Removal of copper from cattle footbath wastewater with layered double hydroxide adsorbents as a route to antimicrobial resistance mitigation on dairy farms

Copper and zincare routinely used inlivestock antimicrobial footbathsincommercial farming. The footbath mix isacosttofarmers, and thedisposal of spent footbath into slurry tanksleads tosoilcontamination,aswellasthe potential for antimicrobial metal resistance and co-selection. This study assesses the potential to mitigate a source of antimicrobial metal resistance in slurry tanks while recovering copper and zinc from spent cattle foot-baths.Thisisthe ﬁ rst study inliteratureto investigate thepotentialofrecoveringcopperfromcattlefootbath so- lutions via any method. The sorbent, Ca 2 Al-EDTA Layered Double Hydroxides (LDH), were used to remove Cu 2+ fromaCu 2 SO 4 ·5H 2 0 solution atdifferenttemperatures.ThemaximumCu 2+ uptakefromtheCu 2 SO 4 ·5H 2 0solu- tion was568 ± 88mg g − 1 . Faster and higher equilibrium uptake was achievedby increasing the temperature of the solution. The sorbent was found to be effective in removing copper and zinc from a commercially available cattle footbath solution ( ﬁ ltered footbath solution Cu 2+ uptake 283 ± 11.05 mg g − 1 , Zn 2+ uptake 60 ± 0.05mgg − 1 ).Thus,thisstudydemonstratestheopportunityforacompletelynovelandpotentiallyeconomically bene ﬁ cialmethodofmitigatingantimicrobialresistanceinagricultureandtheenvironment,whilealsoproviding a new valuable copper and zinc waste stream for secondary metal production. © 2018 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/).


H I G H L I G H T S
• Copper and zinc are routinely used as antimicrobials in cattle footbaths. • Potential antimicrobial resistance (AMR) mitigation method examined. • Layered Double Hydroxides successfully removed copper and zinc from cattle footbath. • The kinetics of all sorption systems fitted well to a pseudo second-order model. • Novel and economically positive way to remove potential source of AMR established.

