Structural Transitions at Ionic Liquids Interfaces

Structural Transitions at Ionic Liquids Interfaces

Benjamin Rotenberg, Mathieu Salanne Sorbonne Universités, UPMC Univ Paris 06, CNRS, Laboratoire PHENIX, F-75005, Paris, France Réseau sur le Stockage Electrochimique de l’Energie (RS2E), FR CNRS 3459, France Maison de la Simulation, USR 3441, CEA - CNRS - INRIA - Université Paris-Sud -Université de Versailles, F-91191 Gif-sur-Yvette, France
Abstract

Recent advances in experimental and computational techniques have allowed for an accurate description of the adsorption of ionic liquids on metallic electrodes. It is now well established that they adopt a multi-layered structure, and that the composition of the layers changes with the potential of the electrode. In some cases, potential-driven ordering transitions in the first adsorbed layer have been observed in experiments probing the interface on the molecular scale or by molecular simulations. This perspective gives an overview of the current understanding of such transitions and of their potential impact on the physical and (electro)chemical processes at the interface. In particular, peaks in the differential capacitance, slow dynamics at the interface and changes in the reactivity have been reported in electrochemical studies. Interfaces between ionic liquids and metallic electrodes are also highly relevant for their friction properties, the voltage-dependence of which opens the way to exciting applications.

Solid-liquid interfaces play key role in many processes, such as catalysis or electrochemical reactions, to mention only chemistry and energy related applications. Despite their importance, our understanding of the molecular-scale structure of such interfaces, where all the essential (electro)chemical processes occur, has long remained limited compared to the case of the corresponding pure solid and liquid phases. Probing directly the interface in experiments is indeed particularly challenging and computer simulations are also more involved due to the symmetry breaking in the direction perpendicular to the interface, which hinders the efficient use of periodic boundary conditions. Indeed, although simulating a few tens of water molecules may be sufficient to investigate the bulk properties of the liquid Lin et al. (2009), a similar number of molecules results in finite-size effects different from that, physically relevant, induced by the presence of the interface.

The past 10 years have witnessed the development of many experimental techniques which are sensitive to molecular arrangements at the interface, such as Scanning Tunneling Microscopy (STM), Sum-Frequency Generation (SFG), Atomic Force Microscopy (AFM), high-energy X-ray reflectivity (XR) or Surface Force Apparatus (SFA). They probe the structure of the liquid via different observables (vibrations, electron density, resistance to shear, etc), thus providing complementary views of the interface: For example the SFG signal is dominated by the innermost adsorbed layer Baldelli (2008); Peñalber and Baldelli (2012); Baldelli (2013), while AFM or XR studies probe several layers of fluid. Atkin and Warr (2007); Hayes et al. (2015); Mezger et al. (2008, 2015) In parallel, the access to high performance computers and the development of new algorithms VandeVondele et al. (2005); Reed et al. (2007) also allowed to simulate more accurately solid-liquid interfaces, shedding a new light on interfacial processes such as adsorption. For example, the combination of STM and Density Functional Theory (DFT) calculations demonstrated that water molecules adsorbed at metal surfaces exhibit a surprisingly rich variety of structures Carrasco et al. (2012). Their arrangement depends on the interplay between the geometry and energetics of the water-metal interaction and of the hydrogen bonding between the water molecules, which varies strongly from one metal to another (and even from one crystal plane of a given metal to another) and with the water coverage of the surface. This structuring impacts the dynamics at the interface Limmer et al. (2013) and ultimately the kinetics of electrochemical processes. Due to the range of length and time scales involved, from the electron transfer event to the local rearrangements of the interfacial fluid, a full understanding of the water-splitting mechanisms from computer simulations will therefore require bridging the gap between ab initio Cheng and Sprik (2012); Nielsen et al. (2015) and classical Willard et al. (2013) approaches.

Here we will focus on a particular class of electrolytes, namely room-temperature ionic liquids (RTILs). They are increasingly used in electrochemistry, with applications ranging from energy storage (batteries, supercapacitors) to electrodeposition Armand et al. (2009). Since they are made of ions, their interfacial properties have long been interpreted following the Gouy-Chapman-Stern theory. However, many of the underlying asumptions are not valid due the very high density of ions – an extreme case considering the absence of solvent in these liquids Kornyshev (2007); Perkin et al. (2013); Lee et al. (2015). A significant number of experiments and molecular simulations have thus been devoted to the study of the interfaces of ionic liquids with a solid Fedorov and Kornyshev (2014); Hayes et al. (2015). The main conclusion arising from XR, AFM, SFA and molecular dynamics (MD) is that the structure perpendicular to the interface is characterized by a strong layering of the liquid Hayes et al. (2015); Mezger et al. (2008); Perkin (2012); Merlet et al. (2013), as expected for a molecular liquid, which extends up to a few nanometres. The local composition of the layers mostly depends on the surface charge of the solid Ivanistsev et al. (2014) and displays strong local correlations due to charge-ordering.

Many recent studies on interfaces of RTILs reported intriguing results, highlighting the role of the molecular structure within the adsorbed layers. As pointed out in an editorial by Kornyshev and Qiao, it is indeed necessary to account for the three-dimensionality of the interface Kornyshev and Qiao (2014). In particular, the formation of an ordered layer of ions has been reported at the interfaces of 1-butyl-3-methylimidazolium-hexafluorophosphate ([Cmim][PF]) with mica Liu et al. (2006) or with vapor Jeon et al. (2012). At electrochemical interfaces, the contact between RTILs with an electrified metal opens the way to voltage-induced ordering transitions within the adsorbed liquid. The universality of such transitions is far from being established, in particular the extent of concerned RTILs-substrate combinations should be clarified. A first objective of this perspective article is therefore to summarize the studies, both experimental and theoretical, in which such transitions have been observed. We then discuss the impact of this finding on the physico-chemical properties of the interface. In particular, the following questions will be addressed: How can we detect structural transitions in experiments and in simulations? Is there a templating action from the solid? What is the main electrochemical signature of these transitions? Is there an impact on the friction properties of the interface? Some of these questions remain open and call for further studies.

