Rational Design of a 3D Crown‐Based Material for the Selective Recovery of Silver Ions from Seawater and e‐Waste

Silver recovery from sustainable sources such as seawater and e‐waste is critical for the conservation of land‐based resources and the reduction of the environmental impacts of e‐waste disposal. However, the abundance of competing metal ions in seawater and in e‐waste makes the recovery extremely challenging. Thus, to effectively capture silver ions under these conditions, the designing of materials with high selectivity, sufficient binding sites, and low affinity to competing metal ions is vital. Herein, we report the design and synthesis of a 3D‐like Schiff‐bridging crown‐based material named AC5, for the selective recovery of Ag+ ions. Through this design, a significant enhancement in the silver ion recovery was achieved with an excellent removal efficiency of up to 99.9%, and a tremendous increase in selectivity of 400 000 – 900 000% when compared to the metal ions (Li+, Na+, K+, Mg2+, and Ca2+) found in seawater, and the heavy metal ions (Cu2+, Cd2+, Ni2+, and Pb2+) found in electronic wastes. Density functional theory and molceular dynamics simulations revealed that the structural geometry of AC5 favors high charge transfer, lowered global hardness, and enhanced ion‐dipole attractions toward Ag+ ions, making the material an excellent candidate for the efficient recovery of silver from desalination brine and spent silver resources.


Introduction
The demand for silver has been on the rise over the years due to its use in jewelry, coins, medical equipment, and the production of electronic components. [1]Its high conductivity, malleability, ductility, and resistance to corrosion make it a valuable metal in various manufacturing industries. [2]In fact, silver is an indispensable metal whose mining dates to 3000 BC and continues to DOI: 10.1002/admi.202300336evolve.According to the United States Geological Survey (USGS), the total world silver reserves are estimated to be 570 000 tons, the majority of which comes from mines, while the rest is recovered from recycled materials. [2]hile the earth's silver reserves are continuously being depleted, there is a need to explore alternative sources of the precious metal such as the ocean and from recycled electronic wastes.
The ocean and e-waste are two important sources of silver ions which, when properly harnessed, could complement the existing mining technologies and help conserve land-based resources. [3]hile the concentration of silver in seawater is extremely low (< 1 ppm), the vast volume of seawater makes it an inexhaustible source of precious metal.Consequently, the total silver reserve in the ocean is estimated at ≈17 million metric tons (over 7 times greater than those found on land). [4]E-waste, on the other hand, is a common source of precious metals such as gold, silver, platinum, and palladium whose improper disposal could constitute various environmental hazards. [5]The recovery of these metals from e-waste is thus critical as it not only helps to conserve scarce resources but also reduces the environmental impacts of e-waste disposal.Despite the abundance of silver in the ocean and in recycled e-waste, the presence of competing metal ions makes the recovery extremely challenging.6a,b] Among the materials exploited for this purpose are crown ethers (CEs) whose unique host-guest complex formation tendency has been utilized to recover several metal ions. [7]7b] Several studies on the use of CEs as recovery agents of silver ions have been reported.Paul et al. [8] reported the catch-release fluorescent dosing and recycling of silver ions on an anthraceneappended CE and demonstrated the practical utilization of such systems in molecular communication.Falaki and co-workers [9] reported the selective recovery of silver and lead ions from singlestream solutions on supported liquid membranes loaded with 18-crown-6 materials.The extraction of silver, cesium, strontium, and other metal ions in fluorinated diluents containing CEs was reported by Smirnov et al. [10] Fissaha and co-workers [11] demonstrated the recovery of silver ions on hydroxylated thio-CEs with various cavity sizes and achieved excellent uptake of 124 -179 mg g −1 within 24 h.While the thia-CEs have achieved remarkable results, their complex synthesis protocol and the time required to achieve maximum uptake make them impractical for real analysis where abundant competing ions co-exist.Others are; the use of crown-functionalized mesoporous silica, [12] crown-based supramolecular nanostructures, [13] aza-crown grafted heterocycles, [14] and the use of polymeric sensors. [15]Despite their numerous potentials, CEs suffer from tremendous interference from co-existing metal ions in e-waste, lowering their selectivity.Moreover, according to the hard-soft acid-base (HSAB) principle proposed by Pearson, [16] the hard Lewis base CEs would bind strongly to hard alkali metal ions present in seawater, consequently blocking the adsorption sites.
To overcome these, it is vital to tune the structural features of the CEs to lower the hardness and provide sufficient binding centers for silver ions, with minimal interference from competing ions.
In this study, we report the recovery of silver ions from seawater and e-waste using a 3D-like Schiff base-bridging CE (AC5).AC5 was synthesized by the acid-catalyzed nucleophilic addition of tris(4-formylphenyl)amine to 4′-aminobenzo-15-crown-5 resulting in the formation of a hierarchical structure with sufficient binding centers for Ag + ions.First-principle DFT simulations supported with classical MD simulations revealed that AC5 exhibited high charge transfer efficacy, lower hardness, greater dipole interactions, and stronger binding affinity to Ag + ions.Solid-liquid and liquid-liquid extraction studies further revealed that AC5 achieved 99.99% recovery of Ag + ions and showed a tremendous selectivity of 400 000 -900 000% compared with metal ions such as Li + , Na + , Mg 2+ , Ca 2+ , and K + present in seawater.Meanwhile, in the presence of metal ions present in ewaste such as Ni 2+ , Cd 2+ , Pb 2+, and Cu 2+ , AC5 exhibited excellent recovery of Ag + ions with efficiency in the range of 94 -99.9%.This study thus presents a strategy to enhance silver recovery from seawater and e-waste and could pave the way for the design of the next generation of materials for effective brine mining.+ , and d) the corresponding frontier orbital distribution of the hydrated ion, calculated at the GGA-PBE level of theory.

