Electrosorption at functional interfaces: from molecular-level interactions to electrochemical cell design

Xiao Su and T. Alan Hatton *
Department of Chemical Engineering, Massachusetts Institute of Technology, Cambridge, Massachusetts 02139, United States. E-mail: tahatton@mit.edu

Received 1st May 2017 , Accepted 26th June 2017

First published on 29th June 2017

Adsorption at charged interfaces plays an important role across all aspects of physical chemistry, from biological interactions within living organisms to chemical processes such as catalysis and separations. With recent advances in materials chemistry, there are a host of modified electrodes being investigated for electrosorption, especially in separations science. In this perspective, we provide an overview of functional interfaces being used for electrosorption, ranging from electrochemical separations such as deionization and selective product recovery to biological applications. We cover the various molecular mechanisms which can be used to enhance ion capacity, and in some cases, provide selectivity; as well as discuss the parasitic Faradaic reactions which often impair electrosorption performance. Finally, we point to the importance of electrochemical configurations, in particular the advantages of asymmetric cell design, and highlight the opportunities for selective electrosorption brought about by redox-mediated systems.

image file: c7cp02822a-p1.tif

Xiao Su

Xiao Su obtained his BASc (Honours) in Chemical Engineering from the University of Waterloo, Canada in 2011, and joined MIT Chemical Engineering as a PhD student with an NSERC post-graduate fellowship. He completed his PhD thesis in 2016 under the supervision Professor T. Alan Hatton (Chemical Engineering) and Professor Timothy F. Jamison (Chemistry), with research focusing on organometallic interfaces for selective electrochemical separations. He is currently undertaking postdoctoral research at MIT, with research interests targeting redox-active interfaces for electrochemical separations, catalysis and sensing.

image file: c7cp02822a-p2.tif

T. Alan Hatton

T. Alan Hatton is the Ralph Landau Professor and Director of the David H. Koch School of Chemical Engineering Practice at MIT. He obtained BSc (Hons) and MSc degrees from the University of Natal, Durban, South Africa, and a PhD from the University of Wisconsin-Madison, all in Chemical Engineering, before joining MIT in 1982. He also holds positions as an Honorary Professorial Fellow at the University of Melbourne, and an Adjunct Professor at Curtin University, both in Australia. Professor Hatton is currently a Co-Director of the MIT Energy Initative (MITEI) Low Carbon Energy Center for Carbon Capture, Utilization and Storage. His research focuses on the synthesis, understanding and application of stimuli-responsive materials for the development of electrochemical processes to facilitate chemical separations and to mediate the transformation of captured waste to useful commodity chemicals. He has over 350 publications and more than 25 patents.

1. Introduction

Electrochemically driven processes have gained strong momentum in recent years due to greater awareness for environmental sustainability and green chemistry, and the rapid development of functional electrodes for a diversity of applications. The search for more efficient energy storage materials has led to an expansive set of available chemistries, ranging from nanostructured carbon to crystalline composites and conjugated polymers.1 At the same time, there has been intense interest in molecular recognition and supramolecular chemistry,2–4 including significant work on the discovery of novel electrochemically responsive species. From an environmental engineering perspective, the development of electrochemical processes for water purification and treatment, including capacitive deionization,5 electrodialysis,6 and separation processes in general,7,8 is occurring at a rapid pace. Electrochemical processes have also been proposed for the extraction of energy through salinity gradients between river and seawater (often referred to as “blue energy” extraction).9,10 In this perspective review, we focus on the recent advances and future directions in the field of electrosorption, the physico-chemical phenomenon that is at the core of most heterogeneous electrochemical processes.

Electrosorption is not only a central mechanism in electrochemical treatment operations, but is also intrinsic to the biological world. Interactions between cations and ion channels play a crucial role in metabolic regulation, dictating the transport rates and selectivity across cell membranes. In biomaterials, modulation of a functional electrochemical interface has been used for bio-sensing, neural prostheses, and even electrochemically mediated tissue engineering, for both cell adhesion and controlled release of therapeutics.11 Adsorption at solid–liquid interfaces is a key step in many chemical processes, e.g., in electrocatalysis, where adsorption of reactant molecules at the catalyst surface is the rate limiting step in many chemical transformations.12 Diffusion, surface bonding and molecular re-arrangement at the surface of an electrocatalyst dictate product selectivity across a range of reactions, including carbon dioxide electroreduction, hydrogen and oxygen evolution, and a wide range of organoelectrochemical processes.13,14

In separations science, the accumulation of ions in the electrical double-layer of a bare carbon or a functionalized electrode forms the basis for capacitive deionization.5 This process is a promising emerging technology in the environmental field, as it is an energy-efficient process to produce potable water from brackish natural sources, and even from wastewater effluents. In many other cases, however, it is the selective sorption of a target molecule in the presence of a multitude of competing species, and not removal of all salts, that is the ultimate goal, whether it is for the recovery of valuable commodity chemicals and pharmaceutical compounds, or simply the removal of certain hazardous pollutants from effluents.7 The modularity of electrosorption-based methods can lead to more efficient process intensification,15 in diverse chemical industries as well as environmental remediation and water treatment plants, by reducing solvent requirements and removing the need for chemical regenerants.

While various forms of carbon remain the focus of much of the research on the development of appropriate electrodes for these processes, chemically functionalized interfaces have garnered newfound attention due to the rich physico-chemical behaviour that they can impart to the electrodes, which often results in a many-fold increase in the specific cell capacitance, and allows control at a molecular level of the interfacial adsorption processes. There have been a variety of methods for creating functional surfaces, both carbon based,16–18 as well as polymeric, including post-synthetic modification, in situ growth, electropolymerization, and simple electrostatic adsorption, many of which have been covered elsewhere in detailed reviews.2,19–21

An understanding of the fundamental mechanisms driving electrosorption is essential for the effective design of novel electrode materials, especially for more efficient ion removal systems. The current perspective provides an overview of the advances in the field of electrochemically driven adsorption, including applications in water purification, chemical separations and bio-electrosorption. We also discuss the innovative routes by which the interfacial adsorption has been modulated by electrochemical potential on both pristine conductive surfaces and chemically functionalized electrodes. Molecular selectivity at the electrochemical interface can derive from inherent physical effects, dependent on electrostatics, ion properties and pore geometry, and from chemical modifications to the electrode surface. In particular, we highlight the importance that overall electrochemical configuration and electrode arrangement has in electrosorption performance, and how careful optimization of tandem chemistries at the anode and cathode can lead to more efficient systems for electrosorption in terms of both uptake capacity and separation factor.

2. Ion-adsorption at electrochemical interfaces

A classical electrochemical configuration is composed of a positive electrode, a negative electrode, and an electrolyte medium. Under charging, in the absence of Faradaic reactions, the electrified surface accumulates counter-ions (ions of the opposite charge) through Coulombic attraction, and expels co-ions (ions of the same charge) through Coulombic repulsion. A key concept in ion adsorption onto a conductive surface is the concept of the electrical double layer.22

The study of electrolyte ions adsorbing at charged interfaces has been extensive through theory, computational simulations and experiments.23–25 As shown in Fig. 1, in the classical Helmoltz model,23 the immediate structure of the double-layer is analogous to a dielectric capacitor, with a distance H between the planes, in which all counter-ions are adsorbed at the electrode surface. The Gouy–Chapman model assumes that the adsorbed ions, treated as point charges, are distributed within a diffuse layer near the charged surface.26,27 Stern28 combined the two models (Gouy–Chapman–Stern model) to describe a more complex structure in which there is a first compact layer of counter-ions which are immobile (which can be referred to as a Stern or Helmoltz layer), and a more diffuse layer in which counter-ions and co-ions coexist and the ions are mobile in accord with the Gouy–Chapman theory. Continuing development of transport and molecular models of the ion distributions near charged interfaces has improved our understanding of electrical-double layer systems,29 especially in porous structures.30,31 From a practical perspective, this accumulation of ions at charged interfaces and within microporous carbon electrodes has been the basis for supercapacitor technologies22 as well as capacitive deionization.5,32,33 Similarly, blue energy extraction using capacitive electrodes essentially relies on the compression and expansion of the electrical double-layer within macroporous and mesoporous carbon electrodes.9,10,34

image file: c7cp02822a-f1.tif
Fig. 1 Schematic representations of electrical double-layer models, with the surface electrical potential represented as Ψs and both the Helmoltz layer of thickness H and the diffuse layer shown. Adapted from ref. 25.

