Battery storage
Electrochemical energy storage — the bridge between intermittent PV and continuous load. Lithium-ion (NMC, LFP) dominates near-term Mars architectures: ~ 250 Wh/kg gravimetric, 4000+ cycles, mature flight heritage (Ingenuity ran on Sony Li-ion). Mars-specific challenges add 15–30 % parasitic thermal-management load to keep cells above −20 °C; deeper-cycle storm-survival sizing pushes battery mass into the tons-per-crew range for solar-only architectures.
Governing equations
Usable stored energy = nominal capacity × voltage × round-trip efficiency. Li-ion round-trip 0.92–0.96; lead-acid 0.75–0.80. [1]
State-of-charge by integration. Real systems also use voltage + model-based estimators because Coulomb counting drifts. [2]
Capacity retention after N cycles. k_1 captures SEI-layer growth (calendar + cycle), k_2 captures cathode degradation. Mars-temperature-cycled Li-ion fades faster than terrestrial. [1]
Parasitic thermal-management load on Mars: sensible heating from ambient to operating temperature plus continuous heat-loss replacement. [3]
Key constants & quantities
| Symbol | Value | Units | Conditions | Description |
|---|---|---|---|---|
| ρ_E,Li-ion | 220–280 ±10 % | Wh / kg (cell level) | — | Gravimetric energy density of commercial automotive Li-ion (NMC 811). Pack-level adds 15–25 % thermal management + structural overhead.[4] |
| ρ_V,Li-ion | 600–800 | Wh / L (cell) | — | Volumetric energy density. Often the binding constraint for stowed launch volume on Mars-bound missions.[4] |
| η_RT,Li-ion | 92–96 | % (round-trip) | — | AC-to-AC round-trip efficiency with inverter loss. Cell-to-cell DC is ~ 97–99 %.[2] |
| N_cycle | 4000–8000 | cycles to 80 % retention | — | Cycle life for LFP at 80 % depth-of-discharge, room T. NMC is 2000–3000; Li-S is 500–1500 (emerging).[1] |
| T_op,min | -20 | °C (Li-ion practical) | — | Minimum operating temperature for Li-ion charging without lithium plating damage. Discharge possible to ~ −40 °C at reduced capacity (~ 50 %).[3] |
| T_thermal-runaway | 130–180 | °C (NMC), 270 (LFP) | — | Onset temperature for thermal runaway. LFP is the safer chemistry — ~ 270 °C and self-limiting; NMC propagates aggressively.[1] |
| Ingenuity battery | 35.75 | Wh (6 × Sony US18650VTC4, 273 Wh/kg) | — | Mars Helicopter battery pack — the flight reference for Mars-surface Li-ion. Survived 72 flights + ~ 3 years cold cycles.[5] |
| m_storm-survival | 2000–5000 | kg battery / 4-crew base / sol of load | — | Order-of-magnitude battery mass for storm-survival sizing of a solar-only architecture. Drives the architectural choice toward PV+nuclear hybrid.[6] |
Operating envelope
Mass balance
Basis: 100 kWh useful storage at battery-pack level, Li-ion NMC chemistry
Inputs
| Lithium-ion cells | 360 | kg (NMC 811, 280 Wh/kg cell) | [4] |
| Pack structure + cooling + electronics | 90 | kg | [4] |
| Thermal-management energy (Mars annual) | 8,000 | kWh | [3] |
- Pack structure + cooling + electronics: BMS, busbars, thermal-management plates, enclosure. ~ 20 % of cell mass.
- Thermal-management energy (Mars annual): Continuous heater + heat-loss replacement to hold pack above −20 °C. ~ 20 % of throughput goes to keeping the battery warm.
On Earth, batteries are nearly parasitic-free. On Mars surface, holding pack T above −20 °C through the night burns ~ 15–25 % of throughput as resistance-heater load. Net useful storage efficiency drops to ~ 75 %.
Variants & trade-offs
Lithium-ion NMC (Tesla 4680 / Ingenuity heritage)
[1]Nickel-Manganese-Cobalt cathode + graphite anode + liquid electrolyte. Highest energy density currently in mass production. Spacecraft heritage in 18650, 21700, and prismatic formats.
