Materials

The substances Cu(NO₃)₂·3H₂O (99.0%), CuCl₂·2H₂O (99.0%), Al(NO₃)₃·9H₂O (99.0%), AlCl₃·6H₂O (99.0%), Mg(NO₃)₂·6H₂O (99.0%), Zn(NO₃)₂·6H₂O (99.0%), Ni(NO₃)₂·6H₂O (99.0%), Cd(NO₃)₂·4H₂O (99.0%), Pb(NO₃)₂ (99.0%), NaNO₃ (99.0%) and NaCl (99.0%) were purchased from Aladdin–E (China). NaOH, HCl and HNO₃ were purchased from Guangzhou chemical reagent CO., LTD (Guangzhou, China). All chemical reagents were used as received. Crystal seeds of hydrotalcite were synthesised by traditional coprecipitation methods (Laipan et al., Reference Laipan, Zhang, Wang, Zhu and Sun2015, Laipan et al., Reference Laipan, Yu, Zhu, Zhu, Smith, He, O’Hare and Sun2023, Laipan et al., Reference Laipan, Chen, Wang, Zhang, Yuan, Zhu and Sun2024) and paratacamite (one type of HMSs) was obtained by a direct precipitation method (Pollard et al., Reference Pollard, Thomas and Williams1989). Acid mine drainage was collected from Fenghuang Village, Baihe County, Ankang City, Shaanxi Province, China. The concentration of the heavy metals of the acid mine drainage are shown in Table 1.

Table 1. Heavy metal ions and their concentrations of the acid mine drainage

Recovery of heavy metals from simulated wastewater using hydrotalcite seeds

First a 1 L solution (pH 3.0) of simulated wastewater with various heavy metal ions (Cu²⁺, Pb²⁺, Ni²⁺, Zn²⁺ and Cd²⁺; each a concentration of 1 mmol/L) and Al³⁺ (2.5 mmol/L) was prepared by using nitrates. Subsequently, 0.5 g of hydrotalcite seeds were added into the solution while adjusting the reaction pH (using 0.1 mol/L NaOH and HNO₃ solution) by an automatic pH controller (Chroma CPH–2, China) and reacted for 12 h under magnetic stirring. To explore the effect of pH, the target reaction pH was maintained at 5∼9. Subsequently, the products were centrifuged to obtain the solids and supernatant. The supernatant was analysed by inductively coupled plasma mass spectrometry (ICP-MS, using an Agilent 7900) to determine the removal efficiency of various heavy metals, while the solid products were freeze dried for further analysis after washing with ultrapure water three times. A reaction system with pH 7.0 was also used to observe removal kinetic process of each heavy metal. The concentrations of the heavy metal cations in the supernatant were determined by sampling at intervals of 1, 2, 5, 10, 15, 30, 45, 60, 120 and 180 minutes.

Controlled experiments were also conducted, using the same experimental procedure as the above experiments but without adding hydrotalcite seeds.

Recovery of heavy metals from acid mine drainage using hydrotalcite seeds

A 200 mL solution of acid mine drainage was used to conduct the experiment. 0.2 g of hydrotalcite seeds were added into the solution while adjusting the reaction pH to 7.0 (using 0.1 mol/L NaOH solution) by an automatic pH controller (Chroma CPH–2, China). The concentrations of heavy metals in the supernatant were determined (using ICP-MS) by sampling at intervals of 5, 10, 30, 60, 90, 120 and 180 minutes. As there is Fe³⁺ in the drainage, no Al³⁺ was added into the drainage during the reaction. Controlled experiments without adding hydrotalcite seeds were also conducted.

Recovery of Cu²⁺ in the presence of different crystal seeds

The system Cu²⁺–Cl^––Al³⁺ was used to test the effects of different crystal seeds (hydrotalcite and a crystalline hydroxide metal salt of paratacamite) on the products recovered. First a 1 L solution (pH 4.5) with 5 mM CuCl₂·2H₂O and 2.5 mM AlCl₃·6H₂O was prepared with the molar ratio of Cl to Cu to Al of 7:2:1, respectively. The crystal seeds of paratacamite or hydrotalcite were then added into the Cu²⁺–Cl^––Al³⁺ system maintaining a crystal seed concentration of 1 g/L. After that, the solution pH was regulated to the targets of pH 5.0 and 7.0 and reacted for 12 h under magnetic stirring. Finally, the products were centrifuged and freeze dried for further analysis.

