Methods and materials for precious metal separation

Abstract

Embodiments generally relate to compositions and methods for separation of precious metals, such as silver, gold, palladium, etc., from a liquid source. Recovering precious metals may including adding a precipitating agent to a liquid source to form a first solution including a first supernatant and a first precipitate, which contains a first metal, and separating the first precipitate from the first supernatant. A first cationic cellulose-based sorbent may be added to the first supernatant to form a second solution including a second supernatant and a second precipitate, which contains a second metal, and separating the second precipitate from the second supernatant. A second cationic cellulose-based sorbent may be added to the second supernatant to form a third solution including a third supernatant and a third precipitate, which contains a third metal, and separating the third precipitate from the third supernatant.

Description

Atty. Ref. No. 0073605-001030 METHODS AND MATERIALS FOR PRECIOUS METAL SEPARATION CROSS-REFERENCE TO RELATED APPLICATIONS

[0001] This patent application is related to and claims the benefit of priority of U.S. ProvisionalApplication 63 / 696,563, filed on September 19, 2024, the entire contents of which are incorporated by reference. STATEMENT REGARDING FEDERALLY SPONSORED RESEARCH DEVELOPMENT

[0002] This invention was made with government support under Grant No. DE-EE0007898awarded by the Department of Energy. The Government has certain rights in the invention. FIELD

[0003] Embodiments relate to compositions and methods for separation of precious metals froma liquid source. In particular, embodiments relate to cellulose-based sorbents configured to selectively precipitate precious metals from liquid waste. BACKGROUND

[0004] The global demand for precious metals, including gold, silver, and palladium, has seen asignificant surge in recent years. These metals are integral to various applications, such as electrical conductors, dental crowns, catalysts, building decorations, jewelry, and as a hedge against inflation in investment portfolios. The global market for precious metals is projected to expand from approximately $240.5 billion in 2021 to an estimated $415.4 billion by 2028, highlighting the increasing reliance on these materials across multiple sectors. This rising demand underscores the necessity for sustainable and efficient methods of sourcing precious metals, particularly from secondary sources.Atty. Ref. No. 0073605-001030

[0005] One of the primary sources of precious metals is waste printed circuit boards (WPCB),which are prevalent in electronic devices. However, the disposal of WPCB poses significant environmental challenges, contributing to the growing issue of waste electric and electronic equipment (WEEE). In 2022, the production of WEEE reached 59.4 million metric tons, with projections indicating a continued upward trend. The effective separation and recovery of precious metals from WEEE are critical not only to meet the increasing demand but also to mitigate the environmental impact associated with traditional disposal methods. Current practices for recovering precious metals from WEEE often involve energy-intensive processes, which can exacerbate the environmental burden.

[0006] The recovery of precious metals from WPCB typically involves mechanical and chemicalprocesses, including pyrometallurgical and hydrometallurgical methods. While techniques such as solvent extraction and electrowinning have been utilized for separation and purification, they often face challenges related to high costs, significant energy consumption, and negative environmental impacts. Adsorption methods have emerged as a promising alternative due to their simplicity and milder processing conditions. Various adsorbents, including reduced graphene oxide and metal-organic frameworks, have been developed; however, they still encounter issues related to high costs, low durability, and selectivity against impurities. Additionally, biosorption using microorganisms offers a sustainable approach but is limited by low recovery capacities and stringent environmental requirements. SUMMARY

[0007] We have developed a process for the selective separation and recovery of precious metalsfrom a multi-metal mixture. This process leverages negatively charged ions and cationic hairy cellulose nanocrystals (CHCNC), along with a CHCNC-coated sorbent, to facilitate effectiveAtty. Ref. No. 0073605-001030 metal ion extraction. By employing a sustainable, bio-based nanotechnology approach, our process aims to address the growing demand for precious metals across various industrial sectors, including electronics, automotive, catalysis, and renewable energy.

[0008] In particular, our process includes an initial step of precipitating a first metal from amulti-metal solution using negatively charged ions, a subsequent step of precipitating a second metal using CHCNC, and a third step of precipitating a third metal using a CHCNC-coated sorbent. This multi-step process offers a promising alternative to conventional separation practices, such as solvent extraction, which are often associated with high costs and significant environmental impacts.

[0009] Ultimately, our process not only enhances the efficiency of precious metal extraction butalso supports the transition toward a circular economy by promoting sustainable practices in precious metal recovery.

[0010] In an exemplary embodiment, a method for recovering precious metals from a liquidsource comprising a first metal, a second metal, and a third metal, comprises adding a precipitating agent to the liquid source to form a first solution comprising a first supernatant and a first precipitate, wherein the first precipitate comprises the first metal; separating the first precipitate from the first supernatant; adding a first cationic cellulose-based sorbent to the first supernatant to form a second solution comprising a second supernatant and a second precipitate, wherein the second precipitate comprises the second metal; separating the second precipitate from the second supernatant; adding a second cationic cellulose-based sorbent to the second supernatant to form a third solution comprising a third supernatant and a third precipitate, wherein the third precipitate comprises the third metal; and separating the third precipitate from the third supernatant.Atty. Ref. No. 0073605-001030

[0011] In some embodiments, the first cellulose-based sorbent comprises cationic hairy cellulosenanocrystals.

[0012] In some embodiments, the cationic hairy cellulose nanocrystals comprise cellulose bodiesbearing cellulose chains protruding therefrom, wherein the cellulose chains are functionalized with cationic groups.

[0013] In some embodiments, the cationic groups are quaternary ammonium groups.

[0014] In some embodiments, the second cationic cellulose-based sorbent comprises cationichairy cellulose nanocrystals immobilized on a substrate.

[0015] In some embodiments, the substrate is microcrystalline cellulose.

[0016] In some embodiments, the precipitating agent comprises one or more compoundsconfigured to react with the first metal via a precipitation reaction.

[0017] In some embodiments, the precipitating agent comprises sodium chloride.

[0018] In some embodiments, the first metal is silver.

[0019] In some embodiments, the second metal is gold.

[0020] In some embodiments, the third metal is palladium.

[0021] In an exemplary embodiment, a method for recovering precious metals from a liquidsource comprising a first metal, a second metal, and a third metal, comprises adding a precipitating agent to the liquid source to form a first solution comprising a first supernatant and a first precipitate, wherein the first precipitate comprises the first metal; separating the first precipitate from the first supernatant; adding cationic hairy cellulose nanocrystals to the first supernatant to form a second solution comprising a second supernatant and a second precipitate, wherein the second precipitate comprises the second metal; separating the second precipitate from the second supernatant; adding cationic hairy cellulose nanocrystals immobilized on aAtty. Ref. No. 0073605-001030 substrate to the second supernatant to form a third solution comprising a third supernatant and a third precipitate, wherein the third precipitate comprises the third metal; and separating the third precipitate from the third supernatant.

[0022] In an exemplary embodiment, a method for recovering precious metals from a liquidsource comprising a first metal, a second metal, and a third metal, comprises directing a feed and a precipitating agent to a first metal removal unit, wherein the precipitating agent is configured to react with the feed to form a first solution comprising a first supernatant and a first precipitate, wherein the first precipitate comprises the first metal; separating the first precipitate from the first supernatant via a first filtration mechanism; adding a first cationic cellulose-based sorbent and the first supernatant to a second metal removal unit, wherein the first cationic cellulose- based sorbent is configured to react with the first supernatant to form a second solution comprising a second supernatant and a second precipitate, wherein the second precipitate comprises the second metal; separating the second precipitate from the second supernatant via a second filtration mechanism; adding a second cationic cellulose-based sorbent and the second supernatant to a third metal removal unit, wherein the second cationic cellulose-based sorbent is configured to react with the second supernatant to form a third solution comprising a third supernatant and a third precipitate, wherein the third precipitate comprises the third metal; and separating the third precipitate from the third supernatant via a third filtration mechanism.

[0023] In some embodiments, the first metal removal unit comprises a centrifuge configured toprecipitate the first metal from the feed.

[0024] In some embodiments, the second metal removal unit comprises a centrifuge configuredto precipitate the second metal from the first supernatant.Atty. Ref. No. 0073605-001030

[0025] In some embodiments, the third metal removal unit comprises a centrifuge configured toprecipitate the third metal from the second supernatant.

[0026] In some embodiments, the second metal removal unit is a bed packed with the firstcationic cellulose-based sorbent.

[0027] In some embodiments, the third metal removal unit is a bed packed with the secondcationic cellulose-based sorbent.

[0028] Other details, objects, and advantages will become apparent as the following descriptionof certain exemplary embodiments thereof proceeds. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] The above and other objects, aspects, features, advantages, and possible applications ofembodiments of the present innovation will be more apparent from the following more particular description thereof, presented in conjunction with the following drawings. Like reference numbers used in the drawings may identify like components.

[0030] FIG. 1 is a flow diagram demonstrating an exemplary method of separating andrecovering precious metals from a multi-metal mixture.

[0031] FIG. 2 is a flow diagram demonstrating an exemplary process of separating andrecovering precious metals from a multi-metal mixture.

[0032] FIG. 3 is a schematic illustration demonstrating selective precious metal recovery froman Au-Pd-Ag ion mixture via three steps, involving NaCl, CHCNC, and a CHCNC-coated sorbent.

[0033] FIG. 4 is a schematic illustration of CHCNC synthesis via periodate-mediated oxidationof cellulose fibrils (yielding DAMC), and subsequent cationization through a Schiff base reaction to functionalize DAMC with quaternary ammonium groups.Atty. Ref. No. 0073605-001030

[0034] FIG. 5 is a graph showing DAMC aldehyde group content, determined by a NH2OH·HCltitration method.

[0035] FIG. 6 is a graph showing CHCNC quaternary ammonium group content, measured via aAgCl titration method.

[0036] FIG. 7 is a representative atomic force microscopy (AFM) image of CHCNC.

[0037] FIG. 8 is an Attenuated Total Reflectance Fourier Transform Infrared (ATR-FTIR)spectra of DAMC and CHCNC.

[0038] FIG. 9 is a schematic illustration of CHCNC interactions with AuCl4- and PdCl42- at pH ~2. CHCNC electrostatically bind AuCl4- and reduce it to Au nanoparticles, successfully separating Au. CHCNC electrostatically bind PdCl42-; however, the aggregates remain colloidally stable at sub-stoichiometric PdCl42-concentrations with respect to the quaternary ammonium groups.

[0039] FIG. 10 is a graph showing the effect of pH on Au precipitation (without CHCNC).

[0040] FIG. 11 is a graph showing the effect of contact time on CHCNC-mediated Au recoverycapacity at varying initial Au concentrations (C0,Au). The contact time excludes centrifugation time (~ 5 min).

[0041] FIG. 12 is a graph showing CHCNC-mediated Au recovery percentage (RAu) at varyingC0,Au.

[0042] FIG. 13 is a graph showing Au recovery capacity (qe,Au) of CHCNC versus equilibriumAu concentration (Ce,Au).

