Zinc-reinforced ZSM-5 subject to a rare-earth magnet and the presence of a legume yields considerable copper extraction

Compounded effects of copper pollution

A trace amount of copper is necessary for life to function, however, due to anthropogenic effects; Cu²⁺ levels may exceed the permitted upper regulatory limit of 0.0015 g/L in drinking water and 0.0013 g/L in industrial effluents (Bilal et al., 2013). As copper has a high affinity for organic matter, it accumulates in sediments (Han et al., 2001; Caillat et al., 2014), so it does in biological systems (Salt et al., 1995; Han et al., 2001; van Hullebusch et al., 2003; Chibuike & Obiora, 2014; Li et al., 2014; Saadani et al., 2016; Usman, Al-Ghouti & Abu-Dieyeh, 2019). Exposure to excessive amounts of Cu²⁺ can lead to various disruptive effects and impairment of biological processes (Dave, 1984; Kramer et al., 2004; Lamb et al., 2012; Saadani et al., 2016; Igiri et al., 2018; Taylor et al., 2020). Cupric stress can have inhibitory and detrimental effects on metabolism, growth, and development of plants in natural and agricultural systems (Chibuike & Obiora, 2014; Shabbir et al., 2020; José Rodrigues Cruz et al., 2022; Schmitz et al., 2023; Xu et al., 2024).

Sulfur is a non-metallic element vital to biogeochemical cycles and biosynthesis (Giordano, Noriciand & Hell, 2005; Lucheta & Lambais, 2012) and is amply available in different forms, such as sulfate SO₄²⁻, which is prevalent in water (World Health Organization (WHO), 2004; Giordano, Noriciand & Hell, 2005; Boone et al., 2012). Sulfate levels in freshwater are generally around 20 mg/L (World Health Organization (WHO), 2004), whilst the Secondary Maximum Contaminant Level (SMCL) for sulfate in drinking water is 250 mg/L (Webb & Czapar, 2013). However, clear guideline values for sulfate in water have not been fully administered yet, see also (Boone et al., 2012). Excess amount of SO₄²⁻ may concur with abundant Cu²⁺ and thus aggravate the burdens of pollution on the environment due to various natural and anthropogenic processes (Tabatabai, 1987), such as the widespread use of copper sulfate CuSO₄ in fertilisers and pesticides as well as in water treatment (van Hullebusch et al., 2003; Boone et al., 2012).

Adsorption of anions, as well as heavy metal cations (Vidal et al., 2009; Liu, Li & Zhang, 2010; Bilal et al., 2013; De Gisi et al., 2016), represents an extraction technique that is affordable and generally environmentally friendly (Montes-Atenas & Valenzuela, 2017). Attaining improved remediation of pollution in ecosystems is, therefore, one of the main driving forces underlying the development of enhanced adsorbents (Bilal et al., 2013; De Gisi et al., 2016; Virgen et al., 2018).

The employment of zeolites in the adsorption of pollutants

Zeolites, having the general formula (M_2/n O·Al₂O₃·xSiO₂·yH₂O), n=cation valence (Sherman, 1999), are a group of natural or synthetic hydrated alumina-silicate minerals with reactive pores of molecular dimensions that allow configurational diffusion and thus entrapment of target anions or cations in the zeolite pores (Masuda, 2003; Widayat & Annisa, 2017). The zeolite polyhedral (tetrahedral) structure consists of a framework of [SiO4]⁴⁻ and [AlO4]⁵⁻ linked to each other at corners by sharing their oxygen atoms in a three-dimensional network with numerous voids (Muraoka et al., 2019). This crystalline microporous alumina-silicate structure varies according to the configuration of the oxygen sharing (Masuda, 2003; Muraoka et al., 2019), where Al³⁺ isomorphous substitution for Si⁴⁺ in the said framework results in a negative charge that is key in the adsorption of heavy metal cations (Erdem, Karapinar & Donat, 2004; Barakat, 2008). The adsorption efficiency is subject to several factors including zeolite crystal structure, pore sizes, quantity, adsorbate concentration, pH, contact time, and affinity to desired adsorbates (Margeta et al., 2013).

