Extraction of valuable metals from acid mine drainage by an electrochemically activated limestone system

Abstract

Acid mine drainage (AMD) contains toxic yet valuable heavy metals. Conventional chemical neutralization using hydrated lime is highly efficient in mitigating heavy metal pollution, yet it generates hazardous solid wastes and is inefficient in recovering critical metal resources. Here, we report the design and demonstration of an electrochemically activated limestone (EAL) system for AMD treatment and valuable metal extraction. The EAL system extracted 82.3-100.0% of different metals (Cu, Cd, Zn) with high purity (~81.3 wt% valuable metals) through electro-reduction and local high pH-mediated precipitation when treating simulated AMD with varied conditions and actual AMD at energy consumption of 0.006-0.180 kW·h g⁻¹ metal. In addition, the effluent after the EAL treatment contains high alkalinity and Ca²⁺, which is proven to have CO₂ sequestration potential. Overall, the EAL treatment reduces total costs by 17.9% and CO₂ emissions by 60.1% compared to conventional lime dosing treatment, demonstrating its economic viability and ecological superiority for the practical remediation of AMD and extraction of valuable metals.

Acid mine drainage (AMD) contains toxic metals that harm ecosystems. Here, Lei et al. developed an electrochemically activated limestone system to treat AMD, extracting 82.3-100% of metals with high purity, reducing costs by 17.9%, and CO₂ emissions by 60.1%.

Introduction

With the high demands for critical metals and minerals, mine wastes (i.e., waste rock and tailings) from the increasing mining practices are becoming a global ecological crisis^1–3. Contaminated drainage released from mine wastes like acid mine drainage (AMD) is one of the most critical issues, particularly given its known adverse effect on mine sites and nearby ecosystems (e.g., biodiversity loss, soil deterioration, and natural water pollution)^4,5. In China, around 7 billion tons of mine drainage are estimated to be discharged annually, with the proportion of AMD accounting for more than 50%. At the same time, the US Forest Service reported that AMD had polluted more than 14,000 kilometers of the United States’s rivers in 2010⁶. Acid mine drainage is rich in toxic yet valuable metals (e.g., cadmium (Cd), copper (Cu), and zinc (Zn)), which are critical for producing modern goods such as batteries, electric vehicles, and mobile phone^7–9. Therefore, another adverse effect of AMD is the release and waste of recoverable amounts of metal resources.

In general, AMD, resulting from the biogeochemical reactions of sulfide-bearing mine wastes when exposed to oxygen and water⁶, contains high concentrations of sulfuric acid (H₂SO₄), which facilitates heavy metal mobilization^10,11. The well-established approach for AMD treatment is chemical neutralization¹², which firstly increases solution pH and subsequently triggers the formation and precipitation of metal hydroxides or carbonates using high alkalinity reagents. Technologies based on this mechanism, such as direct lime addition and limestone (CaCO₃)-based facilities, have been widely used in active or passive AMD treatment. Nevertheless, these remediation processes typically produce heavy metal-loaded sludge that is challenging to separate from drainage through physical sedimentation or interception due to their high moisture nature¹³. In this context, most precipitated heavy metals may leach back into water bodies, severely threatening the ecosystem. Furthermore, the recovery of valuable metal elements is complicated by the extra cost required to dewater sludges and purify wastes¹⁴. Therefore, sustainable technology is urgently needed to achieve simultaneously critical metals extraction and pH amelioration for efficient AMD treatment (Fig. 1a).

Fig. 1. Electrochemically activated limestone (EAL) system for extracting heavy metals from acid mine drainage (AMD).

a Comparison and illustration of the conventional lime dosing and the EAL treatments for the remediation of acid mine drainage. Schematic showing ions migration and electrode interface reactions in the conventional electrochemical system (b) and the EAL system (c) for heavy metals-containing AMD.

In contrast to traditional chemical neutralization processes, scientists have discovered the benefits of using electrochemistry to extract metals from wastewater^15,16. The standard electrochemical method applies a direct current (DC) electric field to a solution to accumulate heavy metal cations by electric attraction^17,18. Meanwhile, interfacial metal ions are electrochemically deposited as zero-valent metallic states through electron transfer on the negative electrode^19–21. Besides, recent studies find that the OH⁻ generated from cathodic water reduction (2H₂O + 2e⁻→H₂+2OH⁻) can create a relatively high pH area near the electrode, which is beneficial to the precipitation of metal cations, further improving the deposition of heavy metals²². However, the practical application of electrochemical technology in AMD treatment is restrained by the substantial energy input required to eliminate the high acidity of AMD via consuming the cathodic electrons (hydrogen evolution reaction, HER: 2H⁺ + 2e⁻→H₂) and cathodically produced OH⁻ (H⁺ + OH⁻→H₂O) (Fig. 1b). Meanwhile, the protons generated from anodic water oxidation (oxygen evolution reaction, OER: 2H₂O→4H⁺ + O₂+4e⁻) are continuously supplied into the solution, further limiting the efficiency of heavy metal extraction. The current state-of-the-art electrochemical technologies focus on developing various modified methods to improve the metal extraction performance or separate the H⁺ from cathodic reduction reaction by using novel electrode materials^23,24, modifying electrochemical reactors with ion exchange membranes^25,26, and changing applied current/potential strategies^27,28. However, three concerns hinder the practical application of these methods: the high cost of novel materials and membranes, the complex operation for field application, and the lack of efficiency in alleviating wastewater pH.

Compared with the above-mentioned alternative methods, continuously neutralizing H⁺ from AMD and electrochemical water splitting was more straightforward and practical. Highly soluble chemical neutralizers like slaked lime and soda ash, though widely used in water treatment, are expensive (e.g., the price of soda ash is over US $450 ton⁻¹), and directly adding them before or during AMD’s electrochemical remediation will cause rapid precipitation of heavy metal ions due to the quick release of alkalinity, limiting the recovery of heavy metals²⁹. In contrast, limestone, mainly composed of calcium carbonate (CaCO₃), has been widely used in active or passive AMD treatment and other wastewater remediation^10,30. It has advantages like inexpensive (less than US $100 ton⁻¹)¹⁰, easy-to-obtain³¹, eco-friendly, and, more importantly, controllable release of alkalinity^32,33. As a result, limestone can perfectly address the efficiency limitation resulting from protons during electrochemical AMD remediation and metal extraction. However, the direct use of limestone is of low efficiency and sluggish kinetics in H⁺ consumption. Furthermore, integrating it directly with conventional electrochemical reactors cannot achieve the expected cathodic metal extraction due to the limestone’s surficial metal chemisorption reaction (e.g., Cd²⁺ + CaCO₃→Ca²⁺ + CdCO₃).

