Experimental Study on Cycle Aging of 3.4 Ah Lithium-Sulfur Pouch Cells: Temperature and Current Investigation

Contents lists available at ScienceDirect Chemical Engineering Journal journal homepage: www.elsevier.com/locate/cej Experimental study on cycle aging of 3.4 Ah lithium–sulfur pouch cells: Temperature and current investigation Dominika Capkovaa,b,∗, Tomas Finsterle c, Kevin M. Ryan a, Tomas Kazda b, Daniel-Ioan Stroed, Vaclav Knapc aDepartment of Chemical Sciences and Bernal Institute, University of Limerick, Limerick, V94 T9PX, Ireland bDepartment of Electrical and Electronic Technology, Faculty of Electrical Engineering and Communication, Brno University of Technology, 616 00, Brno, Czech Republic cDepartment of Electrotechnology, Czech Technical University in Prague, 166 27 Prague, Czech Republic dDepartment of Energy, Aalborg University, 9220 Aalborg East, Denmark A R T I C L E I N F O Keywords: Lithium–sulfur pouch cell Battery degradation Cycle aging Temperature influence C-rate influence A B S T R A C T High energy density sulfur cathodes are among the most promising alternatives to conventional intercalation cathodes for next-generation lithium-ion batteries. However, the practical implementation of lithium–sulfur (Li–S) systems is limited by rapid capacity fade and poor cycling stability. These issues are primarily driven by the polysulfide shuttle effect, wherein soluble higher lithium polysulfides, generated at the high voltage discharge plateau, migrate between the electrodes, resulting in active material loss. In an attempt to approach the commercial application of Li–S batteries, an in-depth investigation of pouch cells under different conditions is inevitable. This study focuses on the cycle aging of pre-commercial 3.4 Ah Li–S pouch cells at different temperatures and current rates using non-destructive techniques. The most negative effect on the performance of the Li–S battery cell is a low temperature of 10 ◦C and 50 ◦C. From the perspective of different charging and discharging currents, the reduced battery lifetime was observed for fast charging at 0.2 C and 0.3 C. The internal resistance increased with the degradation of the battery cell and is more pronounced in the low voltage plateau. To maximize the cycle life of the Li–S batteries, the optimal cycling conditions are at around 30 ◦C, charging at 0.1 C and discharging at 0.2 C. 1. Introduction High energy density electrochemical energy storage devices are attracting increasing interest in the technology field due to their widespread applications, e.g., electric vehicles, portable electronic devices, and aerospace [1–4]. Currently, the gravimetric energy density of lithium-ion (Li-ion) batteries is about 200–300 Wh kg−1 at the cell level, depending on material chemistry [5,6]. However, to address industry demands, energy density higher than 500 Wh kg−1 has to be reached in the near future [6–8]. Incorporation of innovative electrode materials such as sulfur can elevate the energy density to and beyond the required limits [9]. Sulfur as a cathode material has several advantages such as high abundance, low price, non-toxicity, which will enable its future largescale application in batteries and use in transportation and energy storage [10]. The main advantage of sulfur is its high theoretical capacity of 1675 Ah kg−1 based on two-electron conversion reaction [11]. Metallic lithium possesses a theoretical specific capacity of ∗Corresponding author at: Department of Chemical Sciences and Bernal Institute, University of Limerick, Limerick, V94 T9PX, Ireland. E-mail address: [email protected] (D. Capkova). 3860 Ah kg−1 and low electrochemical potential (−3.04 V vs standard hydrogen electrode) [12]. The coupled lithium–sulfur (Li–S) battery can achieve a theoretical energy density of 2500 Wh kg−1 with an average cell voltage of 2.1 V [13,14]. However, Li–S batteries are still in the pre-commercial stage due to several fundamental issues. The conductivity of sulfur is low (5 × 10−30 S cm−1) and insulating characteristics limits the redox kinetics [15]. The different density of sulfur and the discharge product Li2S cause a volumetric expansion of ∼ 80%, leading to pulverization of the electrode [16]. The sulfur reduction process generates higher polysulfides with a long chain (Li2S8, Li2S6, Li2S4) and subsequently lower polysulfides with a short chain (Li2S2, Li2S). Higher polysulfides might dissolve in the electrolyte, followed by their shuttling between electrodes, resulting in a loss of active material, fast capacity fade, and low Coulombic efficiency [17,18]. Several strategies have been applied to mitigate negative issues related to sulfur cathode material, e.g., porous carbons [19,20], metal oxides [21,22], and metal–organic https://doi.org/10.1016/j.cej.2025.170341 Received 30 July 2025; Received in revised form 16 October 2025; Accepted 29 October 2025 Chemical Engineering Journal 525 (2025) 170341 Available online 1 November 2025 1385-8947/© 2025 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY license ( http://creativecommons.org/licenses/by/4.0/ ). D. Capkova et al. frameworks (MOFs) [23,24]. However, most studies focus on lab-scale material analysis in coin cells, while the behavior of scaled-up pouch cells may be different [25–27]. Pouch cells have attracted significant attention due to their enhanced applicability in real-world implementation compared to coin cells. In order to fill in the gap and accelerate the commercialization of Li–S batteries, investigation of high capacity Li–S pouch cells is essential. Studies of Li–S pouch cells are still in an early stage compared to fundamental research on Li–S batteries. Manthiram et al. [28] studied the MOF-modified separator in Li–S pouch cells. The cell without separator modification failed after 3 cycles and the cell with the MOFmodified separator maintains a capacity of ∼ 75% after 30 cycles. The application of a macroporous catalytic cathode with double-end binding sites in Li–S pouch cells delivered ∼ 74% capacity retention after 80 cycles [29]. A three-way electrolyte (TWE) with ternary solvents composed of 1,2-dimethoxyethane (DME), di-isopropyl sulfide (DIPS), and 1,3,5-trioxane (TXA) were employed to mitigate sulfur side reactions [30]. The pouch cells constructed with TWE showed a capacity retention of ∼ 65% after 27 cycles. Sulfur nanoparticles were synthesized and implemented in a pouch cell. The optimized sulfur nanoparticles delivered a capacity retention of ∼ 71% after 150 cycles [31]. The utilization of 2,2,6,6-tetramethylpiperidinyloxy (TEMPO) as a mediator to suppress the lithium dendrites and accumulate dead lithium during cycling extended the life of high-loading Li–S pouch cells to 70 cycles [32]. The application of a bifunctional separator based on intrinsic microporosity polymer demonstrated 40 cycles with high sulfur loading and lean electrolyte [33]. Most studies focus on pouch cells up to 1 Ah and investigation of high capacity pouch cells are reported only in several studies. The failure analysis of 4 Ah pouch cells has been performed, where the cells crossed the end-of-life (EOL) (< 80% of initial capacity) after 16 cycles [34]. The performance of 6 Ah Li–S pouch cells was investigated, in the fourth cycle cell failure was observed, and a low voltage plateau was not observed [35]. 19 Ah Li–S pouch cells were analyzed, however, they reached EOL after 60 cycles [36]. In addition, data from this study were used for aging analysis and state-of-health (SOH) estimation. There were several studies [37–40] focused on state-of-charge (SOC) estimation and equivalent circuit network model parameterization of 3.4, 12, 14, and 19 Ah Li–S pouch cells. The models for Li–S cells to predict features during charge and discharge process were presented by Offer et al. [41–43]. Moreover, in our previous research, we have investigated 3.4 Ah Li-pouch cells during calendar aging at different temperatures and various states of discharge and analysis of the same cells using electrochemical impedance spectroscopy [44,45]. Herein, pre-commercial 3.4 Ah Li–S pouch cells are in-depth investigated during cycling aging using non-destructive techniques for battery degradation diagnostics. This work extends our prior investigation on the calendar aging mechanisms in Li–S pouch cells [44]. The cells were cycled at five different temperatures, and the influence of five different charge/discharge currents was studied. The total capacity, capacity of high and low voltage plateaus, and Coulombic efficiency are assessed in depth. In order to perform in-depth analysis of these cells, incremental capacity analysis was performed and the peak characteristics were studied. The shuttle current was measured and its evolution was monitored under various degradation conditions. Cell impedance was studied by electrochemical impedance spectroscopy (EIS) and direct current (DC) resistance was obtained from current pulses. The extended cycle life of LI-S pouch cells was observed for cells cycled at 30 ◦C and charged/discharged at 0.1 C/0.2 C. The most detrimental impact on cell performance had a high charging current of 0.2 C and 0.3 C, as well as low and high (10 ◦C and 50 ◦C) temperatures. 2. Methodology Pre-commercial long-life Li–S pouch cells (OXIS Energy) are used in this study with a rated capacity of 3.4 Ah. They possess a sandwich Table 1 Specifications of the prototype Li–S pouch cell. Parameter Value Nominal Capacity 3.4 Ah Nominal Voltage 2.05 V Maximum Voltage 2.45 V Minimum Voltage 1.5 V Nominal Charging Current 0.34 A (0.1 C) Nominal Discharging Current 0.68 A (0.2 C) Temperature Operation Range +5 ◦C to +80 ◦C Weight Approx. 