Characterisation methods

X-ray diffraction (XRD) patterns of the products were measured on a Bruker D8 ADVANCE X-ray diffractometer using CuKα radiation operating at 40 kV and 40 mA. The patterns were recorded over the 2θ range from 3 to 80° with a scan speed of 3°/min using a bracket sample holder. Scanning electron microscopy (SEM) images were obtained by FEI Verios 460 high-resolution field emission SEM. Scanning transmission electron microscopy (STEM) and element-distribution mapping images were collected by FEI Talos F200S high-resolution TEM.

Effects of crystal seeds on heavy metals removal

The impact of hydrotalcite crystal seeds on the removal of heavy metals from simulated wastewater and actual acid mine drainage was assessed. Simulated wastewater with mixed heavy metal ions (Cu²⁺, Pb²⁺, Ni²⁺, Zn²⁺ and Cd²⁺) was used to test whether the hydrotalcite crystal seed method is applicable to the wastewater with multiple heavy metals; additonaly, to reveal if the coexistent heavy metal cations would affect the removal of each other. Results indicated an improvement of heavy metal removal efficiency of pH ∼7–9 by using the crystal seed method (Fig. 1g–j; thus, pH ≥ 7 was selected in the following experiments). At pH 6, the removal of Cu, Cd and Pb was improved whereas Ni and Zn removal decreased. Under a more acid solution (e.g. pH 5), the removal of heavy metals was significantly lower than that of the seed-free system. The reasons for the above differences under acid solution (pH 5 and 6) should be that an acid condition would result in dissolution of the hydrotalcite. Additionally, hydrotalcite removes heavy metal cations mainly via ligand exchange with the H⁺ (on the surface hydroxyl groups of hydrotalcite), surface precipitation and isomorphic substitution (Liang et al., Reference Liang, Zang, Xu, Tan, Hou, Wang and Sun2013, He et al., Reference He, Qiu, Hu and Liu2018). Under an acid solution, the ligand exchange between heavy metal cations and surface hydroxyl groups (H⁺) would be suppressed by the free H⁺ ions and thus caused the decrease of removal rate. According to the experimental results, the addition of hydrotalcite crystal seeds can effectively improve the heavy metals removal efficiencies with an increase of 18%–47%, and for all heavy metals, the total heavy metal removal rate can be increased by 31.8% (Fig. 1a–e, under pH 7).

Figure 1. Simultaneous removal of multiple heavy metal ions from simulated wastewater and actual acid mine drainage in the systems with and without the hydrotalcite seed: (a)∼(e) removal kinetics under pH 7 from simulated wastewater; (f)∼(j) removal efficiencies under different pH from simulated wastewater; (k) and (l) removal kinetics of multiple heavy metal ions from actual acid mine drainage under pH 7. The concentration of each heavy metal in simulated wastewater is 1 mmol/L; the concentration of the hydrotalcite seed in all systems is 1 g/L; the contact time in all systems is 3 h.

