Evaluation of the Immobilization of Coexisting Heavy Metal Ions of Pb²⁺, Cd²⁺ , and Zn²⁺ from Water by Dairy Manure-Derived Biochar: Performance and Reusability

Abstract

Heavy metals including Cd, Pb, and Zn are prevalent stormwater and groundwater contaminants derived from natural and human activities, and there is a lack of cost-effective treatment for their removal. Recently, biochar has been increasingly recognized as a promising low-cost sorbent that can be used to remediate heavy metal contaminated water. This study evaluates the immobilization/release performance of dairy manure-derived biochar (DM-BC) as a sustainable material for competitive removal of coexisting heavy metal ions from water and explains the underlying mechanism for regeneration/reusability of biochar. Results showed that the metal ions exhibited competitive removal in the order of Pb²⁺ ≫ Zn²⁺ > Cd²⁺. The pH played a decisive role in influencing metal ion speciation affecting the electrostatic attraction/repulsion and surface complexation. Higher pH led to greater removal for Pb²⁺ and Cd²⁺, whereas Zn²⁺ showed maximum removal at pH ≈ 7.5. Diffuse reflectance infrared spectroscopy, scanning electron microscopy, and X-ray diffraction confirmed the interactions and precipitation reactions of oxygen-containing functional groups (e.g., ─OH, $\text{C}{\text{O}}_3^{2-}$, and Si─O) as key participants in metal immobilization. Langmuir, Freundlich, and Redlich–Peterson isotherm modeling data showed varied and unique results depending on the metal ion and concentration. The removal kinetics and model fitting showed that the three steps of intraparticle diffusion might be more representative for describing the immobilization processes of metal ions on the external surface and internal pores. In the flow-through columns, DM-BC effectively retained the mixed metal ions of Cd²⁺, Pb²⁺, and Zn²⁺ showing 100% removal for the duration of the column run over three cycles of regeneration and reuse.

Introduction

Pollution associated with heavy metals causes serious health and safety concerns because of their toxicity and persistence in the environment (Mishra et al. 2019). Although some heavy metals are nutritionally essential in trace amounts, most become harmful or carcinogenic when their concentrations exceed certain tolerance levels (US EPA 2018; Sfakianakis et al. 2015; Langston 2018; Nandi et al. 2012; Fäth et al. 2018).

Heavy metals such as cadmium (Cd), zinc (Zn), and lead (Pb) release into environments from point sources and nonpoint sources (Du et al. 2015; Ismail et al. 2016; Cowden and Aherene 2019; Rahman et al. 2019). A nonpoint source of particular concern is stormwater runoff. Stormwater collects an array of contaminants, including Cd, Pb, and Zn, from a variety of impermeable and semi-permeable surfaces in urban areas (Sandoval et al. 2019). Cd, Pb, and Zn are included in the list of 126 priority pollutants regulated by the US EPA’s Clean Water Act (CWA), which provides guidance for national pollutant discharge elimination systems (NPDES) (US EPA 2010). These metals are three of the top five heavy metals that contaminate surface water, groundwater, and soil and pose significant concerns to public health and ecological well-being (Liu et al. 2018; US EPA 2018, 2019). Because of the human health risks and other ecotoxicological effects, the US EPA has set maximum contamination levels (MCLs) for heavy metals in drinking water, including Cd (0.005 mg L⁻¹), Pb (0.015 mg L⁻¹), and Zn (5.0 mg L⁻¹) (US EPA 2018).

To comply with the NPDES permits, meet MCLs, and improve environmental health and safety, the eradication of heavy metals from stormwater and groundwater becomes necessary. Common treatment methods, such as coagulation and flocculation, chemical precipitation, ion exchange, reverse osmosis, membrane separation, filtration processes, and electrochemical techniques have proved expensive because they either require specialized chemicals/reagents and apparatus or coproduce a considerable quantity of metal-containing hazardous wastes (Gunatilake 2015; Sikdar and Kundu 2018; Gupta et al. 2015; Crini et al. 2019; Bolisetty et al. 2019).

Considering the presented drawbacks of common treatment technologies, sorption is regarded as a remedy for the substantial volume of heavy metal polluted water through immobilization onto cost-effective materials (Xue et al. 2012; Inyang et al. 2016). For the sake of this research, the removal mechanism “sorption” is defined to include adsorption, precipitation, coprecipitation, electrostatic attraction, and ion exchange. Currently, an extensive assortment of carbonaceous sorbents, such as activated carbon (AC), is used in the elimination processes of heavy metals from water. However, the depletion of coal-based products because of climate change is causing a commercial resource crisis for AC generation, leading to the urgency of alternative sorbents (Chen 2015).

Biochar is receiving attention as a low-cost sorbent for the remediation of polluted water. By pyrolyzing organic matter (e.g., forest and plant debris and animal waste) in an oxygen-depleted atmosphere, the carbon-rich byproduct, biochar, is produced. The physicochemical characteristics of biochar are variable depending on the feedstock type, pyrolysis temperature, and various prefeedstock and postfeedstock treatments (Singh et al. 2017). These physical parameters are useful in determining the removal processes and long-term effectiveness of biochar as a remedial material for heavy metal contaminated water (Ahmad et al. 2014; Jiang et al. 2016; Godwin et al. 2019). Research has primarily focused on monometal systems but has also explored the competitive nature of heavy metal extraction by different biochar (Doumer et al. 2016; Gazi et al. 2018; Godwin et al. 2019; Ni et al. 2019; Shan et al. 2020; Abdin et al. 2020). The competitive immobilization/release behavior and underlying mechanism for heavy metal removal by biochar is still an area in need of research, especially concerning the remediation of multimetal polluted water. Of the wide range of biochar feedstocks, the production of dairy manure-derived biochar (DM-BC) creates positive effects on sustainable waste management and environmental protection. Thus, it is important to expand the understanding of mechanistic preferences and competitive removal of coexisting heavy metals using DM-BC.

The main objectives of this investigation are to evaluate the immobilization/release performance of DM-BC as a sustainable material for the removal of metals in a complex competitive aqueous system and to explain the underlying processes for regeneration/reusability of DM-BC. In addition, uncovering the capacity and reusability of DM-BC could create a pathway for it to be applied as a promising replacement for conventional sorbents, such as AC.

Materials and Methods

Chemical Reagents

All chemicals used in this study were reagent grade of 99% purity or better. Chemicals were purchased from Fischer Scientific, Thermo Scientific, or Sigma-Aldrich. A complete list of chemicals can be found in the Supplemental Materials (Table S1).

Characterization of Dairy Manure-Derived Biochar

DM-BC was supplied by collaborators from an industrial vendor, who did not disclose the specific pyrolysis conditions. The DM-BC was sieved through a 2-mm sieve and stored in closed vessels until usage.