50.7 g Nominal dimension 145 mm × 78 mm × 5.6 mm Specific Energy Density 137.5 Wh kg−1 Volumetric Energy Density 114.4 Wh L−1 Fig. 1. Illustration of applied measurement methodology (a) and RPT procedure (b). structure where sulfur, carbon, and binder on the current collector form the cathode, lithium metal is used as the anode, and the separator is saturated by a sulfolane-based electrolyte. Specifications of investigated 3.4 Ah Li–S pouch cells are presented in Table 1. The details of applied non-destructive electrochemical techniques (incremental capacity analysis (ICA), differential voltage analysis (DVA), shuttle current, and electrochemical impedance spectroscopy (EIS)) are described in our previous study [44]. The methodology for assessing cycling aging consists of the cycling test and a reference performance test (RPT), which was performed at 30 ◦C periodically after every 20 cycles, as shown in Fig. 1(a). The RPT procedure was followed based on [46] and it is illustrated in Fig. 1(b). The methodology of the RPT test was as follows: 1. Pre-condition cycle to reset ’the cumulative history’ of the cell; 2. The measurement of actual charge and discharge capacity; 3. A set of charging and discharging pulses at various DOD/SOC levels to obtain the resistance of the cell; 4. For every even RPT measurement of the shuttle current or EIS. The investigated cells were cycled in a voltage window from 1.5 V to 2.45 V (or charging time limit of 11 h). The first group of Li–S Chemical Engineering Journal 525 (2025) 170341 2 D. Capkova et al. Table 2 Cycling conditions of investigated Li–S pouch cells for aging. Temperature [◦C] 10 20 30 40 50 C-rate CHA DCH 0.1 0.2 X X X X X 0.1 0.5 X 0.1 1.0 X 0.2 0.2 X 0.3 0.2 X pouch cells was cycled at five different temperatures of 10, 20, 30, 40, and 50 ◦C using a charging current of 0.1 C and a discharging current of 0.2 C. The second group of cells was tested using a consistent temperature of 30 ◦C and different charge/discharge currents. The testing conditions are stated in Table 2. Each test case was performed on two Li–S cells to demonstrate the reproducibility of the experiment. Cell cycling and RPT measurements were performed on the Digatron BTS 600 battery tester. FuelCon Evaluator Battery Test Station was used for the EIS measurements. During the RPTs and cycling aging, the cells were stored in a controlled temperature environment in thermal chambers. The reference temperature for the RPTs was 30 ◦C. Before starting the test, cells were allowed to thermo-dynamically stabilize for two hours. The evolution of capacity and Coulombic efficiency at different temperatures and charge/discharge currents is discussed in Section 3. The data interpreted in this study are obtained from RPT measurements. The DVA analysis presented in this study (Section 4 is based on [47]. Original data were carefully denoised using the Savitzky-Golay filtering method (settings: 3rd order and 21 sample window for voltage, 1st order, and 101 sample window for capacity) and moving average (with windows of 3 samples and 2 samples for voltage and capacity, respectively). To obtain a DV curve, a derivative calculation of voltage to charging capacity was performed. The methodology for the direct shuttle current measurement, which is described in Section 5, is based on [48]. The cells were fully charged and left in an open-circuit state until OCV value was reached. Subsequently, the cells were maintained in a constant voltage charging mode at the detected OCV, while the current was monitored for two hours to determine the steady-state value, referred to as the shuttle current. Internal resistance and EIS measurements were investigated in detail and are evaluated in Section 6. Internal resistance was measured during discharging using a current pulse at 0.2 C. The EIS measurements were conducted over a frequency range of 10 kHz to 10 mHz using a voltage amplitude of 3 mV. The EIS spectra were fitted using MATLAB software with the ‘Zfit’ function [49]. An equivalent electrical circuit, shown in Fig. 2, was proposed to analyze the electrochemical behavior of Li–S batteries. The details of the applied circuit and its fitting process can be found in [45]. The EIS spectra consist of two semicircles (SC) and an inclined line a low frequency region. In this model, a constant phase element (CPE) replaces the capacitor to account for the system’s non-ideal behavior. The electrolyte resistance, denoted as Re, represents the contribution of ohmic resistance. The first semicircle in the spectrum is characterized by Rint and CPEint , corresponding to interphase contact resistance and the capacitance within the sulfur electrode bulk. The second semicircle represents charge transfer resistance and double-layer capacitance at the electrode surface, described by Rct and CPEdl. Finally, the straight-line segment at the spectrum’s end represents the diffusion process, modeled by Rdif f , CPEdiff , and the Warburg element (W). The EIS spectra are available maximally up to 100 cycles, due to the limited availability of the measurement device in the laboratory. As part of this study, the evolution of cell resistance under different operating conditions was investigated, while the other elements of the equivalent circuit model were not analyzed in detail due to the substantially higher interpretive value of resistance related to battery degradation. Fig. 2. Equivalent electrical circuit used for fitting the experimental EIS data and an example of