Patent application title: Devices and methods of copper recovery
Brackenbury Darron (Moraga, CA, US)
Chulheung Bae (San Ramon, CA, US)
Brian J. Dougherty (Emeryville, CA, US)
Stephen Harrison (Benicia, CA, US)
Stephen Harrison (Benicia, CA, US)
IPC8 Class: AC25C112FI
Class name: Electrolytic synthesis (process, composition, and method of preparing composition) preparing single metal copper produced
Publication date: 2010-04-15
Patent application number: 20100089763
Contemplated configurations and methods allow for effective removal of
copper from aqueous media by selectively concentrating copper in a first
stage, wherein competing metal ions are subjected to a redox reaction to
thereby increase selective concentration. Copper is then plated in a flow
through electrode from a relatively concentrated copper solution that is
depleted from competing non-copper metals. In preferred aspects, the
first stage provides a copper-depleted effluent that includes zinc and/or
manganese for further recovery.
1. A method of removing copper from an aqueous medium,
comprising:providing an aqueous medium comprising copper and a second
metal, wherein the copper and the second metal are in an ionic
form;performing a redox reaction in the aqueous medium under conditions
that change valence of the second metal;removing the second
metal;selectively enriching copper relative to the second metal to
thereby produce a copper enriched solution; andelectrolytically
depositing copper on a flow-through electrode.
2. The method of claim 1 wherein the second metal is ferrous iron (Fe-II), and wherein the redox reaction is an oxidation.
3. The method of claim 2 wherein the aqueous medium further comprises at least one of a zinc ion and a manganese ion.
4. The method of claim 1 wherein the redox reaction comprises oxidation with O2 of the second metal.
5. The method of claim 4 wherein the oxidation is performed to achieve a ferric iron concentration of at least 90% of total iron ionic species.
6. The method of claim 4 wherein the oxidation is performed under conditions that do not change the valence of the copper.
7. The method of claim 1 wherein the step of selectively enriching copper is performed using an ion exchange resin.
8. The method of claim 7 wherein the copper enriched solution comprises copper at a concentration of at least 3000 ppm.
9. The method of claim 1 wherein the flow through electrode comprises graphite felt.
10. The method of claim 1 wherein the step of depositing the copper produces a copper depleted catholyte that is used in the step of selectively enriching copper.
11. The method of claim 2 wherein the step of enriching further produces a copper-depleted solution, and subjecting the copper-depleted solution to an ion exchange step to thereby isolate at least one of a zinc ion and a manganese ion.
12. A remediation system comprising:a redox reactor that is configured to receive an aqueous medium comprising copper and a second metal, wherein the copper and the second metal are in an ionic form;wherein the redox reactor is further configured to perform a redox reaction in the aqueous medium under conditions that change valence of the second metal;a removal system that is configured to remove the second metal;a concentration system that is fluidly coupled to the redox reactor and that is further configured to selectively enrich copper relative to the second metal to thereby produce a copper enriched solution; anda first electrochemical cell that is fluidly coupled to the concentration system and that is further configured to include a flow-through electrode onto which copper is platable.
13. The system of claim 12 wherein the copper ion is a cupric ion, wherein the second metal is ferric iron (Fe-III), wherein the aqueous medium further comprises ferrous iron (Fe-II), and optionally further comprises at least one of a zinc ion and a manganese ion, and wherein the redox reaction is an oxidation.
14. The system of claim 13 wherein the redox reactor is further configured to oxidize at least 90% of the ferrous iron to ferric iron.
15. The system of claim 12 wherein the concentration system comprises an ion exchange resin.
16. The system of claim 12 wherein the concentration system is configured to provide a copper depleted eluent that is enriched in zinc.
17. The system of claim 12 wherein the flow through electrode comprises a carbon felt.
This application is a continuation-in-part of our International
patent application with the serial number PCT/US06/37747, which was filed
Sep. 26, 2007.
FIELD OF THE INVENTION
The field of the invention is remediation of aqueous media, especially remediation of water contaminated with copper and optionally other metals (particularly iron).
BACKGROUND OF THE INVENTION
Metal recovery from water, and especially copper recovery from acid mine drainage is often technically complex and/or economically challenging as copper is frequently present in relatively small quantities and numerous other non-copper metals interfere with its isolation. Consequently, vast quantities of copper contaminated water have accumulated over decades and remain untreated.
For example, the Berkeley open pit copper mine ceased operations in 1982 and since then, groundwater from the surrounding basin leaked into the pit. As the water passed through metal deposits, the sulfuric acid content of the water increased and leached heavy metals and other contaminants from the earth, accumulating to over 17 billion gallons of wastewater. Among other contaminants, arsenic, cadmium, and zinc are present in substantial quantities, and copper was measured to be above 180 ppm. Not surprisingly, such wastewater is now considered a resource. Conceptually, copper can be recovered from solution by precipitation, adsorption, or electrowinning.
Copper precipitation is typically relatively simple, and several processes are known in the art. As most of such processes rely on hydroxide formation, enormous quantities of base are required where the wastewater is acidic. Moreover, as copper is often present in relatively dilute form in such wastewaters, mixing and separation of the precipitate are impractical in most cases. Other known processes, for example, employ ferric and ferrous iron as reagents to promote formation of copper ferrite precipitate (see e.g., U.S. Pat. No. 6,238,571). While such process tend to overcome at least some of the disadvantages of purely pH based copper precipitation, large scale applications will still often be impractical.
Alternatively, copper can also be bound or complexed using adsorbent media, and there are many adsorbent media known in the art. For example, copper can be complexed with sulfides, or organic chelating agents (e.g., EDTA, various thiocarbamates) to either form a precipitate or to be adsorbed by a secondary sorbent (e.g., activated charcoal). However, such methods often require additional treatments and will therefore not be economically attractive in many circumstances. Polymeric adsorbents typically avoid difficulties otherwise associated with complexing agents as they may act as a sold phase carrier or adsorbing filter. Most commonly, the polymeric materials are polyethyleneimine-based and may or may not include functional groups as described in U.S. Pat. Nos. 4,741,831 and 5,387,365. However, recovery of copper from such materials is often problematic and requires additional steps and/or materials. To overcome such difficulties, certain polymeric materials may be employed to adsorb and desorb copper as described in U.S. Pat. App. No. 2005/0173349. Alternatively, an ion exchange resin may be used that binds copper and from which copper is electrolytically desorbed as described in U.S. Pat. No. 5,482,632. While such adsorption is conceptually simple, difficulties are often encountered with binding of non-copper metals, decomposition of organic material and redox processes during electroelution.
