comminution

Comminution (crushing & grinding)

Process Semi-native mining
TRL Mars
Energy intensity
Required by
0
Requires
3

Reduces mined rock and regolith to the particle size that liberates mineral grains for downstream separation — typically crushing from run-of-mine to centimeters, then grinding to tens or hundreds of microns. It is the dominant energy consumer in any ore operation, governed by the Bond work index, with most input energy lost as heat. On Mars the prize is the waste heat (useful) and the dust (hazardous), and the constraint is grinding-media supply.

Last reviewed: 2026-06-14

Governing equations

Bond's comminution law: specific energy W (kWh/t) to reduce feed of 80%-passing size F₈₀ to product P₈₀ (both in µm), scaled by the material work index Wᵢ. The industry sizing equation since 1952. [1]

Comminution efficiency: the fraction of input energy that becomes new surface area. The overwhelming balance is heat, sound, and plastic deformation — why grinding dominates the plant energy budget. [2]

Tumbling-mill power draw scales with charge density, the 2.5 power of mill diameter, length, and fraction of critical speed — the relation that makes large mills efficient on Earth and problematic to scale down for a small colony. [2]

Critical speed — where centrifuging begins. It scales with √g, so at Mars 0.38 g critical speed drops to ~62 % of the Earth value and grinding dynamics shift: a recalculation, not a copied data sheet. [2]

Key constants & quantities

Symbol Value Units Conditions Description
Wᵢ (hard basalt) 18–25 kWh/t Bond work index for hard mafic rock — Martian basalt sits in this band, harder to grind than average terrestrial ore.[1]
W (typical grind) 10–25 kWh/t Specific grinding energy to reach liberation size (~75-150 µm P₈₀) for hard rock — the number that multiplies every tonne processed.[2]
W (crushing) 0.5–3 kWh/t Specific energy for crushing stages (coarse size reduction) — an order of magnitude below grinding, hence stage crushing before mills.[2]
P₈₀ liberation 40–150 µm Typical grind size for mineral liberation; finer grinds liberate more but cost disproportionately more energy (Bond's inverse-root law).[2]
Media wear 0.1–1 kg steel / t ore Grinding-media consumption — a recurring steel demand that, on Mars, ties comminution capacity directly to the EAF/steel chain.[3]
N_c ratio (Mars) 0.62 × Earth critical speed Critical speed scales as √g; at 0.38 g tumbling mills must be re-tuned and run slower for the same cataracting action.[2]

Operating envelope

ParameterRangeUnitsSource
Feed size (F₈₀) 1000 – 500000 µm (ROM to crushed) [2]
Product size (P₈₀) 40 – 150 µm [2]
Grinding specific energy 10 – 25 kWh/t [1]
Mill fraction of critical speed 65 – 80 % [2]
Mill ball charge 25 – 40 % of volume [2]

Mass balance

Basis: 1 t ore ground from ROM to 100 µm P₈₀

Inputs

Run-of-mine ore/regolith 1 t [2]
Electrical energy 18 kWh [1]
Grinding media (steel) 0.4 kg [3]
  • Electrical energy: Crushing ~2 + grinding ~16 for hard basalt at Wᵢ ≈ 20.
  • Grinding media (steel): Consumed by wear; high-chrome or forged steel from the local EAF chain.

Outputs

Liberated mineral particles 1 t [2]
Waste heat 17 kWh [2]
Fugitive dust 1 hazard [4]
  • Liberated mineral particles: At 100 µm P₈₀, ready for magnetic/gravity/flotation separation.
  • Waste heat: ~95 % of input energy — recoverable at low grade for habitat/process preheat.
  • Fugitive dust: Fine perchlorate-bearing respirable dust — the dominant safety hazard of dry comminution on Mars.
TRL · Earth
9/ 9
TRL · Mars
4/ 9
Crushing and grinding are mature to the point of cliché on Earth. Mars-side, the unit operations are unflown but mechanically simple; the open questions are reduced-gravity mill dynamics, grinding-media supply before local steel matures, and dry-dust containment. NASA ISRU testbeds have demonstrated regolith crushing at TRL 4-5 in simulant.[5]
Energy budget
18 kWhe / t ore ground to 100 µm (hard basalt, Wᵢ ≈ 20) [1]

Comminution is a net heat EXPORTER: nearly all electrical input degrades to ~40-80 °C waste heat. At settlement scale this is the single largest controllable load in the extractive plant — a 10 t/h mill draws ~180 kW, a meaningful fraction of an early power budget.