Nyquist plot of the Li–S pouch cell at 70 % DOD at the beginning of life. 3. Capacity and coulombic efficiency evolution The normalized capacity over cycling at 0.1 C/0.2 C at different temperatures is illustrated in Fig. 3(a). The cells cycled at a high temperature of 50 ◦C showed the most significant capacity decline only after 40 cycles. The fast capacity decrease continued until the endof-test (EOT), which was achieved after 220 cycles with a capacity retention of around 40%. Another fast EOT with low capacity retention was achieved with cells cycled at a low temperature of 10 ◦C. The capacity retention of around 40% was obtained after 160 cycles. The longest cycle life was observed for cells cycled at 30 ◦C. The capacity retention was up to ∼70% until 340 cycles, and after 400 cycles it was around 40%. The dependence of the C-rate during charging and discharging at 30 ◦C is shown in Fig. 3(b). The influence of charge/discharge currents was analyzed using stable charge or discharge current, respectively. The cells were charged by 0.1 C and discharged by 0.2, 0.5, and 1 C. The performance of the cells was similar up to 120 cycles. However, after 120 cycles, the cells discharged by 1 C reached their EOT, and it was not possible to perform more cycles. The cells discharged by 0.5 C survived 160 cycles with a capacity retention of 60%. The best performance was observed for cells discharged by 0.2 C, where the cycle life was significantly improved to 340 with a capacity retention of 70% and 400 cycles with a capacity retention of 40% in total. In order to investigate the influence of different charge currents, the cells were charged at 0.1, 0.2, and 0.3 C and discharged at 0.2 C. It was found that a high charging rate (0.2 C or 0.3 C) has a stronger influence on the cell performance compared to a high discharging rate. The cells charged at 0.2 C and 0.3 C reached their EOT only after 60 and 40 cycles, respectively, with a capacity retention of around 45%–30%. While a slow charging rate at 0.1 C achieved 360 cycles with a capacity retention of around 70%. The normalized discharge capacity for the first cycle is shown in Fig. S1 and varies significantly with temperature, discharge rate, and charge rate. At lower temperatures or higher C-rates, reduced ion mobility and increased polarization lead to lower capacities, while moderate temperatures and slower rates enhance electrochemical kinetics and improve performance. At high temperatures, faster polysulfide diffusion and increased reaction kinetics can accelerate side reactions and the growth of interfacial resistance, leading to earlier capacity loss and faster EOT. Chemical Engineering Journal 525 (2025) 170341 3 D. Capkova et al. Fig. 3. Evolution of (a) total capacity, capacities obtained from (c) high and (d) low voltage plateaus, and (b) Coulombic efficiency for various temperatures. Conversely, at low temperatures, sluggish lithium-ion transport and slower sulfur redox kinetics result in incomplete sulfur utilization and heterogeneous Li2S deposition, which also contributes to performance decline. Similarly, high charge/discharge currents increase polarization, reduce reaction efficiency, and exacerbate electrode passivation, all of which shorten the effective cycle life. Other relevant parameters are analyzed in the following sections. In an effort to perform an in-depth analysis of the capacity behavior of the Li–S cells during aging, the discharge curves were closely examined by separation of the capacity of high (∼2.38 V) and low voltage (∼2.03 V) plateaus (see Fig. 3c-f). The high voltage plateau describes the reduction of sulfur to higher (long-chain) polysulfides, and the low voltage plateau is attributed to the further reduction to lower (shortchain) polysulfides. At the beginning of the test, all cells have the ratio between high and low voltage plateau around 30:70. However, with battery degradation the ratio is changing, and the capacity of the low voltage plateau is decreasing more significantly than the high voltage plateau. This phenomenon is visible for cells cycled at 10 and 20 ◦C, and even more significant differences in the ratio are for high-rate charging, where the ratio for both plateaus is around 50:50, which corresponds to the dramatic decrease in capacity. The low voltage plateau capacity usually represents 70% of the total capacity. As the battery is degrading, the capacity drop for the low voltage plateau is more notable then for the high voltage plateau due to a higher contribution to the total capacity. For the rest of the cells, the ratio between plateaus is stable until they reach very low capacities (< 0.5). Coulombic efficiency can evaluate the performance and stability of the battery cycling and is the ratio of the discharge capacity to the charge capacity expressed in a percentage. Its behavior for the analyzed Li–S pouch cells is presented in Fig. 3g,h. Coulombic efficiency is quite stable at around (96% and 98%) for cells aged at lower temperatures (10 and 20 ◦C). For cycling temperatures higher than 30 ◦C, Coulombic efficiency decreases down to 70%–60% up to 100 cycles. Afterwards, it stabilizes for the cells cycled at 30 ◦C. This behavior can be attributed to the increasing shuttle effect when cycling the battery at higher temperatures. However, when the cells approach the EOT and low capacities, Coulombic efficiency is increasing due to shortened low voltage plateau and overall discharge curve. Regarding the impact of Chemical Engineering Journal 525 (2025) 170341 4 D. Capkova et al. Fig. 4. Charging curve at 0.1 C and the corresponding DV curve at the beginning of the life of the Li–S pouch cell at 30 ◦C. C-rate on the Coulombic efficiency, the most notable decrease was observed for cells charged at 0.3 C and discharged at 0.2 C. However, a slow charge at 0.1 C and fast discharge at 1 C showed the highest values of Coulombic efficiency due to the short discharge time during cycling suppressing side reactions (shuttle effect, passivation layer). With the decrease of the discharge rate to 0.5 C and 0.2 C, Coulombic efficiency slightly decreased up to 100 cycles. Separate images for each condition are shown in Fig. S2-S6. 4. Differential voltage analysis (DVA) The battery degradation during aging were further investigated using differential voltage analysis (DVA) to identify loss of active material. DVA was performed on charging curves due to better control of the charging process than discharging during the real-life battery operation. Fig. 4 illustrates the charging curve and corresponding DV curve. DV curve contains a peak around 2.35 V, representing the sharp rise part between the plateaus and dividing the DV curve into two regions. The first region describes the low voltage plateau, and the second part corresponds to the high voltage plateau. This peak was fitted using an open code ‘peakfit’ in MATLAB [50] to quantitatively analyze the results and investigate the changes in peak height, position, and width. The peak width was analyzed at half the height. Moreover, the voltage kink around 2.2 V at the beginning of the charging process is presented in a DV curve as a loop. This loop was not further analyzed, we focused only on peak investigation as it is related to voltage plateaus. The peak changes in position, height, and width were analyzed for all investigated temperatures and charge/discharge rates (see Fig. 5a-f). The peak position can initially decrease, but in long term, with battery degradation, it shifts towards a higher potential (see Fig. 5a,b). For the temperature test, the fastest EOL was reached for 10 ◦C followed by 50 ◦C. This trend is visible in a peak shift towards the higher potential of about 2.39 V. Moreover, the peak shift for this case was observed much earlier than that for the rest of the cells. The peak position higher than 2.35 V was reached only after 40 cycles for cells cycled at 10 ◦C, while for cells at 50 ◦C it was 180 cycles, and for cells at 30 ◦C it was after 380 cycles. For different discharge current rates, a similar increase in peak position was observed for discharge at 0.5 C and 1 C at 120 and 140 cycles, respectively, compared to 380 cycles for 0.2 C. The fast charging at 0.3 C showed peak shift only after 20 cycles, where the capacity was only 70%. When charging at 0.2 C, the capacity dropped to 70% after 40 cycles, and the peak shift to higher potentials was also visible. Thus, the peak position appears to be a good health indicator for the LiS cells. The fastest degradation for the temperature tests was for cells cycled at 10 ◦C, where the peak was the shortest and widest of all tested cells. However, the peak height was stable almost to the EOT, and the peak width was increasing with the battery aging. Different peak evolution was for the second fastest reached EOT for cells cycled at 50 ◦C. The peak was the tallest and narrowest of all tested cells. The peak height continuously increased for one cell, and the other increased at the beginning and remained stable until the EOT. The peak width was similar for both cells and slowly increased with the battery degradation. The cells with the longest lifetime, cycled at 30 ◦C, showed a peak with the height and width somewhere in between the cells cycled at 10 and 50 ◦C. Thus, the peak height and width indicate different degradation mechanisms across temperature conditions. The current influence was not very significant for the peak height, but there were observed some changes for the peak width. The most significant changes were for the fast charging, the peak width increased. Overall, with the battery degradation, the DV peak width increased, and various degradation mechanisms led to different peak changes. The observed shift in peak positions is interpreted as a consequence of progressive loss of electrochemically active sulfur and increased electrode polarization, likely due to the growth of interfacial resistance and the accumulation of insulating passivation layer on the cathode. These effects reduce the kinetics of sulfur redox reactions, increase overpotential, and lead to decreased reversibility and capacity fading over cycling [51,52]. The degradation of the DVA peak is more significant high/low temperatures and high charge current which correlates with capacity degradation (Fig. 3). Individual images for each condition are presented in Fig. S7-S11. 