Adsorption and Electrochemistry of Carbon Monoxide at the Ionic

May 1, 2019 - Department of Chemistry, Oakland University, Rochester, Michigan 48309, United States. Electrochemical gas sensor methods: Figure S1...
1 downloads 0 Views 1MB Size
Download PDF
Article Cite This: J. Phys. Chem. B XXXX, XXX, XXX−XXX

pubs.acs.org/JPCB

Adsorption and Electrochemistry of Carbon Monoxide at the Ionic Liquid−Pt Interface Yongan Tang, Xiaojun Liu, Jordan McMahan, Anil Kumar, Asim Khan, Michael Sevilla, and Xiangqun Zeng* Department of Chemistry, Oakland University, Rochester, Michigan 48309, United States

Downloaded by BETHEL UNIV at 22:07:37:431 on May 24, 2019 from https://pubs.acs.org/doi/10.1021/acs.jpcb.8b11602.

S Supporting Information *

ABSTRACT: In this work, CO adsorption and oxidation processes were studied with cyclic voltammetry and anodic adsorptive stripping chronoamperometry in two structural different ionic liquids (ILs) (i.e., 1-butyl-1methylpyrrolidinium bis(trifluoromethylsulfonyl)imide [Bmpy][NTf2] and 1butyl-3-methylimidazolium bis(trifluoromethylsulfonyl)imide [Bmim][NTf2]). Multiple redox processes were observed in the ILs. During the anodic oxidation processes, the NTf2− anion is oxidized to form NTf•2 radical and CO is oxidized to CO2 and produces a proton in these ILs when a trace amount of water is present. The products of oxidation processes (NTf•2 radical and proton) can be reduced at cathodic processes. Results show that the cation in these ILs can facilitate the formation of an electrolyte−electrode interface structure that influences the amount of CO adsorbed as well as the subsequent CO oxidation current and charge. By selecting the anodic and cathodic potentials, we developed an innovative electroanalytical method for CO sensing based on a simple double-potential adsorptive stripping chronoamperometry. The method allows calibration of the concurrent NTf2− anion and CO redox processes as well as the double-layer charging and discharging processes in the IL with the presence of a trace amount of water providing quantitative analysis of CO concentration with high accuracy and sensitivity. The reported method is the first work to show that quantitative CO detection can be achieved in the presence of complex dynamic interfacial processes in the ILs. The trace water present in the ILs is beneficial for CO oxidation, but a large amount of water is detrimental for the CO oxidation in ambient condition.

■

INTRODUCTION In recent years, ionic liquids (ILs) have attracted intense interest as electrolytes, solvents, and reaction media in both fundamental and applied research in electrochemistry, analytical chemistry, catalysis, materials, pharmaceuticals, and biotechnology.1−3 ILs, composed of organic cations or anions, have negligible vapor pressure at ambient pressure and high thermal stability that reduces hazards associated with flash points and flammability. They are typically more chemically and electrochemically stable than traditional aqueous and organic solvents. By exploiting their synthetic flexibility, ILs can be tailored to be chemically independent via designing one ion to deliver one function and the second ion to deliver a different, completely independent function. In the past decade, we and others have carried out systematic investigations to explore ILs as a new type of solvent, electrolyte, and sensing material for miniaturized electrochemical sensor implementation.4,5 One of the major issues is that water is an environmental variable in ambient conditions for IL-based gas sensors because complete removal of water from the ILs is difficult. On the one hand, the presence of the trace water in the ILs could affect the sensing reactions and properties of the electrode−electrolyte interface; on the other hand, trace water can enable a redox reaction that is desirable for electrochemical sensor development. For example, our early work showed that © XXXX American Chemical Society

trace water in a hydrophobic aprotic IL 1-butyl-1-methylpyrrolidinium bis(trifluoromethylsulfonyl)imide ([Bmpy][NTf2]) acts as a proton source that facilitates the electrooxidation of acetaldehyde.6 By controlling the electrode potentials, several methods including cyclic voltammetry (CV) and potential step methods (single-potential, doublepotential, and triple-potential step methods) were established for the quantification of acetaldehyde in the “IL−trace water electrolyte”.6 Recent work shows that water plays a decisive role in surface charging and IL layering at hydrophilic and ionizable substrates such as mica, while the effect is rather small and potential dependent at hydrophobic and strongly ionophilic electrified gold surfaces.7 Thus, the understanding of the adsorption and reactions occurring at the IL−electrode interface in ambient conditions where trace water is present in the IL is critical for the use of ILs as electrolytes for highly sensitive and selective electrochemical sensor development. In this work, we studied CO adsorption and oxidation processes at the IL−electrode interface via CV and anodic adsorptive stripping chronoamperometry (ASC). Our goal is to exploit the benefits of ILs and the unique CO adsorption Received: December 2, 2018 Revised: May 1, 2019 Published: May 1, 2019 A

DOI: 10.1021/acs.jpcb.8b11602 J. Phys. Chem. B XXXX, XXX, XXX−XXX

Article

The Journal of Physical Chemistry B and oxidation processes at the IL−Pt electrode interface to develop new analytical methods for sensing CO in ambient conditions. CO is a major atmospheric pollutant, and CO poisoning can lead to death. Consequently, the monitoring and detecting of CO in our environment is of great importance. Many analytical CO sensors based on semiconductors have been developed.8−10 The main drawback of these sensors is that metal oxide sensors tend to suffer from baseline drifts upon interaction with poisoning species. They frequently require high temperature and are not selective. Because CO can be easily oxidized, electrochemical CO sensors were developed to quantitatively detect CO with high sensitivity using low cost and low power electrochemical transducers. Commercial electrochemical CO sensors have been on the market for over 50 years.11 The majority of electrochemical CO sensors are based on CO oxidation on the Pt electrode and use H2SO4 or HNO3 as electrolytes. This reaction has been used as a model reaction for the fundamental study of reaction mechanisms and the surface properties for fuel cells as well as for CO sensors.12−14 The strong adsorption of CO on the Pt electrode poisons the Pt catalyst and is a major challenge to be addressed in the hydrogen fuel cell research. The oxidation of adsorbed CO on the Pt electrode follows a typical heterogeneous catalytic reaction that produces CO2 and a proton. The mechanism of CO oxidation in aqueous electrolytes has been widely investigated. However, very few studies of CO oxidation in ILs have been carried out.15−17 ILs eliminate the problems of dry out and corrosive acid electrolytes in aqueous-based electrochemical CO sensors and will increase the stability and extend the device life time for both fuel cell and CO sensor applications. Thus, it is important to understand the role of ILs on CO adsorption and oxidation for the development of an IL-based electrochemical CO sensor. In contrast to the electrical double layer (EDL) typically observed in the aqueous electrolytes, the ions of ILs are strongly oriented at the electrode surface into an ordered layer structure18 and multiple ion pair layers could form at the IL− electrode interface.19−21 It was shown that the potentialdependent interface of an IL at the metal electrode is very sensitive to the electrode surface conditions, such as proton adsorption on an oxide electric interface22 and the adsorption of CO on a metal electrode.18,23,24 Recently, in situ electrochemical attenuated total reflection-surface enhanced infrared reflection absorption spectroscopy revealed that the interfacial structures of CO-covered Pt electrode in the ILs are different from those in conventional electrolytes. Electrooxidation of adsorbed CO on the Pt electrode in ILs is shown to be sensitive to the crystalline orientation of the Pt surface.15 Additionally, because the complete removal of water from the ILs is difficult, the trace water in the IL is suggested to provide the adsorbed oxide layer, Pt−OH, on the Pt surface that facilitates the CO oxidation in the IL.15,25 However, whether CO oxidation processes in the IL with the presence of trace water could be utilized for the detection and quantitative analysis of CO concentration was not studied. As shown in Scheme 1, we hypothesize that (i) trace water could be utilized as a reactant to provide surface-adsorbed OH on the Pt to facilitate CO oxidation in the ILs; (ii) a trace amount of water is beneficial for CO oxidation while a larger volume of water will alter the IL−electrode and IL properties which hinder the CO oxidation; and (iii) specific IL cations improve the interfacial properties of the IL−electrode for CO