The study of the electrical double layer is not restricted to carbon electrodes, or aqueous solutions solely. The detailed structure of the double layer, especially at an ionic liquid/solid interface, has been extensively studied using molecular dynamics (MD) tools.35,36 These molecular level simulations have been used to accurately predict capacitances for energy storage materials, and even capture the effect of pore size effects.37 All-atom molecular dynamics simulations have also been performed to study Stern layer conductance and charge inversion effects at the electrochemical interface of silica nanochannels,38,39 in which the silanol chemistry at the surface plays an important role in determining electrosorption behavior. Furthermore, for silica/water interfaces, the Stern layer model has been carefully investigated using non-equilibrium molecular dynamics in comparison to the classical continuum theory, in an attempt to improve and provide a more realistic description.40

For adsorption of ionic species, symmetric electrolytes are usually adequate for aqueous supercapacitors used in energy storage applications, and are generally the target in simple capacitive deionization. Discrimination between different ions during the electrosorption process can, however, be important for chemical process separations, molecular sensing and even electrocatalysis.7 Selectivity can be achieved based on ion size and charge alone (purely electrostatic effects), and can be enhanced further through chemical functionalization of the electrodes, either by immobilized ligands or redox-active species. As discussed below, an understanding of both the physical and chemical driving forces at a molecular level is crucial for the design of more efficient electrosorption processes.

2.1. Ion-selectivity by physical effects

Electrosorption selectivity can occur solely by physical effects within porous carbon or metal structures, as ions can differ in size, charge and hydration radius. This topic has been studied extensively through both simulations and experiments, in which there is a complex interplay of factors dictating ion-selectivity towards various counter ions.41 Of special interest for both fundamental electrochemistry and practical water treatment applications is the electrosorption selectivity between ions of different valencies and sizes within nanopores, based purely on electrical and geometrical effects. Through Grand Canonical Monte-Carlo (GCMC) simulations and qualitative experimental observations,41 it has been observed that higher valency counter-ions tend to have an energetic advantage in electrosorption due to more favourable charge screening, while smaller monovalent counter-ions tend to have better access to the pores. There is a strong dependence of both these effects on surface charge density (e.g. applied potential or imposed current), and as such, the final electrosorption selectivity results from a competition between pore accessibility and charge screening. As charge density increases, divalent ions tend to adsorb more strongly than monovalent ions due to the more favourable energetic screening, until a point is reached at which, under very high surface charge densities, size-exclusion effects begin to dominate and monovalent smaller ions are again preferred.41,42Fig. 2a illustrates some of the experimental results from these studies. Interpretation of these observations must account for the hydrated size of the ions and accessibility to the pores of the actual porous electrode used in the experiments, where size-exclusion can bias the results towards preferential adsorption of the smaller ion, as was observed for K+ over Ca2+ under very low potential bias.41 Pore-size (micropore or mesopore) can also then play an important role in dictating the capacitance and electrosorption capacity of nanostructured carbon materials.43,44
image file: c7cp02822a-f2.tif
Fig. 2 (a) Experimental electrosorption with carbon electrodes of a mixture of monovalent cations and a mixture of a monovalent vs. a divalent cation. Reprinted with permission from ref. 41. Copyright 2008 from AIP Publishing. (b) Selectivity between monovalent and divalent cations from a flow-capacitive deionization cell. Reprinted with permission from ref. 45. Copyright 2015 from Elsevier.

The challenges and sometimes contradictory results of ion-selectivity in electrosorption from multicomponent electrolytes have been analyzed through a careful theoretical and experimental study, and results indicated that preferential adsorption depended both on thermodynamic equilibrium criteria and on adsorption kinetics.45 In a mixture of sodium and calcium (5[thin space (1/6-em)]:[thin space (1/6-em)]1 Na+[thin space (1/6-em)]:[thin space (1/6-em)]Ca2+), it was observed that the monovalent ion was first adsorbed due to its higher mobility, but that under further charging the divalent species was preferred, ascribed to energetic effects (Fig. 2b).45 Similar electrostatic preferences were observed with anions based on purely double-layer effects – smaller ions with smaller hydrated radii exhibited higher adsorption capacities, such as chloride over sulfate ions.46 However, if the adsorption capacity was normalized by ion valence, the adsorption selectivity accounting for charge follows the sequence trivalent anions > divalent anions > monovalent anions. It was also observed for monovalent ions that although smaller anions such as chloride have higher accessibility and initial selectivity, chemical affinity of the surface groups can eventually lead to ion-exchange with a more preferential target such as nitrate.46 Furthermore, the detailed transport of ions into porous conductive electrodes during electrosorption has been tackled experimentally with advanced spectroscopy techniques. Effective diffusion coefficients of ions within mesoporous carbon were measured using neutron imaging,47 and multi-nuclear magnetic resonance studies have been used to probe ion adsorption in microporous carbide electrodes.48

These studies serve to show that careful consideration of a range of factors is necessary to predict ion selectivity based on electrostatics, as it is dependent not only on energetics of sorption, but also on pore geometry and even adsorption process time. Furthermore, pure pristine surfaces are required for accurate measurements, as unaccounted surface chemistry can easily lead to spurious conclusions. From a practical perspective, it should be noted that unless very high charge densities are utilized (voltage or current), electrosorption selectivity of ions with non-functionalized electrodes based on electrostatics often yields quite moderate separation factors. It remains important to understand these inherent selectivities with carbon porous materials, as these effects may play a significant role in the energetics and performance of deionization systems for treatment of real waste or brackish water, in which there can be complex mixtures of electrolytes that are often asymmetric in charge and size.

2.2. Ion-selectivity by redox-mediated chemical effects

In both deionization and chemical separations, synthetic modification of the electrode surfaces is a promising approach for achieving higher ion capacities, and even more interestingly, selective electrosorption of dilute contaminants from solution. Indeed, ion-selectivity through specific chemical effects is at the heart of sensing and molecular recognition applications, in which the interactions are not purely electrostatic in nature, but depend also on charge-transfer phenomena, such as hydrogen bonding and covalent interactions, many of which can be electrochemically tuned.49,50 In particular, Faradaic species, which can undergo distinct changes in oxidation state based on an electron-transfer process, are efficient chemical and biological sensors, and have played a crucial role in the development of novel ion-selective functional materials.

Redox-active interfaces for selective electrochemical separations have been categorized extensively in a recent review.7 Cation-selective separations can be achieved with Faradaic intercalation compounds, often crystalline in nature, or with conjugated conductive polymers. Anion-selective separations have been the most widely studied with polymeric materials or surface immobilized anion-exchange ligands, due to their more favourable oxidation processes. Heterogeneous organic or organometallic interfaces, in particular metallopolymers, have been shown to be powerful platforms for selective anion sorption.51–54 Conjugated and redox-active polymers present an interesting class of materials for electrosorption from either aqueous or organic solutions, with the development of battery-type polymer materials offering a wide range of chemistries.55 Reversible, electrochemically mediated ion-doping has been demonstrated for poly(vinyl)ferrocene,56,57 polyaniline,58,59 polypyrrole,60 and polythiophene among others. Furthermore, both crystalline and polymeric materials can easily be synthesized or post-synthetically modified onto a range of substrates. For example, advances in the electropolymerization of pyrroles or thiophenes with pendant functional groups have offered a rich library of heterogeneous Faradaic reactions to be used in sensing and potential separations.2

Interest in metallopolymers, which contain main-group or d-block transition metals, or even f-block elements, in the polymer segments, has expanded rapidly recently in terms of both synthesis and characterization.51,54 Poly(vinyl)ferrocene was one of the first synthesized metallopolymers,51 and, due to the fast Faradaic electron transfer and electrochemical control of the redox-center, has been used in applications ranging from wound healing to supercapacitors. For selective electrosorption, ferrocene-based electrodes showed remarkable selectivity towards carboxylates, sulfonates and phosphonates in the presence of a greater than 30-fold excess of competing electrolytes in both aqueous and organic media (Fig. 3a).53 The materials platform, poly(vinyl)ferrocene conjugated carbon-nanotube electrodes (PVF-CNT) (Fig. 3b), allows for adsorption and release of these small molecules with separation factors >180 and ion adsorption capacities >250 mg g−1 based on adsorbent weight.53,61

image file: c7cp02822a-f3.tif
Fig. 3 (a) Redox-mediated electrosorption of target anions. (b) Surface structure of poly(vinyl)ferrocene electrodes. (c) Ferrocene–anion interactions optimized through DFT calculations. (c) Stoichiometric adsorption of carboxylates to ferrocene units in the presence of 33-fold excess competing perchlorate electrolyte. Reprinted with permission from ref. 53. Copyright 2016 from Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim.

The underlying mechanism for the strong selectivity of these electrodes for organic anions over other electrolytes was determined through a combination of spectroscopic observations and density-functional theory calculations to be an unusual carbon-mediated hydrogen bonding between the metallocene ring and the carboxylates, activated through electrochemical oxidation (Fig. 3c).53 The extra binding interaction provided by the directional charge transfer between the carboxylate oxygens and the cyclopentadienyl moieties enables attainment of high separation factors for the target analytes over electrolytes at stoichiometric ratios with respect to the iron centers (Fig. 3d),53,62 and has even hinted at significant selectivity between different carboxylates themselves based purely on the electronic structure.53,61

Anion recognition and sensing capabilities of metallocenes have been utilized for detection of a wider range of analytes in homogeneous sensor systems, and there is great potential to exploit these supramolecular interactions further in the separations field,7 as well as include more complex supramolecular polymers in the functionalization of the redox electrodes.63 Understanding the basic molecular interactions at functionalized interfaces is crucial for a variety of applications driven by electrosorption, as will be seen in the next few sections.

