Ionic liquids with cyano anions have long been used because of their unique combination of low-melting temperatures, reduced viscosities, and increased conductivities. Recently we have shown that cyano anions in ionic liquids are particularly interesting for their potential use as electron donors to excited state photo-acceptors [B. Wu et al., J. Phys. Chem. B 119, 14790–14799 (2015)]. Here we report on bulk structural and quantum mechanical results for a series of ionic liquids based on the 1-ethyl-3-methylimidazolium cation, paired with the following five cyano anions: SeCN, SCN, N(CN)2, C(CN)3, and B(CN)4. By combining molecular dynamics simulations, high-energy X-ray scattering measurements, and periodic boundary condition DFT calculations, we are able to obtain a comprehensive description of the liquid landscape as well as the nature of the HOMO-LUMO states for these ionic liquids in the condensed phase. Features in the structure functions for these ionic liquids are somewhat different than the commonly observed adjacency, charge-charge, and polarity peaks, especially for the bulkiest B(CN)4 anion. While the other four cyano-anion ionic liquids present an anionic HOMO, the one for Im2,1+/B(CN)4 is cationic.

Ionic liquids (ILs) continue to increase in importance for energy technology applications.1–6 Most modern ILs combine a set of inorganic or organic anions with a set of largely organic cations to generate a huge multiplicity of combinations leading to designer solvent properties. The charged and polar parts of the anionic and cationic species are now well known to separate from the non-polar components, leading to local nanophase segregation.7–33 One of the dominant structural features of bulk ILs is the order between adjacent anions and cations that is primarily caused by strong Coulomb interactions. Though Coulomb and hydrogen bond interactions make important contribution to the potential energy, many studies have shown that van der Waals interactions also play an important role in the structural organization of ILs.34–38 

In this work we compare structural features obtained from both high-energy X-ray scattering and molecular dynamics simulations for the series of five cyano-anion ILs shown in Fig. 1. So that we can focus solely on the structural trends of this series of cyano-anion based ILs, we have chosen a cation that is known not to induce intermediate-range structural order. These five ILs are based on the 1-ethyl-3-methylimidazolium cation (Im2,1+) paired with one from a series of anions having one to four cyano groups: selenocyanate (SeCN), thiocyanate (SCN), dicyanamide (N(CN)2), tricyanomethanide (C(CN)3), and tetracyanoborate (B(CN)4).

FIG. 1.

Molecular structures of the cyano-anions and the Im2,1+ cation with the atom label scheme.

FIG. 1.

Molecular structures of the cyano-anions and the Im2,1+ cation with the atom label scheme.

Close modal

Cyano-functionalized anions produce some of the most fluid and conductive ILs with low melting temperatures and low viscosities. For this reason, a number of studies have been performed to determine their suitability for use as electrolytes in dye sensitized solar cells.39–43 Wu et al. have recently investigated the potential of these anions to be used as electron donors in charge transfer reactions instead of only as a solvation medium.44 Nevertheless, there have been fewer structural studies on the cyano-anion ILs45,46 relative to the number of studies on ILs with fluorinated anions, such as bis(trifluoromethylsulfonyl)amide (NTf2), BF4  and PF6.25–29,47–49 A very recent computational study by Weber and Kirchner reported on dynamical and structural aspects of Im2,1+/SCN, Im2,1+/N(CN)2, and Im2,1+/B(CN)4, including a significant discussion on hydrogen bonding and cationic stacking interactions.45 Though details and procedures are not the same as those presented here, we generally find qualitative or quantitative agreement with their work when the results can be precisely compared. Besides adding data for two additional ILs, Im2,1+/SeCN and Im2,1+/C(CN)3, the current work also presents high-energy X-ray scattering data for quantitative comparison with our simulation analysis, as well as a detailed characterization of the electrochemical stability window for this set of five cyano-anion ILs.

Im2,1+/SeCN and Im2,1+/SCN were synthesized according the procedure described in Refs. 39 and 50. Im2,1+/N(CN)2 and Im2,1+/C(CN)3 were purchased from Iolitec, while Im2,1+/B(CN)4 was purchased from EMD Millipore USA. All ionic liquids samples were dried for 48 h on a vacuum Schlenk line at 10−2 Torr and 40 °C. The sample containers were back-filled with argon and transferred to an argon glove box. Inside the glove box, the IL samples were transferred to 2.0 mm outer diameter X-ray quartz capillaries, purchased from Hampton Research (HR6-150) and sealed using capillary wax (HR4-328). The sample height inside the quartz capillary was about 25 mm. The quartz capillary was flame sealed after being cooled for 10 min in a copper cylinder holder immersed in liquid nitrogen.

High-energy X-ray scattering data were collected at beamline 11-ID-C at the Advanced Photon Source at Argonne National Laboratory (APS, ANL) following protocols described previously.25–29 The X-ray photon energy was 114.99 keV, for a wavelength of λ = 0.1078 Å, and data were analyzed over a range of scattering vectors ranging from 0.2 < q < 14 Å−1 in reciprocal space. The X-ray beam cross section was 0.2  ×  0.2 mm2. The X-ray scattering patterns were recorded using a Perkin Elmer amorphous Silicon 1621 CN3-EHS detector. The sample-to-detector distance, beam center and the detector tilt were calibrated using the data collected from standard cerium oxide sample. The sample-to-detector distance was 766.5 mm. X-ray diffraction data were collected with 10 min exposures (6 s/frame, 20 frames/file and 5 files/data-set). A Cryostream open-flow temperature control system (Oxford Cryosystems) was used to maintain the sample temperature at 295 K.

Raw X-ray scattering data were integrated and graphed as intensity versus scattering wave vector q using the Fit2D software.51 The magnitude of the scattering vector is defined as q = (4π/λ)sin(θ), where 2θ is the scattering angle and λ is the X-ray wavelength. The observable in the scattering experiment is the coherent X-ray scattering intensity, Icoh(q), from which we derive the structure function of the liquid. The structure function is related to the coherent part of the total scattered intensity as described in the equation

S(q)=Icoh(q)ixifi2(q)ixifi(q)2,
(1)

xi and fi are the atomic fraction and X-ray form factor for atomic species i, respectively.

Integrated scattering data for both the IL sample and the background, measured using an empty 2.0 mm quartz capillary, were then used as input to the PDFgetX2 program52 to obtain the total scattering structure function, S(q), with corrections for sample absorption, Compton scattering, and multiple scattering. We note that corrections for sample absorption and multiple scattering are extremely small for such high energy X-rays in transmission mode geometry. In addition to standard corrections, an offset value was applied to the background data (empty 2.0 mm quartz capillary). A positive offset value was necessary due to the high dark current for this particular type of detector.

The IL simulations were carried out for systems containing 1000 ion pairs in a cubic box with periodic boundary conditions using the GROMACS package.53,54 We performed the equilibration of the simulation box as in our previous studies:28,55 we started with constant pressure and temperature simulations on systems in which ions have 1% of their charge and slowly increased the charges to 100%. This protocol was then followed by a 4 ns temperature annealing scheme in which the temperature of the system was increased to 500 K and then decreased back to the target value of 295 K. During this intermediate state, we used the NPT ensemble with v-rescale thermostat and Berendsen barostat56 as coded in GROMACS to control the temperature and pressure. As a final production step all systems were further run in the NPT ensemble for 6 ns at 295 K and a pressure of 1 bar. In this final step, the temperature and pressure were controlled by using the Nosé-Hoover thermostat57–59 and the Parrinello-Rahman barostat,60 respectively. The equations of motion were integrated using a time step of 1 fs using the leap-frog algorithm. Coulomb and Lennard–Jones cutoffs were set to 1.5 nm. The Particle Mesh Ewald (PME) method61,62 with an interpolation order of 6 and Fourier grid spacing of 0.8 Å was used to account for electrostatic interactions.

