Mimetite precipitation on Pb-clinoptilolite: an effective approach for arsenate removal from water

The experiments were conducted in polyethylene tubes at ambient laboratory temperature, which was approximately 25°C, without active temperature control. All solutions were prepared using double-distilled water and analytical-grade chemicals obtained from Thermo Scientific Inc. (Waltham, MA, USA) and Chempur Inc. (Piekary Śląskie, Poland), Sodium arsenate dibasic heptahydrate (Na₂HAsO₄·7H₂O) was used as the source of arsenate ions, while lead nitrate (Pb(NO₃)₂) provided the lead ions. Chloride ions were introduced via sodium chloride (NaCl). The pH was adjusted dropwise using 0.5 M HNO₃ or 1 M NaOH, with the pH measured using a glass electrode and an ELMETRON (Zabrze, Poland) CPC-401 pH meter.

Zeolite rock, serving as the carrier of lead in the further experiments, was sourced from the Bystré deposit (Slovak Republic) and supplied by Zeocem, a.s. (Bystré, Slovakia). It was rich in clinoptilolite with identified admixtures of orthoclase, illite, and quartz. The material has been thoroughly characterized in previous studies (see, for example, Solińska & Bajda, 2022; Wołowiec et al., 2017, and the literature cited therein). Initially, in its calcium form, it was transformed into sodium-exchanged clinoptilolite by reacting with a 2 M NaCl solution for 24 hr. The Na-modified clinoptilolite was then prepared for arsenate sequestration by first performing Pb²⁺ sorption, followed by thorough washing to remove excess Pb and desorb loosely bound ions, ensuring the zeolite was primed for subsequent arsenate interactions.

Aqueous Pb(II) solutions were prepared at concentrations of 100, 500, 1000, 2000, and 4000 mg Pb(II)/L with no pH adjustment. Sorption experiments were conducted in batch systems, where 40 mL of solution at varying Pb concentrations was mixed with 0.8 g of Na-transformed clinoptilolite. The slurries were shaken at room temperature for 24 hr. This was shown in previous studies to be sufficient for equilibrium between metal ions in solution and those sorbed on clinoptilolite (Bajda et al., 2004; Mozgawa & Bajda, 2005). Afterward, the samples were centrifuged at 4500 rpm for 10 min, filtered through 0.2-μm syringe filters, and analysed for Pb. Atomic Absorption Spectroscopy (AAS) was used to measure Pb concentrations in the solutions before and after the sorption experiments. The experiments were repeated in triplicate. The amount of Pb sorbed was calculated based on the difference between the initial measured and equilibrium concentrations. The sorption per 0.8 g of zeolite was calculated as: (C_i−C_e) × V. To express sorption per kilogram of zeolite (mg/kg), this value was then divided by the zeolite mass (0.8 g) and multiplied by 1000, according to the equation: (C_i−C_e) × (V/m) × 1000. C_i was the initial measured concentration of Pb (mg/L), C_e was the equilibrium concentration (mg/L), V was the volume of solution in litres (0.04 L), and m was the mass of the zeolite in grams (0.8 g).

Pb-sorbed clinoptilolite resulting from the experiment at 4000 mg Pb/L was thoroughly washed with redistilled water and then centrifuged at 4500 rpm for 5 min. The Pb concentration in the solution resulting from the 7th washing cycle has dropped below the AAS detection limit (equal to 0.1 mg Pb/L). The solid was washed, air-dried, and subjected to acid digestion using 25 mL of hydrofluoric acid (HF), a few drops of perchloric acid (HClO₄), and 20 mL of hydrochloric acid (HCl) per 0.3849 g of zeolite. The digested samples were then analysed for total Pb content.