In still other known approaches, electrochemical recovery of copper is achieved by plating elemental copper from an aqueous solution. However, plating efficiency is critically dependent on the copper concentration and is therefore not practicable in most configurations. To improve plating efficiencies, spouted bed cathodes can be employed as described by Spiegel et at in U.S. Pat. No. 6,298,996. To even further increase the cathode surface area, carbon felt cathodes were employed for copper plating in specific configurations as taught by Sunderland in U.S. Pat. No. 5,690,806. While such cathodes typically increase current efficiency, numerous difficulties nevertheless remain. Among other things, Sunderland's configurations are limited to copper concentrations in wastewater of less than 50 ppm. Larson describes in U.S. Pat. App. No. 2004/0168909 devices and methods for flow through carbon felt electrodes for metal recovery from ore processing units, and especially gold recovery where carbon felt electrodes are used to form the lixivants. Under certain conditions, metals are also deposited in the negative electrode. However, these and other known processes fail to remedy problems associated with competing electrochemical reactions of non-target metals.
Thus, while numerous compositions and methods for devices and methods of copper recovery from aqueous media are known in the art, all or almost all of them suffer from one or more disadvantages. Therefore, there is still a need for improved devices and methods of copper recovery.
SUMMARY OF THE INVENTION
The present invention is directed to configurations and methods for metal recovery, and especially for copper recovery from aqueous solutions where other non-copper metals are also present (most typically, non-copper metals include ferric and ferrous iron, zinc, and/or manganese). In especially preferred aspects, the copper-containing solution is first subjected to a redox reaction to reduce presence of metals with valences that would interfere with enrichment and/or plating. Copper is then selectively concentrated from the treated medium to produce a copper enriched solution from which copper is plated onto a flow through electrode. While in some aspects of the inventive subject matter the redox reaction is a reduction (e.g., ferric iron to ferrous iron), the redox reaction may also be an oxidation reaction (e.g., ferrous iron to ferric iron where zinc recovery is also desired).
Among other advantages, contemplated configurations and methods allow for selective copper enrichment from low to moderate concentrations without significant interference from non-copper metals. Thus, high current efficiencies and high copper purity are achieved while producing a copper depleted solution that may be further used for recovery of remaining metals (e.g., zinc and manganese).
In one aspect of the inventive subject matter, a method of removing copper from an aqueous medium includes a step of providing an aqueous medium comprising copper and a second metal, wherein the copper and the second metal are in an ionic form. In another step, a redox reaction is performed in the aqueous medium under conditions that change valence of the second metal, and in yet another step, copper is selectively enriched relative to the second metal to thereby produce a copper enriched solution from which copper is electrolytically deposited on a flow-through electrode.
Most typically, the second metal is ferric iron (Fe-III), and the aqueous medium also includes ferrous iron (Fe-II), and optionally at least one of a zinc ion and a manganese ion. Preferably, the redox reaction comprises electrolytic reduction of the second metal, and the reduction is performed to achieve a ferric iron concentration of less than 30% of total iron ionic species. It is also preferred that the reduction is performed under conditions that do not change the valence of the copper. In other preferred embodiments, the redox reaction comprises non-electrolytic oxidation of ferrous to ferric iron under conditions that do not change the valence of the copper. In still further preferred aspects, an ion exchange resin is used to selectively enrich the copper to produce a copper enriched solution with copper at a concentration of at least 3000 ppm and a copper depleted pass fraction. From the so copper depleted pass fraction, zinc can then be removed in a resin that preferentially chelates or otherwise binds zinc. Especially suitable flow through electrodes will comprise graphite felt, and the step of depositing the copper produces a copper depleted catholyte that is used in the step of selectively enriching copper.
In another aspect of the inventive subject matter, a remediation system will include a redox reactor configured to receive an aqueous medium comprising copper and a second metal, wherein the copper and the second metal are in an ionic form, and wherein the redox reactor is further configured to perform a redox reaction in the aqueous medium under conditions that change valence of the second metal. A concentration system is fluidly coupled to the redox reactor and configured to selectively enrich copper relative to the second metal to thereby produce a copper enriched solution, and a first electrochemical cell is fluidly coupled to the concentration system, wherein the cell is further configured to include a flow-through electrode onto which copper is platable.
Most typically, the copper ion is a cupric ion, the second metal is ferric iron (Fe-III), and the aqueous medium may further comprise ferrous iron (Fe-II), and optionally comprise at least one of a zinc ion and a manganese ion. Similar to the method contemplated above, the redox reactor preferably comprises a second electrochemical cell that may advantageously be configured to reduce ferric iron to ferrous iron without reducing the cupric ion to elemental copper. Alternatively, the second cell may also be an electrochemical or non-electrochemical reactor in which oxidation of ferrous to ferric iron is performed. In further preferred aspects, the concentration system comprises an ion exchange resin, wherein the exchange resin preferentially binds ferric iron relative to ferrous iron. In most preferred aspects, the concentration system is configured to provide a copper depleted eluent that is enriched in the second metal, and/or the redox reactor is configured to increase selectivity in the concentration system. Where desirable, the first electrochemical cell may be configured to provide a copper depleted eluent to the concentration system.
Various objects, features, aspects and advantages of the present invention will become more apparent from the following detailed description of preferred embodiments of the invention.
BRIEF DESCRIPTION OF THE DRAWINGS
FIG. 1 is a schematic illustration of an exemplary copper removal process according to the inventive subject matter.
FIG. 2 is a schematic illustration of an exemplary electrochemical cell for copper plating according to the inventive subject matter.
FIG. 3 is a graph showing the binding capacity of an exemplary resin for selected metal ions.
FIG. 4 is a graph showing the cell voltage and Fe3+ and Cu2+ concentration profiles during copper plating.
FIG. 5 is a graph depicting overall current efficiency in copper plating according to the inventive subject matter.
FIG. 6 is a graph depicting Fe3+ concentration in the initial samples over consecutive cycles of plating.
FIG. 7 is a graph depicting energy consumption to recover 1 kg of copper at selected Cu2+ concentrations over consecutive cycles of plating.
FIG. 8 is a graph depicting required electrode area for copper plating over consecutive cycles of plating.
FIG. 9 is a graph depicting zinc breakthrough on selected resins.