The full set of parameters including partial charges used for our simulations is provided in the supplementary material.63 The force field parameters for Im2,1+ and N(CN)2 were obtained from the well-established Canongia-Lopes and Pádua (CL&P) force field;64,65 the force field parameters for other cyano anions come from a combination of different sources as well as our own ab initio calculations. For SCN, parameters were obtained from Refs. 66 and 67. In the case of C(CN)3, parameters come from a combination of sources including the CL&P potentials, charges computed by us and OPLS-AA parameters as included in the GROMACS database.53,54,64,68 Parameters for B(CN)4 were obtained from Refs. 64 and 69. OPLS-AA type parameters for SeCN were not available in the literature so we created new ones based on those of SCN. The Se–C distance was set to conform to experiments,70 the Lennard–Jones parameters for the Se atom were reparameterized to better approximate the measured structure function and ab initio charges were generated. Atomic partial charges for SeCN and C(CN)3 were fitted using the CHelpG protocol implemented in the Gaussian09 code71 at the MP2 level of theory using the aug-cc-pvtz basis set.

The structure function S(q) was computed as

S(q)=ρ0ijxixjfi(q)fj(q)0L/24πr2(gij(r)1)sin(qr)qrW(r)dr[xifi(q)]2,
(2)

where gij is the pair distribution function for atomic species of type i and j. xi and xj are the fractions of atoms of type i and j, and fi(q) and fj(q) are the X-ray atomic form factors.72 ρ0 and L are the total number density and the periodicity of the simulation box, respectively. The Lorch function W(r) is used to reduce the effect of finite truncation of g(r) at large values of r.73,74

In order to calculate the condensed phase electrochemical potential window and other ab initio quantities, we considered a system with 8 ion pairs periodically replicated. Equilibration consisted first of classical MD simulations following a similar protocol to that used for the larger system followed by 300 steps of ab initio conjugate gradient optimization using the SIESTA package75–77 during which the box size was kept fixed. SIESTA automatically applies periodic boundary conditions. The goal of this optimization step was not to bring the liquid to a minimum energy state but instead to adjust bond lengths, angles, and dihedrals derived from the classical simulations to the DFT potential energy surface. This is consistent with prior work from our group and others.78–81 

In this work we used non-relativistic Troullier-Martins pseudopotentials82 and the Perdew, Burke, and Ernzerhof (PBE)83 implementation of the generalized gradient approximation (GGA). Our pseudo-potentials were generated using the ATOM program and energies were carefully tested for the various possible excitations of each element. We used the double zeta plus polarization basis set with an energy shift of 25 meV after testing that even in the case of the larger Se atoms this basis set gave an accurate representation of the energy. The mesh cutoff was set to 250 Ry which translates to a real space grid resolution of about 0.1 Å. We conducted the Brillouin zone sampling only to the Γ-point; this is reasonable for our large disordered liquid systems. The electronic temperature was taken to be 300 K.

The anodic and cathodic limits for the ILs are associated with the highest occupied molecular orbital (HOMO) and lowest unoccupied molecular orbital (LUMO), respectively.78,84 Considering the potentials corresponding to the anodic limit (VAL) and the cathodic limit (VCL) as the energies of the HOMO and LUMO levels of the ILs, respectively, one can write

VCL=εLUMOeandVAL=εHOMOe.
(3)

In Eq. (3)e is the electronic charge, and εHOMO, εLUMO are the energies of the HOMO and LUMO levels, respectively. Thus, the EPW can be written as

EPW=VALVCL=εLUMOεHOMOe.
(4)

A first test of the accuracy of a force field is to compare the simulated vs. experimental densities; these results are presented in Table I for Im2,1+ combined with the family of cyano-anions. The agreement between simulation and experiment is quite reasonable. The large increase in density from SCN to SeCN is expected based on the similar anionic volume but larger mass of Se. The rest of the series follows a trend based mainly on volume, where larger anionic volumes correlate with lower densities.

TABLE I.

Simulated and experimental densities (g cm−3) for Im2,1+ with indicated cyano anions. Experimental viscosities (cP) are also shown. The experimental values were obtained at 298.15 K and 1 atm, and the simulated densities were obtained at 295 K and 1 atm.

Ionic liquidSim. densityExp. densityExp. viscosity
Im2,1+/SeCN 1.407 … 25.0 (Ref. 39
Im2,1+/SCN 1.155 1.117 (Ref. 85), 1.117 (Ref. 8629.2 (Ref. 85
Im2,1+/N(CN)2 1.113 1.104 (Ref. 85), 1.101 (Ref. 8618.8 (Ref. 85
Im2,1+/C(CN)3 1.087 1.082 (Ref. 85), 1.082 (Ref. 8617.3 (Ref. 85
Im2,1+/B(CN)4 1.080 1.036 (Ref. 85), 1.036 (Ref. 6922.5 (Ref. 85
Ionic liquidSim. densityExp. densityExp. viscosity
Im2,1+/SeCN 1.407 … 25.0 (Ref. 39
Im2,1+/SCN 1.155 1.117 (Ref. 85), 1.117 (Ref. 8629.2 (Ref. 85
Im2,1+/N(CN)2 1.113 1.104 (Ref. 85), 1.101 (Ref. 8618.8 (Ref. 85
Im2,1+/C(CN)3 1.087 1.082 (Ref. 85), 1.082 (Ref. 8617.3 (Ref. 85
Im2,1+/B(CN)4 1.080 1.036 (Ref. 85), 1.036 (Ref. 6922.5 (Ref. 85

Fig. 2 shows the measured liquid structure functions S(q) graphed in the range between 0.2 ≤ q ≤ 14.0 Å−1 as compared to those obtained computationally. This provides another test of force field accuracy as well as a starting point for analysis of structural correlations in the liquid phase across the set of cyano-anions. Whereas the match between experimental and computational S(q) is not perfect, it is quite acceptable and represents a good basis for the interpretation of experimental data based on simulations. The most relevant region in reciprocal space can be found below 2.5 Å−1. In this region, ionic liquids often display two features. We have often been able to characterize these as arising from adjacency correlations and periodic charge alternation. For systems with anionic or cationic alkyl substitutions of length n ≥ 6, alternation between polar and non-polar domains can lead to a third peak in the structure function S(q) for small values of q, often referred to as a first sharp diffraction peak (FSDP).87,88 Because Im2,1+ has no significant non-polar substituents, a FSDP for q < 0.5 Å−1 is not observed. We must therefore identify only the signatures of periodic charge alternation and shorter range adjacency correlations.

FIG. 2.

Comparison between experiments and simulations of the X-ray structure function S(q) for Im2,1+/SeCN, Im2,1+/SCN, Im2,1+/N(CN)2, Im2,1+/C(CN)3, and Im2,1+/B(CN)4 at 295 K. For clarity, vertical offsets of +2, +4, +6, and +8 are applied for Im2,1+/C(CN)3, Im2,1+/N(CN)2, Im2,1+/SCN, and Im2,1+/SeCN, respectively.

FIG. 2.

Comparison between experiments and simulations of the X-ray structure function S(q) for Im2,1+/SeCN, Im2,1+/SCN, Im2,1+/N(CN)2, Im2,1+/C(CN)3, and Im2,1+/B(CN)4 at 295 K. For clarity, vertical offsets of +2, +4, +6, and +8 are applied for Im2,1+/C(CN)3, Im2,1+/N(CN)2, Im2,1+/SCN, and Im2,1+/SeCN, respectively.

Close modal

One way to unambiguously identify alternations in the liquid phase is by partitioning S(q) into a unique sum of subcomponents. A judicious choice of subcomponents – polar head groups of cations and anions in the case of charge alternation, or polar and apolar in the case of polarity alternation – leads to what we have termed as peaks and antipeaks separating length scales associated with charge and polarity alternation from shorter range (larger q) correlations that are associated either with intramolecular phenomena or adjacency between ions.89,90 Fig. 3 exemplifies this procedure by highlighting the q range where charge alternation occurs, clearly identified by two peaks and an antipeak (also emphasized with a vertical line). This alternation, shown by the presence of two peaks and one antipeak, is associated with the typical periodicity of ions of the same charge separated by those of opposite charge. Peaks or features in S(q) for the range of q between the charge alternation region and q ∼ 2.5 Å−1 include contributions from many short-range intramolecular and nearest-neighbor interactions.

FIG. 3.

Partitioning of S(q) into cation-anion, cation-cation, and anion-anion correlations. Vertical lines indicate the position in q space associated with charge alternation.

FIG. 3.

Partitioning of S(q) into cation-anion, cation-cation, and anion-anion correlations. Vertical lines indicate the position in q space associated with charge alternation.

