The Interesting Engineering[11] headline caught my eye, “NASA battery survives -274°F extreme cold, could power future lunar rovers and tools” and immediately I needed to dig deeper to see what was really behind it and the reality can be summarised more correctly: “The headline number is survival, not operation. NASA’s SPEARS solid-state prototypes survived exposure down to −170 °C and up to 100 °C without damage. They still worked at −80 °C, with reduced performance, and performed well between −60 °C and 80 °C in thermal vacuum testing at Langley. They can’t deliver power at −170 °C. The idea is to let the cells cold-soak through the lunar night, then warm them with a small heater before use. That avoids keeping them warm continuously.”

On 14 October 2026 NASA reported that small solid-state prototype cells from its SPEARS project came through thermal vacuum testing after exposure to −170 °C, and still delivered power at −80 °C [1]. The headline figure has been widely repeated as “survives −274 °F”. For a battery engineer the more useful question is what was actually shown, what sits behind it, and what a lunar pack would need from the cell. This post pulls together NASA’s release, the conference material the team has put on the NASA Technical Reports Server, and the published requirements for the most likely customer, the Artemis Lunar Terrain Vehicle. It is a companion to our Solid State Battery hub and the Battery Heating page.
What NASA has said
SPEARS (Solid Power for Extreme Environment Adaptability and Resilience in Space) is led from NASA Glenn, with materials development and testing at NASA Langley and flight test boards being built at NASA Ames [1]. The cells in the release are coin cells, which matters for how far we can read across to a pack. The test results split into two quite different claims, and it is worth keeping them apart.
| Claim | Temperature (°C) | What it means |
|---|---|---|
| Survival (no damage after exposure) | −170 to +100 | Cold soak, then warm and use |
| Good operation | −60 to +80 | Usable working window |
| Operation, reduced performance | down to −80 | Works, but no rate or capacity stated |
| Operation at −170 °C | − | Not claimed; cells must be warmed first |

The design idea is simple: let the cells go cold through the lunar night, then use a small heater to bring them up to operating temperature when power is needed, rather than keeping them warm continuously [1].
NASA has not said which cathode, anode or electrolyte the tested cells use beyond “ceramic- or polymer-based”, and has not published capacity retention after the cold soak.
What sits behind it: the NASA conference material
There is no formal NASA Technical Report on SPEARS yet. The team has presented at the APS Global Physics Summit, the Space Power Workshop and the ECS meeting in 2026 [3][4][10], and the slides from the ACS Fall 2026 meeting are public [2]. Those slides are the most detailed source and tell us three useful things.
It grew out of SABERS. SPEARS follows NASA’s earlier solid-state lithium-sulfur programme for electric aircraft, SABERS, which ran from 2019 to 2024 [2]. The SABERS team reports more than 500 Wh/kg at unit-cell level (cathode, solid electrolyte and lithium anode only, so not a packaged cell) and more than 20 mAh/cm² areal capacity at under 2 MPa stack pressure, against the 10 to 200 MPa often used in lab cells [2]. They also say plainly that power and cycling remained the hard part. See our Lithium Sulfur page for the chemistry.
The electrolyte screen is where the low-temperature work starts. The team screened 12 solid electrolytes, eight sulfides and four halides, measuring conductivity by impedance from −30 °C to 150 °C on pressed pellets [2]. Their four leading candidates at −30 °C are below. The stated SPEARS target is adequate conductivity at −40 °C and below [2].
| Electrolyte | Type | Ea (eV) | σ at −30 °C (mS/cm) |
|---|---|---|---|
| “Sulfide #1” (not named) | Sulfide | 0.119 | 2.30 |
| Cl-rich argyrodite (Li5.5PS4.5Cl1.5) | Sulfide | 0.159 | 1.63 |
| Li6PS5Cl (baseline argyrodite) | Sulfide | 0.189 | 0.68 |
| “Halide #1” (not named) | Halide | 0.168 | 0.35 |
Cell formats are deliberately practical. The team says most academic solid-state work uses split cells with heavy fixtures, and that SPEARS is aiming at coin, pouch or cylindrical cells that can be field tested, built with dry processing and polymer binders [2]. A solid-state coin cell has already flown on a SmallSat launched in January 2025 [2]. The slides do not disclose the SPEARS electrode chemistries or any cycle data, so we cannot yet say whether the cells in the thermal vacuum test are lithium-sulfur, a layered oxide or something else.