I Evidences for structural transitions at ionic liquid interfaces

i.1 Experimental studies

To our knowledge, the first studies dealing with interfacial phase transitions in Coulomb fluids were conducted by Freyland et al. Freyland (2008). Their in situ STM study of the interface between the [Cmim][PF] and the (111) face of gold reported the formation of Moiré-like patterns at potentials greater than -0.2 V with respect to a platinum reference electrode. These were attributed to the formation of an ordered adlayer of PF. At negative potentials, the STM images were consistent with the formation a layer of anions with the () structure, indicating a two-dimensional ordering transition at this interface. It is worth noting that these observations closely follow a previous work performed on the adsorption of iodine from aqueous solutions on similar gold surfaces Tao and Lindsay (1992). A further study by the same authors on the electrodeposition of Cd on Au(111) in a chloroaluminate ionic liquid has also revealed the formation of an ordered AlCl adlayer Pan and Freyland (2007).

Figure 1: a) High-resolution STM image (5.4 nm7.5 nm) of the BMP adlayer on Au(111) at 1.4 V and b) proposed structural model, showing two domains of the () BMP adlayer separated by a translational domain boundary (indicated by dashed line). Reproduced with permission from reference 30; Copyright: Wiley, 2015.

When changing both the nature of the anion (from PF to BF) and of the surface of gold in contact with the RTIL (from (111) to (100)), Su et al. have also evidenced the existence of potential-driven ordering transitions Su et al. (2009). Increasing the potential from 0.3 V, an ordered layer of anions is formed between 0.1 and 0.4 V. On the contrary, when scanning in the negative potentials direction, they first observed a loose film-like layer which was attributed to a disordered adsorption of Cmim cations. Then, for potentials lower than 0.95 V, perpendicularly oriented double-row strips were observed. These strips were assigned to the formation of micelle-like arrangements of aligned Cmim cations. These structures also formed with PF and SOCF anions, but not on (111) surfaces of gold Su et al. (2009), which shows that in this case a structural commensurability of the adsorbed layer and the metal surface is necessary for the formation of ordered structures.

This conclusion was confirmed in a study using in situ video-STM to probe the (111) Au interface with a RTIL composed of a different cation, namely 1-butyl-1-methylpyrrolidinium (BMP) associated with the bis(trifluoromethylsulfonyl)imide anion (TFSIWen et al. (2015). Stable images could only be obtained for negative potentials below 1 V. For such potentials the images showed the formation of ordered structures. Several distinct arrangements of cations were proposed in order to interpret the observations at various potentials; one of them is shown in 1. In both structures proposed by Wen et al., the cation rings are adsorbed on the surface; in contrast, the alkyl chains lie flat on the surface only for the lower charge density (hence lateral cation density in the adsorbed fluid) and extend into the perpendicular direction for the higher density. The lattice parameters for the adlayer superstructure decrease accordingly and may change symmetry, resulting for the densest packing in a square lattice which differs from the hexagonal substrate. Finally, the video-STM furter allowed the first direct observation of the dynamical evolution of the adsorbed liquid. In particular, it was found that the fluctuations occur mainly at the boundaries between ordered domains.

All the structural transitions observed with STM have so far involved Au electrodes because this metal can be produced as single crystals with well-defined surfaces. Elbourne et al. have recently used another technique, in situ amplitude-modulated AFM, to study interfaces of highly ordered pyrolytic graphite (HOPG) instead Elbourne et al. (2015). This substrate presents the advantage of having flat surfaces with high area, and to avoid the surface reconstructions or etching which can occur in the case of gold Su et al. (2009). These authors studied the effect of applied potential on the adsorbed layer structure for a [Cmim][TFSI] ionic liquid. At the open-circuit potential, well-defined rows are present on the surface. Unlike previous works, in which the ordered structures were apparently formed of only one type of ions, the unit cell is composed of an anion-cation-cation-anion arrangement Elbourne et al. (2015). This structure changes markedly with surface potential or when relatively low concentrations of lithium or chloride ions are present in the RTIL.

The variety of systems in which transitions are observed clearly show that it is a common feature of metal/ionic liquids interfaces. However, the few works reported so far raise very interesting questions. In particular, all the ordered structures proposed so far to interpret the STM data are composed of a single species only. However, the relatively small applied potentials which are used (1 V) may not be sufficent to fully separate cations from anions. In the case of the AFM study, the anion-cation-cation-anion rows result in an overall neutral layer but there is a strong charge imbalance on the nanometre scale. A few hypotheses can thus be proposed to explain the observations: i) there may be a specific absorption of the ions on gold with the formation of partially covalent bonds Anderson et al. (2014) ii) on top of the observed layer, there could be an oppositely charged layer of ions which is not observed by the experiment and iii) it remains possible that the proposed structures, which are only based on the relative size of the ions (keeping the possibility of some kind of conformational ordering, for example only the imidazolium rings of the cations would lie parallel to the surface), are not the correct ones. Using additional techniques such as SFG, which is sensitive to the orientation of the ions Baldelli (2013), could possibly shed a complementary light on this issue. Another open question is whether commensurability between the adsorbed layer and the metal substrate is necessary to observe a transition. Here also, the recent study performed with HOPG electrodes Elbourne et al. (2015) suggests that it is not the case, and that the electric fluctuations at an homogeneous and flat metallic surface are sufficient to trigger ordering transitions in the interfacial layer of RTIL. In addition, the use of carbon electrodes in this study demonstrates that a perfect metallic behavior is not necessary to induce such transitions.

i.2 Computer simulations

Figure 2: Snapshots of typical ordered structures observed in computer simulations. Top left: First adsorbed layer of a LiCl molten salt electrolyte on the (100) surface of an aluminum electrode at a negative potential. Adapted from reference 34. Top right: First adsorbed layer of a [Cmim][PF] RTIL on a graphite electrode at a neutral potential. Adapted from reference 35. Bottom: First two adsorbed layer of a simplified RTIL on a charged Lennard-Jones wall with a large negative surface charge density. Reprinted from reference 36, Copyright 2013, with permission from Elsevier.