5. Evaluation of shuttle current To investigate the polysulfide shuttling, the shuttle current was analyzed for cells cycled at different temperatures and current rates for charging and discharging. The interpolated value of the shuttle current at 2.4 V is depicted in Fig. 5g,h. The evolution of the shuttle current is in correlation with Coulombic efficiency (Fig. 3g,h). The shuttle current is stable for cells cycled at lower temperatures (10 and 20 ◦C). Based on this observation, it can be concluded that the shuttle effect did not cause accelerated battery degradation for cells cycled at 10 and 20 ◦C. The electrolyte viscosity is higher at low temperatures, leading to slow ionic transport and polysulfide migration. For temperatures of 30, 40, and 50 ◦C, the shuttle current increases up to ∼100 cycles and then slowly decreases. The shuttle current decrease can be attributed to the fully saturated electrolyte with higher polysulfides [53]. The maximum amount of polysulfides in the electrolyte was reached with an increasing shuttle current, and subsequently, the shuttle current decreased. The shuttle current evolution for different discharging/charging currents is shown in Fig. 5h. The discharge currents of 0.2 C and 0.5 C caused an initial increase in the shuttle current up to 100 cycles. However, for the cell discharged by 1 C, the shuttle current increase is much less significant than for 0.2 C and 0.5 C, probably due to reduced discharge time and mitigation of polysulfide shuttle by usage of a high current. The shuttle current evaluation for different charging is strongly affected by a short cell lifetime, and it is not possible to analyze the shuttle current of these cells. Chemical Engineering Journal 525 (2025) 170341 5 D. Capkova et al. Fig. 5. (a)–(f) Peak parameter evaluations from DVA for the cells cycled at different temperatures and charging/discharging currents. (g),(h) Interpolated shuttle current for 2.4 V for all investigated pouch cells. 6. Internal resistance and electrochemical impedance spectroscopy (EIS) The measurement of internal resistance by current pulses at 0.2 C was performed to analyze the evolution of the resistance during the battery lifetime. The obtained resistance for a 1-s discharge at a high (10% DOD) and low (70% DOD) voltage plateau is shown in Fig. 6. In general, the resistance increases with battery degradation which may be caused by electrode cracking, loss of electrical connectivity between particles, the formation and growth of Li2S insulating layer, and the electrode material delamination for the current collector [54]. In addition, the resistance is higher in the low voltage plateau compared to the high voltage plateau due to the formation of lower insoluble polysulfides, which have insulating characteristics [55]. Faster reaction kinetics at 50 ◦C leads to the formation of a dense and inhomogeneous passivation layer on the cathode, which is electronically insulating and can block active sides leading to an increase in impedance. In a contrast, the cycling temperature of 10 ◦C leads to sluggish Li+ transport and slower polysulfide conversion rates lead to non-uniform and incomplete Li2S deposition. Instead of forming a continuous layer, Li2S precipitates in isolated islands or clusters, leaving parts of the cathode surface underutilized. This heterogeneity reduces active sulfur utilization, increases local current density, and contributes to capacity loss over repeated cycles [56]. On the anode site, high temperature (50 ◦C) can lead to decomposition and dissolution of the SEI layer. The resulting SEI layer becomes thicker, more heterogeneous, and less stable. These compositional changes reduce Li+ transport efficiency and lead to continuous lithium consumption, contributing to impedance growth and capacity fading. Reduced temperature (10 ◦C) slows the formation and growth of the SEI layer and limits the mobility of Li+ within it. The resulting SEI is often less uniform and thinner, which can temporarily reduce parasitic reactions but may also be mechanically fragile. Over cycling, polysulfide crossover combined with uneven SEI formation can lead to localized lithium deposition, intermittent SEI breakdown, and gradual impedance growth [57,58]. The temperature influence on the resistance in the high voltage plateau was negligible; the resistance increased only for the cells at Chemical Engineering Journal 525 (2025) 170341 6 D. Capkova et al. Fig. 6. Measured internal resistance for a 1-s discharge with a 0.2 C current at different (a) temperatures and (b) charge/discharge currents. the EOT (see Fig. 6a). The resistance in the low voltage plateau increases faster for cells cycled at 10 ◦C, which may be caused by higher electrolyte viscosity and decreased binder conductivity at low temperatures [59]. Low