The experimental procedures and methods to characterize DM-BC are described in our previous study (Wallace et al. 2019). The pH of the biochar was measured as follows. Briefly, 2.5 g of biochar was weighed in a 50-mL polypropylene tube with 25 mL of DI water or 10 mM NaCl. The sample was shaken at 200 rpm for 1 h, removed from the shaker and left standing for 30 min, and then measured for pH. The values of pH at the point of zero charge (pH_PZC) for each biochar were determined using a modified method described by Tan et al. (2008).

The measurement of specific surface area for the biochar was completed using a Quantachrome NOVA 2000e Surface Area and Pore Size Analyzer (BET). To study the mineralogy of the biochar, X-ray diffraction (XRD) analysis was executed on a Rigaku Miniflex X-ray diffractometer (Ultima IV, Riigaku, Japan). Surface morphology, elemental compositions of the biochar were captured by a Leo-Zeiss 1450VPSE scanning electron microscope (SEM, Carl Zeiss Microscopy, US) equipped with an EDAX Genesis 4000 XMS SYSTEM 60 energy-dispersive spectrometer (EDS). Surface functional groups were examined using diffuse reflectance infrared Fourier transform spectroscopy (DRIFTS) with a Bruker Vertex 70 Fourier Transform Infrared Spectroscopy (FTIR) system. SEM/EDAX and DRIFTS analysis were completed on the DM-BC before and after the sorption of heavy metals. Comprehensive details for all methods are provided in the Supplemental Materials.

Batch Experiments

Batch experiments were carried out to explore the competitive immobilization/release performance of mixed heavy metals onto DM-BC. In this study, mixed heavy metals in both chloride and nitrate salt forms were used to evaluate the effect of different anions of equal ionic strength on the competitive removal of mixed metal ions. All batch tests including sorption kinetics and isotherms used 100 mg of biochar sample added to a 50-mL polypropylene tube and mixed with 30 mL metal stock solution. The solution contained 10 mM NaCl or 10 mM NaNO₃ background with each metal (1.0 mM unless otherwise noted) in the chloride salt forms: cadmium chloride (CdCl₂), lead nitrate (Pb(NO₃)₂ (substituted because of the insolubility of lead chloride), and zinc) chloride (ZnCl₂); or nitrate salts forms: cadmium nitrate (Cd(NO₃)₂), lead nitrate (Pb(NO₃)₂), and zinc nitrate (Zn(NO₃)₂).

During the experimental operation, the solution pH was not adjusted and allowed to drift freely to equilibrium. The tubes were incubated for 24 h on an agitator shaker at a constant speed (200 rpm) at ambient room temperature. Prior to metals analyses, liquid samples were filtered through a 0.22-μm mixed cellulose ester (MCE) membrane filter (Sigma-Aldrich, US). Samples were fixed with 0.30 mL of 70% nitric acid, diluted 10 times, and stored at 4°C to await metals analysis. Each batch experiment was conducted in duplicate, and the results are presented as averaged values with calculated standard deviations.

To understand the effects of solution pH on the sorption of mixed heavy metals using DM-BC, batch experiments were conducted as described with Cd²⁺, Pb²⁺, and Zn²⁺ ions (1.0 mM) in both chloride (10 mM NaCl) and nitrate (10 mM NaNO₃) systems. The pH was adjusted using 1 M HCl or NaOH from 3 to 11 with the increment of one unit. Liquid samples were acidified and analyzed as detailed in the batch experiment section.

Sorption Kinetics

For the sorption kinetics experiments, the tubes were incubated under static conditions at ambient room temperature (25°C). The pH values were allowed to drift freely and became stable at 5.8 ± 0.16. Liquid samples were collected at time intervals of 2, 4, 6, 10, 24, 48, 96, and 168 h and were acidified with 0.5 mL of 30% HNO₃ and stored for metal analysis by ICP-OES. The experimental data were fitted with commonly used removal kinetic models including a pseudo-first order, PFO model (Lagergren 1898), a pseudo-second order (PSO) model (Ho and McKay 1999), and intraparticle diffusion models (IDM) (Cheung et al. 2007; Wu et al. 2001). The model parameters are summarized in Table S2.

Sorption Isotherms

For the sorption isotherm experiments, individual metal ion (Cd²⁺, Pb²⁺, and Zn²⁺) solutions were made at concentrations of 6, 12, 24, 48, 96, 192, and 768 mg L⁻¹. Samples were incubated for 24 h on an agitator shaker (200 rpms) at ambient room temperature (25°C).The pH was adjusted to 7.00 ± 0.25 at time zero. If the pH drifted over the duration of the experiment, it was corrected to 7.00 ± 0.25 at 24 h. In this study, 10 mM NaCl or NaNO₃ was used as a background electrolyte to maintain solution ionic strength. After incubation, the liquid samples were prepared and stored as described in the batch experiment section for metal analysis. Results were modeled using Langmuir, Freundlich, and Redlich–Peterson isotherm models (Table 4).

Table 4.

Model parameters of Freundlich, Langmuir and Redlich–Peterson isotherms for the sorption Cd2þ and Zn2þ in single metal system

	Freundlich			Langmuir			Redlich-Peterson
Material	KF [(mg g−1) (mg L−1)−1/n]	n	R 2	KL (L mg−1)	qmax (mg g−1)	R 2	Kr (L mg−1)	αR (L mg −1)β	β	R 2
(Concentrations in the range of 6 to 768 mg L−1)
DM-BC Cd2+	1.8 ± 0.3	2.2 ± 0.1	0.977	0.03 ± 0.0	34.4 ± 2.3	0.974	1.2 ± 0.4	0.2 ± 0.2	0.7 ± 0.1	0.992
Zn2+	5.1 ± 0.2	2.9 ± 0.2	0.924	0.17 ± 0.0	26.4 ± 2.2	0.939	3.9 ± 2.9	0.1 ± 0.5	1.1 ± 1.2	0.927
(Concentrations in the range of 6 to 96 mg L−1)
DM-BC Cd2+	1.5 ± 0.4	1.7 ± 0.3	0.831	0.08 ± 0.01	16.8 ± 2.5	0.973	1.0 ± 0.0	0.001 ± 0.002	p2.059	0.998
Zn2+	1.1 ± 0.2	1.7 ± 0.3	0.771	0.07 ± 0.03	14.9 ± 4.4	0.842	N/C	N/C	N/C	N/C

Removal of Mixed Metal Ions of Cd², Pb², and Zn² in Column Study

DM-BC was used as an individual sorbent in continuous fixed-bed columns to investigate the immobilization/release behavior of mixed metal ions of Cd²⁺, Pb²⁺, and Zn²⁺ in a chloride system and to illustrate the removal capacities and stability through the regeneration-reuse processes. Glass columns (cross-sectional area as 4.91 cm² and height as 30 cm) were used as fixed-bed (30 cm of the bed depth) up-flow reactors and packed with 42.5 g of DM-BC. During the operation of each column, the influent containing mixed metal ions of Cd²⁺, Pb²⁺, and Zn²⁺ at 1.0 mM each in 10 mM NaCl were pumped through the packed columns in an up-flow mode with a peristaltic pump at a flow rate of 1.0 mL min⁻¹ at ambient room temperature. Control columns were operated using only 10 mM NaCl without the inclusion of heavy metals. The pore volume was measured at 22.6 ± 0.3 mL (n = 2) in the treatment columns and 26.9 ± 0.3 mL (n = 2) in the the control columns.

