Energy and water are deeply interconnected, and each sector is both central to society and under increasing stress. Innovations in materials will be a powerful tool in efforts to overcome these challenges by providing sustainable solutions to treating water and rendering it fit-for-purpose with minimal expenditure of energy and other resources. Interfaces between components of water systems and the water-based fluids themselves govern the performance of the vast majority of water treatment and conveyance processes. This perspective examines many of these interfaces, ranging from those in sorbents and sensors to membranes and catalysts, and surveys opportunities for scientists and engineers to reveal new insights into their function and, thereby, to design novel technologies for next-generation solutions to our collective energy-water challenges.

Energy and water represent two of the most impactful sectors for society, and they are intimately intertwined (Fig. 1). Globally, demand for both of these commodities has relentlessly risen for decades, and this trend is projected to continue for the foreseeable future,1,2 putting increasing stress on limited resources and the environment. Without massive amounts of fresh water, modern energy production operations would be impossible. Water for cooling and steam generation in power plants, for example, represents nearly half of all water withdrawals in the United States.3 Hydroelectric power plants, of course, also depend centrally upon water resources. But water demand for the energy sector does not stop there. Refining of petroleum, whether in cracking or coking facilities, consumes about half a gallon of water for every gallon of product,4 and water withdrawal/return at these sites is far more than that. Hydraulic fracturing, too, is a heavy water consumer. For fracking operations, water is sourced from either groundwater or a surface supply to prepare hydraulic fracturing fluids (blends comprising primarily water, mixed with proppant, and additives). These fluids are injected into the well to create cracks within the targeted rock formation, releasing the fossil resources, but also vast amounts of aqueous fluid that returns to the surface. This so-called produced water often represents many times the volume of the oil and gas recovered, and it must be treated and then either reused or disposed.5 Biofuels, too, are water-intensive. Not only must the crops themselves generally be irrigated, but the processing of these crops into fuels also consumes substantial amounts of water, for boiler and cooling tower water, use in the dryer system, and so on. Collectively, nearly 30 gallons of water are consumed for each gallon of ethanol produced.6 

FIG. 1.

Examples of interconnectivity of energy and water.

FIG. 1.

Examples of interconnectivity of energy and water.

Close modal

Just as the energy sector relies heavily on water resources, so does the water sector rely on energy. Treatment of drinking water and, especially, wastewater, and distribution of water to and from these treatment plants, consumes about 4% of all electricity in the United States.7 In regions where water resources must be transported large distances or across significant elevation changes, the demand can be far above that value (e.g., it is nearly 20% in California). The principal energy demands for water treatment include aeration, solids processing, and, most of all, pumping.

The energy and water sectors are both so central to modern society that one cannot imagine how we would function where the flow to cease when the tap is turned on or the lights not to come on when the switch is flipped. Yet, both energy and water resources are under increasing stress worldwide. From a supply standpoint, of course, the total amount of water on Earth is essentially constant. The location and quality of water, however, varies considerably over time. Our waterways are increasingly polluted, both by our own activities and by effects derived from a warming planet. Many coastal aquifers, for example, are suffering from saltwater intrusion as sea level continues its relentless rise. Climate change is also shifting the timing and nature of precipitation. Storm intensity is increasing, meaning that water is delivered in sudden torrents that reservoirs and soils cannot retain. More precipitation is also falling as rain rather than snow, which means that it does not reside in the mountains and release gradually during the springtime melt, but rather runs off during the winter months when demand is relatively low. Groundwater, a major source for agriculture and industrial/domestic use globally, is being consumed unsustainably in almost all cases, rapidly lowering water levels in aquifers. Together, these factors translate to ever-increasing water scarcity. Switching to the demand side of the equation, the prospects are equally daunting. Demand for water is projected to increase substantially in the coming decades. This swelling demand is spurred by both a growing population and continuing development. As individuals shift to consuming more meat in their diets and begin to use larger amounts of electricity or consumer products, their water footprint increases dramatically. With skyrocketing demand and diminishing supply for a material that is both essential and not substitutable, crises are inevitable.

Many solutions to this challenge are based in policy, consumer behavior, and infrastructure modernization. Scientists, too, will be needed to secure a sustainable future, and material interfaces play a particularly central role in this arena.8 Interfaces between solid materials and aqueous solutions dictate the properties and performance of everything from sorbents to pipes to membranes. While such interfaces in natural systems, where aqueous fluids interact with minerals in the sub-surface, are critical to both energy and water sectors, this perspective will focus on interfaces in manufactured systems. Readers interested in sub-surface interfaces may refer to excellent reviews in the literature.9–11 In this perspective, water/solid interfaces will be explored first in the context of non-reactive interactions, where interplay of affinities, microstructure, and electrostatics govern behavior, and then in the context where chemical reactions take place, adding further complexity to the system. This is not meant to be an exhaustive review of these topics, which are each broad and deep fields of study with more than a century of scientific exploration, but rather a high-level introduction and examination of some particular areas with potential for major impact. Interfaces provide a rich diversity of properties considering the myriad relevant solid materials, each with its own microstructure and composition, and the aqueous solutions, varying in dissolved species, ionic strength, and more. This complexity represents both a challenge to those trying to develop predictive understanding of water/solid interfaces and an opportunity to design and exploit innovative interactions.

Many factors contribute to interactions between a solid surface and the components of an aqueous solution—including the water itself. Microstructure, electrostatics, dispersion forces, and entropic effects all can play a role. Even when one excludes the possibility for chemical reaction between these species, the complex interplay of the various interactions can challenge today's most sophisticated models. A thorough understanding of water/solid interfaces, however, is critical to the design of next-generation sorbents, sensors, and membranes that exhibit selectivity, resilience, and energy-efficiency.

Cleaning water generally means separating water from various undesirable constituents suspended or dissolved in the water. Many separations challenges with water are best approached using affinities to bind the targeted species through sorption. Sorbent materials either absorb or adsorb other materials. This can be accomplished either by electrostatic/chemical affinity or via physical trapping. Absorption is the process by which a fluid dissolves or permeates into another liquid or solid, and this process is far less common in water treatment, with the possible exception of superabsorbent polymers occasionally used to solidify waste. Adsorption, in contrast, is a powerful and widespread mechanism for cleaning water. Adsorption entails adhesion of atomic or molecular species to an interface, generally utilizing a porous substrate to maximize the surface-to-volume ratio. Typical adsorbents include activated carbon,12 biochar,13 silica gel,14 zeolites,15 and polymers.16 Activated carbon—a highly porous material generally formed into pellets or powder—is the most extensively used, driven by the abundance and low cost of the material. Activated carbon is manufactured by heating carbon sources in the absence of oxygen while exposing them to carbon dioxide or steam. This environment induces the formation of graphitic microstructures containing pores, the dimensions of which can be adjusted via process conditions. Despite their apparent simplicity, these materials are effective sorbents for a wide range of organic substances. Biochar is related to activated carbon, derived from thermal decomposition of biomass in oxygen-limited conditions, and has received particular attention for its penchant for capture of heavy metals.13,17,18 Whereas activated carbon often targets non-polar species, silica gel is an effective strategy for sorption of polar hydrocarbons. Silica gel is most often synthesized in reactions of sodium silicate with an acid, forming porous, amorphous silicon dioxide. Pore size can then be tuned by manipulation of the aging conditions. In addition to hydrocarbons, there are also reports of porous silica materials adsorbing heavy metals.19 Another class of silicate material, zeolites, offers highly uniform pores with diameters comparable to small molecules, enabling them to selectively bind particular species from an aqueous solution. Perhaps the greatest sorption design flexibility, however, comes from polymer-based sorbents. Functionalization of polymer chains with chemical moieties allows one to insert virtually any affinity interaction, which can then be further tuned by manipulating the morphology of the network of pores making up the sorbent material.

When treating water, sometimes adsorption involves a trade between the sorbent and the water in a process called ion exchange, in which ions from the solution are swapped with those residing on the sorbent surface. The sorbent interface in these cases can be a zeolite such as those described above, natural minerals such as clay or, most commonly, a polymer known as an ion exchange resin. Perhaps the most common use of ion exchange is in water softening, where multivalent ions (e.g., Ca2+ and Mg2+) from the source water replace monovalent ions (Na+, K+, or H+) present in the resin, driven by the tighter binding of the former with the polymer's anionic groups. Ion exchange systems have a cyclic operation because the process is consumptive, and the sorbent must be periodically regenerated by washing away the accumulated ions and returning the material to its initial condition.

