0 Introduction

Nowadays, Cu, Zn, and Pb are highly demanded heavy metals in the market and it is projected that the supply would reach its maximum in 2030 to 2050^[1]. This implies the rise of water pollution by these heavy metals from electronic industry, mining, paints industry, and glass industry in the near future. A recent report published by Kenai Watershed Forum found out that fertiliser, traffic intensification as well as general urbanization contribute to Zn and Cu pollution in Kenai River Watershed^[2]. Furthermore, research done by Sankson et al.^[3] reported that the combined sewer overflows discharge in Poland consist of 170-1890 μ g/L Zn and 31-590 μ g/L Cu. Moreover, Voghji River was highly polluted with heavy metal consisting of 1.28-82.5 μ g/L Cu and 0.94-105.5 μ g/L Zn due to mining activities^[4].

Heavy metals cause deleterious effects to living organisms.Heavy metals could enter human body through different routes, such as inhalation of toxic gas and ingestion of contaminated drinking water and food. High exposure to heavy metal could lead to serious health effects and in extreme situation, it can lead to death. Cu²⁺ can cause stomach-ache, irritation in eyes, noses, mouths, and headache, while Zn²⁺ causes stomach cramps, nausea, and respiratory disorder^[5]. Additionally, plants which are exposed to high level of Cu²⁺ will prohibit root elongation as well as root cell membrane^[6]. Similarly, Zn²⁺ also causes growth inhibition in macroalgae, which is economically significant and a fast-growing commodity^[7].

In Malaysia, the allowable discharge for Cu²⁺ according to standard A is 0.2 mg/L and 1 mg/L for standard B^[8]. Meanwhile, the allowable discharge for Zn²⁺ is 2 mg/L for both standard A and B. To ensure good quality of life, it is compulsory to treat the wastewater to the allowable discharge limit. Chemical precipitation, coagulation-flocculation, ion exchange, and membrane filtration are the commonly used method to treat heavy metal polluted water. However, most of these methods have disadvantages including high energy consumption which results in high operating cost, low efficiency of heavy metal removal, and formation of toxic sludge where further treatment is required^[9]. Lately, research has been conducted on genetically modified micro-organism to remove heavy metal such as Cd²⁺ and Pb^2+[10]. However, this is a newly introduced feature, which is not suitable for practical application for now.

Adsorption is a well-known separation method, and it is recognized by numerous scientists for wastewater applications in recent years^[11]. It is because adsorption shows high efficiency in heavy metals removal at reasonable cost and minimizes the chemical and biological sludge produced^[12]. Furthermore, regeneration of adsorbents can be performed in most of the adsorption process, which further reduces the maintenance cost. Isotherm and kinetics studies are commonly conducted in adsorption related research. This is because kinetic study is important to understand the adsorption mechanism and identify the rate of the controlling steps. Pseudo first order (PFO) and pseudo second orders (PSO) are commonly studied to investigate the adsorption mechanism. In PFO, low constant value, K₁, indicates slow adsorption process while high constant value, K₂, reveals an increase in adsorption rate^[13]. Table 1 shows the list of adsorbents used for heavy metals adsorption. Generally, PSO presented the adsorption mechanism well with coefficient of determination, R², which is approximately 0.99.^[13-16] q_e represents the absorbed amount at equilibrium.

Zeolite is a popular adsorbent due to its inherent high porosity and excellent metal binding capacity. Numerous factors such as type of zeolite, pH, contact time, mass of adsorbent, and concentration of feed solution affect the performance of zeolite in heavy metal removal. Thus, the objective of this study is to identify an ideal operating condition to achieve maximum removal rate by considering the effect of pH and contact time on the performance of zeolite. Besides, kinetic study was also conducted to find out the adsorption mechanism.

1 Materials and Methods 1.1 Materials

Analytical grade zeolite crystalline alumiosilicates, which consist of silica and alumina, were supplied by ChemSoln. The zeolite was firstly sieved using 200 μ m × 500 μ m sieves and treated in the oven at 100 ℃ for 24 h before use. CuCl₂ and ZnCl₂ were used as received.