Computational Studies
The optimized structural geometries of benzo-15-crown-5 and AC5 are presented in Figure 1a,b.The corresponding structural features are depicted in Table S1 (Supporting Information).The geometrically optimized structures did not show any implicative variations across all bonds within the crown cavity, implying that the 3D array of the crown ethers in AC5 did not result in structural distortions of the crown moiety.Both molecules exhibit the C1 low symmetry point group, and the bond properties are almost identical.The crown structures remain non-distorted and the intra-atomic bond distances within the crown rings remain almost the same.The energy-minimized aqua-coordinated silver ions, Ag(H 2 O) 4 + , and the corresponding frontier orbital distributions are presented in Figure 1c,d, while those of lithium, sodium, potassium, magnesium, and calcium ions are depicted in Figure S1 (Supporting Information).The hydrated ions displayed the expected coordination numbers and the M n+ -O bond distances agreed with experimental findings using the extended X-ray absorption fine structure (EXAFS) and the large-angle Xray scattering (LAXS) techniques. [17]The HOMO-LUMO energy gap (ΔE) further revealed that the Ag + ions had the lowest value of 3.083 eV in contrast to the Li + , Na + , K+, Mg 2+ and Ca 2+ ions with the values of 6.732, 6.166, 6.174, 6.136 and 6.234 eV, respectively exhibits high possibility of electronic transitions, high polarizability and stronger back-acceptance of non-bonding electrons when interacting with the crown molecules.
Meanwhile, the frontier orbital distribution and the fundamental electronic properties of the crown molecules are presented in Figure 2. Evidently, the structural orientation of AC5 favors charge transfer processes as the ΔE value is significantly lowered (1.678 eV) when compared to the benzo-15-crown-5 (4.090 eV).The electronic properties revealed that AC5 having a lower global hardness () of 0.839 eV and stronger dipole moment (μ) of 6.22 Debye exhibits the matching structural features for the recovery of Ag + ions which equally has the  of 1.541 eV in accordance with the HSAB principle. [16]The benzo-15-crown-5 in contrast recorded  and μ values of 2.045 eV and 4.047 Debye, respectively lowering its attraction for the Ag + ions, thus making it non-selective in the presence of competing alkali and alkaline earth metal ions.Consequently, AC5 has shown promising aptness for the selective recovery of silver ions from seawater with minimal interference from competing ions.
The preferential centers of interaction of the metal ions on AC5 were characterized and visualized using the molecular electrostatic potential (MEP) maps as presented in Figure S2 (Supporting Information).MEP is a graphical illustration of the electronic density of the surface of a material. [18]It predicts the local reactivity on the material surface.Regions of low electron density are characterized by blue color and represent the active centers prone to nucleophilic attack, while the regions of high electron density are depicted in red which represents the centers prone to electrophilic interactions.An inspection of the MEP of the molecules revealed that whilst the benzo-15-crown-5 exhibited greater dipole interactions within the crown cavity, three possible interaction centers were identified on AC5; the macrocyclic crown cavity, the sp 2 -hybridized Schiff nitrogen atoms and tetrahedrally-coordinated sp 3 nitrogen atoms on the tris(4formylphenyl)amine moiety.These centers were identified as Φ 1 , Φ 2, and Φ 3 , and their preferential interactions with the metal ions were studied.
Using first-principles DFT simulations, we further investigate the interactions of both materials with the metal ions in seawater and in e-waste (on AC5 only).All the metal ions were placed at 2.5 Å above the crown nanopores on benzo-15-crown-5 and at the Φ 1 , Φ 2, and Φ 3 centers on AC5, and allowed to optimize.The results are presented in Figures 3 and 4, and Figures S3-S7 (Supporting Information).The respective interaction energies were estimated using Equation 2 and presented in Figure 5 and Figure S8 (Supporting Information).In the case of the benzo-15-crown-5, the metal ions can be seen encapsulated within the crown cavities due to the ion-dipole attractions by the polyether oxygen atoms.Weaker attractions were exerted on the metal ions due to the qualitative HSAB principle and in accordance with the ionic radii of the ions.Consequently, interaction energies (E int ) of −5.60, −32.9, −28.0, −14.3 and −33.7 kcal/mol were estimated for Li + , Na + , K + , Mg 2+, and Ca 2+ ions, respectively.
The interaction of the metal ions on AC5, in contrast, was highly exergonic and yielded many folds higher than those recorded on the crown.This can be attributed to the lowering in the crown hardness and the surge in the dipole moment resulting from the structural modifications.The Φ 1 active center recorded comparable E int among the metal ions due to the non-selectivity of the crown units. [19]The Φ 2 and Φ 3 centers on the other hand exhibited increasing selectivity to silver ions due to the increased polarizability imparted by the aromatic benzene rings around the nitrogen atoms.This lowered the hardness in these regions and promoted the dipole attraction toward silver ions.Consequently, E int of −870 and -990 kcal/mol were estimated for silver ions at Φ 2 and Φ 3 , respectively.These values are indeed highly negative and imply the spontaneity of the adsorption of silver ions on AC5, which demonstrates the feasibility of silver recovery from seawater and from e-waste.
The analysis of the non-covalent interactions (NCI) between the metal ions and benzo-15-crown-5 and AC5 further revealed the nature of interactions of the materials with the metal ions.NCI analysis is an index of electron density and its derivatives and enables the visualization of the nature of the intermolecular interactions between two interacting systems. [20]It utilizes the plot of the reduced density gradient (RDG) and the electron density,  where: Based on the sign of the electron density ( 2 ), the NCI can be classified as either bonded ( 2 < 0) or non-bonded ( 2 > 0).Consequently, the 3D gradient maps are represented in colors as; deeper blue which represents regions of strong attractive interactions, green which depicts van der Waals interactions, and deeper red which represents regions of strong repulsive and steric interactions.The NCI iso-surface plots of the interactions of benzo-15-crown-5 with the metal ions (Figure S9, Supporting Information) revealed that the encapsulation of the metal ions within the crown nanopores was strictly driven by van der Waals attractions with a few instances where the ions experienced repulsions due to steric effect in the vicinity of the rings.Meanwhile, on the AC5 surface, silver ions were slightly repelled at the Φ 1 and the Φ 2 centers, whereas at Φ 3 position strong van der Waals attraction was experienced enabling the strong binding of the ions (Figure 6).Other metal ions encountered weaker van der Waals interactions at all positions (Figures S10 and S11, Supporting Information) resulting in weaker binding on the AC5 surface.Consequently, AC5 has demonstrated a stronger affinity for silver ions in agreement with the estimated interaction energies.
The energetics of electron distribution on the surface of isolated AC5 molecules and during the interactions with the metal ions were further investigated.The partial density of states (PDOS) of the surface atoms of AC5 and the interactions with Ag + ions at the Φ 1 , Φ 2, and Φ 3 positions are presented in Figure 7.
The PDOS plots revealed an overlap of the Ag s-orbitals with the surface p-orbitals at 0.02 eV (close to the Fermi level), in addition to the significant lowering of the peak intensities.This suggests the hybridization of the orbitals and the subsequent occupation of the states by the s-electrons of Ag.Similarly, the lowering of the band gap near the Fermi level upon interaction with the Ag + ions corresponds to the stabilization of the complexes upon interaction.Lastly, the peaks shifted drastically to more negative energies without broadening which was attributed to interactions resulting from non-covalent hybridizations, in agreement with the NCI analysis.Similar drastic shifts in energies and lowering of peak intensities were observed in the interactions of AC5 with other metal ions, both in seawater (Figures S12 and S13, Supporting Information) and in e-waste (Figure S14, Supporting Information).However, the non-appearance of the hybrid peak at 0.02 eV in these complexes signifies their weak interactions with AC5.Consequently, PDOS analysis further strengthens the interaction energies and the NCI analysis which predicts the preference of AC5 for silver ions.