[0043] FIG. 14 is a graph showing the effect of varying C0,Au on CHCNC hydrodynamic sizedistributions.Atty. Ref. No. 0073605-001030

[0044] FIG. 15 is a graph showing the effect of contact time on the hydrodynamic size ofCHCNC at varying C0,Au.

[0045] FIG. 16 is a graph showing qe,Au at varying C0,Au in a NaCl solution (3200 ppm).

[0046] FIG. 17 is a graph showing the effect of NaCl concentration on the CHCNChydrodynamic size.

[0047] FIG. 18 is a schematic illustration of monovalent anion effects of CHCNC.

[0048] FIG. 19 is a graph showing precipitated Pd without CHCNC at varying pH.

[0049] FIG. 20 is a graph showing the effect of contact time on CHCNC-mediated Pd recoverycapacity (qe,Pd) at varying initial Pd concentrations (C0,Pd). The centrifugation time (5 min) is excluded.

[0050] FIG. 21 is a graph showing the effect of C0,Pd on CHCNC-mediated Pd recoverypercentage (RPd).

[0051] FIG. 22 is a graph showing the effect of C0,Pd on CHCNC-mediated Pd recovery capacity.

[0052] FIG. 23 is a graph showing the effect of C0,Pd on CHCNC hydrodynamic equivalent size.

[0053] FIG. 24 is a graph showing hydrodynamic size of CHCNC incubated with Pd for varyingincubation time.

[0054] FIG. 25 is a schematic illustration of CHCNC interactions with varying concentrations ofPdCl42-.

[0055] FIG. 26 shows transmission electron microscopy (TEM) images of (left) precipitate phaseand (right) supernatant of AuCl4--CHCNC mixture.

[0056] FIG. 27 shows TEM images of (left) precipitate phase and (right) supernatant of PdCl42--CHCNC mixture.Atty. Ref. No. 0073605-001030

[0057] FIG. 28 includes graphs showing X-ray photoelectron spectroscopy (XPS) analysis of(left) AuCl4- and (right) PdCl42-(pH=2).

[0058] FIG. 29 shows (left) the XPS Au 4f peaks and (right) EDS mapping of CHCNC-Auprecipitates with the scanning electron microscope (SEM) image in the inset.

[0059] FIG. 30 shows (left) the XPS Au 4f peaks and (right) EDS mapping of CHCNC-Ausupernatants with the SEM image in the inset.

[0060] FIG. 31 shows (left) the XPS Pd 3d peaks and (right) EDS mapping of CHCNC-PdCl42-precipitates with the SEM image in the inset.

[0061] FIG. 32 shows (left) the XPS Pd 3d peaks and (right) EDS mapping of CHCNC-PdCl42-supernatants with the SEM image in the inset.

[0062] FIG. 33 is a graph showing X-ray diffraction (XRD) spectroscopy conducted on CHCNC.

[0063] FIG. 34 is a graph showing XRD spectroscopy conducted on AuCl4- -CHCNC precipitate.

[0064] FIG. 35 is a graph showing XRD spectroscopy conducted on PdCl42—CHCNCprecipitate.

[0065] FIG. 36 is a graph showing XRD spectroscopy conducted on PdCl42—CHCNCsupernatant.

[0066] FIG. 37 is a schematic illustration showing the electrostatic interactions of PdCl42- and thequaternary ammonium group of CHCNC, resulting in stable CHCNC- PdCl42-aggregates that do not undergo precipitation at C0,Pd< 53 ppm and CHCNC dosage = 500 ppm.

[0067] FIG. 38 is a schematic illustration showing a CHCNC-coated microcrystalline celluloses(e.g., MINC+) and its composition.Atty. Ref. No. 0073605-001030

[0068] FIG. 39 demonstrates formation of the CHCNC-coated microcrystalline celluloses(MCC) via the covalent linkage between DA and CHCNC based on a Schiff base reaction, which simultaneously takes place with DA polymerization to PDA (not illustrated).

[0069] FIG. 40 is a graph showing the content of cationic groups of the CHCNC-coated MCC,measured by titration.

[0070] FIG. 41 shows optical microscopy images of MCC and CHCNC-coated MCC.

[0071] FIG. 42 is a graph showing Pd (C0,Pd = 10 or 200 ppm)recovery percentage of CHCNC,MCC, and CHCNC-coated MCC.

[0072] FIG. 43 includes graphs showing recovery of Pd at (left) C0,Pd = 10 ppm, and (right) C0,Pd= 200 ppm (pH = 2).

[0073] FIG. 44 is a process flow diagram of precious metal recovery via successive separationsteps.

[0074] FIG. 45 shows distribution coefficient (Kd) values of metals separated by (top left) NaClin the first step, (top right) CHCNC in the second step, and (bottom) CHCNC-coated MCC in the third step, for an Ag-Au-Pd mixture.

[0075] FIG. 46 shows graphs of (left) precious metal R of NaCl in the leachate (step 1), and(right) the corresponding Kdvalues for leachate components.

[0076] FIG. 47 shows graphs of (left) precious metal R of CHCNC in the leachate (step 2), and(right) the corresponding Kdvalues for leachate components.

[0077] FIG. 48 shows graphs of (left) precious metal R of CHCNC-coated MCC in the leachate(step 3), and (right) the corresponding Kdvalues for leachate components.Atty. Ref. No. 0073605-001030

[0078] FIG. 49 is a plot demonstrating CHCNC-mediated Au recovery capacity and contact timecompared with the existing natural and synthetic adsorbents. MOF stands for metal–organic frameworks.

[0079] FIG. 50 is a plot demonstrating CHCNC-coated sorbent -mediated Pd recovery capacityand contact time compared with the existing natural and synthetic adsorbents. DETAILED DESCRIPTION

[0080] The following description is of exemplary embodiments and methods of use that arepresently contemplated for carrying out the present invention. This description is not to be taken in a limiting sense, but is made merely for the purpose of describing the general principles and features of various aspects of the present invention. The scope of the present invention is not limited by this description.

[0081] Embodiments relate to methods and processes of selectively separating and recoveringmetals (e.g., precious metals) from a liquid source. As used herein, the term “liquid source” may refer to any liquid material or medium that contains a plurality of metal ions dissolved or otherwise provided therein. The number of metals provided in the liquid source is not particularly limited. In some embodiments, the liquid source may include at least two metals, namely a first metal and a second metal. In other embodiments, the liquid source may include at least three metals, namely a first metal, a second metal, and a third metal. In still other embodiments, the liquid source may include at least four metals, namely a first metal, a second metal, a third metal, and a fourth metal.

[0082] As nonlimiting examples, exemplary liquid sources may include naturally occurringbodies of water (e.g., rivers, lakes, etc.), industrial process streams, industrial waste streams, leachates from mineral ores, solutions generated from chemical processes, and / or the like. In oneAtty. Ref. No. 0073605-001030 example, the liquid source may be formed from secondary sources, such as end-of-like electric and electronic equipment and products.

[0083] Referring to FIG. 1, a method 100 of selectively separating and recovering metals from aliquid source may include a step 110 of adding a precipitating agent to the liquid source to form a first solution. In particular, the precipitating agent is configured to selectively precipitate a first metal from the liquid source. For example, the precipitating agent may separate and recover the first metal via a precipitation reaction.

[0084] In some embodiments, the precipitating agent may be sodium chloride (NaCl) or anycompound able to precipitate the first metal form the liquid source.

[0085] In some embodiments, the first metal may be silver or a silver-like metal.

[0086] In one exemplary embodiment, the precipitating agent is NaCl and the first metal issilver. For example, the NaCl may react with a compound containing silver (e.g., silver nitrate (AgNO3)) to form a first solution including an insoluble silver chloride (AgCl) precipitate.

[0087] The method 100 may further include a step 115 of removing the precipitate, whichincludes the first metal, from the first solution. After addition of the precipitating agent to the liquid source, the solution may be vortexed and / or centrifuged to form a first supernatant and a first precipitate. The first precipitate may be separated from the first supernatant (e.g., via a filter, skimmer, or other suitable device), and the first supernatant may be passed for further processing.

[0088] The method 100 may further include a step 120 of adding a first cationic cellulose-basedsorbent to the first supernatant from step 115 to form a second solution. In particular, the first cationic cellulose-based sorbent is configured to selectively precipitate a second metal from the first supernatant. For example, the first cationic cellulose-based sorbent may separate andAtty. Ref. No. 0073605-001030 recover the second metal from the first supernatant via electrostatic attraction and reduction reactions.

[0089] In some embodiments, the first cationic cellulose-based sorbent may include cationichairy cellulose nanocrystals (CHCNC). The CHCNC include cellulose bodies bearing cellulose chains protruding therefrom. The cellulose chains are functionalized with cationic groups, thus providing the sorbent with a positive charge. In such embodiments, the sorbent may absorb the negatively charged metal or compound containing the metal via electrostatic attraction with the cationic groups, followed by interactions with hydroxyl groups of the CHCNC and reduction to a metal ion. This reduction process leads to the precipitation of the metal ion / nanoparticle from the solution.

[0090] In some embodiments, the cationic groups may be quaternary ammonium groups.

[0091] In some embodiments, the second metal may be in the form of a negatively chargedcomplex, such that the cationic sorbent may readily bind the complex.

[0092] In some embodiments, the second metal may be gold or a gold-like metal. In someembodiments, gold may be in the form of a negatively charged complex. As nonlimiting examples, gold may be in the form of AuCl4-, AuCl3(OH)-, AuCl2(OH)2-, and / or the like.

[0093] In one exemplary embodiment, the first cationic cellulose-based sorbent is CHCNC andthe second metal is gold. For example, the CHCNC may interact with the negatively charged gold complex to precipitate metallic gold / gold nanoparticles.

[0094] The method 100 may further include a step 125 of removing the precipitate, whichincludes the second metal and may be in the form of metal nanoparticles, from the second solution. After addition of the sorbent to the first supernatant, the solution may be vortexed and / or centrifuged to form a second supernatant and a second precipitate. The second precipitateAtty. Ref. No. 0073605-001030 may be separated from the second supernatant (e.g., via a filter, skimmer, or other suitable device), and the second supernatant may be passed for further processing.

[0095] The method 100 may further include a step 130 of adding a second cationic cellulose-based sorbent to the second supernatant from step 125 to form a third solution. In particular, the second cationic cellulose-based sorbent is configured to selectively precipitate a third metal from the second supernatant. For example, the second cationic cellulose-based sorbent may separate and recover the third metal or a compound including the third metal from the second supernatant via electrostatic interactions.

[0096] In some embodiments, the second cationic cellulose-based sorbent may include CHCNC,as described above, immobilized / coated on a substrate. The substrate is not particularly limited and may be any material suitable to accommodate immobilization of the CHCNC to its surface. The substrate may include, but not be limited to, microcrystalline cellulose (MCC), silica gel, and / or the like. It is contemplated that binding the CHCNC to the substrate does not compromise the cationic groups, as it only involves aldehyde groups of the CHCNC.

