Physical, Static, and Kinetic Analysis of the Electrochemical Deposition Process for the Recovery of Heavy Metal from Industrial Wastewater

Through the electrodeposition technique, toxic metals in wastewater can be removed and deposited on a chosen substrate with excellent selectivity. In this work, we use this technique to extract lead cations from simulated wastewater by using fluorine-doped tin oxide (FTO) substrate at various temperatures. In situ tracking of lead nucleation at advanced stages has been achieved by chronoamperometry. According to the experimental results, the theoretical models developed to study the kinetic growth of lead deposits in 2D and 3D are in good agreement. Nucleation rate and growth rate constants, for example, were found to be strongly influenced by temperature. Cottrell's equation is used to calculate the diffusion coefficient. X-ray diffraction, scanning electron microscopy, and energy-dispersiveX-ray techniques were used to investigate and characterize the lead deposits. The reported results could provide insight into the optimization of electrodeposition processes for heavy metal recovery from wastewater and electronic wastes.


Introduction
Treatment and recycling of wastewater and electronic waste (e-waste) have become increasingly necessary due to the presence of precious, critical, and strategic metals as well as the environmental impact of metal recovery [1][2][3]. e-waste is highly heterogeneous and includes a wide range of materials, such as metals, plastics, and ceramics, which are nonbiodegradable pollutants. Mercury, lead, arsenic, cadmium, chromium, and cadmium, for example, are classifed as the most common poisonous heavy metals, even at low levels of concentration [4,5]. Both the U.S. Environmental Protection Agency and the International Agency for Cancer Research classify them as human carcinogens [6]. For instance, the levels of contamination in aquatic and terrestrial animals attributed to anthropogenic activities in Indonesia were found to have exceeded the tolerable limits of international standards, according to some reports [7][8][9]. Lead species are considered the most hazardous pollutants. It is extensively used in many diferent industries to fabricate, for instance, batteries, ammunition, metal products, and many other electronic devices. For example, in the United States in 2021, about 990,000 tons of secondary lead were produced from recycled materials, an amount equivalent to 62% of apparent domestic consumption. Nearly all the secondary lead was recovered from old scrap, mostly lead-acid batteries [10]. For these reasons, so far, numerous methods for recovering metal from e-waste have been suggested. Among them, one can cite cementation, ion exchange, liquid-liquid extraction (solvent extraction), and adsorption [11]. Te major disadvantages of these methods are their slow kinetics, low adsorption capacity, and high cost. Te precipitation of insoluble metal hydroxides is another used process [12]. New biochemical and biotechnological technologies are emerging. Even so, using these technologies poses a range of challenges [13]. Indeed, they involve chemical reactions that require the use of large quantities of expensive organic solvents and other polluting chemicals. Consequently, it is a priority to recover metals from e-waste via metallurgical extraction due to the fast depletion of natural mineral ores and the limited geographical availability of critical and/or strategic metals ( Figure 1).
Terewith electrodeposition is an eco-friendly technology with convenient and precise controls as well as low energy consumption. It has been proven to be very efective in the treatment of e-waste [14]. It allows nanostructures to be deposited with a high degree of control [15,16]. Te electrodeposition process is either totally ohmic-regulated or mixed ohmic-difusion controlled depending on the concentration of Pb (II) ions [17,18]. Chen et al. [19] used a potentiostatic technique to depose Zn and Pb at varied temperatures and potentials into Cu foil. As known, the electrode structure could alter the electroplated deposits by modifying their nucleation energy, size, and shape. Indeed, FTO substrate is a widely used material for metal recovery due to its inert surface. It can be used to study the efects of metal-metal interaction on the growth and nucleation process [20,21]. Lead electrodeposition onto a fuorinedoped tin oxide (FTO) substrate from a nitrate solution was investigated by Rebey et al. [22]. To the best of our knowledge, the coupling between the dynamics of lead recovery and the characterization of its deposits in FTO as a function of temperature has not yet been fully investigated.
Tis work intends to control the metal deposits on an FTO substrate and comprehend the processes that occur at the metal/substrate interfaces at diferent temperatures using an electrochemical cell connected to a chiller. Depending on the inquiry, either in situ chronoamperometry or cyclic voltammetry (CV) mode is used. Following that, we focus on how the 2D and 3D kinetic growth models relate to the experimental results and the theoretical development of lead deposits. X-ray difraction (XRD), scanning electron microscopy (SEM), and energy-dispersiveX-ray (EDX) techniques were used to characterize the Pb deposits.