MD Simulations
The diffusion of the ions in aqueous systems comprising AC5 was studied using classical MD simulations.Amorphous cells comprising AC5 molecules, water molecules, and metal ions were constructed and subjected to dynamics simulations on the NVT ensemble.The mean square displacement (MSD) of the diffusion of the ions is presented in Figure 8.The group I and II metal ions experienced minimal electrostatic attraction by the AC5 molecules, resulting in faster diffusion in the aqueous system (Figure 8a).Similarly, the heavy metal ions (Figure 8b) due to their weak polarizabilities tend to encapsulate within the crown cavities but are not favorably captured due to their respective ionic sizes.This in turn weakens their interactions with AC5, promoting their fast diffusion.The Ag + ions on the other hand, preferentially interact at the Φ 2 and Φ 3 centers on AC5 via electrostatic attractions, slowing down their diffusion rate, compared to the co-existing metal ions.Furthermore, the hydrated Ag + ions can be seen undergoing partial dehydration to shed their water molecules and bind strongly.
We further conducted radial distribution function (RDF) analysis to illustrate the extent of the interaction of Ag + ions in bulk solution and upon dehydration and interacting with the active centers of AC5.The results are presented in Figure 9.A sharp peak was observed at 1.06 Å, indicating strong electrostatic attractions between the material and Ag + ions, in line with the hydration shell of the ions. [21]The peaks at 1.49, 1.80, 2.15, 2.50, and 3.45 Å, correspond to the close-range interactions between the active centers and the ions.Beyond 4.0 Å, the peaks eventually flatten due to the solvation of the active centers.The partially hydrated systems exhibit similar behavior with a slight decrease in intensity, affirming the ability of the coordinated water molecules.These observations further attest to the stronger iondipole attractions exerted by AC5 on Ag + ions, and the potential  of the material for the practical recovery of silver from seawater and e-waste.

