61st Annual Report on Research 2016

The goal of this project is to understand and design ionic liquid systems to carry out electrochemical reduction of CO₂ to valuable hydrocarbon commodities. Ionic liquids (IL) are solutions of pure ions, usually made from large, organic cations paired either with similarly large, organic anions or small inorganic anions. These mixtures of ions remain liquid at, and below, room temperatures due to the steric repulsions between bulky ions, and the result is a conductive, low-vapor pressure fluid with intriguing bulk phase and interfacial phase properties.

Our work in this area has begun by investigating two major questions regarding ILs: 1) how do ILs behave near (electrode) surfaces, and 2) what effects does absorbed water have on the performance and behaviour of IL materials. In our first study, A series of IL films supported on three different solid surfaces were examined using spectroscopic methods. We found long-ranging ordered structures in the liquid phase that developed over periods ~ 30 minutes - 120 minutes. We characterized the systems by recording systematic changes in Infrared absorption profiles and second-harmonic generation measurements, as reported in our Langmuir paper. The ordering occurs at significant distances from the surface (up to several microns) and appears to be independent of substrate material, cation structure, overlying vapor phase, and the presence of water impurities. We examined a series of IL fluids that varied the structure of the cation, but maintained a constant anion, bis(trifluoromethyl)sulfonylimide). We found that all of the liquids tested exhibited the ordering behavior, with an ordering time that correlated with individual fluids’ bulk viscosities.

These results are important because they describe a highly-ordered interfacial phase of matter that extends orders of magnitude farther from the solid surface than has been previously reported for any liquid system. While it is widely accepted that some degree of molecular ordering and phase transition occurs in fluids near surfaces and in fluids under shear, observation of an organized fluid layer extending some microns from a solid surface is, to our knowledge, unprecedented until now. Ultimately, we propose that the very strong intermolecular interactions of the ILs contribute to their ordering to such a long distance. We also propose that the relatively high viscosity of IL slows the rate of reorientation, making the adoption of the locally preferred orientation significantly slower (10’s of minutes) than what is seen for surface induced (re)organizations of molecular liquids (picoseconds). This may explain why the effect has not been seen before.

The second area we have invested in the sorption of water into IL media. The incorporation of water into ILs is nearly impossible to prevent, and water has a significant effect on nearly all the practical applications of ILs, especially in the electrochemical reduction of carbon dioxide. However, the presence of water in ILs and its effect on basic physicochemical properties is hardly studied. In order to better understand and to be able to predict the role of water in ILs, we conducted a series of experiments to establish a range of these effects for two model ionic liquids: ethylammonium nitrate (EAN), and butyltrimethylammonium bis(trifluoromethylsulfonyl)imide (N1114 TFSI). EAN represents a ‘protic’ IL, which as a class are more hydroscopic than the ‘aprotic’ ILs, such as N1114 TFSI. We exposed these ILs to controlled dry and humid environments, and reported (in ACS Omega) the water sorption rates for these liquids (270 ± 30 ppm/min for EAN and 30 ± 3 ppm/min for N1114 TFSI), the electrochemical response to the water, as well as the infrared spectra for the dry and wet fluids. We took care to report accuracy and precision associated with common methods for reporting water content as these are highly variable across the field.

The interactions of water vary drastically in different ILs. This is clearly indicated by their water absorption rates (listed above) and by IR peak profiles for the O-H stretches of water. EAN shows a broad symmetric stretching peak indicative of a hydrogen bonding network which incorporates water clusters, whereas N1114 TFSI shows two well-resolved peaks characteristic of water monomers. Voltammetry data indicate that water is likely ‘condensing’ on the silver electrode for concentrations > 3500 ppm water. The electrical current due to silver electrode oxide reduction at -1.1 V vs Fc/Fc⁺ in N1114 TFSI correlates directly with the water concentration (i.e. a higher [water] gives a larger oxide reduction current). These results confirm that adventitious water strongly affects electrode stability, and that the concentration of water might be probed directly in-situ, by monitoring the associated oxide peak current.

Ultimately, this study shows that water must be closely monitored and controlled. Without proper experimental control, meaningful comparisons of data from IL experiments from different research groups or institutions will be difficult. This work also leads into an important area of work for the CO2 reduction project in understanding the chemical environment of water in the IL. The distinct OH vibrational profiles of water in the protic vs aprotic ILs will direct our upcoming studies on proton transport in ILs to support proton coupled electron transfer reactions of carbon dioxide.

Our results add significant new information to the understanding of IL interfaces, molecular orientation in materials, and IL materials’ behaviors. These data encourage additional studies to find if long-range ordering may be observed in a broader range of ILs as a possible avenue for forming regular 3-D molecular networks.