Scheme 1. CO Adsorption and Oxidation at Pt Electrode in an IL with the Presence of Water

oxidation, providing a more sensitive CO sensor. The reactant CO and product CO2 have widely different adsorptivity on the Pt electrode as well as solubility in the ILs.6,26−28 This property can be utilized for developing new electroanalytical methods for gas sensing utilizing the change of both the IL double-layer properties and redox properties of the analyte. To validate our hypothesis, we selected two different NTf2-based hydrophobic aprotic ILs with different cation structures [i.e., [Bmpy][NTf2] and 1-butyl-3-methylimidazolium bis(trifluoromethylsulfonyl)imide ([Bmim][NTf2])]. Imidazolium-based cations adsorbed on the Pt electrode form different IL−electrode interface double-layer structures in comparison with Bmpy+ cations.29,30 Different water concentrations in these ILs were tested to investigate the effect of water concentration on CO oxidation and detection. We compared the adsorption behavior of CO at the IL−Pt electrode interface and subsequent electro-oxidation of adsorbed CO to CO2 in these ILs using voltammetry and chronoamperometry. Our results show that the current and charge of the CO oxidation and the double-layer charging and discharging processes differ in these two ILs. The amount of CO adsorbed on the Pt electrode differs in these two structurally different ILs, likely because of the different electrified IL−Pt interface structure. A trace amount of water present in the hydrophobic ILs is shown to facilitate the CO oxidation process, but a larger volume of water led to changes in the IL properties and complicated redox mechanisms in the ILs. The CO molecule has a very small dipole of 0.12 debye versus H2O which is 1.85 debye. Both cations should have little interaction with CO. Even though the adsorbed Pt−OH is involved in CO oxidation, we did density functional theory (DFT) calculations to study the binding activity of the cation of the IL with OH− as the first step to understand possible IL− electrode interface interactions. The mechanism of the electrochemical behaviors of CO adsorption and oxidation in these two ILs is proposed. Furthermore, a new electroanalytical method capitalized on the unique adsorption properties of CO on the IL and Pt electrode interface for CO detection via double-potential anodic stripping chronoamperometry was demonstrated. Both types of ILs can be utilized for CO detection, and the unique CO adsorption and oxidation mechanisms found at each IL−Pt interface provide tenability for the analytical sensing of CO.

■

EXPERIMENTAL SECTION Chemicals and Reagents. [Bmpy][NTf2] (99%) and [Bmim][NTf2] (99%) were purchased from IoLiTec Inc.31 The viscosity of [Bmpy][NTf2] and [Bmim][NTf2] is 79 and 54 cP at 0.101 MPa (1 atm), respectively.32 Different batches B

DOI: 10.1021/acs.jpcb.8b11602 J. Phys. Chem. B XXXX, XXX, XXX−XXX

Article

The Journal of Physical Chemistry B of the ILs could contain different amounts of trace water. High-purity gases (nitrogen, 99.99%) were obtained from Airgas Great Lakes (Independence, OH). The CO gas is from Praxair Inc. (100%, CAS 630-08-0, part number 2.5-G). Electrochemical Gas Sensor and Characterization Methods. A back flow electrochemical gas cell shown in Supporting Information Figure S1 was used in this study.33−35 A platinum gauze (Sigma-Aldrich) was used as the working electrode, placed directly on top of a gas-permeable Teflon membrane (Zitex, Chemplast, Incorporated, Wayne, NJ). The geometric area of the Pt gauze electrode is 0.64 cm2. The analyte, CO, was introduced via a gas inlet, and the concentration was controlled with a mass flow controller (MKS Instruments, Inc.). Counter and reference electrodes (0.5 mm Pt wires) were placed into the ILs, separated by filter paper to avoid short circuit. IL (150−300 μL) was added into the electrochemical cell as the electrolyte. The barriers formed by filter papers and the IL which is relatively viscous could minimize the amount of CO to diffuse to the quasi Pt reference electrode and adsorb on it that can affect the stability of quasi Pt reference electrode potential. Electrochemical Characterization Methods and Instruments. The CO oxidation experiments were carried out by CV and anodic ASC using electrochemical potentiostats. The details of these experiments and some results are shown in Supporting Information Figures S1 and S6−S11 and Tables S3 and S4. Computational Study. The wb97x-D functional implemented in the Gaussian 1636 suite of programs was used to compute the energetics and equilibrium thermodynamics of reactions including the effect of the IL environment (ε = 11.7) by use of the integral equation formalism of the polarized continuum model (IEF-PCM) of Tomasi et al.17 The structures of NTf2 anion and Bmpy+ and Bmim+ cations was fully optimized using the 6-31++G** basis set. The computational calculation results are shown in Table S2 and Figures S3−S5.