Materials.
In deionized distilled water, the electrolyte solution, which is sodium nitrate (NaNO 3 ) (Termo Fisher Scientifc, USA, 99.99%), was dissolved until a homogeneous, colorless, transparent liquid was obtained. High purity lead (II) nitrate (Pb(NO 3 ) 2 ) (Termo Fisher Scientifc, USA, 99.99%) was used as a Pb 2+ ionic precursor. Te obtained solution was continuously stirred until the complete dissolution of the chemicals. Te electrical recovery process was carried out using a glass substrate made from FTO conducting glass of resistance 20 ± 0.3 Ω/square, and the glass is 2.2 mm thick with dimensions of 25 mm × 25 mm (Ossila Ltd, UK). Te substrates were de-ionized after being washed with isopropyl alcohol to get them ready for experimentation. In order to complete the cleaning procedure, the substrates were heated in an electronic furnace at 80°C for 10 minutes. We run the electrochemical deposition experiments in a three-electrode cell in which three diferent electrodes (the working, counter, and reference) are placed in the same electrolyte solution. A platinum wire is used as the counter electrode, and the substrate is used as the working electrode. An Ag/AgCl electrode is used as the reference electrode.

Voltammetry and Chronoamperometry.
Te cyclic voltammetric I-V and chronoamperometric measurements were carried out at varying temperatures (5°C, 20°C, and 35°C) using a HEKA potentiostat/galvanostat PG510 controlled by POSTMASTER software, at a scan rate of 60 mV/s. Te electrochemical experiments were performed in an aqueous solution of 0.1 M Pb(NO 3 ) 2 in 0.4 M NaNO 3 . Cyclic voltammograms were measured in the voltage interval [-1.5 V, +1.5 V], while the chronoamperograms were recorded at the potential of -0.8 V.

Characterisation.
Te structure and phase identifcation of the recovered electrodeposits were investigated using the following: (i) An X-ray difraction analysis device (PanAlytical MPDPRO difractometer) equipped with CuKα radiation of 1.540Å over the range 20°-80°equipped with a linear X'Celerator detector using copper anticathode (λKα1/α2 � 1.540560/1.544330Å). X-ray difraction patterns are obtained in the 2θ range of 20-80°; the step size is 0.02; and the time per step is 30 s. (ii) A scanning electron microscope (SEM), an FEI Quanta 600 microscope and (iii) energy-dispersiveX-ray (EDX) coupled with the SEM were used.

Voltammetric and Chronoamperometric Behavior.
Te static behavior of lead electrosorption onto an FTO substrate is shown by the cyclic voltammograms recorded for the diferent temperatures of 5, 20, and 35°C (respectively, shown by the green, blue, and red lines in Figure 2). As shown, the reference I-V cycle without Pb 2+ (black line) has a symmetrical shape and does not show any current peak. However, in the presence of lead, the current intensity measured through the voltage window seems somewhat sensitive to Pb 2+ cations. Tere are clearly identifable current density peaks at the potential of -0.8 V for all the studied samples at diferent temperatures. As the temperature increases, the peak intensity increases as well.
Tese peaks indicate that electrochemical processes are successfully occurring on the electrode surface through the reduction of Pb 2+ cations to Pb, with a strong dependence on the temperature parameter. As demonstrated later, the higher the temperature of the cell, the greater the density of the deposit of Pb [23,24].
Te in-situ evolution of the current transient is shown in Figure 3. Tey demonstrate that the nucleation rate changes with temperature. As shown, three distinct parts describe the whole process. Each current density reaches its threshold noted as j m (maximum current density) at a time t m . Following that, they remain constant or decrease slightly. Te frst stage is characterized by a relatively stable current intensity, noted as the silent part. Te duration of this stability is strongly afected by the temperature. Indeed, it decreases as the temperature increases. As the current density rises signifcantly, it enters the second stage of the nucleation process. As the temperature increases, the slope of the current density toward time increases clearly. During the third stage, the current density is practically stable. During this plateau-like shape phase, the Pb deposits across the FTO surface reach their equilibrium. As reported by González-García and his coworkers [25][26][27], electrochemical deposition is a very complex process and is strongly infuenced by experimental conditions. Hence, these experimental results should be interpreted in light of theoretical, structural, and microstructural considerations.  In terms of pretreatment, there are two types: manual processing, which involves sorting, separating, cleaning, emptying, dismantling, decontaminating, and segregating, and mechanical processing, which involves shredding, milling, grinding, and separating through eddy current or air stream classifers.

