Li-ion battery thermal management
Boiling immersion and internal channels that keep fast-charging cells near-isothermal.
The window
Lithium-ion cells want to live between 15 and 40 °C with less than 5 °C of gradient across the cell. A Samsung 18650 discharged at 8C in still air reaches nearly 90 °C. Every conventional cooling scheme, air, cold plates, phase-change material, fights contact resistance and the can’s own conductivity from the outside.
Two-phase immersion
Submerge the cell in a dielectric fluid whose boiling point sits inside the battery’s window and the surface is pinned near saturation by boiling itself, with no thermal interface in the way. I built the rig: an acrylic vacuum chamber, an eight-channel Arbin cycler, a chilled condenser, Keysight and NI data acquisition on surface thermistors and chamber pressure, and a planned active pressure control that sets the saturation temperature on demand. The working fluid is SF33, boiling at 33 °C, benchmarked against Novec 7000 and water.
In full immersion an 18650 under constant-current discharge stays within a few degrees of the fluid’s saturation temperature from 4C all the way to 10C, with about one degree of axial gradient, where the same cell in air runs 40–60 °C hotter. Half immersion shows the other side of the coin: the un-immersed end drives large axial gradients, which is exactly the effect pack designers need to avoid. Two-by-two packs in printed holders and long-cycle aging under immersion are in progress.
Internal cooling of prismatic cells
Large prismatic EV cells have an air-filled void between the jelly roll and the can. Thin aluminium coolant channels in that void take heat out from inside the can rather than through a can, an interface and an external plate. The modeling progressed from a one-dimensional boundary-value model in Python, to a COMSOL electro-thermal model with a two-RC equivalent circuit fitted from published pulse data for a 58 Ah NMC cell, to an Ansys Fluent battery model on geometry validated by X-ray CT and checked against published 2C discharge data. A modeling paper is in draft. The published precursor is the Applied Thermal Engineering 2025 paper on internal flow cooling of cylindrical cells, which I co-authored.
Status
Ongoing within a multi-thrust battery program. Next: a hollow-core cell under immersion, 100-cycle aging with impedance spectroscopy before and after, and cell-spacing studies for pack-level boiling.