Systems and methods for integrated solar photodialysis

Abstract

A photodialysis device that can desalinate saline streams is described herein. The device can have a desalination rate that is significantly faster than current solar thermal desalination technology. Salt is removed from water by passing ionic current derived from sunlight through this water using dye-sensitized membranes. The device can serve as a distributed, low-cost technology that can efficiently and effectively desalinate low salinity sources of water, which is particularly useful for agriculture and drinking water applications.

Claims

1. A deionization system (200) for deionizing a fluid, said system comprising: a) a chamber (210) for containing said fluid; and b) two or more ion-exchange (IX) membranes (110) arranged in the chamber so as to partition the chamber into two or more chamber compartments, wherein at least one of the IX membranes comprises a photoactive dye covalently bound to said IX membrane, whereupon exposure to a photon source, said photoactive dye is configured to undergo a regenerative and reversible light-driven dissociation reaction or light-driven association reaction to generate or remove a positively-charged mobile ion and/or a negatively-charged mobile ion, wherein ions in the fluid are transported between chamber compartments to affect deionization of the fluid in at least one of the compartments.

2. The system of claim 1, wherein the IX membranes are comprised of one or more layers, wherein at least one of the layers is photoactive for light-driven ion pumping.

3. The system of claim 1, wherein said layers are stacked optically in series to create a tandem membrane structure.

4. The system of claim 1, wherein the IX membranes comprise a cation-exchange membrane, an anion-exchange membrane, a photoactive-dye-sensitized polymer membrane, or a combination thereof.

5. The system of claim 1, wherein the IX membrane comprises a photoactive ion pumping membrane that operates with symmetric ion distributions.

6. The system of claim 1, wherein the IX membrane comprises a photoactive ion pumping membrane that transports only positively-charged ions or only negatively-charged ions.

7. The system of claim 1, wherein the photon source is sunlight, ultraviolet light, visible light, near-infrared light, or combination thereof.

8. The system of claim 7, wherein the IX membranes divide the chamber into two-compartments such that an inner compartment is configured to be deionized, and an outer compartment is configured to be concentrated with ions, or vice versa.

9. The system of claim 1, wherein the photoactive dye is an organic photoacid, a photoacidic quantum dot, a photoacidic inorganic or organometallic metal complex, an organic photobase, a photobasic quantum dot, a photobasic inorganic or organometallic metal complex, or an analogous dye where any ion is released or bound due to photoexcitation.

10. The system of claim 9, wherein the photoactive dye is according to any one of the following compounds: ##STR00026## ##STR00027## ##STR00028##

11. The system of claim 9, wherein the photoactive dye is a photoacid comprising a donor core and at least one acceptor component bound to the donor core.

12. The system of claim 11, wherein the acceptor component is according to any one of the following structures: ##STR00029##

13. The system of claim 11, wherein the donor core is according to any one of the following structures: ##STR00030##

14. The system of claim 9, wherein the photoactive dye is a photoacidic quantum dot having a plurality of ligands.

15. The system of claim 14, wherein the photoacidic quantum dot is according to the following structure: ##STR00031##

16. The system of claim 9, wherein the photoactive dye is a photobase according to any one of the following compounds: ##STR00032##

17. The system of claim 9, wherein the photoactive dye is a photoacidic inorganic metal complex according to any one of the following structures: ##STR00033## ##STR00034##

18. The system of claim 9, wherein the photoactive dye is a photoacidic inorganic metal complex comprising a metal group and a dithiol ligand modified with a protonated group.

19. The system of claim 9, wherein the photoactive dye is a photobasic inorganic metal complex according to any one of the following structures: ##STR00035##

20. A method of deionizing a fluid stream containing ions, said method comprising: a) providing a deionization system according to claim 1 for deionizing a fluid; b) adding said fluid stream to the deionization system provided in (a), wherein the fluid is disposed in the chamber of said system and is contacting the IX membranes; and c) exposing the IX membranes to a photon source, which causes the photoactive dye in the membrane to undergo a regenerative and reversible light-driven dissociation reaction or light-driven association reaction to generate or remove a positively-charged mobile ion and/or a negatively-charged mobile ion, that drive transport of the ions into or out of the fluid, thereby changing the ion content of the fluid.

Description

BRIEF DESCRIPTION OF THE SEVERAL VIEWS OF THE DRAWING(S)

(1) This patent application contains at least one drawing executed in color. Copies of this patent or patent application publication with color drawing(s) will be provided by the Office upon request and payment of the necessary fee.

(2) The features and advantages of the present invention will become apparent from a consideration of the following detailed description presented in connection with the accompanying drawings in which:

(3) FIGS. 1A, 1B, 1C, and 1D show device arrangements of a two compartment-chambered system for photodialysis.

(4) FIGS. 2A, 2B, 2C, and 2D show device arrangements for photodialysis using a three compartment-chambered system. FIG. 2D shows a plurality of capillary tubes in a bundle, each capillary tube having a cation-exchange membrane, an anion-exchange membrane, and an intervening dye-sensitized polymer membrane transecting the tube, as a non-limiting example. Photovoltaic action occurs upon photoexcitation of the dye-sensitized membrane, which causes salt ions to move from outside of the tube to inside the tube. Note on FIG. 2C, the black dots represent platinized electrodes and the redox reactions shown only to be used and occur in the prototypes.

(5) FIGS. 3A and 3B show device arrangements for photodialysis using four compartments.

(6) FIG. 4A shows a non-limiting schematic of a photoacid-modified bipolar IX membrane utilized in the present invention.

(7) FIG. 4B shows integrated solar photodialysis desalination using the membrane in FIG. 4A.

(8) FIG. 4C shows an example the membrane modified with a hydroxypyrene derivative, 8-hydroxypyrene-1,3,6-tris(2-aminoethylsulfonamide).

(9) FIGS. 5A, 5B, and 5C depict non-limiting examples of possible membrane configurations.

(10) FIG. 6A shows a non-limiting schematic of a photoacid dye sensitization cycle using a photoacid-dye-sensitized ion-exchange membrane.

(11) FIG. 6B shows a prior art schematic of desalination by traditional electrodialysis.

(12) FIG. 6C shows current density vs. potential curves for three membranes: Nafion-CS3 single layer, commercial Fumatech FBM, and Nafion/PVA/AEM three layer, each separating aqueous 10 mM HClO.sub.4 and aqueous 10 mM KOH.

(13) FIG. 7A is a photodialysis prototype of a two compartment system, FIG. 7B shows a photodialysis prototype using a four compartment system is accordance with FIG. 3A.

(14) FIG. 7C shows a two-compartment cell-in-a-container device is accordance with FIG. 1C, wherein the inner compartment is desalinated. This prototype mimics the ion-pumping behavior of an optimal solar-powered sodium pump by using two reversible hydrogen electrodes with low catalytic overpotential to generate ionic current.

(15) FIG. 8 shows potentiostatic chronoamperometry data for desalination using a four-compartment cell arrangement.

(16) FIG. 9A shows a cyclic voltammogram of the two-compartment straw device of FIG. 7C at different concentrations of aqueous HCl to mimic different stages of deionization. The ability of this device to desalinate water at voltages relevant to a theoretical light-driven ion pumping membrane was tested and the results thereof are shown.

(17) FIG. 9B shows an image of the inner compartment for a device like that in FIG. 7C, which here is a miniaturized version of that in FIG. 7C to minimize resistive losses.