[0097] It is contemplated that the CHCNC has a higher affinity for the second metal than othermetals present in the solution (e.g., the third metal).

[0098] In some embodiments, the third metal may be in the form of a negatively chargedcomplex, such that the cationic sorbent may readily bind the complex.

[0099] In some embodiments, the second metal may be palladium or a palladium-like metal. Insome embodiments, palladium may be in the form of a negatively charged complex. As nonlimiting examples, gold may be in the form of PdCl42-and / or the like.

[0100] In one exemplary embodiment, the second cationic cellulose-based sorbent isCHCNC coated on a substrate and the third metal is palladium. For example, the CHCNC-coatedAtty. Ref. No. 0073605-001030 substrate may interact with the negatively charged palladium complex to precipitate metallic palladium / palladium nanoparticles. While CHCNC alone may bind to the palladium to form CHCNC-Pd aggregates, the aggregates may remain colloidally stable in the solution and may not undergo significant phase separation, thus complicating recovery of the palladium.

[0101] The method 100 may further include a step 135 of removing the precipitate,which includes the third metal and may be in the form of metal nanoparticles, from the third solution. After addition of the sorbent to the second supernatant, the solution may be vortexed and / or centrifuged to form a third supernatant and a third precipitate. The third precipitate may be separated from the third supernatant (e.g., via a filter, skimmer, or other suitable device), and the third supernatant may be passed for further processing.

[0102] In one exemplary embodiment, a liquid source including at least silver, gold, andpalladium ions may be provided. NaCl may be provided as a precipitating agent to the liquid source to selectively precipitate an AgCl precipitate. CHCNC may then be added to the resulting supernatant to selectively precipitate gold in the form of metallic gold / gold nanoparticles. A CHCNC-coated substrate may then be added to the resulting supernatant to precipitate palladium in the form of metallic palladium / palladium nanoparticles, thus achieving selective, sequential separation of silver, gold, and palladium.

[0103] The process utilized to carry out the above described method may be a semi-batchprocess or a continuous process.

[0104] Referring to FIG. 2, a feed 205 (e.g., liquid source) can be directed to a first metalremoval unit 215. A precipitating agent 210 can also be directed to the first metal removal unit 215, such that the precipitating agent 210 and the feed 205 together form a first solution. TheAtty. Ref. No. 0073605-001030 precipitating agent 210 is configured to react with the feed 205 within the first metal removal unit 215 and to selectively precipitate a first metal.

[0105] The first metal removal unit 215 may be a tank, reactor, column, etc. The firstmetal removal unit 215 may be operated at a predetermined temperature and a predetermined pressure. The predetermined temperature and / or the predetermined pressure may be selected to promote reaction between the feed 205 and the precipitating agent 210.

[0106] The first solution may be separated into a first precipitate 220 and a firstsupernatant 225, in which the first precipitate 220 includes the first metal. In some embodiments, a filtration mechanism (not shown), such as a filter, skimmer, or other suitable device, may be used to separate the first solution into the first precipitate 220 and the first supernatant 225.

[0107] The first supernatant 225 may be directed to a second metal removal unit 235. Insome embodiments, a first cationic cellulose-based sorbent 230 can also be directed to the second metal removal unit 235, such that the first cationic cellulose-based sorbent 230 and the first supernatant 225 together form a second solution. In other embodiments, the second metal removal unit 235 may include a bed packed with the first cationic cellulose-based sorbent 230. In either case, the first cationic cellulose-based sorbent 230 is configured to react with the first supernatant 225 within the second metal removal unit 235.

[0108] The second metal removal unit 235 may be a tank, reactor, column, etc. Thesecond metal removal unit 235 may be operated at a predetermined temperature and a predetermined pressure. The predetermined temperature and / or the predetermined pressure may be selected to promote reaction between the first supernatant 225 and the first cationic cellulose- based sorbent 230.Atty. Ref. No. 0073605-001030

[0109] The second solution may be separated into a second precipitate 240 and a secondsupernatant 245, in which the second precipitate 240 includes the second metal. In some embodiments, a filtration mechanism (not shown), such as a filter, skimmer, or other suitable device, may be used to separate the second solution into the second precipitate 240 and the second supernatant 245.

[0110] The second supernatant 245 may be directed to a third metal removal unit 255.In some embodiments, a second cationic cellulose-based sorbent 250 can also be directed to the third metal removal unit 255, such that the second cationic cellulose-based sorbent and the second supernatant together form a third solution. In other embodiments, the third metal removal unit 255 may include a bed packed with the second cationic cellulose-based sorbent 250. In either case, the second cationic cellulose-based sorbent 250 is configured to react with the second supernatant 245 within the third metal removal unit 255.

[0111] The third metal removal unit 255 may be a tank, reactor, column, etc. The thirdmetal removal unit 255 may be operated at a predetermined temperature and a predetermined pressure. The predetermined temperature and / or the predetermined pressure may be selected to promote reaction between the second supernatant 245 and the second cationic cellulose-based sorbent 250.

[0112] The third solution may be separated into a third precipitate 260 and a wastewaterstream 265, in which the third precipitate 260 includes the third metal. In some embodiments, a filtration mechanism (not shown), such as a filter, skimmer, or other suitable device, may be used to separate the third solution into the third precipitate 260 and the wastewater stream 265.

[0113] The metals recovered from the methods and processes described herein may beused in a number of different applications, including but not limited to, electronic and electricalAtty. Ref. No. 0073605-001030 applications (e.g., electrical conductors, connectors, switches, capacitors, resistors, and the like), automative applications (e.g., catalytic converters, electrical contacts, sensors, and the like), catalysis and chemical processing (e.g., catalysts and the like), dental applications (e.g., crowns, fillings, bridges, implants, and the like), construction and architecture, and jewelry / personal accessories. EXAMPLES

[0114] Materials: Sheets of northern bleached delignified softwood kraft pulp wereprovided by Resolute Inc. (Canada). Hydrochloric acid (HCl, ACS reagent, 37 %), gold(III) chloride solution (HAuCl4, 99.99 % trace metal basis, 30 wt% in dilute HCl, also known as auric chloride and gold chloride), potassium tetrachloropalladate(II) (K2PdCl4, ACS reagent, 99.0 %), silver nitrate (AgNO3, >99 %), sodium chloride (NaCl, >99.5 %), calcium chloride dihydrate (CaCl2⋅2H2O, for molecular biology, ≥99.0 %), sodium hydroxide (NaOH, ACS Reagent, >97%), nitric acid (HNO3, ≥90.0 %), hydroxylamine hydrochloride (NH2OH⋅HCl, ReagentPlus®, 99%), sodium (meta)periodate (NaIO4, >99.0 %), ethylene glycol (C2H6O2, ReagentPlus®, >99 %), sodium chlorite (NaClO2, 80 %), hydrogen peroxide (H2O2, 30 wt%), isopropyl alcohol (CH3CHOHCH3, anhydrous, >99.7 %), ethylenediaminetetraacetic acid (EDTA, C10H16N2O8, 99.4 %), dopamine hydrochloride (DA, C8H11NO2, >99 %), microcrystalline cellulose (MCC, 99 %), Tris buffer (1.0 M, pH =7.0), (2-hydrazinyl-2-oxoethyl)-trimethylazanium chloride (Girard’s reagent T, GT, C5H14ClN3O, 98 %) and WhatmanTM filter papers (grade 1, circles) were purchased from MilliporeSigma, USA. Mica sheets (V1 grade) and AFM stainless steel disks were obtained from Ted Pella Inc. and SPI supplies (USA), respectively. Copper grids (400 mesh) coated with a Formvar / Carbon film and uranyl acetate solution (2 % w / v) were provided by Electron Microscopy Sciences, USA. Isopropyl alcohol (IPA, propanol) and plastic centrifugeAtty. Ref. No. 0073605-001030 tubes (15 mL and 50 mL) were supplied by KOPTEC (USA) and VWR International (USA), respectively. Unless otherwise specified, ultrapure (Milli-Q) water (resistivity =18.2 MΩ cm at 25℃), produced via passing deionized (DI) water through an ultrafilter (Biopak Polisher, Millipore, USA), was used in all experiments. Thiourea (ACS reagent, ≥99.0 %, Sigma-Aldrich, USA), sodium chlorite (NaClO, 12.5 %, Alliance Chemical, USA), sulfuric acid (H2SO4, 95–98 % w / v, VWR International, USA), ferric ion (Fe(III), ≥99 %, Reade International, USA) were purchased. A mixture of char and metals obtained from processing WPCB through a thermolyzer process was provided by CHZ Technologies LLC., USA.

[0115] Methods – CHCNC Synthesis and Characterization: The CHCNC wassynthesized via an established protocol. In brief, 1 g of softwood pulp was sliced into thin pieces (roughly 2 ×2 cm2) and soaked in DI water for at least 24 h. The wet pulp was disintegrated in a fruit blender (Cleanblend, USA) for 10 min, resulting in a pulp slurry, followed by vacuum filtration. To initiate a periodate oxidation reaction for dialdehyde modified cellulose (DAMC) synthesis, the wet pulp containing DI water (total volume =65 mL), 1.32 g of NaIO4, and 3.8 g of NaCl were mixed in a beaker. The mixture was stirred for 42 h at room temperature while covered to avoid light- induced periodate deactivation. To terminate the reaction, 1 mL of ethylene glycol was added to the mixture, quenching the remaining unreacted NaIO4. DAMC fibrils were then rinsed at least 5 times with DI water. Each rinse involved dispersing the fibrils in 50 mL of DI water, stirring for 5 min, and vacuum filtration.

[0116] To convert the dialdehyde groups of DAMC to quaternary ammonium groups,DAMC fibrils were reacted with GT. To this end, 1.1 g of GT was added to a suspension of never-dried DAMC in DI water (total volume =50 mL), followed by stirring for 24 h at roomAtty. Ref. No. 0073605-001030 temperature. The pH was maintained at 4.5 ±0.2 by the intermittent addition of a NaOH solution (0.5 M) during the first 5 h of reaction. Afterward, the fibrils were thoroughly rinsed with DI water five times and vacuum filtered. Gel-like fibrils were redispersed in DI water (total volume =100 mL) and heated to 60℃ for 2 h while stirring continuously. The resulting suspension was sonicated using a probe sonicator (diameter =1 / 2 in., Qsonica Q500, USA) at 60 % amplitude for 10 min in an ice bath and centrifugated (10,000×g, 15 min) to remove non-fibrillated fibers, which were negligible. Then, isopropyl alcohol (30 mL) was added to the supernatant (50 mL) at a 1:1.6 vol ratio, and the gel-like precipitate (i.e., CHCNC) was collected by centrifugation (8000×g, 10 min). The CHCNC were finally redispersed in 50 mL of DI water and purified by dialysis (Spectra / Por dialysis bags, MW cutoff =6–8 kDa) against DI water for 3 days. The water was exchanged twice daily. The amount of DAMC and CHCNC aldehyde groups, as well as CHCNC and MINC+ (e.g., CHCNC coated on MCC) quaternary ammonium groups, were determined. Additionally, the CHCNC morphology was investigated using AFM imaging.

