Designing Pyrrolidinium-Based Ionic Liquid Electrolytes for Energy Storage: Thermal and Electrical Behaviour of Ternary Mixtures with Lithium Salt and Carbonates

Abstract

Ionic liquids (ILs) have attracted increasing attention due to their unique physicochemical properties. Among them, 1-Methyl-1-propylpyrrolidinium bis(trifluoromethanesulfonyl)imide, emerges as an ideal candidate for fundamental studies in electrochemical applications. This work aims to deepen the understanding of its conductivity performance, and potential interaction with added metal salts, providing insight into its applicability in advanced energy storage systems. Firstly, binary mixtures with ethylene carbonate have been prepared to improve the transport properties and select the optimal concentration of both components. Subsequently, lithium salt was added to design the adequate electrolyte. The thermal and electrochemical characterisation of these mixtures was performed by differential scanning calorimetry (DSC) thermogravimetric analysis (TGA) and Broad Band Dielectric Spectroscopy (BBDS). The results reveal a wide liquid range for the ternary systems studied, extending below −80 °C and above 120 °C. Additionally, they exhibit notably high conductivity values at room temperature (ranging from 0.2 S·m⁻¹ for the most concentrated to 0.70 S·m⁻¹ for the lowest concentrated), which highlights their suitability for advanced electrochemical applications, including but not limited to batteries. This extended liquid phase mitigates, or potentially eliminates, some of the most common issues associated with current electrolytes, such as undesired crystallisation at low temperatures. In this paper, a new promising electrolyte, consisting of a ternary mixture obtained by adding lithium salt to the eutectic composition of [C₃C₁Pyrr][TFSI] and ethyl carbonate is proposed.

1. Introduction

One of the biggest challenges facing modern society is the transition to a sustainable model based on renewable resources, which requires solving our heavy reliance on fossil fuels. This transition is essential to mitigate climate change, reduce carbon emissions, and ensure long-term energy security. However, achieving this shift is not straightforward and involves overcoming significant obstacles, such as the intermittency of renewable energy sources like solar and wind, whose production is variable, dependent on weather conditions and time of day.

To address this challenge, it is crucial to improve the performance of existing energy storage technologies, which play a key role in stabilising the grid, allowing for excess energy generated during periods of high production to be stored and then released during periods of low production. The current scenario of the battery market is dominated by lithium-ion batteries (LIBs), which have become the preferred choice due to their numerous advantages over other battery systems, such as high specific capacity and voltage, absence of memory effect, excellent cycling performance, low self-discharge, and wide temperature range of operation [1,2]. Nevertheless, many of the currently available energy storage devices, like LIBs, face limitations in terms of energy capacity, charging times, lifespan, and cost. Enhancing these devices to store larger quantities of energy, charge more efficiently, last longer, and remain cost-effective is critical for enabling the widespread adoption of renewable energy.

Carbonates are essential components of both current lithium-ion batteries (LIBs) and the interesting alternative of lithium metal batteries (LMBs), which use more concentrated electrolytes to increase their performance [3] and in new configurations under study as, for example, high-entropy electrolytes [4]. Among the many carbonates available, ethylene carbonate (EC) stands out due to its high dielectric constant, a desirable characteristic for an electrolyte. However, EC has a relatively high melting point (36 °C), which limits its performance at lower temperatures [4,5]. To address this issue, linear carbonates such as dimethyl carbonate (DMC) or diethyl carbonate (DEC), can also be added to the previous electrolyte mixture to improve their fluidity by reducing viscosity and lowering the melting temperature. Consequently, LiPF₆/linear carbonate(s)/EC [6] has become one of the most widely used and popular electrolyte formulations in today’s industry due to its balanced performance in terms of conductivity, thermal stability, and operational temperature range. Nevertheless, linear carbonates pose significant safety concerns due to their low flash point, which is near room temperature [7]. This makes them highly flammable under thermal, electrical, or mechanical stress, potentially leading to severe consequences such as fire and/or explosion [8], especially in battery packs composed of thousands of cells. The combination of ionic liquids (ILs) with cyclic carbonates has emerged as a promising electrolyte for various electrochemical applications, providing a safer and more reliable option compared to traditional linear carbonate-based electrolytes [8,9]. Different carbonates and even water are also cosolvents of greatest interest as is widely pointed out in recent literature [10,11,12,13,14,15,16].