Teoretical Approach. Te elaboration of the crystalline materials involves phenomena of nucleation and growth.
Electro-crystallization is the study of these two phenomena under the infuence of an electric feld. According to Amblard [28], these two phenomena compete with each other. Tey infuence the kinetics, structure, and properties of the deposits. For example, the faster the nucleation speed, the fner the grains that form the deposit. Teoretical electrocrystallization models have been proposed to demonstrate the nucleation and growth modes during the electrodeposition process. Te nucleation process is generally described in two types: instantaneous and progressive processes. Te growth process is typically divided into three categories [29]: (i) Te two-dimensional (2D) growth mode, or Frank-Van der Merwe mechanism, is generally found in the case where the metal and the substrate are of the same chemical nature. (ii) Te 3D growth mode, or Volmer-Weber mechanism, can be exploited to produce nanostructures. (iii) Stranski-Krastanov's mechanism begins with a 2D growth mode followed by 3D growth. According to Stackelberg [30]: (i) In the 2D growth process, the current density can be expressed by the following equations for instantaneous and progressive nucleation: where j 2D is the current density, k 2D is the lateral growth rate constants (mol cm −2 s -l ), z � +2 is the Pb valency, F is the Faraday constant (F � 96485 C.mol − 1 ), t is the time, h is the layer height (cm), N 0 is the total number of active centers (cm −2 ), A 2D is the nucleation rate (nuclei cm −2 s −1 ), M is the atomic weight (g.mol −1 ), and ρ is the density (g.cm −3 ) of the deposit. (ii) In the 3D growth process, the current density can be expressed by the following equations for instantaneous and progressive nucleation: where k and k ′ are, respectively, the lateral and vertical growth rate constants and A 3D is the nucleation rate. (ii) In the Stranski-Krastanov mechanism, the current density can be expressed through a combination of equations (1) or (2) and (3) or (4).
Furthermore, in Harrison and Tirsk's [31] studies, the current density is divided into two parts: the frst one is due to 3D crystal growth, which can be given in equations (3) or (4). Te second part is the current caused by an outward growth on a substrate base plane at a free surface uncovered by growing nuclei, j f . j f is expressed by equations (5) or (6) for instantaneous and progressive nucleation, respectively, as follows: where k 0 is the growth rate constant on the base plane of the substrate. As shown in Figure 3, temperature infuences the shape of current-time transients. As a result, temperature strongly afects the nucleation and growth processes. Te behaviour of current-time characteristics is theoretically analysed at three temperatures (5°C, 20°C, and 35°C). Experimental results have been simulated and examined with 2D and 3D models according to equations (1) to (6). For the current density measured at 5°C (black curve), the ftting is described by the following equation, which is the combination of three parts: where i 0 � zFk 0 , where t ind is the induction time.
Te closest ft confrms that nucleation and crystal growth at 5°C begin with progressive nucleation and 2D crystal growth. At an induction time, t 0 , the second process of progressive nucleation and 3D crystal growth starts, as described by Stranski-Krastanov. Te chronoamperometric curve of the electrodeposition of lead, obtained at 20°C (blue curve in Figure 3), shows that two quasi-plateaus appear at shorter and longer times, which follows the model of 3D growth [32]. To obtain further information regarding this process, diferent equations (from equations (1) to (6)) were tested. Te most accurate ft of the experimental data is obtained for equation (9), derived from equations (3)-(5). Generally, this equation (9) indicates that after an induction time, there is instantaneous nucleation followed by progressive 3D nucleation: 4 Scientifca where where k s and k s ′ are, respectively, the lateral and vertical growth rate constants for the secondary growth process. At 35°C, the current-time transient curve has a steeper slope (the green curve in Figure 3). Tis mechanism can be described by progressive nucleation and 3D growth or the Volmer-Weber model [29]. Te best ft can be achieved with the following equation, which is a combination of equations (4) and (6): (11) Figure 4 shows the values of log (k 0 ), log k { ) ′ ), and A 3D as functions of temperature. Tey are derived from the bestftting model. From 5°C to 35°C, the rate constant, k 0 , exhibits linear dependence on temperature (black curve). Te vertical growth rate constant, k ′ , is found to be nonlinear for the frst 3D growth process (red curve). Indeed, k ′ increases as the temperature increases from 5°C to 20°C and then becomes practically stable from 20°C to 35°C. We also investigate the nucleation rate, A 3D , assuming that the growth rates are the same in both directions. Our study shows that it decreases with increasing temperature, based on the blue curve.