Close modal

A few things are worth highlighting. First, as expected, alternations in real space often give rise to cancellations in the liquid structure function S(q). It is not uncommon to see peaks and antipeaks that are large compared to the overall S(q) intensity; this includes situations for which complete destructive interference cancels the charge alternation feature from the overall S(q).23,25,27–30,89,90 For the case of liquids that do not display a FSDP resulting from polarity alternation, this leads to a single peak at q values consistent with adjacency correlations only. One must emphasize that the absence of a charge alternation peak in the overall S(q) only indicates perfect destructive interference for peaks and antipeaks but never absence of charge alternation symmetry, since this charge alternation is the hallmark of all ionic liquids. Considering Fig. 2 in the relevant region below 2.5 Å−1, we can see a clear distinction between liquids containing SCN, SeCN, and N(CN)2 when compared to those containing C(CN)3 and B(CN)4. In the case of the first three liquids one notices a single peak (and in some cases an incipient shoulder) at smaller q values. The only peak at q values between 1.8-1.9 Å−1 arises from what we have previously called adjacency correlations.89 Comparison with Fig. 3 confirms that for SCN, SeCN, and N(CN)2, destructive interference between peaks and antipeaks is the reason why S(q) does not show a well defined peak associated with charge alternation.

In contrast, B(CN)4 and C(CN)3 show two clear features in S(q) below 2.5 Å−1. However, defying established intuition,27–30,90 the feature at lowest q value cannot be assigned to charge alternation. One hint that the peak at lower q values does not arise from charge alternation is that this feature appears for q > 1 Å−1. With bulky anions such as B(CN)4 and C(CN)3 one should expect charge alternation to occur at characteristically larger distances (lower q) than in the other three liquids composed of smaller cyano anions. If we carefully review Fig. 3 we see that indeed charge alternation for B(CN)4 and C(CN)3 is at q values significantly lower than 1 Å−1. In each case two clear peaks and an antipeak confirm this. Just as in the case of SCN, SeCN, and N(CN)2, for these two systems charge alternation also appears to be abolished from the overall S(q) because of fortuitous cancellation of peaks and antipeaks. What then is the origin of the peak at 1.3 Å−1 in the experimental S(q)?

In the different families of ionic liquids we have studied to date, we have repeatedly assigned the first peak in S(q) for q ≤ 2 Å−1 to either intramolecular or intermolecular adjacency correlations. These correspond to a myriad of different shorter range interactions associated either with intramolecular phenomena or with local solvation interactions at distances smaller or equal to the typical separation of ions with the same charge that are intercalated by ions of opposite charge. It appears that when the volume of the anions is large so as to significantly block access of cations to other cations this adjacency peak splits into two. This is because the intermolecular cationic contribution to adjacency correlations necessarily shifts to lower q values. We therefore see the expected cation-anion and anion-anion adjacency correlations close to q ∼ 2 Å−1, but also a cation-cation correlation at longer distances (smaller q, but still larger than that for charge alternation). In other words, for the larger anions, both peaks in the total S(q) are due to adjacency correlations. This is a novel result that departs from the common intuition we had developed in prior studies. Please also see Fig. S163 where for the Im2,1+/B(CN)4 system charge alternation and the two adjacency peaks are highlighted in the context of the full S(q).

To understand this observation of two distinct adjacency peaks in S(q), we focus on the center-of-mass spatial distribution functions surrounding a cation depicted in Fig. 4. Such SDFs facilitate our understanding of adjacency phenomena, as charge repetition linked to periodic charge alternation only gets established beyond the second solvation shell. For small anions, the distribution is preferentially aligned off-axis in equatorial regions about the ring hydrogens. Density from cations can also be found at distances not much larger than the first solvation shell of anions. Weber and Kirchner suggested that the significant density in the cationic spatial distribution above and below the imidazolium ring results from π-stacking interactions between the Im2,1+ cations.45 Fig. 4 shows that the density distributions for Im2,1+/SCN and Im2,1+/SeCN are very similar. As the anionic volume increases, bands of anionic solvation appear above and below the central cationic ring and the bulkier anions displace the cationic ring stacking. This is most prominently observed for the B(CN)4 and C(CN)3 anions. As this occurs the q value at which anions and cations contribute to “adjacency” correlations with the central cation splits resulting in two peaks in the overall S(q). The cationic contribution is observed at smaller value of q ≈ 1.3 Å−1, corresponding to larger distances.

FIG. 4.

Spatial Distribution Functions (SDFs) about the Im2,1+ cation for the five cyano-anion ILs. Anionic densities are shown in blue; cationic densities in red. Results are averaged over 6000 liquid snapshots from our MD trajectories. The isosurface values for anionic and cationic centers-of-mass around a central cation in units of nm−3: SeCN 7.42 (anion), 4.44 (cation); SCN 8.02 (anion), 4.65 (cation); N(CN)2 6.50 (anion), 4.19 (cation); C(CN)3 5.01 (anion), 3.48 (cation); B(CN)4 4.32 (anion), 3.14 (cation). The TRAVIS software was used to generate this figure.91 

FIG. 4.

Spatial Distribution Functions (SDFs) about the Im2,1+ cation for the five cyano-anion ILs. Anionic densities are shown in blue; cationic densities in red. Results are averaged over 6000 liquid snapshots from our MD trajectories. The isosurface values for anionic and cationic centers-of-mass around a central cation in units of nm−3: SeCN 7.42 (anion), 4.44 (cation); SCN 8.02 (anion), 4.65 (cation); N(CN)2 6.50 (anion), 4.19 (cation); C(CN)3 5.01 (anion), 3.48 (cation); B(CN)4 4.32 (anion), 3.14 (cation). The TRAVIS software was used to generate this figure.91 

Close modal

A complementary display of the charge ordering is obtained by plotting the SDFs for the cationic and anionic center-of-mass probabilities about a central cyano-anion, as shown in Fig. 5. The red cationic densities in Fig. 5 show symmetries indicative of those for the central anion. The second shell anionic densities, reflecting one of the contributions to the same-type charge-charge correlation, are of course more averaged, but nevertheless show high symmetry for the C(CN)3 and B(CN)4 anions with C3v and Td point group symmetry, respectively.

FIG. 5.

Spatial Distribution Functions (SDFs) of the Im2,1+ cation center-of-mass about each of the five anion centers-of-mass. Anionic densities are shown in blue; cationic densities in red. Results are averaged over 6000 liquid snapshots from our MD trajectories. The isosurface values for anionic and cationic centers-of-mass around a central cation in units of nm−3: SeCN 5.00 (anion), 6.90 (cation); SCN 5.25 (anion), 6.20 (cation); N(CN)2 4.65 (anion), 5.60 (cation); C(CN)3 4.15 (anion), 4.95 (cation); B(CN)4 3.62 (anion), 4.75 (cation). The TRAVIS program was used to generate this figure.91 

FIG. 5.

Spatial Distribution Functions (SDFs) of the Im2,1+ cation center-of-mass about each of the five anion centers-of-mass. Anionic densities are shown in blue; cationic densities in red. Results are averaged over 6000 liquid snapshots from our MD trajectories. The isosurface values for anionic and cationic centers-of-mass around a central cation in units of nm−3: SeCN 5.00 (anion), 6.90 (cation); SCN 5.25 (anion), 6.20 (cation); N(CN)2 4.65 (anion), 5.60 (cation); C(CN)3 4.15 (anion), 4.95 (cation); B(CN)4 3.62 (anion), 4.75 (cation). The TRAVIS program was used to generate this figure.91 

Close modal

Further analysis of correlations in real space for this set of ILs is provided in the supplementary material.63 Fig. S263 shows selected pair distribution functions between cationic hydrogen and possible anionic hydrogen bonding partners. Since the hydrogen connected to C2 is always the most likely hydrogen bonding partner, in Fig. S3 we show a joint distance/angle distribution function between C2, H2, and terminal N atom of the different anions.63 For the three ILs discussed both by Weber and Kirchner and in this work, our classical force field results are in excellent agreement with the ab initio methods.45 

Table II summarizes the calculated and the experimental EPWs for our five cyano-anion ILs. With the exception of Im2,1+/SeCN for which to the best of our knowledge the EPW is not available, Table II shows the very good agreement one can establish between periodic boundary condition DFT calculations and electrochemical data. The EPW of these five ILs increases in the following order: Im2,1+/SeCN < Im2,1+/SCN < Im2,1+/C(CN)3 < Im2,1+/N(CN)2 < Im2,1+/B(CN)4. If ionic liquids based on cyano-anions are to be used as electron donors,44 it is important to fully understand the states from which such electrons may originate. To this aim we computed ion-based projected density of states (PDOS) that provide information about the nature of the electron donating states in the condensed phase.