ILs are ionic compounds with a melting point below 100 °C, consisting of an organic cation and a weakly coordinated inorganic or organic anion. They are particularly attractive due to their non-flammability, negligible vapour pressure, good ionic conductivity, wide electrochemical window, and high chemical and thermal stability. These unique properties make ILs highly versatile for a broad range of applications, including their use as solvents, lubricants, and electrolytes [7].

Among the virtually unlimited number of ILs, N-methyl-N-propylpyrrolidinium bis(trifluoromethanesulfonyl)imide ([C₃C₁Pyrr][TFSI]) was selected for this work due to the favourable characteristics of pyrrolidinium-based ILs and their suitability for electrochemical applications [17]. Regarding the lithium salt, lithium bis(trifluoromethanesulfonyl)imide (LiTFSI) was chosen because it offers several advantages, including higher ionic conductivity, no moisture sensitivity, and greater thermal and electrochemical stability compared to LiPF₆ [18].

This study takes a step forward from previous research on pyrrolidinium-based IL electrolytes [19] by introducing ternary mixtures that incorporate lithium salt and ethylene carbonate into [C₃C₁Pyrr][TFSI]. The aim is to explore robust and reliable formulations that optimise their performance for energy storage applications. These ternary systems are designed to improve key properties such as thermal stability and ionic conductivity, while addressing limitations observed in traditional electrolytes. This work represents a significant advancement in the development of smart electrolytes for next-generation energy storage devices, offering enhanced safety and efficiency.

2. Materials and Methods

2.1. Chemicals

To study the potential of ionic liquid-based systems for electrochemical applications, we selected 1-Methyl-1-propylpyrrolidinium bis(trifluoromethylsulfonyl)imide ([C₃C₁Pyrr][TFSI]) as the IL and lithium bis(trifluoromethylsulfonyl)imide (LiTFSI) as the lithium salt. Both compounds share the [TFSI]⁻ anion, a deliberate choice aimed at maximising salt solubility and minimising the complexity of the resulting ionic environment. Ethylene carbonate (EC), a solvent widely used in conventional battery electrolytes, was incorporated into the formulations due to its low volatility and favourable electrochemical properties [20].

Table 1. Identification of chemicals used in this work.

Name	Molecular Mass (g·mol−1)	Structure	Short Name CAS Number	Purity Provenance
1-Methyl-1-propylpyrrolidinium bis(trifluoromethylsulfonyl) imide	408.4		[C3C1Pyrr][TFSI] 223437-05-6	0.99 a Iolitec, Heilbronn, Germany
Lithium bis(trifluoromethylsulfonyl) imide	287.09		LiTFSI 90076-65-6	>0.99 a Sigma Aldrich, San Luis, MI, USA
Ethylene carbonate	88.06		EC 96-49-1	>0.99 a Acros Organics, Waltham, MA, USA

^a indicated by the provider.

summarises the name, molecular mass, chemical structure, CAS number, purity, and provenance of the chemicals used in this work. Nine binary mixtures of [C₃C₁Pyrr][TFSI] with EC across the entire concentration range and three ternary mixtures of [C₃C₁Pyrr][TFSI] + EC + LiTFSI (0.2 m, 0.5 m, and 1 m) were analysed in this study.

