High Efficiency Solar Desalination Accompanying Electrocatalytic

Subscriber access provided by CARLETON UNIVERSITY

Article

High Efficiency Solar Desalination Accompanying Electrocatalytic Conversions of Desalted Chloride and Captured Carbon Dioxide Byeong-ju Kim, Guangxia Piao, Seonghun Kim, So Young Yang, Yiseul Park, Dong Suk Han, Hokyong Shon, Michael R Hoffmann, and Hyunwoong Park ACS Sustainable Chem. Eng., Just Accepted Manuscript • DOI: 10.1021/ acssuschemeng.9b02640 • Publication Date (Web): 16 Aug 2019 Downloaded from pubs.acs.org on August 21, 2019

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 33 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Sustainable Chemistry & Engineering

High Efficiency Solar Desalination Accompanying Electrocatalytic Conversions of Desalted Chloride and Captured Carbon Dioxide Byeong-ju Kim,1 Guangxia Piao,1 Seonghun Kim,1 So Young Yang,1 Yiseul Park,2 Dong Suk Han,3 Ho Kyong Shon,4 Michael R. Hoffmann,5 and Hyunwoong Park1,* 1 2

School of Energy Engineering, Kyungpook National University, Daegu 41566, Korea

Department of Chemical Engineering, Pukyong National University, Busan 48513, Korea 3

4

Center for Advanced Materials (CAM), Qatar University, Doha 2713, Qatar School of Civil and Environmental Engineering, University of Technology, Sydney, Post Box 129, Broadway, NSW, 2007, Australia

5

Linde + Robinson Laboratory, California Institute of Technology, Pasadena, CA 91125, U.S.A.

*To whom correspondence should be addressed: (H. Park) E-mail: [email protected]; Tel: +82-53-950-8973

1

ACS Paragon Plus Environment

ACS Sustainable Chemistry & Engineering 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Abstract The sustainability of conventional water- and energy-associated systems is being examined in terms of water–energy nexus. This study presents a high-efficiency, off-grid solar desalination system for saline water (salinity 10 and 36 g L−1) that accompanies electrocatalytic oxidations of chloride and, consequently, urine via oxidized chlorine species, while concomitantly producing formate from captured CO2. A variable number of desalination cell arrays is placed between a double-layered nanoparticulate titania electrocatalyst (Ti/IrxTa1−xOy/nano-TiO2; denoted as n-TEC) anode and a porous dendrite Bi cathode. A potential bias to the n-TEC and Bi pair initiates the transport of chloride and sodium ions in the saline water to the anode and cathode cells, respectively, at an ion transport efficiency of ~100% and a specific energy consumption of ~1.9 kWh m−3. During the desalination, the n-TEC anode catalyzes the conversion of the transported chloride into reactive chlorine species, which in turn mediate the decomposition of urine in the anode cell. Concurrent with the anodic process, formate is continuously produced at a Faradaic efficiency of >95% from the CO2 captured in the catholyte. When a photovoltaic cell (power conversion efficiency of ~18%) is coupled to the stack device with five desalination cells, the three independent processes synergistically proceed at a maximum overall solar-to-desalination system efficiency ~16% and a maximum solar-toformate chemical energy conversion efficiency of ~7%.

Keywords Water-energy nexus; Electrocatalysis; Reactive chlorine species; CO2 reduction; Desalination

2


Page 2 of 33

Page 3 of 33 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Introduction Coastal areas have long been favored as an ideal location for residences and energy production. Currently, 90% of the world’s urban areas are coastal, and 68% of the world’s population has been predicted to live in an urban area by 2050.1 The sustainability of urban areas is essentially based on the secure availability of water and energy. Ironically, most thermochemical power plants (a representative “not in my backyard” facility) are located in coastal areas as well because of the availability of seawater as a heat-exchanging medium. Coal-fired power plants require a large quantity of seawater (e.g., ~40 tons per second for 1 GW-level power generation) while discharging wastewater and emitting more than 1 kg of CO2 per kWh.2 The sustainability of these systems is being examined in terms of water–energy nexus.3-4 The present study shows that such sustainability can be partially achieved by a novel hybrid off-grid electrocatalytic solar desalination system accompanying the oxidation of desalted chloride for the remediation of wastewater and the conversion of captured CO2 to carbon-neutral, valued-added chemicals (Scheme 1). Electrocatalytic systems are effective for the direct production of reactive intermediate species (e.g., OH and O3) and stabilized chemicals (e.g., H2 and H2O2) via oxidation and reduction reactions, respectively.5-6 The coupling of the redox reactions has been shown to work synergistically with renewable electrical grids (i.e., photovoltaic systems) for operation of the overall process.7 Given that most water contains chlorides (>0.3 mM), the chloride oxidation reaction (COR) competes with the water oxidation reaction (E(Cl2/Cl−) = 1.36 V; E(O2/H2O) = 1.23 V; E(OH/H2O) = 2.7 V) and produces reactive chlorine species (RCSs) such as Cl, Cl2−, and HClO/ClO−.8-10 RCSs are powerful oxidants,11 diffusing well into bulk and enabling homogeneous reactions without presorption onto substrates.12-13 Despite this 3



uniqueness, the effects of RCSs become relevant when the chloride concentration is sufficiently high (typically, >5 mM).14-17 Brackish water and seawater can be considered as vast chloride supplies (salinities of 1–10 g L−1 and 36 g L−1, respectively) and can therefore be used for RCSmediated electrocatalysis. With this possibility in mind, we have recently demonstrated a proof-of-concept-level solar desalination system with semiconductor photoanodes (TiO2 and WO3) and Pt cathodes for RCS-mediated water treatment and H2 production, respectively, while desalting saline and seawater.14 Despite a successful demonstration, the system has several limits associated with materials (e.g., the theoretically low photoconversion efficiency of wide-bandgap semiconductors18-19 and structurally ineffective sunlight utilization with vertically standing photoanodes) and electrolytes (e.g., the pH between the anolyte and catholyte build up during the desalination and electrochemical reaction). Consequently, the specific energy consumption (SEC) of the system was 4.4 kWh m−3 for 50% desalination, whereas the energy recovery was as low as ~0.8 kWh m−3.14 To overcome such challenges, we here present our design of a photovoltaics-coupled, electrocatalytic desalination cell-stack system that i) desalinates saline water and seawater in flow-type desalination cells, ii) carries out electrocatalytic COR with concomitant remediation of water with human urine, and iii) captures CO2 and simultaneously converts it into formate (Scheme 1). Urea and urine were chosen as model substrates; urea is the largest component of urines (~0.33 M),20 which are discharged in quantities of 240 million tons per day.17 The concentration of chloride in human urine is typically ~170 mM.20 In addition, formic acid and its salts are the most cost-efficient ($0.4 kWh−1 for formate vs. $0.035 kWh−1 for methanol)21 among CO2 conversion liquid products and are much easier to store and transport than gaseous 4


Page 4 of 33

Page 5 of 33 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


products. They are widely used in preservatives, the rubber and leather industries, and the dyeing and finishing of textiles, reaching a global market of ~$880 million in 2027. For the COR, we synthesized double-layered oxide electrodes composed of an Ir and Ta mixed oxide underlayer and a nanoparticulate TiO2 overlayer (denoted as n-TEC).16,22 For the CO2 reduction reaction (CO2RR), a porous Bi foam was synthesized via electrodeposition. There are several electrocatalyts for CO2RR,23-25 and Bi is considered the most suitable due to its high selectivity for formate production. Other cathodic reactions (e.g., the production of H2 and H2O2) can be considered; however, a high pH (>12) during electrocatalysis is not desirable for such reactions (Faradaic efficiency of ~50%). We used the catholyte as a solvent for CO2 capture and successfully converted the captured CO2 into formate with a Faradaic efficiency of >95% at circum-neutral pH (6.57).