To determine the equilibrium product of the reaction and define the reaction efficiency in terms of arsenate sequestration from solution, the Pb-modified zeolite, containing approximately 70 g of bound Pb per kg, was reacted with arsenate solutions (0.5 g of Pb-modified zeolite per 40 mL of solution containing 50 mg As(V)/L). The suspensions were shaken for 7 days in the presence of chloride ions (20 mg Cl/L) at pH 2 and 7. These two pH values were selected to represent contrasting conditions: pH 7 corresponds to near-neutral environments where mimetite is highly stable and exhibits minimal solubility, while pH 2 simulates acidic environments such as acid mine drainage, where mimetite stability is significantly reduced and lead release from Pb-modified zeolite may occur (Bajda, 2010; Magalhães & Silva, 2003). The solution was sampled at 24 hr and 7 days and analysed for As (UV-Vis spectrophotometry) and Pb (AAS). Solid samples were collected after 1 week, washed three times with water, centrifuged at 4500 rpm, air-dried, and analysed using X-ray powder diffraction (XPRD) and scanning electron microscope with an energy dispersive spectrometer (SEM-EDS).

Concentrations of As(V) in solutions were measured by spectrophotometry (Lenoble et al., 2003) using the molybdenum blue method, with a detection limit of 20 μg As(V)/L, on a HITACHI U-800 UV-Vis spectrophotometer at 870 nm. Pb concentrations were determined by atomic absorption spectroscopy (AAS) using a Savant AA spectrometer (GBC Scientific Equipment (Melbourne, Australia); detection limit: 0.1 mg Pb/L, λ = 217 nm).

Previously ground in an agate mortar, solid samples were analysed by XPRD using a Rigaku SmartLab diffractometer (2–72° 2θ, step size 0.02°, 1 s/step). Phases were identified with the PDF-2 JCPDS database and XRAYAN software (KOMA, Warsaw, Poland; Marciniak et al., 2006). The morphology and elemental composition of the solids were characterized using a FE-SEM, Quanta 200 FEG, FEI Company, now part of Thermo Fisher Scientific, Hillsboro, OR, USA equipped with EDS. Air-dried powders and crystals were mounted on adhesive tape and imaged under low vacuum without any coating.

The results of lead immobilization, expressed as a function of initial and equilibrium Pb(II) concentrations, are presented in Figure 1. These findings align with previous studies (e.g., Bajda et al., 2004; Bektaş & Kara, 2004; Günay et al., 2007; Inglezakis et al., 2007; Mozgawa & Bajda, 2005; Mozgawa et al., 2009; Oter & Akcay, 2007), confirming that clinoptilolite exhibits excellent sorption capacity for Pb(II), removing over 98% of lead from solutions containing up to 1000 mg Pb/L. At higher initial concentrations, approximately 82% and 44% of Pb(II) were removed from solutions containing 2000 mg/L and 4000 mg/L, respectively (Table 1).

Figure 1.

(A) Sorption of Pb(II) by clinoptilolite as a function of the initial Pb(II) concentration in solution. (B) Sorption isotherm of Pb(II) on clinoptilolite, illustrating the relationship between equilibrium Pb(II) concentration and the amount sorbed.

The greatest Pb uptake occurred in the experiment with an initial concentration of 4000 mg Pb/L, where 1714 mg Pb/L was removed from solution. The sorption capacity, calculated from the difference in Pb concentration before and after the experiment, reached approximately 85,700 mg Pb/kg of zeolite. The Pb-sorbed zeolite was subsequently washed repeatedly with redistilled water until the Pb concentration in the rinsate dropped below 0.1 mg/L and then air-dried. Chemical analysis of the solid revealed that the amount of non-desorbed lead retained by the zeolite was 74 g Pb/kg. Given this high uptake capacity, this material was selected for subsequent experiments.

The reaction of Pb-modified zeolite (Fig. 3a) with a solution containing AsO₄³⁻ and Cl⁻ ions at initial pH values of 2 and 7 resulted in the immediate crystallization of lead chloro-arsenate (mimetite, Pb₅(AsO₄)₃Cl; Figures 2b and 3b), effectively sequestering arsenate from the solution. Mimetite was detected by XPRD after the experiment. A comparison of XPRD patterns before and after exposure to the arsenate—and chloride—containing solution confirmed that zeolite (containing admixtures of orthoclase, quartz, and illite) and mimetite were the only crystalline phases present (Fig. 2). Mimetite precipitated heterogeneously as needle-like crystals on the zeolite surface (Fig. 3c), partially covering most of the grains and forming incrustations (Figs. 3c, d). Additionally, some mimetite crystals crystallized near the primary phase but not directly on its surface, suggesting homogeneous precipitation. These crystals reached a size of up to 2.5 μm (Fig. 3e).