voltage plateau is attributed to the insulating lower polysulfides they might form a growing insulating layer and it can be visible as a resistance increase with battery degradation. This trend is also visible in our results (Fig. 6c). Different discharge currents did not show significant differences in resistance between cells (Fig. 6b). However, the fast charging current (0.2 C and 0.3 C) caused a significant increase in the resistance, see Fig. 6b,d. The resistance was more than 2 times higher at the EOT, 20-40 cycles for high current and 380 cycles for low current. This observation indicates a loss of electrical conductivity in the electrode material, probably due to cracking caused by high charging current, large volumetric changes, and the instability of the lithium anode under high polarization conditions. During fast charging, Li+ are deposited rapidly onto the anode, which can lead to non-uniform lithium plating, dendrite formation, and localized SEI breakdown. These effects accelerate parasitic reactions with the electrolyte, consume active lithium, and increase cell impedance, contributing to faster capacity fade [60]. Images for each condition are provided separately in Fig. S12-S16. EIS measurements were conducted at various DODs during the RPT process. In this study, we focus on the EIS spectra at 15% and 70% DOD to compare the resistances associated with the high and low voltage plateaus, in a manner analogous to the internal resistance assessment from current pulse measurements. The evolution of the EIS spectra at different temperatures and C-rates during aging are shown in Fig. 17–26. The impact of aging conditions on the evolution of calculated resistances across different temperatures is illustrated in Fig. 7 and charge/discharge rates in Fig. 8. The electrolyte resistance R𝑒, blue line Figs. 7and 8, increases with battery degradation due to higher electrolyte viscosity caused by dissolved higher polysulfides. In general, R𝑒 is slightly higher in the high voltage plateau (10% DOD) where higher polysulfides are formed. From the temperature point of view, the highest electrolyte resistance was obtained for the lowest temperature of 10 ◦C due to an additional increase of the electrolyte viscosity caused by low temperature. The most significant influence of the R𝑒 from the current perspective had a high charging current of 0.2 C and 0.3 C. These cells reached their EOT only after 60 and 40 cycles, and the electrolyte resistance increased to ∼150 and ∼250 mΩ, respectively. The orange line in Figs. 7and 8 represents interphase contact resistance R𝑖𝑛𝑡, which is slightly higher for the low voltage plateau compared to the high voltage plateau. This might be caused by an insulating characteristic of lower polysulfides, Li2S precipitation, and the formation of a passivation layer [61]. R𝑖𝑛𝑡 appears to be fairly stable for different temperatures, as no significant capacity variation was observed. A high charging current caused a significant increase in the R𝑖𝑛𝑡, probably due to electrode cracks and an increasing insulating passivation layer on the electrode surface with enhanced cell degradation. The charge transfer resistance (R𝑐𝑡) is represented with a yellow line in Figs. 7and 8 is lower in the high voltage plateau compared to the low voltage plateau, attributed to the electrochemical reduction of elemental sulfur to soluble higher polysulfides, which facilitates more rapid charge transfer kinetics in the high voltage region. The increase in (R𝑐𝑡) in the low voltage plateau may hinder the reduction kinetics of higher polysulfides, leading to decreased formation of soluble intermediates and the preferential precipitation of insoluble species such as (Li2S2 and Li2S), which contribute to the formation of a passivation layer. The growth of the passivation layer may result in an increase of the (R𝑐𝑡). The highest (R𝑐𝑡) was reached for the cells cycled at the temperature of 10 ◦C due to slower electrochemical kinetics and thicker SEI layer at low temperatures. A high discharging current of 1 C and charging current of 0.3 C increased the R𝑐𝑡. The R𝑐𝑡 increase using high currents during cycling can be caused by high overpotential or SEI layer [62,63]. The continuous deposition of poorly soluble low polysulfides on the cathode surface leads to the formation of a blocking layer, which is expressed by diffusion resistance [55]. The diffusion resistance (R𝑑𝑖𝑓 𝑓 ) is shown in Figs. 7and 8 using green color and is higher in the low voltage plateau compared to the high voltage plateau due to suppressed diffusion of higher polysulfides in the high voltage plateau. The higher values of R𝑑𝑖𝑓𝑓 in the low voltage plateau could be caused by the formation of lower polysulfides and an insulating blocking layer on the Chemical Engineering Journal 525 (2025) 170341 7 D. Capkova et al. Fig. 7. Resistance evolution in the high and low voltage plateau for cells cycled at different temperatures. electrode surface. R𝑑𝑖𝑓𝑓 was the highest for 10 ◦C and 50 ◦C as the degradation of these cells was the most significant. Similar behavior was observed for cells discharged using 1 C and charged 0.2 C and 0.3 C. The percentage change of resistances to the beginning-of-test (BOT) up to 100 cycles is summarized in Table S1 for cycling at different temperatures and Table S2 for cycling at different charge/discharge C-rates. The most significant percentage change after the first cycling round was for the cell cycled at 50 ◦C, and subsequent cycling increased the percentage change compared to BOT. However, the percentage change was even more significant for the cell charged/discharged by 0.3 C/0.2 C. These statements are in agreement with capacity decrease, a more pronounced capacity decline can lead to high resistance increase. 