(18) FIGS. 10A and 10B show numerical modeling results for NaCl leakage during desalination using the device of FIG. 7C.

(19) FIG. 11A shows potentiostatic chronoamperometry data for deionization of aqueous 50 mM HCl in the two-compartment tube setup.

(20) FIG. 11B shows a cyclic voltammogram of the system after chronoamperometry indicating deionization of the inner compartment to aqueous 10 mM HCl. Curves were transferred to the first quadrant to make interpretation easier.

(21) FIG. 12A shows a cyclic voltammogram using different-sized chambers with labeled resistance in aqueous 50 mM HCl.

(22) FIG. 12B shows images of chambers with approximate ion transport path lengths for the entire cell.

(23) FIG. 13A shows a non-limiting design of a multiple-compartment cell with pumping.

(24) FIG. 13B shows a cyclic voltammogram using the flow device with pumping and containing aqueous 50 mM HCl inside and outside of the device.

(25) FIG. 13C shows a non-limiting embodiment of a multiple-compartment cell with pumping via tubes connected to said device.

(26) FIG. 14A shows a plot of the effective current (mol/s) as a function of the concentration of aqueous HCl and best-fit line.

(27) FIG. 15A shows a 4-layer PINP membrane photovoltage under 405 nm illumination (shaded regions) wetted by aqueous 10 mM HClO.sub.4 on both sides and measured using four-electrode chronopotentiometry at zero applied current. The two current-carrying electrodes were platinum meshes and the two potential-sensing electrodes were KCl-saturated calomel reference electrodes. Illumination was from the PN side or the PI side.

(28) FIG. 15B shows a 4-layer PINP membrane photovoltage under 405 nm illumination (shaded regions) wetted by aqueous 10 mM HCl saturated with forming gas (5% H.sub.2 in N.sub.2) on both sides and measured using four-electrode chronopotentiometry at zero applied current. The two current-carrying electrodes were carbon rods and the two potential-sensing electrodes were platinized platinum mesh reversible hydrogen reference electrodes, with illumination from the PN side.

DETAILED DESCRIPTION OF THE INVENTION

(29) Following is a list of elements corresponding to a particular element referred to herein:

(30) TABLE-US-00001 100 ion-pumping membrane system 110 membrane 120 photoactive dye 200 deionization system 210 chamber for containing fluids

ABBREVIATIONS

(31) bpy, 2,2′-bipyridine

(32) bq, biquinolinc

(33) AEM, anion-exchange membrane

(34) BPM, bipolar membrane

(35) CEM, cation-exchange membrane

(36) IX, Ion-Exchange

(37) ISPD, integrated solar photodialysis

(38) PINP, refers to a 4-layer ion pump as cation-exchange membrane (Nafion 212; 50 μm thick), poly(p-phenylene oxide) with covalently-bound aminopyrene photoacid dye, anion-exchange membrane (Sustainion X37; 50 μm thick), and then another cation-exchange membrane (Nafion 212; 50 μm thick).

(39) RO, reverse osmosis

(40) As used herein, the term “Ion-pumping”, when used in conjunction with membranes, processes, applications, and the like, encompasses the terms “ion-exchange”, “ion-transpory”, “ion-conductive”, and “ion-selective”, and may be used interchangeably. The term “IX” refers to both ion-exchange and ion-transport and is defined as the transport of ions from one solution to another with (exchange) or without replacement by other similarly charged ions. IX membranes such as anion-exchange membranes or cation-exchange membranes predominantly transport ions of a single charge.

(41) As defined herein, the term “photoactive dye” is a compound that can undergo a regenerative light-driven dissociation reaction or light-driven association reaction to generate or remove a positively-charged mobile on and/or a negatively-charged mobile ion. Referring to FIG. 6A, one of ordinary skill in the art would understand that the photoactive dye can generate a positively-charged on or a negatively-charged ion upon photoexcitation, and then remove the same charge-type of ion from solution, thus regenerating itself. In some embodiments, the process is not strictly reversible and the photogenerated positively-charged ion is a proton and there is a photogenerated negatively-charged hydroxide ion. In other alternative embodiments, other ions may be generated or removed with light, including, but not limited to Na.sup.+, potassium ion (K.sup.+), magnesium ion (Mg.sup.2+), calcium ion (Ca.sup.2+), Cl.sup.−. These ions may also be responsible for generating transient power, similar to the H.sup.+ and OH.sup.− processes, and may be reversibly released and bound during the sensitization process.

(42) As defined herein, a bipolar membrane is an IX membrane composed of an anion-exchange layer and a cation-exchange layer. For the most common uses of conventional bipolar membranes, applying an electric bias helps facilitate rapid and efficient water dissociation inside the membrane to generate additional protons (H.sup.+) and hydroxide ions (OH.sup.−). Traditional electrodialysis cells must convert electronic power into ionic power. The present invention utilizes photoactive dyes to generate or remove positively-charged and/or negatively-charged ions using solar energy, and upon transport of these ions and regeneration of the dyes, the solution or membrane has a change in net charge. The light-driven ion pumping membrane of the present invention bypasses the step of applying an external electric bias and converting electronic power into ionic power and will therefore thermodynamically save about 85% of the energy required by an electrodialysis cell that performs water electrolysis to drive ionic current, which in some embodiments, also drive electrodialysis of seawater to potable water.

(43) Any of the modified IX membranes or bipolar membranes described herein may be used in an IX process such as desalination for treating wastewater, salt water, or any electrolyte solution. Desalination is an IX application known to one of ordinary skill in the art. Examples of which are disclosed in, but not limited to, U.S. Pat. No. 8,764,957, the specification of which is incorporated herein by reference. These membranes can be arranged in specific patterns and in desired quantities depending on the ultimate application and desired ion-transport processes. Additional details and principles of said IX membranes may be found in co-owned and co-pending U.S. application Ser. No. 15/698,324, which published as US2018-0065095A1, the specification(s) of which is/are incorporated herein in their entirety by reference. Briefly, the IX membrane features regenerative photoactive dyes that are covalently-bonded to a membrane or separator for ion-pumping. The photoactive dye-functionalized membranes can be arranged with other IX membranes, which serve as selective contacts to afford photovoltaic action and therefore form a power-producing membrane that pumps ions for use in driving an IX process, such as desalination and electrodialysis. The photoactive dye is regenerative such that upon exposure to a photon source, the photoactive dye undergoes a light-driven dissociation reaction or light-driven association reaction to generate or remove a positively-charged mobile ion and/or a negatively-charged mobile ion for use in the IX process, and a second mobile ion is generated or removed upon regeneration of the ground-state dye to its initial state also for use in the IX process. In preferred embodiments, the photoactive dye may, for example, be an organic photoacid, a photoacidic quantum dot, a photoacidic inorganic or organometallic metal complex, an organic photobase, a photobasic quantum dot, a photobasic inorganic or organometallic metal complex, or an analogous dye where any ion is released or bound due to photoexcitation.