The inventors have unexpectedly discovered that various metals, and especially metals at low to moderate concentrations (e.g., less than 1000 ppm) can be effectively and selectively recovered from an aqueous medium in which other metals are present by using a process that employs a redox reaction to reduce enrichment and/or plating interference from other metals, that selectively enriches the target metal to form a enriched medium, and that then plates the target metal from the enriched medium in a flow through electrode.
Contemplated configurations and methods are particularly desirable for treatment of copper containing acid mine drainage solutions and has been shown to produce high quality copper in a current efficient and economical manner. Such achievement is particularly notable as mine waste solutions are often richer in iron, zinc, and sulfuric acid than the target metal copper, and may further have significant buffering capacity (e.g., due to calcium, magnesium, and aluminum sulfates).
In previously known copper recovery systems and methods, iron typically interfered significantly with various process steps, including enrichment and plating. The inventors have now discovered that such interference can be substantially reduced, if not even entirely eliminated by combining a redox reaction (in which iron is reduced to ferrous or oxidized to ferric iron) with a metal enrichment process, wherein the enrichment process is substantially insensitive to the iron in either oxidized or reduced form.
In an exemplary process described in more detail below, Berkeley Pit water served as the source of acid mine drainage water (which is known to comprise thousands of tons of copper, zinc, iron, manganese, and smaller amounts of other metals). This water is acidic at a pH of about 2.8, has a sulfate concentration of about 10,000 ppm, with a copper concentration of 178 ppm, manganese of 240 ppm, zinc of 600 ppm, and iron of 1000 ppm. It should also be appreciated that the buffering capacity of this water was determined have a combined load of about 1270 ppm of calcium, magnesium, and aluminum salts. Using known processes, it would seem highly improbable that copper could be recovered from this mixture in an economical manner. Unless otherwise specified or dictated by the context, it should be noted that the terms "copper" and "metal" as used herein refers to both the ionic species as well uncharged elements (plated element).
It should still further be appreciated that electrochemical capture of metals on the cathode of an electrochemical cell is usually inefficient. This is particularly true where the metal concentrations are relatively low (e.g., below 1000 ppm, more problematic below 500 ppm, and even more problematic below 200 ppm), and where additional metals with redox states in aqueous solution (e.g., iron) will act as shuttle species between the anode and cathode and compete with the plating process (Fe+2-e.sup.- at the anode → Fe+3; Fe+3+e.sup.- at the cathode → Fe+2). Conventional tank electrolyzers as used in the industry for electrowinning of copper/zinc will not properly operate under such circumstances and low metal concentration as the shuttle species and the anodes and cathodes are in the same electrolyte. Electrowinning from dilute mixed metal acid mine waste demands advanced reactor designs and materials to minimize impact of redox ions, poor mass transfer of target ions, propensity to electrolyze water and evolve hydrogen instead of metal deposition, and to minimize production of metal deposits that are granular, dendritic, and re-oxidized when removed from the electrolyte.
In contrast, the configurations and systems according to one aspect of the inventive subject matter will first reduce non-target metals and the concentration of metals with valence states that would interfere with concentration and/or plating of the target metal. Then, the pretreated and concentrated solution is subjected to electroplating while reject fluid can be further processed to isolate the non-target metals. FIG. 1 illustrates a schematic overlook of the devices and methods presented herein. Here, in one exemplary copper isolation process, a solution containing copper ions, ferric iron, ferrous iron, zinc ions and manganese ions is first subjected to a redox reaction to electrolytically reduce ferric iron to ferrous iron, preferably without reducing the copper ions (and other non-copper metal ions) as indicated in box 110. The so treated solution is then passed through a cation exchange column that binds copper ions preferentially over other metal ions as indicated in box 120.
It should be noted that the prior reduction step of ferric iron to ferrous iron will greatly reduce binding of iron ions to the resin as preferred resins will typically not bind ferrous iron and ferric iron only to a relatively small extent. Copper ion depleted eluent 122 may then be stored for further processing (e.g., to remove the non-copper metal ions). The resin with the bound copper ions is then treated with eluent to remove the copper ions from the resin as depicted in box 130. Consequently, the eluent is preferentially enriched in copper ions (typically at least 10-fold in concentration) while at the same time non-copper metal ions are removed into the copper depleted eluent 122. Thus, the concentration of non-copper ions in the copper ion-enriched elution will be significantly reduced (e.g., at least 5-fold, more typically at least 10-fold in concentration) and also include significantly reduced (e.g., at least 5-fold, more typically at least 10-fold in concentration) concentrations of ionic species that may interfere (e.g., as redox shuttle) with the subsequent electroplating in box 140. Electroplating is preferably performed in a flow through electrode, which most preferably includes carbon felt as cathode material. Where desired, the copper ion depleted solution after electroplating is then stored or routed for further use as eluent in the ion exchange column of box 120. The plated elemental copper can then be further refined to desired purity as depicted to in box 142.
Therefore, the inventors contemplate a method of removing copper from an aqueous medium in which in one step an aqueous medium comprising copper and a second metal is provided, wherein the copper and the second metal are in an ionic form. In another step, a redox reaction is performed in the aqueous medium under conditions that changes valence of the second metal, and in yet another step, copper is electrolytically deposited as elemental copper on a flow-through electrode. Viewed from a different perspective, the inventors also contemplate a remediation system that includes a redox reactor that is configured to receive an aqueous medium comprising copper and a second metal, wherein the copper and the second metal are in an ionic form, wherein the redox reactor is further configured to perform a redox reaction in the aqueous medium under conditions that change valence of the second metal. In such systems, a concentration system is fluidly coupled to the redox reactor and is further configured to selectively enrich copper relative to the second metal to thereby produce a copper enriched solution. Contemplated systems also include a first electrochemical cell that is fluidly coupled to the concentration system and that is further configured to include a flow-through electrode onto which copper is platable.
Alternatively, and according to another preferred aspect of the inventive subject matter, non-target metals and especially ferrous iron are oxidized and removed from the solution, preferably by precipitation, flocculation, filtration, or other process as suitable for the oxidized species. While not limiting to the inventive subject matter, the non-target metal can be oxidized by electrochemical processes, one or more redox reaction, and/or air-oxidation (e.g., sparging air or other O2-containing gas through the solution) or alternative oxidizing agent such as hydrogen peroxide or even electrochemically. Most preferably, oxidation is performed such that at least 90%, more typically at least 95%, and most typically at least 99% of all ferrous iron is converted to ferric iron.