The removal capacity at the point of breakthrough (q_B) is defined as the amount of removal when the effluent concentration of the metals reaches 10% or lower of the initial influent concentration of 1.0 mM. The removal capacity at the point of exhaustion (q_E) is defined as the effluent concentration of metal ions reaching 90% or higher of the influent concentration. After the point of exhaustion was observed, the column was left standing up to 24 h to allow most of the remaining pore water to drain by gravity overnight. The column was then sparged with N₂ gas for 5 min at 20 psi to ensure all residual pore water was expelled from the column.

Fresh pore volumes of 10 mM NaCl were subsequently run through the column to promote desorption in the exhausted columns until no metal ions were detected or no further decrease of metals was detected in the effluent. The effluent was collected every 22 min in the collection vessels on the fraction collector, and the pH was measured immediately. Liquid samples were filtered with a 0.22 μm filter, acidified with 0.5 mL of 30% HNO₃, and analyzed for metals by ICP-OES.

Once the desorption experiment was completed, the column was regenerated by dewatering as specified and soaked with 2.0 M HCl. HCl (2 M) has been reported to be an effective reagent to desorb heavy metals from carbonaceous sorbent materials including AC (Rao et al. 2009; Anirudhan and Sreekumari 2011) and biochar (Vilvanathan and Shanthakumar 2018; Kołodyńska et al. 2017). Finally, the columns were again rinsed with 10 mM NaCl to remove all the sorbed metals and then air-dried under N₂ gas. Two more cycles of sorption-desorption were run on the columns to evaluate the effectiveness of the regenerated DM-BC on the competitive removal of mixed metal ions of Cd²⁺, Pb²⁺, and Zn²⁺.

The Thomas model [Eq. (1)] is commonly applied to calculate the removal performance in a continuous fixed-bed column (Thomas 1944)

$$\mathrm{l}\text{n}\left(\frac{C_0}{C_e}-1\right)=\frac{k_T{q}_{T}M}{𝒬}-k_T{C}_{0}t$$ (1)

Where k_T = Thomas rate constant (mL=min mg); q_T = equilibrium metal ions uptake per g of biochar (mg g⁻¹); C₀ = influent metal ions concentration (mg L⁻¹); C_e = effluent metal ions concentration at time t (mg L⁻¹); M = mass of biochar (g); 𝒬 = filtration velocity (mL min⁻¹); and t = time of influent passed through the column. The parameters k_T and q_T are calculated from the plot of ln[(C₀/C_e) − 1 versus time (t). Additional details of QA/QC procedures for experiments are found in the QA/QC section of the Supplemental Materials.

Results and Discussion

Batch Removal of Mixed Heavy Metal Ions

Among the three coexisting heavy metal ions, Pb²⁺ ions showed much higher preferential removal over the other two metal ions from water (Fig. 1). In addition, no substantial difference was observed for the removal of heavy metal ions by DM-BC in either nitrate or chloride system. Results confirmed the competitive metal removal in the order of Pb²⁺≫ Zn²⁺> Cd²⁺ n both chloride and nitrate systems.

Fig. 1.

Removal of mixed metal ions of Cd²⁺, Pb²⁺, and Zn²⁺ (each at 1.0 mM) after 24 h in a chloride (10 mM NaCl) or nitrate (10 mM NaNO₃) system. The pH was stable at 5.8 ± 0.2. A 100% removal would be equivalent to the amount of sorption of 33.6 mg g⁻¹ for Cd²⁺, 62.16 mg g⁻¹ for Pb²⁺, and 19.61 mg g⁻¹ for Zn²⁺.

The observed favored removal of Pb²⁺ ions over Zn²⁺ and Cd²⁺ ions on the DM-BC is congruent with previous studies using bio-char and other sorbents (Ding et al. 2016a, b; Park et al. 2016) and is attributed to the physicochemical properties of Pb²⁺ ions, such as a smaller hydrated radius, a higher electronegativity, and a lower hydrolysis constant (pKH for Pb²⁺ at 7.71, Cd²⁺ at 9.0, Zn²⁺ at 10.1) (Park et al. 2016). Therefore, the Pb²⁺ ion is more favorably removed through inner-sphere sorption and surface complexation than Cd²⁺ and Zn²⁺.

DM-BC Characterization Results

The physiochemical characteristics including specific surface area (SSA), particle size, pH, and point of zero charge (pH_PZC) were determined for DB-BC and summarized in Table 1. The SSA was measured at about 158.6 m² g⁻¹, which is higher than the typical range of 8 to 132 m² g⁻¹ (Leng et al. 2021). The larger surface area provides more potential adsorption sites for the removal of heavy metals. In addition, the measured pH_PZC was approximately 10.5, suggesting that basic functional groups were dominant on the biochar surface.

Table 1.

Physiochemical characteristics of DM-BC (n = 2)

Physiochemical characteristics	DM-BC
SSA (m2 g−1)	158.6 ± 2.7
Size (mm)	≤2
pH (DI water)	10.4 ± 0.14
pH (10 mM NaCl)	9.9 ± 0.0
pHPZC	10.5 ± 0.05

Effect of Surface Functional Groups

As indicated by DRIFTS analysis (Fig. 2), the functional groups on the surface of DM-BC include carbonate/calcite ($\text{C}{\text{O}}_3^{2-}$, 1,430 cm⁻¹), phenolic (─OH, 1,390−1310 cm⁻¹), aliphatic (─CH₃, ─CH₂, 2990–2840 cm⁻¹), and silicate (Si─O, 1,030 cm⁻¹). These functional groups are commonly identified on the surface of biochar (Azargohar et al. 2014; Sing et al. 2017). The presence of functional groups such as ─OH, $\text{C}{\text{O}}_3^{2-}$ and Si─O can influence the removal of metal ions from aqueous solution (Gu et al. 2019; Uddin 2017; Vhahangwele et al. 2015). The ─OH surface groups provide ion exchange sites where the divalent metal ions Me²⁺ replace H⁺ ions, $\text{C}{\text{O}}_3^{2-}$ promotes heavy metal precipitation when $\text{C}{\text{O}}_3^{2-}$ is released to form metal carbonates (e.g., PbCO₃) (Yang et al. 2019). Changes of IR absorbance peak intensity identify evidence of functional group influence in the biochar at equilibrium compared to the pristine DM-BC. This is determined in the increased peak height of the metal sorbed biochars for functional groups $\text{C}{\text{O}}_3^{2-}$ and Si─O, as well as the slight broadening of the OH⁻ peaks.