Figure 1. CV of (a) [Bmpy][NTf2], (b) [Bmim][NTf2] in nitrogen (black) and in 5% v/v CO (red). Pt electrode, scan rate: 100 mV/s. The potential window is between 0 and 1.7 V.

result from the presence of trace water.37−39 These peaks are relatively reversible. The broad anodic peak between 1.4 and 1.7 V is ascribed to the NTf2− anion oxidation to NTf•2 radical and the oxidation of the Pt.40 In the cathodic scan, the NTf•2 radical can be reduced between 0.8 and 1.2 V.40 The peak assignments of these redox processes are consistent with our early work.40−42 In the presence of CO, the oxidation of adsorbed CO presents an irreversible oxidation peak between 1.4 and 1.7 V, as shown in Figure 1a,b. The CO oxidation peak is broad suggesting the oxidation of adsorbed CO at different Pt crystalline sites with a range of the oxidation potentials in these two ILs.41 Our results are consistent with the results of CO oxidation in the ILs at different Pt single-crystal electrodes.43 Because of the competition of CO, NTf2− and trace water for the adsorption sites on the Pt electrode, it is expected that NTf2− anion oxidation peak should be much smaller than those in the background CV (Figure 1). As more CO adsorbs, NTf2− and water adsorptions on the Pt electrode are reduced. Figure 1 shows that the proton reduction peak and the hydrogen oxidation peak are a little larger than those in the background CV. There is also an increase of NTf2• reduction peak after the CO oxidation. It is rationalized that the proton produced by CO oxidation can quickly neutralize the product of NTf2− by forming HNTf2 that can be quickly removed from the Pt electrode surface.44 This process increases the rate of NTf•2 reduction to form NTf2−. The proton produced because of CO oxidation also increases the proton reduction peak current. The nature and structure of the cation and anion can change the IL−electrode interface properties.45,46 It has been previously reported that different IL anions can shift the CO

■

RESULTS AND DISCUSSION Characterization of CO Adsorption and Oxidation in [Bmpy][NTf2] and [Bmim][NTf2] by CV. It is well known that CO adsorbs directly on the Pt electrode surface. The complexities involved in CO adsorption and oxidation on the Pt electrode with the ILs are reflected in the strong sensitivity of this reaction to the Pt surface structure and the adsorbed anions.15 Electrochemical CO oxidation at three different Pt(hkl) electrodes corresponding to the three basal crystallographic planes were shown to occur at different potentials in 1ethyl-3-methylimidazolium tetrafluoroborate [C2mim][BF4] and 1-ethyl-3-methylimidazolium bis(trifluoromethylsulfonyl)imide [C2mim][NTf2] electrolytes. The catalytic activity of Pt(hkl) toward CO oxidation is dependent on the availability of sites for adsorption of OH radicals from the trace water present in the ILs. We performed CV experiments using the ILs at ambient conditions with less than 1000 ppm water (Supporting Information Figure S2). Figure 1 shows the CV results with the ILs with and without CO in [Bmim][NTf2] and [Bmpy][NTf2], respectively. Without CO, the CVs show similar features in both [Bmim][NTf2] and [Bmpy][NTf2]. In the anodic scan, there are two anodic peaks, and two corresponding cathodic peaks were observed in the cathodic scan. The redox peaks between 0 and 0.4 V are attributed to proton reduction and hydrogen oxidation processes which C

DOI: 10.1021/acs.jpcb.8b11602 J. Phys. Chem. B XXXX, XXX, XXX−XXX

Article

The Journal of Physical Chemistry B Table 1. Redox Processes in the IL at Double-Potential Step Chronoamperometry no CO

anodic stripping processes (step from 0.1 to 1.6 V) double-layer charging process (Qox‑EDL) (1) NTf2− ↔ NTf•2 + e−

cathodic reduction process (step from 1.6 to 0.1 V) (1) double-layer discharging process (Qred‑EDL) (2) NTf•2 + e− ↔ NTf2− (3) NTf2H + e− → NTf2− + 1/2H2 (trace water)

total − charge/blank Qox‑blank = Qox‑EDL + QNTf2 with CO (1) double-layer charging process (2) NTf2− ↔ NTf•2 + e− (3) CO + OH → CO2 + H+ +e total Qox‑CO = Qox‑EDL + QNTf2− + Qox‑CO charge/CO

Q red ‐ blank = Q red ‐ EDL + Q NTf • + Q red ‐ H+‐ b

ΔQox = Qox‑CO − Qox‑blank = Qox‑EDL + QNTf2− + Qox‑CO − Qox‑EDL‑b − QNTf2− calibrated charge (ΔQ) ≈ Qox‑CO

ΔQ red = Q red ‐ CO − ΔQ red ‐ blank = Q red ‐ EDL + Q NTf

2

(1) double-layer discharging process (2) NTf•2 + e− ↔ NTf2− (3) NTf2H + e− → NTf2− + 1/2H2 Qred‑CO = Qox‑EDL + QNTf2− + Qred‑H+ 2

−

+ Q red ‐ H+‐ b

− Q red ‐ EDL − Q NTf • − Q H+ ≈ Q red ‐ H+ ∝ Q ox ‐ CO 2

It is expected that, at low CO adsorption coverage while no competition If Qred‑EDL is not affected by the product CO2, then ΔQred = Qred‑CO − between CO and NTf2− anion adsorption at Pt surface, in the oxidation Qred‑blank = Qred‑H+. The Qred‑H+ is directly related to [CO]ads based process, if Qox‑EDL is not affected by CO adsorption, then ΔQox = Qox‑CO on the reaction: CO + OH → CO2 + H+ +e − Qox‑blank = Qox‑CO

oxidation potential.15,42 As shown in Figure 1, there are subtle differences between the CO oxidation peak and proton redox peaks in [Bmpy][NTf2] and [Bmim][NTf2]. This suggests that the cation of the ILs can also affect the adsorption of CO and water on the Pt electrode. Bmim+ cation is aromatic, whereas the Bmpy+ cation contains mainly hydrocarbons; thus, these two ILs can form different double-layer structures at the IL− electrode interface through van der Waals and Coulombic interactions with the charged surface and cause the subtle difference in CO oxidation peak potential. DFT Calculations. The interaction of CO with the cation of the IL (Bmim+ or Bmpy+) is expected to be very weak because there is no significant dipole moment on the CO. DFT calculations using wb97x-D/6-31+G** and wb97x-D/6-31+ +G** methods including dispersion interactions are used in the calculations. The IL environment by use of the IEF-PCM solvation model (ε = 11.7) was used in the calculation, and thermodynamic free energies [free energy (G)] and enthalpies [enthalpy (H)] at 298 K were evaluated via a vibrational analysis. These energies are presented in the Supporting Information. The electrostatic potential surfaces of [Bmpy][NTf2] and [Bmim][NTf2] are shown in Figure S3. The calculation shows that the enthalpy change for the interaction of Bmpy+ and Bmim+ with CO are −0.24 and −0.17 kcal/mol, respectively. This confirms that there is no significant interaction between CO and either Bmpy+ or Bmim+. We also performed theoretical calculations employing DFT (wb97x-D) to estimate the interaction energies between OH− with Bmpy+ or Bmim+ as a first step to understand IL− electrode interface properties. It is expected that a stronger interaction of OH− with the cation of the IL will decrease the interaction of OH− on the Pt electrode surface. We optimized using the wb97x-D/6-31+G** method including the IL environment by use of the IEF-PCM approach (ε = 11.7) (Figure S4). The calculated thermodynamic enthalpies in atomic units for each molecule are listed in the Supporting Information after Figure S3 and in Table S1. The enthalpy of interaction of Bmpy+ with OH− to form [Bmpy][OH] is −13.5 kcal/mol, whereas the enthalpy of interaction of Bmim+ with OH− to form [Bmim][OH] is −21.0 kcal/mol. These DFT results show that Bmim+ has a stronger interaction with OH− compared with the interaction of Bmpy+ with OH−. The differing behavior of the cations of these two ILs can influence

