The increased demand for lithium-ion batteries (LIBs), and the stain this will place on supply, has sparked significant interest in sodium-ion batteries (NIBs). Sodium is cheaper and more abundant than Li, making NIBs a more sustainable option for certain applications. While numerous NIB electrode materials which seek to maximise electrochemical properties have been developed, it is also important to consider other aspects of sustainability when developing NIBs, such as synthesis methods and the components used to cell fabrication. The aims of this thesis is to use a low energy intensity synthesis method for Na-ion cathode materials and investigate the use of environmentally friendly binders in cathode preparation using polymers derived from nature.
Na2/3Ni1/3Mn2/3O2 (NNM) is a strong candidate for use in Na-ion battery cathodes, due to its relatively high theoretical capacity of 173 mAh g-1 and working voltage around 3.7 V Na+/Na, giving it one of the highest theoretical energy densities compared to other Na-ion transition metal layered oxides. NNM can form two polymorphs depending on final calcination temperature: a high temperature phase which occurs above 800 °C, and a low temperature phase which should have lower energy requirements during synthesis. This low temperature phase is often overlooked, due to the challenges associated with forming phase pure materials at low temperatures using solid state synthesis, since high temperatures are required to fully react the NiO precursor. In previous work, however, this low temperature NNM phase was successfully synthesised at 550 °C using an emergent method known as biotemplating. The work presented herein takes the next steps in the understanding and application of this synthesis method.
Biotemplating is a synthesis method that utilises water-soluble naturally-occurring polymers to spatially distribute a homogenous mixture of different metal cation precursors. During initial stages of calcination, the polymer prevents recrystallisation of the reagent phases, preserving the atomic level mixing found in solution. As the template combusts during calcination, numerous nucleation sites form, allowing for reactions to occur quickly and at lower temperatures vs solid state. Thus, less energy-intensive reaction conditions are required for synthesis of product phases. Biotemplating is therefore a promising synthesis method for complex oxides, because some polymers are able to interact with a number of different cations in solution due the deprotonated functional groups.
To gain a better understanding of the factors which affect biotemplating, the crystallisation process with different dextran concentrations was investigated and was compared to solid state synthesis. It was found that with sufficient dextran, the low temperature NNM phase began to crystallise at a much lower temperature of 300 °C, whereas in solid state synthesis it appears at 500 °C. In addition, the final phase fractions after calcination were significantly impacted depending on the concentration of dextran; the higher the concentration, the lower the amount of secondary phases, such as NiO.
The low temperature phase synthesised with a low concentration of dextran had a large phase fraction of NiO, and was likely to be Ni-deficient in the transition metal layers. It exhibited a lower initial discharge capacity compared to stoichiometric NNM phase, because the main capacity contribution is from the Ni2+/3+/4+ redox, although the cycling stability was found to be better. This improvement was attributed to vacancies in the transition metal layers, which are able to better accommodate structural distortions during electrochemical cycling, as found in the literature. This brings opportunity to investigate alternative methods for improving cycling stability of these types of materials.
Regardless of synthesis conditions, the low temperature phase NNM exhibited poor cycling stability when charged to 4.5 V. This was found to be due to a phase transition which occurs during cycling, and is associated with volume change and structural instability. To further stabilise the structure to higher voltages, a substitutional doping scheme was investigated, introducing some Cu in place of Ni inNa2/3Ni1/3-xCuxMn2/3O2 (x = 1/12, 1/6 and 1/3). Again dextran biotemplating synthesis was used, to preserve the lower reaction temperatures and atomic level mixing. Specimens with Cu content of x= 1/12 and 1/6 formed a minimal amount of secondary phases (<2 wt.%), further demonstrating the ability to synthesise complex oxides using biotemplating. Specimen x = 1/6 exhibited improved cycling stability at extended voltage range of 2 – 4.5 V, with a capacity retention of 77.6 % after 50 cycles, whereas NNM (x = 0) was 53 %.
Finally, in order to consider the materials used in the wider battery architecture, water-based binders were investigated. Binders play a key role in maintaining the structural integrity of the electrode but often use toxic and expensive solvents, and polymers with low flexibility. Here I investigate the water stability of NNM cathodes, and thereafter the use of xanthan gum as a water-based binder. The electrochemical performance was comparable to electrodes prepared with more traditional materials. This shows that biopolymers are a viable option for binders in such electrode materials.
It has been demonstrated that dextran biotemplating is a facile synthesis method for complex Na-ion layered oxide cathodes. It enables early onset crystallisation at low temperatures of the desired phase, thus reducing the overall required energy intensity for synthesis. The investigated materials have shown to be stable against water intercalation, thus allowing the use of sustainable water-soluble binders in cathodes. Here, biopolymers have shown great versatility in the cathode preparation process, improving the sustainability across different aspects of Na-ion batteries.