Here, we report an electrochemically activated limestone (EAL) system (Fig. 1c) to achieve simultaneous acidity amelioration and heavy metals extraction toward AMD treatment. Our system proposed an anode-specifically packed limestone that, in addition to directly neutralizing AMD, limestone can focus on consuming the anodically produced H⁺, which not only prevents H⁺ from consuming cathodic electrons and alkalinity but also improves limestone’s utilization and avoids its surficial metal adsorption. Meanwhile, under the work of the electric field, metal ions migrate onto the negative cathode, being directly deposited or precipitated on the cathode surface due to electrochemically created local high pH via hydrogen evolution reaction. The EAL system achieved effective pH amelioration and over 82.3–100.0% extraction of different heavy metals (copper, cadmium, zinc) using simulated AMD with varied conditions and actual AMD. The corresponding Technico-economic analysis (TEA) and Life Cycle Assessment (LCA) indicate that the EAL treatment is 17.9% cheaper and emits 60.1% less CO₂ than the conventional lime dosing treatment. This research potentially provides a sustainable avenue to achieve metal recovery alongside AMD remediation.

Results

Design and performance of the EAL system

The EAL system (Fig. 2a and Supplementary Fig. 1) involves an anodic packing column and a cathodic extraction area. The porous packing column is filled with particle limestone (CaCO₃) (Supplementary Fig. 2 and Supplementary Table 1), in which the rod-shaped Ru/Ir-coated titanium anode is fixed at the center of the porous column, vertically against the tubular stainless-steel cathode. Thanks to the structure of the packed anode system, the cathodic extraction area is separated from the neutralization zone. We first assessed the pH amelioration process of the EAL system in simulated AMD treatment by applying a current density of 2.4 A m⁻² and filling limestone particles in a solid-to-liquid ratio (g: L) of 50.

Fig. 2. Treatment performance of the EAL system.

a Design and structure of the EAL system. b, c Changes of pH (b), Ca ions, and dissolved inorganic carbon (DIC) concentrations after 5-h (data presented as mean ± s.d. (n  =  3)) (c) between the EAL treatment and two other treatments on treating simulating AMD without heavy metals. ‘Limestone’, directly adding limestone with the same solid-to-liquid ratio as the EAL treatment to the simulated AMD. ‘No filling’ uses the same conditions as the EAL system, except no limestone is assembled in the anode area. Current density, 2.4 A m⁻². d Comparison of Cd concentration between the EAL treatment and two other treatments on treating simulated AMD spiked with 20 mg L⁻¹ Cd (data presented as mean ± s.d. (n  =  3)). e Distribution of Cd after the three treatments. f Cathodic extraction and specific energy consumption of the EAL system after 4 h of treatment on simulated AMD spiked with varied concentrations of Cd (2–200 mg L⁻¹) (data presented as mean ± s.d. (n  =  3)). Current density, 2.4 A m⁻². g Cathodic extraction and specific energy consumption of the EAL system after 2.5 h, 4 h, and 15 h treatments on Cd-containing simulated AMD (20 mg L⁻¹) with varied initial pH (2.0–4.0), respectively (data presented as mean ± s.d. (n  =  3)). Current density, 2.4 A m⁻². h Cathodic extraction and specific energy consumption of the EAL system on treating simulated AMD spiked with varied heavy metals (20 mg L⁻¹ Cu, Cd, and Zn) (data presented as mean ± s.d. (n  =  3)). Current density, 2.4 A m⁻².

Figure 2a shows that the electrolysis treatment with no limestone filling failed to ameliorate pH due to the simultaneous water splitting and water combination in bulk solution. In contrast, the limestone alone treatment could only increase the solution pH from 3.0 to 7.5 through carbonate neutralization. However, the EAL system is most efficient in neutralizing AMD and raising bulk pH among the three systems (Fig. 2b), and the released Ca²⁺ and dissolved inorganic carbon (DIC) concentrations were also significantly higher than in the limestone alone system (Fig. 2c). Notably, the local pH at the anode remained around 3.0 due to the oxygen evolution reaction, while the bulk solution pH increased to 9.7 within 5 h. This indicates that the anode-filled limestone in the EAL system reacted with the H⁺ from OER, reducing its diffusion into the bulk solution. This reaction promotes OH⁻ accumulation via HER, thereby increasing both the solution and cathodic local pH, which is advantageous for subsequent metal extraction processes. Furthermore, while CO₂ may be generated from the dissolution of CaCO₃ at the anode under low pH conditions³⁴, the observed molar ratio of DIC to Ca²⁺ after 5-h treatment was 1.4, aligned with a previous study³⁵. The DIC concentration under open-air conditions was 67.3% higher than under N₂ conditions (without CO₂). These findings suggest that the treated AMD can capture CO₂ not only from the dissolution of CaCO₃ but also from the surrounding air.

We then investigated the effect of current densities on the EAL system’s effectiveness (Supplementary Fig. 3a). The Cd removal efficiency increased significantly from 3.0% to 99.0% as the current density rose from 0.0 to 2.4 A m⁻². However, further increasing the current density to 4.8 A m⁻² did not enhance the removal efficiency and led to an excessively high final pH (Supplementary Fig. 3b). Considering energy consumption and environmental factors, subsequent experiments were performed at an optimized current density of 2.4 A m⁻².