2.2. Differential Scanning Calorimetry

The DSC measurements were carried out using a DSC Q2000 from TA Instruments (Waters-TA Instruments. New Castle, DE, USA) with hermetically sealed aluminum pans. Samples weighing between 5 and 8 mg were analysed to determine their thermal behaviour and phase transitions. Transition onset temperatures were identified from the DSC curves during the reheating and re-cooling steps, with an estimated uncertainty of ±2 °C at a 95% confidence level [21].

Two experimental procedures varied depending on the mixture composition. For the systems without ethylene carbonate (EC), the procedure involved an initial heating of the sample from room temperature to 125 °C at a rate of 5 °C min⁻¹, followed by a step at 125 °C for 10 min to remove volatile impurities. After this, the sample was cooled to −85 °C at 5 °C min⁻¹, held isothermally at −85 °C for 5 min, and then reheated to 100 °C. Finally, a cooling step to −85 °C at 10 °C min⁻¹ was applied. In the case of mixtures containing EC, the initial heating step was omitted to prevent EC evaporation. Instead, the protocol began with a cooling ramp from room temperature to −85°C at 5 °C min⁻¹, followed by the subsequent steps as described above.

2.3. Thermogravimetric Analysis

The thermal stability of different binary and ternary mixtures was evaluated using thermogravimetric analysis (TGA) in dynamic mode to assess the short-term thermal stability and identify the upper limit of the liquid range. The experiments were conducted using a Mettler Toledo DSC/TGA1 instrument operated in dynamic mode. The samples, weighing between 3 and 5 mg, were placed in an open platinum pan and subjected to a temperature range from 50 to 800 °C at a rate of 10 °C· min⁻¹ under a nitrogen atmosphere, with a purge gas flow of 20 cm³ min⁻¹ [22,23].

2.4. Ionic Conductivity

The measurement protocol for ionic conductivities determination followed the approach detailed in our earlier studies [19]. Broadband dielectric spectroscopy was used for the ionic conductivity measurements employing an Agilent RLC precision meter HP 4284A (Agilent Technologies, Santa Clara, CA, USA), with a precision of 0.05%. The frequency range employed was between 20 Hz and 0.2 MHz to accurately fit the ohmic regime. The ionic conductivity was calculated from the intercept obtained by fitting the ohmic region of the imaginary part of the dielectric constant, selecting the region where the slope is −1.00 ± 0.02 in the log(ε) vs. log(ω) representation, as established by J. Leys [24]. The samples were placed between two stainless-steel electrodes and secured in a Swagelok cell, ensuring that the electrode separation remained constant during the experiment. Temperature control was achieved by placing the Swagelok cell inside a Memmert ICP400 climate chamber, where measurements were pursued across a temperature range from 0 °C to 50 °C, starting by the highest temperature.

3. Results and Discussion

3.1. Differential Scanning Calorimetry (DSC)

DSC curves of pure IL and binary mixtures of [C₃C₁Pyrr][TFSI] + EC are shown in Figure 1a. While the pure IL exhibits single peaks upon heating (at 9 °C) and cooling (at −11 °C), corresponding to melting and crystallisation processes, respectively, the DSC heating profiles of the mixtures are characterised by two endothermic peaks for all EC concentrations, except for the x = 0.5 mixture, which shows a single peak at −19 °C. This melting point, decreasing upon the addition of EC, has been observed for various IL-carbonate mixtures [25]. The remaining mixtures, which exhibit two endothermic peaks upon heating, interestingly present the first endothermic peak also at −19 °C. This behaviour suggests the possible formation of different polymorphs depending on the EC concentration. The onset temperatures and the enthalpies of every transition obtained from DSC curves are listed in

Table 2. Temperature (in °C) of melting $T_m$, freezing $T_c$, and cold crystallisation $T_{cc}$ enthalpy of melting (ΔH_m) and freezing (${\mathsf{\Delta }H}_c$) obtained from DSC curves.

