Heat at the source: how peak cell current sets the thermal budget of a lithium-ion pack

Thermal management of a battery pack is usually discussed as a cooling problem: how much air, how much coolant, how big a plate. That framing quietly assumes the heat is a given and the only question is how fast it can be carried away.

It is worth turning the question around. The heat is not a given. It is generated inside the cell, by a mechanism with a known governing equation, and the dominant term in that equation is set by the current waveform the pack is asked to accept. Change the waveform and you change the heat at the point where it is made, rather than chasing it once it exists.

This note sets out that argument in full: where the heat comes from, why the peak matters more than the average, what the elevated-temperature literature says follows from it, what a capacitive buffer on the DC link changes about the picture, and — stated plainly, because it is the part most often skipped — what would have to be measured before anyone claimed a thermal benefit for a specific pack.

1. The heat is made inside the cell

Total heat generation in an operating lithium-ion cell has two parts. The reversible entropic term is associated with the electrochemistry itself and changes sign between charge and discharge. The irreversible term is dominated, in most working conditions, by ohmic dissipation:

P = I²R

where I is the cell current and R its internal resistance — the sum of ionic resistance in the electrolyte and separator, electronic resistance in the current collectors and tabs, and charge-transfer resistance at both interfaces.

The square is the whole argument. A cell asked for 100 A dissipates four times the heat of the same cell at 50 A, not twice. A cell at 200 A dissipates sixteen times. There is no engineering cleverness in that relationship; it is Joule's law, and it applies to every cell ever built regardless of chemistry.

Put a duty cycle against it. A cell held to 0.3C by something upstream, against the same cell cycled at a continuous 1.0C:

P₀.₃ / P₁.₀ = (0.3 / 1.0)² = 0.09

A 91% reduction in the resistive heating term. I want to be exact about the status of that figure, because it gets quoted as though it were a measurement. It is arithmetic. It follows necessarily from Joule's law given the assumption that current is genuinely held to 0.3C, and it is worth precisely as much as that assumption is. It is not a claim about pack life, cooling load or range.

2. Why the peak matters more than the average

Two duty cycles can carry identical energy and identical average current, and put the cell in quite different thermal states.

Heat generation follows the square of the instantaneous current, so it is dominated by the peaks. Heat removal, by contrast, is roughly linear in the temperature difference between the cell and its surroundings, and it is slow: a cell has a thermal mass, and the path from its interior to the coolant has a thermal resistance. The product of the two is a time constant, typically in the range of minutes for a cell in a pack — far longer than the seconds-long transients that produce the peaks.

The consequence follows directly. A short, high-current transient deposits its heat faster than the pack can remove it. The energy goes into raising the temperature of the cell itself, and the cooling system only catches up afterwards. Averaging the current over the cycle and computing I²R from the average understates the heat generated, sometimes by a large factor, because the mean of the square is not the square of the mean.

There is a second asymmetry, which is spatial. Resistance is not uniform inside a cell — current density concentrates near the tabs — so the hottest point is hotter than the average cell temperature, and the core is hotter than the surface a thermocouple can reach. Both gaps widen as current rises.

3. What temperature does to the rest of the cell's life

Heat matters because degradation is temperature-activated. The dominant ageing mechanisms in a lithium-ion cell — growth of the solid electrolyte interphase, electrolyte decomposition, loss of cyclable lithium — follow Arrhenius kinetics, so their rates rise exponentially with temperature rather than proportionally. The working rule of thumb from the ageing literature is a roughly doubled reaction rate per 10 °C rise (Vetter et al., 2005).

That is why the I²R term is worth attention out of proportion to its size in watts. It is not merely a load on the cooling system. It is the input to an exponential, and it is generated in the place least able to shed it.

4. The chain the review literature describes

Thermal runaway is the well-characterised end of that same temperature axis, and the mechanism is documented in the review literature rather than being anybody's proprietary insight. The sequence, as set out by Feng et al. (2018) and in the earlier thermal review by Bandhauer, Garimella and Fuller (2011), is a chain of exothermic reactions with staggered onset temperatures:

SEI decomposition begins at the low end of the range, around 80–120 °C depending on cell chemistry and history. It is exothermic, so it contributes to further heating.
Separator softening and melt follows, typically in the 130–160 °C range for polyolefin separators, which can bring the electrodes into direct contact.
Cathode decomposition begins higher still, above roughly 200 °C for many chemistries, releasing oxygen that participates in subsequent reactions.

What makes the chain self-sustaining is that each step generates heat faster than the previous one, so once the rate of heat generation exceeds the rate of heat removal, the system no longer needs an external source. The temperature at which that crossover happens is a property of the cell and its thermal environment, not of the cell alone.

The relevant observation for a system designer is the one at the start of the chain, not the end. Every step is entered from below. Whatever keeps the cell's peak internal temperature further from the first onset gives the thermal management system more margin to work with, and the first onset is approached through ordinary self-heating long before any abuse condition is involved.