We subsequently explored the merits of the EAL system in targeting Cd-containing simulated AMD without limestone or power supply. Figure 2d illustrates that electrolysis with no limestone filling showed a poor Cd removal efficiency of only 2.7% in 4 h under 2.4 A m⁻². In addition, the bulk pH of the simulated AMD was maintained at 3.0 (Supplementary Fig. 4). Likewise, while the limestone alone treatment (solid-to-liquid ratio, 50 g L⁻¹) neutralized the wastewater pH to 6.7 and released 54.3 mg L⁻¹ Ca²⁺, it failed to remove Cd from the simulated AMD. However, with conditions identical to those of the above two systems, the EAL system successfully increased bulk pH from 3.0 to 9.7. During the pH amelioration process, more than 99.0% of Cd was removed from the AMD, meeting the discharge limit value (0.1 mg L⁻¹)²⁵. Analyzing the distributions of Cd in different treatments implies that the limestone alone treatment mainly removed Cd through adsorption and coprecipitation, while the EAL treatment reduced the Cd coverage on the limestone (Fig. 2e). Specifically, the EAL system successfully extracted 96.8% Cd onto the cathode, 35.9 times higher than the electrochemical treatment without limestone. This might be attributed to two processes: (1) the electric field drove the Cd²⁺ from the solution to the negative cathode, which reduced the contact between Cd²⁺ and limestone³⁶; (2) the H⁺ released from anodic water oxidation continuously reacted with limestone and prevented them from surficial precipitation of Cd²⁺. Likewise, the outstanding performance of the EAL system compared to direct electrolysis (without limestone filling) suggests that the limestone filling in the EAL system could constantly consume the H⁺ ions in AMD and H⁺ produced via OER, which eliminates the inhibition of H⁺ on cathodic metal electrodeposition. It is worth mentioning that the EAL system not only achieves pH amelioration and metal removal but also separates removed metals (mainly on the cathode) with chemical neutralization agents, such as limestone. This is not possible by direct chemical neutralization, in which the metals are removed but mixed with chemical agents, thus making it challenging to extract.

Then, we further evaluate the capacity of the EAL system in treating simulated AMD with different metal concentrations (2, 20, and 200 mg L⁻¹ Cd). The concentration range covers most reported Cd concentrations in real mine drainages³⁷. Regardless of the initial Cd concentration, the EAL system exhibited substantially high Cd extraction efficiency and pH amelioration ability (Supplementary Fig. 5). The Cd extraction capacity of the EAL system increased significantly with initial Cd concentration. Consequently, the specific energy consumption decreased drastically from 0.18 kW h g⁻¹ Cd at 2 mg L⁻¹ Cd²⁺ to 0.03 kW h g⁻¹ Cd at 20 mg L⁻¹ Cd²⁺ (Fig. 2f). With an initial concentration of 200 mg L⁻¹, the EAL system achieved a high Cd extraction efficiency of 68.0% with a very low specific energy consumption of 0.006 kW h g⁻¹ Cd after 4 h of operation without showing a saturation trend (Fig. 2f and Supplementary Fig. 5a).

Furthermore, we looked at the influence of initial pH on Cd extraction from the AMD. The results show that the treatment duration and specific energy consumption of the EAL system are higher when treating more acidic AMD under the same operating conditions (Supplementary Fig. 6a and Fig. 2g), which is reasonable since the high H⁺ concentrations favor HER that competes with Cd²⁺ ions for electrons. However, by extending the electrolysis time, the EAL treatment could raise the solution pH from 2.0 to 7.8 and extract 98.0% of Cd. This suggests that the ability of the EAL system to achieve pH amelioration and Cd extraction mainly depends on the transferred number of cathodic electrons. Therefore, in the EAL system, we can play with the current density or reaction time to supply sufficient electrons for acidity elimination and extraction of heavy metals in treating AMD with varied pH.

On top of Cd²⁺, we used the EAL system to treat two other simulated AMD spiking with 20.0 mg L⁻¹ Cu and Zn, also common valuable metal elements in AMD^37–39. We also compared its performance with the limestone alone treatment and the electrolysis treatment with no limestone. Figure 2h and Supplementary Fig. 7 illustrate that electrolysis alone treatment (without limestone) can only achieve sufficient Cu removal. Specifically, the Cu concentration in the AMD was reduced from 20.0 to 0.3 mg L⁻¹ after 2.5 h treatment at 2.4 A m⁻². In contrast, it performed poorly in removing Cd and Zn under the same conditions. According to the electrochemical motive force (EMF) series and the simulation using HSC Chemistry 6.0^40,41 (Supplementary Fig. 8), this phenomenon might be ascribed to the fact that the electrolysis system could directly reduce and deposit Cu²⁺ with high redox potential into the metallic state (E(Cu²⁺/Cu⁰) = +0.342 V, versus standard hydrogen electrode (SHE))⁴², but unable to achieve the deposition of heavy metal cations with low redox potential (i.e., Cd (E(Cd²⁺/Cd⁰) = −0.403 V, versus SHE) and Zn (E(Zn²⁺/Zn⁰) = −0.762 V, versus SHE) due to the high concentration of H⁺ (Supplementary Figs. 4 and 6d).

In contrast, the EAL system increased the pH of three simulated AMD from 3.0 to 7.0-9.5, effectively reducing the concentrations of Cu, Cd, and Zn below their limit values. More importantly, Fig. 2h shows that the EAL system successfully extracted 96.9% of Cu and 82.3% of Zn onto the cathodes, 20% and 101.9 times higher than the electrolysis treatment (Supplementary Fig. 7f). The results presented herein clearly demonstrate the feasibility and broad applicability of the EAL system in achieving simultaneous pH amelioration and metal resource recovery toward AMD treatment.