Material Synthesis and Characterizations
Having established from first principles DFT and classical MD simulations that AC5 exhibits the potential for the practical recovery of silver ions, we moved on to synthesize the material following the procedure in the literature, [22] with slight modifications.The brownish non-crystalline powder having flake-like morphologies with irregular sizes was obtained (Figure 10a,b).The TEM images revealed the flake-like materials stacked above each other, and the nanopores of the crown rings were depicted as dark regions (Figure 10c,d).The elemental mapping of the surface of AC5 (Figure 10eg) revealed the distribution of carbon, nitrogen, and oxygen atoms.
The FTIR spectra of the starting monomers (4′-aminobenzo-15-crown-5 and tris(4-formylphenyl)amine) and AC5 are presented in Figure 11a.The peak corresponding to the ─C═O vibrational stretch can be seen at 1720 cm −1 on tris(4-formylphenyl)amine.This peak is, however absent on AC5, confirming the successful protonation and subsequent condensation of the carbonyl oxygen.The primary amine -NH stretch on 4′-aminobenzo-15-crown-5 appeared at 3450 cm −1 , which is absent on AC5.The appearance of a weak peak at 1690 cm −1 on the spectrum of AC5 corresponding to the imine ─C═N stretch, and the disappearance of the -NH stretch band affirms the successful addition of the two monomers via the imine linkage to form the 3D Schiff-bridging crownbased material.The appearance of the broad -OH peak at 3500 cm −1 could be attributed to the partial protonation of the oxygen atoms within the crown macrocyclic cavity.The X-ray diffraction (XRD) and the selected area electron diffraction (SAED) spectra of AC5 (Figure 11b,c) revealed its non-crystalline nature, with a pattern typical of amorphous materials.The thermogravimetric analysis (TGA) profile (Figure 11d) revealed two major weight losses at 300 °C and 650 °C, which correspond to the combustion of fractions of the molecule with the release of NOx, CO 2 , and H 2 O, respectively.The elemental analysis (Figure 11e) identified the constituent elements of AC5 to be carbon, hydrogen, nitrogen, and oxygen with percent weights of 63.05, 6.85, 6.45, and 23.65%, respectively.

