Dissolution of NaTi2(PO4)3 in Alkaline Solutions and Its Implications for Use As an Aqueous Anode

Abstract

There are two primary qualities batteries used for grid-scale storage should have (1) a low levelized cost of energy storage (LCOE) or a low $/kWh delivered over the lifetime of the battery and (2) a high degree of safety which includes low toxicity, low flammability, and low corrosiveness. Aqueous rechargeable alkali-ion batteries offer a promising alternative to Li-ion batteries for grid-scale electrochemical energy storage installations because of their lower cost of and higher safety. However, with the exception of systems that utilize activated carbon based anodes, demonstrated chemistries show significant capacity fade. Although a decent amount of research has been done on failure mechanism of intercalation cathodes in aqueous solution, few studies have been carried out to elucidate the role of the anode material in battery failure. Sodium super ionic conductor (NASICON) type NaTi2(PO4)3 is one of the most promising materials for aqueous anodes. It has good chemical stability, a high theoretical specific capacity of 133 mAh/g, low cost of manufacture, and a favorable redox potential. However, it is often reported to be cycled in neutral aqueous solution, which when given its low redox potential implies there will be some amount of water decomposition occurring. Water decomposition occurs due to both direct electrochemical hydrogen evolution, as well as direct chemical oxidation of the fully sodiated NaTi2(PO4)3. In this work we have set out to quantify the inherent chemical stability of NaTi2(PO4)3 in alkaline solutions using inductively coupled mass spectrometry (ICP-MS), x-ray diffraction (XRD), Fourier transform infrared spectroscopy (FT-IR), scanning electron microscopy (SEM), cyclic voltammeter (CV), and in-situ pH measurements. It was found that at pH > 11 dissolution begins to occur but does not become significant until pH > 12.7. At the highest tested pH an unidentified phase with an elongated morphology becomes thermodynamically favorable and begins to precipitate from solution. This secondary phase acts as a sink for dissolved titanium allowing for the continual dissolution of NaTi2(PO4)3 rather then eventual saturation of solvated titanium species. The in-situ pH measurement indicates that upon cycling in neutral solution the electrode undergoes a pH swing from lower pH to higher pH due to the generation of OH- ions, followed by their diffusion out of the electrode. This increase followed by decrease in pH exacerbates the dissolution of NaTi2(PO4)3 and leads to a continual dissolution precipitation reaction. The combination of FTIR and XRD indicate that these precipitated phases were a mixture of layered titantes of the form NaxH2-xTinO2n+1-yH2O. Although NaTi2(PO4)3 appears to be susceptible to dissolution at higher pH, the measured dissolution does not account for the overall capacity fade seen upon cycling. The authors are currently investigating the effect of gas generation and entrapment on the loss of electrochemical surface area and its effect on the apparent capacity fade.