Looking forward, sorbents will continue to rely on the proven technologies described above, but there is also a pressing need for innovative sorbents that exhibit improved specificity or other superior characteristics.20 A number of materials have been developed to this end. Metal-organic frameworks (MOFs), related to zeolites, consist of metal ions or clusters coordinated to organic ligands. MOFs generally form microporous structures, which have seen substantial application in gas sorption,21 but opportunities are emerging to use these materials for water treatment as well.22 Zhang et al., for example, have developed a MOF (ZJU-101) derived by modification of a previously reported material (MOF-867) in which methyl groups were added to the pyridyl sites to form the cationic framework specifically designed to adsorb hexavalent chromium, a troublesome human carcinogen [Fig. 2(a)].23 

FIG. 2.

(a) Illustration of the ion exchange process in ZJU-101 demonstrating how Cr2O72− ions were captured and illustration of the coulombic attraction between the positive ligand and negative Cr2O72−. Adsorption isotherms for dichromate adsorption over MOF-867 and ZJU-101, Ce: equilibrium concentration of adsorbate, and Qe: the amount of Cr2O72− adsorbed. Adapted from Ref. 23. (b) Adsorption rates of Congo red (CR) and methylene blue (MB) on porous BN nanosheets, respectively. The insets show the corresponding photographs. Adsorption isotherms of CR and MB on porous BN nanosheets, respectively. Qe (mg g−1) is the amount of dyes adsorbed at equilibrium and Ce (mg l−1) is the equilibrium solute concentration. The insets present the photographs of the BN porous nanosheets with CR and MB composites before (upper) and after (bottom) recovering by heating at 400 °C for 2 h, respectively. Reproduced with permission from W. W. Lei et al., Nat. Commun. 4, 1777 (2013). Copyright 2013 Springer Nature. (c) Still images taken from a repetitive sorption/compression/re-sorption cycle of a treated polyurethane foam sponge adsorbing dyed silicone oil (top row) and Anadarko crude oil (bottom row) from the surface of a water bath. Reproduced with permission from E. Barry et al., J. Mater. Chem. A 5, 2929 (2017). Copyright 2017 The Royal Society of Chemistry. (d) Kinetics of dyes removal by nFeMCH. The inset demonstrates the magnetic separation of dye-laden-nFeMCH. Reproduced with permission from M. Khan and I. M. C. Lo, J. Hazard. Mater. 322, 195 (2017). Copyright 2017 Elsevier.

FIG. 2.

(a) Illustration of the ion exchange process in ZJU-101 demonstrating how Cr2O72− ions were captured and illustration of the coulombic attraction between the positive ligand and negative Cr2O72−. Adsorption isotherms for dichromate adsorption over MOF-867 and ZJU-101, Ce: equilibrium concentration of adsorbate, and Qe: the amount of Cr2O72− adsorbed. Adapted from Ref. 23. (b) Adsorption rates of Congo red (CR) and methylene blue (MB) on porous BN nanosheets, respectively. The insets show the corresponding photographs. Adsorption isotherms of CR and MB on porous BN nanosheets, respectively. Qe (mg g−1) is the amount of dyes adsorbed at equilibrium and Ce (mg l−1) is the equilibrium solute concentration. The insets present the photographs of the BN porous nanosheets with CR and MB composites before (upper) and after (bottom) recovering by heating at 400 °C for 2 h, respectively. Reproduced with permission from W. W. Lei et al., Nat. Commun. 4, 1777 (2013). Copyright 2013 Springer Nature. (c) Still images taken from a repetitive sorption/compression/re-sorption cycle of a treated polyurethane foam sponge adsorbing dyed silicone oil (top row) and Anadarko crude oil (bottom row) from the surface of a water bath. Reproduced with permission from E. Barry et al., J. Mater. Chem. A 5, 2929 (2017). Copyright 2017 The Royal Society of Chemistry. (d) Kinetics of dyes removal by nFeMCH. The inset demonstrates the magnetic separation of dye-laden-nFeMCH. Reproduced with permission from M. Khan and I. M. C. Lo, J. Hazard. Mater. 322, 195 (2017). Copyright 2017 Elsevier.

Close modal

Heavy metals and other ions have also been targeted by carbon nanotube sorbents27 and, increasingly, by the extended family of two-dimensional (2D) materials.28,29 These materials possess particularly high surface area, which translates to large sorption capacity for materials attracted to the basal planes—including organic, non-ionic species.24,30,31 Lei et al., for example, reported porous boron nitride (BN) nanosheets with very high specific surface area that exhibited excellent sorption performance for oils, solvents, and dyes [Fig. 2(b)].24 The BN sorbent captured up to 33 times its mass in oils and organic solvents. Such 2D material sorbents can sometimes be regenerated by thermal treatment, which overcomes the binding of the adsorbed species, releasing them from the surface so the sorbent can be reused.

Reusability, as with heat treatment of 2D material sorbents or washing of ion exchange resins, is often overlooked as a powerful metric for these materials. Reuse of sorbents can markedly reduce the life-cycle cost and increase the sustainability of a treatment process. A simple regeneration strategy relevant for certain adsorption scenarios is mechanical compression. Polymeric foam sponges are promising candidates for this approach. Several groups have recently reported sponges with modified interfacial properties as sorbents for oil and organic solvents,25,32,33 predominantly for applications in water bodies where fluid selectivity is vital.34 After extracting oil from the water with such a sponge, it can be compressed, with the captured fluids recovered in a containment vessel, thereby rendering the sorbent ready for another cycle of adsorption and compression [Fig. 2(c)]. Yet another methodology for sorbent reuse is to implement magnetic separation to concentrate the captured species and minimize energy required for regeneration of the sorbent.26,35 Khan and Lo reported a sorbent based on nanoscale γ-Fe2O3 magnetic cationic hydrogel (nFeMCH).26 This sorbent was applied for the removal of two acid dyes (Acid Red 27 and Acid Orange 52), for which it exhibited a rapid sorption rate over 30 consecutive rounds of sorption–desorption, facilitated by magnetic separation of the superparamagnetic material [Fig. 2(d)].

Specificity is also of central importance when designing next-generation sorbents. Ideally, one would tailor interfacial properties to adsorb a particular targeted solute. Achieving such a goal would allow for capture of challenging water contaminants such as nutrients36,37 and heavy metals.38 Of equal importance is the enticing potential of capturing one specific material from a complex solution. Bioactive molecules, such as endocrine disruptors,39 have been gaining increasing attention as recalcitrant pollutants in wastewater. An adsorbent with high specificity could sequester these compounds while allowing non-hazardous organic materials to pass through. Adsorption of ions, including ions of a particular charge, is already within the purview of current water treatment technologies. But if ion-specific sorption were effective, energy investment in resource recovery from waste streams and treatment of hazardous waters would be dramatically reduced.

In some ways, insights derived from selective adsorption can be applied to a different challenge in water—sensing. Water sensing is a broad field and can be generally classified as either sensing of chemical content or measurement of volume/flow. The focus here is on the former, with a particular emphasis on proximity- and/or contact-based evaluation. Such sensors often rely on either electrochemical/electrical or optical measurements. One strategy, for example, is to chemically functionalize an interface to selectively attract certain ions to the gate of a transistor, where the concentration is measured by current modulation resulting from the charged ion species (ion-sensitive field-effect transistor).40 Central challenges related to sensing technologies are specificity, sensitivity, usability/reusability, and cost. Researchers are leveraging recent developments in materials science to meet these challenges. Interface functionalization chemistry is enabling more selective adhesion. For optical sensors, inexpensive semiconductor light-emitting diodes, lasers, and detectors (including plasmonics) are now available. Sensitive field effect transistors can enhance gate-driven electrical sensing.

