Measuring and Enhancing the Ionic Conductivity of Chloroaluminate Electrolytes for Al-Ion Batteries

At the core of the aluminum (Al) ion battery is the liquid electrolyte, which governs the underlying chemistry. Optimizing the rheological properties of the electrolyte is critical to advance the state of the art. In the present work, the chloroaluminate electrolyte is made by reacting AlCl3 with a recently reported acetamidinium chloride (Acet-Cl) salt in an effort to make a more performant liquid electrolyte. Using AlCl3:Acet-Cl as a model electrolyte, we build on our previous work, which established a new method for extracting the ionic conductivity from fitting voltammetric data, and in this contribution, we validate the method across a range of measurement parameters in addition to highlighting the model electrolytes’ conductivity relative to current chloroaluminate liquids. Specifically, our method allows the extraction of both the ionic conductivity and voltammetric data from a single, simple, and routine measurement. To bring these results in the context of current methods, we compare our results to two independent standard conductivity measurement techniques. Several different measurement parameters (potential scan rate, potential excursion, temperature, and composition) are examined. We find that our novel method can resolve similar trends in conductivity to conventional methods, but typically, the values are a factor of two higher. The values from our method, on the other hand, agree closely with literature values reported elsewhere. Importantly, having now established the approach for our new method, we discuss the conductivity of AlCl3:Acet-Cl-based formulations. These electrolytes provide a significant improvement (5–10× higher) over electrolytes made from similar Lewis base salts (e.g., urea or acetamide). The Lewis base salt precursors have a low economic cost compared to state-of-the-art imidazolium-based salts and are non-toxic, which is advantageous for scale-up. Overall, this is a noteworthy step at designing cost-effective and performant liquid electrolytes for Al-ion battery applications.


ESI 1. Electrochemical cells
The two electrochemical cells used are schematically shown in Figure S1. The simultaneous CV/QCM measurements were done in a purpose-built cell, made from PEEK material, to physically fit the QCM resonator electrodes (Figure S1a-b); (a) is the main cylinder and (b) is the bottom cap. The total volume inside of the PEEK cylinder is 20 mL. Standard o-rings were used to prevent any liquid leaks and protect the QCM resonator electrode.
The EIS measurements were performed in a jacketed glass cell ( Figure S1c). The total volume inside of the jacketed cell is 70 mL. The schematic shows the two different liquid layers with the ILA electrolyte in a brown colour on the bottom and the less dense viscous paraffin oil layer on top in a grey colour. The two cylindrical wires represent the Al electrodes used in the broadband EIS measurements. The paraffin oil layer is not a conducting liquid, so its presence does not impact on the measured data.

ESI 2. EIS conductivity cell calibration
The ionic conductivity of a solution can be measured from broadband EIS data. Here we have used seven different conductivity standard solutions (i.e., 44,479, 11,419, 8,863, 1,249, 885, 442, and 74 µS cm -1 ; at 19 °C) to calibrate our homemade jacketed glass cell. Figure S2a shows a Bode plot of the impedance modulus as a function of frequency for the conductivity standard solutions. The plateau region at higher frequencies (i.e., 10 4 to 10 5 Hz) gives a visual estimation of the resistance, and we see the resistance decreases with more conductive solutions (i.e., the dashed black arrow points towards higher conductivities) as expected. The resistance values are quantitatively extracted from fitting EIS data to an electrochemical equivalent circuit. Here the data fit very well to a simple R-CPE model (see colour matched solid lines in Figure S2a; inset shows equivalent circuit), which consists of a resistor (R) in series with a constant phase element (CPE) to represent a non-ideal capacitor. The cell constant (K; cm -1 ), which is dependent on the geometrical arrangement of the electrodes and electrode surface area, was determined from resistance (R; Ω) measurements of known conductivity (σ; S cm -1 ) standard solutions using the equation σ = K/R. Specifically, the gradient of a plot of σ versus 1/R will yield the cell constant. From this calibration in Figure S2b we find a cell constant K = 0.377 (± 0.002) cm -1 that can be used to measure conductivities of other electrolyte solutions.

ESI 3. Mole ratio dependent viscosity data
Electrolyte viscosity values were estimated from QCM data as shown previously. 1 In Figure S3 we show the effect the mole ratio of LA:LB on the viscosity. The AlCl 3 :Acet-Cl 2:1 liquid shows the lowest viscosity at 37 ± 4 cP. At mole ratios both below and above 2:1 (LA:LB) we can see the viscosity increases, and a sharper increase is observed at more Lewis acidic ratios. The error (standard deviation) in the measured data also increases as we move away from the 2:1 mole ratio. Overall, we find the 2:1 mole ratio to be the optimal composition from our tests. Figure S3. Mole ratio dependent viscosity data for the AlCl 3 :Acet-Cl system.
The i-E curve fitting method was used to investigate several other ILA electrolytes prepared in a similar way. Data in Figure S4 show overlaid experimental CV traces and corresponding i-E fits for (a) AlCl 3 :EMIM-Cl, (b) AlCl 3 :acetamide, and (c) AlCl 3 :urea. All formulations use a 1.50:1 mole ratio. All CVs show linear i-E behaviour in the anodic scan with fitting operable over >1.5 volts. The AlCl 3 :EMIM-Cl 1.50:1 liquid in Figure S4a shows similar behaviour to our AlCl 3 :Acet-Cl 2:1 liquid but the higher currents yield a higher conductivity of 19.7 mS cm -1 . The AlCl 3 :acetamide ( Figure S4b) and AlCl 3 :urea ( Figure S4c)