7. Conclusions Research on Li–S pouch cells remains in its early stages relative to the extensive fundamental studies on Li–S battery chemistry. To accelerate the successful commercialization of Li–S batteries, the cycling conditions for extended cycle life must be investigated on the scale of pre-commercial and commercial pouch cells (capacities > 1 Ah), as certain degradation phenomena are not observable in laboratory-scale formats such as coin cells or hand-assembled pouch cells with capacities below 1 Ah. This study is a continuation of our previous exploration focused on the calendar aging of Li–S pouch cells [44]. As far as the authors are aware, the in-depth analysis of the large-scale Li–S pouch cell degradation has not been reported. Therefore, this study should fill in the gap and contribute to the way forward toward a successful commercialization of Li–S batteries. The aging of pre-commercial 3.4 Ah Li–S pouch cells was thoroughly examined during the battery cycle life using non-destructive electrochemical techniques. To identify the optimal cycling conditions, different temperatures (10, 20, 30, 40, and 50 ◦C) and various combinations of charge/discharge currents (0.1 C/0.2 C, 0.1 C/0.5 C, 0.1 C/1 C, 0.2 C/0.2 C, and 0.3 C/0.2 C) were investigated. The main challenges related to battery degradation are the shuttle effect, high electrolyte viscosity at low temperature, and high resistance. These phenomena are likely to be caused by the formation and growth of a passivation layer from insulating discharge products, electrode cracks, and deteriorated interconnection of particles. The correlation between the capacity fading, increasing resistance, and shifted DV peak to higher potentials was observed, as well as in decreasing Coulombic efficiency and increasing shuttle current. Chemical Engineering Journal 525 (2025) 170341 8 D. Capkova et al. Fig. 8. Calculated resistance values in the high and low voltage plateau for cells cycled using various charge/discharge currents. The cycle life of Li–S pouch cells could be extended by operating them at optimal temperature (30 ◦C) and using the less harmful C-rate of 0.1 C for charging and 0.2 C for discharging. Using optimal conditions, the cycle life was extended to ∼340 cycles with > 70% capacity retention. The battery aging was accelerated using a low temperature of 10 ◦C and 50 ◦C and using a high charge current of 0.3 C or 0.2 C and a discharge current of 1 C. The fast capacity fade at low temperature could be caused by high electrolyte viscosity, the insulating passivation layer, and increased resistance. Cycling at a high temperature of 50 ◦C might accelerate side reactions as the activation energy is lower. A high charging current was demonstrated as the most powerful method to destroy the Li–S rapidly. The capacity decreased to ∼70% of the initial capacity only after 20 cycles using a charging current of 0.3 C and ∼40 cycles for 0.2 C. Investigating the aging of Li–S pouch cells is essential for the next step towards bringing Li–S cells to market. The analysis and diagnosis of large-scale cells is very time demanding due to tedious experiments in an effort to perform in-depth electrochemical analysis. This study can potentially serve as a reference for the design of experimental procedures and can be expanded upon by further studies to investigate large-scale battery cells. CRediT authorship contribution statement Dominika Capkova: Writing – original draft, Visualization, Project administration, Investigation, Funding acquisition, Data curation, Conceptualization. Tomas Finsterle: Formal analysis, Data curation. Kevin M. Ryan: Writing – review & editing, Supervision, Project administration, Funding acquisition. Tomas Kazda: Writing – review & editing, Supervision, Project administration, Funding acquisition. Daniel-Ioan Stroe: Writing – review & editing, Supervision, Resources, Methodology, Conceptualization. Vaclav Knap: Writing – review & editing, Supervision, Project administration, Methodology, Funding acquisition, Data curation, Conceptualization. Declaration of competing interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper. Acknowledgments The authors would like to thank OXIS Energy for providing lithium– sulfur battery cells. D.C. acknowledges the EU Horizon 2023 research Chemical Engineering Journal 525 (2025) 170341 9