Preparation of granular activated carbons from composite of powder activated carbon and modified β-zeolite and application to heavy metals removal

Zeolites, as microporous aluminosilicates with crystalline structures, are extensively used as selective adsorbents in the environmental aspects or for catalysts in various chemical processes. Zeolites, such as β zeolite (a 12-ring aperture (3D) high-silica zeolite), have been broadly employed for diverse purposes encompassing separation and adsorption processes (Kunkeler et al. 1998; Motlagh-Bahadory-Esfahani & Faghihian 2014).

Activated carbons (with high porosity and large surface area) are applied in a variety of forms including granular activated carbon (GAC) and powder activated carbon (PAC). Typical applications of PAC are use in sugar decolorization, food processing and pharmaceuticals. GAC is frequently used in continuous processes of gas and liquid phase applications. GAC has a number of advantages over PAC including lower pressure drop, regeneration and reusability, high hardness and low attrition (Oh et al. 2009; Al-Rahbi & Williams 2016).

In the recent decades, different methods have been proposed to produce the shaped activated carbons (Garcia-Garcia et al. 1997; Lim et al. 2010). In order to improve the properties of the shaped activated carbons, using binders are a conventional method. In several studies, a range of binders were used including ammonium phenolic resin, lignosulphonate (ALSA), araldite resin, PVA, sodium salt and humic acid (Shi et al. 2003; Smith et al. 2012), clays and organic binder (Carvalho et al. 2006; Lim et al. 2010; Saeidi & Lotfollahi 2015). In addition to the effects of the binders (organic and inorganic) in shaping the GAC, previous researches examined the effects of particle size distribution of PAC on the properties of their products. They concluded that controlling the particle size distribution of PAC improved the products quality (Saeidi & Lotfollahi 2014, 2015a, 2015b, 2016).

Recently, the study of zeolites/activated carbon composite is considered as a new material which has properties of both zeolites and carbon. The zeolites/activated carbon composites were extensively used in scientific and experimental studies as a result of the adsorptive features of zeolite and activated carbon and their extensive uses in the wastewater industries and gas purification. Composite materials produced from activated carbon and zeolite materials contain hydrophobic and hydrophilic properties. For that reason, this new composite is appropriate for sorption of both organic materials and metallic ions from aqueous and gaseous phases (Babic et al. 2011; Foo & Hameed 2011; Ma et al. 2012). Zhu et al. (2016) produced a set of spherical samples by activated carbon, ZSM-5, and phenolic resin as binder. Their results demonstrated that the produced samples (with high surface area) showed well removal effects for benzene, suitable iodine number and an acceptable bactericidal effect.

Heavy metals (such as cadmium (Cd), zinc (Zn), and lead (Pb) ions) in wastewater streams have attracted the attention of the environmental researchers due to their toxicity and threat to human beings. Therefore, removal of heavy metals from wastewater is one of the most significant environmental and economic topics (Amer et al. 2017; Bahabadi et al. 2017). Between the several methods for removing the heavy metals from the wastewater streams, the adsorption process is an attractive alternative because of its low cost, simple operation, high efficiency, and flexible design.

In the present study, a mixture of methylcellulose (MC), β-zeolite, Cu-modified β-zeolite, bentonite and PAC was used for producing GAC as a new composite. The effects of the binders, β-zeolite and Cu-modified β-zeolite compositions on the mechanical strength and the adsorption properties of the produced samples were examined. The sorption of Zn²⁺, Cd²⁺ and Pb²⁺ ions from aqueous solutions using modified GAC was also investigated under different experimental conditions. The kinetic models were used in this study to analyze the adsorption rate data. The Langmuir and Freundlich isotherm models were applied to analyze the experimental equilibrium data.

