Researchers at Columbia University and the Massachusetts Institute of Technology have demonstrated a technique for guiding lithium-ion deposition inside batteries using externally applied magnetic fields, a development that addresses one of the most persistent safety and performance barriers in next-generation battery design. Published in Science in late 2025, the study showed that a modest magnetic field of 0.5 Tesla — comparable to what a small permanent magnet produces — could redirect lithium plating away from dangerous needle-like dendrite formations and toward uniform, compact layers on the anode surface.
The Dendrite Problem
Lithium-metal anodes have long been considered the ultimate goal for battery engineers. Replacing the graphite anode in a conventional lithium-ion cell with pure lithium metal could increase energy density by 40%–70%, potentially enabling EV batteries that store more than 500 Wh/kg. The obstacle has always been dendrites — microscopic lithium spikes that grow during charging, pierce the separator membrane, and cause short circuits that can lead to thermal runaway and fire.
Decades of research have attacked the dendrite problem through chemical additives, solid electrolytes, and mechanical pressure. The Columbia–MIT team took a fundamentally different approach: exploiting the fact that lithium ions carry an electrical charge and therefore respond to magnetic Lorentz forces when they move through an electromagnetic field.
In controlled experiments, cells equipped with a compact magnetic array around the anode exhibited zero detectable dendrite formation across 500 charge–discharge cycles at current densities of 4 mA/cm² — roughly four times the rate at which dendrites typically begin to appear in unprotected lithium-metal cells.
How It Works
The mechanism relies on magnetohydrodynamic (MHD) convection. When lithium ions travel through the electrolyte under the influence of a perpendicular magnetic field, they experience a force that creates a gentle circular flow pattern in the electrolyte near the anode surface. This convection disperses concentration gradients — the uneven buildup of ions that seeds dendrite nucleation — and promotes even lithium deposition.
- Field strength required: 0.3–0.5 Tesla, achievable with neodymium permanent magnets
- Weight penalty: Estimated at 3%–5% of pack mass for the magnetic array
- Cycle life tested: 500 cycles at 4 mA/cm² with no dendrite detection via post-mortem SEM imaging
- Coulombic efficiency: 99.7%, approaching the 99.9% threshold considered necessary for commercial viability
“We are not changing the chemistry of the battery. We are changing the physics of how lithium moves. That distinction matters because it means this approach can, in principle, be layered on top of any lithium-metal cell design.” — Prof. Yuan Yang, Department of Applied Physics and Applied Mathematics, Columbia University
Path to Practical Application
The research team acknowledged several hurdles before the technique could reach production vehicles. The permanent magnets add weight and cost. The MHD effect diminishes at low current densities — meaning it is most beneficial during fast charging, where dendrite risk is highest, but less impactful during slow overnight charging. And the interaction between magnetic fields and battery management electronics would require careful shielding and calibration.
Still, the approach has attracted early-stage interest. QuantumScape, the Silicon Valley solid-state battery developer backed by Volkswagen, confirmed it is evaluating magnetic-field-assisted deposition in its R&D program. Samsung SDI filed a related patent in South Korea in November 2025 describing a “magnetically enhanced anode structure” for lithium-metal cells.
If the weight and integration challenges can be resolved, magnetic control of lithium deposition unlocks the long-sought combination of lithium-metal energy density and lithium-ion-level safety — a pairing that would meaningfully accelerate the timeline for EVs with 600+ km range and 15-minute fast charging. For now, the Columbia–MIT results represent a proof of concept that has moved the idea from theoretical physics into the realm of engineering optimization.