3. Electrosorption in biological applications

Electrosorption and associated molecular transport play a key role in the context of biological systems and biomaterials. The transport of calcium across cell membranes is dictated by voltage-gating through pores,64–66 where interactions of the permeating ions with the pore walls contribute to cation selectivity.67,68 In addition, the interaction of proteins and other biomolecules with electrochemical interfaces has garnered intense attention in recent years,11,69 with applications in areas such as enzymatic fuel cells and immobilized catalysis,20,70–72 bio-electronic systems,69 and biosensors.73–76

For proteins, the adsorption onto a surface can dramatically modulate the final film stability and activity, especially for enzyme-based devices,69 which are often limited by biocompatibility issues and electron accessibility. As emphasized in a recent review on biomimetic approaches for attaching enzymes to electrodes, the optimization of the distance between active sites is important,69 as well as the wiring between them. With redox enzymes, for example, polymers or organometallic mediators may be needed to facilitate electron-transfer between a metal surface and the entrapped enzymes.69 Proteins can either be adsorbed through charge interactions onto electrode surfaces, or be immobilized directly onto electrode surfaces through the help of conductive polymers. Through electrostatics, charged groups of an electrode can be selected based on the protein isoelectric point, to enhance surface attraction (Fig. 4a). Glucose oxidase enzymes were functionalized covalently with 3-carboxymethylpyrrole, and subsequently attached to the electrodes via co-polymerization of the monomer attached to the enzyme with pyrrole with preservation of enzyme activity.11,77 This covalent “wiring” effect has been shown to yield much higher activities than simple entrapment by charge balancing within a similar polypyrrole film. Finally, in a bio-inspired approach to preserve the native function of proteins, redox enzymes have been immobilized on a covalently attached lipid bilayer, resulting in significant retention of enzyme integrity after immobilization in this more native-like environment (Fig. 4b).78 The resulting architecture was formed through the self-assembly of proteoliposomes onto a surface with cholesterol tethers, and the resulting interface was demonstrated to facilitate electrochemical oxygen reduction.

image file: c7cp02822a-f4.tif
Fig. 4 (a) Charged protein interactions with a surface and (b) membrane protein immobilized onto a gold surface through the use of a tethered lipid bilayer, to provide a biomimetic environment. Reprinted with permission from ref. 69. Copyright 2017 from the Royal Society of Chemistry.

In biomolecular sensing, the adsorption of the analyte at the interface of the electrode is coupled with a change in electrical response, based on either current or potential. Highly sensitive DNA sensors can be made based on electrochemical principles,79 with polymer-modified electrodes, homogeneous redox-reporters or even functionalized nanoparticles. The main aim of sensor development is specificity and sensitivity, and in this regard electrochemical methods have a great advantage over optical and mass sensors in that the electronic signal is obtained directly, without the need for additional signal transduction.79,80 Electrochemical adsorption also plays a key role in protein-sensing specificity in detection systems as varied as carbon nanotubes, silica and nanoparticles.81–83 In field-effect biosensors (bioFETs) for example, atomistic simulations have shown that the double layer, in particular the Stern layer, plays a key role in mediating the interaction between the dissolved ions, DNA and the electrified interface.84 When the biomolecule binds at the electrode, the sensors detect perturbations in the electrical field. The work suggested that mixed electrolyte solutions, in particular, a divalent salt such as magnesium, could contribute to enhancing the electrical response.84

In bioseparations, use of electrochemical means has been shown to improve the modularity of the separation process and remove the need for chemical regenerants. In particular, since many biomacromolecules are charged and have different properties, selective electrosorption can be achieved through intrinsic charge modulation with a conductive surface, and even further enhanced through electrode functionalization. In a very recent work,85 poly(vinyl)ferrocene dip-coated electrodes have been shown to enable modulation of protein interactions very effectively, enabling the potential-controlled adsorption and release of various proteins (Fig. 5a), and achieve varying adsorption capacity and separation factors based on differences in size, charge and even charge distribution (Fig. 5b). Electrosorption onto these PVF–CNT electrodes was also shown to be an effective way to immobilize enzymes, with up to 90% retention of activity of chymotrypsin and lysozyme even under oxidizing conditions, as only low voltages were required for this immobilization (Fig. 5c). This shows that functional redox-interfaces potentially can provide an interesting alternative to packed bed bio-reactors – we can now enable enzyme attachment through a soft electrochemical modulation of the sorption process.

image file: c7cp02822a-f5.tif
Fig. 5 (a) Use of heterogeneous redox-electrodes for modulating surface–protein interactions through electrochemical control. (b) Adsorption capacity of PVF-CNT for various proteins when compared to their isoelectric point (pI) (LYS – lysozyme, MYO – myoglobin, HRP – horse radish peroxidase, R-A – ribonuclease, a-CHY – alpha-chymotrypsin, BSA – bovine serum albumin). (c) Comparison of the enzymatic activity of alpha-chymotrypsin before adsorption, immobilized on the surface, and after release. Reprinted with permission from ref. 85. Copyright 2017 from the American Chemical Society.

Overall, electrosorption of biological compounds is a promising frontier for electrochemistry. From a purely synthetic perspective, the development of novel protein and enzyme immobilization routes can enhance the final electron transfer properties. At the same time, successful attachment or recovery of biologically active species is often a delicate process – for enzymes, a hydrated environment at the interface, mimicking their native environment, helps maintain their stability and function.69 Further developments in biological electrosorption must improve electrochemical performance as well as biocompatibility, as these two often walk hand in hand.

4. Electrosorption in environmental separations

Providing potable water to the world remains one of the main engineering and scientific challenges of the 21st century. Water purification and treatment is a multifaceted problem, ranging from household-level filtration, through industrial effluent and municipal wastewater treatment, all the way to national-scale programs for seawater and brackish water desalination.86 For desalination, membrane-based methods such as reverse osmosis and electrodialysis occupy a large share of the market.87 Electrosorption-based techniques, on the other hand, inherently do not rely on the use of a membrane (though it can help improve performance), but rather on the direct accumulation of ions within the double-layers established at the electrode interfaces.

Porous, high charge capacity electrodes form the basis for capacitive deionization (CDI), an emerging technology for removing ionic species from aqueous streams that has received intense attention over the past decade, both at a fundamental research level as well as for scale up and commercialization.5,33 In many ways, capacitive deionization (CDI) is one of the most direct applications of electrosorption. There have been detailed reviews on all aspects of CDI research, ranging from the materials chemistry, electrode configurations, fields of applications and energetic as well as economic viability of these methods.88–90 In particular, carbon-materials have been a major focus, with strong advances in the making of aerogels and activated carbon more efficient in the recovery of salts.91–93 Through the tuning of pore size,94 surface chemistry,95,96 and a range of other materials properties,44 salt removal capacity can be improved significantly.

There have been a series of architectures proposed for capacitive deionization systems (Fig. 6 and 7), which vary in terms of the flow direction of the feed stream, the size and form factors of the electrode materials, the arrangements of the electrodes, and the presence or absence of a membrane. The orientation of the electrodes (perpendicular or parallel) to the flow direction can be varied, for instance, such as in flow-through capacitive deionization. When compared to the conventional flow-between type systems (flow parallel to electrode plates), flow-through capacitive deionization, accomplished by placing the electrodes perpendicular to the feedstream, has been shown to provide faster kinetics and the possibility of addressing higher feed concentration reductions.97 In membrane capacitive deionization (MCDI), an anion exchange membrane at the anode, and a cation exchange membrane at the cathode help prevent co-ion expulsion back to the fluid stream and thus enhance capacitive deionization, with lower energy requirements than conventional CDI.89 In another variation, the active sorptive material itself flows as a particle suspension, a process referred to as flow electrode capacitive deionization (FCDI).22,98 In such a configuration, the electrode configuration is actually in a fluidized-bed type of setting, with a recirculation that allows for adsorption in a flow cell and regeneration in a separate chamber, allowing for continuous operation (see Fig. 7). Recent advances in this configuration have steadily increased the particle loading in the flowable bed to 35 wt% of carbon, with strong desalination performance for feedwater up to 20 mM salt concentration.98 Flow-electrodes have also been combined with membranes to improve salt removal efficiencies.22

image file: c7cp02822a-f6.tif
Fig. 6 Flow-diagrams adapted from Porada et al., showing the various modes for CDI – (a) flow-by mode, (b) flow-through mode, (c) ion-pumping and (d) batch adsorption and discharge. Reprinted with permission from ref. 8, originally adapted from ref. 33. Copyright 2016 from Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim.

image file: c7cp02822a-f7.tif
Fig. 7 Various configurations for flow electrode capacitive deionization. Reprinted from ref. 5. Copyright 2015 from the Royal Society of Chemistry.

