Sparks, 1986)
PFO
qt = qe[1 −exp(-k1pt)]
ln(qe-qt)=lnqe-k1pt
ln(qe-qt) vs. t
qe=exp(intercept),
(Lagergren, 1898)
k1p=-(slope) Type(I)
PSO
qt =
k 2p q e2 t
Type(II)
1 + q e k 2p t Type(III)
Type(IV)
t 1 t = + q t k 2p q e2 q e
t/qt vs. t
1 1 1 1 =( ) + qt k 2p q 2e t q e
1/qt vs. 1/t
k2p=(slope2)/intercept
1 qt ) k 2 pq e t
qt vs. qt/t
qt = k 2p qe2 − k 2p qe q t t
qt/t vs. qt
q t = qe − (
qe=slope-1 qe=intercept-1
(Ho&
k2p=( intercept 2)/slope
McKay;1998;1999;2000)
qe=intercept k2p= -1/(slope× intercept) qe= - intercept/slope k2p=( slope 2)/ intercept
Table 6. R2, ARE and model parameters of different kinetic models by linear regression method for the adsorption of Zn(II) onto the ACABPEX Initial Zn(II) Conc.
59
69
78
98
108
128
136
146
155
165
R2
0.937
0.976
0.973
0.983
0.974
0.976
0.976
0.974
0.980
0.980
ARE
9.801
9.707
10.80
10.62
11.51
10.74
9.757
9.318
10.01
8.087
0.397
0.348
0.328
0.297
0.293
0.303
0.301
0.298
0.285
0.298
17.74
15.19
14.48
13.42
14.46
17.06
17.73
17.30
15.98
19.87
0.938
0.980
0.985
0.975
0.989
0.989
0.991
0.993
0.987
0.990
11.31
7.147
5.824
9.183
5.164
5.950
4.633
4.111
6.228
4.716
4.495
4.477
4.459
4.208
4.618
4.609
5.244
5.181
4.899
5.624
0.365
0.390
0.403
0.450
0.422
0.429
0.383
0.388
0.418
0.370
0.962
0.992
0.991
0.985
0.982
0.984
0.980
0.979
0.986
0.982
21.17
27.07
27.69
29.52
32.63
31.84
34.65
35.27
31.24
34.31
13.72
12.34
12.99
13.78
14.18
14.59
14.20
14.33
14.81
14.45
(mg/L)
Elovich
β(g mg
)
α(mg g
h-1)
-1
-1
2
R
ARE FP
k(mg g
-1
-v
h )
v 2
R
ARE PFO
qe (mg g
-1
k1p (h
(I)
)
0.418
0.192
0.176
0.158
0.147
0.145
0.152
0.146
0.155
0.163
2
R
0.995
0.998
0.997
0.998
0.996
0.995
0.995
0.993
0.996
0.996
ARE
6.984
2.961
3.878
3.974
4.777
5.825
3.507
3.529
5.293
3.060
13.51
15.87
16.95
18.52
19.23
19.61
19.23
19.61
20.00
19.61
0.035
0.019
0.016
0.013
0.012
0.012
0.013
0.013
0.012
0.014
0.942
0.989
0.978
0.996
0.977
0.985
0.979
0.949
0.982
0.968
6.730
4.026
5.899
3.028
6.224
6.228
7.413
9.646
6.260
7.786
13.33
14.93
15.15
17.24
16.67
16.67
16.39
15.87
16.95
16.67
-1
)
qe(mg g
-1
)
k2p(g mg
-1
-1
h )
2
R (II)
ARE qe(mg g
-1
)
k2p(g mg
0.035
0.024
0.024
0.017
0.021
0.021
0.026
0.030
0.023
0.029
2
R
0.866
0.966
0.939
0.978
0.931
0.928
0.917
0.858
0.930
0.902
ARE
7.009
3.538
5.066
3.143
5.849
6.084
6.222
8.445
5.845
6.428
13.61
15.32
15.90
17.40
17.51
17.66
17.27
16.89
17.91
17.45
-1
PSO
(III)
qe(mg g
-1
h )
)
k2p(g mg
0.032
0.022
0.020
0.016
0.017
0.017
0.021
0.024
0.019
0.024
2
R
0.866
0.966
0.939
0.978
0.931
0.928
0.917
0.858
0.930
0.902
ARE
7.687
3.287
4.612
3.206
5.592
5.920
5.521
7.214
5.690
5.653
14.25
15.53
16.28
17.56
18.01
18.20
17.80
17.78
18.40
18.07
0.027
0.021
0.019
0.016
0.016
0.016
0.019
0.019
0.017
0.021
-1
(IV)
-1
qe(mg g
-1
k2p(g mg
-1
h )
)
-1
-1
h )
Table 7. ARE and model parameters of various kinetic models by Non-linear regression method for the adsorption of Zn(II) onto ACABPEX Initial Zn(II)
Elovich ARE
FP
Parameters
ARE
PFO
Parameters
ARE
PSO
Parameters
ARE
Parameters
Conc.
β
α -1
-1
K -1
(g mg )
(mg g h )
-1
v
qe
-v
-1
(mg g h )
k1p
qe
-1
(mg g )
(h )
-1
k2p -1
-1
(mg g )
(g mg h )
59
7.47
0.444
20.883
10.52
4.513
0.321
11.49
12.18
0.376
4.632
13.75
0.025
69
8.53
0.380
21.252
6.83
4.278
0.390
12.92
12.53
0.320
2.917
15.77
0.019
78
9.64
0.371
20.927
5.68
4.365
0.428
13.84
13.04
0.297
3.833
16.82
0.016
98
8.57
0.336
18.001
8.89
4.072
0.462
11.086
13.01
0.340
2.914
17.91
0.015
108
9.44
0.354
20.905
5.04
4.622
0.436
14.257
13.57
0.299
4.612
19.08
0.012
128
9.52
0.338
20.916
5.58
4.462
0.457
14.680
13.49
0.316
5.294
17.37
0.017
136
8.88
0.358
26.575
4.467
5.209
0.398
15.761
13.83
0.333
3.925
17.60
0.018
146
8.08
0.353
24.862
3.964
5.269
0.372
21.557
17.75
0.146
3.375
18.01
0.016
155
8.65
0.328
22.268
5.940
4.806
0.454
14.515
14.23
0.306
4.748
18.51
0.015
165
7.51
0.336
27.847
4.409
5.541
0.403
15.735
14.50
0.333
2.996
18.43
0.017