Sample	${\mathit{T}}_{\mathit{m}}$	${\mathit{T}}_{\mathit{c}}$	${\mathit{T}}_{\mathit{c}\mathit{c}}$	ΔHm/J g−1	ΔHcc; ΔHc/J g−1
Mixtures (1 − x) EC + x [C3C1Pyrr][TFSI]
1.0	9	−11	--	36	33
0.7	−18|−8	−37|−49	--	42	35
0.5	−19		52	51	33
0.4	−19|−10	--	−61	63	53
0.3	−19|−3	--	−59	73	60
0.25	−19|0	--	−58	77	60
0.2	−19|5	−59	−44	87	13/50
0.15	−19|8	−51	−60	92	7/63
0.1	−19|15	−42		100	85
Ternary mixtures ([C3C1Pyrr][TFSI]/EC/LiTFSI)
0.2 m	−25	--	−39	2.6	2.6
0.5 m	--	--	--	--	--
1 m	--	--	--	--	--

. A typical supercooling effect, characteristic of ILs, is also observed in all these samples, with crystallisation temperatures below 0 °C in all the cases and with T_m-T_c > 20 °C.

Figure 1b shows the phase diagram, where the 0.5 molar ratio mixture can be considered the eutectic composition. This behaviour is in good agreement with the findings of Fox et al. [26] and Hofmann et al. [25] who studied mixtures of pyrrolidinium-based ionic liquids and 1-Ethyl-3-methylimidazolium bis(trifluoromethanesulfonyl)azanide, respectively, with EC as a solvent, and reported similar eutectic composition. Furthermore, it can be observed that the melting enthalpy increases with the addition of EC, obtaining lower values than the corresponding pure EC [27].

From the eutectic IL + EC mixture, ternary mixtures (IL + EC + LiTFSI) with three different salt concentrations, 0.2, 0.5, and 1 m (mol·(kg IL + EC mixture)⁻¹) were prepared.

Figure 2 shows the DSC curves of ternary mixtures of 0.5 [C₃C₁Pyrr][TFSI] + 0.5 EC and LiTFSI. The addition of lithium salt practically vanishes phase transitions over the entire studied temperature range, except for the mixture with the lowest lithium salt concentration, for which smooth cold crystallisation and melting peaks are observed. This vanishing is the typical behaviour after the salt addition, as has been widely reported previously [19,20,21].

3.2. Thermogravimetric Analysis (TGA)

The thermal stability of pure ionic liquids (ILs) has been extensively studied in previous papers. For instance, Chen et al. [28] reported an onset decomposition temperature of 435 °C for [C₃C₁Pyrr][TFSI] under similar experimental conditions. Furthermore, the thermogravimetric analysis of binary mixtures performed by several research groups has shown that the addition of lithium salt does not significantly affect the thermal stability of mixtures compared to pure IL [29,30,31]. Therefore, the main objective of this part was to evaluate the changes on the thermal stability in the newly formulated binary and ternary mixtures.

Figure 3 shows the thermogravimetric (TG) and derivative thermogravimetric (DTG) curves of binary mixtures of [C₃C₁Pyrr][TFSI] and ethylene carbonate (EC), with EC molar fractions of 0.3, 0.5, 0.85, and 0.9. Figure 4 presents the TG and DTG curves of the ternary mixtures containing EC and [C₃C₁Pyrr][TFSI] (at 0.5 molar fraction) with three different concentrations of LiTFSI.

TG and DTG curves of binary mixtures exhibit two well-defined decomposition steps. An onset temperature of approximately 150 °C is obtained from TG curves, regardless of the EC molar fraction. Furthermore, it can be seen that the greater the amount of ethylene carbonate (EC) in the binary mixture, the greater the initial mass loss. This mass loss can be attributed to the organic compound EC. Svoboda et al. [32] stated in their review that the decomposition temperature of EC is around 250 °C, but can begin to degrade at 120 °C in the presence of alkali hydroxides, as is the case here presented. Our findings are in good agreement with Le et al. [33] who observed a similar behaviour in binary systems consisting of imidazolium- and ammonium-based ionic liquids mixed with EC. They explained that the initial peak suggests that introducing EC into the IL medium strengthens ion–dipole interactions (interactions between different types of molecules) while partially disrupting the weaker dipole–dipole interactions among EC molecules (interactions between similar molecules). As a result, only a portion of the weaker EC–EC interactions are overcome as temperature increases. Notably, the degradation temperature of the EC-IL mixtures is still higher than that of the conventional LiPF₆/EC-DMC (1:1) electrolyte. The second peak appears at 435 °C (onset temperature), being also independent of the EC molar fraction, which is in concordance with the results obtained by Chen et al. [28]. These observations provide clear evidence that EC limits the thermal stability of the binary samples.