Experimental Synthesis and Characterization of Electrodes Double-layered electrocatalytic oxides deposited onto Ti substrates (Nanopac, Korea) were synthesized following a typical coating–annealing process published elsewhere.8,16,22 In brief, Ti substrates (~2 mm thick) were coated with Ir and Ta at a molar ratio of 0.73/0.27 followed by annealing at ~500 C for 1 h. Then, TiO2 nanoparticles synthesized via a typical sol–gel process were sprayed onto the material as an overcoat, which was then calcined at 450 C for 5 h. Our previous studies on the these materials showed that the underlayer is essential for ohmic contact and durability22 and that the TiO2 overcoat required metal dopants (e.g., Bi, Nb, or Sb) for electrocatalysis.26 However, a further study revealed that dopant-free TiO2 5



exhibited essentially the same performance as the doped one.16,27 As-synthesized doublelayered nanoparticulate TiO2 electrocatalysts (Ti/IrxTa1−xOy/TiO2; denoted n-TEC) were characterized by X-ray photoelectron spectroscopy (XPS, Quantera SXM, ULVAC-PHI) and X-ray diffraction (XRD, D/Max-2500, Rigaku) to examine their elemental composition and crystalline structures, respectively. The morphology of the samples was examined by fieldemission scanning electron microscopy (FE-SEM, S-4800, Hitachi). In addition, field-emission electron probe microanalysis (EPMA, JXA8530F(5CH), JEOL) was used to estimate the composition and thickness of each layer. Porous Bi foam electrodes were synthesized on Cu substrates (Alfa Aesar, 99.9%, 0.254 mm thick) via electrodeposition. Prior to Bi deposition, the Cu substrates were polished with sandpaper (Chayoung, P220) and rinsed with ethanol (Duksan, 99.5%) and deionized water (18 M cm, Human Corp.) followed by sonication in an ultrasonic bath (UC-10, Jeiotech) with HCl (Junse, 35–37%) for 5 min. They were then immersed into aqueous solutions of Bi(NO3)3·5H2O (0.1 M, Junsei, 98%) and H2SO4 (Duksan, 95%), to which a constant current density of 5 A cm−2 was applied for 5 s from a DC power supply (Agilent E3633A).

Electrocatalysis and Chemical Analysis The electrochemical behavior and electrocatalytic performance of as-synthesized electrodes (n-TEC or Bi foam; working electrode, 1 × 1 cm2) was examined using a typical three-electrode configuration with a saturated calomel electrode (SCE, reference electrode) and Pt foil (counter electrode) in air-equilibrated aqueous electrolytes of NaCl and NaClO4 for nTEC- and CO2-equilibrated aqueous solutions (NaOH with and without K2SO4) for Bi foam. The three electrodes were placed in a customized, single-compartment Teflon reactor. Linear 6


Page 6 of 33

Page 7 of 33 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


sweep voltammograms (LSVs) were obtained at a scan rate of 5 mV s−1 using a potentiostat/galvanostat (Ivium). If necessary, the Pt counter electrode was separated from the working electrodes in two-compartment reactors separated with proton exchange membranes (Nafion 117, Chemours). For bulk electrolysis for the decomposition of urea (CH4N2O, Junsei) and CO2 conversion, the n-TEC and Bi foam electrodes were biased at constant potentials of 1.5–1.9 VSCE and −1.5 VSCE, respectively. Desalination-coupled electrocatalysis hybrid reactors were designed (Scheme 1). In a three-cell (batch-type) stack with a basic unit configuration (Scheme 1a), each n-TEC and Bi was placed in the anode cell (AN) with aqueous urea (2 mM, anolyte) and in the cathode cell (CA) with K2SO4 (2 mM, Sigma-Aldrich, 30 mL, catholyte), respectively. The middle cell (i.e., the desalination cell, DS) was filled with saline water (NaCl 0.171 M or 10 g L−1, 20 mL). The anode cell and desalination cell were separated by an anion-exchange membrane (AEM, AMI7001S, Membrane International), and the desalination cell and cathode cell were separated by a cation-exchange membrane (CEM, CMI-7000s, Membrane International). In addition, a flow-type stack reactor in which the number of desalination cells could be varied was fabricated (Scheme 1b). In this expandable stack device, the anode cell and cathode cell were the same as in the batch-type stack, whereas the number of desalination cells containing saline water (NaCl 10 g L−1) or seawater (salinity 36 g L−1) varied between 1 and 5. The desalination cells were connected to each other, and the saline water (total 40 mL) was circulated through the cells at a flow rate of 5 mL min−1. The cells between each two desalination cells were filled with the same saline water (40 mL) that was circulated among the cells. As the desalination proceeded, salts were gradually accumulated in these cells (hereafter denoted as concentration cells, CNs). The overall process was operated by a DC power supply (E3647A, Keysight 7



Technologies) connected to the anode and cathode pair. For off-grid processes, the anode/cathode pair in the stack device with five desalination cells was coupled to a monocrystalline Si photovoltaic array (area of 1.75 cm2; JSC of ~6.23 mA cm−2; VOC of ~4.3 V; fill factor of ~67%; power conversion efficiency of ~18%; M4030-4V, Minisolar) under simulated sunlight (AM 1.5G, 100 mW cm−2, ABET Tech). Three different PV-coupled electrocatalyses were performed: (Case A) n-TEC anode with 2 mM urea / NaCl 10 g L−1 / Bi cathode with 2 mM K2SO4; (Case B) n-TEC anode with urine / seawater (36 g L−1) / Bi cathode with 2 mM K2SO4; (Case C) n-TEC anode with urine / seawater (36 g L−1) / Bi cathode with 0.1 M K2SO4 (purged with CO2). This study used artificial urine and seawater. The urine was synthesized with 0.33 M urea, 5.92 mM creatinine, 33.5 mM NaCl, and 19.2 mM KCl according to the literature28 and was subsequently diluted tenfold; the seawater was prepared with a product from Instant Ocean® (36 g L−1). During electrocatalysis, aliquots in the anode cell were intermittently sampled and quantitatively analyzed for urea. Details of the analytical method for urea are available elsewhere.14 The concentrations of free chlorine and total chlorine (i.e., free chlorine plus combined chlorine) were determined with N,N-diethyl-p-phenylenediamine (DPD) reagent (Hach methods 10101 and 10102). The total organic carbon (TOC) of the anolyte was estimated using a TOC analyzer (TOC-L, Shimadzu). Other anions (Cl− and NO3−) and cations (Na+ and NH4+) were analyzed using ion chromatographs (IC, Thermo Scientific, DIONEX ICS-1100), as described elsewhere.14 Formate was quantified using a high-performance liquid chromatograph (Young Lin 9100) equipped with a C18 inverse column (4.6 mm × 150 mm, Thermo); details of the analysis method are available elsewhere.24,29-30 The Faradaic efficiency (FE), specific energy consumption (SEC, either for 50% 8


Page 8 of 33

Page 9 of 33 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


desalination or for conductivity of 500 S cm−1), ion transport efficiency (ITE), and solar-todesalination (STD) system efficiency were estimated by the following equations: FE (%) = (Amount of formate × 2F) × 100% / (I × t)

(1)

SEC (kWh m−3) = E × I × t / (saline water volume)

(2)

ITE (%) = (Amount of transported monovalent ion) / (I × t)

(3)

STD (%) = E × I × 100% / (100 mW cm−2 × PV area)

(4)

where F, E, I, and t are the Faraday constant (96,485 C mol−1), operating voltage (V), operating current (mA), and time (s), respectively. Finally, the solar-to-formate energy conversion efficiency was estimated on the basis of the reaction Gibbs energy (rG°, 295.46 kJ mol−1) assuming the overall reaction of the COR and the CO2RR (see Supporting Information).