(44) Referring to FIGS. 1A-15B, the present invention features ISPO devices equipped with ion-pumping, photoactive-dye-sensitized IX membranes and methods for deionizing a fluid. In some embodiments, the present invention features a deionization system (200) for deionizing a fluid. As shown in FIGS. 1A and 1B, in one embodiment, the present invention features a deionization system for deionizing a fluid using a two-compartment chambered system. The system may comprise a chamber (210) for containing a fluid to be de-ionized and two IX membranes (110) arranged in the chamber so as to partition the chamber into two or more chamber compartments. At least one of the IX membranes has a photoactive dye (120) covalently bound to itself (referred to as a dye-sensitized IX membrane). Upon exposure to a photon source, the photoactive dye is configured to undergo a regenerative light-driven dissociation reaction or light-driven association reaction. This reaction generates or removes a positively-charged mobile ion and/or a negatively-charged mobile ion causing the ions in the fluid to be transported between chamber compartments to affect deionization of the fluid in one of the compartments. The photon source may be sunlight, ultraviolet light, visible light, near-infrared light, or a combination thereof. Non-limiting embodiments of two compartment systems are shown in FIGS. 1A-1D. In these embodiments, two membranes are required with at least one being photoactive: one to pump an ion and the other to allow the oppositely charged ion to be pulled by the energy of the pumped ion, thus net transporting a neutral species, such as NaCl.

(45) In another embodiment, the deionization system (200) for deionizing a fluid uses a four-compartment chambered system (FIGS. 3A and 38). The system comprises a chamber (210) for containing the fluid and a plurality of ion-exchange (IX) membranes (110) arranged in the chamber so as to partition the chamber into a four-compartment chamber. At least one of the IX membranes comprises a photoactive dye covalently bound to the IX membrane (referred to as a dye-sensitized IX membrane). Upon exposure to a photon source, the photoactive dye is configured to undergo a regenerative light-driven dissociation reaction or light-driven association reaction to generate or remove a positively-charged mobile ion or a negatively-charged mobile ion causing the ions in the fluid to be transported between chamber compartments to affect deionization of the fluid in at least one of the compartments. For instance, one compartment is configured to be deionized, another compartment is configured to be concentrated with salt, and the other two compartments are configured to be either acidic or basic.

(46) In other embodiments, the present invention features a method of deionizing a fluid stream containing ions. The method uses a deionization system that includes chambers for containing the fluid and IX membranes for deionizing a fluid as described above. After adding the fluid stream to the deionization, system, the fluid is disposed in the chamber of the system and is in contact with the IX membranes. The membrane that has the covalently-bound photoactive dye is then exposed to a photon source, which causes the photoactive dye in the membrane to undergo a regenerative light-driven dissociation reaction or light-driven association reaction to generate or remove a positively-charged mobile ion or a negatively-charged mobile ion. This enables transport of the ions into or out of the fluid, thereby changing the ion content of the fluid.

(47) Referring to FIGS. 4A-4C, in some embodiments, the present invention features a system that uses an ion-pumping membrane system (100) in an IX process. The system may comprise an IX membrane (110) having a photoactive dye (120) covalently bound to said membrane (110). Without wishing to be bound by a particular theory or mechanism, upon exposure to a photon source, such as sunlight, ultraviolet, visible or near-infrared light, the photoactive dye is configured to undergo a regenerative light-driven dissociation reaction or light-driven association reaction to generate or remove a positively-charged mobile ion and/or a negatively-charged mobile ion. In some embodiments, the photoactive dye (120) may be covalently bound to at least a surface of the IX membrane (110) as shown in FIG. 4C. In other embodiments, the IX membrane (110) may be modified throughout their thickness with the dye.

(48) In some embodiments, the system may further comprise a second membrane juxtaposed or attached to the IX membrane, thus forming a membrane system (100) that is a bipolar membrane as shown in FIG. 5B. For example, the IX membrane may be a cation-exchange membrane (110a) and the second membrane may be an anion-exchange membrane (110b).

(49) In some embodiments, the present invention may feature a desalination system comprising a desalination chamber, the two-layer membrane system described above, a second cation-exchange membrane, and a second anion-exchange membrane. In one embodiment, the membrane system, the second cation-exchange membrane, and the second anion-exchange membrane may be disposed in the desalination chamber such that the membranes partition the desalination chamber into three chamber sectors, as shown in FIGS. 2A-2D. Further still, the cation-exchange membranes and the anion-exchange membranes may be arranged in an alternating pattern. During the IX process, fluids containing salt ions may be disposed within the desalination chamber, and photoexcitation of the photoactive dye in the dye-sensitized membrane causes the regenerative light-driven dissociation reaction or light-driven association reaction to generate or remove a positively-charged mobile ion and/or a negatively-charged mobile ion, which enables transport of the salt ions in the chamber sector disposed between the second cation-exchange membrane and the second anion-exchange membrane to cross the respective membrane, thereby changing the salt ion content from the fluids in said chamber sector.

(50) In other embodiments, the IX membrane (110) may comprise a polymer membrane (110c) to which the photoactive dye is covalently bound. In some embodiments, the membrane system may further comprise a cation-exchange membrane (110a) and an anion-exchange membrane (110b), in addition to the polymer IX membrane (110c).

(51) In one embodiment, the polymer IX membrane (110) (membrane (110c)) may be juxtaposed or attached to the cation-exchange membrane (110a) and the anion-exchange membrane (110b), thus forming a membrane system (100) that is a three-layer bipolar membrane as shown in FIG. 5C. A desalination system comprising a desalination chamber, the aforementioned membrane system, a second cation-exchange membrane, and a second anion-exchange membrane is also provided herein. Similar to the embodiments shown in FIGS. 2A-2D, the membrane system, the second cation-exchange membrane, and the second anion-exchange membrane may be disposed in the desalination chamber such that the membranes partition the desalination chamber into three chamber sectors, and where the IX process occurs as previously described. Preferably, the cation-exchange membranes and the anion-exchange membranes may also be arranged in an alternating pattern.

(52) In other embodiments, the IX membrane (110c) may be disposed between the cation-exchange membrane (110a) and the anion-exchange membrane (110b), but not necessarily juxtaposed in between. According to another embodiment, the desalination system of the present invention may comprise a desalination chamber, the IX membrane (110c), the cation-exchange membrane (110a), and the anion-exchange membrane (110b). Similar to the embodiments shown in FIGS. 2A-2D, the membranes may be disposed in the desalination chamber such that the membranes partition the desalination chamber into three chamber sectors, and where the IX process occurs as previously described.

(53) In further embodiments, the cation-exchange membrane (110a) may be connected edgewise to the anion-exchange membrane (110b) so as to form a cylindrical IX tube. The diameter of the IX tube can vary greatly, but to desalinate seawater to potable water using sunlight likely requires that the diameter of the IX tube be less than about 5 mm. This configuration of the desalination system may improve device efficiency by minimizing ohmic ionic voltage losses. In some embodiments, the IX membrane (110c) may be disposed within the IX tube such that the IX membrane (110c) transects the cation-exchange membrane (110a) and the anion-exchange membrane (110b) at the edge connections, as shown in FIG. 2D. Without wishing to be bound by a particular theory or mechanism, photoexcitation of the dye-sensitized membrane (110c) can enable transport of salt ions into or out of the IX tube by crossing the respective cation-exchange membrane (110a) or anion-exchange membrane (110b). This overall design is analogous to that described above with the general design strategy of connecting one side of the IX membrane to the other side of the lX membrane by a continuous fluid that has at least one intervening IX membrane. In other embodiments, at least two intervening IX membranes may be used.