1.2 Methodology

0.5 g zeolite was exposed to 10×10^-6 of Cu²⁺ and Zn²⁺ solutions at pH 3, 5, 7, 9, and 11. The pH of the solution was adjusted by adding either 0.1 N of NaOH or 0.1 N of HCl into the solutions. Sample was collected and filtered using syringe filter (Thermo Fisher, 0.22 micron) after 60 min to evaluate the adsorptive performance of zeolite. In order to find out the effect of contact time on heavy metal adsorption rate, sample was collected every 30 min within 150 min. The concentration of Cu²⁺ and Zn²⁺ before (C_i) and after (C_f) was measured using Atomic Absorption Spectrometer (Agilent Technologies, Model: 240 AA).

The percentage of removed Cu²⁺ and Zn²⁺ was calculated by using the following equation:

$ {\rm{Percentage}}\;{\rm{of}}\;{\rm{removal}} = \frac{{\left( {{C_{\rm{i}}} - {C_{\rm{f}}}} \right)}}{{{C_{\rm{i}}}}} \times 100{\rm{\% }} $

(1)

The kinetic of the adsorption process was determined by using PFO (Eq.(2)), PSO (Eq.(3)), Ritchie's equation (Eq.(4)), Boyd's external diffusion equation (Eq.(5)), Weber and Moris (W & M) (Eq.(6)), and Elovich (Eq.(7)). The equations are listed as below:

$ {\rm{log}}\left( {{q_{\rm{e}}} - {q_t}} \right) = \log {q_{\rm{e}}} - {K_1}t $

(2)

$ \frac{t}{{{q_t}}}{\rm{ = }}\frac{1}{{{K_2}q_{\rm{e}}^2}} + \frac{t}{{{q_{\rm{e}}}}} $

(3)

where q_e and q_t are the amount of adsorbate (solute) adsorbed by the adsorbent (g/mg zeolite) at equilibrium and at time t. K₁and K₂ are defined as the constant values for PFO and PSO models.

$ {q_t} = {q_\infty } - {q_\infty }{\left( {1 + \left( {n - 1} \right) \propto t} \right)^{\frac{1}{{n - 1}}}} $

(4)

$ {q_t} = {q_\infty }\left( {1 - {{\rm{e}}^{ - Rt}}} \right) $

(5)

$ {q_t} = {k_{{\rm{W\& M}}}}{t^{0.5}} $

(6)

$ {q_t} = \frac{1}{b}\ln \left( {1 + abt} \right) $

(7)

The amount of adsorbate adsorbed on adsorbent at infinity time is labeled as q_∞, while n is known as the number of active sites on the adsorbent occupied by adsorbate in Ritchie's equation. R is the rate constants in Boyd's external diffusion equation while k_{W & M} is the intraparticle diffusion coefficient. For Elovich model, a is the initial adsorption rate constant while b is the desorption rate constant.

The following statistic parameters were used to evaluate the validity of the kinetic models.

$ {R^2} = \frac{{\sum {{{\left( {{q_{{\rm{mean}}}} - {q_{{\rm{cal}}}}} \right)}^2}} }}{{\sum {{{\left( {{q_{{\rm{cal}}}} - {q_{{\rm{mean}}}}} \right)}^2}} + \sum {{{\left( {{q_{{\rm{cal}}}} - {q_{\exp }}} \right)}^2}} }} $

(8)

$ {\chi ^2} = \frac{{\sum {{{\left( {{q_{\exp }} - {q_{{\rm{cal}}}}} \right)}^2}} }}{{q_{{\rm{cal}}}^2}} $

(9)

$ {\rm{SSE}} = \sum {{{\left( {{q_{\exp }} - {q_{{\rm{cal}}}}} \right)}^2}} $

(10)

$ MSE = \frac{1}{{{N_{\exp }}}}{\sum {\left( {{q_{\exp }} - {q_{{\rm{cal}}}}} \right)} ^2} $

(11)

R², Eq.(1), residual sum of square (SSE) and mean square error (MSE) are used.