Results and Discussion Electrocatalytic Activities of n-TEC and Bi Electrodes The XRD patterns of the as-synthesized n-TEC samples show a mixed crystalline structure of TiO2 (anatase : rutile = 82 : 18) and IrO2 (101). No evidence for tantalum oxides was found in either the XRD or the XPS analysis (Figure S1 and S2 in Supporting Information). These weak or absent signals of Ir and Ta are attributed to low Ta and Ir contents (XPS detection limit 0.1–1 at%) and/or to the thick TiO2 overlayer limiting beam penetration (typically ~10 nm) to the IrxTa1−xOy underlayer. The cross-sectional SEM image clearly shows a ~1 m-thick interfacial, waved layer within the n-TEC, whereas the top surface in the 6–8 m-thick upper 9



region was uniform (Figure 1a inset and S1). The EPMA analysis further revealed that the bottom region was nearly 100% Ti, whereas the upper region was constituted primarily by Ti and O, indicating that a crystalline TiO2 layer was formed on the Ti substrate. By contrast, Ir, Ta, and O were found only in the interfacial region; Ir was rather uniform, whereas Ta was localized, likely because of its low concentration. The waving Ir and Ta distributions were attributed to the nonuniform Ti substrate surface. The absence of Ir and Ta in the upper region suggests that they did not intrude into the TiO2 overlayer during the calcination stage. The LSVs of the as-synthesized n-TEC electrodes in aqueous sodium chloride (NaCl) and sodium perchlorate (NaClO4) solutions were compared (Figure 1a). No obvious urea effect on the LSVs was observed in either electrolyte (Figure S3), indicating insignificant interaction (i.e., adsorption) of urea with the electrode. Nevertheless, the urea bulk electrolysis was obviously affected by the electrolyte (Figure 1b and S4). In NaClO4, urea was not decomposed at E = 1.5 VSCE for 6 h despite a stable current density of ~1.5 mA cm−2 (Figure S3 inset). By contrast, urea was decomposed in a reaction with pseudo-first-order kinetics (Ct = C0ekt, where k is the rate constant and t is time), with k value of 1.6 h−1 in NaCl at E = 1.5 VSCE, which were 4.8 and 2 times greater than those at E = 1.7 and 1.9 VSCE in NaClO4, respectively. Assuming that the urea undergoes a one-electron oxidation reaction, the Columbic efficiency (i.e., k/I) with NaCl is 0.33 C−1 at 1.5 V, which is ~12 and 7 times greater than those with NaClO4 at 1.7 and 1.9 VSCE, respectively. This chloride-enhanced kinetics is attributed to the RCS-mediated decomposition of urea in solution bulk (i.e., homogeneous oxidation).14,31 RCSs are mobile and can be converted to chlorides upon reduction on a counter electrode (i.e., cathode),8,12 retarding the decomposition of the urea.17 This retardation effect was confirmed by a decrease in the k value (0.17 h−1) by a factor of 9 when the membrane between the working and counter 10


Page 10 of 33

Page 11 of 33 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


electrodes was removed and the reaction proceeded in a single cell (SC in Figure 1b).7-8 The Faradaic efficiency (FE) of the RCS generation in the single cell was estimated to be 30-50%. The RCS-mediated urea decomposition was studied in greater detail with 50 mM NaCl. At 1.5 VSCE, urea was completely removed in 2 h (Figure 1c), the kinetics of which was 2.6fold faster than that in 20 mM NaCl (k values of 4.2 and 1.6 h−1 in 50 and 20 mM NaCl, respectively) because of the greater chloride concentration. Total chlorine was evolved immediately after electrocatalysis began; its quantity reached the maximum in 4 h (Phase I: formation of chloro-organic compounds, including N-chlorinated urea, at ~1.4 mM), decreased until ~7 h (Phase II: decomposition of the chloro-organic compounds), and then increased again (Phase III: addition of formed free chlorine to nondecomposed chloro-organic compounds of ~0.5 mM). This N-shaped curve is typically observed in the chlorination process, particularly in the case of ammonia,9,32 where the breakpoint between Phases II and III is attributed to the formation of free chlorine. Consistent with this hypothesis, free chlorine was detected only after the breakpoint, reaching the same concentration as the total chlorine. The electrocatalysis times for the breakpoint were slightly different among tests (Figure S5), likely because of differences in performance of the synthesized n-TEC electrodes; however, the overall tendency was the same. The TOC of urea did not change until the time (~2 h) for the completion of the urea chlorination and the production of approximately half the peak total chlorine (virtually equivalent to chloro-organic compounds in the absence of free chlorine). On the other hand, the TOC approached zero soon after the breakpoint (~7 h), in agreement with the evolution of free chlorine. Although ammonia is a typical intermediate in the urea decomposition and nitrate is often found as a final decomposition product of ammonia,14,17,33 neither ammonia nor nitrate was found even after prolonged electrocatalysis beyond the time of zero TOC. Ammonia 11



appears to be transient and/or present at concentrations below the detection limit. The reaction mechanism is discussed in detail in the Supporting Information (Figure S6). As a cathode, Bi was electrodeposited onto Cu foils. As-synthesized Bi samples showed a highly crystallized rhombohedral structure (Figure S7). Notably, they displayed a highly porous structure with ~50 m-pores due to the H2 bubble formation during the reductive synthesis, whereas the main framework of Bi has a dendrite-like configuration (Figure 1a inset; see Figure S8 for more images). This structure should be beneficial for mass transfer at gas– liquid–solid interfaces. For the CO2RR, bicarbonate (HCO3−) is widely used because of its buffering effect and some synergistic effects.23-24,29,34 However, the present study used NaOH (0.1 M, pH ~12.8) since the solution pH was previously found to increase to ~13 owing to proton-coupled cathodic reactions as Na+ was accumulated during electrodesalination processes.14 When CO2 was purged, the solution pH decreased to ~6.7 (Figure 1d inset) because of consumption of hydroxide (CO2 + OH−  HCO3−). The Bi electrodes showed an Eon of ca. −1.1 VSCE (i.e., −0.46 VRHE) (Figure 1a) and produced formate with a Faradaic efficiency (FE) of >90% at E = −1.5 VSCE (Figure 1d). This activity is comparable to those reported in the literature.35-36 However, coupling the CO2RR and the anodic reaction (i.e., COR) requires an overall voltage 2.5 V. To reduce the Eon of the CO2RR, K2SO4 was added onto NaOH. K2SO4 is a low-cost, noncarbonaceous chemical (i.e., no carbon source for formate); in addition, the K+ ion can promote the CO2RR because it is larger than the Na+ ion.37-38 With the addition of K2SO4, the Eon shifted to ca. 0.9 VSCE and formate production was enhanced (Figure 1d). However, the FE was slightly decreased to ~80% because of the low pH of 5.5– 6., which decreased the fraction of HCO3− beneficial for the CO2RR.39-40

12


Page 12 of 33

Page 13 of 33 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Desalination-Coupled Electrocatalysis The electrocatalytic redox reactions with n-TEC and Bi electrode pairs were coupled with the desalination process of saline water (NaCl 0.171 M or 10 g L−1) by placing the desalination cell (DS) between an anode cell (AN with 2 mM urea) with the n-TEC and a cathode cell (CA with 2 mM K2SO4) with the Bi (Scheme 1a). To evaluate the anodic reactions in the three-cell (batch-type) stack, a constant potential of 1.5 VSCE (Ea) was applied to the nTEC (as a working electrode; hence Bi functioned as a counter electrode and virtual cathode) (Figure 2a). I was trace in the initial stage of ~2 h and gradually increased with time; the rate of increase in I was greater with time, suggesting that the reaction accelerated. The I reached a plateau at ~9 mA in ~13 h and then suddenly decreased to zero at ~17 h (shutdown point). Meanwhile, the overall stack voltage (i.e., potential difference between n-TEC and Bi electrodes; Estack) was maintained at 2.5 V and abruptly increased beyond the voltage limit (12 V) of the multimeter in ~17 h. During this process, ions (Cl− and Na+) in the desalination cell were transported to the anode and cathode cells, respectively (Figure 2b). The equal Cl− in the anode cell and desalination cell confirmed the quantitative transport of chloride from the desalination cell to the anode cell (the same was true for Na+). Notably, the Cl− was trace in the initial stage and accelerated with time. The similar time profiles of Cl− and I suggest that the ion transport was attributable solely to I (i.e., an electrochemical process) and that the trace I in the initial stage resulted from low ionic concentrations in the anode and cathode cells. The ion transport efficiency (ITE) of ~100% confirms this hypothesis. As the ionic concentration decreases in the desalination cell and increases in the anode and cathode cells, the overall solution conductivity should change with time. At the shutdown point (~17 h), the chloride concentrations in the desalination cell and anode cell were ~7.65 mM and ~109 mM, 13



