Solid-State Fast Charging: What the Test Data Actually Shows

Every solid-state battery announcement over the past year or so has come with a headline charge time: 15 minutes, 12 minutes, sometimes under 10.

Feels like a limbo dance at a bar, getting lower and lower as we all get drunk on the lemonade.

Lined up next to each other the claims look almost interchangeable, but the amount of evidence sitting behind each number is wildly different: a multi-year, methodically disclosed test programme with real cycling data in one case, and a forward-looking company target with no public test data at all in another. Sorting out which is which matters more than the round number itself if the plan is to design around this technology rather than just read about it.

The Best-Documented Case: QuantumScape

QuantumScape has published more methodology behind its fast-charge numbers than most of the field, but the numbers come from at least three separate tests, run years apart on different cell builds, and they get flattened into a single “fast charge” headline more often than the underlying data supports.

The earliest disclosure, from a 2021 battery showcase, ran single-layer prototype cells at a 1C rate: a one-hour charge, not a fast charge and reported over 1,000 cycles to a median of better than 90% energy retention, close to room temperature, commercially-relevant current density and full depth of discharge [1]. In the same showcase, a separate and much harsher test was built explicitly around fast charging: a simulated race-track drive cycle combining aggressive acceleration and braking with repeated 15-minute charges. That test delivered on the order of 100 full cycles, retaining 90% of energy at 100% depth of discharge [1]. The 1,000-cycle figure and the fast-charge figure are not the same test, and quoting one in support of the other overstates how much fast-charge cycling had actually been demonstrated at that point.

A year later, in January 2022, QuantumScape published a test aimed squarely at fast charging: 400 consecutive 4C (15-minute) charges from 10% to 80% state of charge, run at both 25°C and 45°C, 3.4 atmospheres of stack pressure and full depth of discharge, retaining more than 80% of initial energy after those 400 cycles [2]. For a 400-mile-range pack, 400 cycles works out to roughly 160,000 driving miles [2]. QuantumScape ran a third-party commercial EV cell through the identical protocol for comparison and reported that it degraded rapidly within a few dozen cycles. A useful baseline, though the comparison cell’s specific chemistry and vendor were not disclosed [2].

The QSE-5 cell now shipping to automakers as B-samples carries a newer, rounder figure: under 15 minutes for the same 10-80% window, quoted for a roughly 5 Ah, multi-layer, automotive-format cell rather than the single-layer research cells above [3]. What hasn’t been re-published alongside that figure is a matching cycle-by-cycle retention curve at the 15-minute rate specifically. A separately-reported cycle-life figure: better than 95% energy retention after 1,000 cycles, attributed to Volkswagen subsidiary PowerCo testing an earlier prototype. However, this doesn’t state what charge rate those 1,000 cycles were actually run at [4], so it would be a mistake to assume it is 1,000 cycles at the 15-minute fast-charge rate rather than a slower one; treat it as a separate, less-documented claim rather than proof the fast-charge number holds for 1,000 cycles.

Another Disclosed Dataset: Nuvvon’s Silicon-Anode Cells

A smaller solid-state developer, Nuvvon, has posted a more granular cycle-life dataset on its own 5Ah silicon-anode NMC cells directly on this site, Dr. Simon Madgwick [5]. The methodology disclosed is more specific than most headline announcements: a 5Ah SiC-NMC “Power” cell cycled continuously at 3C charge / 5C discharge, full depth of discharge, in a 25°C chamber under 10psi of fixture pressure, reaching 1,014 cycles to 91% state of health at the time of posting [5]. A 3C charge is a 20-minute charge over the full state of charge window, simple maths gives 14 minutes for the 10 to 80% SoC window.

*No pressure! Nuvvon 5Ah SiC NMC solid state, Dr. Simon Madgwick / Nuvvon Inc.*

The detail worth flagging for the stack-pressure question raised below is a follow-up post comparing the same cell cycled with no external fixture pressure at all: after 450 cycles, the zero-pressure cell was tracking the same degradation curve as the 10psi cell that had already passed 1,000 cycles [6]. If that holds over the full cycle count, it would suggest Nuvvon’s solid polymer electrolyte tolerates the volume change of a silicon anode without needing the applied pressure that garnet and sulfide solid electrolytes typically rely on.