Silver Recovery
The recovery of silver ions using AC5 was conducted in simulated seawater comprising 30 ppm each of Li + , Na + , K + , Mg 2+, and Ca 2+ ions.The results are presented in Figure 12a-c.The corresponding digital photos of the recovery solutions are presented in Figure S15 (Supporting Information).The structural geometry of AC5 lowers the hardness and promotes the affinity for Ag + ions.The relatively harder groups I and II ions were weakly attracted toward the active centers, resulting in lower removal efficiency.AC5 thus, achieved 99.9% removal of Ag + ions, while the groups I and II ions such as Li + , Na + , K + , Mg 2+, and Ca 2+ ions recorded efficiencies of 4.61, 8.13, 9.09, 18.3, and 18.1%, respectively, which implies the high selectivity of AC5 to Ag + ions (Figure 12a).The equilibrium uptake capacity (Q e ) measured for each ion (Figure 12b) put silver ahead with a value of 23.1 mg g −1 , in contrast to the values of 1.50, 2.81, 2.13, 6.62, and 6.11 mg g −1 achieved for Li + , Na + , K + , Mg 2+, and Ca 2+ ions, respectively.The corresponding distribution coefficients (K d ) of AC5 for the metal ions (Figure 12c) show the extreme selectivity of the material toward Ag + ions.Thus, K d values of 0.048, 0.088, 0.100, 0.225, and   0.221 L g −1 were measured for Li + , Na + , K + , Mg 2+ and Ca 2+ ions, respectively.The Ag + ions in contrast achieved a remarkable K d of 964.3 L g −1 , up to 900 000% higher in selectivity than the groups I and II ions.Similar results were achieved with liquid-liquid extractions using dichloromethane (Figure S16, Supporting Information).
In the presence of the common heavy metal ions Cu 2+ , Pb 2+ , Cd 2+, and Ni 2+ present in e-wastes, AC5 further achieved an excellent recovery of Ag + ions, maintaining the removal efficiency of 99.9% (Figure 12d), and uptake capacity up to 30.1 mg g −1 and K d of 791.2 L g −1 (Figures S17 and S18, Supporting Information).These results agree with the predicted reactivity of AC5 from DFT and MD simulations, and demonstrate the extreme selectivity of AC5 to Ag + ions, making it a potential material for the practical recovery of silver from seawater and electronic wastes.
Furthermore, the effect of pH on the recovery of Ag + ions was explored in the range 3 to 11 as presented in Figure 13.Interestingly, the recovery of Ag + ions was significantly lowered at pH 3 and 5 due to the partial protonation of the crown oxygen atoms, the formation of insoluble silver complexes, and the electrostatic repulsion of positively charged H 3 O + and Ag + ions.The maximum removal efficiency was thus reached at pH 7, which was maintained until pH 9. The sudden drop in efficiency at pH 11 could be attributed to the formation of the sparingly soluble AgOH which limits the availability of Ag + ions during the recovery.Meanwhile, a higher chloride concentration of 2000 ppm did not limit the recovery of Ag + ions on AC5 as the material maintained an excellent recovery of 99.9% in the pH range 7-9, eliminating the need for pH adjustment prior to the recovery in practical seawater samples.
We further studied the effect of high copper concentrations as found in e-wastes on the recovery of Ag + ions using AC5.The concentration of Ag + ion was maintained at 30 ppm, whereas that of Cu 2+ ion was varied in the ratio 1:1, 5:1, 10:1, and 20:1, and the results are presented in Figure S19 (Supporting Information).Obviously, the increase in concentration of Cu 2+ ions did not alter the recovery efficiency of Ag + ions significantly as the host-guest interactions are mainly controlled by the HSAB attraction principle.Thus, about 90% of Ag + ions were recovered even at a higher Cu 2+ ion concentration of 600 ppm, making AC5 a potential material for silver recovery in electronic wastes.