In the example of ferrous iron being the non-target metal, oxidation will produce iron-III-oxides, which can be readily removed using conventional methods. Where needed, removal may be assisted by microfiltration to achieve single-digit ppm (or even lower) iron concentrations. Once the iron is removed from the solution, the iron-depleted solution can then be passed trough one or more ion-exchange resins with selectivity towards copper ions and other metal ions (most preferably zinc). In sequential systems, it is typically preferred that the first exchange resin has a high specificity towards copper ions (e.g., Amberlite IRT to 748I) and elutes a solution that is then passed over the second resin (e.g., Purolite S950). Thus, it should be recognized that rather than attempting to remove various species of iron or other contaminating non-target metals, iron and other contaminating non-target metals are either converted to a redox state that does not significantly interferes with subsequent concentration and/or plating, or converted to a redox state in which substantially all of the contaminating non-target metal can be removed (e.g., to less than 10 ppm, more preferably less than 1 ppm, and most preferably less than 0.1 ppm) without interfering with subsequent concentration and/or plating.
With respect to contemplated target metals (i.e., metals that are plated in contemplated configurations and methods) it should be appreciated that numerous metals other than copper are also deemed suitable and alternative metals include various noble metals (gold, silver, platinum, etc.), various heavy metals (cadmium, chromium, etc.), uranium, etc., all of which may be in various oxidation states. However, it is generally preferred that the metals have at least a single positive charge. Suitable target metals may be present as dissolved salts (i.e., as free ions), in complex with one or more ligands, or as organo-metallic compound. Therefore, the nature of contemplated aqueous solutions will vary substantially, and may include various wastewaters (e.g., from circuit board manufacture, mine drainage, landfill, etc.), process fluids (e.g., semiconductor manufacture, chemical plant, etc.), contaminated well water, and all other aqueous media that have at least one metal ion in a concentration of less than 2000 ppm, more typically of less than 1000 ppm, even more typically of less than 500 ppm, and most typically between about 100 ppb and 300 ppm.
Consequently, the nature of the second metal may vary considerably, and it should be appreciated that the second metal may be present in various oxidation states, and/or that additional metals may be present. Thus, suitable second metals include iron, chromium, zinc, molybdenum, nickel, cadmium, silver, etc. However, especially contemplated second metals include those that will act as shuttle species in a plating process of the target metal and/or that will co-enrich in an enrichment process of the target metal. Thus, especially contemplated second metals include ferric iron (Fe-III) and ferrous iron (Fe-II), which may be accompanied by zinc ion and/or a manganese ions (such metal composition is typical for various acid mine drainage solutions). Second metal ions will typically be present in total concentrations of between 1 ppm and 300 ppm, more typically between 300 ppm and 500 ppm, even more typically between 500 ppm and 1000 ppm, and most typically between 1000 ppm and 3000 ppm, and even higher.
With respect to contemplated redox reactions it is contemplated that suitable reactions include electrochemical reactions on an anode and/or cathode, and chemical reactions with a reductant and/or oxidizing agent. However, particularly preferred processes comprise electrolytic reduction of the second metal. Alternatively, or additionally, chemical reaction with a redox reagent may also be employed. For example, where copper is to be isolated in an aqueous solution that includes both ferric and ferrous iron, it is typically preferred to reduce ferric iron to ferrous iron to a final ferric iron concentration of less than 30%, more preferably less than 20%, even more preferably less than 10%, and most preferably less than 5% of total iron ionic species. Such reduction will advantageously not only significantly reduce redox shuttle of ferric iron but also allow the ferrous iron to pass through the ion exchange resin (and with that be removed) where the copper concentration step includes an ion exchange resin. On the other hand, where the redox reaction is an oxidation, it is contemplated that the final ferric concentration is at least 90%, more preferably at least 95%, and most preferably 99% of total iron ionic species. As discussed above, all suitable manners of oxidation are deemed suitable; however, especially preferred oxidations include chemical oxidation with an O2-containing gas, O3, or a reagent that liberates O2, O3, or an oxygen radical. Thus, viewed from a different perspective, the redox reactor is configured and/or operated to increase selectivity in the concentration system. Most typically, reduction or oxidation (chemical or electrolytic) will be performed under conditions that do not change the valence of the copper ion (or other target metal).
Various options of selectively enriching the target metal (e.g., copper) are known in the art. However, particularly preferred options employ an ion exchange resin that binds the target ion preferably at a higher affinity than non-target metals. In such and other processes, it is generally preferred that the target metal is enriched at least 3-fold, more preferably at least 5-fold, even more preferably at least 10-fold, and most preferably at least 20-fold in absolute concentration whereas non-target metals are enriched to lesser degree, and most preferably depleted by at least 2-fold, more preferably at least 5-fold, even more preferably at least 10-fold, and most preferably at least 20-fold in absolute concentration. In especially preferred aspects, and particularly where copper is the target metal, suitable ion exchange resins will preferentially bind ferric iron relative to ferrous iron. Thus, the reduction of ferric to ferrous iron will dramatically reduce overall iron ion concentration in the copper enriched exchange eluent.
Alternatively, or additionally, the target metal may also be selectively enriched using various known processes, including membrane filtration (which may or may not be assisted by complexation or chelation of target or non-target ions) and/or precipitation. For example, non-target metals may be removed by an increase in pH (ferric iron precipitates above pH 3.5 while ferrous iron remains dissolved). In other examples, copper may be enriched by ferrite precipitation.
Preferred enrichment processes will generate aqueous media with target metal (e.g., copper) concentrations of at least 1000 ppm, more typically at least at least 2000 ppm, even more typically at least 3000 ppm, and most typically at least 4000 ppm, while the total non-target metal concentration is preferably below 3000 ppm, more preferably below 2000 ppm, and most preferably below 1000 ppm. In further preferred aspects, it is contemplated that the concentration system is configured to provide a copper depleted eluent or other process fluid that is enriched in the second metal, wherein valuable second metals especially include zinc, chromium, molybdenum, and arsenic.