TABLE II.

Calculated EPWs, in units of V, for the five cyano-anion ILs. The standard deviation σsim is calculated for 10 snapshots each separated by 1 ns.

Ionic liquidsEPWexpEPWsimσsim
Im2,1+/SeCN … 2.75 0.09 
Im2,1+/SCN 2.60 (Refs. 50 and 922.95 0.14 
Im2,1+/C(CN)3 2.90 (Refs. 93 and 943.00 0.05 
Im2,1+/N(CN)2 3.0, 3.50 (Refs. 95–973.37 0.11 
Im2,1+/B(CN)4 4.0, 4.6 (Refs. 98 and 994.37 0.07 
Ionic liquidsEPWexpEPWsimσsim
Im2,1+/SeCN … 2.75 0.09 
Im2,1+/SCN 2.60 (Refs. 50 and 922.95 0.14 
Im2,1+/C(CN)3 2.90 (Refs. 93 and 943.00 0.05 
Im2,1+/N(CN)2 3.0, 3.50 (Refs. 95–973.37 0.11 
Im2,1+/B(CN)4 4.0, 4.6 (Refs. 98 and 994.37 0.07 

Fig. 6 shows that in all cases except for Im2,1+/B(CN)4, the HOMO levels are predominantly anionic. In other words, the anodic limit is likely set by the oxidation of the cyano anions. However, for Im2,1+/B(CN)4, the HOMO states are mostly cationic. This means that if the solvent were to donate an electron, the cationic species would be the donor. For all systems except Im2,1+/B(CN)4 (where LUMO states are of mixed cationic-anionic nature), the LUMO states are mostly cationic. Such ionic-species dependent cathodic and anodic limits have been observed for other ILs before.84 Recently Wu et al. demonstrated that anions in all of these liquids except for Im2,1+/B(CN)4 could act as electron donors to a photoexcited acceptor, 9, 10-dicyanoanthracene.44 The fact that the anion in Im2,1+/B(CN)4 is not a good electron donor is not surprising in light of our current findings where we see that the anodic limit is cationic. A significantly larger EPW is associated with the change in the anodic limit for Im2,1+/B(CN)4.

FIG. 6.

The cationic and anionic projected density of states (PDOS) for five different ILs considered here. The Fermi energy is set to zero in all cases. The line broadening function for the PDOS calculation was set to 0.2 eV.

FIG. 6.

The cationic and anionic projected density of states (PDOS) for five different ILs considered here. The Fermi energy is set to zero in all cases. The line broadening function for the PDOS calculation was set to 0.2 eV.

Close modal

To better understand which atoms within the ions contribute the most to possible donor states, we further partitioned the PDOS into atomic contributions. Fig. 7 shows that for Im2,1+/SeCN and Im2,1+/SCN, the HOMO energy contributions are greatest from selenium/sulfur atoms with smaller contributions from nitrogen and much smaller from carbon. For Im2,1+/N(CN)2, the nitrogen contribution which is distributed across the central and CN components is significantly more important than the carbon contribution. The situation is different for Im2,1+/C(CN)3, where the nitrogen and carbon contributions to the HOMO are similar. In this case, it is the central carbon and not the carbon atoms in the CN groups that contribute to the HOMO state. As expected, anionic contributions to HOMO in the case of Im2,1+/B(CN)4 are not important, instead it is cationic ring carbon that contributes the most to the HOMO state.

FIG. 7.

Atomic contributions to cationic and anionic projected density of states (PDOS). The Fermi energy is set to zero. The line broadening for the PDOS calculation was set to 0.2 eV.

FIG. 7.

Atomic contributions to cationic and anionic projected density of states (PDOS). The Fermi energy is set to zero. The line broadening for the PDOS calculation was set to 0.2 eV.

Close modal

Fig. 8 highlights the nature of HOMO and LUMO states in the liquid phase in the case of Im2,1+/C(CN)3 and Im2,1+/B(CN)4. Similar images of the HOMOs and LUMOs for the Im2,1+ ILs with SeCN, SCN, and N(CN)2 are provided in Fig. S4 of the supplementary material.63 The HOMO state is predominantly localized on cations for Im2,1+/B(CN)4, whereas the opposite is true for the other four ILs. We also notice that for some systems it is more than one ion that contributes to a given state.

FIG. 8.

Highest Occupied Molecular Orbitals (HOMOs) and Lowest Unoccupied Molecular Orbitals (LUMOs) for Im2,1+/C(CN)3 and Im2,1+/B(CN)4. The cationic or anionic nature of HOMO and LUMO states in the case of our other systems is consistent with that of Im2,1+/C(CN)3 as can be appreciated from Fig. S4 in the supplementary material. The blue square denotes the edges of the periodic box. Atoms shown outside the box correspond to periodic images.

FIG. 8.

Highest Occupied Molecular Orbitals (HOMOs) and Lowest Unoccupied Molecular Orbitals (LUMOs) for Im2,1+/C(CN)3 and Im2,1+/B(CN)4. The cationic or anionic nature of HOMO and LUMO states in the case of our other systems is consistent with that of Im2,1+/C(CN)3 as can be appreciated from Fig. S4 in the supplementary material. The blue square denotes the edges of the periodic box. Atoms shown outside the box correspond to periodic images.

Close modal

ILs with cyano-substituted anions are widely used because they can display both relatively low viscosities and wide EPWs. We have used MD simulations, X-ray scattering, and periodic boundary condition first principles calculations to interrogate the nature of low viscosity cyano-anion ionic liquids. We find that structure in these liquids is dominated by charge alternation and adjacency close range correlations. Normally such adjacency correlations appear as a broad single peak in S(q) near 2 Å−1. It turns out that when anions are bulky and block access to cations from other cations such as in the case of C(CN)3 and B(CN)4, one can observe not one but two adjacency peaks. In these two systems, the peak at lowest q value should not be mistaken for charge alternation that occurs at even lower q and is completely masked by cancellation of peaks and antipeaks. Spatial correlation functions show that smaller cyano-anions solvate imidazolium often off-axis but along the direction of possible hydrogen bonds whereas cations often appear above and below the ring. Instead as the anions become bulkier anionic solvation appears like a band that covers above and below the ring. This likely explains the shift of the cation-cation adjacency peak to smaller q values around 1.3 Å−1 and the split into two different adjacency peaks in the total S(q).

First principles studies show that in all cases but Im2,1+/B(CN)4, the anodic limit is anionic. One of our groups has recently shown that anions in all of these systems except for B(CN)4 can act as effective electron donors to an excited state acceptor. Im2,1+/B(CN)4 is different in that the liquid band gap and consequently the EPW is significantly larger and the anodic limit is cationic making this a poor electron donating system. This characteristic may be advantageous for a number of electronic or electrochemical technologies.

L.F.O.F. and M.C.C.R. thank the Brazilian FAPESP and CNPq for financial support. Work at Rutgers and Iowa was supported by the U.S. Department of Energy, Office of Basic Energy Sciences, Division of Chemical Sciences, Geosciences, and Biosciences under Contract Nos. DE-SC0001780 (B.W., M.L., and E.W.C.) and DE-SC0008644 (K.B.D. and C.J.M.). High energy X-ray experiments at APS beamline 11-ID-C were supported by the U.S. Department of Energy, Office of Science, Office of Basic Energy Sciences, under Contract No. DE-AC02-06CH11357. We thank Dr. James F. Wishart, Brookhaven National Laboratory, for providing a sample of Im2,1+/B(CN)4. K.B.D. thanks Dr. Juan Carlos Araque for instructive discussions.