Many recent studies have leveraged fluorescence quenching as a reliable method for detecting trace analytes in solution. The adsorption selectivity of MOFs, which can be designed with native fluorescence, can be a powerful tool to this end. The idea is to observe the emitted light, the intensity of which is diminished when the targeted species binds to the interface. Researchers have demonstrated MOFs with extremely sensitive detection of Fe3+ ions,41 detection of chromate ions in the presence of large excess of other anions,42μg/L detection of strategically important uranyl ions,43 and even, interestingly, detection of trace concentrations of water itself within other solvents.44 Supramolecular and macromolecular systems have also been utilized for selective binding of species in aqueous solutions coupled with optical detection. Luminescent coordination polymers have been reported, for example, with sub-ppb level sensitivity for trivalent chromium.45 By exploiting competitive binding interactions in arrays comprising multiple sensing components, it is possible to identify simultaneously a wide range of analytes. Fluorescent supramolecular gels with 22 binding sensors have been reported with capability to identify 14 different ions, ranging from halides to various metals to protons.46 

As with sorbents, carbon and porous silica materials play a major role in sensor materials, in this case predominantly in the context of electrochemical detection. Metal ions, especially lead and copper, are a common target given their prevalence in buildings' water pipes, as are other metals carrying health concerns such as cadmium. Gold-doped carbon foams47 as well as graphene and reduced graphene oxide (rGO) composites48,49 have been shown to successfully detect such ions with nanomolar sensitivity. rGO composites have also been developed for nM-μM electrochemical sensing of nitrate and nitrite,50,51 two ions widespread in natural water systems primarily as a result of agricultural runoff and representing both a direct health risk and an aggravating factor for devastating eutrophication of water bodies. Aromatic amines are extensively used for the manufacture of chemicals, including dyes, pharmaceuticals, agrochemicals, and chelating agents, among others. These compounds often end up polluting water, and mitigation requires sensitive detection. Composites containing mesoporous silica have shown promise in sensing molecules such as hydrazine and nitrobenzene using electrochemical methods.52 

Field-effect transistors (FETs) are often used in capacitance and other electronic water sensor technologies, and detection times and sensitivities are seeing unprecedented improvements in recent years as a result of the introduction of 2D materials.53 This class of materials is essentially entirely made up of interfaces, so the ratio of surface area available to interact with analytes is enormous, while the distance over which electronic effects must be transported or detected is miniscule. 2D materials as diverse as graphene, black phosphorous, and molybdenum disulfide have been reported by Chen et al., with demonstrated limits of detection in the ppb range for various heavy metals, and in some cases even lower.54–56 As the library of 2D materials grows, so too will the potential for this class of materials to impact sensor technology.

Membranes are thin materials that moderate the transport of species through them depending on their physical or chemical properties. These components are critical elements in water purification,57,58 including desalination,59,60 decontamination, and disinfection. The aim for membranes in this context is to remove undesirable species, including ions, solutes, pathogens, and particles, while allowing water to pass through. There are many types of membranes, but they can generally be divided into two categories: isotropic and anisotropic. Isotropic membranes are chemically and structurally homogeneous through their cross section, and they usually rely on size-exclusion as the means for executing separation. Anisotropic membranes are heterogeneous structurally—and sometimes chemically—through their cross section, and they can take many different forms depending on how they are fabricated. Phase-separation membranes (sometimes called Loeb–Sourirajan membranes) have a gradient in pore size and porosity across the membrane thickness but are chemically homogeneous. Composite membranes are both chemically and structurally heterogeneous,61 and there is a special class of these membranes, called Janus membranes,62 which has emerged recently and will be discussed at the end of this section.

Another way to categorize membranes is in terms of the size of species that are excluded (Fig. 3). Macroscopic objects are excluded using particle filtration, with microfiltration (MF), ultrafiltration (UF), and nanofiltration (NF) targeting progressively smaller species. Reverse osmosis (RO) membranes are generally considered non-porous and are used to separate out ions and small molecules via selective solution–diffusion transport. Most membranes have a broad distribution of pore sizes, with the large-pore pathways dominating the transport behavior. This sometimes limits their effectiveness in filtering out targeted species since undesirable materials can find a way through the membrane. A leading challenge in membrane science, therefore, is to narrow pore size distributions to reduce energy consumption and provide superior overall performance. The focus for this section, however, is specifically on interfaces in membranes rather than on the nature of the support material or the pore geometry per se.

FIG. 3.

Schematic illustration of membrane filtration spectrum. Reverse osmosis, nanofiltration, ultrafiltration, microfiltration, and conventional particle filtration differ principally in the average pore diameter of the membranes. Reverse osmosis membranes are so dense that the pores are considered as non-porous. Reproduced with permission from A. Lee et al., Environ. Sci. 2, 17 (2016). Copyright 2016 The Royal Society of Chemistry.

FIG. 3.

Schematic illustration of membrane filtration spectrum. Reverse osmosis, nanofiltration, ultrafiltration, microfiltration, and conventional particle filtration differ principally in the average pore diameter of the membranes. Reverse osmosis membranes are so dense that the pores are considered as non-porous. Reproduced with permission from A. Lee et al., Environ. Sci. 2, 17 (2016). Copyright 2016 The Royal Society of Chemistry.

Close modal

Tailoring membrane properties necessitates understanding many complex interactions at interfaces. Membrane interfaces can comprise a spectrum of different materials. Inorganic membranes are often nanocrystalline solids with complex interactions, structure, and dynamics of water near their surfaces. Polymer or other soft matter membranes are similarly affected by the environment and interactions near their interfaces. A prominent attribute of membranes is that water and solutes are confined within pores, a factor that influences the structure and dynamics of these species.

A ubiquitous issue with membrane interfaces is fouling, a process by which fluid flux through the membrane is undesirably reduced.63 Membrane fouling can be the result of adsorption of organic or biological species or precipitation and cake formation of inorganic species, any of which can block the pores and restrict flow. These phenomena are particularly prevalent in water treatment applications, since the raw water coming into direct contact with the membranes generally contains a number of detrimental solutes or suspended materials. Many different strategies have been developed to deal with membrane fouling, but as of yet none are sufficiently effective or efficient. One approach is to pre-treat water before it encounters membranes, but such processes involve substantial energy investment. Today, fouling is often addressed by periodic cleaning or replacement of fouled membranes. Mild cleaning, such as by back-flowing water, can remove loosely adhered species, but harsh chemical cleaning is necessary to address most foulants, and these treatments tend to degrade the membranes themselves, shortening their lifetime. Membrane interface design offers an alternative route, building resilience into the membranes themselves such that they have passive or active antifouling properties.

A common approach for passive resistance to fouling is to create hydrophilic interfaces, often involving polyethylene glycol (PEG).64 This polymer attracts a protective hydration layer to the surface of the membrane, which serves to minimize access to the interface for potential foulants. Simply coating a membrane with PEG is ineffective, as the coating will generally not be robustly adhered in an operational setting. Fan et al. presented a novel strategy involving a polyethersulfone (PES) membrane matrix with PEG as both surface-modifier and pore-forming agent, and m-trihydroxybenzene (MTB) as a hydrogen bond donor to secure the PEG to the PES pore walls and thereby prepare anti-fouling UF membranes [Fig. 4(a)].65 

FIG. 4.

(a) Schematic for the formation of anti-fouling PES membranes achieved using surface segregation of PEG mediated by MTB. Reproduced with permission from X. Fan et al., J. Membr. Sci. 499, 56 (2016). Copyright 2016 Elsevier. (b) Schematics of various fabrication strategies to prepare Janus membranes. Reproduced with permission from H.-C. Yang et al., Angew. Chem. Int. Ed. 55, 13398 (2016). Copyright 2016 Wiley.

FIG. 4.

(a) Schematic for the formation of anti-fouling PES membranes achieved using surface segregation of PEG mediated by MTB. Reproduced with permission from X. Fan et al., J. Membr. Sci. 499, 56 (2016). Copyright 2016 Elsevier. (b) Schematics of various fabrication strategies to prepare Janus membranes. Reproduced with permission from H.-C. Yang et al., Angew. Chem. Int. Ed. 55, 13398 (2016). Copyright 2016 Wiley.

Close modal

Another popular hydrophilization strategy relies on zwitterionic molecules. In some cases, these species are integrated directly into the membrane matrix, with some fraction of the zwitterionic species segregating at the interface. Li et al., for example, prepared an amphiphilic zwitterionic copolymer poly(vinylidene fluoride)-graft-poly(sulfobetaine methacrylate) (PVDF-g-PSBMA) and used it as an amphiphilic additive in the fabrication of PVDF membranes.66 Often, however, researchers aim to deposit the zwitterionic material directly onto the membrane's surface to maximize its impact. Zhao et al. exploited enriched poly(N,N-dimethylamino-2-ethylmethacrylate) (PDMAEMA) chains on the pore walls, transforming them into zwitterionic poly(carboxybetaine methacrylate) (PCBMA) by quaternization with 3-bromopropionic acid.67 Others have used grafting to achieve a similar goal. Li et al. grafted the zwitterionic monomer SBMA onto the surface of PVDF membranes.68 Recently, a novel peptide approach was introduced by Piatkovsky et al., who coated UF and RO membranes with a block copolymer of polystyrene (PS) bound to an alternating lysine-glutamic acid peptide and observed good anti-fouling performance.69 Many other types of surface treatment have been reported to increase hydrophilicity and thereby reduce fouling, ranging from GO70 to plasma treatment71 to ALD over-coating with hydrophilic oxide.72 

A shortcoming of these hydrophilic interfaces is that they tend to succumb to foulants that can coalesce and spread on the surface such as oils. A more nuanced interface can be designed that includes a mixture of hydrophilic and hydrophobic moieties or, alternatively, fluorinated species to overcome this issue. These interfaces are said to exhibit “fouling-release” properties.73–76 Examples of this approach include using a mixture of PEG and polydimethylsiloxane (PDMS) or grafting fluorinated polyamine on polydopamine (PDA) to form low-free-energy microdomains to impede the accumulation of foulants. Even these more advanced membranes will eventually surrender to fouling in sufficiently dirty water streams. In these cases, one can turn to active anti-fouling strategies, where the foulants are degraded or detached using aggressive reactions. These approaches will be addressed in the section below on reactive interfaces.

