Anal Bioanal Chem DOI 10.1007/s00216-012-6525-2

REVIEW

Electrode interfaces switchable by physical and chemical signals for biosensing, biofuel, and biocomputing applications Evgeny Katz & Segiy Minko & Jan Halámek & Kevin MacVittie & Kenneth Yancey

Received: 28 August 2012 / Revised: 23 October 2012 / Accepted: 24 October 2012 # Springer-Verlag Berlin Heidelberg 2012

Abstract This review outlines advances in designing modified electrodes with switchable properties controlled by various physical and chemical signals. Irradiation of the modified electrode surfaces with various light signals, changing the temperature of the electrolyte solution, application of a magnetic field or electrical potentials, changing the pH of the solutions, and addition of chemical/biochemical substrates were used to change reversibly the electrode activity. The increasing complexity in the signal processing was achieved by integration of the switchable electrode interfaces with biomolecular information processing systems mimicking Boolean logic operations, thus allowing activation and inhibition of electrochemical processes on demand by complex combinations of biochemical signals. The systems reviewed range from simple chemical compositions to complex mixtures modeling biological fluids, where the signal substrates were added at normal physiological and elevated pathological concentrations. The switchable electrode interfaces are considered for future biomedical applications where the electrode properties will be modulated by the biomarker concentrations reflecting physiological conditions. Keywords Modified electrode . Signal-responsive material . Switchable electrode . Biocomputing . Logic gate . Biosensor

Published in the topical collection Bioelectroanalysis with guest editors Nicolas Plumeré, Magdalena Gebala, and Wolfgang Schuhmann. E. Katz (*) : S. Minko : J. Halámek : K. MacVittie : K. Yancey Department of Chemistry and Biomolecular Science, Clarkson University, Potsdam, NY 13699-5810, USA e-mail: [email protected]

Introduction A new concept of chemically modified electrode surfaces, introduced in 1970s and 1980s, was originally aimed at obtaining electrocatalytic, bioelectrocatalytic, and photoelectrocatalytic properties [1–3]. Rapid development of various chemically modified electrodes [4], particularly using self-assembling methods of surface modification pioneered in the 1980s [5, 6], resulted in progress in bioelectrochemical systems [7, 8] used in biosensors [9, 10] and biofuel cells [11, 12]. Broadening the research area and merging it with studies in materials chemistry, particularly in the field of stimuli-responsive materials [13–15], resulted in novel properties of modified electrodes. Modified electrode surfaces functionalized with stimuli-responsive materials attached to electrode surfaces as self-assembled monolayers or thin films produced switchable/tunable properties controlled by external signals [16, 17], thus resulting in the development of switchable biosensors [18] and switchable fuel cells [19], electrochemical information processing systems [20], and other novel electrochemical devices. “Switchable/tunable” properties of the electrode interfaces mean interfaces with reversibly variable properties (permeability for soluble redox species, redox activity of immobilized species) in response to the external physical or chemical signals. In the case of “switchable” interfaces, they undergo a sharp transition between two distinct physical states, whereas “tunable” interfaces allow a smooth transition between the different states. Depending on the properties of the material, various modified electrode surfaces with the switchable behavior controlled by light signals [21], a magnetic field [22, 23], temperature changes [24], an applied electrical potential [25, 26], and chemical/biochemical inputs [27, 28] have been designed. Recent advances in logic processing of chemical [29] and biochemical [30, 31] signals have resulted in increasing complexity of the signal-

E. Katz et al.

responsive materials and have allowed the realization of electrode interfaces with sophisticated properties controlled by complex combinations of chemical signals. Integrating chemical/biochemical computing systems with signal-responsive electrode interfaces allowed the integration of Boolean logic with the switchable electrode properties [32]. This article gives an overview of switchable electrochemical interfaces reversibly changing their properties in response to various physical and chemical signals, particularly emphasizing the importance of scaling up the complexity of the signal-processing systems by application of biomolecular logic systems integrated with signal-responsive interfaces and glimpses of the diverse challenges and opportunities in the near future.

Modified electrodes switchable by light signals Immobilization of photoisomerizable molecules (spiropyran [33, 34], azobenzene [35, 36], diarylethene [37, 38], and phenoxynaphthacenequinone [39] derivatives) on electrode surfaces allowed reversible switching of the electrode activity upon application of light signals with different wavelength. Photochemically activated interfaces of different complexity, ranging from polymeric thin films [37, 38] or self-assembled monolayers [39] to sophisticated molecular “machines” [40], were engineered to switch on–off electrochemical reactions. Some of the photoisomerizable species (e.g., diarylethene [37, 38] and phenoxynaphthacenequinone [39]) were directly responsible for alteration of the electrode activity, being redox-active in one of the isomeric states and redox-inactive when their structures were changed by the photochemical reaction. The redox-active states were used to mediate electrocatalytic cascades amplifying the interfacial redox transformations, whereas the inactive states inhibited these reactions. Another mechanism of photochemically activated switching of electrochemical reactions was based on the changing the shape or charge of the photoisomerizable molecules, thus resulting in “command interfaces” regulating access of diffusional redox species to the conducting support [33–35] or activating supramolecular electron-transporting “machines” [40]. For example, reversible photoisomerization of spiropyran moieties associated with an electrode surface resulted in protonation–deprotonation of the isomeric states (note that the spiropyran state is neutral, whereas the merocyanine state is protonated and positively charged) (Fig. 1), thus affecting the interfacial reactions of charged diffusional redox species [33, 34]. For example, cytochrome c, being a positively charged redox probe, has demonstrated reversible switchable electrochemical behavior on this interface [41]. Cyclic voltammograms recorded on the spiropyran state of the interface generated by its irradiation with visible light at 495 nm demonstrated reversible redox transformations of

Fig. 1 The electrochemical process of cytochrome c controlled by light signals applied to an electrode surface functionalized with spiropyran (SP)/merocyanine (MRH+) photoisomerizable species and pyridine units. Cytox and Cytred denote the initial oxidized state and the electrochemically generated reduced state of cytochrome c, respectively. (Adapted from [41], with permission; copyright American Chemical Society, 1995)

cytochrome c promoted by pyridine units in the mixed monolayer, whereas UV irradiation (350 nm<λ<395 nm) resulted in the inhibition of the cytochrome c redox changes owing to repulsion of cytochrome c from the positively charged merocyanine interface (Fig. 2). The reversible on/ off switching of the cytochrome c electrochemical reaction was achieved when visible and UV irradiation signals were applied stepwise to the modified electrode surface (Fig. 2, inset). Primary redox transformations of cytochrome c were also amplified by biocatalytic reactions in the presence of cytochrome c specific enzymes, thus allowing photochemical control over complex biocatalytic cascades [41].