Fig. 2.

DRIFTS analysis of pristine and metal sorbed DM-BC in the chloride system using DM-BC for the removal of Cd²⁺, Pb²⁺, and Zn²⁺ ions at 12.4, 28.9, 11.4 mg g⁻¹ respectively. Samples were incubated at room temperature over 24 h at pH 7.

XRD analysis for pristine DM-BC displays the distinct quartz, calcite, and graphite patterns (Fig. 3). The metals retained on DM-BC primarily formed crystalline carbonate minerals, including otavite (CdCO₃), cerussite (PbCO₃), hydrocerussite (Pb₃(CO₃)₂)(OH)₂), and smithsonite (ZnCO₃), although XRD peaks of these metal carbonates overlap those of calcite. XRD data also indicated the structure of wulfingite (Zn(OH)₂. The mineralogical characterizations from XRD analysis present) consistent evidence congruent with DRIFTS analysis that surface complexation and/or precipitation of metal carbonate and/or hydroxides play an important role in controlling the removal of mixed metal ions from aqueous solution by DM-BC.

Fig. 3.

X-ray diffractograms of pristine DM-BC and metals loaded DM-BC. Single metal ion of Cd²⁺, Pb²⁺, or Zn²⁺ at an initial concentration of 96 mg L⁻¹ was sorbed onto 100 mg of DM-BC7 in 10 mM NaCl. The pH was controlled at 7. Metal sorption capacity was 12.4 ± 0.03, 28.9 ± 0.1 and 11.4 ± 0.7 mg g⁻¹ for Cd²⁺, Pb²⁺, and Zn²⁺ ions, respectively.

SEM-EDS analysis was operated for the sorption of single metal ions of Cd²⁺, Pb²⁺, or Zn²⁺ on DM-BC from the sorption isotherm experiments (768 mg L⁻¹) (Fig. S1). Results show that the SEM images of DM-BC do not have a uniform porous surface. Sorption/complexation and precipitation as aggregates were not determined on the DM-BC, although these patterns could apply for DM-BC. The intense EDS peaks verified the presence of each metal bonded on the biochar’s surface.

Influences of Solution pH

pH can affect the speciation of metal ions and solubility of metal hydroxides; however, these effects vary between individual metal hydroxides (Pagnanelli et al. 2000; Sheng et al. 2004). Fig. 4 reveals that the removal efficacy of mixed metal ions (Cd²⁺, Pb²⁺, and Zn²⁺) by DM-BC was enhanced with increasing pH up to 10 in both the chloride and nitrate systems, except for Zn²⁺ ion removal. The exception of Zn²⁺ ions removal may be connected to the release of dissolved organic matter (DOM) because metal complexation with DOM increases at higher pH which amplifies metal solubility (Borggaard et al. 2019; Weng et al. 2002). For the removal of Pb²⁺ ions, results revealed that as the solution pH increased to 7, the Pb²⁺ ions removal by DM-BC increased and attained a maximum of 100% removal at pH 7 and then became steady with the continuous increase of pH. When the solution pH was below 7, the removal of Pb²⁺ ions was most likely due to surface sorption onto biochar. As the solution pH increased above 7, removal via precipitation also came into play when the Pb²⁺ ions precipitated as Pb(OH)₂ (Sočo and Kalembkiewicz 2016; Sheng et al. 2004; Lodeiro et al. 2006). However, the production of cerus-site (PbCO₃), hydrocerussite (Pb₃(CO₃)₂ (OH)₂) as suggested the XRD patterns, and the strong presence of carbonate in the pristine biochar indicates that cerussite and hydrocerussite were probably the dominant species formed during removal. By contrast, the sorption of Zn²⁺ ions only occurred when the solution pH was 6 and higher. The removal of Zn²⁺ ions by DM-BC achieved the maximum of 70% around pH 7.5 but fell to zero with the continuous rise of pH to 8 and higher. The reduced removal may be connected to the Zn complexation with organics from the DM-BC that kept Zn in soluble forms (Borggaard et al. 2019). As for the Cd²⁺ ions, negligible removal of Cd²⁺ ions was detected in the chloride system, while the removal in the nitrate system increased at pH 8 and plateaued at pH 10.

Fig. 4.

Effects of solution pH on the competitive removal of mixed metal ions of Cd²⁺, Pb²⁺, and Zn²⁺ (1.0 mM each) in a chloride (10 mM NaCl) or nitrate (10 mM NaNO₃) system by DM-BC. The solution pH was adjusted from 3 to 10 at 1-unit increments.

The pH_PZC for DM-BC was measured to be 10.5 and is within the published range of biochar pH_PZC values (Karunanayake et al. 2017; Dewage et al. 2018; Suliman et al. 2016; Gogri 2017). As the solution pH rose to 10, nearing DM-BC’s pH_PZC, the surface charges of biochar changed from mostly positive to near neutral, alleviating the repulsion and enhancing the sorption of metal cations. Again, the removal of the three metal ions followed the favored order of Pb²⁺≫ Zn²⁺> Cd²⁺, which is in accord with preceding observations and literature reports (Park et al. 2016; Xu et al. 2013).

The removal for the metal ions was credited to (1) decreased H⁺ ion concentration that allowed the biochar surface to become negatively charged and enhance the attraction of the metal ions in keeping with metal ion speciation chemistry (Sočo and Kalembkiewicz 2016; Lodeiro et al. 2006); and (2) immobilization of metal ions on biochar via precipitation reactions (Lei et al. 2019).

Removal Kinetics

The removal kinetics of mixed metal ions of Cd²⁺, Pb²⁺, and Zn²⁺ by DM-BC were modeled in both chloride and nitrate systems (Fig. 5). Solution pH buffered naturally and was stable at 5.8 for the duration of the experiments. The results demonstrated that there was no significant difference in removal patterns seen between chloride and nitrate systems.

Fig. 5.

Removal kenitics of mixed metal ions of Cd²⁺, Zn²⁺, and Pb²⁺ (1.0 mM each) by DM-BC at pH 5.8 in chloride (10 mM NaCl) or nitrate (10 mM NaNO₃) systems.

The removal of Pb²⁺ by DM-BC in the chloride system increased, achieving a removal capacity of 59.5 mg Pb²⁺ g⁻¹ biochar (>95%) at 24 h and then slowly rose to a maximum removal capacity at 62.0 mg Pb²⁺ g⁻¹ biochar (99.5%) at 168 h. DM-BC displayed a gradually increased removal of Cd²⁺ and Zn²⁺ ions over the experimental duration, reaching the highest removal capacity at 8.7 mg g⁻¹ biochar (26%) and 9.9 mg g⁻¹ biochar (50%) at 168 h. The experimental data were analyzed and modeled using the PFO, PSO, and IDM, which are summarized in Table 2. In summary, the PSO model best describes the removal kinetics of mixed metal ions of Cd²⁺, Pb²⁺, and Zn²⁺, evidencing that chemisorption was the rate-limiting mechanism for the retention of Cd²⁺, Pb²⁺, and Zn²⁺ ions on DM-BC (Inyang et al. 2016; Momčilović et al. 2011). Although the PFO model showed R² values above 0.85 for the removal of Cd²⁺ and Zn²⁺ ions, these values were not considered to be reliable owing to the high uncertainty reflected in the standard deviations.