(54) In some embodiments, for any of the designs or number of compartments, the cross-sectional shapes of the compartments can be a circle, oval, square, rectangle, triangle, or any shape. In other embodiments, for any of the designs or number of compartments, the compartments can be nested or not. The black dots in the Figures represent platinized electrodes and the redox reactions shown only occur in the prototypes. In some embodiments, the light-driven ion pump can pump ions across the dye-sensitized membrane and no electrodes or redox chemistry will be driven. There is an additional voltage loss required to drive the redox reactions and thus, a device running on a solar photovoltaic cell wired to electrodes to drive redox reactions that drive desalination is even more wasteful of energy.

(55) Another embodiment of the desalination system may comprise a desalination chamber and a plurality of the IX tubes as shown in FIG. 2D. The plurality of IX tubes may be bundled together such that the IX tubes are parallel to each other. The bundle of IX tubes may be disposed within the desalination chamber. During the IX process, fluids containing salt ions are disposed within the desalination chamber, and photoexcitation of the photoactive dye in the dye-sensitized membrane (110c) of each IX tube causes the regenerative light-driven dissociation reaction or light-driven association reaction to generate or remove a positively-charged mobile ion and/or a negatively-charged mobile ion, which enables transport of the salt ions into or out of the lX tubes by crossing the respective CEM (110a) or AEM (110b), thereby changing the salt ion content of the fluids between each IX tube.

(56) According to other embodiments, the present invention may feature a method of desalinating a fluid containing salt ions. The method may comprise adding said fluid to any of the desalination systems described herein, such that the fluid is disposed in a desalination chamber of said system and contacts the membranes. The method further comprises exposing the dye-sensitized membrane to a photon source, which causes the photoactive dye in the dye-sensitized membrane to undergo a regenerative light-driven dissociation reaction or light-driven association reaction to generate or remove a positively-charged mobile ion and/or a negatively-charged mobile ion that enable transport of the salt ions into or out of the fluid, thereby changing the salt ion content of the fluid.

PHOTOACTIVE DYES

(57) Consistent with previous embodiments, the membranes of the desalination systems and ISPD devices utilizes a photoactive dye for generating ions upon exposure to a photon source. Without wishing to be bound to a particular theory or mechanism, the photoactive dye is configured to undergo a regenerative light-driven dissociation reaction or light-driven association reaction to generate or remove a positively-charged mobile ion and/or a negatively-charged mobile ion. Preferably, the photoactive dye is capable of being covalently bound to a substrate, such as a polymer or IX membrane. Consistent with previous embodiments, the photoactive dye may be a photoacid, a photobase, or a dye capable of performing light-driven ion release or capture. The following examples of photoactive dyes and syntheses thereof are not intended to limit the invention in any way. Equivalents or substitutes are within the scope of the invention.

PHOTOACIDS

(58) As defined herein, a photoacid is a molecule that becomes more acidic when it undergoes a regenerative light-driven dissociation reaction upon absorbing photons. Photoacids differ from photoacid generators (PAGs) in that PAGs undergo photodissociation irreversibly.

(59) In one embodiment, a visible-light-absorbing photoacid dye molecule that resembles hydroxypyrene with free amine groups that can be bound to polymers (FIG. 4C) was synthesized. Specifically, a congener of the state-of-the-art visible-light-absorbing organic photoacid, pyranine, was synthesized by appending three amine groups to its hydroxypyrene core to allow for covalent attachment to polymers. As illustrated in Scheme 1, first, the alcohol group of the tris(sulfonato)hydroxypyrene (pyranine) starting material is protected, then the three sulfonates were activated to sulfonyl chlorides, and then converted to sulfonamide groups using N-Boc-ethylenediamine followed by deprotection using trifluoroacetic acid.

(60) ##STR00001##

(61) In some embodiments, the photoacid dye molecule may comprise an 8-Aminopyrene-1,3,6-trisulfonic acid trisodium. A non-limiting reaction scheme of synthesizing said photoacid may be according to Scheme 2:

(62) ##STR00002##

(63) In other embodiments, the photoactive dye may be a strong photoacid comprising a “Donor-pi-Acceptor” structure or the plurality of any of these groups. Non-limiting example of acceptors include the following:

(64) ##STR00003##
where

(65) ##STR00004##

(66) Examples of donors include, but are not limited to, the following:

(67) ##STR00005##
where

(68) ##STR00006##

(69) In one embodiment, an acceptor-donor-acceptor photoacid with a phenol donor core may be synthesized according to Scheme 3.

(70) ##STR00007##

(71) In another embodiment, an acceptor-donor-acceptor photoacid with a pyrrole donor core may be synthesized according to Scheme 4.

(72) ##STR00008##

(73) In other embodiments, the photoacid dye molecule may have a long half-life in the photodissociated state. A non-limiting reaction scheme of synthesizing said photoacid may be according to Scheme 5:

(74) ##STR00009##

(75) In further embodiments, the photoacid may be according to any of the following structures:

(76) ##STR00010## ##STR00011## ##STR00012##

(77) In alternative embodiments the photoacid may comprise compounds such as quantum dots. A non-limiting example of a quantum dot photoacid is as follows:

(78) ##STR00013##

(79) In other alternative embodiments, the photoacid may comprise compounds such as inorganic coordination compounds. Without wishing to be bound by a particular theory or mechanism, the use of inorganic compounds as a photoacid, and alternatively as a photobase, is enabled by a photoreaction where the inorganic compound absorbs light, an electron moves from or to a metal of the inorganic compound and to or from a ligand of the inorganic compound. The end location of the electron will most likely be a more basic site and the initial location of the electron will most likely be a more acidic site. For instance, phenanthroline can have an NH.sub.2 group that receives an electron where —NH.sub.2+H.sup.+.fwdarw.—NH.sub.3.sup.+. Other examples of common acceptor ligands include bpy or bq, which also have a series of energies. In some embodiments, the electron can often originate on the metal and thus, the functional group that loses a proton should be bound to the metal. In other embodiments, if a coordination compound is insoluble in water, then said compound can be made water soluble by adding an —SO.sub.3 group, or the compound can be bound to a polymer and immersed in water as an insoluble film.

(80) In some embodiments, the photoacidic inorganic compounds may be according to the following:

(81) ##STR00014##

(82) In other embodiments, the photoacidic inorganic compounds may comprise a hydride bound to a metal, such as Ir or Ru. The photoacidic inorganic compounds may compose a ligand that can donate electron density to the metal so that photoexcitation results in more oxidation of the ligand than the metal. In one embodiment, the inorganic compounds having ligands that can donate electron density may be according to the following:

(83) ##STR00015##

(84) Non-limiting examples of photoacidic inorganic compounds include the following:

(85) ##STR00016## ##STR00017##

(86) According to further embodiments, the photoacidic inorganic compounds may comprise a metal complex having a metal group, such as Pt, Pd, and Ni, and a dithiol (S-containing) ligand modified with a protonated group, such as —OH or —NH.sub.3.sup.+. Non-limiting examples of metal complexes that may be modified with a protonated group (not shown) include the following:

(87) ##STR00018##

(88) In an exemplary embodiment, the inorganic metal complex may be synthesized to be water soluble and exhibit full charge transfer after photoexcitation, resulting in formal oxidation and reduction of two ligands, as shown in Scheme 6.