Modified electrodes switchable by temperature changes Modification of electrode surfaces with a thermosensitive polymer, poly(N-isopropylacrylamide) (PNIPAM) brush,

Electrode interfaces switchable by physical and chemical signals

(ITO) electrode demonstrated reversible electrochemical behavior of the [Fe(CN)6]3− redox probe at low temperature (12 °C) when the PNIPAM film is permeable, whereas increasing the temperature to 42 °C resulted in the restructuring of the PNIPAM film, inhibiting penetration of the redox species to the conducting support and switching off the electrochemical reaction (S. Minko et al., unpublished work) (Fig. 4). The activation–inhibition of the electrochemical reaction was reversible and repeatable upon cyclic changes of the solution temperature. Thermoswitchable electrodes might find important biomedical applications in biosensors, cell culture scaffolds, and biofuel cells differently operating at physiologically variable temperature.

Magnetoswitchable electrode interfaces loaded with magnetic nanospecies/microspecies

Fig. 2 Cyclic voltammograms obtained for cytochrome c at different isomeric states of the electrode interface: SP produced by visible light (a); MRH+ generated by UV irradiation (b). The inset illustrates reversible switching between the activated and inhibited electrochemical process of cytochrome c modulated by light signals. SCE saturated calomel electrode. (Adapted from [41], with permission; copyright American Chemical Society, 1995)

allowed temperature control over electrochemical reactions [24, 42]. The temperature-controlled swelling–shrinking phase transition of the PNIPAM brush due to the temperature-dependent interaction of the monomer units with water resulted in different permeability of the modified interface for diffusional redox species. In the swollen state, the PNIPAM brush is permeable for the redox probe, whereas in the shrunken state the dense brush blocks transport of the probe to the electrode surface. Alternating the temperature back and forth between high and low values is used to turn off and on transport of ions across the PNIPAM brush, respectively (Fig. 3). For example, cyclic voltammograms recorded on a PNIPAM-functionalized indium tin oxide

Fig. 3 Temperature-controlled reversible on/off switching of an electrode surface functionalized with a poly(N-isopropylacrylamide) (PNIPAM) brush. ITO indium tin oxide

Magnetic nanoparticles/microparticles [22, 43–46] and nanowires [23, 47–49] coated with organic shells containing redox species were used to switch on/off electrochemical reactions upon their translocation in the presence of an external magnetic field. Magnetic Fe3O4 microparticles chemically functionalized with redox species (e.g., quinone, ferrocene, or viologen derivatives) were reversibly attracted to an electrode surface and then suspended in a solution [22, 43–46]. A horizontally positioned flat electrode allowed particle translocation between the electrode surface and solution simply by moving a permanent magnet between two positions: below the electrode to attract the magnetic

Fig. 4 Cyclic voltammograms of the PNIPAM-modified ITO electrode obtained in the presence of 1 mM [Fe(CN)6]3− in 0.1 M phosphate buffer (pH 7.0) and 0.1 M sodium perchlorate at different temperatures: 12 °C (a) and 42 °C (b). The inset illustrates reversible changes of the peak current upon stepwise temperature changes: steps 1 and 3 correspond to the off state at 42 °C, whereas steps 2 and 4 correspond to the on state at 12 °C. Potential scan rate, 10 mV/s

E. Katz et al. Fig. 5 Electrochemical and bioelectrocatalytic reactions of redox-modified magnetic particles controlled by an external magnetic field. A Switching on and off the electrochemical reaction of the redox groups (R) associated with the magnetic particles upon attracting them to the electrode surface and lifting them up, respectively. B Switching on and off the bioelectrocatalytic oxidation of glucose in the presence of glucose oxidase (GOx) and ferrocene (Fc)-functionalized magnetic particles upon magnetoinduced translocation of the particles between the surface and solution states. (Adapted from [22], with permission)

particles to its surface and above the electrode to remove particles from the surface and suspend them in the solution (Fig. 5A). While being suspended in the solution, the redoxfunctionalized magnetic particles were unable to interact directly with the electrode surface, thus demonstrating no electrochemical response (Fig. 5A, right). However, when attracted to the electrode surface, they electrically communicated with the conducting support, thus demonstrating electrochemical behavior characteristic of surface-confined redox species (Fig. 5A, left). For example, differential pulse voltammetry demonstrated switchable electrochemical behavior of quinone-modified particles due to magnetically induced vertical translocation between the surface and solution states (Fig. 6). Stepwise repositioning of the magnet below and above the electrode resulted in the reversible activation and inhibition of the electrochemical process, respectively (Fig. 6, inset). The magnetoinduced on/off switching of the electrochemical reactions performed by the redox species associated with the magnetic particles was used to control biocatalytic cascades mediated by the redox species (e.g., using ferrocene-modified particles for communicating with glucose oxidase, GOx, for bioelectrocatalytic oxidation of glucose). While in intimate contact with the electrode surface, the redox-active ferrocene species mediated the bioelectrocatalytic reaction of GOx (Fig. 5B, left); on the other hand, when suspended in the solution, the redox species were unable to mediate the electron transport between the enzyme and the electrode (Fig. 5B, right). In other examples, adaptive multifunctional magnetic nanowires were designed to control operation of electrochemical sensors and microfluidic devices, providing

external control over electrocatalytic and bioelectrocatalytic processes upon application of a magnetic field [47–49]. The opposite effect on interfacial reactions was demonstrated for hydrophobic magnetic nanoparticles [43, 44, 50]. Magnetic nanoparticles functionalized with hydrophobic organic shells were originally dispersed in a nonaqueous liquid phase immiscible with water (e.g., toluene) which was the second liquid phase above the aqueous solution. Upon

Fig. 6 Differential pulse voltammograms obtained for quinonefunctionalized magnetic particles attracted to the electrode surface (a) and lifted up (b) by the external magnetic field. The inset illustrates reversible changes of the peak current upon stepwise translocation of the magnetic particles between the surface and solution states. (Adapted from [22], with permission)

Electrode interfaces switchable by physical and chemical signals Fig. 7 Reversible magnetocontrolled translocation of hydrophobic magnetic nanoparticles between the surface (A) and solution (B) states resulting in inhibition and activation of diffusional electrochemical processes, respectively. (Adapted from [44], with permission; copyright American Chemical Society, 2004).