Variants & trade-offs

Jaw / cone crushing (coarse stages)

[2]

Compressive crushing of run-of-mine rock down to centimeters — the low-energy front end before any mill.

Materials: Manganese-steel crushing surfaces · Heavy frame + eccentric drive
  • Lowest specific energy of any size-reduction step
  • Robust, simple, tolerant of tramp material
  • Wear parts (Mn-steel liners) are castable locally
  • Limited reduction ratio — needs grinding downstream for liberation
  • Heavy machine mass to import or fabricate

When preferred: Always — the first stage of any hard-rock circuit.

Tumbling mills (ball / SAG)

[2]

Rotating drum where steel balls (or the ore itself, in SAG) cascade and grind the charge to fine product.

Materials: Steel grinding media · Replaceable mill liners · Large low-speed drive + bearings
  • Proven workhorse; handles high throughput to fine sizes
  • SAG mode reduces media demand by using coarse ore as grinding media
  • Highest energy and media consumption in the plant
  • Critical-speed dynamics shift at 0.38 g — needs re-tuning, not transplant
  • Large rotating mass; bearing and drive are major imports

When preferred: Bulk tonnage to liberation size once steel-media supply exists.

High-pressure grinding rolls (HPGR)

[3]

Two counter-rotating rolls squeeze a packed bed at high pressure — inter-particle comminution at markedly lower energy than tumbling mills.

Materials: Studded wear-resistant roll surfaces · Hydraulic loading system
  • 15-30 % lower specific energy than ball milling for many ores
  • Minimal grinding media — decouples comminution from steel supply
  • Micro-cracks the product, easing downstream grinding and leaching
  • Roll surfaces are specialist wear imports
  • Sensitive to feed moisture and tramp metal

When preferred: Energy- and media-constrained Mars plant — the efficiency-first choice.

Stirred / vibratory fine mills

[2]

Bead or vibratory mills for ultra-fine grinding (<20 µm) where tumbling mills become hopelessly inefficient.

Materials: Ceramic grinding beads · Stirrer or vibratory drive
  • Reaches fine sizes economically where ball mills cannot
  • Compact; gravity-independent stirred action suits 0.38 g
  • Throughput-limited; a polishing stage, not a bulk grinder
  • Ceramic media supply

When preferred: Liberating fine-grained ores and producing feed for precision metallurgy.

Failure modes

Mode Cause Detection Mitigation
Respirable dust release (safety-critical)[4] Dry crushing and grinding of perchlorate- and silica-bearing regolith generates fine respirable dust; in a habitat-adjacent plant this is a chronic crew-health and equipment hazard. Particulate monitors in the plant enclosure; pressure-drop on dust collection. Enclosed circuits at negative pressure, cyclone + bag/HEPA collection, wet grinding where water can be spared; plant sited in its own ventilation zone.
Liner and media wear-out[3] Abrasive hard rock grinds away mill liners and consumes grinding media continuously. Power-draw trend, throughput decline, liner thickness inspection at relines. Local Mn-steel liner casting and grinding-media production; HPGR to minimize media demand; scheduled reline campaigns.
Mill overload / centrifuging at wrong speed[2] Running a tumbling mill near Earth-calibrated critical speed at 0.38 g centrifuges the charge — grinding stops, power spikes. Power signature, acoustic/vibration monitoring, product-size drift. Re-derive critical speed for Mars gravity (≈62 % of Earth value); variable-speed drive with charge-motion sensing.
Tramp metal damage[3] Excavation hardware fragments enter the circuit and jam or fracture crushing surfaces. Metal detectors on feed conveyors; sudden power transients. Magnetic tramp removal ahead of crushers; protected-discharge crusher designs.
Bearing / drive failure on large mills[6] The mill trunnion bearing and drivetrain are the highest-load rotating components in the plant. Vibration and temperature monitoring, lubricant analysis. Installed-spare philosophy, local bearing reconditioning (precision-bearings node), modular smaller mills over a single large one.

Mars adjustments

The planet pre-ground the surface for free[7]

Impact: Eons of impact gardening and aeolian abrasion left a thick mantle of fine regolith. For any target liberated at silt/sand sizes, the single most expensive unit operation in Earth mineral processing is already done.

Mitigation: Bias the extractive flowsheet toward minerals recoverable from natural fines; reserve grinding for cemented and hard-rock sources.

Reduced gravity changes mill dynamics[2]

Impact: Critical speed, charge trajectory, and classification cut-points all depend on g. Earth mill data sheets do not transfer; cataracting/cascading transitions occur at different speeds at 0.38 g.