While capacitive deionization has been suggested as an efficient desalination process, especially for brackish water,87,99,100 there are barriers to its implementation at higher salinities,101 and studies on fouling and increase of electrode robustness are still required.5,101 Efforts are on-going to understand these limitations and overcome them in order to implement capacitive deionization for high-salinity applications. A combined experimental and computational modeling work was carried out to investigate the desalination of high-salinity solutions, including the use of a novel neutron imaging technique to view lithium ions within mesoporous carbon.102 It was observed that at high salinity, the discharge process could be slowed down due to ion accumulation during the regeneration step within the pores, which leaves diffusion (as opposed to electromigration) transport as the main ion recovery process.102 Further chemical development of carbon electrodes, different operation modes (such as flow-through electrodes) or the use of membranes could potentially mitigate these problems going forward.

With these limitations and future challenges in mind, capacitive deionization does offer an alternative to established methods such as reverse osmosis (RO) and electrodialysis (ED) that could result in lower energetic costs for specific applications,103 and represents a promising electrosorption-based method for practical implementation. Furthermore, as recent reviews and studies have pointed out,5,7,62 Faradaic materials have the opportunity to play an important role in improving charge efficiency, voltage windows, and ion capacity. Battery-type materials have been used for achieving significant salt removal capacities at comparable energy costs to reverse osmosis,104 as well as efficient “mixing entropy batteries” for blue energy generation.105

For small molecules, the applications of electrosorption are not only limited to brackish water deionization, but also encompass the treatment of industrial effluents or specialized environmental separations.106 Electrosorption has emerged as a method to further improve both rate and capacity of adsorption of various chemical species through the imposition of an electrical field, and offers a regeneration route dependent purely on switching the electrochemical potential, thus decreasing operating costs. Regeneration of fixed-bed adsorbents is usually either energetically intensive due to the high temperatures needed to regenerate it (up to 70% of the operating costs such as for activated carbon),106 or there is a need for a high concentration of chemical regenerants. Adsorption of wastewater contaminants aided by an electrical field has been studied extensively with carbon materials,106,107 especially for organic pollutants such as dyes and aromatic organic acids.108–110

Activated carbon, whether aided or not by an electrical field, has been recognized as one of the most efficient adsorbents for wastewater contaminants due to its high capacity and interaction with organic compounds.111 Its surface is often both chemically and physically heterogeneous, with carboxylic and phenolic groups that can provide surface acidity and additional interactions with species in solution.106 For the adsorption of electrically neutral (uncharged) contaminants, a small bias by the electrochemical potential of the surface charge is needed to adsorb the species; however, these can be easily expelled through the application of an opposite potential beyond the potential of zero charge (Epzc).106Fig. 8a qualitatively illustrates these relations by showing that potentials anodic to Epzc result in higher anion adsorption, while cathodic potentials result in higher cation adsorption, and uncharged species bind most strongly close to the potential of zero charge. This has been shown to be an efficient removal system for a range of phenolic compounds ranging from sulfonates, carboxylates to alcohols, as well as pyridines, with an adsorption/desorption cycle shown for methylquinolinium at +0.6 V, +0.3 V and 0 V vs. Ag/AgCl (Fig. 8b). As can be seen, the electrode desorbs the cationic methylquinolinium species at higher potentials (+0.6 V in this particular system).

image file: c7cp02822a-f8.tif
Fig. 8 (a) Representation of adsorption dependence of the various species at different potentials, relative to the point of zero charge. Adapted from ref. 106. (b) Adsorption/desorption cycle for methylquinolinium at activated carbon surfaces. Reprinted with permission from ref. 106. Copyright 1998, Chapman and Hall.

In wastewater treatment, selective electrosorption of cations was effected with redox-active films decades ago through the so-coined “electrically switched ion exchange” (ESIX) systems, which were initially proposed with nickel hexacyanoferrate systems for adsorbing cesium from nuclear rejects.112–114 Chemically, Prussian Blue compounds and their analogues have exhibited ion-sieving properties as-synthesized,115–117 but such interactions have been reported to be enhanced under an applied electric potential. Fig. 9 presents the crystal structure and electron microscopy images of copper hexacyanoferrate particles. The mechanism for sorption of ions into these crystals has been shown to be complex, with diffusion of cesium occurring spontaneously, followed by the formation of a new crystalline phase.116,118

image file: c7cp02822a-f9.tif
Fig. 9 (a) Scanning electron microscopy (SEM) images of K2CuFe(CN)6 crystals prepared by local growth. (b) Representation of the crystal structure of K2CuFe(CN)6. Reprinted with permission from ref. 116. Copyright 2004 Elsevier Inc.

Since the proposal of ESIX, analogous concepts have been pursued with different materials and ions, such as polyaniline for nitrate,119 composites of Mn-O2/PPy/PSS for lithium,120 and mixed zirconium phosphate/polyaniline films for targeting lead (Pb2+).121 A main challenge with the crystalline materials, especially hexacyanoferrates, is the stability of the electrodes, and thus various methodologies for improving their morphology and composition have been proposed.122,123 Despite these hurdles, the highly selective intercalation mechanism within these hexacyanoferrate compounds still makes them among the most popular electrochemically switched materials for selectively targeting cationic wastewater contaminants. Redox-active polymeric materials, on the other hand, are in general more suited for anion-selective binding due to their more favourable oxidation reactions.

The use of non-Faradaic affinity ligands on the electrode surfaces and the modulation of electrochemical potential for the recovery of uranium has been proposed in a very recent work.124 Uranium was extracted from seawater, in the presence of high ionic strength and competing ions, through a “half-wave rectified alternating current electrochemical method” (HW-ACE),124 with significant increases in extraction capacities over physicochemical methods (∼2000 mg g−1 by HW-ACE) at 2000 ppm of uranium concentration in spiked seawater over 24 h of adsorption. As shown in Fig. 10, an electrode was functionalized with amidoxime polymers, which have demonstrated adsorption capacities for uranium for inherent physico-chemical adsorption. However, this adsorption capacity was enhanced electrochemically through the use of an alternating potential which removed non-specific ion adsorption from the electrode surface through coulombic repulsion, and allowed for ion-rearrangement and growth of the attached uranium oxide particles. The uranyl ions remained on the surface due to the specific amixodime ligands.124

image file: c7cp02822a-f10.tif
Fig. 10 Depicted is the concept of using modulating potential during adsorption to allow for freeing active sites of undesired ions. The electrodes are functionalized with amidoxime for the extraction of uranyl ions. Reprinted from ref. 124. Copyright 2017 Nature Publishing Group.

5. Faradaic reactions and asymmetric electrochemical design

In the discussions so far, we have considered mostly the interactions and physico-chemical processes at the interface of a single, functional electrode. However, the general configuration of the electrochemical cell often plays an important role in the performance of the process or in measurements of electrochemical properties. The spacing, form factor, and importantly, the chemistry at the interface of the electrodes can all ultimately impact the behaviour of the electrochemical system. Often in analytical electrochemical studies, the counter-electrode, be it the cathode or the anode, is neglected in favour of studies focusing solely on the working electrode. However, in practical implementation of electrochemical processes, the efficiency of the counter-electrode can have a significant impact on both the electrochemical charging behaviour of the working electrode, and on the solution chemistry.

In a general scheme, the adsorption of certain ions at the counter-electrode will probably, depending on the concentration and ionic strength, affect counter-ion and co-ion distribution at the working electrode, especially for desalination purposes. Second, and related to ion-adsorption, are the charge storage properties, as the kinetics of electrosorption must be well matched on both electrodes for the cell to operate efficiently. Third, and often neglected in many studies, are the Faradaic side-reactions at one electrode that often impact the overall solution quality, whether it be degradation of the electrode, breakdown of the electrolyte, catalytic conversion of an impurity species, or even destruction of the solvent medium itself. In aqueous systems, a range of water and oxygen reactions can occur; these reactions can produce or consume protons and hydroxides, and in non-aqueous media, the decomposition of the solvent or electrolyte can occur through a range of complex pathways.125 Some of the aqueous reactions that can occur at the cathode or anode, depending on the pH of the solution, are shown below:125,126


2H+ + 2e → H2

2H2O + 2e → H2 + 2OH

O2 + 2H2O + 4e → 4OH


2H2O → O2 + 4H+ + 4e

Csurface + H2O → C[double bond, length as m-dash]Osurface + 2H+ + 2e

The selectivity towards the different water splitting and oxygen reduction reactions depends on both potential and pH, but it has been observed that these solvent-based reactions affect the pH significantly as well as impact the deionization performance.126 The parasitic side-reactions are detrimental as protons and hydroxides compete with salt adsorption, and continuous cycling often degrades the carbon electrode.90,126,127 A lowering of the operating voltages can often minimize the impact of these reactions, but also alter EDL-based electrosorption for capacitive-type technologies severely.

