Quantifying Non-Uniform Internal SEI Growth under Dynamic Cross-Plane Core Thermal Gradients

Core

Cross-plane thermal gradients in thick-electrode lithium-ion cells drive localized current crowding and uneven solid electrolyte interphase development. Under a continuous 3C load, dynamic discharge in a 21700 or 46800 cylindrical cell can push the temperature delta between the center mandrel and the outer can wall past 12 Kelvin. Large-format prismatic cells between 100 Ah and 280 Ah exhibit cross-layer differences over 15 Kelvin during fast-charge pulses when bottom cold plates provide the only cooling.

This thermal gradient continuously skews local overpotentials across individual jelly-roll wraps and stacked electrode pairs.

Parasitic reduction of ethylene carbonate and linear carbonates at the graphite surface follows Arrhenius kinetics. The hot core layers experience rapid solvent reduction, whereas the colder outer layers develop high charge-transfer resistance. Nominal life models often treat the active electrode stack as an isothermal block, which overlooks how lower electrolyte viscosity and faster kinetics in the inner wraps concentrate local current density away from the chilled periphery.

Under a continuous 3C discharge rate with single-side liquid chilling at 15 degrees Celsius, a 50 Ah prismatic cell produces a 14.2 Kelvin cross-plane core-to-skin temperature spread.

During the opening stages of a high-rate charge, the interior anode layers take on higher instantaneous current. As parasitic film passivates these core graphite particles, cyclable lithium inventory is lost non-uniformly. The outer layers must then take up excess current at lower temperatures, pushing local graphite potential below zero volts versus Li/Li+ and triggering secondary lithium plating.

Ignoring these internal gradients when designing fast-charge profiles leads directly to early pack degradation, sudden capacity rollover, and module imbalances that cannot be rebalanced.

Kinetics

Spatial Reaction Rate Divergence

Degradation rates across the cell reflect coupled electrochemical and thermal dynamics. The modified current density for solvent reduction depends strictly on interfacial temperature and overpotential at any given point. Higher temperatures in the core scale up the standard reaction rate constant, setting the local rate of film growth through the electrode thickness and along the winding radius.

Above 45 degrees Celsius, lithium hexafluorophosphate breaks down much faster. The resulting solvent decomposition consumes cyclable lithium ions and precipitates inorganic compounds like lithium fluoride and lithium carbonate alongside organic alkyl carbonates. Table 1 outlines the kinetic and physical parameters for cross-plane film evolution across a 65 mm radius jelly roll.

Electrode layers age at noticeably different speeds. The innermost wraps build up thick, dense passivating layers dominated by inorganic salts. Outer wraps carry thinner films but develop localized mechanical fractures from uneven lithium-ion flux, causing local state-of-charge swings between the inner and outer wraps to diverge over extended cycling.

Does Dynamic Cycling Compound Spatial Film Inhomogeneity?

Dynamic power profiles cycle between sharp Joule heating and convective cooling, preventing the cell from ever settling into thermal equilibrium. Because the internal gradient keeps shifting, the primary reaction front moves continuously along the radius.

During regenerative braking pulses, the warmer core takes the brunt of the charge current because its localized ohmic and charge-transfer resistances are lower. Once the pulse ends, internal relaxation currents flow from the more heavily charged core to the undercharged outer layers. These internal balancing currents continue driving parasitic reactions during rest periods, completely hidden from external pack-level monitoring.

Higher local temperatures accelerate passivating layer growth while cooler regions accumulate plating risk.

Gradient

Coupled Physics of Cross-Plane Transport

Thermal conduction through stacked cell layers is strongly anisotropic. Across electrode sheets, separators, and current collectors, cross-plane thermal conductivity sits between 0.5 and 1.2 W/m K, compared to in-plane conductivity of 20 to 35 W/m K along the copper and aluminum foils. Heat generated deep inside thick jelly rolls struggles to cross the separator stack toward surface cooling paths.

This cross-plane gradient sets up an uneven ionic conductivity field in the liquid electrolyte. Bulk electrolyte conductivity climbs with temperature while tortuosity stays largely fixed, meaning central anode particles face lower liquid-phase overpotentials and draw ionic flux toward the interior.