respectively, indicating that the process shutdown was attributable to the low conductivity in the desalination cell. The pH of the saline water remained unchanged at pH ~6 (Figure S10). During the first 2 h, the urea concentration changed insignificantly (Figure S11) because of the trace Cl− (~50 mol; 1.7 mM). However, after 4 h (Cl− ≈ 150 mol; 5.0 mM), >80% of the urea was decomposed; in 8 h, virtually all of the urea was decomposed. The total chlorine then reached a plateau and decreased to coincide with the free chlorine in ~14 h, while the FE of the RCS generation was ~15%. The TOC began to decrease before the plateau of the total chlorine production and became zero as the amounts of total and free chlorines became similar. Neither ammonia nor nitrate was found beyond 18 h. These behaviors of urea, total and free chlorines, and TOC were the same as in the electrolysis of urea presented in Figure 1c. In contrast to the anodic reaction, the CO2RR with the desalination occurred differently from that without the desalination (Figure 1d vs. 2c). The formate production with desalination was trace, and its FE was poor (~10%) until ~9 h. Despite an increase in the FE to ~40% in the late stage, the low activity persisted until ~90% desalination of saline water (0.171 M NaCl). The cathode potential (Ec, estimated by subtracting Ea from Estack) varied between −1.5 and −1.8 VSCE, which was sufficient to drive the CO2RR. Accordingly, the poor CO2RR should be attributed to the catholyte, not the cathode potential. Most of the CO2 was apparently hydrolyzed to bicarbonate (HCO3−) in the initial stage; the formation of HCO3− consumed hydroxide ions and decreased the catholyte pH (Figure 2d). After the pH reached ~7 in ~12 h, the purged CO2 participated in the reduction reaction on the Bi. Given the solubility of CO2 in aqueous solutions (~33 mM), the bicarbonate formed from the purged CO2 should enhance the catholyte conductivity as well as catalyze the CO2RR. For a controlled CO2RR, a constant potential of −1.5 VSCE was applied to the Bi 14


Page 14 of 33

Page 15 of 33 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


electrode (as a working electrode; hence n-TEC functioned as a counter electrode and virtual anode) as the catholyte (0.1 M K2SO4) was purged with CO2 (Figure 2e). The increase in conductivity of electrolytes resulted in several positive effects. First, the overall performance of the stack was improved, with a constant I of ~6 mA. Accordingly, the formate production started immediately after the electrolysis (i.e., no induction period) and an FE >90% was maintained (Figure 2c). Second, the shutdown time of the device was ~13 h, indicating that less time was required for the complete desalination (Figure 2a vs. 2e). In consistent to this, the time for zero conductivity (i.e.,  < 500 S cm−1) of saline water decreased from 20 h (2 mM K2SO4) to 14 h (0.1 M K2SO4) (Figure 2f). Finally, the decomposition of urea occurred without delay because of faster chloride transport from the desalination cell to the anode cell (Figure 2f).

Off-Grid Desalination with Urine Remediation and CO2 Reduction Although the aforementioned batch-type three-cell stack was used to successfully demonstrated a simultaneous operation of the three independent processes, the specific energy consumption (SEC) for the 50% desalination (~9.0 kWh m−3) was estimated to be much higher than that of state-of-the-art reverse osmosis desalination (~1.8 kWh m−3).4 The contributions of the energy saved for water treatment and the energy gain by chemicals (e.g., H2) to the SEC were estimated to be beneficial (~0.12 and 1 kWh m−3, respectively).14 To reduce the SEC, we designed an expandable flow-type stack with desalination cell arrays (Scheme 1b).41 For comparison, a DC voltage of 3.5 V (i.e., EDC = 3.5 V) was applied on the basis of the batchtype stack performance (Figure S11). The overall process tendency for 1-DS was similar to the batch-type in terms of the time-profiled changes in I, ionic concentration (proportional to ) in 15



the desalination cell, and urea decomposition in the anolyte, leading to an SEC of ~10 kWh m−3 for 50% desalination (Figure 3a). With increasing DS number, I values were substantially reduced because of proportional increases in the device resistance resulting from membrane (resistance of each membrane: max. 40 /sq) and distance between the anode and cathode (Figure S12). Despite the reduced I values, the decrease in  of the saline water (i.e., the degree of desalination) became faster with increasing DS number (Figure 3b) because of an enlarged interface between the desalination cell and the concentration cell. As a result, the ITE increased from ~97% to ~270 and ~450% for 3- and 5-DS, respectively (Figure 3b inset), and the SEC substantially decreased to ~3 and ~1.9 kWh m−3, respectively. A slight decrease in the normalized ITE with 3- and 5-DS (270%/3 = 450%/5 = 90%) was attributed to the slow intercell ion transport. However, the urea decomposition rate in the anolyte was nearly unaltered (k ~ 0.69, 0.73, and 0.76 h1 for 1-, 3-, and 5-DS, respectively) (Figure 3a inset) because i) the urea decomposition proceeded predominantly by RCSs and ii) the rate of chloride transport to the anode cell was not strongly influenced by the number of DS (i.e., the similar normalized ITE values). The as-fabricated stack with 5-DS was connected with a commercially available, lowcost monocrystalline Si PV cell for stand-alone operation of the overall process. To match the power required for the 5-DS stack (3.5 VDC and max. ~5 mA cm−2; see Figure 3a), the employed PV (power conversion efficiency of ~18%; see Figure 3c inset) was re-sized. Three different conditions were considered (denoted as Cases A, B, and C). In Case A (2 mM urea / 10 g L−1 NaCl / 2 mM K2SO4 as an anolyte, saline water, and catholyte, respectively), coupling the PV and the n-TEC/Bi electrode pair resulted in an Estack of ~4.2 V, which then decreased to ~3.9 V (Figure 3c). I changed with time following a volcano-fashion with peak at ~6 mA in 16


Page 16 of 33

Page 17 of 33 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


~3.5 h, which is similar to the corresponding plot for the DC-powered system (Figure 3a). The conductivity of saline water in the desalination cell decreased monotonically with time, reaching ~8.7 mS cm−1 (50% desalination) and ~0.5 mS cm−1 in ~5 and ~9 h, respectively (Figure 3d). The decomposition of urea was completed in 3 h without production of any Ncompound (ammonia or nitrate) during prolonged electrocatalysis (Figure S13). All of these behaviors were the same as those for the DC-powered system, indicating the successful operation with a PV under simulated sunlight (AM 1.5; 100 mW cm−2). The SEC of the PV system (~2 kWh m−3) was similar to that of the DC-powered 5-DS stack (~1.9 kWh m−3). Finally, the PV-hybrid stack was tested for human urine decomposition and simultaneous seawater desalination (Case B: urine / seawater (36 g L−1) / 2 mM K2SO4 as the anolyte, saline water, and catholyte, respectively). The overall tendency of Case B was the same as that of Case A. However, Case B showed a larger I with a smaller Estack (Figure 3c) because of the higher solution conductivities in the anolyte and saline water. The solution conductivity of seawater decreased linearly with time, and its rate of decrease was nearly the same as that in Case A (Figure 3d). In Case C (urine with 0.1 M K2SO4 / seawater (36 g L1) / 0.1 M K2SO4 with CO2 purging, as an anolyte, saline water, and catholyte, respectively), stabilized I and Estack were obtained from the initial stage, whereas the desalination rate was nearly the same as that of Case B (except for the initial period to ~5 h). A comparison of Cases A, B, and C indicates that the desalination proceeds solely via the electrochemical process and that the effect of the solution conductivity difference on the ion transport was insignificant. The advantage of the concentrated electrolytes (Case C) is accelerated ion transport even in the initial stage. This promoted the COR and urea decomposition in the anode cell (Figure S14), and led to linear production of formate in the cathode cell at a Faradaic efficiency close to ~100% 17



(Figure 3e). The production of ammonia in Case C appears to result from a higher concentration of urea ([urea in urine]0 ~33 mM) than the concentration of RCSs (Figure S15). The coupling of the COR and the CO2RR enabled us to estimate a solar-to-formate energy conversion efficiency of ~7% in the initial stage and ~5% in the final stage (see Supporting Information). In addition, the overall solar-to-desalination (STD) device efficiency for Case A was estimated to gradually increase from ~4% in the initial stage to ~10% in 5–6 h (50% desalination), reaching ~9% at nearly 100% desalination ( = 0.5 mS cm−1 in ~9.43 h; Figure 3f). The STD efficiency of Case C was highest at ~16% in the initial stage and decreased to ~13% in the final stage, similar to the STD efficiency of Case B. These results indicate the present device can use a maximum of ~90% of PV electricity for the operation of triple-hybrid reactions.