In the energy storage field, the choice of electrochemical configuration is often allied closely with the selection of the materials chemistry at each electrode. At a general level, cell configuration can be classified into symmetric arrangements and asymmetric arrangements, as shown in Fig. 11.62 In most systems used in deionization, symmetric, non-Faradaic electrodes are utilized due to their reliance on double-layer charge, which follows to a great extent the configurations used in supercapacitor work. For Faradaic systems, however, asymmetric arrangements have been shown to have much higher energy storage capacities, in terms of both capacitance and energy density.128–131 Asymmetric electrochemical capacitors, especially those with fast and reversible Faradaic processes, are able to extend the operating voltage windows of these energy storage devices in aqueous electrolytes (Fig. 12).132 Manganese-oxide based electrode materials have much higher performance metrics in an asymmetric configuration, and depending on the chosen counter-electrode, the specific energies can vary from 20 W h kg−1 to over 100 W h kg−1.132

image file: c7cp02822a-f11.tif
Fig. 11 Summary of possible cell arrangements. Configuration (A) represents the conventional, non-Faradaic symmetric system. (B) Has only one electrode redox-functionalized, the counter is only conductive. (C) Denotes a symmetric Faradaic system with both electrodes functionalized with the same redox-species. (D) Shows an asymmetric system in which one of the electrodes is in a different oxidation state, but they are chemically identical. (E) Presents a fully asymmetric system in which the anode and the cathode have redox-functionalities that are distinct. Adapted from ref. 62.

image file: c7cp02822a-f12.tif
Fig. 12 Ragone plots for a series of asymmetric and symmetric electrochemical capacitors operating in aqueous media. (a) Manganese-oxide (MnO2) based systems and (b) alternative systems based on anthraquinone (CAQ), polypyrrole (PPY), poly(3,4-ethylenedioxythiophene) (PEDOT), and dihydroxybenzene (CBDH). Other components compared include AC (activated carbon) and carbon (C), and metal oxides (RuO2, NiO). Reprinted with permission from ref. 132. Copyright 2011 from Materials Research Society.

In desalination, even with non-Faradaic electrodes, the concept of asymmetrically functionalized electrodes has been shown to minimize co-ion effects in reducing electrosorption efficiency.96,133 This enhanced CDI performance has been demonstrated with immobilized charged groups on the carbon substrate (Fig. 13)96 – CNT surfaces were coated with either sulfonate groups (–SO3, referred to as CS) or primary amino groups (–NH3+, referred to as CN), depending on whether positive or negative charge was desired. It was found that salt removal efficiency increased with the use of a CS//CNT cell or a CN//CNT cell, but that the CS//CN cell did not offer any salt removal efficiency advantages, although it did increase the electrosorption rate significantly, which could be due to the altered properties of the pore walls.96 The asymmetric use of a Nafion-activated carbon composite vs activated carbon has also been shown to increase the desalination capacity by over 30% due to reduction of co-ion repulsion effects.133

image file: c7cp02822a-f13.tif
Fig. 13 Non-Faradaic cell assemblies of functionalized CNT electrodes. Reprinted with permission from ref. 96. Copyright 2011 Elsevier.

The presence of immobilized surface charges can extend the voltage windows for CDI operation and result in higher salt adsorption capacities.134 In addition, extensive studies on the effect of electrode asymmetry on both capacitive and membrane capacitive deionization performance with carbon electrodes have indicated that electrode performance is often related to the point of zero charge Epzc of carbon.135 Through cycling and oxidation of the carbon (formation of oxygenated groups on the carbon surface), the Epzc was shifted, with different deionization performance depending on its value. With properly tuned zero charge potentials (reported to be 0.5 V Epzc cathode and −0.1 V Epzc anode), asymmetric membrane capacitive deionization cells were shown to have enhanced ion adsorption capacity and to mitigate side-reactions.135 Similarly, it has been noted that conductive polymer coatings also reduce losses from parasitic effects and enhance the cycling stability of carbon electrodes at the anode.136

For Faradaic ion-selective systems, the use of asymmetric arrangements becomes even more important – protons and hydroxides from parasitic reactions can severely affect the ion-selective behaviour of functionalized interfaces.62 Electrosorption, especially with Faradaic surfaces, can be highly pH sensitive, and the dramatic pH excursions that can occur at counter-electrodes can affect the ion-binding phenomena at the working electrode dramatically. To address this challenge, a combination of molecular chemistry and engineering design on the electrochemical cell configuration becomes important. With organometallic interfaces based on asymmetric metallocenes, the matching of the electron-transfer kinetics (see Fig. 14) between the cathode and the anode yields energetically efficient asymmetric Faradaic systems with very high uptake capacities and separation factors.62 Cobalt-organometallic electrodes, more specifically poly(2-(methacrylolyoxy)ethyl cobaltocenium (PMAECoCp2-CNT)) and poly[η5-(1-oxo-4-methacryloyloxy-butyl)cyclopentadienyl]cobalt(η4-tetraphenylcyclobutadiene) (PCoCpCb),137,138 were used as asymmetric counter electrodes to poly(vinyl)ferrocene (PVF), and both PVF-CNT//PMAECoCp2-CNT and PVF-CNT//PCoCpCb showed an order of magnitude improvement in ion-selective adsorption of organic pollutants, in the presence of 100-fold excess competing electrolyte. Parasitic side-reactions at the cathode were shown to be completely supressed, with measurements in both the liquid-phase (pH, reflecting OH production) and the gas headspace (for possible evidence of H2 formation) pointing to close to 100% Faradaic efficiency for the self-exchange reactions of the organometallics. Furthermore, the redox-window for the asymmetric operation was as low as +0.6 V, thus potentially leading to lower energy costs when compared to regular CDI processes, which require voltage windows between 1.2 and 1.5 V. Functionalization of both electrodes with complementary chemistries not only preserves the overall solution chemistry, but also enables tandem removal of a diverse range of anionic and cationic contaminants in solution together; the PVF-CNT//PCoCpCb cell was able to remove effectively both quinchlorac (anionic carboxylate) and paraquat (aromatic pyridinium) contaminants present initially at 100 micromolar concentrations, in the presence of a 100-fold excess of sodium perchlorate competing electrolyte (>40 mg g−1 uptake capacity in each electrode) – see Fig. 14c.

image file: c7cp02822a-f14.tif
Fig. 14 (a) Schematic representation of configuration E, in which the anode and cathode are coated with redox polymers. The absence of the redox polymer on the cathode allows for parasitic side reactions to occur. (b) PVF-CNT//PMAECoCp2-CNT asymmetric system with matched charges. (c) Adsorption efficiency of an asymmetric cell with PMAECoCp2-CNT when compared with a conductive counter, for removal of dilute emerging concern pollutants in the micromolar regime, in the presence of 100-fold competing electrolyte. Reprinted with permission from ref. 62. Copyright 2017 from Royal Society of Chemistry.

Finally, through similar principles, an asymmetric system composed of the PVF-CNT electrode and an electrode with polyanthraquinone grafted onto vertically aligned carbon nanotubes (PAQ-VACNTs) was shown to be highly effective for the tandem selective removal of carboxylates (formate, acetate, benzoate) through the specific interactions of carboxylates with poly(vinyl)ferrocene, and lithium through its binding to the redox-quinone centers, in both aqueous and organic systems, and with an order of magnitude higher uptake capacity than those reported for other CDI technologies (shown in Fig. 15).61 As with the cobalt counter-electrodes, the electrochemical charge of the PAQ-VACNT cathode was carefully matched with the metallopolymer anode to maximize electrochemical performance. The complete electrochemical swing process is shown in detail in Fig. 15a, in which a mixture of anions and cations is introduced to the electrochemical chamber, and under potential each redox-electrode undergoes its favourable Faradaic process, and adsorbs the targeted ionic species selectively. The valuable products thus adsorbed can be recovered by release into a fresh solvent phase on reversal of the potential stimulus.

image file: c7cp02822a-f15.tif
Fig. 15 (a) Schematic of the operation of an asymmetric pseudocapacitor system for selective recovery of lithium and carboxylates, using the PVF//PAQ system. (b) Comparison of electrosorption capacity of PVF-CNT//PAQ-CNT systems with carbon and other literature values. Reprinted with permission from ref. 61. Copyright 2016 from American Chemical Society.

The selectivity shown by the asymmetric systems so far, especially when coupled with redox-active electrodes, makes them interesting candidates for high-value added chemical process separations and ion-selective applications.