magnetoinduced attraction to the electrode surface, they generated a hydrophobic thin film on the conducting interface, thereby isolating the electrode surface from the aqueous solution (Fig. 7, part A), inhibiting all electrochemical reactions for soluble redox species and introducing a very large interfacial electron-transfer resistance as measured by impedance spectroscopy (Fig. 8A). Retraction of the hydrophobic nanoparticles from the electrode surface and their placement into the nonaqueous solution on top of the aqueous electrolyte regenerated a bare electrode surface and allowed diffusional redox species to react with the conducting support (Fig. 7, part B) with a small electron-transfer resistance (Fig. 8B) [44]. A nontrivial effect was observed for electrode surfaces functionalized with redox species—the electrochemical response for them was not inhibited upon magnetoinduced deposition of the hydrophobic nanoparticles on top the modified electrode interface [43, 60]. This allowed Fig. 8 Nyquist plot (Zim vs. Zre) for the Faradaic impedance measurements performed upon attraction of the hydrophobic magnetic particles to the electrode surface (A) and their lifting up (B) by repositioning of the external magnet. The measurements were performed in the presence of a 1 mM 1:1 K3[Fe(CN)]6/K4[Fe(CN)6] mixture and upon biasing the working electrode at 0.17 V. (Adapted from [44], with permission; copyright American Chemical Society, 2004)

discrimination of the electrochemical reactions of the diffusional and surface-confined redox species. The redox species soluble in the aqueous electrolyte reacted with the conducting support only when the hydrophobic magnetic nanoparticles were removed from the electrode surface, whereas the surface-confined redox species demonstrated their redox activity in both interface states: in the absence and presence of the hydrophobic nanoparticles on the surface. However, the mechanism of their electrochemical reactions was affected by the interface state—they demonstrated electrochemical behavior typical for an aqueous environment when the hydrophobic nanoparticles were removed from the surface. In contrast, when the hydrophobic film composed of the attracted magnetic nanoparticles was formed on the electrode surface, the surface-confined species became immersed in the nonaqueous microenvironment, demonstrating a redox process typical for the conditions when electron transfer cannot be accompanied by proton transfer.

E. Katz et al.

The separation of diffusional and surface-confined electrochemical reactions as well as the alteration of electrochemical mechanisms reversibly controlled by the external magnetic field in the presence of hydrophobic magnetic nanoparticles allowed switching between different bioelectrocatalytic [43, 50] and photoelectrocatalytic [50, 51] reactions. The magnetoswitchable nanoparticle–electrode interfaces were applied for biosensing switchable between different analytes [52], for changing interfacial processes between single-electron and multielectron transfer [53], and for realization of “write–read–erase” memory devices [54, 55]. Another way of controlling electrode interfacial properties was developed to generate a “forest” of conducting nanowires in the presence of an external magnetic field [56]. These standing nanowires were composed of gold shell/CoFe2O4 magnetic core nanoparticles, self-assembled in the presence of a magnetic field, to generate an array of

nanoelectrodes extending the electrode support and increasing the total electrode area (Fig. 9A). This resulted in the enhancement of the electrochemical response for a diffusional redox probe by about sixfold as followed by cyclic voltammetry (Fig. 9B). The process was reversed when the external magnet was removed, thus switching off the magnetic field and resulting in disaggregation of the conducting nanowires. The primary electrochemical reaction of the diffusional redox species was coupled with the bioelectrocatalytic oxidation of glucose in the presence of GOx, resulting in the amplification of the biocatalytic cascade at the nanostructured array on the electrode surface (Fig. 9A). The magnetoswitchable nanoelectrode array can operate as a general platform for electrochemical biosensors with the enhanced current output signals controlled by the structure of the selfassembled nanowires. Sophisticated adaptive bioelectronic devices controlled by external magnetic signals are feasible upon application of magnetoswitchable electrochemical and bioelectrochemical systems, finding numerous applications in different areas of bioelectronics, biosensor technology, and biomedical systems [23].

Modified electrodes switchable by applied potentials resulting in electrochemical transformations at functional interfaces

Fig. 9 A Magnetic-field-controlled reversible assembly of gold-coated magnetic nanoparticles and their use as a nanostructured electrode with the enhanced ability of electrochemical oxidation of ferrocenemonocarboxylic acid coupled with glucose oxidation biocatalyzed by GOx. B Cyclic voltammograms of ferrocenemonocarboxylic acid (0.1 mM) obtained in the absence (a) and presence (b) of the magnetic field. (Adapted from [56], with permission; copyright American Chemical Society, 2008)

Electrical potentials applied to chemically modified electrode surfaces can significantly change their interfacial properties owing to chemical transformations induced by the variable potential. Modification of electrode surfaces with redox polymers, which can be electrochemically converted to different oxidation states, allows significant variation of electron-transfer, optical, and wettability properties of the functionalized electrodes by application of different potentials [57, 58]. Incorporation of metallic species into polymer matrices [59, 60] or their self-assembly in nanostructured ensembles on electrode interfaces [61–63] can amplify the potential-induced surface changes even more. For example, Cu2+ cations associated with a poly(acrylic acid) (PAA) matrix immobilized on an electrode surface were reversibly reduced to yield metallic clusters and then oxidized back to the cationic state [60] (Fig. 10A). This reversible electrochemical transformation resulted in dramatic changes in the conductivity of the polymer layer, which was very high in the presence of metallic clusters, but was significantly decreased when they were oxidized to the ionic state (Fig. 10B). The modulation of the interfacial conductivity by the applied potential was particularly applied for the design of switchable biofuel cells [64], which can also find bioelectroanalytic applications as self-powered biosensors [65].

Electrode interfaces switchable by physical and chemical signals

Fig. 10 A Reversible transformations of a poly(acrylic acid) (PAA) polymer film loaded with Cu2+ cations upon application of reductive (−0.5 V; SCE) and oxidative (+0.5 V) potentials to the modified electrode. B I–V curves of the polymer film electrochemically switched between states with different conductivity: the Cu2+/PAA electronically nonconducting state after application of a potential of +0.5 V (a); the Cu0/PAA electronically conducting state after application of a potential of −0.5 V (b). The inset illustrates reversible variation of the film resistance derived from the I–V curves: points a measured at +0.5 V and points b measured at −0.5 V. (Adapted from [60], with permission)

Electrode surface modification with polyelectrolytes sensitive to pH allows new applications and is a new approach to switchable interfaces controlled by electrochemically induced reactions [25, 26]. Polyelectrolytes associated with electrode surfaces were switched between permeable and nonpermeable forms upon local interfacial pH changes generated by electrochemical reactions, thus being switchable by potentials applied to the modified electrode surface. For example, a poly(4-vinylpyridine) (P4VP)-brush-modified ITO electrode was used to reversibly switch the interfacial activity upon electrochemical signals [25]. Electrochemical reduction of oxygen resulted in consumption of hydrogen ions at the electrode interface, thus yielding a local pH increase triggering deprotonation and restructuring of the P4VP brush on the electrode surface (Fig. 11A). The initial swollen state of the protonated P4VP brush (pH 4.4) was permeable for anionic [Fe(CN)6]4− redox species (Fig. 11B, curve a), whereas the electrochemically produced local pH of 9.1 resulted in the deprotonation of the polymer brush. The hydrophobic shrunken state of the polymer brush produced was impermeable for the anionic