Table 3. Characteristic thermogravimetric properties (temperatures in °C and W_onset in %) of binary and ternary studied systems obtained from TGA curves.

Sample	${\mathit{T}}_{\mathit{o}\mathit{n}\mathit{s}\mathit{e}\mathit{t}}$	${\mathit{T}}_{\mathit{e}\mathit{n}\mathit{d}\mathit{s}\mathit{e}\mathit{t}}$	${\mathit{W}}_{\mathit{o}\mathit{n}\mathit{s}\mathit{e}\mathit{t}}$	${\mathit{T}}_{\mathit{p}\mathit{e}\mathit{a}\mathit{k}}$
Mixtures (1 − x) EC + x [C3C1Pyrr][TFSI]
0.7	136/436	194/465	98/72	182/458
0.5	135/435	198/468	97/66	175/455
0.15	144/430	196/466	89/36	178/454
0.1	148/424	197/465	86/28	182/453
Ternary mixtures ([C3C1Pyrr][TFSI]/EC/LiTFSI)
0.2 m	138/436	200/478	96/66	172/458
0.5 m	133/432	204/478	95/68	174/458
1 m	138/428	210/480	95/73	176/478

shows the characteristic thermogravimetric parameters of the ionic liquid (IL), including T_onset (representing the beginning of the mass loss process), T_endset (indicating the end of the mass loss process), W_onset (mass loss percentage at onset temperature), T_peak (maximum degradation rate in the DTG curve).

Figure 4 and Table 3 show that the initial mass loss remains constant as the amount of lithium salt increases, indicating that lithium addition does not affect the thermal stability of the mixtures compared to the binary ones, following the same trend. An approximate 17–18% loss is observed in this first mass loss step for both the ternary mixtures and the binary mixture of pure IL with EC at the same molar fraction (0.5:0.5) shown in Figure 3b. In line with the findings of Le et al. [33], the incorporation of LiTFSI salt results in a lower boiling point for binary and ternary mixtures with regard to the corresponding to free EC. This salt addition may promote interactions between free EC and either Li⁺ or TFSI⁻ within the complex mixture. Additionally, as previously mentioned, the second mass loss peak remains constant upon adding lithium salt, a behaviour already reported in earlier studies [34,35].

3.3. Broadband Dielectric Spectroscopy

Dielectric measurements were carried out upon cooling, taking advantage of the supercooling effect in these samples, pointed out in the previous Section 3.1, avoiding phase transition in this study. Figure 5a shows the ionic conductivity, determined from the imaginary part of the dielectric constant against frequency, of binary mixtures with EC against the molar fraction of ionic liquid at different temperatures.

Table 4. Ionic conductivities in S.m⁻¹ of the different studied mixtures: pure IL (x = 1) [19] binary mixtures with EC and ternary mixtures with LiTFSI at the 0.5:0.5 mixture of EC:[C₃C₁Pyrr][TFSI].