- Cell voltage
- 3–4.2 V
- Energy density
- 240–280 Wh/kg cell
- Highest energy density of any mature chemistry
- Mars-flight heritage (Ingenuity 2021–)
- Mass production drives lowest $/kWh ($100/kWh pack level)
- Fast charge rates support PV-feed peaks
- Thermal runaway 130–180 °C — propagates aggressively in pack
- Cobalt + nickel are hard imports for Mars settlement
- Plating below 0 °C charge — strict thermal management
- Calendar fade ~ 2 %/year independent of cycling
Lithium iron phosphate (LFP, BYD Blade)
[1]Lower-energy-density cousin of NMC with vastly better thermal safety (runaway onset 270 °C) and 2–3× cycle life. The mainstream automotive choice as of mid-2020s.
- Cell voltage
- 2.8–3.6 V
- Energy density
- 160–180 Wh/kg cell
- Thermal runaway threshold 270 °C — much safer pack architecture
- Cycle life 2–3× NMC
- No cobalt or nickel — easier supply chain for Mars
- Lower calendar fade than NMC
- ~ 30 % lower energy density than NMC
- Lower nominal voltage (3.2 V) means more cells in series for same bus voltage
- Less mature in spacecraft applications
Vanadium redox flow (grid-scale)
[2]Liquid electrolyte tanks pumped through a separate electrochemical stack. Power and energy decoupled — you can scale energy by adding tank volume without changing stack hardware.
- Energy density
- 25–50 Wh/kg system
- Cycle life
- 10000–20000 cycles
- Decouples power from energy — scale energy by tank size
- Effectively unlimited cycle life
- Non-flammable — no thermal runaway
- Electrolyte recyclable; no rare-metal supply issue beyond V
- 10× lower energy density than Li-ion
- Requires pumps + plumbing — moving parts
- V supply is constrained (~ 80% from China terrestrially)
- Freezing of aqueous electrolyte at low T — heating overhead
Failure modes
| Mode | Cause | Detection | Mitigation |
|---|---|---|---|
| Thermal runaway (safety-critical)[1] | Internal short circuit (mechanical, dendrite, manufacturing defect) ignites the electrolyte; exothermic reaction propagates cell-to-cell. | Cell voltage drops while temperature spikes; off-gas detector (HF, CO). | LFP chemistry where mass budget allows; intumescent fire-blanket between cells; cell-level fusing; isolation valves between pack modules; pressure-relief vent stack. |
| Lithium plating from sub-zero charging[3] | Charging Li-ion below 0 °C plates metallic lithium on the anode instead of intercalating. Plated Li forms dendrites, grows toward cathode, eventually shorts. | Capacity fades faster than calendar model; impedance spectroscopy shows characteristic shift. | Hard interlock: no charge below 0 °C cell T. Pre-heat to 5 °C before allowing PV input. Pulse-warm via internal resistance during discharge if necessary. |
| Capacity fade (calendar + cycle)[1] | SEI layer growth, cathode crystal-structure degradation, electrolyte breakdown. Continues whether the battery is used or not. | Capacity test at full discharge; impedance measurement; coulometric SOC drift. | Operate in 10–90 % SOC window; storage at 50 % SOC + low T; replace pack at 80 % retention. |
| BMS sensor drift[2] | Voltage / temperature sensors drift; SOC estimate diverges from reality. | Periodic full charge-discharge cycle reveals discrepancy. | Multiple redundant sensors; periodic recalibration cycles; model-based estimation that fuses voltage + Coulomb + impedance. |
| Cold-soak capacity loss (reversible)[3] | Below −20 °C, electrolyte ionic conductivity drops by orders of magnitude; effective capacity halves. | Discharge capacity at constant load drops in winter; recovers when warm. | Insulated battery enclosure; resistive trickle heater; design system for cold-derated capacity in worst-case scenarios. |
| Connector / busbar corrosion[1] | Atmospheric moisture + dust deposit at connector interfaces; over years, contact resistance rises. | I²R loss climbs; thermal imaging shows hot connector. | Sealed connector blocks; periodic inspection + cleaning; redundant parallel paths. |
Mars adjustments
Cold-soak thermal management[3]
Impact: Mars surface T ranges −90 °C to +20 °C. Li-ion operating window is −20 to +60 °C; charging requires > 0 °C. Holding cells warm through the night costs 15–25 % of throughput as heater energy.