Fig. 11 A Electrochemically induced pH changes resulting in reversible switching of a poly(4-vinylpyridine) (P4VP) brush between on (left) and off (right) states allowing and restricting penetration of the anionic species to the electrode surface. B Cyclic voltammograms obtained on the modified electrode in the presence of 0.5 mM K4[Fe (CN)6] prior to the application of the potential to the electrode (a) and after application of −0.85 V to the electrode in the presence of O2 (b). The inset illustrates the reversible switching of the peak current upon “closing” the interface by the electrochemical signal resulting in local pH changes and restoration of the electrode activity by stirring the solution. (Adapted from [25], with permission; copyright American Chemical Society, 2010)

redox species, thus fully inhibiting its redox process at the electrode surface (Fig. 11B, curve b). Application of this approach to different interfacial polymer systems will allow the realization of a vast range of switchable electrodes, with the externally controlled activity useful for application in biosensors and biofuel cells.

Electrode interfaces switchable by chemical and biochemical signals Many different approaches were realized to design electrode interfaces switchable by chemical and biochemical signals, usually using receptor units associated with electrode surfaces selectively binding chemical species and thus alternating interfacial properties [66–68]. Recent advances in materials science, particularly related to stimuli-responsive polymers, allowed a new approach to switchable electrodes controlled by chemical and biochemical processes yielding pH changes: pH-sensitive polymer thin films or brushes tethered to electrode surfaces [14, 15] demonstrated pH-switchable interfacial behavior when a neutral state of the polyelectrolyte was impermeable for ionic redox species, thus switching the

E. Katz et al.

electrode activity off, whereas the ionized state of the polymer was permeable for ionic redox species of the opposite charge, allowing their access to the conducting support and switching the electrode reaction on [69, 70]. Mixed polymer brushes allowed more sophisticated switchable behavior of modified electrode surfaces. For example, poly(2-vinylpyridine) (P2VP) and PAA tethered to an ITO electrode as a mixed brush demonstrated three differently charged states controlled by an external pH value: positively charged due to protonation of the P2VP component (pH 3), neutral when the charges of P2VP and PAA are compensated (pH 4.5), and negatively charged when the PAA component is dissociated (pH 6) (Fig. 12A) [69, 70]. The pH-controlled switching between the different charges allowed discrimination between electrochemical reactions of oppositely charged redox species: [Fe (CN)6]4− and [Ru(NH3)6]3+. The negatively charged [Fe (CN)6]4− redox anions accessed the conducting support and electrochemically reacted only when the P2VP component of the mixed-polymer brush was protonated and positively charged. On the other hand, the positively charged [Ru (NH3)6]3+ redox cations reached the electrode support only when the polymeric film was negatively charged owing to dissociation of PAA. The neutral state of the modified interface was not permeable for all ionic species, thus resulting in the off state of the modified electrode for both electrochemical reactions of differently charged redox species. Thus, pH changes in the supporting electrolyte solution resulted in

reversible changes in the electrode activity from the electrochemical reaction of [Fe(CN)6]4− to the redox process of [Ru (NH3)6]3+ and back (Fig. 12B, C) [69, 70]. In the next step of developing chemically switchable electrode interfaces, redox-active species were bound to a pH-responsive polymer film associated with an electrode surface [71]. Os(dmo-bpy)2 redox groups (dmo-bpy is 4,4′-dimethoxy-2,2′-bipyridine) were covalently bound to P4VP-brush chains grafted on an ITO electrode. The redox species linked to the polymer brush were electrochemically active at pH<5, when the polymer is protonated and swollen and the polymer chains are flexible. When the pH of the electrolyte solution was increased above 6, the polymer chains were deprotonated and the shrunken state of the polymer produced did not demonstrate electrochemical activity. The electrode switching off originated from the poor mobility of the polymer chains in the shrunken state inhibiting contact between the polymer-bound redox species and the conducting support. Indeed, a low density of the redox species in the polymer film did not allow electrons to hop between them and their electrochemical activity was observed only upon translocation of the polymer chains in the swollen state bringing the redox species a short distance from the conducting support (Fig. 13A). The reversible switching on/off of the modified electrode was followed by cyclic voltammetry measured at different pH values (Fig. 13B), resulting in modulation of the electrochemical

Fig. 12 A Permeability of a poly(2-vinylpyridine) (P2VP)/PAA mixed-polymer brush for differently charged redox species controlled by pH values: the positively charged protonated P2VP component of the mixed brush allows access of the negatively charged [Fe(CN)6]4− redox anions (left); the neutral hydrophobic polymer thin film inhibits access of all ionic species to the electrode (middle); the negatively charged dissociated PAA component of the mixed brush allows access

of the positively charged [Ru(NH3)6]3+ redox cations (right). B The differential pulse voltammograms obtained for the mixed-polymer brush in the presence of 0.5 mM [Fe(CN)6]4− and 0.1 mM [Ru (NH3)6]3+ at different pH values of the solution: from 3.0 (a) through 4.0, 4.35, 4.65, and 5.0 to 6.0 (f). C The dependence of the peak current on the pH for the anionic, [Fe(CN)6]4−, (a) and cationic, [Ru(NH3)6]3+, (b) species. (Adapted from [70], with permission)

Electrode interfaces switchable by physical and chemical signals

Fig. 13 A Reversible pH-controlled activation–inactivation of an OsP4VP-modified electrode. B Cyclic voltammograms of the modified electrode recorded at different pH values: 7.0 (a), and 3.0 (b). The inset illustrates reversible switching of the modified electrode activity upon stepwise variation of the pH: steps 1, 3, and 5 correspond to pH 7.0; steps 2 and 4 correspond to pH 3.0. (Adapted from [71], with permission; copyright American Chemical Society, 2008)

responses by stepwise changes of the solution pH (Fig. 13B, inset). The switchable electrode was applied for pHcontrolled bioelectrocatalytic oxidation of glucose in the presence of soluble GOx, thus opening the way for designing glucose biosensors with switchable activity [71]. Further sophistication of the system allowed switchable bioelectrocatalytic glucose oxidation upon performing enzymatic reactions changing the pH values in the electrolyte solution [28]. Two soluble hydrolytic enzymes—esterase and urease—were used to change the pH in situ, thus affecting the activity of the modified electrode with regard to mediating bioelectrocatalytic oxidation of glucose in the presence of soluble GOx. Ethyl butyrate added to the solution was hydrolyzed by esterase, resulting in the formation of butyric acid and lowering the pH to about 3.8, thus enabling the bioelectrocatalytic oxidation of glucose mediated by the modified electrode (Fig. 14A) [28]. Then the addition of urea to the solution resulted in the formation of ammonia biocatalyzed by urease, restoring the initial pH of about 6.5 and disabling the glucose oxidation at the modified electrode. The bioelectrocatalytic activity of the modified electrode switchable in situ by the enzymatic reactions was analyzed by cyclic voltammetry (Fig. 14B). The