Is the separator the problem at −80 °C?
We can get a feel for this with the published activation energies. Assuming simple Arrhenius behaviour carries on below the measured range:
where σ is ionic conductivity, σref the measured value at Tref = 243 K (−30 °C), Ea the activation energy, kB Boltzmann’s constant (8.617 × 10−5 eV/K) and T the temperature in kelvin. Taking 193 K (−80 °C) and a 30 µm separator, in line with the 20 to 30 µm solid electrolyte films NASA Glenn has reported [5], the area-specific resistance of the separator alone is thickness divided by conductivity.
| Electrolyte | σ at −60 °C (mS/cm) | σ at −80 °C (mS/cm) | 30 µm separator ASR at −80 °C (Ω·cm²) |
|---|---|---|---|
| Sulfide #1 | 1.0 | 0.53 | 5.7 |
| Cl-rich argyrodite | 0.56 | 0.23 | 13 |
| Li6PS5Cl | 0.19 | 0.066 | 46 |
| Halide #1 | 0.11 | 0.044 | 68 |
At 1 mA/cm², a 5.7 Ω·cm² separator costs under 6 mV. Even the baseline argyrodite costs under 50 mV. Our reading is that for the best sulfide, the bulk separator is not what limits a cell at −80 °C. The losses that matter will be at the interfaces and in the composite cathode, where ions have to travel a tortuous path through catholyte and cross solid-solid contacts whose charge transfer also slows sharply in the cold. That is consistent with the team’s own summary that ion transport, and particularly interfacial contact, is the key problem [2]. It also explains why the release says “reduced performance” without a rate: a cell can work at −80 °C at a low current while still being unusable at a rover’s peak draw. Our Cathode-Electrolyte Interface page covers why the catholyte is usually the bottleneck.
What the mission needs from the cell
SPEARS is a technology development project, not a battery built to a mission specification, and NASA has not assigned it to a vehicle [1]. The nearest published requirements come from the Lunar Terrain Vehicle and from NASA’s work on surviving the lunar night.
- Night length. A full lunar night is 354 hours, but carefully chosen south-pole sites never see more than 125 hours of darkness over a 10-year life, and that is the figure being used to size the LTV [6]. NASA’s LTV request for information set at least 85 hours as the minimum, with 125 hours preferred [7].
- Vehicle. The same RFI asked for a 10-year life, two suited crew, at least 800 kg payload and 20 km range at the south pole without recharging [7].
- Environment. Night-time lows at the poles reach around 50 K [8], and NASA quotes −200 °C for the south pole [1]. Neither is inside the −170 °C survival figure yet.
- The mass at stake. NASA Glenn estimates that keeping LTV-class batteries warm through the night with heaters would take 15 to 30 kWh, or 100 to 200 kg of extra battery [8]. That is the prize for a cell that can simply go cold.
SPEARS is not the only route to that prize. Glenn’s Lunar Power Hibernation work takes conventional liquid Li-ion cells, disconnects them from the bus before they freeze and restarts at lunar dawn using cryogenic-capable electronics running from the solar array [9]. In preliminary tests, commercial 18650 cells (LG M36 and Molicel M35A) survived repeated 14-day dwells at about 50 K in vacuum and recovered afterwards with little loss of discharge capability, and a demonstrator power supply cold-started at 57 K after a cold soak [8]. So the honest comparison for SPEARS is not “solid-state versus a liquid cell that dies in the cold”. It is solid-state versus a hibernating liquid cell, and the solid-state case has to win on safety, vacuum tolerance, energy density or cycle life through repeated deep thermal cycles.
Warming up is not free
The “small heaters” in the release still need energy, and a cold-soaked cell cannot supply it. A rough upper bound: take 1 kg of cell from −170 °C to −40 °C, a rise of 130 K. Using the room-temperature specific heat of about 1,000 J/kg·K from our Thermal hub:
Specific heat falls at cryogenic temperatures, so the true figure will be lower, but the order of magnitude matters. NASA has not published an energy density for the SPEARS cells. If we assume 200 to 250 Wh/kg (our assumption, in the range of today’s production cells), that is a meaningful fraction of the cell’s own energy, and it has to come from somewhere else: the solar array at dawn, a separate warm battery, or a radioisotope heater. This is the same architecture question the hibernation team faces, and it means the BMS and power electronics have to work cold before the battery does. Our Battery Heating and Cold Temperature Charge and Discharge pages cover the terrestrial version of the problem.
