A NaS cell holds molten sodium (m.p. 98 °C) and molten sulfur/sodium‑polysulfide (m.p. 115 °C) apart with a beta‑alumina solid electrolyte (BASE) wall that conducts only Na⁺ ions — everything runs above ~300 °C so both electrodes stay liquid. Discharge reaction:
2 Na + x S → Na₂Sx (x decreasing from ≈5 toward ≈2 as discharge proceeds)
The open-circuit voltage is set by which two-phase (or single-phase) sulfur/polysulfide region the cathode composition sits in, giving the cell's signature staged, multi-plateau discharge curve:
V_ocv ≈ 2.076 V liquid S₈ + Na₂S₅ two-phase coexistence (SOC ≳ 60%)
V_ocv 2.076→1.78 V single-phase liquid Na₂Sx, x sliding (60% ≳ SOC ≳ 40%)
V_ocv ≈ 1.78 V solid Na₂S₃ / Na₂S₂ two-phase coexistence (SOC ≲ 40%)
Terminal voltage subtracts an internal-resistance drop, V = V_ocv − I·R_int. R_int comes from the Na⁺ conductivity of the ceramic, which follows an Arrhenius law — hotter cells conduct better and lose less voltage under load:
σ(T) = σ₀ · exp(−E_a / k_B T), R_int ∝ 1/σ(T)
- Discharge / Charge / Hold — sets the direction ions cross the electrolyte; Na⁺ leaves the anode pool (its level visibly drops) and migrates into the sulfur cathode on discharge, and reverses on charge.
- Current rate — the C-rate driving Na⁺ flux; higher current drains state-of-charge faster and produces a larger I·R_int voltage sag.
- Cell temperature — raises ionic conductivity of the beta-alumina, shrinking R_int and the voltage sag for the same current.
The lower chart plots terminal voltage against integrated capacity as the cell runs, tracing out the real staged discharge/charge curve live — watch the flat plateau give way to the sloping single-phase region and then settle onto the lower plateau. Real cells (e.g. NGK's NAS units) run this exact chemistry at grid scale for multi-hour storage, prized for high round-trip efficiency and no self-discharge — the tradeoff is the standing 300+ °C operating temperature simulated here.