Fig. 14 A Switchable bioelectrocatalytic oxidation of glucose controlled by external enzymatic reactions. B Cyclic voltammograms obtained for the switchable bioelectrocatalytic glucose oxidation when the system is in the initial off state, pH about 6.5 (a), enabled by the ethyl butyrate input signal, pH about 3.8 (b), and inhibited by the urea reset signal, pH about 7.5 (c). The inset illustrates switching of the bioelectrocatalytic current on and off by the enzyme-processed biochemical signals. (Adapted from [28], with permission; copyright American Chemical Society, 2008)

reversible stepwise transition between the active and inactive states demonstrated the possibility to modulate the bioelectrocatalytic activity by biochemical signals (Fig. 14B, inset). The system was further improved by immobilization of the enzymes (esterase and urease) on the electrode surface, thus integrating the pH-switchable polymer with the signal-processing enzymes producing local pH changes triggering the switching on/off of the electrode [72].

E. Katz et al.

Switchable electrodes controlled by biomolecular computing systems Usually switchable electrodes are controlled by one kind of external signal using a variable physical parameter (irradiation at a specific wavelength, changing the temperature, magnetic field, or electrical potential applied to the electrode) or (bio)chemical signals defined as the absence or presence of one specific species, as outlined already. In very few examples, switchable electrodes were controlled by combinations of different kinds of physical and chemical signals, e.g., variable magnetic field and absence/presence of a chemical [73], irradiation of an electrode, and application of a potential to it [74]. Still these examples are rare and the systems are hardly integrated in complex chemical devices. The input signals used are completely different in nature from the output signals; thus, the systems cannot be concatenated in networks of higher-complexity processing signals in several steps, applying various logic rules at each step of their operation. Unexpectedly, the solution came from a very different area of unconventional chemical computing [75–77] allowing further sophistication of switchable electrochemical interfaces. Molecular, supramolecular [29, 78–82], and biomolecular [30, 31] systems mimicking Boolean logic gates and their networks allowed the realization of systems processing many chemical input signals via preprogrammed rules implemented in the chemical composition and producing the output signal in the form of other chemical species which could be applied again as an input for further information processing steps. Fig. 15 The biochemical logic gate with the enzymes used as input signals to activate the gate operation followed by the reset function. A The AND gate based on GOx- and invertasecatalyzed reactions. B The pH changes generated in situ by the AND gate for different combinations of the input signals: 0,0 (a), 0,1 (b), 1,0 (c), and 1,1 (d). The inset contains a bar diagram showing the pH changes as the output signals of the AND gate. C The truth table of the AND gate showing the output signals in the form of pH changes generated for different combinations of the input signals. D Equivalent electronic circuit for the biochemical AND–reset logic operations. (Adapted from [97], with permission; copyright American Chemical Society, 2009)

Particularly, the design of sophisticated information processing systems became possible when complex biomolecular reactions were applied for processing multiple biochemical inputs [30, 31]. The result of the biochemical chain reactions mimicking computer operations was connected to stimuliresponsive materials, thus resulting in high-complexity switchable electrode interfaces [17, 83]. Different biochemical systems mimicking Boolean logic operations (AND, OR, XOR, INHIB, NOR, etc., gates) [84–88] and their sophisticated networks performing complex logic algorithms [89–91] were designed using enzyme systems. These systems have been functionally [92, 93] and sometimes structurally [72] integrated with stimuliresponsive materials associated with electrode surfaces, thus allowing logic control over electrochemical reactions using complex patterns of various chemical signals. Importantly, the functional integration of biomolecular information processing systems with the pH-switchable polymer membranes and brushes described earlier allowed a high level of sophistication of the signal-switchable electrodes. Most chemical [29, 78–82] and biochemical [30, 31] systems designed for mimicking Boolean logic operations generate output signals in the form of small concentration changes which are followed by optical or electrochemical means. To use them for switching the states of stimuli-responsive materials (particularly those associated with electrode interfaces), the output signal should be in a form acceptable by the materials, thus resulting in changes in them. For example, small concentration changes of NADH/NAD+ cofactor optically [94, 95] or electrochemically [96] readable as the logic output from

Electrode interfaces switchable by physical and chemical signals

Fig. 16 The signal-responsive membrane associated with an ITO electrode and coupled with the enzyme-based logic gates. A Atomic force microscopy topography images (10×10 μm2) of the membrane with the closed (a) and open (b) pores. B The electron-transfer resistance, R et , of the switchable interface derived from impedance

spectroscopy measurements obtained with different combinations of the input signals. Left‐blue bars correspond to the AND logic gate and right‐brown bars correspond to the OR logic gate. (Adapted from [97], with permission; copyright American Chemical Society, 2009)

several recently developed enzyme logic systems cannot be used directly for switching the states of presently known stimuli-responsive materials. On the other hand, pH changes generated by many enzyme logic systems [92] could easily be used to control the states of many polymer-based signalresponsive systems which are sensitive to pH [14, 15]. Therefore, enzyme systems performing AND/OR Boolean logic operations and producing pH changes upon biocatalytic reactions have been designed and coupled with nanostructured signal-responsive materials associated with electrode interfaces [97]. To exemplify the functional coupling of enzyme logic systems and pH-responsive materials, an AND logic

gate was composed of an aqueous solution containing sucrose, O2, and urea, and GOx and invertase were applied as input signals (Fig. 15A) [97]. The presence of the enzymes at an optimized concentration was defined as logic input 1, whereas their absence was considered as logic input 0. When the logic input combination 1,1 was applied (meaning the presence of both enzymes), the reaction chain was activated, including conversion of sucrose to glucose catalyzed by invertase, followed by the oxidation of glucose catalyzed by GOx, thus yielding gluconic acid and an acidic pH. The absence of either or both enzymes (input signals 0,1; 1,0; 0,0) did not allow the completion of the chain reaction and resulted in no pH