In order to simulate electrochemical systems, it is necessary to fix the potential of the electrode. In classical molecular dynamics or Monte-Carlo, this can be done by various methods Merlet et al. (2013). Our approach consists in treating the partial charges carried by the electrode atoms as additional degrees of freedom which fluctuate during the simulation. Their values are determined at each time step from a self-consistent calculation Siepmann and Sprik (1995); Reed et al. (2007); Merlet et al. (2013). In such simulations, the electrochemical cell consists in a wide slab of electrolyte held between two electrodes with different voltages. Like in experiments, the potentials are not absolute. The only fixed quantity is the potential difference between the two electrodes , although it is also possible to calculate the potential of each electrode with respect to the bulk liquid in the case of flat electrodes; we will note this potential . In the following, we will assimilate an ordering transition to an abrupt change in the structure observed when changing the potential. However, it is worth noting that first-order transitions are associated with a discontinuity in an order parameter and a corresponding singularity in a partition function, which are not easy to prove in simulations Chandler (1987). This point will be further discussed in the next section.

Using this simulation approach, a first example of voltage-driven transition was reported for a rather exotic system, formed with a high temperature molten salt (LiCl) and an aluminum electrode with its (100) surface in contact with the liquid Pounds et al. (2009); Tazi et al. (2010). An advantage of this system is that a polarizable force field could be built directly from accurate DFT calculations using a generalized force-matching approach Pounds et al. (2009).For potential drops across the interface more negative than 1.76 V, which corresponds to the point of zero charge (PZC), the molten salt adopted a disordered structure at the interface, while for larger potentials an ordered structure was obtained Pounds et al. (2009). This structure, which is shown on the top panel of 2, was commensurate with the aluminum substrate, and a strong alignment of the dipole components of the chloride anion and the normal of the surface was observed for large potentials Tazi et al. (2010). Interestingly, no transition was observed when the plane of the metal was changed to (110) instead of (100), but a different ordered structure was then obtained showing that an epitaxial mechanism is at play, whereby the molten salt adapts it structure to that of the electrode surface.

In a recent work, Kirchner et al. studied interfaces between primitive models of ionic liquids and solid surfaces with various net charges (i.e. the electric potential was not controlled) Kirchner et al. (2013). At certain charge densities (  C cm) the structure of the adsorbed layer of cations undergoes a structural transition to a surface-frozen monolayer of densely packed counter-ions with a Moiré-like structure. At lower surface charge densities (i.e. lower than  C cm), they even observed the formation of an herring-bone structure arising from the superposition of two ordered monolayers of ions (see the bottom panel of 2). These findings provide an interesting support for the STM studies discussed above, but it is worth noting that the charge densities employed are somewhat larger than the experimental ones – they would correspond to potentials which are above the electrochemical window of typical RTILs.

Going towards more realistic models, an ordering transition was reported from molecular simulations for the interface between [Cmim][PF] (for which a coarse-grained force field was used) and an electrified surface of graphite Merlet et al. (2014). The presence of the ordered structure could be monitored by computing the in-plane structure factor in the first layer of the adsorbed liquid. This structure factor was liquid-like on a wide range of potentials, but it showed some strong Bragg-like peaks suggesting a 2-dimensional lattice-like organization for both the anions and cations, which is shown in the top-right panel of 2. This ordered structure contained on average as many anions as cations, and it was also observed by Kislenko et al. in simulations of the same RTIL (with an all-atom model) adsorbed on an uncharged surface of graphite Kislenko et al. (2009). By using importance sampling techniques, it could be shown by Merlet et al. that this structure was the most stable one for small positive potentials ( V) and metastable for small negative electrode potentials. It is worth noting that similarly to the experimental work of Elbourne et al. involving HOPG electrodes, no commensurability with the electrode surface seems necessary to observe such ordered structures.

So far, no ordering transitions have been observed using more elaborate, all-atom models of RTILs in contact with electrodes at fixed potential. In particular, the adsorption of [Cmim][PF] and [Cmim][BF] on electrified surfaces of gold was studied by Hu et al.Hu et al. (2013) but they did not report any ordering transition similar to the experimental observations by STM.

These simulation results, while confirming the possibility of transitions in the adsorbed layer of the fluid, also open their share of questions. Future works will need to address the issue of finite-size effects, since there must be a commensurability between the formed ordered structure and the simulation cell. Timescales are important too, since metastable states may be much longer-lived than the typical simulation times, which are on the order of the nanosecond only due to the computational cost. There is therefore a possibility that the reported transitions are artefacts of the simulation setups, but the similarities with experimental findings seem to weaken this hypothesis. The question of specific interactions with surfaces such as gold will also have to be treated. This requires in turn the development of accurate force fields for this purpose. First steps in this direction have recently been made in the case of carbon materials Pensado et al. (2014).