the IL−Pt electrode interface double-layer structure and the subsequent CO oxidation currents and charges. The charge for CO oxidation is higher in [Bmpy][NTf2] than [Bmim][NTf2]. However, the charge is the sum of the CO oxidation and NTf2− anion oxidation and the double-layer charge. Thus, the charge needs to be calibrated for the blank charge for the CO quantification as detailed in the later sections. Characterization of CO Adsorption and Oxidation in [Bmpy][NTf2] and [Bmim][NTf2] with or without Added Water at a Wide Potential Window by CV. The adsorption of water and CO at the Pt electrode was studied in the aqueous electrolyte, suggesting surface coverage-dependent formation of carbon monoxide monohydrate.47 We further characterize CO oxidation of the two ILs at a wide potential window (0− 2.5 V). The ILs were from the same vendor, but different batches were used in these experiments. It is expected that the amount of trace water present in the ILs from different batches could be different. This could affect the redox potential of CO. Figures S6−S8 shows the CVs of CO oxidation with the original IL and IL with 2.3% added water. It is clear that adding more water in the ILs results in significant changes in the CVs of the redox processes of CO, proton, and NTf2− in both ILs. Water can increase the conductivity of the ILs which reduces the IR drop in the IL electrolyte. This can lead to a shift in the reference electrode potential. Without the presence of CO, the added water in the IL results in a shift of the proton reduction peak to a more negative value in both ILs. Water could also affect the IL−electrode interface structure because of its adsorption. In the presence of a greater quantity of water, the peak charge of CO oxidation decreased significantly in [Bmim][NTf2] versus [Bmpy][NTf2]. This indicates that a significant amount of OH adsorbed at the IL−Pt interface decreased CO adsorption in [Bmim][NTf2]. As shown in Figure S8 (the Supporting Information), the CO oxidation peak potential shifted negatively in [Bmim][NTf2] with additional water added. In contrast, there is significant change in the CO oxidation peak potential, current, and shape in [Bmpy][NTf2] when 2.3% additional water is added to the IL. Recent studies show that water can hydrogen bond with ILs; water clusters form linear chains of hydrogen-bonded molecules almost exclusively. At a low water content, the ions are selectively coordinated by individual water molecules, but their ionic network is largely unperturbed.7,48 Although the D

DOI: 10.1021/acs.jpcb.8b11602 J. Phys. Chem. B XXXX, XXX, XXX−XXX

Article

The Journal of Physical Chemistry B

achieved by evaluating charge instead of current. Charge is the integral of current over time and charge is more favorable for data analysis for potential step experiments because integration decreases the noise level. According to the Faraday law, Q = nFN (n is the number of electron transfer, F is the Faraday constant, and N is the number of moles of the redox species). N is proportional to the moles of CO on the electrode surface which in turn is dependent on the surface coverage of CO. The surface coverage of CO depends on the time of CO adsorption and CO concentration in the gas phase. The greater the CO concentration is in the gas phase, the shorter the time is needed for a full surface coverage of CO to be reached. As shown in Table 1, obtaining the charge from the current versus time curves for both anodic and cathodic faradic and nonfaradic processes with or without CO, quantitative analysis of CO and new physical insights into the IL−electrode double layer and its influence on CO adsorption and oxidation processes in the ILs can be obtained. For example, we will be able to understand whether the double-layer structures of the IL−electrode will change at varying CO adsorption coverages. If a quantitative relationship between CO concentration and the charge exists in both anodic oxidation and cathodic reduction processes after subtracting the blank charge, then the double layer of the IL−Pt electrode can be considered constant during the potential step experiments and will be independent of the amount of the CO adsorption. A fundamental understanding of the CO oxidation process in the ILs is essential for the development of the new electroanalytical methods for CO detection exploring the unique characteristics of both the adsorption and redox behavior of CO at the IL− electrode interface. Relationship of CO Oxidation Charge Versus CO Concentration in the ILs. In ASC experiments, the adsorption time of CO at 0.1 V was varied to control the surface concentration of CO when a constant CO concentration in gas phase was maintained. This experiment was repeated with a range of CO concentrations in gas phase between 0.5 and 10% (v/v). Figures S9−S11 (the Supporting Information) illustrate the details of the ASC experiments; Tables S3 and S4 summarize the integrated charge (Q) data from these ASC experiments. The charge Q versus [CO] concentration curves for oxidation processes (Figure 2) and reduction processes (Figure 3) were analyzed and compared for [Bmpy][NTf2] and [Bmim][NTf2], respectively. The charge (the y axis) shown in Figures 2 and 3 was calibrated by subtracting the charge of the blank exp. (identical exp. condition without adsorbed CO shown in Tables S3 and S4). Figure 2a shows the oxidation charge versus CO concentration ranges from 0.5 to 10% in [Bmpy][NTf2] when the adsorption time was controlled between 5 and 120 s. There is an obvious concentration-dependent feature at each adsorption time tested in [Bmpy][NTf2]. There are two quantitative positive linear relationships of charge versus concentration of CO, one in 0−2% and the other in 2.5− 10% CO. These results support that the double-layer structure and properties are negligibly affected by the CO adsorption and the redox processes at the electrode interface. It also suggests several important characteristics of the [Bmpy][NTf2]−Pt interface: (1) there are plenty of adsorption sites at [Bmpy][NTf2]−Pt interface for CO; (2) the CO adsorption at [Bmpy][NTf2]−Pt interface is very fast and the mass transport of CO to the IL−electrode interface by diffusion is the ratelimiting step for the current response. Thus, CO concentration