Figure 2.

XPRD patterns of: (A) Natural zeolite (clinoptilolite) enriched in admixtures of orthoclase (O), illite (I), and quartz (Q), (B) Clinoptilolite after modification through sorption of Pb(II) and (C) Precipitate resulting from reaction of Pb-modified zeolite with AsO₄³⁻ aqueous solution at pH 7 (50 mg As/L). XPRD, X-ray powder diffraction.

Figure 3.

SEM images in BSE mode display the morphology of Pb-modified clinoptilolite and mimetite precipitates. (A) Pb-modified clinoptilolite, before reaction with an arsenate solution, exhibited a granular surface structure. (B) After reaction with an arsenate solution (50 mg As/L; pH 7) containing Cl⁻ ions (20 mg Cl⁻/L), mimetite (Pb₅(AsO₄)₃Cl) precipitated. (C) Lead chloro-arsenate formed as needle-like crystals distributed across the surface of the Pb—modified zeolite. (D) These needles developed as incrustations enveloping the grains of the primary Pb-clinoptilolite. (E) The mimetite needles reached lengths of approximately 2.5 μm, exhibiting characteristic hexagonal cross-sections. BSE, backscattered electron; SEM, scanning electron microscopy.

Arsenate removal varied significantly, reaching 67.7% at pH 2 but increasing to 99.88% at pH 7 within 24 hr. After 7 days, uptake at the more acidic pH improved to 74.84%, indicating a time-dependent process that may involved gradual mimetite formation, while at neutral pH, it remained unchanged (Table 2). The effect of pH can be attributed to both the stability of the mimetite (Pb₅(AsO₄)₃Cl) phase and the predominant arsenate species present in solution. Mimetite exhibits higher stability at neutral pH, remaining a solid phase that effectively sequesters arsenate from the solution (Bajda, 2010; Magalhães & Silva, 2003). At low pH, Pb-modified zeolite may have partially dissolved, releasing Pb²⁺ into the solution, which could lead to re-dissolution or hindered precipitation of mimetite, resulting in lower arsenate retention.

Furthermore, arsenate speciation may have played a crucial role in the precipitation process. At pH 7, arsenate exists primarily as HAsO₄²⁻, which readily interacts with Pb²⁺, favouring mimetite formation. In contrast, at pH 2, arsenate is predominantly in the neutral H₃AsO₄ form, which has a lower affinity for Pb²⁺, potentially limiting the precipitation reaction (Bajda, 2010; Smedley & Kinniburgh, 2002). These findings highlighted the importance of pH in controlling arsenate removal efficiency.

After 7 days of reaction in the presence of AsO₄³⁻, the lead concentration ranged from <0.02 mg/L at pH 7 to 20 mg/L at pH 2. In contrast, in the control experiment, where Pb-modified zeolite was dissolved under identical conditions but in the absence of arsenate (only Cl⁻ was present), lead release was significantly higher, reaching 43 mg/L at pH 2. This is consistent with substantial zeolite dissolution under acidic conditions. The Pb concentration remained significantly lower at higher pH levels, at 0.61 mg/L.

The ionic strength and pH of the solution influence not only the efficiency of heavy metal sorption but also the stability of zeolite and the potential re-release of contaminants into the medium (e.g., Cama et al., 2005; Deliyanni et al., 2003; Payne & Abdel-Fattah, 2004, 2005; Wilkin & Barnes, 1998). At low pH, Pb concentrations greatly exceeded the drinking water standard of 0.01 mg/L, indicating significant mobilization of this toxic heavy metal into solution. In contrast, Pb levels remained below the detection limit of the AAS method, suggesting favourable conditions for practical arsenate sequestration under neutral conditions. Additionally, since the tested solutions contained only arsenate and/or chloride ions, the release of Pb may be less significant in more complex water compositions. The increased arsenate sequestration and minimal lead release at neutral pH conditions suggest that Pb-modified zeolite could be particularly effective in arsenic-contaminated waters with moderate pH levels.