6. Summary and outlook

The rapid development of electrode materials and range of applications have contributed to make electrosorption an exciting topic for research in recent years. For biological, chemical and environmental processes, adsorption at a charged interface plays an important, and oftentimes crucial, role in dictating electrochemical performance. Complex physico-chemical mechanisms are often at play at a molecular level – understanding and tuning of these processes have allowed engineers and scientists to enhance electrosorption capacity and achieve remarkable selectivity, even in multicomponent ionic mixtures. As seen through numerous examples, chemical functionalization and design can help overcome the inherent thermodynamic limitations of conventional conductive interfaces, and achieve orders of magnitude increase in selectivity, often through the aid of redox-mediated systems such as intercalation crystals or electroactive polymers. At the same time, fundamental studies on electrical double-layers and the design of porous carbon adsorbents have advanced the performance of capacitive deionization processes immensely. Biological applications represent a novel frontier for electrochemical processes, with a rich pool of topics and associated application areas ranging from healthcare to bioengineering. Challenges going forward include improving the robustness of functional electrodes (carbon-based or otherwise), finding new chemistries to target ever more dilute and varied contaminants, and, through molecular design, to improve the electron-transfer rate and enhance the binding energy of single-site receptors.

From a macroscopic perspective, it has been seen that the process configuration and arrangement of the electrochemical cell can play an important role in dramatically improving electrosorption performance. In particular for applications in separations science, such as deionization and selective electrosorption, parasitic Faradaic reactions can be extremely harmful as they degrade the electrode, significantly affect the solution pH and divert current densities which could otherwise be used for desired processes. Development of asymmetric arrangements has been shown to mitigate these effects, and even suppress them completely, when properly designed electrode chemistries are utilized.

Electrochemical adsorption is a highly multidisciplinary field, with interest from chemistry, physics, biology, engineering and materials science, and at all levels, from the fundamental ionic studies to the engineering implementation for industrial applications. Going forward, we expect that further electrochemical design can expand the library of asymmetric systems for a variety of applications, for both bulk deionization and selective separations. In addition to the crystalline Faradaic electrodes, we expect that polymer-based or hybrid composite systems will be used more extensively in the field of electrosorption, as the molecular binding behaviour can be more easily tuned through synthetic methods. In terms of applications, purification or wastewater treatment will continue to garner intense and deserved interest, with novel configurations such as particle flow-deionization contributing to the improvement in salt-removal efficiency.

At the same time, selective electrochemical separations can be expected to play a more central role in the future development of the field, due to their economic importance for industry, in addition to exciting opportunities for physico-chemical investigations. With the implementation of highly tuned chemical binding sites, we can envision extensions of electrochemically based sorption methods to complex chemical synthesis purification, precious metal recovery and biological purification or assays, where these technologies can offer sustainable and modular alternatives for process intensification, with lower solvent usage and energetic costs, and no need for chemical regenerants. Finally, we expect rapid growth in both the creation of new materials for separating the minority ion from complex mixtures, and a focus towards more detailed understanding of interfacial electrochemistry through the application of both molecular and macroscopic modelling tools.


We would like to thank the J-WAFS Seed Fund program for research funding.