detailed mechanisms of water effects on the CO adsorption and oxidation processes in the ILs need further investigation, this study suggests that trace water in the original ILs (less than 1000 ppm) is sufficient and beneficial for CO oxidation processes, and adding additional water is detrimental to these processes possibly due to the competing adsorption of water with CO on the Pt surface. Double-Potential Step Stripping Chronoamperometry of CO Adsorption and Oxidation in [Bmpy][NTf2] and [Bmim][NTf2]. CV is an excellent technique to aid the understanding of redox processes, but it is not the best sensing technique. Because CO adsorbs on the Pt electrode strongly, anodic ASC was used for CO sensor development. In ASC, the analyte is first deposited (if it is a metal ion) or adsorbed (if it is a neutral molecule such as CO here) at the working electrode at a constant potential. After an accurately measured period, the analyte is oxidized and stripped from the working electrode. The deposition or adsorption step enables the electrochemical preconcentration of the analyte, that is, the concentration of the analyte at the surface of the working electrode is far greater than it is in bulk solution. Owing to this preconcentration step, stripping methods yield the lowest detection limits of all voltammetric methods. As summarized in Table S2 and also shown in Table 1 below, the adsorption and oxidation of CO occur concurrently with the redox processes of NTf2− anion and proton but at different electrode potentials in the ILs. As shown in Figures S6−S8 in the Supporting Information, water affects the peak potential and peak charge of these redox processes. Thus, it is important to perform CV in the electrochemical sensor to determine the peak potential to select the best applied potential for amperometric CO sensing.49 The electrochemical sensor with CV shown in Figure 1 was used to carry out chronoamperometry experiments. On the basis of Figure 1, the CO oxidation potential is ∼1.6 V. This potential was selected as the anodic stripping potential for the CO sensor test. The CO adsorption potential was set at 0.1 V. At 0.1 V, the cation should also adsorb at the IL−Pt electrode interface. As shown in Figure S9 in the Supporting Information, we control the amount of CO adsorption by controlling the CO adsorption time at 0.1 V ranging from 5 to 120 s while keeping the CO concentration in the gas-phase constant. Then, the working electrode potential is stepped to 1.6 V to allow the CO oxidation process to occur for an accurately measured period of time (i.e., 60 s). At 1.6 V potential, although NTf2− anion oxidation to NTf2• radical occurs concurrently, the CO oxidation process dominates (as shown in Figures 1 and S6− S8). The electrode potential is then stepped back to 0.1 V where the NTf•2 radical and proton reduction processes occur. In these experiments (Figure S9), CO at a specific concentration in the gas phase was maintained throughout. It is known that CO chemisorbs at the Pt surface and CO2 physisorbs at the Pt electrode.50 Thus, the strong adsorption of CO at Pt electrode and high viscosity of IL could reduce the likelihood of CO in the gas phase to be readsorbed at the electrode interface after CO was oxidized to CO2 at the experimental timescale. The multiple processes during the double-potential steps are summarized in Tables 1 and S2. During these double-potential step experiments, besides the faradic processes, the nonfaradic processes of charging and discharging at IL−Pt electrode interface occur concurrently. Because of the multiple redox processes in the ASC, the quantitative determination of the CO concentration is best E

DOI: 10.1021/acs.jpcb.8b11602 J. Phys. Chem. B XXXX, XXX, XXX−XXX

Article

The Journal of Physical Chemistry B

Figure 3. Reduction charge vs CO concentration in gas phase from Table S4 in (a) [Bmpy][NTf2] and (b) [Bmim][NTf2], the charge is calibrated by subtracting the blank.

Figure 2. Oxidation charge vs CO concentration in gas phase from Table S3 in (a) [Bmpy][NTf2] (b) [Bmim][NTf2], the charge was calibrated by subtracting the charge obtained in the blank experiment (i.e., without CO at identical experimental conditions).

which hinders the accessibility of additional CO onto the electrode surface.51−53 Furthermore, the higher binding activity of OH− with Bmim+ led to a low concentration of OH− anions on the Pt electrode surface that can facilitate the oxidation of CO.52 The smaller signal of CO oxidation in [Bmim][NTf2] in Figure 3b compared with those signals observed in [Bmpy][NTf2] (Figure 3a) supports this rationalization. In general, the CO oxidation charge in [Bmpy][NTf2] is larger than that obtained from [Bmim][NTf2], which is in accord with our conclusion that the double-layer structure in [Bmim][NTf2] hinders the adsorption of CO onto the electrode, as well as the higher affinity of OH− to [Bmim]+ cation. At 0.1 V, the higher affinity of Bmim+ compared with Bmpy+ at the Pt electrode will reduce the free sites available for CO adsorption on Pt and make it a rate-limiting process in [Bmim][NTf2]. In contrast, in [Bmpy][NTf2], the large Bmpy+ cation allows the IL−electrode interface to have sufficient free space for CO adsorption which contributes to the greater amount of CO adsorption in [Bmpy][NTf2]. This charge difference for CO oxidation in these two ILs on the Pt surface is consistent with the current versus concentration data (Figures S9−S10). The difference in viscosities of [Bmpy][NTf2] (74 cp) and [Bmim][NTf2] (59 cp) may also contribute to the observed difference in the charge and current versus CO concentration results.32 Similar phenomena have been previously reported by Herrero and co-workers where lower viscosity of [Bmim][NTf2] resulted in higher current density for CO oxidation than [Bmim][BF4] which has much higher viscosity than [Bmim][NTf2].15 Further in situ spectroelectrochemical investigations are required to fully

in the gas phase can be quantified by stripping chronoamperometry. The fast adsorption of CO at the [Bmpy][NTf2]−Pt electrode interface could potentially exclude the interference from other species in the gas phase. In contrast, significantly different ASC experimental results were observed with [Bmim][NTf2] versus those with [Bmpy][NTf2] when the same sets of experiments were performed. As shown in Figure 2b, there is a negative correlation of charge versus [CO] when the CO concentration is less than 2% and the charge remains almost constant when CO concentration is greater than 2%. The negative charge suggests an overcalibration by subtracting the blank charge. The negative concentration-dependent feature of CO in [Bmim][NTf2] at low CO concentration suggests that the double-layer charge in the blank is bigger than the double-layer charge in the presence of CO. This can be rationalized by the adsorption of CO which changes the double-layer structure at the [Bmim][NTf2]−Pt interface to be more ordered which in turn reduces the doublelayer thickness. The compact structure and stronger adsorption of Bmim+ on Pt results in a more ordered interface structure. The adsorption of CO at the interface may induce further ordering. This rationalization is supported by other studies in which the coadsorbate-CO at the electrode surface is suggested to induce an ordering which was reported in the case of benzene mixed with CO at Pd, Rh, and Pt(111) electrode surfaces. Benzene alone adsorbs in a disordered or weakly ordered manner at room temperature. The CO adsorption has limited Pt surface accessible sites because of the competing adsorption of Bmim+ in [Bmim][NTf2].51 At higher CO concentrations, the [Bmim][NTf2] has limited free space F

DOI: 10.1021/acs.jpcb.8b11602 J. Phys. Chem. B XXXX, XXX, XXX−XXX

Article

The Journal of Physical Chemistry B

versus [CO] curve can be used for the quantitative analysis of the CO concentration. The slope of charge versus [CO] curve indicates the sensitivity of the quantification. An intercept of zero shows a perfect calibration. The results at varying adsorption times allow the determination of time needed to obtain a monolayer of CO adsorption coverage. The linear range of the data is the analytical range where CO can be quantitatively measured. Tables S5 and S6 (the Supporting Information) summarize the linear regression fits of the charge versus [CO] at different CO adsorption times in [Bmpy][NTf2] and [Bmim][NTf2], respectively. Perfect calibration should show a linear regression fit that has zero intercept and a coefficient of linearity R close to 1. In both [Bmpy][NTf2] and [Bmim][NTf2], our results show that at 30 s adsorption time gives the best linear fit at CO concentration (range 0−2.0%) and also at high CO concentration (2.5−10%) based on CO oxidation in the anodic potential step experiments. The best adsorption time for quantification of CO using the cathodic reduction process is 60 s for low CO concentration (range 0−2.0%) as well as at higher CO concentration (2.5−10%). Interestingly, the negative correlation for the reduction process is quantitatively related to CO concentration in both low and high [CO] in all adsorption times tested. This shows that the calibration method could be effective even though the background faradic and nonfaradic processes are dominant in the cathodic processes. The intercepts for both concentration region changes slightly with the adsorption times for the oxidation processes but more changes are shown in the anodic processes at short adsorption times.