Fig. 17 A The biocatalytic cascade used for the logic processing of the chemical input signals and producing in situ pH changes as the output signal. B The equivalent logic circuitry for the biocatalytic cascade. C Cyclic voltammograms obtained for an ITO electrode modified with a P4VP brush in the initial off state, pH about 6.7 (a), the on state enabled by the logic system at pH about 4.3 (b), and in situ reset to the off state (c). The inset illustrates reversible switching of the electrode activity. D Anodic peak currents, Ip, for the 16 possible input combinations. The dotted lines show threshold values separating logic 1, undefined, and logic 0 output signals. (Adapted from [32], with permission; copyright American Chemical Society, 2009)

E. Katz et al.

changes in the system (Fig. 15B). Thus, the system demonstrated AND logic behavior (Fig. 15B, inset, with the characteristic truth table in Fig. 15C). Another biochemical reaction (hydrolysis of urea biocatalyzed by urease) was used to reset the system to the initial pH state (Fig. 15D). A similar approach was used to realize an OR logic gate operating as two parallel reactions biocatalyzed by different enzyme inputs, but resulting in pH changes as the output of both reactions [97]. The AND/OR enzyme-based logic gates were applied to control a pH-sensitive membrane associated with an electrode surface, thus switching the electrode interface between the on state (when the pores are open; Fig. 16A, image b) and the off state (when the pores are closed; Fig. 16A, image a) [97]. The electrode surface activation and inhibition was controlled by the enzyme logic gates and followed by impedance spectroscopy (Fig. 16B). A similar approach was applied to control logically an electrode interface modified with a pH-sensitive polymer brush [98] as well as nanostructured assemblies [99, 100]. Further sophistication of the biomolecular information processing systems was achieved by assembling several logic gates in logic circuitries [89, 90] and their functional integration with switchable electrode surfaces [32]. For example, the logic network composed of three enzymes (alcohol dehydrogenase, glucose dehydrogenase, and GOx) (Fig. 17A) operating as a circuitry of four concatenated logic gates (Fig. 17B) processed four chemical input signals (NADH, acetaldehyde, glucose and oxygen). The biomolecular chain reaction resulted in pH changes controlled by all four chemical input signals applied. The in situ produced pH changes were applied to a pH-sensitive polymer-brushmodified electrode, resulting in it changing from the off state, when the electrochemical reactions were inhibited, to the on state, when the interface was electrochemically active. The switching properties of the electrode were analyzed by cyclic voltammetry (Fig. 17C), being controlled by four input signals in 16 variants (Fig. 17D) according to the Boolean logic encoded in the logic circuitry. The initial steps in this research were aimed only at demonstrating the coupling between biomolecular computing systems and modified electrodes with switchable interfaces. However, practical use of similar systems is highly feasible in the area of “smart” multi-signal-processing biosensors and actuators, particularly for biomedical applications [101, 102], e.g., for the analysis of pathophysiological conditions corresponding to different injuries [103–105]. Whereas for the demonstration of the biocomputing concept convenient arbitrary concentrations of chemical inputs were used (usually logic 0 was represented by the absence of the corresponding species), in biomedical applications the logic levels of the biochemical input signals are defined on the basis of their biomedical meaning: logic 0 and logic 1 input

Fig. 18 A The biocatalytic cascade used for the logic processing of the biomarkers characteristic of liver injury, resulting in situ pH changes and activation of the electrode interface. B The pH changes generated in situ by the biocatalytic cascade activated with various combinations of the input signals of the two biomarkers, alanine transaminase (ALT) and lactate dehydrogenase (LDH): 0,0 (a), 0,1 (b), 1,0 (c), and 1,1 (d). The dotted line corresponds to the pKa of the P4VP brush. C Cyclic voltammograms obtained for an ITO electrode modified with a P4VP polymer brush in the initial off state, pH 6.3 (a) and the on state enabled by the ALT—LDH input combination 1,1, pH 4.75 (b). (Adapted from [106], with permission; copyright American Chemical Society, 2011)

levels correspond to normal physiological and elevated pathological concentrations of biomarkers, respectively, which may appear with a small difference. For example, biomarkers characteristic of liver injury—alanine transaminase and lactate dehydrogenase—were processed by a biocatalytic system functioning as a logic AND gate (Fig. 18A) [106]. The NAD+ output signal produced by the system upon its activation in the presence of both biomarkers was then biocatalytically converted to the pH decrease. The acidic pH produced by the system as a response to the biomarkers (Fig. 18B) resulted in the activation of the

Electrode interfaces switchable by physical and chemical signals

electrode interface modified with the pH-sensitive polymer brush (Fig. 18C).

15. 16. 17. 18.

Summary

19.

Numerous switchable electrode interfaces controlled by various physical and chemical signals, as well as by their combinations, have been designed in the last two decades. A novel facet in this research was established when biomolecular information processing systems were integrated with the switchable electrodes, thus allowing control of electrochemical reactions by biochemical or even physiological signals. A further increase in the system complexity was achieved by using enzyme systems [107, 108] together with immune-biorecognition systems [109] or with microbial cells [110, 111] for processing various input signals and their combinations and finally for activation or inhibition of modified electrode surfaces and biofuel cells [112, 113]. Implantable biofuel cells [114–116] producing electrical power on demand depending on physiological conditions are feasible as the result of this research. The present developments in the area of switchable modified electrodes are based on the application of a multidisciplinary approach, and their future developments and those of bioelectronics in general will require contributions from electrochemists and specialists in materials science and unconventional biomolecular computing. Acknowledgments This research was supported by the National Science Foundation (grants DMR-0706209, CCF-0726698, CCF1015983) and by the Office of Naval Research (grant ONR N0001408-1-1202).

20.

21. 22. 23. 24. 25. 26. 27. 28. 29.

30.

31. 32. 33. 34. 35. 36. 37.

38.

References

39. 40.

1. Murray RW (1984) In: Bard AJ (ed) Electroanalytical chemistry, vol 13. Dekker, New York, pp 191–368 2. Murray RW (1980) Acc Chem Res 13:135–141 3. Wrighton MS (1986) Science 231:32–37 4. Abruña HD (1988) Coord Chem Rev 86:135–189 5. Bain CD, Troughton EB, Tao Y-T, Eval J, Whitesides GM, Nuzzo RG (1989) J Am Chem Soc 111:321–335 6. Nuzzo RG, Fusco FA, Allara DL (1987) J Am Chem Soc 109:2358–2368 7. Rusling JF, Forster RJ (2003) J Colloid Interface Sci 262:1–15 8. Willner I, Katz E (2000) Angew Chem Int Ed 39:1180–1218 9. Wang J (2008) Talanta 75:636–641 10. Gooding JJ (2008) Electroanalysis 20:573–582 11. Moehlenbrock MJ, Minteer SD (2008) Chem Soc Rev 37:1188–1196 12. Barton SC, Gallaway J, Atanassov P (2004) Chem Rev 104:4867– 4886 13. Stuart MAC, Huck WTS, Genzer J, Muller M, Ober C, Stamm M, Sukhorukov GB, Szleifer I, Tsukruk VV, Urban M, Winnik F, Zauscher S, Luzinov I, Minko S (2010) Nature Mater 9:101–113 14. Minko S (2006) Polym Rev 46:397–420

41. 42. 43. 44. 45. 46. 47. 48.