Cold temperature performance of a cell is often tricky as we can look at just discharge and ignore charging, only to find that the application needs somewhere to sink a source of power. That charging power might be very small and initially ignored as being important. Hence, the need to fully understand the system and it’s operation.
Two pack-level questions are also left open. First, stack pressure: a fixture that holds pressure on a solid-state stack has to do so through a 270 K swing, with steel, aluminium and ceramic all contracting by different amounts. NASA’s low-pressure approach helps, but we found no published data on pressure retention through cold cycling. See Stack Pressure and Interfacial Resistance for Pack Designers. Second, charging: removing the liquid does not remove the risk of lithium plating at low temperature on a lithium-metal or graphite anode, so a charge-temperature limit will still be needed.
What to take from this
- Read “survives −170 °C” as a storage claim. The working figure is −60 °C, with reduced performance to −80 °C at an undisclosed rate.
- The electrolytes are good enough in the bulk. On NASA’s own numbers the best sulfide separator is not the limit at −80 °C; interfaces and the composite cathode are. Expect future SPEARS results to be about those.
- It is still a coin cell at an early sample level. No capacity, cycle life, energy density or chemistry has been published for the tested cells. Our reading is that, in Cell Sample Maturity terms, this is pre-A sample level; NASA has not used that language.
- The benchmark is a hibernating Li-ion pack, not a heated one. SPEARS has to beat cryo-tolerant liquid cells with cold-start electronics, which NASA is developing in parallel.
The direction makes sense. Our view is that solid-state cells are a better long-term fit for repeated deep cold soaks in vacuum than liquid cells, because there is nothing to freeze, boil or leak. The data to prove it, cycle life through dozens of simulated lunar nights in a pouch-format cell, is what to watch for next. The ECS and Space Power Workshop decks [3][4] are the ones most likely to hold low-temperature cell data and are worth requesting from the authors.
References
- NASA Develops Batteries to Power Humanity’s Return to Moon, NASA Langley Research Center, 14 October 2026.
- Lin, Y. et al., Challenges in Solid-State Batteries toward Applications in Extreme Space Environment, ACS Fall Meeting, Chicago, August 2026 (presentation, NTRS 20260007982).
- Lee, R. Y. et al., Developing Low Temperature All-Solid-State-Batteries for Aerospace Applications, 249th ECS Meeting, May 2026 (NTRS 20260004579; no preview available).
- Lin, Y. et al., Materials Challenges for Solid-State Batteries in Aerospace Applications, 43rd Space Power Workshop, April 2026 (NTRS 20260002690; no preview available).
- Next Generation Batteries for Electric Aviation and Space, NASA Glenn Research Center (NTRS 20240000923).
- Surviving Night at the Lunar South Pole: Exploring Viability of Radioisotope Power Systems for a Crewed Rover, TFAWS 2023.
- NASA Posts RFI for Lunar Terrain Vehicle, Potomac Officers Club (secondary summary of the NASA JSC RFI on SAM.gov).
- Uguccini, N., Surviving the Lunar Night, Space Power Workshop 2024 (presentation).
- Power Hibernation for Low-Cost Solar Powered Lunar Missions, NASA Glenn Research Center (NTRS 20230015563).
- Lin, Y. et al., Solid-Solid Interfaces in Solid-State Batteries, APS Global Physics Summit, Denver, March 2026 (NTRS 20260002075; no preview available).
- NASA battery survives -274°F extreme cold, could power future lunar rovers and tools, Interesting Engineering
- Adeli, P. et al., Boosting Solid-State Diffusivity and Conductivity in Lithium Superionic Argyrodites by Halide Substitution, Angewandte Chemie International Edition 58 (2019) 8681–8686.
See also: Solid State Battery, Solid Electrolyte Comparative Matrix, Cathode-Electrolyte Interface & Catholyte Engineering, Stack Pressure & Interfacial Resistance for Pack Designers, Things about Solid State Packaging, Thermal Runaway & Safety Realities in Solid-State Cells, Lithium Sulfur, Battery Heating, Temperature Limits of a Battery, Cell Energy Density Roadmap to 2035
Get the next teardown in your inbox
New teardowns, benchmarks and design data from BatteryDesign.net, free by email. Unsubscribe any time.