Table 2.

Parameters of the pseudo-first order, pseudo-second order, and intraparticle diffusion models for the removal of mixed metal ions at pH 5.8

								Intraparticle diffusion
	Pseudo-first order			Pseudo-second order				Step 1a		Step 2b
Biochar metal	k1 (h−1)	qe(calc) (mg g−1)	R 2	k2 (g/mg h)	qe(exp)	qe(calc)		ki1 (g/mg h0.5)	R 2	ki2 (g/mg h0.5)	R 2
					(mg g−1)		R 2
DM-BC
Cd2+ (Cl−)	0.02 ± 1.8 × 10−3	1.47	0.961	0.016	8.71	9.05	0.991	0.43 ± 0.84	0.858	4.16 ± 1.18	0.735
Cd2+ ($\text{N}{\text{O}}_3^{-}$)	0.03 ± 5.0 × 10−3	1.26	0.856	0.028	8.71	8.85	0.998	0.26 ± 0.51	0.961	5.75 ± 1.09	0.533
Pb2+ (Cl−)	0.03 ± 1.1 × 10−3	2.97	0.660	0.006	61.94	62.81	0.999	2.99 ± 5.7	0.880	57.50 ± 1.56	0.619
Pb2+ ($\text{N}{\text{O}}_3^{-}$)	0.03 ± 8.0 × 10−3	3.01	0.717	0.006	61.94	62.62	0.999	2.05 ± 3.98	0.952	58.05 ± 0.32	0.959
Zn2+ (Cl−)	0.01 ± 1.0 × 10−3	1.89	0.957	0.007	9.97	10.12	0.966	0.28 ± 0.54	0.885	2.57 ± 0.53	0.969
Zn2+ ($\text{N}{\text{O}}_3^{-}$)	0.01 ± 1.7 × 10−3	1.89	0.847	0.006	9.97	9.90	0.933	0.11 ± 0.23	0.979	1.79 ± 1.13	0.884

^aFirst step represents the instantaneous external surface removal and fast-pace gradual removal.

^bSecond step reflects the slow equilibrium stage of intraparticle diffusion.

Compared with metals removal by other biochars, DM-BC showed similar, and in some cases, a better kinetic model fit (Table 3). The results further support DM-BC having a promising potential to efficiently remove heavy metal ions from water in multimetals systems. However, very few studies are available evaluating the PFO and PSO models in mixed metal systems since most literature has focused on single metal systems. Therefore, future research needs to focus more on the removal kinetics in mixed metal systems by biochar.

Table 3.

Best fit model parameters of pseudo-first order (PFO), pseudo-second order (PSO) and intraparticle diffusion model (IDM) for the removal of Pb²⁺,Cd²⁺, and Zn²⁺ ions using biochars

Biomass type [temperature (°C)]	Metal	Model	Parameter 1	Parameter 2	R 2	pH	Reference
Corn Straw (600)	Cd2+	PFO	k1 (min−1) = 0.027	qe1 (mg g−1) = 25.87	0.969	4–6	Zhou et al. (2018)
		PSO	k2 (g mg−1 min−1) = 2.04 × 10−3	qe2 (mg g−1) = 26.32	0.997
Modified Corn Straw (600)	Cd2+	PFO	k1 (min−1) = 0.019	qe1 (mg g−1) = 94.06	0.968	4–6	—
		PSO	k2 (gmg-1min-1) = 3.47 × 10−4	qe2 (mg g−1) = 100.0	0.998
Activated Sludge (80)	Cd2+	PFO	k1 (min−1) = 0.062	qe1 (mmol g−1) = 0.167	0.987	6	Kusvuran et al. (2012)
		PSO	k2 (min−1)=0.762	qe2 (mmol g−1) = 0.167	0.993
		IDM	ki1 (mmol g min−1) = 28.7 × 10−3	ki2 (mmol g min−1) = 2.3 × 10−3	0.989(ki1), 0.992(ki2)
	Pb2+	PFO	k1 (min−1) = 0.046	qe1 (mmol g−1) = 0.155	0.960	6	—
		PSO	k2 (min−1) = 0.632	qe2 (mmol g−1) = 0.155	0.996
		IDM	ki1 (mmol g min−1) = 24.3 × 10−3	ki2 (mmol g min−1) = 2.01 × 10−3	0.996(ki1), 0.929(ki2)
Water Hyacinths (450)	Cd2+	PFO	k1 (min−1) = 0.0123	qe1 (mg g−1) = 43.94	0.84	5	Ding et al. (2016a)
		PSO	k2(min−1) = 0.00042	qe2 (mg g−1) = 46.79	0.94
	Pb2+	PFO	k1 (min−1) = 0.0162	qe1 (mg g−1) = 44.78	0.82	5	—
		PSO	k2(min−1) = 0.00055	qe2 (mg g−1) = 47.33	0.94
Corn Straw (300)	Zn2+	PFO	k1 (h−1) = 0.130	qe1 (mg g−1) = 7.11	0.921	5	Chen et al. (2011)
		PSO	k2 (gm g−1 h−1) = 0.006	qe2 (mg g−1) = 8.20	0.999
Hardwood (450)	Zn2+	PFO	k1 (h−1) = 0.141	qe1 (mg g−1) = 2.63	0.923	5	—
		PSO	k2 (gm g−1 h−1) = 0.009	qe2 (mg g−1) = 3.14	0.988
Peanut Shell (500)	Pb2+	PSO	k2 (gm g−1 h−1) = 0.928	qe2 (mg g−1) = 38.0	0.997	5	Wang et al. (2015)
Canna indica (500)	Cd2+	PFO	k1 (h−1) = 27.6	qe1 (mg g−1) = 98.09	0.994	5	—
		PSO	k2 (gm g−1 h−1) = 1.57	qe2 (mg g−1) = 98.52	0.999
		IDM	ki1 (g/mgh0.5) = 5.09	—	0.884

Sorption Isotherms

In this study, batch experiments were conducted to illustrate the sorption isotherm for each individual metal ion on DM-BC. The experimental data were modeled to fit the Langmuir, Freundlich, and Redlich–Peterson isotherm equations, except for the removal of Pb²⁺ ions. The Pb²⁺ ions achieved 100% removal by DM-BC for the tested concentrations, representing the range of high contamination levels of 6 to 96 mg L⁻¹ and higher concentrations (192 and 768 mg L⁻¹). Fig. 6 shows the Langmuir, Freundlich, and Redlich–Peterson isotherm model fitting for the removal of Cd²⁺ and Zn²⁺ ions (6 to 768 mg L⁻¹) using DM-BC. For the sorption of Cd²⁺ and Zn²⁺ ions on DM-BC, the R² values reflect a best fit to the Redlich–Peterson isotherm (0.992) for Cd²⁺ ions and to the Langmuir isotherm (0.939) for Zn²⁺ ions. The β value (0.7) for the Redlich–Peterson model was lower than the unity, suggesting that the Cd²⁺ ions had not reached maximum coverage onto both the homogenous and heterogeneous surfaces of DM-BC. The Langmuir isotherm suggests that sorption of Zn²⁺ ions formed a monolayer on a homogenous surface of DM-BC.