49.

50.

Luzinov I, Minko S, Tsukruk VV (2004) Prog Polym Sci 29:635–698 Pita M, Katz E (2009) Electroanalysis 21:252–260 Bocharova V, Katz E (2012) Chem Rec 12:114–130 Laocharoensuk R, Bulbarello A, Hocevar SB, Mannino S, Ogorevc B, Wang J (2007) J Am Chem Soc 129:7774–7775 Wang J, Musameh M, Laocharoensuk R, Gonzalez-Garcia O, Oni J, Gervasio D (2006) Electrochem Commun 8:1106–1110 Shipway AN, Katz E, Willner I (2001) In: Sauvage J-P (ed) Molecular machines and motors. Structure and bonding, vol 99. Springer, Berlin, pp. 237–281 Katz E, Shipway AN, Willner I (2002) In: Sekkat S, Knoll W (eds) Photoreactive organic thin films. Elsevier, San Diego, pp 219–268 Willner I, Katz E (2003) Angew Chem Int Ed 42:4576–4588 Wang J (2008) Electroanalysis 20:611–615 Song S, Hu N (2010) J Phys Chem B 114:5940–5945 Tam TK, Pita M, Trotsenko O, Motornov M, Tokarev I, Halámek J, Minko S, Katz E (2010) Langmuir 26:4506–4513 Tam TK, Pita M, Motornov M, Tokarev I, Minko S, Katz E (2010) Adv Mater 22:1863–1866 Hou KY, Yu L, Severson MW, Zeng XQ (2005) J Phys Chem B 109:9527–9531 Tam TK, Zhou J, Pita M, Ornatska M, Minko S, Katz E (2008) J Am Chem Soc 130:10888–10889 Katz E (ed) (2012) Molecular and supramolecular information processing: from molecular switches to logic systems. WileyVCH, Weinheim Katz E (ed) (2012) Biomolecular information processing—from logic systems to smart sensors and actuators. Wiley-VCH, Weinheim Katz E, Privman V (2010) Chem Soc Rev 39:1835–1857 Privman M, Tam TK, Pita M, Katz E (2009) J Am Chem Soc 131:1314–1321 Lion-Dagan M, Katz E, Willner I (1994) J Am Chem Soc 116:7913–7914 Willner I, Lion-Dagan M, Katz E (1996) Chem Commun 623–624 Liu NG, Dunphy DR, Atanassov P, Bunge SD, Chen Z, Lopez GP, Boyle TJ, Brinker CJ (2004) Nano Lett 4:551–554 Liu ZF, Hashimoto K, Fujishima A (1990) Nature 347:658–660 Wesenhagen P, Areephong J, Landaluce TF, Heureux N, Katsonis N, Hjelm J, Rudolf P, Browne WR, Feringa BL (2008) Langmuir 24:6334–6342 Browne WR, Kudernac T, Katsonis N, Areephong J, Hielm J, Feringa BL (2008) J Phys Chem C 112:1183–1190 Doron A, Portnoy M, Lion-Dagan M, Katz E, Willner I (1996) J Am Chem Soc 118:8937–8944 Willner I, Pardo-Yissar V, Katz E, Ranjit KT (2001) J Electroanal Chem 497:172–177 Willner I, Lion-Dagan M, Marx-Tibbon S, Katz E (1995) J Am Chem Soc 117:6581–6592 Yin Z, Zhang J, Jiang L-P, Zhu J-J (2009) J Phys Chem C 113:16104–16109 Katz E, Baron R, Willner I (2005) J Am Chem Soc 127:4060–4070 Katz E, Sheeney-Haj-Ichia L, Basnar B, Felner I, Willner I (2004) Langmuir 20:9714–9719 Katz E, Sheeney-Haj-Ichia L, Willner I (2002) Chem Eur J 8:4138–4148 Wang J, Kawde AN (2002) Electrochem Commun 4:349–352 Laocharoensuk R, Bulbarello A, Mannino S, Wang J (2007) Chem Commun 3362–3364 Loaiza OA, Laocharoensuk R, Burdick J, Rodriguez MC, Pingarron JM, Pedrero M, Wang J (2007) Angew Chem Int Ed 46:1508–1511 Wang J, Scampicchio M, Laocharoensuk R, Valentini F, Gonzalez-Garcia O, Burdick J (2006) J Am Chem Soc 128:4562–4563 Willner I, Katz E (2006) Langmuir 22:1409–1419