understand the role of the cation and anion of the ILs on the adsorption mechanism of CO on the Pt surface in these ILs. This will help to optimize the parameters for CO sensors, thus resulting in reliable, sensitive, and selective IL-based electrochemical CO sensors. Relationship of Reduction Charge Versus CO Concentration in the ILs after the ASC Experiments. As shown in Table S2, the CO oxidation will produce a proton which can be reduced when the potential is stepped from 1.6 V back to 0.1 V. Additionally, protons from the trace water in the original IL can also be reduced. As shown in Figures S9−S11, when the potential was stepped back from 1.6 to 0.1 V, a noticeable reduction current was presented in all ASC experiments at varying CO concentrations. Figure 3a,b compares the reduction charge versus CO concentrations tested at different adsorption times (5−120 s) and different CO concentrations (0−10%). It is found that in [Bmpy][NTf2], there is a good quantitative linear relationship between the reduction charge versus CO concentrations between 0 and 2.5%, and the charge signals reach a plateau between 3 and 5% of CO. This result further supports that the double-layer structure of [Bmpy][NTf2]−Pt remains constant and does not change with the adsorption of CO. It also demonstrates better linearity when compared with the CO oxidation processes as shown in Figure 3a. This result not only does support the suitability of the [Bmpy][NTf2] as the electrolyte for CO electrochemical sensor development but also shows that the proton generated from CO oxidation is better for quantification as only those CO molecules undergoing oxidation are quantified. Interestingly, as shown in Figure 3b, there is a negative correlation of reduction charge versus CO concentration between 0 and 2% CO and a positive correlation of reduction charge versus [CO] between 2.5 and 10% in [Bmim][NTf2]. This is consistent with those observed in Figure 3b. The negative concentration-dependent feature of CO in [Bmim][NTf2] at low CO concentration suggests that the charge of the double layer and the proton reduction from the trace water in the blank is larger than those in the presence of CO. In [Bmpy][NTf2], the high amount of free space likely contributes to fast diffusion of CO2 to the bulk and little change of the interface structure because of the CO 2 production. In [Bmim][NTf2], slower diffusion of CO2 to the bulk results in the accumulation of CO2 on the Pt surface, which leads to a decrease of the double-layer thickness because of higher ordering. At higher CO concentrations, the negative correlation changes to the positive correlation. This is rationalized as more protons are being generated from CO oxidation. The charge due to proton reduction increases at higher CO concentration. Furthermore, high CO concentration generates more CO2 that can significantly change the adsorption of Bmim+ cations on the electrode surface by reorganization of surface cations. This could lead to an increase of the double-layer thickness and decrease of the double-layer charge. Both effects will make the total reduction charge higher than that of the blank and lead to positive charge value in the y axis in Figure 3b. Analytical Methods Based on the Double-Potential Step Chronoamperometry. The ASC results above show that both the double-layer charging/discharging and the IL and trace water redox processes can be calibrated by subtracting the charge obtained in the blank experiments (i.e., without the presence of CO) summarized in Table 1 and data shown in Tables S3 and S4 in the Supporting Information. The charge

■

CONCLUSIONS In this work, we studied the CO adsorption and oxidation in [Bmpy][NTf2] and [Bmim][NTf2] in the presence of trace water by CV and ASC. In addition to CO oxidation, there are multiple other redox processes, that is, the anion NTf2− oxidation and proton reduction processes in these ILs. We systematically characterized the multiple redox processes in the ILs by using a double-potential ASC technique between 0.1 and 1.6 V. We vary the amounts of CO adsorbed by controlling the adsorption time of CO at 0.1 V. The multiple redox processes and the double-layer structure, which change during the potential step experiments in the ILs, can be effectively calibrated by subtracting the charge of the blank (i.e., without CO). This allows us to demonstrate an innovative electrochemical CO sensor utilizing the unique properties of CO interfacial chemistry at the IL−Pt electrode interface. The CO sensor demonstrated here is simple and sensitive and can be used in ambient conditions with no need to remove water from the IL which is not practical and sometimes impossible for real world sensing application when ILs are used as solvents and electrolytes. The CO sensor method is compared between the two ILs. The cations Bmpy+ and Bmim+ are found to contribute to different electrochemical behaviors of CO oxidation in these ILs. It is found that the adsorption of CO is dependent on the IL−Pt interface structure and that the structure depends on the nature the IL cation. The CO adsorption process is found to be relatively slow at the IL−Pt interface and the amount of CO adsorption is time dependent. DFT studies suggest that the different binding activities of OH− and CO between two cations Bmpy+ and Bmim+ could contribute to different electrochemical double-layer behaviors. Our results show that quantification of G