Fig. 6.

Single metal ion sorption isotherms using metal concentrations ranging from 6 to 768 mg L⁻¹ over 24 h. The solution pH was controlled at 7.

The experimental data were also examined for isotherm model parameters only at concentrations in the range of 6 to 96 mg L⁻¹ (Fig. 7 and Table 4). No significant difference was observed between the high concentration and lower range, suggesting the levels of metal ions were not an influencing factor in this study. The Redlich–Peterson model best represents the sorption isotherm of Cd²⁺ ions on DM-BC with an R² of 0.998; whereas the sorption of Zn²⁺ ions best fits the Langmuir isotherm, but with an R² value of only 0.84. Thus, multiple active sites on the heterogeneous surface of DM-BC exhibited unique affinities for the removal of Cd²⁺ and Zn²⁺ ions.

Fig. 7.

Single metal ion sorption isotherms using metal concentrations ranging from 6 to 96 mg L⁻¹ over 24 h. The solution pH was controlled at 7.

The sorption isotherms from this study were compared to published data including single metal systems (Table 5) and mixed metals systems (Table 6). No congruent conclusions can be formed regarding the best fitting of isotherm models and the calculated removal capacity for Cd²⁺, Zn²⁺, and Pb²⁺ ions. This is mainly due to the unique physicochemical characteristics of biochar made from various feedstock under different pyrolysis conditions.

Table 5.

Single metal removal using biochar: best fit of sorption isotherm model parameters

Biochar type [temperature (°C)]	Metal	Model	Parameter 1	Parameter 2	R 2	pH	Reference
Pig Manure (400)	Cd2+	Langmuir	qmax (mg g−1) = 107	KL (1 mg−1) = 0.002	0.969	6	Kotodynska et al. (2012)
		Freundlich	KF (mg g−1) = 2.07	n = 3.82	0.970
	Pb2+	Langmuir	qmax (mg g−1) = 175	KL (1 mg−1) = 0.011	0.996	6
		Freundlich	KF (mg g−1) = 5.99	n = 4.95	0.920
	Zn2+	Langmuir	qmax (mg g−1) = 62.3	KL (1 mg−1) = 0.005	0.985	5
		Freundlich	KF (mg g−1) = 2.89	n = 2.10	0.985
Diary Manure (400)	Cd2+	Langmuir		KL (1 mg−1) = 0.001	0.990	6
		Freundlich		n = 3.06	0.830
	Pb2+	Langmuir		KL (1 mg−1) = 0.002	0.990	6
		Freundlich		n = 4.04	0.858
	Zn2+	Langmuir		KL (1 mg−1) = 0.006	0.970	5
		Freundlich		n = 2.70	0.928
Peanut Hull (400)	Pb2+	Langmuir	qmax (mg g−1) = 49.9	KL (1 mg−1) = 0.59	0.912	5	Wang et al. (2015)
		Freundlich	KF (mg g−1) = 25.1	1 = 0.119	0.968
Medicine Residue (400)	Pb2+	Langmuir	qmax (mg g−1) = 82.5	KL (1 mg−1) = 0.58	0.932	5
		Freundlich	KF (mg g−1) = 40.5	n = 0.121	0.976
Canna indica (600)	Cd2+	Langmuir	qmax (mg g−1) = 140	KL (1 mg−1) = 1.03	0.876	5	Cui et al. (2016)
		Freundlich	KF (mg g−1) = 52.8	n = 0.26	0.740
Pinewood (300)	Pb2+	Langmuir	qmax (mg g−1) = 3.89	KL (1 mg−1) = 0.36	0.98	5	Liu and Zhang (2009)
		Frenudlich	KF (mg g−1) = 1.75	n = 4.77	0.47
Rice husk (300)	Pb2+	Langmuir	qmax (mg g−1) = 1.84	KL (1 mg−1) = 0.21	0.92	5
		Freundlich	KF (mg g−1) = 0.35	n = 2.07	0.95
Corn straw (300)	Zn2+	Langmuir	qmax (mg g−1) = 11.0	KL (1 mg−1) = 0.232	0.998	5	Chen et al. (2011)
		Freundlich	KF (mg g−1) = 2.84	n = 3.336	0.898
Hardwood (450)	Zn2+	Langmuir	qmax (mg g−1) = 4.54	KL (1 mg−1) = 0.061	0.998	5
		Freundlich	KF (mg g−1) = 0.72	n = 2.827	0.941
Pine Bark (400/450)	Cd2+	Langmuir	qmax (mg g−1) = 0.34	KL (1 mg−1) = 0.0002	0.743	5	Mohan et al. (2007)
		Freundlich	KF (mg g−1) = 0.40	n = 0.35	0.778
	Pb2+	Langmuir	qmax (mg g−1) = 3.0	KL (1 mg−1) = 0.226	0.963
		Freundlich	KF (mg g−1) = 1.28	n = 0.15	0.961
Oak Wood (400/450)	Cd2+	Langmuir	qmax (mg g−1) = 0.37	KL (1 mg−1) = 0.037	0.575	5
		Freundlich	KF (mg g−1) = 0.23	n = 0.12	0.584
	Pb2+	Langmuir	qmax (mg g−1) = 2.62	KL (1 mg−1) = 0.163	0.908
		Freundlich	KF (mg g−1) = 0.77	n = 0.22	0.948

Table 6.