An actual acid mine drainage with the coexistence of heavy metal cations (Fe³⁺, Cu²⁺ and Zn²⁺) and anion (Cr(VI)) (Table 1) was also employed to test whether the hydrotalcite crystal seed method is applicable to actual wastewater. Without addition of hydrotalcite crystal seeds, Fe³⁺, Cu²⁺ and Cr(VI) can be removed completely by pH neutralisation, but the removal rate of Zn²⁺ presented an abnormal decrease trend as reaction time increase (Fig. 1k). This abnormal decrease trend can also be found in the previous study about removal of Zn²⁺ by the Zn²⁺–Al³⁺ coprecipitation method (Laipan et al., Reference Laipan, Chen, Wang, Zhang, Yuan, Zhu and Sun2024). However, this abnormal decrease trend can be eliminated by the addition of hydrotalcite crystal seeds (Fig. 1l). Our previous research findings indicate that the precipitation of Zn²⁺ and the formation of Zn-LDH are not synchronous, leading to the re-release of Zn²⁺ back into the solution (Laipan et al., Reference Laipan, Chen, Wang, Zhang, Yuan, Zhu and Sun2024). Zn²⁺ may initially form precipitates and be adsorbed onto ferric hydroxide, then gradually evolves into Zn-LDH, during which the formation of positively charged Zn-LDH results in some Zn²⁺ being re-released into the solution. This phenomenon did not occur in simulated wastewater; we speculate that this may be due to the coexistence of five divalent metal ions in the simulated wastewater, including ions with a precipitation pH close to that of Zn²⁺, which easily leads to the coprecipitation of Zn, forming a multi-component LDH of Zn with one or more of Cu/Ni/Cd/Pb. In actual acid mine wastewater, only Cu²⁺ and Zn²⁺ are present as divalent metal ions, with Cu²⁺ being in a lower concentration (24 ppm for Cu and 53 ppm for Zn). The amount of Zn removed by coprecipitation with Cu²⁺ to form Cu/Zn-LDH should be relatively small. Additionally, the large difference in precipitation pH of Cu²⁺ and Zn²⁺ may inhibit the coprecipitation removal of Zn. Fe³⁺, Cu²⁺ and Zn²⁺ can be effectively removed in the presence of hydrotalcite seeds, however Cr(VI) removal decreased as time increased. This is because the removal of Cr(VI) by LDHs faces competition from water-soluble carbon dioxide. As indicated in Fig. 1l, reaction for 50–60 min should be an effective way for realising high removal efficiency for all the heavy metals.

Effects of crystal seed on heavy metal recovery product

The solid products generated from simulated wastewater (with mixed heavy metals of Cu²⁺, Pb²⁺, Ni²⁺, Zn²⁺ and Cd²⁺) in the treatments with and without hydrotalcite seeds were characterised by XRD. The results show that there are two crystalline products of hydrotalcite and hydrocerussite (Pb₃(CO₃)₂(OH)₂) (Fig. 2). Without hydrotalcite seeds, the dominant product is hydrocerussite with a small amount of impurity phase of a LDH. However, when hydrotalcite seeds were added into the reaction system the dominant recovered product became the LDH phase at higher pH, with a small amount of the impurity phase of hydrocerussite. During the reaction, only the by-product of hydrocerussite was observed, and beneficially, the phase of the recovered products can be regulated by adjusting the reaction pH. The above results suggests that the use of crystal seeds can selectively induce the formation of certain kind of products of heavy metals.

Figure 2. XRD patterns of the recovered products from the simulated wastewater with mixed heavy metal ions (Cu²⁺, Pb²⁺, Ni²⁺, Zn²⁺ and Cd²⁺) with or without hydrotalcite seeds addition.

To further reveal the effect of crystal seeds on product selectivity, a simplified system of Cu²⁺–Al³⁺–Cl^– with addition of different crystal seeds was used. Taking into account that copper ions can complete precipitation at low pH of generally <6.5, the experimental pH values were chosen as 5 and 7. For the Cu²⁺–Al³⁺–Cl^– reaction system without crystal seeds, the products are mixtures of LDHs and paratacamite (Cu₂(OH)₃Cl) (Fig. 3a). When Cu²⁺ encountered both Cl^– and Al³⁺, the dominant crystalline solids are paratacamite with a small amount of LDH at low Al³⁺ concentrations. However the proportion of LDH increased with increasing Al³⁺ concentration. As the LDH and paratacamite are products of the Cu²⁺–Al³⁺–Cl^– system, the LDH (hydrotalcite) and paratacamite were employed as two different kinds of crystal seed. According to Fig. 3b, only paratacamite was formed at pH 5.0 when paratacamite had been applied as the crystal seed; but when pH increased to 7.0 the crystalline products reverted to a mixture of LDHs and paratacamite, and the LDH became the dominant product. Interestingly, when hydrotalcite seeds were added into the Cu²⁺–Cl^––Al³⁺ system, the LDH became the only crystalline product at both pH 5.0 and 7.0. Notably, there are two kinds of LDH (i.e. hydrotalcite and CuAl-LDH) in the resultant solid of the hydrotalcite seeding system under pH 7.0, suggested by the two diffraction peaks (at 2θ of c.a. 11°, belonging to (003) of the LDH) and their corresponding secondary diffraction peaks (at 2θ of 23–24°, belonging to (006) of the LDH). Regarding the single diffraction peak belonging to (003) of the LDH for the product in the hydrotalcite seeding system at pH 5.0, the reason may be that the content or concentration of formed CuAl-LDH is low. The above results suggest that a crystal seed can change the fate of heavy metal ions by inducing formation of specific minerals. That is, the recovered products of heavy metal ions can be controlled by the addition of different crystal seeds.