(89) ##STR00019## ##STR00020##

PHOTOBASES

(90) According to some embodiments, the photoactive dye may be a photobase, which can covalently bind to the membranes described herein. Upon light absorption, the photobase can generate an OH.sup.−, instead of H.sup.+ as in the case of photoacids, and net form an H.sup.+ during the regeneration process, as opposed to OH.sup.− for the photoacids. Examples of photobases include, but are not limited to, the following acridine-core photobases:

(91) ##STR00021## ##STR00022##

(92) In one embodiment, an acridine-core photobase may be synthesized according to the reaction shown in Scheme 7.

(93) ##STR00023##

(94) In alternative embodiments, the photobase may comprise a photobasic metal complex. In one embodiment, the photobasic metal complex may have the following structure:

(95) ##STR00024##

(96) Finite-element device physics models were created and simulations performed on ideal light-driven ion pumps infiltrated with water whose properties were modeled as being the same as bulk water. Under solar-simulated conditions, the models show that the membranes can generate a photovoltage of −0.35 V. This may be a lower bound to the true maximum theoretical efficiency given that bulk water properties change in polymers. Irrespective, because of the energy requirements, the implementation of the three-compartment architecture for generation of acid, base, and desalinated water will require a more complex tandem light-absorbing membrane architecture in order to obtain the required voltage to drive the cell. This can be achieved using a membrane or architecture that has additional layers in a so-called tandem structure. That is, given photoactive one-layer, two-layer, three-layer or n-layer (where n is an integer) ion pumps, several of these can be stacked optically in series. As long as some light transmits to the bottom materials and there is a low-resistance contact between each photoactive material, each will generate a photocurrent. The overall observed photocurrent will be limited in the serial stack by the ion pump with the smallest photocurrent, but the photovoltage will be the sum of the photovoltages from each photoactive material. This architecture can generate larger voltages than a single unit and will allow a device architecture with acid, base, and desalinated water to operate. In fact, this architecture will be useful in general, which is described in greater detail below.

(97) Using a single light-driven ion-pumping membrane that is not a tandem structure, there are two potential ways to overcome the voltage mismatch between the required voltage and the maximum attainable photovoltage in some embodiments, one option is to use a two-compartment cell (FIG. 1A or 1B) and a sodium-pumping and/or chloride-pumping membrane that does not generate acid or base and therefore only requires a minimum of ˜0.1 V to desalinate brackish water and ˜0.2 V to desalinate sea water, each to potable water. This voltage requirement is well within the theoretical photovoltage range of light-driven ion-pumping membranes. As such, an electrochemical cell and device that could mimic the best theoretical ion-pumping membrane were used to experimentally determine the feasibility of the present invention. In other embodiments, another option to overcome the voltage mismatch is to use a four-compartment cell (FIGS. 3A and 3B) where a gradient in alkaline proton concentration generates an asymmetry in the materials that is not as extreme as the case shown in FIG. 2A.

(98) In some embodiments, the two-compartment cell is advantageous because it is simple, will have fewer resistive losses, and has no energetic penalty to overcome acid or base gradients. In some embodiments, the four-compartment cell is advantageous because it can use a light-driven proton pumping membrane, where light drives proton transport. The membrane required to operate with no acid/base gradient is already possible. As shown in FIG. 6C, most two-layer and some three-layer bipolar membranes exhibit efficient and ohmic ionic conduction irrespective of the direction of polarization bias, and some two-layer and most three-layer bipolar membranes rectify current like diodes and make effective light-driven ion pumps. By affixing these two membranes together, one has a single membrane that presents the same type of membrane on each side (i.e. cation-exchange or anion-exchange). As such, this stack can be immersed in a single pH condition and exhibit photovoltaic action as a light-driven ion pump that is very well suited for the arrangement shown in FIGS. 3A and 3B. Irrespective, while the four-compartment cell has advantages over the three-compartment cell, it still has some limitations to producing acid/base. The production of acid/base can be circumvented through use of a direct light-driven sodium or chloride pump. Chloride pumping proteins exist in nature, which suggests that it is possible to fabricate an artificial structure with this same function but using chloride analogs of photoacidic molecules. Analogous to photoacids, one way to accomplish this is to use light to alter the equilibrium binding constant of a Cl.sup.− ion, or other anion/cation, to the dye molecules.

(99) Two electrodes that perform Faradaic electron-transfer reactions t their interfaces at ˜0 V overpotential on two faces of an insulating sheet of plastic are equivalent to an ion pumping membrane that does not drive any Faradaic chemistry. As such, the reversible hydrogen electrode (2H.sup.++2e.sup.−.fwdarw.H.sub.2) was selected for the Faradaic reactions, as this reaction is catalytically facile at platinum and therefore swill exhibit small electrocatalytic overpotentials at low photon-flux-relevant current densities. This arrangement is presented in FIG. 7A for the two-compartment cell and FIG. 7B for the four-compartment cell.

(100) In FIG. 7A, the electrode on the left removes protons from solution via H.sub.2 evolution, and the electrode on the right adds protons to solution via H.sub.2 oxidation. The net chemistry is equivalent to a membrane that pumps protons across it. This net change in charge distribution tomes a chloride through the anion-exchange membrane (AEM), deionizing the left chamber. NaCl was selected to replace HCl because the H.sub.2 chemistry only changes local H.sup.+ concentration and thus HCl is necessary to most accurately mimic the ion pumping behavior. Another embodiment could use an AG/AgCl electrode or an Na.sup.+ intercalation material as a surrogate for generating and removing Cl.sup.− and Na.sup.+, respectively. Use of H.sup.+ instead of Na.sup.+ makes the electrolyte substantially more conductive. In order to account for this difference in calculations, a correction factor was introduced that was based on the aqueous ionic conductivities of H.sup.+ and Na.sup.+ and the aqueous cation transference numbers of HCl and NaCl. Based on these numbers, in order to accurately predict how a sodium pumping membrane would behave in the presence of aqueous NaCl, instead of aqueous HCl, the current must be divided by 3.28. In order to minimize resistive losses in the electrolyte, a cell that was as small as was reasonable to make was fabricated. This allows the fastest desalination rate per unit volume of fresh water produced. For versatility, maximum efficiency, and with maximal surface area of light-powered membrane exposed, a device with small tubes that are easily fabricated (e.g. by electrospinning, by extrusion) were made using 3D printing or other methods.

EXAMPLES

(101) The following are non-limiting examples of the present invention. It is to be understood that said example is not intended to limit the present invention in any way. Equivalents or substitutes are within the scope of the present invention.

Example 1: Two-Compartment System

(102) Representative embodiments of two-compartment arrangements are shown in FIGS. 1A-1D. The embodiments in FIGS. 1C and 1D can be manufactured as two nested tubes. The two-compartment arrangements have the lowest energy requirements and thus the smallest voltage losses. In fact, only ion transport current losses and desalination energy requirements would be present. There is no loss for performing additional reactions as no other reactions occur. This requires a light-driven pump of directly Cl.sup.+ or Na.sup.+ or other ions that are present in high concentrations, and not H+/OH— which in water are present in low concentrations. This basic reactivity via a dye (example of structure shown below) that changes its binding affinity for Cl.sup.−.