Contemplated electrochemical cells include all known electrochemical cells in which anolyte and catholyte are separated by a separator and in which copper can be plated from a copper containing solution. Therefore, suitable electrochemical cells may be closed and static systems in which electrolyte is loaded prior to reduction of the metal or open dynamic cells in which electrolyte is circulated (or recirculated) from catholyte and/or anolyte tanks. However, particularly preferred systems include open dynamic cells. It should be noted that electrolyte (re)circulation can be accomplished with pumps and other equipment well known in the art. Similarly, there are numerous electrode configurations and materials known in the art, and all configurations and materials are deemed suitable for use herein. However, particularly preferred anode configurations include flat panel electrodes or flat mesh electrodes. It is still further preferred that the anode is disposed in the anode compartment such that space between the anode and the separator defines a relatively narrow anolyte flow path. Viewed from a different perspective, it is generally preferred that the anode is as close as possible to the separator. Anode materials are preferably corrosion resistant and will include platinum coated titanium mesh, Magnelli phase titanium suboxide, etc. With respect to suitable anolytes, it should be noted that all anolytes known for electrowinning of metals are deemed suitable, and that the choice of suitable anolyte is well within the scope of a person of ordinary skill in the art.
Similarly, contemplated cathode materials are corrosion resistant to the catholyte and will provide a relatively large surface area to allow efficient copper plating at even relatively low copper concentrations. Therefore, especially contemplated cathode materials include carbon-based materials, and particularly carbon felt, glassy carbon, and carbon fiber, all of which may be partially pyrolyzed, activated, or otherwise treated. As used herein, the term "carbon felt" refers to a textile material that predominantly comprises randomly oriented and intertwined carbon fibers, which are typically fabricated by carbonization of organic felts (see e.g., IUPAC Compendium of Chemical Terminology 2nd Edition (1997)). Most typically, organic textile fibrous felts are subjected to pyrolysis at a temperature of at least 1200° K., more typically 1400° K., and most typically 1600° K. in an inert atmosphere, resulting in a carbon content of the residue 90 wt %, more typically 95 wt %, and most typically 99 wt %. Furthermore, contemplated carbon felts will have a surface area of at least about 0.01-100 m2/g, and more typically 0.1-5 m2/g, most typically 0.3-3 m2/g, and where the carbon felt is activated, will have a surface area (BET) of more than 100-500 m2/g, more typically at least about 500-800 m2/g, even more typically at least about 800-1200 m2/g, and most typically at least about 1200-1500 m2/g, or even more.
Depending on the organic textile material and carbonization conditions, the carbon felt may be graphitic, amorphous, have partial diamond structures (added or formed by carbonization), or a mixture thereof. In contrast, reticulated or vitreous (glassy) carbon is formed from carbonized thermosetting organic polymer foams that generally have a non-fibrous, open or closed cellular architecture. While not preferred as high surface area material in conjunction with the teachings presented herein, reticulated or vitreous (glassy) carbon may also be used. Most preferably, the carbon felt is prepared from carbonized organic textile fibrous felts and has a surface area of about 0.1-5 m2/g to about 1200 m2/g and even higher (where the carbon felt is activated). While the exact configuration is of the carbon felt may be variable, it is typically preferred that the carbon felt will have a thickness to allow for a flow path from one side to the other of the felt of between 0.1 cm and 10 cm, and even more preferably between 0.5 cm and 5 cm.
Especially preferred cathodes are configured such that at least 50%, more typically at least 75%, even more typically at least 90%, and most typically at least 95% of the catholyte flow from one side of the cathode through the volume of the cathode to the other side of the cathode (which is proximal to the separator). Thus, especially preferred cathodes are flow through cathodes. Current can be fed to the flow through material of the cathode in numerous manners, and all known manners are deemed suitable for use herein.
With respect to the separator, it is generally preferred that the separator maintains the anolyte separate from the catholyte and only allows charged species to cross. Preferably, the charged species are limited to proton exchange, and therefore, all known proton exchange membranes are deemed suitable for use herein (e.g., NAFION® [perfluorosulphonate cation exchange membrane commercially available from Dupont]) may be advantageously employed in contemplated electrochemical cells as such membrane has desirable resistance to the catholyte and anolyte.
In still further preferred aspects, the cell is configured and operated such that the target metal is substantially completely plated onto the cathode. For example, where the target metal is copper, it is preferred that copper is deposited in an amount of at least 90%, more typically at least 95%, and most typically at least 98% of the amount present in the catholyte before operation of the electrochemical cell. Therefore, the copper depleted catholyte may be reused as eluent in the step of selectively enriching copper. Viewed from a different perspective, the inventors contemplate an electrochemical cell that is configured to provide a copper depleted eluent to the concentration system.
In the following examples, copper was recovered from controlled aqueous solutions and Berkely pit acid mine water in a process that included a reduction step to convert ferric to ferrous iron, a selective enrichment step that increase copper concentration, and a plating step that almost completely removed any copper from the treated and concentrated solution. It should be particularly appreciated that the processes and devices presented herein allow for an economic and conceptually simple recovery of copper without interference of non-copper metals at concentrations originally significantly higher than copper.
FIG. 2 schematically depicts an exemplary electrowinning cell 200 in which the cell body is formed from HDPE plates 210 (typically high-density polyethylene or other corrosion resistant material) in which flow channels are formed to allow entry and exit of catholyte and anolyte. The cathode compartment 212 and anode compartment 214 are separated by NAFION membrane 220. In the anode compartment, anolyte flows (arrows) across the anode 230, which is coupled to spacer block 232 to reduce the space between the anode 230 and the separator 220. The anode 230 is preferably a platinum coated titanium mesh. The cathode 240 in the cathode compartment 212 is preferably a graphite felt 242 that is mounted on a titanium current feeder 244. Graphite felt 242 is preferably held in place by plastic mesh 246. It should be especially noted that the catholyte flows (arrows) into the cathode compartment 212 and from there through the graphite felt 242 before entering the space defined by the separator 220, felt 242, and catholyte entry and exit ports. Therefore, the flow of the electrolyte and the current are parallel.
Initial experiments with direct electroplating of copper from untreated Berkeley Pit water on high surface area carbon felt electrodes yielded a copper deposit that assayed at 99.9% copper. Further electro-refining of the copper off the carbon felt electrode gave an purity of 99.999% copper. The refining solution used was a typical copper refining solution (see e.g., Industrial Electrochemistry, P28, Pletcher, D. Chapman Hall 1982). While the initial results were promising with respect to the ability to selectively plate copper from untreated wastewater at reasonable efficiency and selectivity, copper deposited only on the outer surface of the felt electrode rather than throughout the felt electrode.