DOI: 10.1021/acs.jpcb.8b11602 J. Phys. Chem. B XXXX, XXX, XXX−XXX

Article

The Journal of Physical Chemistry B

(9) Chesler, P.; et al. Nanostructured Sno2-Zno Composite Gas Sensors for Selective Detection of Carbon Monoxide. Beilstein J. Nanotechnol. 2016, 7, 2045−2056. (10) Nwaboh, J. A.; Qu, Z.; Werhahn, O.; Ebert, V. Interband Cascade Laser-Based Optical Transfer Standard for Atmospheric Carbon Monoxide Measurements. Appl. Opt. 2017, 56, E84−E93. (11) Stetter, J. R.; Korotcenkov, G.; Zeng, X.; Tang, Y.; Liu, Y. Electrochemical Gas Sensors: Fundamentals, Fabrication, and Parameters. In Chemical Sensors Comprehensive Sensor Technologies; Korotcenkov, G., Ed.; Momentum Press, LLC: New York, 2011; Vol. 5, pp 1−89. (12) Gao, F.; McClure, S. M.; Cai, Y.; Gath, K. K.; Wang, Y.; Chen, M. S.; Guo, Q. L.; Goodman, D. W. CO Oxidation Trends on PtGroup Metals from Ultrahigh Vacuum to near Atmospheric Pressures: A Combined in Situ Pm-Iras and Reaction Kinetics Study. Surf. Sci. 2009, 603, 65−70. (13) Freund, H.-J.; Meijer, G.; Scheffler, M.; Schlögl, R.; Wolf, M. CO Oxidation as a Prototypical Reaction for Heterogeneous Processes. Angew. Chem., Int. Ed. 2011, 50, 10064−10094. (14) Twigg, M. V. Progress and Future Challenges in Controlling Automotive Exhaust Gas Emissions. Appl. Catal., B 2007, 70, 2−15. (15) Hanc-Scherer, F. A.; Sánchez-Sánchez, C. M.; Ilea, P.; Herrero, E. Surface-Sensitive Electrooxidation of Carbon Monoxide in Room Temperature Ionic Liquids. ACS Catal. 2013, 3, 2935−2938. (16) Ejigu, A.; Johnson, L.; Licence, P.; Walsh, D. A. Electrocatalytic Oxidation of Methanol and Carbon Monoxide at Platinum in Protic Ionic Liquids. Electrochem. Commun. 2012, 23, 122−124. (17) Tomasi, J.; Mennucci, B.; Cammi, R. Quantum Mechanical Continuum Solvation Models. Chem. Rev. 2005, 105 (8), 2999−3094. (18) Baldelli, S. Surface Structure at the Ionic Liquid−Electrified Metal Interface. Acc. Chem. Res. 2008, 41, 421−431. (19) Hayes, R.; Borisenko, N.; Tam, M. K.; Howlett, P. C.; Endres, F.; Atkin, R. Double Layer Structure of Ionic Liquids at the Au(111) Electrode Interface: An Atomic Force Microscopy Investigation. J. Phys. Chem. C 2011, 115, 6855−6863. (20) Lockett, V.; Horne, M.; Sedev, R.; Rodopoulos, T.; Ralston, J. Differential Capacitance of the Double Layer at the Electrode/Ionic Liquids Interface. Phys. Chem. Chem. Phys. 2010, 12, 12499−12512. (21) Hayes, R.; El Abedin, S. Z.; Atkin, R. Pronounced Structure in Confined Aprotic Room-Temperature Ionic Liquids. J. Phys. Chem. B 2009, 113, 7049−7052. (22) Yuan, H.; Shimotani, H.; Tsukazaki, A.; Ohtomo, A.; Kawasaki, M.; Iwasa, Y. Hydrogenation-Induced Surface Polarity Recognition and Proton Memory Behavior at Protic-Ionic-Liquid/Oxide ElectricDouble-Layer Interfaces. J. Am. Chem. Soc. 2010, 132, 6672−6678. (23) Wang, Z.; Guo, M.; Mu, X.; Sen, S.; Insley, T.; Mason, A. J.; Král, P.; Zeng, X. Highly Sensitive Capacitive Gas Sensing at Ionic Liquid-Electrode Interfaces. Anal. Chem. 2016, 88, 1959−1964. (24) Wang, Z.; Guo, M.; Baker, G. A.; Stetter, J. R.; Lin, L.; Mason, A. J.; Zeng, X. Methane-Oxygen Electrochemical Coupling in an Ionic Liquid: A Robust Sensor for Simultaneous Quantification. Analyst 2014, 139, 5140−5147. (25) Chi, X.; Tang, Y.; Zeng, X. Electrode Reactions Coupled with Chemical Reactions of Oxygen, Water and Acetaldehyde in an Ionic Liquid: New Approaches for Sensing Volatile Organic Compounds. Electrochim. Acta 2016, 216, 171−180. (26) Ohlin, C. A.; Dyson, P. J.; Laurenczy, G. b. Carbon monoxide solubility in ionic liquids: determination, prediction and relevance to hydroformylationElectronic supplementary information (ESI) available: further experimental details. See http://www.rsc.org/suppdata/ cc/b4/b401537a/. Chem. Commun. 2004, 1070−1071. (27) Choe, S. J.; Park, D. H.; Huh, D. S. Adsorption and Dissociation Reaction of Carbon Dioxide on Pt(111) and Fe(111) Surface: Mo-Study. Bull. Korean Chem. Soc. 2000, 21, 779−784. (28) Gunasooriya, G. T. K. K.; Saeys, M. Co Adsorption Site Preference on Platinum: Charge Is the Essence. ACS Catal. 2018, 8, 3770−3774.

CO can be achieved in both [Bmpy][NTf2] and [Bmim][NTf2] ILs based on either oxidation processes or reduction processes by selecting an optimum adsorption time. Thus, the anodic ASC method could be used to study the heterogeneous catalysis of CO oxidation in structurally different ILs and to further optimize the selection of the best IL electrolytes for the development of next-generation miniaturized real-time and continuous CO sensors with enhanced sensitivity and selectivity.

■

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.jpcb.8b11602. Details of electrochemical sensor, experimental details as well as some of the supporting experimental results; total thermodynamic enthalpies; main redox processes in the ILs with and without CO; integrated charge of CO oxidation and reduction double layer charge; and linear fits (PDF)

■

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. ORCID

Yongan Tang: 0000-0002-9818-3372 Anil Kumar: 0000-0002-9979-7798 Michael Sevilla: 0000-0001-8799-5458 Xiangqun Zeng: 0000-0003-3867-226X Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS We acknowledge the support of NIH (R01ES022302) for this work. We thank Tongtong Chen’s help for the IR experiment and Dr. Arun Sridhar’s help for additional experiment in the revision of this work.

■

REFERENCES

(1) Rogers, R. D.; Voth, G. A. Ionic Liquids. Acc. Chem. Res. 2007, 40, 1077−1078. (2) Han, X.; Armstrong, D. W. Ionic Liquids in Separations. Acc. Chem. Res. 2007, 40, 1079−1086. (3) MacFarlane, D. R.; Forsyth, M.; Howlett, P. C.; Pringle, J. M.; Sun, J.; Annat, G.; Neil, W.; Izgorodina, E. I. Ionic Liquids in Electrochemical Devices and Processes: Managing Interfacial Electrochemistry. Acc. Chem. Res. 2007, 40, 1165−1173. (4) Rehman, A.; Zeng, X. Methods and Approaches of Utilizing Ionic Liquids as Gas Sensing Materials. RSC Adv. 2015, 5, 58371− 58392. (5) Rehman, A.; Zeng, X. Ionic Liquids as Green Solvents and Electrolytes for Robust Chemical Sensor Development. Acc. Chem. Res. 2012, 45, 1667−1677. (6) Xiao, C.; Zeng, X. In Situ Eqcm Evaluation of the Reaction between Carbon Dioxide and Electrogenerated Superoxide in Ionic Liquids. J. Electrochem. Soc. 2013, 160, H749−H756. (7) Cheng, H.-W.; Stock, P.; Moeremans, B.; Baimpos, T.; Banquy, X.; Renner, F. U.; Valtiner, M. Characterizing the Influence of Water on Charging and Layering at Electrified Ionic-Liquid/Solid Interfaces. Adv. Mater. Interfaces 2015, 2, 1500159. (8) Birkefeld, L. D.; Azad, A. M.; Akbar, S. A. Carbon Monoxide and Hydrogen Detection by Anatase Modification of Titanium Dioxide. J. Am. Ceram. Soc. 1992, 75, 2964−2968. H