G R A P H I C A L A B S T R A C T
a b s t r a c t a r t i c l e i n f o

Introduction
Agricultural systems have been reported to be one of the key routes for the development and spread of antimicrobial resistance (AMR) within the environment (Department of Health, 2014). Heavy metals such as copper and zinc, have a long history of being used as antimicrobials in medicine and agriculture (Hobman and Crossman, 2015). Footbaths containing up to 10% copper and zinc sulphate are used on dairy farms for mass preventative treatment of cattle lameness caused by bacterial infections such as digital dermatitis (Holzhauer et al., 2012). After use, these footbath solutions are usually disposed of in slurry tanks, where they can reside for up to several months, before commonly being spread on to the land (Laven and Logue, 2006). Despite metals being resident in slurry tanks, heavy metals are not subject to reduction in this medium (Baker et al., 2016;Yu et al., 2017). If heavy metals are allowed to accumulate in the environment, they could co-select antibiotic resistance and metal resistance in microorganisms (Guo et al., 2018;Seiler and Berendonk, 2012), and increase levels of copper and zinc on farmland where slurry is spread. Thus, eliminating the disposal of waste cattle footbath into slurry tanks could potentially not only provide a plentiful source of recoverable copper and zinc, but also reduce the environmental impact of spreading slurry while mitigating this potential source of AMR co-selection and metal resistance (Department of Health, 2014;Hobman and Crossman, 2015).
AMR has been declared "a major global threat to public health" by the World Health Organisation, and the estimated associated economic costs by 2050 could be as high as 100 trillion US Dollars (Hobman, 2016). Antibiotic resistance can be acquired by bacteria via mutations of target proteins and/or regulatory genes or motifs, or through acquisition of resistance genes from other microorganisms (Pal et al., 2017). Metal resistance is a common phenotype in many microorganisms that are exposed to metals in their habitats (Hobman and Crossman, 2015), and antibiotic and metal resistance genes are frequently colocated on mobile genetic elements, providing the potential for coselection. Copper has traditionally been used in agriculture, and is increasingly being utilised in areas such as hospitals as an antimicrobial material due to its broad-spectrum activity against bacteria, viruses, yeasts and fungi (Schmidt et al., 2016). In addition to its use in cattle footbaths and as a fungicide, copper is also added to animal feed to promote growth due to its influence on the gut microbiota (Hobman and Crossman, 2015). However, the use of copper in animal feed has been associated with increased levels of AMR in microbial populations from farm animals and animal faeces . Although AMR has traditionally been viewed as a clinical issue, the propagation of resistance to antimicrobials is actually an environmental challenge, and both human and agricultural uses of antimicrobials have contributed to this situation (Department of Health, 2014;Summers, 2006). The influence of conditions where bacteria are stored for long periods in environments such as slurry tanks, which contain biologically relevant levels of antibiotics, and metals such as copper are crucial in understanding and combating these multi-drug resistance and transmission phenomena (Hobman and Crossman, 2015). Thus, removing AMR co-selection drivers such as metals in cattle footbaths in slurry tanks can only be seen as a positive step in the fight against AMR propagation.
Waste cattle footbaths could potentially be a significant source of recycled copper and zinc. A typical cattle footbath contains 25 kg of footbath mix and 200 L of water per 50-200 cows. Given that there are approximately 1.9 million dairy cattle in the UK (Bate, 2016), potentially up to 400 million L of footbath is being disposed every year through slurry tanks. Despite rising demand for copper, the environmental and economic impacts of increased production and declining ore grades means that copper mining is becoming more prohibitive and expensive (Northey et al., 2014). The main challenge of future raw material extraction is the escalating unit energy costs of production (Frenzel et al., 2017), and secondary production of copper requires ≤85% less energy than primary sources (Fu et al., 2017). Hence, secondary sources of copper will increasingly play a major role in copper supply. Sources of untapped waste copper, such as cattle footbaths, could be targeted as sources for recycled copper, providing farmers with a much-needed source of income and reduced waste burden. However, the recovery of copper from cattle footbaths via any method has not been explored to date.
The EU Water Framework Directive 2000/60/EC (WFD) (European Community, 2000) aims to deliver integrated river basin management for the whole of Europe. A key component of the Directive are the ecological targets for surface waters. Copper and zinc form part of the List II Group of metals which have a deleterious effect on the aquatic environment, and are subject to reduction programmes according to Directive 2006/11/EC on pollution caused by certain dangerous substances discharged into the aquatic environment (European Community, 2006). Heavy metals are non-biodegradable and can accumulate in living organisms (Fu and Wang, 2011). Some heavy metal ions are known to be toxic and carcinogenic, and are of particular concern in wastewater treatment (Burakov et al., 2018). Approaches to heavy metal removal in wastewater are flotation (Feng et al., 2017a(Feng et al., , b, 2018 and chemical precipitation (Rojas, 2014). Greatly influencing the solubility of heavy metal hydroxides is pH where more basic pH leads to a reduction in solubility, which can then be removed via sedimentation (Hamid et al., 2016). However, this results in large quantities of low density, highly hydrated sludge, which has subsequent disposal issues (Fu and Wang, 2011). Adsorption is an economically and technologically viable method of removing heavy metals from wastewaters (Ali et al., 2018), particularly for low density sludge with low pH values (González et al., 2015). Layered double hydroxides (LDHs) are emerging as an effective absorbent for wastewater treatment due to their low cost, simple preparation, and high sorption efficiency, while extending the opportunity for waste valorisation (Liang et al., 2013). LDHs are known as hydrotalcite-like compounds or anionic clays, containing positively charged layers and exchangeable anions in the interlayer (Chen and Song, 2013). They consist of brucite-like layers, with a partial M II for M III substitution, leading to an excess of positive charge, compensated with anions situated in the interlayer, together with water molecules (Pérez et al., 2006). LDHs can be used as scavengers of dyes (Mohamed et al., 2018), phosphates (Li et al., 2016), oxoanions (Goh et al., 2008), heavy metals (Asiabi et al., 2017), and phenols and pesticides (Choy et al., 2007). This is due to their anion exchange, acid base and adsorption capability, with LDHs removing heavy metals via chemical precipitation and chelation (Liang et al., 2013). Chemical precipitation occurs due to localised pH buffering effects from the high concentration of OH − . LDHs have been shown to be an effective solid matrix for immobilising metals without substantially altering the environmental pH (González et al., 2015). The majority of studies exploring the potential of LDHs to recover copper have used Mg-Al LDHs for copper (Cu 2+ ) adsorption from simulated wastewater, with Cu 2+ uptake between 20 and 181 mg g −1 in the literature (Anirudhan and Suchithra, 2008;Gong et al., 2011;González et al., 2015;Huang et al., 2015;Li et al., 2017;Ma et al., 2014Ma et al., , 2016Yang et al., 2016;Yue et al., 2016). Higher levels of Cu 2+ uptake have been reported for Zn-Al LDHs (450 mg g −1 (Chen and Song, 2013)) and for Ca-Al LDHs (381 mg g −1 (Rojas, 2014)). Ca-Al LDHs are usually used in cement fastening (Raki et al., 2004) and anion adsorption (Ghosal and Gupta, 2015), but demonstrated have high removal capacities and affinity for heavy metal ions (Rojas, 2014). Temperature is known to increase the uptake of copper for Mg-Al LDHs (Kameda et al., 2010), but the influence of temperature kinetics on Cu 2+ uptake using Ca-Al LDHs has not been investigated.
This study explores the potential to remove copper from cattle footbaths and present the first successful investigation into the removal of copper and zinc from a commercially available cattle footbath powder mix solution. Isothermal and kinetic analysis of copper removal from a simulated wastewater was conducted using Ca 2 -Al EDTA LDHs. Finally, the paper presents the real-world implications and applications of the work, and how this can be applied to aide AMR metal resistance and coselection mitigation in dairy farms.

