Synthesis and Comparative Electrochemical Studies of Bottlebrush Solid Polymer Electrolytes (SPEs) for Li-Ion Transport

Abstract

Solid polymer electrolytes in solvent-free lithium batteries may overcome some of the disadvantages of liquid electrolytes such as flammability and instability.1 Thereby, the polymer electrolyte acts as both ion transport medium and electrical separator between the electrodes. Compared to rigid separators (e.g. fiber glass) a higher shape flexibility is a further advantage, as these SPEs can compensate volume changes of the electrodes by elastic and plastic deformation.2 Poly(ethylene glycol) (PEG) possesses one of the highest ionic conductivities among solvent-free SPEs but suffers from a conductivity drop below its melting temperature about 50-60°C due to high crystallinity depending on the molecular weight.1 In addition to linear PEG polymers, ion-conducting bottlebrush graft copolymers can be obtained by attaching PEG side chains to a polymer backbone in order to reduce the crystallinity maintaining very high molecular weight.In this work we synthesized five new bottlebrush polymers using free radical polymerization as well as ring-opening metathesis polymerization (ROMP). These brush polymers contain different lengths of PEG side chains (1 kg mol-1 and 2 kg mol-1) and two different backbones, poly(methacrylate) and poly(norbornene). We present the influence of the polymer architecture on mechanical stability, ionic conductivity, Li-ion transport number and electrochemical stability of a series of SPEs obtained thereof by mixing with different amounts of lithium bis(trifluoromethanesulfonyl)imide (LiTFSI). We also compare the results to the respective linear PEG counterparts. The differences in ionic conductivity were analyzed and correlated with respect to thermal properties such as T g and ΔH melt in order to understand the fundamental factors which influence the properties of solvent-free PEG-containing bottlebrush SPEs.We also examined the implications of changing a linear polymer system to a brush architecture for potential applications in batteries and correlate the occurring processes in the cell with the distribution of relaxation times (DRT). Detailed comparative measurements under similar cell configurations for diverse O/Li ratios in a temperature range of 25 to 80 °C for different SPEs were carried out to elucidate structure-property relationships. The interesting findings are that by applying a brush architecture, we suppress the crystallinity of PEG and improve the mechanical strength without losing ionic conductivity. We obtained conductivities in the range of 10-3 to 10-4 S cm-1 for solvent-free SPEs. Furthermore, the best ionic conductivities for any system correlate strongly with their respective T g.3 Nevertheless, there are still several unresolved questions regarding these bottlebrushes compared to linear PEG in terms of interfacial as well as bulk processes in both blocking steel (ionic conductivity) and lithium (lithium plating/stripping, interfacial resistance) electrode setups. In addition to the polymer component, the lithium salt has a major effect on the properties of the electrolyte. LiTFSI is probably the most common Li-ion source in SPEs.4 Besides that, lithium borate salts have gained high interest due to their high thermal stability, cost-effectiveness, favorable solid electrolyte interface (SEI) formation and ionic conductivities in the same range as LiTFSI.5 For example, Lithium bis(oxalate)borate (LiBOB) and its asymmetric counterpart Lithium difluoro(oxalate) borate (LiDFOB) are stable in organic solvents and the electrochemical stability is higher than 4.5 V vs. Li/Li+.6 Different salts in an electrolyte can influence the Li-ion transport as well as the processes at the interfaces or the formation of the SEI. This again requires a detailed comparative analysis.For this, we prepared promising bottlebrush polymer electrolytes (1 kg mol-1 PEG sidechain) containing LiBOB and LiDFOB and subsequently analyzed them electrochemically by impedance spectroscopy. The measurement data was finally interpreted by the extended Distribution of Relaxation Times (eDRT). Both SPEs showed similarities in the lithium-ion conducting process in possessing one major, resistive-capacitive bulk conductivity mechanism. The resulting, temperature-dependent conductivities were evaluated and are in the range of 10-4 to 10-5 S cm-1.During cycling, the SPEs showed increased interface resistances over time, which are higher than respective bulk resistances. By applying eDRT, the time-dependent formation of an interphase layer in the SPEs is identified, separated from the slower charge transfer process and quantified. Thus, the electrolytes cannot be considered electrochemically stable against metallic Li, which is similar for liquid electrolytes.7 References (1) Scrosati, B.; Energy Environ. Sci. 2011, 4, 3287.(2) Janek, J.; Nat. Energy 2016, 1, 16141.(3) Rosenbach, D.; ACS Appl. Energy Mater. 2019, 2, 3373–3388.(4) Etacheri, V.; Energy and Environmental Science 2011, pp 3243–3262.(5) Xu, K. Chem. Rev. 2014, 114, 11503–11618.(6) Liu, Z.; Coord. Chem. Rev. 2015, 292, 56–73.(7) Hahn, M.; Electrochim. Acta 2020, 344, 136060. Figure 1