Extracted heavy metals and possible extraction pathways

After electrolysis and EAL treatments, we further characterized and compared the heavy metal species on the stainless-steel cathodes to explore the heavy metal extraction mechanism. Many micro-sized particles are formed on the cathode surface in the EAL system after treating three heavy metals-containing AMD (Supplementary Fig. 10). In contrast, in the absence of limestone filling, we only found micro-sized particles in treating Cu-loading AMD. The cathode surface remained smooth after treating Cd- and Zn-loaded AMD, suggesting an absence of metal deposition. Subsequently, we probed the elemental composition of the electrodes’ surface after AMD treatment (Fig. 3a–c). The EDS analysis detects a significant Cu signal on the cathode after electrochemical treatment of Cu containing AMD with or without filling limestone. The other elemental compositions were similar to the raw cathode (Supplementary Fig. 9). This agrees with our hypothesis that copper ions can be deposited onto the cathode as metallic copper in AMD, given its higher redox potential (0.342 V). However, in treating Cd- and Zn-loaded AMD, the cathode, after electrolysis without CaCO₃ filling, exhibited similar element compositions as the raw electrode with Cd and Zn signals below the detection limit, implying an absence of metal deposition.

Fig. 3. Characterizations of extracted heavy metals in three heavy metals-containing solutions after the EAL and electrolysis treatment.

a–c EDS results of extracted heavy metals in three heavy metals-containing simulated AMD after the EAL and electrolysis treatment; d–f XPS results of extracted heavy metals in three heavy metals-containing simulated AMD after the EAL and electrolysis treatment; g–i XRD spectra of extracted heavy metals in three heavy metals-containing simulated AMD after the EAL and electrolysis treatment.

In contrast, the cathode surface after the EAL treatment appeared to have firm peaks of Cd and Zn. Meanwhile, detecting significant C and O signals suggests the formation of Cd and Zn hydroxides or carbonates on the cathode. Considering the low redox potentials of Cd (−0.403 V) and Zn (−0.762 V) ions, the EAL extraction might be due to the precipitation at the high pH zone near the cathode after increasing the solution pH^43,44. To verify this hypothesis, we conducted an XPS analysis to determine the chemical state of the heavy metals extracted by the electrolysis and the EAL systems (Fig. 3d–f). The XPS Cu 2p showed peaks at 932.8 and 952.6 eV with weak satellite peaks for Cu extracted with or without limestone, indicating that most copper species on the cathode were in a metallic state²⁸. XRD-GI spectra further confirmed the formation of metallic Cu on both systems’ cathodes (Fig. 3g). For the XPS spectra of Cd and Zn, Cd 3d and Zn 2p peaks observed in the electrolysis treatment without limestone indicated divalent states of Cd and Zn. Conversely, XPS analysis of the EAL treatment showed that the extracted Cd and Zn species were a mixture of metallic and divalent states, implying that the EAL extraction process involved electrochemical reduction and precipitation. XRD-GI characterization confirmed that the cathodic deposits after treating Cd-loaded AMD were metallic Cd (PDF#01-072-8273), and CdCO₃ (PDF#97-003-3662) (Fig. 3h). Given the poor XRD spectrum collected from the cathode after treating a low concentration of Zn (20 mg L⁻¹), we conducted additional XRD measurements on the cathode after treating simulated AMD with a high Zn concentration (200 mg L⁻¹) (Supplementary Fig. 11). The XRD results reveal that the deposited product on the electrode matches the PDF card of Zn₅(OH)₆(CO₃)₂ (PDF#01-072-1110) (Fig. 3i). These findings explain the superior performance of the EAL system in extracting low redox potential heavy metals.

The difference in extraction mechanisms between the two systems is primarily due to the change in pH (H⁺ concentration). During the electrolysis treatment without limestone filling, the high proton concentration constantly consumes cathodic electrons, which inhibit the reduction of the heavy metals with low redox potential (i.e., Cd and Zn). In contrast, the EAL system effectively consumed protons, eliminating its competition towards cathodic electrons and forming the cathodic high pH area, finally facilitating heavy metal reduction and precipitation.

Taking Cd-containing AMD as an example, we evaluated the properties of deposits on the cathode over varied time frames. The EDS results demonstrated that the surface of the cathodes after 2 h (intermediate) and 4 h (final) treatments have similar elemental compositions (Supplementary Fig. 12 and Fig. 3b), while the relative contents of elements differed significantly (Fig. 4a). The relative atomic contents of C (11.4%) and O (13.8%) on the cathode after 4 h of treatment were significantly higher than those at 2 h of treatment, implying that the Cd extracted at the later stage was mainly in the form of carbonate or hydroxide precipitates. These results also suggest that Cd ions could be extracted in the early stage via simultaneous electro-reduction and local pH-mediated precipitation, which opens the door for possible selective extraction. In addition, at the later stage, the Cd extraction primarily depends on precipitation due to the increasing bulk pH from 6.7 to 9.7 and carbonate accumulation resulting from limestone neutralization. We further investigated this hypothesis by conducting XPS and XRD characterizations on two samples (Fig. 4b, c and Supplementary Fig. 13). After the initial 2-h treatment using the EAL system, the Cd content extracted on the cathode consisted primarily of 38.1% Cd(0) and 61.9% Cd(II). Concurrently, the XRD spectra of this sample exhibited a distinct signal matching the standard PDF card for metallic cadmium (PDF#99-000-0525). However, following a 4-h treatment, the cathode’s relative Cd(0) content decreased to 23.1%, with Cd(II) increasing to 76.9%. Furthermore, the XRD spectra displayed prominent peaks corresponding to otavite (CdCO₃, PDF#99-003-3662), whereas the peaks attributed to zero-valent Cd became less pronounced. These exciting results further evidence the sequence of the electro-reduction and local pH-mediated precipitation in treating low redox potential metals loaded AMD by the EAL system. Additionally, at the cathode after 4-h treatment, XPS results demonstrated a weak Ca 2p signal, mainly in CaCO₃ (Supplementary Fig. 14), agreeing with the minimal calcium content (0.9%, Fig. 5c) in the cathode deposition products.

Fig. 4. Possible extraction pathway of the EAL system.

a Elemental compositions of the cathodes at the half and whole reaction time of Cd extraction from simulated AMD. b Comparison of chemical states of the extracted Cd at the half and whole reaction time. c XRD-GI spectra of the cathodes at the half and whole reaction time. d A schematic representation of the EAL system for heavy metals extraction from AMD (E_h represents the electrode potential of the cathode and the redox potential of cations in solution).