Ii Impact of the transitions on physico-chemical properties

We now turn to the consequences of structural transitions within the adsorbed fluid on the physico-chemical properties of the interface. Specifically, we discuss the impact of voltage-induced transitions on the electrochemical response of the electrode-RTIL interface in terms of differential capacitance, cyclic voltammograms and electrochemical reaction, as well as on the mechanical response (solid-liquid friction).

ii.1 Peaks in the differential capacitance

The differential capacitance measures the response of the average surface charge density to changes in the voltage :

(1)

By definition, a capacitor corresponds to a voltage-independent differential capacitance. However, the charge of the electrode reflects the composition and the charge distribution within the interfacial liquid. As a result, one should expect a signature of abrupt structural changes at voltages corresponding to putative phase transitions in the electrode charge, hence peaks in the corresponding differential capacitance. While experimentally such peaks have indeed been observed Su et al. (2009); Cannes et al. (2013); Costa et al. (2015), their possible link with changes in the structure or the interface has been difficult to demonstrate until recently, due to the experimental challenges of in situ imaging and the occurence of other processes such as surface reconstruction of the electrode.

Indeed, in their STM study of [Cmim][BF] on a (100) gold electrode, Su et al. have also measured the capacitance of the interface. They observed a fivefold increase in this quantity in the potential region of transition from anion adsorption to cation adsorption. In the case of the in situ video-STM study of the [BMP][TFSA] on a (111) gold electrode Wen et al. (2015), thanks to the high temporal resolution or the video-STM technique, the authors were able to visualize the evolution of the interfacial fluid during cyclic voltammetry (CV) experiments. The cyclic voltammogram displays two current peaks associated with two surface transitions which could also be linked to the formation of the ordered cationic structures, such as the one shown on 1, upon increasingly negative surface charge density.

Figure 3: (a) Calculated probability distribution of the charge density of graphite electrodes in contact with the [Cmim][PF] ionic liquid with respect to the applied potential . The two-dimensional graph of the distribution employs a logarithmic scale with lines separated by a difference of 0.5 and is plotted as a function of in (b). Note the fat tails in the distribution, and the markedly nonlinear shifts with changing voltage. (c) Differential capacitance, , as a function of . Adapted from reference 35.

In computer simulations, it is relatively straightforward to calculate the capacitance of the interface in constant potential simulations. The generic method consists in simulating an electrochemical cell at various voltage and extracting the average surface charge. Then the plot is differentiated, which provides the differential capacitance through 1. However, close to a transition, a large peak in the capacitance is expected, so that many voltages should in principle be sampled in this region. An alternative was recently proposed, which consists in using importance sampling methods Limmer et al. (2013); Merlet et al. (2014). In short, by using the whole distribution of surface charges during the simulations, it is possible to sample the probability distributions of any variable as continuous functions of the applied potential. There is in principle no need to acquire more data close to the transition, the only requisite is to have a good overlap between the histograms of surface charges from the various voltages.

The probability distribution of the charge density of graphite electrodes in contact with the [Cmim][PF] ionic liquid Merlet et al. (2014) obtained with this approach is shown on 3(a). The figure shows the probability distribution on a logarithmic scale. It is clear that there are three branches along which the distribution of the surface charge distribution shifts almost linearly upon increasing the potential. These branches are separated by more complex changes in the distribution around particular voltages. We will focus on the one occuring at  = 0.9 V since this is the potential for which the order-disorder transition discussed above occurs. 3(b) shows the distribution at three applied voltages (0.8 V, 0.9 V and 1.0 V). They are characteristic of a first-order phase transition. Away from phase coexistence (at 0.8 V and 1.0 V), they show the presence of “fat tails”, which are due to the small probability of obtaining the metastable phase. At the transition the distribution displays hints of bimodality, which is expected if the two phases are equiprobable. However it would be necessary to simulate larger systems to fully conclude on this point.

The differential capacitance computed from these simulations is shown on 3(c). Note that much better statistics could again be obtained compared to the usual method involving 1 by using the Johnson-Nyquist relation,

(2)

where is the surface of the electrode and are the fluctuations in the electrode surface charge density. A large peak in the differential capacitance is observed at the applied voltage where the transition occurs, which is consistent with the experimental findings of Su et al. Note that again, larger systems should lead to a singular charge-density transition in a macroscopic limit Chandler (1987). Our simulations therefore confirm that the presence of large peaks in experimental measures of the capacitance of an interface can indeed be the signature that a potential-driven transition is occuring.

ii.2 Hysteresis and slow dynamics

The above-mentioned domain boundaries between phases also have important implications by themselves, due to the entailed free energy cost. In three dimensions, this would be a surface free energy. In the present case the topology of the boundary between interfacial domains at the surface of the electrode remains to be clarified Kornyshev and Qiao (2014). As a result, annealing these boundaries, either between grains of otherwise identical domains, or between different domains, requires overcoming the corresponding free energy barrier. In practice, the consequences of these barriers are observable as long time scales in the dynamics of the interface or as hysteresis in cyclic voltammetry.

Uysal et al. reported a potential-dependent hysteresis at an electrified graphene/RTIL interface Uysal et al. (2014). X-ray reflectivity measurements during cyclic voltammetry and potential step measurements are used to probe the electronic density in the direction perpendicular to an epitaxial graphene surface, within the adsorbed [Cmim][TFSI] ionic liquid. The resulting profiles were consistent with that obtained from MD simulations, by assuming a combination of two limiting structures, with weights varying as a function of applied voltage. The structure evolves very slowly after a potential step, with processes occuring over time scales exceeding 10 s. In addition, the CV scans exhibit significant (scan rate dependent) hysteresis. While in this work the authors safely indicated that the nature of the apparent barrier and the associated mechanism require further investigation, these observations clearly point to the crucial role of structural transitions and the associated domain boundaries in the observed hysteresis and slow dynamics. Another manifestation of slow processes occuring at the ionic liquid / electrode interface was reported by Roling et al., who have carefully analysed the capacitance spectra on a broad range of frequencies Roling et al. (2012); Druschler et al. (2012). Although no particular ordering transition was observed by STM, these authors concluded that the slower capacitive process could be related to structural reorganisations of the gold surface or to strong rearrangements in the first adsorbed layer of ions.