Beyond dealing with fouling, there are many reasons why one might want to adjust the interfaces of a membrane to change affinities for components of aqueous solutions, including water itself.61,77 Increasing hydrophilicity will have an advantage of reducing the pressure required to push water through the membrane, thereby saving energy. Delicate balancing of hydrophilicity and related affinities can translate into not only more efficient transport but also more effective selectivity, as in the case where lithium chloride monohydrate and titanium dioxide nanoparticles were dispersed in PVDF to simultaneously increase flux and favor passage of certain ions.78 Intricate control over electrostatics is particularly important for separations in solutions with mixed ion populations. Ultrathin sheets of 2D materials such as MXenes, e.g., Ti3C2Tx, where T represents terminating functional groups (O, OH, and F), have demonstrated charge- and size-selective rejection of ions and molecules while maintaining high water flux.79 Differential sieving in such systems is a function of both the hydration radius and charge of the ions. Another recent pursuit is to draw inspiration from biological membranes, in which natural systems have engineered transmembrane proteins with precise (sometimes dynamic) conformations and electrostatics to transport specific ions or water with extreme efficiency and selectivity. Researchers are both using such proteins directly in water treatment80 and developing artificial transmembrane water channels to achieve selective transport.81 

As alluded to at the beginning of this section, there is a special class of anisotropic membranes called Janus membranes, which take distinct advantage of opposing interfacial affinities on their two faces.62 Janus films are prevalent in nature, such as the lotus leaf, which has one face that is superhydrophobic and another that is hydrophilic and superoleophobic, affording the leaf with both self-cleaning and anti-fouling properties. Janus membranes, by adding pore networks connecting the opposing faces, allow the two faces to work collaboratively and enable unusual transport behavior. The interfacial properties can be related to wettability, charge, or morphology, among others. Fabrication of Janus membranes can be challenging. Many methods one might use to modify an interface involve wet chemistry that will tend to be drawn into the pores of a membrane by capillary action, thereby modifying the membrane homogeneously rather than only on the desired face. Various methodologies have been developed to overcome this challenge, and they can generally be divided into asymmetric fabrication and asymmetric decoration categories [Fig. 4(b)], depending on whether the Janus configuration is created as part of the membrane structure itself or if it is realized by post-modification.

Among the interesting applications for Janus membranes is directional liquid transport.82 With careful control over the interfacial properties, for example, one can create a “water diode.” Following an asymmetric fabrication approach using electrospinning, Wu et al. prepared a fibrous film having hydrophobic polyurethane (PU) on one side and hydrophilic crosslinked poly(vinyl alcohol) (c-PVA) on the other.83 Water spreads on the PVA and has a large contact angle on the PU. A water drop placed on the PU will be pulled through the membrane to the hydrophilic side, facilitated by the large Laplace pressure exerted on the nearly circular droplet. In contrast, the water spread on PVA will not penetrate to the interior. For this application, as for most applications involving Janus membranes, the position of the boundary between the two properties within the membrane is critical to balance their relative influence and achieve cooperative transport. One can similarly prepare oil diodes, with the two sides being oleophobic and oleophilic, respectively. (Hydrophobic/hydrophilic Janus membranes often will not work for direction oil permeation because hydrophilic surfaces are generally also oleophilic in air.) Oil/water separation is therefore a common application for Janus membranes.84 Their related use for demulsification is as well, for example, using hydrophilic polyamine and superhydrophobic PDMS, with the former exposed to the emulsion.85 In such a system, the polyamine will destabilize the emulsion through electrostatic interactions, coalescing the oil into larger droplets, which are subsequently pulled through the PDMS layer. Janus membranes have wide-ranging applications in energy-water systems, ranging from catalytic contactors for wastewater treatment to osmotic energy harvesting at confluences of fresh and saline waterways.86 

In many water systems, the conditions are such that, in addition to all of the complex interactions outlined in Sec. II, there is also the possibility for chemical reactions to take place. These reactions can be intentional, as in the case of catalytic degradation of foulants, pollutants, and pathogens, or undesirable, as in the corrosion of pipes. In all cases, though, much remains to be learned about interfacial reactivity in aqueous systems.

The vast majority of our water infrastructure is simply for conveyance, that is, pipes used to move water to its point of use and to move it to a disposal or treatment location after its use. There is also a lot of infrastructure tied up in cooling coils, storage vessels/basins, and the like. All of these components suffer from unwanted chemical reactions at their interface with the water, namely, corrosion and leaching. Corrosion is influenced by a number of different factors, including the chemistry of the water and/or soil in contact with the interface (pH, dissolved oxygen, ionic strength, and mineral content), temperature (high temperatures speed up chemical reactions), and hydrodynamics (high velocity or sudden changes in direction create turbulent conditions that accelerate corrosion).87,88

When considering corrosion of water pipes, metal pipes immediately come to mind. Concrete sewers, though, also experience corrosion.89 Carbon dioxide from air can react with the calcium hydroxide in concrete to form calcium carbonate in a process called carbonatation [Ca(OH)2 + CO2 → CaCO3 + H2O]. This process actually strengthens concrete, but it lowers pH, which ends up corroding most metals. When saline water contacts concrete pipes, chlorides can leach calcium hydroxide. Sulfates are also present in saline water and cause corrosion in concrete, but sulfates more commonly originate from acid rain or from bacteria reducing hydrogen sulfide gas. Water can also simply dissolve minerals present in concrete, especially calcium, leaching ions into the fluid. Secondary corrosion of concrete can be initiated by bacteria, which are plentiful in untreated sewage. It is not the bacteria themselves that cause corrosion, rather, sulfate-reducing bacteria produce hydrogen sulfide, which other bacteria living in the space above the water level oxidize to sulfuric acid. This acid facilitates dissolution of carbonates and produces sulfates, which are themselves corrosion-inducers as described above.

Water distribution pipes are often constructed of cast iron, with supply lines within buildings typically using copper or, in older buildings, lead. Iron pipes will corrode through oxidation reactions, particularly when there is organic waste in the line releasing hydrogen sulfide, which is oxidized to sulfuric acid.90 Iron leaching from corroding pipes is the leading cause of water discoloration in potable water systems. Leaching of other metals, however, is generally of greater concern from the standpoint of human health. Copper and lead are both a concern in this regard, and their corrosion is accelerated in the presence of chlorine,91,92 which is routinely added to drinking water supplies as a disinfectant. As discussed above for concrete pipes, bacteria can also exacerbate corrosion in metal pipes.93 The fact that corrosion of water system components is still so pervasive today is a testament to how difficult it is to prevent. Nonetheless, with a greater understanding of interfacial properties and processes in these complex environments, development of improved leaching- or corrosion-resistant coatings—or even of corrosion-reversing processes—are conceivable.94 

Water remediation chemistry faces a broad array of solvated and suspended pollutants. Many of these contaminants can be addressed using sorbents and/or membranes, but challenges with fouling and disposal issues with pollutant-enriched waste preclude these technologies from fully tackling water treatment. Chemically reactive strategies combining advanced oxidative95–97 and reductive98,99 catalytic processes offer the complementary capability of degradation of pollutants while at the same time opening an enticing new opportunity for energy and resource recovery. In the context of interfaces, electrocatalysis and photocatalysis are the dominant subjects and will be the primary focus for this section. Homogeneous catalysis is certainly important in water treatment, but it is outside the scope of this perspective.