Mitigation: Variable-speed drives, re-derived operating curves, and gravity-independent options (HPGR, stirred mills) preferred.

Energy is the binding constraint[1]

Impact: At 10-25 kWh/t, grinding can dominate the early settlement power budget. Comminution scale, not ore grade, may set how much metal a colony can make.

Mitigation: Efficiency-first equipment (HPGR), recover the ~95 % waste heat, schedule grinding to power-surplus hours.

Grinding media couples to the steel economy[3]

Impact: Consuming up to ~1 kg steel per tonne ore, tumbling mills create a recurring steel demand that competes with structural needs until local steel is abundant.

Mitigation: Media-light circuits (HPGR, SAG, autogenous), local high-chrome media casting once metallurgy matures.

Dry-dust hazard is amplified[4]

Impact: Perchlorate toxicity and fine silicate particulates make uncontrolled comminution dust a crew-health hazard in a way Earth dust-control practice does not fully cover.

Mitigation: Sealed negative-pressure circuits, high-efficiency collection, and where water permits, wet grinding to suppress dust entirely.

Alternatives & substitutes

Use naturally fine regolith (skip primary grinding)[7]

  • Mars surface fines are already silt-to-sand sized — much beneficiation needs no grinding at all
  • Eliminates the largest energy load for those feeds
  • Liberation only works if the target mineral is already free at the natural grain size
  • Hard-rock and cemented deposits still require full comminution

When preferred: Magnetic separation of fines, sulfate/perchlorate brine extraction — process the dust the planet already ground.

Thermal / chemical pre-weakening[3]

  • Microwave or thermal-shock pre-treatment cracks grain boundaries, cutting grinding energy
  • Direct chemical attack (acid digestion) can bypass fine grinding for some minerals
  • Added thermal energy and equipment; net benefit is ore-specific
  • TRL low for Mars conditions

When preferred: High-work-index ores where grinding energy dominates.

Requires

References

  1. Bond, F. C. (1952). The Third Theory of Comminution. Transactions AIME, 193, 484–494. — Origin of the Bond Work Index and the comminution energy law W = 10·Wi·(1/√P₈₀ − 1/√F₈₀) — still the industry sizing basis for crushers and mills.
  2. Wills, B. A., & Finch, J. A. (2016). Wills' Mineral Processing Technology: An Introduction to the Practical Aspects of Ore Treatment and Mineral Recovery, 8th Edition. Butterworth-Heinemann. doi:10.1016/C2010-0-65478-2 — The canonical mineral-processing text: comminution, classification, gravity/magnetic/flotation separation, dewatering, flowsheet design.
  3. Dunne, R. C., Kawatra, S. K., & Young, C. A. (Eds.) (2019). SME Mineral Processing and Extractive Metallurgy Handbook. Society for Mining, Metallurgy & Exploration. ISBN 978-0-87335-385-4. — Comprehensive practitioner reference across comminution, separation, hydro/pyrometallurgy, materials handling, and plant operations.
  4. Gaier, J. R., Ellis, S., & Hanks, N. C. (2002). Aeolian removal of dust types from photovoltaic surfaces on Mars. NASA Glenn Research Center, NASA/TM-2002-211837. NASA/TM-2002-211837. — Mars dust deposition + removal mechanisms on optical / radiator surfaces; α_s and ε degradation rates.
  5. Mueller, R. P., Smith, J. D., Schuler, J. M., Nick, A. J., Gelino, N. J., et al. (2016). Design of an Excavation Robot: Regolith Advanced Surface Systems Operations Robot (RASSOR) 2.0. NASA Kennedy Space Center, ASCE Earth + Space Conference 2016. doi:10.1061/9780784479179.018 — NASA Mueller RASSOR design: counter-rotating bucket-drum architecture for low-g excavation; demonstrated 2014-2016 in Mars regolith simulant.
  6. Harris, T. A., & Kotzalas, M. N. (2006). Rolling Bearing Analysis, 5th Edition (Essential Concepts of Bearing Technology + Advanced Concepts of Bearing Technology). CRC Press. ISBN 978-0-8493-7183-7. — Definitive precision-bearing engineering reference: design + materials + lubrication + L10 fatigue life + applications.
  7. McLennan, S. M., Sephton, M. A., Beaty, D. W., Hecht, M., et al. (2014). Planning for Mars Returned Sample Science: Final Report of the MSR End-to-End International Science Analysis Group. NASA Mars Exploration Program Analysis Group (MEPAG). — Mars surface materials properties and ISRU planning; basis for water extraction system sizing.