Fig. 5. Extracted product evaluation and effluent carbon capture capacity.

a Chemical neutralization treatment (limestone dosing) and precipitates composition after treating mixed metals-containing simulated AMD (data presented as mean ± s.d. (n  =  3)). b The combined electrolysis and limestone treatment and cathodic deposits composition after treating mixed metals-containing simulated AMD (data presented as mean ± s.d. (n  =  3)). Current density, 2.4 A m⁻². c EAL treatment and cathodic deposits composition after treating mixed metals-containing simulated AMD (data presented as mean ± s.d. (n  =  3)). Current density, 2.4 A m⁻²; d Changes of pH, Ca²⁺ concentration, and DIC concentration during the EAL extraction and carbon capture (data presented as mean ± s.d. (n  =  3)).

Moreover, we examined system operation using a three-electrode setup with an Ag/AgCl reference electrode. In the EAL system treating simulated AMD without heavy metals, the initial working voltage ranged from 3.0 to 3.1 V and increased to 3.6–3.7 V after 1 h (Supplementary Fig. 15). Correspondingly, the initial cathode potential was −0.23 V, which dropped to −0.57 V during the 1–2 h period and then rose back to −0.4 V after 2 h. The observed drop in cathode potential during the 1–2 h period is likely due to the accumulation of hydrogen ions at the cathode (the pH near the cathode is lower than the solution pH at the beginning of 1 h, Fig. 1b), resulting in concentration polarization and a subsequent decrease in potential. As the pH increased and the hydrogen ion concentration decreased, the cathode potential increased.

In the EAL system treating simulated AMD containing Cd, the cathode potential behavior between 0–1 h was similar to that observed in the simulated AMD without Cd. However, between 1–2 h, the cathode potential did not drop to −0.57 V but instead fluctuated between −0.4 V and −0.3 V. This behavior can be attributed to the reduction of Cd²⁺ at the cathode, as confirmed by XPS results (Supplementary Fig. 13), which indicated that cathodic Cd reduction occurred during the 0–2 h period. The reduction of Cd²⁺ at the cathode consumes electrons, counteracting the concentration polarization effect and preventing the cathode potential from dropping to −0.57 V.

Based on the above systematic investigations, we proposed a plausible heavy metal extraction mechanism of the EAL system (Fig. 4d). First, aqueous H⁺ and metal cations migrate onto the cathode under the work of the electric field. Metal cations with high redox potentials are rapidly reduced and deposited on the cathode. Meanwhile, bulk H⁺ ions are simultaneously consumed by limestone neutralization and cathodic reduction. Second, as the concentrations of high-redox potential heavy metals and H⁺ decrease, the heavy metal cations with low redox potential begin to gain cathodic electrons to achieve electro-reduction and electrodeposition. At the same time, the generation of OH⁻ ions from HER raises the local pH at the cathode surface, facilitating the formation of heavy metal precipitates. Third, as the overall solution pH rises and the cathodic potential further declines, the same heavy metal cations with low redox potential remain, which cannot be directly reducible. At this juncture, the release of carbonate ions from the limestone neutralization reaction at the anode, along with the high pH environment formed at the cathode through water electrolysis, collaboratively promotes the formation of stable metal carbonates on the cathode, thereby achieving the extraction of heavy metal cations with varied electrochemical characteristic and solubility.

Evaluation of deposited products and carbon capture potential

The metal content and purity are essential for subsequent industrial valorization. We compared metal extraction efficiency and quality of recovered products in limestone alone treatment, combined electrolysis and limestone dosing, as well as the proposed EAL treatment (Fig. 5a–c). As shown in Supplementary Fig. 16, limestone alone treatment (solid-to-liquid ratio, 50 g L⁻¹) is insufficient in extracting metals, with a negligible removal efficiency of all three metals (<3.2%). This could be due to the relatively large particle size and surface passivation, which limits AMD pH amelioration by directly dosing limestone. Besides, the mechanism of heavy metals removal is mainly through alkali precipitation, which results in the mixture of valuable metals and dosed limestone, causing the final precipitates with an extremely low content of three metals (Fig. 5a).

In the electrolysis combined with limestone dosing system, under a current density of 2.4 A m⁻² and the same limestone dosage as limestone alone treatment, the pH only increased to 6.2 after 5 h, and the concentrations of Cd and Zn could not be effectively reduced to discharge standards (Supplementary Fig. 17). More importantly, only 78.1% of Cu, 75.4% of Cd, and 58.5% of Zn were deposited on the cathode surface, and 21.9% of Cu, 19.6% of Cd, and 29.4% of Zn formed precipitates in the solution, mixed with the added limestone, resulting in low purity of the cathodic deposits, ranging from 16.1 wt% to 21.0 wt% (Fig. 5b).

In contrast, our EAL system concentrated limestone in the anode region, effectively consuming the H⁺ released by the oxygen evolution reaction. The system increased the solution pH from 3.0 to 9.3 (Supplementary Fig. 18b) and reduced the concentrations of Cu, Cd, and Zn to levels below discharge limits (<0.1 mg L⁻¹) (Supplementary Fig. 18a). More importantly, over 95.0% of Cu, Cd, and Zn were extracted from the solution and deposited on the cathode, significantly improving the purity of the cathodic products (27.0 wt% of Cu, 27.4 wt% of Cd, and 26.9 wt% of Zn). This demonstrates that the EAL system, by filling limestone in the anode region, effectively reduced the impact of hydrogen ions on cathodic deposition, enhanced metal extraction efficiency, ensured the metal purity of the deposits, and minimized the mixing of metals with limestone.

After metal extraction by the EAL treatment, the SEM images of the cathodes showed many micro-sized particles (Supplementary Fig. 20a), while the EDS result suggests that these particles are mainly composed of elements Cu, Cd, Zn, C, and O (Supplementary Fig. 20b). Besides, XRD-GI characterization identified that these micro-sized particles were Cu/Cd metallic alloy (PDF#97-8956) and Zn₅(OH)₆(CO₃)₂ (PDF#01-072-1100) (Supplementary Fig. 20c). This evidence further demonstrated that the EAL system can effectively extract heavy metals through reduction and precipitation in mixed heavy metals loaded AMD.