Figure 3. XRD patterns of Cu–bearing minerals generated from different systems: (a) crystalline products from Cu²⁺–Cl^––Al³⁺ systems at pH 7.0 with different Cu:Al molar ratios; (b) crystalline products from different crystal seeds (paratacamite and hydrotalcite) induced Cu²⁺–Cl^––Al³⁺ systems at pH 5.0 and 7.0 with a Cu:Al molar ratio of 2. At pH 5.0, paratacamite induced Cu²⁺ to form paratacamite; at pH 7.0 the products are a mixture of paratacamite and LDH, with the LDH as the dominant product. However with hydrotalcite seeds, the products are only LDH at both pH 5.0 and 7.0.

The products were further analysed by SEM and STEM. Products from the paratacamite seeding system at pH 5.0 show a poorly developed octahedral nanocrystals-like shape, which is similar with the reported shape of paratacamite particle (Liu et al., Reference Liu, Inoué, Zhu, H. and Hochella2017), further supporting evidence for the formation of paratacamite (Fig. 4a,b). Products from paratacamite seeding system at pH 7.0 show both flaky and nano-sized particle morphology, suggesting the coexistence of LDHs and paratacamite (Fig. 4c,d). When hydrotalcite was used as the crystal seed, only the flaky morphology of a LDH was observed (Fig. 4e–h), in good agreement with XRD results.

Figure 4. SEM images of the Cu-bearing solids produced from the paratacamite and hydrotalcite inducing Cu²⁺–Cl^––Al³⁺ system at pH 5.0 and 7.0. (a, b) Products from paratacamite system at pH 5.0; (c, d) products from paratacamite system at pH 7.0; (e, f) products from hydrotalcite system at pH 5.0; (g, h) products from hydrotalcite system at pH 7.0.

Transmission electron microscopy results also confirm only nano-sized particles in the products from the paratacamite seeding system at pH 5.0, whereas there was a coexistence of flaky and particle morphology in the products at pH 7.0 (Fig. 5a,b), further demonstrating the crystalline product of paratacamite and mixture of paratacamite and LDH at pH 5.0 and 7.0, respectively. The solely flaky morphology in the products from hydrotalcite seeding systems suggests the products are LDHs only (Fig. 5c,d). Additionally, elemental mapping results provide visual evidence for the formation of different Cu-bearing minerals at different systems. In the paratacamite seeding systems, the distributions of Cu and Cl are highly consistent in the products generated at pH 5.0 (Fig. 5e). However, in the products resultant from pH 7.0, though some of the distribution of Cu are highly consistent with Cl, more of the Cu distribution coincides with Al (Fig. 5f). These results further support the formation of a mixture of LDHs and paratacamite, with LDH as the dominant product. As for the hydrotalcite seeding systems, all of the Cu, Al and Cl are highly consistent in distribution in the products (Fig. 5g,h), indicating a single type of product. Based on the analysis of XRD, SEM, TEM and element-distribution mapping results, the product is CuAl-LDH with Cl^– as a counterion in the interlayer in the hydrotalcite seeding systems. In summary, addition of crystal seeds during heavy metal removal/recovery by precipitation can change the product, and different crystal seeds present different induction effects on product selectivity, which is illustrated in Fig. 6.

Figure 5. TEM and element-distribution maps of the products of the mineral inducing Cu²⁺–Cl^––Al³⁺ systems. (a, e) products from the paratacamite seed system at pH 5.0; (b, f) products from the paratacamite seed system at pH 7.0; (c, g) products from hydrotalcite seed system at pH 5.0; (d, h) products from hydrotalcite seed system at pH 7.0.

Figure 6. Reaction systems and their corresponding products.