Desorption and Regeneration
The desorption and regeneration of adsorbents are essential to their practical applications.Thus, we conducted the desorption of the recovered silver ions in this study in a 1:1 v/v solution of 0.01% dithizone in ethanol and 0.1 m HCl.While the acid weakens the interactions of the AC5 with Ag + ions by the partial protonation of the adsorbent, the sulfur-containing dithizone chelates with the ions to form metal-dithizone complex. [23]Further acidification of these complexes releases the Ag + ions and the AC5 is regenerated with a thorough washing in a slightly alkaline solution (pH 9), and re-used without any further treatment.As shown in Figure 14a, the regenerated AC5 exhibited a negligible loss in capacity and maintained the recovery > 99% even after 5 cycles of removal-regeneration.The desorption efficiency was also maintained > 95%, making the regeneration strategy effective for multiple cycles.Meanwhile, the uptake capacity and the desorption ratio (Figure 14b) were excellently maintained with no significant decline in recyclability after 5 consecutive cycles.Lastly, the structural features of the regenerated AC5 molecules were checked by conducting FTIR and SEM analysis (Figure 14c,d).The results show that the structural features of AC5 after several adsorption-desorption cycles were retained as all the peaks on the original material were present, and the structural morphology was retained, suggesting recyclability.

Silver Recovery from Real Seawater
Most materials are effective in simulated seawater but may fail when evaluated in a practical seawater medium.Thus, the efficacy of these materials which are often reported in lab-based simulated seawater may not translate to real seawater samples.Therefore, the potential of the practical utilization of AC5 for silver recovery from seawater and other aqueous sources was investigated using real seawater samples collected from the seafront  in Khobar, Eastern province of Saudi Arabia.The samples were filtered using fine acid-treated filter paper, and the silver content was analyzed using ICP-OES.The measured Ag + ions concentration was 5.67 ppb.The sample was further spiked with 1000 ppb of standard Ag + ions and both the un-spiked and the spiked samples were added to a pre-weighed amount of AC5, and subjected to the recovery process.Interestingly AC5 achieved remarkable recovery of Ag + ions in both samples (Table 1) reaching the efficiency of 97.4 and 97.0%.These results validate the predicted change in the structural behavior of the crown molecules and demonstrate the extreme selectivity of AC5 to Ag + ions in a complex matrix such as seawater.Meanwhile, the performance comparison among related materials from the literature (Table 2) showed that AC5 outperforms other crown and non-crown-based materials in the selectivity toward silver ions.

Conclusion
In conclusion, we have successfully explored the potential of the practical recovery of silver from seawater and e-waste on a novel 3D-like Schiff-bridging crown-based material, named AC5.The designed crown has shown exceptional performance in the recovery of silver ions from seawater, due to the synergistic action of the crown moiety, the Schiff bridge, and the linker active sites.First principles DFT and classical MD simulations revealed that AC5 exhibited high charge transfer efficacy, sufficient binding sites, lowered global hardness, and significantly enhanced iondipole attractions toward Ag + ions.We also demonstrated that in the presence of groups I and II metal ions (Li + , Na + , Mg 2+ , Ca 2+ , and K + ) found in seawater, AC5 achieved excellent recovery of Ag + ions with a removal efficiency of up to 99.9% and a tremendous unprecedented selectivity of 400 000 -900 000%.In the presence of heavy metal ions (Cu 2+ , Cd 2+ , Ni 2+, and Pb 2+ ) found in electronic wastes, however, the novel material achieved remarkable selectivity with removal efficiency in the range of 94-99.9%.This study reports a strategy for the rational design of effective materials for silver recovery from seawater and spent silver sources, a strategy that could be further explored for a sustainable brine economy.