Cattle footbath and slurry preparation
Duplicate samples of used (6 h old) cattle footbath samples were collected from a high-performance dairy farm in the East Midlands of the UK. High copper content footbath powder (Hoof trimming limited, product code 0924: 25 kg to 200 L) is used on a weekly basis on the farm to protect against cattle lameness caused by bacterial infections. Cu 2+ /Zn 2+ footbaths were prepared according to manufacturer's instructions as follows; 25 kg was mixed with 200 L of water to give a 5% Cu 2+ and Zn 2+ solution. The footbath was stirred before collection and was sampled from different parts of the footbath. The used footbath samples were sterilized at 121°C for 15 min. A 500 g portion of the same footbath powder was used for adsorption analysis experiments. The pH of the liquid samples was measured using a benchtop pH meter (HI5222, Hanna Instruments) with HI1131B glass body, double junction, combination pH electrode with HI7662-T temperature probe. Prior to taking the pH readings, the electrode was calibrated using buffers of pH 4.0, 7.01, and 10. After each reading the glass electrode was rinsed by deionised water before measuring the pH of the next sample.

Layered double hydroxide (LDH) synthesis and characterisation
In this study, two Ca-Al LDHs were formed for characterisation purposes; Ca 2 Al-EDTA and Ca 2 Al-NO 3 . Ca-Al LDHs have historically formed with nitrate ion (NO 3 − ) interlayer rather than carbonate (CO 3 2− ) type (Raki et al., 2004). NO 3 − has been found to have the lowest affinity for the LDH layers in ion exchange compared with CO 3 2− , SO 4 2− , Cl − (Bontchev et al., 2003;Châtelet et al., 1996). Due to this low affinity, the characterisation aimed to asses if the NO 3 − could be replaced by EDTA by adding sufficient quantities of EDTA during the synthesis process. Calcium-based LDHs exhibit a different crystal structure to other LDH materials, and preferentially form with nitrate ions in the interlayer rather than carbonate ions (Renaudin and François, 1999). LDH materials with Magnesium, and transition metals in the brucite like layers form preferentially with CO 3 2− ions, and as a result, are not as well suited to continuous synthesis with Na 2 EDTA precursors (Pavlovic et al., 2009;Pérez et al., 2006). Continuous hydrothermal synthesis was carried out in the same counter-current flow reactor arrangement ( Fig. 1) previously reported (Lester et al., 2012), with reaction conditions similar to those reported (Clark et al., 2017). Briefly, a mixed metal salt solution containing Ca(NO 3 ) 2 ·4H 2 O (0.067 M) and Al(NO 3 ) 3 ·9H 2 O (0.033 M) was flowed up into the reactor at a rate of 10 mL min −1 , while the base (0.125 M NaOH) and Na 2 EDTA (0.0167 M) solution was heated to 100°C temperature in a 2 kW band heater and flowed down into the reactor at 20 mL min −1 . Pressure in the system was maintained at 15 MPa by a back-pressure regulator (Pressure Tech, UK). Following synthesis, samples were centrifuged, washed thoroughly and oven dried overnight at 70°C.
Powder X-Ray Diffraction (XRD) analysis was completed using a Bruker D8-Advance diffractometer using Cu Kα radiation (λ = 1.5418 Å) in the 2θ range 5°-70°at a rate of 0.75°min −1 . XRD was used to establish LDH purity. The morphology of particles was elucidated with a Siemens XL30 (SEM) with an acceleration voltage of 20 kV. Nitrogen adsorption, using a Micromeritics Tristar II 3020 instrument at −196°C, and the Brunauer-Emmett-Teller (BET) method were used to calculate surface area. Thermal stability was ascertained through thermogravimetric analysis (TGA) on a TA Q500 TGA instrument between 25°C and 700°C at a heating rate of 5°C min −1 under N 2 flow at 40 mL min −1 . Fourier transform infra-red spectra were collected using a Bruker tensor FTIR ATR instrument between 500 and 1000 cm −1 and a resolution of 4 cm −1 .