Recently, Limmer proposed a detailed study of these effects using a coarse-grained model capturing strong inter-ionic correlations Limmer (2015). Its limited computational cost compared to molecular simulations allowed for a systematic finite-size scaling analysis, which demonstrated the first-order nature of the fluctuation-induced transition and spontaneous charge density ordering at the interface, in the presence of an otherwise disordered bulk solution, already observed in molecular simulations Merlet et al. (2014). A crucial step in this demonstration is the extensive growth of the free energy barrier between phases analogous to the ones observed in reference 47, which indeed implies hysteresis and long time scales.

ii.3 Impact on reactivity

Structural changes in the ionic liquid at the interface also have implications on the local environment of other species in the liquid, in particular electro-active species. This in turn may result in changes in their reactivity. A direct observation of this feature has recently been reported by Garcia-Rey and Dlott, who studied CO reduction on a polycrystalline Ag electrode, with 1-ethyl-3-methylimidazolium tetrafluoroborate [Cmim][BF] containing 0.3 mol% water as electrolyte Garcia Rey and Dlott (2015). Such systems have been shown to reduced the overpotential for CO reduction. SFG and IR were used to probe the surface field experienced by the adsorbed CO molecules produced by the electrochemical reduction of CO. From the CO Stark shift, a sudden increase of the field at the electrode surface was observed at the threshold potential for CO reduction, which could be traced back to a structural transition within the RTIL – even though no information could be obtained on the nature of these structural changes. Nevertheless, this study illustrates the potential benefit of exploiting the peculiar structure of ionic liquid interfaces and the voltage-driven changes thereof (with potentially much greater diversity than in solvent-based electrolytes) for electrochemical reactions.

ii.4 Voltage-dependent friction

Figure 4: Friction forces as functions of load and applied potential recorded for a sharp AFM tip sliding on Au(111) in [BMP[FAP]. Each data point represents the average over a full scan frame of 14 nm side length. The potential is switched between 1.5 and 0.7 V for each value of the applied normal load. Reprinted with permission from reference 52; Copyright 2012 by the American Physical Society.

Finally, voltage-driven changes in the structure and composition of the interfacial fluid also have implications from the dynamical point of view. Sweeney et al. conducted nanotribology experiments to probe the lubrication properties of 1-butyl-1-methylpyrrolidinium tris(pentafluoroethyl) trifluorophosphate ([BMP][FAP]) confined between silica colloid probes or sharp silica tips and a Au(111) substrate, using AFM Sweeney et al. (2012). As the composition of the adsorbed layers is tuned by the electrode potential, from cation-enriched to anion-enriched, the friction also evolves. 4 illustrates that these variations are directly linked to the nature of the sliding plane, which may correspond to cation or anion layers, depending on the electrode potential and on the normal load exerted on the confined fluid. While the voltage-driven structural changes on the microscopic interfacial structures remain to be investigated, such studies open the way to a new tuning of frictional forces at the molecular scale without changing the substrate.

More detailed information on the role of key microscopic and macroscopic factors can be obtained using molecular simulations, such as load, shear velocity, surface topology and length of alkyl side chains in the ionic liquid Mendonça et al. (2013). Simulations with fixed surface charge density (instead of potential) have further evidenced two mechanisms underlying friction changes in such systems, namely charge effects on normal and in-plane ordering in the film, as well as swapping between anion and cation layers at the surface Fajardo et al. (2015).

Iii Summary and outlook

There is now a large body of experiments pointing towards the existence of potential-driven transitions at the interface between ionic liquids and metallic electrodes. However, as discussed above, the exact structure and composition of the ordered phases remain open questions. Computer simulations bring some theoretical support on the question, but they are still scarce because of the technical difficulty associated with the use of constant applied potential ensemble. They also suffer from sampling issues (both in size and time) which render the observation and the characterization of the transitions difficult. The current works, in which the interactions are determined using classical force fields may also be limited if particular bonding occurs at the interface. The recent inclusion of constant voltage methods in DFT-based molecular dynamics packages Golze et al. (2013) may open new opportunities for tackling this difficult problem, as it was shown recently in the context of nanotribology. Page et al. (2014); Li et al. (2015)

In addition, the role of many parameters remain to be established. For example, how do the composition of RTIL and the possible presence of impurities affect the occurence of ordering transitions? What is the impact of the temperature of the systems? Also many applications of RTILs use them in the presence of a solvent, which will also impact the whole structure of the electric double layer. Finally, although it is clear that the nature of the substrate plays a strong role, it is not sure that there is always a commensurability between the ordered structure of the liquid and the metal. Topological defects at the surface of the metal may also play a predominant role, Black et al. (2015) and it is likely that corrugation effects can modify the formation and/or the detection of ordered layers: Recent simulations have shown that the heterogeneous nucleation of ice at a surface depended markedly on the morphology of the latter. Fitzner et al. (2015) Additional works with varying metal electrodes will allow to better understand these issues.

Whether these transitions will have practical applications remains an open question, but they clearly impact a lot the physico-chemical properties: peaks in the differential capacitances, slow dynamics at the interface, varying reactivity and voltage-dependent friction properties have already been reported. Overall such transitions reinforce the view of RTILs as solvents with multifaceted properties, with a composition that can be specifically tailored to a given task.