[0117] Methods – PM Recovery Experiments: Stock solutions of HAuCl4, K2PdCl4, orAgNO3, containing 1000 ppm of Au, Pd, or Ag, respectively, were prepared by diluting the HAuCl4solution (30 wt% in dilute HCl), dissolving K2PdCl4, or dissolving AgNO3in 100 mL of ultrapure water, respectively.

[0118] Elemental analysis of solutions was performed using Thermo iCAP 7400inductively coupled plasma-atomic emission spectrometry (ICP- AES) or Aligent 7800 inductively coupled plasma–mass spectroscopy (ICP-MS) at Penn State’s Laboratory for Isotopes and Metals in the Environment (LIME) or Center for Critical Minerals (C2M), respectively. The quality control was performed through the analysis of blank and replicate samples with varying dilution factors, as well as internal and standard solutions with knownAtty. Ref. No. 0073605-001030 elemental concentrations (i.e., ICP-MS Multi Element Solution Standard with 48 and 13 Components (10 ppm) from High-Purity Standards company (HPS), and Precious Metals ICP- MS Standard from Thomas Scientific).

[0119] The separation experiments of Au and Pd chloride complexes were conductedusing CHCNC as an adsorbent, and for Ag recovery, NaCl was used as a precipitating agent. Varying concentrations of Au, Pd, or Ag solutions, ranging from 1 ppm to 800 ppm, were prepared via the successive dilution of corresponding stock solutions with ultrapure water, (total volume =7 mL, including CHCNC or NaCl and the acid used for pH adjustment). The pH of Au or Pd solutions was adjusted to 2 using a HCl solution (0.5 M), and a HNO3solution (0.1 M) was used for adjusting the pH of Ag solution to 2. The same amount of CHCNC (dosage or concentration =500 ppm) was added to the Au or Pd solutions, followed by vortexing for desired incubation times (1, 10, 30, 60, 120, 240, or 360 min). To separate Ag as AgCl based on the metric ratio of Cl- (NaCl)-to-Ag+(AgNO3), a NaCl solution (3 M) was added to the Ag solutions to obtain a NaCl final concentration of 18.7 mM, followed by vortexing for 5 min. Samples were then centrifuged at 6000×g for 5 min, supernatants (5 mL) were gently separated, and the concentration of unadsorbed (free) Au or Pd or unprecipitated Ag was measured using the ICP- MS. The recovery percentage (R, %) and recovery capacity at equilibrium (qe, mg g-1) were calculated using equations (1) and (2), respectively: ^^^^^^(%) = ^^ ^^× ^^100 (1)(2)Atty. Ref. No. 0073605-001030

[0120] where C0 (mg mL-1) denotes the initial ion concentration, Ce (mg mL-1) is theequilibrium ion concentration, V (mL) is the total solution volume, and m (g) is the mass of CHCNC or NaCl added to the solutions.

[0121] The effect of contact time on CHCNC-mediated Au or Pd adsorption was studiedvia incubating CHCNC (dosage =500 ppm) in an Au or Pd solution (C0 =200 ppm and 800 ppm). The solution pH was adjusted to 2 using a HCl solution (0.5 M), and the total solution volume was 7 mL, including CHCNC and the acid used for pH adjustment. CHCNC were added to the Au or Pd solution, followed by vortexing for varying durations, including 1, 10, 30, 60, 120, 240, or 360 min at room temperature. The mixtures were then centrifuged at 6000×g for 5 min, and supernatants (5 mL) were gently separated for Ce measurement using the ICP-AES.

[0122] The CHCNC hydrodynamic size and corresponding aggregates after Au or Pdadsorption were investigated via dynamic light scattering (DLS, Malvern Zetasizer Nano series, UK) spectroscopy at 90◦scattering angle and room temperature. The initial concentrations of each PM ion were varied from 10 ppm to 800 ppm while maintaining the CHCNC dosage constant at 500 ppm. The total volume was 1 mL, including a HCl solution (0.5 M) used to decrease the pH to 2. For the DLS measurements, 70 μL of Au-CHCNC or Pd-CHCNC dispersions were transferred to ZEN2112 low-volume quartz cuvettes (Malvern, UK) without further dilution. The Z-average of intensity, also known as cumulants mean, was registered as the hydrodynamic equivalent size of nanoparticles.

[0123] Methods – Characterization of CHCNC and PM-CHCNC Aggregates: To identifythe functional groups of DAMC or CHCNC, a FTIR spectrometer (Fisher Scientific, USA) with a Diamax ATR accessory was used, following the Bouguer-Beer-Lambert law. Samples were dried at ~ 37℃ overnight and deposited directly onto the ATR crystal, analyzed at a fixedAtty. Ref. No. 0073605-001030 infrared light angle of incidence (45°). The tip of pressure clamp was lowered to apply maximum pressure. A total of 100 scans were recorded in the transmission mode, ranging from 4000 cm1to 500 cm1, with a resolution of 6 cm1. The spectra were then averaged to obtain representative spectra, and the absorbance data were calculated via referencing the spectrum of clean, bare diamond.

[0124] XPS was conducted on dried CHCNC, HAuCl4, K2PdCl4, Au-CHCNCaggregates, and Pd-CHCNC aggregates using a VersaProbe III instrument (Physical Electronics, Germany) following the standard ISO 15472:2001. The XPS instrument was outfitted with a monochromatic aluminum (Al) kα X-ray source (photon energy, hν =1,486.6 eV) and a concentric hemispherical analyzer. Low energy electrons (<5 eV) and positive argon ions were used to neutralize the surface charges. The binding energy axis was calibrated using copper (Cu 2p3 / 2at 932.62 eV and Cu 3p3 / 2at 75.1 eV) and gold (Au 4f7 / 2at 83.96 eV) foils after sputter- assisted cleaning. Survey spectra were obtained over a range of 0–1200 eV, and the high- resolution spectra of core level electrons were recorded with a pass energy of 23.5 eV and an energy step size of 0.1 eV. The spectra were analyzed using CasaXPS software (Version 2.3.25, accessed via Penn State MCL) to deconvolute the peaks and determine the elemental composition, as well as the chemical state of samples.

[0125] TEM images of Au-CHCNC and Pd-CHCNC aggregates were acquired using aTecnai G2 microscope, equipped with a 200 kV LaB6 electron source (Field Electron and Ion, FEI Company, USA). To prepare samples for imaging, the final concentrations of PM (Au or Pd) and CHCNC were maintained constant at 800 ppm and 500 ppm, respectively. The total sample volume was 1 mL including a HCl solution (0.5 M) used for adjusting pH to 2. After vortexing for 5 min, the aqueous samples were centrifuged at 6000×g for 5 min, and supernatants andAtty. Ref. No. 0073605-001030 precipitates were separated for imaging without any additional modifications. A drop (5 μL) of precipitate or supernatant was placed on a copper grid (400 mesh) coated with a Formvar / Carbon film, incubated for ~ 2 min, and the excess liquid was absorbed with a piece of filter paper. A drop of uranyl acetate solution (2 % w / v) was then placed on the grids, incubated for ~ 2 min, and the excess liquid was recovered. Grids were then dried at room temperature overnight, followed by TEM imaging. The CHCNC length and width were manually measured for at least 50 particles using the ImageJ software (Version 1.53t).

[0126] To examine the elemental composition of Au-CHCNC or Pd-CHCNC aggregates,a field emission scanning electron microscopy (FE-SEM, Carl Zeiss Merlin®, Germany), equipped with an Oxford EDS (Ultim Max 100) and AZtec software (Version 4.2 SP1, accessed via Penn State Nanofabrication Laboratory), was used for imaging and elemental analysis. Elemental mapping was performed at an accelerating voltage of 15 kV. Samples were prepared for elemental mapping via mixing Au or Pd solutions with CHCNC to reach a final concentration of 800 ppm and 500 ppm (adsorbent dosage), respectively. The total sample volume was 1 mL, including a HCl solution (0.5 M) used for pH adjustment to 2. After vortexing for 5 min, mixtures were centrifuged at 6000 ×g for 10 min, and the supernatants and precipitates were separated and used for analyses. To this end, samples were frozen at -80 ◦C and lyophilized at 0.01 mbar for 48 h using a FreeZone benchtop freeze dryer (Labconco, USA).

[0127] To examine the crystalline structure of CHCNC, Au-CHCNC, and Pd- CHCNCaggregates, XRD was conducted using an X-ray diffractometer (Panalytical Empyrean, Malvern, USA) with a radiation source (λ =1.54 Å). Samples were prepared via mixing an Au or Pd solution (final concentration =800 ppm) with CHCNC (dosage =500 ppm). The total sample volume was l mL, including a HCl solution (0.5 M) used for pH adjustment to 2. After vortexingAtty. Ref. No. 0073605-001030 for 5 min, samples were centrifuged at 6000×g for 5 min, supernatants and precipitates were separated, and independently oven-dried at 37℃ overnight. Before conducting the XRD analysis, dry samples were grounded into a fine powder, and a small amount of each sample was deposited on the holder. The fractograms were collected at a voltage of 45 kV and a current of 40 mA using a PIXcel3DX-ray detector at 2θ =10° to 50° (θ denotes the angle of incidence).

[0128] Methods – Immobilization of CHCNC on MCC (e.g., MINC+): To recover lowconcentrations of Pd and enable adsorbent reuse, CHCNC were immobilized on MCC via a mussel-inspired technique to yield MINC+. A previous study showed that a 2 mg mL-1DA solution in 10 mM Tris buffer (pH =8.5), incubated for 24 h, enabled spontaneous formation of a thin film. Here, 200 mg of MCC and 80 mg of DA were mixed with a CHCNC dispersion (final concentration =10 mg mL-1) and Tris buffer (0.4 mL, 1 M) to reach a total volume of 40 mL, including a NaOH solution used for pH adjustment to 8.5. The reaction proceeded in 50 mL centrifuge tubes, covered with aluminum foil, for 24 h at room temperature via agitation on a nutating mixer (60 rpm, Fisherbrand, USA). The suspension color changed from pale yellow to black as a result of DA polymerization and oxidation to polydopamine (PDA). The suspension was then centrifugated at 15000×g for 5 min, the supernatant was decanted, and precipitates were rinsed with ultrapure water (40 mL), followed by centrifugation (3000×g, 5 min) to recover any unreacted CHCNC or DA. This process was repeated three times. The product (e.g., MINC+) was then dried in an oven at 37℃ overnight for further experiments.