Electrocatalysis for water treatment is considered an advanced oxidation processes (AOP),95 and it has received increasing attention because of its high efficacy.100–102 Anodic oxidation (AO) is a heterogeneous electrocatalytic process that occurs at the surface of the anode. AO is related to the oxygen evolution reaction (OER) for water splitting, but the aim is to create reaction intermediates, such as hydroxyl radicals, rather than oxygen as the complete OER product. These oxygen-containing intermediates are highly reactive and able to break down a spectrum of organic materials into progressively smaller molecules. Degradation of organic pollutants through AO is a multi-step process, beginning with direct electron transfer to the surface of the anode and formation of reactive oxygenated species (ROS) via oxidation of water at the interface. The ROS (peroxides, superoxide, hydroxyl radical, or singlet oxygen) can react directly with organic pollutants at the interface or form hydrogen peroxide, releasing it into solution for subsequent degradation reactions. Degradation by •OH or H2O2 can be accomplished through partial oxidation or total mineralization, converting organic pollutants entirely into carbon dioxide and water. AO efficiency depends, among other factors, on the catalyst materials on the surface. Higher potential for oxygen evolution at the catalyst will correlate with weaker interaction between the catalyst surface and hydroxyl groups and, therefore, greater chemical reactivity for degradation processes. Catalysts traditionally used for OER (e.g., IrO2 and Pt) typically have strong •OH binding, limiting their utility for AO. Other catalysts, however, that were cast aside due to poor OER performance, such as first and second row transition metal oxides and boron-doped diamond, exhibit weak hydroxyl binding and are therefore more efficient AO catalysts.103 In an illustration of the effectiveness of transition metal catalysts performing AO for wastewater treatment, Cho et al. reported AO of real human waste using a semiconductor electrode with a mixed particle coating of bismuth oxide and titanium dioxide [Fig. 5(a)].104 

FIG. 5.

(a) Variation in absorbance spectra along with the electrolysis time and absorbance at 425 nm as a function of electrolysis time in potentiostatic electrocatalysis experiments. The inset visualizes the color variation. Reproduced with permission from K. Cho et al., RSC Adv. 4, 4596 (2014). Copyright 2014 The Royal Society of Chemistry. (b) (Left) Decay of atenolol concentration vs. electrolysis time by EF treatment at different current values. (Right) Decay of concentrations of (• and ▪) atenolol and (○ and □) p-hydroxybenzoic acid by EF treatment of solutions containing 0.15 mM of each organic compound with current of 30 (circles) and 60 mA (squares). The corresponding kinetic analyses are given in the inset panels. Reproduced with permission from I. Sirés et al., Water Res. 44, 3109 (2010). Copyright 2010 Elsevier. (c) Absorption spectra of visible light induced degradation of methyl orange (MO) as a function of wavelength and time in the presence of N-TiO2 membrane. Effect of N-TiO2 thickness on the photocatalytic degradation of MO using bare ceramic as a control system (indicated in grey). Reproduced with permission from A. Lee et al., Adv. Sustainable Syst. 1, 1600041 (2017). Copyright 2017 Wiley. (d) Schematic of a microbial electrolysis cell generating hydrogen from wastewater.

FIG. 5.

(a) Variation in absorbance spectra along with the electrolysis time and absorbance at 425 nm as a function of electrolysis time in potentiostatic electrocatalysis experiments. The inset visualizes the color variation. Reproduced with permission from K. Cho et al., RSC Adv. 4, 4596 (2014). Copyright 2014 The Royal Society of Chemistry. (b) (Left) Decay of atenolol concentration vs. electrolysis time by EF treatment at different current values. (Right) Decay of concentrations of (• and ▪) atenolol and (○ and □) p-hydroxybenzoic acid by EF treatment of solutions containing 0.15 mM of each organic compound with current of 30 (circles) and 60 mA (squares). The corresponding kinetic analyses are given in the inset panels. Reproduced with permission from I. Sirés et al., Water Res. 44, 3109 (2010). Copyright 2010 Elsevier. (c) Absorption spectra of visible light induced degradation of methyl orange (MO) as a function of wavelength and time in the presence of N-TiO2 membrane. Effect of N-TiO2 thickness on the photocatalytic degradation of MO using bare ceramic as a control system (indicated in grey). Reproduced with permission from A. Lee et al., Adv. Sustainable Syst. 1, 1600041 (2017). Copyright 2017 Wiley. (d) Schematic of a microbial electrolysis cell generating hydrogen from wastewater.

Close modal

The classical Fenton reaction, which uses a mixture of Fe2+ and H2O2, is a homogeneous chemical reaction used extensively in water treatment. When electrochemically produced H2O2 is involved, the process is called an electro-Fenton (EF) process.97,107 These processes produce additional •OH from the H2O2 via cooperative interactions with ferrous ions. Hydroxyl radicals are substantially more reactive than the peroxide, so EF processes can improve the efficiency of organic pollutant degradation. Sirés et al. compared AO and EF decontamination of solutions containing β-blockers (e.g., atenolol), which are pharmaceutical pollutants prevalent in water bodies, and found significant enhancement though the use of EF [Fig. 5(b)].105 

In recent years, researchers have been increasingly introducing another energy source to facilitate catalytic water treatment: light. As with electrocatalysis, the working principle of photocatalysis is AOP.108 Light energy from the sun or a manufactured light source is converted to oxidative species in the presence of water and oxygen, which then degrade organic compounds. Photocatalysis offers potential advantages of ambient operating pressure and temperature and thorough conversion of contaminants.109 Many semiconductor materials have been studied as photocatalysts for water treatment, but by far the most extensively studied system is TiO2.

TiO2 is most active for photon energies in the UV region, from 300 to 390 nm, it is stable throughout multiple photocatalytic cycles, and it generally does not release toxic byproducts. The intended mechanism is that electron–hole pairs created by excitation with photons will separate, and the electron will react with scavenger molecules such as oxygen. The hole can react with water to form reactive species, such as •OH, which can degrade organic pollutants or pathogens through radical oxidation and charge transfer as previously discussed. Some studies have aimed to shift light absorption into the visible range to reduce costs, potentially even exploiting sunlight. Lee et al., for example, reported nitrogen doping of TiO2 to create electronic states in the semiconductor gap and thereby harvest blue light, depositing this doped oxide onto membranes and degrading organic molecules [Fig. 5(c)].106 Looking beyond titania, many of the same classes of materials that have been explored for sorbents and membranes also appear in photocatalysis studies. There are many water-stable MOFs,110 for example, with light-harvesting capabilities that can generate ROS under illumination.111 Surely, there are also opportunities for 2D materials to play in this space. MoS2, for example, is known to be catalytically active and is a direct bandgap semiconductor when in monolayer form.112 

Interfaces that produce ROS for pollutant and pathogen degradation have an additional benefit of having active anti-fouling function. There are also materials that, however, specialize in active anti-fouling such as cationic polymeric materials.113,114 These systems present antimicrobial groups such as quaternary ammonium, phosphonium, and guanidinium. Electrostatic interactions between these cationic groups and bacteria that make up biofoulants, which are predominantly anionic, can damage the organisms' sufficiently to cause death. Graphene-based nanomaterials are also effective as anti-fouling surface modifiers.115,116 In addition to physically puncturing microorganisms, GO and rGO materials can cause oxidative stress through lipid peroxidation, thereby offering multi-faceted antimicrobial function.

Microbes also can serve a positive function in interfacial water treatment systems, facilitating catalytic reactions that recover valuable resources from wastewater. Biological catalysis is the result of chemical processes across hierarchical levels of organization, from atomic-scale control of active-site structures in individual enzymes and enzyme cascades to microbial community networks sharing coordinated functions. Waste streams have a plethora of materials that microorganisms can transform into beneficial products. Bio-electrohydrogenesis through microbial electrolysis cells (MECs), for example, is a promising technology for generating hydrogen by degradation of organic waste [Fig. 5(d)]. In such a MEC, microbial activity takes place on the anode, produced protons, and then traverses a membrane to the cathode, where hydrogen gas is evolved. Exoelectrogenic microorganisms can also generate biofuels, methane, and other useful inorganic and organic chemicals in microbial fuel cells (MFCs). Ongoing optimization of microbe-electrode interfaces will be central to the development of future enhanced MFC and MEC devices.117–120 

Water stress—and even full-blown water crises—are an increasing fact of life for our global society. It is clear that, in addition to progressive water policy and significant infrastructure investment, we will need a plethora of new technologies for treating water with energy-efficiency and low cost. Deeper understanding of interfaces between material surfaces and aqueous solutions is necessary to enable these technologies. There is a surprising complexity surrounding the interplay of water, solutes, microstructure, electrostatics, hydrodynamics, and reactivity at interfaces. New insights into this environment will serve as a foundation for sorbents capable of capturing specific, sometimes trace solutes from multi-component solutions and having straightforward reusability to minimize waste. Similar understanding of solute-surface interactions will be critical in the design of sensors capable of detecting analytes of concerns amid a sea of chemically similar species. Interfaces take on an additional role in confined geometries such as those in the pore network of membranes, where they influence not only affinities but also the transport of fluids and their constituent materials. Informed manipulation of surface charge and other properties will empower scientists to engineer novel transport phenomena such as unidirectional or highly selective flow, for example, of ions. Water conveyance pipes represent the vast majority of our infrastructure systems, and engineered interfaces have the potential to dramatically extend the lifetime of these components and improve human health by minimizing corrosion and leaching of concrete or metals. Many pollutants present in the water that are challenging to mitigate using separation methods can be attacked with catalysis, degrading them into less harmful species. Heterogeneous electro- and photocatalysts can effectively break down organic molecules and pathogens, primarily through efficient generation of reactive oxygen species, and new materials and interfacial engineering strategies have the potential to dramatically improve performance. Fouling of surfaces, an omnipresent nuisance for water systems that necessitates increased energy consumption and use of harsh chemical treatments, can be alleviated through a combination of passive and active strategies with carefully designed interfaces. Collectively, these advances will represent a major step in our efforts to secure a sustainable and prosperous water future.

