Rice husk (RH), the outer covering of a rice kernel, is an abundant agricultural byproduct that can be source of anode materials for lithium-ion batteries (LIBs). From RHs both carbon (from the organic constituents), and silicon (RHs contain about 10/15% wt of hydrated silica) can be derived. As anode material for LIBs, Si has a theoretical capacity ten folds that of standard graphite electrodes, but it is subjected to huge volumetric expansion upon lithiation (> 300% for bulk) which leads to high mechanical instability and thus to rapid battery failure. Compositing nanosized Si domains with C is an effective route towards mechanical stability while also increasing the anode conductivity. The carbon obtained by direct carbonization of RH at high temperature in inert atmosphere behaves like an hard carbon, so it is also suitable for sodium intercalation. In this work, different C/SiO2 and C/SiO2/Si composites derived from RH are tested in half-cell configuration vs. Li as well as Na, both with conventional (LP30) and non-conventional glyoxal based electrolyte, namely 1M LiTFSI in TEG/PC (3:7) and 1M NaTFSI in TEG/PC.

In this work,three different C/SiO₂ RH-based composite anodes are investigated. Two of the composites are obteined by carbonization of the RH in a tubular oven under Ar atmosphere either up to 800°C (sample RH800) or up to 1000°C (RH1000). These two samples were then used as active material for preparing electrodes without any further treatment. The third sample, after the same carbonization at 800°C, has undergone a magnesiothermic reduction at 700°C in a tubular oven under Ar. The thus obtained sample (RHMgR) has been washed with 1M HCl and filtered with distilled water to remove by-products (mainly MgO). All the samples have been characterized by X-Ray diffraction (XRD), BET surface analysis, X-ray photon absorption spectroscopy (XPS), scanning-electron microscopy elemental differential analysis (SEM-EDX). Electrodes were prepared by depositing (Doctor Blade coating) on Cu foil a slurry made of active material (90%), carbon black (5%) and CMC (5%) in distilled water.

All the samples were characterized by cyclic voltammetry (CV) and galvanostatically cycled in Li-half cells with standard LP30 electrolyte and the alternative glyoxal-based electrolyte. Directly carbonized samples have also been similarly tested in Na-half cells.

Preliminary results show that the directly carbonized samples behave like hard-carbons: no clear redox peaks during CVs, good capacity retention at 1C cycling (calculated on graphite-rates). For this reason, the first two sample have been investigated also in half-cell configuration. 

The glyoxal-based electrolyte with Li-salt performances are comparable to that of LP30.