[0129] Methods – MINC+ enabled Pd Recovery and Reusability: To investigate MINC+-mediated cyclic Pd recovery, Pd solutions (10 ppm or 200 ppm) were prepared via the successive dilution of K2PdCl4stock solution with ultrapure water. MINC+ (dosage =5000 ppm) was added to the Pd solutions (total sample volume =7 mL, including the HCl used for pH adjustment to 2),Atty. Ref. No. 0073605-001030 vortexed for 2 h at room temperature while covered with an aluminum foil, and centrifuged (6000×g, 5 min). Supernatants (5 mL) were gently separated, and the unadsorbed Pd concentration in the supernatants was measured using the ICP-AES. R and qewere then calculated using Equations (1) and (2), respectively.

[0130] To recover the adsorbed Pd from MINC+, EDTA was added to Pd- adsorbedMINC+ suspensions, desorbing Pd via a ligand-exchange reaction. EDTA has been previously used for Pd desorption from surfactant gels. Briefly, the Pd (10 ppm or 200 ppm)-adsorbed MINC+ (35 mg) was suspended in 7 mL of an EDTA solution (10 mM) at 60℃, and vortexed for 5 min. After cooling to room temperature, the dispersions were centrifugated at 6000×g for 5 min, and the supernatants (5 mL) were separated for the ICP-AES analysis. Desorption percentage (D) was calculated using equation (3): ^^^^ = ^^ ( �^^× � ^^× ^^100% (3)

[0131] where C0,Pd (mgequilibrium desorbed Pd concentration, and R is the recovery percentage in each cycle.

[0132] Methods – Leachate Preparation: The char-metal mixtures obtained from thethermolysis processing of shredded WPCB (CHZ Technologies LLC.) were used for this study. The samples first underwent a selective liberation process to liberate malleable and brittle components in varying size fractions. The liberated materials were then processed using gravity separation techniques, i.e., wet jigging (Alljig Minijig, Allmineral, Germany), shaking table (Deister Model 15-S Laboratory Concentrating Table, Deister Concentrator Company, USA), and multi gravity separator (MGS, Micro MGS, Gravity Mining, UK) for treating +0.595 mm, - 0.595 mm +0.149 mm, and -0.149 mm size fractions, respectively (+ and - denote larger than andAtty. Ref. No. 0073605-001030 smaller than, respectively), and magnetic separation to separate and concentrate metals, including Cu, Al, PM, and ferrous metals. The concentrate product from MGS and shaking table were subsequently leached. A three-step leaching process was used for the selective dissolution of base metals in the first stage, Au and Ag in the second stage, and Pd in the last stage.

[0133] In the first stage, mainly base metals were leached as M(SO4)n (M stands formetal) using H2SO4and H2O2, according to reactions I-IV. The test conditions were adopted from the optimum leaching conditions. The concentrated product (210 g) was leached using a 2 M H2SO4solution and 20 % v / v H2O2at room temperature, 1:100 solid:liquid mass ratio, and stirring rate of 450 rpm. This stage leached 99.95 % of Cu and 98.70 % of Al, with minimal PM dissolution (i.e., Au ~ 0.07 %, Ag ~ 0.09 %, and Pd ~ 0.15 %) from the concentrate product. The solution was then filtered, and the remaining solid residue (~ 25 g) was subjected to the subsequent leaching stages. Cu + H2O2+ H2SO4→ CuSO4+ 2H2O, ΔG° = -77.9 kcal mol-1(I) Zn + H2O2+ H2SO4→ ZnSO4+ 2H2O, ΔG° = -128.0 kcal mol-1(II) Fe + H2O2+ H2SO4→ FeSO4+ 2H2O, ΔG° = -115.8 kcal mol-1(III) Ni + H2O2+ H2SO4→ NiSO4+ 2H2O, ΔG° = -101.2 kcal mol-1(IV)

[0134] Cyanide, halide, thiosulfate, and thiourea are among the reagents used in industryfor the recovery of Au. Cyanide is the most commonly used lixiviants in the industry; however, it has been gradually replaced with thiourea and thiosulfate due to its toxicity and high chemical consumption. While we examined the efficacy of cyanide, thiosulfate, and thiourea leaching for the recovery of Au and Ag, thiourea was selected as the lixiviant for the second step.

[0135] The thiourea leaching process was selected because of faster kinetics for Au andAg, lower environmental impact compared with cyanide leaching, and lower operating costs andAtty. Ref. No. 0073605-001030 sensitivity to the process parameters compared with the thiosulfate leaching, which requires high chemical consumption. The reaction of Au in a H2SO4solution with thiourea and ferric ions (as an oxidant) is presented in reaction V. 2Au + 2Fe3++ 4CS(NH2)2+ SO42-→ [Au(CS(NH2)2)2]SO4+ 2Fe2+(V)

[0136] For thiourea leaching, the solid residue from the first leaching stage was mixedwith a solution of thiourea (0.315 M), H2SO4(0.5 M), and Fe (III) (0.107 M) and stirred at 450 rpm for 72 h at 25℃ and 1:100 solid: liquid mass ratio. This leaching stage recovered additional 0.02 % of Cu and 0.03 % of Al, and mainly recovered Au (97.57 %) and Ag (96.5 %) along with Pd (5.25 %) from the concentrate product. The leaching solution was filtered, and the remaining solid residue (~ 20 g) was subsequently leached (third stage) to recover Pd and the remaining Au and Ag.

[0137] Chloride leaching was used in the third stage. The solid residue of the secondstage (~ 20 g) was mixed with a solution of HCl (5 M), NaClO (10 % v / v), and H2O2(1 % v / v) at a 1:100 solid:liquid mass ratio. The solution was stirred at 450 rpm and 65℃ for 7 h, and then filtered. The remaining solid along with the leachate of all three stages were analyzed using the ICP-MS. This stage leached additional 0.02 % of Cu, 0.21 % of Al, 1.03 % of Au, 3.28 % of Ag, and 83.97 % of Pd. The leachate solutions were then used in PM recovery experiments.

[0138] Methods – Selective Recovery of PM using CHCNC and MINC+: The Au and Pdchloride complexes and Ag+were selectively recovered using CHCNC, MINC+, and NaCl, respectively. Tri-metal model mixtures of Au, Pd and Ag with identical concentrations (10 ppm or 200 ppm each) were prepared by diluting the stock ion solutions (1000 ppm) using ultrapure water and adjusting the final pH to 2 using a HNO3solution (0.1 M, total volume =30 mL, including 500 ppm of CHCNC, 5000 ppm of MINC+, or 50 mM of NaCl). The WPCB leachateAtty. Ref. No. 0073605-001030 obtained from the second stage leaching, containing a mixture of varying ions (shown in Table S1, Supporting Information), was used as received (initial pH =1.9, total volume =10 mL per sample) to which CHCNC (dosage =500 ppm), MINC+(dosage =5000 ppm), or NaCl (final concentration =50 mM) was added. The experiments were designed in three consecutive steps in which NaCl was first used as a precipitating agent for Ag separation, followed by using CHCNC and MINC+ as adsorbents for Au and Pd in the second and third steps, respectively. At each step, after centrifugation (6000×g), 2 mL of supernatants was gently separated to quantify the concentrations of free metals via the ICP-AES. The detail of each step is as follows.

[0139] Step 1 (Ag Recovery). To precipitate Ag as AgCl at a suprastoichiometric molarratio of Cl--to-Ag+(Cl- : Ag+= 50:0.093 for 10 ppm Ag+in the tri-metal model mixtures, 50:1.854 for 200 ppm Ag+in the tri-metal model mixtures, and 50:1.22 for 132 ppm Ag+in the WPCB leachates), a NaCl solution (3 M) was added to the metal mixtures / leachates to reach a final NaCl concentration of 50 mM. After vortexing for 5 min, mixtures were centrifuged (6000×g, 5 min), and a portion of supernatants (2 mL) was separated for the ICP-AES.

[0140] Step 2 (Au recovery). The remaining supernatants from Step 1 were transferred to15 mL centrifuge tubes, and CHCNC were added to the solutions (dosage =500 ppm) to reach a total volume of 10 mL. After vortexing for 5 min, solutions were centrifuged (6000×g, 5 min), and a portion of supernatants (2 mL) were separated for the ICP-AES.

[0141] Step 3 (Pd recovery).MINC+ were added to the remaining supernatants (dosage=5 mg mL-1) to reach a total volume of 8 mL. After vortexing for 120 min using a nutating mixer (60 rpm, Fisherbrand, USA), solutions were centrifuged at 6000×g for 5 min, and the supernatants (2 mL) were separated for the ICP-AES.Atty. Ref. No. 0073605-001030

[0142] To evaluate the selectivity of NaCl, CHCNC, and MINC+ in the recovery of Ag,Au, and Pd, respectively, the distribution coefficient (Kd) was calculated using equation (4): ^^^^� = ^^� � ��^^ (4)

[0143] where C0 (mg mL-concentrations of metals in each step, respectively, V is the solution volume (mL), and m is the adsorbent / precipitant mass (mg). Kd, a parameter for comparing ion recovery capacities, was used to assess the affinity of adsorbents (CHCNC or MINC+) to individual ions in the ion mixtures. A higher Kdindicates a stronger affinity of metal ions to the adsorbent.

[0144] Methods – Statistical Analysis: All the experiments were performed in at leastthree replicates and the data were presented as mean ±standard deviation (SD). Significant differences among mean values were determined via the two-way analysis of variance (ANOVA) for more than three distinct groups, followed by the Tukey–Kramer post hoc test. The symbols *, **, ***, and **** represent p-value (p) <0.05, p <0.01, p <0.001, and p <0.0001, respectively.p ≥0.05 was considered non-significant (ns). The statistical analyses were conducted usingGraphPad Prism software (Version 9.4.1).

[0145] Results – CHCNC synthesis and characterization: The synthesis of nanocellulose-based adsorbents for PM recovery is based on the conversion of delignified cellulose pulp to CHCNC. FIG. 4 presents the CHCNC synthesis via two-step oxidation and Schiff base reactions, followed by heating (60℃, 2 h) and sonication (10 min, in an ice bath). Cellulose fibrils are first converted to DAMC via a NaIO4- mediated oxidation reaction during which cellulose vicinal diols are cleaved at C2-C3 positions and converted to aldehyde groups. The DAMC aldehyde content is 5.9 ±0.2 mmol g-1, obtained from pH titration using the NaOH volume required toAtty. Ref. No. 0073605-001030 neutralize the HCl, released from the reaction between NH2OH⋅HCl and aldehyde groups (a representative curve is shown in FIG. 5).

[0146] The DAMC is then reacted with GT in which the condensation of primary aminegroups with DAMC carbonyl groups occurs, yielding quaternary ammonium-modified cellulose (QAMC) fibrils. CHCNC are obtained by disintegrating the QAMC fibrils via heating and sonication at 60℃ for 10 min, followed by IPA-mediated precipitation. FIG. 6 shows a representative conductometric titration curve to obtain the quaternary ammonium group (N+(CH3)3) content (2.0 ±0.2 mmol g-1) of CHCNC. Assuming that all the quaternary ammonium groups of CHCNC ware yielded by the reaction between the primary amine group of GT and the DAMC aldehyde groups (5.9 ±0.2 mmol g-1), the CHCNC aldehyde content is estimated to be ~3.9 ±0.2 mmol g-1.