Two things are worth saying clearly here. Onset temperatures vary substantially between chemistries — LiFePO₄ is well known in the literature for higher onset temperatures and a less energetic cathode decomposition than nickel-rich oxides — so the numbers above are ranges, not thresholds. And self-heating is only one route to the chain: mechanical damage, an internal short from a manufacturing defect, overcharge and external heating all enter it by paths that have nothing to do with the duty cycle.

5. What a capacitive buffer changes about the picture

An electric double-layer capacitor stores charge electrostatically at the electrode–electrolyte interface. There is no phase change, no intercalation and no diffusion limit of the kind a battery has, which is why it will accept and release a large current for a short time, hundreds of thousands of times, without meaningful degradation.

That makes it a poor way to store energy and an excellent way to store transients. A duty cycle is not one thing: it is a low-frequency energy demand with a high-frequency power demand sitting on top of it, and nothing requires both to be served by the same device. The architecture in the Hybrid Power Pack patents partitions them — chemistry handles the steady-state energy case, architecture handles the transient-stress case — with a capacitor bank on the DC link intercepting the peaks.

Expressed in the terms of this note, the buffer acts on the generation side of the thermal problem rather than the removal side. It attenuates the peaks that dominate I²R, which is the term that is both largest and worst-timed. Nothing about it changes R, the cell chemistry, the onset temperatures, or the cell's behaviour under abuse.

6. What would actually demonstrate a thermal benefit

Here is the part that decides whether any of the above is engineering or marketing.

Two independent test reports exist on the Hybrid Power Pack hardware. The Central Institute of Road Transport, Pune — one of the vehicle testing agencies designated under Rule 126 of India's Central Motor Vehicles Rules — tested Model HPP35A and recorded 10,000 cycles at 60 °C, report No. B34000, 20 March 2020. DRDO's R&DE(E) establishment, Pune, issued an environmental test report, No. 020623ESG/ETF, 21 June 2023, covering discharge at +70 °C under sustained 450 A.

Both are endurance and environmental evaluations of a single configuration. Neither was run against a battery-only control, and neither measured thermal-runaway behaviour. They establish that the hardware survives a demanding thermal envelope; they do not, and were not designed to, establish a comparison.

The test that would settle it is not exotic, and it is worth specifying so that anyone can hold the result to it:

Two packs, one variable. Identical cells, identical count, identical thermal path and enclosure. One with the capacitive buffer, one without. Anything else that differs makes the comparison unreadable.
A defined duty cycle with real transient content, published in full, and run identically on both.
Instrumented cell temperature — surface at several points including near the tabs, and core where a reference cell allows it, since the surface reading understates the interior.
Comparative type-testing to IEC 62619 and IEC 62133-2 against that duty cycle, which is the recognised framework for this class of claim.
For any statement about runaway specifically, the recognised abuse protocols — nail penetration, overcharge, external short, thermal propagation — run on both configurations, by a third party, with the full report published rather than summarised.

Until that comparison exists, the defensible statement is the narrow one: the physics of I²R and of Arrhenius kinetics is not in dispute, the architecture acts on the term those equations identify, and the size of the effect on a specific pack is a measurement that has not yet been made.

The claim worth making

The industry's default answer to cell heating is a better cell or a bigger cooler, and enormous value has come from both. But a meaningful share of the heat a cell produces is attributable to the current waveform it is handed, and that waveform is an architectural choice — available now, on cells that already exist, without waiting for the next chemistry.

That is a narrower claim than "we prevent thermal runaway." It is also one I can defend in a room full of people whose job is to find the hole in it, which is the only kind of claim worth putting a patent number next to.

References
Vetter, J. et al. (2005). Ageing mechanisms in lithium-ion batteries. Journal of Power Sources, 147(1–2), 269–281.
Bandhauer, T. M., Garimella, S. & Fuller, T. F. (2011). A critical review of thermal issues in lithium-ion batteries. Journal of The Electrochemical Society, 158(3), R1–R25.
Feng, X. et al. (2018). Thermal runaway mechanism of lithium ion battery for electric vehicles: A review. Energy Storage Materials, 10, 246–267.

Hemant K. Rohera is an inventor and engineer in Pune, India, sole named inventor on three granted patent families — bioelectronic medical devices, hybrid energy storage and vehicle power electronics. The Hybrid Power Pack family is granted and in force as IN 301517 · US 10,523,019 B2 · EP 3 320 595 B1 (validated in Germany, the United Kingdom and Switzerland) · JP 6644883 · CA 2,991,527 C · MX 377263 B · VN 40455 · ZA 2018/00747 · EA 035682 (in force in Russia), priority 10 July 2015. Available for licensing, assignment or joint development. ORCID: 0009-0005-3275-1743

The complete patent register, with a grant number and grant date for every entry, and the research record, with abstracts and DOIs for the deposited notes, are published at hemant-rohera.vercel.app/patent-register.html and hemant-rohera.vercel.app/research.html.

Search This Blog

Hemant Rohera — Bioelectronics, Energy Storage and Power Electronics