(103) ##STR00025##

(104) FIGS. 1A and 1B show how direct light-driven ion pumps of Cl.sup.− (FIG. 1A) or Na.sup.+ (FIG. 1B) are able to drive oppositely charged ions with minimum energy requirement and no need for starting with, making, wasting energy for, or dangers associated with acid and/or base. Additional embodiments of the present invention may include other dyes that will bind/release ions other than H.sup.+ after absorbing light. FIGS. 1C and 1D are designs only for measurement purposes of the processes during desalination as they generate/consume H.sup.+ or OH.sup.− and thus require that the solution has those species in it, and not NaCl. In this sense, they do not perform desalination but rather analogous deionization as a completely reasonable model reaction for ultimate desalination with a light-driven ion pump. Electrochemical reactions cannot as easily generate/consume Na.sup.+/Cl.sup.− in the present model cells to (tiredly mimic the light-driven ion pump concept. The reduction and oxidation reactions are in this case shown to switch locations between designs to demonstrate how in both cases deionization of the inner compartment can result.

Example 2: Three-Compartment System

(105) FIGS. 2A-2D show representative three-compartment arrangements. For the first ions desalinated, there is ever-so-slightly greater than zero voltage required because all of the NaCl chambers have the same concentration (FIG. 2A, initial state before desalination). Ideally after desalination (FIG. 2B), the product HCl and NaOH should be as dilute as possible. For example, at 1 M HCl and NaOH there is a 0.83 V potential requirement on top of that required for desalination which is ˜0.1 V to take brackish water to potable water and ˜0.2 V to take sea water to potable water; this acid/base value decreases by 0.12 V per order of magnitude decrease in HCl and NaOH concentration; thus, at 0.1 M HCl and NaOH, it requires 0.71 V on top of that required for desalination. All of these values and those in all of the calculations are based on room temperature (25° C.) operation where RT/F=0.059 V. The projected area of each compartment (volume in 3D) can be adjusted to result in less concentrated acid and/or base; however, larger compartments may have more losses due to ion transport and a smaller salt compartment means that it has to be changed/flowed more often to bring in new water to be desalinated. Notwithstanding, if NaCl starts at 0.1 M (brackish water) but the salt compartment is ten times smaller than each of the acid and base compartments, then once the salt compartment reaches 0.01 M NaCl (potable water), by removing 0.09 M NaCl, the acid and base compartments will only be 0.009 M NaOH and 0.009 M HCl due to the volumetric dilution. This would mean there is a 0.59 V voltage loss to pump the final ions into that acid and base, but recall that the voltage loss is zero for the first ions passed. Also, the voltage to desalinate the final ions is 0.12 V due to the single order-of-magnitude gradient in salt at the end. Thus, the voltage loss ranges from 0 to 0.71 V (0.59 V+0.12 V) during desalination of the 0.1 M NaCl to 0.01 M NaCl with making salty acid and salty base side products.

(106) FIG. 2C is another embodiment of the three-compartment arrangement. This design uses nested tubes and can be more amenable to manufacturing and production than the design that trifurcates the inside of a small tube (as shown in FIGS. 2A-2B).

(107) In some embodiments, cylinders or chambers may be flowed, the acid and base compartments could be flowed faster than the salt compartment(s) to minimize buildup of acid and base and minimize wasted voltage. In other embodiments, a photovoltage is generated with the light-driven ion pump when there is no acid/base gradient across the membrane; the initial state requires no acid base gradient. In other embodiments, the designs may have different area membranes such that if the dye-sensitized IX membrane is small then when a light-driven ion pump is used its current will be smaller than if it were the same size as the CEM and AEM. This means that smaller current (i) is passed through the CEM and AEM and electrolytes, which means that less voltage is wasted due to ion transport following Ohm's Law (V=iR). As the resistance (R) is dictated by the properties of the materials and solutions, only current is varied by the light-driven ion pump and so a smaller current means that the ohmic ionic voltage loss (V) is small. Smaller voltage losses indicate that the device will more effectively utilize the energy in light to drive desalination.

Example 3: Four-Compartment System

(108) Representative images of two four-compartment arrangements operating with base (FIG. 3A) and acid (FIG. 3B), this design may be advantageous over the three-compartment arrangement because when one starts with more concentrated base in the top two compartments and at the same concentration so that moving the first ions incurs no voltage loss due to the add and base present, as was the case for the three-compartment arrangements (FIG. 2A), then when base is generated and lost during operation the initial concentration will not change much from the condition of the same amount of base and thus little voltage loss will be incurred due to gain or loss of base. As shown, during measurements all reduction reactions are in the top, left compartments and oxidation reactions in the top, right compartments and that is why the desalinated compartment switches places in the two designs. This arrangement can easily be flipped in measurements and in practice by switching the directional arrangement of the light-driven H.sup.+/OH.sup.− pumping membrane. This embodiment also requires a light-driven H.sup.+/OH.sup.− pump that operates with no acid/base gradient.

(109) For example, assume all four compartments have the same projected area, even though this design can also benefit from having smaller salt compartments than the acid or base compartments. Then with 1 M NaOH in the top two compartments and 0.1 M NaCl (brackish water) in the bottom two compartments, initially there is no voltage loss to overcome in order to drive net desalination because the ion transport process reactions between the basic compartments have no initial concentration difference and drive the opposite reactions of gain of NaOH and loss of NaOH (left to right), and the ion transport process reactions between the salt compartments have no initial concentration difference and drive the opposite reactions of loss of NaCl and gain of NaCl (left to right). Once this design is desalinated to brackish water, the compartments would have 1.09 M NaOH (top, left), 0.91 M NaOH (top, right), 0.01 M NaCl (bottom, left), and 0.19 M NaCl (bottom, right). At that point, the voltage loss to drive one more series of ions through this circuit as ionic current would be 0.0046 V for the base/base transport process reactions (as RT/F log (1.09/0.91)) and 0.076 V for the salt/salt desalination transport process reactions (as RT/F log (0.19/0.01)). This means that the total voltage loss including desalination would equal 0.080 V at the end, which is small compared to (I) voltages required to generate 1 M acid and 1 M base in the three-compartment design, (ii) the 1.23 V required in an electrodialysis cell that drives water electrolysis, and (iii) most of the ionic ohmic ionic voltage losses at modest currents, and this 0.080 V total voltage loss is below the reasonable maximum photovoltages predicted from a light-driven ion pump of ˜0.35 V.

(110) While this design and the small voltage loss means that a light-driven H.sup.+/OH.sup.− pump could be practical, a challenge with this design is that aqueous base needs to be supplied initially and replenished. However, since each base compartment becomes equally more basic and less basic over time, if the contents of the compartments are ultimately mixed they would revert back to their initial basic condition, assuming there was no parasitic base leakage elsewhere. This means that the initial base can be reused. One way to slow parasitic leakage of base is, for example, if the base compartments start at 0.1 M NaOH, which will be less leaky than if they started at 1 M NaOH as predicted by Donnan Exclusion Theory. In this case the initial voltage loss would be 0 V and the final loss once 0.1 M NaCl is desalinated to 0.01 M NaCl in the bottom, left compartment would be 0.15 V (as 0.076 V+0.076 V, for making each of NaOH and NaCl in the concentrated sides and losing each from the dilute sides), which is still a small voltage requirement.

Example 5: Experimental

(111) Materials: Neasepta® anion-exchange membrane (Neosepta AHA), Sustainion anion-exchange membrane (Sustainion X37), Nafion NR212 (Ion Power), Pt wire, Pt mesh, Pt foil, lead acetate, chloroplatinic acid, poly(acrylonitrile-co-butadiene-co-styrene) (ABS), Parafilm, aqueous HCl, NaOH, MilliQ water, HM Digital electrical conductivity monitor model DM-2EC.