Targets, Not Yet Test Data: Toyota and Samsung SDI

Toyota and Samsung SDI’s numbers sit in a different category again: both are stated targets for future commercial products rather than published test results. Toyota has said its sulfide-electrolyte solid-state cell is intended to support a 10% to 80% charge in around 10 minutes, alongside a 600-plus mile range target, with commercial production planned for 2027-2028 [7]. Samsung SDI’s own published roadmap points to a 9-minute, 8%-to-80% charge target on its all-solid-state cell by around 2027, alongside a stated 20-year design-life goal [8]. Neither company has published the kind of cycle-by-cycle retention curve, test temperature or C-rate methodology that QuantumScape disclosed for its own 2021 and 2022 results above; both are, at this stage, roadmap commitments rather than measured data anyone outside the company can independently check.

Why Fast Charge and Cycle Life Are the Same Problem

The reason none of this is simple is physical, not just a matter of companies being cagey with data. Fast charging pushes lithium ions into the anode at a high plating current density, and above a chemistry’s so-called critical current density, a solid electrolyte tends to grow dendrites (lithium filaments that penetrate the separator and short the cell) rather than plating smoothly. That ceiling is a genuine material limitation rather than a design choice: Brown University engineers testing LLZTO, a garnet-type solid electrolyte prized for its high ionic conductivity, found it dendrites readily at high charging rates in its normal state [9].

What the same study also found is that the ceiling isn’t fixed. Heating one face of the electrolyte while cooling the other put the material under compressive stress and increased its critical current density three-fold, using just a 20°C gradient across the cell [9]. The researchers believe this could plug into a pack’s existing thermal management hardware, since cycling already generates the heat needed to create the gradient, rather than requiring new components [9]. That is a genuinely promising result, and it illustrates the point of this whole page well: it is a single research-scale demonstration on one electrolyte chemistry, published weeks before this was written, not yet a large-sample, long-cycle, production-format result. It will likely show up as a talking point on the next battery-day slide long before anyone can independently verify it at the pack level.

This post has been built based on the support of and sponsorship from: illumion, Custom Power, Quarto Technical Services, Eatron Technologies and About:Energy

What This Actually Means for a Pack Designer

Read across everything above, the useful takeaway isn’t “solid-state fast charging works” or “it doesn’t”, it is a short checklist for reading the next headline number. Ask what rate the cycle-life figure was actually measured at, and whether it is the same test as the fast-charge claim or a different, slower one quietly reported alongside it. Ask whether the cell tested is the same format and layer count as the product actually being sold, since single-layer research cells and multi-layer automotive-format cells do not necessarily behave the same way under fast charge. Ask whether an independent lab ran the test or whether the number is entirely self-reported. And treat any charge-time figure that omits temperature, applied pressure and depth-of-discharge as incomplete, because all three change the result significantly.

In summary it feels like the smaller solid-state developer, Nuvvon, is on the more solid path based on the quality and amount of data they are sharing.

References

A Discussion of QuantumScape’s Battery Technology Performance Results, QuantumScape, 2021
QuantumScape Data Shows Industry-First 15-minute Fast Charging for Hundreds of Consecutive Cycles, QuantumScape, 2022
A First Look at the QSE-5 B Sample, QuantumScape, 2024
QuantumScape Starts Shipping QSE-5 Sample Cells, Battery-News.de
Nuvvon Power 5Ah Cell, Dr. Simon Madgwick / Nuvvon Inc., via BatteryDesign.net
No pressure! Nuvvon 5Ah SiC NMC solid state, Dr. Simon Madgwick / Nuvvon Inc., via BatteryDesign.net
Toyota touts 10-minute EV charging, solid-state battery due in 2027, Green Car Reports
SAMSUNG SDI to Present Essence of Super-Gap Battery Technology at InterBattery 2024, Samsung SDI
New strategy addresses persistent problem in next-generation solid-state batteries, Brown University