ZSM-5 (Zeolites Socony Mobil-5) is a synthetic microporous zeolite of MFI framework type (Priyadi & Mukti, 2015; Wojciechowska et al., 2019), with a higher silica-to-alumina ratio, good thermal stability, and high contact surface area. With its unique framework structure, ZSM-5 possesses unique selectivity properties and thus is commonly used as an adsorbent and as a catalyst in the petrochemical industry (Widayat & Annisa, 2017; Wojciechowska et al., 2019; Niu et al., 2022; Yuan et al., 2022; Zachariou et al., 2023). These properties together with acid resistance and pores of 5.1–5.6 Å (Shirazi, Jamshidi & Ghasemi, 2008), make ZSM-5 a good adsorbent of heavy metals (Wojciechowska et al., 2019) and a suitable candidate for extracting Cu²⁺ (1.28 Å diameter) from contaminated media.

As the porosity of ZSM-5 is key for its adsorption capacity (Liu et al., 2021), thermal, as well as chemical modification, are commonplace to increase mesoporosity and hence adsorption capacity (Rac et al., 2020). However, the adsorption ability of ZSM-5 has not received sufficient examination thus far (Priyadi & Mukti, 2015). The adsorption of heavy metals may improve by chemical modification of the zeolite via ion exchange or wet impregnation (Kinger et al., 2000; Ciobanu et al., 2008; Wojciechowska et al., 2019). The chemical modification can enhance the performance, selectivity, and functionality of the zeolite compared to the raw unmodified form (Kinger et al., 2000; Cieśla et al., 2019; Wojciechowska et al., 2019).

Chemically modified forms of ZSM-5 are used in desulfurisation through selective adsorption of sulfur containing compounds (Garcia & Lercher, 1992; Liu, Li & Zhang, 2010; Liu et al., 2021), e.g., from refinery streams (Dehghan & Anbia, 2017; Sarda et al., 2012). Furthermore, the adsorption of pharmaceutical water contaminants has been achieved by Rac et al. (2020) using thermally and chemically treated de-silicated ZSM-5 with resultant increased mesoporosity. However, synthesis and chemical modification consume time and energy and are usually less environmentally friendly (Muhammad, 2018; Cieśla et al., 2019). Therefore, the production of physically modified zeolite forms through novel means with the potential to be more effective in capturing environmentally toxic pollutants is desirable. Interestingly, Alswata et al. (2017) demonstrated that hybridising zeolite with nano-particles of zinc oxide (ZnO), via co-precipitation, could yield considerable extraction of heavy metals from aqueous solutions. Nevertheless, physical rather than chemical modification of zeolites, such as ZSM-5, is still underdeveloped.

Electrochemical series and magnetism

According to the electrochemical series, any metal with higher reactivity can be oxidised, whilst it reduces and displaces a lesser reactive metal from its soluble salt compounds. For the displacement reaction to occur, the target species must be less reactive (van Straten & Ehret, 1939). For instance, by reacting with zinc solid Zn_(s) (reducing agent), Cu²⁺ of CuSO₄ (oxidising agent) accepts electrons donated by the Zn_(s) and transitions accordingly from its aqueous state to a solid state, as shown in the following equations (Sarda, Handa & Arora, 2016):

(1) $$\mathrm{Z}{\mathrm{n}}_{(\mathrm{s})}+\mathrm{C}{\mathrm{u}}_{(\mathrm{a}\mathrm{q})}^{2+}\to \mathrm{Z}{\mathrm{n}}_{(\mathrm{a}\mathrm{q})}^{2+}+\mathrm{C}{\mathrm{u}}_{(\mathrm{s})}$$

(2) $$\mathrm{Z}{\mathrm{n}}_{(\mathrm{s})}\to \mathrm{Z}{\mathrm{n}}^{2+}+2{\mathrm{e}}^{-}$$

(3) $$\mathrm{C}{\mathrm{u}}^{2+}+2{\mathrm{e}}^{-}\to \mathrm{C}{\mathrm{u}}_{(\mathrm{s})}.$$

It is striking, however, that modifying (i.e., altering) the surface of zeolite using a reducing agent such as zinc (post-transitional metal) to accompany the sorption of heavy metal cations with the ability to electrochemically reduce the cations to metal solids has never been tested before despite being highly plausible. Such a dual functionality of pollutant extraction could be further enhanced, especially the electrochemical facet, by employing a static magnetic field. The latter has been demonstrated to accelerate and increase the displacement and deposition of copper on a zinc solid substrate (Udagawa et al., 2014). To date, an experimental test of capturing heavy metal cations using zeolites physically modified by a reducing metal, in a solid form, under strong magnetism represents a knowledge gap to fill.