Sample	273 K	283 K	293 K	298 K	313 K	323 K
x	Mixtures (1 − x) EC + x [C3C1Pyrr][TFSI]
1.0	0.1138(37)	0.200(11)	0.308(21)	0.397(38)	0.645(59)	0,77(11)
0.85	0.1878(24)	0.3010(52)	0.4542(99)	0.538(12)	0.837(32)	1.023(43)
0.7	0.2253(31)	0.3688(70)	0.553(13)	0.645(16)	1.007(43)	1.242(61)
0.5	0.3783(86)	0.573(24)	0.806(48)	0.960(29)	1.36(18)	1.681(86)
0.4	0.467(11)	0.689(22)	0.927(35)	1.084(40)	1.531(75)	1.809(98)
0.3	0.588(15)	0.832(28)	1.092(44)	1.261(63)	1.704(99)	2.03(13)
0.25	0.618(13)	0.875(21)	1.153(34)	1.306(40)	1.783(69)	2.026(83)
0.2	0.677(28)	0.921(42)	1.180(66)	1.351(83)	1.76(15)	2.02(18)
0.15	0.713(15)	0.994(29)	1.291(43)	1.461(50)	1.903(87)	2.21(10)
0.1	0.1160(21)	0.924(76)	1.131(98)	1.28(12)	1.63(16)	1.79(18)
Ternary mixtures EC:[C3C1Pyrr][TFSI] (0.5:0.5) + [LiTFSI]
0.2 m [LiTFSI]	0.271(30)	0.406(48)	0.566(77)	0.696(96)	1.02(13)	1.26(16)
0.5 m [LiTFSI]	0.1432(94)	0.233(24)	0.362(40)	0.463(58)	0.648(85)	0.81(11)
1 m [LiTFSI]	0.06474(40)	0.1238(12)	0.2098(29)	0.2769(43)	0.4580(93)	0.564(16)

shows the ionic conductivity values of the studied systems. A clear maximum at a molar fraction of approximately 0.15 is observed for the binary mixtures, which smoothly decreases as the amount of ionic liquid increases. Although viscosity grows with the IL molar fraction [25,36], the initial rise in conductivity at low concentrations is attributed to the increase in the number of charge carriers up to x_IL = 0.15. Then, the formation of ion aggregates and the increase in viscosity lead to a gradual decrease in the ionic conductivity with the IL concentration [37,38]. This behaviour is typical of IL and carbonate mixtures, as observed by other authors [39]. It is related to the fact that while EC—a solid organic solvent at room temperature—is not ionically conductive, its high dielectric constant favors the dissociation of the ions from the ionic liquid, thereby enhancing ionic conductivity of mixtures.

Similar conductivity values were found by Le et al. [33] for imidazolium and ammonium-based systems using the same lithium salt.

To our knowledge, scarce previous studies analysing the transition temperatures, thermal stability, and ionic conductivity of ternary mixtures IL + carbonate + lithium salt for energy storage applications can be found in the literature. Among them, the works of Le et al. [33] and García-Garabal et al. [36] can be highlighted. In the first case, the authors report changes in different properties of ammonium- and imidazolium-based ILs (with the [TFSI]⁻ anion moiety) upon the addition of EC cosolvent to prepare the electrolyte solution for lithium-ion battery, observing an enhancement of ionic conductivity and ion mobility related to the decrease in electrolyte viscosity. The work of García-Garabal et al. [36] presents a study of the same ionic liquid, [C₃C₁Pyrr][TFSI], with LiTFSI mixed with different concentrations of other organic solvents (acetonitrile, γ-butyrolactone, and dimethyl sulfoxide), obtaining ionic conductivities up to 3.1 S m⁻¹ for mixtures with acetonitrile.

The ionic conductivity was also fitted against the molar fraction with the Casteel-Amis equation (Equation (1)) and represented in Figure 5a:

$$\sigma ={\sigma }_{max}{\left({\displaystyle \frac{x}{x_{max}}}\right)}^a\mathit{exp}\left(b{\left(x-x_{max}\right)}^2-{\displaystyle \frac{a}{x_{max}}}\left(x-x_{max}\right)\right)$$

(1)

where σ_max is the maximum ionic conductivity, x_max is the [C₃C₁Pyrr][TFSI] molar fraction at which the maximum conductivity is reached, and a and b are empirical parameters [39,40]. The fitting parameters are summarised in

Table 5. Fitting parameters of Casteel-Amis equation (Equation (1)) for all studied temperatures.