Conclusions We developed a multicell stack desalination device composed of an anode cell for the chloride oxidation reaction, cathode cell for the CO2 reduction reaction, and desalination cells (with concentration cells). In the electrocatalytic remediation of urea (in urine) as a model water pollutant, double-layered n-TEC electrodes converted chloride into RCSs that efficiently mediate the urea oxidation. As the TOC of urea decreased, neither ammonia nor nitrate was identified as a byproduct, indicating the virtual mineralization of urea and its chlorinated intermediates into CO2 and N2. Concurrent with the anodic reaction, the as-synthesized porous Bi electrodes produced formate at a Faradaic efficiency greater than 90%. When the n-TEC and Bi pair were coupled in desalination/concentration cell arrays in the configuration of an 18


Page 18 of 33

Page 19 of 33 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


expandable flow-type stack, the three independent processes (desalination, urea decomposition, and formate production) synergistically proceeded at a specific energy consumption of ~1.9 kWh m−3 for 50% desalination. Such hybrid processes were successfully operated upon coupling with a low-cost PV, leading to a maximum overall solar-to-desalination device efficiency of~16% and to a maximum solar-to-formate energy conversion efficiency of ~7%. The uniqueness of the as-designed system is that the kinetics of the urea decomposition is independent of the reactor configuration, whereas the overall ion transport efficiency increases almost linearly with the number of desalination cells. Furthermore, the catholyte is found to be used as a CO2 capturing solution in which formate as a value-added chemical is simultaneously produced at high Faradaic efficiency. This hybrid system can be applied to seaside towns suffering from limited access to drinking water and water reuse systems, as well as to CO2emitting facilities located in coastal areas (e.g., coal-fired power plants discharging low-salinity wastewater used for/by residents and seawater used as a heat-exchanging water, and emitted CO2).

 Associated Content Supporting Information The Supporting Information is available free of charge on the ACS Publications website at DOI: XXXX. Efficiency estimation, XRD, XPS, SEM, EPMA, electrocatalysis, desalination, hybrid reactions, and reactor photo.

19



 Authors Information Corresponding Author *E-mail: [email protected] (H. Park) ORCID Byeong-ju Kim: 0000-0002-0807-4878 Guangxia Piao: 0000-0003-4415-5461 Seonghun Kim: 0000-0001-5474-4249 So Young Yang: 0000-0002-1783-0276 Yiseul Park: 0000-0003-3256-1032 Dong Suk Han: 0000-0002-4804-5369 Ho Kyong Shon: 0000-0002-3777-7169 Michael R. Hoffmann: 0000-0001-6495-1946 Hyunwoong Park: 0000-0002-4938-6907 Author Contributions B.K., G.P., S.K. and H.P. designed and performed experiments, and S.Y.Y., Y.P., D.S.H., H.K.S. and M.R.H. discussed the experimental results. H.P. wrote this manuscript.

Notes The authors declare no competing financial interest. 20


Page 20 of 33

Page 21 of 33 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


 Acknowledgement We are grateful to the Korea CCS R&D Center (KCRC) (no. 2014M1A8A1049354) for financial support. This research was partly supported by the National Research Foundation of Korea (2019R1A2C2002602, 2018R1A6A1A03024962, and 2019M1A2A2065616) and the Korea Ministry of Environment as Waste to Energy-Recycling Human Resource Development

Project

(YL-WE-17-001).

S.Y.

Yang

is

grateful

to

the

NRF

(2017R1C1B1005179). Y. Park is grateful to the Next-Generation Carbon Upcycling Project (2017M1A2A2043123). This publication was made possible by a grant from the Qatar National Research Fund under its National Priorities Research Program (NPRP 101210-160019).

21



References 1. World Urbanization Prospects: The 2018 Revision, Department of Economic and Social Affairs/Population Division, United Nations; 2018. 2. Environment Baseline, Volume 1: Greenhouse Gas Emissions from the U.S. Power Sector, Department of Energy, U.S.A.; 2016. 3. The Water-Energy Nexus: Challenges and Opportunities, Department of Energy, U.S.A., 2014. 4. Elimelech, M.; Phillip, W. A. The future of seawater desalination: Energy, Technology, and the Environment. Science 2011, 333, 712-717, DOI 10.1126/science.1200488. 5. Panizza, M.; Cerisola, G. Direct and mediated anodic oxidation of organic pollutants. Chem. Rev. 2009, 109, 6541-6569, DOI 10.1021/cr9001319. 6. Brillas, E.; Sires, I.; Oturan, M. A. Electro-fenton process and related electrochemical technologies based on fenton’s reaction chemistry. Chem. Rev. 2009, 109, 6570-6631, DOI 10.1021/cr900136g. 7. Park, H.; Vecitis, C. D.; Choi, W.; Weres, O.; Hoffmann, M. R. Solar-powered production of molecular hydrogen from water. J. Phys. Chem. C 2008, 112, 885-889, DOI 10.1021/jp710723p. 8. Park, H.; Vecitis, C. D.; Hoffmann, M. R. Electrochemical water splitting coupled with organic compound oxidation: the role of active chlorine species. J. Phys. Chem. C 2009, 113, 7935-7945, DOI 10.1021/jp810331w. 9. White, G. C. Handbook of Chlorination. 2nd ed.; Van Nostrand-Reinhold: New York, 1985. 10. Deborde, M.; von Gunten, U. Reactions of chlorine with inorganic and organic compounds during water treatment - Kinetics and mechanisms: A critical review. Water Res 2008, 42, 13-51, DOI 10.1016/j.watres.2007.07.025. 11. Wardman, P. Reduction potentials of one-electron couples involving free radicals in aqueous solution. J. Phys. Chem. Ref. Data 1989, 18, 1637-1755, DOI 10.1063/1.555843. 12. Park, H.; Vecitis, C. D.; Hoffmann, M. R. Solar-powered electrochemical oxidation of organic compounds coupled with the cathodic production of molecular hydrogen. J. Phys. Chem. A 2008, 112, 7616-7626, DOI 10.1021/jp802807e. 13. Kim, S.; Choi, S. K.; Yoon, B. Y.; Lim, S. K.; Park, H. Effects of electrolyte on the electrocatalytic activities of RuO2/Ti and Sb-SnO2/Ti anodes for water treatment. Appl. Catal., B 2010, 97, 135-141, DOI 10.1016/j.apcatb.2010.03.033. 14. Kim, S.; Piao, G.; Han, D. S.; Shon, H. K.; Park, H. Solar desalination coupled with water remediation and molecular hydrogen production: A novel solar water-energy nexus. Energy Environ. Sci. 2018, 11, 344-353, DOI 10.1039/C7EE02640D. 15. Yang, S. Y.; Kim, D.; Park, H. Shift of the reactive species in the Sb-SnO2electrocatalyzed inactivation of E-coli and degradation of phenol: effects of nickel doping and electrolytes. Environ. Sci. Technol. 2014, 48, 2877-2884, DOI 10.1021/es404688z. 16. Cho, K.; Hoffmann, M. R. BixTi1–xOz Functionalized Heterojunction Anode with an Enhanced Reactive Chlorine Generation Efficiency in Dilute Aqueous Solutions. Chem. Mater. 2015, 27, 2224-2233, DOI 10.1021/acs.chemmater.5b00376. 17. Kim, J.; Choi, W. J. K.; Choi, J.; Hoffmann, M. R.; Park, H. Electrolysis of urea and urine for solar hydrogen. Catal. Today 2013, 199, 2-7, DOI 10.1016/j.cattod.2012.02.009. 18. Jacobsson, T. J. Photoelectrochemical water splitting: An idea heading towards 22