[0147] FIG. 7 presents an AFM image of CHCNC, which are needle-like nanoparticles,similar to the conventional strong acid-yielded cellulose nanocrystals (CNC). Although the hairs of CHCNC are not discernible in the AFM image, evidence for hairs in HCNC has been confirmed using techniques such as DLS and acoustic attenuation spectroscopy. The length and width of nanocrystals are 143 ± 46 nm and 3 ± 1 nm, respectively. Additionally, the hydrodynamic equivalent size of CHCNC, analyzed using the DLS spectroscopy, is 109 ± 2 nm. Since the particles are assumed as spheres in the DLS size measurement, the CHCNC hydrodynamic equivalent size falls within the range of nanocrystal length and width, measured using the AFM images.

[0148] To identify the functional groups of CHCNC, FIG. 8 presents the ATR-FTIRspectra of DAMC and CHCNC. In the spectrum of DAMC, the peak at 882 cm-1is assigned toAtty. Ref. No. 0073605-001030 C-O in the hemiacetal linkage, resulted from the interaction of aldehyde groups with the neighboring hydroxyl (O-H) groups. The small peak at ~ 1730 cm-1and broad peak at ~ 3313 correspond to the carbonyl (C=O) stretching vibrations of aldehyde groups and O-H stretching vibrations associated with the inter- or intra-molecular hydrogen bonds in the cellulose structure, respectively. For the CHCNC spectrum, the peaks at 3265 cm-1, 1689 cm-1, and 1476 cm-1are attributed to the stretching vibrations of O-H, C=O, and C-H of methyl groups in the quaternary ammonium of GT, respectively. The new peak at 1551 cm-1may be associated with the formation of imine bonds (C=N), confirming the successful substitution of quaternary ammonium groups, or N� H bending vibrations in the GT. The peak at 927 cm-1is assigned to nitrogen-nitrogen (N- N) bond in the GT.

[0149] To assess the CHCNC efficacy in recovering PM from aqueous media, they aredispersed (dosage = 500 ppm) in Au or Pd chloride solutions. FIG. 9 schematically presents proposed interactions of AuCl4- and PdCl42-with CHCNC. Au and Pd form negatively charged chloride complexes at an acidic pH (~ 2), thus we hypothesize that the positively charged CHCNC electrostatically bind to them, recovering Au via reduction (i.e., Au nanoparticle formation) and yielding Pd-CHCNC aggregates that do not phase separate if the CHCNC charge is not neutralized.

[0150] Results – CHCNC-mediated Au recovery: Au ions are in the form of a negativelycharged chloride complex at pH ~ 2, thus CHCNC readily binds them. At higher pH, as shown in FIG. 10, ~ 20 wt% and > 46 wt% of Au precipitate at pH = 4 and 6, respectively. At pH > 3, Au can be in the forms of AuCl4-, AuCl3(OH)-, and AuCl2(OH)2-. Thus, pH = 2 is suitable for CHCNC recovery experiments as no pH-mediated metal precipitation occurs. FIG. 11 presents the Au recovery capacity (qe,Au) of CHCNC versus contact (incubation) time at varying initial AuAtty. Ref. No. 0073605-001030 concentrations (C0,Au). The AuCl4-recovery by CHCNC (500 ppm) reaches equilibrium instantly regardless of C0,Au. The maximum qe,Auis ~ 398 ± 6 mg g-1at C0,Au = 200 ppm, which increases to ~ 764 ± 6 mg g-1at C0,Au= 400 ppm and to ~ 752 ± 166 mg g-1at C0,Au= 800 ppm in ~ 5 min. Accordingly, the contact time is fixed at 5 min (excluding the centrifugation time) for all the Au recovery experiments. FIGS. 12 and 13 show the RAuand qe,Auof CHCNC (500 ppm) at C0,Au= 10 – 800 ppm, respectively. At C0,Au< 200 ppm, RAu≥ 94 %. By increasing C0,Aufrom 200 ppm to 800 ppm (contact time = 5 min), RAudecreases to ~ 49.1 ± 0.3 % (FIG. 12). The complete recoveryof Au at low concentrations implies that reduction is the main recovery mechanism instead of AuCl4--mediated CHCNC aggregation and precipitation. Stoichiometrically, 197 ppm of Au fully neutralizes 500 ppm of CHCNC with a charge content of 2 mmol g-1, based on the 1:1 molar ratio of quaternary ammonium groups on the CHCNC to AuCl4-. AuCl4- is first adsorbed to the quaternary ammonium groups of CHCNC via electrostatic attraction, followed by interactions with CHCNC hydroxyl (–OH) groups and reduction to Au+or metallic Au0. Nanoparticle-based ion removal is typically challenging at low adsorbate-to- adsorbent ratios because stable colloids do not undergo phase separation / precipitation. The reduction process leads to the precipitation of Au from the solution, enabling effective Au recovery even in dilute solutions. Additionally, the maximum qe,Au of CHCNC is 772 ±21 mg g-1(FIG. 13), which is about 2 times higher than the theoretical recovery capacity (394 mg g-1, based on the electrostatic interactions), suggesting the involvement of a reduction process.

[0151] CHCNC colloidal behavior during AuCl4- adsorption is investigated viameasuring the hydrodynamic equivalent size of CHCNC-Au aggregates at varying C0,Auand contact times. FIG. 14 shows the scattered light intensity versus hydrodynamic equivalent size ofAtty. Ref. No. 0073605-001030 CHCNC or CHCNC-Au aggregates at C0,Auvarying from 0 to 800 ppm after 2 h of contact. CHCNC size distribution is monomodal. At C0,Au=10 ppm or 50 ppm, size distribution became bimodal with a small peak at 11 ±3 nm, associated with Au nanoparticles, and a large peak at 93 ±11 nm or 105 ±12 nm, respectively, corresponding to CHCNC-Au aggregates. The peak at 106 ±16 nm for CHCNC shifts to lower hydrodynamic sizes at C0,Au=10 ppm or 50 ppm. This shift occurred as a result of CHCNC hair shrinkage, following charge screening by AuCl4-. At higher C0,Au, e.g., 200 ppm, 600 ppm, and 800 ppm, size distributions are monomodal, which may be attributed to the formation of Au and / or CHCNC-Au aggregates. FIG. 15 shows the changes in CHCNC-Au hydrodynamic equivalent size over time. After 10 min of incubation, the hydrodynamic size at C0,Au=200 ppm, 400 ppm, and 800 ppm is 32 ±1 nm, 43 ±2 nm, and 68 ±1 nm, respectively, which increase by increasing contact time to 360 min, indicating a time- independent trend.

[0152] To examine the effect of ionic strength on qe,Au and recovery time, FIG. 16presents qe,Auversus contact time at C0,Au=200 ppm or 400 ppm in a NaCl solution (concentration =3200 ppm). The CHCNC- mediated AuCl4- adsorption is time-dependent regardless of C0,Au, requiring ~ 2 h to reach a plateau at ~ 71 % (C0,Au=200 ppm) and ~ 66 % (C0,Au=400 ppm) of NaCl-free recovery capacity (FIG. 11). This may be explained by the introduction of negatively charged ions (Cl-) with the addition of NaCl, screening the charge of quaternary ammonium groups and causing the hairs on the CHCNC to shrink. Additionally, since the CHCNC reaction sites may become less accessible as a result of hair shrinkage, the reaction time between AuCl4- and CHCNC may increase. FIG. 17 presents the hydrodynamic equivalent size of CHCNC at varying NaCl concentrations. Following a similar trend to AuCl4- effect on CHCNC, the hydrodynamic size decreases from 133 ±7 nm to 42 ±4 nm when the NaClAtty. Ref. No. 0073605-001030 concentration increases from 0 ppm to 800 ppm and then increases to 104 ±4 nm by increasing the NaCl concentration to 3200 ppm. The initial decrease in the hydrodynamic size is a result of hair shrinkage (contraction), while the subsequent increase is likely associated with CHCNC colloidal aggregation stemming from the Cl--mediated charge screening and reduced electrostatic repulsion. FIG. 18 presents a schematic of CHCNC interactions with varying concentrations of AuCl4-. The AuCl4- induces hair shrinkage via charge screening and subsequently colloidal aggregation as its concentration is further increased. Note that this scheme does not show the CHCNC-mediated Au reduction.

[0153] Results – CHCNC-mediated Pd recovery: Tetrachloropalladate ions are in theform of chloride complexes (PdCl42-) at pH =2, thus positively charged CHCNC are tested for Pd recovery. FIG. 19 presents the percentage of precipitated Pd at varying pH, showing that pH =2 is suitable to conduct the CHCNC-mediated ion recovery experiments to avoid pH-induced precipitation. FIG. 20 shows qe,Pdof CHCNC versus contact time at varying initial Pd concentrations (C0,Pd). The Pd recovery capacity of CHCNC reaches a plateau after ~ 120 min at C0,Pd=200–800 ppm, implying that CHCNC-mediated Pd recovery is regulated not only by the electrostatic interactions but also by other mechanisms, such as PdCl42-self-aggregation. Accordingly, the contact time is maintained at 120 min for the rest of Pd recovery experiments.

[0154] FIGS. 21 and 22 present RPd and qe,Pd of CHCNC (500 ppm) at C0,Pd varying from10 ppm to 800 ppm, respectively. As shown in Fig. 3b, at C0,Pd<200 ppm, RPdis only ~ 10 %. When the CHCNC are partially saturated with PdCl42-, the CHCNC- PdCl42-aggregates with a sufficiently high charge content remain colloidally stable in the solution and do not undergo significant phase separation. By increasing C0,Pdfrom 10 ppm to 400 ppm, RPdincreases from 11 ±1 % to 61 ±1 %; however, by further increasing C0,Pdfrom 400 ppm to 800 ppm, RPddecreasesAtty. Ref. No. 0073605-001030 to 33 %. Assuming the Pd recovery is mediated by electrostatic interactions, the decrease in RPdat C0,Pd>400 ppm may be a result of CHCNC active site saturation, i.e., quaternary ammonium groups, leaving the excess PdCl42-in the solution. The stoichiometric Pd recovery capacity of CHCNC based on the molar ratio of N+(CH3)3on the CHCNC PdCl42-(2:1) in the solution is ~ 106 mg g-1(quaternary ammonium content =2 mmol g-1, binding 1 mmol of PdCl42-per gram of CHCNC). FIG. 22 shows that the maximum qe,Pdof CHCNC is 559 ±36 mg g-1, which is about 5 times higher than the theoretical recovery capacity (106 mg g-1). The discrepancy between experimental and theoretical recovery capacity may be associated with the contribution of other recovery mechanisms and PdCl42-self-aggregation. The PdCl42-aggregation in aqueous solutions has been reported, where halogen electron pairs undergo charge transfer to a Group 10 metal atom.

