Impact of the Sulfurized Polyacrylonitrile Cathode Microstructure on the Electrochemical Performance of Lithium-Sulfur Batteries

Abstract

The growing demand for advanced energy storage systems requires the development of next-generation battery technologies with superior energy density and cycle stability, with lithium-sulfur (Li-S) batteries representing a promising solution. Sulfur-containing polyacrylonitrile cathodes (SPAN) for Li-S batteries are a significant advancement for this next-generation battery chemistry, addressing the major issue of limited cycle life encountered in conventional carbon/sulfur composite cathodes. In the presented study, the influence of available ionic and electronic conduction pathways within the cathode on the electrochemical performance of SPAN-based Li-S batteries is studied in details. To this end, a series of SPAN cathodes with different microstructures is prepared by adapting the compression degree of calendering. Mechanical and morphological characterizations confirm a pronounced springback effect due to a characteristic elastic deformation behavior of SPAN. Electrochemical impedance spectroscopy (EIS) shows increased cathode impedance values with multiple overlapping processes in the high- to mid-frequency region in highly compressed SPAN cathodes. Moreover, while the (first) discharge capacity is unaffected, the subsequent charge capacity decreases substantially for highly compressed cathodes. The electrochemical experiments and electrochemical continuum simulations confirm that this phenomenon is mainly due to the disturbance of the electronic percolation pathways caused by the springback behavior during calendering.

Keywords: Lithium–sulfur battery; calendering; electrochemical characterization; electrochemical impedance spectroscopy; mechanical characterization; numerical simulation; sulfurized polyacrylonitrile.

Figure 1

a) Adhesion strength of SPAN as function of compression grade; b) Multi‐point composite (black) and two‐point (red) electrical resistivity of SPAN electrodes as a function of the compression rate; Fitting equation, parameter and residual values for (a) and (b) can be found in Table S1 (Supporting Information); c) pore size distribution before and after calendering for different compression rates, two measurements per compaction rate have been conducted (dark and light color); d) nanoindentation behavior of SPAN, compared with popular LIB electrodes; SEM/EDS cross‐section images: e) pristine SPAN; f) SPAN with 44% compaction rate (big active material particles are encircled exemplarily).

Figure 2

Nyquist presentation of symmetrical OCV impedance data of a) four different calendering grades after 2 h during OCV; b) SPAN in comparison to an ideal porous carbon/sulfur (C/S) composite electrode; OCV impedance behavior over one day for c) uncalendered SPAN cathode and d) SPAN cathode with 62% compaction rate; equivalent models for fitting of e) porous cathodes (TLM) and f) affected SPAN cathodes (R‐RQ‐RQ‐Q).

Figure 3

Fitted resistance contributions of the uncalendered and calendered cathodes using an TML or R‐RQ‐RQ‐Q equivalent model; a) bulk electrolyte resistance R_bulk, b) inter‐particle resistance in high‐frequency regime P2, c) pore resistance, d) inter‐particle resistance in mid‐frequency regime.

Figure 4

Cycling data of cathodes with various compression rates: a) 1st cycle with 3.5 V upper cut‐off potential; b) comparison of 3.0 and 3.5 V upper cut‐off potential; c) capacity and coulombic efficiency of 3.5 V cells; d) effect of C‐rate toward voltage plateaus and capacity of a medium compressed electrode.

Figure 5

Effect of varying homogeneous electric conductivity on impedance response of symmetric cell at OCV. Solid lines represent simulated spectra using the continuum model. Circles indicate experimental data of a) the uncalendered electrode (grey circles) and b) highly calendered electrode (62% compression rate, yellow circles). Real part corrected by high frequency intersect. a) Conductivity determined by multi‐point measurements in the electrode plane. b) Conductivity range determined by two‐point measurements.

Figure 6

Impact of heterogeneity in conductive networks on discharge performance at C/3. Colors indicate layers with a) low electronic (0.001 mS cm⁻¹) and b) ionic (β  =  10) mobility. The thickness of the layers is varied between 0 and 4 µm. The inset shows the corresponding impedance simulations.