Metal analysis
Multi-element analysis of diluted sterilized total footbath samples was undertaken by Inductively Coupled Plasma Mass Spectrometry (ICP-MS, iCAP-Q, Thermo Fisher Scientific) and individual element analysis of the experimental samples was undertaken using Atomic Absorption Spectrometry (AAS 272, Perkin Elmer). For ICP-MS analysis, the samples were first prepared via a total digestion method. 0.5 g liquid Fig. 1. A) Flow diagram of a continuous supercritical water reactor with animated depiction of nozzle reactor: P1 and P2 -Gilson HPLC pumps (P1 flow 20 mL min −1 P2 flow 10 mL min −1 ), V1 and V2check valves, P3 2 kW band heater, R 1 counter current reactor, P4 heater exchanger, P 5 45 μm filter, V 3 Pressure Tech back pressure regulator. B) Schematic diagram of counter current reactor: T 1 and T 2 -Thermocouples. C) Animated depiction of reactor in operation. and 0.2 g solid samples were mixed with 10 mL of concentrated nitric acid and then digested in a microwave digester (MARS X, CEM) at 190°C , 38 Bar, for 45 min. Two milliliter of concentrated hydrochloric acid and 2 mL of concentrated hydrofluoric acid was added, and the digestion was repeated for 25 min. Thirty milliliter of boric acid was then added, and the digestion was repeated for 15 min. The digested samples were then diluted for analysis by ICP-MS. The ICP-MS was run employing three operational modes, including (i) a collision-cell (Q cell) using He with kinetic energy discrimination (He-cell) to remove polyatomic interferences, (ii) standard mode (STD) in which the collision cell is evacuated and (iii) hydrogen mode (H 2 -cell) in which H 2 gas is used as the cell gas. Samples were introduced from an autosampler (Cetac ASX-520) incorporating an ASXpress™ rapid uptake module through a PEEK nebulizer (Burgener Mira Mist). Internal standards were introduced to the sample stream on a separate line via the ASXpress unit and included Ge (10 μg L −1 ), Rh (10 μg L −1 ) and Ir (5 μg L −1 ) in 2% trace analysis grade (Fisher Scientific, UK) HNO 3 . External multi-element calibration standards (Claritas-PPT grade CLMS-2 from SPEX Certiprep Inc.) included Ag, Al, As, Ba, Be, Cd, Ca, Co, Cr, Cs, Cu, Fe, K, Li, Mg, Mn, Mo, Na, Ni, P, Pb, Rb, S, Se, Sr, Tl, U, V and Zn, in the range 0-100 μg L −1 (0, 20, 40, 100 μg L −1 ). A bespoke external multi-element calibration solution (PlasmaCAL, SCP Science) was used to create Ca, Mg, Na and K standards in the range 0-30 mg L −1 . Phosphorus, boron and sulphur calibration utilised in-house standard solutions (KH 2 PO 4 , K 2 SO 4 and H 3 BO 3 ). In-sample switching was used to measure B and P in STD mode, Se in H 2 -cell mode and all other elements in He-cell mode. Peak dwell times were 10 mS for most elements with 150 scans per sample. Sample processing was undertaken using Qtegra™ software (Thermo-Fisher Scientific) utilising external crosscalibration between pulse-counting and analogue detector modes when required.

Metal sorption studies
The adsorption experiments aimed to assess the potential of the LDH to adsorb copper from simulated copper solutions, and copper and zinc from the footbath powder solution used on the dairy farm. The first set of experiments used copper (II) sulphate pentahydrate (Cu 2 SO4·5H 2 O) ≥ 98% (Sigma-Aldrich) dissolved in deionised water as a simulated wastewater. The final set of experiments analysed the sorption of copper and zinc from cattle footbath solution made from the commercially sourced footbath powder mix (Section 2.1).
All solutions were prepared with deionised water (18 MΩ Milli Q, Millipore System). For all experiments, a 1 L aqueous solution Cu 2 SO4·5H 2 O was prepared, and 200 ml was dispersed to 4 conical flasks. The flasks were placed in the incubator (Series 1A, LMS) and allowed to equilibrate for 1 h with constant stirring. Subsequently, 3 of the flasks had Ca 2 Al-EDTA LDH sorbent added, while the fourth acted as a control. At specified intervals sample aliquots were taken, and then centrifuged. After centrifuging, the supernatants were quantitatively diluted and used to determine the metal levels with AAS (Section 2.4).
For the isotherm experiments, 0.05 g of Ca 2 Al-EDTA LDH sorbent was added to 200 mL of aqueous Cu 2 SO 4 ·5H 2 O solution (100-1000 mg L −1 ). All isotherm experiments were conducted at 20°C within the incubator. Samples were taken from the base solution (t = 0) and then after 60 min. The metal adsorbed per gram of sorbent (q e ) was calculated according to the equation: where q e (mg g −1 ) is the equilibrium uptake, C 0 (mg L −1 ) is the initial Cu 2+ concentration, C e (mg L −1 ) is the Cu 2+ concentration at equilibrium, V (mL) is the volume of adsorbate and m (g) is the mass of adsorbent.

Footbath waste adsorption studies
In order to establish industrial feasibility of the EDTA modified LDH for Cu 2+ removal, the adsorbent was utilised in adsorption studies where the adsorbate was cattle footbath waste rather than a simulated wastewater solution of Cu 2 SO 4 ·5H 2 O. The footbath wastewater was used both filtered and unfiltered to establish the effect of the suspended solids (TSS) on the adsorption capacity.