References

  • Lin et al. (2009) Lin, I.-C.; Seitsonen, A. P.; Coutinho-Neto, M. D.; Tavernelli, I.; Rothlisberger, U. Importance of van der Waals Interactions in Liquid Water. J. Phys. Chem. B 2009, 113, 1127–1131.
  • Baldelli (2008) Baldelli, S. Surface Structure at the Ionic Liquid-Electrified Metal Interface. Acc. Chem. Res. 2008, 41, 421–431.
  • Peñalber and Baldelli (2012) Peñalber, C. Y.; Baldelli, S. Observation of Charge Inversion of an Ionic Liquid at the Solid Salt-Liquid Interface by Sum Frequency Generation Spectroscopy. J. Phys. Chem. Lett. 2012, 3, 844–847.
  • Baldelli (2013) Baldelli, S. Interfacial Structure of Room-Temperature Ionic Liquids at the Solid-Liquid Interface as Probed by Sum Frequency Generation Spectroscopy. J. Phys. Chem. Lett. 2013, 4, 244–252.
  • Atkin and Warr (2007) Atkin, R.; Warr, G. G. Structure in Confined Room-Temperature Ionic Liquids. J. Phys. Chem. C 2007, 111, 5162–5168.
  • Hayes et al. (2015) Hayes, R.; Warr, G. G.; Atkin, R. Structure and Nanostructure in Ionic Liquids. Chem. Rev. 2015, 115, 6357–6426.
  • Mezger et al. (2008) Mezger, M.; Schröder, H.; Reichert, H.; Schramm, S.; Okasinski, J. S.; Schröder, S.; Honkimäki, V.; Deutsch, M.; Ocko, B. M.; Ralston, J.; et al. Molecular Layering of Fluorinated Ionic Liquids at a Charged Sapphire (0001) Surface. Science 2008, 322, 424–428.
  • Mezger et al. (2015) Mezger, M.; Roth, R.; Schröder, H.; Reichert, P.; Pontoni, D.; Reichert, H. Solid-Liquid Interfaces of Ionic Liquid Solutions – Interfacial Layering and Bulk Correlations. J. Chem. Phys. 2015, 124, 164707.
  • VandeVondele et al. (2005) VandeVondele, J.; Krack, M.; Mohamed, F.; Parrinello, M.; Chassaing, T.; Hutter, J. QUICKSTEP: Fast and Accurate Density Functional Calculations Using a Mixed Gaussian and Plane Waves Approach. Comp. Phys. Commun. 2005, 167, 103–128.
  • Reed et al. (2007) Reed, S. K.; Lanning, O. J.; Madden, P. A. Electrochemical Interface Between an Ionic Liquid and a Model Metallic Electrode. J. Chem. Phys. 2007, 126, 084704.
  • Carrasco et al. (2012) Carrasco, J.; Hodgson, A.; Michaelides, A. A Molecular Perspective of Water at Metal Interfaces. Nat. Mater. 2012, 11, 667–674.
  • Limmer et al. (2013) Limmer, D. T.; Willard, A. P.; Madden, P.; Chandler, D. Hydration of Metal Surfaces Can Be Dynamically Heterogeneous and Hydrophobic. Proc. Natl. Acad. Sci. U.S.A. 2013, 110, 4200–4205.
  • Cheng and Sprik (2012) Cheng, J.; Sprik, M. Alignment of Electronic Energy Levels at Electrochemical Interfaces. Phys. Chem. Chem. Phys. 2012, 14, 11245–11267.
  • Nielsen et al. (2015) Nielsen, M.; Björketun, M. E.; Hansen, M. H.; Rossmeisl, J. Towards First Principles Modeling of Electrochemical Electrode-Electrolyte Interfaces. Surf. Sci. 2015, 631, 2–7.
  • Willard et al. (2013) Willard, A. P.; Limmer, D. T.; Madden, P. A.; Chandler, D. Characterizing Heterogeneous Dynamics at Hydrated Electrode Surfaces J. Chem. Phys. 2013, 138, 184702.
  • Armand et al. (2009) Armand, M.; Endres, F.; MacFarlane, D. R.; Ohno, H.; Scrosati, B. Ionic-Liquid Materials for the Electrochemical Challenges of the Future. Nat. Mater. 2009, 8, 621–629.
  • Kornyshev (2007) Kornyshev, A. A. Double-Layer in Ionic Liquids: Paradigm Change? J. Phys. Chem. B 2007, 111, 5545–5557.
  • Perkin et al. (2013) Perkin, S.; Salanne, M.; Madden, P.; Lynden-Bell, R. Is a Stern and Diffuse Layer Model Appropriate to Ionic Liquids at Surfaces? Proc. Natl. Acad. Sci. U.S.A. 2013, 110, E4121.
  • Lee et al. (2015) Lee, A. A.; Vella, D.; Perkin, S.; Goriely, A. Are Room-Temperature Ionic Liquids Dilute Electrolytes? J. Phys. Chem. Lett. 2015, 6, 159–163.
  • Fedorov and Kornyshev (2014) Fedorov, M. V.; Kornyshev, A. A. Ionic Liquids at Electrified Interfaces. Chem. Rev. 2014, 114, 2978–3036.
  • Perkin (2012) Perkin, S. Ionic Liquids in Confined Geometries. Phys. Chem. Chem. Phys. 2012, 14, 5052–5062.
  • Merlet et al. (2013) Merlet, C.; Rotenberg, B.; Madden, P. A.; Salanne, M. Computer Simulations of Ionic Liquids at Electrochemical Interfaces. Phys. Chem. Chem. Phys. 2013, 15, 15781–15792.
  • Ivanistsev et al. (2014) Ivanistsev, V.; O’Connor, S.; Fedorov, M. V. Poly(a)morphic Portrait of the Electrical Double Layer in Ionic Liquid Electrochem. Commun. 2014, 48, 61–64.
  • Kornyshev and Qiao (2014) Kornyshev, A. A.; Qiao, R. Three-Dimensional Double Layers J. Phys. Chem. C 2014, 118, 18285–18290.