Bentonite (specific surface area of 16 m²/g) and methyl cellulose (400 cps) as binders were provided from Shah industrial mining company (Iran) and BDH (UK), respectively. PAC with iodine number of 1,270 mg/g (ASTM standard D4607–94) was purchased from Jacobi commercial company (Sweden). Sodium thiosulfate (Na₂S₂O₃), potassium iodide (KI), iodine (I), starch, hydrochloric acid (HCl) and sodium hydroxide (NaOH) were obtained from Merck Company (Germany). β-Zeolite was purchased from Zeolyst International Company (USA). The chemicals used for the study were analytical grades of Lead nitrate (Pb(NO₃)₂.H₂O), cadmium nitrate (Cd(NO₃)₂.4H₂O), zinc nitrate (Zn(NO₃)₂.6H₂O) and copper nitrate (Cu(NO₃)₂.3H₂O) all purchased from Sigma-Aldrich. These chemicals were used to prepare the metal solutions by dissolving them in double distilled water.

Preparation of cu-modified β-zeolite was carried out by an ion-exchange method (Azizi & Ehsani-Tilami 2013). In this method, 0.1 g β-zeolite was mixed with 10 mL aqueous solution of copper nitrate (Cu(NO₃)₂.3H₂O) (1 M) at pH = 4 in a 50 mL reaction flask. Cu ion-exchanged β-zeolite was prepared at 298 K for 4 h. The bright blue solid phase was filtered, washed with distilled water. Then the solid was dried at 343 K for 24 h and calcinated at 823 K for 4 h.

β-zeolite and the modified samples were characterized by a series of complementary analytical techniques. In this work, for preparing the cu-modified β-zeolite, 0.1 g β-zeolite was mixed with 10 mL aqueous solution of copper nitrate (Cu(NO3)2.3H2O) (1 M). Powder X-ray diffraction (XRD) was obtained using a GBC MMA X-ray diffractometer (Cu Kα as radiation resource) in the scope of scanning angle (2θ) 5°–50° with scanning rate of 5°/min. Fourier transform infrared spectroscopy was performed (Nicolet 400 D Impact spectrum single FT-IR device) at the scanning scope of 4,000–400 cm⁻¹ with the resolution of 4 cm⁻¹.

In order to form the zeolite + PAC + binders composite, different proportions of PAC, MC as organic binder, β-zeolite or Cu-modified β-zeolite, bentonite as inorganic binder and water was mixed. The mixing was performed in a laboratory mixer with 300 rpm for 20 min in a water bath at 80 °C (The temperature was regulated by a thermostatic bath with the precision of ±0.2 °C). The mixture's plasticity related to the ratio of binders in the mixture (Saeidi & Lotfollahi 2015). The components were weighed on an analytical balance (A&D Company, GF-600, Japan) with an accuracy of 0.001 g. The produced mixture was injected into a cylindrical mold (1 cm diameter and 1 cm long). Then, the produced GACs were dried at 100 °C for 48 h. It should be noted that the uniform mixing using high-speed mixer improves the adsorption capacities of the produced samples (Saeidi & Lotfollahi 2015).

The FT-IR spectra of GAC is presented in Figure 1. The peak at 3,453.6 cm⁻¹ can be attributed to the O–H stretching of the hydroxyl groups and intermolecular hydrogen bonds. The characteristic bands at 2,700–2,900 cm⁻¹ were considered as the C-H stretching due to presence the CH and CH₂ groups of the cellulose structure. The peak at 2,352.5 cm⁻¹ was related to the C–H stretching of methyl groups. The characteristic bands at region 1,000–1,250 cm⁻¹ were related to the Si–O–C structure (peaks at 1,188.4 cm⁻¹ and 1,066.7 cm⁻¹). In addition to, the existence of Si–C bonding was confirmed by the peak at 764.4 cm⁻¹. Another, peaks were attributed to the β-zeolite and AC structure. These changes in the peaks and the creation of various functional groups indicated that the new adsorbent (GAC) has been produced.

Figure 2 demonstrates the XRD spectra of the β-zeolite and the Cu modified β-zeolite studied in this work. The location of diffraction lines which remained unchanged demonstrated that the structure of β-zeolite remained intact after introducing the Cu cations. The XRD patterns of Cu modified β-zeolite as a CuO phase, which can be defined with the main diffraction peaks at 2θ = 32.42°, 35.54°, 38.75°, 48.74°, indicated that the oxidation of Cu²⁺ could be occurred during the thermal treatment.