Cattle footbath characterisation
Cattle footbaths have acidic pH values to ensure maximum solubility of the copper ions. The pH of the used footbath and fresh footbath solutions analysed in this study were 2.9 and 2.7 pH respectively, which is typical for solution containing heavy metal ions (Hashim et al., 2011) and prevents the precipitation of the copper hydroxide (Pavlovic et al., 2009). As manure accumulates in the footbath, due to the cows excreting as they walk through the footbath, the pH will increase, with the footbath becoming ineffective at a pH levels ≥ 4.5. Acidifiers such as sodium bisulphate are often added to the mix to safely the lower pH, thus increasing the solubility of the copper and extending its effectiveness. Fig. 2 shows that there are significant levels of sodium in the footbath (4879 mg L −1 ), which is likely to be due to the Sodium Bisulphate. The ICP also shows that the copper (6184 mg L −1 ) and zinc (8176 mg L −1 ) are at the expected levels of around 5%. The ICP shows that the waste footbath also includes high levels of other elements, which may adsorb to the LDH sorbent. The adsorption of these metals were not examined during this study. The full ICP analysis can be found in the supplemental data, Table S1. The pH of the LDH powder in deionised water was 11.2, and the pH of the Cu 2 SO 4 ·5H 2 O solution was 5.7.

Structural characterisation of LDH materials
In order to establish if EDTA was present in the LDH after synthesis, Ca 2 Al-NO 3 and Na 2 EDTA were also characterised to allow for assessment of crystal structure and surface area changes. The XRD patterns of the Ca 2 Al-EDTA and Ca 2 Al-NO 3 (Fig. 3) show that EDTA modified Ca 2 Al LDH exhibits some NO 3 intrusion into the interlayer region. The Ca 2 Al-EDTA sample is also less crystalline than the Ca 2 Al-NO 3 sample, with reflections at higher angles less defined than those in the NO 3 intercalated LDH. The lattice parameters for the Ca 2 Al-NO 3 material are 0.6 nm and 1.7 nm for a and c respectively. However, for the Ca 2 Al-ETDA LDH, there are multiple reflections at low Bragg angles, with reflections at 6.3°and 7.3°. The implication is that the reflections are both relating to EDTA in the interlayer region, but with different orientations (Rojas, 2014). Lattice parameters in the c plane relating to the reflections at 6.3°and 7.3°are 2.8 nm and 2.4 nm respectively.
The FTIR spectra for Ca 2 Al-EDTA, Ca 2 Al-NO 3 show broad absorption peaks between 3700 and 3000 cm −1 , relating to hydroxyl stretching in the metal hydroxide layers and hydrogen bonded water (Fig. 4).
The Ca 2 Al-NO 3 sample also shows a strong absorption band 1400-1300 cm −1 , which corresponds to the nitrate in the interlayer region. The spectra also show a relatively high NO 3 − content in the Ca 2 Al-EDTA sample, matching with the diffractograms. The absorption band between 1500 cm −1 and 1600 cm −1 represents the C_O bond in the CH 2 \ \C_O part of the EDTA. Fig. 5 shows SEM images of Ca 2 AL-EDTA LDH formations. There is a clear formation of distinct hexagonal platelets with size 1-3 μm across hexagonal surface for Ca 2 AL-EDTA (Fig. 5a) and 2-5 μm across hexagonal surface for Ca 2 Al-NO 3 (Fig. 5b). Additionally, some larger clusters of agglomerated LDH have formed, but the hexagonal plate-shape morphology is still present. The smaller platelet size and more disordered structure to the agglomerates is attributed as cause of the increased surface area (S BET ) in Ca 2 Al-EDTA when compared with Ca 2 Al-NO 3 . The S BET for Ca 2 Al-EDTA is 17.1 ± 0.1 m 2 g −1 and Ca 2 Al-NO 3 is 5.5 m 2 g −1 . The TGA and DTA diagrams (Supplementary data, Fig. S1) exhibit the typical thermal decomposition of LDH materials (Rojas, 2016). The first stage is the removal of surface (105°C) and bound (225°C) water, followed by dehydroxylation and decarbonation (up to 580°C for Ca 2 Al-NO 3 and 660°C for Ca 2 Al-EDTA). Thus, the LDH materials are thermally stable in the metal recovery temperatures of 10-40°C.