Differential material expansion magnifies the problem. Graphite expands about ten percent along the c-axis at stage-one lithiation. Uneven lithiation across the cell radius sets up local mechanical stresses, leaving high-temperature anode zones to deal with both accelerated film growth and mechanical shear from cyclic volume changes.

Quantification Methodologies

Tracking radial film growth requires multi-physics finite element models paired with reduced-order electrochemical estimators that account for layer-resolved heat dissipation and local overpotential shifts.

1. Electrochemical Impedance Spectroscopy separates bulk resistance shifts from interfacial charge-transfer resistance across distinct frequency decades between 10 kHz and 10 mHz.
2. Galvanostatic Intermittent Titration quantifies local solid-state diffusion coefficients and overpotentials at incremental open-circuit voltage plateaus.
3. High-Precision Coulometry isolates total parasitic side-reaction currents by tracking coulombic inefficiency down to parts-per-million resolution.
4. Direct Heat Flux Sensing records transient heat generation split between reversible entropic heat and irreversible polarization losses.

UL 9540A test reports without embedded multi-point thermocouple logs obscure internal core temperature peaks during fast charge qualification.

Simulations show that a 10 Kelvin cross-plane gradient maintained across 1,000 equivalent full cycles causes a 300 percent difference in passivating film thickness between the inner mandrel and outer casing. Whether acoustic emission monitoring can reliably detect inner-wrap breakdown without destructive teardown remains an open question.

Dissection

Post-Mortem Spatial Characterization

Confirming non-uniform film growth directly requires destructive teardown inside an argon glovebox. Cells are brought down to zero volts or a set cut-off before opening, allowing technicians to pull anode samples from specific radial positions: the core near the center pin, the mid-roll region, and the outer wrap against the steel or aluminum can.

Harvested samples are washed with dimethyl carbonate to strip away residual salt deposits before analysis. Combining several analytical tools highlights clear structural and chemical differences across the layers.

• Scanning Electron Microscopy exposes cross-sectional morphology and film topography, resolving thickness variations from 15 nanometers on outer layers to over 80 nanometers in core regions.
• X-Ray Photoelectron Spectroscopy determines the elemental ratio of organic carbon-oxygen compounds to inorganic fluorides across depth profiles sputtered by argon ion beams.
• Inductively Coupled Plasma Mass Spectrometry measures the absolute consumption of cyclable lithium across isolated single-layer electrode harvest disks.
• Focused Ion Beam Milling prepares pristine cross-sections of active graphite particles, exposing internal microcracks induced by asymmetric lithiation stress.

Is Post Mortem Destructive Metrology Transferable to Pack Models?

Physical characterization provides essential calibration points for lifetime models. Post-test inspections show that core graphite particles suffer severe binder breakdown and active surface loss, with dense surface layers clogging pore openings, driving up local tortuosity, and choking liquid diffusion.

Capacity loss is frequently misattributed to uniform degradation across the entire electrode area.

Ledger

Overlooking internal thermal gradients skews total cost of ownership and lifecycle financial models. Storage integrators and automakers frequently base warranty reserves on datasheet cycle life measured under ideal conditions. A cell rated for 3,000 cycles at 25 degrees Celsius often falls below 1,400 cycles under fast-charging demands when pack-level cooling applies heat unevenly across cell boundaries.

Localized surface film growth accelerates internal resistance in large-format cells, creating capacity mismatches that passive balancing shunts cannot correct. As a result, storage operators are forced into unplanned augmentation spending when system capacity drops below performance guarantees years earlier than modeled.

In a 100 MWh utility installation, sizing calculations based on linear isothermal aging underestimate degradation by 22 percent under daily two-hour discharge cycles with bottom-plate chilling. The resulting early replacement cycles introduce substantial unbudgeted cell replacement and disposal freight costs over a ten-year operating window.

Procurement teams counter these financial risks by building gradient-specific validation criteria into supply agreements. Robust specifications require cell suppliers to provide spatial degradation mapping and cycle data collected under thermal gradients that match the actual pack cooling geometry.

Structuring warranties around IEC 62620 Annex H shifts financial exposure by pegging capacity retention targets to multi-axis thermal gradient tests rather than single-cell oven baselines.