The FT-IR spectra of β-zeolite and Cu modified β-zeolite are presented in Figure 3. The characteristic bands at 1,070–1,090 cm⁻¹ observed in both spectra were considered as the Si–O stretching of Si–O–Si structure. The bands within the ranges 1,000–1,300 cm⁻¹ and 400–700 cm⁻¹ were attributed to the β-zeolite structure (Figure 3). Based on the comparison of Figure 3(a) and 3(b), it can be concluded that the intensity of diffraction lines decreased as the Cu cation was introduced. Comparing two spectra for β-zeolite and Cu-modified β-zeolite (Figure 3(a) and 3(b)) indicated that the bands of Cu-modified β-zeolite decreased sharply at 1,089 cm⁻¹ and 1,224 cm⁻¹. They are powerful evidences for the presence of copper. The same trend was observed in the other bands. In order to obtain more information about the existence of Cu in the β-zeolite (Figure 3(b)), the bands at range of 465.16 and 568.76 cm⁻¹ are related to Cu²⁺ interacting with oxygen in the β-zeolite structure. The results indicated that Cu has affected all functional groups in the β-zeolite.

In this work, PAC, MC, β-zeolite, Cu modified β-zeolite and bentonite were used to produce the GAC. The surface area of produced samples depends on the value of organic binder (MC) (Saeidi & Lotfollahi 2016). Increasing the content of MC (wt%) leads to decreasing surface area of produced samples. Therefore, the minimum amount of MC should be determined to reach good shape and high mechanical strength accompanied with suitable surface area. For this reason, the amount of MC was increased in order to reach an appropriate amount. Increasing the MC content to 5 wt% led to production of samples with good qualities both in appearance and mechanical strength. The samples with less than this value (1, 2, 3, 4 wt%) did not have good shape and mechanical strength. The MC content was increased to 6, 7 and 8 wt% in order to study its effects on the properties of the produced GAC samples. Increasing MC content from 5 to 8 wt% in the constant inorganic binder led to a decrease in the surface areas of the produced samples. However, the mechanical strength of the samples increased and the iodine number of the samples decreased with increasing MC content in the produced samples. The results of the mechanical strength show an improvement in the mechanical strength of the samples with the increase in the bentonite content (Table 1).

As can be seen in Table 1, by increasing the β-zeolite content in the initial mixture from 0 to 3.5 wt%, the iodine number of the samples increased from 1,210.8 to 1,259.1 mg/g and the mechanical strength of the samples reduced from 222.78 to 191.37 N. In order to improve properties of the produced GAC, β-zeolite was modified by copper and the properties of modified samples were compared with the properties of the original β-zeolite. Using Cu for modification of β-zeolite led to higher adsorption (iodine number) compared to other metals due to an increased number of active sites after modification (Wen et al. 2016; Bahabadi et al. 2017). The results indicated that by using the Cu modified β-zeolite content, the iodine number of samples increased (Table 2). For example, the iodine number of samples containing 5 wt% MC, 1 wt% β-zeolite, and 9 wt% bentonite in the composite, which was 1,230.4, increased to 1,251.3 mg/g when Cu-β-zeolite was used in the composite. The mechanical strength of samples was slightly increased from 218.15 to 218.79 N.

Sorption experiments have been conducted to find the optimum pH and contact time as very important parameters in the adsorption process. The produced GACs, with different amounts of Cu modified β-zeolite or β-zeolite, MC, and bentonite (wt%), were used for removing Zn⁺², Cd⁺² and Pb²⁺ ions from aqueous solutions. Zinc, lead and cadmium poisoning are important health issues in every country (Ghasemian-Lemraski & Sharafinia 2016). Ion solutions were prepared by dissolving a suitable amount of Cd(NO₃)₂.4H₂O, Pb(NO₃)₂.H₂O and Zn(NO₃)₂.6H₂O in 1,000 ml of distilled water. A series of 100 ml conical flasks was used in this study. The method involved filling each flask with 100 ml of Zn, Cd and Pb ion solutions with a concentration of 50 ppm and adding 0.75 g of the sorbent to the solution. The flasks were shaken at room temperature by a mechanical shaker. At the end of the test, the sorbent was filtered out and washed using deionized water. The concentration of Zn, Cd and Pb ions in the solution was obtained by AAS.