Experimental Section
DFT Methodology: To provide insights into the host-guest interactions of the crown-based material and the metal ions, first-principles DFT simulations were conducted on Materials Studio, using the Perdew-Burke-Ernzerhof generalized gradient approximation (PBE-GGA) method.This method was selected due to its superior description of electronic sub-systems, and even so, it provided a decent level of accuracy when studying molecular-level interactions. [33]All the structures were geometrically relaxed on the DMol 3 module while enforcing the spin-unrestricted command. [34]A self-consistent field (SCF) threshold of 10 −5 Ha was imposed to provide an exact self-consistent charge density and to offer variational freedom during the search for the wavefunctions of the vacant states.The maximum force tolerance was set to 2.0 × 10 −3 Ha Å −1 , whereas the energy tolerance was maintained at 1.0 × 10 −5 Ha.
To simulate the aqueous media, the conductor-like screening model (COSMO) was used and the solvent was chosen as water.The quantum chemical reactivity descriptors such as the electronegativity (), global hardness (), and electron affinity (E A ) were estimated following the DFT-Koopman theorem of the energies of the highest-occupied molecular orbital (E HOMO ) and the lowest-unoccupied molecular orbital (E LUMO ). [35]he host-guest interaction energies were estimated using the equation: Where, E H-G , E H, and E G represent the free energies of the host-guest complexes, the isolated host molecule, and the isolated guest ions, respectively.
Molecular Dynamics Simulations: MD simulations were conducted on the Forcite module of Materials Studio, using the universal forcefield (UFF), the choice of which was attributed to its wider coverage of the periodic table and its decent predictions of the geometries and conformational energies of organic molecules, main group elements, and metal complexes. [36]Aqueous simulation boxes with dimensions 30 × 30 × 30 Å comprising of the host molecules previously optimized on the DMol 3 module, the metal ions, counterions, and water molecules were constructed and geometrically optimized using the congruent gradient algorithm, [37] followed by equilibration on the NPT ensemble for 1000 ps, at pressure and temperature of 1 bar and 298.15 K, respectively.The initial velocities of the molecules were set as random while the timestep during the dynamics simulations was 1 × 10 −3 ps.The temperature and pressure were controlled by the Nose-Hoover thermostat and the Berendsen barostat, during which the trajectory frames were output after every 5000 steps.The Ewald summation method was used to treat the long-range Coulombic interactions, whereas the attractive and repulsive interactions were estimated using the Lennard-Jones method at a cutoff range of 18.5 Å.The equilibrated systems were then subjected to NVT dynamics simulations for 1500 ps while maintaining the same simulation conditions.
The diffusion of the ions through the aqueous systems was estimated by calculating the mean square displacement (MSD) using the equation: where N is the number of ions in the system to be averaged, r i (0) and r i (t) are the position vectors of the i ion at the start and at time t, respectively.The description of the guest ions in the domain of the host molecule was estimated by conducting a pair correlation function (PCF) analysis using the equation: where a and b represent the host molecule and the guest ions, r is the distance between them, V is the volume of the entire system, and N a and N b are the number of particles of a and b, respectively.Others are N ab the number of similar particles of a and b, and r ai and r bj the 3D coordinates of a in i and b in j, respectively.Chemicals and Reagents: All the chemicals and reagents were of high purity and used as-received without further purifications.These include; tris(4-formylphenyl)amine, 97%; 4′-aminobenzo-15-crown-5, 97%; N,Ndimethyl formamide (DMF), anhydrous 99.8%; hydrochloric acid, ACS reagent 37% and dichloromethane (DCM), ACS reagents 99.9% and absolute ethanol all purchased from Sigma Aldrich.Others are; dithizone, from CDH chemicals, and standard solutions of Ag, Li, Na, K, Mg, Ca, Ni, Cu, Cd, and Pb from Fluka chemicals.
The morphology of the material was analyzed on JEOL JSM-6701F field emission scanning electron microscope (FESEM) fitted with an energydispersive X-ray (EDX) spectrometer, and on JEOL JEM-2100F field emission transmission electron microscope (FETEM), whereas the diffraction patterns were obtained on a Rigaku Miniflex-II X-ray diffractometer using CuK radiation.The sample was scanned at the rate of 0.03 °C min −1 in the 2 range of 5 -80°.FTIR measurements were conducted on Nicolet iS5 FTIR spectrometer in the range of wavenumber 400 -4000 cm −1 .TGA measurements were conducted on SDT Q600 TGA and DSC analyzer, while the elemental analysis was conducted on Perkin Elmer EA-2400 CHNS/O elemental analyzer.
Recovery Methods: The silver recovery measurements were conducted by placing 20 mL of freshly prepared 30 ppm solution of the metal ions in a glass vial and 5 mg of the material was added.The mixture was sonicated for 10 min and then stirred for 2 h.Thereafter the mixture was left undisturbed for 1 h at the end of which the supernatant solution was collected in a syringe, filtered, and analyzed by ICP-OES.The recovery efficiency (), the equilibrium uptake capacity (Q e ), and the distribution coefficient (K d ) were estimated using the equation: where, C o , C e , m, and V are the starting metal ion concentration (ppm), the equilibrium concentration of the metal ions (ppm), the mass of AC5 added (g), and the volume of the solution (mL or respectively.Similarly, liquid-liquid extraction was carried out by mixing 10 mL of AC5 dispersed in DCM with 10 mL of freshly prepared 30 ppm solution of the metal ions in a glass vial and stirred at 300 rpm for 2 h.Thereafter, the mixture was left to settle for 3 h and the two layers were collected in a separatory funnel.The aqueous solution was filtered and analyzed for the metal ions using ICP-OES and the (%), Q e (mg g −1 ), and K d (L g −1 ) were estimated.
The exhausted materials were regenerated by treatment with a combination of 0.01% dithizone in ethanol and 0.1 m HCl in the ratio (1:1) for 3 h to desorb the adsorbed metal ions and restore the removal capacity.The regenerated materials were dried in the oven at 60 °C overnight and added to another fresh solution of the metal ion to conduct subsequent removal tests.The desorbed metal ions were collected and measured by ICP-OES and the desorption ratio was calculated using the equation: where, Q s and Q e are the amount of metal ions in the desorbed solution and the amount adsorbed, respectively.