The relationship between enthalpy of formation and crystallisation behaviour

Enthalpies of formation from the elements (∆_f H° at 298.15 K) of different products were employed to reveal the different crystallisation behaviours of Cu²⁺ in the presence of ions and different crystal seeds. In general, thermodynamic properties can be estimated by treating a compound as a mixture of binary compounds in which the metal and anion have similar coordination environments (kumar Allada et al., Reference kumar Allada, Navrotsky, Berbeco and Casey2002). For example, for a LDH phase of M(II)_1–xM(III)_x(OH)₂(A^n–)_{x /n}·mH₂O (anions, A^n–; divalent metal, M(II); trivalent metal, M(III)), previous studies indicated that the enthalpies of formation can be estimated as a weighted sum of the components (equation 1) (kumar Allada et al., Reference kumar Allada, Navrotsky, Berbeco and Casey2002, Evans and Slade Reference Evans and Slade2006). Therefore, the enthalpies of formation of CuAl-LDH (Cu_1–xAl_x(OH)₂Cl_x·mH₂O) and paratacamite (Cu₂(OH)₃Cl) can be estimated according to equations (1–3):

(1)\begin{align} &{{\Delta _f}H{^\circ _{M{{\left( {II} \right)}_{1 - x}}M{{\left( {III} \right)}_x}{{\left( {OH} \right)}_2}{{\left( {{A^{n - }}} \right)}_{x/n}}m{H_2}O}}}\nonumber\\ &\quad = \frac{x}{n}\Delta {H_{M\left( {II} \right){A_{2/n}}}} + x\Delta {H_{M\left( {III} \right){{\left( {OH} \right)}_3}}}\nonumber\\ &\qquad+ \left( {1 - 2x} \right)\Delta {H_{M\left( {II} \right){{\left( {OH} \right)}_2}}} + m\Delta {H_{{H_2}O}}\end{align}

(2)\begin{equation}{\Delta _f}H{^\circ _{C{u_2}{{\left( {OH} \right)}_3}Cl}} = \frac{1}{2}\Delta {H_{CuC{l_2}}} + \frac{3}{2}\Delta {H_{Cu{{(OH)}_2}}}\end{equation}

(3)\begin{align} {{\Delta _f}H{^\circ _{C{u_{1 - x}}A{l_x}{{\left( {OH} \right)}_2}C{l_x}\cdot m{H_2}O}}} &= \frac{x}{2}\Delta {H_{CuC{l_2}}} + x\Delta {H_{Al{{\left( {OH} \right)}_3}}}\nonumber\\ &\quad+ \left( {1 - \frac{{3x}}{2}} \right)\Delta {H_{Cu{{\left( {OH} \right)}_2}}} + m\Delta {H_{{H_2}O}} \end{align}

For LDHs, according to the previous reports, the value of x is generally in the range 0.2∼0.33 (Evans and Slade, Reference Evans and Slade2006, Fan et al., Reference Fan, Li, Evans and Duan2014), and m = 1–(3x/2) (Miyata, Reference Miyata1980); the value of m should therefore, be in the range 0.51∼0.70. Accordingly, a range of enthalpies of formation of CuAl-LDH can be estimated. The enthalpies of formation of different compounds are given in Table 2 [the enthalpies of formation of the binary compounds are from ‘Lange’s handbook of chemistry’ (Dean Reference de Souza E Silva, de Mello, Menezes Duarte, Montenegro, Araujo, de Barros Neto and da Silva1999)]. According to the calculation, Cu₂(OH)₃Cl possess much lower enthalpies of formation (–778.6 KJ/mol) than that of Cu(OH)₂ (–450.4 KJ/mol), indicating paratacamite is more stable and thermodynamically favourable. CuAl-LDH and paratacamite present close values of enthalpies of formation; this should be the reason for the fact that the product of Cu²⁺–Cl^––Al³⁺ systems are a mixture of paratacamite and LDHs. Increasing the concentrations of Al³⁺ in Cu²⁺–Al³⁺–Cl^– systems will generally increase the x value of CuAl-LDH (Cu_1–xAl_x(OH)₂Cl_x·mH₂O), and thus favours the formation of LDHs, in good agreement with XRD results (Fig. 3a).