  1. G. P. Wang, L. Zhang and J. J. Zhang, Chem. Soc. Rev., 2012, 41, 797–828 RSC .
  2. S. J. Higgins, Chem. Soc. Rev., 1997, 26, 247–257 RSC .
  3. J. W. Steed, Chem. Soc. Rev., 2009, 38, 506–519 Search PubMed .
  4. J. H. R. Tucker and S. R. Collinson, Chem. Soc. Rev., 2002, 31, 147–156 Search PubMed .
  5. M. E. Suss, S. Porada, X. Sun, P. M. Biesheuvel, J. Yoon and V. Presser, Energy Environ. Sci., 2015, 8, 2296–2319 CAS .
  6. R. K. Nagarale, G. S. Gohil and V. K. Shahi, Adv. Colloid Interface Sci., 2006, 119, 97–130 Search PubMed .
  7. X. Su and T. A. Hatton, Adv. Colloid Interface Sci., 2017, 244, 6–20 Search PubMed .
  8. X. Su and T. A. Hatton, “Electrosorption” in Kirk-Othmer Encyclopedia of Chemical Technology, John Wiley & Sons, Inc., 2016, pp. 1–11 Search PubMed .
  9. D. Brogioli, R. Zhao and P. M. Biesheuvel, Energy Environ. Sci., 2011, 4, 772–777 CAS .
  10. K. Sharma, Y. H. Kim, S. Yiacoumi, J. Gabitto, H. Z. Bilheux, L. J. Santodonato, R. T. Mayes, S. Dai and C. Tsouris, Renewable Energy, 2016, 91, 249–260 Search PubMed .
  11. M. J. Higgins, P. J. Molino, Z. L. Yue and G. G. Wallace, Chem. Mater., 2012, 24, 828–839 Search PubMed .
  12. C. Zinola, Electrocatalysis: Computational, Experimental and Industrial Aspects, CRC Press, Taylor & Francis Group, Florida, US, 2010 Search PubMed .
  13. H. Lund and O. Hammerich, Organic Electrochemistry, Marcel, Dekker Inc., New York, 2001 Search PubMed .
  14. X. Su, L. Bromberg, K. J. Tan, T. F. Jamison, L. Padhye and T. A. Hatton, Environ. Sci. Technol. Lett., 2017, 4, 161–167 Search PubMed .
  15. Y. H. Kim, L. K. Park, S. Yiacoumi and C. Tsouris, Annu. Rev. Chem. Biomol. Eng., 2017, 8, 359–380 Search PubMed .
  16. I. Dumitrescu, P. R. Unwin and J. V. Macpherson, Chem. Commun., 2009, 6886–6901 Search PubMed .
  17. D. Vairavapandian, P. Vichchulada and M. D. Lay, Anal. Chim. Acta, 2008, 626, 119–129 Search PubMed .
  18. Y. Zhu, D. K. James and J. M. Tour, Adv. Mater., 2012, 24, 4924–4955 Search PubMed .
  19. R. D. A. Hudson, J. Organomet. Chem., 2001, 637, 47–69 Search PubMed .
  20. T. Ahuja, I. A. Mir, D. Kumar and Rajesh, Biomaterials, 2007, 28, 791–805 Search PubMed .
  21. S. Sadki, P. Schottland, N. Brodie and G. Sabouraud, Chem. Soc. Rev., 2000, 29, 283–293 Search PubMed .
  22. S. I. Jeon, H. R. Park, J. G. Yeo, S. Yang, C. H. Cho, M. H. Han and D. K. Kim, Energy Environ. Sci., 2013, 6, 1471–1475 CAS .
  23. H. Helmholtz, Ann. Phys., 1879, 243, 337–382 Search PubMed .
  24. A. J. Bard and L. R. Faulkner, Electrochemical Methods: Fundamentals and Applications, Wiley, 2nd edn, 2000 Search PubMed .
  25. H. Wang and L. Pilon, J. Phys. Chem. C, 2011, 115, 16711–16719 CAS .
  26. D. L. Chapman, Philos. Mag., 1913, 25, 475–481 Search PubMed .
  27. M. Gouy, J. Phys. Theor. Appl., 1910, 9, 457–468 Search PubMed .
  28. O. Stern, Z. Elektrochem. Angew. Phys. Chem., 1924, 30, 508–516 CAS .
  29. R. Burt, G. Birkett and X. S. Zhao, Phys. Chem. Chem. Phys., 2014, 16, 6519–6538 Search PubMed .
  30. G. Feng, R. Qiao, J. Huang, B. G. Sumpter and V. Meunier, ACS Nano, 2010, 4, 2382–2390 Search PubMed .
  31. P. M. Biesheuvel, Y. Q. Fu and M. Z. Bazant, Phys. Rev. E: Stat., Nonlinear, Soft Matter Phys., 2011, 83, 1–17 Search PubMed .
  32. Y. Oren, Desalination, 2008, 228, 10–29 Search PubMed .
  33. S. Porada, R. Zhao, A. van der Wal, V. Presser and P. M. Biesheuvel, Prog. Mater. Sci., 2013, 58, 1388–1442 Search PubMed .
  34. D. Brogioli, R. Ziano, R. A. Rica, D. Salerno and F. Mantegazza, J. Colloid Interface Sci., 2013, 407, 457–466 Search PubMed .
  35. C. Merlet, D. T. Limmer, M. Salanne, R. van Roij, P. A. Madden, D. Chandler and B. Rotenberg, J. Phys. Chem. C, 2014, 118, 18291–18298 CAS .
  36. C. Merlet, B. Rotenberg, P. A. Madden and M. Salanne, Phys. Chem. Chem. Phys., 2013, 15, 15781–15792 Search PubMed .
  37. C. Merlet, B. Rotenberg, P. A. Madden, P. L. Taberna, P. Simon, Y. Gogotsi and M. Salanne, Nat. Mater., 2012, 11, 306–310 Search PubMed .
  38. C. D. Lorenz, P. S. Crozier, J. A. Anderson and A. Travesset, J. Phys. Chem. C, 2008, 112, 10222–10232 CAS .
  39. C. D. Lorenz and A. Travesset, Phys. Rev. E: Stat., Nonlinear, Soft Matter Phys., 2007, 75, 1–5 Search PubMed .
  40. H. Zhang, A. A. Hassanali, Y. K. Shin, C. Knight and S. J. Singer, J. Chem. Phys., 2011, 134, 1–15 Search PubMed .
  41. C. H. Hou, P. Taboada-Serrano, S. Yiacoumi and C. Tsouris, J. Chem. Phys., 2008, 129, 1–9 Search PubMed .
  42. M. Valisko and D. Boda, J. Phys. Chem. C, 2007, 111, 15575–15585 CAS .
  43. C. H. Hou, C. D. Liang, S. Yiacoumi, S. Dai and C. Tsouris, J. Colloid Interface Sci., 2006, 302, 54–61 Search PubMed .
  44. C. Tsouris, R. Mayes, J. Kiggans, K. Sharma, S. Yiacoumi, D. DePaoli and S. Dai, Environ. Sci. Technol., 2011, 45, 10243–10249 Search PubMed .
  45. R. Zhao, M. van Soestbergen, H. H. M. Rijnaarts, A. van der Wal, M. Z. Bazant and P. M. Biesheuvel, J. Colloid Interface Sci., 2012, 384, 38–44 Search PubMed .
  46. Z. L. Chen, H. T. Zhang, C. X. Wu, Y. S. Wang and W. Li, Desalination, 2015, 369, 46–50 Search PubMed .
  47. K. Sharma, H. Z. Bilheux, L. M. H. Walker, S. Voisin, R. T. Mayes, J. O. Kiggans, S. Yiacoumi, D. W. DePaoli, S. Dai and C. Tsouris, Phys. Chem. Chem. Phys., 2013, 15, 11740–11747 Search PubMed .
  48. A. C. Forse, J. M. Griffin, H. Wang, N. M. Trease, V. Presser, Y. Gogotsi, P. Simon and C. P. Grey, Phys. Chem. Chem. Phys., 2013, 15, 7722–7730 Search PubMed .
  49. P. D. Beer and E. J. Hayes, Encyclopedia of Supramolecular Chemistry, Marcel Dekker, Inc., New York, 2003 Search PubMed .
  50. P. D. Beer and P. A. Gale, Angew. Chem., Int. Ed., 2001, 40, 486–516 Search PubMed .
  51. G. R. Whittell and I. Manners, Adv. Mater., 2007, 19, 3439–3468 Search PubMed .
  52. L. X. Ren, C. G. Hardy, S. F. Tang, D. B. Doxie, N. Hamidi and C. B. Tang, Macromolecules, 2010, 43, 9304–9310 Search PubMed .
  53. X. Su, H. J. Kulik, T. F. Jamison and T. A. Hatton, Adv. Funct. Mater., 2016, 26, 3394–3404 Search PubMed .
  54. M. A. Vorotyntsev and S. V. Vasilyeva, Adv. Colloid Interface Sci., 2008, 139, 97–149 Search PubMed .
  55. S. Muench, A. Wild, C. Friebe, B. Haupler, T. Janoschka and U. S. Schubert, Chem. Rev., 2016, 116, 9438–9484 Search PubMed .
  56. S. Bruckenstein, A. T. Fensore and A. R. Hillman, J. Electrochem. Soc., 1998, 145, L24–L26 Search PubMed .
  57. I. Jureviciute, S. Bruckenstein and A. R. Hillman, J. Electroanal. Chem., 2000, 488, 73–81 Search PubMed .
  58. Y. Kong, Y. Sha, S. K. Xue and Y. Wei, J. Electrochem. Soc., 2014, 161, H249–H254 Search PubMed .
  59. Q. Hao, W. Lei, X. Xia, Z. Yan, X. Yang, L. Lu and X. Wang, Electrochim. Acta, 2010, 55, 632–640 Search PubMed .
  60. T. Raudsepp, M. Marandi, T. Tamm, V. Sammelselg and J. Tamm, Electrochim. Acta, 2014, 122, 79–86 Search PubMed .
  61. D. S. Achilleos and T. A. Hatton, ACS Appl. Mater. Interfaces, 2016, 8, 32743–32753 CAS .
  62. X. Su, K. J. Tan, J. Elbert, C. Ruttiger, M. Gallei, T. F. Jamison and T. A. Hatton, Energy Environ. Sci., 2017, 10, 1272–1283 CAS .
  63. U. S. Schubert and C. Eschbaumer, Angew. Chem., Int. Ed., 2002, 41, 2892–2926 Search PubMed .
  64. B. Corry, T. W. Allen, S. Kuyucak and S. H. Chung, Biophys. J., 2001, 80, 195–214 Search PubMed .
  65. J. Yang, P. T. Ellinor, W. A. Sather, J. F. Zhang and R. W. Tsien, Nature, 1993, 366, 158–161 Search PubMed .
  66. B. Hille, Ionic Channels of Excitable Membranes, Sinauer, Sunderland, MA, 2001 Search PubMed .
  67. D. Boda, M. Valisko, B. Eisenberg, W. Nonner, D. Henderson and D. Gillespie, J. Chem. Phys., 2006, 125, 034901 Search PubMed .
  68. D. Boda, M. Valisko, B. Eisenberg, W. Nonner, D. Henderson and D. Gillespie, Phys. Rev. Lett., 2007, 98, 168102 Search PubMed .
  69. P. O. Saboe, E. Conte, M. Farell, G. C. Bazan and M. Kumar, Energy Environ. Sci., 2017, 10, 14–42 CAS .
  70. M. Gerard, A. Chaubey and B. D. Malhotra, Biosens. Bioelectron., 2002, 17, 345–359 Search PubMed .