This material is based upon the work supported by program development funding from Argonne National Laboratory, provided by the Director, Office of Science, of the U.S. Department of Energy under Contract No. DE-AC02-06CH11357.

1.
U.S. Energy Information Administration
, “
Annual energy outlook 2018 with projections to 2050
” (
2018
).
2.
United Nations Educational Scientific and Cultural Organization
, “
The UN World water development report 2015, water for a sustainable world
” (
2015
).
3.
C. A.
Dieter
,
K. S.
Linsey
,
R. R.
Caldwell
,
M. A.
Harris
,
T. I.
Ivahnenko
,
J. K.
Lovelace
,
M. A.
Maupin
, and
N. L.
Barber
, “
Estimated use of water in the United States county-level data for 2015: U.S. Geological Survey data release
” (
2017
).
4.
Jacobs Consultancy
, “
Water consumption in US petroleum refineries
” (
2016
).
5.
U.S. EPA
, “
Hydraulic fracturing for oil and gas: Impacts from the hydraulic fracturing water cycle on drinking water resources in the United States (final report)
” (
2016
).
6.
V.
Tidwell
,
A. C.
Sun
, and
L.
Malczynski
, “
Biofuel impacts on water
” (
2011
).
7.
C.
Copeland
and
N. T.
Carter
, “
Energy-water nexus: The water sector's energy use
” (
2017
).
8.
U. S. Department of Energy Office of Basic Energy Sciences
, “
Basic research needs for energy and water
” (
2018
).
9.
D. L.
Sparks
, in
Soil Physical Chemistry
, 2nd ed., edited by
D. L.
Sparks
(
CRC Press
,
Boca Raton
,
1998
).
10.
C. I.
Steefel
,
C. A. J.
Appelo
,
B.
Arora
,
D.
Jacques
,
T.
Kalbacher
,
O.
Kolditz
,
V.
Lagneau
,
P. C.
Lichtner
,
K. U.
Mayer
,
J. C. L.
Meeussen
,
S.
Molins
,
D.
Moulton
,
H.
Shao
,
J.
Šimůnek
,
N.
Spycher
,
S. B.
Yabusaki
, and
G. T.
Yeh
,
Comput. Geosci.
19
,
445
(
2015
).
11.
C.
Liu
,
Y.
Liu
,
S.
Kerisit
, and
J.
Zachara
,
Rev. Mineral. Geochem.
80
,
191
(
2015
).
12.
F. L.
Fu
and
Q.
Wang
,
J. Environ. Manag.
92
,
407
(
2011
).
13.
H.
Li
,
X.
Dong
,
E. B.
da Silva
,
L. M.
de Oliveira
,
Y.
Chen
, and
L. Q.
Ma
,
Chemosphere
178
,
466
(
2017
).
14.
S. V.
Patwardhan
,
Chem. Commun.
47
,
7567
(
2011
).
15.
S. B.
Wang
and
Y. L.
Peng
,
Chem. Eng. J.
156
,
11
(
2010
).
16.
E.
Ramirez
,
S. G.
Burillo
,
C.
Barrera-Diaz
,
G.
Roa
, and
B.
Bilyeu
,
J. Hazard. Mater.
192
,
432
(
2011
).
17.
Q.
Ma
,
Y.
Yu
,
M.
Sindoro
,
A. G.
Fane
,
R.
Wang
, and
H.
Zhang
,
Adv. Mater.
29
,
1605361
(
2017
).
18.
D.
Kołodyńska
,
J.
Krukowska
, and
P.
Thomas
,
Chem. Eng. J.
307
,
353
(
2017
).
19.
Z. J.
Ma
,
H. J.
Ji
,
Y.
Teng
,
G. P.
Dong
,
J. J.
Zhou
,
D. Z.
Tan
, and
J. R.
Qiu
,
J. Colloid Interface Sci.
358
,
547
(
2011
).
20.
J.
Płotka-Wasylka
,
N.
Szczepańska
,
M.
de la Guardia
, and
J.
Namieśnik
,
TrAC Trends Anal. Chem.
77
,
23
(
2016
).
21.
L.
Pan
,
M. B.
Sander
,
X. Y.
Huang
,
J.
Li
,
M.
Smith
,
E.
Bittner
,
B.
Bockrath
, and
J. K.
Johnson
,
J. Am. Chem. Soc.
126
,
1308
(
2004
).
22.
P.
Kumar
,
A.
Pournara
,
K.-H.
Kim
,
V.
Bansal
,
S.
Rapti
, and
M. J.
Manos
,
Prog. Mater. Sci.
86
,
25
(
2017
).
23.
Q.
Zhang
,
J.
Yu
,
J.
Cai
,
L.
Zhang
,
Y.
Cui
,
Y.
Yang
,
B.
Chen
, and
G.
Qian
,
Chem. Commun.
51
,
14732
(
2015
).
24.
W. W.
Lei
,
D.
Portehault
,
D.
Liu
,
S.
Qin
, and
Y.
Chen
,
Nat. Commun.
4
,
1777
(
2013
).
25.
E.
Barry
,
A. U.
Mane
,
J. A.
Libera
,
J. W.
Elam
, and
S. B.
Darling
,
J. Mater. Chem. A
5
,
2929
(
2017
).
26.
M.
Khan
and
I. M. C.
Lo
,
J. Hazardous Mater.
322
,
195
(
2017
).
27.
A. A.
Gouda
and
S. M.
Al Ghannam
,
Food Chem.
202
,
409
(
2016
).
28.
H.
Rashidi Nodeh
,
W. A.
Wan Ibrahim
,
M. A.
Kamboh
, and
M. M.
Sanagi
,
Chemosphere
166
,
21
(
2017
).
29.
M.
Ghidiu
,
J.
Halim
,
S.
Kota
,
D.
Bish
,
Y.
Gogotsi
, and
M. W.
Barsoum
,
Chem. Mater.
28
,
3507
(
2016
).
30.
F.
Perreault
,
A. F.
de Faria
, and
M.
Elimelech
,
Chem. Soc. Rev.
44
,
5861
(
2015
).
31.
H. C.
Bi
,
X.
Xie
,
K. B.
Yin
,
Y. L.
Zhou
,
S.
Wan
,
L. B.
He
,
F.
Xu
,
F.
Banhart
,
L. T.
Sun
, and
R. S.
Ruoff
,
Adv. Funct. Mater.
22
,
4421
(
2012
).
32.
Q.
Wang
,
H.
Wang
,
S.
Xiong
,
R.
Chen
, and
Y.
Wang
,
ACS Appl. Mater. Interfaces
6
,
18816
(
2014
).
33.
E.
Barry
,
J. A.
Libera
,
A. U.
Mane
,
J. R.
Avila
,
D.
DeVitis
,
K.
Van Dyke
,
J. W.
Elam
, and
S. B.
Darling
,
Environ. Sci.: Water Res. Technol.
4
,
40
(
2018
).
34.
Q.
Ma
,
H.
Cheng
,
A. G.
Fane
,
R.
Wang
, and
H.
Zhang
,
Small
12
,
2186
(
2016
).
35.
J.
Lalley
,
C.
Han
,
X.
Li
,
D. D.
Dionysiou
, and
M. N.
Nadagouda
,
Chem. Eng. J.
284
,
1386
(
2016
).
36.
A.
Bhatnagar
and
M.
Sillanpaa
,
Chem. Eng. J.
168
,
493