Characterizations.
To correlate the chronoamperometry results, the morphologies of Pb deposit particles obtained at diferent temperatures are characterized by the SEM technique. Figure 5 shows that the density of the individual crystals changes considerably with temperature, which enables the correlation of the lead (Pb) deposit morphology to their chronoamperograms. At 5°C, it is seen that randomly distributed individual crystals with sizes in the order of 5 μm were formed on the substrate (Figure 5(a)). Te crystals are characterized by fat faces and sharp corners and edges. At 35°C (Figure 5(c)). An agglomeration of crystals is observed. Te distance between the deposits is lowered and one observes that sand-rose-like morphology develops at T = 20°C ( Figure 5(b)). Tis can be explained by 3D crystal growth as described by the Stranski-Krastanov mechanism mentioned above. Te obtained surface morphologies are in line with data obtained by chronoamperometry analysis [33].
Terefore, the nucleation mode of lead metal is strongly infuenced by the concentration of cationic precursor, allowing the correlation of the morphology of the deposit with its CV. Previous studies attribute the shape of recovered metal deposits to the applied CV mode and deposition rate [32,34]. Other physical factors such as the concentration of the ionic solution, the bathing temperature, and time are also taken into account to determine the shape and size of the recovered deposits. Tese include fernlike dendrites [35], needle-like [36], dendritic [32], and honeycomb-like structures [34].
Quantitative analysis of the obtained flm is carried out using EDX. Results presented in Figure 6 show the presence of lead (Pb), oxygen (O), silicium (Si), and tin (Sn). Teir percentage is given in the inset table. Te results obtained from Figure 6 indicate Pb (23.52%), O (20.58%), Si (2.35%), and Sn (53.54%). Sn is the major constituent of the sample since it is the principal element of the substrate, Florine (F) doped SnO 2 . F is missed maybe due to its low doped percentage. Si is present in a low percentage because it is completely covered by the FTO substrate. Te presence of Pb is the result of the electrodeposition process. Notably, we only provide the result for 20°C in Figure 6. For the sake of the fgure's clarity, the other temperatures (5°C and 35°C) are not included because they behave in the same way.
Te X-ray patterns of the obtained samples at 5°C (T5), 20°C (T20), and 35°C (T35) are compared to Pb(NO 3 ) 2 as starting materials and FTO as a substrate on which Pb is deposed (Figure 7). Tis fgure shows the disappearance of lead nitrate in the three samples. Teir patterns show peaks corresponding to FTO (SnO 2 : reference code 01-077-0452) in addition to extra lines observed at 22.5°, 29.5°, and 39.7°. No overlapping of the measured extra lines with those reported in the literature was found. Since the Pb (II) cations were electrochemically reduced at a potential of -0.8 V, the obtained deposit should be lead (lead oxide is excluded). A deep Xray difraction study is in progress to determine the structure of this new phase.
Te crystallite size "D" is determined from XRD diffracted patterns through the Scherrer equations [36,37]. Te following equation is obtained: where K � 0.89 is a constant, λ � 1.5406Å, and θ and β are the difraction angle and the corresponding full width at halfmaximum of the observed peak, respectively. Lead is characterized mainly by two peaks observed at 2θ � 26.2°and 2θ � 39.7°. Te size of particles is calculated from the frst peak since the second corresponds to an overlapping of the (300) and (221) lines. For the three samples obtained at T � 5, 20, and 35°C, the calculated size is 45, 44, and 43 nm, respectively (Table 1). Te crystallinity index (CI), which considers the contribution of the amorphous and crystalline phases, can be investigated by diferent methods, including XRD and NMR [36,38,39]. In this work, the crystallinity of deposited lead is calculated using X-ray difraction data. Te percentage of the CI of the sample toward temperature was determined for the entire difracted pattern using the following formula:   Figure 7: Te XRD pattern of the lead deposited on an FTO substrate (with a potential pulse at −0.8 V at diferent temperatures). Te patterns of Pb(NO 3 ) 2 and FTO are given for comparison. 6 Scientifca Te obtained values are 67.3, 69.6, and 71.4% for T � 5°C, 20°C, and 35°C, respectively (Table 1). Tese values indicate an increase in crystallinity with temperature. Moreover, the increase of the difracted peak related to lead (for example, the peak at 39.82°, Figure 8), as the temperature increases, is consistent with the SEM results ( Figure 5), due to the agglomeration of the individual crystals.

Conclusion
Pb (II) cations were electrochemically reduced at a potential of −0.8 on the FTO substrate at diferent temperatures. Te kinetic parameters of electrodeposition processes were determined from the theoretical analysis of chronoamperometry data. A reasonable agreement between the values of difusion coefcients determined by applying the Cottrell equation and the nonlinear ftting method was achieved. Te density number of active sites and the nucleation rate constant have been discussed. According to our results, the morphology of deposit particles and chronoamperometry curves are well correlated. After comparison with what was reported in the literature, XRD and EDX analyses suggest the formation of a novel lead phase. Tis study highlights the importance of controlling the recovery process of toxic e-waste from industrial water.

Data Availability
Data are available on request from the corresponding author.

Conflicts of Interest
Te authors declare that they have no conficts of interest.