Figure 1 .
Figure 1.The optimized structural geometries of a) benzo-15-crown-5 and b) AC5, c) the optimized geometry of hydrated Ag(H 2 O) 4 + , and d) the corresponding frontier orbital distribution of the hydrated ion, calculated at the GGA-PBE level of theory.

Figure 2 .
Figure 2. Frontier orbital distribution of a) Benzo-15-crown-5 and b) AC5.The fundamental electronic properties of both molecules calculated at the GGA-PBE level of theory are presented in (c).

Figure 4 .
Figure 4.The structural geometries of the interaction of AC5 with Ag + ions at the a) Φ 1 , b) Φ 2 , and c) Φ 3 positions and the corresponding frontier orbital distributions.

Figure 5 .
Figure 5.The interaction energies of AC5 with the metal ions at the a) Φ 1 , b) Φ 2 , and c) Φ 3 positions.

Figure 7 .
Figure 7.The partial density of states of the a) surface atoms of AC5, and the interactions of AC5 with Ag + ions at the b) Φ 1 , c) Φ 2 , and d) Φ 3 positions.The Fermi level represented by the broken lines (E f ) is set to 0 eV in all cases.

Figure 8 .
Figure 8.The MSD-t plots of the diffusion of the metal ions a) in seawater and b) in e-waste in aqueous system comprising of AC5.The constructed amorphous simulation box is shown as an inset in (a), while the snapshot of the attraction of partially dehydrated Ag + ions approaching the active center is shown as an inset in (b).

Figure 9 .
Figure 9. (Left) The interactions of AC5 with Ag + ions at the a) Φ 1 , b) Φ 2 , and c) Φ 3 positions and the corresponding RDF plots of the interactions of the isolated and the hydrated ions (right).

Figure 10 .
Figure 10.The powder of a) AC5, b) the SEM micrograph, c,d) the TEM, and e-g) the elemental mapping of the elements on the surface of AC5, showing the distribution of carbon, nitrogen, and oxygen atoms, respectively.

Figure 12 .
Figure 12. a) The recovery efficiency, b) uptake capacity, and c) the distribution coefficients of the typical metal ions present in seawater on AC5 during silver recovery from 30 ppm aqueous solutions.Silver recovery in the presence of competing metal ions present in e-waste is presented in (d).

Figure 13 .
Figure 13.a) The recovery efficiency of Ag + ions at different pH, and b) the solutions containing Ag + ions prior to the addition of AC5 (top) and after the recovery before filtering and collection of AC5 (bottom).

Figure 14 .
Figure 14.a) The silver recovery cycles on AC5 and b) the corresponding cycling performance.The FTIR spectrum of c) AC5 at the end of 5 cycles is presented, while d) the SEM micrograph is shown.

Table 1 .
The recovery of silver in practical seawater using AC5.

Table 2 .
Performance comparison among related materials.The highest reported in the literature; ± recovery efficiency in synthetic water; # recovery efficiency in real seawater/wastewater; N.R: not reported.