Based on this and other observations, the inventors decided to selectively concentrate copper in the wastewater, preferably via ion exchange. However, as copper concentration by ion exchange was expected to also enrich other non-copper metals (which typically occur in higher concentrations), resins with selectivity for copper needed to be identified. Specifically, suitable resins should provide enrichment of copper without concentration of iron, zinc, and manganese. To that end, several cation exchange resins were tested with large samples of Berkeley pit water, and selected commercially available resins were identified as binding copper with desirable affinity.
Among other feasible options, the inventors found Amberlite IRT 748I as particularly suitable as copper eluted other ions like Zn+2, Fe+2, and Mn+2. The ion exchange resin became almost completely occupied with copper. However, Amberlite IRT 748I bound Fe+3, the least abundant state of iron in Berkeley Pit waters, with significant affinity. It was thus necessary to ensure that as much iron as possible was in the reduced ferrous state prior to reaching the ion exchange resin. Among other reasons for removing ferric iron from the copper was the fact that ferric iron acts as a shuttle species in an electrochemical cell. Conceptually, various options for reducing ferric iron to ferrous iron are available: Chemical treatment (e.g., with sulfur dioxide, hydrazine sulfate, sodium metabisulfite etc.), or electrochemical reduction.
A standard lab cell with a carbon felt cathode, a coated titanium anode, and a Nafion separator was used to reduce Fe3+ from a concentration of 700 mg/l to about 250 mg/l in 4.5 L of Berkeley Pit (BP) water. Table 1 below summarizes typical results. Iron speciation was done using 1,10 Phenanthroline and UV-VIS.
TABLE-US-00001 TABLE 1 Iron Reduction Using an Electrochemical Cell TIME CURRENT SAMPLE (MIN) (A) VOLTAGE (V) FE(III)(MG/L) Fe-HH-Initial 0 0 0 708 Fe-HH-10 10 2.5 9 586 Fe-HH-20 20 2.5 8.5 470 Fe-HH-30 30 2.5 8 420 Fe-HH-40 40 2.5 7.5 250
As an alternative method, sodium metabisulfite, Na2S2O5, was used to reduce Fe3+ to Fe2+. One mole of sodium metabisulfite reduces 4 moles of Fe3+. The speciation of iron was done via titration with Ce4+. When sodium metabisulfite was used in excess, it was successful in reducing Fe3+, however when the salt was added to BP water a dark orange color was formed which inhibited its use in an ion exchange column.
It should be noted that both methods are economic as so little reduction is required to reach a point of diminishing returns and the loss of capacity of the resin due to the presence of remaining Fe+3. Thus, the electrochemical process was adopted as the most acceptable. Similarly, where the iron ions were oxidized, air sparging and subsequent filtration proved economically most attractive.
Based on the preliminary data obtained, the inventors then contemplated that Berkeley
Pit water is treated first to convert Fe+3 to Fe+2, and to feed the so treated wastewater to an ion exchange bed that strips the wastewater of copper until breakthrough occurs. The treated water is collected for zinc and manganese recovery, which can be performed using numerous manners known in the art. The copper rich solution stripped from the ion exchange column is then treated in the carbon felt cell to plate the copper. When copper reaches the prescribed thickness the electrode is removed and the copper removed mechanically from the cathode. The copper free solution is then used to further strip more copper from the next batch of ion exchange resin.
Ferric iron contents in Berkley Pit are 25% to 30% of the total 1000 ppm. This ratio had increased in the samples received due to oxidation. To simulate the conditions in the pit, BP water was reduced electrochemically to 30% Fe(III) and was used in a beaker test with IRC 748I under a nitrogen blanket to prevents oxidation. The resin captured 60% of total copper, 14% of total iron, and 8% of total zinc. More detailed results and further experiments are provided below.
Ion Exchange Capture
The first set of beaker tests were performed on DOWEX® M4195 and the following AMBERLITE® resins: IRC 747, IRC 748I, and GT73. It was found that DOWEX® M4195 and IRC 748I had higher copper capacities than GT73. AMBERLITE® resin IRC 747 did not have a high capacity for copper. FIG. 3 shows the equilibrium capacity of each resin for copper, iron, and zinc respectively (in the presence or absence of ascorbic acid as reductant). Among the resins tested, DOWEX® M4195 had the highest capacity for zinc.
An electrochemical cell as depicted in FIG. 2 was employed to plate copper from the regenerant of the IRC 748I ion exchange resin column. Pt/Ti mesh and carbon felt were used as anode and cathode, respectively. Here, the Pt/Ti mesh electrode was bent and spot welded onto a Ti plate as shown in the FIG. 2 and a HDPE plastic block was inserted between the Pt/Ti mesh and Ti plate current feeder to give less space between the anode and a separator. A 10 cm thick carbon felt cathode was threaded onto a Ti mesh current feeder together with a plastic mesh which pushed the felt electrode against the Ti mesh current feeder (resulting in better contact between the felt electrode and current feeder). The projected area of the carbon felt electrode was 11×9.5 cm2. The anodic and cathodic compartments were separated with an ion exchange membrane (Nafion® 450), and 10% H2SO4 and the regenerant were utilized as anolyte and catholyte, respectively. Both electrolytes entered at the bottom of the cell and left at the top of the cell. The catholyte flow process is in "flow-through" mode where flow of solution and current are parallel.
The catholyte was fed from the cathode surface which faced the Ti mesh current feeder towards the ion exchange membrane as seen in FIG. 2. It should be appreciated that the catholyte flow path was made in this way in order to suppress the fast copper dendrite growth on the surface of the carbon felt facing the membrane during the plating process. From the preliminary experiments with opposite catholyte flow path, it was observed that the copper dendrite grew fast on the side of the carbon felt electrode facing the membrane since that side was nearer to the positive electrode and less copper ions were sent to the other side of the felt electrode resulting in poorer copper plating along the whole depth of the carbon felt electrode. Between each cell part, Tygon gaskets were situated for better sealing.