E. Katz et al. 51. Katz E, Willner I (2005) Angew Chem Int Ed 44:4791–4794 52. Katz E, Willner I (2005) Chem Commun 4089–4091 53. Katz E, Lioubashevski O, Willner I (2006) Chem Commun 1109– 1111 54. Katz E, Willner I (2005) Chem Commun 5641–5643 55. Katz E, Willner I (2006) Electrochem Commun 8:879–882 56. Jimenez J, Sheparovych R, Pita M, Narvaez Garcia A, Dominguez E, Minko S, Katz E (2008) J Phys Chem C 112:7337–7344 57. Raitman OA, Katz E, Willner I, Chegel VI, Popova GV (2001) Angew Chem Int Ed 40:3649–3652 58. Chegel V, Raitman O, Katz E, Gabai R, Willner I (2001) Chem Commun 883–884 59. Zheng L, Xiong L (2006) Colloids Surf A 289:179–184 60. Chegel VI, Raitman OA, Lioubashevski O, Shirshov Y, Katz E, Willner I (2002) Adv Mater 14:1549–1553 61. Riskin M, Basnar B, Chegel VI, Katz E, Willner I, Shi F, Zhang X (2006) J Am Chem Soc 128:1253–1260 62. Riskin M, Basnar B, Katz E, Willner I (2006) Chem Eur J 12:8549–8557 63. Riskin M, Katz E, Willner I (2006) Langmuir 22:10483–10489 64. Katz E, Willner I (2003) J Am Chem Soc 125:6803–6813 65. Katz E, Bückmann AF, Willner I (2001) J Am Chem Soc 123:10752–10753 66. Diamond D, McKervey MA (1996) Chem Soc Rev 25:15–24 67. Yang DH, Ju M-J, Maeda A, Hayashi K, Toko K, Lee S-W, Kunitake T (2006) Biosens Bioelectron 22:388–392 68. Gabai R, Sallacan N, Chegel V, Bourenko T, Katz E, Willner I (2001) J Phys Chem B 105:8196–8202 69. Motornov M, Tam TK, Pita M, Tokarev I, Katz E, Minko S (2009) Nanotechnology 434006 70. Tam TK, Pita M, Motornov M, Tokarev I, Minko S, Katz E (2010) Electroanalysis 22:35–40 71. Tam TK, Ornatska M, Pita M, Minko S, Katz E (2008) J Phys Chem C 112:8438–8445 72. Bocharova V, Tam TK, Halámek J, Pita M, Katz E (2010) Chem Commun 46:2088–2090 73. Pita M, Tam TK, Minko S, Katz E (2009) ACS Appl Mater Interfaces 1:1166–1168 74. Baron R, Onopriyenko A, Katz E, Lioubashevski O, Willner I, Wang S, Tian H (2006) Chem Commun 2147–2149 75. Calude CS, Costa JF, Dershowitz N, Freire E, Rozenberg G (eds) (2009) Unconventional computation. Lecture notes in computer science, vol. 5715 Springer, Berlin 76. Adamatzky A, De Lacy Costello B, Bull L, Stepney S, Teuscher C (eds) (2007) Unconventional computing. Luniver, Beckington, UK 77. Gheorghe M (2005) Molecular computation models: unconventional approaches. Idea Group Publishing, Hershay, PA 78. de Silva AP, Uchiyama S (2007) Nat Nanotechnol 2:399–410 79. Szacilowski K (2008) Chem Rev 108:3481–3548 80. Credi A (2007) Angew Chem Int Ed 46:5472–5475 81. Pischel U (2007) Angew Chem Int Ed 46:4026–4040 82. Andreasson J, Pischel U (2010) Chem Soc Rev 39:174–188 83. Katz E, Bocharova V, Privman V (2012) J Mater Chem 22:8171–8178 84. Baron R, Lioubashevski O, Katz E, Niazov T, Willner I (2006) J Phys Chem A 110:8548–8553 85. Strack G, Pita M, Ornatska M, Katz E (2008) Chembiochem 9:1260–1266 86. Privman V, Zhou J, Halámek J, Katz E (2010) J Phys Chem B 114:13601–13608

87. Melnikov D, Strack G, Zhou J, Windmiller JR, Halámek J, Bocharova V, Chuang M-C, Santhosh P, Privman V, Wang J, Katz E (2010) J Phys Chem B 114:12166–12174 88. Zhou J, Arugula MA, Halámek J, Pita M, Katz E (2009) J Phys Chem B 113:16065–16070 89. Niazov T, Baron R, Katz E, Lioubashevski O, Willner I (2006) Proc Natl Acad Sci USA 103:17160–17163 90. Privman V, Arugula MA, Halámek J, Pita M, Katz E (2009) J Phys Chem B 113:5301–5310 91. Tam TK, Pita M, Katz E (2009) Sens Actuators B 140:1–4 92. Pita M, Minko S, Katz E (2009) J Mater Sci Mater Med 20:457– 462 93. Minko S, Katz E, Motornov M, Tokarev I, Pita M (2011) J Comput Theor Nanosci 8:356–364 94. Halámek J, Zhou J, Halámková L, Bocharova V, Privman V, Wang J, Katz E (2011) Anal Chem 83:8383–8386 95. Bocharova V, Halámek J, Zhou J, Strack G, Wang J, Katz E (2011) Talanta 85:800–803 96. Zhou N, Windmiller JR, Valdés Ramírez G, Zhou M, Halámek J, Katz E, Wang J (2011) Anal Chim Acta 703:94–100 97. Tokarev I, Gopishetty V, Zhou J, Pita M, Motornov M, Katz E, Minko S (2009) ACS Appl Mater Interfaces 1:532–536 98. Wang X, Zhou J, Tam TK, Katz E, Pita M (2009) Bioelectrochemistry 77:69–73 99. Motornov M, Zhou J, Pita M, Tokarev I, Gopishetty V, Katz E, Minko S (2009) Small 5:817–820 100. Motornov M, Zhou J, Pita M, Gopishetty V, Tokarev I, Katz E, Minko S (2008) Nano Lett 8:2993–2997 101. Wang J, Katz E (2011) Isr J Chem 51:141–150 102. Wang J, Katz (2010) Anal Bioanal Chem 398:1591–1603 103. Zhou J, Halámek J, Bocharova V, Wang J, Katz E (2011) Talanta 83:955–959 104. Halámek J, Bocharova V, Chinnapareddy S, Windmiller JR, Strack G, Chuang M-C, Zhou J, Santhosh P, Ramirez GV, Arugula MA, Wang J, Katz E (2010) Mol Biosyst 6:2554–2560 105. Halámek J, Windmiller JR, Zhou J, Chuang M-C, Santhosh P, Strack G, Arugula MA, Chinnapareddy S, Bocharova V, Wang J, Katz E (2010) Analyst 135:2249–2259 106. Privman M, Tam TK, Bocharova V, Halámek J, Wang J, Katz E (2011) ACS Appl Mater Interfaces 3:1620–1623 107. Tam TK, Pita M, Ornatska M, Katz E (2009) Bioelectrochemistry 76:4–9 108. Amir L, Tam TK, Pita M, Meijler MM, Alfonta L, Katz E (2009) J Am Chem Soc 131:826–832 109. Tam TK, Strack G, Pita M, Katz E (2009) J Am Chem Soc 131:11670–11671 110. Li Z, Rosenbaum MA, Venkataraman A, Tam TK, Katz E, Angenent LT (2011) Chem Commun 47:3060–3062 111. Arugula MA, Saffarini D, Katz E, He Z (2012) Chem Commun 48:10174–10176 112. Zhou M, Zhou N, Kuralay F, Windmiller JR, Parkhomovsky S, Valdés-Ramírez G, Katz E, Wang J (2012) Angew Chem Int Ed 51:2686–2689 113. Katz E, Pita M (2009) Chem Eur J 15:12554–12564 114. Szczupak A, Halámek J, Halámková L, Bocharova V, Alfonta L, Katz E (2012) Energy Environ Sci 5:8891–8895 115. Halámková L, Halámek J, Bocharova V, Szczupak A, Alfonta L, Katz E (2012) J Am Chem Soc 134:5040–5043 116. Schröder U (2012) Angew Chem Int Ed 51:7370–7372