Beyond metal recovery, given the increase of pH and hardness ions (2.4 mM Ca²⁺ and 0.4 mM Mg²⁺), we explored the benefits of the EAL system in direct air CO₂ capture (DAC) at room temperature (22.0 ± 2.0 °C). As illustrated in Fig. 5d, after 15 h of exposure to air, the pH of the solution decreased from 9.3 to 8.8, while the concentration of inorganic carbon increased from 4.1 to 4.8 mM. Since the effluent was filtered through a 0.45 μm filter, the increased inorganic carbon should not come from any solids. Instead, it probably results from CO₂ entering the solution from the air and reacting with OH⁻ to form HCO₃⁻ (CO₂+OH⁻→HCO₃⁻)⁴⁵, according to the total charge (cation, anion) neutralization (Supplementary Note 1). Meanwhile, the concentration of calcium ions decreased from 2.4 to 2.2 mM, indicating that some of the Ca²⁺ began to form precipitates under alkaline conditions (Ca²⁺ + HCO₃⁻→CaCO₃+H⁺)^46,47. At the subsequent 80 h of reaction, the DIC concentration decreased from 4.8 to 4.6 mM and finally achieved equilibrium, while the concentration of Ca²⁺ declined gradually to 2.0 mM. This is primarily due to the simultaneous precipitation of CaCO₃ and the direct capture of CO₂ from the air. Meanwhile, the solution pH gradually decreased and stabilized at 8.3, within the permissible discharge range (pH 6.0–9.0).

Furthermore, we collected and characterized the bulk precipitates after DAC. The SEM images illustrated that the precipitates had needle, columnar, or fibrous shapes (Supplementary Fig. 20a), and the EDS results (Supplementary Fig. 20b) suggested they mainly consisted of Ca, C, and O elements. XRD characterization (Supplementary Fig. 20c) further clarified that the micro-sized particles were aragonite, CaCO₃ (PDF#99-00013), proving the formation of calcium carbonate precipitates during the DAC experiment. These results demonstrate that the EAL system, on top of extracting high-purity valuable metals from AMD, has the potential to capture CO₂ from the air, ultimately restoring the wastewater to natural water levels.

Extraction of heavy metals from iron-rich and real AMD

In AMD, high concentrations of ferrous ions (Fe²⁺) are expected. On exposure to air, Fe²⁺ is anticipated to be oxidized to ferric (Fe³⁺) by dissolved oxygen and generate ferric hydroxide (Fe(OH)₃) with the increase of pH during treatment^10,48. The generated Fe(OH)₃ has a known ability to remove heavy metals via adsorption, flocculation, and (co-)precipitation, which potentially affects the heavy metal extraction performance of the EAL system⁴⁹. Therefore, we clarified the influence of Fe²⁺ in the established EAL system by treating the simulated iron-rich AMD (spiked with 20 mg L⁻¹ Cu, Cd, Zn, and 200 mg L⁻¹ Fe²⁺). Figure 6 shows that extended treatment durations are required to reduce the heavy metal concentrations to below discharge limits, compared to the absence of Fe²⁺. This is due to the competition for OH⁻ by iron ions for forming iron hydroxides (Fig. 6b). Notably, the generation of iron sludges showed no or a slight impact on the extraction of heavy metals. Figure 6c exhibited that the EAL system still could extract ~100.0% of Cu, ~100.0% of Cd, 94.3% of Zn, and 44.5% of Fe onto the cathode under a current density of 2.4 A m⁻², demonstrating the promising performance of the EAL system in extracting heavy metals from AMD in the presence of Fe²⁺. Most importantly, it is worth mentioning that the coexisting Fe ions do not significantly interfere with valuable metal extraction at the cathode, as the iron sludge mainly forms in bulk wastewater (45.9% of Fe).

Fig. 6. Performance of the EAL system in treating high ferrous simulated AMD and real AMD.

a Concentrations of heavy metals in iron-rich AMD during the EAL extraction. b Changes in pH, Ca, and Fe concentrations during the EAL extraction. c Distributions of Cu, Cd, Zn, and Fe in iron-rich AMD after the EAL extraction. d Changes of ion concentrations in real acid mine drainage before and after the EAL treatments. e Changes of pH in real acid mine drainage before and after the EAL treatments. f Distributions of metal ions and rare earth elements in real AMD after the EAL treatments. Conditions: current density = 2.4 A m⁻². g Costs for the conventional lime dosing and the EAL treatments. h CO₂ emission, human toxicity, and ecotoxicity for the conventional lime dosing and the EAL treatments.

Furthermore, we conducted a continuous operation under the same conditions to evaluate the long-term performance of the EAL system. During 480 h of treatment, the EAL system demonstrated a stable heavy metals removal (>50%) and pH amelioration ability with consistent voltage changes (Supplementary Fig. 21a), suggesting that the system is highly stable in both decontamination and energy consumption under long-term operation. Besides, with long-term operation, the metal deposits grow into large volumes. They could be efficiently collected through physical scraping methods without compromising the integrity of the cathode (Supplementary Fig. 21b). Hence, the cathode can be readily cleaned and reused directly after collecting deposits, thereby recycling the electrode. More importantly, due to the effects of cathodic reduction and local high pH-induced precipitation, the cathodic metal deposits mainly in the form of metallic states and hydroxides or carbonates precipitates, which effectively decreased the mixture of alkali metals with low redox potential (e.g., Ca, Mg) or high solubility (e.g., K, Na) and suspend solid like SiO₂. Therefore, the products obtained from the EAL process could exhibit a higher purity (Fig. 5) of metals compared to the precipitates collected from conventional hydrated lime treatment (Supplementary Fig. 19 and Note 2) and conventional mineral ores used in metal extraction industries⁵⁰. This further suggests that cathodic products collected in the EAL system have a high potential for direct use in critical metal production industries, and its relatively high critical metal content further makes it easier to separate during specific metal extraction processes and reduces the generation of waste materials.