Page 22 of 33

Page 23 of 33 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


obsolescence? Energy Environ. Sci. 2018, 11, 1977-1979, DOI 10.1039/C8EE00772A. 19. Chen, Z.; Dinh, H. N.; Miller, E. Photoelectrochemical Water Splitting: Standards, Experimental Methods, and Protocols. Springer: 2013, DOI 10.1007/978-1-4614-8298-7. 20. Putnam, D. F. Composition and Concentrative Properties of Human Urine, NASA Contractor Report (NASA CR-1802); 1971. 21. Sridhar, N.; Hill, D. Carbon Dioxide Utilization. Electrochemical Conversion of CO2 - Opportunities and Challenges, DNV; 2011. 22. Park, H.; Bak, A.; Ahn, Y. Y.; Choi, J.; Hoffmannn, M. R. Photoelectrochemical performance of multi-layered BiOx-TiO2/Ti electrodes for degradation of phenol and production of molecular hydrogen in water. J. Hazard. Mater. 2012, 211, 47-54, DOI 10.1016/j.jhazmat.2011.05.009. 23. Deb Nath, N. C.; Choi, S. Y.; Jeong, H. W.; Lee, J.-J.; Park, H. Stand-alone photoconversion of carbon dioxide on copper oxide wire arrays powered by tungsten trioxide/dye-sensitized solar cell dual absorbers. Nano Energy 2016, 25, 51-59, DOI 10.1016/j.nanoen.2016.04.025. 24. Choi, S. K.; Kang, U.; Lee, S.; Ham, D. J.; Ji, S. M.; Park, H. Sn-coupled p-Si nanowire arrays for solar formate production from CO2. Adv. Energy Mater. 2014, 4, 1301614, DOI 10.1002/aenm.201301614. 25. Lee, C. H.; Kanan, M. W. Controlling H+ and CO2 reduction selectivity on Pb electrodes. ACS Catal. 2015, 5, 465-469, DOI 10.1021/cs5017672. 26. Kesselman, J. M.; Weres, O.; Lewis, N. S.; Hoffmann, M. R. Electrochemical production of hydroxyl radical at polycrystalline Nb-doped TiO2 electrodes and estimation of the partitioning between hydroxyl radical and direct hole oxidation pathways. J. Phys. Chem. B 1997, 101, 2637-2643, DOI 10.1021/jp962669r. 27. Huang, X.; Qu, Y.; Cid, C. A.; Finke, C.; Hoffmann, M. R.; Lim, K.; Jiang, S. C. Electrochemical disinfection of toilet wastewater using wastewater electrolysis cell. Water Res. 2016, 92, 164-172, DOI 10.1016/j.watres.2016.01.040. 28. Khan, L. B.; Read, H. M.; Ritchie, S. R.; Proft, T. Artificial urine for teaching urinalysis concepts and diagnosis of urinary tract infection in the medical microbiology laboratory. J. Microbiol. Biol. Educ. 2017, 18, DOI 10.1128/jmbe.v18i2.1325. 29. Kang, U.; Park, H. A facile synthesis of CuFeO2 and CuO composite photocatalyst films for production of liquid formate from CO2 and water over a month. J. Mater. Chem. A 2017, 5, 2123-2131, DOI 10.1039/C6TA09378G. 30. Kang, U.; Yoon, S. H.; Han, D. S.; Park, H. Synthesis of aliphatic acids from CO2 and water at efficiencies close to the photosynthesis limit using mixed copper and iron oxide films. ACS Energy Lett. 2019, DOI 10.1021/acsenergylett.9b01281. 31. Ahn, Y. Y.; Yang, S. Y.; Choi, C.; Choi, W.; Kim, S.; Park, H. Electrocatalytic activities of Sb-SnO2 and Bi-TiO2 anodes for water treatment: Effects of electrocatalyst composition and electrolyte. Catal. Today 2017, 282, 57-64, DOI 10.1016/j.cattod.2016.03.011. 32. Pressley, T. A.; Bishop, D. F.; Roan, S. G. Ammonia-Nitrogen Removal by Breakpoint Chlorination. Environ. Sci. Technol. 1972, 7, 622-628, DOI 10.1021/es60066a006. 33. Cho, K.; Kown, D.; Hoffmann, M. R. Electrochemical treatment of human waste coupled with molecular hydrogen production. RSC Adv. 2014, 4, 4596-4608, DOI 10.1039/C3RA46699J. 34. Kang, U.; Choi, S. K.; Ham, D. J.; Ji, S. M.; Choi, W.; Han, D. S.; Abdel-Wahabe, A.; Park, H. Photosynthesis of formate from CO2 and water at 1% energy efficiency via copper iron oxide catalysis. Energy Environ. Sci. 2015, 8, 2638-2643, DOI 10.1039/c5ee01410g. 23



35. Han, N.; Wang, Y.; Yang, H.; Deng, J.; Wu, J.; Li, Y.; Li, Y. Ultrathin bismuth nanosheets from in situ topotactic transformation for selective electrocatalytic CO2 reduction to formate. Nature Commnun. 2018, 90, 1320, DOI 10.1038/s41467-018-03712-z. 36. Kim, S.; Dong, W. J.; Gim, S.; Sohn, W.; Park, J. Y.; Yoo, C. J.; Jang, H. W.; Lee, J.-L. Shape-controlled bismuth nanoflakes as highly selective catalysts for electrochemical carbon dioxide reduction to formate. Nano Energy 2017, 39, 44-52, DOI 10.1016/j.nanoen.2017.05.065. 37. Ayemoba, O.; Cuesta, A. Spectroscopic evidence of size-dependent buffering of interfacial pH by cation hydrolysis during CO2 electroreduction. ACS Appl. Mater. Interfaces 2017, 9, 27377-27382, DOI 10.1021/acsami.7b07351. 38. Singh, M. R.; Kwon, Y.; Lum, Y.; Ager, J. W., III.; Bell, A. T. Hydrolysis of electrolyte cations enhances the electrochemical reduction of CO2 over Ag and Cu. J. Am. Chem. Soc. 2016, 138, 13006-13012, DOI 10.1021/jacs.6b07612. 39. Dunwell, M.; Lu, Q.; Heyes, J. M.; Rosen, J.; Chen, J. G.; Yan, Y.; Jiao, F.; Xu, B. The central role of bicarbonate in the electrochemical reduction of carbon dioxide on gold. J. Am. Chem. Soc. 2017, 139, 3774-3783, DOI 10.1021/jacs.6b13287. 40. Hursan, D.; Janaky, C. Electrochemical reduction of carbon dioxide on nitrogen-doped carbons: Insights from isotopic labeling studies. ACS Energy Lett. 2018, 3, 722-723, DOI 10.1021/acsenergylett.8b00212. 41. Kim, Y.; Logan, B. E. Microbial desalination cells for energy production and desalination. Desalination 2013, 308, 122-130, DOI 10.1016/j.desal.2012.07.022.