Phytoremediation

Phytoremediation is a common form of bioremediation (Sen Gupta, Yadav & Tiwari, 2020) that employs plants to extract/remove toxic pollutants (Salt et al., 1995; Suman et al., 2018; Usman, Al-Ghouti & Abu-Dieyeh, 2019). The main biological mechanisms of phytoremediation observed for heavy metal extraction are through uptake and deposition into biological mass in the plant root system (rhizofiltration) Furthermore, subsequent translocation of toxic metals to accumulate in the shoot system may occur, a process also referred to as phytoextraction (Salt et al., 1995). Tolerant plants are, therefore, useful in cleansing and phytostabilising affected surroundings by sequestering toxic cations (Salt et al., 1995; Han et al., 2001; van Hullebusch et al., 2003; Chibuike & Obiora, 2014; Li et al., 2014; Saadani et al., 2016; Usman, Al-Ghouti & Abu-Dieyeh, 2019).

Various wild or natural plants are shown to phytostabilise heavy metals (e.g., Xiao et al., 2008; Ahmadpour et al., 2015; Xu et al., 2019; Usman, Al-Ghouti & Abu-Dieyeh, 2019), generally due to the negative charge of the cellular walls that can attract and compartmentalise heavy metal cations (Jiang, Liu & Liu, 2001; Matijevic, Romic & Romic, 2014; Usman, Al-Ghouti & Abu-Dieyeh, 2019). For example, when exposed to a group of metals including copper, a low shrub succulent, Tetraena qataranse (Hadidi) Beier & Thulin, accumulated 22.4 mg/kg of copper in the roots, whilst 6.2 mg/kg translocated into the shoots, compared to 5.7 mg/kg left in the soil (Usman, Al-Ghouti & Abu-Dieyeh, 2019). Moreover, plants are natural bio-adsorbents of sulfate (Sadeghalvad et al., 2021), as plants assimilate then store or metabolise the anion (Calderwood & Kopriva, 2014).

Interestingly, terrestrial plants can grow in aquaculture for the remediation of pollutants. For example, brown mustard, Brassica juncea (L.), and alpine pennycress, Thlaspi caerulescens (J.Presl & C.Presl), have been shown to reduce the levels of a combination of toxic metals, including copper, through phytoextraction, over several days of application (Salt et al., 1995). Whereas, through rhizofiltration via the plant root system, sunflower Helianthus annuus (L.) can substantially reduce copper levels amongst other heavy metals overnight (Salt et al., 1995). The Fabaceae family includes species that are generally tolerant accumulators of heavy metals (Reeves et al., 2017; Yadav, Singh & Jadeja, 2021). For example, the broad bean, Vicia faba (L.), is a popular global cash crop (Karkanis et al., 2018). This plant is rich in protein with robust shoots and roots where the root system colligates associated symbiotic nitrogen-fixing rhizobacteria, adding to the value of V. faba as green manure (Hamdi, 1982; Chen et al., 2018). Having considerable biomass, along with the said symbiotic association, corroborates the applicability of broad beans as a sustainable low-cost augmented phytoremediator of copper (Teng et al., 2015; Saadani et al., 2016). Moreover, V. faba has considerable tolerance to the uptake of metallic contaminants (Abdel Hamed Abdel Latef & Abu Alhmad, 2013; Teng et al., 2015), as the plant can phytostabilise and accumulate copper in its tissues (Nadgórska-Socha et al., 2013; Matijevic, Romic & Romic, 2014; Alobaidi, 2016), but that is relative to copper content in the medium (Brennan & Bolland, 2003). However, Cu²⁺ concentrations of 0.159 g/L of CuSO₄ in hydroponic culture have been reported to have a genotoxic effect on V. faba after 42-h exposure (Souguir et al., 2008). Longer exposure to Cu²⁺ from treatment with CuSO₄ results in oxidative stress, disruption of uptake of other elements, and inhibition of growth (Mourato, Martins & Campos-Andrada, 2009; Abdel Hamed Abdel Latef & Abu Alhmad, 2013; Fatnassi et al., 2015; Alobaidi, 2016; Benouis et al., 2021). That is in addition to necrosis that occurs at higher exposure levels ≥0.048 g/L (Fatnassi et al., 2015; Alobaidi, 2016). Trapping copper and sulfur using aquacultured V. faba combined with metal-reinforced zeolite subject to magnetism is yet to be examined.