DOI: 10.1021/acs.jpcb.8b11602 J. Phys. Chem. B XXXX, XXX, XXX−XXX

Article

The Journal of Physical Chemistry B

(50) Freund, H.-J.; Roberts, M. W. Surface Chemistry of Carbon Dioxide. Surf. Sci. Rep. 1996, 25, 225−273. (51) Tang, Y.; Wang, Z.; Chi, X.; Sevilla, M. D.; Zeng, X. In Situ Generated Platinum Catalyst for Methanol Oxidation Via Electrochemical Oxidation of Bis(Trifluoromethylsulfonyl)Imide Anion in Ionic Liquids at Anaerobic Condition. J. Phys. Chem. C 2016, 120, 1004−1012. (52) Liu, Z.; Huang, Z. Y.; Cheng, F. F.; Guo, Z. H.; Wang, G. D.; Chen, X.; Wang, Z. Efficient Dual-Site Carbon Monoxide ElectroCatalysts Via Interfacial Nano-Engineering. Sci. Rep. 2016, 6, 33127. (53) Sanchez-Castillo, M. A.; Couto, C.; Kim, W. B.; Dumesic, J. A. Gold-Nanotube Membranes for the Oxidation of CO at Gas-Water Interfaces. Angew. Chem., Int. Ed. 2004, 43, 1140−1142.

(29) Rivera-Rubero, S.; Baldelli, S. Surface Spectroscopy of Roomtemperature Ionic Liquids on a Platinum Electrode: A Sum Frequency Generation Study. J. Phys. Chem. B 2004, 108, 15133−15140. (30) Xu, S.; Xing, S.; Pei, S.-S.; Ivaništšev, V.; Lynden-Bell, R.; Baldelli, S. Molecular Response of 1-Butyl-3-Methylimidazolium Dicyanamide Ionic Liquid at the Graphene Electrode Interface Investigated by Sum Frequency Generation Spectroscopy and Molecular Dynamics Simulations. J. Phys. Chem. C 2015, 119, 26009−26019. (31) Burrell, A. K.; Sesto, R. E. D.; Baker, S. N.; McCleskey, T. M.; Baker, G. A. The Large Scale Synthesis of Pure Imidazolium and Pyrrolidinium Ionic Liquids. Green Chem. 2007, 9, 449−454. (32) Xiao, C.; Rehman, A.; Zeng, X. Dynamics of Redox Processes in Ionic Liquids and Their Interplay for Discriminative Electrochemical Sensing. Anal. Chem. 2012, 84, 1416−1424. (33) Wang, Z.; Lin, P.; Baker, G. A.; Stetter, J.; Zeng, X. Ionic Liquids as Electrolytes for the Development of a Robust Amperometric Oxygen Sensor. Anal. Chem. 2011, 83, 7066−7073. (34) Chao, Y.; Yao, S.; Buttner, W. J.; Stetter, J. R. Amperometric Sensor for Selective and Stable Hydrogen Measurement. Sens. Actuators, B 2005, 106, 784−790. (35) Stetter, J. R.; Penrose, W. R.; Yao, S. Sensors, Chemical Sensors, Electrochemical Sensors, and Ecs. J. Electrochem. Soc. 2003, 150, S11−S16. (36) Frisch, M. J.; et al. Gaussian 09, Revision B.01, Wallingford CT, 2009. (37) de Souza, R. F.; Padilha, J. C.; Gonçalves, R. S.; Rault-Berthelot, J. Dialkylimidazolium Ionic Liquids as Electrolytes for Hydrogen Production from Water Electrolysis. Electrochem. Commun. 2006, 8, 211−216. (38) Meng, Y.; Aldous, L.; Belding, S. R.; Compton, R. G. The Formal Potentials and Electrode Kinetics of the Proton/Hydrogen Couple in Various Room Temperature Ionic Liquids. Chem. Commun. 2012, 48, 5572−5574. (39) Craig, J. H. Adsorption of hydrogen and carbon monoxide on platinum. Surf. Sci. 1981, 111, L695−L700. (40) Tang, Y.; Wang, Z.; Chi, X.; Sevilla, M. D.; Zeng, X. In Situ Generated Platinum Catalyst for Methanol Oxidation Via Electrochemical Oxidation of Bis(Trifluoromethylsulfonyl)Imide Anion in Ionic Liquids at Anaerobic Condition. J. Phys. Chem. C 2016, 120, 1004−1012. (41) Tang, Y.; Zeng, X. Electrochemical Oxidation of Hydrogen in Bis(Trifluoromethylsulfonyl)Imide Ionic Liquids under Anaerobic and Aerobic Conditions. J. Phys. Chem. C 2016, 120, 23542−23551. (42) Wang, Z.; Kumar, A.; Sevilla, M. D.; Zeng, X. Anaerobic Oxidation of Methane to Methyl Radical in Ntf2 Based Ionic Liquids. J. Phys. Chem. C 2016, 120, 13466−13473. (43) Montiel, M. A.; Solla-Gullón, J.; Sánchez-Sánchez, C. M. Electrochemical Reactivity and Stability of Platinum Nanoparticles in Imidazolium-Based Ionic Liquids. J. Solid State Electrochem. 2016, 20, 1043−1052. (44) Tang, Y.; He, J.; Gao, X.; Yang, T.; Zeng, X. Continuous Amperometric Hydrogen Gas Sensing in Ionic Liquids. Analyst 2018, 143, 4136−4146. (45) Khan, A.; Gunawan, C. A.; Zhao, C. Oxygen Reduction Reaction in Ionic Liquids: Fundamentals and Applications in Energy and Sensors. ACS Sustainable Chem. Eng. 2017, 5, 3698−3715. (46) Freemantle, M. An Introduction to Ionic Liquids; Royal Society of Chemistry, 2010. (47) Stonehart, P. Interactions between adsorbed carbon monoxide and water on platinum. Electrochim. Acta 1973, 18, 63−68. (48) Niazi, A. A.; Rabideau, B. D.; Ismail, A. E. Effects of Water Concentration on the Structural and Diffusion Properties of Imidazolium-Based Ionic Liquid-Water Mixtures. J. Phys. Chem. B 2013, 117, 1378−1388. (49) Liu, X.; Chen, X.; Xu, Y.; Chen, T.; Zeng, X. Effects of Water on Ionic Liquid Electrochemical Microsensor for Oxygen Sensing. Sens. Actuators, B 2019, 285, 350−357. I

DOI: 10.1021/acs.jpcb.8b11602 J. Phys. Chem. B XXXX, XXX, XXX−XXX