1.
N. V.
Plechkova
and
K. R.
Seddon
, “
Applications of ionic liquids in the chemical industry
,”
Chem. Soc. Rev.
37
,
123
150
(
2008
).
2.
H.
Weingärtner
, “
Understanding ionic liquids at the molecular level: Facts, problems, and controversies
,”
Ang. Chem., Int. Ed.
47
,
654
670
(
2008
).
3.
J. F.
Wishart
, “
Energy applications of ionic liquids
,”
Energy Environ. Sci.
2
,
956
961
(
2009
).
4.
T.
Torimoto
,
T.
Tsuda
,
K.
Okazaki
, and
S.
Kuwabata
, “
New frontiers in materials science opened by ionic liquids
,”
Adv. Mater.
22
,
1196
1221
(
2010
).
5.
E. W.
Castner
, Jr.
,
C. J.
Margulis
,
M.
Maroncelli
, and
J. F.
Wishart
, “
Ionic liquids: Structure and photochemical reactions
,”
Annu. Rev. Phys. Chem.
62
,
85
105
(
2011
).
6.
D. R.
MacFarlane
,
N.
Tachikawa
,
M.
Forsyth
,
J. M.
Pringle
,
P. C.
Howlett
,
G. D.
Elliott
,
J. J. H.
Davis
,
M.
Watanabe
,
P.
Simon
, and
C. A.
Angell
, “
Energy applications of ionic liquids
,”
Energy Environ. Sci.
7
,
232
250
(
2014
).
7.
M. G.
Del Popolo
and
G. A.
Voth
, “
On the structure and dynamics of ionic liquids
,”
J. Phys. Chem. B
108
,
1744
1752
(
2004
).
8.
Y.
Wang
and
G. A.
Voth
, “
Unique spatial heterogeneity in ionic liquids
,”
J. Am. Chem. Soc.
127
,
12192
12193
(
2005
).
9.
Y.
Wang
and
G. A.
Voth
, “
Tail aggregation and domain diffusion in ionic liquids
,”
J. Phys. Chem. B
110
,
18601
18608
(
2006
).
10.
J. N. A.
Canongia Lopes
and
A. A. H.
Pádua
, “
Nanostructural organization in ionic liquids
,”
J. Phys. Chem. B
110
,
3330
3335
(
2006
).
11.
L. P. N.
Rebelo
,
J. N. C.
Lopes
,
J. M. S. S.
Esperança
,
H. J. R.
Guedes
,
J.
Łachwa
,
V.
Najdanovic-Visak
, and
Z. P.
Visak
, “
Accounting for the unique, doubly dual nature of ionic liquids from a molecular thermodynamic and modeling standpoint
,”
Acc. Chem. Res.
40
,
1114
1121
(
2007
).
12.
W.
Jiang
,
Y.
Wang
, and
G. A.
Voth
, “
Molecular dynamics simulation of nanostructural organization in ionic liquid/water mixtures
,”
J. Phys. Chem. B
111
,
4812
4818
(
2007
).
13.
J. N.
Canongia Lopes
,
K.
Shimizu
,
A. A. H.
Pádua
,
Y.
Umebayashi
,
S.
Fukuda
,
K.
Fujii
, and
S.-i.
Ishiguro
, “
A tale of two ions: The conformational landscapes of Bis(trifluoromethanesulfonyl)amide and N,N-Dialkylpyrrolidinium
,”
J. Phys. Chem. B
112
,
1465
1472
(
2008
).
14.
K.
Fujii
,
Y.
Soejima
,
Y.
Kyoshoin
,
S.
Fukuda
,
R.
Kanzaki
,
Y.
Umebayashi
,
T.
Yamaguchi
,
S.-i.
Ishiguro
, and
T.
Takamuku
, “
Liquid structure of room-temperature ionic liquid, 1-ethyl-3-methylimidazolium bis-(trifluoromethanesulfonyl) imide
,”
J. Phys. Chem. B
112
,
4329
4336
(
2008
).
15.
R.
Atkin
and
G. G.
Warr
, “
The smallest amphiphiles: Nanostructure in protic room-temperature ionic liquids with short alkyl groups
,”
J. Phys. Chem. B
112
,
4164
4166
(
2008
).
16.
K.
Fujii
,
S.
Seki
,
S.
Fukuda
,
T.
Takamuku
,
S.
Kohara
,
Y.
Kameda
,
Y.
Umebayashi
, and
S.-i.
Ishiguro
, “
Liquid structure and conformation of a low-viscosity ionic liquid, N-methyl-N-propyl-pyrrolidinium bis(fluorosulfonyl) imide studied by high-energy X-ray scattering
,”
J. Mol. Liq.
143
,
64
69
(
2008
).
17.
S.
Fukuda
,
M.
Takeuchi
,
K.
Fujii
,
R.
Kanzaki
,
T.
Takamuku
,
K.
Chiba
,
H.
Yamamoto
,
Y.
Umebayashi
, and
S.-i.
Ishiguro
, “
Liquid structure of N-butyl-N-methylpyrrolidinium bis(trifluoromethanesulfonyl) amide ionic liquid studied by large angle X-ray scattering and molecular dynamics simulations
,”
J. Mol. Liq.
143
,
2
7
(
2008
).
18.
Y.
Umebayashi
,
W.-L.
Chung
,
T.
Mitsugi
,
S.
Fukuda
,
M.
Takeuchi
,
K.
Fujii
,
T.
Takamuku
,
R.
Kanzaki
, and
S.-i.
Ishiguro
, “
Liquid structure and the ion-ion interactions of ethylammonium nitrate ionic liquid studied by large angle X-ray scattering and molecular dynamics simulations
,”
J. Comput. Chem., Jpn.
7
,
125
134
(
2008
).
19.
R.
Hayes
,
S. Z.
El Abedin
, and
R.
Atkin
, “
Pronounced structure in confined aprotic room-temperature ionic liquids
,”
J. Phys. Chem. B
113
,
7049
7052
(
2009
).
20.
R.
Kanzaki
,
T.
Mitsugi
,
S.
Fukuda
,
K.
Fujii
,
M.
Takeuchi
,
Y.
Soejima
,
T.
Takamuku
,
T.
Yamaguchi
,
Y.
Umebayashi
, and
S.
Ishiguro
, “
Ion-ion interaction in room temperature ionic liquid 1-ethyl-3-methylimidazolium tetrafluoroborate studied by large angle X-ray scattering experiment and molecular dynamics simulations
,”
J. Mol. Liq.
147
,
77
82
(
2009
).
21.
E. W.
Castner
, Jr.
and
J. F.
Wishart
, “
Spotlight on ionic liquids
,”
J. Chem. Phys.
132
,
120901
(
2010
).
22.
K.
Fujii
,
R.
Kanzaki
,
T.
Takamuku
,
Y.
Kameda
,
S.
Kohara
,
M.
Kanakubo
,
M.
Shibayama
,
S.-i
Ishiguro
, and
Y.
Umebayashi
, “
Experimental evidences for molecular origin of low-Q peak in neutron/x-ray scattering of 1-alkyl-3-methylimidazolium bis(trifluoromethanesulfonyl)amide ionic liquids
,”
J. Chem. Phys.
135
,
244502
(
2011
).
23.
H. V. R.
Annapureddy
,
H. K.
Kashyap
,
P. M.
De Biase
, and
C. J.
Margulis
, “
What is the origin of the prepeak in the x-ray scattering of imidazolium-based room-temperature ionic liquids?
,”
J. Phys. Chem. B
114
,
16838
16846
(
2010
).
24.
R.
Hayes
,
S.
Imberti
,
G. G.
Warr
, and
R.
Atkin
, “
Pronounced sponge-like nanostructure in propylammonium nitrate
,”
Phys. Chem. Chem. Phys.
13
,
13544
13551
(
2011
).
25.
C. S.
Santos
,
H. V. R.
Annapureddy
,
N. S.
Murthy
,
H. K.
Kashyap
,
E. W.
Castner
, Jr.
, and
C. J.
Margulis
, “
Temperature-dependent structure of methyltributylammonium Bis(trifluoromethylsulfonyl)amide: X-ray scattering and simulations
,”
J. Chem. Phys.
134
,
064501
(
2011
).
26.
C. S.
Santos
,
N. S.
Murthy
,
G. A.
Baker
, and
E. W.
Castner
, Jr.
, “