2 Results and Discussion 2.1 Effect of pH

Fig. 1 shows the adsorptive performance of zeolite in Cu²⁺ and Zn²⁺ under varied pH condition. Cu²⁺ removal increased from 30.16% to 82.15% when the pH of the solution increased from pH 3 to pH 7. However, the removal rate remained constant when the pH value increased from 7 to 11. This is because under acidic condition, the presence of free H⁺ behaved as the competitor to both Zn²⁺ and Cu²⁺. This reduces the number of active sites on the zeolite to adsorb Zn²⁺ and Cu²⁺. This finding is similar to recent findings obtained by Refs. [17-18] where the presence of H⁺ reduced the amount of Pb²⁺ adsorbed to both natural and Fe (Ⅲ) modified zeolite alginate beads. Similar adsorptive trend was found in Zn²⁺ solution as shown in Fig. 1, where the percentage of removal rate increased from 64.43% at pH 3 to 90.87% at pH 7 and the removal rate remained constant when the pH is further increased to pH 11.

Comparatively, adsorptive performance of zeolite towards Zn²⁺ was higher compared with Cu²⁺, where the maximum removal rate was ~90% for Zn²⁺ and ~80% for Cu²⁺. This is due to the effect of hydration enthalpy (ΔH_hyd), where ΔH_hyd=-1955 kJ/mol for Zn²⁺ and ΔH_hyd=-2010 kJ/mol for Cu^{2+ [19-20]}. Hydration enthalpy describes the amount of energy required to detach water molecules from cations. In this case, Cu²⁺ needs to release higher amount of energy to detach from water compared with Zn²⁺, thus its interaction with zeolites is lower compared with Zn^2+[21]. As a consequence, the percentage of removal for Cu²⁺ is lower compared with Zn²⁺ under the same pH condition.

2.2 Effect of Contact Time

Fig. 2 shows the effect of contact time on Cu²⁺ and Zn²⁺ removal rate. Generally, the removal rate increased from t=30 min to t=120 min. Equilibrium point achieved at t=120 min, where no further increment in terms of removal rate was found after t=150 min. This is because the number of active sites of zeolites is higher at the beginning of the adsorption process. Rapid adsorption process occurred in the first 30 min, where approximately 60% removal was found for both Cu²⁺ and Zn²⁺. At t=60 min, the removal rate reduced, where additional 15% and 10% removal rate was recorded for Cu²⁺ and Zn²⁺, respectively. This is due to the reduction in the number of active sites, which is available for adsorption. At the saturation point, no active sites are available for either Cu²⁺ or Zn²⁺. Hence, the constant removal rate, which was 94% for Cu²⁺ and 90% for Zn²⁺, was found after 120 min.

q_e of zeolite calculated in this study was 0.36 mg Cu²⁺mg/g zeolite and 0.376 mg Zn²⁺mg/g zeolite when 10×10^-6 Cu²⁺ and Zn²⁺ were used as the feed. Compared with adsorbents listed in Table 1, the adsorptive capacity of the zeolite is very much lower, although the reported removal rate was comparable which is within 90%-94%. Similar to a study done by Wang et al.^[22], the adsorption capacity of SiO₂ encapsulated natural zeolite was 12.3 g/mg and 9.0 g/mg for Cu²⁺ and Zn²⁺. The removal rate of Cu²⁺ and Zn²⁺ were 98% when 10×10^-6 Cu²⁺ and Zn²⁺ were used as the feed. This suggested that the performance of the zeolite used in this study can be further enhanced by reducing the particle size of zeolite, Fe coating^[17], and surfactant modification^[23].

Compared with the findings observed in Fig. 1, the adsorptive performance of zeolite was higher for Zn²⁺ (~90%) compared with Cu²⁺ (~80%). Without pH adjustment, the effect of hydration radii governs the adsorption process. Hydration radii show the size of ion after it interacts with water molecules. It equals to 4.3 Å and 4.19 Å for Zn²⁺ and Cu²⁺, respectively^[20]. Thus, Cu²⁺ with smaller size compared with Zn²⁺ reached the fine pores in zeolite and higher adsorption rate is achieved.