This laboratory study presents a novel approach that amalgamates techniques from the fields of biology, physics, chemical engineering, and analytical science for pollutant extraction from freshwater contaminated with copper sulfate. By combining the use of zinc-reinforced ZSM-5 zeolite (through the physical embedding of zinc metal coarse powder on part of the zeolite surface), a companion rare-earth (neodymium) static super magnet, and/or an accompanying aquacultured faba bean, we provide a timely and novel solution to treat heavily polluted aquatic systems. See Fig. 1 for the experimental setup and design. The current article is peer-reviewed based on our revised preprint (Khudr et al., 2023).

Experimental set-up

The aqueous medium of the experiment mimicked freshwater used in aquaculture (Klüttgen et al., 1994). The medium was standardised, reproduced in bulk, stored in the lab environment, and aerated with an aquarium air pump for 12 h before the experiment. The medium consisted of a mixture of 0.33 g/L of sea salt (Sigma Aldrich©, Gillingham, UK), 23 ml/L of 117.6 g/L calcium chloride dehydrate CaCl₂.2H₂O (Fisher Scientific©, Loughborough, UK), 22 ml/L of 25.2 g/L sodium bicarbonate NaHCO₃ (Honeywell©, Strahlenbergerstr, Germany), and 1 ml/L of 0.07 g/L selenium dioxide SeO₂ (Merck KGaA©, Darmstadt, Germany) dissolved in Millipore Milli-Q de-ionised high purity water (pH = 7).

The pollutant was dry crystal copper sulfate pentahydrate CuSO₄·5H₂O (Sigma Aldrich©, Gillingham, UK). Imitating circumstances of potential leakage or spillage associable with anthropogenic activities, 0.7 g/L of the pollutant was applied. Copper extraction was conducted using a complex treatment: (1) zinc-reinforcement of zeolite: The zeolite of choice (ZSM-5) was applied in its raw (zinc-free form) as the baseline, or zinc-reinforced forms (via partial zinc gel coating or zinc plasma sputtering), −/+ rare-earth magnet, in the absence of aquacultured faba bean plant V. faba. (2) Vicia faba as the model bio-trap of Cu²⁺ was applied alone. (3) The zinc-gel-reinforced form (−/+ the magnet) was accompanied by the V. faba bio-trap (Fig. 1).

Zeolite reinforcement

Commercial grade raw zeolite ZSM-5 pellets, with Si/Al of 50, were acquired by the Department of Chemical Engineering (CE), The University of Manchester, from Pingxiang Naike Chemical Industry Equipment Ltd© (Pingxiang, Jiangxi, China). These zeolite pellets are stable synthetics. An amount of 7 g of bacterial agar powder (Sigma Aldrich©, St. Louis, MO, USA) was used to reinforce half of the surface of the zeolite pellet with coarse zinc powder, particle size about 0.3–1.5 mm (14–50 mesh ASTM), EMSURE® Reag. Ph Eur (Sigma Aldrich©, Gillingham, UK). The gel powder was dissolved in 100 ml of high-purity deionised water then heated on a hot plate and stirred continuously until it became homogeneous and viscous. Subsequently, ZSM-5 pellets were systematically semi-immersed individually in the done gel, using a forceps, and then coated/dusted with 0.5 g of zinc coarse powder. The resulting zinc-gel-reinforced zeolite pellets were left to dry for 4 days to allow the gel to solidify into a hard resin-like structure, holding the zinc fine metal pieces firmly onto the gel-surfaced part of the zeolite. The experiment took place at room temperature and the pH and conductivity were measured several times during the experiment using a Eutech PC 2700 probe (Thermo Fisher Scientific, Waltham, MA, USA).