[0155] To examine the colloidal behavior of CHCNC during the Pd recovery process, thehydrodynamic size of CHCNC- PdCl42-aggregates at varying contact times and C0,Pdare measured. FIG.23 shows the hydrodynamic equivalent size of CHCNC at varying C0,Pd. At C0,Pd<200 ppm, the particle size is in the range of 65–75 nm, which is less than the CHCNC size (109 ±2 nm, C0,Pd=0 ppm), possibly as a result of ion-mediated cellulose hair shrinkage. By increasing C0,Pdfrom 200 ppm to 800 ppm, the hydrodynamic size increases from 67 ±5 nm to 513 ±28 nm, implying the formation of colloidal CHCNC- PdCl42-aggregates. Stoichiometrically, the quaternary ammonium groups of 500 ppm CHCNC are neutralized by 53 ppm of PdCl42-(2 mmol g-1N+(CH3)3×1 / 2 ×0.0005 mg L-1CHCNC ×106.42 mg mmol-1=0.053 mg mL-1=53 ppm). Therefore, the suprastoichiometric Pd recovery is likely associated with the self-aggregated PdCl42-, resulting in a significantly higher qe,Pd(~ 5 times) comparedAtty. Ref. No. 0073605-001030 with the theoretical recovery capacity based on individual ions. FIG. 24 presents the effect of contact time on the hydrodynamic equivalent size of CHCNC at varying C0,Pd. After 6 h of contact, the CHCNC hydrodynamic equivalent size changes from 133 ±5 nm to 126 ±5 nm at C0,Pd=200 ppm, to 403 ±26 nm at C0,Pd=400 ppm, and to 2006 ±101 nm at C0,Pd=800 ppm. The significant changes in CHCNC hydrodynamic size at C0,Pd=400 and 800 ppm show time- dependent CHCNC- PdCl42-colloidal aggregation, which is likely because of the increased ionic strength in the solution and shielding the electrostatic repulsion between CHCNC particles. FIG. 25 shows the schematic of CHCNC and their hairs interacting with varying concentrations of PdCl42-, wherein the PdCl42-induce hair shrinkage via neutralizing quaternary ammonium groups and reducing the static repulsion, leading to colloidal aggregations.

[0156] Results – PM recovery mechanisms: To further investigate the underlyingmechanisms of CHCNC- mediated Au and Pd recovery, the CHCNC-Au and CHCNC- PdCl42-aggregates are characterized via TEM, SEM-EDS mapping, XPS, and XRD. All characterizations in this section are conducted on samples prepared via mixing CHCNC (dosage =500 ppm) with AuCl4- or PdCl42-solutions (final concentration of each metal ions =800 ppm). Given that the quaternary ammonium groups of 500 ppm CHCNC are neutralized by 197 ppm of Au or 53 ppm of Pd, relying on the electrostatic interactions, the molar ratio of CHCNC (500 ppm):metal (800 ppm) is stoichiometric, resulting in an excess of ions within the system. To examine the morphology of CHCNC-Au and CHCNC- PdCl42-aggregates, FIG. 26 shows the TEM images of aggregates in precipitates and supernatants. TEM images of Au precipitates (FIG. 26) present that the AuCl4-adsorption by CHCNC leads to the formation of flower-like nanostructures. We hypothesize that CHCNC simultaneously adsorb the ions and provide nucleation sites, reducing and stabilizing Au nanoparticles.Atty. Ref. No. 0073605-001030

[0157] Since the Au is in the form of anionic chloride complexes at pH = 2, it is readilyadsorbed by CHCNC via electrostatic interactions. As a result, Au3+reduces to Au+and / or metallic Au0by the neighboring –OH groups of CHCNC and precipitates, rendering the active sites available for excess AuCl4-.

[0158] In the TEM image of Au supernatant (FIG. 26), rod-like CHCNC with length of143 ± 46 nm are observed, confirming that CHCNC exist in the supernatant. In the TEM image of Pd precipitate (FIG. 27), Pd nanoparticles are observed, suggesting the precipitation of CHCNC PdCl42-aggregates or self-aggregated PdCl42-. The TEM image of supernatant (FIG. 27) shows Pd nanoparticles upon interacting with CHCNC.

[0159] To investigate the chemical states and distributions of Au and Pd, the precipitatesand supernatants of CHCNC-Au and CHCNC- PdCl42-mixtures are characterized by XPS and SEM-EDS mapping, respectively. FIG. 28 presents the XPS spectra of HAuCl4 and K2PdCl4(dissolved in ultrapure water, adjusted pH to 2, and dried) without CHCNC, as controls. All the

[0160] Au species of HAuCl4 are in the form of either Au3+or Au+, ~ 86 % of Pd speciesof K2PdCl4are in the form of PdCl42-, and ~ 14 % of Pd is in the metallic form, Pd0. The presence of metallic Pd0may be a result of partial reduction of K2PdCl4during the sample preparation (e.g., exposure to the light or heat).

[0161] FIGS. 29-32 presents the XPS spectra, including the curve fittings of Au 4f andPd 3d peaks for CHCNC-Au and CHCNC- PdCl42-aggregates in the supernatants or precipitates, along with the respective SEM-EDS mapping. As observed in the Au 4f XPS spectrum of CHCNC-Au precipitates (FIG. 29), more than ~ 84 % of Au precipitates as the metallic Au0and ~ 16 % as the ionic (Au+) form, implying significant CHCNC-mediated Au ion reduction to Au0. FIG. 29 presents SEM-EDS results of CHCNC-Au precipitates. The EDS-derived weightAtty. Ref. No. 0073605-001030 percentages of carbon (C) and oxygen (O) atoms are ~ 54 % and ~ 22 %, respectively, showing that CHCNC is found in the precipitate as a result of binding AuCl4- (Au weight percentage ~ 16 %). The EDS of CHCNC-Au precipitate (inset of FIG. 29) shows the Au distribution, indicating Au aggregate formation. In FIG. 30, the XPS spectrum of CHCNC-Au supernatant shows the absence of metallic Au0, while indicating Au+~ 42 % and Au3+~ 58 %. FIG. 30 presents the SEM-EDS results of Au supernatant, showing ~ 18 % Cl, suggesting excess AuCl4- in the supernatants, which is not recovered by CHCNC. Note that a fraction of Cl may be related to the HCl used for the pH adjustment of mixtures to 2. The lower weight percentage of C (~ 30 %) and O (~ 15 %) in the supernatant compared with those in the precipitate may also indicate that some CHCNC participate as a result of electrostatic attraction and reduction reaction. The EDS of CHCNC-Au supernatant shows a near-uniform distribution of Au species as significantly smaller particles than those in the precipitate. In FIG. 31, the XPS spectrum of CHCNC- PdCl42-precipitate shows that more than ~ 85 % of Pd is in the form of chloride complexes. Additionally, PdO ~ 15 % is detected, which may be a result of Pd oxidation. FIG.31 presents SEM-EDS results of CHCNC- PdCl42-precipitate. The C and Cl weight percentage is ~ 40 % and ~ 22 %, respectively, implying that CHCNC and Pd chloride complexes precipitate. The EDS shows the distribution of Pd in magenta. As observed in FIG. 32, the XPS spectrum of CHCNC- PdCl42-supernatant has PdCl42-~ 91 % and PdO ~ 9 %. FIG. 32 presents the SEM-EDS of CHCNC- PdCl42-supernatant. The EDS-derived weight percentage of C atoms is the highest (~ 40 %), followed by Cl (~ 25 %), implying that a fraction of CHCNC- PdCl42-aggregates remains colloidally stable in the supernatant without undergoing precipitation.

[0162] FIGS. 33-36 presents the XRD patterns of CHCNC, CHCNC-Au (onlyprecipitate), and CHCNC- PdCl42-(supernatant and precipitate). For CHCNC, the typicalAtty. Ref. No. 0073605-001030 cellulose peak for the (200) crystal plane is observed at 2θ ~ 22.6°. For the CHCNC-Au precipitate, peaks at 38.2°, 44.5°, 64.7°, and 77.8° correspond to Au crystal planes of (111), (200), (220), and (311), respectively. These observations confirm that Au nanoparticles exist in the Au-CHCNC precipitates. For CHCNC- PdCl42-cipitate, the broad peak at 2θ ~ 21.8° may be attributed to the CHCNC, confirming that the CHCNC- PdCl42-aggregates exist in the precipitate. The sharp peaks in the XRD patterns of CHCNC- PdCl42-precipitate and supernatant may be attributed to Pd crystal planes. Similar diffraction patterns have been observed for polycrystalline K2PdCl4, mechanically pre-treated in air and acetylene. Overall, the XRD findings show the successful separation of Au particles and PdCl42-using CHCNC.

[0163] Results – MINC+-mediated cyclic Pd adsorption-desorption: To overcome thecolloidal limitations of Pd recovery using unsaturated CHCNC at sub-stoichiometric ratios, CHCNC are immobilized on a model substrate (MCC) via our MINC technology. FIG. 37 schematically shows the electrostatic interactions between CHCNC and PdCl42-, resulting in stable CHCNC- PdCl42-aggregates in the solution when C0,Pd<53 ppm. FIG. 38 presents a schematic of MINC+ preparation via PDA-mediated anchorage of CHCNC to MCC, converting the nanoscale CHCNC to a microscale, phase-separated adsorbent. As shown in FIG. 39, CHCNC aldehyde groups react with DA primary amine groups via a Schiff base reaction, attaching the CHCNC to MCC surface as a result of DA polymerization. We hypothesize that the binding of CHCNC to MCC does not compromise the cationic groups as it only involves the aldehyde groups of CHCNC. To test this, conductometric titration of MINC+, shown in FIG. 40, is conducted using AgNO3to quantify the CHCNC quaternary ammonium group content. The titration results show that there is 1.1 ±0.2 mmol of quaternary ammonium groups per gram of MINC+, indicating that positive charges are partially preserved after grafting CHCNC ontoAtty. Ref. No. 0073605-001030 MCC (the charge content of free CHCNC =2.0 ±0.2 mmol g-1). This reduction may be due to the CHCNC hairs being trapped inside the PDA layer, making them partially inaccessible.

[0164] FIG. 41 presents the optical microscopy images of MCC and MINC+, whichshow that the dimensions of these particles are at the microscale, enabling phase separation. FIG. 42 presents the RPdof CHCNC, MCC, and MINC+ at C0,Pd=10 or 200 ppm. Compared withCHCNC with RPd ≤11.7 ±0.4 % at C0,pd =10 ppm and RPd ≤46 ±11 % at C0,Pd =200 ppm, theMINC+-mediated Pd recovery reached RPd~ 88.2 ±0.4 % at C0,Pd=10 ppm and RPd~ 91.1 ± 0.8 % at C0,Pd=10 ppm. This is because of CHCNC immobilization on a phase-separated substrate. Importantly, the MCC (control) does not significantly recover the ions.