According to Tuzcu and Atalay^[24], the pH of 10×10^-6 Zn²⁺ solution is 6.3 while the pH for 10×10^-6 Cu²⁺ is 6. The pH values reduced from 5.7 to 5.5 when the concentration of Cu²⁺ increased from 50×10^-6 to 100×10^-6. Similarly, when the concentration of Zn²⁺ increased from 50×10^-6 to 100×10^-6, the pH decreased from 6 to 5.9. It is reasonable to deduce that as the concentration of Zn²⁺ and Cu²⁺ reduced, the pH of the solution will go towards neutral. During the adsorption process, pH increased as the concentration of Zn²⁺ in solution reduced. This explains similar Zn²⁺ removal rate observed at t=120 min and at pH 7, which is approximately 90%. Meanwhile, this also suggests that by adjusting the pH of Zn²⁺ solution to minimum pH7, the maximum Zn²⁺ removal rate 90% can be achieved within 60 min. Similar trend was also observed in Cu²⁺ solution, where zeolite adsorbed 82.15% of Cu²⁺ at pH≥7, and without pH adjustment, 75% of Cu²⁺ was removed at t=60 min. However, this removal rate did not treat the water to the allowable discharge limit in Malaysia, which is 1 mg/L. Thus, the need to increase the contact time from 60 to 120 min was required to achieve 94% of Cu²⁺ removal, as shown in Fig. 2. At this point, the allowable discharge limit was met, where the discharge concentration was 0.523 mg/L.

2.3 Kinetic Study

Kinetic study for adsorption process is important as it provides valuable insights on the reaction pathways as well as the sorption reaction mechanism. It also describes the adsorbate uptake rate, which affects the residence time of adsorbate update at the solid-liquid interface. Fig. 3 shows that the amount of adsorbed Cu²⁺ on activated carbon predicted by the 6 kinetic models is very close to the experimental data, except for W & M and Boyd's external diffusion equation. W & M model assumes diffusion of adsorbate is the slowest step while Boyd's external diffusion equation assumes internal diffusion is the rate determining step. Hence, both internal and external diffusion contributed less impact to the overall Cu²⁺ adsorption process, compared with adsorption reaction on the active site in this study.

Table 2 presents the constant values of the respective kinetic models. Empirical models such as PFO and PSO are lack of physical meaning^[25]. Elovich is commonly used to model chemisorption while Ritchie's equation assumes adsorption is dominated by the reaction at active sites. It is notable that Ritchie's equation and Elovich model predicted the experimental data well with R² value of 0.9977. Hence, this is difficult to justify which model fitted the data the best based on the R² value alone. Other statistical parameters such as SSE, MSE, and χ² are presented in Table 2 to verify the models, and it shows that Ritchie's equation is the best model with the lowest SSE, MSE and χ² values. This showed that adsorption at active sites is the rate of determining the steps in this study^[25].

Similarly, Zn²⁺ adsorption data was also predicted by using 6 kinetic models, which covers the external diffusion, internal diffusion, and adsorption on active sites. The results in Fig. 4 and Table 3 show that Elovich predicted Zn²⁺ adsorption the best with the highest R² value of 0.9913 and the lowest SSE(0.000783), χ²(0.002473), and MSE (0.000131) values compared with the other models. This indicates that the role of active sites in zeolite played an important role in determining the heavy metals removal rate. This adsorption process can be enhanced by modifying the surface of active site by considering the exchange of metal ion with the functional groups on the surface of adsorbent.

3 Conclusions

Results obtained from this study show that zeolite exhibited promising adsorptive performance for Cu²⁺ and Zn²⁺ in both neutral and alkali conditions. 82.15% Cu²⁺ and 90.87% Zn²⁺ removal rate was found at pH 7. Under the same pH condition, selectivity towards Zn²⁺ is higher compared with Cu²⁺, which may be due to the effect of hydration enthalpy. Without pH adjustment, selectivity towards Cu²⁺ is higher compared with Zn²⁺ due to smaller hydration radii of Cu²⁺, which is 0.11 Å smaller than Zn²⁺. Besides, rapid adsorption was found in the first 30 min of the experiment and saturation point was achieved at 120 min for both Zn²⁺ and Cu²⁺. Kinetic study showed that adsorption reaction at the active site on zeolite dominated the heavy metal removal process. Hence, the performance of zeolite could be enhanced by modifying the surface properties such as surfactant coating, where the exchange of metal ion with the functional groups on the surface of zeolite is considered.