24


Page 24 of 33

Page 25 of 33 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Scheme Caption Scheme 1. Schematic of the desalination–electrocatalysis hybrid systems ((a) DC-powered batch-type stack with one desalination cell and (b) PV-coupled flow-type stack with five desalination cells). In (a), an n-TEC (Ti/IrxTa1−xOy/TiO2) anode and porous Bi cathode are placed in two different cells and saline water is placed in a middle cell (i.e., desalination cell). An electrical bias initiates the desalination, transporting anions (i.e., chloride) from the desalination cell to the anode cell via an anion-exchange membrane (AEM) and cations (i.e., sodium) to the cathode cell via a cation-exchange membrane (CEM). As the chloride is gradually enriched in the anolyte, the n-TEC anode oxidizes chloride to reactive chlorine species (RCS), which mediate the decomposition of urine in the anolyte. Similar to the anodic process, the accumulation of sodium ions in the catholyte increases the solution conductivity and simultaneously produces NaOH at pH 13. When the catholyte is purged with CO2, the catholyte pH decreases to ~7 and formate is produced via the CO2 reduction reaction. In (b), desalination cells are connected to each other and the saline water is circulated within them. As the desalination proceeds, ions are accumulated in the cells (i.e., concentration cells) located between each two desalination cells (hence, #concentration cells = #desalination cells − 1). The concentration cells are connected to each other. In principle, the transfer of one electron via an electrical circuit induces the transport of a number of ions equal to the number of desalination cells. Note that the initial salinity of the desalination cell and the concentration cell is the same before the reaction.

Figure Captions Figure 1. (a) Linear sweep voltammograms of as-synthesized n-TEC electrodes in aqueous solutions of NaCl and NaClO4 (each 20 mM) and of Bi foam electrodes in aqueous solutions of NaOH (0.1 M) and NaOH + K2SO4 (each 0.05 M) purged with CO2 (at equilibrated pH values of ~6.7 and ~5.5, respectively; see Figure 1d inset). The upper inset shows a crosssectional SEM image and the elemental mappings of n-TEC (see Figure S1 for more details of the surface characterization of n-TEC); the bottom inset displays the SEM images (top view) of Bi foams. (b) Electrocatalytic decomposition of urea (2 mM) in aqueous NaCl and NaClO4 25



electrolytes (each 20 mM) using n-TEC electrodes maintained at 1.5–1.9 V vs. SCE. See Figure S4 for more data. (c) Bulk electrolysis of urea (2 mM) using n-TEC at E = 1.5 V vs. SCE in aqueous NaCl (50 mM) and simultaneous time changes of (total and free) chlorines and TOC. (d) Electrocatalytic formate production from CO2 and the corresponding Faradaic efficiencies (FEs) (inset: changes in pH during electrocatalysis) using Bi foam electrodes at −1.5 V vs. SCE in CO2-purged aqueous solutions of NaOH (0.1 M) and NaOH + K2SO4 (each 0.05 M).

Figure 2. Desalination–electrocatalysis hybrid reactions using n-TEC anode and Bi cathode pairs in three-cell (batch-type) stacks (see Scheme 1a). (a) Time-profiled changes in the current (I), potential (vs. SCE; Ea for the anode and Ec for the cathode), and the stack voltage (Estack). The n-TEC anode (with 2 mM urea) was maintained at 1.5 V vs. SCE, and the potential biased to the Bi cathode (with 2 mM K2SO4 purged with CO2) was estimated from the difference between the Estack and the Ea. An aqueous solution of NaCl at 0.171 M (10 g L−1) was used as saline water. (b) Changes in the amounts of ions (Cl− and Na+) in the anode cell (AN, 30 mL), cathode cell (CA, 30 mL), and desalination cell (DS, 20 mL) during the hybrid reactions of Figure 2a. In addition, the ion transport efficiency (ITE) was estimated for Cl− and Na+. See Figure S11 for the changes in the concentrations of urea, total chlorine, free chlorines, and TOC as functions of time. (c) Formate production and its Faradaic efficiency in the cathode cell with 2 mM and 0.1 M K2SO4 purged with CO2 (conditions in Figure 2a and 2e, respectively). (d) pH changes in catholytes containing 2 mM K2SO4 with and without CO2 purging. (e) Changes in the current (I), potentials (vs. SCE; Ea for the anode and Ec for the cathode) and the stack voltage (Estack) as functions of time. The Bi cathode (with 0.1 M K2SO4 purged with CO2) was maintained at −1.5 V vs. SCE, whereas the potential at the n-TEC anode (with 2 mM urea + 0.1 M K2SO4) was estimated by difference between the Estack and the Ec. An aqueous solution of NaCl at 0.171 M (10 g L−1) was used as saline water. (f) Saline water conductivity () in the desalination cell and urea concentration in the anode cell under the conditions in Figure 2a and 2e.

26


Page 26 of 33

Page 27 of 33 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 3. (a and b) Electrodesalination–electrocatalysis hybrid reactions in a DC-powered multidesalination cell-stack with n-TEC anode and Bi cathode pairs at 3.5 VDC. 1-, 3-, and 5DS represent the numbers of desalination cells. Anolyte: 2 mM urea; catholyte: 2 mM K2SO4; desalination cell: 0.171 M NaCl. Time-profiled changes in (a) current and (b) solution conductivity () of the saline water with different numbers of desalination cells are shown. The insets in (a) and (b) show the change in urea concentration with time and the averaged ion transport efficiency (ITE), respectively. The numbers on the current profiles in (a) refer to the specific energy consumption (SEC, kW m−3) for 50% desalination. (c–f) Three case operations of a PV-coupled 5-DS stack device (see Scheme 1b). Case A: 2 mM urea / 0.171 M NaCl / 2 mM K2SO4. Case B: Urine (1/10-diluted) / seawater (36 g L−1) / 2 mM K2SO4. Case C: Urine (1/10-diluted) with 0.1 M K2SO4 / seawater (36 g L−1) / 0.1 M K2SO4. (c) Time-profiled changes in current and voltage of the device (inset: IV curve of the employed PV; area of 1.75 cm2, ISC of ~6.23 mA cm−2, VOC of ~4.3 V, fill factor of ~67%, and power conversion efficiency of ~18%). (d) Changes in the conductivity of saline water and ITE (inset). (e) Formate production and Faradaic efficiency. (f) Solar-to-desalination efficiency.

27



Scheme 1. Schematic of the desalination–electrocatalysis hybrid systems ((a) DC-powered batch-type stack with one desalination cell and (b) PV-coupled flow-type stack with five desalination cells). In (a), an n-TEC (Ti/IrxTa1−xOy/TiO2) anode and porous Bi cathode are placed in two different cells and saline water is placed in a middle cell (i.e., desalination cell). An electrical bias initiates the desalination, transporting anions (i.e., chloride) from the desalination cell to the anode cell via an anion-exchange membrane (AEM) and cations (i.e., sodium) to the cathode cell via a cation-exchange membrane (CEM). As the chloride is gradually enriched in the anolyte, the n-TEC anode oxidizes chloride to reactive chlorine species (RCS), which mediate the decomposition of urine in the anolyte. Similar to the anodic process, the accumulation of sodium ions in the catholyte increases the solution conductivity and simultaneously produces NaOH at pH 13. When the catholyte is purged with CO2, the catholyte pH decreases to ~7 and formate is produced via the CO2 reduction reaction. In (b), desalination cells are connected to each other and the saline water is circulated within them. As the desalination proceeds, ions are accumulated in the cells (i.e., concentration cells) located between each two desalination cells (hence, #concentration cells = #desalination cells − 1). The concentration cells are connected to each other. In principle, the transfer of one electron via an electrical circuit induces the transport of a number of ions equal to the number of desalination cells. Note that the initial salinity of the desalination cell and the concentration cell is the same before the reaction.