Table 2. Enthalpies of formation (298.15 K) from hydroxide and chlorate components, and elements

However, as enthalpies of formation of CuAl-LDH (–829.9 ∼ –793.0 KJ/mol) is slightly lower than that of paratacamite (–778.6 KJ/mol), the former has higher stability than the latter. Thus, the dominant product should be a LDH instead of paratacamite in Cu²⁺–Cl^––Al³⁺ systems without crystal seeds. But it was suggested by the results that the paratacamite was the dominant product in these systems without crystal seeds. The following two reasons may be responsible for why the theory does not agree with the fact. On the one hand, it is widely believed that the metal hydroxide is formed first and subsequently crystalises into a LDH (Boclair and Braterman Reference Boclair and Braterman1999, Boclair et al., Reference Boclair, Braterman, Jiang, Lou and Yarberry1999, Evans and Slade Reference Evans and Slade2006, Yang et al., Reference Yang, Liu, Zhu, Chen, Wei, Chen, Xian and He2011, Chang et al., Reference Chang, Wu, Song, Kuang, Lei, Wang and Sun2013, Grégoire et al., Reference Grégoire, Ruby and Carteret2013, Paikaray et al., Reference Paikaray, Gomez, Hendry and Essilfie-Dughan2014). As suggested by the enthalpies of formation of Cu(OH)₂ and Cu₂(OH)₃Cl, the latter is a more thermodynamic favourable product. Therefore, paratacamite will form first which prevents the formation of Cu(OH)₂, and eventually prevents the formation of CuAl-LDH. On the other hand, the formation of CuAl-LDH may be controlled primarily by dynamics rather than thermodynamics, and paratacamite may be a more dynamically preferable product than CuAl-LDH. The fact that the addition of crystal seeds can change products may be because the hetero-surface of crystal seeds can act as crystal nucleus and provides active sites, so nucleation becomes energetically favourable and therefore will change the crystallisation behaviour and accelerate the crystallisation process (Yazdanpanah et al., Reference Yazdanpanah, Testa, Perala, Jensen, Braatz, Myerson and Trout2017).

Possible reason for the induction effect of crystal seed on heavy metal removal and product

According to the results of this work, the addition of hydrotalcite and paratacamite crystal seeds can both enhance the removal of heavy metals and crystallisation of products, and the type of the product is closely related to the types of the added crystal seed. It is believed that the following aspects may be the cause of the improvement of heavy metal removal efficiency by a crystal seed method. (1) The crystal seed is also an adsorbent with high surface reactivity that can rapidly accumulate ions on its surface (Zhang et al., Reference Zhang, Chen, Yu, Yu and Qiu2022a, Laipan et al., Reference Laipan, Yu, Zhu, Zhu, Smith, He, O’Hare and Sun2023, Li et al., Reference Li, Jing, Zhang, Zhu, Yu, Chen and Qiu2023); thus, local supersaturation may be easily reached, leading to interfacial precipitation or co-precipitation. As indicated in the introduction, adsorption, the commonly occurring process in Nature, can sufficiently enrich concentrations of ions around the solid surface, causing a local supersaturated state and formation of surface precipitates (Yan et al., Reference Yan, Yu, Yan, Wu, Ning, Qi and Han2017, Siebecker et al., Reference Siebecker, Li and Sparks2018, Deng et al., Reference Dean2019). (2) The addition of hydrotalcite seeds can promote the heterogeneous nucleation pathway of a crystal, increasing the crystallisation efficiency and growth process of a crystal (Sear, Reference Sear2006, Yazdanpanah et al., Reference Yazdanpanah, Testa, Perala, Jensen, Braatz, Myerson and Trout2017). Additionally, the seeds can also cause local oversaturation of a low concentration solution via its adsorption or other effects, and thus results in induced crystallisation (Li et al., Reference Li, Sun, Wang, Xu and Yan2015, Deng et al., Reference Dean2019). (3) Previous studies have demonstrated that under a solid–water interface the solid could induce the crystallisation of ions even under a low saturated or unsaturated solution environment (Qian and Botsaris, Reference Qian and Botsaris1998, Liu et al., Reference Liu, Xu, Huang, Cheng and Seo2021, Yang et al., Reference Yang, Zhao, Zhu and Zhang2023). This may be because adding crystal seeds can provide extra surface energy (Yazdanpanah et al., Reference Yazdanpanah, Testa, Perala, Jensen, Braatz, Myerson and Trout2017). The extra surface energy may also be the reason for the different crystallisation behaviours of systems with and without paratacamite and hydrotalcite seeds.