  71. F. Ricci and G. Palleschi, Biosens. Bioelectron., 2005, 21, 389–407 Search PubMed .
  72. S. Cosnier, Electroanalysis, 1997, 9, 894–902 Search PubMed .
  73. P. N. Bartlett, P. R. Birkin, J. H. Wang, F. Palmisano and G. De Benedetto, Anal. Chem., 1998, 70, 3685–3694 Search PubMed .
  74. D. Belanger, J. Nadreau and G. Fortier, J. Electroanal. Chem., 1989, 274, 143–155 Search PubMed .
  75. S. Cosnier and A. Lepellec, Electrochim. Acta, 1999, 44, 1833–1836 Search PubMed .
  76. S. Cosnier, Biosens. Bioelectron., 1999, 14, 443–456 Search PubMed .
  77. B. F. Y. Yonhin and C. R. Lowe, J. Electroanal. Chem., 1994, 374, 167–172 Search PubMed .
  78. L. J. C. Jeuken, S. D. Connell, P. J. F. Henderson, R. B. Gennis, S. D. Evans and R. J. Bushby, J. Am. Chem. Soc., 2006, 128, 1711–1716 Search PubMed .
  79. T. G. Drummond, M. G. Hill and J. K. Barton, Nat. Biotechnol., 2003, 21, 1192–1199 Search PubMed .
  80. J. Wang, G. D. Liu and A. Merkoci, J. Am. Chem. Soc., 2003, 125, 3214–3215 Search PubMed .
  81. R. J. Chen, H. C. Choi, S. Bangsaruntip, E. Yenilmez, X. W. Tang, Q. Wang, Y. L. Chang and H. J. Dai, J. Am. Chem. Soc., 2004, 126, 1563–1568 Search PubMed .
  82. J. J. Gooding, Electrochim. Acta, 2005, 50, 3049–3060 Search PubMed .
  83. X. L. Luo, A. Morrin, A. J. Killard and M. R. Smyth, Electroanalysis, 2006, 18, 319–326 Search PubMed .
  84. B. M. Lowe, Y. Maekawa, Y. Shibuta, T. Sakata, C. K. Skylaris and N. G. Green, Phys. Chem. Chem. Phys., 2017, 19, 2687–2701 Search PubMed .
  85. X. Su, J. Hubner, M. J. Kauke, D. Dalbosco, J. Thomas, C. Gonsalez, E. Zhu, M. Franzreb, T. F. Jamison and T. A. Hatton, Chem. Mater., 2017 DOI:10.1021/acs.chemmater.7b01699 .
  86. M. A. Shannon, P. W. Bohn, M. Elimelech, J. G. Georgiadis, B. J. Marinas and A. M. Mayes, Nature, 2008, 452, 301–310 Search PubMed .
  87. M. A. Anderson, A. L. Cudero and J. Palma, Electrochim. Acta, 2010, 55, 3845–3856 Search PubMed .
  88. S. Porada, L. Borchardt, M. Oschatz, M. Bryjak, J. S. Atchison, K. J. Keesman, S. Kaskel, P. M. Biesheuvel and V. Presser, Energy Environ. Sci., 2013, 6, 3700–3712 CAS .
  89. R. Zhao, P. M. Biesheuvel and A. van der Wal, Energy Environ. Sci., 2012, 5, 9520–9527 CAS .
  90. A. Hemmatifar, J. W. Palko, M. Stadermann and J. G. Santiago, Water Res., 2016, 104, 303–311 Search PubMed .
  91. C. J. Gabelich, T. D. Tran and I. H. Suffet, Environ. Sci. Technol., 2002, 36, 3010–3019 Search PubMed .
  92. J. Biener, M. Stadermann, M. Suss, M. A. Worsley, M. M. Biener, K. A. Rose and T. F. Baumann, Energy Environ. Sci., 2011, 4, 656–667 CAS .
  93. Z. H. Huang, Z. Y. Yang, F. Y. Kang and M. Inagaki, J. Mater. Chem. A, 2017, 5, 470–496 CAS .
  94. L. Zou, L. Li, H. Song and G. Morris, Water Sci. Technol., 2010, 61, 1227–1233 Search PubMed .
  95. C. J. Yan, L. D. Zou and R. Short, Desalination, 2014, 333, 101–106 Search PubMed .
  96. J. Yang, L. D. Zou and N. R. Choudhury, Electrochim. Acta, 2013, 91, 11–19 Search PubMed .
  97. M. E. Suss, T. F. Baumann, W. L. Bourcier, C. M. Spadaccini, K. A. Rose, J. G. Santiago and M. Stadermann, Energy Environ. Sci., 2012, 5, 9511–9519 CAS .
  98. G. J. Doornbusch, J. E. Dykstra, P. M. Biesheuvel and M. E. Suss, J. Mater. Chem. A, 2016, 4, 3642–3647 CAS .
  99. P. Xu, J. E. Drewes, D. Heil and G. Wang, Water Res., 2008, 42, 2605–2617 Search PubMed .
  100. L. Zou, G. Morris and D. Qi, Desalination, 2008, 225, 329–340 Search PubMed .
  101. F. A. AlMarzooqi, A. A. Al Ghaferi, I. Saadat and N. Hilal, Desalination, 2014, 342, 3–15 Search PubMed .
  102. K. Sharma, Y. H. Kim, J. Gabitto, R. T. Mayes, S. Yiacoumi, H. Z. Bilheux, L. M. H. Walker, S. Dai and C. Tsouris, Langmuir, 2015, 31, 1038–1047 Search PubMed .
  103. A. Subramani and J. G. Jacangelo, Water Res., 2015, 75, 164–187 Search PubMed .
  104. M. Pasta, C. D. Wessells, Y. Cui and F. La Mantia, Nano Lett., 2012, 12, 839–843 Search PubMed .
  105. F. La Mantia, M. Pasta, H. D. Deshazer, B. E. Logan and Y. Cui, Nano Lett., 2011, 11, 1810–1813 Search PubMed .
  106. A. Ban, A. Schafer and H. Wendt, J. Appl. Electrochem., 1998, 28, 227–236 Search PubMed .
  107. A. S. Koparal, Y. Yavuz and U. B. Ogutveren, Water Environ. Res., 2002, 74, 521–525 Search PubMed .
  108. O. Gercel, Sep. Sci. Technol., 2016, 51, 711–717 Search PubMed .
  109. X. F. Sun, B. B. Guo, L. He, P. F. Xia and S. G. Wang, AIChE J., 2016, 62, 2154–2162 Search PubMed .
  110. E. Ayranci and O. Duman, J. Hazard. Mater., 2006, 136, 542–552 Search PubMed .
  111. G. Crini, Bioresour. Technol., 2006, 97, 1061–1085 Search PubMed .
  112. K. M. Jeerage and D. T. Schwartz, Sep. Sci. Technol., 2000, 35, 2375–2392 Search PubMed .
  113. M. A. Lilga, R. J. Orth, J. P. H. Sukamto, S. M. Haight and D. T. Schwartz, Sep. Purif. Technol., 1997, 11, 147–158 Search PubMed .
  114. S. D. Rassat, J. H. Sukamto, R. J. Orth, M. A. Lilga and R. T. Hallen, Sep. Purif. Technol., 1999, 15, 207–222 Search PubMed .
  115. P. A. Haas, Sep. Sci. Technol., 1993, 28, 2479–2506 Search PubMed .
  116. C. Loos-Neskovic, S. Ayrault, V. Badillo, B. Jimenez, E. Garnier, M. Fedoroff, D. J. Jones and B. Merinov, J. Solid State Chem., 2004, 177, 1817–1828 Search PubMed .
  117. M. Pyrasch, A. Toutianoush, W. Q. Jin, J. Schnepf and B. Tieke, Chem. Mater., 2003, 15, 245–254 Search PubMed .
  118. S. Ayrault, B. Jimenez, E. Garnier, M. Fedoroff, D. J. Jones and C. Loos-Neskovic, J. Solid State Chem., 1998, 141, 475–485 Search PubMed .
  119. M. H. Ansari and J. B. Parsa, Sep. Purif. Technol., 2016, 169, 158–170 Search PubMed .
  120. X. Du, G. Q. Guan, X. M. Li, A. D. Jagadale, X. L. Ma, Z. D. Wang, X. G. Hao and A. Abudula, J. Mater. Chem. A, 2016, 4, 13989–13996 CAS .
  121. Q. Zhang, X. Du, X. L. Ma, X. G. Hao, G. Q. Guan, Z. D. Wang, C. F. Xue, Z. L. Zhang and Z. J. Zuo, J. Hazard. Mater., 2015, 289, 91–100 CrossRef CAS PubMed .
  122. W. Chen and X. H. Xia, Adv. Funct. Mater., 2007, 17, 2943–2948 CrossRef CAS .
  123. Y. H. Lin and X. L. Cui, Chem. Commun., 2005, 2226–2228 RSC .
  124. C. Liu, P.-C. Hsu, J. Xie, J. Zhao, T. Wu, H. Wang, W. Liu, J. Zhang, S. Chu and Y. Cui, Nat. Energy, 2017, 2, 17007 CrossRef CAS .
  125. D. Pletcher, A first course in electrode processes, Royal Society of Chemistry, Cambridge, UK, 2009 Search PubMed .
  126. Y. Bouhadana, M. Ben-Tzion, A. Soffer and D. Aurbach, Desalination, 2011, 268, 253–261 CrossRef CAS .
  127. D. He, C. E. Wong, W. W. Tang, P. Kovalsky and T. D. Waite, Environ. Sci. Technol. Lett., 2016, 3, 222–226 CrossRef CAS .
  128. H. B. Li, M. H. Yu, F. X. Wang, P. Liu, Y. Liang, J. Xiao, C. X. Wang, Y. X. Tong and G. W. Yang, Nat. Commun., 2013, 4, 1894 CrossRef CAS PubMed .
  129. F. Beguin, V. Presser, A. Balducci and E. Frackowiak, Adv. Mater., 2014, 26, 2219–2251 CrossRef CAS PubMed .
  130. Z. J. Fan, J. Yan, T. Wei, L. J. Zhi, G. Q. Ning, T. Y. Li and F. Wei, Adv. Funct. Mater., 2011, 21, 2366–2375 CrossRef CAS .
  131. C. Zhou, Y. W. Zhang, Y. Y. Li and J. P. Liu, Nano Lett., 2013, 13, 2078–2085 CrossRef CAS PubMed .
  132. J. W. Long, D. Belanger, T. Brousse, W. Sugimoto, M. B. Sassin and O. Crosnier, MRS Bull., 2011, 36, 513–522 CrossRef CAS .
  133. W. S. Cai, J. B. Yan, T. Hussin and J. Y. Liu, Electrochim. Acta, 2017, 225, 407–415 CrossRef CAS .
  134. X. Gao, S. Porada, A. Omosebi, K. L. Liu, P. M. Biesheuvel and J. Landon, Water Res., 2016, 92, 275–282 CrossRef CAS PubMed .
  135. A. Omosebi, X. Gao, J. Landon and K. L. Liu, ACS Appl. Mater. Interfaces, 2014, 6, 12640–12649 CAS .
  136. X. Gao, A. Omosebi, N. Holubowitch, A. Liu, K. Ruh, J. Landon and K. Liu, Desalination, 2016, 399, 16–20 CrossRef CAS .
  137. C. Ruttiger, V. Pfeifer, V. Rittscher, D. Stock, D. Scheid, S. Vowinkel, F. Roth, H. Didzoleit, B. Stuhn, J. Elbert, E. Ionescu and M. Gallei, Polym. Chem., 2016, 7, 1129–1137 RSC .
  138. L. X. Ren, C. G. Hardy and C. B. Tang, J. Am. Chem. Soc., 2010, 132, 8874–8875 CrossRef CAS PubMed .

This journal is © the Owner Societies 2017