Adsorption isotherms
To establish if it was possible to recover Cu 2+ from waste cattle footbaths, a scoping study was conducted using simulated wastewater with 200 mg L −1 Cu 2 SO 4 ·5H 2 O and 0.05 g of Ca 2 Al-EDTA LDH. Fig. 6 shows the effect of contact time on the sorption capacity of Cu 2+ ions by Ca 2 Al-EDTA LDH. The majority of the sorption occurs rapidly within the process, and then stabilises to a much lower rate of adsorption, with equilibrium being reached after 45 min. For this scoping study,  56% (469 mg g −1 ) of the Cu 2+ ions were removed after 60 min. The final pH of the solution was 5.7, which is similar to previous studies (Huang et al., 2015).
The data from the adsorption studies was modelled according to the linear forms of the Langmuir (Eq. (2)) and Freundlich (Eq. (3)) isotherms: where q m is the maximum adsorption capacity of the LDH, K l is the Langmuir constant.
where K f and n are Freundlich constants. Fig. 7 shows the Cu 2+ isotherm at 20°C for Ca 2 Al-EDTA LDH. The shape of the isotherm corresponds to a H type isotherm based on Giles Classification (Giles et al., 1960), and has a similar profile to that of a Cu 2+ uptake isotherm previously reported (Rojas, 2014) for a Ca-Al LDHs. As Rojas noted, Ca -Al LDHs have a high affinity for Cu 2+ , which is also higher than Mg-Al and Zn-Al LDHs for heavy metals. The application of the Langmuir (Eq. (2)) and Freundlich (Eq. (3)) isothermal equations to the experimental data (Table 1) show that the Langmuir model fitted most closely with the isothermal experimental data, with a regression coefficient (R 2 ) of 0.996. This means that the sorption process occurs in a monolayer, and that the distribution of active sites on the sorbent is homogeneous. The low 1/n value in the Freundlich model suggests a potential for heterogeneity within the material surface. This could be due to the fact that there are crystals where there is Ca 2 Al-NO 3 and some crystals where there is Ca 2 Al-EDTA. This would result in a difference in adsorption mechanics between the materials as there would be no intercalation where the NO 3 − is intercalated. For this reason the Langmuir isotherm is focused on in the text. Based on the Langmuir model, the maximum theoretical Cu 2+ uptake (q m ) is 568 mg g −1 . The affinity for copper uptake is significantly higher than that obtained for Mg-Al or Zn-Al LDHs (Zubair et al., 2017), and Ca-Al LDH (Rojas, 2014). As noted by Rojas, the precipitation mechanism, leading to the formation of the corresponding heavy metal hydroxides (Rojas, 2014), determined the removal behaviour due to the similar q e values at high C e . Thus, the higher uptake of copper is due to the complexation of the copper in the EDTA interlayer, in addition to the precipitation mechanism.

Sorption kinetics
In order to establish rates of adsorption; pseudo-first (Eq. (4)) and pseudo-second order (Eq. (5)) rate equations Guzman-Vargas et al., 2016;Ni et al., 2007;Saiah et al., 2008;Sari and Tuzen, 2009) and the intra-particle diffusion model, developed by Weber and Morris (Doǧan and Alkan, 2003;Guzman-Vargas et al., 2016), were applied to adsorption data. The pseudo-first order rate equation can be expressed in its linear form as : where q t is the Cu 2+ uptake at a given time (mg g −1 ), t is the given time (min) and k 1 (min −1 ) is the pseudo-first order rate constant. The pseudo-second order rate equation can be expressed linearly as : where k 2 is the pseudo-second order rate constant (g mg −1 min −1 ). The intra-particle diffusion model can be presented as (Guzman-Vargas et al., 2016): where k id (mg g −1 min −1/2 ) is the intra-particle diffusion rate constant. The kinetics of sorption describes the removal of Cu 2+ ions by the LDH sorbents. In this study, the kinetics were analysed for Cu 2+ removal from a 200 mg L −1 Cu 2+ solution using Cu 2 SO 4 ·5H 2 O at 20°C, 30°C, and 40°C. Fig. 8 shows the effect of contact time on the Cu 2+ uptake in the presence of Ca 2 Al-EDTA LDH at 20°C, 30°C, and 40°C. There is a notable increase in uptake and rate of uptake as temperature increases. The most rapid increase in uptake during the first stage of sorption was for 40°C (63% after 5 min) and reached equilibrium in the shortest time. The highest removal rate (498 mg g −1 ) was also at 40°C. The lowest final uptake was at 20°C (427 mg g −1 ). Previous studies for Cu 2+ uptake with Mg-AL LDHS have taken between 2 and 24 h to equilibrate (González et al., 2015), which suggests that the rapid uptake and equilibrium of Ca 2 Al-EDTA LDHs is due to a different uptake mechanism to Mg-AL LDHs.
The experimental kinetic data was fit to the pseudo-first-order (Eq. (4)), pseudo-second-order (Eq. (5)) as well as the Weber and Morris intra-particle diffusion (Eq. (6)) kinetic model equations, the parameters and linear regression coefficients (R 2 ) of which are presented in Table 2. Based on the correlation coefficient, the pseudo-second-order was the best fit for all the experimental data. This has been noted for several other similar studies for Mg-Al LDHs (González et al., 2015;Huang et al., 2015;Ma et al., 2014;Yang et al., 2016), but has not previously been noted for Cu 2+ uptake kinetics of Ca-Al type LDHs. Additionally, the q e values obtained via the pseudo-second-order model were similar to those obtained experimentally. This suggests that the pseudo-second-order model will work well for this LDH for the whole-time range. Thus, the second order sorption mechanism appears to be dominant, and the overall rate of metal uptake was controlled by a chemisorption process (Ho, 2006). The fit of the experimental data to the intra-particle diffusion model Eq. (6) provides an insight into the overall rate of sorption ( Table 2). The intercept value (C) provides an indicator of the boundary layer thickness, with higher values representing a greater boundary layer effect. Table 2 shows that the intercept values obtained are much higher than those obtained in other studies (González et al., 2015), suggesting that intra-particle diffusion plays a strong role in the sorption process. Additionally, the high k ip values indicate that there are many active sites on the surface of the Ca 2 Al-EDTA LDH for Cu 2+ ions to attach to, thus enhancing the diffusion rate.