Multimetal removal using biochar: best fit of sorption isotherm model parameters Table 7. Background element release (mg L⁻¹) of control columns vs. metal removal columns

Biochar type [temperature (°C)]	Metal	Model	Parameter 1	Parameter 2	R 2	Reference
Sesame Straw (700)	Cd2+	Langmuir	qmax (mg g−1) = 5	KL (1 mg−1) = 0.04	0.978	Park et al. (2016)
		Freundlich	KF (mg g−1) = 1.2	1/n = 0.232	0.971
	Pb2+	Langmuir	qmax (mg g−1) = 88	KL (1 mg−1) = 0.03	0.988
		Freundlich	KF (mg g−1) = 0.29	1/n = 0.279	0.994
	Zn2+	Langmuir	qmax (mg g−1) = 7	KL (1 mg−1) = 0.04	0.986
		Freundlich	KF (mg g−1) = 1.4	1/n = 0.279	0.997
	Cr2+	Langmuir	qmax (mg g−1) = 21	KL (1 mg−1) = 0.05	0.992
		Freundlich	KF (mg g−1) = 1.8	1/n = 0.495	0.950
	Cu2+	Langmuir	qmax (mg g−1) = 40	KL (1 mg−1) = 0.03	0.956
		Freundlich	KF (mg g−1) = 2.4	1/n = 0.5137	0.985
Dairy Manure (350)	Pb2+	Langmuir	qmax (mmol kg−1) = 789	Kl (1 mmol−1) = 4.9	0.97	Xu et al. (2013)
		Freundlich	KF (mmol kg−1) = 704	n = 2.48	0.92
	Cu2+	Langmuir	qmax (mmol kg−1) = 297	Kl (1 mmol−1) = 3.01	0.97
		Freundlich	KF (mmol kg−1) = 203	n = 4.09	0.88
	Zn2+	Langmuir	—	—	—
		Freundlich	—	—	—
	Cd2+	Langmuir	—	—	—
		Freundlich	—	—	—
Rice Husk (350)	Pb2+	Langmuir	qmax (mmol kg−1) = 79.9	Kl (1 mmol−1) = 0.14	0.98
		Freundlich	—	—	—
	Cu2+	Langmuir	qmax (mmol kg−1) = 27.4	Kl (1 mmol−1) = 0.14	0.98
		Freundlich	—	—	—
	Zn2+	Langmuir	—	—	—
		Freundlich	—	—	—
	Cd2+	Langmuir	—	—	—
		Freundlich	—	—	—
Chicken Bone (600)	Cd2+	Langmuir	qmax (mg g−1) = 53	KL (1 mg−1) = 0.039	0.957	Park et al. (2015)
		Freundlich	KF (mg g−1) = 4.39	1/n = 0.442	0.986
	Cu2+	Langmuir	qmax (mg g−1) = 107.5	KL (1 mg−1) = 0.069	0.984
		Freundlich	KF (mg g−1) = 7.87	1/n = 0.548	0.985
	Zn2+	Langmuir	qmax (mg g−1) = 43.9	KL (1 mg−1) = 0.026	0.956
		Freundlich	KF (mg g−1) = 3.82	1/n = 0.399	0.971

However, a few studies show that the sorption of Cd²⁺, Pb²⁺, and Zn²⁺ ions using biochar derived from pig manure, dairy manure, and anaerobically digested sludge fit well to the Langmuir isotherm, suggesting monolayer sorption on a finite number of identical surface sites of biochar (Kołodyńska et al. 2012; Ni et al. 2019). The Langmuir isotherm also well described the sorption of Pb²⁺ ions on biochar made from pinewood (Liu and Zhang 2009) and pine bark (Mohan et al. 2007). In addition, Cui et al. (2016) showed that the Langmuir model described the removal of Cd²⁺ ions by C. indica derived biochar, in which the sorption mechanism was attributed to precipitation, ion exchange, complexation with functional groups, and coordination with π electrons. By comparison, the removal of Zn²⁺ ions using corn straw and hardwood-derived biochars demonstrated a best fit for the Langmuir isotherm (Chen et al. 2011), which is consistent with the results in this study for the isotherms of individual metal ion of Cd²⁺, Pb²⁺, and Zn²⁺ (6 to 96 mg L⁻¹). The Freundlich isotherm exhibited the best fit for the sorption of Pb²⁺ ions by biochar made from peanut hull and medicine residue (Wang et al. 2015), rice husk (Liu and Zhang 2009), and oak wood (Mohan et al. 2007), indicating heterogeneous sorption affinity.

The DM-BC showed similar or higher model calculated sorption capacity (q_max) than those from many of the reported biochars, including corn straw, rice husk, and pine wood. Similar phenomena were observed for the removal of mixed metal ions, and studies consistently confirm that Pb²⁺ shows preferential removal in a multimetal system with a variety of biochar feedstocks (Shan et al. 2020), although much fewer studies are available in the literature. Further research in this area is warranted.

Column Studies

In addition to the batch experiments and solid phase characterization, column studies were carried out to explore the competitive removal and immobilization stability of mixed metal ions of Cd²⁺,Pb²⁺, and Zn²⁺ on DM-BC in a continuous flow-through system over three cycles of regeneration and reuse. The influent of metal columns contained the mixed metal ions of Cd²⁺, Pb²⁺, and Zn²⁺ with 10.0 mM NaCl, while the influent of control columns consisted of only 10.0 mM NaCl (i.e., free of mixed metal ions).

Leachable Elements from Control and Treatment Columns

To understand the leachable metals and other elements released from the DM-BC in a continuous flow-through system, metal columns and control columns filled with pristine DM-BC were operated in duplicate over a period of 315 min (equal to 12 pore volumes). In the control columns, no metal ions of Cd²⁺, Pb²⁺, and Zn²⁺ were detected in the effluent during the entire period, nor in the acid wash (2 M HCl). The biochar showed release of background elements including [Al], [Ca], [K], [Mg], [Na], and [Si] (Fig. 8). The competitive removal of mixed metal ions of Cd²⁺, Pb²⁺, and Zn²⁺ along with the release of background elements, [Al], [Ca], [K], [Mg], [Na], and [Si] on DM-BC is presented in Fig. 9.

Fig. 8.

Release of elements from pristine DM-BC control columns flushed with 10 mM NaCl over 315 min. pH (free drift) was stable at 10.5 (±0.08).

Fig. 9.

Breakthrough of mixed metal ions Cd²⁺, Pb²⁺, and Zn²⁺ (each at 1 mM) with 10 mM NaCl along with release of background elements [Al], [Ca], [K], [Mg], [Na], and [Si] from DM-BC columns.

The concentration of dissolved elements from pristine DM-BC in control columns decreased overall for [Al], [Na], [Mg], and [Ca], while [K] and [Si] showed the release of less than 1 mg L⁻¹ over 315 min. In addition, [Al] and [Mg] were released in the control column but were not detected in the treatment columns. In the treatment columns, high initial release at 15 min was noticed at the concentrations of 558.7 and 25.0 mg L⁻¹, respectively, for [Al] and [Mg]. Conversely, elements released in the treatment columns but not in the control were [K] and [Si]. The [K] ion showed an initial release at 15 min at a concentration of 11,097.7 mg L⁻¹ that decreased over the duration of the cycle to 613.3 mg L⁻¹. [Si] was initially released to 25.7 mg L⁻¹ after 15 min, then increased to 45.4 mg L⁻¹, but then decreased to 22.6 mg L⁻¹. Elements released in both control and metals columns were [Ca] and [Na]. The [Ca] was released approximately twice as much in the metal columns compared to the control columns, whereas the initial release of[Na] was 5 times higher concentration in the control columns compared to the metal columns. Table 7 compares the release of leachable elements in the control columns and the metal columns.