Mitigation: Insulated enclosure (vacuum-jacketed, MLI); waste-heat coupling from nearby cryocooler hot-side or habitat HVAC; redundant heaters; battery-bank co-location with continuously-running loads.
Dust storm capacity buffer[6]
Impact: For solar-only architectures, multi-week dust storms require battery storage for the entire storm duration. Sizing this leads to 2–5 t of Li-ion per 4-crew base.
Mitigation: PV + nuclear hybrid for crewed missions reduces storm-survival battery to days, not weeks. Hydrogen buffer via electrolysis is mass-efficient at this scale.
Surface radiation environment[3]
Impact: Galactic cosmic rays and solar protons can flip BMS firmware bits and degrade cell electrolyte over years. Lower flux than LEO but unshielded by atmosphere or magnetosphere.
Mitigation: Radiation-hardened BMS electronics; ECC memory; programmable redundancy; cell-level shielding via regolith berming for fixed installations.
Deep storage cycling for 26-month mission cadence[1]
Impact: Mars cargo missions may sit on the surface 26 months unused. Calendar fade + extreme T cycling without active management can degrade unused packs significantly.
Mitigation: Storage protocol: pack at 50 % SOC, insulated enclosure, periodic awakening for self-check + balance. LFP chemistry tolerates this far better than NMC.
Lower gravity affects thermal-management plumbing[3]
Impact: Liquid-cooled packs rely partly on natural convection. 0.38 g reduces convection-driven flow; forced circulation pumps must compensate.
Mitigation: Pumped-loop liquid cooling sized for 0.38 g convection; phase-change materials as thermal buffers; air-cooled (Mars CO₂) instead of liquid for smaller packs.
Alternatives & substitutes
Hydrogen-as-storage (electrolysis + fuel cell)[2]
- Decoupled energy from power — store unlimited duration in H₂ tanks
- Uses water-electrolysis infrastructure already deployed for ISRU
- No cold-soak penalty (gas is unaffected)
- Round-trip efficiency ~ 35–40 % vs Li-ion 92 %
- Mass is in the tank pressure-vessel, not the hydrogen
- Two separate subsystems vs one battery
When preferred: Multi-week or seasonal energy buffer where Li-ion mass is prohibitive; combined with existing ISRU infrastructure.
Flywheel energy storage[2]
- No degradation with cycles — mechanical, not electrochemical
- High power density (fast discharge)
- No cold-soak issue
- Low energy density (~ 100 Wh/kg system level)
- Bearing wear is lifetime-limiter
- Gyroscopic effects on lighter Mars habitats
- Self-discharge ~ 5 %/h
Requires
Inputs
References
- (2018). Battery Technologies for Grid-Scale Energy Storage. Annual Review of Chemical and Biomolecular Engineering, 9, 333-355. doi:10.1146/annurev-chembioeng-060817-084218 — Comprehensive review of Li-ion, LFP, NaS, redox flow chemistries; cycle life, safety, applications.
- (2020). Energy Storage Grand Challenge: Energy Storage Market Report 2020. U.S. Department of Energy, Office of Energy Efficiency and Renewable Energy. DOE/GO-102020-5497. — DOE energy storage technology comparison: chemistries, lifetimes, roundtrip efficiency, deployment scale.
- (2007). Performance Characterization of Lithium-Ion Cells for Aerospace Applications. NASA Glenn Research Center, NASA/TM-2007-214958. NASA/TM-2007-214958. — NASA Glenn Li-ion testing at low temperature, cold-soak performance, aerospace cycling models.
- (2019). BatPaC: Battery Performance and Cost Model for Lithium-Ion Batteries. Argonne National Laboratory, ANL-19/16. ANL-19/16. — BatPaC model — industry-standard battery pack mass, volume, and cost modeling.
- (2021). The Ingenuity Helicopter on the Perseverance Rover. Space Science Reviews, 217(4), 56. doi:10.1007/s11214-021-00815-w — Mars Helicopter — Li-ion 18650 battery flight; first powered flight on another planet; 3 yr operational data.
- (2008). Mars Year 28 Global Dust Storm: Optical Depth and Atmospheric Effects. Journal of Geophysical Research: Planets, 113(E10), E10006. doi:10.1029/2008JE003133 — Global Mars dust storm characterization; τ measurements, impact on surface insolation.