Communication: X-ray scattering from ionic liquids with pyrrolidinium cations
,”
J. Chem. Phys.
134
,
121101
(
2011
).
27.
H. K.
Kashyap
,
C. S.
Santos
,
H. V. R.
Annapureddy
,
N. S.
Murthy
,
C. J.
Margulis
, and
E. W.
Castner
, Jr.
, “
Temperature-dependent structure of ionic liquids: X-ray scattering and simulations
,”
Faraday Discuss.
154
,
133
143
(
2012
).
28.
H. K.
Kashyap
,
C. S.
Santos
,
R. P.
Daly
,
J. J.
Hettige
,
N. S.
Murthy
,
H.
Shirota
,
E. W.
Castner
, Jr.
, and
C. J.
Margulis
, “
How does the ionic liquid organizational landscape change when nonpolar cationic alkyl groups are replaced by polar isoelectronic diethers?
,”
J. Phys. Chem. B
117
,
1130
1135
(
2013
).
29.
H. K.
Kashyap
,
C. S.
Santos
,
N. S.
Murthy
,
J. J.
Hettige
,
K.
Kerr
,
S.
Ramati
,
J.
Gwon
,
M.
Gohdo
,
S. I.
Lall-Ramnarine
,
J. F.
Wishart
,
C. J.
Margulis
, and
E. W.
Castner
, Jr.
, “
Structure of 1-alkyl-1-methylpyrrolidinium bis(trifluoromethylsulfonyl)amide ionic liquids with linear, branched, and cyclic alkyl groups
,”
J. Phys. Chem. B
117
,
15328
15337
(
2013
).
30.
J. J.
Hettige
,
H. K.
Kashyap
,
H. V. R.
Annapureddy
, and
C. J.
Margulis
, “
Anions, the reporters of structure in ionic liquids
,”
J. Phys. Chem. Lett.
4
,
105
110
(
2012
).
31.
Y.
Shen
,
D. F.
Kennedy
,
T. L.
Greaves
,
A.
Weerawardena
,
R. J.
Mulder
,
N.
Kirby
,
G.
Song
, and
C. J.
Drummond
, “
Protic ionic liquids with fluorous anions: Physicochemical properties and self-assembly nanostructure
,”
Phys. Chem. Chem. Phys.
14
,
7981
7992
(
2012
).
32.
M.
Kofu
,
M.
Nagao
,
T.
Ueki
,
Y.
Kitazawa
,
Y.
Nakamura
,
S.
Sawamura
,
M.
Watanabe
, and
O.
Yamamuro
, “
Heterogeneous slow dynamics of imidazolium-based ionic liquids studied by neutron spin echo
,”
J. Phys. Chem. B
117
,
2773
2781
(
2013
).
33.
R.
Hayes
,
G. G.
Warr
, and
R.
Atkin
, “
Structure and nanostructure in ionic liquids
,”
Chem. Rev.
115
,
6357
6426
(
2015
).
34.
S.
Kossmann
,
J.
Thar
,
B.
Kirchner
,
P.
Hunt
, and
T.
Welton
, “
Cooperativity in ionic liquids
,”
J. Chem. Phys.
124
,
174506
(
2006
).
35.
C.
Spickermann
,
J.
Thar
,
S. B. C.
Lehmann
,
S.
Zahn
,
J.
Hunger
,
R.
Buchner
,
P. A.
Hunt
,
T.
Welton
, and
B.
Kirchner
, “
Why are ionic liquid ions mainly associated in water? A car-parrinello study of 1-ethyl-3-methyl-imidazolium chloride water mixture
,”
J. Chem. Phys.
129
,
104505
(
2008
).
36.
J.
Thar
,
M.
Brehm
,
A. P.
Seitsonen
, and
B.
Kirchner
, “
Unexpected hydrogen bond dynamics in imidazolium-based ionic liquids
,”
J. Phys. Chem. B
113
,
15129
15132
(
2009
).
37.
K.
Wendler
,
S.
Zahn
,
F.
Dommert
,
R.
Berger
,
C.
Holm
,
B.
Kirchner
, and
L.
Delle Site
, “
Locality and fluctuations: Trends in lmidazolium-based ionic liquids and beyond
,”
J. Chem. Theory Comput.
7
,
3040
3044
(
2011
).
38.
A. S.
Pensado
,
M.
Brehm
,
J.
Thar
,
A. P.
Seitsonen
, and
B.
Kirchner
, “
Effect of dispersion on the structure and dynamics of the ionic liquid 1-ethyl-3-methylimidazolium thiocyanate
,”
Chemphyschem
13
,
1845
1853
(
2012
).
39.
P.
Wang
,
S. M.
Zakeeruddin
,
J. E.
Moser
,
R.
Humphry-Baker
, and
M.
Gratzel
, “
A solvent-free, SeCN-/(SeCN)(3)(-) based ionic liquid electrolyte for high-efficiency dye-sensitized nanocrystalline solar cells
,”
J. Am. Chem. Soc.
126
,
7164
7165
(
2004
).
40.
D.
Kuang
,
S.
Uchida
,
R.
Humphry-Baker
,
S. M.
Zakeeruddin
, and
M.
Graetzel
, “
Organic dye-sensitized ionic liquid based solar cells: Remarkable enhancement in performance through molecular design of indoline sensitizers
,”
Angew. Chem., Int. Ed.
47
,
1923
1927
(
2008
).
41.
J.
Scheers
,
P.
Johansson
, and
P.
Jacobsson
, “
Anions for lithium battery electrolytes: A
,”
J. Electrochem. Soc.
155
,
A628
A634
(
2008
).
42.
M.
Marszalek
,
Z. F.
Fei
,
D. R.
Zhu
,
R.
Scopelliti
,
P. J.
Dyson
,
S. M.
Zakeeruddin
, and
M.
Gratzel
, “
Application of ionic liquids containing tricyanomethanide [C(CN)3] or tetracyanoborate [B(CN)4] anions in dye-sensitized solar cells
,”
Inorg. Chem.
50
,
11561
11567
(
2011
).
43.
D.
Zhou
,
Y.
Bai
,
J.
Zhang
,
N.
Cai
,
M.
Su
,
Y.
Wang
,
M.
Zhang
, and
P.
Wang
, “
Anion effects in organic dye-sensitized mesoscopic solar cells with ionic liquid electrolytes: Tetracyanoborate vs dicyanamide
,”
J. Phys. Chem. C
115
,
816
822
(
2011
).
44.
B.
Wu
,
M.
Liang
,
M.
Maroncelli
, and
E. W.
Castner
, Jr.
, “
Photoinduced bimolecular electron transfer from cyano anions in ionic liquids
,”
J. Phys. Chem. B
119
,
14790
14799
(
2015
).
45.
H.
Weber
and
B.
Kirchner
, “
Complex structural and dynamical interplay of cyano-based ionic liquids
,”
J. Phys. Chem. B
120
,
2471
2483
(
2016
).
46.
N.
Vergadou
,
E.
Androulaki
,
J.-R.
Hill
, and
I. G.
Economou
, “
Molecular simulations of imidazolium-based tricyanomethanide ionic liquids using an optimized classical force field
,”
Phys. Chem. Chem. Phys.
18
,
6850
6860
(
2016
).
47.
O.
Russina
,
A.
Triolo
,
L.
Gontrani
, and
R.
Caminiti
, “
Mesoscopic structural heterogeneities in room-temperature ionic liquids
,”
J. Phys. Chem. Lett.
3
,
27
33
(
2012
).
48.
A.
Triolo
,
O.
Russina
,
B.
Fazio
,
G. B.
Appetecchi
,
M.
Carewska
, and
S.
Passerini
, “
Nanoscale organization in piperidinium-based room temperature ionic liquids
,”
J. Chem. Phys.
130
,
164521
(
2009
).
49.
O.
Russina
,
A.
Triolo
,
L.
Gontrani
,
R.
Caminiti
,
D.
Xiao
,
J.
Hines
,
G.
Larry
,
R. A.
Bartsch
,
E. L.
Quitevis
,
N.
Pleckhova
, and
K. R.
Seddon
, “
Morphology and intermolecular dynamics of 1-alkyl-3-methylimidazolium bis(trifluoromethane)sulfonylamide ionic liquids: Structural and dynamic evidence of nanoscale segregation
,”
J. Phys.: Condens. Matter
21
,
424121