(112) Fabrication of a two-compartment cell: The smaller compartment of this electrochemical cell was produced by 3D printing ABS plastic into a rectangular tube with a 2 mm×2 mm inner compartment that was 6 cm long. The walls of this compartment were 2 mm thick. An ABS cover was printed that matched the size of the front face of the tube. A Neosepta AHA membrane was attached to the tube by taping the top of the tube to the membrane. Next, the ABS cover was taped over the bottom of the tube, enclosing the membrane. Acetone was applied to the edges of the cover where they contacted the tube. The ABS dissolved and re-solidified around the membrane, creating an effective seal. The seal was confirmed by injecting excess electrolyte into the top of the tube. Electrolyte only entered or exited through the top of the tube, indicating that there were no holes where water can leak through via simply flowing. Since no pressure was applied to this cell, leaks from pressurization were not expected. It is important to apply the acetone such that the edge of application is angled downward. Otherwise, ABS will dissolve on the front face of the membrane and make the Neosepta AHA insulating. After the acetone dried, the top of the cell was cut, and the bottom section of tape was removed, leaving a tube with a Neosepta AHA face. The larger compartment was 3D printed from ABS and made large enough to house the smaller compartment.

(113) Fabrication of a two-compartment cell with pumps: A multi-compartment device was fabricated as shown in FIG. 13A. The 2 outer plates were 3D printed in VeroClear plastic using an Objet Polyjet 3D printer. Sealing was provided with 4 Viton gaskets laser cut with the designs shown in the figure. Gasket 1 allowed the flow of acid and base through inlet and outlet holes, while providing contact of water with the respective membrane (CEM=Cation-Exchange Membrane, or AEM=Anion-Exchange Membrane). Gasket 2 allowed the flow of base and acid, while providing contact with the middle photoactive membrane that generates current. The device is designed to pump acid/base into the device using peristaltic tubing, while it is immersed in the desalinating water.

(114) Fabrication of a four-compartment cell: The four-compartment cell was produced by 3D printing ABS into four separate quarter circle compartments that were held together with a worm clamp. Two Nation membranes were sandwiched individually between two sheets of Parafilm. A hole was punched in the Parafilm to expose the Nafion to the electrolyte in the contacting cell compartments. The Parafilm was used to improve the seal between Nation and the walls of the compartment. These membranes were placed in contact with the compartments containing electrodes. A Neosepta AHA membrane was sandwiched between Parafilm in a similar way and placed between the compartments that did not contain electrodes.

(115) Platinum black electrode fabrication: Platinum wire, mesh, and foil electrodes were prepared by soldering the desired piece of platinum to a steel wire, enclosing the soldered section in a glass tube, and epoxying over the tube to seal the soldered section away from electrolyte. Platinization was performed via two-electrode chronoamperometry where a potential between −1 V and −4 V was applied to the platinum working electrode versus a platinum mesh counter electrode for 1 to 10 minutes.

(116) Two-compartment cell cyclic voltammogram (CV): In a typical measurement, the 2 mm×2 mm tube was loaded with aqueous 50 mM, 30 mM, or 10 mM HCl via injection with a needle and syringe. About 1.5 cm of electrolyte was removed from the top of the cell, leaving about 2.5 cm of electrolyte wetting the Neosepta AHA membrane. The cell was attached to a Pt black mesh electrode by wrapping the electrode around the top of the tube with Parafilm. The Pt was near the center of the wetted region of the membrane on the opposite side of the cell. A Pt black wire electrode was placed down the tube and served as the working electrode. The platinum mesh served as the counter electrode. This tube was hung through a Teflon lid into another compartment filled with aqueous 50 mM HCl. The lid had a hole in it that was larger than the tube and electrode but smaller than the Parafilm ring at the top of the tube. Thus, the device hung by the Parafilm into the larger compartment. A plastic tube attached to a hydrogen gas tank was fed to the bottom of the outer compartment and bubbled hydrogen into the cell. The water lines of the outer and inner compartments were approximately the same. CV scans were taken with a Biologic potentiostat, typically sweeping from +0.01 V to −0.4 V at 5 mV/s.

(117) Two-compartment cell CV using the device with mechanical pumps: In a typical measurement, the device was put into a two-compartment arrangement. The setup allowed electrolyte to be flowed into a small chamber between Neosepta AHA and a Pt black foil working electrode. A Pt black foil counter electrode was placed on the opposite side of a Viton sheet from the working electrode and exposed to the outer chamber. CV measurements were taken exactly as those in the two-compartment tube.

(118) Two-compartment cell CA: In a typical measurement, the device and container were setup in an idential way to the two-compartment tube CV with aqueous 50 mM HCl in the inner compartment, Potentiostatic chronoamperometry was performed at −0.35 V until the system passed −0.36 mA of current. This current value was chosen because it corresponded to the current that the device passed at −0.35 V when aqueous 10 mM HCl was loaded into the cell. After this current was reached, CV measurements were taken with the parameters typical of a two-compartment CV.

(119) Four-compartment cell CA: In a typical measurement, aqueous 50 mM NaCl and aqueous 50 mM NaOH were loaded into the compartments labeled with these electrolytes in FIGS. 3A and 3B. Platinum black mesh electrodes were placed into the compartments containing basic solution and 5 V was applied between the electrodes for 2 hours. Then the electrolyte in the desalinated compartment was unloaded, its ion concentration was measured with a conductometer, and its pH was approximated with pH paper.

(120) Results

(121) Using the four-compartment cell, aqueous 50 mM NaCl was desalinated by applying a large voltage (5 V) across the cell for an extended period of time (2 hours). The major goal of this experiment was to determine whether it was possible to desalinate aqueous 50 mM NaCl and recover potable water. Leakage of base across the membrane from the concentrating base chamber into the desalination chamber was a concern due to the large concentration difference of OH.sup.− across the Nafion membrane. If too much base leaks, the water will not be potable for drinking or usable for agricultural or other purposes. FIG. 8 shows potentiostatic chronoamperometry data measured using a four-compartment cell to desalinate aqueous 50 mM NaCl.

(122) After desalination, the concentration of the desalinated compartment was determined to be 6 mM NaCl, based on measurements from an electrical conductivity monitor that had been calibrated for use with aqueous NaCl. The pH was initially 6 and after the experiment the pH was measured to be 8-9. This suggests that some base leaked from the concentrated base compartment. Because the four-compartment system is much more difficult to work with from a manufacturing and practicality perspective, the simpler two-compartment design was investigated, which most likely requires the use of a light-driven chloride or sodium pump.

(123) FIG. 9A presents CVs of the two-compartment cell at different conditions during desalination. As is evidenced from the data, a decrease in the concentration of salt (ionic strength) results in an increase in the resistance of the electrolyte. To predict the amount of time it would take to desalinate water at 0.35 V, the current at 0.35 V for the CVs was averaged over at least three trials per concentration. A linear dependence of current on concentration was assumed and used to create an equation for current as a function of concentration/molarity. As molarity is dependent on amount of charge passed, which is the time integral of the current, the concentration as a function of time was calculated using a first-order differential equation created from the data. FIG. 9B shows an image of a smaller version of the inner compartment of the device of FIG. 7C, which is a miniaturized tube used to decrease resistive losses.