Adsorption thermodynamics
The thermodynamic parameters of Gibbs free energy change ΔG°, standard enthalpy ΔH°, and standard entropy ΔS°of the kinetic data were also analysed to gain a greater insight into the effect of temperature on the adsorption. The thermodynamic parameters ΔG°, ΔH°, and ΔS°were calculated using the following equations: where R is the ideal gas constant (8.314 kJ kmol −1 K −1 ), T the temperature (K) and K d the distribution coefficient. The plot of ln K d against 1/T Fig. 6. Effect of contact time on sorption capacity (q t ) of Ca 2 Al-EDTA LDH (0.05 g) for copper (II) sulphate pentahydrate solution at an initial concentration of 200 mg L −1 . N = 3 with error bars of standard deviation. Fig. 7. Sorption isotherm of Cu 2+ by Ca 2 Al-EDTA LDH in the copper (II) sulphate pentahydrate solution. q e is the equilibrium sorption capacity (mg g −1 ), and C e is the equilibrium concentration of metal ions in solution (mg L −1 ). N = 3 with error bars of standard deviation. gives a straight line (Fig. 9); the slope and the intercept correspond to ΔH°/R and ΔS°/R, respectively (Zaghouane-Boudiaf et al., 2012). ΔG°v alues at different temperatures used in experiments are calculated using the following equation: The thermodynamic parameters obtained from the analysis are presented in Table 3. The negative values of ΔG°at different temperatures indicate a thermodynamically favourable process due to the spontaneous nature of the adsorption. The change of free energy for chemisorption is a range of −80 to 400 kJ mol −1 (González et al., 2015). The ΔG°w ere between −6.18 and −4.55 kJ mol −1 , which indicates that the nature of adsorption was spontaneous and confirmed the affinity of material for Cu 2+ . These values were in the middle of physisorption and chemisorption, and thus the process was a physical adsorption enhanced by a chemical effect. The decreasing values of ΔG°with increasing temperature suggest that the process increases in spontaneity at higher temperature. The positive ΔH°(19.3 kJ mol −1 ) shows that the adsorption process was endothermic, and the positive ΔS°( 81.3 kJ mol −1 ) of the system indicates an increase in the randomness at the interface adsorbent/adsorbate during adsorption.