Mimetite crystallized due to the reaction of Pb(II) desorbed from zeolite with arsenate and chloride ions present in aqueous solution: 5Pb2+desorbed from zeolite+3AsO43−contamination+Cl−supplied extra=Pb5AsO43 Clprecipitated on zeolite 5{{\text{Pb}}^{2+}}_{\text{desorbed}\ \text{from}\ \text{zeolite}}+3{{\text{AsO}_{4}^{3-}}_{\text{contamination}}}+\text{C}{{\text{l}}^{-}}_{\text{supplied}\ \text{extra}}={{\text{Pb}}_{5}}{{\left( {{\text{AsO}}_{4}} \right)}_{3}}{{\text{Cl}}_{\text{precipitated}\ {\text{on}}\ {\text{zeolite}}}}

The solution likely became oversaturated immediately upon mixing, resulting in the precipitation of lead chloro-arsenate. The Pb-modified zeolite served as a nucleation substrate, providing active sites for the formation of the secondary phase. These sites facilitated local solution supersaturation with respect to mimetite (Pb₅(AsO₄)₃Cl), promoting the heterogeneous precipitation (incrustations) of new phase crystals. When the local concentrations of lead, arsenate, and chloride ions in the bulk solution in the vicinity of zeolite reached a certain level, mimetite could precipitate within the solution volume, independent of the zeolite surface. Under these conditions, homogeneous nucleation of mimetite occurred, with crystals forming in the liquid phase and subsequently precipitating in the spaces between zeolite grains rather than as a layer on the substrate surface.

The Pb-modified zeolite served as the sole source of Pb(II) required for mimetite precipitation. While lead does not desorb from the zeolite in pure water, the presence of Na⁺, As(V), and Cl⁻ in solution promoted the formation of mimetite. Lead desorption is likely facilitated by ion exchange with Na⁺ ions, which replace Pb²⁺ in the zeolite structure. Additionally, the local depletion of Pb²⁺ at the mineral-water interface near the zeolite surface, driven by mimetite precipitation, may further promote lead release from the zeolite. The reaction predominantly occurs at the zeolite surface, suggesting that the rate of mimetite crystallization exceeds that of Pb²⁺ desorption. However, both euhedral crystals of mimetite and incrustations on the surface of zeolite were observed, indicating homogeneous and heterogeneous precipitation. Initially, desorbed Pb²⁺ reacted with AsO₄³⁻ and Cl⁻ available in the vicinity of the zeolite grains, resulting in heterogeneous precipitation and the formation of mimetite incrustations on the zeolite surface (Figs. 3b–d). This suggests that the desorption of lead was slower than the precipitation of mimetite. Once the local supply of AsO₄³⁻ and Cl⁻ was depleted, desorbed Pb²⁺ ions migrated into the bulk solution, leading to homogeneous precipitation of mimetite as needle-like crystals with hexagonal cross-sections, dispersed between the clinoptilolite grains (Fig. 3e). In this case, supersaturation and subsequent precipitation occurred within the bulk solution. A similar phenomenon was reported for mimetite precipitation on goethite (Kleszczewska-Zębala et al., 2016). These observations imply that the advection of AsO₄³⁻ and Cl⁻ ions was slower than the desorption of Pb²⁺. Arsenate and chloride ions were supplied by diffusion and advection from mixing the suspension. Slow diffusion enabled the distribution of lead ions necessary for mimetite crystallization on Pb-zeolite (Kleszczewska-Zębala et al., 2016; Manecki et al., 2006).

The presence of aqueous AsO₄³⁻ and Cl⁻ and precipitation of mimetite may have induced lead desorption in two ways. Firstly, the precipitation of mimetite acted as a trap for Pb ions in the solution, altering the equilibrium between lead-bound zeolite and the surrounding solution. This could increase the Pb desorption by Le Chatelier’s principle. Furthermore, the interaction between arsenate ions and surface-sorbed lead may have weakened the bond between the lead and the surface, leading to its desorption and subsequent precipitation as mimetite.