(
2011
).
37.
H. B.
Yin
,
Y.
Yun
,
Y. L.
Zhang
, and
C. X.
Fan
,
J. Hazard. Mater.
198
,
362
(
2011
).
38.
B.
Saha
and
C.
Orvig
,
Coord. Chem. Rev.
254
,
2959
(
2010
).
39.
P.
Verlicchi
,
M.
Al Aukidy
, and
E.
Zambello
,
Sci. Total Environ.
429
,
123
(
2012
).
40.
C. S.
Lee
,
S. K.
Kim
, and
M.
Kim
,
Sensors
9
,
7111
(
2009
).
41.
B.
Wang
,
Q.
Yang
,
C.
Guo
,
Y.
Sun
,
L.-H.
Xie
, and
J.-R.
Li
,
ACS Appl. Mater. Interfaces
9
,
10286
(
2017
).
42.
W.
Liu
,
Y.
Wang
,
Z.
Bai
,
Y.
Li
,
Y.
Wang
,
L.
Chen
,
L.
Xu
,
J.
Diwu
,
Z.
Chai
, and
S.
Wang
,
ACS Appl. Mater. Interfaces
9
,
16448
(
2017
).
43.
W.
Liu
,
X.
Dai
,
Z.
Bai
,
Y.
Wang
,
Z.
Yang
,
L.
Zhang
,
L.
Xu
,
L.
Chen
,
Y.
Li
,
D.
Gui
,
J.
Diwu
,
J.
Wang
,
R.
Zhou
,
Z.
Chai
, and
S.
Wang
,
Environ. Sci. Technol.
51
,
3911
(
2017
).
44.
L.
Chen
,
J.-W.
Ye
,
H.-P.
Wang
,
M.
Pan
,
S.-Y.
Yin
,
Z.-W.
Wei
,
L.-Y.
Zhang
,
K.
Wu
,
Y.-N.
Fan
, and
C.-Y.
Su
,
Nat. Commun.
8
,
15985
(
2017
).
45.
T.-Y.
Gu
,
M.
Dai
,
D. J.
Young
,
Z.-G.
Ren
, and
J.-P.
Lang
,
Inorg. Chem.
56
,
4668
(
2017
).
46.
Q.
Lin
,
T.-T.
Lu
,
X.
Zhu
,
T.-B.
Wei
,
H.
Li
, and
Y.-M.
Zhang
,
Chem. Sci.
7
,
5341
(
2016
).
47.
W.
Xiong
,
L.
Zhou
, and
S.
Liu
,
Chem. Eng. J.
284
,
650
(
2016
).
48.
Z.
Guo
,
D. D.
Li
,
X. K.
Luo
,
Y. H.
Li
,
Q. N.
Zhao
,
M. M.
Li
,
Y. T.
Zhao
,
T. S.
Sun
, and
C.
Ma
,
J. Colloid Interface Sci.
490
,
11
(
2017
).
49.
S.
Chaiyo
,
E.
Mehmeti
,
K.
Žagar
,
W.
Siangproh
,
O.
Chailapakul
, and
K.
Kalcher
,
Anal. Chim. Acta
918
,
26
(
2016
).
50.
H.
Bagheri
,
A.
Hajian
,
M.
Rezaei
, and
A.
Shirzadmehr
,
J. Hazard. Mater.
324
,
762
(
2017
).
51.
Y.
Haldorai
,
J. Y.
Kim
,
A. T. E.
Vilian
,
N. S.
Heo
,
Y. S.
Huh
, and
Y.-K.
Han
,
Sens. Actuators B
227
,
92
(
2016
).
52.
R.
Gupta
,
P. K.
Rastogi
,
V.
Ganesan
,
D. K.
Yadav
, and
P. K.
Sonkar
,
Sens. Actuators B
239
,
970
(
2017
).
53.
S.
Mao
,
J.
Chang
,
H.
Pu
,
G.
Lu
,
Q.
He
,
H.
Zhang
, and
J.
Chen
,
Chem. Soc. Rev.
46
,
6872
(
2017
).
54.
A.
Maity
,
X.
Sui
,
C. R.
Tarman
,
H.
Pu
,
J.
Chang
,
G.
Zhou
,
R.
Ren
,
S.
Mao
, and
J.
Chen
,
ACS Sens.
2
,
1653
(
2017
).
55.
G.
Zhou
,
H.
Pu
,
J.
Chang
,
X.
Sui
,
S.
Mao
, and
J.
Chen
,
Sens. Actuators B
257
,
214
(
2018
).
56.
G.
Zhou
,
J.
Chang
,
H.
Pu
,
K.
Shi
,
S.
Mao
,
X.
Sui
,
R.
Ren
,
S.
Cui
, and
J.
Chen
,
ACS Sens.
1
,
295
(
2016
).
57.
A.
Lee
,
J. W.
Elam
, and
S. B.
Darling
,
Environ. Sci.: Water Res. Technol.
2
,
17
(
2016
).
58.
M. A.
Shannon
,
P. W.
Bohn
,
M.
Elimelech
,
J. G.
Georgiadis
,
B. J.
Marinas
, and
A. M.
Mayes
,
Nature
452
,
301
(
2008
).
59.
M.
Elimelech
and
W. A.
Phillip
,
Science
333
,
712
(
2011
).
60.
L. F.
Greenlee
,
D. F.
Lawler
,
B. D.
Freeman
,
B.
Marrot
, and
P.
Moulin
,
Water Res.
43
,
2317
(
2009
).
61.
J.
Yin
and
B.
Deng
,
J. Membr. Sci.
479
,
256
(
2015
).
62.
H.-C.
Yang
,
J.
Hou
,
V.
Chen
, and
Z.-K.
Xu
,
Angew. Chem. Int. Ed.
55
,
13398
(
2016
).
63.
R. N.
Zhang
,
Y. N.
Liu
,
M. R.
He
,
Y. L.
Su
,
X. T.
Zhao
,
M.
Elimelech
, and
Z. Y.
Jiang
,
Chem. Soc. Rev.
45
,
5888
(
2016
).
64.
S. R. V.
Castrillon
,
X. L.
Lu
,
D. L.
Shaffer
, and
M.
Elimelech
,
J. Membr. Sci.
450
,
331
(
2014
).
65.
X.
Fan
,
Y.
Su
,
X.
Zhao
,
Y.
Li
,
R.
Zhang
,
T.
Ma
,
Y.
Liu
, and
Z.
Jiang
,
J. Membr. Sci.
499
,
56
(
2016
).
66.
J. H.
Li
,
M. Z.
Li
,
J.
Miao
,
J. B.
Wang
,
X. S.
Shao
, and
Q. Q.
Zhang
,
Appl. Surf. Sci.
258
,
6398
(
2012
).
67.
Y. F.
Zhao
,
L. P.
Zhu
,
Z.
Yi
,
B. K.
Zhu
, and
Y. Y.
Xu
,
J. Membr. Sci.
440
,
40
(
2013
).
68.
M. Z.
Li
,
J. H.
Li
,
X. S.
Shao
,
J.
Miao
,
J. B.
Wang
,
Q. Q.
Zhang
, and
X. P.
Xu
,
J. Membr. Sci.
405
,
141
(
2012
).
69.
M.
Piatkovsky
,
H.
Acar
,
A. B.
Marciel
,
M.
Tirrell
, and
M.
Herzberg
,
J. Membr. Sci.
549
,
507
(
2018
).
70.
J.
Lee
,
H. R.
Chae
,
Y. J.
Won
,
K.
Lee
,
C. H.
Lee
,
H. H.
Lee
,
I. C.
Kim
, and
J. M.
Lee
,
J. Membr. Sci.
448
,
223
(
2013
).
71.
L.
Zou
,
I.
Vidalis
,
D.
Steele
,
A.
Michelmore
,
S. P.
Low
, and
J.
Verberk
,
J. Membr. Sci.
369
,
420
(
2011
).
72.
J.
Nikkola
,
J.
Sievanen
,
M.
Raulio
,
J.
Wei
,
J.
Vuorinen
, and
C. Y.
Tang
,
J. Membr. Sci.
450
,
174
(
2014
).
73.
C.
Wenjuan
,
S.
Yanlei
,
P.
Jinming
,
D.
Yanan
,
Z.
Xueting
, and
J.
Zhongyi
,
Adv. Funct. Mater.
21
,
191
(
2011
).
74.
Y. F.
Li
,
Y. L.
Su
,
X. T.
Zhao
,
X.
He
,
R. N.
Zhang
,
J. J.
Zhao
,
X. C.
Fan
, and
Z. Y.
Jiang
,
Acs Appl. Mater. Interfaces
6
,
5548
(
2014
).
75.
X. T.
Zhao
,
Y. L.
Su
,
Y. F.
Li
,
R. N.
Zhang
,
J. J.
Zhao
, and
Z. Y.
Jiang
,
J. Membr. Sci.
450
,
111