  • Liu et al. (2006) Liu, Q. X.; Zein El Abedin, S.; Endres, F. Electroplating of Mild Steel by Aluminium in a First Generation Ionic Liquid: A Green Alternative to Commercial Al-Plating in Organic Solvents Surf. Coat. Technol. 2006, 201, 1352–1356.
  • Jeon et al. (2012) Jeon, Y.; Vaknin, D.; Bu, W.; Sung, J.; Ouchi, Y.; Sung, W.; Kim, D. Surface Nanocrystallization of an Ionic Liquid. Phys. Rev. Lett. 2012, 108, 055502.
  • Freyland (2008) Freyland, W. Interfacial Phase Transitions in Conducting Fluids. Phys. Chem. Chem. Phys. 2008, 10, 923–936.
  • Tao and Lindsay (1992) Tao, N. J.; Lindsay, S. M. In Situ Scanning Tunneling Microscopy Study of Iodine and Bromine Adsorption on Gold(111) under Potential Control. J. Phys. Chem. 1992, 96, 5213–5217.
  • Pan and Freyland (2007) Pan, G.-B.; Freyland, W. In Situ STM Investigation of Spinodal Decomposition and Surface Alloying During Underpotential Deposition of Cd on Au(111) from an Ionic Liquid. Phys. Chem. Chem. Phys. 2007, 9, 3286–3290.
  • Wen et al. (2015) Wen, R.; Rahn, B.; Magnussen, O. M. Potential-Dependent Adlayer Structure and Dynamics at the Ionic Liquid/Au(111) Interface: A Molecular-Scale In Situ Video-STM Study. Angew. Chem., Int. Ed. 2015, 54, 6062–6026.
  • Su et al. (2009) Su, Y.-Z.; Fu, Y.-C.; Yan, J.-W.; Chen, Z.-B.; Mao, B.-W. Double Layer of Au(100)/Ionic Liquid Interface and its Stability in Imidazolium-Based Ionic Liquids. Angew. Chem., Int. Ed. 2009, 48, 5148–5151.
  • Elbourne et al. (2015) Elbourne, A.; McDonald, S.; Voitchovsky, K.; Endres, F.; Warr, G. G.; Atkin, R. Nanostructure of the Ionic Liquid-Graphite Stern Layer. ACS Nano 2015, 9, 7608–7620.
  • Anderson et al. (2014) Anderson, E.; Grozovski, V.; Siinor, L.; Siimenson, C.; Lust, E. In Situ STM Studies of Bi(111)—1-Ethyl-3-Methylimidazolium Tetrafluoroborate + 1-Ethyl-3-Methylimidazolium Iodide Interface Electrochem. Commun. 2014, 46, 18–21.
  • Tazi et al. (2010) Tazi, S.; Salanne, M.; Simon, C.; Turq, P.; Pounds, M.; Madden, P. A. Potential-Induced Ordering Transition of the Adsorbed Layer at the Ionic Liquid / Electrified Metal Interface. J. Phys. Chem. B 2010, 114, 8453–8459.
  • Merlet et al. (2014) Merlet, C.; Limmer, D. T.; Salanne, M.; van Roij, R.; Madden, P. A.; Chandler, D.; Rotenberg, B. The Electric Double Layer Has a Life of Its Own. J. Phys. Chem. C 2014, 118, 18291–18298.
  • Kirchner et al. (2013) Kirchner, K.; Kirchner, T.; Ivanistsev, V.; Fedorov, M. V. Electrical Double Layer in Ionic Liquids: Structural Transitions from Multilayer to Monolayer Structure at the Interface. Electrochim. Acta 2013, 110, 762–771.
  • Siepmann and Sprik (1995) Siepmann, J. I.; Sprik, M. Influence of Surface-Topology and Electrostatic Potential on Water Electrode Systems. J. Chem. Phys. 1995, 102, 511–524.
  • Merlet et al. (2013) Merlet, C.; Péan, C.; Rotenberg, B.; Madden, P. A.; Simon, P.; Salanne, M. Simulating Supercapacitors: Can We Model Electrodes As Constant Charge Surfaces? J. Phys. Chem. Lett. 2013, 4, 264–268.
  • Chandler (1987) Chandler, D. Introduction to Modern Statistical Mechanics; Oxford University Press, 1987.
  • Pounds et al. (2009) Pounds, M.; Tazi, S.; Salanne, M.; Madden, P. A. Ion Adsorption at a Metallic Electrode: An Ab Initio Based Simulation Study. J. Phys.: Condens. Matter 2009, 21, 424109.
  • Kislenko et al. (2009) Kislenko, S.; Samoylov, I.; Amirov, R. Molecular Dynamics Simulation of the Electrochemical Interface Between a Graphite Surface and the Ionic Liquid BMIMPF. Phys. Chem. Chem. Phys. 2009, 11, 5584–5590.
  • Hu et al. (2013) Hu, Z.; Vatamanu, J.; Borodin, O.; Bedrov, D. A Molecular Dynamics Simulation Study of the Electric Double Layer and Capacitance of BMIMPF and BMIMBF Room Temperature Ionic Liquids Near Charged Surfaces Phys. Chem. Chem. Phys. 2013, 15, 14234–14247.
  • Pensado et al. (2014) Pensado, A. S.; Malberg, F.; Costa Gomes, M. F.; Pádua, A. A. H.; Fernández, J.; Kirchner, B. Interactions and Structure of Ionic Liquids on Graphene and Carbon Nanotubes Surfaces. RSC Adv. 2014, 4, 18017–18024.
  • Cannes et al. (2013) Cannes, C.; Cachet, H.; Debiemme-Chouvy, C.; Deslouis, C.; de Sanoit, J.; Le Naour, C.; Zinovyeva, V. A. The Double Layer at [BuMeIm][TfN] Ionic Liquid-Pt or C Materials Interfaces. J. Phys. Chem. C 2013, 117, 22915–22925.
  • Costa et al. (2015) Costa, R.; Pereira, C. M.; Silva, A. F. Structural Ordering Transitions in Ionic Liquids Mixtures. Electrochem. Commun. 2015, 57, 10–13.
