Mechanism of reaction in NaAlCl sub 4 molten salt batteries with nickel felt cathodes and aluminum anodes. II. Experimental results and comparison with model calculations

• Published in: Journal of the Electrochemical Society Vol. 140; no. 12; pp. 3380 - 3390
• Main Authors: Knutz, B C, Berg, R W, Hjuler, H A, Bjerrum, N J
• Format: Journal Article
• Language: English
• Published: 01.12.1993
• ISSN: 0013-4651

The battery systems Al/NaCl-AlCl sub 3 -Al sub 2 X sub 2 /Ni-felt (X=S, Se, Te) and the corresponding system without chalcogen (X) were studied experimentally at 175 deg C. Charge-discharge experiments, performed on cells with NaCl saturated melts, showed that systems containing dissolved chalcogen are advantageous over the chalcogen-free systems in rate capability and cyclability. Results of gravimetric analysis and Raman spectroscopy of the electrolytes confirmed the exchange of chalcogen between cathode and electrolyte during cycling. Cathode reactions were studied by coulometric titrations (performed on cells with slightly acidic NaCl-AlCl sub 3 melts and small amounts of chalcogen) and compared with model calculations. Cells containing chalcogen revealed at least three voltage plateaus during cycling. The lowest plateau is associated with formation/decomposition of essentially Ni sub y S sub z and Ni sub y Se sub z in the sulfide and selenide system, respectively. Cells containing selenide revealed extra capacity below the Ni sub y Se sub z -plateau, most probably associated with an Al sub v Ni sub y Se sub z compound. On the second plateau of sulfide systems NiCl sub 2 or a Ni sub y S sub z Cl sub 2y-2z compound with y > (4.4plus/minus0.2). Formation of z occurs during charging. Reduction of the formed compound to Ni takes place via consumption of sodium chloride. For acidic melts, sulfide at the cathode was found to be present as a large anion with two aluminum atoms held by a sulfur and a chlorine atoms, and each aluminum atom bearing two chlorine atoms.