28


Page 28 of 33

NaCl NaClO4

0 Bi foam

NaOH

30

2.0

Urea Free Cl

1.5

Total Cl

1.0 10 0.5 0.0

0 0

-5

d

NaOH + K2SO4

1.2 1.5 1.8 2.1

1.5 Cl @1.5 V (SC)

1.0 ClO4 @1.9 V

ClO4 @1.7 V

0.5

Formate (mol)

ClO4 @1.5 V

Cl @1.5 V

0.0 0

6

8

Time (h) 100 80 8

100

2.0

4

125

E (V vs. SCE)

b

2

150

75

pH

-10 -1.8 -1.5 -1.2 -0.9

6 5

50

60

7

40 0

20

40

Time (min)

60

20 NaOH

25

Mixed

0 2

4

6

0 -20

0

Time (h)

20

TOC

TOC (mg/L)

5

I (mA)

c

n-TEC

10

20

30

40

50

60

Time (min)

Figure 1. (a) Linear sweep voltammograms of as-synthesized n-TEC electrodes in aqueous solutions of NaCl and NaClO4 (each 20 mM) and of Bi foam electrodes in aqueous solutions of NaOH (0.1 M) and NaOH + K2SO4 (each 0.05 M) purged with CO2 (at equilibrated pH values of ~6.7 and ~5.5, respectively; see Figure 1d inset). The upper inset shows a crosssectional SEM image and the elemental mappings of n-TEC (see Figure S1 for more details of the surface characterization of n-TEC); the bottom inset displays the SEM images (top view) of Bi foams. (b) Electrocatalytic decomposition of urea (2 mM) in aqueous NaCl and NaClO4 electrolytes (each 20 mM) using n-TEC electrodes maintained at 1.5–1.9 V vs. SCE. See Figure S4 for more data. (c) Bulk electrolysis of urea (2 mM) using n-TEC at E = 1.5 V vs. SCE in aqueous NaCl (50 mM) and simultaneous time changes of (total and free) chlorines and TOC. (d) Electrocatalytic formate production from CO2 and the corresponding Faradaic efficiencies (FEs) (inset: changes in pH during electrocatalysis) using Bi foam electrodes at −1.5 V vs. SCE in CO2-purged aqueous solutions of NaOH (0.1 M) and NaOH + K2SO4 (each 0.05 M).

29


FE (%)

10

Urea,

a

Urea (mM)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Chlorines (mM)

Page 29 of 33


4

3

2

3

Cl in DS

80

Na in DS

2

Cl in AN

60 40

Na in CA

1

Cl_ITE

20

Na_ITE

0 0

5

10

15

Formate (0.1 M)

1.2

60

FE (2 mM) FE (0.1 M)

40

0.6

FE (%)

80

Formate (2 mM)

20 5

10

15

20

I @-1.5 V vs. SCE (Bi) Estack (V)

9

Ea (V vs. SCE)

6 3

18

6

12

5

10

15

20

Without CO2 purging With CO2 purging

0

5

10

15

20

1.0

100 mM

6 100 mM

0.5

2 mM

0.0 0

20

2.0 1.5

2 mM

0

0 0

15

6

f

2

0

10

8

12

4

5

4

0 0

0 0

10

 (mS/cm)

12

d 12

100

1.8

0.0

e

0

20

Catholyte pH

0

E (V)

Formate (mmol)

c

Time (h)

5

10

15

20

Time (h)

Figure 2. Desalination–electrocatalysis hybrid reactions using n-TEC anode and Bi cathode pairs in three-cell (batch-type) stacks (see Scheme 1a). (a) Time-profiled changes in the current (I), potential (vs. SCE; Ea for the anode and Ec for the cathode), and the stack voltage (Estack). The n-TEC anode (with 2 mM urea) was maintained at 1.5 V vs. SCE, and the potential biased to the Bi cathode (with 2 mM K2SO4 purged with CO2) was estimated from the difference between the Estack and the Ea. An aqueous solution of NaCl at 0.171 M (10 g L−1) was used as saline water. (b) Changes in the amounts of ions (Cl− and Na+) in the anode cell (AN, 30 mL), cathode cell (CA, 30 mL), and desalination cell (DS, 20 mL) during the hybrid reactions of Figure 2a. In addition, the ion transport efficiency (ITE) was estimated for Cl− and Na+. See Figure S11 for the changes in the concentrations of urea, total chlorine, free chlorines, and TOC as functions of time. (c) Formate production and its Faradaic efficiency in the cathode cell with 2 mM and 0.1 M K2SO4 purged with CO2 (conditions in Figure 2a and 2e, respectively). (d) pH changes in catholytes containing 2 mM K2SO4 with and without CO2 purging. (e) Changes in the current (I), potentials (vs. SCE; Ea for the anode and Ec for the cathode) and the stack voltage (Estack) as functions of time. The Bi cathode (with 0.1 M K2SO4 purged with CO2) was maintained at −1.5 V vs. SCE, whereas the potential at the n-TEC anode (with 2 mM urea + 0.1 M K2SO4) was estimated by difference between the Estack and the Ec. An aqueous solution of NaCl at 0.171 M (10 g L−1) was used as saline water. (f) Saline water conductivity () in the desalination cell and urea concentration in the anode cell under the conditions in Figure 2a and 2e.

30


Urea (mM)

6

100

4

ITE (%)

6

-Ec (V vs. SCE)

Ions (mmol)

12

I @1.5 V vs. SCE (n-TEC) Estack (V)

9

I (mA)

b

12

E (V)

a

I (mA)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 30 of 33

2.0 1.0 0.5

18

a

9.98

1-DS 3-DS 5-DS

1.5

0.0

10

0

2

4

6

Time (h)

8

3.02

5

1.88

b

12 500

5-DS

400

6

ITE (%)

I (mA)

15

Urea (mM)

20

3-DS

300 200

1-DS 1-DS

3-DS

100

0 2

4

6

8

5-DS

0

0 0

0

2

4

1

6

2

3

4

5

4.0 4 3.5

Case A Case B Case C

c

0 0

5

40

4.5

10

15

Case C Formate FE

0 8

12

16

20

e

100 80 60 40 20 0

24

500 Case A

400

Case B Case C

300 0

20

6

12

18

24

Time (h)

d

0

d

0

2 1

30

20

3

4

8

10

3.0

Time (h)

0

 (mS/cm)

0

ITE (%)

PV E (V)

2

50

5.0

E (V)

I (mA/cm2)

8 6 4 2 0

FE (%) STD Efficiency (%)

I (mA)

8

6

Time (h)

Time (h) 10

Formate (mM)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


 (mS/cm)

Page 31 of 33

6

12

18

24

Time (h) 18

f

12 Case A

6

Case B Case C

0 0

Time (h)

4

8

12

16

20

Time (h)

Figure 3. (a and b) Electrodesalination–electrocatalysis hybrid reactions in a DC-powered multidesalination cell-stack with n-TEC anode and Bi cathode pairs at 3.5 VDC. 1-, 3-, and 5DS represent the numbers of desalination cells. Anolyte: 2 mM urea; catholyte: 2 mM K2SO4; desalination cell: 0.171 M NaCl. Time-profiled changes in (a) current and (b) solution conductivity () of the saline water with different numbers of desalination cells are shown. The insets in (a) and (b) show the change in urea concentration with time and the averaged ion transport efficiency (ITE), respectively. The numbers on the current profiles in (a) refer to the specific energy consumption (SEC, kW m−3) for 50% desalination. (c–f) Three case operations of a PV-coupled 5-DS stack device (see Scheme 1b). Case A: 2 mM urea / 0.171 M NaCl / 2 mM K2SO4. Case B: Urine (1/10-diluted) / seawater (36 g L−1) / 2 mM K2SO4. Case C: Urine (1/10-diluted) with 0.1 M K2SO4 / seawater (36 g L−1) / 0.1 M K2SO4. (c) Time-profiled changes in current and voltage of the device (inset: IV curve of the employed PV; area of 1.75 cm2, ISC of ~6.23 mA cm−2, VOC of ~4.3 V, fill factor of ~67%, and power conversion efficiency of ~18%). (d) Changes in the conductivity of saline water and ITE (inset). (e) Formate production and Faradaic efficiency. (f) Solar-to-desalination efficiency. 31


24


For Table of Contents Use Only

This study presents a solar desalination system accompanying electrocatalytic oxidations of desalted chloride and production of formate from captured CO2.

32


Page 32 of 33

Page 33 of 33 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


This study presents a solar desalination system accompanying electrocatalytic oxidations of desalted chloride and production of formate from captured CO2. 283x185mm (96 x 96 DPI)


High Efficiency Solar Desalination Accompanying Electrocatalytic

Recommend Documents