We further test the performance of the EAL system in natural acid mine drainage to demonstrate its performance in realistic applications. The AMD is collected downstream from a copper-sulfur tailings pond in Shaoguan, Guangdong Province, China (Supplementary Table 3). By applying a 2.4 A m⁻² of current density, the EAL system successfully reduced the concentrations of metal elements, particularly the heavy metals (i.e., Cu, Cd, and Zn) to below the limited value (Fig. 6d), with the wastewater pH ameliorated from 2.5 to 7.8 (Fig. 6e). In addition, the EAL system also simultaneously reduces the concentrations of aluminum, iron, manganese, and rare earth elements (REE, e.g., yttrium, lanthanum, cerium), further achieving wastewater purification. It is noteworthy that, although the acid mine drainage was rather complex with many coexisting metal ions, the EAL system could still extract valuable metal ions, that is, 99.5% of Cu, 95.8% of Cd, 85.8% of Zn, and 61.8% of REE onto the cathode (Fig. 6f), indicating the effectivity of the EAL system in actual AMD application and valuable resources recovery. The concentration of removed heavy metals and REE on the cathode surface provides an easy way to separate these valuable ions from complicated wastewater, making future upcycling more promising.

Technico-economic analysis and life cycle assessment

To assess the economic feasibility of the EAL system, a comparative techno-economic analysis of the EAL system and conventional chemical neutralization techniques was conducted (Supplementary Note 3 and Supplementary Tables 3, 4). The EAL treatment at a commercial scale has not been demonstrated. Therefore, we estimate material use, energy use, and cost by scaling up laboratory processes for a hypothetical processing plant designed to treat an acid mine drainage with a calculated acidity of 500 mg L⁻¹ (total as H₂SO₄, equal to 510.2 mg CaCO₃ L⁻¹ (Eq. (2)) and a flow rate of 190 m³ h⁻¹ operating over a 20-year lifetime as estimated in previous studies^51–53. The details of the two process circuits are illustrated in Supplementary Fig. 22. Figure 6g shows that the total cost of conventional lime treatment is approximately US $72,587,000. In comparison, the price of the EAL treatment is estimated to be approximately US $61,583,000, which is 17.9% lower than the conventional lime treatment. This advantage is expected to be more significant considering the value of the recovered metals and the avoided pollution from heavy metal sludge.

Utilizing metal extraction for AMD treatment represents a shift from traditional chemical neutralization to electrically mediated resource recovery (54.0% of the total cost used for electricity, Fig. 6g). To contextualize this shift, we performed and compared the life-cycle assessment (LCA) for the EAL treatment and for conventional lime dosing systems that use high environmental footprint hydrated lime and produce a large volume of hazardous sludge (Supplementary Tables 6 and 7). For the conventional lime dosing treatment, the global warming potential (GWP), human toxicity (HT), and ecotoxicity (ET) were estimated to be 1.33 kg of CO₂ equivalent, 1.05 × 10⁻⁵ of critical toxicity unit for humans (CTUh), and 251.07 of critical toxicity unit for ecosystems (CTUh), respectively (Fig. 6h). With the EAL treatment, nearly all the environmental impacts were significantly reduced. CO₂ emission,  HT, and ET dropped to 0.53 kg, 9.06 × 10⁻⁶ of CTUh, and 217.05 of CTUh, respectively, which decreased GWP by more than 60.1%, HT by 13.7% and ET by 13.6% when compared with those of conventional lime dosing treatment.

Importantly, we performed LCA and TEA analysis based on the experimentally measured voltage values, which include the energy consumed for the side reactions of electrodes and additional resistance introduced by the anodic-packed limestone. Therefore, we are confident that the EAL system holds significant potential to provide even greater advantages in AMD treatment and metal resource recovery with further research and parameter optimization (e.g., using more efficient electrode materials, optimizing anode filling structure).

Discussion

Modern society heavily relies on mining to extract metal resources to produce electronics, i.e., mobile phones, batteries, and electronic cars. Mining is often energy-intensive and environmentally unfriendly, i.e., producing large quantities of acidic mine drains. Today’s strategy for AMD treatment is mainly about dosing lime, which produces toxic heavy metal-loaded hazardous waste that requires careful treatment and wastes strategic metal resources. However, we regard AMD as a pool for heavy metal extraction, moving from simply AMD remediation to metal mining from AMD. To this end, we proposed a technique-feasible, and even carbon-negative approach that combines conventional lime dosing with electrowinning, establishing an electrochemically activated limestone system. Innovatively, we designed an electrochemical cell that allows precise control of lime dosing strictly in the anode region. Unlike conventional electrochemical reactors, the EAL system incorporates a chemical proton consumption path and utilizes the local pH at electrodes, which can simultaneously ameliorate pH and extract heavy metals from acid mine drainage under varied conditions. In this way, we ensure that the removed valuable metals are not mixed with limestone or low-valued iron sludge, making metal extraction highly possible. More interestingly, the EAL system is proven effective in recovering critical rare earth elements during the actual AMD treatment process. Beyond the excellent resource extraction performance, the EAL-treated AMD also contains high concentrations of calcium ions and has the potential for CO₂ sequestration from the air, the potential of which is yet to be explored further. Our TEA and LCA indicated that the EAL treatment could reduce total cost by 17.9% and CO₂ emissions by 60.1% compared with the conventional lime dosing treatment. The neutralized water and heavy metals obtained from the EAL system could potentially offset industrial production activities. Overall, our established EAL system shows technical, economic, and ecological superiority in extracting valuable metals from AMD. We expect this system to inspire innovations in other heavy metal-rich waste stream management and contribute to the United Nations sustainable development goals (SDGs) by securing access to clean water (SDG 6) and promoting sustainable industrial production (SDG 12).