(124) Deviation of Time to Desalinate 50 mM Salt to 10 mM.

(125) The best fit line in FIG. 14 is: i=3.68*10.sup.−7 M+6.56*10.sup.−10=C.sub.1M+C.sub.2, where C.sub.1 and C.sub.2 are constants and M is concentration (M).

(126) $\mathrm{Since}=\frac{\mathrm{dQ}}{\mathrm{dt}},$
wherein Q is charge (C) and t is time (s), and

(127) $\frac{Q}{V}=M_0-M\left(t\right),$
where V is volume (L), M.sub.0 is initial concentration (M) and M(t) is the concentration at any time, then

(128) $\frac{1}{V}\frac{\mathrm{dQ}}{\mathrm{dt}}=-\frac{\mathrm{dM}}{\mathrm{dt}}=\frac{i}{V}=\frac{1}{V}\left(C_{1}M+C_2\right),\mathrm{and}\frac{\mathrm{dM}}{\mathrm{dt}}+\frac{C_1}{V}M=-\frac{C_2}{V}.$

(129) Multiplying by an exponential term one gets

(130) $\frac{\mathrm{dM}}{\mathrm{dt}}{e}^{\frac{C_{1}t}{V}}+\frac{C_1}{V}{\mathrm{Me}}^{\frac{C_{1}t}{V}}=-\frac{C_2}{V}{e}^{\frac{C_{1}t}{V}},$
which simplifies to

(131) $\frac{\mathrm{dMe}\frac{C_{1}t}{V}}{\mathrm{dt}}=-\frac{C_2}{V}{e}^{\frac{C_{1}t}{V}}.$
Integration yields

(132) $\mathrm{Me}\frac{C_{1}t}{V}=-\frac{C_2}{V}{e}^{\frac{C_{1}t}{V}}+C_3,$
which simplifies to

(133) $M=-\frac{C_2}{C_1}+C_3{e}^{\frac{C_{1}t}{V}}.$
For initial M.sub.0=50 mM,

(134) $0.05=-\frac{C_2}{C_1}+C_3,$
and given the best-fit line, C.sub.3=0.0518, V=0.1 mL, and M.sub.final=10 nM,

(135) $0.01=-0.0018+0.0518e^{-3.68*{10}^{-3}{t}_{\mathrm{desal}}},$
and thus, t.sub.desat=402 s.

(136) After solving this equation for a total volume of 0.1 mL, it was calculated that it would take 402 seconds to deionize the chamber to aqueous 10 mM HCl. After correcting for the conversion of resistance to that expected for NaCl, 1 mL of water, and 1 cm.sup.2 of active area, the desalination time was determined to be ˜6600 seconds. This is ˜2.7 times faster than the theoretical maximum for the solar thermal distillation process, and therefore a promising means to desalinate water of this salinity using sunlight.

(137) It is known that even small amounts of NaCl can leak across IX membranes, so the rate of leakage between a 2 mm long chamber containing aqueous 10 mM NaCl in contact with a compartment containing aqueous 50 mM NaCl was numerically modeled and simulated. The leakage rate was determined to be slow enough that only a <4% change in salinity would occur over the course of the 1300 sec required to desalinate the chamber, therefore resulting in a chamber with aqueous 10.39 mM NaCl instead of aqueous 10.00 mM NaCl. Therefore, the leakage rate will be considerably smaller than the productive rate of desalination. FIG. 10A shows the normalized integral of NaCl concentration over the 2 mm chamber. The total of the integral is an important value that shows a nearly negligible rate of ion crossover (FIG. 10B).

(138) FIGS. 11A and 11B show representative potentiostatic chronoamperometry data for desalination of the inner compartment. The experiment was terminated when the system passed 0.36 mA of current, which is the average current passed during a CV experiment when the inner compartment contained aqueous 10 mM HCl. As further evidence that the water was deionized, a CV of the system after performing a long-time potentiostatic chronoamperometry experiment clearly matches the CV data taken when aqueous 10 mM HCl was in the center compartment, indicating that aqueous ˜10 mM HCl was in the center compartment after the chronoamperometry experiment. Given the complexities in loading the device with electrolyte, the total amount of charge passed during the experiment was used to compute the volume of electrolyte present instead of measuring it directly. After averaging over five trials, it was determined that ˜0.10 mL of electrolyte was in the compartment. This took 430 seconds to desalinate, which implies that after converting to 1 mL of desalinated aqueous NaCl with 1 cm.sup.2 of active area, it would take ˜7000 seconds to desalinate 1 mL of water, which is ˜2.5 times faster than the solar thermal distillation process. These results agree well with the predicted desalination time based on CV data.

(139) Referring to FIGS. 12A and 12B, in order to demonstrate the importance of the size of the device, the resistance was tested as a function of path length for three different cells. Cells were loaded with aqueous 50 mM HCl and placed in a beaker filled with aqueous 50 mM HCl. CVs were taken as described in the previous two-compartment experiments. These data suggest that the resistance changes linearly with the path length.

(140) FIG. 13B presents data collected on the device shown in FIGS. 13A and 13C, which includes pumps to circulate and collect water streams. Pumping could be achieved using excess solar energy (probably from infrared light) that is not utilized by the ion pump to heat the water anisotropically and create density gradients and convective flows. Since this device with pumps attached is larger than the previously measured device, the rate of desalination is slower with this device. Still, since the device that has pumps attached dearly passes current, it demonstrates that flow can be incorporated into smaller, more effective devices with small ionic ohmic ionic voltage losses.

CONCLUSIONS

(141) The present invention demonstrates the feasibility of a solar-powered ion pump water desalination system. This system is ideally suited for tower salinity brackish water desalination where crossover of any salt ions is low and permselectivity is high. After testing a series of devices, a two-compartment, three-compartment, or four-compartment cell device may be used. In preferred embodiments, a simpler two-compartment cell architecture that can incorporate a light-driven sodium or chloride pump was configured to lower resistive losses in the cell and remove the use of caustic base or acid, which is a safety hazard when located near water for drinking or agriculture. In some embodiments, if a sodium or chloride pumping membrane has a near-ideal photovoltage and a modest photocurrent, or utilized a tandem structure, then aqueous 50 mM NaCl may be desalinated faster than from ideal solar thermal distillation.

(142) Various modifications of the invention, in addition to those described herein, will be apparent to those skilled in the art from the foregoing description. Such modifications are also intended to fall within fire scope of the appended claims. Each reference cited in the present application is incorporated herein by reference in its entirety.

(143) Although there has been shown and described the preferred embodiment of the present invention, it will be readily apparent to those skilled in the art that modifications may be made thereto which do not exceed the scope of the appended claims. Therefore, the scope of the invention is only to be limited by the following claims. In some embodiments, the figures presented in this patent application are drawn to scale, including the angles, ratios of dimensions, etc. The reference numbers recited in the below claims are solely for ease of examination of this patent application, and are exemplary, and are not intended in any way to limit the scope of the claims to the particular features having the corresponding reference numbers in the drawings. In some embodiments, the figures are representative only and the claims are not limited by the dimensions of the figures. In some embodiments, descriptions of the inventions described herein using the phrase “comprising” includes embodiments that could be described as “consisting essentially of” or “consisting of”, and as such the written description requirement for claiming one or more embodiments of the present invention using the phrase “consisting essentially of” or “consisting of” is met.