(
2009
).
50.
J. M.
Pringle
,
J.
Golding
,
C. M.
Forsyth
,
G. B.
Deacon
,
M.
Forsyth
, and
D. R.
MacFarlane
, “
Physical trends and structural features in organic salts of the thiocyanate anion
,”
J. Mater. Chem.
12
,
3475
3480
(
2002
).
51.
A. P.
Hammersley
,
S. O.
Svensson
,
M.
Hanfland
,
A. N.
Fitch
, and
D.
Hausermann
, “
Two-dimensional detector software: From real detector to idealised image or two-theta scan
,”
High Pressure Res.
14
,
235
248
(
1996
).
52.
X.
Qiu
,
J. W.
Thompson
, and
S. J. L.
Billinge
, “
PDFgetX2: A GUI-driven program to obtain the pair distribution function from X-ray powder diffraction data
,”
J. Appl. Crystallogr.
37
,
678
(
2004
).
53.
B.
Hess
,
C.
Kutzner
,
D.
van der Spoel
, and
E.
Lindahl
, “
GROMACS 4: Algorithms for highly efficient, load-balanced, and scalable molecular simulation
,”
J Chem. Theory Comput.
4
,
435
447
(
2008
).
54.
D.
Van Der Spoel
,
E.
Lindahl
,
B.
Hess
,
G.
Groenhof
,
A. E.
Mark
, and
H. J. C.
Berendsen
, “
GROMACS: Fast, flexible, and free
,”
J. Comput. Chem.
26
,
1701
1718
(
2005
).
55.
J. J.
Hettige
,
H. K.
Kashyap
, and
C. J.
Margulis
, “
Communication: Anomalous temperature dependence of the intermediate range order in phosphonium ionic liquids
,”
J. Chem. Phys.
140
,
111102
(
2014
).
56.
H. J. C.
Berendsen
,
J. P. M.
Postma
,
W. F.
van Gunsteren
,
A.
DiNola
, and
J. R.
Haak
, “
Molecular dynamics with coupling to an external bath
,”
J. Chem. Phys.
81
,
3684
3690
(
1984
).
57.
S.
Nosé
, “
A unified formulation of the constant temperature molecular dynamics methods
,”
J. Chem. Phys.
81
,
511
519
(
1984
).
58.
S.
Nosé
, “
A molecular dynamics method for simulations in the canonical ensemble
,”
Mol. Phys.
52
,
255
268
(
1984
).
59.
W. G.
Hoover
, “
Canonical dynamics: Equilibrium phase-space distributions
,”
Phys. Rev. A
31
,
1695
1697
(
1985
).
60.
M.
Parrinello
and
A.
Rahman
, “
Polymorphic transitions in single crystals: A new molecular dynamics method
,”
J. Appl. Phys.
52
,
7182
7190
(
1981
).
61.
T.
Darden
,
D.
York
, and
L.
Pedersen
, “
Particle mesh Ewald: An Nlog(N) method for Ewald sums in large systems
,”
J. Chem. Phys.
98
,
10089
10092
(
1993
).
62.
U.
Essmann
,
L.
Perera
,
M. L.
Berkowitz
,
T.
Darden
,
H.
Lee
, and
L. G.
Pedersen
, “
A smooth particle mesh Ewald method
,”
J. Chem. Phys.
103
,
8577
8593
(
1995
).
63.
See supplementary material at http://dx.doi.org/10.1063/1.4955186 for more detailed supplementary information.
64.
J. N. C.
Lopes
,
J.
Deschamps
, and
A. A. H.
Pádua
, “
Modeling ionic liquids using a systematic all-atom force field
,”
J. Phys. Chem. B
108
,
2038
2047
(
2004
).
65.
J. N. C.
Lopes
and
A. A. H.
Padua
, “
Molecular force field for ionic liquids III: Imidazolium, pyridinium, and phosphonium cations; chloride, bromide, and dicyanamide anions
,”
J. Phys. Chem. B
110
,
19586
19592
(
2006
).
66.
A.
Chaumont
and
G.
Wipff
, “
Solvation of Ln(III) lanthanide cations in the [BMI][SCN], [MeBu3N][SCN], and [BMI]5[Ln(NCS)8] ionic liquids: A molecular dynamics study
,”
Inorg. Chem.
48
,
4277
4289
(
2009
).
67.
A. B.
Pereiro
,
J. M. M.
Araujo
,
F. S.
Oliveira
,
C. E. S.
Bernardes
,
J. M. S. S.
Esperanca
,
J. N.
Canongia Lopes
,
I. M.
Marrucho
, and
L. P. N.
Rebelo
, “
Inorganic salts in purely ionic liquid media: The development of high ionicity ionic liquids (HIILs)
,”
Chem. Commun.
48
,
3656
3658
(
2012
).
68.
W. J.
Jorgensen
,
D. S.
Maxwell
, and
J.
Tirado-Rives
, “
Development and testing of the OPLS all-atom force field on conformational energetics and properties of organic liquids
,”
J. Am. Chem. Soc.
118
,
11225
11236
(
1996
).
69.
T.
Koller
,
J.
Ramos
,
N. M.
Garrido
,
A. P.
Fröba
, and
L. G.
Economou
, “
Development of a united-atom force field for 1-ethyl-3-methylimidazolium tetracyanoborate ionic liquid
,”
Mol. Phys.
110
,
1115
1126
(
2012
).
70.
L. J.
Farrugia
,
C. J.
Carmalt
, and
N. C.
Norman
, “
Syntheses and X-ray crystal structures of the bismuth(III) thiocyanate and selenocyanate complexes K-3(dmpu)(4) Bi(SCN)(6) and K-3(dmpu)(4) Bi(SeCN)(6) (dmpu=N,N’-dimethylpropyleneurea)
,”
Inorg. Chim. Acta
248
,
263
266
(
1996
).
71.
M. J.
Frisch
,
G. W.
Trucks
,
H. B.
Schlegel
,
G. E.
Scuseria
,
M. A.
Robb
,
J. R.
Cheeseman
,
G.
Scalmani
,
V.
Barone
,
B.
Mennucci
,
G. A.
Petersson
,
H.
Nakatsuji
,
M.
Caricato
,
X.
Li
,
H. P.
Hratchian
,
A. F.
Izmaylov
,
J.
Bloino
,
G.
Zheng
,
J. L.
Sonnenberg
,
M.
Hada
,
M.
Ehara
,
K.
Toyota
,
R.
Fukuda
,
J.
Hasegawa
,
M.
Ishida
,
T.
Nakajima
,
Y.
Honda
,
O.
Kitao
,
H.
Nakai
,
T.
Vreven
,
J. A.
Montgomery
, Jr.
,
J. E.
Peralta
,
F.
Ogliaro
,
M.
Bearpark
,
J. J.
Heyd
,
E.
Brothers
,
K. N.
Kudin
,
V. N.
Staroverov
,
R.
Kobayashi
,
J.
Normand
,
K.
Raghavachari
,
A.
Rendell
,
J. C.
Burant
,
S. S.
Iyengar
,
J.
Tomasi
,
M.
Cossi
,
N.
Rega
,
J. M.
Millam
,
M.
Klene
,
J. E.
Knox
,
J. B.
Cross
,
V.
Bakken
,
C.
Adamo
,
J.
Jaramillo
,
R.
Gomperts
,
R. E.
Stratmann
,
O.
Yazyev
,
A. J.
Austin
,
R.
Cammi
,
C.
Pomelli
,
J. W.
Ochterski
,
R. L.
Martin
,
K.
Morokuma
,
V. G.
Zakrzewski
,
G. A.
Voth
,
P.
Salvador
,
J. J.
Dannenberg
,
S.
Dapprich
,
A. D.
Daniels
,
Farkas
,
J. B.
Foresman
,
J. V.
Ortiz
,
J.
Cioslowski
, and
D. J.
Fox
, gaussian 09, Revision D.01, Gaussian, Inc., Wallingford, CT, 2009.
72.
International Tables for Crystallography
, edited by
E.
Prince
(
International Union of Crystallography
,
2006
), Vol.
C
.
73.
E.
Lorch
, “
Neutron diffraction by germania, silica and radiation-damaged silica glasses
,”
J. Phys. C: Solid State Phys.
2
,
229
(
1969
).
74.
J.
Du
,
C. J.
Benmore
,
R.
Corrales
,
R. T.
Hart
, and
J. K. R.
Weber