Methods

Materials and reagents

The Ru/Ir-coated titanium anode was purchased from Wenghe Metal Materials Co., Ltd (Hefei, China). The stainless-steel cathode was provided by Maize Wire Mesh Product (Anping, China). Analytical reagents of Cadmium sulfate heptahydrate (CdSO₄·7H₂O), sodium sulfate (Na₂SO₄), cooper sulfate (CuSO₄), zinc sulfate (ZnSO₄), ferrous sulfate (FeSO₄) and sodium hydroxide (NaOH) were purchased from Macklin Biochemical Co., Ltd (Shanghai, China). Sulfuric acid (98% H₂SO₄) and nitric acid (65%–68% HNO₃) were bought from Sinopharm Chemical. Limestone particles were purchased from Jinxin Mineral (Xuancheng, China). The simulated AMD used in this study was prepared by spiking 30 mM Na₂SO₄ with specific concentrations of heavy metal cations, and its initial pH was adjusted to 3.00 ± 0.05 using 1 M H₂SO₄. This preparation of the simulated AMD is based on the previous reports⁵⁴. The real AMD collected from a copper-sulfur tailings pond was tested and used after filtration through a 0.22-µm filter to remove particles and microorganisms, and its detailed composition was illustrated in Supplementary Table 2.

Experimental apparatus and procedure

Electrochemical experiments were conducted in a cylindrical electrochemical cell (an inner diameter of 90 mm, an inner height of 100 mm, and a working volume of 500 mL) (Supplementary Fig. 1). The packing area (an inner diameter of 20 mm, an inner height of 80 mm) was located in the middle of the reactor and filled with 25 g of limestone particles (average diameter of 1–3 mm, 33.8 wt% Ca) (Supplementary Fig. 2), in which a rod-shaped titanium anode coated by Ru/Ir (a diameter of 3 mm, a length of 20 mm) was assembled in the center. A tubular stainless-steel cathode (a diameter of 74 mm and a height of 80 mm) with an effective area of 84.8 cm² was placed around the packing area (Supplementary Fig. 1). All electrochemical experiments were conducted under constant current provided by a direct current power supply (MN-155D, Zhaoxin Co., Ltd) at 20 ± 0.5 °C and then stirred magnetically (200 rpm).

Software simulation and calculational methods

The software package of HSC chemistry 6.0⁴¹ was applied for the simulation of heavy metal speciation in response to the electrochemical treatment.

The specific energy consumption in kW h g⁻¹ metal ions of the EAL system was calculated by:

$$\mathrm{Specific}\mathrm{energy}\mathrm{consumption}\left(\mathrm{kW}\mathrm{h}{\mathrm{g}}^{-1}\right)=\frac{\widehat{U}\times I\times t}{m}$$ 1

where Ū is the averaged DC voltage (V), I is current (A) over the electrochemical process of specific time (h), and m refers to the amount of heavy metal extracted from wastewater in mg over the processing time.

The acidity of the simulated AMD and real AMD was calculated by:

$$\mathrm{Acidity}\left(\mathrm{mg}{\mathrm{CaCO}}_3{\mathrm{L}}^{-1}\right)=50\times \left\{\frac{3\times \left(c\left({\mathrm{Fe}}^{2+}\right)+c\left({\mathrm{Fe}}^{3+}\right)\right)}{56}+\frac{3\times c\left({\mathrm{Al}}^{3+}\right)}{27}+\frac{2\times c\left({\mathrm{Mn}}^{2+}\right)}{55}+1000\times {10}^{-\left(\mathrm{pH}\right)}\right\}$$ 2

Where, c(Fe²⁺), c(Fe³⁺), c(Al³⁺) and c(Mn²⁺) refer to the mass concentration of Fe²⁺, Fe³⁺, Al³⁺ and Mn²⁺ in solution.

Liquid analysis

We determined the contents of metal ions, rare earth elements, and the elemental compositions of the limestone, the cathodic extractions, and the bulk precipitates using ICP-OES (iCAP7400 Duo MFC, Thermo Scientific) applying 0.45 μm filtration, nitric acid dissolution, and proper dilution. The pH was measured by a pH meter (SevenExcellence, METTLER TOLEDO) directly after the liquid samples were taken from bulk solutions. Notably, the liquid sample used for the electrode’s local pH measurement was collected by the Rhizon MOM Soil Moisture Sampler (MOM, Rhizosphere Research Products), immobilized near the electrodes during the electrochemical process. We used the ion chromatograph (IC, Dionex Aquion RFIC, Thermo Fisher) to determine the concentration of anions in the real acid mine drainage. Besides, a TOC analyzer (TOC Multi N/C 3100, Analytik Jena AG) was used to check the dissolved inorganic carbon concentration of liquid samples.

Solids characterization

For the limestone, the cathodic extractions, and the bulk precipitates, we used Scanning Electron Microscopy coupled with energy-dispersive X-ray Spectroscopy (SEM-EDS) and X-ray diffraction analysis (XRD) to check the morphology, elemental distribution, and phase composition. In addition, the samples’ chemical state was determined by an X-ray photoelectron spectrometer (XPS, PHI 5000 Versaprobe III, ULVAC-PHI) with the monochromatic Al Kα radiation at a voltage of 15 kV and current of 5 mA.

Life cycle assessment

Simapro version 10.1 software was applied to establish scenarios and assess the environmental impacts through ReCiPe Midpoint (H) v1.05 methodology. We obtained background information from Ecoinvent version 3 database⁵⁵. The system boundaries of conventional lime dosing and the EAL treatments are illustrated in Supplementary Fig. 22 for the different scenarios. Supplementary Table 5 shows the main inputs and outputs of the scenarios under investigation. Detailed calculations are included in the supplementary material (Supplementary Note 4).

Author contributions

W.L. and Y.L. conceived the study and designed experiments; W.L. performed and J.L. helped with experiments; W.L. analyzed the data and assembled figures; J.W. and Z.Z. helped with the data analysis. W.L. and Y.L. wrote the manuscript; D.Z. helped with the manuscript revision; All authors commented on the paper.

Peer review

Peer review information

Nature Communications thanks Denys Villa-Gomez, and the other, anonymous, reviewer(s) for their contribution to the peer review of this work. A peer review file is available.

Data availability

The Source Data underlying the figures of this study are available with the paper. All raw data generated during the current study are available from the corresponding authors upon request. Source data are provided with this paper.

Competing interests

The authors declare no competing interests.

Supplementary information

The online version contains supplementary material available at 10.1038/s41467-025-62045-w.