Potential for metal recovery from cattle footbath waste
While removing metal from simulated wastewater shows the potential for wastewater treatment using adsorbents, it is only by testing the adsorbents on real world wastewaters that the true capacity of LDH adsorbents can be explored. For Cu 2+ removal from a commercial footbath solution, the kinetics were analysed at 20°C in an incubator from filtered and unfiltered 200 Cu 2+ mg L −1 cattle footbath mix solutions. The aim was to analyse if the sediment inhibits copper uptake. For the filtered solution, the zinc concentration was also measured after treatment. Fig. 8 shows the Cu 2+ uptake for filtered and unfiltered footbath solutions and the Zn 2+ uptake for the filtered footbath. Overall the percentage Cu 2+ uptake is lower (filtered 50% and unfiltered 49% - Fig. 10) than that of the simulated wastewater (67%) after 60 min (Fig. 6). However, the uptake did not fully reach equilibrium within this time, and thus there is potential that the final uptake could be higher. The final uptake was 283 ± 11 mg g −1 filtered and 279 ± 22 mg g −1 unfiltered, which was lower than that of the simulated Cu 2 + only wastewater (568 mg g −1 ). However, the pH of the footbath solution (3.92) was lower than that of the simulated wastewater (4.97). In aqueous solutions, Cu 2+ has been shown to be the dominant species at pH below 6 (Li et al., 2017), and Gong et al. (2011) found that solutions with an initial pH below 4 showed low initial uptake due to partial dissolution of the sorbent by acidic hydrolysis. Thus, the lower initial uptake could be related to the lower pH of the footbath solution, indicating that pH control is crucial to optimising the solution. No dissolution was experienced due to pH in this study. In addition to pH, the other elements and compounds within the footbath mix (Fig. 2) could be interfering with the sorption of the copper by occupying active sites on the LDH surface. It is important to note that the copper used in the footbath would be of a lower grade than the reagent grade (N98% purity) used for the simulated portion of this study. There is still a high initial uptake of copper (34% for filtered footbath) but is lower than the 44% uptake for the simulated wastewater. There is also the potential for competition for the same sorption sites, which has previously been noted for multi-component solutions and Ca-Al LDHs (Rojas, 2014).  As with the simulated wastewater study, the kinetic analysis showed that the pseudo second order model fits best with the Cu 2+ experimental data (R 2 = 0.998 filtered, R 2 = 0.994 unfiltered), with q e experimentally being close to that obtained with the model (Table 2). This shows that the pseudo-second-order model is applicable to single and multicomponent solutions for this sorbent, and that chemisorption processes also dominate the uptake mechanisms. The fit of the Zn 2+ experimental data was poorer, but the pseudo-second-order model still fitted well (R 2 = 0.919), with q e experimentally once more being close to that obtained with the model. While the uptake values of Zn 2+ were low (60 mg g −1 ) compared to Cu 2+ (283 mg g −1 ), they are still relatively high compared to the reported uptake of multiple metals on Mg-Al LDHs (Zubair et al., 2017), which use reagent grade metal solutions. The results imply that there is a stronger affinity for Cu 2+ over Zn 2+ , which has also been noted for Pb 2+ and Cd 2+ for other LDHs (González et al., 2014(González et al., , 2015. Furthermore, the ICP data (Fig. 2) showed that several other metals were present in the footbath. Their removal potential was not investigated in this study as the focus was on the AMR metals of zinc and copper. However, the potential and affinity of LDHs to remove all of these metals requires further investigation.
The results demonstrate for the first time that it is possible to remove Cu 2+ and Zn 2+ ions from commercially available acidic pH cattle footbath solution using LDH adsorbents. Thus, these adsorbents could be used as heavy metal scavengers for low pH wastewaters with high pollutant concentrations. Research on novel sorbent materials rarely includes an assessment of scale up potential. Large scale manufacture of sorbent materials is required for them to be a feasible alternative to granular activated carbon, which is currently used at industrial scale. The LDHs used in this study were produced at gram scale using a laboratory scale counter-current flow reactor (Clark et al., 2017) and now at pilot and full scale . To date, few systems have been demonstrated in literature at larger scales . Large scale continuous synthesis of LDHs is a relatively uncommon field, and only in batch systems . However the continuous hydrothermal synthesis of LDHs and organic-modified LDHs utilises the same reactor design as some other continuously produced nanomaterials, which has been demonstrated on large scale (Caramazana-González et al., 2017;Caramazana et al., 2018;Dunne et al., 2016a, b;Lester et al., 2018;Munn et al., 2015).
It is known that high levels of copper and zinc in slurry wastewater disposed to land can lead to high levels of the metals in soil and crops (Jondreville et al., 2003). Given the potential for AMR metal resistance and co-selection due to bacterial exposure to high metal concentrations (Guo et al., 2018), there is a need to review the standard method of disposing of cattle footbath into farm slurry tanks. Currently there are no specific policy requirements which control the amount of footbath waste which can be disposed of in slurry tanks, or any limits on the duration the footbath waste can be stored before being spread on to the land. Further work is required to explore the influence of footbath solutions on AMR in slurry tanks. Furthermore, the results of this study fit with the objectives of the EU circular economy action plan (European Community, 2015) to boost the market for secondary raw material and water reuse. Given that there is the potential for up to 400 million L of waste cattle footbath to be used annually in the UK, the use of these footbaths is a considerable cost to farmers which could be mitigated by recovering and reusing the metals in the spent footbath solution, as well as being a large secondary source of copper and zinc sulphate. Thus, copper and zinc recovery from cattle footbath   wastes could be a substantial business opportunity for the farming community.

Conclusions
This study assesses the potential to remove copper from cattle footbaths. Isothermal and kinetic analysis of copper removal from a simulated wastewater have been presented for Ca 2 -Al EDTA LDHs. This paper presents the first successful investigation into the removal of copper and zinc from a commercially available cattle footbath powder mix solution.
The Ca 2 -Al EDTA LDHs used in this study removed a substantial amount of Cu 2+ from a Cu 2 SO 4 ·5H 2 0 solution (568 mg g −1 ). Increasing the temperature of the solution resulted in accelerated uptake, with higher equilibrium uptake reached in a shorted time. The pseudosecond-order model was found to be the most appropriate to describe the kinetics for both single and multi-component solutions investigated for the Ca 2 Al-EDTA LDHs. Thus, the overall rate of metal uptake is controlled by chemisorption processes. The study showed for the first time that it is possible to adsorb Cu 2+ and Zn 2+ from a commercially available cattle footbath solution using adsorbents. While the uptake rate and final equilibrium value were lower than that of the single component solution, the uptake was higher than those recorded in literature for Mg-Al LDHs (filtered footbath solution Cu 2+ uptake 283 mg g −1 , Zn 2 + uptake 60 mg g −1 ).
The study has shown that Ca 2 Al-EDTA LDHs have the potential to effectively remove large quantities of copper and zinc from cattle footbath solutions without having to alter the environmental pH of the solution. Given that potentially nearly 400 million L of cattle footbath waste being disposed of annual into slurry tanks in the UK alone, there is a significant opportunity to access a valuable copper and zinc waste stream for secondary copper production, as well as remove a potential route of AMR co-selection and metal resistance on dairy farms.