The pH of a solution is one of the most important operational parameters in the removal of heavy metal ions from solution with the adsorption process. The effect of initial pH on the sorption of Zn, Cd and Pb from solutions was investigated for all of the produced sorbents at room temperature. In these experiments, 0.75 g of the sorbents was added to 100 mL of a 50 ppm solution at different pHs in the range of 3–7 (shown in Figure 4). As can be observed in Figure 3, increasing the pH led to an increase in the sorption of Zn, Cd and Pb, which reached a maximum value at pH 5.5 and then decreased. It can be seen from Figure 3 that pH values less than 5 are not suggested for the adsorption process. The low adsorption capacity at low pH can be justified because of destruction in the structure of zeolites (Apiratikul & Pavasant 2008; Malamis & Katsou 2013). Therefore, the structure of zeolite may collapse in the presence of acids (Rao et al. 2006) especially at a low pH. The amount of damage to the structure of zeolite depends on the pH of the solution.

The adsorption of metal ions on the produced GAC with different values of MC content, bentonite and β-zeolite or Cu modified β-zeolite (wt%) were conducted as a function of contact time. The experiments were carried out in a metal ion concentration of 50 ppm, pH 5.5 (maximum value) and 0.75 gram of the sorbents at room temperature. It was observed that the amount of Zn, Cd and Pb sorption increased with an increase in the contact time. As shown in Figures 5 and 6, efficiency of removal first followed an increase and then a stationary state was observed. It was concluded that 30 min was adequate for sorption to attain equilibrium. The order of maximum sorption capacity for GAC (MC = 5 wt%, Cu modified β-zeolite = 3 wt% and bentonite = 7 wt%) at contact time of 30 min, in a unit of mg·g⁻¹ was: Pb²⁺ (6.506) > Cd²⁺ (6.393) > Zn²⁺ (6.073). The maximum capacity was obtained after 30 min of sorption and at pH 5.5 by GAC (MC = 5 wt%, Cu modified β-zeolite = 3 wt% and bentonite = 7 wt%) as the most effective adsorbent, and the efficiencies of removal were 97.6%, 95.9% and 91.1% for lead, cadmium and zinc, respectively (Table 3).

The sorption kinetic parameters are in Tables 4–6. The sorption capacities of Zn, Cd and Pb ions by GAC (MC = 5, Cu-β-zeolite = 3, bentonite = 7) wt% as the best sorbent versus time are plotted in Figure 7. As can be seen in Figure 7 and Tables 4–6, the pseudo-second order model could be used to describe the sorption kinetic behavior of Zn, Cd and Pb on sorbents more satisfactorily than the first-order kinetics model based on its higher regression coefficient (R² > 0.99) value.

The aim of this study was preparation of GACs with different proportions of PAC, MC as an organic binder, β-zeolite or Cu-modified β-zeolite as a reinforcing adsorption and bentonite as an inorganic binder. The produced GACs were tested for evaluating their mechanical strength, adsorption capacity (iodine number) and their capability for removing zinc, cadmium and lead ions from aqueous solutions. The produced GAC was prepared with the minimum amount of MC (5%wt) and bentonite (6.5–10 wt%) to obtain samples with suitable shape. To increase the adsorption capacity of the produced GAC, β-zeolite and Cu modified β-zeolite were used. The results showed that by using β-zeolite modified with copper, the adsorption capacity of samples was increased while their mechanical strength remained stable. The GAC samples produced by Cu-modified β-zeolite was found to be a very good adsorbent for separating metal ions from solutions. The results indicated that the optimum pH for removal of metal ions was 5.5 and contact time at 30 min was adequate to reach equilibrium sorption. The maximum capacity of sorption by GAC (MC = 5 wt%, Cu modified β-zeolite = 3 wt% and bentonite = 7 wt%) as the most effective adsorbent was 97.6%, 95.9% and 91.1% for Pb, Cd and Zn, respectively. This adsorbent had acceptable mechanical strength. The kinetic experimental data were described with the sorption kinetic models (pseudo-second and pseudo-first order kinetic models). The sorption kinetic data were well coordinated with the pseudo-second order kinetic model.