(
2014
).
76.
X. T.
Zhao
,
Y. L.
Su
,
W. J.
Chen
,
J. M.
Peng
, and
Z. Y.
Jiang
,
J. Membr. Sci.
415
,
824
(
2012
).
77.
P.
Wang
and
T. S.
Chung
,
J. Membr. Sci.
474
,
39
(
2015
).
78.
E.
Yuliwati
and
A. F.
Ismail
,
Desalination
273
,
226
(
2011
).
79.
C. E.
Ren
,
K. B.
Hatzell
,
M.
Alhabeb
,
Z.
Ling
,
K. A.
Mahmoud
, and
Y.
Gogotsi
,
J. Phys. Chem. Lett.
6
,
4026
(
2015
).
80.
S. S.
Shenvi
,
A. M.
Isloor
, and
A. F.
Ismail
,
Desalination
368
,
10
(
2015
).
81.
X. B.
Hu
,
Z. X.
Chen
,
G. F.
Tang
,
J. L.
Hou
, and
Z. T.
Li
,
J. Am. Chem. Soc.
134
,
8384
(
2012
).
82.
H.
Wang
,
J.
Ding
,
L.
Dai
,
X.
Wang
, and
T.
Lin
,
J. Mater. Chem.
20
,
7938
(
2010
).
83.
J.
Wu
,
N.
Wang
,
L.
Wang
,
H.
Dong
,
Y.
Zhao
, and
L.
Jiang
,
Soft Matter
8
,
5996
(
2012
).
84.
X.
Tian
,
H.
Jin
,
J.
Sainio
,
R. H. A.
Ras
, and
O.
Ikkala
,
Adv. Funct. Mater.
24
,
6023
(
2014
).
85.
Z.
Wang
,
Y.
Wang
, and
G.
Liu
,
Angew. Chem.
128
,
1313
(
2016
).
86.
H.-C.
Yang
,
Y.
Xie
,
J.
Hou
,
A. K.
Cheetham
,
V.
Chen
, and
S. B.
Darling
, “
Janus Membranes: Creating Asymmetry for Energy Efficiency
,”
Adv. Mater.
(in press).
87.
I. S.
Cole
and
D.
Marney
,
Corrosion Sci.
56
,
5
(
2012
).
88.
I. J. H. G.
Vreeburg
and
D. J. B.
Boxall
,
Water Res.
41
,
519
(
2007
).
89.
J.
Vollertsen
,
A. H.
Nielsen
,
H. S.
Jensen
,
T.
Wium-Andersen
, and
T.
Hvitved-Jacobsen
,
Sci. Total Environ.
394
,
162
(
2008
).
90.
P.
Sarin
,
V. L.
Snoeyink
,
J.
Bebee
,
K. K.
Jim
,
M. A.
Beckett
,
W. M.
Kriven
, and
J. A.
Clement
,
Water Res.
38
,
1259
(
2004
).
91.
N.
Boulay
and
M.
Edwards
,
Water Res.
35
,
683
(
2001
).
92.
M.
Edwards
and
A.
Dudi
,
J. - Am. Water Works Assoc.
96
,
69
(
2004
).
93.
N. B.
Hallam
,
J. R.
West
,
C. F.
Forster
, and
J.
Simms
,
Water Res.
35
,
4063
(
2001
).
94.
F.
Liu
,
X. H.
Lu
,
W.
Yang
,
J. J.
Lu
,
H. Y.
Zhong
,
X.
Chang
, and
C. C.
Zhao
,
Desalination
313
,
18
(
2013
).
95.
F. C.
Moreira
,
R. A. R.
Boaventura
,
E.
Brillas
, and
V. J. P.
Vilar
,
Appl. Catal. B
202
,
217
(
2017
).
96.
J. J.
Pignatello
,
E.
Oliveros
, and
A.
MacKay
,
Crit. Rev. Environ. Sci. Technol.
36
,
1
(
2006
).
97.
I.
Sirés
,
E.
Brillas
,
M. A.
Oturan
,
M. A.
Rodrigo
, and
M.
Panizza
,
Environ. Sci. Pollution Res.
21
,
8336
(
2014
).
98.
B. P.
Vellanki
,
B.
Batchelor
, and
A.
Abdel-Wahab
,
Environ. Eng. Sci.
30
,
264
(
2013
).
99.
P. K.
Tandon
and
S. B.
Singh
,
Environ. Chem. Lett.
14
,
15
(
2016
).
100.
C. A.
Martinez-Huitle
and
S.
Ferro
,
Chem. Soc. Rev.
35
,
1324
(
2006
).
101.
E.
Guinea
,
C.
Arias
,
P. L.
Cabot
,
J. A.
Garrido
,
R. M.
Rodríguez
,
F.
Centellas
, and
E.
Brillas
,
Water Res.
42
,
499
(
2008
).
102.
E.
Brillas
,
I.
Sirés
, and
M. A.
Oturan
,
Chem. Rev.
109
,
6570
(
2009
).
103.
M.
Panizza
and
G.
Cerisola
,
Chem. Rev.
109
,
6541
(
2009
).
104.
K.
Cho
,
D.
Kwon
, and
M. R.
Hoffmann
,
RSC Adv.
4
,
4596
(
2014
).
105.
I.
Sirés
,
N.
Oturan
, and
M. A.
Oturan
,
Water Res.
44
,
3109
(
2010
).
106.
A.
Lee
,
J. A.
Libera
,
R. Z.
Waldman
,
A.
Ahmed
,
J. R.
Avila
,
J. W.
Elam
, and
S. B.
Darling
,
Adv. Sustainable Syst.
1
,
1600041
(
2017
).
107.
P. V.
Nidheesh
and
R.
Gandhimathi
,
Desalination
299
,
1
(
2012
).
108.
D.
Cambié
,
C.
Bottecchia
,
N. J. W.
Straathof
,
V.
Hessel
, and
T.
Noël
,
Chem. Rev.
116
,
10276
(
2016
).
109.
M. N.
Chong
,
B.
Jin
,
C. W. K.
Chow
, and
C.
Saint
,
Water Res.
44
,
2997
(
2010
).
110.
C.
Wang
,
X.
Liu
,
N.
Keser Demir
,
J. P.
Chen
, and
K.
Li
,
Chem. Soc. Rev.
45
,
5107
(
2016
).
111.
J.
Park
,
D.
Feng
,
S.
Yuan
, and
H. C.
Zhou
,
Angew. Chem. Int. Ed.
54
,
430
(
2015
).
112.
K.
Peng
,
L.
Fu
,
H.
Yang
,
J.
Ouyang
, and
A.
Tang
,
Nano Res.
10
,
570
(
2017
).
113.
L.
Wang
,
B.
Su
,
C.
Cheng
,
L.
Ma
,
S.
Li
,
S.
Nie
, and
C.
Zhao
,
J. Mater. Chem. B
3
,
1391
(
2015
).
114.
S.
Waheed
,
A.
Ahmad
,
S. M.
Khan
,
S-e
Gul
,
T.
Jamil
,
A.
Islam
, and
T.
Hussain
,
Desalination
351
,
59
(
2014
).
115.
K.
Krishnamoorthy
,
M.
Veerapandian
,
L.-H.
Zhang
,
K.
Yun
, and
S. J.
Kim
,
J. Phys. Chem. C
116
,
17280
(
2012
).
116.
F.
Perreault
,
A. F.
de Faria
,
S.
Nejati
, and
M.
Elimelech
,
ACS Nano
9
,
7226
(
2015
).
117.
B. E.
Logan
and
K.
Rabaey
,
Science
337
,
686
(
2012
).
118.
D.
Pant
,
G.
Van Bogaert
,
L.
Diels
, and
K.
Vanbroekhoven
,
Bioresource Technol.
101
,
1533
(
2010
).
119.
A.
Kundu
,
J. N.
Sahu
,
G.
Redzwan
, and
M. A.
Hashim
,
Int. J. Hydrogen Energy
38
,
1745
(
2013
).
120.
P. A.
Selembo
,
M. D.
Merrill
, and
B. E.
Logan
,
Int. J. Hydrogen Energy
35
,
428
(
2010
).