Temperature/K	σ/S m−1	xmax	a	b
273	0.725 (92)	0.117 (80)	0.43 (53)	0.8 (13)
283	0.983 (48)	0.130 (38)	0.44 (26)	0.75 (70)
293	1.258 (47)	0.131 (39)	0.35 (22)	0.39 (63)
298	1.413 (37)	0.156 (27)	0.50 (22)	0.84 (63)
313	1.839 (27)	0.175 (17)	0.45 (13)	0.46 (40)
323	2.109 (44)	0.191 (25)	0.41 (21)	0.15 (67)

.

As shown, the Casteel-Amis equation provides a good fit for all temperature data series, except for the lowest IL molar fraction (x_IL = 0.10) at 273 K, whose ionic conductivity is significantly lower than the expected value. After performing the measurement, the sample was found to be completely solid, which explains the observed low ionic conductivity. When comparing different temperature data sets, it is found that increasing the temperature shifts the maximum ionic conductivity to higher IL molar fractions (Table 5). This shift is attributed to the reduction in viscosity and ion aggregation, which enhances the mobility of charge carriers and causes a shift in the maximum ionic conductivity [41].

To further study the conduction mechanism and the ionic conductivity dependence with IL concentration, the 298 K data series was fitted with the pseudolattice theory [42], the resulting fit is presented in Figure 5b. This theory gives a universal curve without fitting parameters, predicting the normalised electrical conductivity of IL with polar solvents as shown in Equation (2) [43]:

$${\displaystyle \frac{\sigma }{{\sigma }_{max}}}={\xi }_{IL}\left(2-{\xi }_{IL}\right)$$

(2)

where, similarly with the previous equation, σ_max is the maximum of the ionic conductivity and ξ_IL is the reduced concentration, defined by Equation (3) as follows:

$${\xi }_{IL}={\displaystyle \frac{c}{c_{max}}}$$

(3)

Being c the ionic liquid concentration, in moles per litre, and c_max the concentration at which the maximum conductivity takes place. Even though the data follow quite well the universal curve, the fitting fails to determine the position and the maximum in ionic conductivity due to a lack of a parameter describing the high effects of high ion concentrations. To compute the effect of concentrated mixtures, an additional term can be added to the universal curve as follows:

$${\displaystyle \frac{\sigma }{{\sigma }_{max}}}={\xi }_{IL}\left(2-{\xi }_{IL}\right)+D{\xi }_{IL}{\left(2-{\xi }_{IL}\right)}^2$$

(4)

where D is a fitting parameter that considers the effect of high ion concentrations. For the 298 K data series, the fitting parameters were found to be σ_max = 1.423(33), c_max = 1.536(40), and D = 0.248(13). As mentioned, the discrepancies between Equations (1) and (2) are caused by the assumption made in Equation (2) that there is no interaction between IL and EC and negligence of complex morphologies caused by high ion concentrations due to interactions between ions, EC molecules and interactions between IL and EC [34].

The temperature also plays a key role in ionic conductivity. Typically, ILs follow the Vogel-Fulcher-Tamman (VFT) equation [16] as follows:

$$\sigma ={\sigma }_{\infty }\exp \left(-{\displaystyle \frac{{\mathrm{E}}_{\mathrm{a}}}{{\mathrm{k}}_{\mathrm{B}}\text{⋅}\left(\mathrm{T}-{\mathrm{T}}_0\right)}}\right)$$

(5)

where ${\sigma }_{\infty }$ is the ionic conductivity at infinity temperature, T₀ is the ideal glass-transition temperature or Vogel temperature, and E_a is the activation energy per molecule. The data series corresponding to x_IL = 1.0, 0.7, 0.5, and 0.15 are represented in Figure 5c and the VFT parameters for all samples are compiled in

Table 6. Fitting parameters of the VFT equation (Equation (5)) for all samples.