Additionally, to reinforce the zeolite with nano-layers of zinc (Testbourne©, Whitchurch, UK), raw ZSM-5 was plasma-sputtered with zinc longitudinally on half of the zeolite pellet surface using a coating machine based on the technology of scanning electron microscopy (FEI Quanta 200 Environmental, FEI Company©, Hillsboro, OR, USA). The sputtering technique allows using an argon plasma stream under a vacuum to separate zinc metal atoms off a zinc disc target by applying a high voltage to initiate plasma gas ion exchange. The plasma sputter anchored the active metal (zinc) atoms onto the surface of the ZSM-5 zeolite pellets as an ultra-thin film. Technical difficulties in materialising this approach using the said machine and the substrate zinc disc resulted in having a very limited number of repeats for this addendum and hence the outcome of the zinc sputtering of zeolite is anecdotal. The zinc-sputtered zeolite treatments (ZnSZ-M) and (ZnSZ+M) were, therefore, excluded from the statistical analysis specified below.

One gram of the zeolite ZSM-5 (raw or zinc-reinforced) was wrapped in a customisable sachet of ultra-fine polyester mesh (~3-micron pores), acquired from JoTech Ltd©, Warwickshire (UK). The mesh enveloping the zeolite was used to create what is termed herein as the ‘drum’ (extraction device), and was employed in every treatment involving either of the zeolite forms. The drum was tethered to a scaffold that enabled it to be submerged in 300 ml of the experimental medium per beaker. Beakers were cling-filmed to control evaporation and 4-ml pipettes were respectively used to penetrate the respective beakers through the resealable cling film, where the medium was pipette-pumped multiple times every sampling time (see below) to homogenise the solution and mimic semi-stagnant water conditions; then the beakers were resealed accordingly.

Preparation of the model bio-trap of copper

Seeds of one cultivar of faba bean V. faba var minor (Harz) (local supplier, Manchester, UK) were used to grow V. faba plants as a model organism for the extraction of Cu²⁺ from the experimental medium. Vicia faba in aquaculture was chosen as a innovative means to extract and immobilise Cu²⁺, building on the fact that this legume is rich with proteins to which copper cations may have a great affinity, and the plant is easy to grow and maintain with robust shoots and roots.

The plants were grown individually in 9-cm round plastic pots using soil compost (Levington F2) and watered when needed. Around 10 days post-germination, each plant was carefully extracted from the compost, the root system was washed thoroughly in water baths to remove compost debris, and then each plant was transferred to a beaker containing 300 ml of the experimental medium. Our pilot trials showed that the plant adapted to the contaminant-free aquatic medium with no signs of root degradation and exhibited significant continuous growth over 14 days. The top of each beaker, where the plant root was submerged, was cling-filmed to minimise evaporation; the plant stem emerged through the cling film. Each plant was kept upright through suspension using a thread attached to a metallic scaffold, with the plant root fully submerged in the experimental medium (aquaculture) (Fig. S1). The plants were supplied with organic spirulina powder (Naturya©, Bath, UK) dissolved and homogenised in the control experimental medium The latter supplement 2 ml of (1 g/L) was injected into the medium every three days through the beaker cling-film seal.

Application of rare-earth magnet

A neodymium magnet disc (25 mm × 10 mm × 6 mm) with a vertical pull of 14.8 kg (N42, ~1.2 Tesla) (First4Magnets®, Tuxford, UK) was employed. Neodymium magnets, permanent potent static magnets made from rare-earth element alloys, are quite affordable and suitable for use at room temperature. The magnet was cling-filmed, to avoid corrosion, and then placed it in the centre of the beaker base underneath the submerged zeolite drum for 180 min starting from the beginning of the experiment. This was done to stimulate copper deposition on the zinc-reinforced surfaces of the zeolite, drawing on the rapid deposition process shown by Udagawa et al. (2014), see Fig. 1 for an illustration of the design. After the time-bound application, the magnet was carefully removed without affecting the integrity of the zeolite drum. The zinc-reinforced zeolite forms (via gel adhesion or plasma sputtering) were tested with and without the presence of the magnet. For the treatment levels where the V. faba bio-trap accompanied the zinc-gel-reinforced zeolite, the root of the hanging plant was placed in close proximity to the magnet, while the zeolite drum was submerged directly over the magnet (Fig. S1). The zeolites (raw or zinc-gel-reinforced) were tested concurrently with the application of the aquacultured V. faba, with and without magnet presence. Due to the aforementioned limitations with the sputtering, the application of the zinc-sputtered zeolite was not tested in the presence of V. faba.