Table 7.

Background element release (mg L−1) of control columns vs. metal removal columns

	[Al]		[Ca]		[K]		[Mg]		[Na]		[Si]
Time	Control	Metal removal	Control	Metal removal	Control	Metal removal	Control	Metal removal	Control	Metal removal	Control	Metal removal
15	558.7	0.1	2.6	4.0	0.7	11,097.9	25.0	0.0	12,660.9	2,379.0	0.0	25.8
45	426.1	0.0	0.7	1.0	1.0	2,564.9	36.6	0.0	2,509.0	560.1	0.0	43.3
75	217.3	0.2	0.0	0.6	0.7	1,319.8	38.6	0.0	1,100.1	295.3	0.0	45.4
180	125.3	0.0	0.0	0.5	0.2	836.3	28.2	0.0	228.5	195.9	0.0	33.0
315	105.1	0.0	0.0	0.5	0.3	613.3	21.2	0.0	465.8	189.5	0.0	22.6

Competitive Removal of Mixed Metal Ions in Treatment Columns

No breakthrough of three mixed metal ions Cd²⁺, Pb²⁺, and Zn²⁺ on DM-BC was observed for the metal columns over the three cycles of regeneration and reuse, suggesting all mixed metal ions were immobilized on the DM-BC inside the columns (Fig. S2). The results illustrated that DM-BC effectively retained the mixed metal ions of Cd²⁺, Pb²⁺, and Zn²⁺. The complete removal of metal ions was attributed primarily to the precipitation of metal hydroxides (Sočo and Kalembkiewicz 2016) and/or surface complexation with mineral components such as $\text{C}{\text{O}}_3^{2-}$ and $\text{Si}{\text{O}}_3^{2-}$ (Gu et al. 2019; Yang et al. 2019) since the effluent pH values were 10.6, 8.2, and 7.7 and stable over the course of the test cycles. As discussed previously, when the solution pH is higher than 7–8, all three metal ions predominantly form metal hydroxides and/or carbonates and immobilize on the surface of DM-BC (Li et al. 2019).

Based on the results from both batch and continuous flow-through column experiments, multiple mechanisms controlling the competitive removal of mixed metal ions of Cd²⁺, Pb²⁺, and Zn²⁺ from water by DM-BC were summarized as follows. The physicochemical properties such as hydrated radius, electronegativity, and hydrolysis constant (pKH) determine the preferential removal of mixed metal ions with the order of Pb²⁺≫ Zn²⁺> Cd²⁺. In addition, the solution pH plays a decisive role in controlling the metal ion species in solution, the surface charge, and solubility of metal minerals and in enhancing the electrostatic attraction/repulsion, surface complexation with mineral components such as $\text{C}{\text{O}}_3^{2-}$ and $\text{Si}{\text{O}}_3^{2-}$, and chemical precipitation (e.g., metal hydroxides) of metals on biochar. Furthermore, multiple active sites on the heterogeneous surface of DM-BC demonstrate different affinities for the mixed metal ions.

Conclusion

In this study, the efficiency and mechanisms of removal of mixed metal ions of Cd²⁺, Pb²⁺, and Zn²⁺ from water in both static and continuous flow-through systems were investigated using DM-BC.DM-BC was examined for physicochemical characterization, surface interaction mechanisms, removal capacity and kinetics in batch tests, and regeneration-reuse behavior in continuous fixed-bed column experiments. DM-BC showed the promising potential to be an effective and reusable material for the long-term remediation of mixed metals polluted water. The main conclusions are summarized:

1. DM-BC showed the competitive removal of mixed metal ions following the preferential order of Pb²⁺≫ Zn²⁺> Cd²⁺. The preferential removal of Pb²⁺ ions over Cd²⁺ and Zn²⁺ ions is attributed to physicochemical properties of Pb²⁺ ions, such as a smaller hydrated radius, higher electronegativity, and lower hydrolysis constant (pKH).

2. The removal sequence of mixed metal ions depends on the special properties of metal ions and their unique interactions with DM-BC under specific solution conditions. Among the various influencing factors, the solution pH plays a decisive role in controlling the metal ion species in solution, surface charge, and solubility of metal minerals. Consequently, the pH affects the electrostatic attraction/repulsion, surface complexation with oxygen-containing functional groups (e.g., ─OH, $\text{C}{\text{O}}_3^{2-}$ and Si─O), and chemical precipitation of metal carbonate and hydroxides on biochar. These interactions and precipitation reactions were suggested using DRIFTS, SEM/EDS, and XRD analysis.

3. Results for the Langmuir, Freundlich, and Redlich–Peterson isotherm models showed that at a range of 6 to 768 mg L⁻¹ the sorption of Cd²⁺ and Zn²⁺ ions on DM-BC showed a best fit to the Redlich–Peterson isotherm. For Cd²⁺, the β value suggests that Cd²⁺ ions had not reached maximum coverage onto both homogenous and heterogeneous surfaces of DM-BC. However, for Zn²⁺, the results indicate that Zn²⁺ ions had reached maximum coverage on the surface of the DM-BC.

   At a lower concentration range of 6 to 96 mg L⁻¹, the Redlich–Peterson model again best represents the sorption isotherm of Cd²⁺ ions on DM-BC, whereas the sorption of Zn²⁺ ions best fit the Langmuir isotherm. Thus, the multiple active sites on the heterogeneous surface of DM-BC demonstrate different affinities for the sorption of Cd²⁺ and Zn²⁺ ions at lower concentrations.

   For the Pb²⁺ ions, 100% removal was observed for the tested concentrations; thus, the isotherms were not modeled for the Pb²⁺ data.

4. The removal kinetics and model fitting suggest that the three steps of intraparticle diffusion might be more representative for describing the immobilization processes of metal ions on the external surface and internal pores, although a PSO model best fits the experimental data.

5. DM-BC retained the mixed metal ions of Cd²⁺, Pb²⁺, and Zn²⁺. DM-BC showed complete removal of Cd²⁺, Pb²⁺, and Zn²⁺ via the precipitation of metal hydroxides and/or surface complexation with mineral components such as $\text{C}{\text{O}}_3^{2-}$ and $\text{Si}{\text{O}}_3^{2-}$ across the three regeneration cycles due to high solution pH from the alkalinity released by the biochar.

6. As a result, this study suggests the feasibility of using waste-derived biochar as a promising sorbent for the remediation of mixed metal ions contaminated water, considering the low-cost waste as feedstock, high efficiency, and regeneration-reuse potentials. Future research should focus on the scale up and cost analysis and optimization as well as real contaminated water for practical application in real contaminated fields.

Data Availability Statement

All data, models, and code generated or used during the study appear in the published article.