Sample	σ∞/S m−1	Ea/eV	T0/K
Mixtures (1 − x) EC + x [C3C1Pyrr][TFSI]
1	12.1 (17)	0.293 (18)	203 (13)
0.85	20.6 (59)	0.0352 (51)	186.0 (74)
0.7	19.8 (36)	0.0310 (30)	192.6 (46)
0.5	19.6 (44)	0.0279 (38)	191.1 (67)
0.4	19.8 (41)	0.0282 (37)	185.6 (68)
0.3	20.4 (42)	0.0285 (40)	179.7 (76)
0.25	13.0 (20)	0.0202 (25)	196.1 (57)
0.2	13.7 (32)	0.0223 (42)	186.9 (97)
0.15	11.8 (10)	0.0179 (14)	199.1 (34)
Ternary mixtures ([C3C1Pyrr][TFSI]/EC/LiTFSI)
0.2 m	30 (14)	0.041 (10)	171 (15)
0.5 m	9.5 (51)	0.0253 (82)	203 (14)
1 m	13.3 (74)	0.0320 (81)	203 (10)

.

Although the VTF equation fits the temperature dependence of the electrical conductivity of all the mixtures very well, the trend between VFT parameters and IL molar fraction is irregular, meaning that the VFT equation cannot describe the effect of IL concentration on ionic conductivity as also shown in other works [19,41].

Regarding ternary mixtures, their behaviour as a function of temperature is represented in Figure 5d. As expected, adding lithium salt to binary mixtures decreases ionic conductivity by increasing the viscosity of the system [37,38]. However, the addition of EC leads to an increase in conductivity of IL+lithium salt mixtures; remarkably, at 298 K, thus, the conductivity of ternary mixtures is more than twice that of the binary mixture without EC, as reported in previous studies [19] and comparable to other ternary mixtures with water [10,12,14], but avoiding the risks of water content in lithium batteries. The findings presented herein, in conjunction with the identified need for further investigation into their physicochemical properties, electrochemical performance, and electrodes compatibility suggest that these ternary mixtures possess considerable potential for application as next-generation electrolytes in advanced energy storage systems.

4. Conclusions

In this work, the electric and thermal properties of [C₃C₁Pyrr][TFSI] and its binary mixtures with EC, and ternary (from the eutectic composition) mixtures with LiTFSI were studied. The main conclusions of the research are the following:

• EC addition provokes the widening of transition peaks in DSC. In addition, melting and crystallisation take place in two steps, which indicates that salt addition and/or EC reduce, or remove, the crystalline behaviour. Binary mixture enthalpies increase with the addition of EC presenting lower values than the corresponding pure EC.
• The binary mixture 0.5 [C₃C₁Pyrr][TFSI] + 0.5 EC corresponds to the eutectic composition whose melting temperature is the lowest of all the studied concentrations.
• Salt addition to the eutectic composition removes the phase transition at high quantities of salt in the range from −80 to 100 °C.
• The values of conductivity of the [C₃C₁Pyrr][TFSI] + EC mixtures are higher than the pure IL and present a maximum at x_IL = 0.15.
• LiTFSI addition to the 0.5 [C₃C₁Pyrr][TFSI] + 0.5 EC mixture diminishes the conductivity; however, the values at room temperature are higher than the required 0.1 S·m⁻¹ for energy storage applications.

In summary, while further studies are required to fully assess their performance and stability under operational conditions, the results obtained here present the proposal of a new promising electrolyte, consisting of a ternary mixture of the ionic liquid ([C₃C₁Pyrr][TFSI]), ethyl carbonate, and lithium salt. This electrolyte exhibits an excellent liquid range, characterised by the absence of crystallisation peaks and high thermal stability, while maintaining high ionic conductivity.