In preferred systems, a peristaltic pump (Masterflex L/STM, Cole-Parmer Instrument company model 7552-02) was employed to circulate the electrolytes with two pump heads (Masterflex® Easy-Load) giving same flow rate to both compartments. The electrolyte flow rate was adjusted by the pump controller (Masterflex® Wash-Dow Modular controller, Cole-Parmer Instrument company model 7552-71) and it was fixed at 20 ml sec-1, which for a flow channel past the electrode of width 11 cm and height 9.5 cm corresponds to a linear flow velocity of 18.18×10-2 cm sec-1. The electrolytes were fed into the electrochemical cell from the 2 L plastic bottle reservoirs through the Tygon tubings and sent back to the reservoirs. The catholyte tubings in the cathodic reservoir were immersed into the electrolyte to prevent any Fe2+ ions in the catholyte from contacting the oxygen in the air and being oxidized to Fe3+ which can possibly oxidize the plated copper on the felt electrode as well as consume charges for its reduction back to Fe2+ causing poorer charge efficiency during the copper plating process. A pressure gauge was connected near the catholyte inlet at the bottom of the cell to monitor the pressure drop before and after the copper plating. Copper plating was performed under constant current at 30 Amp using a Maccor unit (computerized battery charger) which also monitored the cell voltage and energy consumption during the experiments.
Exemplary Recovery Process Using Berkeley Pit Water
Reduction and Ion Exchange Process
Berkeley Pit water was taken and electrochemically reduced with calculated time under a fixed current to convert Fe3+ ions to Fe2+ as described above before it was sent to the ion exchange resin column which was found not to capture the Fe2+ ions. The pretreated water was passed through the ion exchange resin (Amberlite IRC 748I, Rohm and Haas) column where Cu2+, Zn2+, and residual Fe3+ were loaded. The fully loaded column was then regenerated with 10 w.t. % H2SO4 solution. Volume (around 1100 ml) of the collected regenerant was measured and the regenerant was sent to another electrochemical cell to carry out copper plating.
Electrochemical Copper Plating
After setting up the flow system with the copper plating cell (configured as shown in FIG. 2), 1500 ml of anolyte and about 1100 ml of catholyte in each reservoir were circulated for 10 minutes, and 2 ml of the initial samples were taken from each reservoir prior to copper plating. Samples were picked up with a syringe from the middle of the reservoirs. Then the Maccor unit was turned on to start the copper plating process under the constant current density at 3000 A m-2. During plating, 2 ml of samples were taken from the cathodic reservoir at every 2-3 minutes until the color of the catholyte turned transparent (from blue green) indicating most of copper was deposited onto the carbon felt electrode. After the final sampling, the Maccor unit was turned off and the final anolyte sample was taken to monitor the amount of the diffused Cu2+ ions in the anodic reservoir.
After the copper plating, the regenerant was filtered to remove copper debris and the filtered regenerant was immediately used to regenerate the successively loaded ion exchange resin column. From 5th cycle concentrated acid was added to the regenerant prior to the next regeneration to adjust the acid concentration. After each experiment, the electrochemical cell was taken apart and all of the cell components were washed with de-ionized water. The used carbon felt electrode was replaced with a fresh felt for each copper plating experiment.
All the samples from the above plating experiments were analyzed with UV and AA spectrometers for iron speciation and Cu2+, Zn2+, and Fe3+ ion concentrations. After plating, half of each sample was mixed with 1,10-phenantholine solution which is complexed with Fe2+ giving orange color in different intensity depending on the concentration of the complex. UV spectrometer was employed to measure the concentration of the complex which indicates the concentration of Fe2+ ions in the samples. Total iron, copper and zinc concentrations were analysed with AA spectrometer. Iron speciation was performed immediately after the copper plating experiment to avoid the possible Fe2+ oxidation back to Fe3+ with the oxygen in the air. 150 ml of regenerant samples were taken before and after the 8th plating to analyze the ion contents in them and the carbon felt cathode was also collected for the analysis after the 8th plating.
Fe3+ Reduction Versus Cu2+ Deposition
Copper plating efficiencies were calculated with Cu2+ concentration change during the experiments. However, the regenerant includes Fe3+ ions which reduction to Fe2+ occurs at higher potential compared to Cu2+ reduction to Cu resulting in the competition between the two species at the beginning of the experiments.
FIG. 4 shows the cell voltage, Fe3+ and Cu2+ concentration profiles during the 1st copper plating from the regenerant. As seen in this figure, Fe3+ and Cu2+ concentration decrease for the first 3 minutes but Fe3+ concentration change gives steeper slope than Cu2+ indicating that Fe3+ reduction consumes more electrical charges than Cu deposition during that time. After fast decrease for the first 3 minutes Fe3+ concentration remains at 550 ppm for the rest of the experiment while Cu2+ concentration starts decreasing faster and the cell voltage slightly increases until 9 minutes showing that Cu plating mainly occurs during this period of time. After 9 minutes of the experiment, the Cu2+ concentration drops much slower and the cell voltage rapidly increases indicating that the hydrogen evolution commences.
Initial drop in the cell voltage (e.g., first 4 minutes) could be explained with the initial copper deposition which makes the electrode surface more conductive with copper than the original bear carbon and hence reduces IR drop, which decreases the overall cell voltage at the beginning of the plating experiments. Successive copper plating experiments have shown similar tendency. It should be noted that the Fe3+ concentration decreases rapidly for the first 3 minutes and then remains at 550 ppm which is about 26% of the initial Fe3+ concentration until the end of the experiment while most of Cu2+ ions has been converted to Cu and massive hydrogen evolution occurred at the end of the experiment. The reason why 26% of Fe3+ ions were not reduced until the end of the experiment could be explained by several possibilities: (a) Some Fe3+ ions are complexed with other ions such as Al in Berkeley Pit water and hard to be electrochemically reduced. These complexes would not affect the copper plating efficiency; (b) Organics prevents further reduction of the Fe3+; (c) Iron shuttle occurs with dissolved oxygen in the regenerant.
(II) Oxidation and Ion Exchange Process
In an alternative process, iron (II) will be oxidized with air or other O2-containing gas to iron (III), for example, by dispersing air in the aqueous solution. The iron oxide so formed is then filtered off using a combination of filtration and micro-filtration using well known manners. A dispersant may be added to the stream to aid in the filtration of the iron oxide. The liquid effluent, now essentially free of iron (II), will be fed to an ion exchange column to remove copper (in a manner as described herein) and then to an second ion exchange column to capture zinc as shown in the examples below. Zinc electrodeposition from sulfate based solutions is well known in the art and all known manners and configurations are deemed suitable for use herein. It is generally preferred that for zinc plating high surface area carbon electrodes are used to improve efficiency of electrochemical zinc recovery.