Characterisation and analysis of copper extraction and materials

Energy Dispersive X-ray Spectrometer (EDAX) was used to perform a chemical microanalysis of the zeolite elemental composition and to assess Cu²⁺ extraction and accumulation onto the zeolite. We characterised the raw unmodified form of the ZSM-5, the zinc-free bare part of the reinforced zeolite, and the zinc-treated (zinc-reinforced) surface of the partly reinforced zeolite. The comparative characterisation helped measure copper extraction and whether the presence of zinc supported or inhibited copper adsorption by the zinc-free part of the zeolite. This was also accompanied by using a Scanning Electron Microscopy-Energy Dispersive X-ray Spectrometer (SEM/EDX) (FEI Quanta 200 Environmental; FEI Company©, Hillsboro, OR, USA) to visualise the morphology/structure of ZSM-5 and to further inspect areas in the zeolite surface for element accumulation and dispersion (copper in particular). Furthermore, supportive characterisation of the zeolite (before and after application) and across different treatments was carried out. X-ray Powder Diffraction (XRD) was done to examine the crystal structure of ZSM-5, using a X-ray diffractometer D8 Discover (Bruker©, Coventry, UK) at room temperature using Cu Kα radiation (λ = 1.5406 Å), from 5 to 90 2θ in 0.02° steps, with 1 s per step with 30 mA and 40 kV generator. Additionally, BET analysis was done to measure the surface area and pore volume of the zeolite using a Micromeritics ChemiSorb 2750 analyser (Micromeritics©, Norcross, GA, USA) at −196 °C, using an argon–helium mixture (10 vol % argon). ZSM-5 samples (~150 mg) were degassed at 300 °C under a vacuum (500 µm) for 240 m. The zeolite samples were preliminarily held in helium at 200 °C for 1 h.

An Inductively Coupled Plasma Optical Emission Spectrometry (ICP-OES) (Vista-PRO, Varian Inc©, Palo Alto, CA, USA) was used to measure the residual concentration of Cu²⁺, relative to the initial copper concentration (0.178 g/L), per treatment over time. A regression test (generalised linear model, glm) in the R environment with a quasiPoisson family (R Core Team, 2021), using the package ‘multcomp’ was applied (Hothorn, Bretz & Westfall, 2008). The explanatory variables were: (1) Treatment effect (six levels: raw zeolite (RZ) as the baseline, the bio-trap (V. faba) alone (F), zinc-gel-reinforced zeolite minus magnet without V. faba (ZnGZ-M-F), zinc-gel-reinforced zeolite plus magnet without V. faba (ZnGZ+M-F), zinc-gel-reinforced zeolite plus magnet with V. faba (ZnGZ-M+F), zinc-gel-reinforced zeolite plus magnet with V. faba (ZnGZ+M+F), (2) sampling time (four levels in minutes: 5, 20, 7,200 and 15,840), (3) the interaction (Treatment X sampling time). The total number of repeats across the treatments analysed statistically was 69, as a limited number of repeats was discarded due to machine or personal errors and/or necrosis of the exposed faba bean bio-trap. The sampling times were adopted based on: (1) Influenced by a neodymium magnet, copper displacement by zinc occurs rapidly (Udagawa et al., 2014) then plateaus quickly, justifying the 5 and 20 min samples. (2) Copper uptake by the bio-trap is gradual and accumulative, justifying later samples at 7,200 and 15,840 min. (3) To avoid detrimental effects of high copper exposure on the bio-trap, the 15,840 min sampling was applied based on our pilot. (4) The treatment lines of the current multifactorial experiment were run simultaneously with systematic sampling.