[0165] To desorb PdCl42- from MINC+, EDTA is used for forming coordination bondswith the MINC+-adsorbed PdCl42-and releasing the Pd-EDTA complex, thus enabling Pd recovery. FIG. 43 show the cyclic adsorption–desorption of PdCl42-at C0,Pd=10 ppm and 200 ppm using MINC+, respectively. As observed in FIG. 43, RPd~ 88.2 ±0.4 % and Pd desorption percentage (DPd) ~ 62 ±5 % in cycle 1 at C0,Pd=10 ppm, reaching 83 ±5 % and 59 ±7 %, respectively, at cycle 5. At C0,Pd=200 ppm (FIG. 43), RPddrops from ~ 91.0 ±0.1 % to ~ 22 ±3 % from cycle 1 to 5, and DPdincreases from 67 ±7 % in cycle 1 to ~ 71 ±7 % in cycle 5. About 30 % of Pd is not desorbed from MINC+ using EDTA, possibly because of self-assembled PdCl42-aggregates, hindering the coordination between Pd2+and EDTA. The reduction in RPdat 200 ppm may be attributed to (i) non-desorbed PdCl42-that remains at MINC+ active sites, preventing further PdCl42-adsorption and / or (ii) binding EDTA to the quaternary ammonium active sites on MINC+ during desorption via electrostatic interactions, gradually decreasing available adsorption sites on MINC+ at each cycle.Atty. Ref. No. 0073605-001030

[0166] Results – Process design for the selective PM recovery: To achieve complete Pdrecovery, CHCNC are replaced with MINC+ in the third step. FIG. 3 shows the schematic of selective PM recovery from an Ag-Au-Pd mixture (initial PM concentration is reduced to 10 ppm to better mimic WPCB leachate) via a new three-step separation process, involving NaCl, CHCNC, and MINC+. The concentration of varying species, including the ions and adsorbents, are shown in FIG. 44. The target of steps 1, 2, and 3 is to selectively recover Ag, Au, and Pd, respectively. FIG. 45 present the Kdvalues in steps 1, 2, and 3, respectively. In step 1 (FIG. 45), Ag is selectively recovered with a Kd~ 53 ± 14 mL mg-1, and the Kdof other PM is near zero. In step 2 (FIG. 45), Au is selectively recovered with a Kd~ 13 ± 1 mL mg-1, and the Kdof other PM is near zero, and in step 3 (FIG. 45), Pd is selectively recovered with a Kd~ 14 ± 4 mL mg-1, and the Kdof other PM is near zero.

[0167] To selectively recover PM from WPCB leachate without interfering with otherbase metals, we use the 3-step separation process developed in the previous section (FIG. 3). FIG. 46 show the R and Kdof NaCl mediated ion recovery in the first step, respectively. Using NaCl, 88.0 ± 0.4 % of Ag is recovered with the highest Kd, showing the selective recovery of Ag via NaCl addition. The second step involves CHCNC to selectively recover Au. FIG. 47 present the R and Kdof CHCNC mediated ion recovery in the second step, respectively. Using CHCNC, 67 ± 1 % of Au is recovered, with the highest Kd, implying the selective recovery of Au using CHCNC. The third step involves MINC+ to selectively recover Pd. FIG. 48 show the R and Kdof MINC+-mediated ion recovery in the third step, respectively. Using MINC+, 92 ± 2 % of Pd is recovered, with the highest Kd, showing the selective recovery of Pd from the leachate using MINC+. The Au remains in the leachate after the second step of recovery, which adsorbs to theAtty. Ref. No. 0073605-001030 CHCNC on the MINC+, resulting in ~ 33 ± 7 % recovery in the third step. Together, the adsorbents, i.e., CHCNC and MINC+, developed via the nanoengineering of cellulose, selectively recover Au and Pd.

[0168] To compare the performance of CHCNC and MINC+ developed in this work withother sorbents, FIG. 49 presents the corresponding Au recovery capacity and contact time reported in the literature. Although CHCNC do not have the highest Au recovery capacity, it reaches ~ 800 mg g-1in only 5 min, representing the shortest contact time among all bio-based materials (highlighted). Notably, CHCNC have a remarkable selectivity for Au in metal ion mixtures, and their recovery capacity and time are comparable to synthetic polydopamine MOF. Additionally, CHCNC outperform other cellulosic adsorbents such as polyaniline-coated acetate cellulose membranes in Au recovery capacity and time and dithiocarbamate modified cellulose in recovery time. FIG. 50 compares Pd recovery capacity and contact time of MINC+ with other bio-based and synthetic sorbents found in the literature. MINC+ has the second highest Pd recovery capacity after pyridine-based porous organic polymers within 2 h, surpassing most of the existing adsorbents for Pd. Note that there are limited reports on Pd selectivity by these adsorbents, except for pyridine-based porous organic polymers and 2,5-bispolystyrene-1,3,4- thiadiazole. Pyridine-based porous organic polymers recovered ~ 100 % of Pd with a recovery capacity of 50 mg g-1from a metal mixture (initial concentration = 5 ppm) overnight. Similarly, 2,5-bis-polystyrene-1,3,4-thiadiazole recovered ~ 100 % of Pd with a recovery capacity of 21.8 mg g-1from a metal mixture, containing Pd (109 ppm) after 4 h. Together, CHCNC-enabled adsorbents provide a sustainable platform for selective PM separation.Atty. Ref. No. 0073605-001030

[0169] It should be understood that modifications to the embodiments disclosed hereincan be made to meet a particular set of design criteria. For instance, the number of or configuration of components or parameters may be used to meet a particular objective.

[0170] It will be apparent to those skilled in the art that numerous modifications andvariations of the described examples and embodiments are possible in light of the above teachings of the disclosure. The disclosed examples and embodiments are presented for purposes of illustration only. Other alternative embodiments may include some or all of the features of the various embodiments disclosed herein. For instance, it is contemplated that a particular feature described, either individually or as part of an embodiment, can be combined with other individually described features, or parts of other embodiments. The elements and acts of the various embodiments described herein can therefore be combined to provide further embodiments.

[0171] It is the intent to cover all such modifications and alternative embodiments as maycome within the true scope of this invention, which is to be given the full breadth thereof. Additionally, the disclosure of a range of values is a disclosure of every numerical value within that range, including the end points. Thus, while certain exemplary embodiments of the apparatus and process and / or utilization and methods of making and using the same have been discussed and illustrated herein, it is to be distinctly understood that the invention is not limited thereto but may be otherwise variously embodied and practiced within the scope of the following claims.

Claims

Atty. Ref. No. 0073605-001030 What is claimed is:

1. A method for recovering precious metals from a liquid source comprising a first metal, a second metal, and a third metal, the method comprising: adding a precipitating agent to the liquid source to form a first solution comprising a first supernatant and a first precipitate, wherein the first precipitate comprises the first metal; separating the first precipitate from the first supernatant; adding a first cationic cellulose-based sorbent to the first supernatant to form a second solution comprising a second supernatant and a second precipitate, wherein the second precipitate comprises the second metal; separating the second precipitate from the second supernatant; adding a second cationic cellulose-based sorbent to the second supernatant to form a third solution comprising a third supernatant and a third precipitate, wherein the third precipitate comprises the third metal; and separating the third precipitate from the third supernatant.

2. The method of claim 1, wherein the first cellulose-based sorbent comprises cationic hairy cellulose nanocrystals.

3. The method of claim 2, wherein the cationic hairy cellulose nanocrystals comprise cellulose bodies bearing cellulose chains protruding therefrom, wherein the cellulose chains are functionalized with cationic groups.Atty. Ref. No. 0073605-001030 4. The method of claim 3, wherein the cationic groups are quaternary ammonium groups.

5. The method of claim 1, wherein the second cationic cellulose-based sorbent comprises cationic hairy cellulose nanocrystals immobilized on a substrate.

6. The method of claim 5, wherein the substrate is microcrystalline cellulose.

7. The method of claim 1, wherein the precipitating agent comprises one or more compounds configured to react with the first metal via a precipitation reaction.

8. The method of claim 5, wherein the precipitating agent comprises sodium chloride.

9. The method of claim 1, wherein the first metal is silver.

10. The method of claim 1, wherein the second metal is gold.

11. The method of claim 1, wherein the third metal is palladium.

12. A method for recovering precious metals from a liquid source comprising a first metal, a second metal, and a third metal, the method comprising: adding a precipitating agent to the liquid source to form a first solution comprising a first supernatant and a first precipitate, wherein the first precipitate comprises the first metal; separating the first precipitate from the first supernatant;Atty. Ref. No. 0073605-001030 adding cationic hairy cellulose nanocrystals to the first supernatant to form a second solution comprising a second supernatant and a second precipitate, wherein the second precipitate comprises the second metal; separating the second precipitate from the second supernatant; adding cationic hairy cellulose nanocrystals immobilized on a substrate to the second supernatant to form a third solution comprising a third supernatant and a third precipitate, wherein the third precipitate comprises the third metal; and separating the third precipitate from the third supernatant.

13. A method for recovering precious metals from a liquid source comprising a first metal, a second metal, and a third metal, the method comprising: directing a feed and a precipitating agent to a first metal removal unit, wherein the precipitating agent is configured to react with the feed to form a first solution comprising a first supernatant and a first precipitate, wherein the first precipitate comprises the first metal; separating the first precipitate from the first supernatant via a first filtration mechanism; adding a first cationic cellulose-based sorbent and the first supernatant to a second metal removal unit, wherein the first cationic cellulose-based sorbent is configured to react with the first supernatant to form a second solution comprising a second supernatant and a second precipitate, wherein the second precipitate comprises the second metal; separating the second precipitate from the second supernatant via a second filtration mechanism; adding a second cationic cellulose-based sorbent and the second supernatant to a third metal removal unit, wherein the second cationic cellulose-based sorbent is configured to reactAtty. Ref. No. 0073605-001030 with the second supernatant to form a third solution comprising a third supernatant and a third precipitate, wherein the third precipitate comprises the third metal; and separating the third precipitate from the third supernatant via a third filtration mechanism.

14. The system of claim 13, wherein the first metal removal unit comprises a centrifuge configured to precipitate the first metal from the feed.

15. The system of claim 13, wherein the second metal removal unit comprises a centrifuge configured to precipitate the second metal from the first supernatant.

16. The system of claim 13, wherein the third metal removal unit comprises a centrifuge configured to precipitate the third metal from the second supernatant.

17. The system of claim 13, wherein the second metal removal unit is a bed packed with the first cationic cellulose-based sorbent.

18. The system of claim 13, wherein the third metal removal unit is a bed packed with the second cationic cellulose-based sorbent.

19. The system of claim 13, wherein the first metal is silver.

20. The system of claim 13, wherein the second metal is gold.Atty. Ref. No. 0073605-001030 21. The system of claim 13, wherein the third metal is palladium.

Images