Ion Exchange Capture
Two resins were considered for zinc recovery from Berkeley Pit water. One was Lewatit VP OC 1026 which is a macroporous resin impregnated with Di-ethylhexyl-phospate (D2EHPA). The other one was Purolite S950 which is a macroporous resin with styrene-divinylbenzene as the back polymer and aminophosphonic acid functional group. FIG. 9 depicts break through curves for zinc on these two resins. The following column tests were done using these resins:
Resin OC1026: Amt Resin Used: 28 g (42 mL) in single column of 1'' diameter. Due to the nature of the resin, glass wool was placed on top of the resin bed to hold it in place. A surrogate solution was made with the following composition (to mimic BP water post copper removal): 695 mg/L Zn, 1249 mg/L Fe, 10 mg/L Cu at pH 2.5.
About 25 bed volumes (1.05 L) of the above solution was passed through the column at 11 BV/hr. The resin was then regenerated with 1M H2SO4 solution (2.5 bed volumes) at 3 BV/hr. The amount zinc up taken by resin was 7 g Zn/L of resin. Zinc breakthrough was after 5 bed volumes.
Purolite S950: Amt Resin Used: 30 g (40 mL) in single column of 1'' diameter. A surrogate solution (2 L) was made with the compositions shown above. There was no need for glass wool since this resin sinks in the column. Flow rate was 12 BV/hr or 8 mL/min. Zinc broke through after 15 bed volumes and the bed was saturated with zinc after 50 bed volumes. The amount of Zn up taken by the resin was 23.5 g Zn/L resin. Zinc was eluted off the resin using 3.5 bed volumes of 1M H2SO4.
Current Efficiency, Energy Consumption, and Specific Electrode Area Required for Copper Plating
Current efficiency (charge or Faradaic efficiency), energy consumption, and specific electrode area are factors required to scale up the system and calculate the total cost for the copper plating process. Current efficiency shows how much charge is utilized for copper plating out of the total charge used for the electrochemical reactions including Fe3+ and Zn2+ reduction and hydrogen evolution during the experiment. It can be calculated:
Wherein Eff. is efficiency, n is the number of electrons involved in the reaction, F is the Faraday's constant (96485) [C mole-1], ΔC is the concentration difference (C1-C2)[mol/l], V is the volume of the solution [l], I is the supplied current [Amp], and t I the time spent between C1 and C2 [sec]. Overall current efficiency was employed, which is accumulative current efficiency. As shown in FIG. 5, maximum overall current efficiency was obtained from 1st cycle at 78%, decreasing until 4th copper plating. From 4th to 8th cycles the overall current efficiencies are similar with around 70% maximum current efficiency. The decrease in overall current efficiency until 4th plating can be explained with the initial Fe3+ concentration in the regenerant for each experiment.
In FIG. 6, the initial concentration of Fe3+ accumulates with the number of cycles and it reaches to the steady state at 9000˜9500 ppm after 3rd cycle. As mentioned above, Fe3+ reduction competes with copper plating at the beginning of the experiments and more Fe3+ concentration would therefore consume more electrons, thus resulting in poorer current efficiency for the copper plating. As seen in FIG. 5, once the initial Fe3+ concentration is constant at around 9000 ppm after the 3rd cycle, the overall current efficiencies are similar to each cycle's except the 9th. The decrease in efficiency for the last cycle could be due to the lower initial copper concentration (see Table 2) and more electrons would be consumed for the Fe3+ reduction during the copper plating (see Table 3). Since 150 ml of sample was taken after 8th copper plating for the analysis, more acid was added to the regenerant to do the final regeneration, which made lower copper concentration in the regenerant for the 9th copper plating.
TABLE-US-00002 TABLE 2 Initial Cu2+ and Fe3+ concentrations for the copper plating experiments Cycle Cu(II) concentration Fe(III) number [ppm] concentration [ppm] [Cu(II)]/[Fe(III)] 1 4043 2137 1.892 2 3693 3712 0.995 3 3734 6837 0.546 4 3654 8847 0.413 5 3859 9459 0.408 6 3939 9515 0.414 7 3648 9280 0.393 8 3783 9618 0.393 9 3216 9513 0.338
TABLE-US-00003 TABLE 3 Initial and last Fe3+ concentrations during the copper plating experiments Cycle Initial Fe(III) Last Fe(III) number concentration [ppm] concentration [ppm] Fe(III) reduction % 1 2137 596 72 2 3712 725 80 3 6837 1558 77 4 8847 2307 74 5 9459 2301 76 6 9515 2102 78 7 9280 1146 88 8 9618 1157 88 9 9513 983 90
FIG. 7 shows how much energy was consumed to recover 1 kg of copper at each Cu2+ concentration. This also indicates that Fe3+ initial concentration affects the copper plating and more Fe3+ ions make the system consume more energy to recover the copper from the regenerate at the beginning of the process. Specific electrode area required for the copper plating has been also calculated and shown in FIG. 8. Two fold of electrode area is required for the copper plating with lowest current efficiency (cycle 9) compared to that with the highest (cycle 1) for the first half of the copper plating experiments (see FIG. 7).
Based on the above contemplations and data, it should be recognized that the previous significant obstacles associated with the presence of iron and its redox performance, low concentration of target ion copper, presence of other transition metals, and high concentration of dissolved salts have been eliminated as contemplated processes combine sample treatment (here: redox reaction and selective enrichment) with effective electrochemical device design (here: dynamic electrochemical cell with flow through electrode).
Thus, specific embodiments and applications of devices and methods of copper recovery have been disclosed. It should be apparent, however, to those skilled in the art that many more modifications besides those already described are possible without departing from the inventive concepts herein. The inventive subject matter, therefore, is not to be restricted except in the spirit of the appended claims. Moreover, in interpreting both the specification and the claims, all terms should be interpreted in the broadest possible manner consistent with the context. In particular, the terms "comprises" and "comprising" should be interpreted as referring to elements, components, or steps in a non-exclusive manner, indicating that the referenced elements, components, or steps may be present, or utilized, or combined with other elements, components, or steps that are not expressly referenced. Furthermore, where a definition or use of a term in a reference, which is incorporated by reference herein is inconsistent or contrary to the definition of that term provided herein, the definition of that term provided herein applies and the definition of that term in the reference does not apply.
Patent applications by Chulheung Bae, San Ramon, CA US
Patent applications by Stephen Harrison, Benicia, CA US
Patent applications in all subclasses Copper produced