, “
A molecular dynamics simulation interpretation of neutron and x-ray diffraction measurements on single phase Y2O3–Al2O3 glasses
,”
J. Phys.: Condens. Matter
21
,
205102
(
2009
).
75.
E.
Artacho
,
E.
Anglada
,
O.
Diéguez
,
J. D.
Gale
,
A.
García
,
J.
Junquera
,
R. M.
Martin
,
P.
Ordejón
,
J. M.
Pruneda
,
D.
Sánchez-Portal
, and
J. M.
Soler
, “
The SIESTA method; developments and applicability
,”
J. Phys.: Condens. Matter
20
,
064208
(
2008
).
76.
J. M.
Soler
,
E.
Artacho
,
J. D.
Gale
,
A.
García
,
J.
Junquera
,
P.
Ordejón
, and
D.
Sánchez-Portal
, “
The SIESTA method for ab initio order-N materials simulations
,”
J. Phys.: Condens. Matter
14
,
2745
(
2002
).
77.
D.
Sánchez-Portal
,
P.
Ordejón
,
E.
Artacho
, and
J. M.
Soler
, “
Density-functional method for very large systems with LCAO basis sets
,”
Int. J. Quantum Chem.
65
,
453
461
(
1997
).
78.
Y.
Zhang
,
C.
Shi
,
J. F.
Brennecke
, and
E. J.
Maginn
, “
Refined method for predicting electrochemical windows of ionic liquids and experimental validation studies
,”
J. Phys. Chem. B
118
,
6250
6255
(
2014
).
79.
C. J.
Margulis
,
H. V. R.
Annapureddy
,
P. M. D.
Biase
,
D.
Coker
,
J.
Kohanoff
, and
M. G. D.
Popolo
, “
Dry excess electrons in room-temperature ionic liquids
,”
J. Am. Chem. Soc.
133
,
20186
20193
(
2011
).
80.
C.
Xu
,
A.
Durumeric
,
H. K.
Kashyap
,
J.
Kohanoff
, and
C. J.
Margulis
, “
Dynamics of excess electronic charge in aliphatic ionic liquids containing the Bis(trifluoromethylsulfonyl)amide anion
,”
J. Am. Chem. Soc.
135
,
17528
17536
(
2013
).
81.
C.
Xu
and
C. J.
Margulis
, “
Solvation of an excess electron in pyrrolidinium dicyanamide based ionic liquids
,”
J. Phys. Chem. B
119
,
532
542
(
2015
).
82.
N.
Troullier
and
J. L.
Martins
, “
Efficient pseudopotentials for plane-wave calculations
,”
Phys. Rev. B
43
,
1993
2006
(
1991
).
83.
J. P.
Perdew
,
K.
Burke
, and
M.
Ernzerhof
, “
Generalized gradient approximation made simple
,”
Phys. Rev. Lett.
77
,
3865
3868
(
1996
).
84.
S. P.
Ong
,
O.
Andreussi
,
Y.
Wu
,
N.
Marzari
, and
G.
Ceder
, “
Electrochemical windows of room-temperature ionic liquids from molecular dynamics and density functional theory calculations
,”
Chem. Mater.
23
,
2979
2986
(
2011
).
85.
C. M. S. S.
Neves
,
K. A.
Kurnia
,
J. A. P.
Coutinho
,
I. M.
Marrucho
,
J. N.
Canongia Lopes
,
M. G.
Freire
, and
L. P. N.
Rebelo
, “
Systematic study of the thermophysical properties of imidazolium-based ionic liquids with cyano-functionalized anions
,”
J. Phys. Chem. B
117
,
10271
10283
(
2013
).
86.
P.
Navarro
,
M.
Larriba
,
E.
Rojo
,
J.
Garcia
, and
F.
Rodriguez
, “
Thermal properties of cyano-based ionic liquids
,”
J. Chem. Eng. Data
58
,
2187
2193
(
2013
).
87.
M.
Wilson
and
P. A.
Madden
, “
Voids, layers, and the first sharp diffraction peak in ZnCl2
,”
Phys. Rev. Lett.
80
,
532
535
(
1998
).
88.
S. M.
Urahata
and
M. C. C.
Ribeiro
, “
Structure of ionic liquids of 1-alkyl-3-methylimidazolium cations: A systematic computer simulation study
,”
J. Chem. Phys.
120
,
1855
1863
(
2004
).
89.
H. K.
Kashyap
,
J. J.
Hettige
,
H. V. R.
Annapureddy
, and
C. J.
Margulis
, “
SAXS anti-peaks reveal the length-scales of dual positive-negative and polar-apolar ordering in room-temperature ionic liquids
,”
Chem. Commun.
48
,
5103
5105
(
2012
).
90.
J. C.
Araque
,
J. J.
Hettige
, and
C. J.
Margulis
, “
Modern room temperature ionic liquids, a simple guide to understanding their structure and how it may relate to dynamics
,”
J. Phys. Chem. B
119
,
12727
12740
(
2015
).
91.
M.
Brehm
and
B.
Kirchner
, “
TRAVIS–A free analyzer and visualizer for monte carlo and molecular dynamics trajectories
,”
J. Chem. Inf. Model.
51
,
2007
2023
(
2011
).
92.
G.-h.
Sun
,
K.-x.
Li
, and
C.-g.
Sun
, “
Application of 1-ethyl-3-methylimidazolium thiocyanate to the electrolyte of electrochemical double layer capacitors
,”
J. Power Sources
162
,
1444
1450
(
2006
).
93.
Y.
Yoshida
,
K.
Muroi
,
A.
Otsuka
,
G.
Saito
,
M.
Takahashi
, and
T.
Yoko
, “
1-ethyl-3-methylimidazolium based ionic liquids containing cyano groups: Synthesis, characterization, and crystal structure
,”
Inorg. Chem.
43
,
1458
1462
(
2004
).
94.
S. A.
Forsyth
,
S. R.
Batten
,
Q.
Dai
, and
D. R.
MacFarlane
, “
Ionic liquids based on imidazolium and pyrrolidinium salts of the tricyanomethanide anion
,”
Aust. J. Chem.
57
,
121
124
(
2004
).
95.
J. N.
Barisci
,
G. G.
Wallace
,
D. R.
MacFarlane
, and
R. H.
Baughman
, “
Investigation of ionic liquids as electrolytes for carbon nanotube electrodes
,”
Electrochem. Commun.
6
,
22
27
(
2004
).
96.
M.
Zistler
,
C.
Schreiner
,
P.
Wachter
,
P.
Wasserscheid
,
D.
Gerhard
, and
H. J.
Gores
, “
Electrochemical characterization of 1-ethyl-3-methylimidazolium thiocyanate and measurement of triiodide diffusion coefficients in blends of two ionic liquids
,”
Int. J. Electrochem. Sci.
3
,
236
245
(
2008
), available online at http://www.electrochemsci.org/papers/vol3/3030236.pdf.
97.
D. R.
MacFarlane
,
S. A.
Forsyth
,
J.
Golding
, and
G. B.
Deacon
, “
Ionic liquids based on imidazolium, ammonium and pyrrolidinium salts of the dicyanamide anion
,”
Green Chem.
4
,
444
448
(
2002
).
98.
G. P.
Pandey
and
S. A.
Hashmi
, “
Studies on electrical double layer capacitor with a low-viscosity ionic liquid 1-ethyl-3-methylimidazolium tetracyanoborate as electrolyte
,”
Bull. Mater. Sci.
36
,
729
733
(
2013
).
99.
S.
Thiemann
,
S.
Sachnov
,
S.
Porscha
,
P.
Wasserscheid
, and
J.
Zaumseil
, “
Ionic liquids for electrolyte-gating of ZnO field-effect transistors
,”
J. Phys. Chem. C
116
,
13536
13544
(
2012
).

Supplementary Material