  • Limmer et al. (2013) Limmer, D. T.; Merlet, C.; Salanne, M.; Chandler, D.; Madden, P. A.; van Roij, R.; Rotenberg, B. Charge Fluctuations in Nanoscale Capacitors. Phys. Rev. Lett. 2013, 111, 106102.
  • Uysal et al. (2014) Uysal, A.; Zhou, H.; Feng, G.; Lee, S. S.; Li, S.; Fenter, P.; Cummings, P. T.; Fulvio, P. F.; Dai, S.; McDonough, J. K.; Gogotsi, Y. Structural Origins of Potential Dependent Hysteresis at the Electrified Graphene/Ionic Liquid Interface. J. Phys. Chem. C 2014, 118, 569–574.
  • Roling et al. (2012) Roling, B.; Drüschler, M.; Huber, B. Slow and Fast Capacitive Process Taking Place at the Ionic Liquid/Electrode Interface. Faraday Discuss. 2012, 154, 303–311.
  • Druschler et al. (2012) Druschler, M.; Borisenko, N.; Wallauer, J.; Winter, C.; Huber, B.; Endres, F.; Roling, B. New Insights Into the Interface Between a Single-Crystalline Metal Electrode and an Extremely Pure Ionic Liquid: Slow Interfacial Processes and the Influence of Temperature on Interfacial Dynamics. Phys. Chem. Chem. Phys. 2012, 14, 5090–5099.
  • Limmer (2015) Limmer, D. Interfacial Ordering and Accompanying Unbounded Capacitance at Ionic Liquid-Metal Interfaces. Arxiv 2015, 1506.02667.
  • Garcia Rey and Dlott (2015) Garcia Rey, N.; Dlott, D. D. A Structural Transition in an Ionic Liquid Controls CO Electrochemical Reduction. J. Phys. Chem. C 2015, 119, 20892–20899.
  • Sweeney et al. (2012) Sweeney, J.; Hausen, F.; Hayes, R.; Webber, G. B.; Endres, F.; Rutland, M. W.; Bennewitz, R.; Atkin, R. Control of Nanoscale Friction on Gold in an Ionic Liquid by a Potential-Dependent Ionic Lubricant Layer. Phys. Rev. Lett. 2012, 109, 155502.
  • Mendonça et al. (2013) Mendonça, A. C. F.; Pádua, A. A. H.; Malfreyt, P. Non-Equilibrium Molecular Simulations of New Ionic Lubricants at Metallic Surfaces: Prediction of the Friction. J. Chem. Theory Comput. 2013, 9, 1600–1610.
  • Fajardo et al. (2015) Fajardo, O. Y.; Bresme, F.; Kornyshev, A. A.; Urbakh, M. Electrotunable Lubricity with Ionic Liquid Nanoscale Films. Sci. Rep. 2015, 5, 7698.
  • Golze et al. (2013) Golze, D.; Iannuzzi, M.; Nguyen, M.-T.; Passerone, D.; Hutter, J. Simulation of Adsorption Processes at Metallic Interfaces: An Image Charge Augmented QM/MM Approach. J. Chem. Theory Comput. 2013, 9, 5086–5097.
  • Page et al. (2014) Page, A. J.; Elbourne, A.; Stefanovic, R.; Addicoat, M. A.; Warr, G. G.; Voïtchovsky, K.; Atkin, T. 3-Dimensional Atomic Scale Structure of the Ionic Liquid-Graphite Interface Elucidated by AM-AFM and Quantum Chemical Simulations. Nanoscale 2014, 6, 8100–8106.
  • Li et al. (2015) Li, H.; Atkin, R.; Page, A. J. Combined Friction Force Microscopy and Quantum Chemical Investigation of the Tribotronic Response at the Propylammonium Nitrate-Graphite Interface. Phys. Chem. Chem. Phys. 2015, 17, 16047–16052.
  • Black et al. (2015) Black, J. M.; Okatan, M. B.; Feng, G.; Cummings, P. T.; Kalinin, S. V.; Balke, N. Topological Defects in Electric Double Layers of Ionic Liquids at Carbon Interfaces. Nano Energy 2015, 15, 737–745.
  • Fitzner et al. (2015) Fitzner, M.; Sosso, G. C.; Cox, S. J.; Michaelides, A. The Many Faces of Heterogeneous Ice Nucleation: Interplay Between Surface Morphology and Hydrophobicity. J. Am. Chem. Soc. 2015, DOI:10.1021/jacs.5b08748.
Comments 0
Request Comment
You are adding the first comment!
How to quickly get a good reply:
  • Give credit where it’s due by listing out the positive aspects of a paper before getting into which changes should be made.
  • Be specific in your critique, and provide supporting evidence with appropriate references to substantiate general statements.
  • Your comment should inspire ideas to flow and help the author improves the paper.

The better we are at sharing our knowledge with each other, the faster we move forward.
""
The feedback must be of minimum 40 characters and the title a minimum of 5 characters
   
Add comment
Cancel
Loading ...
248102
This is a comment super asjknd jkasnjk adsnkj
Upvote
Downvote
""
The feedback must be of minumum 40 characters
The feedback must be of minumum 40 characters
Submit
Cancel

You are asking your first question!
How to quickly get a good answer:
  • Keep your question short and to the point
  • Check for grammar or spelling errors.
  • Phrase it like a question
Test
Test description